PRIMER ON

KIDNEY DISEASES
Second Edition

NATIONAL KIDNEY FOUNDATION

SCIENTIFIC ADVISORY BOARD

CHAIRMAN

SANDRA SABATINI
Departments of Physiology and Internal Medicine
Texas Tech University Health Sciences Center
Lubbock, Texas

CONSULTING EDITORS

Garabed Eknoyan

James V. Donadio, Jr.

Department of Medicine
Baylor College of Medicine
Houston, Texas

Division of Nephrology
Mayo Clinic
Rochester, Minnesota

Shaul G. Massry, M.D.

William E. Mitch

Division of Nephrology
University of Southern California
School of Medicine
Los Angeles, California

Renal Division
Emory University
Atlanta, Georgia

PRIMER ON
KIDNEY DISEASES
Second Edition
EDITOR

ARTHUR GREENBERG
Division of Nephrology
Department of Medicine
Duke University
Durham, North Carolina

ASSOCIATE EDITORS
Alfred K. Cheung

Thomas M. Coffmann

Division of Nephrology and Hypertension

University of Utah School of Medicine
and Veterans Affairs Medical Center
Salt Lake City, Utah

Division of Nephrology
Department of Medicine
Duke University and Durham
Veterans Administration Medical Center
Durham, North Carolina

Ronald J. Falk

J. Charles Jennette

Department of Medicine
Division of Nephrology
University of North Carolina
Chapel Hill, North Carolina

Department of Pathology
and Laboratory Medicine
University of North Carolina
Chapel Hill, North Carolina

ional Kidney Foundation1

ACADEMIC PRESS
San Diego

London

New York

Boston

Sydney

Tokyo

Toronto

The opinions expressed and approaches recommended are those of the

authors and not those of the National Kidney Foundation or Academic Press.
Great care has been taken by the authors and editors to maintain the accuracy
of the information contained herein. However, neither Academic Press nor
the National Kidney Foundation nor the editors and authors can be held
responsible for errors or for any consequences arising from the use of this
information.

This book is printed on acid -free paper.

Copyright 1998, 1994 by National Kidney Foundation
All Rights Reserved.
No part of this publication may be reproduced or transmitted in any form or by any
means, electronic or mechanical, including photocopy, recording, or any information
storage and retrieval system, without permission in writing from the publisher.

Academic Press
a division of Harcourt Brace & Company
525 B Street, Suite 1900, San Diego, California 92101-4495
http://www.apnet.com
Academic Press Limited
24-28 Oval Road, London NW1 7DX
http://www.hbuk.co.uk/ap/
Library of Congress Cataloging-in-Publication Data
Primer on kidney diseases / edited by Arthur Greenberg ... (et al.}.
p.
cm.
Includes index. ISBN
0-12-299090-0
1. Kidneys-Diseases. 2. Nephrology. I. Greenberg, Arthur,
date. II. National Kidney Foundation
[DNLM: 1. Kidney Diseases. 2. Kidney-Pathology. WJ 300 N2773
1994]
RC902.N28 1994
616.6' l- dc20
DNLM/OLC
for Library of Congress
94-26276
CIP
PRINTED IN CAN ADA
98 99 00 01

02 FR

CONTENTS
Contributors
ix
Foreword to the First Edition Foreword to the
xiii
Second Edition Preface to the First Edition Preface xv
to the Second Edition
xvii
xix

10. Metabolic Alkalosis

80

Edward R. Jones

11. Respiratory Acidosis and Alkalosis

SECTION I

86

Nicolaos E. Madias

STRUCTURE AND FUNCTION OF

THE KIDNEY AND THEIR
CLINICAL ASSESSMENT

12. Approach to Acid-Base Disorders

92

Martin Goldberg

13. Disorders of Potassium Metabolism

98

Michael Allan

I. Overview of Renal Function and Structure

14. Disorders of Calcium and Phosphorus

Homeostasis

3
Josephine P. Briggs, Wilhelm Kriz, and
Jurgen B. Schnermann

1. Clinical Evaluation of Renal Function

106

David A. Bushinsky

20

15. Edema and the Clinical Use of Diuretics

114

David H. Elliso n

Andrew S. Levey

3. Urinalysis

27

Arthur Greenberg

4. Hematuria

36

Clifford E. Kashtan

5. Proteinuria

42

Sharon Anderson

SECTION 3

GLOMERULAR DISEASES
16. Glomerular Clinicopathologic Syndromes

127

/. Charles Jennette and Ronald J. Falk

6. Renal Imaging Techniques

47
17. Immunopathogenesis of Renal Disease

Howard J. Mindell and Jonathan T. Fairbank

141

Diane M. Cibrik and John R. Sedor

18. Minimal Change Nephropathy

SECTION 2

ACID-BASE, FLUID, AND

ELECTROLYTE DISORDERS
7. Hyponatremia and Hypoosmolar Disorders

19. Membranoproliferative Glomerulonephritis

57

Daniel Batlle

20. Focal Segmental Glomerulosclerosis

160

Gerald B. Appel

64

21. Membranous Nephropathy

164

Sharon G. Adler and Cynthia C. Nast

Paul M. Palevsky

9. Metabolic Acidosis

153

Vivette D'Agati

Joseph G, Verbalis

8. Hypernatremia

149

Norman J. Siegel

71

22. IgA Nephropathy and Related Disorders

170

Bruce A. Julian

23. Goodpasture's Syndrome and Anti -GBM Disease

Caroline O.S. Savage

175

VI

Contents

SECTION 4

38. Cholesterol Atheroembolic Renal Disease

THE KIDNEY IN
SYSTEMIC DISEASE
24. Renal Function in Congestive Heart Failure

39. Pigment Nephropathy

273

James P. Knochel

183

40. Acute Interstitial Nephritis

277

Catherine M. Meyers

William T. Abraham and Robert W. Schrier

25. Renal Function in Liver Disease

269

Arthur Greenberg

188

Murray Epstein

41. Management of Acute Renal Failure

282

Florence N. Hutchison

26. Poststreptococcal and Other Infection-Related

Glomerulonephritides

193

SECTION 6

Asha Moudgil, Arvind Bagga, Ragini Fredrick, and

Stanley C. Jordan

27. Renal Involvement in Systemic Vasculitis

DRUGS AND THE KIDNEY

200

J. Charles Jennette and Ronald J. Falk

28. Renal Manifestations of Systemic Lupus

Erythematosus and Other Rheumatic
Disorders

42. Nonsteroidal Anti-inflammatory Drugs and the

Kidney

291

Vardaman M. Buckalew, Jr.

208

James E. Balow

43. Principles of Drug Therapy in Renal Failure

298

Arasb Ateshkadi

29. Diabetic Nephropathy

215

Julia Breyer

30. Dysproteinemias and Amyloidosis

SECTION 7

220

HEREDITARY RENAL DISORDERS

Paul W. Sanders

31. Renal and Urologic Complications of Cancer and

Its treatment

227

44. Sickle Cell Nephropathy

309

Jon I. Scheinman

Marc B. Garnick

32. Hemolytic Uremic Syndrome/Thrombotic

Thrombocytopenic Purpura

45. Polycystic and Acquired Cystic Diseases

232

Richard L. Siegler

46. Alport's Syndrome and Related Disorders

33. Renal Manifestations of HIV

237

Patricia Y. Schoenfeld

313

Patricia A. Gabow

318

Martin C. Gregory

47. Medullary Cystic Disease

323

Ellis D. Avner

SECTION 5

ACUTE RENAL FAILURE

SECTION 8
34. Pathophysiology of Acute Renal Failure

247

Robert Safirstein

35. Approach to the Patient with Acute Renal

Failure

253

Robert D. Toto

36. Renal Failure Caused by Therapeutic Agents

260
48. Tubulointerstitial Disease

Thomas M. Coffman

37. Acute Renal Failure Due to Metabolic

Derangements
Christof Westenfelder

TUBULOINTERSTITIAL
NEPHROPATHIES AND
DISORDERS OF THE
URINARY TRACT
329

William F. Finn

266

49. Lithium-Induced Renal Disease

Gregory L. Braden

332

Contents

VII

50. Lead Nephrotoxicity

335

65. Outcome of End Stage Renal Disease Therapies

422

Wendy E. Bloembergen

William F. Finn
51. Hyperoxaluria
Dawn S. Milliner

337

52. Medullary Sponge Kidney

343

66. Progression of Renal Disease

429

Thomas H. Hostetter
67. Management of the Patient with Chronic Renal
Disease

Ellis D. Avner
53. Renal Papillary Necrosis

345

433

Paul L. Kimmel
68. Nutrition and Renal Disease

Garabed Eknoyan
54. Obstructive Uropathy

348

440

Bradley J. Maroni
69. Renal Osteodystrophy and Other Musculoskeletal

Stephen M. Korbet
55. Vesicoureteral Reflux and Reflux Nephropathy

356

Jorge A. Velosa
56. Nephrolithiasis

360

Alan G. Wasserstein

Complications of Chronic Renal Failure

448

James A. Delmez
70. Cardiac Function and Cardiac Disease in
Renal Failure

455

Robert N. Foley and Patrick S. Parfrey

57. Urinary Tract Infections

366

Walter E. Stamm

71. Neurological Manifestations of Renal Failure

459

Cosmo L. Fraser

58. Renal and Urinary Tract Neoplasia

371

Robert R. Bahnson

72. Hematologic Manifestations of Renal Failure

465

Jonathan Himmelfarb
73. Endocrine Manifestations of Renal Failure

472

Eugene C. Kovalik

SECTION 9

74. The Evaluation of Prospective Renal

THE KIDNEY IN SPECIAL

CIRCUMSTANCES

Transplant Recipients

477

Bertram L. Kasiske
75. Renal Transplantation: Immunosuppression and

59. The Kidney in Infants and Children

383

R. Ariel Gomez

Postoperative Management

482

Douglas J. Norman

60. The Kidney in Pregnancy

388

Susan H. Hou
61. The Kidney in Aging

SECTION 11

395

HYPERTENSION

Jerome G. Porush

76. Pathogenesis of Hypertension

Stephen C. Textor

SECTION 10

77. Essential Hypertension

CHRONIC RENAL FAILURE AND

ITS THERAPY
62. The Uremic Syndrome

Joseph V. Natty, Jr.

78. Renovascular Hypertension
403

R. Vanholder
63. Hemodialysis and Hemofiltration

Laura P. Svetkey
79. Therapy of Hypertension

408

John M. Flack

491

496

501

506

Alfred K. Cheung
64. Peritoneal Dialysis
Beth Piraino

416

Index

577

CONTRIBUTORS

Numbers in parentheses indicate the pages on which the authors' contributions begin.

William T. Abraham, MD (183), Department of Medicine, University of Colorado School of Medicine, Denver, Colorado 80262
Sharon G. Adler, MD (164), Harbor-UCLA Medical Center,
Division of Nephrology, Torrance, California 90509
Michael Allon, MD (98), Division of Nephrology, University of
Alabama at Birmingham, Birmingham, Alabama 35233
Sharon Anderson, MD (42), Division of Nephrology and Hy pertension, Oregon Health Sciences University, Portland, Oregon 97201
Gerald B. Appel, MD (160), Columbia Presbyterian Medical
Center and Columbia University College of Physicians
and Surgeons, New York, New York 10032
Arasb Ateshkadi, PhD (298), Department of Pharmacy Practice, College of Pharmacy, University of Utah, Salt Lake
City, Utah 84112
Ellis D. Avner, MD (323,343), Department of Pediatrics, Case
Western Reserve University School of Medicine and
Rainbow Babies and Childrens Hospital of University
Hospitals of Cleveland, Cleveland, Ohio 44106
Arvind Bagga, MD (193), Department of Pediatric Nephrology,
Cedars-Sinai Medical Center, Los Angeles, California
90048
Robert R. Bahnson, MD (371), Division of Urology, Ohio State
University Medical Center, Columbus, Ohio 43201
James E. Balow, MD (208), NIDDK, National Institutes of
Health, Bethesda, Maryland 20892
Daniel Batlle, MD (71), Division of Nephrology/Hypertension,
Department of Medicine, Northwestern University Medical School, Chicago, Illinois 60611
Wendy E. Bloembergen, MD (422), University of Michigan, Ann
Arbor, Michigan 48103
Gregory L. Braden, MD (332), Renal Division, Baystate Medical Center, Springfield, Massachusetts 01199
Julia Breyer, MD (215), Division of Nephrology, Vanderbilt
University Medical Center, Nashville, Tennessee 37232
Josephine P. Briggs, MD (3), NIDDK, National Institutes of
Health, Bethesda Maryland 20892
Vardaman M. Buckalew, Jr., MD (291), Department of Internal
Medicine, Section on Nephrology, Bowman Gray School
of Medicine, Winston-Salem, North Carolina 27157

David A. Bushinsky, MD (106), University of Rochester School

of Medicine and Dentistry and Nephrology Unit, Strong
Memorial Hospital, Rochester, New York 14642
Alfred K. Cheung, MD (408), Division of Nephrology and Hy pertension, University of Utah School of Medicine, Salt
Lake City, Utah 84112
Diane M. Cibrik, MD (141), Department of Medicine, School of
Medicine, Case Western Reserve University, Cleveland,
Ohio 44109
Thomas M. Coffman, MD (260), Division of Nephrology, De partment of Medicine, Duke University and Durham
Veterans Affairs Medical Center, Durham, North Carolina 27705
Vivette D'Agati, MD (153), Columbia University College of
Physicians and Surgeons, New York, New York 10032
James A. Delmez, MD (448), Renal Division, Washington University School of Medicine, St. Louis, Missouri 63110
Garabed Eknoyan, MD (345), Department of Medicine, Baylor
College of Medicine, Houston, Texas 77030
David H. Ellison, MD (114), Section of Nephrology, Yale University School of Medicine, New Haven, Connecticut
06510
Murray Epstein, MD (188), University of Miami School of Medicine, Miami, Florida 33125
Jonathan T. Fairbank, MD (47), College of Medicine, University
of Vermont and Division of Nuclear Medicine, Fletcher
Allen Health Care, Burlington, Vermont 05401
Ronald J. Falk, MD (127,200), Department of Medicine, Division of Nephrology, University of North Carolina,
Chapel, Hill, North Carolina 27599
William F. Finn, MD (329,335), Division of Nephrology and
Hypertension, Department of Medicine, University of
North Carolina School of Medicine, Chapel Hill, North
Carolina 27599
John M. Flack, MD (506), Department of Medicine, Wayne
State University Medical School, Harper Hospital, Detroit, Michigan 48201
Robert N. Foley, MD (455), Division of Nephrology, Memorial
University of Newfoundland, St. John's, Newfoundland,
Canada A1B 3V6
Cosmo L. Fraser, MD (459), Department of Medicine, University of California, San Francisco, and Geriatric Nephrol-

Contributors

ogy, Veterans Affairs Medical Center, San Francisco,

California 94121
Ragini Fredrich, MD (193), Department of Pediatric Nephrology, Cedars -Sinai Medical Center, Los Angeles, California 90048
Patricia A. Gabow, MD (313), PKD Research, University of
Colorado Health Sciences Center, Denver, Colorado
80262
Marc B. Garnick, MD (227), Harvard Medical School, Brookline, Massachusetts 02146
Martin Goldberg, MD (92), Temple University School of Medicine, Philadelphia, Pennsylvania 19140
R. Ariel Gomez, MD (383), Department of Pediatrics, University of Virginia, Charlottesville, Virginia 22908
Arthur Greenberg, MD (27,269), Division of Nephrology, Department of Medicine, Duke University, Durham,
North Carolina 27705
Martin C. Gregory, MD (318), University of Wasatch Clinics,
Salt Lake City, Utah 84112
Jonathan Himmelfarb, MD (465), Division of Nephrology,
Maine Medical Center, Portland, Maine 04102
Thomas H. Hostetter, MD (429), Renal Division, University of
Minnesota, Minneapolis, Minnesota 55455
Susan H. Hou, MD (388), Section of Nephrology, Rush Medical
College, Chicago, Illinois 60612
Florence N. Hutchison, MD (282), Medical University of South
Carolina and Ralph H. Johnson Department of Veterans
Affairs Medical Center, Charleston, South Carolina
29403
J. Charles Jennette, MD (127,200), Department of Pathology
and Laboratory Medicine, University of North Carolina,
Chapel Hill, North Carolina 27599
Edward R. Jones, MD (80), The Germantown Hospital and
Medical Center, Philadelphia, Pennsylvania 19144
Stanley C.Jordan, MD (193), Department of Pediatric Nephrology, Cedars -Sinai Medical Center, Los Angeles, California 90048
Bruce A. Julian, MD (170), Department of Medicine, Division
of Nephrology, University of Alabama at Birmingham,
Birmingham, Alabama 35294
Clifford E. Kashtan, MD (36), Division of Pediatric Nephrology,
Department of Pediatrics, University of Minnesota Medical School, Minneapolis, Minnesota 55455
Bertram L Kasiske, MD (477), Division of Nephrology, Department of Medicine, University of Minnesota School of
Medicine, Minneapolis, Minnesota 55415

Eugene C. Kovalik, MD (472), Division of Nephrology, Duke

University Medical Center and Veterans Affairs Medical
Center, Durham, North Carolina 27710
Wilhelm Kriz, MD (3) Institute for Anatomy and Cell Biology,
University of Heidelberg, Heidelberg 1, Germany
Andrew S. Levey, MD (20), Tufts University School of Medicine and New England Medical Center, Boston, Massachusetts 02111
Nicolaos E. Madias, MD (86), Tufts University School of Medicine and Division of Nephrology, New England Medical
Center, Boston, Massachusetts 02111
Bradley J. Maroni, MD (440), Renal Division, Department of
Medicine, Emory University School of Medicine, Atlanta, Georgia 30322
Catherine M. Meyers, MD (277), Renal-Electrolyte Division,
Department of Medicine, University of Pennsylvania
School of Medicine, Philadelphia, Pennsylvania 19104
Dawn S. Milliner, MD (337), Division of Nephrology, Mayo
Clinic, Rochester, Minnesota 55905
Howard J. Mindell, MD (47), College of Medicine, University of
Vermont and Division of Diagnostic Radiology, Fletcher
Allen Health Care, Burlington, Vermont 05401
Asha Moudgil, MD (193), Department of Pediatric Nephrology, Cedars -Sinai Medical Center, Los Angeles, California 90048
Joseph V. Nally Jr., MD (496), Department of Nephrology and
Hypertension, The Cleveland Clinic Foundation, Cleveland, Ohio 44195
Cynthia C. Nast, MD (164), University of California at Los
Angeles and Cedars -Sinai Medical Center, Los Angeles,
California 90509
Douglas J. Norman, MD (482), Oregon Health Sciences University, Portland, Oregon 97201
Paul M. Palevsky, MD (64), University of Pittsburgh School of
Medicine and Veterans Affairs Pittsburgh, Healthcare
System, Pittsburgh, Pennsylvania 15204
Patrick S. Parfrey, MD (455), Division of Nephrology, Memo rial University of Newfoundland, St. John's, Newfoundland, Canada A1B 3V6
Beth Piraino, MD (416), Renal-Electrolyte Division, Department of Medicine, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania 15213
Jerome G. Porush, MD (395), Division of Nephrology and Hy pertension, Brookdale University Hospital and Medical
Center 11212
Robert Safirstein, MD (247), Division of Nephrology, University of Texas Medical Branch, Galveston, Texas 77555

James P. Knochel, MD (273), Department of Internal Medicine,

Presbyterian Hospital, Dallas, Texas 75231

Paul W. Sanders, MD (220), Division of Nephrology, University of Alabama at Birmingham, Birmingham, Alabama 35233
Caroline O.S. Savage, MD (175), Renal Unit, The Queen Elizabeth Hospital, Edgbaston, Birmingham B15 2TH,
United Kingdom

Stephen M. Korbet, MD (348), Section of Nephrology, Rush

Medical College, Chicago, Illinois 60612

Jon I. Scheinman, MD (309), Pediatric Nephrology, Medical

College of Virginia, Richmond, Virginia 23298

Paul L Kimmel, MD (433), Department of Medicine, George

Washington University Medical Center, Washington,
DC 20037

Contributors

Jurgen B. Schnermann, MD (3), Department of Physiology, University of Michigan Medical School, Ann Arbor, Michigan 48103
Patricia Y. Schoenfeld, MD (237), University of California Renal
Center at San Francisco General Hospital, San Francisco,
California 94110
Robert W. Schrier, MD (183), Department of Medicine, University of Colorado School of Medicine, Denver, Colorado 80262
John R. Sedor, MD (141), Department of Medicine, School of
Medicine, Case Western Reserve University, MetroHealth Medical Center, Cleveland, Ohio 44109
Norman J. Siegel, MD (149), Department of Pediatrics, Yale
University School of Medicine, New Haven, Connecticut 06520
Richard L. Siegler, MD (232), University of Utah School of
Medicine, Salt Lake City, Utah 84132
Walter E. Stamm, MD (366), Division of Allergy and Infectious
Diseases, University of Washington, Seattle, Washington 98195
Laura P. Svetkey, MD (501), Duke Univers ity Medical Center,
Durham, North Carolina 27705

XI

Stephen C. Textor, MD (491), Mayo Medical School and Division of Hypertension, Mayo Clinic, Rochester, Minnesota 55905
Robert D. Toto, MD (253), Department of Internal Medicine,
Division of Nephrology, Va nderbilt University Medical
Center, Nashville, Tennessee 37232
Prof. Dr. R. Vanholder (403), Department of Internal Medicine,
Nephrology Unit, University Hospital, Gent, B-9000
Belgium
Jorge A. Velosa, MD (356), Division of Nephrology, Mayo
Clinic, Rochester, Minnesota 55905
Joseph G. Verbalis, MD (57), Division of Endocrinology and
Metabolism, Georgetown University, Washington, DC
20007
Alan G. Wasserstein, MD (360), Renal Electrolyte and Hypertension Division, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania 19104
Christof Westenfelder, MD (266), Division of Nephrology, University of Utah and Veterans Affairs Medical Center,
Salt Lake City, Utah 84148

FOREWORD TO THE FIRST EDITION

more than one occasion the possibility of publishing a

01primer
on kidney diseases has been entertained by the

of gratitude is due to all the authors for their valuable

contributions to the Primer. For bringing everyone's effort
to its final fruition we are grateful to our publishers, Academic Press, Inc., and especially to Jasna Markovac, Editorin-Chief, who was kind enough to act as our Biomedical
Editor throughout the developmental phase of the Primer.
A word of explanation concerning the cover design: the
frame is based on the symbols for the elements developed
by John Dalton at the beginning of the nineteenth century.
The four corner symbols, clockwise, beginning in the upper
left corner represent sodium, phosphorus, potassium, and
magnesium. The symbols in the boxes framing the cover,
beginning from each corner, are those of hydrogen, oxygen,
ammonia, carbon dioxide, and carbon. The figures on the
cover are reproductions from the drawings of Leonardo
da Vinci. The background is his classic drawing of man in
equilibrium, befitting for a text on the kidney whose ultimate function is to maintain equilibrium in man by maintaining the "milieu interieur." Superimposed is a drawing
of a scarred left kidney and atrophied right kidney from
Leonardo's Anatomical Notebooks, yet another befitting
diagram for a text on kidney diseases.

professional membership of the National Kidney

Foundation. On none of those occasions did the project
progress beyond a consideration of such a notion. It was
only some three years ago, when Dr. Arthur Greenberg
submitted a written proposal for a nephrology primer to
the Scientific Advisory Board of the National Kidney
Foundation that action on this project was initiated. That it
got into motion is entirely due to his perseverance and
dedication. That the primer is now a reality is due to the
commitment and leadership provided by its Editor, Arthur
Greenberg, and his Associate Editors: Alfred K. Cheung,
Thomas M. Coffman, Ronald J. Falk, and J. Charles
Jennette. The work of the Consulting Editors: Joel D.
Kopple, Neil Kurtzman, Manuel Martinez-Maldonado,
and Shaul G. Massry has been relatively easier, but nevertheless demanding and extremely important to providing
an independent external review process to the text. On
behalf of the National Kidney Foundation I thank the Editors and Consulting Editors for their time and effort in
making the Primer on Kidney Diseases a reality. A debt

Garabed Eknoyan
1994

XIII

FOREWORD TO THE SECOND EDITION

he reservation with which work on the First Edition

of the Primer on Kidney Diseases was begun is now
history. The successful reception it received has
been a rewarding experience for everyone involved in its
production. Thus, it was with greater confidence and enthusiasm that work on the Second Edition was undertaken.
The Editors were gracious enough to accept the challenge
for a revised and updated Primer. They are the only constants of this new edition. To provide continuity, while at
the same time allowing for new input, it was decided from
the outset to invite new contributors to author at least one
third of the chapters. The same rule was applied to the
Consulting Editors. The change was deemed necessary to
maintain the vitality of the Primer, and this policy will be
continued in the future. A very special note of thanks is
due to the former authors and Consulting Editors who
contributed so much to the success of the First Edition.

All of the improved features of the Second Edition reflect

the creativity and commitment of the Editor, Arthur
Greenberg, and his Associate Editors: Alfred K. Cheung,
Thomas M. Coffman, Ronald J. Falk, and J. Charles Jennette. As with the First Edition, each of the chapters was
reviewed by the Editors, as well as one of the Consulting
Editors: James V. Donadio, Garabed Eknoyan, Shaul G.
Massry, William E. Mitch, and Sandra Sabatini. On behalf
of the National Kidney Foundation, I thank the Editors
and Consulting Editors for their diligent effort in bringing
the Second Edition to fruition. A special thanks goes to
returning and new authors of the Primer. Without their
contributions, the revision would not have been possible.
The support received from the Publishers, and particularly
that of Jasna Markovac, has been invaluable. They have
been as understanding, cooperative, and responsive as anyone could wish from a publisher.
Garabed Eknoyan

PREFACE TO THE FIRST EDITION

the past twenty years, at least a half dozen

During
excellent multivolume texts in nephrology or renal

overly technical citations. Reference to these reviews and

the multivolume nephrology texts will direct the interested
reader to primary sources.
Chapters were thoroughly reviewed before acceptance.
Each was read by a member of the Editorial Board, by the
Editor, and then by a member of the National Kidney
Foundation Scientific Advisory Board serving as a Consulting Editor. Authors were then asked to revise their manuscripts and the review process was repeated. We are grateful for the authors' efforts to distill the essential content
into the smallest possible space and for their unfailing cooperation with the entire editorial process. Although a general
consensus was not always achievable, we believe that the
chapters all reflect views accepted by a broad range of nephrologists.
In keeping with its overall goals, the National Kidney
Foundation has a specific educational mission: to encourage
"dissemination of both existing and newly acquired knowledge to the professional community." The editors hope
that in serving this purpose, the Primer will facilitate the
care of patients with kidney disease. Perhaps it will also
serve as an invitation to students and medical residents to
look more closely at the specialty that focuses on kidney
disease, nephrology.
I am most grateful to the Consulting Editors and, especially, to the Co-Editors for their enthusiasm and commitment at every stage of this pro ject. Finally, imitation is the
sincerest form of flattery. As a student and house officer,
I found the Primer on the Rheumatic Diseases (now in its
tenth edition) an excellent summary of that field. Drs. H.
Ralph Schumacher, Jr., and Arthur Grayzell of the Arthritis Foundation generously provided many suggestions that
were very helpful in initiating work on this first Primer on
Kidney Diseases. The Editors and the National Kidney
Foundation hope this new book will prove equally handy,
and they welcome your suggestions for improvement of
future editions.

physiology have appeared, testifying to the expansion of

clinical and basic science information in renal medicine.
Many books covering fluid and electrolyte disorders,
dialysis, or renal transplantation have also been published
during this time. Collectively, they function superbly as
reference works for nephrologists and nephrology fellows.
They are less useful students or busy clinicians. The purpose
of this book is different. Like any primer, it is an introductory textbook. It is targeted at medical students, house
officers, and primary care physicians. For them, we hope to
provide a fresh, succinct, and accessible but comprehensive
overview of clinical nephrology.
The Primer begins with a description of the normal kidney and of techniques used to evaluate kidney function
and anatomy. Subsequent sections cover the glomerular
diseases, tubulointerstitial diseases, acute renal failure,
chronic renal failure, and hypertension. Reviews of dialysis
and renal transplantation offer information useful to the
non-nephrologist called upon to care for patients approaching end-stage renal disease or patients with stable functioning renal allografts. The final chapters consist of practical
discussions of acid-base and electrolyte disorders, stressing
the pragmatic application of renal physiology to clinical
problems.
Contributors to the Primer were asked to adhere to strict
page limits, chosen to reflect the relative importance of the
topic and the depth of the pertinent medical literature.
Determining the proper length was an imprecise business.
Some chapters may appear longer than warranted by the
prevalence of the disease condition they describe, because
we did not want to substitute brevity for clarity. Additional
details may be obtained from the suggested readings that
follow each chapter. These sources are just that; we thought
recent reviews and key primary articles would be more
valuable than a detailed listing of potentially obscure or

Arthur Greenberg
1994

xvii

PREFACE TO THE SECOND EDITION

are gratified by the warm response to the

ThefirstEditors
edition of the Primer. The book is too big to fit in

the cooperation of Jasna Markovac and Tari Paschall at

Academic Press, we have been able to render the photomicrographs of the urine sediment and renal pathology in
color.
To keep the Primer fresh, it was our original intent to
rotate authors periodically. Accordingly, roughly one third
of the chapters have new authors. The Editors gratefully
acknowledge the hard work of these new contributors and
the efforts of returning authors, all of whom have updated
the text and references of their chapters. All authors, old
and new, cooperated patiently and graciously with the
Primer's editorial process, which is also unchanged. Each
chapter was read by the Editor, an Associate Editor, and
a member of the NKF Scientific Advisory Board acting as
a Consulting Editor. Requests for revision were solicited,
and the review process was repeated before chapters
were accepted.
I want to acknowledge the support of Gary Eknoyan,
who this year serves as NKF President. With Shaul Massry,
he was instrumental in launching the first edition and
guided this revision as well. The assistance of three new
Consulting Editors, Sandra Sabatini, James Donadio, and
William Mitch, was invaluable, as was the assistance of
both veterans. I am again indebted to Alfred Cheung, Tom
Coffman, Ron Falk, and Charles Jennette for their critical
insights, estimable advice, and hard work.

the pocket of a white coat, but many more students and

house officers than expected are carrying it and reading it
while rotating on nephrology services. Our goals for this
new edition, which appears 3 years after the original, have
not changed. The book is still targeted at students, house
staff, and practitioners, for whom we have tried to provide
a comprehensive, accessible, and pragmatic summary of
the management of renal diseases and fluid and
electrolyte disorders.
Readers of the first edition will notice some modifications. The book is now divided into 11 sections, covering
renal function and its assessment, electrolyte disorders,
glomerular disease, the kidney in systemic disease, acute
renal failure, drugs and the kidney, hereditary renal dis eases, tubulointerstitial diseases, the kidney in special circumstances, chronic renal disease, and hypertension. Two
chapters dealing with the characteristics of kidney function
in the very young and the very old have been added, along
with a separate chapter introducing the tubulointerstitial
diseases. The chapter serving as an overview of glomerular
diseases now includes an outstanding series of schematic
diagrams of glomerular pathology contributed by Charles
Jennette. Analgesic abuse nephropathy and the effects of
NSAIDs on the kidney are now covered together. With

Arthur Greenberg

xix

SECTION
STRUCTURE AND FUNCTION
OF THE KIDNEY AND THEIR
CLINICAL ASSESSMENT

OVERVIEW OF RENAL FUNCTION

AND STRUCTURE
JOSEPHINE P. BRIGGS, WILHELM KRIZ, AND JURGEN B. SCHNERMANN

BASIC CONCEPTS

The Kidney and Homeostasis

Functions of the Kidney

Numerous functions of the body proceed optimally only

when body fluid composition and volume are maintained
within an appropriate range. For example:

The main functions of the kidneys can be categorized

as follows:
1. Maintenance of body composition. The volume of
fluid in the body, its osmolarity, electrolyte content
and concentration, and its acidity are all regulated
by the kidney by variation in urine excretion of
water and ions. Electrolytes regulated by changes
in urinary excretion include Na + , K+ , Cl+ , Ca + ,
Mg2+, and PO+ 4.
2. Excretion of metabolic endproducts and foreign
substances. The kidney excretes a number of
products of metabolism, most notably urea, and a
number of toxins and drugs.
3. Production and secretion of enzymes and
hormones.
a. Renin is an enzyme produced by the granular
cells of the juxtaglomerular apparatus and
catalyzes the formation of angiotensin from a
plasma globulin, angiotensinogen. Angiotensin is
a potent vasoconstrictor peptide and contributes
importantly to salt balance and blood pressure
regulation.
b. Erythropoietin, a glycosylated, 165-amino-acid
protein produced by renal cortical interstitial
cells, stimulates the maturation of erythrocytes
in the bone marrow.
c. 1,25-Dihydroxyvitamin D3, the most active form
of vitamin D 3, is formed by proximal tubule
cells . This steroid hormone plays an important
role in the regulation of body calcium and
phosphate balance.
In later chapters of this Primer, the pathophysiological
mechanisms and consequences of derangements in kidney
function are discussed in detail. This chapter reviews the
basic anatomy of the kidney and the normal mechanisms
for urine formationglomerular filtration and tubular
transport.

Primer on Kidney Diseases, Second Edition

Cardiac output and blood pressure are dependent on

optimum plasma volume.
Most enzymes function best over rather narrow
ranges of pH or ion concentration.
Cell membrane potentials depend on K+
concentration.
Membrane excitability depends on Ca +
concentration.
The principal job of the kidneys is the correction of perturbations in the composition and volume of body fluids that
occur as a consequence of food intake, metabolism, environmental factors, and exercise. Typically, in healthy people, such perturbations are corrected within a matter of
hours, so that in the long term, body fluid volume and
the concentration of most ions do not deviate much from
normal set points. In many disease states, however, these
regulatory processes are disturbed, resulting in persistent
deviations in body fluid volumes or ionic concentrations.
Understanding these disorders requires an understanding
of the normal regulatory processes.
The Balance Concept

The maintenance of stable body fluid composition requires that appearance and disappearance rates of any substance in the body balance each other. Balance is
achieved when:
Ingested amount f Produced amount = +
Excreted amount Consumed amount.
For a large number of organic compounds, balance is the
result of metabolic production and consumption. However,
electrolytes are not produced or consumed by the body;
balance is achieved by adjusting excretion to match intake.
Therefore, when a person is in balance for sodium, potas sium, and other ions, the amount excreted must equal the
amount ingested. Since the kidneys are the principal organs

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

Briggs, Kriz, and Schnermann

where regulated excretion takes place, urinary excretion of

such solutes closely follows the dietary intake. A central
theme of physiology of the kidneys is understanding the
mechanisms by which urine composition is altered to maintain the body in balance.
Body Fluid Composition

To a large extent, humans are composed of water. Adipose tissue is low in water content; thus, in obese people,
the fraction of body weight that is water is lower than that
in lean individuals. As a consequence of slightly greater
fat content, women, on the average, contain less water than
men, about 55% instead of 60%. Useful round numbers to
remember for bedside estimates of body fluid volumes are
given in Table 1. Typical ionic compositions of the intracellular and extracellular fluid compartments are given in Table 2.

TABLE 2

Typical Ionic Composition of Plasma and Intracellular Fluid

Plasma
(mEq/L)
Cations
K+
Na+
Ca2+
(ionized)
Mg2+
Total cations
Anions

cr

HCOj
Phosphates
Protein
Other
Total anions

Intracellular
fluid
(mEq/L)

4
143

140
12

2
1
150 mEq/L

0.001
38
190 mEq/L

104
24
2 14

4
12
40
50
_88
190 mEq/L

6
150 mEq/L

KIDNEY STRUCTURE

The kidneys are two bean-shaped organs lying in the

retroperitoneal space, each weighing about 150 g. The kidney is an anatomically complex organ, consisting of many
different types of highly specialized cells, arranged in a
highly organized three-dimensional pattern. The functional
unit of the kidney is called a nephron (there are approximately 1 million nephrons in one human kidney); each
nephron consists of a glomerulus and a long tubule which
is made of a single layer of epithelial cells (the nephron is
depicted schematically in Fig. 1). The nephron is segmented
into distinct parts proximal tubule, loop of Henle, distal
tubule, collecting ducteach with a typical cellular appearance and special functional characteristics.
The nephrons are packed tightly together to make up
the kidney parenchyma, which can be divided into regions.
The outer layer of the kidney is called the cortex; it contains
all the glomeruli, much of the proximal tubule, and some
of the more distal portions as well. The inner section, called
the medulla, consists largely of the parallel arrays of the
loops of Henle and collecting ducts. The medulla is formed
into cone-shaped regions, called pyramids (the human kid-

TABLE I

Bedside Estimates of Body Fluid Compartment Volumes

Example for 60-kg patient
Remember
Total body water = 60% x
60% x 60 kg = 36 L 2/3 X 36 L
body wt Intracellular water
= 2/3 total
= 24 L 1/3 x 36 L = 12 L 1/4 X
body water Extracellular
water = 1/3 total
12 L = 3 L 3 L H - (1-0.40) =
body water Plasma
water = 1/4
6.6 L
extracellular water
Blood volume = Plasma water
1-Hct

ney typically has 7 to 9), which extend into the renal pelvis.
The tips of the medullary pyramids are called papillae. The
medulla is important for concentration of the urine; the
extracellular fluid in this region of the kidney has much
higher solute concentration than plasmaas much as four
times higher, with highest solute concentrations reached
at the papillary tips.
The process of urine formation begins in the glomerular
capillary tuft, where an ultrafiltrate of plasma is formed.
The filtered fluid is collected in Bowman's capsule and
enters the renal tubule to be carried over a circuitous
course, successively modified by exposure to the sequence
of specialized tubular epithelial segments with different
transport functions. The proximal convoluted tubule, which
is located entirely in the renal cortex, absorbs approximately two-thirds of the glomerular filtrate. Fluid remaining at the end of the proximal convoluted tubule enters
the loop of Henle, which dips down in a hairpin configuration into the medulla. Returning to the cortex, the tubular
fluid passes close by its parent glomerulus at the juxtaglomerular apparatus, then enters the distal convoluted tubule
and finally the collecting duct, which courses back through
the medulla, to empty into the renal pelvis at the tip of
the renal papilla. Along the tubule, most of the glomerular
filtrate is absorbed, but some additional substances are
secreted. The final product, the urine, enters the renal pelvis
and then the ureter, collects in the bladder, and is finally
excreted from the body.
RENAL CIRCULATION
Anatomy of the Circulation

The renal artery, which enters the kidney at the renal

hilum, carries about one-fifth of the cardiac output; this
represents the highest tissue-specific blood flow of all larger
organs in the body (about 350 mL/min 100 g tissue). As a

Overview of Renal Function and Structure

os ; is , -

FIGURE I Organization of the nephron. The human kidney is

made up of a million nephrons, two of which are shown schematically here. Each nephron consists of the following parts: glomerulus
(1), proximal convoluted tubule (2), proximal straight tubule (3),
thin descending limb of the loop of Henle (4), thin ascending limb
(5), thick ascending limb (6), macula densa (7), distal convoluted
tubule (8), and connecting tubule (9). Several nephrons coalesce
to empty into a collecting duct, which has three distinct regions:
the cortical collecting duct (10), the outer medullary collecting
duct (11), and the inner medullary collecting duct (12). As shown,
the deeper glomeruli give rise to nephrons with loops of Henle
which descend all the way to the papillary tips, while the more
superficial glomeruli have loops of Henle that bend at the junction
between the inner and outer medulla.

consequence of this generous perfusion, the renal arteriovenous O2 difference is much lower than that of most other
tissues (and blood in the renal vein is noticeably redder in
color than that in other veins). The renal artery bifurcates
several times after it enters the kidney and then breaks
into the arcuate arteries, which run, in an arch-like fashion,
along the border between the cortex and the outer medulla.
As shown in Fig. 2, the arcuate vessels give rise, typically
at right angles, to interlobular arteries, which run to the
surface of the kidney. The afferent arterioles supplying the
glomeruli come off the interlobular vessels.
Two Capillary Beds

The renal circulation is unusual in that it breaks into

two separate capillary beds: the glomerular bed and the
peritubular capillary bed. These two capillary networks are

FIGURE 2 Organization of the renal vascular system. The renal

artery bifurcates soon after entering the kidney parenchyma and
gives rise to a system of arched-shaped vessels that run along the
border betweeen the cortex and the medulla. In this diagram, the
vascular elements surrounding a single renal pyramid are shown.
The human kidney typically has seven to nine renal pyramids.
Here the arterial supply and glomeruli are shown in black, and
the venous system is shown in grey. The peritubular capillary
network which arises from the efferent arterioles is omitted, for
simplicity. The vascular elements are named as follows: interlobar
artery and vein (1 and la), arcuate artery and vein (2 and 2a),
interlobular artery and vein (3 and 3a), stellate vein (4), afferent
arteriole (5), efferent arteriole (6), glomerular capillaries (7), jux tamedullary efferent arteriole, supplying descending vasa recti (8),
and ascending vasa recti (9).

arranged in series so that all the renal blood flow passes

through both. As blood leaves the glomerulus, the capillaries coalesce into the efferent arteriole, but almost immediately the vessels bifurcate again to form the peritubular
capillary network. This second network of capillaries is the
site where tubular reabsorbate is returned to the circulation. Pressure in the first capillary bed, that of the glomerulus, is rather high (about 40 to 50 mm Hg), while pressure
in the peritubular capillaries is similar to that in capillary
beds elsewhere in the body (about 5 to 10 mm Hg).
About 25% of the plasma that arrives at the glomerulus
passes through the filtration barrier to become the filtrate.
Blood cells, most of the proteins, and about 75% of the
fluid and small solutes stay in the capillary and leave the
glomerulus via the efferent arteriole. This postglomerular
blood, which has a relatively high concentration of protein

Briggs, Kriz, and Schnermann

and red cells, enters the peritubular capillaries, where the

high oncotic pressure from the high protein concentration
facilitates the reabsorption of fluid. The peritubular capillaries coalesce to form venules and eventually the renal
vein.
Medullary Blood Supply

The blood supplying the medulla is also postglomerular:

specialized peritubular vessels, called vasa recta, arise from
the efferent arterioles of the glomeruli nearest the medulla
(the juxtamedullary glomeruli). Like medullary renal tubules, these vasa recta form hairpin loops dipping into
the medulla.

ulus is a ball of capillaries, consisting of endothelial cells,

and surrounded by specialized epithelial cells. Directly adherent to the basement membrane that surrounds the capillary loops is an inner layer of epithelial cells called the
glomerularpodocytes. These are large, highly differentiated
cells that form an array of lace-like foot processes over the
outer layer of these capillaries. An outer epithelial capsule,
called Bowman's capsule, acts as a pouch to capture the
filtrate and direct it into the beginning of the proximal
tubule. As shown in the accompanying figures, the capillaries are held together by a stalk of cells, called the glomeru lar mesangium.
Glomerular Filtration Barrier

GLOMERULUS
Structure

The structure of the glomerulus is shown schematically

in Fig. 3 and in the photomicrographs in Fig. 4. The glomer-

Urine formation begins at the glomerular filtration barrier. The glomerular filter through which the ultrafiltrate
has to pass consists of three layers: the fenestrated endothelium, the intervening glomerular basement membrane, and
the podocyte layer (Fig. 5). This complex "membrane" is
freely permeable to water and small dissolved solutes, but
retains most of the proteins and other larger molecules, as
well as all blood particles. The main determinant of passage
through the glomerular filter is molecular size. A molecule
like inulin (5 kDa) passes freely through the filter, and
even a small protein like myoglobin (16.9 kDa) is filtered
to a large extent. Substances of increasing size are retained
with increasing efficiency until at a size about 60 to 70 kDa
the amount filtered becomes very small. Filtration also
depends on ionic charge, and negatively charged proteins,
such as albumin, are retained to a greater extent than would
be predicted by size alone. In certain glomerular diseases,
proteinuria develops because of loss of this charge selectivity.
Ultrafiltration in the Glomerulus

GBM

Filtrate formation in the glomerulus is governed by the

same forces, often called Starling forces, which determine
fluid transport across blood capillaries in general. The glomerular filtration rate (GFR) is equal to the product of
the net filtration pressure, the hydraulic permeability, and
the filtration area,
-UP

FIGURE 3 Anatomy of the glomerulus and juxtaglomerular apparatus. Schematic diagram of a section of a glomer ulus and its
juxtaglomerular apparatus. Structures shown are as follows: afferent arteriole (AA), efferent arteriole (EA), macula densa (MD),
distal tubule (D), juxtaglomerular granular cell (G), sympathetic
nerve endings (N), mesangial cell (M), extraglomerular mesangial
cell (EGM), endothelial cell (E), epithelial podocyte (PO), with
foot process (F), parietal epithelial cell (PE), glomerular basement
membrane (GBM), urinary space (US), urinary pole (UP), and
proximal tubule (P).

GFR = Lp X Area X Pn e t,
where Lp is the hydraulic permeability, and ,Pnet is the
net ultrafiltration pressure. Net ultrafiltration pressure or
effective filtration pressure is the difference between the
hydrostatic and oncotic pressure difference across the capillary loop,
P n e t = AF -

ATT

= (PG C - P B ) - n o c - !!),

where P is hydrostatic pressure, II is oncotic pressure, and

the subscripts GC and B refer to the glomerular capillaries
and Bowman's space.

I. Overview of Renal Function and Structure

FIGURE 4 Structure of the glomerulus. (a) A light micrograph of a glomerulus, showing the afferent arteriole (A),
efferent arteriole (E), macula densa (M), Bowman's capsule (BC), and beginning of the proximal tubule (PT). The
typical diameter of a glomerulus is about 100 to 150 /u,m, which is just barely visible to the naked eye. (b) Higher power
view of glomerular capillary loops, showing the epithelial podocyte (P), endothelial cells (E) and mesangial cells (M).

Changes in GFR can result from changes in the permeability/surface area product (Lp x Area) or from changes
in net ultrafiltration pressure. One factor influencing Pne,
is the resistance in the afferent and efferent arterioles.
An increase in resistance in the afferent arteriolar (before
blood gets to the glomerulus) will decrease PGC and GFR.
However an increase in resistance as blood exits through
the efferent arteriole will tend to increase Poc and GFR.
Changes in Pnel can also occur as a result of an increase
in renal arterial pressure, which will tend to increase PGC
and GFR. Obstruction of the tubule will increase PB and
decrease GFR, and a decrease in plasma protein concentration will tend to increase GFR.
Determination of GFR

GFR is measured by determining the urinary excretion

of a marker substance that must fulfill the key requirement
that the amount filtered per minute is equal to the amount
excreted in the urine per minute. This requirement is met
if this substance (1) is neither absorbed nor secreted by the
renal tubules, (2) is freely filterable across the glomerular
membranes, and (3) is not metabolized or produced by the
kidneys. These critical properties are ideally met by inulin,
a large sugar molecule with a molecular weight of about
5000. Inulin is often infused in experimental studies to
measure GFR. An endogenous substance that has similar
properties and is used in the clinical setting is creatinine.
As shown in Table 3, the formula for GFR derives from
a simple rearrangement of the statement indicating that
the amount filtered per minute equals the amount excreted
per minute. A formula of this general form, called the
clearance formula, denotes the volume of plasma (mL/
min) cleared of a particular substance by excretion into
the urineits clearance rate. In the case of inulin, the
clearance of inulin is equal to GFR. The clearance of creatinine is slightly'greater than GFR (15 to 20%) because the

excreted amount exceeds the amount filtered as a result

of some tubular secretion of creatinine. GFR is typically about 100 mL/min for women and 120 mL/min
for men.
Juxtaglomerular Apparatus

Tightly adherent to every glomerulus, in between the

entry and the exit of the arterioles, is a plaque of distal
tubular cells called the macula densa, which is part of the
Juxtaglomerular apparatus. This cell plaque is in the distal
tubule, at the very terminal end of the thick ascending limb
of the loop of Henle, right before the transition to the
distal convoluted tubule. This is a special position along
the nephron, because at this site NaCl concentration is
quite variable. Low rates of flow result in a very low salt
concentration at this site, 15 mEq/L or less, while at higher
flow rates salt concentration rises to 40 to 60 mEq/L. NaCl
concentration at this site regulates glomerular blood flow,
through a mechanism called tubuloglomerular feedback;
increases in salt concentration cause a decrease in glomerular blood flow.
The other unique cells that make up the Juxtaglomerular
apparatus are the renin-containing Juxtaglomerular granular cells. Renin secretion is also regulated locally by salt
concentration in the tubule at the macula densa. In addition, the granular cells have extensive sympathetic innervation, and renin secretion is controlled by the sympathetic
nervous system.
TUBULAR FUNCTION: BASIC PRINCIPLES
Absorption and Secretion in the Renal Tubules

The glomerular filtrate undergoes a series of modifications before becoming the final urine. These changes consist
of removal or absorption and addition or secretion of sol-

Briggs, Kriz, and Schnermann

Podocyte Prob

FIGURE 5 Structure of the glomerular capillary loop and the filtration barrier, (a) A
single capillary loop showing the endothelial and foot process layers and the attachments
of the basement membrane to the mesangium. Pressure in the glomerular capillary bed
is substantially higher than in other capillaries. As shown in the diagrammatic insert, the
mesangium provides the structural supports which permit these cells to withstand these
high pressures, (b) The glomerular filtration barrier.

utes and fluid, absorption and secretion indicate directions

of transport, not mechanisms.
1. Absorption. Absorption, the movement of solute or
water from tubular lumen to blood, is the
predominant process in the renal handling of Na + ,
Cl~, H2O, HCO 3~ glucose, amino acids, protein,
PO4+ , Ca + , Mg2+, urea, uric acid, and others.

2. Secretion. Secretion, the movement of solute from

blood or cell interior to tubular lumen, is important
in the renal handling of H+ , K+ , NH 4+ , and a number
of organic acids and bases.
Substances can move into or out of the tubule either by
the transcellular pathway, which requires traversing the
luminal and the basolateral cell membranes, or by the para-

I. Overview of Renal Function and Structure

TABLE 3

Tubular fluid

Derivation of the Formula for GFR

Step 1:

Step 2:

Step 3:

Filtered amount of inulin = Excreted amount of

inulin
( GFR GFR in) = (U- m - V)
where GFR is glomerular filtration rate,
GFR in is the concentration of inulin in the
filtrate
C/i,, is the concentration of inulin in urine V is
the urine flow rate
Inulin is freely filterable, so its concentrations in
plasma and filtrate are identical. Hence
Filtered amount of inulin = GFR Ptn where
/"] is the concentration of inulin in plasma.
Rearranging, this produces
G F R P i n = U i n V and
hence
U- X V
G F R = ^ 2 - -

cellular pathway between cells (Fig. 6). Many specialized

membrane proteins participate in the movement of substances across cell membranes along the renal tubule. Some
of the important membrane transport mechanisms, together with examples of substances that use these mechanisms, and proteins that are important for these processes
are given in Table 4.

Paracellular
pathway

Transcellular
pathway

Tight,
junction

Apical cell
membrane

Segmentation of the Nephron

One of the more striking characteristics of the renal

tubule is dramatic cellular heterogeneity. Early renal anatomists recognized that there are marked differences in the
appearance of the cells of the proximal tubule, loops of
Henle, and distal tubule. These different nephron segments
also differ markedly in function, distribution of important
transport proteins, and responsiveness to drugs such as
diuretics that inhibit transport.
K+
i

Proximal Tubule

The proximal tubules absorb the bulk of filtered small

solutes. These solutes are present in proximal tubular fluid
at the same concentration as in plasma. Approximately
60% of the filtered Na+ , Cl~, K+ , Ca+ , and water and more
than 90% of the filtered HCO3~ are absorbed along the
proximal tubule. This is also the segment that normally
reabsorbs virtually all the filtered glucose and amino acids
by Na-dependent cotransport. An additional function of
the proximal tubule is phosphate transport, which is regulated by parathyroid hormone. In addition to these reabsorption functions, secretion of solutes also occurs along
the proximal tubule. The terminal portion of the proximal
tubule, the S3 or pars recta, is the site of secretion of
numerous organic anions and cations, a mechanism used
by the body for eliminating a number of drugs and toxins.
The proximal tubule, as shown in Fig. 7, has a prominent
brush border, extensive interdigitated foot basolateral in-

N a+

Basolateral cell
membrane . 1
Capillary basement membrane

FIGURE 6 General scheme for epithelial transport. The driving

force for solute movement is primarily generated by the action of
the Na,K-ATPase in the basolateral membrane. Solut e and water
can move through either a paracellular pathway between cells or
a transcellular transport pathway, which requires movement across
both luminal and basolateral membranes. From BM Koeppen,
BA Stanton, Renal Physiology, Mosby, St. Louis, 1992. Wit h per-

foldings, and large prominent mitochondria, which supply

the energy for Na,K = ATPase.
Loop of Henle

The loop of Henle consists of the terminal or straight

portion of the proximal tubule, thin descending and ascending limbs, and the thick ascending limb and is important
for generation of a concentrated medulla and for dilution
of the urine. The thick ascending limb is often called the
diluting segment, since transport along this water impermeant segment results in development of a dilute tubular
fluid. The thick ascending limb is also a major site of Mg 2+
reabsorption along the nephron. The principal luminal
transporter expressed in this segment is the Na + -K+ -2Cl
cotransporter, which is the target of diuretics such as furosemide. The morphology of loop of Henle epithelia is illustrated in Fig. 8.
Distal Nephron

The distal nephron, which includes the distal convoluted

tubule, the connecting tubule, and the cortical and medul-

10

Briggs, Kriz, and Schnermann

TABLE 4

Types of Membrane Transport Mechanisms Used in the Kidney

Mechanism
Facilitated or carrier mediated
Active transport (pumps)
Coupled transport Cotransport
Countertransport Osmosis

Examples of substances
Glucose, urea Na+, K+, H+, Ca2+
Cr, glucose, amino acids, formate, phosphate
Bicarbonate, H+ H2O

lary collecting duct, is the portion of the nephron where

final adjustments in urine composition, tonicity, and volume are made. Distal segments are the sites where the
most critical regulatory hormones, like aldosterone and
vasopressin, regulate acid and potassium excretion and to
determine final urinary concentrations of potassium, sodium, and chloride.
Both distal convoluted tubule and connecting tubule
have well-developed basolateral infoldings with abundant
mitochondria, like the proximal tubule, although they are
easily distinguished from proximal tubule by the lack of

Examples of transport protein

GLUT1 carrier, urea carrier Na,KATPase, H-ATPase, Ca ATPase
Na- K- Cl cotransporter C1/HCO3
exchanger, Na/H antiporter Water
channels (aquaporins)

brush border (Fig. 9). The distal convoluted tubule is the

principal site of action of thiazide diuretics.
The collecting duct cells are cuboidal, and their basolateral folds do not interdigitate extensively. When there is
a sizable osmotic gradient, and water moves across this
epithelium, the spaces between cells become wide. The
collecting duct changes in its appearance as it travels from
the cortex to the papillary tip (Fig. 10). In the cortex there
are two different cell types in the collecting duct: principal
cells and intercalated cells. Principal cells are the main site
of salt and water transport, and intercalated cells are the

S
3

FIGURE 7 Proximal tubule. The proximal tubule consists of three segments: SI, S2, and S3. The left panel shows
schematic diagrams of the typical cells from these three segments, the right panel shows a cross section of the SI
segment. The SI begins at the glomerulus, and extends several millimeters, before the transition to the S2 segment.
The S3 segment, which is also called the proximal straight tubule, descends into the renal medulla to the inner medulla.
The proximal tubule is characterized by a prominent brush border, which increases the membrane surface area by a
factor of about 40-fold. The ba solateral infoldings, which are lined with mitochondria, are interdigitated with the
basolateral infoldings of adjacent cells (in these diagrams, processes that come from adjacent cells are shaded). These
adaptations are most prominent in the first parts of the proximal tubule, and are less well developed later along the
proximal tubule.

I. Overview of Renal Function and Structure

TA
L

FIGURE 8 Loop of Henle. The loop of Henle makes a hairpin loop into the medulla. Segments included in the loop
are terminal portion of the proximal tubule, the thin descending (tDL), and ascending limbs (tAL), as well as the thick
ascending limb (TAL). The left panel shows schematic drawings of cell morphology, the right panel shows a cross- section
through the thin descending limb in the outer medulla. The thin limbs, as their names suggest, are shallow epithelia
without the prominent mitochondria of more proximal segments. The thick limb, in contrast, is a taller epithelium with
basolateral infoldings and well- developed mitochondria. This segment is water impermeable, and transport along this
segment is important for generation of interstitial solute gradients, and a low salt concentration and dilute fluid in the
tubular lumen.

DCT

CT

FIGURE 9 Distal convoluted tubule. The distal convoluted tubule is customarily divided into two parts: the true distal
convoluted tubule (DCT, shown schematically on the left, and in cross-section on the right) and the connecting tubule
(CT), where cell morphology is somewhat more simil ar to collecting duct.

II

Briggs, Kriz, and Schnermann

12

CCD

PC

* 1C

OMCD

- ,

IMCD

FIGURE 10 Collecting duct. The collecting duct changes its morphology as it travels from cortex to medulla.
In the cortex there are two cell typesprincipal cells (PC) and intercalated cells (1C). Appearance is shown
schematically on the left and in a cross- section on the right.

key sites for acid-base regulation. The medullary collecting

duct in its most terminal portions comes increasingly to
resemble the tall cells typical of transitional epithelium.
SALT AND VOLUME REGULATION
Absorption of Sodium

Because of its high extracellular concentration, large

amounts of Na + and its accompanying anions are present
in the glomerular filtrate, and the absorption of this filtered
Na + is in a quantitative sense the dominant work performed
by the renal tubules. The amount of Na + absorbed by the
tubules is the difference between the amount of Na + filtered
and the amount of Na + excreted,
Na + absorption = Filtered Na' Na + Excretion
or
Na + absorption = (GFR X PNl,) - (V X [7Na ),
where t/Na is the urinary Na+ concentration and PNa is the

plasma Na + concentration. With a GFR of 120 mL/min

and a plasma Na + concentration of 145 mEq/L, 17.4 mEq
of Na + is filtered every minute or about 25,000 mEq or
575 g of Na per day. Since only about 100 to 250 mEq of
Na is excreted per day (this reflects the average intake
provided by a typical Western diet), or can estimate that
the tubule reabsorbs somewhat more than 99% of the fil-

tered Na. The fractional excretion of Na+ (FENa) is defined

as the fraction of filtered Na+ excreted in the urine. Using
creatinine as a GFR estimate, FENa is calculated from
FENa = Excreted Na _

f/ Na X V
Filtered Na ~ P Na X GFR

___ t/_Na XV
/TSJa

_ t/Na/P Na

FENa is usually less than 1%. However, this value depends

on Na intake and can vary physiologically from nearly 0%
at extremely low intakes to about 2% at extremely high
intakes. FENa can also exceed 1% in disease states where
tubular transport of Na is impaired (for example, in most
cases of acute renal failure).
Mechanisms of Na + Absorption
Tubular Na + absorption is a primary active transport
process driven by the enzyme Na,K-ATPase. In renal epithelial cells, as in most cells of the body, this pump translocates Na + out of cells (and K+ into cells) and thereby lowers
intracellular Na concentration (and elevates intracellular
K+ concentration). A key for the generation of net Na +
movement from tubular lumen to blood is the asymmetrical
distribution of this enzyme: it is present exclusively in the
basolateral membrane (the blood side) of all nephron seg-

I. Overview of Renal Function and Structure

ments, but not in their luminal membranes. Delivery of Na

to the pump sites is maintained by Na entry into the luminal
side of the cells along a favorable electrochemical gradient.
Since Na permeability of the luminal membrane is much
higher than that of the basolateral membrane, Na entry is
fed from the luminal Na pool. The asymmetric permeability
is due to the presence of a variety of different transport
proteins or channels exclusively in the luminal membrane.
A number of these luminal transporters are the target
molecules for diuretic action. Principal entry mechanisms
for Na + and Cl~ in the different nephron segments (and
effective diuretics) are:
1. Early proximal: Na + -dependent cotransporter, Na + /
H+ exchanger
2. Late proximal: Na + /H+ exchanger, Cl~/anion
exchanger
3. Thick ascending limb of the loop of Henle:
Na + -K+ -2Cl~ cotransporter (furosemide-sensitive
carrier)
4. Distal convoluted tubule: Na + /Q~ cotransporter
(thiazide-sensitive carrier)
5. Collecting duct: Na + channel (amiloridc-sensitive
channel)
Regulation of NaCI Excretion
+

Because Na salts are the most abundant extracellular

solutes, the amount of sodium in the body (the total body
sodium) determines extracellular fluid volume. Therefore,
excretion or retention of Na + salts by the kidneys is critical
for the regulation of extracellular fluid volume.1 A disturbance in volume regulation, particularly enhanced salt retention, is common in disease states. The sympathetic, nervous system, the renin-angiotensin-aldosterone system
(RAS), atrial natriuretic peptide (ANP), and vasopressin
represent the four main regulatory systems that change
their activity in response to changes in body fluid volume.
These changes in activity mediate the effects of body fluid
volume on urinary Na excretion.
Sympathetic Nervous System

A change in extracellular fluid volume is sensed by

stretch receptors on blood vessels, principally those located
on the low pressure side of the circulation in the thorax,
for example, in the vena cava, cardiac atria, and pulmonary
vessels. A decreased firing rate in the afferent nerves from
these volume receptors enhances sympathetic outflow from
cardiovascular medullary centers. Increased renal sympathetic tone enhances renal salt reabsorption and can decrease renal blood flow at higher frequencies. In addition
to its direct effects on renal function, increased sympathetic
1

Plasma Na concentration does not correlate at all with total

body sodium or the fullness of the extracellular fluid spaces. In
fact, a low serum Na can be observed in states in which there is
either excess total body sodium or deficiency of total body sodium.
However, plasma Na concentration is the principal determinant
of extracellular fluid osmolarity. In general, abnormalities in Na
concentration arise from defects in tonicity regulation, not volume regulation.

13

outflow promotes the activation of another salt retaining

system: the RAS.
Renin-Angiotensin System

Renin is an enzyme that is formed by and released from

granular cells in the wall of renal afferent arterioles near
the entrance to the glomerulus. These granular cells are
part of the juxtaglomerular apparatus (see Fig. 3). Renin
is an enzyme that cleaves angiotensin I from angiotensinogen, a large circulating protein made principally in the liver.
Angiotensin I, a decapeptide, is converted by angiotensinconverting enzyme to the biologically active angiotensin
II. Renin is the rate-limiting step in the production of angiotensin II and it is therefore the plasma level of renin that
determines plasma angiotensin II. The three principal
mechanisms in control of renin release are:
1. Macula densa mechanism. Macula densa refers to a
group of distinct epithelial cells in the wall of the thick
ascending limb of the loop of Henle, where it makes contact
with its own glomerulus. At this location, NaCI concentra
tion is between 30 and 40 mEq/L, and it varies as a direct
function of tubular fluid flow rate; i.e., it increases when
flow rate is high and decreases when flow rate is low. A
decrease in NaCI concentration at the macula densa stron
gly stimulates renin secretion, whereas an increase inhibits
it. The connection to the regulation of body fluid volume
is the dependence of the flow rate past the macula densa
cells upon body Na content: flow. It is high in states of Na
excess and low in Na depletion.
2. Baroreceptor mechanism. Renin secretion is stimu
lated by a decrease in arterial pressure, an effect believed
to be mediated by a "baroreceptor" in the wall of the
afferent arteriole responding to pressure, stretch, or
shear stress.
3. /3-adrenergic stimulation. An increase in renal sympa
thetic activity or in circulating catecholamines stimulates
renin release through /3-adrenergic receptors on the juxta
glomerular granular cells.
Angiotensin II has direct and indirect effects to promote
salt retention. It enhances Na+ reabsorption in the proximal
tubule (stimulation of Na/H exchange) and, because it is
a potent renal vasoconstrictor, it may reduce GFR by reducing glomerular capillary pressure or plasma flow. Angiotensin II affects salt balance indirectly by stimulating the
production and release of the steroid hormone aldosterone
from the zona glomerulosa of the adrenal gland. Aldosterone acts on the collecting duct to augment salt reabsorption (and K+ secretion).
Atrial Natriuretic Factor

Atrial natriuretic factor (ANF) is a peptide hormone

that is synthesized by atrial myocytes and released in response to increased atrial distension. Thus, ANF secretion
is increased in volume expansion and inhibited in volume
depletion. The main cause of the ANF-induced natriuresis
is an inhibition of Na + reabsorption along the collecting
duct, but an increase in GFR may sometimes play a contributory role.

14

Briggs, Kriz, and Schnermann

Vasopressin or Antidiuretic Hormone (ADH)

Vasopressin is regulated primarily by body fluid osmolarity. However, in states in which intravascular volume is
depleted, the set point for vasopressin release is shifted,
so that for any given plasma osmolarity, vasopressin levels
are higher than they would be normally. This shift promotes
water retention to aid in restoration of body fluid volumes.

Tubular fluid

Fusion of vesicles
containing water
channels

WATER AND OSMOREGULATION

Regulation of Body Fluid Osmolarity

When water intake is low or water is lost from the body

(in hypotonic fluids such as sweat, for example), the kidneys
conserve water by producing a small volume of concentrated urine. In dehydration, urine production is less than
a liter per day (less than 0.5 mL/min) and the osmotic
concentration may reach 1200 mOsm/kg H2O. When water
intake is high, urine flow may increase to as much as
14 L/day (10 mL/min) with an osmolality substantially
lower than that of plasma (75 to 100 mOsm/kg). These
wide variations in urine volume and osmotic concentration
do not obligatorily affect the excretion of the daily solute
load. Thus, the daily solute excess of about 1200 mOsm/
day may be excreted in 12 L of urine (with a Uosm of 100
mOsm/L) or in 1 L (with a t/ osm of 1200 mOsm/L). The
hormone responsible for the regulatory changes in urine
volume and tonicity is ADH (synonym: vasopressin).
Role of ADH in Osmolarity Regulation

ADH is a nine-amino-acid peptide produced by neurons

located in the supraoptic and pareventricular nuclei of the
hypothalamus. It is stored in and released from granules
in nerve terminals that are located in the posterior pituitary
(neurohypophysis). The release of ADH is exquisitely sensitive to changes in plasma osmolarity, with increases in
Posm above a threshold of about 285 mOsm/kg leading to
increases in ADH secretion and plasma ADH concentrations. As has been pointed out, the actual set point for
release depends on body fluid volume as well.
The most important function of ADH is the regulation
of water permeability of the distal portions of the nephron,
particularly the collecting duct. As shown schematically in
Fig. 11, ADH binds to receptors (R) in the basolateral
membrane of collecting duct cells. This activates adenylate
cyclase (AC) to form cAMP. cAMP activates a protein
kinase, which leads to the phosphorylation of undefined
proteins. This phosphorylation causes membrane fusion of
vesicles that contain preformed water channels. The result
is an up to 20-fold increase in water permeability of the
apical (luminal) membrane of collecting duct cells. Upon
removal of ADH, water channels are rapidly removed from
the apical membrane by endocytosis.
Tubular Water Absorption

At each point along the nephron, the osmotic pressure

of the tubular fluid is lower than that in the interstitial space.
This transtubular osmotic pressure difference provides the

H3 0

Blood

FIGURE 11 Mechanism of action of ADH on the collecting duct.

ADH combines with a basolateral receptor (R) which is coupled
to adenyl cyclase (AC). Generation of cyclic AMP leads to activation of protein kinase A, and phosphorylation of target proteins.
Preformed water channels contained in vesicles located under
the luminal membrane are inserted into the luminal membrane,
resulting in a tenfold increase in its water permeability. From BM
Koeppen, BA Stanton, Renal Physiology, Mosby, St. Louis, 1992.
With permission.

driving force for tubular water reabsorption. The rate of

fluid absorption in a given nephron segment is determined
by the magnitude of this gradient and the osmotic water
permeability of the segment. Even though the osmotic pres sure difference across the proximal tubule epithelium is
very small (3 to 4 mOsm/L), the rate of fluid absorption
is very high because this segment has a very high water
permeability. In contrast, osmotic gradients across the thick
ascending limb may be as high as 250 mOsm/L, and yet
virtually no water flows across this segment because it is
highly water impermeable. This segment dilutes the urine
because it absorbs Na + and Cl~ without water.
In contrast to the invariability of water conductivity in
the proximal tubule and the thick ascending limb, water
permeability in the collecting duct can be altered under
the influence of ADH. If ADH is absent, water permeability and water absorption are low, and the hypotonicity
generated in the thick ascending limb persists along the
collecting duct. As a consequence, a dilute urine is excreted.
If ADH is present, the collecting duct becomes quite water
permeable, and water is reabsorbed until the tubular fluid
in the collecting duct equilibrates with the hypertonic interstitium. The final urine in this case is osmotically concentrated and has a low volume.
Medullary Hypertonicity

To allow osmotically driven water absorption, the osmotic concentration in the medullary interstitium must be
slightly higher than that in the collecting duct lumen. Thus,
when a final urine with an osmolarity of 1200 mOsm/kg is
excreted, the medullary interstitium at the tips of the papillae must be a little higher than 1200 mOsm/kg. The genera tion of such a unique extracellular environment is achieved

15

I. Overview of Renal Function and Structure

by a counter-current multiplication system that exists in the

renal medulla in the form of the countercurrent arrangement of descending and ascending limbs of the loops of
Henle.
Countercurrent Multiplication

In two adjacent tubes with flow in opposite directions,

the fluid can attain an osmotic concentration difference in
the system's longitudinal axis that can by far exceed that
seen at each level along,it. This principle of countercurrent
multiplication requires energy expenditure and the pres ence of unique differences in membrane characteristics
between the two limbs of the system.
The countercurrent multiplier represented by the loops
of Henle is believed to generate an osmotic gradient because:
1. Active Nad transport across the ascending limb
(the so-called single effect of the countercurrent
system) generates an osmotic difference between
tubular fluid and surrounding local interstitium.
2. A low water permeability of the ascending limb
prevents dissipation of this gradient.
3. A high water permeability of the descending limb
permits equilibration of descending limb contents
with the surrounding local interstitium.

isoosmotic to plasma. In Step 2, NaCl is absorbed from the

ascending limb into the interstitium until a gradient of
200 mOsm/kg is reached. In Step 3, the fluid in the descending limb equilibrates osmotically with the interstitium by
water movement out of the tubule. In Step 4, the hypertonic
fluid is presented to the TAL with an increased solute
concentration in the region near the tip of the system.
Again active NaCl transport along the ascending limb establishes a 200 mOsm/kg gradient increasing interstitial
concentrations and by water abstraction descending limb
contents. Note that concentrations near the tip begin to be
higher than those near the base. Continued operation of
such a mechanism will gradually result in generation of a
gradient of hypertonicity, with the highest osmolarities at
the papillary tip.
The tubular fluid leaving the ascending limb of the loop
of Henle countercurrent multiplier is hypotonic. However,
the medullary interstitium has been osmotically "charged."
Since the collecting ducts on their way to the papillary tip
return into the hypertonic medullary environment, their
content can now be concentrated by water flow along an
osmotic gradient.
Role of Urea in the Countercurrent Mechanism

How such a system can result in progressive increases

in osmotic concentration along the corticopapillary axis is
shown in Fig. 12. In Step 1 (time zero), the fluid in the
descending and ascending limbs and the interstitium is

In addition to Na + and Cl~, urea is the other major solute

present in the renal medulla in an osmotically concentrated
form. Urea enters the medulla by reabsorption across the
collecting duct. Marked differences in the permeability to
urea allow reabsorption to proceed only across the terminal
portions of the medullary collecting duct. In the early por-

Descending Interstitium

Ascending

limb
300

300

(D

limb
400 200

300

400

400

200

300 300 300

400 400 400

200

400 400

400

200

400

400

200

400

300
300

300

300

300

300

300

300 400

200

400

300

300 400
300

300

200
300

300 400

200

300 350

150

350

400 400 350

150

350

500

300

500

300

500

300

500

35
0
50
35 15
0
0 0

15 - - ^ 30Repeat steps 4-6

500
0
v 0
30

70
0
100
30 30 10
0
0 0 0

70
0
100
0

50
120 120 100
0 0
0
80
0

FIGURE 12 The process of countcrcurrent multiplication. See text for details. From BM Koeppen, BA
Stanton, Renal Physiology, Mosby. St. Louis, 1992. With permission.

16

iggs, Kriz, and Schnermann

tions of the collecting duct, urea permeability is low and

reabsorption of urea cannot occur. Since water leaves the
tubule under the influence of ADH, the urea staying behind
is progressively concentrated. As a consequence, a substantial urea gradient develops, providing the driving force for
urea reabsorption when the permeability to urea permits
it. The contribution of urea accumulation to osmotic water
absorption along the inner medullary collecting duct must
be sizable since about half of inner medullary tonicity is
accounted for by urea. Therefore, a reduction in urea synthesis by reducing protein intake markedly impairs the
concentrating ability of the kidneys.
Comparison between Volume Regulation
and Osmoregulation
Osmoregulation is under control of a single hormonal
system, ADH, whereas volume regulation is under control
of a set of redundant and overlapping control mechanisms .
Lack or excess of ADH results in defined and rather dramatic clinical syndromes of excess water loss or water retention. In contrast, a defect in a single volume regulatory
mechanism generally results in more subtle abnormalities,
because of the redundant regulatory capacity from the
other mechanisms. Thus, excess aldosterone results in a
mild volume retention followed by "'escape" and return
to normal Na + excretion, due to the action of the other
mechanisms. Similarly, excess ANF probably produces
only a modest decrement in volume, with no persistent
abnormality in Na + excretion. Severe salt-retaining states,
such as liver cirrhosis and congestive heart failure, are
characterized by activation of all the volume regulatory
mechanisms. Finally, the symptoms characteristic of disorders of Osmoregulation and of volume regulation are different, with hypo- and hypernatremia being the hallmark of
deranged Osmoregulation, and edema or hypovolemia resulting from deranged volume regulation.
REGULATION OF BODY FLUID POTASSIUM
AND ACIDITY

Both potassium and hydrogen ions are present in body

fluids at low concentrations, about 4 to 4.5 mEq/L for K+
and about 40 nEq/L for H+ . Both ions show a number
of features:
1. Relatively small deviations in either K+ or H+ concen
trations can be life threatening, and therefore the regula
tion of K+ and H+ concentration requires control systems
with high sensitivity and precision.
2. Constancy of both K+ and H+ ion concentration over
the long term is achieved by regulated excretion of these
ions in the urine. However, in both cases, other mechanisms
exist that provide immediate protection against excessive
deviations of plasma concentrations from normal.
3. Regulation in the renal excretion of both K+ and H+
is caused to a large extent by variation in the secretion of
these ions by collecting ducts. The principal cell of the
collecting duct is the cell type responsible for regulated K+

secretion; the intercalated cell is the cell type responsible

for H+ secretion (see Fig. 10).
4. The rate of both K+ and H+ secretion is increased
by aldosterone.
5. A primary derangement of K+ balance can cause an
acidity disturbance and a primary acidity disturbance can
derange K+ homeostasis.
Regulation of Body Fluid Potassium
Distribution of Body K +
Due to the presence of Na,K-ATPase in virtually all cell
membranes, K+ is mostly in the intracellular space. Of the
3500 mEq of body potassium, only about 1 to" 2% is present
in the extracellular space, where it has a concentration of
4 to 5 mEq/L. The remainder (about 98%) is stored in cells.
The distribution has the potential risk in that the release
of even a small amount of K+ from intracellular stores can
elevate plasma K+ concentration substantially (for example, in insulin deficiency, cell lysis, severe exercise). On the
other hand, the distribution of K+ between the extracellular
and intracellular space is utilized as a means to buffer acute
changes in plasma K+ concentrations. For example, the
administration of an acute oral K+ load induces much
smaller changes in plasma K+ concentration than would
occur if all absorbed K+ were to remain in the extracellular
space. K+ ions are shifted into cells under the stimulatory
influence of insulin and epinephrine. The effect of both
hormones reflects mainly an activation of the Na,KATPase. Another important factor determining K+ distribution is plasma H+ concentration. An increase in H+ ions
causes uptake of H+ into cells and intracellular buffering
and this uptake to some extent occurs in exchange for K+ .
Thus, acidosis will tend to increase plasma K+ , and alkalosis
will tend to decrease it.
Renal Handling of K +
K+ homeostasis in the long term requires the excre tion of an amount equivalent to the daily K+ intake (50 to
150 mEq). This represents a fractional K+ excretion (FEK+)
of about 10%, much higher than FE^. About 60 to 70%
of filtered K+ is absorbed along the proximal tubule and
further reabsorption of K+ takes place in the thick ascending limb of the loop of Henle, so that only about 10% of
filtered K+ enters the distal tubule. Along the collecting
duct, K+ is both secreted and absorbed. Collecting duct K+
secretion increases when dietary K+ intake is elevated. On
the other hand, when intake is low, collecting duct K+
secretion virtually ceases and absorption is dominant. Thus,
while K+ absorption along the proximal tubule and the
loop of Henle does not change very much depending on
intake, collecting duct K secretion is variable, and this
variability accounts almost completely for the variation in
urinary K+ excretion.
Mechanisms of K + Secretion
K+ secretion across the collecting duct epithelium utilizes
the transcellular route. K+ uptake across the basolateral
membrane is driven by Na,K-ATPase, which elevates intra-

I. Overview of Renal Function and Structure

17

cellular K+ concentration to a level above electrochemical

equilibrium. K+ can then move along a favorable gradient
from cell interior to tubule lumen utilizing potassium channels in the luminal membrane.
Three major variables determine the rate at which K+
is secreted by collecting duct cells:

tion in segments prior to the collecting duct (in the loop

of Henle and in the distal tubule respectively) increase the
flow of fluid past the collecting duct cells which causes
increased K+ secretion. In addition, the diuretics cause
volume depletion, which stimulates aldosterone secretion.

1. Changes in the activity of Na,K-ATPase affect uptake

and thereby intracellular K+ concentration. An increase in
pump activity will increase intracellular Kh levels and will
tend to stimulate K+ secretion.
2. Changes in the electrochemical gradient affect the
driving force for K+ movement across the luminal mem
brane. Both an increase in intracellular K+ concentration
and in the lumen negative transepithelial potential differ
ence will increase the driving force and will tend to increase
K+ secretion.
3. Changes in the permeability of the luminal membrane
determine the amount of K+ that can be secreted for a
given driving force. Thus, an increase in luminal K+ conduc
tance will increase K+ secretion.

Regulation of Body Fluid Acidity

Regulation of K+ Excretion

1. Plasma K + concentration. One important determinant

of K+ excretion is plasma K+ concentration. For example,
the change in K+ excretion following a change in dietary
K+ intake is mediated by an increase in plasma K+ . The
effect of plasma K+ on secretion is induced partly by a
direct effect on intracellular K+ concentration.
2. Aldosterone. At any level of plasma K'. K+ secretion
will also depend upon plasma aldosterone levels. Aldoste
rone enhances K+ secretion by activation of Na,K-ATPase
and by an increase in K+ permeability of the luminal mem
brane. Aldosterone is partly responsible for the diet induced increase in K+ excretion since its production and
secretion are directly stimulated by plasma K+ concentra
tion. This effect is independent of angiotensin.
3. Tubular flow rate. An increase in tubular flow rate
past the K+ -secreting cells stimulates, and a decrease re
duces, K+ secretion. This effect is the consequence of flowdependent changes in the K+ gradient across the apical
membrane. K+ secretion causes an increase in K+ concen
tration in the tubular fluid, which eventually decreases the
K+ gradient and the rate of K+ secretion. An increase in
flow diminishes the rate of rise of luminal K' concentration
so that a more favorable K+ gradient for K+ secretion
is maintained.
4. Distal sodium delivery. When more Na * is delivered
to the distal nephron, if reabsorption increases, the net
electrical charge in the lumen will become more negative.
This favorable electrochemical gradient will tend to in
crease urinary K+ secretion.
5. Hydrogen ions. A decrease in FT ion concentration
in alkalotic states causes a stimulation of K* secretion. This
effect is mediated by the rise in intracellular K+ concentra
tion that occurs in alkalosis.
Diuretics and K+ Excretion

Diuretics increase tubular flow rate. Agents such as loop

diuretics and thiazides that inhibit NaCl and water absorp -

Basic Considerations

Maintenance of the extracellular pH around 7.4 depends

on the operation of buffer systems that accept H+ when it
is produced and liberate H+ when it is consumed. The state
of the demand on total body buffering can be determined
by assessing the behavior of the HCO3/CO2 system, which
is the major extracellular buffer. The law of mass action
for this buffer system states that
[HC03]
pH = 6.1 + log [C0 ]
2
Since [CO2] equals the solubility
coefficient times the this can be rewritten
[HC03
pH = 6.1 + log 0.03 X
This, the familiar HendersonHasselbach equation tells us that pH constancy depends on
a constant ratio in the concentration between the two
buffer components. If this ratio increases because either
HCO3 increases or CO2 decreases, pH will increase
(alkalosis). If the ratio decreases, because either HCO3
decreases or CO2 increases, pH decreases' (acidosis).
Regulation of HCO3 is mainly a function of the kidneys,
and regulation of CO2 is a respiratory function. The
regulation of HCOj concentration by the kidneys
consists of two main comp onents:
1. Absorption of HCOj. Because of the high GFR and
because plasma HCOj concentrations are also relatively
high (24 mEq/L), large amounts of HCO are filtered. Re
trieval of this filtered HCOs is absolutely essential for acidbase balance. It is important to note that this process of
renal HCO 3 absorption does not add new HCO3 to the
blood, but merely prevents a loss of filtered HCO3 into the
urine. Therefore, renal HCO3 absorption cannot correct
an existing metabolic acidosis.
2. Excretion of H + ions. Under normal dietary condi
tions, approximately 40 to 80 mmol H+ ions are generated
daily (mostly sulfuric acid from the metabolism of sulfurcontaining amino acids). These H+ ions are buffered and
therefore consume HCO. The kidneys must excrete these
H+ ions to regenerate the HCO3 pool (this second task can
therefore also be labeled as generation of "new" HCOi).
Mechanisms of Bicarbonate Absorption

Filtered HCO (about 4300 mEq/day) is efficiently absorbed by renal tubules, predominantly the proximal tubules, so that under normal acid-base conditions very little
HCO3 is found in the urine. As a rule, all tubular HCO3
absorption is the consequence of H+ ion secretion, and not

Briggs, Kriz, and Schnermann

18
of direct absorption of HCO 3 ions. H+ ion s are continuously
generated inside the cells from the dissociation of H 2 O (or
by CO 2 reacting with H 2 O) and transported into the lumen.
In the lumen, secreted H + ions combine with filtered HCOi
to form carbonic acid, which is broken down to CO 2 and
H 2 O in a reaction that is catalyzed by a carbonic anhydrase
located in the apical brush border membrane. CO 2 a n d
H 2 O are then absorbed passively. The OH - ions which are
generated in the cell during this process combine with CO 2
to form HCOi, a reaction catalyzed by a cytosolic carbonic
anhydrase. HCOi exits across the basolateral side of the
cell and returns to the blood in association with Na + . The
net balance of this process can be expressed as
H 2O + CO, <- H 2CO 3 <- HCOj + H* <- H 2O -> OH" + CO 2 - HCOi + Na'
(tubular lumen)
(cell interior)
(blood)

Specific
transport
proteins in renal epithelial cells cause t h e H+ and HCOi to
move in the right directions. Two different mechanisms,
both located in the apical mem brane, are responsible for
t h e m o v e m e n t o f p r o t o n s i n t o the tubular fluid:
1. The first is a NaVH + exchanger that is driven by the
N a + gradient and is found in the proximal tubule. In terms
of mEq transported, it contributes most to HCOi ab
sorption.
2. The second is a primary active transport of H + i o n s .
An H +- ATPase has been found in the luminal membrane
of one class of intercalated collecting duct cells. There is
also some evidence for the presence of an H,K -A T P a s e
similar to that found in parietal cells of the gastric mucosa.
A c t i v e H + ion transport is respo nsible for the secretion of
smaller amounts of H + t h a n N a + / H + exchange, but it can
proceed against a steeper gradient.
There are also at least two mechanisms for the transport
of HCOi across the basolateral membrane. The movement
of HCOi can be coupled to the movement of Na ions and
this is the major exit mechanism in the proximal tubule.
In the collecting duct, HCOi exit occurs predominantly
through a basolateral Cl/HCOi exchanger (equivalent to
the band 3 protein of red cells).
Bicarbonate Secretion
W h i l e net HCOi transport for the whole kidney is always
in the reabsorptive direction, certain intercalated cells in
the cortical portion of the collecting duct can actually se crete HCOi. The HCOi -secreting cells have a polarity that
is the reverse of the H + -s ecreting cells; i.e., they possess a
b a s o l a t e r a l H+ - A T P a s e a n d p r o b a b l y a l u m i n a l C l / H C O i
exchanger. HCOi secretion may be important during con sumption of a diet providing base equivalents and for the
correction of metabolic alkalosis.
Excretion of H+ Ions (Formation of New HCOj)
Urinary acid excretion cannot to any significant extent
occur as free H + ions. The absolute minimum urinary pH
in humans is about 4.5, corresponding to an H + concentra -

tion of only 0.03 mEq/L. Since about 40 to 80 mEq of H +

must be excreted per day, it is clear that most H + ions must
be excreted in a bound or buffered form. Excretion of
b o u n d H + is achieved by: (1) the titration of luminal nonbi carbonate buffers and (2) by the renal synthesis and excre tion of ammonium ions.
Titratable Acidity
Binding of secreted H+ ions to filtered nonbicarbonate
buffer anions leads to the formation and excretion of uri nary titratable acidity (titratable acidity is defined as the
number of moles NaOH that has to be added to bring urine
pH back to 7.4). The ability to buffer H + ions depends on
its dissociation constant (pK) and the quantity of buffer.
Under normal conditions only the HPO4 2 / H 2 P O ; i b u f f e r
is present in amounts sufficient to act as intratubular H +
acceptor. This buffer pair has a pK of 6.8 and is excreted
at a daily rate of about 50 mmol. Using the Henderson Hasselbach equation for the phosphate buffer (pH 6.8 + log
[HPO4 2 ]/[H 2 PO4]) the following relations can be calculated
(considering only that fraction of total PO;; that is actual l y
e x c r e t e d , a b o u t 2 5 t o 3 0 % o f t h e f i l t e r e d P O 4 load):
HPOi2
pH (mmol/day)

H 2 POj
(ininol/day)

H + buffered
(mmol/day)

Filtrate
End proximal

7.4
6.8

40
25

10
25

0
15

Urine

4.8

0.5

49.5

39.5

This table shows that the buffer capacity o f HPO^ 2 can

be fully utilized if the intratubular pH is lowered suffi ciently. In some situations other urinary buffers become
important. In diabetic ketoacidosis large amounts of /3 hydroxybutyrate are excreted (e.g., 300 mmol/L). Even
though this buffer has a pK of 4.8 it will carry up to
1 5 0 m m o l H + ions per liter.
Ammonium Excretion
The second form of bound H + ions in the urine is ammo nium. The excretion of NHJ is equivalent to the generation
of HCOi or excretion of H + . Glutamine, formed in the liver
from glutamate and extracted from the blood by uptake
mechanisms in the luminal and basolateral membranes of
renal proximal tubule cells, is the major source of urinary
ammonium. Ammonium is generated in the proximal tu bule by a metabolic pathway in which the degradation of
glutamine to glutamate and further to a - k e t o g l u t a r a t e
yields 2 NH} ions and 2 HCOi ions (rather than NH 3 , CO 2 ,
a n d H2 O). While the NHJ ions are secreted through distinct
transport pathways into the lumen of the proximal tubule,
the new H COi ions are added to the blood HCOi pool.
It is essential that the NHJ that is formed by renal proxi mal tubules is preferentially secreted into the tubular lumen
and then excreted in the urine. If the generated NHJ was
absorbed by the renal tubular epith elium (or secreted pref erentially into the blood), it would be used to form urea.
Ureagenesis forms H+ ions which would consume the pro duced HCOi and thereby negate net base production. This
is shown in the following reactions:

I. Overview of Renal Func tion and Structure

2NH 4 + 2CO2 -> urea + H2O + 2H+ or

19
200

600

600-

800

2NH4 + 2HCOa -> urea + CO2 + 3H2 O

(urea: H2NCONH2).
Urinary H+ excretion in the form of NH4 is on the order
of 40 to 50 mmol/day. Renal NH4 formation and excretion
are greatly enhanced in metabolic acidosis. Failure of proximal tubules to generate NH4 is the main reason that chronic
renal failure leads to metabolic acidosis.

400-

Reabsorbed

200-

Regulation of H + Ion Secretion

1. Intracellular pH. Systemic pH changes, whether

caused by changes in plasma HCO^ (metabolic) or by
changes in P Co 2 (respiratory), alter H+ secretion (and there
fore HCO>3 absorption). Intracellular acidification, as oc
curs in acidosis, stimulates H+ secretion and intracellular
alkalinization (alkalosis) inhibits it.
2. Aldosterone. In addition to affecting Na+ absorption
and K+ secretion, aldosterone stimulates H'' secretion by
collecting ducts.
3. Potassium. Changes in plasma K+ concentration can
affect H+ secretion, in part by changing intracellular pH.
Thus, hypokalemia increases intracellular acidity and stim
ulates H+ ion secretion. While the effect of hypokalemia
alone is relatively small, a marked stimulation of H+ secre
tion results when hypokalemia occurs with high plasma
aldosterone levels. In this situation, which can occur in
primary hyperaldosteronism or with administration of di
uretics, metabolic alkalosis may be generated by the
kidneys.
RENAL HANDLING OF GLUCOSE AND AMINO ACIDS

An important function of the renal tubule is retrieval of

the glucose and amino acids that are present in glomerular
filtrate and which would be lost to the body if they were
not reabsorbed. To a large extent, this is a proximal tubule
function, and disordered glucose and amino acid transport
is characteristic of diseases that disturb proximal tubular
function.
Glucose transport by the proximal tubule occurs via a
transport protein present in the luminal membrane that
carries a glucose molecule together with a sodium ion, the
glucose-sodium cotransporter. This transporter uses the
sodium concentration gradient (sodium concentration is,
of course, higher outside the cell than in) to drive the
movement of glucose into the cell. Glucose then diffuses
out of the cell across the basolateral membrane, a process
facilitated by a second carrier protein. The resulting reabsorption process is highly efficient, and in normal circumstances practically all the filtered glucose is removed from
the proximal tubule fluid and glucose is virtually absent
from urine.
When plasma glucose concentration rises, increasing
amounts of glucose are filtered, and at a certain point the
filtered load of glucose exceeds the capacity of the proximal

400

PG

(mg%)

FIGURE 13 Glucose reabsorption. Shown is a typical titration

curve for renal glucose reabsorption. At plasma glucose concentrations less than approximately 200 mg%, the filtered glucose is
completely reabsorbed, and no glucose is excreted in the urine.
When plasma glucose exceeds this level, the filtered load of glucose
exceeds the transport capacity of the tubule, and glucose apears
in the urine.

transport mechanisms. This maximum reabsorption rate is

called the tubular transport maximum for glucose (TmG;
Fig. 13). When glucose delivery exceeds the TmG, the excess
glucose is excreted in the urine.
Many of the same principles apply to the reabsorption
of amino acids. Also a function of the proximal tubule,
amino acid absorption is also highly effective. For most
amino acids, less than 1% of the amount filtered escapes
into the urine. A number of different luminal and basolateral transport proteins are needed to remove the amino
acids from the glomerular filtrate. A specific transporter
carries the dibasic amino acids, L-arginine and L-lysine, and
another carrier is responsible for removal of the acidic
amino acids from the tubular fluid. There also are luminal
transporters which, like the sodium-glucose cotransporter,
exploit the sodium concentration gradient for cotransport
of certain amino acids together with sodium. Other carrier
molecules in the basement membrane facilitate the exit of
the amino acids from the cell.
Bibliography

For a more thorough introduction to normal kidney

function:
Berne RM, Levy MN: Physiology, 3rd ed. Chapters 41-43, MosbyYear Book Inc., St. Louis; 1993. '
Koeppen B, Stanton B: Renal Physiology. Mosby-Year Book, St.
Louis, 1992.
Guyton AC: Textbook of Medical Physiology, 8th ed. Saunders,
New York, Chapters 25-31, 1991.
Rose BD: Clinical Physiology of Acid-Base and Electrolyte Disorders, 4th ed. McGraw- Hill, New York, 1994.
Vander A: Renal Physiology, 4th ed. McGraw-Hill, New York,
1991.
Shayman JA: Renal Physiology, Lippincott, Philadelphia, 1995.

For more detail on topics covered in this chapter:

20

Andrew S. Levey

Adrogue HJ, Madias NE: Changes in plasma potassium concentration during acute acid- base disturbances. Am J Med 71:456467, 1981.
Baylis C: Glomerular nitration dynamics. In: Lote CJ (ed) Advances in Renal Physiology. Grune & Stratton, London, pp.
33-83, 1986.
Beauwkes R, Bonventre JV: Tubular organization and vasculartubular relations in the dog kidney. Am J Physio! 229:695703, 1975.
Debnam ES, Unwin RJ: Hyperglycemia and intestinal and renal
glucose transport: implications for diabetic renal injury. Kidney
Int 50:1101-1109, 1996.

Kaplan B, Batlle DC: Regulation of potassium balance and metabolism. In: Jacobsen HR, Striker GE, Klahr S (eds). Principles
and Practice of Nephrology. Mosby, St. Louis, 1995.
Levine SD: Diuretics. Med Clin North Am 73:271-282, 1989.
Levey AS: Measurement of renal function in chronic renal disease.
Kidney Int 38:167-184, 1990.
Lifton RP: Genetic determinants of human hypertension. Proc
Natl Acad Sci USA 92:8545-8551, 1995.
Robertson GL: Physiology of ADH Secretion. Kidney ta32(Suppl
21):S-20, 1987.
Schrier RW: The edematous patient. In Schrier RW (ed) Manual
of Nephrology. Little Brown, Boston, 1990.

CLINICAL EVALUATION OF
RENAL FUNCTION
ANDREW S. LEVEY

INTRODUCTION

This chapter focuses on the clinical assessment of the

glomerular filtration rate (GFR) because it is generally
considered the best index of overall renal function in health
and disease. A brief consideration of other renal functions
is important before discussing the pathophysiology and
clinical evaluation of the GFR in detail.
Glomerular Function

Abnormalities in glomerular function may be classified

into three general categories: alterations in the filtration
of water and small solutes (assessed as the GFR); alterations in the filtration barrier to macromolecules, manifested as proteinuria; and alterations in the continuity of
the glomerular capillary wall, manifested as hematuria. Pro teinuria and glomerular diseases causing hematuria are
discussed in Chapters 4 and 5.
Tubular Function

Because of the diverse functions of the tubules, abnormalities in tubular function are more difficult to classify.
In general, abnormalities are due to defects in the reabsorp tion of water or solutes or the secretion of solutes in local-

Primer on Kidney Diseases, Second Edition

ized nephron segments. They may occur in isolation or in

combination, with or without glomerular abnormalities.
When they occur in combination and with glomerular abnormalities, there is usually widespread renal parenchymal
injury. The severity of the impairment in some tubular
functions, for example, ammonia excretion and urinary
concentration, may correlate better with the severity of
structural damage than does the impairment in GFR. When
abnormalities in tubular function occur in isolation, they
give rise to a number of diverse syndromes, such as renal
tubular acidosis and nephrogenic diabetes insipidus.
GLOMERULAR FILTRATION RATE

Virtues of the GFR as an index of renal function are

that it is a direct measure of renal function, it is reduced
prior to the onset of symptoms of renal failure, the impairment in GFR correlates with the severity of some of the
structural abnormalities observed in chronic renal diseases,
and signs and symptoms of uremia occur when GFR falls
below 5-10 ml/min. Nonetheless, it is not an ideal index.
The GFR is difficult to measure, and the measurements
can be insensitive for detecting renal disease, estimating
its severity, and monitoring its progression.
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

2. Clinical Evaluation of Renal Function

21

Normal Range for GFR

Because GFR varies directly with renal size, which in

turn varies with body size, GFR is conventionally factored
by body surface area. In normal humans, GFR is approximately 125 ml min"1 1.73 nr2 or 180 L day" 1
1.73 m~ 2. The value of 1.73 m2 reflects the average value
for young men and women. Despite adjustment for body
surface area, however, the normal GFR for women is approximately 8% lower than that for men. GFR declines
with age; in men, the decrement is approximately 10 ml
min~' 1.73 irr2 per decade after the age of 40 years (Chapter 61). Hence, the normal GFR in 80-year-old men is
approximately 80 ml min^1 1.73 m" 2. Cross-sectional
studies in normal women indicate roughly similar results.
In children, normal adult values are obtained after 2 years
of age (see Chapter 59).
Determinants of GFR

As described in Chapter 1, glomerular filtration is a pas sive process of ultrafiltration of plasma across the semipermeable glomerular capillary wall. In principle, the level
of GFR is the sum of the filtration rates of each of the
approximately two million nephrons
GFR = NX SNGFR,

(1)

where N is the number of nephrons and SNGFR is single

nephron GFR. The determinants of SNGFR are expressed
in the equation
(2)
SNGFR = Lp X area X Pn c l,
where Lp is hydraulic permeability, area is filtration
area, and Pnet is net ultrafiltration pressure, the difference
between the mean transcapillary hydrostatic (AP) and
oncotic (A;r) pressure gradients. Although not explicit in
this equation, SNGFR varies directly with the renal plasma
flow, because of the relationship among renal plasma flow,
AP, andAw.
It is not possible to measure the number of nephrons,
SNGFR, or its determinants in humans. These factors have
been quantitated in animals with experimentally induced
renal diseases. In many conditions, AP is increased, thereby
sustaining SNGFR and GFR at a level higher than would
be expected from the decline in Lp X area and the number
of nephrons. As a result, the GFR may not accurately
reflect the extent of renal injury. GFR is also affected
by conditions other than renal disease (Table 1). These

conditions affect SNGFR rather than the number of nephrons and may alter the level of GFR in healthy individuals
and in patients with renal disease.
Measurement of GFR from Renal Clearance

GFR cannot be measured directly; it is assessed from

renal clearance. The renal clearance of a substance x is
given by the equation
Cx =

(3)

where Cx is the clearance of x, Ux is the urinary

concentration of x, V is the urine flow rate, and Px is the
average plasma concentration of x during the interval of
urine collection. The term UXV is thus the urinary excretion
rate of x. The value for clearance is related to the
efficiency of excretion; the greater the excretion per unit
concentration in plasma, the higher the clearance.
If substance x is freely filtered across the glomerular
capillary wall and excreted only by glomerular filtration,
its rate of glomerular filtration is equal to its rate of urinary excretion,
GFR X P x = U X V,

(4)

where the term GFR X Px is the filtered load of x. By

substitution into Eq. (3),
C, = GFR.

(5)

Hence, substance x would be defined as an "ideal filtration

marker" whose renal clearance could be used to assess
GFR.
If substance x also undergoes tubular reabsorption or
secretion, then clearance does not equal GFR. The value
for the renal clearance of x would depend not only on the
level of renal function, but also on the mechanism of renal
excretion. Because UXV is the algebraic sum of the rates
of excretion of substance x by glomerular filtration, tubular
reabsorption, and tubular secretion, the clearance of x
would exceed GFR for substances that are filtered and
secreted and would be less than GFR for substances that
are filtered and reabsorbed. Much investigation has been
directed to identifying exogenous and endogenous substances, termed filtration markers, which fulfill the criteria
for an ideal filtration marker and permit accurate, precise,
and convenient assessment of GFR from measurement of
renal clearance.
Relation of GFR to Plasma Solute Concentrations

TABLE I

Conditions Other Than Renal Disease That Affect GFR

Pregnancy
Reduced renal perfusion
Marked surfeit or deficit of extracellular fluid volume
Nonsteroidal anti-inflammatory drugs Acute protein
load and habitual protein intake Blood glucose control
(in diabetics)
Level of arterial blood pressure and class of antihypertensive
agent used for therapy

An important concept for the following discussion is the

steady state of solute balance. A steady state with regard
to substance x is achieved when its rate of generation in
body fluids (from either endogenous production or exogenous intake) is constant and equal to its rate of disappearance from body fluids (from either excretion or metabolism). In these circumstances, the plasma concentration of
x is constant. In the steady state, if substance x is excreted
only in the urine, its generation and urinary excretion are
equal, as depicted in the equation

22

Andrew S. Levey
= U X V,

Estimation of GFR from Creatinine Clearance

(6)

where Gx is the generation rate of x. An important

corollary is that, in the steady state, the rate of generation
of solutes excreted in the urine can be assessed from their
rate of urinary excretion.
From rearrangement of Eq. (3) and substitution of Eq.
(5), it becomes clear that the steady stale plasma concentration of an ideal filtration marker is determined by its rate
of production and its excretion by glomerular filtration.
GFR GFR
Thus, Px is inversely related to GFR, and GFR can
be assessed from the level of P,.

Because of differences among individuals in creatinine

generation and excretion, creatinine clearance provides a

(7)

COMMON CLINICAL MEASUREMENTS TO ASSESS GFR

In clinical practice, GFR is usually estimated from the

serum level or renal clearances of the endogenous filtration
markers, creatinine and urea.
Estimation of GFR from Serum Creatinine

The use of serum creatinine as an index of GFR is based

on two assumptions: (1) creatinine is an ideal filtration
marker whose clearance (Ccr ), therefore, equals GFR; and
(2) creatinine balance is in the steady state with constant
generation (Gcr ) and excretion (C/cr V). Under these conditions, the serum level (Pcr ) would be inversely related to
GFR.
PC, =

{/cr V/GFR = CC,/GFR.

(8)

Neither assumption is entirely correct. Nonetheless,

the level of serum creatinine concentration is generally an
adequate index of GFR for clinical decision making.
However, in some circumstances, serious errors can result
from estimation of GFR from serum creatinine. These are
discussed in a later section.
The magnitude of GFR is sufficient to maintain a low
serum concentration of creatinine. The normal value is
approximately 0.8 to 1.2 mg/dl, with slightly lower values
in women than in men. Figure 1 shows hypothetical changes
in creatinine production, excretion, balance, and serum
creatinine concentration following a 50% decrement in
GFR. A new steady state is reached when the serum creatinine concentration rises sufficiently so that the filtered load
of creatinine, and hence its urinary excretion, again equals
its generation. In the new steady state, the reduced level
of GFR is reflected by a reciprocal increase in the steadystate plasma concentration (Fig. 2).
Thus, in the steady state, changes in serum creatinine
reflect changes in GFR. Because of their reciprocal relationship, a large change in GFR is required to raise serum
creatinine from the normal to the elevated range. However,
once serum creatinine is elevated, even small changes in
GFR raise it substantially more. Expression of the serum
creatinine level as its reciprocal (1/P0) allows the magnitude of decline in GFR to be more clearly appreciated.

1
DAY

FIGURE I Effect of a sudden decrease in GFR on creatinine

excretion, production, balance, and serum creatinine concentration. Reprinted with permission from Kassirer JP: N Engl J Med
285:385-389, 1971.

Per Cent of Normal GFR

80

60

40

20

100
320

32 -

28 _

280 1

24 -

i
o
24O "

,-*

8
f

1
11
- ii

20

L
itii/
ii

16

-^

12

S
1

8
4

____
120

//
-z2&
i -- 1 -- j^j^-1 -- 1 i
100

80

60

40

200
I6O i.
C

120 |
z

80 g
5
40

m
O

20

GFR (ml/min )

FIGURE 2 Relationship of GFR to steady-state levels of serum

creatinine and BUN. Reprinted with permission from Kassirer JP:
N Engl J Med 285:385-389, 1971.

2. Clinical Evaluation of Renal Function

23

more accurate estimate of GFR than serum creatinine.

Most commonly, creatinine clearance is calculated from a
24-hour urine collection for creatinine and a single measurement of serum creatinine, assuming a steady state:
Ccr = Ucr V/Pcr .

(9)

A shorter timed urine specimen collection under supervision reduces inaccuracy due to errors in timing and completeness of collection.
Another advantage of estimating GFR from creatinine
clearance rather than from serum creatinine is that a steady
state is not required. Additional serum measurements can
be obtained to estimate the average serum concentration
during the urine collection. Measurements can be made
either at the midpoint 0f the urine collection or at the
beginning and end of the urine collection and averaged.
Estimation of Creatinine Clearance from Serum Creatinine

As discussed later, creatinine generation can be approximated from age, gender, and body size. Therefore, in the
steady state, creatinine clearance can be estimated from
these factors and the serum level, without urine collection.
The formula derived by Cockcroft and Gault (1976) is
particularly simple to use,
Ccr = ((140 - age) X weight)/(Pcr X 72)
(10)
(men)
Ccr = (((140 - age) X weight)/(Pcr X 72)) X 0.85,
(women)

Estimation of GFR from BUN and Urea Clearance

Urea is also excreted principally by glomerular filtration;

hence, the steady-state urea level also varies inversely with
GFR (Fig. 2). However, due to the renal tubular reabsorption of urea and the many factors that affect urea generation, the serum level and renal clearance of urea are poor
indexes of GFR. Nonetheless, there are special circumstances in which it is useful to measure them.
Urea is usually measured as urea nitrogen, and for historical reasons, the serum level is commonly referred to as
the blood urea nitrogen (BUN). The normal value is approximately 8 to 12 mg/dl; thus, the normal ratio of BUN
to creatinine is approximately 10:1. In principle, a reduction in GFR would lead to a rise in BUN but would not alter
the BUN: creatinine ratio. As discussed later, deviations
in the ratio may indicate alterations in urea reabsorption
or generation.
Most important is the well-recognized relationship of the
level of renal function, the BUN level, and clinical features
of uremia. A useful "rule of thumb" is that a BUN level
greater than 100 mg/dl is associated with a higher risk of
complications in both acute and chronic renal failure and
may indicate the need to initiate dialysis. However, as dis cussed below, if urea generation is reduced, the BUN may
remain below this level even if GFR declines to very low
levels and uremic symptoms arise.

(11)

Measurement of GFR Using Exogenous Filtration Markers

where Ccr is in milliliters per minute, age is in years, weight

is in kilograms, and Pcr is in milligrams per deciliter. These
formulas were derived from individuals without renal dis ease. In patients with renal disease, however, there is a wide
range of differences between the estimates of creatinine
clearance and simultaneously measured GFR.

Inulin, a 5200-Da polymer of fructose, meets the criteria

for an ideal filtration marker, and its renal clearance is
regarded as the "gold standard" for the measurement of
GFR. However, the classic method of inulin clearance requires a continuous intravenous infusion and bladder catheterization and the chemical assay for inulin is difficult.

TABLE 2

Clinical Conditions That Cause Errors in the Estimation of GFR from Measurement of Creatinine Clearance
or Serum Creatinine"
Effect on
Condition

Comment

Plasma ketosis
None
Increase Interference with the picric acid assay for creat inine. Interference
Therapy with certain cephalosporins or
None
Increase with the picric acid and iminohydrolase assays for
fluctyosine Therapy with
creatinine, respectively. Inhibition of tubular secretion of creatinine.
cimetidine or
Decrease
Increase
trimethoprim trimethoprim
Ingesting cooked meat
Increase
Increase
Transient increase in GFR and creatinine generation.
Restriction of dietary protein
Decrease
Decrease
Sustained decrease in GFR and creatinine generation.
Vigorous prolonged exercise
None
Increase
Increase in muscle creatinine generation.
Muscle wasting
None
Decrease
Decrease in muscl e creatinine generation.
Muscle growth
None
Increase
Increase in muscle creatinine generation.
Renal disease*
Increase
Decrease
Decrease in GFR, but stimulation of tubular secretion of
creatinine, and possible decrease in creatinine generation.
" Ccr, creatinine clearance; Fcr, serum creatinine; GFR, glomerular filtration rate. Modified from Levey AS: Assessing the effectiveness
of therapy to prevent the progression of renal disease. Am J Kidney Dis 22:207-214, 1993.
* Effects on Cc r and P c r relative to effects on GFR (i.e., Ccr is higher than expected and P cr is lower than expected for the reduction in
GFR, see text).

24

Andrew S. Levey

Hence, this method is not suitable in practice and remains

largely a research tool.
To simplify the assessment of GFR, investigators have
studied other clearance methods, using a bolus injection
and alternative filtration markers, such as low-molecularweight compounds chelated to radioisotopes. Urine is collected after voiding or the urine collection is omitted and
plasma clearance is measured. In the United States, the
two most frequently used markers are [125I]iothalamate
and [99m Tc]DTPA. Simultaneous measurements of renal
clearance of these markers, by alternative methods, and of
inulin, by classic techniques, reveal similar although not
identical results. These alternative methods for measuring
GFR have become popular in clinical research in recent
years and can be used in practice when estimates of GFR
from endogenous filtration markers are likely to be inaccurate, as discussed below.

tubule. In normal individuals, creatinine secretion accounts

for 5 to 10% of excreted creatinine; hence, clearance exceeds GFR by approximately 10 ml min"1 1.73 m" 2. In
contrast, creatinine secretion is enhanced in patients with
reduced GFR, leading to a greater disparity between creatinine clearance and GFR (Fig. 3). Moreover, the magnitude
of creatinine secretion is variable among individuals and
over time. Therefore, the value for creatinine clearance
overestimates the GFR, and the magnitude of the overes timation is not predictable. The most serious errors occur within the range of GFR from 40 to 80 ml min"1
1.73 m~2, where creatinine clearance can remain normal.
Consequently, it can be difficult to detect the onset or early
progression of the decline in GFR from measurements of
creatinine clearance.
Some drugs, notably trimethoprim and cimetidine, competitively inhibit creatinine secretion, thereby reducing
creatinine clearance and raising the serum creatinine
concentration, despite no effect on GFR. Indeed, some
investigators have recommended assessing GFR from measurements of creatinine clearance during high-dose cimetidine treatment. Clinically, it can be difficult to determine
whether the serum creatinine has risen because of a druginduced inhibition of secretion or a decline in GFR. A clue
that inhibition of creatinine secretion is responsible is an
unchanged blood urea nitrogen concentration, since urea
clearance is unaffected by these drugs.

LIMITATIONS TO CLINICAL ASSESSMENT OF GFR

The following sections describe the renal handling and

metabolism of creatinine and urea and the measurement
of creatinine. The purpose is to highlight clinical conditions
in which errors may occur in the interpretation of the serum
level or renal clearance of creatinine or urea as an index
of GFR (Tables 2 and 3). With appropriate awareness of
these limitations, the careful practitioner can usually make
a satisfactory assessment of the level of renal function from
these measurements.

Creatinine Metabolism

Creatinine is derived largely from the metabolism of

creatine in muscle, which originates in part from dietary
creatine contained in meat. Cooking meat converts a variable fraction of creatine to creatinine which is absorbed and

Renal Handling of Creatinine

Creatinine (molecular mass, 113 Da) is both freely filtered by the glomerulus and also secreted by the proximal

TABLE 3

Clinical Conditions That Cause Errors in the Estimation of GFR from Measurement of Urea Clearance or BUN"
Effect on
Condition
Dehydration
Reduced renal perfusion (volume depletion, congestive

BUN
Increase

Comment
Increased urea reabsorption

Increase

Reduced GFR, increased urea reabsorption,

Increase

Decrease

Increase

Decrease

Reduced urea reabsorption

Increased GFR, reduced urea reabsorption

None

Decrease

Reduced urea generation

None
None

Increase
Increase

Increased urea generation

Increased urea generation

Decrease*

Variable*

Decreased GFR, increased urea reabsorption,

Decrease
Decrease

heart failure)
Overhydration
Increased renal perfusion (volume expansion,
pregnancy, syndrome of inappropriate ADH

increased urea generation

secretion)
Restriction of dietary protein
Increased dietary protein

Accelerated catabolism (fever, trauma, GI bleeding, cell

lysis, therapy with tetracycline or corticosteroids)

Liver disease

decreased urea generation. Decreased GFR, no change in urea reabsorption, decreased urea generation (if dietary protein
Renal disease is restricted).
" CUrea, urea clearance; BUN, blood (serum) urea nitrogen; GFR, glomerular filtration rate.
* Effects on Curea and BUN relative to effects on GFR (e.g., Curoa is lower than expected for the reduction in GFR).

2. Clinical Evaluat ion of Renal Function

100 -

25

1 --------------------------------------------------------- - --------

240 -

90 80 -

200 T3

o>

70 -

e
aT

160 -

6 Oc
c
o o> 50 -

120 -

o
E 40 -

'.

13
"-'

c/ 30 >

f* "* '

20 40 -

1.0 -

} -----1
i ------r1---------\ 1 ------1
] ----- 1\ -----r \
~i \-------1
2O 4O 60 80 1OO 120 140 16O

t:v.^

"** ' ^, v **'"'" ! " * . . . ' . . - . '

Cmuim, ml/mm/1.73m2

20

40

60

80

100

120

140

/] 73m

160

180

FIGURE 3 Relationship of creatinine clearance (left) and serum

creatinine (right) to GFR (inulin clearance) in patients with renal disease. Horizontal lines indicate lower limit of normal
for creatinine clearance (left) and upper limit of normal for serum creatinine (right). Vertical lines indicate lower limit of
normal for GFR. Diagonal line (left) is the line of identity. Shaded areas indicate the proportion of patients in whom
the level of creatinine clearance or serum creatinine is normal despite reduced GF R. Reprinted with permission from
Shemesh et al: Kidney Int 28:830-838, 1985, and Levey et al: Anna Rev Med 39:469-490, 1988.

enters the creatinine pool. Hence, creatinine generation

parallels total muscle mass, and to a lesser extent, meat
intake. Lower values are found in women, children, the
elderly, malnourished individuals, and those with restricted
meat intake, leading to lower serum creatinine levels. As
a corollary, in these groups, serum creatinine may remain
normal despite a reduction in creatinine clearance. Therefore, in detecting a reduction in GFR, an elevated serum
creatinine is an even less-sensitive indicator than a reduced
creatinine clearance. Indeed, GFR can be reduced to as
low as 20 ml mirr1 1.73 m" 2, despite normal serum
creatinine concentration (Fig. 3).
Factors affecting creatinine generation are especially important in patients with chronic renal disease. Meat intake
is often restricted in conjunction with a low-protein diet,
and malnutrition and muscle wasting are common. These
factors tend to blunt the rise in serum creatinine as creatinine clearance declines. In addition, extrarenal elimination
of creatinine can occur, contributing to maintenance of
serum creatinine at a lower level. For these reasons, the
serum creatinine level tends to overestimate the creatinine
clearance, which, as discussed above, tends to overestimate
the level of GFR. Hence, it can be difficult to assess the
level of GFR from the serum creatinine concentration or
changes in GFR from changes in serum creatinine.

Creatinine Measurement

The most important technical difficulty is the imprecision

in serum creatinine assays at low values that are characteris tic of the normal range, especially in women and children.
Hence, it can be difficult to determine whether changes in
the serum level within the normal range represent analytical variability or changes in GFR. In addition, various substances can interfere with the serum assays, leading to spurious elevations in the serum level (Table 2).
Renal Handling of Urea

Urea (molecular mass, 60 Da) is freely filtered by the

glomerulus. However, 35 to 65% of the filtered load is
reabsorbed by the proximal and distal tubules. Hence, urea
clearance is less than GFR. Urea reabsorption varies inversely with renal blood flow and the state of hydration
(Table 3). Dehydration, reduced renal perfusion, or the
early phase of obstruction of the urinary tract causes increased urea reabsorption, decreased urea clearance, and
a high BUN: creatinine ratio. Conversely, overhydration
or increased renal perfusion causes decreased urea reabsorption, increased urea clearance, and a reduced BUN:
creatinine ratio. The BUN: creatinine ratio is useful in the
differential diagnosis of acute or chronic renal insuffi-

Andrew S. Leve y

26

ciency. Reduced renal perfusion (e.g., extracellular volume contraction or congestive heart failure) or recent-onset
urinary tract obstruction usually causes the BUN: creatinine ratio to rise to >20:1. In contrast, in acute tubular
necrosis and most causes of chronic renal disease, the
BUN: creatinine ratio remains normal (approximately
10:1). As discussed below, the BUN : creatinine ratio also
can be elevated in states of increased urea generation associated with acute or chronic renal insufficiency.
Another useful application of the measurement of urea
clearance is to assess the overall level of renal function
in patients with advanced chronic renal disease. As GFR
declines, urea reabsorption is reduced and affected less by
the state of hydration. By coincidence, the proportion of
the filtered load of urea that is reabsorbed approximates
the proportion of excreted creatinine that is secreted. In
patients with GFR <15 ml mirr1 1.73 irr2, the average
of the values for urea and creatinine clearance approximates the GFR. In patients with chronic renal failure
treated by dialysis, residual renal function is generally measured by urea clearance. Values for urea clearance <1 ml/
min indicate negligible residual renal function.
Urea Metabolism

Urea is the end product of protein catabolism and is

synthesized primarily by the liver. Approximately one
quarter of synthesized urea is metabolized in the intestine
to carbon dioxide and ammonia. The ammonia thus generated returns to the liver where it is reconverted to urea.
In the steady state, the level of BUN is an index of urea
generation. Dietary protein intake is the principal determinant of urea generation (Table 3). A high dietary protein
intake raises the BUN and consequently the BUN: creatinine ratio. Other factors that increase urea generation in-

clude accelerated catabolism of endogenous protein, as

occurs during febrile illnesses, trauma, gastrointestinal
bleeding, tumor lysis or therapy with tetracyclines or corticosteroids (Table 3). On the other hand, urea generation
may be reduced in liver disease leading to a BUN: creatinine ratio <10:1.
Bibliography
Cockcroft DW, Gault MH: Prediction of creatinine clearance from
serum creatinine. Nephron 16:31-41, 1976.
Kassirer JP: Clinical evaluation of kidney function: Glomerular
function. N Engl J Med 285:385-389, 1971.
Levey AS: Use of measurements of GFR to assess the progression
of renal disease. Semin Nephrol 9:370-379, 1989.
Levey AS: Nephrology forum: Measurement of renal function in
chronic renal disease. Kidney Int 38:167-184, 1990.
Luke RG: Uremia and the BUN. N Engl J Med 305:1213-1215,
1981.
Miller TR, Anderson RJ, Linas SL, Henrich WL, Berns AS, Gabow PA, Schrier RW: Urinary diagnostic indices in acute renal
failure: A prospective study. Ann Int Med 89:47-50, 1978.
Perrone RD, Madias NE, Levey AS: Serum creatinine as an index
of renal function: New insights into old concepts. Clin Chem
38:1933-1953, 1992.
Rolin HA, Hall PM, Wei R: Inaccuracy of estimated creatinine
clearance for prediction of iothalamate glomerular filtration
rate. Am J Kidney Dis 4:48-54, 1984.
Roubenoff R, Drew H, Moyer M, Petri M, Whiting-O'Keefe Q,
Hellmann DB: Oral cimetidine improves the accuracy and precision of creatinine clearance. Ann Intern Med 113:501-506,1990.
Shemesh O, Golbetz H, Kriss JP, Myers BD: Limitations of creatinine as a filtration marker in glomerulopathic patients. Kidney
Int 28:830-838, 1985.
Smith HW: Diseases of the kidney and urinary tract. In: The
Kidney: Structure and Function in Health and Disease. Oxford
University Press, New York, Chap. 26, pp. 836-887, 1951.
Walser M: Progression of chronic renal failure in man. Kidney Int
37:1195-1210, 1990.

URINALYSIS
ARTHUR GREENBERG

Specific Gravity

The microscopic examination of the urine sediment is

an indispensable part of the evaluation of patients with
renal insufficiency, proteinuria, hematuria, urinary tract
infection, or nephrolithiasis. This task must not be delegated; it should be performed personally. The relatively
simple chemical tests performed in the routine urinalysis
provide important information about a number of primary
renal and systemic disorders. Examination of the urine
sediment gives valuable clues about the renal parenchyma.
The features of a complete urinalysis are listed in Table 1.

The specific gravity of a fluid is the ratio of its weight

to the weight of an equal volume of distilled water. The
urine specific gravity is a conveniently determined but inaccurate surrogate for osmolality. Specific gravities of
1.001 to 1.035 correspond to an osmolality range of 50 to
1000 mOsm/kg. A specific gravity near 1.010 connotes isosthenuria, with a urine osmolality matching that of plasma.
Relative to osmolality, the specific gravity is elevated when
dense solutes such as protein, glucose, or radiographic contrast are present. Formulas that correct specific gravity for
glycosuria or proteinuria are available but of little practical
utility. It is better to measure osmolality directly.
Three methods are available for specific gravity measurement. The hydrometer is most accurate, but it requires a
sufficient volume of urine to float the hydrometer as well as
equilibration of the specimen to the hydrometer calibration
temperature. The second method is based on the wellcharacterized relationship between urine specific gravity
and refractive index. Refractometers calibrated in specific
gravity units are commercially available; they require only
a drop of urine. Finally, the specific gravity may also be
estimated by dipstick.
The specific gravity is used to determine whether the
urine is or can be concentrated. During a solute diuresis
accompanying hyperglycemia, diuretic therapy, or relief of
obstruction, the urine is isosthenuric. In contrast, with a
water diuresis due to overhydration or diabetes insipidus,
the specific gravity is typically 1.004 or lower. In the absence
of proteinuria, glycosuria, or iodinated contrast administration, a specific gravity of more than 1.018 implies preserved
concentrating ability. Measurement of specific gravity is
useful in differentiating between prerenal azotemia and
acute tubular necrosis (ATN) and in assessing the import
of proteinuria observed in a randomly voided urine.

SPECIMEN COLLECTION AND HANDLING

Urine should be collected with a minimum of contamination. A clean catch midstream collection is preferred. If
this is not feasible, bladder catheterization is appropriate
in adults; the risk of a urinary tract infection following a
single catheterization is insignificant. Suprapubic aspiration
is employed in infants. In the uncooperative male patient,
a clean, freshly applied condom catheter and urinary collection bag may be used. Urine in the collection bag of a
patient with an indwelling bladder catheter is subject to
stasis. Fresh urine may be collected by withdrawing urine
from above a clamp placed on the drainage tube.
The chemical composition of the urine changes with
standing and the formed elements degenerate over time.
The urine is best examined fresh; a brief period of refrigeration is acceptable. Because bacteria multiply at room temperature, bacterial counts from unrefrigerated urine are
unreliable. High urine osmolality and low pH favor cellular
preservation. These two characteristics of the first voided
morning urine give it particular value in evaluating for
suspected glomerulonephritis.
PHYSICAL AND CHEMICAL PROPERTIES OF THE URINE
Appearance

Chemical Composition of the Urine Dipstick Methodology

Normal urine is clear, with a faint yellow tinge due to

the presence of urochromes. As the urine becomes more
concentrated, its color deepens. Bilirubin, other pathologic
metabolites, and a variety of drugs may discolor the urine
or change its smell. Suspended erythrocytes, leukocytes, or
crystals may render the urine turbid. Conditions associated
with a change in the appearance of the urine are listed in
Table 2.

Primer on Kidney Diseases, Second Edition

Tedious individual assays of protein, glucose (or other

reducing substances), or ketones have been supplanted by
more convenient dipstick methods. The urine dipstick is a
plastic strip to which paper tabs impregnated with chemical
reagents have been glued. The reagents in each strip are
chromogenic. After timed development, the color on the
paper segment is compared to a chart. Some reactions
27

Copyright 1 998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

28

Arthur Greenberg

TABLE I

TABLE 2

The Routine Urinalysis

Selected Substances That May Alter the Physical Appearance

or Odor of the Urine
Color change
Substances

Appearance Specific
gravity Chemical tests
(Dipstick)
pH
Protein
Glucose
Ketones
Blood
Urobilinogen
Bilirubin
Nitrites

White
Pink/red/brown

Yellow/orange/brown

Leukocyte esterase
Microscopic examination (formed elements)
Crystals: Urate; calcium phosphate, oxalate, or carbonate,
triple phosphate; cystine; drugs Cells: Leukocytes,
erythrocytes, renal tubular cells, oval fat
bodies, transitional epithelium, squamous Casts:
Hyaline, granular, RBC, WBC, tubular cell,
degenerating cellular, broad, waxy, lipid-laden Infecting
organisms: Bacteria, yeast, trichomonas, nematodes
Miscellaneous: spermatozoa, mucous threads, fibers, starch,

Brown/black

Blue or green
green/brown

hair, and other contaminants

Odor

are highly specific. Others are sensitive to the presence of

interfering substances or extremes of pH. Heavy discoloration of the urine with bilirubin or blood may obscure the
color changes.

Sweet or fruity
Ammoniacal
Maple syrup
Musty or mousy
Sweaty feet

pH

Rancid

The pH test pads employ indicator dyes that change

color with pH. The physiologic urine pH ranges from 4.5
to 8.0. The determination is most accurate if done promptly,
because growth of urea-splitting bacteria and loss of CC>2
raise pH. In addition, bacterial metabolism of glucose may
produce organic acids and lower pH. These strips are not
sufficiently accurate to be used for the diagnosis of renal
tubular acidosis.
Protein

Protein measurement uses the protein-error-of-indicators principle. The pH at which some indicators change
color varies with the protein concentration of the bathing
solution. Protein indicator strips are buffered at an acid
pH near their color change point. Wetting them with a
protein containing specimen induces a color change. The
protein reaction may be scored from trace to 4+ or by
concentration. Their equivalence is as follows: trace, 5 to
20 mg/dL; l + , 30 mg/dL; 2+, 100 mg/dL; 3+, 300 mg/
dL; 4+, greater than 2000 mg/dL. Highly alkaline urine,
especially after contamination with quaternary ammonium
skin cleansers, may produce false-positive reactions. Protein strips are highly sensitive to albumin but less so to
globulins, hemoglobin, or light chains. When light-chain
proteinuria is suspected, an acid precipitation assay must
be employed. An acid that denatures protein is added to
the urine specimen and the density of precipitate is related

Chyle, pus, phosphate crystals

Erythrocytes, hemoglobin, myoglobin,
porphyrins, beets, senna, cascara,
levodopa, methyldopa,
deferoxamine, phenolphthalein and
congeners, food colorings,
metronidazole, phenacetin
Bilirubin, urobilin, phenazopyridine
urinary analgesics, senna, cascara,
mepacrine, iron compounds,
nitrofurantoin , riboflavin, rhubarb,
sulfasalazine, rifampin, fluorescein,
phenytoin
Methemoglobin, homogentisic acid
(alcaptonuria), melanin (melanoma),
levodopa, methyldopa
Biliverdin, Pseudomonas infection,
dyes (methylene blue and indigo
carmine), triamterene, vitamin B
complex, methocarbamol, indican,
phenol, chlorophyll
Substance or condition

Ketones
Urea-splitting bacterial infection Maple
syrup urine disease Phenylketonuria
Isovaleric or glutaric acidemia or excess
butyric or hexanoic acid
Hypermethioninemia, tyrosinemia

to the protein concentration. Urine

negative by dipstick but positive with sulfosalicyclic acid is
highly suspicious for light chains. Tolbutamide, high-dose
penicillin, sulfon-amides, and radiographic contrast may
give false-positive turbimetric reactions. If the urine is
very concentrated, the presence of a modest amount of
protein is less likely to correspond to significant
proteinuria on a 24-hour basis. The protein indicator used
for routine dipstick analysis is not sensitive enough to
detect microalbuminuria. Special dipsticks are available
for screening for microalbuminuria in incipient diabetic
nephropathy (see Chapters 5 and 29).
Blood

Reagent strips for blood rely on the peroxidase activity

of hemoglobin to catalyze an organic peroxide with subsequent oxidation of an indicator dye. Free hemoglobin pro duces a homogeneous color. Intact red cells cause punctate
staining. False-positive reactions occur if the urine is contaminated with other oxidants such as povidone-iodine,
hypochlorite, or bacterial peroxidase. Ascorbate causes
false negatives. Myoglobin is also detected, because it has
intrinsic peroxidase activity. A urine that is positive for
blood by dipstick but shows no red cells on microscopic

3. Urinalysis

examination is suspect for myoglobinuria or hemoglobinuria. Pink discoloration of serum may occur with hemolysis,
but free myoglobin is seldom present in a concentration
sufficient to change the color of plasma. A specific assay
for urine myoglobin will confirm the diagnosis.
Specific Gravity
Specific gravity reagents strips actually measure ionic
strength using indicator dyes with ionic-strength-dependent pK a's. They do not detect glucose or nonionic radiographic contrast.
Glucose
In contrast to the obsolete copper sulfate-based tablet
tests that detected any reducing substances, modern dipstick reagent strips are specific for glucose. They rely on
glucose oxidase to catalyze the formation of hydrogen peroxide which reacts with peroxidase and a chromogen to
produce a color change. High concentrations of ascorbate
or ketoacids reduce test sensitivity. However, the degree
of glycosuria occurring in diabetic ketoacidosis is sufficient
to prevent false negatives despite ketonuria.
Ketones
Ketone reagent strips depend on the development of a
purple color after acetoacetate reacts with nitroprusside.
They do not detect acetone or /3-hydroxybutyrate. False
positives may occur in patients taking levodopa.
Urobilinogen
Urobilinogen is a colorless substance produced in the
gut from metabolism of bilirubin. Some is excreted in feces
and the rest reabsorbed and excreted in the urine. In obstructive jaundice, bilirubin does not reach the bowel, and
urinary excretion of Urobilinogen is diminished. In other
forms of jaundice, Urobilinogen is increased. The urobilinogen test is based on the Ehrlich reaction in which diethylaminobenzaldehyde reacts with Urobilinogen in acid medium to produce a pink color. Sulfonamides may produce
false positives and degradation of Urobilinogen to urobilin
false negatives.
Bilirubin
Bilirubin reagent strips rely on the chromogenic reaction of bilirubin with diazonium salts. Conjugated bilirubin
is not normally present in the urine. False positives
may be observed in patients receiving chlorpromazine
or phenazopyridine. False negatives occur in the presence
of ascorbate.
Nitrite

This screening test for bacteriuria relies on the ability

of gram negative bacteria to convert urinary nitrate to
nitrite which activates a chromogen. False-negative results
occur with infection with enterococcus or other organisms
that do not produce nitrite, when ascorbate is present, or
when urine has not been retained in the bladder long
enough (approximately 4 hours) to permit sufficient pro duction of nitrite from nitrate.

29
Leukocytes
Granulocyte esterases can cleave pyrrole amino acid es ters, producing free pyrrole that subsequently reacts with
a chromogen. The test threshold is 5 to 15 WBC/HPF.
False negatives occur with glycosuria, high specific gravity,
cephalexin or tetracycline therapy, or excessive oxalate
excretion. Contamination with vaginal debris may give a
positive test result without true urinary tract infection.
MICROSCOPIC EXAMINATION OF
THE SPUN URINARY SEDIMENT
Specimen Preparation and Viewing

Standardization of procedure ensures comparability of

serial urine specimens. Twelve milliliters of urine should
be spun in a conical centrifuge tube for 5 minutes at 1500
to 2000 rpm (450g). After centrifugation, the tube is inverted and drained. The pellet is resuspended in the few
drops of urine that remain in the tube after inversion by
flicking the base of the tube gently with a finger or with
the use of a pipette. Care should be taken to fully suspend
the pellet without excessive agitation. A drop of urine is
poured onto a microscope slide or transferred with the
pipette. The drop should be sufficient in size that a standard
22 X 22-mm coverslip just floats on the urine. If too little
is used, the specimen rapidly dries out. If an excess of urine
is applied, it will spill onto the microscope objective or
stream distractingly under the coverslip. Rapid commercial
urine stains or the Papanicolaou stain may be used to enhance detail. Most nephrologists prefer the convenience of
viewing unstained urine. Subdued light is necessary. The
condenser and diaphragm are adjusted to maximize contrast and definition. When the urine is dilute and few
formed elements are present, detection of motion of objects
suspended in the urine ensures that the focal plane is correct. One should scan the urine at low power (100X) to
obtain a general impression of its contents before moving
to high power (400X) to look at individual fields. It is useful
to scan large areas at low power and move to high power
when a structure of interest is identified. Cellular elements
should be quantitated by counting or estimating the number in at least ten representative high power fields. Casts
may be quantitated by counting the number per low power
field, although most observers use less specific terms such
as rare, occasional, few, frequent, and numerous.
Cellular Elements
The principal formed elements of the urine are listed in
Table 1. The figures constitute an atlas of selected
formed elements.
Erythrocytes

Red blood cells (RBC; Figs. 1A and IB) may find their
way into the urine from any source between the glomerulus
and the urethral meatus. The presence of more than two
or three erythrocytes per HPF is usually pathologic. Erythrocytes are biconcave discs 7 /^m in diameter. They become

30

Arthur Greenberg

FIGURE I Cellular elements in the urine. In this and subsequent figures (Figs. 2-4), all photographs were made from
unstained sediments and, except as specified, photographed at 400X original magnification. (A) Nondysmorphic
red blood cells. Note that they appear as uniform, biconcave disks. (B) Dysmorphic red blood cells from a patient with
IgA nephropathy. Their shape is irregular, with membrane blebs and spicules. (C) White blood cells from a patient
with an indwelling bladder catheter. Numerous individual (small arrows) and budding yeast (single thick arrow) as well
as hyphal forms (open arrow) are present. (D) Renal tubular epithelial cells. Note the variability of shape. The erythrocytes in the background are much smaller. (E) Squamous epithelial cells. (F) Transitional epithelial cells in a characteristic
clump.

3. Urinal ysis

FIGURE 2 Casts. (A) Hyaline cast. (B) "Muddy brown" granular casts and amorphous debris from a patient with ATN
(original magnification, 100X). (C) Waxy cast (open arrows) and granular cast (solid arrow) from a patient with lupus
nephritis and a telescoped sediment. Note the background hematuria. (D) Tubular cell cast. Note the hyaline cast matrix.
(E) Red blood cell cast. Background hematuria is also present.

31

32

Arthur Greenberg

crenated in hypertonic urine. In hypotonic urine, they swell

or burst, leaving ghosts. Erythrocytes originating in the
renal parenchyma are dysmorphic, with spicules, blebs, submembrane cytoplasmic precipitation, membrane folding,
and vesicles. Those originating in the collecting system
retain their uniform shape. Some experienced observers
report success differentiating renal parenchymal from collecting system bleeding by systematic examination of erythrocytes using phase contrast microscopy.
Leukocytes

Polymorphonuclear leukocytes (PMN; Fig. 1C) are approximately 12 /am in diameter and are most readily recognized in a fresh urine before their multilobed nuclei or
granules have degenerated. Swollen PMNs with prominent
granules displaying Brownian motion are termed "glitter"
cells. PMNs indicate urinary tract inflammation. They may
occur with intraparenchymal diseases such as glomerulonephritis or interstitial nephritis. They are a prominent feature of upper or lower urinary tract infection. In addition,

they may appear with periureteral inflammation as in regional ileitis or acute appendicitis.
Renal Tubular Epithelial Cells

Tubular cells (Fig. ID) are larger than PMNs, ranging

from 12 to 20 /^m. Proximal tubular cells are oval- or eggshaped and tend to be larger than the cuboidal distal tubular cells, but since their size varies with urine osmolality,
they cannot be reliably differentiated. In hypotonic urine,
it may be difficult to distinguish tubular cells from swollen
PMNs. A few tubular cells may be seen in a normal urine.
More commonly, they indicate tubular damage or inflammation from ATN or interstitial nephritis.
Other Cells

Squamous cells (Fig. IE) of urethral, vaginal, or cutaneous origin are large, flat cells with small nuclei. Transitional
epithelial cells (Fig. IF) line the renal pelvis, ureter, bladder, and early urethra. They are rounded cells several times
the size of leukocytes and often occur in clumps. In hypo-

FIGURE 3 Lipid. (A) Oval fat bodies, bright field illumination. (B) Same field under polarized light. Characteristic
Maltese cross shown at arrow. (C) Lipid-laden cast, bright field illumination. (D) Same field under polarized light.
Characteristic Maltese cross shown at arrow.

3. Urinalysis

33

tonic urine, they may be confused with swollen tubular

epithelial cells.
Casts and Other Formed Elements

Casts are appropriately named. As demonstrated by immunofluorescence studies, they consist of a matrix of
Tamm-Horsfall urinary mucoprotein in the shape of the
distal tubular or collecting duct segment in which they
were formed. The matrix has a straight margin helpful in
differentiating casts from clumps of cells or debris.
Hyaline Casts

Hyaline casts consist of mucoprotein alone. Because

their refractive index is very close to that of urine, they
may be difficult to see, thus requiring subdued light and
careful manipulation of the iris diaphragm. Hyaline casts
are nonspecific. They occur in concentrated normal urine
as well as numerous pathologic conditions (Fig. 2A).
Granular Casts

Granular casts consist of finely or coarsely granular material. Immunofluorescence studies show that fine granules
derive from altered serum proteins. Coarse granules may
result from degeneration of embedded cells. Granular casts
are nonspecific but usually pathologic. They may be seen
after exercise or with simple volume depletion and as a
finding in ATN, glomerulonephritis, or tubulointerstitial
disease (Figs. 2B and 2C).
Waxy Casts

Waxy casts or broad casts are made of hyaline material

with a much higher refractive index than hyaline casts,

hence their waxy appearance. They behave as though they

are more brittle than hyaline casts and frequently have
fissures along their edge. Broad casts form in tubules that
have become dilated and atrophic due to chronic parenchymal disease (Fig. 2C).
Red Blood Cell Casts

RBC casts indicate intraparenchymal bleeding. The hallmark of glomerulonephritis, they are less frequently seen
with tubulointerstitial disease. RBC casts have been described along with hematuria in normal individuals after
exercise. Fresh RBC casts retain their brown pigment and
consist of readily discernible erythrocytes in a tubularshaped cast matrix (Fig. 2D). Over time, the heme color
is lost along with the distinct cellular outline. With further
degeneration, RBC casts are hard to distinguish from
coarsely granular casts. RBC casts may be diagnosed by
the company they keep. They appear in a background of
hematuria with dysmorphic red cells, granular casts, and
proteinuria. Occasionally, the evidence for intraparenchymal bleeding is a hyaline cast with embedded red cells.
These have the same pathophysiologic implication as
RBC casts.
White Blood Cell Casts

WBC casts consist of white cells in a protein matrix.

They are characteristic of pyelonephritis and are useful in
distinguishing this disorder from lower tract infection. They
may also be seen with interstitial nephritis and other
tubulointerstitial disorders.
Tubular Cell Casts

These casts consist of a dense agglomeration of sloughed

tubular cells or just a few tubular cells in a hyaline matrix.

TABLE 3

Common Urinary Crystals

Description
Crystals found in acid urine
Amorphous

Rhomboid prisms
Rosettes Bipyramidal
Dumbbell shaped
Needles

Composition

Comment

Uric acid
Sodium urate

Cannot be distinguished from amorphous

phosphates except by urine pH, may be
orange tinted by urochromes

Uric acid Uric

acid Calcium
oxalate Calcium
oxalate
Uric acid, sulfa drugs, radiographic
contrast

Hexagonal plates
Cystine
Crystals found in alkaline urine
Amorphous Coffin-lid (beveled
rectangular prisms)
Granular masses or dumbbells
Yellow- brown masses with or
without spicules

Phosphates
Triple (magnesium ammonium) phosphate
Calcium carbonate
Ammonium biurate

Also termed "envelope-shaped"

Clinical history provides useful
confirmation. Sulfa may resemble
sheaves of wheat; urate and contrast
crystals are thicker.
Presence may be confirmed with
nitroprusside test
Indistinguishable from urates except by pH
Seen with urea- splitting infection and
bacteriuria Larger than amorphous
phosphates

Arthur Greenberg

\ i1?! -: I";' * s
/ >.%! \',

Ti-t

v,.</
!
ft...

t-

FIGURE 4 Crystals. (A) Hexagonal cystine and bipyramidal or "envelope-shaped" oxalate (arrows). Photo courtesy
of Dr. Thomas O. Pitts. (B) "Coffin -lid"- shaped triple phosphate. (C) Dumbell- shaped oxalate. (D) Rhomboid urate.
(E) Needle- shaped urate.

3. Urinalysis

35

They occur in concentrated urine, but are more characteris tically seen with the sloughing of tubular cells that occurs
with ATN (Fig. 2E).
Bacteria, Yeast, and Other Infectious Agents

Bacillary or coccal forms of bacteria may be discerned

even on an unstained urine. Examination of a gram stain
preparation of unspun urine allows estimation of the bacterial count. One organism per HPF of unspun urine corresponds to 20,000 organisms/mm3. Individual and budding
yeasts and hyphal forms occur with Candida infection or
colonization. Candida organisms are similar in size to red
cells but they are greenish spheres, not biconcave discs.
When budding forms or hyphae are present, yeast are obvious (Fig. 1C). Trichomonas organisms are identified by
their teardrop shape and motile flagellum.
Lipiduria

In the nephrotic syndrome with lipiduria, tubular cells

reabsorb luminal fat. Sloughed tubular cells containing fat
droplets are called oval fat bodies. Fatty casts contain lipid
laden tubular cells or free lipid droplets. By light microscopy, lipid droplets appear round and clear with a green
tinge. Cholesterol esters are anisotropic; cholesterol-containing droplets rotate polarized light, producing a Maltese
cross appearance under polarized light. Triglycerides appear similar by light microscopy, but they are isotropic.
Crystals, starch granules, mineral oil, and other urinary
contaminants are also anisotropic. Before concluding that
anisotropic structures are lipid, the observer must compare
polarized and bright field views of the same object (Fig. 3).
Crystals

Crystals may be present spontaneously or precipitate

with refrigeration of a specimen. They may be difficult to
identify because of similar shapes; the common urinary
crystals are described in Table 3. The pH is an important
clue to identity, because solubility of a number of urinary
constituents is pH dependent. The three most distinctive
crystal forms are cystine, calcium oxalate, and magnesium
ammonium (triple) phosphate. Cystine crystals are hexagonal plates resembling benzene rings. Calcium oxalate crystals are classically described as envelope shaped, but when
viewed as they rotate in the urine under the microscope,
they appear bipyramidal (Fig. 4A). "Coffin lid"-shaped
triple phosphates are rectangular with beveled ends (Fig.
4B). Oxalate may also occur in dumbell-shaped crystals
(Fig. 4C). Urate may have several forms, including rhomboids (Fig. 4D) or needles (Fig. 4E).
Characteristic Urine Sediments

The urine sediment is a rich source of diagnostic information. Occasionally, a single finding, e.g., cystine crystals, is
pathognomonic. More often, the sediment must be considered as a whole and interpreted in conjunction with clinical

and other laboratory findings. Several patterns bear emphasis.

In the acute nephritic syndrome, the urine may be pink
or pale brown and turbid. Blood and moderate proteinuria
are present by dipstick. The microscopic examination
shows RBCs and RBC casts as well as granular and hyaline
casts. WBC casts are rare. In the nephrotic syndrome, the
urine is clear or yellow. Increased foaming may be noted
because of the elevated protein content. In comparison to
the sediment of nephritic patients, the nephrotic sediment
is bland. Hyaline casts and lipiduria with oval fat bodies
or lipid-laden casts predominate. Granular casts and a few
tubular cells may also be present along with a few RBCs.
With some forms of chronic glomerulonephritis, a telescoped sediment is observed. This term refers to the pres ence of the elements of a nephritic sediment together with
broad or waxy casts indicative of tubular atrophy and dipstick findings of heavy proteinuria (Fig. 2C). In pyelonephritis, WBC casts and innumerable WBCs are present
along with bacteria. In lower tract infection, WBC casts
are absent. The sediment in ATN shows tubular cells, tubular cell casts, and muddy brown granular casts (Fig. 2B).
The typical urinary findings in individual renal disorders
are discussed in their respective chapters.
Bibliography
Birch DF, Fairley KF, Becker GJ, Kincaid- Smith P: A Color Atlas
of Urine Microscopy. New York, Chapman & Hall, 1994.
Braden GL, Sanchez PG, Fitzgibbons JP, Stupak WJ, Germain
MJ: Urinary doubly refractile lipid bodies in nonglomerular
renal disease. Am J Kidney Dis 11:332-337, 1988.
Fairley KF, Birch DF: Hematuria: a simple method for identifying
glomerular bleeding. Kidney Int 21:105-108, 1982.
Fassett RG, Owen JE, Fairley J, Birch DF, Fairley KF: Urinary redcell morphology during exercise. Br Med J 285:1455-1457,1982.
Graff L: A Handbook of Routine Urinalysis. Lippincott, Philadelphia, 1983.
Kincaid- Smith P: Haematuria and exercise-related haematuria. Br
MedJ 285:1595-1597, 1982.
McQueen EG: Composition of urinary casts. Lancet 2:397-398,
1966.
Raymond JR, Yarger WE: Abnormal urine color: differential diagnosis. Southern Med J 81:837-841, 1988.
Rutecki GJ, Goldsmith C, Schreiner GE: Characterization of proteins in urinary casts. Fluorescent-antibody identification of
Tamm-Horsfall mucoprotein in matrix and serum proteins in
granules. N Engl J Med 284:1049-1052, 1971.
Schumann GB, Harris S, Henry JB: An improved technique for
examining urinary casts and a review of their significance. Am
J Clin Pathol 69:18-23, 1978.
Schumann GB, Schweitzer SC: Examination of the urine. In: Henry
JB (ed) Clinical Diagnosis and Management by Laboratory
Methods. 18th ed. Saunders, Philadelphia, pp. 387-444, 1991.
Stamey TA, Kindrachuk RW: Urinary Sediment and Urinalysis.
A Practical Guide for the Health Professional. Saunders, Philadelphia, 1985.
Voswinckel P: A marvel of colors and ingredients. The story of
urine test strips. Kidney Int 46:S3 - S7, 1994.
Zimmer JG, Dewey R, Waterhouse C, Terry R: The origin and
nature of anisotropic urinary lipids in the nephrotic syndrome.
Ann Intern Med 54:205-214, 1961.

HEMATURIA
CLIFFORD E. KASHTAN

INTRODUCTION

Hematuria, while potentially benign, is frequently a harbinger of disease in the kidneys or urinary tract. As such
it requires the conscientious attention of the clinician. In
this section the causes of hematuria will be reviewed, and
an approach to its diagnostic evaluation presented. Differences between the pediatric and adult populations will be
stressed, as will easily avoided pitfalls that may trap the
unwary. The reader is encouraged to consult relevant chapters for detailed descriptions of the various entities mentioned herein.
EPIDEMIOLOGY

Hematuria, the presence of excessive numbers of red

blood cells (RBCs) in the urine, may be macroscopic
(gross), in which the urine is pink, red, or cola-colored, or
microscopic, i.e., visible only with aid of a microscope.
Normal individuals may excrete as many as 104 to 105 RBCs
in the urine during a f2-hour period, corresponding to
several RBCs in the sediment of a randomly collected,
centrifuged urine specimen examined with a high-power
objective. Thus, a working definition of hematuria is the
presence of greater than 4 RBCs per high-power field
(HPF) of urine sediment.
Studies of school children have indicated that the prevalence of microhematuria is high (about 4%). fn the majority
of children with a urine specimen showing microhematuria,
follow-up urinalyses are normal, and these children do not
develop urinary tract pathology. For this reason, in a child
with isolated microhematuria, a second specimen should
be obtained before an extensive evaluation.
Among an unselected population of adults over 35 years
of age, the prevalence of asymptomatic microhematuria
(>1 RBC/HPF of sediment) was found to be about 13%.
The positive predictive value of microhematuria for serious
urologic disease was low overall, being most useful in elderly men. As in the pediatric population, patients with
excessive numbers of RBCs in several urine samples are
more likely to have significant underlying urinary tract pathology.
The causes of hematuria (Tables 1 to 3) are age-dependent. Although glomerular causes predominate in children
and young adults, fewer than 5% of cases of hematuria in
patients over 40 years of age result from glomerular lesions.

Primer on Kidney Diseases, Second Edition

Neoplasms of the urinary tract account for 15 to 20% of

patients with hematuria after 40 years of age, and approximately two-thirds of these are carcinomas of the bladder.
In children, the neoplastic disorders that cause hematuria
are Wilms tumors and rhabdomyosarcomas of the bladder,
and make up only a small fraction of cases of hematuria.
However, because early recognition and treatment of
Wilms tumor result in excellent outcomes, prompt imaging
of the upper urinary tract is imperative in the child younger
than 3 years of age with hematuria. Urinary tract calculi
are a common cause of hematuria in adults. While calculi
may cause hematuria in children, their incidence is low
except in certain defined populations (see below). The neonate with hematuria is much more likely to have renal
vein thrombosis or renal cortical infarction (associated with
umbilical arterial catheters) than are older children or
adults. Urinary tract infection (UTI) is a frequent cause
of hematuria in children and adults, with prostate disease
being particularly important in older males.
The epidemiologic characteristics of hematuria clearly
affect the approach to evaluation. Although the yield from
cystoscopy is relatively high in the adult patient with hematuria, cystoscopy is rarely indicated in the child. Children
with hematuria should be evaluated by a pediatrician or
nephrologist, with referral to a urologist if initial evaluation
suggests cystoscopy will aid in localizing the source of hematuria.
It should be recalled that urine dipsticks identify myoglobin and filtered free hemoglobin as well as hemoglobin
released by RBCs undergoing lysis on the reagent strip
(see Chapter 3). For this reason, dipsticks may be used to
screen patients for hematuria but results need to be validated by microscopic examination of the urine. Similarly,
a variety of synthetic and natural products can alter the
appearance of the urine, and urine microscopy will quickly
determine whether the change in color is due to the pres ence of RBCs. In infants, urates can cause brownish diaper
stains that may be mistaken for hematuria.
EVALUATION (Fig. I)
Level I: History, Physical Examination, and Urinalysis
(Table 4)

The initial evaluation of the patient with hematuria

consists of a careful history and physical examination,
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

4. Hematuria

37
TABLE I

TABLE 2

Hematuria: Glomerular Causes

Hematuria: Nonglomerular Renal Causes

Proliferative glomerulonephritis
Primary
IgA nephropathy Postinfectious
glomer ulonephritis Membranoproliferative
glomerulonephritis Idiopathic rapidly
progressive (crescentic)
glomerulonephritis Fibrillary glomerulonephritis Secondary
(associated with systemic diseases) Nephritis of anaphylactoid
(Henoch- Schonlcin) purpura Systemic lupus erythematosus
Antiglomerular basement membrane nephritis (including
Goodpasture syndrome) Systemic vasculitis Chronic
bacteremia (subacute bacterial endocarditis,
"shunt nephritis") Essential mixed
cryoglobulinemia Nephritis associated
with hepatitis B or C
Nonproliferative glomerulopathies Minimal change
nephrotic syndrome Focal (global or segmental)
glomerulosclerosis Membranous nephropathy
Hemolytic uremic syndrome
Familial glomerular diseases Alport syndrome Thin glomerular
basement membrane disea se ("benign"

familial hematuria)
Fabry disease Nailpatella syndrome

as well as analysis of the urine. It is the aim of the history

and physical to establish whether the hematuria is isolated
or associated with findings that may indicate its origin and
the likelihood that it represents significant urinary tract
disease.
History

Increased frequency of urination and dysuria (pain or

burning with urination) often accompany UTI. Hesitancy
(difficulty in initiating the urine stream), a weak stream,
and dribbling are suggestive of bladder obstruction from
a calculus, tumor, or enlarged prostate. Loin pain is associated with the loin-pain hematuria syndrome, while colicky
flank pain that radiates into the groin is a frequent complaint in patients with a ureteral calculus and in some patients with papillary necrosis. Arthralgia/arthritis and
rashes suggest a systemic inflammatory disorder such as
anaphylactoid (Henoch-Schonlein) purpura (in which
crampy abdominal pain and melena may also be present),
systemic lupus erythematosus (SLE) or systemic vasculitis.
Hematuria following a prodrome of bloody diarrhea suggests the hemolytic-uremic syndrome. A history of pharyngitis or skin infection 1 to 2 weeks prior to the onset of gross
hematuria suggests poststreptococcal glomerulonephritis.
Hematuria may occur in patients with an inherited or druginduced coagulopathy, but is rarely the presenting complaint. A history of deafness may be present in the patient

Neoplasms Renal cell carcinoma

Wilms tumor

Benign cysts

Angiomyolipoma (tuberous sclerosis)

Multiple myeloma
Vascular Renal infarct Renal vein thrombosis Malignant
hypertension Arteriovenous malformation
Loin-pain hematuria syndrome
Hypercalciuria
Idiopathic
Metabolic Hyperparathyroidism Hyperoxaluria
Hyperuricosuria Cystinuria
Polycystic kidney disease (autosomal
dominant) Medullary cystic
disease/familial
juvenile nephronophthisis Medullary sponge
Familial kidney
Analgesic abuse
Sickle cell disease and trait
Renal tuberculosis
Diabetes mellitus
Papillary necrosis Obstructive uropathy
Alcoholism
Ankylosing spondylitis
Any cause
Drug-induced acute interstitial nephritis
(especially semisynthetic penicillins)
Renal contusion or laceration ExerciseHydronephrosis induced hematuria
Drugs
Trauma

with Alport syndrome (hereditary nephritis),

as may a family history of hematuria, renal failure, or
deafness. Because transient (24 to 48 hours) hematuria may
be seen in individuals who exert themselves strenuously
(exercise-induced hematuria), a history of exercise and
physical activity should be obtained. Travel to tropical
areas may expose an individual to Schistosoma
haematobium. Children from some areas of Southeast Asia
are prone to the development of calcium oxalate stones of
the urinary tract, perhaps due to dietary factors. A family
history of urinary tract stones should also be sought.
In the patient referred for evaluation of hematuria, characterization of the pattern of hematuria is useful. Patients
with persistent microhematuria and episodic gross hematuria frequently have glomerular disease, while those with
intermittent or continuous asymptomatic microhematuria,
or episodic gross hematuria with complete clearing of the
urine between episodes, are less likely to have significant
renal pathology. Episodic gross hematuria provoked by
mild upper respiratory infections is not unusual in patients
with glomerular disease, particularly IgA nephropathy.

38

Clifford E. Kashtan

TABLE 3

Hematuria: Extrarenal Causes

Calculi

Ureter, bladder, prostate

Neoplasms Transitional cell carcinoma (renal pelvis,

ureter, bladder) Adenocarcinoma, benign
hypertrophy
(prostate) Squamous cell carcinoma
(urethra)
Infections Acute cystitis, prostatitis, urethritis
Bacterial
Chlamydia trachornatis
Tuberculosis Schistosomiasis
Drugs
Cyclophosphamide (hemorrhagic cystitis)
Anticoagulants (heparin, coumadin)
Trauma

Genital or anal
bleeding

Contusion or laceration
Exercise-induced hematuria Foreign body of
urethra or bladder Decompression of severely
distended bladder
Vulvovaginitis
Vaginal foreign body
Anal fissure

Brown or cola-colored urine suggests hematuria arising

in the kidney, while pink or red urine suggests extrarenal
hematuria. Blood clots usually, but not invariably, indicate
bleeding from an extrarenal source.
Physical Examination
Elevated blood pressure and/or peripheral edema in the
patient with hematuria indicates glomerular pathology. A
rash may accompany anaphylactoid purpura nephritis, systemic vasculitis, mixed cryoglobulinemia, or SLE. Swollen,
inflamed joints in hematuric patients suggest anaphylactoid
purpura nephritis, SLE, vasculitis, or cryoglobulinemia.
RBCs arising in the vagina or vulva may mix with urine
in girls with vulvovaginitis or a vaginal foreign body, or
girls who have been the victims of sexual abuse, giving the
appearance of hematuria. Similarly, bleeding from an anal
fissure can present as apparent hematuria in children who
wear diapers. Prostatitis, benign prostatic hypertrophy
(BPH) and carcinoma of the prostate, each of which may
be associated with hematuria, can be detected by rectal
examination. Because BPH is such a frequent finding in
elderly men, it should not be assumed to be the cause of
hematuria until other causes are excluded. In adolescents
and adults with hematuria, sexually transmitted diseases
(STDs) need to be considered. Examination of the penis
or vulva for discharge may be indicated.
Urinalysis
Proteinuria accompanying hematuria strongly indicates
glomerular disease. Proteinuria may be detected by dipstick
(1+ or greater) or timed collection (greater than 4 mg
irr2 h~' in the child and 150 mg/24 h in the adult). In the
patient with hematuria and proteinuria, evaluation should
start with the glomerular causes of hematuria (Table 1).

Patients with gross hematuria may excrete up to 500 mg

protein/24 h, due to the blood in the urine. Like proteinuria,
RBC casts are indicative of glomerular disease, but are
more difficult to document. Pyuria (the presence of white
blood cells in the urine) is frequently present in patients
with UTI or STD. Crystals with typical morphologic characteristics are frequently visible in the urine of patients with
hematuria due to calculi composed of cystine or oxalate.
Some clinicians employ phase-contrast microscopy of the
urine to assess RBC morphology. The presence of dysmorphic RBCs suggests renal parenchymal disease, while their
absence indicates bleeding from an extrarenal site (see
Chapter 3).
Level II: Other Studies
A patient's history, physical examination, and urinalysis
will often provide the clinician with a strong suspicion regarding the origin of the hematuria. This suspicion will
direct the ensuing evaluation, leading to the correct diagnosis quickly while avoiding unnecessary procedures.
Glomerular Hematuria (Table I)
As noted above, the presence of proteinuria and/or RBC
casts in the patient with hematuria indicates glomerular
disease. Similarly, persistent microhematuria with episodic
gross hematuria suggests a glomerular process. While renal
biopsy is the definitive diagnostic procedure in this clinical
situation, it is not without risk and should be reserved for
select patients.
"*
Once hematuria has been localized to the glomerulus,
levels of serum complement, antistreptococcal antibodies,
antinuclear antibodies, creatinine, and blood urea nitrogen
(BUN) should be measured. Other blood studies should
be obtained in selected patients, e.g., antineutrophil cytoplasmic antibodies in patients suspected of systemic vasculitis; antiglomerular basement membrane antibodies in patients with a rapidly progressive acute nephritic syndrome,
with or without pulmonary hemorrhage; and cryoglobulins
when mixed cryoglobulinemia is suspected.
A negative family history for renal disease or hematuria
does not rule out a diagnosis of hereditary nephropathy,
such as Alport syndrome, thin glomerular basement membrane disease (also known, somewhat misleadingly, as benign familial hematuria), or autosomal dominant polycystic
kidney disease, because some patients with these diseases
represent new mutations or the offspring of asymptomatic
disease carriers. Complete evaluation of the family of a
patient with hematuria requires obtaining urinalyses on all
accessible family members. Patients suspected of having
Alport syndrome should undergo audiologic examination
and ophthalmologic evaluation, since certain ocular abnormalities (anterior lenticonus, yellowish perimacular flecks,
posterior polymorphous dystrophy) are diagnostic or very
suggestive of this disease.
Poststreptococcal glomerulonephritis remains an important cause of the acute nephritic syndrome (hematuria
and two or more of the following: proteinuria, edema, hy-

39

4. Hematuria
HEMATURIA1
History, Physical Exam, urinalysis, serum

proteinuria, high BP,

creatinine, BUN
RBC casts, dysmorphic
dysuria, colicky pain
RBCs

NONGLOMERULAR

HEMATURIA OF
UNCERTAIN ORIGIN

pyuria,
dysuria

GLOMERULAR

consider urgent biopsy

for rapidly declining GFR urine culture, STD culture
stone

abdominal
flat plate,
renal ultrasound 3

ydronephrosis, cyst,
mass, other

BIOPSY

normal
normal
low

additional
testing 4
cryoglobulins
ANA
ASO, anti-DNase B
HBsAG
anti-HC

40 yrs

stone
work-up further work-up

cystoscopy
additional
IVP
testing 5
CT, MRI
< 40 yrs
ANCA anti- GBM
serum/urine PEP, IEP audiogram, eye exam PT, PTT sickle
normal
screen
CONSIDER BIOPSY FOR:
high urine Ca
V further
progressive
azotemia
work-up

C3

YES

determination of prognosis proteinuria

unexplained findings

low

ANA
anti -strep
biopsy?

positive

family

---- ^- audiogram

urinalyses

eye exam

biopsy?
TREATMENT
consider cystoscopy, IVP

FIGURE I (1) This algorithm's design is based on two questions: (a) Is the hematuria glomerular or nonglomerular?
(b) Given the patient's age, what conditions must be promptly identified/excluded [e.g., Wilms tumor in child <3 yr,
urinary tract malignancy in adult >40 yr, rapidly progressive glomerulonephritis (RPGN)]? (2) In practice, C3, antinuclear
antibodies (ANA), and antistrep antibodies are often obtained at presentation. (3) This is the first imaging test to obtain
in the child <3 yr. (4) Nonstreptococcal infections may provoke a proliferative glomerulonephritis (GN) similar to
poststreptococcal disease. Biopsy may not be needed in the patient with typical features of postinfectious GN if C3 and
urinalysis normalize appropriately. Some patients with low C3 and positive antistrep antibodies have features atypical
of poststreptococcal GN (e.g., nephrotic syndrome, progressive azotemia) and need a biopsy to confirm the diagnosis.
Biopsy is indicated in patients with systemic lupus erythematosus and evidence of clinically significant GN, in order to
establish the type and severity of lupus nephritis. Children with features typical of anaphylactoid purpura usually do not
need a biopsy unless they have nephrotic-range proteinuria or reduced creatinine clearance. (5) These studies are indicated
in selected patients, e.g., anti- GBM antibodies in the patient with RPGN or pulmonary hemorrhage, ANCA and CRP
in patients with suspected vasculitis, serum/urine protein and immunoelectrophoresis for suspected multiple myeloma,
audiogram and eye exam for suspected Alport syndrome.

pertension, and reduced glomerular filtration rate). Kidney

biopsy is usually not indicated in the patient with acute
nephritic syndrome, hypocomplementemia, and evidence
of recent streptococcal infection. Kidney biopsy is usually

indicated in the patient with the acute nephritic syndrome

who does not have evidence of streptococcal infection or
hypocomplementemia, and in the great majority of patients
in whom hematuria is accompanied by proteinuria. In pa-

40

Clifford E. Kashtan

TABLE 4

Hematuria: Initial Evaluation

History

Physical

Urine

Frequency, hesitancy
Force of stream

Bleeding tendency
Deafness

Dysuria Loin
pain Renal
colic
Arthralgia
Rash
Blood pressure
Edema Rash

Family history
Exercise history
Travel history
Medications

Purpura
Arthritis
Proteinuria
Casts
Pyuria

Prostatic enlargement or
tenderness
Bacteriuria
Crystals

Flank mass Genital or

anal bleeding Deafness

tients with isolated hematuria, the clinical utility of kidney

biopsy must be considered on a case-by-case basis. Diagnostic accuracy requires the performance of immunofluorescence and electron microscopy as well as light microscopy on all renal biopsy specimens from patients with
glomerular hematuria.
Up to 20% of children with steroid-responsive, minimal
change nephrotic syndrome present with microscopic hematuria. Therefore, the presence of hematuria in the child
of appropriate age (6 months to 6 years) with nephrotic
syndrome should not change the initial approach (a trial
of oral prednisone).
Nonglomerular Hematuria (Tables 2 and 3)

The history, physical examination, and urinalysis may

suggest the likely source of bleeding in the patient with
isolated hematuria and may direct the subsequent evaluation. For example, a history of renal colic suggests a urinary
tract calculus. Since more than 90% of calculi contain calcium, most will be revealed by a plain film of the abdomen.
Intravenous pyelography (IVP) may be used to precisely
localize radiopaque stones in the urinary tract, or to dis cover radiolucent calculi. Suspicion of a urinary tract calculus should increase if the patient has a history of diffuse
bowel disease or bowel resection (secondary hyperoxaluria) or a family history of urolithiasis (see Chapter 56).
The presence of pyuria, with or without dysuria, suggests
UTI or STD. In the patient with suspected UTI, a sterilely
collected urine specimen should be sent for routine bacterial culture and expectant antibiotic therapy should be initiated. Specific cultures for agents causing STD (e.g., chlamydia, Neisseria gonorrhea) may be indicated.
Individuals of African ancestry with hematuria should
be evaluated for sickle hemoglobinopathy, which accounts
for about one-third of cases of hemal uria in this population.
Sickle cell trait frequently first presents as hematuria (see
Chapter 44).
Loin pain -hematuria syndrome is characterized by recurrent episodes of loin/lumbar pain, which is usually unilat-

eral and may be quite severe, associated with gross or

microscopic hematuria. Most patients are women, 20 to 40
years of age. Renal biopsy findings are essentially normal,
while renal angiographic findings are variable.
The medication history may be revealing in patients with
hematuria. Administration of oral or intravenous cyclophosphamide may be associated with hemorrhagic cystitis,
ureteritis, and pyelitis, either acutely or as a late event.
This complication of cyclophosphamide therapy is seen
most commonly now in patients undergoing bone marrow
transplantation, but may also complicate treatment of systemic vasculitis and SLE. Diagnosis of hemorrhagic cystitis
is made by cystoscopy. Certain drugs, especially the semisynthetic penicillins, may produce acute interstitial nephritis; pyuria and acute renal failure are the two main features,
but hematuria may also be observed (Chapter 40). Diagnosis of this condition may require renal biopsy. The
long-term, excessive use of phenacetin- or acetaminophencontaining analgesics may result in analgesic-abuse nephropathy, one manifestation of which may be hematuria
due to papillary necrosis. Patients with diabetes mellitus
or urinary tract obstruction are also susceptible to papillary
necrosis (see Chapter 53). Chronic administration of furosemide to premature infants with bronchopulmonary dysplasia may be associated with urolithiasis.
When the history, physical examination, and urinalysis
provide no clues to the origin of hematuria, a renal ultrasound serves as a useful screening procedure. Kidneys that
are hydronephrotic for any reason may bleed after minimal
trauma. Autosomal dominant polycystic kidney disease can
usually be diagnosed by ultrasound by the third decade of
life. Renal ultrasound will also identify simple renal cysts
which, in the absence of other causes, may be the source of
hematuria. Some renal tumors will be visible by ultrasound.
Renal vein thrombosis often produces substantial renal
enlargement; this condition should be suspected in the neonate with hematuria, or the nephrotic patient who presents
with the new onset of gross hematuria and unilateral flank
pain. Renal ultrasound will also frequently identify the
patient with nephrocalcinosis, although care must be taken
in the interpretation of hyperechoic renal pyramids in
the neonate.
Further evaluation when history, physical examination,
urinalysis, and renal ultrasound fail to provide a diagnosis
depends heavily on the age of the patient. Because of the
higher incidence of urinary tract malignancy in older individuals, patients over 40 years of age should undergo intravenous pyelography (see Chapter 6) and cystoscopy, and
urinary cytology should be performed. These studies have
a lower yield in younger patients, in whom urinary tract
malignancies are infrequent. However, prompt imaging by
ultrasound is mandatory in children younger than 3 years
with hematuria, in order to detect Wilms tumor. Renal
ultrasound will also detect renal vein thrombosis or hydronephrosis in the neonate with hematuria.
In children and young adults, asymptomatic gross or
microscopic hematuria is frequently due to idiopathic hypercalciuria. This condition may be associated with hematuria in the absence of any radiographically detectable cal-

4. Hematuria

culi, although the risk of developing a stone is high. There

is frequently a family history of urolithiasis. Idiopathic
hypercalciuria may reflect excessive intestinal uptake of
calcium (absorptive hypercalciuria) or insufficient tubular reabsorption of calcium (renal hypercalciuria). Hypercalciuria is present when the urinary calcium excretion is
greater than 4 mg kg"1 day" 1 (0.1 mmol kg"1 day" 1).
Renal biopsy should be considered in the patient with
persistent microhematuria (>6 to 12 months) in whom
other evaluations are unrevealing. A variety of glomerular
disorders, including many of those listed in Table 2, may
present as isolated hematuria. The degree of parental or
patient anxiety may influence the timing of biopsy in
these cases.
On occasion, hematuria occurs secondary to a vascular
anomaly, such as arteriovenous fistula, hemangioma, or
varices of the kidney or ureter. Renal arteriography may
be revealing in such cases. The yield of renal arteriography
is probably highest when cystoscopy documents unilateral
upper urinary tract bleeding in a patient with a normal
renal ultrasound and CT scan.
Bibliography
Andres A, Praga M, Bello I, Diaz-Rolon JA, Gutierrez-Millet V,
Morales JM, Rodicio JL: Hematuria due to hypercalciuria and
hyperuricosuria in adult patients. Kidney Int 36:96-99, 1989.
Blumenthal SS, Fritsche C, Lemann J: Establishing the diagnosis
of benign familial hematuria: The importance of examining the
urine sediment of family members. JAMA 259:2263-2266,1988.
Burden RP, Dathan JR, Etherington MD, Guyer PB, Maclver
AG: The loin -pain/hematuria syndrome. Lancet 1:897-900,
1979.
Copley JB: Isolated asymptomatic hematuria in the adult. Am J
MedSci 291:101-111, 1986.
Copley JB, Hasbasrgen JA: "Idiopathic" hematuria: A prospective
evaluation. Arch Intern Med 147:434-437, 1987.

41
Dodge WF, West EF, Smith EH, Bunce H: Proteinuria and hematuria in schoolchildren: epidemiology and early natural history.
/ Pediatr 88:327-347, 1976.
Fairley KF, Birch DF. Hematuria: a simple method for identifying
glomerular bleeding. Kidney Int 21:105-108, 1982.
Froom P, Ribak J, Benbassat J. Significance of microhaematuria
in young adults. Br Med J 288:20-22, 1984.
Garcia CD, Miller LA, Stapleton FB. Natural history of hematuria
associated with hypercalciuria in children. Am J Dis Child
145:1204-1207, 1991.
Hayashi M, Kume T, Nihira H: Abnormalities of renal venous
system and unexplained renal hematuria. / Urol 124:12-16,1979.
Ingelfinger JR, Davis AE, Grupe WE: Frequency and etiology of
gross hematuria in a general pediatric setting. Pediatrics 59:557561, 1977.
Jones DJ, Langstaff RJ, Holt SD, Morgans BT: The value of
cystourethroscopy in the investigation of microscopic haematuria in adult males under 40 years: A prospective study of 100
patients. Br J Urol 62:541-545, 1988.
Mohr DN, Offord KP, Owen RA, Melton LJ: Asymptomatic microhematuria and urologic disease: A population-based study.
JAMA 256:224-229, 1986.
Schuster GA, Lewis GA: Clinical significance of hematuria in
patients on anti-coagulant therapy. J Urol 137:923-925, 1987.
Siegel AJ, Hennekens CH, Solomon HS, Boeckel BV: Exerciserelated hematuria: Findings in a group of marathon runners.
JAMA 241:391-392, 1979.
Tiebosch ATMG, Frederick PM, van Breda Vriesman PJC, et al:
Thin -basement-membrane nephropathy in adults with persistent
hematuria. N Engl J Med 320:14-18, 1989.
Trachtman H, Weiss RA, Bennett B, Griefer I: Isolated hematuria
in children: Indications for a renal biopsy. Kidney Int 25:9499, 1984.
Vehaskari VM, Rapola J, Koskimies O, Savilahti E, Vilska J,
Hallman N: Microscopic hematuria in schoolchildren: Epidemiology and clinicopathologic evaluation. J Pediatr 95:676-684,
1979.

PROTEINURIA
SHARON ANDERSON

Proteinuria characterizes almost every form of glomerular disease and contributes to all of the complications of
the nephrotic syndrome. Indeed, proteinuria or one of its
complications is often the first indication that renal disease
is present.

in part by electrostatic repulsion. Though molecular shape

also appears to be important, less is known about this parameter.
Finally, the third component is the proximal tubule.
Some proteins are in fact filtered, but are then catabolized
and reabsorbed in the proximal tubule. Thus, the amount
of protein in the urine depends on how much gets there,
how well it passes through the glomerular barrier, and
whether or not the proximal tubule degrades it if the protein
does get into the glomerular filtrate.

RENAL PROTEIN HANDLING

To understand the pathogenesis of proteinuria, it is first

important to review the handling of protein by the normal
kidney. There are three major components of proteinuria.
The first is the amount of protein being presented to the
glomerular capillary (the filtered load), which is dependent
both on the protein concentration in the plasma and on
the glomerular filtration rate (GFR). The second major
determinant relates to permeability (ability to traverse the
glomerular capillary barrier), which is dependent both on
the integrity of the glomerular capillary wall and on specific
attributes of the particular protein molecule. The glomerular capillary barrier normally presents a formidable barrier
to protein. The glomerular capillary wall consists of three
layers (Fig. 1): the fenestrated endothelium, the glomerular
basement membrane (GBM), and the epithelial cells which
are attached to the basement membrane by foot processes.
The primary barrier to filtration is the GBM.
Experiments using test macromolecules, such as dextran,
have shown that factors determining the ability of a given
protein molecule to pass through the GBM are molecular
size, charge, and shape. The glomerular capillary wall is
highly permeable to small solutes and water, but limits
passage of larger molecules. Inulin (MW 5200, 16 A) is
completely filtered, while albumin (MW 69,000, 36 A) is
filtered only to a small degree. Below MW 60,000, there is
a progressive increase in filtration. The inverse relationship
between molecular size and degree of filtration suggests
that "pores" (functional, not structural) may exist within
the GBM through which only molecules of a certain size
may pass. Molecular charge is another important determinant. Anionic dextrans are filtered to a very limited degree;
by comparison, filtration of neutral and cationic dextrans
is much higher. The glomerular capillary wall contains sialoproteins and proteoglycans (such as heparan sulfate)
that are negatively charged. Most circulating macromolecules are anionic within the physiologic pH range. Thus,
filtration of albumin and other macromolecules is limited

Primer on Kidney Diseases, Second Edition

CLINICAL PROTEINURIA: DEFINITION AND

MEASUREMENT

In normal kidneys, low-molecular-weight proteins and

small amounts of albumin (about 500 to 1500 mg/day) are
filtered. These proteins enter the proximal tubule, where
they are almost completely reabsorbed and catabolized
by proximal tubular cells. The net result is daily protein
excretion of about 40 to 80 mg. Of this amount, about 30
to 50 mg consists of Tamm-Horsfall protein, a mucoprotein
secreted by tubular cells in the thick ascending loop of
Henle. Abnormal proteinuria is defined as excretion of
>150 mg/day.
Presence or absence of abnormal proteinuria is first detected by dipstick, which is read as 0 to 4+, corresponding
to concentrations of 0 to >500 mg/dL. The most accurate
quantitation is performed on a 24-hour urine collection.
For this purpose, urinary protein concentration (mg/dL) is
determined [for example, by precipitation with sulfosalicylic acid (SSA)], and then multiplied by total volume (dL/
day) to obtain 24-hour protein excretion (in mg/day). In
most situations, the results obtained with dipstick and with
quantitative methods correlate fairly well. However, there
is one important clinical situation in which the two methods
differ. The dipstick is most sensitive to albumin, whereas
quantitative methods detect all proteins. The dipstick may
be negative when low-molecular-weight proteins, rather
than albumin, are present in the urine. This occurs in certain
tubular diseases, where proximal tubule catabolism of filtered proteins is impaired; or in diseases where there is
excessive production and filtration of low-molecularweight proteins, exceeding the proximal tubule's capacity
to catabolize filtered proteins. The most important clinical
example of this situation is multiple myeloma; the urine
may contain large amounts of protein (detectable by quan-

42

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

5. Proteinuria

43

Bowman's / ^B / Space
'PodocyteV
Slit diaphragm
[Lamina rara externa
Lamina densa |
Lamina rarainterna
Epithelial cell

GBM

Endothelial cellsSl<r='
Capillary lumen

FIGURE I The glomerular barrier consists of the capillary endothelium, glomerular basement membrane, and glomerular epithelial cell.

titative methods) and yet be negative or only trace positive

by dipstick. Such a condition should be suspected when
the amount of protein determined quantitatively is disproportionately elevated, compared to the dipstick element.
The diagnosis may be confirmed by urinary and serum
electrophoreses, as described below.
An convenient alternative to 24-hour urine collection is
to compare the concentrations (in mg/dL) of protein and
creatinine (UprM /UCl ratio) on a random spot urine specimen.
Since urinary creatinine excretion is relatively constant
(20 to 25 mg kg" 1 day" 1 for men, 15 to 20 for women)
an increase in the ratio indicates an increase in urinary
protein excretion. The urinary protein/creatinine ratio is a
fairly accurate estimate of total daily protein excretion
and can be useful when used serially in a given patient. The
normal ratio is <0.2 (less than 200 mg/day); values of 1.0
and 3.5 represent total protein excretion of 1.0 and 3.5
g/day, respectively.
Recently, it has been recognized that diabetes, essential
hypertension, and some other diseases may manifest "microalbuminuria," which is defined as daily urinary albumin
excretion of 30 to 300 mg/day. This level of albuminuria
is above the normal range, yet below the limits of detectability by dipstick. Microalbuminuria is thought to be the
earliest sign of nephropathy in Type I diabetes mellitus,
and a powerful predictor of cardiovascular risk in Types I
and II diabetes and other conditions. Microalbuminuria
can be detected using sensitive laboratory techniques, when
specifically requested. It is recommended that patients with
Type I diabetes be screened annually, starting at 5 to 10
years after diagnosis of diabetes. Patients with Type II
diabetes should be screened at the time of diagnosis, and
then annually thereafter. While a 24-hour urine collection
remains the gold standard, many clinicians have opted for
the more easily obtainable spot urine microalbumin/creatinine ratio, with a value >30 /ug albumin/mg creatinine
representing clinical rnicroalbuminuria. (See Chapter 29.)
MECHANISMS OF PROTEINURIA

Proteinuria can be usefully classified according to four

major mechanisms: (1) functional proteinuria, (2) overpro-

duction (prerenal) proteinuria, (3) glomerular proteinuria,

and (4) tubular proteinuria.
Functional proteinuria is the term used for a transient
increase in protein excretion which occurs in various physiologic and experimental situations, in the absence of renal
disease. For example, fever, emotional stress, acute medical
illness, and infusions of norepinephrine or angiotensin II
can all cause significant but transient increases in urinary
protein excretion. Urinary protein excretion is increased
two- to threefold during and immediately after heavy exercise. The mechanism for these forms of proteinuria most
likely relates to glomerular hemodynamic changes, or possibly to changes in glomerular pore size, rather than to
alterations in the GBM.
Some normal subjects also exhibit "orthostatic proteinuria," defined as protein excretion which is abnormally
elevated in the upright position, but normal with recumbency. This benign condition is found in 2 to 5% of adoles cents, but only rarely over the age of 30 years. Orthostatic
proteinuria is usually less than 1 g/day and may be detected
using split (16-hour day, 8-hour night) urine collections.
Patients with postural or orthostatic proteinuria will have
increased excretion while upright, and a normal level (<150
mg/day when extrapolated to 24 hours) in the recumbent
specimen. If protein excretion is increased in both specimens, the patient is considered to have persistent, rather
than orthostatic, proteinuria.
Overproduction (prerenal) proteinuria occurs when
there is an increased plasma concentration of one or more
filterable proteins, thereby increasing the filtered protein
load, thus increasing excretion. There may be enhanced
excretion of heavy chains, light chains, or other immunoglobulin fragments. The differential diagnosis includes
Bence Jones proteinuria, myoglobinuria, hemoglobinuria, lysozymuria (myelomonocytic leukemias), and /32microglobulinuria (neoplasia).
Glomerular loss of proteins may result from several different mechanisms. The most important are: (1) loss of the
negative charges in the GBM; (2) an increase in effective
pore size or number, due to direct damage to the GBM or
change in GBM structure due to loss of anionic proteins; or
(3) disease-related changes in glomerular hemodynamics,
particularly glomerular capillary pressure (Poc)- Studies in
various experimental models have confirmed that diseases
such as glomerulonephritis may be associated with abnormalities in the synthesis, composition, and negative charge
density of the GBM, including changes in heparan sulfate
content or incorporation into the GBM.
Hemodynamic factors also influence the ultrafiltration
of macromolecules. Infusion of vasoactive hormones such
as angiotensin II and norepinephrine, which increase PGc,
increase proteinuria in experimental animals. In addition,
glomerular diseases which are characterized by glomerular
capillary hypertension (increased PGC ) are regularly associated with proteinuria, while pharmacologic or dietary maneuvers like ACE inhibition or protein restriction that reduce PGC reduce proteinuria.
The differential diagnosis of glomerular proteinuria includes all glomerular diseases (e.g., focal glomerulonephri-

44

Sharon Anderson

tis, poststreptococcal glomerulonephritis, minimal change

disease, focal sclerosis, membranous glomerulonephritis,
diabetic nephropathy).
Tubular proteinuria occurs when there is impaired tubular reabsorption of normally filtered proteins. The differential diagnosis includes hereditary diseases (e.g., Wilson's
disease), toxic injury (cadmium, lead, gentamicin), metabolic causes (hypokalemia), Fanconi syndrome, chronic interstitial nephritis, and miscellaneous other causes (diuretic
phase of acute tubular necrosis and amylase excretion in
acute pancreatitis).
LABORATORY EVALUATION OF PROTEINURIA

Proteinuria is usually first detected by dipstick on a routine screening urinalysis. In occasional patients, proteinuria
will be intermittent (present on some urinalyses, but not
on others). This pattern may represent any of a number
of minor glomerular or tubular lesions, and prognosis is
often favorable if the patient does not progress to persistent
proteinuria. Once proteinuria is persistent, further evaluation is indicated.
The next step is to quantify the amount. In general,
>3 g/day is virtually always glomerular. Proteinuria <3 g/
day is nondiagnostic and could represent a prerenal, glomerular, or tubular origin. If the origin of the proteinuria
is unclear, the source of the urinary protein may next be
evaluated by urine protein electrophoresis (Fig. 2). If the
protein is >70% albumin, the source is glomerular. A tubular source will cause excretion of globulins more than albumin, with multiple peaks (representing a number of different proteins). A prerenal source usually reveals a single
globulin peak, representing the protein whose plasma concentration is unusually high.
Once the origin of the proteinuria (prerenal, glomerular,
or tubular) is determined, subsequent evaluation can begin.

Pre
3 a 2 a. Alb Alb

Pre y
(3 a2<*i Alb Alb

Glomerular
Pre-renal
FIGURE 1 Schematic examples of
urinary electrophoresis. The normal pattern (upper left) consists of
a number of peaks, with albumin being the largest. In tubular
proteinuria (upper right), a broad band representing multiple
proteins is seen in the globulin range. In glomerular proteinuria

(lower left), a large albumin peak is seen. When proteinuria is

prerenal in origin, a single large monoclonal globulin peak
(reflecting the circulating paraprotein) will be found.

For prerenal sources, the next steps may include a serum

protein electrophoresis and/or serum and urine immunoelectrophoreses to further characterize the individual protein being excreted. The diagnosis can then usually be
made, for example, by bone marrow or lymph node biopsy;
further renal evaluation may not be necessary. For tubular
sources, the clinical scenario usually indicates the etiology.
More extensive evaluation, often including a renal biopsy,
is needed when the urinary protein is glomerular in origin.
Almost every form of glomerular disease (including both
nephritic and nephrotic syndromes) is characterized by
protein uria. Though a biopsy is frequently needed, occasional patients with relatively low amounts of proteinuria
and stable renal function may be followed, and biopsy
deferred until evidence of progression or complications
arises. Again, the clinical setting, and other laboratory tests,
may be useful. Determining the presence of a nephritic
syndrome, a nephrotic syndrome, or a systemic illness helps
to begin the differential diagnosis. Longstanding diabetes,
together with presence of diabetic retinopathy and absence
of signs of acute renal injury (e.g., red cell casts, active
sediment), usually indicates diabetic nephropathy. Pres ence of other systemic diseases known to involve the glomerulus (e.g., systemic lupus erythematosus) can also indicate the diagnosis . Longstanding hypertension, with
hypertensive nephrosclerosis, may be associated with proteinuria which is usually below the nephrotic range. Pres ence of pulmonary involvement suggests one of the pulmonary-renal syndromes (e.g., Goodpasture syndrome,
small vessel vasculitis), while measurement of serum complement levels, antineutrophil cytoplasmic antibodies
(ANCA), and cryoglobulin levels can all be helpful in narrowing the differential diagnosis. In many cases, however,
if these tests are negative or nonspecific, a renal biopsy is
required for definitive diagnosis.
CLINICAL CONSEQUENCES OF PROTEINURIA

Loss of albumin and other plasma proteins into the urine

is the hallmark of the nephrotic syndrome, and contributes
to virtually all of the systemic complications of this disorder.
"Subnephrotic proteinuria" refers to urinary protein excretion of <3.5 g/day; the source may be prerenal, glomerular,
or tubular. Severe clinical signs and symptoms are rare
with this lower grade of proteinuria. However, in patients
with the nephrotic syndrome (>3.5 g/day), the complications of proteinuria can be severe and even life-threatening.
Proteins are essential for all body processes, and exces sive loss of these various proteins damages numerous systems, resulting in diverse complications (Table 1; Fig. 3).
Some hormones, trace metals, and vitamins are lost in the
urine. Alterations in coagulation factors, such as loss of
antithrombin III, lead to problems with spontaneous
thromboemboli. Loss of immunoglobulins and parts of the
complement cascade can lead to problems with immunity
and increased rates of infection. Massive urinary albumin
losses create many of the most critical complications of the
nephrotic syndrome. Though not yet proven, it has been
suggested that the excessive load of albumin going through

45

5. Proteinuria

TABLE I

THERAPY OF PROTEINURIA

Clinical Complications of Proteinuria

Successful treatment of the underlying glomerular disorder is, of course, the primary therapeutic goal, as this will
both preserve renal function and reduce or eliminate proteinuria. However, some forms of glomerular disease have
no specific therapy available, and other forms are not always successfully treated. In such cases, the presence of
severe and symptomatic proteinuria may necessitate therapy aimed at reducing proteinuria specifically. Several therapeutic interventions have been successfully used to lower
urinary protein excretion.
Dietary protein restriction may lower proteinuria in
some, though not all, patients. This intervention works, in
part, by reducing both GFR and PGC, as well as by effects
on glomerular permselectivity. For this purpose, the lowest allowable level of protein restriction is 0.6 g kg" 1
day" 1, plus protein added to match urinary losses. Thus,
a 70-kg patient with 10 g of proteinuria per day would
receive a daily allotment of 52 g of protein, representing
0.6 g kg" 1 day" 1, plus 10 additional grams to compensate
for urinary losses. This therapy, which may also help to
slow the progression of renal disease, should be used only in
close collaboration with a renal dietician to avoid problems
with negative nitrogen balance.
Angiotensin-converting enzyme (ACE) inhibitors have
also proven useful in some cases. While effective reduction
in blood pressure with any agent may offer some benefit,
ACE inhibitors appear to be the most consistently effective.
They reduce proteinuria, in part, by lowering efferent arteriolar resistance, thereby reducing PQO Other mechanisms

Hypoalbuminemia
Edema
Increased hepatic lipoprotein synthesis, hyperlipoproteinemia
Increased platelet aggregability
Increased tubular protein reabsorption Possible
tubular dysfunction, tubular damage
Loss of proteins carrying vitamins, hormones, and minerals
Trace mineral deficiencies Hypocalcemia, Vitamin D
deficiency
Loss of immunoglobulins
Reduced cellular immunity, increased susceptibility to
infection
Alterations in coagulation factors Spontaneous
thromboembolism, renal vein thrombosis
Negative nitrogen balance, malnutrition
Alterations in drug metabolism (due to reduced protein
binding)
Diuretic resistance

the proximal tubule may lead to eventual tubular damage

and dysfunction. Hypoalbuminemia contributes to massive
edema formation, since albumin is the primary oncotic
force keeping fluid inside the vascular space. Hypoalbuminemia also leads to hyperlipoproteinemia, with very high
(>400 g/dL) serum cholesterol levels, since hypoalbuminemia stimulates hepatic synthesis of lipoproteins.

Increased glomerular permeability

Urinary losses of
proteins carrying
hormones, metals
and vitamins
Alteratio ^
coagulat ^ ^
factors

iltrationof
jrotems

i
Increased \ ^
plasma [
is in
nn

_____ ^

^^es'o^
immunoglobulins

\
Reduced cellular
Increased ^
immunity ------------ infections

Thromboembolism
nuria >

Malnutrition

Album Increased tubular ^s*

reabsorption of ^r
/ filtered protein
Tubular
Tubular catabolism >
of albumin

Lipiduria
t

Hyperlipoproteinemia
Increased hepatic
synthesis of
lipoproteins

[ Hypoalbuminemia |
Edema

damage

FIGURE 3 Major complications of the nephrotic syndrome. See text for discussion. Reproduced from Bernard DB: Kidney Int 33:1 184-1202, 1988.

46
appear to be involved as well, including a direct effect
on glomerular permselectivity. For maximum benefit, they
should be given in conjunction with diuretic therapy and/
or a low sodium diet, since continuation of liberal sodium
intake may abolish the antiproteinuric efficacy of these
drugs. However, ACE inhibition should be instituted cautiously. Patients with moderate to severe renal insufficiency
may develop a rise in the serum creatinine, acute renal
failure, or hyperkalemia with ACE inhibitors, and therefore this therapy should be avoided in patients at risk for
these complications. ACE inhibitors are used most safely
when GFR is not severely impaired. Other antihypertensive drugs have been less consistent in effect, but could
certainly be tried if ACE inhibition is not possible. Limited,
short-term studies suggest that angiotensin II receptor antagonists may also be effective in reducing proteinuria.
Though less likely to cause cough, these drugs probably
will have a side effect profile similar to ACE inhibitors.
Nonsteroidal anti-inflammatory drugs in high doses are
also effective in reducing proteinuria, probably because
they also tend to reduce GFR. While they can be useful,
they can be used only when GFR is not severely impaired,
to avoid the risks of acute renal failure and hyperkalemia.
Because of the risk of lowering GFR, and the high incidence
of gastrointestinal side effects at high doses, these agents
are not as widely used to lower proteinuria, but may be
considered in severe cases.
Bibliography
Anderson S, Kennefick TM, Brenner BM: Systemic and renal
manifestations of glomerular disease. In: Brenner BM, Rector
FC, Jr (eds) The Kidney, 5th ed. Philadelphia, Saunders, pp.
1981-2010, 1996.
Bennett PH, Haffner S, Kasiske BL, et al: Screening and management of microalbuminuria in patients with diabetes mellitus:
Recommendations to the Scientific Advisory Board of the Na-

Sharon Anderson
tional Kidney Foundation from an ad hoc committee of the
Council on Diabetes Mellitus of the National Kidney Founda tion. Am J Kidney Dis 25:107-112, 1995.
Bernard DB. Extrarenal complications of the nephrotic syndrome.
Kidney Int 33:1184-1202, 1988.
Bernard DB, Salant DJ: Clinical approach to the patient with
proteinuria and the nephrotic syndrome. In: Jacobson H, Striker
G, Klahr S (eds) Principles and Practice of Nephrology, 2nd ed.
St. Louis, Mosby-Year Book, pp. 110 -121, 1995.
Brenner BM, Hosteller TH, Humes DH: Molecular basis of proteinuria of glomerular origin. N Engl J Med 298:826-833, 1978.
Carlson JA, Harrington JT: Laboratory evaluation of renal function. In: Schrier RW, Gollschalk CW (eds) Diseases of the Kidney, 5th ed. Little, Brown, Boston, pp. 361 -405, 1993.
Deen WM, Myers BD, Brenner BM: The glomerular barrier to
macromolecules: Theoretical and experimental considerations.
In: Brenner BM, Stein JH (eds) Nephrotic Syndrome. Churchill
Livingstone, New York, pp. 1-29, 1982.
Ginsberg JSM, Chang BS, Matarese RA, Garella S: Use of single
voided urine samples to eslimale quantitative proteinuria. N
Engl J Med 309:1543-1546, 1983.
Kanwar YS, Liu ZZ, Kashihara N, Wallner El: Current status of
the structural and functional basis of glomerular nitration and
proteinuria. Semin Nephrol 11:390-413, 1991.
Kaplan NM. Microalbuminuria: A risk factor for vascular and renal
complications. Am J Med 92(Suppl. 4B):4B-8 S-4 B-12S, 1992.
Kaysen GA. Hyperlipidemia in the nephrotic syndrome. Am J
Kidney Dis 12:548-551, 1988.
Kaysen GA, Myers BD, Couser WG, Rabkin R, Felts IM: Mechanisms and consequences of proteinuria. Lab Invest 54:479498, 1986.
Llach F: Hypercoagulability, renal vein Ihrombosis, and other
Ihrombolic complications of nephrotic syndrome. Kidney Int
28:429-439, 1985.
Morrison G, Audel PR, Singer I: Clinically important drug interactions for the nephrologist. In: Bennett WM, McCarron DA,
Brenner BM, Stein JH (eds) Pharmacotherapy of renal disease
and hypertension. Churchill Livingstone, New York, pp. 4997, 1987.

RENAL IMAGING TECHNIQUES

HOWARD J. MINDELL AND JONATHAN FAIRBANK

such as US, may provide more detail regarding the renal

parenchyma, the IVU remains an inexpensive, widely available, "low tech" test that gives excellent detail of the entire
urinary tract, especially of the collecting system. The IVU
remains the first choice in most patients with hematuria,
pyuria, or flank pain. Other indications for IVU include
urothelial malignancy, tuberculosis, papillary necrosis, and
a variety of congenital anomalies. When the collecting system must be examined but IVU is inadequate because renal
failure limits excretion of radiographic contrast and urinary
tract visualization, retrograde urography or percutaneous
antegrade urography may be needed. The former requires
cystoscopy and ureteral catheter placement; the latter is
performed after imaging guided puncture of the calyceal
system. The IVU is used to evaluate renal donors and
in the pre- and postoperative assessment of patients with
nephrolithiasis and other lesions that can be treated with
endoscopic manipulation of the ureter.
The contrast medium employed for intravenous urography and other studies is subject to two classes of reactions:
idiosyncratic and toxic, such as acute renal failure. Approximately 0.1 to 0.2% of patients develop an idiosyncratic
reaction characterized by urticaria, wheezing, dyspnea, or
chest pain which may be followed by hypotension. The
case fatality rate is estimated to be 1 in 85,000. Pretreatment
with glucocorticoids or the use of low osmolality contrast
media may reduce the incidence of this reaction. Substitution of the latter for conventional media is restricted because these agents are 10 times as costly. Low osmolarity
agents are therefore typically restricted to patients with a
history of prior contrast reaction, a strong generalized allergic history, or asthma. The second, toxic reactions, include
acute renal failure. This occurs independent of the idiosyncratic reaction just described.
The risk of contrast nephropathy is highest in patients
with diabetic nephropathy, intermediate in patients with
renal insufficiency due to other causes, and low or normal
for diabetics without renal disease. There are no absolute
contraindications to the use of IV contrast, but it should
be used with circumspection in such settings as advanced
age, debility, congestive heart failure, and multiple my eloma. Patients with volume depletion are at greatest risk.
Intravascular volume should be restored by fluid adminis tration before contrast is given. A recent study suggested
that overnight expansion with 0.5% saline affords protec-

GENERAL CONSIDERATIONS

Investigation of patients with renal disorders often requires obtaining images of the kidneys and urinary tract.
Although more helpful in evaluating renal masses or disorders of the urinary outflow tract than intrinsic renal parenchymal disease, imaging studies can either establish the
general pathway for further investigation or lead to a specific diagnosis. In general, the choice of studies proceeds
from less to more invasive or expensive studies, unless
the expensive or invasive study is much more likely to
be definitive. The current medical-economic environment
favors the use of the simplest, safest, and cheapest approach
that can answer the question at hand.
ULTRASONOGRAPHY

Ultrasonography (US) is entirely safe, independent of

renal function, capable of multiplanar display, and does
not require contrast agents or prior patient preparation.
However, it lacks specificity in many instances (for example, in differentiating among renal parenchymal diseases),
and technical problems may arise in large patients in whom
tissue degrades the interrogating sound waves or when
intestinal gas reflects sound, and prevents delineation of
the underlying structure. Although US can exquisitely demonstrate vascular occlusive disease, in many patients it is
inapplicable because of the technical reasons cited above.
Doppler US, even with the use of color to show flow direction, has not enjoyed wide or universal success in screening
for renovascular hypertension. In addition to the limitations noted, multiple renal arteries may be impossible to
sort out by US alone. Nevertheless, US is a mainstay of
renal imaging and is frequently the first study chosen. US
can easily differentiate hydronephrosis (Fig. 1) from intrinsic renal parenchymal disease and renal cysts from solid
tumors. It is also a useful method for delineating perinephric collections, pyelonephritic scars, and nephrocalcinosis. US can assist in evaluating renal transplants and
can guide a variety of interventional approaches to the
kidney, including biopsy, aspiration, and percutaneous
nephrostomy.
EXCRETORY (INTRAVENOUS) OROGRAPHY

Intravenous urography (IVU) can noninvasively show

the entire urinary tract. While other imaging modalities,

Primer on Kidney D iseases, Second Edition

47

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

Howard J. Mindell and Jonathan Fairbank

48

pyelonephritis and its complications and in renal cystic

disorders. With arterial occlusive disease, CT can show
perfusion defects, where contrast fails to opacify segments
of the renal parenchyma, and recent work has found that
3D reconstruction with helical CT may show the renal
arterial vessels directly. With venous occlusive disease, CT
can show flow abnormalities and stasis in the affected kidney and can directly demonstrate thrombus in the renal
vein. CT can show urinary calculi that are not visible by
radiography (urates) and unenhanced helical CT is a promising new technique in the diagnosis of renal colic.
MAGNETIC RESONANCE IMAGING

FIGURE I Ultrasound showing a sagittal section through the right

kidney. The single black arrow indicates a dilated calyx in the
lower pole. Note how the calyx connects to the dilated renal pelvis
(black, fluid filled). Multiple open arrows indicate the margin of
the renal cortex.

tion in patients who are not volume depleted (see Chapter 36).
It also bears emphasis that these comments about side
effects of administration of radiographic contrast apply no
matter what the procedure for which contrast is employed.
They are not specific to IVU and concern about contrast
administration must be taken into account when a CT with
contrast or angiogram is ordered. Finally, the visualization
of the urinary tract is diminished when the serum creatinine
exceeds 2 to 3 mg/dL; the IVU is unlikely to successfully
visualize the collecting system when the serum creatinine
exceeds 4 mg/dL. Therefore, the risk of increased contrast
administration should be weighed against its diminished
potential benefit in these patients.
PLAIN ABDOMINAL RADIOGRAPH

The plain abdominal radiograph (KUB), while a standard prelude (scout film) for the IVU, may be requested
alone or in conjunction with US. Renal or ureteric calculi
may be visible on the KUB, although the former may require oblique views to confirm an intrarenal location (as
opposed to the more anterior gallbladder, etc.), and the
latter may require IVU to differentiate from pelvic phleboliths.

Magnetic resonance imaging (MRI) depends on first

aligning the hydrogen nuclei (protons) of body tissues with
a powerful magnetic field and then applying radiofrequency
(RF) pulses. Energy released in these circumstances can
be measured and used to create anatomic images dependent on characteristics of the tissues and the introduced
magnetic and RF energy sources. MRI offers superb tissue
contrast, with multiplanar imaging capabilities, although
calcification is not well shown. MRI is safe, with no ionizing
radiation, although MR contrast agents such as gadolinium
may rarely induce patient reactivity. MR may not be suitable for use in claustrophobic patients or those with implanted ferromagnetic devices, such as pacemakers. MRI
has a limited role in defining specific parenchymal lesions
at present, although its sensitivity to iron in hemoglobin
permits its use to image the kidneys in patients with paroxysmal nocturnal hemoglobinuria, myoglobinuria, and epidemic Korean fever. MRI can beautifully show the renal
vasculature, and magnetic resonance angiography (MRA)
is being evaluated as a potential replacement for conventional angiography or a rival to computed tomographic
angiography (CTA) in assessing various arterial lesion,
such as renal artery stenosis). MRI has the following main
roles in renal imaging:
1. Delineating complex renal masses, where CT is not
definitive.
2. Staging renal neoplasms, particularly in evaluating
for renal vein or inferior vena caval extension of
tumor.
3. Diagnosing renovascular lesions.
4. Where renal failure or contrast media reactivity
preclude the use of other modalities. Magnetic
resonance urography (MRU) has been recently
introduced as an alternative to the IVU in highly
selected circumstances.

COMPUTED TOMOGRAPHY

Computed tomography (CT) offers far greater contrast

resolution than conventional radiography, with detailed
anatomic cross-sectional anatomic imaging unaffected by
overlying structures such as bone or gas. Virtually the entire
urinary system and retroperitoneum are well visualized on
CT. CT plays the main role in staging renal neoplasms
and has superseded IVU in trauma. CT may be useful in

RADIONUCLIDE IMAGING

For nuclear medicine studies, radiopharmaceutical

agents with specific renal handling characteristics are employed (see Table 1). These agents are administered and
then the patient is imaged with a gamma camera that can
record the number of counts emitted and the location of
their source. This permits quantitation of function in a

6. Renal Imaging Techniques

49
TABLE I

Most Commonly Used Radiopharmaceuticals in

Renal Imaging

99m'

Mechanism of
Major clinical
renal action usefulness
Glomerular Perfusion
filtration P a r e n c h y m a l I m a g i n g E s t i m a t e
G F R Excretion

99n

Tubular binding
and tubular
secretion

Radionuclide
T c-D T P A

T c- D M S A

99m

T c-G H P

99m

T c-M A G 3

" G a -citrate

99m

Pyelonephritis Estimation
of tubular
mass (i.e., cortical
scar)

Glomerular
Perfusion
filtration and
Excretion
tubular
E s t i m a t i o n o f t u b u l a r mass
secretion and
tubular binding
High renal extraction
Tubular secretion and useful images
and glomerular even with
filtration
moderate renal
dysfunction
Estimate ERPF
N/A P y e l o n e p h r i t i s I n t e r s t i t i a l n e p h r i t i s R e n a l
abscess
N/A R e n a l a b s c e s s

Tc-, 11 'In labeled

WBCs

region as well as anatomic delineation. Nuclear medicine

techniques are particularly useful in assessing the adequacy
of renal perfusion and in determining whether the outflow
tract is intact. Renal parenchymal integrity may also be as sessed.
The scanning method is varied according to the nature
of the information sought. When anatomic information
is desired, the scanning interval for each image typically
encompasses several minutes. Late views may be obtained
after a delay of hours or even 1 to 2 days. In contrast, when
flow is being studied, images are typically of only a few
seconds in duration. The resolution of nuclear medicine
studies is poor. Anatomic information is of limited quality.
Flow studies are, therefore, the major use of nuclear medicine techniques. Single-photon emission computed tomography (SPECT) is utilized at some centers to afford greater
detail. The most common uses of nuclear medicine studies
are listed below.
1. Measurement of renal function. Radionuclide studies
permit calculation of glomerular filtration rate (GFR) and
effective renal plasma flow (ERPF), even in cases of renal
impairment. Appropriate radionuclides are injected intravenously and images obtained. The most accurate calculations of GFR and EPFR are obtained by withdrawing blood
samples at predetermined intervals and using standard
clearance calculations, but strictly count-based computer

imaging methods now closely rival this method, without

requiring blood samples.
2. Measurement of "split" renal function. To determine
whether nephrectomy is warranted or safe, it is often im
portant for the surgeon to know the relative contribution
to total renal function of each kidney. Computer-enhanced
scan techniques can determine the contribution of each
kidney to ERPF or GFR. Split renal function measure
ments are also helpful in follow up of surgical procedures
that relieve unilateral obstruction.
3. Renovascular hypertension. Differential renal blood
flow studies using 99mTc-DPTA or 99mTc-MAG3 pre- and
postadministration of an ACE inhibitor such as captopril in
an appropriately screened hypertensive patient with intact
renal function will detect renovascular disease with a sensi
tivity and specificity exceeding 90%. (see chapter 78, Fig. 3.)
4. Contraindication to contrast media. While radionuclide methods do not provide the anatomic detail of other
imaging methods, they may provide information regarding
arterial perfusion, differential renal function, and urinary
tract obstruction. Radionuclide methods may be particu
larly suitable for patients with renal failure or severe con
trast media sensitivity.
Aside from pregnancy, there are no contraindications to
the use of radionuclides for diagnostic purposes.
5. Evaluation of renal transplants. Radionuclide imaging
may detect impaired blood flow at the renal arterial anastomotic site, urinary tract obstruction, and extravasation of
urine due to disruption at the ureteric anastomosis. This
modality complements US for these purposes.
6. Obstructive uropathy. The diuretic renogram (Fig. 2)
is very helpful, particularly in children, who demonstrate an
MIN
FIGURE 1 R a d i o n u c l i d e , d i u r e t i c r e n o g r a p h y . T h e p l a t e s h o w s
1184

888

592

296

an accumulation of counts over the kidneys, 30 minutes after

intravenous injection, of furasemide and the tracer MAG3. The
left kidney (left) is normal. Note the flat curve on the right , indicat ing that pooled isotope remains in the kidney, without a normal
fall off as on the left side; the right side is partially obstructed.

Howard j. Mindell and Jonathan Fairbank

50

enlarged renal pelvis by US or IVU. This test distinguishes

between true obstruction at the ureteropelvic junction and
nonobstructive hydronephrosis. Furosemide is injected before or after administration of 99m Tc DTP A or MAG 3,
renal images are obtained, and a renogram curve of activity
in each kidney is constructed. Dilatation without significant
obstruction is suggested when radionuclide accumulation
in the dilated area is reduced at high urine flow rates.
7. Renal infection or scar. The IVU may appear normal
in patients with pyelonephritis, and the differentiation
between this and cystitis may be difficult. 99m Tc-DMSA
and glucoheptonate (GHP) are renal cortical scanning
agents. Areas of inflammation or scar will demonstrate
no uptake. The agents are also helpful in determining
the amount of remaining renal cortex in children with
chronic urinary tract infections and vesicoureteric reflux.
99m
Tc- or m ln-labeled WBCs may be used to identify
renal abscesses. In patients with low WBC counts, gallium67 may be used.
ANGIOGRAPHY

Intravenous administration of contrast medium has not

proved suitable for renal angiography, and direct intraarterial injections, either into the main renal arteries or selectively into smaller branches, are generally necessary in
this modality.
Renal angiography can be used for diagnostic and/or
therapeutic purposes.
Diagnostic Angiography

1. Suspected renal artery lesions. Renal angiography is

the definitive imaging procedure where renovascular hypertension is suspected. It is indicated when medical therapy for hypertension is ineffective and the patient is a
candidate for surgical or angioplastic therapy. As stated

earlier, CTA and MRA are under investigation as replacements for conventional angiography in this area. Renal
vein sampling for renin levels may assist selecting patients
for treatment. Patients with acute occlusion or thrombosis
of the renal arteries, embolic processes, or posttraumatic
renal vascular injury may be candidates for renal angiography. In patients with polyarteritis nodosa, selective renal
angiography is the only method with sufficient resolution
to detect the characteristic tiny peripheral aneurysms.
2. Unexplained hematuria. Angiography may be re
quired to investigate the cause for unexplained hematuria.
Vascular malformations may be suspected on US or CT
but generally will require angiography for definitive delin
eation, especially if therapy is planned.
3. Renal transplantation. Angiography may be needed
to map the renal arterial system in prospective renal donors,
where multiple renal arteries or vascular anomalies may
complicate transplant surgery. CTA is used for this purpose
at some transplant centers. Angiography is required to
diagnose posttransplant renal artery stenoses or occlusions.
5. Renal vein disorders. Although renal vein thrombosis
or occlusion can be shown by CT or MRI, rarely direct
renal venography may be required, for instance if other
modalities are not diagnostic.
6. Miscellaneous. Complex renal masses, or complica
tions of polycystic disease or trauma, may require angiog
raphy.
Interventional Angiography

Angiography has become increasingly important as a

therapeutic maneuver rather than a simple diagnostic technique. Balloon angioplasty may be used to dilate renal
artery stenosis, both in the native and in the transplanted
kidney (Fig. 3). Ostial lesions, previously thought to be
refractory to this method, may be amenable to treatment
with endovascular stinting. Further details on the use of

FIGURE 3 (a) Computed tomographic angiography (CTA) showing the computer-generated 3D reconstruction of the
aorta and the main renal arteries, possible after a simple IV injection of contrast media. The arrow indicates a significant
stenosis in the main right renal artery, (b) Selective right renal angiogram after successful angioplasty, in the same patient
as (a), with an arrow showing the now patent right main renal artery. Figures courtesy of Dr. Ken Najarian, FAHC,
Burlington, Vt.

6. Renal Imaging Techniques

51

angioplasty in renal artery stenosis are provided in Chapter

78. Angiography may also be used to embolize bleeding
sites within the kidney that may have resulted from trauma
or as a complication of interventional procedures such as
a percutaneous biopsy. Angiography may also be used for
selected infusion of a thrombolytic agent for treatment of
renal artery or renal vein thrombosis.
Voiding Cystourethrography

This study is useful in demonstrating vesicoureteric reflux. Voiding Cystourethrography (VCU) can be performed
with fluoroscopic or radionuclide technique, depending on
local preferences.
CHOICE OF IMAGING PROCEDURE FOR SPECIFIC
CLINICAL SITUATIONS
Arterial Disease

US can readily diagnose aortic aneurysm, but CT is more

accurate where aneurysmal bleeding is suspect, and CT or
CTA are better for surgical planning, especially to show
the exa ct levels of the renal arteries. CT can diagnose acute
renal infarction by showing parenchymal perfusion defects
with a peripheral "rim sign" of collateral flow. As previously suggested, although color Doppler US can demonstrate renovascular disease such as renal artery stenosis in
individual patients, it is not recommended as a screening
tool for renovascular hypertension. In selected patients
where significant hypertension is not amenable to medical
control, and where surgical or angioplastic therapy is contemplated, a combination of captopril renography and/or
renal angiography is required. Because of the frequent
occurrence of double, or even triple, renal arteries, US
cannot reliably rule out significant renal artery stenosis. On
the other hand, even where US clearly shows a renovascular
lesion, angiography is always needed as a precursor to
angioplasty or surgical intervention.
Congenital Disorders

The most common congenital renal anomalies for which

imaging is used are the various forms of renal ectopia,
fusion anomalies such as horseshoe kidney (Fig. 4), and
obstructive processes of the ureteropelvic junction (UPJ)
or distal ureter. US and IVU are the primary imaging
tools in evaluating these lesions, although occasionally a
as horseshoe kidney pres ents as an incidental finding at
CT or with a radionuclide study, even a bone scan. UPJ
obstruction can be further evaluated by diuretic urography
or radionuclide renography.
Inflammatory Diseases

In acute pyelonephritis, the IVU is usually not indicated

and in any case ordinarily not diagnostic. In selected patients, CT can diagnose acute pyelonephritis by showing
characteristic peripheral radiolucent perfusion defects in

FIGURE 4 IVU demonstrating a horseshoe kidney. A black

dashed line has been drawn to show the lower margin of the renal
parechyma as it crosses the mid line to fuse with the opposite
kidney.

the renal parenchyma, frequently with edema in the perinephric fat. When urinary tract infections follow an atypical
or protracted course, US, and with better specificity CT, can
delineate infected renal cysts, renal abcesses, or infected
perirenal collections. Either US or CT can guide interventional procedures for diagnosing and treating these lesions.
Tuberculosis causes strictures of the renal collecting systems or ureters, and the diagnosis may be suggested by
IVU. Chronic pyelonephritis, frequently associated with
reflux nephropathy, produces scarred, and in advanced dis ease, shrunken kidneys readily shown on IVU or cross
sectional imaging.
Neoplasia

The main grouping of renal tumors divides renal parenchymal masses from urothelial lesions (see Chapter 58).
Although cystic or solid renal parenchymal masses may be
first detected by IVU, they require US and/or CT for further differentiation and staging (Fig. 5). MRI is available
in some instances where more conventional imaging is not
definitive, where the patient is sensitive to iodinated intravascular contrast media, or where a question about renal
vasculature is unresolved. Urothelial tumors, nearly always
transitional cell carcinoma (TCC), are best shown on IVU
or retrograde pyelographythe latter serving as a guide
for brush biopsy. In selected patients, cross sectional im-

Howard J. Mindell and Jonathan Fairbank

52

chyma, but rarely pinpoints the diagnosis, which may require renal biopsy, the latter amenable to US guidance. The
main goal of imaging is to rule out obstruction, polycystic
kidney disease, renal papillary necrosis, vascular disease,
or reflux nephropathy. For patients presenting in renal
failure, US is an excellent noninvasive method to differentiate intrinsic parenchymal disease from obstructive uropathy.
UROLITHIASIS

FIGURE 5 C T showing renal carcinoma. This cross section

through the level of the kidneys was obtained using oral and IV
contrast media. The lower black arrow points to the irregulary
radiolucent, necrotic tumor mass. The second black arrow, higher
in position as the reader views the illustration, points to a filling
defect in the inferior vena cava, representing tumor thrombus. A
normal cava would be homogeneously bright with enhancing
blood, like the aorta.

aging such as CT may be needed to diagnose or confirm

TCC of the renal pelvis or ureter.

Most urinary calculi are calcified and therefore readily

detectable by KUB. Radiolucent stones appear as filling
defects in the contrast opacified area on IVU. These techniques are the main imaging studies employed for nephrolithiasis. Recently, helical unenhanced CT has shown promise in the diagnosis of renal colic due to ureteric calculi.
CT can also demonstrate radiolucent calculi such as uric
acid stones that are not seen on KUB. As the IVU can
precisely locate the position of a ureteral calculus, it is
used to plan therapy for renal or urinary calculi, although
retrograde pyelography may be needed, especially during
the course of stent placement or extracorporeal shock wave
lithotripsy (ESWL). The role of ultrasound in patients with
nephrolithiasis is limited. It is very sensitive in showing tiny
intrarenal calculi or nephrocalcinosis but of only limited
applicability in diagnosing ureteric stones as the mid and
distal ureter are only rarely visualized by US.

Cystic Disorders

Renal Transplantation

Benign renal cysts occur in up to 30% of normal adults

and can nearly always be readily differentiated from solid
neoplasm by US. Complex cysts may require CT for definitive evaluation. Both infantile autosomal recessive and
adult autosomal dominant polycystic renal disease are
readily diagnosed by US; the former may be detected in
utero, and the latter lends itself to US for screening where
indicated (see Chapter 45). Medullary sponge kidney (tubular ectasia) is diagnosed by IVU (see Fig. 1, Chapter 52).

Renal transplantation donors need at least an IVU preoperatively, and as previously indicated, depending on institutional preferences, angiography or CTA to assess vascular anatomy. In recipients posttransplantation, color flow
Doppler US documents the status of vascular perfusion to
the transplanted kidney, and US can document hydronephrosis, renal volumes, and extrarenal collections such as
lymphocele, but cannot reliably differentiate rejection from
acute tubular necrosis. Radionuclide studies can also assess
perfusion or obstruction, and rarely retrograde or even
percutaneous antegrade pyelography is required to assess
ureteral patency. Angiography may be needed to show
renal artery stenoses to the transplant.

Trauma
Helical (spiral) CT, where very rapid CT of the entire
upper abdomen can be done in a single breath hold, has
largely supplanted the IVU for evaluating renal trauma,
because of its speed and accuracy, and its multiorgan
sweep. CT also assesses the renal vascular supply, although angiography is needed in selected cases or major
vascular trauma, where diagnosis and/or angiographic
therapy is required.
Parenchymal disorders
Renal parenchymal disease consists of a broad range of
processes, from inflammatory or immunologic disorders to
toxic or ischemic lesions. Imaging can show small kidneys,
and US may demonstrate echogenic, abnormal paren-

Ureteral Obstruction
Bilateral ureteric obstruction causing renal failure is rare
in adults in the absence of malignancy. Although specific
lesions such as retroperitoneal fibrosis must be considered,
US can show hydronephrosis, but ureteric assessment will
require IVU, or cross-sectional imaging such as CT or MRI
to demonstrate the retroperitoneum, or retrograde pyelography. Acute unilateral ureteric obstruction, commonly
due to renal colic with passage of a ureteric calculus is
usually readily diagnosed by IVU (Fig. 6) although as previously suggested unenhanced CT is emerging as an effective technique in this area.

6. Renal Imaging Techniques

53

lesioni.e., if US is done to delineate a renal tumor, the

renal vein may be also shown by color flow Doppler US.
CT and MRI can also show the renal veins, as well as the
IVC. Where isolated renal vein thrombosis is suspected,
contrast CT can be definitive, both in showing the enlarged
renal vein and collateral circulation that may develop and
in showing the actual thrombus as a filling defect in the
opacified renal vein. Again, where CT is suboptimal for
technical reasons or the patient is reactive to iodinated
contrast media, MRI can be diagnostic of disorders of the
IVC or renal veins.
Bibliography

FIGURE 6 IVU shows acute obstruction from a ureteral calculus.

The inset is a section from the "scout" KUB, showing a 5 mm
opaque ureteric calculus (white arrow). The 30-minute film after
IV contrast injection shows right hydronephrosis above (black
arrow with slender tail), with hydroureter (paired black arrows)
tapering to the calculus seen on the KUB (black arrow with
thick tail).

Renal Vein Lesions

The renal vein can be evaluated as part of the crosssectional imaging procedure used to study the primary

Blaufox MD, ed: Evaluation of Renal Function and Disease with

Radionudides. 2nd ed. Karger, Basel, 1989.
Bude RO, Rubin J A: Detection of renal artery stenosis with Doppler sonography: It is more complicated than originally thought.
Radiology 196:612-613, 1995.
Dodd GD, Tublin ME, Shah A, Zajko AB: Review. Imaging of
vascular complications associated with renal transplants. AJR
157:449-459, 1991.
Dunnick NR, Cohan RH: Commentary. Cost, corticosteroids, and
contrast media. AJR 162:527-529, 1994.
Mindell HJ, Cochran ST: Commentary. Current perspectives in
the diagnosis and treatment of urinary stone disease. AJR
163:1314-1315, 1994.
Pollack HM, ed: Clinical Urography. Saunders, Philadelphia, 1990.
Pollack HM, Wein AJ: State of the art imaging of renal trauma.
Radiology 172:297-308, 1989.
Resnick MI, Rifkin MD: Ultrasonography of the Urinary Tract.
3rd ed. Williams and Wilkins, Baltimore, 1991.
Rubin GD, Dake MD, Napel S, McDonnell CH, Jeffrey RB:
Three-dimensional spiral CT angiography of the abdomen: Initial clinical experience. Radiology 186:147-152, 1993.
Solomon R, Warner C, Mann D, D'Elia J, Silva P. Effects of saline,
mannitol and furosemide on acute decreases in renal function
produced by radiocontrast agents. N Engl J Med 331:14161420, 1994.
Steinberg EP, Moore RD, Powe NR, et al: Safety and cost effectiveness of high -osmolality as compared with low-osmolality
contrast material in patients undergoing cardiac angiography.
N Engl J Med 326:425-430, 1992.
Zagoria RJ, Berchtold RE, Dyer RB: Staging of renal adenocarcinoma: Role of various imaging procedures. AJR 164:363-370,
1995.

SECTION 2
ACID-BASE, FLUID, AND
ELECTROLYTE DISORDERS

HYPONATREMIA AND
HYPOOSMOLAR DISORDERS
JOSEPH G. VERBALIS

tration of Na+ per liter of plasma is artifactually decreased

because of the increased relative proportion occupied by
lipid or protein. Fortunately, measure ment of plasma [Na + ]
by ion-specific electrodes, which is now commonly employed by most clinical laboratories, is much less influenced
by high concentrations of lipids or proteins than is measurement of plasma [Na + ] by flame photometry, although such
errors can nonetheless still occur. However, because direct
measurement of plasma osmolality is based on the colligative properties of solute particles in solution, measured
plasma osmolality will not be affected by the increased
lipids or proteins. Second, high concentrations of effective
solutes other than Na + can cause relative decreases in
plasma [Na + ] despite an unchanged plasma osmolality; this
commonly occurs with hyperglycemia. Misdiagnosis can be
avoided again by direct measurement of plasma osmolality,
or by correcting the plasma [Na + ] by 1.6 mEq/L for each
100 mg/dL increase in plasma glucose concentration above
100 mg/dL.

The incidence of hyponatremia depends on the nature

of the patient population and also the criteria used to establish the diagnosis. Hospital incidences of 15 to 22% are
common if hyponatremia is defined as plasma [Na + ]
<135 mEq/L, but only 1 to 4% of patients have a plasma
[Na + ] <130. Although most cases are mild, hyponatremia is
important clinically because: (1) acute severe hyponatremia
can cause substantial morbidity and mortality; (2) mild
hyponatremia can progress to more dangerous levels during management of other disorders; (3) general mortality
appears to be higher in patients with even asymptomatic
hyponatremia; and (4) overly rapid correction of chronic
hyponatremia can produce severe neurological deficits
and death.
Definitions

Hyponatremia is of clinical significance only when it reflects corresponding hypoosmolality of the plasma. Plasma
osmolality can be measured directly by osmometry, or calculated as
Posm (mOsm/kg H2O) = 2 X plasma [Na + ] (mEq/L) +
glucose (mg/dL)/18 + BUN (mg/dL)/2.8.

PATHOGENESIS

The presence of significant hypoosmolality always indicates excess water relative to solute in the ECF. Because
water moves freely between the ICF and ECF, this also
indicates an excess of total body water relative to total
body solute. Imbalances between water and solute can be
generated initially either by depletion of body solute more
than body water or by dilution of body solute from increases
in body water more than body solute (Table 1). It should
be recognized, however, that this distinction represents an
oversimplification, because most hypoosmolar states include components of both solute depletion and water retention (e.g., isotonic solute losses, as occurs during an acute
hemorrhage, do not produce hypoosmolality until the subsequent retention of water from ingested or infused hypotonic fluids causes a secondary dilution of the remaining
ECF solute). Nonetheless, this concept has proven useful
because it provides a simple framework for understanding
the diagnosis and therapy of hypoosmolar disorders.

Both methods produce comparable results under most conditions, as does simply doubling the plasma [Na + ]. However, total osmolality is not always equivalent to "effective"
osmolality, sometimes referred to as the "tonicity" of the
plasma. Solutes compartmentalized to the extracellular
fluid (ECF) are effective solutes, because they create osmotic gradients across cell membranes leading to osmotic
movement of water from the intracellular fluid (ICF) to
ECF compartments. Solutes that freely permeate cell membranes (urea, ethanol, methanol) are not effective solutes,
since they do not create osmotic gradients across cell membranes and therefore are not associated with secondary
water shifts. Only the concentration of effective solutes
in plasma should be used to determine whether clinically
significant hypoosmolality is present.
Hyponatremia and hypoosmolality are usually synonymous, but with two important exceptions. First, pseudohyponatremia can be produced by marked elevation of plasma
lipids and/or proteins. In such cases the concentration of
Na + per liter of plasma water is unchanged, but the concen-

Primer on Kidney Diseases, Second Edition

DIFFERENTIAL DIAGNOSIS

The diagnostic approach to hypoosmolar patients should

include a careful history (especially concerning medica57

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

58

Joseph G. Verbalis
TABLE I

Pathogenesis of Hypoosmolar Disorders"

Depletion (primary decreases in total body solute + secondary
water retention)*:
1. Renal solute loss
Diuretic use
Solute diuresis (glucose, marmitol)
Salt-wasting nephropathy
Mineralocorticoid deficiency
2. Nonrenal solute loss
Gastrointestinal (diarrhea, vomiting, pancreatitis, bowel
obstruction)
Cutaneous (sweating, burns)
Blood loss
Dilution (primary increases in total body water secondary
solute depletion)':
1. Impaired renal free water excretion
A. Increased proximal reabsorption
Hypothyroidism
B. Impaired distal dilution
Syndrome of inappropriate antidiuresis (SIAD)
Glucocorticoid deficiency
C. Combined increased proximal reabsorption and
impaired distal dilution
Congestive heart failure
Cirrhosis Nephrotic
syndrome
D. Decreased urinary solute excretio n
Beer potomania
2. Excess water intake
Primary polydipsia
Dilute infant formula
Modified from Verbalis (1995).
6
Virtually all disorders of solute depletion are accompanied by
some degree of secondary retention of water by the kidneys in
response to the resulting intravascular hypovolemia; this mechanism can lead to hypoosmolality even when the solute depletion
occurs via hypotonic or isotonic body fluid losses.
c
Disorders of water retention primarily cause hypoosmolality
in the absence of any solute losses, but in some cases of SIAD
secondary solute losses occur in response to the resulting intravascular hypervolemia and this can then further aggravate the dilutional hypoosmolality in the plasma (but note that this pathophysiology does not likely contribute to the hyponatremia of edemaforming states such as congestive heart failure and cirrhosis, since
in these cases multiple factors favoring sodium retention will result
in an increased total body sodium).

tions), clinical assessment of the ECF volu me status, a

thorough neurological evaluation, plasma electrolytes, glucose, BUN and creatinine, uric acid, calculated and/or directly measured plasma osmolality, and simultaneous urine
electrolytes and osmolality. A definitive diagnosis is not
always possible at the time of presentation, but an initial
categorization according to the patient's clinical ECF volume status will allow a determination of the appropriate
initial therapy in the majority of cases (Fig. 1).

Decreased ECF Volume (Hypovolemia)

Clinically detectable hypovolemia, generally determined
most sensitively by careful measurement of orthostatic
changes in blood pressure and pulse rate, always indicates
some degree of solute depletion. Elevation of plasma BUN
is a useful laboratory correlate of decreased ECF volume.
Even isotonic or hypotonic volume losses can lead to hypoosmolality if water or hypotonic fluids are ingested or infused as replacement. A low urine Na+ concentration (t/Na)
in such cases suggests a nonrenal cause of the solute depletion, whereas a high t/Na suggests renal causes of solute
depletion (Table 1). Diuretic use is the most common cause
of hypovolemic hypoosmolality, and thiazides are more
commonly associated with severe hyponatremia than are
loop diuretics such as furosemide. Although this represents
a prime example of solute depletion, the pathophysiological
mechanisms underlying the hypoosmolality are complex
and are composed of multiple potential components including free water retention. Furthermore, many such patients
do not present with clinical evidence of marked hypovolemia, mainly because ingested water has been retained in
response to nonosmotically stimulated vasopressin (AVP)
secretion, as is often true for all disorders of solute depletion. To further complicate diagnosis, the t/Na may be high
or low depending on when the last diuretic dose was taken.
Consequently, almost any suspicion of diuretic use mandates careful consideration of this diagnosis. A low plasma
[K+] is an important clue to diuretic use, because few other
disorders that cause hyponatremia and hypoosmolality also
cause appreciable hypokalemia. Whenever the possibility
of diuretic use is suspected in the absence of a positive
history, a urine screen for diuretics should be performed.
Most other etiologies of renal or nonrenal solute losses
causing hypovolemic hypoosmolality will be clinically apparent, although some cases of salt-wasting nephropathies
(chronic interstitial nephropathy, polycystic kidney disease,
obstructive uropathy, or Bartter's syndrome) or mineralocorticoid deficiency (Addison's disease) may be challenging
to diagnose during early phases of these diseases.
Normal ECF Volume (Euvolemia)
Virtually any disorder associated with hypoosmolality
can potentially present with a volume status that appears
normal by standard methods of clinical evaluation. Because
clinical assessment of volume status is not very sensitive,
normal or low levels of plasma BUN and uric acid are very
helpful laboratory correlates of relatively normal, or even
slightly expanded, ECF volume. In these cases a low t/Na
suggests a depletional hypoosmolality secondary to ECF
losses with subsequent volume replacement by water or
other hypotonic fluids. As discussed earlier, such patients
may appear euvolemic by all the usual clinical parameters
used to assess hydrational status. Primary dilutional disorders are less likely in the presence of a low [/ Na (^30 mEq/
L), although this pattern can occur in hypothyroidism. A
high [7Na (>30 mEq/L) generally indicates a dilutional hypoosmolality such as syndrome of inappropriate antidiure-

7. Hyponatremia and Hypoosmolar Disorders

59

HYPOOSMOLALITY

HYPONATREMIA

DECREASED

INCREASED

HYPERGLYCEMIA OR
<100
PSEUDOHYPONATREMIA?

ECF VOLUME STATUS?

D
E
P
L
E
T
I
O
N

NORMAL URINE

OSMOLALITY?

.
SOLUTE REPLETION f
NSSK+

H
Y

URINE Na ?
>30

DIURETIC USE?

\\ _ A
- ^ADRENAL

INSUFFICIENCY? ) 1", 2 -

^---------- V

(jHYPOTHYROIDISM?
/^IMPROVEMENT^ "|~V
v
INPosM?
J

RENAL FAILURE?

EVALUATE ETIOLOGY

P
O
O
S
M
O
L
A
L
I
T
Y

OF

CNS SYM

EVALUATE/RX UNDERLYING DISEASE

ASYMPTOMATIC?^
M
HYPERTONIC SALINE fPLASMA
Na + >120? VV
FUROSEMIDE
+
\APLASMA Na >20?/

FLUID RESTRICTION,
QIC DRUGS ASSOCIATED
WITH { H2O EXCRETION

NO FURTHER
RX

-N-f

SUSTAINED
HYPOOSMOLALITY?
(OR ABNORMAL H^ LOAD)

EVALUATE
ETIOLOGY
OF
SIAD

CHRONIC RX:
1.
2.
3.
4.

FLUID RESTRICTION
DEMECLOCYCLINE
UREA
(AVP ANTAGONISTS)

FIGURE I Schematic summary of the evaluation and therapy of hypoosmolar patients. The dark arrow in the center
emphasizes that the presence of central nervous system dysfunction due to hyponatremia should always be assessed
immediately, so that appropriate therapy can be started as soon as possible in symptomatic patients while the outlined
diagnostic evaluation is proceeding. N, no; Y, yes; ECF, extracellular fluid volume; NSS, normal (isotonic) saline; Rx,
treat; 1, primary; 2, secondary; Posm , plasma osmolality; d/c, discontinue; SIAD, syndrome of inappropriate antidiuresis.
Numbers referring to osmolality are in mOsm/kg H2O, numbers referring to plasma Na+ concentration are in mEq/L.
Modified from Verbalis, 1995.

60

Joseph G. Verbalis
TABLE 2

Criteria for the Diagnosis of SIAD

Essential
1. Decreased effective osmolality of the extracellular fluid
(P osm < 275 mOsm/kg H2 O).
2. Inappropriate urinary concentration (t/osm > 100 mOsm/kg
H2 O with normal renal function) at some level of
hypoosmolality.
3. Clinical euvolemia, as denned by the absence of signs of
hypovolemia (orthostasis, tachycardia, decreased skin turgor,
dry mucous membranes) or hypervolemia (subcutaneous
edema, ascites).
4. Elevated urinary sodium excretion while on a normal salt
and water intake.
5. Absence of other potential causes of euvolemic
hypoosmolality: hypothyroidism, hypocortisolism (Addison's
disease or pituitary ACTH insufficiency) and diuretic use.
Supplemental
6. Abnormal water load test (inability to excrete at least 90%
of a 20 mL/kg water load in 4 h and/or failure to dilute Uosm
to < 100 mOsm/kg H2 O).
7. Plasma AVP level inappropriately elevated relative to
plasma osmolality.
8. No significant correction of plasma [Na+] with volume
expansion but improvement after fluid restriction.
Note. Modified from Verbalis (1995).

sis (SIAD) (Table 1). SIAD is the most common cause of

euvolemic hypoosmolality. Although originally described
as the syndrome of inappropriate antidiuretic hormone
secretion (SIADH), approximately 10 to 20% of patients
who meet all established criteria for SIADH do not have
measurably elevated plasma AVP levels, so it is actually
more accurate to use the term SIAD rather than SIADH to
describe this entire group of disorders. The clinical criteria
necessary for a diagnosis of SIAD remain as defined by
Bartter and Schwartz in 1967 (Table 2), but several points
deserve emphasis. First, tru e ECF hypoosmolality must be
present and hyponatremia secondary to pseudohyponatremia or hypoglycemia excluded. Second, urinary osmolality must be inappropriate for plasma hypoosmolality. This
does not require a / osm >Posm , but simply that the urine
osmolality is greater than maximally dilute (i.e., f/ osm >
100 mOsm/kg H2O). Furthermore, urine osmolality need
not be inappropriately elevated at all levels of Posm but
simply at some level under 275 mOsm/kg H2O, since in
patients with a reset osmostat AVP secretion can be suppressed at some level of osmolality resulting in maximal
urinary dilution and free water excretion at plasma osmolalities below this level. Although some consider a reset
osmostat to be a separate disorder rather than a variant
of SIAD, such cases nonetheless illustrate that some hypoosmolar patients can exhibit an appropriately dilute urine
at some, though not all, plasma osmolalities. Third, clinical
euvolemia must be present to diagnose SIAD, and this
diagnosis cannot be made in a hypovolemic or edematous
patient. This does not mean that patients with SIAD cannot
become hypovolemic for other reasons, but in such cases

it is impossible to diagnose the underlying SIAD until the

patient is rendered euvolemic. The fourth criterion, renal
salt wasting, has probably caused the most confusion regarding SIAD. The importance of this criterion lies in its
usefulness in differentiating hypoosmolality caused by a
decreased relative intravascular volume in which case renal
Na + conservation occurs, from dilutional disorders in which
urinary Na + excretion is normal or increased due to ECF
volume expansion. However, t/ Na can also be high in renal
causes of solute depletion such as diuretic use or Addison's
disease, and conversely patients with SIAD can have a
low urinary Na+ excretion if they subsequently become
hypovolemic or solute depletedconditions sometimes
produced by imposed salt and water restriction. Consequently, although high urinary Na+ excretion is the rule in
most patients with SIAD, its presence does not confirm
this diagnosis nor does its absence rule out the diagnosis.
The final criterion emphasizes that SIAD remains a diagnosis of exclusion, and the absence of other potential causes
of hypoosmolality must always be verified. Glucocorticoid
deficiency and SIAD can be especially difficult to distinguish, because hypocortisolism can cause elevated plasma
AVP levels and also has direct renal effects to prevent
maximal urinary dilution. No patient should be diagnosed
as having SIAD without a thorough evaluation of adrenal
function, preferably via a rapid ACTH stimulation test.
Many different disorders have been associated with SIAD,
and these can be divided into several major etiologic groups
(Table 3).
Some cases of euvolemic hyponatremia do not fit particularly well into either the dilutional or depletional catego-

TABLE 3

Common Etiologies of SIAD

Tumors Pulmonary/mediastinal (bronchogenic
carcinoma,
mesothelioma, thymoma) Nonchest (duodenal
carcinoma, pancreatic carcinoma,
ureteral/prostate carcinoma, uterine carcinoma,
nasopharyngeal carcinoma, leukemia)
Central nervous system disorders
Mass lesions (tumors, brain abscesses, subdural hematoma)
Inflammatory diseases (encephalitis, meningitis, lupus)
Degenerative/demyelinative diseases (Guillain -Barre; spinal
cord lesions) Miscellaneous (subarachnoid hemorrhage,
head trauma, acute
psychosis, delirium tremens, pituitary stalk section)
Drug induced
Stimulated AVP release (nicotine, phenothiazines, tricyclics,
serotonin reuptake inhibitors) Direct renal effects
and/or potentiation of AVP effects
(desmopressin, oxytocin, prostaglandin synthesis inhibitors)
Mixed or uncertain actions (chlorpropamide, clofibrate,
carbamazepine, cyclophosphamide, vincristine)
Pulmonary diseases
Infections (tuberculosis, aspergillosis, pneumonia, empyema)
Mechanical/ventilatory (acute respiratory failure, COPD,
positive pressure ventilation)

61

7. Hyponatremia and Hypoosmolar Disorders

ries. Chief among these is the hyponatremia that sometimes

occurs in patients who ingest large volumes of beer with
little food intake for prolonged periods, frequently called
"beer potomania." Even though the volume of fluid ingested may not seem sufficiently excessive to overwhelm
renal diluting mechanisms, in these cases free water excretion is limited by very low urinary solute excretion thereby
causing water retention and dilutional hyponatremia. However, because such patients have very low Na+ intakes as
well, it is likely that relative depletion of body Na+ stores
also is a contributing factor to the hypoosmolality in
some cases.
Increased ECF Volume (Edema, Ascites)

The presence of hypervolemia, as detected clinically by

the presence of edema and/or ascites, indicates whole body
sodium excess and hypoosmolality in these patients suggests a relatively decreased intravascular volume and/or
pressure leading to water retention as a result of both
elevated plasma AVP levels and decreased distal delivery
of glomerular filtrate. Such patients usually have a low
t/Na because of secondary hyperaldosteronism, but under
certain conditions the I7Na may be elevated (e.g., glucosuria
in diabetics, diuretic therapy). Hyponatremia generally
does not occur until fairly advanced stages of diseases such
as congestive heart failure, cirrhosis, and nephrotic syndrome, so diagnosis is usually not difficult. Renal failure
can also cause retention of both sodium and water, but in
this case the factor limiting excretion of excess body fluid
is not decreased effective circulating volume but rather
decreased glomerular filtration. It should be remembered
that even though many edema-forming states have secondary increases in plasma AVP levels as a result of decreased
effective arterial blood volume, they are not classified as
cases of SIAD, since they fail to meet the criterion of
clinical euvolemia (Table 2, point 3). Although it can be
argued that this represents a semantic distinction, it is important with regard to segregating identifiable etiologies
of hyponatremia that are associated with different methods
of evaluation and therapy. Primary polydipsia also can
cause hypoosmolality in a small subset of patients with
underlying SIAD, particularly psychiatric patients with
long-standing schizophrenia on neuroleptic drugs, or even
more rarely in patients with normal kidney function if the
volumes ingested exceed the maximum renal free water
excretory rate of approximately 1000 mL/h. However, these
patients rarely manifest overt signs of volume excess since
water retention alone without sodium excess does not cause
clinically apparent hypervolemia.
Although incidences will vary according to the population being studied, sequential analysis of hyponatremic patients admitted to a large university teaching hospital revealed that approximately 20% were hypovolemic, 20%
had edema-forming states, 33% were euvolemic, 15% had
hyperglycemia-induced hyponatremia, and 10% had renal
failure. Consequently, euvolemic hyponatremia generally
constitutes the largest single group of hyponatremic patients found in this setting.

CLINICAL MANIFESTATIONS OF HYPONATREMIA

The clinical manifestations of hyponatremia are largely

neurological and primarily reflect brain edema resulting
from osmotic water shifts into the brain. Significant symp toms generally do not occur until plasma [Na + ] falls below
125 mEq/L, and the severity of symptoms can be roughly
correlated with the degree of hypoosmolality. However,
individual variability is marked, and, for any single patient,
the level of plasma [Na + ] at which symptoms will appear
cannot be predicted. Furthermore, several factors other
than the severity of the hypoosmolality also affect the degree of neurological dysfunction. Most important is the
period during which hypoosmolality develops. Rapid development of severe hypoosmolality is frequently associated
with marked neurologic symptoms, whereas gradual development over several days or weeks is often associated with
relatively mild symptomatology despite profound degrees
of hypoosmolality. This is because the brain can counteract
osmotic swelling by excreting intracellular solutes (including potassium and organic osmolytes). Because this is a
time-dependent process, rapid development of hypoosmolality can result in brain edema before this adaptation occurs, but with slower development of the same degree of
hypoosmolality brain cells can lose solute sufficiently rapidly to prevent brain edema and neurological dysfunction.
Underlying neurological disease also affects the level of
hypoosmolality at which CNS symptoms appear; moderate
hypoosmolality is of little concern in an otherwise healthy
patient, but can cause morbidity in a patient with an underlying seizure disorder. Nonneurological metabolic disorders (hypoxia, hypercapnia, acidosis, hypercalcemia, etc.)
similarly can affect the level of plasma osmolality at which
CNS symptoms occur. Recent clinical studies have suggested that menstruating females and young children may
be particularly susceptible to the development of neurological morbidity and mortality during hyponatremia, especially in the acute postoperative setting. The true clinical
incidence as well as the underlying mechanisms responsible
for these sometimes catastrophic cases remains to be determined.
THERAPY

Despite some continuing controversy regarding the optimal speed of correction of osmolality in hyponatremic patients, there is now a relatively uniform consensus about
appropriate therapy in most cases (Fig. 1). If any degree
of clinical hypovolemia is present, the patient should be
considered to have a solute depletion-induced hypoosmo lality and treated with isotonic (0.9%) NaCl at a rate appro priate for the estimated volume depletion. If diuretic use
is known or suspected, saline should be supplemented with
potassium (30 to 40 mEq/L) even if plasma [K+ ] is not low
because of the propensity of such patients to have total
body potassium depletion. Most often the hypoosmolar
patient will be clinically euvolemic, but several situations
will dictate a reconsideration of potential solute depletion
even in the patient without clinically apparent hypovo-

62
lemia: a decreased [7Na, any history of recent diuretic use,
and any suggestion of primary adrenal insufficiency. Whenever a reasonable likelihood of depletional rather than
dilutional hypoosmolality exists, it is appropriate to treat
initially with a trial of isotonic NaCl. If the patient has
SIAD, no harm will have been done with a limited (1 to
2 L) saline infusion, because such patients will simply excrete excess NaCl without significantly changing their Posm.
However, this therapy should be abandoned if plasma [Na+]
does not improve, because longer periods of continued
isotonic NaCl infusion can worsen the hyponatremia by
virtue of gradual water retention. The treatment of euvolemic hypoosmolar patients will vary depending on their
presentation. A patient meeting all criteria for SIAD except
that Uosm is low should simply be observed since this may
represent spontaneous reversal of a transient form of
SIAD. If there is any suspicion of either primary or secondary adrenal insufficiency, glucocorticoid replacement
should be started immediately after completion of a rapid
ACTH stimulation test. Prompt water diuresis following
initiation of glucocorticoid treatment strongly supports glucocorticoid deficiency, but the absence of a quick response
does not negate this diagnosis since several days of glucocorticoids are sometimes required for normalization of
^"osm- Hypervolemic hypoosmolar patients are generally
treated initially by diuresis and other measures directed at
their underlying disorder. Such patients rarely require any
therapy to increase plasma osmolality acutely, but often
benefit from varying degrees of sodium and water restriction to reduce body fluid retention.
In any significantly hyponatremic patient one is faced
with the question of how quickly the plasma osmolality
should be corrected. Although hyponatremia is associated
with a broad spectrum of neurological symptoms, sometimes leading to death in severe cases, too rapid correction
of severe hyponatremia can produce pontine and extrapontine myelinolysis a brain demyelinating disease that also
can cause substantial neurological morbidity and mortality.
Recent clinical and experimental results suggest that optimal treatment of hyponatremic patients entails balancing
the risks of hyponatremia against the risks of correction
for each patient. Several factors should therefore be considered in making a treatment decision in hyponatremic patients: the severity of the hyponatremia, the duration of the
hyponatremia, and the patient's symptomatology. Neither
sequelae from hyponatremia itself nor myelinolysis after
therapy is very likely in patients whose serum [Na +] remains
>120mEq/L, although significant symptoms can develop
even at higher serum [Na +] levels if the rate of fall of
plasma osmolality has been very rapid. The importance of
duration and symptomatology relates to how well the brain
has volume -adapted to the hyponatremia, and consequently its degree of risk for subsequent demyelination
with rapid correction. Cases of acute hyponatremia (^48
hours in duration) are usually symptomatic if the hyponatremia is severe (i.e., 120 mEq/L). These patients are
at greatest risk from neurological complications from the
hyponatremia itself and should be corrected to higher
plasma [Na +] levels promptly. Conversely, patients with

Joseph G. Verbalis
more chronic hyponatremia >48 hours in duration) who
have minimal neurological symptomatology are at little
risk from complications of hyponatremia itself, but can
develop demyelination following rapid correction. There
is no indication to correct these patients rapidly, and they
should be treated using slower-acting therapies such as
fluid restriction.
Although the above extremes have clear treatment indications, most hyponatremic episodes will be of indeterminate duration and patients will have varying degrees of
milder neurological symptomatology. This group presents
the most challenging treatment decisions, since the hyponatremia will have been present sufficiently long to allow
some degree of brain volume regulation, but not enough
to prevent some brain edema and neurological symptomatology. Most recommend prompt treatment of such patients because of their symptoms, but using methods that
allow a controlled and limited correction of their hyponatremia. Reasonable correction parameters consist of a maximal rate of correction of plasma [Na + ] in the range of
1 to 2 mEq L" 1 rr1 as long as the total magnitude of
correction does not exceed 25 mEq/L over the first 48
hours. Some argue that these parameters should be even
more conservative with maximal correction rates of
<0.5 mEq L" 1 h"1 and magnitudes of correction that
do not exceed 12 mEq/L in any 24-hour period. Treatments
for individual patients should be chosen within these limits
depending on their symptomatology. In patients who are
only moderately symptomatic, one should proceed at the
lower recommended limits of 0.5 mEq L^ 1 Ir1 , while
in those who manifest more severe neurological symptoms
an initial correction at a rate of 1 to 2 mEq L" 1 Ir1 would
be more appropriate. Controlled corrections are generally
best accomplished with hypertonic (3%) NaCl solution
given via continuous infusion, because patients with euvolemic hypoosmolality such as SIAD generally will not respond to isotonic NaCl. An initial infusion rate can be
estimated by multiplying the patient's body weight, in kg,
by the desired rate of increase in plasma [Na +], in mEq
L'1 Ir1 (e.g., in a 70-kg patient an infusion of 3% NaCl
at 70 mL/h will increase plasma [Na +] by approximately
1 mEq L'1 Ir1 , while infusing 35 mL/h will increase
plasma [Na +] by approximately 0.5 mEq L'1 hr1 ). Furosemide (20 to 40 mg iv) should be used to treat volume
overload. It should be used preemptively in patients with
known cardiovascular disease. Patients with diureticinduced hyponatremia usually respond well to isotonic
NaCl and do not require 3% NaCl. Regardless of the initial
rate of correction chosen, acute treatment should be interrupted once any of three endpoints is reached: (1) the
patient's symptoms are abolished, (2) a safe plasma [Na +]
(generally a 120 mEq/L) is achieved, or (3) a total magnitude of correction of 20 mEq/L is achieved. It follows from
these recommendations that serum [Na +] levels must be
carefully monitored at frequent intervals (at least every 4
hours) during the active phases of treatment in order to
adjust therapy so that the correction stays within these
guidelines. Regardless of the therapy or rate initially chosen, it cannot be emphasized too strongly that it is only

7. Hyponatremia and Hypoosmolar Disorders

necessary to correct the plasma osmolality acutely to a safe

range rather than completely to normonatremia. In some
situations patients may spontaneously correct their hyponatremia via a water diuresis. When the hyponatremia is
acute (e.g., psychogenic polydipsia with water intoxication), such patients do not appear to be at risk for subsequent demyelination; however, in cases where the hyponatremia has been chronic (e.g., hypocortisolism)
intervention should be considered to limit the rate and
magnitude of correction of plasma [Na +] (e.g., administration of desmopressin 1 to 2 /j,g iv or infusion of hypotonic
fluids) using the same endpoints as for active correction.
Treatment of chronic hyponatremia entails choosing
among several suboptimal therapies. An important exception is patients with the reset osmostat syndrome; because
the hyponatremia of such patients is not progressive but
rather fluctuates around their reset level of plasma [Na +],
no therapy is generally required. For most other cases of
mild-to-moderate SIAD, fluid restriction represents the
least toxic therapy by far and is the treatment of choice.
This should always be tried as the initial therapy, with
pharmacologic intervention reserved for refractory cases
where the degree of fluid restriction required to avoid hypoosmolality is so severe that the patient is unable, or unwilling, to maintain it. If pharmacologic treatment is necessary,
the preferred drug at present is the tetracycline derivative
demeclocycline, which causes nephrogenic diabetes insipidus, thereby decreasing urine concentration. The effective
dose of demeclocycline ranges from 600 to 1200 mg/day;
several days of therapy are necessary to achieve maximum
effects, so one should wait 3 to 4 days before increasing the
dose. Demeclocycline can cause reversible nephrotoxicity,
especially in patients with cirrhosis; renal function should
b e monitored and the medication stopped if increasing
azotemia occurs. Several other drugs (diphenylhydantoin,
opiates, ethanol) can decrease AVP hypersecretion in selected cases, but responses are unpredictable. The current
unavailability of an ideal therapeutic agent for chronic
SIAD will likely change in the near future with development of specific AVP antidiuretic receptor antagonists.
Such agents will likely become the treatments of choice
for SIAD. However, their use to correct established hyponatremia will require judicious adherence to the same
guidelines already established for other therapies to prevent complications from brain demyelination.

63
Bibliography
Anderson RJ, Chung H-M, Kluge R et al: Hyponatremia: A prospective analysis of its epidemiology and the pathogenetic role
of vasopressin. Ann Intern Med 102:164-168, 1985.
Arieff AI, Llach F, Massry SG: Neurological manifestations and
morbidity of hyponatremia: Correlation with brain water and
electrolytes. Medicine 55:121-129, 1976.
Ayus JC, Wheeler JM, Arieff AI: Postoperative hyponatremic
encephalopathy in menstruant women. Ann Intern Med
117:891-897, 1992.
Bartter FC, Schwartz WB: The syndrome of inappropriate secretion of antidiuretic hormone. Am J Med 42:790-806, 1967.
Berl T: Treating hyponatremia: Damned if we do and damned if
we don't. Kidney Int 37:1006 -1018, 1990.
Robertson GL, Aycinena P, Zerbe RL: Neurogenic disorders of
osmoregulation. Am J Med 72:339-353, 1982.
Robertson GL: Posterior pituitary. In: Felig P, Baxter JD, Frohman
LA (eds) Endocrinology and Metabolism. McGraw-Hill, New
York, pp. 385-432, 1995.
Saito T, Ishikawa S, Abe K, et al: Acute aquaresis by the nonpeptide arginine vasopressin (AVP) antagonist OPC-31260 im proves hyponatremia in patients with syndrome of inappropriate
secretion of antidiuretic hormone (SIADH). J Clin Endocrinol
Metab 82:1054-1057, 1997.
Schrier RW: Pathogenesis of sodium and water retention in highoutput and low-output cardiac failure, nephrotic syndrome, cirrhosis and pregnancy. N Engl J Med 319:1065-1072; 11271134, 1988.
Sterns RH: Severe symptomatic hyponatremia: Treatment and
outcome. A study of 64 cases. Ann Intern Med 107:656-664,1987.
Sterns RH, Ocdol H, Schrier RW et al: Hyponatremia: Pathophysiology, diagnosis, and therapy. In: Narins RG (ed) Disorders of
Fluid and Electrolytes. McGraw-Hill, New York, pp. 583-615,
1994.
Verbalis IG: Hyponatremia: Epidemiology, pathophysiology, and
therapy. Curr Opin Nephrol Hypertens 2:636-652, 1993.
Verbalis JG: Inappropriate antidiuresis and other hypoosmolar
states. In: Becker KG (ed) Principles and Practice of Endocrinology and Metabolism. Lippincott, Philadelphia, pp. 265-276,
1995.
Verbalis JG: The syndrome of inappropriate antidiuretic hormone
secretion and other hypoosmolar disorders. In: Schrier RW,
Gottschalk CW (eds) Diseases of the Kidney. Little, Brown &
Co., Boston, pp. 2393-2427, 1996.
Zerbe R, Stropes L, Robertson G: Vasopressin function in the
syndrome of inappropriate antidiuresis. Anna Rev Med 31:315327, 1980.

HYPERNATREMIA
PAUL M. PALEVSKY

itary gland. In the kidney, arginine vasopressin modulates

the hydraulic permeability of the collecting duct. In the
absence of vasopressin, the collecting duct is relatively impermeant to water. Vasopressin exerts its effect on the
collecting duct through the activation of V2-vasopressin
receptors located on the basolateral aspect of the tubular
epithelium. The V2-receptor is coupled to adenylate cyclase
by GTP-binding proteins; receptor binding activates adenylate cyclase, which catalyzes the conversion of adenosine
triphosphate (ATP) to the second messenger, cyclic-AMP
(cAMP). Through incompletely elucidated mechanisms,
cAMP stimulates the insertion of the aquaporin-2 water
channel into the apical cell membrane. The resulting increase in the hydraulic permeability of the apical cell membrane permits the passive reabsorption of water from the
collecting duct into the isotonic cortical and hypertonic
medullary interstitium. The excretion of a concentrated
urine is therefore dependent on the generation and maintenance of the corticomedullary osmotic gradient as well as
the utilization of the gradient through the tubular response
to vasopressin secretion.
The osmotic regulation of vasopressin secretion is extremely sensitive. Below a body fluid osmolality of 280 to
285 mmol/kg, vasopressin secretion is inhibited and plasma
vasopressin levels are virtually undetectable. As body fluid
osmolality increases above this threshold, vasopressin secretion increases linearly with increases in body fluid osmo lality of as little as 1 to 2% resulting in detectable increases
in plasma vasopressin levels. The renal response to changes
in vasopressin secretion is also extremely sensitive. Urine is
maximally dilute when vasopressin secretion is suppressed;
urinary concentration increases linearly as plasma vasopressin levels rise in response to rising plasma tonicity, with
maximal urinary concentration achieved at vasopressin levels that correspond to a plasma osmolality of approximately
295 mmol/kg.
While renal water conservation is important in preventing further renal water losses, it is not sufficient for the
prevention of progressive hypertonicity or the restoration
of plasma tonicity to normal. The ultimate defense against
the development of hypertonicity and hypernatremia is the
osmotic stimulation of thirst and its resultant increase in
water ingestion. Thirst is also mediated by hypothalamic
osmoreceptors located in the anterior wall of the third
ventricle. These thirst osmoreceptors, although in proxim-

Hypernatremia is one of the two cardinal disturbances

of water homeostasis. Decreases in total body water relative to total body electrolyte are characterized by an increase in the electrolyte concentration in all body fluids.
In the intracellular compartment this is manifested by a
decrease in cell volume and an increase in the intracellular
potassium concentration. In the extracellular space, the
primary manifestation is an increase in the sodium concentration, resulting in the laboratory finding of hypernatremia.
Hypernatremia does not imply an abnormality in sodium homeostasis. Rather it is a disturbance in the sodium
concentration resulting from a deficit of water relative
to salt. Total body sodium content is the primary determinant of extracellular volume and in the setting of
intact water homeostasis does not have a primary effect on
the sodium concentration. Alterations in sodium balance
result in isotonic volume expansion or volume depletion
unless also accompanied by a defect in water homeostasis. Hypernatremia may occur, however, with normal,
increased or decreased total body sodium content,
manifest by corresponding alterations in extracellular
volume.
Hypernatremia is a common clinical problem, with a
prevalence in hospitalized patients of 0.5 to 2%. In adults,
two distinct groups of hypernatremic patients may be
identified. Patients developing hypernatremia prior to
hospital admission are generally elderly and debilitated
and often have an intercurrent acute infection. In contrast,
patients developing hypernatremia during hospitalization
have an age distribution similar to that of the general
hospital population. In these patients, hypernatremia is an
iatrogenic complication usually associated with impaired
thirst or restricted access to water combined with an
inadequate prescription for water administration.
REGULATION OF WATER HOMEOSTASIS

The physiologic response to hypertonicity includes both

renal water conservation and stimulation of thirst. Hypertonicity is sensed by osmoreceptors located adjacent to the
anterior wall of the third ventricle in the hypothalamus.
Activation of these osmoreceptors stimulates the secretion
of arginine vasopressin by neurons whose cell bodies are
in the supraoptic and paraventricular nuclei in the hypothalamus and whose axons terminate in the posterior pitu-

Primer on Kidney Diseases, Second Edition

64

8. Hypernatremia

65

ity to the osmoreceptors modulating vasopressin secretion,

are anatomically distinct. Impulses from these osmoreceptors are projected to higher levels in the cerebral cortex,
where they result in the perception of thirst and waterseeking behavior. The osmotic threshold for thirst is approximately 5 mmol/kg greater than the osmotic threshold
for vasopressin secretion; once this threshold is exceeded,
thirst increases proportionately as body fluid osmolality
rises.
PATHOPHYSIOLOGY

Hypernatremia results when there is net water loss or

hypertonic sodium gain. In the normal individual, thirst
is stimulated by any rise in body fluid tonicity. Water
ingestion increases and the hypernatremic state is rapidly
corrected. Therefore, sustained hypernatremia can occur
only when the thirst mechanism is impaired and water
intake does not increase in response to hypertonicity.
Although increased water losses or hypertonic sodium
gain usually contribute to the development of hypernatremia, abnormalities in the osmoregulation of thirst or
in thirst perception or the inability to gain access to
water are required for the development of sustained hypernatremia.
The importance of thirst in the pathogenesis of hypernatremia is illustrated by patients with severe diabetes insipidus. These patients are unable to concentrate their urine
and excrete large volumes of dilute urine, occasionally in
excess of 10 to 15 L/day. Under normal circumstances they
do not develop hypernatremiain response to their renal
water losses, thirst is stimulated and they maintain body
fluid tonicity at the expense of profound secondary polydipsia. If, however, they are unable to drink in response to
thirst, as may occur during an intercurrent illness, hypernatremia rapidly develops.
Isolated defects in thirst (primary hypodipsia) are uncommon but may result from any process involving the
hypothalamus in proximity to the anterior wall of the third
ventricle (Table 1). Lesions may include a wide range of
intrcranial pathology, including primary and metastatic

TABLE I

Defects in Thirst

Primary hypodipsia
Hypothalamic lesions affecting the osmostat

Trauma
Craniopharyngioma or other primary suprasellar tumor

Metastatic tumor
Oranulomatous disease
Vascular lesions

Essential hypernatremia
Geriatric hypodipsia
Secondary hypodipsia
Cerebrovascular disease
Dementia Delirium
Mental status changes

tumors, granulomatous diseases, vascular abnormalities

(most commonly involving the anterior communicating artery), trauma, and hydrocephalus. Hypodipsia has also
been described in elderly patients (geriatric hypodipsia) in
whom overt hypothalamic pathology is absent. Essential
hypernatremia is a rare disorder in which there is an upward
resetting of the thresholds for both thirst and vasopressin
secretion. Patients with this condition are able to concentrate and dilute their urine, albeit around an elevated body
fluid osmolality, and have an elevated set point for thirst
perception. More commonly, impaired thirst results from
diffuse neurologic disease which interferes with the less
well-defined cerebral cortical pathways subserving thirst
perception and water ingestion (secondary hypodipsia).
Thus, defects in thirst are commonly associated with cerebrovascular disease, dementia, and acute illnesses that result in delirium or obtundation.
CLINICAL CLASSIFICATION

Although impaired thirst and restricted water intake underlie the development of sustained hypernatremia, the
hypernatremic states are most commonly classified on the
basis of the associated water loss or electrolyte gain and
the corresponding changes in extracellular fluid volume
(Table 2). Pure water deficits are associated with minimal
change in total body sodium and relative preservation of
extracellular fluid volume. When hypotonic fluid deficits
are present, hypernatremia coexists with total body sodium

TABLE 1

Classification of Hypernatremia on the Basis of

Associated Changes in Extracellular Volume

Pure water deficit (normal extracellular volume)

Diabetes insipidus
Hypothalamic
Nephrogenic Increased
insensible losses
Hypotonic fluid deficit (decreased extracellular volume)
Renal losses
Diuretic administration
Osmotic diuresis
Postobstructive diuresis
Polyuric phase of ATM"
Gastrointestinal losses
Vomiting
Nasogastric drainage
Enterocutaneous fistulae
Diarrhea
Cutaneous losses
Burn injuries
Excessive perspiration
Hypertonic sodium gain (increased extracellular volume) Salt
ingestion Hypertonic NaCl Hypertonic NaHCO3 Total
parenteral nutrition
" ATN, acute tubular necrosis.

Paul M. Palevsky

66
depletion and extracellular fluid volume contraction. Hypertonic sodium gain results in hypernatremia and extracellular fluid volume expansion, (see Fig. 1)
Pure Water Deficits

Isolated water deficits, unless of extreme magnitude, are

generally not associated with clinical evidence of intravas cular volume depletion. Only one-third of a pure water
deficit is derived from the extracellular compartment, and
only one-twelfth from the intravascular compartment. In
a 70-kg individual, a 5% decrease in total body water, which
would result in an increase in the plasma sodium concentra tion of approximately 7 mmol/L, results in a reduction in
ECF volume of less than 700 mL and intravascular volume
of less than 200 mL. This modest degree of volume depletion is generally not detectable on physical examination
and may be manifest only by mild prerenal azotemia. With
more severe water deficits, hemodynamically significant
intravascular volume depletion may develop.
Pure water deficits may occur with impaired thirst unaccompanied by increased water losses. In patients with sig nificant hypodipsia, voluntary water ingestion may not be
sufficient to replace obligate gastrointestinal and insensible

water losses. Despite maximal renal water conservation,

progressive hypernatremia will result if supplemental water
intake is not provided. More commonly, pure water deficits
develop in the setting of increased insensible or renal
water losses. Insensible water losses are approximately
0.6 mL kg"1 h"1 , or about 1 L/day in the average adult.
These losses are not subject to osmotic regulation but may
be increased by a wide variety of factors including fever,
exercise, increased ambient temperature, and hyperventilation. Increased renal electrolyte-free water losses result
from diabetes insipidus. Patients with increased insensible
losses manifest oliguria with a maximally concentrated
urine (urine osmo lality >700 mmol/kg). In contrast, patients with diabetes insipidus have polyuria and a less than
maximally concentrated urine.
Diabetes Insipidus

Diabetes insipidus results either from failure of the

hypothalamic-pituitary axis to synthesize or release adequate amounts of vasopressin (hypothalamic diabetes insipidus) or from failure of the kidney to produce a concentrated urine despite appropriate circulating vasopressin
levels (nephrogenic diabetes insipidus). In both forms of
the disorder, the inability of the kidney to concentrate the

<20

mmol/L

Isolated
Hypodipsia
(Table 1)

Increased
Insensible Losses
(Table 2)

700 mmol/kg

>40

mmol/L
Hypothalamic
Diabetes Insipidus
(TableS)

Response

Nephrogenic
Diabetes Insipidus
(Table 4)
FIGURE I

Algorithm for the evaluation of hypernatremia.

67

8. Hypernatremia

urine leads to polyuria and secondary polydipsia. If water

intake is adequate to replace urinary electrolyte-free water
losses, hypertonicity and hypernatremia will not develop.
Thus, hypernatremia is not a hallmark of diabetes insipidus.
Diabetes insipidus may occur in either complete or partial forms. In complete hypothalamic diabetes insipidus,
vasopressin secretion is absent, resulting in the production of large volumes of dilute urine (urine osmolality
<150 mmol/L). In partial hypothalamic diabetes insipidus,
vasopressin secretion is detectable but subnormal, and a
less severe defect in renal water conservation is present.
Similarly, in complete nephrogenic diabetes insipidus, renal
responsiveness to the hydroosmotic effect of vasopressin
is absent, whereas an impaired response occurs in partial
nephrogenic diabetes insipidus.
Any pathologic process involving the hypothalamicpituitary axis may lead to vasopressin deficiency and hypothalamic diabetes insipidus (Table 3). Common etiologies
include pituitary surgery, head trauma, primary and metastatic tumors, leukemia, hemorrhage, thrombosis, and granulomatous diseases. Diabetes insipidus following head
trauma or surgery may be transient, may be permanent,
or may follow a triphasic course. The transient form is most
common, with an abrupt onset followed by resolution over
a period of days to weeks. In the triphasic pattern, there
is an initial period of vasopressin deficiency lasting 2 to 4
days as the result of axonal injury, a 5- to 7-day period of
inappropriate vasopressin release thought to result from
release of hormone by degenerating neurons, and finally
permanent diabetes insipidus following neuronal death.
In 30% of patients, hypothalamic diabetes insipidus is
idiopathic, most probably occurring on an autoimmune

TABLE 3

Hypothalamic Diabetes Insipidus

Pituitary surgery
Head trauma
Neoplasia Primary: Dysgerminoma, craniopharyngioma,
suprasellar
pituitary tumors
Metastatic: Carcinomas of the breast and lumphoma
Leukemia
Vascular lesions
Aneurysms
Cerebrovascular accidents Sheehan's syndrome
(postpartum pituitary hemorrhage)
Infections
Encephalitis
Meningitis
Tuberculosis
Syphilis
Granulomatous disease
Sarcoidosis
Histiocytosis
Autoimmune
Vasopressin -neurophysin gene mutations

basis. A rare, hereditary form is due to mutations in the

vasopressin -neurophysin gene which result in an abnormal
structure and processing of the vasopressin prohormone
and ultimately cell death of the vasopressin-secreting
neurons.
The term nephrogenic diabetes insipidus should be restricted to those situations in which there is an intrinsic
abnormality of the collecting duct that leads to vasopressin
insensitivity or hyporesponsiveness. Patients with chronic
tubulointerstitial renal disease may be unable to generate
or maintain a normal corticomedullary osmotic gradient
and are therefore unable to concentrate their urine normally. However, they rarely have significant polyuria and
should not be considered to have nephrogenic diabetes insipidus.
Nephrogenic diabetes insipidus may be either acquired
or congenital in origin (Table 4). The acquired form is
far more common than the congenital and is most often
associated with pharmacologic therapy with lithium or
demeclocycline. Both agents have been demonstrated to
inhibit intracellular generation of cAMP in response to
vasopressin. In addition, demeclocycline also inhibits the
intracellular action of cAMP. Acquired nephrogenic diabetes insipidus may also result from obstructive uropathy,
hypercalcemia, or severe hypokalemia.
The congenital form of nephrogenic diabetes insipidus
has been identified in multiple kindreds and is usually inherited in a sex-linked pattern with variable penetrance in
hemizygous females. The genetic defect has been localized
in the majority of kindreds to the vasopressin V2 -receptor
gene; however, mutations in the aquaporin-2 gene have
been identified in rare patients with an autosomal recessive
form of nephrogenic diabetes insipidus.
The polyuria of diabetes insipidus needs to be differentiated from other forms of polyuria, including primary polydipsia, solute diuresis secondary to glycosuria, mannitol,
urea, or diuretics, or resolving acute renal failure. Polyuria
due to solute diuresis can usually be excluded by demonstrating that the urine osmolality is less than 150 mmol/kg.
In patients with severe polyuria, formal dehydration testing
is usually unnecessary and may result in severe hypernatremia and hypotension. Dehydration testing may be re-

TABLE 4

Nephrogenic Diabetes Insipidus

Drug induced
Lithium
Demeclocycline
Methoxyflurane
Amphotericin B
Electrolyte Disorders
Hypercalcemia
Hypokalemia
Obstructive uropathy
Congenital
Vasopressin V2 -receptor mutations
Aquaporin -2 mutat ions

68

Paul M. Palevsky

quired, however, to diagnose partial forms of diabetes insipidus.

During a dehydration test, the patient is placed on a
strict fast with special care taken to ensure that the patient
consumes no fluids. During the test, urine osmolality is
measured hourly and plasma osmolality every 4 to 6 hours.
Water deprivation is continued until body weight has declined by 3%, plasma osmolality has reached 295 mmol/kg,
or urine osmolality has reached a plateau (variation of less
than 5% over 3 hours). In patients with severe diabetes
insipidus, urine osmolality will remain less than plasma
osmolality, whereas in partial diabetes insipidus urine osmolality will be greater than plasma osmolality although
submaximally concentrated (Table 5). The urinary response to exogenous vasopressin usually differentiates the
hypothalamic form from the nephrogenic form. Measurement of plasma vasopressin levels at maximal dehydration
(prior to exogenous vasopressin), although not routinely
available, may be extremely useful in equivocal cases.
Hypotonic Fluid Deficits
Patients with inadequately replaced hypotonic fluid
losses will develop both hypernalremia and extracellular
volume depletion. Unlike patients with pure water losses,
these individuals manifest the classic findings of intravascular volume depletion, i.e., tachycardia, hypotension, and
decreased central venous pressure. Hypotonic fluid may
be lost from the skin, gastrointestinal tract, or kidney. Cutaneous losses of electrolyte-containing hypotonic fluids may
be significant in patients with severe burn injuries and patients with increased sensible perspiration (as opposed to
insensible transpirational skin loss). The majority of gastrointestinal fluids, with the exception of pancreatic and biliary
secretions, are also hypotonic. Protracted vomiting, nasogastric drainage, and diarrhea commonly contribute to the
development of hypovolemic hypernatremia. Excessive renal hypotonic fluid losses are most commonly due to diuretic therapy but are also associated with osmotic diureses
due to glucose, mannitol, or urea. Postobstructive diuresis,
the polyuric phase of acute tubular necrosis, adrenal insufficiency, and a variety of chronic renal diseases, especially
medullary cystic disease and renal tubular acidosis (types
I and II), are also associated with renal-salt wasting and
hypotonic urinary losses. Hypovolemic hypernatremia may

also develop in patients with "third-space" fluid losses (e.g.,

bowel obstruction, pancreatitis, peritonitis) if the seques tration of isotonic fluid is combined with inadequate replacement of ongoing electrolyte-free water losses.
Urinary indices are helpful in ascertaining the source
of hypotonic fluid losses. When the losses have a renal origin, the urine sodium concentration is usually elevated
(>20 mmol/L) and the urine is less than maximally concentrated. In contrast, nonrenal losses are associated with renal
sodium avidity (urine sodium <10 mmol/L) and a maximally concentrated urine (urine osmolality >700 mmol/kg).
Hypertonic Sodium Gain
Hypertonic sodium gain produces hypernatremia and
extracellular volume overload. When both thirst and renal
function are intact, the volume and tonicity disturbances
are transient. The increase in tonicity stimulates thirst, with
a resultant increase in water intake that corrects the hypertonicity. In addition, the volume expansion stimulates a
natruresis. Persistent hypernatremia implies impaired thirst
or restricted water intake. Hypertonic sodium gain may
result from the accidental ingestion of large quantities of
sodium salts but is more commonly iatrogenic, resulting
from the administration of hypertonic sodium chloride or
sodium bicarbonate solutions, inappropriate electrolyte
prescription for parenteral hyperalimentation solutions, or
inappropriate sodium supplementation of enteral nutrition.
Errors in formula preparation may result in hypernatremia
in infants.
Hypertonicity Due to Nonelectrolyte Solutes
Although all hypernatremia results in body fluid hypertonicity, hypertonicity can also occur in the absence of hypernatremia. The accumulation of osmotically active nonelectrolyte solutes in the extracellular compartment may result
in the development of hypertonicity accompanied by a
normal, or even depressed, serum sodium concentration.
The accumulation of a nonelectrolyte solute increases the
osmolality of the extracellular fluids. If the solute cannot
enter the intracellular compartment, water will exit the
intracellular space. The net result is intracellular dehydration, cell shrinkage, expansion of the extracellular compartment, and dilution of the extracellular fluid sodium concen-

TABLE S
Diagnosis of Diabetes Insipidus

Urine osmolality
Disorder

After dehydration

After exogenous vasopressin

Plasma vasopressin level

Hypothalamic DI
Complete Partial

<300 mmol/kg
>300 mmol/kg

>50% increase
>10% increase

<1.0 pg/mL
<1.5 pg/mL

Nephrogenic DI
Complete Partial

<300 mmol/kg
>300 mmol/kg

<50% increase
<10% increase

>5.0 pg/mL
>2.0 pg/mL

8. Hypernatremi a

69

tration. It is important to recognize nonhypernatremic

hypertonicity in order to avoid institution of inappropriate
therapy for the resultant dilutional hyponatremia.
The most common nonelectrolyte solute associated causing hypertonicity is glucose. In patients with hyperglycemia, the serum sodium concentration usually declines by
1.6-2.0 mmol/L for each 100 mg/dL increase in plasma
glucose above normal. Isonatremic and hyponatremic hypertonicity have also been described following administration of mannitol, maltose, sorbitol, glycerol, and radiocontrast agents. If suspected, the presence of an unmeasured
solute may be confirmed by comparison of the measured
plasma osmolality with an estimate of plasma osmolality
derived from measurement of serum sodium, glucose, and
urea nitrogen concentrations:
^osm = (2 X [Na + ]) + glucose (mg/dL)/18 +
BUN (mg/dL)/2.8.
The presence of an "osmolar gap" between these two values for plasma osmolality suggests the presence of an unmeasured solute. The responsible solute can usually be
ascertained by review of medications administered and may
be confirmed by specific biochemical testing.

Hypertonicity

H,O
Electrolytes

Rapid Adaptation

Osmolytes H,0
Chronic Adaptation
FIGURE 2 Brain
adaptation
to
hypernatremia. Brain
cell
volume is indicated
b y the
size of the oval;
intracellular osmolality by the density of
shading. Following the acute onset of hypernatremia, there is a
rapid loss of water from the intracellular compartment resulting
in a decrease in brain cell volume and an increase in brain cell
osmolality. Adaptive processes are then activated which restore
brain cell volume to normal. An initial phase of rapid adaptation,
occurring over the first 24 hours, consists of electrolyte uptake into
brain cells, partially restoring brain volume. In chronic adaptation,
which is complete by Day 7, organic solutes accumulate within
brain cells leading to restoration of brain volume to normal
although intracellular osmolality remains elevated.

CLINICAL MANIFESTATIONS

The major clinical manifestations of hypernatremia

result from alterations in brain water content. In response
to hypertonicity, fluid shifts from the intracellular compartment into the extracellular compartment, maintaining
osmotic equilibrium and resulting in a decrease in intracellular volume. In the central nervous system, acute
hypernatremia is associated with a rapid decrease in
water content and brain volume (Fig. 2). Within 24
hours, however, adaptive processes result in the uptake
of electrolyte into brain cells with a partial restoration
of brain volume. Subsequently, there is an increase in
intracellular organic solute content, primarily through the
accumulation of amino acids, polyols, and methylamines,
which restores brain volume to normal. The accumulation
of these intracellular solutes ("idiogenic osmoles") has
important therapeutic implications. Although they minimize cerebral dehydration during hypertonicity, their accumulation increases the risk of cerebral edema during
the treatment of hypernatremia.
The clinical manifestations of hypernatremia reflect its
rapidity of onset, its duration, and its magnitude. In
severe acute hypernatremia, brain shrinkage may be
substantial, placing traction on the venous sinuses and
intracerebral veins and causing their rupture. The resulting intracerebral and subarachnoid hemorrhage may produce irreversible neurologic defects or death. The manifestations of less profound hypernatremia are nonspecific
and include nausea, muscle weakness, and fasciculations,
and alterations in mental status ranging from lethargy
to coma. Seizures, while uncommon in chronic hypernatremia, may develop after initiation of therapy in as
many as 40% of patients.

The mortality rate associated with hypernatremia has

been reported to range from 40% to more than 70%, depending on the magnitude of hypernatremia and its rapidity
of onset. The majority of deaths, however, are not a direct
consequence of the hypernatremia but result from underlying illnesses. In a recent study, the mortality in 103 consecutive hypernatremic patients was 41%; however, in only 16%
of the patients did the hypernatremia contribute to the
cause of death. When mortality was analyzed on the basis
of adequacy of therapy, hypernatremia contributed to mortality in 25% of patients in whom it persisted for more than
72 hours compared to only 8% of patients in whom the
hypernatremia was promptly treated and resolved within
72 hours.
TREATMENT

The treatment of hypernatremia is water. The existing water deficit should be repleted and any ongoing
electrolyte-free water losses replaced. The water deficit
may be estimated based on the current serum sodium concentration (Sfla) and body weight, using the assumption
that total body water is approximately 60% of body weight:
Water Deficit = 0.6 X [Body Weight (kg)] X
[(Na + ]/140) - 1].
Despite inaccuracies inherent in this formula, the calculation provides a useful approximation for initiating water re placement.

70

In acute hypernatremia, repletion of the water deficit

may be rapid. The electrolytes that accumulate in the
brain during acute hypernatremia are rapidly extruded
into the extracellular compartment during treatment, minimizing the risk of cerebral edema. In contrast, overly
rapid therapy of chronic hypernatremia may produce
cerebral edema if the water replacement occurs more
rapidly than the brain can dissipate the accumulated
organic solutes, a process requiring approximately 24 to
48 hours. Well-controlled studies to ascertain the optimal
treatment of chronic hypernatremia do not exist. However, based on knowledge of the rate of solute loss by
the brain following chronic hypernatremia, prompt but
gradual correction is most prudent. In symptomatic patients, rapid water replacement should be provided initially; however, the serum sodium concentration should
be reduced by no more than 1 to 2 mmol-L~1-h~1. Once
symptoms have resolved, replacement of the remainder
of the water deficit should occur over 24 to 48 hours.
Throughout treatment, the patient's neurologic status
should be monitored carefully; deterioration after an
initial improvement in neurologic symptoms suggests the
development of cerebral edema and mandates temporary
discontinuation of water replacement.
No individual regimen of water replacement is of documented superiority. Water may be administered enterally,
either orally or by nasogastric tube, or intravenously. Intravenous repletion can consist of either hypotonic saline or
5% dextrose in water; pure water cannot be administered
intravenously as the local hypotonicity at the site of administration can produce severe intravascular hemolysis. When
glucose containing solutions are used, the blood glucose
concentration should be monitored carefully and insulin
therapy initiated, if necessary, to forestall hyperglycemia.
In addition to replacing the calculated water deficit, ongoing fluid losses must be replaced. Urinary, gastrointestinal, and other losses should be quantified and should be
replaced on the basis of their volume and electrolyte content. Insensible losses must also be replaced, recognizing
that they increase by approximately 20% for each 1C increase in body temperature.
Prompt attention to reducing excessive water losses is
also important. Insensible losses may be reduced by normalizing body temperature with cooling blankets and antipyretics. Hyperglycemia should be controlled and protein
loading decreased in order to limit osmotic diuresis. Nasogastric drainage can be reduced by therapy with H2receptor antagonists or proton pump inhibitors. Diarrhea
can be reduced by altering enteral feeding, treating infectious causes, discontinuing cathartic agents, or administering antidiarrheal agents. Specific therapy for diabetes insipidus should be initiated, when appropriate.
Hypothalamic diabetes insipidus is readily treated with
hormone replacement. Although the native hormone,
aqueous arginine vasopressin, is available, treatment is
most often with desamino-8-D-arginine vasopressin
(dDAVP), a synthetic analog of vasopressin with a longer
half-life and less vasoconstrictive effects. Hormone replacement therapy is generally ineffective in nephrogenic

Paul M. Palevsky

diabetes insipidus. Restriction of dietary protein and

sodium intake may attenuate the polyuria by reducing
obligate urinary solute excretion. Thiazide diuretics may
also be of benefit. Thiazides directly inhibit urinary diluting capacity and indirectly decrease distal tubular delivery
by inducing mild intravascular volume contraction which
increases proximal tubular sodium reabsorption. The net
effect is to decrease delivery of free water to the collecting
duct and to enhance vasopressin-independent water reabsorption, thereby moderating the polyuria. Nonsteroidal
anti-inflammatory drugs are also useful as adjunctive
therapy. Amiloride is useful in the treatment of lithiuminduced nephrogenic diabetes insipidus. It is postulated
that in addition to its diuretic effects amiloride blocks
lithium entry into collecting duct cells, thereby reducing
cellular toxicity.
Initial treatment of the patient with coexistent hypernatremia and extracellular volume depletion should be directed at restoring intravascular volume. Frank circulatory
compromise should be promptly treated with isotonic saline
or colloid solutions. Once adequate volume replacement
has been achieved, treatment should then be directed toward replacing the water deficit. Treatment of hypernatremia in patients with volume overload generally requires
both water repletion and solute removal. Because hypernatremia often develops rapidly in these patients, the compensatory mechanisms defending brain volume are ineffective
and neurologic symptoms may be accentuated. In addition,
the concomitant intravascular volume expansion may result
in pulmonary edema and exacerbate respiratory failure.
Treatment must therefore be promptly instituted in order
to prevent neurologic and cardiopulmonary complications.
Because water repletion will further exacerbate intravascular volume overload, a loop diuretic should be simultaneously administered to facilitate solute excretion. In patients
with massive volume overload or renal failure, initiation
of hemodialysis or hemofiltration may be necessary.
Bibliography
Ayus JC, Armstrong DL, Arieff AI: Effects of hypernatremia in
the central nervous system and its therapy in rats and rabbits.
J Physiol 492:243-255, 1996.
DeRubertis FR, Michelis MF, Beck N, Field JB, Davis BB: "Essential" hypernatremia due to ineffective osmotic and intact volume
regulation of vasopressin secretion. / Clin Invest 50:97-110,
1971.
Fitzsimons JT: Physiology and pathophysiology of thirst and so dium appetite. In: The Kidney: Physiology and Pathophysiology.
2nd ed. Raven Press, New York, pp. 1615-1648, 1992.
Gullans SR, Verbalis JG: Control of brain volume during hyperosmolar and hypoosmolar conditions. Annu Rev Med 44:289301, 1993.
Holtzman EJ, Ausiello DA: Nephrogenic diabetes insipidus:
causes revealed. Hasp Pract 29:67-82, 1994.
Lien YH, Shapiro JI, Chan L: Effects of hypernatremia on organic
brain osmoles. / Clin Invest 85:1427-1435, 1990.
Miller M, Dalakos T, Moses AM, et al: Recognition of partial
defects in antidiuretic hormone secretion. Ann Intern Med
73:721-729, 1970.
Palevsky PM, Bhagrath R, Greenberg A: Hypernatremia in hospitalized patients. Ann Intern Med 124:197-203, 1996.

9. Metabolic Acidosis

71

Phillips PA, Rolls BJ, Ledingham JGG, et al: Reduced thirst after
water deprivation in healthy elderly men. NEnglJ Med 311:753759, 1984.
Robertson GL: Regulation of vasopressin secretion. In: Seldin
DW, Giebisch G (eds) The Kidney: Physiology and Pathophysiology. 2nd ed. Raven Press, New York, pp. 1595-1613, 1992.
Ross EJ, Christie SBM: Hypernatremia. Medicine 48:441-478,
1969.

Seckl JR, Dunger DB: Diabetes insipidus: current treatment recommendations. Drugs 44:216-224, 1992.
Snyder NA, Feigal DW, Arieff AI: Hypernatremia in elderly patients. Ann Intern Med 107:309-319, 1987.
Zerbe RL, Robertson GL: A comparison of plasma vasopressin
measurements with a standard indirect test in the dif ferential diagnosis of polyuria. N Engl J Med 305:1539-1546,
1981.

METABOLIC ACIDOSIS
DANIEL BATLLE

Metabolic acidosis is an acid-base disorder characterized

by a fall in blood bicarbonate concentration and a fall in
blood pH (i.e., a rise in the hydrogen ion concentration
[H+]). Respiratory compensation provides for a predictable decrease of the blood CC>2 tension (pCO2). For each
1 mEq decrease in bicarbonate concentration the pCO2
falls by 1.0 to 1.5 mm Hg. The importance of respiratory
compensation on blood pH is illustrated with an example
(Fig. 1). The presence of an inappropriate response suggests the existence of a mixed acid-base disturbance. Severe metabolic acidosis, unlike respiratory alkalosis, cannot
be fully compensated despite maximal hyperventilation.
Accordingly, if the blood pH is not reduced when bicarbonate is very low, a respiratory alkalosis must be present.
Metabolic acidosis occurs when bicarbonate is lost from
the body (via the gastrointestinal tract or the kidney), when
the kidneys fail to regenerate bicarbonate via adequate
acid excretion, or when bicarbonate is consumed in the
titration of excessive acid produced endogenously or from
the ingestion of acid-producing compounds.
THE PLASMA ANION GAP IN THE INITIAL EVALUATION
OF METABOLIC ACIDOSIS

The plasma anion gap (AG) is useful in the evaluation

of the type of metabolic acidosis and the initial classification
of its various causes. The AG is simply a calculation that
allows the clinican to infer whether there has been a change
in unmeasured anions or cations in plasma.
The use of the AG takes advantage of the principle of
electroneutrality, which dictates that the number of posi-

Primer on Kidney Diseases, Second Edition

tively charged particles in any solution must be equal to

the number of negatively charged particles. The major unmeasured anions in plasma include albumin, phosphate,
sulfate, and other organic anions. The major unmeasured
cations in plasma include calcium, magnesium, and other
less-abundant cations. That the AG actually reflects the
difference between unmeasured anions (UA) and the unmeasured cations (UC) can be easily appreciated by examining the following basic equations: Since the sum of all
cations = the sum of all anions,
(Na + + K + ) + UC = (Cl- + HCCV) + UA
(Na + + K + ) - (Cr - HC(V) = UA - UC.
Accordingly, the AG reduces to
AG = UA - UC
and
AG = ([Na + ] + [K+ ]) - ([Cr] + [HCCV])
(normally 14 2 mEq/L).
Due to the small magnitude of changes in potassium concentration in the plasma, potassium is often omitted from
the calculation of the AG. Thus,
AG = [Na + ] - ([Cr + [HCCV])
(normally 10 2 mEq/L).
The AG provides a convenient tool to estimate whether
there has been a change in either UA or UC even before
such a change can be documented by direct measurements.
For instance, in metabolic acidosis due to the addition of
non-chloride-containing acid (e.g., lactic acid) the added
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

Daniel Batlle

72

Respiratory fatigue
pH

6.81

Initial
pH

Progressive acidosis

7.11

PCO2 32

PCO2 16

HCOg 5

HCOg

pH

6.81

PCO2 16
HCOg 2.5

Alkali therapy
pH

7.29

PCO2 16
HCO3 7.5
FIGURE I The impact on pH of minimal changes in either pCO2 or
HCO3 when plasma bicarbonate is very low is illustrated.

protons buffer the bicarbonate while the retained anion

(lactate) increases the unmeasured anions. The increase
in the AG from lactate will match the fall in plasma bicarbonate. In contrast, when a chloride-containing acid is
added to the blood, the hydrogen ion titrates bicarbonate
while exogenous chloride ion is largely retained by the
kidneys. This results in a rise in the chloride concentration
equivalent to the fall in bicarbonate; the AG will not
change. This is described as hyperchloremic or normal AG
metabolic acidosis. When the decrement in plasma bicarbonate is not matched by an equivalent increment either
in plasma chloride or in the plasma AG, a mixed hyperchloremic and high AG metabolic acidosis is present. For
instance, a mixed hyperchloremic/high AG metabolic acidosis can develop in a setting where the metabolic acidosis
originates from two different mechanisms (e.g., diarrhea
in patient with preexisting renal failure or shock).
Because the AG reflects changes in UA or UC, certain
situations other than metabolic acidosis can alter the
plasma AG. The level of plasma albumin influences the
plasma AG and should be taken into consideration. For
every fall of albumin of 10 g/L, the AG drops by about 4
mEq/L. An abnormally high concentration of unmeasured
cations, as seen with the accumulation of abnormal cationic
paraproteins in patients with IgG myeloma, can depress
the AG. These interpretative pitfalls in the use of the
plasma AG stress the need to pay careful attention to
all the laboratory variables and to look for appropriate
clinical correlates.
HYPERCHLOREMIC METABOLIC ACIDOSIS (OR
NORMAL AG TYPE OF METABOLIC ACIDOSIS)

Hyperchloremic metabolic acidosis should be distinguished from chronic respiratory alkalosis. In either acidbase disorder plasma chloride is elevated and plasma bicarbonate is reduced. An arterial blood gas is needed to
properly diagnose each acid-base disorder.

With the administration of HC1 or other HCl-generating

compounds a hyperchloremic metabolic acidosis develops
as blood bicarbonate is titrated, while plasma chloride increases proportionally by the retention of exogenous chloride. One basic alteration underlying the generation of
hyperchloremic metabolic acidosis is the loss of bicarbonate, usually in the urine or in the stools (Table 1). In individuals with metabolic acidosis associated with chronic diarrhea due to protracted laxative abuse, a chronic volume
deficit ensues despite avid renal sodium retention. The
virtual absence of urine sodium in some of these patients
may impede a normal distal acidification response to the
prevailing acidemia. This is because distal sodium delivery
is necessary for optimal distal hydrogen ion secretion.
When sodium excretion is increased to normal by salt replacement, urine pH falls and acid excretion increases resulting in amelioration of the metabolic acidosis.
Diversion of urine through intestinal segments can result
in hyperchloremic metabolic acidosis, hypokalemia, and
other electrolyte abnormalities, e.g., hypomagnesemia and
hypocalcemia. Various intestinal segments have been used
as conduits to substitute for the bladder. The anastomosis
of a ureter into the sigmoid colon (ureterosigmoidostomy)
almost always results in hyperchloremic metabolic acidosis.
Consequently this technique was supplanted by the ileal
conduit. Hyperchloremic metabolic acidosis, however, is
still a problem in about 10% of patients with an ileal conduit, particularly those where the ileal conduit is obstructed.
RENAL CAUSES OF HYPERCHLOREMIC
METABOLIC ACIDOSIS

Ingestion of an average protein-containing diet generates

a net fixed acid excess of approximately 1 mEq/kg body
wt daily during the course of metabolism. Addition of acid
(HA) to the blood compartments results in the titration of
the bicarbonate anion according to the equation

TABLE 1

Causes of HypercWoremic (or Normal AG)

Metabolic Acidosis
Administration of chloride -containing acid
NH4 C1, HC1 Hyperalimentation
Cholestyramine
Bicarbonate wastage
GI tract (diarrhea, ileus, fistula, villous adenoma) Urinary
tract diversions to intestine (ureterosigmoidostomy, ileal
conduit)
Impaired renal H+ secretion and reduced NH4 excretion Distal
RTA (hypokalemic and some hyperkalemic types)
Posthypocapnia (transient)
Impaired NH3 formation and reduced NH4 excretion
Advanced renal insufficiency (GFR < 20 mL/min)
Hyperkalemia Aldosterone deficiency

9. Metabolic Acidosis

HA + Na HCO 3 ^ H2 O + CO 2 + Na A.
In addition to reclaiming all filtered bicarbonate, the
kidney must excrete the acid anion and regenerate the
bicarbonate that was consumed in the initial titration by
the acid. Regeneration of bicarbonate is largely accomplished by hydrogen ion secretion in the distal nephron.
Hydrogen ions secreted by the renal tubules are titrated
by urinary buffers, primarily ammonia. Due to its favorable
pK (6.8) and relatively high concentration in the urine,
phosphate is also a major urinary buffer and contributes
greatly to titratable acid excretion. Increased ammonium
excretion is, by far, the major mechanism by which the
kidneys can regenerate bicarbonate. Net renal acid excretion (NAE) is calculated as urinary ammonium excretion
(NH4) plus urinary titratable acid (TA) minus bicarbonate
(HCO3~) or other potential bases (PB) that are excreted
in the urine:
NAE ^ (NH 4 + TA) - (HCO 3 + PB).
The kidneys may fail to provide for an adequate excretion of acid because of either decreased acid excretion
(NH4 , TA, or both) or increased alkali excretion (i.e.,
HCO3~ wastage). The former mechanism underlies an
array of syndromes collectively referred to as distal renal
tubular acidosis. The latter mechanism accounts for the
development of proximal renal tubular acidosis. With advanced renal failure, NH4+ excretion is also reduced, resulting in metabolic acidosis which is usually mixed (i.e., hyperchloremic and high AG).

73
OVERVIEW OF DEFECTS IN COLLECTING TUBULE
ACIDIFICATION CAUSING THE VARIOUS DISTAL
RTA SUBTYPES

The collecting tubule is the major site of urinary acidification within the distal nephron. This nephron segment
displays axial heterogeneity both anatomically and functionally. In cortical collecting tubules, active sodium reabsorption generates an electrical potential (lumen-negative)
that favors the secretion of H+ and potassium. In contrast,
in outer medullary collecting tubules, H+ secretion does
not appear to be under the influence of sodium transport
and is not accompanied by potassium secretion. The collecting tubule contains a proton pump at its luminal surface.
This pump, a proton-translocating ATPase, secretes H+ in
an electrogenic manner and can operate independently of
sodium transport and despite an unfavorable transtubular
electrical gradient. Alterations causing an impairment of
collecting tubule acidification by primarily interfering with
active H+ secretion are referred to as secretory types of
distal RTA (Table 2). Alterations in transepithelial voltage
caused by either inhibition of sodium transport or enhancement of lumen to cell chloride transport can also reduce
the rate of H+ secretion, albeit indirectly, and are referred
to as voltage-dependent types of distal RTA. A primary
H+ secretory defect, by definition, is limited to a defect in
either the H,K-ATPase pump or the H,K-ATPase pump
(Table 2).
Most patients with distal RTA appear to have a secretory
defect due to H-ATPase failure. Hypokalemia can develop

PROXIMAL RENAL TUBULAR ACIDOSIS

TABLE I

In normal individuals, the urine bicarbonate concentration is usually very low (less than 1 mEq/24 h). In patients
with proximal renal tubular acidosis (type II RTA), a large
fraction of the filtered load of bicarbonate is excreted when
the concentration of bicarbonate in plasma is normal or
even when it is moderately reduced. In contrast, the urine
of such patients is virtually bicarbonate-free when the concentration of bicarbonate in plasma falls below a critical
level referred to as the renal threshold. The renal bicarbonate threshold in patients with proximal RTA varies between 15 and 20 mEq/L (normal about 24 mEq/L). When
plasma bicarbonate is below the renal bicarbonate threshold, urine pH falls to levels almost as low as those seen in
normal subjects (less than 5.5). This feature is helpful in
distinguishing proximal RTA from some types of distal
RTA where urine pH is higher than 5.5 despite severe
acidemia (type I RTA and some cases of hyperkalemic
RTA; see below).
In both children and adults, bicarbonate wastage may
occur as a part of the Fanconi syndrome, a generalized
defect in proximal tubular transport that results in inhibition of reabsorption of glucose, phosphate, uric acid, and
amino acids. In this setting, glycosuria develops at normal
plasma glucose concentrations, and reduced plasma phosphate and plasma uric acid levels as well as aminoaciduria
may be observed.

Classification of DRTA
Type

Example
+

Permeability defects H
backleak Enhanced HCO5
secretion
Secretory defects
Diffuse collecting tubule
H-ATPase defect
Medullary collecting tubule
H-ATPase defect Diffuse
H-ATPase
secretory defect Diffuse
collecting H,KATPase defect
Rate-dependent defects
Impaired Na + transport
(Na+ channel defect)
Enhanced CI transport
Aldosterone deficiency
Aldosterone resistance
Reduced urinary buffers
Increased intracellular pH

Amphotericin B
Unknown
Chronic kidney transplant
rejection Nephrocalcinosis
(some cases)
Obstructive nephropathy
Endemic hypokalemic DRTA
(?)
Amiloride, trimethoprim
Pseudohypoaldosteronism
type II Selective
aldosterone
deficiency
Pseudohypoaldosteronism
type I
Hyperkalemia Cytosolic
carbonic anhydrase
II deficiency

Daniel Batlle

74

in such patients as a consequence of secondary hyperaldosteronism and accelerated potassium secretion in the distal
nephron. Theoretically, a defect in the H,K-ATPase pump
could better explain the development of severe hypokalemia in some of these cases but evidence that such a defect
causes distal RTA in humans is so far lacking. As both
potassium wastage and impaired H+ secretion are characteristic of hypokalemic distal RTA, this remains an attractive hypothesis.
In a typical patient with distal (type I or classic) RTA,
urine pH during spontaneous acidosis or after acid loading
cannot be lowered below 5.5 and ammonium exc retion is
reduced. The latter can be inferred at the bedside from
the finding of a positive urine anion gap (UAG; see below).
In such patients, the urine pCO2 measured after bicarbonate loading does not increase normally (i.e., above 70 mm
Hg), reflecting the reduced rate of collecting tubule H+
secretion. The latter feature is useful in distinguishing a
secretory or voltage-dependent defect from a permeability
defect. Permeability defects (i.e., amphotericin B-induced
RTA) do not impair the generation of a normally high
urine
Hyperkalemic RTA

Two pathogenic subtypes of hyperkalemic distal RTA

that are frequently encountered in adults with underlying
renal disease have been well described (Table 3). One

TABLE 3

Causes of Hyperkalemic Metabolic Acidosis

Selective aldosterone deficiency
Low renin
Hyporeninemic hypoaldosteronism (e.g., diabetic
nephropathy)
Prostaglandin synthesis inhibitors
Normal or high renin
Normoreninemic hypoaldosteronism
Hyperreninemic hypoaldosteronism in critically ill patients
Convert ing enzyme inhibitors
Corticosterone methyloxidase deficiency
Heparin therapy
Cyclosporine
Hyperkalemic distal renal tubular acidosis

Obstructive uropathy
Sickle cell hemoglobinopathies
Renal amyloidosis
Acute hypersensitivity interstitial nephritis

Amiloride administration
Trimethoprim'1

Cyclosporine
Aldosterone resistance
Pseudohypoaldosteronism type I (infants)
Pseudohypoaldosteronism type II (Gordon's syndrome) Adult
aldosterone hyporesponsiveness and renal insufficiency
Spironolactone administration
" Metabolic acidosis is usually nol a feature of trimethoprim induced hyperkalemia.

subtype, which corresponds to the animal model of selective aldosterone deficiency (SAD), is characterized by hyperkalemic hyperchloremic metabolic acidosis associated
with low plasma and urinary aldosterone levels, reduced
ammonium excretion, and preserved ability to lower urine
pH below 5.5. This constellation of findings is also referred
to as type IV distal RTA. The findings associated with
SAD or type IV RTA are typified by the syndrome of
hyporeninemic hypoaldosteronism. While hyperkalemia is
the usual manifestation of this syndrome, the development
of hyperchloremic metabolic acidosis is also very common
(more than 75% of cases). In these subjects, acidosis develops because of impaired acid excretion secondary to reduced urinary buffer (ammonia) availability. Ammonia formation in these patients is suppressed as a result of both
hyperkalemia and aldosterone deficiency.
In the other subtype of hyperkalemic RTA, ammonium
excretion is also reduced but, characteristically, urine pH
cannot be lowered below 5.5 not only during acidemia, but
also after stimulation of sodium-dependent distal acidification by the administration of either sodium sulfate or loop
diuretics. In this type (referred to as hyperkalemic distal
RTA), plasma aldosterone levels may be normal or elevated but are more often reduced. The finding of low aldosterone levels suggests the existence of a combined defect,
that is, SAD combined with a tubular defect that interferes
with the ability to maximally lower urine pH despite the
presence of acidosis. The mechanism underlying this subtype of hyperkalemic distal RTA, which was first described
in patients with obstructive uropathy, was originally attributed to a failure to generate a favorable transtubular voltage gradient (lumen negative) in the collecting tubular that
interferes with both H+ and K+ secretion. A recent study
in patients with this subtype of hyperkalemic distal RTA
suggests, however, that the defect in H+ secretion is related
to H-ATPase dysfunction and is independent of the associated defect in K+ secretion.
RTA: CLINICAL FEATURES AND TREATMENT

A common feature of children with proximal and distal

RTA is failure to thrive. Acidemia, by mechanisms not yet
fully delineated, interferes with normal growth. Distal RTA
may also lead to nephrolithiasis or nephrocalcinosis.
The treatment of distal RTA consists of alkali therapy.
When started early in the course of the disease, alkali
therapy normalizes growth. The alkali requirements of
these patients are usually small compared with those of
patients with proximal RTA. A dose of 1 to 2 mEq/kg
body wt of sodium bicarbonate daily is sufficient in most
cases. Many patients with hereditary distal RTA have a
bicarbonate wastage tendency during the first years of life
that gradually abates with advancing age. Hence, the dose
of alkali needed to correct the acidosis decreases after 6
years of age. Alkali can be provided in the form of a sodium
citrate solution, which is well tolerated because it causes
less abdominal distention than bicarbonate. Each milliliter
of citrate solution provides 1 mEq of bicarbonate after

9. Metabolic Addosis

hepatic conversion of citrate into bicarbonate. Potassium

supplements are indicated in patients with hypokalemia.
The hyperkalemic types of distal RTA usually develop in
patients with an underlying tubulointerstitial renal disease
and/or with moderate renal insufficiency who have aldosterone deficiency. SAD is particularly common in patients
with either type I and type II diabetes, especially those
who have developed nephropathy. Other common causes
are listed in Table 3. Patients with hyperkalemic distal RTA
should be differentiated from those with pure SAD because
some patients with the latter condition are more likely
to respond to treatment with mineralocorticoids. Tubular
hyporesponsiveness to the action of exogenous mineralocorticoid, however, is common in some patients with SAD.
High doses of mineralocorticoid may be needed to correct
the hyperkalemia.
Hyperkalemia can also be treated using exchange resins
such as sodium polystyrene sulfonate, which increase gas trointestinal potassium excretion. Correction of hyperkalemia has the additional salutory effect in that it improves
the acidosis by increasing ammonium excretion. Loop diuretics are effective in increasing potassium excretion, and
they also ameliorate the metabolic acidosis associated
with SAD.

75
1. Hyperchloremic acidosis with normal or low plasma potassium
Urine anion
gap
Negative

pH during acidemia

Look for extrarenal causes

(i.e., diarrhea), consider
proximal RTA

UAG = (Na + K) - (Cl + HCO 3 ).

In a relatively acid urine (pH < 6.5), urine bicarbonate
concentration is very low, and its contribution to the UAG
can be considered negligible. Normally, the UAG has a
negative value (0 to -50) provided that a stimulus for
ammonium excretion (i.e., acidosis) is present. The UAG,
by contrast, is positive in patients with various types of
defects in distal urinary acidification. This reflects the in ability of patients with distal RTA to excrete ammonium
appropriately in the face of spontaneous acidosis.
Accordingly, the UAG is useful in the separation of
those patients with reduced ammonium excretion because
of impaired distal acidification from those patients with
hyperchloremic metabolic acidosis from other causes,
namely, gastrointestinal bicarbonate losses (Fig. 2). Normal
subjects with diarrhea or given an acid load to produce
metabolic acidosis display a negative UAG. The negative
UAG in patients with diarrhea reflects an abundance of
ammonium in the urine since the kidney's response to
acidosis is appropriate (i.e., to excrete acid). In patients
with diarrhea, renal acidification is greatly stimulated by
acidemia, and the urine pH should be low. In some patients,

I >5.5
I Distal RTA

<5.5

2. Hyperchloremic acidosis with hyperkalemia Urine

Measure urine NH4,

urine p C O ? after
bicarbonate loading.
If abnormal, distal
RTA is present.

anion gap
Positive

Negative
Look for extrarenal
causes of hyperkalemia
(drugs, etc.)

Urine pH during acidemia

I
<5.5
I
Plasma
aldosterone

DIAGNOSTIC APPROACH TO HYPERCHLOREMIC

METABOLIC ACIDOSIS

After history and physical examination, the next step

in the diagnostic approach to hyperchloremic metabolic
acidosis is to determine whether urinary acidification is
normal (Fig. 2). Ideally, this should involve measurement
of urine pH, titratable acidity, and ammonium excretion.
Whether ammonium is appropriately increased or inappropriately low for the prevailing acidosis can be inferred at
bedside by a simple calculation of the UAG.

Positive Urine

i Low
I
Plasma
cortisol
I

Normal
I
Selective
aldosterone
deficiency

FIGURE 2

I
>5.5

I
Hyperkalemic DRTA
look for specific causes
\
(e.g., obstructive uropathy)

Normal or high
I
Aldosterone
resistance
i Low
I
Adrenal
insufficiency

Diagnostic approach to hyperchloremic acidosis.

however, urine pH rises because of the addition of large

amounts of ammonia buffer into the collecting tubule. In
addition, many of these patients are volume-depleted from
protracted diarrhea, so that the delivery of sodium to the
collecting tubule is impaired. In this situation, urine pH
may not be lowered below 5.5 despite acidemia, until distal
sodium delivery is restored by the administration of salt
or diuretics. The findings of hypokalemia, hyperchloremic
metabolic acidosis, and a high urine pH could erroneously
suggest the diagnosis of distal RTA in these patients. The
key differential finding is the UAG, which is positive if
the patient has distal RTA but negative if the patient has
hypokalemic hyperchloremic metabolic acidosis associated
with diarrhea. Patients with proximal RTA display a negative UAG insofar as they are capable of excreting substantial amounts of ammonium. Chronic respiratory alkalosis
can also increase the UAG because in this condition, uri-

76

Daniel Batlle

nary ammonium excretion is reduced as part of a normal

adaptation to hypocapnia.
Provided that respiratory alkalosis is excluded, patients
with a diminished serum bicarbonate concentration who
have a positive UAG should be suspected to have distal
RTA, which can be separated into hyperkalemic and hypokalemic (classic) distal RTA on the basis of the plasma
potassium (Fig. 2). The diagnosis of distal RTA can be
confirmed by the finding of a urinary pH above 5.5 in
the face of spontaneous or ammonium chloride-induced
metabolic acidosis. In general, if the patient is spontaneously acidotic (pH < 7.35), there is no need to induce
severe acidosis in order to evaluate urinary acidification.
If the patient is suspected of having distal RTA and acidosis
is mild or not present, ammonium chloride can be given
(0.1 g/kg body wt daily for 3 consecutive days). Measurements of urine pH, urinary ammonium, and titratable acidity should be performed at the end of Day 3 to confirm or
exclude the diagnosis of distal RTA. It should be noted
that there are patients with a defect in distal acidification
in whom urine pH may be less than 5.5 in the face of
metabolic acidosis. This typically occurs in patients with
hyperkalemic RTA associated with SAD (see Fig. 2). In
some patients without hyperkalemia the ability to lower
urine pH below 5.5 may also be preserved, even when distal
acidification assessed by the urine pCO2 after bicarbonate
loading is defective (i.e., urine pCO2 does not increase
above 50-60 mm Hg; normal >70 mm Hg) (Table 2). This
has been well documented in individuals receiving chronic
lithium therapy, subjects with medullary sponge kidney,
and some patients with idiopathic hypercalciuria. These
patients can be considered to have incomplete distal RTA
since metabolic acidosis is often absent.
HIGH AG TYPES OF METABOLIC ACIDOSIS
Renal Failure

Renal failure, both acute and chronic, is a common cause

of metabolic acidosis that is usually classified as a high AG
type (Table 4). It should be appreciated, however, that the
fall in plasma HCOs associated with renal failure is mainly
due to an absolute decrease in ammonium excretion as a
result of reduced renal mass and that a reciprocal increase

in the AG does not need to occur. Variable degrees of

anion retention (sulfate, phosphate, and organic anions)
occurring as renal insufficiency progresses determine
whether a high AG metabolic acidosis develops. The predominant pattern among patients with end stage renal dis ease on maintenance hemodialysis is a mixed one, that is, a
combined high AG and hyperchloremic metabolic acidosis.
In patients with chronic renal disease an increase in ammonia production by remaining nephrons is capable of
maintaining normal acid-base balance until the glomerular
filtration rate has dropped below 20 to 25 mL/min. Some
of the metabolic abnormalities with chronic renal insufficiency, however, may interfere with urinary acidification.
In particular, chronic hyperkalemia worsens the acidosis
by suppressing ammoniagenesis. An absolute decrease in
overall renal ammonia production ensues when GFR is less
than 20 mL/min and acidosis develops. Even with advanced
renal insufficiency the urine is acid (pH < 5.5) suggesting
that the collecting tubules of residual nephrons are capable
of adequate H+ secretion. In fact, when acidosis is seen in
patients with moderate degrees of renal failure, a distal
acidification defect should be suspected.
The acidosis is usually nonprogressive and relatively
asymptomatic. The goal of alkali therapy in this group is
mainly to avert the consequences of protracted acidosis on
bone tissue which may be through a direct demineralization
effect of acidemia or through its effects on the hormones
that regulate calcium metabolism, namely, parathyroid hormone and vitamin D.
Lactic Acidosis

Lactate is formed in the cell cytosol from pyruvate. The

enzyme catalyzing this reaction is lactate dehydrogenase
(LDH). Under normal conditions, the rates of lactic acid
regeneration and utilization are matched (about 15-20
mEq/day). The liver and, to a lesser extent, the kidneys
are the major sites of lactic acid uptake and metabolism.
Lactic acidosis, resulting from the accumulation of excess
lactate, is usually seen in severely ill patients many of whom
have tissue hypoxia due to tissue hypoperfusion from hypotension with or without hypoxemia. When the tissue oxygen
supply does not meet the cellular oxygen demand, type A
lactic acidosis occurs.

TABLE 4

Causes of Metabolic Acidosis with Increased Plasma Anion Gap

Characteristic features and other comments'1

Etiology

Major circulating anion

Ketoacidosis

/3-Hydroxybutyrate, acetoacetate

Diabetes, alcohol, starvation

Lactic acidosis

Lactate

Shock, tissue hypoxia, liver failure

Renal failure

Sulfate, phosphate, variety of organic acids

Methanol

Formate

Ethylene glycol

Glycolate, lactate

Salicylates
Toluene
Paraldehyde

Variety of organic acids, salicylate

Hippurate
Acetate?

<f> BUN
Increased osmolal gap, hyperemic optic disk
Increased osmolal gap; ARF, urinary oxalate crystals
Concomitant respiratory alkalosis, tinnitus, fever

"ARF, acute renal failure; RTA, renal tubular acidosis.

It also causes distal RTA and hypokalemia

Very rare

9. Metabolic Acidosis

77

Some patients with lactic acidosis, however, lack any

evidence of tissue hypoxia. This condition is known as type
B lactic acidosis and almost always results from imp aired
lactate clearance due to hepatic failure. In type A lactic
acidosis, the diagnosis is usually made on the basis of the
clinical setting (shock, hypoxemia, severe anemia). Type
B lactic acidosis should be suspected when there is evidence
of liver impairment or ingestion of substances known to
produce this condition. Type B lactic acidosis can also occur
with large tumors producing a large cell turnover and a
marked increase in pyruvate production.
A distinctive type of lactic acidosis can occur in patients
with bacterial overgrowth due, for example, to jejunoileal
bypass or small-bowel resection. In this situation, intestinal
bacteria can metabolize glucose into D-lactic acid, which
then accumulates in the blood.
Treatment of lactic acidosis is directed toward correcting
the underlying cause. Levels of plasma lactate higher than
20 mEq/L reflect very severe acidosis and are often associated with a high mortality. In monitoring the patient, one
should not rely on an isolated lactate level because of
the rapidly changing nature of the underlying condition.
Monitoring the plasma AG provides an estimate of serum
lactate changes over time and may obviate measurement
of serial lactate levels.
The administration of bicarbonate is usually indicated
when blood pH is very low (i.e., >7.10) despite adequate
respiratory compensation. In lactic acidosis, to counteract
the excessive lactic acid production, large amounts of bicarbonate are usually needed and this may lead to increased
systemic CO2 generation, a rise in cell CO2 and thus a fall
in intracellular pH. Despite this theoretical problem, the
use of bicarbonate therapy in cases of severe acidemia is
advisable, provided that blood pCO2 is reduced. A more
specific approach to correct lactic acidosis entails the use of
dichloroacetate. This agent, now available in some hospital
pharmacies, is effective in lowering lactate levels by stimu lating pyruvate dehydrogenase and stimulating mitochondria! pyruvate oxidation. Despite clear-cut improvement
of the acidosis, however, controlled studies have failed to
demonstrate an improvement in survival in patients with
lactic acidosis treated with this agent.
Diabetic Ketoacidosis

This condition is characterized by hyperglycemia of variable degree and metabolic acidosis secondary to overproduction of ketoacids. The hormonal profile of diabetic ketoacidosis comprises low insulin levels, hyperglucagonemia,
high cortisol levels, high level of circulating catecholamines,
and elevated growth hormone levels. These hormonal alterations trigger and maintain the ketogenesis. The accumulation of acetoacetic and /3-hydroxybutyric acids results in
metabolic acidosis when these acids rapidly dissociate to H+
and the corresponding ketoanions. Some of the acetoacetic
acid is nonenzymatically converted to acetone. Acetone is
a volatile ketone that is excreted via the lungs producing
a characteristic odor to the breath.

Although the acidosis is usually of the high AG type, a

striking hyperchloremic type of acidosis may be seen when
ketones are rapidly lost in the urine. If volume contraction
is severe, ketoacids are retained and the AG is elevated.
In this situation, the retained ketoacids provide a source
of potential bicarbonate. This may contribute to a fast
recovery from acidosis owing to the equimolar conversion
of retained ketone salts to bicarbonate.
Metabolic acidosis in diabetic ketoacidosis is often severe. The diagnosis of ketoacidosis is established by demonstrating ketonemia or ketonuria, metabolic acidosis, and
hyperglycemia (Fig. 3). Ketonuria may be undetectable if
most of the offending acetoacetic acid has been converted
to /3-hydroxybutyric acid, which is not detected on the
routine urine dipstick. Lack of a marked elevation in blood
sugar may result in delayed diagnosis. Uncontrolled hyperglycemia usually has caused an osmotic diuresis with a
substantial loss of sodium, chloride, and potassium. Many
such patients therefore develop various degress of prerenal
azotemia. A significant total body potassium deficit is usually present even in the presence of a normal serum potas sium concentration. Some patients may have hyperkalemia,
which is mainly due to impaired potassium transport into
cells caused by the insulin deficiency. Significant phosphate
depletion is usually pres ent as a result of poor food intake
and urinary phosphate wasting.
The goal of therapy is to reverse and halt the ketogenesis
and to replace the electrolyte and water deficits. Insulin is
needed to inhibit ketogenesis and break the cycle of lipolysis and ketogenesis. Insulin preferably should be adminis tered as a low-dose constant intravenous infusion. Insulin
should not be stopped as glucose concentrations approach
the normal range. Rather, glucose should be infused and
insulin continued until the ketosis has cleared. Although
these patients often have more of a water than sodium
deficit, isotonic saline is usually needed initially to achieve
a rapid correction of the volume deficit. Correction of potassium deficit should be initiated early, even when plasma
potassium concentration is normal, provided that insulin
has been given and an adequate urine output is present.
The role of bicarbonate therapy in patients with diabetic
ketoacidosis has been debated. In general, the use of bicarbonate is warranted when the acidosis is severe, since its
potential benefits seem to outweigh the risks. Bicarbonate
should be administered to patients whose arterial blood
pH is less than 7.20 despite adequate hyperventilation and
to those with a bicarbonate concentration below 10 mEq/
L. During recovery, a high AG metabolic acidosis is sometimes replaced by hyperchloremic acidosis. The excretion
of ketoanions, which represent potential bicarbonate, coupled with the administration of intravenous saline infusion
and avid renal retention of chloride leads to hyperchloremia. The development of hyperchloremic acidosis is an
indication for continuation of bicarbonate therapy.
Starvation Ketoacidosis

During starvation, lipolysis and gluconeogenesis provide

the body with the needed calories for survival. The low

78

Daniel Batlle
Serum ketones
Urine ketones
I
Positive

I
Negative-.

Blood glucose

Lactic acidosis: lactate levels

Renal failure: BUN/creatinine

Intoxications: calculate osmolal gap

Wide
High

Normal or low

Diabetic ketoacidosis
Ethanol?

Fasting ketoacidosis
Ethanol?
FIGURE 3

Ethylene glycol
Methanol

Normal (near 0)
ASA
intoxication Dlactic acidosis

Diagnostic approach to high anion gap metabolic acidosis.

insulin levels induced by hypoglycemia and the associated

increase in glucagon, epinephrine, cortisol, and growth hormone levels stimulate lipolysis with release of fatty acids.
The accumulation of ketone bodies may result in metabolic
acidosis which is often of mild severity. Increased net acid
excretion by the kidney may correct the acidosis over the
next few days, obviating the need for bicarbonate therapy.
Previously malnourished individuals, however, may have
inadequate ammoniagenesis and thus develop a more severe acidosis. Patients may develop metabolic alkalosis
with refeeding, which may be related to the abrupt cessation of ketogenesis while urinary acidification remains stimulated.
Alcoholic Ketoacidosis

This condition is characterized by acidosis secondary to

increased ketogenesis and variable blood glucose levels in
the absence of clinical diabetes. Typically, the condition
develops following an alcoholic binge and an episode of
vomiting and starvation. Unlike starvation ketoacidosis,
the acidosis in alcoholic ketoacidosis may be severe. The
combination of low insulin levels (related to low food intake), high levels of epinephrine (which may be related,
in part, to withdrawal of alcohol), and elevation of cortisol
levels result in the stimulation of glycogenolysis, gluconeogenesis, and lipolysis. The net result is acidosis of variable
severity, hyperglycemia or hypoglycemia, and a clinical
picture easily confused with diabetic ketoacidosis.
There is usually a prompt improvement in response to
hydration and glucose administration without need for exogenous insulin. Bicarbonate should be given if the acidosis
is severe (blood pH <7.20). Therapy also includes correcting concomitant electrolyte disturbances such as hypophosphatemia, hypomagnesemia, and hypokalemia.
Salicylate Intoxication

Two distinct forms of intoxication may result from salicylate ingestion: (1) an acute form, which usually follows an

acute overdose ingested accidentally or intentionally; and

(2) a chronic form, which results from gradual accumulation
of the drug during long-term administration. Acid-base
disturbances are predominant in the acute form. Typical
manifestations of acute salicylate intoxication include intense hyperventilation, fever, stupor, coma, and convulsions. Tinnitus, deafness, and vertigo are common complaints particularly in patients with chronic salicylism.
Other laboratory abnormalities include hypouricemia,
which is due to inhibition of renal tubule urate reabsorption
by high salicylate levels.
The diagnosis of salicylate intoxication is usually obvious
from the history and can be confirmed by checking the
urine for salicylates using ferric chloride which produces
a purple color if salicylate is present. Severe toxicity occurs
when blood salicylate levels reach 100 mg/dL. In chronic
salicylis m, however, dose nomograms are not useful and
death can occur at low blood salicylate levels.
Combined metabolic acidosis and respiratory alkalosis
is a frequent finding in adult patients with acute salicylate
intoxication. In children, metabolic acidosis is predominant, often overshadowing the attendant respiratory alkalosis while in adults the respiratory alkalosis usually predominates. Respiratory alkalosis is the result of direct
stimulation of the respiratory center by salicylate. The
amount of salicylate required to produce this effect is much
less than that required to produce metabolic acidosis. The
mechanism of metabolic acidosis in patients with salicylate
intoxication is not totally clear. Only a small fraction of
the decrement in plasma bicarbonate and the increment in
unmeasured anions may be directly attributed to salicylic
acid accumulation. Ketoacidosis as well as lactic acidosis
may contribute to the metabolic acidosis.
Removal of the salicylate from the body starts with gas tric lavage if the patient presents within 6 hours of drug
administration. Repeated oral administration of oralactivated charcoal enhances whole-body drug elimination
and interferes with drug absorption. Sodium bicarbonate

9. Metabolic Addosis

79

not only corrects the acidosis but also augments the renal
clearance of salicylates by producing an alkaline diuresis
(urine pH > 7.0) and may limit the amount of the drug
that enters the cerebrospinal fluid and brain cells. Hemodialysis is a very efficient way of removing salicylate from
the body and also helps correct the acidosis and other
electrolyte abnormalities.
Ethylene Glycol

Ethylene glycol is commonly found as a component of

antifreeze and as a solvent in a number of industrial applications. It can cause damage to the central nervous system
(CNS), the kidneys, and the cardiopulmonary system. Ethylene glycol is metabolized by alcohol dehydrogenase to
glycoaldehyde, which is further metabolized to glycolic,
glyoxilic, and oxalic acidsall potentially toxic compounds. Suppression of the citric acid cycle and altered
intracellular redox state may lead to lactic acid overproduction. A high concentration of ethylene glycol in plasma
will result in a disparity between measured and calculated
plasma osmolality. This high plasma "osmolal gap" may
also be seen in ethanol and methanol intoxications as well
as in some cases of diabetic ketoacidosis (Fig. 2). The
plasma osmolality may be estimated according to the
formula
BUN
(Na X 2) +
28
Calcium oxalate crystalluria is a
hallmark of this type of intoxication and may cause
obstructive renal failure when crystals precipitate in the
renal tubules. The urine should be examined for the
presence of the crystals which may be needle shaped
(monohydrate) or envelope shaped (dihy-drate).
Treatment includes supportive measures, diuresis, bicarbonate, ethanol administration and the provision of thiamine and pyridoxine supplements. Ethanol competes with
ethylene glycol metabolism by alcohol dehydrogenase and
thus slows the latter's metabolism. A serum ethanol level
of 100 mg/dL can be achieved by either nasogastric or
intravenous administration of ethanol. Hemodialysis may
be indicated to quickly lower ethylene glycol plasma levels,
to remove other potentially toxic metabolites, and to correct the acidosis.
Methanol

MethanoC or wood alcohol, is a common solvent and

fuel component. It may be ingested as a substitute for
ethanol, or in suicide attempts. Methanol is metabolized via
alcohol dehydrogenase to formaldehyde which is rapidly
converted to formate, the anion responsible for the AG
and most of the toxicity. It can cause severe optic nerve
as well as CNS injury. Symptoms may include blurred vision
which suggests serious toxicity, respiratory depression, cya-

nosis, altered mental status, and cardiovascular collapse.

There is a latent period before the onset of symptoms that
averages 12 to 24 hours but that may be delayed up to 3
to 4 days if there is concomitant consumption of ethanol.
Acidosis, often severe, is related to the accumulation of
formate as well as lactate, ketoacids, and other organic
acids. An osmolal gap is usually present. The finding of
optic nerve swelling coupled with the absence of crystalluria strongly suggests the diagnosis of methanol as opposed
to ethylene glycol intoxication. Therapy includes osmotic
diuresis, administration of ethanol, bicarbonate infusion,
and hemodialysis.
Bibliography
Adrogue HJ, Wilson H, Boyd AE, Suki WN, Eknoyan G: Plasma
acid-base patterns in diabetic ketoacidosis. N Engl J Med
307:1603-1610, 1982.
Batlle D: Hyperchloremic acidosis. In: Giebisch G and Seldin D
(eds) The Regulation of Acid -Base Balance. Raven Press, New
York, pp. 107-121, 1989.
Batlle D, Flores G: Underlying defects in distal renal tubular
acidosis. New understandings. Am J Kidney Dis 27(6):896915, 1996.
Batlle D, Arruda JAL, Kurtzman NA: Hyperkalemic distal renal
tubular acidosis associated with obstructive uropathy. N Engl /
Med 304:373-380, 1981.
Batlle D, Grupp M, Gaviria M, Kurtzman NA: Distal renal tubular
acidosis with intact ability to lower urine pH. Am J Med 72:751758, 1982.
Batlle D, vonRiotte A, Schlueter W: Urinary sodium in the evalua tion of hyperchloremic metabolic acidosis. N Engl J Med
316:140-144, 1987.
Batlle D, Hizon M, Cohen E, Gutterman C, Gupta R: The use
of the urinary anion gap in the diagnosis of hyperchloremic
metabolic acidosis. N Engl J Med 318:594-599, 1988.
Brenner RJ, Spring DB, Sebastian A, et al.: Incidence of radio graphically evident bone disease, nephrocalcinosis, and nephrolithiasis in various types of renal tubular acidosis. New Engl J
Med 307:217-221, 1982.
Emmett M, Seldin D: Overproduction acidosis. In: Giebisch G
and Seldin D (eds) The Regulation of Acid-Base Balance. Raven
Press, New York, pp. 391 -429, 1989.
Fernandez PC, Cohen RM, Feldman GM: The concept of bicarbonate distribution space: The crucial role of body buffers. Kidney Int 36:747-752, 1989.
Gennari JR, Rimmer JM: Acid- base disorders in end stage renal
disease: Part I. Semin Dialysis 3:81-85, 1990.
Green J, Kleeman CR: Role of bone in regulation of systemic
acid- base balance. Kidney Int 39:9 -26, 1991.
Kurtzman NA: Renal tubular acidosis: A constellation of syndromes. Hasp Pract 173-198, 1987.
Oh MS, Phelps KR, Traube M, Barbosa- Saldiar JL, Boxhill C,
Carroll HJ: D-Lactic acidosis in a man with the short -bowel
syndrome. N Engl J Med 301:249-252, 1979.
Peterson CD, Collins AI, Hines JM, Bullock ML, Keane WF:
Ethylene glycol poisoning: Pharmacokinetics during therapy
with ethanol and hemodialysis. N Engl J Med 304:21-23, 1981.
Stacpoole PW, Wright EC, Baumgartner TG, et al: Dichloroacetate for lactic acidosis in adults: A controlled clinical trial of
dichloroacetate for treatment of lactic acidosis in adults. N Engl
J Med 327:1564-1569, 1992.

METABOLIC ALKALOSIS
EDWARD R. JONES

Metabolic alkalosis is characterized by a primary increase in the plasma bicarbonate concentration. The
resultant alkalemia suppresses ventilation, yielding a secondary increase in pCO2. An elevated bicarbonate concentration may also be observed as compensation for
the hypercapnia characteristic of respiratory acidosis. Dis tinguishing between these disturbances, each associated
with an elevated bicarbonate level and hypercapnia, requires a blood gas determination to define pH. The history
and physical examination are critical in distinguishing
between these two disorders and in identifying patients
who have both metabolic alkalosis and respiratory acidosis. The distinctive physical characteristics of patients
with chronic lung disease are helpful clues. Formulas
and confidence bands to predict metabolic adaptation to
hypercapnia and define the presence of simple metabolic
alkalosis, respiratory alkalosis, or mixed disturbances are
discussed in Chapter 12.
In a study of 13,000 arterial blood gases in 3300 hospitalized patients, metabolic alkalosis was the most frequent
acid-base disturbance. Although its direct cause is poorly
defined, there appears to be an increased morbidity and
mortality in surgical patients with multisystem organ fa ilure
and metabolic alkalosis (blood pH > 7.55). Therefore the
recognition and appropriate correction of metabolic alkalosis are important.
CLINICAL MANIFESTATIONS

Patients with metabolic alkalosis manifest symptoms related to the cause of the alkalosis, i.e., postural symptoms,
weakness, and thirst are associated with volume depletion;
neuromuscular symptoms and arrhythmias are associated
with hypokalemia. The clinical presentation is generally a
(Consequence of the alkalemia, which exerts significant effects on the central nervous and neuromuscular systems,
cardiovascular system, and pulmonary vasculature as well
as contributing to various metabolic derangements. These
signs and symptoms occur in both metabolic and respiratory alkalosis. It is difficult to differentiate most of the side
effects of alkalemia from the numerous other concomitant
metabolic derangements in patients with metabolic alkalosis such as hypocalcemia and hypomagnesemia.
Alkalemia may contribute to the changes in mental status
seen in patients with metabolic alkalosis and to a lowering
Primer on Kidney Diseases. Second Edition

of the seizure threshold. Tetany and neuromuscular irritability including carpopedal spasms and positive Chvostek
or Trousseau signs may be seen.
Ventricular and supraventricular irritability as well as
increased sensitivity to digoxin appears to be more frequent. Antiarrhythmics may be ineffective until the alkalemia is corrected. Decrements in cardiac output and cardiovascular instability are common.
Alkalemia decreases ionized calcium by shifting ionized
calcium onto plasma proteins. Therefore, despite normo calcemia, the alkalemic patient may present with seizures or
carpopedal spasms. A decrease in urinary calcium excretion
may also be observed; its clinical significance is not well defined.
Alkalemia shifts potassium intracellularly, resulting in
hypokalemia. In addition, metabolic alkalosis causes potassium depletion which is generally correlated with the
magnitude of the alkalosis. The consequences of potassium
depletion are discussed in Chapter 13.
Metabolic alkalosis causes an elevated plasma anion gap.
Alkalosis stimulates glycolysis, more specifically phosphofructokinase activity, thereby increasinglactate production.
The plasma lactate level may increase as much as 5 mEq/
L; this can raise the anion gap by an equivalent amount.
Volume contraction concentrates plasma proteins and increases the unmeasured anion. Finally, alkalemia is associated with the loss of protons from plasma proteins, uncovering negative charges and contributing to the rise in
unmeasured anion.
Alkalemia also alters the oxyhemoglobin curve, shifting
the curve leftward and potentially limiting oxygen availability. The clinical significance of this is unknown.
PATHOPHYSIOLOGY

Recognition and appropriate intervention to correct the

many processes that result in metabolic alkalosis demand
a basic understanding of the pathogenesis of metabolic
alkalosis. Under normal conditions, an excess of bicarbonate from any source would be excreted by the kidney by
suppressing renal acid excretion and bicarbonate reabsorption. A sustained rise in serum bicarbonate concentration
can only occur if two factors are present. First there must
be a source of the alkali (i.e., the generation of alkalosis),
and there must be factors in place that retain the alkali or
limit the ability of the kidney to excrete bicarbonate (i.e.,
80

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

1 0 . M e t abolic Alkalosis

81

maintenance of the alkalosis). Under normal conditions, a

rise in serum bicarbonate from any source would be excreted by the kidney by suppressing renal acid excretion
and bicarbonate reabsorption. In those clinical settings
where metabolic alkalosis occurs, bicarbonate excretion is
minimized as a result of factors which both generate and
maintain the alkalosis.
Generation of Metabolic Alkalosis

The generation of bicarbonate (Table 1) results from

either the renal or extrarenal loss of acid or from gain of
base to the extracellular fluid. Net negative hydrogen ion
balance (acid excretion exceeding production) results in a
rise in extracellular fluid (ECF) bicarbonate concentration.
The loss of acid translates into equimolar gain of bicarbonate.
Secretion of hydrogen ion from the gastric mucosa is
associated with the addition of bicarbonate to blood. Vomiting and gastric drainage are common causes of metabolic
alkalosis. Intestinal acid loss can also occur with chloridelosing diarrhea (congenital or acquired chloridorrhea).

TABLE I

Pathophysiologic Approach to Metabolic Alkalosis

Generation of Metabolic Alkalosis
Loss of acid from ECF
Nonrenal acid loss
Gastrointestinal fluid losses
Gastric losses Chloride-losing
diarrhea
Cellular shifts: Potassium for hydrogen exchange
Renal acid losses Acid secretion due to diuretics
Nonreabsorbable anions: enhanced distal
sodium -hydrogen exchange Severe potassium
depletion Gain of bicarbonate to the ECF Conversion of
bicarbonate equivalents (citrate, acetate,
lactate)
Bicarbonate administration: parenteral or oral
Resin exchanges plus antacids Unchanged
bicarbonate content Contraction of ECF
bicarbonate Posthypercapnic state
Maintenance of Metabolic Alkalosis
Decreased GFR (decreased filtered load of bicarbonate)
Enhanced proximal bicarbonate reabsorption
Reducedv arterial blood volumeabsolute or effective
Potassium depiction (also decreased GFR)
Hypercapnia
Decreased parathyroid hormone
Hypercalcemia Increased distal tubular
bicarbonate reabsorptio n
Mineralocorticoid excess
Hypokalemia
Nonreabsorbable anion
Chloride depletion
ECF, extracellular fluid; GFR, glomerular filtration rate.

A minor rise in serum bicarbonate concentration can

occur due to a translocation of protons across cells in exchange for potassium during hypokalemia. This mechanism
has little or no clinical significance.
Renal loss of acid is probably the most common cause
of metabolic alkalosis. Loop diuretics like furosemide as
well as the thiazides stimulate acid excretion by increasing
distal nephron sodium delivery and by causing secondary
hyperaldosteronism. Hypokalemia enhances proton secretion and aimnoniagenesis thus increasing net acid excretion.
Excessive alkali gains can be seen during the administration of bicarbonate or its equivalents during cardiopulmo nary resuscitation, with massive blood transfusions (sodium
citrate is used as an anticoagulant), or the administration
of total parenteral nutrition formula (acetate). The addition
of alkali from the gastrointestinal tract occurs during the
concomitant use of exchange resins (Kayexelate) and
poorly absorbable antacids.
Contraction alkalosis results from the loss of chloriderich, bicarbonate-free fluid which increases serum bicarbonate concentration by 3 to 5 mEq/L. The metabolic alkalosis commonly seen with loop diuretics results predominately from enhanced renal acid excretion and the
concomitant hypovolemia, and not contraction of ECF bicarbonate alone.
Maintenance of Metabolic Alkalosis

A sustained rise in bicarbonate concentration implies

decreased renal bicarbonate excretion. Under normal circumstances, the addition of alkali to the ECF volume is
associated with volume expansion and hypochloremia. The
normal kidney senses the volume expansion. The alkali
is excreted along with sodium, and the hypochloremia is
corrected. Sustained elevations in bicarbonate concentration require one of the following: a decreased glomerular
filtration rate (GFR), enhanced proximal tubular bicarbonate reabsorption (both of the latter are volume-dependent
stimuli), or increased distal tubular bicarbonate reabsorption (a volume-independent mechanism).
Decrements in GFR translate into diminished filtered
bicarbonate loads, limiting bicarbonate excretion. Enhanced proximal tubular bicarbonate reabsorption will
sustain the ECF bicarbonate and maintain the alkalosis.
Reduced effective arterial blood volume (EABV) is associated with decreased bicarbonate excretion. Hypokalemia
and hypercapnia stimulate proton secretion and thus bicarbonate addition to blood. Hypokalemia may also decrease
GFR, further limiting bicarbonate excretion.
Enhanced distal nephron bicarbonate reabsorption is an
important factor in maintaining metabolic alkalosis. Persis tent hypermineralocorticoidism, hypokalemia, and the
presence of a poorly reabsorbable anion stimulate acid
excretion and subsequent addition of alkali to blood.
The importance of hypochloremia in the maintenance
of metabolic alkalosis is not established. Some suggest that
chloride depletion itself may maintain the metabolic alkalosis and that this effect is independent of plasma volume.

Edward R. Jones

82
Indeed, chloride depletion appears to enhance bicarbonate
reabsorption while distal bicarbonate secretion decreases.
Differences in location of the chloride/bicarbonate (C\~/
HCOs) exchanger within intercalated cells of the cortical
collecting tubule plays a role in maintenance of metabolic
alkalosis. Type A intercalated cells have the transporter
on the basolateral membrane. Type B cells are thought to
have these transporters on the luminal side. Hydrogen ion
secretion via the H-ATPase pump and passive chloride
cosecretion are enhanced on the luminal side during decreased chloride delivery. There is a concomitant enhancement of chloride entry into type A cells with subsequent
HCOs addition to blood. The Cl /HCOl exchange is reversed in type B cells; thus, enhanced chloride secretion
into the lumen results in decreased bicarbonate secretion.
As a result, both cell types play important roles in maintaining metabolic alkalosis in chloride-depleted states.
In essence, the recognition of metabolic alkalosis and
its therapeutic intervention require an understanding that
subsequent metabolic alkalosis is a result of the dependent
processes of generating alkali and maintaining the alkalosis.
Correction of either component alone does not resolve the
alkalosis. For instance, cessation of vomiting alone will not
lower serum bicarbonate. Correction of the hypovolemia
along with cessation of vomiting are both required to permit normalization of the ECF bicarbonate concentration.
Clinical Conditions Resulting in Metabolic Alkalosis
(Fig. I ;Table 2)
In keeping with the pathophysiology presented above,
metabolic alkalosis is best categorized according to the
volume status of the patient. Hypovolemic, chloridedeficient patients excrete less than 15 mEq/L of chloride
on a random urine. These patients have been classified
as having saline-responsive alkalosis. Indeed, when their
plasma volume is restored with sodium chloride, a bicarbonaturia ensues.
In contrast, the ECF volume expanded (chloride-rich)
patient excretes greater than 20 mEq/L of chloride. These
patients, who do not respond to sodium chloride adminis tration, have saline-resistant alkalosis.
Metabolic Alkalosis Associated with ECF Volume Contraction
An increase in bicarbonate concentration can be generated either from renal or extrarenal sources. Its maintenance results from the hypovolemia and hypokalemia that
enhance proximal bicarbonate reabsorption and from the
decreased GFR that limits bicarbonate excretion. In addition, the hypokalemia and excess mineralocorticoid state
enhance distal acid excretion and bicarbonate addition
to blood.
Renal Alkalosis
Thiazide, loop, and mercurial diuretics cause hypokalemic metabolic alkalosis, with secondary hypokalemia, hypovolemia, and hyperaldosteronism. Metabolic alkalosis

occurs in 3 to 50% of patients receiving diuretics. Thiazide

diuretics increase bicarbonate concentration by only 2 to
1 mEq/L. The use of loop diuretics either alone or in combination with metalazone, particularly in patients with anasarca, can increase bicarbonate concentration by 15 to
20 mEq/L.
Laboratory findings vary depending on whether the patient is still taking the diuretic. Continued diuretic consumption results in a natriuresis with a high urinary chloride, despite hypovolemia. Once the diuretic is stopped,
urinary sodium and chloride fall below 15 mEq/L. This
electrolyte pattern may occur with the surreptitious use of
diuretics for weight loss, and may be diagnosed by a urinary
diuretic screen.
The administration of poorly reabsorbable anions (e.g.,
ampicillin, carbenicillin, or other penicillins) results in an
obligatory natriuresis with enhanced distal tubular potassium and proton losses. The resultant hypokalemia, hypovolemia, and metabolic alkalosis can be blunted or aborted
with adequate potassium supplementation and volume repletion.
Bartter's syndrome is a rare disorder whose pathophysiology is still incompletely understood. The clinical
presentation comprises hypokalemia, metabolic alkalo sis, hypomagnesium, hyperuricemia, hyperreninism, and
hyperaldosteronism but without hypertension. Prostaglandin production is increased. Patients are hyperchloruric,
with potassium wasting. The majority of patients are less
than 25 year olds, with common presenting symptoms of
polyuria, paresthesia, and muscle weakness and cramping.
Recent studies suggest that patients with Bartter's syndrome have a mutation in the sodium: potassium: 2chloride cotransporter or in a potassium secretory transporter in the thick limb of the ascending loop of Henle.
An additional syndrome which needs differentiation
from Bartter's syndrome is Gitelman's syndrome. Gitelman's syndrome is a rare entity associated with hypokalemia, hypomagnesemia, and hypocalciuria. Urinary chloride levels are low and there is no juxtaglomerular
apparatus hyperplasia. There is impaired transport in the
cortical thick limb of the loop of Henle and reduced sodium
reabsorption in the distal tubule. The former defect probably accounts for the magnesium wasting and the latter for
the hypocalciuria. Indeed, data suggest that there is a defect
in the gene for the thiazide-sensitive distal tubular sodiumchloride cotransporters.
Nonrenal Alkalosis
Vomiting is associated with the loss of water, electrolytes,
and protons from the ECF. The loss of hydrogen ion into
the gastric lumen is associated with delivery of equimolar
quantities of bicarbonate into gastric venous blood increas ing the filtered load. The loss of 3 or 4 L of gastric content
will result in marked electrolyte and fluid losses. The hypovolemia results in hyperaldosteronism, causing a marked
kaliuresis, profound potassium depletion and subsequent
hypokalemia (the gastric potassium losses are trivial in this
setting), hypochloremia, and severe metabolic alkalosis.

10. Metabolic Alkalosis

83
Pathophysiologic approach to patients with metabolic alkalosis
Hyperbicarbonatemia
Primary disturbance

Variable
Variable

Volume status Blood

Normal

pressure Bicarbonate

Increased

Decreased
Increased

content

Decreased

Variable

Increased

Low or normal

Normal

Increased

Mineralocorticoid effect

Variable

Increased

Increased
Variable

Plasma aldosterone
Decreas ed
Plasma renin

Decreased

Decreased

Decreased

Increased

Increased
Diuretic use
Nonreabsorbable anions
Magnesium depletion
Posthypercapnia
Refeeding Bartter's
syndrome Gitelman's
syndrome
Hypercalcemia and
hypoparathyroidism

Gastric losses
Chloride-wasting
diarrhea
Chloride-deficient
baby formula
Villous adenoma

Renal artery
stenosis
Accelerated or
malignant HBP
Renin-secreting
tumor

FIGURE I

TABLE 2

Differential Diagnosis of Metabolic Alkalosis

Urine chloride <15 mEq/L (chloride-responsive)

Gastric fluid losses Stool losses: Chloride-losing
diarrhea Diuretic therapy" (diuretic screen
negative) Posthypercapnic state Refeeding
Urine chloride >20 mEq/L (chloride-resistant)
Primary hyperaldosteronism Exogenous steroids
Bartter's syndrome Cushing's syndrome Severe
hypokalemia Alkali loading Hypercalcemia
" Urinary chloride will vary depending on proximity to ingestion of diuretics. Generally, a negative
diuretic screen is associated with remote diuretic use
and a low urine chloride.

Primary
hyperaldosteronism

Endogenous mineralocorticoids Cushing's

syndrome Adrenal
enzyme deficiency
Liddle's syndrome
Exogenous
mineralocorticoids
Licorice
Chewing tobacco
Carbenolone

Exogenous
bicarbonate load
Bicarbonate (oral)
Bicarbonate
precursor's

Contraction
Cellular shifts

acetate
citrate Anion
exchange resin
plus antacids Milkalkali syndrome
Overshoot alkalosis

Approach to metabolic alkalosis.

The clinical setting of vomiting is best approached by evalu ating the various phases of development of the alkalosis.
The early developmental phase is associated with a rapid
rise in bicarbonate, mild hypovolemia, and hypochloremia.
The filtered load of bicarbonate exceeds the reabsorptive
capacity of the proximal tubule. Therefore bicarbonaturia
with an alkaline urine is present. Despite hypovolemia,
there is a natriuresis due to an obligatory bicarbonate diure sis. Natriuresis with hypovolemia is also seen with diuretic
use. Kaliuresis is marked, whereas the urine is chloride
deplete. There is a positive urinary anion gap (denned
as the urinary sodium plus potassium minus chloride). A
positive urinary anion gap with alkaline urine implies that
bicarbonate is the unmeasured anion; when the urine pH
is low, nonreabsorbable anions (ketones, lactate, penicillin,
etc.) account for the gap. (See Chapter 9 for a more extensive discussion of the urine anion gap.) Once gastric losses
end, and the patient is profoundly hypovolemic (maintenance phase), proximal sodium reabsorption is maximized,
GFR is depressed, bicarbonaturia is eliminated, and sodium
is reabsorbed. The kaliuresis is lessened but persists, and
approximates the low urine chloride plus an undefined

quantity of anion. Urine pH falls because

of the effect

84

Edward R. Jones

of aldosterone to stimulate distal proton secretion, and

because the urinary anion gap is no longer significantly
positive. During the correction of the alkalosis (reparative
phase), as plasma volume is restored and GFR improves,
bicarbonaturia resumes as do the natriuresis, alkaline urine,
and positive urinary anion gap. The urine chloride remains
low until volume repletion is complete. The common feature in these phases is the low urinary chloride with persis tent kaliuresis.
The provision of chloride as sodium and potassium salts
to avoid volume depletion can prevent significant metabolic
alkalosis. The use of parenteral H2-receptor blockers will
limit gastric acid losses. Once alkalosis is present, both the
generative and maintenance phase of the alkalosis must
be resolved. It is critical to the gastric losses while also
replenishing plasma volume in these saline-responsive patients with sodium chloride and appropriate KC1 replacement.
Metabolic Alkalosis Associated with ECF Volume Expansion

The kidneys are responsible for both the generation and

the maintenance of metabolic alkalosis in this setting, but
the alkalosis is not responsive to saline. These patients have
urinary chloride excretions greater than 20 mEq/L.
The saline unresponsive alkaloses are generally divided
into those associated with or without hypertension. Further
categorization according to renin/aldosterone profiles may
also be helpful. The latter are generally not measured in
clinical practice; this characterization is most helpful as a
conceptual tool.
Hypertension w ith Mineralocorticoid Excess

Whether endogenous or exogenous, mineralocorticoid

excess is associated with enhanced salt reabsorption with
volume expansion, hypokalemia, kaliuresis, chloruresis, increased net acid excretion (enhanced ammoniagensis), and
decreased GFR. The severity of the metabolic alkalosis is
related to the severity of the potassium depletion.
High renin, mineralocorticoid excess states are seen most
commonly with renal vascular hypertension. Approximately 20% of patients with renal artery stenosis present
with severe hypokalemia and metabolic alkalosis. Malignant hypertension is generally associated with hyperreninemia due to renal ischemia and 10 to 20% of these cases are
accompanied by hypokalemic metabolic alkalosis. Reninsecreting tumors of juxtaglomerular apparatus origin are
associated with hypertension and metabolic alkalosis.
Low renin, mineralocorticoid excess states are most commonly seen with primary hyperaldosteronism (Conn's syndrome). Hypokalemia, metabolic alkalosis, hypernatremia,
and hypertension are hallmarks of this low renin state. It
accounts for less than 1 to 2% of all cases of hypertension.
Marked potassium depletion (as great as 1000 mEq has
been reported), together with the hyperaldosteronism, are
necessary to produce the alkalosis, because aldosterone
administration alone or hypokalemia alone only marginally
increase serum bicarbonate concentration. The serum bi-

carbonate concentration has been reported to be above

30 mEq/L in 85% of cases and above 35 mEq/L in 40%
of cases.
Hypertension with Low Renin and Aldosterone

States characterized by an excess of nonaldosterone mineralocorticoids or mineralocorticoid effect will produce

identical clinical findings as an excess of aldosterone, although aldosterone levels will be suppressed. The clinical
syndromes include Cushing's syndrome, exogenous steroid
administration, and adrenal enzyme deficiencies (11/3- and
17a-hydroxylase). Use of licorice containing glycyrrhizic
acid has also been associated with metabolic akalosis. Glycyrrhizic acid does not exert direct mineralocorticoid activity, rather it inhibits 11/3-hydroxysteroid dehydrogenase.
This enzyme is responsible for the conversion between
active steroid cortisol and inactive cortisone. Liddle's syndrome is characterized by mutations that result in constitutive activation of the amiloride-sensitive collecting duct
sodium channel that is ordinarily subject to stimulation by
mineralocorticoids.
Volume Expanded States with Low Renin
and Mineralocorticoid

Exogenous consumption of sodium bicarbonate, calcium

carbonate, baking soda, Turns (10 mEq of alkali per tablet)
or Rolaids (4.6 mEq per tablet) in the face of decreased
renal excretion of bicarbonate can cause metabolic alkalosis.
Consumption of poorly absorbable antacids (particularly
aluminum hydroxide gels used to control hyperphosphatemia or sucralf ate, together with sodium polystyrene sulfonate (to treat hyperkalemia, will generate bicarbonate from
the gastrointestinal tract. The resin binds cations in the gut
resulting in a more soluble form of bicarbonate which is
absorbed by the intestinal cell from the lumen. Renal failure perpetuates the alkalosis by impeding excretion of the
excess bicarbonate added from the gastrointestinal tract.
Large volume blood transfusion can result in metabolic
alkalosis. Each unit of whole blood contains 17 mEq of
sodium citrate; packed red cells contain 5 mEq. The administered citrate is converted to bicarbonate, resulting in alkalosis.
Milk-alkali syndrome is a rare but frequently mentioned
cause of metabolic alkalosis. The consumption of calciumcontaining absorbable antacids (Oscal, oyster shell calcium,
Turns, etc.) along with large quantities of milk (2-3 q/day)
can produce this disorder. Hypercalcemia, hypoparathyroidism, hypercalciuria, and alkalinuria favor renal
calcium-phosphorous deposition, resulting in renal failure.
DIAGNOSIS

The recognition and therapeutic intervention demand

appreciation of the underlying pathophysiology. A thorough history and physical examination are necessary, centering on potential sources of alkali; particular attention

10. Metabolic Alkalosis

85

should be given to the volume status of the patient. The

use of the change in bicarbonate and change in anion gap
or the so-called "delta-delta" gaps can uncover a metabolic
alkalosis. Acid added to the extracellular fluid dissociates,
yielding a proton and its related anion. The former is buffered by bicarbonate such that a rise in anion gap is reflected by an equimolar fall in bicarbonate. This results in
a delta anion to delta bicarbonate (delta-delta) of one. A
delta anion gap greater than the delta bicarbonate (presuming a normal starting bicarbonate) implies that excess alkali
was present, and thus infers the presence of a metabolic
alkalosis along with a metabolic acidosis. For example,
if a patient has a bicarbonate level of 20 mEq/L (delta
bicarbonate of 5 mEq/L) and an anion gap of 25 mEq/L,
the delta-delta ratio is 2:1 assuming a normal anion gap
of 12 mEq/L. Alternatively, to yield an anion gap of
25 mEq/L, 13 mEq/L of bicarbonate are titrated. The starting bicarbonate concentration must have been 33 mEq/L
(12 + 20). Either means of assessment implies that both a
metabolic alkalosis and acidosis are present.
Evaluation of urinary electrolytes to define causes of
metabolic alkalosis and inferentially plasma volume presupposes normal renal and adrenal function, and the absence of diuretics. Screening for the latter may be necessary.
A urinary chloride less than 15 mEq/L implies decreased
effective or absolute arterial blood volume or very low
salt intake. In the absence of edema, volume expansion
to restore plasma volume results in bicarbonaturia and
subsequent correction of the alkalosis as long as the genera tion of the alkalosis is also uncovered and corrected. In
the presence of edema and low urinary chloride, improving
EABV rather than saline infusions (which will only worsen
the edema) are necessary.
Urinary chloride concentrations greater than 20 mEq/L, in
the absence of polyuria, suggest that plasma volumes are
normal or increased. Obviously patients with salt-wasting
states in the face of volume depletion are an exception.
Such patients include those with diuretic use, salt-wasting
nephropathies, and nonreabsorbable anions in the urine
(the early and late phases of vomiting, or penicillin use).
If they are excluded on clinical grounds, mineralocorticoid
excess or Bartter's syndrome are likely culprits. Hypokalemic metatabolic alkalosis from occult vomiting or diuretic
consumption can be differentiated from Bartter's syndrome, but only with difficulty. Indeed some authorities
have referred to these two conditions as pseudo-Bartter's
syndrome. The urinary electrolytes are especially helpful
in differentiating between these states. >
TREATMENT

As discussed earlier, correcting the processes responsible

for the generation of the alkali is necessary. Antiemetics,
H2-blockers, stopping diuretics, and preventing and correcting hypokalemia are critical. Defining and reestablish-

ing plasma volume are mandatory. Rarely, severe alkalemia

can be treated with acetazolamide (250-500 mg) with appropriate replacement of saline and potassium.
Severe metabolic alkalosis in the setting of acute or
chronic renal failure, where bicarbonaturia (to correct the
alkalosis) is not anticipated, may require an acidifying agent
particularly in the symptomatic patient. The goal is to lower
arterial pH below 7.5. Hydrochloric acid, in a concentration
of 0.1 or 0.2 M (100-200 mEq/L) is infused into a central
vein. The HC1 distributes initially in the ECF. Therefore
the amount to be infused approximates 20% of body wt
times the predicted reduction in bicarbonate (in mEq/L)
necessary to lower pH. The frequent determination of arterial blood gases is mandatory. Finally, in the anuric patient,
an alternative is dialysis against dialysate with reduced
bicarbonate content.
Bibliography
Bastani B, Purcell H, Hemken P, et al.: Expression and distribution
of renal vacuolar proton -translocating adenosine triphosphatase
in response to chronic acid and alkali loads in the rat. J Clin
Invest 88:126, 1991.
Bettinelli A, Bianchetti MG, Borella P, et al: Genetic heterogeneity in tubular hypomagnesemia-hypokalemia with hypocalciuria
(Gitelman's syndrome). Kidney Int 47:547, 1995.
Carrneiro AV, Sebastian A, Cogan MG: Reduced glomerular nitration rate can maintain a rise in plasma bicarbonate concentration in humans. Am J Nephrol 7:450, 1987.
Cogan MG, Liu FY, Berger BE, et al: Metabolic alkalosis. Med
Clin North Am 67:903, 1983.
Gabow PA: Disorders associated with an altered anion gap. Kidney
Int 27:472, 1985.
Halperin ML, Kamel KS, Narins RG: Use of urine electrolytes
and osmolality: Bringing physiology to the bedside. In: Narins
RG, Stein JH. Diagnostic Techniques in Renal Diseases. Vol.
52. Contemporary Issues in Nephrology. Churchill Livingstone,
New York, p. 9, 1992.
Hodgkin JE, Soepranoff, Chan DM: Incidence of metabolic alkalemia in hospitalized patients. Crit Care Med 8:725, 1980.
Kassirer JP, London AM, Goldman DM, et al: On the pathogenesis
of metabolic alkalosis in hyperaldosteronism. Am J Med
49:306, 1970.
Koch SM, Taylor RW: Chloride ion in intensive care medicine.
Crit Care Med 20:227, 1992.
Narins RG, Jones ER, Townsend R: Metabolic Acid Base Disorders: Pathophysiology, Classification, and Treatment. Fluid, Electrolytes and Acid Base Disorders. Churchill Livingstone, New
York, pp. 335-385, 1985.
Norris SH, Kurtzman NA: Does chloride play an independent role
in the pathogenesis of metabolic alkalosis. Semin Nephrology
8:101, 1988.
Seldin DW, Rector FC Jr: The generation and maintenance of
metabolic alkalosis. Kidney Int 1:305, 1972.
Stone DK, Xie XS: Proton translocating ATPases: Issues in struc ture and function. Kidney Int 33:767, 1988.
Tsukamoto T, Kobayashi T, Kawamoto K, et al: Possible Discrimination of Gitelman's Syndrome from Bartter's Syndrome by
renal clearance study: Report of two cases. Am J Kidney Dis
25:637, 1995.

RESPIRATORY ACIDOSIS AND ALKALOSIS

NICOLAOS E. MADIAS

RESPIRATORY ACIDOSIS

equation, it is not possible for pCO2 to reach values much

higher than 80 mm Hg while the level of pO2 is still compatible with life. Extreme hypercapnia can only be seen during
oxygen administration and, in fact, is often the result of
uncontrolled oxygen therapy.

Respiratory acidosis, or primary hypercapnia, is the

acid-base disturbance initiated by an increase in carbon
dioxide tension of body fluids. Hypercapnia acidifies body
fluids and elicits an adaptive increment in plasma bicarbonate that should be viewed as an integral part of the respiratory acidosis. Arterial carbon dioxide tension (pCO2) measured at rest and at sea level is greater than 45 mm Hg in
simple respiratory acidosis. Lower values of pCO2 might
still signify the presence of primary hypercapnia in the
setting of mixed acid-base disorders (e.g., eucapnia, rather
than the expected hypocapnia, in the presence of metabolic acidosis).

Secondary Physiological Response

An immediate increment in plasma bicarbonate concentration that is accounted for by titration of nonbicarbonate
body buffers occurs in response to acute hypercapnia. This
adaptation is complete within 5 to 10 minutes from the
rise in pCO2. On average, plasma bicarbonate increases by
about 0.1 mEq/L for each mm Hg acute increment in/?CO2;
as a result, plasma hydrogen ion concentration rises by
about 0.75 nEq/L for each mm Hg acute rise in pCO2.
Therefore, the overall limit of adaptation of plasma bicarbonate in acute respiratory acidosis is quite small; even
whenpCO2 rises to levels of 80 to 90 mm Hg, the increment
in plasma bicarbonate does not exceed 3 to 4 mEq/L. Moderate hypoxemia does not alter the adaptive response to
acute respiratory acidosis. On the other hand, preexisting
hypobicarbonatemia (whether due to metabolic acidosis or
chronic respiratory alkalosis) enhances the magnitude of
the bicarbonate response to acute hypercapnia, whereas
such a response is diminished in hyperbicarbonatemic
states (whether due to metabolic alkalosis or chronic respiratory acidosis). Other electrolyte changes observed in
acute respiratory acidosis include a mild rise in plasma
sodium (1 to 4 mEq/L), potassium (0.1 mEq/L for each 0.1
unit fall in pH), and phosphorus, and a small decrease in
plasma chloride and lactate concentrations. A small reduction in the plasma anion gap is also observed, reflecting
the fall in plasma lactate and the acidic titration of
plasma proteins.
The adaptive increase in plasma bicarbonate concentration observed in the acute phase of hypercapnia is amplified
markedly during chronic hypercapnia as a result of generation of new bicarbonate by the kidneys. Both proximal and
distal acidification mechanisms contribute to this adaptation, which requires 3 to 5 days for completion. The renal
response to chronic hypercapnia includes chloruresis and
generation of hypochloremia. On average, plasma bicarbonate increases by about 0.3 mEq/L for each mm Hg
chronic increment in pCO2; as a result, plasma hydrogen
ion concentration rises by about 0.3 nEq/L for each mm

Pathophysiology

Hypercapnia develops whenever carbon dioxide excre tion by the lungs is insufficient to match carbon dioxide
production, thus leading to positive carbon dioxide balance.
Hypercapnia could result from increased carbon dioxide
production, decreased alveolar ventilation, or both. Overproduction of carbon dioxide is usually matched by increased excretion such that generation of hypercapnia is
prevented. However, patients with marked limitation in
pulmonary reserve and those receiving constant mechanical
ventilation might experience respiratory acidosis due to
increased carbon dioxide production. Established clinical
circumstances include increased physical activity, augmented work of breathing by the respiratory muscles, shivering, seizures, fever, and hyperthyroidism. Increments in
carbon dioxide production might also be imposed by the
administration of large carbohydrate loads (greater than
2000 kcal/day) and parenteral nutrition to semistarved, critically ill patients as well as during the decomposition of
bicarbonate infused in the course of treating metabolic
acidosis. By far, most cases of respiratory acidosis reflect
a decrease in alveolar ventilation. Decreased alveolar ventilation can result from decreased minute ventilation, increased dead space ventilation, or a combination of both.
The major threat from carbon dioxide retention in patients breathing room air is the associated obligatory hypoxemia. Thus, in the absence of supplemental oxygen,
patients suffering respiratory arrest develop critical hypoxemia within a few minutes, long before extreme hypercapnia ensues. Because of the constraints of the alveolar gas

Primer on Kidney Diseases, Second Edition

86

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

11. Respiratory Acidosis and Alkalosis

87

Hg chronic rise in pCO2. Empirical observations indicate

a limit of adaptation of plasma bicarbonate on the order
of 45 mEq/L. The renal response to chronic hypercapnia
is not altered appreciably by dietary sodium or chloride
restriction, moderate potassium depletion, alkali loading,
or moderate hypoxemia. It is currently unknown to what
extent renal insufficiency of variable severity limits the
renal response to chronic hypercapnia. Obviously, patients
with end stage renal disease cannot mount a renal response
to chronic hypercapnia and, thus, they are subject to severe
acidemia. The degree of acidemia is more pronounced in
patients receiving hemodialysis rather than peritoneal dialysis because the former treatment maintains, on average,
a lower plasma bicarbonate concentration. Recovery from
chronic hypercapnia is crippled by a chloride-deficient diet.
In this circumstance, despite correction of the level of
pCC>2, plasma bicarbonate concentration remains elevated
as long as the state of chloride deprivation persists, thus
creating the entity of "posthypercapnic metabolic alkalosis." Chronic hypercapnia is not associated with appreciable changes in the plasma concentrations of sodium, potas sium, phosphorus, or anion gap.

Etiology

Respiratory acidosis can develop in patients with normal

or abnormal airways and lungs. Tables 1 and 2 present
causes of acute and chronic respiratory acidosis, respectively. This classification takes into consideration the usual
mode of onset and duration of the various causes and
emphasizes the biphasic time course that characterizes the
secondary physiological response to hypercapnia. Primary
hypercapnia can result from disease or malfunction within
any element of the regulatory system controlling respiration, including the central and peripheral nervous system,
the respiratory muscles, the thoracic cage, the pleural space,
the airways, and the lung parenchyma. Not infrequently,
more than one cause contributes to the development of
respiratory acidosis in a given patient. Chronic lower airways obstruction resulting from bronchitis and emphysema
is the most common cause of chronic hypercapnia.
Clinical Manifestations

Clinical manifestations of respiratory acidosis arising

from the central nervous system are collectively known

TABLE I

Causes of Acute Respiratory Acidosis 0

Normal airways and lungs

Abnormal airways and lungs

Central nervous system depression

General anesthesia Sedative
overdosage Head trauma
Cerebrovascular accident Central
sleep apnea Cerebral edema Brain
tumor Encephalitis

Upper airways obstruction

Coma-induced hypopharyngeal obstruction
Aspiration of foreign body or vomitus
Laryngospasm or angioedema Obstructive
sleep apnea Inadequate laryngeal intubation
Laryngeal obstruction postintubation

Neuromuscular impairment High spinal

cord injury Guillain -Barre syndrome
Status epilepticus Botulism, tetanus
Crisis in myasthenia gravis Hypokalemic
myopathy Familial hypokalemic periodic
paralysis
Drugs or toxic agents (e.g., curare, succinylcholine, aminoglycosides,
organophosphorus)

Lower airways obstruction

Generalized bronchospasm Severe
asthma Bronchiolitis of infancy and
adults

Ventilatory restriction Rib fractures with flail chest Pneumothorax

Hemothorax Impaired diaphragmatic function (e.g., peritoneal
dialysis, ascites)

Pulmonary perfusion defect Cardiac arrest

Severe circulatory failure Massive
pulmonary thromboembolism Fat or air
embolism

latrogenic events Misplacement or displacement of airway cannula during

anesthesia or
mechanical ventilation
Bronchoscopy-associated hypoventilation or respiratory arrest Increased
CO2 production with constant mechanical ventilation (e.g., due
to high carbohydrate diet or sorbent-regenerative hemodialysis)
'Adapted from Madias and Adrogue (1991).

Disorders involving pulmonary alveoli Severe

bilateral pneumonia Infant or adult
respiratory distress syndrome Severe
pulmonary edema

88

Nicolaos E. Madias
TABLE 2

Causes of Chronic Respiratory Acidosis"

Normal airways and lungs

Abnormal airways and lungs

Central nervous system depression

Sedative overdosage
Methadone/heroin addiction
Primary alveolar hypoventilation (Ondine's curse)
Obesity-hypoventilation syndrome (Pickwickian
syndrome) Brain
tumor
Bulbar
poliomyelitis

Upper airways obstruction Tonsillar and

peritonsillar hypertrophy Paralysis of vocal
cords Tumor of the cords or larynx Airway
stenosis post prolonged intubation Thymoma,
aortic aneurysm

Neuromuscular impairment
Poliomyelitis Multiple sclerosis
Muscular dystrophy
Amyotrophic lateral sclerosis
Diaphragmatic paralysis
Myxedema Myopathic disease
Ventilatory restriction Kyphoscoliosis,
spinal arthritis Obesity Fibrothorax
Hydrothorax Impaired diaphragmatic
function

Disorders involving pulmonary alveoli

Severe chronic pneumonitis
Diffuse infiltrative disease (e.g., alveolar proteinosis)
Interstitial fibrosis

Lower airways obstruction

Chronic obstructive lung disease (bronchitis, bronchiolitis, bronchiectasis,
emphysema)

" Adapted from Madias and Adroguc (1991).

as "hypercapnic encephalopathy" and include irritability,

inability to concentrate, headache, anorexia, mental cloudiness, apathy, confusion, incoherence, combativeness, hallucinations, delirium, and transient psychosis. Progressive
narcosis or coma might develop in patients receiving oxy gen therapy, especially those with an acute exacerbation
of chronic respiratory insufficiency in whom pCO2 levels
up to 100 mm Hg or even higher can occur. In addition,
frank papilledema (pseudotumor cerebri) and motor dis turbances, including myoclonic jerks, napping tremor identical to that observed in liver failure, and seizures might
develop. The occurrence of neurological symptomatology
in patients with respiratory acidosis depends on the magnitude of the hypercapnia, the rapidity with which it develops,
the severity of the acidemia, and the degree of accompanying hypoxemia.
The hemodynamic consequences of respiratory acidosis
reflect a variety of mechanisms, including a direct depressing effect on myocardial contractility. An associated sympathetic surge, sometimes intense, leads to increases in plasma
catecholamines, but, during severe acidemia (generally
blood pH below 7.20), receptor responsiveness to catecholamines is markedly blunted. Hypercapnia results in systemic vasodilation by a direct action on vascular smooth
muscle; this effect is most obvious in the cerebral circulation
where blood flow increases in direct relation to the level
of pCQ?. By contrast, carbon dioxide retention can produce
vasoconstriction in the pulmonary circulation as well as
the kidneys; in the latter case, the hemodynamic response
might be mediated via an enhanced sympathetic activity.

The composite effect of these inputs is such that mild to

moderate hypercapnia is usually associated with an increased cardiac output, normal or increased blood pressure,
warm skin, a bounding pulse, and diaphoresis. However,
when hypercapnia is severe or considerable hypoxemia is
present, decreases in both cardiac output and blood pres sure might be observed. Concomitant therapy with vasoactive medications or the presence of congestive heart failure might further modify the hemodynamic response.
Cardiac arrhythmias, particularly supraventricular tachyarrhythmias not associated with major hemodynamic compromise, are common. They do not result primarily from
the hypercapnia, but rather reflect the associated hypoxemia and sympathetic discharge, concomitant medication,
electrolyte abnormalities, and underlying cardiac disease.
Salt and water retention is commonly observed in sustained
hypercapnia, especially in the presence of cor pulmonale.
In addition to the effects of heart failure on the kidney,
multiple other factors might be involved, including the prevailing stimulation of the sympathetic nervous system and
the renin-angiotensin-aldosterone axis, the increased renal
vascular resistance, and the elevated levels of antidiuretic
hormone and cortisol.
Diagnosis
In general, one never should rely on clinical evaluation
alone to assess the adequacy of alveolar ventilation. Whenever hypoventilation is suspected, arterial blood gases
should be obtained. If the patient's acid -base profile re -

I. Respiratory Acidosis and Alkalosis

veals hypercapnia in association with acidemia, at least an
element of respiratory acidosis must be present. However,
hypercapnia might be associated with a normal or an alkaline pH because of the simultaneous presence of additional
acid-base disorders. Information from the patient's his tory, physical examination, and ancillary laboratory data
should be utilized for an accurate assessment of the acidbase status.
Therapeutic Principles

Treatment of acute respiratory acidosis must be directed

at prompt removal of the underlying cause whenever possible. Immediate therapeutic efforts should focus on establishing and securing a patent airway, restoring adequate
oxygenation by delivering an oxygen-rich inspired mixture,
and providing adequate ventilation in order to repair the
abnormal gas composition. As noted, acute respiratory acidosis poses its major threat to survival not because of hypercapnia or acidemia, but because of the associated hypoxemia. Swift restoration of pCC>2 to normal levels is the
therapeutic goal. Mechanical ventilation should be employed if conservative therapy fails to correct hypoxemia
and hypercapnia promptly. Despite its overall merit, nonin vasive oximetry should not substitute for arterial blood gas
measurements in titrating FjO2. The presence of a component of metabolic acidosis is the primary indication for
alkali therapy in patients with acute respiratory acidosis.
Administration of sodium bicarbonate to patients with simple respiratory acidosis is not only of questionable efficacy,
but also involves considerable risk. Concerns include pHmediated depression of ventilation, enhanced carbon dioxide production due to bicarbonate decomposition, and volume expansion. Alkali therapy might have a special role
in patients with severe bronchospasm by restoring the responsiveness of the bronchial musculature to /3-adrenergic
agonists. Successful management of intractable asthma in
patients with blood pH below 7.00 by administering sufficient sodium bicarbonate to raise blood pH to above 7.20
has been reported.
Recent evidence indicating that lung overdistention can
induce lung injury has led to a new strategy for management
of disorders requiring mechanical ventilation, such as adult
respiratory distress syndrome and severe airflow obstruction. The strategy entails prescription of tidal volumes of
5 to 7 mL/kg body wt (instead of the conventional level of
10 to 15 mL/kg body wt) to achieve a plateau airway pres sure of s30 cm HoO. Because an increase in pCO2 might
develop, this approach is termed permissive hypercapnia.
If the resultant hypercapnia reduces blood pH below 7.15
to 7.20, some physicians prescribe bicarbonate; however,
this strategy is controversial and others intervene only for
pH values on the order of 7.00. Uncontrolled observations
suggest that permissive hypercapnia affords improved clinical outcomes but controlled studies are lacking. Heavy
sedation and neuromuscular blockade are frequently
needed with this therapy. Following discontinuation of neuromuscular blockade, some patients develop prolonged
weakness or paralysis. Contraindications to permissive hy-

89
percapnia include cerebral edema, increased intracranial
pressure, and convulsions; depressed cardiac function and
arrhythmias; and severe pulmonary hypertension.
Unfortunately, only rarely can one remove the underlying cause of chronic respiratory acidosis. Nonetheless, maximizing alveolar ventilation with relatively simple maneuvers is often rewarding. Such maneuvers include treatment
with antibiotics, bronchodilators, or diuretics; avoidance of
irritant inhalants, tranquilizers, or sedatives; and elimination of retained secretions. Administration of adequate
quantities of chloride (usually as the potassium salt) prevents or corrects a complicating element of metabolic alkalosis (commonly diuretic induced) that can further dampen
the ventilatory drive. Acetazolamide might be used as an
adjunctive measure but care must be taken to avoid potas sium depletion. Potassium and phosphate depletion should
be corrected, because they might contribute to the development or the maintenance of respiratory failure by impairing
the function of skeletal muscles. The use of pharmacologic
stimulants of ventilation has been generally disappointing
but a measure of benefit can be derived by some patients
with central sleep apnea, obesity-hypoventilation syndrome, or chronic obstructive lung disease. In contrast to
acute respiratory acidosis, administration of oxygen to patients with long-standing CC>2 retention should be carried
out cautiously aiming at apO2 of about 60 mm Hg. Injudicious use of oxygen therapy can produce further reductions
in alveolar ventilation and aggravate hypercapnia dramatically. Restoration of the chronically elevated pCO2 to or
toward normal via assisted ventilation, if indicated, should
proceed gradually over many hours to a few days. Overly
rapid reduction in pCO2 in such patients risks the development of sudden, substantial alkalemia with the attendant
hazards of a major reduction in cardiac output and cerebral
blood flow, arrhythmias (including predisposition to digitalis intoxication), and generalized seizures. In the absence
of a complicating element of metabolic acidosis and with
the possible exception of the severely acidemic patient
with intense generalized bronchoconstriction undergoing
mechanical ventilation, there is no role for alkali adminis tration in chronic respiratory acidosis.
RESPIRATORY ALKALOSIS

Respiratory alkalosis, or primary hypocapnia, is the

acid-base disturbance initiated by a reduction in carbon
dioxide tension of body fluids. Hypocapnia alkalinizes body
fluids and elicits an adaptive decrement in plasma bicarbonate that should be viewed as an integral part of the respiratory alkalosis. The level of pCO2 measured at rest and at
sea level is lower than 35 mm Hg in simple respiratory
alkalosis. Higher values of pCO2 might still indicate the
presence of an element of primary hypocapnia in the setting
of mixed acid-base disorders (e.g., eucapnia, rather than
the anticipated hypercapnia, in the presence of meta bolic alkalosis).
Pathophysiology

Primary hypocapnia most commonly reflects pulmonary

hyperventilation due to increased ventilatory drive. The

90

Nicolaos E. Madias

latter results from signals arising from the lung, the peripheral chemoreceptors (carotid and aortic), the brain stem
chemoreceptors, or influences originating in other centers
of the brain. Additional mechanisms for the generation of
primary hypocapnia include maladjusted mechanical ventilators, the extrapulmonary elimination of carbon dioxide
by a dialysis device or extracorporeal circulation (e.g.,
heart-lung machine), and decreased carbon dioxide production (e.g., due to sedation, skeletal muscle paralysis,
hypothermia) in patients receiving constant mechanical
ventilation.
A condition termed pseudorespiratory alkalosis occurs
in patients with profound decrements in cardiac output
but relative preservation of respiratory function. In these
patients, there is venous (and tissue) hypercapnia due to
prolongation of transit time, resulting in a greater than
normal addition of carbon dioxide per unit of blood travers ing the systemic capillaries. On the other hand, arterial
blood evidences hypocapnia due to increased ventilationto-perfusion ratio causing increased remo val of carbon
dioxide per unit of blood traversing the pulmonary circulation. Absolute carbon dioxide excretion is decreased,
however, and body carbon dioxide balance is positive.
Therefore, respiratory acidosis, rather than respiratory alkalosis, is present. In addition, arterial blood might reveal
normoxia or hyperoxia despite the presence of severe hypoxemia in venous blood. Thus, both arterial and mixed (or
central) venous blood sampling is needed to assess the
acid-base status and oxygenation of patients with critical
hemodynamic compromise.
Secondary Physiological Response
Adaptation to acute hypocapnia is characterized by an
immediate decrement in plasma bicarbonate that is accounted for principally by titration of nonbicarbonate body
buffers. This adaptation is completed within 5 to 10 minutes
after the onset of hypocapnia. Plasma bicarbonate falls, on
average, by approximately 0.2 mEq/L for each mm Hg
acute decrement in pCO2; consequently, plasma hydrogen
ion concentration decreases by about 0.75 nEq/L for each
mm Hg acute reduction in pCCX The limit of this adaptation of plasma bicarbonate is on the order of 17 to 18 mEq/
L. Concomitant small rises in plasma chloride, lactate, and
other unmeasured anions balance the fall in plasma bicarbonate; each of these components accounts for about one
third of the bicarbonate decrement. Small decreases in
plasma sodium (1-3 mEq/L) and potassium (0.2 mEq/L
for each 0.1 unit rise in pH) might be observed. Severe
hypophosphatemia can occur in acute hypocapnia due to
translocation of phosphorus into the cells.
A larger decrement in plasma bicarbonate occurs in
chronic hypocapnia as a result of renal adaptation to the
disorder and involves suppression of both proximal and
distal acidification mechanisms. Completion of this adaptation requires 2 to 3 days. Plasma bicarbonate decreases,
on average, by about 0.4 mEq/L for each mm Hg chronic
decrement in pCO2', as a consequence, plasma hydrogen
ion concentration decreases by approximately 0.4 nEq/L

for each mm Hg chronic reduction in pCO2. The limit of

this adaptation of plasma bicarbonate is on the order of
12 to 15 mEq/L. About two thirds of the fall in plasma
bicarbonate is balanced by a rise in plasma chloride concentration, the remainder reflects an increase in plasma unmeasured anions; part of the remainder is due to the alkaline titration of plasma proteins, but most remains
undefined. Plasma lactate does not rise in chronic hypocapnia, even in the presence of moderate hypoxemia. Similarly,
no appreciable change in the plasma concentration of sodium occurs. In sharp contrast with acute hypocapnia, the
plasma concentration of phosphorus remains essentially
unchanged in chronic hypocapnia. Although plasma potas sium is in the normal range in patients with chronic hypocapnia at sea level, hypokalemia and renal potassium wasting have been described in subjects in whom sustained
hypocapnia was induced by exposure to high altitude. Patients with end stage renal disease are obviously at risk of
developing severe alkalemia in response to chronic hypocapnia because they cannot mount a renal response. Such
a risk is higher in patients receiving peritoneal dialysis
rather than hemodialysis because the former treatment
maintains, on average, a higher plasma bicarbonate concentration.
Etiology
Primary hypocapnia is the most frequent acid-base dis turbance encountered, occurring in normal pregnancy and
high-altitude residence. Table 3 lists the major causes of
respiratory alkalosis. Most are associated with the abrupt
appearance of hypocapnia but, in many instances, the process might be sufficiently prolonged to permit full chronic
adaptation to occur. Consequently, no attempt has been
made to separate these conditions into acute and chronic
categories. Some of the major causes of respiratory alkalosis are benign, whereas others are life threatening. Primary
hypocapnia is particularly common among the critically ill,
occurring either as the simple disorder or as a component
of mixed disturbances. Its presence constitutes an ominous
prognostic sign, with mortality increasing in direct proportion to the severity of the hypocapnia.
Clinical Manifestations
Rapid decrements in pCC>2 to half the normal values or
lower are typically accompanied by paresthesias of the
extremities, chest discomfort, circumoral numbness, lightheadedness, confusion, and infrequently, tetany or generalized seizures. These manifestations are seldom present in
the chronic phase. Acute hypocapnia decreases cerebral
blood flow, which, in severe cases, might reach values less
than 50% of normal, resulting in cerebral hypoxia. This
hypoperfusion has been implicated in the pathogenesis of
the neurological manifestations of acute respiratory alkalosis along with other factors, including hypocapnia per se,
alkalemia, pH-induced shift of the oxyhemoglobin dissociation curve, and decrements in the level of ionized calcium

11. Respiratory Acidosis and Alkalosis

91
Diagnosis

TABLE 3

Causes of Respiratory Alkalosis0

Hypoxemia or tissue hypoxia
Central nervous system
Decreased inspired O2 tension
stimulation
High altitude
Voluntary
Bacterial or viral pneumonia
Pain
Aspiration of food, foreign
Anxiety
body, or vomitus
Psychosis
Laryngospasm
Fever
Drowning
Subarachnoid hemorrhage
Cyanotic heart disease Severe
Cerebrovascular accident
anemia Left shift deviation of
Meningoencephalitis
HbO 2
Tumor
curve
Trauma
Hypotension* Severe
Drugs or hormones
circulatory failure* Pulmonary
Nikethamide, ethamivan
edema
Doxapram
Stimulation of chest receptors
Xanthines
Pneumonia
Salicylates
Asthma
Catecholamines
Pneumothorax
Angiotensin II
Hemothorax
Vasopressor agents
Flail chest
Progesterone
Infant or adult respiratory
Medroxyprogesterone
distress syndrome Cardiac
Dinitrophenol
failure Noncardiogenic
Nicotine
pulmonary
Miscellaneous
edema
Pregnancy
Pulmonary embolism
Sepsis
Interstitial lung disease
Hepatic failure Mechanical
hyperventilation Heat
exposure Recovery from
metabolic acidosis
"Adapted from Madias and Adrogue (1991).
* Might produce pseudorespiratory alkalosis.

and potassium. Some evidence indicates that cerebral blood

flow returns to normal in chronic respiratory alkalosis.
Actively hyperventilating patients manifest no appreciable changes in cardiac output or systemic blood pressure.
By contrast, acute hypocapnia in the course of passive
hyperventilation typically observed during mechanical ventilation in patients with depressed central nervous system
or those under general anesthesia, frequently results in
a major reduction in cardiac output and systemic blood
pressure, increased peripheral resistance, and substantial
hyperlactatemia (exceeding 2 mEq/L). Neither cardiac arrhythmias nor clinically evident signs or symptoms of coronary insufficiency develop in actively hyperventilating
normal volunteers. However, patients with underlying coronary artery disease might occasionally suffer hypocapniainduced coronary vasoconstriction resulting in angina pectoris, ischemic electrocardiographic changes, and arrhythmias. Increased cardiac excitability has been attributed to
the same factors that have been held responsible for the
enhanced excitability of the central nervous system associated with acute hypocapnia.

Careful observation can detect abnormal patterns of

breathing in some patients, yet marked hypocapnia can be
present without a clinically evident increase in respiratory
effort. Thus, arterial blood gases should be obtained whenever hyperventilation is suspected. In fact, the diagnosis of
respiratory alkalosis, especially the chronic form, is frequently missed; physicians often misinterpret the electro lyte pattern of hyperchloremic hypobicarbonatemia as indicative of normal anion gap metabolic acidosis. If the
patient's acid-base profile reveals hypocapnia in association with alkalemia, at least an element of respiratory alkalosis must be present. Primary hypocapnia, however, might
be associated with a normal or an acidic pH due to the
concomitant presence of other acid-base disorders. Notably, mild degrees of chronic hypocapnia commonly leave
blood pH within the high-normal range. As always, proper
evaluation of the patient's acid-base status requires careful
assessment of the history, physical examination, and ancillary laboratory data. Once the diagnosis of respiratory alkalosis has been made, a search for its cause should be carried
out. The diagnosis of respiratory alkalosis can have important clinical implications: It often provides a clue to the
presence of an unrecognized, serious disorder (e.g., sepsis)
or signals the severity of a known underlying disease.
Therapeutic Principles

Management of respiratory alkalosis must be directed

toward correcting the underlying cause, whenever possible.
Taking measures to treat the respiratory alkalosis itself is
commonly not required, because the disorder, especially
in its chronic form, leads to minimal or no symptoms and
poses little risk to health. A notable exception is the patient
with the anxiety-hyperventilation syndrome; in addition to
reassurance or sedation, rebreathing into a closed system
(e.g., a paper bag) might prove helpful by interrupting the
vicious cycle that can result from the reinforcing effects of
the symptoms of hypocapnia. Considering the risks of severe alkalemia, sedation or, in rare cases, skeletal muscle
paralysis and mechanical ventilation might be required to
correct temporarily marked respiratory alkalosis. Management of pseudorespiratory alkalosis must be directed at
optimizing systemic hemodynamics.
Bibliography
Adrogue HJ, Rashad MN, Gorin AB, Yacoub J, Madias NE:
Assessing acid-base status in circulatory failure. Differences between arterial and central venous blood. N EnglJ Med 320:13121316, 1989.
Al-Awqati Q: The cellular renal response to respiratory acid-base
disorders. Kidney Int 28:845-855, 1985.
Arbus GS, Hebert LA, Levesque PR, Etsten BE, Schwartz WB:
Characterization and clinical application of the "significance
band" for acute respiratory alkalosis. A' Engl J Med 280:117123, 1969.
Brackett NC Jr, Cohen JJ, Schwartz WB: Carbon dioxide titration
curve of normal man. Effect of increasing degrees of acute
hypercapnia on acid- base equilibrium. N Engl J Med 272:612, 1965.

Martin Goldberg

92
Brackett NC Jr, Wingo CF, Muren O, Solano JT: Acid- base response to chronic hypercapnia in man. N Engl J Med 280:124130, 1969.
Feihl F, Ferret C: Permissive hypercapnia. How permissive should
we be? Am J Respir Crit Care Med 150:1722-1737, 1994.
Gennari FJ, Kassirer JP: Respiratory alkalosis. In: Cohen JJ, Kassirer JP (eds) Acid -Base. Little, Brown, Boston, pp. 349-376,
1982.
Krapf R, Beeler I, Hertner D, Hulter HN: Chronic respiratory
alkalosis. The effect of sustained hypcrventilation on renal regulation of acid-base equilibrium. N Engl J Med 324:1394-1401,
1991.

Madias NE, Adrogue HJ. Respiratory acidosis and alkalo sis. In:
Adrogue JH (ed) Contemporary Management in Critical Care:
Acid-Base and Electrolyte Disorders. Churchill Livingstone,
New York, pp. 37-53, 1991.
Madias NE, Cohen JJ: Respiratory acidosis. In: Cohen JJ, Kassirer
JP (eds) Acid-Base. Little, Brown, Boston, pp. 307 -348, 1982.
Madias NE, Cohen JJ: Adaptation to respiratory acidosis and
alkalosis. In: Fishman AP (ed) Pulmonary Diseases and Disorders. 2nd ed. McGraw-Hill Book Company, New York, Vol. 1,
pp. 289-298, 1988.
Madias NE, Wolf CJ, Cohen JJ: Regulation of acid-base equilibrium in chronic hypercapnia. Kidney Int 27:538-543, 1985.

APPROACH TO ACID-BASE DISORDERS

MARTIN GOLDBERG

Acid-base disorders occur commonly in medical and

surgical patients. They are particularly important in severely ill hospitalized patients. They may be present as
single (simple) disorders or as a combination of simple
disorders (mixed acid-base disturbances). Systematic diagnostic recognition and correction of these simple or mixed
disorders is important because they may not only impact
on the patient's prognosis, but also may provide clues to
the nature of the underlying primary disease(s). Before
discussing a systematic approach to the diagnosis of acidbase disorders, several fundamental terms require definition.
ACIDEMIA VS ALKALEMIA

These terms represent abnormal hydrogen ion concentrations of blood: either higher (acidemia) or lower (alkalemia) than the normal range of 35 to 45 nEq/L(pH = 7.357.45).
[H+ ] = hydrogen ion concentration pH =
log 1/[H+ ] = -log [H+ ]

Blood pH = 6.10 + log (24/0.03 X 40)

= 6.10 + log (20/1) = 7.40.
The Henderson equation (modified) is
[H+ ] = 24 X pCO2/[HCO5].
Normally,
[H+ ] = 24 X (40/24) = 40 nEq/L.
ACIDOSIS VS ALKALOSIS

Alkalosis is a primary process (e.g., loss of H+ from

vomiting) which, if unopposed, would produce alkalemia.
Acidosis is a primary process (e.g., retention of [H+ ] due
to renal failure), which, if unopposed, would produce acidemia.
These are pathophysiological processes or abnormal
states which, if unopposed by therapy or disease, would
cause deviations of extracellular fluid [H+ ] from normal
levels. They are defined independently of the [H+ ] or pH
because two or more concomitant processes may produce
either no net change in [H+ ] and/or may modify the body's
adaptive mechanisms to a single disorder.

The Henderson-Hasselbalch equation is

RESPIRATORY VS METABOLIC DISTURBANCES

pH = pK + log ([HCO5l/[0.03 X pCO2).

Normally,

Primer on Kidney Diseases, Second Edition

Respiratory disturbances are those caused by abnormal pulmonary elimination of CO2, producing an excess
Copyright 1998 by the National Kidney Foundation
Alt rights of reproduction in any form reserved.

12. Approach to Acid-Base Disorders

(acidosis) or deficit (alkalosis) of H2CO3 (in equilibrium

with/?CO2) in extracellular fluid. These primarily alter the
pCO2 in the denominator of the Henderson-Haseelbalch
equation. Compensatory (adaptive) adjustments involve
changes in [HCOa] accumulation in body fluids. Adaptation via the kidneys is slow (3-4 days); hence there are
operationally four respiratory disorders: acute and chronic
respiratory acidosis; acute and chronic respiratory alkalosis.
Metabolic disturbances are those caused by excessive
intake, metabolic production, or losses of fixed (nonvolatile) acids or bases in the extracellular fluid. These are
reflected by changes in [HCOi] in blood and therefore alter primarily the numerator of the HendersonHasselbalch ratio. Adaptation to metabolic disorders is
relatively rapid (12-36 hours); hence there are two: metabolic acidosis (a process which lowers plasma [HCO])
and metabolic alkalosis (a process which raises plasma
[HC05]).
COMPENSATORY (ADAPTIVE) RESPONSES

Primary changes in the metabolic component ([HCOi])

stimulate adaptive changes in ventilation, producing
changes in pCO2. Additional adaptive changes occur in
extracellular and intracellular buffers and in the kidney
(adjustments in H+ secretion, HCOi reabsorption and secretion, and generation of HCOi).
Primary changes in the respiratory component (pCO2)
stimulate adaptive changes in [HCOs] via reactions with
extra- and intracellular buffers and by slower (renal) adjustments in H+ secretion, HCOs reabsorption and secretion,
and HCOi generation.
As a rule, compensation restores [H+ ] or pH toward
normal, but not to complete normality.
SIMPLE (SINGLE) VS MIXED DISORDERS

A simple disorder includes the primary process with the

initial changes in [H+ ],pCO2 or [HCOs] and all compensatory processes in reaction to these initial changes.
A mixed disorder is the simultaneous occurrence of two
or more simple disturbances in a patient. Mixed disorders
may be additive or counterbalancing regarding their net
effect on [H+ ] or pH; they are frequently difficult to diagnose and reflect serious illness. A mixed disorder in which
there are three simultaneous primary acid-base disorders
present is commonly referred to as a triple acid-base disturbance.
ACID-BASE MAPS AND FORMULAS IN DIAGNOSIS OF
ACID-BASE DISORDERS

The adaptive responses for the six simple disorders (including the acute and chronic respiratory disorders) have
been quantified experimentally. The 95% confidence limits
can be defined, and several formulas have been developed
to reflect the ranges of compensation. The acid-base map

93
provides the graphic representation of these ranges. (See
Table 1 and Fig. 1).
In a patient with a clinical condition suggesting a simple
acid-base disorder, values of pH, pCO2, and [HCOs] lying
within the ranges defined by the formulas or map are compatible with the diagnosis of the specified simple disorder.
This does not, however, rule out the possibility of a mixed
disorder due to two or more counterbalancing processes,
the net effects of which might produce values lying in the
normal range, or in an area of a simple disorder. Furthermore, whereas values lying outside the range of a simple
disorder suggest the diagnosis of a mixed disorder, a mixed
disorder may sometimes be simulated during a transient
state in which the adaptive processes to a simple disorder
have not yet been completed.
ANION GAP

The anion gap (AG) is the concentration of unmeasured

anions (proteinate, phosphates, sulfates, organic acid
anions) which are typically associated with H+ as it is initially generated in body fluids. It is commonly calculated as
AG = plasma [Na + ] - (plasma [CT] +
plasma [HCOi]).
Normal values for AG are 6-13 mEq/L. An AG >
15 mEq/L generally indicates one type of metabolic acidosis
(i.e., due to accumulation of organic acids as occurs in
lactic acidosis, diabetic ketoacidosis, or ingested toxins). An
exception to this occurs during infusions of fluids containing
salts of organic anions (e.g., lactate, amino acids, high doses
of penicillins).
APPROACH TO THE DIAGNOSIS OF
ACID-BASE DISORDERS

Appropriate diagnosis of acid-base disorders involves

analysis and synthesis of all relevant information based on
the patient's history, physical examination, and laboratory
data. Use of data from one of these areas alone is insufficient to define the various processes that determine the
primary disorder and the body's adaptive responses. An
outline of the important steps in this approach is as follows:
1. Suspect acid-base disturbances from the history.
2. Suspect acid -base disturbances from the physical ex
amination.
3. Evaluate the venous total CO2 (fCO2), [Cl~], [K+ ],
and AG to support suspected diagnoses and to suggest
possible additional primary disorders.
4. Obtain arterial pH, pCO2 , and [HCOs] to establish
definitive diagnoses using formulas, acid-base map, and
limits of compensation.
A careful review of the patient's history may reveal several commonly encountered clinical conditions which are
typically associated with one or more acid-base disturbances (Table 2). The presence of these conditions should
cause one to add the potential disorders to the differential
diagnosis. One needs to assess the specific disease states,

94

Martin Goldberg
TABLE I

Patterns of Arterial Blood Changes and Adaptation in Simple Acid-Base Disorders

Blood acid-base pattern

Primary disorder
Metabolic acidosis
Metabolic alkalosis
Respiratory acidosis
Acute
Chronic
Respiratory alkalosis
Acute
Chronic

PH

[H+l

[HCOJ]

'T - a

T6

Jl

\b
T T"

t
T

I
1

I"
I I"

Adaptive response

pCO2

Limits of adaptation

I"

pCO 2 = 1.5 [HCOj] + 8 2

ApCO 2 = 0.5 A[HCOi]

pCO2 not <10 mm Hg

pCO 2 not >55 mm Hg

ta

AIHCOj] = 0.1 ApCO 2

A[HCO3] = 0.4 ApCO 2

[HCOj] not >30 mEq/L

[HCOi] not >45 mEq/L

A[HC03] = 0.2 ApCO 2

[HCOj] not <17-18 mEq/L

A[HCO3] = 0.5 ApCO 2

[HCOj] not < 12-15 mEq/L

r
i"
i"

" Initial event.

6
Secondary adaptive response.
Note: Arrows indicate direction of change from normal. A double arrow indicates that the magnitude of the change is considerably
greater in the chronic disorder compared to the acute disorder. Units for pCO2 are mm Hg, and units for [HCOj] mEq/L. A = change
from normal.

nM/1

CUTE
RESPIRATORY
ALKALOSIS

CHRONIC
RESPIRATORY
ACIDOSIS
METABOLIC
ALKALOSIS

50 60
70 mm

= 8.5
80

90

ACID-BASE MAP
FIGURE I The acid-base map. Shaded areas represent the 95% confidence limits for zones of adaptation of
the simple acid-base disorders. Numbered diagonal lines represent isopleths of plasma bicarbonate concentration.
This map has been modified and updated from Goldberg M, Green SB, Moss ML, Marbach MS, Garfinkel D.
JAMA 223:269-275, 1973.

95

12. Approach to Acid-Base Disorders

TABLE 2
Common Clinical Conditions Associated With Acid-Base Disorders

Condition

Metabolic
acidosis

Metabolic
alkalosis

Respiratory
acidosis

Respiratory
alkalosis

Cardiovascular disease
Cardiopulmonary arrest
Pulmonary edema
CNS disease Diabetes
mellitus
Drugs Diuretics
Poisonings
Fever
GI
disease
Diarrhea
Vomiting/gastric suction
Hepatic failure
Hyperkalemia
Hypokalemia
Pulmonary disease Acute
asthma COPD/respiratory
failure Emboli Pneumonia
Renal disease Renal failure
Renal tubular acidosis
Sepsis
" High anion gap metabolic acidosis. 6
During treatment and recovery.

various drugs, the presence of mechanical ventilation, and

nasogastric suctionall of which may be either causative of
primary acid-base disorders or which may modify adaptive
responses. A history of renal failure or diarrhea suggests
metabolic acidosis; pneumonia, sepsis, or hepatic failure
should raise the suspicion of respiratory alkalosis; therapy
with the common diuretics or vomiting should alert one to
the possibility of metabolic alkalosis. In particular, certain
clinical catastrophes such as cardiac arrest, septic shock,
various intoxications, and the heptorenal syndrome typically involve mixed acid-base disorders (see Table 2).
The physical examination may provide additional clues
to potential acid-base disorders or may provide evidence
to support those disorders suspected from the history. For
example, normocalcemic tetany is compatible with severe
alkalemia, and cyanosis may indicate severe hypoxia and
the possibility of respiratory acidosis or lactic metabolic
acidosis. Rapid and/or deep respirations in the absence
of cardiac or pulmonary failure are compatible with the
hyperventilation associated with severe metabolic acidosis.
Metabolic alkalosis is common in patients with signs of
extracellular fluid volume contraction, while high fever typically stimulates respiratory alkalosis.

Careful analysis of the routine laboratory data including

the venous electrolytes may provide useful quantitative
information enabling the identification of the possible predominant acid-base disturbance and clues to additional
disorders. The presence of azotemia strengthens support
for the diagnosis of renal metabolic acidosis.
Most useful is examination of the venous plasma tCO2,
[Cl~], [K+ ], and the AG. In the clinical laboratory, measure ment of venous plasma or serum electrolytes includes an
estimation of the tCO2- This measurement reflects the sum
of the numerator and the denominator of the Henderson
or Henderson-Hasselbalch equations, i.e, [HCO] +
H2CO3 + dissolved CO2 gas. Since normally and in pathophysiological states the [HCOs] is >90% of the tCO2, then
the venous tCO2 is a reasonable approximation of the venous [HCOs]. An abnormal plasma [HCOs] indicates an
acid-base disorder (Table 1). An increased [HCOs] indicates either a primary metabolic alkalosis or an adaptation
to respiratory acidosis. A decreased [HCOi] indicates either primary metabolic acidosis or an adaptation to respiratory alkalosis.
Applying knowledge of the limits of adaptation (Table
1) may help eliminate some diagnostic considerations. For

96
example, if the plasma [HCOs] < 12 mEq/L, which exceeds
the limits of adaptation (Table 1) to respiratory alkalosis,
then metabolic acidosis is definitely present. Conversely,
if plasma [HCOi] > 45 mEq/L, exceeding the adaptive
limit for respiratory acidosis, then metabolic alkalosis may
be diagnosed. A plasma [HCOi] in the normal range, on the
other hand, does not exclude either a mixture of disorders
which have opposing effects on [HCOi] (e.g., metabolic
acidosis + metabolic alkalosis such as may occur in a patient
with renal failure with a high AG metabolic acidosis who
is also vomiting due to uremia and is losing HC1), or acute
respiratory disorders in which the early adaptive changes
in [HCOi] are small (Table 1).
Evaluating changes in plasma [Cl~] may provide additional insights into possible acid-base disorders. Remember, however, that changes in [Cr] are important from
the standpoint of acid-base disorders only when they are
disproportionate to changes in plasma [Na +]. When this
occurs the changes in plasma [Cr] are typically associated
with reciprocal changes in plasma [HCOi]. Thus in primary
metabolic alkalosis and in the adaptation to chronic respiration acidosis, plasma [Cl~] falls as [HCOi] rises, whereas
in primary metabolic acidosis (with normal AG) and in the
adaptation to chronic respiratory alkalosis, plasma [Cr]
rises as [HCOi] decreases.
Abnormalities in plasma [K+] are common in patients
with acid-base disorders. There is, however, no consistent
quantitative relationship between changes in pH or
[HCOi] and changes in [K+]. This is because [K+] is influenced by many metabolic and physiological factors besides
acid-base disturbances. Alterations in plasma [K+] are
more pronounced in metabolic than in respiratory disorders. A rise in serum [K+] is more likely to be associated
with an acute metabolic acidosis with normal AG and is
least likely to be observed in lactic acidosis. In general a
change in [K+] is useful only as a qualitative indicator in
acid-base disorders; a high K+ plus a low [HCOi] implies
metabolic acidosis, and a low [K+] with high [HCOi] suggests metabolic alkalosis.
Calculation of the AG is essential in the evaluation of
acid-base disorders. An elevated AG commonly signifies
the presence of a metabolic acidosis regardless of the
change in plasma [HCOi]. One should be aware, however,
that uncommonly the AG may be moderately increased
(5-9 mEq/L) by severe volume contraction, metabolic alkalosis, and infusions of sodium salts of unmeasured anions
(e.g., sulfates, lactate, carbenicillin). Nevertheless, the AG
is a valuable tool in suspecting the presence of metabolic
acidosis: An AG > 30 mEq/L almost always signifies a
metabolic acidosis, and an AG between 20 and 30 mEq/L
usually signifies metabolic acidosis.
Examination of the relationship between the change in
AG (AAG) and the change in [HCOi] (AHCOi) may be
useful in diagnosing mixed acid-base disorders. In simple
high AG metabolic acidosis each mEq of H+ accumulation
is associated with 1 mEq of unmeasured anion accumulation; further, each mEq of H+ has consumed 1 mEq of
HCOi. Therefore increases in the AG are associated with
corresponding decreases in plasma [HCOi], and AAG ap-

Martin Goldberg
proximates AHCOi. If AAG > AHCOi, this suggests the
presence of an additional process that generates HCOi
(either a metabolic alkalosis or chronic respiratory acidosis), whereas AAG < AHCOi suggests the additional pres ence of a hyperchloremic metabolic acidosis or a chronic
respiratory alkalosis.
Arterial blood gases are used to confirm the predominant
acid-base disorder and to identify and confirm mixed dis turbances. From information derived from the history,
physical examination, and the routine analyses of venous
blood, potential acid-base disturbances can be identified,
and, in many instances, a reasonable acid-base diagnosis
can be established without obtaining arterial blood gases.
On the other hand, it is extremely difficult to diagnose
acute respiratory disturbances and most mixed acid-base
disorders without knowledge of arterial pH, pCO2 , and
[HCOi]. Thus if these are distinct possibilities or if the
patient is critically ill (a setting in which mixed disorders
are common), arterial studies should be obtained.
The data on arterial pH, pCO2 , and [HCOi] should be
analyzed and applied to the acid-base formulas, to the
acid-base map, and to knowledge of limits of adaptation
(see Table 1 and Fig. 1). The results of these analyses
should be coordinated with the differential and potential
diagnostic inferences derived from the history, physical
examination, and routine laboratory data. Before proceeding further, however, one must ensure the internal consis tency of the blood values and their compatibility with the
Henderson-Hasselbalch or Henderson equations. This can
readily be accomplished using the acid-base map (which
also serves as a nomogram for these equations) or by converting pH values to [H+] and applying the Henderson
equation. In the pH range 7.20 to 7.50, each 0.1 unit change
in pH from a normal value of 7.40 corresponds to 10 nEq/
L change in [H+ ] from a normal value of 40 nEq/L.
If the pH is abnormal then the major acid-base disturbance may be identified. Acidemia denotes a predominant
acidosis, and alkalemia indicates a predominant alkalosis.
Reference should then be made to the patterns for each
acid-base disorder as summarized in Table 1. This will
enable confirmation of the major disorder. For example,
acidemia, a low [HCOi], and a lowpCO2 support the diagnosis of metabolic acidosis. In order to rule out an additional primary disorder as a component of a mixed disturbance, the data must be compared to the expected range
of adaptation by either plotting the data on the acid-base
map (Fig. 1), using one of the formulas for adaptive response in Table 1. When the pCO2 is above or below the
predicted range of compensation for a metabolic disturbance or when the [HCOi] is above or below the predicted
levels for a respiratory disorder, a mixed disturbance is
present.
Although many mixed acid-base disorders can be diagnosed using the arterial blood values in conjunction with
the map or formulas, the final diagnostic conclusions require a coordinated synthesis of information derived from
the clinical and laboratory data in conjunction with the
arterial data. Thus, use of the map alone will not identify
triple mixed acid-base disorders. Values lying within a

12. Approach to Acid-Base Disorders

specific zone on the map (while suggesting a simple disorder) are also compatible with a mixed disorder with counterbalancing effects on pH, [HCOi], orpCO2 . In order to
dissect out the various components of a mixed disorder,
we must apply information on changes in AG, plasma [Cl~],
and the plasma [K+]. In fact a normal plasma [HCOi] or
a normal pH or a normal pCO2 by themselves do not rule
out mixed disorders; nor are normal levels of all three
variables simultaneously imcompatible with mixed dis orders.
Mixed disorders should be suspected in the following circumstances:
1. pH is normal and:
a. [HCOi] is high (mixed metabolic alkalosis +
respiratory acidosis).
b. [HCOi] is low (mixed metabolic acidosis +
respiratory alkalosis).
c. [HCOi] is normal and AG is high (mixed
metabolic acidosis + metabolic alkalosis).
2. [HCOi] is normal and:
a. pH is in the acidemic range (mixed chronic
respiratory acidosis and metabolic acidosis).
b. pH is in the alkalemic range (mixed metabolic
alkalosis and chronic respiratory alkalosis).
3. pCO2 and [HCOi] are shifted from normal in
opposite directions.
4. AG is elevated and clinical/laboratory data suggest
a diagnosis other than a simple metabolic acidosis
(metabolic acidosis is a component of a mixed
disorder).
5. When AG is high and the AAG + AHCOi (see
previous discussion of AG and AAG). Remember
that a mixed disorder may include two different
types or etiologies of a simple disorder (e.g., a high
AG + a normal AG metabolic acidosis or an acute

97
respiratory disorder superimposed on an underlying
chronic respiratory disorder).
In conclusion, a successful approach to resolving the
sometimes complex dilemmas associated with the diagnosis
of acid-base disorders involves the application of a systematic, orderly, and logical series of steps. Building on a
knowledge of pathophysiology, these steps involve a coordinated analysis and re-synthesis of information derived
from taking a proper history, performing an adequate physical examination, and obtaining relevant laboratory data.
Final success is measured not only by the intellectual satis faction of the physician in making the correct evaluation(s),
but also by the well being of the patient who has benefited
from the appropriate corrective therapy.
Bibliography
Bia M, Thier SO: Mixed acid-base disturbances: A clinical approach. Med Clin North Am 65:347-361, 1981.
DuBose TD Jr, Cogan MG, Rector FC Jr: Acid-base disorders.
In: Brenner BM, (ed) The Kidney. 5th ed. Saunders, Philadelphia, pp. 929-998, 1996.
Goldberg M, Green SB, Moss ML, Marbach MS, Garfinkel D:
Computer-based instruction and diagnosis of acid- base disorders. JAMA 223:269-275, 1973.
Goodkin DA, Krishna GG, Narins RG: The role of the anion gap
in detecting and managing mixed acid- based disorders. Clin
Endocrinol Metab 13:333-350, 1984.
Hamm L: Mixed acid- based disorders. In: Kokko JP, Tannen RL
(eds) Fluids and Electrolytes. Saunders, Philadelphia, pp. 490495, 1990.
Kraut JA, Madias NE: Approach to diagnosis of acid-base disorders. In: Massry SG, Glassock RJ (eds) Textbook ofNephrology.
Williams and Wilkins, Baltimore, pp. 487-493, 1995.
Laski ME, Kurtzman NA: Acid- base disorders in medicine. Dis
Mon 42:51-125, 1996.
McCurdy DK: Mixed metabolic and respiratory acid-base disturbances: Diagnosis and treatment. Chest 62:35S-44S, 1972.

DISORDERS OF POTASSIUM METABOLISM

MICHAEL ALLON

MECHANISMS OF POTASSIUM HOMEOSTASIS

Aldosterone directly increases the activity of the Na,KATPase in the collecting duct cells, thereby stimulating
secretion of potassium into the tubular lumen. Medical
conditions that impair aldosterone production or secretion
(e.g., diabetic nephropathy, chronic interstitial nephritis)
or drugs that inhibit aldosterone production or action [e.g.,
nonsteroidal antiinflammatory drugs (NSAIDs), angiotensin converting enzyme (ACE) inhibitors, heparin, spironolactone] decrease potassium secretion by the kidney. Conversely, medical conditions associated with increased
aldosterone levels (primary aldosteronism, secondary aldosteronism due to diuretics or vomiting) increase potassium
excretion by the kidney. Although there is profound secondary hyperaldosteronism in congestive heart failure and
cirrhosis, each of these conditions may be associated with
hyperkalemia due to decreased delivery of sodium. Many
diuretics increase renal potassium excretion by a number
of mechanisms, including high distal sodium delivery , high
urine flow rate, metabolic alkalosis, and hyperaldosteronism due to volume depletion. Poorly controlled diabetes
commonly increases urinary potassium excretion, because
it leads to an osmotic diuresis with a high urinary flow rate
and high distal delivery of sodium.
Reabsorption of sodium in the collecting duct occurs
through selective sodium channels, creating an electronegative charge within the tubular lumen relative to the tubular
epithelial cell, which in turn promotes secretion of cations
(K+ and H+ ) into the lumen. Therefore, drugs that block the
sodium channel in the collecting duct decrease potassium
secretion. Conversely, in Liddle's syndrome, a rare genetic
disorder, this sodium channel is constitutively open, resulting in avid sodium reabsorption and excessive potassium secretion.

Total body potassium is about 3500 mmol. Approxi mately 98% of the total is intracellular, primarily in skeletal
muscle, and to a lesser extent in liver. The remaining 2%
(about 70 mmol) is in the extracellular fluid. Two systems
help maintain potassium homeostasis. The first regulates
potassium excretion (kidney and intestine). The second
regulates potassium shifts between the extracellular and
intracellular fluid compartments.
External Potassium Balance

The average American diet contains about 100 mmol of

potassium per day but dietary potassium intake may vary
widely from day to day. To maintain potassium balance, it
is necessary to increase potassium excretion when dietary
potassium increases, and to decrease potassium excretion
when dietary potassium decreases. Normally, the kidneys
excrete 90 to 95% of dietary potassium, with the remaining
5 to 10% excreted by the gut. Potassium excretion by the
kidney is a relatively slow process. It takes 6 to 12 hours
to excrete an acute potassium load.
Renal Handling of Potassium

To understand the physiologic factors that determine

renal excretion of potassium, it is critical to review the main
features of tubular potassium handling. Plasma potassium
is freely filtered across the glomerular capillary into the
proximal tubule. It is subsequently completely reabsorbed
by the proximal tubule and loop of Henle. In the distal
tubule and the collecting duct, potassium is secreted into
the tubular lumen. For practical purposes, urinary excretion of potassium can be viewed as a reflection of potassium
secretion into the lumen of the distal tubule and collecting
duct. Thus, any factor that stimulates potassium secretion
increases urinary potassium excretion; conversely, any factor that inhibits potassium secretion decreases urinary potassium excretion.

Adaptation in Renal Failure

In patients with renal failure, the kidney compensates

by increasing the efficiency of potassium excretion. Clearly,
there is a limit to renal compensation, and a significant loss
of kidney function impairs the ability to excrete potassium,
thereby predisposing to a positive potassium balance and
a tendency to hyperkalemia. In most patients with chronic
renal failure, overt hyperkalemia does not occur until the
creatinine clearance falls below 10 mL/min. Serum aldosterone levels are elevated in many patients with chronic renal
failure. Aldosterone stimulates the activity of both Na,KATPase and H,K-ATPase, thereby promoting secretion

Physiologic Regulation of Renal Potassium Excretion

Five major physiologic factors stimulate distal potassium

secretion, thereby increasing excretion: (1) aldosterone,
(2) high distal sodium delivery, (3) high urine flow rate,
(4) high [K+ ] in tubular cells and (5) metabolic alkalosis.

Primer on Kidney Diseases, Second Edition

98

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

13. Disorders of Potassium Metabolism

of potassium in the collecting duct and defending against

hyperkalemia. These adaptive mechanisms are less effective in patients with acute renal failure, as compared with
chronic renal failure. Therefore, severe hyperkalemia occurs more frequently in the former group.
A subset of patients with chronic renal failure fail to
increase aldosterone levels appreciably; as a result, they
develop hyperkalemia at moderate levels of renal insufficiency (CrCl < 50 mL/min), typically in association with
hyperchloremic, normal anion gap metabolic acidosis. This
condition termed type IV renal tubular acidosis, is most
commonly associated with diabetic nephropathy and
chronic interstitial nephritis. Moreover, administration of
drugs that inhibit aldosterone production or secretion (e.g.,
ACE inhibitors, angiotensin II antagonists, NSAIDs, heparin) may provoke hyperkalemia in patients with mild to
moderate chronic renal failure.
Intestinal Potassium Excretion
Like the renal collecting duct, the small intestine and
colon secrete potassium. Aldosterone stimulates potassium
excretion by the gut. In normal individuals, intestinal potassium excretion plays a minor role in potassium homeostasis. However, in patients with significant renal failure,
intestinal potassium secretion is increased three- to fourfold, thereby contributing significantly to potassium homeostasis. This adaptation is limited, and is inadequate to
compensate for the loss of excretory function in patients
with advanced renal failure.
Internal Potassium Balance
Overview
Extracellular fluid [K] is ~4 mEq/L, whereas the intracellular [K] is 150 mEq/L. Because of the uneven distribution of potassium between the fluid compartments, a relatively small net shift of potassium from the intracellular
to the extracellular fluid compartment produces marked
increases in plasma potassium. Conversely, a relatively
small net shift from the extracellular to the intracellular
fluid compartment produces a marked decrease in plasma
potassium. Unlike renal excretion of potassium, which requires several hours, potassium shift between the extracellular and intracellular fluid compartment (also referred
to as extrarenal potassium disposal) is extremely rapid,
occurring within minutes.
Clearly, in patients with advanced renal failure, whose
capacity to excrete potassium is marginal, extrarenal potassium disposal plays a critical role in the prevention of
life-threatening hyperkalemia following potassium-rich
meals. The following example will illustrate this important
principle. Suppose that a 70-kg dialysis patient with a
serum potassium of 4.5 mmol/L eats one cup of pinto beans
(~35 mmol potassium). Initially, the dietary potassium
is absorbed into the extracellular fluid compartment
(0.2 X 70 = 14 L). This amount of dietary potassium will
increase the serum potassium by 2.5 mmol/L (35 mmol/14
L). In the absence of extrarenal potassium disposal, the
patient's serum potassium would rise acutely to 7.0 mmol/

99
L, a level frequently associated with serious ventricular
arrhythmias. In practice, the increase in serum potassium
is much smaller, due to efficient physiologic mechanisms
that promote potassium shifts into the intracellular fluid
compartment.
Effects of Insulin and Catecholamines on Extrarenal
Potassium Disposal
The two major physiologic factors that stimulate transfer of potassium from the extracellular to the intracellular
fluid compartments are insulin and epinephrine. The stimulation of extrarenal potassium disposal by insulin and /32 adrenergic agonists are both mediated by stimulation of the
Na,K-ATPase activity, primarily in skeletal muscle cells.
Interference with these two physiologic mechanisms (insulin deficiency or /32 -adrenergic blockade, respectively) predisposes to hyperkalemia. On the other hand, excessive
insulin or epinephrine levels predispose to hypokalemia.
The potassium-lowering effect of insulin is dose related
within the physiologic range of plasma insulin and is independent of its effect on plasma glucose. Even the low physiologic levels of insulin present during fasting promote extrarenal potassium disposal. In nondiabetic individuals
hyperglycemia stimulates endogenous insulin secretion,
thereby decreasing the serum potassium. In insulindependent diabetics, endogenous insulin production is limited, and significant hyperglycemia may occur. Hyperglycemia results in plasma hypertonicity, which promo tes
potassium shifts out of the cells, and produces paradoxic
hyperkalemia.
The potassium-lowering action of epinephrine is mediated by /32 -adrenergic stimulation and is blocked by nonselective /3-blockers, but not by selective /3i-adrenergic blockers. a-Adrenergic stimulation promotes shifts of potassium
out of the cells into the extracellular fluid compartment,
tending to increase serum potassium. Epinephrine is a
mixed a- and /3-adrenergic agonist, such that its net effect
on serum potassium reflects the balance between its /3adrenergic (potassium-lowering) and a-adrenergic (potas sium-raising) effects. In normal individuals the /3adrenergic effect of epinephrine predominates over the aadrenergic effect, such that the serum potassium decreases. In contrast, the a-adrenergic effect of epinephrine
on potassium shifts is much more prominent in patients
with severe renal failure; as a result, dialysis patients are
refractory to the potassium-lowering effect of epinephrine.
Effect of Acid-Base Disorders on Extrarenal
Potassium Disposal
As a general rule, metabolic alkalosis shifts potassium
into the cells, whereas metabolic acidosis shifts potassium
out of the cells. However, the nature of the metabolic
acidosis determines its effect on serum potassium. Thus,
mineral acidosis (i.e., hyperchloremic, normal anion gap
metabolic acidosis) typically shifts potassium out of the
cells, thereby predisposing to hyperkalemia. In contrast,
organic metabolic acidosis (e.g., lactic acidosis) does not
affect the serum potassium. Bicarbonate administration to
individuals with normal renal function decreases serum

100

Michael Allon

potassium, but this effect is mainly due to enhanced urinary

excretion of potassium. In contrast, bicarbonate adminis tration to dialysis patients (in whom the capacity for urinary
potassium excretion is negligible) does not lower plasma
potassium acutely. Moreover, bicarbonate administration
does not potentiate the potassium-lowering effects of insulin or albuterol in dialysis patients.
LA B O R A T O R Y T E S T S F O R D I F F E R E N T I A L D I A G N O S I S
OF POTASSIUM DISORDERS
Differential Diagnosis of Hypokalemia and Hyperkalemia
The clinical history, review of medications, family his tory, and physical examination are sufficient in the rapid
differential diagnosis of the etiology of most potassium
disorders. In selected patients the etiology of hypokalemia
or hyperkalemia is not apparent, and additional specialized
laboratory tests may be useful. Measurements of the fractional excretion of potassium (FEK ) and transtubular potassium gradient (TTKG) may be useful in distinguishing
between renal and nonrenal etiologies of hyperkalemia and
hypokalemia. The general principle underlying these tests
is that the kidney compensates for hyperkalemia by increas ing potassium excretion, and compensates for hypokalemia
by decreasing potassium excretion. When the serum potas sium valve is abnormal and potassium excretion has not
compensated appropriately, a renal etiology for the serum
potassium abnormality is suggested. The optimal use of
FEK or TTKG is the differential diagnosis requires that
these values be obtained before the potassium abnormality
(hyperkalemia or hypokalemia) is corrected.
Fractional Excretion of Potassium.
FEK is the percentage of potassium filtered into the proximal tubule that appears in the urine. It represents potas sium clearance corrected for glomerular filtration rate
(GFR) or C1 K/C1 CR. Since the clearance of any substance
can be calculated from UV/P, this ratio (in bold type) can
be algebraically transformed to [( UKV/ PJ ^/ UCRV/PCR)] X
100%. The V in the numerator and denominator cancel
out, giving the simplified formula
[U K./SK] X 100%
[/CR/ SCR] '

where t/K and /CR are the concentrations of potassium

and creatinine in the urine, and S K and SCR are the corresponding serum concentrations. For an individual with normal renal function on an average dietary potassium intake,
the FEK is approximately 10%. When hypokalemia is due to
extrarenal causes (low potassium diet, GI losses, potassium
shifts into cells) the kidney conserves potassium, and the
FEK is low. In contrast, hypokalemia due to renal potassium
losses is associated with an increased FEK. Similarly, in
the setting of hyperkalemia, a high FEK is consistent with
appropriate renal compensation and suggests an extrarenal
etiology, whereas a low FEK suggests a renal etiology. If
a urine creatinine measurement is not available, one can

often use [7K alone to differentiate between renal and extrarenal causes. Specifically, in a hypokalemic patient, t/K >
20 mEq/L suggests a renal etiology, whereas I/ K < 20 mEq/
L suggests an extrarenal etiology.
Transtubular Potassium Gradient
The TTKG is a formula that estimates the potassium
gradient between the urine and the blood in the distal
nephron. It is calculated from [UK/(Uosln/Posm)]/P K, where [7osm
and Fosm are the urine and plasma osmolalities. The
numerator is an estimate of the luminal potassium concentration. The Uosm/Posm term is included to correct for the rise
in UK that is due purely to water abstraction and concentration of the urine overall. TTKG values have been
derived from empiric measurements in normal individuals
under a variety of physiologic conditions. In a normal individual under normal circumstances, the TTKG is about 6
to 8. Hypokalemia with a high TTKG suggests excessive
renal potassium losses, whereas hypokalemia with a low
TTKG suggests an extrarenal etiology. Similarly, hyperkalemia with a low TTKG suggests a renal etiology, whereas
hyperkalemia with a high TTKG is consistent with an extrarenal etiology.
Several factors limit the utility of the FEK and TTKG
in the differential diagnosis of potassium disorders. The
FEK and TTKG are increased when dietary potassium is
increased, and decreased when dietary potassium is decreased. Furthermore, in patients with chronic renal failure
there is an adaptive increase in potassium excretion per
functioning nephron, such that FEK and TTKG increase.
This means that the "normal" value for a given individual
can vary substantially, making it difficult to determine the
significance of a high or low FEK or TTKG.
HYPOKALEMIA
Hypokalemia vs Potassium Deficiency
It is important to distinguish between potassium deficiency and hypokalemia. Potassium deficiency is the state
resulting from a persistent negative potassium balance, i.e.,
potassium excretion exceeding potassium intake. Hypokalemia refers to a low plasma potassium concentration. Hy pokalemia can be due either to potassium deficiency (inadequate potassium intake or excessive potassium losses) or
to net potassium shifts from the extracellular to the intracellular fluid compartment. A patient may have severe potas sium depletion without manifesting hypokalemia. An important example is a patient presenting with diabetic
ketoacidosis. Due to hyperglycemia, such patients have
typically undergone an osmotic diuresis for several days,
leading to a high rate of renal potassium excretion and
potassium deficiency. However, as a result of insulin deficiency, there is a concomitant shift of potassium out of the
cells into the extracellular fluid compartment. At presentation to the hospital, such patients are frequently normokalemic or even hyperkalemic. Once they are treated with
exogenous insulin, there is a rapid shift of potassium back
into the cells, and within a few hours the patients develop

13. Disorders of Potassium Metabolism

significant hypokalemia. Conversely, patients hospitalized
with an acute myocardial infarction commonly have hypokalemia due to stress-induced catecholamine release and
enhanced extrarenal potassium disposal, even though they
may have a normal external potassium balance.
Clinical Disorders Associated with Hypokalemia

Table 1 lists the most common causes of hypokalemia.

The kidney can avidly conserve potassium, such that hypokalemia due to inadequate potassium intake is a rare event
requiring prolonged starvation ("tea and toast diet").
Therefore, hypokalemia is usually due to excessive potassium losses from the gut or the urine or due to potas sium shifts from the extracellular to the intracellular fluid
compartments. Prolonged vomiting causes potassium
losses, in small part due to potassium present in the gastric
juice (10 mEq/L), but mainly from renal losses due to
secondary aldosteronism resulting from volume depletion.
Severe diarrhea, either due to disease or laxative abuse,
results in significant potassium excretion in the stool.
Excessive renal potassium losses as a cause of hypokalemia occur in a number of clinical syndromes. Conceptually, it is useful to consider separately hypokalemia associ-

TABLEI

Causes of Hypokalemia
Inadequate potassium intake (severe malnutrition)
Extrarenal potassium losses
Vomiting Diarrhea
Hypokalemia due to urinary potassium losses
Diuretics (loop, thiazides, acetazolamide)
Osmotic diuresis (e.g., hyperglycemia)
Hypokalemia with hypertension
Primary aldosteronism
Glucocorticoid remediable hypertension
Malignant hypertension
Renovascular hypertension
Renin-secreting tumor
Essential hypertension with excessive diuretics
Liddle's syndrome
11/3 -hydroxysteroid dehydrogenase deficiency
Genetic
Drug-induced (chewing tobacco, licorice, some French
wines)
Congenital adrenal hyperplasia
Hypokalemia with a normal blood pressure
Distal RTA (type I)"
Proximal RTA (type II)
Bartter's syndrome
Gitelman's syndrome
Hypomagnesemia (cisplatinum, alcoholism, diuretics)
Hypokalemia due to potassium shifts Insulin
administration Catecholamine excess (acute
stress) Familial periodic hypokalemic
paralysis Thyrotoxic hypokalemic paralysis
" RTA, renal tubular acidosis.

101
ated with hypertension and hypokalemia associated with
a normal blood pressure. When hypokalemia is associated
with hypertension, measurements of plasma renin and aldosterone may be helpful in the differential diagnosis. Several
physiologic observations are relevant:
1. Aldosterone, a mineralocorticoid, stimulates sodium
reabsorption and potassium secretion in the
collecting duct.
2. The physiologic stimulus for aldosterone secretion
is activation of the renin-angiotensin axis.
Moreover, aldosterone-induced sodium retention
suppresses the renin-angiotensin axis by negative
feedback.
3. Glucocorticoids at high concentrations bind to
mineralocorticoid receptors and mimic their
physiologic actions.
4. Glucocorticoids are stimulated by
adrenocorticotropic hormone (ACTH), and
suppress ACTH production by negative feedback.
Primary aldosternonism is due to autonomous (nonrenin-mediated) secretion of aldosterone by the adrenal
cortex. This results in avid sodium retention and potassium
secretion by the distal nephron. The patients present with
volume-dependent hypertension, hypokalemia, and metabolic alkalosis. Biochemical evaluation reveals a high and
nonsuppressable serum aldosterone level with a low plasma
renin. Abdominal CT scan reveals either a unilateral adrenal adenoma or bilateral adrenal hyperplasia. The former
is treated surgically, and the latter with spironolactone.
Glucocorticoid-remediable aldosteronism (GRA) is a rare,
autosomal dominant condition in which there is fusion of
the 11/3-hydroxylase and aldosterone synthase genes. As a
result, aldosterone secretion is stimulated by ACTH, and
can be suppressed by an exogenous mineralocorticoid,
dexamethasone. Patients with GRA have a very similar
clinical presentation to those with primary aldosteronism
(volume-dependent hypertension, hypokalemia, high serum aldosterone, and low serum renin), except that they
are younger and have a family history of hypertension.
Patients with renovascular hypertension, renin-secreting
tumors, and severe malignant hypertension may also present with severe hypertension and hypokalemia. In contrast
to patients with primary aldosteronism, these patients have
secondary aldosteronism, i.e., high serum renin and aldosterone levels. Of course, patients with essential hypertension may also have hypokalemia and high plasma renin
and aldosterone levels if they are treated with loop or
thiazide diuretics and are volume depleted.
Patients with 11/3-hydroxysteroid dehydrogenase deficiency, a rare genetic disorder, have a defect in the conversion of cortisol to cortisone in the peripheral tissues. This
results in high tissue cortisol levels that activate the mineralocorticoid receptors, producing hypokalemia and hypertension. Such patients have low serum renin and aldosterone levels. Chewing tobacco, certain brands of licorice,
and some French red wines contain glycyrrhizic acid, which
inhibits 11/3-hydroxysteroid dehydrogenase. Ingestion of
these foods may produce hypokalemia, volume-dependent

Michael Allon

102

hypertension, and low serum renin and aldosterone levels,

similar to the clinical presentation of congenital 11/3hydroxysteroid dehydrogenase deficiency.
Patients with congenital adrenal hyperplasia have a deficiency of 11 j3-hydroxylasse, an enzyme required in the common pathways for mineralocorticoids and glucocorticoids.
These patients have low serum renin and aldosterone levels, high levels of DOCA (a mineralocorticoid), and high
levels of androgen. Males have early puberty, and females
exhibit virilization, with hirsutism and clitoromegaly. This
condition improves with exogenous corticosteroids to suppress ACTH.
Liddle's syndrome is a rare autosomal dominent disorder
caused by a defect of the collecting duct sodium channel.
It is constitutively open, resulting in an unregulated increase of sodium absorption and potassium secretion at
that site. Patients present with hypokalemia, hypertension,
and volume overload. Their biochemical profile reveals
a low serum renin and aldosterone. The patients' blood
pressure and serum potassium improves dramatically with
inhibitors of the sodium channel, such as amiloride.
Hypokalemia due to excessive renal potassium excretion
is also seen in a number of clinical conditions in which
hypertension is infrequent. Both type I (distal) and type
II (proximal) renal tubular acidosis (RTA) are associated
with kaliuresis and hypokalemia; both conditions present
with a normal anion gap metabolic acidosis. Type I RTA
is frequently associated with hypercalciuria and calcium
oxalate kidney stones. Type II RTA is rare in adults, and
is often associated with a generalized defect in proximal
tubular function, manifesting as glycosuria (with a normal
serum glucose), hypophosphatemia with phosphaturia, and
a low serum uric acid with uricosuria (see Chapter 9).
Bartter's syndrome is a rare familial disease characterized by hypokalemia, metabolic alkalosis, hypercalciuria,
normal blood pressure, and high plasma renin and aldosterone levels. Recently, it has been associated with a mutation
in the renal Na-K-2Cl transporter, which is responsible for
active sodium reabsorption in the thick ascending limb of
Henle. These patients act as if they are chronically ingesting
loop diuretics; for this reason, they are difficult to distinguish clinically from patients with surreptitious diuretic
ingestion. Patients with Gitelman's syndrome, a variant of
Bartter's syndrome, differ in that they have hypocalciuria
and hypomagnesemia. Recently, Gitelman's syndrome has
been linked to a mutation in the renal thiazide-sensitive
Na + -Cr transporter. These patients act as if they are chronically ingesting thiazide diuretics.
Familial hypokalemic periodic paralysis is a rare, autosomal dominant disorder in which affected individuals develop periodic episodes of severe muscle weakness in association with profound hypokalemia, due to rapid shifts of
potassium from the extracellular to the intracellular fluid
compartment. Interestingly, even when patients have complete paralysis, the diaphragm and bulbar muscles are
spared, such that the patient is able to breathe, swallow,
talk, and blink. The paralysis resolves within hours of potassium ingestion. The patients are asymptomatic and have
a normal serum potassium between the acute episodes.

Thyrotoxic hypokalemic paralysis is an unusual manifestation of hyperthyroidism, seen primarily in Asian patients.
The clinical presentation is similar to that of hypokalemic
periodic paralysis, except that the paralytic episodes cease
when the hyperthyroidism is corrected.
Drug-Induced Hypokalemia

A number of drugs have the potential to cause hypokalemiaeither by stimulating renal potassium excretion or
by blocking extrarenal disposal. Exogenous mineralocorticoids mimic the effects of aldosterone, thereby stimulating
distal potassium secretion. Glucocorticoids possess some
mineralocorticoid activity; treatment with high doses of
these agents has a similar effect. Most diuretics, including
loop diuretics, thiazide diuretics, and acetazolamide increase renal potassium excretion. A number of drugs, including alcohol, diuretics, and cisplatinum, cause renal
magnesium-wasting and hypomagnesemia. For reasons
that are not well understood, hypomagnesemia impairs renal potassium conservation. Thus, these patients may have
associated hypokalemia that is refractory to potassium supplementation until the magnesium deficit is corrected. (Paradoxically, cyclosporine may produce hypomagnesemia in
conjunction with /zyperkalemia.)
Drugs that promote extrarenal potassium disposal may
also result in hypokalemia. This phenomenon can be seen
after the administration of an acute dose of insulin. Similarly, j32-agonists (either intravenous or nebulized), including albuterol and terbutaline, frequently result in acute hypokalemia.
Clinical Manifestations of Hypokalemia

Hypokalemia may produce electrocardiographic abnormalities, including a flattened T wave and a U wave (Fig.
1). Hypokalemia also increases the risk of ventricular arrhythmias in patients with ischemic heart disease or patients taking digoxin. Severe hypokalemia is associated with
variable degrees of skeletal muscle weakness, even to the
point of paralysis. On rare occasions, diaphragmatic paralysis from hypokalemia can lead to respiratory arrest. There
may also be decreased motility of smooth muscle, manifesting itself as ileus or urinary retention. Rarely, severe hypokalemia may result in rhabdomyolysis.
Severe hypokalemia also interferes with the urinary concentrating mechanism in the distal nephron, and leads to
nephrogenic diabetes insipidus. Such patients have a low
urine osmolality in the face of high serum osmolality, and
are refractory to vasopressin.
Treatment of Hypokalemia

The acute treatment of hypokalemia requires potassium

supplementation, which can be given either intravenously
or orally. The correlation between serum potassium and
total potassium deficit in hypokalemia patients is quite
poor. A given patient's serum potassium is a reflection of
both external potassium balance and transcellular potas sium shifts. The fraction of administered exogenous potas -

13. Disorders of Potassium Metabolism

103
nuts, and legumes; see Table 2). In some patients, chronic
oral potassium supplementation is necessary.

Serum
potassium
Ventricula
r
(mEq/L)

HYPERKALEMIA

10
Prolonged P-R
Auricular standstill,
intraventricular block
Hyperkalemia
interval,
high T wave
depressed S-T segment
High T wave
Normal

JA-

3.5
Low T wave

_/^-Jl_x-^

Normal
Hypokalemia

3.0
Low T wave, high U wave

2.5

Low T wave,
high U wave,

low S-T segment

I
ntervals

FIGURE I A schematic representation of EKG changes and the

serum potassium levels at which such changes are typically seen.
From Seldin DW, Giebisch G, eds: The Regulation of Potassium
Balance, Raven Press, New York, 1989.

sium that remains in the extracellular fluid compartment is

variable. Thus, it is difficult to predict how mu ch potassium
replacement a given hypokalemic patient will require.
Without adequate monitoring, it is possible to give too
much potassium, making the patient hyperkalemic. Therefore, one should give repeated small doses of potassium,
with frequent checks of the serum potassium value.
Oral potassium administration is safer than the intravenous route, and is less likely to produce an overshoot in
the serum potassium. Each oral dose should not exceed 20
to 40 mEq of potassium. Intravenous KC1 should be reserved for severe, symptomatic hypokalemia (<3.0 mEq/
L) or for patients who cannot ingest oral potassium. Intravenous KC1 should not be given any faster than 10 mmol/
h in the absence of continuous EKG monitoring. The serum
potassium should be rechecked every 2 to 3 hours to confirm a clinical response and to avoid an overshoot.
Correction of the underlying medical condition may pre vent recurrence of hypokalemia after its correction. If the
patient has a chronic condition associated with persistent
urinary potassium losses, such that hypokalemia is likely
to recur, the patient should be encouraged to increase the
intake of foods high in potassium (especially fresh fruits,

Pseudohyperkalemia is a factitious elevation of the serum potassium due to in vitro release of potassium from
blood cells. It may be seen with in vitro hemolysis, thrombocytosis, or severe leukocytosis. Pseudohyperkalemia due
to hemolysis is readily apparent because the serum is pink.
Pseudohyperkalemia due to severe thrombocytosis or leukocytosis can be confirmed by drawing simultaneous blood
samples in tubes with and without anticoagulant; if the
serum potassium is significantly higher than the plasma
potassium, this confirms the diagnosis.
True hyperkalemia is caused by a positive potassium
balance (increased potassium intake or decreased potas sium excretion) or an increase in net potassium shift from
the intracellular to the extracellular fluid compartment.
Table 3 provides a list of the most common causes of
hyperkalemia. In practice, most patients who develop severe hyperkalemia have more than one contributory factors. For example, a patient with moderate renal failure
due to diabetic nephropathy may be medicated with an
ACE inhibitor and have mild hyperkalemia. However,
when started on indomethacin for acute gouty arthritis, the
patient rapidly develops severe hyperkalemia.
Drug-Induced Hyperkalemia
Many drugs have the potential to cause hyperkalemia
either by inhibiting renal potassium excretion or by blocking extrarenal disposal (Table 4). Most individuals taking
these drugs will not develop hyperkalemia. Patients at
greatest risk are those with renal failure, especially if they
have a high dietary potassium intake or are on an additional
medication that predisposes to hyperkalemia. Most diuretics (loop diuretics, thiazide diuretics, acetazolamide) increase urinary potassium excretion and tend to cause hypokalemia. However, potassium-sparing diuretics inhibit
urinary potassium excretion and predispose to hyperkalemia by one or two mechanisms. Spironolactone is a competitive inhibitor of aldosterone; it binds to the aldoster-

TABLE 2

Potassium Content of Selected Foods

Food
Pinto beans (1 cup)
Raisins (1 cup)
Honeydew (i melon)
Nuts (1 cup)
Blackeyed peas (1 cup)
Collard greens (1 cup)
Banana (1 medium)
Tomato (1 medium)
Orange (1 large)
Milk (1 cup)
Potato chips (10)

Potassium (mg)
1370
1106

939
688
625
498
440
366
333
351
226

Potassium (mEq)

35
28
24
18
16
13
11
9
9
9
6

104

Michael Allon
TABLE 3

TABLE 4

Causes of Hyperkalemia

Mechanisms for Drug-induced Hyperkalemia

Pseudohyperkalemia
Hemolysis
Thrombocytosis Severe
leukocytosis Fist
clenching
Decreased renal excretion Acute or chronic renal failure
Aldosterone deficiency (e.g., type IV renal tubular acidosis)
Frequently associated with diabetic nephropathy, chronic
interstitial nephritis, or obstructive nephropathy
Adrenal insufficiency (Addison's disease) Drugs that
inhibit potassium excretion (see Table 3) Kidney
diseases that impair distal tubule function
Sickle cell anemia
Systemic lupus erythematosus
Abnormal potassium distribution
Insulin deficiency /3-blockers
Metabolic or respiratory acidosis
Familial hyperkalemic periodic paralysis
Abnormal potassium release from cells
Rhabdomyolysis Tumor lysis syndrome

one receptors in the collecting duct, thereby inhibiting

Na,K-ATPase activity, and indirectly limiting potassium
secretion. Interestingly, the immunosuppressive drug
cyclosporine also blocks Na,K-ATPase activity in the distal
nephron. Two other potassium-sparing diuretics, amiloride
and triamterene, bind to the sodium channel in the collecting duct. Blocking this channel inhibits sodium reabsorption in the distal nephron, thereby limiting the establishment of the electrochemical gradient required for
potassium secretion. Two antibiotics, trimethoprim (one
of the components of Bactrim) and pentamidine, have also
been shown to block the sodium channel in the collecting
duct and to predispose patients to hyperkalemia. In addition, trimethoprim has been shown to inhibit the collecting
tubule H,K-ATPase.
Because aldosterone plays an important role in enhancing renal potassium excretion in patients with renal failure,
drugs that inhibit aldosterone production (either directly
or indirectly) predispose such patients to hyperkalemia.
Angiotensin II is a potent stimulator of aldosterone production in the adrenal cortex. ACE inhibitors inhibit the production of angiotensin II, thereby decreasing aldosterone
levels. Similarly, angiotensin II receptor blockers, such as
losartan, also inhibit aldosterone production. Prostaglandin
inhibitors (NSAIDs) inhibit the production of renin,
thereby indirectly decreasing aldosterone production. This
effect is seen even with "renal-sparing NSAIDs," such as
sulindac. Heparin has been shown to inhibit the production
of aldosterone in the renal cortex directly, primarily by
decreasing the number and affinity of angiotensin II receptors in the zona glomerulosa. This effect occurs even with
the low doses of subcutaneous heparin used for prophylaxis
of venous thrombosis in hospitalized patients. The immu -

Decrease renal potassium excretion

Block sodium channel in the distal nephron
Potassium- sparing diuretics: amiloride, triamterene
Antibiotics: trimethoprim, pentamidine
Block aldosterone production
ACE inhibitors (e.g., captopril, enalapril, lisinopril,
benazepril)
Angiotensin II antagonists (e.g., losartan)
NSAIDs Heparin Tacrolimus
Block aldosterone receptors
Spironolactone
Block Na,K- ATPase activity in the distal nephron
Cyclosporine
Inhibit extrarenal potassium disposal Block /3 2 adrenergic mediated extrarenal potassium
disposalnonselective /3 -blockers (e.g., propranolol,
nadolol, timolol) Block Na,K-ATPase activity in skeletal
muscles; digoxin
overdose (not therapeutic doses) Inhibit
insulin release (e.g., somatostatin)
Potassium release from injured cells Drug-induced
rhabdomyolysis (e.g., lovastatin, cocaine) Drug-induced tumor
lysis syndrome (chemotherapy agents in acute leukemias,
high-grade lymphomas)
Drug-induced acute renal failure

nosuppressive drug tacrolimus may also cause hyperkalemia by inhibiting aldosterone synthesis.
Given the stimulation of extrarenal potassium dis posal by /3-adrenergic agonists, it is not surprising that /32antagonists can predispose to hyperkalemia. This effect is
seen primarily with nonselective /3-blockers (e.g., propranolol and nadolol), rather than with selective /3rblockers.
There is significant systemic absorption of topical ftblockers, and severe hyperkalemia may rarely be provoked
by timolol eyedrops. Drugs inhibiting endogenous insulin
release, such as somatostatin, have rarely been implicated
as a cause of hyperkalemia in patients with renal failure.
Presumably, long-acting somatostatin analogs, such as octreotide, would have a similar effect on serum potassium.
Digoxin overdose causes inhibition of Na,K-ATPase activity in skeletal muscle cells, and may manifest with hyperkalemia. This effect is rarely seen at therapeutic doses of
the drug.
Finally, drugs can cause hyperkalemia indirectly by causing release of intracellular potassium from injured cells (e.g .,
rhabdomyolysis with lovastatin and cocaine, or tumor lysis
syndrome occurring spontaneously or when chemotherapy
is administered to patients with acute leukemia or high grade
lymphoma). Moreover, drug-induced acute renal failure
may be associated with secondary hyperkalemia.
Clinical Manifestations of Hyperkalemia
Hyperkalemia may produce progressive electrocardiographic abnormalities, including peaked T waves, flattening

13. Disorders of Potassium Metabolism

or absence of P waves, widened QRS complexes, and sine

waves (see Fig. 1). The major risk of severe hyperkalemia
is the development of life-threatening ventricular arrhythmias. Severe hyperkalemia with EKG changes is a medical emergency.
Severe hyperkalemia, like severe hypokalemia, can cause
skeletal muscle weakness, even to the point of paralysis
and respiratory failure. Hyperkalemia impairs urinary acidification by decreasing collecting tubule apical H,KATPase, which may result in a renal tubular acidosis (type
IV RTA). Hyperkalemia stimulates endogenous aldosterone secretion.
Treatment of Hyperkalemia

Severe hyperkalemia associated with electrocardiographic changes (see Fig. 1) is a life-threatening state requiring emergent intervention. If the patient's EKG is suspicious for hyperkalemia, one should initiate therapy
without waiting for the laboratory confirmation. If the patient has renal failure, urgent dialysis is required for removal of potassium from the body. Because of the inevitable delay in initiating dialysis, the following temporizing
measures must be initiated promptly.
1. Stabilize the myocardium. Acute administration of
intravenous calcium gluconate does not change
plasma potassium, but does transiently improve the
EKG. The effect is almost immediate. Give 10 mL
of calcium gluconate over 1 minute. If there is no
improvement in the EKG appearance within 3 to 5
minutes, the dose should be repeated.
2. Shift potassium from the extracellular to the
intracellular fluid, in order to rapidly decrease the
serum potassium. This involves administration of
insulin and a /32 -agonist.
a. Intravenous insulin is the fastest way to lower
the serum potassium. The plasma potassium starts
to decrease within 15 minutes. Intravenous glucose
is given concurrently to prevent hypoglycemia.
Give 10 units of regular insulin and 50 mL of 50%
dextrose (1 ampule of D50 ) as a bolus, followed by
a continuous infusion of 5% dextrose at 100 mL/h
to prevent late hypoglycemia. Never give dextrose
without insulin for the acute treatment of
hyperkalemia; in patients with inadequate
endogenous insulin production, the resulting
hyperglycemia can produce a paradoxical increase
in serum potassium.
b. Beta-agonists. Give 20 mg of albuterol (a
j62 -agonist) by inhalation over 10 minutes. The
onset of action is 30 minutes. Use the concentrated
form (5 mg/mL) of the drug to minimize the
volume that needs to be inhaled. The dose required
to lower plasma potassium is considerably higher
than that used to treat asthma, because only a small
fraction of nebulized albuterol is absorbed
systemically. Thus, 0.5 mg of intravenous albuterol
(not available in the US) produces a comparable
change in plasma potassium to that seen after 20 mg

105

of nebulized albuterol. The potassium-lowering

effect of albuterol is additive to that of insulin.
c. Sodium bicarbonate. This is NOT useful
acutely, because it takes 3 to 4 hours for an effect
on serum potassium to start. Moreover, bicarbonate
administration is still indicated if the patient has
severe metabolic acidosis (serum bicarbonate <10
mmol/L).
3. Once the previous temporizing measures have been
performed, further interventions are done to
remove potassium from the body.
a. Diuretics. These only work if the patient has
adequate kidney function.
b. Kayexalate. This resin-exchanger removes
potassium from the blood via the gut, in exchange
for an equal amount of sodium. It is relatively
slow-acting, requiring 1 to 2 hours before plasma
potassium decreases. Each gram of resin removes
0.5 to 1.0 mmol of potassium. Give 50 g in 30 mL
sorbitol by mouth or 50 g in a retention enema.
The rectal route is faster and more reliable.
c. Hemodialysis. This is the definitive treatment
for patients with advanced renal failure and severe
hyperkalemia.
For patients with moderate hyperkalemia, not associated
with electrocardiographic changes, it is frequently sufficient
to provide one or two doses of exchange resin, and to
discontinue the drugs predisposing to hyperkalemia.
To prevent a recurrence of hyperkalemia once the acute
treatment has been provided, the following measures are
useful:
1. Counsel the patient on dietary potassium restriction,
40 to 60 mEq/day (Table 2).
2. Avoid medications that interfere with renal excretion
of potassium, e.g., potassium-sparing diuretics, NSAIDs,
and ACE inhibitors.
3. Avoid drugs that interfere with potassium shifts from
the extracellular to the intracellular compartment, e.g.,
/3-blockers.
4. In selected patients, chronic medication with loop di
uretics can be used to stimulate urinary potassium ex
cretion.
5. Specific therapy may be indicated for the underlying
etiology, when available. For example, patients with adre
nal insufficiency require replacement with exogenous glucocorticoids and mineralocorticoids. In patients with hyperkalemic periodic paralysis (a rare, autosomal dominant
disorder in which affected individuals develop periodic epi
sodes of severe muscle weakness in association with pro
found hyperkalemia), prophylactic aerosolized albuterol
can prevent both exercise-induced hyperkalemia and mus
cle weakness.
Bibliography
Allon M: Hyperkalemia in end-stage renal disease: mechanisms
and management. J Am Soc Nephrol 6:1134-1142, 1995.
Allon M: Treatment and prevention of hyperkalemia in end-stage
renal disease. Kidney Int 43:1197-1209, 1993.

106

David A. Bushinsky

DuBose TD: Hyperkalemic hyperchloremic metabolic acidosis:

Pathophysiologic insights. Kidney Int 51:591-602, 1997.
Ethier JH, Kamel KS, Magner PO, Lcmann J, Halperin ML: The
transtubular potassium concentration in patients with hyperkalemia and hypokalemia. Am J Kidney Dis 15:309-315, 1990.
Farese RV, Biglieri EG, Shackleton CHL, Irony I, Gomez-Fontes
R: Licorice-induced hypermineralocorticoidism. N Engl J Med
325:1223-1227, 1991.
Field MJ, Giebisch GJ: Hormonal control of renal potassium excretion. Kidney Int 27:379-387, 1985.
Kamel KS, Halperin ML, Faber MD, Steigerwalt SP, Heilig CW,
Narins RG: Disorders of potassium balance. In: Brenner BM
(ed) The Kidney. Saunders, Philadelphia, pp. 999-1037, 1996.
Krishna GG, Steigerwalt SP, Pikus R, et al: Hypokalemic states.
In: Narins RG (ed) Clinical Disorders of Fluid and Electrolyte
Metabolism. McGraw-Hill, New York, pp. 659-696, 1994.
Lifton RP, Dluhy RG, Powers M, et al: A chimaeric 11 betahydroxylase/aldosterone synthase gene causes glucocorticoid-

remediable aldosteronism and human hypertension. Nature

355:262-265, 1992.
Salem MM, Rosa RM, Batlle DC: Extrarenal potassium tolerance
in chronic renal failure: implications for the treatment of acute
hyperkalemia. Am J Kidney Dis 18:421-440, 1991.
Shimkets RA, Warnock DG, Bositis CM, et al: Liddle's syndrome:
Heritable human hypertension caused by mutations in the beta
subunit of the epithelial sodium channel. Cell 79:407-414,1994.
Simon DB, Nelson-Williams C, Bia MJ, et al: Gitelman's variant
of Bartter's syndr ome, inherited hypokalaemic alkalosis, is
caused by mutations in the thiazide -sensitive Na-Cl cotransporter. Nature Genet 12:24-30, 1996.
Simon DB, Karet FE, Hamdan JM, et al: Bartter's syndrome,
hypokalaemic alkalosis with hypercalciuria, is caused by mutations in the Na-K-2Cl cotransporter NKCC2. Nature Genet
13:183-188, 1996.
Tannen RL: Approach to the patient with altered potassium concentration. In: Kelley WN (ed). Textbook of Internal Medicine.
Lippincott, Philadelphia, pp. 848 -855, 1992.

DISORDERS OF CALCIUM AND

PHOSPHORUS HOMEOSTASIS
DAVID A. BUSHINSKY

CALCIUM Calcium
Homeostasis
Distribution

The vast majority (99.5%) of total body calcium is contained within the bone mineral while only about 0.1% of
body calcium is in the extracellular fluid. The mineral
phases of bone provide a reservoir of calcium for the
smaller extra- and intracellular pools.
The concentration of calcium in the cell cytosol is approximately 100 nmol/L or approximately one-thousandth of
the extracellular calcium concentration. Within the cell,
the mitochondria and the sarcoplasmic and endoplasmic
reticula contain the highest concentrations of calcium.
Serum Concentration

In humans the concentration of serum calcium is maintained at a constant level between 9.0 and 10.4 mg/dL or
2.25 to 2.60 mmol/L. Of the total serum calcium approxi -

Primer on Kidney Diseases. Second Edition

mately 40% is protein bound, especially to albumin and to

a lesser extent globulins and other proteins. Approximately
10% of the calcium is complexed to phosphate, citrate,
carbonate, and other anions; the remaining 50% exists in
the ionized form. The ionized calcium and the complexed
calcium constitute what is filtered by the kidney, the ultrafiltrable calcium, which has a concentration of approximately 1.5 mmol/L.
The concentration of ionized calcium is of physiological
importance and remains remarkably constant even with
marked variations in the levels of total calcium. Increases
in total calcium concentration lead to an increase in calcium
binding to albumin, lessening the increase in ionized calcium concentration. Although a fall in serum albumin of
1 g/dL is usually associated with a fall of 0.8 mg/dL in
total calcium concentration, the proportional fall in ionized
calcium will be less. The fall in albumin will lessen the
fraction of bound calcium and result in only a small decrease in the concentration of ionized calcium. Alterations
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

107

14. Disorders of Calcium and Phosphorus Homeostasis

in systemic pH will alter calcium binding by albumin. A fall

in pH of 0.1 unit will cause approximately a 0.1 mEq/L rise
in the concentration of ionized calcium as the increase in
hydrogen ions displaces calcium from albumin. However,
the concentration of ionized calcium is difficult to predict
using these "rules of thumb." As most clinical laboratories
have the required electrode for the direct measurement of
ionized calcium, clinicians should request its measurement
and become comfortable with its interpretation.
Intestinal Absorption

The average American diet contains approximately

800 mg (20 mmol) of calcium of which there is generally a
net absorption of approximately 160 mg (4 mmol). Calcium
absorption varies widely depending on other dietary components that may bind calcium (oxalate or phosphate) or
promote absorption (lactose) and the level of serum
1,25(OH)2D3.1,25(OH)2D3 is the principal hormonal regulator of intestinal calcium absorption. Calcium is absorbed
in the duodenum, jejunum, and ileum. Calcium moves
across the brush border down its concentration gradient
by a l,25(OH)2D3-facilitated mechanism, through the cell
bound to l,25(OH)2D3-induced calcium binding proteins,
and then is transported across the basolateral membrane
against a steep gradient utilizing a calcium ATP-ase or
through a sodium for calcium exchange mechanism.
Renal Excretion

A 70-kg male with a glomerular filtration rate of 180 L/

day filters approximately 270 mmol of calcium per day
(180 L/day X 1.5 mmol/L). This quantity of calcium, over
10 g, is far more than the entire extracellular fluid calcium
content. The kidney reabsorbs approximately 98% of the
ultrafiltered calcium, leading to a urine calcium excretion of
approximately 4 mmol/day. Approximately 65% of filtered
calcium is reabsorbed in the proximal tubule. Proximal
reabsorption is linked to sodium reabsorption and does
not appear to be under hormonal control. Approximately
25% of filtered calcium is reabsorbed in the loop of Henle.
In this segment, the lumen-positive voltage provides a
strong driving force for calcium reabsorption. In addition,
there appears to be active transport of calcium. The distal
convoluted tubule is responsible for approximately 8% of
calcium reabsorption, and is the major site for the regulation of urine calcium excretion. In this segment, active
calcium transport occurs against both electrical and chemical gradients. Distal convoluted tubule calcium reabsorption is stimulated by parathyroid hormone (PTH) and phosphorus depletion and is inhibited by metabolic acidosis and
adrenocortical excess. Thiazide diuretics increase reabsorp tion, whereas loop diuretics, such as furosemide, decrease
reabsorption. There appears to be some, albeit quite small,
calcium reabsorption in the collecting tubule.
In health, daily urine calcium excretion equals intestinal
calcium absorption. However, during renal insufficiency,
urine calcium excretion falls due not only to the decreased
glomerular filtration rate but to enhanced calcium reabsorption secondary to the usual increase in PTH. The increase in PTH is multifactorial; it appears secondary to the

decrease in renal 1,25(OH) 2D3 synthesis, resulting from

a decrease in renal mass and from an increas e in serum
phosphorus due to a decrease in phosphorus excretion.
The fall in 1,25(OH)2D3 decreases calcium absorption, resulting in hypocalcemia and subsequent increased PTH
secretion. In addition, 1,25(OH)2D3 itself has a significant
inhibitory effect on PTH secretion. Recent studies have
shown that increased levels of serum phosphorus will increase serum levels of PTH independent of alterations in
calcium or 1,25(OH)2D3. Renal osteodystrophy is covered
in detail in Chapter 69.
Regulation of Serum Levels

Large amounts of calcium must be transported through

the extracellular fluid during periods of rapid skeletal
growth and pregnancy, yet the concentration of serum calcium usually varies by less than 10%. This precise regulation
of serum calcium concentra tion is accomplished by a complex interaction of the intestine, bone, and kidney involving
the principal calcium-regulating hormones: PTH and
1,25(OH) 2D3. PTH stimulates renal calcium reabsorption
and bone mineral turnover and increases the serum level
of 1,25(OH)2 D3. A principal role of 1,25(OH)2D3 is to
stimulate intestinal calcium absorption. Numerous cellular
activities are regulated by calcium or are dependent on a
stable serum calcium concentration for normal function.
Alterations in serum calcium concentration result in disturbances of cellular, especially neuronal, function.
DISORDERS OF CALCIUM HOMEOSTASIS
Hypocalcemia

Hypocalcemia may be defined as a reduction in the ionized component of serum calcium. Patients who have a
decrease in total serum calcium concentration may, or may
not, have a reduction in ionized calcium, as hypoalbuminemia is very prevalent in hospitalized patients and because
albumin binds to the majority of protein-bound calcium.
If there is doubt about the diagnosis, ionized calcium should
be measured directly.
Clinical Presentation

The clinical presentation of patients with hypocalcemia

correlates with the rapidity and magnitude of the fall in
serum calcium, but symptoms are manifest in many patients
with a total serum calcium of 7.0 mg/dL or less. The principal clinical findings are neurological. Perioral paresthesias
are followed by carpopedal spasm. Occasionally, patients
develop laryngeal stridor. Tetany may be evoked by acute
respiratory alkalosis caused by hyperventilation, due to the
rapid fall in ionized calcium concentration. As the pH rises
(a decrease in hydrogen ion concentration), hydrogen ions
dissociate from albumin, promoting increased calcium
binding.
Two clinical signs may indicate hypocalcemia: Chvostek's and Trousseau's. Chvostek's sign is evoked by tapping
the facial nerve and observing a grimace. Trousseau's is
evoked by inflating a blood pressure cuff 3 mm Hg above

108

David A. Bushinsky

the systolic pressure for at least 3 minutes and observing

spasm of the outstretched hand. Trousseau's sign is more
specific, as Cvostek's sign is present in approximately 10%
of normocalcemic individuals.
Patients with hypocalcemia may also present with generalized seizures which represent whole body tetany. The
electrocardiogram in hypocalcemic patients has a characteristic prolonged corrected QT and ST intervals. There
may be peaked T waves, arrhythmias, and heart block.
Causes (Table I)

Chronic Renal Insufficiency. Decreased glomerular nitration decreases renal phosphorus excretion, resulting in
hyperphosphatemia. The increased serum phosphorus not
only complexes with serum calcium producing hypocalcemia but also downregulates the la-hydroxylase responsible for the renal conversion of 25(OH)D3 to 1,25(OH)2D3.
Chronic renal insufficiency also results in a reduction of
functional renal mass and decreased 1,25(OH)2D3 production. Serum levels of 1,25(OH)2D3 are low, resulting in
decreased intestinal calcium absorption and hypocalcemia.
Levels of PTH and the osteoblastic enzyme alkaline phosphatase tend to be elevated.
Following Parothyroidectomy. Surgical reduction of parathyroid mass in patients with secondary or tertiary hyperparathyroidism usually leads to profound hypocalcemia,
due to bone remineralization. This hungry bone syndrome
may require prolonged and vigorous calcium replacement.
Hypoparathyroidism. Both idiopathic and postsurgical
hypoparathyroidism result in a deficiency of PTH. Renal
calcium excretion is increased and 1,25(OH)2D3 production
and bone turnover decreased. Serum levels of PTH are
either low for the level of serum calcium or undetectable.
Pseudohypoporathyroidism. In this hereditary disorder,
the target cell response to PTH is decreased. PTH level is
elevated and, in most patients, cyclic-AMP does not rise
normally in response to PTH. Patients also commonly have
shortened metacarpals and metatarsals in addition to short
stature, obesity, and heterotopic calcification.
Malignant Disease. The most frequent cause of a decrease in total calcium concentration in ill patients with

malignant disease is a decrease in serum albumin concentration; ionized calcium may be normal. Certain malignancies, such as prostate and breast cancer, may cause
enhanced osteoblastic activity and accelerated bone formation, resulting in hypocalcemia. In the tumor lysis syndrome, the rapid cell destruction in response to chemotherapy may result in an increase in serum phosphorus which
then complexes serum calcium, resulting in hypocalcemia.
Rhabdomyo/ysis. Cellular injury, especially due to rapid
crush injuries, causes a rapid release of cellular phosphorus
which complexes with extracellular calcium, resulting in
hypocalcemia. Associated renal failure blocks the expected
PTH-induced rise in 1,25(OH)2D3.
Hypomognesemio. Hypocalcemia and hypomagnesemia
frequently coexist and are often due to decreased absorption of dietary divalent cations or poor dietary intake. Hy pomagnesemia may impair PTH secretion and interfere
with its peripheral action.
Acute Pancreatitis. Acute pancreatitis leads to the release
of pancreatic lipase which degrades retroperitoneal and
omental fat which then binds calcium in the peritoneum
removing it from the extracellular fluid and resulting in
hypocalcemia. Hypomagnesemia and hypoalbuminemia
also have been reported to contribute to the hypocalcemia
of acute pancreatitis.
Septic Shock. Endotoxic shock is associated with hypocalcemia through mechanisms which are not well delineated.
As myocardial function correlates directly with ionized calcium concentration, hypocalcemia in this condition may be
responsible, in part, for the hypotension.
Vitamin D Deficiency. There are numerous causes of vitamin D deficiency including renal insufficiency, as noted
above. Dietary deficiency is uncommon in the United States
because vitamin D is added to various foods. This fatsoluble vitamin is subject to malabsorption and levels may
be low in chronic liver disease and primary biliary cirrhosis.
Anticonvulsant therapy with any of several agents increases
the turnover of vitamin D into inactive compounds and
results in a decrease in serum levels of 1,25(OH)2D3 .
Treatment

TABLE I Principal
Causes of Hypocalcemia

Chronic renal insufficiency

Following parathyroidectomy

Hypoparathyroidism
Pseudohypoparathyroidism
Malignant disease

Rhabdorayolysis
Hypomagnesemia
Acute pancreatitis

Septic shock
Vitamin D deficiency

Acute. Patients with symptoms attributable to hypocalcemia must be treated promptly. In addition, asymptomatic
patients with a serum calcium of approximately 7.0 mg/dL
or less also should be treated prophylactically. Intravenous
calcium is the mainstay of treatment. Calcium gluconate
may be administered as 10 mL of a 10% solution (0.47 mEq
calcium/mL) administered over 10 minutes. Patients on
digoxin may require EKG monitoring. To avoid precipitation of the insoluble salt calcium carbonate, calcium must
not be given in the same intravenous line as bicarbonate.
If a patient is acidemic and hypocalcemic, in general the
hypocalcemia must be treated first because the acidemia
increases the proportion of ionized calcium and thus pro-

14. Disorders of Calcium and Phosphorus Homeostasis

tects the patient against symptomatic hypocalcemia. A constant infusion of calcium gluconate (50 mL of 10% calcium
gluconate in 500 mL of D5W) may be administered every
8 hours.
Chronic. In patients with normal renal function, chronic
hypocalcemia is treated by administering oral calcium and
1,25(OH)2D3. The former, often given as calcium carbonate, must provide at least 1 g of elemental calcium each
day. 1,25(OH)2D3 therapy will promote intestinal calcium
absorption; however, in the case of hypoparathyroidism, it
will also increase urine calcium excretion because PTH is
not present to promote renal calcium reabsorption. Thus,
the goal of therapy is to keep serum calcium concentration
at the lower limit of normal, approximately 8.0 mg/dL,
and to keep urine calcium excretion below 350 mg/day.
Thiazide diuretics may be utilized to help prevent hypercalciuria and to promote normocalcemia. 1,25(OH)2D3 administered without oral calcium may result in resorption
of bone mineral and osteopenia.
Hypercalcemia
Hypercalcemia is defined as an increase in the concentration of serum ionized calcium.
Clinical Presentation
The clinical signs and symptoms of hypercalcemia correlate with the rapidity and magnitude of the elevation in
serum calcium. As with hypocalcemia, neurologic abnormalities predominate. Often patients present with drowsiness and lethargy followed by headache and irritability.
Confusion may be followed by stupor and coma. Muscle
weakness, emotional problems, and depression may occur.
Polyuria is frequent in patients with hypercalcemia due
to a renal concentrating defect. The excess calcium is
thought to impair the renal response to antidiuretic hormone. Hypercalcemia can lead to a decline in renal function
by a variety of mechanisms. Polyuria may lead to volume
depletion and prerenal azotemia. Excess calcium excretion
may produce nephrocalcinosis, especially in the presence
of alkaline urine which decreases the solubility of calcium
phosphate complexes.
Calcium directly increases cardiac contractility, and patients with hypercalcemia are thought to have an increased
incidence of hypertension. Hypercalcemia shortens the corrected QT interval, may broaden the T waves, and may
produce first-degree AV block. Anorexia, nausea, and severe vomiting are associated with hypercalcemia, as is constipation.
Causes (Table 2)
Primary Hyperparathyroidism. Accounting for more than
50% of patients, primary hyperparathyroidism is the leading cause of hypercalcemia. Patients are typically women
60 years of age or older. Most have a benign adenoma of
a single parathyroid gland, whereas others have hyperplasia
of all four glands. Parathyroid carcinoma is extremely rare.
The elevated PTH levels increase both renal calcium reab-

109

TABLE 2

Principal Causes of Hypercalcemia

Primary hyperparathyroidism

Malignancy
Renal failure

Following renal transplant

Thiazide diuretics
Immobilization

Milk-alkali syndrome
Granulomatous disease

Familial hypocalciuric hypercalcemia

Thyrotoxicosis

Vitamin intoxication
Theophylline toxicity

sorption and phosphorus excretion. Proximal tubule bicarbonate reabsorption is impaired and, as expected, urinary
cyclic-AMP is elevated. The excess PTH also increases the
serum level of 1,25(OH)2D3, increasing enhanced intestinal
calcium absorption and bone turnover, with bone resorption predominating over bone formation.
Approximately 6% of patients with calcium-containing
kidney stones have hyperparathyroidism. In this case the
increased filtered load of calcium exceeds the increased
renal calcium reabsorption, leading to hypercalciuria, renal
stone formation, and occasionally nephrocalcinosis. Increased urinary bicarbonate alkalinizes the urine, promoting precipitation of calcium phosphorus complexes. With
current automated blood chemical analysis, hypercalcemia
from hyperparathyroidism is generally detected before
stone formation and/or nephrocalcinosis occurs.
Malignancy. Malignancy is the second most frequent
cause of hypercalcemia. Both direct bone destruction by
the growing tumor and secretion of calcemic factor(s) by
malignant cells may produce hypercalcemia. Patients with
squamous cell lung carcinoma and metastatic carcinoma of
the breast develop hypercalcemia most frequently, whereas
patients with myeloma, T-cell tumors, renal cell carcinoma,
and other squamous cell tumors are also prone to hypercalcemia. Many tumors produce a PTH-related peptide (PTHrP) in which the first 13 amino acids are very similar to
those in PTH and which binds to PTH receptors in the
kidney and bone. Although PTHrP is not detected on standard PTH assays, commercial assays are now available.
Other tumors produce factors such as transforming growth
factor-a or interleukin-1, cytokines such as lymphotoxin,
or hormones such as 1,25(OH)2D3.
In general, the malignancy is evident when patients present with hypercalcemia. The finding of a low level of PTH
and an elevated level of PTH-rP supports the diagnosis.
While hypercalcemia due to an occult malignancy is rare,
associated symptoms such as weight loss and fatigue should
prompt a search for tumors that are frequently associated
with hypercalcemia, such as those in the lung, kidney, and
urogenital tract.

no

David A. Bushinsky

Renal Failure. Hypocalcemia, and not hypercalcemia, is

generally found in patients with renal failure. In these patients, it is important to determine if the hypercalcemia
was responsible for the renal failure. The hypercalcemia
caused by diseases such as sarcoidosis, myeloma, immobilization, and milk-alkali syndrome frequently causes renal
failure. Hypercalcemia may occur during the recovery
phase of rhabdomyolysis -induced renal failure because calcium recently deposited in muscle and soft tissues is rapidly mobilized.
Hypercalcemia is frequently observed during excessive
1,25(OH) 2D3 replacement therapy of patients on dialysis,
especially if they are simultaneously being given large
amounts of oral calcium as a phosphorus binder. Some
patients may develop severe secondary hyperparathyroidism with marked hyperplasia of the parathyroid glands and
subsequent hypercalcemia. In addition, patients may have
aluminum intoxication, a disorder characterized by low
bone turnover, which predisposes them to hypercalcemia.
These aluminum intoxicated patients often have modestly
elevated PTH and alkaline phosphatase levels (see Chapter 69).

25(OH)D3 to 1,25(OH)2 D3 by the granulomatous tissue.

This phenomenon has been observed in anephric patients.

Following Renal Transplant. Frequently, patients on longterm dialysis develop parathyroid hyperplasia leading to
autonomous secretion of PTH. If these patients then
receive a successful renal transplant, the PTH secretion
continues and hypercalcemia may develop due to enhanced, PTH-induced, renal calcium reabsorption. The hypercalcemia is generally mild and tends to decrease over the
ensuing 6 to 12 months as the hypertrophied parathyroid
glands involute; however, patients with prolonged marked
hypercalcemia may require surgical parathyroidectomy.

Treatment

Thiazide Diuretics. Thiazide diuretic therapy is often as sociated with mild hypercalcemia due to increased renal
calcium reabsorption. If the hypercalcemia does not resolve
when the thiazides are discontinued, then the patient must
be investigated for other causes of hypercalcemia.
/mmobifization. Immobilization leads to a rapid increase
in bone resorption and may lead to hypercalcemia, especially if there is any decrement in renal calcium excretion.
This disorder is most prevalent in younger patients who
sustain a traumatic spinal cord injury.
Milk-Alkali Syndrome. Ingestion of large amounts of
calcium-containing nonabsorbable antacids may lead to hypercalcemia, alkalemia, nephrocalcinosis, and renal insufficiency. While this disorder has become less common because ulcers are being treated with antibiotics and agents
that inhibit gastric acid secretion, the use of calcium and
alkali preparations is increasing in efforts to prevent and
treat osteoporosis.
Gronu/omatous Disease. Granulomatous diseases, such
as sarcoidosis, tuberculosis, and leprosy, may produce hypercalciuria and hypercalcemia due to the conversion of

Familial Hypocalciuric Hypercalcemia. This autosomal

dominant disorder causes mild hypercalcemia, hypophosphatemia, and reduced renal calcium excretion (often less
than 100 mg/day). These patients have a mutation in the
recently identified calcium receptor that causes a reduction
in receptor activity. PTH levels are normal and parathyroidectomy is not indicated.
Thyrotoxicosis. Excess thyroid hormone can stimulate osteoclastic bone resorption, resulting in hypercalcemia. The
hypercalcemia is generally mild and coexistent hyperparathyroidism must be excluded.
Vitamin Intoxication. Vitamin D intoxication, as noted
above, is often observed in patients with end-stage renal
disease treated with 1,25(OH)2D3. Vitamin D intoxication
can also occur in food or vitamin faddists. Vitamin A excess
can also cause hypercalcemia and may be seen in the
same patients.
Treatment of hypercalcemia should be directed at the
underlying disorder. However, if the patient is symptomatic
and the serum calcium level is 13 mg/dL or higher acute
therapy is indicated. The mainstay of treatment in patients
with reasonable renal and cardiac function is intravenous
saline. Rates as high as 200 to 250 mL/h of normal saline
may be used to facilitate renal calcium excretion. Furosemide is calciuric and is often necessary (40 mg iv in patients
with normal renal function) during volume repletion with
saline to avoid pulmonary congestion. Thiazide diuretics
must be avoided because they decrease renal calcium excretion. Patients with renal insufficiency may be unable to
excrete the sodium load and thus hemodialysis against a
low calcium bath may be necessary to lower serum calcium
concentration. Patients with cardiac or renal insufficiency
are at particular risk for volume overload during saline
treatment and may require central monitoring and a more
modest normal saline infusion.
Patients with hypercalcemia due to enhanced intestinal
calcium absorption may benefit from a reduction of dietary
calcium intake. Normally, Americans consume approximately 800 mg of calcium per day, reducing this by onethird is generally effective. However this should only be a
short-term remedy, as the provision of a low calcium diet
in a patient with normal renal function and excess
1,25(OH)2D3 will promote bone demineralization. Glucocorticoids (initial dose 40 mg of prednisone) will inhibit
intestinal calcium absorption but may require 1 week for
maximal effect.
Patients with hypercalcemia mediated by enhanced osteoclastic bone resorption (usually patients with malignancy), will benefit from the osteoclastic inhibitor calcitonin (4-8 units/kg sc or im 6-12 hours). However, the
hypocalcemic effect of calcitonin is often transient. The
bisphosphonate etidronate (7.5 mg kg body wt" 1 day" 1

III

14. Disorders of Calcium and Phosphorus Homeostasis

infused iv for 3 days) lowers serum calcium in a few days

with a maximal effect in 7 days. Pamidronate (30-90 mg/
day given as a single 24-hour infusion for 3 days) is a more
potent bisphosphonate than etidronate. Both bisphosphonates have been shown to cause renal insufficiency and
should be avoided in patients with renal impairment. Gallium nitrate will effectively lower serum calcium; however,
it is contraindicated in patients with renal insufficiency or
in those receiving other nephrotoxic agents.
Oral phosphorus can be utilized to treat hypercalcemia.
To prevent induction of soft-tissue and calcium degeneration, it should be employed only if the serum phosphorus
is at or below the lower limit of normal. However, this
form of outdated therapy is best avoided.
PHOSPHORUS
Phosphorus Homeostasis
Distribution

The majority (90%) of total body phosphorus is contained within the bone mineral while 10% is contained
within the cells and 1% in the extracellular fluid.
Serum Concentration

The concentration of serum phosphorus in adults ranges

from 2.5 to 4.5 mg/dL or 0.81 to 1.45 mM. Serum phosphorus levels are highest in infants and decrease in childhood
reaching the normal adult levels in late adolescence. Approximately 70% of blood phosphorus is termed organic
and contained within phospholipids; the remaining 30% is
termed inorganic. Of the inorganic phosphorus, 85% is free
and circulates as monohydrogen or dihydrogen phosphate
or is complexed with sodium, magnesium or calcium. The
remainder is protein bound.
Intestinal Absorption

Dietary phosphorus intake varies between 800 and

1850 mg of PO4/day (26-60 mmol/day). Approximately
25 mmol/day is absorbed. Diets that are adequate in calories and protein generally contain adequate phosphorus.
Phosphorus absorption is regulated by 1,25(OH)2D3 via a
sodium-dependent active transport mechanism principally
in the duodenum and by a passive phosphorus concentration dependent mechanism in the jejunum and ileum. The
absorption of phosphorus is hindered by complex forma tion in the intestine; both aluminum and calcium can form
insoluble complexes with phosphorus that hinder its absorption. During renal failure, phosphorus absorption continues. The formation of intestinal phosphorus complexes
with aluminum and/or calcium is used to decrease intestinal
absorption in these patients who are unable to excrete
absorbed phosphorus. Use of calcium to complex intestinal
phosphorus is favored over aluminum because the latter
is toxic to many organs including brain and bone (see Chapters 69 and 71).

1.25 mM filters approximately 200 mmol of phosphorus

per day as approximately 85% of serum phosphorus is ultrafiltrable (180 L/day X 1.25 mmol/L X 0.85). Because urine
phosphorus excretion averages about 25 mmol/day, approximately 12.5% of the glomerular filtrate is excreted in
the urine. The bulk of phosphorus reabsorption (approximately 85%) occurs in the proximal tubule. Proximal phosphorus transport across the apical membrane occurs by a
rate-limiting sodium-phosphorus cotransport. Phosphorus
resorption occurs in the pars recta but not in the loop of
Henle. A small quantity of phosphorus is reabsorbed in
the distal convoluted tubule, especially in the absence of
PTH. Whether there is phosphorus reabsorption more dis tally is not known.
Parathyroid hormone is the major hormonal regulator
of phosphorus excretion. Variation in dietary phosphorus
intake is the major nonhormonal regulator. PTH is phosphaturic as are 1,25(OH)2 D3, extracellular fluid volume
expansion, and a high phosphorus diet. A low phosphorus
diet leads to a decrease in renal excretion. The hypophosphaturia of phosphorus depletion overrides the hyperphosphaturia induced by PTH. During renal insufficiency urine
phosphorus excretion remains relatively constant until the
glomerular filtration rate falls to about 25% of normal.
Phosphorus excretion is maintained despite the fall in glomerular filtration rate and filtered load of phosphorus by
the phosphaturic effects of elevated PTH levels. (See Renal
Calcium Excretion above.)
Regulation of Serum Levels

Serum levels of phosphorus are not as tightly regulated

as those of calcium. They may vary by as much as 50%
over the course of a day. Although there is diurnal variation
in phosphorus levels, alterations in dietary phosphorus intake are principally responsible for the swings in serum
levels. In addition, serum levels are controlled by hormones
such as PTH and 1,25(OH)2D3 and by the status of extracellular fluid volume and renal function.
DISORDERS OF PHOSPHORUS HOMEOSTASIS
Hypophosphatemia

A decrease in the level of serum phosphorus may or may

not reflect total body phosphorus content because only 1%
of body phosphorus is in the extracellular fluid. Moderate
hypophosphatemia may be defined as a serum phosphorus
between 1 mg/dL and the lower limit of normal and severe
hypophosphatemia as a serum level below 1 mg/dL. While
moderate hypophosphatemia is generally asymptomatic,
severe hypophosphatemia is associated with a variety of
clinical disturbances and, occasionally, death. Hypophosphatemia may result from a decrease in intake, and increase
in excretion, or a shift of phosphorus from the extracellular
environment into cells.

Renal Excretion

Clinical Presentation

A 70-kg male with a glomerular filtration rate of 180 L/

day and a mean serum phosphorus concentration of

Patients with severe hypophosphatemia may have neuro logic dysfunction characterized by weakness, paresthesias,

112

David A. Bushinsky

confusion, seizures, and coma. The weakness may be associated with muscle edema and rhabdomy olysis. Respiratory
muscle paralysis may result in death.
Causes (Table 3)

Alcohol Related. Many chronic ethanol abusers are hypophosphatemic on hospital admission or become hypophosphatemic with treatment. Although the etiology of their
hypophosphatemia is multifactorial, it is due in large part
to poor oral intake.
Refeeding. When patients are fed after prolonged poor
intake or starvation the calories provide a stimulus for
tissue growth and utilization of phosphorus in phosphory lated intermediates such as ATP. If the diet does not contain adequate phosphorus then severe hypophosphatemia
may develop after several days. Likewise, if total parenteral
nutrition (TPN) solutions contain inadequate phosphorus
then hypophosphatemia may become evident as the patient
regains body mass.
Diabetes Meffitus. In severe diabetic ketoacidosis, especially of prolonged duration, there is excessive urinary
phosphorus loss, which is accompanied by poor oral phosphorus intake, that may lead to hypophosphatemia and
also hypokalemia. The severity of the hypophosphatemia
may become manifest only during treatment of the diabetic ketoacidosis.
Alkalosis. Both respiratory and metabolic alkalosis induce
hypophosphatemia; however, hypophosphatemia is far
more severe in prolonged respiratory alkalosis. The extracellular alkalosis appears to cause intracellular alkalosis
which results in a shift of phosphorus into the intracellular
space and hypophosphatemia.

Treatment

The treatment of hypophosphatemia is best directed at

reversing the underlying disease or nutritional process.
Moderate hypophosphatemia can often be treated with
oral phosphorus supplementation in the form of milk (an
excellent source of phosphorus containing about 1 g of
phosphorus per liter), or Neutra-Phos tablets (250 mg of
phosphorus). Oral phosphorus can cause diarrhea which
is usually seen at doses of over 1 g/day. Severe hypophosphatemia usually indicates a significant loss of total body
phosphorus. In asymptomatic patients if oral replacement
is possible, the patient should receive at least 3 g/day for
1 week. Often the patient will be symptomatic and require
intravenous therapy. In this case, treatment may be initiated with 2 mg/kg body wt of phosphorus as the sodium
salt infused over 6 hours. Serum phosphorus levels must
be checked frequently during replacement. Intravenous
phosphorus may produce hypocalcemia and metastatic calcification especially if the calcium phosphorus product exceeds 60. Concurrent hypokalemia and hypomagnesemia
are often found in patients with hypophosphatemia.
Hyperphosphatemia

In adults, hyperphosphatemia is defined as an increase

in the concentration of serum phosphorus to greater than
5.0 mg/dL. Because cells contain abundant phosphorus,
hemolysis of the collected sample should be excluded, especially if there is associated hyperkalemia. Hypophosphatemia may be caused by decreased phosphorus excretion, an
increase in phosphorus load, or a shift of phosphorus from
the cells into the extracellular fluid.
Clinical Presentation

Alcohol related

The clinical presentation of patients with hyperphosphatemia is dominated by the associated fall in serum calcium.
As serum phosphorus rises there is a reciprocal fall in serum
calcium. This fall is multifactorial but includes a decrease
in 1,25(OH)2D3 synthesis, leading to a decrease in intestinal
calcium absorption and formation of calcium phosphorus
complexes resulting in ectopic calcification especially of
previously injured tissues. The symptoms of patients with
hyperphosphatemia then are those of patients with hypocalcemia, tetany, seizures, and decreased myocardial contractility. In addition, ectopic calcification may occur in
virtually any organ in the body, especially when the calcium
phosphorus product exceeds 60. Calcification is especially
prominent in proton-secreting organs, such as the stomach
or kidney, in which basolateral bicarbonate secretion results in an increase in pH that promotes calcium hydrogen
phosphate (brushite) precipitation. On a chronic basis, the
hyperphosphatemia of renal failure can result in secondary
hyperparathyroidism and renal osteodystrophy.

Refeeding
Diabetes mellitus

Causes (Table 4)

Postrenal Transplant. Hypophosphatemia is commonly

observed following renal transplantation. The etiology is
multifactorial but related to a persistent elevation in serum
PTH levels and an intrinsic renal tubular defect in phosphorus reabsorption that results in hyperphosphaturia. In addition, glucocorticoids used in immunosuppression inhibit
renal phosphorus transport.
Urinary Loss. Renal tubular acidosis, hypokalemia, hypomagnesemia, and other renal tubular disorders are associated with an increased urinary excretion of phosphorus.

TABLE 3

Principal Causes of Hypophosphatemia

Alkalosis
Postrenal transplant

Urinary loss
Total parenteral nutrition

Renal Failure. With a fall in the glomerular filtration rate

below approximately 25 ml per minute renal excretion of
phosphorus is less than intestinal absorption and hyperphosphatemia ensues. Increased levels of phosphorus have

113

14. Disorders of Calcium and Phosphorus Homeostasis

TABLE 4
Principal Causes of Hyperphosphatemia

Renal failure
Hypoparathyroidism
Cell injury
Exogenous administration

been shown to directly increase the level of PTH. The

increased levels of phosphorus also suppress renal
1,25(OH)2D3 production, resulting in less calcium absorption. In addition, the excess phosphorus may bind serum
calcium if the calcium phosphorus product exceeds 60. Levels of serum 1,25(OH)2D3 fall due to the reduction in mass
of the renal tissue that converts 25(OH)D3 to 1,25(OH)2D3.
The resulting hypocalcemia and low levels of 1,25(OH)2D3
further increase PTH secretion and urine phosphorus excretion, lowering the level of serum phosphorus at the cost
of hyperparathyroidism. In spite of the additional phosphorus excretion per nephron, with more profound renal insufficiency, serum phosphorus increases as the patient is unable to excrete the absorbed phosphorus. During chronic
renal failure the hyperphosphatemia is less severe than
during the same decrement in glomerular filtration rate in
acute renal insufficiency when there has been insufficient
time for adaptation (see Chapter 69).
Hypoparathyroidism. PTH promotes renal phosphorus
excretion and hyperphosphatemia results when there is a
lack of PTH or a resistance to its action.

sorption as does increasing urine pH with sodium bicarbonate or acetazolamide. In patients with hyperphosphatemia
and acute renal failure, glucose and insulin transiently increases cellular phosphorus uptake; however, phosphorus
is best removed from the extracellular space with dialysis.
The successful treatment of hyperphosphatemia of
chronic renal failure requires a coordinated effort among
the patient, dietician, and physician. A diet devoid of phosphorus is unpalatable; however, dietary phosphorus can be
reduced substantially with proper dietary supervision. Both
hemodialysis and peritoneal dialysis remove phosphorus;
but even with a low phosphorus diet they cannot restore
phosphorus balance. Agents that bind dietary phosphorus
and prevent its absorption are generally necessary to prevent and treat hyperphosphatemia. Calcium salts given with
meals are effective in binding dietary phosphorus. Doses
are gradually increased until the serum phosphorus is less
than approximately 5 mg/dL. If hypercalcemia occurs before the serum phosphorus is controlled, then the dialysate
calcium may be lowered and aluminum gels added to bind
phosphorus. If patients present with severe hyperphosphatemia (serum phosphorus greater than 6.5 mg/dL), the
serum phosphorus should be lowered with aluminumbinding gels prior to adding calcium salts to avoid soft
tissue calcification. However, aluminum is toxic to the
brain, bone, and bone marrow and should be avoided whenever possible (see Chapter 69 for details).
ACKNOWLEDGMENTS
This work was supported by grants AM-39906 and AR-33949
from the National Institutes of Health.

Bibliography

Cell Injury. During effective chemotherapy, especially of

rapidly lysing cells such as lymphomas, there is a marked
increase in phosphorus release from cells which exceeds
renal excretory capacity. The hyperphosphatemia is often
associated with hyperkalemia and hyperuricemia and leads
to hypocalcemia. Phosphorus can also be released from
cells during acute rhabdomyolysis, crush injuries, or tissue
infarction. Hyperphosphatemia is especially severe when
the release of cellular phosphorus occurs in the setting of
acute renal failure.
Exogenous Administration. Excess phosphorus in TPN
solutions or phosphorus in laxatives may result in hyperphosphatemia. Phosphorus-containing enemas can result
in hyperphosphatemia and should be avoided in patients
with renal failure. Excess vitamin D not only increases
calcium absorption but phosphorus absorption as well and
can result in hyperphosphatemia.
Treatment

The treatment of hyperphosphatemia must be directed

at the underlying cause. If renal function is intact then
phosphaturia should be promoted. Extracellular fluid volume expansion with saline lowers renal phosphorus reab-

Brown EM, Pollack M, Seidman, CE, et al: Calcium -ion-sensing

cell-surface receptors. N Engl J Med 333:234-240, 1995.
Bushinsky DA, Krieger NS: Integration of calcium metabolism in
the adult. In: Coe FL, Fauus MJ (eds) Disorders of Bone and
Mineral Metabolism, Raven Press, New York, pp. 417-432,1992.
Bushinsky DA, Krieger NS: Role of the skeleton in calcium homeostasis. In: Seldin DW, Giebisch G (eds) The Kidney: Physiology and Pathophysiology, Raven Press, New York, pp. 23952430, 1992.
Kovach KL, Hruska KA: Phosphate balance and metabolism. In:
Jacobson HR, Striker GE, Klahr S (eds) The Principles and
Practice of Nephrology, Mosby, St. Louis, pp. 986-992, 1995.
Kumar R: Calcium metabolism. In: Jacobson HR, Striker GE,
Klahr S (eds) The Principles and Practice of Nephrology, Mosby,
St. Louis, pp. 964-971, 1995.
Levi M, Cronin RE, Knochel JP: Disorders of phosphate and
magnesium metabolism. In: Coe FL, Favus MJ (eds) Disorders
of Bone and Mineral Metabolism, Raven Press, New York, pp.
587-610, 1992.
Mundy GG: Hypercalcemia in hematologic malignancies and in
solid tumors associated with extensive localized bone destruc tion. In: Favus MJ (ed) Primer on the Metabolic Bone Diseases
and Disorders of Mineral Metabolism, Raven Press, New York,
pp. 173-176, 1993.
Suki WN, Rouse D: Renal transport of calcium, magnesium, and
phosphorus. In: Brenner BM (ed) The Kidney, 5th ed. Saunders,
Philadelphia, pp. 472-515, 1996.

EDEMA AND THE CLINICAL USE OF

DIURETICS
DAVID H. ELLISON

Edema is usually a manifestation of expanded extracellular fluid (ECF) volume, whether it results from conges tive heart failure, cirrhosis, nephrotic syndrome, or renal
failure. Extracellular fluid volume expands when the
kidneys retain NaCl in excess of dietary NaCl ni take.
Renal NaCl retention may reflect an adaptive response
to inadequate effective arterial blood volume, as in patients with congestive heart failure. Renal NaCl retention
may also reflect a pathological response of kidney tubules
to damage, as in patients with acute renal failure. Regardless of its cause, the best treatment for edema is to
restrict dietary NaCl intake and to correct the primary
disorder. Despite these interventions, or when they are
impossible, ECF volume frequently remains expanded
unacceptably and, for this reason, diuretics are among
the most commonly prescribed drugs. All of the diuretics
used to treat edema increase both Na and water excretion,
regardless of the underlying cause of edema. They are
powerful drugs which, if used carefully, play an important
role in treating symptomatic edema. The prompt and
dramatic improvement of symptoms when intravenous
diuretics are administered to a patient suffering from
acute pulmonary edema remains one of the most gratifying responses in clinical medicine.
In addition to their use for edema, diuretic drugs are
indicated for a wide variety of nonedematous disorders.
Specific details of diuretic treatment of hypertension, acute
renal failure, nephrolithiasis, and hyponatremia are dis cussed in other chapters in this volume. This chapter focuses on the renal mechanisms of diuretic action and the
use of diuretics to treat edema.

Na is reabsorbed. All of the diuretic drugs in clinical use

act primarily on the renal tubules to inhibit Na reabsorption
and increase fractional Na excretion.
A simple and clinically useful classification of diuretic
drugs is based on the sites and mechanisms of their actions
along the nephron (see Table 1). All active NaCl reabsorption by renal epithelial cells is driven by the Na,K-ATPase
pump. This transport protein, which is present at the basolateral membrane (the blood side) of epithelial cells along
the nephron, uses metabolic energy (derived from hydrolysis of ATP) to extrude Na from the cell into the bloodstream and to move K into the cell. The action of the Na,KATPase keeps the cellular Na concentration low and the
cellular K concentration high. It also contributes to making
the cell interior electrically negative with respect to the
ECF. The low cellular Na concentration and the cellnegative voltage drive positively charged Na ions into the
cell across the luminal membrane from tubule fluid. Although Na,K-ATPase pumps are present at the basolateral
cell membranes of nearly all epithelial cells, each nephron
segment possesses unique apical mechanisms that permit
Na to move across the luminal membrane; these specific
transport pathways at the luminal membrane form the cellular basis of diuretic action. Together, active Na extrusion
from the basolateral membrane and passive Na entry across
the luminal membrane permit vectorial Na transport in the
absorptive direction.
Proximal Tubule Diuretics

Approximately two thirds of filtered water and NaCl are

reabsorbed along the proximal tubule. Sodium moves down
its electrochemical gradient from tubule lumen into proximal tubule cells coupled to the movement of other solutes
against their electrochemical gradients; among these solutes are glucose, amino acids, and phosphate. Bicarbonate
and chloride are indirectly coupled to Na absorption. Because the epithelium is electrically "leaky" (highly permeable to ions), large transepithelial ion gradients do not
develop and solute absorption along this segment remains isoosmotic.
An important pathway by which Na crosses the luminal
membrane of proximal tubule cells involves electroneutral
exchange of Na + for H+ (see Fig. 1). H+ ions that are

THE PHYSIOLOGICAL BASIS OF DIURETIC ACTION

The amount of NaCl excreted by the kidneys is the difference between the quantity filtered by the glomeruli and
the quantity reabsorbed by the renal tubules. Assuming a
normal glomerular filtration rate (150 L/day) and a normal plasma Na + concentration (150 mM), 23 mol of
Na + is filtered each day in normal humans (equivalent to
about 3 Ib of table salt). To maintain a normal fractional
Na excretion (FENa) of <1%, more than 99% of the filtered

Primer on Kidney Diseases. Second Edition

114

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

David H. Ellison

115
TABLE I

Physiological Classification of Diuretic Drugs"-1"

Proximal diuretics

Loop diuretics

DCT diuretics

CD diuretics

Carbonic anhydrase inhibitors

Acetazolamide

Na+- K--2Cl- inhibitors

Furosemide
Bumetanide
Torscmide Ethacrynic
Acid

Na- Cl inhibitors
Hydrochlorothiazide
Metolazone
Chlorthalidone
Indapamide' Many
others

Naf channel blockers

Amiloride
Triameterene
Aldosterone antagonists
Spironolactone

" Reproduced with permission from Ellison DH: Ann Intern Med 114:886-894, 1991.
b
DCT, distal convoluted tubule; CD, collecting duct. ' Indapamide may have other
actions as well.

extruded across the luminal membrane of proximal cells

titrate bicarbonate (HCOj), which has been filtered by the
glomeruli, forming carbonic acid (H2CO3). Carbonic acid
then dehydrates to CC>2 and H2O, a reaction catalyzed by
the enzyme carbonic anhydrase in the brush border of
proximal tubule cells. Via these events, Na 1 and HCOs
are functionally reabsorbed across the luminal membrane
into the cell. For transepithelial NaHCO3 reabsorption to
continue at steady state, Na + and HCC>3 must leave the

cell across the basolateral membrane via the Na,K-ATPase

pump and via a NaHCO3 transport pathway. Carbonic anydrase located within proximal tubule cells generates H+
ions for extrusion across the apical membrane and bicarbonate ions which exit across the basolateral membrane.
Carbonic anhydrase inhibitors interfere with enzyme activity both inside the cell and within the brush border. Their
action in the brush border inhibits Na +/H+ exchange by
slowing the rate at which carbonic acid dehydrates, thereby

Lumen

Blood

Lumen
Na

DCT
Diuretics

Na Channel
Blockers

DCT

Lumen
HC03

FIGURE I Predominant sites and mechanisms of action of clinically important diuretic drugs. Patterns identify sites of
action along the nephron and corresponding cell types. PT, ^H proximal tubule; LH, loop of Henle; H i M I TAL, thick
ascending limb cell; DT, distal tubule; ^H DCT, distal convoluted tubule cell; CD, collecting duct; V/A PC, principal
cell; CA, carbonic anhydrase inhibitors are important in their ability to reduce Na+ reabsorption by the renal proximal
tubule. Note that Na* channel blockers probably act along the connecting tubule K\N as well as the collecting duct V/,\_
Spironolactone (not shown) is a competitive aldosterone antagonist and acts primarily in the cortical collecting tubule.
Reproduced with permission from Ellison DH: Ann Intern Med 114:887, 1991.

Blood

15. Edema and the Clinical Use of Diuretics

116

acidifying tubule fluid. Their action inside the cell inhibits

NaHCO3 reabsorption by impairing the intracellular production of HCOs, interfering with basolateral exit of
HCOi. The net result of carbonic anhydrase inhibition is
impaired Na + , HCOj Cl~, and water reabsorption by the
proximal tubule and increased renal Na + , Cr, HCO5 and
water excretion. When administered acutely, these drugs
provoke a moderate alkaline diuresis. When administered
chronically, their natriuretic potency is relatively weak.
Their limited potency reflects several compensatory processes that occur when carbonic anhydrase inhibitors are
administered. First, when NaHCO, reabsorption along the
proximal tubule is inhibited, much of the solute and fluid
that escape reabsorption by the proximal tubule can be
reabsorbed by more distal nephron segments. Second, inhibition of solute reabsorption along the proximal tubule
leads to increased solute delivery to the macula densa. This
activates the tubuloglomerular feedback mechanism, which
suppresses glomerular filtration rate (GFR) and decreases
the amount of Na + , Cl~, and HCO, that is filtered. Finally,
alkaline diuresis induces metabolic acidosis; when serum
HCC>3 concentrations decline, less HCO-j is filtered and
the carbonic anhydrase-dependent component of Na + reabsorption falls.
Because carbonic anhydrase inhibitors are relatively
weak diuretics and because they often result in metabolic
acidosis, their use as diuretic drugs is limited. They are
commonly employed, however, to treat open angle glaucoma, where they reduce the formation of aqueous humor
by as much as 50%. Furthermore, they can be used to
prevent acute mountain sickness and to treat metabolic
alkalosis at times when Cl~ cannot be administered because
of ECF volume expansion. This is especially useful when
respiratory drive is compromised by metabolic alkalosis;
careful use of carbonic anhydrase inhibitors may correct
alkalosis and improve respiratory drive. Carbonic anhydrase inhibitors may also be used in conjunction with other
classes of diuretics to induce diuresis in otherwise resistant
patients (see "Resistance to Diuretics," below).
Loop Diuretics

Approximately 25% of the filtered Nad is reabsorbed

along the loop of Henle. Transcellular NaCl reabsorption
along the medullary and cortical thick ascending limbs is
driven by Na,K-ATPase at the basolateral membrane. An
electroneutral pathway at the luminal membrane carries 1
Na + , 1 K+ , and 2C1~ from tubule fluid into the cell, driven
by the electrochemical gradient for Na + (see Fig. 1). Much
of the K+ that is taken up via this pathway recycles across
the luminal membrane through K' channels. The Na + -K+ 2C1~ pathway, therefore, generates net NaCl reabsorption
and (because of the K+ recycling) a voltage across the
wall of the tubule that is oriented with the lumen positive
relative to blood.
Loop diuretics such as furosemide, bumetanide, and
torsemide inhibit the action of the Na + -K+ -2Cl~ pathway
directly. These diuretics are anions that circulate bound to
protein so that very little diuretic reaches tubule fluid via

filtration. Instead, they are secreted into the lumen of the

proximal tubule by the organic anion transport pathway
and travel downstream to the thick ascending limb where
they bind to a Cl~ site on the transport protein. Although
the mechanism by which ethacrynic acid inhibits NaCl reabsorption is not as clear, its net effect on transport along
the thick ascending limb is qualitatively similar. Loop diuretics are potent ("high ceiling") drugs that promote the
excretion of Na + and Cl~, together with K+ . Although they
inhibit K+ reabsorption along the thick ascending limb,
their effects on K+ excretion reflect predominantly their
tendency to increase K+ secretion along the distal nephron
(see "Complications of Diuretic Treatment" below). Loop
diuretics increase magnesium and calcium excretion. Bumetanide, furosemide, and torsemide reduce the magnitude
of the lumen positive voltage in the thick ascending limb.
This tends to impair Ca 2+ and Mg 2+ reabsorption and appears to account for the actions of these drugs on Ca 2+ and
Mg 2+ excretion.
Loop diuretics impair the ability of the kidney to elaborate urine that is either very concentrated or very dilute.
The Na+ -K+ -2Cr pathway removes Na+ and Cl~ from the
lumen as fluid courses up the thick ascending limb. Because
this segment of the nephron is impermeable to water, solute
removal dilutes the tubule fluid. By blocking the predominant solute removal pathway, these agents inhibit free water generation. The action of the Na + -K+ -2Cl~ pathway
also provides the "single effect" that is responsible for
countercurrent multiplication. Solute removal from the
thick ascending limb contributes to generating a high solute
concentration in the medullary interstitium which drives
water reabsorption from the medullary collecting tubule.
By blocking the Na+ -K+ -2Cr pathway, loop diuretics inhibit the kidney's ability to generate a concentrated urine;
this is one reason that these diuretics can be useful for
treatin g patients with the syndrome of inappropriate antidiuretic hormone secretion.
Loop diuretics have important hemodynamic effects
both within the kidney and systemically. They increase
secretion of vasodilatory prostaglandins and often reduce
cardiac preload, when administered acutely. In some situations, however, they elicit a vasoconstrictor response that
may impair cardiac performance acutely; this anomalous
response may be blocked by angiotensin-converting enzyme (ACE) inhibitors. Loop diuretics tend to maintain
or increase the GFR, even in the face of ECF volume
depletion, because they block the tubuloglomerular feedback mechanism and because diuretic-induced prostaglandin secretion dilates the afferent arteriole.
Distal Convoluted Tubule Diuretics (DCT Diuretics)

The distal tubule (the nephron segment just beyond the

loop of Henle) reabsorbs 5 to 10% of the filtered NaCl.
As in other nephron segments, the intracellular concentration of Na + is maintained low by Na,K-ATPase. Sodium
and Cl" enter the cell across the luminal membrane via
a distinct Na-Cl cotransport pathway (see Fig. 1). DCT
diuretics are anions that, like the loop diuretics, circulate

David H. Ellison

117

in the bloodstream bound to protein and are secreted into

the lumen of the proximal tubule by the organic anion
transport pathway. They are carried downstream to the
distal tubule, where they bind to the Cl~ site on the Na-Cl
transport protein and inhibit its action. Because the distal
tubule is relatively water impermeable, NaCl reabsorption
along the DCT contributes to urinary dilution. DCT diuretics therefore impair urinary diluting capacity, but they have
no effect on urinary concentrating ability. Most DCT diuretics, with the possible exception of metolazone, become
less effective when the GFR declines below 40 mL/min.
DCT diuretics increase magnesium excretion but, in contrast to loop diuretics, they inhibit urinary calcium excretion. Two mechanisms have been invoked to explain the
effects of DCT diuretics on calcium excretion. First, DCT
diuretics stimulate calcium reabsorption along the proximal
tubule because they contract ECF volume. Second, DCT
diuretics stimulate calcium reabsorption along the distal
tubule through their action on the Na-Cl cotransporter.
When this pathway is blocked intracellular concentrations
of Na + and Cr decline. Low intracellular Cl~ concentrations hyperpolarize the cell membrane voltage which opens
dihydropyridine-sensitive calcium channels in the luminal
membrane and stimulates Na+ -Ca2+ exchange at the basolateral cell membrane. Diuretic-induced reductions in intracellular Na + concentrations may also stimulate Na + Ca2+ exchange at the basolateral cell membrane. Both
processes increase calcium reabsorption. The effects of
DCT diuretics on calcium excretion form the basis for the
use of these drugs to reduce the incidence of calcium nephrolithiasis.
Collecting Duct Diuretics

Sodium reabsorption by the collecting duct system, which

amounts to only 3% of the filtered NaCl load, is primarily
electrogenic (current generating), unlike transport along
more proximal segments. Current moves because Na + enters cells across the luminal membrane through ion channels. As in the other segments, the concentration of Na+
inside collecting duct (principal) cells is maintained below
electrochemical equilibrium by the action of the Na,KATPase. As Na + moves out of the lumen, it generates a
voltage across the tubule wall that is oriented with the
lumen negative, relative to the blood. This lumen-negative
voltage helps to drive K+ movement in the secretory direction. Although Na + and K+ do not traverse the same channel, their transport is coupled functionally by the transepithelial voltage.
Two major groups of diuretics act predominantly in the
collecting duct. Sodium channel blockers, such as triamterene and amiloride, act from the lumen to inhibit Na+ movement through Na+ channels in principal cells. Because these
drugs impair Na + movement, the transepithelial voltage
falls, inhibiting K+ secretion secondarily. This effect accounts for their K+ sparing action. It should be emphasized
that, although amiloride inhibits renal Na + -H+ exchange
in the proximal tubule, the proximal effect probably does
not contribute to its diuretic action in humans because the

concentrations of amiloride achieved in the lumen of the

proximal tubule during oral administration are insufficient
to interfere with Na + -H+ exchange. The second class of
collecting duct diuretics (CD diuretics) is represented by
spironolactone, a competitive antagonist of aldosterone.
Aldosterone, a mineralocorticoid hormone secreted by the
adrenal gland in response to renin or high serum potassium
concentrations, stimulates Na+ reabsorption and K+ secretion along the collecting duct, and increases the magnitude
of the transepithelial voltage. By inhibiting the action of
aldosterone, spironolactone causes mild natriuresis and
potassium retention.
Because the collecting duct reabsorbs only a small percentage of the filtered Na + load, collecting duct diuretics
are relatively modest in potency. In the past, their use as
sole agents has been limited to situations in which excessive
aldosterone secretion plays a central pathogenic role; in
patients with cirrhotic ascites, for example, spironolactone
has been reported to be more effective than loop diuretics,
as a single agent. Further, when hypertension is caused by
adrenocortical hyperplasia, adequate blood pressure control can often be obtained with oral amiloride. The most
common use of the collecting duct diuretics is to prevent
excessive potassium wasting when combined with other,
more potent, diuretics. Recently, interest in the utility of
CD diuretics has been stimulated by suggestions that non
renal actions of aldosterone may contribute to the pathogenes is of congestive heart failure and by suggestions that
addition of these drugs to a traditional diuretic regimen
may reduce the number of premature ventricular contractions in patients with congestive heart failure. A large multicenter trial is under way to examine the role of spironolactone in patients with congestive heart failure.
Osmotic Diuretics

Unlike other classes of diuretics, osmotic diuretics do

not interfere directly with specific transport proteins but
rather act as osmotic particles in tubule fluid. Water reabsorption throughout the nephron is driven by the osmotic
gradients that are generated by solute transport. When an
agent such as mannitol is administered, it is filtered but
very poorly reabsorbed. Because the mannitol is retained
in the tubule lumen, the osmolality of tubule fluid remains
higher than normal, inhibiting fluid reabsorption. NaCl reabsorption is also inhibited, in this case because solute
reabsorption dilutes tubule fluid, predisposing to NaCl
backflux. Thus, these drugs tend to increase the excretion
not only of fluid but of Na + , K+ , Cl~, bicarbonate and other
solutes. The urinary osmolality during osmotic diuresis
tends to approach that of plasma, regardless of the state
of hydration. Osmotic diuretics increase renal blood flow
and wash out the medullary solute gradient, effects which
contribute to the diuretic-induced impairment in urinary
concentrating capacity.
Osmotic diuretics have been used to prevent acute renal
failure following cardiopulmonary bypass, rhabdomyolysis,
and radiocontrast exposure. Mannitol is also frequently
employed to reduce cerebral edema, first by osmotic fluid

118

15. Edema and the Clinical Use of Diuretics

removal from the brain and then by promoting diuresis.

Although data in the settings of cardiopulmonary bypass
and rhabdomyolysis are inconclusive, a recent, wellcontrolled study of patients exposed to radiocontrast agents
indicated that renal failure was more common in the group
that received mannitol than in the group that received half
normal saline alone. Thus, mannitol appears to have no
place in preventing contrast nephropathy.
ADAPTATION TO DIURETIC DRUGS

When a Loop diuretic is administered acutely, Na+ and

fluid excretion increase transiently. This natriuresis is followed by a period of positive Nad balance, termed "postdiuretic NaCl retention" (see Fig. 2). The net effect of the
diuretic on ECF volume during a 24-hour period is determined by the sum of NaCl losses during diuretic action
(excretion > intake), and NaCl retention during periods
when the drug concentration is low (intake > excretion).
Factors that influence the relation between natriuresis and
postdiuretic NaCl retention include the dietary NaCl intake, the dose of diuretic, its half-life, and the frequency
with which it is administered. When loop diuretics are administered once daily to patients ingesting a high NaCl diet
postdiuretic NaCl retention compensates entirely for NaCl
losses during the period of drug action; net Na balance

remains neutral from the first day. When NaCl intake is

restricted, Na avidity during the postdiuretic period cannot
overcome the initial NaCl losses, Na balance is negative,
and ECF volume declines. This relation between dietary
NaCl intake and the net effect of diuretics accounts for the
central role of dietary NaCl restriction in effective diuretic therapy.
Even when diuretic treatment does induce negative NaCl
and fluid balance initially, neutral NaCl balance is achieved
after several days to weeks because other adaptive mechanisms come into play, limiting the magnitude of the diuretic
response (see "The Braking Phenomenon" (Fig. 2). Several mechanisms contribute to adaptation during chronic
diuretic treatment. Contraction of the ECF volume, at least
relative to pretreatment levels, may stimulate secretion of
renin, aldosterone, and antidiuretic hormone which mediate renal NaCl and fluid retention. Contraction of the
ECF volume may increase the activity of renal nerves
which stimulates renal NaCl retention via direct effects on
renal tubules. Contraction of the ECF volume may also
reduce renal perfusion pressure and the GFR. In addition
to adaptations that depend on changes in ECF volume,
however, specific intrarenal effects of diuretics may also
contribute to adaptation. Loop diuretics inhibit solute reabsorption by the thick ascending limb of Henle's loop,
thereby increasing solute delivery to and solute reabsorp-

Dietary Na Intake
(140mmol/day)

Short Term Adaptation 'PostDiuretic NaCl Retention'

Chronic Adaptation
The Braking Phenomenon'

FIGURE 2 Effects of a loop diuretic on urinary Na+ excretion. Bars represents 6 hours. Black bars indicate periods
during which urinary Na+ excretion exceeds dietary intake. Hatched areas indicate post diuretic NaCl retention, periods
during which dietary Na* intake exceeds urinary Na+ excretion. Changes in the magnitude of natriuretic response during
several days are indicative of diuretic "braking." Dotted line indicates dietary Na+ intake per 6-hour period. Inset is
effect of diuretics on weight (and ECF volume) during several days of diuretic administration. From Ellison DH:
In Sedin DW, Giebisch G (eds) Diuretic Agents: Physiology and Pharmacology, Academic Press, San Diego, 1997.
With permission.

David H. Ellison
tion from the distal nephron. When solute delivery to the
distal tubule is increased chronically (as during long-term
diuretic therapy), distal tubule cells undergo substantial
hypertrophy. This structural change is associated with increases in the ability of these cells to transport NaCl which
participate in returning the patient to neutral NaCl balance.
When these adaptive mechanisms occur prior to the
achievement of acceptable levels of ECF volume, they may
contribute importantly to diuretic resistance, as discussed
below.
The goal of diuretic treatment of edema is not simply
to increase urinary NaCl or fluid excretion. Instead, the
goal is to reduce ECF volume to a clinically acceptable
level and to maintain that volume chronically. To achieve
this goal, urinary NaCl excretion must increase initially
(Fig. 2), but excretion rates of NaCl and fluid return to
pretreatment levels once steady state returns. Thus, during
successful diuretic treatment of edema, when the patient's
weight has stabilized, urinary NaCl excretion matches dietary intake; it is not increased above normal values.
COMPLICATIONS OF DIURETIC TREATMENT

The most common complications of diuretic treatment

result directly from the effects of these drugs on renal
fluid and electrolyte excretion and include ECF volume
depletion, hyponatremia, and hypokalemia. Although both
DCT and loop diuretics predispose to hypokalemia, disorders of K+ homeostasis occur more frequently with DCT
diuretics. Serum K+ concentration declines by approxi mately 0.5 mM, when DCT diuretics are administered without KC1 supplementation. Mild hypokalemia (<3.5 mM)
occurs in up to 48% of patients during treatment with DCT
diuretics (50 mg hydrochlorothiazide); moderate hypokalemia (<3.0 mM) occurs in up to 17%. During furosemide
treatment mild hypokalemia occurs in 5% of patients, moderate hypokalemia in less than 1%. Frank hypokalemia
during treatment of hypertension with a DCT diuretic
should alert the clinician that increased renin or aldosterone secretion may be responsible for the hypertension (as
occurs in Conn's syndrome or renovascular disease).
Several mechanisms contribute to the tendency of loop
and DCT diuretics to cause hypokalemia. First, both classes
of diuretics increase fluid flow through the distal nephron,
the site at which K' secretion determines urinary K+ excretion rates. High fluid flow rates stimulate K+ secretion
directly. Second, both loop and DCT diuretics stimulate
secretion of aldosterone which also increases K+ secretion
along the distal nephron. Finally, both DCT and loop diuretics predispose to hypomagnesemia which contributes
to the development of hypokalemia through unknown
mechanisms. Hypokalemia has several adverse consequences. These include ventricular arrhythmias, especially
during the administration of digitalis glycosides or when
hypomagnesemia is present, and glucose intolerance. Hypokalemia may make control of blood pressure more difficult.
Methods to prevent or treat hypokalemia during diuretic
therapy include: (1) using the lowest effective diuretic dose

119
(especially for hypertension; see Chapter 35), (2) supplementing dietary K+ (best administered as KC1, (3) restricting dietary Na + , (4) preventing hypomagnesemia, and
(5) using CD (K+ sparing) diuretics together with loop or
DCT diuretics. Serum concentrations of Na + and K+ should
be monitored in every patient who is treated with diuretics
and most patients should be encouraged to consume a diet
rich in K+ and low in Na+ . Many physicians treat patients
whose serum K+ concentration falls below 3.5 mM, although some have suggested that K+ concentrations between 3.0 and 3.5 mM do not require treatment. Certainly,
if a patient is at risk for complications of hypokalemia,
e.g., patients receiving digitalis glycosides or patients with
hepatic cirrhosis, K concentrations should be maintained
above 3.5 mM. Of note, adding a CD diuretic not only
corrects hypokalemia in many patients, but may also prevent hypomagnesemia; hypomagnesemia may act synergis tically with hypokalemia to predispose to ventricular arrhythmias.
Hyperkalemia is a complication of the CD diuretics.
Hyperkalemia occurs most commonly in patients with renal
failure or in patients taking ACE inhibitors. Triamterene
metabolism is impaired in patients with cirrhosis. In addition, the drug precipitates hyperkalemia in this group of patients.
Mild metabolic alkalosis occurs frequently during treatment with loop and DCT diuretics. These drugs promote
urinary losses of NaCl (leaving HCOs behind). Further,
they increase aldosterone secretion, which stimulates the
proton secretion directly. Metabolic alkalosis can exacerbate hepatic encephalopathy and can inhibit respiratory
drive. Severe metabolic alkalosis is often a manifestation
of overly aggressive therapy. Loop diuretics or combination
diuretic therapy (see below) may also lead to excessive
ECF volume depletion and vascular collapse.
Hyponatremia may develop during treatment with loop
diuretics, but this complication is more common with DCT
diuretics (thiazides and their congeners). Some patients
treated with thiazide diuretics develop severe and potentially life-threatening hyponatremia, often several days to
weeks after initiation of diuretic therapy. The tendency for
hyponatremia to recur can be assessed by measuring body
weight and serum [Na] before and 6 to 8 hours after a
single dose of DCT diuretic. If the serum [Na] falls below
136 mM following diuretic administration, the patient is at
high risk for severe hyponatremia. In these patients, DCT
diuretics appear to stimu late excessive fluid intake which
contributes to weight gain and hyponatremia. Thiazide diuretics and, less commonly, loop diuretics also predispose
to glucose intolerance, hyperlipidemia, and hyperuricemia,
when administered chronically. Although the mechanisms
by which these complications develop are not completely
clear, hypokalemia and ECF volume contraction may contribute. Serum concentrations of glucose, lipids, and uric
acid should be monitored in patients on chronic diuretic
treatment.
Some complications of diuretic treatment are drug or
group specific and reflect toxic side effects; allergic interstitial nephritis is an idiosyncratic reaction to diuretics that

120

15. Edema and the Clinical Use of Diuretics

may precipitate skin rash and acute renal failure. Ototoxicity is a toxic effect to loop diuretics that occurs most
commonly when high doses are administered rapidly (furosemide iv > 15 mg/min) to patients with renal insufficiency.
Triamterene can cause renal stones and may precipitate
acute renal failure when administered with indomethacin.
Spironolactone causes gynecomastia, especially in patients
with cirrhosis of the liver.
DIURETIC TREATMENT OF EDEMA

Edema is a manifestation of disordered Nad homeostasis. The NaCl retention often reflects a physiological response to what the kidneys perceive as inadequate arterial
blood volume, as occurs in congestive heart failure. In other
situations, NaCl retention may reflect an abnormal renal
response, resulting from damage to the kidney, as occurs
in renal failure. In either case, therapeutic maneuvers
should be aimed first at correcting the primary disorder.
Often, however, such maneuvers are not available or do
not contact the ECF volume adequately and more direct
methods of effecting NaCl removal are needed. Before
initiating treatment with diuretic drugs, it is important to
begin the patient on a diet low in NaCl. Extracellular fluid
volume varies directly with NaCl intake, both in normal
and edematous individuals. For patients with mild ECF
volume expansion, a "no added salt" diet may be appropriate (4 g Na/day); for more severe edema, a low Na diet
(2 g Na/day) should be prescribed. Even when dietary
restriction alone is unsuccessful and diuretic drugs are administered, the dietary Na intake must be restricted below
4 g/day for diuretics to be effective. A second important
consideration before initiating diuretic therapy is to improve the general management of the patient by discontinuing, when possible, drugs that predispose to NaCl retention
or interfere with diuretic efficacy. Nonsteroidal antiinflammatory drugs (NSAIDs) promote renal NaCl retention
directly and interfere with the efficacy of loop and DCT
diuretics. Many vasodilators promote edema; minoxidil
frequently causes significant ECF volume expansion; calcium channel blockers promote edema despite intrinsic
natriuretic properties, perhaps through local vasodilation.
Other antihypertensive drugs may also predispose to NaCl
retention by reducing renal perfusion.
Once the decision to initiate diuretic therapy has been
made, the initial choice of drug and dosage depends on
the underlying cause of edema and its severity. Hypertension often responds to very low doses of a DCT diuretic
(e.g., 12.5 mg/day of hydrochlorothiazide), doses which
tend to cause few side effects. Cirrhotic edema and ascites
frequently respond to spironolactone (100-300 mg daily);
Spironolactone may be more effective than furosemide in
these patients. Moderate edema associated with congestive
heart failure may respond to a DCT diuretic such as hydrochlorothiazide, in doses of 25 to 50 mg/day; some studies
suggest that a DCT diuretic may reduce ECF volume more
effectively than loop diuretics in patients with very mild
congestive heart failure. But when edema from congestive

heart failure, cirrhosis, or nephrotic syndrome is more than

mild, when renal failure is present, or in the presence of
pulmonary congestion or severe symptoms, loop diuretics
are the drugs of choice.
Loop diuretics have the highest natriuretic potency, are
active at all levels of renal function, and act rapidly even
following oral administration. The drugs have steep dose
response curve; as the dose is increased, there is little response until a critical threshold is reached, above which
diuretic effectiveness increases rapidly to a maximum.
When a loop diuretic is administered in a dose that exceeds
the threshold, most patients experience an increase in urine
output that is noticeable during the several hours after
diuretic ingestion. To be effective, each dose of loop diuretic must exceed this threshold. When initiating oral diuretic therapy, a target is set for weight loss and a low dose
of loop diuretic (20 mg furosemide or its equivalent) is
begun once daily. If urine output rises during 4 to 6 hours
after diuretic ingestion, the same dose is continued on a
daily basis (unless weight loss exceeds the target value). If
urine output does not rise, the patient may double the dose
the following day (to 40 mg once daily). If there is no
response, the dose can be doubled each day until a response
is obtained or until the maximum safe dose is achieved. In
normal individuals, 40 mg of furosemide orally produces
maximal diuresis, but in patients with edema or renal failure, larger doses are frequently necessary (even in renal
failure, there is little to be gained by increasing doses beyond 240 mg furosemide or 8 mg bumetanide, iv). Often,
the dose that elicits an increase in urine output can be
continued indefinitely, because adaptive mechanisms such
as those discussed above, bring the patient back into NaCl
balance once ECF volume has been reduced. Sometimes,
however, patients may be maintained with lower doses than
were necessary to elicit diuresis initially, once control of
the ECF volume is achieved.
DIURETIC RESISTANCE: CAUSES AND TREATMENT

Control of ECF volume expansion can be attained in

most edematous patients using the approach outlined
above. In some circumstances, however, moderate or high
doses of loop diuretics do not reduce ECF volume to the
desired level even when used appropriately, and thus the
patient is deemed resistant to diuretic therapy. The determination of what ECF volume is acceptable depends on
many factors including the severity of the underlying dis ease, patient preference, and co-morbid illness. When further reductions in ECF volume are necessary, a systematic
approach (see Fig. 3) to diuretic resistance usually leads
to a treatment regimen that is safe and is effective. One of
the most common causes of apparent res istance to diuretic
drugs is dietary indiscretion because, as discussed above,
dietary NaCl excess abrogates the effect of most diuretic
regimens; the influence of dietary NaCl intake is most pronounced for the loop diuretics. If the patient's weight is
stable, dietary compliance can be assessed by measuring
the amount of Na excreted during 24 hours. A urinary Na

121

David H. Ellison

[INADEQUATE DIURETIC RESPONSE!

I DETERMINE TARGET RESPONSE]

CORRECT IDENTIFIED
PROBLEM AND REASSESS

Diuretic Resistance
Persists

[Measur
e 24

Diuretic Resistance
Persists
DOUBLE DAILY
DOSE
(to maximal safe
dose) _________
MAINTAIN UNTIL
TARGET
RESPONSE ACHIEVED

ADD DCT OR CD DIURETIC

MAINTAIN UNTIL
TARGET
RESPONSE ACHIEVED

[INTRAVENOUS OR CONTINUOUS INFUSION]

FIGURE 3 Algorithm for diuretic resistant patients. Regimens for combination therapy are
given in the text. Continuous infusions can be administered according to the following schedules; furosemide: 20 to 80 mg load followed by 2 to 80 mg/h; bumetanide 1 mg load followed
by 0.2 mg/h; torsemide 25 rag load followed by 1 to 50 mg/h. NSAIDs, nonsteroidal antiinflammatory drugs. From Wilcox CS: In The Kidney, Brenner B (ed) 5th ed. Saunders, Philadelphia, 1996.

excretion rate greater than 100 to 120 mmol (equivalent

to 2-3 g Na per day) indicates both that the patient is
ingesting too much NaCl and that true diuretic resistance
is not present; Na excretion rates near or above L20 mmol/
day should be sufficient to effect weight loss when patients
ingest less than this quantity on a daily basis. Of course,
dietary compliance cannot always be ensured, and more

intensive regimens (see below) may provide effective diuresis for patients who continue to ingest too much NaCl.
Absorption of many diuretics is variable. The gastrointestinal absorption of furosemide varies by as much as 60%
from day to day in a single individual, averaging around
50 to 60% (this effect is true of both Lasix and other brands
of furosemide). Gastrointestinal absorption may be slowed

122
further by edema of the gut, such as occurs in some patients
with congestive heart failure. When there is concern about
bioavailability, it may be advantageous to use torsemide
or bumetanide, rather than furosemide, because bioavailability exceeds 80% for these drugs. If a more bioavailable
drug is not effective, iv therapy may be necessary until
edema is controlled; at this time, diuretic absorption may
improve again and oral therapy may once again become effective.
Once a loop or DCT diuretic drug has been absorbed
into the bloodstream, it reaches the lumen of the kidney
tubule via the organic anion secretory pathway located
in the proximal tubule. This pathway also interacts with
NSAIDs and probenecid, as well as with endogenous
anions that accumulate in renal failure. When NSAIDs
have been administered to patients with renal failure, diuretic secretion into the lumen of the proximal tubule is
inhibited and less diuretic reaches its active site for any
given serum concentration. To overcome the inhibition,
higher serum levels are needed. This is one reason why
high doses of diuretic drugs are required to elicit diuresis
in patients with renal failure. Of note, although the ratio
of equipotent doses of furosemide to bumetanide is 40:1
in patients with normal renal function, it falls to 20:1 in
patients with renal failure because nonrenal furosemide
clearance is impaired. Torsemide has both renal and nonrenal clearance. Therefore, its clearance rate is relatively
stable in the face of either renal or hepatic disease. In
general, when switching from intravenous to oral furosemide, twice the oral dose is given. When switching from
intravenous to oral torsemide, the conversion is one to one.
Diuretic resistance is common in patients with the nephrotic syndrome. Hypoalbuminemia reduces the serum
concentration of diuretics because it increases their volume
of distribution (diuretics are extensively protein bound).
Further, hypoalbuminemia may predispose to renal vasoconstriction. Because albumin is filtered by the abnormal
glomeruli, renal clearance of diuretics is actually increased,
but the diuretic in the tubule lumen may be inactive because
it is bound to filtered albumin in tubule fluid and is not free
to interact with the transport protein. In these situations,
increasing the dose, changing from oral to intravenous therapy, or infusing diuretic mixed together with albumin may
improve the therapeutic response. Experimental approaches to patients with nephrotic syndrome include administering inhibitors of diuretic -protein interactions.
Inadequate renal perfusion from any cause compromises
diuretic effectiveness. For patients in whom low cardiac
output contributes to Na retention, low dose dopamine
(2-4 fj,g kg" 1 mkr1) may increase renal plasma flow and
increase urine flow (dopamine may also increase urine flow
in patients with acute renal failure), but the effects of "renal
dose" dopamine are controversial and poorly documented.
When edema results from cirrhosis of the liver, removal
of ascitic fluid by paracentesis or peritoneovenous shunting
may improve renal function and Na + excretion. Arterial
hypoxemia causes renal vasoconstriction resulting in antinatriuresis, which reverses promp tly when the arterial oxy gen tension increases above 60 mm Hg. Renal vasoconstric-

15. Edema and the Clinical Use of Diuretics

tors, such as NSAIDs and adrenergic agonists may also

lead to diuretic resistance, in part by reducing GFR, and
should be avoided in the diuretic-resistant patient. The
effects of drugs used to reduce cardiac afterload on renal
NaCl excretion are complex. When ACE inhibitors or nitroprusside increase cardiac output effectively, they may
stimulate natriuresis and reduce edema. On the other hand,
when aggressive therapy reduces blood pressure beyond a
critical threshold (which may be surprisingly high in patients with severe vascular disease), it may lead to NaCl
retention and even renal failure. This is especially common
during concomitant administration of NSAIDs, when bilateral renal artery stenosis is present, or during very aggres sive diuretic therapy.
Not uncommonly, simple approaches to diuretic resis tance fail. Several strategies are available to achieve effective control of ECF volume in such patients. First loop
diuretics may be administered more frequently than once
a day, especially if postdiuretic Na retention is contributing
importantly to NaCl retention. Recall, however, that each
dose must be above the diuretic threshold to be effective.
Second, a diuretic of another class may be added to a
regimen that contains a loop diuretic. This strategy produces true synergism; the combination of agents is more
effective than the sum of the responses to each agent alone.
DCT diuretics are the class of drug most commonly combined with loop diuretics, although synergism occurs when
loop diuretics are combined with carbonic anhydrase inhibitors and theophylline as well. These drugs act synergistically for several reasons. First, loop diuretics increase NaCl
delivery to the distal tubule, a site at which NaCl transport
depends on the luminal NaCl concentration. Loop diuretics, therefore, stimulate NaCl reabsorption in the DCT
diuretic-sensitive segment. Adding a DCT diuretic will inhibit NaCl transport along the stimulated segment. Second,
when loop diuretics are administered chronically, cells in
the distal tubule enlarge, further increasing their ability to
transport NaCl. DCT diuretics inhibit the increased NaCl
reabsorption that accompanies hypertrophy of the distal
nephron and therefore counteract the effects of hypertrophy. Third, as discussed above, the efficacy of loop diuretics
reflects the balance between natriuresis during diuretic action and Na retention when diuretic concentrations decline.
DCT diuretics have longer half lives than loop diuretics.
The former therefore prevent or attenuate NaCl retention
during the periods between doses of loop diuretics, thereby
increasing the latter's net effect. Thus, at least three mechanisms contribute to the ability of DCT diuretics to act
synergistically with loop acting drugs.
When two diuretics are combined, the DCT diuretic is
generally administered some time before the loop diuretic j
(1 hour is reasonable) in order to ensure that NaCl transport
in the distal nephron is blocked when it is flooded with
solute. When intravenous therapy is indicated, chloro-thiazide
(500-1000 mg) may be employed. Metolazone is the DCT
diuretic most frequently combined with loop diuretics because
its half-life is relatively long (as formulated in Zaroxolyn) and
because it has been reported to be effective even when renal
failure is present. Other thiazide and

David H. Ellison

thiazide-like diuretics, however, may be equally effective.

The dramatic effectiveness of combination diuretic therapy
is accompanied by complications in a significant number
of patients. Massive fluid and electrolyte losses have led
to circulatory collapse during combination therapy and patients must be followed carefully. The lowest effective dose
of DCT diuretic should be added to the loop diuretic regimen; patients can frequently be treated with combination
therapy for only a few days and then must be placed back
on a single drug regimen. When continuous combination
therapy is needed, low doses of DCT diuretic (2.5 mg metolazone or 25 mg hydrochlorothiazide) administered only
two or three times per week may be sufficient.
For hospitalized patients who are resistant to diuretic
therapy, a different approach is to infuse loop diuretics
continuously (see Fig. 3). Continuous diuretic infusions
have several advantages over bolus diuretic administration.
First, because they avoid peaks and troughs of diuretic
concentration, continuous infusions prevent periods of positive NaCl balance (postdiuretic NaCl retention) from occurring. Second, continuous infusions are more efficient
than bolus therapy (the amount of NaCl excreted per mg
of drug administered is greater). Third, some patients who
are resistant to large doses of diuretics given by bolus
have responded to continuous infusion. Fourth, diuretic
response can be titrated; in the intensive care unit where
obligate fluid administration must be balanced by fluid
excretion, excellent control of NaCl and water excretion
can be obtained. Finally, complicatio ns associated with high
doses of loop diuretics, such as ototoxicity, appear to be less
common when large doses are administered as continuous
infusion. Total daily furosemide doses exceeding 1 g have
been tolerated well when administered over 24 hours. On e
approach is to administer a loading dose of 20 mg furosemide followed by a continuous infusion of 4 to 60 mg/h.
In patients with preserved renal function, therapy at the
lower dosage range should be sufficient. When renal failure
is present, higher doses may be used, but patients should
be monitored carefully for side effects such as ECF volume
depletion and ototoxicity.
Most patients who are deemed resistant to diuretics respond to these approaches. Rather than lack of efficacy,
side effects of diuretic therapy such as increases in serum
creatinine concentration often limit the ability to reduce
ECF volume further. Obtaining effective control of ECF'

123

volume without provoking complications requires a thorough understanding of diuretic physiology and a commitment to use diuretics rationally and carefully. When used
in this manner, diuretics remain among the most powerful
drugs in clinical medicine.
Bibliography
Andreasen F, Eriksen UH, Fuul SJ, et al.: A comparison of three
diuretic regimens in heart failure. Eur J Clinical Invest 23:234239, 1993.
Barr CS, Lane CC, Hanson J, Arnott M, Kennedy N, Struthers AD:
Effects of adding spironolactone to an angiotensin -converting
enzyme inhibitor in chronic congestive heart failure secondary
to coronary artery disease. Am J Cardiol 76:1259-1265, 1995.
Denton MD, Chertow GM, Brady HM: "Renal- dose" dopamine
for the treatment of acute renal failure: Scientific rationale,
experimental studies and clinical trials. Kidney Int 49:4-14,1996.
Dormans TP, Van Meyel JJ, Gerlag PC, Tan Y, Russel FG, Smits
P: Diuretic efficacy of high dose furosemide in severe heart
failure: Bolus injection versus continuous infusion. / Am Coll
Cardiol 28:376-382, 1996.
Ellison DH: The physiologic basis of diuretic synergism: Its role in
treating diuretic resistance. Ann Intern Med 114:886-894, 1991.
Ellison DH: Diuretic drugs and the treatment of edema: From
clinic to bench and back again. Am J Kidney Dis 23:623-643,
1994.
Fliser D, Schroter M, Neubeck M, Ritz E: Coadministration of
thiazides increases the efficacy of loop diuretics even in patients
with advanced renal failure. Kidney Int 46:482-488, 1994.
Leary WP, Reyes AJ: Renal excretory actions of diuretics in man:
Correction of various current errors and redefinition of basic
concepts. In: Reyes AJ, Leary WP (eds) Progress in Pharmacology, pp. f53-f66, 1988.
Martin SJ, Danziger LH: Continuous infusion of loop diuretics
in the critically ill: A review of the literature. Crit Care Med
22:1323-1329, 1994.
Rose BD: Diuretics. Kidney Int 39:336-352, 1991.
Rudy DW, Voelker JR, Greene PK, Esparza FA, Brater DC:
Loop diuretics for chronic renal insufficiency: A continuous
infusion is more efficacious than bolus therapy. Ann Intern Med
115:360-366, 1991.
Solomon R, Werner C, Mann D, D'Elia J, Silva P: Effects of saline,
mannitol, and furosemide on acute decreases in renal function
induced by radiocontrast agents. A' Engl J Med 33f:f4161420, 1994.
Wilcox CS: Diuretics. In: Brenner BM (ed) The Kidney, 5th ed.
Saunders, Philadelphia, 1996, pp. 2299-2330.
Wilcox CS, Mitch WE, Kelly RA, et al: Response of the kidney
to furosemide: I. Effects of salt intake and renal compensation.
J Lab Clin Med f 02:450-458, 1983.

SECTION 3
GLOMERULAR DISEASES

GLOMERULAR CLINICOPATHOLOGIC
SYNDROMES
J. CHARLES JENNETTE AND RONALD j. FALK

the general population. Recurrent gross hematuria may be

superimposed on asymptomatic microscopic hematuria, or
may occur in isolation. The patient observes urine discoloration, which often is described as tea or Cola colored.
All hematuria is not of glomerular origin. In fact, most
hematuria is not of glomerular origin. Glomerular diseases
cause less than 10% of hematuria in patients with no proteinuria, with almost 80% being caused by bladder, prostate, or urethral disease. Hypercalciuria and hyperuricosuria also can cause asymptomatic hematuria, especially
in children.
Microscopic examination of the urine can help determine
whether hematuria is of glomerular or nonglomerular origin. Chemical (e.g., osmotic) trauma to red blood cells as
they pass through the nephron causes structural changes
that are not present in red blood cells that have hemorrhaged directly into the urine from a gross parenchymal
injury in the kidney (e.g., a neoplasm) or from a lesion in
the urinary tract (e.g., renal pelvis traumatized by stones
or an inflamed bladder). Dysmorphic red blood cells that
have transited the urinary tract from the glomeruli usually
have lost their biconcave configuration and hemoglobin,
and often have multiple membrane blebs, sometimes producing acanthocytes and "Mickey Mouse cells." The presence of red blood cell casts and other sediment abnormalities typical of glomerulonephritis also supports a
glomerular origin for hematuria.
Published series of renal biopsies carried out in patients
with asymptomatic hematuria show differences in the frequencies of identified underlying glomerular lesions. Differences in the nature of the population analyzed (e.g.,
military recruits vs routine physical examination patients),
and differences in pathologic analysis (e.g., failure of the
earlier studies to recognize thin basement membrane nephropathy) account for the observed disparities. The data
presented in Table 3 are derived from patients with hematuria who underwent diagnostic renal biopsy. The data in the
first column equate with asymptomatic hematuria, and are
similar to other recent series. In these patients with hematuria, less than 1 g/24 hr proteinuria and serum creatinine
less than 1.5 mg/DL, the three major findings were no
pathologic abnormality (30%), thin basement membrane
nephropathy (26%), and IgA nephropathy (28%). Whereas

Injury to glomeruli results in a variety of signs and symp toms of disease, including proteinuria caused by altered
permeability of capillary walls, hematuria caused by rupture of capillary walls, azotemia caused by impaired filtration of nitrogenous wastes, oliguria or anuria caused by
reduced urine production, edema caused by salt and water
retention, and hypertension caused by fluid retention and
disturbed renal homeostasis of blood pressure. The nature
and severity of disease in a given patient is dictated by the
nature and severity of glomerular injury.
Specific glomerular diseases tend to produce particular
syndromes of renal dysfunction, although multiple glomerular diseases can produce the same syndrome (Tables 1
and 2). The diagnosis of a glomerular disease requires recognition of one of these syndromes, followed by collection
of data to determine which specific glomerular disease is
present. Alternatively if reaching a specific diagnosis is not
possible or necessary, the physician should focus efforts on
narrowing the differential diagnosis to a likely candidate
disease.
Evaluation of pathologic features identified in a renal
biopsy specimen may be required for definitive diagnosis.
The pathologic features of various glomerular diseases are
described in the corresponding chapters of this book. Figure 1 depicts some of the clinical and pathologic features
used to resolve the differential diagnosis in patients with
antibody-mediated glomerulonephritis, Figures 2-5 illustrate the distinctive ultrastructural features of some of the
major categories of glomerular disease, Figure 6 illustrates
some of the major patterns of immune deposition identified
by immunofluorescence microscopy.
ASYMPTOMATIC HEMATURIA AND RECURRENT
GROSS HEMATURIA

Hematuria is usually defined as greater than 3 red blood

cells per high power field observed by microscopic examination of a centrifuged urine sediment (see Chapter 3).
Asymptomatic hematuria is hematuria that a patient is
unaware of and that is not accompanied by clinical manifestations of nephritis or nephrotic syndrome, that is, without
azotemia, oliguria, edema, or hypertension. Asymptomatic
microscopic hematuria has a prevalence of 5 to 10% in

Primer on Kidney Diseases, Second Edition

127

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

128

J. Charles Jennette and Ronald j. Falk

TABLE I

Clinical Manifestations of Glomerular Diseases, and Representative Diseases That

Present with These Manifestations
Asymptomatic proteinuria
Focal segmental glomerulosclerosis
Mesangioprolifcrative GN
Nephrotic syndrome
Minimal change glomerulopathy
Membranous glomerulopathy
Idiopathic (Primary)
Secondary (e.g., lupus) Focal
segmental glomerulosclerosis
Mesangioproliferative GN Type I
membranoproliferative GN Type II
membranoproliferative GN Fibrillary
GN
Diabetic glomerulosclerosis
Amyloidosis Light-chain
deposition disease
Asymptomatic microscopic hematuria
Thin basement membrane nephropathy
IgA nephropathy
Mesangioproliferative GN Alport's
syndrome
Recurrent gross hematuria Thin
basement membrane nephropathy IgA
nephropathy Alport's syndrome

Acute nephritis
Acute diffuse proliferative GN
Poststreptococcal GN
Poststaphylococcal GN Focal or
diffuse proliferative GN
IgA nephropathy
Lupus nephritis
Type I membranoproliferative GN
Type II membranoproliferative GN
Fibrillary GN
Rapidly progressive nephritis
Crescentic GN Anti- GBM
GN Immune complex GN
ANCA GN
Pulmonary-renal vasculitic syndrome
Goodpasture's (Anti- GBM) syndrome
Immune complex vasculitis
Lupus ANCA
vasculitis
Microscopic polyangiitis
Wegener's granulomatosis
Churg- Strauss syndrome
Chronic renal failure
Chronic sclerosing GN

Note that the same manifestations can be caused by different diseases, and that the same
disease can manifest in different ways. GN, glomerulonephritis; GBM, glomerular basement
membrane; ANCA, antineutrophil cytoplasmic autoantibodies.

thin basement membrane nephropathy virtually always

manifests as asymptomatic hematuria or recurrent gross
hematuria, IgA nephropathy can manifest any of the syndromes listed in Table 1 (see Chapter 22).
Alport syndrome is an hereditary disease caused by a
defect in the genes that code for basement membrane type
IV collagen (see Chapter 46). In affected males, Alport
syndrome initially manifests as asymptomatic microscopic
hematuria, sometimes with superimposed episodes of gross
hematuria. The hematuria usually begins in the first decade
of life. Progressively worsening proteinuria and renal insufficiency eventually develop, although the rate of progres sion is quite variable. Affected females, who are almost
always heterozygous, often have intermittent microscopic
hematuria but may have no other manifestations of renal disease.
Renal biopsy is not usually performed to evaluate asymp tomatic hematuria. Renal biopsy diagnoses rarely affect
treatment in patients with asymptomatic hematuria, but
occas ional patients will be found to have disease that might
benefit from treatment (e.g., the one patient with early
crescentic glomerulonephritis identified in the cohort of
patients in the first column of Table 3). Renal biopsy is
also of some prognostic value. For example, thin basement
membrane nephropathy has a better prognosis and a much

greater propensity for familial occurrence than IgA nephropathy. Many patients with asymptomatic hematuria
are subjected to repeated invasive urologic evaluations until a definitive diagnosis can be made. In these patients,
additional urologic evaluation can be prevented if renal
biopsy provides a diagnosis.
In renal biopsy specimens, thin basement membrane nephropathy is diagnosed based on thinning of the glomerular
basement membrane lamina densa (Fig. 3), whereas Alport's syndrome is suspected if there is marked lamination
of the lamina densa. The presence of mesangial immune
deposits with a dominance or codominance of immunohis tologic staining for IgA is diagnostic for IgA nephropathy
(Fig. 6).
ACUTE GLOMERULONEPHRITIS AND RAPIDLY
PROGRESSIVE GLOMERULONEPHRITIS

Acute and rapidly progressive glomerulonephritis often

present with acute onset of manifestations of nephritis,
such as azotemia, oliguria, edema, hypertension, proteinuria, and hematuria with an "active" urine sediment that
often contains red blood cell casts, pigmented casts, and
cellular debris. Rapidly progressive glomerulonephritis
leads to a 50% or greater loss of renal function within

16. Glomerular C linicopathologic Syndromes

129
TABLE 2

Tendencies of Glomerular Diseases to Manifest Nephrotic and Nephritic Features

Nephroti
c
features

Nephritic
features

Minimal change glomerulopathy

Membranous glomerulopathy
Diabetic glomerulosclerosis

Amy loidosis
Focal segmental glomerulosclerosis

Fibrillary glomerulonephritis
Mesangioproliferative glomerulopathy"
Membranoproliferative glomerulonephritis''
Proliferative glomerulonephritis"
Acute diffuse proliferative glomerulonephritis''
Crescentic glomerulonephritisd
" Mesangioproliferative and proliferative glomerulonephritis (focal or diffuse) are structural manifestations of a number of glomerulonephritides, including IgA nephropathy
and lupus nephritis.
Both type I (mesangiocapillary) and type II (de nse deposit disease).
' Often a structural manifestation of acute poststreptococcal glomerulonephritis.
Can be immune complex mediated, antiglomerular basement membrane antibody
mediated, or associated with antineutrophil cytoplasmic autoantibodies.
Note t hat most diseases can manifest both nephrotic and nephritic features, but there
usually is a tendency for one to predominate. From Jennette JC, Mandal AK (eds):
Diagnosis and Management of Renal Disease and Hypertension. Carolina Academic Press,
Durham, NC, 1994. With permission.

weeks to months. If renal failure is severe, manifestations

of uremia develop, such as nausea and vomiting, hiccups,
dyspnea, lethargy, pericarditis, and encephalopathy. Severe
volume overload can cause congestive heart failure and
pulmonary edema.
The pathologic processes that most often produce the
clinical manifestations of acute or rapidly progressive nephritis are inflammatory glomerular lesions. The nature
and severity of the glomerular inflammation correlate with
the clinical features of the nephritis (Fig. 7). Note in Fig.
7 that the structural stages of glomerular inflammation can
change over time, and this is reflected by changes in the
clinical manifestations of nephritis.
The structurally least severe injury is mesangial hyperplasia alone, which usually causes asymptomatic proteinuria
or hematuria rather than overt nephritis. Proliferative glomerulonephritis, which may be focal (affecting less than
50% of glomeruli), or diffuse (affecting greater than 50%
of glomeruli), is characterized histologically not only by
the proliferation of glomerular cells (e.g., mesangial, endothelial, and epithelial cells), but also by the influx of leuko cytes, especially neutrophils and mononuclear phagocytes.
Necrosis and sclerosis also may be present.
Lupus nephritis provides a paradigm of the interrelationships among pathogenic mechanisms, pathologic consequences, and clinical manifestations of glomerular disease
(Fig. 4). The mildest expression of lupus nephritis (mesangioproliferative lupus glomerulonephritis, class II lupus nephritis) is induced by exclusively mesangial localization of
immune complexes that usually causes only mild nephritis
or asymptomatic hematuria and proteinuria. Localization

of substantial amounts of nephritogenic immune complexes

in the subendothelial zones of glomerular capillaries where
they are adjacent to the inflammatory mediator systems in
the blood induces overt glomerular inflammation (focal or
diffuse proliferative lupus glomerulonephritis, class III or
IV lupus nephritis) and usually causes severe clinical manifestations of nephritis. Qualitative and quantitative characteristics of the pathogenic immune complexes that result
in localization predominantly in subepithelial zones where
they are not in contact with the cellular inflammatory mediator systems in the blood induces membranous lupus glomerulonephritis (class V lupus nephritis). In this form the
nephrotic syndrome rather than nephritis is usually observed. As the nephritogenic immune response in a given
patient changes over time and in response to treatment,
transitions may occur between different lupus nephritis
phenotypes (see Chapter 58).
The structurally most severe form of active glomerulonephritis is crescentic glomerulonephritis, which manifests
clinically as rapidly progressive glomerulonephritis. In patients with new onset renal disease who have a nephritic
sediment and a serum creatinine >3 g/24 hr, glomerulonephritis with crescents is the most common finding in renal
biopsy specimens (Table 3). Crescents are proliferations
of cells within Bowman's capsule that include both mononuclear phagocytes and glomerular epithelial cells. Since
crescent formation is a response to glomerular rupture, it
is a marker of severe glomerular injury. Crescents do not
indicate the cause of glomerular injury, because many different pathogenic mechanisms can cause crescent formation. There is no consensus on how many glomeruli should

]. Charles Jennette and Ronald J. Falk

130

ANTIBODY MEDIATED GLOMERULONEPHRITIS

Glomerular immune
complex localization with
granular IF staining

Circulating ANCA with

paucity of glomerular IF
immunoglobulin staining

i
vasculitis
with
no
asthma
or
granulomas

I
Goodpasture's
Syndrome

Anti-GBM
GN

ANCA
GN

Microscopic
Polyangiitis

acute
mesangio- GBM
strep/staph capillary dense
infection
changes deposits

T
IgA and
no
vasculitis

Wegener's
Granulomatosis

Churg-Strauss
Syndrome

subepithelial
deposits

other
features

T
20nm
fibrils

IgA and SLE

systemic
vasculitis

Circulating anti-GBM
antibodies with linear
glomerular IF staining
IgA
Nephropathy

H-S
Purpura

Lupus
Nephritis

Acute PostInfectious GN

Type I
MPGN

Type II
MPGN

Membranous
GN

Fibrillary
GN

Many
Others

FIGURE I Features that distinguish different immunopathologic categories of antibody-mediated glomerulonephritis.

GBM, glomerular basement membrane; IF,
immunofluorescence
microscopy; ANCA,
anti-neutrophil
no
granulomas
eosinophilia,
cytoplasmic autoan- systemic tibodies;
GN, and no
with lung
without
asthma and
glomerulonephritis;
H- S, Henoch- Schonlein;
hemorrhage
lung
vasculitis
asthma
granulomas
MPGN,
membranoproliferative

glomerulonephritis.

have crescents before the term crescentic glomerulonephritis can be applied. Most pathologists use the term when
more than 50% of glomeruli have crescents, but the percenter of glomeruli with crescents should be specified in the
diagnosis, even if it is less than 50% (e.g., IgA nephropathy
with focal proliferative glomerulonephritis and 25% crescents). Within a specific pathogenic category of glomerulonephritis (e.g. anti-GBM disease, lupus nephritis, IgA nephropathy, poststreptococcal glomerulonephritis), the

higher the percentage of glomeruli with crescents the worse

the prognosis. Among pathogenetically different glomerulonephritides, however, the pathogenic category may be
more important in predicting outcome than the presence
of crescents. For example, a patient with poststreptococcal
glomerulonephritis with 50% crescents has a much better
prognosis for renal survival, even without immunosuppres sive treatment, than a patient with anti-GBM glomerulonephritis or ANCA glomerulonephritis with 25% crescents.

Normal Glomerular Capillary

Idiopathic

Membranous Glomerulopathy

Minimal Change Glomerulopathy

Secondary
Membranous Glomerulopathy

Diabetic Glomerulosclerosis

Amyloidosis

FIGURE 2 Ultrastructural changes in glomer ular capillaries of glomerular diseases that cause the
nephrotic syndrome. Normal glomerular capillary: Note the visceral epithelial cell with intact foot
processes (green), endothelial cell with fenestrations (yellow), mesangial cell (red) with adjacent mesangial matrix (light gray), and basement membrane with lamina densa (dark gray) that does not completely
surround the capillary lumen but splays out as the paramesangial basement membrane. Minimal change
glomerulopathy: note the effacement of foot processes and microvillous transformation. Diabetic glomerulosclerosis: Note the thickening of the lamina densa and expansion of mesangial matrix. Idiopathic
membranous glomerulopathy: Note the subepithelial dense deposits with adjacent projections of basement membrane (see also Fig. 5). Secondary membranous glomerulopathy: Note the mesangial and
small subendothelial deposits in addition to the requisite subepithelial deposits. Amyloidosis: Note the
fibrils within the mesangium and capillary wall. From JC Jennette, with permission.

Thin Basement Membrane

Nephropathy

Proliferative Lupus
Glomerulonephritis

Mesangioproliferative
Glomerulonephritis

Type I Membranoproliferative
Glomerulonephritis

Acute
Postinfectious
Glomerulonephritis

Type II Membranoproliferative
Glomerulonephritis

FIGURE 3 Ultrastructural changes in glomerular capillaries of glomerular diseases that

cause hematuria and the nephritic syndrome. Thin basement membrane nephropathy: Note
the thin lamina densa of the basement membrane. Mesangioproliferative glomerulonephritis
(e.g., mild lupus nephritis or IgA nephropathy): Note the mesangial dense deposits and
mesangial hypercellularity. Acute diffuse proliferative glomerulonephritis (e.g., poststreptococcal glomerulonephritis): Note the endocapillary hypercellularity contributed to by leukocytes, endothelial cells, and mesangial cells and the dense deposits, including not only
conspicuous subepithelial "humps" but also inconspicuous subendothelial and mesangial
deposits. Proliferative lupus glomerulonephritis (see also Fig. 4): Note the extensive subendothelial and mesangial dense deposits. Type I membranoproliferative glomerulonephritis
(mesangiocapillary glomerulonephritis): Note the subendothelial deposits with associated
subendothelial interposition of mesangial cytoplasm and deposition of new matrix material
resulting in basement membrane replication. Type II membranoproliferative glomerulonephritis (dense deposit disease): Note the intramembranous and mesangial dense deposits.
From JC Jennctte, with permission.

133

16. Glomerular Clinicopathologic Syndromes

FIGURE 4 Ultrastructural features of the major classes of lupus nephritis. The sequestration
of immune deposits within the mesangium in class II (mesangioproliferative) lupus glomerulonephritis causes only mesangial hyperplasia and mild renal dysfunction. Substantial amounts of
subendothelial immune deposits, which are adjacent to the inflammatory mediator systems of
the blood, cause focal (class 111) or diffuse (class IV) proliferative lupus glomerulonephritis with
overt nephritic signs and symptoms. Localization of immune deposits predominantly in the
subepithelial zone causes membranous (class V) lupus glomerulonephritis, which usually manifests predominantly as the nephrotic syndrome. From JC Jennette, with permission.

This importance of pathogenic category in predicting

the natural history of glomerulonephritis indicates that the
pathologic separation of glomerulonephritis into the light
microscopic morphologic categories in Fig. 7 is not sufficient for patient management. In addition to establishing
the type of glomerular terms present, the pathogenic or
immunopathologic category of disease must be determined
as well. If a renal biopsy is performed, this determination
is usually based on immunohistology and electron microscopy (Figs. 1-6). Immunohistology reveals the presence or
absence of immunoglobulins and complement components.

The distribution (e.g., capillary wall, mesangium), pattern

(e.g., granular, linear), and composition (e.g., IgAdominant, IgG-dominant, IgM-dominant) of immunoglobulin is useful for determining specific types of glomerulonephritis (as will be discussed in detail in later chapters in
this primer).
Table 4 gives the frequencies of the major immunopathologic categories of glomerulonephritis in patients with cres cents who have undergone renal biopsy. The immune complex category contains a variety of diseases, including lupus
nephritis, IgA nephropathy, and poststreptococcal glomer-

134

J. Charles Jennette and Ronald J. Falk

FIGURE 5 Ultrastructural stages in the progression of membranous glomerulopathy. Stage

I has subepithelial electron-dense immune complex deposits without adjacent projections of
basement membrane material. Stage II has adjacent GBM projections that eventually surround
the electron-dense immune deposits in stage III. Stage IV has a markedly thickened GBM with
electron lucent zones replacing the electron-dense deposits. From JC Jennette, with permission.

ulonephritis. Note that the majority of patients with >50%

crescents have little or no immunohistologic evidence for
immune complex or anti-GBM antibody localization within
glomeruli (i.e., the term is pauciimmune). More than 80%
of these patients with pauciimmune crescentic glomerulonephritis have circulating antineutrophil cytoplasmic autoantibodies (ANCA).
Because both the structural severity (e.g., morphologic
stages in Fig. 7) and immunopathologic category of disease
(e.g., the categories given in Figs. 1 -6, such as IgA nephropathy, lupus nephritis, anti-GBM disease) are important in
predicting the course of disease in a patient with glomerulonephritis, the most useful diagnostic term should include

information about both (e.g., focal proliferative IgA nephropathy, diffuse proliferative lupus glomerulonephritis,
crescentic anti-GBM glomerulonephritis).
Because they often are immune-mediated inflammatory
diseases, many types of glomerulonephritis are treated with
corticosteroids, cytotoxic drugs or other antiinflammatory
and immunosuppressive agents. The aggressiveness of the
treatment, of course, should match the aggressiveness of
the disease. For example, active class IV lupus nephritis
warrants immunosuppressive treatment, whereas class II
lupus nephritis dose not.
The two most aggressive forms of glomerulonephritis
are anti-GBM crescentic glomerulonephritis and ANCA

Membranous Glomerulopathy (IgG)

Type I MPGN (C3)

Postinfectious GN (C3)

Class IV Lupus GN (IgG)

Amyloidosis (Lambda LC)

Type II MPGN (C3)

IgA Nephropathy (IgA)

Anti-GBM GN (IgG)

FIGURE 6 Immunofluorescence microscopy staining patterns for membranous glomerulopathy: Note the global granular capillary wall staining for IgG. AL amyloidosis: Note the
irregular fluffy staining for light chains. Type I membranoproliferative glomerulonephritis
(MPGN): Note the peripheral granular to band-like staining for C3. Type II MPGN:
Note the band-like capillary wall and coarsely granular mesangial staining for C3. Acute
postinfectious glomerulonephritis: Note the coarsely granular capillary wall staining for
C3. IgA nephropathy: Note the mesangial staining for IgA. Class IV lupus nephritis:
Note the segmentally variable capillary wal l and mesangial staining for IgG. Anti-GBM
glomerulonephritis: Note the linear GBM staining for IgG.

J. Charles Jennette and Ronald J. Falk

136
TABLE 3

Renal Disease in Patients with Hematuria Undergoing Renal Biopsy

No abnormality Thin
BM nephropathy IgA
nephropathy GN
without crescents"
GN with crescents"
Other renal disease''
Total

Prot <1
Cr <1.5
30%
26%
28%
9%
2%
5%
100%
n = 43

Prot 1-3
2% 4%
24%
26%

Cr 1.5-3.0
1% 3%
14%
37%

24%
20%
100%
n = 123

21%
24%
100%
n = 179

Cr >3
0%
0% 8%
23%
44%
25%
100%
n = 255

" Proliferative or necrotizing GN other than IgA nephropathy or lupus nephritis.

Includes causes for the nephrotic syndrome, such as membranous glomerulopathy and focal
segmental glomerulosclerosis.
Note: This is an analysis of renal biopsy specimens evaluated at by the University of North Carolina
Nephropathology Laboratory. Patients with systemic lupus erythematosus have been excluded from
the analysis. GN, glomerulonephritis; BM, basement membrane; P, proteinur ia (g/24hr); Cr, serum
creatinine (mg/dL). Derived from Caldas MLR, Jennette JC, Falk RJ, Wilkman AS: NC Glomerular
Disease Collaborative Network: What is found by renal biopsy in patients with hematuria? Lab Invest
62:15A, 1990.

crescentic glomerulonephritis. The most important factor

in renal outcome is early diagnosis and treatment. Both
diseases are treated with immunosuppressive regimens
(e.g., pulse methylprednisolone and intravenous or oral
cyclophosphamide). Plasmapheresis is usually added to the

regimen for anti-GBM disease. Immunosuppressive treatment generally can be terminated after 4 to 5 months in
patients with anti-GBM glomerulonephritis with very little
risk of recurrence (see Chapter 23). The initial induction
of remission for ANCA-glomerulonephritis often is carried

LIGHT MICROSCOPIC MORPHOLOGY

CHRONIC
GLOMERULONEPHRITIS

MESANGIOPROUFERATIVE
GLOMERULOPATHY
END STAGE KIDNEY

ASYMPTOMATIC
HEMATURIA &/or
PROTEINURIA

ACUTE
NEPHRITIS

RAPIDLY PROGRESSIVE
NEPHRITIS

CHRONIC
NEPHRITIS

CLINICAL MANIFESTATIONS

FIGURE 7 Morphologic stages of glomerulonephritis (top of diagram) aligned with the usual
clinical manifestations (bottom of diagram). Certain glomerular diseases, such as anti-GBM and
ANCA glomerulonephritis, usually have crescentic glomerulonephritis with rapid progression of
renal failure if nol promptly treated. Others, such as lupus nephritis, have a predilection for
causing focal or diffuse proliferative glomerulonephritis with variable rates of progression
dependent on the activity of the glomerular lesions. Some, such as IgA nephropathy, tend to
begin as mild mesangioproliferative lesions but may progress into more severe proliferative
lesions. Poststreptococcal glomerulonephritis and similar diseases typically initially develop a
very active acute proliferative glomerulonephritis but then resolve through a mesangioproliferative phase to normal: Still others, such as IgM mesangial nephropathy, rarely progress past
the mesangioproliferative phase. Reprinted with permission from Jennette JC, Mandal AK:
Syndrome of glomerulonephritis. In: Mandal AK, Jennette JC (eds), Diagnosis and Management
of Renal Disease and Hypertension (2nd Ed.). Carolina Academic Press, Durham, NC, 1994.

137

16. Glomerular Clinicopathologic Syndromes

TABLE 4

Frequency of Immunopathologic Categories of Crescentic Glomerulonephritis in More Than

3000 Consecutive Nontransplant Renal Biopsies Evaluated by Immunofluorescence Microscopy
in the University of North Carolina Nephropathology Laboratory
Any crescents
(n = 540)

>50% crescents
( = 195)

Arteritis in biopsy
(n = 37)

51% (277/540)

61% (118/195)"

84% (31/37)

44% (238/540)

29% (56/195)

14% (5/37)c

5% (25/540)*

11% (21/195)

3% (imy

Immunohistology
Pauciimmune
(<2 + Ig)
Immune complex

(^2 + Ig)
Anti-GBM

" 70 of 77 patients tested for ANCA were positive (91%). (44 P -ANCA and 26 C- ANCA)
b
c

3 of 19 patients tested for ANCA were positive (16%). (2 P-ANCA and 1 C-ANCA)
Four patients had lupus and one had poststreptococcal glomerulonephritis.

This patient also had a P -ANCA (MPO-ANCA).

From Jennette JC, Falk RJ: The pathology of vasculitis involving the kidney. Am J Kidney Dis 24:1994.

out for 6 to 12 months, and even then there is an approximately 25% risk for recurrence that will require additional
immunosuppression.
GLOMERULONEPHRITIS ASSOCIATED WITH
SYSTEMIC DISEASES

Some patients with acute or rapidly progressive glomerulonephritis have a pathogenetically related systemic dis ease. Immune complex-mediated glomerulonephritides
that are induced by infections may have an antecedent or
concurrent infection, such as streptococcal pharyngitis or
pyoderma preceding acute poststreptococcal glomerulonephritis or hepatitis C infection concurrent with type I membranoproliferative glomerulonephritis. As noted earlier,
glomerulonephritis with any of morphologic expressions
shown in Fig. 7, as well as membranous glomerulopathy,
can be caused by systemic lupus erythematosus (Fig. 4).
Because glomeruli are vessels, glomerulonephritis is a
frequent manifestation of systemic small vessel vasculitides,
such as Henoch-Schonlein purpura, cryoglobulinemic vas culitis, microscopic polyangiitis (microscopic polyarteritis),
Wegener's granulomatosis, or Churg-Strauss syndrome.
Henoch-Schonlein purpura is caused by vascular localization of IgA-dominant immune complexes, which manifests
as IgA nephropathy in the glomeruli. Cryoglobulinemic
vasculitis is caused by cryoglobulin deposition in vessels,
and often is associated with hepatitis C infection. In glomeruli, cryoglobulinemia usually causes type I membranopro liferative glomerulonephritis. Microscopic polyangiitis,
Wegener's granulomatosis, or Churg-Strauss syndrome
have a paucity of immune deposits in vessel walls and are
associated with circulating ANCA. ANCA glomerulonephritis is characterized pathologically by fibrinoid necrosis
and crescent formation, and often manifests as rapidly pro gressive renal failure. Patients with vasculitis -associated
glomerulonephritis typically have clinical manifestations
of vascular inflammation in multiple organs, such as skin

purpura caused by dermal angiitis, hemoptysis caused by

alveolar capillary hemorrhage, abdominal pain caused by
gut infarcts, and mononeuritis multiplex caused by vasculitis in the epineural and perineural vessels of peripheral
nerves.
A distinctive and severe clinical presentation for glomerulonephritis is pulmonary -renal vasculitic syndrome,
which usually has rapidly progressive glomerulonephritis
combined with pulmonary hemorrhage. Table 1 lists the
most common causes for pulmonary-renal vasculitic syndrome. Histologic and immunohistologic examination of
involved vessels, including glomeruli in renal biopsy specimens, is useful in making a definitive diagnosis (Fig. 1)
Serologic analysis for anti-GBM antibodies, ANCA, and
markers for immune complex disease (e.g., antinuclear antibodies, cryoglobulins, antihepatitis C and B antibodies,
complement levels) also may indicate the appropriate diagnosis (Fig. 1).
ASYMPTOMATIC PROTEINURIA AND
NEPHROTIC SYNDROME

When proteinuria is severe, it causes the nephrotic

syndrome. Less severe proteinuria, or severe protein uria of short duration, may be asymptomatic. The nephrotic syndrome is characterized by massive proteinuria
(>3 g 24hr1 1.73 nr 2 ), hypoproteinemia (espe cially hypoalbuminemia), edema, hyperlipidemia, and lipiduria. The most specific microscopic urinalysis finding is
the presence of oval fat bodies. These are sloughed tubular
epithelial cells that have resorbed some of the excess lipids
and lipoproteins in the urine (see Chapter 3, Fig. 3).
Severe nephrotic syndrome predisposes to thrombosis
secondary to loss of hemostasis control proteins (e.g., antithrombin III, protein S, and protein C), infection secondary
to loss of immunoglobulins and, possibly, accelerated atherosclerosis because of the hyperlipidemia. Dehydration
and inactivity may increase the risk for venous thrombosis

138
in nephrotic patients. In nephrotic patients with frequent
bacterial infections, administration of intravenous gamma
globulin may be required.
Any type of glomerular disease can cause proteinuria. In
fact, proteinuria is a very sensitive indicator of glomerular
damage. All proteinuria, however, is not of glomerular
origin. For example, tubular damage can cause proteinuria,
but rarely more than 2 g/24 hr.
As noted in Table 2, some glomerular diseases are more
likely to manifest the nephrotic syndrome than others, although virtually any form may cause the nephrotic syndrome. The two primary renal diseases that most often
manifest as nephrotic syndrome are minimal change glomerulopathy and membranous glomerulopathy, and the
two secondary forms of renal disease that most often manifest as nephrotic syndrome are diabetic glomerulosclerosis
and amyloidosis (specifically, AL and AA amyloidosis).
Age is a major influence on the frequency of causes for
the nephrotic syndrome. In children under 10 years old,
about 80% of the nephrotic syndrome is caused by minimal
change glomerulopathy. Throughout adulthood, minimal
change glomerulopathy accounts for only 10 to 15% of
primary nephrotic syndrome.
Membranous glomerulopathy is the most common cause
of primary nephrotic syndrome in adults, but it accounts
for less than 50% of cases. As shown in Fig. 8, a variety of
glomerular diseases account for the remaining cases of
nephrotic syndrome that are identified at the time of renal
biopsy. The data in Fig. 8 are derived from patients with
nephrotic range proteinuria who have undergone renal biopsy. The frequency of causes for the nephrotic syndrome
that are not always examined by renal biopsy, especially
diabetic glomerulosclerosis, are not accurately represented
in Fig. 8.
Membranous glomerulopathy is most frequent in the
fifth and sixth decades of life. It is characterized pathologically by numerous subepithelial immune complex deposits
(Figs. 2,5, and 6). The glomerular lesion evolves over time,
with progressive accumulation of basement membrane material around the capillary wall immune complexes (Fig.
5) and eventual development of chronic tubulointerstitial
injury in those patients with progressive disease (see Chapter 21). If the Heymann nephritis animal model is analogous
to human disease, idiopathic (primary) membranous glomerulopathy may be caused by autoantibodies specific for
antigens on visceral epithelial cells, which would allow immune complex formation in the subepithelial zone but not
in the subendothelial zone or mesangium of glomeruli. In
addition to the numerous subepithelial immune deposits,
membranous glomerulopathy secondary to immune complexes composed of antigens and antibodies in the systemic
circulation often has immune complex deposits in the mesangium, and may have small subendothelial deposits (Fig.
2). Thus, the ultrastructural identification of mesangial or
subendothelial deposits should raise the level of suspicion
for secondary membranous glomerulopathy, such as membranous glomerulopathy caused by a systemic autoimmune
disease (e.g., lupus, mixed connective tissue disease, autoimmune thyroiditis), infection (e.g., hepatitis B or C, syphi-

, Charles jennette and Ronald J. Falk

Other
Amyloidoisis
Diabetic gs
Membranoproliferative gn
Proliferative gn

Minimal change gp
Focal segmental gs

Membranous gp
1
6
2
5
55

35

45

65 Age

FIGURE 8 Diagram demonstrating the approximate frequency

of different renal diseases in patients with the nephrotic range
proteinuria who had renal biopsies that were evaluated in the
University of North Carolina Nephropathology Laboratory. Note
the variation in frequency with age. The proliferative glomerulonephritis category includes all forms of proliferative glomerulonephritis, including lupus nephritis, IgA nephropathy, IgM mesangial
nephropathy and others, gn, glomerulonephritis; gp, glomerulopathy; gs, glomerulosclerosis. Reprinted with permission from Jennette JC, Mandal AK: The nephrotic syndrome. In: Mandal AK,
Jennette JC (eds), Diagnosis and Management of Renal Disease
and Hypertension (2nd ed.). Carolina Academic Press, Durham,
NC, 1994.

lis), or neoplasm (e.g., lung or gut carcinoma). In very

young and very old patients, the likelihood of secondary
membranous glomerulopathy is greater, although still uncommon. Membranous glomerulopathy occurring in young
patients raises the possibility of systemic lupus erythematosus or hepatitis B infection, and in very old patients raises
the possibility of occult carcinoma.
Both type I and II membranoproliferative glomerulonephritis (MPGN) typically manifest mixed nephrotic and
nephritic features, sometimes accompanied by hypocomplementemia and C3 nephritic factor, which is an autoantibody against the C3 convertase of the alternative complement activation pathway (see Chapter 19). Both types often
have glomerular capillary wall trn'ckening and hypercellularity by light microscopy. Type I MPGN (mesangiocapillary glomerulonephritis) is characterized ultrastructurally
by subendothelial immune complex deposits that stimulate
subendothelial mesangial interposition and replication of
basement membrane material, whereas type II MPGN
(dense deposit disease) has pathognomonic intramembranous dense deposits (Fig. 3). Both types have extensive
glomerular staining for C3 (Fig. 6), with type I having more
immunoglobulin staining than type II. Type I MPGN may

16. Glomerular Clinicopathologic Syndromes

be secondary to cryoglobulinemia, neoplasms, or chronic

infections (e.g., hepatitis C and B, and infected prostheses,
such as a ventriculoatrial shunt).
When taken as a group, focal and diffuse proliferative
glomerulonephritides account for a substantial proportion
of patients who have nephrotic range proteinuria. Patients
with proliferative glomerulonephritis and marked proteinuria usually also have features of nephritis, especially hematuria. Included in this group would be patients with lupus
nephritis and IgA nephropathy who have nephrotic
range proteinuria.
Amyloidosis as a cause for the nephrotic syndrome is
most frequent in older adults (Fig. 8). Currently in the
United States, amyloid causing the nephrotic syndrome is
approximately 75% AL amyloid rather than AA, and AL
amyloid is approximately 75% lambda rather than kappa
light-chain composition. Patients with kappa light-chain
paraproteins and the nephrotic syndrome are more likely
to have light-chain deposition disease (i.e., nodular sclerosis
without amyloid fibrils) rather than amyloidosis (see Chapter 30). Amyloid composition can be determined by immu nofluorescence microscopy (Fig. 6).
CHRONIC GLOMERULONEPHRITIS AND END STAGE
RENAL DISEASE

Most glomerular disease, with possible exceptions being

uncomplicated minimal change glomerulopathy and thin
basement membrane nephropathy, can progress to chronic
glomerular sclerosis with progressive renal failure and
eventually to end stage renal disease (ESRD). Chronic
glomerular disease is the third leading cause of ESRD in the
United States, following hypertensive and diabetic renal
disease in frequency. Clinicopathologic studies of different
glomerular diseases have revealed marked differences in
their natural histories. Diseases such as anti-GBM and
ANCA crescentic glomerulonephritis have high risk for
rapid progression to ESRD unless treated. Other diseases
such as IgA nephropathy and focal segmental glomerulosclerosis, have more indolent but persistent courses. A few
glomerulonephritides, for example poststreptococcal glomerulonephritis, may initially manifest a rather severe nephritis, but usually resolve completely with little risk for
progression to ESRD. And some diseases are very unpredictable. For example, membranous glomerulopathy may
remit spontaneously, have persistent nephrosis for decades
without renal failure, or progress over several years to
ESRD.
Chronic glomerulonephritis is characterized pathologically by varying degrees of glomerular scarring, which is
always accompanied by cortical tubular atrophy, interstitial
fibrosis, interstitial infiltration by chronic inflammatory
cells, and arteriosclerosis. As the glomerular, interstitial,
and vascular sclerosis worsen, they eventually reach a point
at which histologic evaluation of the renal tissue cannot
reveal the initial cause for the renal injury, and a pathologic
diagnosis of ESRD is all that can be concluded.
Clinically, chronic glomerulonephritis that is progressing
to ESRD eventually results in uremia that must be man-

139

aged by dialysis or renal transplantation. As the term implies, patients with uremia have accumulation of nitrogenous wastes (urea, uric acid, creatinine) in the blood. Other
clinical manifestations of uremia include nausea and vomiting, hiccups, anorexia, pruritis, lethargy, pericarditis, myopathies, neuropathies and encephalopathy (see Chapter 62).
RENAL BIOPSY: INDICATIONS AND METHODS

In a patient with renal disease, a renal biopsy provides

tissue that can be used to determine the diagnosis, indicate
the cause, predict the prognosis, direct treatment, and collect data for research, although not all potential applications are accomplished by every renal biopsy.
Renal biopsy is indicated in a patient with renal disease
when all three of the following conditions are met: (1) the
cause cannot be determined or adequately predicted by less
invasive diagnostic procedures; (2) the signs and symptoms
suggest parenchymal disease that can be diagnosed by
pathologic evaluation; and (3) the differential diagnosis
includes diseases that have different treatments, different
prognoses, or both. Situations in which a renal biopsy serves
an important diagnostic function include nephrotic syndrome in adults, steroid-resistant nephrotic syndrome in
children, glomerulonephritis in adults other than clear-cut
acute poststreptococcal glomerulonephritis or lupus nephritis, and acute renal failure of unknown cause. In some
renal diseases for which the diagnosis is relatively certain
from clinical data, a renal biopsy may be of value not
only for confirming the diagnosis but also for assessing the
activity, chronicity, and severity of injury; for example,
in patients with suspected lupus nephritis. Although the
diagnosis is strongly supported by positive serologic results
in patients with anti-GBM and ANCA-glomerulonephritis,
the extremely toxic treatment that is used for these diseases
warrants the additional level of confirmation that a renal
biopsy provides. A renal biopsy also provides information
about the severity and potential reversibility of the glomerular damage. Table 5 demonstrates the types of native
renal disease that have prompted renal biopsy among the
nephrologists who refer specimens to the University of
North Carolina Nephropathology Laboratory. Approxi mately 80% of these biopsies were performed by nephrologists in community practice. Diseases that typically cause
nephrotic syndrome were the most frequent impetus for
biopsy, followed by diseases that cause nephritis.
Contraindications to percutaneous renal biopsy include
an uncooperative patient, solitary kidney, hemorrhagic diathesis, uncontrolled severe hypertension, severe anemia
or dehydration, cystic kidney, hydronephrosis, multiple renal arterial aneurysms, acute pyelonephritis or perinephric
abscess, renal neoplasm, and ESRD.
Clinically significant complications of renal biopsy are
relatively infrequent but must be kept in mind when determining the risk-benefit ratio of the procedure. Small perirenal hematomas that can be seen by imaging studies (e.g.,
ultrasound) are relatively common if looked for carefully.
Gross hematuria occurs in <10% of patients, arteriovenous

140

J. Charles jennette and Ronald j. Falk

TABLE 5

Frequency of Various Diagnoses Among 7257 Renal Biopsies Evaluated in the University of North Carolina
Nephropathology Laboratory
Diseases that often cause nephrotic syndrome (42%)

3067

Diseases that often cause hematuria and nephritis (29%)

2109

Idiopathic membranous glomerulopathy

847

Focal segmental glomerulosclerosis (FSGS)

Minimal change glomerulopathy
Diabetic glomerulosclerosis

768
398
246

Type I membranoproliferative GN

190

Acute diffuse proliferative (postinfectious) GN

86

Mesangioproliferative GN
Amyloidosis
Clq nephropathy

145
108
99

Thin basement membrane nephropathy

82

Lupus nephritis (all classes)

IgA nephropathy

636
538

Other immune complex proliferative GN

375

Pauciimmune/ANCA GN

301

Anti-GBM GN

56

Alport's syndrome

35

Collapsing variant of FSGS

87

Glomerular tip lesion variant of FSGS

Fibrillary GN

65
59

Light-chain deposition disease

26

Type II membranoproliferative GN
Preeclampsia/eclampsia
Immunotactoid glomerulopathy
Collagenofibrotic glomerulopathy

14
6
6
3

Miscellaneous other diseases (3%)

No pathologic lesion identified* (2%)

141

371

Adequate tissue with nonspecific abnormalities* (5%)

370

126
101
69

Inadequate tissue for definitive diagnosis* (6%)

417

Diseases that often cause acute renal failure" (5%)

Thrombotic microangiopathy (all types)
Acute tubulointerstitial nephritis
Acute tubular necrosis

Atheroembolization

34

Light-chain cast nephropathy

31

Coritcal necrosis

10

Diseases that often cause chronic renal failure (8%)

583

Arterionephrosclerosis
Chronic sclerosing GN

229
166

End stage renal disease

114

Chronic tubulointerstitial nephritis

74

199

" Other than glomerulonephritis.

Specimens with nonspecific abnormalities (e.g., interstitial fibrosis, tubular atrophy, glomerular scarring, arteriosclerosis), specimens
with no identifiable pathologic abnormality (e.g., in a patient with asymptomatic hematuria), and some specimens with inadequate tissue
for definitive diagnosis (e.g., a very small specimen with only a few glomeruli but with negative immunofluorescence microscopy) may
provide useful clinical information, especially with respect to ruling out diseases that were in the differential diagnosis.

GN, glomerulonephritis.

fistula in <1%, hemorrhage that requires surgery in <1%,

and mortality in <0.1%.
Current percutaneous needle biopsy procedures usually
employ localization of the kidney by real-time ultrasound
guidance, determination of kidney location and depth by
ultrasound immediately prior to biopsy, or CT guided localization of the kidney. Many varieties of biopsy needles
have been used over the yearsmost of which are effective
in experienced hands. Recently, there has been a major
shift toward utilization of spring-loaded disposable gun devices.
Light microscopy alone is not adequate for the diagnosis
of most native kidney diseases, although it may be adequate
for assessing the basis for renal allograft dysfunction during
the first few weeks after transplantation. All native kidney
biopsies should be processed for at least light microscopy
and immunofluorescence microscopy. Most renal pathologists advocate performing electron microscopy on all native
kidney biopsies, but some believe that tissue for electron
microscopy should be fixed as a contingency and electron
microscopy performed only if the other microscopic findings suggest that it will be useful.
The needle biopsy core should be examined with a magnifying glass (e.g., 15X) or a dissecting microscope to con-

firm that renal tissue is present and to determine whether

it is cortex or medulla. When gently prodded and pulled
with forceps, adipose tissue is mushy and strings out, skeletal muscle tissue falls apart into little clumps, and renal
tissue maintains a cylindrical shape. At 15X magnification,
adipose tissue looks like clusters of tiny fat droplets (cells),
skeletal muscle is red-brown with irregular fiber bundles,
and renal tissue is pale pink-tan. Glomeruli in the renal
cortex appear as reddish blushes or hemispheres projecting
from the surface of the core. Straight red striations produced by the vasa recta are markers for the medulla. When
there is extensive glomerular hematuria, the convoluted
tubules in the cortex appear as red corkscrews. Once the
tissue landmarks are identified, portions of tissue should
be separated for processing for light, immunofluorescence,
and electron microscopy.
In our experience with renal biopsy specimens sent to
us from more than 100 different nephrologists per year,
most of whom are in private practice, approximately 6%,
of renal biopsy specimens are inadequate for a definitive
diagnosis (Table 5). The most common inadequacy is renal
tissue with too little or no cortex. This can be remedied
by beginning the sampling procedure with the biopsy needle just barely into the outer cortex. Obviously, if the biopsy

1 7 . I m m u n o p a t h o g e n e s i s o f Renal Disease

141

needle is inserted too deeply into or through the cortex,

the sampling procedure will obtain only medulla. Even
specimens that are considered inadequate for a definitive
diagnosis may provide useful information. For example,
in a patient with the nephrotic syndrome, a renal biopsy
specimen that has no glomeruli for light or electron microscopy, but has one glomerulus that stains negative for immu noglobulins and complement by immunofluorescence microscopy, rules out any form of immune complex
glomerulonephritis, such as membranous glomerulopathy,
and focuses the differential diagnosis on minimal change
glomerulopathy vs. focal segmental glomerulosclerosis.
Bibliography
Appel GB: Renal biopsy: The clinician's viewpoint. In: Silva FG,
D'Agati VD, Nadasdy T (eds) Renal Biopsy Interpretation.

Churchill Livingstone, New York, pp. 21-29, 1996. Bolton

WK: Goodpasture's syndrome. Kidney Int 50:1753-1766,
1996. Cameron JS: The nephrotic syndrome and its
complications. Am
J Kidney Dis 10:157-171, 1987.

Cohen AH, Nast CC, Adler SG, Kopple JD: Clinical utility of
kidney biopsies in the diagnosis and management of renal disease. Am J Nephrol 9:309-315, 1989.
Couser WG: Rapidly progressive glomerulonephritis: Classification, pathogenetic mechanisms, and therapy. Am J Kidney Dis
11:449-464, 1988.
Dische F, Parsons V, Taube D: Thin-basement -membrane nephropathy. N Engl J Med 320:1752-1753, 1989.
Galla JH: IgA nephropathy. Kidney Int 47:377-387, 1995.
Glassock RJ, Cohen AH: The primary glomerulopat hies. Disease
A Month 42:329-383, 1996.
Jennette JC, Falk RJ: Diagnosis and management of glomerulonephritis and vasculitis presenting as acute renal failure. In: Mandal
AK, Hebert LA (eds), Medical Clinics of North America: Renal
Failure and Transplantation. Saunders, Philadelphia, pp. 893908, 1990.
Mariani AJ, Mariani MC, Macchioni C, Stams UK, Hariharan A,
Moriera A: The significance of adult hematuria: 1,000 hematuria
evaluations including a risk-benefit and cost-effectiveness analysis. J Urol 141:350-355, 1989.
Tiebosch ATMG, Frederik PM, van Breda Vriesman PJC, et al.:
Thin-basement-membrane nephropathy in adults with persistent
hematuria. N Engl J Med 320:14-28, 1989.

IMMUNOPATHOGENESIS OF
RENAL DISEASE
DIANE M. CIBRIK AND JOHN R. SEDOR

The mechanisms of renal injury that lead to progressive

disease are becoming better understood. Both humoral
and cell-mediated immunity are implicated in the primary
processes (i.e., the mechanisms that induce kidney injury)
and the secondary events (i.e., the cascade of tissue
responses activated by the initial renal damage) that
mediate renal inflammation and fibrosis. Experimental
models of kidney disease can be initiated by intrarenal
antibody deposition or by immune effector cells. The
extent and character of subsequent tissue immunologic
injury is influenced by the ongoing accessibility and
distribution of the kidney antigens, the biochemical characteristics of the antibody deposited and the specific,
secondary inflammatory mediators induced by the primary
injury process. Nonimmunologic mechanisms, although

Primer on Kidney Diseases, Second Edition

not the subject of this chapter, also can propagate ongoing

injury. These processes include adaptive hemodynamic
and physical forces that can cause intraglomerular hypertension and abnormal intravascular stress and strain (see
Chapter 66). Although not well studied in renal model
systems, information obtained with other experimental
systems shows that these stimuli can alter the proteins
synthesized by the target cells in a manner that contributes
to persisting inflammatory tissue injury.
This chapter presents an overview of basic immunology,
including the pertinent differences between humoral and
cellular immunity and the concepts of self-tolerance and
autoimmunity. With these fundamental mechanisms of immune injury in mind, the immunopathogenesis of kidney
disease will be explored as it relates to in situ immune
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

142

Diane M. Cibrik and John R. Sedor

complex disease, circulating immune complex disease,

and cell-mediated injury of the kidney. Finally, the role
of complement activation in the context of renal disease
will be discussed. Although these pathways of immunemediated renal injury are dealt with individually within
this chapter, many of these effector pathogenetic mechanisms often act in concert in vivo to cause renal damage
(Fig. 1).
IMMUNE RESPONSE

A specific immune response can be generated by both

the cellular and humoral immune systems (Fig. 2). T cells,
for the most part, direct the cellular immune response,
whereas antibodies derived from B cells direct the humoral
immune response. Antigen-presenting cells (APC) in conjunction with T cells are essential in the initiation of the
immune response. Examples of APCs include monocytes,
macrophages, dendritic cells, and B cells. These cells express multi-subunit, major histocompatibility complex
(MHC) proteins on their cell surfaces. These MHC molecules consist of class I and class II molecules which differ
in their function, chemical structure, cell distribution, and
length of peptide fragment that they will bind. Class I
MHC molecules are present on all nucleated cells, whereas
constitutive class II MHC molecule expression is restricted
to specific cells such as macrophages, B cells, certain endothelial cells, and some activated T cells. The expression of
class I MHC molecules on most cell types can be increased
by proinflammatory mediators such as interferons as an
amplification mechanism for certain T cell responses.

Cytokine-inducible increases in class II molecules on mononuclear phagocytes are also an important means of increas ing expression on these cells from constitutively low levels.
Both in vitro and in vivo studies suggest that cytokines can
stimulate nonimmune cells to express class II molecules,
but the functional consequences of these observations remain unclear. Self or nonself peptide fragments which interact with MHC molecules can be generated by one of two
intracellular pathways: (1) the degradation of cytosolic pro teins (e.g., viral proteins) within proteasomes with subsequent processing to permit interaction with class I MHC
molecules and transport to the cell surface (i.e., presentation), or (2) the endocytosis of exogenous proteins (self or
nonself) from the external milieu for association with, and
presentation by, class II MHC proteins. The subunits of
both classes of MHC molecules associate to form a
"groove" or pocket where the peptide fragments bind to
specific amino acid residues and are subsequently monitored by surveying T cells. Class I and class II are recognized by counter-receptors on CD8+ or CD4+ (helper) T
cells, respectively. Although the MHC is extremely polymorphic within a given species, an individual's complement
of MHC proteins can bind only a distinct subset of peptide
fragments. This restriction influences the character of an
individual immune response against a specific antigen.
Surveying T cells interact with the peptide:MHC complex through the highly variable regions of the T cell receptor (TCR). The binding of the MHC with the TCR confers
specificity on the immune response, but this association
alone is not sufficient to induce T cell activation. In order
to activate T cells, a second interaction is necessary between

Antigen-sensitized T cells

Antibody Deposition

Complement

[C5a,

C3b

Neutrophils, macrophages
_______ platelets _______

Renal |
Epithelium

Macrophages
Fibroblasts

MAC

Enzymes, Eicosanoids, Oxidants, Procoagulants, Core-forming proteins

Amplification of Inflammation
Further Damage to Glomerular or Tubulointerstitial Cells

..A
I Mesangial
Cells

Attenuation of glomerular or Tubular Basement Membranes Accompanying Fibrosis

FIGURE \ Effector pathways of the immune response which mediate renal injury in animal models. From
Couser WG: J Am Soc Nephrol 1:13-29, 1990.

17. Immunopathogenesis of Renal Disease

Lysis
Endogenously
of
synthesized
target
cell
antigen

143
Extracellular
antigen

Antigen
specific
activated

Clonal expansion
with antibody
production
FIGURE 2 Activation of the immune response. Antigenpresenting cells (APC) present endogenous or exogenous antigen
via MHC class I or class II molecules, respectively, to the T -cell
receptor (TCR). The interaction between the TCR and MHC: antigen complex, in conjunction with costimulator activities (see
text), induces T-cell activation and proliferation and APC activation. Activated APCs and T cells release cytokines that promote
cytotoxic T lymphocyte (CTL) formation and antigen-specific,
clonal B-cell expansion.

cell surface molecules present on the APC (i.e., B7 or

ICAM-1) and their counter-receptors expressed on the T
cell. This interaction, known as costimulation, aids in T
cell-APC adhesion and stimulates the intracellular signal
transduction events necessary for T cell activation. By requiring two signals for activation, the T cell can monitor self
from nonself (foreign) peptides and control its subsequent
responses. Recognition of the first signal (MHC: peptide
complex binding to the TCR) without the second, costimu latory signal may lead to anergy, or may lead to deletion
of self-reactive T cells from the repertoire in some instances. The production and release of soluble mediators
(i.e., cytokines from APCs and T cells) also requires both
T cell activation signals. These mediators, in turn, can amplify the immune response by enhancing APC activation,
T cell proliferation, and monoclonal B cell expansion. Thus,
many interactive signals are required to generate an effective immune response.
Humoral immunity is also important in generating an
extracellular immune response. B cells can be activated to

secrete antibodies by one of two mechanisms. In the first,

B cells can bind specific antigen directly via crosslinking
immunoglobulin (Ig) molecules present on the cell surface,
resulting in stimulation of biochemical signals within the
cell. Together with the help of cytokines, these intracellular
signals stimulate B cell proliferation and differentiation
into antibody-forming cells. In the second mechanism, B
cells act as APCs by internalizing antigen bound weakly
to surface Ig. Processed antigen is loaded onto class II MHC
molecules and brought to the cell surface where it can
be recognized by CD4+ T cells. In conjunction with the
secretion of cytokines by T cells and other costimulatory
activities, this B cell-T cell interaction stimulates antigenspecific, B cell proliferation and differentiation, and clonal
expansion with subsequent antibody production (Fig. 2).
Autoimmunity, either directly or indirectly, has been
implicated in the immunopathogenesis of many renal dis eases. Knowledge of the mechanisms that control selftolerance is important to understanding the immunopathogenesis of kidney disease. Humoral and cellular immune
responses against self antigens can result in autoimmune
diseases. However, many normal individuals produce autoantibodies or autoreactive T cells when challenged by a
foreign protein but fail to mount an autoimmune response.
Why don't these individuals develop an autoimmune dis ease? Immunologic tolerance is a physiological state in
which a potentially destructive immune response does not
develop against self-antigens. Several specific immune
mechanisms protect against potentially, self-destructive
processes. Clonal deletion, clonal anergy, and T cell peripheral suppression are three ways in which immunologic tolerance is achieved within the immune system. In clonal
deletion, self-reactive T and B cells in the thymus and bone
marrow, respectively, bind to autologous antigen during
maturation and are lost via negative selection. Functional
inactivation of T cells (clonal anergy) occurs when APCs,
presenting self-antigen, lack secondary costimulatory signals, such as B7. In peripheral suppression, cellular (suppressor T cells) and humoral molecules (antiinflammatory
cytokines) can suppress autoreactive lymphocytes. Because
of these three mechanisms, immunologic tolerance is maintained and autoimmune disease states are only rare and
aberrant events.
Autoimmunity and disease occur when all mechanisms
of self-tolerance are overcome and an immune response
against self-antigens is generated. Several important concepts about autoimmunity which are applicable to renal
disease have evolved from the experimental and clinical
literature. First, the response to a specific self-antigen occurs on a spectrum ranging from immunologic tolerance to
immune self-destruction. In the aggregate, environmental
factors such as microbial infection, genetic background, and
specific immunologic abnormalities influence the response.
Aberrant function of a single control component alone is
insufficient to generate an autoimmune response. Second,
autoimmune diseases can be systemic or organ-specific.
Finally, multiple effectors mediate tissue injury in autoimmune diseases.

144

Diane M. Cibrik and John R. Sedor

A number of mechanisms have been postulated to cause

loss of self-tolerance. First, peripheral tolerance may be
overcome. T and B-cell anergy is induced by antigen recognition in the absence of APC costimulatory activities or Tcell help, respectively. If local infection or inflammation
stimulates the expression of costirnulator molecules on
APCs or activates helper T cells that can interact with
potentially autoreactive B cells, anergy may be lost. Second,
antigen-independent stimuli may activate self-reactive lymphocytes not depleted during development. For example,
previously anergic B-cell clones can be stimulated nonspecifically by endotoxin from microorganisms or by viral infection to cause polyclonal B-cell activation that includes
self-reactive B cells. Third, antibodies or activated T cells
produced in the normal immune response to a pathogen
may recognize (i.e., crossreact with) native or modified self
proteins. One hypothesized mechanism of poststreptococcal glomerulonephritis suggests that antibodies against
streptococci crossreact with normal basement protein or
glomerular cell surface protein epitopes, a phenomenon
termed antigenic or molecular mimicry. Antigenic crossreactivity between foreign and self proteins can also occur
if the self protein is modified by complexing with haptens,
such as drugs, or by partial degradation to expose neoantigenic determinants. Fourth, loss of self-tolerance may also
occur when a neoantigen sequestered during development
is exposed by tissue damage and viewed as foreign. This
mechanism has been implicated in human Goodpasture's
disease (see below). Finally, loss of suppressor T-cell function or excessive T-cell help may create an imbalance between suppressor and helper influences to result in loss
of immunologic tolerance. Excessive cytokine production
may "break" T-cell tolerance or stimulate previously silent
B-cell clones to produce high levels of antibody. Through
specific and nonspecific immune mechanisms, autoimmunization plays a critical role in the loss of self-tolerance and
has been implicated in the pathogenesis of both experimental and human kidney disease.

IMMUNE MECHANISMS OF GLOMERULAR INJURY

Immune mechanisms are central to the pathogenesis of
most types of glomerulonephritis. These mechanisms can
be divided into two categories; humoral and cell-mediated.
Most of the focus in the past has been on antibodymediated injury. Two forms have been established: in situ
immune complex disease and circulating immune complex
disease (Fig. 3). Although these mechanisms of immune
injury are not mutually exclusive, each produces a distinct
immunofluorescence pattern within the glomerulus. The
requisite antibodies may be generated through the autoimmunization mechanisms described above. However, antibody deposition by itself is not sufficient to cause renal
injury. Cell-mediated tissue injury can occur through the
release of cytokine growth factors or other mediators of
inflammation. In addition, in vitro results suggest that inflammatory cells may directly injure the glomerulus, as in
pauci-immune glomerulonephritis. Activation of the complement cascade, either by the classical or alternative pathway, also plays a significant role in glomerular damage.
Humoral Immunity
In Situ Immune Complex Disease
The binding of antibody to antigen (i.e., formation of an
immune complex) localized to within a specific tissue is
known as in situ immune complex formation. In situ immune complex disease can occur in one of two ways: either
by antibodies binding to "native" antigen or to "planted"
antigen. Native antigens are intrinsic glomerular protein
constituents . Two well-established experimental models of
"native" antigen immune complex disease exist: antiglomerular basement membrane (GBM) nephritis which models human Goodpasture's disease and Heymann's nephritis,
an experimental model of human membranous nephropathy (Fig. 3). In anti-GBM nephritis, antibodies are directed
against intrinsic fixed antigens within the glomerular base-

In Situ

Circulating
Immune complex deposition

Anti -GBM

Heymann

Subepithelial
deposit
(Rare)
Basement
membrane
Subendothelial
deposit
Circulating
complex

Antigen
Antibody

FIGURE 3 Antibody -mediated glomerular disease. Circulating Immune complexes (A) deposit in the subendothelial
space and subsequently can be modified and rearranged to facilitate movement within the glomerulus. In situ immune
complex formation is illustrated by the deposition of anti-GBM antibodies in anti-GBM disease (B) or anti-megalin
antibodies in Heymann nephritis (C). From Cotran RS, Kumar V, Robbins SL: Pathologic Basis of Disease. WB Saunders
Co., Philadelphia 1994.

145

17. Immunopathogenesis of Renal Disease

ment membrane. In the experimental model, antibodies

generated in rabbits immunized with rat kidney tissue are
injected into normal rats with the subsequent development
of anti-GBM nephritis. These antibodies can also crossreact
with basement membrane present in lung alveoli, yielding
simultaneous lung and kidney lesions, or Goodpasture's
syndrome. In this syndrome, the human basement membrane antigen recently has been identified as two discontinuous epitopes within the noncollagenous domain of the
3 chain of collagen type IV. The current model of the
pathogenesis of Goodpasture's syndrome proposes that
these antigenic epitopes are concealed during the normal
assembly of type IV collagen into basement membrane.
Subsequent exposure of these privileged epitopes, possibly
caused by modification of the GBM in response to smoke,
hydrocarbons, infections, or other precipitating factors results in the formation of an antigenic structure, antibody
production, and inflammation.
In the Heymann's nephritis model, rats are immunized
with preparations of proximal tubular brush border. The
animals subsequently develop renal lesions that resemble
human membranous nephropathy. The rat Heymann antigen has been identified as gp330 or megalin, a 330-kD
glycoprotein member of the LDL receptor superfamily
which is localized in the clathrin-coated pits of visceral
epithelial cells and on the proximal tubule brush border.
Antibody binding to gp330 results in alternative pathway
complement activation with subsequent shedding of the
immune complexes from the cell surface into the subepithelial space. In membranous nephropathy, no cellular infiltrates are seen on light microscopy, most likely because
the immune deposits form within the subepithelial surface
of the glomerular capillary wall and are not in direct contact
with circulating immune cells. However, these immune deposits can interact with soluble mediators to initiate an
inflammatory response.
Comp lement is the best studied of these mediator systems. Experimental evidence demonstrates that complement pathways are activated by the immune complexes
deposited within kidney (see below). The complement
membrane-attack complex (MAC, C5b-9) can initiate glomerular capillary wall damage that results in proteinuria,
although the specific mechanisms by which MAC causes
proteinuria are unknown. In vitro studies have demonstrated that MAC is a potent stimulus for glomerular cells
to produce oxidants, proteas es, prostaglandins, and cytokines. Through these processes, MAC may be responsible
for the alteration in the GBM permeability seen in membranous nephropathy. Megalin, like other members of the
LDL receptor superfamily, may bind and clear proteases
that escape through the glomerular filtration barrier. When
antibody binding interferes with normal, receptormediated endocytosis to fix megalin and any associated
proteins in immune deposits, megalin-associated enzymes
may degrade the matrix constituents of the glomerular
basement membrane or activate other soluble mediator
systems.
In contrast to these native antigens, "planted" antigens
are not intrinsic to the glomerulus. They may become

trapped within the glomerulus because of charge and size

characteristics, molecular configuration, or carbohydrate
content. For example, segments of the glomerular filtration
barrier can interact with positively charged proteins which
act as "planted" antigens to form in situ immune complexes. Occasionally, immunoglobulins themselves are
"planted" antigens and attract antiidiotypic antibodies,
such as rheumatoid factors, to augment in situ immune
complex formation. "Planted" antigens fall into two
groups: endogenous and exogenous. Examples of endogenous antigens are DNA, IgA, and other immunoglobulins.
DNA has an affinity for the moieties present on the GBM,
whereas immunoglobulin aggregates can get trapped within
the mesangium. In one rat model for anti-GBM nephritis,
rabbit anti-rat GBM antibodies, in amounts insufficient to
directly cause glomerular injury, can serve as foreign antigen and initiate an antiidiotypic response. The possibilities
for exogenous antigens are endless. Drugs, lectins, and
infectious agents can have an affinity for glomerular constit uents. Thus "planted" antigens can either be trapped nonspecifically within the glomerulus or specifically bind to
glomerular constituents.
Circulating Immune Complex Disease

In this form of immune complex disease, soluble circulating antigen-antibody complexes become passively trapped
within the glomerulus, subsequently resulting in renal injury through complement activation, leukocyte infiltration,
and often glomerular cell proliferation. Unlike the antibodies that form in situ immune complexes, the antibodies that
make up these complexes have no specificity for glomerular
components. Although usually cleared by the reticuloendothelial phagocytes, circulating immune complexes can deposit within the glomerulus and other vessel walls if they
have appropriate physical and chemical characteristics.
Due to their charge and size characteristics, circulating
immune complexes initially lodge in the subendothelial
space, rarely progressing further into the basement membrane. As a result subendothelial deposits or paramesangial
deposits are commonly identified in renal biopsies. Once
formed, however, deposited immune complexes can be
modified and rearranged in situ as a result of changing
antigen:antibody ratios. Free antibody will bind to other
antigenic epitopes, increasing the size of the immune complex deposits. Excess antigen may actually dissolve already
formed deposits. Several studies have suggested that this
ongoing in situ modification can facilitate movement from
one glomerular site, such as the subendothelial space, to
another, such as the subepithelial glomerular basement
membrane.
The noncovalent nature of the antigen and IgG interaction may allow the circulating immune complex to separate
within the kidney circulation into component antibody and
antigen. Intrarenal immune complex formation can result
from the free antigen planting within the GBM followed
by subsequent binding of antibody. These models suggest
that circulating immune complexes do not directly deposit
in the kidney. However, several studies in which covalently
(and therefore irreversibly)-linked antigen and antibody

Diane M. Cibrik and John R. Sedor

146
complexes were injected into rodents clearly have demonstrated glomerular immune complex deposition, suggesting
that circulating complexes can directly deposit in the kidney. The net charge of the injected complexes determined
the site of their intraglomerular localization.
The classic experimental models for glomerular injury
induced by circulating immune complexes are acute and
chronic serum sickness. In these models, rabbits are immu nized with a heterologous protein, bovine serum albumin
(BSA). Glomerulonephritis, which is characterized by subendothelial and mesangial immune deposits, develops
after the animals develop anti-BSA antibodies. Circulating
immune complexes also have been implicated in the pathogenesis of human renal diseases such as the autoimmune
nephritides associated with cryoglobulinemia and lupus erythematosus, poststreptococcal glomerulonephritis, and
tumor-antigen associated membranous nephropathy. Human serum sickness, also associated with circulating immune complex-mediated renal disease, has become rare
but can be seen with antivenom therapy for poisonous
snake, fish, and spider bites and with experimental therapies using mouse monoclonal antibodies. As already indicated in the discussion of in situ complex disease (above),
circulating immune complexes can contain either endogenous or exogenous antigens. Examples of endogenous
antigens identified in circulating immune complexes and
implicated in human glomerulonephritis include DNA,
thyroglobulin, thyroid antigens, immunoglobulins, red cell
stroma, and tumor neoantigens. Exogenous antigens implicated in glomerulonephritis are primarily derived from microbes, such as bacteria (streptococci) and viruses (hepatitis
B and C). Fungi and parasites also can serve as exogenous
antigen. Soluble foreign serum proteins (cow, goat etc.),
food antigens, and chemicals, such as gold and mercury,
act as exogenous antigens to generate circulating immune
complexes that can result in serum sickness or renal disease.
Cell-Mediated Immunity
Many types of glomerulonephritis exhibit hypercellularity upon light microscopic examination, reflecting infiltration of the kidney by inflammatory cells such as neutrophils,
monocytes, T cells, or platelets, or proliferation of resident
glomerular cells. The location of the glomerular immune
complex deposits controls, in part, whether inflammatory
cells will be involved in mediating glomerular injury. As
previously discussed, immune complexes at the subepithelial site are separated from the circulation by the GBM,
limiting leukocyte access. Therefore, as exemplified by
membranous nephropathy, a nonproliferative lesion results
because complement-derived chemoattractants and antibody cannot interact directly with circulating immune cells.
In contrast, immune deposits within subendothelial or mesangial sites are more readily accessible to circulating inflammatory cells and can lead to cell adhes ion, activation,
and transmigration across the capillary wall. Many human
renal inflammatory diseases, such as diffuse proliferative
lupus nephritis, IgA nephropathy, type I membranoproliferative glomerulonephritis, Henoch-Schonlein purpura,

and some types of acute postinfectious nephritis result from

deposition of immune complexes at mesangial and subendothelial sites. Once localized, inflammatory cells mediate
immune injury by releasing soluble mediators such as cytokines, proteolytic enzymes, or reactive oxygen metabolites (Table 1). These mediators can amplify the existing
immune response, directly injure nearby glomerular cells,
or degrade or denature the matrix protein constituents of
the glomerular filtration barrier.
Mediators of immune glomerular injury can be divided
into two groups: primary and secondary. Antibody (dis cussed previously) and T cells are primary mediators,
whereas complement, other inflammatory cells, and their
secretagogues are secondary mediators of the immune response. Recently, a number of studies using animal models
of glomerulonephritis have strongly implicated T-cell mediated, immune mechanisms of kidney injury, which are independent of antibody deposition. For example, transfer of
sensitized T cells from donors with progressive glomerulonephritis to normal recipients can initiate glomerular injury
in the absence of immune complex deposition. In a rat

TABLE I

Immune Mediators Implicated in Renal Injury

Complement
Membrane attack complex (C5b-9)
C3b C5a
Reactive oxygen metabolites
Hydroxy radicals Hydrogen
peroxides Superoxides
Hypochlorous acids
Enzymes
PMN myeloperoxidase
Neutral serine proteases (elastase, cathepsin G)
Cysteine proteases (cathepsin B & H)
Metalloproteinases (gelatinase)
Cell adhesion molecules
Cytokin es and growth factors (secretory mediators)
IL-1/3
TNFa
TGF/3
Basic FGF
Prostaglandins
PDGF
Plasminogen activator
Eicosanoids (thromboxanes, leukotrienes)
Polypeptide growth factors
Coagulation cascade
Thrombin Plasmin
Plasminogen activators
Plasminogen activator inhibitors
Note: A wide variety of proinflammatory and immunomodulatory mediators have been associated with
renal immune/inflammatory injury. The above list is
not intended to be exhaustive.

17. Immunopathogenesis of Renal Disease

model of granulomatous nephritis with proteinuria induced

by a chemically reactive form of a well-studied hapten,
adoptive transfer of T cells from hapten-immunized animals, but not administration of antihapten antibody, induced disease in naive recipients. These studies suggest
that intrinsic kidney cells can present antigen in vivo and
that cell-mediated immune mechanisms can cause renal
injury. In a rat model of nephrotoxic serum nephritis, depletion of CD8+ T cells prior to the administration of nephro toxic serum completely prevented proteinuria, although
the amount of antibody present in the glomeruli of CD8+
T-cell-depleted animals was similar to control animals. In
another experimental model employing chickens, in which
B cell development was arrested by neonatal bursectomy,
immunization with GBM induces a severe proliferative
glomerulonephritis characterized by T cells within the glomeruli without evidence of anti-GBM antibodies.
Besides T cells, other inflammatory leukocytes, such as
neutrophils, macrophages, and platelets, can migrate into
and localize within the glomerulus to promote immune
injury. Release of cytokines, proteases, and/or oxidants
from these cells (Tables 1 and 2) can damage the GBM
resulting in detachment of the epithelial cells foot processes
from the filtration barrier with subsequent protein leakage.
Effacement of epithelial foot processes is an electron micrographic hallmark of proteinuric renal disease. Neutrophils
can infiltrate the glomerulus via Fc receptor-mediated adCirculating cells
TABLE I

Cellular Sources of Immune Mediators

PMNs

Macrophages

Platelets

Renal cells
Mesangial cells

Endothelial cells

Olomerular epithelial cells

Tubular epithelial cells

Reactive oxygen metabolites

Proteases
Myeloperoxidase
Reactive oxygen metabolites
Cytokines
Prostaglandins
Platelet-derived growth factor
Eicosanoids
Coagulation factors
Platelet -activating factor
Eicosanoids
Platelet -derived growth factor
Reactive oxygen metabolites
Neutral proteases
Cytokines
Prostaglandins
Plasminogen activator
Platelet -derived growth factor
Basic fibroblast growth factor
Adhesion molecules
Cytokines
Growth factors
Proteolytic enzymes
Megalin
Oxygen radicals
Class II MHC proteins
Cytokines

147

herence, interactions with adhesion molecules expressed

on endothelial cell surfaces, and by the actions of chemoattractants such as C5a from complement activation. The
major soluble mediators released by neutrophils are reactive oxygen metabolites such as peroxide, hydroxyl radicals,
oxygen free radicals, and superoxides. Myeloperoxidase, a
positively charged enzyme found within polymorphonuclear neutrophils, binds to anionic sites within the glomerulus and catalyzes the reaction between peroxide and halide
anion to generate hypohalous acids. These acids are
thought to directly damage the GBM, leading to proteinuria. In addition to these reactive oxygen metabolites, activated neutrophils release neutral serine proteases such as
elastase and cathepsin G that have been shown to degrade
the GBM in vitro. Macrophages localize within the glomerulus in response to chemoattractant cytokines (chemo kines) derived from cell-mediated immune reactions or
in response to antibody deposition via their Fc receptor.
Activated macrophages secrete numerous proinflammatory and immunomodulatory molecules, including IL-1/3
and TNF-a, TGF -jS, prostaglandins, and lipid mediators
such as platelet-activating factor and lipoxygenase products
(see Tables 1 and 2).
Many of these molecules enhance the immune response
or alter the character and functions of resident glomerular
cells. For example, platelet-derived growth factor stimu lates mesangial cell proliferation, which in turn, leads to
the increased production of mesangial cell proteases, resulting in glomerular injury. TGF-/3 activates mesangial cells
to increase extracellular matrix deposition, which leads to
scarring and glomerulosclerosis. Antibodies, complement,
and immune complexes as well can activate mesangial cells
to produce various cytokines that can amplify the immune
response or promote glomerular injury and sclerosis. Platelets localize within the glomerulus in a complementdependent manner during immune-mediated injury and
release growth factors and eicosanoids which may influence
glomerular permeability. In models of experimental glomerulonephritis, depletion of platelets by antiplatelet antibody reduces proteinuria significantly, supporting a role
for platelets in glomerular injury. Although each of the
inflammatory cells has been discussed individually, each,
along with its many soluble mediators, can act in concert
with the others to produce immune injury (Fig. 1).
Complement pathway activation may produce immunemediated renal injury in one of two ways: by recruitment
of inflammatory cells to the site of immune injury or by
generation of the C5b-9 membrane attack complex (MAC).
C5a, a product of the complement cascade activation attracts neutrophils to the site of immune complex deposition
and leads to their adherence via their C3b receptors. In
Heymann nephritis, complement depletion abolishes pro teinuria. In contrast, leukocyte depletion has little effect
on indices of renal injury. These results suggest that proteinuria in Heymann nephritis is a complement dependent
process mediated by MAC and that complement-induced
renal injury can occur without involvement of inflammatory
cells. The mechanisms by which MAC causes proteinuria
have yet to be determined. In vitro complement compo-

148

Diane M. Cibrik and John R. Sedor

nents activate glomerular epithelial cells (GEC) to produce

proinflammatory mediators such as oxidants, proteases,
and cytokines which may alter the permeability of the
GBM. MAC may also stimulate the GEC to produce excess
extracellular matrix components that lead to an altered
GBM architecture and ni creased permeability. Lastly,
MAC may alter the binding of extracellular matrix proteins
to their receptors (integrins) on GEC, leading to detachment of the GEC and proteinuria.
The alternative pathway of complement activation has
been specifically implicated in the pathogenesis of membranoproliferative glomerulonephritis (MPGN) and IgA nephropathy. Seventy percent of patients with a particular
type of MPGN (type II) have C3 nephritic factor present
in their sera. This factor, instead of properdin, binds and
stabilizes alternative C3 convertase. The C3 nephritic factor
is an autoantibody to the alternative C3 convertase, suggesting that MPGN is an autoimmune disease. In IgA nephropathy, IgA codeposits along with C3 and properdin
in the mesangium. In both diseases, early complement components are usually absent in glomeruli, suggesting a role
for the alternative pathway in the immunopathogenesis of
both these glomerulonephritides.
In contrast to the glomerulonephritides discussed above,
which are associated with antibody-mediated immune in jury, necrotizing and crescentic glomerulonephritis, also
known as pauci-immune glomerulonephritis, lacks any evidence of glomerular antibody deposition. The immunopathogenesis of this disease remains unclear, but approximately 90% of patients with pauci-immune vasculitis or
glomerulonephritis have circulating antineutrophil cytoplasmic antibodies (ANCA) (see Chapter 27). Examples
of ANCA-associated diseases are polyarteritis nodosa, leukocytoclastic angiitis, and Wegener's granulomatosis.
ANCAs have been implicated in the pathogenesis of these
diseases and can be classified into two groups based on
their characteristic immunohistochemical staining patterns
of neutrophils. C-ANCA yields a cytoplasmic staining pattern, whereas P-ANCA gives a peri nuclear staining pattern.
The antibodies are specific for enzymes present within the
primary granules of neutrophils. For C-ANCA, the enzyme
is serine protease 3 and for P-ANCA it is myeloperoxidase.
The mechanisms by which ANCA cause glomerulonephritis remain to be elucidated. In vitro evidence suggests that
these antibodies stimulate neutrophils to release toxic oxy gen radicals and proteolytic enzymes, which subsequently
cause fibrinoid necrosis. Antimyeloperoxidase ANCA antibodies along with TNF and endotoxin can stimulate neutrophils to damage human endothelial cells in a dosedependent manner in vitro.
TUBULOINTERSTITIAL NEPHRITIS

Humoral and cell-mediated immune mechanisms similar

to those characterized in the immunopathogenesis of experimental glomerular disease also have been implicated
in primary tubulointerstitial immune injury. In situ immune
complexes can form and circulating immune complexes can
deposit within the tubulointerstitium. Possible mechanisms

of tubulointerstial immune complex deposition include

"overflow" of glomerular immune complex "binding" sites
resulting in tubulointerstitial immune complex formation,
cross-reactivity between glomerular and tubulointerstitial
antigenic epitopes and simultaneous production of autoreactive antibodies against multiple self-antigens.
Cell-mediated pathways of immune injury appear to be
the primary mechanisms of tubulointerstitial nephritis and
have been best characterized in a model of tubulointerstitial nephritis produced by antitubular basement membrane
antibodies. Although initiated by heterologous antibodies
in susceptible rodents, antibody-mediated injury is minimal. Rather, sensitized T cells are required for disease in
both an acquired model as well as in an inbred mouse
strain (kdkd) that spontaneously develops tubulointerstitial nephritis. The target antigen, 3M-L, is synthesized by
proximal tubular epithelial cells. Antibody binding to its
antigenic epitope activates the mechanisms of immune injury described in detail in the preceding sections.
Tubulointerstitial inflammation almost always accompanies proteinuric renal disease, and some of the most intriguing recent work in renal immunopathogenesis has attempted to elucidate the molecular signals that underlie
this link between glomerular and interstitial pathology.
Proximal tubular cells actively reabsorb proteins via
clathrin-coated pits via a receptor or constitutive pathway.
Reabsorbed proteins are degraded by lysosomes into
amino acids. The increase in proteinuria that accompanies
many renal diseases may overwhelm the capacity of the
tubular epithelial cell to perform this function. Accumulation of abnormally high amounts of protein in the endolysosomes leads to organelle rupture and leak of lysosomal
enzymes into the cytoplasm. In addition, protein congestion
may increase intracellular oxidant stress, a process known
to activate a number of signaling enzymes and transcription
factors. For example, the activity of the transcription factor
NF-x-B regulates expression of many proinflammatory cytokines and is increased by oxygen radicals. Thus, tubular
protein overload, a nonimmunologic stimulus of renal injury, can activate interstitial immune mechanisms to cause
tubulointerstitial inflammation.
Bibliography
Abbas A, Lichtman AH, Pober JS: Cellular and Molecular Immunology. WB Saunders Company, Philadelphia, 1994.
Bolton WK, Tucker FL, Sturgill BC: New avian model of experimental glomerulonephritis consistent with mediation by cellular
immunity. / Clin Invest 73:1263276, 1984.
Brady HR, Papuyianni A, Serhan CN: Leukocyte adhesion promotes biosynthesis of lipoxygenase products by transcellular
routes. Kidney Int 45:S-90- S97, 1994.
Cotran RS, Kumar V, Robbins SL: Pathologic Basis of Disease.
WB Saunders Company, Philadelphia, pp. 171-240 and 927990, 1994.
Couser WG: Mediators of immune glomerular injury. Clin Investig
71:808-811, 1993.
Couser WG: Pathogenesis of glomerulonephritis. Kidney Int
44:S519-S526, 1993
Emancipator SN, Sedor JR. Cy tokines and Renal Function. In:
Remick DO, Kunkel SL (eds) Cytokines in Health and Disease;

18. Minimal Change Nephropathy

149

Physiology and Pathophysiology. Dekker, New York, pp. 467488, 1992. Ewert BH, Jennette JC, Falk RJ: Antimyeloperoxidase antibodies
stimulate neutrophils to damage human endothelial cells. Kidney
Int 41:375-383, 1992. Farquhar MG: Molecular analysis of the
pathological autoimmune
antigens of Heymann nephritis. Am J Pathol 148:1331-1337,
1996. Glassock RJ: The glomerulopathies. In: Schrier RW (ed)
Renal
and Electrolyte Disorders. Little, Brown, Co., Boston, pp. 727806, 1992. Kawasaki K, Yaoita E, Yamamoto T, et al.:
Depletion of CDS
positive cells in nephrotoxic serum nephritis of Wky rats. Kidney
Int 41:1517-1526, 1992.
Kelly CJ. Tomaszewski J, Neilson EG: Immunopathogenic mechanisms of tubulointerstitial injury. In: Brenner BM, Tisher CC
(eds) Renal Pathology. 2nd ed., Lippincott, Philadelphia, 2. pp.
699-722, 1996

Makker SP: Mediators of immune glomerular injury. Am J Nephrol

13:324-336, 1993. Neilson EG: Pathogenesis and therapy of
interstitial nephritis.
Kidney Int 35:1257-1270, 1989 Paul WE: The immune
system: An introduction. In: Paul WE
(ed) Fundamental Immunology. Raven Press, New York, pp.
1-20, 1993 Remuzzi G, Ruggenenti P, Benigni A:
Understanding the nature
of renal disease progression. Kidney Int 51:2 -15, 1997.
Rennke HG, Klein PS, Sandstrom DJ et al.: Cell mediated injury
in the kidney: Acute nephritis induced in the rat by azobenzenearsonate. Kidney Int 45:1044-1056, 1994. Roy S, Scherer MT,
Briner TJ et al.: Murine MHC polymorphism
and T-cell specificities. Science 244:572-574, 1989 Sedor JR:
Cytokines, growth factors and glomerulonephritis
immune-mediated renal disease. Semin Nephrol 12:428-440,
1992.

MINIMAL CHANGE NEPHROPATHY

NORMAN J. SIEGEL

OVERVIEW AND TERMINOLOGY

HISTOPATHOLOGY

Minimal change nephropathy or minimal change disease

(MCD) is a histopathologic lesion which is almost always
associated with the nephrotic syndrome at the onset of
disease. Other terms such as lipoid nephrosis, nil disease,
and idiopathic nephrotic syndrome have been used, interchangeably, with minimal change nephropathy. Because of
the finding of lipids in renal tubule cells and the urine of
patients with nephrotic syndrome, the term lipoid nephrosis
has been used to describe patients with steroid-responsive
nephrotic syndrome. However, this term is not synonymous
with MCD because other histologic lesions may also show
a response to steroid therapy. The term nil disease is an
oversimplification of the histopathology seen in patients
with minimal change disease because it refers predominantly to the findings by light microscopy alone. Idiopathic
nephrotic syndrome actually refers to a wide variety of
histopathologic lesions which are associated with primary
nephrotic syndrome. Thus, while each of these terms is
prevalent in the literature, contemporary understanding of
the natural history and therapeutic options requires more
precise terminology.

Minimal change nephropathy is defined on light microscopy by a lack of definitive alteration in glomerular structure. While the degree of alterations which may remove a
biopsy from this category has been debated, it is generally
agreed that the cellularity of the glomerulus must be minimal and that the tubular and interstitial structures must
also be normal. There may be some doubly refractile appearance or lipid droplets in the tubule cells but there
is no evidence of tubular atrophy or interstitial fibrosis.
Immunofluorescent staining also shows no change from
normal and an absence of immunoglobulin or complement
protein deposition. The most obvious and consistent finding
in patients with MCD is a characteristic fusion of epithelial
foot processes seen on electron microscopy (Fig. 1). There
is normal fenestration of the endothelial cells lining the
capillary loop; the glomerular basement membrane is uniform in thickness and structure, but the epithelial cells
show swelling and continuous contact with the glomerular
basement membrane. Because a biopsy is susceptible to
sampling error, it must be remembered that lesions which
affect only some glomeruli, such as focal and segmental

Primer on Kidney Diseases, Second Edition

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

Norman J. Siegel

ISO

BOW SP

FIGURE I Electron micrograph of glomerular capillary loop in

patient with MCD. The endothelial cell (End) and basement membrane are normal in content and structure. The epithelial cell
(Epith) demonstrates diffuse swelling of the pseudopods, footprocess fusion. There are no electron dense deposits.

glomerulosclerosis, may be inadvertently misdiagnosed

as MCD.
The role of a renal biopsy in the initial management of
patients who present with nephrotic syndrome is controversial. In most cases, children are treated with a course of
steroids and biopsies are reserved for those who are steroid
resistant. Because of the lower prevalence of steroidresponsive lesions, adolescents and adults are usually biopsied prior to treatment. For those patients of any age with
a complicated clinical presentation or course of disease, an
assessment of histopathologic changes is essential.
CLINICAL PRESENTATION

Although generally thought to be a childhood disease,

MCD presents in both children and adults. The insidious
onset of nephrotic syndrome, usually manifested by edema
formation, is the most common presentation. In children,
the onset of disease is after the first year of life, with a
peak incidence between 24 and 36 months of age and a
strong male predominance. In preadolescent children, 85
to 95% with idiopathic or primary nephrotic syndrome will
have MCD. In adolescents and young adults, the prevalence

of MCD declines to approximately 50% and the male predominance begins to disappear. In patients over the age
of 40 years with primary nephrotic syndrome, the incidence
of MCD is 20 to 25% and there is a nearly equal distribution
between males and females.
In patients presenting with the typical features of MCD,
a relatively "pure" nephrotic syndrome is usually observed.
Nephritic features such as hypertension, hematuria, and
reduced renal function are relatively uncommon. Any one
of these nephritic features may occur in 15 to 20% of patients with MCD but the combination of two or more of
these factors is decidedly unusual and should make one
consider a different diagnosis. The finding of gross hematuria or red cell casts would generally not be considered
compatible with the diagnosis of MCD. Thus, the predominant clinical features are those of nephrotic syndrome with
heavy proteinuria, low serum albumin, edema formation,
and elevated serum cholesterol. Other than the findings of
oval fat bodies which appear as maltese crosses on examination of the urine under a polarized lens (see Chapter 3,
Fig. 3) the urinalysis is normal. Serum complement levels
are normal and antinuclear antibodies and cryoglobulins
absent. Serum immunoglobulins may be abnormal with a
reduction of serum IgG levels to 20% or less of normal
values, a less severe reduction in IgA, and a mild increase
in IgM and IgE levels.
Although MCD is usually associated with a primary or
idiopathic nephrotic syndrome, secondary MCD also occurs (Table 1). In 80 to 90% of children, MCD is idiopathic
although cases associated with the ingestion of heavy metals
such as mercury or lead as well as acquired immunodeficiency syndrome have been reported. In adult patients,
especially the elderly, the association of this MCD with
nonsteroidal anti-inflammatory drugs is particularly important because of their very frequent use. Hodgkin's dis ease and lymphoproliferative disorders must be kept in

TABLE I

Secondary Causes of Minimal

Change Nephropathy
Drugs
Nonsteroidal antiinflammatory agents
Ampicillin/penicillin
Trimethadione
Toxins
Mercury
Lead Bee
stings
Infection
Mononucleosis
HIV
Immunizations
Tumors
Hodgkin's lymphoma
Other lymphoproliferative diseases
Carcinoma
Obesity

18. Minimal Change Nephropathy

mind when older patients present with MCD. Although

most patients with obesity and nephrotic syndrome have
focal segmental glomerulosclerosis, some of these patients have responded to steroids or been documented to
have MCD.
RESPONSE TO THERAPY

The benchmark for therapy in MCD is the use of corticosteroids. No other cause of nephrotic syndrome is as exquisitely sensitive to treatment with steroid therapy as MCD.
Because of the high prevalence of this lesion in children,
disappearance of proteinuria in response to the oral administration of prednisone is considered diagnostic for MCD.
Characteristically, these young children respond to treatment with a diuresis and clearing of proteinuria within
about two weeks of initiation of prednisone therapy, usually
at a dose of 2 mg kg ' day"' not to exceed 60 mg/day.
Some children may have a slower response to initial therapy
and not clear their proteinuria for 6 to 8 weeks. A response
to therapy in 80 to 90% of adolescents and adults with
MCD has also been documented. However, in these age
groups, the response is slower and a prolonged period of
therapy, in some cases up to 16 weeks, may be required
before complete remission of proteinuria is achieved. Unfortunately, this distinct effect of steroids on MCD has
been used to presume that remission of proteinuria while
on prednisone implies underlying MCD.
The clinical course of MCD is frequently described in
terms of the patient's response to steroids (Table 2). Complete remission is defined as complete resolution of proteinuria for at least 3 to 5 consecutive days; a partial remission
is a reduction in the degree of proteinuria without complete
clearing; and a relapse is a reoccurrence of proteinuria for
at least 3 to 5 consecutive days. Children and adolescents
tend to have complete remissions whereas partial remis sions are more frequent in adult patients. The clinical outcome for children with MCD, as related to steroid therapy,
is outlined in Fig. 2. A similar, although not identical, pattern can be expected in adults. About 10% of patients
initially treated with steroid therapy will not respond to
treatment and will have early steroid resistance. Alternative methods of administration of steroids have been at-

TABLE 2

Clinical Course of Minimal Change Disease as Related to

Steroid Therapy
Steroid responsive
Complete remission of proteinuria within 8-12 weeks with
infrequent relapses
Frequently relapsing steroid dependent Relapses occur during
the taper of steroids Relapses occur at rate of twice every 6
months or six times
every 18 months Relapses occur within 2 weeks of
cessation of therapy
Steroid resistant Failure to obtain a remission
within 12 weeks

151
Initial Response to Steroids

Subsequent Clinical Course

" 20 No Relapses 70
Relapsing Course

90 Remission
100
Patients With MCD
I 1 0 Steroid Resistant

FIGURE 1 Clinical outcome

for children with MCD.

40 Steroid Responsive
30 Frequently Relapsing
Steroid Dependent
_

5 Late Steroid Resistant

C, 8 Respond to Alternative Therapy 2

Refractory to Treatment

tempted in this group of patients,

but have not been particularly successful. These patients
frequently respond to cytotoxic or immunosuppressive
therapies.
Of the patients who achieve a remission on initial steroid
therapy, 75 to 85% experience one or more relapses (Fig.
2). In those patients who are destined to follow a relapsing
course, the first relapse usually occurs within 6 to 12 months
of the onset of their disease, although in some patients the
first relapse may be delayed as long as 24 to 30 months.
In patients with a relapsing course, 50 to 65% can be expected to have a steroid-responsive clinical profile with
frequent relapses occurring over 3 to 5 years. However,
for 25 to 30% of patients with MCD, a more protracted
clinical course occurs and their diseas e is described as being
frequently relapsing or steroid dependent. A small proportion of these patients may develop late steroid resistance
at which point a repeat renal biopsy may show evolution
to other histopathologic patterns such as focal segmental
glomerular sclerosis. Thus, the majority of patients with
MCD have a good response to initial steroid therapy, and
a disease course characterized by relapses of their nephrotic
syndrome but, overall, an excellent long-term prognosis.
Patients with frequently relapsing/steroid-dependent
MCD present the greatest therapeutic challenge. In the
long run, the relapsing nature of their disease is likely to
disappear; however, over the short term, deleterious side
effects of continued steroid therapy may occur
particularly in adults and geriatric patients. Most often,
when a patient has a frequently relapsing/steroid-dependent clinical pattern, an attempt is made to decrease the
frequency of relapses by maintaining the patient on chronic
low dose corticosteroid therapy. Many patients will tolerate
low doses of prednisone on a daily or alternate day basis
without significant side effects. For those patients who require large doses of medication to remain in remission or
who develop significant side effects, alternative therapy
must be considered.
Cyclophosphamide and chlorambucil have proved to be
the most effective alternative therapies for patients with
MCD. These drugs have been shown to be effective in
both children and adults with frequently relapsing/steroiddependent nephrotic syndrome. Typically, prednisone will
be used to induce remission, then an 8- to 12-week course
of cyclophosphamide in a dose of 2 mg kg" 1 day" 1 or
chlorambucil in a dose of 0.1 to 0.2 mg kg~' day" 1 will
be added. The most pronounced and dramatic effect is the
prolonged period of remission of the nephrotic syndrome

152
which is achieved in patients treated with these agents.
After cessation of the cytotoxic agent, 30 to 40% of patients
will have no subsequent relapses. Among relapsers, the
average period of remission is 18 to 24 months. Subsequent
relapses are usually more steroid responsive than prior to
cytotoxic therapy. Thus, steroid-related side effects such
as a cushingoid appearance or growth retardation can be
markedly diminished and reversed. In patients unable to
tolerate steroids, it is also possible to induce a remission
of the nephrotic syndrome with either of these agents alone;
in the elderly, alkylating agents have been used as initial
therapy. These agents have a number of serious side effects,
including cystitis, alopecia, leukopenia, gonadal toxicity,
potential for malignancy formation, and seizures; their use
requires caution and careful judgment.
Other medications have also been used to treat patients
with either steroid refractory or frequently relapsing/steroid-dependent disease. Cyclosporine has been reported
to be effective in some adult and pediatric patients with
an initial steroid-resistant course. Its primary utility is to
achieve a steroid-sparing effect because the relapse rate
is high after Cyclosporine is discontinued. The prolonged
period of remission produced by treatment with cyclophosphamide or chlorambucil is not observed. In addition, longterm therapy with Cyclosporine may be complicated by
nephrotoxicity. Despite multiple clinical trials, Cyclosporine
remains a late therapeutic option for patients with MCD.
Levamisole has also been used to achieve a steroidsparing effect.
For patients unresponsive to these therapeutic interventions, symptomatic control of edema formation with salt
restriction and diuretics are the mainstays of therapy. If
complications of anasarca occur, such as pleural or pericardial effusions, severe hyponatremia, or fascitis, the infusion
of albumin can be beneficial to mobilize the extracellular
fluid and prevent pulmonary or cardiac decompensation.
This therapy, however, should be undertaken with caution
because the shifts in intravascular volume associated with
albumin infusion can result in severe hypertension or congestive heart failure and because the beneficial effect is
short term since the albumin is rapidly excreted in a patient
with heavy proteinuria. Proteinuria can be reduced with
nonsteroidal antiinflammatory drugs (NSAIDS) and angiotensin converting enzyme (ACE) inhibitors. These therapies may be of benefit in patients who are refractory to
treatment of their nephrotic syndrome, have severe proteinuria, and who are developing malnutrition because of
the quantity of protein lost in the urine. However, the
reduction in proteinuria in patients treated with either of
these medications is, in large part, related to alterations in
renal blood flow and glomerular filtration rate rather than
a direct effect on the mechanisms of the proteinuria. Consequently, patients with MCD who are treated with either
NSAIDS or ACE inhibitors are particularly susceptible to
alterations in intravascular volume and renal perfusion and
are prone to the development of acute renal failure or
hyperkalemia. While these drugs are generally well tolerated and may be beneficial in a selected subset of patients,
careful monitoring of patients receiving them is essential.

Norman J. Siegel
COMPLICATIONS AND LONG-TERM OUTCOME

The primary complications of MCD are related to persis tent nephrotic syndrome or to side effects of therapy. The
most common complications of steroid therapy include cushingoid facies, stria, and acne, as well as cataracts. In general, these medications are much better tolerated in children than in adults. The induction of a prolonged remission
by a cytotoxic agent permits regression of the majority of
the steroid-related side effects. Stria and cataracts usually
persist; however, catch-up growth frequently occurs and
the cushingoid appearance disappears. Complications of
cytotoxic agents are substantial but can be avoided with
the careful dosing of these drugs, limitations of their use
to short courses, and appropriate precautions such as adequate hydration. Indeed, in the majority of studies in children and adults, minimal side effects of either chlorambucil
or cyclophosphamide have been reported and the drugs
have been well tolerated. One of the most disturbing side
effects of cyclophosphamide therapy, gonadal toxicity, appears to be dose related and reversible in the majority of
young patients treated with this medication.
Peritonitis is an important complication of the nephrotic
syndrome for those patients particularly children, who are
unable to achieve a remission of their proteinuria. Peritonitis occurs during periods of severe edema formation particularly when ascites is present. The most common infecting
agent is Streptococcus pneumonia. However, infections
with Escherichia coll and Hemophilus influenzae have also
been reported. Prior to antibiotic therapy, peritonitis was
the major cause of death in children and adults with MCD.
Reversible acute renal failure has also been reported in
patients with MCD. This complication appears to occur
much more frequently in adult patients than in children.
It is usually associated with a state of severe edema formation. A period of rapid weight accumulation immediately
precedes the onset of the renal failure. The pathophysiology of acute renal failure in patients with MCD is poorly
understood and cannot be clearly related to intravascular
volume depletion, acute tubular necrosis, or vascular obstruction such as renal vein thrombosis.
The development of chronic renal failure in patients
with MCD is rare in children and adults with a steroidresponsive clinical course. Patients at highest risk are those
who either do not have an initial response to steroid therapy
or those who become late nonsteroid responders (Fig. 2).
In both of these situations the possibility that the histopathologic lesion may be different or have evolved into a pattern
different than MCD must be considered and may be the
dominant factor in overall prognosis. For the majority of
children the relapsing nature of their disease will begin to
dissipate about 10 years after onset and the majority will
be free of proteinuria after puberty. However, late relapses
after long-term remissions of the nephrotic syndrome in
patients who have had their initial episode at a very young
age have been well documented. Similarly, adult patients
with MCD have a very good prognosis with 85 to 90%
survival rate 10 years or more after the onset of disease.
The major morbidity is related to complications of therapy.

19. Membranoproliferative Glomerulonephritis

153

Bibliography
Berns JS, Gaudio KM, Krassner LS, et al.: Steroid-responsive
nephrotic syndrome of childhood: A long-term study of clinical
course, histopathology, efficacy of cyclophosphamide therapy,
and effects on growth. Am J Kidney Dis 9:108-114, 1987.
Habib R, Kleinknecht C: The primary nephrotic syndrome of
childhood: Classification and clinicopathologic study of 406
cases. In: Sommers SC (ed) Pathology Annual. AppletonCentury-Crofts, New York, pp. 417-474, 1971.
International Study of Kidney Disease in Children: Nephrotic syndrome in children: Predictio n of histopathology from clinical
and laboratory characteristics at time of diagnosis. Kidney Int
13:159-165, 1996.

Nolasco F, Cameron JS, Heywood EF, et al.: Adult -onset minimal

change nephrotic syndrome: A long-term follow-up. Kidney Int
29:1215-1223, 1986.
Siegel NJ, Goldberg B, Krassner LS, Hayslett JP: Long-term
follow- up of children with steroid-responsive nephrotic syndrome. J Pediatr 81:251-258, 1972.
Smith JD, Hayslett JP: Reversible renal failure in the nephrotic
syndrome. Am J Kidney Dis 21:201-213, 1992.
White RHR, Glasgow EF, Mills RJ: Clinicopathologic study of
nephrotic syndrome in childhood. Lancet 1:1353-1359, 1970.
Korbet SM: Management of idiopathic nephrosis in adults, including steroid-resistant nephrosis. Curr Opin Nephrol Hypertens
4(2): 169-16, 1995.

MEMBRANOPROLIFERATIVE
GLOMERULONEPHRITIS
VIVETTE D'AGATI

Membranoproliferative glomerulonephritis (MPGN) is

a morphologic entity which is primarily defined at the light
microscopic level. The membranoproliferative pattern of
glomerular injury is characterized by mesangial proliferation and thickening of the glomerular capillary walls due
to peripheral capillary wall immune deposits, mesangial
interposition, and double contours ("tram-tracks") of the
glomerular basement membrane, each of which may be
present to varying degrees. Because MPGN manifests mes angial proliferation and extension of cells into the peripheral capillaries, thereby causing accentuation of the
glomerular lobularity, the terms "mesangiocapillary
glomerulonephritis" and "lobular glomerulonephritis"
have been used synonymously. An important concept in
approaching this entity is that MPGN is a nonspecific pattern of glomerular response to injury and as such includes
primary (or idiopathic) forms which are subdivided into
types I, II, and III, as well as a large number of secondary
forms listed in Table 1. These secondary conditions are
actually more common than the idiopathic forms and include various systemic autoimmune, infectious, and neoplastic disorders. Secondary causes of MPGN must be excluded by careful integration of the renal histologic,
immunofluorescence, and electron microscopic findings

Primer on Kidney Diseases, Sec ond Edition

with the clinical and laboratory features, before a diagnosis

of primary MPGN can be made. Whereas most cases of
MPGN occurring in children are idiopathic, secondary
forms tend to predominate in the adult population.
PATHOLOGY
Idiopathic MPGN

Idiopathic MPGN occurs in several morphologically dis tinct forms, each of which may correspond to a different
clinicopathologic disease entity. Type I is the most common. It consists of mesangial expansion owing to increased
mesangial cell number and matrix containing mesangial
immune deposits, sometimes imparting a nodular quality
to the mesangium Fig. 1. The peripheral capillary walls are
thickened by extensive circumferential mesangial interposition, subendothelial deposits, and double contours of the
glomerular basement membrane best delineated with the
Jones methenamine silver stain. The degree of glomerular
hypercellularity varies from mild to marked, and the severe
forms generally have highly exaggerated lobularity. Infiltrating neutrophils and monocytes may contribute to the
glomerular hypercellularity. Glomeruli tend to be affected
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

154

Vivette D'Agati
TABLE I

Membranoproliferative Glomerulonephritis
Primary or idiopathic

Type 1

Mesangiocapillary glomerulonephritis

Type II

Dense deposit disease

Type III

With mixed features of type I MPGN and membranous GN (Burkholder et al., 1970)
With complex intramembranous deposits (Strife et al., 1984; Anders et al., 1977)

Secondary (associated with other diseases)

With Immune deposits
Infections
Cryoglobulinemia (predominantly type II, also type III)
Hepatitis B, Hepatitis C, Endocarditis, Visceral abscesses, Infected ventriculoatrial shunts, Malaria,
Schistosomiasis, Mycoplasma, EBV, HIV
Autoimmune diseases
SLE, Rheumatoid art hritis, Sjogren's syndrome
Dysproteinemias
Light chain deposition disease, Cryoglobulinemia type I or II
Waldenstrom's macroglobulinemia, Immunotactoid and Fibrillary glomerulopathy
Without immune deposits
Chronic liver disease
Cirrhosis, ar antitrypsin deficiency
Thrombotic microangiopathies (subacute phase)
HUS/TTP, Anti-phospholipid antibody syndrome, Radiation nephritis, Sickle cell anemia, Transplant
glomerulopathy

FIGURE I MPGN type I showing lobular accentuation of the tuft

with global mesangial hypercellularity and sclerosis, with nodular
enlargement of the mesangium (arrow). There are frequent double
contours in association with subendothelial deposits (arrowhead)
(Jones methenamine silver, X300).

uniformly, in a global and diffuse manner, although mild

cases may have more focal and segmental lesions. By immunofluorescence, the immune deposits usually consist of immunoglobulins and complement components distributed in
a combined granular mesangial pattern and granular to
semilinear subendothelial pattern which outlines the lobular contours. Although IgG and C3 usually predominate,
weaker and more variable staining for IgM, IgA, and Clq
may also be found. A minority of cases will exhibit staining
for C3 only without codeposits of immunoglobulin. By electron microscopy, the mesangial areas are expanded by mes angial proliferation and mesangial immune deposits, with
increasing mesangial sclerosis and nodularity as the disease
progresses. Mesangial cells migrate into the peripheral capillaries by interposing their cell processes between the original glomerular basement membrane (GBM) and the
endothelium ("mesangial interposition" or "peripheral
mesangial extension"), often in association with subendothelial electron dense deposits Fig. 2. There is debate
whether monocytes or endothelial cells are also capable of
cellular interposition. In this peripheral location, irregular
neomembrane is produced internal to the original GBM,
producing a double contour or tram-track, which may be
partial or circumferential.
Type II MPGN is the least common form of idiopathic
MPGN, also known as "dense deposit disease." The degree
of mesangial hypercellularity is highly variable and less
uniform than in type I, and double contours of the GBM
tend to be less well developed Fig. 3. The defining feature
is a band-like thickening of the lamina densa of the GBM,
some tubular basement membranes, and the basal lamina of
Bowman's capsule by intramembranous deposits of PASpositive, silver-positive material. By electron microscopy,
these consist of distinctive highly electron dense deposits

155

19. Membranoproliferative Glomerulonephritis

FIGURE 2 MPGN type I. Electron micrograph showing narrowing of the capillary lumen (L) by subendothelial deposits, circum ferential extension of mesangial cells and subendothelial neomembrane, producing a double contour (X3000).

which replace and transform the lamina densa, producing

smooth, ribbon-like thickenings Fig. 4. More nodular mesangial electron dense deposits are also usually found. By
immunofluorescence there is typically positivity for C3 only
in a linear or double-contoured distribution along the glomerular basement membranes and as nodular "ring forms"
in the mesangium. More focal and often discontinuous linear C3 deposits are also found in Bowman's capsule and
some tubular basement membranes.
Type III MPGN is a relatively uncommon and controversial variant. Two subtypes have been defined. The first
displays combined features of MPGN type I and membranous glomerulonephritis with variable deposits in mesangial, subendothelial, and subepithelial locations. These usually consist of immunoglobulins (particularly IgG) and
complement (C3 and Cl), although up to half of cases
may have staining for C3 only. By electron microscopy the
glomerular capillary walls exhibit a complex thickening due
to electron dense deposits in mixed subendothelial and
subepithelial locations, often with associated tramtracks on
the inner aspect, and basement membrane spikes separating the subepithelial deposits on the outer aspect. Correspondingly, at the light microscopic level, silver stain highlights the combined GBM double contours and spikes,

FIGURE 3 MPGN type II. A severe example with exuberant

lobular accentuation of the tuft by mesangial proliferation. There
is band-like thickening of th e glomerular basement membranes
(arrows) (hematoxylin -eosin, X300).

giving a complex, woven, and multilayered appearance to

the glomerular capillary wall. The second variant is less
well accepted. It resembles a hybrid between type I and
type II MPGN, with combined subendothelial and extensive intramembranous deposits which cause highly irregular thickenings of the glomerular capillary wall, often extending from the subendothelial to the subepithelial aspect.
These intramembranous deposits lack the smooth, ribbonlike appearance of the dense deposits in type II disease. The
intramembranous deposits are silver negative and cause
severe disruption and fraying of the lamina densa. Correspondingly, by light microscopy, the glomerular basement
membranes appear markedly thickened and usually fail to
take the silver stain. Tubular basement membrane deposits
occur with less regularity than in MPGN type II.
Secondary MPGN

Although the many entities listed in Table 1 share membranoproliferative features at the light microscopic level,
they can usually be distinguished from the primary form
by integrating clinical and serologic data with the immunofluorescence and electron microscopic features. Some secondary forms of MPGN have immune complex deposits
similar to the primary forms, whereas others have mesangiocapillary proliferative features without immune deposits

156

Vivette D'Agati

frequently implicated in MPGN. In MPGN secondary to

cryoglobulinemia, large intracapillary glomerular deposits
("protein thrombi") are common. By electron microscopy,
glomerular electron dense deposits may be of the usual
granular type or exhibit an organized annular-tubular (25
to 35 nm) or fibrillar substructure (Fig. 5). In some patients
with cryoglobulinemic glomerulonephritis, the deposits
may be relatively scanty at both the immunofluorescence
and ultrastructural level. Vasculitis with arterial cryoglobulin deposits can also sometimes be demonstrated on renal biopsy.
PATHOGENESIS

FIGURE 4 MPGN type II. Electron micrograph showing smooth

ribbon-like dense deposits within the lamina densa of two adjacent
glomerular capillaries (X2000).

(as in chronic thrombotic microangiopathy). For example,

in light chain deposition disease, the deposits consist of
light chain only (usually kappa), without associated Ig
heavy chain or complement. In thrombotic microangiopathies, the subendothelial material found between the double contours of GBM consists of organizing products of
coagulation rather than immune deposits and thus stains
predominantly for fibrinogen. In fibrillary glomerulopathy,
the deposits consist of polyclonal IgG and complement and
have a characteristic ultrastructural appearance of randomly oriented fibrils 16 to 24 nm in diameter.
In cryoglobulinemia, the glomerular deposits reflect the
composition of the cryoprecipitate detected in the serum.
In type I cryoglobulinemia, these consist of a single monoclonal immunoglobulin component, (such as IgG kappa).
In the more common type II cryoglobulinemia, the deposits
usually stain for both IgG and IgM but with IgM dominance
since the cryoglobulins typically consist of a monoclonal
IgM kappa component complexed to polyclonal IgG. C3
is usually more intense than Clq. Type II cryoglobulinemia
is the form most frequently identified in "mixed essential"
cryoglobulinemia, of which most cases have now been
linked to chronic hepatitis C infection. Type III cryoglobulins consisting of two polyclonal immunoglobulins are common in autoimmune and infectious diseases but are less

Although the nature of the antigenic stimulus in most

forms of MPGN is unknown, the demonstration of circulating immune complexes (ICs) in over half of patients and
hypocomplementemia in nearly three quarters strongly
supports an 1C pathogenesis. A process of 1C deposition
in the mesangium and glomerular capillary wall followed
by complement activation, leukocyte chemotaxis, platelet
activation, and release of cytokines mitogenic for mesangial
cells is likely central to its mediation. Experimental models
of chronic serum sickness produced by repeated injection of
foreign antigen closely mimic MPGN, although the relative
contribution of passive deposition of preformed ICs and
in situ 1C formation following local Ag planting are in
dispute. In models of passive administration of preformed
ICs, complexes of intermediate size and variable charge
(cationic, neutral, or anionic) tend to deposit in the mesangium and subendothelial zones and could be augmented
by complement activation and binding of antiidiotypic antibodies or rheumatoid factors. Experimentally, cationic antigens have a particular affinity for the negatively charged
lamina rara interna and could act as a source of subendothelial Ag planting. Hepatitis B viral antigens (surface,
core, and e antigen), hepatitis C nucleocapsid/core protein
C22-3, and DNA-histones are examples of antigens which
have been identified in the circulating ICs or glomerular
deposits of patients with secondary forms of MPGN due
to infections and SLE. Whereas many cases of MPGN
type I follow an upper respiratory tract infection or other
infectious process, the inciting antigen in the idiopathic
forms remains unknown.
The pathogenesis of MPGN type II is likely to be quite
different from the schema of 1C deposition outlined above.
In dense deposit disease, no circulating ICs have been identified and glomerular deposits consist of C3 only, without
Ig. These patients have evidence of systemic dense deposits
affecting splenic sinusoids and Bruch's membrane of the
retina, and some have associated partial lipodystrophy.
Lipid extraction techniques applied to the ultrathin sections
for electron microscopy cause the dense deposits to lose
their electron dense appearance, suggesting that they may
consist of lipids rich in unsaturated fatty acids, although
their chemical composition and origin are unknown. It has
been postulated that factor D of the complement system, a
serine protease "adipsin" synthesized by adipocytes which
converts factor B of the preconvertase C3bB to its activated

19. Membranoproliferative Glomerulonephritis

157

FIGURE 5 Cryoglobulinemic glomerulonephritis. Electron micrograph showing subendothelial electron dense, deposits with organized substructure (X10,000). The inset shows a higher
magnification of the subendothelial deposits with hollow annular-tubular structures approximately 30 nm in diameter (X24,000).

form C3bBb, may provide a pathogenetic link between

partial lipodystrophy and MPGN type II.
CLINICAL FEATURES AND OUTCOME
Idiopathic MPGN

Complement activation and hypocomplementemia are

integral features of all three types of MPGN, and thus
the term "hypocomplementemic glomerulonephritis" was
used in the past. Although there is considerable overlap,
the patterns of complement activation tend to differ between the three forms. In type I, serum C3 and CH50 levels
are reduced in up to one half of patients. Concentrations
of Clq, C4, properdin, and factor B are borderline or reduced in less than half of cases. These findings are consis tent with the hypothesis of both classic complement pathway activation by ICs as well as alternative pathway
complement activation. Inherited complement deficiencies
(C2, C3, C6, C7, and C8) and reduced hepatic synthesis of
complement likely contribute to the hypocomplementemia
in some patients with MPGN, particularly types I and III.
In type II, there is persistent and often profound depression
of C3, with normal levels of early complement components
Clq and C4. This persistent marked C3 reduction is likely
mediated by C3 nephritic factor (Nfa), an autoantibody to
C3bBb, the C3 convertase of the alternative pathway,
which is demonstrable in more than 70% of patients with
MPGN type II. By binding to the C3 convertase, C3 nephritic factor stabilizes the convertase and prevents the
action of C3b inactivator, thereby maintaining the convert-

ase in a continually activated state. C3 nephritic factor is

not unique to MPGN type II but can be identified in about
a quarter of patients with type I disease and some autoimmune conditions such as SLE, cryoglobulinemia, and shunt
nephritis. In type III MPGN, C3 levels are reduced in up
to 50% of patients, C4 levels are usually normal, and C3
nephritic factor is absent. The presence of reduced C5, C6,
C7, and C9 in some patients with type III MPGN has
been linked to a separate terminal pathway nephritic factor
(NFt) which activates complement more distally in the
complement cascade.
In its primary form, MPGN is most common in older
children and young adults, (ages 7 to 30). It is equally
prevalent in both sexes, with white racial predominance.
The renal manifestations include nephrotic and nephritic
features, often present in combination. In two thirds of
patients with MPGN type I, features of nephrotic syndrome
dominate the clinical presentation. Other presenting features include asymptomatic hematuria with erythrocyte
casts, acute nephritic syndrome, subnephrotic proteinuria
with stable or slowly progressive renal insufficiency, and
recurrent episodes of gross hematuria. An upper respiratory tract infection including streptococcal pharyngitis commonly precedes first clinical recognition of renal disease.
The failure of the nephritis to resolve, with persistence of
proteinuria, hematuria, and hypocomplementemia beyond
6 weeks, are ominous clues that the disease is other than
acute poststreptococcal glomerulonephritis, which must
then be confirmed by renal biopsy.
MPGN type I tends to be a slowly progressive disease
and there are few spontaneous remissions. The reported

158

Vivette D'Agati

10-year renal survival is 50%. Median renal survival for

both children and adults ranges from 9 to 12 years. Treatment may lead to partial or complete clinical remission in
a subgroup of patients. Repeat renal biopsies performed
after a prolonged course of therapy have documented regression of the mesangiocapillary lesions but with frequent
persistence of mesangial hypercellularity.
Type II MPGN is much rarer than MPGN type I, constituting less than 5% of all cases of idiopathic MPGN seen
at Columbia Presbyterian Medical Center. Mean age of
onset is slightly younger than in type I and common presentations include acute nephritic syndrome or recurrent macroscopic hematuria. Median renal survival ranges from 5
to 12 years. The course of MPGN type II is generally more
aggressive than type I, although up to 20% of patients may
exhibit a relatively stable, more indolent course.
In all types of MPGN, features which portend a poor
outcome include greater percentage of sclerotic glomeruli,
numerous crescents and more severe tubular atrophy, and
interstitial fibrosis at biopsy. Negative clinical prognostic
features include elevated serum creatinine and hypertension at biopsy as well as a course of severe, unremitting
nephrotic syndrome. Outcome appears to be better in patients with asymptomatic hematuria and subnephrotic proteinuria who have more focal and milder membranoproliferative features on renal biopsy.
Secondary MPGN

A major cause of secondary MPGN is cryoglobulinemia. Cryoglobulins are immunoglobulins which share the
physical property of precipitation in the cold. Three
main types have been identified. Type I cryoglobulins consist of a single monoclonal immunoglobulin. This type occurs
in dysproteinemias such as B cell lymphoma, Waldenstrom's macroglobulinemia, and myeloma. Type II and
III cryoglobulins are mixed cryoglobulins containing two
immunoglobulins, often with rheumatoid factor activity.
Type II cryoglobulins consist of one monoclonal immunoglobulin (usually IgM) complexed to a polyclonal IgG.
Type III cryoglobulins consist of two polyclonal immunoglobulins (usually IgG and IgM). These mixed forms are
often found in autoimmune diseases, chronic infections,
chronic liver disease and, in the case of type II, some dysproteinemias. Recently, the majority of cases of mixed essential cryoglobulinemia have been linked to hepatitis C
virus (HCV) infection. In fact, IgG directed to HCV core
(C22-3) protein has been found to be enriched in the circulating cryoglobulins.
Clinical manifestations of cryoglobulinemia resemble
those of multisystem vasculitis. These include palpable purpura due to leukocytoclastic vasculitis, arthralgias and arthritis, distal necroses and skin ulcerations, Raynaud's phenomenon, peripheral neuropathy, abdominal pain, and
glomerulonephritis of the MPGN type, sometimes with
associated vasculitis of renal arterioles, small arteries
and veins.
A significant fraction of cases of apparently "idiopathic"
MPGN in adults may actually be related to HCV infection.

However, the incidence of HCV infection in patients with

MPGN has not been studied in any systematic way. The
increasing recognition of hepatitis C-associated MPGN in
adults has only become possible with the availability of
highly sensitive serum assays for antibody to HCV [including second generation enzyme-linked immunoassay (EIA
II) and recombinant immunoblot assay (RIBA) which test
for antibody to four HCV antigens] as well as polymerase
chain reaction (PCR) for viral genome in patient serum.
Acute hepatitis due to HCV is often subclinical and anicteric but leads to chronic active hepatitis in a significant
percentage of cases. The development of glomerulonephritis and cryoglobulinemia are late complications of hepatitis
C infection. Among patients with HCV-associated MPGN,
only one fifth have signs of liver disease but two thirds
have mild elevations of their liver transaminases. Primary
risk factors for HCV infection in this group include IVDA
in over half and blood transfusions in about one fifth. Other
cases may be caused by perinatal transmission from mother
to fetus. Cryoglobulinemia can be detected in about two
thirds of patients but less than half have extrarenal manifestations (such as purpura, arthralgias, neuropathy, and cardiomyopathy). Decreased CH50 is detectable in 90%, with
reduced C4 in 70%, reduced C3 in 50%, and positive rheumatoid factor in more them 70%. In some cases, diagnosis
of HCV-associated MPGN may precede by months or years
the clinical demonstration of cryoglobulinemia. About one
third of cases of HCV-associated MPGN never develop
evidence of cryoglobulinemia. The most common renal
presentation is proteinuria (which is nephrotic in about
70%) and mild azotemia. Hematuria and nephritic features
are present in about 25%. The most common pattern of
glomerulonephritis is MPGN type I (>80%), with fewer
cases of MPGN type III or acute exudative glomerulonephritis.
TREATMENT
Idiopathic MPGN

In neither children nor adults is the optimal therapy for

idiopathic MPGN clearly defined. Only limited data are
available for types II and III MPGN, and most of the
randomized clinical trials are flawed by short-term followup periods of only 1 to 4 years. Interpretation of the findings
is made difficult by the failure of earlier studies to identify
HCV as an associated risk factor.
The major treatment modalities employed in children
and adults with idiopathic MPGN include corticosteroids
and antiplatelet agents. In the pediatric population, early
studies showed stabilization of renal function in most children treated with alternate day steroids (2 mg/kg every
other day for 1 year, tapered gradually to 20 mg every
other day for up to 5 to 10 years). In a subsequent controlled
trial by the International Study of Kidney Disease in Children, alternate day prednisone administered for 3 to 4 years
led to stabilized or improved renal function in 61% of
patients, compared to 12% of controls. However, treatment
failures occurred in 40% of the prednisone-treated group

159

19. Membranoproliferative Glomerulonephritis

compared to 55% of those receiving placebo. Problems

related to steroid toxicity, such as hypertension and seizures, in children treated with this regimen have led to
trials of shorter courses of steroids coupled with aggressive
antihypertensive therapy. In one uncontrolled pediatric
study of 19 patients, prednisone 2 mg/kg daily was administered for a maximum of 8 weeks, followed by taper to
20 mg every other day for 2 to 3 years, in conjunction with
one or more antihypertensive medications. At 2 years following onset of treatment, there was significant improvement in GFR and proteinuria in most patients, with reversion to inactive urinary sediment.
A role for corticosteroid therapy in adults with idiopathic
MPGN is less clear. Convincing evidence of benefit has
been difficult to demonstrate in several retrospective studies. On the other hand, one study has demonstrated that
antiplatelet agents (aspirin 975 mg/day and dipyridamole
225 mg/day) administered for 1 year reduced the incidence
of renal failure at 3 to 5 years, but without a difference
in outcome at 10 years. These findings suggest that more
prolonged therapy may be required. A small controlled
trial of aspirin (500 mg/day) and dipyridamole (75 mg/day)
versus placebo, together with antihypertensives and protein
restriction for 3 years, showed greater reduction in proteinuria in those treated with aspirin and dipyridamole. Larger,
long-term studies are required to better define the role of
these agents in children and adults.
In sum, conformity of opinion on how to treat idiopathic
MPGN is lacking. Based on the available data, it seems
judicious to treat children with idiopathic MPGN who have
significant active disease and nephrotic syndrome with a
prolonged course of low-dose corticosteroids. A similar
regimen may be considered in adults with type I MPGN
in whom secondary forms (in particular, HCV infection)
have been excluded. Treatment has not been shown to be
efficacious for type II or type III MPGN in adults. The
role of antiplatelet agents as first-line therapy remains to
be defined in long-term studies. Nonetheless, antiplatelet
therapy appears a reasonable alternative to steroids in selected cases.
Compared to earlier 10-year renal survivals of approximately 50%, more recent studies of patients receiving various therapies for MPGN have reported improved survivals
of 60 to 85% for both children and adults. This apparently
improved outcome is likely due in part to bias in the study
design, since survival for the treated group was calculated
from onset of therapy whereas survivals in historical controls were calculated from time of clinical onset of disease.
Secondary MPGN
Choice of therapy in secondary forms of MPGN depends
on the underlying condition. For example, membranoproliferative variants of lupus nephritis will usually respond to
immunosuppressive regimens used to treat other examples
of class IV lupus with comparable activity. Partial or complete remission can be attained in MPGN secondary to
bacterial endocarditis or infected atrioventricular shunts
with use of appropriate antimicrobial therapy. In hepatitis

B-associated MPGN, the glomerulonephritis often resolves

spontaneously and immunosuppressive therapy may be
contraindicated because of the increased risk of activation
of the hepatitis.
In hepatitis C-associated MPGN with or without cryoglobulinemia, therapeutic strategies have included prednisone, pulse methylprednisolone, cyclophosphamide, plasmapheresis, and interferon-a. When 19 patients with
HCV-associated MPGN were treated for 6 months or more
with interferon-a, there was a significant reduction in pro teinuria (5.8 to 2.0 g/day), but no significant change in renal
function at the end of the study period. In a controlled
Italian study of 53 patients with HCV-associated cryoglobulinemia, of whom two thirds had clinical evidence of
renal disease, patients randomized to receive interferon
for 6 months had significant improvement in cutaneous
vasculitis, circulating cryoglobulin levels, and renal function
compared to those receiving conventional therapy. Unfortunately, in this and a number of other studies, relapses
of renal disease and cryoglobulinemia are common after
discontinuation of interferon therapy, suggesting that longterm interferon use may be more beneficial. Interferon-a's
efficacy correlates with its ability to reduce viral burden;
however, significant side effects including fever, flu-like
symptoms, and myalgias may limit its use in some patients.
A course of aggressive immunosuppressive therapy (pulse
methylprednisolone followed by oral steroids and cyclophosphamide) is usually reserved for patients with severe
acute vasculitic multisystem manifestations but runs the
risk of promoting HCV viral replication or lymphoma in
the long term. In patients with severe acute disease, plasmapheresis (two or three times weekly) for up to 4 weeks has
been used successfully to remove the circulating cryoglobulin, and can be administered in conjunction with immu nosuppressive therapy. Cryofiltration, which precipitates
out the cryoglobulins by cooling of the patient's plasma
and then reinfuses the patient's own rewarmed plasma, can
obviate the need for replacement fluid. Most centers now
advocate plasmapheresis combined with an immunosuppressive regimen that includes steroids and cyclophosphamide to control the severe acute systemic vasculitis,
whereas long-term therapy with interferon alone remains
the mainstay for the chronic phase of the disease.
TRANSPLANTATION

For those patients with idiopathic MPGN who progress

to end stage renal disease and receive a renal transplant,
recurrence of MPGN in the allograft is a frequent occurrence, particularly within the first 6 months to 1 year posttransplant. Approximately 30% of transplanted children
with MPGN type I develop recurrence, some in several
successive allografts. Up to 40% of these recurrences lead
to graft failure. In type II MPGN, recurrence is even more
common and may be as high as 80 to 100%, depending on
how carefully it is sought. However, recurrences tend to
be clinically mild and fewer than 10 to 20% of recurrences
are responsible for graft loss. The recurrence rate of MPGN
type III is not known. Recurrences must be differentiated

Gerald B. Appel

160
from transplant glomerulopathy, which also has membranoproliferative features but lacks immune deposits. Rare
cases of de novo MPGN related to hepatitis C infection
have also been reported in allografts.
Bibliography
Anders D, Agricola B, Sippel M, et al.: Basement membrane
changes in membranoproliferative glomerulonephritis. II. Characterization of a third type by silver impregnation of ultra thin
sections. Virchows Arch [Pathol Anat] 376: 1 -19, 1977.
Burkholder PM, Marchand A, Krueger RP: Mixed membranous
and proliferative glomerulonephritis. Lab Invest 23: 459-479,
1970.
Cameron JS, Turner DR, Heaton J, et al.: Idiopathic mesangio capillary glomerulonephritis. comparison of types I and II in
children and adults and long-term prognosis. Am J Med 74:
175-192, 1983.
D'Amico G, Fornasieri A: Cryoglobulinemic glomerulonephritis:
A membranoproliferative glomerulonephritis induced by hepatitis C virus. Am J Kidney Dis 25: 361-369, 1995.
Donadio JV, Anderson CF, Mitchell JC, et al.: Membranoproliferative glomerulonephritis. A prospective clinical trial of antiplatelet therapy. N Engl J Med 310: 1421-1426, 1984.
Donadio JV, Offord KP: Reassessment of treatment results in
membranoproliferative glomerulonephritis. Am J Kidney Dis
14: 445-451, 1989.
Ford DM, Briscoe DM, Shanley PF, Lum GM: Childhood membranoproliferative glomerulonephritis type I: Limited steroid therapy. Kidney Int 41: 1606-1612, 1992.
Habib R. Gubler MC, Loirat C, et al.: Dense deposit disease: A
variant of membranoproliferativ e glomerulonephritis. Kidney
Int 7: 204-215, 1975.

Habib R, Kleinknecht C, Gubler MC, et al.: Idiopathic membranoproliferative glomerulonephritis in children. Report of 105 cases.
Clin Nephrol 1: 194-214, 1973.
Holley KE, Donadio JV, Jr: Membranoproliferative glomerulonephritis. In: Tisher CC, Brenner BM (eds) Renal Pathology, with
Clinical and Functional Correlations. 2nd ed., Lippincott, Philadelphia, pp. 294-329, 1994.
Johnson RJ, Gretch DR, Couser WG, et al.: Hepatitis C virusassociated glomerulonephritis. Effect of alpha-interferon therapy. Kidney Int 46: 1700-1704, 1994.
Johnson RJ, Gretch DR, Yamabe H, et al.: Membranoproliferative
glomerulonephritis associated with hepatitis C virus infection.
N Engl J Med 328: 465-470, 1993.
Johnson RJ, Willso n R, Yamabe H, et al.: Renal manifestations
of hepatitis C virus infection. Kidney Int 46: 1255-1263, 1994.
McEnery PT, McAdams AJ, West CD: The effect of prednisone
in a high-dose, alternate-day regimen on the natural history of
idiopathic membranoproliferative glomerulonephritis. Medicine
64: 401-424, 1985.
Misiani R. Bellavita P, Fenili D, et al.: Interferon alfa - 2a therapy
in cryoglobulinemia associated with hepatitis C virus. N Engl J
Med 330: 751-756, 1994.
Strife CF, Jackson EC, McAdams AJ: Type III membranoproliferative glomerulonephritis: Long-term clinical and morphologic
evaluation. Clin Nephrol 21: 323-334, 1984.
Tarshish P, Bernstein J, Tobin JN, Edelmann CM: Treatment of
mesangiocapillary glomerulonephritis with alternate-day prednisoneA report of the International Study of Kidney Disease
in Children. Pediatr Nephrol 6: 123-130, 1992.
Varade WS, Forristal J, West CD: Patterns of complement activation in idiopathic membranoproliferative glomerulonephritis
types I, II and II. Am J Kidney Dis 16: 196-206, 1990.
West CD: Childhood membranoproliferative glomerulonephritis:
An approach to management. Kidney Int 29: 1077-1093, 1986.

FOCAL SEGMENTAL GLOMERULOSCLEROSIS

GERALD B. APPEL

Focal segmental glomerulosclerosis (FSGS) is a clinical

and histopathologic entity with a pattern of glomerular
injury which may occur in an idiopathic or secondary form
(see Table 1). The most common manifestation is proteinuria, which can range from minor amounts to nephrotic
levels. FSGS accounts for less than 15% of cases of idiopathic nephrotic syndrome in children, but is more frequent
in adults. Recent studies performed at several institutions

Primer on Kidney Diseases, Second Edition

have documented an increased incidence of FSGS in biopsies of adult patients that is especially noteworthy among
nephrotic patients. FSGS appears to be the most common pattern of idiopathic nephrotic syndrome among
African-Americans and may now be the most common
pattern in all races. Idiopathic FSGS frequently progresses
to end stage renal failure (ESRD). While potentially treatable, the appropriate type of therapy, duration of treat Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

20. Focal Segmental Glomerulosc lerosis

TABLE I

Etiologies of Focal Sclerosis

Primary idiopathic
Focal segmental glomerulosclerosis
Variants of minimal change disease
IgM nephropathy Diffuse
mesangial hypercellularity
Variants of FSGS
Glomerular tip lesion
Collapsing FSGS
Secondary

Unilateral renal agenesis

Renal ablationremnant kidney

Sickle cell disease

Morbid obesity (with or without sleep apnea)
Congenital cyanotic heart disease

Heroin nephropathy
HIV nephropathy

Aging kidney
Healed focal proliferative or necrotizing GN
Vesicoureteral reflux
FSGS, focal segmental glomerulosclerosis; GN,
glomerulonephritis.

ment, and response rate to treatment are all subjects of

controversy.
PATHOLOGY

There are no uniformly agreed on diagnostic criteria for

FSGS. The pathologic diagnosis of FSGS usually depends
on identifying in some glomeruli (focal lesions) areas of
glomerulosclerosis in parts of the glomerular tufts (segmental lesions) (Fig. 1). In addition, fusion or effacement of foot
processes is found to some extent in all of the glomeruli,

161
including those unaffected by areas of segmental sclerosis.
The term "idiopathic" FSGS may be used only if the lesions
of other types of focal glomerulonephritis that could heal
as focal sclerosing lesions are absent and if there is no
evidence of immune complex deposition by electron microscopy (EM). Focal areas of IgM and C3 deposition isolated to the areas of segmental sclerosis are thought to
result from entrapment of immunoglobulin and complement components rather than from true immune complex
deposition. The remainder of the glomerular tuft and the
glomeruli unaffected by glomerulosclerosis typically have
some degree of foot process effacement noted by EM, but
they do not have evidence of immune complex deposition
by IF. Although large amounts of proteinuria and uniform
foot process fusion may be present, biopsies taken early
in the course of FSGS, when renal function is still normal,
show few glomeruli with segmental sclerosing lesions and
almost no global sclerosis. At a later stage, as renal function
deteriorates, many glomeruli will show segmental or global
sclerosis. Some investigators believe the segmental sclerosing lesions are initially present in the juxtamedullary glomeruli and then spread outward in time to involve the
remaining renal cortex. Interstitial fibrosis is also a common
finding in biopsies with significant glomerulosclerosis. The
mechanism responsible for this damage is unclear but may
relate to changes in the postglomerular circulation, absorption of filtered proteins and lipoproteins across tubular
epithelia, or incitement of cytokine and growth factors by
abnormally filtered substances. The presence of increased
tubulointerstitial damage correlates with a poor renal prognosis.
Several variants of FSGS deserve comment. The glomerular tip lesion is characterized by swelling, vacuolation,
and proliferation of epithelial cells and, later, sclerosis and
hyalinosis in the segment of the glomerulus adjacent to
the origin of the proximal tubule. The remainder of the
glomerulus has changes similar by light microscopy (LM)
and EM to those seen in minimal change nephrotic syndrome. Another variant of FSGS is associated with focal
or global glomerular capillary collapse and sclerosis with
visceral epithelial cell swelling similar to that seen in HIV
glomerulopathy (see Fig. 1, Chapter 33). This so-called
"collapsing" or "malignant" variant of glomerulosclerosis
is more common in African-Americans, and has a distinctive and more ominous clinical course than other forms of
idiopathic FSGS. In addition, a number of lesions that
are often considered variants of minimal change nephrotic
syndrome may evolve into FSGS, including IgM nephropathy (a presentation of minimal change by LM but with IF
positivity for IgM and, in some patients, mesangial dense
deposits on EM), and diffuse mesangial hypercellularity
(mild proliferation of cells limited to the glomerular mesangium).
PATHOGENESIS

FIGURE I Glomerulus from a patient with focal segmental glomerulosclerosis showing perihilar sclerosis with adhesion to Bowman's capsule (periodic acid- Schiff stain). Courtesy of Dr. J. C.
Jennette.

By definition, the pathogenesis of idiopathic FSGS is

unknown. Some patients who appear on initial biopsy to
have minimal change disease evolve into FSGS, which can

162

Gerald B. Appel

be demonstrated on a subsequent biopsy. While the sample

of glomeruli observed on initial biopsy in some of these
patients may have missed segmental lesions present in only
a few juxtamedullary glomeruli, other patients with repeated relapses of the nephrotic syndrome after discontinuation of steroid therapy seem more convincingly to have
experienced evolution of the lesion. Moreover, all the glomeruli in classic FSGS have fusion of foot processes and
are responsible for the proteinuria. As in minimal change
nephrotic syndrome, the loss of the charge barrier of the
glomerular capillary wall may allow negatively charged
albumin to pass through the altered capillary wall into
Bowman's space. These alterations may be in response
to lymphokine production or other circulating humoral
mechanisms. Evidence for the latter includes studies describing a circulating permeability factor that promotes the
in vitro permeability of glomeruli to albumin. The presence
of this permeability factor, a 50 kDa protein that is not
an immunoglobulin, has been used to predict the rapid
development of proteinuria in the allograft of some FSGS
patients who reach ESRD and undergo transplantation.
Moreover, some patients with recurrent FSGS in the allograft respond to plasmapheresis or to use of a protein
absorption column with a reduction in proteinuria.
Proteinuria in FSGS is often less selective than in minimal
change disease (vide infra), implying leakage of larger macromolecules through "larger pores" in the glomerular basement membrane. Drug-induced (e.g., puromycin nephrosis,
adriamycin nephrosis) minimal change lesions in some animals can develop a picture of FSGS with nonselective pro teinuria. In one such model, the lifting off of the visceral
epithelial cells from the glomerular basement membrane
has been correlated with the nonselective proteinuria.
The pathogenesis of the sclerosing lesions and their pro gressive nature is also debated. Humans with idiopathic
FSGS or remnant kidney FSGS initially have a high glomerular filtration rate (GFR) and evidence of hyperfiltration,
suggesting that hyperfiltration and increased intracapillary
glomerular pressure may be mediators of FSGS. Likewise,
patients with glomerulomegaly, as seen in remnant kidneys
or due to obesity or hypoxemia in sleep apnea, also have
a high incidence of the nephrotic syndrome and FSGS.
FSGS without increased glomerular capillary pressure or
glomerulomegaly may relate to hyperlipidemia or intraglomerular coagulation.
It is possible and even likely that the pattern of FSGS
seen on biopsy represents a common pathway for a number
of distinct entities with different pathogenetic mechanisms
and clinical courses.
CLINICAL FEATURES

Most patients with idiopathic FSGS present with either

asymptomatic proteinuria or the full nephrotic syndrome.
The 10 to 30% of cases with asymptomatic proteinuria
are typically children whose proteinuria is detected during
routine pediatric checkups or camp or sports physicals; in
adults, detection of asymptomatic cases occurs most often
at military induction examinations, routine gynecologic or

obstetric check-ups, and insurance or employment physicals. Patients with the nephrotic syndrome present with
edema.
Hypertension is found in 30 to 50% of children and adults
with FSGS. Microscopic hematuria is found in 25 to 75%
of these patients: likewise, a decreased GFR is noted at
presentation in 20 to 30% of these patients. Daily urinary
protein excretion ranges from less than 1 g to 20 to 30 g/
day. Proteinuria is typically nonselective, i.e., it contains
higher molecular weight proteins as well as albumin. Nevertheless, albumin still comprises the largest component of
the urine protein. Complement levels and other serologic
tests are normal. Occasional patients will have glycosuria,
aminoaciduria, phosphaturia, or a concentrating defect indicating tubular damage as well as glomerular injury.
DIAGNOSIS

The diagnosis of FSGS requires a renal biopsy. Early

on, only a minority of the glomeruli show the segmental
sclerosing lesion. Even these lesions may have a predilection for the juxtamedullary region of the kidney, so the
renal biopsy may look identical to that of minimal change
nephrotic syndrome. This is especially likely when a very
superficial biopsy contains only a small number of glomeruli. Likewise, a small sample of glomeruli in a biopsy in an
older adult may show some glomeruli identical to minimal
change disease and one or two globally sclerotic glomeruli.
They may be the result of FSGS or merely the obsolescent
glomeruli that are found in the kidneys of older individuals
without renal disease. The finding of tubulointerstitial damage in such biopsies should suggest the possibility of unobserved scarred glomeruli and FSGS, but in neither situation
can the diagnosis of FSGS be firmly established. Clinically,
patients who are thought to have a minimal change pattern
nephrotic syndrome with a poor response to corticosteroids
or other immunosuppressive agents are likely to have
FSGS. The biopsy may also provide clues to a secondary
form of FSGS. In heroin nephropathy there is often more
severe tubulointerstitial disease than in idiopathic FSGS.
In HIV nephropathy there is often a collapsing variant
of glomerulosclerosis with global rather than segmental
involvement, and tubuloreticular inclusions are commonly
found on EM. In patients with remnant kidneys and
hyperfiltration-induced FSGS there is often less effacement
of the foot processes than in idiopathic FSGS.
THERAPY AND OUTCOME

Although variable in the individual patient, the course

of untreated FSGS is usually one of progressive proteinuria
and declining GFR. Patients with asymptomatic proteinuria typically develop the nephrotic syndrome over time.
Only a small minority of patients experience a spontaneous
remission of proteinuria or the nephrotic syndrome. Both
children and adults usually develop ESRD 5 to 20 years
from presentation. Some patients with the collapsing variant (or so-called malignant FSGS) will have a more rapid
course to ESRD in 2 to 3 years. The features which have

20. Focal Segmental Glomerulosderosis

been associated with a more rapid progression to renal

failure in idiopathic FSGS include higher grade proteinuria
(> 10-15 g day), a higher serum creatinine at time of
biopsy, a greater degree of glomerulosclerosis on biopsy,
and more tubulointerstitial damage on renal biopsy.
Idiopathic FSGS may recur in the transplanted kidney
with severe proteinuria and the nephrotic syndrome. Patients who present with more severe degrees of proteinuria
and a more rapid course to renal failure are at greater risk
for recurrence in the allograft.
The therapy of FSGS is controversial. There are few
randomized, controlled trials upon which to base decisions.
In general, those patients with a sustained remission of
their nephrotic syndrome are unlikely to progress to ESRD,
whereas those with unremitting nephrotic syndrome are
likely to have progression. In most studies before 1985,
only 10 to 30% of patients appeared to respond to a course
of corticosteroids with a remission of proteinuria. Moreover, only a low response rate to other immunosuppressive
agents such as azathioprine, cyclophosphamide, and chlorambucil was recorded, and the relapse rate after treatment
was also high. As a result most American nephrologists
considered FSGS to be unresponsive to therapy and did
not advocate immunosuppressive treatment.
A seminal Canadian study in 1987 noted that 44% of
children with FSGS responded with a remission of proteinuria. While the response rate for treated adults was similar
(39%), most adults did not receive any specific therapy.
Recent studies of the use of immunosuppressives in FSGS
have confirmed initial response rates of 25 to 60%. A collaborative Italian study using much longer courses of prednisone, cyclophosphamide, and/or azathioprine found 60%
of FSGS patients to have a complete remission of the nephrotic syndrome. Those patients with a complete remis sion had an excellent long-term renal survival, without the
occurrence of ESRD. An uncontrolled trial in children
using combined pulse steroids and long-term immunosuppression with corticosteroids and cytotoxics has also found
a 60% complete remission rate and a 16% partial remission
rate of the nephrotic syndrome and a low rate of progres sion to renal failure. A recent American study of more
than 50 nephrotic adults with nephrotic syndrome due to
FSGS also showed a better than 50% response rate as well
as long-term improvement in renal survival in the treated
group. Finally a number of trials have used low dose
cyclosporine ( 4-6 mg kg"1 day"1, for 2-6 months )
to treat steroid-resistant FSGS patients. Three controlled
trials have had complete and partial remission rates of 60
to > 90% with use of cyclosporine vs 17 to 33% in the
placebo group. The largest of these studies, The North
American Collaborative Study of Cylosporine in Nephrotic
Syndrome, found 12% complete remission and > 70 %
complete or partial remission with use of cyclosporine in
steroid-resistant FSGS. Even some FSGS patients who
have been unresponsive to cytotoxic agents may respond to
this therapy. Unfortunately, some patients will experience
renal damage from the cyclosporine.
At present the ideal regimen to treat idiopathic FSGS
is unknown. Many clinicians would not treat patients with

163
subnephrotic levels of proteinuria and little damage on
their renal biopsies. Others would treat these patients with
angiotensin converting enzyme (ACE) inhibitors to reduce
proteinuria and its side effects but would not use immunosuppressives to treat the specific glomerular lesion. Patients
at increased risk of renal failure such as those with nephrotic range proteinuria, elevated serum creatinines, and
interstitial scarring on biopsy may be treated with a prolonged course (6-9 months) of daily or every other day
corticosteroids (starting with 60 mg of prednisone daily or
120 mg every other day and tapering to lower doses after
several months), or other immunosuppressive medication
in the hope of inducing a remission of the nephrotic syndrome and preventing eventual ESRD.
For patients with secondary forms of FSGS, treatment
of the primary etiology, although rarely possible, is the
first step in management. Patients with FSGS secondary
to obesity and heroin nephropathy have had remissions of
proteinuria after weight reduction or cessation of drug use,
respectively. Use of ACE inhibitors and other nonspecific
methods to reduce proteinuria and the manifestations of
the nephrotic syndrome may also be beneficial. Immunosuppressive medications have not proved effective in any
form of secondary FSGS. In those patients with either
primary idiopathic or secondary forms of the FSGS who
remain nephrotic, control of fluid retention and edema can
be managed with salt restriction and diuretics. In addition,
attention should be given to hypertension control with antihypertensive medication, hyperlipidemia control with diet
and antihyperlipidemic medications, and perhaps to prevent hyperfiltration with low protein diets.
Bibliography
Arturo M, Biava C, Amend W, Tomlanovich S, Vincenti F: Recur rent focal glomerulosclerosis: Natural history and response to
treatment. Am J Med 92:375, 1992.
Banfi G, Moriggi M, Sabadini E, Fellin G, D'Amico G, Ponticelli
C: The impact of prolonged immunosuppression on the outcome
of idiopathic focal-segmental glomerulosclerosis with nephrotic
syndrome in adults. Clin Nephrol 36: 53-59, 1991.
Barisoni L, Valeri A, Radhakrishnan J, Nash M, Appel GB,
D'Agati V: FSGS: A 20 years epidemiologic study. J Am Soc
Nephrol 5:347, 1994.
Cattran D, Greenwood C, Bernstein K, et al.: Results of a 6 month
randomized trial of cyclosporine vs. placebo in adults with FSGS.
J Am Soc Nephrol 6:413, 1995.
D'Agati V: Nephrology forum. The many masks of FSGS. Kidney
Int 46:1223-1241, 1994.
Dantal J, Bigot E, Bogers W, et al.: Effect of plasma protein
absorption on protein excretion in kidney transplant recipients
with recurrent nephrotic syndrome. N Engl, J Med 330: 7-14,
1994.
Detwiler RK, Falk RJ, Hogan SL, Jennette JC: Collapsing glomerulopathy: A clinically and pathologically distinct variant of
FSGS. Kidney Int 45: 1416-1424, 1994.
Fogo A: Internephron heterogeneity of growth factors and sclerosis. Kidney Int 45:S24-s26, 1994.
Haas M, Spargo BH, Coventry S: Increasing incidence of FSGS
among adult nephropathies: A 20-year renal biopsy study. Am
J Kidney Dis 26: 740-750, 1995.

164

Sharon G. Adler and Cynthia C. Nast

Jennette JC, Marquis A, Falk RJ, Bodick N: Glomerulomegaly

in focal segmental glomerulosclerosis but not in minimal change
glomerulopathy. Lab Invest 62: 48A, 1990.
Korbet SM, Schwartz M, Lewis EJ: Primary FSGS: Clinical course,
and response to therapy. Am J Kidney Dis 23:773-783, 1994.
Lieberman KV, Tejani A, for the NY-NJ Pediatric Nephrology
Study Group: A randomized double-blind placebo -controlled
trial of cyclosporine in steroid resistant FSGS in children. J Am
Soc Nephrol 7: 56-63, 1996.
Pei Y, Cattran D, Delmore T, Katz A, Lang A, Ranee P: Evidence
suggesting under-treatment of adults with idiopathic focal segmental glomerulosclerosis. Am J Med 82: 938-944, 1987.
Ponticelli C, Rizzoni G, Edefonti A, Alt et al.: A randomized trial
of cyclosporine in steroid-resistant idiopathic nephrotic syndrome. Kidney Int 43:1377-1384, 1993.

Rennke H, Klein PS: Pathogenesis and significance of non-primary

focal and segmental glomerulosclerosis. Am J Kidney Dis 13:
443-455, 1989.
Rydell JJ, Korbet SM, Borok RZ, Schwartz MM: FSGS in adults:
Presentation, course, and response to treatment. Am J Kidney
Dis 25:534-542, 1995.
Savin VJ, Artero M, Sharma R, et al.: Circulating factor associated
with increased glomerular permeability to albumin in recurrent
focal segmental sclerosis. N Engl J Med 334:878-882, 1996.
Tune BM, Kirpekar R, Sibley RK, Reznik VM, Griswold WR,
Mendoza SA: Intravenous methylprednisolone and alkylating
agent therapy of prednisone-resistant pediatric FSGS: A longterm follow-up. Clin Nephrol 43: 84-88, 1995.
Valeri A, Barisoni L, Appel GB, Seigle R, D'Agati V:Idiopathic
collapsing FSGS: A clinicopathologic study. Kidney Int 50:17341746, 1996.

MEMBRANOUS NEPHROPATHY
SHARON G. ADLER AND CYNTHIA C. NAST

Membranous nephropathy is a histologic diagnosis based

on the presence of subepithelial deposits along an often
thickened glomerular basement membrane. In the nephropathology literature, the terms epi-, peri-, or extramembranous have occasionally been substituted for membranous,
and the term glomerulonephritis has been used interchangeably with glomerulopathy and nephropathy. Membranous nephropathy is the histological diagnosis in 15 to
25% of adult patients with no associated systemic illness
who undergo renal biopsies for heavy proteinuria. In this
setting, membranous nephropathy is referred to as "idiopathic." However, in up to 30% of patients with renal
biopsy-proven membranous nephropathy, medications or
associated systemic illnesses appear to play a pathogenetic
role in the initiation of the glomerular disorder. Table 1
summarizes those entities reported to be associated with socalled "secondary" membranous nephropathy. Idiopathic
and secondary membranous nephropathy are often distinguishable by clinical, laboratory, or histological features,
and therapeutic strategies for treatment tend to differ.
However, it should be noted that in routine clinical practice,
it often proves difficult or impossible to rigorously exclude
underlying causes. In fact, on occasion membranous nephropathy may precede by many months or even years the

Primer on Kidney Diseases, Second Edition

clinical presentation of an associated systemic illness. While

a relatively common cause of nephrotic syndrome in adults,
membranous glomerulopathy in children occurs less frequently than minimal change disease.
CLINICAL FEATURES

The clinical presentation is characterized by heavy proteinuria in more than 80% of patie nts, most often in association with the full expression of the nephrotic syndrome.
The remaining patients have asymptomatic proteinuria. A
minority have accompanying microscopic hematuria. The
onset of symptomatology tends to be insidious. Hypertension and azotemia are usually not features at initial presentation, although hypertension may be present in a minority
of patients at presentation, and certainly develops as progressive renal insufficiency ensues. Males tend to be affected more often and more severely than females. Although idiopathic membranous nephropathy can occur at
any age, 80 to 90% of patients are older than 30 years at
diagnosis. The older the patient, the greater the likelihood
that there is associated malignancy. This may be present
in up to 20% of patients over the age of 60.
Manifestations of the nephrotic syndrome may be quite
severe in patients with membranous nephropathy, includCopyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

165

21. Membranous Nephropathy

TABLE I

Clinical Entities Associated with "Secondary"

Membranous Nephropathy
Immunologic DisordersSystemic lupus erythematosus, mixed
connective tissue disease, rheumatoid arthritis, Sjogren's
syndrome, dermatomyositis, sarcoidosis, Hashimoto's
thyroiditis, dermatitis herpetiformis, myasthenia gravis,
Gullain- Barre syndrome, Weber- Christian panniculitis,
bullous pemphigoid, anticardiolipin antibody syndrome.
NeoplasmsCarcinoma (lung, colon, breast, stomach, kidney,
esophagus, carotid body), melanoma, leukemia, lymphoma
(especially non-Hodgkin's).
InfectionsHepatitis B, secondary or congenital syphilis,
quartan malaria, leprosy, schistosomiasis, filariasis, scabies.
MedicationsPenicillamine, organic gold, mercury-containing
compounds, captopril, probenicid, trimethadione, nonsteroidal
anti-inflammatory drugs.
Miscellaneous Sickle cell anemia, Fanconi's syndrome,
sclerosing cholangitis, systemic mastocytosis, volatile
hydrocarbon exposure, Kimura's disease, Gardner-Diamond
syndrome, diabetes, de novo in renal allografts.

ing anorexia, malaise, edema, and occasionally anasarca.

Ascites, pericardial, and pleural effusions are less common
in adults, even with severe nephrotic syndrome, than in
children. Hyperlipidemia is common. It is characterized by
increases in total and LDL cholesterol, Apo B, C-II, CIII, triglycerides, VLDL, and sometimes Lp(a), and by
abnormalities in the distribution of HDL cholesterol, particularly an increase in HDL fraction 3 and a decrease in
HDL fraction 2. In the general population these changes
are associated with an increase in atherogenesis and cardiovascular disease. By inference, it is presumed that these
same risks are also conferred on patients with nephrotic
syndrome with these changes, but this has not been studied
longitudinally in this population. Growth retardation in
children with severe nephrotic syndrome is a difficult clinical problem. Predisposition to arterial and venous thromboses have been noted. The incidence of renal vein thrombosis is high. The precise frequency is difficult to ascertain
since its determination would require venography, an invasive measure that is rarely undertaken unless symptomatically indicated. Similarly, pulmonary embolus has been
reported, rarely as the presenting manifestation of the nephrotic syndrome.
PATHOLOGY Primary
(Idiopathic) Membranous Nephropathy

On light microscopy, glomeruli have mesangial regions

of normal width which do not display hypercellularity. In
very early membranous nephropathy, the glomeruli may
appear completely normal. With more well-developed lesions, capillary walls are thickened with patent capillary
lumina, occasionally with a rounded configuration. Subepithelial fuchsinophilic deposits are observed with Masson's
trichrome staining. Methenamine silver stain discloses sub-

epithelial projections ("spikes") along the capillary walls,

corresponding to the deposition of new basement membrane material adjacent to the deposits (Fig. 1). When
capillary walls are tangentially sectioned, the deposits and
"spikes" may appear as holes within the basement membranes. In advanced membranous glomerulopathy the deposits are often within the capillary basement membrane,
having been completely covered by the new basement
membrane material, and capillary walls may display "double contours." Crescent formation is uncommon in membranous nephropathy. When present, crescents may be as sociated with late stages of the disease or, rarely, the
formation of antiglomerular basement membrane antibodies.
Immunofluorescence microscopy always demonstrates
strong granular capillary wall staining for IgG, with C3
and both light chains usually present in the same pattern.
Weaker IgA and IgM staining are infrequently observed.
The ultrastructural features of membranous nephropathy have been used to classify the disease into four stages.
Stage I is associated with normal light microscopy; by electron microscopy there are few to many small or mediumsize electron-dense deposits along the capillary walls in a
subepithelial position, with effacement of the overlying
visceral epithelial foot processes. Stage II is characterized
by more numerous and larger deposits with the formation
of subepithelial projections of basement membrane material lateral to the deposits ("spikes"). In stage III the new
extracellular material surrounds the deposits which become
incorporated into the capillary basement membrane. Stage
IV has markedly thickened capillary walls with rarefaction
and lucent areas in the intramembranous deposits and layering of the basement membrane. In all stages, resolution
of the disease is associated with fading of the deposits as
they undergo resorption, and fresh homogeneous deposits
are observed when the disease is active. Mesangial deposits
are very rarely encountered in primary membranous nephropathy.
Secondary Membranous Nephropathy

The capillary wall changes are similar to those of the

primary disease, although there may be fewer deposits.
Mesangial regions are frequently expanded and hypercellular, most strikingly in lupus nephritis; mesangial deposits
are common in secondary forms, particularly those related
to infections. Subendothelial deposits may occur infrequently. In membranous lupus nephritis, glomerular staining for all immunoglobulins and Clq is common, as are
Subendothelial and extraglomerular deposits and tubuloreticular structures in endothelial cell cytoplasm. Recurrent
and de novo membranous nephropathy occurring in renal
transplants in the absence of other glomerular abnormalities usually appears histologically similar to primary membranous glomerulopathy.
PATHOGENESIS

Membranous nephropathy is an immune-mediated dis ease in which immune complexes localize in the subepithe-

166

Sharon G. Adler and Cynthia C. Nast

ii*g,.M

SSifeit'^t,.w
FIGURE I (A) Glomerulus from a patient with primary membranous nephropathy. Capillary walls are slightly thickened
without mesangial abnormalities [silver stain (Jones), X230]. (B) Glomerulus from a patient with membranous lupus
glomerulonephritis. Note the mesangial hypercellularity in addition to the thickened capillary walls [silver stain (Jones),
X230]. (C) Glomerular capillary walls showing subepithelial projections (spikes) of silver-positive basement membrane
material (arrows) [silver stain (Jones, X700]. (D) Glomerulus showing granular capillary wall staining by immunofluorescence using anti-IgG antibody in primary membranous nephropathy (X480).

Hal aspect of glomerular capillary walls. Antigen-antibody

complexes could develop at this location by the production
of immune complexes in situ (the most likely mechanism)
or by the deposition of circulating immune complexes.
Immune complexes forming in situ at the subepithelial
capillary wall site can involve the binding of a circulating
antibody to an intrinsic glomerular antigen. In one experimental form of membranous nephropathy (Heymann's nephritis), the intrinsic antigen is a glycoprotein synthesized
by glomerular visceral epithelial cells. In situ complex formation can also be initiated by the deposition of an extrinsic
"planted" antigen. Size and charge are critical factors in
the localization of "planted antigens" to the subepithelial
aspects of capillary walls with subsequent immunoglobulin
deposition. Cationic proteins have been "planted" by intravenous injection to induce membranous nephropathy experimentally. Antigens may also become trapped in the
capillary wall by direct binding, binding related to biochemical factors, or autoimmune mechanisms. Low-avidity antibody predisposes to subepithelial deposit formation. Finally, circulating immune complexes with a suitable
cationic charge and of the appropriate size may traverse the
capillary basement membrane, lodging in the subepithelial

space and binding to anionic sites in the capillary wall.

However circulating complexes are not uniformly found in
patients with membranous nephropathy and currently are
not thought to account for the majority of cases of this dis sorder.
The antigen(s) involved in primary human membranous
nephropathy are not known. In secondary membranous
nephropathy, many of the antigens have not been dis cerned; however, hepatitis B surface and e antigens have
been identified in immune deposits, as have thyroid antigens in occasional patients with thyroiditis. In both primary
and secondary forms, complement is activated at the capillary wall site and appears to be related to the development
of proteinuria, but it is not required for the development of
the morphologic lesion. The finding of urinary complement
membrane attack complex (C5b-C9) has been suggested
as a diagnostic test for following disease activity. It may
be more useful when the glomerular antigen is intrinsic
than when it is "planted."
DIAGNOSIS

Idiopathic membranous nephropathy is a diagnosis

based on histological findings in a patient whose history,

21. Membranous Nephropathy

167

physical examination, and appropriate laboratory testing

exclude the likelihood of a secondary cause. Approximately
80% of patients have poorly selective proteinuria. The majority of patients have proteinuria exceeding 3.5 g per day.
A full-blown nephrotic syndrome is common, including
hypoalbuminemia, hyperlipidemia, and lipiduria. The urinary sediment is usually inactive and characterized by
heavy proteinuria. Lipiduria and oval fat bodies may be
present when the nephrotic syndrome occurs. Microhematuria occurs in less than a quarter of patients. Serum complement levels are usually normal. Low serum complement
levels suggest a secondary form of membranous nephropathy or an alternative diagnosis. Serum protein electrophoresis commonly shows broad band increments in /3 and y
globulins as a nonspecific feature of the nephrotic syndrome. The following laboratory tests are usually negative
or normal in patients with idiopathic membranous glomerulopathy, and abnormalities should suggest a secondary
form of the disorder: antinuclear antibodies, anti-DNA
antibodies, rheumatoid factor, glycosylated hemoglobin,
hepatitis antigen or antibody, thyroid antibodies, carcinoembryonic antigen, and cryoglobulins. Circulating immune complexes, when sought, have been found in idiopathic membranous nephropathy by some but not all
investigators and is not diagnostically helpful. Increased
risk for the development of idiopathic membranous glomerulopathy appears to be associated with the presence of
the histocompatibility antigens DR3 in Caucasians and Chinese and DR2 in Japanese.
COURSE

The course of untreated idiopathic membranous nephropathy is highly variable. Among patients initially pres enting with the nephrotic syndrome and a normal serum
creatinine level, the probability of a spontaneous complete
remission and a stable glomerular filtration rate (GFR) for
up to 20 years is approximately 30%. The probability of a
spontaneous partial remission with stable GFR is approximately 25%. An additional 20-25% of patients experience
persistent nephrotic syndrome with stable or very slowly
progressive loss of GFR. Twenty to twenty-five percent of
patients progress to end-stage renal disease over a 20- to
30-year period of follow-up. On initial presentation, it is
difficult to predict the likely outcome of an individual patient. However, the following factors have been suggested
as portending risk for a poor outcome: significant tubulointerstitial fibrosis and elevated serum creatinine at the time
of initial renal biopsy, male sex, older age at onset, heavy
proteinuria (e.g., >10 g/day), hypertension, and stage IV
lesions. The outcome in patients with secondary membranous nephropathy is affected by the associated underlying
disorders. Complete remissions are reported in patients
with treated infections, occasionally with treated neoplasms, and often in association with discontinuation of
offending drugs. Complete or partial remissions may also
be achieved with immunosuppressive therapy for patients
with membranous nephropathy secondary to immunologic disorders.

THERAPY

Controversy and uncertainty continue to envelop the

decision-making process with regard to the treatment of
idiopathic membranous nephropathy. These difficulties
emerge largely as a result of the clinical heterogeneity of
these patients. Given the overall good prognosis for most,
the lack of uniform benefit from treatment regimens involv ing steroids and/or cytotoxics, and the inherent risks involved with the employment of these agents, many have
advocated palliative therapy for the patient with normal
renal function and few adverse symptoms. This might begin
with sodium restriction and diuretics, taking care not to
induce a state of marked prerenal azotemia.
A number of strategies have been advocated to diminish
heavy proteinuria. Nonsteroidal anti-inflammatory drugs
(NSAIDS) have been reported to decrease proteinuria in
some patients with the nephrotic syndrome. An accompanying decrement in GFR is almost always seen, though the
decrease in proteinuria is often greater. Indomethacin and
meclofenamate have been the drugs most often used for
this purpose. Despite their utility in occasional patients
with heavy proteinuria, for the majority of patients with
nephrotic syndrome, the overall impact of NSAIDS in substantially diminishing proteinuria appears disappointing.
Dietary protein restriction in the range of 0.6-0.8 g kg~J
day" 1 has been suggested as a further means of diminishing
proteinuria, although longterm studies regarding nutritional safety are still necessary in patients with heavy proteinuria.
Small, prospective clinical trials performed in patients
with membranous nephropathy have demonstrated reductions in proteinuria averaging approximately 50% in patients treated with ACE inhibitors. Furthermore, a randomized prospective trial involving 583 patients with
chronic renal insufficiency of varying etiologies (192 of
whom had glomerular disease), demonstrated that the use
of an ACE inhibitor (in this case, benazepril) slowed the
progression of chronic renal insufficiency. Over the 3-year
follow-up period, the risk reduction for a doubling of serum
creatinine was 53%. ACE inhibitors offer the obvious collateral benefit of blood pressure control in these patients,
and a recent analysis of data from the Modification of
Renal Disease Study suggests that in patients with >1 g
of proteinuria, a mean arterial pressure of <92 mm Hg
slows the rate of renal functional decline. Thus, ACE inhib itors offer benefits for those patients with membranous
nephropathy for whom immunosuppressive therapy is
not indicated.
The hyperlipidemia of nephrotic syndrome may contribute to accelerated atherogenesis. Although dietary fat restriction is reasonable, it rarely normalizes the hyperlipidemia of nephrosis. Use of the newer HMG-CoA reductase
inhibitors (3-hydroxy -3-methylglutamyl-coenzyme A) or
agents such as gemifibrizol often lower serum LDL choles terol and triglyceride levels substantially in nephrotics.
However, currently it is unclear whether this will have longterm benefit on atherogenesis. For the patient experiencing
hypercoagulability with venous, or less commonly arterial,

168
thrombosis or pulmonary emboli, long-term anticoagulation for the duration of the nephrotic syndrome is recommended.
Patients at higher risk for progressive disease, patients
who actually have progressive disease, and those with particularly severe nephrotic complications are candidates for
immunosuppressive therapy. Historically, corticosteroid
therapy used alone without the addition of cytotoxic drugs
had long been advocated as the mainstay of treatment for
idiopathic membranous nephropathy. Three large prospective randomized clinical trials examined the usefulness of
steroids in achieving remissions in proteinuria and preventing progressive loss of kidney function. The earliest of these
three studies suggested that remissions of proteinuria were
more common and progressive loss of renal function was
less frequent in steroid-treated than in control patients.
However, later studies failed to confirm these findings.
Thus, currently, most available data do not suggest benefit
from the use of corticosteroids alone in the therapy of most
patients with idiopathic membranous nephropathy.
Additional data suggest that treatment for idiopathic
membranous glomerulopathy may be improved by combining corticosteroids with cytotoxic agents. In the largest series of such patients with the longest follow-up, control
patients received no specific therapy. During a 6-month
course, treated patients were given intravenous methylprednisolone followed by daily oral prednisone during
months 1, 3, and 5 and oral chlorambucil during months
2, 4, and 6. After a median follow-up of 5 years, the treated
group experienced more partial and complete remissions
of nephrotic syndrome and had lower serum creatinines
than did the untreated controls. However, no patient with
a serum creatinine >1.7 mg/dL at entry participated in this
trial. Additional analyses showed that while the treatment
regimen increased the probability of remission compared
to controls, this probability was somewhat reduced by the
presence of a serum creatinine >1 mg/dL and proteinuria
>5 g/day. Thus, it has been argued that the data cannot be
applied to many of the patients most needy of an aggressive
therapeutic approach. To determine whether the chlorambucil contributed significantly to the beneficial effects of
the regimen, an additional study was performed comparing
combination treatment with steroids alone. While more
patients given the steroids plus chlorambucil were in remis sion in the first 3 years of follow-up than those given steroids alone, by Year 4 the differences were no longer statis tically significant. Thus, the apparent dilemma in the use
of combined prednisone and chlorambucil is twofold. First,
in order to maximize response rate, patients need to be
treated early, thereby exposing to risk patients who might
otherwise undergo spontaneous remissions. Second, given
the relapses that may occur in the treated group and the
late spontaneous remissions that may occur in the untreated
group, the added benefit of cytotoxics over the long term
are still not yet clearly defined.
In studies attempting to address the first dilemma, steroids have been used either with chlorambucil or with
longer-term cyclophosphamide in small numbers of patients with idiopathic membranous nephropathy and more

Sharon G. Adler and Cynthia C. Nast

advanced renal insufficiency. Limited success has been reported in diminishing proteinuria and/or improving renal
function using chlorambucil or oral cyclophosphamide.
However, patients with renal insufficiency who are treated
with alkylating agents tend to experience more severe side
effects, including leukopenia and clinically significant,
sometimes life-threatening infections. To further complicate matters, in a recently reported prospective, randomized controlled trial using intravenous cyclophosphamide
with steroids in patients with membranous nephropathy
and renal insufficiency, alkylator therapy was not shown
to be a better treatment than steroids alone. Furthermore
in a randomized controlled trial involving 18 patients with
membranous nephropathy and progressive renal insufficiency, steroids and low-dose chlorambucil, but not steroids
and intravenous cyclophosphamide were effective in improving renal function. The efficacy of cyclosporine in treating patients with membranous nephropathy who have a
poor prognosis was examined by Cattran et al. Seventeen
patients who had been followed for 1 year with dietary
therapy alone and who continued to have proteinuria and
a decline in creatinine clearance of at least 8 mL/min were
randomized to receive either cyclosporine or no additional
treatment for 12 months. The patients treated with
cyclosporine had improved proteinuria and creatinine
clearance compared to the untreated group, and 6 of 8
maintained their decrement in proteinuria up to 9 months
after cyclosporine discontinuation. However, the treated
group had a higher incidence of hypertension and a larger
number of transient rises in serum creatinine. Although
only a small number of patients have been studied to date,
and follow-up is short, these data nevertheless suggest that
cyclosporine therapy may be useful in selected patients
with membranous nephropathy. Confirmatory data in
larger groups of patients with longer term follow-up are
needed to better evaluate the usefulness of this therapy.
Intravenous immunoglobulin has been used in a small number of patients, but more experience will be required to
assess its potential.
To summarize, the treatment of idiopathic membranous
nephropathy remains highly controversial and should be
individualized. To a great extent, the choices regarding
therapy have more to do with the patient's perception of
the benefits and risks of the therapies and the illness than
they do about the physician's sense of them. It is therefore
incumbent on physicians to provide patients with a balanced view of the available information regarding treatment options, along with their specific recommendation for
treatment, in the hope that the patient will be able to
make an informed consent regarding therapy. In low-risk
patients, (e.g., those with normal renal function, relatively
low proteinuria, and manageable nephrotic syndrome),
nonspecific therapies are indicated. These might include
salt restriction and diuretics for the management of edema,
angiotensin converting enzyme (ACE) inhibitors and perhaps nonsteroidal anti-inflammatory drugs to decrease pro teinuria and HMG-CoA reductase inhibitors to improve
hyperlipidemia. Anticoagulation may be used to treat a
thrombotic event, and perhaps also as prophylaxis against

21. Membranous Nephropathy

thrombotic events, particularly in patients with serum albumin <1.5 mg/dL. In high-risk patients, particularly those
with azotemia, histological evidence of tubulointerstitial
fibrosis, very heavy proteinuria, hypertension, or morbid
events from the nephrotic syndrome, the use of steroids in
combination with chlorambucil or oral cyclophosphamide,
or perhaps, in selected patients, cyclosporine or intravenous
immunoglobulin may be beneficial. However, the prescription of these agents should be tempered by the knowledge
that they may be least effective when used under the circumstances of highest risk and that their side-effect potential in this setting may be heightened.
Therapies for the secondary forms of membranous nephropathy are guided by the nature of the underlying cause.
Lupus membranous nephritis is discussed in Chapter 28.
Discontinuation of offending drugs and treatment of infections frequently induce complete remission. The use of
interferon for the treatment of chronic active hepatitis B
has been reported to achieve complete remission in some
patients with associated membranous nephropathy. Remis sion appears more likely in patients who were infected
during adulthood, compared to patients who were infected
perinatally or in early childhood. Of note, the liver disease
of patients with hepatitis B and membranous nephropathy
may be exacerbated by the use of high-dose long-term
steroids. Thus, steroids are currently discouraged as a sole
treatment for patients with membranous nephropathy and
hepatitis. Since malignancy is uncommon as an underlying
cause of membranous nephropathy in patients under the
age of 50, an exhaustive search for malignancy seems unwarranted unless a neoplastic process is suggested by a
thorough history, physical examination, chest X-ray, and
routine clinical chemistries. In older patients, the addition
of mammography in women, stool guaiac examinations,
and endoscopic examination of the colon may be considered. It should be noted that even this would not detect
all potential underlying neoplasms, as the origin of an underlying tumor may include numerous organs, and the neoplasm may be occult at the time of initial presentation with
nephropathy. Medical or surgical treatments of neoplasms
have occasionally been associated with remission. In these
patients, recurrence may be heralded by a relapse of the
nephrotic syndrome.
Bibliography
Austin HA III, Antonyvich TT, MacKay K, Bourapas DT, Balow
JE: NIH conference. Membranous nephropathy. Ann Intern
Med 110:672-82, 1992.

169
Bruns FJ, Adler S. Fraley DS, Segel DP: Sustained remission
of membranous glomrulonephritis after cyclophosphamide and
prednisone. Ann Int Med 114:725-730. 1991
Burstein DM, Korbet SM, Schwartz MM: Membranous glomerulopathy and malignancy. Am J Kidney Dis 22:5 -10, 1993
Cattran DC, Greenwood C, Ritchie S, et al.: A controlled trial of
cyclosporine in patients with progressive membranous nephropathy. Canadian Glomerulonephritis Study Group. Kidney Int
47:1130-1135, 1995.
Falk RJ, Hogan SL, Muller KE, Jennette JC, and the Glomerular
Disease Collaborative Network: Treatment of progressive membranous glomerulopathy: A randomized trial comparing cyclo phosphamide and corticosteroids with corticosteroids alone.
Ann Int Med 116:438-445, 1992
Gabbai FB, Goshwa LC, Wilson CB, Blantz RC: An evaluation
of the development of experimental membranous nephropathy.
Kidney Int 31:1267-1278, 1987
Maschio G, Alberti D, Janin G, et al.: Effect of the angiotensinconverting-enzyme inhibitor benazepril on the progression of
chronic renal insufficiency. N Engl J Med 334:939-945, 1996.
Noel LH, Zonetti M, Droz D. Long-term prognosis of idiopathic
membranous glomerulonephritis. Am J Med 66:82-90, 1979
Peterson JC, Adler S, Burkart JM, et al.: Blood pressure control,
proteinuria, and the progression of renal disease. The Modification of Diet in Renal Disease Study. Ann Intern Med 123:754762, 1995.
Ponticelli C, Zucchelli P, Imbasciati E, Cagnole L, Pozzi C, Passerini P, Grassi C, Limido D, Pasquali S, Volpini T, Sasdelli M,
Locatelli F: Controlled trial of methylprednisolone and chlorambucil in idiopathic membranous nephropathy. N Engl J Med
310:946-950, 1984
Ponticelli C, Zucchelli P, Passerini P, Cesana B, and the Italian
Idiopathic Membranous Nephropathy Treatment Study Group:
Methylprednisolone plus chlorambucil as compared with methylprednisolone alone for the treatment of idiopathic membranous nephropathy. N Engl J Med 327:599-603, 1992
Reichert LJM, Huysmans FTM, Assmann K, Koene RAP, Wetzels
JFM: Preserving renal function in patients with membranous
nephropathy: Daily oral chlorambucil compared with intermit tent monthly pulses of cyclophosphamide. Ann Intern Med
121:328-333, 1994.
Ronco P, Allegri L, Brianti E, Chatelet F, VanLeer EHG, Verroust
P: Antigenic targets in epimembranous glomerulonephritis.
Appl Pathol 7:85-98, 1989
Rostoker G, Ben Maadi A, Remy P, Lang P, Lagrue G, Weil B:
Low-dose angiotensin -converting enzyme inhibitor captopril to
reduce proteinuria in adult idiopathic membranous nephropathy: a prospective study of long-term treatment. Nephrol Dial
Transplant 10:25-21, 1995.
Schiepatti A, Mosconi L, Peina L, et al.: Progression of untreated
patients with idiopathic membranous nephropathy. N Engl J
Med 329:85-89, 1993.
Wheeler DB, Bernard DB: Lipid abnormalities in the nephrotic
syndrome: causes, consequences and treatment. Am J Kidney
Dis 23:331-346, 1994.

IgA NEPHROPATHY AND

RELATED DISORDERS
BRUCE A. JULIAN

In 1968 Berger and Hinglais in France described the

unique renal immunohistologic features of IgA nephropathy. In the subsequent 29 years, it has been recognized as
the most common form of primary glomerulonephritis in
the world and accounts for about 10% of patients reaching
endstage renal failure in many countries. Because of the
enthusiasm of its champion and evolving controversies
about the pathogenesis, the eponym Berger's disease has
gained increasing favor. The same renal immunohistologic
features may be found in patients with nephritis due to
Henoch-Schonlein purpura, perhaps the systemic form of
the disease process causing IgA nephropathy. These two
entities have been grouped as primary IgA nephropathy,
in distinction to secondary IgA nephropathy in patients
with other disord ers in which the IgA deposits in the renal
mesangium due to increased production or decreased extrarenal clearance: alcoholic cirrhosis, dermatitis herpetiformis, psoriasis, ankylosing spondylitis, celiac disease, inflammatory bowel disease, carcinoma at various sites, and
mycosis fungoides.

protein of the classical pathway. In renal biopsy specimens

with IgG, the minimal or no staining for Clq is a useful
criterion to distinguish IgA nephropathy from lupus nephritis.
IgA deposits in the capillaries of skin biopsy specimens
have been reported in some patients with IgA nephropathy
without cutaneous manifestations of a vasculitis. This finding is often cited as support for the hypothesis that IgA
nephropathy and Henoch-Schonlein purpura constitute the
spectrum of one disease. However, the frequency of this
finding is variable and a skin biopsy does not replace a
renal biopsy for the diagnosis.
Light Microscopy

The characteristic light microscopic finding is mesangial

enlargement due to proliferation of mesangial cells, increased mesangial matrix, or both (Fig. 2). The abnormalities may be focal (not all glomeruli affected) and segmental
(only a portion of a glomerulus affected). Most patients
exhibit mesangial changes in all glomeruli, whereas a few
patients show no significant glomerular abnormality. Segmental sclerosing lesions are found in many patients. Proliferation of the glomerular epithelial cells causing crescents
is observed occasionally and is more common when the IgA
deposits extend to the glomerular basement membranes, or
in patients with acute renal insufficiency or macroscopic
hematuria. Circumferential crescents are rare. Focal necro sis of a glomerular capillary suggesting a vasculitis may
be found in patients with Henoch-Schonlein purpura and
occasionally in the biopsy samples from patients with primary IgA nephropathy
The severity of the findings in the interstitium (fibrosis,
tubular damage, infiltration of lymphocytes) and vascular
sclerosis generally correlates with the extent of the glomerular pathology. A few patients undergoing renal biopsy at
the time of macroscopic hematuria with acute renal insufficiency have sloughing of the tubular epithelial cells ("acute
tubular necrosis") with red blood cell casts.

PATHOLOGY
Immunohistology

The diagnosis of IgA nephropathy requires immunofluorescence examination of kidney tissue. IgA is the predominant or sole immunoglobulin and is in the mesangium of all
glomeruli, even those with a normal histologic appearance
(Fig. 1). In occasional patients the IgA deposits extend to
the peripheral capillary loops. The IgA is predominantly
of the IgAl subclass. The intensity of staining for lambda
light chains often exceeds that for the kappa isotype. In
about three fourths of patients a second immunoglobulin,
IgG or IgM (sometimes both), is detected with staining
intensity equal to or less than that for IgA. Staining for C3
(the third component of complement) is usually observed in
areas with IgA, although the deposits do not always coincide. Other components of the alternative complement
pathway, including properdin and factor H G31H), as well
as the membrane attack complex (C5b -9) are often detected. C4 binding protein is detected in about one third of
patients, sometimes accompanied by a second complement

Primer on Kidney Diseases, Second Edition

Electron Microscopy

The mesangium is enlarged by proliferation of mesangial

cells with abundant cytoplasm, often with increased mesan170

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

171

22. IgA Nephropathy and Related Disorders

FIGURE I Direct immunofluorescence microscopy of a glomerulus with normal architecture by light microscopy using FTIC-conjugated anti-human IgA. Staining is l imited
to the mesangial areas (x350).

FIGURE 2 Light microscopic histology of glomerulus with increased mesangial cells and
matrix (periodic acid- Schiff, x350).

gial matrix. Electron-dense deposits in the mesangial and

paramesangial areas correspond to the immune deposits.
Some deposits may extend into the subendothelial area
of the glomerular basement membrane. Deposits in the
subepithelial and intramembranous areas are uncommon.
One third of patients exhibit abnormalities of the glomerular basement membranes: focal thinning, splitting, and lamination. The focal distribution of these changes distinguishes IgA nephropathy from the more diffuse patterns

in Alport's syndrome and thin basement membrane syndrome.

CLINICAL FEATURES
Epidemiology

Primary IgA nephropathy may be diagnosed in individuals of all ages, but is most common in children and young

172

Bruce A. Julian

adults. Males are affected two to three times more often

than females. The frequency of the disease varies substantially between countries. In southern Europe, Asia, and
Australia IgA nephropathy accounts for 20 to 40% of patients with primary glomerular disease. Japan has the highest frequency, probably due to widespread screening of
school children for urinary abnormalities. The United
States and Canada have the lowest frequencies, and IgA
nephropathy is rare in central Africa.
The wide range in disease prevalence may reflect differences in criteria for performing the requisite renal biopsy,
as well as genetically determined differences in the development of renal disease. In the United States the prevalence
of IgA nephropathy is high among some Native American
groups, but very low in African Americans. Several large
families with multiple members with IgA nephropathy have
been described. The clinical features may vary widely
among these relatives; some have Henoch-Schonlein purpura. Familial IgA nephropathy has not yet been linked
to an HLA antigen, restriction fragment length polymorphism, or other genetic marker. The mode of inheritance
of familial IgA nephropathy remains uncertain.
Clinical Presentation

Patients with primary IgA nephropathy usually present

with one of three syndromes (Table 1). The most distinctive
is an episode of macroscopic hematuria concurrent with
an infection of the upper respiratory tract. In contrast to
a 10-14 day delay for the onset of macroscopic hematuria
in patients with poststreptococcal glomerulonephritis, the
urine becomes red or tea-colored 1 to 2 days after the onset
of symptoms; this feature has been labeled "synpharyngitic" hematuria. The patient is usually asymptomatic, but
sometimes reports malaise, fatigue, or myalgia. Hypertension and peripheral edema of the nephritic syndrome, common in poststreptococcal glomerulonephritis, are rare.
Some patients, especially children, experience loin pain.
Bleeding may last hours to many days; episodes may recur
months or years later with pharyngitis, a febrile illness of
another cause, or extreme exercise. Macroscopic hematuria
is more commo n in children than adults, and the interval
between episodes lengthens with increasing age. A few
patients with macroscopic hematuria develop acute renal
dysfunction that generally improves with supportive therapy; dialysis is occasionally necessary.

TABLE I

Clinical Presentation of IgA Nephropathy

Common
Synpharyngitic macroscopic hematuria
Asymptomatic microscopic hematuria
Henoch- Schonlein purpura
Uncommon
Nephrotic syndrome
Chronic renal insufficiency with hypertension

A second presentation is asymptomatic microscopic hematuria with variable degrees of proteinuria. Such patients
are often adults undergoing routine urinalysis as part of
an insurance or employment physical. Proteinuria is usually
less than 1 g/day. These patients rarely experience macroscopic hematuria after ascertainment.
Henoch-Schonlein purpura is a third common mode of
presentation and develops much more frequently in children than in adults. Patients also manifest skin, joint, and
intestinal involvement. The skin lesions are characteristically more numerous on the legs and lower trunk. The rash
is frequently morbiliform early in the course and rapidly
becomes purpuric. Arthralgias are not usually accompanied
by features of inflamed joints. Gastrointestinal symptoms
may be striking, including severe abdominal pain, ileus, or
bloody diarrhea. Nephritis often develops after the initial
symptoms subside; therefore, a urinalysis should be
checked at a 1-month follow-up visit. Some patients, especially young children, never exhibit nephritis.
A less common presentation is the nephrotic syndrome.
These patients may have advanced glomerular disease with
hypertension and renal insufficiency. Alternatively, the nephrotic syndrome may be the initial clinical feature of patients with normal renal function and normal blood pres sure. Proteinuria in the latter setting is often selective,
comprised primarily of smaller proteins such as albumin.
Renal biopsy shows histologic features similar to minimalchange glomerulonephritis, but with diffuse staining for
IgA in all mesangia. Whether these few patients comprise
a subset of IgA nephropathy patients or have two separate
diseases is controversial. Chronic renal insufficiency and
hypertension (occasionally malignant) are the initial manifestations for a minority of patients.
Laboratory Findings

Because the glomerular IgA is apparently derived from

the circulation, much attention has focused on the serum
levels of IgA or IgA-containing immune complexes. Serum
IgA concentrations are increased in about 50% of patients.
However, this finding is neither unique to patients with
IgA nephropathy nor consistent when multiple assays are
performed for an individual patient over several months.
Serum concentrations of complement components are usually normal. Levels of circulating IgA-containing immune
complexes are frequently increased. The complexes may
contain 'C3, IgG, or both, but an antigen such as a viral
or dietary protein has not been identified. Discovery of
circulating IgA-fibronectin complexes in some patients
sparked interest as a possible screening tool, but the specificity and sensitivity are too poor. Renal biopsy remains
the standard for the diagnosis.
Natural History

IgA nephropathy is a chronic disease. Although clinical

remission (resolution of microscopic hematuria and proteinuria with a normal serum creatinine concentration) may
develop in as many as one third of patients several years

22. IgA Nephropathy and Related Disorders

after renal biopsy, disappearance of the IgA from glomeruli

has been only rarely documented. Renal function progres sively worsens in about 40% of patients, about half of whom
reach endstage renal failure after 20 years of clinically
apparent disease. For patients with modest proteinuria at
biopsy, an increase in magnitude of proteinuria often portends a rise in serum creatinine concentration. About 10%
of patients with renal insufficiency develop the nephrotic
syndrome. Nearly one third of patients exhibit a benign
course with chronic microscopic hematuria and normal serum creatinine concentration; proteinuria generally remains less than 1 g/day.
Hypertension in patients with primary IgA nephropathy
is at least as common as in patients with membranous or
membranoproliferative glomerulonephritis, but probably
less frequent than in patients with idiopathic focal segmental glomerulosclerosis. Hypertension is more common in
patients whose biopsy specimens show vascular sclerosis.
Malignant hypertension develops in about 5% of patients.
Among patients whose urinalysis becomes normal with
prolonged observation, half remain hypertensive.
Prognostic Markers

Clinicopathological parameters shown by multivariate

analysis in studies at separate centers to predict a poor
long-term clinical outcome are shown in Table 2. These
features are more useful for a group of patients than for
an individual. Patients with a history of macroscopic hematuria also have a better long-term outlook. This clinical
feature is more common in pediatric patients who may
have had a shorter duration of disease than adults. Nonetheless, it remains a marker for a good prognosis even after
taking patient age into account. Variations in the gene
encoding angiotensin converting enzyme may predict the
clinical course. In some, but not all, studies, patients homo zygous for the D (deletion) allele had a worse prognosis
than those homozygous the I (insertion) allele. The prognostic value of other clinical variables, including older age
at diagnosis (with an unknown duration of disease) and
male gender, are less well defined. In contrast, several laboratory or histologic parameters have not been helpful in
predicting outcome, including serum IgA concentration,
intensity of immunofluorescence staining for IgA in the
biopsy specimen, and the isotype of any additional immu -

TABLE2
Markers of Poor Outcome in IgA Nephropathy

Hypertension
Proteinuria greater than 2 g/24 hr
Renal insufficiency at biopsy
Biopsy changes
Severe proliferative and sclerotic changes in glomeruli
Interstitial fibrosis

Vascular sclerosis
Extension of IgA deposits to peripheral capillary walls

173

noglobulin in the mesangium. The long-term prognosis differs between some ethnic groups, perhaps reflecting varied
methods of ascertainment and differences in criteria for
performing renal biopsies.
PATHOGENESIS

The pathogenesis of IgA nephropathy is not yet understood. There is no evidence, however, that the process is
different in patients with macroscopic hematuria or familial
disease as compared with those without these clinical features. Multiple mechanisms likely play a role in the primary
and secondary forms of the disease. Research has focused
on the characteristics of the IgA molecules in the mesangium, the source of these molecules, the mechanism of
deposition in the mesangium, and the inflammatory response that culminates in glomerular damage.
Primary IgA Nephropathy

Immunofluorescence and electron microscopic findings

of granular mesangial deposits of IgA and C3 resemble the
pattern in experimental immune complex glomerulonephritis. However, it is uncertain whether the IgA and C3
are deposited as an immune complex from the circulation
or whether they combine locally through an in situ mechanism. IgAl is the predominant IgA subclass in the renal
deposits, whereas the other IgA subclass, IgA2, is rarely
detected. For technical reasons, the relative contribution
of the molecular forms, monomers and polymers, is difficult
to establish. The question of polymeric vs monomeric form
in the mesangium is relevant for the induction of inflammation damaging glomeruli. Polymeric IgA is much more active phlogistically than monomeric IgA.
Several observations support the hypothesis that the
mesangial IgA is derived from the circulation: recurrence
of mesangial IgA in a transplanted allograft, clearance of
renal IgA deposits in an allograft inadvertently transplanted from a donor with subclinical IgA nephropathy,
restriction of the mesangial IgA to the IgAl subclass, and
increased plasma levels of IgAl (with normal levels of
IgA2) and IgAl-containing immune complexes.
Increased levels of these complexes could be caused by
increased synthesis or decreased hepatic catabolism. Increased production of IgAl has been shown, but alone is
is not usually sufficient to cause IgA nephropathy. For
example, patients with IgAl myeloma or infection with
HIV have high plasma levels of IgAl but rarely exhibit
glomerular deposits. Thus, a qualitative abnormality of circulating IgA appears necessary for mesangial deposition.
Recent studies have shown that IgA molecules in the blood
of patients with IgA nephropathy contain decreased numbers of galactose residues in the O-linked glycans in the
hinge region. This abnormality may decrease uptake by
hepatic receptors, thereby bypassing metabolism in the
liver and increasing the amount circulating through the
kidneys. Moreover, the decreased content of galactose may
enhance binding of IgA molecules to renal mesangial cells,
leading to greater glomerular sclerosis and damage. Some

174

Bruce A. Julian

of the circulating IgA complexes also contain fibronectin

that facilitates attachment of IgA to collagen in the mesangial matrix, but this binding is not specific for patients with
IgA nephropathy. In addition, IgA molecules may also
attach to type V collagen in the matrix by a mechanism
independent of fibronectin.
The process by which mesangial deposits of IgA damage
glomeruli has not been elucidated. Activation of complement appears important, as in other types of glomerulonephritis. Immunohistologic findings suggest involvement
predominantly of the alternative pathway, although the
classical complement pathway may participate in some patients. Cells infiltrating glomeruli from the circulation may
accentuate the inflammatory response by secreting cytokines. Furthermore, mesangial cells synthesize an array of
inflammatory mediators and express receptors for some of
them. On activation, the mesangial cells proliferate and
secrete extra matrix.
Secondary IgA Nephropathy
The pathogenetic mechanisms of secondary IgA nephropathy are uncertain. Decreased hepatic clearance of
IgA-containing immune complexes has been postulated for
patients with alcoholic cirrhosis. Increased systemic exposure to dietary antigen leading to increased production of
IgA has been proposed for patients with celiac disease,
inflammatory bowel disease, and dermatitis herpetiformis.
However, the basis for glomerular deposition of IgA in
patients with various other diseases, such as carcinoma of
the respiratory or gastrointestinal tract, is unknown.
TREATMENT

Treatment of secondary IgA nephropathy is generally

directed toward the associated medical condition. Clearance of mesangial IgA deposits has been described in patients whose principal disorder resolved. Alternatively, no
consensus has emerged about a disease-specific approach
for patients with primary IgA nephropathy. Unfortunately,
only a few controlled treatment trials have been conducted.
One approach to treatment is to separate patients into two
clinical categories: acute and chronic.
Acute Disease

Acute renal dysfunction affects a small minority of patients. The renal biopsy specimen may show glomerular
epithelial crescents. No consensus has been reached about
the approach to these patients, but treatment has included
glucocorticoids, cytotoxic agents, anticoagulants, or plasmapheresis, alone or in combination. Many centers favor
glucocorticoid therapy, often given as three 0.5 to 1.0 g
intravenous doses of methylprednisolone followed by an
oral prednisone regimen for several months. Assessment
of the apparent benefit is complicated by spontaneous resolution of renal dysfunction in some patients. Early
treatment-associated improvement in renal function is frequently shortlived.

For patients with macroscopic hematuria and tubular

necrosis without glomerular crescents, renal function often
returns to baseline with supportive therapy with control of
blood pressure and dialysis as necessary.
A few patients with normal renal function develop the
nephrotic syndrome with light microscopic histologic features of minimal-change glomerulonephritis. Treatment
with glucocorticoids clears the nephrotic syndrome, but the
IgA deposits usually persist.
Chronic Disease
Most patients with chronic IgA nephropathy are hypertensive and control of their blood pressure is the cornerstone of therapy. An angiotensin converting enzyme
(ACE) inhibitor appears to be the agent of choice. Not
only is blood pressure usually well controlled, but the rate
of decline of creatinine clearance is frequently slowed or
stabilized, possibly due to better control of intraglomerular
capillary hypertension or decreased cytokine-induced glomerular sclerosis. Proteinuria often improves, but a decrement does not always predict a favorable effect on creatinine clearance. Enthusiasm for this class of agents recently
increased because patients homozygous for the D allele of
the gene encoding angiotensin converting enzyme had the
highest circulating levels of the enzyme, and the largest
decrement in proteinuria after treatment.
Use of an ACE inhibitor in normotensive patients is
more controversial. These patients often have a normal
serum creatinine and may become hypotensive with treatment. In one study of adults, therapy for 4 months significantly decreased proteinuria; however, no study has evaluated the long-term preservation of renal function in this
setting.
Because macroscopic hematuria frequently coincides
with an upper respiratory tract infection, infection may
accentuate glomerular damage, perhaps by increasing synthesis of IgA or various cytokines that augment the inflammatory process. Although chronic antibiotic prophylaxis has no proven benefit, tonsillectomy in some studies
decreased the magnitude of hematuria; however, its effects
on renal function have not been adequately evaluated. Phenytoin decreases serum IgA concentrations and treatment
may decrease the frequency of macroscopic hematuria, but
histologic features do not improve.
In the absence of a disease-specific therapy, investigators
have tried various other approaches. In a nonrandomized
study, children with histologic features associated with progressive disease treated with alternate-day prednisone preserved renal function with resolution of proteinuria and
hematuria. In a separate study, daily prednisone therapy of
adults preserved renal function when the initial creatinine
clearance was greater than 70 ml/min. Fish oil rich in
omega-3 fatty acids given for 2 years was found to preserve
creatinine clearance in patients with proteinuria at least
1 g/day in a mu lticenter, placebo-controlled, randomized
study. Proteinuria was not altered. Therapy was well tolerated: No bleeding complication or increase in plasma cholesterol or triglyceride concentrations was observed. Prior

22. IgA Nephropathy and Related Disorders

175

results with fish oil had been more controversial; only two
of four trials showed a beneficial effect. Treatment for 9
months with pooled intravenous followed by intramuscular
immunoglobulins slowed the decline in renal function and
lessened proteinuria in one study, but the effect was frequently shortlived after stopping therapy. Several immunosuppressive or antiinflammatory agents, alone or in combination, have not consistently shown benefit: azathioprine,
cyclosporine, aspirin, dipyridamole, dapsone, and danazol.
Unfortunately, there is little agreement about these therapeutic options.
Two general treatment options may be helpful. Moderate
salt restriction usually enhances an antihypertensive regimen. Also, restriction of dietary protein intake to 0.6 g/kg
body weight per day may slow loss of renal function in
patients with various chronic renal diseases such as IgA
nephropathy, but patients should be monitored for protein malnutrition.
For patients reaching endstage renal failure, transplantation affords an excellent option. Recurrent disease (the
appearance of the immunohistologic features) has been
documented in about 50% of patients 2 years posttransplant, but loss of the allograft from recurrent disease is
uncommon. Kidneys from living-related donors are not at
greater risk than cadaveric allografts for recurrent disease.
BIBLIOGRAPHY
Berger J: Recurrence of IgA nephropathy in renal allografts. Am
J Kidney Dis 12:371-372, 1988.

Cameron JS: Henoch-Schonlein purpura: clinical presentation.

Contrib Nephrol 40:246-249,1984.
Cattran DC, Greenwood C, Ritchie S: Long-term benefit of angio tensin -converting enzyme inhibitor therapy in patients with severe immunoglobulin A nephropathy: A comparison to patients
receiving treatment with other antihypertensive agents and to
patients receiving no therapy. Am J Kidney Dis 23:247-254,1994.
Donadio JV Jr, Bergstralh EJ, Offord KP, et al.: A controlled trial
of fish oil in IgA nephropathy. N Engl J Med 331:11941199,1994.
Galla JH: IgA nephropathy. Kidney Int 47:377-387,1995.
Isbels LS, Gyory AZ: IgA nephropathy: Analysis of the natural
history, important factors in the progression of renal disease,
and a review of the literature. Medicine 73:79-102, 1994.
Julian BA: Treatment options in IgA nephropathy. Nephrology
3:103-108, 1997.
Mestecky J, Tomana M, Crowley-Nowick PA, et al.: Defective
galactosylation and clearance of IgAl molecules as a possible
etiopathogenetic factor in IgA nephropathy. Contrib Nephrol
104:172-182,1993.
Mustonen J, Pasternack A: Associated diseases in IgA nephropathy. In: Clarkson AR (ed) IgA Nephropathy. Martinus Nijhoff
Publishing, Boston, pp. 47-65, 1987.
van Es LA: Pathogenesis of IgA nephropathy. Kidney Int 41:17201729,1992.
Wyatt RJ, Kritchevsky SB, Woodford SY, et al.: IgA nephropathy:
Long-term prognosis for pediatric patients. / Pediatr 127:913919, 1995.
Yoshida H, Kon V, Ichikawa I: Polymorphisms of the renin -angiotensin system genes in progressive renal diseases. Kidney Int
50:732-744,1996.

GOODPASTURE'S SYNDROME AND

ANTI-GBM DISEASE
CAROLINE O.S. SAVAGE

INTRODUCTION

Goodpasture's syndrome is defined as the concurrence

of pulmonary hemorrhage and focal necrotizing glomerulo nephritis. This pulmonary-renal syndrome is seen most
commonly in the context of systemic vasculitis and thenmost frequently in the presence of antineutrophil cyto-

Primer on Kidney Diseases, Second Edition

plasm antibodies (ANCA). Confusingly, Goodpasture's

name is also associated with the occurrence of antiglomerular basement membrane (GBM) antibodies. Today, many
prefer to use the term Goodpasture's disease to specifically
describe those patients with focal necrotizing glomerulonephritis, often accompanied by pulmonary hemorrhage, who
demonstrate autoimmunity to the GBM.
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

Caroline O.S. Savage

176
Causes of Goodpasture's syndrome, specifically pulmo nary-renal vasculitic syndromes, are given in Table 1.
Goodpasture's syndrome needs to be distinguished from
the presence of lung and renal disease occurring together
either with (e.g., sarcoidosis) or without (e.g., presence of
pulmonary emboli, infection, or cancer occurring in patients with diverse types of nephritis) immediate pathogenic
relationship. This chapter deals specifically with Goodpas ture's disease or anti-GBM disease.
GOODPASTURE'S DISEASE
Clinical and Pathological Features

Goodpasture's disease has a peak incidence in the third to

sixth decades but all ages may be affected and there is a
secondary peak in incidence in the sixth and seventh decades. The full syndrome with pulmonary hemorrhage and
nephritis is more common in young males, perhaps reflecting smoking habits, wherever elderly females tend to pres ent with nephritis alone. Whites are more commonly affected than individuals of African ancestory. Patients with
overt pulmonary hemorrhage tend to present at an earlier
stage and before oliguria develops, while in the absence of
overt pulmonary hemorrhage patients often present in renal failure. Smoking predisposes to development of pulmo nary hemorrhage. Pulmonary hemorrhage occurs in about
two thirds of patients in most series.
Features of pulmonary hemorrhage include symptoms
of continuous or episodic dyspnea, and of hemoptysis. Hemoptysis may be minimal or massive and absence of this
feature does not exclude presence of pulmonary hemorrhage. Symptoms can develop rapidly over hours. Signs
may include tachypnea, cyanosis in severe cases, and inspiratory crackles or bronchial breathing over the basal lung
fields. The chest radiograph usually shows patchy or diffuse
infiltrates involving the central lung fields; the changes are
usually but not always symmetrical (Fig. 1). Blood gas

FIGURE I Chest radiograph of a patient with Goodpasture's disease who had pulmonary hemorrhage.

FIGURE 2 Acute crescentic glomerulonephritis associated with

Goodpasture's syndrome.

analysis reveals a reduced pQ-i. A rise in transfer factor,

the rate of transfer of a small concentration of CO in the
inspired air onto hemoglobin, is indicative of bleeding
within alveoli; correction should be made for hemoglobin
concentration because transfer factor varies directly with
this. Significant blood loss into the lungs may cause a fall
in circulating hemoglobin concentration, while actual
bleeding may be visualized at bronchoscopy. Occasionally
subclinical hemorrhage leads to iron deficiency anemia.
Lung biopsy and histology may document intra-alveolar
red blood cells or hemosiderin-laden macrophages. Immu nofluorescence and immunoperoxidase staining is difficult
in the lung, but when possible, studies have demonstrated
linear fixation of IgG along the pulmonary basement membrane.
Nephritis is associated with hematuria on urinanalysis;
the urinary sediment reveals red blood cell casts and dis torted red blood cells, due to bleeding directly from glomeruli into renal tubules. Proteinuria may be present but is
usually less than 5 g /24 hr. If renal function is not already
impaired on measurement of BUN and Serum creatinine
levels, it may rapidly and inexorably deteriorate if treatment is not instituted as an emergency. Kidney size is usually normal on renal ultrasound and percutaneous renal

23. Goodpasture's Syndrome and Anti-GBM Disease

177

lar C3 is present in about two thirds of biopsies. It is worth

remembering that other conditions can cause linear staining on direct immunostaining including systemic lupus ery thematosus, diabetes mellitus, normal autopsy kidney, cadaveric kidney after perfusion, and renal transplants in
patients with Alport's syndrome.
Diagnosis

The diagnosis of Goodpasture's disease is confirmed by

presence of linear IgG. (detected by direct immunofluorescence), along the GBM in renal biopsy tissue and by
detection of circulating anti-GBM antibodies. These latter
may be detected within serum samples by using specific
ELISA or, as used in some reference laboratories, by radioimmunoassay. In both, the principle is to coat purified
GBM components that have been digested by collagenase
(or, as is increasingly used, NClaSIV peptides) onto a
solid-phase support such as plastic microtiter plates. Diluted serum is applied and anti-GBM antibodies bind to
the GBM components. Binding is subsequently detected
using anti-human IgG reagents labeled a marker agent.
Concurrence with Other Diseases

FIGURE 3 Immunoperoxidase staining showing linear fixation of

IgG along the GBM in Goodpasture's syndrome.

biopsy permits a histological renal diagnosis. Typically,

there is a diffuse nephritis with focal or total necrosis of
glomerular tufts and segmental or circumferential cellular
crescents around some or all glomeruli (Fig. 2). By immunofluorescence or immunoperoxidase staining, linear IgG
fixation is present along the GBM (Fig. 3) and may be
accompanied by other immunoglobulins. Linear or granu-

TABLE I

Pulmonary-Renal Vasculitic Syndrome

Disease Diagnostic test assay

Microscopic polyangiitis
Wegener's granulomatosis
Goodpasture's disease Systemic
lupus erythematosus ChurgStrauss syndrome
Henoch-SchSnlein purpura
Beliefs disease Rheumatoid
vasculitis Penicillamine
1

ANCA, antineutrophil
cytoplasm antibodies.

ANCA"
ANCA
Anti-GBM antibodies
Anti-DNA antibodies
ANCA present in about half
of cases IgA deposition in
skin and
kidney Clinical
only Rheumatoid
factor Drug history

A number of well-documented cases of membranous

glomerulonephritis have evolved into Goodpasture's dis ease with presence of circulating and fixed anti-GBM antibodies. Patients with Goodpasture's disease and anti-GBM
antibodies may sometimes have clinical evidence of vasculitis with circulating ANCA. Finally, Goodpasture's disease
can arise within the transplanted kidneys of patients with
Alport's syndrome. The native kidneys of these individuals
lack the Goodpasture antigen (the underlying inherited
defect of a5(IV)-chains causes a secondary loss of 3(IV)chains) predisposing them to develop an immune response
to the previously unseen antigen that is present within
the allograft.
Autoantigens and Autoantibodies

The pathogenesis of Goodpasture's disease is believed

to be directly linked to development of autoimmunity to
the noncollagenous 1 domain of the a3 chain of type IV
collagen (NCla3IV). The type IV collagen molecule is
composed of three a(IV) -chains which are folded into a
globular NCI domain at the C-terminal region. There are
five chains for type IV collagen and a sixth gene has been
identified. The al(IV) and 2(IV) chains are ubiquitous
while the a 3(IV), a 4(IV) and a 5(IV)-chains colocalize
and are only present in specialized basement membranes,
including those within the kidney, lung, eye, cochlea, and
choroid plexus.
Development of autoantibodies to NClaSIV is closely
linked to clinical evidence of disease, and passively injected
anti-GBM antibodies can cause disease in subhuman primates and in sheep. Binding of autoantibodies to glomerular or pulmonary basement membranes initiates
an inflammatory response with complement deposition,
leukocyte recruitment, and eventual tissue injury.

178

Caroline O.S. Savage

Genetics

Many autoimmune diseases demonstrate association

with antigens of the major histocompatibility (MHC) complex. Goodpasture's disease shows a strong association with
HLA-DR2. HLA -DR2 has been divided into two specificitiesDRwlS and DRwl6each of which consists of a
number of alleles. More recent studies using molec ular analysis of HLA class II genes in Goodpasture's
disease has shown that the association is with the
DRwlS (DRB1*1501 and *1502) alleles. These alleles are
in strong linkage disequilibrium with DQw6 (DQA1*01
and DQB1*06) and so there is also an association with this
entire haplotype.
Environmental Influences
There are a number of reports suggesting geographical
case clustering on occasion, possibly indicative of environmental factors that might initiate or reveal disease. Smoking or infections may act to aggravate tissue injury and so
reveal disease. There are also reports of association with
exposure to organic solvents or hydrocarbons.
Therapy
Therapies are directed toward removing the pathogenic
anti-GBM antibodies as quickly as possible, preventing
further synthesis and reinstituting tolerance to NClaSIV.
This was the reasoning behind the introduction of plasma
exchange, used in conjunction with powerful immunosuppressive therapy, in the 1970s. Most therapeutic protocols
now use a combination of prednisolone, cyclophosphamide,
and plasma exchange (Table 2). Response to treatment
can be monitored by serum creatinine, chest radiograph
changes, improvement in lung transfer factor, and by monitoring anti-GBM antibody liters. Blood count measurements are important not only to exclude further falls in
hemoglobin due to lung hemorrhage but also to ensure that
cyclophosphamide therapy does not induce life-threatening
bone marrow suppression. It is important to avoid infection
as far as is possible not only because the patient is immunosuppressed but also because infection may precipitate wors ening of disease and tissue injury during the early stages
of treatment.

TABLE 2

Treatment Protocol for Goodpasture's Syndrome

Prednisolone
1 mg kg ' 24 hr 1 orally (maximum 80
mg). Weekly dose reduction to 20 mg
and then more slowly to 1-2 years
Cyclophosphamide 2.5 mg kg" 1 d"1 orally (maximum 150
mg/d). Continue for 4 months, white blood
cell count permitting, then change to
azathioprine for 1-2 year
Plasma exchange Daily 4 L exchanges for 4.5% albumin for 14
days or until anti-GBM antibodies are
undetectable

The observation that patients presenting with end stage

renal failure who are dialysis dependent and who have a
severe crescentic renal biopsy rarely recover renal function
has led to the recommendation that such patients should
be spared immunosuppressive therapy. Predictors of a poor
renal prognosis include a serum creatinine greater than
7 mg/dL and crescent formation in more than 50% of the
glomeruli on renal biopsy. The exception would be those
individuals who have concurrent pulmonary hemorrhage
who should receive full immunosuppressive therapy. The
use of methylprednisolone has been advocated instead of
plasma exchange in this setting and while there are no
controlled clinical trials that directly compare use of plasma
exchange with methylprednisolone, methylprednisolone
may be associated with a greater risk of opportunistic infections, which may considerably worsen the prognosis. In
those patients in whom immunosuppressive therapy is withheld, anti-GBM antibody liters fall spontaneously with a
mean time to baseline values of about 11 months. Once
anti-GBM antibodies are undetectable, these patients may
be safely transplanted.
Plasma Exchange
Plasma exchange involves the extracorporeal separation
of plasma from cellular elements of blood, with return of
cells to the individual together with a substitute fluid for
the removed plasma. Automated methods are based on
either centrifugation principles, with continuous or discontinuous flow techniques, or nitration principles. In both
methods, blood is processed in an extracorporeal circuit
and exchange of 2 to 4 L may be achieved in 2 hours.
The procedure requires vascular access that is capable of
withstanding flow rates of 60 to 150 mL/min (e.g., via antecubital veins or indwelling catheters inserted into the internal jugular or subclavian veins), adequate anticoagulation
(e.g., heparin, acid citrate dextrose B form), and suitable
replacement fluids or combinations of fluids (e.g., albumin
solutions, fresh frozen plasma, synthetic plasma expanders,
crystalloids). Fresh frozen plasma (300-400 ml) may be
given at the end of the exchange when there is fresh pulmo nary hemorrhage or within 24 hours of an invasive procedure.
Complications include the following; anaphylactoid reactions, most common with use of fresh frozen plasma; metabolic alkalosis in patients with renal failure due to infusion
of solutions containing high citrate and low chloride levels;
citrate toxicity especially if liver disease is present; hemorrhage particularly in patients with thrombocytopenia or
defective clotting; thrombosis related to access procedures;
posttransfusion hepatitis; cardiovascular complications as sociated with administration of sodium citrate, ionic alterations, and use of central vascular access; pulmonary edema
caused by alveolar lesions induced by presence of leukocyte
alloantibodies in replacement plasma; nausea, vomiting,
and rigors due to endotoxin contamination of replacement fluids.
Course of the Disease
Effective immunosuppressive therapy usually leads to
abolition of the circulating anti-GBM antibodies, although

2 3 . G o o d p a s t u r e ' s S y n d r o m e a n d A n t i -GBM Disease

linear antibody may remain fixed within the kidney for a

year or more. It is possible but highly unusual for patients
with Goodpasture's disease to relapse, unlike patients with
systemic vasculitis who relapse frequently. In Goodpas ture's disease, even severe lung lesions have the capacity
to resolve completely, and it is unusual for patients who
have survived serious lung hemorrhage to be left with significant pulmonary impairment or with radiological evidence of fibrosis. The same is not true of the nephritis;
patients presenting with a creatinine above 6-7 mg/dL, with
oligoanuria, or with extensive crescent formation within the
renal biopsy, do not recover renal function.
Bibliography
Burns AP, Fisher M, Li P, et al.: Molecular analysis of HLA class
II genes in Goodpasture's disease. Q J Med 88:93-100, 1995.
Fleming SJ, Savage COS, McWilliam LJ, et al.: Anti-GBM antibody mediated nephritis complicating transplantation in a patient with Alport's syndrome. Clin Nephrol 46:857-859, 1988.
Flores JC, Taube D, Savage COS, et al.: Clinical and immunological evolution of oliguric anti-GBM nephritis treated by haemodialysis. Lancet 1:5-8, 1986.
Haworth SJ, Savage COS, Carr D, et al.: Pulmonary hemorrhage
complicating Wegener's granulomatosis and microscopic polyarteritis. Br Med J 290:1775-1778, 1985.

179
Lerner RA, Glassock RJ, Dixon FJ: The role of anti-glomerular
basement membrane antibody in the pathogenesis of human
glomerulonephritis. / Exp M ed 126:989-1004, 1967.
Merkel F, Pullig O, Marx M, et al.: Course and prognosis of antibasement membrane antibody (anti-BM-Ab)-mediated disease:
Report of 35 cases. Nephrol Dial Transplant 9:372- 376,
1994
Niles JL, Bottinger EP, Saurina GR, et al.: The sy ndrome of
lung hemorrhage and nephritis is usually an ANCA-associated
condition. Arch Intern Med 156:440 -445, 1996.
Phelps RG, Turner AN: Goodpasture's syndrome; New insights
into pathogenesis and clinical picture. J Nephrol 9:111-117,1996.
Saus J, Wieslander J, Langeveld JPM, et al.: Identification of the
Goodpasture antigen as the alpha3(IV) chain of collagen IV.
/ Bio! Chem 263:13374-13380, 1988.
Savage COS, Pusey CD, Bowman C, et al.: Antiglomerular basement membrane antibody mediated disease in the Br itish Isles
1980-1984. Br Med J 292:301-304, 1986.
Short AK, Esnault VLM, Lockwood CM: Antineutrophil cytoplasm antibodies and antiglomerular basement-membrane antibodies - 2 coexisting distinct autoreactivities detectable in patients with rapidly progr essive glomerulonephritis. Am J Kidney
Dis 26:439-445, 1995.
Turner AN, Rees AJ: Goodpasture's disease and Alports syndromes. Anna Rev Med 47:377-386, 1996.

SECTION 4
THE KIDNEY IN
SYSTEMIC DISEASE

RENAL FUNCTION IN CONGESTIVE

HEART FAILURE
WILLIAM T. ABRAHAM AND ROBERT W. SCHRIER

kidney as the ultimate effector organ of body fluid regulation is intact. Thus, in such edematous states, the kidney
must be responding to extrarenal signals from the afferent
(i.e., "sensor") limb of a volume regulatory system by decreasing renal sodium and water excretion. The exact nature of these afferent mechanisms has been a major focus
of research.
A proposed body fluid volume regulation hypothesis indicates that the relative integrity or "fullness" of the arterial
circulation constitutes the primary afferent signal through
which the kidneys either increase or decrease their excretion of sodium and water (Fig. 1). This hypothesis explains
how an increase in the volume of blood on the venous side
of the circulation may cause a rise in total blood volume
while a decrease in the relative volume of blood in the
arterial circulation may promote continued renal sodium
and water retention. A decrease in cardiac output is the
most obvious means whereby a decrease in arterial circulatory integrity may occur. However this cannot be the only
afferent signal for underfilling of the arterial circulation
because several edematous disorders, such as cirrhosis,
high-output cardiac failure, pregnancy, and large arteriovenous fistulae, are associated with increased card iac output.
Thus, peripheral arterial vascular resistance and the compliance of the arterial vasculature has been proposed as
the second determinant of the fullness of the arterial circulation. Peripheral arterial vasodilation therefore provides
another afferent signal for arterial underfilling which causes
renal sodium and water retention. Thus, a diminished cardiac output could initiate sodium and water retention in
low-output heart failure, whereas arterial vasodilation is
the proposed mechanism in high-output CHF and other
edematous states (Fig. 1).
Arterial blood pressure is not a sensitive index of the
presence of arterial underfilling in cardiac failure or other
edematous states, because of the rapidity of the compensatory efferent (i.e., "effector") responses. Rather, it is only
with advanced disease states that the compensatory responses to arterial underfilling are insufficient to maintain
mean arterial pressure. Even if sensitive methods for accurately measuring the volume of blood in the art erial circulation (which is generally less than 2% of total body fluid)
were available, the absolute measured arterial blood vol-

Chronic congestive heart failure (CHF) affects more

than 4 million Americans, or approximately 1 to 1.5% of
the U.S. population. For those age 65 to 74 years, the
prevalence of CHF is about 5%, while for those older than
75 years, the prevalence of CHF exceeds 10%. Currently,
there are 700,000 new cases of CHF diagnosed each year
in the United States. Moreover, CHF is now the most
common hospital discharge diagnosis for those over the age
of 65 years and the fourth leading cause of hospitalization in
US adults. In 1992, the estimated total direct costs of heart
failure in the US exceeded 10 billion dollars, while hospitalization costs for heart failure surpassed those for myocardial infarction and cancer combined.
A major cause of morbidity and mortality in patients with
symptomatic CHF relates to alterations in renal function
which cause avid sodium and water retention. Sodium and
water retention in patients with CHF results in expansion of
intravascular fluid volume, which leads to increased cardiac
preload and subsequently to an enlargement of extravascular and interstitial fluid volumes manifested clinically as
pulmonary and peripheral edema. This increase in cardiac
preload contributes to the two most common symp toms of
heart failuredyspnea and fatigueand may contribute
to the progressive decline in cardiac performance that comprises the natural history of CHF. Knowledge of the pathogenesis of this renal sodium and water retention permits
the rational use of conventional and investigational medical
therapy in CHF.
AFFERENT MECHANISMS INVOLVED IN THE SODIUM
AND WATER RETENTION OF CHF

In normal humans, an expansion of the extracellular fluid

(ECF) volume due to an increase in sodium and water
intake is associated with an increase in renal sodium and
water excretion followed by restoration of the normal ECF
volume. However, in certain edematous states, including
cardiac failure, cirrhosis, and pregnancy, avid renal sodium
and water retention persists despite expansion of the ECF
volume. In fact, this clinical paradox of continued renal
sodium and water retention despite total body sodium and
water excess defines these edematous disorders. In these
clinical circumstances, it is clear that the integrity of the

Primer on Kidney Diseases, Second Edition

183

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

William T. Abraham and Robert W. Schrier

184

BODY FLUID VOLUME REGULATION

HYPOTHESIS

I Cardiac
Output

1 Peripheral
Vascular Resistance

1 Fullness of the
Arterial Circulation

T Non-Osmotic
Vasopressin
Release

T Sympathetic
Nervous System
Activity

T Renin-AngiotensinAldosterone System
Activity

Diminished Renal
Hemodynamics
and
Renal Sodium and Water
Excretion

FIGURE I Proposed mechanism of renal sodium and water retention in high -output and lowoutput cardiac failure. A decrease in cardiac output initiates sodium and water retention in
low-output heart failure. In high -output CHF, arterial vasodilation starts the sequence of
events leading to diminished renal hemodynarm'cs and reduced renal sodium and water
excretion. From Schrier RW: N Engl J Med 319:1065-1072, 1988. (With permission from
the Massachusetts Medical Society.)

ume could be increased as compared to that of normal

subjects despite relative arterial underfilling caused by peripheral arterial vasodilation.
Afferent Volume Receptors

The afferent volume receptors must reside in the arterial

vascular compartment and predominate over low-pressure
receptors. While stimulation of the low-pressure volume
receptors of the thorax, including the atria, right ventricle,
and pulmonary vessels, results in enhanced release of atrial
natriuretic peptide (ANP) and suppression of arginine vasopressin (AVP), patients with advanced CHF exhibit avid
sodium and water retention despite the presence of elevated atrial pressures and increased circulating concentrations of ANP. Thus, the high-pressure baroreceptors in the
left ventricle, carotid sinus, aortic arch, and juxtaglomerular
apparatus have been implicated as the primary afferent
receptors involved in this volume regulatory system. Unloading of these ventricular and arterial receptors results
in diminution of the tonic inhibitory effect of afferent vagal
and glossopharyngeal neural traffic to the central nervous
system and initiates the nonosmotic release of vasopressin,
activation of the renin-angiotensin-aldosterone system, and
stimulation of the adrenergic nervous system. The activation of these three neurohumoral vasoconstrictor systems,
by altering renal hemodynamics and tubular function, mediates the avid renal sodium and water retention characteristic of patients with advanced CHF.

EFFERENT MECHANISMS INVOLVED IN THE SODIUM

AND WATER RETENTION OF CHF

As noted, the changes in renal hemodynamics and tubular sodium and water handling that occur in CHF are primarily mediated by activation of neurohormonal systems.
In this regard, the previously mentioned major neurohormonal vasoconstrictor systems predominate in advanced
or decompensated CHF. However, vasodilating hormones
such as the natriuretic peptides and renal prostaglandins
may play important counterregulatory roles. The renal effects of activation of these neurohormonal systems in CHF
are reviewed below and are summarized in Tables 1 and 2.
Changes in Renal Hemodynamics in CHF

The glomerular filtration rate (GFR) is usually normal

in mild heart failure and is reduced only as cardiac performance becomes more severely impaired. Renal vascular
resistance is increased with a concomitant decrease in renal
blood flow (RBF). Generally, RBF decreases in proportion
to the decrease in cardiac performance. Therefore the filtration fraction, the ratio of GFR to RBF, is usually increased in patients with cardiac failure. This increased
filtration fraction is a consequence of predominant
constriction of the efferent arterioles within the kidney.
This efferent arteriolar vasoconstriction is mediated by in creased renal adrenergic nerve activity and angiotensin II.
These changes in renal hemodynamics alter the hydrostatic
and oncotic forces in the peritubular capillaries to favor

24. Renal Function in Congestive Heart Failure

185

TABLE I

TABLE 2

Renal Effects of Neurohormonal Vasoconstrictors in

Congestive Heart Failure

Renal Effects of Hormonal Vasodilators in Congestive

Heart Failure

Renal nerves
Promote efferent greater than afferent arteriolar constriction
Enhance sodium reabsorption in the proximal tubule
Stimulate renal renin release
Angiotensin II
Promotes efferent greater than afferent arteriolar constriction
Enhances sodium reabsorption in the proximal tubule
Stimulates adrenal aldosterone synthesis and release
Aldosterone
Enhances sodium reabsorption and potassium and hydrogen
ion secretion in the collecting duct
Arginine vasopressin
Increases water reabsorption in the cortical and medullary
collecting duct Increases sodium chloride reabsorption
in the medullary
ascending limb of Henle's loop
From Hosenpud JD, Greenberg BH (eds): Congestive Heart
Failure: Pathophysiology, Diagnosis, and Comprehensive Approach to Management. Springer-Verlag, New York, pp. 161-173,
1994. With permission.

increased proximal tubular reabsorption of sodium and

water.
Renal Effects of Neurohormonal Vasoconstrictor
Activation in CHF
As mentioned, increased renal adrenergic nerve activity
and angiotensin II cause efferent arteriolar vasoconstriction which increases the filtration fraction and moves the
balance of hemodynamic forces in the peritubular capillaries in favor of enhanced proximal tubular sodium reabsorp tion. In addition to this indirect effect to increase sodium
reabsorption in the proximal tubule, both renal adrenergic
nerves and angiotensin II exh ibit a direct effect in the
proximal tubule to enhance sodium reabsorption. Renal
nerves also stimulate renal renin release which contributes
to the increase in circulating angiotensin II concentration
seen in CHF, and angiotensin II stimulates adrenal aldosterone synthesis and release. The latter enhances sodium
reabsorption in the collecting duct. In support of a role for
increased renal adrenergic nerve activity in the sodium and
water retention of CHF is the observation that intrarenal
adrenergic blockade results in a natriuresis in experimental
animals and humans with CHF. For example, in rats with
CHF, the intravenous administration of the central alpha 2
receptor agonist, clonidine, results in a marked decrease
in renal adrenergic nerve activity and a concomitant increase in urinary volume excretion.
A role for the renin-angiotensin-aldosterone system in
the sodium retention of CHF is suggested by the observation that urinary sodium excretion inversely correlates with
plasma renin activity and urinary aldosterone excretion
in patients with CHF. However, the administration of an
angiotensin converting enzyme (ACE) inhibitor to patients

Natriuretic peptides
Increases glomerular filtration rate
Promotes diminished sodium reabsorption in the collecting
duct
Suppresses renin activity
Inhibits aldosterone synthesis and release
Possibly inhibits vasopressin release
Renal prostaglandins Promote renal vasodilation Decrease
tubular sodium reabsorption in the ascending limb
of Henle's loop Inhibit vasopressin hydroosmotic action in
the collecting duct
From Hosenpud JD, Greenberg BH (eds): Congestive Heart
Failure: Pathophysiology, Diagnosis, and Comprehensive Approach to Management. Springer-Verlag, New York, pp. 161-173,
1994. With permission.

with CHF does not consistently increase urinary sodium

excretion despite a consistent decrease in plasma aldosterone concentration. The simultaneous fall in blood pressure
due to decreased circulating concentrations of angiotensin
II or an excessive fall in glomerular capillary pressure due
to decreased angiotensin II and increased bradykinin may
obscure the natriuretic response to ACE inhibition. The
observation that furosemide-induced natriuresis is augmented by very low doses but not by standard doses of
an ACE inhibitor is compatible with this hypothesis. In
addition, an important role for aldosterone in the sodium
retention of CHF has been demonstrated using the specific
aldosterone antagonist, spironolactone.
Arginine vasopressin increases water reabsorption in the
cortical and medullary collecting duct and increases sodium
chloride reabsorption in the medullary ascending limb of
Henle's loop. Two lines of evidence implicate the nonosmotic release of AVP in the abnormal water retention of
CHF. In experimental CHF, the absence of a pituitary
source of AVP is associated with normal or near normal
water excretion. In addition, the use of investigational antagonists to the renal, or V2, receptor of AVP may reverse
the abnormal water retention in experimental and human CHF.
Renal Effects of Hormonal Vasodilator Activation in CHF
The natriuretic peptides, ANP and brain natriuretic peptide (BNP), increase GFR, promote diminished sodium
reabsorption in the collecting duct, and suppress renin activity and aldosterone synthesis and release. The tubular
sites of action for these neurohormonal agents are shown
in Fig. 2. Experimental observations suggest that the natriuretic peptides play an important role in attenuating the
renal sodium and water retention associated with CHF.
Circulating concentrations of ANP and BNP are elevated
in CHF; however, the natriuretic effect of both endogenous
and exogenous natriuretic peptides is often blunted in CHF

186

William T. Abraham and Robert W. Schrier

M M U I b

Renal Nerves r /

Ang II

UlJUf

xAldo

Aldo
r

K* H*

T T
Na* H20
V

j /

Cortex

K*-

H,0^.

Medulla

v. _______ s
Collecting
Loop of Henle

E2 results in a significant reduction in urinary sodium excretion and suggests a possible counterregulatory role for vasodilating prostaglandins in CHF. Moreover, it has been
well documented that the administration of a cyclooxygenase inhibitor to patients with CHF may result in acute
reversible renal failure, an effect that is thought to be due
to inhibition of renal prostaglandins.

-*-^-

IMPLICATIONS FOR MEDICAL THERAPY OF CHF

-AVP

H 2 0 ^ l -AVP
^
______ ) 1 C
Na*
ANP/BNP

Duct

FIGURE 2 Sites of action of some important neurohormonal systems which alter the tubular handling of sodium and water in
CHF. Renal nerves and angiotensin II (Ang II) enhance sodium
reabsorption in the proximal tubule. Aldosterone (Aldo) increases
sodium reabsorption and promotes potassium and hydrogen ion
secretion in the distal nephron. Arginine vasopressin (AVP) acts to
increase water reabsorption in the cortical and medullary collecting
duct. Atrial (ANP) and brain (BNP) natriuretic peptides inhibit sodium reabsorption in the collecting duct. From Hosenpud
3D, Greenberg BH (eds): Congestive Heart Failure: Pathophysiology, Diagnosis, and Comprehensive Approach to Management.
Springer- Verlag, New York, pp. 161-173,1994. With permission.

patients. Based on the observation of a positive correlation

among endogenous ANP levels, urinary cyclic guanosine
monophosphate (cGMP, the second messenger for the natriuretic effect of ANP and BNP in vivo), and cardiac filling
pressures, it has been suggested that the natriuretic peptide
resistance seen in CHF patients may be due to decreased
sodium delivery to the distal tubular site of natriuretic
peptide action rather than a direct impairment of the
receptor-signal transduction system involving cGMP. This
possibility may be analogous to the impairment in aldosterone escape seen in patients with CHF and other chronic
sodium-retaining states. This diminished distal tubular sodium delivery hypothesis of natriuretic peptide resistance
is also supported by the observation that maneuvers which
increase distal tubular sodium delivery (e.g., low-dose mannitol, renal denervation, and angiotensin II blockade) restore renal responsiveness to ANP in CHF and in other
edematous disorders. In patients with advanced CHF the
natriuretic response to BNP correlates most strongly with
the change in distal tubular sodium delivery during BNP
infusion. Other mechanisms, such as intracellular factors
acting beyond cGMP, may also contribute to natriuretic
peptide resistance in CHF.
Renal prostaglandins promote renal vasodilation, decrease sodium reabsorption in the ascending limb of
Henle's loop, and inhibit AVP hydroosmotic action in the
collecting duct. The administration of acetylsalicylic acid
to patients with moderate CHF and a normal sodium intake
in doses that decrease the synthesis of renal prostaglandin

Medical therapy of CHF with diuretics and direct-acting

vasodilators may further activate neurohormonal vasoconstrictor systems and diminish circulating concentrations of
ANP. The emergence of diuretic resistance or vasodilator
tolerance in patients with chronic heart failure may be due,
in part, to further activation of vasoconstrictor mechanisms
induced by these therapeutic agents. For example, the administration of the loop diuretic, furosemide, increases
plasma norepinephrine, renin activity, and aldosterone and
decreases plasma ANP in patients with acute or chronic
heart failure. These neurohormonal effects of diuretic therapy may contribute to the development of diuretic resis tance, because increased renal nerve activity and angiotensin II enhance proximal tubular sodium reabsorption
thereby obscuring the beneficial effect of a diuretic that
acts in the loop of Henle or in a more distal segment of
the nephro n. Continuous vasodilator therapy with nitroglycerin results in the rapid development of drug tolerance
and in fluid retention, occurring simultaneously with activation of the renin-angiotensin-aldosterone system. That this
nitrate tolerance is prevented by the concomitant adminis tration of an ACE inhibitor further implicates drug-induced
activation of neurohormonal vasoconstrictor mechanisms
in the fluid retention associated with these agents.
This further activation of neurohormonal vasoconstrictor systems in response to diuretic or nitrate administration
is compatible with the aforementioned arterial underfilling
hypothesis in CHF. Diuretic- or vasodilator-induced reductions in cardiac preload, although effective in decreasing
the congestive signs and symptoms of CHF, move the myocardial length-tension relationship in a direction that further diminishes ventricular performance. In contrast, the
administration of a positive inotrope to patients with CHF
results in an upward and leftward shift of the Starling curve
and an increase in ventricular performance which may restore the integrity of the arterial circulation and improve
the neurohormonal milieu.
These observations suggest that inhibition of vasoconstrictor mechanisms by neurohormonal antagonists or positive inotropes may be more beneficial than or additive to
nonspecific diuretic or vasodilator therapy in CHF. The
proven beneficial effects of ACE inhibition on symptoms,
hemodynamics, and survival in heart failure patients support the routine use of these agents in CHF. However, as
mentioned above, the administration of an ACE inhibitor
during heart failure does not consistently increase urinary
sodium excretion. Adrenergic-blocking agents, renin antagonists, angiotensin II antagonists, AVP antagonists, natriuretic peptides, vasodilating prostaglandins, and a vari-

187

24. Renal Function in Congestive Heart Failure

ety of positive inotropes are all under active investigation

in heart failure patients. A recent study in human heart
failure suggests that angiotensin II receptor antagonism
may exert a more favorable effect on the kidney when
compared to ACE inhibition. However, this observation
awaits confirmation by large clinical trials. V2 receptor AVP
antagonists, now in phase I-II clin ical development in the
United States, are promising agents for reversing the dilutional hyponatremia of CHF.
Low or "renal" doses of dopamine (1-2 /j.g kg"1 min"1)
may improve renal blood flow and GFR, promote natriuresis via a direct effect on the proximal tubule, and facilitate
diuresis in some patients with advanced decompensated
CHF. However, the indiscriminate use of dopamine in patients with heart failure should be avoided, because largescale randomized clinical efficacy trials have not been performed. In a careful study of six diuretic-sensitive CHF
patients, dopamine did not improve the responsiveness to
furosemide. Thus, the empiric use of dopamine should
probably be reserved for patients with decompensated
CHF and some degree of diuretic resistance. In this setting,
urine output should be followed prior to and during dopamine infusion to assess its clinical efficacy and indication
for continued administration.
Finally, the pharmacologic management of CHF is fre quently complicated by the coexistence of chronic renal
failure. The use of ACE inhibitors in such patients may be
complicated by altered pharmacokinetics and worsening
renal dysfunction. The former concern applies to all but
one of the ACE inhibitors and may be overcome by modest
dose reduction. The latter complication is seen most frequently in the setting of advanced heart failure, where
preservation of glomerular capillary pressure may be particularly dependent on angiotensin II-induced efferent arteriolar constriction and maintenance of systemic blood pres sure. However, with careful drug titration, many patients

with heart failure and chronic renal insufficiency will tolerate ACE inhibition. As the GFR falls below 30 mL/min, the
efficacy of distal tubular diuretics (thiazides and potassiumconserving agents) is lost. Thus, loop diuretics and metolazone are preferred in such patients. Finally, digoxin intoxication is common in patients with coexistent cardiac and
renal disease. Because digoxin is eliminated primarily via
the kidneys, dose adjustment is necessary as the GFR falls.
Bibliography
Abraham WT: New neurohormonal antagonists and natriuretic
peptides in the treatment of congestive heart failure. Coronary
Artery Dis. 5:127-136, 1994.
Abraham WT: Physiology and therapeutic implications of natriuretic peptides in chronic heart failue. Heart Failure 12:5572, 1996.
Abraham WT, Schrier RW: Renal salt and water handling in
congestive heart failure. In: Hosenpud JD, Greenberg JH (eds)
Congestive Heart Failure: Pathophysiolo gy, Differential Diagnosis, and Comprehensive Approach to Therapy. Springer- Verlag,
New York, pp. 161 -173, 1993.
Abraham WT, Schrier RW: Use of furosemide in the treatment
of congestive heart failure. In: Aspects of Diuretic Therapy with
Furosemide: An Update. International Clinical Practice Series.
Wells Medical Ltd., Kent, pp. 23-33, 1993.
Abraham WT, Schrier RW: Cardiac failure, liver disease, and
nephrotic syndrome. In: Schrier RW, Gottschalk CW. (eds)
Diseases of the Kidney. 6 th ed. Little, Brown Co., Boston, pp.
2353-2392, 1996.
Schrier RW. Pathogenesis of sodium and water retention in high output and low-output cardiac failure, nephrotic syndrome, cirrhosis, and pregnancy. N Engl J Med 319:1065-1072 (Part 1);
1127-1134 (Part 2), 1988.
Schrier RW: Bo dy fluid volume regulation in health and disease:
A unifying hypothesis. Ann Intern Med 113:155 -159, 1990.
Schrier RW. A unifying hypothesis of body fluid volume regulation: The Lilly Lecture 1992. J Royal Coll Physicians London
1992; 26:295-306.
Schrier RW: An odyssey into the milieu interieur: Pondering the
enigmas. / Am Soc Nephrol 2:1549-1559, 1992.

RENAL FUNCTION IN LIVER DISEASE

MURRAY EPSTEIN

proposed that ascites formation in cirrhotic patients begins

when a critical imbalance of Starling forces in the hepatic
sinusoids and splanchnic capillaries causes an excessive
amount of lymph formation, which exceeds the capacity of
the thoracic duct to return it to the circulation. Consequently, excess lymph accumulates in the peritoneal space
as ascites. Simultaneously, vasodilatation and opening of
arteriovenous shunts expands the space in which the plasma
volume is contained; renal sodium retention is activated
and the relative disparity between plasma volume and the
circulatory space is perceived as "effective" plasma volume
contraction. The renal retention of sodium may be successful in restoring absolute total plasma volume to a normal
or even supranormal value; however, because of the enlargement of the circulatory space in which this volume is
contained, a state of relative or "effective" plasma volume
contraction persists. This may explain the apparent paradox
of unremitting renal sodium retention even with an apparently expanded absolute plasma volume. The term "effective" plasma volume refers to the part of the total circulating volume that is effective in stimulating volume receptors.
The concept is elusive because the actual volume receptors
remain incompletely defined. The diminished effective volume is thought to constitute an afferent signal that triggers
events leading to augmentation of salt and water reabsorption by the renal tubule. Thus, the traditional formulation
implies that the renal retention of sodium is a secondary
rather than primary event.
In contrast, the "overflow" hypothesis for ascites formation suggests that the primary event is the inappropriate
retention of excessive sodium by the kidney with a resultant
expansion of plasma volume. In the setting of abnormal
Starling forces (both portal venous hypertension and a
reduction in plasma colloid osmotic pressure) in the portal
venous bed and hepatic sinusoids, the expanded plasma
volume is sequestered preferentially in the peritoneal space
with ascites formation. Thus, according to this formulation,
renal sodium retention and plasma volume expansion precede rather than follow the formation of ascites.
A third hypothesis to account for renal sodium and water
retention in cirrhosis has been proposed. This theory,
termed the "peripheral arterial vasodilation hypothesis,"
postulates that peripheral vasodilation is the initial determinant of intravascular underfilling, i.e., that an imbalance
between an expanded capacitance and the available volume
constitutes a diminished effective volume. This "peripheral

A variety of alterations in renal function and electrolyte

metabolism frequently accompany liver disease. These diverse complications vary from those that have little clinical
significance to others that constitute serious derangements
requiring therapeutic intervention. This chapter emphasizes abnormalities of renal sodium and water handling,
and on the hepatorenal syndrome.
RENAL SODIUM HANDLING
Clinical Features

Patients with Laennec's cirrhosis manifest a remarkable

capacity for sodium chloride retention; indeed, they frequently excrete urine that is virtually free of sodium.
Extracellular fluid accumulates excessively and eventually
becomes manifest as clinically detectable ascites and
edema. It should be emphasized that cirrhotic patients
who are unable to excrete sodium will continue to gain
weight and accumulate ascites and edema as long as
dietary sodium content exceeds maximal urinary sodium
excretion. If access to sodium is not curtailed, the relentless retention of this ion may lead to the accumulation
of vast amounts of ascites (on occasion up to 20 L).
Weight gain and ascites formation promp tly cease when
sodium intake is limited.
The abnormality of renal sodium handling in cirrhosis
is not a static and unalterable condition. Rather, patients
with cirrhosis may undergo a spontaneous diuresis followed by a return to avid salt retention. While a significant
number of patients who are maintained on a sodiumrestricted dietary program may undergo a spontaneous
diuresis, there is inadequate information concerning the
incidence of this phenomenon and the predictability of
its occurrence in any part icular patient. The primary
renal excretory abnormality causing fluid retention is a
disturbance of sodium rather than water excretion. Many
sodium-retaining patients with ascites and edema can
excrete large volumes of dilute urine when given excessive
amounts of water without sodium.
Pathogenesis

The pathogenetic events leading to the deranged sodium

homeostasis of cirrhosis are exceedingly complex and remain the subject of continuing controversy. At least two
hypotheses have been advanced. Traditionally, it has been

Copyright 1 998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

Primer on Kidney Diseases, Second Edition

1 88

25. Rena l Function in Liver Disease

vasodilatation" theory is not a separate hypothesis, but

rather represents a revision of the "underfill" theory.
Efferent Factors
Initial attempts to explain the renal abnormalities responsible for sodium retention focused on the decrement
in glomerular filtration rate (GFR) that frequently occurs
in patients with advanced liver disease. A number of observations indicated, however, that a decrease in GFR cannot
constitute the major determinant of the abnormalities in
renal sodium handling, because sodium retention occurred
despite normal or supranormal GFR. Thus, renal sodium
retention accompanying cirrhosis is due primarily to enhanced tubular reabsorption rather than to alterations in
the filtered load of sodium.
The mediators for the enhanced tubular reabsorption of
sodium in cirrhosis and their relative role in this process
have not been elucidated completely. Several mechanisms
have been suggested, including (1) hyperaldosteronism,
(2) alterations in intrarenal blood flow distribution, (3) an
increase in renal sympathetic nervous system activity,
(4) alterations in eicosanoids including a relative diminution of renal vasodilatory prostaglandins, (5) a relative impairment of renal kallikrein generation, (6) alterations in
atrial natriuretic peptide, and (7) modulation of a possible
humoral natriuretic factor.

MANAGEMENT OF ASCITES AND EDEMA

The approach to the cirrhotic patient with ascites should

be grounded on the realization that ascites, unless massive,
may not require treatment per se. The initial goal of any
treatment program should be an attempt to obtain spontaneous diuresis by consistent and scrupulous adherence to
rigid dietary sodium restriction (500 mg per day). While
the frequency with which such dietary management successfully relieves ascites is unsettled, if feasible such a diet
should be prescribed to all patients since it is impossible
to predict which patients will respond. When the response
to dietary management is inadequate or when the imposition of rigid dietary sodium restriction is not feasible because of cost or unpalatability of the diet, diuretic agents
should be prescribed.
If there is no compelling reason for rapidly mobilizing
excessive fluid, treatment should begin with one of the
distal potassium-sparing diuretics. It is reasonable to start
with spironolactone, 100 mg/day twice daily. If this dosage
of spironolactone does not induce a natriuresis, the dosage
may be increased in stepwise fashion every 3 to 5 days to
a maximum of 400 mg/day. Spironolactone's onset of action
is slow, requiring 3 to 5 days for the peak effect to become
manifest, and thus natriuresis will not occur during the
initial 2 to 3 days of therapy. If no natriuresis occurs with
the maximum dosage of spironolactone, one can add a loop
diuretic such as furosemide to the regimen.
Loop diuretics most be delivered into the lumen of the
renal tubule to produce a pharmacologic effect. Studies in
patients with cirrhosis disclose wide variability in response,

189
but some patients clearly have diminished ability to attain
sufficient drug in the urine to cause a response. Such patients may require 2 to 2 i times greater doses of loop
diuretics to attain "normal" amounts at the site of action.
Hence, a reasonable therapeutic strategy is first to adminis ter conventional doses of a loop diuretic (e.g., 40 mg of IV
furosemide = 80 mg of oral furosemide = 1 mg of iv or oral
bumetamide). If no response or an inadequate response
ensues, the dose may be increased by at most 2 i fold.
There appears to be no justification for higher doses. Such
a strategy should compensate for any changes in the disposition of loop diuretics in patients with liver disease and still
be safe. Because of recent trends emphasizing reduction
of hospital length of stay, many clinicians short-circuit this
approach by utilizing large-volume paracentesis.
Large-Volume Paracentesis
Because large-volume paracentesis may induce hypovolemia, hyponatremia, renal failure, and encephalopathy,
until recently its use in the treatment of refractory ascites
has been limited. Previous studies have demonstrated a
varied decrease in plasma volume following paracentesis.
Furthermore, it has been suggested that ascites per se may
have a negative influence on cardiac function. Recently,
several groups of investigators have reevaluated the effects
of large-volume (4 to 25 L per day until disappearance
of ascites) or total paracentesis (complete mobilization of
ascites in only one paracentesis session) and concluded
that it constitutes effective and relatively safe therapy for
mobilizing ascites. Extensive studies by the Barcelona
group have clearly established a role for large-volume paracentesis in the therapeutic armamentarium of refractory
ascites. Their experience and that of others would dictate
that there are relatively few side effects associated with
this procedure. The intravenous administration of albumin
or other plasma expanders is initiated at the end of the
procedure to obviate the adverse effects on renal function.
In cases of massive ascites treated by total paracentesis
(>8 L), half of the dose of plasma expanders should be
administered at the end of the procedure and half 6 hours
later to prevent excessively rapid plasma volume expansion, which may lead to variceal bleeding. In most patients,
peripheral edema rapidly reabsorbs following mobilization
of ascites and usually disappears within the first 2 days
after treatment.

RENAL WATER HANDLING IN CIRRHOSIS

Clinical Features
Impairment of renal diluting capacity occurs frequently
in cirrhosis. Hyponatremia, the expression of this impaired
capacity to excrete water, is a commonly encountered clinical problem in cirrhotic patients. Although this abnormality
may reflect the severity of the hepatic disease, the available
data suggest difficulty in correlating the impairment of maximal water diuresis with specific clinical features, such as
the degree of ascites or the level of jaundice.

190

Murray Epstein

The majority of patients with compensated liver disease

excrete water normally wherever those with decompensated hepatic illness manifest widely varying responses to
oral loading. Prospective studies have indicated that the
transition from compensation to decompensation, or vice
versa, is not necessarily accompanied by concomitant
changes in renal water handling.
Pathogenesis

The mechanism responsible for the impairment of water

diuresis in cirrhosis is not fully established. Four factors
have been proposed: (1) enhanced antidiuretic hormone
(ADH) activity; (2) a decreased delivery of filtrate to the
diluting segments of the nephron; (3) a decrease in the
release of prostaglandins; and (4) an increase in renal sympathetic nervous system activity.
Treatment of Hyponatremia

An impaired diluting ability has important management

implications for many patients with advanced liver disease.
Because hyponatremia connotes a dilutional state secondary to an impaired capacity to excrete water, fluid restriction constitutes the basis for treatment. Appropriate fluid
restriction will eventually repair this abnormality, regard less of the degree of dilutional impairment.
Unfortunately, this seemingly simple goal is often elusive. In many hospitals, adherence to the physician's orders
for fluid restriction is erratic. Because paramedical and
nursing personnel may not consider the fluids administered
with medication as part of fluid intake, patients who are
supposedly on rigid fluid restriction regimens may be receiving as much as 1500 and even 2000 mL fluid per day.
Because of this problem, it is sometimes necessary to resort
to absolute fluid restriction (0 to 200 mL/day) for the first
few days in order to initiate normalization of serum sodium.
Of interest, several vasopressin antagonists under development promise to become imp ortant adjuncts normal type
not hold in the management of refractory hyponatremia.
SYNDROMES OF ACUTE AZOTEMIA

Acute renal failure occurs with increased frequency in

patients with hepatic and biliary disease. While acute azotemia may often be due to classic acute renal failure, cirrhotic
patients can also develop a unique form of renal failure
called the hepatorenal syndrome.
Hepatorenal Syndrome

Progressive oliguric renal failure commonly complicates

the course of patients with advanced hepatic disease. While
this condition has been designated by many names including "functional renal failure" and "the renal failure of
cirrhosis," the more appealing, albeit less specific, term
"hepatorenal syndrome" (HRS) is widely applied. Hepatorenal syndrome may be defined as unexplained renal
failure occurring in patients with liver disease in the ab-

sence of clinical, laboratory, or anatomical evidence of

other known causes of renal failure.
Clinical Features

The clinical presentation and course of HRS are characterized by marked variability. HRS develops usually in
patients who have advanced chronic hepatic failure and
portal hypertension. Although this form of renal failure
can complicate the course of virtually all forms of severe
liver disease, most studies performed in the United States
have identified alcoholic cirrhosis as the underlying dis order.
Renal failure may develop with great rapidity, occasionally occurring in patients in whom normal GFR and concentrating ability were documented a few days before the onset
of HRS. Although HRS may follow events that reduce
effective blood volume, including abdominal paracentesis,
vigorous diuretic therapy, and gastrointestinal bleeding, it
can occur in the absence of an apparent precipitating factor.
Patients with cirrhosis often develop HRS following admis sion to the hospital, raising the question as to whether
events in the hospital might precipitate HRS. Generally,
HRS develops in patients who have ascites, which is often
tense, as well as other clinical stigmata of portal hypertension and chronic liver disease. Rarely, HRS can occur with
little evidence of severe hepatic dysfunction.
A substantial body of evidence lends strong support to
the concept that the renal failure in HRS is functional in
nature. Direct evidence is derived (1) from the demonstration that kidneys transplanted from patients with HRS are
capable of resuming normal function in the recipient and
(2) by the return of renal function when the patient with
HRS successfully receives a liver transplant. Laboratory
studies confirm the severity of hepatic disease. Prothrombin
time is prolonged, serum albumin concentration is reduced,
and bilirubin concentration is increased although the degree of jaundice is extremely variable. Most patients have
a modest decrease in systemic blood pressure. Urine
volume is reduced (<800 mL/day), oliguria is common
(<400 mL/day), but anuria is rare. The serum creatinine
and blood urea nitrogen (BUN) are elevated but seldom
approach the high values seen in patients with endstage
renal failure due to primary renal disease. The degree of
renal impairment is underestimated by our reliance on the
BUN and serum creatinine concentrations as indices of
renal function. These parameters are misleading because
BUN levels tend to be depressed in cirrhotics owing to
reduced protein intake from anorexia and limited hepatic
synthesis of urea. Similarly, endogenous creatinine production declines as muscle wasting develops in proteinmalnourished cirrhotics, thereby lowering serum creatinine
concentration at any level of creatinine clearance. The
urine is usually acid; mild proteinuria, granular casts, and
microscopic hematuria are not infrequent. The urine is
virtually free of sodium chloride (<10 mEq/L) and modestly concentrated (500 to 700 mOsm/kg H2O). Occasionally, urine sodium concentrations of 20 to 30 mEq/L may
be seen. This may be due to superimposed acute tubular

25. Renal Function in Liver Disease

191

necrosis (ATN) or acute metabolic alkalosis due to vomiting; in the former, urinary chloride would be commensurate
with that of sodium, while in the latter, urinary chloride
would be low. Toward the end of the clinical course, urinary
sodium excretion may increase and urine osmolality decrease toward isotonicity, probably due to superimposed
ATN. Urinary lysozyme, a marker of tubule cell injury, is
also low in contrast to the levels in typical ATN.
The prognosis of HRS is grave. The majority of patients
die within a few weeks of onset of azotemia, although 5
to 10% of the patients may survive several months of mild
azotemia. Complete recovery in those with advanced renal
failure is rare.

creatinine is <20:1 in ATN. Proteinuria is absent or minimal in HRS.

In summary, the finding of a low urinary sodium concentration in the presence of oliguric ATN precludes the diagnosis of ATN. Only when prerenal failure and ATN are
excluded can one establish the diagnosis of HRS.
Because HRS and prerenal azotemia have similar urinary
diagnostic indices, one must often use a functional maneuver, i.e., the administration of volume expanders, to differentiate between the two entities. In this regard, it should
be underscored that our frame of reference for the cirrhotic
patient may be quite different from that pertaining to other
disease states. The degree of volume expansion necessary
to replete the cirrhotic patient may at times be marked,
occasionally requiring the infusion of massive amounts
of colloid.

Differential Diagnosis

As has been discussed earlier, the abrupt onset of oliguria

in a cirrhotic patient may have other etiologies than HRS.
Prerenal causes are important to differentiate, particularly
because they constitute reversible conditions if recognized
and treated in the incipient phase. Volume contraction or
cardiac pump failure may present as a "pseudohepatorenal" syndrome. As already mentioned, it is common for
patients with alcoholic cirrhosis to develop classic ATN.
In many instances, the differentiation from HRS can be
made readily by recognition of the precipitating event and
characteristic laboratory findings. Table 1 lists laboratory
features helpful in differentiating the three principal causes
of acute azotemia in patients with liver disease. The most
common finding in the urine of HRS patients is a strikingly
low sodium concentration, usually less than 10 mEq/L, and
occasionally as low as 2 to 5 mEq/L. Similarly, prerenal
azotemia is associated with low urinary sodium concentrations. In contrast, patients with oliguric ATN fre quently have urinary sodium concentrations exceeding
30 mEq/L, sometimes higher. Both HRS and prerenal azotemia manifest well-maintained urinary concentrating ability characterized by a urine to plasma osmolality ratio (U/
Posm) >1.0, whereas ATN patients excrete an isoosmotic
urine. Urine to plasma creatinine ratio (U/P creatinine)
is >30 in both prerenal failure and HRS, whereas U/P

Treatment

The management of HRS is discouraging in view of the

absence of any reproducibly effective treatment modality.
Because knowledge today about the pathogenesis of HRS
remains inferential and incomplete, therapy must be primarily supportive. The initial step in management is not
to equate decreased renal function with HRS, but rather
to search diligently for and treat correctable causes of azotemia such as volume contraction, cardiac decompensation,
and urinary tract obstruction. The diagnosis of ATN should
be considered, since the cirrhotic patients with ATN may
recover, if supported with dialytic therapy.
Once the correctable causes of renal functional impairment are excluded, the mainstay of therapy is careful restriction of sodium and fluid intake. A number of specific
therapeutic measures have been attempted, but few have
proved to be of practical value. Attempts at volume expansion with different exogenous expanders result in only transient improvement in renal hemodynamics and function
without significant improvement in the outcome. Similarly,
attempts at reinfusion of ascites utilizing peritoneal fluid
which has been concentrated have not provided any lasting improvement.

TABLE I

Differential Diagnosis of Acute Azotemia in Patients with Liver Disease

Important differential urinary findings
Biochemical characteristics

Prerenal azotemia

Hepatorenal syndrome

<30:1

>30:1

At least 100 mOsm > plasma

osmolality
Normal

At least 100 mOsm > plasma

osmolality
Unremarkable

Urine sodium concentration (mEq/L)

Urine to plasma creatinine ratio
Urine osmolality
Urine sediment

Acute renal failure

(Acute tubular necrosis)
<30" <20:1 Equal to
plasma osmolality
Casts, cellular debris

" It has recently been appreciated that radiocontrast agents and sepsis may lower urinary sodium concentration in patients with acute
tubular necrosis.

192
In view of the prominent role assigned to renal cortical
ischemia in the pathogenesis of HRS, it is not altogether
surprising that numerous attempts to treat HRS with vasodilators have their made. Intrarenal infusion of nonspecific
vasodilators such as acetylcholine improve renal blood flow
but do not augment GFR. Similarly, blockade of vasoconstrictor a-adrenergic nerves by intrarenal infusion of phentolamine or phenoxybenzamine has no significant effect on
GFR. Infusions of vasodilator prostaglandins to correct a
possible renal prostaglandin deficiency have been unrewarding.
Dialysis has been reported to be ineffective in the management of HRS. Increasing experience, however, suggests
that such a sweeping condemnation should be qualified.
Although most of the early published literature indeed
suggests a dismal prognosis for patients who are dialyzed,
such reports have dealt with patients with chronic endstage liver disease. In a few instances, dialysis has been
undertaken with an ultimately favorable outcome in HRS
patients with acute hepatic disease. These observations indicate that in selected patients, i.e., patients with acute
hepatic dysfunction in whom there is reason to believe
that renal failure may reverse with resolution of the acute
hepatic insult, dialytic therapy is indicated. Recently, the
role of dialysis has been expanding as a supportive measure
for patients awaiting hepatic transplantation.
Because many patients undergoing liver transplantation
with or without acute renal failure exhibit hemodynamic
instability, treatment with continuous hemofiltration may
be better tolerated. Fluid removal may facilitate overall
management and allow liberal replacement of clotting factors and other fluids during the perioperative period. In
addition, continuous renal replacement therapy may help
control hepatic encephalopathy and the associated increases in intracranial pressure.
There has been a flurry of enthusiasm for the use
of the peritoneal jugular shunt (LeVeen shunt) in the
management of HRS. Because the underlying abnormality
is thought to be maldistribution of extracellular fluid with
a resultant diminished effective blood volume, attention
has focused on developing procedures that can redistribute body fluids between compartments, so that the central
compartment is replenished at a time when ascites is decreasing.
Only two prospective randomized studies of the role of
the PVS in the treatment of HRS have been performed.
The first prospectively compared the effects of the PVS
shunt (n = 10) or medical therapy (n = 10) on renal function and mortality in 20 patients with the HRS associated
with alcoholic liver disease. After 48 to 72 hours, body
weight and serum creatinine were increased with medical
therapy and decreased in patients with the shunt. Despite
improvement of renal function only one patient with the
PVS had a prolonged survival (210 days). In the second,
there were seven long-term survivals in a group of 14 patients treated with PVS, but the results were not statistically
significant when compared to those of a group of 19 patients
undergoing medical therapy. The mean half-life of patients
treated with the shunt did not differ significantly from that

Murray Epstein

of controls. Based on the available data, it is apparent that

a beneficial role for the PVS in the treatment of the HRS
has not been established.
Transjugular Intrahepatic Porto -systemic Shunt
Recently, a few anecdotal reports have appeared describing improvement in renal function in HRS patients following insertion of a transjugular intrahepatic porto-systemic
shunt (TIPS). The rationale for this procedure is similar
to that for the establishment of a side-to-side portacaval
shunt, thereby creating a portal to systemic vascular pathway that serves to decompress portacaval system. Although
TIPS obviates the need for performing major vascular surgery, it is not as simple and innocuous a procedure as
some of its adherents propose. It is operator-dependent,
requiring skilled and experienced interventional radiologists for its successful insertion. The available data supporting its use consist mainly of a few preliminary and anecdotal reports.
Orthotopic Liver Transplantation
Orthotopic liver transplantation (OLTX) has recently
become the accepted treatment for end-stage liver disease.
Of interest, many of these patients present with varying
degrees of concomitant renal dysfunction, including HRS.
OLTX has been reported to reverse HRS acutely. A large,
recent study reported a good, long-term survival with return of acceptable renal function for prolonged periods
after OLTX. Of More than 800 patients undergoing transplantation in Dallas, 59 had HRS. Of the patients surviving
liver transplant, four required subsequent renal transplant
for nonreturn of function. Of the remaining patients without hepatorenal syndrome (more than 650), two required
subsequent kidney transplant. These results and others
clearly indicate that patients with HRS have good survival
rates following liver transplant action, although the rates
are statistically inferior compared with patients without
HRS. With aggressive pre- and posttransplant management, one can anticipate excellent results following OLTX
in patients with HRS.
Acknowledgments
Portions of this chapter have been adapted from two earlier
reviews by the author: Epstein M: "Renal sodium handling in liver
disease" and "The hepatorenal syndrome". Both are in Epstein
M (ed) The Kidney in Liver Disease, 4th ed., Hanley & Belfus,
Philadelphia, 1996.

Bibliography
Epstein M: The LeVeen shunt for ascites and hepatorenal syndrome. N Engl J Med 302:628-630, 1980.
Epstein M: Derangements of renal water handling in liver disease.
Gastroenterology 89:1415-1425, 1989.
Epstein M: Atrial natriuretic factor in patients with liver disease.
Am J Nephrol 9:89-100, 1989.
Epstein M: Renal effects of head-out wat er immersion in humans:
A 15-year update. Physiol Rev 72:563-621, 1992.

26. Poststreptococcal and Infection-Related GN

Epstein M, Berk DP, Hollenberg NK, et al.: Renal failure in the
patient with cirrhosis: the role of active vasoconstriction. Am J
Med 49:175-185, 1970.
Epstein M: Hepatorenal syndrome. In: Epstein M (ed.) The Kidney
in Liver Disease. 4th ed. Hanley & Belfus, pp. 75-108, 1996.
Epstein M. Renal sodium handling in liver disease. In: Epstein M
(ed.) The Kidney in Liver Disease. 4th ed. Hanley & Belfus, pp.
3-31, 1996.
Gines P, Arroyo V, Vargas V, et al.: Paracentesis with intravenous
infusion of albumin as compared with peritoneovenous shunting
in cirrhosis with refractory ascites. N Engl J Med 325:829835, 1991.
Gonwa TA, Wilkinson AH: Liver transplantation and renal function: Results in patients with and without hepatorenal syndrome.
In: Epstein M (ed.) The Kidney in Liver Disease. 4th ed. Philadelphia, Hanley & Belfus, pp. 529-542, 1996.
Levy M: Pathophysiology of ascites formation . In: Epstein M (ed.)
The Kidney in Liver Disease. 4th ed. Philadelphia, Hanley &
Belfus, pp. 179-220, 1996.

193
Linas SL, Schaffer JW, Moore EE, Good JT Jr, Giansiracusa R:
Peritoneovenous shunt in the management of the hepatorenal
syndrome. Kidney Int 30:736-740, 1986.
Perez GO, Epstein M, Oster JR: Dialysis, hemofiltration, and
other extracorporeal techniques in the treatment of the renal
complications of liver disease. In: Epstein M (ed.) The Kidney
in Liver Disease. 4th ed. Philadelphia, Hanley & Belfus, pp.
517-528, 1996.
Stanley MM, Ochi S, Lee KK, et al.: Peritoneovenous shunting
as compared with medical treatment in patients with alcoholic
cirrhosis and massive ascites. N Engl J Med 321:1632-1638,
1989.
Vaamonde CA. Renal water handling in liver disease. In: Epstein
M (ed.) The Kidney in Liver Disease. 4th ed. Philadelphia, Hanley & Belfus, pp. 33-74, 1996.
Somberg KA. Transjugular intrahepatic portosystemic shunt in
the treatment of refractory ascites and hepatorenal syndrome.
In: Epstein M (ed.) The Kidney in Liver Disease. 4th ed. Philadelphia, Hanley & Belfus, pp. 507-516, 1996.

POSTSTREPTOCOCCAL AND OTHER

INFECTION-RELATED
GLOMERULONEPHRITIDES
ASHA MOUDGIL, ARVIND BAGGA, RAGINI FREDRICH, AND STANLEY C. JORDAN

Glomerulonephritis (GN) is characterized by proliferation and inflammation of the glomeruli secondary to immune complex deposition within the glomerular capillary
walls and the mesangium. GN may occur as a consequence
of infections or autoimmune diseases or may be idiopathic.
This chapter focuses on GN resulting from infections. Nephritogenic antigens derived from infectious agents may
either bind directly to glomerular sites forming in situ immune complexes with cognate antibody, or circulating immune complexes may become trapped in the glomeruli.
Either mechanism initiates inflammation of the glomeruli
through complement activation and chemotaxis of polymorphonuclear leukocytes. Immunochemical analysis of
immune complexes isolated from affected glomeruli has

Primer on Kidney Diseases, Second Edition

demonstrated antigenic structures specific for infectious

agents.
The onset of clinical symptoms may be acute, with hypertension, edema, gross hematuria, variable degrees of proteinuria, oliguria, and azotemia. Some patients may have
an insidious onset of edema, hematuria, proteinuria, and
renal insufficiency. Patients presenting with features of
acute GN (AGN) often have evidence of a recent infection
of the pharynx or skin with group A j3-hemolytic streptococci. However, other bacterial, viral, and parasitic agents
have also been implicated in the pathogenesis of GN (Table
1). In patients with GN associated with infections other
than group A j8-hemolytic streptococci, the onset is often
insidious and the course protracted. Some of these patients
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

194

Moudgil, Bagga, Fredrich, and Jordan

TABLE I

Infectious Agents Associated with Immune

Complex Glomerulonephritis

Bacterial
Group A /3-hemolytic streptococci
Staphylococcus aureus
Staphylococcus epidermidis Gramnegative bacilli Streptococcus
pneumoniae Treponema pallidum
Salmonella typhi Meningococcus
Leptospirosis
Viral
Hepatitis B and C Cytomegalovirus
Enteroviruses Measles Parvovirus
Oncornavirus Mumps virus Rubella
Varicella

Parasitic
Plasmodium malariae
Toxoplasma Filaria
Schistosomia
Trichinella
Trypanosome
Rickettsial Scrub
typhus
Fungal
Coccidioides immitis

may develop chronic renal failure. In developed countries,

GN associated with nonstreptococcal infections is assuming
greater importance. This is thought to be secondary to a
decline in the incidence of group A streptococcal infections
and a relative increase in the incidence of other infectionrelated glomerulonephritides in alcoholics and intravenous
drug abusers.
POSTSTREPTOCOCCAL GLOMERULONEPHRITIS

Poststreptococcal glomerulonephritis (PSGN) is the

most common form of GN in children. If follows either
pharyngeal or skin infection with group A /3-hemolytic
streptococci. Streptococcal pharyngitis occurs more commonly during the winter and early spring, whereas impetigo
is more common in summer. Only certain serologic types
of streptococci, known as "nephritogenic strains," are associated with PSGN. The nephritogenic strains associated
with pharyngeal infections include protein M types 1, 3, 4,
6,12, 25, and 49 and those with skin infections types 2, 49,
55, 57, and 60. Serotype 49 is the most common strain
associated with PSGN.
The occurrence of PSGN following infection with a nephritogenic strain is variable and dependent on environmental, host, and genetic factors. The risk of developing
AGN is 10 to 15% during epidemics of infection with these
strains. Subclinical episodes occur 4 to 10 times more frequently than overt clinical disease. Asymptomatic contacts
of patients with PSGN may have abnormal urinary findings,
most commonly hematuria.
Clinical Features
Poststreptococcal glomerulonephritis is primarily a dis ease of children, mostly occurring between 5 and 15 years

of age; it is rare under 2 years and over 40 years. A male

preponderance (2:1) is reported. The onset of GN is usually
abrupt and preceded by streptococcal pharyngitis or pyoderma by 1 to 4 weeks. This latent period is 7 to 14 days for
the postphary ngitic and 14 to 28 days for the postpyoderma
forms of GN.
Hematuria and edema are the most common complaints
(Table 2). Gross hematuria, classically described as colacolored, smoky, reddish brown urine, is seen in 70% and
microscopic hematuria in virtually all patients. Typically
the edema is periorbital and worse in the morning. Weight
gain may occur without evidence of edema. The etiology
of fluid accumulation is multifactorial; contributing factors
include reduction in glomerular filtration rate, sodium and
water retention, and, to a lesser degree, proteinuria. Most
patients with PSGN have oliguria. Anuria is infrequent and
if persistent suggests rapidly progressive GN.
Hypertension is usually mild to moderate and most likely
secondary to increased sodium and water retention. The
hypertension may be severe enough to cause hypertensive
encephalopathy with headache, visual disturbances, altered
sensorium, coma, or convulsions. Serious neurologic manifestations have been described even in the absence of severe hypertension. Circulatory congestion secondary to
expansion of the extracellular fluid volume is present in the
majority of patients and may result in dyspnea, orthopnea,
cough, cardiomegaly, gallop rhythm, pleural effusions, and
pulmonary edema. In such cases, an incorrect diagnosis of
primary cardiovascular disease may be made if examination
of the urinary sediment is omitted. Pallor frequently develops in patients with PSGN, secondary to dilutional anemia
and edema. Systemic symptoms such as mild fever, nausea,
vomiting, and abdominal pain may be present, especially
in children.
Patients with PSGN may not have all the clinical features
mentioned above. Therefore, the diagnosis of PSGN should
be considered in patients with sudden unexplained onset
of edema, oliguria, hypertension, or encephalopathy. Subclinical cases may be discovered on examination of the
urine sediment during investigation of family contacts of
patients with overt disease.
Because there is a latent period between streptococcal
pharyngeal infection and the onset of hematuria, PSGN is

TABLE 2

Clinical Manifestations of Acute

Poststreptococcal Glomerulonephritis
Common
Hematuria
Edema
Hypertension
Oliguria
Uncommon Anuria and
renal failure

Variable
Circulatory congestion Encephalopathy
Headache
Somnolence
Convulsions Systemic symptoms
Mild fever

Nausea
Abdominal pain
Pallor

26. Poststreptococcal and Infection -Related GN

often referred to as "postpharyngitic nephritis." Viral and

bacterial upper respiratory infections may trigger inflammation in an already damaged glomerulus resulting in concomitant, or "synpharyngitic hematuria." This is commonly
observed with IgA nephropathy, Henoch-Schonlein purpura, familial nephritis, and membranoproliferative GN.
Synpharyngitic nephritis can also occasionally be seen with
streptococcal infections and represents an exacerbation of
an underlying chronic renal disease. In these patients, hematuria abates rapidly, in contrast to PSGN where it may
last for more than 1 year.

195

Blood Chemistry
Serum creatinine levels are usually normal, although
blood urea nitrogen may be increased disproportionately.
Renal tubular function including the urine concentrating
ability is usually preserved. The fractional excretion of sodium is less than 0.5% during the acute phase of the illness.
Dilutional hyponatremia and hyperchloremic metabolic
acidosis with mild hyperkalemia, consistent with the diagnosis of type IV renal tubular acidosis, may occur. Renal
tubular acidosis is considered to occur from suppression
of plasma renin activity and aldosterone secretion by an
expanded extracellular volume. Serum albumin is often
reduced because of dilution and urinary losses of protein.

bears no relation to the development of nephritis or its

severity. In streptococcal skin infections, binding of the
streptolysin to skin lipids may blunt the immunologic response to this antigen, resulting in a low ASO liter. In
these siluations, ADNase B and AH are more specific and
positive in 90% of Ihe cases. Antibodies againsl endoslreplosin (an antigen derived from nephrilogenic streptococci)
are usually increased for a prolonged duration in PSGN.
The Streptozyme lesl, a slide agglutination tesl, delecls
antibodies againsl erythrocytes coaled wilh a mixlure of
slreplococcal exlracellular antigens (slreplolysin O, hyaluronidase, DNase B, slreplokinase, and NADase). II has
nol proved lo be a reliable subslilule for ASO and ADNase
B because of a high rale of false positives and negatives.
In 1986, Ihe World Heallh Organization advised againsl
ils use.
Depression of hemolylic complemenl aclivily (CH 50)
and C3 levels is seen in 90 to 100% of cases of active PSGN.
Similar depression of C3 is seen in siblings of palienls wilh
PSGN despite absence of hematuria. The early complemenl componenls (Clq, C2, C4) may also be mildly reduced, initially. Serum properdin and C5 levels are decreased in 60% of palienls, usually in association wilh
reduced C3 levels, suggesting activation of Ihe alternative
complement pathway. C3 nephritic factor activity may be
detected initially bul becomes undeleclable wilh normalization of C3 levels. There is no correlation between the
depression of C3 level and severity of nephritis. C3 levels
relurn lo normal wilhin 6 lo 8 weeks after Ihe onsel of
dis ease in more lhan 90% of palienls. Persislenlly low C3
levels suggesl underlying membranoproliferative GN, lupus nephritis, or GN related lo endocarditis or occull vis ceral abscess.
Circulating cryoglobulins and immune complexes may
be found in most patienls in Ihe firsl few weeks of Ihe
illness. High-molecular-weighl fibrinogen complexes are
presenl in severe cases. Circulating antibodies lo basement
membrane consliluents (type IV collagen and laminin) may
also be found.

Cultures for Streptococci

Despite a latent period between the streptococcal infection and PSGN, throat or skin cultures may show group A
streptococci. Family contacts of patients may have positive
cultures even in the absence of nephritis. Because about
20% of children are asymptomatic carriers of streptococci
in the nasopharynx, a positive throat culture cannot be
used as the sole criterion to diagnose PSGN.

Hematology
The normocylic normochromic anemia is usually dilulional ralher lhan secondary lo hematuria or hemolysis.
Shortened erylhrocyte survival, secondary to rapid elimination of red cells coated with immune complexes, has also
been described. Thrombocylopenia is rarely reported. Fibrinogen, factor VIII, and plasmin are elevated acutely and
are Ihoughl lo correlate wilh disease aclivily.

Serologic Findings
Antibodies against extracellular products of streptococci
constitute indirect evidence of recent infection. These
antibodies include antistreptolysin O (ASO), antistreptokinase, antihyaluronidase (AH), anlideoxyribonucle ase B (ADNase B) and antinicotyladenine dinucleotidase
(NADase). The ASO tilers begin to rise 10 to 14 days after
streptococcal pharyngitis, peak at 3 to 4 weeks, and then
decrease. Early antibiotic treatment can blunt the rise in
titers of these antibodies. The magnitude of ASO liters

Indications for Renal Biopsy

Laboratory Findings
Urinalysis
Hematuria is present in virtually all patients and may be
gross or microscopic. The majority of red blood cells in
the urinary sediment are dysmorphic. Red cell casts are
present in fresh urine in 60 to 85% of hospitalized patients.
Leukocyte casts, tubular epithelial cell casts, and granular
casts are common. Proteinuria is often present, usually less
than 2 g irr2 day"1; its severity often parallels that of
hematuria. In 30% of the patients, proteinuria may be in
the nephrotic range. Proteinuria is nonselective and often
contains fibrin degradation products.

Renal biopsy is rarely indicated in PSGN, bul should be

considered in patients with: (1) atypical presentation such
as prolonged oliguria, anuria, severe proteinuria, marked
azolemia, and no serologic evidence of preceding slreplococcal infection; (2) significant hypertension and macroscopic hematuria for 2 to 3 weeks or persistent proteinuria
wilh or wilhoul hemaluria lasting more than 6 monlhs; and
(3) persistency low C3 levels. Renal biopsy is done lo

196

Moudgil, Bagga, Fredrich, and Jordan

assess prognosis or detect the presence of other underlying

glomerular diseases such as systemic lupus erythematosus,
antineutrophil cytoplasmic autoantibody associated vasculitis, or membranoproliferative GN.
Pathology
As indicated above, renal biopsy is generally not required
for the diagnosis of PSGN. Information on pathology has
been derived from patients with severe nephritis or from
autopsy material. Primary acute abnormalities include a
generalized and diffuse endocapillary and mesangial cell
proliferation accompanied by infiltration with polymorphonuclear leukocytes, monocytes, and eosinophils within the
capillary lumina and mesangium. The glomeruli may enlarge and show lobular accentuation of the tufts (Fig. 1).
Epithelial cells are rarely involved in the proliferative process; significant crescent formation is limited to patients
with severe disease. Crescents in 50% or more glomeruli
suggests a guarded prognosis for complete recovery. In
cases biopsied after clinical resolution, the hypercellularity
subsides beginning with disappearance of the neutrophils,
and the capillary lumina become patent. However, glomer-

uli may retain some abnormal morphological features and

diffuse or segmental mesangial proliferation may persist
for prolonged periods.
On electron microscopy, the presence of discrete
electron-dense, dome-shaped subepithelial deposits is characteristic (Fig. 2). These "humps" are present early in the
clinical course and diminish after 4 to 6 weeks. Small mesangial and subendothelial deposits also are usually present.
Incomplete clinical resolution may occur when they are
present in large numbers and in contiguous areas. Immunofluorescence examination shows deposition of IgG and C3
in a granular pattern along the glomerular basement membrane and in the mesangium. Most deposits disappear after
a few weeks, but mesangial deposits may persist for prolonged periods.
Pathogenesis
Poststreptococcal glomerulonephritis is an immune complex disease. However, the precise nature of the antigens
involved in the formation of nephritogenic immune complexes is unknown. Soluble streptococcal antigenic products in the glomeruli have been detected inconsistently.

'*. '**;
**

FIGURE I Light microscopy of acute poststreptococcal glomerulonephritis showing a hypercellular glomerulus with polymorphonuclear leukocytes within the capillary lumen. Part of the second
glomerulus in the figure shows increased endocapillary proliferation with narrowing of the lumen (hematoxylin -eosin). Courtesy
of Dr. A. H. Cohen.

_. . . <

FIGURE 2 Electron micrograph of a glomerular capillary from a

patient with acute poststreptococcal glomerulonephritis showing
a large dome-shaped subepithelial electron-dense deposit (arrow).
Effacement of foot processes is present. Courtesy of Dr. A. H.
Cohen.

26. Poststreptococcal and Infection-Related GN

The reasons for this inconsistency include possible denaturization of antigenic material in the glomeruli and saturation
of antigen with specific antibody, therefore blocking recognition by immunochemical probes. Endostreptosin has
been identified in the mesangium, but not in the "humps"
that are characteristic of the disease.
IgG, C3, and streptococcal extracellular antigens have
been detected in circulating immune complexes of patients
with PSGN. However, similar complexes are also found in
patients with impetigo without nephritis. Since streptococcal antigens are not always nephritogenic, other mechanisms may be involved. It has been suggested that these
antigens may induce alterations of autologous IgG or glomerular components rendering them immunogenic or
cross-reactive with normal glomerular constituents. Another possible mechanism is that endostreptosin, or other
cationic cytoplasmic antigens, may bind to glomerular
structures and serve as "fixed" or "planted" antigen for
development of in situ immune complexes. Glomerular
immune complex deposition may then result in complement activation, activation of kinin or coagulation cascades,
release of polymorphonuclear chemotactic factors, and
acute glomerular injury.
Prevention and Treatment

A long-term protective immunity against nephritogenic

streptococci cannot be achieved with active immunization
and complete eradication of the organism in the population
is also not feasible. Early antimicrobial therapy of affected
persons and immediate family members may prevent the
spread of streptococcal infections but does not prevent the
development of PSGN. It may, however, attenuate the
severity of clinical disease. Widespread use of antimicrobials and better hygiene has greatly reduced the incidence
of PSGN in developed countries.
Treatment of patients with acute PSGN is symptomatic.
Fluid and sodium restriction and judicious use of loop diuretics are useful in patients with circulatory congestion,
hypertension, and edema. Severe hypertension with or
without encephalopathy is an emergency and usually responds to treatment with nifedipine and furosemide. Calcium channel blockers and loop diuretics may also be used
for treatment of moderate hypertension. Because plasma
renin activity is reduced in these patients, angiotensinconverting enzyme inhibitors and /3-blockers could be predicted to be less useful. Despite this, captopril has been
shown to lower blood pressure and increase creatinine
clearance in these patients.
Bed rest has no beneficial effect but has been traditionally recommended for treatment of severe hypertension,
hematuria, and marked edema. Restriction of protein may
be necessary in the presence of marked azotemia. Hyperkalemia may be treated with ion exchange resins. Occasionally dialysis may be necessary. Antibiotic therapy has no
influence on the course of the disease; however, appropriate therapy should be instituted for patients with positive
cultures for group A j8-hemolytic streptococci. Since subsequent encounters with nephritogenic strains of group A /3-

197

hemolytic streptococci are unlikely, penicillin prophylaxis

is not recommended.
Course and Prognosis

The acute clinical episode of PSGN is usually self-limited

and most patients will undergo diuresis within 7 to 10 days
of onset of the illness. Hypertension and azotemia subside
within 1 to 2 weeks. Hematuria and proteinuria disappear
by 6 months in more than 90% of children. Microscopic
hematuria may persist for 1 to 2 years. In adults, proteinuria
may be present in 60% for more than 1 year and in 30%
for 2 or more years.
Follow-up of pediatric patients with PSGN has revealed
a paucity of chronic sequelae. In the majority of patients,
with no evidence of preexisting renal disease, heavy proteinuria or extensive crescentic glomerular lesions at the
onset, the prognosis is excellent. However, in children with
disease severe enough to require hospitalization, the outcome may not always be benign. Most investigators believe
that the severity of disease at the onset correlates with the
development of chronicity and chronic glomerulonephritis
does not evolve from mild or unrecognized PSGN.
In contrast to children, the prognosis in adults is less
favorable. Long-term studies of patients with PSGN have
reported progression to endstage renal failure in some instances . Data from Trinidad and South Australia, however,
have indicated that recovery is the rule even though complete histological and clinical healing may take as long as
9 years. The long-term prognosis of PSGN occurring in
association with epidemics appear to be more favorable
than those occurring sporadically. This may represent selection bias.
Since immunity to streptococcal M-protein is type specific and long lasting and nephritogenic serotypes are limited in number, recurrent episodes of PSGN are rare.
GLOMERULONEPHRITIS ASSOCIATED WITH
INFECTIONS OTHER THAN GROUP A
/S-HEMOLYTIC STREPTOCOCCI
GN Associated with Sepsis, Bacterial Endocarditis, and
Visceral Abscesses

GN may be seen in association with a variety of infections

other than group A /3-hemolytic streptococci (Table 1). It
may also occasionally occur with group D streptococci,
gonococci, and gram-negative rods. The incidence of GN
due to staphylococcal and gram-negative infections has
increased in the past two or three decades, particularly in
immunocompromised adults, alcoholics, intravenous drug
abusers, and the elderly. Some of these patients develop
bacterial endocarditis and visceral abscesses. The clinical
features in these patients may be indistinguishable from
PSGN; occasionally, however, the onset may be insidious
and the course protracted. GN is of varying severity and
may show focal or diffuse endocapillary proliferation with
or without crescents. The total complement and C3 levels
are usually decreased. Cryoglobulins and circulating immune complexes are frequently detected. It is important

198

Moudgil, Bagga, Fredrich, and Jordan

to distinguish these patients from PSGN patients because

appropriate antibiotic therapy leads to resolution of symp toms. Valve replacement surgery may be indicated in patients with bacterial endocarditis and persisent infection
despite proper antimicrobial treatment. Surgery may also
be required to drain a valve ring abscess. Delay in institution of proper treatment may lead to increased patient
mortality and or incomplete renal recovery leading to
chronic renal failure.
Shunt Nephritis

GN associated with a chronically infected shunts used

for relieving hydrocephalus and infected vascular grafts or
venous catheters is known as shunt nephritis. Stapylococcus
epidermidis, an organism found on normal skin, is the most
common infecting organism and may be cultured either
from the blood or from the shunt. Other organisms implicated include Staphylococcus aureus and gram-negative
bacteria. Clinical features are usually indolent and include
fever, arthralgias, lethargy, weight loss, pallor, purpuric
rash, lymphadenopathy, and hepatosplenomegaly. Patients
may have microscopic hematuria, proteinuria, and impaired renal function. Gross hematuria, hypertension, and
nephrotic syndrome are uncommon. Serum complement
levels are usually low. Serum concentrations of factor B
and properdin are only occasionally decreased, indicating
predominant activation of the classical complement pathway. Rheumatoid factor, circulating immune complexes,
and cryoglobulins are frequently present in the serum. Renal histology generally shows features of membranoproliferative GN and sometimes of focal proliferative GN.
Renal biopsy is not usually indicated. Most patients recover
with removal of shunts and antibiotic therapy. Some patients may be left with residual renal impairment.

proliferative GN, mesangial proliferative GN, and vasculitic involvement of the kidney may also occur. The patients
generally present with nephrotic syndrome, but may also
present with AGN. These patients have circulating HBsAg,
and a high proportion also show HBe antigenemia. Serum
C3 and C4 levels are usually decreased, and cryoglobulins
and circulating immune complexes are often detected.
HBV particles may be seen in the glomerular capillary wall
on electron microscopy, and HBsAg and HBeAg have been
detected in the glomerular capillaries by immunofluorescence.
Hepatitis C virus infection is also associated with the
development of GN. The spectrum of glomerular lesions
found in association with hepatitis C infection is similar
to that seen in hepatitis B. Hepatitis C infection is more
commonly associated with proliferative GN, whereas hepatitis B infection is more common in membranous nephropathy. Figure 3 shows light microscopy of a patient with hepatitis C related GN illustrating mesangial proliferation in
some capillary loops whereas others are unaffected. It also
shows thickening of glomerular basement membrane in
some capillaries. Patients may develop hematuria, proteinuria, or nephrotic syndrome. The pathogenesis of GN is
thought to be similiar to that associated with hepatitis B.

Glomerulonephritis Associated with Syphilis

Renal involvement is known to occur in congenital and,

rarely, acquired syphilis. Patients generally present with
nephrotic syndrome, but may occasionally show features
of AGN. The renal disease is usually associated with other
manifestations of congenital syphilis, including hepatosplenomegaly, skin rash, and mucosal involvement. The
diagnosis is confirmed by positive serologic test results for
syphilis. Complement levels of Clq, C4, C3, and C5 are
decreased in congenital syphilis and return to normal following treatment with penicillin. The serum levels of complement are usually normal in acquired syphilis. Membranous nephropathy or diffuse proliferative GN may be seen
on renal biopsy. Renal disease is thought to be secondary
to an immunologically mediated reaction to treponemal
antigens. Treatment of both congenital and acquired syphilis results in rapid improvement of renal manifestations.
Glomerulonephritis Associated with Hepatitis B and C

Hepatitis B infection is associated with glomerular dis ease, (usually a membranous nephropathy), membrano-

FIGURE 3 Light microscopic histology of hepatitis C related glomerulonephritis showing focal mesangial proliferation and thickening of the capillary wall (periodic acid- Schiff). Courtesy of
Dr. A. H. Cohen.

199

26. Poststreptococcal and Infection -Related GN

Hepatitis C particles have, however, not been conclusively

demonstrated in the glomerular capillaries, but have been
identified in cryoglobulins and immune complexes.
Immunosupressive therapy is not indicated in the treatment of hepatitis B or hepatitis C associated GN since it
impairs the host's ability to clear these viruses from the
circulation and may actually stimulate viral replication.
Some patients, particularly those with hepatitis B associated GN, may go into spontaneous remission. Antiviral
therapy with interferon-a may be useful, but relapses are
common (see Chapter 19)
Other Infections Reported in Association with GN
GN secondary to typhoid fever, brucellosis, leptospirosis,
cytomegalovirus, measles, mumps, and infectious mononucleosis has also been described. The nephritis seen with
these organisms is usually mild and often presents with
hematuria and proteinuria. The absence of bacteriological
and serological evidence of a prior streptococcal illness and
the temporal association with these infections sugges t the
diagnosis of infectious GN of nonstreptococcal origin.
GN Associated with Parvovirus BI9 and Hantavirus
Infection with parvovirus B19, in patients with sickle cell
disease, has been recently reported to cause focal segmental
GN. These patients presented with nephrotic syndrome
approximately 1 week after the onset of aplastic crisis.
Most of these patients showed progression to chronic renal
failure. Focal segmental glomerulosclerosis was seen in one
patient who was biopsied late in the course of disease.
Hantavirus infection, originally discovered in Korea, is
known to cause hemorrhagic fever with renal involvement.
A few cases have been described in the United States and
other developed countries. It primarily causes tubulointerstitial nephritis and acute renal failure. Slight glomerular
mesangial changes may be seen on renal biopsy in some patients.
With the availabilty of newer and powerful diagnostic
tools such as polymerase chain reaction/insitu hybridization
to detect infectious agents, the list of viruses and other
infectious agents associated with GN will probably continue to grow. These advances will help us to understand

more about the pathogenesis of so called "idiopathic GN"

and even focal segmental glomerulosclerosis.
Bibliography
Arze RS, Rashid H, Morley R, et al.: Shunt nephritis: report of
2 cases and review of the literature. Clin Nephrol 19 (1):48
53, 1983
Baldwin DS, Gluck MC, Schacht RG, et al.: The long term course
of poststreptococcal glomerulonephritis. Ann Intern Med 80
(3):342-358, 1974.
Beaufils M, Liliane MM, Sraer JD, et al.: Acute renal failure of
glomerular origin during visceral abscesses. N EngI J Med 295
(4):185-189, 1976.
Centers for disease control and prevention. Update:Hantavirus
disase in United States. MMWR 42:6 12-614, 1993.
Conjeevaram HS, Hoofnagle JH, Austin HA, et al.: Long term
outcome of hepatitis B virus related glomerulonephritis after
therapy with interferon alfa. Gastroenterology 109 (2):540546, 1995.
Dodge WF, Spargo BH, Travis LB, et al.: Poststrepto coccal glomerulonephritis: A prospective study in children. N EngI J Med
286 (6):273 -278, 1972.
-Cumber SC, Chopra S: Hepatitis C: A multifaceted disease. Review of extrahepatic manifestations. Ann Intern Med 123 (8):
615-620, 1995.
Hruby Z, Kuzniar J, Rabczynski J, et al.: The variety of clinical and
histopathologic presentations of glomerulonephritis associated
with latent syphilis. Int Urol Nephrol 24(5):541-547, 1992.
Montseny JJ, Meyrier A, Kleinknecht D, et al.: The current spectrum of infectious glomerulonephritis. Experience with 76 patients and review of the literature. Medicine 74 (2): 63-73, 1995.
Neugarten J, Gallo GR, Baldwin DS: Glomerulonephritis in bacterial endocarditis. Am J Kidney Dis 3(5): 371-379, 1984.
Potter EV, Lipschultz SA, Abi dh S, et al.: Twelve to seventeen
year follow-up of patients with poststreptococcal acute glomerulonephritis in Trinidad. N EngI J Med 307 (12):725-729, 1982.
Ronco P, Verroust P, Morel-Maroger L: Viruses and glomerulonephritis. Nephron 31 (2):97-102, 1982.
Simckes AM, Spitzer A: Poststreptococcal acute glomerulonephritis. Pediatr Review 16 (7): 278-279, 1995.
Seligson G, Lange K, Majeed MA, et al.: Significance of endostreptosin antibody litres in poststreptoccal glomerulonephritis. Clin
Nephrol 24 (2):69-75, 1985.
Vogl W, Renke M, Mayer-Eichberger D, et al.: Long-term progno sis for endocapillary glomerulonephritis of poststreptococcal
type in children and adults. Nephron 44 (l):58-65, 1986.
Wierenga KJ, Pattison JR, Brink N, et al.: Glomerulonephritis
after human parvovirus infection in homozygous sickle-cell disease. Lancet 346 (8973):475-476, 1995.

RENAL INVOLVEMENT IN
SYSTEMIC VASCULITIS
J. CHARLES JENNETTE AND RONALD J. FALK

The kidneys are affected by many systemic vasculitides

(Fig. 1). Large vessel vasculitides, such as giant cell (temporal) arteritis and Takayasu arteritis, can narrow the abdominal aorta or renal arteries, resulting in renal ischemia and
renovascular hypertension. Medium-sized vessel vasculitides, such as polyarteritis nodosa and Kawasaki disease,
also can reduce flow through the renal artery and may affect
intrarenal arteries, resulting in infarction and hemorrhage.
Small vessel vasculitides, such as microscopic polyangiitis
(microscopic polyarteritis), Wegener's granulomatosis,
Henoch-Schonlein purpura, and cryoglobulinemic vasculitis, frequently involve the kidneys, especially glomerular
capillaries.

nary arteries, and main renal arteries. Polyarteritis nodosa

often involves the renal arteries, whereas Kawasaki disease
only rarely affects the renal arteries. These diseases also
may involve small arteries, such as arteries within the parenchyma of skeletal muscle, liver, heart, pancreas, spleen,
and kidney. By the definitions in Table 1, these vasculitides
do not affect vessels smaller than arteries (i.e., arterioles,
capillaries, venules), and thus do not cause glomerulonephritis. The presence of arteritis and glomerulonephritis
indicates some form of small vessel vasculitis, such as microscopic polyangiitis.
Histologically, the acute arterial injury of Kawasaki dis ease and polyarteritis nodosa is characterized by focal ves sel wall fibrinoid necrosis and infiltration of inflammatory
cells. Fibrinoid necrosis results from plasma coagulation
factors spilling into the necrotic areas, where they are activated to form fibrin. Early in the inflammatory process,
neutrophils predominate, but later mononuclear leukocytes are most numerous. Thrombosis may occur at the site
of inflammation, resulting in infarction. Focal necrotizing
injury to vessels erodes into the vessel wall and adjacent
tissue producing an inflammatory aneurysm, which may
rupture and cause hemorrhage.
Although small vessel vasculitides may affect mediumsized arteries, they have a predilection for small vessels
such as arterioles, postcapillary venules (e.g., in the dermis),
and capillaries (e.g., in glomeruli and pulmonary alveoli)
(Fig. 1). As described in Table 1, a variety of clinically and
pathogenetically distinct forms of small vessel vasculitis
have in common focal necrotizing inflammation of small
vessels. In the acute phase, this injury is characterized histologically by segmental fibrinoid necrosis and neutrophil
infiltration (Fig. 2), sometimes with secondary thrombosis.
The neutrophils often undergo karyorrhexis (leukocytoclasia). With chronicity, mononuclear leukocytes become
predominant and fibrosis develops.
The distinctive features of each form of vasculitis are
described in Table 1 and Fig. 1. For example, Wegener's
granulomatosis has necrotizing granulomatous inflammation, Churg-Strauss syndrome has eosinophilia and asthma,
Henoch-Schonlein purpura has IgA-dominant vascular immune deposits, cryoglobulinemic vasculitis has cryoglobulins, and microscopic polyangiitis has none of these features.

PATHOLOGY

As depicted in Fig. 1 and described in Table 1, different

types of systemic vasculitis have a predilection for affecting
different vessels within the kidney. In addition, different
types of vasculitis have different histologic and immunohis tologic features.
Giant cell (temporal) arteritis and Takayasu arteritis
have a predilection for the aorta and its major branches,
and only rarely cause clinically significant renal disease,
although asymptomatic pathologic involvement is more
common. Giant cell (temporal) arteritis often involves the
extracranial branches of the carotid arteries, including the
temporal artery . Some patients, however, do not have temporal artery involvement, and patients with other types
of vasculitis (e.g., microscopic polyangiitis and Wegener's
granulomatosis) may have temporal artery involvement.
Therefore, temporal artery disease is neither a required
nor a sufficient pathologic feature of giant cell (temporal) arteritis.
Histologically, both giant cell (temporal) arteritis and
Takayasu arteritis are characterized by focal granulomatous inflammation, often with multinucleated giant cells.
With chronicity, the inflammatory injury evolves into fibrosis and frequently results in vascular narrowing.
Polyarteritis nodosa and Kawasaki disease have a predilection for medium-sized arteries (i.e., main visceral arteries), such as the mesenteric arteries, hepatic arteries, coro-

Primer on Kidney Diseases, Second Edition

200

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

Lobar/Lo
bular
Arcuate Artery
Interlobular Artery
Arteriole
Glomerulus

Large
Vessel
Vasculitis

Aorta
Medium-Sized
Vessel
Vasculitis

Small
Vessel
Vasculitis

granulomatous granulomatous
arteritis in a
arteritis in a
patient >50
patient <50

Takayasu
Arteritis

immune complexes in

necrotizing
arteritis without
MCLN syndrome

necrotizing
arteritis with
MCLN syndrome
synrome
MCLN sy

Polyarteritis
Nodosa

Kawasaki
Disease

vessel walls with IF

immunoglobulin staining

T
other sources
for immune
complexes
Other Im Cx
Vasculitis

cryoglobulins IgA-dominant
in blood and
vessel wall
vessel walls immune deposits
Cryoglobulinemic
Vasculitis

H-S
Purpura

circulating ANCA with

paucity of vascular IF
immunoglobulin staining

T
SLE or
rheumatoid
arthritis
Lupus/Rheumatoid
Vasculitis

T
vasculitis with
no asthma or
granulomas
Microscopic
Polyangiitis

Wegener's
Granulomatosis

Churg-Strauss
Syndrome

FIGURE I Predominant distribution of renal vascular involvement by systemic vasculitides and diagnostic clinical and
pathologic features that distinguish among them. The width of the black triangles indicates the predilection of small,
medium, and large vessel vasculitides for various portions of the renal vasculature. Note that medium -sized renal arteries
can be affected by large, medium, and small vessel vasculitides, but arterioles and glomeruli are affected by small vessel
vasculitides alone based on the definitions in Table 1. MCLN; mucocutaneous lymph node syndrome.

granulomas
and no
asthma

eosinophilia,
asthma and
granulomas

J. Charles Jennette and Ronald J. Falk

202
TABLE I

Names and Definitions of Vasculitis Adopted by the Chapel Hill Consensus Conference on the Nomenclature of
Systemic Vasculitis
Large vessels vasculitis"
Giant cell (temporal) arteritis Granulomatous arteritis of the aorta and its major branches, with a predilection for the extracranial
branches of the carotid artery. Often involves the temporal artery. Usually occurs in patients older than 50
and often is associated with polymyalgia rheumatica.
Takayasu arteritis Granulomatous inflammation of the aorta and its major branches. Usually occurs in patients younger than 50.
Medium-sized vessel vasculitis"
Polyarteritis nodosa (classic Necrotizing inflammation of medium -sized or small arteries without glomerulonephritis or vasculitis in
polyarteritis nodosa)
arterioles, capillaries, or venules.
Kawasaki disease

Arteritis involving large, medium -sized, and small arteries, and associated with mucocutaneous lymph node
syndrome. Coronary arteries are often involved. Aorta and veins may be involved. Usually occurs in
children.
Small vessel vasculitis0

Wegener's granulomatosis*' Granulomatous inflammation involving the respiratory tract, and necrotizing vasculitis affecting small to
medium-sized vessels, e.g., capillaries, venules, arterioles, and arteries. Necrotizing glomerulonephritis is
common.
Churg-Strauss syndrome1 ,b,c Eosinophil-rich and granulomatous inflammation involving the respiratory tract and necrotizing
vasculitis affecting small to medium -sized vessels, and is associated with asthma and blood eosinophilia
Microscopic polyangiitis
(microscopic
polyarteritis)6 ''7
Henoch-Schonlein purpura''

Necrotizing vasculitis with few or no immune deposits affecting small vessels, i.e., capillaries, venules, or
arterioles. Necrotizing art eritis involving small and medium -sized arteries may be present. Necrotizing
glomerulonephritis is very common. Pulmonary capillaritis often occurs.
Vasculitis with IgA- dominant immune deposits affecting small vessls, i.e., capillaries, venules, or
arterio les. Typically involves skin, gut, and glomeruli, and is associated with arthralgias or arthritis.

Essential cryoglobulinemic Vasculitis with cryoglobulin immune deposits affecting small vessels, i.e., capillaries, venules, or
vasculitis''
arterioles, and associated with cryoglobulins in serum. Skin and glomeruli are often involved.
Cutaneous leukocytoclastic Isolated cutaneous leukocytoclastic angiitis without systemic vasculitis or glomerulonephritis.
angiitis
Note. Modified from Jennette JC, Falk RJ, Andrassy K, et al.: Nomenclature of systemic vasculitides: the
proposal of an international consensus conference. Arthritis Rheum, 1994;37:187-192.
" Large artery refers to the aorta and the largest branches directed toward major body regions (e.g., to the extremities and the head
and neck); medium - sized artery refers to the main visceral arteries (e.g. renal, hepatic, coronary and mesenteric arteries), and small vessel
refers to the distal arterial radicals that connect with arterioles (e.g., renal arcuate and interlobular arteries), as well as arterioles, capillaries,
and venules. Note that some small and large vessel vasculitides may involve medium -sized arteries; but large and medium -sized vessel
vasculitides do not involve vessels smaller than (other than) arteries.
Strongly associated with antineutrophil cytoplasmic autoantibodies (ANCA).
c
May be accompanied by glomerulonephritis and can manifest as nephritis or pulmonary-renal vasculitic syndrome.

The glomerular lesions of microscopic polyangiitis,

Wegener's granulomatosis, and Churg-Strauss syndrome
(i.e., the ANCA vasculitides) are identical pathologically
and are characterized by segmental fibrinoid necrosis and
crescent formation (Fig. 3). The glomerulonephritis of
Henoch-Schonlein purpura is identical to IgA nephropathy. The glomerulonephritis of cryoglobulinemic vasculitis is a secondary form of type I membranoproliferative glomerulonephritis (mesangiocapillary glomerulonephritis).
Leukocytoclastic angiitis of medullary vasa recta (Fig.
4) also occurs in the ANCA-vasculitides, and rarely may
be severe enough to cause papillary necrosis.

PATHOGENESIS

"Vasculitides are caused by the activation of inflammatory

mediator systems in vessel walls. The initiating event (i.e.,
the etiology), however, is unknown for most vasculitides.
An immune response to heterologus antigens (e.g., hepatitis B or C antigens in some forms of immune complex
vasculitis) or autoantigens (e.g., endothelial antigens in Kawasaki disease) is presumed to be the etiologic event in
many vasculitides, but this can be documented in only a few
patients. Table 2 categorizes vasculitides based on putative
immunologic mechanisms.
Except for cellular vascular rejection in transplanted
organs, there is no well-documented example of T-cell-

27. Renal Involvement in Systemic Vasculitis

FIGURE 1 Renal interlobular artery with fibrinoid necrosis from

a patient with microscopic polyangiitis (Masson trichrome stain).

mediated vasculitis. Primarily because of the pattern of

inflammation, cell-mediated inflammation has been incriminated in the pathogenesis of giant cell (temporal) arteritis
and Takayasu arteritis, but there is no strong experimental
support for this.
The vasculitides listed in the immune complex mediated
category in Table 2 all immunohistologic evidence for ves sel wall immune complex localization, i.e., granular staining
for immunoglobulins and complement. In these vasculitides, the presence of antibodies bound to antigens in vessel
walls activates humoral inflammatory mediator systems
(e.g., complement, coagulation, fibrinolytic and kinin systems), which attract and activate neutrophils and monocytes. These activated leukocytes releaser generate toxic
oxygen metabolites and release enzymes that cause matrix
lysis and cellular apoptosis, resulting in necrotizing inflammatory injury to vessel walls.
This same final pathway of inflammatory injury also can
be reached by antibodies binding to antigens that are inte-

203

FIGURE 4 Medullary vasa recta with leukocytoclastic angiitis

from a patient with Wegener's granulomatosis (hematoxylin
and eosin ).

gral components of vessel walls. The best documented

example of this is anti-GBM antibody mediated glomerulonephritis and Goodpasture's syndrome. A less welldocumented example of direct antibody attack is Kawasaki
disease. There is evidence that patients with Kawasaki dis ease have antiendothelial antibodies that react with antigens that are expressed on the surface of stimulated endothelial cells.
A very important group of necrotizing systemic small
vessel vasculitides, which frequently involve the kidneys,
has no immunohistologic evidence for vascular immune

TABLE 2

Putative Immunologic Causes of Vasculitis

Immune complex mediated Henoch- Schonlein
purpura" Cryoglobulinemic vasculitis" Lupus
vasculitis" Serum sickness vasculitis"
Rheumatoid vasculitis Polyarteritis nodosa
Infect ion-induced immune complex vasculitis"
Viral (e.g., hepatitis B and C virus)
Bacterial (e.g., streptococcal)
Direct antibody attack mediated
Goodpasture's syndrome (antibasement membrane antibodies)"
Kawasaki disease (antiendothelial antibodies)
ANCA-associated and possibly ANCA-mediated
Wegener's granulomatosis" Microscopic
polyangiitis" Churg- Strauss syndrome"
Cell mediated
Allograft cellular vascular rejection
Giant cell (temporal) arteritis
Takayasu arteritis
" May be accompanied by glomerulonephritis and can manifest
as nephritis or pulmonary-renal vasculitic syndrome.

FIGURE 3 Glomerulus with segmental fibrinoid necrosis with

red (fuchsinophilic) fibrinous material and an adjacent cellular
crescent from a patient with ANCA-small vessel vasculitis (Masson
trichrome stain).

|. Charles Jennette and Ronald j. Falk

204
complex localization or direct antibody binding. This paucity of immune deposits has fostered the designation pauciimmune for this group of vasculitides, which includes
microscopic polyangiitis (microscopic polyarteritis), Wegener's granulomatosis, and Churg-Strauss syndrome. The
pathogenesis of these vasculitides is unknown, but their
close association with antineutrophil cytoplasmic autoantibodies (ANCA) has lead to speculation that they result
from ANCA-induced leukocyte activation and vascular
injury.
ANCA are autoantibodies specific for proteins within
the granules of neutrophils and lysosomes of monocytes.
They often are detected in patient serum by indirect immu nofluorescence microscopy using alcohol-fixed normal human neutrophils as substrate. Using this assay, two patterns
of neutrophil staining discriminate between the two major
subtypes of ANCA: cytoplasmic staining (C-ANCA) and
perinuclear staining (P-ANCA). Using specific immunochemical assays, such as enzyme-linked immunosorbent
assays or radioimmunoassays, most C-ANCA are specific
for a neutrophil and monocyte protein ase called proteinase
3 (PR3) and most P-ANCA are specific for myeloperoxidase (MPO).
One hypothesis about the pathogenesis of ANCAassociated vasculitides proposes that ANCA react with cytoplasmic antigens (e.g., PR3 and MPO) that are released
at the surface of cytokine-stimulated leukocytes, causing
the leukocytes to adhere to vessel walls, degranulate, and
generate toxic oxygen metabolites. ANCA antigens also
may become planted in vessel walls or even produced by
endothelial cells, thus providing a nidus for in situ immune
complex formation in vessel walls . If such in situ formation
is present, it must be at a level that cannot be detected by
immunofluorescence microscopy, because of the pauciimmune character of ANCA vasculitis.
CLINICAL FEATURES

The clinical features of systemic vasculitides are extremely varied and are dictated by the category of vasculitis, the type of vessel involved, and the organ system dis tribution of vascular injury. Irrespective of the type of
vasculitis, most patients will have accompanying constitutional features of inflammatory disease, such as fever,
arthralgias, myalgias, and weight loss. These are probably
caused by increased circulating levels of proinflamma tory cytokines.
Giant cell (temporal) arteritis and Takayasu arteritis typically present with evidence for ischemia in tissues supplied
by involved arteries. Patients with Takayasu arteritis often
develop claudication (especially in the upper extremities),
absent pulses, and bruits. Approximately 40% of patients
with Takayasu arteritis develop renovascular hypertensio n,
but this is rare with giant cell arteritis.
Giant cell (temporal) arteritis can affect virtually any
organ in the body, but signs and symptoms of involvement
of arteries in the head and neck are the most common
clinical manifestations. Superficial arteries (e.g., the tempo-

ral artery) may be swollen and tender. Arterial narrowing

causes ischemic manifestations in affected tissues, such as
headache, jaw claudication, and loss of vision. About half
of the patients with giant cell (temporal) arteritis have
polymyalgia rheumatica, which is characterized by aching
and stiffness in the neck, shoulder girdle, or pelvic girdle.
Medium-sized vessel vasculitides (i.e., polyarteritis nodosa and Kawasaki disease) often present with clinical evidence for infarctio n in multiple organs, such as abdominal
pain with occult blood in the stool and skeletal muscle
pain and cardiac pain with elevated serum muscle enzymes.
Laboratory evaluation often demonstrates clinically silent
organ damage, such as liver injury with elevated liver function tests and pancreatic injury with elevated serum amylase.
Polyarteritis nodosa frequently causes multiple renal
infarcts and aneurysms. Unlike microscopic polyangiitis, polyarteritis nodosa usually does not cause severe renal failure. Rupture of arterial aneurysms with massive
retroperitoneal or intraperitoneal hemorrhage is a life threatening complication of polyarteritis nodosa.
Kawasaki disease almost always occurs in young children
and has a predilection for coronary, axillary, and iliac arteries. Kawasaki disease is accompanied by the mucocutaneous lymph node syndrome, which includes fever, nonpuru lent lymphadenopathy, and mucosal and cutaneous
inflammation. Although the renal arteries frequently are
involved histologically, clinically significant renal disease
is very rare in patients with Kawasaki disease.
The small vessel vasculitides often present with evidence
for inflammation in vessels in multiple organs, but may
initially manifest with involvement of only one organ, followed by development of disease in other organs. Hematuria, proteinuria, and renal insufficiency caused by glomerulonephritis are frequent clinical features of all of the small
vessel vasculitides listed in Table 1. Other manifestations
include purpura caused by leukocytoclastic angiitis in dermal venules and arterioles, abdominal pain and occult
blood in the stool from mucosal and bowel wall infarcts,
mononeuritis multiplex from arteriolitis and arteritis in
peripheral nerves, necrotizing sinusitis from upper respiratory tract mucosal angiitis, and pulmonary hemorrhage
from necrotizing alveolar capillaritis.
In addition to the features just described, which are common to patients with all types of small vessel vasculitis,
patients with Wegener's granulomatosis and Churg-Strauss
syndrome have distinctive clinical features that set them
apart. Patients with Wegener's granulomatosis have necro tizing granulomatous inflammation in the upper or lower
respiratory tract, and rarely in other tissues (e.g., skin,
orbit). In the lungs, this inflammation produces irregular
nodular lesions that can be observed by radiography. These
lesions may cavitate and hemorrhage, but massive pulmo nary hemorrhage in patients with Wegener's granulomatosis is usually caused by capillaritis rather than granulomatous inflammation. By definition, patients with ChurgStrauss syndrome have blood eosinophilia and a history of

205

27. Renal Involvement in Systemic Vasculitis

asthma. They also develop eosinophil-rich tissue inflammation, especially in the lungs and gut.
DIAGNOSIS

Multisystem disease in a patient with constitutional signs

and symptoms of inflammation, such as fever, arthralgias,
myalgias, and weight loss, should raise suspicion of systemic
vasculitis. Data that will resolving the differential diagnosis
include the patient's age, organ distribution of injury, concurrent syndromes (e.g., mucocutaneous lymph node syndrome, polymyalgia rheumatica, asthma), type of vessel
involved (e.g., artery, arteriole, venules capillary), lesion
histology (e.g., granulomatous, necrotizing), lesion immu nohistology (e.g., immune deposits, pauciimune), and laboratory data (e.g., cryoglobulins, ANCA) (Fig. 1).
Signs and symptoms of tissue ischemia along with angiography demonstrating irregularity, stenosis, occlusion, or,
less commonly, aneurysms of large and medium-sized arteries should suggest giant cell (temporal) arteritis or Takayasu arteritis. A useful discriminator between giant cell arteritis and Takayasu arteritis is age, with the former rarely
occurring before 50 years of age and the latter rarely occurring after 50. The presence of polymyalgia rheumatica is
a clinical marker for giant cell arteritis.
Polyarteritis nodosa and Kawasaki disease cause organ
ischemia and have a preference for internal viscera, such
as the heart, kidneys, liver, spleen, and gut. Arteritis in
skeletal muscle and subcutaneous tissues causes tender
nodules that can be identified on physical examination.
Medium-sized artery aneurysms observed angiographically
(e.g., in renal arteries) indicate some type of vasculitis, but
this is not very disease-specific because giant cell (temporal)
arteritis, Takayasu arteritis, polyarteritis nodosa, Kawasaki
disease, Wegener's granulomatosis, microscopic polyarteritis, and Churg-Strauss syndrome can all produce them.
Kawasaki disease occurs almost always in children under
5 years old and is accompanied by the mucocutaneous
lymph node syndrome.
A small vessel vasculitis should be suspected when there
is evidence for inflammation of vessels smaller than arteries, such as glomerular capillaries (hematuria and proteinuria), dermal postcapillary venules (purpura), and alveolar
capillaries (hemoptysis). To discriminate among the small
vessel vasculitides, evaluation of serology data, vessel immunohistology, or concurrent nonvasculitic disease (e.g.,
asthma, eosinophilia) is required (Fig. 1).
Evaluation of vessels in biopsy specimens, such as glo merular capillaries in renal biopsies or alveolar capillaries
in a lung biopsies, can be helpful, especially if immunohis tology is performed. The pauciimmune vasculitides (i.e.,
microscopic polyangiitis, Wegener's granulomatosis,
Churg-Strauss syndrome) will lack immune deposits, antiGBM disease will have linear immunoglobulin deposits,
and the immune complex vasculitides will have granular
immune deposits.

Serology, especially, ANCA analysis is useful in differentiating among the small vessel vasculitides. Wegener's
granulomatosis, microscopic polyangiitis (microscopic
polyarteritis), and Churg-Strauss syndrome are strongly
associated with ANCA. As depicted in Fig. 5, most patients
with active untreated Wegener's granulomatosis have CANCA (PR3-ANCA). A minority of patients will have PANCA (MPO-ANCA). Therefore, C-ANCA are a very
sensitive serologic marker for active Wegener's granulomatosis; however, C-ANCA are not completely specific for
Wegener's granulomatosis because some patients with CANCA will have systemic small vessel vasculitis without
granulomatous inflammation (i.e., microscopic polyangiitis) and others will have pauciimune necrotizing and cres centic glomerulonephritis alone. Approximately 80% of patients with microscopic polyangiitis have either C-ANCA
or P-ANCA, with most having P-ANCA (MPO-ANCA).
Patients with Churg -Strauss syndrome usually have either
P-ANCA or C-ANCA. A minority of patients with immune-complex-mediated vasculitis or anti-GBM disease
will have concurrent ANCA-associated disease.
Diagnostic serologic tests for immune-complex-mediated vasculitides include assays for circulating immune
complexes (e.g., cryoglobulins in cryoglobulinemic vasculitis), assays for antibodies known to participate in immune
complex formation or to mark the presence of a disease
that generates immune complexes (e.g., antibodies to hepatitis B or C, streptococci, DNA), and assays for consump tion or activation of humoral inflammatory mediator system components (e.g., assays for reduced complement
components or for activated membrane attack complex).
THERAPY AND OUTCOME

All of the vasculitides discussed in this chapter respond

to anti-inflammatory or immunosuppressive therapy. The
aggressiveness of treatment should match the aggressiveness of the disease.
Takayasu arteritis and giant cell (temporal) arteritis usually respond well to high-dose corticosteroid treatment
(e.g., 1 mg/kg body wt per day prednisone) during the
acute phase of the disease, followed by tapering and lowdose maintenance for several months to a year depending
on disease activity. Occasional patients with severe disease
or steroid toxicity benefit from cytotoxic agents (e.g., cyclophosphamide). If present, renovascular hypertension
should be controlled. After the inflammatory phase is past
and the sclerotic phase has developed, reconstructive vas cular surgery may be required to improve flow to ischemic
tissues, especially in patients with Takayasu arteritis.
Corticosteroid treatment is not recommended for Kawasaki disease because it appears to worsen coronary artery
disease, which is the most life-threatening aspect of Kawasaki disease. The preferred treatment is a combination of
aspirin and high-dose intravenous gamma globulins. This
controls the inflammatory manifestations of the disease
(e.g., the mucocutaneous lymph node syndrome), prevents

206

J. Charles Jennette and Ronald j. Falk

IMMUNE

P-ANCA
(MPO-ANCA)
DISEASE

GLQMERULONEPHRmS
ALONE

PULMONARYRENAL
VASCULIT1C
SYNDROME

ANT1-GBM
DISEASE

WEGENER'S
GRANULOMATOSIS

COMPLEX
DISEASE
C-ANCA
(PR3-ANCA)
DISEASE

FIGURE 5 Relationship of vasculitic clinicopathologic syndromes to immunopathologic categories of vascular injury in patients with crescentic glomerulonephritis. The circles represent
the major immunopathologic categories of vascular inflammation that affect the kidneys, and
the sh aded ovals the clinicopathologic expressions of the vascular inflammation. Note that
clinical syndromes can be caused by multiple immunopathologic processes, e.g., pulmonaryrenal vasculitic syndrome can be caused by anti-GBM antibodies (i.e., Goodpasture' s syndrome), immune complex localization (e.g., systemic lupus erythematosus), or ANCAassociated disease (e.g., microscopic polyangiitis and Wegener's granulomatosis). Reproduced
with permission from Jennette JC: Anti-neutrophil cytoplasmic autoantibody -associated disease: A pathologist's perspective. Am J Kidney Dis 1991;18:164-170.

thrombosis of injured arteries, and limits of coronary artery

involvement. With appropriate treatment, more than 90%
of patients with Kawasaki disease have complete resolution
of the disease.
Most patients with Henoch-Schonlein purpura have mild
self-limited disease that requires only supportive care. Arthralgias are relieved by nonsteroidal antiinflammatory
drugs. Corticosteroid treatment is beneficial in patients who
have severe abdominal pain caused by intestinal vasculitis.
The treatment of severe glomerulonephritis in patients with
Henoch-Schonlein purpura is controversial. There is anecdotal evidence that aggressive crescent glomerulonephritis
should be treated with high-dose corticosteroids, cytotoxic
agents, and/or plasmapheresis, but this has not been documented in controlled trials (see Chapter 28).
The course and treatment of cryoglobulinemic vasculitis
depend in part on the presence of associated diseases. In
patients with multiple myeloma or other lymphoproliferative diseases, treatment should be directed at the primary
disease. Up to 25% of patients with cryoglobulinemic vasculitis caused by hepatitis C may respond to interferon-a
treatment. In patients with severe vascular inflammation,

treatment with corticosteroids, cytotoxic drugs, and plas mapheresis may be required (see Chapter 19).
High-dose corticosteroids (e.g., pulse methylprednisolone) and cytotoxic agents (e.g., cyclophosphamide) are
recommended for controlling severe major organ damage
caused by microscopic polyangiitis, Wegener's granulomatosis, and Churg-Strauss syndrome. In patients with mild
disease, or disease of limited extent (e.g., polyarteritis nodosa confined to the skin or Wegener's granulomatosis
limited to the upper respiratory tract), corticosteroids alone
may be adequate. Trimethoprim-sulfamethoxazole may be
adequate treatment for mild Wegener's granulomatosis of
the upper respiratory tract, and perhaps for preventing
nasal relapse in patients with Wegener's granulomatosis
that is in remission.
Important indicators for aggressive treatment in patients
with microscopic polyangiitis and Wegener's granulomatosis are the development of rapidly progressive glomerulonephritis or massive pulmonary hemorrhage. Patients with
these complications should be treated with high-dose corticosteroids (e.g., pulse methylprednisolone 7 mg -kg^-day" 1
for 3 days followed by daily oral prednisone) and oral

27. Renal Involvement in Systemic Vasculitis

or intravenous cyclophosphamide. Oral cyclophosphamide can be initiated at a dose of 2 mg kg"1 day'1 ,

or intravenous cyclophosphamide can be begun at
0.5 g m~ 2 month"1 adjusted upward to 1 g/m2 based on
the leukocyte count. The optimum duration of cyclophosphamide treatment is controversial. One approach is to
stop treatment when there is disease remission, or after 6
months. Plasmapheresis and high-dose intravenous gamma
globulin also may be beneficial, but this is not well documented.
As many as 80% of ANCA vasculitis patients will enter
remission with aggressive immunosuppression, but approximately a third will have a relapse within 2 years. There is
controversy over how best to treat relapses. A frequent
approach is to re-treat overt vasculitic relapses with a repeat
course of corticosteroids and cyclophosphamide.
Bibliography
Agnello V, Chung RT, Kaplan LM: A role for hepatitis C virus
infection in type II cryoglobulinemia N Engl J Med 327:14901495, 1992.
Churg J, Churg A. Idiopathic and secondary vasculitis: A review.
Mod Pathol 2:144-60, 1989.
DeRemee RA, McDonald TJ, Harrison EG, et al.: Wegener's
granulomatosis. Anatomic correlates, a proposed classification.
Mayo Clin Proc 51:777-781, 1976.
Falk RJ, Hogan S, Carey TS, Jennette JC: Clinical course of antineutrophil cytoplasmic autoantibody -associated glomerulonephritis and systemic vasculitis: the Glomerular Disease Collaborative Network. Ann Intern Med 113:656-663, 1990.
Hagen EC, Ballieux BE, van Es LA, Daha MR, van der Woude
FJ: Antineutrophil cytoplasmic autoantibodies: A review of the

207
antigens involved, the assays, and the clinical and possible pathogenetic consequences. Blood 81:1996-2002, 1993.
Hoffman GS, Kerr GS, Leavitt RY, et al.: Wegener's granulomatosis: An analysis of 158 patients. Ann Intern Med 116:488-498,
1992.
Hunder GG, Arend WP, Bloch DA, C et al.: The American College
of Rheumatology 1990 criteria for the classification of vasculitis.
Arthritis Rheum 3:1065-1067, 1990.
Jennette JC, Falk RJ: The pathology of vasculitis involving the
kidney. Am J Kidney Dis 24:130-141, 1994.
Jennette JC, Falk RJ, Andrassy K, et al.: Nomenclature of systemic
vasculitides: proposal of an international consensus conference.
Arthritis Rheum 37:187-192, 1994.
Leung DYM, Collins T, Lapierre LA, et al.: Immunoglobulin M
antibodies present in the acute phase of Kawasaki syndrome
lyse cultured vascular endothelial cells stimulated by gamma
interferon. / Clin Invest 77:1428-1435, 1986.
Lie JT: Systemic and isolated vasculitis: A rational approach to
classification and pathologic diagnosis. Pathol Annu 24(Pt 1):25114, 1989.
Niles JL, Pan GL, Collins AB, et al.: Antigen-specific radioimmunoassays for anti-neutrophil cytoplasmic antibodies in the diagnosis of rapidly progressive glomerulonephritis. J Am Soc
Nephrol 2:27-36, 1991.
Nachman PH, Hogan SL, Jennette JC, Falk RJ: Treatment response and relapse in ANCA-associated microscopic polyangiitis and glomerulonephritis. / Am Soc Nephrol 7:23-32, 1996.
Ronco P, Verroust P, Mignon F, et al.: Immunopathological studies
of polyarteritis nodosa and Wegener's Granulomatosis: A report
of 43 patients with 51 renal biopsies. Q J Med 206:212-223,1983.
Serra A, Cameron JS, Turner DR, et al.: Vasculitis affecting the
kidney: Presentation, histopathology and long-term outcome.
Q J Med 53:181-207, 1984.
Smith DL. Spontaneous rupture of a renal artery aneurysm in
polyarteritis nodosa: Critical review of the literature and report
of a case. Am J Med 87:464-467, 1989.

RENAL MANIFESTATIONS OF SYSTEMIC

LUPUS ERYTHEMATOSUS AND OTHER
RHEUMATIC DISORDERS
JAMES E. BALOW

The kidney is affected in many rheumatic diseases as

well as by some of the drugs used to treat them. Some
of these topics are covered elsewhere: systemic vasculitis
(Chapter 27), membranoproliferative glomerulonephritis and cryoglobulinemia (Chapter 19), dysproteinemias
and amyloidosis (Chapter 30), and nonsteroidal antiinflammatory drugs and the kidney (Chapter 42). This
chapter focuses primarily on systemic lupus erythematosus,
Henoch-Schonlein purpura, and systemic sclerosis, which
regularly and often dramatically affect the kidney.

nephritis is all too frequently unrecognized until full-blown

nephritic and/or nephrotic syndromes emerge. Renal insufficiency is not a common early manifestation because compensatory hypertrophy and hyperfiltration in less affected
nephrons often mask significant injury in others. Nonetheless, the diligent physician should be able to recognize renal
involvement early in the course of lupus nephritis when
intervention will be the most effective.
A renal biopsy is often used in defining the prognosis
and formulating plans for treatment of lupus nephritis.
Clinicopathologic correlations in the various forms of lupus
nephritis are shown in Table 3. Mesangial nephropathy can
be considered the lowest common denominator of lupus
nephritis. In some cases, mesangial abnormalities are the
expression of intrinsically mild lupus nephritis which does
not warrant specific treatment. More commonly, mesangial
nephropathy represents the early stage of glomerular dis ease in transition to proliferative or membranous forms of
lupus nephritis, but criteria for preemptive treatment have
not been defined. Focal and diffuse proliferative lupus nephritis exhibit pathologic renal lesions which are qualitatively similar and, in essence, represent mild and severe
forms of proliferative disease. Thus, focal and diffuse proliferative lupus nephritis are separated arbitrarily by the pres ence of pathologic lesions which obliterate glomerular capillaries in less than or more than 50% of nephrons,
respectively. Typical lesions of proliferative lupus nephritis
include endocapillary hypercellularity and "wire loop" lesions which represent large but irregular subendothelial
immune complex deposits affecting certain glomerular capillaries (Fig. 1).
In general, the functional consequences of proliferative
lupus nephritis are related to the extent of obliterative
lesions along the filtration surface of glomerular capillaries.
However, many active proliferative lesions are reversible
with effective therapy (e.g., hypercellularity, inflammatory
infiltrates) and renal prognosis depends on the extent of
lesions that heal mostly by scarring (e.g., cellular crescents,
fibrinoid necrosis) or are already irreversible prior to therapy (e.g., glomerular sclerosis, fibrous crescents, tubular
atrophy, interstitial fibrosis). Analysis of renal biopsies is

SYSTEMIC LUPUS ERYTHEMATOSUS

Systemic lupus erythematosus (SLE) is the classic autoimmune disease, with deposition of immune complexes and
infiltration of inflammatory cells in diverse tissue sites. The
spectrum of systemic involvement is reflected in part by
the multiplicity of standard criteria for diagnosis and classification of SLE listed in Table 1. For research purposes,
four or more of these criteria are conventionally used to
classify patients with a diagnosis of SLE. For clinical purposes, diagnosis of SLE may be appropriate with less than
four of the formal criteria, especially in the presence of
other signs, such as alopecia and Raynaud's phenomenon.
Autoantibodies are prominent in the diagnosis of SLE. A
synopsis of the autoantibodies found in SLE and the other
rheumatic diseases with renal components discussed in this
chapter is shown in Table 2.
Chronic fatigue, arthralgias, and skin rashes are among
the most intrusive symptoms and signs of SLE. Paradoxically, major visceral disease, while representing a greater
threat to long-term survival, often produces fewer and less
imposing subjective manifestations and can easily go undetected until substantial damage has accrued. Such is the case
with renal involvement where asymptomatic hematuria or
proteinuria are common initial manifestations and are often
overlooked or misinterpreted as signs of minor genitourinary tract disorders. It is critically important for the clinician, at a minimum, to ask laboratory personnel to examine
the urine in light of suspicion of glomerular disease, or
preferably to examine the urine sediment personally. Lupus

Primer on Kidney Diseases, Second Edition

208

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

28. Renal Manifestations of Systemic Lupus Erythematosus And Other Rheumatic Disorders

209

TABLE I

Diagnosis of Systemic Lupus Erythematosus

Criterion
1. Malar rash
2. Discoid rash

3. Photosensitivity
4. Oral ulcers

Description

Erythema (flat or raised) of malar eminences

Erythematous, scaling plaques with variable skin atrophy
Sunlight-induced or exacerbated rashes
Mucous membrane ulcerations

6. Serositis

Nonerosive, nondeforming arthritis of small joints

Pleuritis, or pericarditis

7. Renal disease

Persistent proteinuria (>0.5 g/d or 3+ dipstick), or cellular (red, white, tubular, or mixed cell) casts

8. Neurologic disease
9. Hematologic disease

Seizures (in the absence of other causes) or psychosis (in the absence of other causes)

5. Arthritis

10. Immunologic disease

Hemolytic anemia or leukopenia (<4000//u,L on two or more occasions), or lymphopenia (<1500/,u,L on

two or more occasions, or thrombocytopenia (<100,000//u,L)
LE cell prep, or anti-DNA antibody, or anti- Sm antibody, or false positive serologic test for syphilis
(STS) Abnormal liter (in absence of predisposing

11. Antinuclear antibody

drugs)

From Tan EM, Cohen AS, Fries JF: Arthritis Rheum 25:1271-1277, 1982.

facilitated by systematic scoring of pathologic lesions and

creation of an activity index (glomerular cellularity, leukocyte exudation, hyaline deposits, necrosis, cellular crescents, and interstitial inflammation) and a chronicity (damage) index (glomerulosclerosis, fibrous crescents, tubular
atrophy, and interstitial flbrosis).
Transitions occur among the various major classes of
lupus nephritis. Mesangial nephropathy may represent an
inherently mild form of disease which may remain as such;
alternatively, mesangial nephropathy may represent an
early stage of disease in evolution to proliferative or membranous forms of lupus nephritis. Proliferative nephropathies may "heal" by regression of hypercellularity and resorption of subendothelial deposits leaving the appearance
of membranous nephropathy (thickened capillary loops
and subepithelial deposits) (Fig. 2). True mixed membranous and proliferative forms of lupus nephritis may coexist;
such cases have a particularly ominous prognosis and treatment should be focused on the proliferative element. Extraglomerular renal vasculopathy occurs in lupus nephritis but

it is rarely due to frank vasculitis. SLE-associated antiphospholipid antibodies seem to increase risk of vasculopathy.
The prognosis of the various forms of lupus nephritis is
estimated from a composite of clinical and pathologic
features. Nephritic urinary sediment and/or worsening nephrotic syndrome, particularly when combined with progressive deterioration of renal function, is ominous and
constitutes a major indication for therapy. The levels of
activity and chronicity indexes in the renal biopsy help to
define prognosis and supplement the clinical indications
for therapy. On renal biopsy, high-grade proliferation, fibrinoid necrosis, cellular crescents, and/or extensive subendothelial immune complex deposits portend a poor prognosis and hence are also considered strong indications for
intensive therapy (Table 3). While isolated laboratory tests
are rarely dependable predictors of an adverse course, falling serum complement levels or rising anti-DNA antibody
liters should prompt an intensified search for other direct
evidence of incipient renal flares. Conversely, fixed nonnephrotic pro teinuria, fixed azotemia, or high chronicity index

TABLE I

Characteristic Autoantibodies in the Renal- Rheumatic Diseases

Disorder

Autoantibodies

Systemic lupus erythematosus

Henoch- Schonlein purpura

Anti-nuclear antibodies (ANA) diffuse pattern, anti-DNA, anti-Sm, anti-phospholipid (cardiolipin)

Systemic sclerosis Mixed

ANA (speckled), anti- Scl-70 (topoisomerase 1), anticentromere

connective tissue disease

ANA, anti- RNP (ribonucleoprotein)

Sjogren's syndrome

ANA, anti- Ro (anti- SS- A), anti-La (anti- SS- B)

Rheumatoid arthritis Behcet's

Rheumatoid factor (mostly IgM antibody to IgG)

syndrome Relapsing

None

polychondritis Familial

None (occasional antibodies to type 2 collagen)

Mediterranean fever

None

None

210

lames E. Balow
TABLE 3

Typical Clinicopathological Features and Correlations in Lupus Nephritis

Morphologic class
I. Normal or minimal abnormality

II. Mesangial nephropathy

Clinical renal manifestations

Usually asymptomatic

Low- grade hematuria and/or proteinuria

Normal renal function

III. Focal proliferative lupus nephritis

Nephritic urinary sediment

Variable but usually nonnephrotic
proteinuria

IV. Diffuse proliferative lupus nephritis

Nephritic and nephrotic syndromes

Hypertension
Variable renal insufficiency

V. Membranous nephropathy

VI. Sclerosing nephropathy

Characteristic renal pathology

Minimal glomerular abnormalities by light
microscopy No glomerular
immune deposits
Mesangial expansion but mostly patent
capillaries Mesangial
immune deposits
Predominantly segmental proliferation,
necrosis, crescents in <50% of glomeruli
Predominantly mesangial and
subendothelial immune deposits
Predominantly global proliferation, necrosis,
crescents in >50% of glomeruli
Variable sclerosis, atrophy, fibrosis
Predominantly mesangial and
subendothelial immune deposits

Nephrotic syndrome

Uniform capillary loop thickening

Subepithelial (and mesangial) immune
deposits

Inactive urinary sediment: broad, waxy

(chronic renal failure) casts

Glomerular obsolescence, tubular atrophy,

and interstitial fibrosis Few
if any immune deposits

on renal biopsy when present in isolation indicate a poor

prognosis and are relative contraindications to treatment.
Corticosteroids are the backbone of treatment of SLE.
However, the threat of pernicious long-term complications
(e.g., osteoporosis, avascular necrosis of bone, cataracts,
and atherosclerosis) mandates that corticosteroid therapy
be used with restraint; this means limited duration of high
doses, frequent attempts at tapering, and alternate day
prednisone therapy whenever possible. Some mild cases of
proliferative lupus nephritis may respond to Corticosteroids
alone. When remission of renal disease is prompt and
nearly complete, simply tapering to alternate-day prednisone over the subsequent month and establishment of a
low maintenance dose may be acceptable. However, in
most forms of proliferative lupus nephritis, the renal response is delayed and incomplete and most consultants
would institute pulse cyclophosphamide as an adjunct to
prednisone within 4 to 8 weeks of starting treatment. Indeed, there is broad agreement that cyclophosphamide
treatment is warranted for severe focal (crescents or necro sis in >25% of glomeruli) and almost all cases of diffuse
proliferative lupus nephritis from the outset. Intermittent
pulse cyclophosphamide is favored because of reduced
long-term toxicity (e.g., bone marrow, gonads, and bladder)
compared to conventional daily oral cyclophosphamide.
Pulse cyclophosphamide is commonly administered
monthly for 6 months as induction therapy; this is followed
by quarterly pulse cyclophosphamide maintenance treatments until remission of nephritis has been sustained for
1 year or longer (see below). Azathioprine has been proposed as an acceptable alternative to the extended courses
of pulse cyclophosphamide during the maintenance period.
The practical roles of cyclosporine or mycophenolate for

either induction or maintenance therapy are uncertain and

undefined. Several controlled studies have failed to show
that plasma exchange therapy is useful in proliferative lupus nephritis.
Stabilization and gradual improvement in clinical and
laboratory features occur in the majority of cases of proliferative lupus nephritis within several weeks of the initiation
of corticosteroid and pulse cyclophosphamide therapy.
However, complete remission of lupus nephritis (i.e., inactive urinary sediment, proteinuria <1 g/day, inactive lupus
serologies) is typically not seen for many months. Exacerbations of lupus nephritis continue to be problematic following all current forms of therapy. Indeed, each major
relapse produces cumulative damage and progressive compromise of renal function no matter how effective is each
induction therapy program. Recent studies have shown that
major exacerbations of severe lupus nephritis are significantly more common after short (6-month) induction
courses of pulse cyclophosphamide than they are after an
extended course of low-intensity maintenance therapy.
Thus, there is a growing consensus that low-intensity maintenance therapy is preferable to repeated episodes of highintensity corticosteroid and cyclophosphamide treatment
for proliferative lupus nephritis.
The renal prognosis of proliferative lupus nephritis has
markedly improved during the last half of the 20th century.
The likelihood of end-stage renal failure or death within
10 years of diagnosis has fallen from more than 80% to
less than 20% in racially diverse populations. The improved
prognosis is due to a composite of medical advances, including more effective antibiotics and antihypertensives, as well
as safer immunosuppressive regimens, particularly pulse
cyclophosphamide therapy. The prognosis of lupus nephri-

28. Renal Manifestations of Systemic Lupus Erythematosus And Other Rheumatic Disorders

211

FIGURE I Diffuse proliferative (Class IV) glomerulonephritis. (A) Light microscopy shows global hypercellularity and
increased mesangial matrix which compromise the filtration surface of the glomerular capillaries. So called "wire loop"
lesions (arrows) formed by massive but irregular immune complex deposits cause irregular thickening of the capillary
loops (hematoxylin and eosin stain). (B) Electron microscopy shows extensive dark-staining subendothelial immune
complex deposits (arrows) disrupting the integrity of the endothelial cells and compromising the capillary lumen. Epithelial
foot process fusion is seen along the outer surface of the glomerular basement membrane. Courtesy of Dr. Sharda Sabnis,
Nephropathology Section, Armed Forces Instit ute of Pathology, Washington, DC.

tis is worse in African-American patients than in comparably treated whites. The basis for this observation is unknown, but differences in racial mix in various clinical
reports may contribute to the discrepancies in apparent
treatment effects in proliferative lupus nephritis.
The indications for treatment of membranous lupus nephropathy are controversial. The prognosis measured by
risk of renal failure is relatively low (approximately 25%
at 15 years). Male gender and massive proteinuria (>10 g/
day) may define subsets of patients with higher risk of renal
progression. Recent reports indicate that the atherogenic
cardiovascular disease and thromboembolic events brought
on by persistent nephrotic syndrome may be additional
reasons for attempting to treat membranous nephropathy.
Angiotensin-converting enzyme inhibitors are commonly
used to reduce proteinuria and HMG-CoA reductase inhib itors are used to reduce hypercholesterolemia, but responses are usually partial until remission of nephrotic
syndrome occurs either spontaneously or with immunosuppressive drug therapy. High-dose alternate-day prednisone

tapered after approximately 2 months is the most commonly used approach. Clinical trials are in progress to evaluate the benefits of adjunctive pulse cyclophosphamide or
low-dose cyclosporine in membranous lupus nephropathy.
HENOCH-SCHONLEIN PURPURA

Henoch-Schonlein purpura (HSP) is a systemic disease,

is found mainly in children and adolescents, and has characteristic features of purpuric rash, arthritis, gastrointestinal
hemorrhage, and glomerulonephritis. The skin lesions are
those of a cutaneous vasculitis, specifically a leukocytoclas tic vasculitis with secondary hemorrhage. There are no
recognized autoantibodies, but Henoch-Schonlein disease
is characterized by IgA deposits in the skin and kidneys,
suggesting a relationship to IgA nephropathy. Transient,
asymptomatic hematuria and proteinuria occur in more
than half of the cases, but they do not warrant treatment.
Hypertension, early appearance of nephrotic syndrome,
persistent nephritic urinary sediment, or progressive azote-

212

lames E. Balow

FIGURE 2 Membranous (Class V) lupus nephropathy. (A) Light microscopy shows uniformly thickened but widely
patent glomerular capillary loops. Hypercellularity is limited to mild mesangial expansion (periodic acid- Schiff stain).
(B) Electron microscopy shows well-preserved but uniformly thickened capillary loops. Extensive dark staining deposits
are seen on the external (subepithelial) side of the glomerular basement membranes. Some examples are marked with
arrows. Courtesy of Dr. Sharda Sabnis, Nephropathology Section, Armed Forces Institute of Pathology, Washington, DC.

mia reflects proliferative glomerulonephritis . More than

50% glomerular cellular crescents forebode an unfavorable
prognosis and are important indications for treatment. Relapsing or chronic courses of Henoch-Schonlein purpura
are unusual, but they increase the likelihood of progres sive nephritis.
Immunosuppressive therapy is warranted in a minority
of patients with Henoch-Schonlein nephritis. In patients
with severe and/or rapidly progressive Henoch-Schonlein
nephritis, high-dose corticosteroids, especially pulse methylprednisolone, are indicated. Cytotoxic drugs, such as cyclophosphamide and azathioprine, and plasma exchange
are often used in refractory cases, but their supplemental
benefits over corticosteroids alone are unproven.
SYSTEMIC SCLEROSIS

Systemic sclerosis or scleroderma is characterized by

hardening of skin and dysfunction of certain internal organs
due to massive connective tissue (notably collagen) accumulation and vasculopathy. Raynaud's syndrome, skin
thickening, digital pulp atrophy (sclerodactyly), and tel-

angectasia are the primary clinical manifestations of systemic sclerosis. Musculoskeletal involvement includes
periarticular tendon and nerve entrapment, flexion contractures, acral osteolysis, and myopathy. Alimentary tract
dysfunction includes esophageal dysmotility and malabsorption, which are major causes of morbidity. Pulmonary
hypertension and interstitial fibrosis, cardiomyopathy, and
scleroderma renal crisis are the major causes of mortality.
While antibodies to nuclear (ANA) and to specific Scl-70
(topoisomerase 1) antigen (Table 2) are present in at least
two-thirds and one-fifth of patients, respectively, very little
is known of the pathogenesis of systemic sclerosis. Current
theories focus on primary disturbances in the vascular endothelium, in the immune and cytokine systems, and in
collagen biosynthesis.
The kidney is affected in about one third of patients with
systemic sclerosis. The clinical spectrum includes reninmediated hypertension, microangiopathic hemolytic anemia (Fig. 3), low-grade proteinuria, and azotemia. In aggressive cases, malignant hypertension and rapidly progres sive renal failure ensue (scleroderma renal crisis). The
kidney pathology includes an occlusive vasculopathy due

28. Renal Manifestations of Systemic Lupus Erythematosus And Oth er Rheumatic Disorders

213

f *Jrj OcP

FIGURE 3 Microangiopathic hemolytic anemia. Peripheral blood

smear showing fragmented red blood cells of various shapes and
sizes, but few platelets (Wright stain). In addition to scleroderma
renal crisis, thrombotic microangiopathy is seen in various forms of
malignant hypertension, disseminated intravascular coagulation,
thrombotic thrombocytopenic purpura, hemolytic uremic syndrome, and necrotizing vasculitis.

to intimal swelling, fibrinoid necrosis and fragmentation,

and duplication of the elastic lamina (Fig. 4). Thrombotic
microangiopathy and rapidly progressive renal failure
sometimes occur in the absence of hypertension and rarely
even before characteristic clinical features indicate a diagnosis of systemic sclerosis.
Although considered an immunologic disorder, systemic
sclerosis does not appear to benefit substantively from currently available immunosuppressive drugs. Corticosteroids
may in fact be deleterious in systemic sclerosis. Overall, 5year survival for diffuse scleroderma is approximately 60%;
in an uncontrolled study of diffuse scleroderma, penicillamine treatment improved survival to 80%. Cardiac, pulmo nary, and especially renal involvement worsens prognosis
significantly. At present, medical therapy in the face of
renal involvement is focused on interrupting the vicious
cycle of vasculopathy, hypertension, and azotemia. The

FIGURE 4 Vasculopathy of scleroderma renal crisis. Interlobular

artery shows severe mucoid intimal thickening wit h fibrin exudation (red, amorphous deposits). Compromise of the vessel lumen
produces a viscous cycle of malignant hypertension and worsening
thrombotic microangiopathy (hematoxylin and eosin stain).

prognosis of scleroderma renal crisis has historically been

extremely poor. However, in recent years, survival has dramatically improved as the result of therapy with angiotensin-converting enzyme inhibitors early in the course (with
the onset of hypertension but before creatinine levels exceed 3 mg/dL). The prognosis of patients with scleroderma
renal crisis treated with angiotensin-converting enzyme inhibitors has improved from <10% to >50% patient and
renal survival. Nearly half of patients needing dialysis have
return of renal function if blood pressure is effectively controlled.
MIXED CONNECTIVE TISSUE DISEASE

Overlap syndromes occur among the rheumatic diseases,

most often with features of systemic lupus, rheumatoid
arthritis, Sjogren syndrome, systemic sclerosis, and polymy ositis. Some authors favor assigning diagnosis to the nearest
discrete connective tissue disease category depending on
the dominant symptoms and signs; others favor using a
separate diagnosis of mixed connective tissue disease

214

lames E. Balow

(MCTD). The latter approach is supported by autoantibodies to extractable nuclear antigens (ENA), particularly to
ribonuclear protein (RNP), which is characteristic of patients with MCTD (Table 2). Renal involvement is variable;
when present in MCTD, the glomerular disease is most
akin to that of mesangial and membranous forms of lupus
nephritis. In dermatomyositis or polymyositis (whether idiopathic or as the predominant component of MCTD),
renal involvement is distinctly uncommon.
SJOGREN'S SYNDROME

Sjogren's syndrome may be primary or may be a secondary feature of rheumatoid arthritis, systemic lupus, MCTD,
systemic sclerosis, or polymyositis. It is characterized by
parotid and other exocrine gland swelling and dysfunction
due to lymphoid infiltration; the resulting mucus secretory
defect produces dry mucous membranes (sicca complex).
Characteristically, Sjogren's is accompanied by profound
hyperglobulinemia and a high frequency of type 1 (distal)
renal tubular acidosis. The long-term positive hydrogen ion
balance must be countered by base (sodium bicarbonate
or sodium citrate) supplementation in order to avert osteoporosis, hypercalciuria, nephrocalcinosis, and nephrolithiasis. Rare cases of glomerulonephritis have been seen in
patients with primary Sjogren's syndrome. The clinical and
pathologic features, as well as indications for treatment,
are akin to those of membranous or proliferative forms of
lupus nephritis.
RHEUMATOID ARTHRITIS

Rheumatoid arthritis is a chronic symmetrical polyarthritis accompanied by periarticular osteoporosis and bone
erosions. Rheumatoid factor (IgM antibody against IgG)
is characteristic and rheumatoid nodules often develop in
patients with high liters of these autoantibodies. Progres sive glomerular disease is uncommon in the natural history
of rheumatoid arthritis. On the other hand, isolated hematuria due to mild mesangial nephropathy is recognized with
increasing frequency, but treatment is usually unnecessary.
Side effects of treatments for rheumatoid arthritis are quite
often the causes of renal disease. Rare cases of nephrotic
syndrome are mostly due to membranous nephropathy;
predilection to membranous nephropathy seems to be increased by gold and penicillamine therapy. Proteinuria and
abnormal urine sediment are considered reasons to discontinue these treatments for rheumatoid arthritis. Secondary
(AA) amyloidosis is an uncommon cause of nephrotic syndrome in severe, debilitated patients with rheumatoid arthritis. Treatment of AA amyloidosis of the kidney is usually ineffective and most cases inexorably progress to endstage renal failure. Proliferative glomerulonephritis caused
by cryoglobulinemia or rheumatoid (necrotizing) vasculitis
is an additional rare cause of renal disease in rheumatoid
arthritis; institution or intensification of cytotoxic drug therapy may be effective in reducing cryoglobulinemia and in
controlling renal vasculitis.

BEH^ET'S SYNDROME

Beliefs syndrome is a male-predominant vasculitic dis ease characterized by painful oral and genital ulcers, arthritis, rash, iritis, uveitis, retinitis, and focal central nervous
system deficits. A thrombotic diathesis leads to vascular
occlusive events which overlap with those of vasculitic lesions. There are no laboratory criteria for this clinical syndrome. Major systemic involvement, including renal dis ease due to a vasculitic-like process, may occur in a small
minority of patients. Rapidly progressive, crescentic glomerulonephritis has been described as a very rare component of Bethel's syndrome.
RELAPSING POLYCHONDRITIS

Relapsing polychondritis is characterized by inflammation of the cartilage of the ears, joints, nose, eyes, and
respiratory tract. Clinically, many of the features are akin
to Wegener's granulomatosis. As in Behget's syndrome,
on rare occasions relapsing polychondritis can escalate to
a severe vasculitis associated with a rapidly progressive
glomerulonephritis.
FAMILIAL MEDITERRANEAN FEVER

Familial Mediterranean fever is an autosomal recessive

disorder characterized by relapsing episodes of fever,
polyserositis, arthritis, and erythematous skin rash. Secondary (AA) amyloidosis is a cause of nephrotic syndrome and
subsequent renal failure in certain susceptible populations,
particularly those of Turkish, Armenian, and Sephardic
Jewish descent. Colchicine is effective in freating the acute
episodes of familial Mediterranean fever and in preventing
secondary amyloidosis.
Bibliography
Austin HA, Balow JE: Henoch- Schonlein nephritis: Prognostic
features and the challenge of therapy. Am J Kidney Dis 2:512520, 1983.
Balow JE, Boumpas DT, Austin HA: Renal Disease. In: Schur P
(ed) The Clinical Management of Systemic Lupus Erythem atosus. 2nd ed. Lippincott-Raven, Philadelphia, pp. 109-126,
1996.
Boumpas DT, Austin HA, Vaughan EM, et al.: Severe lupus nephritis: Controlled trial of pulse methylprednisolone versus two
different regimens of pulse cyclophosphamide. Lancet 340:741745, 1992.
Chang-Miller A, Okamura M, Torres VE, et al.: Renal involvement
in relapsing polychondritis. Medicine (Baltimore) 66:202-217,
1987.
Churg J, Bernstein J, Glassock RJ: Renal Disease: Classification
and Atlas of Glomerular Diseases. Igaku-Shoin, New York, 1995.
Donnelly S, Jothy S, Barre P: Crescentic glomerulonephritis in
Beliefs syndromeResults of therapy and review of the literature, din Nephrol 31:213-218, 1989.
Donohoe JF: Scleroderma and the kidney. Kidney Int 41:462477, 1992.
El-Mallakh RS, Bryan RK, Masi AT, Kelly CE, Rakowski KJ:
Long-term low-dose glucocorticoid therapy associated with re-

29. Diabetic Nephropathy

215

mission of overt renal tubular acidosis in Sjogren's syndrome.

Am J Med 79:509-514, 1985.
Goldstein AR, White RH, Akuse R, Chantler C: Long-term follow- up of childhood Henoch- Schonlein nephritis. Lancet 339:
280-282, 1992.
Grishman E, Churg J, Needle MA, Venkataseshan VS (eds): The
Kidney in Collagen- Vascular Diseases. Raven Press, New
York, 1993.
Helin HJ, Korpela MM, Mustonen JT, Pasternak AI: Renal biopsy
findings and clinicopathologic correlations in rheumatoid arthritis. Arthritis Rheum 38:242-247, 1995.
Kitridou RC, Akmal M, Turkel SB, Ehresmann GR, Quismorio
FP, Massry SG: Renal involvement in mixed connective tissue

disease: A longitudinal clinicopathologic study. Semin Arthritis

Rheum 16:135-145, 1986. Klippel JH, Dieppe PA (eds).
Practical Rheum atology. Mosby,
London, 1995. Said R, Hamzeh Y, Said S, Tarawneh M, alKhateeb M: Spectrum
of renal involvement in familial Mediterranean fever. Kidney
Int 41:414-419, 1992. Steen VD. Renal involvement in systemic
sclerosis. Clin Dermatol
12:253-258, 1994. Steen VD, Costantino JP, Shapiro AP,
Medsger TA: Outcome
of renal crisis in systemic sclerosis: relation to availability of
angiotensin converting enzyme (ACE) inhibitors. Ann Intern
Med 113:352-357, 1990.

DIABETIC NEPHROPATHY
JULIA BREYER

Diabetic nephropathy is the leading cause of end-stage

renal disease (ESRD) in the United States. Patients with
diabetes currently account for 35% of all the patients enrolled in the Medicare ESRD program. Sixty-three percent
of the patients with diabetic nephropathy have type II
diabetes mellitus. The cost to Medicare for ESRD patients
with diabetes mellitus exceeded $2 billion per year in 1990.
In addition to its significant impact on health care costs,
the development of diabetic kidney disease has a dramatic
impact on the morbidity and mortality of patients with
diabetes. It has been estimated that patients with type I
diabetes mellitus and nephropathy as manifested by proteinuria have a 100-fold greater risk of dying relative to a
nondiabetic population. Diabetic patients without proteinuria have only a 2-fold increase in relative mortality.

and that a pool of susceptible patients becomes exhausted.

Thus, if a patient survives 35 years of diabetes without
developing nephropathy, the patient is at extremely low
risk of doing so in the future. Studies of type II diabetes
mellitus in Pima Indians and in western Europe show results similar to those of studies of patients with type I
diabetes mellitus; approximately 40% of these patients will
also develop diabetic nephropathy.
Diabetic African Americans, Mexican Americans,
American Indians, Maori, and Polynesians are among the
minority populations who have a greater risk of developing
diabetic nephropathy and a more rapid progression to
ESRD. It is not known whether these racial differences
are due to genetic, environmental, or other factors.
PATHOGENESIS

EPIDEMIOLOGY

Diabetic nephropathy with proteinuria rarely develops

before ten years duration of diabetes. The cumulative incidence of proteinuria in patients with type I diabetes after
40 years duration of diabetes is 40%. However, the annual
incidence of diabetic nephropathy in type I patients peaks
just before 20 years duration of diabetes and thereafter
declines. This suggests that the risk of developing diabetic
nephropathy is not constant over the duration of diabetes
Primer on Kidney Diseases, Second Edition

Many factors have been postulated to be important in

the pathogenesis of diabetic nephropathy. Normal kidneys
transplanted into patients with diabetes develop diabetic
lesions, where kidneys from patients with diabetes that
were inadvertently transplanted into nondiabetic ESRD
patients resolved their diabetic lesions. These observations
strongly support the notion that exposure to the diabetic
milieu rather than an intrinsic defect in the kidneys predis poses patients to the development of diabetic nephropathy.
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

216

Julia Breyer

Not only hyperglycemia but also insulin deficiency, augmented glucagon and growth hormone levels, and increased ketone bodies have been implicated in the pathogenesis of diabetic nephropathy. Recently, it has been
demonstrated that hyperglycemia leads to the accumulation of advanced glycosylation endproducts in tissues in
patients with diabetes. It has been postulated that this, at
least in part, may be responsible for the endorgan complications of diabetes. Glucose forms chemically reversible early
glycosylation products with proteins at a rate proportional
to the glucose concentration. Some of these early glycosylation endproducts on long-lived proteins undergo a complex
series of chemical rearrangements, forming covalent crosslinks that do not reverse with correction of hyperglycemia.
The presence of these irreversibly cross-linked advanced
glycosylation endproducts (AGE) results in a series of interactions that produce the endorgan damage seen in diabetes, at least in animal models. AGEs accumulate faster
than normal in patients with diabetes and their accumulation parallels the severity of renal disease.
In studies in diabetic rats and in humans with diabetes,
unilateral renal artery stenosis has been reported to protect
the kidney distal to the blocked renal artery from developing the morphologic changes of diabetes. This suggests that
exposure to the systemic blood pressure is an important
factor in the development of diabetic kidney disease. Additionally, abnormal intraglomerular pressures have been
demonstrated to contribute to the development of glomerulosclerosis in the diabetic rat. It has been demonstrated
in diabetic rats that decreasing intraglomerular pressures
by decreasing angiotensin II-mediated efferent arteriolar
vasoconstriction preserves the structure and function of
the glomerulus. Thus, both systemic blood pressure and
intraglomerular pressure appear to have an important role
in the development of diabetic kidney disease.
Finally, other factors have been implicated in the pathogenesis of diabetic nephropathy including dietary protein
excess, alterations in the renin angiotensin axis, altered

prostaglandin production, elevated kinin production, and

elevated atrial naturetic factor levels.
NATURAL HISTORY

At the onset of type I diabetes, functional changes are

present in virtually all patients (Fig. 1). Within a few years,
morphologic changes occur in the kidneys of most patients.
Early nephropathy is manifested by the development of
microalbuminuria. Hyperfiltration, elevations in the mean
arterial blood pressure, and poor blood sugar control can
all be features of this early period. Approximately 176
years after the onset of type I diabetes, frank proteinuria is
manifested. Once proteinuria is established, renal function
inexorably declines with 50% of patients reaching ESRD
7 to 10 years after the onset of proteinuria.
Patients with type II diabetes in whom the true time of
onset of diabetes (e.g., the Pima Indians) is known appear
to have a very similar natural history. However, for most
patients with type II diabetes, the time of onset of diabetes
is unknown and they have likely had diabetes for many
years by the time the diagnosis is made. Thus, they may
have proteinuria and declining renal function even at the
time their diabetes is diagnosed.
Functional Changes
At the time of diagnosis of diabetes, functional changes
present in the kidney include increased kidney size, albuminuria that reverses with blood sugar control, and glomerular hyperfiltration. Increased glomerular filtration rate
(GFR) at the time of diagnosis of type I diabetes has been
shown to be predictive of the development of clinically
overt diabetic nephropathy on average 20 years later. Glomerular hyperfiltration in animal models of diabetes is associated with increments in renal plasma flow, due to renal
vasodilatation, and increases in the glomerular transcapillary hydraulic pressure gradient, which lead to glomerulosclerosis.

Early Nephropathy |
Microalbuminuria
Rising Blood Pressure
0
21
1
1
1 T
Onset of
\
Diabetes
\
Functional Changes
tGFR

Reversible

Albuminuria
\ Kidney size

11-23 I

I
*,
^^\
^

13-25
I

Onset of
Rising
Proteinuria Serum Creatinine

HTN
15-27
I
ESRD

Structural Changes
J Glomerular Basement
Membrane Thickness
Mesangial Expansion

FIGURE I The natural history of diabetic nephropathy. HTN, hypertension; ESRD, end
stage renal disease.

29. Diabetic Nephropathy

217
Pathology

At the onset of type I diabetes, renal biopsies are normal.

Within 1 1/2 to 2 1/2 years glomerular basement membrane
(GBM) thickening, a sensitive indicator for the presence
of diabetes, appears. In addition to GBM thickening, other
distinctive lesions include nodular (Kimmelstiel-Wilson)
and diffuse forms of intracapillary glomerulosclerosis, the
capsular drop lesion, the fibrin cap, and mesangial matrix
expansion (Fig. 2). These changes, along with afferent and
efferent arteriolar hyalinosis and increased renal extracellular membrane albumin and IgG localization in the glomeruli as demonstrated by immunofluorescence microscopy are diagnostic of diabetic nephropathy. Mesangial
expansion correlates with creatinine clearance although no
correlation exists between the GBM thickening and clinical
renal function. Diabetes is primarily a vascular (including
glomerular) disease; however, altered structure and function of the renal tubules and interstitium are also present
(i.e. interstitial fibrous and tubular atrophy). The degree
of interstitial fibrosis correlates with the reduction in GFR.
However, no one has yet demonstrated that renal biopsies
done before the onset of proteinuria will distinguish those
patients who will later develop proteinuria and declining
renal function from those who will not. Biopsies from patients with type II diabetes are indistinguishable from biopsies in type I patients.
Microalbuminuria and Early Nephropathy

A sensitive assay for urinary albumin excretion has

shown that subclinical elevations of urinary albumin excretion in diabetic patients without overt proteinuria predict
the later development of proteinuria and ESRD. Normal
urinary albumin excretion is less than 30 mg/24. Patients
with overt proteinuria have elevated urinary albumin excretion in excess of 300 mg/24 h. With this sensitive assay,
patients with urinary albumin excretion between 30 and
300 mg/24 h can be identified, and this is the range that
defines microalbuminuria. The presence of microalbuminuria in type I diabetes patients is highly predictive of progres -

sion to proteinuria and renal insufficiency during the next

10 to 15 years. Patients with type II diabetes also develop
microalbuminuria which predicts progression to pro
teinuria and declining renal function. Excessive cardio
vascular mortality has been reported in patients with type
11 diabetes and microalbuminuria. Due to premature car
diovascular deaths, there is a discrepancy between the num
ber of patients with early nephropathy and those who de
velop ESRD.
Elevated blood pressure has also been demonstrated to
precede and perhaps predict the development of diabetic
nephropathy. Patients with microalbuminia or overt nephropathy are predisposed to hypertension, as indicated
by elevated sodium-lithium countertransport activity in
erythrocytes and a strong parental history of hypertension,
compared to patients with diabetes but without nephropathy. Poor glycemic control also synergistically increases
the risk of nephropathy in patients who are genetically
predisposed to hypertension.
Proteinuria and Declining Renal Function

The presence of persistent proteinuria heralds the overt

phase of diabetic nephropathy. Once proteinuria is established, renal function inevitably declines. The proteinuria
can be as little as 500 mg/24 h or greater than 20 g/24 h.
In the vast majority of patients, GFR begins to decline
with the onset of proteinuria, with the average time of
ESRD being 10 years after the onset of proteinuria (Fig.
1). Renal function can also be adversely affected by other
complications of diabetes, such as neurogenic bladder, urinary tract infection, and papillary necrosis. As renal function declines, uremia ensues. Insulin requirements decrease
in uremia in part because the kidney is responsible for 30
to 40% of the catabolism of insulin and also because patients lose their appetite as renal function declines. Diabetes is also the most common cause of type IV renal tubular
acidosis, which results in hyperkalemia and a hyperchloremic metabolic acidosis. Thus, patients with this tubular
defect can have hyperkalemia and acidosis despite relatively well-preserved GFR.
SCREENING

FIGURE 2 Diffuse and early nodular diabetic glomerulosclerosis.

The peripheral structures at the right that consist of hyalinized
eosinophilic material are fibrin caps, (periodic acid- Schiff stain).

Based on current recommendations, all patients with

type I diabetes of greater than 5 years duration should
have annual screens for microalbuminuria to identify this
high-risk population. It is important to specifically order
testing for urinary albumin excretion or microalbuminuria
since routine urine protein determinations are not sensitive
enough to detect microalbuminuria. A complete 24-hour
urine collection, an overnight urine collection, which can
be extrapolated to 24 hours, or a spot morning urine with
measurement of the albumin: creatinine ratio may be employed. As shown in Table 1, hyperglycemia, uncontrolled
hypertension or congestive heart failure, urinary tract infections, and excessive physical exercise must be considered
as potential causes of transient microalbuminuria. Thus, if
the urinary albumin excretion rate is elevated in a single

218

Julia Breyer
TABLE I

Measurement of Microalbuminuria
Test IDDM patients of greater than 5 years duration every
year. Test NIDDM patients at the time of diagnosis and
every year.
Rule out causes of transient microalbum inuria: hyperglycemia,
UTI, PE, essential HTN, CHF, water loading.
If the albumin excretion rate is elevated, repeat three times
over 3-6 months to define persistent microalbuminuria.
Normal albumin excretion: <30 mg/24 h; microalbuminuria:
30-300 mg/24 h; proteinuria: >300 mg/24 h
IDDM, insulin- dependent diabetes mellitus; NIDDM, noninsulin-dependent diabetes mellitus; UTI, urinary tract infection;
PE, physical exercise; HTN, hypertension; CHF, congestive
heart failure.

collection, persistent microalbuminuria should be confirmed with three additional collections over 3 to 6 months.
Patients with type II diabetes should be screened at the
time of diagnosis and yearly thereafter.
When the patient with diabetes develops proteinuria, an
evaluation for the cause of proteinuria should be undertaken. Ninety to 95% of type I patients with diabetic nephropathy will have diabetic retinopathy. The absence of
retinopathy should make one suspect a cause of proteinuria
other than diabetic nephropathy. In patients with type I
diabetes, proteinuria from diabetic nephropathy occurs in
a particular time frame: 176 years after the onset of diabetes. If a patient with type I diabetes first manifests proteinuria after 5 years of diabetes or after 30 years of diabetes
they are unlikely to have diabetic nephropathy as the cause.
Patients should be evaluated carefully for the presence of
other systemic diseases that can cause proteinuria such as
hepatitis B, systemic lupus erythematosus, or amyloidosis.
Renal ultrasounds should be performed to rule out anatomic abnormalities. If any evidence of another disease
process is present, a renal biopsy should be considered.
The presence of significant hematuria or red blood cell
casts should also prompt consideration of a renal biopsy.
However, tor the vast majority of patients with diabetes
and proteinuria who present with retinopathy, a benign
urinalysis, no evidence of another systemic disease, and
proteinuria in the appropriate time frame, a renal biopsy
is not necessary for diagnosis.
TREATMENT OF PROGRESSIVE DISEASE
The major therapeutic interventions that have been evaluated include: improved glycemic control, antihypertensive
therapy, treatment with angiotensin converting enzyme
(ACE) inhibitors, and restriction of dietary protein intake.
Tight control of blood glucose prevents the development
of and ameliorates established diabetic nephropathy in animal studies. Multiple epidemiologic studies have implicated
hyperglycemia in the pathogenesis of the long-term complications of diabetes in humans. Recently, the Stockholm
Diabetes Intervention Study demonstrated a beneficial ef-

fect of intensive blood sugar control in patients with type

I diabetes and established complications.
The Diabetes Control and Complications Trial (DCCT)
randomly assigned 1441 patients with type I diabetes to
either intensive therapy, achieving a mean hemoglobin AiC
of 7%, or conventional therapy, achieving a mean hemoglobin AiC of 9%. At baseline, the patients were in two cohorts: 726 with no evidence of end organ complications (the
primary cohort), and 715 patients with early complications
primarily of mild retinopathy (the secondary cohort). In
both cohorts, microalbuminuria developed in fewer patients in the intensive therapy group than in the conventional therapy group, with a risk reduction of 34% within
the primary cohort and 43% in the secondary cohort. The
risk of developing proteinuria was reduced by 56% in the
secondary cohort. The chief adverse effect associated with
intensive blood sugar control was a two-to three fold increase in severe hypoglycemic episodes requiring assis tance. In the secondary analysis, the beneficial effects of
intensive blood sugar control were a continuous function,
so that any decrease in hemoglobin A^ values was associated with decreased risk of developing diabetic nephropathy. Thus, intensive blood sugar control prevents the development and progression of early diabetic nephropathy. In
patients with type II diabetes, far fewer studies have been
completed, although several are ongoing to examine the
impact of intensive blood sugar control in preventing the
development of diabetic nephropathy.
Multiple studies have demonstrated that in type I diabetes patients with proteinuria and declining renal function,
lowering the systemic blood pressure slows the rate of
decline in renal function and improves patient survival.
Similar results have been found in patients with type II
diabetes. Also, in patients with early diabetic nephropathy
manifested by microalbuminuria, ol wering the systemic
blood pressure has been demonstrated to lower urinary
albumin excretion rates. Thus, even in early diabetic kidney
disease, the lowering of systemic blood pressure appears
to be of benefit. The definition of adequate blood pressure
control in patients with diabetic nephropathy is unclear.
The current target recommendation is a blood pressure at
or below 135/85 mm Hg. It has also been suggested that
the goal of blood pressure therapy should be a documented
decrease in urinary albumin exc retion rate or proteinuria.
Studies in animals with diabetic kidney disease demonstrated that ACE inhibitors preserved renal structure and
function independent of their effect on systemic blood pres sure. In these diabetic animals, the mechanism of protection was interruption of the vasoconstrictor effect of angiotensin II on the efferent arteriole, leading to lower
glomerular intracapillary pressures while preserving renal
plasma flow. Treatment with ACE inhibitors in patients
with microalbuminuria has been reported to decrease the
urinary albumin excretion rate and decrease the progres sion to overt proteinuria.
In a clinical trial, patients with type I diabetes and diabetic nephropathy (manifested by >500 mg proteinuria per
24 hours and serum creatinine =2.5 mg/dL) were randomized to receive either captopril 25 mg three times a day or

219

29. Diabetic Nephropathy

placebo. The primary outcome was time to doubling of

serum creatinine to at least 2 mg/dL, which approximated
halving the patient's GFR. Patients receiving captopril had
a 48% risk reduction for doubling of serum creatinine and
a 50% risk reduction for the secondary outcome of time to
death, dialysis, or transplantation compared to the placebo
group. The renoprotective effect of ACE inhibition was
independent of its effect on systemic blood pressure. ACE
inhibitors were equally efficacious in normotensive and
hypertensive patients. Thus, the use of ACE inhibition is of
clear benefit in patients with type I diabetes and established
nephropathy. A beneficial effect of ACE inhibitors in patients with type II diabetes is less well established.
A beneficial effect of calcium channel blockers in diabetic nephropathy has not been established. Far fewer studies have been performed and the available studies were
done with small numbers of patients, most of whom had
type II diabetes. Recently it has been suggested that a
combination of ACE inhibitors and calcium channel blockers may be of benefit. But again, few data support this
recommendation.
In many experimental animal models of renal disease,
including diabetes, high dietary protein intake accelerates
the deterioration of renal function. Several studies done
in small numbers of patients with diabetes and diabetic
nephropathy have suggested a beneficial effect of low protein diets in slowing the rate of decline of renal function.
A variety of other interventions have been attempted in
small studies in either diabetic rats or humans, including
the use of aldose reductase inhibitors, treatment with inhibitors of the formation of AGES, treatment of dyslipidemia,
treatment with octreotide, treatment with antiplatelet
drugs, treatment with thromboxane synthetase inhibitors,
and treatment with the protease inhibitor camostate mesilate. None of those interventions have as yet been
proven efficacious.
The 40% of patients with diabetes who develop diabetic
nephropathy must be identified as early in their course as
possible. Interventions noted above can be implemented
to slow and halt the progression of their renal disease even
at very early stages such as when the patient has only
microalbuminuria. Once persistent elevation of the urinary
albumin excretion is found, the patient should be referred
to a nephrologist for further evaluation and for consideration of implementation of specific therapeutic interventions. The nephrologist can also evaluate the patient to
ensure that no other cause of the proteinuria besides diabetic nephropathy is present. In addition to implementing
specific therapeutic interventions it is important to help
preserve the patients remaining renal function by avoiding
other insults such as excessive use of nonsteroidal antiinflammatory agents, the unnecessary use of aminoglycoside
antibiotics, and by properly managing urinary tract infections or distinctive uropathy due to diabetic neurogenic
bladder. Patients with established diabetic nephropathy are
at particularly high risk for developing acute renal failure
secondary to exposure to IV contrast agents. Because of
their high incidence of vascular complications, patients with
diabetes are often candidates for radiocontrast administra-

tion for cardiac catherizations or other diagnostic procedures. Minimizing exposure to radiocontrast agents plays
an important role in preserving residual renal function. If
exposure to a radiocontrast agent is necessary, ensuring
adequate intravascular volume at the time of the study is
critical, (see Chapters 6 and 36).
Lastly, patients with diabetes develop uremia and require
renal replacement therapy earlier than patients with other
forms of renal disease. Patients with diabetes will often
have a lower GFR for any given serum creatinine than a
nondiabetic patient with renal insufficiency. Thus, it is not
uncommon for a patient with diabetes to be near ESRD,
even with a serum creatinine as low as 2 to 3 mg/dL. Diabetic patients also develop symptoms of uremia at higher
GFR levels (10-15 mL/min) than patients with nondiabetic
kidney disease. Establishing a vascular access for hemodialysis may be difficult because of coexistent vascular disease.
Because diabetic patients have a high prevalence of cardiovascular disease, their transplantation evaluations may require extensive cardiac testing. For all of these reasons,
dialysis teaching and planning as well as transplant evaluations should begin early.
TREATMENT OF ESRD

Once patients with diabetic nephropathy reach ESRD,

their options are hemodialysis, peritoneal dialysis, or transplantation. Figure 4 in Chapter 65 shows the first year
mortality rates for dialysis patients by year of first ESRD
therapy and primary cause of ESRD. Patients with diabetes
have the highest first year mortality throughout the years
shown. Encouraging, however, is the overall decline in adjusted one year death rates for patients with diabetes from
1983 through 1993. Patients with diabetes who receive a
cadaveric renal transplant also have a higher morbidity and
mortality than the nondiabetic transplant patient. However, 2-year renal graft and patient survival rates in the
type I patient with a living related donor transplant are
comparable to nondiabetic ESRD patients. Survival is
longer for the patient with diabetes who receives a living
related donor transplant compared to a patient who remains on dialysis. Combined renal pancreas transplants
have a higher morbidity rate than renal transplant alone
but have improved glycemic control and motor and sensory nerve function, as well as less extensive morphologic
changes of diabetic nephropathy in the allograft. No benefits are accrued in pancreatic polypeptide secretion, preservation of kidney function, or retinopathy, however.
The increased morbidity and mortality rates in ESRD
patients with diabetes are primarily secondary to complications of atherosclerotic disease. Multiple investigators have
demonstrated that the risk of cardiovascular disease in type
I patients with renal disease far exceeds that in duration
matched type I patients without nephropathy (Fig. 3).
There are not only increased rates of cardiovascular events,
but also excessive cerebral and peripheral vascular complications in patients with diabetes on dialysis or posttransplant. Malnutrition, as evidenced by low serum albumin
values and infections, also complicate the renal replace-

Paul W. San ders

220

40

with
nephropathy
<1> (0
U 01
C U>

30-

'5r 20c ra
. <u

101
!
without nephropathy
3 O

oo
Years After Onset of Proteinuria
FIGURE 3 Cumulative incidence of coronary heart disease in
patients with () and without (O) nephropathy. (Reprinted from
Breyer, 1992, with permission.)

tional Kidney Foundation from an ad hoc committee of the

Council on Diabetes Mellitus of the National Kidney Foundation. Am J Kidney Dis 25:107-112, 1995.
Breyer JA: Diabetic nephropathy in insulin - dependent patients.
Am J Kidney Dis 6:533-547, 1992.
Breyer JA: Medical management of nephropathy in type I diabetes mellitus: Current recommendations. J Am Soc Nephrol
6:1523-1529, 1995
DeFronzo RA: Diabetic nephropathy: Etiologic and therapeutic
considerations. Diabetes Rev 3:510-564, 1995.
The Diabetes Control and Complications Trial Research Group.
The effect of intensive treatment of diabetes on the development
and progression of long-term complications in insulin-dependent
diabetes mellitus. N Engl J Med 329:977-986, 1993.
The Diabetes Control and Complications Research Group: Effect
of intensive therapy on the develo pment and progression of
diabetic nephropathy in the Diabetes control and Complications
Trial. Kidney Int 7:1703-1720, 1995.
Humphrey LL, Ballard DJ, Frohnert PP, et al.: Chronic renal
failure in non-insulin-dependent diabetes mellitus. Ann Intern
Med 111:788-796, 1989.
Lewis EJ, Hunsicker LG, Bain RP, et al.: The effect of angio tensin -converting enzyme inhibition on diabetic nephropathy.
TV Engl J Med 329:1456-1462, 1993.
Mattock MB, Morrish NJ, Viberti G, et al.: Prospective study of
microalbuminuria as predictor of mortality in NIDDM. Diabetes
41:736-741, 1992.
Mogensen CE: Microalbuminuria as a predictor of clinical diabetic
nephropathy. Kidney Int 3:673-689, 1987.
Ritz E Stefanski A: Diabetic nephropathy in type II diabetes Am
J Kindey Dis 27:167-194, 1996.
Tuck ML, Corry, DB: Pathophysiology and management of hypertension in diabetes. Annu Rev Med 42:533-58, 1991.

ment therapy of patients with diabetes. Thus, despite advances in renal replacement therapies, malnutrition, infection, and atherosclerotic disease shorten the life of the
patient with diabetes and ESRD.
Bibliography

Bennett PH, Haffner S, Kasiske BL, et al.: Screening and manage ment of microalbuminura in patients with diabetes mellitus:
Recommendations to the scientific Advisory Board of the Na-

DYSPROTEINEMIAS AND AMYLOIDOSIS

PAUL W. SANDERS

LIGHT-CHAIN-RELATED RENAL DISEASES

Overview

Following the description of an abnormal urinary protein

initially identified by Dr. William Macintyre and subsequently characterized by Dr. Henry Bence Jones in 1847,

Bence Jones proteins have been found to represent immu -

Primer on Kidney Diseases, Second Edition

noglobulin light chains, and the multiple renal lesions associated with deposition of these proteins have been studied
extensively. Plasma cells synthesize light chains that become part of the immunoglobulin molecule. Each light
chain possesses two independent globular regions, termed
constant and variable domains. The variable domain forms
part of the antigen-binding site and derives from rearrange-

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

221

3 0 . D y s p r o t e i n e m i a s a n d A m yloidosis

ment of more than 20 gene segments. Thus, despite similar

biochemical properties, no two light chains are identical.
Disulfide bonding among two light chains with higher mo lecular weight proteins (heavy chains) occurs during or
shortly after heavy-chain synthesis, forming the classical
immunoglobulin G molecule that is then secreted. A slight
excess production of light, compared to heavy, chains appears to be required for efficient immunoglobulin synthesis,
but may result in release of free light chains. Light chains
can also form homodimers through disulfide bonding to
another light chain before secretion.
Once in the circulation, light chains are handled simi larly to other low-molecular-weight proteins, which are
cleared by the kidney. Unlike albumin, monomers (22
kDa) and dimers (44 kDa) are readily filtered at the
glomerulus and reabsorbed by the proximal tubule. In the
proximal tubule, endocytosis of light chains occurs through
a single class of receptors with relative selectivity for these
proteins. The isoelectric point of light chains does not influence reabsorption rate. After endocytosis, lysosomal enzymes hydrolyze the proteins, and the amino acid components are returned to the circulation. Reabsorption is
saturable and allows the delivery of light chains to the distal
nephron and their ultimate appearance in the urine as
Bence Jones proteins.
Urinary Bence Jones proteins possess unusual heat solubility properties. When heated to 60C, the proteins precipitate, but upon further heating to 100C, the proteins resolubilize. Although this unusual thermal property was the
standard screening test for the presence of Bence Jones
proteins, it is very insensitive and positive only when significant quantities are excreted in the urine. The qualitative
urine dipstick test for protein also has low sensitivity for
light chains. In healthy adults, the approximate urinary
concentration of light chain proteins, which are polyclonal
because of escape into the circulation of small amounts of
free light chain produced during normal immunoglobulin
assembly, is 2.5 /ug/mL. Urinary light chain concentration
is generally between 20 and 500 /xg/mL in patients with
monoclonal gammopathies of undetermined significance,
and is usually much higher (range 0.02-11.8 mg/mL) in
patients with multiple myeloma or Waldenstrom's macroglobulinemia. The amount of light chain excreted is often insufficient for detection using turbidimetric and heat
tests. In addition, because of the insensitivity of routine
serum protein electrophoresis (SPEP) and urinary protein
electrophoresis (UPEP), these tests are no longer recommended. Identification therefore rests with antibody detection assays (immunofixation electrophoresis or immunoelectrophoresis) using serum and urine. Immunofixation
electrophoresis is very sensitive and detects monoclonal
light chains and immunoglobulins even in very low concentrations. The causes of monoclonal light-chain proteinuria,
a hallmark of plasma cell dyscrasias, are listed (Table 1).
The most common cause of light chain proteinuria is multiple myeloma.
Virtually every compartment of the kidney can be damaged from monoclonal light-chain deposition (Table 2).
Histologic examination of necropsy specimens from 57 pa-

TABLE I

Causes of Bence Jones Proteinuria

Multiple myeloma
AL-amyloidosis

Light-chain deposition disease

Waldenstrom's macroglobulinemia
Monoclonal gammopathy of undetermined significance
Heavy-chain disease (fj.) (rare)
Lymphoproliferative disease (rare)
Rifampin therapy (rare)

tients demonstrated renal lesions in 48 to 49%. Sixty-five

percent of those patients with renal lesions had cast nephropathy, or "myeloma kidney," while 21% had AL-amyloidosis and 11% had monoclonal light chain deposition
disease (LCDD). The light chain takes the center stage in
the pathogenesis of these renal lesions. The type of renal
lesion induced depends on the physicochemical properties
of the light claim, and the simultaneous occurrence of two
or more of these lesions in the same patient is unusual.
Cast Nephropathy
Pathology

Cast nephropathy is a noninflammatory tubulointerstitial

renal lesion. Characteristically, multiple intraluminal proteinaceous casts are identified mainly in the distal portion
of the nephrons (Fig. 1). Casts may also be seen in the
proximal tubule, or even in glomeruli when they are abundant. The casts are usually acellular, homogeneous, and
eosinophilic with multiple fracture lines. Immunofluores cence and immunoelectron microscopy confirm that the
casts contain light chains and Tamm-Horsfall glycoprotein.
Persistence of the casts produces giant cell inflammation
and tubular atrophy that typify myeloma kidney. Glomeruli
are usually normal.

TABLE 2

Light-Chain-Related Renal Lesions

Glomerulopathies
Light-chain depositio n disease
AL-amyloidosis
Cryoglobulinemia
Fibrillary glomerulonephritis (rare)

Immunotactoid glomerulopathy (rare)

Tubulointerstitial lesions
Cast nephropathy ("myeloma kidney")
Fanconi syndrome
Proximal tubule injury (acute tubular necrosis)
Tubulointerstitial nephritis (rare)
Vascular lesions
Asymptomatic Bence Jones proteinuria
Hyperviscosity syndrome Neoplastic cell
infiltration (rare)

222

Paul W. Sanders

FIGURE I Renal biopsy tissue from a patient with myeloma cast

nephropathy showing tubules with multinucleated giant cells that
contain cast material. Some of the cast material has angular or
rhomboid contours. Note that there is also tubular epithelial atrophy as well as interstitial edema, fibrosis, and infiltration by mono nuclear leukocytes (hematoxylin -eosin stain).

Clinical Features

Renal failure from this lesion may present acutely or as

a chronic progressive disease and may develop at any stage
of the disease. The diagnosis is usually evident when
chronic bone pain, pathologic fractures, and hypercalcemia
are complicated by proteinuria and renal failure. Recently,
more patients present with symptoms of renal failure or
undefined proteinuria; further evaluation then confirms a
malignant process. Cast nephropathy should therefore be
considered when proteinuria (often more than 3 g/day),
particularly without concomitant hypoalbuminemia or albuminuria, is found in a patient who is in the fourth decade
of life or older. Hypertension is not common in cast nephropathy. The diagnosis may be confirmed by finding
monoclonal immunoglobulins or light chains in the serum
and urine and typical intraluminal cast formation on kidney
biopsy. Virtually all patients with cast nephropathy have
detectable monoclonal light chains in the urine or blood.
Pathogenesis

Myeloma casts contain Tamm-Horsfall glycoprotein and

occur initially in the distal nephron, which provides an
optimum environment for precipitation of light chains
(Table 3). Casts occur in part because light chains coaggre-

gate with Tamm-Horsfall glycoprotein. Tamm-Horsfall gly coprotein, which is synthesized exclusively by cells of the
thick ascending limb of the loop of Henle, is the major
fraction of urinary protein and the predominant constituent
of urinary casts in normal individuals. Cast-forming Bence
Jones proteins bind to a common portion of the peptide
backbone of Tamm-Horsfall glycoprotein; binding results
in coaggregation of these proteins and subsequent occlusion of the tubule lumen by precipitated protein complexes.
Intranephronal obstruction and renal failure ensue. Light
chains with high affinities for Tamm-Horsfall glycoprotein
are potentially nephrotoxic. Intravenous infusion of nephrotoxic human light chains in rats elevates proximal tubule
pressure and decreases single nephron glomerular filtration
rate; intraluminal protein casts can be identified in these
kidneys.
Coaggregation of Tamm-Horsfall glycoprotein with light
chains also depends upon the ionic environment and the
physicochemical properties of the light chain, because not
all patients with myeloma develop cast nephropathy, even
when the urinary excretion of light chains is significant.
Increasing concentrations of sodium chloride or calcium,
but not magnesium, facilitate coaggregation. The loop diuretic furosemide augments coaggregation and accelerates
intraluminal obstruction in vivo in the rat. Finally, the lower
tubule fluid flow rates of the distal nephron allow more
time for light chains to interact with Tamm-Horsfall glycoprotein and subsequently to obstruct the lumen. Situations
where flow rates are reduced, such as volume depletion,
can accelerate tubule obstruction or convert nontoxic light
chains into cast-forming proteins.
Treatment and Prognosis

The principles used to treat cast nephropathy include

decreasing circulating light chain and preventing coaggregation of light chain with Tamm-Horsfall glycoprotein
(Table 4). Prompt and effective chemotherapy should start
on diagnosis of multiple myeloma. Standard treatment employs alkylating agents and steroids. Intermittent treatment
with melphalan and prednisone decreases circulating levels
of light chains and stabilizes or improves renal function in
two thirds of patients who present with renal failure. Fanconi syndrome may also resolve with this therapy. An alternative chemotherapeutic regimen that includes vincristine,
adriamycin, and either methylprednisolone (VAMP) or
dexamethasone (VAD) has been used successfully. VAD
can induce a remission more rapidly than melphalan and
prednisone and allow faster reduction in the amount of

TABLE 3

Factors Affecting Cast Formation

TABLE 4

Concentration and type of light chain

Concentration and carbohydrate content of Tamm-Horsfall
glycoprotein Distal
nephron [NaCl] Distal
nephron [Ca2+] Tubule
fluid flow rate
Presence of furosemide or radiocontrast agents
Acidic urine

Standard Therapies for Cast Nephropathy

Chemotherapy to decrease light-chain production
Increase free water intake to 2-3 L per day as tolerated
Treat hypercalcemia aggressively Avoid exposure to
furosemide, radiocontrast agents, and
nonsteroidal anti-inflammatory agents
Alkalinize urine

223

30. Dysproteinemias and Amyloidosis

circulating light chains. Dosing of VAD may be easier since

the elimination of melphalan, which is mainly excreted in
the urine, is problematic in patients with renal failure. Thus,
VAD may be particularly beneficial in the setting of renal
failure related to deposition of light chains, and the physician can rapidly determine the efficacy of such an approach.
Myeloablative therapy followed by allogeneic bone marrow transplantation may also prove effective in controlling
renal failure in myeloma, but it has a significant mortality
and its use is currently limited to the small population who
are suitable and have an HLA -compatible relative. The role
of autologous bone marrow transplantation is uncertain.
Plasmapheresis removes light chains from bloodstream effectively over a short period and is useful primarily in the
setting of acute renal failure or hyperviscosity syndrome.
Prevention of aggregation of light chains with TammHorsfall glycoprotein is the second cornerstone of therapy.
Volume repletion, normalization of electrolytes, and avoidance of complicating factors such as furosemide, radiocontrast material, and nonsteroidal anti-inflammatory agents
are helpful in preserving and improving renal function.
Daily fluid intake, up to 3 L in the form of electrolyte-free
fluids, should be encourated as long as defects in osmoregulation do not manifest. Alkalinization of the urine with
oral sodium bicarbonate (or citrate) to keep the urine pH
>7 may also be therapeutic and should be attempted.
Hypercalcemia occurs in more than 25% of patients with
multiple myeloma. In addition to being directly nephro toxic, hypercalcemia enhances the nephrotoxicity of light
chains. In most cases, treatment of volume contraction with
the infusion of saline will correct the hypercalcemia. Loop
diuretics also increase calcium excretion, but furosemide,
because it may facilitate nephrotoxicity from light chains,
should not be administered until the patient is clinically
euvolemic. Glucocorticoid therapy (prednisone, 60 mg/
day) is helpful for acute management of the multiple my eloma as well as hypercalcemia. However, the response to
the latter is not dramatic. Rarely, more aggressive management is necessary. The bisphosphonates, pamidronate and
etidronate, are used to treat moderate hypercalcemia (serum calcium >3.25 mmol/L, or 13 mg/dL) unresponsive to
other measures. Though the exact mechanism of action is
debated, bisphosphonates lower serum calcium by interfering with osteoclastic bone resorption. While hypercalcemia
of myeloma responds to bisphosphonates, these agents are
nephrotoxic and should be administered only to euvolemic
patients. Therapy with pamidronate allows out-patient
management of mild hypercalcemia. Although their efficacy is transient, use of the bisphosphonates may allow time
for chemotherapy and hydration to prevent recurrence of
hypercalcemia.
Dialysis should be used as indicated. Perhaps 5 to 10%
of patients who are dialysis dependent because of cast
nephropathy will regain sufficient renal function to stop
dialysis. Renal transplantation has also been successfully
performed in selected patients with multiple myeloma in remission.
Renal failure decreases survival in multiple myeloma.
Median survival for patients with renal failure is 20 months,

compared with a median survival of 20 to 40 months for

the general population of patients with myeloma. Major
predictors of survival include the stage of the disease, decline in serum creatinine concentration at 1 month into
treatment to 3.4 mg/dL, and response to chemotherapy.
Response to chemotherapy is important: Median survival
time is 36 months for responders, but only 10 months for
nonresponders.
Other Tubulointerstitial Renal Lesions

Proximal tubule injury, including Fanconi syndrome, and

tubulointerstitial nephritis can also rarely occur. Fanconi
syndrome may precede overt multiple myeloma. Plasma
cell dyscrasia should therefore be considered in the differential diagnosis when this syndrome occurs in adults. More
severe damage to proximal tubule epithelium can produce
clinical manifestations of acute renal failure. The mechanism of damage to the proximal epithelium is related to
accumulation of toxic light chains in the endolysosomal
system. An inflammatory tubulointerstitial nephritis with
cellular infiltrates including eosinophils and active tubular
damage has also been described. A careful search usually
detects subtle light chain deposits along the tubular basement membranes almost exclusively in the areas of interstitial inflammation. This interstitial inflammatory pattern is
sometimes associated with glomerular involvement, and
cast formation is rare. Without detection of the deposits
of light chains along the basement membrane, this lesion
may be mistakenly considered to be a hypersensitivity reaction to drugs, most notably nonsteroidal antiinflamma tory agents.
Light- Chain Deposition Disease

Light-chain deposition diseas e (LCDD) is a systemic

disease but typically manifests renal failure related to a
glomerular lesion associated with nonamyloid electrondense granular deposits of monoclonal light chains with or
without heavy chains. LCDD may be accompanied by other
clinical features of multiple myeloma or another lymphoproliferative disorder, or may be the sole manifestation of
a plasma cell dyscrasia.
Pathology

Nodular glomerulopathy with distortion of the glomerular architecture by deposition of amorphous, eosinophilic
material is the most common pathologic finding observed
with light microscopy (Fig. 2). These nodules, which are
composed of light chains and extracellular matrix proteins,
begin in the mesangium, and their appearance is reminis cent of diabetic nephropathy. Less commonly, other glomerular morphologic lesions besides nodular glomerulopathy can be seen in LCDD. Immunofluorescence microscopy
demonstrates the presence of monotypical light chains in
the glomeruli. Under electron microscopy, linear deposits
of light-chain proteins are present in a subendothelial position along the glomerular capillary wall, along the outer
aspect of tubular basement membranes, and in the mesan-

Paul W. Sanders

224

While light chains play an important pathogenetic role

in these glomerular lesions, heavy chains, along with light
chains, can be identified in the deposits, prompting some
authors to suggest the term monoclonal light-chain and
heavy-chain deposition disease. In those specimens, the
punctate electron-dense deposits appear larger and more
extensive than those deposits that contain only light chains,
but it is unclear whether the clinical course of these patients
differs from the course of isolated light-chain deposition
without heavy-chain components.
Treatment and Prognosis

FIGURE 2 Gloraerulus from a patient with K light-chain deposition disease showing marked nodular mesangial matrix expansion
with slight hypercellularity (hematoxylin -eosin stain).

gium. Tubular cell damage is also notable and obvious in

some patients, even in the early stage of the disease.

The treatment of LCDD is difficult. Randomized controlled trials for treatment of this light chain-related renal
disease are unavailable, but these patients appear to benefit
from the same chemotherapy as that given for multiple
myeloma, particularly if renal failure is mild at presentation. Five of eight patients with serum creatinine concentrations less than 4.0 mg/dL at the time of diagnosis did not
progress with chemotherapy, while 9 of 11 patients with
higher creatinine concentrations progressed to end-stage
renal failure despite therapy. The 5-year survival is approximately 70%, and is reduced by coexistent myeloma.

Clinical Features

The clinical presentation is typical of a progressive glomerulonephritis. The major symptoms of LCDD are proteinuria, sometimes in the nephrotic range, microscopic
hematuria, and renal failure. Albumin and monoclonal
light chains are dominant proteins in the urine, and the
presence of albuminuria and other findings of nephrotic
syndrome are important clues to the presence of a glomerular lesion. The amount of excreted light chain is usually less
than that found in cast nephropathy. Although extrarenal
manifestations of overt multiple myeloma can manifest at
presentation or over time, a majority (74%) of patients
with LCDD will not develop myeloma or other malignant
lymphoproliferative disease. Other organ dysfunction, es pecially liver and heart, can develop and is related to deposition of light chains in those organs. Development of renal
failure in untreated patients is common. Because renal
manifestations generally predominate and are often the
sole presenting features, the diagnosis of their plasma cell
dyscrasia is not infrequently first made during the evaluation of unexplained renal abnormalities. Renal biopsy,
which is necessary to establish the diagnosis, is recommended early in the course.
Pathogenesis

LCDD has a pathogenesis different from that of cast

nephropathy. The response to light-chain deposition includes expansion of the mesangium by extracellular matrix
proteins to form nodules and eventually glomerular sclerosis. Recent experimental studies have shown that mesangial
cells exposed to light chains obtained from patients with
biopsy-proven LCDD produce transforming growth factor|8, which serves as an autacoid to stimulate these same cells
to produce matrix proteins, including type IV collagen,
laminin, and fibronectin. Thus, transforming growth factor/3 plays a central role in glomerular sclerosis from LCDD.

AL-Amyloidosis

Amyloid represents a family of proteins. Fifteen different

types of amyloid have been identified and are named according to the precursor protein that is involved in production of the amyloid. AL-amyloidosis, which is also known
as "primary amyloidosis," represents a plasma cell dyscrasia that is usually characterized by systemic deposition
of amyloid and a mild increase in monoclonal plasma cells
in the bone marrow. However, about 20% of patients with
AL-amyloidosis have overt multiple myeloma or other
lymphoproliferative disorder. In AL-amyloidosis, the amy loid deposits are composed of immunoglobulin light chains.
AL-amyloidosis is a systemic disease characterized by
extracellular deposition of AL-amyloid in a variety of organs (Table 5). Cardiac infiltration frequently produces
congestive heart failure and is a common presenting manifestation of primary amyloidosis. Infiltration of the lungs
and gastrointestinal tract is also common, but often produces few clinical manifestations. Dysesthesias, orthostatic
hypotension, diarrhea, and bladder dysfunction from peripheral and autonomic neuropathies can also occur. Amy loid deposition can also produce an arthropathy that resembles rheumatoid arthritis, a bleeding diathesis, and a variety
of skin manifestations that include purpura. Kidney involvement is common in primary amyloidosis.
Pathology

Glomerular lesions are the dominant renal features of

AL-amyloidosis, and are characterized by the presence of
mesangial nodules and glomerular sclerosis (Fig. 3). In
the early stage, amyloid deposits are usually found in the
mesangium and are not associated with an increase in mes angial cellularity. Deposits may also be seen along the
subepithelial space of capillary loops in more advanced

225

30. Dysproteinemias and Amyloidosis

TABLE 5

Relative Frequency of Organ Infiltration by Light Chains in LCDD and AL-Amyloidosis

Isotype
LCDD
AL-amyloid

K> A
> K

Renal
++++A
+++

Cardiac

Organ involvement
Liver
Neurologic

+++
+

Note: +, uncommon but occurs during the course of the disease;

stages. Imrnunohistochemistry demonstrates that the deposits consist of light chains. On electron microscopy, the
deposits are characteristic randomly oriented, nonbranching fibrils 7 to 10 nm in diameter. In some cases of
early amyloidosis, glomeruli may appear normal on light
microscopy. Careful examination, however, will identify
scattered monotypic light chains on immunofluorescence
microscopy. Ultrastructiiral examination with immunoelectron microscopy to reveal the fibrils of AL-amyloid may be
required to establish the diagnosis. As the disease advances,
mesangial deposits progressively enlarge to form nodules
that compress the filtering surfaces of the glomeruli and
cause renal failure. Epithelial proliferation and crescent
formation are rare in AL-amyloidosis.
There are significant differences between amyloidosis
and LCDD. For amylo id deposition to occur, amyloid P
glycoprotein must also be present. The amyloid P component is not part of the amyloid fibrils, but binds them. This
glycoprotein is a constituent of normal human glomerular
basement membrane and elastic fibrils. In contrast to ALamyloid, in LCDD the light-chain deposits are punctate,
granular, and electron-dense and are identified in the mesangium and/or subendothelial space, Amyloid P component is absent. Amyloid has characteristic tinctorial
properties and stains with Congo red, which produces an
apple-green birefringence when the tissue section is examined under polarized light, and with thioflavin T and S.
These special stains are not taken up by the granular light-

FIGURE 3 Glomerulus from a patient with AL amyloidosis showing segmentally variable accumulation of amorphous acidophilic
material that is effacing portions of the glomerular architecture
(hematoxylin-eosin stain).

+++

H+

GI
rare

Pulmonary
rare

H, extremely common during the course of the disease.

chain deposits of LCDD. Another difference between these

lesions is the tendency for K light chains to compose the
granular deposits of LCDD, while usually A light chains
constitute AL-amyloid. Both diseases can involve organs
other than the kidney (Table 5). Amyloidosis more commonly involves the gastrointestinal tract and lungs, whereas
deposits of LCDD infiltrate the liver more frequently
than amyloidosis.
Clinical Features
Proteinuria and renal insufficiency are the two main renal
manifestations of AL-amyloidosis. Proteinuria ranges from
asymptomatic nonnephrotic proteinuria to nephrotic syndrome. More than 90% of patient have monoclonal light
chains in the urine or blood, and renal insufficiency is pres ent in 58 to 70% at the time of diagnosis. Isolated microscopic hematuria is not common in AL-amyloidosis.
Biopsies of the kidney or other affected organ with ultrastructural studies and immunohistochemistry are important
tools to establish the diagnosis. Scintigraphy using 123Ilabeled P component, which binds to amyloid, can often
diagnose the degree of organ involvement from amyloid infiltration.
Pathogenesis
The pathogenesis of AL-amyloidosis is not completely
understood. Recently, internalization and processing of
light chains by mesangial cells has been shown to cause
amyloid formation. The finding that N-terminal sequences
of light-chain fragments in amyloid are identical to the
sequence of soluble light chains suggests that proteolytic
cleavage of light chains may play a role in causing amyloid.
Presumably, intracellular oxidation or proteolysis of light
chains allows formation of amyloid, which is then extruded
into the extracellular space. With continued production of
amyloid, the mesangium expands, compressing the filtering
surface of the glomeruli and producing progressive renal
failure.
Treatment and Prognosis
A recent randomized trial compared colchicine therapy
alone (0.6 mg twice daily) to colchicine plus melphalan
and prednisone in doses used for multiple myeloma. A 1year course of intermittent melphalan and prednisone
along with colchicine increased survival from 6.7 to 12.2
months, thus supporting the use of chemotherapy in ALamyloidosis not associated with multiple myeloma.
Colchicine should probably be continued indefinitely but

226

Paul W. Sanders

the total dose of melphalan should be limited 600 mg because melphalan promotes development of myelodysplasia
or leukemia. In another retrospective study, patients with
nephrotic syndrome, a normal serum creatinine concentration, and no echocardiographic evidence of cardiac amy loidosis had the best response rate (39%) to melphalan and
prednisone. Response in this subset required 11.7 months
of treatment, but the median survival was 89.4 months,
with 78% surviving 5 years. Eleven of seventeen patients
who responded to treatment had complete resolution of
nephrotic syndrome and 6 others had a 50% reduction in
urinary protein excretion; only 3 of the 17 had persistent
nephrotic-range proteinuria. Toxicity was significant in this
series; acute leukemia or myelodysplasia developed in 7 of
the responders. Survival for patients with AL-amyloidosis
averages 12.2 months with chemotherapy. At present, there
are no controlled trials showing that patients who do not
respond to melphalan and prednisone will respond to more
aggressive chemotherapeutic regimens.
Fibrillary Glomerulonephritis and
Immunotactoid Glomerulopathy
Fibrillary glomerulonephritis is a rare disorder characterized ultrastructurally by the presence of amyloid-like, randomly arranged, fibrillary deposits in the capillary wall.
Unlike amyloid, these fibrils are thicker (18 to 22 nm) and
Congo red and thioflavin T stains are negative (Fig. 4).
Most patients with fibrillary glomerulonephritis do not
have a plasma cell dyscrasia; however, occasionally a
plasma cell dyscrasia is present, so screening is advisable.
Patients typically manifest nephrotic syndrome and varying
degrees of renal failure; progression to end stage renal
failure is the rule. No standard treatment for the idiopathic
variety is currently available. Immunotactoid, or microtubular, glomerulopathy is usually associated with a plasma

cell dyscrasia. The deposits in this lesion contain thick

(>30 nm), organized, microtubular structures that are located in the mesangium and along capillary walls. Cryoglobulinemia, (see Chapter 19), should be considered in the
differential diagnosis and ruled out clinically. Treatment
of the underlying plasma cell dyscrasia is indicated for this
rare disorder.
WALDENSTROM'S MACROGLOBULINEMIA

This disorder constitutes about 5% of monoclonal gammopathies and is characterized by the presence of a monoclonal population of lymphocytoid plasma cells. This condition clinically behaves more like lymphoma, although the
malignant cell secretes IgM (macroglobulin). IgM is not
excreted and accumulates in the plasma to produce hyperviscosity syndrome. Lytic bone lesions are uncommon, but
hepatosplenomegaly and lymphadenopathy are frequently
identified. Hyperviscosity syndrome produces neurologic
symptoms, visual impairment, bleeding diathesis, renal failure, and symptoms of hypervolemia. Renal failure is usually
mild but occurs in about 30% of patients. Hyperviscosity
syndrome and precipitation of IgM in the glomerular capillaries are the most common causes of renal failure. About
10 to 15% of patients develop AL-amyloidosis, but cast
nephropathy is rare.
Because of the typically advanced age at presentation
(sixth to seventh decade) and slowly progressive course,
the major therapeutic goal is relief of symptoms. Treatment is generally plasmapheresis for hyperviscosity syndrome, followed by alkylating agents alone. All patients
with monoclonal IgM levels >3 g/dL should have a
serum viscosity check. Plasmapheresis is indicated in
symptomatic patients and should be continued until symp toms resolve and serum viscosity normalizes. Initial chemotherapy is usually chlorambucil, which is adjusted to
control serum IgM concentration and organomegaly, without inducing cytopenias. Severe renal failure requiring renal replacement therapy is uncommon. Median survival is
about 3 years and is related to the advanced age at onset
of this disorder.
Bibliography

FIGURE 4 Electron micrograph of a glomerulus from a patient

with fibrillary glomerulonephritis. Note the randomly arranged
relatively straight fibrils with a diameter of approximately 20 nm.
The overall ultrastructural appearance resembles amyloid except
that the fibrils are approximately twice as thick.

Alexanian R, Dimopoulos, m: The treatment of multiple myeloma.

N Engl J Med 330:484-489, 1994.
Buxbaum JN, J Chuba JV, Hellman GC, et al.: Monoclonal immunoglobulin deposition disease: Light chain and light and heavy
chain deposition diseases and their relation to light chain amy loidosis. Ann Intern Med 112:455-464, 1990.
Ganeval D, Noel L-H, Preud'Homme J-b, et al.: Light-chain deposition disease: Its relation with AL-type amyloidosis. Kidney Int
26:1 -9, 1984.
Ganeval D, Rabian C, Guerin V, et al.: Treatment of multiple
myeloma with renal involvement. Adv Nephrol 21:347-370
1992.
Gertz MA, Kyle RA, Greipp PR: Response rates and survival in
primary systemic amyloidosis. Blood 77:257-262, 1991.
Heilman RL, Velosa JA, Holley KE, et al.: Long-term followup and response to chemotherapy in patients with light-chain
deposition disease. Am J Kidney Dis 20:34-41, 1992.

31. Renal and Urologic Complications of Cancer

227

Ivanyi B: Frequency of light chain deposition nephropathy relative

to renal amyloidosis and Bence Jones cast nephropathy in a
necropsy study of patients with myeloma. Arch Pathol Lab Med
114:986-987, 1990.
Jones HB: Papers on chemical pathology: Prefaced by the Gulstonian Lectures, read at the Royal College of Physicians, 1846.
Lancet 2:88-92, 1847.
Preud'homme J-L, Aucouturier P, Touchard G, et al: Monoclonal
immunoglobulin deposition disease (Randall type). Relationship with structural abnormalities of immunoglobulin chains.
Kidney Int 46:965-972, 1994.
Sanders PW, Herrera G A, Kirk KA, et al.: Spectrum of glomerular
and tubulointerstitial renal lesions associated with monotypical
immunoglobulin light chain deposition. Lab Invest 64:527537, 1991.

Sanders PW, Herrera GA: Monoclonal immunoglobulin light

chain-related renal diseases. Semin Nephrol 13:324-341, 1993.
Skinner M, Anderson JJ, Simms R, et al.: Treatment of 100 patients
with primary amyloidosis: A randomized tria l of melphalan,
prednisone, and colchicine versus colchicine only. Am J Med
100:290-298, 1996. Solomon A, Weiss DT, Kattine AA:.
Nephrotoxic potential of
Bence Jones proteins. N EnglJ Med 324:1845-1851, 1991.
Tagouri YM, Sanders PW, Pickens MM, et al.: In vitro AL-amyloid
formation by rat and human mesangial cells. Lab Invest 74:290302, 1996.
Zhu L, Herrera GA, Murphy -Ullrich JE, et al.: Pathogenesis
of glomerulosclerosis in light chain deposition disease: Role
for transforming growth factor-/3. Am J Path ol 147:375-385,
1995.

RENAL AND UROLOGIC COMPLICATIONS OF

CANCER AND ITS TREATMENT
MARC B. GARNICK

Nephrourologic complications of cancer constitute a

major source of morbidity and oftentimes mortality in
patients with neoplastic disorders. These complications
may take many forms. Cancer may be associated with
glomerulonephritis, often of the immune complex variety.
Urinary tract obstruction may cause significant functional
and structural changes within the kidney, pelvis, and
ureter, leading to acute or chronic renal failure as well
as predisposing to urinary tract infection. Malignant cells
may arise in or invade the renal parenchyma, leading to
renal dysfunction. Cancer therapy itself may also cause
renal and urothelial damage. Finally, abnormalities of
urinary tract function can occur secondary to spontaneous
or chemotherapy-induced tumor cell lysis in which the
kidney is overwhelmed by breakdown products of rapid
cellular dissolution.
GLOMERULAR DISEASE

Secondary nephrotic syndrome has been reported with

adenocarcinomas of the gastrointestinal tract, including
colon, rectum, stomach, and pancreas, as well as with

Primer on Kidney Diseases, Second Edition

lung, breast, ovary, skin, prostate, renal cell, and other

tumors. Secondary membranous nephropathy is the histologic lesion in two thirds of patients with nephrotic
syndrome and carcinomas. The nephropathy is immunecomplex-mediated; carcinoembryonic and other antigens
have been eluted from renal tissue of some patients.
Less frequently, proliferative lesions with a nephritic
picture are found.
Hodgkin's disease is associated with nephrotic syndrome, which is usually due to minimal change nephropathy. Other histologic patterns, including amyloidosis, have
been observed. The activity of the minimal change nephropathy often parallels activity of the underlying lymphoma. The nephrotic syndrome responds to therapy of
the lymphoma and a relapse of proteinuria may herald
relapse of the tumor. Non-Hodgkin's lymphomas are as sociated with membranous nephropathy. Minimal change
nephropathy or proliferative glomerulopathies are less frequently found.
With the use of interferons (both a and -y), rare cases
of reversible, immune-mediated glomerulonephritis have
been reported.
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

228

Marc B. Garnick

COMPLICATIONS RESULTING FROM ANATOMIC

ABNORMALITIES RELATED TO THE PRIMARY
DISEASE PROCESS
Urinary Tract Obstruction

Retroperitoneal tumors causing ureteral obstruction can

occur with any solid neoplasm. Urologic and gynecologic
cancers, together with lymphomas, are the most common
causes of ureteral obstruction. Anatomically, the retroperitoneal lymph nodes which are often the drainage site for
urologic or gynecological cancers or the sites of lymphomatous infiltration are along paraaortic, paracaval, and periureteral locations in the retroperitoneal space. Lymphadenopathy of mild to moderate degree which is strategically
placed along the course of the ureter can result in unilateral
hydronephrosis in a relatively short time. Likewise, retroperitoneal lymph node metastases or primary retroperitoneal tumors can attain a large size without ever causing
ureteral obstruction because of their location with respect
to the course of the ureter. Prostate cancer causes distal
urinary tract obstruction due to obstruction at the bladder
neck. This is also the case for advanced cervical cancer
which, in addition to causing obstruction of the bladder
neck, may infiltrate the retroperitoneal surfaces and obstruct the ureters. Testicular cancer, especially seminoma,
can invade the retroperitoneal space and result in large
bulky tumors capable of obstructing both ureters simultaneously. In addition, metastatic neoplasms of many sources
can invade the retroperitoneal areolar tissues forming a
plaque-like sheet of tumor, sometimes thin, that can involve
and obstruct the ureters. Adenopathy may not be evident
on imaging studies, making diagnosis difficult. Metastatic
breast cancer, for example, may result in a desmoplastic
reaction which may mimic primary retroperitoneal fibrosis.
Regardless of what the cause of urinary obstruction is,
prompt diagnosis and treatment are imperative, since the
ability to deliver potentially nephrotoxic drugs that are
curative, such as cisplatin, will be compromised early on if
there is irreparable renal damage resulting from obstruction.
The management of obstructive uropathy due to cancer
requires consideration of the underlying malignancy and
its location, prognosis, and expected response to therapy.
Obstructive uropathy is discussed in detail in Chapter 54.
Renal Parenchyma! Infiltration

Renal failure resulting from bilateral replacement of the

kidneys by tumor has been reported in leukemia and
lymphoma. While 30 to 50% of autopsied patients with
leukemia may show infiltration of the kidneys by leukemic
cells, actual compromise of renal function is rare. The leukemic infiltrate is usually diffuse, but may be nodular or
focal and may result in as much as a ten-fold increase
in kidney size. Radiographically, renal enlargement and
elongation of the calyces are common. Among the leukemias, the form most commonly associated with acute renal
failure is acute lymphocytic leukemia. Radiation to the
kidneys in doses of 600 to 1000 rads accompanied by cyto-

toxic chemotherapy is the treatment of choice for this condition.

Renal parenchymal infiltration by Hodgkin's and nonHodgkin's lymphomas has been reported in one third of
autopsied cases. Clinically, the diagnosis is suggested by
flank masses or enlarged renal outlines on radiograms, but
uremia from lymphomatous renal infiltration can also occur
in normal-sized kidneys. Other clues incude mild, nonnephrotic proteinuria or hypertension, the latter presumably
due to distention of intrarenal arteries. Pathologically, multiple discrete kidney nodules appear more frequently than
diffuse enlargement. Despite the high prevalence of lymphomatous renal infiltration, however, death directly attributable to acute renal failure secondary to such tumor invasion
is uncommon. In the absence of a known diagnosis of
lymphoma, acute renal failure of unknown etiology merits
an extensive evaluation which often includes a kidney biopsy. In the setting of widespread lymphoma and enlarged
kidneys, however, an invasive diagnostic procedure is usually not indicated. When lymphoma has been shown to be
the cause of acute renal failure, radiotherapy and chemo therapy in conjunction with dialysis have transiently improved renal function.
Myloma and other plasma cell dyscrasias may also cause
renal dysfunction related to structural abnormalities of the
kidney. These findings are discussed in Chapter 30.
COMPLICATIONS SECONDARY TO CANCER THERAPY
Genitourinary Hemorrhage

One of the most dramatic clinical events in cancer management is the development of uncontrollable urinary tract
hemorrhage. Although in the past this was often due to
bleeding from the bladder secondary to radiation therapy
and/or cyclophosphamide or ifosphamide treatment, it is
fortunately an uncommon event. Now, the use of Mesna
with ifosphamide has greatly reduced the urothelial damage
from this alkylating agent. Mesna is able to inactivate the
toxic metabolites of ifosphamide on uroepithelial surfaces.
Bleeding diathesis superimposed upon urothelial damage
from toxic drugs or their metabolites in the urine is most
often the etiology of urinary tract hemorrhage. Additional
agents implicated in urinary tract hemorrhage, however,
include L-asparaginase, actinocyceri-D, mitomycin-C,
mithramycin, and 6-mercaptopurine. Patients with cancer
often receivie anticoagulants and may receive antibiotics
with anticoagulant effects. Together these may place the
urinary tract at increased risk for bleeding.
The management of hemorrhagic cystitis requires the
combined efforts of urologist and oncologist alike. Catheter
drainage and continuous irrigation, intravesical cautery,
and intravesical therapy with formalin have all been used
with reported success. For refractory bleeding, surgical approaches including cystotomy with installation of phenol
and ligation of the bladder vessels have been required.
Urinary diversion and removal of the organ to stop lifethreatening bleeding have on occasion been necessary.

31. Renal and Urologic Complications of Cancer

229

Drug- Specific Effects

Cyclophosphamide

Cyclophosphamide has a wide spectrum of urological

effects, including hemorrhagic cystitis (see above). The
most commonly described defect of renal function is impaired water excretion, most commonly seen with the high
doses (usually greater than 50 mg/kg) used for bone marrow
transplantation. Laboratory manifestations of cyclophosphamide-induced impaired water excretion include hyponatremia, decreased serum osmolality, and elevated urinary
sodium, but normal serum vasopressin levels. These abnormalities probably result from a direct toxic effect of cyclophosphamide metabolites on distal tubules and collecting
ducts. Hypotonic solutions and solutions that induce hyponatremia should be avoided when effecting the diuresis
needed to prevent hemorrhagic cystitis.

Table 1 lists potentially nephrotoxic agents.

Cisplatin
With its wide spectrum of activities, cisplatin is one of
the most commonly used anticancer agents. It is primarily
excreted by the kidney and accumulates in renal tubular
cells selectively. Platinum resorption in the tubules leads
to enzyme inhibition, alkylation of specific molecules, and
generalized cellular injury that is worse in the distal nephron. Pathologic changes in the kidney include interstitial
edema and tubular dilatation with relative glomerular sparing. Cisplatin damage may lead to acute renal failure, but,
more commonly, a gradual and irreversible decline in renal
function occurs with cumulative dosing. The presence of
chloride favors the persistence of cisplatin rather than the
formation of the active chemotherapeutic species, aquated
dihydroxydiamino platinum. Therefore, vigorous hydration with saline should be employed when cisplatin is administered, with the goal of maintaining a urine output in
excess of 200 mL/h before and for at least 18 hours after
the dose. Diuretics may be employed. Most nephrotoxicity
is traceable to underhydration and inattentive supervision
of urinary volumes.

Streptozotocin

Renal damage is common following Streptozotocin therapy and may present as tubular dysfunction with phosphate
wasting or a complete Fanconi syndrome with aminoaciduria, proteinuria, or renal insufficiency. Careful and diligent
monitoring of renal function must occur following adminis tration.

TABLE I

Potentially Nephrotoxic Chemotherapeutic Agents

Risk
Drug

High

Int

Type
Low

Time Course

Acute

Chronic

Specific tubular damage

limned

Delayed

X
X
X
X
X
X

Alkylating agents
Cisplatin

Cyclophosphamide
Streptozotocin"

X
X

Semustine (methyl CCNU)

X
X

Carmustine (BCNU)

Lomustine (CCNU)

Antimetabolites

Methot rexate6
Cytocine arabinoside (Ara-C)

X
X
X
X
X

5-Fluorouracil (5-FU)'

5-Azacytidine
6-Thioguanine
Antitumor antibiotics
Mitomycin''
Mithramycin f
Adriamycin

X
X
X
X
X

X
X

X
X
X
X
X

X
X

X
X
X

Biologic agents
Interferon-a
Interferon-y

Corynebacterium parvum
Interleukin-2

X
X

X
X

X
X
X
X

From Weber B, Garnick MB, Rieselbach RE: Nephropathies due to antineoplastic agents. In: Massry SG, Glassock RJ (eds) Textbook of
Nephrology. 2nd ed., Vol 1. Williams & Wilkins, Baltimore, pp. 818-822, 1989. With permission. " Fanconi's syndrome as the most
severe manifestation. b Only seen with intermediate to high dose regimens. c Only seen where given in combination with mitomycin C.
Hemolytic- uremic syndrome as the most severe manifestation. e Frequent
with antineoplastic doses, rare in doses used for hypercalcemia.

Marc B. Garnick

230
Semustine (CCNU)

Nitrosoureas are excreted primarily in the urine. Doserelated renal dysfunction occurs with cumulative semustine
doses, usually after 1500 mg/m2 has been administered. The
onset of renal dysfunction may also be delayed for years
following completion of therapy. Pathologic examination
demonstrates glomerular and interstitial fibrosis.
Methotrexate

Methotrexate competitively binds to dihydrofolate reductase, the enzyme responsible for converting folic acid
to reduced folate cofactors. Drug-induced nephrotoxicity
can be a prominent feature of higher dose methotrexate
therapy. Although the exact mechanism of damage is unclear, three explanatory hypotheses have been suggested.
The first relates to methotrexate or methotrexate metabolites precipitating in the renal tubules causing an obstructive nephropathy. Methotrexate also may have a direct
effect on renal tubular cells altering regeneration of epithelial cells and ion secretory channels and other metabolic
processes with a secondary feedback decrease in GFR.
A third possibility relates to a direct effect on glomular
perfusion causing a direct decrease in glomular filtration
rate.
Toxicity is dose-related. The urinary solubility of methotrexate and its metabolites is increased at higher urine pH.
Thus hydration and urinary alkalinization have been shown
to be useful in minimizing toxicity. Rescue agents such as
leucovorin, thymidine, carboxypeptidase, and dihydrofolate reductase that correct or bypass the effect of methotrexate have also been employed.
Mitomycin C

Mitomycin C, an antitumor antibiotic isolated from

Streptomyces caespitosis, is primarily used in the treatment
of gastrointestinal and breast cancers. Nephrotoxicity is
dose-related and cumulative. Clinically, a rising blood urea
nitrogen and creatinine and proteinuria are the most common features.
The hemolytic uremic syndrome may be a severe consequence of mitomycin C therapy, usually after 60 mg of the
drug has been administered. Characterized by endothelial
damage leading to micro-angiopathic hemolytic anemia
(fragmented red blood cells on smear), irreversible uremia,
and thrombocytopenia, this syndrome may also be associated with hypertension, neuropathies, and pulmonary
edema. Although most commonly associated with mitomy cin C, other chemotherapies have been implicated in the
development of this syndrome. Once recognized, blood
products should be avoided, as this may exacerbate
symptoms.
Mithramycin

Mithramycin is rarely used as an antineoplastic agent; it

is more frequently employed to treat cancer-related hypercalcemia. The doses required for hypercalcemia, 25 /ug/kg,
are much lower than the antitumor dose and rarely result
in renal dysfunction. However, mithramycin may be nephrotoxic in up to 40% of patients receiving antitumor doses

of 125 to 250 yu,g/kg. Usually mithramycin nephrotoxicity is

characterized by proteinuria and a diminution in creatinine
clearance, but hemolytic uremic syndrome may also occur.
Biological Response Modifiers

Biological response modifiers are receiving widespread

use in the treatment of many disorders including neoplasia,
chemotherapy-induced cytopenias, and viral hepatitis.
Interferon-cc interferon y, corynebacterium parvum, interleukin-2, and the newly introduced colony stimulating factors, G-CSF and GM -CSF, have all occasionally been associated with nephrotoxicity. The most prominent of these
is interleukin-2, administered with or without associated
LAK cell therapy. Early studies with IL-2 reported a nearly
50% incidence of oliguria and azotemia. In one study of
adoptive immunotherapy with IL-2,10 of 20 patients developed creatinine elevations between 2.1 and 10 mg/dL during the course of therapy, although these elevations were
usually rapidly reversible on cessation of therapy. Renal
physiology studies with IL-2 administration have demonstrated universal diminution in GFR and reduction in
FENa, similar to what is seen in endotoxic shock. With
more recent alterations in dosing schedules, IL-2 nephrotoxicity has been diminished, although not eliminated. In
one study of weekly "low-dose" IL-2, reversible nephrotoxicity occurred in 2 of 33 patients.
Renal Complications of Bone Marrow Transplantation, HighDose Chemotherapy, and Total Body Irradiation

Over the past decade, the use of high-dose chemotherapy

(HDC; with or without total body irradiation), accompanied by bone marrow transplantation and peripheral blood
stem cell progenitor support, has assumed an important
role in oncologic practice. Although commonly used in
selected patients with lymphoma, breast cancer, and certain
leukemias, this treatment modality continues to be investigated in other hematologic neoplasms, such as myeloma
and solid tumors such as melanoma and testis cancer. Many
of the chemotherapeutic agents listed in Table 1 have intrinsic nephrotoxicity when used in conventional doses.
When super high doses are administered, a broadened
range of renal derangements may occur, especially when
concomitant total body irradiation (TBI) is administered
as part of the preparative regimen prior to transplantation.
The additional need for potentially nephrotoxic antibiotics,
antifungals, and immunosuppressants may increase the incidence of nephrotoxicity.
In one study of 84 patients treated with HDC and TBI
for hematologic neoplasms, the presence of significant renal
dysfunction was correlated with the dose of TBI given and
the presence of graft vs host disease (GVHD). Ninetythree percent (93%) of patients receiving lower doses of
TBI without GVHD were free of renal dysfunction at 18
months posttransplantation compared to 52% of patients
who received higher radiation doses and GVHD. Renal
shielding should be recommended and renal doses carefully
monitored in these patients. Another study with longer

31. Renal and Urologic Complications of Cancer

231

follow-up reported frequent renal dysfunction in recipients

of transplantation with accompanying TBI.
The concomitant use of nonchemotherapeutic nephrotoxins, such as cyclosporine, amphotericin, and aminoglycoside antibiotics, must be scrupulously monitored to avoid
synergistic adverse effects. The incidence of renal toxicity
may be diminished by monitoring blood levels of these
agents, as well as performing detailed pharmacokinetic profiling of the individual chemotherapeutic agents. In one
study, the development (and subsequent avoidance) of potentially fatal nephrotoxicity was predicted early in the
course of therapy by using real-time pharmacokinetic
guided dosing.

nephropathy may be insidious occurring as late as 10 years

following the radiation therapy treatment. Patients may
present with asymptomatic renal failure and a clinical syndrome suggestive of chronic glomerulonephritis or tubulointerstitial disease. Today, the most effective treatment
for radiation nephropathy is prevention through careful
localization of the kidneys during treatment field planning
and limitation of dosage and volume. It appears that the
delivery of 200 cGy in 10 fractions in a 10-week period is a
reasonably safe treatment plan for large abdominal tumors
that are going to encompass both kidneys. There is also
some potential for synergistic radiation nephropathy to
occur in association with nephrotoxic agents such as dactinomycin.

Metabolic Complications of Tumor Lysis

Bibliography
Abelson HT. Garnick MB: Renal failure induced by cancer chemotherapy. In: Rieselbach RE, Garnick MB (eds) Cancer and
the Kidney. Lea & Febiger, Philadelphia, pp 769-813, 1982.
Alpers CE, Cotran RS: Neoplasia and glomerular injury. Kidney
Int 30:465-473, 1986
Averbuch SD, Austin H, Sherwin S, et al.: Acute interstitial nephritis with the nephrotic syndrome following recombinant leukocyte A interferon therapy for mycosis fungoides. A' Engl J Med
310:32-35, 1984
Cadman E, Lundberg W, Bertino J: Hyperphosphatemia and hypo calcemia accompanying rapid cell lysis in a patient with Burkitt's
lymphoma and Burkitt cell leukemia. Am J Med 62:283, 1977.
DeFronzo RA, Colvin OM, Braine H, et al.: Cyclophosphamide
and the kidney. Cancer 33:483-491, 1973
Garnick MB: Urologic complications of cancer and its treatment.
In: Holland JF, Frei III E (eds) Cancer Medicine, 3rd ed. Lea &
Febiger, Philadelphia, pp. 2323 -2331, 1993.
Garnick MG, Mayer RJ: Acute renal failure associated with neoplastic disease and its treatment. Semin Oncol 5:155-165, 1978.
Garnick MB, Mayer RJ, Abelson HT: Acute renal failure associated with cancer treatment. In: Brenner BM, Lazarus JM (eds)
Acute Renal Failure, 2nd ed. Churchill-Livingstone, New York,
pp. 621-657, 1988.
Harmon WE, Cohen HT, Schneeberger E, et al.: Chronic renal
failure in children treated with methyl CCNU. N Engl J Med
300:1200-1203, 1979
Leblond V, Sutton L, Jacquiaud C, et al.: Evaluation of renal
function in 60 long-term survivors of bone marrow transplantation. J Am Soc Nephrol 6:1661-1665, 1995.
Miralbell R, Bieri S, Mermillod B, et al.: Renal toxicity after
allogeneic bone marrow transplantation: The combined effects
of total-body irradiation and graft -versus-host disease. J Clin
Oncol 14:579-585, 1996.
Morel-Maroger Striker L, Striker GE: Glomerular lesions in malignancies. Contrib Nephrol 48:111-124, 1985
Price TM, Murgo AJ, Keveney IJ, et al.: Renal failure and hemolytic anemia associated with mitomycin C. Cancer 55:51-56,
1989.
Rieselbach RE, Garnick MB: Renal diseases induced by antineoplastic agents. In: Schrier RW, Gottschalk CW (eds) Diseases
of the Kidney, 5th ed., Vol 2. Little Brown. Boston, pp. 11651186, 1993.
Wright JE, Elias A, Tretyakov O, et al.: High -dose ifosphamide,
caraboplatin and etoposide pharmacokinetics: Correlation of
plasma drug levels with renal toxicity. Cancer Chemother Pharmacol 36:345-351, 1995.

Acute uric acid nephropathy may occur in a cancer patient when there is a rapid turnover of cellular contents,
usually secondary to lysis of tumors sensitive to chemotherapy or radiation. Plasma levels of urate become elevated
due to nucleic acid breakdown. In response to hyperuricemia, the kidney increases the tubular secretion of urate.
Uric acid solubility decreases at low pH. Because of the
acidification mechanisms within the distal tubule, uric acid
precipitation and tubular obstruction may occur.
Uric acid nephropathy most often occurs in association
with chemotherapy responsive tumors such as leukemias
and lymphomas, especially Burkitt's lymphoma. It is important to anticipate the development of this complication
prior to the initiation of therapy. Vigorous hydration, pro phylactic use of allopurinol both before and during antineoplastic therapy, and meticulous attention to serum electrolytes, especially potassium, calcium, and phosphate, are
crucial in the overall management of patients undergoing
cytoreductive therapy. Urate nephropathy and the tumor
lysis syndrome are discussed in detail in Chapter 25.
Radiation Nephropathy

Although uncommon today, radiation damage to the

kidney had been a major complication of curative therapy
of large abdominal and retroperitoneal neoplasms in which
both kidneys were included in the radiation therapy port.
Physiologically, alterations in glomerular filtration rate, renal plasma flow, and tubular excretory capacity occur following as little as 400 cGy delivered to the bulk of both
kidneys. Urinalysis findings with radiation nephropathy include microscopic hematuria, albuminuria, and hyaline and
granular casts. Pathologically the kidneys are scarred and
shrunken with evidence of arteriolar fibrinoid necrosis, tubular atrophy and glomerulosclerosis. These changes usually occur over 6 to 12 months following the delivery of
greater than 2300 cGy to both kidneys during a 4- to 5week period. Acute radiation nephropathy presents with
renal dysfunction and signs and symptoms referable to
cardiovascular compromise secondary to hypertension,
which may be accelerated or malignant. Chronic radiation

HEMOLYTIC UREMIC
SYNDROME/THROMBOTIC
THROMBOCYTOPENIC PURPURA
RICHARD L SIEGLER

Hemolytic uremic syndrome (HUS) and thrombotic

thrombocytopenic purpura (TTP) are both characterized
by a thrombotic microangiopathy (TMA) in target organs
(Fig. 1). However, the epidemiology, clinical expression,
response to therapy, and natural history of classic postdiarrheal (D+) HUS and TTP are sufficiently distinct to warrant
viewing them as different, but similar, disorders. It must be
acknowledged, however, that some cases of nondiarrheal
(atypical) HUS, especially those cases seen in adults, are
indistinguishable from TTP. This has led to the concept that
HUS and TTP should be viewed as variable expressions of
a single disorder. TTP was first described by Moschowitz
in 1925 when he reported a 16-year-old female who succumbed to anemia and multiorgan TMA. Although Gasser
is credited with first describing childhood HUS 30 years
later, we owe much of our understanding of the clinical
features and natural history of postdiarrheal (D+) HUS to
the observations of Gianantonio in the 1960s and 1970s.
A major contribution was also made by Karmali in 1983
when he recognized the syndrome's association with cytotoxin producing enterohemorrhagic Escherichia coli (e.g.,
E. coli O157:H7).

ous outbreaks. Cattle intestines are natural reservoirs for

the bacteria. Therefore, contamination of beef with cattle
feces during the slaughtering process sets the stage for
HUS. Hamburger is the single most commonly recognized
vector because surface contamination of sides of beef used
to make hamburger results in the mixing of pathogens
throughout the beef patty. Thereafter, only thorough cooking kills the toxin-producing organisms. Classic postdiarrheal (D+) HUS occurs predominately during the warmer
months of the year; there is no clear-cut gender predilection. The median age for children is 2 years (mean age 4
years), but it also occurs in adolescents and adults. The
incidence varies between different regions of the United
States, but the overall annual incidence is probably about
1 to 1.5 cases for every 100,000 children less than 18 years
of age. The incidence rises dramatically in infants aged 1
to 2 years (e.g., 7 cases/100,000). There are few incidence
data available for D+ HUS in adults. A study of E. coli
O157:H7 associated HUS in Washington State, however,
demonstrated an annual incidence of approximately 1 case/
100,000 population for those ages 20 to 49 years, 0.5 case/
100,000 population for those 50 to 59 years of age, and
close to 2 cases/100,000 population for those 60 years of
age and older. Washington experiences more cases than
most regions of the country, so the overall national incidence in adults is probably much lower.
Ten percent of childhood HUS and more than one-hajf
of adult cases are not preceded by diarrhea. Nondiarrheal
(D~) HUS, also known as atypical HUS, can be secondary or idiopathic. As with TTP, it can occur in association
with nonenteric infections, during pregnancy and the postpartum period, or with primary glomerulonephritis, malignant hypertension, drugs, bone marrow transplant, cancer, and collagen vascular dis orders such as systemic
lupus erythematosus (SLE) and scleroderma. The most
common nonenteric infection associated with HUS is
pneumococcal pneumonia or meningitis, but HIV is also
emerging as a causative agent. Preeclampsia can progress
to HUS, which can also occur as late as 3 months following a normal delivery. Examples of drug-related causes in-

HEMOLYTIC UREMIC SYNDROME

Epidemiology

Ninety percent of childhood HUS, but less than 50% of

adult cases, are preceded by diarrhea (D+) that is usually
bloody (Table 1). There is persuasive evidence that the
diarrhea is caused by enterohemorrhagic E. coli (EHEC)
that produce potent cytotoxins known as Shiga-like toxins
(SLT) or verotoxins. About 90% of EHEC cases in the
United States are of the O157:H7 serotype, but dozens of
other toxin -producing EHEC serotypes exist.
Vectors for EHEC include meat, especially beef hamburger and fermented salami. Other vectors include contaminated water, fruits, and vegetables, as well as unpasteurized apple juice, apple cider, and dairy products.
Person-to-person spread has been responsible for numer-

Primer on Kidney Diseases, Second Edition

232

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

32. HUS/TTP

233
Pathogenic Cascades

FIGURE I A glomerulus from a child with postdiarrheal HUS,

showing microthrombi of glomerular capillaries.

elude oral contraceptives, mitomycin, and cyclosporine.

Approximately 10% of patients with systemic sclerosis
(scleroderma) experience rapid onset of acute renal failure with clinical and histological features of HUS. Most
also have malignant hypertension. Very severe hypertension, without associated scleroderma can, on occasion,
also progress to secondary HUS. Although secondary HUS
accounts for the majority of D~ cases in adulthood, it is
rare in children. The bulk of the D~ cases in children, and
a minority in adults, are idiopathic. Some of these cases
are familial.

The pathogenic cascade for classic postdiarrheal HUS is

becoming better understood. Ingested EHEC colonize the
distal ileum and large intestines via bacterial fimbrial attachment and cause a severe (hemorrhagic) colitis. This
severe inflammatory response facilitates the entry of Shigalike toxins, and probably lipopolysaccharide, into the circulation. Shiga-like toxins are 71-kDa protein subunit toxins
comproised of five B subunits and one A subunit; they
exist in several forms (i.e., SLT-1, SLT-2, SLT-2 variants).
Following absorption into the circulation, the SLT B subunits rapidly attach to glycolipid (GB3 ) receptors in the
gut, kidneys, and, occasionally, other organs (e.g., brain,
pancreas). The single A subunit is then internalized and
inhibits protein synthesis, thereby causing cell injury or
death. Even though microvascular endothelial cell injury
is central to the TMA that characterizes HUS/TTP, it is
not clear that the endothelial cell is the initial renal target
for the toxin. There is some evidence that the renal tubular
epithelial cell is initially injured in D+ HUS.
Shiga-like -toxins-mediated endothelial cell injury results in thickened glomerular capillary walls due to endothelial cell swelling, detachment of endothelial cells from
the underlying basement membrane, and accumulation of
fluffy material between the endothelial cells and basement
membrane. The endothelial cell injury also results in the
release of von Willebrand factor (vWF) multimers and
platelet activating factor and activation of platelets, and,
to a lesser extent, the coagulation cascade. Concomitantly,
there are perturbations of multiple additional pathogenic
cascades, including production of cytokines (e.g., tumor
necrosis factor, interleukin-6, interleukin-8), as well as
increased synthesis of thromboxane and endothelin coupled with decreased production of prostacyclin (PGy,
which together impair the renal microcirculation by facilitating both microthrombotic obstruction and renal vasoconstriction.
The pathogenesis for D~ HUS is less well understood.
Secondary HUS is presumably caused by drugs (e.g., oral
contraceptives, mitomycin, cyclosporine) or condition (e.g.,
malignant hypertension, cancer, SLE, scleroderma) that
either damage endothelial cells or favor microthrombus
formation (e.g., anticardiolipin antibodies, reduced PGIa
production). Idiopathic D HUS is sometimes familial (autosomal dominant or recessive). Low serum complement
concentrations are common, and, in some families, there
is deficient production of the antithrombotic prostaglandin PGI 2 .

TABLE 1
Epidemiology

Age (mean)
Gender
Seasonality
Incidence
Antecedents

D + HUS

TTP

4 yr
Equal
Warm months
1-1.5/100,000
(<18 years age)
Diarrhea

35 years
70% women
All year
0.37/100,000
(all ages)
Varied

Prodrome

Classic postdiarrheal HUS is preceded by colitis that

becomes bloody in about 75% of patients. Most children
also experience vomiting. Abdominal pain can be severe
and is sometimes initially misdiagnosed as acute appendicitis or ulcerative colitis. Rectal prolapse, intussusception,
and bowel necrosis with perforation can occur. Children

234

Richard L. Siegler

with nondiarrheal (D ) HUS usually have vomiting without

diarrhea. Low-grade fever occurs in almost half, and upper
respiratory infection in about one-fourth.
Clinical Features

After about 1 week of diarrhea (range 1-14 days) (Table

2), the onset of HUS is heralded by pallor, diminished
urine output, lethargy, and irritability and, in close to 10%
of cases, seizures. Jaundice, petechia, or purpura occur
occasionally. Mild cases, with little or no renal failure, and
examples of incomplete syndrome (i.e., lacking the triad
of hemolytic anemia, thrombocytopenia, and acute renal
failure) have been reported. Most children experience oliguric renal failure that lasts on average about 1 week. Close
to one-half have a period of anuria with a duration of 3
days. Prolonged oliguria occurs occasionally.
Hypertension is present in the majority of cases of D+
HUS. Usually it is mild to moderate and labile, and resolves
by the time of discharge from the hospital.
Seizures and other forms of severe central nervous system involvement used to be seen in close to 50% of cases.
Now, however, as a result of more timely diagnoses, greater
attention to fluid and electrolyte management, and earlier
initiation of dialysis, metabolic encephalopathy (e.g., seizures, coma, semicoma) occurs in only about 20% of cases.
A 3 to 5% incidence of stroke or generalized cerebral
edema is still observed.
Pancreatic involvement is seen in approximately 20% of
cases. It is usually expressed as elevated serum amylase
and lipase enzyme concentrations, with or without pain,
tenderness, or vomiting; pseudocysts or necrosis have not
been reported. Of greater concern is insulin -dependent
diabetes mellit us that has been reported to occur in close
to 10% of cases; on occasion it can be permanent. On
rare occasions the TMA can involve the heart, lungs, eyes,
parotid gland, skin, or muscle.
Although the clinical features of nondiarrheal HUS in
children can be very similar to those of classic D+ disease,
there is a subset of D~ patients who present with the insidi-

TABLE 2
Features
D+HUS

TTP

Severe colitis

Usually

Rarely

Fever
Encephalopathy
Thrombocytopenia
Hemorrhage, multiorgan
Hematuria/proteinuria
Oligoanuria
Renal failure
Multiorgan involvement
Death"
Recurrences

Mild

Prominent
>90%
Severe
Usually

' With optimal therapy.

20%
Moderate
Rarely
100%
Usually
Usually
Occasionally

5%
1%

75%
Occasionally
Occasionally
Usually

15%
20%

ous onset of nonoliguric renal failure and malignant hypertension that often leads to congestive heart failure. The
clinical expression of D~ HUS in adults is generally severe.
Cases associated with preeclampsia are usually accompanied by fetal distress, liver involvement, and, in about onethird of cases, disseminated intravascular coagulation. Postpartum HUS typically includes severe renal failure and
hypertension. Fever and neurological involvement are frequent. Cancer (especially gastric and metastatic carcinoma)
and mitomycin-associated HUS are associated with a very
malignant and rapidly progressive clinical course.
Laboratory features include microangiopathic hemolytic
anemia, thrombocytopenia, and azotemia. The anemia is
largely due to mechanical fragmentation of red blood cells
(RBC) as they pass through partially occluded renal vessels.
It is characterized by fragmented RBCs (e.g., burr cells,
shistocytes) on blood smear and elevated serum lactic dehydrogenase concentrations. A low platelet count can be documented in about 95% of cases, but severe thrombocytopenia occurs infrequently. Prothrombin times and partial
thromboplastin times are usually normal. Increased serum
concentrations of hepatic enzymes, uric acid, triglycerides,
and bilirubin are commonly seen.
Management

Supportive therapy continues to be the cornerstone

in managing children with classic postdiarrheal HUS.
Numerous therapies, including heparin, antiplatelet
agents, fibrinolytic agents, vitamin E, prostacyclin, intravenous IgG, and infusion of fresh frozen plasma, have been
tried, but have been found to be either too risky or of
no, or only marginal, benefit. High doses of intravenous
furosemide may decrease the need for dialysis in some patients.
In contrast to TTP (vide infra), the value of plasma
exchange in D+ HUS is unclear. Although there are
anecdotal reports suggesting benefit, there was only one
prospective controlled clinical trial in children; it included
both D+ and D~ patients. The results of this study suggest
that it may be helpful in certain high-risk patients. The
response, if any, to plasma exchange is certainly not
dramatic, however, and in no way approaches the response seen in most cases of TTP. Moreover, the difficulty
in obtaining vascular access in small infants, the high
volume of blood required for operation of some systems,
the expense, and the risk of infection in infusing fresh
frozen plasma, coupled with the usually favorable outcome for those treated conservatively, argue against its
routine use. There is a temptation to initiate plasma
exchange for those doing poorly, especially for those
with severe central nervous system dysfunction. Even
so, if used, plasma exchange should be done with the
understanding that it is of unproven value in improving
the outcome in these unfortunate patients.
There is greater enthusiasm for using plasma exchange
in children with atypical (nondiarrheal) HUS, and some

235

32. HUS/TTP
reports suggest benefit compared to historical controls who
did not receive plasma exchange. There are no published
prospective controlled studies to support its recommended
use, however. Plasma manipulation (infusion or exchange)
is commonly used to treat D HUS in adults, especially if
there is neurological involvement or a relapsing course.
Anecdotal reports suggest benefit, but its efficacy has not
been tested in large prospective trials.
HUS occurring during pregnancy usually resolves following delivery; cases secondary to scleroderma and associated malignant hypertension benefit from treatment with
angiotensin-converting enzyme (ACE) inhibitors.
Therefore, meticulous attention to fluid and electrolyte
balance, control of any seizures or hypertension, blood
transfusions for severe anemia, the appropriate use of dialysis, and aggressive nutritional support constitute appropriate management. Packed red blood cells should be administered for severe anemia. Since some survivors will
eventually require a renal transplant, consideration should
be given to using leukocyte-poor blood in order to avoid
presensitization. Patients with D pneumococcal-related
HUS should receive blood products that are free of antiT antibody. Maintaining the hematocrit between 33 and
35% may improve CNS oxygen delivery in those with severe
encephalopathy. Platelet transfusions should be used sparingly and limited to those with severe bleeding or those
about to undergo potentially dangerous invasive vascular
procedures (e.g., placement of subclavian dialysis lines),
since administering platelets may provide substrate for additional platelet microthrombi deposition. Total parenteral
nutrition will be necessary for those with severe colitis.
Because most children have severe hypertriglyceridemia,
it may not be possible to administer lipids. Because pancreatic TMA with subsequent insulin-dependent diabetes mellitus can occur, frequent blood sugar measurements are
necessary, especially in those receiving high concentrations
of dextrose.
Outcome

Approximately 5% of children die during the acute phase

of D+ HUS, usually from extrarenal disease. Brain involvement (e.g., infarction, generalized edema) accounts for the
majority of fatal complications, but bowel infarction, as
well as heart or lung TMA, can be fatal. An additional
5% are left with chronic brain damage or end-stage renal
disease (ESRD). The remaining 90% experience functional
recovery, though about 5% are left with chronic hypertension and approximately half have either proteinuria or diminished glomerular filtration rate (GFR). Sometimes
chronic abnormalities appear after a period of apparent
recovery. Approximately 10% are left with both chronic
proteinuria and low GFR and are therefore at high risk
for eventually developing ESRD. Those who experience
prolonged oliguria (e.g., >10 days) are at very high risk for
chronic renal damage. Additional sequelae include rectal
stricture (3%), diabetes mellitus (8%) that can be perma-

nent, and neurological deficits (e.g., hemiparesis) from

stroke (3%). Sequelae secondary to involvement of the
retina, heart, or lungs are rare (<1%). Although a recurrence occurs in only about 1% of D+ cases, it happens in
approximately 10% of those with ESRD who receive a
kidney transplant. Children with D HUS, however, experience about a 20% recurrence rate. Moreover, in most parts
of the world, those with D HUS experience greater mortality and more ESRD than those with D+ disease. The outcome is worse in adults. Overall mortality is approximately
15 to 30%, chronic renal failure occurs in 20 to 30% of
survivors, and recurrence is seen in about 25% of cases.
This difference in outcome appears to be largely, but not
entirely, due to the much higher incidence of D cases
in the adult population. Adults who have familial HUS
experience a high incidence of ESRD, and postpartum,
cancer, and mitomycin -associated HUS patients experience
a 50% or greater mortality rate.
THROMBOTIC THROMBOCYTOPENIC PURPURA
Epidemiology

Although there are a few reports of TTP following

EHEC (e.g., E. coli O157:H7), enteric infections, postdiarrheal TTP is uncommon. It usually occurs during or after
pregnancy, in association with certain drugs (e.g., birth
control medications, mitomycin, and cyclosporine), systemic disorders (e.g., systemic lupus erythematous and scleroderma) and malignant hypertension. More reports of
AIDS-associated TTP are appearing in the literature.
In contrast to D+ HUS, there is no seasonal variation
for TTP. There is a higher incidence of disease in women,
who account for about 70% of all cases. Even though
the disease occurs at all ages, it is rare during infancy
and in the very elderly (>80 years of age). In most
series the peak incidence is during the third and fourth
decades, with an average age of about 35 years. The
incidence of TTP appears to be increasing, but it cannot
be entirely attributed to the AIDS epidemic. The current
estimate is 0.37 case per 100,000 persons per year. African
Americans are afflicted more often than whites, with a
relative risk ratio of 3.4.
Pathogenic Cascades

The pathogenic cascade of TTP has not been studied

to the same extent as that of D+ HUS, but it is probably
similar. Even though it is assumed that damage to endothelial cells of involved organs is also pivotal in TTP,
the cause of the damage is usually obscure. Moreover,
patients with TTP, but usually not those with HUS, have
very large vWF multimers in their circulation during the
acute phase of their disease; multimers tend to persist
in those with chronic relapsing TTP. The very large
multimeric forms of vWF strongly react with platelets
and promote their aggregation.

Richard L Siegler

236
Prodrome
TTP is only rarely preceded by diarrhea, but may follow
a viral-flu-like syndrome characterized by fatigue, fever,
abdominal pain, nausea, and vomiting.
Clinical Features
In contrast to classic D+ HUS, patients with TTP generally present with fever (Table 2), bleeding and, in about
half of the cases, CNS dysfunction. Brain involvement eventually occurs in approximately 90% of cases and is usually
the dominant feature of the syndrome. Headache, somnolence, confusion, aphasia, hemiparesis, and minor motor
irritability (e.g., tremor, jerkiness) are very common. Seizures occur in about one-third of patients, and coma in
approximately 10%. The neurological findings often fluctuate and can be fleeting. Hemorrhage, in the form of
gastrointestinal (GI) bleeding, is seen in the majority of
cases, and purpura is very common. Retinal lesions occur
in close to 20% of cases. Although the majority of patients
show signs of renal involvement (i.e., hematuria, proteinuria, and azotemia), in contrast to HUS, oliguric renal
failure with severe azotemia is uncommon. However, lifethreatening involvement of other vital organs (e.g., heart,
lungs) is seen much more frequently than in HUS. Laboratory findings are similar to those described for HUS, except
that the thrombocytopenia tends to be more severe. This
may explain the higher incidence of epistaxis, menorrhagia,
purpura, and GI bleeding.
Management
The supportive therapy guidelines for HUS mentioned
above also apply to treatment of TTP, but in contrast to
D+ HUS, there are also specific therapies of value. It is
difficult to determine the relative efficacy of the various
specific therapies, however, since in most reported series
two or more therapies were administered concurrently.
Even so, there is a consensus that plasma therapies offer
the greatest benefit, and they should be initiated as soon
as the diagnosis is established. Plasma exchange is probably
the most effective modality and with good response in 70
to 90% of cases. Its efficacy is assumed to result from removal of harmful mediators of TMA, coupled with the
replacement of factors needed to inhibit the disease process. The initial replacement solution should be fresh frozen
plasma; those who fail to respond can sometimes be salvaged by changing the replacement solution to cyrosupernatant, which lacks the very largest VWF multimers. There
is no consensus as to frequency and duration of therapy,
but seven consecutive daily treatments followed by alternate day exchanges in those showing improvement, using
a total exchange volume per treatment of 1 to 1.5 times
the patients predicted plasma volume, is reasonable. Treatment should be continued until remission is achieved. It is
not certain that plasma exchange is superior to the infusion
of fresh frozen plasma alone. The better results reported

with plasma exchange may be because plasma exchange

allows the infusion of much larger volumes of fresh frozen
plasma, especially in those with oligoanuria. Periodic infu sions of fresh frozen plasma (without plasma exchange)
are often effective in preventing relapses in those with
recurrent TTP.
Vincristine is probably of value in those who fail to
respond to plasma exchange. Corticosteroids, iv immunoglobulins (gamma globulin), and antiplatelet agents (e.g.,
dipyridamole, dextran 70, ticlodipine) have all been used,
usually in combination with other modalities, making if
difficult to evaluate their efficacy. Splenectomy may be
helpful for those who fail to respond to plasma manipulation and vincristine, or who continue to experience
recurrences in spite of prophylactic infusions of fresh
frozen plasma. However, this procedure carries a very
high risk.
In summary, plasma exchange is the treatment of choice
and should be instituted promptly. Those who fail to respond can be treated with vincristine. Splenectomy should
be reserved for those who fail to benefit from less extreme therapies.
Outcome
Both mortality and recurrence are higher in TTP than
in D+ HUS. Without treatment, the death rate approaches
90%; even with modern plasma therapy there is approxi mately a 15% mortality rate. Uncontrollable multisystem
involvement (e.g., brain, heart, lungs) accounts for most
deaths. Recurrences occur in approximately 20% of cases.
The relative infrequency of TTP, compared to D+ HUS,
has hampered the acquisition of long-term outcome information.
Bibliography
Bell WR, Braine HG, Ness PM, Kickler TS: Improved survival in
thrombotic thrombocyto penic purpurahemolytic uremic syndomre. Clinical experience in 108 patients. N Engl J Med
325:398-403, 1991.
Hayward CPM, Sutton DMC, Carter Jr WH, et al.: Treatment
outcomes in patients with adult thrombotic thrombocytopenic
purpurahemolytic uremic syn drome. Arch Intern Med
154:982-987, 1994.
Melnyk AMS, Solez K, Kjellstrand CM: Adult hemolytic-uremic
syndrome. A review of 27 cases. Arch Intern Med 155:20772084, 1995.
Remuzzi G: HUS and TTP: Variable expression of a single entity.
Kidney Int 32:292-308, 1987.
Remuzzi G, Ruggenenti P: The hemolytic uremic syndrome. Kidney Int 47:2 -19, 1995.
Rock GA, Shumak KH, Buskar NA, et al.: Comparison of
plasma exchange with plasma infusion in the treatment of thrombotic thrombocytopenic purpura. N Engl J Med 325:393-397,
1991.
Ruggenenti P: Thrombotic thrombocytopenic purpura and related
disorders. Hematol/Oncol Clin North Am 4:219-241, 1990.
Ruggenenti P, Remuzzi G: The pathophysiology and management
of thrombotic thrombocytopenic purpura. Eur J Haematol
56:191-207, 1996.

33. Renal Manifestations of HIV

237

Schieppati A, Ruggenenti P, Cornejo RP, et al.: Renal function

at hospital admission as a prognostic factor in adult hemolytic
uremic syndrome. J Am Soc Nephrol 2:1640-1644, 1992.
Siegler RL: Management of hemolytic-uremic syndrome. J Pediatr
112:1014-1020, 1988.
Siegler RL: Spectrum of extrarenal involvement in postdiarrheal
hemolytic-uremic syndrome. J Pediatr 125:511-518, 1994.
Siegler RL: The hemolytic uremic sydnrome. Pediatr Clin North
Am 42:1505-1529, 1995.

Siegler RL: Atypical hemolytic-uremic syndrome: A comparison

with postdiarrheal disease. J Pediatr 128:505-511, 1996.
Siegler RL, Milligan MK, Burningham TH, Christofferson RD,
Chang S-Y, Jorde LB: Long-term outcome and prognostic indicators in the hemolytic-uremic syndrome. J Pediatr 118:195200, 1991.
Siegler RL, Pavia AT, Christofferson RD, Milligan MK: A 20
year population-based study of postdiarrheal hemolytic uremic
syndrome in Utah. Pediatrics 94:35-40, 1994.

RENAL MANIFESTATIONS OF HIV

PATRICIA Y. SCHOENFELD

Human immunodeficiency virus (HIV) infection causes

structural as well as functional abnormalities of the kidney
that can occur at all stages of the illness. However, most
of the acute complications are more common during later
stages of disease, when clinical acquired immunodeficiency
syndrome (AIDS) is present. Most clinical interest has focused on the occurrence of specific patterns of glomerular
disease, such as HIV-associated nephropathy (HIVAN),
but immune complex glomerulonephritis occurs in 35% of
biopsied patients with HIV infection and other forms of
renal disease that may be related to HIV co-infections such
as hepatitis C (HCV) and hepatitis B (HBV) infection are
also being recognized with greater frequency. It is well
recognized that fluid, electrolyte, and acid-base disturbances and associated endocrine abnormalities occur with
regularity in HIV-infected patients. Acute renal failure due
to drug-induced nephrotoxicity or acute tubular necrosis
associated with hypovolemia and septic shock is common
in sick, hospitalized patients. Finally, end-stage renal failure
(ESRD) requiring dialysis is growing in frequency and patients are surviving longer with more successful AIDS therapy and appropriate dialysis treatment.
FLUID AND ELECTROLYTE ABNORMAL/TIES
Hyponatremia

Of the commonly encountered fluid and electrolyte dis turbances summarized in Table 1, the most frequent abnormality is hyponatremia. This disorder has been found in

Primer on Kidney Diseases, Second Edition

30 to 60% of patients hospitalized with HIV infection. More

than half of those patients develop hyponatremia from
volume depletion, usually due to extrarenal losses such as
diarrhea and vomiting. Inadequate volume replacement or
restoration of losses with dilute fluids results in dilutional
hyponatremia superimposed on absolute sodium depletion.
Clinical management includes replacement of salt and water deficits, as well as measures to treat the underlying
cause of volume depletion.
Hyponatremia can also occur as a result of inappropriate
secretion of antidiuretic hormone (SIADH) associated with
common pulmonary and intracranial diseases such as Pneumocystis carinii pneumonia, toxoplasmosis, and tuberculosis. Treatment of SIADH consists of fluid restriction, result ing in negative water balance and a rise in serum sodium.
Persistent release of ADH due to infections that are slowly
responsive to treatment may also be managed with the use
of demeclocycline 600 to 1200 mg/day to block the action
of ADH on the renal tubule (see Chapter 7).
HYPOKALEMIA

Abnormalities of serum potassium concentration are

summarized in Table 1. Severe diarrhea due to gastrointes tinal infection can often present with volume depletion,
hypokalemia, and metabolic acidosis. Treatment of the underlying cause of diarrhea is of principal benefit, but most
patients require large quantities of volume with bicarbonate and potassium replacement to correct the electrolyte
abnormalities. A number of drugs frequently used in paCopyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

238

Patricia Y. Schoenfeld
TABLE I

Fluid and Electrolyte Abnormalities in HIV Infection

Disorders of serum sodium concentration
Hyponatremia
Extrarenal salt and water loss with dilute fluid replacement
Syndrome of inappropriate antidiuretic hormone secretion
(SIADH)
Adrenal dysfunction ? Renal salt wasting
Hypernatremia Drug- induced nephrogenic
diabetes insipidus
Disorders of serum potassium concentration
Hypokalemia Severe diarrhea
Drug-induced tubular dysfunction
Hyperkalemia
Adrenal insufficiency or isolated mineralocorticoid
deficiency
Renal failure
Primary renal disease Drug induced
renal disease Hyporeninemic
hypoaldosteronism Drug related
Trimethoprim P entamidine
Acid-base disorders Respiratory
alkalosis and acidosis Metabolic
acidosis Bicarbonate loss from
diarrhea Renal acidosis from renal
failure Lactic acidosis
Other disorders
Hypocalcemia
Foscarnet
Pentamidine
Hypercalcemia
CMV infection
MAC complex
Lymphoma
Hyperphosphatemia
Foscarnet induction therapy (2nd week)
Nephrogenic diabetes insipidus
Foscarnet
Amphotericin B Increased serum uric
acid concentration
Rif ampin
Ethambutol
Pyrazinamide
Didanosine (ddl)

tients with AIDS can also result in hypokalemia, due to

tubular dysfunction, such as that seen with amphotericin
B (see Chapter 3).
HYPERKALEMIA

Hyperkalemia is most often seen in the setting of renal

insufficiency associated with acute or chronic renal parenchymal disease. It may also be a manifestation of mineralo-

corticoid deficiency due to either Addison's disease or more

selective defects in aldosterone biosynthesis. Renal tubular
abnormalities caused by interstitial nephritis, as well as
adrenal dysfunction, have been reported to result in deficient production of renin and aldosterone [hyporeninemic
hypoaldosteronism (HRHA)], which is clinically manifested by metabolic acidosis, hyperkalemia, and mild hyponatremia. Clinical HRHA has been reported to occur in
50% of patients with unexplained persistant hyperkalemia.
This disorder, termed type IV renal tubular acidosis, will
often respond clinically to treatment of the underlying dis order and the use of loop diuretics or fludrocortisone to
correct the distal tubular defects in potassium and acid
secretion (see Chapters 9 and 13).
Pentamidine has been shown to inhibit distal nephron
reabsorption of sodium by blocking apical sodium channels
in a fashion similar to the potassium-sparing diuretic amiloride. This results in a decrease in the electrochemical
gradient that drives distal potassium secretion and
thus can result in hyperkalemia. High-dose trimethoprimsulfamethoxazole (20 and 100 mg kg"1 day" 1, respectively) and trimethoprim dapsone has also been reported
to cause an increase in serum potassium in 20 to 53%
of patients, due to the action of trimethoprim, which
also mimicks that of amiloride. This side effect usually
occurs 7 to 10 days after the start of treatment and can
increase further with continuing therapy. Patients require
close monitoring of serum potassium to prevent lifethreatening hyperkalemia.
ACID-BASE DISORDERS

The most common acid-base disturbances in AIDS patients are respiratory alkalosis and respiratory acidosis due
to pulmonary and central nervous system opportunistic
infection. Respiratory alkalosis is often the initial disturbance, but as lung disease worsens and gas exchange deteriorates, hypercapnia and respiratory acidosis will occur.
Metabolic acidosis occurs as a result of several different
causes listed in Table 1.
1. Hyperchloremic metabolic acidosis occurs in
patients with large diarrheal losses of alkali in the
stool secondary to intestinal infection. This type of
acidosis can also present as renal tubular acidosis
due to distal tubular dysfunction associated with
drug toxicity or in patients with isolated
hypoaldosteronism or, more rarely, adrenal
insufficiency.
2. Patients with acute or progressive chronic renal
insufficiency develop metabolic acidosis, which may
be accompanied by an increase in anion gap.
3. Lactic acidosis manifested by markedly elevated
blood lactate levels, has been reported in a small
number of patients, but without associated
hypoxemia, tissue hypoperfusion, malignancy, or
sepsis. Some patients were taking Zidovudine,
which suggests a possible relationship between the
drug and mitochondrial dysfunction. While this is a
rare complication, its serious nature requires

33. Renal Manifestations of HIV

239

prompt diagnosis and treatment with conventional

measures as well as thiamine.
Other Electrolyte Disturbances
Abnormalities in serum calcium, phosphorus, magnesium, and uric acid concentrations have been reported,
although the causes of these disturbances are not well understood. The use of Foscarnet has been found to produce
a dose-related drop in ionized calcium with normal total
serum calcium. It is also frequently associated with transient hyperphosphatemia during the second week of induction therapy. Granulomatous infections, disseminated cytomegaloyirus infection, human T cell leukemia, and
lymphoma can cause hypercalcemia, with the latter being
due to the production of 25-hydroxylase by tumor cells
resulting in increased hydoxylation of 25-hydroxyvitamin
D to dihydroxyvitamin D3. Other drugs listed in Tables 1
and 2 can cause various disturbances in these electrolytes.
Infection with P. carinii pneumonia is associated with macrophage dysfunction, which has been reported to result
in abnormal vitamin D metabolism and elevated 1,25dihydroxyvitamin D levels, which corrects with treatment
of the pneumonia. Hypercalcemia may be missed due to the
frequent occurrence of hypoalbuminemia in HIV-infected
patients. However, measurement of the ionized calcium
level will reveal the presence of this complication.

ACUTE RENAL FAILURE

Acute renal failure (ARF) has been reported to occur
in 3 to 4 % of HIV-infected patients in New York and
Miami; another center in New York reported an incidence
of 20% in a retrospective analysis of AIDS patients in the
mid-1980s. These figures compare to an incidence rate of
4 to 5% in hospitalized non-HIV-infected patients. The
cause of ARF in 88 patients from the latter study was
volume depletion in 30%, drug toxicity in 42%, and shock
or hemodynamic instability in 17%. Others have reported
drug toxicity (59%) or sepsis, either directly or indirectly
(75%), to be the major cause of ARF. Table 2 summarizes
the major factors causing ARF, including some of the more
common drugs responsible for nephrotoxicity. Both oliguric and nonoliguric forms of renal failure can be encountered, and the natural course of the renal disease is similar
to that associated with other acute illness. Survival and
recovery from ARF are usually dependent on the nature
and severity of the underlying illness, and the mortality
rate in AIDS patients is about 50 to 60%a figure that is
not different from the general population. Dialytic and
conventional support measures are often successful in reversing renal insufficiency, particularly in patients whose
HIV infection is not as advanced.
STRUCTURAL RENAL DISEASE
HIV-Associated Nephropathy (HIVAN)

TABLE 2
Cause of Acute Renal Failure in HIV -lnfected Patients
Drug toxicity

ATN

TMP -SMX
Foscarnet
Amphotericin B
+ + Rifampin
Acyclovir
Dapsone (rare)

+++
+++
+/
+ + + Pentamidine
+++
++
+
+

Sulfadiazine
Interferon-a

TIN

Other
Hyperkalemia
Crystalluria
Hypokalemia
Hyperkalemia
Crystalluria/
obstruction
Papillary necrosis

++
MPGN

Acute tubular necrosis

Rhabdomyolysis
Hemolytic uremic syndrome/thrombotic thrombocytopenic
purpura
Postinfectious GN Plasmacytic

This unique clinical and histopathological renal disease

is thought to be the direct result of HIV in the kidney. It
occurs in about 5 to 10% of HIV-infected patients, but the
prevalence and geographical distribution are not uniform
and depend on specific risk factors for HIV which are race,
gender, and social behavior dependent. There is a striking
predilection of HIVAN for African-American individuals,
as is also true for focal segmental glomerulosclerosis
(FSGS) associated with intravenous drug use (IDU). This
is in contrast to IgA nephropathy, which has been reported
primarily in whites, and other forms of immune complex
disease, which occur in all racial groups. HIVAN is 7 to
10 times more common in men than women, and 30 to
60% have a history of IDU. Thus, in the United States, the
typical patient with HIVAN is a young black male with
a history of IDU who presents with the typical clinical
presentation shown in Table 3. HIVAN has been reported

TABLE 3
Clinical Manifestations of HIVAN

interstitial nephritis
Obstructive uropathy
Crystalluria (acyclovir, sulfadiazine, uric acid) Tumor
lysis syndrome (increased PO4 , uric acid)
Retroperitoneal fibrosis Ureteral compression
(tumor, lymph nodes)
ATN, Acute tubular necrosis; TIN, tubulointerstitial nephritis;
GN, glomerulonephritis; MPGN, membranoproliferative glomerulonephritis

Heavy proteinura
Azotemia Enlarged
kidneys Normal blood
pressure
Rapid progression to ESRD without treatment
Improved renal function and delayed progression with
combination antiviral treatment
ESRD, end-stage renal disease.

240

Patricia Y. Schoenfeld

to occur in about 8% of children with AIDS from the

University of Miami.
The pathology of HIVAN has been well characterized
and is summarized in Table 4 and Figs. 1 and 2. There are
characteristic glomerular, tubulointerstitial, and ultrastructural lesions. The glomeruli show diffuse epithelial cell
changes with prominent collapse of the glomerular tuft.
Most biopsies show FSGS and obsolescent glomeruli; earlier stages have visceral epithelial cell swelling and hypertrophy with periodic acid-Schiff positive inclusion droplets
which stain with immunoglobulin sera on immunofluores cence. There are typically large dilated tubules containing
hyaline occlusive casts and prominent lymphocytic infiltration of the interstitium. Ultrastructural changes consist of
numerous tubuloreticular structures in the endoplasmic reticulum of glomerular and endothelial cells and in interstitial leukocytes. Nuclear bodies are noted in greater frequency. The ultrastructural changes are not unique to
HIVAN, as they are also seen in idiopathic FSGS and
heroin nephropathy. They may represent a common viral
marker for different forms of renal disease. Collectively,

TABLE 4

Pathology of the Kidney in HIV Infection

Interstitial renal disease
Infections
CMV
Cryptococcosis
Histoplasmosis
Mycobacteria
Fungal agents (candida, aspergillosis)
Malignancy
Kaposi's sarcoma
Lymphoma Acute
tubular necrosis
HIV-associated nephropathy (HIVAN)
FSGS
Widespread visceral epithelial changes
Global collapse of the glomerular tuft
Interst itial fibrosis
Tubular dilatation with proteinaceous casts
Ultrastructural changes
Tubuloreticular inclusions
Nuclear bodies
Granulofibrillary transformation
Other glomerular disease
Immune complex glomerulonephritis
Membranoproliferative GN (HCV and cryoglobulinemia,
HBV related)
IgA nephropathy
Membraneous nephropathy
Postinfectious GN
Diffuse proliferative GN
Mesangial hyperplasia
Minimal change disease AA
Amyloidosis
Note: CMV, cytomegalovirus; FSGS, focal segmental glomerulosclerosis; GN, glomerulonephritis; HBV, hepatitis B virus; HCV,
hepatitis C virus

FIGURE I Glomerulus from a patient with HIV-associated nephropathy showing extensive global collapse of capillaries and
hypertrophy and slight hyperplasia of visceral epithelial cells. Also
note interstitial fibrosis and tubular atrophy (henatoxylin -eosinmethionine stain). Courtesy of Dr. J.C. Jennette.

the histological findings above are the basis for defining

HIVAN as a distinct renal disease, different from the other
causes of FSGS.
The pathogenesis of HIVAN has been intensely studied
over the past 10 years, and the accumulated data provide
substantial evidence that HIVAN is caused by localization
of the virus in the kidney. However, the mechanism by
which HIV leads to glomerulosclerosis remains unknown.
Mesangial cells of the normal kidney express CD4 antigen,
which is the receptor for the gp!20 envelope protein of
HIV. HIV core p24 antigen has been localized by immunochemistry in tubular epithelial cell cytoplasm. The HIV
genome has been found in tubular and glomerular epithelial cells using in situ hybridization and a cDNA nucleic
probe in patients with HIVAN, but not in patients with
immune-mediated glomerulonephritis or HIV seronegative
patients. Several investigators have succeeded in infecting
glomerular cells by exposing them to HIV in vitro. With
more sensitive PCR techniques, HIV DNA has been detected in HIV-infected patients with proteinuria, and the
genome was present in all renal cell types except interstitial
cells. All these studies suggest that the HIV genome is
ubiquitous in renal tissue of HIV-infected patients, regardless of type of renal disease. However, the genome itself
is probably not enough for the development of sclerosis,
and other mechanisms are necessary for the induction of
the disease. Other possible factors in the pathogenesis of
renal disease, especially the interstitial pathology, include
the presence of many different infectious agents that have
been found in kidneys of patients dying with AIDS. Cytokines and growth factors are important in the progression
of idiopathic FSGS and may contribute to the rapid progression in HIV patients.
Until recently the clinical course of HIVAN has been
one of inexorable progression to ESRD in 6 to 12 months.
As summarized in Table 5, different types of treatment
have been attempted with some modest success to date.
These include the use of antiviral agents such as the nucleo-

33. Renal Manifestations of HIV

241

FIGURE 2 Electron micrograph of glomerular capillary from a patient with HIV-associated nephropa-thy,
demonstrating endothelial tuhuloreticular inclusions (arrow).

side analogues Zidovudine (AZT) and /or didanosine

(ddl), which have been used by several groups who reported temporary remission or delay of progression of renal
disease for as long as 33 months. The most benefit with
these agents appears to be in patients who are treated early
and before azotemia develops. The use of immunomodulator therapy has been reported to improve renal function
and to delay progression in small number of patients. However, steroid-induced complications were significant and
patients relapsed when the drug was stopped. Recently,
even greater hope for HIV-infected patients with renal
disease has emerged with the institution of combination
antiviral therapy which includes nucleoside analogues and
a protease inhibitor, such as AZT, 3TC, and Indinavir.
Several reports have demonstrated striking falls in viral

TABLE 5

Treatment of HIVAN
Nucleoside analogues
AZT (zidovudine)
ddl (Didanosine)
3TC (Lamivudine)
Immunomodulators
Steroids
Cyclosporine
Angiotensin-converting enzyme inhibitors
Combination nucleoside analogue and protease inhibitor
antiretroviral therapy AZT, 3TC, Indinavir

load with the use of protease inhibitors, and preliminary

experience with this type of drug regimen in predialysis
patients and one dialyzed patient in our clinics has shown
improvement in renal function associated with large decreases in viral load measurements to undetectable levels
(less than 500 copies). When to start such treatment remains controversial, but all agree that it should be used in
advanced HIV disease and many would use it in patients
with CD4 counts below 300 to 400 or with viral loads greater
than 5000 copies. Use in patients with renal disease is not
well defined, but it would seem logical to start treatment
when HIV-related kidney disease is diagnosed, especially
HIVAN. Experience with this regimen in the treatment of
immune-mediated renal disease in HIV-infected patients
(see below) is not available. It is recommended that viral
load measurements be used for following the course of
disease and obtained twice at 1-month intervals to establish
a baseline, then quarterly, and 4 weeks after establishing
new therapy.

Immune-Mediated Renal Disease

As mentioned earlier, 25 to 50% of patients with HIV

infection have been found to have renal histopathological
abnormalities on renal biopsy consistent with a diagnosis
of immune-mediated renal disease, which may be associated with some features of HIVAN. Membranoproliferative and diffuse proliferative glomerulonephritis, membranous, and IgA nephropathy have been found on renal
biopsy in patients with HIV infection. The significance of

242

Patricia Y. Schoenfeld

such disparate types of renal disease is unclear, but the

possible pathogenetic mechanisms include the deposition
of HIV antigen-antibody immune complexes in response
to infectious agents or as a specific immune complex directly causing renal disease. In some cases, the renal disease
may be indirectly related or even unrelated to the HIV,
but influenced by the patient's genetic makeup, infectious
complications of AIDS, HIV therapy, or differences in
systemic host responses.
The diagnoses most commonly reported include IgA nephropathy, focal proliferative, and membranoproliferative
glomerulonephritis. IgA nephropathy occurs mostly in
whites and tends to be mild or slowly progressive compared
to the more rapid course of HIV AN. The relationship of
IgA nephropathy to HIV is unclear, but increased circulating levels of IgA and the presence of circulating immune
complexes containing IgA are common findings in HIVinfected patients and may result from abnormal IgA regulation, which leads to deposition in the kidney and renal
disease.
HCV and HIV co-infection are common, especially in
patients with a history of IDU as the primary risk factor
for these infections. Such patients presenting with hematuria and cryoglobulinemia frequently have membranoproliferative glomerulonephritis on renal biopsy, a finding in
about 50% of HIV-infected patients presenting with hematuria to our renal clinic. It is unclear whether the cryoprecipitate is related to anti-HCV associated antibodies or
whether immune complexes containing HIV antigens are
involved as well.
Other forms of renal disease including minimal change
disease, AA amyloidosis, hemolytic uremic syndrome/
thrombotic thrombocytopenic purpura, and tumor invasion
of the kidneys have been reported with less frequency.
END-STAGE RENAL DISEASE

While renal transplantation is not currently an option

for most patients with ESRD and HIV, many patients are
treated successfully with chronic dialysis therapy. The incidence of centers providing dialysis for Patients with HIV
has increased from 11% in 1985 to 37% in 1994. During
1994, 1.5% or 3144 dialyzed patients had HIV; however
since only 28% of centers tested for the AIDS virus at entry
to dialysis, this probably represents an underestimate of
the prevalence of dialyzed patients with HIV. Incidence
figures are not available. However, in the dialysis program
at San Francisco General Hospital, 15% of new dialysis
patients each year are positive for HIV.
The choice of dialysis modality has implications not pres ent in the non-HIV-infected patient, including risk of dis ease transmission, care of patients with communicable dis eases, ethical issues related to the care of the terminally ill,
and shortened survival. For many of these considerations,
peritoneal dialysis (PD) has had distinct medical advantages , including fewer invasive procedures, less exposure
to potential infections, and possibly less stimulation of the
immune system by blood membrane interactions. Staff who
care for HIV-infected patients on PD have fewer risks

of needle stick, less equipment involved, and less overall

contact with patients with communicable diseases. Potential disadvantages to PD therapy include increased frequency of peritonitis in some patients, peritonitis due to
fungal or other opportunistic organisms, and technique failure as advanced AIDS-related complications may limit the
ability to do self-care.
Survival of dialyzed patients with HIV has been recently
reviewed for a 10-year period from 1985 to 1995 in a group
of 62 patients treated at the University of California Renal
Center at San Francisco Hospital. The median survival for
all patients was 10 months compared to 17 months for
AIDS patients without ESRD followed by the Department
of Public Health in San Francisco. In patients with CD4
counts above 200 cells/mL, survival was substantially better,
33 months compared to 7 months in patients with CD4
counts below 200. Patients able to do PD had a significantly
longer survival than patients treated with hemodialysis, but
some of this advantage may be due to earlier stage of
disease in those patients. Survival was positively correlated
with CD4 count, as well as serum albumin concentration,
which fell progressively with disease progression. HIVrelated cognitive motor complex was more common in
dialyzed patients than reported for the general AIDS population, and also correlated significantly with overall survival.
The diagnosis and clinical care of HIV-infected patients
with renal disease are challenging and requires concerted
effort from nephrology and AIDS care professionals.
Appropriate antiviral agents, prophylactic care to prevent
complications, and adequate dialysis therapy are all required to prolong survival. New regimens of combination
antiviral therapy, including protease inhibitors, assessment
of the stage of illness with viral load measurements, and
earlier treatment of infection and renal disease offer
new hope for the delay of progression to ESRD and
for optimizing survival in those treated with chronic dialysis.
Bibliography
Carpenter CCJ, Fischl MA, Hammer SM, et al.: Antiretroviral
therapy for HIV infection in 1996. Recommendations of an
International Panel. JAMA 276: 146-154, 1996.
Cohen AH: Renal pathology of HIV-associated nephropathy. In:
Kinmel PL, Berns JS (eds) Renal and Urologic Aspects of
HIV Infection. Churchill Liningstone, New York, pp. 155-180,
1995.
D'Agati V, Appel GB: HIV infection and the Kidney. J Am Soc
Nephrol 8:138-152, 1997.
Humphreys MH: Human immunodeficiency virus-associated renal
disease. Kidney Int 48: 311-320, 1995.
Humphreys MH, Schoenfeld P: Electrolyte, acid-base, and endocrine disturbances in patients with HIV infection. In: Kinmel
PL, Berns JS (eds). Renaland Urologic Aspects of HIV Infection.
Churchill Livingstone, New York, pp. 27-40, 1995.
Mellors JW, Rinaldo CR, Gupta P, et al.: Prognosis in HIV-1
infection predicted by the quantity of virus in the plasma. Science
272: 1167-11170, 1996.
Pardo V, Strauss J, Abitbol C, Zilleruelo G: Renal disease in
children with HIV infection: In: Kimmel PL, Berns JS (eds)

3 3 . R e n a l M a n i festations of HIV
Renal and Urologic Aspects of HIV Infection. Churchill Livingstone, New York, pp. 135-153, 1995.
Rao TK, Friedman EA, Nicastri AD: The types of renal disease
in the aquired immunodeficiency syndrome. N EnglJ Med 316:
1062-1068, 1987.
Rao TK, Friedman EA: Outcome of severe renal failure in patients
with acquired immunodeficiency syndrome. Am J Kidney Dis
25: 309-398, 1995.

243
Rao TKS: Renal complications in HIV disease. Management of
the HIV-Infected patient, Part I. In Medical Clinics of North
America 80 (6), 1996.
Schoenfeld P, Mendelson M, Rodriquez R: Survival of ESRD
patients with HIV infection. J Am Soc Nephrol 6:561, 1995.
Schoenfeld P, Rodriquez R: Renal aspects of HIV disease. In:
Cohen PT, Sande MA, Volberding PA (eds) The AIDS Knowledge Base. 3rd ed. Little, Brown & Co., Boston, 1997, in press.

SECTION 5
ACUTE RENAL FAILURE

PATHOPHYSIOLOGY OF ACUTE
RENAL FAILURE
ROBERT SAFIRSTEIN

The syndrome of acute renal failure (ARF) is defined

as a reduction of glomerular filtration rate (GFR) that is
often reversible. The syndrome may occur in three clinical
settings: (1) as an adaptive response to severe volume
depletion and hypotension with structurally and functionally intact nephrons; (2) in response to cytotoxic insults to
the kidney when both renal structure and function are
abnormal; and (3) when the passage of urine is blocked.
Thus ARF may be classified as prerenal, intrinsic, and
postrenal. While this classification is useful in establishing
a differential diagnosis, it is now evident that many pathophysiologic features are shared among the different categories. The intrinsic form of the syndrome may be accompanied by a well-defined sequence of events: an initiation
phase characterized by daily increases in serum creatinine
and reduced urinary volume; a maintenance phase, where
GFR is relatively stable and urine volume may be increased; and a recovery phase in which serum creatinine
falls and tubule function is restored. This sequence of
events is not always apparent and oliguria may not be
present at all. The reason for this lack of uniform clinical
presentation is most probably a reflection of the variable
nature of the injury. It is also useful to classify ARF as
oliguric or nonoliguric based on the daily urine excretion.
Oliguria is defined as a daily urine volume of less than 350
to 400 mL/day. Stratification of the renal failure along these
lines helps in decision making, such as the timing of dialysis,
and seems to be an important criterion for response to
therapy (see below). This chapter considers the pathophysiology of the syndrome, focusing especially on the intrinsic
form of the disease, and introducing newer concepts of
what causes the syndrome, based on recent observations
in human and animal forms of the disease. This better
understanding of ARF has lead to newer approaches to
treatment that is hoped will improve outcome in this group
of patients.

appear indifferent to the stress (Fig. 1). Frank necrosis, as

is often seen experimentally, is not prominent in the vast
majority of human cases. Necrosis is usually patchy, involv ing individual cells or small clusters of cells, sometimes
resulting in small areas of denuded basement membrane.
Less obvious injury is more often noted, including loss of
brush borders, flattening of the epithelium, detachment
of cells, intratubular cast formation, and dilatation of the
lumen. While proximal tubules show many of these
changes, injury to the distal nephron can also be demonstrated when human biopsy material is closely examined.
The distal nephron is also the site of obstruction by desquamated cells and cellular debris.
Apoptosis has been noted in ischemic and nephrotoxic
forms of ARF. This form of cell death differs from frank
necrosis in that it requires the activation of a regulated
program that leads to DNA fragmentation, cytoplasmic
condensation, and cell loss without precipitating an inflammatory response (Table 1). In contradistinction to necrosis, the principal site of apoptotic cell death is the dis tal nephron.
Disruption of the cell cytoskeleton seems to be an important determinant of many of the early morphologic
changes, especially in ischemic injury. Loss of the integrity
of the actin cytoskeleton leads to flattening of the epithelium, with loss of the brush border, loss of focal cell contacts
and subsequent disengagement of the cell from the underlying substratum. Membrane proteins, including the integrins
and the Na, K-ATPase, redistribute in the plasma membrane as cells lose their polarity. The functional impact of
these changes is great as cells lose their capacity to achieve
vectorial transport, and the redistribution of the adhesion
molecules from the basolateral membrane sites provokes
intratubular obstruction. These sublethal changes contribute considerably to the severe impairment of function (see
section later on treatment of ARF), and may help explain
the often disproportionate decline in renal function when
compared to the minor morphologic changes observed.

MORPHOLOGY OF ARF
Molecular Responses to Renal Injury: Implications for Cell
Fate

The changes in renal epithelial morphology that accompany ARF are subtle. At least four cellular fates can be
identified in ARF: cells may die either by frank necrosis
or in apoptosis; they may replicate and divide; or they may

Primer on Kidney Diseases, Second Edition

In sections of the kidney taken from patients with ARF,

regeneration and necrosis coexist, indicating how early the
247

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

248

Robert Safirstein

FIGURE I Representative photomicrographs of outer stripe of outer medulla of rat

kidney 5 days after cisplatin injection (5 mg/kg body wt) demonstrating cell fate during
acute renal failure. (A) Necrosis of the S3 segment of the proximal convoluted tubule is
apparent (NT). The solid arrows show regenerating tubules, indicated by the uptake of
3
H-thymidine. (B) The open arrow shows an apoptotic body. Thick ascending limbs (Tal)
and collecting ducts (CD) are without apparent morphological damage.

attempt to repair the kidney is initiated in the course of

renal failure. Even in its most severe form, few patients who
survive initial dialysis require long-term dialysis, indicating
how effective this repair process is. The regeneration process is accompanied by increased renal DNA synthesis and
is proceeded by prominent changes in gene expression. The
changes in renal gene expression can be grouped in at least
three major categories (Table 2).
Many of the genes that are expressed after renal injury
are involved in cell cycle regulation and are similar to
those expressed when growth factors are added to cells to
stimulate them to enter the growth cycle. Although the
endogenous growth factors that serve this response have
not been identified, administration of growth factors exogenously has been shown to ameliorate and hasten recovery
from ARF (see Section later on treatment of ARF).

Another group of genes are proinflammatory and chemotactic and may be responsible for the apparent inflammatory aspects of ARF. Depletion of neutrophils, and
blockade of neutrophil adhesion each reduce renal injury
following ischemia indicating that the inflammatory response is in part responsible for some features of ARF. This
mechanism may be especially prominent in posttransplant
ARF. This proinflammatory state may also be important
in the prescription of dialysis, especially in sepsis where
consideration should be given to whether the procedure
will enhance or reduce cytokine production (see below).
These two aspects of the renal molecular response to
injurythe increases in protooncogene and chemokine expressionresemble what is observed in cells exposed to
adverse environmental conditions such as ionizing radiation, oxidants, and hypertonicity, and have been termed

TABLE I

Comparison between Apoptotic and Necrotic Cell Death

Stimuli
Occurrence
Adhesion between cells
Nucleus
Nuclear chromatin
DNA cleavage
Phagocytosis by other cells
Inflammation

Apoptosis

Necrosis

Physiologic
Single cells
Lost (early)
Convolution of nuclear outline and breakdown (karyorrhexis)
Compaction in uniformly dense masses
Internucleosomal, "laddering" appearance of distinct fragments on
agarose gels
Present
Absent

Pathologic
Groups of cells
Lost (late)
Disappearance (Karyolysis)
Clumping not sharply denned
Random; "smear" pattern on
agarose gels
Absent
Present

249

34. Pathophysiology of Acute Renal Failure

TABLE 2
Molecular Responses to Renal Ischemia

Increased gene expression

Genes involved in cell fate determinations: regeneration,
apoptosis
Transcription factors: c-jun, c-fos
Cyclin dependent kinase inhibitors: p21
Genes involved in inflammation
Chemokines: MCP -1, IL-8
Adhesion molecules: ICAM-1, integrins
Decreased gene expressionloss of mature phenotype
Prepro epidermal growth factor Tamm-Horsfall
protein Aquaporin -2 Sodium proton exchanger 3
(NHE3)

the stress response (Fig. 2). This response is thought to be

a major determinant of whether cells survive the insult or
not, and might be necessary for the repair of injured cells. In
some circumstances, the stress response leads to apoptosis
rather than survival, and is probably a function of the duration of the stress response, its degree, and the specific cell in
which the response takes place. In the ischemia/reperfusion
model of renal injury, which may also involve DNA damage, these genes are expressed for the most part in the
distal nephron, a site of both cell survival and apoptosis
(see above).
A summary of these important features of the renal stress
pathway and its possible consequences is given in Fig. 2.
It can be seen that initiation of the stress response may
ultimately determine much of the proinflammatory, reparative, cytoreductive, and functional aspects of renal failure.
Particular limbs of the response can be targeted for up-or

Nephrotoxins

downregulation to limit injury or improve function. The

recent data using growth factors (presumably to tip the
balance between cell gain and cell loss), as well as the data
on antiadhesion molecules to reduce inflammation, support
the notion that the renal stress response is an appropriate
target for therapy (see discussion on treatment of ARF).
The last group of changes involves what appears to be a
loss of the mature phenotype of the kidney, because
many of the changes in gene expression involve the loss
of proteins that are only expressed maximally during the
maturation of the kidney. These include the prepro epidermal growth factor and Tamm-Horsfall protein genes, whose
functions are unknown in the kidney, but also include the
downregulation of important membrane transporter genes,
such as Aquaporin-2, and NHE3. The loss of these latter
proteins may be responsible, in part, for the tubular reabsorptive defects typical of ARF (see below).
Pathophysiology of the Cell Injury

The mechanism of the changes in cell viability during

renal injury are complex and incompletely understood.
Most of the experimental data have been derived from the
ischemia-reperfusion model of acute renal failure and have
focused on necrotic cell death. Because as many as 50%
of patients have ischemia-induced ARF, the observations
should be relevant to a large portion of the patients at risk.
As mentioned above, different stresses initiate common
biochemical events, so that understanding the relevant
pathways of one stress will most likely be applicable to
others. Such studies have focused on three biochemical
pathways: intracellular calcium homeostasis, reactive oxy gen species, and phospholipases.
Depletion of intracellular ATP that accompanies ischemia increases the cytosolic concentration of calcium. Such

DNA Damage

Oxidative Stress

ACTIVATION OF THE RENAL STRESS RESPONSE

Transcription factors Secreted Cbemokines Structural Proteins

CELL FATE
FIGURE 2

INFLAMMATION

FUNCTION

The stress response and its consequences.

Robert Safirstein

250
increases in intracellular calcium can damage epithelial
cells by activating proteases and phospholipases, and can
disrupt cellular integrity. Controversy remains, however,
about the importance of this change even in the generation
of ischemic injury.
Restoration of renal blood flow after ischemia produces a
burst of reduced oxygen species from a variety of processes.
Resultant lipid, protein, and DNA damage could lead to
necrosis. Direct tests of their role in human ARF by the
use of oxygen scavengers have been disappointing so far,
however.
Members of the phospholipase family, which hydrolyze
membrane lipids, could contribute to ischemic renal injury
as they appear to do in other organs injured by ischemia.
Several products of phospholipid breakdown are vasoconstrictive and chemotactic and could participate in the functional and cytotoxic events of renal failure. Adequate as sessment of their role in the cytotoxicity of ischemia awaits
the availability of useful inhibitors of the various members
of the family.
PATHOPHYSIOLOGY OF ABNORMAL FUNCTION
Reduced GFR

From a consideration of the forces and flows at the glomerular capillary vascular bed it is possible to form a conceptual framework in which to analyze the causes of reduced GFR in ARF (Table 3).
Intrarenal vasoconstriction is the dominant mechanism
for the reduced GFR in acute renal failure, especially in
its initial phases. Reduction in renal blood flow reduces
net ultrafiltration pressure because it increases the rate of
rise of irac along the length of the glomerular capillary,
thus reducing net ultrafiltration pressure. If accompanied
by a reduction in arterial pressure or selective afferent
arteriolar vasoconstriction, PGC would also fall. The mediators of this vasoconstriction are unknown but tubule injury
seems to be an important concomitant finding. There is
some evidence experimentally that a low ultrafiltration coefficient may also play a role in reducing GFR during acute
renal failure, presumably due to mesangial cell contraction

TABLE 3

Determinants of Glomerular Filtration Rate

GFR = LpA x P n e t
LpA is the ultrafiltration coefficient, consisting of the hydraulic
permeability and area of the glomerular membrane
.P net = \P ATT is the net ultrafiltration pressure across the
entire glomerular capillary, where
AP = PGC - P P T is the hydrostatic pressure difference
between the glomerular capillary and Bowman's space, P Gc
is the glomerular capillary hydrostatic pressure, and P P T is
the intratubular hydrostatic pressure
ATT is

the net glomerular oncotic pressure, the difference

between glomerular oncotic pressure ( TT GC) and tubule
oncotic pressure ( JT> T)

and consequent reduction in the area available for filtration. While obstruction to the outflow of urine into the
collecting system is an obvious cause of reduced net ultrafiltration, less obvious is the intratubular obstruction that
results from sloughed cells and cellular debris that evolves
in the course of renal failure. The importance of this mechanism is highlighted by the improvement in renal function
that follows relief of such intratubular obstruction. Also,
when obstruction is prolonged (longer than a few hours),
intrarenal vasoconstriction is prominent. Damaged renal
epithelium are also abnormally permeable to inulin, and
thus backleak of glomerular filtrate may be an additional
mechanism of renal failure. Each of these mechanisms
vasoconstriction, mesangial cell contraction, tubular obstruction, and backleakcontribute individually or in combination during the course of ARF.
Apart from the increase in basal renal vascular tone
observed during ARF, it is also true that the stressed renal
microvasculature is more sensitive to otherwise tolerated
manipulations of the renal vascular bed. For example, the
use of nonsteroidal antiinflammatory drugs in patients with
severe liver disease may precipitate ARF. As a result, a
prerenal state may progress to an intrinsic form of renal
failure and thus reduce GFR further. Prolonged vasoconstriction may evolve into intrinsic ARF especially when
there is concomitant large vessel arterial disease. This latter
form of renal failure is often induced by the use of angiotensinconvertingenzyme inhibitors and/or diuretics. The vas culature of the injured kidney has an impaired vasodilatory
response and loses its autoregulatory behavior. This latter
phenomenon has important clinical relevance as the frequent reduction in systemic pressure during intermittent
hemodialysis may provoke additional damage that could
delay recovery of ARF.
PATHOPHYSIOLOGY OF THE
CONCENTRATING DEFECT

Another physiologic hallmark of intrinsic ARF is the

failure to concentrate urine maximally. The defect is not
responsive to pharmacologic doses of vasopressin and is
postreceptor in nature. The injured kidney fails to generate
and maintain a high medullary solute gradient. Because
the accumulation of solute in the medulla depends on normal distal nephron function, this is yet another example of
the role of the distal nephron in the pathophysiology of
ARF. The mechanism of this defect is not simply a function
of lethally injured proximal tubules, because necrosis is not
prominent in the distal nephron, but rather may involve
more subtle affects on function, such as the observed loss
of Aquaporin 2 expression following renal ischemia. The
failure to excrete a concentrated urine even in the face of
oliguria is a helpful diagnostic tool to distinguish prerenal
from intrinsic renal disease.
DIAGNOSTIC INDICES OF GLOMERULAR AND
TUBULAR FUNCTION USED IN THE DIFFERENTIAL
DIAGNOSIS OF ARF

A rapid rise in serum creatinine, which is sometimes

associated with a reduced urine volume, declares the pres -

3 4 . P a t h ophysiology of Acute Renal Failure

ence of ARF. Given the conceptual framework discussed

above, determination of whether the cause of the renal
failure is prerenal, intrinsic, or postrenal usually requires
only a few noninvasive tests, in addition to history and
physical.
If a prerenal cause of ARF is suggested by history, confirmation of the adaptive nature of the response can be
obtained by examining the urine and searching for evidence
of enhanced renal water and solute reabsorption. The findings of a highly concentrated urine with a low pH in the
absence of cellular elements is suggestive adaptive responses to volume depletion. These findings, combined
with a disproportionate rise of blood urea nitrogen as compared to creatinine and an elevated serum concentration
of uric acid, all point to prerenal causes.
A widely used aid to discriminate between prerenal and
intrinsic ARF is the determination of the fractional excretion of sodium,

251
TABLE 4

Therapeutic Targets of Treatment

Offsetting
Calcium channel blockage Atrial
vasoconstriction natriuretic factor Endothelin
blockade Nitric oxide regulation
Antiadhesion strategies,
Limiting inflammation anti-ICAM
antiintegrins
Biocompatable membranes
Cytokine absorbing biomembranes
Altering cell outcome
Dialysis dose
adjustments

Growth factors and "survival" factors

Hi flux membranes Continuous
arteriovenous
hemodialysis (CAVHD) Continuous
venovenous
hemodialysis (CVVHD) Continuous
venovenous
hemofiltration (CVVH)

FENa = (t/NaAPNa)/(t/ Cr/ X 100.

Low (<1%) fractional excretion indicates salt and water
avidity and is consistent with prerenal causes of renal failure. Lethal and sublethal renal cell injury, on the other
hand, would lead to diminished salt reclamation and failure
to reach maximum urine concentration, and, as a result,
FENa rises and exceeds 1%. This index is helpful when the
patient is oliguric.
A reliance on any of these determinations alone is hazardous because the regulation of salt and water metabolism
is complex and other factors besides the fullness of the
intravascular space may be present. For example, the fractional excretion of sodium may be low in intrinsic ARF
when there are other comorbid events that enhance salt
reabsorption. Coexistent heart and liver disease enhance
renal sodium retention even in the diseased kidney. Also,
the common use of diuretics may also confound interpretation of high fractional sodium excretion even when prerenal
causes predominate. Accurate history and physical are a
great help in this regard.
Postrenal causes are detected by renal sonography. Renal perfusion scans do not usually help distinguish among
the various causes of reduced renal perfusion except when
there is asymmetric perfusion suggestive of renal vascular
lesions. In this case, the presence of such lesions should
be confirmed by MRI and/or renal angiography.
TREATMENT OF ARF RE: IMPLICATIONS DERIVED FROM
NEWER UNDERSTANDING OF ITS PATHOPHYSIOLOGY

The mortality of ARF is still high, especially in those

patients who require dialysis. This high mortality is in part
due to a sicker and older population of patients receiving
potentially nephrotoxic medical and surgical therapies. Recent insights into the pathophysiology of ARF have provided newer targets for therapy with the hope of improving
outcome in these patients. Such approaches are summarized in Table 4.

Therapeutic Interventions Directed at Renal Vasodilation

Although the effector pathway for the intense vasoconstriction during ARF is as yet unknown, a variety of therapeutic approaches have been used to limit the decline in
GFR with inconsistent results. Calciumchannel blockade
to relax the renal vasculature and ameliorate renal failure
may be useful in ARF seen after renal transplantation,
cyclosporine, and radiocontrast dyes. It has not been effective in most other forms of renal failure. Dopamine, which
vasodilates the normal renal vasculature and increases sodium excretion, has not been effective clinically and is
associated with significant side effects, especially in the
critically ill patient. Atrial natriuretic peptide may improve
renal function in oliguric ARF patients, but not in those
who are nonoliguric. Isotonic saline infusion at relatively
modest rates, (which would increase renal blood flow and
provoke a diuresis), has been shown to ameliorate radiocontrast-induced ARF in patients with modest reduction of
renal function before exposure. Antagonism of the potent
renal vascoconstictive effects of endothelin shows promise
experimentally in ischemic renal failure, but there is no
clinical experience at present. Finally, modifying nitric oxide production has also been shown to protect kidneys
experimentally. Identification of the effector pathways responsible for the renal vascoconstriction offers the hope
of more successful therapy.
Approaches Based on Modifying the Inflammatory
Aspects of ARF

Several aspects of the renal stress response are proinflammatory in nature, including the increased expression
of the potent monocyte and neutrophil chemotactic chemokines monocyte chemotoctic protein 1 (MCP-1) and
interleukin-8 (IL-8). Overproduction of cytokines may be

Robert Safirstein

252

ISIGNAL
TRANSDUCT

RENAL CELL
STRESS

TRANSCRIPTION
PROGRAM

DEATH

REPAIR
SURVIVAL FACTORS
GROWTH FACTORS
CYTOKINES

FIGURE 3

Survival factors and cell fate during renal cell stress.

a factor in many aspects of renal failure, including vasoconstriction and leukocyte invasion. Blockade of intracellular
adhesion molecule I and integrin-mediated adhesion is a
promising new approach, which is mediated perhaps by
interfering with inflammation. Strategies directed against
the integrins may also operate by reducing intratubular
obstruction, as stated above. Additional insight into how
these chemokines work in the kidney will most likely yield
additional approaches.
Survival Factors

As dicussed above renal stress initiates a transcriptional

program that is intimately involved in cell fate. Some cells
participating in this response will survive and repair,
whereas others will die by apoptosis. What determines
whether a cell will recover from such injury or undergo
cell death by necrosis or apoptosis is probably a function
of the severity of the stress, the specific changes in gene
regulation that the cell is capable of mounting, and the
availability of survival factors in the cell's external milieu.
An example of these internal and external influences on
cell survival after toxic exposure is given by the sensitivity
of cells derived from patients with Xeroderma pigmentosa.
Such cells lack a critical DNA repair pathway and do not
survive otherwise tolerated doses of radiation. Also, the
survival of bone marrow cells damaged by cancer chemo therapeutic agents may be prevented by addition of IL3. Figure 3 provides the conceptual framework for this
approach. The provision of survival factors, such as trophic
cytokines and growth factors, in addition to accelerating
entry of cells into replicative phases of the cell cycle and
hence increase the rate at which cells reline injured tubules,
may also alter an apoptotic or even necrotic outcome to
one of survival and repair. Early success with exogenously
administered growth factors such as EGF and IGF (among
others) show promise in this regard. Survival factors identified during early morphogenesis of the kidney are also
likely to be used in treatment of the injured kidney in this
way, or be used prior to exposure to mitigate the injury.

Renal Replacement Therapy

The decision to offer renal replacement therapy should

be made cautiously, given the instability of the renal vasculature, the hypotension acute intermittent hemodialysis
provokes, and the proinflammatory state of the injured
kidney. For these reasons, hemodialysis via high flux biocompatible membranes that limit the frequency of hypotensive episodes and diminish cytokine production show great
promise in the treatment of ARF.
SUMMARY

(Great strides) have been made in the understanding of

the pathophysiology of ARF. Morphologic and molecular
information have broadened our notions of how the kidney
loses function and recovers from cytotoxic insults. The
study of the stress response and its consequences, the cell
cycle and its regulation, as well as continued advances in
the understanding of the embryology of the kidney, will
likely yield specific targets for therapy and newer survival
factors that could limit injury and hasten repair.
Bibliography
Chan L, Chittinandana A, Shapiro JI, Stanley P F Schrier RW:
Effect of an endothelin -receptor antagonist on ischemic acute
renal failure. Am J Physiol 266:F135-F138, 1994.
Chertow GM, Christiansen CL, Cleary PD, Munro C, Lazarus JM:
Prognostic stratification in critically ill patients with acute renal
failure requiring dialysis. Arch Intern Med 155:1505-1511,1995.
Conger J, Robinette JB, Hammond WS: Differences in vascular
reactivity in models of ischemic acute renal failure. Kidney Int
39:1087-1097, 1991.
Di Mari JF, Saggi S, Aronson P, Safirstein R: Renal ischemia
reperfusion injury reduces Aquaporin -2 and NHE3 Expression.
/ Am Sec Nephrol 7:1823, 1996.
Donohoe JF, Venkatachalam MA, Bernard DB, Levinsky N: Tubular leakage and obstruction after renal ischemia: structuralfunctional correlations. Kidney Int 13:208-222, 1978.
Goligorsky MS, Lieberthal W, Racusen LC, Simon EE: Integrin
receptors in renal tubular epithelium: new insights into patho physiology of acute renal failure. Am J Physiol 264:F1, 1993.

35. Approach to the Patient with Acute Renal Failure

253

Hakim RM: Clinical implications of hemodialysis membrane biocompatibility. Kidney Int 44:484-494, 1993.
Holbrook NJ, Fornace AJ, Jr: Response to adversity: molecular
control of gene activation following genotoxic stress. New Biol
3:825-833, 1991.
Kelly KJ, Williams LWW, Colvin RB, Bonventre .IN: Antibody
to intracellular adhesion molecule-I protects the kidney against
ischemic injury. Proc Natl Acad Sci USA 91:812, 1994.
Noiri E, Peresleni T, Miller F, Goligorsky MS: In vivo targeting
of inducible NO synthace with oligodeoxynucleotides protects
rat kidney against ischemia. J Clin Invest 97:2377-2383, 1996.
Nouwen E, Verstrepen W, Buyssens N, Zhu M, De Broe M:
Hyperplasia, hypertrophy, and phenotypic alterations in the distal nephron after acute proximal tubular injury in the rat. Lab
Invest 70:479-493, 1994.
Paller MS: Effect of neut rophil depletion on ischemic renal injury
in the rat. J Lab Clin Med 113:379-386, 1989.
Safirstein R, Miller P, Dikman S, Lyman N, Shapiro C: Cisplatin
nephrotoxicity in rats: defect in papillary hypertonicity. Am J
Physiol 241.-F175-F185, 1981.
Safirstein R, Bonventre JV: Molecular response to ischemic and
nephrotoxic acute renal failure. In: Schlondorff D, Bonventre

JV (eds) Molecular nephrology: Kidney function in health and

disease. D. Marcel Dekker, New York, 1995, pp. 839-854.
Schumer M, Colombel MC, Sawczuk IS: Morphologic, biochemical, and molecular evidence of apoptosis during the reperfusion
phase after brief periods of renal ischemia. Am J Pathol 140:831 838, 1992.
Solez K, Finckh ES: Is there a correlation between morphologic
and functional changes in human acute renal failure? Data of
Finckh, Jeremy, and Whyte re-examined twenty years later. In:
Solez K, Whelton A (eds) Acute Renal Failure: Correlations
between Morphology and Function. Dekker New York, 1984.
Solomon R, Werner C, Mann D, D'Elia J, Silva P: Effects of saline,
mannitol, and furosemide to prevent acute decreases in renal
function induced by radiocontrast agents. N Engl J Med
331:1416-1420, 1994.
Thadhani R, Pascual M, Bonventre JV: Acute renal failure.
N Engl J Med 334:1448-1460, 1996.
van Bommel E, Bowry N, So K et al: Acute dialytic support for
the critically ill: Intermittent hemodialysis versus continuous
arteriovenous hemodiafiltration. Am J Nephrol 15:192-200,
1995.

APPROACH TO THE PATIENT WITH ACUTE

RENAL FAILURE
ROBERT D. TOTO

INTRODUCTION

Acute renal failure (ARF) is denned as an abrupt reduction in renal function sufficient to result in azotemia. Establishing the correct diagnosis and correctly assessing the
cardiovascular and volume status of the patient are the key
elements in patient management. As such, the approach
to the patient with ARF requires a careful history and
physical examination in combination with routine laboratory tests including complete blood count, blood biochemis tries urinalysis, and urine electrolytes. Renal imaging is
often useful, and in some cases, a renal biopsy is necessary.

Primer on Kidney Diseases, Second Edition

Fortunately, in most instances the primary cause of ARF

can be identified from the history, physical examination,
urinalysis, and renal sonogram.
ACUTE VS CHRONIC RENAL FAILURE

Distinguishing acute from chronic renal failure can be

difficult. A history of chronic symptoms of fatigue, anorexia, nocturia, pruritus, or restless legs may point to
chronic renal failure. The renal sonographic finding of small
contracted kidneys signals chronic renal failure; normalsize kidneys can be observed with either acute or chronic
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

254

Robert D. Toto

renal failure. Although rare, radiographic findings of renal

osteodystrophy are also helpful when present since hyperparathyroid changes occur only in chronic renal failure.
CAUSES OF ARF
ARF is often multifactorial, particularly in an ICU patient. For example, septic shock causing ischemia in combination with a nephrotoxic antibiotics or myoglobinuria is
not an uncommon combination in this setting. ARF is usually classified according to whether the renal circulation
(prerenal), the renal parenchyma (intrinsic renal) or the
urinary flow tract (postrenal) are altered as shown in
Table 1.

TABLE I

Causes of ARF
Prerenal (decreased renal blood flow)

Hypotension from any cause

Congestive heart failure
Volume depletion
Renal loss: Addisonian crisis, vomiting, DKA
Extrarenal loss: Vomiting, diarrhea, burns, excessive
sweating
Hypercalcemia
Prostaglandin inhibition (NSAIDs), cyclosporine, ACE
inhibitors, other
Intrarenal
Ischemia

Hypoperfusion from systemic hypotension of any cause

Sepsis sy ndrome
Atheroembolism
Thromboembolism: Renal artery thrombosis, aortic
thrombosis
Thrombotic microangiopathy from TTP, HUS/verotoxins,
drugs
Systemic necrotizing vasculitides (e.g., polyarteritis nodosa)
Toxins
Endogenous
Myoglobin, hemoglobin, uric acid, calcium-phosphate
complexes
Exogenous
Drugs including aminoglycosides, cephalosporins, acyclovir,
amphotericin
Chemotherapeutic agents, e.g., cisplatinum, methotrexate,
mitomycin C
Inflammation Acute glomerulonephritis, acute tubulointerstitial
nephritis
Tumor
Tumor infiltration, myeloma kidney
Postrenal (Obstruction) Prostate hypertrophy Intratubular
obstruction: Uric acid, calcium oxalate, acyclovir,
others Intra- and extraluminal: Stone, tumor, fibrosis, clot,
abscess
NSAIDS, nonsteroidal anti-inflammatory drugs; TTP, thrombotic thrombocytopenic purpura; ACE inhibitors, angiotensinconverting enzyme inhibitors

Prerenal Azotemia
In prerenal azotemia, renal blood flow is markedly decreased, usually because of intense, compensatory renal
afferent arteriolar vasoconstriction. Because renal plasma
flow rate is the main determinant of glomerular filtration
rate (GFR), filtration rate decreases markedly. Common
clinical situations in which this is the major or only mechanism of ARF include hypotension from any cause, severe
extracellular fluid volume depletion, hemorrhage, gastrointestinal losses, and third-space losses from burns, peritonitis, pancreatitis, or liver disease, as well as renal salt losses,
severe congestive heart failure, and hepatorenal syndrome.
Various drugs including nonsteroidal antiinflammatory
drugs, cyclosporine, osmotic diuretic agents (including glucose and mannitol), and angiotensin-converting enzyme
inhibitors can precipitate ARF when mild degrees of volume depletion or CHF are present. Vascular diseases may
also cause prerenal ARF by reducing glomerular blood
flow. Among the responsible disorders are renal artery
emboli, thrombotic occlusion of the aortorenal bifurcation,
and atheroembolic renal disease. Despite the fall in GFR,
renal tubular structure and function remain normal. Vasculitides and thrombotic thrombocytopenic purpura/hemo lytic uremic syndrome may also lead to ARF in this fashion,
but more commonly they also cause significant glomerular injury including endothelial damage, increased permeability to proteins, and inflammatory infiltration. In prerenal azotemia, increased reabsorption of sodium and
water with elaboration of a concentrated urine with a low
sodium concentration is typical. As urea reabsorption is
als o increased, a disproportionate increase in BUN relative to creatinine occur; hence the BUN/5cr ratio is often
> 20:1.
Intrarenal ARF
The most common cause of intrinsic renal failure is acute
tubular necrosis (ATN). This lesion is typically found after
ischemic or toxic insults. Tubular necrosis leads to sloughing of tubular cells into the lumina with partial or total
obstruction of nephron flow, thereby contributing to the
reduction in GFR. The urinalysis typically reveals granular
casts and renal tubular cells indicative of injured tubules.
Damaged tubules do not transport solutes normally, leading to elaboration of a dilute urine with a high sodium
concentration. Acute glomerulonephritis leads to ARF because of impaired ultrafiltration across the glomerular basement membrane. Damage to glomerlar capillaries leads to
proteinuria and hematuria; red blood cell casts may be
observed. Another common cause of intrinsic renal failure
is acute interstitial nephritis caused by drugs or other toxins
(see Chapter 40). In this situation interstitial inflammation
leads to tubular injury and necrosis. Pyuria, hematuria,
proteinuria with granular casts, and renal tubular epithelial
cells in the urine reflect the inflammation and tubular damage in the kidney. Also, elaboration of a dilute urine with
a high sodium concentration is typical.

255

35. Approach to the Patient with Acute Renal Failure

Postrenal ARF
Obstruction of the collecting system generally must involve both kidneys or a solitary kidney to cause significant
renal failure. Obstruction of the urinary tract at any level
may cause acute renal failure (see Chapter 54). Thus bladder outlet obstruction from prostate enlargement, tumor,
or urethral stricture; ureteral obstruction from tumor,
stone, papillae or fibrosis; or even massive crystal (uric
acid, acyclovir, calcium oxalate) deposition in the tubules
can cause acute obstruction leading to renal failure. Distension of the bladder may cause pain and a palpable mass.
Obstruction should be considered in patients with acute
anuria, particularly in those with a recent history of polyuria
alternating with oliguria.
HISTORY

Obtaining a careful history is essential to determining

the etiology of ARF. The history should focus initially in
two key areas: (1) factors known to lower renal perfusion
and (2) known and potential nephrotoxins. Attention
should be directed to recent events that could lead to alterations in cardiovascular and volume status and use of drugs/
toxins should be sought. For example, patients with volume
depleted states may present with orthostatic dizziness, presyncope, or syncope as their main problem. In this instance,
a careful history of factors that cause volume depletion
such as vomiting, diarrhea, excessive sweating, burns, and
renal salt wasting (e.g., diabetic ketoacidosis) should be
sought. In contrast, headaches caused by hypertension, dyspnea caused by pulmonary congestion, and peripheral
edema caused by extracellular fluid volume expansion are
common complaints in patients with congestive heart failure or renal salt retention due to glomerulonephritis
or other glomerular/vascular disorders (e.g., hemolyticuremic syndrome or atheroembolic renal disease) causing
ARF. A history of recent trauma with blood loss or muscle
trauma should raise the possibility of renal failure due to
ischemia and/or myoglobin-induced tubular necrosis. Fever, skin rash, and joint pains should raise the possibility
of a rheumatic disease such as SLE, a vasculitis, endocarditis, or a drug allergy causing intrinsic renal failure. A history
or dyspnea or pulmonary hemorrhage should raise the possibilities of Goodpasture's syndrome, Wegener's granulomatosis, Churg-Strauss vasculitis and pulmonary edema
due to volume overload from glomerulonephritis. Constitutional and non-specific symptoms including low grade fever,
malaise, myalgias, arthralgias, weakness and fatigue are
common in patients presenting with acute renal failure due
to rapidly progressive glomerulonephritis.
Both endogenous and exogenous toxins can give rise to
renal failure (Table 1). Therefore, a history of nephrotoxin
exposure is an extremely important component of the evaluation of a patient with ARF. A careful and thorough
review of the patient's history (and chart) for nephrotoxin
exposure is an extremely important component of the evaluation of a patient with ARF. Over-the-counter drugs,
including acetaminophen, antibiotics, antihypertensives,

and poisons, should be considered in all patients in which

the cause of ARF is inapparent. For example, occult inges tion of ethylene glycol can lead to renal failure with prominent metabolic acidosis (see Chapter 9). A patient may
survive the initial metabolic acidosis and present later with
renal failure, sometimes in association with cranial nerve
(especially VI) palsies or other neurologic effects.
Endogenous toxins include myoglobin, hemoglobin, and
calcium-phosphorus complexes. ARF may be the presenting finding in patients with nontraumatic rhabdomyolysis
from cocaine use or. infections, and hemoglobinuric renal
failure may be the initial presentation in a patient with an
acute intravascular hemolytic event. Occupational exposure history is also important because various heavy metals
can cause acute tubular necrosis. For example, exposure
to mercury vapor or liquid, lead, cadmium, or other heavy
metals can be encountered by welders and miners.
A history of recent infections is important. A wide variety
of infections can cause renal failure either by direct invasion
of the kidney or indirectly by activation of the immune
system in which the kidney is an innocent bystander. It is
important to review the patient's urine color, volume, and
pattern of output. Abrupt anuria suggests an acute obstruction, severe acute glomerulonephritis, or a sudden vascular
catastrophe such as renal artery emboli or atherosclerotic
occlusion of the aortorenal bifurcation. A gradually diminishing urine output may indicate a urethral stricture or, in
an older male, bladder outlet obstruction due to prostate
enlargement. Painless, gross hematuria in the setting of
ARF suggests acute glomerulonephritis, whereas painful
gross hematuria suggests obstruction of the ureter by tumor, blood clots, or necrotic renal papillae.
PHYSICAL EXAMINATION

The physical examination may provide many clues to

the underlying cause and potential therapy for ARF.
Skin
Examination of the skin for petechiae, purpura, and ecchymoses provides clues to inflammatory and vascular
causes of renal failure, including infectious diseases, thrombocytopenic purpura, and disseminated intravascular coagulation. Also, skin infarction from embolic phenomena and
cutaneous vasculitis presenting as palpable purpura occur
frequently in patients with septic shock, atheroembolic dis ease, systemic vasculitis , and infective endocarditis and
should be looked for carefully (Fig. 1).
Ocular Findings
Examination of the eye may reveal evidence of uveitis
(interstitial nephritis and necrotizing vasculitis), ocular
muscle paralysis (ethylene glycol poisoning and necrotizing
vasculitis), signs of severe hypertension, atheroembolic lesions (Hollenhorst crystals), and Roth spots (endocarditis)
Cytoid bodies (cotton wool exudates) may be seen in systemic lupus erythematosus (SLE). In addition, ophthal-

Robert D. Toto

256

balance and weights is helpful in establishing an accurate

estimate of volume status. Pulse rate and blood pressure
should be measured in the supine and standing (or seated
with legs dangling) positions. Close inspection of the neck
for jugular venous pulse level, careful examination of the
heart and lungs, and an assessment of peripheral edema
are essential.
The volume status of patients with prerenal azotemia
may vary substantially. Characteristic hypotension, low
urine output, and low urine sodium are observed both
with severe extracellular volume depletion and with severe
congestive heart failure. The decrease in arterial blood
volume in the former leads to findings similar to those
found with the decrease in effective arterial volume (despite extracellular fluid volume expansion) in the latter.
Importantly, the management of these two presentations
is completely opposite; patients with heart failure require
vasodilators, inotropes, and diuretics, whereas the volumedepleted patient requires saline, blood products, etc. Evidence for volume depletion, including orthostatic hypotension, dry mucous membranes, decreased skin turgor as
well as signs of sepsis, congestive heart failure, or cardiac
tamponade, should be sought in patients with low blood
pressure or overt hypotension. Often it is difficult to assess
the volume status from the examination alone, particularly
in elderly, obese, or severely edematous patients. In such
cases, placement of a central venous catheter or a pulmo nary artery catheter to measure right heart pressures, cardiac output, and systemic vascular resistance may be necessary.
Severe hypertension should raise the possibility of renal
failure due to malignant nephrosclerosis, glomerulonephritis, vasculitis, or atheroembolic disease. Cardiac murmurs
may be a sign of endocarditis, which can produce fulminant
glomerulonephritis with ARF. A pericardial friction rub in
a patient with ARF not only is an indication for emergency
hemodialysis but also may be a premonitory sign of impending cardiac tamponade. Hypotension due to tamponade can be temporarily stabilized by increasing cardiac
preload with a rapid intravenous fluid bolus.
Abdominal Examination

FIGURE I Cutaneous lesions in patients with (a) atheroembolic

disease (b) systemic necrotizing vasculitis; (c) cryoglobulinemia.

moplegia may be present in patients with systemic necrotizing vasculitis with renal failure.
Cardiovascular and Volume Status

A careful determination of the cardiovascular and volume status is the most important aspect in the diagnosis
and initial management of ARF. Evaluating daily intake
and output and body weight are extremely useful in estimating volume status. A table illustrating the trends in fluid

Abdominal examination may reveal signs of urinary obstruction (palpable bladder), tenderness in the upper quadrants (associated with ureteral obstruction or renal infarction), or ascites which may be associated with hepatic failure,
severe nephrotic syndrome, or Budd-Chiari syndrome. An
abdominal bruit should raise the possibility of severe atherosclerotic disease, which can engender renal failure from
a number of related disorders, including renal artery stenosis, thrombosis of the real artery origin, or atheroembolic
renal disease. Obstruction may be accompanied by a flank
mass due to tumor or retroperitoneal fibrosis.
Extremities

Examination of the extremities for signs of edema, tissue

ischemia, muscle tenderness from rhabdomyolysis, and ar-

257

35. Approach to the Patient with Acute Renal Failure

thritis (e.g., SLE, rheumatoid arthritis, relapsing polychondritis, infections) may provide clues to the diagnosis of
renal failure. Nail findings of hypoalbuminemia (paired
bands of pallor in the nailbedMuerhcke's lines) may be
a clue to underlying nephrotic syndrome which may rarely
predispose a patient to ATN.
Neuropsychiatric Abnormalities

Neuropsychiatric abnormalities are common in renal failure and range from signs of uremic encephalopathy (confu sion, somnolence, stupor, coma, seizures) to neurologic abnormalities associated with specific diseases. As mentioned
above, cranial nerve palsies can be seen in patients with ethylene glycol poisoning and vasculitides, including Wegener's
granulomatosis and polyarteritis nodosa. Altered and
changing mental status is common in both thrombotic microangiopathy and systemic atheroembolism.
DIAGNOSTIC TESTS
Serum Biochemistry Tests

Increases in BUN and creatinine are hallmarks of renal failure. Typically, the serum creatinine increases 1 to
2 mg dL" 1 day"1; however, increases of > 5mg dL" 1
day" 1 can occur in patients with rhabdomyolysis since muscle is a major source of creatine, the precursor of creatinine.
In cases of prerenal azotemia and in some patients with
obstructive uropathy, the BUN/Scr ratio is elevated above
20:1 because of enhanced reabsorption of urea. Also, in
cases where significant upper gastrointestinal bleeding occurs, the BUN/,Scr ratio may increase further as digested
protein from blood is absorbed and metabolized by the
liver.
Urinalysis

The urinalysis is the most important test in the initial

evaluation of ARF (see Chapter 3). An abnormal urinalysis
strongly suggests intrinsic renal failure. Gross color changes
in the urine may be seen with various intrinsic renal dis eases. The urine in acute tubular necrosis from ischemia
and toxins frequently appears "dirty" brown and opaque
owing to the presence of tubular casts. Reddish brown
urine or "cola-colored" urine is present in patients with
acute glomerulonephritis and is some times observed in
patients with pigment-associated tubular necrosis including
myoglobinuria and hemoglobinuria. Dipstick tests positive
for proteinuria and blood are particularly useful for diagnosing intrinsic renal diseases, including glomerulonephritides, acute interstitial nephritis, toxic and infectious causes
of tubular necrosis, and vascular diseases. Examination of
the spun urine sediment is extremely helpful for differentiating prerenal from intrarenal causes of renal failure. As
shown in Table 2, the urine sediment typically reveals granular "muddy" casts and renal tubular cells in ATN. Interstitial nephritis is often accompanied by pyuria, WBC casts,
and microhematuria. In allergic interstitial nephritis, eosinophiluria may be present on Wright and Hansel's stains

but eosinophiluria is nonspecific and can be seen in urinary

tract infections, glomerulonephritis, and atheroembolic renal disease as well. Glomerulonephritis is heralded by hematuria and RBC casts. In addition, granular casts, fat
globules, and oval fat bodies may be seen in glomerulopathies associated with heavy proteinuria. Uric acid crystals
may be present in ATN associated with acute uric acid
nephropathy, calcium oxalate crystals may be present in
ethylene glycol poisoning with acute renal failure due to
nephrocalcinosis, and acetaminophen crystals may be observed in ATN from acute poisoning with this drug. Some
of the ordinary findings in ATN are shown in Chapter 3,
Figs. ID, 2B, and 2D.
Urine Electrolytes

In a patient with ARF, urine electrolytes can serve as

an indicator of the functional integrity of the renal tubules.
The determination is most helpful when performed in the
setting of hypotension and oliguric renal failure. The single
most informative test is the fractional excretion of sodium
(FENa). FENa is defined as:
((Urine [Na]/Plasma [Na])/
(Urine [Cr] / Plasma [Cr])) X 100.
In patients with prerenal azotemia, the FENa is usually
< 1% and in ATN it is usually >1%. However, there are
exceptions. For example, some patients with ATN caused
by severe burns, radiocontrast nephropathy, and underlying liver disease have a FE^a < 1%. This is probably due
to severe renal vasoconstriction with low tubular flow rate
and heterogeneous damage to the tubules. In addition,
FENa is < 1% in some patients with glomerulonephritis as
tubular function remains intact with increased, rather than
decreased proximal tubular sodium reabsorption. Diuretic
administration may confuse the interpretation of the FENa.
For example, a patient with prerenal azotemia due to administration of a loop diuretic may have a FENa > 1%.
Therefore, the FENa should never be relied on as the sole
means of determining the cause of ARF. Note also that if
only a small amount of urine is available, the supernatant
remaining after the spun sediment is examined may be sent
for Urine [Na] and Urine [Cr] to permit calculation of
the FENa.
Assessing Urine Output

Knowledge of the urine output is important not only

from the standpoint of etiology but also and most importantly from the standpoint of management and outcome.
Oliguria is defined as a 24-hour urine output between 100
and 400 mL/day. Anuria is defined as a urine output
^ 100 mL/day. Nonoliguric renal failure occurs with a urine
output > 400 mL/day. In patients with suprapubic discomfort and an obviously distended bladder or in patients with
a history of declining urine output or documented oliguria,
a bladder catheter should be placed temporarily to relieve
or rule out bladder outlet obstruction. Patients with oliguric
renal failure are easier to manage and have a lower overall

Robert D. Toto

258
TABLE 2
Differential Diagnosis of ARF: Urinalysis, Water,

and Sodium Metabolism

Water metabolism

Diagnosis
Prerenal
Intrarenal
Tubular necrosis
Interstitial nephritis
Glomerulonephritis

Urinalysis
Normal, or hyaline casts
Granular and epithelial cell casts Pyuria,
hematuria, mild proteinuria, granular and epithelial
cell casts, eosinophils Hematuria, marked
proteinuria, RBC casts, granular casts Normal or
hematuria, mild proteinuria

Vascular disorders
Postrenal

Normal or hematuria, granular casts, pyuria

Sodium handling
Na
<20 mEq/L

FENa
<1%

<40 si

>20 mEq/L
>20 mEq/L

sl%
>1%

>1

<40

>20 mEq/L

<1%

>1

<40

>20 mEq/L

<1%

<40

>20 mEq/L

>1%

U/P Osm

U/P Cr

>1

>40

<1
<40

si

U, urine; P, plasma; Osm, osmolality; Cr, creatinine; Na, sodium; FEN a, fractional excretion of sodium.

mortality rate than oliguric or anuric patients (see discussions on management and survival in accompanying sections). Measurement of subsequent daily urine output is
an important feature of management of patients (see Chapter 41).
Fluid Challenge
In patients with suspected prerenal azotemia from significant intravascular volume depletion, an intravenous infusion of normal saline, colloid (such as albumin or dextran), or blood products may be helpful. If volume
depletion is present, the fluid challenge should improve
renal blood flow and result in increased urine output and
correction of renal failure. This maneuver usually consists
of an infusion of 1 to 2 L of normal saline administered over
a 2 to 4-hour period depending on the clinical judgment of
the treating physician. Close and careful bedside monitoring of vital signs, physical examination, and urine output
are essential. Failure of this maneuver to improve the vital
signs and urine output can help to point to intra- or postrenal causes of renal failure. Caution must be exercised during fluid challenge because of the potential for producing
pulmonary edema in patients with congestive heart failure
or intrinsic failure that do not respond to volume expansion.
Renal Imaging
Ultrasound
Renal ultrasonography is particularly useful for evaluating the urinary collecting system, including the calyces,
pelvis, and bladder, for obstruction. The test is readily
available, noninvasive, accurate, reliable, and reproducible.
Technically it is easy to perform in most patients; however,
difficulty visualizing the kidney does occur in obese patents,
in patients with distention from abdominal gas, and in some
patients with massive ascites and retroperitoneal fluid collections. Typically, obstructive uropathy is manifest by dilatation of the collecting system and ureters (see Chapter 6,
Fig. 1, and Chapter 54, Fig. 2). However, in some cases
of early obstruction or ureteral encasement by tumor or

fibrosis, ureteral and renal pelvis dilatation may not be

detected by ultrasound. The parenchymal pattern on ultrasound may be abnormal with increased echogenicity in the
presence of intrinsic renal disease. However, this finding
is nonspecific and cannot be used to determine the etiology
of intrinsic renal failure. Moreover, in most cases of biopsyproven ATN, the renal parenchymal pattern on ultrasound
appears completely normal.
Doppler Scans
Doppler scans are used to evaluate renal parenchymal
blood flow. However, because reduced renal blood flow
occurs in both prerenal and intrarenal causes of renal failure, this test is of little diagnostic value in patients with
ARF. One exception is the patient with thrombotic obliteration of the renal circulation, as the absence of flow to the
kidneys can easily be discerned by this technique.
Nuclear Scans
Radionuclide imaging with 99mTc-DTPA or 131I-hippuran
can be used to assess renal blood flow and tubular function
in ARF. However, the utility of such testing is similar to that
of Doppler flow studies. Unfortunately, a marked delay in
tubular excretion of radionuclide occurs in both prerenal
and intrarenal diseases, and this technique is therefore of
little value in most patients with ARF.
Magnetic Resonance Imaging
MRI may have some utility in diagnosing intrinsic renal
failure. For example, increase in TVweighted signal may
occur in patients with inflammatory renal disease such as
acute tubulointerstitial nephritis and glomerulonephritis.
Whether it is cost effective in the setting of ARF remains
to be determined.
Renal Angiography
A renal angiogram is helpful in patients with ARF due
to vascular disorders including renal artery stenosis with
ARF from angiotensin-converting enzyme inhibition, renal
artery emboli, aortic atherosclerosis with acute aortorenal
occlusion, and in cases of systemic necrotizing vasculitis such

35. Approach to the Patient with Acute Renal Failure

259
a therapeutic perspective. For instance, patients presenting
with the clinical syndrome of rapidly progressive glomerulonephritis should undergo renal biopsy unless there is
an overt contraindication since effective renal preserving
therapy is available and should be instituted as soon as
possible. The biopsy is also particularly useful in cases of
suspected interstitial nephritis from drugs. For example, in
a febrile patient with endocarditis receiving broad spectrum
antibiotic coverage, it may be impossible to distinguish
between glomerulonephritis and acute tubulointerstitial
nephritis on clinical grounds alone. The empiric use of
glucocorticoids to treat suspected allergic interstitial nephritis could be considered without a biopsy, but their risk
would be high in this setting. In most instances, a biopsy
can be carried out safely despite ARF. Clinical judgment is
important in determining candidates for biopsy and routine
precautions should always be taken. Imaging with sonogram, CT, or MRI reduce the risk of biopsy complications
in patients who are acutely ill.
Bibliography

FIGURE 2 Renal angiogram in a patient with acute renal failure

caused by systemic nccrotizing vasculitis. The early phase of this
selective injection of the left renal artery shows marked attenuation
of cortical blood flow and multiple aneurysms of medium-sized
arcuate arteries.

as polyarteritis nodosa and Takayasu's arteritis (Fig. 2).

The latter can affect the distal aorta and present with advanced renal failure.
Renal Biopsy

Renal biopsy is the only diagnostic test that can establish

the cause of intrinsic renal failure with absolute certainty.
Furthermore, several studies have shown that biopsy in the
setting of acute renal failure often (up to 40% of the time)
reveals unexpected findings. Therefore, renal biopsy should
be considered in any patient with ARF in whom the etiology is unknown and histologic diagnosis is important from

Abuelo JG: Diagnosing vascular causes of renal failure. Ann Intern

Med 123: 601-614, 1995. Alkhunaizi AM, Schrier RW:
Management of acute renal failure:
New perspectives. Am J Kidney Dis 28: 315-328, 1996.
Barrett BJ: Contrast nephrotoxicity. J Am Soc Nephrol 5: 125137, 1994. Davda RK, Guzman NJ: Acute renal failure. Prompt
diagnosis is
key to effective management. Postgrad Med 96: 89-92, 1994.
Fogazzi GB: Crystalluria: A neglected aspect of urinary sediment
analysis. Nephrol Dial Transplant 11: 379-387, 1996. Hock R,
Anderson R: Prevention of drug- induced nephrotoxicity
in the intensive care unit. J Crit Care 10: 33-43, 1995. Hou
SH, Bushinsky DA, Wish JB, Cohen JJ, Harrington JT:
Hospital-acquired renal insufficiency: A prospective study.
Am J Med 74:243-8, 1983.
Liano F, Pascual J: Epidemiology of acute renal failure: A prospective, multicenter community-based study. Madrid acute renal
failure study group. Kidney Int 50:811-8, 1996. Pascual J,
Liano F, Ortuno J: The elderly patient with acute renal
failure. / Am Soc Nephrol 6:144-153, 1995. Porter GA:
Contrast -associated nephropathy: Presentation, patho physiology and management. Miner Electrolyte Metab 20: 232243, 1994.

RENAL FAILURE CAUSED BY

THERAPEUTIC AGENTS
THOMAS M. COFFMAN

Compounds used for diagnostic and therapeutic purposes are common causes of renal insufficiency. The kidney
is a frequent target for toxic injury because it is a major
route of excretion for a variety of drugs. As a part of the
excretory process, these materials may be greatly concentrated in the urinary space and within renal tubular cells,
enhancing their potential to cause local toxicity. Also, the
rate of blood flow per gram of tissue weight in the kidney
is relatively high, resulting in exaggerated exposure of renal
endothelial cells and glomeruli to circulating substances.
Since most renal functions are dependent on tightly regulated blood flow patterns, agents that imp air these hemodynamic relationships may interfere with the ability of the
kidney to maintain normal homeostasis.
Drug toxicity in the kidney is manifested through the
same clinical syndromes that are associated with kidney
diseases of other causes. As depicted in Table 1, these
include acute and chronic renal failure, and nephrotic syndrome. Moreover, a single agent may have more than one
effect on kidney functions. The particular clinical manifestation of nephrotoxicity is determined by the dose and
duration of exposure, chemical properties of the agent, as
well as factors within the individual patient such as age,
volume status, and genetic background. This chapter describes a general approach to nephrotoxicity and, reviews
the renal effects of some common causative agents. More
detailed discussions of individual agents or syndromes associated with toxic renal injury can be found in other chapters
(see Chapters 31, 40, 42, and 49).

sensitive and specific than serum creatinine, none of these

have yet found widespread clinical application.
The clinical syndromes caused by drugs mimic those
associated with kidney diseases of other causes. Thus, when
the etiology of renal failure is being investigated, the possible role of therapeutic agents should always be considered
and a detailed medication history is an extremely important
part of the clinical evaluation. Temporal associations between the appearance of a kidney abnormality and medication changes must be documented. In patients who are
receiving compounds known to be nephrotoxic, the plan
for clinical management should include careful monitoring
of renal function, avoidance of clinical risk factors and, in
some cases, monitoring of serum drug levels. Also, the ro le
of nephrotoxins in exacerbating renal failure from other
causes should be considered. For example, in the hospitalized patient with acute tubular necrosis, the potentially
additive detrimental effects of aminoglycosides, radiocontrast, or other nephro toxins must be recognized. Moreover,
renal clearance of some drugs will be substantially reduced
in such patients and doses must be lowered appropriately
(see Chapter 43).
Acute Renal Failure from Therapeutic Agents
In approaching any patient with acute kidney dysfunction, a potential causative role for therapeutic agents should
always be considered. As illustrated in Table 1, mechanisms
of acute renal failure related to drugs can be roughly categorized as prerenal/hemodynamic, intrarenal, or postrenal/
obstructive syndromes. As described in Chapter 35, this
separation can be extremely helpful in identifying the etiology and directing management of patients with acute renal
failure (ARF) from any cause.

DIAGNOSIS

The possibility of drug-induced nephrotoxicity should be

considered when the serum creatinine concentration rises
during administration of a therapeutic agent. Because of
the nonlinear relationship between serum creatinine and
glomerular filtration rate (GFR), a substantial reduction
in GFR is necessary before toxic injury can be appreciated
clinically. This point is particularly important to consider
in the context of agents that may cause chronic nephropathy. In this case, kidney injury may not be detected until
40 to 50% of kidney function has been irreversibly lost.
While a number of potential diagnostic markers, such as
urinary excretion of various tubular enzymes, have been
evaluated as indicators of renal toxicity that might be more

Prerenal Azotemia: Hemodynamically Mediated Renal

Insufficiency Associated with Drugs
As shown in Table 1, several classes of therapeutic agents
including cyclosporine, tacrolimus, radiocontrast, nonsteroidal anti-inflammatory drugs (NSAIDS), and angiotensin
converting enzyme (ACE) inhibitors can cause a syndrome
of abnormal kidney function that resembles prerenal azotemia. Similar to prerenal azotemia from other causes, hemo dynamic renal dysfunction caused by drugs can be associated with low urine sodium excretion. While such renal
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

Primer on Kidney Diseases, Second Edition

26
0

36. Renal Failure Caused by Therapeutic Agents

TABLE I

Renal Syndromes Caused by Therapeutic Agents

Clinical syndrome Causative agents
Acute renal failure
Prerenal/hemodynamic Cyclosporine, tacrolimus,
radiocontrast, amphotericin B, ACE
inhibitors, NSAIDs
Intrarenal Acute tubular
Aminoglycosides, amphotericin B,
necrosis
cisplatin, certain cephalosporins
Penicillins, cephalosporins,
Acute interstitial
sulfonamides, rifampin, NSAIDs,
nephritis
interferon, interleukin -2
Acyclovir, analgesic abuse,
Postrenal/obstructive methylsergide
Lithium, analgesic abuse,
cyclosporine, tacrolimus, cisplatin,
Chronic renal failure
nitrosoureas
Nephrotic syndrome

Gold, NSAIDs, penicillamine,

captopril, interferon

Note: This is a representative, but not

exhaustive list of etiologi-cal agents. Please refer to individual
Primer chapters for more complete listings.
ACE inhibitors, angiotensin converting enzyme inhibitors;
NSAIDs, nonsteroidal anti-inflammatory drugs.

dysfunction is usually reversible when the offending agent

is discontinued, ischemic damage may result if the insult
is prolonged or particularly severe.
Drugs can cause hemodynamically mediated renal dysfunction through several mechanisms. Agents such as
cyclosporine, tacrolimus, radiocontrast, and amphothericin
cause intense vasoconstriction in the kidney that reduces
renal blood flow and glomerular perfusion. These compounds do not seem to affect vascular tone directly but
may stimulate production of other vasoconstrictors such
as endothelin or thromboxane Aa. Cyclosporine, tacrolimus, and amphotericin can produce renal insufficiency in
normal subjects who have no underlying renal circula tory abnormalities.
In contrast, hemodynamic kidney dysfunction associated
with NSAIDs generally occurs in patients with preexisting
compromise of renal perfusion. NSAIDs inhibit the
cyclooxygenase isoenzymes (also called PGH synthases 1
and 2), turning off the synthesis of prostaglandins. In normal subjects, renal prostaglandin production is low and
administration of NSAIDs has very little effect on renal
function. However, as an adaptive mechanism, production
of vasodilator prostaglandins increases in situations in
which renal perfusion is threatened. Inhibiting production
of these vasodilator compounds by NSAIDs can cause precipitous declines in renal blood flow and GFR. This syndrome is most often seen in patients with volume depletion, heart failure, and preexisting renal disease (see
Chapter 42).
Similarly, acute renal failure following administration of
ACE inhibitors usually occurs in patients with underlying

261
abnormalities of the renal vasculature and circulation. This
syndrome is most commonly seen in patients with conges tive heart failure on diuretics, in patients with severe bilateral renal artery stenosis, patients with critical renal artery
stenosis in a single functioning kidney, and in patients with
vascular disease and nephrosclerosis. ACE inhibitors lower
blood pressure by inhibiting the conversion of angiotensin I
to angiotensin II. Angiotensin II is a potent vasoconstrictor
that acts to increase peripheral resistance. It also induces
preferential constriction of efferent arterioles within the
glomerulus helping to maintain GFR when renal blood
flow is compromised. In the clinical settings described
above, ACE inhibitors cause ARF by reducing systemic
blood pressure while simultaneously reducing transglomerular pressure due to the fall in postglomerular, efferent
arteriolar resistance. As with other forms of drug-induced
hemodynamic renal insufficiency, kidney function usually
returns to baseline when the ACE inhibitor is discontinued.
Based on their pharmacology, the new type 1 (ATI) angiotensin receptor blockers would be expected to have similar
effects in susceptible patients.
Intrarenal ARF: Acute Tubular Necrosis and Acute
Interstitial Nephritis Caused by Drugs

Drug-induced acute renal failure from intrarenal mechanisms can be divided into two entities with distinct clinical
and pathophysiologic characteristics: acute tubular necrosis
(ATN) and acute interstitial nephritis (AIN). ATN is associated with drug administration shares many of the clinical
features of ATN from other causes. This form of ARF
can be seen following administration of agents that are
primarily excreted by the kidney, such as aminoglycoside
antibiotics, amphotericin B, and chemotherapeutic agents
such as cisplatin. Nephrotoxicity often results from direct
toxic effects of the compound on renal tubular cells although other hemodynamic mechanisms may play a role.
In this setting, the onset of ARF is often nonoliguric and
may be slow to develop. If nephrotoxicity is not detected
and administration of the causative agent is continued,
oliguric ARF may develop. The urinalysis is characteristically bland and may show modest proteinuria, tubular epithelial cells, and noncellular casts. Generally, tubular toxicity will abate when the offending agent is discontinued
although there may be a lag before complete recovery
of renal function occurs. An exception to this may occur
following repetitive exposure to tubular toxins that can
result in chronic, irreversible renal impairment.
In AIN, drug exposure causes ARF through a syndrome
of intrarenal inflammation. This disorder is described in
detail in Chapter 40 and is characterized by inflammatory
cell infiltration of the renal interstitium with reduced GFR
and renal blood flow. Systemic signs of hypersensitivity,
including rash, arthralgias, and fever, may also occur. The
urinalysis reflects active renal inflammation and usually
contains red cells, white cells, and occasional cellular casts
with nonglomerular levels of proteinuria. Eosinophiluria
can also be observed but is not pathognomonic. Common
causative agents include penicillins, cephalosporins, sulfonamide analogs, rifampin, and NSAIDs. AIN usually resolves after the offending agent is removed, although some

262

Thomas M. Coffman

have recommended a limited course of corticosteroids to

hasten recovery.
Obstructive Nephropathy Associated with
Therapeutic Agents

Drug-associated obstruction can occur at several anatomic sites: intratubular, intraureteral, and extrinsic ureteral
obstruction from retroperitoneal fibrosis. These obstructive
syndromes have been associated with specific causative
agents. For example, the antiviral agent acyclovir can cause
ARF due to the precipitation of the drug, which is relatively
insoluble, within renal tubular lumens. In analgesicassociated nephropathy (discussed in Chapter 42), patients
may present with symptoms of acute ureteral obstruction
due to sloughing of necrotic renal papillary tissue. Methylsergide has been associated with retroperitoneal fibrosis
causing obstructive nephropathy. In one survey, however,
drug-induced retroperitoneal fibrosis made up less than 3%
of cases of patients with this unusual syndrome.
Chronic Renal Failure from Therapeutic Agents

Chronic renal failure caused by drugs is usually manifested as a chronic tubulointerstitial process. This syndrome
of chronic interstitial nephropathy has been associated with
a number of structurally diverse agents, including lithium,
analgesics, cyclosporine, cisplatin, and nitrosureas. Chronic
interstitial disease caused by a drug most often presents
as an elevation in serum creatinine, which may be slowly
progressive. However, abnormalities of renal tubular function may be a predominant feature. Such abnormalities
include renal tubular acidosis, concentrating defects, defective potassium secretion, and tubular proteinuria. On histologic examination, interstitial fibrosis, tubular atrophy, and
infiltration of the renal interstitium with chronic inflammatory cells are observed. The urinalysis may contain white
cells and red cells with modest levels of proteinuria. While
patients with drug-induced chronic interstitial nephropathy
may progress to end stage renal disease requiring renal
replacement therapy, the course of the disease can usually
be stabilized or reversed if the offending agent is identified
and discontinued.
Nephrotic Syndrome Associated with Therapeutic Agents

Glomerulopathy with proteinuria may be caused by several drugs, including gold, penicillamine, and NSAIDs. Affected patients often present with proteinuria, edema, and
hypoalbuminemia. Pathologically, membranous nephropathy has been associated with all of the agents listed above,
whereas minimal change nephropathy has been seen in
patients taking certain NSAIDs and penicillamine. In most
cases, proteinuria remits when the agent is discontinued.
However, in a few cases, renal injury has progressed after
the drug is stopped.
SPECIFIC AGENTS WHICH CAUSE RENAL FAILURE
Antibiotics

As a class of drugs, antibiotics are the most common

cause of clinically recognized drug-induced renal failure.

Within this group, aminoglycosides are responsible for the

majority of episodes of nephrotoxicity in hospitalized patients. Aminoglycosides most commonly cause an ATN
picture, while other antibiotics such as penicillins, rifampin,
and sulfonamide more commonly produce AIN. In practice, antibiotics are often administered to patients who are
severely ill with other coexistent processes that can independently affect renal function or may aggravate and potentiate nephrotoxicity. Thus, in an individual patient, a
specific causative role for an antibiotic may be difficult
to discern.
Aminoglycosides

Aminoglycosides are amphophilic, cationic antibiotics

that are used to treat serious gram-negative bacterial infections. They are by far the most common cause of antibioticassociated renal insufficiency in hospitalized patients. Depending on the criteria used for defining nephrotoxicity,
the reported incidence of nephrotoxicity ranges between
7 and 36% of patients receiving aminoglycosides. Serum
protein binding of aminoglycosides is minimal, they are
freely filtered at the glomerulus, and renal excretion is the
major route of elimination. Aminoglycosides accumulate
within the renal cortex reaching saturation within the first
3 days of treatment. Accumulation of drug within cortical
tubular cells probably causes toxicity, although the specific
cellular mechanisms of aminoglycoside-induced ARF have
not been completely defined. Evidence of tubular cell abnormalities and injury may be seen by both light and electron microscopy.
The usual clinical presentation of aminoglycoside nephrotoxicity is a rising BUN and creatinine that typically
appears 5 to 7 days into the antibiotic course. Renal insufficiency may occur earlier in the presence of risk factors.
Frank azotemia may be preceded by the development of
a concentrating defect manifested as polyuria and the urinalysis most commonly shows modest proteinuria with
noncellular casts and occasional tubular epithelial cells.
Characteristically, renal function deteriorates progressively
but the process is reversible if the diagnosis is suspected
and the aminoglycoside is discontinued. However, there
may be a lag time before renal function begins to improve.
This is probably related to the kinetics of accumulation
of aminoglycosides in renal cortical tissue, since urinary
excretion of aminoglycosides has been detected for days
to weeks after administration is discontinued.
While virtually every patient who receives aminoglycosides is at some risk of developing renal toxicity, there is
a positive association between the dose and duration of
therapy and the risk of developing renal failure. Tailoring
aminoglycoside doses to maintain drug levels within a defined therapeutic range serves to minimize the risks of
toxicity while maintaining bactericidal concentrations of
antibiotic. Single daily dose regimens have been advocated
by some authors as an approach to avoid aminoglycoside
nephrotoxicity, but the benefits of these regimens have not
been clearly demonstrated. Because aminoglycosides are
primarily excreted by the kidney, increased aminoglycoside
levels can be both a cause and a marker of nephrotoxicity.
Monitoring peak and trough drug levels along with serum

36. Renal Failure Caused by Therapeutic Agents

263

creatinine every 2 to 3 days is prudent, but daily monitoring

may be required in the unstable patient with a serious
infection and fluctuating level of renal function.
Several risk factors for aminoglycoside nephrotoxicity
have been identified and are illustrated in Table 2. When
aminoglycoside therapy is being initiated, these characteris tics may be used to identify patients at high risk of developing toxicity for more intensive monitoring and to modify
factors such as volume status and electrolyte abnormalities.
When possible, alternative antibiotic choices might also
be considered in high-risk patients. All members of the
aminoglycoside family can potentially cause nephrotoxicity. While gentamicin is the most extensively utilized, tobramycin has a similar bactericidal profile and exhibits less
nephrotoxicity, at least in animal models. Amikacin probably has an intermediate potential for nephrotoxicity between gentamicin and tobramycin.

striction with a hemodynamically mediated reduction in

GFR. Second, amphotericin is highly bound to cell membranes and causes damage that affects membrane integrity
and permeability. In the kidney, this membrane injury is
thought to be the basis for the potassium and magnesium
wasting, inability to maximally concentrate urine, and distal
tubule acidification defects that are characteristic of amphotericin nephrotoxicity. These abnormalities, along with
an abnormal urine sediment, usually precede the development of clinically apparent azotemia. Renal failure is usually nonoliguric and progressive, but will slowly abate when
the amphotericin is discontinued. However, high doses and
repetitive exposure to amphotericin can cause permanent
kidney damage and chronic renal failure. Volume depletion
potentiates nephrotoxicity and it has been suggested that
sodium loading and volume expansion may prevent or ameliorate renal injury.

Cephalosporins

Acyclovir

Cephalosporins are semisynthetic /3-lactam derivatives

that have broad spectrum bactericidal activity. While they
are generally tolerated well by patients, renal insufficiency
is an infrequent but well-defined complication of cephalosporin therapy. Two forms of renal failure have been described with Cephalosporins: ATN and AIN. A profile of
tubular toxicity has been best documented with cephaloridine and cephalothin especially when higher doses are used.
A rank order potential for Cephalosporins to produce proximal tubular toxicity has been defined in animal studies
as cephaloglycin > cephaloridine > > cefaclor > cephazolin > cephalothin > cephalexin and ceftazidime. Combination therapy with aminoglycosides or furosemide may
increase the risk for cephalosporin-associated ATN.

Acyclovir is an effective and relatively nontoxic antiviral

agent that is widely used to treat herpes virus infections.
When given by the oral route, acyclovir is essentially devoid
of significant renal toxicity. However, nephrotoxicity has
been described in a small number of patients who have
received intravenous courses of acyclovir, particularly at
high doses (>500 mg/m2). Acyclovir undergoes tubular
secretion in the kidney and renal tissue levels increase
substantially during treatment. The mechanism of ARF is
thought to be precipitation of the relatively insoluble drug
within tubular lumens causing obstructive nephropathy.
The urine sediment may contain red cells and white cells
with needle-shaped birefringent crystals. Renal failure generally resolves when the acyclovir is discontinued. Risk
factors for toxicity are volume depletion and bolus adminis tration of drug. However, ARF has been observed despite
adequate fluid repletion and the use of continuous infusion protocols.

Amphotericin B

Amphotericin B is a polyene antibiotic that is the treatment of choice for the majority of serious fungal infections.
Unfortunately, this agent produces a number of side effects,
with nephrotoxicity being the most clinically problematic.
In some series, the degree of nephrotoxicity is roughly
proportional to the total cumulative dose received. At least
two mechanisms mediate the adverse effects of amphotericin in the kidney. First, the drug causes acute renal vasocon-

TABLE 2

Risk Factors for the Development of

Aminoglycoside Nephrotoxicity

Prolonged course of treatment (>10 days)

Volume depletion
Sepsis
Preexisting renal disease
Hypokalemia
Elderly patient
Combination therapy with certain Cephalosporins (particularly
cephalothin) Concomitant exposure to other
nephrotoxins (i.e.,
radiocontrast, amphotericin B, cisplatinum)
Gentamicin > amikacin > tobramycin

Vancomycin

Vancomycin is a glycopeptide antibiotic developed in

the 1950s for use against penicillin-resistant gram-positive
infections. Recently, usage of vancomycin has increased
because of changes in resistance profiles as well as its favorable pharmacokinetics in patients on dialysis. Vancomycin
is also used frequently to treat methicillin-resistant Staphylococcus aureus infections and as primary treatment for
pseudomembranous colitis caused by Clostridium difficile.
In the early years of its use, azotemia was reported in up
to 25% of patients who received this agent. However, this
was probably related to impurities present in the early
commercial preparations of the drug. Recent series suggest
that the incidence of nephrotoxicity is <7% with modern
vancomycin preparations. Clinical and animal studies suggest that coadministration of vancomycin with aminoglycosides may enhance the risk of developing nephrotoxicity.
Pentamidine

Approximately 25% of patients treated with pentamidine

may experience a fall in GFR that is reversible when the
drug is discontinued. Nephrotoxicity with pentamidine may

264

Thomas M. Coffman

be more common in AIDS patients and has been associated

with significant hyperkalemia. Although the incidence of
renal problems is reduced with inhaled preparations, renal
insufficiency has been reported in association with aerosolized pentamidine use. However, a specific role for pentamidine is often difficult to identify due to the presence of
other drugs or comorbid conditions that might als o affect
renal function.
Radiocontrast
The administration of radiocontrast is a relatively common cause of ARF in hospitalized patients. However, the
reported incidence of contrast nephropathy from published
studies is quite variable. This variation relates to differences
in criteria for defining the syndrome, the period of observation after the contrast administration, and the prevalence
of risk factors in the population studied. Preexisting renal
insufficiency is the most important and best-documented
risk factor and contrast-induced nephrotoxicity is rare in
patients with normal renal function. Other potential risk
factors are listed in Table 3. While animal studies suggested
that low osmolality, nonionic contrast agents are less nephrotoxic than conventional high osmolality, ionic agents,
prospective clinical studies have failed to demonstrate differences in nephrotoxicity in patients treated with ionic vs
nonionic contrast.
The pathogenesis of radiocontrast nephropathy results
from the vasoactive effects of radiocontrast. In animals,
contrast injection initially causes vasodilatation of the renal
circulation, followed by intense and persistent vasoconstriction. The etiology of this vasoconstrictive phase is not clear,
but may include reduced production of vasodilator prostaglandins or changes in intracellular calcium. Patients with
contrast nephrotoxicity typically develop a rise in their
serum creatinine within 24 hours after the radiocontrast
study, sometimes associated with oliguria. The urinary sediment is unremarkable and the fractional excretion of sodium is typically very low, consistent with the hemodynamic
etiology of renal impairment.
Renal impairment is usually transient, although occasional patients will require support with acute dialysis. The
clinician's efforts should be directed toward prevention

TABLE 3

Risk Factors for the Development of Acute Renal Failure

Following Radiocontrast Administration
Preexisting renal dysfunction
Diabetic nephropathy Severe
congestive heart failure Volume
depletion Elderly patient Multiple
myeloma Large volumes of
radiocontrast
Concomitant treatment with ACE inhibitors, NSAIDs, or
exposure to other nephrotoxins
ACE inhibitors, angiotensin converting enzyme inhibitors;
NSAIDs, nonsteroidal anti-inflammatory drugs.

of contrast nephropathy by avoiding unnecessary studies,

particularly in patients with risk factors. If contrast adminis tration is unavoidable, the high-risk patient should be given
0.45% saline intravenously at a rate of 1 mL per kilogram
of body weight per hour beginning 12 hours before the
procedure and continued for an additional 12 hours afterward. This regimen was found to provide better protection
against acute decreases in renal function induced by radiocontrast than hydration plus mannitol or furosemide, which
in the past had been suggested to be helpful in this setting.
Concomitant administration of other nephrotoxic agents
should be avoided and the amount of contrast used during
the study should be kept to a minimum. As noted above,
low osmolality, nonionic agents exert no clear benefit in
preventing nephrotoxicity.
Immunosuppressive Drugs
Cyclosporine
Cyclosporine is an unusual cyclic peptide compound that
inhibits early events involved in T cell activation and is
extremely effective in suppressing transplant rejection.
Since its introduction in 1983, Cyclosporine has become
the primary component of immunosuppressive regimens
administered to patients with virtually every type of organ
graft. In addition, Cyclosporine has demonstrated efficacy
in the treatment of autoimmune disorders such as psoriasis and uveitis. However, the clinical applications of
Cyclosporine have been limited by the frequent occurrence
of nephrotoxicity and especially concern over the potential
for developing chronic, irreversible renal injury.
Cyclosporine nephrotoxicity can be manifested by three
distinct clinical syndromes: acute reversible renal dysfunction, chronic interstitial nephropathy, and thrombotic microangiopathy. Acute cyclosporine nephrotoxic ity is the
predominant renal abnormality seen within the first 6 to
12 months after initiating treatment and is characterized
by an acute or subacute reduction in renal function that
is often dose dependent. Generally, renal dysfunction is
nonprogressive and reverses when the dose is lowered or
the drug is discontinued. Virtually every patient who receives therapeutic doses of cyclosporine will experience a
component of persistent, reversible reduction in GFR and
renal blood flow. The mechanism of acute cyclosporine
nephrotoxicity is hemodynamic and results from the ability
of cyclosporine to induce intense renal vasoconstriction.
NSAIDs may enhance acute effects of cyclosporine on
renal hemodynamics. Cyclosporine does not cause renal
vasoconstriction directly, but may act through stimulating
production of other vasoconstrictor compounds such as
thromboxane A 2, endothelin, or leukotrienes.
In renal transplant recipients within the first year after
transplant, it is often difficult to clinically differentiate acute
cyclosporine nephrotoxicity from acute rejection. A stable
or slowly progressive increase in serum creatinine that reverses when the cyclosporine dose is reduced suggests
nephrotoxicity. Renal biopsy can be helpful in this setting
since aggressive inflammatory cell infiltrates are usually
absent in acute cyclosporine nephrotoxicity and their pres ence in a biopsy specimen would suggest ongoing rejection.

36. Renal Failure Caused by Therapeutic Agents

In addition, prominent vacuolization of proximal tubular
epithelial cells has been described in acute cyclosporine
toxicity. While serum cyclosporine levels are frequently
used to monitor therapeutic efficacy and to prevent toxicity,
there is only a rough correlation between serum levels and
clinical events.
Chronic cyclosporine nephrotoxicity is defined by the
development of interstitial fibrosis with reduced levels of
GFR in patients receiving long-term cyclosporine treatment. Generally, 6 to 12 months of treatment are required
before signs of chronic nephropathy become apparent. Because of the irreversible nature of the morphologic abnormalities, this form of toxicity is more ominous than the
acute form. Histologically, chronic cyclosporine nephrotoxicity is characterized by focal or striped medullary interstitial fibrosis. Often these changes are accompanied by tubular atrophy and obliterative arteriolar changes. In more
advanced cases, diffuse interstitial fibrosis with focal and
segmental glomerular sclerosis can be seen. In renal transplant patients, these changes may be difficult to differentiate from the typical features of chronic rejection. While
the mechanism of chronic cyclosporine nephrotoxicity is
not known, it is likely that cumulative dose, arterial hypertension, and immunologic injury contribute to the development of the lesion. In addition, animal studies suggest that
severe sodium depletion may potentiate the development
of renal fibrosis associated with cyclosporine adminis tration.
In a small number of patients who have been treated
with cyclosporine, fulminant ARF occurs with thrombocytopenia, consumptive coagulopathy, and thrombotic microangiopathy. This syndrome has been described in both
kidney and bone marrow allograft recipients. In renal transplants, the histologic changes resemble those of hemolyticuremic syndrome and the occurrence of this disorder carries a poor prognosis for kidney graft survival. However,
patients usually recover when the cyclosporine is discontinued and some of these patients have been successfully
retransplanted using other immunosuppressives. The
mechanism of this devastating syndrome is not known but
imbalances of vasoconstrictor to vasodilator eicosanoids
have been suggested to play a role.
Cyclosporine is metabolized primarily through the action
of hepatic P450 microsomal enzymes. Thus, agents that
influence the activity of this enzyme system can cause significant changes in cyclosporine metabolism. Generally,
drugs that reduce the rate of cyclosporine metabolism, such
as erythromycin, ketoconazole, and verapamil, produce increased serum levels and potentiate toxicity. On the other
hand, agents that increase the rate of cyclosporine metabolism, such as phenytoin, phenobarbital, and rifampin, may
reduce serum levels and thus may blunt therapeutic efficacy.
Tacrolimus

Tacrolimus (formerly known as FK506) is a macrolide

antibiotic with profound immunosuppressive properties.
Despite its complete lack of structural similarity with

265
cyclosporine, the effects of tacrolimus on cytokine expres sion and T lymphocyte activation are identical to those of
cyclosporine. The molecular target for both tacrolimus and
cyclosporine is calcineurin, a protein phosphatase that is
required for signaling by the T cell receptor. However,
tacrolimus is more than 100 times more potent than
cyclosporine in inhibiting this pathway. In view of its enhanced potency, the efficacy of tacrolimus has been compared to cyclosporine in a series of clinical trials. In these
studies, tacrolimus reduced the incidence of rejection; however, patient and graft survival were not prolonged. Moreover, the side effects of tacrolimus, including nephrotoxicity, were similar to and, in some cases, more severe than
those in cyclosporine-treated patients. Both acute and
chronic nephrotoxicity have been described with tacrolimus, but it has been suggested that hypertension occurs less
frequently, which, if confirmed, would be a major clinical
advantage. The observation that these chemically dissimilar
compounds have in vivo effects that are essentially identical
underscores the key role for calcineurin inhibition in both
the efficacy and toxicities of tacrolimus and cyclosporine.
Bibliography
Appel GB, Given DB, Levine LR, Cooper GL: Vancomycin and
the kidney. Am J Kidney Dis 8:75-80, 1986.
Bennett WM: The nephrotoxicity of immunosuppressive drugs.
Clin Nephrol 43(Suppl l):S3 -57, 1995.
Branch RA: Prevention of amphotericin B-induced renal impairment. Arch Intern Med 148:2389-2394, 1988.
Humes HD, Weinberg JM, Knauss TC: Clinical an d pathophysio logic aspects of aminoglycoside nephrotoxicity. Am J Kidney
Dis 2:5-25, 1982.
Kaloyanides GJ: Antibiotic-related nephrotoxicity. Nephrol Dial
Transplant 9(Suppl 4):130-134, 1994.
Kopp JR, Klotman PE: Cellular and molecular mechanisms of
cyclosporin nephrotoxicity. J Am Soc Nephrol 1:162-179,1990.
Parfrey PS, Griffiths SM, Barrett BJ, et al.: Contrast -material induced renal failure in patients with diabetes mellitus, renal insufficiency, or both: a prospective controlled study. N Engl J Med
320:143-149, 1989.
Sawyer MH, Webb DE, Balow JE, Straus SE: Acyclovir-induced
renal failure. Am J Med 84:1067-1071, 1988.
Schwab SJ, Hlatky MA, Pieper KS, et al.: Contrast nephrotoxicity:
A randomized controlled trial of a nonionic and an ionic radio graphic contrast agent. N Engl J Med 320:149-153, 1989.
Solomon R, Werner C, Mann D, D'Elia J, Silva P: Effects of saline,
mannitol, and furosemide on acute decreases in renal function
induced by radiocontrast agents. N Engl J Med 331:1416-142,
1994.
Tune BM, Hsu C-Y, Fravert D: Cephalosporin and carbacephem
nephrotoxicity: Roles of tubular cell uptake and acylating potential. Biochem Pharmacol 51(4):557-561, 1996.
US Multicenter FK506 Liver Study Group. A comparison of tacro limus (FK506) and cyclosporine for immunosuppression in liver
transplantation. N Engl J Med 331:1110-1115, 1994.
Welty TE, Copa AK: Impact of vancomycin therapeutic drug
monitoring on patient care. Ann Pharmacother 28:1335-1339,
1994.
Whelton A. Therapeutic initiatives for the avoidance of aminoglycoside toxicity. / Clin Pharmacol 25:67-81, 1985.
Zager RA: Endotoxemia, renal hypoperfusion, and fever: Interactive risk factors for aminoglycoside and sepsis-associated acute
renal failure. Am J Kidney Dis 20(3):223-230, 1992.

ACUTE RENAL FAILURE DUE TO

METABOLIC DERANGEMENTS
CHRISTOF WESTENFELDER

Acute renal failure (ARF) can complicate acute elevations of (1) uric acid, causing uric acid nephropathy;
(2) calcium, causing hypercalcemic nephropathy; and
(3) phosphate, causing hyperphosphatemic nephropathy.
Most patients with tumor lysis syndrome or with myoglobinuric ARF present with both severe hyperuricemia and
hyperphosphatemia. All three complications can occur
when patients with malignancy-induced hypercalcemia re
ceive chemotherapy.

which occurs when the distal tubular solubility of these

purines is exceeded. Such an increase is observed in patients
with severe hyperuricemia and, less commonly, in markedly
dehydrated patients, in whom serum UA levels may be
normal. With dehydration, UA solubility can be exceeded
because the urine in the distal tubule is hyperconcentrated.
Urinary acidification by the collecting duct is another factor
that contributes to the development of UAN. With a decrease in urinary pH, UR is converted to the less soluble
purine UA. In most patients with UAN, all three pathogenic mechanisms are contributory.

ACUTE URIC ACID NEPHROPATHY

Homeostasis and Renal Handling of Uric Acid (UA)

Uric acid nephropathy (UAN) is a reversible form of

ARF that develops when intraluminally precipitated uric
acid (UA) and monosodium urate (UR) crystals cause extensive obstruction of renal collecting ducts. This occurs
most commonly as part of the tumor lysis syndrome, i.e.,
when patients with myelo- or lymphoproliferative malignancies receive chemotherapy. The syndrome may also
occur spontaneously, when tumor burden is high and cell
turnover rapid. The high rate of nucleic acid breakdown
leads to grossly augmented UA production, hyperuricemia,
and hyperuricosuria. Reversal or prevention of UAN is
successful in approximately 95% of patients.

Uric acid is the xanthine oxidase-dependent end product

of endogenous and dietary nucleoproteins. The UA pool
in the adult is 1 g and the daily turnover 500 to 700 mg.
Of the latter, 80% appears in the urine and 20% is metabolized by colonic bacteria. Eighty percent of the UA pool
is derived from the catabolism of endogenous nucleic acids
and nucleotides, the remainder from the diet.
In the kidney, UR and UA are subjected to (1) glomerular filtration, (2) almost complete proximal tubular reabsorption, (3) proximal tubular secretion, and (4) proxi mal
and a minor degree of distal tubular "postsecretory" reabsorption. The fractional excretion of UR is 10%, which
corresponds to the 24-hour urinary excretion of 500 to
700 mg. Volume expansion and some organic acids, e.g.,
salicylates in high doses or probenecid, reduce UR reabsorption. Volume contraction, on the other hand, enhances
UR reabsorption and causes hyperuricemia. Hyperuricemia also develops when proximal tubular secretion of UR
via the organic acid transporter is inhibited by competing
organic acids, e.g., salicylates in low doses, probenecid,
pyrazinamide, lactate, and ketoacids. Finally, UR excretion
increases as glomerular filtration decreases and BUN and
serum creatinine rise. As a consequence, plasma UR levels
in patients with end-stage renal disease plateau at 12 to
13 mg/dL.

Pathology

On gross examination, the papillae contain yellow, linear

striations that outline the course of collecting ducts obstructed by precipitated UA and UR. On microscopic examination, the tubular lumens contain needle-shaped birefringent crystals. Proximal tubules may be dilated due to
obstruction. The glomeruli are spared, except for mild dilatation of Bowman's space. If acute uric nephropathy is
unresolved, acute and chronic inflammatory cells, tubulointerstitial granulomas, edema, and medullary microtophi appear.

Clinical Presentation
Pathogenesis

Predictably, marked hyperuricemia (> 20 mg/dL), particularly when combined with dehydration or acidemia, causes
acute UAN. A sudden fall in urine output, complete anuria,

Critical to the development of UAN is the extensive

obstruction of collecting ducts by UA and UR crystals,

Primer on Kidney Diseases, Second Edition

266

Copyright 1998 6 by the National Kidney Foundation

Alt rights of reproduction in any form reserved.

267

37. Acute Renal Failure Due to Metabolic Derangements

and rarely flank pain or ureteral colic may be noted. When

severe, the acute loss in renal function leads to progressive
azotemia, symptoms of uremia, volume overload, hyperkalemia, hyperphosphaternia, and metabolic acidosis. Unless
pretreated with allopurinol and adequate volume expansion, up to 90% of patients who receive chemo - or radiation
therapy for an advanced myelo- or lymphoproliferative
malignancy (see Table 1) will develop hyperuricemia and
secondary UAN. At least 30% of these develop severe
hyperkalemia and hyperphosphatemia, both major complications of the tumor lysis syndrome. Since marked hyperphosphatemia per se can cause ARF and severe hypocalcemia, patients with tumor lysis require particularly
aggressive and prompt therapy. On rare occasions, dehydration alone has been found to precipitate UAN in patients with breast, gastric, or myelo- or lymphoproliferative malignancies.
Patients with the rare Lesch-Nyhan syndrome develop
recurrent episodes of UAN, severe hyperuricemia, gout,
and encephalopathy with self-mutilation. These complications result from a complete, X-linked deficiency of
hypoxanthine-guanine-phosphoribosyltransferase. Overproduction of UR due to other genetic disorders is an
uncommon cause of UAN. Occasionally, uricosuric agents
such as salicylates and probenecid can precipitate UAN.
A purine-rich diet per se does not cause UAN.
Diagnosis
Typically, patients with suspected UAN present with
ARF and severe hyperuricemia as chemotherapy for a
myelo- or lymphoproliferative malignancy is initiated. A

TABLE I

Metabolically Induced Acute Renal Failure

Uric acid nephropathy
Hyperuricosuria (>600 mg/24 h) with hyperuricemia Myelo-,
lymphoproliferative malignancies with tumor lysis
Metastatic cancers of breast, lung, stomach with tumor lysis
Rhabdomyolysis, crush injury, seizures, thyrotoxicosis
Lesch-Nyhan syndrome
Hyperuricosuria (>600 mg/24 h) without hyperuricemia
Uricosuric drugs: High -dose salicylates, probenecid
Fanconi syndrome, acquired, genetic
Hypercalcemic nephropathy Malignancies: Squamous cell
carcinoma of lung, lymphoma,
multiple myeloma, kidney, breast
Primary hyperparathyroidism
Sarcoidosis, granulomatous diseases
Vitamin D or A intoxication Milk-alkali
syndrome, immobilization
Hyperphosphatemic nephropathy
Myelo-, lymphoproliferative malignancies with tumor lysis
Metastatic cancers of breast, lung with tumor lysis
Rhabdomyolysis, crush injury, seizures, thyrotoxicosis, bowel
necrosis In renal insufficiency: PO4 enemas, PO4 laxatives,
TPN with
PO4

serum urate level greater than 20 mg/dL should suggest

the diagnosis because values are seldom this high in other
causes of renal failure. Similarly, a urine uratexreatinine
excretion ratio greater than 1 (calculated from concentrations expressed in mg/dL) may be helpful as it suggests
that the hyperuricemia results from overproduction and is
associated with hyperuricosuria. In contrast, volume depletion and other states that cause hyperuricemia by reducing
urate excretion tend to be associated with low ratios. Diuretics induce volume contraction and compete with urate
for secretion as do other organic acids, such as salicylates,
probenecid, lactate, and ketoacids. A high ratio is not an
invariable finding in UAN. UA crystalluria occurs frequently, but it is neither a constant nor pathognomonic
finding.
Hyperphosphatemia, hypocalcemia, and hyperkalemia
are features of the tumor lysis syndrome that may accompany UAN. However, due to muscle destruction, these
findings and hyperuricemia may also occur with rhabdomy olysis. In rhabdomyolysis -induced ARF, plasma CK and
myoglobin levels are elevated. Other distinguishing features of this disorder are discussed in Chapter 39.
The diagnosis of UAN may be more difficult if spontaneous tumor lysis is the presenting feature of the malignancy.
In addition, other causes of cancer-associated ARF such
as direct renal infiltration with tumor, ureteral obstruction,
sepsis, and administration of nephrotoxic antibiotics or chemotherapeutic agents must be considered. The diagnosis
is usually made clinically and a kidney biopsy is generally
not indicated.
Therapy and Outcome
With appropriate therapy, recovery of renal function
in UAN approaches 100%, and attendant morbidity and
mortality are mostly a function of the underlying disease.
Prevention of UAN in high-risk patients with myelo- and
lymphopro liferative malignancies is successful in more than
95% of cases. The duration of UAN depends on the age
of the patient and other complicating factors. These include
the presence of underlying renal disease or renal insufficiency, pyelonephritis, vascular disease, the concomitant
use of nephrotoxic antibiotics, and the hemodynamic and
nutritional status.
Prophylaxis
Patients with myelo- and lymphoproliferative malignancies or with extensive solid tumors are at greatest risk for
UAN. In these, baseline 24-hours UR excretion is frequently >1000 mg/24 hr, while serum UA levels may be
misleadingly normal. When dehydration or acidemia occurs, UAN develops. The situation is similar in HGPRT
deficiency, and rarely occurs with gout. Allopurinol, maintenance of euvolemia, a high urine output (>2.5 L/day),
and normal acid-base balance are effective preventive
measures. The same principles apply when such a patient
is to undergo cancer therapy; volume deficits must be corrected with parenteral normal saline and oral fluids, and a
urine output of >2.5 L/day established. Furosemide may

268

Christof Westenfelder

be added once hydration is assured. Optimally, the urine

is alkalinized to a pH >7.0, using iv sodium bicarbonate, or
Shohl's solution, and, if needed, additional acetazolamide.
Since the latter measures can cause hypokalemia, serum
potassium levels must be monitored and corrected. Allopurinol is begun 1 to 3 days prior to cancer therapy. When
renal function is impaired, the allopurinol dose must be
reduced. Allopurinol raises plasma and urinary hypoxanthine and xanthine levels, occasionally causing ARF due
to intrarenal xanthine precipitation, i.e., xanthine nephropathy.
Established UAN
Once oligoanuric UAN has developed, uremic symptoms
and other complications develop as in any patient with
severe ARF. Conservative therapy, as described above,
may suffice if renal clearance is above 30 ml/min. Daily
hemodialysis is highly effective in reducing the UR pool
and in treating uremic and other complications. With severe
hyperphosphatemia (>10 mg/dL), a low calcium dialysate
must be used to prevent soft tissue calcification and ARF
due to intrarenal CaPO4 precipitation. The latter can also
occur when patients with high CaPO4 products, as seen in
tumor lysis, receive bicarbonate for the purpose of urinary alkalinization.
HYPERCALCEMIC NEPHROPATHY

Hypercalcemic nephropathy results from the effects of

acute or chronic hypercalcemia, which, in 50% of the patients, is the consequence of an underlying malignancy (Table 1). In addition to acute renal failure the renal complications of hypercalcemia include tubular defects such as renal
tubular acidosis and nephrogenic diabetes insipidus, as well
as nephrocalcinosis and nephrolithiasis. The severity of
renal insufficiency depends on the duration and magnitude
of hypercalcemia and concurrent hyperphosphatemia and
hyperuricemia. Unless renal insufficiency is advanced, correction of hypercalcemia improves kidney function, reverses nephrocalcinosis, and reduces kidney stone formation.
Acute hypercalcemia decreases creatinine clearance via
intrarenal vasoconstriction. In addition, high calcium levels
reduce the reabsorption of Na, K, Mg, and H2 O, potentially
lead to volume contraction, prerenal azotemia, hypernatremia, polydipsia, polyuria, and, in 20 to 30% of patients,
hypokalemia and hypomagnesemia. Acute and severe hypercalcemia (>15 mg/dL), especially with concomitant
hyperphosphatemia, can precipitate reversible ARF by
causing renal vasoconstriction and distal tubular CaPO4
precipitation.
Chronic hypercalcemia produces calcification and subsequent necrosis of collecting duct cells. Cellular debris and
CaPO4 crystals cause distal obstruction and proximal tubular dilatation. Nephron loss results from tubular atrophy,
interstitial scarring, diffuse nephrocalcinosis, and stones.
Furthermore, evolving distal renal tubular acidosis may
aggravate nephrocalcinosis and kidney stone formation.

Once GFR is 30 mL/min or less, hyperphosphatemia develops and serum calcium falls.
More than 50% of patients with hypercalcemia develop
hypertension due to increased peripheral vascular resis tance and high cardiac output. Passage of a kidney stone
may be the first manifestation of hypercalcemia. With severe hypercalcemia, oliguria, hypovolemia, confusion, lethargy, and muscle weakness develop.
The diagnosis of hypercalcemic nehropathy rests on the
simultaneous demonstration of renal insufficiency and hypercalcemia. The urinalysis may show cellular casts, RBCs,
and calcium phosphate or oxalate crystals. Nephrocalcinosis, stones, and obstruction are diagnosed radiographically-, or by ultrasonography.
Correction of acute hypercalcemia and attendant volume
deficits promptly improves renal function by reversing renal
vasoconstriction and tubular obstruction. Whenever renal
function allows, volume expansion with normal saline and
furosemide are employed to increase urinary calcium excretion. Patients with poor renal function require intensive
hemodialysis with a low calcium bath. This will effectively
reduce serum calcium levels and correct uremic and other
electrolyte abnormalities. For other calcium-lowering therapies see Chapter 14.
The prognosis of hypercalcemic nephropathy depends
on the underlying disorder. Primary hyperparathyroidism
and most other benign causes of hypercalcemia are eminently treatable and recovery from hypercalcemic nephropathy should be expected. The prognosis is more guarded
when the hypercalcemia is due to malignancy.
HYPERPHOSPHATEMIC NEPHROPATHY

Extreme hyperphosphatemia (>20 mg/dL) is commonly

complicated by soft tissue calcifications, hypocalcemia, and
ARF. In general, hyperphosphatemia is a prominent finding of tumor lysis seen in patients who receive chemotherapy for malignancies such as Burkitt's ly mphoma, lymphoblastic leukemia, metastatic small cell lung carcinoma, and
adenocarcinoma of the breast. In addition, hyperphosphatemia can complicate rhabdomyolysis, malignant hyperthermia, and extensive bowel necrosis (Table 1). Release
of cellular PO4 , K, nuclear proteins, and LDH produces
hyperphosphatemia, hyperkalemia, hyperuricemia, and
markedly elevated LDH levels. In addition, excessive PO4
administration to patients with renal insufficiency causes
symptomatic hyperphosphatemia.
ARF results from CaPO4 precipitation in collecting ducts
and renal interstitium, impairing renal function by obstruction and interstitial inflammation. Renal histology is similar
to that seen in hypercalcemic nephropathy. Rhabdomyolysis may cause myoglobinuric ARF. Hyperphosphatemic
patients are also at high risk for the development of UAN
and may thus undergo urinary alkalinization. In the pres ence of profound hyperphosphaturia, alkalinization may
lead to additional CaPO4 precipitation and a further deterioration of renal function. In patients with a creatinine clearance >50 mL/min, volume expansion with normal saline,
modest urinary alkalinization, and acetazolamide diuresis

269

38. Cholesterol Atheroembolic Renal Disease

will rapidly lower plasma PC>4 levels and improve renal

function. Glucose and insulin can be used to induce a transient intracellular shift of PO4 and K. With poor renal
function and uremic manifestations, intensive dialysis is required.
Bibliography
Benabe JE, Martinez-Maldonado M: Hypercalcemic nephropathy.
Arch Intern Med 138:777-779, 1978. Cohen LF, Balow JE,
Magrath IT, Poplack DG, Ziegler JL: Acute
tumor lysis syndrome: A review of 37 patients with Burkitt's
lymphoma. Am J Med 68:486-491, 1980. Conger JD: Acute
uric acid nephropathy. Semin Nephrol 1:69
74, 1981. Emmerson BT, Thompson L: The spectrum of
hypoxanthineguanine phosphoribosyltransferase deficiency. Q J Med
166:423-440, 1973. Hande KR, Noone RM, Stone WJ: Severe
allopurinol toxicity:
Description and guidelines for prevention in patients with renal
insufficiency. Am J Med 76:47-56, 1984. Jones DP, Mahmoud
H, Chesney RW: Tumor lysis syndrome:
Pathogenesis and management. Pediatr Nephrol 9:206-212,
1995. Kelly WN: Pharmacologic approach to the maintenance
of urate
homeostasis. Nephron 14:99-115, 1975.

Kelton J, Kelley WN, Holmes EW: A rapid method for the diagnosis of uric acid nephropathy. Arch Intern Med 138:612-615,1978.
Kjellstrand CM, Campbell DC, von Haritzsch B, Buselmeier TJ:
Hyperuricemic acute renal failure. Arch Intern Med 133:349359, 1974.
Klinenberg JR, Kippen I, Bluestone R: Hyperuricemic nephropathy: Pathologic features and factors influencing urate deposition.
Nephron 14:88-98, 1975.
Lins LE: Reversible renal failure caused by hypercalcemia: A
retrosp ective study. Acta Med Scand 203:309-314, 1978.
Rieselbach RE, Bentzel CJ, Cotlove E, Frei E, Freireich EJ: Uric
acid excretion and renal function in the acute hyperuricemia of
leukemia: Pathogenesis and therapy of acute uric acid nephropathy. Am J Med 37:872-884, 1964.
Rieselbach RE, Steel TH: Influence of the kidney upon urate
homeostasis in health and disease. Am J Med 56:665-675,1974.
Simmonds HA, Cameron JS, Morris GS, Davies PM: Allopurinol
in renal failure and the tumor lysis syndrome. Clin Chim Acta
160:189-195, 1986.
Stryer L: Biosynthesis of nucleotides. In: Biochemistry, 3rd ed.
Freeman New York, pp. 601-621, 1988.
Tungsanga K, Boonwicht D, Lekhakula A, Sitprija V: Urine uric
acid and urine creatinine ratio in acute renal failure. Arch Intern
Med 144:934-937, 1984.
Zusman J, Brown DM, Nesbit ME: Hyperphosphatemia, hyperphosphaturia and hypocalcemia in acute lymphoblastic leukemia. N Engl J Med 289:1335-1340, 1973.

CHOLESTEROL ATHEROEMBOLIC
RENAL DISEASE
ARTHUR GREENBERG

Cholesterol atheroembolic renal disease results when

cholesterol crystals and other debris separate from atheromatous plaques, flow downstream, and lodge in small renal
arteries, producing luminal occlusion, ischemia, and renal
dysfunction. Depending on the source and distribution of
emboli, renal disease may be the sole or predominant manifestation or simply one feature of a systemic illness characterized by multiorgan ischemia or infarction.
Early, autopsy-derived descriptions of renal atheroembolism overemphasized a catastrophic presentation with
irreversible renal failure, intestinal infarction, and death
from intra -abdominal sepsis. Recently, atheroembolism

Primer on Kidney Diseases, Second Edition

has been recognized as a cause of occult or reversible renal

failure. Recovery of renal function may follow extended
survival on renal replacement therapy.
PATHOLOGY

The initial lesion in cholesterol atheroembolism is obstruction of a medium-sized or small artery by atheromatous debris. Arterioles and capillaries are less commonly
affected. Lesions may occur in any organ. Cholesterol dis solves in formalin used to process tissue for routine histologic examination; crystals are not seen in tissue sections
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

270

Arthur Greenberg

FIGURE I Cholesterol atheroembolus occluding the lumen of an interlobular renal

artery. Needle-like clefts (solid arrows) are present along with a macrophage/multinucleated giant cell reaction (open arrow) (methenamine silver-trichrome stain, 450X).
(Courtesy of Dr. S.I. Bastacky.) Reproduced with permission from Am J Kidney Dis
29:334-344, 1997.

unless special fixatives are used. However, a characteristic

cleft marks the space where the needle-like crystals had
been (Fig. 1). The size of the artery affected is typically
around 200 /u,m, but may range from 55 to 900 jam. The
earliest lesion consists of cholesterol crystals and thrombus.
After dissolution of the thrombus, macrophages engulf the
cholesterol, but the predominant reaction is endothelial.
New endothelium covers the crystals. If the vessel wall is
eroded by the crystals, an intense perivascular inflamma -

tory response with giant cells is established (Fig. 1). As a

late finding, concentric fibrosis, particularly involving the
adventitia, occurs. Finally, there is recanalization of small
vascular channels. Crystals may reach the glomerular capillary loops (Fig. 2). Crystal dissolution in vivo is slow; in
experimental models, cholesterol clefts persist for long as
9 months after embolization. The main glomerular finding
is that of ischemia, with glomerular collapse and basement
membrane wrinkling. In some patients, focal segmental

FIGURE 2 Glomerular findings in cholesterol at heroembolism. A distended glomerular

capillary loop adjacent to the glomerular hilum contains a multinucleated macrophage
with two cholesterol clefts (arrow) (hematoxylin-eosin stain, 150X). (Courtesy of Dr.
Dennis Borochovitz.)

3 8 . C h o l e s t e r o l A t h e r o e m b o lic Renal Disease

glomerulosclerosis with glomerular collapse and epithelial

cell prominence occurs. This finding accounts for some of
the cases associated with nephrotic range proteinuria.
PATHOGENESIS

The classic autopsy description by Flory noted choles terol atheroembolism solely in patients with erosive
plaques. The prevalence of atheroembolism paralleled the
severity of aortic disease. Less severe atherosclerotic lesions or plaques covered by thrombus do not pose a risk
of atheroembolism.
Embolization may be spontaneousparticularly with severe aortic diseasebut mechanical disruption of plaque
during angiographic or surgical procedures usually precedes it. Table 1 lists predisposing factors. Regardless of
the area primarily targeted for imaging, passage of a catheter along the ascending or descending aorta proximal to
the renal arteries confers a risk of embolization to the
kidneys. Renal artery angioplasty may be a particular risk.
The site of any concurrent nonrenal embolization depends
on the path of the catheter.
Thrombus overlying atheromatous plaque may bind and
immobilize friable debris. Anticoagulation or thrombolysis
removes this protective covering; atheroembolism has been
reported after heparin, warfarin, or thrombolytic therapy
without angiography.
CLINICAL FEATURES

As expected of a process that complicates severe atherosclerosis, risk factors for atherosclerosis as well as evidence
of disseminated atherosclerotic disease are commonly pres ent. The incidence is higher in smokers. Up to 75% of
patients are male. The mean age at diagnosis is in the midseventh decade. Fewer than 5% of patients are below age
50. Table 2 lists other accompanying or predisposing features. Notably, diabetes mellitus is not commonly observed.
Cholesterol atheroembolism is notable for its highly variable severity. Its manifestations depend on both the extent
of renal involvement and the extrarenal sites affected. Mas -

TABLE I

Risk Factors for Cholesterol Atheroembolic

Renal Disease

271
TABLE 2

Associated Findings in 221 Patients with

Cholesterol Atheroembolism0
Finding
% of cases
61
Hypertension Coronary artery
Cerebrovascular disease Congestive

4
4

heart failure Diabet es mellitus

25

disease Aortic aneurysm

21
" From Fine MJ, Kapoor W, Falanga
V: Angiol-ogy 38:769-784, 1987.

21

sive and widespread embolization in the multiple cholesterol emboli syndrome presents catastrophically with fever,
stroke, acute renal failure, abdominal pain and gastrointes tinal bleeding due to bowel infarction, intra-abdominal sepsis, and death. The frequency of organ involvement is summarized in Table 3.
Autopsy reports are skewed toward patients with severe
involvement. Recently, milder disease, with predominant
renal and cutaneous involvement, has been recognized.
Typically, the renal course is characterized by slowly deteriorating renal function. The daily increase in serum creatinine may be as little as 0.1 to 0.2 mg/dL and progression
to end stage disease may occur over 30 to 60 days or even
longer. Patients may also present over many months with
the insidious development of renal insufficiency. Occasional patients present with heavy proteinuriawith or
without associated clinical features of atheroembolism.
Skin involvement includes livedo reticularis of the lower
extremities due to occlusion of small arteries as well as
cyanosis of the toes. This classic "blue toes" lesion occurs
in spite of preservation of distal pulses (Chapter 35, Fig.
1A). Digital ulceration may occur along with severe pain.
Cutaneous involvement may be overlooked unless specifically sought.
Features of gastrointestinal involvement include abdominal pain, anorexia, weight loss, and bleeding that can range
from a positive stool test for occult blood to brisk hemorrhage. Infarction and sepsis may occur. Central nervous
system involvement includes stroke or diffuse cortical dysfunction due to widespread embolization as well as the
scotomata, field cuts, or blindness that accompany the classic, but rare, retinal Hollenhorst plaque.

latrogenic
Surgery on the aorta proximal to the renal arteries
Aortic aneurysm repair Coronary artery bypass
grafting Cardiac valve surgery Other
Angiography or angioplasty
Intra-aortic balloon pump circulatory augmentation
Anticoagulation or thrombolytic therapy
Spontaneous
Severe ulcerating atherosclerosis

DIAGNOSIS

The diagnosis of atheroembolism relies on a high index

of suspicion in patients at risk. During the acute phase,
leukocytosis and an elevated erythrocyte sedimentation
rate may be present. Eosinophilia is observed in approximately one half of affected individuals; hypocomplementemia occurs rarely. Hyperamylasemia suggests pancreatic
involvement. Although nephrotic range proteinuria may

272

Arthur Greenberg
TABLE 3

Histologic Involvement at Autopsy"

% of cases
Kidney

75

Spleen

52

Pancreas

Gastrointestinal tract

52
31

Adrenal glands

20

Liver
Brain

17
14

Skin

6
" From Fine MJ, Kapoor W, Falanga V: Angiol-ogy
38: 769-784, 1987.

occur, proteinuria is typically modest and the urine sediment nonspecific.

Principal differential diagnostic considerations include
radiocontrast-induced acute renal failure, ischemic acute
tubular necrosis, systemic vasculitis, allergic interstitial nephritis, cryoglobulinemia, myeloma, hypertensive nephrosclerosis, and renal artery stenosis. Radiocontrast nephropathy (see Chapter 36) is readily distinguished from
atheroembolism by its more rapid course. In dye nephropathy, renal failure occurs immediately, with renal function
reaching a nadir within 3 or 4 days, and substantial recovery
usually occurring over a similar period. In atheroembolism,
the onset of renal failure may be delayed and progression
is slower. Recovery occurs lateif at all.
Most instances of cholesterol atheroembolism are diagnosed clinically. If a tissue diagnosis is deemed necessary,
biopsy of an area of livedo reticularis is the least invasive
approach. A muscle biopsy may also be used. Although
renal biopsy showing cholesterol atheroemboli is definitive,
alternative means of diagnosis should be considered first.
THERAPY AND OUTCOME

Spontaneous atheroembolism occurs only in patients

with more severe vascular involvement. Shedding of atheromatous debris often continues with a progressively declining course. In patients with atheroembolism after vascular
surgery or angiography, embolization may be limited to
the initial episode. Stabilization and gradual improvement
may follow as small vessel inflammation subsides and re-

canalization with restoration of blood flow occurs. A typical

patient will have gradually lessening anorexia and abdominal and digital pain with subsequent healing of digital ischemic lesions. Management focuses on supportive measures
with local care of digital ischemia, analgesia for pain, and
digital amputation if tissue is not viable. Anticoagulation
should be stopped and repeat angiographic procedures
avoided. Hemodialysis or peritoneal dialysis may be successfully employed. Some patients recover sufficient renal
function to permit discontinuation of dialysis ; in others,
renal failure is irreversible. Recent reports note that abdominal or transesophageal ultrasonography may be used
to localize diseased areas of aorta. Cholesterol atheroembolism occurs in an elderly population with extensive atherosclerotic disease and a high prevalence of hypertensive
cardiovascular disease. Such patients present a formidable
surgical risk; the role of endarterectomy or resection of
diseased aortic segments has not been established.
Bibliography
Colt HG, Begg RJ, Saporito JJ, Cooper WM, Shapiro AP: Cholesterol emboli after cardiac catheterization. Eight cases and review
of the literature. Medicine 67:389-400, 1988.
Fine MJ, Kapoor W, Falanga V: Cholesterol crystal embolization:
A review of 221 cases in the English literature. Angiology
38:769-784, 1987.
Flory CM: Arterial occlusions produced by emboli from eroded
aortic atheromatous plaques. Am J Pathol 21:549-565, 1945.
Greenberg A, Bastacky SI, Iqbal A, Borochovitz D, Johnson JP:
Focal segmental glomerulosclerosis associated with nephrotic
syndrome in cholesterol atheroembolism: Clinicopathologic correlations. Am J Kidney Dis, 29:334-344, 1997.
Kasinath BS, Corwin HL, Bidani AK, Korbet SM, Schwartz MM,
Lewis EJ: Eosinophilia in the diagnosis of atheroembolic renal
disease. Am J Nephrol 7:173-177, 1987.
Kassirer JP: Atheroembolic renal disease. N Engl J Med 280:812818, 1969.
Mannesse CK, Blankestijn PJ, Man in 't Veld AJ, Schalekamp
MADH: Renal failure and cholesterol crystal embolization: A
report of 4 surviving cases and a review of the literature. Clin
Nephrol 36:240-245, 1991.
McGowan JA, Greenberg A: Cholesterol atheroembolic renal disease. Report of 3 cases with emphasis on diagnosis by skin
biopsy and extended survival. Am J Nephrol 6:135-139, 1986.
Thadhani RI, Carma go CA, Xavier RJ, Fang LST, Bazari H:
Atheroembolic renal failure after invasive procedures. Natural
history based on 52 histologically proven cases. Medicine 74:350358,1995.
Tunick PA, Perez JL, Kronzon I: Protruding atheromas in the
thoracic aorta and systemic embolization. Ann Intern Med
115:423-427, 1991.

PIGMENT NEPHROPATHY
JAMES P. KNOCHEL

Pigment nephropathy is a term describing acute renal

failure caused by rhabdomyolysis with myoglobinuria or
hemolysis with hemoglobinuria.

myophosphorylase deficiency (McArdle's syndrome) and

carnitine palmityl transferase deficiency.
Acquired Metabolic Myopathies

RHABDOMYOLYSIS

Potassium deficiency is a typical example of an acquired

metabolic myopathy. Potassium deficiency impairs glycogen synthesis in muscle. Since glycogen is a major energy
substrate during anaerobic work, physical work performed
by a potassium deficient individual can result in rhabdomyolysis. Potassium deficiency also interferes with the normal increase of muscle blood flow with exe rcise. This can
impose damage by ischemia.

Rhabdomyolysis results from injury to muscle cells, causing leakage of their contents into the blood and urine.
Trauma or recurrent attacks of cramps or pain during exercise and excretion of cola-colored urine are classic events
in rhabdomyolysis. Although some patients experience few
symptoms, most demonstrate tender, stiff, or firm muscles.
Severe weakness or even paralysis may occur due to extensive necrosis or hyperkalemia. Serum creatinc phosphokinase (CK) is virtually always elevated in rhabdomyolysis.
Idiopathic muscle necrosis in patients with diabetes mellitus is a rare exception.

Hypoxia/lschemia

Carbon monoxide poisoning, by causing hypoxia because

of carbon monoxyhemoglobin formation, is a recognized
cause of acute rhabdomyolysis. Severe congestive heart
failure may cause modest rhabdomy olysis.

Causes
Exertional and Traumatic Rhabdomyolysis
Normal subjects can develop rhabdomyolysis after intense exercise. Violent, repetitive activities or a grand mal
seizure are good examples. Presumably, exhaustive exercise may not only directly injure structural components of
muscle cells, but also deplete energy stores and disrupt
cellular transport, thereby allowing calcium to accumulate
in the cell. Calcium overload in turn activates proteolytic
enzymes that result in cell death. Many factors lower the
threshold for injury. These include poor physical condition
or preexistent injury, typified by alcoholic myopathy. For
any given unit of work, women unexplainably show much
less rhabdomyolysis than men. Volume depletion and exercise in the heat, perhaps by causing overheating of muscle
and reduced blood flow, are potentiating factors. Eccentric
muscle contractions (running downhill) are more likely to
cause rhabdomyolysis than concentric contractions (running uphill). Fasting lowers the threshold presumably by
limiting substrates for muscle contraction. Finally, associated illnesses that primarily affect skeletal muscle, typified
by influenza, reduce the threshold for injury induced by
exercise. Death in cases of crush syndrome is often related
to the complications of rhabdomyolysis.

Drugs

Many drugs cause subtle muscle cell injury or frank rhabdomyolysis. Although hundreds of drugs have been implicated, those of major concern include cocaine, heroin,
amphetamines, HMG-CoA inhibitors, and fibric acid derivatives used as cholesterol-lowering agents, especially when
given in conjunction with cyclosporine, nicotinic acid, or
erythromycin. The combination of lovastatin and gemfibrozil may also cause rhabdomyolysis. Drugs that appear
to interfere with key glycolytic enzymes such as isoniazide hydrazide, which impairs activity of glycogen phosphorylasemay also be implicated.
Special Causes
A number of substances are directly myotoxic. These
include clostridial toxin in gas gangrene, the proteolytic
enzymes in snake venom, or the myotoxin that accumulates
in skeletal muscle of the quail when it ingests sweet parsley
seeds. Hypothyroidism consistently causes elevated CK
and sometimes causes frank rhabdomyolysis.
Laboratory Diagnosis

The CK-MM is the dominant isoform in skeletal muscle

and the most sensitive test to confirm the diagnosis. Without ongoing necrosis, CK peaks at 12 to 36 hours and has

Inherited Metabolic Myopathies

A number of specific enzyme derangements are responsible for exertional rhabdomyolysis. Classic examples are

Primer on Kidney Diseases, Second Edition

273

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in anyform reserved.

274

James P. Knochel

a half life of about 48 hours (Fig. 1). The CK-MB isoform,

commonly employed to detect myocardial injury, may constitute 5% of total CK in skeletal muscle of highly trained
athletes. Healthy persons undergoing training exercises often show CK enzyme elevations as high as 10,000 IU/L.
Patients who are seriously ill from a variety of causes commonly show elevated CK-MM levels ranging from normal
to 10,000 IU/L. These include congestive heart failure with
hypoxia, sepsis, and a variety of acute infections. However, in rhabdomyolysis of clinical importance, elevations
of this enzyme are higher, commonly reaching levels of
100,000 IU/L or higher. In extreme cases, CK may approach
3.0 million IU/L. Measurements of other muscle enzymes
in rhabdomyolysis such as aldolase, lactic dehydrogenase,
or transaminases provide no additional useful information.
A transient elevation of serum creatinine disproportionate
to the elevation of BUN is commonly seen in acute
rhabdomyolysis. This results from release of creatine phosphate from muscle which is spontaneously dehydrated to
creatinine. The usual ratio of urea nitrogen to creatinine
in serum is 10:1. Ratios of five or less shortly after onset
suggest acute rhabdomyolysis. Uric acid in serum may exceed 40 mg/dL. Purines released from injured muscle are
converted to urate in the liver. Hyperuricemia of this magnitude is seldom seen in other conditions, even acute tumor
lysis induced by chemotherapy. Leukocytosis is common
in rhabdomyolysis of any etiology. Hypoalbuminemia is
an ominous sign and implies major capillary damage with
leakage of plasma components out of the vascular space.
On rare occasions, capillary damage may be so extensive
that erythrocytes also escape into interstitial tissues. This
results in shock with an acute reduction of hematocrit in
the absence of bleeding or hemolysis. A urine sodium concentration above 20 mEq/L suggests tubular injury. However, the urine sodium concentration may be low in cases
60,000

Serum creatine kinase

activity 19ylo d* pushups

50,000
2

40,000

O
| 30,000
CD

OT

20,000
10,000

6
8
Time (days)
12

to exhaustion
14 10

FIGURE I The course of serum CK activity

in a 19-year-old healthy man who developed rhabdomyolysis of
the deltoid and triceps muscles as a result of performing pushups
to the point of absolute exhaustion.

of myoglobinuria and, accordingly, this finding in pigment

nephropathy may be less helpful than in other oliguric
settings. Hyperkalemia is often observed as a consequence
of potassium release from damaged muscle cells. Profound
hypocalcemia, with serum calcium values below 3.0 mg/
dL, may result from hyperphosphatemia and trapping of
calcium in injured muscle. Transient resistance to parathyroid hormone followed by hypercalcemia during the recovery phase may be observed. Glucose may be detected in
the urine despite a normal glucose concentration in serum.
This apparently represents proximal tubular injury as a
result of pigment damage to the tubular epithelium. Finally,
pigmented casts may be seen in the urine.
HEMOLYSIS AND HEMOGLOBINURIA

Visible hemoglobinemia and hemoglobinuria occur only

in instances of major intravascular hemolysis. In extravas cular hemolysis, hemoglobin released into tissues is
promptly metabolized to bilirubin and the only pigment
found in the urine will be increased quantities of urobilinogen.
Transfusion with mismatched blood is the most common
cause of hemoglobin-induced nephropathy. Mycoplasma
infection appears to be the most commo n cause of hemolysis mediated by infection-associated immunologic mechanisms. Patients undergoing transurethral resection of the
prostate may develop major intravascular hemolysis as a
result of absorption of glycine solution from the prostatic
bed. Saline is not used in the perfusate because of the
hazards of electrical conduction by ionic solutions when
cautery is employed. The glycine is metabolized, allowing
electrolyte-free water to induce hemolysis. An example of
red cell oxidant injury is that mediated by application of
copper sulfate to white phosphorus burns. The term "black
water fever" refers to hemoglobinuria and acute renal failure in patients with falciparum malaria. Recent evidence
suggests that drugs used to treat malaria in addition to the
disease itself may be responsible for hemolysis or, alternatively, that some cases represent frank myoglobinuria as
a result of rhabdomyolysis associated with the malarial
infection. Laboratory findings in patients with intravascular
hemolysis include elevations of lactic dehydrogenase which
may be striking, slight hyperbilirubinemia, and in some
instances hyperkalemia and hyperphosphatemia. Plasma
hemoglobin levels will be markedly elevated. After hemo lysis, haptoglobin levels will be abnormally low. The urine,
similar to that from patients with rhabdomyolysis, may
show pigmented casts as well as free hemoglobin or its
product, hematin.
DISPOSITION OF HEME PIGMENTS IN PLASMA

There is a fundamental difference between the fate of

hemoglobin and that of my oglobin after their release into
plasma. Free hemoglobin in plasma saturates haptoglobin
at a concentration of approximately 100 mg/dL. Because
the hemoglobin haptoglobin complex is a large molecule,
it is not filtered by the glomerulus. Accordingly, hemoglo-

275

39. Pigment Nephropathy

bin will not appear in the urine until plasma haptoglobin

becomes saturated and total hemoglobin exceeds 100 mg/
dL. Hemoglobin becomes visible at a concentration of
100 mg/dL. In contrast to hemoglobin, myoglobin has no
specific binding protein. Any myoglobin entering plasma
is readily filtered by the glomerulus. Thus, the presence
of stained serum suggests hemolysis, not rhabdomyolysis.
Accordingly, if heme pigment is present in the urine, a
clear serum suggests rhabdomyolysis and myoglobinuria.
Because dipstick tests are exquisitely sensitive for detection
of heme pigment in the urine, there is no practical need
to perform more sophisticated and expensive tests to either
detect or differentiate myoglobin or hemoglobin.
MECHANISMS OF PIGMENT NEPHROPATHY

When acute renal failure complicating crush injury was

described shortly after World War II by Bywaters and
Beall, it was recognized that when pigmenturia occurs in
a setting of hypotension, decreased renal perfusion, and
an acid urine, acute tubular necrosis is likely to follow.
Subsequent experimental studies showed that if arterial
volume, blood pressure, and renal perfusion were intact
before the event, infusion of muscle extracts, myoglobin,
or hemoglobin had no adverse effects on renal function.
Experimental studies also showed that infusion of saline,
bicarbonate solutions, or mannitol before pigmenturia was
induced protected against acute renal failure.
The classic pathological findings in pigment nephropathy
include proximal tubular necrosis and distal tubular obstruction by pigmented casts. The mechanisms underlying
proximal tubular injury have been explored in detail. Once
the myoglobin or hemoglobin molecule is filtered by the
glomerulus and enters the proximal tubular fluid, it probably enters the proximal tubular cell by endocytosis. Hypothetically, the intact heme pigment molecule then is taken
up by lysosomes. The pH of the lysosomes is sufficiently low
that the heme pigment splits into its globin and ferrihemate
components. Ferrihemate is then transported out of the
tubular cell, requiring expenditure of ATP. In the presence
of renal ischemia, hypoxia, and a critical reduction of ATP
stores incident to active transport, the cell becomes injured.
Distal tubular obstruction by casts appears to increase the
amount of heme pigment absorbed by the proximal tubular
cells and, accordingly, increases its toxicity. In addition,
the iron portion of the ferrihemate molecule may generate
toxic superoxide radicals. Cytochrome P450 may be a
source of iron that catalyzes formation of these products.
Pigment-induced nephrotoxicity is attenuated by substances, such as deferoxamine that scavenge free oxygen
radicals. Experimentally, the kidney itself induces increased formation of heme oxygenase to protect against
continued or repeated exposures to heme proteins. Heme
pigments may also bind endogenous nitric oxide and
thereby cause vasoconstriction. Another theory holds that
intracellular stores of glutathione become exhausted and
this source of protection against oxidation damage is lost.
Although mannitol may be a free hydroxyl scavenger, its
major effect is to act as a diuretic and dilute the concentra-

tion of heme proteins within the tubular lumen, thus possibly decreasing their nephrotoxicity. In addition, mannitol
may enhance excretion of heme proteins, especially if they
have been solubilized by alkalinization of the urine with
bicarbonate. Unfortunately, no study has conclusively
shown that either mannitol or bicarbonate exerts protection
if given after the pigment is administered. Clearly, gelification of heme proteins in the distal tubule, a process requiring interaction with Tamm-Horsfall proteins, would be
ameliorated by any intervention that decreases their concentration or renders them less likely to form gels. Accord ingly, infusion of bicarbonate to alkalinize the urine has
been recommended for 50 years. Nevertheless, it has not
been clearly shown whether the prophylactic effect of bicarbonate is related to alkalinization or simply its osmotic
diuretic effect. Recent evidence suggests that intestinal endotoxins may be absorbed under several conditions, such
as in exhaustive exercise, trauma, or sepsis, thereby generating a cascade of cytokines (tumor necrosis factor and
other interleukins) that might also play a role in the pathogenesis of tubular injury. This could be of particular importance in patients who sustain massive rhabdomyolysis as a
result of a heavy exertion in the heat. Finally, renal tubular
cell injury has also been related to accumulation of calcium
in the cytoplasm which activates destructive proteases or
phospholipases. The cytoprotective effect of drugs that pre vent calcium entry in the cells as well as several amino acids
that either protect against the harmful effects of hemepigments (glycine) or promote their excretion (lysine) have
also been reported.
TREATMENT OF PIGMENT NEPHROPATHY

In severe rhabdomyolysis, it is critical to correct hypovolemia and shock by infusion of blood or crystalloid to maintain perfusion of vital organs. Although its use is controversial, most physicians advocate mannitol in a single
intravenous dose of 12.5 or 25 g over 15 to 30 minutes. A
total of 12.5 g or 70 mmol of mannitol (MW = 180) infused
into a total body water of 40 L results in only a trivial
increase of plasma osmolality, extracellular, or intraarterial volume. If this substance results in increased renal
perfusion and more rapid clearance of the heme pigment
from the kidney, its use is clearly justifiable. A single dose
of furosemide is also recommended to reduce sodium transport and oxygen utilization by the kidney, thus theoretically
preserving vital energy stores in the renal tubular cells. Its
diuretic effect possibly reduces the concentration of heme
proteins in the distal nephron. At the same time, because
of the potential of increasing solubility of heme proteins,
one should administer sodium bicarbonate to alkalinize the
urine. In some patients, it is extremely difficult to achieve
adequate urinary alkalinization because of the severity of
the associated metabolic acidosis.
Hyperkalemia potentially causes arrhythmias, a fall in
cardiac output, cardiac arrest, muscle paralysis, or respiratory failure if untreated. In severe rhabdomyolysis, even
modest hyperkalemia may cause cardiotoxicity because of
the simultaneous hypocalcemia that nearly always occurs.

lames P. Knochel

276
Calcium ions electrically oppose the cardiotoxic effects of
potassium. A serum potassium value of only 6.5 mEq/L
may be cardiotoxic when serum calcium is profoundly decreased. For this reason, the serum level of potassium
should not be relied upon as the sole marker of the severity
of cardiotoxicity; rather, the electrocardiogram itself must
be followed. Acute cardiotoxicity is most quickly and effectively treated by infusing calcium chloride, which produces
instantaneous improvement in the electrocardiogram. Unfortunately, infused calcium salts are rapidly deposited in
injured tissue so that their effect is transient. In addition,
calcium infusions do not lower serum potassium concentra tion. Specific maneuvers to reduce serum potassium concentration must be introduced simultaneously. These include glucose and insulin infusions and therapy with potent
/3-sympathctic agonists such as albuterol. By convention,
sodium bicarbonate infusions are also employed to shift
potassium into cells. However, the efficacy of bicarbonate
for treatment of hyperkalemia has been questioned. It
should be kept in mind that injured muscle cells are less
likely to take up potassium and these therapies may be
less effective than expected. Thus, in some cases of hyperkalemia, despite treatment, potassium ions are released at
such a rapid rate from injured muscle cells that dialysis
becomes mandatory. If possible, 20% sorbitol containing
the sodium potassium exchange resin clisodium polystyrene
disulfonate should be administered as soon as possible, by
either the oral or the rectal route, or both. Although this
measure is not effective as an acute treatment of hyperkalemia, it will help to blunt the rise of potassium over the
next 2 or 3 days. Hypocalcemia per se does not usually
require treatment unless calcium sails must be given for
hyperkalemia or for a hypocalcemia-induced decrease in
cardiac output. Despite hypocalcemia, tetany is very rare
in these patients, probably because of the muscle injury
and the antitetanic effect of metabolic acidosis. In those
patients given large quantities of calcium salts intravenously, subsequent mobilization of calcium from injured
tissue during the diuretic phase may result in potentially
fatal hypercalcemia. Accordingly, calcium salts should not
be used unless necessary. Hyperphosphatemia should be
treated by administration of phosphate binding antacids,
debridement of necrotic tissue in trauma cases, or dialysis.
Some authorities have advocated plasma exchange to
remove heme proteins and other proteinaceous components of destroyed cells from the circulation. However, this
is not an established form of treatment because of the rapid
clearance of pigment from plasma by the liver and spleen.
One must be vigilant for infection in any patient with acute
renal failure. Other complications seen include disseminated intravascular coagulation and thrombocytopenia. In
most instances, this complication clears spontaneously
within a few days but may require fresh frozen plasma
or specific components to facilitate blood clotting. Adult

respiratory distress syndrome (ARDS) and multiple organ

failure are also important complications.
One must be alert for development of fascial compartment compression syndromes. This is most apt to involve
muscles of the lateral thigh, the gastrocnemius, the anterior
tibial, and the gluteus maximus. Almost any muscle can be
involved. When muscle injury occurs, protein components
of muscle cells decompose and form powerful osmotic
forces that pull water into the muscle cell. If the muscle is
in a fascial compartment, the pressure rapidly increases and
will soon exceed arterial pressure. Ischemia and necrosis
follow. This can account for a secondary rise of CK in
serum or the appearance of irreversible nerve damage.
Measurement of tissue pressure in such locations can accurately predict the occurrence of a compartment syndrome.
Prophylactic treatment by fasciotomy can prevent secondary tissue necrosis. Current investigations suggest that mannitol may also be helpful toward preventing the compartment syndrome.
Bibliography
Baliga R, Zhang Z, Baliga M, Shah SV: Evidence for cytochrome
P-450 as a source of catalytic iron in myoglobinuric acute renal
failure. Kidney Int 49:362-369, 1996.
Better OS, Stein JH: Early management of shock and prophylaxis
of acute renal failure in traumatic rhabdomyolysis. N Engl J
Med 322:825-829, 1990.
Bywaters EG, Beall D: Crush injuries with impairment of renal
function. Br Med J 1:427-432, 1941.
Curry SC, Chang D, Connor D: Drug- and toxin -induced rhabdomyolysis. Ann Emerg Med 18:1068-1084, 1989.
Gabow PA, Kaehny WD, Kelleher SP: The spectrum of rhabdomyolysis. Medicine 61:141-149, 1982.
Haller RG, Knochel JP: Metabolic myopathies. In: Johnson RT,
Griffin JW (eds) Current Ther. Neural. Dis., Decker, pp. 397 402, 1993.
Knochel JP: Catastrophic medical events with exhaustive exercise:
White collar rhabdomyolysis. Kidney Int 38:709-719, 1990.
Knochel JP: Hematuria in sickle cell trait. Arch Intern Med
123:160-165, 1969.
Knochel JP: Hypophosphatemia and rhabdomyolysis. Am J Med
92:455-457, 1992.
Knochel JP: Mechanisms of rhabdomyolysis. Curr Opin Rheum atol
5:725-731, 1993.
Knochel JP: Potassium deficiency and rhabdomyolysis. / Mol Cell
Cardiol 27:241-249, 1995.
Miller JH McDonald RK: The effect of hemoglobin on renal function in the human. J Clin Invest 30:1033-1040, 1951.
Nath KA, Balla G, Vercollotti GM, et al.: Induction of heme
oxygenase is a rapid and protective response in rhabdomyolysis
in the rat. / Clin Invest 90:267-270, 1992.
Umpierrez GE, Stiles RG, Kleinbart J, Krendel DA, Watts NB:
Diabetic muscle infarction. Am J Med 101:245-250, 1996.
Wakabayashi Y, Kikuno T, Ohwada T, Kikawada R: Rapid fall
in blood myoglobin in massive rhabdomyolysis and acute renal
failure. Inten Care Med 20:109-112, 1994.
Zager RA: Rhabdomyolysis and myohemoglobinuric acute renal
failure. Kidney Int 49:314-326, 1996.

ACUTE INTERSTITIAL NEPHRITIS

CATHERINE M. MEYERS

AIN, increased expression of LFA-1 and VLA -4 cell surface receptors, as well as their respective ligands ICAM-1
and VCAM-1, is generally observed in areas of mononuclear cell infiltration. Recent studies have extended these
observations by examining the role of monocyte chemotactic peptide-1 (MCP-1), a potent chemoattractant and activating factor for monocytes, in interstitial nephritis. Renal
expression of MCP-1 is markedly upregulated in AIN and
correlates directly with the level of monocyte infiltration
and interstitial damage. Further support for the cellmediated hypothesis is also derived from the observation
of in vivo (delayed-type hypersensitivity responses) and in
vitro (lymphoblast transformation) activation on repeat
exposure to specific inciting agents.
Humorally mediated events may also be important in
eliciting some forms of tubulointerstitial injury as biopsies
in occasional patients with drug-induced lesions (rifampin,
methicillin, and phenytoin) have shown IgG and complement deposition along the tubular basement membrane
(TBM). Circulating anti-TBM antibodies have also been
reported in such settings.
One hypothesis concerning immune recognition of the
interstitium suggests that portions of infectious particles or
drug molecules may cross-react with or alter endogenous
renal antigens. An immune response directed against these
inciting agents would therefore also target the interstitium.
Although it is tempting to speculate on the relevance of
these cross-reactive antigens in interstitial disease, the nephritogenic potential of such a response has not been tested
within an experimental system.

Primary interstitial nephropathies constitute a diverse

group of diseases that elicit interstitial inflammation associated with renal tubular cell damage. This process typically
spares both glomerular and vascular structures. Traditionally, interstitial nephritis has been classified morphologically and clinically into acute and chronic forms. Acute
interstitial nephritis (AIN) generally induces rapid deterioration in renal function, and elicits marked interstitial inflammatory responses characterized by interstitial edema
with varying degrees of tubular cell damage, as well as
mononuclear cell infiltrates consisting primarily of lymphocytes (Fig. 1). Eosinophils, macrophages, plasma cells, and
neutrophils may also be apparent within these infiltrates.
In some cases of AIN, interstitial granuloma formation is
also observed. Most commonly this form of granulomatous
interstitial nephritis is associated with either drug- or
infection-induced renal inflammation. By contrast, chronic
tubulointerstitial disease follows an indolent course, characterized by tubulointerstitial fibrosis and atrophy associated with interstitial mononuclear cell infiltration. AIN is
not an uncommon cause of renal dysfunction, and should
always be considered in the differential diagnosis of acute
renal failure (ARF). Moreover, estimates from recent large
clinical studies suggest that AIN comprises approximately
10 to 15% of reported cases of ARF.
PATHOGENESIS

Despite the varied inciting factors of tubulointerstitial

nephritis in humans (Table 1), the striking similarity of
induced interstitial lesions, which consist primarily of T cell
lymphocytes, suggests that immune-mediated mechanisms
are important either in initiating the interstitial damage or
in amplifying primary interstitial injury from nonimmune
causes. Studies from experimental models of interstitial
disease suggest that both humoral and cell-mediated immune mechanisms are relevant effector pathways for inducing interstitial injury. Cell-mediated events likely play a
prominent role in most forms of human disease, in view
of the preponderance of T cell lymphocytes (CD4+ and
CD8+ ) present within interstitial infiltrates, generally in
the absence of antibody deposition. Immunohistochemical
studies conducted on biopsies obtained in drug-induced
AIN also indicate the importance of cell-cell interactions
in intrarenal inflammation, as there is a significant increase
in interstitial expression of cellular adhesion molecules. In

Primer on Kidney Diseases, Second Edition

CLINICAL FEATURES

AIN occurs in four distinct clinical settings (Table 1). It

may occur as a result of drug therapy, systemic or local
infection, immunologic disease, or an idiopathic lesion
without an apparent precipitating cause. AIN is observed
in all age groups; however, older patients appear predis posed to developing ARF. Systemic manifestations of a
hypersensitivity reaction, such as fever, rash, and arthralgias, are nonspecific findings that may accompany AIN. In
such cases, an erythematous maculopapular rash involves
the trunk and proximal extremities. Hypertension and
edema are not characteristic of AIN, although they have
been noted with s pecific drug-induced lesions. Other non277

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

278

Catherine M. Meyers

FIGURE I Acute interstitial nephritis. Light microscopic findings demonstrate the loss of normal
tubulointerstitial architecture with a dense mononuclear cell infiltrate (double arrows) and some
evidence of tubular dilatation and atrophy (single arrow). Note that the renal tubules are displaced
by infiltrating mononuclear cells, edema, and mild interstitial fibrosis. (Courtesy of Dr. Michael
P. Madiao, University of Pennsylvania, Philadelphia, PA.)

specific constitutional symptoms, as well as flank pain with

gross hematuria, have been variably reported.
The spectrum of urinary abnormalities (Table 2) comprises microscopic hematuria that at times may be macroscopic, sterile pyuria, and white blood cell casts. Red blood
cell casts have also been reported, albeit rarely, in AIN.
Eosinophiluria, with more than 1% of urinary leukocytes
positive by Hansel's stain, is suggestive evidence of AIN,
but can be seen in other forms of renal injury and inflammation. In a recent single-center study, the diagnostic accuracy of eosinophiluria in AIN was examined. Fifty-one
patients with various renal diseases, including 15 individu-

als with a confirmed diagnosis of AIN, were evaluated in

this study. The sensitivity of eosinophiluria in this setting
was 40% and the specificity was 72%, with a positive predictive value of only 38%. These data illustrate the association
of eosinophiluria with various morphologic forms of renal
inflammation and demonstrate that it is not a reliable indicator of AIN.
Mild proteinuria, generally less than 1 g/day, is frequently
observed in AIN. Nephrotic range proteinuria with acute
disease has been reported, however, with nephropathies in duced by nonsteroidal anti-inflammatory drugs (NSAIDs),
and rarely by ampicillin, rifampin, and interferon-a ther-

TABLE 2

TABLE I

Acute Interstitial Nephritis

Drugs Antibiotics (especially penicillin analogs,
cephalosporins,
sulfonamides, and rifampin) Nonsteroidal
anti-inflammatory drugs Diuretics (especially
thiazides and furosemide)
Infections
Direct infection of renal parenchyma
Associated with a systemic infection
Immunologic disorders Systemic
lupus erythematosus Sjogren's
syndrome Mixed essential
cryoglobulinemia Acute allograft
rejection
Idiopathic

Urinary sediment
Urinary protein
excretion

Laboratory Findings in Acute Interstitial

Nephritis
Erythrocytes, leukocytes
(eosinophils), leukocyte casts

Fractional excretion <1 g/day, rarely >1 g/day (NSAIDs) Usually

of sodium
>1
Proximal tubular
defects
Glucosuria, bicarbonaturia,
phosphatur ia, aminoaciduria, proximal
Distal tubular
RTA
defects
Medullary defects Hyperkalemia, sodium wasting, distal
RTA
Sodium wasting, urine-concentrating
defects
RTA, renal tubular acidosis.

40. Acute Interstitial Nephritis

apy. Serologic studies, such as anti-DNA antibodies, antinuclear antibodies (ANA), and complement levels, are typically normal in AIN, unless it occurs in the setting of a
systemic autoimmune disorder. Recent case reports also
note the presence of antineutrophil cytoplasmic antibodies
(ANCA), a serologic marker for systemic vasculitis, in
some patients during the acute phase of interstitial nephritis. Elevated pANCA titers have been observed in druginduced AIN (omeprazole, ciprofloxacin) and cANCA in
the tubulointerstitial nephritis and uveitis (TINU) syndrome. Of note, two cases of ciprofloxacin-induced AIN
also had evidence of necrotizing vasculitis on renal biopsy.
The clinical relevance of ANCA titers in AIN is unclear,
however, as the overall incidence, antigen specificity, and
pathogenicity of these antibodies have not been established. Urinary fractional excretion of sodium is greater
than
1 in many patients with AIN, but is not a reliable
diagnostic indicator. Biochemical abnormalities (Table 2)
reflective of the tubular damage induced by the inflamma
to ry process are also observed in these patients. The
pattern of tubular dysfunction observed varies with the
principal site of injury. Lesions affecting the proximal
tubule result in renal glucosuria, aminoaciduria, phosphaturia, uricosuria, and proximal renal tubular acidosis (type
2 RTA). Distal tubular lesions result in an inability to
acidify urine (type 1 RTA), secrete potassium, and regu
late sodium balance. Medullary lesions interfere with
maximal urinary concentration and promote polyuria. A
conside rable degree of overlap in these proximal and
distal abnormalities, however, may be apparent clinically.
Renal ultrasound in affected patients typically reveals
normal or enlarged kidneys, depending on the degree
of interstitial edema. Renal gallium scanning has been
advocated in some centers to distinguish AIN from other
causes of ARF, primarily acute tubular necrosis, but this
test is neither sensitive nor specific. In view of the
nonspecific nature of many of these clinical feature of
AIN, a definitive diagnosis can only be made through
renal biopsy.
CLINICAL COURSE AND THERAPY

The spectrum of renal dysfunction in AIN ranges from

mild, self-limited disease to oliguric renal failure requiring
dialysis therapy. As this renal lesion is generally reversible,
even when it presents with severe renal impairment, the
overall prognosis is quite favorable. Recovery of renal function may occur over weeks to several months. Some patients have persistent tubular defects or residual renal impairment, or both. Progression to end-stage renal disease,
however, has been reported with all forms of AIN. Moreover, a rapidly progressive fibrosing interstitial nephritis
has recently been described in clusters of patients in weight
loss programs who ingested Chinese herbal preparations
tainted with nephrotoxic plants (Aristolochia fangchi). Renal disease in all affected individuals was irreversible, with
many patients requiring dialysis therapy within 1 year of
presentation.

279

Clinical studies suggest that a less favorable prognosis

in AIN correlates with extensive interstitial infiltrates, interstitial fibrosis, tubular atrophy, and interstitial granuloma formation on renal biopsy. Other factors that correlate with persistent renal disease are advanced patient age,
preexisting renal disease, and a protracted course of oliguric ARF (>3 weeks). In the recent literature, chronic
renal insufficiency following a bout of AIN is most commonly induced by NSAIDs.
The therapy of AIN consists primarily of supportive
measures, after possible inciting factors such as drugs or
infections have been eliminated. The role of corticosteroids in treating AIN has not been clearly elucidated.
Rapid improvement and complete recovery of renal function following steroid therapy in several drug-induced
lesions have suggested their therapeutic utility in this
disorder. Two empirically derived steroid regimens have
been employed in drug (primarily antibiotic)-induced
AIN. One protocol consists of parenteral methylprednisolone (0.5 to 1.0 g) administration for 1 to 4 days, and
the other, more commonly employed regimen, consists
of high-dose prednisone therapy (1 mg kg'1 day"1)
given daily for 7 to 14 days. Some clinical investigators
have not corroborated the steroid responsiveness of this
renal lesion, however, and randomized controlled studies
have not been conducted. Steroids are not employed in
infection-related AIN, but may be useful in the treatment
of nephritogenic responses in systemic immunologic disorders. Although experimental mo dels have suggested a
disease-protective role for cyclophosphamide and cyclosporine in interstitial nephritis, similar studies have not
been conducted in human disease.
DISTINCT CAUSES OF AIN
Drug-Induced AIN

The list of drugs that reportedly induce AIN is quite

extensive (Table 3). Many of these cases of AIN are reports
of single cases, however, and have developed in patients
exposed to a number of different medications. Druginduced AIN is a rare idiosyncratic reaction that occurs in
a small subset of patients exposed to a particular medication. It is not dose dependent, and typically recurs on repeat
exposure to the same or closely related drug. As seen in
Table 3, implicated drugs have diverse chemical structures,
although within a class of related drugs, structural similarity
can lead to cross-reactive sensitivities. This has been observed particularly with /3-lactam drugs, in that penicillininduced nephropathies have been exacerbated with cephalosporin therapy. A few drugs, most notably penicillins,
rifampin, NSAIDs, and sulfonamide derivatives, account
for the majority of reported cases of AIN, and their characteristic features will be discussed below.
Penicillins

The /3-lactam antibiotics, predominantly the penicillins,

are the most common cause of drug-induced AIN. The

280

Catherine M. Meyers

TABLE 3

Drug- Induced Acute Interstitial Nephritis

Antibiotics
Penicillin analogs
Methicillin, ampicillin, penicillin, nafcillin, carbenicillin,
oxacillin, amoxicillin, mezlocillin, llucloxacillin
Cephalosporins Cephalothin, cefotetan, cephradine, cephalexin,
cefoxitin,
cefazolin, cefaclor, cefotaxime
Sulfonamide derivatives
Sulfamethoxazole, cotrimoxazole
Other antibiotics
Rifampin, gentamicin, kanamycin, vancomycin, acyclovir,
ciprofloxacin, aztreonam, erythromycin, azithromycin,
ethambutol, tetracyclines, nitrot'urantoin
Nonsteroidal anti-inflammatory drugs
Fenoprofen, ibuprofen, naproxen, indomethacin, tolmetin,
zomepirac, diflusinal, sulindac, phenylbutazone, aspirin,
phenacetin, mefenamic acid, 5-aminosalicylic acid
Diuretics
Thiazides, furosemide, triameterene, chlorthalidone
Miscellaneous Medications
Phenytoin, allopurinol, cimetidine, ranitidine, phenobarbital,
azathioprine, cyclosporine, aldomet. carbamazepine,
diazepam, phenylpropanolamine, captopril, clolfibrate, antiinterfcron, interleukin 2, anti-CD4 monoclonal antibodies,
ticlopidine, quinine, propylthiouracil, omeprazole,
streptokinase, Chinese herbs
Note: Drugs reported with greatest frequency are shown in
italics.

largest number of cases occurred with methicillin, which is

no longer used in clinical practice. AIN has been reported
with most of the penicillin analogs more routinely prescribed (Table 2), although with much lower incidences
than with methicillin. Renal failure in this setting has been
observed most commonly in older children and young
adults. The classic hypersensitivity triad of fever, rash, and
eosinophilia in the setting of ARF may occur in up to
30% of patients with /3-lactam-induced AIN. Oliguric renal
failure has been reported in approximately 30% of these
patients. Clinical studies suggest a beneficial role for steroids in this patient population, although as previously
stated, randomized controlled studies have not been performed. AIN has developed during treatment of a variety of
infections, such as endocarditis, osteomyelitis, pneumonia,
cellulitis, and abscesses. An underlying infection is clearly
not requisite for inducing this reaction, however, as several
patients given prophylactic antibiotics have subsequently
developed AIN.
NSAIDs

NSAIDs are among the most widely prescribed medications in clinical practice. They mediate a number of adverse
renal side effects, largely as a result of their inhibition of
renal prostaglandin synthesis, and are extensively discussed
in Chapter 42 of this text. AIN is a less commonly observed

side effect of these medications. Propionic acid derivatives

appear to cause a disproportionate number of cases (two
thirds of cases have been attributed to fenoprofen, ibupro fen, and naproxe n), although AIN has been reported with
most NSAIDs currently available (Table 3). Unlike other
drug-induced reactions, AIN occurs following long-term
exposure to the medication, and has been reported 2 weeks
to 18 months after initiation of therapy. Patients tend to
be nonoliguric females older than 60 years of age, who
generally lack systemic manifestations of a hypersensitivity
reaction. They may present with hypertension, edema, and
nephrotic range proteinuria (up to 80% of cases). Interesting histologic features of induced interstitial lesions are the
associated minimal change glomerulopathy and occasional
interstitial granuloma formation apparent on renal biopsy.
Renal disease generally improves after the drug is discontinued, with or without steroid therapy, although chronic
renal insufficiency and end stage renal disease are not infrequent complications.
Rifampin

Numerous cases of rifampin-induced AIN have occurred

during treatment of tuberculosis. Most of these cases have
developed with intermittent therapy or upon restarting
rifampin after a lapse in uneventful daily therapy. Patients
typically complain of flu-like symptoms such as fever,
chills, malaise, and headache. Unlike other drug-induced
lesions, flank pain and hypertension are common in this
form of AIN. Moreover, oliguric ARF occurs frequently,
and dialysis is required in approximately two thirds of
affected patients. In many cases, this reaction has occurred
within hours of a single dose of rifampin. Some patients
have developed thrombocytopenia, hemolysis, or abnormalities in liver function in addition to AIN. Histologically, evidence of acute tubular necrosis may be apparent
along with AIN. In a few cases, an associated proliferative glomerulonephritis has also been observed. Circulating antirifampin antibodies, as well as IgG deposition
along the TBM, have been reported in some affected
patients. As AIN has developed in patients receiving
concurrent rifampin and prednisone therapy, there is no
evidence to suggest that steroids play a therapeutic role
in this disease.
Sulfonamide Derivatives

Drug-induced AIN was first described in the setting of

Sulfonamide administration. The majority of cases of sulfonamide derivative-induced AIN occur with combination
sulfamethoxazole and trimethoprim therapy. Thiazides and
furosemide have also been associated with a few cases
of AIN, some of which have developed in patients with
preexisting renal disease. The associated hypersensitivity
triad of fever, rash, and eosinophilia with renal dysfunction
is variably present in affected patients. In addition to the
characteristic histologic features of AIN, some biopsies
have revealed a predominance of eosinophils within interstitial infiltrates, as well as interstitial granuloma formation.

281

40. Acute Interstitial Nephritis

Isolated case reports have suggested beneficial effects of

steroid therapy in treating this variant.
Infections

AIN was first described in the preantibiotic era in the

setting of diphtherial and streptococcal infections. It is
now apparent that AIN complicates the clinical course
of a number of bacterial, viral, fungal, and parasitic
infections, as listed in Table 4. This inflammatory response
within the kidney may occur as a result of direct renal
infection, i.e., pyelonephritis , or as a reaction to a systemic
infection. Pyelonephritis, the most common cause of acute
infectious interstitial nephritis, typically presents with
fever, costovertebral tenderness, dysuria, pyuria, bacteriuria, and leukocytosis. Renal function is unimp aired unless complicated by urinary tract obstruction. Characteris tic renal parenchymal lesions consist of focal areas of
neutrophils throughout the interstitium. Pyelonephritis
responds well to antibiotic therapy and is discussed more
extensively in Chapter 57 of this text. In contrast to
pyelonephritis, other infection-associated interstitial processes occur in the absence of urinary tract infection.
Interstitial infiltrates are frequently perivascular and comprised of mononuclear cells, predominantly T cell lymphocytes. As previously discussed, the pathogenesis of such
immune targeting in these infections is not well understood, although cross-reactive determinants may play a
role in immune recognition of interstitial structures.
Infection-associated interstitial nephritis is generally transient and renal function improves with appropriate therapy of the systemic illness; however, chronic progression
of renal insufficiency has been reported.
Immune Disorders

Although glomerulonephritis is the most common renal

manifestation of systemic immunologic disorders, predomi-

TABLE 4

Infections Associated with Acute Interstitial Nephritis

Bacterial infections
Streptococcus, diptheria, brucella, legionella. pneumococcus,
tuberculosis
Viral Infections
Epstein -Barr virus, cytomegalovirus, BK virus, Hantaan
virus, measles (rubeola), human immunodeficiency virus,
herpes simplex virus type 1
Fungal Infections
Candidiasis, histoplasmosis
Other Infections Toxoplasmosis, leishmaniasis,
schistosomiasis, Rocky
Mountain spotted fever, ehrlichiosis, malaria, mycoplasma,
leptospirosis, syphilis, ascaris lumbricoides
Note: Infections associated with direct renal infection are shown
in italics.

nant interstitial pathology can be seen in systemic lupus

erythematosus, Sjogren's syndrome, and mixed essential
cryoglobulinemia. Most affected patients present with nonoliguric renal failure, and biochemical evidence of tubular
dysfunction. In addition to the typical pathological features
of AIN, biopsies from many of these patients also reveal
immune-complex and complement deposition along the
TBM, and occasionally within interstitial vessels. Concurrent glomerular pathology may also be apparent. Standard
therapeutic modalities in these immunologic disorders consist of corticosteroids and cytotoxic agents. Acute renal
allograft rejection, a distinct subset of immunologic disorders, also induces acute interstitial inflammation. These
lesions are characterized by marked interstitial mononuclear infiltrates, consisting of recipient T cell lymphocytes.
Allograft rejection is treated with high-dose (pulse) steroids, as well as antibody therapy targeting activated recipient T cells, and is discussed further in Chapter 75 of this
text.
Idiopathic
In approximately 10 to 20% of biopsy-proven cases of
AIN, no precipitating cause is detected. Systemic manifestations of a hypersensitivity reaction are generally absent
in the majority of these idiopathic cases, which often pres ent with nonoliguric renal failure. One subset of these apparently idiopathic lesions, the TINU syndrome, has been
described in approximately 30 adolescent or adult females
since 1975. Anterior uveitis can precede, accompany, or
follow AIN. Bone marrow granuloma formation may be
another common feature of this syndrome. Although the
cause of TINU syndrome is not known, an autoimmune
nature is suggested by occasional positive serologies, such
as cANCA, rheumatoid factor, and ANA, in affected patients. Both renal and ocular changes respond to a brief
course of steroid therapy, although a few patients have
developed progressive renal disease.
Bibliography
Cameron JS: Immunologically mediated interstitial nephritis: primary and secondary. Adv Nephrol 18:207-248, 1989.
Colvin RB, Fang LST: Interstitial nephritis. In: T isher CC, Brenner
BM (eds) Renal Pathology with Clinical and Functional Correlations. Lippincott, Philadelphia, pp. 728-776, 1989.
Dodd S: The pathogenesis of tubulointerstitial disease and mechanisms of fibrosis. Curr Topics Pathol 88:51-67, 1995.
Heptinstall RH: Interstitial nephritis. A brief review. Am J Pathol 83:214-236,1976
Jones CL, Eddy AA: Tubulointerstitial nephritis. Fed Nephrol
6:572-586, 1992.
Kannerstein M: Histologic kidney changes in the common acute
infectious diseases. Am J Med Sci 203:65-73, 1942.
Kida H, Abe T, Tomosugi N, et al.: Prediction of the long-term
outcome in acute interstitial nephritis. Clin Nephrol 22:55-60,
1984.
Kleinknecht D: Interstitial nephritis, the nephrotic syndrome, and
chronic renal failure secondary to nonsteroidal anti-inflammatory drugs. Sem in Nephrol 15:228-235, 1995.

282

Florence N. Hutchison

Magil AB, Tyler M: Tubulointerstitial disease in lupus nephritis:

A morphometric study. Histopathology 8:81-87, 1984.
Neilson EG. Pathogenesis and therapy of interstitial nephritis.
Kidney Int 35:1257-1270, 1989.
Pusey CD, Saltissi D, Bloodworth L, et al.: Drug associated acute
interstitial nephritis: Clinical and pathologic features and the
response to high - dose steroid therapy. Q J Med 52:194-211,
1983.
Ruffing KA, Hoppes P, Blend A, et al.: Eosinophils in urine revisited. Clin Nephrol 41:163-166, 1994.

Shih DF, Korbet SM, Rydel JJ, et al.: Renal vasculitis associated
with ciprofloxacin. Am J Kidney Dis 26:516-519, 1995.
Simon AHR, Alves-Filho G, Ribeiro-Alve s MAVF: Acute
tubulointerstitial nephritis and uveitis with antineutrophil
cytoplasmic antibody. Am J Kidney Dis 28:124 - 127,
1996.
Vanherweghem JL, Depierreux M, Tielemans C et al.: Rapidly
progressive interstitial renal fibrosis in young women: Association with slimming regimen including Chinese herbs. Lancet
341:387-391, 1993.

MANAGEMENT OF ACUTE RENAL FAILURE

FLORENCE N. HUTCHISON

Morbidity and mortality from acute renal failure (ARF)

have improved significantly over past decades due in large
part to improved supportive care and early dialysis. Since
the kidney usually recovers from an episode of ARF within
days or weeks, the goal of management of ARF is to prevent the consequences of the loss of kidney function and
to prevent further renal injury. To develop a rational management plan, it is important to understand the normal
function of the kidney because the symptoms and complications of renal failure derive from the loss of these functions.
Homeostasis refers to the process by which the internal
body milieu is maintained in a constant state despite
marked changes in the external environment and variations
in the intake of solute, water, and food. The normal kidney
is critical to the maintenance of body homeostasis, particularly with regard to volume homeostasis, electrolyte balance, acid-base balance, and excretion of nitrogenous
products of protein metabolism. Loss of these functions
can result in the pulmonary edema, hyperkalemia, acidosis,
and uremia which characterize severe renal failure. Although the causes of ARF are myriad, and specific interventions may be indicated for certain types of ARF, the consequences of ARF can be prevented in most cases by good
conservative management, regardless of the etiology.
Therapy of ARF should be initiated as soon as renal
injury has been detected. Thus, the first step in management
is early recognition that renal failure is present. Since the
relationship between serum creatinine concentration and
glomerular filtration rate (GFR) is not linear, but exponen-

Primer on Kidney Diseases, Second Edition

tial, an increase in serum creatinine will not become clinically evident until the GFR is reduced to less than about
50 mL/min. Many episodes of ARF with less severe renal
dysfunction are not recognized and do not represent clinically significant events. However, the relationship between
GFR and serum creatinine also implies that any increase
in serum creatinine above the value previously determined
for a given patient represents significant loss of renal function. This point is particularly pertinent in cachectic or
elderly patients whose baseline serum creatinine may be
only 0.8 mg/dL because of the smaller amount of creatinine
generated from a reduced muscle mass. With an increase
to 1.5 mg/dL the serum creatinine may still be within the
range of normal values for most clinical laboratories, but
this represents a significant increase in serum creatinine
for this patient and indicates that ARF has occurred. It is
also critical to determine the cause of renal dysfunction
and to identify reversible causes of renal insufficiency such
as prerenal azotemia or urinary tract obstruction. The differential diagnosis of acute renal insufficiency is detailed
in Chapter 35.
MANAGEMENT OF VOLUME HOMEOSTASIS

Successful volume homeostasis permits maintenance of

a constant internal circulatory and extracellular volume
despite consumption of varying quantities of water and salt
and varying insensible loss of water. The presence of pedal
or sacral edema, or overt pulmonary edema in the setting
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

41. Management of Acute Renal Failure

of ARF implies that water and/or salt intake has exceeded

the capacity of the injured kidney to excrete the water and
salt load. This situation can be easily anticipated in the
oliguric (<400 mL/day of urine) or anuric (no urine output)
patient, but frequently complicates nonoliguric ARF as
well.
Most patients with ARF lose the ability to either concentrate or dilute the urine and will excrete a constant volume
of urine regardless of fluid intake. Likewise, the ability to
excrete a sodium load may be severely curtailed. As a
result, urine output and sodium excretion will rarely match
spontaneous fluid and salt intake and either volume overload or volume depletion can result. A typical example is
a patient with acute tubular necrosis whose urine output
is fixed at 1200 mL/day. If he receives parenteral nutrition
at a rate of 2000 mL/day along with intravenous antibiotics,
he will gradually develop volume overload and edema unless adjustments are made to the volume of fluid adminis tered. If parenteral nutrition is withheld and he is unable
to eat, he will become volume depleted unless parenteral
fluids are administered.
Management of volume homeostasis requires that the
patient be examined daily to assess the volume status. This
is accomplished by measurement of supine and standing
blood pressure and pulse to note orthostatic changes suggestive of volume depletion, examination of skin turgor
and hydration of mucous membranes, auscultation of the
lungs for evidence of pulmonary congestion, a general exam
for edema, detailed review of daily input and output records, and accurate measurement of serial daily weight
changes.
Each patient with ARF should have a specific individualized prescription for fluid and sodium intake. As a general
rule, the patient who is euvolemic on exam should be provided with a volume of fluid equal to daily urine output
with an additional 300 to 500 mL/day to replace insensible
water losses. A sodium intake of 2 g/day or less should be
prescribed. Patients with increased insensible fluid loss such
as those with severe diarrhea and large burns will have a
much larger fluid requirement. Since insensible fluid loss
cannot be accurately measured, it is imperative that a patient's volume status be assessed on a daily basis and the
fluid prescription modified as necessary. All oral fluid intake as well as intravenous fluids and medications should
be included in the patient's fluid prescription. The patient
with clinical evidence of volume overload should be restricted to a fluid intake less than the daily urine output.
A trial of diuretics is worthwhile since some cases of oliguric
ARF may be converted to a nonoliguric state. Low-dose
dopamine infusion (0.3 /u,g kg'1 min'1) has also been
advocated to increase urine output in oliguric ARF, but it
has not been shown to be efficacious in controlled trials.
Patients with clinical evidence of volume depletion should
be provided additional volume to achieve a euvolemic state.
It should be emphasized that the goal of volume management is to maintain euvolemia. If a patient is allowed to
remain hypovolemic because of concern that pulmonary
edema may occur, the sustained hypovolemia may worsen
renal injury or delay recovery from renal failure. Increased

283
fluid administration may be required during the polyuric
recovery phase of ARF.
Hypernatremia and hyponatremia are observed frequently in patients with ARF. Since abnormal serum sodium concentrations are caused by disorders of water metabolism, prevention of hypo- or hypernatremia is linked
to volume management. The capacity to modify water excretion is impaired in ARF, as demonstrated by the finding
of a persistantly isosmotic urine. Hyponatremia is most
common and results from an excess of free water relative
to solute, while hypernatremia results when free water intake is inadequate. Although excessive oral intake of water
can result in hyponatremia in the patient with ARF and
inadequate oral intake will produce hypernatremia, the
most common cause of either disorder in hospitalized patients with ARF is incorrect administration of intravenous
fluids. Hypotonic saline solutions are frequently selected
to avoid volume overload, but provide a large proportion
of electrolyte free water. For example, administering
1000 mL of 0.45% saline is equivalent to giving 500 mL of
isotonic (normal) saline and 500 mL of electrolyte free
water. In contrast, 1000 mL of 0.9% saline is isotonic and
provides no electrolyte free water. Most parenteral antibiotics are administered in 5% dextrose in water; their full
volume is electrolyte-free water. Another source of excess
free water intake is from enteral and parenteral feeding
solutions, which are formulated with a low sodium concentration. Although prepared enteral solutions may be isosmotic or hypertonic, the solute consists predominantly of
carbohydrates and protein that are taken up by cells and
metabolized leaving electrolyte free water behind. Parenteral nutritional solutions are usually hyperosmotic when
administered, but again the osmoles consist mainly of dextrose and amino acids which are metabolized leaving a
large quantity of solute-free water.
The volume of free water required for an individual
patient to maintain osmolar balance must be determined
empirically and will vary considerably depending on the
type and quantity of insensible fluid losses. For example,
burn injuries induce large volumes of insensible fluid loss,
but the fluid is isosmotic. In contrast, insensible losses from
diarrhea and perspiration contain little solute and may
dramatically increase the patient's electrolyte free water
requirements. In the absence of abnormally high insensible
fluid loss, a patient should be provided either orally or
intravenously with 300 to 500 mL/day of electrolyte free
water as part of the total volume prescription. In patients
receiving hypotonic enteral nutritional solutions, the serum
sodium concentration should be monitored regularly. If it
decreases in the absence of volume overload, sodium can
be added to the enteral solution as table salt.
In the intensive care unit, clinical assessment of a patient's volume status may be confounded by surgical
wounds, severe pneumonia, or edema due to altered capillary permeability. In this setting, measurements of central
venous pressure and pulmonary capillary wedge pressure
are important adjuncts to monitor the patient's volume
status. Frequently these patients are receiving multiple parenteral medications and parenteral nutrition which consti-

284

Florence N. Hutchison

tute a large obligatory volume load. Most parenteral medications can be given slowly in a concentrated solution to
minimize the volume administered and the parenteral nutrition prescription should be written to provide optimal
calories and protein in a minimum volume. Attention
should also be given to the solution in which drugs are
administered. Most drugs can be administered in normal
saline rather than dextrose and water if hyponatremia is
a problem.
MANAGEMENT OF ELECTROLYTE HOMEOSTASIS

Hyperkalemia can be a serious consequence of ARF.

Normally serum potassium concentration is very tightly
regulated by the shifts of potassium between the intra and extracellular compartments. The kidney contributes to
overall potassium homeostasis by excreting into the urine
the approximately 100 mEq of potassium that is ingested
in the average diet so that total body potassium remains
constant. Although shifts between intra- and extracellular
potassium may contribute to hyperkalemia in ARF, the
main cause of hyperkalemia in renal failure is excessive
intake of potassium relative to the injured kidney's reduced
excretory capacity.
Prevention of hyperkalemia in the patient with ARF can
usually be accomplished by restriction of potassium intake.
Since certain types of foods, such as fruits, chocolate, and
nuts, contain large quantities of potassium, eliminating
these foods from the diet is often adequate to prevent
hyperkalemia. The potassium content of the diet should be
routinely specified for the patient with ARF and typically
should be limited to less than 50 mEq/day. Potassium
should be omitted from parenteral fluids. It is important
not to overlook other nondietary exogenous sources of
potassium. These include drugs such as potassium penicillin
G, saturated solution of potassium iodide (SSKI), and potassium phosphate, dietary additives such as salt substitutes, which contain potassium chloride or citrate, and intravenous fluid such as lactated Ringer's and parenteral
nutrition solutions, which typically are formulated with potassium chloride or potassium phosphate. Finally, drugs
that impair renal potassium excretion, such as potassiumsparing diuretics, nonsteroidal anti-inflammatory agents,
and angiotensin-converting enzyme inhibitors, should be
avoided.
If dietary restriction of potassium is inadequate to prevent hyperkalemia, potassium binding resins such as sodium polystyrene sulfonate (Kayexalate) can be utilized.
The binding resin is usually administered orally as 15 to
30 g in a solution of 20% sorbitol. It exchanges sodium
for potassium in the bowel, there by increasing intestinal
excretion of potassium. Potassium binding resins can also
be administered rectally as a retention enema. This is particularly valuable if an ileus is present; the time of onset
of action will be faster by this route. Although the resin is
quite effective in removing potassium, it should not be
considered a substitute for dietary potassium restriction.
Since sodium is exchanged for potassium, chronic use of
these compounds delivers a large sodium load and can

worsen volume overload or hypernatremia. Additionally,

the binding resin causes constipation and should not be
used orally in patients with intestinal ileus or obstruction.
Diarrhea is a predictable consequence when the binding
resin is administered repeatedly with sorbitol and may complicate acidosis by causing intestinal bicarbonate loss. The
diarrhea may be intolerable to the patient.
Hyperkalemia that develops despite effective restriction
of potassium intake and that cannot be easily corrected
with potassium binding resin is an indication for dialysis,
but should also be a cue to initiate an investigation for
the source of the potassium. If diet or drugs cannot be
implicated, one should consider endogenous causes of hyperkalemia such as hyperglycemia and insulinopenia, severe acidosis, hemolysis, rhabdomyolysis, ischemic tissue
injury, or other causes of myonecrosis (see Chapter 13).
Since potassium-lowering therapies, including dialysis,
have only a transient effect, appropriate management of
hyperkalemia requires identification and specific treatment
of the cause of hyperkale mia.
Hypocalcemia frequently complicates ARF, but clinically significant hypocalcemia is rare, as is hypomagnesemia. However, failure to recognize these disorder can produce life-threatening tetany and cardiac arrhythmias. Most
reported cases of symptomatic hypocalcemia in patients
with ARF have been attributed to hypomagnesemia caused
by urinary magnesium wasting in association with cisplatin,
amphotericin B, or aminoglycosides. Although decreased
synthesis of 1,25-dihydroxyvitamin D by the injured kidney
may contribute to hypocalcemia by reducing intestinal calcium absorption, hypomagnesemia inhibits synthesis and
release of parathyroid hormone and functional hypoparathyroidism is a primary mechanism responsible for hypocalcemia. Hypocalcemia may als o result as a consequence of
transfusion of large quantities of blood products preserved
in citrate or as a consequence of phosphorus administration. Prevention and management of this problem can be
accomplished by generous supplementation of both electrolytes. Calcium may be supplemented orally, usually 3
to 4 g/day in divided doses, but symptomatic hypocalcemia
should be treated with parenteral calcium. Calcium gluconate and calcium chloride are available as 10% solutions
containing approximately 10 and 28 mg/mL, respectively.
Either solution should be diluted in saline or dextrose in
water and administered slowly through a peripheral vein
to avoid cardiac toxicity. Magnesium can be repleted orally
as magnesium oxide or parenterally as intramuscular magnesium sulfate, usually 2 g as a single dose.
Phosphorus ingested in the diet and in drugs accumulates
in renal failure and may result in hyperphosphatemia. In
patients with ARF this is rarely of clinical consequence
except when very high serum phosphorus levels contribute
to hypocalcemia, as in rhabdomyolysis, tumor lysis syndrome, and hypercatabolic states. Hyperphosphatemia can
be effectively prevented by dietary protein restriction as
described below and by oral administration of aluminum
hydroxide gels or calcium carbonate or calcium acetate
with meals.

285

41. Management of Acute Renal Failure

MANAGEMENT OF ACID- BASE HOMEOSTASIS

Consumption of a normal American diet by an individual

weighing 70 kg produces 60 to 70 mEq of acid daily which
must be excreted by the kidney to maintain normal acidbase balance. Accumulation of these acids in excess of the
body's buffering capacity in patients with ARF can produce
an anion gap acidosis. However, it is unusual for acidosis
to be severe or to occur early in the course of ARF, and
acidosis should not be attributed exclusively to ARF until
other causes of acidosis have been eliminated, e.g., lactic
acidosis, ketoacidosis. The presence of ARF may complicate acidosis of any cause if the kidney is unable to increase
acid excretion in response to an acid load or is unable to
reclaim bicarbonate filtered into tubular fluid. Acidosis not
due specifically to ARF should be treated by removing the
source of acid generation or bicarbonate loss as would be
done in patients without ARF and is discussed in Chapter 9.
Since the majority of the ingested acid load is derived
from protein metabolism, dietary protein restriction will
slow the development of the acidosis of ARF. A diet containing approximately 0.8 g/kg body wt of good quality
protein usually satisfies the nutritional requirement for protein and minimizes acid production. Restriction of nonpro tein calories may worsen acidosis since body muscle pro teins will be catabolized for the total energy requirement in
the absence of other dietary energy sources. Many patients
with ARF are anorexic yet have increased caloric requirements due to physiological stress, so inadequate calorie
intake is common and should be assessed carefully. The
assistance of a skilled dietitian can be invaluable in determining the patient's true calorie and protein requirements.
If acidosis develops despite dietary protein restriction or
if protein restriction is not desired because of hypercatabolism with sepsis, burn, or a postoperative state, a base equivalent can be administered to buffer the acidosis. Sodium
bicarbonate is most commonly used, but sodium acetate
can also be used in parenteral solutions. Both compounds
have the disadvantage of containing sodium and may
worsen volume overload. Bicarbonate can be administered
orally as tablets on a regular basis to compensate for continuing generation of acid. Parenteral bicarbonate should
be administered as an isotonic solution. A solution containing approximately 150 mEq/L can be mixed by adding 3
ampoules of premixed hypertonic sodium bicarbonate
(50 mEq/50 mL) to 1000 mL of 5% dextrose in water.
Use of these ampoules, which contain 1000 mEq sodium
bicarbonate/L without dilution, is not recommended unless
serum pH is less than 7.1 and then only as a temporizing
measure until the cause of the acidosis can be reversed or
dialysis initiated since repeated administration will produce hypernatremia.
MANAGEMENT OF UREMIA

Uremia is a syndrome resulting from the accumulation

of nitrogenous products of protein metabolism that are
normally excreted by the kidney. Early symptoms of uremia

are fatigue, lethargy, mental dullness, hiccups, anorexia,

and nausea. It may be difficult to determine whether these
symptoms are due to ARF or to intercurrent illness. More
serious symptoms are myoclonus, confusion, delirium or
coma, seizures, and pericarditis. The blood urea nitrogen
(BUN) concentration correlates best with symptoms of
uremia, but there is considerable variability among patients. As a general rule, few patients have symptoms when
the BUN is less than 70 mg/dL and most patients will
experience some symptoms when the BUN is greater than
100 mg/dL.
Since the accumulation of nitrogenous waste products is
dependent on dietary protein intake, the degree of loss of
renal function, and the duration of renal failure, it is not
possible to predict which patients will develop uremic
symptoms. In most cases the physican cannot alter either
the degree of renal insufficiency or the duration of ARF,
and must rely on protein restriction to prevent uremia and
dialysis to ameliorate uremic symptoms. As noted previously, a protein intake of 0.6 to 0.8 g/kg body wt is considered the minimum amount adequate to supply nutritional
requirements for protein.
NUTRITIONAL MANAGEMENT IN ARF

Most episodes of ARF occur as a complication of other

serious illness or injury and studies have clearly demonstrated that adequate nutritional support substantially improves survival in patients in critical care units. Unfortunately, the dietary restrictions required for management
of ARF, as well as the anorexia of uremia, may be counterproductive of efforts to optimize nutritional support. For
example, very low protein diets (20 g protein/day) have
been advocated to prevent uremia and avoid dialysis. However, the possible benefits of avoiding short-term dialysis
by severe protein restriction must be weighed against the
potential increase in morbidity and mortality due to impaired nutrition. This level of protein restriction is not recommended.
Parenteral nutrition is frequently necessary in seriously
ill patients with ARF, but the composition of the parenteral
solution must be modified in accordance with the loss of
renal function. The minimum recommended protein intake
in patients with ARF is 0.6 to 0.8 g kg body wt" 1 day" 1
and may be considerably higher in very ill hypercatabolic
patients. The assistance of an experienced nutritionist to
calculate protein catabolic rate and calorie requirements
is invaluable in estimating actual protein requirements so
that excessive or inadequate amino acid administration can
be avoided. Carbohydrate and lipid should be maximized
with a target of providing 30 to 35 kcal kg body wt"1
day"1. Provision of adequate carbohydrate and lipid calories will reduce protein requirements and limit symptoms
of ARF. Since protein metabolism is the source of endogenous acid generation in renal failure, a base should be given
in an amount adequate to buffer the acid generated from
metabolism of the amino acids. Bicarbonate cannot be
added to parenteral nutrition solutions without causing
precipitation of calcium salts, so sodium acetate is used as

286

Florence N. Hutchison

a base equivalent. Since approximately 60 to 80 mEq of

acid is generated each day, a similar amount of sodium
acetate should be administered. Fluid volume should be
limited to the minimum required to deliver adequate caloric intake. The sodium concentration (sodium chloride +
sodium acetate) should be adjusted to provide approximately 300 to 500 mL as electrolyte-free water as discussed
previously. For example, if a patient is given 2000 mL/day
the total sodium concentration should be 100 mEq/L.
This is approximately equivalent to 1500 mL of isotonic
fluid and 500 mL of free water. If the patient is receiving
1000 mL/day, a sodium concentration of approximately
75 mEq/L will provide the equivalent of 500 mL of isotonic
fluid and 500 mL of electrolyte-free water. In general, potassium, magnesium, and phosphorus should not be added
to parenteral nutrition solutions in patients with ARF. If
a deficiency of one of these electrolytes exists, it is better
to replete that solute specifically than to administer potentially toxic solute on a continuous basis.
DIALYSIS IN ARF

In ARF, the role of dialysis is to prevent morbidity associated with complications of ARF and provide temporary
support until the renal insufficiency resolves. The decision
to initiate dialysis and the frequency of dialysis are based
on the patient's clinical condition rather than a particular
numerical value of BUN or serum creatinine concentration.
Dialysis is indicated in ARF for management of specific
problems. Pericarditis and other uremic symptoms can be
resolved only with dialysis and are considered absolute
indications for dialysis. Other problems, such as volume
overload, hyperkalemia, and acidosis, are considered relative indications for dialysis; that is, dialysis should be instituted when conservative management has failed or is not
practical. For exa mple, an oliguric or anuric patient who
requires mechanical ventilation for volume overload pulmonary edema should be dialyzed urgently since fluid restriction and diuretics will not be effective. The indications
for dialysis are summarized in Table 1.
Since dialysis is a temporizing measure in this setting,
the dialysis prescription may be considerably different from
that typically used for a patient on chronic dialysis. For

TABLE I

Indications for Dialysis

Absolute indicationsDialysis is the only possible treatment

Uremic symptoms
Uremic pericarditis Relative indicationsDialysis will be
required if conservative
management of these abnormalities is unlikely to be
successful or is contraindicated
Volume overload

Hyperkalemia
Metabolic acidosis
Other electrolyte abnormalities

example, in contrast to the chronic dialysis patient who

receives 3 to 4 hours of hemodialysis three times a week,
hypercatabolic oliguric ICU patients may require daily dialysis to limit uremic symptoms and to remove the fluid
administered with parenteral nutrition and medications.
Or a patient who presents with ARF and life-threatening
hyperkalemia may require a single dialysis to correct the
hyperkalemia and may then be maintained successfully
with conservative management. The decision to initiate
dialysis must be weighed against the risks associated with
large bore dialysis catheter placement and dialysis. Bleeding, infection, and pneumothorax may complicate catheter
placement, and dialysis may induce hypotension, mental
confusion, and cardiac arrythmias, particularly in unstable
critically ill patients.
The details of the dialysis prescription are beyond the
scope of this discussion, but it is worthwhile to mention
several variations on hemodialysis that are used routinely
in patients with ARF. Conventional intermittent hemodialysis is most commonly used. Continuous renal replacement
therapy may be preferred in patients whose major complication is volume overload, whose medical regimen requires
ongoing administration of large volumes of fluid, or whose
blood pressure is too unstable to permit hemodialysis. In
its simplest form, this procedure provides volume removal
only, but variants permit successful solute clearance (dialysis) as well. Peritoneal dialysis is also an effective modality
for management of ARF and can provide both fluid removal and clearance of solute. Dialysis is discussed extensively in Chapters 63 and 64. The decision to initiate dialysis
and the selection of the most appropriate mode of therapy
requires the participation of a nephrologist experienced in
these procedures.
DRUG MANAGEMENT IN ARF

Many frequently used drugs are eliminated through the

kidney and doses must be modified in patients with ARF.
In the setting of chronic renal failure, the physician can
reliably estimate GFR on the basis of the patient's steadystate serum creatinine concentration and make adjustments
in drug doses, but this is not the case with ARF. When GFR
falls abruptly, the serum creatinine concentration rises as
creatinine accumulates in the body until a new steady-state
level is reached. Thus, in ARF, the rate of increase in
serum creatinine concentration is a more accurate indicator of the severity of renal failure than the serum creatinine concentration on a given day. Serum creatinine is
also dependent on body muscle mass, so an increase of
0.5 mg dL~' day'1 in an elderly patient with reduced
muscle mass may indicate that GFR is <10 mL/min, but
a similar rate of increase in serum creatinine concentration
in a muscular young athelete may indicate that GFR is
relatively well preserved at 30 to 40 mL/min.
A vital component of management of the patient with
ARF is a review of all medications to identify those that
are excreted in the urine and, especially, those that are
toxic at high concentrations, such as digitalis compounds,
or that may cause further renal injury, such as aminoglyco-

41. Management of Acute Renal Failure

TABLE 2
Guidelines to Management of Acute Renal Failure
1. What is the patient's volume status?
Normal:
Fluid intake = urine output + 300-500 mL/day
Sodium intake = 2 g/day
Overloaded: Fluid intake < urine output
Sodium intake < 2 g/day
Try loop diuretic
Consider dialysis
Depleted: Restore volume with isotonic saline, then
prescribe fluid intake = urine output +
300-500 mL/day
Sodium intake = 2 g/day
2. Is the patient hyperkalemic?
No: Potassium intake = 50 mEq/day
Yes: Look for source of potassium
Eliminate parenteral potassium
Reduce dietary potassium intake to < 50 mEq/day
Potassium binding resin
Consider dialysis
3. Is the patient acidemic?
No: Protein intake = 0.8 g k"1 day"1 , 30 kcal- k" 1 day"1
Yes: Look for cause of acidosis
Reduce protein intake to 0.6 gm kg" 1 day"1 ; maintain
30 kcal kg"1 day"1
Oral bicarbonate or isotonic intravenous bicarbonate
Consider dialysis
4. Is the patient ureinic?
No: Protein intake = 0.8 g kg" 1 day" 1 , 30 kcal kg" 1
day"1 Yes: Reduce protein intake to 0.6 g kg" 1 day"1 ;
maintain
30 kcal kg"1 day"1 Check for
gastrointestinal bleedin g Consider
dialysis
5. Is the patient receiving medication?
No: Check any new medications for toxicity and adjust doses
Yes: Stop nephrotoxic drugs
Adjust medication doses
Check drug levels
6. Is the patient's nutrition adequate?
No: Provide balanced nut rition with 30-35 kcal kg" 1
day"1 and 0.6-0.8 g kg" 1 day"1 protein Assess
need for enteral or parenteral nutrition Yes: Reassess
periodically

sides. Compounds containing magnesium and phosphorus

are commonly used as antacids and cathartics and can induce hypermagnesemia and hypocalcemia in patients with

287

ARF. Aminoglycoside antibiotics and most vasopressor

agents are removed with dialysis and dosing must be adjusted to compensate for dialytic losses. A variety of texts
and handbooks provide modified dosing guidelines for specific drugs for use in patients with renal failure and in
dialysis patients. Measurements of creatinine clearance using the average of the serum creatinine concentrations measured before and at the end of the urine collection and
serum drug levels are also critical adjuncts to appropriate
management. Drug prescribing in renal failure is discussed
further in Chapter 43.
Table 2 is a checklist providing guidelines for basic management of the patient with ARF. Good management requires early recognition of renal failure, anticipation and
prevention of potential problems, and ongoing evaluation
of the patient's clinical condition.
Bibliography
Denton MA, Chertow GM, Brady HR: "Renal- dose" dopamine
for the treatment of acute renal failure: Scientific rationale,
experimental studies and clinical trials. Kidney Int 49:4 -14,
1996.
Feld LG, Cachero S, Springate JE: Fluid needs in acute renal
failure. Pediatr Clin North Am 37:337-350, 1990.
Centric A, Cledes J: Immediate and long term prognosis in acute
renal failure in the elderly. Nephrol Dial Transplant 6:86-90,
1991.
Kierdo H: Continuous versus intermittent treatment: Clinical results in acute renal failure. In: Siebert, HG, Mann, H, Stummvoll,
HK (eds) Continuous Hemofiltration. Contrib. Nephrol., pp.
1-12, 1991.
Manian FA, Stone WJ, Alford RH: Adverse antibiotic effects
associated with renal insufficiency. Rev Infect Dis 12:236-249,
1990.
Norton DF, Franklin SS: Acute renal failure in the critical care
setting. Acute Care 13:127-156, 1987.
Pascual J, Orofino L, Liano F, et al.: Incidence and prognosis of
acute renal failure in older patients. / Am Geriatr Soc 38:2530, 1990.
Schetz M, Lauwers PM, Ferdinande P: Extracorporeal treatment
of acute renal failure in the intensive care unit : a critical view.
Intensive Care Med 15:349-357, 1989.
Schneeweiss B, Graninger W, Stockenhuber F, et al.: Energy metabolism in acute and chronic renal failure. Am J Clin Nutr
52:596-601, 1990.
Schrier RW, Gardenschwartz MH, Burke TJ: Acute renal failure:
Pathogenesis, diagnosis, and management. Adv Nephrol 10:213240, 1981.
Speigel DM, Ullian ME, Zerbe GO, Berl T: Determinants of
survival and recovery in acute renal failure patients dialyzed in
intensive-care units. Am J Nephrol 11:44-47, 1991.

SECTION 6
DRUGS AND THE KIDNEY

NONSTEROIDAL ANTI-INFLAMMATORY
DRUGS AND THE KIDNEY
VARDAMAN M. BUCKALEW, JR.

The American public consumes billions of doses of overthe-counter (OTC) and prescription analgesics each year.
These consist primarily of drugs from two distinct classes:
(1) aspirin and other nonsteroidal anti-inflammatory drugs
(NSAIDs) and (2) acetaminophen and other nonnarcotic
analgesics. These analgesics are associated with adverse
events involving a number of organs, including the kidney.
Although the absolute event rate is probably low, the renal
effects of these drugs are important because of the large
numbers of individuals exposed to them.
Adverse renal events fall into two categories: those that
occur acutely in individuals taking NSAIDs, and a chronic
renal disease, analgesic nephropathy (AN), which occurs
in individuals who consume analgesics habitually. This
chapter discusses the epidemiology, pathophysiology, diagnosis, and treatment of these two types of adverse events.

control group given only acetaminophen. In a multivariate

analysis, the only risk factors for NSAID-associated acute
renal failure were age greater than 65 and coronary artery
disease. In a study of an elderly population (mean age 87
years), 15 of 114 patients (13%) given NSAIDs developed
a greater than 50% increase in serum urea nitrogen level
(but no change in serum creatinine), which was reversible
on stopping the drug. Multivariate analysis showed that
concomitant use of loop diuretics and higher doses of drug
correlated with acute renal failure.
Mechanism

Decreased Renal Blood Flow. The most common type of

NSAID-induced acute renal failure is due to decreased
synthesis of renal vasodilator prostaglandins (PGs) PGE 2
and PGI2, leading to decreased renal blood flow and glomerular filtration rate (GFR). Renal blood flow is not dependent on these PGs in normal individuals. However,
renal blood flow becomes dependent on their intrarenal
production in certain clinical states, making these individuals susceptible to NSAID-induced acute renal failure
(Table 1).
In volume depletion, including decreased effective volume in generalized edema, traumatic or septic shock, and
general anesthesia, increased PG synthesis is probably stimulated by the renal vasoconstrictor effects of sympathetic
nervous system and renin-angiotensin system activation.
The stimulus for PG production is not so clear when renal
mass is reduced, as occurs in renal transplantation and
other renal diseases. Thus, patients with renal disease of
any etiology may develop NSAID-induced renal failure,
although those with immune-mediated glomerulonephritis
and urinary obstruction may be particularly susceptible.
The well-known decrease in renal blood flow and GFR
with age may underlie the increased susceptibility of the
elderly to this condition. However, there may be other
factors which have not yet been identified. The association
of NSAID-induced acute renal failure with diuretic use
may be related to the edema states for which these drugs
are used or to their ability to stimulate vasodilator PG
production. Triamterene may be a more potent stimulator
of vasodilator PG production than other diuretics for reasons which are not clear, and its use with NSAIDs is not

ACUTE EFFECTS FROM NONSTEROIDAL

ANTI-INFLAMMATORY DRUGS
Acute Renal Failure
Epidemiology

The incidence and prevalence of acute renal failure due

to NSAIDs are not well documented. Available estimates
vary widely depending on the methodology used, the setting in which the study was done, and the population observed. A search of a data base from a large health maintenance organization in which more than 100,000 outpatients
were exposed to NSAIDs over an 11-year period revealed
that none were hospitalized for any of their known renal
complications. While the methodology used in this study
would not have detected adverse events treated without
hospitalization, these results indicate a very low overall
rate of acute renal failure or any other serious complication,
at least in this urban, largely middle-class population.
Two studies provide data on the incidence of acute renal
failure in hospitalized patients. In a city hospital that serves
primarily an indigent population, acute renal failure occurred in 343 of 1908 inpatients (18%) receiving ibuprofen.
In most patients, there was a clinically insignificant rise in
serum creatinine, and the incidence of severe acute renal
failure was only 0.9%. Furthermore, acute renal failure was
no more likely to occur in patients given NSAIDs than in a

Primer on Kidney Diseases, Second Edition

291

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

292

Vardaman M. Buckalew, Jr.

TABLE I

Predisposing Factors for NSAID-lnduced

Hemodynamic Acute Renal Failure
Decreased effective blood volume
Congestive heart failure
Cirrhosis
Nephrotic syndrome
Anesthesia Shock
Decreased absolute blood volume
Hemorrhage Sodium and water
depletion
Chronic renal failure
Acute renal failure due to contrast and obstruction
Drugs Diuret ics
Cyclosporine
Advanced age Renal

available OTC (ibuprofen, naproxen, and ketoprofen). All

NSAIDs, including aspirin, have the potential for causing
acute, hemodynamic renal failure. The likelihood of an
individual drug causing acute renal failure is related to its
potential for inhibiting renal PG synthesis. Of the commonly used agents, indomethacin is the most potent, and
aspirin the least. Some studies have suggested that sulindac,
an indoleacetic acid derivative, may have lower toxicity.
This drug does have a unique metabolism, in that the active
metabolite formed from the prodrug is inactivated by an
intrarenal enzyme system. The renal sparing effect of sulindac, if it exists, is not absolute, however, and cases of acute
renal failure have been reported. This has been explained
as being due to individual differences in the rate at which
the metabolite is inactivated or to high plasma levels of
this metabolite which may occur after 10 days of therapy
in patients with renal failure.

transplantation
Alterations in Electrolyte and Water Excretion

recommended. Cyclosporine increases urinary excretion of

vasodilator PGs, probably because of its renal vasoconstrictor effect. As a result, the drug increases the risk of NSAIDinduced acute renal failure in renal and heart transplant
patients as well as nontransplant patients receiving it.
In most cases of hemodynamic acute renal failure, the
decrease in renal blood flow is purely functional and recovery is rapid on stopping the drug. In some cases, however,
renal biopsy has demonstrated histologic changes of acute
tubular necrosis, the presence of which may delay recovery.
Interstitial Nephritis (IN). This complication, which appears to be an idiosyncratic reaction, particularly to the
propionic acid derivatives (ibuprofen, naproxen, and fenoprofen), is associated with the nephrotic syndrome in about
90% of cases. In contrast to acute interstitial nephritis
(AIN) associated with other drugs, NSAID-induced AIN
is associated with a low incidence of hypersensitivity symp toms, eosinophilia, and eosinophiluria and exhibits a predominance of lymphocytes rather than eosinophils in the
renal interstitium. This latter phenomenon may explain
why NSAID-induced IN is virtually the only type of IN
that is accompanied by nephrotic syndrome since the glomerular lesion is thought to be due to inflammatory mediators such as leukotrienes produced by the interstitial lymphocytes.
In patients with the nephrotic syndrome, minimal change
disease is usually found on renal biopsy. A few cases of
minimal change glomerulopathy without interstitial nephritis have been observed after long-term (>3 months) use
of NSAIDs. This entity has been reported with propionic
acid and indoleacetic acid derivatives.

Causative Agents
Seven chemical classes of NSAIDs are available in oral
form. At this time only the propionic acid derivatives are

NSAIDs may cause several alterations in water and electrolyte excretion. These effects are due either directly or
indirectly to inhibition of PG synthesis .
Sodium Excretion

PGE2 and PGI 2 increase renal sodium excretion by a

variety of mechanisms, including renal vasodilation and
direct inhibition of tubular sodium reabsorption. It is not
surprising, therefore, that inhibition of PG synthesis by
NSAIDs causes decreased renal sodium excretion. This
effect, however, is relatively minor and is not often clinically significant. Because most diuretics stimulate vasodilator PG production, NSAIDs may interfere with the natriuretic effect of diuretics in both edema states and in
hypertension. In the latter, this effect may cause blood
pressure to be a few mmHg higher than usual.
Potassium Excretion

PGE2 and PGI 2 increase renin release by a direct effect

on the juxtaglomerular cells. Consequently, NSAIDs have
been reported to cause hyperkalemia by inducing hyporeninemic hypoaldosteronism. A secondary mechanism for this
effect may be a decrease in sodium delivery to the distal
tubule and a decrease in urine flow. Risk factors for this
adverse event include renal failure, preexisting hypoaldosteronism, advanced age, and the concomitant use of drugs
that interfere with potassium excretion such as potassiumsparing diuretics and angiotensin -converting enzyme inhibitors.
Water Excretion

PGE2 facilitates water excretion by decreasing Na + reabsorption in the loop of Henle and by inhibiting the hydroosmotic effect of vasopressin. Accordingly, NSAID adminis tration may interfere with water excretion, causing the
development of hyponatremia. Too few cases have been
reported to identify the risk factors for this rare complication with certainty. However, two groups that may be at

42. Nonsteroidal Anti -inflammatory Drugs and the Kidney

risk are the elderly and infants treated with NSAIDs for
patent ductus arteriosus.
Management
In evaluating patients presenting with any of the adverse
events listed above, NSAIDs should be suspected as an
etiologic agent. It is important to remember that these
drugs may be used either as self-medication or by prescription. The index of suspicion should be especially high in
the groups at special risk, particularly the elderly and those
with preexisting renal disease, congestive heart failure, and
hepatic cirrhosis. A history of use of drugs known to increase the risk of NSAID complications should be sought.
Individuals with predisposing factors should be discouraged from taking NSAIDs. Patients known to be taking
NSAIDs regularly, especially those with increased risk and
those taking large doses, should have their renal function
checked frequently. The drugs should be discontinued with
the appearance of any renal sids effects. Concomitant use
of NSAIDs and an orally active PGEi analog, misoprostol,
has been reported to blunt the decrease in renal blood flow
in some patients with renal failure. Misoprostol has also
been shown to reduce the renal vasoconstrictor effects of
cyclosporine and could reduce the risk of NSAID-induced
acute renal failure in transplantation. However, the place of
this drug in the long-term management of these conditions
remains to be determined.
Acute renal failure from hemodynamic factors should
be differentiated from that due to interstitial nephritis.
This can usually be done with a urinalysis, because 90% of
patients with the latter have heavy proteinuria. Acute renal
failure due to hemodynamic factors should reverse rapidly,
usually within 3 to 5 days. Occasionally, recovery is delayed
by the development of acute tubular necrosis, a condition
that may be suspected on the basis of the typical urinary
sediment. Recovery will also be slow when acute renal
failure is due to interstitial nephritis. Steroids have been
reported to improve this condition, but it may also remit
spontaneously. No guidelines have yet been developed for
steroid use. Renal biopsy may be indicated in patients
whose renal failure does not respond quickly to cessation
of the drug.
ANALGESIC NEPHROPATHY

It can be argued that the discipline of renal epidemiology

began with the studies in the early 1950s of the association
between regular consumption of analgesics mixtures and
chronic renal failure. These studies suggested that the daily
ingestion of analgesics containing aspirin and phenacetin
caused a syndrome of chronic renal failure due to what
was then called "chronic pyelonephritis" and which we
now call "chronic tubulointerstitial nephropathy." Over
the intervening 40 years, the literature on this entity has
multiplied so that today there is general agreement among
nephrologists and epidemiologists about its principal features. When the major offending agent was shown to be
phenacetin, this popular analgesic was removed from the

293
market in most countries around the world. Despite this
attempt to prevent analgesic nephropathy (AN), it continues to be a problem even in countries where phenacetin
is no longer available, including the United States. This
suggests that phenacetin was not the sole cause of the
condition and indicates the need for further investigations
and perhaps additional public health measures.
Epidemiology
The Syndrome of Ha bitual Analgesic Use
Numerous studies in many countries around the world
have documented the existence of a population of individuals who habitually use OTC analgesics for a variety of selfdiagnosed indications (Table 2). Estimates of the prevalence of habitual analgesic consumption, defined as daily
ingestion for at least 1 year, vary widely geographically
from as low as 2% to as high as 30%. The explanation for
this wide variation is not clear. The highest prevalence is
among industrial workers and hospital patients, and the
lowest in community-based populations. Within these populations women are more likely to be habitual users than
men. The nature of the preparations available has some
influence, with the highest prevalence being in those areas
where phenacetin was most popular. Inclusion of caffeine
or codeine in the preparations may also increase habitual
use. The effect of the recent changes in the pharmacopeia
of OTC analgesics on habitual consumption has not yet
been studied.
AN occurs in a subset of those who use analgesics habitually. Thus, the geographical variation in the prevalence of
AN noted above is due mainly to the geographical variation
in the prevalence of habitual analgesic use.
Which Analgesics Are Used Habitually?
The type of preparations available for OTC consumption
vary with time and location. In the United States, two major
changes have been made in the makeup of OTC analgesics:
phenacetin was removed from the market in late 1983, and
the first nonaspirin NSAID, ibuprofen, became available
in 1984. These alterations have changed the prevalence of
drugs used by habitual consumers (Table 3). The prevalence of aspirin use has declined, but that of total NSAID
use (aspirin plus nonaspirin NSAIDs) has actually risen
slightly. In addition, the increment in acetaminophen use
is almost equal to the prevalence of phenacetin use when
it was available.
The risk for developing AN currently must be evaluated
in the light of this changing pattern of analgesic consump -

TABLE 2

Indications for Habitual Analgesic Use

Headache
Arthritis Sleep
aid Backache
Don't feel well

294

Vardaman M. Buckalew, Jr.

TABLE 3

TABLE 4

Prevalence of Habitual Use of Individual Analgesics in the

United States

Risk of Renal Disease from Individual Analgesics

1978-1979 (%)
Aspirin
NSAIDs

Acetaminophen
Phenacetin

55
0
19
26

1980-1982 (%) 1991 (%)

38
22
20
20

39
23
38
0

tion. Some authors have assumed, based on the decline in

prevalence of AN observed in some countries following
the removal of phenacetin from the market, that AN was
due entirely to phenacetin and will therefore disappear.
There is good reason to doubt this optimistic scenario.

Relative risk"
Drug
Phenacetin
Acetaminophen

ESRD

2.66-19.05
(n = 5)
2.1, 4.06

Chronic renal failure

5.11, 8.1
( = 2)
3.21

Aspirin

1.0, 2.5
( = 2)

1.3

NSAIDs

1.0

2.1

Note: n = number of studies, if > 1.

" All risks greater than 1.3, adjusted for use of other analgesics,
are statistically significant, except in one study in the phenacetin
group wtih a risk of 4.

The Relative Risk of AN

The relative risk, or odds ratio, of developing either

chronic renal failure (CRF) or end-stage renal disease
(ESRD) in habitual users of analgesics varied between 2.4
and 8 in eight studies from 1969 to 1992, when the makeup
of OTC analgesic preparations was changing. In six of these
eight, the risk of CRF or ESRD from individual analgesics
could be estimated; phenacetin in five, acetaminophen in
three, aspirin in three, and nonaspirin NSAIDs in two
(Table 4).
The risk for developing ESRD from habitual phenacetin
use varied from about 3 to 19, and for CRF from 5 to 8
with a cumulative threshold dose of about 0.5 to 1.0 kg.
The risk of ESRD from habitual acetaminophen use varied
from 2 to 4 in two studies and was 3.2 for CRF in one
study. One study suggested consumption of as little as 360
pills containing acetaminophen (about 0.1 to 0.2 kg total
dose) may increase risk of ESRD.
Of three studies of aspirin, one from the United States
showed no risk of CRF and another showed no risk of
ESRD. However, one study from Spain showed a significant risk of ESRD of 2.5. Two studies from the United
States of nonaspirin NSAIDs have been reportedone in
subjects with chronic renal failure and one with ESRD. In
the study of CRF from North Carolina, the overall risk
was slightly increased at* 2.1; however, a risk of 16.6 was
found in men over 65 years of age. In the study of ESRD
from the mid-Atlantic area, there was no risk overall in
subjects taking more than 360 but less than 5000 pills;
however, in subjects taking more than 5000 pills lifetime
the risk was 8.8.
These epidemiologic data suffer from the problem of
accurately quantitating analgesic consumption by questionnaire or interview. In addition, case-control studies may
overestimate risk if patients with renal disease have symp toms which might increase analgesic consumption, socalled "confounding by indication." Although these studies
are not conclusive and further research is clearly needed,
they do suggest the following tentative conclusions. First,
habitual acetaminophen use increases the risk of renal dis ease at a total dose no higher than for phenacetin, but the
risk appears to be somewhat less. A possible explanation

for this difference will be discussed below. Second, habitual

NSAID use may increase the risk of renal disease, especially in the elderly, but the threshold dose is much higher
than that for acetaminophen. Third, the risk of renal disease
from habitual aspirin use, if any, must be lower than that
for either acetaminophen or nonaspirin NSAIDs.
/Mixtures vs Single Agents. A recent epidemiological study
of AN from Europe suggests that AN occurs only in individ uals consuming analgesic mixtures. Based on this and other
evidence, the National Kidney Foundation recommended
in a recent position paper that analgesic mixtures be made
available by prescription only.
The hypothesis that only analgesic mixtures cause renal
disease may explain the apparent lower risk of acetaminophen compared to phenacetin noted above. Phenacetin was
always marketed as a mixture with other analgesics, usually
aspirin plus caffeine, whereas acetaminophen is available
both as a single agent and in mixtures with aspirin and
other agents. Although the risks calculated from case control studies have been adjusted statistically for the use of
other drugs, the question of whether habitual use of single
analgesics increases risk of renal disease has not been answered.
Risk by Diagnosis. Classically, the nephropathy associated
with analgesic use is a chronic interstitial nephritis. Nevertheless, two questionnaire-based studies show an increased
risk of habitual acetaminophen consumption in patients
with other renal diagnoses including hypertensive nephrosclerosis and diabetic nephropathy. Although these observations suggest that habitual acetaminophen use might contribute to the progression of established renal disease,
confounding by indication is a particular problem in these
studies because patients with known renal disease may
consume acetaminophen preferentially. Nevertheless, the
potential public health implications of these studies are
important and further research is indicated.
The Prevalence of AN

The incidence and prevalence of AN in the dialysis population are reported periodically in ESRD registries in Eu-

42. Nonsteroidal Anti-inflammatory Drugs and the Kidney

rope, Australia, and now in the United States. In Europe,
the average incidence of new cases starting dialysis was 2%
in 1990, down from 3% in 1981. However, the rate, although
declining, remains high in Switzerland (12%) and Belgium
(11%). In Australia, the 1992 rate was 9%, down from over
20% in the early 1970s. The decline has been attributed to
legislation limiting OTC availability of analgesic mixtures.
The United States Renal Data System (USRDS) indicates
that from 1989 to 1993, the diagnosis of AN was made in
0.8% of new dialysis patients. This rate was unchanged
from the 1988 to 1991 data. The USRDS does not publish
data on geographic variation; past reports indicated that
AN was more common in the southeastern United States
by a factor of about 5. In a recent population based study
in the mid-Atlantic region (Maryland, northern Virginia,
and the District of Columbia), the population-attributable
risk for habitual acetaminophen use was 8 to 10%. This
ten-fold discrepancy between the incidence of AN in the
USRDS and that predicted from a community-based case
control study could be due to geographical variation, or
an underdiagnosis of AN by nephrologists, or to an overes timate of risk based on case-control methodology.
There are very few estimates of the incidence and prevalence of AN in the population of patients with chronic
renal failure not on dialysis. During the phenacetin era in
the 1970s, two studies estimated that 7% (in Philadelphia)
and 13% (in northwest North Carolina) of patients with
chronic renal failure had AN. It is likely that this figure is
lower now than 20 years ago.
Pathogenesis

Concepts of the pathogenesis of AN are based primarily

on studies in animal models. The results of studies in which
single analgesics and various combinations have been fed
to rats have, however, been somewhat controversial. In
many cases, very large doses have been required to cause
renal lesions. In general, combinations of analgesics, especially those which include aspirin, have been more nephrotoxic than single agents. In fact, aspirin is the most nephrotoxic of all analgesics studied in the rat. Most studies agree
that dehydration increases the nephrotoxicity of these
agents.
Chronic interstitial inflammation and papillary necrosis
are the primary lesions of AN. This is probably due to the
concentration of the offending agents, especially acetaminophen and salicylates, by the countercurrent mechanism.
Acetaminophen is metabolized to toxic oxygen radicals by
cytochrome P450 mixed-function oxidases located in the
renal cortex and outer medulla and by prostaglandin endoperoxidase synthases in the renal papilla. Aspirin potentiates the toxicity of these radicals by depleting the kidney
of reduced glutathione necessary for their detoxification.
A second proposed mechanism invokes a decrease in
renal papillary blood flow in patients consuming analgesics
containing aspirin or other NSAIDs that could inhibit renal
PG production. In addition, a papillary microangiopathy
of unknown pathogenesis which could interfere with the
papillary circulation, has been demonstrated in autopsy

295
studies. A lipofuscin pigment in many tissues has been
described in autopsies of humans dying with AN. It is
thought to be an oxidized polymer of unsaturated fatty
acids, probably arising from the effect of the aforementioned acetaminophen derived radicals. The role of this
pigment in the pathogenesis of AN is not known.
Caffeine is present in many of the analgesic combinations
taken by patients with AN., Its role in the pathogenesis of
AN is not known. Most speculation has centered on the
possibility that caffeine increases analgesic consumption
by causing "caffeine withdrawal" headaches. However, caffeine may also play a role in the nephrotoxicity of acetaminophen as an adenosine antagonist. Adenosine diminishes
the transport-associated respiration in the medullary thick
ascending limb. Cell necrosis by oxygen radicals might be
enhanced if respiration were stimulated by adenosine antagonism.
Urinary tract malignancies have been attributed most
often to the N-hydroxylation metabolites of phenacetin.
Because acetaminophen is also metabolized to potential
carcinogens, it is not clear why the risk of urinary tract
malignancies has been associated to date only with habitual
use of phenacetin and not acetaminophen.
The Clinical Syndrome
History

The key to making the diagnosis is obtaining an accurate

history of the type and amount of analgesic consumption.
One-third or more of patients with AN will deny consuming
analgesics and even those who admit taking them may
underestimate the amount taken. The best approach to
ascertaining that analgesics are used regularly is to question
the patient about those chronic conditions for which analgesics might be taken (Table 2). Most patients with these
conditions will readily describe them in detail but they may
be reluctant to admit how much analgesic they take. The
situation is not unlike that of obese patients underestimating the amount of food they eat. A frequently employed
strategy to obtaining an accurate assessment of analgesic
consumption is to ask family members. Another clue to
the presence of habitual analgesic consumption is a history
of peptic ulcer disease. Up to one third of patients with
AN will have such a history.
Patients with AN may not be habitually consuming analgesics at the time they are seen by the nephrologist for
CRF or ESRD. Not infrequently, the patient presenting
with renal failure of unknown etiology will give a history
of some chronic pain condition in the past which lasted for
some significant period of time, but is no longer present,
and for which analgesics had habitually been consumed.
An important criterion for making the diagnosis is the
total dose of analgesic consumed. Most patients who admit
to habitual analgesic use have a favorite preparation, the
contents of which can be ascertained and a cumulative
total dose estimated. Not all patients with AN habitually
consume OTC preparations; some may be taking these
preparations by prescription. It should also be kept in mind
that patients with AN may also abuse other substances

296

Vardaman M. Buckalew, Jr.

such as alcohol, sedatives, and perhaps other recreational

drugs, and tobacco.
Physical Examination
The findings on physical examination in patients with
AN are nonspecific. Some investigators have noted a typical
personality type in women with this syndrome. They may
be excessively dependent and passive-aggressive with a low
threshold for pain and features of hypochondriasis. It has
also been noted that women with this syndrome appear to
be prematurely aged. Generalized atherosclerosis may be
found, because this condition may occur with increased
frequency in patients with AN compared to patients with
other types of renal disease.
Laboratory Findings
The laboratory findings may be typically those of chronic
tubulointerstitial nephritis as discussed in Chapter 48. Abnormalities found with increased frequency in patients with

AN include sterile pyuria and bacteriuria. When the syndrome was caused by phenacetin, an anemia out of proportion to the degree of renal failure was frequently seen. This
may not be observed in the syndrome caused by other analgesics.
Imaging of the kidney using ultrasonography, computed
tomography (CT), or pyelography may show reduced renal
size. The renal contour may have an irregular, wavy appearance described as "bumpy." Papillary necrosis may be manifest as renal medullary calcification. On ultrasound, this
calcification is typically seen surrounding the central sinus
and may appear as a complete or incomplete garland. However, renal contours may be difficult to discern on ultrasound in patients with advanced renal disease because of
increased echogenicity. Furthermore, it is difficult to precisely localize renal calcium deposits on ultrasound, and
papillary calcifications may be incorrectly diagnosed as vas cular calcifications or renal calculi. A recent study from
Europe explored the use of CT without contrast to evaluate

FIGURE I (A) CT scan without contrast in a 42 -year-old female with a 20-year history of daily
analgesic use for headache. Medullary calcifications are prominent without bumpy contours. (B)
CT scan without contrast in a 56-year-old female with a 35-year history of habitual analgesic
use for headaches. Bumpy renal contours and medullary calcifications are apparent. Renal
volume was 61.3 mm on right and 59.2 mm on left (see text).

42. Nonsteroidal Anti -inflammatory Drugs and the Kidney

renal size, contour, and calcifications in subjects with AN.

Renal size was measured as the sum of both sides of a
rectangle enclosing the kidney at the level of the renal
vessels. Renal volume was considered decre ased if this sum
was less than 103 mm in males and 96 mm in females.
Renal contour was considered "bumpy" if three or more
indentations were observed at the level where the most
number of indentations were identified. Calcifications were
localized as cortical, papillary, or central. The combination
of reduced renal size and either bumpy contours or renal
medullary calcification had a sensitivity and specificity of
90% for AN in patients with ESRD and 87% and 100%
respectively in CRF without ESRD. These findings are
illustrated in Fig. 1, which shows CT scans without contrast
in two patients with CRF and a history of habitual analgesic ingestion.
An increased incidence of renal parenchymal, pelvic, and
ureteral carcinoma has been documented in AN. However,
this was probably due to phenacetin and may no longer
be observed.
Diagnosis

Criteria for diagnosing AN, divided into major and minor categories, are summarized in Table 5. A high index
of suspicion is indicated in those with CRF or ESRD of
unknown etiology and two or three minor criteria, especially if there is a history of a chronic pain syndrome and
peptic ulcer disease. Although always inferential, the diagnosis is indicated with a high degree of certainty if two
or more major criteria are present, es pecially if accompanied by some of the minor criteria. In some cases, the
diagnosis of AN only can be suspected when the patient
has a chronic pain syndrome but denies heavy analgesic
consumption.
Treatment

The successful treatment of AN depends on identifying

habitual analgesic ingestion and either convincing the patient to discontinue the practice or to switch to a "safe"
analgesic. Unfortunately, there may not be a completely

TABLE 5

Diagnostic Criteria for AN

Major
History of daily analgesic consumption >1 year Small,
bumpy kidneys and renal calcification on ultrasound
Decreased renal volume, bumpy contours, and medullary
calcification on CT without contrast
Minor
Chronic pain syndrome History of peptic
ulcer disease Dependent personality,
hypochondriasis Chronic
tubulointerstitial nephritis Sterile pyuria
Bacteriuria

297

safe analgesic that can be used habitually in patients with

CRF. Aspirin taken alone may be the safest analgesic in
these patients, but should not be given in combination
with acetaminophen. Acetaminophen used alone is safe for
episodic use, although, as noted above, this drug if used
habitually may contribute to progression of renal failure
in established renal disease of any cause. NSAIDs have
been shown to decrease GFR in patients with CRF of any
cause and therefore should probably be avoided. Propoxy phene taken alone is safe, but it should be used with caution
in combination with acetaminophen. Tramadol (Ultram)
is a new centrally acting analgesic with weak opioid activity
and has no known renal toxicity. This drug can be tried in
patients with renal disease with severe pain syndromes in
whom habitual analgesics are needed. Nephrologists should
monitor the type and amount of analgesic consumed by all
patients with CRF, especially those whose renal failure
is progressing.
Several studies have suggested that avoidance of nephrotoxic analgesics prevents progression of renal failure from
AN. However, as with any type of renal disease, progres sion in AN may be affected by factors other than continued
analgesic ingestion. Hypertension, if present, should be
adequately controlled. The use of protein restricted diets
in patients with chronic renal failure of any cause is dis cussed in Chapter 68. Patients with AN may develop acute
papillary necrosis, urinary tract infections, or kidney stones.
The treatment of these conditions is discussed elsewhere
in this book.
Bibliography
Backris GL, Starke U, Heifets M, et al.: Renal effects of oral
prostaglandin supplementation after ibuprofen in diabetic subjects: A double-blind, placebo-controlled, multicenter trial.
J Am Soc Nephrol 5:1684-1688, 1995.
Beard K, Perera DR, Jick H: Drug-induced parenchymal renal
disease in outpatients. J Clln Pharmacol 28:431-435, 1988.
Bennett WM, DeBroe ME: Analgesic nep hropathy a prevent able renal disease. N Engl J Med 320(19):1269-1271, 1989.
Buckalew VM, Schey HM: Renal disease from habitual antipyretic
analgesic consumption: an assessment of the epidemiologic evidence. Medicine 11(1):291-303, 1986
Elseviers MM, De Broe ME: Analgesic nephropathy in Belgium
is related to the sales of particular analgesic mixtures. Nephrol
Dial Transplant 9:41-46, 1994.
Elseviers MM, De Broe ME: A long-term prospective study of
analgesic abuse in Belgium. Kidney Int 48:1912-1919, 1995.
Elseviers MM, De Schepper A, Corthouts R, et al.: High diagnostic
performance of CT scan for analgesic nephropathy in patients
with incipient to severe renal failure. Kidney Int 48:13161323, 1995.
Gurwitz JH, Avorn J, Ross-Degnan D, et al.: Nonsteroidal antiinflammatory drug-associated azotemia in the very old. JAMA
264:471-475, 1990.
Henrich WL (ed): Analgesic nephropathy. Am J Kidney Dis
28(Suppl. 1):S1-S70, 1996.
Johnson AG, Nguyen TV, Day, RO: Do nonsteroidal antiinflammatory drugs affect blood pressure? Ann Intern Med
121:289-300, 1994.

298

Arasb Ateshkadi

Klag MJ, Whelton PK, Perneger TV: Analgesics and chronic renal
disease. Curr Opin Nephrol Hyperten 5:236-241, 1996.
Murray MD, Brater DC, Tierney WM, ot al.: Ibuprofen-associated
renal impairment in a large general internal medicine practice.
Am J Med Sci 299:222-229,1990.
Palmer BF: Renal complications associated with use of nonsteroidal anti-inflammatory agents. J Invest Med 43:516-533, 1995.
Perneger TV, Whelton PK, Klag MJ: Risk of kidney failure associated with the use of acetaminophen, aspirin, and nonsteroidal
antiinflammatory drugs. N Engl J Med 331:1675-1679, 1994.
Sandier DP, Burr R, Weinberg CR: Nonsteroidal anti-inflammatory drugs and the risk for chronic renal failure. Ann Intern
Med 115(3):165 -172, 1991.

Sandier DP, Smith JC, Weinberg CR, et al.: Analgesic use and
chronic renal failure. N Engl J Med 320:1238-1243, 1989.
Toto RD, Anderson SA, Browri-Cartwright D, et al.: Effects of
acute and chronic dosing of NSAIDs in patients with renal
insufficiency. Kidney Int 30:760-768, 1986.
Warren GV, Korbet SM, Schwartz MM, et al.: Minimal change
glomerulopathy associated with nonsteroidal antiinflammatory
drugs. Am J Kidney Dis 13:127-130, 1989.
Incidence and causes of treated ESRD. U.S. Renal Data System,
USRDS 1995 Annual Data Report, The National Institutes of
Health, National Institute of Diabetes, Digestive and Kidney
Diseases, Bethesda, MD. Am J Kidney Dis 26(Suppl. 2):S39S50, 1995.

PRINCIPLES OF DRUG THERAPY IN

RENAL FAILURE
ARASB ATESHKADI

Patients with either acute or chronic renal failure typically receive multiple drugs to manage the associated complications. On the average, patients with chronic renal failure receive more than seven drugs; nearly one-fourth take
10 or more. Patients with renal impairment suffer from a
high incidence of adverse drug reactions. Hence, familiarity
with principles of drug dosage adjustment for patients with
renal impairment is of critical importance for any clinician
responsible for prescribing and monitoring pharmacotherapy. The importance of this assertion is illustrated by a
recent study which showed that as many as 44% of patients
with an estimated creatinine clearance (Ccr) of less than
40 mL/min received excessive doses of drugs. The average
dose was 2.5 times higher (range: 1.07 to 6.45) than the
maximum recommended for the patient's degree of renal dysfunction.
Currently, the drug approval process in the United States
requires drug dosing recommendations for patients with
renal failure as part of the product labeling. For most marketed drugs, this requirement was not in effect at the time
the drug was approved, and such information may not
be part of the product package. Numerous sources are
available for guiding dosage adjustment in renal failure.
Unfortunately, much of the information contained in these

Primer on Kidney Diseases, Second Edition

guidelines is based on single-dose pharmacokinetic studies,

rather than multiple-dose pharmacokinetic or pharmacodynamic investigations. Creatinine clearance, the marker
used to estimate glomerular filtration rate (GFR), may
grossly overestimate actual GFR (see Chapter 2). Therefore, it is recommended that the available sources always
be interpreted with caution. When such guidelines are absent or inadequate, the clinician must use knowledge of
the drug's pharmacokinetics and pharmacodynamics to decide on the need for, and the manner of, dosage adjustment.
Fortunately, most drugs follow a first-order elimination
process (i.e., linear pharmacokinetics), which means that
as the dose is doubled, serum concentration is also doubled.
This makes the effect of dosage adjustment on serum concentration predictable. A few drugs, such as phenytoin and
high-dose salicylate, exhibit nonlinear pharmacokinetics,
which means that the doubling of the dose results in a
disproportionate rise in plasma concentration of the drug.
The dosing of such drugs is often a greater challenge.
EFFECT OF RENAL FAILURE ON PHARMACOKINETICS

Drugs used in clinical practice can be divided into three

categories based on their route of elimination: (1) category
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

43. Principles of Drug Therapy in Renal Failure

299

I drugs (e.g., aminoglycosides) are primarily eliminated by

the kidneys; (2) category II drugs (e.g., ethanol, acetaminophen) are primarily eliminated by nonrenal routes; and
(3) category III drugs (e.g., bumetanide, nafcillin, vancomy cin) are eliminated by a combination of renal and nonrenal
routes. The need for dosage adjustment is often more critical for category I drugs, but drugs that undergo nonrenal
elimination may also require dosage adjustment in patients
with renal failure.
To fully appreciate pharmacokinetic changes in renal
failure, it is important to become familiar with some of the
commonly used pharmacokinetic terminology. Bioavailability (expressed as a percentage) is the fraction of the
dose absorbed from the site of administration (e.g., subcutaneous, oral/gastrointestinal). The apparent volume of distribution (Vd; expressed in liters) is a mathematical concept
that relates the amount of a drug in the body to the serum
concentration. It is important to note that Kd does not
correspond to any specific anatomic space within the body.
Half-life (fa', expressed in hours) is the time required to
decrease the serum concentration of the compound by half.
For example, a volume-overloaded patient receiving an
aminoglycoside antibiotic may exhibit a prolonged fa, not
only because of decreased renal function (i.e., decreased
renal clearance, Clrenai)> but also due to an increase in the

drug's yd . Table 1 lists drugs whose fa and Cl are affected

by end-stage renal disease (ESRD) and dialysis.
Effect of Renal Failure on Drug Absorption

In patients with renal failure, a number of factors can

affect drug absorption: uremic gastroparesis, changes in
gastric pH due to high concentrations of gastric ammonia,
gut wall edema, and alterations in first-pass metabolism.
The altered gastric pH affects the absorption of drugs such
as ferrous sulfate, ketoconazole, itraconazole, and dapsone.
Gut wall edema, which may occur during severe generalized edema, may affect the absorption of furosemide.
First-pass metabolism refers to the intestinal or hepatic
metabolism and inactivation of a drug, often by the cytochrome P450 enzyme system. Drugs with a high first-pass
metabolism include propranolol, verapamil, and lidocaine.
Drugs with decreased bioavailability in renal failure include
furosemide and pindolol. Drugs with increased bioavailability in renal failure include dextropropoxyphene, dihydrocodeine, oxprenolol, and propranolol. Before a decision
is made to decrease the dose of a drug with increased
bioavailability, it is also important to appreciate the pharmacodynamic changes that occur in renal failure. For example, because the sensitivity of /3-receptors is decreased in

TABLE I

Effect of End-Stage Renal Disease, Hemodialysis, and Peritoneal Dialysis on the Clearance and Half-Life of Selected Drugs

Effect of hemodialysis"
tin
Drug
Amikacin

Azlocillin
Aztreonam
Cefazolin

Cefoxitin
Cefuroxime
Cefotaxime
Ceftazidime
Ciprofloxacin
Clavulanic acid
Clindamycin
Erythromycin
Gentamicin
Imipenem
Metronidazole
Nafcillin
Ofloxacin
Piperacillin
Sulfamethoxazole
Ticarcillin
Tobramycin
Trimethoprim
Vancomycin

Normal
1.6
0.9
2.0
2.2
0.8
1.3
0.9
1.8
4.4
1.0
2.2-3.3
2.1 2.2
0.9
7.9
1.0
5-8
1.2
10
1.2
2.5
14
6.9

(h)
ESRD

39
5.1
7.0
28.0
13-25
15-22
2.5
26.0
8.4-12
4.3
1.9-3.4
4.0
53
2.9
7.7
2.1
28-38
3.9
13.3
14.8
58 2640 161

tm (h)
3.8-5.5
2.2
2.7
2.6-5.0
4
3.5
1.9-3.4
2.8
3.0-5.5
NR
1.6-3.1 0.8
5.2-11.3
1.0
2.8
NR
NR
1.3-2.4
3.2-11.1
3.4
4.3-6.7
5.0-9.4
NR

Cl
(niL/niin)
30-36
ND
43
NR
NR
NR
14-40
27-50
29.6-47
141
NR
28.5
24-47
84 58125
0
116
74 2184
33
31-70
29-66
16.1

Effect of peritoneal
dialysis'
tin (h)
18-26
2.5
7.0
32
NR
11.8
2.9-4.4
8.7
ND
ND
ND
ND
8.5
ND
5.6
ND
ND
ND
13-18
10.6
25 1724 3043

Cl
(mL/min)
6.7 ':

ND
2.1
NR

1.5
4.7
NR
8.5
ND
ND
ND
ND
12.5
ND
15.8
ND
ND

ND
1.2
7.2
4.7 5.1
2.3-14.2

" Effect during a conventional hemodialysis treatment.

b
Effect during a CAPD exchange.
Note: ti/2, terminal half-life; Cl, dialysis clearance (mL/min); ND, no data; NR, not reported; ESRD, end stage renal disease.

300

Arasb Ateshkadi

renal failure, the dose of propranolol may actually need to

be increased, despite its increased bioavailability in such patients.
In addition to these effects on drug metabolism, it is
important to note that certain drugs commonly used in
renal failure, such as cationic preparations used as phosphate binders (aluminum and calcium products ), may also
reduce the absorption of certain drugs, such as ampicillin,
digoxin, and isoniazid.
Effect of Renal Failure on Distribution
Several drug properties determine the extent to which
a drug distributes throughout the body. These include the
drug's molecular size, hydrophilicity, plasma protein binding, and tissue binding. Drugs that are hydrophilic (e.g.,
aminoglycosides) tend to distribute mostly to the intravas cular space and as a result have a low Vd - Likewise, drugs
that are highly protein bound (e.g., furosemide) mostly
tend to stay in the intravascular space. Conversely, drugs
that are lipophilic (e.g., quinidine) or highly tissue bound
(e.g., digoxin, cyclosporine) have a high Vd because they
distribute to larger areas of the body. Knowledge of the
drug's Kd is important for the purpose of calculating a
loading dose (see below). In renal failure, the Vd may
change due to changes in body composition, and decreased
plasma protein binding. The body composition may change
because of uremia-induced malnutrition and excess fluid
volume. Reduction in protein binding of drugs is clinically
significant only for agents that are highly bound to albumin
(i.e., acidic and neutral drugs) (Table 2), and may in patients with renal disease, occur for three main reasons:
hypoalbuminemia; drug displacement from binding sites
on albumin by accumulated organic acids; and diminished
affinity of albumin binding sites for drugs. With reduced
protein binding, a therapeutic free (active) serum concentration will occur at a lower total serum concentration. If
the dose of the affected drug is not properly targeted to
achieve a total serum concentration appropriate for pa -

dents with normal renal function, toxicity will occur

(e.g., phenytoin).
Effect of Renal Failure on Drug Elimination
Effect on Hepatic Metabolism
Renal failure reduces the nonrenal elimination of drugs
metabolized by the P450IID6 isoenzyme (e.g., debrisoquine, quinidine, metoprolol, codeine). This may be due
to the presence of a circulating inhibitor that competes
with the drug for P450 binding sites.
Renal failure may decrease hepatic deacetylation, acetylation, hydroxylation, O-demethylation, N-demethylation,
conjugation, and sulfoxidation of drugs. Renal disease may
affect reversible reactions such as conjugation, such that
inactive conjugated metabolites that are not renally excreted are converted back to the parent compounds. Examples of drugs with decreased nonrenal (hepatic) clearance
in ESRD are listed in Table 3.
Renal failure may also increase clearance of drugs like
phenytoin and propranolol by hepatic oxidative pathways.
Other drugs with increased hepatic clearance in ESRD
include bumetanide and nafcillin. Of particular interest is
the elimination behavior of nafcillin. In healthy individuals,
nafcillin undergoes both renal (60%) and nonrenal (40%)
elimination; however, in ESRD, nonrenal clearance of nafcillin increases to the extent that no dosage adjustment
is required.
Effect on Renal Metabolism
Virtually all the polypeptide hormones are metabolized
by the kidneys. Hence, the clearance of these substances
(e.g., insulin, parathyroid hormone) is greatly reduced in
renal failure. Renal tissues contain many of the enzymes
found in the liver. Most of the mixed-function oxidases
(i.e., P450 enzymes) are present in the renal cortex. How-

TABLE 3

Example of Drugs with Reduced Nonrenal

Elimination in End Stage Renal Disease
TABLE 2

Example of Drugs with Reduced Protein Binding in

Renal Failure

Barbiturates
Cephalosporins (some)
Clofibrate
Diazoxide

Acyclovir
Aztreonam
Captopril

Cefotaxime
Cimetidine

Furosemide

Cortisone
Imipenem
Metoclopramide

Penicillins (some)

Nimodipine

Phenytoin

Prednisone

Salicylate

Procainamide

Sulfonamides
Valproate

Procaine

Warfarin

Zidovudine

Sulindac

43. Principles of Dr u g T h e r a p y i n R e n a l F a i l u r e

ever, the kidneys have only 15% of the metabolic activity

of the liver, and drug clearance may not be significantly
affected by renal disease.
Effect on Renal Excretion
Renal handling of drugs includes filtration at the glomerulus and reabsorption and secretion by the tubules. The
extent of reduction in renal drug elimination depends on
both the fraction of the drug excreted by the kidneys and
the degree of renal insufficiency. When available, the serum
concentration of the parent compound and any active or
toxic metabolite may help guide dosing (e.g., procainamide
and its active metabolite W-acetylprocainamide). Management of drug levels is essential for drugs with a narrow
therapeutic index, such as digoxin and aminoglycosides.
EFFECT OF RENAL FAILURE ON PHARMACODYNAMICS
Patients with renal failure may exhibit alterations in the
degree of response to some drugs, such as diuretics, benzodiazepines, and antihypertensive agents. While such pharmacodynamic alterations are sometimes explainable by
changes in the drug's pharmacokinetics (e.g., reduced clearance), physiologic aberrations that occur during renal failure can also affect how a patient responds to the pharmacologic effects of a drug. Altered pharmacodynamics in renal
failure is typically due to: (1) change in receptor sensitivity,
as occurs with insulin and /3-receptor antagonists; or (2) a
more severe form of the disorder under treatment, such as
hypertension and volume overload. Diuretic resistance is
perhaps the most common form of renal failure-induced
alteration in pharmacodynamics. Patients with renal dis ease commonly require much higher doses of diuretics to
achieve a natriuretic response similar to that of healthy
individuals. Other examples of altered pharmacodynamics
in renal failure are increased sensitivity to alprazolam and
reduced effect of normal doses of commonly used antihypertensive drugs because of fluid overload.
ADJUSTING DOSAGE FOR RENAL FAILURE
In deciding whether to adjust the dose of a drug in a
patient with renal failure, it is also important to take note
of the underlying renal disorder, because it may cause alterations in pharmacokinetics or pharmacodynamics of the
agent. For example, the rate and extent of absorption of
a drug may be decreased in a patient with diabetic nephropathy and diabetic gastroparesis.
In adjusting drug doses for patients with renal failure,
clinicians should not rely excessively or exclusively on nomograms, because these do not take into account all the
variability that may exist in patients. Nomograms may be
used for initial therapy, but only with close monitoring of
the patient, and careful consideration of the following ques tions:
1. What is the patient's estimated renal function? Is it
changing?
2. What is the extent of the drug's renal elimination?

301
3. Are the drug's metabolites pharmacologically or lexi
cologically active? If so, do they accumulate in renal
failure?
4. Should the loading dose of the drug be altered?
5. Should the maintenance dose of the drug be adjusted?
6. Is the patient on dialysis?
7. What is the mode of dialysis?
8. If the patient is receiving hemodialysis or hemofiltration, which filter is being used?
9. How often and how efficient is the dialysis?
Estimation of GFR
Creatinine clearance (Ccr ) remains, despite its limitations, the method of choice for estimating GFR in the
clinical setting. Ccr can be determined by direct measurement via urine collection (see Chapter 2). Alternatively,
the Cockroft -Gault method is commonly used in the clinical
setting to estimate Ccr . This equation, however, can be used
only when renal function is stable,
(140 - age) X LEW (kg)
72 X [Cr] (multiply by 0.85 (1)
for females),
=

where [Cr] is serum creatinine and LEW is lean body

weight (kg), calculated as
LBW for males = 50 kg + 2.3 X inches >5 feet ,2)
LEW for females = 45.5 kg + 2.3 X inches >5 feet.
Route of Drug Elimination
Information on the routes of drug elimination can be
obtained from various sources, including drug monographs
(e.g., product package insert), and reviews of drug dosing
in renal failure (see bibliography). As a general rule, drugs
with extensive renal elimination (category I) will require
some degree of dosage adjustment, whereas drugs that are
extensively metabolized extrarenally (category II) need no
dosage adjustment. The need for dosage adjustment for
drugs that fall in category III (dual routes of elimination)
is based on clinical pharmacokinetic studies.
Metabolite Accumulation
Hepatic drug metabolism mostly leads to production of
hydrophilic compounds, which are then excreted in the
urine. Some of these accumulated metabolites may have
pharmacologic and/or toxicologic activity (Table 4). For
example, meperidine is metabolized to normeperidine. Although it has no narcotic effect, normeperidine can lower
the seizure threshold as it accumulates in uremic patients.
Another example is the class la antiarrhythmic agent (sodium channel blocker) procainamide. This drug is metabolized to JV-acetylprocainamide, which accumulates in renal
failure and acts as a class III antiarrhythmic agent (potas sium channel blocker). For certain drugs, the inactive metabolite may accumulate in renal failure and be converted
back to the active parent compound (e.g., diflunisal, ketoprofen, clofibrate).

302

Arasb At e s h k a d i
TABLE 4

Example of Parent Compounds and Their Pharmacologically or Toxicologically

Active Metabolites
Parent drug

Metabolite

Metabolite activity

Allopurinol

Oxypurinol

Azathioprine

6-Mercaptopurine

Inhibitor of xanthine oxidase

Immunosuppressant

Cephalothin

Desacetylcephalothin

Antibacterial potency 50% of parent compound

Chlordiazepoxide

Oxazepam

Anxiolytic

Chlorpropamide
Clofibrate
Daunorubicin
Diazepam
Doxorubicin
Meperidine

2-Hydroxychlorpropamide

Hypoglycemic

Chlorophenoxyisobutyric acid

Hypolipidemic with direct muscle toxicity

Daunorubicinol

Cytotoxic
Anxiolytic

Doxorubicinol
Normcperidine

Cytotoxic

Procainamide

/V-acetylporcainamide

Class III antiarrhythmic with possible cardiac toxicity

Propoxyphene

Norpropoxyphene

Narcotic analgesic

Propranolol

4-Hydroxypropranolol

Rifampin

Desacetylrifampin

/3-Receptor antagonist
Antimicrobial activity

Sulfadiazine

Acetylsulfadiazine

Nausea, vomiting, rash

Oxazepam

Lowers seizure threshold; psychotic changes

Dosage Adjustment in Renal Failure

In many instances where dosage adjustment is required

in renal failure, literature or product labeling recommendation suffices; however, as previously mentioned, these recommendations must always be interpreted and used with
caution.
Drug dosage adjustment must be made on a case-by-case
basis, after considering concomittant medical conditions,
other administered medicines, and severity of renal failure.
Some drugs require dosing based on LEW (e.g., aminoglycosides), whereas others are based on total body weight
(e.g., heparin, vancomycin).
Nonrenal disease, which may require additional dosage
adjustment (e.g., hepatic disease), should be identified. If
a therapeutic drug concentration is rapidly needed, as may
be the case for a severe infection or an arrhythmia, a loading
dose is calculated
Loading dose = Vd x C n

(3)

where Cmax is the desired peak serum concentration.

As can be seen from this equation, the drug's Vd must be
known. In renal failure, the Fd may increase (e.g., digoxin,
aminoglycosides), decrease (e.g., phenytoin), or remain unchanged. Determining a maintenance dose is often more
challenging and is established on the basis of: (1) desired
peak and trough concentrations of the drug (if known)
and (2) estimated or calculated individual pharmacokinetic
parameters. Two methods are used individually or together
to adjust the maintenance dose: dosage reduction and interval extension. With some cardiovascular drugs, maintenance of a steady concentration is important; hence, dosage
reduction is used while maintaining normal intervals. Conversely, aminoglycoside antibiotics require high peak con-

centrations for efficacy and low trough concentrations to

minimize toxicity; therefore, both dosage reduction and
interval expansion are often needed.
For many drugs with a narrow therapeutic index for
which determination of serum concentrations are clinically
available (Table 5), therapeutic drug monitoring is neces sary. To better interpret a serum or plasma concentration,

TABLE 5

Drugs Which Require Therapeutic

Drug Monitoring
Amikacin

Carbamazepine
Cyclosporin
Digoxin

Gentamicin
Lidocaine
Lithium
Methotrexate
Netilmicin

Phenobarbital
Phenytoin
ProcainamideW-acetylprocainamide
Quinidine

Theophylline
Tobramycin

Valproate
Vancomycin

43. Principles o f D r u g T h e r a p y i n R e n a l F a i l u r e

the drug dose, and the route and time of administration

must be known. It must be noted that toxicity can still
occur while the serum drug concentration is maintained
within the therapeutic range. The use of digoxin in patients
with hypokalemia is one such example.
Dosage Adjustment in Dialysis

The degree to which a drug is removed via dialysis determines whether a supplemental dose is needed or not. Although the peritoneal membrane has larger pore sizes than
standard hemodialysis membranes, drug clearance during
peritoneal dialysis is usually much lower due to a substantially reduced dialysate flow rate (see Chapters 63 and 64).
Drug Dosing in Hemodialysis

The amount of drug removed by hemodialysis is approximately the product of drug concentration in the dialysate
and the dialysate volume. This value divided by the total
body stores of the drug prior to dialysis (product of predialysis serum concentration and volume of distribution) yields
the actual fraction of drug removed by dialysis (FDR). In
addition, the dialysis membrane can bind a certain fraction
of the drug, which will not appear in the dialysate. Measure ment of dialysate drug concentrations is not practical in
the clinical setting. Hence, several methods to predict the
FDR have been proposed (see bibliography). The FDR is
an index of the relative dialyzability of the drug and has
been used to determine whether dosage adjustment is
needed.
Endogenous clearance of a drug in the period between
dialyses (i.e., interdialytic period) may also affect drug dosing. Knowledge of the amount of drug removed both during
and between dialysis treatments is often necessary to design
an optimal therapeutic regimen. The clearance and halflife of several drugs both on and off dialysis are presented
in Table 1. Patients may require postdialytic or interdialytic
dosing when the maintenance of serum concentrations
within a narrow range is necessary (e.g., aminoglycosides).
Accurate information on the dialysis kinetics of drugs is
not always readily available; hence, complicated pharmacokinetic equations have been developed to aid in dosing
patients on hemodialysis. Although a trained pharmacotherapist is sometimes able to estimate the fraction of a
dose removed based on the properties of both the drug
and the dialysis procedure, most clinicians find it easier to
follow standard guidelines for drug dosing. Such guidelines
are available from the product package insert, and dosing
tables and nomograms when such recommendations are
absent or inadequate, the clinician must use knowledge of
certain properties of the drug and the dialysis procedure
to make an educated guess regarding how much of the
drug has been removed. Factors that favor dialysis of drugs
are listed in Table 6. Water-soluble drugs with molecular
weights less than 500 Da and low plasma-protein binding
can be expected to be removed by dialysis. The charge
characteristic of some drugs favor their adsorption to the
hemodialysis membrane. Such drugs are removed from the
blood without appearing in the dialysate (e.g., aminoglyco-

303

TABLE 6

Factors That Increase Drug Dialyzability

Drug properties
Molecular weight <500 D9 "
High water solubility Small
volume of distribution No
erythrocyte partitioning Low
nonrenal eliminat ion
Dialysis properties Membrane
composition and charge" (drug dependent)
Large membrane surface area (e.g., high efficiency)
Large membrane pore size (e.g., high flux)
High dialysate flow rate or volume"
High blood flow rate"
Dialysate composition (drug dependent)
High ultrafiltration rate
Long dialysis treatment times
Long dialysate dwell time*
" For hemodialysis.
6
For peritoneal dialysis.

sides binding to polyacrylonitrile membrane). Drugs with

smaller Vd are distributed mostly in the intravascular space
and hence are more likely to be available for removal by
dialysis. Conversely, drugs which partition to the erythrocytes (e.g., cyclosporine) are less available for dialysis removal. Perhaps the two most important dialyzer-related
properties for drug removal are membrane pore size and
surface area. The use of larger-pore (i.e., high-flux) and/
or larger surface-area (i.e., high-efficiency) dialysis membranes increase the dialyzability of drugs. High-flux membranes may remove compounds with molecular weights
ranging from 5000 to 20,000 Da. As a result, drugs that
are not dialyzed by standard dialysis membranes will be
dialyzed when high-flux membranes are employed. For example, vancomycin (molecular weight, 1442 Da) which has
a t-i/2 of 180 hours with standard dialysis membranes, has
a ?i/2 of 56 hours with high-flux, high-efficiency membranes
(Fig. 1).
Drugs that are efficiently removed by hemodialysis are
administered postdialysis in amounts equal to the amount
lost during dialysis. For example, during a typical hemodialysis session approximately half of the total body amount
of an aminoglycoside is removed, and a dose equal to 50%
of the maintenance dose will be required postdialysis.
Drug Dosing in Continuous Ambulatory Peritoneal
Dialysis (CARD)

Factors that increase drug removal by peritoneal dialysis

include a low molecular weight, low protein binding, a small
Vd , a rapid rate of equilibration between tissue binding sites
and blood, and a limited amount of nonrenal metabolism
and excretion (Table 6). As a general rule, if the amount
of drug removed by daily peritoneal dialysis is greater than

304

Arasb Ateshkadi

90

80 ^
60 -

s.
50h

o
40
0.4

0.6

0.8
1
1.2
1.4
1.6
Membrane Surface Area (m2)

1.8

FIGURE I Effect of dialysis membrane surface area on extracorporeal clearance of vancomycin.

20 to 30% of the administered dose, larger doses must be

administered. Example of drugs that require higher doses in
CAPD include the aminoglycosides, most cephalosporins,
and probably vancomycin (inconsistent data reported).
Many other drugs do not require a dosing regimen different
from those for a nondialyzed patient, because of their low
clearance by CAPD. Drugs with a high yd (e.g., digoxin),
high nonrenal clearance (e.g., cimetidine), or high plasma
protein binding (e.g., tricyclic antidepressants) are not
cleared to any significant extent by CAPD.
Because fewer data are available on the effect of
CAPD on pharmacokinetics of drugs, consideration of
certain general principles allows for a reasonable prediction of whether dosing adjustment is required. Estimation
requires the knowledge of several factors that influence
drug movement across the peritoneal membrane (Table
6). These include the physicochemical and pharmacokinetic properties of the drug, the physiology and anatomy
of the dialysis (peritoneal) membrane, and the dialysis
regimen. To assess the amount of drug removed by
CAPD, both the peritoneal clearance (C1PD) and the Vd
of the drug must be known. Similar to calculating the
total body clearance, Clpo can be calculated by dividing
the total amount of drug recovered in the spent dialysate
from one or more dwell periods by the area under the
plasma concentration-time curve during the same period
AUC0-t. Alternatively, if at the end of each dwell period,
equilibrium is reached between the drug concentrations
in blood and the peritoneal cavity, and if there is no
significant plasma protein binding, the C1PD is roughly
equivalent to the dialysate flow rate. Given that in CAPD
the dialysate flow rate rarely exceeds 5 to 7 mL/min,
the amount of drug removed by CAPD is limited. In
estimating the supplemental dose of a drug, the amount
of drug or the fraction of the dose removed by CAPD
(FDRpo) may be determined from the formula
FDRp n =
(4)
C l t Clp
ota
l X /u

where Cl,ola i is the systemic drug clearance and / u is the

unbound drug fraction. In this equation, Cl totai includes

both the peritoneal (i.e., C1PD) and the endogenous (i.e.,

renal and nonrenal) clearances of the drug. The endogenous clearance may be estimated from the literature values
for patients with ESRD. The unbound drug fraction is
important, because plasma protein binding of the drug is
the main limiting factor for diffusion of a drug into the
peritoneal cavity. Ideally, /u values from ESRD patients
should be used. If not available, data from healthy subjects
may be substituted. This equation does not take into account protein loss from the peritoneal cavity. Because the
quantity of protein loss from the peritoneal cavity is small
compared to the amount of protein in the systemic circulation, this loss does not substantially affect drug pharmacokinetics.
Pharmacokinetics of Intraperitoneally Administered Drugs
In patients on CAPD, intraperitoneal administration of
drugs is the preferred route in at least three clinical situations: (1) for the administration of antibiotics in patients
with CAPD peritonitis; (2) for the administration of insulin
and (3) for the administration of deferoxamine in the treatment of aluminum overload. The intraperitoneal adminis tration of a low-molecular-weight drug, such as most antibiotics, rapidly achieves a high local drug concentration
within the peritoneal cavity. This concentration gradient
leads to rapid diffusion of the drug from the peritoneal
cavity into the systemic circulation. The reverse movement
from the systemic circulation into the peritoneal cavity is
much slower and restricted. As a result, intraperitoneal
antibiotics used in the treatment of peritonitis (e.g., aminoglycosides, cephalosporins) are 50 to 80% bioavailable during a 6-hour dwell period. During bouts of peritonitis, the
permeability of the peritoneal membrane increases, and
as a result, both the rate and extent of intraperitoneal
absorption of certain drugs (e.g., vancomycin, gentamicin,
various /3-lactam antibiotics) increase and remain so for
days after the peritonitis episode.
Drug Dosing in Continuous Renal Replacement
Therapy (CRRT)
Since dialysis is not employed during continuous hemofiltration, solute removal with CAVH or CVVH occurs
exclusively via ultrafiltration (i.e., convective movement).
CAVHD and COVHD rely on diffusion as well as convective clearance. As compared to standard hemodialysis, solute clearance per unit time may be lower during CRRT;
however, due to its continuous nature, the overall clearance
may be higher. For example, urea clearance during CRRT
may be only 15 to 30 mL/min, versus approximately
200 mL/min for intermittent hemodialysis; however, because CRRT occurs 24 hours per day, the overall weekly
clearance of urea for CRRT is 151 to 302 L vs. 48 L for
intermittent hemodialysis. Drug removal during CRRT is
affected by some of the same factors described above and
listed in Table 6. Most of the hemofilters are the same as
high-flux hemodialysis filters, and remove molecules with
molecular weights of 5000 to 20,000 Da; hence, the molecular weight of the drug is usually not a rate-limiting factor
for drug removal. As for hemodialysis, however, highly

305

4 3 . P ri n c i p l e s o f D r u g T h e r a p y i n R e n a l F a i l u r e
TABLE 7

Total Clearance, Extracorporeal Clearance, and Adjustment Factor for Continuous

Renal Replacement Therapy

Drug
Amikacin

Clanuric (mL/min)
2

0.7-0.8

AFfor
ultrafiltration
of 0.5 L/h
4-5

'total I" ! extra * ^--

Atenolol

13

0.39

1.6

Carbenicillin

11

0.35

1.5

Cefalexin

15

0.32

1.5

Ceftazidime

12

0.38

1.6

Cefuroxime

0.47

1.9

Cisplatin

0.29

1.4

Fluconazole

0.55

2.2

Flucytosine

Ganciclovir

16

0.34

Gentamicin

0.6-0.7

Lamoxatam

10

0.33

1.5

Lithium

0.6

2.5
1.5

14

0.37

1.6

Methotrexate

0.33

1.5

Netilmicin

0.3-0.6

1-3

Phenobarbital

0.45

1.8

Sotalol

11

0.43

1.8

Tetracycline

15

0.31

1.4

Tobramycin

0.6-0.7

Clanuric, total clearance in anuric patients; Clex t ra: Cltotal, ratio of extracorporeal (CRRT)
clearance to total endogenous clearance; AF, adjustment factor.

protein -bound drugs are not significantly removed by

CRRT.
Limited data are available for drug dosing in CRRT. It
should be further emphasized that the available literature
values recommend dosing alterations based on various
blood flow settings, filter types, and ultrafiltration rates,
which may not be applicable to the specific patient. Because
patients who receive CRRT are critically ill, drug dosing
adjustments must be made with utmost care. When available, plasma or blood concentrations of drugs should be
obtained and monitored, especially for drugs with a narrow
therapeutic index (e.g., aminoglycosides, antiarrhythmics).
In the absence of dosing guidelines, a trained pharmacotherapist may be consulted to calculate the amount of drug
that may be removed, based on the drug's sieving coefficient (Si). The sieving coefficient of a substance is the
relationship between drug concentration in the ultrafiltrate
(Cu() and the average drug concentration in the plasma or
serum, as calculated from arterial (CA) and venous (Cv)
concentrations, and describes the ability of a drug to pass
from the systemic circulation to the ultrafiltrate:
Si = (CA + Cy)/2'
Si values near 1.0
indicate free drug movement
across the membrane, whereas Si values near 0 indicate
complete lack

(5)

of drug movement to the ultrafiltrate. Drugs with a high

Si value include aminoglycosides, vancomycin, and some
cephalosporins. If no data on Si are available, then the
extent of the drug's plasma protein binding can provide a
rough estimate, since there is a good correlation between
/u and Si. If the /u value is to be used, ideally values from
patients with acute renal failure should be used. If not
available, data from with ESRD may be used.
Supplemental dosing may be required during CRRT if
the ratio of extracorporeal clearance to endogenous clearance is greater than 0.25. If the nonrenal clearance is
>60 mL/min, supplemental dosing during CRRT is not required.
In CRRT, the procedure for calculating a loading dose
is identical to that described above for patients with renal
failure. To calculate a maintenance dose, the dose recommended for anuric patients is multiplied by an adjustment
factor (AF), as derived from the literature. Alternatively,
the AF may be used to adjust the dosage interval,
- AF'
(6)
where T ' is the dosing interval for CRRT patients and r is
the dosing interval for anuric patients. Table 7 lists the
total and extracorporeal clearances, as well as AF values for
some commonly used drugs likely to be removed by CRRT.

306
Bibliography
Anders MW: Metabolism of drugs by the kidney. Kidney Int
18:636-647, 1980. Anderson RJ. Drug prescribing for patients
in renal failure. Hasp
Pract Off Ed 18:145-149, 1983. Ateshkadi A, Matzke GR:
Dialysis therapy. In: Herfindal ET,
Gourley DR (eds) Textbook of Therapeutics: Drug and Disease
Management, 6th ed. Baltimore, Williams & Wilkins, pp. 429464, 1996.
Bakris GL, Talbert R: Drug dosing in patients with renal insufficiency. A simplified approach. Postgrad Med 94:153-156, 1993.
Bennett WM: Guide to drug dosage in renal failure. Cttn Pharmacokinet 15:326-354, 1988. Bennett WM, Aronoff GR,
Morrison G, et al.: Drug prescribing
in renal failure: dosing guidelines for adults. Am J Kidney Dis
3:155-193, 1983. Bjornsson TD: Nomogram for drug dosage
adjustment in patients
with renal failure. Clin Pharmacokinet 11:164-170, 1986.
Davis BB, Mattammal MB: Renal metabolism of drugs and xenobiotics. Nephron 27:187-196, 1981.

Arasb Ateshkadi
Gibson TP: Renal disease and drug metabolism: An overview.
Am J Kidney Dis 8:7^17, 1986. Keller E: Peritoneal kinetics
of different drugs. Clin Nephrol
30(Suppl. l):S24-528, 1988.
Lee CC: The assessment of fractional drug removal by extracorporeal dialysis. Biopharm Drug Disposition 3:163-173,
1982. Reetze Bonorden P, Bohler J, Keller E: Drug dosage in
patients
during continuous renal replacement therapy. Pharmacokinetic
and therapeutic considerations. Clin Pharmacokinet 24:362379, 1993. Swan SK, Bennett WM: Drug dosing guidelines in
patients with
renal failure. West J Med 156:633-638, 1992. Talbert RL: Drug
dosing in renal insufficiency. J Clin Pharmacol
34:99-110, 1994. Touchette MA, Slaughter RL: The effect of
renal failure on hepatic
drug clearance. Ann Pharmacother 25:1214-1224, 1991.
Turnheim K: Pitfalls of pharmacokinetic dosage guidelines in renal
insufficiency. Ear J Clin Pharmacol 40:87-93, 1991.

SECTION 7
HEREDITARY RENAL DISORDERS

SICKLE CELL NEPHROPATHY

JON I. SCHEINMAN

Clinical Features

The recognized renal syndromes in sickle cell disease

(SCD) include hematuria, (which may herald renal papillary necrosis) tubular dysfunction, and proteinuria associated with a sclerosing nephropathy (Table 1).

Gross hematuria can be one of the most dramatic clinical

events in SCD. It can occur at any age, and its origin is
usually unilateral, with greater propensity to be on the left.
Gross hematuria can also be associated with sickle trait,
and it has been shown to be more often associated with
sickle cell trait than sickle cell anemia, postulated to be
due to the associated greater genetic frequency of the sickle
heterozygote in sickle trait. Thus, the workup of the patient
with hematuria should include screening for sickle hemo globin, especially in the absence of hypercalciuria or evidence of glomerular disease, and renal "sickle crisis" is the
likely diagnosis when gross, painless hematuria occurs in
a known HbSS or HbSA patient. Notably, severe pain
makes the presumptive diagnosis of sickle hematuria less
likely.
RPN has been postulated to have common etiologic factors with hematuria in sickle cell patients, but the clinical
relationship of hematuria and necrosis is variable. Papillary
necrosis may or may not be present in sickle cell patients
with hematuria, and pain is not linked to either sickle
cell hematuria or papillary necrosis. Papillary necrosis detected in sickle cell patients is usually without signs or
symptoms. IVP surveys of patients without hematuria or
pain have shown a higher prevalence of papillary necrosis
in sickle cell patients than in those without hemoglobinopathy.
For the patient with known sickle cell hemoglobin presenting for the first time with gross hematuria, assays of
coagulation parameters including platelets are appropriate.
Renal and bladder ultrasound or intravenous pyelogram
may reveal papillary necrosis (Fig. 1), but can also rule out
bleeding from a tumor. Consideration of tumor is especially
important in view of the recently described renal medullary
carcinoma that appears almost exclusively in patients with
sickle hemoglobin. Increased echo density of the medullary
pyramids is common in sickle cell disease, even without
papillary necrosis. A later finding in papillary necrosis is
calcification of the medullary pyramids in a "garland" pattern surrounding the pelvis. A "shadowing" echo density
seen in nephrocalcinosis is different from that resulting
from other causes . It is probably unnecessary to perform
contrast urography to visualize the renal architecture unless
urinary obstruction is apparent by ultrasound. When urog-

GROSS AND MICROSCOPIC HEMATURIA

Pathogenesis

Sickling in the renal medulla causes vascular obstruction, proximal congestion, and red blood cell (RBC)
extravasation, resulting in hematuria. Specific conditions
in the medulla predispose to sickling: the pO2 of 35 to 40
mm Hg in the medulla is below the 45 mm Hg threshold
for sickling; the very high osmolality causes water to move
out of RBCs and concentrates the sickle hemoglobin,
thereby aggravating sickling; and the low medullary pH
also promotes sickling. In pathologic investigation, severe
renal medullary vascular congestion is found in the kidneys
of patients with gross hematuria. Diffuse microscopic extravasation occurs, and a specific bleeding point cannot
be defined.
Renal papillary necrosis (RPN) may occur in some sickle
cell patients with hematuria, both those patients with sickle
cell disease and those with sickle cell trait. This RPN seems
causally related to the medullary vascular congestion created by increased sickling induced by specific conditions
in the medulla. When RPN has been studied in patients
with hematuria, it appears as a diffuse fibrotic process with
areas of active extravasation. Acute necrosis of large areas
of the medulla is not observed. Collecting ducts appear
within areas of amorphous collagenous scar, and vasa recta
are selectively destroyed. The findings associated with RPN
in sickle cell patients with hematuria resemble those in
animal models with chemically induced RPN in which an
initial dilatation and engorgement of vessels is followed by
small, focal repeated infarctions in the papillae.
The pathologic findings of the RPN in SCD differ from
those in analgesic abuse nephropathy or diabetes mellitus.
In analgesic abuse nephropathy, the most serious fibrotic
lesions are in the descending limbs of the loops of Henle
and peritubular capillaries, and vasa recta are typically
spared. In diabetes mellitus, massive infarction occurs with
a sloughing of papillae that may cause obstruction.

Primer on Kidney Diseases, Second Edition

309

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

310

Ion I. Scheinman
TABLE I

Renal Findings in Sickle Cell Nephropathy

Hematuria Medullary
congestion Renal
papillary necrosis
Medullary calcification
Medullary carcinoma
Tubular dysfunction
Concentration defect
Increased sodium and phosphate reabsorption
Decreased proton and potassium secretion
Increased urate secretion
Glomerular sclerosis Hyperfiltration
Glomerular hypertrophy
Proteinuria/nephrotic sydnrome Focal
segmental glomerulosclerosis

raphy is performed, the calyceal clubbing of RPN in SCD

is not accompanied by cortical scarring, as is seen in pyelonephritic clubbing.
Treatment

Conservative management is appropriate in view of the

benign pathology of sickle cell hematuria. Probably the
most effective mode of conservative management is maintaining a high rate of urine flow by using hypotonic fluid
intake (4 L/day) and diuretics (thiazide or loop diuretics
such as furosemide). Diuresis can diminish medullary osmolarity and possibly alleviate sick ling in the vasa recta,
which in turn may also help prevent papillary necrosis. In
addition, the high urine flow rate will promote clearance
of clots from the bladder. Bed rest is often recommended,
presumably to avoid dislodging clots, but there is little
evidence to show that it is beneficial.

Another mode of conservative management is alkalinization of the urine with 8 to 12 g NaHCOs per day, which
may reduce the tendency of the acid environment in the
renal papillae to promote sickling. Alkalinizing the patient
to increase hemoglobin O2 affinity is theoretically helpful,
but not of proven value.
Proceeding to less conservative treatments, transfusions
may be necessary both to correct blood loss and, as in any
other sickle crisis, to increase the proportion of normal
HbAA cells.
Although e-amino caproic acid (EACA) may be used
in sickle cell hematuria to facilitate clotting in the urinary
tract by inhibiting fibrinolysis, its use carries a risk of systemic thrombosis and is not advised unless other measures
have failed. In one series, 4 of 12 cases required EACA
after failures of treatment with volume and alkalinization.
The effective adult dose of EACA is 8 g per day. Lower
doses may be adequate to arrest hematuria. It may be
worthwhile to start with 3 g and increase the dose until
hemostasis occurs.
Other modes of treatment of sickle cell hematuria can
be considered. In vitro, hyponatremia can induce swelling
of RBCs, reduce the effective HbS concentration, and make
cells less likely to sickle. Theoretically, a combination of
vasopressin and hypotonic fluid administration could decrease plama osmolality and accomplish the same goals in
vivo. However, because induced hyponatremia has not
been proved safe or effective in humans and because other
measures are usually sufficient, induced hyponatremia
is not recommended. Experimental data suggest that
angiotensin-converting enzyme (ACE) inhibition can induce a 50% increase in papillary blood flow, which could
theoretically prevent RPN, but because increased blood
flow could acutely aggravate hematuria, ACE inhibition is
also not advised. Uncontrolled bleeding may be approached by arteriographic localization and local embolization of the involved renal segment. Pelvic irrigation with
a sclerosing agent such as oxychlorosene has been used.
Nephrectomy will be required only rarely.
TUBULAR DYSFUNCTION

Because the pathology in sickle cell disease primarily

involves the medulla, the functions of the collecting ducts
and juxtamedullary nephrons are most affected. Accordingly, a renal concentrating defect is the most commonly
recognized renal problem in SCD.
Pathogenesis

FIGURE I (A) Tomographic pyelography of an 18-year-old patient presenting with abdominal pain and hematuria. Papillary
necrosis is evident from blunted medullary cavities. (B) Ultrasono graphic visualization of the same kidney. The middle pole exhibits
deep extensions into the papilla, likely sinus tracts, typical of the
"papillary form" of renal papillary necrosis. The parenchyma adjacent to calyces is echodense, reflecting the fibrosis.

Urinary Concentration and Water Reabsorption

Normally, the highest urine concentration is achieved
by the collecting ducts of juxtamedullary nephrons which
extend deepest into the medulla. The continued low-grade
sickling and medullary congestion in SCD result in loss of
the normal conditions for medullary water reabsorption
and thus cause decreasing urine concentration. Vasopressin
generation is normal in SCD, yet medullary damage results
in a concentrating defect not responsive to vasopressin.

311

44. Sickle Cell Nephropathy

A maximum urinary osmolality in the range of 450 mOsm/

kg is not a severe limitation for the SCD patient because
this degree of concentration can be achieved by the outer
medulla. Defective urinary concentration can often be restored in children with SCD by "hypertransfusion" therapy, presumably reversing the inner medullary congestion.
However, in adults, the concentrating defect is not revers ible; medullary fibrosis results in an irreversible destruction
of collecting ducts.
The renal concentrating defect in SCD does not involve
urinary diluting capacity, which normally depends on solute
absorption in the ascending loop of cortical nephrons that
leaves a dilute urine. The renal cortical area is relatively
uninvolved in SCD, and urinary dilution is appropriate.
Tubular Secretion
Tubular secretion is altered in the kidneys of SCD patients. Effective renal plasma flow (ERPF) in SCD patients
is increased, as indicated by increase in maximum hippurate
tubular transport, a fechnique used because most hippurate
is removed during a single pass through the kidney. Creatinine secretion in SCD is also increased, resulting in a creatinine clearance higher than expected for the glomerular
filtration rate (GFR). Increased uric acid excretion likewise
occurs, but a normal plasma uric acid level is maintained
because urate production is increased in SCD patients.
Proximal Tubular Reabsorption
Proximal tubular reabsorption is also altered in SCD
patients. Increased proximal sodium reabsorption is noted,
causing less distal sodium delivery in response to saline
infusion. The response to diuretics may be diminished because it is dependent on sodium delivery out of the proxi mal nephron. Phosphate reabsorption is similarly increased
in the proximal tubules and may result in hyperphosphatemia during times of sickling and RBC destruction.
Acid and Potassium Excretion
Acid and potassium excretion are variably affected in
SCD. Although it might be expected considering the severe
involvement in SCD of the longer juxtamedullary nephrons
where bicarbonate is normally most effectively reabsorbed,
bicarbonate wasting or proximal renal tubular acidosis is
not commonly seen in SCD. It is assumed that cortical
nephrons are able to adequately compensate. Hydrogen
ion (proton) secretion is performed by the "intercalated"
collecting duct cells that are most prominent in the cortical
segment of the collecting ducts. Because SCD does not
typically involve this cortical area, a severe acidification
defect is not usual. Still another factor in normal renal acid
excretion, ammonium generation, is not affected by SCD.
Titratable acid and ammonium generation in response to an
ammonium chloride challenge in SCD patients are slightly
decreased, but a maximal stimulus can produce normal
acidification. Potassium excretion is impaired in SCD patients, but hyperkalemia is rare without renal failure. An
observed decrease in acid and potassium excretion in SCD
suggests that aldosterone deficiency is responsible, but renin and aldosterone values and their responses to volume

depletion are normal. Electrolyte abnormalities resemble

those in type IV renal tubular acidosis but actually result
from an aldosterone independent end organ failure secondary to medullary fibrosis.
Possible Role of Prostaglandins in the Regulation of Renal
Function in SCD
In general, decreased medullary transport and increased
proximal activity have been ascribed to vasodilatory prostaglandin activity, possibly a compensation for defects in
medullary sodium and water conservation. Data have come
primarily from indomethacin inhibition of prostaglandin
synthesis. Measurements of prostaglandin excretion are not
demonstrably abnormal. In normal patients, indomethacin
does not change GFR or ERPF. In SCD, both GFR and
ERPF are decreased with indomethacin, suggesting that the
high GFR and ERPF may be maintained by prostaglandins.
Sodium retention is induced by indomethacin in both normals and SCD patients, but in normals, water reabsorption
also increases, resulting in no change in serum osmolality.
In SCD, water retention does not equally accompany sodium retention, and plasma osmolality increases. Compared to normals, SCD patients experience a greater fall
in the fractional excretion of sodium after indomethacin.
Urinary dilution in response to water loading, which is
otherwise normal in SCD, is diminished by indomethacin
but not in normals. Taken together, these findings are best
explained by decreased delivery of sodium to the distal
diluting segment.
Clinical Features
Clinically significant abnormalities in tubular function
are more likely in patients developing renal insufficiency
with failure of compensating mechanisms. The diagnosis
of renal tubular abnormalities is evident from routine laboratory tests, which may show acidosis, hyperphosphatemia,
hyperuricemia, and hyperkalemia.
Treatment

The tubular disorders in SCD usually require no treatment when renal function is normal. A failure of urinary
concentration may result in increased susceptibility to dehydration. Early treatment of diarrhea or emesis will diminish that risk. Acidosis may require early treatment in the
patient with SCD. Diuretics, especially the thiazides, may
aggravate hyperuricemia. The edema accompanying severe
anemia may be difficult to treat because of the diminished
diuretic response. In the presence of renal insufficiency,
potassium released from erythrocytes during severe hemo lysis may exceed the patient's ability to excete potassium.
/3-Adrenergic blockade and ACE inhibition are potentially
aggravating factors.
PROTEINURIA AND FOCAL
SEGMENTAL GLOMERULOSCLEROSIS

Proteinuria is now recognized as the harbinger of progressive renal insufficiency in SCD. The incidence of renal

312

Ion I. Scheinman

failure has been estimated to be as great as 10 to 25% in

HbSS or Hbc.
Pathogenesis
In sickle cell disease nephropathy, glomeruli are hypertrophied, and the hypertrophy correlates morphologically
with ERPF and GFR. Proteinuria in SCO results from a
functional "leak" in the size-dependent nitration barrier
related to the high capillary flow in hypertrophied glomeruli. Hyperfiltration or increased GFR is common in young
patients with SCD, and ERPF increases even more than
GFR. Glomerular hypertrophy also relates to focal segmental glomerular sclerosis (FSGS) in SCD patients and
is of greater magnitude than with idiopathic FSGS. The
proteinuria of sickle cell nephropathy is closely associated
pathologically with FSGS. The development of renal failure
in SCD parallels the progression of FSGS and interstitial
fibrosis.
Pathological lesions of SCD nephropathy vary. The earliest distinctive \esion in SCD g\ometu\opat\iy is a svfoendothelial zone lucency in the glomerular capillary loop. This
may foreshadow either capillary collapse or a mesangial
proliferative process. Focal tubulointerstitial fibrosis is
present adjacent to sclerotic glomeruli. In addition to these
findings, hemosiderin granules are seen in the tubular cytoplasm as a residual of hemolysis. There is little evidence
of immune complex deposition in SCD patients with heavy
proteinuria, suggesting that the proteinuria does not result
from an inflammatory reaction.
Diagnosis
The upper limit of the normal GFR in SCD is never
certain, whether determined with the "gold standard" inulin clearance or with the other clearance methods commonly used. In addition, an accurate measure of GFR is
not necessary for most clinical purposes. However, a decrease over time in any measure of GFR in a patient with
SCD, especially when accompanied by proteinuria, is ominous.
Proteinuria in a patient with SCD should be quantitated,
and renal function should be assessed. Diseases other than
sickle cell glomerulopathy should be ruled out, but biopsy
will not often be needed in the SCD patient with insidious
development of proteinuria. Sudden edema and massive
proteinuria could indicate a process other than SCD. The
presence of hypertension is unusual in SCD without nephropathy. Recent data suggest that patients who develop
FSGS are more likely to be severely anemic but do not
have greater frequency of sickle crises.
Treatment

Because it is likely that some aspect of the sickle cell

condition predisposes to nephropathy, it is reasonable to
attempt to minimize sicklingdespite the finding that the

nephropathy of SCD is not obviously related to frequency

of sickle crises.
In treating SCD nephropathy, FSGS should also be addressed. Those factors of coexisting disease states that promote FSGS should be corrected to the extent possible. The
development of FSGS in animal models is accelerated by
a high protein intake, suggesting a treatment in humans
with protein restriction. However, because restriction of
protein intake may increase the risk of delayed growth and
development that is already present in children with SCD,
only the avoidance of an unusually high protein intake is
advisable. It is uncertain whether alterations in hemodynamics can affect the progression of FSGS to renal insufficiency in SCD. Glomerular hyperperfusion and proteinuria
may be mediated by increased glomerular capillary pressure. Although conventional antihypertensives decrease
blood pressure, they do not appear to alter proteinuria.
However, reduction of glomerular capillary pressure by
ACE inhibition could protect the glomemlus from FSGS.
The reduction of proteinuria and glomerular sclerosis by
ACE inhibition in. tats has prompted human trials. In a 2week ttial oi ena\apii\ therapy in 10 patients with mild
SCD nephropathy, proteinuria diminished by 57% but rebounded after treatment withdrawal. Blood pressure and
GFR did not change significantly. A long-term trial is
needed to determine whether ACE inhibition is beneficial in
preventing the progression of sickle cell nephropathy to
renal failure.
Several therapies are available for advanced nephropathy in SCD patients. Treatment of renal failure should
be considered even in the presence of a serious primary
disease; dialysis can be undertaken, and renal transplantation has been as successful in SCD as in other diseases.
Although the anemia in SCD patients with end stage
renal disease is resistant to conventional doses of erythro poietin, larger doses may be necessary and appropriate.
While it has been proposed that erythropoietin could
increase the proportion of fetal hemoglobin, this has not
been observed. However, if erythropoietin is administered
in combination with hydroxyurea, the proportion of fetal
hemoglobin appears to increase. Recently, bone marrow
transplantation has been used for unusual cases of sickle
cell anemia and may be relevant to treatment of SCD nephropathy.
Bibliography
Allon M: Renal abnormalities in sickle cell disease. Arch Intern
Med 150:501-504, 1990. Allon M, Lawson L, Eckman JR,
Delaney V, Bourke E: Effects
of nonsteroidal anti-inflammatory drugs on renal function in
sickle cell anemia. Kidney Int 34:500-506, 1988. Bakir AA,
Hathiwala SC, Ainis H, et al.: Prognosis of the nephrotic
syndrome in sickle glomerulopathy: A retrospective study. Am
J Nephrol 7:110-115, 1987. Bhathena DB, Sondheimer JH:
The glomerulopathy of homozygous sickle hemoglobin (SS) disease: Morphology and pathogen*'
esis. J Am Soc Nephrol 1:1241-1252, 1991. Chatterjee SN:
National study in natural history of renal allografts

313

45. Pol/cystic and Acquired Cystic Diseases

in sickle cell disease or trait: A second report. Transplant Proc
19:33-35, 1987.
Chauhan PM, Kondlapoodi P, Natta CL: Pathology of sickle cell
disorders. Pathol Ann 18:253-276, 1983
deJong PE, Statius van Eps LW: Sickle cell nephropathy:
New insights into its pathophysiology. Kidney Int 27:711-717,
1985.
Falk RJ, Scheinman JI, Phillips G, Orringer E, Johnson A, Jennette
JC: Prevalence and pathologic features of sickle cell nephropathy and response to inhibition of angiotensin converting enzyme.
N Engl J Med 326:910-915, 1992.

Mapp E, Karasick S, Pollack H, Wechsler RJ. Karasick D: Uroradiological manifestations of S-hemoglobinopathy. Semin Roentgenol 22:186-194, 1987. Powars DR, Elliott-Mills DD, Chan L,
et al.: Chronic renal failure
in sickle cell disease: Risk factors, clinical course, and mortality.
Ann Intern Med 115:614 -620, 1991. Sklar AH, Campbell HT,
Caruana RJ, et al.: A population study
of renal function in homozygous sickle cell anemia. Int JArtif
Organs 13:231-236, 1990. Tomson CRV: End stage renal disease
in sickle cell disease: Future
directions. Postgrad Med J 68:775-778, 1992.

POLYCYSTIC AND ACQUIRED

CYSTIC DISEASES
PATRICIA A. GABOW

Renal cysts are a common, if dramatic, end point for a

wide variety of renal insults. Renal cysts occur as part of
both congenital and hereditary disorders or can be acquired
later in life. Cysts can be single or multiple, replacing virtually all the renal parenchyma in some disorders; they can
be clinically inconsequential or contribute to serious complications and sequelae.
CONGENITAL AND SYNDROMIC CYSTIC DISEASES

Congenital disorders occur in the course of fetal development but are not necessarily heritable. Numerous congenital or rare heritable syndromes include renal cysts (Table
1). These disorders are usually diagnosed in utero or early
in life and are defined by a characteristic array of abnormalities. The diversity of these syndromes with cysts serves to
underscore the nonspecific nature of renal cystic disease.
In these disorders the renal cysts often do not pose the
most significant clinical problem.
HEREDITARY RENAL CYSTIC DISEASES

The four important hereditary cystic renal diseases are

autosomal recessive polycystic kidney disease (ARPKD),
autosomal dominant polycystic kidney disease (ADPKD),
von Hippel-Lindau disease, and tuberous sclerosis (TS).

Primer on Kidney Diseases. Second Edition

The latter three have an autosomal dominant pattern of

inheritance and the chromosomal location or locations of
a putative gene have been identified.
Autosomal Recessive Polycystic Kidney Disease
ARPKD is a rare disease with incidence estimates ranging from 1:6000 to 1:40,000. A gene for this disease is
located on chromosome 6. The diversity of phenotypic
presentation in terms of age of onset and systemic manifestations had led some clinicians to question whether the
disorder represents a single genetic disease. However, gene
linkage analysis to date has not demonstrated genetic heterogeneity. The identification of the gene location has permitted prenatal diagnosis with gene linkage analysis in some
families. The disorder can be diagnosed in utero in many
fetuses after the 24th week of gestation by ultrasonography
from the findings of large hyperechoic kidneys and often
oligohydramnios. However, there is no definitive information on how frequently an affected fetus will have a noncharacteristic ultrasonogram. Although prenatal ultrasonography is appropriate in families with a previously
affected child, it is not yet clear that a normal study excludes
disease development later in life. Often the diagnosis of
ARPKD is made after birth and is prompted by the finding
of abdominal masses. Children with large kidneys in utero
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

314

Patricia A. Gabow
TABLE I

Syndromes with Renal Cysts

Sporadic disorders Facioauricularvertebral syndrome
Chromosomal abnormalities
Trisomy 21, 18, 13, 9
Triploidy
Deletion of 17p, 4p Duplication of lOq, 3q, Iq
Sex chromosome abnormalities 45,X; 47,XYY
X-linked recessive disorders
Oculocerebrorenal syndrome of Lowe
Orofacial-digital syndrome
Autosomal recessive disorders
Mcckel's syndrome Miranda's
syndrome Smith Lemli-Opitz
syndrome Lissencephaly type
II Baret- Biedl syndrome
Hydrolethalus syndrome Fryns
syndrome
Orofacial-digital syndrome type I Short-ribpolydactyl syndrome type I and II
Acrocephalopolydactylous dyspl asia
Cerebrohepatorenal syndrome (Zellweger) Glutaric
acidemia type II Neonatally lethal
adrenoleukodystrophy Chondysplasia calcificans
punctate, rhizomelic type
From Gilberg-Barness EF, Opitz JM, Barness LA:
In: Spitzer A, Avner ED (eds) Inheritance of Kidney
and Urinary Tract Diseases. Kluwer Academic, Boston,
pp. 327 -400, 1990, with kind permission from Kluwer
Academic Publishers.

often have immediate respiratory distress from pulmonary

hypoplasia and pneumothorax; these complications can result in perinatal death.
ARPKD is characterized by the development of renal
cysts, or, more correctly, tubular ectasia of the collecting
ducts, and congenital hepatic fibrosis. In older children,
the kidneys appear large with increased echogenicity in the
medulla and macrocysts on ultrasonograms. Pathologically
the liver displays increased numbers of dilated bile ductules
and fibrosis in the portal areas. Hepatic macrocysts are
unusual. The clinical manifestations of the hepatic disease
include cholangitis and portal hypertension with esophageal varices and gastrointestinal bleeding. The renal manifestations include hypertension, which is often severe, urinary tract infection, and, ultimately, renal failure.
Children who present early in life tend to show a predominance of renal abnormalities; those who present later in
childhood tend to have a predominance of hepatic manifestations. Recent studies have shown a better prognosis than
previously reported; children who survive the first month
of life do quite well, with a 78% probability of surviving
beyond 15 years of age. Early identification and aggressive
treatment of the hypertension may improve the renal
disease prognosis in these children. A few children with

ARPKD have successfully undergone simultaneous renal

and liver transplantation.
The other major diagnoses to be considered in children
who present with renal cysts are TS, ADPKD glomerulocystic disease, and the syndromes listed in Table 1. The
presence of other characteristic phenotypic abnormalities
and the pattern of inheritance provide valuable information that facilitates making specific diagnosis. Although
ADPKD cannot be differentiated from ARPKD solely on
the basis of ultrasonographic findings, the presence of dis crete renal cysts, particularly in a fetus, favors ADPKD.
The single most reliable means of differentiating between
these two hereditary childhood cystic diseases is to perform renal ultrasonography in the parents. The diagnosis
ARPKD is favored if neither parent has the characteristic
findings of ADPKD. In some instances, a liver and/or renal
biopsy in the affected child may be helpful. Although congenital hepatic fibrosis has also been reported in ADPKD,
its virtual 100% occurrence in ARPKD and its rare occurrence in ADPKD strongly favor the diagnosis of ARPKD.
Autosomal Dominant Polycy stic Kidney Disease
ADPKD is both the most common hereditary renal dis ease and one of the most common hereditary disease in
this country. It affects 1:400 to 1:1000 Americans and is
the fourth most common cause of end stage renal disease
(ESRD). At least three different genes can produce the
disorder. The ADPKD 1 gene is responsible for approximately 90% of the disease in the white population of Euro pean ancestry. Recently cloned, this gene is located on the
short arm of chromosome 16 and is a very complicated
large gene. The gene product, polycystin, is thought to be a
membrane glycoprotein involved in cell-cell or cell-matrix
interactions. The ADPKD2 gene, which is on chromosome
4, has also been cloned. The chromosomal location of the
ADPKD3 gene is yet to be determined. The frequency of
the gene distribution in other racial and ethnic groups has
not been defined.
The identification of the location of the ADPKD genes
has added gene linkage analysis to the diagnostic armamentarium for this disease. This technique permits the diagnosis
to be made in utero or in children or adults prior to the
development of renal cysts if multiple family members can
be tested. Blood samples from at least two affected family
members are necessary to identify the DNA marker that
is linked to the ADPKD gene. Commercially available
probes make this technique applicable in most ADPKD1
families; probes are now available for ADPKD2 families.
Nonetheless, gene linkage analysis is not generally needed
to make the diagnosis. Ultrasonography remains the diagnostic test of choice because it is sensitive, relatively inexpensive, safe and able to provide information about the
degree of structural renal involvement, while requiring only
the participation of the individual wishing the information
(Fig. 1). Recent studies have demonstrated that approximately 3 to 24% of individuals with ADPKD1 will have a
normal renal ultrasonogram before 30 years of age. In
children the percentage of false negative ultrasonograms

45. Pol/cystic and Acquired Cystic Diseases

315

FIGURE I Renal ultrasonogram of a 44-year-old male with autosomal dominant poly cystic kidney disease. The kidney is moderately enlarged and contains multiple cysts of
variable size.

is even higher, approaching 40% in children under age 5

years. Therefore, gene linkage analysis is likely to be of
value in establishing the gene status only in these individuals and in fetuses. However, there are caveats regarding
ADPKD screening of asymptomatic children and adults,
because it has the potential for adverse psychological consequences and can affect insurability. Moreover, genetic
screening of fetuses does not predict the course of the
disease. In contrast, children diagnosed in utero or in the
first year of life via a characteristic renal ultrasonogram
tend to have a more severe disease than individuals diagnosed later in life and have an approximately 27% chance
of renal failure or renal insufficiency before adulthood.
ADPKD is not only a renal disease; it is a systemic
disorder with a myriad of manifestations (Table 2). The
renal manifestations of clinical importance include hypertension, infection, nephrolithiasis, acute and chronic pain,
and renal failure. The renal cysts are responsible for many
of these complications (Fig. 2). Hypertension occurs in
about 30% of ADPKD children, 60% of adults prior to
ESRD, and about 80% of ADPKD patients with ESRD.
Hypertension is more common in individuals with more
severe cystic involvement and may be ameliorated at least
temporarily, in adults, by cyst decompression. It has been
suggested that the cysts alter the intrarenal circulation creating areas of renal ischemia with activation of the reninangiotensin-aldosterone system. Renin and aldosterone
levels are higher in ADPKD hypertensive patients than in
patients with a similar degree of essential hypertension.
Because of the role of the renin-angiotensin system in
ADPKD hypertension, angiotensin converting enzyme inhibitors are often the drug of choice in treating hypertension in these patients. However, there are no long-term
studies comparing the renal outcomes in patients treated
with different antihypertensive drugs.

TABLE 1

Manifestations of Autosomal Dominant Polycystic

Kidney Disease in Adults
Extrarenal manifestations
Gastrointestinal Hepatic cysts
Cholangiocarcinoma"
Congenital hepatic fibrosis"
Pancreatic cysts Diverticula
Cardiovascular
Cardiac valvular abnormalities Intracranial
aneurysms Thoracic and abdominal aortic
aneurysms (?)
Miscellaneous cysts
Ovarian cysts
Arachnoid cysts"
Pineal cysts"
Splenic cysts"
Renal manifestations
Anatomical Renal
cysts Renal
adenomas
Complications
Hypertension
Hematuria and/or hemorrhage
Acute and chronic pain
Urinary tract infectionbladder, interstitium, cysts
Nephrolithiasis Nephromega ly Renal failure
" Rare.

316

FIGURE 2 A nephrectomy specimen of a kidney from a patient

with autosomal dominant polycystic kidney disease. The kidney
appears moderately enlarged with cysts of varying sizesome of
which appear hemorrhagic. Reprinted by permission of The New
England Journal of Medicine, Gabow PA: 329:332-342,1993. Massachusetts Medical Society. All rights reserved.

Acute abdominal, flank, or back pain are common and

can result from renal and extrarenal complications. One
renal etiology of pain is infection in the interstitium or the
cysts. ADPKD patients who appear to have simple cystitis
or pyelonephritis can be treated as one would treat any
other patients with these maladies. A patient with presumed pyelonephritis who fails to demonstrate an appropriate response should be considered to have a renal cyst
infection. Renal cyst infections are also suggested by the
presence of discrete renal tenderness, positive blood cultures, and absence of white blood cell casts. The failure of
antibiotic therapy reflects the lack of cyst penetration of a
majority of the antibiotics. Those antibiotics with good cyst
fluid accumulation include trimethoprim-sulfamethoxa zole, ciprofloxacin, and chloramphenicol; hence they are
the antibiotics of choice for patients with presumed cyst infection.
Nephrolithiasis, which occurs in as many as 34% of
ADPKD patients, can also present as an acute pain syndrome. Stone formation likely reflects both intrarenal urinary stasis due to cysts and metabolic abnormalities includ-

Patricia A. Gabow

ing decreased citrate excretion. Acute pain can also be

caused by cyst hemorrhage; this is not always accompanied
by hematuria. Diagnosis of this complication is usually possible by computed tomography.
Chronic disabling pain occurs in a small group of
ADPKD patients. These patients can benefit from cyst
decompression procedures, such as surgical unroofing of
cysts, percutaneous or laparoscopic cyst decompression,
and cyst drainage and alcohol sclerosis, which prevents
rapid fluid reaccumulation.
Although renal failure is the most serious renal complication of ADPKD, it is not an invariable outcome. Only
about 50% of ADPKD patients progress to ESRD before
60 years of age. The factors associated with a worse renal
outcome include the ADPKD 1 gene (in contrast to the
ADPKD2 gene), male gender, African origin, sickle hemo globin, earlier age of presentation, hypertension, episodes
of gross hematuria, and urinary tract infections in males.
Data on the effect of pregnancy on renal outcome are
variable, but several studies suggest that more than three
or four pregnancies, particularly in hypertensive women,
is associated with a worse long-term renal outcome. In
experimental models of renal cystic disease a variety of
interventions including treatment with taxol, prednisone,
sodium and potassium bicarbonate, and lovastatin as well
as protein restriction have been shown to ameliorate the
renal cystic disease. A few human studies have been performed, the largest being the Modification of Diet in Renal
Disease (MDRD) study, which failed to show efficacy of
protein restriction or rigorous control of blood pressure
on renal disease progression. However, the patients had
advanced renal disease and the study was of relatively short
duration. Hence the efficacy of these interventions early
in the course of disease remains undetermined.
Extrarenal cystic manifestations include hepatic cysts,
perhaps ovarian cysts, and rarely, pancreatic, splenic, and
arachnoid cysts. Hepatic cysts occur in about 60% of the
overall ADPKD population. However, hepatic cysts are
very rare in childhood. Their prevalence increases with age
with a peak occurrence in the fifth to sixth decade. Massive
hepatic cystic disease appears to be mainly a disorder of
ADPKD women; pregnancy is a major correlate of hepatic
cyst number and size. Although massive hepatic cystic dis ease can result in substantial abdominal distention and
pain, it only very rarely affects hepatic function; other
causes should always be sought for abnormal liver function
tests in an ADPKD patient. Hepatic cysts can occasionally
become infected; both antibiotics and drainage appear to
be required for recovery. The other extrarenal cystic manifestations usually have no clinical sequelae.
Extrarenal noncystic manifestations include gastrointes tinal, cardiovascular, and musculoskeletal abnormalities.
Diverticular disease is more frequent and more severe
among ADPKD patients with ESRD than in non-ADPKD
ESRD patients and age-matched controls. One of the most
common extrarenal manifestations is cardiac valvular abnormalities, particularly mitral valve prolapse which occurs
in 26% of ADPKD patients compared to 2% of a non-

45. Pol/cystic and Acquired Cystic Diseases

ADPKD control group. It is often accompanied by palpitations and atypical chest pain.
One of the most devastating manifestations of ADPKD
is rupture of an intracranial aneurysm. About 5 to 10% of
ADPKD patients have intracranial aneurysms. However,
the risk may not be evenly distributed across all families; some data suggest that aneurysms cluster in certain
ADPKD families. Although it is clear that aneurysms can
be detected before rupture by noninvasive methods such
as dynamic computed tomography or magnetic resonance
angiography, screening of all ADPKD patients for aneurysms is not cost-effective. Therefore, the current recommendations are to screen only those patients who have
a family history of aneurysm, those undergoing elective
surgery with a high likelihood of developing hypertension
during surgery, or those engaged in an activity in which
sudden loss of consciousness would be associated with
grave consequences to themselves or others. Of course, any
patient with acute symptoms compatible with aneurysmal
rupture requires immediate investigation. However,
chronic headaches do not identify patients with aneurysms.
Von Hippel-Lindau Disease

This systemic disease is an uncommon autosomal dominant disorder which occurs in 1:36,000 live births. The
defective gene is on chromosome 3. The eye, the central
nervous system (CNS), the pancreas, the adrenal glands,
the epididymis, and the kidney may be involved. The manifestations of the disease include retinal angiomas and cerebellar and spinal hemangioblastomas, pancreatic cysts and
islet cell carcinomas, and pheochromocytomas. The latter
may be bilateral. The epididymal abnormalities include
cystic and solid epididymal masses. Renal cysts and renal
cell carcinomas are observed. As many as 75% of patients
with this disease will have renal cysts. These cysts pathologically are more like those in acquired cystic disease and less
like simple cysts (see below) in that the cyst epithelium
displays a histologic spectrum from normal-appearing single-layered epithelium to epithelial hyperplasia to renal
cell carcinoma. It is estimated that between 38 and 55% of
patients with Von Hippel-Lindau disease develop renal
cell carcinoma which can be bilateral and multicentric in
up to 60% of the cases. Renal cell carcinoma occurs on
average in the fourth and fifth decade, considerably earlier
than in the general population. It accounts for 20 to 50%
of the deaths in these patients.
Because of the devastating CNS and renal complications,
regular screening by imaging the brain and abdomen in
at-risk family me mbers has been recommended. Genetic
screening permits limiting these studies to those family
members who are found to be gene carriers.
Tuberous Sclerosis

TS is another autosomal dominant disease with similarities to ADPKD. As with ADPKD, different genes can produce the disease; these genes occur on chromosomes 9 and
16very close to the ADPKD1 gene. TS is characterized

317

by epilepsy, mental retardation, skin abnormalities including adenoma sebaceum and ash-leaf spots, and hamartomas
of many organs including angiomyolipoma of the kidneys
and renal cysts. Angiomyolipomas occur in 40 to 60% of
patients with or without renal cysts, and renal cysts occur
alone in about 18% of patients. In neonates or young children the occurrence of isolated renal cysts can lead to the
misdiagnosis of ARPKD or ADPKD. Particularly in the
presence of aggressive cystic disease, renal failure can occur. In fact, renal disease is the most common serious complication of this disease in patients over 30 years of age.
ACQUIRED CYSTIC DISEASES

There are several types of acquired cystic disease: simple

renal cysts, hypokalemia-related cystic disease, and acquired cystic disease (ACD). Simple renal cysts increase
with age, occurring in less than 0.1% of children and up to
20% of individuals over age 50. Although cysts are usually
single and unilateral, they may be multiple and bilateral.
However, the latter should raise the question of a polycystic
renal disease. On ultrasonography simple cysts are smoothwalled without internal debris. As they are only very rarely
malignant, they require no further evaluation, if these characteristic ultrasonographic features are present.
Multiple renal cysts have been described in patients with
hypokalemia secondary to hyperaldosteronism. In some of
these patients the renal cysts regressed after successful
resection of the adrenal adenoma. The cellular proliferative
stimulus of hypokalemia may be pathogenetic in this
setting.
ACD refers to a specific disorder in which renal cysts
develop in the kidneys of both adults and children with
ESRD due to a noncystic renal disorder. Although its
pathogenesis remains to be defined, abnormal cell proliferation and intratubular obstruction with oxalate crystals
have been suggested. Risk factors for the development of
ACD include the duration of renal insufficiency, including
years on dialysis, and in some studies male gender and
African ancestry. ACD occurs in 7 to 22% of patients with
renal failure who are not on dialysis, 40% of patients who
have been on dialysis for 3 years, and 80 to 90% of patients
who have been on dialysis for 4 to 10 years. In some patients
a successful renal transplant leads to the regression of
ACD. In contrast, ACD may occur in the transplanted
kidneys of patients with chronic rejection. Most ACD patients are asymptomatic; among symptomatic patients the
most common manifestations are pain, hematuria, and fever. In addition, cyst hemorrhage and retroperitoneal
bleeding (presumably from cyst rupture), erythrocytosis,
hypercalcemia, and hypoglycemia have also been reported.
Both benign and malignant renal neoplasms are common;
the most serious sequela of ACD is renal cell carcinoma.
This complication has been estimated to be 50 times more
frequent in dialysis patients with ACD than in the general
population. As with the disease itself, male gender and
African ancestry appear to be risk factors for the development of malignancy. Although agreement on the metastatic
potential of these tumors is lacking, it is clear that meta-

318

Martin C. Gregory

static disease does occur in at least some ACD patients.

The survival is similar to that of other patients with renal
cell carcinoma. The occurrence of malignancy and its potential to metastasize has raised concern about the need
for screening. Although some decision analysis data have
suggested that screening is not cost-effective, many clinicians recommend a screening program. One guideline includes an ultrasonogram at the onset of dialysis and followup ultrasonography every 3 years thereafter. There is also
rationale for opting to screen only males in good health
who have been on dialysis for more than 7 years. Because
metastatic potential appears to relate to tumor size, any
tumor greater than 2 cm is a strong indicator for nephrectomy.
Acknowledgments
This work was supported by Grant 5 P01 DK34039, Human
Polycystic Kidney Disease (PKD), awarded by the Department
of Health and Human Services, Public Health Service, NIDDK,
and the Clinical Research Center, Grant MORR-00051 from the
General Clinical Research Centers Research Program of the Division of Research Resources, National Institutes of Health.

Bibliography
Chapman AB, Rubinstein D, Hughes R, et al.: Intracranial aneurysms in autosomal dominant polycystic kidney disease. N Engl
J Med 327:916-920, 1992.
Choyke PL, Glenn GM, Walther MM, Patronas NJ, Linehan WM,
Zbar B: von Hippel-Lindau disease: Genetic, clinical, and imaging features. Radiology 194:629 -642, 1995.
Pick GM, Gabow PA: Hereditary and acquired cystic disease of
the kidney (review). Kidney Int 46:951-964, 1994.

Pick GM, Duley IT, Johnson AM, Strain JD, Manco-Johnson ML,
Gabow PA: The spectrum of autosomal dominant polycystic
kidney disease in children. J Am SocNephrol 4:1654-1660,1994.
Gabow PA, Johnson AM, Kaehny WD, et al.: Factors affecting the
progression of renal disease in autosomal-dominant polycystic
kidney disease. Kidney Int 41:1311-1319, 1992.
Gilbert -Barness EF, Opitz JM, Barness LA: Heritable malformations of the kidney and urinary tract. In: Spitzer A, Avner ED
(eds) Inheritance of Kidney and Urinary Tract Diseases. Kluwer
Academic, Boston, pp. 327-400, 1990.
Guay-Woodford LM: Autosomal recessive polycystic kidney disease: Clinical and genetic profiles. In: Watson ML, Torres VE
(eds) Polycystic Kidney Disease. Oxford University Press, New
York, pp. 237-266, 1996.
Harris PC, Ward CJ, Peral B, Hughes J: Polycystic kidney disease
1: Identification and analysis of the primary defect. / Am Soc
Nephrol 6:1125-1133, 1995.
Huston J, Torres VE, Sulivan PP, Offord KP, Wiebers DO: Value
of magnetic resonance angiography for the detection of intracranial aneurysms in autosomal dominant polycystic kidney disease. J Am Soc Nephrol 3:1871-1877, 1993.
Ishikawa I: Acquired cystic disease: Mechanisms and manifestations. Semin Nephrol 11:671-684, 1991.
Klahr S, Breyer JA, Beck GJ, et al.: Dietary protein restriction,
blood pressure control, and the progression of polycystic kidney
disease. / Am Soc Nephrol 5:2037-2047, 1995.
Linehan WM, Lerman MI, Zbar B: Identification of the vo n
Hippel-Lindau (VHL) gene. JAMA 273:564-570, 1995.
Roach ES, Delgado MR: Tuberous sclerosis. Dermatol Clin
13:151-165, 1995.
Sarasin FP, Wong JB, Levey AS, Meyer KB: Screening for acquired cystic kidney disease: A decision analytic perspective.
Kidney Int 48:207-219, 1995.
Torres VE, King BF, Holley KE, Blute ML, Gomez MR: The
kidney in the tuberous sclerosis complex. Adv Nephrol 23:4370, 1994.

ALPORT'S SYNDROME AND

RELATED DISORDERS
MARTIN C. GREGORY

Alport's syndrome is a disease of collagen that affects

the kidneys always, the ears often, and the eyes occasionally. Cecil Alport described the association of hereditary
hematuric nephritis with hearing loss in a family whose

Primer on Kidney Diseases, Second Edition

affected males died in adolescence. Genetic advances have

broadened the scope of the condition to include ocular
defects, platelet abnormalities, late-onset renal failure, and
normal hearing in some families. At least 80% of kindreds
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

319

46. Alport's Syndrome and Related Disorders

are X-linked and over half of those result from a mutation

of COL4A5, the gene located at Xq22 that codes for the a5
chain of type IV collagen. Autosomal recessive inheritance
occurs in perhaps 5% of cases and autosomal dominant
inheritance has been shown in a few kindreds with associated thrombocytopathy and in very rare kindreds with otherwise typical Alport's syndrome.
JUVENILE AND ADULT FORMS

This distinction is fundamental to the understanding of

Alport's syndrome. Renal failure tends to occur at a similar
age in all male hemizygotes in a given kindred, but this
age can differ widely between kindreds. Uremia in males
occurs in childhood or adolescence in some families, and
in the late thirties in others. The families with early onset
of renal failure in males are termed "juvenile" and those
with renal failure in middle age are called "adult" type
nephritis. Extrarenal manifestations tend to be more prominent in the juvenile kindreds. Moreover, because males in
juvenile kindreds do not commonly survive to reproduce,
these kindreds tend to be small and frequently arise from
new mutations. Adult type kindreds are typically much
larger and new mutations occur infrequently.
CLASSICAL GENETICS

In most kindreds inheritance is X-linked. This was suggested by classical pedigree analysis, strengthened by formal likelihood analysis, proved by extremely tight linkage
to restriction fragment length polymorphic markers
(RFLPs), and finally cemented by identification of mutations.
MOLECULAR GENETICS

Causative mutations of COL4A5, the gene coding for

the a5 chain of type IV collagen, appear consistently in
many kindreds. Deletions, point mutations, and splicing
errors occur. Deletions and splicing errors result in severe
renal disease and severe hearing loss. Missense mutations
may cause juvenile disease with hearing loss or adult dis ease with or without hearing loss. Deletions involving the
5' end of the COL4A5 gene and the 5' end of the adjacent
COL4A6 gene occur consistently in families with esophageal and genital leiomyomatosis. Homozygotes or mixed
heterozygotes for mutations of COL4A3 or COL4A4 can
develop autosomal recessive Alport's syndrome. Some
cases of autosomal recessive Alport's syndrome may be
homozygotes for a COL4A4 mutation that causes benign
familial hematuria.

nephritis are usual but not inevitable after retransplantation. The anti-GBM antibodies developing after transplantation are heterogeneous. All will stain normal GBM but
only some stain epidermal basement membrane. In most
kindreds with Alport's syndrome the GBM of affected
males fails to stain in the normal fashion with anti-GBM
sera, and the GBM of female heterozygotes stains in an
interrupted fashion. Certain sera have in common activity
against 3(IV) and also epidermal basement membrane.
The "Alport antigen" that these sera recognize is a
26-kDa monomer belonging to the a3(IV) chain.
BIOCHEMISTRY

The open meshwork of interlocking molecules of type IV

collagen that forms the framework of glomerular basement
membrane is composed of heterotrimers of a chains. Most
of these heterotrimers consist of two al(IV) chains and
one a2(IV) chain, but an important minority contain an
a3(IV), an a4(IV), and an a5(IV) chain. The primary
chemical defect in Alport's syndrome most commonly involves the 5(IV) chain, but faulty assembly of the a3,4,5
heterotrimer produces similar pathology in glomerular,
aural, and ocular basement membranes regardless of which
a chain is defective. As an illustration of failure of normal
heterotrimer formation, many patients whose genetic defect is in the gene coding for the a5(IV) chain lack demonstrable <*3(IV) chains in GBMs.
PATHOLOGY

In young children light microscopy may be normal or

near normal. An increased number of "fetal" glomeruli
may be seen. As disease progresses, interstitial and tubular
foam cells may become quite prominent, although they
can also be found in many other conditions. Eventually
progressive glomerular sclerosis and interstitial scarring
develop (Fig. 1). Routine immunofluorescence examina-

IMMUNOCHEMISTRY

Patients with Alport's syndrome lack a component of

the glomerular basement membrane (GBM). After transplantation, about 10% of Alport males develop anti-GBM
nephritis, presumably because tolerance to this normal antigen has not been acquired. Recurrences of anti-GBM

FIGURE I High-power photomicrograph of foam-filled tubular

and interstitial cells in a renal biopsy from a patient with Alport's
syndrome. Relatively normal proximal tubular cytoplasm stains
red in the tubules on the left and at the top. The remaining cells
appear "foamy" because of the spaces left where lipids have been
eluted during processing.

Martin C. Gregory

320

FIGURE 2 (A) Medium power electron micrograph of normal human glomerular basement membrane. The lamina
densa of the basement membrane is of uniform density, texture, and thickness throughout its length. (B) Electron
micrograph of glomerular basement membrane from a patient with Alport's syndrome at the same magnification as (A).
The glomerular basement membrane varies in thickness. It is thinner than normal at the top left of the capillary loop.
At middle left the lamina densa is split into two layers and at bottom right the basement membrane is very broad and
frays into several lamellae. Electron micrographs kindly supplied by Dr. Daniel A. Terreros.

tion is negative. The characteristic features are seen by

electron microscopy (Fig. 2). The GBM is thickened up to
two or three times its normal thickness, split into several
irregular layers, and frequently interspersed with numerous
electron dense granules about 40 nm in diameter. In florid
cases of juvenile types of disease the lamellae of basement
membrane material may branch and rejoin in a complex
"basket weave" pattern. Early in the development of the
lesion, thinning of the GBM may predominate or may be
the only abnormality visible. The abnormalities in children
or adolescents with adult-type Alport's syndrome may be
unimpressive or suggestive of thin basement membrane
disease.

less extensively. Urinary erythrocytes are dysmorphic, and

red cell casts can usually be found. Proteinuria is very
variable; occasionally it reaches nephrotic levels.
Hemizygous males inevitably progress to end stage rend
disease (ESRD). This occurs at widely variable ages that
are fairly constant within each family. Heterozygous females are generally much less severely affected. Around
one fifth of them will develop ESRD, usually after the age
of 50, but renal failure in girls in their teens and even
younger does occur.
In families with autosomal inheritance, females are affected as severely and as early as males. Renal failure often
occurs before the age of 10 years in homozygotes with
autosomal recessive Alport's syndrome.

CLINICAL FEATURES
Renal Features

Nonrenal Features

Uninterrupted microscopic hematuria is present from

birth in affected males. Hematuria may become visible
after exercise or during fever; this is more common in
juvenile kindreds. Microscopic hematuria has a penetrance
of approximately 90% in heterozygous females in adulttype kindreds. In juvenile kindreds, the penetrance of hematuria in females may be lower, but this has been studied

Hearing Loss
Bilateral high-frequency cochlear hearing loss is present
in many but not all kindreds. Some authors do not use the
term Alport's syndrome unless there is substantial hearing
loss in the family. Regardless of terminology, the important
clinical point is that X-linked nephritis progressing to
ESRD can occur in families without overt hearing loss.

46. Alport's Syndrome and Related Disorders

Expectation of hearing loss has resulted in many diagnoses

being missed and has introduced a substantial ascertainment bias against those with normal hearing. In families
with juvenile type disease, hearing loss is almost universal
in male hemizygotes and common in severely affected female heterozygotes.
Hearing loss affects all frequencies but is maximal at 6
through 8 kHz. The loss is also severe at 8 to 20 kHz, but
these frequencies are not covered by conventional audiometry. In adult-type Alport's syndrome with hearing loss,
there is no perceptible deficit until age 20, but loss progresses to 60-70 dB at 6-8 kHz at ages over 40. Hearing
loss occurs earlier in juvenile kindreds. The kinetics are not
well established, but many gradeschoolers or adolescents
require hearing aids.
Ocular Defects
These appear confined to juvenile kindreds. Myopia, arcus juvenilis, and cataracts occur but lack diagnostic specificity. Three changes that are present in a minority of kindreds but that are nearly diagnostic are anterior lenticonus,
posterior polymorphous corneal dystrophy, and retinal
flecks. Anterior lenticonus is a forward protrusion of the
anterior surface of the ocular lens. It results from a weakness of the type IV collagen forming the anterior lens
capsule. The resulting irregularity of the surface of the lens
causes an uncorrectable refractive error. The retina cannot
be clearly seen by ophthalmoscopy and with a strong positive lens in the ophthalmoscope the lenticonus can often
be seen through a dilated pupil as an "oil drop" or circular
smudge on the center of the lens. Retinal flecks are small
yellow or white dots scattered around the macular or in
the periphery of the retina. If sparse they may be difficult
to distinguish from small hard exudates.
Leiomyomatosis
Several X-linked families show the precocious development of striking leiomyomas of the esophagus and female
genitalia in association with Alport's syndrome. These tumors are frequently large and multiple. They may bleed
or obstruct and their resection can be difficult. All families
described so far have had a deletion at the 5' end of the
COL4A5 gene that extends into the COL4A6 gene.

321
certainty whether a particular individual carries a defective gene.
The key to diagnosis is to suspect the possibility of Alport's syndrome in any patient with otherwise unexplained
hematuria, glomerulopathy, or renal insufficiency. In many
cases the familial nature of the condition will not be immediately apparent. Inquiry into the family must be detailed
and insistent. The patient is a usually a young male.
Chances are that he knows little of his distant relatives,
but his mother will likely know more of the family details.
Male relatives linked to the patient through one or more
females may have renal failure. Check a urine sample from
the patient's parents, particularly his mother, for microscopic hematuria. Hearing loss is a helpful clue, but it is
crucial to remember that hearing loss is neither a sensitive
nor a specific marker of Alport's syndrome. Indeed, it is
neither necessary nor sufficient for the diagnosis. Most
patients with hearing loss and renal disease do not have
Alport's syndrome, but rather a variety of other disorders
most commonly glomerulonephritis and a banal cause for
hearing loss such as noise exposure or aminoglycoside
therapy.
TREATMENT

There is no specific treatment for Alport's syndrome.

General measures to retard the progression of renal failure,
such as effective treatment of hypertension and modest
protein restriction appear warranted, but are unproved. As
for other forms of progressive renal disease, angiotensinconverting enzyme inhibitors may offer a specific advantage. Males should wear hearing protection in noisy surroundings. Hearing aids improve but do not completely
correct the hearing loss. Tinnitus is generally resistant to
all forms of therapy; hearing aids may render it less disruptive by amplifying ambient sounds.
Retinal lesions do not appear to affect vision and thus
need no therapy. The serious impairment of vision caused
by lenticonus or cataract cannot be corrected with spectacles or contact lenses. Lens removal with reimplantation
of an intraocular lens is standard and satisfactory treatment.
RE L A T E D D I S O R D E R S Familial Thin
Basement Membrane Disease

DIAGNOSIS

No single feature is pathognomonic. The diagnosis is

made by finding hematuria in multiple family members,
together with a history of renal failure in related males,
and reinforced by a biopsy showing characteristic ultrastructural changes in the proband or a relative. Immunofluorescence examination of the biopsy should include
staining with fluoresceinated anti-GBM antibody; this will
help distinguish Alport's syndrome from familial thin GBM
disease. In families with a previously defined mutation,
molecular diagnosis of affected males and gene carrying
females is possible. In relatively large families without a
known mutation, segregation analyses can decide with near

Familial thin basement membrane disease or benign familial hematuria is an autosomal dominant basement membrane glomerulopathy. Some cases result from mutations
of the COL4A4 gene at 2q35-q37. Ultrastructurally the
GBM is uniformly thinned to about half its normal thickness. There is no disruption or lamellation of the GBM,
nor are any other abnormalities of the glomeruli, tubules,
vessels, or interstitium visible by light, immunofluorescence, or electron microscopy. Renal insufficiency does
not occur. Longevity is unaffected by this condition and
survivors into the ninth decade are recorded. Minor degrees
of lamellation of the GBM and hearing loss have been
described in some families. The significance of these find-

Martin C. Gregory

322
ings is uncertain in view of the ease with which adulttype Alport's syndrome can be mistaken for familial thin
basement membrane disease.
Once the precise diagnosis is established the patient and
family can be spared further invasive tests and an appropriate prognosis given to them and to insurers. Unfortunately, the distinction between Alport's syndrome and benign familial hematuria is not always easy to make.
Intermediate forms appear to exist.
To be sure of the pattern of inheritance requires a large
pedigree with accurate diagnoses in all the family members.
A single mistaken diagnosis from incidental renal disease,
inaccurate urinalysis, or incomplete penetrance may vitiate
the conclusions about the pattern of inheritance in the
entire pedigree. Even biopsy evidence is fallible. Early
cases of Alport's syndrome may show ultrastructural
changes indistinguishable from those of benign familial
hematuria. This is particularly likely to occur if a child
from an adult-type Alport's kindred is biopsied. Stability
of serum creatinine for several years in this child is as
compatible with adult-type Alport's syndrome as it is with
benign familial hematuria. It is not clear whether lamellation is an occasional feature of thin GBM disease, or
whether some families with adult-type Alport's syndrome
were misdiagnosed as familial thin GBM disease. The interpretation is further clouded because in some families with
familial thin GBM disease, homozygous individuals have
Alport's syndrome. In these families, autosomal dominant
thin GBM disease and autosomal recessive Alport's syndrome are caused by the same mutations.
Alport's Syndrome with Thrombocytopathy
(Epstein's Syndrome)

This uncommon autosomal dominant variant of Alport's

syndrome associates moderate thrombocytopenia with severe hearing loss and renal failure in both males and females. The platelets are much larger than normal (about
7 /ji.m in diameter), and there is a mild or moderate bleeding
tendency. In some families there are inclusion bodies
(Fechtner granules) in leukocytes.
Autosomal Recessive Alport's Syndrome

A few children who develop severe renal disease before

the age of 10 years have homozygous or mixed heterozygous mutations of the genes for the a3 or o4 chains of type
IV collagen. Boys and girls are equally severely affected.
The heterozygous parents may or may not show hematuria.
APPROACH TO THE PATIENT WITH
HEREDITARY NEPHRITIS

Although Alport's syndrome is less common than polycystic kidney disease, it is probably more common than is
generally appreciated. Important differential diagnoses of
hematuria in young persons are IgA nephropathy or other
glomerulonephritis, renal calculi, or medullary sponge kidney. The differential diagnosis of familial renal disease with

hematuria includes familial thin GBM disease and familial

IgA nephropathy, and, of course, polycystic kidney disease.
Familial renal diseases without hematuria that might be
confused include polycystic kidney disease, medullary cystic disease, and some poorly defined forms of inherited
glomerular and tubulointerstitial renal disease. If a patient
with unexplained hematuria or renal insufficiency has a
family history of hematuria or renal failure, the family
history should be extended, concentrating particularly on
the mother's male relatives. Finding hearing loss strengthens, and finding specific ocular lesions greatly strengthens,
suspicion for Alport's syndrome. Renal biopsy will generally be indicated in one or two family members, but once
the diagnosis of a hereditary basement membrane nephropathy is established in a family, it is hard to justify further
biopsies unless there are features suggesting another diagnosis. The extent of investigation will be guided by your
judgment and will relate inversely to the strength of the
family history. For example, a young man on the line of
descent of a known Alport family whose urine contains
dysmorphic erythrocytes needs minimal investigation, perhaps nothing more than a serum creatin ine unless there
are additional clinical features suggesting a systemic dis ease. With an uncertain family history, an intravenous pyelogram or abdominal flat plate and renal sonogram, ANA,
and possibly anti-GBM antibody, ANCA, and renal biopsy
may be appropriate. Genetic testing is presently of very
limited applicability in sporadic cases and small kindreds
because most families have "private" mutations. Patients
may wish to enroll in research studies that may eventually
identify the mutation in their family. There are a dozen or
more small families and two very large kindreds in the
US with mutations for which specific mutation tests are
available. In these families, direct mutation analysis can
quickly establish whether an individual is a gene carrier.
It is not yet clear whether the two common mutations
are sufficiently widespread to justify screening for them in
adults with unexplained renal failure or before undertaking
a renal biopsy in an adult or child with hematuria.
Patients with any hereditary nephropathy should be informed of the nature of their disease and perhaps given a
copy of the renal biopsy report in order to avoid unneces sary further investigation. Similar considerations apply to
family members who are potentially gene carriers. Those
with Alport's syndrome should be followed for elevation
of blood pressure or serum creatinine. The frequency of
follow-up will depend on the anticipated age of onset of
renal deterioration in the family and will become closer as
this age is approached. Those with familial thin basement
membrane disease should be checked about every 2 years,
because some may ultimately have Alport's syndrome.
Bibliography
Alport AC: Hereditary familial congenital haemorrhagic nephritis.
Br Med J 1:504-506, 1927. Barker DF, Fa in PR, Goldgar DE,
et al.: High - density genetic
and physical mapping of DNA markers near the X-linked Alport
syndrome locus: Definition and use of flanking polymorphic
markers. Hum Genet 88:189-194, 1991.

47. Medullary Cystic Disease

323

Barker DF, Hostikka SL, Zhou J, et al.: Identification of mutations

in the COL4A5 collagen gene in Alport syndrome. Science
248:1224-1227, 1990.
Dische FE, Brooke IP, Cashman SJ, et al.: Reactivity of mono clonal antibody PI with glomerular basement membrane in thin membrane nephropathy. Nephrol Dial Transplant 4:611-617,
1989.
Epstein CJ, Sahud MA, Piel CF, et al.: Hereditary macrothrombo cytopathia, nephritis and deafness. Am J Med 52:299-310,1972.
Gleeson MJ: Alport's syndrome: Audiological manifestations and
implications. J Laryngol Otol 98:449-465, 1984.
Govan JA: Ocular manifestations of Alport's syndrome: A hereditary disorder of basement membranes? BrJ Ophthalmol 67:493503, 1983.
Gregory MC, Atkin CL. Alport's syndrome, Fabry's disease, and
nail-patella syndrome. In: Schrier RW, Gottschalk CW (eds)
Diseases of the Kidney, 6th ed. Little, Brown, Boston, pp. 561 590, 1997.
Gregory MC, Terreros DA, Barker DF, et al.: Alport syndrome
Clinical phenotypes, incidence, and pathology. In: Tryggvason

K (ed) Molecular Pathology and Genetics of Alport Syndrome.

Karger, Basel, pp. 1-28, 1996.
Hinglais, N, Grunfeld JP, Bois E: Characteristic ultrastructural
lesion of the glomerular basement membrane in progressive
hereditary nephritis (Alport's syndrome). Lab Invest 27:473487, 1972.
Hostikka SL, Eddy RL, Byers MG, Hoyhtya M, Shows TB, Tryggvason K: Identification of a distinct type IV collagen alpha chain
with restricted kidney distribution and assignment of its gene
to the locus of X chromosome-linked Alport syndrome. Proc
Natl Acad Sci USA 87:1606-1610, 1990.
Kashtan CE, Kleppel MM, Gubler M- C. Immunohistologic findings in Alport syndrome. In: Tryggvason K (ed) Molecular Pathology and Genetics of Alport Syndrome. Karger, Basel, pp.
142-153, 1996.
Lemmink HH, Nilleson WN, Mochizuki T, et al.: Benign familial
hematuria due to mutation of the type IV collagen ;ga4 gene.
/ Clin Invest 98:1114-1118, 1996.
Tiebosch AT, Frederik PM, van Breda Vriesman PJ, et al.: Thin basement-membrane nephropathy in adults with persistent hematuria. N Engl J Med 320:14-18, 1989.

MEDULLARY CYSTIC DISEASE

ELLIS D. AVNER

Medullary cystic disease is a renal cystic disorder characterized by tubular cystic lesions in corticomedullary regions
of the kidney. It is an uncommon, genetically determined
tubulointerstitial nephritis that presents in children and
young adults and inexorably progresses to end stage renal disease.
DEFINITIONS AND EPIDEMIOLOGY

Medullary cystic disease is a distinct clinicopathological

entity that is part of a group of congenital tubulointerstitial
nephropathies known as the juvenile nephronophthisis
medullary cystic disease (JN-MCD) complex (Table 1).
Although genetically heterogeneous, all of the diseases of
the JN-MCD complex share common morphologic and
functional renal alterations, which lead to similar clinical
features and clinical course. The gene responsible for autosomal recessive juvenile nephronophthisis has been localized by linkage analysis to chromosome 2p, but genes responsible for the other variants of the complex have not yet

Primer on Kidney Diseases, Second Edition

been localized or identified. The nosology of the diseases

comprising the JN-MCD complex has been quite confusing,
and they have been presented under a variety of terms
including uremic medullary cystic disease, cystic disease
of the renal medulla, familial juvenile nephronophthisis,
autosomal-recessive nephronophthisis, autosomal domi nant nephronophthisis, Fanconi's nephronophthisis, saltlosing nephritis, and uremic sponge kidney. The overall
incidence and prevalence of JN-MCD is unknown. The
diseases appear to be uncommon, although more than 300
cases have been reported in the literature and JN has been
reported to cause 10 to 30% of pediatric end stage renal
disease in European centers. Pooled data demonstrate a
much lower prevalence (less than 5%) in the North American pediatric end stage renal disease population.
PATHOLOGY AND PATHOGENESIS

In general, no significant differences in renal pathology

have been demonstrated for the individual conditions that
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

324

Ellis D. Avner
TABLE I

The Juvenile Nephronophthisis-Medullary Cystic Disease (JN- MCD) Complex"

Genetics

Percentage of
JN-MCD cases

AR

50

Mean age at
onset year
10

Renal-retinal dysplasia

AR

15-20

10

Medullary cystic disease

AD

Nonfamilial sporadic

15-20
15

28
17

Variant
Juvenile nephronophthisis

Associated features
Hepatic fibrosis, cerebellar ataxia, neurocutaneous
dysplasia, skeletal abnormalities
Hyperuricemia, gouty arthritis

Note: AR, autosomal recessive; AD, autosomal dominant.

" Modified from Bernstein J, Gardner KD: In: Contran RS, Brenner BM (eds) Tubulointerstitial Nephropathies. Churchill-Livingstone,
New York, pp. 335-358,1983; and Hildebrand F, Jugners P, Grunfeld J-P: In: Schrier RW, Gottschalk CW (eds) Diseases of the Kidney,
6th ed. Little, Brown, Boston, pp. 499-520, 1996.

comprise the JN-MCD complex. Early in the course of

the disease, there are few renal structural changes which
emphasize the functional nature of the renal tubular defects
(see below). The characteristic feature of more advanced
cases is severe tubular atrophy with interstitial fibrosis and
inflammatory cell infiltration, tubular basement membrane
thickening, periglomerular fibrosis, and patchy glomerular
obsolescence. Ultrastructural analysis demonstrates thickening and loss of definition of tubular basement membranes. These changes progress to produce symmetrically
small kidneys with severe, diffuse cortical atrophy.
A major feature of diseases of the JN-MCD complex is
renal medullary cysts. Cysts measuring 1 to 2 mm in diameter are generally concentrated along the corticomedullary
junctions, and have been localized by microdissection to
the distal convoluted and medullary collecting tubules. Although the presence of characteristic cysts has been considered a diagnostic criterion, many cases have been identified
in which cysts are absent or detected only late in the course
of disease progression. Such cases are otherwise morphologically, genetically, and clinically identical to cases in
which medullary cysts are a prominent feature, suggesting
that medullary cysts are neither of primary importance in
causing clinical manifestations or tubulointerstitial nephritis nor necessary for disease progression to renal failure.
The pathogenesis of tubulointerstitial nephritis and medullary cyst formation in JN-MCD remains unknown. The
Ultrastructural findings of irregularly thickened and lamellated tubular basement membranes, as well as the decreased or absent expression of certain tubular basement
membrane antigens, have suggested a primary biochemical
or Ultrastructural abnormality of the tubular basement
membrane. Studies in a murine model of JN-MCD suggest
an immunopathogenic basis in which a defect in regulatory
T-cell function leads to functional inactivation of suppres sor T cells, loss of tolerance to tubular antigens, and facilitated expression of effector T cells that mediate tubulointerstitial injury. Further study into the genetics and
pathophysiology of these diseases is required to validate
or disprove these attractive hypotheses.
CLINICAL FEATURES

All variants of JN-MCD complex are characterized by

the insidious onset of renal failure. The most common

symptoms at presentation are polydipsia, polyuria, and enuresis. Additional symptoms include weakness and pallor.
Short stature and failure to thrive may be prominent pres enting symptoms in children.
Unless at-risk individuals with parents or siblings affected with JN-MCD are monitored for the early symptom
of polyuria, most patients present with a reduced glomerular filtration rate (GFR). Anemia is present in most patients
at the time of diagnosis and has been considered by some,
but not all, investigators to be disproportionately severe
relative to the degree of renal dysfunction. Impairment of
urinary concentrating ability is the earliest pathophysiological feature of JN-MCD. It has been documented prior to
any decrease in GFR and may be present with minimal
histological abnormalities. In some families with autosomal
recessive variants of the JN-MCD complex, obligatory heterozygotes have demonstrated similar decreases in urinary
concentrating ability. Unfortunately, this is not a consistent
finding and cannot be utilized to identify heterozygotes for
the autosomal recessive conditions. Renal salt wasting is
another characteristic feature of JN-MCD and has been
reported in 20 to 60% of affected patients. Salt wasting
presumably reflects the severity and distribution of histological changes, as it does in other tubulointerstitial nephropathies, and may occur with or without the presence
of medullary cystic lesions. An additional characteristic
clinical feature of JN-MCD is the paucity of urinary abnormalities. Proteinuria and hematuria are exceedingly rare
and reported only in some individuals with autosomal dominant variants of the complex. Pyuria and documented urinary tract infection are also uncommon, as is hypertension
prior to end-stage renal disease. The absence of these clinical features may help differentiate JN-MCD from the other
genetically determined polycystic kidney diseases.
As noted in Table 1 one-fourth to one-third of all patients
with autosomal-recessive JN-MCD have associated retinal
changes and are clinically classified as the renal-retinal
dysplasia variant. Retinal findings are characterized by pro gressive tapetoretinal degeneration, and visual impairment
may be present in early infancy or childhood. Fundoscopic
alterations are present in all patients with renal-retinal
dysplasia by the age of 10 years, and the diagnosis can be
confirmed with electroretinography. Some reported pa-

47. Medullary Cystic Disease

tients with this clinical variant also have hepatic fibrosis,

cerebellar ataxia, neurocutaneous dysplasia, and various
skeletal abnormalities. The sporadic nature of these associations, as well as the irregular clustering in some families,
makes these associa tions difficult to further classify. Retinal
findings as well as the noted multiorgan associations have
not been reported in the autosomal dominant or nonfamilial sporadic variants of the JN-MCD complex.
The diagnosis of JN-MCD should be suspected in the
child or young adult who presents with progressive renal
failure and normal to small kidneys in association with the
clinical and laboratory features noted above. The differential diagnosis may include renal hypoplasia-dysplasia, congenital obstructive uropathies, autosomal dominant or recessive polycystic kidney disease, bilateral impairment of
the renal circulation, or other forms of chronic tubulointers titial nephritis. A complete history, with particular attention to family history, physical examination, examination
of the urinary sediment, and radiographic imaging (ultrasonography, with computed tomography and arteriography
as indicated) will be adequate to establish the diagnosis of
JN-MCD in most cases. In rare instances, a renal biopsy
may be indicated to definitively establish a histopathological diagnosis. However, pathognomonic medullary cysts
may not be apparent on needle biopsy specimens. The gene
responsible for the autosomal recessive variant of JN-MCD
has been linked to human chromosome 2p. Thus, in "genetically informative" families, diagnosis by genetic linkage
analysis (pre- or postnatal) may be possible.
PROGNOSIS AND THERAPY

JN-MCD is characterized by variable, but inexorable

progression to end-stage renal disease. No specific factors
have been identified which modify the rapidity of progres sion or clinical course. Genetic counseling is a cornerstone
of therapy for families with genetically defined variants of
the complex. If affected individuals are identified prior

325
to the development of end stage renal disease, adequate
hydration and sodium supplementation are critical to avoid
dehydration and superimposed prerenal insults. This is particularly true in infants and young children who are subject
to episodic febrile illnesses and intercurrent episodes of
gastroenteritis. Children with severe polyuria secondary to
renal-concentrating defects and salt wasting may require
nasogastric or gastrostomy tube feedings to maintain hydration and provide adequate nutrition. Anemia is treated
with recombinant human erythropoietin and other manifestations of progressive uremia are treated in standard
fashion. Ultimately, dialysis and renal transplantation are
utilized to treat end-stage renal failure. JN-MCD has not,
to date, been reported to recur in transplanted kidneys.
Bibliography
Antignac C, Kleinknecht C Habib R: Toward the identification
of a gene for familial juvenile nephronophthisis (autosomal recessive medullary cystic kidney disease). Adv Nephrol Necker
Hasp 24:379-33, 1995.
Avasthi PS, Erickson DG, Gardner KD: Hereditary renal-retinal
dysplasia and the medullary cystic disease-nephronophthisis
complex. Ann Intern Med 84:157-161, 1976.
Bernstein J, Gardner KD: Hereditary tubulo interstitial nephritis.
In: Contran RS, Brenner BM, Stein JH (eds) Tubulointerstitlal
Nephropathies. Churchill-Livingstone, New York, pp.335-358,
1983.
Cohen AH, Hoyer JR: Nephronophthisis, a primary tubular basement membrane defect. Lab Invest 55:564-574, 1986.
Hildebrandt F, lungers P, Grunfeld J-P: Medullary cystic and
medullary sponge renal disorders. In: Schrier RW, Gottschalk
CW (eds) Diseases of the Kidney, 6th ed. Little Brown, Boston,
pp. 499-520, 1996.
Kelly CJ, Neilson EG: Medullary cystic disease: An inherited form
of autoimmune interstitial nephritis? Am J Kidney Dis 10: 389395, 1987.
Kleinknecht C, Habib R: Nephronophthisis. In: Cameron S, Davison AM, Grunfeld JP, Kerr D, Ritz E (eds) Oxford Textbook of
Clinical Nephrology, Oxford Press, Oxford, pp. 2188-2197,1992.
Mongeau JG, Worthen HG: Nephronophthisis and medullary cystic disease. Am J Med 43:345-355, 1967.

SECTION 8
TUBULOINTERSTITIAL
NEPHROPATHIES AND DISORDERS OF
THE URINARY TRACT

TUBULOINTERSTITIAL DISEASE
WILLIAM F. FINN

DEFINITION AND CLASSIFICATION

of titratable acid and urinary ammonia excretion, leading

to metabolic acidosis. This may be present in conjunction
with a defect in the secretion of potassium and result in
type IV renal tubular acidosis.

Chronic structural abnormalities involving the renal tubules and interstitum develop in two circumstances. In the
first, tubulointerstitial nephritis is found as the primary
lesion after prolonged exposure to various therapeutic or
environmental agents or in association with a number of
systemic illnesses (Table 1). Such primary chronic interstitial nephritis accounts for up to 15 to 30% of cases of end
stage renal disease. In the second, tubulointerstitial changes
occur as a result of progressive glomerular and vascular
injury. This has come to be recognized as a significant
determinant of the final outcome of these conditions.

DIAGNOSIS

The clinical diagnosis of tubulointerstitial nephritis is

often a diagnosis of exclusion, made when glomerular dis ease is absent. The histologic diagnosis of tubulointerstitial
nephritis requires evidence of an inflammatory infiltrate
accompanied by interstitial fibrosis. In the absence of other
definable disease, presumptive evidence includes a history
of significant exposure to an offending agent or the coexistence of a condition known to be associated with tubulointerstitial disease, the demonstration of abnormalities of
tubular function out of proportion to the reduction in glomerular function, and a urinalysis marked by the absence
of heavy proteinuria with a tendency for leukocyturia
rather than hematuria.

HISTOLOGY

The characteristic lesion is composed of an interstitial

inflammatory mononuclear cell infiltrate containing lymphocytes mostly T cells and occasional plasma cells.
This acute cellular infiltrate is generally accompanied by
interstitial edema, disruption of the tubular basement membrane, and dissolution of normal interstitial architecture.
The transition to a chronic process occurs with the development of interstitial fibrosis, which is accompanied by tubular ectasia, tubular atrophy, and a marked increase in extracellular matrix. Eventually, glomerular and vascular
structures are involved and fibrotic and sclerotic changes
occur throughout the kidney.

Urinalysis

Evaluation of the urine in patients with chronic interstitial nephritis yields variable results, except that marked
proteinuria in uncommon; by quantitative tests, a trace-to1 + protein may be present; by qualitative analyses, usually
less than 1 g of protein per 24 hours, and often less than
500 mg is found. Microscopic examination may disclose a
preponderance of white blood cells and occasionally white
cell casts and granular casts. A general estimate of integrated tubular function may be obtained by determining
the capacity of the kidneys to concentrate or dilute the
urine in response to water deprivation or administration,
the ability to excrete an administered acid load, and the
precis ion with which sodium balance is maintained.

CLINICAL MANIFESTATIONS

The manifestations of tubulointerstitial nephritis depend

on the extent of injury (diffuse or focal), the tubular segments most severely involved (proximal or more distal),
and the degree of compensation achieved by the less severely involved nephrons. When proximal tubules are damaged, such substances as sodium, glucose, amino acids, uric
acid, and low-molecular-weight proteins, which are ordinarily reabsorbed or metabolized at this site, appear in the
urine. A decrease in bicarbonate reabsorption may lead to
proximal renal tubular acidosis. Damage to more distal
structures, including the loop of Henle, the distal convoluted tubule, and the collecting duct, is accompanied by an
inability to maximally concentrate and dilute the urine.
The defect in the ability to concentrate the urine may result
in polyuria. Damage in this area may decrease excretion

Primer on Kidney Diseases, Second Edition

Proteinuria

The normal values of the major urinary proteins are

listed in Table 2. In contrast to the high-molecular-weight
proteins, a finite amount of low-molecular-weight proteins
is normally filtered and then reabsorbed by proximal tubular cells. When the reabsorptive capacity of the proximal
tubular epithelium is disrupted, various low-molecularweight proteins appear in the urine.
329

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

330

William F. Finn
TABLE I

TABLE 2

Causes of Chronic Tubulointerstitial Disease

Normal Values for Urinary Proteins in Humans

Category Examples
Therapeutic agents/
Analgesics
occupational or
NSAIDs
environmental agents Chemotherapeutic agents
(cisplatin, nitrosoureas)
Immunotherapeutic agents
(cyclosporine, FK-506)
Heavy metals (lead, cadmium)
Lithium
Immunologic conditions Renal allograft rejection
Amyloid
Vasculitis
Cryoglobulinemia
Sjogren's syndrome
SLE
Wegener's granulomatosus
Hematopoietic/neoplastic Sickle cell disease Multiple myeloma
diseases
Light chain disease Dysproteinemias
Lymphoproliferative disease
Mechanical disorders Ureteral obstruction
Vascular diseases

Hereditary/genetic

Radiation Hypertension
Atheromatous emboli
Karyomegalic interstitial
nephritis Medullary cystic disease/
nephronophthis complex Polycystic
kidney disease Hereditary
nephritis

Miscellaneous conditions Balkan nephropathy Chinese herb

nephropathy Mycotoxins Sarcoidosis
Metabolic disorders Hypercalcemia
Hypokalemia
Uric acid nephropathy
Oxalate nephropathy
Cystinosis
Infections Systemic Local
SLE, systemic lupus erythomatosus.

High-Molecular-Weight Proteinuria
The distinction between "glomerular" proteinuria and
"tubular" proteinuria is based on the quantity and quality
of the proteins measured in the urine. The appearance in
the urine of serum proteins with a molecular weight in
excess of 40,000 to 50,000 Da is an early marker of glomerular damage. The commonly measured high-molecularweight proteins include albumin, transferrin, and IgG.
Low-Molecular-Weight Proteinuria
In contrast, tubular proteinuria is due to excretion of
low-molecular-weight proteins. The two most commonly

Urinary protein

MN (da)

Tamm-Horsfall

Normal value
14.3 (5.1-25.5)

glycoprotein"
Albumin"

69,000

5.2 (2.8-15)

of, -Microglobulin"

29-33,000

4.20 6

IgG"

146,000

1.2 (0.35-2.7)

Transferrin"

77,000

0.17 (0.08-0.83)

Cystatin C

13,300

0.113 0.125 mg/L

Protein 1"

18,700

0.092 (0.02-0.3)

/32 -Microglobulin"

11,800

0.062 (0.021-0.142)

Retinol-binding protein" 21,400

0.055 (0.03-0.13)

Lysozyme

0.015 (0.002-0.12) mg/L

" mg/g creatinine

mentioned as potential markers of renal tubular damage

are /32-microglobulin (/32-m) and retinol binding protein.
Other low-molecular-weight proteins include ai-microglobulin (a-i-m), protein 1, amylase, lysozyme, ribonuclease, and cystatin C. Due to its molecular weight and
small radius, /32-m is readily filtered at the glomerulus.
Approximately 99.9% is reabsorbed by the proximal tubular epithelial cells and ultimately catabolized. A very small
amount appears in the urine. The urinary excretion of /32m is considerably increased in cases of renal tubular impairment. Since /32-m undergoes degradation at urinary pH of
5.5 or less, accurate measurement demands alkalinization
of the urine. Retinol-binding protein is a low-molecularweight protein synthesized in the liver, where it binds to
retinol. Once the retinol is released at peripheral sites,
the binding protein is rapidly eliminated from plasma by
glomerular filtration. It is then reabsorbed and catabolized
by proximal tubular cells. Because of its stability in acid
urine, the assay of urinary retinol binding protein is preferred over that of /32-m.
Tamm-Horsfall Protein
Tamm-Horsfall glycoprotein is the most abundant protein of renal origin in normal urine and is the major constituent of urinary casts. Tamm-Horsfall glycoprotein is synthesized by cells of the thick ascending limb of the loop of
Henle, and is excreted in the urine at a relatively constant
rate. Urinary excretion can increase following injury to the
distal tubule.
Enzymuria
The interpretation of urinary enzyme liters is founded
on the premise that the sole source of high-molecularweight enzymes is damaged tubular cells. However, it is
hard to correlate specific disease states with the presence
or absence of enzymuria. In addition, a relationship between the severity of cellular injury and the magnitude of
enzymuria has been difficult to establish. This has been

48. Tubulointerstitial Disease

331

due in part to the observation that urinary enzyme activity

is affected by various factors that are independent of cellular integrity, i.e., urinary pH, osmolarity, and the presence
of various enzyme inhibitors or activators. However, in
addition to normal cell shedding, enzymes also gain urinary
access because of altered cell membrane permeability, increased rate of enzyme synthesis, and frank cell necrosis.
Only a limited number of enzymes have been generally
accepted as valuable urinary biomarkers. These include Nacetyl-/3-D-glucosaminidase, alanine aminopeptidase, and
intestinal alkaline phosphatase. It is important to emphasize that the measurement of these enzymes, while of considerable theoretical interest, is of limited clinical utility
(Table 3).
N-Acetyl-/3-glucosaminidase is found in both the straight
(S3) segment of proximal tubular cells and the distal nephron as a lysosomal enzyme. In humans, it has its highest
activity in the S3 segment, with less activity in the collecting
duct. Elevated urinary levels accompany proximal tubular
cell injury. It may also been found in the urine of patients
with various forms of glomerular disease, obstructive uropathy, and nephrosclerosis. Other nonspecific increases in
urinary activity that limit its usefulness as a diagnostic tool
have been described.
Alanine aminopeptidase is restricted to the proximal tubule. Increased excretion has been reported in a variety of
renal diseases including pyelonephritis, glomerulonephritis, urologic cancers and renal transplant rejection. In addition, increased excretion has been reported in association
with many well-defined nephrotoxins. Although not specific in discriminating between glomerular and tubular dis ease, it is very sensitive to acute tubular injury.
Intestinal alkaline phosphatase and nonspecific tissue alkaline phosphatase are two urinary isoenzymes that have
elicited interest as potential segment specific markers of

TABLE 3

Some Enzymes Used as an Index of Nephrotoxicity

Enzymes Cellular location
Alanine aminopeptidase
Alkaline phosphatase
y-Glutamyltransferase
Maltase
Trehalase
Glutamic oxaloacetic transaminase

Cytosol

Glutamic pyruvic transaminase

the human nephron. Intestinal alkaline phosphatase is expressed on the brush border of the tubuloepithelial cells
of the S3 segment of the proximal tubule. The intestinal
alkaline phosphatase released in urine has its origin in the
kidney and is considered to be a specific and sensitive
marker for alterations of the S3 segment. In the kidney,
tissue alkaline phosphatase is localized on the brush border
in all segments of the proximal tubule. By measuring both
enzymes, judgments as to the involvement of S1-S2 vs. S3
segments can be achieved in experimental models.
Tubular Antigens
The urinary excretion of specific proximal tubular antigens is increased in a variety of renal diseases. Monoclonal
antibodies to membrane and other cellular derived antigens
are new, sensitive, specific, and readily available markers
of renal cell injury. Increased excretion of tubular antigens
occurs in exposure to cadmium, hydrocarbons, cisplatin,
and radiographic contrast media. Monoclonal antibodies
to human brush-border antigens have been produced and
used in investigational studies. Their use may permit detection of site-specific renal damage.
MECHANISM OF FIBROGENESIS
Infiltrating cells stimulate fibrogenesis through the release of various cytokines and growth factors. In response,
resident cells in the kidney undergo phenotypic changes
and acquire smooth muscle and fibroblastic characteristics.
Active in this process may be transforming growth factorfi (TGF-/3), one of the more potent stimulators of collagen
and noncollagen basement membrane components. Examples of the latter include proteoglycans and fibronectin.
TGF-/3 also inhibits matrix degrading enzymes such as collagenase and metalloproteinases. The net result is that for
a time the excessive deposition of extracellular matrix can
be removed, but eventually complete resolution becomes
unlikely. TGF-/3 also controls the interaction of cells with
the extracellular matrix by regulating the expression of
various cell adhesion molecules such as integrins. Plateletderived
Brush
bordergrowth factor also stimulates chemotaxis, influences the production of extracellular matrix, and regulates
its subsequent metabolism. Added to this are the interleukins which modulate inflammatory and immune responses
by regulating the growth, differentiation, and mobility of
effector cells. In as yet unknown manner, angiotensin II
may be linked to the irreversible changes of tubulointerstitial fibrosis.

Lactate dehydrogenase
Malate dehydrogenase
ALAcetyl-/3-D-glucosaminidase

Lysosome

Acid phosphatase
/3-Glactosidase
/3-Glucosidase
/3-Glucuronidase
Glutamate dehydrogenase

Mitochondria

PRIMARY VASCULAR AND GLOMERULAR

RENAL DISEASE
In many forms of renal disease due to primary glomerular
injury, there is a rema rkable inverse correlation between
the extent of tubulointerstitial damage and the glomerular
filtration rate. This observation has directed attention to
the important role of the tubulointerstitial compartment
in the progression of renal disease.

Gregory L Braden

332
The most obvious explanation to account for the tubulointerstitial injury that occurs with primary glomerular
disease is that inflammatory glomerular lesions are accompanied by acute interstitial mononuclear cell infiltrates
which later lead to chronic fibrotic changes by mechanisms
described above. Other mechanisms have been suggested.
For instance, it is clear that persistent high-grade proteinuria is associated with a progressive loss of renal function
and the development of tubulointerstitial disease and renal
fibrosis. With the loss of permselectivity that occurs with
glomerular injury, various circulating plasma proteins
usually prevented from passing this barriermay find their
way into the glomerular filtrate and tubular fluid where
they exert toxic effects on epithelial cells. Some of these
proteins may precipitate within the tubular lumen, thereby
forming casts and obstructing tubular fluid flow. Preglomerular vasoconstriction follows, with a decrease in peritubular
capillary blood flow, eventual tubular atrophy and further
interstitial inflammation. As tubular reabsorptive mechanisms are stimulated, lysosomal degradative enzymes may
spill into the cytosol and cause damage. Alternatively, filtered low-molecular-weight cytokines may bind to tubular
epithelial cells and lead to activation of the alternate complement pathway.
Without question, immunologic processes involving either the cell-dependent or the humoral pathway may be

responsible for the ongoing damage. A release of chemo tactic lipids may further the inflammatory response. In addition, changes in tubular oxidative metabolism leading to
an increase in renal ammoniagenesis and the generation
of reactive oxygen species may contribute to the progres sive lesions. Ammonia may interact with the third component of complement with activation of the alternate complement cascade and subsequent immunologic cellular
injury. Cytokine-induced generation of nitric oxide and
the transferrin-related release of iron with accumulation
in tubular cell lysosomes may be associated with oxidative
cellular injury. Finally, arachidonic acid metabolites, particularly thromboxane A 2, may stimulate extracellular matrix
protein formation and in this way contribute to the progres sion of renal disease.
Bibliography
Eddy AA: Experimental insights into the tubulointerstitial disease
accompanying primary glomerular lesions. / Am Soc Nephrol
5:1273-1287, 1994.
Dodd S: The pathogenesis of tubulointerstitial disease and mechanisms of fibrosis. Curr Top Pathol 88:117-143, 1995.
Jones CL, Eddy AA: Tubulointerstitial nephritis. Pediatr Nephrol
6:572-586, 1992.
Nath KA: Tubulointerstitial changes as a major determinant in
the progression of renal damage. Am J Kidney Dis 20:1-17,1992.
Sedor JR: Cytokines and growth factors in renal injury. Semin
Nephrol 12:428-440, 1992.

LITHIUM-INDUCED RENAL DISEASE

GREGORY L. BRADEN

Lithium is the best treatment for manic-depressive illness, but therapy with this drug has been associated with
side effects in many body systems, including the kidneys.
Lithium is freely filtered at the glomerulus, reabsorbed
like sodium at several tubular sites, and concentrated
in the renal medullacircumstances that might favor
lithium-induced renal disease. Although lithium initially
was thought to produce only functional renal disorders,
such as nephrogenic diabetes insipidus (NDI), additional
disorders now attributed to lithium include renal tubular acidosis, chronic interstitial nephritis, and nephrotic
syndrome.

Primer on Kidney Diseases, Second Edition

NEPHROGENIC DIABETES INSIPIDUS

Polydipsia occurs in up to 40% and polyuria greater than

3 L/day occurs in up to 20% of patients treated with lithium.
Despite the high prevalence of polyuria, urine volume is
rarely increased enough to require cessation of lithium
therapy. Most humans with lithium-induced polyuria are
unresponsive to the administration of exogenous antidiuretic hormone (ADH). This could occur either due to abnormalities in the medullary osmotic gradient that drives
ADH-mediated water reabsorption or due to direct inhibition of the tubular hydroosmotic effects of ADH. Chronic
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

49. Lithium-induced Renal Disease

administration of lithium diminishes the medullary and
papillary osmolar gradients due to depletion of urea without affecting sodium chloride concentrations. However,
lithium-induced polyuria is mainly due to direct inhibition
of the ADH-dependent aspects of water conservation. Only
one lithium-treated patient has been reported to have a
defect in the pituitary release of ADH indicative of central
diabetes insipidus.
Lithium impairs the hydro-osmotic response to ADH by
several mechanisms. First, there is abundant physiological
and biochemical evidence that lithium directly inhibits the
adenylate cyclase system in the mammalian distal nephron,
leading to decreased generation of the second messenger,
cyclic adenosine 3',5'-monophosphate (cAMP), the putative stimulus for transepithelial water movement. In addition, lithium has been shown to inhibit transepithelial water
movement after the administration of exogenous cAMP,
suggesting inhibition of water flow at a site distal to the
generation of cAMP.
Although lithium-induced NDI usually improves after
lithium withdrawal, some patients have persistent concentrating defects lasting for years. Amiloride has been shown
to reduce urinary volume and enhance the concentrating
ability of patients with lithium-induced NDI. Amiloride
blocks distal tubular reabsorption of lithium, thus lowering
the renal tissue lithium level. In addition, amiloride is a
diuretic and the combination of dietary salt restriction and
the natriuretic effect of the diuretic induces mild extracellular fluid volume depletion which decreases the glomerular
filtration rate and lessens urinary volume. Thiazide diuretics have also been utilized for this purpose, but they may
raise serum lithium levels. They are more useful in patients
with symptoms that persist after lithium discontinuation.
Interestingly, amiloride has been shown to be effective
without inducing any changes in creatinine clearance or
serum lithium levels, suggesting that the inhibition of lithium uptake in the distal nephron may be the most important
factor for its efficacy. Indomethacin has been effective in
a few patients, presumably by inducing a significant fall in
the glomerular filtration rate. As with diuretics, this may
lower urinary volume. In addition, indomethacin inhibits
synthesis of urinary prostaglandins that inhibit the tubular
action of ADH. As with thiazides, use of nonsteroidal antiinflammatory drugs (NSAIDs) in patients still on lithium
requires great caution and frequent monitoring of lithium
levels. Finally, a few patients with severe NDI will require
cessation of lithium and the substitution of either carbamazepine or valproic acid to treat the manic-depressive illness.
RENAL TUBULAR ACIDOSIS

Lithium impairs distal hydrogen ion secretion in at least

50% of treated patients. During administration of intravenous sodium bicarbonate, distal nephron bicarbonate delivery is increased. In normal humans, this stimulates hydro gen ion excretion by the distal nephron proton pumps.
The additional protons are buffered by bicarbonate; the
carbonic acid formed then dissociates and raises the urinary
pCO2 from 40 to 80 mm Hg or greater. This distal nephron

333
response is impaired in lithium-treated patients; however,
there is no evidence for renal bicarbonate wasting (such
as in proximal renal tubular acidosis), and the excretion of
the urinary buffer, ammonium, is normal. Despite this defect in urinary acidification, the serum bicarbonate and pH
remain normal. Taken together, these studies indicate that
lithium induces incomplete distal renal tubular acidosis.
Thus, patients treated with lithium are prone to systemic
acidosis during stressful conditions such as sepsis or catabolic states.
ACUTE RENAL FAILURE/LITHIUM INTOXICATION

Acute renal failure due to biopsy-proven acute tubular

necrosis may occur in lithium-intoxicated patients. Nephrotic syndrome due to minimal change disease may occur
concomitantly with acute tubular necrosis. Whether lithium
can directly cause tubular necrosis or whether tubular necrosis in these patients is secondary to hemodynamic factors
is unknown.
Lithium intoxication is classified into three grades of
severity based on the blood level (mild, <2.5 mEq/L; moderate, 2.5 to 3.5 mEq/L; severe, >3.5 mEq/L). Toxic effects
include nausea, ataxia, tremors, twitching, muscle rigidity,
disordered consciousness, seizures, and coma. The pathophysiology of lithium intoxication may be due to its ability
to induce dehydration and salt depletion secondary to its
inhibitory effect on ADH action and the acute effects of
lithium to induce a natriuresis, leading to extracellular fluid
volume depletion, activation of the serum-angiotensin system, a decrease in glomerular filtration rate, and, in turn,
increased serum lithium levels. In addition, angiotensinconverting enzyme inhibitors and NSAIDs may cause renal
dysfunction leading to lithium intoxication, especially in
patients on high therapeutic doses of lithium. Close monitoring of serum lithium levels is warranted after initiation
of therapy with these agents.
Patients with lithium intoxication should be admitted to
the hospital, since seizures can occur at any time. The drug
should be discontinued and, if there is an acute ingestion,
either gastric lavage or ipecac should be given. Restoration
of depleted extracellular fluid volume should occur with
the administration of intravenous 0.9% normal saline if
there is no hypernatremia present. For those patients with
mild intoxication with a serum level of 2.5 mEq/L or less,
saline diuresis may be enough to enhance renal lithium
clearance. If there is no history to preclude vigorous volume
expansion, then up to 6 L of saline can be given daily until
the lithium level decreases to the nontoxic range. With this
therapy, the lithium level should fall appro ximately 1 mEq/
day. However, patients who are euvolemic usually do not
respond to saline and/or forced diuresis. For those patients
with moderate or severe toxicity, hemodialysis is the therapy of choice and may need to be repeated if serum lithium
levels rebound after the first hemodialysis treatment. Continuous renal replacement therapy may be employed at
centers where hemodialysis is unavailable. This modality
effectively removes lithium and is not complicated by post
therapy rebound.

334

Gregory L Braden

NEPHROTIC SYNDROME

Lithium causes nephrotic syndrome in a small number

of patients. Renal biopsies have demonstrated minimal
change disease in the majority of patients and focal segmental glomerulosclerosis in a smaller number. Several reports
demonstrate that patients may undergo a complete remis sion of the nephrotic syndrome upon withdrawal of lithium.
Minimal change disease may recur after readministration
of lithium.
CHRONIC INTERSTITIAL NEPHRITIS

The association of lithium with chronic interstitial nephritis was first reported in 1977, when a renal biopsy study
described an increase in interstitial fibrosis, focal nephron
atrophy, or both, in lithium-treated patients compared to
age-matched controls. In this study, 80% of the lithiumtreated patients had decreased creatinine clearances. Subsequently, a number of retrospective and uncontrolled
studies found the prevalence of renal insufficiency to range
from 3 to 20% in patients treated long term with lithium,
but other disorders causing renal insufficiency were not
always excluded. In contrast, additional studies compared
lithium-treated patients to a more suitable control group,
i.e., psychiatric patients not receiving lithium, and found
no differences in serum creatinine or creatinine clearance.
Renal biopsies were not performed.
Prospective studies demonstrate that chronic lithium
therapy in psychiatric patients can significantly impair the
glomerular filtration rate compared to similarly matched
psychiatric patients who never received lithium. Renal biopsies in lithium-treated patients have demonstrated increased interstitial fibrosis and a unique tubular lesion consisting of microcyst formation due to cystic dilatation of
the tubules.
Taken together, these studies indicate that a small number of patients treated with lithium develop progressive
renal damage associated with chronic interstitial nephritis.
However, no patient has yet been reported to develop end
stage renal failure from lithium use. Baseline renal function

studies including serum creatinine, BUN, and creatinine

clearance should be performed prior to the initiation of
lithium therapy and thereafter measured yearly. Patients
who demonstrate deterioration in renal function should
have lithium therapy withdrawn and either carbamazepine
or valproic acid initiated. For those patients who can only
be managed on lithium, lithium therapy can probably be
continued with careful monitoring of renal function over
time while maintaining the serum lithium level within the
therapeutic range.
Bibliography
Allen HM, Jackson RL, Winchester MD, Deck LV, Allon M:
Indomethacin in the treatment of lithium-induced nephrogenic
diabetes insipidus. Arch Intern Med 149:1123-1126, 1989.
Batlle D, Gaviria M, Grupp M, Arruda JAL, Wynn J, Kurtzman
NA: Distal nephron function in patients receiving chronic lithium
therapy. Kidney Int 21:477-485, 1982. Batlle DC, von Riotte
AB, Gaviria M, Grupp M: Amelioration
of polyuria by amiloride in patients receiving long-term lithium
therapy. N Engl J Med 312:408-414, 1985. Boton R, Gaviria
M, Batlle DC: Prevalence, pathogenesis, and
treatment of renal dysfunction associated with chronic lithium
therapy. Am J Kidney Dis 10:329-345, 1987. Leblanc M,
Raymond M, Bonnardeaux A, et al.: Lithium poisoning
treated by high-performance continuous arteriovenous and venovenous hemodiafiltration. Am J Kidney Dis 27:365-372, 1996.
Lehmann K, Ritz E: Angiotensin-converting enzyme inhibitors
may cause renal dysfunction in patients on long-term lithium
treatment. Am J Kidney Dis 25:82-87, 1995. Okusa MD,
Crystal LJT. Clinical manifestations and management
of acute lithium intoxication. Am J Med 97:383-389, 1994.
Tarn VKK, Green J, Schwieger J, Cohen AH: Nephrotic syndrome
and renal insufficiency with lithium therapy. Am J Kidney Dis
27:715-720, 1996. Hestbech J, Hansen HE, Amdisen A,
Olsen S: Chronic renal
lesions following long-term treatment with lithium. Kidney Int
12:205-213, 1977. Jorkasky DK, Amsterdam JD, Oler J, et al.:
Lithium-induced renal
disease: a prospective study. Clin Nephrol 30:293-302, 1988.
Walker RG, Bennett WM, Davies BM, Kincaid-Smith P: Structural
and functional effects of long-term lithium therapy. Kidney Int
21:513-519, 1982.

LEAD NEPHROTOXICITY
WILLIAM F. FINN

HISTORY

rocyte protoporphyrin becomes elevated when blood lead

concentrations reach about 20 /ug/mL (1.0 /xmol/L), approximately twice the average blood level in the United States.

The relationship between lead exposure and chronic interstitial nephritis was first recognized in 1862. Since then
it has become widely appreciated that renal damage may
occur in persons who have had a heavy exposure to lead
over a number of years. Whether exposure to lead in the
general population leads to impaired renal function is not
known. Also of uncertain nature is the relationship between
childhood plumbism and the later development of chronic
lead nephropathy.

PATHOPHYSIOLOGY
Markers of Effect

Low lead exposure has little definable effect on the creatinine clearance (Ccr ), whereas moderate exposure may be
accompanied by a slight state of hyperfiltration. However,
a negative correlation between blood lead and the Ccr has
been found, such that a 10-fold rise in blood lead concentration was associated with as much as a 10 to 13 mL/min
reduction in the Ccr . Thus, lead exposure may impair renal
function in the population at large, although it is possible
that renal impairment itself may lead to an increase in the
blood lead concentration.
Proteinuria of either a low- or high-molecular-weight is
an inconsistent finding. Albuminuria tends to occur only
with advanced disease. With elevated blood lead levels, an
increase in urinary retinol binding protein or /32-microglobulin may be found. Likewise, elevated blood lead levels
may be accompanied by an increase in urinary N-acetyl-/3D-glucosaminidase activity. Lower urinary excretion of 6keto-PGFic, and an enhanced excretion of thromboxane
(TxB2) have been found, but no deleterious effect on renal
function has been identified.

RISK FACTORS
Populations at Risk

Occupational exposure to lead occurs in smelters, miners, painters, and plumbers, and in workers involved in
the manufacture of storage batteries, pottery, and pewter.
Environmental exposure occurs through the use of inadequately glazed pottery, the consumption of "moonshine"
whiskey, and the ingestion of lead-containing food or water.
Lead pipes and soldered joints are important sources of
lead in drinking water, while soil contaminated with lead
from industrial activities is an important source of lead in
foodstuffs. It should be noted that lead is most harmful to
children under 6 years of age in that they absorb about
50% of the ingested lead. In comparison, adults absorb
only 5 to 10%. For children, flaking lead-based paint and
dust in old homes are important sources of lead. The removal of lead from gasoline and paint has significantly
reduced the lead exposure in children and adults alike.

Mechanism of Toxicity

In addition to the age at which exposure occurs, the

amount of lead absorbed and the severity of the disease
may be influenced by the amount of calcium in the diet,
the presence of iron deficiency, and exposure to sunlight
and vitamin D. Iron deficiency enhances lead-induced hepatic metallothionein synthesis and in this way may promote the accumulation of lead in the kidney.

The nephrotoxicity is due in part to the fact that urinary

elimination is a major route of lead excretion from the
body. Nuclear inclusion bodies in proximal tubule epithelial
cells that have a high sulfhydryl group content probably
include a lead-binding protein that sequesters intracellular
lead. A major toxic effect involves the inhibition of cellular
respiration and energy production as a result of the accumulation in mitochondria and the ensuing defects in mitochondrial structure and function. Interaction with membranes and enzymes disrupts calcium metabolism, glucose
homeostasis, and ion transport processes.

Markers of Exposure

CLINICAL MANIFESTATIONS

The most useful screening tests for excessive lead exposure are blood lead and erythrocyte protoporphyrin. Eryth-

Three forms of lead intoxication occur: (1) acute in organic lead poisoning from the inhalation of lead oxide,

Markers of Susceptibility

Primer on Kidney Diseases, Second Edition

335

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

William F. Finn

336
(2) lead poisoning caused by inhalation or absorption
through the skin of tetraethyl lead, and (3) chronic inorganic lead poisoning from inorganic lead in the form of
lead oxide, lead carbonate, or similar compounds. Initially,
the highest concentrations can be found in the kidney. The
lead concentration is 50 times greater in the cortex than
in the medulla. From the kidney, lead is excreted in the
urine or redistributed to other tissues, most notably bone.
The rate of urinary excretion of lead depends on the duration of exposure as well as on the absolute body burden.
At least two types of renal impairment may be found in
association with lead poisoning. In the first and more acute
form, generalized defects in proximal tubular function result in the Fanconi syndrome with amino-aciduria, glycosuria, and phosphaturia. Serum uric acid levels are generally
elevated because of a specific defect in its tubular secretion.
Occasionally, the urine contains cells with eosinophilic intranuclear inclusions composed of lead and protein. These
abnormalities most often occur in children following several months of heavy lead ingestion. Although this impairment is generally rapidly reversible, it is likely that some
cases go on to develop chronic lead nephropathy.
Chronic lead nephropathy, the second type of renal impairment associated with lead poisoning, is an indolent
disease difficult to separate from other forms of chronic,
slowly progressive renal insufficiency. In the absence of a
history of acute lead nephropathy, the diagnosis is suspected when the course of the renal disease is protracted,
without definable cause, and associated with symmetrical
contraction of both kidneys. Urine protein excretion tends
to be less than 1 g/24 h and is marked by the presence of
low-molecular-weight proteins such as /32-rnicroglobulin or
retinol-binding protein. Evidence of excessive lead absorp tion is supplied by determining urinary lead excretion after
administration of the calcium disodium salt of ethylene
diamine tetraacetic acid (EDTA), a lead chelator, as described below. Chronic lead nephropathy is generally considered to be irreversible.
Hypertension

A weak positive association exists between blood pres sure and lead exposure. Some have found an increase in
mobilizable lead in hypertensive patients with normal renal
function. At least one large epidemiological study has
shown that blood lead levels correlate directly with diastolic
blood pressure. Chronic lead exposure suppresses plasma
renin concentration, blunts the renin response to sodium
deprivation, and may be the cause of hyporeninemic hypoaldosteronism in hypertensive patients.
Saturnine Gout

Chronic lead toxicity is associated with saturnine gout.

Since gout is unusual in other forms of chronic renal failure,
chronic lead nephropathy should be considered a patient
with chronic renal failure and symptomatic gout. This is
particularly true in women, as there is an overwhelming
male predominance of idiopathic gout. Indeed, "gouty

nephopathy" is more likely to be a manifestation of lead

nephropathy rather than urate nephropathy.
Renal Cancer

There is a relationship between lead-induced chronic

renal disease and renal adenocarcinoma. In fact, an increased incidence of renal cancer occurs among leadexposed workers even without a statistically increased rate
of chronic renal disease. Variability in individual susceptibility may be explained by differences in lead-binding pro teins among lead-exposed persons.
PATHOLOGY

With acute lead nephropathy, alterations in mitochondrial structure and cytosolic and acid-fast intranuclear inclusion bodies can be seen. With chronic lead nephropathy,
a nonspecific focal interstitial nephritis predominates with
tubular atrophy resulting in shrunken kidneys. The characteristic inclusion bodies are often absent. The histopathology is virtually indistinguishable from that of nephro sclerosis. The finding of a variety of immunoglobulin
deposits in glomerular capillaries and tubular basement
membranes, suggests that immune mechanisms may be involved in these changes.
DIAGNOSIS

The most useful screening tests for excessive lead exposure in adults are blood lead levels and erythrocyte protoporphyrin. Blood lead levels represent recent exposure
along with some degree of equilibration with lead present in
soft tissues and bone. Erythrocyte protoporphyrin provides
evidence of impaired heme synthesis from lead toxicity.
Erythrocyte protoporphyrin concentrations are not useful
in screening for low-level exposure to lead, and thus blood
lead levels should be used as the primary screening method
for children and newborns.
In cases with chronic low-level lead exposure, a lead
mobilization test may provide evidence of excessive lead
absorption. The test is performed by measuring urinary
lead excretion following the administration of EDTA.
After baseline determinations, 1 g of EDTA is given twice,
8 to 12 hours apart. During this time, a 24-hour urine
specimen is collected. An excessive body lead burden is
indicated by excretion of more than 1000 jU,g of lead per
day. Alternatively, 1 g of EDTA in 250 mL of 5% glucose
solution may be given intravenously over the course of 1
hour. Urine is collected at 24-hour intervals for 3 consecutive days or longer depending on the level or renal function.
Excretion of more than 600 /ug is indicative of excess
body lead.
Often, the diagnosis of chronic lead nephropathy rests
on clinical evidence of indolent, slowly progressive renal
failure marked by shrunken kidneys and histologic evidence of renal cortical atrophy, tubular fibrosis, and proliferation of the interstitial tissue. The finding of low-molecular-weight proteinuria, the de novo appearance of gout, and

51. Hyperoxaluria

337

an appropriate history of environmental or occupational

exposure to lead further support the diagnosis.
TREATMENT

Community prevention measures are necessary for all

children with blood lead levels > 10 /ag/dL, nutritional intervention for levels >15 /ug/dL, medical and environmental
intervention for levels >20 /ug/dL, chelation therapy for
levels s45 /ug/dL, and immediate intervention for levels
>70 /xg/dL. In adults with blood lead levels between 40
and 80 /Ag/dL, symptoms or findings consistent with lead
intoxication require treatment. Workers with blood lead
levels >50 /Ag/dL must be removed from occupational lead
exposure until their blood lead levels fall below 40 /tg/dL.
Workers whose Wood lead exceeds 70 to 80 /ttg/dL may
require chelation therapy.
In the absence of marked interstitial fibrosis, patients
who have been demonstrated to have an excessive body
burden of lead may be treated with long-term, low-dose
chelation therapy, 1 g CaNa 2EDTA (with procaine) intramuscularly three times a week. CaNa 2EDTA is not nephrotoxic when administered at an appropriate dosage even in
patients with compromised renal function. Normalizing the
EDTA lead mobilization test defines the end point of therapy, although actual documentation of improved renal
function is limited.

Bibliography
Bernard AM, Vyskocil A, Reels H, Kritz J, Kodl M, Lauwerys
R: Renal effects in children living in the vicinity of a lead smelter.
Environ Res 68:91-95, 1995.
Fowler B A: Mechanisms of kidney cell injury from metals. Environ
Health Perspec 100:57-63, 1993.
Kim R, Rotnitsky A, Sparrow D, Weiss S, Wager C, Hu H: A
longitudinal study of low-level lead exposure and impairment
of renal function. The Normative Aging Study. JAMA 275:1177 1181, 1996.
Moel DI, Sachs HK: Renal function 17 to 23 years after chelation
therapy for childhood plumbism. Kidney Int 42:1226-1231,
1992.
Nolan CV, Shaikh ZA: Lead nephrotoxicity and associated disorders: biochemical mechanisms. Toxicology 73:127-146, 1992.
Payton M, Hu H, Sparrow D, Weiss ST: Low-level lead exposure
and renal function in the Normative Aging Study. Am J Epidemiol 140-.&11-&M, W94.
Rosello J, Gelpi E, Mutti A, De Broe M, Stolte H: Nephron
target sites in chronic exposure to lead. Nephrol Dial Transplant
9:1740-1746, 1994.
Staessen JA, Lauwerys RR, Buchet JP, et al.: Impairment of renal
function with increasing blood lead concentrations in the general
population. The Cadmibel Study Group. N EnglJ Med 327:151156, 1992.
Verberk MM, Williams TE, Verplanke AJ, De Wolff FA: Environmental lead and renal effects in children. Arch Environ Health
51:83-87, 1996.

HYPEROXALURIA
DAWN S. MILLINER

The pathologic properties of oxalate result from the

relative insolubility of calcium oxalate in body fluids
and tissues. Calcium oxalate crystals form readily in
supersaturated urine and lead to urolithiasis. Seventyfive percent of stones formed in the upper urinary tract
are composed of calcium oxalate. In certain circumstances
such as primary hyperoxaluria, oxalate-induced renal tubular and parenchymal injury leads to compromise of
renal function followed by calcium oxalate deposition in

Primer on Kidney Diseases, Second Edition

multiple organ systems and severe multisystem disease.

Idiopathic hyperoxaluria associated with urolithiasis is
the most common hyperoxaluric state. Metabolic overproduction of oxalate, as in the primary hyperoxalurias,
accounts for the most severe consequences of hyperoxaluria. Secondary forms of hyperoxaluria, including
enteric hyperoxaluria and ingestion of drugs and toxins,
are also encountered but have variable clinical manifestations.
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

338

Dawn S. Milliner
PHYSIOLOGY OF OXALATE

Oxalate is an end product of metabolism and is not

degradable by any known biologic system of humans. Elimination of oxalate depends almost entirely on renal excretion. The daily balance among absorption, metabolic production, and excretion is such that plasma oxalate levels
in healthy individuals are much lower than urine covalues
and are on the order of 2.5 /umol/L. Endogenous metabolic
production accounts for 80 to 90% of the oxalate excreted
in the urine. Conversion of precursors to oxalate occurs in
the liver, and the relevant metabolic pathways are only
partially understood (Fig. 1). Primary substrates include
glycine and ascorbate. However, exogenous ascorbic acid,
even in large amounts, does not result in substantial increases in oxalate production in normal individuals. Hepatic peroxisomal alanine glyoxylate aminotransferase
(AGT) is important in converting glyoxylate to glycine,
thereby diverting glyoxylate from oxalate production. Pyridoxine acts as a cofactor in the AGT pathway and can
sometimes be used to therapeutic benefit. Glycerate dehydrogenase is also believed to play an important role, although its specific site in the enzyme pathway is not as
well understood.
Ten to twenty percent of oxalate in the urine derives
from dietary intake. Oxalate is a constituent of many foods,
particularly those of plant origin. Estimates of the oxalate
content of the average Western diet range from 80 to
175 mg/day. In healthy subjects, only 2 to 14% of ingested
oxalate is absorbed. Bioavailability for absorption varies
considerably depending on other dietary constituents. Dietary calcium binds to oxalate in the lumen of the intestinal
tract and has a large effect on the proportion of dietary
oxalate absorbed. Availability of dietary oxalate for absorption may also be influenced by oxalate degrading bacteria residing in the intestinal tract. Concentrations of these
bacteria of up to 108/g of stool have been demonstrated.
The quantitative importance and role of oxalate degrading
bacteria has yet to be established.
Absorption of oxalate under normal conditions can occur
across gastric mucosa and all segments of the intestine,
although most oxalate from food is absorbed in the small

bowel. Patients with total colectomy appear to absorb and

excrete the same amount of oxalate as individuals with
an intact colon. Oxalate absorption does occur across the
normal large intestine, and it is at this site that enhanced
oxalate absorption occurs in patients with enteric hyperoxaluria. Absorption of dietary oxalate occurs early after a
meal as demonstrated by increases in urine oxalate occurring within 2 to 4 hours following ingestion of high oxa late foods.
Data regarding renal handling of oxalate are limited.
There is general agreement that oxalate is freely filtered
and then undergoes bidirectional transport along the renal
tubule. In humans, the proximal tubule appears to play a
major role in both absorption and secretion. The proximal
tubule secretion of oxalate exhibits some of the characteris tics of an active, carrier-mediated transport process. Oxa late is one of many anions secreted by the proximal tubule.
Although it has been speculated that oxalate shares a common secretory pathway with other organic anions such as
urate or PAH, recent studies of isolated membrane vesicles
have suggested that the oxalate transporters are distinct
but may share common partners such as sodium with other
transporters. Whether overall tubular handling of oxalate
results in net secretion or reabsorption remains controversial. Studies in healthy subjects suggest net reabsorption,
while those in patients with disorders of high oxalate production suggest net secretion.
The upper limit of normal for urinary oxalate excretion
is approximately 0.5 mmol/24 hours. Normal values vary
somewhat depending on the assay method employed. Enzymatic and ion chromatographic techniques are the most
reliable. There is no difference in urine oxalate excretion
between males and females, and during adult life there is
no difference by age. In growing children and adolescents,
urine oxalate excretion normalized to body surface area,
is the same as for adults. Acidification of the urine to a
pH of less than 2 with HC1 during or immediately after
the collection is important to ensure accurate results. Acidification prevents precipitation of oxalate and also prevents
in vitro autoconversion of excreted ascorbate to oxalate.
Colonic excretion of oxalate has been demonstrated in
animal models of chronic renal failure, and there have been

Glycolate

Type I
Peroxisomal alanine: glyoxylate
aminotransferase deficiency

Type II
Serine

Hydroxpyruvate

D-glyceric dehydrogenase
deficiency
D-glycerate

Q!-keto-/S-hydroxyadipate

L-glycerate

Oxalate

FIGURE I Pathways of glyoxylate and hydroxypyruvate metabolism in primary hyperoxaluria. Reproduced with permission of Lippincott Raven Publishers from Smith LH: Urolithiasis. In: Schrier RW, Gottschalle CW (eds.) Diseases of
the Kidney, 4th ed., Little Brown, and Company, Boston, MA, pp. 785-813, 1988.

339

51. Hyperoxaluria

preliminary observations of increased concentrations of

oxalate-degrading bacteria in the large intestine of patients
with chronic renal failure. A role for intestinal secretion
of oxalate in chronic renal failure is suggested but remains
to be established.
METABOLIC OVERPRODUCTION OF OXALATE
The Primary Hyperoxalurias

Two types of primary hyperoxaluria (PH) have been

well described. Both are inherited as autosomal recessive
disorders. In type I PH (PH-I), a deficiency or mistargeting
of hepatic peroxisomal AGT results in accumulation of
glyoxylate which is then converted to oxalate and glycolate
(Fig. 1). Patterns of enzyme expression in patients with
PH-I are varied and include absence of both immunoreactive AGT and AGT catalytic activity (approximately 30%
of PH-I patients), presence of immune reactive AGT but
absence of enzymatic activity (approximately 25% of PHI), and presence of immunoreactive AGT and present but
reduced enzymatic activity (approximately 40%). In patients with reduced AGT enzymatic activity, the AGT is
most often selectively mistargeted from the peroxisomes
to the mitochondria. When located in mitochondria, the
AGT is ineffective in glyoxylate metabolism. Hepatic enzyme profiles do not correlate well with clinical severity
and have been of limited prognostic usefulness to date.
The gene encoding AGT is located at the tip of the long
arm of chromosome 2. A number of mutations and polymorphisms have been identified, corresponding to enzymatic phenotypes.
Patients with PH-I most often present clinically with
urolithiasis and demonstrate active stone formation. The
diagnosis is strongly suggested by a 24-hour urine oxalate
value of greater than 1.2 mmol (105 mg)/1.73 m2 and an
elevated urine glycolate. In patients with typical features
and no other identifiable causes of hyperoxaluria, these
findings may be sufficient to establish the diagnosis. As
many as 20 to 25% of patients with PH-I will have normal
urine glycolate excretion. In this circumstance, liver biopsy
for AGT analysis is needed to confirm PH-I. Some patients
with PH-I first come to medical attention with end stage
renal failure and have no antecedent history of urolithiasis.
Due to the low GFR, urine studies may not be helpful, but
the diagnosis is suggested by elevated plasma oxalate and
plasma glycolate concentrations. Interpretation of plasma
oxalate values in dialysis patients is complicated by the
high plasma values seen in patients with renal failure of
other causes. In hemodialysis patients, predialysis plasma
oxalate levels often range from 30 to 70 /xmol/L and are
difficult to distinguish from the values of 40 to 120 ju-mol/
L typically seen in PH patients on dialysis. Most patients
with end stage renal disease due to PH will have dense
nephrocalcinosis demonstrable on a plain film of the kidneys (Fig. 2). Since this is rarely seen in other clinical
settings, it is a valuable clue to the diagnosis. Abundant
oxalate deposits will be evident on kidney biopsy (Fig. 3).
If renal failure has been present for mo re than a few

FIGURE 2 Abdominal radiograph showing dense nephrocalcinosis in a patient with primary hyperoxaluria and end stage
renal disease.

months, bone marrow examination may be helpful in detecting oxalate deposits. Due to the difficulty in interpreting
laboratory results, most patients with end stage renal dis ease and suspected PH require liver enzyme analysis to
confirm the diagnosis, differentiate between PH-I and PHII, and provide a basis for decisions regarding renal replacement therapy.
The diagnosis of PH-I is most often made during the
first or second decade of life, although the disease may
remain unrecognized in some patients until the third or
fourth decade. As long as renal function is well preserved,
the excess oxalate is excreted by the kidneys, the plasma
oxalate level remains normal or only mildly elevated, and
the primary clinical problem is urolithiasis. With time, pro-

FIGURE 3 Histologic section of a kidney of a patient with end

stage renal disease due to PH-I. Note the ext ensive oxalate deposits
in the tubules and interstitium.

340

gressive damage to kidneys occurs as a result of parenchymal deposition of calcium oxalate as well as urolithiasis,
infection, urologic procedures for stone removal. When the
GFR declines to 20 to 30 mL min"1 1.73 m" 2, plasma
oxalate concentration increases, setting the stage for systemic calcium oxalate deposition. Bone marrow, trabecular
bone, myocardium, the cardiac conduction system, the vas cular system, retina, and soft tissues are the most frequently
affected. Most dialysis regimens cannot keep pace with
daily production. Patients maintained on dialysis almost
always continue to accumulate oxalate in multiple organ
systems. Prior to the implementation of aggressive treatment regimens, including transplantation, nearly all patients died before the third decade of life. Even with intensive dialysis or transplantation, patients with systemic
oxalosis are very ill and pose substantial management challenges.
Treatment strategies for PH-I must take into account
individual patient characteristics, including renal function.
Twenty to thirty percent of patients with PH-I respond to
pharmacologic doses of pyridoxine (a cofactor in the AGT
enzyme pathway) with a reduction in oxalate production
to normal or near normal. Accordingly, all patients suspected of having PH-I should receive a trial of pyridoxine
of 5 to 10 mg kg" 1 day" 1 for a minimum of 3 months. In
patients with adequate renal function, serial urine oxalate
determinations will confirm a response. In patients with
advanced renal failure, assessment of pyridoxine responsiveness is difficult, though plasma oxalate and glycolate
are sometimes helpful. When the response is uncertain,
pyridoxine should be continued. Most pyridoxine responsive patients require no more than 5 to 7 mg kg"1 day'1.
A few appear to respond only to very high-dose pyridoxine
(up to a gram per day in adult patients). Peripheral neuropathy can occur with long-term use of very high-dose pyridoxine (>10 mg kg'1 day" 1). Although it has not been
reported in PH-I patients. The risk of neuropathy due to
high-dose pyridoxine in this population is unknown.
High oral fluid intake to maintain a urine volume of 3
to 6 L daily (depending on individual oxalate production)
is a mainstay of management in all patients other than
those with end stage renal failure. Orthophosphate at 30
to 40 mg kg" 1 day" 1 of elemental phosphorus (in divided
doses) has been shown to decrease calcium oxalate crystalluria, decrease calcium oxalate supersaturation, and increase inhibitors of calcium oxalate crystal formation in
the urine of patients with PH-I. Orthophosphate also appears to be beneficial with regard to long-term preservation
of renal function. Ph osphate therapy should be considered
in all patients with adequately preserved renal function
(i.e., GFR >30 mL min"1 1.73 m" 2). It should be used
with caution at lower GFRs and is contraindicated in patients with end stage renal disease due to the risk of hyperphosphatemia. Citrate may be beneficial in patients with
PH-I as well. Magnesium has been advocated, but to date
there have been no intermediate or long-term studies to
substantiate its effectiveness. Low oxalate diets are of limited benefit in primary hyperoxaluria since metabolic overproduction accounts for nearly all the oxalate excreted.

Dawn S. Milliner

Patients with advanced renal insufficiency pose particular problems, and management is best individualized, ideally with assistance from someone who has expertise in
this disorder. Since the majority of patients progressively
accumulate tissue oxalate while on dialysis, the goal should
be early transplantation, ideally at a GFR of approximately
20 mL min"1 1.73 m" 2 and before a large oxalate tissue
burden develops. If there is a delay in transplantation, daily
hemodialysis or a combination of hemodialysis 3 to 4 days
per week along with daily peritoneal dialysis is necessary to
minimize oxalate tissue deposition. Pyridoxine responsive
patients and those with minimal tissue oxalate may be best
served by kidney transplantation alone. This is particularly
true if a living donor is available and transplantation can
be done expeditiously. Other patients may benefit more
from combined liver and kidney transplantation. Loss of
renal allografts due to rapid oxalate deposition continues
to be a major problem with both kidney and combined
liver/kidney transplantation. This is due to large tissue oxa late stores that typically accumulate before transplantation
and continue to be mobilized and excreted for months
to years after transplantation. Patients with PH require
intensive posttransplant monitoring and treatment interventions specifically directed to minimize damage to the
allograft. A transplant center experienced in the management of PH-I patients is preferred.
Type II primary hyperoxaluria (PH-II) is a rare disorder
that is due to deficiency of hepatic glycerate dehydrogenase. The cause of increased oxalate production is unclear.
Hydroxypyruvate accumulation may indirectly increase oxalate synthesis from glyoxylate. The reduction of hydroxy pyruvate to L-glycerate enhances the oxidation of glyoxy late to oxalate in a coupled reaction catalyzed by lactic
dehydrogenase (Fig. 1). The resulting increase in glycerate
production appears to account for the hyperglyceric aciduria characteristic of PH-II.
Patients with PH-II typically present in the first two
decades of life with urolithiasis. Urine oxalate excretion
tends to be lower in PH-II patients than PH-I but is usually
greater than 1.0 mmol (88 mg) 1.73 m" 2 24 h'1, and
there is considerable overlap in the amount of oxaluria
and in clinical features in types I and II. The diagnosis is
suggested by a markedly elevated urine oxalate without an
identifiable secondary cause, elevated urine glycerate, and
normal urine glycolate. The diagnosis can be confirmed by
hepatic enzyme analysis. Treatment consists of a high oral
fluid intake and, in most patients, Orthophosphate therapy.
Oral citrate and/or magnesium may be considered. Stone
formation is easier to control, and there is better preservation of renal function over time in PH-II patients when
compared with PH-I.
The term type III primary hyperoxaluria has been used
by some authors to describe atypical hyperoxaluria syndromes. The small number of patients described to date
have heterogeneous features, and some were evaluated
before reliable assays were available for oxalate, glycolate,
and glycerate and before hepatic enzyme analysis was available. Until such patients can be well characterized and

341

SI. Hyperoxaluria

diagnostic criteria developed, the use of this term is best

avoided.
Oxalate Precursors: Drugs and Toxins

Intake of drugs or toxins that are metabolized to oxalate

causes marked hyperoxaluria and can result in acute renal
failure. Ethylene glycol (found in antifreeze) ingestion is
encountered more frequently than others. Epidemics of
acute renal failure due to diethylene glycol contamination
of paracetamol elixirs have occurred in recent years in
Bangladesh, Nigeria, and Haiti. Ethylene glycol ingestion
should be considered in unexplained acute renal failure
accompanied by a severe anion gap metabolic acidosis not
attributable to other causes. The acidosis is caused by the
accumulation of glycolic acid. The glycolic acid is further
metabolized to oxalic acid and formic acid. The symptom
complex may include central nervous system changes, cardiovascular instability, and respiratory insufficiency. As little as 100 mL of ethylene glycol can be lethal. Since the
initial metabolic degradation step involves alcohol dehydrogenase, early intravenous infusion of ethanol can slow
the metabolic degradation and is of therapeutic benefit.
Hemodialysis is effective in clearing the ethylene glycol
as well as the toxic metabolites and should be initiated
promptly. Measurement of blood ethylene glycol concentration is valuable if obtained before metabolic conversion
to glycolic acid. Ethylene glycol ingestion is suggested by
calcium oxalate crystalluria and can be confirmed by a
urine oxalate level or, if the patient is anuric, by measuring
the plasma oxalate. A renal biopsy will show characteristic
deposition of calcium oxalate crystals in tubules and interstitium.
Other agents metabolized to oxalate include methoxy flurane anesthetics and xylitol. Mild hyperoxaluria has also
been observed in premature infants on parenteral nutrition.
There has been controversy regarding the potential for
ingested ascorbic acid to undergo endogenous metabolism
to oxalate. Recent studies suggest that reports of high urine
oxalate under such circumstances represent in vitro autoconversion of ascorbate to oxalate rather than in vivo metabolism. Oral doses of up to 2 g/day of ascorbic acid do
not produce an increase in urine oxalate.

fatty acids, freeing oxalate to be absorbed; and (3) increased

permeability of the colon to oxalate due to fatty acids. All
of these factors are influenced by diet composition. A high
oxalate diet, especially if dietary calcium is low or fat content is high, will be particularly disadvantageous and can
have a pronounced effect on the amount of oxalate absorbed. Patients with enteric hyperoxaluria often have very
active stone formation, and some patients have developed
renal calcium oxalate deposition and renal failure. The
hyperoxaluria is moderate, most often 0.8 to 1.1 mmol
(70-97 mg) 1.73 m" 2 24 rr1. In adults, enteric hyperoxaluria
is by far the most common cause of moderate hyperoxa luria.
The most effective treatment is to attenuate or eliminate
the malabsorption. A low fat diet can be very helpful for
patients with steatorrhea. Cholestyramine should be considered in patients with bile acid malabsorption. A low
oxalate, high calcium diet should be prescribed. Occasionally, calcium supplements are recommended. Increased
oral fluid intake as tolerated is important in improving the
solubility of oxalate in the urine. Malabsorption can cause
increased lithogenicity of the urine by a variety of mechanisms. Hyperoxaluria is typically only one component. Others include increased urine osmolality due to fluid loss in
the gastrointestinal tract, hyperuricosuria, hypomagnesemia, and hypocitric aciduria. The multifactorial aspects of
urinary tract stone formation in patients with malabsorp tion should be taken into account and treatment regimens
individualized. Some patients with difficult stone problems
and persistently low urine pH may benefit from citrate
or bicarbonate administration. Potassium or magnesium
supplements may be needed.
Dietary Factors

In normal subjects, an increase in dietary animal protein

from 55 to 89 g/day results in an increase in urinary oxalate
of up to 24%. This is because meat proteins contain amino
acids that are partially metabolized to oxalate. Low calcium
diets result in higher oxalate excretion due to reduced
binding of oxalate to calcium in the intestinal lumen, with
increased absorption of the unbound oxalate. The contribution of dietary oxalate to urine oxalate excretion is less
than that of either protein or calcium but can be a significant
factor when the diet is also low in calcium.

INCREASED ABSORPTION OF OXALATE

Enteric Hyperoxaluria

HYPEROXALURIA IN IDIOPATHIC UROLITHIASIS

Enteric hyperoxaluria has now been recognized in a variety of clinical circumstances including small bowel resection, small bowel bypass for obesity, primary small bowel
disease with malabsorption, chronic pancreatitis, and liver
diseases such as primary biliary cirrhosis and chronic cirrhosis of other causes. The common feature is intestinal malabsorption of other substances with a resultant increased absorption of oxalate in the colon. Factors responsible for
the enhanced oxalate absorption include: (1) Increased delivery of malabsorbed bile acids to the colon thereby increasing colonic permeability to oxalate; (2) complexing of
calcium in the lumen of the intestine with malabsorbed

Twenty to thirty percent of patients with idiopathic uro lithiasis have hyperoxaluria. Due to the high prevalence of
urolithiasis in the general population, this form of hyperoxaluria accounts for a large majority of the patients with
hyperoxaluria encountered in any clinical practice. The
degree of idiopathic hyperoxaluria is modest (typically
in the range of 0.5 to 0.8 mmol (44 to 70 mg) 1.73 nr2
24 Ir1 ) but has an effect on urinary supersaturation of
calcium oxalate that exceeds equivalent increases in urine
calcium concentrations. There may be more than one cause
for the hyperoxaluria in this condition. Enhanced gastroin-

Dawn S. Milliner

342
testinal absorption of oxalate, an abnormality of cellular
oxalate transport, and alterations in glyoxylate metabolism
may all play a role.
Enhanced absorption of dietary oxalate (up to 16 to 20%
of ingested oxalate) has been found in a number of studies
of patients with idiopathic urolithiasis. However, most studies involved manipulations of dietary calcium, and it is
unclear whether the increased absorption of dietary oxalate
was a primary feature or was a byproduct of the lower
calcium content of the study diets.
Abnormalities of cellular transport of oxalate may occur
in idiopathic urolithiasis and have implications for intestinal absorption and excretion and for renal tubular secretion
of oxalate. An abnormal transmembrane flux rate of oxa late in red blood cells is characteristic of patients with
idiopathic calcium oxalate stone disease and is not present
in patients with secondary forms of calcium urolithiasis.
This trait has been shown to be autosomal, monogenic,
and dominant with complete penetrance but variable expressivity. It has now been described in several ethnic
groups. Renal cortical and papillary epithelial cells and
colonic epithelial cells share many features of oxalate transport with RBCs. The increased RBC oxalate flux found in
patients with idiopathic calcium oxalate stone disease is
associated with faster intestinal absorption and a higher
renal clearance of oxalate. Some patients with idiopathic
hyperoxaluria have an increased fractional excretion of
oxalate that can exceed 1. Identification of this transport
abnormality may have implications for treatment, since the
transport is returned to normal by thiazide diuretics. To
date, however, there have been no reports of the effect of
thiazides on the urine oxalate in patients with this RBC
transport abnormality.
Elevations of urine glycolate have been observed in a
small number of patients thought to have idiopathic hyperoxaluria suggesting increased metabolic production of oxa late. Administration of pyridoxine has been reported to
reduce both the oxalate and glycolate to normal in these
circumstances. Additional studies are needed to more fully
characterize glyoxylate metabolism in such patients.
Patients with idiopathic hyperoxaluria are most effectively managed with a high oral fluid intake to reduce the
concentration of oxalate in the urine and a low oxalate
diet that contains an adequate amount of calcium. Foods
and beverages with high oxalate bioavailability such as
peanuts, almonds, pecans, chocolate, spinach, rhubarb, tea,
and colas should be avoided. If these measures are insufficient to control stone formation, additional treatment is
warranted. Neutral phosphate and potassium citrate have
been shown to be effective in idiopathic calcium oxalate
urolithiasis. If there is concomitant hypercalciuria, addition
of a thiazide may be considered.

Bibliography
Allison MJ, Cook HM, Milne DB, et al.: Oxalate degradation by
gastrointestinal bacteria from humans. J Nutr 116:455, 1986.
Brinkley LJ, Gregory J, Pak CYC: A further study of oxalate
bioavailability in foods. J Urol 144:94-96, 1990. Chlebeck PT,
Milliner DS, Smith LH: Long-term prognosis in primary
hyperoxaluria type II (L-Glyceric Aciduria). Am J Kid ney Dis
23:255-259, 1994.
Cochat P, Scharer K: Should liver transplantat ion be performed
before advanced renal insufficiency in primary hyperoxaluria
type 1? Pediatr Nephrol 7:212-218, 1993.
Danpure CJ, Rumsby G: Enzymology and molecular genetics of
primary hyperoxaluria type 1. Consequences for clinical management. In: Khan SR (ed) Calcium Oxalate in Biological Systems. CRC Press, New York, pp. 189-205, 1995. Danpure CJ,
Smith LH: The primary hyperoxalurias. In: Coe FL, Favus MJ,
Pak CYC, Parks HJ, Preminger GM (eds) Kidney Stones: Medical
and Surgical Management. Lippincott-Raven, Philadelphia, pp.
859-881, 1996.
Hanif M, Mobarak MR, Ronan A, et al.: Fatal renal failure caused
by diethylene glycol in paracetamol elixir: The Bangladesh epidemic. Br Med J 311:88-91, 1995.
Hatch M, Freel RW: Oxalate transport across intestinal and renal
epithelia. In: Khan SR (ed) Calcium Oxalate in Biological Systems. CRC Press, New York, pp. 217-238, 1995. Hillman RE:
Primary hyperoxalurias. In: Scriver CR, Beaudet AL, Sly WS,
Valle D (eds) Metabolic Basis of Inherited Diseases, 6th ed.,
McGraw-Hill, New York, pp. 933-944, 1989. Johnson CM,
Wilson DM, O'Fallon WM, et al.: Renal stone epidemiology: A
25-year study in Rochester, Minnesota. Kidney Int 16:624-631,
1979.
Lindsjo M: Oxalate metabolism in renal stone disease with specific
reference to calcium metabolism and intestinal absorption.
Scand J Urol Nephrol 119:1 -53, 1989.
Marangella M, Fruttero B, Bruno M, et al.: Hyperoxaluria in
idiopathic calcium stone disease: Further evidence of intestinal
hyperabsorption of oxalate. Clin Sci 63:381-385, 1982.
Marangella M, Petrarulo M, Vitale C, et al.: Plasma and urine
glycolate assays for differentiating the hyperoxaluria syndromes. /
Urol 148:986-989, 1992.
Milliner DS, Eickholt JT, Bergstralh E, et al.: Primary hyperoxaluria: Results of long-ter m treatment with orthophosphate and
pyridoxine. N EnglJ Med 331:1553-1558, 1994. Mitwalli A,
Ayiomamitis A, Grass L, et al.: Control of hyperoxaluria with large
doses of pyridoxine in patients with kidney stones. Int Urol
Nephrol 20:353-359, 1988. Robertson WG: Epidemiology of
urinary stone disease. Urol Res
18(1):S3 -S8, 1990.
Smith LH: Enteric hyperoxaluria and other hyperoxaluric states.
In: Coe FL, Brenner BM, Stein JH (eds) Contemporary Issues
in Nephrology. Churchill Livingstone, New York, pp. 136164, 1980.
Wilson DM, Liedtke RR: Modified enzyme-based colorimetric
assay of urinary and plasma oxalate with improved sensitivity
and no ascorbate interference: Reference values and sample
handling procedures. Clin Chem 37:1229 -1235, 1991.

MEDULLARY SPONGE KIDNEY

ELLIS D. AVNER

Medullary sponge kidney (MSK) is a renal cystic disorder

characterized by tubular cystic lesions in the inner medullary regions of the kidney. Medullary sponge kidney is a
common, and in most instances, nonheritable, disorder in
which ectasia and cyst formation localized to intrapyramidal or intrapapillary segments of renal collecting tubules
are generally asymptomatic unless complicated by nephro lithiasis, hematuria, or urinary infection.

ence of remnants of embryonal tissue in some cases, and the

reported association with other congenital abnormalities
suggest that MSK is a congenital disorder of abnormal
collecting duct development.
CLINICAL FEATURES

In the absence of complications, MSK is a benign asymp tomatic condition in which glomerular filtration rate is normal and structural changes in medullary collecting tubules
are associated with mild impairments of urinary concentration or defects in acidification that are of no clinical concern. Proteinuria, other abnormalities of the urinary sediment, and hypertension are not seen in the absence of
complications.
MSK may present symptomatically with nephrolithiasis,
hematuria, or urinary tract infection during the second or
third decades of life, although symptoms have been reported in children as young as 2 years of age. Five to
twenty-one percent of patients with nephrolithiasis have
underlying MSK. In such patients, metabolic studies reveal
a high incidence of hypercalciuria (40 to 50%), which may
lead to mild secondary hyperparathyroidism in some patients. It is believed that hypercalciuria, in combination
with increased urinary pH due to mild urinary acidification
defects and urinary stasis in ectatic tubules leads to stone
formation in MSK. Increased urinary oxalate and decreased urinary citrate reported in some patients may also
contribute to nephrolithiasis. Stones in MSK are generally
composed of calcium oxalate and apatite, although concurrent infection with urease producing organisms can lead
to aggressive struvite stone formation.
Gross hematuria has been reported in 10 to 20% of patients with MSK and intermittent microscopic hematuria
is present in the majority. Hematuria may be unrelated to
stones or infection and may be secondary to hypercalciuria
or increased fragility of ectatic or cystic tubules that may
bleed with minor or inapparent trauma.
Urinary tract infection may also frequently complicate
the course of MSK in the presence or absence of nephrolithiasis. Patients may initially present with urinary tract
infections, and more than two infections occur in 35% of
female and 5% of male patients with MSK. Older reports
reveal chronic pyelonephritis in 10 to 15% patients with
MSK, although the current prevalence is probably much
lower due to improvements in diagnosis and treatment of

DEFINITIONS AND EPIDEMIOLOGY

MSK is a relatively common renal development abnormality in which ectasia and cyst formation of medullary
collecting ducts give rise to a sponge-like appearance of
the renal medulla. The frequency of MSK in the general
population is estimated to be between 1 :5000 and
1:20,000, although its true prevalence is unknown because
patients are asymptomatic unless their course is complicated by nephrolithiasis, hematuria, or infection. MSK has
been detected in 0.5 to 10% of unselected patients being
studied with intravenous pyelograpy for urological disease.
Most cases of MSK are believed to be sporadic, nonheritable, developmental abnormalities. However, familial
clustering has been reported, and pedigrees in which the
disease is transmitted in an autosomal-dominant fashion
have been described. Since MSK is frequently asymptomatic and difficult to diagnose in mildly affected individuals,
the number of heredofamilial cases may be underreported.
PATHOLOGY AND PATHOPHYSIOLOGY
MSK is characterized pathologically by ectasia and cyst
formation localized to the intrapyramidal or intrapapillary
segments of renal medullary collecting tubules. The disease
may be asymmetrical and focal, with one or more renal
pyramids demonstrating dilated collecting tubules and multiple small cysts measuring 1.0 to 7.5 mm in diameter. Microdissection studies have revealed the cysts may communicate with each other, with ectatic collecting tubules, or
directly with the minor calyces. Extensive changes can lead
to overall enlargement of affected pyramids and calyces.
The etiology of tubular ectasia and cyst formation in
MSK is unknown. Proposed theories, such as fetal collecting tubular obstruction or structural alteration secondary to
hypercalciuria, are unsubstantiated and have little scientific
basis. The failure of structural lesions to progress, the pres -

Primer on Kidney Diseases, Second Edition

343

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

Ellis D. Avner

344

urinary tract infections over the past two decades. No studies have actually documented an increased incidence of
urinary tract infections in uncomplicated MSK when compared to control populations, and there is no reported increase of risk factors for renal infection (such as vesicoureteral reflux, bladder dysfunction, or bladder outlet
obstruction) in affected patients. However, in MSK complicated by obstructing stones, urinary stasis probably represents a risk factor for renal infection.
MSK has been associated with a variety of developmental abnormalities of the kidneys and other organ systems, including congenital hemihypertrophy and, less frequently, the Ehlers-Danlos syndrome, congenital pyloric
stenosis, Marfan syndrome, Caroli disease, cardiac malformations, Beckwith-Wiedemann syndrome, congenital anodontia, polycystic kidney disease, horseshoe kidney, and
renal malrotation or duplication. The common embryopathic basis of these associations is unknown.
MSK is diagnosed by intravenous pyelography. The dilated collecting tubules, -when opacified by contrast medium, appear as linear striations leading to a "blush-like"
pattern of the renal medulla or as spherical cysts leading
to a "bunch of grapes" or "bouquet of flowers" pattern
(Fig. 1). The dilated collecting tubules fill poorly (or not
at all) by retrograde pyelography. With severe involvement,

the renal pyramids and corresponding calyces as well as

the entire kidney, may be enlarged. With superimposed
nephrolithiasis, dense clusters of stones in the dilated and
cystic medulla are often appreciated. Mild or asymptomatic
cases may be difficult to diagnose and identification depends on optimal opacification of the pelvocalyceal system
with good visualization of each individual calyx and infundibulum. There is generally no role for sonography, computerized tomography, or arteriography in the diagnosis
of MSK.
The characteristic pyelogram in an asymptomatic individual or in association with nephrolithiasis, hematuria, or
urinary tract infection is generally diagnostic of MSK. Mild
or localized disease may present a radiographic differential
diagnosis of autosomal-recessive polycystic kidney disease,
calyceal diverticuli, renal tuberculosis, renal papillary necrosis, or other causes of nephrocalcinosis (such as distal
renal tubular acidosis). However, these conditions are
readily differentiated from MSK on clinical grounds.
PROGNOSIS AND THERAPY

Unless complicated by nephrolithiasis, hematuria, or infection, MSK is an asymptomatic condition which is not
associated with any increased morbidity or mortality.
Asymptomatic patients should be regularly screened for
hypercalciuria and bacteruria. Hematuria in MSK is a benign condition which resolves spontaneously and requires
no specific therapy in the absence of nephrolithiasis or
infection. It is unclear whether treatment of hypercalciuria,
if present, will reduce the incidence of hematuria in MSK.
Nephrocalcinosis is treated with standard regimens of fluid
intake and, if hypercalciuria is present, thiazide diuretics.
Inorganic phosphates have also been reported to be particularly effective in treating nephrolithiasis for patients with
MSK. Urinary tract infections are treated with standard
antimicrobial regimens. Low-dose urinary tract antimicrobial prophylaxis is recommended for patients with recurrent infections.
Bibliography

FIGURE I Intravenous pyelogram of a patient with MSK. The

classical medullary pyramidal "blush " is demonstrated (arrowhead) in the lower pole. (Figure kindly provided by W. Bush,
M.D. and W. Winters, M.D., Department of Radiology, University
of Washington.)

Abehouse BS, Abehouse GA: Sponge kidney: A review of the

literature and a report of five cases. J Urol 84:252-267, 1960.
Hildebrandt F, Jungers P, Grunfeld J-P: Medullary cystic and
medullary sponge renal disorders. In: Schrier RW, Gottschalk
CW (eds) Diseases of the Kidney, 6th ed. Little Brown, Boston,
pp. 499-520, 1996.
O'Neil M, Breslav NA, Pak CY: Metabolic evaluation of nephrolithiasis in patients with medullary sponge kidney. JAMA
245:1233-1236, 1987.
Osther PJ, Mathiasen H, Hansen AB, Nissen HM: Urinary acidification and urinary excretion of calcium and citrate in women
with bilateral medullary sponge kidney. Urol Int 52(3):126130, 1994.
Palubinskas AJ: Renal pyramidal structure opacification in excretory urography and its relation to medullary sponge kidney.
Radiology 81:963-970, 1963.
Yendt ER: Medullary sponge kidney. In: Gardner KD, Bernstein
J (eds) The Cystic Kidney. Kluwer, Dordrecht, pp. 379-392,1990.

RENAL PAPILLARY NECROSIS

GARABED EKNOYAN

blood flow sufficiently to result in ischemia of this relatively

underperfused region.
A second factor predisposing the papilla to necrosis is
the ability of the tubule to concentrate solute to in this
region. Although necessary to promote water reabsorption,
this has a deleterious effect when potentially nephrotoxic
agents are concentrated in the medulla, with the greatest
accumulation being in the papillary tip. This explains the
prevalence of renal papillary necrosis in individuals who
abuse analgesics (phenacetin, acetaminophen, salicylates).
While the coadministration of these agents provides a biochemical basis of their cytotoxicity, by producing an oxidant
stress and blocking its reduction, it is their concentration
in the medulla that localizes the initial and major injury
caused by analgesic abuse to the papilla. Experimental
studies have demonstrated that abolition of medullary hypertonicity by water diuresis results in a reduction in the
concentration of analgesics in the papilla and provides protection from RPN.
Another aspect of medullary hypertonicity relevant to
papillary necrosis may be its detrimental effect on the normal phagocytic function of polymorphonuclear leukocytes,
which may predispose to infection. Urinary tract infection
was once considered a principal cause of papillary necrosis.
However, while urinary tract infection is present in most
patients with papillary necrosis, it is not a uniform finding. It
may be a secondary complication superimposed on necrotic
foci, particularly when the necrotic tissue interferes with
the normal outflow of urine, i.e., obstruction occurs. Independent of infection, obstruction of the urinary tract can
cause RPN, because of reduced medullary blood flow. Following acute obstruction, an initial period of vasodilatation
and increased blood flow is followed by significant vasoconstriction with a decline in blood flow that persists for the
duration of obstruction. These changes in blood flow appear to be mediated by cytokines released by infiltrating
monocytes which invade the interstitium shortly after obstruction. That this monocyte-mediated vasoconstriction
should exert its most detrimental effect in the papilla is
not unexpected given the sparse blood flow of this region
and its dependence on vasodilatory prostanoids.
In the clinical conditions which have been associated
with papillary necrosis more than one causative factor (obstruction, infection, diabetes, analgesic abuse, and NSAID
use) is present in over half of the patients who develop

Necrosis of the renal parenchyma may affect the cortex

or the-medulla. Unlike cortical necrosis which is an acute
and often catastrophic event, necrosis of the medulla is a
chronic event which generally pursues an insidious course
and is often localized to the inner zone of the medulla
and more specifically the papilla. Renal papillary necrosis
(RPN) occurs in a relatively limited number of individuals
afflicted with an apparently disparate group of diseases
(Table 1).
PATHOGENESIS

Restriction of the necrotic lesions to the papilla can be

ascribed to the unique structural and functional features
of this region. The first is the blood supply to the papilla.
The rich vascular plexus formed by the descending and
ascending vasa rectae is principally devoted to the countercurrent exchange mechanism necessary to maintain medullary hypertonicity. The small fraction of medullary blood
flow that serves a nutrient function is provided by capillaries that branch off for this purpose. Hence, total medullary
blood flow cannot be equated with tissue supply. Additionally, the size of the vessels and their intercommunications
and branching nutrient capillaries gradually decrease during the course of their descent to the inner zone such that
the tip of the papilla has only small terminal vessels with
sparse intercommunications. As there is a three- to fourfold
increase in interstitial mass in the inner zone of the medulla
as compared to that of the cortex and medulla, the net
effect is a relatively poor blood supply to the parenchyma
of the papilla compared with the remainder of the kidney.
Conditions associated with occlusive lesions of the small
blood vessels of the kidney (diabetes mellitus, sickle hemo globinopathy, transplanted kidney) therefore predispose
to ischemic necrosis of the renal papilla. Furthermore, over
50% of RPN cases are observed in individuals greater than
60 years of age (except in sickle hemoglobinopathies). The
propensity of the elderly to arteriosclerosis has been implicated as a further cause of reduced medullary blood flow.
Another feature of the papillary vasculature that predis poses it to papillary necrosis is its apparently greater dependence on vasodilator prostanoids. On a mole per tissue
weight basis the ratio of prostaglandin synthetase activity
of the papilla to that of the medulla and cortex is 100:10:1.
Agents that inhibit cyclooxygenase activity [nonsteroidal
anti-inflammatory drugs (NSAIDs)] could compromise

Primer on Kidney Diseases, Second Edition

345

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

346

Garabed Eknoyan
TABLE I

Clinical Conditions Associated with Renal

Papillary Necrosis
Condition

Frequency

Diabetes mellitus

50-60%

Urinary tract obstruction

10-40%

Analgesic abuse

15-20%

Sickle hemoglobinopathy

10-15%

Renal allograft rejection

<5%

Pyelonephritis

<5%

Note: The figures indicate the frequency wi th

which each cause has been noted in major reviews
of RPN cases.

RPN. As such, although each of these clinical conditions

(Table 1) alone may cause papillary necrosis, the coexis tence of more than one predisposing condition increases
the risk of papillary necrosis. Thus, the diabetic who abuses
analgesics would be more prone to develop medullary injury. To the extent that the region injured by diabetes now
produces less vasodilatory prostanoids it is more prone to
further vascular injury (due to NSAIDs). Additionally, the
necrotic focus can become the nidus of an infection; should
it slough it could cause obstruction. Once established, a
vicious cycle of vasospasm, vascular occlusion, infection,
and obstruction may lead to full-blown papillary necrosis.
Thus, papillary necrosis may be considered to be the result
of an overlap phenomenon where multiple detrimental factors operate in concert to produce medullary injury.
COURSE AND CLINICAL MANIFESTATIONS

The necrotic process may be localized to one papilla or

involve several. Both kidneys are involved in 65-70% of
cases. Patients who have a unilateral lesion at the time of
initial diagnosis will often develop papillary necrosis in
the other kidney as well. The process begins with foci of
coagulative necrosis , consistent with ischemic necrosis,
which coalesce and extend to involve the rest of the tissue.
Depending on the localization of the initial necrotic process
it may assume the medullary form, in which the necrosis
is in the innermost medullary region while the fornices and
papillary tip are viable, or the papillary form, in which the
fornices and papillary tip are destroyed. The medullary
form is more common in sickle cell hemoglobinopathies.
The necrotic lesions have a well-demarcated sharp border and proceed to form a sequestrum which may either
calcify, slough, or be resorbed leaving a sinus tract or cavity
at its site. Cavity and sinus formation is more common in
the medullary form while calcification and sloughing is
more common in the papillary form. The sloughing is generally associated with hematuria, which can be massive. The
passage of sloughed necrotic tissue may be associated with
lumbar pain and ureteral colic which is clinically similar to
that of nephrolithiasis. The necrotic tissue or the stagnant
urine in the cavities may be the nidus of a urinary tract

infection that can either be chronic, smoldering and recurrent, or present in an acute, fulminant form. With the advent of antibiotics and improved management of urinary
tract infection, the central role once attributed to infection
in the development and course of RPN has diminished considerably.
The course of papillary necrosis is variable. Occasionally,
it will present as an acute disease with septicemia and
rapidly progressive renal failure. Sometimes, it will pursue
a protracted but symptomatic course with recurrent episodes of urinary infection or renal colic. More commonly,
the lesions will remain totally asymptomatic, with the papillary RPN detected as an incidental finding on urography
or unexpected discovery at postmortem examination.
The level of renal insufficiency which develops depends
on the number of papillae involved. While some loss of
renal reserve is expected to result from any renal parenchymal necrosis, renal failure does not always occur. Even
when the lesions are bilateral, if they affect only a few of
the papillae on each side, sufficient unaffected nephrons
remain to maintain adequate renal function. Even when
several papillae are affected, localization of the necrosis to
the papillary tips results in the loss of only the juxtamedullary nephrons whose loops descend to the papillary tip
while the cortical nephrons, which terminate in the outer
zone of the medulla, are spared leaving sufficient functioning nephrons to maintain homeostasis. As more of the
papillae necrose and cortical scarring develops renal failure
will ultimately occur.
Because the deep nephrons are primarily affected, an
inability to concentrate the urine maximally is an early
manifestation. Consequently, polyuria and nocturia are
common and may be a presenting complaint, can be elicited
in the history, or may be demonstrated by appropriate
testing.
Proteinuria is common (70 to 80% of cases) but is usually
only modest (<2 g/day). Pyuria is also common (60 to 80%
of cases). Microscopic hematuria (20 to 40%) and gross
hematuria (20%) are less common.

DIAGNOSIS

In symptomatic patients, the diagnosis can be made on

finding portions of necrotic tissue in the urine, the pathologic examination of which establishes their papillary origin. A deliberate search for papillary fragments should be
made by straining the urine through a filter.
In the absence of a tissue diagnosis excretory urography
has been the best method to establish a diagnosis. Unfortunately, the radiologic changes do not become apparent
until the lesions are advanced and the papillae are shrunk
or sequestered (see Fig. 1). Ultrasonography and computerized tomography are less sensitive but, on occasion, can
be helpful in patients with poor renal function where excretory urography cannot visualize the outflow tracts and is
associated with increased susceptibility to dye-induced nephropathy.

S3. Renal Papillary Necrosis

347
THERAPY

In the absence of a causative factor which can be avoided

(analgesics) or surgically corrected (obstruction), therapy
is directed toward associated complications. Control and/
or eradication of urinary infection is essential. The blood
sugar of diabetics should be well controlled. NSAID should
be avoided. Control of hypertension, as in any other renal
disease, is important. Anti-hypertensive agents which reduce renal blood flow (/3-blockers, thiazides) are best
avoided. Volume depletion, which compromises renal
blood flow, and dehydration, which increases blood viscosity, should be prevented.
Bibliography

FIGURE I Retrograde pyelogram of a diabetic patient with papillary necrosis. The calyces are all blunted. One (arrow) demonstrates a sloughed papilla nearly encircled by contrast the "ring"
sign. (Courtesy Dr. William L. Campbell.)

Bach PH, Hardy TL: Relevance of animal models to analgesicassociated renal papillary necrosis. Kidney Int 28:605-613,
1985.
Eknoyan G, Qunibi WY, Grissom RT, Tuma SN, Ayus JC: Renal
papillary necrosis: An update. Medicine 61:55-73, 1982.
Griffin MD, Bergstralh EJ, Larson TS: Renal papillary necrosisA sixteen year clinical experience. J Am Soc Nephrol 6:248256, 1995.
Groop L, Laasonen L, Edgren J: Renal papillary necrosis in patients with IDDM. Diabetes Care 12:198-202, 1989.
Lagergren C, Ljungqvist A: The intrarenal arterial pattern in renal
papillary necrosis. Am J Pathol 41:633-643, 1962.
Russell GI, Bing RF, Thurston H, Swales JD: Haemodynamic
consequences of experimental papillary necrosis in the rat. Clin
Sci 72:599-604, 1987.
Sabatini S, Eknoyan G (eds): Renal papillary necrosis. Semin Nephrol 4:1-106, 1984.
Shyan-Yih C, Porush JG, Faubert PE: Renal medullary circulation:
Hormonal control. Kidney Int 37:1 -13, 1990.

OBSTRUCTIVE UROPATHY
STEPHEN M. KORBET

INTRODUCTION

is directly linked to the duration and degree of obstruction,

early identification is crucial.

Obstructive uropathy refers to the structural or functional interference with normal urine flow anywhere along
the urinary tract from the renal tubule to the urethra. The
resultant increase in pressure within the urinary tract proximal to the obstruction contributes to a number of structural
and physiologic changes. Hydronephrosis, dilatation of the
calyces and renal pelvis, is the anatomic outcome of an
obstructive process that affects the collecting system distal
to the renal pelvis (Fig. 1). Hydroureter (the term applied
to ureteral dilatation) can accompany hydronephrosis
when the level of obstruction is distal to the ureteropelvic junction. The functional and pathologic changes of
the kidney that often ensue are referred to as obstructive
nephropalhy.
In 1993. almost 400,000 patients were discharged from
the hospital with a diagnosis of obstructive uropathy. The
overall prevalence of obstructive uropathy based on the
presence of hydronephrosis at autopsy is approximately
3%, with men and women affected equally. This obviously
underestimates the true incidence of the disorder, as temporary conditions such as nephrolithiasis would not be included. The frequency and causes of obstruction vary
among males and females and with age. Up to the age of
20 years, the frequency of obstruction is similar in males and
females. Strictures of the urethra or ureter and neurologic
abnormalities account for most causes of obstruction identified at autopsy in those patients < 10 years of age. The
obstruction rate is higher in women between age 20 and
60, primarily because of obstruction from pregnancy and
gynecologic cancers. Above age 60 years, obstructive uropathy is more common in males due to benign prostatic
hypertrophy or prostate cancer.
If left untreated, urinary tract obstruction can result in
progressive, irreversible loss of renal function and end stage
renal disease if both kidneys are affected or if only a solitary
kidney is present. However, obstructive uropathy is one of
the few potentially curable forms of renal disease. It should
therefore be considered in the differential diagnosis of any
patient presenting with unexplained acute or chronic renal
failure. Since the overall success of therapeutic intervention

Primer on Kidney Diseases, Second Edition

CLASSIFICATION, CLINICAL AND

LABORATORY MANIFESTATIONS

Urinary tract obstruction is often classified according to

the duration, location, and degree of the obstructive
process. The duration of obstruction is described as acute
(hours to days), subacute (days to weeks) and chronic
(months to years). The location (Tables 1-3) can be anywhere from the renal tubule to the urethral meatus and thus
can affect one (unilateral obstruction) or both (bilateral
obstruction) collecting systems. Finally, the degree of obstruction may be partial or complete.
These basic attributes of the obstructive process ultimately determine the clinical and laboratory manifestations
with which a patient presents. Patients with an acute obstruction typically present with the abrupt onset of pain.
With unilateral obstruction at the level of the pelvis or
ureter, severe flank pain results, often described as colicky
in nature when due to an intraluminal process such as
nephrolithiasis or papillary necrosis. If obstruction occurs
at the level of the bladder outlet, suprapubic pain and
fullness may be experienced. It may be accompanied by
urinary frequency and urgency if the outlet obstruction is
partial, or anuria if it is complete. Flank tenderness on
percussion or a suprapubic mass may be present on physical
examination in patients with outlet obstruction. Laboratory
manifestations of unilateral obstruction are often limited
to an abnormal urinalysis if the patient has two kidneys.
With intrinsic forms of obstruction, microscopic or gross
hematuria may be observed, and with either intrinsic or
extrinsic processes, secondary infection may result in pyuria
and bacteriuria. In conditions leading to bilateral obstruction (or unilateral obstruction in patients with a solitary
kidney), laboratory features of acute renal failure will be
observed.
The presenting clinical features associated with subacute
or chronic obstruction are generally more subtle and insidious in nature. Patients may be asymptomatic. Instead of
severe pain, the development of vague symptoms such as
flank or suprapubic fullness may be described, depending
on the location of the obstruction. In addition, patients

348

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

349

54. Obstructive Uropathy

TABLE I

Acquired, Intrinsic Causes of Obstructive Uropathy

Intraluminal
Intrarenal
Tubular precipitation of crystals or proteins
Acyclovir
Bence Jones proteins
Methotrexate Sulfonamides
Uric acid Extrarenal Blood
clots Fungus balls
Nephrolithiasis Papillary
necrosis

FIGURE I Intravenous pyelogram demonstrating bilateral, severe dilatation of the renal pelvis, calyces, and ureter in an elderly
male with prostate cancer.

may experience frequency, polyuria, or nocturia. They may

also have difficulty initiating or stopping urination along
with urgency if bladder outlet obstruction is present. The
physical findings may include a flank mass from a hydronephrotic kidney or a suprapubic mass extending to the umbilicus, due to a greatly distended bladder. Laboratory evaluation of the urine can be similar to that seen in acute
obstruction but may include proteinuria (usually less than
2 g/24 hr). Impairment in renal function will also be observed in patients with bilateral disease, as evidenced by
laboratory features of chronic renal failure, hyperkalemia,
renal tubular acidosis, and an inability to concentrate urine.
The differential diagnosis of obstructive uropathy is extensive (Tables 1-3). In addition to the features described
above, clinical and laboratory characteristics unique to the
individual disorders should be considered and pursued in
the evaluation of a patient with obstructive uropathy.
ETIOLOGY Acquired
Obstructive Uropathy

The acquired forms of obstructive uropathy (Tables 1

and 2) are often classified based on the location of the
obstructive process as intrinsic (obstruction occurring
within the urinary tract) or extrinsic (obstruction resulting
from external compression of the urinary tract).
Intrinsic disorders leading to obstruction are divided into
intraluminal and intramural processes. Intraluminal obstruction may be the result of renal tubular obstruction,

Intramural
Anatomic
Tumors (benign or malignant)
Renal pelvis Ureter Bladder
Urethra
Strictures (ureteral or urethral)
Granulomatous disease
Infections
Gonococcal urethritis
Nongonococcal urethritis
Chlamydia trachomatis
Ureaplasma urealyticum
Schistosomiasis haematobium
Tuberculosis
Instrumentation or trauma
Radiation therapy Functional
Cerebrovascular accident
Diabetes mellitus
Multiple sclerosis
Parkinson's disease
Spinal cord abnormalities or injury
Drugs: Anticholinergic agents, disopyramide, levodopa

otherwise referred to as intrarenal obstruction. This is most

often identified with acute renal failure in multiple my eloma from the precipitation of Bence-Jones proteins in the
tubules (myeloma kidney) and in the tumor lysis syndrome
(generally with a lymp homa) where spontaneous cell turn over or chemotherapy-induced malignant cell lysis leads
to the massive production and subsequent precipitation of
uric acid crystals within the tubules (see Chapter 31). A
number of drugs are also associated with intrarenal obstruction due to precipitation or crystal formation within the
renal tubules including sulfadiazine, sulfamethoxazole,
acyclovir, and methotrexate. The predisposition for obstruction in these conditions is enhanced in the setting of
volume contraction with the excretion of a concentrated,
acidic urine. Of the extrarenal causes of obstruction, nephrolithiasis is the most common, particularly in young men.
The most common form of stone is composed of calcium,

Stephen M. Korbet

350
TABLE 2

Acquired, Extrinsic Causes of Obstructive Uropathy

Reproductive system
Females Uterus
Pregnancy
Prolapse
Tumors
Cervical carcinoma
Fibroadenoma Uterine
carcinoma Ovary Abscess
Cysts Tumor Tubules
Pelvic inflammatory disease
Males Prostate
Adenocarcinoma of the prostate
Benign prostatic hyperplasia
Gastrointestinal system
Appendicitis Colorectal
carcinoma Crohn's
disease Diverticulitis
Pancreatitis
Vascular system
Arterial aneurysms
Abdominal aortic
Iliac
Venous
Ovarian vein thrombophlebitis
Retrocaval ureter
Retroperitoneal diseases
Fibrosis
Idiopathic
Drug related
Inflammatory
Hemorrhage
latrogenic complication of surgery
Infection
Pelvic lipomatosis
Radiation therapy
Tumor
Primary (lymphoma, sarcomas)
Metastatic (cervix, bladder, breast, prostate, etc.)

most often calcium oxalate (see Chapter 56). Papillary necrosis can lead to ureteral obstruction and may be seen
in sickle cell disease, diabetes mellitus, amyloidosis, and
analgesic abuse. In addition, gross hematuria with blood
clots resulting from any cause, especially renal trauma,
polycystic kidney disease, IgA nephropathy, or sickle cell
trait may lead to extrarenal obstruction. Intramural obstruction can be divided into anatomic or functional causes.
Of the anatomic abnormalities that lead to urinary obstruction (tumors and strictures), transitional-cell carcinomas of
the renal pelvis and ureter account for the highest proportion. Of particular note is that patients with analgesic ne-

TABLE 3

Congenital Causes of Obstructive Uropathy

Ureter
Ureteropelvic junction obstruction
Ureteroceles
Ectopic ureter
Ureteral valves
Megaureter
Bladder
Myelodysplasias (neurogenic)
Bladder diverticula
Urethra
Prune-belly syndrome
Urethral diverticula
Posterior urethral valves

phropathy are at increased risk for the development of

transitional cell carcinoma of the urinary tract. Ureteral or
urethral strictures may result from infection, trauma, or
postradiation therapy for pelvic tumors. Worldwide, schis tosomiasis haematobium infection is a considerable problem affecting nearly 100 million people. Ova deposited in
the walls of the distal ureter and bladder cause inflammation which leads to ureteral stricture and fibrosis and contracture of the bladder in 50% of chronically infected patients. The incidence of bladder cancer is also increased in
these patients. Rarely, obstructive uropathy may result
from ureteral strictures due to granulomatous disease or
urethral stricture due to gonococcal and nongonococcal
infections. Functional obstruction results from an abnormality (neuromuscular), leading to an alteration in the normal dynamic response of the urinary tract. In neurologic
disorders resulting in injury to upper motor neurons, involuntary bladder contraction (spastic bladder) results; with
lower motor neuron injury, the bladder becomes flaccid
and atonic. Either condition may lead to abnormalities in
the forward flow of urine and an increase in residual urine
volume that can result in obstructive uropathy with vesicoureteral reflux. Almost 90% of patients with multiple
sclerosis develop bladder dysfunction, which is also common in patients with long-standing diabetes mellitus. It may
complicate Parkinson's disease as well as cerebrovascular
accidents. Medications that are known to alter the neuromuscular activity of the bladder can result in decreased
contractility or tone and thus urinary retention, with an
incidence as high as 10% with some drugs (levodopa and
disopyramide). The use of these agents may be particularly
problematic when a preexisting obstructive condition such
as benign prostatic hyperplasia is present.
The most common cause of extrinsic obstruction in
women is pregnancy (see Chapter 60). In up to 90% of
pregnant women, some degree of ureteral dilatation will
be observed by the third trimester. This has been attributed
to pressure by the gravid uterus on the pelvic brim and
affects the right ureter more than the left. However, ureteral dilatation may be seen as early as the first trimester,
possibly as a result of hormonal (progesterone) effects on

351

54. Obstructive Uropathy

peristalsis. This process is often asymptomatic and resolves

spontaneously after delivery. Rarely, bilateral ureteral obstruction during pregnancy can lead to acute renal failure.
The second most common cause of urinary obstruction in
women is carcinoma of the cervix, with obstruction seen
in 30% of patients, usually a result of direct extension. In
older women, uterine prolapse can lead to hydronephrosis,
and this may occur in up to 80% of patients if there is total
prolapse. In these women the ureters are trapped between
the levator muscles and the fundus of the prolapsed uterus.
Endometriosis occasionally leads to pelvic inflammation
with fibrosis and ureteral obstruction. Pelvic inflammatory
disease is associated with significant obstruction in up to
40% of patients if a tubo-ovarian abscess is present.
Benign prostatic hyperplasia is the most common cause
of obstructive uropathy in older men, resulting in symptoms
of outlet obstruction in 50 to 75% of males over age 50
and significant hydronephrosis in 10%. Overall, adenocarcinoma of the prostate is second only to carcinoma of the
cervix as the leading form of extrinsic obstruction due to
tumor. Obstruction from prostate cancer (or any pelvic
malignancy) can be due either to direct extension of tumor
to the bladder outlet or ureters, or to metastases to the
ureters or surrounding lymph nodes.
In addition to colorectal carcinomas (ureteral metastases), a number of gastrointestinal disorders are associated
with obstruction (often unilateral) resulting from local infection and/or inflammatory processes. Vascular diseases,
the most common of which is abdominal aortic aneurysm,
will lead to obstruction from retroperitoneal fibrosis (inflammatory aneurysms) or from direct pressure of the expanding aneurysm. Rarely, systemic diseases associated
with vasculitis (systemic lupus erythematosus, polyarteritis
nodosa, Wegener's granulomatous, and Henoch-Scho'nlein
purpura) have been associated with obstruction.
A number of conditions involving the retroperitoneal
space may result in ureteral obstruction by fibrosis or direct
invasion and compression. Periureteral fibrosis can be observed postradiation, trauma, surgery, granulomatous dis ease, or infection and is associated with obstruction from
some abdominal aneurysms. Retroperitoneal fibrosis has
been linked to the use of a number of drugs, including
methysergide, bromocriptine, and /3-blockers. Idiopathic
retroperitoneal fibrosis occurs in some patients without an
obvious cause. It is predominantly a disease of men (3:1)
in the fifth and sixth decades of life. Dull, non colicky flank
pain of insidious onset is the presenting symptom in 80% of
patients. Fever, weight loss, and nonspecific gastrointestinal
complaints may also be present. On intravenous pyelography, medial deviation of the ureters is a characteristic feature. Although the etiology of this entity is unknown, the
histologic findings are those of an inflammatory process
involving collagen and fibrosis. Surgical release of the ureters (ureterolysis) is often successful in relieving obstruction and pain symptoms. The fibrous tissue appears to extends out from the aorta, encasing and drawing the ureters
medially, and may be up to 6 cm thick. Primary tumors
involving the retroperitoneal space that are frequently associated with obstruction include lymphoma (Hodgkin's dis ease) and reticulum-cell sarcoma. In addition, metastatic

spread to the retroperitoneum of a number of carcinomas,

most commonly cervix (30%) and bladder (20%), but including breast, prostate, colon, and ovary, can also result
in ureteral obstruction. The most likely cause of obstruction
in patients who have undergone pelvic irradiation for malignancy is recurrent tumor if obstruction occurs within 2
years of therapy and radiation fibrosis thereafter.
Congenital Obstructive Uropathy
Of the congenital causes of obstructive uropathy (Table
3), ureteropelvic junction obstruction and posterior urethral valves are the most common. If severe enough, obstruction may have its onset in utero and lead to major
renal abnormalities in the developing fetus. Early in development, obstruction results in a kidney that appears dysplastic. Obstruction later leads to a kidney with cortical
cysts and a reduced nephron mass. Those fetuses in whom
obstruction develops late in gestation have features similar
to those seen postnatally, such as hydronephrosis and renal
parenchymal thinning. In some cases, obstruction may not
manifest itself until childhood. Ureteropelvic junction
(UPJ) obstruction is considered the most common cause
of hydronephrosis in infancy and early childhood. In childhood, the majority of patients are males, whereas in adults
females predominate. In infancy, UPJ obstruction is bilateral in 30% of cases an uncommon finding in adults. The
presentation in children is that of an abdominal mass with
flank pain or abdominal pain and failure to thrive. In adults,
the pain is episodic and often precipitated by high urine
flow rates. Abnormal peristalsis due to a derangement in
the smooth muscles of the renal pelvis has been proposed
as a primary mechanism for UPJ obstruction. Additionally,
a hyperdistensible renal pelvis, incapable of draining completely, has been proposed in some cases. Less often, UPJ
obstruction may result from crossing blood vessels, fibrous
bands, or strictures. Posterior urethral valves, leading to
outlet obstruction, are seen strictly in males and best diagnosed by a voiding cystourethrogram. Presentation in infancy consists of a palpable bladder and kidneys with
marked renal insufficiency. Older children will often pres ent with urgency or enuresis. One of the more unusual
causes of obstructive uropathy is the prune-belly syndrome.
Predominantly seen in males, this consists of the triad of
deficiency of abdominal muscles (resulting in loose, wrinkled, redundant skin over the abdomen appearing like a
"prune"), cryptorchidism, and hydroureteronephrosis. The
obstruction is bilateral, with abnormal ureteral peristalsis
and prostatic hypoplasia implicated as possible mechanisms.
Nonobstructive Urinary Tract Dilatation
Dilatation of the urinary tract may occur without evidence of obstruction. However, if the underlying cause
is chronic, impaired renal function and atrophy of renal
parenchyma may result. This can be seen with vesicoureteral reflux, acute pyelonephritis, and high flow states (such
as diabetes insipidus or psychogenic polydipsia). The ureteral dilatation characteristic of pregnancy occurs as early

352

Stephen M. Korbet

as the first trimester and involves the right collecting system

more often than the left. In the majority of cases, the
ureteral dilatation is thought to result from decreased peristalsis, rather than mechanical obstruction, due to the muscle relaxant effects of progesterone. The dilatation resolves
within a few weeks of delivery and therefore does not
result in functional or pathologic renal impairment unless
complicated by infection.
PATHOPHYSIOLOGY

Our understanding of the consequences of urinary tract

obstruction on renal function are primarily derived from
the effects of short-term (24 hours) complete obstruction
in experimental animals. The alterations in renal function
that result are divided into those that affect either glomerular function or tubular function.

lar cells, but leukocytes (macrophages and T lymphocytes)

which infiltrate soon after obstruction. In addition to their
effects on vascular resistance, All and thromboxane A2
alter GFR by producing mesangial contraction, which leads
to a decrease in the ultrafiltration coefficient. It is of interest
that pretreatment with angiotensin-converting enzyme
(ACE) inhibitors and inhibitors of thromboxane A 2 synthesis has been shown to counteract the effects on GFR and
renal plasma flow in experimental models of obstruction.
The degree of improvement in GFR after release of
obstruction in animals is related to the duration of obstruction. Complete recovery of GFR is observed after obstruction of up to 7 days, 70% recovery with 14 days of obstruction, 30% recovery with 28 days of obstruction, and
essentially no recovery of renal function after obstruction
of 56 days.
Tubular Function

Glomerular Function

Glomerular filtration rate (GFR) declines progressively

after complete obstruction. Within the first few hours of
acute ureteral obstruction, there is an increase in proximal
tubular pressure, the magnitude of which is partially dependent on hydration status (greater with increased hydration).
This results in a decrease in net glomerular filtration pres sure (net glomerular filtration pressure = glomerular filtration pressure - intratubular pressure) and thus a decrease
in GFR. Simultaneously, an increase in the production of
prostacyclin and PGE2 leads to afferent arteriolar dilatation (vasodilative phase) and an increase in renal blood flow
which increases glomerular filtration pressure. However,
since the increase in glomerular filtration pressure is not
proportional to that of the intratubular pressure, a decrease
in net glomerular filtration pressure persists, resulting in a
GFR that is 80% of the preobstruction value.
At 4 to 5 hours after obstruction, intratubular pressure
begins to decline due to ongoing reabsorption of sodium
and water by the nephron, dilatation of the collecting system, and lymphatic absorption of solute and water. A decrease in renal blood flow due to an increase in afferent
arteriolar resistance (vasoconstrictive phase) also ensues,
leading to a decrease in glomerular filtration pressure. Since
glomerular filtration pressure declines at a faster rate than
the intratubular pressure, a further decrease in net glomerular filtration pressure occurs and GFR continues to decline. Thus, the relatively higher intratubular pressure remains a significant factor responsible for the ongoing
decrease in GFR which can be as low as 20% of normal
by 24 hours.
The vasoconstrictive phase is mediated by angiotensin
II (All), thromboxane A2, antidiuretic hormone (ADH),
and a decrease in endothelium-derived relaxing factors
(EDRF) or nitric oxide production. The increase in intrarenal production of All is a consequence of decreased delivery of sodium and chloride to the macula densa and the
effects of vasodilating prostaglandins on renin release. The
source for the increased synthesis of thromboxane in the
obstructed kidney appears to be not only intrinsic glomeru-

Abnormalities in reabsorption of sodium and water and

excretion of potassium and hydrogen ions are characteristic
of obstructive nephropathy. In the acute phases of obstruction, there is an initial increase in sodium and water reabsorption secondary to underperfusion of the distal nephron.
This is evident by reductions in urinary sodium concentration, fractional excretion of sodium, and free water clearance. In clinical practice, this results in urinary indices that
can have a prerenal appearance with a urinary sodium of
<20 mEq/L, fractional excretion of sodium of <1%, and
a urine osmolality of >500 mOsm. With more prolonged
or chronic obstruction, subsequent alterations in tubular
function result in a decrease in sodium and water reabsorption which is characterized by a urinary sodium of
>20 mEq/L, fractional excretion of sodium of >1%, and
a urine osmolality of <350 mOsm.
After the release of chronic (>24 hour) obstruction,
there is decreased sodium reabsorption by the nephrons,
leading to an increased fractional excretion of sodium. The
decrease in sodium reabsorption results from a reduction
in Na,K-ATPase activity in the nephron. Additionally, an
associated increase in excretion of calcium, phosphate, and
magnesium parallels sodium excretion. The ability to concentrate urine is impaired after release of obstruction and
results in an increased fractional excretion of water. The
concentrating defect is due to multiple factors, including
(1) decreased medullary tonicity from washout of solute
due to an increase in medullary blood flow and decreased
sodium chloride reabsorption at the thick ascending limb
of Henle, (2) decreased response to ADH, and (3) a decrease in the number of functional juxtamedullary nephrons. The greater natriuresis and diuresis observed after
release of bilateral ureteral obstruction (BUO) compared
with unilateral renal obstruction (UUO) are attributed to
the accumulation of sodium and water, urea retention, retention of other impermeable solutes, and higher levels
of atrial natriuretic peptide which occur in BUO but
not UUO.
Abnormalities in potassium and hydrogen excretion are
common in obstructive uropathy. Hyperkalemia dispropor-

54. Obstructive Ur opathy

tionate to the degree of renal dysfunction results as the

fractional excretion of potassium is lower in obstructive
uropathy than in other forms of renal disease. The decrease
in potassium secretion is attributed to a reduction in the
secretion or response of the distal tubule to aldosterone
or a combination of both. Hyperkalemic-hyperchloremic
acidosis is also frequently seen in patients with obstructive
uropathy. Explanations include a defect in hydrogen ion
secretion and maximal urine acidification (distal renal tubular acidosis), or a defect in secretion of aldosterone secondary to a decrease in renin production (hyporeninemic aldosteronism or type IV renal tubular acidosis). A decrease
in sodium reabsorption in the distal nephron has been
suggested as an additional factor. By reducing the degree
of intraluminal negativity this would decrease the voltagedependent secretion of both potassium and hydrogen. Finally, reduced levels of in H+ -ATPase (a major transport
pathway) in the cortical and medullary collecting ducts
have been observed in obstructive uropathy and may further explain the diminished hydrogen ion excretion. The
acidifying defect is usually reversible but may persist.
RENAL PATHOLOGY

The pathology of hydronephrosis is similar, irrespective

of the underlying cause of obstruction. In early hydronephrosis the kidney is enlarged and edematous with an increase in the renal pelvic cavity and blunting of the renal
papilla. Later, there are retraction and dimpling of the
papilla, which is most evident at the upper and lower poles.
Microscopically, the renal cortex appears normal but the
tubules may be dilated and Tamm-Horsfall protein is pres ent in Bowman's space. A principal lesion in the medulla
and papilla is ischemic atrophy associated with flattening
and atrophy of tubular epithelium and interstitial fibrosis.
As the hydronephrosis advances, the atrophy becomes even
more pronounced. The kidney essentially transforms into
a fluid-filled sac with loss of papillae and marked thinning
of the cortex and medulla. With progressive sclerosis and
fibrosis, few recognizable renal features remain.
The morphologic changes in obstructive nephropathy
are primarily attributed to the ischemia from the marked
reduction in renal blood flow. However, another contributing factor is invasion of the interstitium early on in obstruction by macrophages and T lymphocytes. Beginning 4 to
12 hours after obstruction, the number of interstitial macrophages increases. Chemoattractants, such as monocyte
chemoattractant peptide-1 and osteopontin, appear to be
released from tubular epithelial cells in response to the
increase in intratubular pressure from ureteral obstruction.
Fibrogenic cytokines, such as transforming growth factor(3 (TGF -/3), produced by the invading macrophages and
T lymphocytes become central in the progressive fibrosis
observed in obstructive nephropathy. TGF-/3 increases matrix synthesis by interstitial fibroblasts and decreases matrix
degradation by downregulating the production of matrix
degradation proteins and promoting the generation of proteinase inhibitors. Vasoactive compounds such as angiotensin II may directly stimulate the production of TGF-/3 by

353

tubular epithelial cells. The use of ACE inhibitor in the

experimental setting significantly decreases the degree of
tubulointerstitial fibrosis seen in obstruction, suggesting a
potential role for ACE inhibitors in the clinical treatment
of obstructive uropathy.
DIAGNOSIS

When the diagnosis of obstructive uropathy is suspected,

a careful history, complete physical, and appropriate laboratory evaluation (including biochemistry profile and urinalysis) are essential. Findings from these evaluations are
invaluable in establishing the presence and functional severity of obstruction, and in determining its etiology. Dilatation of urinary tract is the radiographic finding characteristically used to confirm the presence of obstruction (Fig. 1).
The diagnostic approach may vary depending on whether
acute or chronic obstruction is suspected.
In acute obstruction, such as in nephrolithiasis, there
may not be sufficient time for identifiable dilatation of the
collecting system to occur. In this setting, the intravenous
pyelogram (IVP) has been considered the procedure of
choice or "gold standard." Whether the use of simpler
and less invasive technology such as ultrasound (US) can
replace the IVP in detecting acute urinary tract obstruction
has been an area of intense interest (Fig. 2a and 2b), as
the contrast required for performing the IVP carries a small
but definite risk. The success of ultrasound in diagnosing
acute obstruction varies substantially (sensitivity reported
to be as low as 50% and as high as 90%). In part, the
variable success of US is attributed to its subjective nature.
It is highly operator dependent with most false negatives
due to cases of grade 1 hydronephrosis (mild dilatation of
the renal sinus central-echo complex) or nondilated obstructive uropathy (i.e., retroperitoneal fibrosis). Since
nephrolithiasis is the leading cause of acute renal colic and
obstruction, and >90% of renal stones are radiopaque,
combining plain abdominal radiography (kidney, ureter,
bladder-KUB) with US increases the sensitivity for detecting obstruction to >95%. Finally, the use of duplex doppler
US, to demonstrate the presence of a high resistive index
(>70%) from increased vascular resistance in obstruction,
has been shown to further reduce the false-negative results
associated with US in both acute and chronic obstruction.
Thus, in situations where obtaining an IVP may be difficult
or the use of contrast a major concern or contraindicated,
the combination of US and KUB may provide a viable
alternative. A limitation of US remains its inability to determine the level and the cause of obstruction as compared
to IVP. Recent enthusiasm is mounting for the use of unenhanced helical (spiral) computed tomography (CT) in the
evaluation of acute renal colic. The scan can image the top
of kidney to the base of the bladder and does not require
contrast. Furthermore, it is more sensitive than combined
KUB and US in identifying causes of renal colic and allows
visualization of the obstructive lesion.
A chronic form of obstruction, UPJ obstruction, is often
associated with symptoms that are acute in nature and
reproduced only during periods of high urine flow rates.

Stephen M. Korbet

354

FIGURE 2 (A) Ultrasound of the right kidney, long axis (outlined by long arrows), demonstrating the normal echodense central area, the renal sinus central echo-complex (between short arrows). (B) Repeat ultrasound of the right
kidney, long axis, in the same patient 3 weeks later after presentation with a 3-day history of right flank pain. Note
hydronephrosis (pyelocalyceal dilatation) as demonstrated by a marked separation of the renal sinus central echo-complex
(between short arrows).

Often a diagnosis of UPJ obstruction requires the use of

diuresis urography, which combines the use of intravenous
furosemide with an IVP in order to evaluate for dilatation
and abnormal emptying of the upper urinary tract during
a state of high urine flow. A similar evaluation can be
obtained with diuresis renography using radionuclides.
In chronic obstruction, sufficient time for dilatation of
the urinary tract has transpired, and if the obstruction is
bilateral, renal insufficiency may be present. In this setting,
the use of intravenous pyelogram contrast is best avoided
and ultrasonography is the procedure of choice with a sensitivity of 98% and a specificity of 75%. A false-negative US
evaluation should be extremely rare in the evaluation of
chronic obstruction. CT is also an accurate technique for
the detection of urinary tract dilatation and may be more
likely than US to identify the obstructing lesion. However,
with conventional CT, contrast is often required. Whether
unenhanced helical (spiral) CT will prove beneficial in the
evaluation of chronic obstruction is undetermined. Ultimately, antegrade or retrograde pyelography may be required to further define the site and cause of obstruction.
Retrograde pyelography may be of particular value when
a nondilated obstructive uropathy is suspected. Although
these studies can be useful when intravenous contrast is
contraindicated due to allergy or renal failure, the risk
of urinary tract infection is of concern, particularly with
retrograde pyelography. To evaluate causes of lower urinary tract obstruction, cystoscopy and urodynamic studies
may be utilized. Urodynamic studies are most worthwhile
when a functional abnormality of the bladder (neurogenic
bladder) is suspected.
Nonobstructive urinary tract dilatation may be differentiated from that due to obstruction by means of diuresis
renography or urography. Diuresis renography utilizes radionuclide imaging and evaluates the pattern of radionuclide elimination by each kidney before and after intrave-

nous furosemide. Prolonged retention of radioactivity or

contrast after diuresis is consistent with obstruction. In
nonobstructed dilatation, however, there is a rapid "washout" of marker with the furosemide-induced diuresis.
TREATMENT

The treatment of obstructive uropathy is dictated by the

underlying cause and its location. It should be obvious that
acute or chronic renal failure resulting from urinary tract
obstruction (obstructive nephropathy) warrants emergent
attention as the potential for permanent renal damage increases with the duration of obstruction. The urgency and
aggressiveness with which obstructive uropathy must be
treated are also determined by the severity of symptoms
as well as by the presence of infection.
Nephrolithiasis, the most common cause of acute unilateral obstruction, can most often be treated with conservative measures such as intravenous fluids and pain medications. As 90% of stones less than 5 mm pass spontaneously
with increased urine flow alone, often no additional treatment is required. However, with increasing stone size, the
likelihood of spontaneous passage diminishes, and the need
for more aggressive measures arises. The urological management of nephrolithiasis is discussed in Chapter 56.
Rarely, chronic or recurrent unilateral obstruction of any
cause may lead to advanced hydronephrosis associated with
severe pain, recurrent pyelonephritis, or pyonephrosis, in
which case a nephrectomy may be indicated, especially if
the renal function of the affected kidney is minimal.
The initial treatment approach for patients presenting
with bilateral urinary tract obstruction and renal failure is
primarily dictated by the location of the obstruction. In
patients with a neurogenic bladder or disorders involving
the bladder outlet, the placement of a urethral catheter

54. Obstructive Uropathy

will often suffice. For patients in whom a urethral catheter

cannot be passed into the bladder, a suprapubic cystostomy
may be required. Lesions obstructing the ureters result in
the need for cystoscopy and stent placement. If a stent
cannot be passed beyond the ureteral obstruction in a retrograde fashion, a percutaneous nephrostomy tube can be
placed and antegrade placement of a stent can be attempted. Placement of a percutaneous nephrostomy is successful in more than 90% of patients, resulting in clinical
improvement in up to 70% of cases. Major complications
(abscess, sepsis, hematomas) occur in less than 5% of patients.
Once acute relief of the obstruction has been achieved,
specific treatment of the underlying disease becomes the
primary focus. For example, in neurogenic disorders of the
bladder, voiding regularly and pharmacologic agents may
be useful. In patients with a spastic bladder, the anticholinergic agents oxybutynin and propantheline bromide are
employed. Particularly in patients with bladder atony, however, intermittent catheterization of the bladder (four times
daily) is required. For men with prostatic hyperplasia the
long-term treatment approach depend on the severity of
the outlet obstruction. If symptoms are minimal, and not
associated with infection or upper urinary tract abnormalities, close observation is appropriate. Cases with mild to
moderate symptoms of prostatism can be managed medically with either alpha antagonists or 5-a-reductase inhibitors. Alpha antagonists (doxazosin or terazosin) act by
relaxing the smooth muscle of the prostate and bladder
neck, thus decreasing urethral pressure and outlet obstruction. Hormonal therapy with a 5-a-reductase inhibitor (finasteride) inhibits the conversion of testosterone to the
active dihydrotestosterone and thereby leads to a reduction
in prostate size. The combined use of these agents may be
beneficial in some patients, as the agents are thought to act
synergistically. For patients with hyperplasia of the prostate
resulting in signs and symptoms of severe obstruction (significant urinary retention or renal failure), surgical intervent ion with transurethral resection of the prostate
(TURP) or transurethral incision of the prostate (TUIP)
is generally required.
In disease processes that lead to irreparable damage of
the lower urinary tract (as in bladder, cervical, or prostate
cancer), a diversion procedure such as an ileal conduit or
percutaneous nephrostomy will be needed. In patients with
obstructive nephropathy secondary to malignancy, percutaneous nephrostomy has relieved the obstruction in >75%
of cases and resulted in a significant increase in survival
(>6 months in 50%) as well as an increase in number of
days spent at home compared to patients in whom the
procedure was not performed. Thus, selected patients can
benefit from this aggressive approach.
The likelihood of regaining renal function postobstruction is dependent on the degree and duration of obstruction. In general, complete recovery of renal function is
anticipated in patients with acute uncomplicated obstruction of short duration (si to 2 weeks), and little to no

355

improvement occurs after severe, complete or partial obstruction of a prolonged duration (>12 weeks). However,
recovery of renal function in patients has been recorded
after obstruction for up to 70 days. Radionuclide renography performed several weeks after relief of obstruction has
been proposed as one method to predict recovery.
Release of bilateral obstruction can result in marked
polyuria (postobstructive diuresis). A number of factors
physiologic and pathologiclead to development of this
condition. Physiologic factors contributing to the diuresis
include excess sodium and water retention, accumulation
of urea, and other nonreabsorbable solutes that act as
osmotic diuretics, and accumulation of atrial natriuretic
peptide. Pathologic factors include decreased tubular reabsorption of sodium, inability to maximally concentrate
urine due to a decreased medullary concentrating gradient
and decreased response to ADH, and increased tubular
flow reducing equilibration time for absorption of sodium
and water. Once excess sodium and water have been
excreted, the potential for severe volume contraction as
well as hypokalemia exists if patients are not carefully
monitored and given appropriate fluid and solute replacement. Urinary output should be measured frequently
during the diuresis (at least every 6 hours and in cases
with large urine outputs, hourly). Once the patient has
diuresed enough to become euvolemic, fluid replacement
(intravenous plus oral) should be given as needed to
prevent volume contraction. A suitable method is replacement of 75% of the urine losses with intravenous fluids
having a solute composition similar to what is excreted
(0.45% normal saline), but careful monitoring to avoid
over- and underhydration is essential. Serum electrolyte
levels should be monitored closely during the diuresis,
at least daily if not more often, and replaced as needed.
The postobstructive diuresis is self-limited, resolving over
several days to a week. Persistence of the polyuria is
often due to overzealous hydration which perpetuates
the solute and water diuresis.
Bibliography

Cronan JJ: Contemporary concepts for imaging urinary tract obstruction. Urol Radio! 14:8-12, 1992.
Curhan GC, Zeidel ML: Urinary tract obstruction. In: Brenner
BM (ed) The Kidney, 5th ed. W.B. Saunders, Philadelphia, pp.
1936-1958, 1996.
Davidson AJ, Hartman DS: The dilated pelvocalyceal system. In:
Davidson AJ, Hartman DS (eds) Radiology of the Kidney and
Urinary Tract, 2nd ed. W.B. Saunders, Philadelphia, pp. 571780, 1994.
Diamond JR: Macrophages and progressive renal disease in experimental hydronephrosis. Am J Kidney Dis 26:133-140, 1995.
Haddad MC, Sharif HS, Shahed MS, et al.: Renal colic: Diagnosis
and outcome. Radiology 184:83-88, 1992.
Harrington KJ, Pandha HS, Kelly SA, Lambert HE, Jackson JE,
Waxman J: Palliation of obstructive nephropathy due to malignancy. Br J Urol 76:101-107, 1995.
Hill GS: Calcium and the kidney, nephrolithiasis, and hydronephrosis. In: Heptinstall RH (ed) Pathology of the Kidney, 4th
ed. Little, Brown, Co., Boston, pp. 1563-1629, 1992.

356

Jorge A. Velosa

Hoffman LM, Suki WN: Obstructive uropathy mimicking volume

depletion. JAMA 236:2096 -2097, 1976.
Katz DS, Lane MJ, Sommer FG: Unenhanced helical CT of ureteral stones: Incidence of associated urinary tract findings. AJR
166:1319-1322, 1996.
Klahr S: Pathophysiology of obstructive nephropathy: A 1991 up date. Semin Nephrol 11:156-168, 1991.
Klahr S: New insight into the consequences and mechanisms of
renal impairment in obstructive nephropathy. Am J Kidney Dis
18:689-699, 1991.
Klahr S, Purkerson ML: The pathophysiology of obstructive nephropathy: The role of vasoactive compounds in the hemody -

namic and structural abnormalities of the obstructed kidney.

Am J Kidney Dis 23:219-223, 1994. Klahr S, Ishidoya S,
Morrissey J: Role of angiotensin II in the
tubulointerstitial flbrosis of obstructive nephropathy. Am J Kidney
Dis 26:141-146, 1995. Spital A, Spataro R: Nondilated
obstructive uropathy due to a
ureteral calculus. Am J Med 98:509-511, 1995. Suki W,
Eknoyan G, Rector FC Jr, Seldin DW: Patterns of nephron
perfusion in acute and chronic hydrinephrosis. / Clin Invest
45:122-131, 1966. Vehmas T, Kivisaari L, Mankinen P, et al.:
Results and complications
of percutaneous nephrostomy. Ann Clin Res 20:423-427, 1988.

VESICOURETERAL REFLUX AND

REFLUX NEPHROPATHY
JORGE A. VELOSA

DEFINITIONS

Vesicoureteral reflux (VUR) is the abnormal retrograde

flow of urine from the bladder into the upper urinary tract.
Primary VUR is a common congenital abnormality of the
vesicoureteric junction. Secondary VUR results from an
anatomic or functional obstruction at any level in the urinary tract, (including inflammatory disorders affecting the
vesicoureteral junction, bladder neck obstruction, surgery
on the ureteric orifice, paraurethral diverticulum, posterior
urethral valves, ureteropelvic junction obstruction, ureteral
duplications, hypospadias, and ureteroceles). Intrarenal reflux is the retrograde flow or urine into the ducts of Bellini
(pyelotubular backflow). The term reflux nephropathy refers to renal scarring associated with VUR. It is a major
cause of end-stage renal disease in children and adolescents, accounts for 5 to 15% of cases of end stage renal
disease in adults less than 50 years of age.

PATHOGENESIS
Vesicoureteral Reflux

The anatomic features of the normal valve mechanism

of the ureterovesical junction are the oblique insertion of

Primer on Kidney Diseases. Second Edition

the ureter into the bladder and an adequate length of the

intramural segment of the ureter. As a result of these anatomic characteristics, the intravesical portion of the ureter
is compressed by the bladder musculature when it contracts
during micturition. Primary VUR may result from congenitally anomalous shortening of the intravesical portion of
the ureter with resultant valve dysfunction. The intravesical
ureter lengthens with growth, increasing the competence
of the valve mechanisms; reflux, therefore, tends to diminish or disappear with increasing age.
Little evidence supports the notion that occult VUR
occurs in asymptomatic normal individuals. Studies of the
urinary tract conducted in humans without urinary tract
complaints or infection showed that only 0.4% had VUR.
However, 50% of children in whom a urinary tract infection
developed before age 1 year and 38% of children with
urinary tract infection before age 12 years were noted to
have VUR. In adults with urinary tract infection, the prevalence is approximately 5%. There is a high familial incidence of VUR, which suggests a hereditary abnormality.
Prospective studies of siblings of patients with VUR
showed an incidence of VUR of 32 to 45%. Sixty percent of
the affected siblings had no previous symptoms to suggest
urinary tract infection, and 56% had severe VUR. Race
influences the incidence of VUR. The Southwest PediCopyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

55. Vesicoureteral Reflux and Reflux Nephropathy

atric Nephrology Study Group reported only 8 AfricanAmerican children of a total of 113 with VUR.
VUR may be unilateral or bilateral, and its severity is variable. Voiding cystographya standardized
techniqueis frequently used to determine the severity of
reflux (Table 1).
The incidence of renal scarring has been related to the
severity of the VUR. By use of the International Classification, renal scarring was identified in 85% of patients with
grade V, 37 to 64% with grade IV, 25% with grade III, and
6 to 14% with grade II VUR. To put these numbers in
perspective, mild and moderate grades of VUR occurred
more often than severe VUR, 70 and 30%, respectively.
Scar Formation

Scar development usually represents the combined effects of VUR, intrarenal reflux, and infection. Two studies
document the importance of infection in the development
of new scars. In one study, 75 children were given lowdose prophylactic antibacterial therapy; after 15 years of
follow-up, only two fresh scars were documented. In the
other, children from various institutions were treated for
symptomatic urinary tract infections, but in many, treatment of infection was delayed. New scars developed in 87
of 148 kidneys in the study.
Sterile Reflux and Urodynamic Factors

There is support for the suggestion that high-pressure

VUR and intrarenal reflux in the infant kidney can produce
damage because of urodynamic factors in the absence of
bacterial infection (sterile reflux). Severe obstructive uropathy with high intrapelvic pressures may be necessary to
produce renal damage, such as in children with posterior
urethral valves or other obstructive congenital anomalies.
However, there is still controversy regarding the frequency
with which such damage occurs clinically. This is in contrast
to the overwhelming evidence that VUR in the presence
of bacterial infection leads to renal scarring in children.

TABLE I

Grading System Adopted by the International Reflux

Study Committee

Grade I

Reflux up the ureter only

Grade II

Reflux up the ureter, pelvis, and calices with no

dilatation and normal caliceal fornices

Same as grade II plus mild or moderate
dilatation and tortuosity of the ureter; no
blunting of the fornices Grade IV
Moderate dilatation or tortuosity or both of the
ureter and moderate dilatation of the pelvis
and calices; complete blunting of fornices
Grade V
Gross dilatation and tortuosity of the ureter,
pelvis, and calices. Absent papillary
impressions in the caliees
Grade III

357
Reflux Nephropathy and Progressive Renal Failure

Progressive renal insufficiency associated with VUR and

reflux nephropathy occurs in two circumstances. In the
most severe cases of bilateral reflux, coarse scar formation
may lead to renal insufficiency. More frequently, however,
the progression to renal insufficiency in patients with reflux
nephropathy is related to complicating factors such as pro teinuria and hypertension associated with glomerulosclerosis.
Hypertension

The most common cause of hypertension in children

and adolescents is reflux nephropathy; it is a cue for the
investigation of reflux in these patients. The risk of development of hypertension increases with the severity of the
reflux nephropathy and renal scarring. In a study of adults
and children, 32 of 41 patients with reflux nephropathy
who progressed to advanced renal insufficiency had hypertension.
Proteinuria and Glomerulosclerosis

Several studies have documented the association of pro teinuria due to glomerulosclerosis with reflux nephropathy.
In patients with glomerulosclerosis, progressive renal insufficiency develops even if VUR is corrected or urinary tract
infections or hypertension are absent. The characteristic
clinical feature of these patients is persistent proteinuria;
the pathologic correlate is focal and segmental glomerulosclerosis involving unscarred segments of the kidney.
In a long-term follow-up of adults with reflux nephropathy, 294 patients over 15 years of age were studied with
regard to renal function, hypertension, and proteinuria. At
presentation, (mean age 17 years), 8.5% had hypertension,
4.8% proteinuria, and 2% renal insufficiency. At follow-up
(mean age 34 years), 38% had hypertension, 31% proteinuria, and 24% renal insufficiency. Proteinuria and renal
insufficiency were significantly more frequent in those with
severe bilateral reflux nephropathy. In another study of
patients with VUR and an abnormal serum creatinine at
presentation, all 34 patients progressed to end-stage renal disease.
The degree of proteinuria correlates with the presence
and extent of glomerular lesions. In a study of 94 adult
patients with primary bilateral VUR, all patients with pro gressive renal insufficiency had significant proteinuria
(a 1.5 g/24 h) at the time of the initial evaluation, and
there was a significant negative correlation between the
24-hour urine protein excretion and the simultaneous determination of renal function. It is now accepted that the
occurrence of proteinuria in patients with reflux nephropathy indicates the development of a focal segmental sclerosing glomerulopathy and the likelihood of progressive renal
insufficiency even in the absence of further scarring.
The pathogenesis of VUR related glomerulopathy is still
unclear. The most attractive explanation is that it results
from compensatory intrarenal hemodynamic changes and
single nephron hyperfiltration in response to the decrease

358

Jorge A. Velosa

in renal mass caused by scarring. Although this hypothesis

is compatible with the clinical course of many patients with
reflux nephropathy, a small number of patients with only
mild scarring may have significant proteinuria and glomerulosclerosis. Moreover, all patients with severe scarring develop proteinuria and progressive renal insufficiency. It is
possible that other factors play a role in the pathogenesis
of the glomerulopathy associated with reflux nephropathy.
CLINICAL MANIFESTATIONS
Diagnosis and Evaluation
Urinary tract infections are the most common manifestations of VUR and reflux nephropathy in infants and children. Current practice requires that children with urinary
tract infection be evaluated for VUR after the first infection
has been treated successfully. However, the type of studies
to be performed and the age limits for follow-up remain
controversial. Prospective studies may settle these important issues.
Several techniques have been used to demonstrate the
presence of VUR: voiding cystourethrography, radionuclide cystography, excretory urography, and cystoscopy.
Voiding cystourethrography is the most precise method
available for demonstrating VUR; the use of low-dose fluoroscopic technique decreases the radiation exposure. Radionuclide voiding cystography is gaining wide acceptance
for assessment and management of patients with VUR and
may be more sensitive than a standard contrast voiding
cystourethrogram. Imp ortant limitations of this technique
are the inability to distinguish accurately the grade of reflux
and to evaluate the urethra at voiding. Cystoscopic evaluation of the bladder and urethra is used less often than in
the past.
Investigations of the upper urinary tract to demonstrate
reflux nephropathy include intravenous urography and radionuclide scanning; ultrasonography also has been used.
Intravenous urography with nephrotomography is the traditional imaging technique for demonstrating reflux nephropathy because it gives the best anatomic detail. The
radiologic hallmarks of reflux nephropathy are the irregularly scarred surface of the kidney and the clubbing of
the underlying calyx, indicating full-thickness scarring of a
renal lobe, most frequently seen in the polar regions (Fig.
1). This is the most common form of reflux nephropathy.
In patients wtih the most severe reflux nephropathy, the
renal damage is diffuse, and there is marked and generalized reduction of renal mass, with uniform papillary change.
The main limitations of intravenous urography are its failure to demonstrate focal scarring in some patients (particularly at the upper poles) and its inability to provide
information regarding renal function. For these reasons,
technetium-99 is now widely used as a complementary investigation (Fig. 2). Technetium-99 scanning [technetium99 dimer-captosuccinic acid (DMSA)] is the most sensitive
method for estimating early renal damage. Reflux nephropathy can also be detected by ultrasonography, but this

FIGURE I The midcortex of the left kidney shows a scar interrupt ing the smooth outline of the kidney due to parenchymal loss.
The right kidney shows severe scarring with blunting and de formed calices.

method is operator dependent, has several limitations, and

does not provide information on renal function, either.
In summary, an intravenous urogram and voiding cystourethrogram or radionuclide cystography are indicated
in children younger than 5 years with urinary tract infection.
The initial evaluation in older children with urinary tract
infection includes intravenous urography or DMSA. In

FIGURE 2 DMSA scan demonstrates decreased size of the right

kidney with irregular contour of the upper pole consistent with
scarring. The left kidney has a normal contour. Calculated function:
80% on the left kidney and 20% on the right kidney.

55. Vesicoureteral Reflux and Reflux Nephropathy

adults with recurrent pyelonephritis or complicated urinary

tract infections, intravenous urography is the technique
of choice complemented with DMSA when information
regarding differential renal function is deemed necessary.
Preadolescent siblings of patients with VUR should have
a screening cystogram for VUR because 30 to 50% of this
population has demonstrated abnormalities.

359

the treatment was continued until puberty or the reflux

spontaneously disappeared. The option of screening for
bacteriuria rather than antibiotic prophylaxis seems reasonable in the setting of vigilant attention and prompt
access to medical care. Periodic urine cultures every 2 to
3 months are indicated.
Comparison of Conservative vs Surgical Treatment

MANAGEMENT OF VESICOURETERAL REFLUX

Despite many publications and several controlled trials,

either completed or under way, there is no universal agreement among nephrologists and urologists regarding management of patients with VU R. It is, however, generally
agreed that the main objective in children with VUR is
to prevent reflux nephropathy. To this end, urinary tract
infections must be recognized promptly, confirmed by culture, and treated aggressively and appropriately with antibiotics. Elimination of bacteriuria with treatment should
be documented by follow-up culture of the urine 48 to 72
hours after treatment initiation. Patients with VUR should
have annual voiding cystography by contrast or radionuclide study until the VUR disappears. Resolution of VUR
should be documented with two consecutive cystograms.
If the kidney imaging studies at the time of diagnosis of
VUR are normal, but the clinical picture is suspicious for
VUR, a second study after 2 years is recommended. Adolescents with no scars or hypertension who had previous
VUR are at minimal risk and require no further surveillance. As adult patients with reflux seldom form new scars,
surveillance imaging studies are not necessary. The majority of adult patients with VUR have multiple episodes of
urinary tract infections, although some have a history of a
single episode and some no history of urinary tract infections. In adult patients with multiple urinary tract infections, prophylactic therapy with trimethoprimsulfamethoxazole or trimethoprim alone is indicated. In young adults
with frequent episodes of acute pyelonephritis, antireflux
surgery is an acceptable alternative to many years of
chronic prophylactic therapy. Patients with hypertension,
proteinuria, or decreased renal function and history of
VUR need attention to these problems. It is possible that
progressive glomerular injury may be minimized by treatment of angiotensin converting enzyme inhibitors and perhaps protein restriction, but formal studies are not available. In women with VUR, the risks of hypertension and
urinary tract infection increase during pregnancy; they require renal function monitoring, control of the hypertension, and early treatment of urinary tract infections. Given
the familial nature of VUR, prenatal or neonatal ultrasonographic screening of the fetus for pelvic dilatation is recommended.
There is general agreement regarding the aggressive
treatment of infection in patients with VUR. Antibiotic
prophylaxis and the treatment of asymptomatic bacteriuria
are the current standards of care, but there is no general
agreement on this issue. The prospective controlled trials
used daily antimicrobial therapy with trimethoprimsulfamethoxazole, trimethoprim alone, or nitrofurantoin;

As a rule, surgery is of no benefit if the patient has

proteinuria, renal insufficiency, or hypertension. There is
still considerable disagreement concerning the indications
for antireflux surgery in very young children because in
many children refluxing will stop spontaneously and because once renal damage occurs, neither conservative nor
surgical treatment reliably prevents hypertension or deterioration of renal function.
Results from prospective controlled trials comparing
conservative and surgical treatment are now setting the
standard for management of VUR. The Birmingham
Reflux Study Group compared conservative management
with surgical correction of moderate and severe VUR
in patients younger than age 15 years. After 5 years of
follow-up, the incidence of urinary tract infections was
not different between the two groups: 50% of patients
treated conservatively had no VUR and 7% had new
renal scars compared with 5% new renal scars in the
surgically treated group. In the International Reflux Study
in Children, patients younger than age 11 years with
grades III and IV VUR were allocated randomly to
conservative and surgical treatment. After 5 years of
follow-up, new renal scars developed in 12% of children
treated medically compared to 13% of children treated
surgically. No significant difference in outcome was found
between conservative and surgical management in terms
of the development of new renal lesions or the progression
of established renal scars. These studies demonstrate that
conservative and surgical treatment yield comparable
results regarding prevention of further renal injury in patients with severe VUR. The Southwest Pediatric Nephrology Study Group investigated the evolution of milder
forms of VUR: grades I, II, and III. Children younger
than age 5 years and without evidence of renal scarring
were treated conservatively and followed for 5 years. In
an interim report, after 3 years of observation, fewer
than 10% of kidneys showed definitive scars, whereas
25% showed renal parenchymal thinning. Publication of
further follow-up data from these two important studies
should be available in the near future.
Studies of adults in our institution treated medically or
surgically showed outcomes similar to those in children.
A group of 27 adults with VUR treated medically was
compared with 67 adults with VUR treated surgically during the same interval. After 13 years of follow-up, renal
function was similar, irrespective of the method of management and appeared to be determined by the severity of
renal scarring, level of renal function at the initial evaluation, and presence of proteinuria. However, 93% of the
patients treated surgically experienced a decrease in the

360

Alan G. Wasserstein

rate of acute pyelonephritis. The presence or absence of

proteinuria is of prognostic value in adults with reflux nephropathy. Urinary protein levels of more than 1.5 g/24 h,
especially when associated with severe renal scarring and
depressed renal function, indicate progressive reflux nephropathy.
In summary, the treatment of choice for children with
stage I or II VUR is medical therapy. The treatment for
severe reflux may be either medical or surgical, and the
choice of treatment depends on the clinical circumstances.
Surgical correction of VUR may be indicated in the following clinical settings: (1) children or adults with recurrent
symptomatic pyelonephritis difficult to control medically,
(2) the presence of gross reflux and ureteral dilatation in
a young child, and (3) patient and family considerations,
in particular those that limit compliance with medical
treatment.
Bibliography
Arant BS Jr: Vesicoureteric reflux and renal injury. Am J Kidney
Dis 17:491-511, 1991. Arant BS: Medical management of mild
and moderate vesicoureteral reflux: Follow-up studies of infants and young children. A

preliminary report of the Southwest Pediatric Nephrology Study

Group. / Urol 148(Pt 2):1683-1687, 1992.
Birmingham Reflux Study Group. Prospective trial of operative versus non-operative treatment of severe vesicoureteric reflux in children: five years' observation. Br Med J 295:237-241,
1987.
Dick PT, Feldman W: Routine diagnostic imaging for childhood
urinary tract infect ions: A systematic overview. / Pediatr 128:
15-22, 1996.
Hodson CJ, Cotran RS: Vesicoureteral reflux, reflux nephropathy,
and chronic pyelonephritis. Contemp Issues Nephrol 10:83120, 1983.
Linshaw, M: Asymptomatic bacteriuria and vesicoureteral reflux
in children. Kidney Int 50:312-329, 1996.
Neves RJ, Torres VE, Malek RZ, et al.: Vesicoureteral reflux in
the adult. IV. Medical vs. surgical management. J Urol 132:882885, 1984.
Smellie JM, Tamminen-Mb'bius T, Oilbing H, et al.: Five-year
study of medical or surgical treatment in children with severe
reflux: radiological renal findings. The International Reflux
Study in Children. Pediatr Nephrol 6:223-230, 1992.
Torres VE, Velosa IA, Holley KE, et al.: The progression of
vesicoureteral reflux nephropathy. Ann Intern Med 92:776784, 1980.
Zhang Y, Bailey RR: A long term follow up of adults with reflux
nephropathy. NZ Med J 108:142-144, 1995.

NEPHROLITHIASIS
ALAN G. WASSERSTEIN

Nephrolithiasis has an annual incidence of 7 to 21 cases

per 10,000 persons and accounts for 7 to 10 of 1000 hospital
admissions in the United States. Men are affected about
four times as often as women. The lifetime prevalence of
kidney stones in white men is about 10% and inj excess of
25% in parts of the "stone belt" in the southeastern United
States. The predominant age of onset is the third to fifth
decade. Incidence has increased in the developed countries in the 20th century, probably as a consequence of
increased dietary animal protein intake. In underdeveloped
countries nephrolithiasis is relatively uncommon and bladder stones predominate.
The most common clinical consequence of nephrolithiasis is stone passage with acute renal colic. Flank pain radiating to the groin, abrupt in onset and extreme in severity,

Primer on Kidney Diseases, Second Edition

is typical. Gross or microscopic hematuria, dysuria or frequency, and nausea, vomiting, and occasionally ileus can
accompany stone passage. Stones that are associated with
nephrocalcinosis (primary hyperparathyroidism, renal tubular acidosis) or with staghorn formation (struvite, cystine) can cause renal failure, but the majority of stones
cause renal failure only if associated with bilateral urinary
obstruction or obstruction in a solitary kidney.
Classification and characteristics of renal calculi are given
in Table 1. Most stones are mixed calcium oxalate/calcium
phosphate or pure calcium oxalate and their pathogenesis is
idiopathic. Predominant calcium phosphate stones usually
reflect alkaline urine and have specific causes such as primary hyperparathyroidis m, renal tubular acidosis, alkali
therapy, or milk-alkali syndrome.
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

361

56. Nephrolithiasis

TABLE I

TABLE 2

Classification and Characteristics of Renal Calculi

Urinary Risk Factors for

Calcium Nephrolithiasis

Type
Calcium
oxalate
Uric acid

Struvite
Cystine

Frequency

Sex

75

10-15

M=F

15-20
1

F
M=F

Crystals

Radiography

EnvelopeRound,
radiodense
Diamond
Round or
staghorn,
radiolucent
Coffin-lidStaghorn,
radiodense
HexagonSlaghorn,
intermediate

Stone disease tends to recur. After a first calcium stone

40% of untreated patients have another stone within 5
years, and an additional 40% within the next 25 years.
Patients who lack renal calcifications on plain abdominal
film are less likely to have recurrent stone formation. Stone
formers show decreased stone formation and decreased
need for urological procedures after they begin medical
follow-up (the "stone clinic effect"); this observation may
be due to regression to the mean or to increased fluid
intake (see below).
PATHOGENESIS

Stone formation can be attributed to increased urinary

concentration of crystalloids, decreased inhibitors, or increased promoter substances (Table 2). Stone formation
begins with nucleation, the association of small amounts
of crystalloid to form submicroscopic particles. Nucleation
generally occurs on existing surfaces (heterogeneous nucleation), such as papillary epithelium. Urine must be saturatedthe concentration product of constituent crystalloids must exceed the solubility productto permit such
association. Greater supersaturation (higher concentrations of calcium and/or oxalate) favors nucleation. Submicroscopic particles would ordinarily wash out in the urine;
but they either are attached to the urothelial surface of the
papilla, or associate rapidly to form larger particles too
large to wash away (aggregation). Stone growth occurs by
aggregation, or by crystal growth, the orderly movement
of ions out of solution onto growing crystal.
Urinary saturation with respect to calcium oxalate is
frequent in the general population. Hence the role of inhibitors must be crucial. Citrate and magnesium form soluble
complexes with calcium and oxalate, respectively. These
complexes do not participate in nucleation. Nephrocalcin
is a highly acidic protein of renal tubular origin containing
the amino acid y-carboxyglutamic acid. It binds to the crystal surface and inhibits nucleation, aggregation, and crystal
growth; some calcium stone formers have abnormal
nephrocalcin that lacks y-carboxyglutamic acid. TammHorsfaU protein inhibits only aggregation; some calcium

Crystalloid Concentration
Hypercalciuria
Hyperoxaluria Low urine
volume
Promoters
Hyperuricosuria
Alkaline pH
Inhibitor Deficiency
Hypocitraturia
Hypomagnesuria
Macromolecules
Nephrocalcin
Uropontin
Tamm-Horsfall protein

stone formers have abnormal Tamm-Horsfall protein that

self-aggregates and loses inhibitory activity. These large
inhibitor proteins may have a major if not predominant
role in stone formation, but they are not as yet measured
in clinical practice.
Not all stone disease can be attributed to metabolic risk
factors. Several conditions predispose to stone formation
because of anatomical derangements, including polycystic
kidney disease, horseshoe kidney, and medullary sponge
kidney. Medullary sponge kidney is a congenital condition
in which the collecting ducts undergo cystic dilatation, predisposing to calcium stones and recurrent urinary tract
infection (but not renal insufficiency). Its characteristic urographic appearance is papillary striation or brush effect.
Medullary nephrocalcinosis may be extensive.
METABOLIC EVALUATION OF NEPHROLITHIASIS

Two kinds of stone activity should be distinguished:

anatomic activity, the movement of an existing stone
(usually into the pelvis or ureter) to cause symptoms;
and metabolic activity, the formation of a new stone,
growth of an existing stone, or formation and passage
of multiple tiny stone fragments ("gravel"). Much as in
the evaluation of a pulmonary nodule by comparison
with prior chest radiographs, serial abdominal films are
needed to determine metabolic activity. Figure 6 in Chapter 6 shows an anatomically active stone. It is highly
desirable that stones be recovered and analyzed: The
discovery of uric acid, struvite, pure calcium phosphate,
or cystine can transform the diagnostic approach. Urine
should be strained during acute stone passage and after
extracorporeal shock wave lithotripsy (ESWL). A first
stone requires limited evaluation beyond stone analysis:
urinalysis, serum calcium, phosphate, uric acid, creatinine,
and blood urea nitrogen. Recurrent stone formers should
have, in addition, 24-hour urine collections for measurement of calcium, phosphate, oxalate, citrate, uric acid,

362

Alan G. Wasserstein

creatinine, volume, pH, urea nitrogen, and sodium, preferably on several occasions. These collections are done
during usual dietary intake; medications that could interfere with urinary determinations (e.g., NSAIDs, diuretics,
calcium supplements, antacids) should be avoided if possible. Analysis of a diet diary for animal protein, calcium,
and oxalate is also useful. The metabolic evaluation and
treatment of calcium and uric acid stones is summarized
in Table 3.
RISK FACTORS FOR CALCIUM NEPHROLITHIASIS

Calcium oxalate stones are the consequence of a wide

variety of urinary abnormalities, including low urinary volume, hypercalciuria, hyperoxaluria, hypocitraturia, and
hyperuricosuria; usually more than one is present in an
individual case.
Low Urinary Volume

Some patients have habitually low fluid intake. Controlled studies have not generally detected low urinary
volumes in stone formers, possibly because stone formation and/or hypercalciuria impair urinary concentrating
ability. Epidemiological studies suggest that calcium stone
formation rises sharply when urinary volume falls below
1100 ml daily. Saturation of urine with regard to relevant
crystalloids falls as water is added to urine, whether in
vitro or by increased oral intake. Universal advice to
increase fluid intake may account for the so-called stone
clinic effect. Patients who continued to form stones in
a stone clinic setting were those who, retrospectively,
increased urine output by an average of 50 mL daily;
those who formed no further stones were those who

increased urine output by 500 mL daily. Stone formers

should strive to excrete at least 2 L of urine daily. A
useful prescription is two eight-ounce glasses of water
every four hours by the clock while awake; dilute apple
juice may be substituted, but high oxalate beverages such
as tea, colas, and citrus juices should not constitute the
bulk of ingested fluid.
Hypercalciuria

Hypercalciuria is defined as daily excretion of more than

300 mg in men or 250 mg in women or 4 mg/kg in either
sex. Distribution of calcium excretion is not normal, and
10% of the stone-free general population has excretion
above these limits. Desirable calcium excretion (to limit
stone formation) is probably significantly lower, in the
range of 200 mg daily.
Hypercalciuria, usually idiopathic, occurs in about 60%
of calcium stone formers. A small proportion (5%) has
primary hyperparathyroidism. Renal tubular acidosis, sarcoidosis, and various familial hypercalciuric syndromes are
rare causes of hypercalciuria. Primary hyperparathyroidism
affects predominantly middle-aged and older women. Hypercalciuria is due to excess renal production of 1,25dihydroxyvitamin D under the influence of parathyroid
hormone. The degree of hypercalcemia is usually mild to
moderate. So-called subtle or normocalcemic hyperparathyroidism occurs when serum calcium fluctuates into the
high-normal range; diagnosis depends on repeated blood
determinations of calcium and parathyroid hormone
(PTH), documenting simultaneous elevation. Some "normocalcemic" hyperparathyroidism is due to an inappropriate normal range: The true upper limit of normal serum
calcium in women (10.1 mg/dL) is lower than that in men.

TABLE 3
Metabolic Evaluation and Treatment of Nephrolithiasis
Abnormality
Calcium stones
Hypercalciuria
Hypercalcemia

Evaluation
Treatment
Measure serum and urine calcium, urine oxalate, citrate, uric acid, magnesium, creatinine, and volume on selfselected diet
Urine sodium and urea nitrogen
PTH and ionized serum calcium

Hyperoxaluria

Assess dietary oxalate and calcium, vitamin C, artificial

sweetener (xylitol), ileal disease or bypass

Hypocitratria

Acid load (if severe)

Serum K, creatinine, urine culture

Hyperuricosuria
Low urine volume
Uric acid stones
Acid urine
Hyperuricosuria

Dietary salt and protein restriction, thiazides,

neutral phosphate
Parathyroidectomy
Dietary oxalate restriction, magnesium
supplements, pyridoxine, calcium
supplements (enteric)
Alkali (potassium citrate)

Assess dietary purine intake

Dietary purine restriction, allopurinol

Measure 24-hour urine volume

Scheduled fluid intake 16 oz. q 4 hr while

awake

Measure urinary uric acid an d postprandial urine pH

R/o chronic diarrhea or gout Assess dietary

Alkali (potassium citrate)

purine

Dietary purine restriction, allopurinol

56. Nephrolithiasis

Some patients with normal total serum calciu m may have

elevations of the ionized fraction. In general, the diagnosis
of primary hyperparathyroidism need not be considered
unless the serum calcium is high (or high-normal); on the
other hand, stone formers with hypercalcemia are overwhelmingly likely to have primary hyperparathyroidism.
Presence of nephrocalcinosis and stone composition exclusively of calcium phosphate should also heighten suspicion
of primary hyperparathyroidism. Parathyroid adenoma is
responsible for primary hyperparathyroidism in 85% of
cases, parathyroid hyperplasia for the remainder. Parathyroid exploration is indicated in all stone formers with primary hyperparathyroidism, regardless of the level of the
serum calcium.
Idiopathic hypercalciuria is predominantly a disease of
men in the third to fifth decade. It is associated with
obesity, hypertension, and possibly affluence, and it is
familial. Intestinal calcium absorption is high, PTH levels
normal or low, and fasting urine calcium excretion elevated in 40%. Increased serum 1,25-dihydroxyvitamin D
levels could explain these findings but are not invariably
present. Increased nonparathyroid hormone-mediated
bone resorption could explain low-normal PTH levels
and high fasting calcium excretion, as well as reduced
bone mineral content and the tendency to negative calcium balance on low-calcium diet in idiopathic hypercalciuria; but bone resorption does not account for high
intestinal calcium absorption or high serum 1,25-dihydroxyvitamin D levels. A group of patients has been described
in which osteoporosis and hypercalciuria were apparently
due to IL-1-mediated bone resorption. Phosphate depletion could account for elevated 1,25-dihydroxyvitamin D
production and consequent abnormalities including low
PTH and high fasting calcium excretion, but hypophosphatemia is found only in a fraction of patients. In clinical
practice there is not a pressing need to delineate the
pathogenesis of idiopathic hypercalciuria. Pak has suggested that absorptive hypercalciuria (type 2) is corrected
by low-calcium diet alone and that reduced bone mineral
density is not found in these patients; but concern about
future development of low bone density should temper
use of low-calcium diet even in this group.
There are several dietary factors that exacerbate idiopathic hypercalciuria. The most important of these is
dietary sodium, which is easily assessed by measuring
24-hr urinary sodium excretion. Dietary sodium increases
renal calcium excretion; proximal tubular sodium reabsorption is reduced, as is the reabsorption of moieties
handled in parallel with sodium, such as calcium. Another
factor is dietary protein, which can be assessed by diet
diary or by urine urea nitrogen. Dietary animal protein
exacerbates calciuria by multiple mechanisms: acid load
increases urine calcium excretion; protein increases glomerular filtration rate and filtered load of calcium; sulfate
in dietary protein binds to calcium and inhibits tubular
reabsorption; and hormonal factors stimulated by dietary
protein, such as insulin, glucagon, and prostaglandins,
are calciuric. Dietary animal protein has been convincingly
associated with nephrolithiasis in epidemiological studies,

363
but dietary protein restriction has not been proven to
ameliorate stone formation in prospective trials. A third
factor is dietary calcium. In normal persons, only 6% of
dietary calcium appears in the urine; hypercalciuria from
dietary calcium excess alone is therefore unusual. In
patients with idiopathic hypercalciuria the proportion of
ingested calcium that is absorbed in the intestine is higher,
but intestinal oxalate absorption in such patients rises
disproportionately during dietary calcium restriction.
Since urinary oxalate may be at least as important as
urinary calcium in the genesis of nephrolithiasis, dietary
calcium restriction may be deleterious even in patients
with hypercalciuria. In a large prospective study of patients without nephrolithiasis, the risk of developing kidney stones was inversely related to dietary calcium intake,
perhaps because of the inverse relation between dietary
calcium and urinary oxalate. Regardless of the presence
of hypercalciuria, patients with nephrolithiasis should be
counseled to maintain at least a moderate calcium intake
(two to three servings daily) because of the twin risks
of increased urinary oxalate and loss of bone mineral.
An even higher level of dietary calcium may be desirable,
but optimum calcium intake in calcium stone-formers
has not been defined. Refined carbohydrates and possibly
alcohol are also calciuric, and high dietary fiber is hypocalciuric. Dietary measures should be tried before drug
therapy in the treatment of idiopathic hypercalciuria:
moderate sodium (100 to 125 mEq), moderate protein
(60 to 70 g/day), moderate calcium (two dairy servings
daily), high fiber, and low refined carbohydrates.
If metabolic activity is present despite high fluid intake
and dietary modification, drug treatment is indicated. Thiazides work by two mechanisms: first, by causing ECF volume depletion and thereby increasing proximal tubular
calcium absorption; second, by stimulating calcium reabsorption directly in the early distal convoluted tubule. A
low-salt diet is essential in order that ECF volume contraction be maintained. Several recent double-blind trials have
confirmed the efficacy of thiazides in reducing stone formation. Tachyphylaxis to thiazide treatment may occur after
several years of successful treatment.
Another treatment that inhibits calciuria and stone formation is neutral orthophosphate. It works at multiple sites
to inhibit calciuria and stone formation: reduced intestinal
calcium absorption, increased tubular calcium absorption,
reduced bone resorption, and increased urinary levels of
pyrophospate, an inhibitor of calcium phosphate nucleation. A potassium preparation of neutral or alkaline pH
is preferred, as acid or sodium increase calciuria. Phosphate
should be given with meals, as it blunts protein-induced
calciuria. Orthophosphate can cause diarrhea, and it is contraindicated in renal insufficiency and infection stones.
Compared to thiazides it has fewer metabolic side effects,
but it has not yet been proven efficacious in a controlled
prospective study. Cellulose phosphate is a nonabsorbed
form of phosphate that binds calcium in the gut. Its use
should be avoided, as it increases urine oxalate excretion
and risks bone mineral loss.

Alan G. Wasserstein

364
Hyperoxaluria

Oxalate is an end product of metabolism that forms

a poorly soluble complex with calcium. Normal excretion
is 15 to 40 mg daily, the bulk derived from endogenous
synthesis. Physicochemical considerations suggest that a
small increment in urinary oxalate has a larger effect on
calcium oxalate saturation than a relatively greater increase in urinary calcium. Hyperoxa luria may be due to
increased synthesis or increased intestinal absorption.
Increased synthesis can be due to congenital enzyme
deficiencies (primary hyperoxaluria types I and II), which
produce extreme elevations of urinary oxalate, dense
nephrocalcinosis , and renal failure in early adulthood
(see Chapter 51). High vitamin C intake (2 g daily) in
susceptible persons can also increase oxalate production,
as can pyridoxine deficiency. Increased absorption can
be due to high oxalate diet, intestinal malabsorption
(enteric hyperoxaluria), or calcium restriction (or cellulose
phosphate) in patients with idiopathic hypercalciuria. Enteric hyperoxaluria occurs particularly in Crohn's disease
and ileal surgery; an intact colon is required. Three
mechanisms have been proposed: (1) intestinal calcium
is bound in soaps and unavailable for calcium oxalate
complex formation, so free oxalate is hyperabsorbed in
the colon; (2) poorly absorbed bile salts stimulate increased colonic transport of oxalate; and (3) poorly absorbed bile salts inactivate an oxalate-metabolizing bacterium (Oxalobacter formigenes) and thereby promote
oxalate absorption. Dietary calcium restriction in patients
with idiopathic hypercalciuria presumably enhances oxa luria by a similar mechanism, reducing luminal calcium
and calcium-oxalate complexation and thereby increasing
free oxalate absorption. Hyperoxaluria is treated by reducing dietary oxalate (tea, citrus juices, colas, spinach,
rhubarb, peanuts, and chocolate) and adding magnesium
supplements with meals (which bind intestinal oxalate).
Pyridoxine is sometimes effective in rare cases of oxalate
overproduction. Enteric oxaluria is treated with calcium
supplements, which also bind intestinal oxalate; urinary
calcium excretion is low and hypercalciuria is not a
concern.
Hypocitraturia

Normal citrate excretion is 300 to 900 mg daily. Citrate

acts as an inhibitor of nucleation by forming insoluble complexes with calcium. Urinary citrate is higher in premenopausal women than in men; it is also higher in pregnancy,
Ideologically a useful response as pregnancy markedly increases urinary calcium excretion. Idiopathic hypocitraturia is found in 10 to 40% of stone formers. Hypocitraturia
also may be due to any cause of intracellular acidosis,
including potassium deficiency, renal failure, distal (type
I) renal tubular acidosis (RTA), chronic diarrheal states
and malabsorption, or acetazolamide. In urinary tract infection, bacteria may metabolize citrate and reduce its urinary excretion.
Distal RTA causes a profound reduction of urinary citrate excretion, usually to less than 50 mg daily; it also

causes hypercalciuria and persistently alkaline urine, but

hypocitraturia is the most important contributor to stone
diathesis. Measurement of urinary citrate is a useful alternative to acid loading to establish the diagnosis of distal
RTA. Nephrocalcinosis and renal failure may occur.
The alkali requirement to correct hypocitraturia (4 mEq
kg'1 day" 1 or more) may be higher than that required to
correct systemic acidosis or hypokalemia (1 to 2 mEq
kg'1 day" 1). Hypocitraturia may be the relevant guide to
alkali dosage in these patients.
Potassium citrate is the treatment of choice of idiopathic
hypocitraturia and of distal RTA, either in sustained release capsules or in liquid form. Sodium citrate or bicarbonate are less desirable, as their sodium content tends to
increase calciuria. Thiazides reduce urinary citrate, presumably because of potassium depletion. Potassium citrate is
the potassium supplement of choice in calcium stone
formers.
Uric Acid

Hyperuricosuria is a risk factor for calcium stone formation, probably because uric acid acts as a surface for hetero geneous nucleation. Dietary purine restriction should suffice to ameliorate hyperuricosuria, which is usually due to
high purine consumption. Allopurinol has been shown to
reduce calcium stone formation in patients with isolated
hyperuricosuria.
OTHER TYPES OF NEPHROLITHIASIS

Uric acid stones occur in two circumstances: excessive

urinary uric acid and persistently acid urine. Only 20% of
patients with uric acid stones have hyperuricosuria. Excessive urinary uric acid is due, with rare exceptions, to dietary
purine overconsumption. Persistently acid urine, due to
deficient renal ammoniagenesis, is a feature of gout and of
idiopathic uric acid stones and of chronic diarrheal disease,
because of loss of alkali equivalent in the stool. Uric acid
is 10 to 20 times more soluble at pH 7 than at pH 5. Hence
urine alkalinization is an effective treatment for most patients with uric acid stones; indeed it is more effective than
allopurinol, which generally reduces uric acid excretion by
about half. Target urine pH is 6.5 to 7.0, monitored by the
patient with nitrazine paper; higher urine pH should be
avoided because of the risk of calcium phosphate stones.
Allopurinol can be added if uric acid production is exces sive (over 1200 mg daily) and if dietary purine restriction
is unsuccessful. Unlike calcium stones, uric acid stones dis solve during alkali therapy (if they are not secondarily
calcified).
Infection stones are the most severe form of nephrolithiasis: they can cause progressive renal failure, urosepsis or
perinephric abscess, intractable urinary tract infection,
pain, and bleeding. Infection stones consist of struvite
(magnesium ammonium phosphate) and apatite (calcium
phosphate). They form only during urinary infection with
urease-positive bacteria, usually Proteus or Providencia

365

56. Nephrolithiasis

species, almost never Escherichia coll These bacteria

cleave urea to ammonia, elevating the urine pH to 8 or
more and favoring precipitation of struvite and apatite.
Bacterial infection is difficult to clear, as the organisms are
inaccessible in stone interstices.
Approximately 40% of patients with struvite stones have
an underlying metabolic abnormality; this abnormality (e.g.,
primary hyperparathyroidism) is presumed to be the original cause of nephrolithiasis. Secondary infection then causes
formation of struvite stones. This pathogenesis is more likely
if patients have had stone passage; patients with primary
struvite stone formation and no other abnormality have
large staghorn calculi and usually do not pass stones.
Treatment consists first of antibiotics, usually penicillin
or ampicillin. Sterile urine is achieved in only about 20% of
patients. Because of the large stone burden, stone removal
usually requires combined percutaneous and extracorporeal lithotripsy. Infusion of acidifying solutions directly into
the kidney can help to dissolve stones and clear fragments.
If stone removal is complete, cure may be hoped for. If not,
residual fragments are a nidus for intractable infection and
recurrent stone formation, and antibiotic suppression is continued indefinitely. The urease inhibitor acetohydroxamic
acid can help to prevent stone formation if urinary tract infection persists. However, it has been associated with hemo lytic anemia, thrombophlebitis, and migraine.
Cystine stones result from an unusual inherited disorder
in which renal tubular reabsorption of cystine, ornithine, arginine, and lysine (COAL) is reduced. Cystine is the disulfide reduction product of cysteine. The solubility of cystine
is 250 to 300 mg/L. In patients with relatively low levels of
cystinuria, high urine volume (4 L/day) will be sufficient to
reduce the concentration of cystine below this threshold. It
is particularly important to continue high fluid intake
throughout the night, as there is disproportionate cystine
excretion at that time. Those with more severe cystinuria
require pharmacotherapy. Penicillamine or tiopronin react
with cystine to form a soluble mixed cysteine disulfide; they
also reduce cystine production. These agents are also indicated for stone dissolution.
UROLOGICAL MANAGEMENT OF STONES

Most symptomatic renal calculi (90%) pass spontaneously. The probability of spontaneous passage depends on
width, length, and location of stone. Ureteral stones 4 mm
or less in width are likely (more than 70%) to pass within
1 year; stones 8 mm or more in width are unlikely (less
than 10%) to pass. Length of stone is less critical than width
in determining rate of stone passage, presumably because
stones orient lengthwise as they pass down the ureter. For
example, a stone 6 mm in length has the same chance of
spontaneous passage (85%) as a stone 3 mm in width.
Stones that come to clinical attention in the lower ureter
are twice as likely to pass as upper ureteral stones. Stones
that are judged to be likely to pass and that are not associated with pain o r obstruction are managed expectantly for

as long as 1 year. Indications for urological intervention

for ureteral stones include obstruction, pain, fever, and
observed or anticipated failure of spontaneous passage. On
the basis of incomplete animal data, it is believed that
complete obstruction requires relief within 2 weeks and
partial obstruction within 4 to 6 weeks (depending on severity of hydronephrosis) to avoid permanent renal injury.
Fever requires emergency decompression of the obstructed
urinary tract with retrograde stent or percutaneous nephrostomy; definitive stone removal is deferred until infection
resolves. Stones in the renal pelvis or upper ureter are
treated with extracorporeal shock wave lithotripsy
(ESWL); stents are placed in cases where large stone burden increases the risk of obstruction by pulverized material
(Steinstrasse) after ESWL. Lower ureteral stones can also
be treated with ESWL but the rate of success is only 70
to 80%; ureteroscopy with basket retrieval or ultrasonic or
laser lithotripsy (for impacted stones) is successful in 98 to
99%. Recommendations for use of ESWL for asymptomatic
intrarenal calculi are conflicting. With calyceal stones we
reserve ESWL for the classic indications of pain, obstruction, or bleeding. However, a stone of 5 mm or more in
the renal pelvis generally should have ESWL to prevent
obstruction. ESWL requires that the distal urinary tract be
free of obstruction so that fragmented material can pass;
ESWL is contraindicated in obstruction, pregnancy, and
aortic or renal artery aneurysm. In pregnancy, an obstructing stone can be managed with a retrograde stent till after
delivery, when ESWL can safely be performed. The success
of ESWL begins to diminish as stones exceed 1 cm and is
poor with stones above 2 cm. In these cases, percutaneous
lithotripsy is combined with ESWL. Cystine stones are
relatively refractory to ESWL and percutaneous techniques may be necessary. ESWL and percutaneous techniques have dramatically reduced the need for open surgical treatment of nephrolithiasis; the latter accounts for less
that 5% of urological interventions. ESWL and percutaneous lithotripsy are not without complications. Loss of renal
function (particularly in the medulla where juxtamedullary
nephrons may be involved), bleeding and hematoma, and
urosepsis may occur even in the best centers. All patients
undergoing either procedure should be screened with a
urine culture and treated with appropriate antibiotics prior
to the procedure. Patients should also be counseled that
delayed hypertension may result, although this remains
controversial.
Bibliography
Coe FL, Parks JH, Asplin JR: Pathogenesis and treatment of
kidney stones. N Engl J Med 327:1141 -1152, 1992.
Consensus Conference: Prevention and treatment of kidn ey
stones. JAMA 260:977-981, 1988.
Curhan GC, Willett WC, Rimm EB, Stampfer MJ: A prospective
study of dietary calcium and other nutrients and the risk of
symptomatic kidney stones. N Engl J Med 328:833-838, 1993.
Goldfarb S: Dietary factors in the pathoge nesis and prophylaxis
of calcium nephrolithiasis. Kidney Int 34:544-555, 1987.

366

Walter E. Stamm

Hosking DH, Erickson SB, van den Berg CJ, Wilson DM, Smith
L: The stone clinic effect in patients with idiopathic calcium
urolithiasis. J Urol 130:1115-1158, 1983.
Lemann J: Composition of the diet and calcium kidney stones.
N Engl J Med 328:880-882, 1993.
Lerner SP, Malachy JG, Griffith DP: Infection stones. J Urol
141:753-758, 1989.
Mulley AG, Carlson KJ, for the Health and Public Policy Commit tee, American College of Physicians: Lithotripsy. Ann Intern
Med 103:626-629, 1985.
Nakagawa Y, Abram V, Kezdy FJ, Kaiser ET, Coe FL: Purification
and characterization of the principal inhibitor of calcium oxalate

monohydrate crystal growth in human urine. J Bio l Chem

256:3936-3944, 1981. Pahira JJ: Management of the patient
with cystinuria. Urol Clin
North Am 14:339-346, 1987. Pak CYC, Fuller C: Idiopathic
hypocitraturic calcium -oxalate
nephrolithiasis successfully treated with potassium citrate. Ann
Intern Med 104:33-37, 1986. Uribarri J, Oh MS, Carroll HJ:
The first kidney stone. 111:10061009, 1989.
Wasserstein AG: Kidney stones: Advice on preventing first episodes and recurrences. Consultant 26:81-104, 1986.

URINARY TRACT INFECTIONS

WALTER E. STAMM

Urinary tract infections can be classified as uncomplicated or complicated. Due to their very different etiology,
pathogenesis and management, uncomplicated and complicated urinary tract infections will be discussed separately.
UNCOMPLICATED URINARY TRA CT INFECTION
Definition

Uncomplicated urinary tract infections (UTI) occur in

patients without physiologic or anatomic abnormalities of
the urinary tract and in the absence of recent urological
surgery or instrumentation. They are community-acquired.
Laboratory evidence of infection typically includes bacteriuria and pyuria; in some cases, hematuria may be present.
Uncomplicated urinary tract infections can be further divided into acute cystitis (lower tract infection) and acute
pyelonephritis (upper tract infection).
Etiology and Incidence

Almost all acute uncomplicated urinary tract infections

occur in women, especially between the ages of 18 and
40. Infections occasionally occur before menarche but the
incidence increases markedly in late adolescence and during the second and third decades of life. In the United
States, an estimated 6 to 7 million cases of acute cystitis
involve young women in this age range annually. At least

Primer on Kidney Diseases, Second Edition

250,000 cases of acute pyelonephritis occur annually. Uncomplicated infections of the urinary tract are rare in males.
Recent studies suggest that uncircumcised infants and adult
homosexual males may be at increased risk. In young males,
uncomplicated urinary tract infection may occasionally
spread from the bladder to the prostate (acute prostatitis)
or the epididymis (acute epididymitis), but such cases are
also infrequent.
Uncomplicated urinary tract infections are caused by a
very narrow spectrum of microbes. Eighty percent are due
to Escherichia coli. Staphylococcus saprophyticus causes 10
to 15% of uncomplicated urinary tract infections in young
women, especially during the summer and fall months.
Proteus mirabilis and Klebsiella are occasional causes (2
to 5%).
Pathogenesis

Uncomplicated urinary tract infections in women are

caused by fecal E. coli that migrate from the perianal skin
to colonize the vaginal introitus and urethra. The organisms
then ascend through the urethra into the bladder and, in
cases of acute uncomplicated pyelonephritis, from the bladder to the kidney via the ureters. Both bacterial and host
factors are important in predisposing to acute uncomplicated urinary tract infection. Sexual intercourse appears to
facilitate migration of bacteria from the periurethral zone
into the bladder. Diaphragm-spermicide use and spermi Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

57. Urinary Tract Infections

cide use alone have also been linked to an increased risk,

probably via spermicide-mediated alterations in the vaginal
flora that result in overgrowth with E. coll Some women
may be predisposed to recurrent urinary tract infections
because their uroepithelial cells have increased number
of receptors for E. coli, a trait that may be genetically
determined. Behavioral factors such as not voiding after
intercourse or postponing voiding also appear to increase
risk of infection. Uropathogenic E. coli (particularly those
E. coli strains that cause urinary tract infection) appear to
possess specific virulence determinants such as fimbriae or
production of hemolysin or aerobactin that enable them
to colonize and infect the urinary tract more efficiently
than strains lacking these virulence factors.

Clinical Manifestations
Typical symptoms of acute uncomplicated cystitis include
dysuria, frequency, urgency, voiding of small urine volumes, and suprapubic or lower abdominal pain. Up to onethird of patients may develop gross hematuria, and many
patients describe foul-smelling or cloudy urine. On examination, suprapubic tenderness may be present. Fever and
costovertebral angle tenderness are generally absent.
Acute pyelonephritis characteristically presents with localized flank, low-back, or abdominal pain accompanied
by fever, chills, sweats, headache, nausea, vomiting, and
malaise. Symptoms of lower tract infection may or may
not be present. On examination, fever and flank tenderness
are present. The severity of illness associated with acute
pyelonephritis ranges from very mild to quite severe, including gram-negative septicemia and necrotizing intrarenal or perinephric abscesses. Approximately 20% of patients with acute uncomplicated pyelonephritis have positive blood cultures.
Differential Diagnosis

Acute bacterial cystitis must be distinguished from other

common genital infections that produce dysuria, most commonly, vulvovaginitis due to yeast, trichomonas, or bacterial vaginosis, or sexually transmitted infections that involve both the urethra and cervix (Chlamydia trachomatis,
Neisseria gonorrhoeae, or herpes simplex virus infection).
Typically, bacterial cystitis is sudden in onset and produces
multiple urinary symptoms (dysuria, urgency, frequency,
and gross hematuria), while urethritis due to sexually transmitted pathogens causes milder symptoms, is more gradual
in onset, and is often characterized by dysuria without other
urinary symptoms. Signs of cervicitis may also be present
in these women. Although dysuria may be the presenting
complaint, on further questioning most women with vaginitis have other symptoms suggesting that diagnosis, including vaginal discharge, vaginal odor, or dyspareunia. The
presence of hematuria (gross or microscopic) strongly suggests bacterial cystitis since women with vaginit is or urethritis caused by STD agents seldom have hematuria.
Acute pyelonephritis must be differentiated from other
intraabdominal conditions seen in young women, most im-

367
portantly pelvic inflammatory disease, appendicitis, ectopic
pregnancy, or ruptured ovarian cyst.
Diagnosis

The suspected diagnosis of acute uncomplicated lower

urinary tract infection can be readily confirmed by microscopic examination of the urine and by urine culture. Essentially all women with acute cystitis have associated pyuria
if a careful microscopic examination of unspun urine is
performed. Additionally, up to 50% may have microscopic
hematuria. The leukocyte esterase dipstick can also be used
to rapidly screen a urine specimen for pyuria but is less
sensitive than a microscopic exam. Culture of midstream
urine will generally demonstrate the etiologic bacterial
agent in concentrations >105 cfu/mL. However, in about
one-third of cases, colony counts will range from 10 2 to
104 colony-forming units per milliliter. The Gram stain of
unspun urine will demonstrate the etiologic agent in cases
with alO5 cfu/mL of bacteria, but will be negative in cases
with lower bacterial counts. Thus, a negative urine Gram
stain does not rule out acute bacterial cystitis in those cases
characterized by low colony counts.
In acute pyelonephritis, urinalysis nearly always demonstrates pyuria, and in 80% of cases, bacteriuria can be seen
on the Gram stain of unspun urine. About 20% of cases
have colony counts <105/mL and thus negative Gram
stains. Urine culture will almost always confirm the presence of the infecting organism, and blood cultures are positive in 20% of patients. The presence of leukocyte casts
ina freshly voided specimen of urine strongly supports the
diagnosis, but casts are demonstrable in only 20 to 40%
of cases.
Treatment

Women with acute uncomplicated cystitis should

generally be treated with 3-day course of trimethoprim,
trimethoprim-sulfamethoxazole, or a fluoroquinolone such
as norfloxacin, ciprofloxacin, ofloxacin, lomefloxacin, or
enoxacin (Table 1). Single-dose therapy results in higher
recurrence rates and is thus not recommended. The frequent occurrence of in vitro resistance to ampicillin, amoxicillin, and first-generation cephalosporins among E. coli
strains causing uncomplicated infections, plus the propensity of these agents to foster recurrences, argues against
their routine use. These agents, however, are the agents of
choice for treatment of cystitis during pregnancy, and
should generally be given for 7 days. Patients who recur
after an initial 3-day course of therapy should be treated
with a 7-day course with one of the above antimicrobials.
Women with multiple recent infections, with symptoms
longer than 7 days, who are pregnant, or who have other
complicating factors should also be treated initially with a
7-day regimen.
Patients with acute uncomplicated pyelonephritis can be
safely treated in the outpatient setting if they have no
nausea or vomiting, no evidence of septicemia, are reliable,
are not pregnant, and have a clear-cut diagnosis. Fourteen

368

Walter E. Stamm
TABLE I

Treatment Regimens for Bacterial Urinary Tract Infections

Condition

Characteristic
pathogens

Acute uncomplicated E. coli, S. saprophyticus, P.

cystitis in women mirabilis, Klebsiella
pneumoniae

Acute uncomplicated
pyelonephritis in
women

E. coli, P. mirabilis,
K. pneumoniae,
S. saprophyticus

Mitigating circumstances
None

3-day regimens: oral trimethoprim -sulfamethoxazole,

trimethoprim, norfloxacin, ciprofloxacin, oftoxacin,
lomefloxacin, or enoxacin*

Diabetes, symptoms for >7

days, recent urinary tract
infection, use of
diaphragm, age >65 yr
Pregnancy

Consider 7-day regimen: oral trimethoprim sulfamethoxazole, trimethoprim, norfloxacin,

ciprofloxacin, ofloxacin, lomefloxacin or enoxacin 6

Mild-to-moderate illness, no
nausea or
vomitingoutpatient
therapy Severe illness or
possible
urosepsishospitalization
required

Oral' trimethoprim -sulfamethoxazole, norfloxacin,

ciprofloxacin, ofloxacin, lomefloxacin, or enoxacin
for 10-14 days

Pregnancyhospitalization
recommended

Complicated urinary
tract infection

E. coli, proteus species,

klebsiella species,
pseudomonas species,
serratia species,
enterococci,
staphylococci

Recommended empirical treatment"

Mild-to-moderate illness, no
nausea or
vomitingoutpatient
therapy Severe illness or
possible
urosepsishospitalization
required

Consider 7-day regimen: oral amoxicillin,

macrocrystalline nitrofurantoin, cefpodoxime
proxetil, or trimethoprim -sulfamethoxazole6

Parenteral"* trimethoprim -sulfamethoxazole,

ceftriaxone, ciprofloxacin, ofloxacin, or gentamicin
(with or without ampicillin) until fever gone; then
oral' trimethoprim -sulfamethoxazole, norfloxacin,
ciprofloxacin, ofloxacin, lomefloxacin, or enoxacin
for 14 days
Parenteral^ ceftriaxone, gentamicin (with or without
ampicillin), aztreonam or trimethoprim sulfamethoxazole until fever gone; then oral'
amoxicillin, a cephalosporin, or trimethoprim sulfamethoxazole for 14 days
Oral' norfloxacin, ciprofloxacin, ofloxacin,
lomefloxacin, or enoxacin for 10-14 days

Parenteral'' ampicillin and gentamicin, ciprofloxacin,

ofloxacin, ceftriaxone, aztreonam, ticarcillin clavulanate, or imipenem-cilastatin until fever
gone; then oral' trimethoprim -sulfamethoxazole,
norfloxacin, ciprofloxacin, ofloxacin, lomefloxacin,
or enoxacin for 14-21 days

" Treatments listed are those to be prescribed before the etiologic agent is known (Gram's staining can be helpful); they can be modified
once the agent has been identified. The recommendations are the authors' and are limited to drugs currently approved by the Food and
Drug Administration, although not all the regimens listed are approved for these indications. Fluoroquinolones should not be used in
pregnancy. Trimethoprim -sulfamethoxazole, although not approved for use in pregnancy, has been widely used. Gentamicin should be
used with caution in pregnancy because of its possible toxicity to eighth-nerve development in the fetus.
b
Multiday oral regimens for cystitis are as follows: trimethoprim -sulfamethoxazole, 160-800 mg every 12 hr; trimethoprim, 100 mg
every 12 hr; norfloxacin, 400 mg every 12 hr; ciprofloxacin, 250 mg every 12 hr; ofloxacin, 200 mg every 12 hr; lomefloxacin, 400 mg every
day; enoxacin, 400 mg every 12 hr; macrocrystalline nitrofurantoin, 100 mg four times a day; amoxicillin, 250 mg every 8 hr; and cefpodoxime
proxetil, 100 mg every 12 hr.
c
Oral regimens for pyelonephritis and complicated urinary tract infection are as follows: trimethoprim -sulfamethoxazole, 160 -800 mg
every 12 hr; norfloxacin, 400 mg every 12 hr; ciprofloxacin, 500 mg every 12 hr, ofloxacin, 200-300 mg every 12 hr, lomefloxacin, 400 mg
every day; enoxacin, 400 mg every 12 hr, amoxicillin, 500 mg every 8 hr; and cefpodoxime proxetil, 200 mg every 12 hr.
d
Parenteral regimens are as follows: trimethoprim-sulfamethoxazole, 160-800 mg every 12 hr; ciprofloxacin, 200 -400 mg every 12 hr;
ofloxacin, 200-400 mg every 12 hr; gentamicin, 1 mg kg" 1 body"1 every 8 hr; ceftriaxone, 1-2 g every day; ampicillin, 1 g every 6 hours;
imipenem-cilastatin, 250-500 mg every 6-8 hr; ticarcillin -clavulanate, 3.2 g every 8 hr; and aztreonam, 1 g every 8 to 12 hr.

days of trimethoprim-sulfamethoxazole, trimethoprim

alone, or a fluoroquinolone are preferred regimens (Table
1). For inpatients, parenteral therapy should be used initially. Depending on the local antimicrobial sensitivity pat-

terns of uropathogens causing acute uncomplicated pyelonephritis, preferred regimens include trimethoprimsulfamethoxazole, gentamicin, a fluoroquinolone, or a
third-generation cephalosporin which should be given in-

369

57. Urinary Tract Infections

travenously until defervescence and signs of clinical improvement occur. In 80% of patients, improvement occurs
within 72 hours, at which time oral therapy can be provided
for the remainder of the treatment course. Extension of
the period of therapy is not required solely because a patient had positive blood cultures. Routine use of ultrasound
or CT scanning in hospitalized cases of acute uncomplicated pyelonephritis has not apparent benefit, but failure
to improve after 72 hours of appropriate antibiotic therapy
indicates the need to seek evidence of obstruction or an abscess.
Prognosis
Acute uncomplicated cystitis responds to effective initial
therapy in over 90% of cases. The remainder respond either
to a second treatment course or more prolonged therapy.
Twenty percent of women with acute uncomplicated cystitis
develop a pattern of frequent recurrences characterized by
episodes occurring as often as once per month or once every
other month. Episodes of cystitis, even when they recur frequently, do not appear to produce long-term sequelae.
Acute uncomplicated pyelonephritis also responds to antimicrobials in over 90% of cases. Failure to respond to initial
antimicrobial therapy should suggest a drug-resistant strain,
an anatomic abnormality, or urinary obstruction. Acute
pyelonephritis generally heals without significant renal scarring and without impairment of renal function. Some women
with acute pyelonephritis develop intranephric or perinephric abscesses that do not respond to antimicrobial therapy and require either percutaneous drainage or surgical intervention along with prolonged antibiotic administration.
Intranephric or perinephric abscesses should be suspected if
a pyelonephritis -like illness does not respond to appropriate
antimicrobial therapy or in patients with fever and unilateral
flank pain for 2 or more weeks. Other risk factors for abscess
formation include diabetes mellitus and nephrolithiasis. Renal ultrasound or a CT scan will establish the presence of
intrarenal and perinephric abscesses.
In women prone to frequent recurrent episodes of cystitis, low-dose antimicrobial prophylaxis reduces the incidence of recurrent infections to nearly zero and can be
used safely for long periods. Successfully used regimens
include daily or thrice weekly trimethoprim (100 mg),
trimethoprim-sulfamethoxazole (^ single-strength tablet
daily), or nitrofurantoin (100 mg daily). Women who can
temporally relate their recurrent infections to intercourse
can use these same regimens after intercourse rather than
on a regular basis. Voiding soon after intercourse also reduces the likelihood of recurrent urinary tract infections.
In diaphragm-spermicide users who develop recurrent urinary tract infections, use of an alternative contraceptive
may be warranted. In postmenopausal women with frequently recurring uncomplicated cystitis, topically applied
intravaginal estrogen may also be an effective means of prevention.
COMPLICATED URINARY TRACT INFECTIONS

Complicated urinary tract infections occur in patients

with obstruction, anatomic, or functional disorders of the

urinary tract. Instrumentation, calculi, and catheterization

also predispose to complicated infection, as does pregnancy
which induces hydroureter, bladder enlargement, and urinary stasis (Table 2).
Etiology and Incidence
Complicated urinary tract infections occur most frequently at the extremes of life. Urinary infections in newborns or infants often indicate an underlying anatomic abnormality in the urinary tract. While such infections occur
in fewer than 2% of infants, their occurrence should prompt
the use of appropriate urological studies to seek an underlying anatomic abnormality. The vast majority of complicated
urinary tract infections occur in older men and women. In
men, the occurrence of urinary tract infections is less than
1% until the fifth to sixth decade, at which time prostatic
hypertrophy and obstruction occur, leading to urological
instrumentation. With aging, hospitalization for other illnesses often leads to urinary catheterization and acquisition
of nosocomial urinary tract infections. Over 10% of hospitalized patients have urinary catheters placed, and approximately 10 to 15% of them develop subsequent bacteriuria.
Catheter-associated urinary tract infections account for
40% of all nosocomial infections, over one million infections annually. In addition, urosepsis secondary to catheterassociated bacteriuria is the most common source of gramnegative bacteremia in hospitalized patients.
The spectrum of bacterial species causing complicated
and hospital-acquired urinary tract infections is much
broader than that seen in community-acquired infections.
While E. coli remains the most common organism, many
other gram-negative bacilli cause complicated urinary infections, including Proteus mirabilis, other proteus species,
providencia, serratia, klebsiella, and pseudomonas. These
organisms generally exhibit more antimicrobial resistance
than those causing uncomplicated urinary infection. In addition, enterococci and pseudomonas are far more frequent
in this setting than in uncomplicated urinary tract infections, and Staphyloccus aureus causes complicated infections as well. Increasingly, yeast (candida) infections are

TABLE 2

Factors Associated with Complicated Urinary

Tract Infections
Indwelling or intermittent urinary catheter

Urinary instrumentation or urological surgery

Postvoid residual urine (>100 mL)
Urinary obstruc tion

Vesicoureteral reflux
Urologic anatomic abnormality
Azotemia, renal dysfunction
Renal transplantation
Neurogenic bladder
Pregnancy
Male patient
Nephrolithiasis
Diabetes

370

Walter E. Stamm

being seen in catheter-associated infections, in diabetics,

and in immunosuppressed patients.
Pathogenesis
As in uncomplicated infections, most bacteria causing
complicated UTI invade the urinary tract in an ascending
manner. Rarely, the kidney is infected secondary to bacteremia. Obstruction to urine flow or other functional abnormalities of the urinary tract impair the normally efficient
ability of the bladder to eliminate bacteria. Instrumentation
of the urinary tract, particularly catheterization, provides
a portal of entry for bacteria and also makes it more difficult
to eradicate bacteriuria with antimicrobial agents. Vesicoureteral reflux may facilitate the ease with which bladder
bacteria can ascend to the kidney, producing pyelonephritis. Interestingly, most of the bacteria causing complicated
urinary infection lack the specific urovirulence determi nants characteristic of strains causing uncomplicated UTI,
suggesting that their pathogenicity relates mainly to host
impairment.

Since by definition, complicated UTIs occur in patients

with abnormal urinary tract anatomy or function, determining which patients should undergo further study is a key
issue. The yield of anatomic evaluation in women with no
risk factors is low and anatomic evaluation in this population should be reserved for women with frequent recurrences or slow response to treatment. Men are much more
likely to have anatomic defects, in particular bladder outlet
obstruction from prostate enlargement, and appropriate
evaluation is in order after a first UTI. Patients presenting
with hypotension or other features of sepsis should also be
evaluated since prompt drainage of obstruction is essential.
Evaluation for reflux or other anatomic defects should be
strongly considered in children, particularly those younger
than 5 years of age, with a first or second UTI.
Depending on the clinical circumstance, computed tomography scanning, ultrasound, or intravenous pyelography may be useful in demonstrating the nature of the uro logic abnormality or obstruction. These procedures may
also be useful in identifying focal infections such as a perinephric or intranephric abscess or nephrolithiasis.
Treatment

Clinical Manifestations
The clinical manifestations produced by complicated
urinary tract infection range widely from asymptomatic
bacteriuria to gram-negative septicemia. Most catheterassociated urinary tract infections, for example, remain
asymptomatic and do not produce clinical manifestations.
Similarly, asymptomatic bacteriuria without an associated
catheter is common in elderly patients. In some patients,
complicated urinary tract infections will present with either
the syndrome of acute cystitis or acute pyelonephritis as
outlined above. The hallmark of such episodes in the complicated setting is their lesser responsiveness to antimicrobial therapy and their tendency to recur after treatment.
In hospitalized patients who develop fever, hypotension,
and signs of septicemia, the urinary tract should be suspected as a possible source despite the absence of urinary
symptoms, especially in patients with recent instrumentation or catheterization.
Differential Diagnosis
The considerations outlined for uncomplicated urinary
tract infections apply in the setting of complicated urinary
tract infections as well. In patients with septicemia, urosepsis must be differentiated from other potential sources of
bacteremia. The presence of recent instrumentation, catheterization, and evidence of bacteriuria and pyuria strongly
suggests urosepsis.
Diagnosis
In symptomatic patients, urinary microscopy and culture
should be used to confirm the presence of infection as
described for uncomplicated UTI. In asymptomatic bacteriuria, at least two successive cultures growing the same
organism in quantities greater than or equal to 105/mL
should be obtained to confirm the diagnosis.

Asymptomatic bacteriuria in the patient with a urinary

catheter, as well as in the elderly noncatheterized patient,
should ordinarily not be treated. Treatment in such patients
should be instituted if clinically apparent infection arises
or in special circumstances such as patients who are neutro penic or have a renal allograft. Asymptomatic bacteriuria
should always be treated in pregnancy due to the markedly
increased risk of ascending infection with attendant complications for both mother and child.
Selection of antimicrobials for treatment of symptomatic
complicated urinary tract infections should take into account the relatively broad array of bacterial species that
cause such infections and the degree of illness in the patient
(Table 1). For septicemia in hospitalized patients, parenteral therapy with broad spectrum agents must be used
initially until the infecting organism is identified. Initial
therapy should provide an effective antimicrobial spectrum
that includes activity against the enterococcus and pseudomonas. Acceptable regimens include ampicillin plus gentamicin, imipenem-ciliastatin, or a third-generation cephalosporin with antipseudomonal activity. Once the etiologic
agent is known, therapy can be specifically targeted to the
infecting agent. In patients with complicated urinary tract
infection who have lesser degrees of illness and can be
managed in the outpatient setting, fluoroquinolones can
be given orally. The duration of therapy for complicated
urinary tract infections ranges from 7 to 21 days, depending
on the clinical circumstance. Prolonged therapy may be
needed in patients with urinary calculi. In catheterized patients with candiduria, amphotericin bladder washes may
be effective, especially if the catheter can be removed. Oral
fluconazole can also be used. The relative efficacy of these
therapeutic approaches has not been established.
Prognosis
Complicated urinary tract infections often recur unless
the underlying anatomic or functional defect can be cor-

58. Re n a l a n d U r i n a r y T r a c t N e o p l a s i a

371

reeled. Infection usually cannot be eradicated in patients

with nephrolithiasis without removal of the calculi, either
by surgical means (in the case of larger or staghorn calculi)
or by lithotripsy (in the case of smaller stones). Patients
with indwelling catheters will continue to develop episodes
of infection despite treatment of individual episodes as they
arise. Infections caused by pseudomonas and enterococci
are especially prone to recurrence. Chronic and recurrent
infections of the urinary tract may lead to suppurative
sequelae or to loss of functioning renal tissue with subsequent impaired renal function.
Prevention

Urinary catheters should not be used unless they are

medically indicated. Sterile insertion and maintenance of a
closed catheter system can reduce the incidence of catheterassociated urinary tract infections. Condom catheters may
be an alternative. As the prevalence of infection increases
with the duration of catheterization, catheters should always be removed as soon as possible. In particular, the
junction between the distal catheter and the collecting system should remain closed, downhill gravity drainage should
be maintained at all times, and sterile technique should be
used whenever urine is drained from the collecting bag.
Detailed guidelines for prevention of catheter-associated
infections have been published elsewhere. Antimicrobial
prophylaxis has no value in chronically catheterized patients or in most patients with intermittent asymptomatic

bacteriuria. Antimicrobial prophylaxis should be provided

to patients undergoing prostatic surgery or other urological
procedures since such prophylaxis prevents postoperative
urinary tract infection and urosepsis. All pregnant women
should be screened for bacteriuria in the first trimester and
treated if positive.
Bibliography
Hooton TM, Scholes D, Hughes JP, et al.: A prospective study of
risk factors for symptomatic urinary tract infections in young
women. N Engl J Med 355:468-474, 1996.
Johnson JR, Stamm WE: Urinary tract infections in women: Diagnosis and treatment. Ann Intern Med 11:906-912, 1989.
Lipsky BA: Urinary tract infections in men: Epidemiology, pathophysiology, diagnosis, and treatment. Ann Intern Med 110:138
150, 1989.
Schaeffer AJ, Jones JM, Dunn JK: Association of in vitro Escherichia coli adherence to vaginal and buccal epithelial cells with
susceptibility of women to recurrent urinary tract infections.
N Engl J Med 304:1062-1066, 1981.
Stamm WE: Catheter-associated urinary tract infections: Epidemiology, pathogenesis, and prevention. Am J Med 91(Suppl
3B):65S-71S, 1991.
Stamm WE, Hooton TM, Johnson JR, et al.: Urinary tract infections: From pathogeneis to treatment. / Infect Dis 159:400406, 1989.
Stamm WE, Hooton TM: Management of urinary tract infections
in adults. N Engl J Med 329:1328-1334, 1993.
Strom BL, Collins M, West SL, Sreisberg J, Weller S: Sexual
activity, contraceptive use, and other risk factors for symptomatic and asymptomatic bacteriuria. Ann Intern Med 107:8 16823, 1987.

RENAL AND URINARY TRACT NEOPLASIA

ROBERT R. BAHNSON

It is estimated that in 1996, approximately 83,000 individuals will be diagnosed with cancer of urinary organs (bladder,
kidney, and other urinary) and roughly 24,000 deaths from
urinary neoplasia will occur. In addition, more than 317,000
men will be diagnosed with cancer of the prostate and more
than 41,000 will die from this common malignancy (Table
1). Neoplasms of the kidney, renal pelvis, ureter, bladder,
and prostate will be discussed separately highlighting their
etiology, pathology, diagnosis, staging, and treatment.

Primer on Kidney Diseases, Second Edition

RENAL NEOPLASMS
Etiology

The most common cancer of the kidney is renal cell

carcinoma. The tumor is more common in men, with a
male : female ratio of roughly 2:1. It is generally a tumor
of adults, with peak incidence between the fifth and seventh
decades of life. A recessive oncogene located on chromo some 3p (short arm) is associated with the formation of
Copyrigh t 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

372

Robert R. Bahnson
TABLE I

Estimated new cases of urinary

organ cancer

Men
Women
Total

Bladder
38,300
14,600
52,900

Kidney and
other urinary
18,500
12,100
30,600

Prostate
317,000

Estimated deaths from urinary

organ cancer

Bladder
7800
3900 11,700

Kidney and
other urinary
7300
4700
12,000

Prostate
41,400

both sporadic and familial renal cell carcinoma. This observation is supported by findings of loss of chromosome 3p
alleles in renal cell carcinomas from patients with von
Hippel-Lindau (VHL) disease. VHL syndrome is a hereditary disease characterized by spinal and cerebellar hemangioblastomas, retinal angiomas, pheochromocytomas, renal
and pancreatic cysts, cystadenomas of the epididymis, and
renal cell carcinomas. From 28 to 45% of VHL patients
will develop renal cell carcinomas of the clear cell type.
The tumors in these patients occur early in life (mean age
at diagnosis is 39 years) and are frequently multiple and
bilateral. No definite causal relationship between environmental carcinogens and the development of renal cell carcinoma has been documented.
Patients with chronic end stage renal disease (ESRD)
often develop acquired cystic disease of the kidney
(ACDK). Approximately 1 to 4% of these patients develop
renal carcinoma. This rate is roughly three to six times
greater than that of the general population.

cyst should be further evaluated by CT. As a single diagnostic imaging study, CT scanning is probably the best method
to evaluate a suspected renal mass lesion as it gives fairly
reliable information about the local extension of tumor,
lymph node involvement, and presence of tumor thrombus
in the renal vein or inferior vena cava. Renal arteriography
is seldom necessary unless required for the planning of
parenchyma sparing surgery.
The issue of screening for renal cell carcinoma in VHL
and ACKD is controversial (see Chapter 45). Since renal
involvement in VHL occurs in 28 to 66% of patients and
bilateral renal cell cancers develop in 63 to 75% of those
with renal lesions it has been recommended by some that
such lesions be followed by annual CT examinations.
ACKD increases significantly in frequency after 3 years
of dialysis. As in VHL, CT is probably the best imaging
technique. Clearly any patient with VHL or ACKD with
symptoms or hematuria should undergo a complete imaging evaluation.

Pathology

Staging

The proximal tubular cell has been identified by electron

microscopy and monoclonal antibodies as the origin of
renal cell carcinoma. The tumor is generally composed of
four common histologic types: clear cell, granular cell, tubulo papillary, and sarcomatoid. These cells grow in solid
sheets, trabecular patterns, or less commonly in papillary
configurations. The sarcomatoid renal cell carcinoma is also
known as the bone-metastasizing variant because of its
predilection for bony metastases. This histologic subtype
is characterized by aggressive clinical behavior and a
poor prognosis.

Staging of kidney carcinoma is best performed using the

tumor, node, metastases (TNM) classification agreed on by
the International Union Against Cancer and the American
Joint Committee on Cancer (Table 3). Cost effective clinical staging of patients can be accomplished with a history
and physical exam, chest X ray, liver function tests, and
serum calcium determination. The most frequent sites of
metastases in decreasing order of frequency are lymph
nodes, lung, liver, and bone. Patients with a palpable liver
or abnormalities of liver functions should have further radiographic or radionuclide imaging of the liver. Symptoms
of skeletal involvement or an increase in serum calcium or
alkaline phosphatase should prompt a radionuclide bone
scan.

Diagnosis

Renal cell carcinoma has often been called the internist's

tumor due to the variety of its presentation and the uncommon number of systemic syndromes that have been reported to occur in patients with the disease (Table 2). The
classic triad of flank pain, palpable mass, and hematuria is
present in only 10% of patients and is often an indication
of advanced disease. The majority of clinically confined
renal tumors are now discovered serendipitously by ultrasound, computed tomography (CT), and magnetic resonance imaging (MRI). As an initial radiologic test, ultrasound can distinguish cystic, solid, and complex renal
masses. Any mass that does not meet the criteria for simple

Treatment

The only successful treatment for patients with localized

renal cell carcinoma is surgical extirpation. Radial nephrectomy, removal of the kidney with its surrounding fascial
envelope (Gerota's) and the ipsilateral adrenal gland, is
the preferred surgical approach. The 5-year survival of
patients with node negative tumors without metastases
ranges from 47 to 82%. The presence of positive lymph
nodes or distant metastasis augurs poorly for survival beyond 2 years. Regrettably, radiation therapy and chemo -

58. Renal and Urinary Tract Neoplasia

373

TABLE 2

Incidence of Systemic Syndromes in Patients

with Renal Cell Carcinoma
Syndrome
Raised erythrocyte sedimentation rate
Hypertension
Anemia
Cachexia, Weight loss
Pyrexia
Abnormal liver function
Raised alkaline phosphatase
Hypercalcemia
Polycythemia
Neuromyopathy
Amyloidosis

%
55.6
37.5
36.3
34.5
17.2
14.4
10.1
4.9
3.5
3.2
2.0

therapy have been of limited value in this disease. Several

studies have failed to confirm an advantage to treated patients who received either pre- or postoperative radiotherapy. Multiple, single agent, and combination chemotherapy
trials have proven that renal cell carcinoma is highly resis tant to commonly employed cytoxic agents. The single most
active drug appears to be vinblastine, but objective response rates are too low to justify its routine use.
The most promising therapy of metastatic renal cell carcinoma in the past decade has been immunotherapy. Nonspecific active, systemic immunotherapy with bacillus
Calmette-Guerin (BCG) or Cornybacterium parvum has
shown only modest success. Cytokine therapy utilizing interferons has shown response rates reported in the 15 to
20% range. Most patients who respond have undergone
nephrectomy, have an excellent pretreatment performance
status, and have limited pulmonary metastases. Rosenberg
and colleagues at NCI have used adoptive immunotherapy
with lymphokine-activated killer (LAK) cells and interleukin-2 (IL-2). They reported a 35% objective response rate
in 72 patients with renal cell carcinoma. Since then, efforts
of this group have focused on using genetically altered
tumor infiltrating lymphocytes (TIL) which are grown from
single-cell tumor suspensions cultured in IL-2. Response
rates of 0 to 33% have been reported in five trials of combination IL-2/TIL in patients with metastatic renal cell carcinoma.

cell carcinoma, the treatment of choice is radical nephrectomy.

Angiomyolipoma
Angiomyolipomas are hamartomas composed of blood
vessels, adipocytes, and abundant smooth muscle. They are
seen commonly in patients with tuberous sclerosis and have
a tendency to multiplicity and bilaterality. The fat content
of these lesions often permits radiographic diagnosis with
CT imaging. Asymptomatic tumors under 4 cm may be
observed with radiographic monitoring. Symptomatic lesions can often be treated with angioinfarction. If surgery
is required, excision of the lesion using renal parenchymal
sparing techniques should be attempted because of the
tendency toward multiplicity and bilaterality.
Sarcomas
Sarcomas represent less than 5% of all malignant tumors
of the kidney. The most common sarcoma is leiomyosarcoma. Others include osteogenic sarcoma, liposarcoma,
carcinosarcoma, fibrosarcoma, rhabdomyosarcoma, and
malignant fibrous histiocytoma. The treatment of choice
for all of these tumors is radical nephrectomy, if possible.
Metastatic Tumors
Because of their rich vascularity, the kidneys are frequently the site of metastatic spread. These tumors are
often asymptomatic and discovered only at autopsy.
Lymphoma is probably the most common metastatic neoplasm of the kidney; it may involve the renal parenchyma
diffusely. With the exception of lymphomatous involvement which may respond to systemic therapy, treatment is
usually unnecessary since the patient's prognosis from their
primary tumor is almost uniformly poor.
RENAL PELVIC AND URETERAL TUMORS
Epithelial tumors of the upper urinary tract are uncommon. They show a male to female predominance of roughly
2:1 and are more common among whites than AfricanAmericans. Tumors of the upper urinary tracts are common
in patients with Balkan nephropathy and are more frequently bilateral. These upper tract tumors are commonly
associated with bladder tumors which occur in 40 to 75%
of patients at some point in time. Patients with bladder
tumors have a slight (2 to 13%) chance of developing ureteral or renal pelvic tumors.

Other Renal Tumors

Etiology

Renal Oncocytoma
Oncocytomas constitute approximately 5% of solid renal
neoplasms and invariably pursue a benign course. Microscopically these tumors are composed of polygonal cells
with intense eosinophilia, granular cytoplasm, and abundant mitochondria. Grossly, on cut section, these tumors
have brown pigmentation and are often associated with a
central scar. The origin of these tumors is thought to be
the distal collecting tubule. Because preoperative diagnosis
is difficult and because oncocytomas can coexist with renal

An increased risk of renal pelvic and ureteral tumo rs

has been reported for patients who smoke, abuse analgesics
(particularly phenacetin), and who have been occupationally exposed to chemicals, petrochemicals, plastics, coke,
coal, asphalt, and tar. Among these, cigarette smoking is the
major risk factor for carcinoma of the upper urinary tract.
Pathology
Transitional cell carcinoma accounts for more than 90%
of all upper tract urothelial tumors. The remaining 10%

374

Robert R. Bahnson
TABLE 3

The Staging of Kidney Cancer

Primary tumor (T)
TX
Primary t umor cannot be assessed
TO
No evidence of primary tumor
Tl
Tumor 2.5 cm or less in greatest dimension, limited, to the kidney
T2
Tumor more than 2.5 cm in greatest dimension, limited to the kidney
T3
Tumor extends into major veins or invades adrenal gland or
perinephric tissues, but not beyond Gerota's fascia T3a Tumor
invades adrenal gland or perinephric tissues, but not
beyond Gerota's fascia
T3b Tumor grossly extends into renal vein(s) or vena cava
T4
Tumor invades beyond Gerotas fascia
Regional lymph nodes (N)
The regional lymph nodes are the abdominal para-aortic, paracaval, and renal hilar.
Laterality does not affect the N classification. The significance of the regional
lymph node metastasis in staging kidney cancer lies in the number and size and
not in whether unilateral or contralateral. NX Regional lymph nodes cannot be
assessed NO No regional lymph node metastasis
Nl Metastasis in a single lymph node, 2 cm or less in greatest dimension N2
Metastasis in a single lymph node, more than 2 cm but not more than 5 cm in
greatest dimension, or multiple lymph nodes, none more than 5 cm in greatest
dimension N3 Metastasis in a lymph node more than 5 cm in greatest
dimension
Distant metastasis (M)
MX
Presence of distant metastasis cannot be assessed
MO
Ml
AJCC/UICC stage grouping
Stage I
Stage II
Stage III

Stage IV

No distant metastasis
Distant metastasis

Tl
T2
Tl
T2
T3a
T3a
T3b
T3b
T4
Any T
Any T

AnyT

NO
NO
Nl
Nl
NO
Nl
NO
Nl
Any N
N2
N3
Any N

consists of adenocarcinoma, squamous cell carcinomas, and

inverted papillomas. Nonurothelial tumors are exceedingly rare.

MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
Ml

Staging
Metastatic evaluation is best performed utilizing CT
scanning to assess the presence and degree of local, regional, and distant disease. Staging is illustrated in Table 4.

Diagnosis
The most common presentation of upper tract tumors
is gross, total (i.e., bloody throughout micturition), painless
hematuria. The presence of string-like clots is suggestive
of an upper tract origin of the bleeding. The diagnosis is
confirmed by radiologic imaging, most often with excretory
or retrograde urography. At the time that retrograde pyelography is undertaken, saline barbotage and/or brush
biopsy of the lesion can be performed. Cellular material
obtained by such maneuvers can be examined cytopathologically to confirm the diagnosis.

Treatment

The treatment of upper tract urothelial tumors almost

always involves surgical excision. Neither radiotherapy nor
chemotherapy has been shown in large series to have a
substantial impact on response or survival. Renal pelvic
tumors are treated by nephroureterectomy with bladder
cuff excision because the risk of ipsilateral ureteral or bladder tumor recurrence is high (30 to 75%). Upper and midureteral tumors are best treated by segmental resection if
the tumors are solitary or low grade, and by nephroureter-

375

58. Renal and Urinary Tract Neoplasia

TABLE 4
Staging Systems of Upper Tract Urothelial Tumors

Finding
Carcinoma in situ
Noninvasive papillary tumor
Tumor involving submucosa
Muscle-invasive tumor
Tumor invading renal parenchyma

Tumor invading peripelvic or

periureteral soft tissue
Invasion of contiguous organs
Lymph node metastases
Distant metastases

GrabstaldCummings
Stage
I '
I
II
III
III
III
IV
IV
IV

1987 UICC
Stage
Tis
Ta
Tl
T2
T3
T3
T4

(Nl-3)
Ml

ectomy if they are multiple or high grade. Distal ureteral

tumors are most effectively treated with distal ureterectomy and ureteroneocystostomy.
BLADDER NEOPLASMS
Bladder cancer is a common urologic tumor that is
broadly divided into two biologically meaningful subtypes.
The first is superficial low grade transitional cell carcinoma
which has as its greatest risk a tendency for frequent recurrence within the bladder. The second is invasive high grade
cancer which often presents in advanced stages and is associated with substantial mortality.
Etiology
Bladder tumors are more frequent in men than women
(approximately 3:1) and in older patients. The peak incidence of the disease is the seventh decade of life. Environmental carcinogens have been postulated as the major causative factor in this cancer. There is a clear causal
relationship between cigarette smoking and bladder cancer.
Smokers have at least a fourfold increased risk of developing bladder tumors, and it is estimated that smoking is
responsible for the development of one third of all bladder
tumors. Occupational exposure to aromatic amines is simi larly known to predispose a patient to the development of
bladder cancer. In particular, a high prevalence of bladder
tumors has been shown in individuals exposed to aniline
dyes. Other occupations linked to bladder cancer development include truck driving, painting, and leather working,
as well as others in which there has been chronic exposure
to volatile chemicals such as dry cleaning. Similar to upper
tract urothelial cancers, analgesic abuse is associated with
an increased risk of bladder tumors. Other putative carcinogens include radiation therapy and cyclophosphamide chemotherapy.
Bladder irritation from chronic infection or a long-term
indwelling foley catheter is associated with the development of squamous cell carcinoma. From 2 to 10% of paraplegics with chronic indwelling catheters develop bladder
cancer, and approximately 80% are of squamous histology.

In the Nile River Valley, schistosomal infestation of the

bladder (bilharziasis) is associated with the development
of squamous cell carcinoma.
Pathology
The overwhelming majority of bladder tumors are transitional cell carcinomas. Although most bladder tumors arise
de novo, a number of proliferative lesions of the bladder
have been identified. These include atypical hyperplasia,
Von Brunn's nests, cystitis cystica, cystitis follicularis, cystitis glandularis, inverted papilloma, nephrogenic adenoma,
squamous metaplasia, and leukoplakia.
Both uroepithelial dysplasia and carcinoma in situ (CIS)
of the bladder have been documented as preneoplastic
lesions. Dysplasia is characterized as mild, moderate, and
severe, and is intermediate between normal uroepithelium
and carcinoma in situ. CIS is a term reserved for poorly
differentiated transitional cell carcinoma confined to the
surface uroepithelium. It may be asymptomatic or associated with symptoms of bladder irritability such as frequency, urgency, or dysuria. At cystoscopy, it often appears
as a velvety patch of erythematous mucosa. CIS is frequently (20 to 75%) noted to coexist (in adjacent or distant
mucosa) with muscle invasive transitional cell carcinomas
of the bladder. Although the natural history of this disease
is not fully known, it is of interest that 20% of patients who
underwent cystectomy for treatment of carcinoma in situ
were found to have microscopically invasive cancer. At
present, the treatment of choice for CIS is intravesical
instillation of BCG which produces complete regression in
approximately 50 to 65% of patients.
As mentioned previously, more than 90% of bladder
cancers are transitional cell carcinoma. These tumors arise
most commonly at the bladder neck and lateral bladder
walls. Roughly 70% are low-grade superficial tumors. While
the majority of patients will develop a recurrence, only a
small percentage of these tumors (10 to 15%) will progress
to the more serious muscle invading tumors. Unfortunately,
80 to 90% of the patients with muscle invasive tumors have
the diagnosis made at their first presentation. Even more
sobering is the fact that nearly 50% of this group will already
have occult distant metastases.
Much less common are squamous cell carcinoma and
adenocarcinomas. Squamous cell carcinomas are frequent
in Egypt because of chronic infection with Schistosoma
haematobium. In the United States they constitute 3 to 7%
of tumors and afflict patients with chronic irritation from
infection, stones, or indwelling catheters. Treatment consists of radical cystectomy, since radiation and chemotherapy are ineffective. Adenocarcinomas represent less than
2% of bladder tumors and are most often seen in patients
born with extrophic bladders or in urachal tumors that
invade the dome of the bladder. Adenocarcinomas respond
poorly to radiotherapy or cytotoxic chemotherapy and are
best treated by exenterative surgery.
Diagnosis
Patients with carcinoma of the bladder usually (85%)
present with gross painless hematuria. Although conven-

376

Robert R. Bahnson

tional cytology, flow cytometry, and quantitative fluores cent image analysis have all been utilized to diagnose bladder tumors, the diagnosis is most reliably confirmed by
cystoscopy and biopsy. Excretory urography is used to rule
out involvement of the upper tracts by tumor and to detect
ureteral obstruction which is almost always a sign of muscle
invasive cancer.
Early detection studies have reported the favorable experience of using chemical reagent strips to test for hematuria. Approximately 20% of the population was positive and
8 to 22% of those further evaluated were found to have
urinary malignancies. Two thirds of the detected tumors
were bladder cancers. Further studies are necessary before
screening can be recommended.
Staging

Clinical staging of bladder patients is essential for planning therapy. The most important information is the pathologic evaluation of the transurethral bladder tumor resection. Grade and depth of penetration of the tumor into
the bladder wall determine if the patient has a low-grade,
superficial or high-grade, infiltrating tumor. Superficial tumors do not require further staging evaluation. Muscle
infiltrating tumors are best staged with CT scanning to rule
out lymph node or visceral metastases. Chest radiography
and serum chemistries should also be performed and if the
alkaline phosphatase is elevated a bone scan should be
considered. The most accurate means of detecting lymph
node metastases is pelvic lymphadenectomy but this is usually performed at the same time as radical cystectomy. The
TNM classification of bladder tumors is best utilized for
staging and is illustrated in Table 5.
Treatment

Patients with superficial tumors can be managed effectively by transurethral resection. Patients should have follow-up cystoscopy at 3 month intervals for 2 years, then
every 6 months for 2 years and annually thereafter. This
can usually be performed in the office using flexible cystoscopes. The troublesome feature of superficial tumors is
their 70% recurrence rate.
Patients who are judged to have a high risk of recurrent
tumor because of increased grade or stage, multiple tumors,
or a history of recurrence can be treated with intravesical
chemotherapy or immunotherapy. Thiotepa, doxorubicin,
and mitomycin C are cytotoxic agents commonly employed
in patients with superficial transitional cell carcinoma. They
have been shown to reduce recurrence rates to 30 to 44%.
Intravesical BCG has been shown in prospective trials to
be highly effective in reducing recurrence rates (0 to 41%),
and is currently the favored method of prophylaxis.
Patients who present with muscle invasive bladder cancer
can be treated with surgery, radiation therapy, or chemo therapy. At present, the most effective local therapy is
radical cystectomy. Pelvic recurrence rates are 10 to 20%,
compared to 50 to 70% with radiation therapy, chemother-

apy, or combination of both. The disadvantage of radical

surgery is the loss of normal bladder function. Radical
cystectomy in men consists of removal of the pelvic lymph
nodes, bladder, and prostate. If the tumor is associated
with diffuse CIS of the bladder or involvement of the prostate, total urethrectomy is included. In women the operation is commonly called an anterior exenteration and includes excision of pelvic lymph nodes, bladder, urethra,
ovary, fallopian tubes, uterus, and the anterior wall of the
vagina. Bladder reconstruction usually consists of a ileal
conduit or a continent urinary reservoir. The continent
diversions may be cutaneous, or in eligible men, brought
down to the urethra. The morbidity of radical cystectomy
and bladder reconstruction is 25 to 35% and mortality rate
is 1 to 2%. Five-year survivals of 75% have been reported
for patients with T2, T3a disease, 44% for T3a, T3b tumors,
and 36% for T4 lesions.
Radiation as monotherapy has not been shown to have
high cure rates in patients with invasive disease. Five-year
survival rates have averaged 20 to 35%. Bladder sparing
protocols utilizing chemotherapy with cisplatin, methotrexate and vinblastine (CMV), and radiation therapy
(4000 cGy) have been reported. If the patients have a
complete response from this treatment, further radiotherapy to a total dose of 6480 cGy is given. Patients with only
a partial response are advised to have a cystectomy.
Chemotherapy has been utilized primarily in metastatic
disease but more recently has been used as an adjunct to
surgery or radiotherapy. Cisplatin is the single most active
agent, but combination chemotherapy with methotrexate,
vinblastine, doxorubicin, and cisplatin (M -VAC) has been
shown in patients with metastatic disease to be superior in
a multicenter randomized trial. However, the complete
response rates are still low and of relatively short duration.
Chemotherapy prior to planned surgery (neoadjuvant)
has been advocated to reduce the risk of disease recurrence
after cystectomy. Proponents of this approach argue that;
(1) micrometastases may be present at the time of diagnosis; (2) micrometastases are eliminated most efficiently
when the volume of tumor is small (3) downstaging of the
tumor prior to surgery can occur, and (4) there may be a
reduction in the chemotherapy access to tumor as a consequence of surgery. As yet, there is no evidence from properly designed trials to favor this approach. Similarly, no
well-designed properly executed trials have been performed which show a definite advantage to the use of adjuvant chemotherapy following surgery. Nevertheless, the
increased effectiveness of combination chemotherapy
makes the concept of adjuvant therapy appealing and some
studies suggest a potential advantage for patients at high
risk for disease recurrence who are treated with adjuvant chemotherapy.
Other Bladder Tumors

Besides the more common transitional, adeno, and squamous cell carcinomas, other epithelial tumors include small
cell carcinoma and carcinosarcoma. Small cell neoplasms

58. Renal and Urinary Tract Neoplasia

377
TABLE 5
Staging for Urinary Bladder Carcinoma

Primary tumor (T) The suffix "m" should be added to the appropriate T
category to indicate
multiple tumors.
The suffix "is" may be added to any T to indicate the presence of associated
carcinoma in situ.
TX Primary tumor cannot be assessed TO No evidence of
primary tumor Ta Noninvasive papillary carcinoma Tis
Carcinoma in situ: "flat tumor" Tl Tumor invades subepithelial
connective tissue T2 Tumor invades superficial muscle (inner
half) T3 Tumor invades deep muscle (outer half) or perivesical
fat T3a Tumor invades deep muscle (outer half) i.
microscopically
ii. macroscopically (extravesical mass) T4 Tumor invades any of
the following: prostat e, uterus, vagina, pelvic wall,
abdominal wall
T4a Tumor invades any: prostate, uterus, vagina T4b Tumor
invades pelvic wall and/or abdominal wall Regional lymph nodes
(N)
Regional lymph nodes are those within the true pelvis; all others are distant
nodes.
NX Regional lymph nodes cannot be assessed
NO No regional lymph node metastasis
Nl Metastasis in a single lymph node, 2 cm or less in greatest dimension N2
Metastasis in a single lymph node, more than 2 cm but not more than 5 cm in
greatest dimension, or multiply lymph nodes, none more than 5 cm in greatest
dimension N3 Metastasis in a lymph node more than 5 cm in greatest
dimension
Distant Metastasis (M)
MX
Presence of distant metastasis cannot be assessed
MO
No distant metastasis Ml
Distant metastasis
AJCC/UICC stage grouping
Stage 0
Stage I
Stage 11
Stage 111
Stage IV

Oa
Ois

Ta
Tis
Tl
T2
T3a
T3b
T4
Any
Any
Any
Any

T
T
T
T

NO
NO
NO
NO
NO
NO
NO
Nl
N2
N3
Any N

are thought to arise from neuroendocrine cells and stain

positively for neuroendocrine markers using immunohistochemistry. Carcinosarcomas are tumors which contain both
epithelial and mesenchymal elements. Both of these aggres sive tumors are treated with radical surgery.
Nonepithelial bladder tumors are quite uncommon and
consist of sarcomas such as neurofibroma, leiomyosarcome,
angiosarcoma, and rhabdomyosarcoma. These tumors are
best treated with surgery. In addition to these tumors, pheochromocytomas, lymphomas, and malignant melonoma

MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
Ml

have all been reported in the bladder. Surgical excision

is preferred except for lymphoma which can be treated
effectively with radiation therapy.
PROSTATE CANCER
Etiology

Prostate cancer is the most common cancer occurring in

adult American men and the second leading cause of cancer

378

Robert R. Bahnson
TABLE 6

Staging for Prostate Carcinoma

Primary tumor (T)
TX
Primary tumor cannot be assessed
TO
No evidence o f primary tumor
Tl
Clinically inapparent tumor not palpable nor visible by imaging
Tla Tumor incidental histologic finding in 5% or less of tissue resected
Tib Tumor incidental histologic finding in more than 5% of tissue resected
Tic Tumor ide ntified by needle biopsy (e.g., because of elevated PSA)
T2
Tumor confined within prostate
T2a Tumor involves half of a lobe or less
T2b Tumor involves more than half of a lobe, but not both lobes
T2c Tumor involves both lobes
T3
Tumor extends through the prostatic capsule
T3a Unilateral extracapsular extension
T3b Bilateral extracapsular extension
T3c Tumor invades seminal vesicle(s)
T4
Tumor is fixed or invades adjacent structures other than seminal vesicles
T4a Tumor invades any of: bladder neck, external sphincter, rectum
T4b Tumor invades levator muscles and/or is fixed to pelvic wall
Regional lymph nodes (N)
The regional lymph nodes are the nodes of the true pelvis.
NX Regional lymph nodes cannot be assessed
NO
No regional lymph node metastasis
Nl
Metastasis in a single lymph node, 2 cm or less in greatest dimension
N2
Metastasis in a single lymph node, more than 2 cm but not more than 5
cm in greatest dimension, or multiple lymph node metastases, none more
than 5 cm in greatest dimension
N3
Metastasis in a lymph node more than 5 cm in greatest dimension

Stage 0
Stage I

Stage II
Stage III
Stage IV

Distant metastasis (M)

MX
Presence of distant metastasis cannot be assessed
MO
No distant metastasis
Ml
Distant metastasis
M i a N o n -regional lymph node(s )
Mlb Bone(s)
Mlc Other site(s)
AJCC/UICC stage grouping
Tla
NO
MO
Gl
Tla
NO
MO
G2, 3-4
Tib
NO
MO
Any G
Tic
NO
MO
Any G
Tl
NO
MO
Any G
T2
NO
MO
Any G
T3
NO
MO
Any G
T4
NO
MO
Any G
Any T
Nl
MO
Any G
Any T
N2
MO
Any G
Any T
N3
MO
Any G
Any T
Any N
Ml
Any G

death. African-Americans and men with a family history

of carcinoma of the prostate are at increased risk of developing the disease. A diet high in fat may predispose to the
development of prostate cancer as well.
Pathology

More than 95% of the prostate cancers are acinar adenocarcionmas. Most are multifocal and arise from the peripheral zone of the prostate.

Diagnosis

Low stage prostate cancer usually is clinically silent.

Asymptomatic tumors are most often discovered by digital
rectal examination during a routine physical or by an elevation of serum prostate specific antigen (PSA). Screening
for cancer of the prostate should be performed annually
beginning at age 50, or in the case of African-Americans
or those with a positive family history, at age 40. Suspicious
digital rectal exam or increased PSA should prompt confirmatory diagnostic needle biopsy of the prostate. This is

58. Renal and Urinary Tract Neoplasia

most often performed conjointly with transrectal ultrasound.
Staging

Prostate cancers are best staged using TNM classification

(Table 6). The most frequent sites of metastasis are the
lymph nodes and bone. It is no lo nger cost effective
to perform routine bone scans in patients with PSA levels
<10 ng/mL. CT and MRI studies are not sensitive or specific for evaluating either T or N stage, and their routine
use should be discouraged. Recently, careful retrospective
study has confirmed the utility of nomograms utilizing clinical stage, Gleason score, and serum PSA to accurately
predict pathologic stage in patients with clinically localized
carcinoma of the prostate. Tumor-derived PSA appears
more likely to circulate in plasma in a protein-bound form
than PSA originating in benign tissue. Preliminary data
suggest that a low ratio of free to bound PSA may discriminate prostate cancer from benign prostate hypertrophy
(BPH). However, these findings require further validation.
Treatment

The treatment of patients with carcinoma of the prostate

is controversial. The debate has been engendered by the
observation that men with carcinoma of the prostate often
are older and die of other diseases before the cancer causes
serious morbidity or mortality. Thus, therapeutic decisions

379
involve an assessment of a patient's general health and life
expectancy. For patients with clinically localized tumors
and a life expectancy of 10 years or more, radical prostatectomy or external beam radiotherapy are the best options.
Other acceptable strategies include observation or androgen deprivation therapy. Androgen deprivation can be accomplished by bilateral orchiectomy or by the use of
LHRH agonists with or without oral antiandrogens. Patients with metastatic disease are best treated with androgen deprivation since nearly 80% of those men will have
a favorable response to therapy.
Bibliography
Catalona WJ: Urothelial tumors of the urinary tract. In: Walsh,
Retik, Stamey, Vaughn (eds) Campbell's Urology, 6th ed. W.B.
Saunders, Philadelphia, pp. 1094 -1158, 1992.
deKernion JB, Belldegrun A: Renal tumors. In: Walsh, Retik,
Stamey, Vaughn (eds) Campbell's Urology, 6th ed. W.B. Saunders, Philadelphia, pp. 1053 -1093, 1992.
Garnick M, Fair WR: Prostate cancer: Emerging concepts. Part
I. Ann Intern Med 125:118-125, 1996.
Garnick MB, Fair WR: Prostate cancer: Emerging concepts. Part
II. Ann Intern Med 125:205 -212, 1996.
Motzer RJ, Bander NH, Nanus D: Renal-cell carcinoma. N Engl
J Med 335:865-875, 1996.
Vogelzang NJ, Scardino PT, Shipley WU, Coffey DS: Comprehensive Textbook of Genitourinary Oncology. Williams & Wilkins,
Baltimore, 1996.
Rosenberg SA: Karnofsky memorial lecture: The immunotherapy
and gene therapy of cancer. / Clin Oncol 10:180, 1992.

SECTION 9
THE KIDNEY IN SPECIAL
CIRCUMSTANCES

THE KIDNEY IN INFANTS AND CHILDREN

R. ARIEL GOMEZ

ANATOMICAL DEVELOPMENT OF THE KIDNEY

In humans, the metanephric or definitive kidney appears

during the fifth week of gestation. Nephrogenesis termi nates by 32 to 36 weeks of gestation. Renal function and
urine production begin about 10 weeks into embryonic life.
The kidneys can be detected by ultrasound by the 18th
week of gestation. Term infants are born with a full complement of nephrons. However, because deeper glomeruli develop earlier than superficial ones (centrifugal maturation),
juxtamedullary glomeruli are larger and more mature than
outer cortical ones. The glomeruli become equal in size by
the 14th postnatal month and reach adult size (200 um) at
3 1/2 years of age. The proximal tubule continues to increase in length well into adult life.

48 hours after birth, creatinine levels are similar to the

maternal values. A week after birth, the full-term neonate
should have a creatinine below 1 mg /dL. These levels
should continue to decrease to about 0.3 mg/dL by 3 months
of postnatal life. In preterm infants, however, creatinine
levels can be high (about 1.5 mg/dL) during the first few
weeks of postnatal life. These levels progressively decrease
to 0.4 to 0.8 mg/dL by 3 to 6 months of postnatal life.
When evaluating a neonate with a high creatinine, serial
measurements are more meaningful than a single determination; creatinine levels should progressively decrease. An
increase in serum creatinine indicates a reduction in GFR,
regardless of the gestational age. The normal values for
creatinine during childhood are shown in Fig. 1.

FUNCTIONAL DEVELOPMENT OF THE KIDNEY

Urine Concentration

Glomerular Filtration Rate

In human fetuses, urine flow rate increases from 0.1 mL/

min at 20 weeks of gestation to 1.0 mL/min at 40 weeks of
gestation. In neonates, urine flow rate decreases again to
around 0.1 mL/min. Fetal urine flow correlates with (and
may be modulated by) fetal arterial pressure. The fetal
urine is hypotonic (100 to 200 mOsm/L), unless stress such
as hypoxemia or volume depletion develops. Nevertheless,
the urinary concentrating ability is lower in the fetus than
in the adult. Several factors contribute to the decreased
urinary concentrating ability. In the fetus, the organization
and anatomical development of the medulla is delayed with
respect to glomerular development. Short loops of Henle
and reduced NaCl transport in the ascending loops of Henle
are also important factors. In addition, preferential blood
flow to the inner portions of the cortex dissipates an already
low intrarenal concentration gradient. Although AVP is
secreted properly in response to volume and osmolar challenges, decreased AVP receptor number or coupling to
cAMP generation likely contribute to the decreased concentrating ability found in infants.
Maximal urinary concentration increases progressively
throughout the first year of life (Table 2). In prema ture neonates, maximal urine osmolality is about 400 to
500 mOsm/L. Later in life, infants born prematurely are
able to concentrate their urine to levels found in term
infants of similar postnatal age. This is probably due to
a rapid maturation of the concentration gradient in the
premature infant.

In newborn babies, glomerular filtration rate (GFR) correlates with gestational age. The GFR increases postnatally
and doubles by 2 weeks of age. When corrected for surface
area, the GFR reaches adult values at 1 to 2 years of age.
The normal values of GFR for infants and children are
shown in Table 1. Factors responsible for the increase in
GFR with maturation include: increase in arterial pressure,
increase in renal blood flow, increase in glomerular permeability, and increase in filtration surface area.
Creatinine clearance can be used to measure the GFR
after 1 month of age providing that urinary creatinine excretion is between 15 and 25 mg kg"1 day'1. However,
in infants it is difficult to obtain 24-hour urine collections.
Shorter collections are not adequate because of variations
in creatinine excretion throughout the day. The following
formula permits estimation of GFR (mL mirT1 1.73 m~2)
without the need for urine collections,
GFR = K L/ScR ,
where K is a constant (0.33 in preterm, 0.45 in term neonates, 0.55 in children and adolescent girls , and 0.70 in
adolescent boys), L. is the length of the infant in cm, and
S a. is serum creatinine (mg/dL). Although this formula is
clinically useful, inulin clearance or iothalamate clearance
should be obtained if a more accurate value of GFR is
needed. Along with changes in GFR, serum creatinine levels vary with age and degree of maturity. During the first

Primer on Kidney Diseases, Second Edition

383

Copyright 1998 by the National Kidney Foundation

All rights of rep roduction in any form reserved.

384

R. Ariel Gomez
TABLE I

TABLE 2

Normal Values of GFR" (ml mirT1 1.73 m" 2)

Maximal Urine Osmolality (mOsm/L)

Preterm (25-28 weeks)

1 week
11.0 5.4
2-8 weeks Preterm (29-34 15.5 6.2
weeks)
1 week
15.3 5.6
2-8 weeks Term
28.7 13.8
5-7 days
1-2 months
50.6 6.8
3-4 months
64.6 5.8
5-8 months
85.8 4.8
9-12 months
87.7 11.9
2-12 years
86.9 8.4
133 27
" From Schwartz et al.: Pediatr
Clin North Am 34: 571-590, 1987 and Greene
MG: The Harriet Lane Handbook. A Manual for
Pediatric House Officers, 12th ed. Mosby, St.
Louis, pp. 1-434, 1991.

MALES

14-18 years
Means SD.

sodium excretion is negatively correlated to gestational

age. The fractional excretion of sodium at 31 weeks of
gestation is approximately 5% and declines to less than 1%
by 2 months after birth. Thus, in contrast to full-term
healthy newborns that are able to conserve sodium and
maintain a positive sodium balance, preterm infants are at
risk of developing a negative sodium balance during the
first weeks of life. It is believed that the high sodium excretion in the premature baby is due to a relative inability of
the distal tubule to respond to aldosterone. Newly born
preterm infants who receive human breast milk or formulas
with a composition similar to that of breast milk containing
relatively low sodium concentrations continue to excrete
large amounts of sodium in the urine and eventually develop a negative sodium balance and hyponatremia. The
hyponatremia can be prevented by the addition of supplemental sodium chloride to the formula.
Although sodium retention in the maturing individual is
essential for growth, the counterpart of this chronically
enhanced sodium reabsorption is that newborns have difficulty in rapidly excreting an acute load of sodium and water.
The renal excretory response to a sodium load increases
progressively during infancy and is fully developed by the

MEAN
&

<j> FEMALES
^

663 133
896 179
1118 154
1362 109

10-12 months

Sodium metabolism in the newborn period is characterized by: (1) a progressive increase in renal reabsorptive capacity, (2) positive balance in the term infant, and
(3) an inability to promptly excrete a solute and volume
load.
About 30% of the dietary sodium is retained by the
healthy infant receiving formulas containing different sodium concentrations. Although sodium intake varies
widely, positive sodium balance is usually maintained, thus
allowing the conservation of sodium which is a necessary
requirement for somatic growth. However, premature infants excrete more sodium than full-term infants. In fact,

515 172

6 days
10-30 days

Sodium

1.0

3 days

0.8

^
Z

0.6

0.4
QC

u
HI 0.2

10

I
12

14

I
16

I
18

J
20

AGE (YEARS)
FIGURE I Values of plasma creatinine in normal males and females. From Schwartz et al.: J Pediatr 88:828-830, 1976. With
permission.

59. The Kidney in Infants and Children

385

first year of life. In addition to the low GFR, enhanced

distal tubular sodium reabsorption is responsible for the
retention of sodium.
Based on the information described above, the sodium
requirements for a term newborn range from 1 to 1.5
mEq kg"1 day" 1, whereas the requirements for a
preterm neonate range from 3 to 5mEq kg'1 day"1. The
sodium content of breast milk is appropriate for the
feeding of the healthy term infant. It should be
remembered, however, that if excess sodium is given,
such as feeding undiluted cow's milk, there is a risk of
inducing sodium and water retention with extracellular
volume expansion and sometimes clinically evident
edema.
Oliguria

Most neonates void within the first 24 hours after birth.

Any newborn baby, whether term or premature, who has
not urinated within the first 24 hours of life should be
evaluated for renal disease. Oliguria in infancy is defined
as urine flow rate less than 1 mL kg'1 hr" 1. This is based
on the usual renal solute load (7 to 15 mOsm kg"1 day" 1)
contributed by feedings and a maximal urinary osmolality
of 500 mOsm/kg. To maintain solute balance, the newborn
will therefore need to urinate about 30 mL kg"1 day"1
(1.25 mL kg"1 hr" 1). In infants, oliguria can be easily
evaluated by calculating the fractional excretion of sodium
(FENa) (see Chapter 2).
Determination of the FENa is convenient because it does
not require a timed urine collection. The FENa is particularly helpful in distinguishing oliguria due to a prerenal
(low FENa) etiology from that due to a renal (high FENa)
cause. In the oliguric term neonate, a FENa higher than 2.5%
indicates a renal cause (such as acute tubular necrosis),
whereas a FENa of less than 2.5% suggests prerenal causes
such as dehydration, hypovolemia, hypoalbuminemia, or
decreased effective plasma volume. As mentioned above,
premature babies normally have higher FENil (around 5%)
and, therefore, the cutoff of 2.5% is not useful until after
10 days of postnatal life. In the child as in adults a FENa
of 1% is utilized as the break point.
Calcium

Fetal calcium levels are higher than maternal values because of active calcium transport across the placenta toward
the fetal side. The fractional reabsorption of calcium by
the loop of Henle increases with maturation. Therefore,
infants excrete more calcium in the urine than older children. Most commonly, hypercalciuria in the newborn is

seen in neonates with bronchopulmonary dysplasia who

have received calciuric drugs such as furosemide or glucocorticoids. Hypercalciuria, in turn, can lead to nephrocalcinosis, urolithiasis, and decreased renal function. When
possible, a thiazide diuretic that reduces urine calcium excretion should be used instead of furosemide. Phosphate
depletion in preterm infants can also lead to hypercalciuria.
With the use of new formulas richer in phosphate, hypercalciuria, and other complications of phosphate depletion such
as osteopenia are less frequently observed.
Urine calcium excretion in the neonate can be estimated
by calculating the calcium/creatinine ratio in a random
urine sample. A ratio above 0.8 in preterm and 0.4 in fullterm infants is considered hypercalciuria. In older children,
a ratio above 0.2 or a calcium excretion in a timed urine
sample of 4mg kg'1 24 hr" 1 are used to diagnose hypercalciuria.
Newborn infants are also prone to hypocalcemia, especially when they are sick. Contributing factors include decreased responses to PTH, high serum phosphate levels,
and the rapid bone mineralization characteristic of the newborn period. The normal range for calcium levels in infants
and children are shown in Table 3.
Phosphate

Serum phosphate concentration is normally higher in the

neonate than in the adult (Table 4). Any neonate with a
serum phosphate below 4 mg/dL should be evaluated for
tubular wasting of phosphate. This can be easily done by
calculating the tubular reabsorption of phosphate (TRP)
after measuring the concentrations of creatinine and phosphate both in urine and plasma:
TRP =

[l -(t/PO/Pp04)/(t/cr/Pcr X

where t/P04 and PP04 are the urinary and plasma concentrations of phosphate (mg/dL), (respectively and PCI and Ua
are the plasma and urinary concentrations of creatinine
(mg/dL), respectively. It should be remembered that the
normal TRP varies with age. After the first week of postnatal life, the TRP should exceed 95% in full-term and 75%
in preterm neonates. After the first month of postnatal life
and throughout adulthood, the TRP should be above 85%.
Values below those mentioned above should suggest Fanconi's syndrome, hyperparathyroidism, or chronic renal
failure.
Potassium

The total body potassium (K+ ) in infants is 40 mEq/kg,

compared with the adults of 50 mEq/kg. Growing infants

TABLE 3

TABLE 4

Calcium Levels According to Age

Normal Phosphorus Levels

_____________________ (mg/dL

(mg/dL)

Premature (<1 week)

Full-term (<1 week)
Child
Adult

6-10
7-12
8.0-10.5
8.5-10.5

100%

4.2-9.0
3.8-6.2
3.5-6.8
3.0-4.5
Newborn
1 year 25 yr

Adult

386

R. Ariel Gomez

maintain a positive K+ balance which is necessary for

growth. Conservation of K+ during infancy is reflected
also by higher serum K+ values (Table 5). The newborn
has a low basal rate of K+ excretion and a decreased ability
to rapidly excrete a K+ load, in comparison with the adult
(Table 5). The fractional excretion of K+ in the newborn
is about half of that in the adult (Table 5). This may be
due to reduced K+ secretion by the cortical collecting
duct, the main site for K+ excretion. The factors that limit
urinary K+ excretion by the principal cells during early life
are: (1) unfavorable electrochemical gradient (low cellular
K+ concentration, decreased Na, K-ATPase activity and
transepithelial voltage), (2) limited membrane permeability to K+, (3) low tubular fluid flow rates, and (4) decreased
sensitivity to mineralocorticoids. Also, enhanced K+ absorption by intercalated cells of the medullary collecting
ducts may be a contributory factor. In fact, medullary collecting tubules (which reabsorb K+) mature earlier than
the principal cells of the cortical collecting duct (which
secrete K+), a factor that may be important in the decreased ability to excrete K+ during early life. Also, as
GFR increases with maturation, fluid delivery to the cortical collecting duct increases promoting K+ secretion. Under normal conditions, the reduced ability of the neonate
to excrete K+ is not manifested clinically. However, hyperkalemia can develop if the neonate is exposed to excessive
exogenous or endogenous (i.e., cell breakdown) K+ loads.
Acid-Base Balance
Infants have lower blood pH and HCO3 than older
children and adults (Table 6). Overall, the newborn kidney
is capable of maintaining normal acid-base status. However, infants are prone to the development of acidosis when
they are sick or receive inadequate nutrition or in response
to an exogenous acid load. The plasma concentration of
HCO3 increases with age, due to an increase in the renal
threshold of HCO3~ with maturation. The approximate
renal threshold of HCO3~ is 18 mEq/L in the premature
and 21 mEq/L in the term neonate. Adult values (24 to
26 mEq/L) are reached by 1 year of life. Several factors
seem to be responsible for the low renal HCO3~ threshold
observed in infancy: low activity of carbonic anhydrase,
extracellular fluid volume expansion contributing to proxi-

TABLE 5
Plasma Levels and Excretion of Potassium in Normal Infants
and Children"

Age
(years)

Plasma (K)
(mEq/liter)

K Clearance
(ml miir1 1.73 nr2)

0-0.3
0.4-1.0
3-10
11-20

5.2 0.8
4.9 0.5

53
14 6

4.2 0.5
4.3 0.3

20 11
21 8

FEK (%)

8.5 3.8
14.6 5.0
14.5 8.9
16.2 8.2

" From Jones DL and Chesney RW: Tubular function. In: Holliday MA, Barrett TM, Avner ED (eds). Pediatric Nephrology, 3rd
ed. Williams and Wilkins, Baltimore, pp. 117-147, 1993.

TABLE 6
Acid-Base Measurements as a Function of Age"

Age
Preterm (1 week)

Preterm (6 weeks)
Term (birth)
Term (1 hr)

3-6 months
21-24 months
3.5-5.4 years
5.5-12 years
12.5-17.4 years

Adult males

pH
7.34 0.06

7.38 0.02
7.24 0.05*
7.37 0.05
7.39 0.03
7.40 0.02
7.39 0.04
7.40 0.03
7.38 0.03
7.39 0.01

paCO 2
(mm Hg)
31 3*
35 6
49 10*
34 9
36 3
35 3
37 4
38 3
41 3
41 2

HCOj
(mEq/liter)
17.2 1.2*
21.9 4.4*

20.0 2.8*
19.0 2.3*
22.0 1.9*
21.8 1.6*
22.5 1.3*
23.1 1.2*
24.0 1.0*
25.2 1.0

" From Schwartz GJ: General principles of acid-base physiology. In: Holliday MA, Barrett TM, Avner E (eds). Pediatric Nephrology, 3rd ed. Williams and Wilkins, Baltimore, pp. 222-246,
1993.
Mean 1 SD. * Significantly different from adult males (p <
.05) by Tukey's test.

mal HCO3 wasting, immaturity of luminal Na -H+ e xchange, and imbalance between filtration and reabsorption
of HCO3 in newly formed nephrons. Also, HCO3~ reabsorption in the distal nephron may not compensate for
HCO3 that escapes the proximal tubule.
Infants have a decreased ability to acidify the urine when
compared to adults under basal conditions. Maximal titratable acid and ammonium excretion increase with age and
achieve adult values (when corrected by GFR) by 2 months
of age. Several factors are responsible for the relative inability of the newborn to excrete an acid load including a
limited availability of urinary buffers such as phosphate and
ammonium. The low GFR and high tubular reabsorption of
phosphate (TRP) prevailing in the neonatal period markedly limit the phosphate available as a urinary buffer. Prior
to 2 months of age, ammonia generation and secretion
are also limited. Immaturity of the collecting ducts (fewer
intercalated cells, fewer proton pumps per cell) and immaturity of carbonic anhydrase activity could also limit distal
acidification. In addition to this impairment in renal excretion infants have an increased exogenous and endogenous
proton load in comparison to adults. In prematures, endogenous acid production can be as high as 2 to 3 mEq kg'1
day'1. A large amount of acid is generated during the
metabolism of proteins. Accretion of calcium into the growing bone results in the release of 0.5 to 1.0 mEq kg"1
day" 1 of H+ that needs to be excreted by the kidney or
neutralized by HCOs absorbed through the gastrointestinal tract. For this reason is during episodes of gastroenteritis, infants are susceptible to metabolic acidosis.
Blood Pressure
The normal blood pressure values vary with age, gender,
stature, and degree of maturation. Accepted normal values
are based on demographic studies. The reader should consult the report of the "Second Task Force on Blood Pres-

59. The Kidney in Infants and Children

387
TABLE 7
Classification of Hypertension by Age Group"

Age group

Significant hypertension

Severe hypertension

Systolic a 96 mm Hg
Systolic a 104 mm Hg
Systolic a 112 mm Hg
Diastolic a 74 mm Hg

Systolic a 106 mm Hg
Systolic a 110 mm Hg
Systolic > 118 mm Hg
Diastolic a 82 mm Hg

Children (3-5 yr)

Systolic a 116 mm Hg
Diastolic a 76 mm Hg

Systolic a 124 mm Hg
Diastolic a 84 mm Hg

Children (6-9 yr)

Systolic a 122 mm Hg
Diast olic a 78 mm Hg

Systolic > 130 mm Hg

Diastolic a 86 mm Hg

Children (10-12 yr)

Systolic a 126 mm Hg
Diastolic a 82 mm Hg

Systolic a 134 mm Hg
Diastolic a 90 mm Hg

Adolescents (13-15 yr)

Systolic a 136 mm Hg
Diastolic a 86 mm Hg
Systolic a 142 mm Hg
Diastolic a 92 mm Hg

Systolic a 144 mm Hg
Diastolic a 92 mm Hg
Systolic a 150 mm Hg
Diastolic a 98 mm Hg

Newborn
7 days 830 days
Infant (< 2 yr)

Adolescents (16-18 yr)

" From the Report of the Second Task Force on Blood Pressure Control in Children:
Pediatrics 79(l):l-25, 1987.

sure Control in Children1987" which provides normative

data and guidelines for detection of children with hypertension. A child is considered hypertensive if the systolic or
diastolic blood pressure is above the 95th percentile for
age and sex on at least three occasions. The classification
of hypertension by age group is shown in Table 7. The
etiology of hypertension varies with the age of the patient.
As a rule, the younger the patient, the more likely it is for
hypertension to be secondary. In the newborn period, renal
artery thrombosis, congenital renal malformations, coarctation of the aorta, and bronchopulmonary dysplasia are
the most common causes. In children between infancy and
6 years of age, renal parenchymal diseases, renal artery
stenosis, and coarctation of the aorta are leading cause.
Between 6 and 10 years of age, renal parenchymal diseases,
renal artery stenosis, and essential hypertension are frequently found. In the adolescent, essential hypertension
and renal parenchymal diseases are the two most prominent causes.
Bibliography
Al-Dahhan J, Haycock GB, Nichol B, Chantler C, Stimmler L:
Sodium homeostasis in term and pre-term neonates. III. Effect
of salt supplementation. Arch Dis Child 59:945-950, 1984.

Chevalier RL: Developmental renal physiology of the low birth

weight pre-term newborn. J Urol 156:714-719, 1996.
Ekblom P: Embryology and prenatal development. In: Holliday
MA, Barrett TM, Avner E (eds) Pediatric Nephrology, 3rd ed.
pp. 2-20, Williams & Williams, Baltimore, 1993.
Gomez RA: Postnatal regulation of water and electrolyte excretion. In Brace RA, Ross MG, Robillard JE (eds) Reproductive
and Perinatal Medicine, Vol. XL Fetal and Neonatal Body Fluids:
The Scientific Basis for Clinical Practice. Perinatology Press,
New York, pp. 307 -318, 1989.
Greene MG: The Harriet Lane Handbook. A Manual for Pediatric
House Officers. 12th ed. Mosby -Year Book, St. Louis, pp. 1434, 1991.
Jones DP, Chesney RW: Tubular function. In Holliday MA,
Barrett TM, Avner ED (eds), Pediatric Nephrology, 3rd ed.
Williams & Wilkins, Baltimore, pp. 117-147, 1993.
Mathew OP, Jones AS, James E, Bland H, Groshong T: Neonatal
renal failure: Usefulness of diagnostic indices. Pediatrics 65:5760, 1980.
Polacek E: The osmotic concentrating ability in healthy infants
and children. Arch Dis Child 40:291, 1965.
Report of the Second Task Force on Blood Pressure Control in
Children1987. Pediatrics 79(1): 1-25, 1987.
Schwartz GJ, Brion LP, Spitzer A: The use of plasma creatinine
concentration for estimating glomerular filtration rat e in infants,
children and adolescents. Pediatric Clin North Am 34:571590, 1987.

THE KIDNEY IN PREGNANCY

SUSAN H. HOU

CHANGES IN RENAL FUNCTION DURING PREGNANCY

and followed by monthly screening urine cultures. All patients with symptomatic pyelonephritis should be hospitalized and treated with intravenous antibiotics. They should
receive intravenous antibiotics until they have been afebrile
for 24 hours at which time they can be switched to oral
antibiotics. Single, large-dose therapy with an antibiotic
which may be effective in nonpregnant patients should not
be used in pregnant women. Chronic suppressive therapy
should be used for the duration of the pregnancy after
recovery from acute pyelonephritis.

Dramatic changes in renal function occur during pregnancy. Renal plasma flow increases to 50 to 70% above
normal during the first two trimesters and remains 40%
above normal in the third trimester. An increase in glomerular filtration rate (GFR) begins by the fourth week of
gestation and peaks at 150% of normal at 13 weeks (Fig.
1). An appreciation of this change is essential for evaluation
of renal disease in pregnant women, because the average
BUN for pregnancy is 9 mg/dL and the average serum
creatinine is 0.7 mg/dL. A BUN of 14 mg/dL or higher or
a serum creatinine of 0.9 mg/dL or higher is indicative of
abnormal renal function. GFR varies with position, particularly late in pregnancy when it may be 20% lower in the
supine position than in the lateral recumbent position. The
increased GFR leads to increased clearance of uric acid
and a decrease in serum uric acid levels the mean being
3 mg/dL. Tubular function also changes. The threshold for
reabsorption of glucose increases, and glycosuria may occur
without hyperglycemia. The osmostat is reset such that
thirst is experienced at a level 10 mOsm below the nonpregnant normal and serum sodium is about 5 mEq/L below
nonpregnant normal.
Anatomical changes occur in the kidney during pregnancy as well. Renal length is increased by 1 to 1.5 cm. A
physiologic dilatation of the ureters occurs, giving rise to
hydronephrosis. Stasis of the urine in the ureters predis poses to urinary tract infection and complicates timed
urine collections.

Hypertension

In normal pregnancy, blood pressure drops in the first

two trimesters as a result of the combination of arterial
vasodilatation mediated by the production of vasodilatory
prostaglandins, and by refractoriness to the pressor effects
of angiotensin. Hypertension in pregnancy is defined as
any rise in systolic blood pressure of greater than 30 mm
Hg or any rise in diastolic blood pressure greater than
15 mm Hg above baseline, or by the use of antihypertensive
medications. The blood pressure may rise by 10 mm Hg
in the third trimester in normal pregnancy, but a blood
pressure of greater than 140/85 is abnormal even in the
third trimester. Blood pressures greater than 125/75 before
32 weeks gestation have been associated with an increase
in the frequency of adverse fetal outcome.
The Joint National Committee on Detection, Evaluation
and Treatment of High Blood Pressure defines four types
of hypertension in pregnancy: preeclampsia-eclampsia,
chronic hypertension, chronic hypertension with superimposed preeclampsia, and transient hypertension (Table 1).
Hypertension can also complicate pregnancy in women
with virtually any type of preexisting renal disease.
Preeclampsia is a disease that occurs only in pregnancy.
It includes the triad of hypertension, proteinuria, and
edema and occurs after the 20th week of gestation. Hyperreflexia is characteristic of the disease. Most affected
women have a mild to moderate decrease in GFR with
serum creatinine levels inappropriately high for pregnancy.
Serum uric acid levels are elevated and hematocrit may
rise as a result of hemoconcentration. Pathologically, preeclampsia is characterized by swelling of glomerular epithelial cells termed glomerular endotheliosis (Fig. 2). It occurs
primarily in first pregnancies and biopsy studies have found
that 75% of the multiparous women given the diagnosis of

Urinary Tract Infection

Urinary tract infection is the most common renal disease

occurring during pregnancy. Asymptomatic bacteriuria,
which occurs in 5% of pregnancies, has been associated
with small for gestational age babies and some argue that
it plays a role in hypertension and anemia. If left untreated,
30% of women with asymptomatic bacteriuria develop pyelonephritis. Treatment of asymptomatic bacteriuria reduces the incidence of pyelonephritis in this group to 3%.
Pyelonephritis is a serious disease in pregnant women. Ten
percent are bacteremic and a transient decrease in GFR
is common. Pyelonephritis has been associated with premature labor and intrauterine fetal death. Asymptomatic bacteriuria should be treated with a 10-day course of antibiotics

Primer on Kidney Diseases, Second Edition

388

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

60. The Kidney in Pregnancy

389

100

%
ERPF
increase

50

--------- GFR
n=25
Non
pregnant

16
2
6

36
Weeks of pregnancy

FIGURE I Increase in glomerular filtration rate and effective

renal plasma flow during pregnancy. From Davison JM: Am J
Kidney Dis 9: 248, 1987. With permission.

TABLE I

Hypertensive Disorders of Pregnancy

Preeclampsia
Primigravida
Onset after 20 weeks gestation
Proteinuria Generalized edema
Patients at extremes of childbearing years
Hyperreflexia
Vasospasm on fundoscopic exam
Multisystem involvement No f
hypertension at follow- up
Chronic hypertension
May occur in any pregnancy Before or
anytime during pregnancy No
proteinuria
May have mild pretibial edema
More common in older age group
May have eye ground changes of essential hypertension
Normal reflexes | risk of placental separation
Hypertension at follow- up
Chronic hypertension with superimposed preeclampsia
30% of women with chronic hypertension Hypertension
before or early in pregnancy | 30 mm Hg systolic, 15 mm
Hg diastolic after 20 weeks
gestation
Proteinuria
May occur in any pregnancy More
common in older age group
Hyperreflexia Generalized edema
Hypertension at follow- up
Transient hypertension No hypertension
prior to pregnancy Third trimester or 24
hours postpartum No proteinuria No
hyperreflexia Predicts hypertension
later in life

FIGURE 2 Glomerulus from a 29-year-old, woman in the 22nd

week of pregnancy who had developed 3.6 g/24 hr proteinuria and
hypertension (preeclampsia). The glomerulus has endotheliosis
with obliteration of most capillary lumens by swollen endothelia
cells (Masson trichrome stain). Courtesy of Dr. J. Charles Jennette.

preeclampsia on clinical grounds have some other renal

disease.
The acronym HELLP (hemolysis, elevated liver enzymes, low platelets) syndrome has been used to describe
severe preeclampsia with manifestations of multisystem
involvement. In addition to its hepatic and hematologic
manifestations, HELLP syndrome may be complicated by
disseminated intravascular coagulation, acute renal failure,
cerebrovascular accident, retinal detachment or heart failure (Table 2). The most serious complication of this condition is eclamptic seizures which may result in maternal
death. Anticonvulsant and antihypertensive medication
may be used to protect the mother while preparing for
delivery. Magnesium sulfate has recently been shown to

TABLE 2

Clinical Manifestations of HELLP Sydrome- b

Maternal complications
Disseminated intravascular coagulopthy
Abruptio placentae
Acute renal failure
Severe acites
Pulmonary edema
Cerebral edema
Retinal detachment
Laryngeal edema
Subcapsular hematoma of the liver
Adult respiratory distress syndrome

21%
16%
8%
8%
6%
1%
1%
1%
1%
1%

Laboratory findings
Platelet count (no/mm3 )
Serum aspartate amino transferase (U/L)
Lactic dehydrogenase (U/L)
Total bilirubin (mg/dL)
Serum creatinine (mg/dL)

7000-99,000
70-6193
564-23,584
0.5-25.5
0.6-16.0

" Data derived from Sibai et al.: Am J Obstet Gynecol 169:

1002-1003, 1993. b n = 442.

390
be more effective than phenytoin in preventing seizures in
preeclamptic women. Intravenous hydralazine and labetolol are effective and safe for treatment of severe hypertension. However, delivery is the definitive treatment, although the manifestations of preeclampsia may appear or
worsen for several days postpartum.
The major mechanism by which preeclampsia causes tis sue damage is intense vasoconstriction. Normal pregnancy
is characterized by resistance to the pressor effects of angio tensin. This resistance is lost in preeclampsia. The production of both the vasodilatory prostaglandin prostacyclin
and the vasoconstrictor thromboxane by vascular endothelial cells are increased in pregnancy. In normal pregnancy,
the effects of prostacyclin predominate. In preeclampsia
the ratio of the two shifts toward thromboxane and vasoconstriction results. There is evidence in small studies of
women at high risk for preeclampsia that low-dose aspirin
prevents its development by inhibiting thromboxane formation. This finding has not been confirmed in several
prospective randomized studies in large numbers of
women. Preeclampsia occurs primarily in first pregnancies.
Chronic hypertension can be diagnosed when hypertension is present before pregnancy or before 20 weeks gestation. Occasionally the diagnosis is difficult when a woman
first seeks prenatal care after 20 weeks gestation or when
hypertension is masked by the pregnancy associated drop
in blood pressure which occurs in the first trimester. Thirty
percent of women with essential hypertension experience
such a drop. Signs such as left ventricular hypertrophy or
hypertensive eye ground changes support the diagnosis
of chronic hypertension. In the absence of superimposed
preeclampsia, pregnant women with chronic hypertension
do not have proteinuria. If proteinuria appears in a woman
with essential hypertension, she should be presumed to
have superimposed preeclampsia. Women who develop hypertension for the first time during pregnancy should be
screened for pheochromocytoma. Although rare, pheochromocytoma carries a 50% maternal mortality rate if the
diagnosis is not made.
There is debate about the effect of chronic hypertension
on pregnant women and their infants and about the target
blood pressure. Most physicians aim for systolic blood pres sures below 140 and diastolic blood pressures between 90
and 100. Nonpharmacologic treatment should include bed
rest and avoidance of alcohol. Salt restriction is not indicated and weight loss is contraindicated.
When pharmacologic treatment is needed, the most commonly used drugs include methyldopa, a variety of /3blockers, labetolol, and hydralazine. The use of calcium
channel blockers is increasing. Methyldopa in doses up to
3 g/day in divided doses has been used for 35 years; no
serious problems have been identified in children exposed
to the drug in utero in intensive follow-up testing. Labetolol
and /3-blockers are the other first-line drugs. Early reports
of fetal growth restriction and inability to tolerate anoxic
stress in the infants of mothers treated with /8-blockers
have not been born out by widespread use. If these firstline drugs are ineffective in controlling blood pressure,
hydralazine in doses up to 200 mg bid can be added to the

Susan H. Hou

regimen. Clonidine has been shown in a single controlled

trial to be as effective as methyldopa. Although calcium
channel blockers have been widely used, there are a few
reservations. In some studies, fetal loss rate has been high,
although pregnancy outcome is better than predicted by
previous obstetric history in the same patients. Nifedipine
may result in profound hypotension when used in conjunction with magnesium sulfate in women with preeclampsia.
Angiotensin converting enzyme (ACE) inhibitors should
not be used in pregnancy. In animal studies, captopril has
been associated with 80 to 93% fetal loss. The use of ACE
inhibitors in humans has been associated with oligohydramnios resulting in neonatal death from hypoplastic lungs and
neonatal anuria resulting in death or the need for dialysis.
Superimposed preeclampsia may complicate pregnancy
in women with chronic hypertension. The woman with
chronic hypertension should be monitored every week after
the 20th week to look for the development of worsening
hypertension, proteinuria, hemoconcentration, elevated
transaminases, thrombocytopenia, elevated uric acid, and
decreased serum albumin. If it appears that she is developing superimposed preeclampsia, she should be hospitalized
and delivered for the same indications as women with simple preeclampsia.
Transient hypertension develops in the third trimester
and resolves postpartum. It is not accompanied by any of
the other manifestations of preeclampsia and is a strong
predictor of chronic hypertension later in life.
CHRONIC RENAL DISEASE IN PREGNANCY
Hypertension

The most serious problem complicated pregnancy in

women with renal disease is hypertension, which can be
seen even in the absence of severe renal insufficiency. In
a 1980 report of 121 pregnancies in 89 women with a wide
variety of renal disease and moderately well-preserved renal function, 35% of pregnancies were complicated by a
blood pressure of > 140/90. In more than half of the hypertensive pregnancies, diastolic blood pressure exceeded
110 mm Hg. Hypertension was more common in women
with glomerular diseases and hypertensive nephrosclerosis
than in women with interstitial diseases.
Conception is less common in women with severe renal
insufficiency, but enough data are available to determine
that hypertension is common in this group as well. A 1996
report contain information on 82 pregnancies in 67 women
with serum creatinine 1.4 mg/dL or greater at conception
or in the first trimester. Forty-eight percent had a mean
arterial blood pressure greater than 105 mm Hg in the
third trimester, compared to 28% in the first trimester.
Hypertension occurs in approximately 60% of pregnancies
in dialysis patients. In six reports of pregnancies in renal
transplant recipients, there were 46 instances of hypertension in 135 pregnancies. Although none of these studies
report any maternal death from hypertension, there are
reports of progression of preeclampsia to eclampsia.

391

60. The Kidney in Pregnancy

The approach to the control of hypertension during pregnancy in women with renal disease is similar to the approach to treatment of women with chronic hypertension,
except that the hypertension of renal disease has a component of volume overload which may require salt restriction,
the judicious use of diuretics, or, in severe renal insufficiency, dialysis. Occasionally, volume status can be uncertain enough that central pressure monitoring is necessary.
Women with renal disease and hypertension are at risk
for superimposed preeclampsia, but the recognition of this
complication may be difficult. Even in pregnancies in
women without known renal disease, discrepancies between clinical diagnosis and pathological findings frequently occur. In the setting of renal disease, many clinical
signs and symptoms occur that simulate preeclampsia, thus
further complicating the diagnostic process. For example,
as discussed below, proteinuria is expected to increase in
women with glomerular disease. Plasma uric acid levels
may be elevated in women with decreased GFR, obscuring
the development of preeclampsia. Even worsening of hypertension or decrease in GFR an occur in the absence of
preeclampsia. In this context, the development of hyperreflexia is useful in making the distinction as it does not
generally develop except in preeclampsia. The occurrence
of elevated liver enzymes, thrombocytopenia, and microangiopathic changes on the peripheral blood smear, which
almost certainly signal the development of preeclampsia,
are late signs. However, it is hoped that diagnosis and
delivery will not be delayed until these signs develop. The
diagnostic and therapeutic dilemma is usually resolved if
blood pressure cannot be adequately controlled. In these
circumstances, the pregnancy is usually terminated. Unfortunately in some series which did not specifically address
women with renal disease, eclampsia occurred with a diastolic blood pressure as low as 80 mm Hg. Because the use
of magnesium sulfate is dangerous in women with severe
renal insufficiency, an alternative anticonvulsant should be
used. If magnesium is used, the loading dose should be
given in two separate infusions with measurement of serum
magnesium level between the two infusions. Serum magnesium should be measured every 2 hours, and additional
magnesium given only when the level falls below 4 mg/dL.
Continuous infusion should not be used with moderate
renal insufficiency, and measurement of serum magnesium
every 2 hours is necessary in women even with mild decrements in GFR.
Proteinuria
Pregnancy in women with all types of renal disease is
associated with an increase in proteinuria. Nephrotic syndrome frequently develops in women who excreted only
a small amount of protein prior to pregnancy, particularly
in women with glomerular diseases. The increased proteinuria is thought to result from an increased filtered load of
protein, a change that occurs to a lesser extent in normal
pregnant women. Heavy proteinuria by itself, without hypertension or worsening renal function, is not associated
with poor pregnancy outcome. When heavy proteinuria

develops, salt restriction may decrease edema formation.

Massive edema frequently develops and may be disabling
for the mother. Even in this situation, diuretics are best
avoided. The risk of low doses of diuretics is relatively low
in the absence of preeclampsia. High doses are generally
required in women with nephrotic syndrome, and little is
known about the effect of high doses of diuretics on the
fetus. Diuresis may aggravate intravascular volume contraction in the mother and increase the risk of thromboembolic events. If bed rest is required in a woman with nephrotic syndrome, low-dose heparin should be used
because of the increased risk of thromboembolic disease.
PREGNANCY IN WOMEN WITH RENAL DISEASE AND
GOOD RENAL FUNCTION

The most important determinant of whether renal function declines during pregnancy is renal function at conception. If renal function is normal at the time of conception,
pregnancy does not accelerate the progression of renal
disease. A 1995 study reported the effect of pregnancy
on the progression of renal disease in 171 women with
glomerulonephritis who became pregnant after the diagnosis of renal disease as compared to 189 women of childbearing age with glomerulonephritis who did not conceive after
the onset of renal disease. There was no difference in the
risk of progression to end stage renal disease between the
two groups. Hypertension and certain histological subgroups were at increased risk for progression to renal failure, but in no subgroup did pregnancy increase the risk.
If renal function is well preserved, women with renal dis ease can undertake pregnancy without increased risk of
progression to end stage renal disease, but other complications such as hypertension and increased proteinuria are
common.
The outcome for infants is somewhat worse than for the
general population. Twenty percent of infants are born
prematurely and approximately 10% of pregnancies which
continue beyond the second trimester end in stillbirth or
neonatal death.
PREGNANCY IN WOMEN WITH CHRONIC
RENAL INSUFFICIENCY

Women who conceive when the serum creatinine is

greater than 1.4 mg/dL have an increased risk that their
renal function will deteriorate more rapidly than if they
do not become pregnant. In 43% of the 82 pregnancies in
women with moderate to severe renal insufficiency mentioned above there was a pregnancy associated decline in
renal function. In 31% of pregnancies, deterioration in renal
function persisted 6 months after pregnancy, and in 10%
there was a deterioration between 6 weeks and 6 months
postpartum. At 12 months postpartum, eight women, seven
of whom had had a pregnancy associated decline in renal
function, had reached end stage renal disease. Deterioration in renal function was more common in women who
were hypertensive during the third trimester, but the pres sure of a normal blood pressure did not guarantee preserva-

392

Susan H. H o u

tion of renal function. A greater degree of renal insufficiency at conception was associated higher with a likelihood
of acceleration of disease during pregnancy.
Despite maternal complications, fetal survival was good
(93%). Other studies indicate that even in women with
renal failure severe enough to require the initiation of
dialysis during pregnancy, fetal survival is 74%.
DIABETIC NEPHROPATHY

The effect of pregnancy on diabetic nephropathy depends on renal function at the time of conception as it does
in other renal diseases. Two studies including 57 pregnancies in women with diabetic nephropathy found no evidence that pregnancy accelerates the course of the disease.
However, these studies included only a handful of women
with seriously impaired renal function. A 1996 report of
pregnancies in 11 women with diabetic nephropathy and
serum creatinine equal to or greater than 1.4 mg/dL found
an acceleration in the progression to end stage renal disease
when compared to the rate of decline of renal function
prior to pregnancy and when compared to a control group.
LUPUS NEPHRITIS

Lupus nephritis is most common in women of childbearing age. Few diseases are capable of producing such fulminant but potentially treatable renal failure. Approximately
half of 365 women with preexisting lupus nephritis in 19
reports suffered an exacerbation of their lupus during pregnancy. While lupus flares are less common in women who
have been in remission for more than 6 months, one third
of women in this group suffer exacerbations of their lupus
during pregnancy. When lupus is active at the time of conception, the risk of renal failure is higher. Lupus flares are
less common in women with mesangial proliferation on
biopsy than in women with diffuse proliferative disease.
Despite its more indolent course in patients who do not
become pregnant, membranous lupus is as likely to give
rise to a flare during pregnancy as is diffuse proliferative
disease. Prednisone and azathioprine can be used in pregnancy with the same precautions that apply to transplant
recipients. Cyclophosphamide is teratogenic when used
during the first trimester. Later in pregnancy, there is concern about the risk of fetal bone marrow suppression. There
have been a few case reports of thyroid cancer in children
exposed to Cyclophosphamide in utero. Therapeutic abortion is advisable if Cyclophosphamide must be used during
the first trimester.
Termination of pregnancy is often recommended in
women with renal failure from lupus nephritis. Although
therapeutic abortion is not accompanied by an increase in
lupus flares when it is done before the exacerbation of
disease, it does not carry any guarantee of controlling the
disease once it has become active. If lupus itself or highdose steroid therapy gives rise to uncontrollable hypertension, it may be necessary to terminate the pregnancy to
facilitate blood pressure control. Fetal loss reaches about
50% in women with lupus nephritis and renal insufficiency.

Fetal loss may occur because of hypertension, re nal failure,

or placental vasculopathy. If the fetus is beyond 34 weeks
gestation when a severe lupus flare or renal failure occurs,
the baby should be delivered, because fetal distress and
demise may occur so rapidly that they are not anticipated
by antenatal testing. The fetus that is viable but not yet
mature should be monitored on a daily basis with non stress
tests and biophysical profiles and delivered if evidence of
fetal distress develops. The infants should be monitored
for several manifestations of neonatal lupus. Skin rashes
and thrombocytopenia may occur because of antibodies
that cross the placenta. These usually resolve over weeks
to months as maternal antibodies are lost. Permanent congenital heart block can develop as a result of antigen antibody deposition in the conducting system of the fetus.
PREGNANCY IN DIALYSIS PATIENTS

Infertility is the rule in chronic dialysis patients. Approximately 0.5% of women of childbearing age treated with
chronic dialysis conceive each year. Although it is common
to increase the total amount of dialysis in women who
become pregnant, the benefit of this practice has not yet
been demonstrated and is currently under investigation.
Experience with more than 300 pregnancies in dialysis patients in the US has shown no advantage of one dialysis
modality over another. Pregnancy in dialysis patients is
accompanied by worsening anemia. Although recombinant
human erythropoietin has been used without ill effects in
a handful of pregnant dialysis patients the dose must be
increased to maintain the prepregnancy hematocrit. The
most serious complication faced by the pregnant dialysis
patient is hypertension, which can be life threatening. Intensive monitoring and aggressive antihypertensive treatment is necessary for the safety of the mother. The only
category of antihypertensive drugs clearly contraindicated
in pregnancy is ACE inhibitors which are associated with
neonatal anuria and neonatal death from hypoplastic lungs.
Thirty to forty percent of pregnancies in women who
conceive after starting dialysis result in surviving infants.
Twenty percent of pregnancy losses result from second
trimester spontaneous abortions. The children born to dialysis patients are almost always premature and frequently
are small for gestatio nal age. The best chance for successful
pregnancy in women with end stage renal disease is renal
transplantation.
PREGNANCY IN RENAL ALLOGRAFT RECIPIENTS

Fertility is often restored in renal transplant recipients

and there have been more than 6000 pregnancies reported
in women with renal transplants. Nonetheless, there are
numerous problems associated with such pregnancies.
Transplant patients are generally advised to wait 1.5 to 2
years after successful transplantation and then to undertake
pregnancy only if renal function is stable with a serum
creatinine of 2 mg/dL or less. Other favorable conditions
are the absence of hypertension, significant proteinuria,
renal pelvocaliceal distension, or evidence of graft rejection

393

60. The Kidney in Pregnancy

and the requirement for only low doses of immunosuppres sives. Women whose graft function is impaired at the time
of conception have a substantial risk of worsening renal
function. There are occasional instances where irreversible
loss of graft function has occurred in women with good
stable renal function and normal blood pressure prior to
and throughout pregnancy.
When renal function deteriorates in a transplant recipient during pregnancy, the differential diagnosis includes
cyclosporine or tacrolimus toxicity, rejection, recurrence
of the primary disease and preeclampsia, as well as other
pregnancy-related causes of renal failure. Cytomegalovirus
infection may be related to graft dysfunction.
Exposure in utero to immunosuppressive drugs is a concern in the infants of transplant recipients. Ideally, women
should wait to conceive until the prednisone dose is 15 mg/
day or lower. Maternal blood levels of prednisolone are
ten times the levels in cord blood. Nonetheless, adrenal
insufficiency occasionally occurs in the infant Even in
infants who appear to be stable should be monitored in a
high-risk nursery for several days after birth. Steroid doses
should be reduced at the same rate as would be done in a
nonpregnant patient. High-dose steroids can safely be used
for the treatment of rejection.
Azathioprine requires conversion in the liver to its active
metabolite, 6-mercaptopurine, and then to 6-thiosinic acid.
During the first trimester when organogenesis is taking
place, the fetus lacks the enzyme inosinate pyrophosphory lase necessary for this conversion, thus affording the fetus
some protection from the effects of the drug. The frequency
of congenital anomalies is only slightly increased in women
taking azathioprine, if at all. It has been found empirically
that the risk of fetal leukopenia can be minimized by keeping maternal leukocyte count over 8500 mm" 3. Azathioprine may contribute to growth retardation in the 20 to
50% of infants who are small for gestational age.
Cyclosporine crosses the placenta easily. Although it has
not been associated with congenital anomalies, it does appear to cause more severe growth retardation than azathioprine. The serum creatinine at conception is generally
higher in women taking cyclosporine than in women taking
only prednisone and, azathioprine, and hypertension is
worse in these women. Cyclosporine levels may vary unpredictably during pregnancy and should be monitored on a
weekly basis.
Although pregnancy in transplant recipients is complicated, the fetal outcome is good, with 92% surviving infants
in pregnancies that go beyond the first trimester.
ACUTE RENAL FAILURE IN PREGNANCY

Acute renal failure (ARF) remains an unusual but lifethreatening complication of pregnancy. ARF requiring dialysis occurs in less than 1 in 10,000 to 15,000 pregnancies
in industrialized countries, whereas the frequency of milder
degrees of renal insufficiency is higher. The greatest risk
of pregnancy-related ARF is late in pregnancy. At one
time, sepsis following illegal abortion accounted for a large
proportion of obstetric ARF, making renal failure more

common early in pregnancy. Sepsis following illegal abortion remains a common cause of renal failure in countries
with poor access to health care. While many organisms have
been implicated in causing postabortal sepsis, Clostridium
welchii accounts for a disproportionate number of the cases
of acute renal failure, because it produces a toxin which
causes hemolysis and renal failure.
Obstruction
Although obstruction is uncommon as a cause of ARF
in pregnancy, it must be considered when ARF occurs in
any setting because of its reversibility. Renal stone formation is not increased during pregnancy, despite an augmentation of calcium absorption from the gut and calcium excretion by the kidney. However, when severe abdominal
pain occurs during pregnancy, renal calculi should be considered. Stones can lead to renal failure if they are bilateral
or occur in a solitary kidney.
Ureteral obstruction by the gravid uterus is unique to
pregnancy. The description of this problem is still limited
to a number of case reports. Polyhydramnios is a common
feature, and several episodes of renal failure have occurred
in patients with a solitary kidney. Ultrasound for the diagnosis of obstruction is difficult to interpret because of the
normal physiologic dilatation of the collecting system. Intravenous or retrograde pyelography may be necessary to
make the diagnosis despite the radiation exposure to the
fetus.
Preeclampsia-Eclampsia
Renal insufficiency with a mild increase in serum creatinine above normal for pregnancy is common in preeclamp sia. In most instances, renal function returns to baseline
quickly after delivery. Persistent renal dysfunction and
frank ARF occur in a minority of cases, and then only
when the disease is very severe prior to delivery. Although
ARF is an uncommon complication of preeclampsia, preeclampsia is common enough that it accounts for a substantial percentage of the cases of pregnancy-associated ARF.
The course is similar to the course of acute tubular necrosis
(ATN) from other causes, and recovery of renal function
is the rule.
Acute Tubular Necrosis and Cortical Necrosis
Common causes of ARF are major obstetric complications including hemorrhage, abruptio placentae, amn iotic
fluid embolism, and retained dead fetus. In many series,
obstetric patients have a better prognosis for survival than
other patients with ATN.
Although ATN is the most common pathologic finding,
most series report a greater frequency of cortical necrosis
in obstetric patients than in other patients with ARF. Obstetric patients account for more than half of the cases of
cortical necrosis. With improved obstetric care, this entity
has become extremely rare.
Both arteriogram and renal biopsy have been used to
distinguish between ATN and cortical necrosis. In patients

394

Susan H. Hou

with cortical necrosis, biopsy shows fibrin thrombi in the

glomerular capillaries with widespread necrosis of the glomeruli, sometimes sparing the juxtamedullary glomeruli.
Re renal arteriogram in cortical necrosis shows an inhomo geneous or absent cortical nephrogram. As there is no
specific treatment for cortical necrosis, establishing diagnosis is useful only for prognostic purposes. Permanent renal
failure and partial recovery are common.
Hemolytic Uremic Syndrome
The hemolytic uremic syndrome (HUS, see chapter 32)
is an unusual cause of renal failure. Its usual occurrence
is several days to 10 weeks after a normal pregnancy. Severe
renal failure is associated with microangiopathic hemolytic
anemia, thrombocytopenia, and variable neurological
symptoms. It is part of a spectrum of diseases that includes
thrombotic thrombocytopenic purpura in which neurological symptoms predominate and renal failure is usually less
severe. Both pregnancy and oral contraceptives have been
associated with .HUS. The primary event in the development of the disease appears to be endothelial damage.
Many of the factors which have been identified as inciting
events for HUS cause endothelial cell damage in tissue
culture. Endothelial cell damage in turn leads to the disruption of the cell processes which prevent thrombogenesis,
with the result being the formation of microthrombi in the
vessels of many organs.
The prognosis for postpartum HUS is poor with a
high mortality rate from central nervous system disease
or uncontrollable bleeding. Permanent renal failure is
the rule in survivors. The observation that prognosis is
improved in TTP with plasma exchange has led to its
use in HUS and there are anecdotal reports of a return
of function.
Acute Fatty Liver of Pregnancy
Acute fatty liver of pregnancy is an unusual complication
of pregnancy, usually presenting in the third trimester. It
most often presents as liver failure with incidentally noted
ARF. The renal failure is a variant of hepatorenal syndrome but it is usually reversible. Approximately 80% of
patients recover with early cesarean section and supportive
care. The availability of liver transplant should further improve the prognosis, but the 80% recovery rate makes it
important not to perform a liver transplant prematurely.
Treatment for the renal failure is supportive. It is generally
mild, but oliguric ATN may occur if there are septic complications.

FOLLOW UP OF WOMEN DEVELOPING RENAL

ABNORMALITIES DURING PREGNANCY

Preeclampsia, even when severe, does not predict hypertension in later life. These women need careful follow-up
until all the manifestations of preeclampsia have resolved.
They can then be treated as healthy women, although they
should be regarded as high-risk patients during subsequent
pregnancies. Those with newly discovered essential hypertension require ongoing treatment and those with transient
hypertension are likely to develop essential hypertension
in later life. A woman who develops new onset proteinuria
during pregnancy should be followed until the proteinuria
resolves, If it persists, a renal biopsy should be conducted
at 6 weeks postpartum. Women who have had ARF should
be evaluated carefully to determine the completeness of
recovery. If there is persistent renal insufficiency, treatment
for the specific disease should be presented if available,
along with general measures to prevent the progression of
renal disease.
Bibliography
Benigni A, Gregorini G, Frulska T, et al.: Effect of low dose aspirin
on fetal and maternal generation of thromboxane by platelets
in women at risk for pregnancy induced hypertension. N Engl
J Med 321:357-362, 1989
Davison JM: Pregnancy in renal allograft recipients: prognosis and
management. Clin Obstet Gynaecol (Bailliere's) 8: 501-525,
1994.
Dunlop W, Davison JM: Renal haemodynamics and tubular function in human pregnancy. Clin Obstet Gynaecol (Bailliere's) 1:
769-787, 1987.
Gallery EDM, Brown MA: Volume homeostasis in normal and
hypertensive human pregnancy. Clin Obstet Gynaecol (Bailliere's) 1: 835-851, 1987
Hou SH: Peritoneal and haemodialysis in pregnancy. Clin Obstet
Gynaecol (Balliere's) 8: 481-500, 1994
Joint National Committee on Detection, Evaluation and Treat ment of Hypertension: National high blood pressure education
program working group report on high blood pressure in pregnancy. Am J Obstet Gynecol 163: 1689-1712, 1990.
Jones DC, Hayslett JP: Outcome of pregnancy in women with
moderate or severe renal insufficiency. N EnglJ Med 335: 226232, 1996.
Jungers P, Houillier P, Forget D, et al.: Influence of pregnancy
on the course of primary chronic glomerulonephritis. Lancet
346:1122-1124, 1995.
Kass EH: Infectious diseases and perinatal morbidity. Yale J Bio!
Med 55: 231-237, 1982
Katz AL, Davison JM, Hayslett JP, Singson E, Lindheimer MD:
Pregnancy in women with kidney disease. Kidney Int 18: 192206, 1980.
Mochizuki M, Morikawa H, Yamasaki M, et al.: Vascular reactivity
in normal and abnormal gestation. Am J Kidney Dis 17:139143, 1991.
Remuzzi G: HUS and TTP: Variable expression of a single entity.
Kidney Int 32: 292-308, 1987

THE KIDNEY IN AGING

JEROME G. PORUSH

ANATOMY OF THE KIDNEY

and may be explained by changes in the vasculature. The

interlobar arteries usually become tortuous and spiraling,
and fibroelastic hyperplasia of arcuate arteries and arterioles is seen progressively with age. The changes are more
pronounced in the cortical vessels, so that perfusion pres sure is relatively well maintained in the juxtamedullary
nephrons.

Aging is associated with numerous anatomical and functional renal alterations (Table 1). Much like other organs,
the kidneys undergo involutional changes with age (Table
2). There is a gradual decline in kidney weight starting in
the fifth decade, with the most marked decrease occurring
between the seventh and eighth decades. The progressive
loss of kidney mass appears to affect the renal cortex more
than the medulla. Up to age 40 the normal human kidney
has 600,000 to 1,200,000 glomeruli. Thereafter, there is a
progressive decrease of 30 to 50% in the number of glomeruli. Below age 50, 95% of the normal population have less
than 10% of sclerotic glomeruli. After age 50, the percent
of sclerotic glomeruli increases, making the distinction between involutional and disease-related sclerosis unclear.
The outer cortical glomeruli are, in general, more extensively involved than the deeper glomeruli.
In addition to glomerular sclerosis, there is a gradual
increase in interstitial fibrosis in the medulla primarily, with
little inflammatory response. Tubular length decreases and
proximal tubular volume of individual nephrons decreases
from a mean of 0.076 mm3 at 20 to 39 years to a mean of
0.050 mm3 at 80 to 100 years. Since the loss of glomerular
mass is proportional to the loss of tubular mass, glomerular
balance is usually well preserved. Lastly, there is an increase
in reduplication ai}d an increase in focal thickening of both
glomerular and tubular basement membranes, probably
due to the accumulation of type IV collagen.
The increase in the fraction of sclerotic glomeruli has
been attributed to the protein-rich diet characteristic of
western society. In the rat, this diet induces a state of
chronic glomerular hyperfiltration and hyperperfusion,
contributing to progressive glomerulosclerosis and agerelated decrease in glomerular filtration rate. An increase
in the function of residual intact nephrons would be expected to lead to an increase in nephron size; however, no
significant correlation is found between glomerular area
and glomerulosclerosis. Another possible explanation for
the progressive increase in sclerotic glomeruli is glomerular
ischemia secondary to an age related decrease in renal
blood flow (see below), since it has been shown that both
age and vascular disease correlate independently with the
percentage of hyalinized glomeruli in the aged.
The mechanism for the decrease in renal blood flow does
not appear to be related to a decrease in cardiac output,

Primer on Kidney Diseases. Second Edition

RENAL BLOOD FLOW AND GLOMERULAR

FILTRATION RATE

In cross-sectional and longitudinal studies, a gradual decrease in renal blood flow (RBF) and glomerular filtration
rate (GFR) is noted with age (Tables 3 & 4). RBF declines
progressively at a rate of approximately 10% per decade,
starting after the fourth decade. After the age of 30, there
is a 7.5 to 8 mL/min decline in the GFR per decade. An
age-adjusted derived creatinine clearance can be determined for males with the formula:
creatinine clearance (mL-mirr'-l.TSrrr 2) =
133- (0.64 x age).
For females, multiply the result by 0.93. For instance, a
60 kg, 90-year-old male is expected to have a creatinine
clearance of 133 -(0.64 X 90) = 75.4 mL-mirrMJSnr2 .
It should be noted that the result of this formula is only
an estimate, and the error inherent can be substantial since
some subjects will maintain stable renal function while
aging (approximately one third of the population), while
a small number will even have an increase in GFR. Serum
creatinine is the most commonly used marker of kidney
function (GFR) in clinical medicine. The decrease in creatinine clearance occurring with age is attended by a parallel
reduction in daily creatinine excretion so that there is frequently no change in the serum creatinine (Table 5). Lean
body mass declines by 12 kg in males and by 5 kg in females
by age 65 to 70, so that the decrease in creatinine excretion
probably reflects a decrease in muscle mass and creatinine
production. Also, because a significant number of the elderly suffer from chronic, debilitating diseases, including
neurogenic diseases, muscle atrophy is quite prevalent.
Thus, an estimate of renal function based solely on the
serum creatinine may be deceptive in the elderly. For a
more meaningful evaluation of kidney function, a creatinine clearance or other measurement of GFR is essential.

395

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

396

Jerome G. Porush

for females, Ccr = (13 X serum albumin + 29)

X weight H- 100 X serum creatinine.

TABLE I

Renal Changes in the Elderly

Diminished renal mass

Dropout of functioning nephrons

Glomerular sclerosis
Increased medullary interstitial fibrosis

Decreased GFR
Decreased RPF

Diminished creatinine production

Impaired sodium conservation
Diminished plasma renin and aldosterone levels
Increased prevalence of hyperkalemia
Impaired ammoniagenesis
Impaired maximal urinary concentration
Impaired maximal water excretion
GFR, glomerular filtration rate; RPF, renal
plasma flow.

Measurement of creatinine clearance requires a timed

urine collection (which is notoriously inaccurate in the elderly because of bladder emptying problems). A formula
developed by Cockcroft and Gault obviates this problem
as it allows calculation of the creatinine clearance from the
subject's age, weight, and serum creatinine:
Creatinine clearance (Ccr) (mL/min) = ((140-age)
X weight in kg) H- (72 X serum creatinine in mg/dL).
For females, multiply the result by 0.85. This formula is
applicable as long as the patient has stable kidney function
and is without severe chronic debilitating disease. In general, this formula overestimates the creatinine clearance in
females, the malnourished, and the debilitated. For these
subjects, another formula can be used:
for males, Ccr = (19 X serum albumin in g/dL + 32)
X weight (kg) -r- 100 X serum creatinine (mg/dL);

Filtration fraction tends to increase with age (Table 3), due

to the fact that GFR falls proportionally less than RPF,
and because there is a higher percentage of functioning
deep cortical nephrons in the elderly.
As noted earlier, advancing age is associated with fibroelastic hyperplasia and hyalinization of small arteries and
arterioles. It has also been found that blood pressure increases with age. The obvious question is whether the decline in renal function with age is mainly a reflection of
the increasing blood pressure. At a mean arterial pressure
(MAP) less than 107 mm Hg, no correlation is found between Ca and blood pressure. For pressures above that
level, however, there appears to be a negative correlation
between Ccr and MAP. Thus, the progressive decline in
GFR with the associated anatomical changes are a function
of age per se, but increasing the blood pressure appears
to accelerate the decline.
Infusion of amino acids in the elderly is associated with
an increase in GFR, similar to that seen in younger individuals. The renal plasma flow response, however, is blunted
and less predictable in the elderly and might be due to the
structural changes rather than an alteration in the response
to vasodilators such as prostacyclin and nitric oxide, both
of which have been found to contribute to the renal vasodilatory response to amino acids. Finally, patients over 60
years of age can undergo compensatory hypertrophy to the
same extent as younger patients as early as 7 months after
unilateral nephrectomy.
TUBULAR FUNCTION

As noted earlier, the anatomical changes affecting the

glomeruli are associated with changes in the tubules, so
that glomerular-tubular balance is generally adequately
maintained in the elderly.

TABLE 2

Anatomical Changes in the Kidney in Various Age Groups"

Age in year

Wt. of kidneys (g) (144)

Size of glomeruli'' (/wri) (182)
Cell number of glomerular
tufts (181)
No. of epithelial cells of
convoluted tubules in a
given area (125,600 ^m2)
(100)
Size of epithelial cell nuclei6
Gum) (100)

Under 39

40-49

50-59

60-69

70-79

Above 80

432 36
(12)
190 1
(12)
156 4
(12)
178 6
(6)

388 24
(23)
199 1
(25)
133 3
(25)
159 4
(15)

366 20
(20)
192 1
(35)
142 2
(36)
152 4
(17)

355 14
(35)
191 1
(43)
125 2
(42)
127 3
(23)

327 11
(32)
189 1
(42)
121 2
(42)
121 3
(23)

295 18
(22)
185 1
(25)
106 2
(25)
118 3
(16)

7.1 0.1
(7)

6.9 0.1
(14)

7.1 0.1
(17)

7.0 0.1
(24)

7.2 0.1
(22)

7.3 0.1
(16)

" From Tauchi H, Tsubi K, Okutomi J: Gerontology 17:87-97, 1971. With permission.
Square root of value area. Note: Number of kidneys examined is
indicated in parentheses.

61. The Kidney in Aging

397
TABLE 3
Changes in GFR and RPF with Age"

Age (yr)
50-59
60-69
70-79
80-89

Insulin clearance (mL

min-1 1.73m- 2)
99.3 14.6
96.0 25.5
89.0 19.9
65.3 20.4

Effective renal blood fl ow

(mL min -1 1.73m -2 )
849 123
775 139
589 133
475 141

Filtration
fraction (%)
21 + 3
22 + 3
26 8
23 4

" From Davies DF, Schock NW: JCI 29:496-507, 1950. With permission.

SODIUM BALANCE
Low Sodium Diet

In general, the elderly do not conserve sodium as well

as younger patients when placed on a low sodium diet.
When routine sodium balance studies are performed, they
are nor always conclusive. However, if one determines the
rate at which balance is achieved, it is evident that the
elderly do not conserve sodium normally. From age 30 to
60 years, dietary sodium restriction is associated with a
half-time for renal sodium conservation of 23.4 1.1 hours
compared to 30.9 2.8 hours for subjects over 60 years
of age.
The underlying mechanism for this difference is not clear,
and is undoubtedly, multifactorial. There is experimental
evidence that during salt restriction both proximal and
distal parts of the nephron participate in the increased
sodium reabsorption. The decrease in activity of the autonomic nervous system with age might contribute to the
inability of the elderly subject to conserve salt normally
since this system is known to contribute to the increased
proximal sodium reabsorption noted during salt restriction.
Despite a reduction in nephron number and cortical blood
flow in the elderly, medullary flow is well preserved. As a
result of this disparate blood flow, medullary washout might
occur, reducing the efficacy of the countercurrent system
and, thereby, decreasing the amount of sodium removed
by the ascending limb of deep nephrons. Furthermore,
as GFR decreases, the solute load per surviving nephron
increases, producing a solute diuresis, which may impair
the tubular ability to conserve sodium.
Plasma renin activity and plasma aldosterone have
been measured in different age groups during unrestricted
(120 mEq Na/day) and low sodium (<10 mEq Na/day)
diets in the supine and standing position. Both plasma renin
activity and plasma aldosterone decrease progressively
with each decade of life, starting after the fourth decade.
The same pattern persists whether subjects are on unrestricted or restricted sodium diets, supine or standing. The
reduced renin levels may be due to the decreased sympathetic activity in the elderly or to a possible decrease in
their ability to activate prorenin to renin. The lower basal
plasma aldosterone level may be explained by a decrease
in adrenal responsiveness to circulating angiotensin. Following salt restriction, plasma aldosterone rises appropriately in the elderly, however, so that it is unlikely that the

level, per se, is responsible for impaired sodium conservation.

Whatever the explanation, the diminished ability of the
elderly to conserve sodium when challenged with a low
sodium diet makes the defense of extracellular volume
precarious. Therefore, sodium restriction should be undertaken with caution in the elderly in order to prevent volume
contraction which may lead to acute reductions in renal
function or symptomatic orthostatic hypotension. The latter is particularly a problem in the elderly because of their
impaired baroreceptor response to the upright position.
Volume Expansion

Subjects over the age of 40 excrete a saline load slower

than younger subjects. This may be explained, in part, by
the decrease in filtered load, as GFR decreases with age.
Compared to younger subjects, the elderly excrete sodium,
potassium, and total solutes at a proportionately higher
rate at night. The baseline plasma level of atrial natriuretic
peptide, a substance that participates in the natriuresis associated with volume expansion, increases with age, possibly
due to a decrease in its catabolic rate. As saline loading in
the elderly results in a significant rise in atrial natriuretic
peptide, the blunted natriuresis after saline loading cannot
be attributed to an inappropriate response.
POTASSIUM BALANCE

The low plasma renin activity and plasma aldosterone

levels in the elderly contribute to their tendency to develop
hyperkalemia, particularly when receiving potassium supplementation. The incidence of hyperkalemia complicating

TABLE 4

Age and Creatinine Clearance"

Creatinine clearance (mL min-1
Age (yr) 1.73m-2 )
45-54
55-64
65-74
75-84

128 2
122 2
110 3
97 3
a

From Rowe JW, Andres R, Tobin

JD, et al: J Gerontol 31:155-163, 1976. With
permission.

398

Jerome G. Porush
TABLE 5

Relationship between Serum and Urine Creatinine and Creatinine Clearance"

Serum Creatinine
(mg/dL)
Age (yr)
40-49
50-59

60-69
70-79
80-89
90-99

Urine Creatinine
(mg kg-1 24H-1)

Creatinine clearance
(mL min" 1 173nr 2 )

(male)

(female)

(male)

(female)

(male)

(female)

1.10
1.16
1.15
1.03
1.06
1.20

1.00
0.99
0.97
1.02
1.05
0.91

19.7
19.3
16.9
14.2
11.7

17.6
14.9
12.9
11.8
10.7

88
81

81
74

72

9.4

8.4

63
54
46
39

64
47
34

' From Kampman M, Siersback-Nielson K, Kirstensen M, et al: Acta Med Scand 196:517-520,1974. With permission.

diuretic therapy with potassium supplementation is reported to be 0.5% in patients under the age of 50, 4.0% in
those 50-60 years, and even higher in older subjects. The
use of drugs that interfere with the distal tubular secretion
of potassium, such as potassium-sparing diuretics, angiotensin converting enzyme inhibitors, angiotensin receptor
blockers, nonsteroidal antiinflammatory drugs, and /3adrenergic antagonists necessitates careful monitoring of
serum potassium in elderly patients.
ACID- BASE BALANCE

Blood pH and bicarbonate levels do not change significantly with age. With advancing age, the kidneys maintain
their ability to maximally acidify the urine in the presence
of appropriate stimuli. Unfortunately, bicarbonate reabsorption by the proximal tubule has not been studied under
different loads. However, since plasma bicarbonate remains unchanged despite a gradual decrease in GFR with
age, it appears that glomerular-tubular balance for this
ion is well maintained. Nevertheless, one would expect the
elderly to excrete a bicarbonate load slower than younger
subjects, making them more prone to the development
of metabolic alkalosis. Under normal circumstances, the
excretion of titratable acid and ammonium is similar among
different age groups. However, after a 2 mmol/kg ammo nium chloride load, urinary excretion of net acid is significantly lower in elderly subjects. This defect is accounted
for solely by a decrease in ammonium excretion, since
titratable acid excretion is not altered with age. This impairment of ammonium excretion represents a reduction in the
maximum capacity of the nephron to synthesize ammonia.
URINARY CONCENTRATION

Aging is associated with a diminished capacity to maximally concentrate the urine. The results of one study of
12 hours of water deprivation showed that the elderly
(mean age = 68 years) achieve a mean urine osmolality of 882 mOsm/kg compared to 1051 mOsm/kg in a group
of middle -aged subjects (mean age = 49 years) and
1109 mOsm/kg H2O in a younger group of subjects (mean
age = 33). To achieve maximal urine concentration the
following features should prevail:

1. A normally functioning hypothalamic -hypophyseal

axis. If anything, it has been shown that there is a significant
age-related increase in osmoreceptor sensitivity. For the
same osmolar stimulus, elderly subjects tend to release
twice as much antidiuretic hormone (ADH) as younger
subjects.
2. Adequate solute delivery to the distal nephron. Al
though GFR decreases with age, no significant relationship
has been found between creatinine clearance and urine
osmolality, suggesting that solute delivery to the distal
nephron is not a limiting factor.
3. A collecting tubule responsive to ADH. Although the
maximum urine osmolality following infusion of large doses
of vasopressin is decreased in the elderly undergoing a
water diuresis, urine osmolality responds normally to stepwise incremental doses of vasopressin, insufficient to maxi
mally concentrate the urine, ruling out any significant im
pairment in the ability of the collecting duct to respond
to vasopressin.
4. A normally functioning ascending limb of Henle.
There are suggestions of a defect in sodium reabsorption
in the ascending limb of Henle's loop in elderly subjects.
This defect, coupled with the previously noted state of
medullary washout, leads to a significant reduction in med
ullary tonicity, and consequently the maximum urinary con
centration. The age-related decrease in urinary concentrat
ing ability and the defective thirst mechanism in the elderly
further increases their susceptibility to volume depletion
when placed on a low Sodium diet.

URINARY DILUTION

During water loading, the minimum urine osmolality

achieved by the elderly has been found to be higher than
that of younger subjects by some, but not all investigators.
However, the hourly free water clearance is consistently
lower in the elderly. The recovery of serum osmolality to
baseline is slower, and the average osmolar clearance lower
in the elderly during water loading. The observed delayed
water excretion may be explained by the loss of nephrons
and subsequent increase in solute load per nephron. It may
also be related to the reduction in the renal synthesis of
prostaglandins seen with aging, which may be associated

61. The Kidney in Aging

with enhanced ADH responsivness. Whatever the specific
mechanism, this inability to normally handle a water load
makes the elderly susceptible to hyponatremia.
TUBULAR REABSORPTION OF GLUCOSE

The maximum rate of renal tubular reabsorption of glucose decreases linearly with age. However, because GFR
decreases as well, glycosuria at normal plasma glucose levels is not seen.
Bibliography
Agarwal BN, Cabebe FG: Renal acidification in elderly subjects.
Nephron 26: 291-295, 1980.
Brenner B, Meyer TW, Hosteller TH: Dielary protein inlake and
Ihe progressive nalure of kidney disease: The role of hernody namically mediated glomerular injury in the palhogenesis of
progressive glomerular sclerosis in aging, renal ablalion and
inlrinsic renal disease. N Engl J Med 307:652-659, 1982.
Cockcrofl DW, Gaull MH: Prediclion of crealinine clearance from
from serum crealinine. Nephron 16:31-41, 1976.
Epslein M, Hollenberg NK: Age as a delerminanl of renal sodium
conservation in normal men. J Lab Clin Med 87: 411-417, 1976
Helderman JH, Veslal RE, Rowe HW, el al.: The response of
arginine vasopressin to inlravenous elhanol and hypertonic saline in man: The impacl of aging. J Gerontol 33:39-47, 1978

399
Hollenberg NK, Adams DF, Solomon HS, el al.: Senescence of
Ihe renal vasculature in normal man. Circ Res 34:309-316, 1974
Kaplan C, Paslernack B, Shah H, el al.: Age-relaled incidence of
sclerotic glomeruli in human kidneys. Am J Pathol 80:227234, 1975.
Kasiske BL: Relationship belween vascular disease and age associated changes in Ihe human kidney. Kidney Int 31: 1153-1159,
1987.
Lindeman RD, Tobin J, Shock NW: Longiludinal sludies on Ihe
rate of decline in renal function wilh age. / Am Gerontol Soc
33: 278-285, 1985
Miller JH, Shock NW: Age differences in Ihe renal lubular response lo anlidiuretic hormone. / Gerontol 8: 446-450, 1953
Porush JG, Fauberl PF: Renal Disease in the Aged. Lillle, Brown, &
Co, Boston, 1992.
Rowe JW, Andres R, Tobin JD, el al.: The effecl of age on crealinine clearance in man: A cross-sectional and longitudinal sludy.
/ Gerontol 31: 155-163, 1976.
Rowe JW, Andres R, Tobin JD, el al.: Age-adjusled standards
for crealinine clearance. Ann Intern Med. 84:567-559, 1976.
Rowe JW, Shock NW, DeFronzo RA: The influence of age on
Ihe renal response lo waler deprivation in man. Nephron 17:
270-278, 1976
Sanaka M, Takano K, Shimakura K, Koike Y, Mineshila S: Serum
albumin for estimating crealinine clearance in Ihe elderly wilh
muscle alrophy. Nephron 73: 137-144, 1996.
Tauchi H, Tsuboi K, Okulomi J: Age changes in Ihe human kidney
of the differenl races. Gerontologia 17: 87-97, 1971.

SECTION 10
CHRONIC RENAL FAILURE AND
ITS THERAPY

THE UREMIC SYNDROME

R. VANHOLDER

factors, such as enhanced renin and angiotensin production

due to decreased blood delivery to the kidneys, sodium
retention, and fluid overload.
Apart from changes in systolic contractility, a decrease
in diastolic compliance of the heart also plays a major
role. The basic mechanism is an increase of the interstitial
component of heart tissue. As a consequence, the heart
shows a decreased capability to relax during diastole, hence
decreasing the margin between fluid overload and hypovolemia. Consequently, patients will develop hypotension
when minor quantities of fluid are removed from the body
by dialysis procedures. It is especially near the end of the
interval between dialyses that they will easily develop hypertension and pulmonary edema. Heart disease in dialysis
patients is also discussed in Chapter 70.

INTRODUCTION

The uremic syndrome is characterized by the deterioration of biochemical and physiologic functions that occurs
with the progression of renal failure. It results from the
retention of substances that are ordinarily removed by the
healthy kidneys; the intake of precursors, mainly via nutritrion, also plays a role. The uremic syndrome also results
from derangements of hormonal and enzymatic homeostasis. The quest tor "the" uremic toxin has been overemphasized. Some researchers and nephrologiots consider one specific toxin or group of toxins as responsible
(e.g., urea, parathormone, /32-microglobulin, the group of
"middle" molecules with molicular weight, from 300 to
12,000 Da), not taking into account the uremic syndrome
as the cumulative result of retention of innumerable compounds and deficiency of others. This chapter reviews the
current knowledge about the uremic syndrome, its clinical
and biochemical characteristics, and the factors playing a
role in its development.

Uremic Encephalopathy and Peripheral Neuropathy

These disorders may develop alone or in combination.

The spectrum of central neuropathy is variable and ranges
from subtle cognitive impairment to coma. Sleep disorders,
mainly sleep-apnea syndrome from upper airway obstruction are among the major causes. Dialysis prevents encephalopathy, so that this complication is encountered only in
exceptional circumstances, such as patients who do not
seek medical help until late in their course, or when dialysis
is abandoned because of grave disease.
Severe peripheral polyneuropathy has also become unusual today, although less pronounced forms still persist,
even in patients treated by apparently adequate dialysis.
Neuroleptic manifestations of renal failure are discussed
in Chapter 71.

CLINICAL CHARACTERISTICS

The uremic syndrome is characterized by a quantitative

and qualitative overall deterioration of performance, affecting the cardiovascular, neurologic, hematologic, immu nologic, and other systems. In addition, malnutrition and
acidosis may play an important pathophysiologic role. Fullblown uremic syndrome results in a complex clinical picture
consisting of fatigue, anorexia, weight loss, itching, muscle
cramps, pericarditis, sensory and concentration disturbances, and thirst. It may eventuate in stupor, coma, and
death (Table 1).
Cardiovascular Anomalies

Anemia

One or more cardiovascular anomalies such as hypertension, congestive heart failure, valvular stenosis or insufficiency, accelerated athero -sclerosis and uremic pericarditis,
occur commonly during the progression of renal failure.
Myocardial dysfunction is related to increased myocardial
calcium content, especially in patients undergoing chronic
dialysis. Increased cytosolic Ca 2+ which is observed in many
uremic patients has been associated with an increase of
peripheral vascular resistance, resulting in hypertension.
It should be stressed, however, that hypertension in the
uremic patient may result from a host of other causative

Primer on Kidney Diseases, Second Edition

Until the introduction of genetically engineered recombinant erythropoietin, uremic anemia was one of the major
reasons for the impaired physical condition of renal failure
patients. This hypoproliferative anemia is mainly due to
inadequate erythropoietin production by the failing kidneys and, to a lesser extent, by defective body iron stores,
vitamin deficiency, hemolysis, and erythrocyte fragility.
The causative role of inhibitors of erythropoiesis remains
a matter of debate. Two proven inhibitors of in vitro hematopoiesis found in uremic plasma are the polyamines sper403

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

404

R. Vanholder

TABLE I

The Uremic Syndrome: Clinical Alterations

Nervous system Stupor, coma
Polyneuritis
Fatigue Dementia Malaise
Insomnia Headache Restless legs Convulsions
Motor weakness
Flapping tremor
Concentration disturbances
Gastrointestinal system Stomatitis
Drowsiness
Pancreatitis
Gastritis Anorexia
Irritability
Gastrointestinal ulcers
Hematological system Anemia
Cramps
Nausea, vomiting
Cardiovascular system Pericarditis
Atheromatosis Edema Decreased
Bleeding
diastolic compliance
Pulmonary system Pleuritis Uremic
Hypertension
Pulmonary edema
Lung
Cardiomyopathy
Skin Pruritus Retarded wound
Hypotension
Melanosis Nail
healing
atrophy
Bone disease Osteodystrophy
Hyperparathyroidism
Amyloidosis (/KM)
Miscellaneous Thirst Weight loss
Calcitriol metabolism defect
Impotence, diminished libido
Uremic foetor
Hypothermia

mine and spermidine. Anemia

and other hemologic feature of uremics are discussed in
Chapter 72.
Disturbances of Coagulation
Disturbances of coagulation may include a varied spectrum of abnormalities ranging from a bleeding tendency
to hypercoagulability. Uremic bleeding tendency has a multifactorial origin: platelet related alterations (adherence,
aggregation), uremic anemia, and hyperparathyroidism. In
earlier studies dating from the 1970s, urea and the guanidines were held responsible for defective coagulation. Recent studies have not confirmed these findings because
coagulation disturbances could only be demonstrated at
concentrations of these compounds that were much higher
than those currently observed today with uremia. Recent
data incriminate nitric oxide (NO) as an important inducer
of defective coagulation in uremia. Hemodialysis patients
also receive regular systemic heparinization.
Immune Deficiency
Susceptibility to infection, increased incidence of cancer,
and inadequate production of antibodies, e.g., against hepa-

titis B, all point to immune deficiency in uremia. The problem here is not entirely due to toxins, since bioincompatibility of dialyzer membranes, vitamin D deficiency, and
anatomical lesions (e.g. dialysis fistula and peritoneal catheters which predispose to infection) may contribute as well.
The circulating number of white blood cells is not altered,
but their function is abnormal.
The factors responsible for the depression of immune
function in uremia have not been identified. Candidates
include a lack of circulating opsonins (e.g., fibronectin),
iron overload, uremic anemia, and accumulation of circulating substances with immune suppressant effect (endomorphins, phenols, indoles, parathyroid hormone, high and
middle molecular weight compounds).
Endocrine Dysfunction
The endocrine dysfunction of uremia (see also Chapter
73) affects carbohydrate metabolism, and thyroid, growth,
and reproductive hormones, resulting in a progressive catabolic state. In addition to disturbances in hormonal states,
other causes of negative nitrogen balance are insufficient
food intake due to medication side effects, gastrointestinal
disease, central nervous system dysfunction, enhanced tis sue breakdown from an apparent chronic low grade inflammatory state, and urinary protein losses. Acidosis inhibits many metabolic functions, including protein
synthesis. Dialysis treatment, although life-saving, may result in a further enhancement of catabolic status and of
nitrogen losses as amino acids are abnormal in hemodialysis
and peritoneal dialysis and albumin is lost directly during
peritoneal dialysis.
Changes in carbohydrate metabolism consist of alterations of glucose metabolism, pancreatic glucose-induced
insulin secretion, and target organ sensitivity to insulin.
Together, these abnormalities result in abnormal glucose
tolerance. Normal to mildly elevated fasting glucose and
high insulin levels are usually found.
Levels of T3 and T4 may be low normal or depressed,
the pituitary responsiveness being abnormal, in spite of a
euthyroid appearance.
Basal levels of growth hormone are elevated in uremia,
in proportion to the degree of renal impairment. Nevertheless, growth retardation is a major problem in uremic children. Contributing factors are malnutrition, acidosis, renal
osteodystrophy, inadequate gonadotropic hormone secretion, and increased concentrations of growth hormone
binding protein.
Advanced renal failure results in reproductive abnormalities in both males and females, related to hormonal disturbances. In uremic women, follicular stimulating hormone
(FSH), progesterone, and estradiol levels are similar and
luteinizing hormone (LH) levels tend to be higher than
those observed in the follicular phase of the menstrual
cycle in normal women. Uremic men may be infertile or
impotent, in association with increased LH and decreased
testosterone levels. Spermatogenesis is compromised, because of increased FSH levels. Prolactin levels may be
elevated, inducing galactorrhea and amenorrhea in women
and impotence in men. Gynecomastia may occur in men.

62. The Uremic Syndrome

405

Uremic Bone Disease

Hyperparathyroidism, aluminum toxicity, vitamin D deficiency and resistance, and accumulation of /32-microglobulin, all contribute to development of renal osteodystrophy.
In addition, uremic retention solutes may induce a general
resistance to the metabolic effects of l,25(OH)2 VitD3. This
topic is covered in Chapter 69.
Pruritus
Pruritus is a disturbing epiphenomenon of uremia, with
unclear pathophysiologic mechanisms. Several different
factors have been incriminated: increased levels of serum
vitamin A, hyperparathyroidism, high skin contents of divalent cations, mast cell proliferation with increased release
of histamine, and abnormal cutaneous innervation. Responsible toxins, if any, have not been incriminated, apart
from parathormone.
Malnutrition
A substantial proportion of uremic patients are undernourished. Protein-calorie malnutrition is associated with
an increase in morbidity and mortality. Contributing factors
include anorexia, nausea, vomiting and the subsequent decreased nutrient intake, hormonal and metabolic derangements, and catabolic factors related to the uremic status
as well as the dialytic treatment of uremia (see Chapter 68).
Acidosis
Acidosis results from the accumulation of protons and
bicarbonate wasting, although the relative contribution of
both factors to uremic acidosis may differ from patient to
patient. Bicarbonate wasting and retention of chloride result in hyperchloremic acidosis. If phosphate and organic
acid (uric acid, hippuric acid, lactic acid) retention is sufficient, an increased anion gap will occur.
Acidosis leads to a host of metabolic alterations, hyperventilation, anorexia, headache, stupor, decreased cardiac
inotropic response, catabolism, and hyperkalemia. This
complication of uremia can easily be corrected by the administration of extra alkali or removal of acid by dialysis.
BIOCHEMICAL ALTERATIONS
Enzymatic and Metabolic Dysfunction
A host of enzymatic and metabolic functions are depressed: gluconeogenesis, lactate dehydrogenase activity,
mitochondrial storage of calcium and oxygen consumption,
alkaline phosphatase isoenzyme activity, insulin degradation, Na,K-ATPase activity, tubular anion transport, pro duction and metabolic clearance of calcitriol, and DNA
repair ability.
Changes in Drug Protein Binding
Two protein binding defects are seen in uremia. One
group of mostly acidic drugs shows decreased binding,

which leads to an increase in the free, active fractions of

the drugs in plasma. Because, for most drugs, total plasma
(bound plus unbound) concentrations are monitored, therapeutic free levels are achieved at a lower than normal
total concentration. At conventional normal levels, toxic
side effects should be expected. Examples of drugs include
theophylline, phenytoin, methotrexate, diazepam, digoxin,
and salicylate. Basic drugs, e.g., propranolol, cimetidine,
clonidine, imipramine, have increased protein binding. This
causes a decrease in the plasma free concentration, diminishing the therapeutic effect. Higher plasma total concentrations are usually required for these drugs.
Attempts to identify the ligands responsible for decreased protein binding have been scant. Several studies
suggest that hippuric acid would be one of the main contrib uting compounds, but its relative importance remains undefined. Other potential competitors are indoxyl sulfate, derivates of furanpropanoic acid and other furancarboxylic
acids, and indole-3-acetic acid. It should be stressed that
alterations in drug pro tein binding are not the only pharmacologic changes in uremia. Decreased renal clearance or
metabolism and changes in distribution volume may also
alter drug effect. In addition, accumulation of active drug
metabolites may augment drug potency or toxicity (see
Chapter 43).
UREMIC SOLUTE RETENTION
Current knowledge indicates that each metabolic or
physiologic change in uremia may be attributed to certain
solutes that are retained as a result of renal failure. Some
substances that are not toxic may be used as markers of retention.
Under normal conditions, molecules with molecular
weights up to approximately 58,000 Da are cleared by
the glomerular filter. Uremic retention products have been
arbitrarily subdivided in several groups, according to their
molecular weight: low molecular weight MW (<300 Da),
e.g., urea (MW: 60), creatinine (MW: 113 Da); middle MW
(300 to 15,000 Da), e.g., parathormone (MW: 9424 Da),
/32-microglobulin (MW: 11,818 Da); and high-molecularweight molecules (>15,000 Da) e.g., myoglobin (MW:
18,800 Da).
A listing of some currently known uremic retention solutes with their molecular weights is given in Table 2. In
the following discussion, the most important organic compounds and their presumed toxicity will be reviewed.
Urea
Urea is responsible for the elevated concentration of
hemoglobin A1C observed in nondiabetic uremics, due
to carbamylation of the protein. Other proteins probably
undergo structural changes in a similar manner. It is accepted that in general, urea produces acute clinical symp toms only at higher concentrations than those usually encountered nowadays. Given the extensive number of
toxicity studies to which urea has been submitted, the number of proven toxic effects is disappointingly limited. Nevertheless, some toxic effects such as inhibition of platelet

406

R. Vanholder
Hippuric Acid

TABLE 2

The Most Currently Known Uremic Solutes with Their

Molecular Weights
Compound
Nitric oxide

Urea
Methylguanidine
Phenol

Phosphate
p-Cresol
Creatinine
Hypoxan thine

Spermidine
Xanthine

Urate
Guanidinosuccinate
Indole-acetate

MW
30
60
73
94
96
108
113
136
145
152
168
175
175

Compound

MW

Guanidinoacetate

177

Hippuratc

179

myo-Inositol
ADMA/SDMA
Dimethylarginine

180
202

Spermine
CMPF

202

Pseudouridine

Indoxyl sulfate
Phenylacctylglutamine
/3-Endorphin
Parathorraone
/32 -Microglobulin

202
240
244
251
264
2465
9425
11818

CMPF, carboxymethylpropylfuranpropionic acid; ADMA,

asymmetric dimethylarginine; SDMA, symmetric dimethylarginine.

aggregation and Na,K-ATPase have been demonstrated.

Conceivably, urea may exert some chronic toxicity even at
the lower concentrations encountered currently. Even in
the absence of any toxicity, urea has been recognized as a
useful marker of uremic solute retention and elimination
in dialysis patients. Careful multicenter studies by the U.S.
National Cooperative Dialysis Study (NCDS) revealed a
direct correlation between urea kinetics and morbidity of
patients on hemodialysis, which resulted in a number of
practical and objective indices of dialysis efficiency, allowing better patient management (see Chapter 63).
Methylguanidine and Other Guanidines
Guanidinosuccinic acid and methylguanidine accumu late in uremic serum at levels that are potentially toxic.
Methylguanidine, when injected to healthy dogs, causes
symptoms of heavy intoxication. Guanidinosuccinic acid
has been shown to interfere with platelet function and with
the production of calcitriol from its precursors in the rat.
Guanidines may also contribute to uremic neuropathy.
Indoxyl Sulfate
Indoxyl sulfate has been held responsible for decreased
binding of drugs to plasma proteins, and to defects of cellular organic acid transport, especially in the renal cortical
tubules. Reduction of serum indoxyl sulfate concentration,
by intraintestinal absorption of the precursor indole, reduces uremic itching. Indoxyl sulfate is not removed effectively during hemodialysis, due to its high protein binding.
Removal by CAPD is more effective.
Due to its substantial protein binding (about 100% in
normals, and 90% in uremics), indoxyl sulfate's intradialytic
behavior does not correspond to that of other small, water
soluble compounds such as urea and creatinine.

Hippuric acid inhibits the transport of a variety of organic

acids, especially in the renal proximal tubule, the chorioid
plexus of the brain, the ciliary body of the eye, the thyroid,
the liver, and erythrocytes. It also interferes with the pro tein binding of acidic drugs, such as phenytoin and theophylline, and with glucose tolerance. Hippuric acid (MW:
179 Da) also behaves like a larger molecule, due to its
protein binding.
Peptides

Concentration changes of vasoactive peptides may play

a role in the regulation of blood pressure in uremic patients.
Endothelin is retained in uremic patients. Atrial natriuretic
factor (ANF-atriopeptide) may be elevated in uremia, especially during states of expanded extracellular fluid volume, e.g., before the start of hemodialysis. The effect of
hemodialysis on vasoactive peptide concentration has recently been evaluated. Whereas endothelin decreased only
with high flux dialyzers, atriopeptide decreased with all
types of dialyzers, irrespective of their pore size, due to
the decrease in intravascular volume that occurred during dialysis.
Parathormone (PTH), a middle molecule with a MW of
approximately 9000 Da, is oversecreted during uremia to
correct for hypocalcemia and hyperphosphatemia. PTH
causes glucose intolerance, inhibition of human platelet
function, cardiomyopathy, inhibition of erythropoiesis, altered red cell osmotic resistance, altered B cell proliferation, synaptosome and T-cell dysfunction, and defects of
fatty acid oxidation.
The molecular weight of PTH makes it a classical middle
molecule. Although PTH removal by dialysis has a rather
weak influence on its plasma concentration, serum PTH
serves as one of the indirect parameters of long-term adequacy of dialysis, since its production is correlated with
long-term phosphate accumulation.
/32-Microglobulin (/32M) (MW approximating 11,000
Da), a breakdown product of the major histocompatibility
complex, is also a middle molecule. Amyloidosis, frequently observed in long-term dialysis patients (see Chapter 69), is caused by deposition of /32M in various sites. It
is not entirely clear whether this disorder is related to
accumulation as such, or other factors such as the enhanced
release of the protein or a result of membrane or dialysate
contact during hemodialysis.
Dialysis -related amyloidosis, is characterized by the deposition of |62M amyloid in bone, causing cysts that may
be painful, and in the tendons, tendon rupture and carpal
tunnel syndrome. The latter results in nerve and motor
dysfunction of the hands.
The intradialytic plasma concentration of /32M is determined by both removal and hemoconcentration. Plasma
/32M levels often increase during dialysis using membranes
with small pores because the proteins are not removed and
also because removal of water causes hemoconcentration.
Removal of /32M is, however, achieved during dialysis using
dialyzers with large pores, but even removal by these mem-

62. The Uremic Syndrome

branes cannot keep pace with the daily generation of /32M.

Patients treated exclusively with polyacrylonitrile (AN69)
membranes with large pore size have less bone amyloidosis
than those treated with small pore size cellulosic membranes. Breakdown products of larger proteins such as /32M
are also potentially deleterious. Two polypeptides extracted
from uremic plasrna ultrafiltrate, one with structural homology with a part of /32M, have immunosuppressive effects.
Organic Phosphates

A high level of organic phosphates may cause uremic

itching. Accumulation of organic phosphates stimulates
parathyroid activity, resulting in hyperparathyroidism.
Phosphates are primarily derived from protein intake and
catabolism as well as other dietary sources. Dialytic phosphate removal is followed by a marked rebound, suggesting
multicompartmental behavior which does not correlate
with urea elimination.
At high concentrations, organic phosphates complex
with calcium and are deposited in the tissue as phosphatecalcium complexes; this causes metastatic calcification in
many tissues, inducing functional disturbances besides itching and conjunctivitis.
FACTORS INFLUENCING PLASMA CONCENTRATION
OF UREMIC SOLUTES

Numerous factors influence the plasma concentration of

uremic solutes, such as protein intake, general metabolic
status, residual renal function, dialysis performance, protein binding, and multicompartimental distribution. Elimi nation via the urine remains important even after dialysis
has been started. For example, at a clearance of 5 mL/min,
the kidneys still clear 50 L/week, to be compared with a
clearance of 15 to 150 L/week (depending on the specific
solute) through artificial kidneys.
CONCLUSIONS

The uremic syndrome results from the accumulation of

multiple toxins rather than of one single substance, and
from the deficiency of certain substances. Clinically employed markers of uremia, such as urea and creatinine, do
not play a major role in the development of the clinical
uremic syndrome.
The source of these uremic toxins is only related in part
to protein breakdown. Whereas urea and organic phosphates are products of protein catabolism, and creatinine
and the guanidines from muscle breakdown, the origin of
many other uremic toxins is less clear. For some, such as
hippuric acid, their origin is a specific amino acid together
with non protein sources. Others have completely different
origins, as is the case for indoxyl sulfate, /32M, and PTH.
Calculations based on urea removal are often used to
guide the dialysis prescription, because urea is a small,
water-soluble non-protein-bound compound that is generated from protein and easily measured. However, the removal and generation of many of the putative uremic toxins

407

discussed above are not readily predicted from the behavior

of urea, because they are hydrophobic, protein bound, or
not products of protein catabolis m.
Depending on the renal replacement strategy, there may
be a dissociation between the removal of small water soluble markers such as urea and creatinine as compared to
toxic retention compounds with a higher molecular weight
(e.g., /32M, PTH) or lipophilicity (e.g., phenols, p-cresol).
Substantial improvement in artificial kidney technology
awaits a more thorough understanding of both the basic
metabolic disturbances that occur in uremia and the toxic
compounds responsible.
Bibliography
Fraser CL, Arieff AI: Nervous system complications in uremia.
Ann Intern Med 109, 143-153, 1988.
Haas T, Million D, Dongradi G: Phosphate kinetics in dialysis
patients. Nephrol Dial Transplant 6 (Suppl 2), 108-113, 1991.
Horl WH, Haag-Weber M, Georgopoulos A, Block LH: Physicochemical characterization of a polypeptide present in uremic
serum that inhibits the biological activity of polymorphonuclear
cells. Proc Natl Acad Sci USA 87, 6353-6357, 1990.
Hsu CH, Patel SR, Young EW, Vanholder R: The biological action
of calcitriol in renal failure. Kidney Int 46, 605-612, 1994.
Kushner D, Beckman B, Nguyen L, et al.: Polyamines in the
anemia of end-stage renal disease. Kidney Int 39,725-732,1991.
Lowrie EG, Laird NM, Parker TF, Sargent JA: Effect of the
hemodialysis prescription on patient morbidity. N Engl J Med
305, 1176-1181, 1980.
Mall G, Rambausek M, Neumeister A, Kollmar S, Vetterlein F,
Ritz E: Myocardial interstitial fibrosis in experimental uremia:
implications for cardiac compliance. Kidney Int 33, 804-811,
1988.
Niwa T, Fujishiro T, Uema K, et al.: Effect of hemodialysis on
plasma levels of vasoactive peptides: endothelin, calcitonin -gene
related peptide and human atrial natriuretic peptide. Nephron
64, 552-559, 1993.
Noris M, Benigni A, Boccardo P, et al.: Enhanced nitric oxide
synthesis in uremia: implications for platelet dysfunction and
dialysis hypotension. Kidney Int 44, 445-450, 1993.
Parfrey PS, Harnett JD, Barre PR. The natural history of myocardial disease in dialysis patients. J Am Soc Nephrol 2,2-12,1991.
Piafsky KM: Disease-induced changes in the plasma binding of
basic drugs. Clin Pharmacokin 5, 246-262, 1980.
Ringoir S, Schools A, Vanholder R. Uremic toxins. Kidney Int 33
(Suppl 24), S4 - S9, 1988.
Ritz E, Matthias S, Seidel A, Reichel H, Szabo A, Horl WH:
Disturbed calcium metabolism in renal failure: pathogenesis and
therapeutic strategies. Kidney Int 42 (Suppl 38), S37-S42,1992.
Vallance P, Leone A, Calver A, Collier J, Moncada S: Accumulation of an endogenous inhibitor of nitric oxide synthesis in
chronic renal failure. Lancet 339, 572-575, 1992.
Vanholder R, Ringoir S: Adequacy of dialysis: a critical review.
Kidney Int 42, 540-558, 1992.
Vanholder R, Ringoir S: Infectious morbidity and defects of phagocytic function in end- stage renal disease: a review. / Am Soc
Nephrol 3, 1541-1554, 1993.
Vanholder R, Patel S, Hsu CH: Effect of uric acid on plasma
levels of 1,25 (OH) 2 D in renal failure. / Am Soc Nephrol 4,
1035-1038, 1993.
Vanholder R, De Smet R, Waterloos MA, et al.: Mechanisms
of uremic inhibition of phagocyte reactive species production:
characterization of the role of p-cresol. Kidney Int 47, 510517, 1995.

HEMODIALYSIS AND HEMOFILTRATION

ALFRED K. CHEUNG

STRUCTURE OF HEMODIALYZERS AND HEMOFILTERS

hemodialysis. High flux hemodialysis utilizes membranes

with large pores in order to rapidly remove water and
solutes that are substantially larger than urea ("middle
molecules") (see Chapter 62). For chronic renal failure
patients, maintenance hemodialysis is usually performed
intermittently, for example, 3 to 5 hours three times a week
either at home or in-center at a dialysis unit.
Hemofiltration is another form of extracorporeal therapy
which removes fluid by convection, i.e., movement of water
across the large pore hemofilt ration membrane into the
ultrafiltrate compartment drags along the solutes which
are dissolved in the water (Fig. IB). A crucial distinction
between hemodialysis and hemofiltration is that fluid must
be removed in hemofiltration for solute removal. Removal
of fluid with its accompanying solutes results in a loss in
the total body mass of the solute, but not a decrease in
the plasma concentration. In order to achieve a substantial
decrease in the concentration, "clean" replacement fluid
is infused intravenously to approximately replace the large
volume of plasma fluid removed in the hemofilter. This
modality is analogous to the glomerulus, which also removes plasma solutes by convection. In the case of the
glomerulus, however, the replacement fluid is the water
and electrolytes that are selectively reabsorbed from the
renal tubules (Fig. 1C). Hemodiafiltration refers to the
combination of hemodialysis and hemofiltration operating
simultaneously using a large pore membrane, i.e., solutes
are removed by both diffusion as well as convection. Hemo filtration and hemodiafiltration are used in the United
States primarily for acute renal failure. Under these circumstance, they are applied continuously for days to weeks,
usually in the intensive care unit, and are therefore termed
"continuous renal replacement therapy." In Europe and
Japan, hemofiltration and hemodiafiltration are also utilized on an intermittent basis to treat chronic renal failure.
During a 4-hour session of maintenance hemofiltration, 16
to 20 L of body fluid is typically exchanged.
Hemoperfusion is the removal of solutes (usually toxins)
from blood by adsorption onto materials, such as charcoal
or resins, in the extracorporeal circuit. Hemoperfusion is
primarily used to treat acute poisoning.

Extracorporeal therapy for renal failure refers to the pro cess in which fluid and solutes are removed from or added
to the patient's blood outside the body. During this process,
blood from the patient is continuously circulated through a
hemodialyzer or hemofilter containing an artificial semipermeable membrane and then returned to the patient.
A typical modern hemodialyzer is composed of several
thousand parallel hollow fibers. The wall of these fibers is
the semipermeable membrane separating the blood in the
fiber lumen from the dialysate outside. The total internal
surface area of all the fibers is usually between 0.5 and
2.0 m2. Less commonly, the membranes are in the form
of flat sheets rather than hollow fibers. Regardless of the
membrane geometry, blood from each parallel flow path
is channelled into the single inlet and single outlet of the
plastic casing. There is also an inlet and an outlet for the
dialysate compartment in which the dialysate is usually
circulated in a single-pass fashion countercurrent to the
blood flow. Hemofiltration employs no dialysate; hemo filters have an inlet and an outlet for blood and a single
outlet for the ultrafiltrate compartment.

TYPES OF EXTRACORPOREAL THERAPY FOR

RENAL FAILURE

Among the numerous modalities of extracorporeal renal

therapy outlined in Table 1, "conventional hemodialysis"
and "high efficiency hemodialysis" are most commonly
used, although "high flux hemodialysis" is gradually gaining popularity. Dialysis removes solutes by diffusion, based
on concentration gradients of solutes between the blood
and dialysate across the semipermeable membrane (Fig.
1A). For example, urea diffuses from the blood to the
dialysate compartment, thereby decreasing the total urea
mass in the body as well as the urea concentration in the
plasma. Conversely, the concentration gradient of bicarbonate usually favors the diffusion of this ion from the
dialysate to the blood compartment. Fluid movement is
not a prerequisite for solute transport in this modality,
although fluid loss from the patient is often desirable. High
efficiency hemodialysis refers to hemodialysis using high
blood flow rates and/or large surface area membrane dialyzers in order to remove urea and other small solutes (such as
urea and potassium) more rapidly than during conventional

Primer on Kidney Diseases, Second Edition

HEMODIALYSIS AND HEMOFILTRATION MEMBRANES

AND MACHINES

Transport and, to a lesser extent, biocompatibility are

two characteristics of dialysis membranes that are of inter-

408

Copyright 1998 by the National Kidney Foundation

AH rights of reproduction in any form reserved.

63. Hemodialysis and Hemofiltration

409

TABLE I
Glossary of Various Types of Extracorporeal Therapy for Renal Failure
Hemodialysis (HD)
Conventional hemodialysis hemodialysis using a conventional low flux (small pore size) membranes. Solute removal is primarily by
diffusion. High efficiency hemodialysishemodialysis using a low flux (small pore size) membrane with high efficiency (KoA) for
removal of
small solutes. Typically achieved by using a larger surface area membrane. High flux hemodialysis hemodialysis using a high flux
(large pore size) membrane. It is more efficient in removing larger solutes,

but may or may not be more efficient than conventional hemodialysis in removing small solutes.
Hemofiltration (HF) Continuous arteriovenous hemofiltration (CAVH)removal of small and larger solutes using a high flux
membrane and convection
rather than diffusion. Blood is accessed from an artery and returned to a vein with the driving force deriving from the systemic
arterial pressure. Because it is performed continuously (usually in the intensive care setting), it is also very effective in removing
large amount of fluid. Continuous venovenous hemofiltration (CVVH)similar to CAVH except that blood is accessed from a
vein and returned to
another vein using a blood pump. Intermittent hemofiltrationhemofiltration performed on an
intermittent basis for chronic renal failure.
Hemodiafiltration (HDF)
Continuous arteriovenous hemodiafiltration (CAVHD) similar to CAVH in that solutes are removed by convection using a high
flux membrane in an acute setting. In addition, dialysate flows continuously through the dialysate compartment in order to
enhance solute removal by diffusion, i.e., it is a combination of hemodialysis and hemofiltration. Continuous venovenous
hemodiafiltration (CVVHD) similar to CAVHD except that blood is accessed from a vein and returned

to another vein using a blood pump. Intermittent hemodiafiltrationhemodiafiltralion performed on an

intermittent basis for chonic renal failure.
Hemoperfusion (HP)removal of solutes by adsorption to charcoal or resin, primarily for treatment of acute poisoning.

est to clinical nephrologists. Although there are exceptions,

hemodialysis membranes that are made from cellulose usually have small pores which restrict the movement of large
solutes, and to some extent, the movement of water. In
contrast, high flux hemodialysis membranes, which can also
be used for hemofiltration, are usually made from synthetic
polymers with large pores that facilitate transport of water
and middle molecules. Modern hemodialysis machines contain ultrafiltration control devices that precisely regulate
the flow of fluid across these porous membranes, thereby
preventing an excessively rapid decrease in intravascular
volume. Because of their high hydraulic permeability (or
ultrafiltration coefficient, see below), high flux membranes
can only be used with machines that are equipped with
these devices.
Biocompatibility of dialysis membranes refers to the interactions that occur as a result of contact of blood with the
membrane. Examples of such interactions include the activation of proteins of the coagulation and complement system, as well as various peripheral blood cells. Biocompatibility is relative, since all dialysis membranes induce reactions
to a certain extent which can manifest, for example, as
thrombosis in the dialyzer and, rarely, acute anaphylactoid
reactions. Recent studies have suggested that patients with
acute renal failure who were dialyzed with membranes that
activate less complement and neutrophils had more rapid
recovery of renal function and lower mortality rates than
patients who were dialyzed with membranes that activate
complement and neutrophils more intensely.

The dialysis machine incorporates many important features including, a pump to deliver blood to the dialyzer at
a constant rate, monitors to ensure that the pressures inside
the extracorporeal circuit are not excessive, a detector for
leakage of red blood cells from the blood compartment
into the dialysate compartment, an air detector and shutoff device to prevent air from entering the patient, a pump
to deliver dialysate, a proportioning system to properly
dilute the dialysate concentrate, a heater to warm the dialysate to approximately body temperature, and conductivity
monitors to check dialysate ion concentrations. These devices ensure the proper, safe, and reliable delivery of blood
and dialysate to the dialyzer where exchange of water and
solutes occurs. Machines employed for intermittent hemo filtration are similar to those for hemodialysis, with the
additional requirements for precise ultrafiltration and fluid
replacement coordination and control.
In contrast, continuous arteriovenous hemofiltration
(CAVH) does not require any machinery. Blood is delivered to the hemofilter from an artery, such as the femoral
artery, via a large bore (14-15 gauge) catheter and returned to a large vein, with the driving force derived directly
from the heart rather than a mechanical pump. Unfortunately, blood flow and therefore ultrafiltration rates are
sometimes erratic in CAVH. Machines have been developed which pump blood from a vein through the hemofilter
and back to a veina technique known as continuous
venovenous hemofiltration (CVVH). The machines for
CVVH are usually simpler in design than conventional

410

Alfred K. Cheung
Blood
compartment

Dialysate

Hemodialysis
membrane ~~
compartment

\ _____
Blood
compartment

Replacement fluid
Hemofiltration
membrane

- Ultrafiltrate
compartment

Fluid
reabsorption

Bowman's
space

Renal
tubules

FIGURE I Schematic representation of solute and fluid transport across the

semipermeable membranes. (A) Hemodialyzer. The plasma concentration of
small solutes (solid circles) in the blood inlet is high. Because of the diffusive
loss across the semipermeable hemodialysis membrane (dotted line), the plasma
concentration in the blood outlet is much lower. The thin arrow across the
dialysis membrane represents a small amount of fluid loss (which is not necessary
for solute removal). High dialysate flow rate is necessary to maintain the concen tration gradient across the dialysis membrane. (B) Hemofilter. Plasma concentration of small solutes in the blood compartment remains unchanged as blood
travels the length of the fiber, and is similar to their concentrations in the
ultrafiltrate. The hemofiltration membrane (broken line) has relatively large
pores, which allow the necessary removal of large volume of fluid (heavy arrow).
Replacement fluid is infused into the blood outlet in order to lower the plasma
concentration of solutes and compensate for the fluid loss. (C) Glomerulus.
Analogous to hemofiltration, plasma concentration of small solutes remains
unchanged throughout the length of the glomerular capillary and is similar to that
in Bowman's space. Fluid removal across the glomerular basement membrane
(broken curve) is large (heavy arrow). Reabsorption of fluid from the renal
tubules lowers the plasma concentration of the solutes.

hemodialysis machines, because dialysate production and

many other devices are absent. One disadvantage of CAVH
and CVVH is that they are less efficient than hemodialysis
in removing urea, potassium, and other small solutes because there is no dialysate to permit diffusion. In order
to improve solute clearance, continuous dialysate flow is
sometimes added to these systems techniques known as
continuous arteriovenous hemodiafiltration (CAVHD)

and continuous venovenous hemodiafiltration (CVVHD).

Unfortunately, the addition of dialysate also adds complexity to these systems.

WATER TRANSPORT AND SOLUTE
CLEARANCE PROFILES

The effectiveness of water transport across a dialysis

membrane is measured as the ultrafiltration coefficient,
which is sometimes imprecisely called the ultrafiltration

411

63. Hemodialysis and Hemofiltration

rate. Ultrafiltration coefficients for conventional hemodialysis membranes are usually 2 to 5 mL hr~' Hg"1. With a
transmembrane pressure of 200 mm Hg, a dialyzer with a
coefficient 2.5 mL hr'1 . Hg'1 will remove 0.5 L of fluid
per hour or 2 L in 4 hours. Ultrafiltration coefficients for
high flux dialysis or hemofiltration membranes are much
higher, at 15 to 60 mL hr" 1 mm Hg"1.
The solute transport properties of dialysis membranes
are usually expressed as the mass transfer-area coefficient,
which is the product of the mass transfer coefficient (K 0)
and membrane surface area (A). The mass transfer coefficient of a dialysis membrane is similar to its diffusivity.
The clearance profile of solutes by conventional analysis
membranes presented in Fig. 2 reflects primarily the KoA
of the membranes and only provides a rough estimation
of what might be achieved clinically. The actual solute
clearances depend also on the blood flow rate, dialysate
flow rate, and removal. Diffusive clearance of solutes by
hemodialysis decreases rapidly with increasing molecular
size. For small solutes such as urea, however, removal per
unit time by hemodialysis is more efficient than that by the
native kidneys. Urea clearance is typically between 180 and
220 mL/min for hemodialysis, which is two times the value
for the glomeruli in two kidneys (often 90 to 110 mL/min
for adults). Because of the limited number of hours (usually
9 to 15) that the patient actually spends on hemodialysis per
week, the total weekly clearance of urea by hemodialysis is
I
Molecular Weight (daltons)

Urea Creatinine

far lower than that achieved by native kidneys (168

hours/week).
Clearance of larger solutes, such as /32-microglobulin
(12,000 Da), by conventional hemodialysis is much lower
than that for urea. Because of the diffusive nature of transport, clearance by high flux membranes also decreases rapidly with molecular size, albeit less rapidly compared to
conventional low flux dialysis membranes. In contrast,
clearance by hemofiltration, even using the same membranes as for high flux hemodialysis, is maintained for solutes up to 1000 Da. This is because the mode of transport
in hemofiltration is convection, which is governed by the
membrane sieving coefficient (defined as ultrafiltrate to
plasma concentration ratio) of the solute rather than its
diffusivity, and the sieving coefficient remains near 1.0 up
to 1000 Da. For higher molecular weight solutes, however, the sieving coefficients, and hence clearances, also decline.
DIALYSATE

The dialysate creates solute concentration gradients to

drive diffusion across the dialysis membrane. The typical
composition of dialysate is shown in Table 2. Sodium is
removed primarily by convection such that Ultrafiltration
of 4 L of isotonic fluid results in 560 mEq or 13 g of
sodium removal without a change in plasma sodium concentration. In fact, dialysate sodium is usually kept at concentrations similar to or higher than that in plasma to avoid
a decrease in plasma sodium concentration. In contrast,
dialysate potassium concentration is often kept low in order
to decrease plasma potassium concentration. Because dialysis patients are generally acidemic, base in the form of
either bicarbonate or acetate is offered. Acetate dialysate is
slightly less expensive but has the disadvantages of inducing
hypoxemia and a transient worsening of metabolic acidosis
because of the initial loss of bicarbonate from the plasma.
It also causes intradialytic hypotension and an ill sensation
in some patients. Proportioning systems in modern ma-

TABLE 2

Composition of Dialysate Commonly Used in

Clinical Hemodialysis
Ions
Na
K+
FIGURE 2 Solute clearance profile of various modalities. The
curves are construc ted based partially on data and partially on
theroretical projection. The actual values may vary depending on
the surface area of the membrane and operating conditions, such
as blood flow rate. Native glomeruli refer to the summation of
all the glomeruli in two normal kidneys. "Glomeruli" instead of
"kidneys" are used because tubular reabsorption lowers the renal
clearance of certain solutes (e.g., urea and glucose) substantially.
612, vitamin B]2 ; /^MG, /3 2 -niicroglobulin.

cr
HCOj
Acetate6
Ca2+
Mg2+

Glucose

Concentrations
132-145" mEq/L
0-4.0 mEq/L
103-110 mEq/L
0-40 mEq/L
2-37 mEq/L
0-3.5 mEq/L
0.5-1.0 mEq/L
0-200 mg/dL

" Higher dialysate sodium concentration is sometimes used at the beginning of the dialysis session
("sodium modeling").
b
Either HCC>3 or acetate is used primarily as
buffer.

412

Alfred K. Cheung

chines have made delivery of bicarbonate dialysate a relatively easy task. Calcium concentration in the dialysate
varies depending on the specific need of the patient. Dialysate magnesium concentration is sometimes lowered so that
the patient can take oral magnesium-containing phosphate
binders. Glucose is usually provided at 200 mg/dL to maintain the plasma glucose level stable for both diabetic and
nondiabetic patients.
VASCULAR ACCESS

An adequate vascular access for hemodialysis should permit blood flow to the dialyzer of 200 to 500 mL/min in adults,
depending on the size of the patient. For acute hemodialysis,
a large vein, such as the femoral vein, is often cannulated
with a double-lumen catheter. One lumen is for extracting
the blood from the patient (so-called "arterial" side even
though it may come from the patient's vein), and the other
returns blood to the patient ("venous" side). Femoral catheters are seldom left in place for more than one dialysis ses sion unless the patient is nonambulatory, because they are
prone to kinking, dislodgement, and infection.
For usage of one to several weeks, an indwelling doublelumen catheter is often placed in an internal jugular or
subclavian vein percutaneously. These catheters can be
infected, dislodged, or occluded by thrombi. In addition,
catheters at these sites predispose the vessels to stenosis.
Stenosis of these vessels can cause outflow obstruction and
severe swelling of the ipsilateral arm, especially if a permanent arteriovenous fistula, with arterialized venous pres sure and augmented blood flow, is present in that arm.
Subclavian veins appear to be more likely to develop stenosis than internal jugular veins and therefore should be
avoided if at all possible. Stiff catheters are incriminated
more frequently than softer ones. Softer catheters with
anchoring cuffs offer a good alternative. These catheters
are usually tunneled subcutaneously in the upper thorax
before entering a proximal vessel or, rarely, inserted via
the groin into the femoral vein. They may be used for
several weeks to over a year. Thrombosis of these semipermanent catheters can be resolved with local instillation
of thrombolytic agents. The original form of temporary
vascular access was the Scribner shunt, a plastic conduit
placed outside the skin with the two internal ends inserted
and anchored to an artery and a vein, respectively. The
disadvantage of this shunt is that it often destroys a potential site for a future permanent access. Scribner shunts are
also prone to infection and carry the potential danger of
rapid blood loss if one or both ends are dislodged. They
are sparingly used nowadays but their high blood flow rate
is an advantage in selected patients.
Long-term vascular access for hemodialysis is usually
established by the creation of an arteriovenous (AV) fistula
in an upper extremity, although a lower extremity or even
an axillary vessel may sometimes be employed. A fistula
is established by connecting an artery to a nearby vein
either by direct surgical anastomosis of the native vessels
or with an artificial vascular graft, for example, one made
of polytetrafluoroethylene (PTFE). Native fistulae are pre-

ferred over PTFE grafts because of their relative longevity

(approximately 80% vs 50% patency in 3 years) and lower
susceptibility to infection. The disadvantages are a need for
sufficiently large native veins and a 4 to 8-week maturation
period during which the wall of the fistula thickens and the
lumen enlarges before the fistula is ready for use. PTFE
grafts can be used earlier, occasionally at one week or less,
but they tend to elicit acute transient local inflammatory
reactions manifested by pain, swelling, and redness which
usually subside spontaneously within a few weeks. PTFE
grafts are also more prone to thrombosis and infection, but
the convenience that they offer has made them a popular
choice among many dialysis personnel and more common
than native fistulae in the United States. Declotting of AV
fistulae can be accomplished surgically, with local infusion
of thrombolytic agents, or, more recently, with mechanical
devices inserted percutaneously to break up the clot.
Stenosis at the outflow tract of AV fistulae occurs frequently. Partial obstruction impedes the flow of cleansed
blood from the dialyzer back to the patient; as a result, the
blood recirculates back to the "arterial" (afferent) limb of
the fistula and decreases the amount of fresh systemic blood
that can be delivered to the dialyzer. Hence, the overall
efficiency of the dialysis process is diminished. Plasma urea
concentration obtained from the arterial (afferent) and
venous (efferent) tubings of the dialyzer, as well as that
obtained from a peripheral vessel have been used to calculate the degree of blood recirculation in the dialysis circuit.
Obstruction of the fistula outflow tract also leads to an
increase in pressure inside the venous (efferent) tubing
during hemodialysis, which has been used as a clue to the
presence of fistula outflow stenosis. The reliability of both
the urea recirculation method and the venous pressure
method in detecting fistula stenosis has recently been ques tioned. Instead, other techniques involving noninvasive devices and the dilution principle have been developed to
assess the total blood flow through AV fistulae. Monitoring
changes in total fistula flow rates over time allows earlier
detection of stenosis. An angiogram (also called "fistulogram" in this context) with the injection of contrast dye
(or carbon dioxide for patients who are allergic to contrast)
remains the gold standard for the confirmation and anatomical definition of AV fistula stenosis. Fistula stenosis is
treated surgically by replacing or bypassing the stenotic
segment. Alternatively, stenosis may be relieved by angioplasty with or without the placement of a stent to keep
the lumen open. If left untreated, most stenotic fistulae
eventually become totally occluded by thrombi. AV fistulae
can be infected by both Gram positive, especially Staphylo coccus, and Gram negative bacteria. Mild infection can be
treated with antibiotics; more severe infection is treated
by resection of the fistula. Although they are more difficult
to treat conservatively, infected artificial grafts do not invariably require surgical removal.
Blood flow through the AV fistula often exceeds 1 L/
min and occasionally 2 L/min, accounting for 20 to 40% of
cardiac output, although the blood flow thro ugh the dialysis
needles inserted into the fistula is considerably lower. This
diversion of cardiac output from the capillary beds can

63. Hemodialysis and Hemofiltration

cause distal ischemia, i.e., steal syndrome. In addition, it
can precipitate or exacerbate congestive heart failure in
certain patients. Abandonment or decreasing the luminal
diameter of the fistula surgically is sometimes necessary.
Early planning and placement of a permanent native AV
fistula before the chronic renal failure patient requires
dialysis is the preferred approach. For patients with poor
veins, artificial grafts will have to be used. When early
placement of an AV fistula is not accomplished, surgical
placement of a semipermanent soft catheter in the proximal
internal jugular vein is a good alternative while waiting for
the AV fistula to mature. For patients with a short life
expectancy or when sites for placement of fistula are no
longer available, these semipermanent catheters are sometimes used indefinitely. If percutaneous temporary catheters are used, they should be placed with minimal trauma
to the vessel during insertion. The internal jugular vein on
the contralateral side of the planned AV fistula is preferrable. The temporary catheter should be removed as soon
as the permanent access becomes functional. Repeat catheterization of femoral veins for individual dialysis sessions,
especially for short periods, is a reasonable alternative.
Anticoagulation

Exposure of blood to the extracorporeal circuit activates

the clotting mechanisms. Heparin is used commonly as the
anticoagulant in clinical hemodialysis. It is often given as
an intravenous bolus of 1000 to 5000 U at the beginning
of the session, followed by either continuous infusion or
intermittent boluses at 500 to 2000 U/hr up until the last
hour of dialysis. In some dialysis units, the activated clotting
time or partial thrombin time is measured periodically for
determining heparin requirement and monitoring for individual patients. In patients in whom bleeding is likely or
potentially disastrous, for example, immediate postoperative patients, lower dose heparin (termed by some "tight"
heparin or "minimum" heparin) is used with careful observation of the extracorporeal circuit for clotting. Hemodialysis can sometimes be performed without using any anticoagulants. Underlying coagulation defects, high blood flow
rates, and periodic flushing of the blood compartment with
saline help prevent clotting under these circumstances.
Regional heparinization involves infusion of heparin into
the arterial blood entering the dialyzer, and neutralizing
the heparin effect by infusion of protamine sulfate in the
venous line. Regional citrate is a similar technique except
that citrate and calcium are used to produce and reverse
anticoagulation, respectively. These procedures are relatively cumbersome and are employed infrequently, although some have advocated their use in patients with a
high risk of bleeding in the acute setting. The antiplatelet
agent prostacyclin has also been used sucessfully by a few
centers as a systemic anticoagulant during clinical hemodialysis. Hypotension can be a significant side effect of this
medication.
INDICATIONS FOR HEMODIALYSIS
AND HEMOFILTRATION

Acute hemodialysis is primarily performed for renal failure and drug overdose. Indications for emergency dialysis

413
in the acute renal failure setting include fluid overload,
hyperkalemia, metabolic acidosis, and uremic signs and
symptoms (see Chapter 41 for more details). Initiation of
dialysis prior to onset of these problems is preferable. If
reversal of the acute renal failure does not appear to be
imminent, prophylactic dialysis is often instituted when the
BUN is around 70 to 80 mg/dL or estimated glomerular
filtration rate is 5 to 10 mL/min, although some have advocated the initiation of dialysis even earlier. Maintenance
dialysis for chronic renal failure is usually started at similar
glomerular filtration rates, unless the clinical condition dictates earlier intervention. Although hemofiltration is more
effective in removing larger solutes than hemodialysis, the
clinical indications for this form of therapy on a chronic
intermittent basis have not been precisely defined.
Continuous extracorporeal therapies, such as CAVH and
CVVH, are particularly useful for patients in the intensive
care unit whose cardiovascular status is too unstable for
rapid fluid removal, as may occur during intermittent hemo dialysis. They are also used in patients from whom removal
of substantial amounts of fluid on a continuous basis is
desired, e.g., patients with multiple trauma receiving parenteral nutrition, blood products, and various intravenous
medications. Clearances of urea and potassium by CAVH
and CVVH are sometimes inadequate to maintain plasma
concentrations of these solutes in the desirable range. Under these circumstances, continuous dialysate flow is added
to the system, i.e., CAVHD or CVVHD is employed. Preliminary results from a randomized trial have suggested
that the employment of continuous ext racorporeal therapies is associated with a greater rate of kidney recovery
from acute renal failure and shorter hospitalization stay
than intermittent hemodialysis, although there was no difference in mortality rates. Peritoneal dialysis is another
form of continuous therapy that can be used in patients
suffering from acute renal failure with unstable hemodynamics, but for technical reasons it has been largely replaced by continuous extracorporeal modalities in this
setting.
QUANTITATION OF HEMODIALYSIS

The amount of chronic hemodialysis that should be delivered to patients has not been well defined. Not infrequently,
an arbitrary duration, blood flow rate, and size of the dialyzer are used based on the experience and intuition of the
nephrologist. Removal of fluid to keep the patient euvolemic, or slightly hypovolemic, after dialysis is often desirable, but this so-called "dry weight" for individuals is often
defined arbitrarily as the weight below which the patient
develops symptomatic hypotension or muscle cramps.
There are, of course, many imprecisions associated with
this approach. For example, the likelihood of developing
hypotension depends not only on the amount of fluid removed, but also on the rate of fluid removal.
Normalization of plasma electrolytes, such as potassium
and hydrogen ions, is obviously important. Guidelines for
removal of other uremic toxins are, however, more difficult
to establish. Urea has been widely used as a marker to

414

Alfred K. Cheung

guide dialysis because it is an index of the production and

accumulation of all nitrogenous waste products derived
from protein metabolism. In addition, the removal of urea
by hemodialysis appears to correlate with clinical outcome.
In urea kinetic modeling, the index Kt/V is often used
for quantitation of the dose of dialysis therapy; K is the
hemodialyzer clearance, t is the duration of the dialysis
session, and V is the volume of distribution of urea in the
body. A Kt/V value of 1.2 is currently considered to be the
minimum amount. Because Kt/V is tedious to determine
precisely, some have simply used the decrease in BUN
during dialysis as an alternative guide. A postdialysis to
predialysis BUN ratio of 0.35 or a "urea reduction ratio"
[calculated as (predialysis BUN minus postdialysis BUN)
divided by predialysis BUN] of 65% is roughly equivalent
to a Kt/V of 1.2. There are some suggestions that a larger
dose of hemodialysis, for example, a Kt/V value of 1.6 to
1.8, is beneficial, although definitive data supporting this
notion are lacking. Delivery of large doses of hemodialysis
requires an adequate vascular access to provide high blood
flow rates, a highly efficient dialyzer, and/or the willingness
of the patient to endure longer dialysis sessions. It must
be emphasized that the use of urea kinetics does not substitute for frequent and diligent clinical evaluation of the
patient. Increasing the amount of dialysis should be considered if the patient exhibits uremic symptoms, regardless of
the Kt/V value. In fact, urea kinetics are not used as the
principal guide for hemodialysis therapy in some centers
in Europe, although they are gradually gaining popularity.
The clinical impact of removal of larger solutes (middle
molecules) is at present unclear, although limited data suggest that they are also important.
Quantitation of acute hemodialysis is even more empirical. The dosage of dialysis provided for the treatment of
exogenous toxins, for example, salicylates and lithium, is
often guided by the plasma levels of the toxin and the
clinical status. Dosage of dialysis for acute renal failure
has been guided by the plasma chemistry, such as BUN,
potassium, and bicarbonate, body fluids, and other clinical
status, with the objective of maintaining the BUN below
~80 mg/dL most of the time. Recently, some nephrologists
have advocated the quantitation of hemodialysis for acute
renal failure using urea kinetics as well, although it is unclear if Kt/V values similar to those targeted for chronic
dialysis are also appropriate in the acute setting. The significance of middle molecule removal in acute renal failure
has not been carefully studied.
COMPLICATIONS OF HEMODIALYSIS

Although hemodialysis is nowadays a relatively safe pro cedure, Several complications may still arise. Some are
inherent side effects of the normal extracorporeal circuit;
some result from technical errors, and others are due to
abnormal reactions of patients to the procedure. Intradialytic hypotension is common and has been attributed variably to body volume depletion, shifting of fluid from extrato intracellular space as a result of decrease in serum osmolality induced by dialysis, impaired sympathetic activity,

vasodilation in response to warm dialysate, as well as splanchnic pooling of blood while eating during dialysis. Treatments include avoiding large interdialytic fluid gain, administration of normal saline, hypertonic saline, hypertonic
glucose, mannitol, or colloids, decreasing dialysate temperature to produce vasoconstriction, and avoidance of eating during dialysis. Isolated ultrafiltration, which removes
fluid in the absence of dialysate and therefore does not
reduce the plasma osmolality, and "sodium modeling",
which tailors dialysate sodium concentrations (135 to
160 mEq/L) throughout the dialysis session, are sometimes
useful to minimize hypotensive events. Cardiac arrhythmias may occur as a result of rapid electrolyte changes,
especially in patients taking digitalis and dialyzed against
very low potassium dialysate (0 to 1 mEq/L). Arrhythmias
can induce or aggravate hypotension and overt or silent
myocardial ischemia.
Muscle cramps, nausea, and vomiting occur commonly
during hemodialysis, and are often a result of rapid fluid
removal. Too rapid removal of urea and other small solutes
may lead to acute dysfunction of the central nervous system, sometimes referred to as disequilibrium syndrome
(see Chapter 71). These patients may experience headache
with nausea and vomiting, altered mental status, seizures,
coma, and death. The pathophysiology of this symptom
complex has not been well defined. Current evidence suggests that it may at least be partially attributed to the rapid
decrease in plasma urea concentration and osmolality, with
consequential fluid shifts into the intracranial compartment
that cause cerebral edema. Severe disequilibrium syndrome
is rare nowadays because hemodialysis is usually initiated
at an early stage when the BUN is still not very high, and
the efficiency of solute removal is often deliberately limited
during the first treatment session.
Anaphylactoid reactions during hemodialysis are rare.
They are manifested by various combinations of hypertension or hypotension, pulmonary symptoms, chest and abdominal pain, vomiting, fever, chills, flushing, urticaria, and
pruritus. Cardiopulmonary arrest and death occasionally
follow. The etiologies are probably multifactorial and may
involve activation of the plasma complement or kinin systems by dialysis membranes, administration of angiotensin
converting enzyme inhibitors concomitant to dialysis using
membranes that intensely activate kinins, or the release of
noxious materials that have contaminated the dialyzers
during the manufacturing or sterilization process. Treatment for these anaphylactoid reactions is largely symptomatic. Preventive measures include thorough rinsing of the
dialyzer before use and avoidance of the type of dialyzer
and/or medications to which a particular patient is hypersensitive. Dialyzers or dialysates that are contaminated
with microorganisms or their toxins can rarely cause fever
and infection. Hepatitis B was prevalent in the 1970s,
whereas hepatitis C infection is more common in hemodialysis units nowadays. The mode of transmission of hepatitis
C in dialysis units has not be well established.
Hypoxemia occurs commonly during hemodialysis when
acetate (instead of bicarbonate) is used as the dialysate
buffer. The primary mechanism appears to be the initial

63. Hemodialysis and Hemofiltration

loss of bicarbonate and carbon dioxide by diffusion into

the dialysate, with subsequent hypoventilation. Dialysis
membrane bioincompatibility may play a role by releasing
mediators that impair gas exchange in some instances. A
decrease in systemic partial oxygen pressure of 10 to
12 mm Hg is not uncommon, which could be deleterious
for patients with underlying cardiopulmonary disease.
An array of technical errors associated with hemodialysis
has been described, but fortunately they occur rarely. Inadequate purification of city water prior to use may result in
high levels of contaminants in the dialysate, resulting in,
for example, aluminum or calcium intoxication. Dialysate
contaminated with chloramine and improper proportioning
or overheating of dialysate by the dialysis machine also
lead to hemolysis. Rupture of the dialysis membrane causes
blood loss into the dialysate and entry of microorganisms
from the dialysate into the blood. Defective blood circuit
and monitoring devices may result in air embolism. Difficult
or improper puncture with the dialysis needle may cause
a local hematoma around the vascular access or external
bleeding, which can be aggravated by intradialytic adminis tration of heparin.
DIALYZER REUSE

Hemodialyzers can be reused repeatedly on the same

patient after thorough cleansing and sterilization, employing various agents such as formaldehyde, glutaraldehyde,
sodium hypochlorite (bleach), and the combination of hydrogen peroxide and peroxyacetic acid. The blood compartment must be thoroughly rinsed to remove all the sterilants prior to the next use, because residual sterilants
infused into the body can be harmful. Inadequate sterilization, on the other hand, has been associated with infection
by microorganisms. In general, reused dialyzers can clear
small solutes as effectively as new dialyzers unless a substantial portion of the hollow fibers is occluded by clotted
blood; in contrast, middle molecule clearance is sometimes
impaired even when small solute clearance is maintained.
Fiber bundle volume is checked after each processing, and
the dialyzer is discarded if the volume is below 80% of that
of a new dialyzer. Other contraindications to reuse include
a disrupted dialyzer casing and hepatitis B infection. Reports on the effect of reuse on long-term clinical outcome
are conflicting. Because of the economic benefits and lack
of definite harmful effects when practiced properly, dialyzer reuse is popular in the United States.
CHOICE OF HEMODIALYZERS

There are several considerations when choosing a hemo dialyzer for clinical use. One important consideration is
the capacity (KoA) of the membrane to clear urea, because
urea removal by dialysis has been shown to correlate positively with patient survival and United States federal regulations have set criteria for the minimum urea removal.

415

Increasing amounts of data suggest that the removal of

middle molecules is also beneficial. Biocompatibility characteristics of dialysis membrane are taken into account
by some, albeit not all nephrologists. Therefore, high flux
dialysis membranes, which are more effective in removing
middle molecules and are in general more biocompatible,
have gained popularity in recent years. Purchase cost and
reusability of the dialyzer are additional concerns.
DRUG USAGE IN HEMODIALYSIS

Hemodialysis removes certain medications which are intended for therapeutic use. The removal of a drug depends
on the properties of the drug as well as the conditions of
the dialysis procedure (see Chapter 43). Guidelines for
dosing medications in renal failure with or without dialysis
have been published. It is imperative to refer to these
publications for individual drugs if the physician is unfamiliar with their use in these settings. For example, different
types of pencillins behave differently and the clearance of
a drug by hemodialysis can be substantially different from
that by peritoneal dialysis. It is also important to remember
that these publications provide only a rough guideline. The
information might be derived from conventional hemodialysis and might not be applicable to high flux dialysis. Finally, the efficacy of the particular dialysis session must be
take into account. A short and inefficient dialysis because
of vascular access problems would remove only small
amount of aminoglycosides and the postdialysis supplemental dose should be adjusted accordingly. Frequently,
monitoring of drug levels is required.
Bibliography
A symposium on practical issues in the use of continuous renal
replacement therapies. Sem in Dialysis 9:79-223, 1996.
Ahmad S, Blagg CR, Scribner BH: Center and home chronic
hemodialysis. In: Schrier RW, Gottschalk CW (eds) Diseases of
the Kidney, 5th ed. Little, Brown, Boston, pp. 3031 -3068, 1993.
Cheung AK, Leypoldt JK: The hemodialysis membranes: A historical perspective, current state and future prospect. Sem in Nephrol, 17:196-213, 1997.
Current concepts in the creation, maintenance, and preservation
of hemodialysis access. Sem in Dialysis 9(suppl 1):S1 -S54,1996.
Feldman HI, Kobrin S, Wasserstein A: Hemodialysis vascular access morbidity. / Am Soc Nephrol 7:523-535, 1996.
Hakim RM, Wingard RL, Parker RA: Effect of the dialysis membrane in the treatment of patients with acute renal failure.
N Engl J Med 331:1338-1374, 1994.
Held PJ, Port FK, Wolfe RA, et al.: The dose of hemodialysis and
patient mortality. Kidney Int 50:550-556, 1996
Lowrie EG, Laird NM, Parker TF, Sargent J A: Effect of hemodialysis prescription on patient morbidity: Report from the National
Cooperative Dialysis Study. TV Engl]Med 305:1176-1181,1981.
Nissenson AR, Fine RN, Gentile DE (eds): Clinical Dialysis, 3rd
ed. Appleton & Lange, Norwalk, 1995.
Suki WN, Massry SG (eds): Therapy of Renal Diseases and Related
Disorders, 2nd ed. Kluwer Academic Publishers, Boston, 1991.
Vanholder RC, Ringoir SM: Adequacy of dialysis: A critical analysis. Kidney Int 42:540-558, 1992.

PERITONEAL DIALYSIS
BETH PIRAINO

Peritoneal dialysis (PD) is a form of dialysis in which

dialysate (content shown in Table 1) is instilled in the
peritoneal cavity, periodically drained, and replaced with
fresh solution via a catheter. The forms of PD are listed
in Table 2. With continuous ambulatory peritoneal dialysis
(CAPD), the patient manually exchanges the dialysate four
to five times each day, as shown in Fig. 1. With continuous
cycler peritoneal dialysis (CCPD), the patient connects the
catheter to the cycler at bedtime. The cycler delivers a
set number of exchanges over 8 to 10 hours, the last fill
constituting the long day dwell. In the morning the patient
disconnects the catheter from the cycler. The increasing
popularity of CCPD (35% of all PD utilization in the United
States in 1995) may be due to the convenience of performing the dialysis connections only twice per day and to the
new cycler models which are smaller and more attractive
to patients.

than positively charged solutes such as potassium. Macro molecules such as albumin cross the peritoneum by mechanisms not completely understood, but probably via lymphatics and through large pores in the capillary membranes.
Dialysate contains a high concentration of dextrose.
Thus, the dialysate is hyperosmolar compared to the serum,
causing ultrafiltration, or fluid removal, to occur. The volume of ultrafiltration depends on the dextrose solution
used for each exchange (Fig. 3), the length of the dwell,
and the individual patient's peritoneal membrane characteristics (see below). With increasing dwell time, transperitoneal glucose absorption diminishes the dialysate glucose
concentration and the osmotic gradient. Ultrafiltration is
therefore decreased with long dwell times, for example
with the overnight exchange on CAPD or the daytime
exchange on CCPD. As the glucose concentration gradient
falls during a dwell, the hydrostatic pressure created by
the dialysate within the abdominal cavity overcomes the
osmotic gradient, resulting in fluid reabsorption into the
systemic circulation. In addition, continuous lymphatic absorption diminishes net fluid removal. A newer osmotic
agent, icodextrin, which is a high molecular weight compound allowing continuous ultrafiltration for up to a 12hour dwell time, appears promising as an alternative to
dextrose, but is not yet commercially available.
The rate of movement of small solutes such as creatinine
between dialysate and blood differs from one patient to
another. Peritoneal function characteristics have been
quantified in the peritoneal equilibration test (PET). In
this standardized test, 2 L of 2.5 g/dL dextrose exchange
are infused and the dialysate:plasma creatinine ratio at the
end of a 4-hour dwell (D/Pcr ) is measured. Using this test
each patient's peritoneal membrane can be categorized as
having high (DIPa > 0.81), high average (0.65 to 0.81),
average (0.65), low average (0.50 to 0.65) or low (<0.5)
peritoneal transport. Patients with a high D/Pcr have rapid
clearance of small molecules but poor ultrafiltration due to
dissipation of the concentration gradient between dialysate
and blood by glucose absorption. Patients with a low Dl
Pcr have lower clearances and generally require increased
numbers of dialysis exchanges and/or increased volume per
exchange to avoid uremia once residual renal function is
lost. Ultrafiltration in this category of patient is usually
excellent. The majority of patients have high average, average, or low average peritoneal transport and do well on
either CAPD or CCPD.

PERITONEAL MEMBRANE

Dialysis involves the movement of small solutes and wa ter across a semipermeable membrane. The peritoneal
membrane is the dialyzing surface in peritoneal dialysis.
The visceral peritoneal membrane covers the abdominal
organs, while the parietal peritoneum lines the abdominal
cavity. The peritoneal membrane consists of a single layer
of mesothelial cells overlying an interstitium in which the
blood and lymphatic vessels lie. The mesothelial cells are
covered by microvilli that markedly increase the 2 m2 surface area of the peritoneum. The normal peritoneum is
only a few millimeters in thickness.
During PD, solutes, such as urea nitrogen, creatinine,
and potassium, move from the peritoneal capillaries across
the peritoneal membrane to the peritoneal cavity, whereas
other solutes, such as bicarbonate and calcium, move from
the dialysale into the peritoneal capillaries. Clearance is
achieved by both diffusion and convection. Small solute
movement is mainly by diffusion and is thus based on the
concentration gradient of the solute between dialysate and
blood. Thus, substances with a low molecular weight such
as creatinine, urea, and potassium, which are not in the
infused dialysate, are effectively removed by PD. With
increasing dwell time, the ratio of dialysate to serum urea
levels approaches one (Fig. 2). Because the peritoneal
membrane has a negative charge, negatively charged solutes such as phosphate will move across it more slowly

Primer on Kidney Diseases, Second Edition

416

Copyright I 998 '& by the National Kidney Foundation

All rights of reproduction in any form reserved.

64. Peritoneal Dialysis

417

D
r
a
Dialysate

TABLE I
Composition of Peritoneal Dialysis Fluid
Sodium (mEq/L) Chloride
(mEq/L) Lactate (mEq/L)
Calcium (mEq/L) Magnesium
(mEq/L) Dextrose (g/dL)

132
102, 96, or 95
35 or 40
2.5 or 3.5
0.5 or 1.5
1.5, 2.5, or 4.25

m Bag

PERITONEAL
CATHETERS

The access for PD is a catheter, which is inserted into

the abdominal cavity by either a surgeon or a nephrologist,
generally using local anesthetic with sedation. Although
there are numerous newer catheter designs, none offers a
significant advantage over the original double-cuffed silastic Tenckhoff catheter which is still used most often. The
intra-abdominal portion of the catheter has multiple perforations through which dialysate flows. With the deep cuff
placed in a paramedian position in the rectus muscle, the
catheter is tunneled through the subcutaneous tissue to
exit laterally and downward. The subcutaneous superficial
cuff is located about 3 cm from the exit site of the catheter.
PD can be initiated immediately if exchange volumes are
small and the patient is kept recumbent. Alternatively, PD
may be deferred until the insertion site is well healed, at
which time the patient is trained to perform CAPD or
CCPD. Hemodialysis can be used if necessary as a temporary measure until PD is initiated.
Several potential catheter-related problems exist. A dialysate leak usually presents as clear fluid draining from the
exit site. The leak can be confirmed by a strongly positive
dipstick for glucose. CAPD is discontinued and hemodialysis used temporarily. Edema of the perineum, penis, scrotum, or abdominal wall generally indicates a subcutaneous
dialysate leak. The location of the leak can usually be
identified using CT with intra-abdominal contrast, following which the defect can be surgically closed. Inadequate
flow in a catheter may be due to constipation (which responds to laxatives), catheter obstruction with fibrin or
omentum, or poor catheter position. Ideally the catheter.
tip should be in the patient's pelvis. A poorly positioned
catheter can sometimes be manipulated using a wire or via

/
/ Small
I Volume
Flush

FIGURE
I
Diagram
of
a
CAPD exchange
using
the Y-set. Between
exchanges,
the
abdomen contains
dialysate and only a
short,
capped extension
tubing attached to
the peritoneal catheter is present. The Y-set consists of tubing
with a full bag of dialysate at one end (hanging at top in the
picture) and an empty drainage bag which is placed on the floor.
Clamps on the tubing direct the flow of fluid by gravity. To begin
the exchange, the patient connects the Y tubing to the short
exchange tubing attached to the catheter. The patient flushes air
from the tubing (Step 1) by opening the clamp to the full
dialysate bag and the drainage bag (keeping the peritoneal
catheter clamped) for a count of 10. In Step 2, the clamp to the
full bag is closed, and the clamp on the peritoneal catheter opened
to allow the spent dialysate to flow into the drainage bag. This
takes approximately 10 minutes. In Step 3, the patient closes the
clamp to the drainage bag and opens the clamp on the tubing
leading to the filled dialysate bag, thus allowing fresh dialysate to
drain into the abdomen. The final step is to close the clamp on the
peritoneal catheter extension tubing, disconnect the Y tubing,
and cap the short extension tubing.

TABLE 2

Forms of Peritoneal Dialysis

Manual procedures
Continuous ambulatory peritoneal dialysis (CAPD)
Daytime peritoneal dialysis (DPD)"
Automated procedures
Continuous cycling peritoneal dialysis (CCPD)
Nocturnal intermittent peritoneal dialysis (NIPD)"
Intermittent peritoneal dialysis (IPD)

Three to four day time dwells plus a long bedtime exchange

Four short day dwells, no night dwell
One to two long day dwells, three to four short nighttime exchanges No
day dwell, multiple short nighttime exchanges
Rapid cycling on an intermittent basis (usually three to four times per week), no
dialysis between treatments

" Clearance inadequate unless the patient has a highly permeable membrane (high D/P cr) and/or significant residual renal function.

418

Beth Piraino

rifampin 600 mg every day for 5 days every 12 weeks are

all effectiv e in reducing S. aureus catheter infections. Pseudomonas aeruginosa catheter infections are also difficult
to resolve and frequently relapse. Ciprofloxacin is used to
treat P. aeruginosa catheter infections, but if P. aeruginosa
peritonitis develops, the catheter must be removed to resolve the infection.
PERITONITIS

DWELL TIME, (HOURS)

FIGURE 1 Dialysate to serum ratio over the period of a dwell

using 1.5 g/dL dextrose, 2 L volume. From Popovich RP, Moncrief
JW, Nolph KD, et al.: Continuous ambulatory peritoneal dialysis.
Ann Intern Med 88:453, 1978; and Nolph KD, Twardowski ZJ,
Popovich RP, Rubin J: Equilibraton of peritoneal dialysis solutions
during long dwell exchanges. / Lab Clin Med 93:246-256, 1979.
Reproduced with permission of Mosby -Ycar Book, Inc., St. Louis,
and American College of Physicians, Philadelphia.

laparoscope to a better position, but catheter replacement

may be required.
Peritoneal catheter infections include infections of the
exit site (erythema or purulent drainage from the exit site)
and tunnel infections (edema, erythema, or tenderness over
the subcutaneous pathway). Staphylococcus aureus is responsible for the majority of catheter infections. S. aureus
exit site infections are difficult to treat (Table 3), with
frequent progression to tunnel infections and peritonitis,
in which case catheter removal is required for resolution.
S. aureus nasal carriage is associated with an increased risk
of S. aureus catheter infections. Treatment of nasal carriers
with intranasal mupirocin twice a day for 5 days each
month, mupirocin applied daily to the exit site, or oral
DWELL TIME, (HOUR)

800 -

FIGURE 3 Ultrafiltration with 1.5, 2.5, and 4.25 g/dL dextrose

dialysate, 2 L volume. From Twardowski ZJ, Khanna R, Nolph
KD: Nephron 42:93-101, 1986. Reproduced with permission of S.
Karger AG, Basel.

Peritonitis is defined as cloudy dialysate in which there

are 100 white blood cells/mm3, more than 50% of which are
polymorphonuclear. The patient generally has abdominal
pain, but may or may not have fever, nausea, or vomiting.
Bacteremia is rare. Gram stain of the effluent is seldom
helpful, except for fungal peritonitis, but cultures are generally positive. In many centers, 20% of peritonitis episodes
result in no growth, predominately due to inadequate culture techniques. The etiologies of peritonitis are given in
Table 4.
Peritonitis remains an important cause of hospitalization,
catheter loss, and transfer to hemodialysis. Peritonitis rates,
originally very high, have decreased to less than 1.0 episodes per dialysis year due to improvements in the procedure for performing the dialysis connections, which have
decreased the risk of peritonitis due to contamination (Fig.
1). Because the flush and drain occur after the connection
to the catheter, small numbers of bacteria present because
of touch contamination at the connection site (predominately S. epidermidis; Table 3) may be flushed into the
drain bag instead of into the patient's abdominal cavity.
As a result, peritonitis rates due to S. epidermidis have
decreased, and S. aureus and enteric organisms account for
a larger proportion of peritonitis episodes than before.
Because patients with these organisms are more symptomatic than those with S. epidermidis peritonitis, peritonitis
has become a less frequent, but more severe complication,
often requiring hospital admission. The catheter removal
rate for peritonitis depends on the infecting microorganism.
Peritonitis with S. epidermidis is less likely to result in
catheter loss than peritonitis due to S. aureus or P. aeruginosa. Fungal peritonitis generally requires catheter removal.
The initial treatment of peritonitis is empiric and designed to cover both Gram positive cocci and Gram negative bacilli. A first-generation cephalosporin, such as cefazolin or cephalothin, may be used in conjunction with an
aminoglycoside or a third-generation cephalosporin, with
subsequent therapy tailored to the culture and sensitivity
results. A listing of antibiotics and the dosing schedule is
given in Table 5. Because of the concern about the emergence of vancomycin-resistant organisms, vancomycin use
should be restricted to treatment of methicillin-resistant
organisms or for patients allergic to cephalosporins.
PERITONEAL DIALYSIS PRESCRIPTION AND
DETERMINATION OF ADEQUACY

Adequacy of PD is determined by both clinical assessment and clearance measurements. The well-dialyzed pa-

64. Peritoneal Dialysis

419
TABLE 3
Treatment of Exit Site and Tunnel Infections
Gram positive catheter infections
Cephalexin Oxacillin
Triraethoprim/sulfamethoxazole
Add rifampin if slow to resolve"
Gram negative catheter infections6
Ciprofloxacin Fleroxacin Ofloxacin
Add ceftazidime if slow to resolvec

250-500 mg po qid, or 250500 mg po qid, or 160/800320/1600 mg po qd 300 mg po

bid
500 mg bid
800 mg po, then 400 mg po qd
400 mg po, then 200 mg po qd
500 mg/L, then 125 mg/L in each exchange

" For Staphylococcus aweus.

Cephalexin or trimethoprim/sulfamethoxazole may also be options. Choice based on
sensitivities; quinolones will be needed for pseudomonas infections. Antacids will interfere
with the absorption of quinolones.
' For Pseudomonas aeuruginosa.

tient has a good appetite, no nausea, minimal fatigue, and

feels well. In contrast, the uremic patient is anorectic with
dysgeusia, nausea, and complains of fatigue. In addition to
these clinical parameters, urea and creatinine clearances
based on 24-hour collections of effluent and urine are used
as laboratory measures of adequacy (Table 6). The clearance of urea nitrogen divided by the urea distribution volume (V) is termed Kt/V. Kt is obtained by multiplying the
effluent: blood urea nitrogen concentration ratio by the 24hour effluent drain volume. (Note that the / in Kt refers
to time and is a carryover from hemodialysis in which the
urea nitrogen clearance is multiplied by the time of the
hemodialysis treatment. This is not applicable to a continuous form of dialysis, but the term Kt/V continues to be
frequently used in continuous peritoneal dialysis to represent K/V). Renal urea nitrogen clearance is added to this.
The daily value is multiplied by 7 to provide a weekly
value. V can be estimated as 60% of weight in males and
55% of weight in females. The creatinine clearance is also
obtained from a 24-hour collection of dialysate, to which is
added the average of the renal creatinine and urea nitrogen
clearance (since creatinine clearance overestimates glomerular filtration rate due to the tubular secretion of creatinine). An adjustment for body surface area is also required
The current recommendations are given in Table 6. These
recommendations are still preliminary and will no doubt
be revised as more research is done in this area.
Clearance of small molecules such as urea nitrogen and
creatinine is dependent on the patient's peritoneal membrane characteristics, the length of the dwell time, and the

TABLE 4

Causes of Peritonitis
Contamination at the time of an exchange
Catheter tunnel infection Transmural, across
bowel wall Bacteremia (rare) Vaginal (rare)

volume of effluent drained per unit time. Many CAPD

patients are prescribed 4 exchanges of 2 L volumes per
day. Four 2 L CAPD exchanges per day with 2 L of ultrafiltration per day represents a drain volume of 70 L/week,
which is inadequate in the absence of renal function for
most patients. Initially, most patients have residual renal
function contributing to the total clearance. As renal function is lost, patients require larger exchange volumes (2.5
or 3 L) and may also need 5 daily exchanges to avoid
uremic symptoms and reach the target values of Kt/V and
creatinine clearance. The fifth exchange may be provided
by use of a device that will deliver a middle-of-the-night
exchange. Larger patients should be started on 2.5 or 3 L
exchange volumes. To achieve adequate clearances on cycler PD, one or two day dwells are needed in addition to
three to four nocturnal exchange volumes of 2.5 to 3 L.
ULTRAFILTRATION FAILURE

PD is an extremely effective method for fluid removal

for most patients. The continuous nature of CAPD allows
a more liberal intake of sodium and water than is possible
for patients on intermittent forms of therapy such as hemo dialysis. The patient adjusts ultrafiltration by chosing the
correct concentration of dextrose dialysate dependent on
fluid intake and volume status.
Poor ultrafiltration occurs in patients who have a high
dialysate: serum creatinine ratio, as previously described,
due to rapid dissipation of the glucose concentration gradient. These patients are best managed with frequent short
dwells and elimination of long dwells, such as with either
diurnal or nocturnal PD (Table 2). Mortality in this group
is higher than in other patients on PD and transfer to
hemodialysis may be necessary.
Some patients initially have normal ultrafiltration but
after a period of time develop ultrafiltration failure due to
changes in the peritoneal membrane. A severe form of
ultrafiltration failure occurs in sclerosing encapsulating
peritonitis, a misnomer because this pathologic condition
is not actually peritonitis, but characterized by marked

Beth Piraino

420
TABLE 5

Antibiotic Doses Used to Treat Peritonitis"

Antibiotic
Ampicillin Aztreonam Cefazolin or
cephalothin Ceftazidime Fluconazole
Gentamicin, netilmicin, tobramycin"
Metronidazole Vancomycin

Initial dose
(mg/L)
125 1000 500
250 200 mg po
0.6 mg/kg 500
mg po/iv 15-30
mg/kg

Subsequent doses
mg/L continuously
(each exchange)
125
250
125
125
4
25

Subsequent doses mg/L

intermittently (one
exchange per day)
No data 1000 qd 500 qd 1000
qd 200 mg po qd 0.6
mg/kg/qd or 20 mg/L qd 500
mg po/iv tid 15-30 mg/kg q
5-7 d

" Intraperitoneal unless otherwise indicated. Patients with residual renal function may require higher doses or more frequent
dosing than once daily. Limited data on use of aminoglycosides in PD patients.

thickening of the peritoneum. Sclerosing peritonitis occurs

in 1 to 2% of the chronic PD population. Multiple etiologies
for sclerosing peritonitis have been identified including
chlorhexidine (used at one time to sterilize connection
sites), acetate (as opposed to the currently used dialysate
which contains lactate), practolol (an antihypertensive
medication), and infectious peritonitis, especially episodes
occurring late in the course of PD. Once sclerosing peritonitis develops, PD can no longer be continued successfully.
NUTRITION IN THE PERITONEAL DIALYSIS PATIENT

Forty percent of PD patients are protein malnourished.

This is in part due to losses of amino acids and protein in
the dialysate, the latter generally 8 to 10 g/d. Peritonitis
markedly increases dialysate protein losses. The patient's
appetite may be suppressed from the absorbed dialysate
dextrose as well as from uremia. Both Kt/V and weekly
creatinine clearance correlate, albeit weakly, with protein
intake, suggesting that a certain minimum dose of dialysis
is required for adequate protein intake. The serum albumin
level is inversely related to both mortality and hospitalization in PD patients. A protein intake of at least 1.2 g kg'1
day" 1 is recommended for PD patients, but many patients
ingest only 0.8 to 1.0 g kg" 1 day" 1. Amino acid dialysate
(in which amino acids replace the dextrose) has been used
on a limited basis as a means of correcting protein malnutrition, but proof of its long-term efficacy is so far lacking.
The calories absorbed from dialysate glucose depend on
the dextrose concentration used (1.5, 2.5, 4.25 g/dL), as
well as on the patient's membrane permeability. The development of obesity, therefore, is not unusual in PD patients,
especially those already overweight at the start of dialysis.
In addition, glucose absorption frequently results in hyperlipidemia, which may contribute to cardiovascular disease.
DIABETICS ON PERITONEAL DIALYSIS

Diabetic glomerulosclerosis is the most common cause

of renal failure in CAPD/CCPD patients. The vast majority
of diabetic patients require insulin on CAPD/CCPD, even

if they did not require insulin prior to the initiation of PD.

This is partly due to glucose absorption from the dialysate
and the associated weight gain. Insulin can be given to
CAPD patients via the intraperitoneal route, the subcutaneous route, or a combination of both. If given intraperitoneally, the patient checks a fingerstick glucose at the time
of each exchange and adds regular insulin to the exchange
based on a sliding scale. Patients on CCPD generally require long-acting subcutaneous insulin (with or without
intraperitoneal regular insulin) for adequate glucose control. The use of intraperitoneal insulin has not been shown
to increase the risk of peritonitis.
OUTCOME IN PERITONEAL DIALYSIS PATIENTS

The death rate of PD patients in the United States from

1991 to 1993 was 30 per 100 dialysis years at risk. The
leading causes of death are cardiovascular disease and infections. In the United States, patients over age 55 had a
greater risk of death on PD compared to hemodialysis, but
these results have not been seen in other countries such as
Italy and Canada, whose PD patients have lower mortality
rates than those in the US. Risk factors for death on PD
include increasing age, the presence of cardiovascular dis ease and/or diabetes mellitus, decreased serum albumin
level, poor nutritional status, and inadequate dialysis. For
CAPD patients, a decrease in the Kt/V by 0.1 U/week
increases mortality by 6%, while a decrease in creatinine
clearance by 5 L week" 1 1.73 m" 2 increases mortality
by 7%.
A high proportion of patients transfer off PD to hemodialysis for a multitude of reasons including peritonitis or exit
site infection, catheter malfunction, inability to perform the
dialysis procedure, and inadequate clearance or ultrafiltration (particularly with loss of residual renal function). In
many cases the patient who loses a catheter due to either
peritonitis or a catheter infection may elect to remain on
hemodialysis permanently. The increasing use of the dis connect systems is associated with improved technique survival on CAPD, primarily due to lower peritonitis rates.

421

64. Peritoneal Dialysis

TABLE 6
Adequacy Measures in Peritoneal Dialysis

Solute

Adjusted to size

Minimum weekly goal

Urea nitrogen clearance

Creatinine clearance

V, distribution of urea
Body surface area

Kt/V = 2.1
Ccr/1.73 m 2 = 60-70 L

Transplantation is the goal for most patients on dialysis.

The allograft and patient survivals of PD patients who are
transplanted are similar to those of transplanted hemodialysis patients. If the transplant does not initially function,
PD can be continued, as long as the peritoneal cavity was
not entered. The peritoneal catheter is generally left in
place for 2 to 3 months until the graft is functioning well.
PERITONEAL DIALYSIS FOR ACUTE RENAL FAILURE

Intermittent peritoneal dialysis (IPD) may be succesfully

used to manage patients with acute renal failure. In this
case, the peritoneal catheter is often inserted percutaneously using a stylet, without a subcutaneous tunnel (which,
however, increases the risk of a leak). Rapid exchanges
are done to maximize small solute clearance, often 1 to 2
exchanges per hour, using a cycler. The patient may be
kept on the cycler for 48 hours or even longer, or IPD may
be performed daily for 10 to 12 hours. Although extremely
effective for volume control and better tolerated in the
hemodynamically unstable patient than hemodialysis,
clearance of small solutes may be inadequate in catabolic
patients or patients on total parenteral nutrition receiving
large protein loads. In addition, in the ICU setting there
is considerable risk of peritonitis. For these reasons, IPD
has been largely replaced by hemodialysis and continuous
hemodiafiltration for the management of acute renal
failure.
Bibliography

Bloembergen WE, Port FK, Mauger EA, et al.: A comparison

of mortality between patients treated with hemodialysis and
peritoneal dialysis. / Am Soc Nephrol 6:177-183, 1996.
CANADA-USA Peritoneal Dialysis Study Group. Adequacy of
Dialysis and Nutrition in Continuous Peritoneal Dialysis: Asso-

ciation with Clinical Outcomes. J Am Soc Nephrol 7:198-207,

1996.
Davies SJ, Bryan J, Phillips L, et al.: Longitudinal changes in
peritoneal kinetics: the effects of peritoneal dialysis and peritonitis Nephrol Dial Transplant 11:498-506, 1996.
Gokal R, Ash SR, Helfrich GB, et al.: Peritoneal catheters and
exit-site practices: Toward optimum peritoneal access Peril Dial
Int 13:29-39, 1993.
Golper TA, Brier ME, Bunke M, et al.: Risk factors for peritonitis
in long term peritoneal dialysis: The network 9 peritonitis and
catheter survival studies Am J Kidney Dis 28:428-436, 1996.
Holley JL, Bernardini F, Piraino B.: Infecting organisms in continuous ambulatory peritoneal dialysis patients on the Y-set. Am J
Kidney Dis 23:569-573, 1994.
Keane WF, Alexander SR, Bailie GR, et al.: Peritoneal dialysisrelated peritonitis treatment recommendations 1996 Update.
Peril Dial Int 16:557-573, 1996.
Khanna R and Nolph KD. The physiology of peritoneal dialysis
Am ] Nephrol 9:504-512, 1989.
Lupo A, Tarachini R, Cancarini G, et al.: Long-term outcome in
continuous ambulatory peritoneal dialysis: A 10-year survey by
the Italian Cooperative Peritoneal Dialysis Study Group Am J
Kidney Dis 24:826-837, 1994.
Piraino B, Bernardini J, Sorkin M: Catheter infections as a factor
in the transfer of continuous ambulatory peritoneal dialysis patients to hemodialysis Am J Kidney Dis 13:365-369, 1989.
Pollock CA, Ibels LS, Caterson RJ, Mahony JF, Waugh DA,
Cocksedge B: Continuous ambulatory peritoneal dialysis: Eight
years of experience at a single center. Medicine 68:293-308,1989.
Twardowski ZJ: Clinical value of standardized equ ilibration tests
in CAPD patients current concepts of CAPD. Blood Purification
7:95-108, 1989.
U.S. Renal Data System, USRDS 1996 Annual Report. VI. Causes
of Death. Am J Kidney Dis 28 (Suppl 2):S93- S102, 1996.
Young GA, Kopple JD, Lindholm B, et al.: Nutritional assessment
of continuous ambulatory peritoneal dialysis patients: An international study. Am J Kidney Dis 17:462-471, 1991.

OUTCOME OF END STAGE RENAL

DISEASE THERAPIES
WENDY E. BLOEMBERGEN

The availability of dialysis and transplantation or renal

replacement therapies has allowed the survival of patients
with end stage renal disease (ESRD), previously a fatal
illness. However, the expected outcomes of an ESRD patient remain inferior to the general population. This chapter
includes an overview of ESRD patient outcomes, assessed
in terms of mortality, but also morbidity, technique survival, quality of life, rehabilitation, and cost. The availability of the United States Renal Data System (USRDS), a
national database of all Medicare-treated ESRD patients
(93% of the U.S. ESRD population), and large cohort studies allows the description of many of these outcome measures.

data on reported cause of death is published by the USRDS.

Figure 2 shows the reported causes of death for the ESRD
population over age 20 from 1991 to 1993. Cardiac causes
combined account for 48% of all deaths. Infection, cerebrovascular disease, and malignancy account for 15%, 6%, and
4%, respectively.
The predominant cause of cardiac deaths is reportedly
cardiac arrest or sudden death, accounting for 39%, followed by acute myocardial infection (MI; 24%), arrhythmia
(14%), cardiomyopathy (10%), and atherosclerotic heart
disease (9%) (Fig. 3). A high prevalence of both coronary
artery disease and left ventricular hypertrophy, which is
already present among patients starting ESRD therapy,
contributes to these cardiac deaths.
Disruption of the skin barrier by the vascular access in
hemodialysis patients and the peritoneal catheter in peritoneal dialysis patients are factors which are partly responsible for the high risk of death due to infection. Comorbid
conditions such as peripheral vas cular disease also contribute to infectious deaths. In addition, there is laboratory
evidence that patients with ESRD have defects in cellular
immunity, neutrophil function, and complement activation.
Figure 3 also shows the reported breakdown of infectious
causes. Septicemia due to the vascular access and due to
peritonitis were each reported to account for 8% of deaths.
It is possible that other access infections present as blood
borne infections in other organs and may be misclassified.

MORTALITY

The mortality of the ESRD population is high relative

to the general population. In 1993, the crude mortality rate
of the US ESRD population was 17.6 deaths per 100 patient
years. For the dialysis population only the rate was 23.3
deaths per 100 patient years. The expected remaining lifetime of a patient with ESRD is one third to one fifth that
of the general population for all age and race categories.
Figure 1 shows that the diagnosis of ESRD carries with it
a prognosis between that for colon and lung cancer. For
example, a healthy 45- to 54-year-old person can expect to
live about 30 more years. The expected remaining lifetime
would be 10 years if this person developed colon cancer,
7 years with ESRD, or 3 years with lung cancer.
The failure of currently available "renal replacement"
therapies to provide ESRD patients with expected survival
similar to the general population is likely due to the high
prevalence of comorbid conditions among the ESRD population (many of which are already present as patients start
ESRD therapy), the inability of dialysis to fully replace all
of the functions of native kidneys, and adverse consequences or side effects of the "replacement therapy," dialysis, or transplantation.

Factors Associated with Mortality

Demographics

Several demographic factors have been shown to be associated with mortality in the ESRD population. Increasing
age, white race, and male gender are associated with higher
mortality. Cause of ESRD is also an important predictor
(Fig. 4). Patients with diabetes as the cause of ESRD have
approximately twice the mortality risk as patients with
ESRD due to glomerulonephritis. Patients with ESRD due
to hypertension have mortality rates between those for
patients with ESRD due to glomerulonephritis and diabetes.

Causes of Death

Comorbidity

Characterization of causes of death helps to better understand the high mortality of the ESRD population. National

The prevalence of various comorbid conditions that are

shown to be associated with worse survival is high among

Primer on Kidney Diseases, Sec ond Edition

42
2

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

423

65. Outcome of End Stage Renal Disease Therapies

35.0

Expected Remaining Lifetime (years)

I U.S. Resident D Colon Cancer EiESRD 0 Lung/Bronchus Cancer

30.0

25.0
20.0
15.0

45-54

55-64
Age Group

FIGURE I Expected remaining lifetime for age groups 45 to 54 and 55 to 64 for U.S. resident
population (1990), and selected subpopulations with chronic disease, including colon cancer
population (1983 -1989), ESRD population (1992), and lung/bronchus cancer population (19831989). From the U.S. Renal Data System 1995 Annual Data Report. National Institutes of
Health, National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD, 1995.

patients treated for ESRD (Table 1). For example, by chart

review, 45% of patients in a national random sample of
hemodialysis patients had coronary artery disease (CAD;
defined as prior history of CAD, MI, abnormal angiogram,
angioplasty, or coronary artery bypass graft), which is associated with a 44% higher risk of mortality compared to
those without CAD, adjusted for demographics and other
comorbid conditions. As the table indicates, patients with
a history of congestive heart failure, arrhythmia including

atrial fibrillation, peripheral vascular disease, cerebrovas cular disease, chronic obstructive lung disease and neoplasm have higher independent risks of death than patients
without these conditions. In a recent study, an ankle-arm
blood pressure index of less than 0.9, a sensitive and specific
measure of peripheral vascular disease, was shown to be
present among 35% of hemodialysis patients and was associated with more than a sevenfold risk of cardiovascular
mortality. In another study, echocardiograms performed

Malignancy 4%
Infection
15%
Unknown
7%
Other Known
20%

Cardiac
48%

Cerebrovascular 6%
FIGURE 2 Percent distribution of causes of death for ESRD patients over age
20 (1991 to 1993). Data from The 1996 USRDS Annual Data Report. National
Institutes of Health, National Institute of Diabetes and Digestive and Kidney
Diseases, Bethesda, MD, 1996.

Wendy E. Bloembergen

424

Cardiac Causes (48%)

Infectious Causes (15%)

Acute
Cardiac

Myocardial
Arrhythmia
14

Cardiac Arrest
39%

Atherosclerotic
HD
9%

Other-4%
AIDS- 5%
Pulmonary
Infection
15%

SepticemiaVascular Access-8%
Septicemiaritonitis-8%

iticemis PVD
18%
Infarction
24%

Cardiomyopathy
10%

Other Causes
43%

FIGURE 3 Percent distribution of specific cardiac causes of death among all cardiac causes and of
specific infectious causes of death for all infectious deaths for ESRD patients. Data from The 1996
USRDS Annual Data Report. National Institutes of Health, National Institute of Diabetes and
Digestive and Kidney Diseases, Bethesda, MD, 1996.

on a cohort of 433 patients starting dialysis found the pres ence of left ventricular hypertrophy in 74%, which was
associated with a higher risk of mortality.
Among patients starting dialysis, low serum albumin has
consistently been shown to be a strong predictor of mortality. Although it is too simplistic to equate albumin with
nutritional status since many other factors affect its concentration, other measures of nutrition have also been predictive of mortality among ESRD patients. In a national random sample of 3400 patients starting hemodialysis, 14%

were undernourished (as indicated in the medical records).

Those who were undernourished had a 26% higher risk
of mortality. Worsened nutritional status as measured by
subjective global assessment and percentage lean body
mass was associated with an increased relative risk of death
among a large cohort of patients starting continuous ambulatory peritoneal dialysis (CAPD) who were followed prospectively. Low serum creatinine, also thought to be a measure of reduced muscle mass possibly due to malnutrition,
has been shown to be predictive of elevated mortality ris k.

Death Rate per 100 Patient Years

1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993

Year of ESRD Incidence

FIGURE 4 Death rates by diagnosis and year of incidence, adjusted for age, race, and sex. From
the U.S. Renal Data System 1996 Annual Data Report. National Institutes of Health, National
Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD, 1996.

425

65. Outcome of End Stage Renal Disease Therapies

TABLE I
Prevalence and Relative Mortality Risk of Comorbid
Conditions among US Hemodialysis Patients (1991)

Comorbid conditions

Coronary artery disease

Congestive heart failure

Arrhythmia
Peripheral vascular disease
Cerebrovascular disease

Chronic obstructive lung

disease
Neoplasm

Prevalence
(%)Relative mortality risk
45
1.44
42
1.62
31
1.51
22
1.62
16
1.31
12
1.62
9

1.32

These data suggest that greater attention to nutritional

therapy may be beneficial to patients with ESRD, although
it may be the case that a patient's nutritional status is simply
a measure of other comorbidity and is not necessarily modifiable by nutritional interventions.

example, one study comparing randomly selected patients

new to ESRD (3376 HD and 681 PD) found higher mortality for those treated with PD compared to HD among
diabetics, but not among nondiabetics after adjustment for
comorbidity differences. Another study using national data
compared mortality among prevalent patients receiving
HD or PD (170,000 years of follow-up) and found a 19%
higher mortality risk among PD-treated patients. This risk
was increasingly large and significant for ages > 55 years
and was accentuated in diabetics and females although it
was also present in nondiabetics and males. An evaluation
of causes of death found that the risk of death was increased
among PD patients for all major cause of death categories
except malignancy. It is not clear if these mortality differences are due to differences in dose of dialysis, compliance,
patient selection, medical quality of care, or a true adverse
treatment effect. The pilot study of a large multinational
randomized clinical trial comparing outcomes of PD and
HD is currently being performed. If the full-scale study is
deemed feasible and proceeds it should provide definitive
answers to questions regarding comparative outcomes of
PD and HD.

Socioeconomic Status

Socioeconomic factors have also been shown to predict

mortality. Previous studies have found higher mortality to
be associated with (1) lower income, (2) lack of family
support and (3) patients who live alone.
Modality

A patient who develops end stage renal failure must

choose between several end stage renal replacement therapy options. Broadly classified these include two main types
of dialysis [hemodialysis (HD) and peritoneal dialysis
(PD)], and transplantation. Several factors should be considered when choosing a treatment modality including coexistent disease, the patient's social or living circumstances,
as well as preferences and beliefs regarding the treatment
options. Relative outcomes are also important in the decision process.
Survival rates for transplanted patients are better than
for dialysis patients. However, patient selection obviously
plays a role because patients too ill to be referred for
transplantation remain on dialysis. A recent study comparing survival of transplanted patients to those who were
receiving dialysis after being included on the transplant
waiting list found substantially better survival among transplanted patients. This study is the best to date to address
this question as the issue of selection bias has been reduced.
Recipients of cadaveric transplants have lower survival
than those of living related transplants. From 1992 to 1993,
1 year survival was 95.5% for living related transplants
and 91% for cadaveric transplants. Among transplanted
patients, there has been an overall trend to improved survival over the past decade (see Chapters 74 and 75).
Although most comparative studies of hemodialysis and
peritoneal dialysis published prior to the 1990s showed no
consistent difference in mortality, several recent studies
have found higher mortality associated with peritoneal dialysis among some specified patient groups in the US. For

Treatment Parameters

There are several treatment parameters which, in epidemiologic studies, have been shown to be associated with
mortality. In several studies, lower dose of dialysis as measured by either Kt/V or urea reduction ratio (URR) among
HD-treated patients and weekly Kt/V or creatinine clearance among PD-treated patients (see Chapters 63 and 64)
was associated with higher mortality. In hemodialysis, dialyzer reuse has been associated with higher mortality in
subgroups of dialysis patients. The use of unmodified cellulosic membranes has also been associated with higher mortality than modified cellulosic or synthetic membranes.
However, to date, no clinical trials have been done to sort
out if these treatment factors rather than other confounding
factors (e.g., dialysis facility quality) are responsible for
the worse outcomes. A large multicenter NIH-sponsored
clinical trial comparing high vs. conventional dose dialysis
is currently in progress.
In the area of transplantation, the use of cyclosporine
has been shown to be associated with improvements in
patient survival.
Standardized Mortality Ratio

The standardized mortality ratio (SMR) is a method

which has been used in the ESRD community to compare
mortality for a dialysis facility relative to the national norm.
It is a ratio of the observed number of deaths divided by
the number of deaths which would be expected given the
specific age, race, gender, and causes of ESRD at the individual facility.
MORBIDITY

There are relatively few US studies available which comprehensively quantitate and characterize the occurrence of

426

Wendy E. Bloembergen

morbid events among the ESRD population in a prospective fashion. Descriptions have more frequently been retrospective and have used number of hospital admissions,
length of admission, or total number of hospitalized days
per calendar year as measures of morbidity. To date, most
published reports are not cause-specific and therefore also
do not differentiate, for example, between a hospitalization
for access surgery as compared to a complication of ESRD.
Limitations imposed by Medicare eligibility rules, health
insurance, and geographic location must also be considered.
USRDS data indicates that on average, the morbidity
of the dialysis population is substantial. In 1993, dialysis
patients ages < 65 were admitted to hospital a mean of
1.3 times per year and were confined to hospital for 11.5
days/year. Forty-two percent of patients had no admissions.
However 8% had five admissions or more, showing the
great degree of variability. Eleven percent were admitted
for more than 30 days/year. Overall, however, there has
been a trend toward decreasing hospitalization rates.
One study of more than 500 hemodialysis patients that
did report cause-specific hospitalization found that more
than 25% of admissions were access related (vascular access
declotting, replacement, infection, or PD catheter placement). The second most frequent reason for admission was
cardiac/circulatory followed by gastrointestinal/metabolic
disorders. The remainder were due to a variety of less
frequent causes. In another study of Canadian ESRD patients the probability of requiring hospitalization for an MI
infarction or angina was approximately 10% per year. There
was a similar probability of developing pulmonary edema
requiring hospitalization or ultrafiltration.
Hospitalization rates increase by patient age, as expected. Females have consistently been shown to have
higher hospitalization rates than males. African-Americans
are hospitalized more than whites at younger ages, whereas
the reverse occurs for older age groups. Native Americans
have the highest rates in general, whereas Asians have
the lowest. Rates for diabetics are higher than those for
nondiabetics in all age groups. There are also substantial
differences in hospitalization by geographic region, with
the highest rates in the South Central, Middle Atlantic,
and Northeast regions. The factors responsible for these
observed differences are not clear.
As one would expect, the presence of comorbid conditions including angina, congestive heart failure, peripheral
vascular disease, low serum albumin, and decre ased activity
level have been associated with increased hospital utilization. Hospitalization rates have been shown to be higher
among PD treated compared to HD-treated patients in all
age groups except for patients > 65. However, hospitalization rates have been steadily falling for PD patients,
whereas those for HD patients have remained relatively
stable. Patients with arteriovenous fistulas also have fewer
hospitalizations than those with PTFE grafts.
TECHNIQUE SURVIVAL

The probability of failure of the chosen modality for

renal replacement therapy may be an important factor in

a patient's modality decision. Among dialytic techniques,

there is a substantially higher failure rate among patients
treated with PD compared to HD. Among patients treated
with HD, less than 5% will have switched to PD at 2 years
whereas among patients initially treated with PD, approximately 15% will have switched to HD. Failure on PD is
most frequently due to peritonitis, and exit site or tunnel
infections, but can also be due to other CAPD-related
complications, inadequate dialysis, or social reasons.
Among patients who receive a transplant, renal allograft
survival is dependent on donor source. In 1993,1-year graft
survival for living related transplants was 91%, and for
cadaveric transplants was 83% (Fig. 5). Five year graft
survival was 73% and 56% for living related and cadaveric
transplants, respectively. Graft survival has improved
steadily over the past decade. Factors which have been
identified as predictors of worse graft survival include donor age over 55 or less than 10 years, prolonged cold ischemic time, older recipient age, HLA mismatch, and female
donor. The use of cyclosporine or FK506 has enhanced
survival rates. Post-transplantation factors which predict
worse graft survival include absence of first day diuresis,
dialysis at 1 week, rejection at discharge, and elevated
discharge serum creatinine.
QUALITY OF LIFE

Quality of life, a more refined outcome measure than

morbidity and mortality, is emerging as an important factor
in the assessment of various treatment choices among patients with ESRD. Interest in factors affecting how best to
improve quality of life has increased tremendously in the
past several years, partly because the quality of life experienced by ESRD patients, as among patients with many
other chronic illnesses, such as arthritis and chronic obstructive lung disease, is often significantly compromised.
The renal failure itself, the underlying disease responsible
for the renal failure and the high degree of associated
comorbidity, are likely contributing factors. Patients on
dialysis often complain of fatigue, lack of energy, anorexia,
depression, decreased libido, musculoskeletal symptoms,
and pruritus. HD is usually required three times a week
for 3 to 4 hours, and can be accompanied by various physical
discomforts, nausea, dizziness, headaches, and lack of energy during or after dialysis. Dietary restrictions are strict
and involvement in work, social, and recreational activities
is frequently reduced. PD requires frequent dialysate exchanges and may be complicated by peritonitis or catheter
tunnel infections. Renal transplant patients may experience
side effects of immunosuppressive therapy including infection, acne, hirsutism, and weigh gain. Medical complications, financial pressures, marital discord, sexual dysfunction, emotional stress, and anxiety about loss or death are
common, regardless of modality.
The measurement of quality of lifethe self-assessed
value of an individual's health experiences is a complex,
multidimensional concept typically measured by a number
of domains including physical functioning, social functioning, psychological/emotional well-being, physical symp -

6 5. Outcome of End Stage Renal Disease Therapies

95

427

Graft Survival
(%)

LRD

91.4

90
85

83.2

80
75 H
70

65

86

87

88

89

90

91

92

1993

60
1984

85

Year of Transplantation
FIGURE 5 One year graft survival by donor type and year, adjusted for age, sex, race, first
transplants only. LRD, living related renal transplant; CAD, cadaveric renal transplant. From the
U.S. Renal Data System 1996 Annual Data Report. National Institutes of Health, National Institute
of Diabetes and Digestive and Kidney Diseases, Bethesda, MD, 1996.

toms, cognitive function, vocational functioning, and overall satisfaction with health. It has been measured by both
generic and ESRD specific quality of life instruments
among this population. Numerous studies have documented poor quality of life in patients with ESRD. For
example, from 30 to 50% of patients on chronic dialysis
are reported to have impaired physical performance capacity, and up to 70% have moderate to severe levels of stress.
Studies have suggested that among the ESRD population,
quality of life varies by demographic factors, comorbidity,
underlying disease, and mode of treatment. It is inferior
among patients with comorbid conditions and those with
diabetes as a cause of renal failure. Among the ESRD
modality choices, transplantation appears to offer substantially better quality of life than hemodialysis as shown by
two recent prospective studies which compared quality of
life of dialysis patients while on the transplant waiting list
and the same patients after transplantation. Data on which
dialysis modality type offers the best quality of life are less
conclusive. Various cross-sectional or small studies comparing quality of life by dialysis type have generally found
highest quality of life among patients treated with home
hemodialysis, followed by PD, and in -center dialysis. However due to study design, these observations may be partly
explained by patient selection. Large longitudinal studies
and the pilot study for a clinical trial aimed to compare
quality of life by dialysis modality are currently in progress.
Both longitudinal studies and a randomized clinical trial
have found superior quality of life among dialysis patients
treated with recombinant human erythropoietin.
REHABILITATION

When Medicare coverage for ESRD began in 1973, the

expectation was that many of those whose lives were pro-

longed would contribute to society through work and taxes.

However, since then the patient population has changed
dramatically with the acceptance of older and sicker patients, and now the majority of patients do not work. Consequently, rehabilitation no longer can be judged by employment alone, but must address the patient's level of function
before ESRD and their return to a comparable status.
Efforts to maintain employment should be made for
patients who work, and vocational counseling should be
provided to unemployed patients who are capable of working. The nephrologist plays a crucial role in the motivation
of patients toward these goals. Participation in an exercise
program has been advocated to enhance rehabilitation outcome for all patients. To the extent possible, transplantation should be encouraged among suitable patients. In a
recent prospective study, the proportion of ESRD patients
employed increased from 30% before transplantation (on
dialysis) to 45% 2 years post-transplantation. This is probably because a more flexible schedule allowed patients to
engage in employment, as well as because of improved
physical and mental abilities. As healthcare costs rise, the
vocational and functional rehabilitation of patients undergoing dialysis or transplantation is receiving more attention
by Congress and the public.
COST

In 1972, Congress elected to cover the cost of ESRD

therapy for eligible patients with ESRD through amendments to Medicare. Currently, approximately 93% of
ESRD patients have their treatment covered by Medicare.
The estimated total monetary cost of treating ESRD in the
US in 1994 was $11.13 billion, of which the portion paid
by Medicare was $8.31 billion. This includes the cost of

428

dialysis and hospitalizations (each approximately 40% of

total cost), as well as supplies, covered medications, physician fees, etc. (remaining 20% of total cost). Average Medicare spending per patient year was $36,900. On average, the
total cost per patient (Medicare, other insurance, patient
obligations, etc.) is approximately 20 to 25% higher. There
is substantial variation by modality. Annual Medicare costs
for HD averages $44,200, $39,400 for PD, and $17,600 for
transplanted patients. In a prospective study assessing the
cost utility of renal transplantation compared with dialysis
in Canada, the cost of dialysis care pretransplant and the
cost of the first year after transplant was similar. At 2 years,
transplantation was clearly associated with better quality
of life and was less costly than dialysis among all subgroups.
Annual costs for all ESRD patients rise steadily with
age, an observation primarily attributable to the decline in
transplantation with increasing age. Diabetic patients are
considerably more costly to treat than nondiabetics. Average medicare payments are slightly higher for females than
males in most age categories. Payments for AfricanAmericans are higher than for whites among patients less
than 65, again partly due to lower transplantation rates.
Acknowledgments
Some of the data reported here have been supplied by the USRDS.
The interpretation and report ing of these data is the responsibility
of the author and in no way should be seen as an official policy
or interpretation of the U.S. Government.
Bibliography
Bloembergen WE, Port FK, Mauger EA, Wolfe RA: A comparison
of mortality between patients treated with hemodialysis and
peritoneal dialysis. J Am Soc Nephrol 6:177-183, 1995.
Canada-USA (CANUSA) Peritoneal Dialysis Study Group: Ade quacy of dialysis and nutrition in continuous peritoneal dialysis:
Association with clinical outcomes. J Am Soc Nephrol 198207, 1996.
Canadian Erythropoietin Study Group. Association between recombinant human erythropoietin and quality of life and exercise
capacity of patients receiving haemodialysis. Br Med J 300:573578, 1990.
Churchill DN, Taylor DW, Cook RJ, et al.: Canadian hemodialysis
morbidity study. Am J Kidney Dis 19:214-234, 1992.
Evans RW, Manninen DL, Garrison LP Jr, et al.: The quality of
life of patients with end-stage renal disease. N Engl J Med
312:553-559, 1985.

Wendy E. Bloembergen
Excerpts from U.S. Renal Data System 1996 Annual Data Report,
Chapters on: Causes of Death; Patient Mortality and Survival;
Hospitalization; Renal Transplantation: Access and Outcomes;
The Economic Cost of ESRD and Medicare Spending for Alternative Modalities of Treatment. National Institutes of Health,
National Institite of Diabetes and Digestive and Kidney Diseases, Bethesda, MD, 1996. Am J Kidney Dis 28:S1 -S165,1996.
Fishbane S, Young S, Plaster E, Adam G, Maesaka JK: Ankle-arm
blood pressure index as a predictor of mortality in hemodialysis
patients. Am J Kidney Dis 27:668-672, 1996.
Foley RN, Parfrey PS, Harnett JD, et al.: Clinical and echocardio graphic disease in patients starting end-stage renal disease therapy. Kidney Int 47:186-192, 1995.
Habach G, Bloembergen W, Mauger EA, Wolfe RA, Port FK:
Hospitalization among U.S. dialysis patients: Hemodialysis versus peritoneal dialysis. J Am Soc Nephrol 5:1940-1948, 19 95.
Held PJ, Port FK, Wolfe RA, et al.: The dose of hemodialysis and
patient mortality. Kidney Int 50:550-556, 1996.
Held PJ, Wolfe RA, Gaylin DS, Port FK, Levin NW, Turenne
MN: Analysis of the association of dialyzer reuse practices and
patient outcomes. Am J Kidney Dis 23:692-708, 1994.
Hakim RM, Held PI, Stannard D, et al.: The effect of the dialysis
membrane on mortality of chronic hemodialysis patients (CHD)
in the U.S. Kidney Int 50:566-570, 1996.
Jones KR. Factors associated with hospitalization in a sample of
chronic hemodialysis patients. Health Serv Res 26:671-699,1991.
Laupacis A, Keown PA, Pus N, et al.: A study of the quality of
life and cost utility of renal transplantation. Kidney Int 50:235242, 1996.
Owen WF, Lew NL, Liu Y, Lowrie EG, Lazarus JM: The urea
reduction ratio and serum albumin concentration as predictors
of mortality in patients undergoing hemodialysis. N Engl J Med
329:1001-1006, 1993.
Port FK: Morbidity and mortality in dialysis patients. Kidney Int
46:1728-1737, 1994.
Port FK, Wolfe RA, Mauger EA, Berling EA, Jiang K: Comparison of survival probabilities for dialysis patients vs cadaveric
renal transplant recipients. JAMA 270:1339-1343, 1993.
Rocco MV, Soucie JM, Reboussin DM, McClellan WM: Risk
factors for hospital utilization in chronic dialysis patients. J Am
Soc Nephrol 7:889-896, 1996.
Russell JD, Beecroft ML, Ludwin D, Churchill DW: The quality
of life in renal transplantation a prospective study. Transplantation 54:656-660, 1992.
Terasaki PI, Yuge J, Cecka JM, Gjertson DW, Takemoto S, Cho
YW: Thirty-year tends in clinical transplantation. In: Terasaki
PI, Cecka JM (eds) Clinical Transplants 1993, UCLA Tissue
Typing Laboratory, Los Angeles, 1994.
Wolfe RA, Gaylin DS, Port FK, Held PJ, Wood CL. Using USRDS
generated mortality tables to compare local ESRD mortality
rates with national rates. Kidney Int 42:991-996, 1992.

PROGRESSION OF RENAL DISEASE

THOMAS H. HOSTETTER

Progression of renal insufficiency to end stage kidney

failure is a process which extends over months to years in
most cases. The term renal insufficiency refers to a reduction in glomerular filtration rate (GFR) but with sufficient
residual filtration that the patient does not require dialysis
or transplantation, the point at which kidney failure or end
stage renal disease (ESRD) is said to occur. The circumstances in which the progression is more rapid are considered in other sections of this Primer. While a wide variety
of renal diseases progress, the patterns of that progression
have substantial clinical, laboratory, and pathological similarities. Indeed, the uremic state in which they culminate
is generally considered as an almost generic entity, irrespective of the initial disease. Furthermore, over the last 15
years, the notion has arisen that common mechanisms underlie this progression at a pathophysiologic level.

The rate of decline in GFR varies between patients and,

to some extent, between diseases. Glomerular diseases,
diabetic glomerulopathy, and hypertension tend to progress faster than tubulointerstitial diseases. However, data
from one large study suggested that patients with polycystic
kidney disease may also progress as rapidly as those with
glomerular injury. The level of renal dysfunction at which
one can predict that a specific patient is likely to progress
to ESRD is uncertain. Clearly, the higher the level of serum
creatinine when a patient is first encountered, the more
likely the patient will progress to uremia. However, a small
fraction of patients with renal insufficiency may remain
relatively stable for protracted periods of time even without
specific therapies. A number of clinical risk factors in addition to disease type associate with the likelihood of progres sion or higher rates of progression. Patients with proteinuria are more likely to suffer inexorable declines in renal
function than those without proteinuria and the degree of
proteinuria is a predictor of that likelihood. Those manifesting the highest rates of protein excretion generally demonstrate the most rapid, and certain, declines. Arterial hypertension also appears to be a risk factor for progression
to ESRD; its therapy is discussed below. Finally, a number
of factors have persistent but more modest effects on the
propensity to progress. These include African-American
descent, male sex, hypercholesterolemia, and smoking.

THE COURSE OF PROGRESSION

While many functions of the kidney tend to fail in a

roughly proportionate manner such that acidosis, anemia,
disorders of bone and mineral metabolism, and more nonspecific uremic symptoms develop concurrently, the course
of progressive renal disease is charted principally by decline
in GFR. Clinically, the trajectory of progressive renal dis ease is typically followed by the serum creatinine. The
relation of serum creatinine level to creatinine clearance
and GFR is, of course, inverse. Indeed, the decline in inverse serum creatinine (I/serum creatinine) has been used
as a clinically expedient means of following the change in
renal function. Without major interventions to alter the
progression of renal disease or the supervention of additional renal damage, the I/serum creatinine tends to decline
in a relatively linear manner for individual patients. The
use of the Cockcroft-Gault formula for calculation of creatinine clearance from serum creatinine, age, weight, and
sex yields an approximate value for GFR which is satisfactory for most clinical indications. Rarely is it useful to
perform multiple measurements of creatinine clearance by
urinary collections or even more accurate measurements
of GFR by infusion of exogenous filtration markers such
as iothalamate or inulin. Rather, serum creatinine, interpreted as its inverse and in the light of body habitus, is
sufficiently accurate for following the progression of dis ease in almost all cases (see also Chapter 2).

Primer on Kidney Diseases, Second Edition

MECHANISMS OF PROGRESSION

Arterial hypertension is not only a primary cause of

chronic renal failure, but also is a risk factor for the progres sion of renal disease initiated by other mechanisms such
as diabetic nephropathy and chronic glomerulonephritis.
In these latter settings, the systemic blood pressure appears
to enhance the progressive decline in renal function. The
pathways whereby hypertension arises after renal injury
from some other primary insult likely involve a combination of sodium retention and persistent activation of various
pressor systems including the sympathetic nervous system,
the renin-angiotensin-aldosterone system, and local vascular mediators such as endothelin activation or nitric oxide
suppression. Whatever the genesis of this increased perfusion pressure, many studies show a direct association between hypertension and the rate of progression. Furthermore, lowering blood pressure with antihypertensives
mitigates the decline of filtration rates, as discussed below.

429

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

430
In addition to the deleterious effects of systemic hypertension, the compensations that nephrons undergo after
surviving the initial insults of a primary renal disease seem
to conspire with arterial hypertension to perpetuate injury.
Before discussing these compensations, it is important to
emphasize that most progressive renal diseases attack the
kidney initially in a heterogeneous manner. For example,
in the course of chronic glomerulonephritis, some glomeruli
may be reduced to obsolescence relatively early in the
disease course, whereas others are spared. The basis for
this heterogeneity of injury is uncertain. An experimental
model of these adapted surviving nephrons has been a
reduction in nephron mass by simple surgical removal or
infarctive ablation of renal tissue, often called the remnant
kidney model. Studies of the residual units in this model
demonstrate that they undergo a complex series of changes.
Broadly, these changes have been termed compensatory
hypertrophy and hyperfunction, and have been viewed as
adaptive, allowing these nephrons to carry a greater filtration rate and other metabolic functions. The increase in
filtration rate in the surviving nephrons compensates for
the more severely damaged or destroyed nephrons and is
driven largely by vasodilatation of the arterioles supplying
those nephrons with increases in their plasma flow and
glomerular capillary pressures. The combination of arterial
hypertension and compensatory afferent vasodilation of
residual nephrons sustains the elevated glomerular pressure. Capillary hypertension has also been demonstrated
in a number of other experimental disease models, some
even without systemic hypertension. Furthermore, a reduction of the capillary pressure by several maneuvers including dietary protein restriction and specific antihypertensive
agents such as angiotensin converting enzyme (ACE) inhibitors has been associated with amelioration of structural
and functional injury to surviving glomeruli. Thus the combination of systemic hypertension and this compensatory
vasodilation of surviving glomeruli increases the risk for
secondary injury along this presumed common pathway of
progressive renal diseas e.
The remnant kidney model involves removal of function
of more than one kidney. Simple unilateral nephrectomy
leads to some of the above compensatory changes but they
are not so great as seen with the larger degrees of ablation
of the remnant model. Also, simple nephrectomy seems to
cause little arterial hypertension and is not associated with
initial scarring in the remaining kidney tissue as occurs in
the remnant model. It is probably for these reasonslesser
compensatory structural and functional adaptations, absence of arterial hypertension and initial scarringthat
simple nephrectomy is associated with considerably less
injury in surviving nephrons. Indeed, long-term clinical
follow-ups of patients undergoing nephrectomy due to
trauma, cancer, or kidney donation, have found little damage accrued by the remaining kidney. Thus, this procedure
does not seem to predispose to progressive injury very
often, if ever, as long as arterial hypertension and other
renal insults are not superimposed.
The cellular mechanisms whereby heightened glomerular pressures such as those seen in the remnant model and

Thomas H. Hostetter
in diabetes lead to scarring and eventual occlusion of these
capillary units are less certain than the fact that it occurs.
In all likelihood, the microvascular hypertensive injury is
multifactorial and includes pressure- or tension-induced
increases in synthesis of mesangial matrix as well as mechanical disruption of the capillary wall, yielding permeability defects and proteinuria. Indeed, proteinuria itself
may contribute to the progression of renal disease and, as
noted above, is one of the strongest quantitative risk factors
for progression of kidney injury. A large number of potential mechanisms have been ascribed to the relationship
between proteinuria and progressive injury. The excessively filtered plasma proteins are, in part, reabsorbed by
proximal tubular cells and, in that process, various small
molecules and substances bound to the proteins are liberated. In particular, the removal of iron from transferrin
may provide a catalyst for the synthesis of locally damaging
reactive oxygen species. Small lipids bound to these filtered
proteins and freed during protein reabsorption may also
have inflammatory or chemotactic properties which would
accentuate tubulointerstitial disease. Finally, the inspissation of filtered proteins due to water abstraction in the
distal nephron may lead to cast formation and intrarenal
obstruction.
In addition to the vascular adaptations of vasodilation
and capillary hypertension, metabolic adaptations of the
nephron, particularly the tubular epithelium, have been
invoked as potentially deleterious factors in the progression
of renal injury. A necessary consequence of increased filtration rates in the residual nephrons is an increase in
tubular reabsorption. This increase in reabsorption of solutes necessitates increases in oxygen consumption and
should generate damaging reactive oxygen species as byproducts of that heightened oxidative process. Ammonia
production by residual nephrons is also augmented, even
though total kidney ammonia production excretion eventually falls, accounting for uremic acidosis. The high levels
of ammonia in the cortex can have inflammatory consequences by reacting with components of the complement
system and producing both direct tissue injury and attraction of leukocytes. Elevations in the calcium-phosphorus
product as well as serum oxalate levels may lead to deposition of these substances in the renal interstitium providing
a nidus for inflammation and tissue destruction. Finally,
disorders of circulating cholesterol and triglycerides are
common in chronic renal disease, especially with proteinuria, and considerable attention has focused on the possibility that hyperlipidemia may induce injury in the renal microvasculature.
Together with the functional adaptations of residual
nephrons, these units increase in size throughout their
length from glomerulus through collecting duct. This process of compensatory growth is stimulated by circulating
factors that are as yet unidentified. Roles for growth in the
functional alterations and ultimately in progressive injury
have been suggested. Specifically, increases in glomerular
size may allow increased filtration but may also confer a
mechanical disadvantage in that the increased capillary
radii will augment wall tension, an effect that would be

66. Progression of Renal Disease

exaggerated by the heightened capillary pressures described above. In addition, enlargement of the glomerular
capillary likely disrupts the permselectivity of the glomerulus. The podocytes covering the capillary and providing
the fine definition of permeability are limited in their ability
to proliferate and hence may become in a sense "overextended" on an enlarged capillary surface. Likewise, expansion of mesangial areas may adversely influence the normal
clearance mechanisms of macromolecules that circulate
through this region. Whether the enlargement of tubular
epithelial cells also has untoward consequences, has yet to
be examined.
In addition to the overall compensatory renal growth
stimulated by as yet unidentified circulating factors, the
production of a large number of locally acting specific
growth, proliferative, and fibrogenic peptides play an important role in the sclerosis, scarring, and remodeling that
occur as renal disease progresses. Although a very large
number of these factors have been identified, the chief
among those demonstrated to contribute to the progression
of experimental and clinical renal disease has been transforming growth factor-^, platelet-derived growth factor,
and endothelin. These factors emanate from both invading
cells (such as monocytes) and native renal cells, including
mesangial and epithelial cells, and interact in complex networks linking them with each other and yet other substances including components of the renin cascade. These
latter factors, specifically angiotensin II and aldosterone,
must have pivotal roles because drugs blocking their production and actions have striking effects in many models
of chronic renal injury and act through pathways in addition
to the classical hemodynamic or pressor ones. For example,
the growth and fibrogenic actions of angiotensin II are
mediated, in part, through transforming growth factor-IB;
however, links to endothelin, platelet-derived growth factor, reactive oxygen metabolites, and other mediators also
exist. The nexus of these local factors has been a complex
but evolving area of research in progressive renal injury.
TREATMENT OF PROGRESSION
Blood Pressure

Control of systemic hypertension is the most important

therapeutic maneuver in slowing the progression of chronic
progressive renal disease. Of course, no exact definition of
acceptable control is available. However, more intensive
control is justified than that applied in primary or essential
hypertension. A mean arterial pressure of 90 to 95 mm Hg
is a currently suggested guideline and derives from the
Modification of Diet in Renal Disease (MDRD) study. The
mean arterial pressure is calculated as the diastolic pressure
plus one third the difference between systolic and diastolic
pressure. Thus, an arterial pressure of 120/80 mm Hg corresponds to a mean arterial pressure of 93 mm Hg.
Considerable attention has focused on which drugs best
controll hypertension in a patient with established renal
disease. Experimental studies principally in models of diabetes and the subtotally ablated rat kidney indicate that

431

drugs interrupting the renin-angiotensin-aldosterone system are most efficacious. Specifically, ACE inhibitors and,
more recently, angiotensin II type I receptor blockers have
lessened injury more than other agents. Clinical trials comparing different antihypertensive regimens have generally
confirmed that ACE inhibitors retard progression more
effectively than comparable drugs. At present, the data for
angiotensin receptor antagonists are not as extensive as for
the ACE inhibitors. Furthermore, the bulk of the clinical
evidence derives from studies of patients with diabetic nephropathy. Nevertheless, studies in other types of renal
disease generally show the greater efficacy of ACE inhibitors for slowing the progression of renal disease. It should
be noted that there are a rather limited number of studies
comparing specific agents. However, some studies have
suggested that calcium channel blockers may also retard
progression. Diuretics are a particularly reasonable additional drug as they enhance the effects of the aforementioned classes and are often especially useful in patients
with renal insufficiency whose hypertension has an important "volume" component. Beyond these choices, there are
no strong rational bases on which to choose among the
other classes of agents for maintaining blood pressure
within the acceptable range.
The administration of ACE inhibitors and angiotensin
receptor blockers in patients with renal insufficiency does
have some hazards. Specifically, reductions in renal function may occur rather promptly, particularly in patients who
have renovascular disease or intrarenal arterial damage.
Severe and clinically significant worsening of renal function
occurs in these settings. The fall in arterial perfusion pres sure and glomerular pressure reduce filtration may actually
exacerbate the functional effects of the vascular stenoses.
Less severe diminutions in filtration rate may occur with
any antihypertensive therapy in patients with renal insufficiency but without major renal vascular stenoses. Such declines in GFR are the result of imperfect autoregulation
(the capacity to maintain constant blood flow and filtration
over a range of arterial perfusion pressure) when renal
disease afflicts the kidney. Although exact guidelines do
not exist, a rise in serum creatinine of more than 1 mg/
dL after starting ACE inhibitor therapy should in most
instances lead to its discontinuation and consideration of
underlying renal vascular disease. Lesser elevations in serum creatinine, if stable, can be accepted and followed
without stopping therapy in most cases. Also, patients may
develop hyperkalemia with these drugs due to suppression
of aldosterone. Therefore, measurements of both serum
creatinine and potassium should be obtained shortly after
beginning these drugs in patients with established renal
insufficiency. In most instances, these compounds are tolerated and have particularly beneficial effects.
Dietary Management

Dietary protein restriction has been recommended as a

means of retarding the progression of renal disease. Several
clinical trials have attempted to assess its role, but the
results have often been confusing. However, at least one

432

Thomas H. Hostetter

meta-analysis of multiple trials has suggested that dietary

protein restriction does lessen progressive renal damage.
It is difficult to specify the exact level of dietary protein
that is both consistent with benefit and also consistent with
stable nitrogen balance. In part, this difficulty resides in
the fact that patients generally eat a greater level of protein
than that they are prescribed, even in careful long-term
clinical trials. In general, patients, including those with significant proteinuria, can maintain a neutral nitrogen balance with 0.6 of high biological value protein per kg body
weight per day. Some data suggest that even more liberal
protein intake (up to 0.75 g kg"1 d-1) still provides
benefit, but in all likelihood, patients would tend to eat yet
more if prescribed higher protein diets and at some point
the efficacy of protein restriction would be lost. This level
of protein restriction (0.6 to 0.75 g kg"1 -d ~*) requires
counseling of the patient by a qualified dietitian. Furthermore, for many patients this restriction represents a considerable change in their style of eating. If patients are begun
on a protein-restricted diet, monitoring of their protein
nutrition is required using follow-up of serum albumin.
Although most patients will eat a greater amount of protein
than prescribed, some will self restrict excessively or may
have other catabolic events that lead to greater protein
requirements. Notably, patients who are receiving glucocorticoid therapy will require higher levels of protein and,
indeed, dietary protein restriction in such patients may not
be worth the risk of protein malnutrition given the adverse
effects of glucocorticoids on protein catabolism. Although
the risks of protein malnutrition in general seem to be
rather modest, it should be carefully avoided because protein depletion is a great liability for patients who enter the
end stage therapies of dialysis or transplantation. While
protein restriction may be a less powerful tool for retarding
progression of renal disease than careful antihypertensive
therapy as outlined and may be more cumbersome, it still
should be offered to patients as yet another means of altering the course of their renal disease.
Other Therapies

Therapies designed to affect metabolic function such as

acid production, calcium phosphorous metabolism, lipid
levels, and reactive oxygen moieties in order to retard the
progression of renal disease remain to be tested in any
significant clinical trials. However, these therapies as well

as those directed at specific intrarenal growth or fibropro liferative factors may be employed in the future.
Bibliography
Bakris GL, Copley JB, Vicknair N, Sadler R, Leurgans S: Calcium
channel blockers versus other antihypertensive therapies on progression of NIDDM associated nephropathy. Kidney Int
50:1641-1650, 1996.
Benigni A, Remuzzi G: Glomerular protein trafficking and progression of renal disease to terminal uremia. Semin Nephrol
16(3):151-159, 1996.
Brenner BM, Lawler EV, Mackenzie HS: The hyperfiltration theory: a paradigm shift in nephtology. Kidney Int 49(6):17741774, 1996.
Cockcroft DW, Gault HW: Prediction of creatinine clearance from
serum creatinine. Nephron 16:31-41, 1976.
Floege J, Johnson RJ: Multiple roles for platelet -derived growth
factor in renal disease. Miner Electrolyte Metab 21(4-5):271282, 1995.
Kasiske BL, Kalil RS, Ma JZ, Liao M, Keane WF: Effect of
antihypertensive therapy on the kidney in patients with diabetes:
a meta-regression analysis. Ann Intern Med 118(2):129-138,
1993.
Ketteler M, Noble NA, Border W. Transforming growth factorbeta and angiotensin II: the missing link from glomerular hyperfiltration to glomerulosclerosis? Annu Rev Physiol 57:279295, 1995.
Klahr S, Levey AS, Beck GJ, et al.: The effects of dietary protein
restriction and blood-pressure control on the progression of
chronic renal disease. N Engl J Med 330:887-884, 1994.
Lewis EJ, Hunsicker LG, Bain RP, ct al.: The effect of angiotensinconverting enzyme inhibition on diabetic nephropathy. N Engl
J Med 329:1456-1462, 1993.
Levey AS, Adler S, Caggiula AW, et al.: Effects of dietary protein
restriction on the progression of advanced renal disease in the
Modification of Diet in Renal Disease Study. Am J Kidney Dis
27(5):652-663, 1996.
Maschio G, Alberti D, Janin G, et al.: The effect of angiotensin
converting-enzyme inhibitor benazepril on the progression of
chronic renal insufficiency. N Engl J Med 334:939-945, 1996.
Nath KA. Tubulointerstitial changes as a major determinant in the
progression of renal damage. Am J Kidney Dis 20(1):1-17,1992.
Pedrini MT, Levey AS, Lau J, Chalmers TC, Wang PH: The effect
of dietary protein restriction on the progression of diabetic and
nondiabetic renal diseases: a meta-analysis. Ann Intern Med
124(7):627-632, 1996.
Rosenberg ME, Smith LJ, Correa-Rotter C, Hostetter TH: The
paradox of the renin-angiotensin system in chronic renal disease.
Kidney Int 45:403-410, 1994.
Zucchelli P, Zuccala A, Borghi M, et al.: Long-term comparison
between captopril and nifedipine in the progression of renal
insufficiency. Kidney Int 42:452-458, 1992.

MANAGEMENT OF THE PATIENT WITH

CHRONIC RENAL DISEASE
PAUL L KIMMEL

STAGES OF CHRONIC RENAL DISEASE

neal dialysis or hemodialysis, or cadaver or living related

renal transplantation. The uremic syndrome is discussed
in detail in Chapter 62. The time when such therapy begins
in a patient whose renal impairment is expected to be
permanent and irreversible marks the onset of ESRD.
Clinical measurement of renal function uses creatinine,
a normal product of muscle metabolism, as an endogenous
marker of GFR. The serum creatinine concentration rises
as the GFR falls, but it is an imperfect filtration marker.
Creatinine is secreted as well as filtered by the kidneys,
and its production rate depends on diet and muscle mass.
The latter falls with advancing age or muscle wasting. In
addition to observing changes in serum creatinine directly,
the progression of renal disease may also be followed by
plotting the reciprocal of the creatinine concentration over
time. In many patients, a straight line can be fit to the serial
changes. A change in the slope of such a plot may reveal
a change in the course of progressive renal disease, such
as the superimposition of acute renal failure upon chronic
renal failure. Alternatively, such changes have been used to
monitor the patient's course on a newly prescribed protein
restricted diet.

The course of patients with chronic renal disease is generally progressive, comprises a continuum, and may be divided into four phases for convenience: loss of renal reserve, renal insufficiency, chronic renal failure, and uremia
and end stage renal disease (ESRD) (Fig. 1). Each phase
presents the clinician with different issues and management challenges.
Loss of renal reserve may not be clinically apparent.
Although glomerular filtration rate (GFR) is within the
normal range, the response to pregnancy, a dietary protein
load, or other stimuli which usually enhance GFR is diminished.
Renal insufficiency is characterized by biochemically apparent diminution in renal function. The BUN and serum
creatinine concentrations are increased and the creatinine
clearance is decreased. Patients usually have no symptoms
of renal disease. Homeostatic mechanisms adequately
maintain circulating calcium, phosphorus, and potassium
levels and balance. This phase of chronic renal disease
may, however, be associated with changes in the levels of
circulating hormones such as calcitriol, parathyroid hormone (PTH), and insulin.
Renal failure is associated with a marked decrease in
renal function, but symptoms, if any, are mild. Patients
may have abnormal calcium and phosphorus levels when
compensatory mechanisms are no longer adequate. Normochromic, normocytic anemia may be present, especially
when the GFR is less than 20mL miir 1 1.73 m~ 2.
As the patient progresses through these three stages, the
clinician must establish the renal diagnosis, treat the specific
disease, and initiate nonspecific therapy such as treatment
of hypertension and the use of an appropriate diet. Control
of serum phosphorus levels may be important to limit the
development or extent of renal osteodystrophy. Evaluation
and treatment of anemia related to chronic renal disease
should be undertaken. Perhaps most importantly, preparation of the patient for renal replacement therapy should
commence.
Uremia is characterized by the onset of symptoms of
renal disease, including nausea, vomiting, malaise, pruritus,
and mental lassitude and inability to concentrate. Treatment of uremic symptoms may involve initiation of perito-

Primer on Kidney Diseases, Second Edition

TREATMENT OF THE PATIENT WITH CHRONIC

RENAL DISEASE

Treatment involves at least five imperatives: (1) identification and elimination of factors which can reversibly decrease renal function, (2) institution of measures to slow the
progression of renal disease, (3) maintenance of nutritional
status while minimizing the patient's symptoms and accumulation of toxic waste products, (4) identification and
treatment of complications of chronic renal disease, and
(5) preparation of the patient for ESRD therapy.
Factors which may interfere with or reversibly decrease
renal function in patients with chronic renal disease are
outlined in Table 1. Patients with an unexpected increase
in BUN and creatinine concentrations who exhibit signs
of volume depletion may respond to volume repletion.
Gastrointestinal bleeding should be excluded as a cause of
volume depletion and increas ed renal insufficiency, particularly when the rise in BUN is proportionally greater than
the rise in serum creatinine. Infection can result in volume
433

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

434

Paul L Kimmel

B
Qg

24 12

I-O oz

6
3.6 2.4

U-

50 25 -

1.2 UJ

75 -

DC1 1
DC
CO
0.6

LLJ

" - UJ
Q.

100

GFR
120
110
100
90 80 70 60 50 40 30 20 10
1

0
2

iiT 1.73mTIME

MONTHS? YEARS?
FIGURE I The course of progressive renal disease. The course of a patient who began with normal renal function,
corresponding to a serum creatinine concentration of 0.6 mg/dL and GFR of 120 mL-mkr'-l.TS m"2 , is shown. Yaxes: The geometric serum creatinine ordinate at the extreme left (A), applies to the solid line of identity (open
circles) and the rectangles depicting the various stages of progressive renal disease. The serum creatinine ordinate
at the right side of the figure (C) should be used with the dashed curve (axe-heads) only. The central ordinate (B)
which describes the percentage of renal function remaining applies generally. Line of identity: This line shows a
decrease in function from 100 to 0% of normal, corresponding to the diminution in GFR from 120 to 0 mL-min~M.73
m" 2 . The matching creatinine scale (A) is geometric. Dashed curve: Since GFR = daily creatinine excretion/serum
creatinine concentration and since the creatinine excretion is equal to daily production, which is constant, this equation
is of the form y = klx, which defines a hyperbola. This relationship remains valid over the course of chronic renal
disease unless the creatinine production changes. The curve emphasizes that large changes in GFR may produce
small and clinically imperceptible changes in serum creatinine concentration when overall renal function is near
normal. Conversely, trivial changes in GFR produce large changes in serum creatinine concentration when the GFR
is low. The rate of progression throughout the delineated stages depends on the nature of the underlying renal disease,
host factors, treatment, and compliance with the medical regimen.

depletion if patients have anorexia or vomiting, have diarrheal illnesses, or lose water through insensible routes because of fever. Sepsis may result in the superimposition of
acute renal failure on established chronic renal disease.
Patients with congestive heart failure and renal disease
may have diminution of renal blood flow as cardiac function
worsens. Therapy directed at improving cardiac hemodynamics may in turn substantially improve renal hemodynamics. Optimizing volume status in patients with liver
disease and ascites or the nephrotic syndrome may similarly
improve renal hemodynamics. Diuretics, aldosterone antagonists and angiotensin-converting enzyme (ACE) inhibitors may play important roles in the management of such
patients by counteracting hormonal factors which favor
sodium retention and enhancing urinary sodium excretion.
The measurement of fractional excretion of sodium
(FENa), a useful guide to renal responses to volume disorders in patients with oliguria, is less helpful in patients with
chronic renal disease. This is the case since the fractional
excretion of many solutes typically increases as renal func-

tion diminishes, a result predicated both by clinical findings

and predictable from the mathematical definition of the
fractional excretion of a
renal clearance of x given solute, x.
FEL =
-,
or,
in
clinical settings,
[P]cr

[P]cr

Early in the course of chronic renal disease, solute balance

and urinary solute excretion in patients are maintained,
even as GFR falls. The combination of maintenance or
increase in the numerator of the FE equation with the
progressive decrease in its denominator dictates that the
FE of a solute tends to increase during the course of the
development of renal insufficiency and chronic renal
failure.
The use of nephrotoxic drugs should be avoided, if possible, in patients with chronic renal disease. Patients with

67. Management of the Patient with Chronic Renal Disease

TABLE I
Factors Which Can Interfere With Evaluation of or
Decrease Marginal Renal Function
Volume disorders
Volume depletion
Poor nutritional intake
Excessive salt restriction
Excessive diuretic administration
Vomiting or diarrhea
Gastrointestinal bleeding
Infection
Volume depletion
Septic vasodilatation
Volume overload
Congestive heart failure
Ascites
Nephrotic syndrome
Nephrotoxic agents Nephrotoxic drugs
Aminoglycoside antibiotics
Nonsteroidal antiinflammatory drugs
Anesthetic agents Chemotherapeutic
agents Radiographic contrast material
Infection Septic acute tubular necrosis; indirect
effects
Uncontrolled hypertension
Hypotension
ACE inhibitors
Pericardial disease
Renal arterial disease
Urinary tract obstruction

435

an important role in limiting the progression of some forms

of chronic renal disease. With advanced renal failure or
renal artery disease, however, use of ACE inhibitors can
result in deterioration of renal function. This can occur in
selected patients as a result of excessive afferent and efferent renal arterioral vasodilatation, and subsequent decrease in the net glomerular hydrostatic pressure favoring
filtration. A secondary decrease in circulating aldosterone
levels may also result in the development of hyperkalemia.
Uremic patients may develop pericardial disease (Table
2). Pericarditis, pericardial effusion, or cardiac tamponade
can interfere with renal blood flow and result in a precipitous drop in GFR in patients with already compromised
renal function. Such events are indications for the initiation
of dialytic therapy. Renal arterial disease may progress
simultaneously with long-standing chronic parenchymal renal disease, and should be suspected in patients with a
history of atherosclerotic vascular disease, smoking, or an
unexpectedly swift loss of GFR, especially after treatment
with ACE inhibitors. Consideration should be given to
renal angiography, in spite of potential complications, in
such clinical settings. Procedures which do not employ contrast, such as magnetic resonance imaging, may provide
useful information in such cases, but rigorous comparative
trials of their place in the management of patients with
renal vascular disease and renal insufficiency are not available. Urinary tract obstruction should be considered, especially if the progression of the patient's renal dysfunction
is unusually rapid, or if the history or physical examination
suggest obstructive uropathy.

Pregnancy
Hypercalcemia, hyperuricemia
Agents which interfere with tubular creatininc secretion"
Cimetidine
Trimethoprim
" These agents increase serum creatinine concentration
without changing glomerular filtration rate.

diabetes mellitus, multiple myeloma, and preexisting renal

insufficiency are at increased risk for the development of
renal impairment from aminoglycoside antibiotics. Nonsteroidal anti-inflammatory drugs (NSAIDs) may decrease
renal function by limiting renal blood flow in patients with
chronic renal disease, especially in the setting of volume
depletion (see Chapter 42). Administration of radiocontrast material can result in temporary, or rarely, permanent
loss of renal function. Cimetidine and trimethoprim interfere with creatinine secretion by the renal tubules and
decrease creatinine clearance, leading to a secondary increase in serum creatinine concentration. Although GFR
is not affected, the use of such substances in patients with
chronic renal disease may lead to confusion in interpreting
clinical results.
ACE inhibitors are effective therapy for treating hypertension in patients with chronic renal dysfunction; they play

FORESTALLING THE PROGRESSION OF RENAL DISEASE

TO ESRD

Hypertension is a common finding in patients with kidney disease, and may lead to progressive renal insufficiency
by causing renal ischemia or increased glomerular pressure.
It has been most often associated with volume overload
secondary to the kidneys' limitation in excreting a daily
ingested sodium load; but hormonal factors such as activation of the renin-angiotensin-aldosterone system may also
play a role. Patients with chronic renal disease and volume

TABLE 2

Therapeutic Measures Employed to Treat the Progression of

Chronic Renal Disease
Treatment of hypertension
Systemic
Treatment with ACE inhibitors
Treatment with alternative antihypertensives
Intraglomerular
Treatment with ACE inhibitors
Dietary protein restriction
Dietary protein restriction
Phosphorus restriction and treatment of hyperphosphatemia
Control of glucose metabolism in patients with diabetes mellitus
Treatment of hyperlipidemia?

Paul L Kimmel

436
overload can be appropriately treated with diuretics. Because of decreased renal blood flow, the doses may need
to be increased to ensure desired pharmacologic effects.
Treatment of hypertension, regardless of the specific antihypertensive regimen, is effective in forestalling the progression of renal disease (Table 2). Studies in patients with
diabetes mellitus have suggested that control of hypertension, achieved by disparate medications, results in diminution of progressive loss of renal function. ACE inhibitors
are effective in reducing proteinuria and the rate of loss
of GFR in patients both with and without diabetic nephropathy, although the mechanism underlying such effects is
not completely understood. Although some studies have
suggested that ACE inhibitors exert their protective effects
independent of effects on systemic blood pressure in patients with nondiabetic renal disease, this is currently controversial. Patients with chronic renal insufficiency taking
ACE inhibitors must be monitored for the development
of decreased GFR and hyperkalemia. Calcium channel
blockers have also been used both to treat hypertension
and in an attempt to slow the progression of renal disease.
Nondihydropyridine calcium antagonists (e.g., diltiazem)
may reduce proteinuria in patients with renal disease, while
dihydropyridine agents (e.g., nifedipine, nicardipine, felodipine, amlodipine, isradipine) may be less effective. A
meta-analysis suggested in diabetic patients with renal insufficiency that, compared with other antihypertensive
agents, dihydropyridine calcium antagonists did not have
the beneficial effects on renal function noted in nondiabetic
patients with renal insufficiency. The risks and benefits of
these calcium channel blockers must be evaluated in each
individual patient, in light of the paucity of long-term data
or well-designed large clinical trials evaluating their effects
in patients with chronic renal insufficiency.
When treated aggressively with potent antihypertensives
such as minoxidil and ACE inhibitors, some patients with
advanced renal insufficiency and malignant hypertension
regain enough function to forestall the need for dialysis.
Occasionally patients who start dialytic therapy regain
enough renal function after control of hypertension is established to discontinue dialysis.
Protein-restricted diets have been used over the last
three decades to limit symptoms of uremia. Changes in
dietary protein intake modulate glomerular hemodynamics
and GFR in patients with chronic renal disease, as well
as normal subjects. Some studies have demonstrated the
efficacy of protein-restricted diets in preserving GFR in
patients with chronic renal insufficiency. The separate roles
of reduced protein intake and effective antihypertensive
therapy in determining renal functional outcomes over time
is currently a matter of investigation. The stage in the
development of renal disease during which protein restriction should be prescribed, the degree of protein restriction
to be employed, and the type of diet administered are
currently under study. Extremely limited protein diets with
ketoacid supplements have also been successfully employed in patients with chronic renal disease. Different
protein restricted diets have been instituted in patients in

with beneficial results (see Chapter 66). It is important,

however, to ensure that the patient maintains overall good
nutritional status with adequate caloric intake while being
treated with a protein-restricted diet (see Chapter 68).
COMPLICATIONS OF CHRONIC RENAL DISEASE

The complications of chronic renal disease are discussed

in other chapters of the Primer. For convenience, they can
be divided into fluid, electrolyte, and acid-base disorders,
and disorders of metabolism and organ-system function
associated with renal dysfunction (Table 3).
Sodium Balance

Sodium balance is usually maintained until approxi mately 90% of GFR is lost. As GFR falls, FENa rises, a
consequence of diminished renal tubular sodium reabsorption. Patients with chronic renal disease cannot readily
respond physiologically to large and/or acute changes in
sodium intake with appropriate changes in urinary sodium
excretiona result primarily of tubular dysfunction. A patient with renal insufficiency habitually ingesting a high
sodium intake may continue to excrete relatively large
amounts of sodium when placed on a low sodium diet.
Inappropriate urinary sodium losses, sodium depletion, and
worsening BUN and creatinine may result. Similarly, patients may develop volume depletion and oliguria, with
increased levels of BUN and creatinine, because of over-

TABLE 3

Complications of Chronic Renal Disease

Fluid, electrolyte and acid- base complications
Volume overload Volume depletion Water
overload (Hyponatremia) Water depletion
(Hypernatremia) Hyperkalemia Hypokalemia
Hypocalcemia Hyperphosphatemia
Hypermagnesemia Metabolic acidosis Metabolic
alkalosis
Disorders of metabolism and organ-system
function Hypertension Renal osteodystrophy
Anemia
Disordered lipid metabolism
Coagulopathy Pericarditis
Gastrointestinal Disorders
Neuropathy and encephalopathy
Sleep disorders Sexual
dysfunction Psychological
disorders Immune disorders
Dermatologic complications

67. Management of the Patient with Chronic Renal Disease

zealous diuretic use or when vomiting, fever, diarrhea, or
gastrointestinal bleeding occur. Clues to such volume mediated superimposed renal insufficiency include clinical signs
of volume depletion and orthostasis. Conversely, such patients may develop pulmonary edema when challenged
with an intravenous sodium load, medications with high
sodium content, or a high salt diet. In such cases, diuretics
or sodium restriction may be effective. Treatment of congestive heart failure in patients with chronic renal disease
may result in a fall in GFR if renal perfusion decreases. As
cardiac and renal disease worsen, it becomes increasingly
difficult to balance the need for diuresis to treat congestive
heart failure with the requirement for adequate intravascular volume to maintain GFR.
Water Metabolism

The limited ability of diseased kidneys to excrete a concentrated or dilute urine may result in the development of
hypernatremia or hyponatremia. The urinary concentrating
mechanism is most affected by renal disease. Therefore,
the response to volume depletion or water deprivation is
affected more than the ability to excrete a water load.
Thirst is usually an adequate guide to maintain a patient's
plasma sodium concentration within the physiologic range.
If a patient with chronic renal disease exp eriences a sudden
change in mental status, and does not receive enough water,
hypernatremia may result. Alternatively, hyponatremia
may develop in patients hospitalized with chronic renal
disease who are given quantities of hypotonic fluids which
exceed the capacity of their kidneys to excrete free water.
Serum sodium concentration, and urine volume and osmolality are usually guides to proper therapy.
Potassium Metabolism

Potassium depletion may occur because of malnutrition,

nausea and vomiting, or diuretic use. Renal potassium excretion is generally conserved until late in the course of
chronic renal disease, but may be limited by oliguria or
marked diminution of GFR. Patients with renal diseases
associated with the syndrome of hyporeninemic hypoaldosteronism, such as diabetic nephropathy or interstitial nephritis, may develop hyperkalemia with relatively mild diminution in renal function, or without concurrent oliguria.
Volume depletion, sodium restriction, and oliguria can impair potassium excretion in such patients, as can the use of
ACE inhibitors, NSAIDs, or potassium-sparing diuretics.
Colonic potassium secretion plays a more important role
in total body potassium homeostasis as renal function deteriorates. Preventing constipation may be important to the
patient's overall potassium homeostasis. Dietary counseling regarding the potassium content of foods can be crucial
in patients with hyporeninemic hypoaldosteronism and
during late stages of chronic renal disease.
Acid-Base Metabolism

Metabolic acidosis develops because of the diseased kidneys' limitation in excreting hydrogen ions, secondary to

437
disordered ammoniagenesis, decreased filtration of phosphate and sulfate compounds, and decreased maximal renal
tubular hydrogen ion secretion. Acidemia may cause negative calcium balance and worsen bone disease. Factors predisposing to metabolic acidosis may be offset by forces
engendering metabolic alkalosis such as vomiting and the
use of base supplements or antacids. Bicarbonate supplementation is used to treat acidemia. Its use may be limited
by sodium overload and hypertension. Sodium or potassium citrate are also base supplements. They should be
used with caution because citrate increases the intestinal
absorption of aluminum. This is a particular problem if
aluminium hydroxide gels are prescribed to bind dietary
phosphorus. In addition, potassium excretory capability
may be limited. Calcium carbonate supplementation may
be a useful alternative. Typically, alkali supplementation is
indicated when the serum bicarbonate concentration drops
below 20 mEq/L.
Divalent Ion and Calcitriol Metabolism

As renal function diminishes, hypocalcemia results from

decreased renal 1,25 (OH)2D3 synthesis, hyperphosphatemia, and resistance to the peripheral action of PTH (see
Chapters 14 and 69). Hyperphosphatemia may be directly
involved in the pathogenesis of the progression of renal dis ease. Patients with a GFR less than 40 mL miir1 1.73 irr2
may have already developed abnormal bone histology,
indicative of renal osteodystrophy, because of diminished
renal calcitriol synthesis and secondary hyperparathyroidism. Phosphorus-restricted diets increase calcitriol levels
and may limit the development of secondary hyperparathyroidism in the early stages of renal disease. In later stages,
hyperphosphatemia may be more difficult to manage with
dietary restriction alone. The early use of low dose calcitriol
supplementation, when monitored closely, may normalize
serum calcium levels, and reduce biochemical features of
secondary hyperparathyroidism and the development of
renal osteodystrophy. Patients treated with calcitriol must
be carefully monitored for the development of hypercalcemia, hyperphosphatemia, and worsened renal function.
Newer vitamin D analogues currently under study may
be useful for the treatment of such disorders with less
associated hypercalcemia.
Anemia in Patients with Chronic Renal Failure

Anemia in patients with renal disease has been characterized as multifactorial; however, the most important mechanism involved in its pathogenesis is diminished renal synthesis of erythropoietin (see Chapter 72). Other possible
causes of anemia should be considered and evaluated as
appropriate. Treatment with synthetic erythropoietin is effective, resulting in a dose-dependent increase in hematocrit. This drug has improved exercise tolerance and quality
of life in patients. Many symptoms previously attributed
to uremia improve when patients with chronic renal disease
are treated with erythropoietin.

438

Paul L Kimmel

Uremic Coagulopathy
Uremic coagulopathy may present a problem for patients
with chronic renal disease undergoing surgical procedures.
Although the platelet count and prothrombin and partial
thromboplastin times are normal, the bleeding time is often
prolonged. The lengthened bleeding time may correlate
with a tendency to bleed during and after surgical procedures. The clinician must be aware of the possibility of
the disorder, and check the patient's bleeding time preoperatively. Desamino-D-arginine-8-vasopressin (dDAVP),
conjugated estrogens, and infusion of cryoprecipitate have
been used successfully to treat such patients (see Chapter
72).
Lipid Metabolism
Patients with chronic renal failure may have several different disorders of lipid metabolism. The role of lipids in
mediating the progression of renal disease is unclear, but
is currently the subject of much research. Patients with
the nephrotic syndrome often have hypercholesterolemia,
specifically associated with high levels of low-density lipoprotein cholesterol. Such disorders may be involved in the
pathogenesis of and lead to a higher incidence of cardiovas cular disease in patients with chronic renal disease, although few well-designed clinical studies address these is sues. Weight loss, dietary modification of cholesterol
intake, and substituting foods low in saturated fats and
cholesterol are probably the first steps in management.
Low protein diets may decrease proteinuria, and therefore
may improve cholesterol levels in patients with the nephrotic syndrome, although there have been few long-term
studies performed in patients. HMG-CoA reductase inhibitors are effective in reducing low density lipoprotein
(LDL)-cholesterol levels in patients with nephrotic syndrome. Treatment with probucol results in decreases in
both LDL and high density lipoprotein (HDL) levels in
patients with hypercholesterolemia. Patients with renal dis ease often have hypertriglyceridemia, associated with increased very-low-density and intermediate-density lipopro tein levels. Although proper diet and exercise may be
recommended safely to any patient with hyperlipidemia,
the optimum use of antihyperlidemic agents in these patients has not been determined.

ease. Depression as well as anxiety about the unknown,

the prospect of starting ESRD therapy, sexual dysfunction,
and loss of income, vocation, and social role may all be
important in the development of depression. There have
been few studies to delineate the diagnosis and treatment
of depression in such patients.
DRUG THERAPY IN PATIENTS WITH CHRONIC
RENAL FAILURE

Renal dysfunction makes the prescription of therapeutic

agents more complicated. Renal disease may impair drug
bioavailability and distribution because of changes in gas trointestinal function, the use of concurrent medications
such as antacids, and changes in plasma protein binding
and total body water levels. In tact hepatic drug metabolism
associated with changed renal drug elimination may lead
to the accumulation of pharmacologically active or inactive
drug metabolites. This may interfere with attempts to monitor drug levels or may change therapeutic effects. In general, the loading dose of an administered drug is not altered
by the presence of renal insufficiency. Antibiotic dosage,
with the exception of aminoglycosides and vancomycin,
may not need adjustment if more than 50% of the patient's
renal function is preserved. However, impaired renal clearance and metabolism may dictate the use of lower maintenance drug dosages. One strategy is to lower the individual
dose but maintain dosing intervals constant. This approach
avoids wide fluctuations in plasma drug levels. Another
strategy is to use regular maintenance doses but to extend
the intervals between doses to achieve comparable blood
or tissue drug levels. This approach may lead to higher
peak and trough drug plasma levels. Free blood or plasma
levels of drugs, especially when the therapeutic margin is
low, should be monitored when feasible in patients with
chronic renal disease. Drug dosing must be changed if renal
function changes concurrently, and must be carefully instituted in elderly patients who may have markedly reduced
levels of GFR despite normal or near-normal BUN and
creatinine levels. Patients with chronic renal insufficiency
should be cautioned to consult their physician before beginning any new medications or over the counter preparations.
Drug dosing in renal failure is discussed in detail in Chapter 43.

Sleep Disorders
Day/night sleep reversal is a classic uremic symptom.
Sleep apnea is an important problem in patients with
chronic renal failure. The syndrome may be associated with
cognitive defects, depressive affect, or hypertension. This
disorder can be accurately diagnosed by polysomnography.
Treatment with continuous positive airway pressure
breathing may result in resolution of symptoms in patients
with chronic renal disease.
Psychological Factors
Little is known about psychological adaptation in patients during early periods of the progression of renal dis -

SURGICAL CONSIDERATIONS IN PATIENTS WITH

CHRONIC RENAL FAILURE

Care must be exercised in the preoperative evaluation

of patients with chronic renal disease before they undergo
surgical procedures, especially if general anesthesia is used.
Attention must be paid to detecting and correcting any
preoperative abnormalities of acid-base balance, potas sium level, or coagulation. If a radiologic study employing
contrast is contemplated, limiting the contrast osmolarity
and dose, volume repletion with half-normal saline, and
optimizing the patient's volume status and urine flow before contrast administration may decrease morbidity.

67. Management of the Patien t with Chronic Renal Disease

439

PREPARATION FOR ESRD THERAPY

to employ early initiation of dialysis or transplantation.

The former plan may be fraught with difficulty as uremic
symptoms and complications multiply, and pharmacologic
and dietary treatment becomes difficult. In consultation
with the physician, however, individual patients may wish
to accept the risks of such an approach. Early initiation
of dialysis may avoid such complications as malnutrition,
neuropathy, and possible progression of cardiovascular dis ease. Early dialysis and transplantation before the advent
of uremic symptoms has been advocated in diabetic patients, as has early transplantation before initiation of
dialytic therapy. In general, patients will start ESRD therapy when creatinine clearance ranges from 0 to 10 mL
mirr1 1.73 m~ 2, and when uremic complications and symptoms supervene. Time of initiation and modality of ESRD
therapy should represent a well-formulated, jointly
achieved patient-physician decision.

As early as feasible, patients with progressive renal functional deterioration should be counseled regarding their
options for treating ESRD. The risks and benefits, as well
as the advantages and disadvantages of hemodialytic therapies, peritoneal dialytic methods, and renal transplantation
should be clearly outlined. It is advantageous to arrange
for a meeting with the patient and his or her family both
to provide support during the first three stages of chronic
renal disease and to inform the family so they can participate in planning for dialysis or transplantation. The team
approach, including nurses, dietitians, social workers, psychologists, psychiatrists, and transplant surgeons and personnel has been successfully employed in such preparatory
sessions. The social worker will often conduct a tour of the
ESRD facilities, coordinate reviews of home-based therapies and transplantation, and investigate the financial and
social resources the family brings to bear at the initiation
of renal replacement therapy. The option of not starting
ESRD therapy, or withdrawing it if therapy does not improve general well-being and quality of life should be dis cussed openly with the patient and family.
Access for dialysis should be considered and obtained
as early as possible and practical. An arteriovenous (AV)
fistula provides the longest lasting alternative and is least
prone to infection. An AV fistula may take several months
to mature, especially in elderly or diabetic patients, who
comprise a large proportion of the incident population
treated for ESRD. A peritoneal access catheter may be
placed at the time therapy is initiated, although many physicians think a 10-to 14-day period before using the catheter
is optimal. Psychological support may be advantageous,
since denial of disease can be a powerful negative factor
in the patient's care at this stage.
It may be difficult to determine when to start renal replacement therapy in an individual patient (Table 4). The
goals of the patient and physician should, however, be
clearly established during the stages of renal failure and
incipient uremia: Whether to wait for the onset of uremic
symptoms, using conservative management to maintain as
long a time period as possible before starting dialysis, or

TABLE 4

Indications for Initiating ESRD Therapy

Uremic symptoms
Hyperkalemia" Metabolic
acidosis" Cognestive heart
failure" Hypertension
Pericarditis Neuropathy
Encephalopathy Uremic
coagulopathy
" Resistant to medical therapy.

Bibliography
Attman PO, Samyuelsson O, Alaovic P: Lipoprotein metabolism
and renal failure. Am J Kidney Dis 26:573-592, 1993.
Bourgoignie JJ, Jacob AI, Sallman AL, Pennell JP: Water, electrolyte and acid-base abnormalities in chronic renal failure. Semin
Nephrol 1:91-111, 1981.
Delmez JA, Slatoplosky E: Hyperphosphatemia: Its consequences
and treatment in patients with chronic renal disease. Am J Kidney Dis 19:303-317, 1992.
Felsenfeld AJ, Llach F: Parathyroid function in chronic renal failure. Kidney Int 43:771-789, 1993.
Goodman WG, Coburn JW: The use of 1,25-dihydroxyvitamin D3
in early renal failure. Anna Rev Med 43:227-237, 1992.
Grundy SM: Management of hyperlipidemia of kidney disease.
Kidney Int 37:847-853, 1990.
Jacobson HR, Striker GE, for the workshop group. Report on a
workshop to develop management recommendations for the
prevention of progression in chronic renal failure. Am J Kidney
Dis 25:103-106, 1995.
Klahr S: Chronic renal failure: Management. Lancet 338:423427, 1991.
Levy AS, Adler S, Caggiula AW, et al.: Effects of dietary protein
restriction on the progression of advanced renal disease in the
Modification of Diet in Renal Disease Study. Am J Kidney Dis
27:652-663, 1996.
Livio M, Benigni A, Remuzzi G: Coagulation abnormalities in
uremia. Semin Nephrol 5:82-90, 1985.
Llach F: Secondary hyperparathyroidism in renal failure: The
trade-off hypothesis revisited. Am J Kidney Dis 25:663-679,
1995.
Ma JZ, Greene EL, Raij L: Cardiovascular risk factors in chronic
renal failure and hemodialysis populations. Am J Kidney Dis
27:652-663, 1996.
Maki DD, Ma JZ, Louis TA, Kasiske BL: Long-term effects of
antihypertensive agents on proteinuria and renal function. Arch
Intern Med 155:1073-1080, 1995.
Malluche H, Faugere M-C: Renal bone disease 1990: An unmet
challenge for the nephrologist. Kidney Int 38:193-211, 1990.
Maschio G, Alberti D, Janin G, et al.: Effect of the angiotensinconverting enzyme inhibitor benazepril on the progression of
chronic renal insufficiency. N Engl J Med 334:939-945, 1996.
Massey ZA, Ma JZ, Louis TA, Kasiske BL: Lipid-lowering therapy
in patients with renal disease. Kidney Int 48:188-198, 1995.
Pedrini MT, Levey AS, Lau J, Chalmers TC, Wang PH: The effect
of dietary protein restriction on the progression of diabetic and

440

Bradley J. Maroni

non-diabetic renal diseases: a metaanalysis. Ann Intern Med

124:627-632, 1996.
Peterson JC, Adler S, Burkart JM, et al.: Blood pressure control,
proteinuria and the progression of renal disease. The Modification of Diet in Renal Disease Study. Ann Intern Med 123:754762, 1995.
Ratcliffe PJ: Molecular biology of erythropoietin. Kidney Int
44:887-904, 1993.
Ritz E, Stefanski A, Rambausek M: The role of the parathyroid
glands in the uremic syndrome. Am J Kidney Dis 26:808-813,1995.
Rudnick MR, Goldfarb S, Wexler L, et al., for the lohexol Cooperative Study: Nephrotoxicity of ionic and non-ionic contrast media
in 1196 patients: A randomized trial. Kidney Int 47:254-261,
1995.

Salem MM, Rosa RM, Battle DC: Extrarenal potassium tolerance

in chronic renal failure: Implications for the treatment of acute
hyperkalemia. Am J Kidney Dis 18:421-440, 1991.
ter Wee PM, De Micheli AG, Epstein M: Effects of calcium antagonists on renal hemodynamics and progression of nondiabetic chronic renal disease. Arch Intern Med 154:11851202, 1994.
Tsukamoto Y, Moriya R, Nagaba Y, Morishita T, Izumida I,
Okubo M: Effect of administration of calcium carbonate to treat
secondary hyperparathyroidism in nondialyzed patients with
chronic renal failure. Am J Kidney Dis 25:879-886, 1995.
Wheeler DC, Bernard DB: Lipid abnormalities in the nephrotic
syndrome: Causes, consequences and treatment. Am J Kidney
Dis 23:331-346, 1994.

NUTRITION AND RENAL DISEASE

BRADLEY J. MARONI

INTRODUCTION

In contrast to carbohydrate and fat, excess protein is not

stored, but degraded to form urea and other nitrogenous
wastes eliminated by the kidney. Moreover, protein-rich
foods contain phosphates, hydrogen ions, and other inorganic ions which are also excreted by the kidney (Fig. 1).
Consequently, patients with chronic renal failure (CRF)
who consume excessive dietary protein will accumulate
nitrogenous wastes and inorganic ions, resulting in the clinical and metabolic disturbances characteristic of uremia.
This is the basis for the observation that limiting dietary
protein (and phosphorus) intake can ameliorate many uremic symptoms and some of the metabolic complications
of CRF (e.g., secondary hyperparathyroidism, metabolic
acidosis, etc.). Another less obvious relationship is also
illustrated in Fig. 1. When protein intake is inadequate,
waste products continue to accumulate due to the degradation of body protein stores.
Therefore, the goals of nutritional therapy in renal
failure are to (1) ensure sufficient dietary intake to
prevent malnutrition, (2) to limit the accumulation of
nitrogenous wastes and metabolic disturbances characteristic of uremia, and (3) to possibly slow the rate of
progression of CRF.

Primer on Kidney Diseases, Second Edition

PROTEIN TURNOVER IN NORMAL PATIENTS AND CRF

Healthy Subjects

In healthy adults consuming adequate (but not excessive)

energy and performing moderate physical activity, the average protein requirement is 0.6 g of protein per kg per
day. The "safe level of intake" (i.e., mean plus two standard
deviations) is 0.75 g of protein per kg per day, an amount
that is adequate for >97.5% of the population. In Western
societies, individuals typically ingest 1 to 2 g of protein per
kg per day; i.e., an intake well above the daily protein
requirement. Because excess dietary protein (and amino
acids) is not stored, the principal response to a surplus of
dietary protein is a marked increase in amino acid oxidation. The nitrogen resulting from amino acid catabolism is
then converted to urea and other nitrogenous wastes which
are eliminated primarily by the kidney. In contrast, the
principal adaptive response to a reduction in dietary protein
intake is a marked decrease in amino acid oxidation, resulting in more efficient utilization of dietary essential amino
acids (EAA). Assuming protein intake meets (or exceeds)
the minimum daily requirement, waste nitrogen excretion
(principally urea) decreases, and nitrogen balance (i.e., nitrogen intake equals output) is achieved with the lower
protein intake.
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

68. Nutrition and Renal Disease

441

DIETARY PROTEIN

1
V

,
+

Pi, H , K

, Na+,
so4 ,N

BODY
PROTEIN
STORES

. ESS. and NON - SS

.
E <
AMINO
ACIDS
1
1
UREA

RENAL EXCRETION

FIGURE I Dietary protein yields essential and nonessential

amino acids which can be used to synthesize body proteins. Protein containing foods in excess of the daily requirement are degraded
producing urea, nonurea nitrogen, and inorganic ions which must
be excreted. Reprinted with permission from Mitch WE: In:
Martinez-Maldonado M ed Handbook of Renal Therapeutics. Plenum, New York, pp. 349-373, 1983.

It should also be appreciated that protein turnover (i.e.,

protein synthesis and degradation) is a dynamic process
resulting in the remodeling of ~1 kg of protein per day.
During fasting, body protein stores (principally skeletal
muscle) are degraded and the amino acids are utilized for
hepatic gluconeogenesis. In response to feeding, anabolism
(i.e., protein synthesis > protein degradation) occurs by
suppressing whole-body protein degradation with or without an accompanying stimulation of protein synthesis. Assuming the diet is adequate, the net response over 24 hours
is neutral nitrogen balance and preservation of lean body
mass. In summary, successful adaptation to a low-protein
diet includes: (1) feeding-induced suppression of wholebody protein degradation with or without stimulation of
protein synthesis; and (2) marked suppression of EAA oxidation.
CRF Patients

Since avoiding malnutrition is a primary goal of nutritional therapy, it is important to recognize that low-protein
diets can be used safely in CRF patients. Fortunately, even
patients with advanced CRF (i.e., GFR ~10 to 15 mL/min)
can maintain neutral nitrogen balance and preserve lean
body mass when dietary protein intake is restricted. They
activate the same adaptive responses to low-protein diets
as healthy subjects; namely, a marked suppression of amino
acid oxidation and feeding-induced inhibition of wholebody protein degradation. In addition, nitrogen balance,
serum proteins, and anthropometries remain normal during
long-term protein restriction. In the Modification of Diet
in Renal Disease (MDRD) study, minor, albeit statistically
significant changes, did occur in some nutritional parameters in the low-protein diet groups, i.e., an increase in serum
albumin and a decrease in serum transferrin and some anthropometric measures. However, abnormal values stabilized following the first few months of diet therapy, and
the average values for the low-protein diet groups remained

within the normal range. Most importantly, a lower protein

intake was not associated with a higher rate of hospitalizations or death.
Low-protein diets (LPD) can also be used safely in patients with the nephrotic syndrome. When nephrotic patients were fed a diet containing 0.8 protein (plus 1 g protein/g proteinuria) and 35 kcal per kg body wt per day,
nitrogen balance was neutral; the principal compensatory
response was a suppression of amino acid oxidation. Amino
acid oxidation correlated inversely with urinary protein
losses suggesting that proteinuria is also a stimulus to conserve dietary EAA. In conclusion, both nephrotic and nonnephrotic CRF patients can activate normal adaptive responses to dietary protein restriction allowing them to
achieve neutral nitrogen balance and maintain lean body
mass.
PROGRESSION OF CRF AND SPONTANEOUS
PROTEIN INTAKE

Recently there has been concern that LPDs cause malnutrition and it has been suggested that these diets be used
cautiously in patients with CRF. These fears arise principally from two observations: (1) evidence of a spontaneous
decrease in protein intake and worsening of some indices
of nutritional status in patients with progressive CRF consuming unrestricted diets; and (2) the association between
hypoalbuminemia and mortality in hemodialysis patients.
In fact, several lines of evidence suggest that LPDs can be
used safely in CRF patients. First, the studies discussed
above indicate that proper implementation of a LPD yields
neutral nitrogen balance with maintenance of normal serum proteins and anthropometric indices during long-term
therapy. Second, recent evidence suggests that hypoalbuminemia in hemodialysis patients is more closely related to
inflammation than to dietary adequacy (vide infra). Without proper dietary education, CRF patients may inadvertently consume protein-rich meals resulting in the accumu lation of nitrogenous wastes and uremic symptoms for
several days. In this setting, satiety may actually represent
anorexia and it is clear that protein intake and nutritional
status can deteriorate if uremic symptoms persist, increas ing the risk of malnutrition. In contrast, if the diet is planned
so that sufficient protein is prescribed to maintain nitrogen
balance while limiting the accumulation of nitrogenous
wastes (Fig. 1), uremic symptoms will not develop and
nutritional status will be maintained.
LPDS AND PROGRESSION OF RENAL DISEASE

An alarming feature of CRF is that once damage is

established (serum creatinine ~2.5 to 3.0 mg/dL) the loss
of renal function usually continues even when the initial
disease process is no longer active. Evidence that an LPD
slows renal progression in animals and the ability to assess
changes in renal function in humans prompted investigators
to examine whether restricting dietary protein might delay
the loss of residual renal function in CRF patients.

442

Bradley J. Maroni
MONITORI NG DIETARY ADEQUACY AND COMPLIANCE

Several prospective randomized trials involving small

groups of CRF patients have concluded that an LPD can
slow the loss of renal function in diabetic and nondiabetic
renal disease. In contrast, the largest clinical trialthe
MDRD Studydid not demonstrate a benefit of dietary
protein restriction on renal progression, at least when an
'intention to treat" analysis was used (i.e., analysis of outcomes regardless of whether patients complied with the
diet). When the MDRD results were analyzed according
to the protein intake actually consumed, a 0.2 g kg'1
day" 1 reduction in protein intake was associated with a
29% slower rate of loss of GFR and a 41% prolongation
in the time to dialysis (P < 0.01). Two meta-analyses (one
including the MDRD study) have also concluded that an
LPD was associated with a 33 to 46% risk of renal failure.
In principle, an LPD could reduce the risk of renal failure
by slowing the progression of renal disease or by ameliorating uremic symptoms. Interestingly, the authors of one of
the meta-analyses estimated that a study would need to
enroll at least 1000 patients to detect a 33% reduction in
the risk of renal failure or death. Neither the MDRD study
has any other clinical trial included this many patients.
In conclusion, despite many supportive observations, it
has not been proved that dietary protein restriction will
slow the progression of renal failure in humans. One approach to the treatment of patients with progressive CRF
is initially to focus on treatment of hypertension, aiming
for a blood pressure <125/75 mm Hg in patients with pro teinuria exceeding 1 g/day, since more aggressive blood
pressure control is associated with a slower rate of progres sion in proteinuric patients. Angiotensin converting enzyme inhibitors should be considered the firstline of therapy, since the rate of progression of renal disease is 50%
slower with these agents in both diabetic and nondiabetic
patients. In motivated patients who are progressing despite
hypertensive treatment, dietary protein restriction is recommended.

When prescribing LPDs it is important to monitor dietary adequacy and compliance. Fortunately, a simple
method is available for estimating protein intake in the
outpatient setting (Table 1). To understand this technique
and its limitations, it must be appreciated that waste nitrogen derived from degraded protein is excreted as urea
and nonurea nitrogen (NUN). Whereas the urea nitrogen
appearance rate (i.e., urea excretion plus accumulation)
closely parallels protein intake, NUN excretion (i.e., nitrogen in feces and urinary ammonia, uric acid, creatinine,
etc.) is relatively constant, averaging 0.031 g N kg"1
day" 1. If the patient is in neutral nitrogen balance (BN),
then nitrogen intake (IN) equals urea nitrogen appearance
(UNA) plus nonurea nitrogen (NUN) (Table 1, Formula
2). When the BUN is unchanging (i.e., steadystate), then
urea nitrogen appearance (UNA) equals the 24-hour urinary urea nitrogen (UUN) excretion (Table 1 Formula 3).
Thus, IN equals UUN + 0.031 g N/kg body wt (Table 1,
Formula 4).
In the example shown in Table 1, because the prescribed
and estimated intake are similar, we can conclude that the
patient is compliant with the protein prescription. If the
estimated intake is less than prescribed, the patient should
be instructed to increase protein intake to reach the goal.
In contrast, if the estimated intake exceeds the prescription
by more than 25% (i.e., the precision of the method), then
the patient is noncompliant or a catabolic illness or condition (e.g., metabolic acidosis) is stimulating the breakdown
of body protein and increasing waste nitrogen production.
If a careful evaluation reveals no abnormality, the patient
should be referred to the dietitian for assistance in achieving compliance.
Dietary recall or diaries are the only available methods
for assessing energy intake in the outpatient setting. Al-

METHODS FOR ASSESSING NUTRITIONAL STATUS

Estimating Protein Intake from the 24- Hour Urinary Urea

Nitrogen Excretion

TABLE I

The criteria for an ideal technique for assessing nutritional status in outpatients would be (1) readily available
and easy to perform, (2) sensitive and reproducible, and
(3) validated in patients with renal failure. Although a
number of methods are available to assess the nutritional
status and/or body composition of patients with renal failure, their utility is compromised in patients with renal dis ease due to: alterations in metabolism caused by renal
disease independent of nutritional status (e.g., albumin,
prealbumin, urinary creatinine excretion), lack of
population-specific reference values (e.g., anthropometries), changes in extracellular (or peritoneal) fluid volume
(bioelectrical impedance); or, not widely available clinically
(e.g., neutron activatio n analysis, dual X-ray photon absorptiometry, NMR spectroscopy). Presently, serial changes
in anthropometries (e.g., edema-free body weight), serum
albumin, and transferrin are typically used to monitor nutritional status (vide infra).

5 N = / N - (UNA + NUN), where NUN = 0.031 g N/kg

body wt
if BN = 0, then 7 N = UNA +0.031 g N/kg body wt
when BUN unchanging, then UNA = UUN and,
/N = UUN +0.031 g N/kg body wt
Example: A 50-year-old man is seen in clinic 1 month following
instruction on a diet providing 0.6 g of protein per kg per day
(i.e., 70 kg x 0.6 protein/kg = 42 g protein).
Weight: 70 kg; UUN = 4.8 g/day; NUN = 0.031 g N X 70 kg
= 2.17 g N/day
if BN = 0, then 7 N = UUN + NUN
= 4.8 + 2.17 = 6.97 g N
= 6.97 g N x 6.25 g protein/g N
= 43.6 g protein/day
Note: N, nitrogen; BN, nitrogen balance (g N/day); 7N, nitrogen
intake (g N/day); BUN, blood urea nitrogen; UNA, urea nitrogen
appearance (g N/day); UUN 24-hour urinary urea nitrogen (g N/
day); NUN, nonurea nitrogen (g N/day).

68. Nutrition and Renal Disease

though their limitations are well recognized, as a compromise between accuracy and compliance, three-day food
diaries can be utilized to monitor energy intake. A skilled
dietitian can facilitate menu planning and patient satisfaction while assisting with the monitoring of patient's compliance and nutritional status. Nutritional status is monitored
by performing serial measurements of anthropometries,
serum albumin, and transferrin. Once patients are comfortable with the diet, they can be seen every 3 months while
(1) estimating protein intake from the 24-hour UUN excretion; (2) estimating caloric intake from food diaries/recall;
(3) monitoring anthropometries, serum albumin and transferrin; and (4) using these measures to provide patient
feedback.
The nutritional status of ESRD patients should also be
monitored, because many individuals consume less than
the recommended intake of protein and calories, and malnutrition is common in this population (vide infra). One
frequently used method to estimate dietary protein intake
is the protein catabolic rate (PCR). The PCR is analogous
to the urea nitrogen appearance rate (UNA) described
above and is calculated from equations relating the predialysis BUN, the normalized urea clearance (Kt), and volume
of distribution of urea (V). In maintenance hemodialysis
patients, urea kinetics are routinely measured to estimate
the delivered dose of dialysis (i.e., Kt/VUKa; see Chapter
63); hence, the PCR is also calculated. In continuous ambulatory peritoneal dialysis (CAPD) patients, expended dialysate is collected during a 24-hour period and urea nitrogen
and protein losses are measured. Dietary protein intake in
clinically stable CAPD patients can then be estimated as:
PCR (g protein/day) = 13 + 7.31 UNA (g N/day) + protein
losses (g/day). Since protein requirements are determined
primarily by lean body mass, PCR is typically normalized
to body weight (i.e., nPCR, g protein kg"1 day" 1). As
discussed previously, PCR will only reflect dietary protein
intake only when patients are in neutral nitrogen balance;
e.g., there are no superimposed catabolic illnesses. In addition to the nPCR, postdialysis weight ("dry weight") and
serum albumin are also routinely monitored. In ESRD
patients with low or declining values of body weight and
serum albumin, or an nPCR <1.0, protein and calorie intake can be estimated from dietary recall or diaries.
FACTORS CONTRIBUTING TO MALNUTRITION
AND INCREASED PROTEIN REQUIREMENTS IN
DIALYSIS PATIENTS

Virtually all studies of nutrition in dialysis patients have

found that these patients are frequently malnourished.
Although the prevalence depends on the indices used to
evaluate nutritional status, 10% of ESRD patients suffer
from severe malnutrition and another 33% from mild to
moderate malnutrition. Factors which increase protein
requirements and contribute to malnutrition include:
(1) inadequate intake due to anorexia induced by uremia,
comorbidity, medications, depression, etc.; (2) superimposed illnesses or catabolic states (e.g., acidosis); (3) increased requirements due to albumin and amino acid losses

443

into dialysate; and (4) catabolism induced by the hemodialysis procedure.

Several lines of evidence indicate that hemodialysis stimulates catabolism, at least when performed using bioincompatible membranes. For instance, cuprophane membranes
activate complement and stimulate the release of interleukin 1 and other cytokines from monocytes. Hemodialysis
with cuprophane membranes stimulates skeletal muscle
catabolism, whereas the use of polyacrilonitrile membranes
does not. Finally, with cuprophane membranes, net protein
balance became more negative during the dialysis procedure due to a suppression of whole-body protein synthesis
as well as dialysate amino acids losses.
NUTRITIONAL STATUS AND MORTALITY IN ESRD

A low serum albumin level is a powerful predictor of

mortality in ESRD patients. Increased mortality is also
associated with low values of BUN (suggesting a lowprotein intake), cholesterol, and creatinine (index of muscle protein mass). Although these findings suggest that
protein-calorie malnutrition increases mortality in dialysis
patients, this is presently unproved. Other factor(s) besides
nutritional status may result in abnormalities in these parameters. Albumin synthesis rates have been found to be
decreased in hypoalbuminemic hemodialysis patients even
with dietary protein intake similar to patients with normal
serum albumin levels. Serum albumin (and albumin synthesis rates) correlated inversely with the serum levels of the
acute-phase reactants, C-reactive protein (CRP) and a 2macroglobulin. A preliminary report has confirmed the
inverse relationship between serum albumin and CRP; in
contrast to serum albumin, CRP was an independent predictor of survival in hemodialysis patients. Thus, serum
albumin levels in hemodialysis patients are determined at
least partly by nonnutritional factors (e.g., inflammation),
and one should not necessarily equate hypoalbuminemia
with malnutrition in this population.
Because laboratory surrogates of nutritional status may
be influenced by nonnutritional factors, the relationship
between other indices of nutrition and survival would be
of great interest. Although it is well established that the
prevalence of malnutrition is increased in the ESRD population (e.g., low anthropometric indices, decreased total
body nitrogen), there is very little information relating nutritional status to patient outcomes (morbidity and mortality). Such studies have rarely been performed, because
they are labor intensive, expensive, and require long-term
follow-up of large numbers of patients with well-defined
risk factors. Complex statistical models using multivariate
analysis are required to evaluate which risk factors are
independent determinants of mortality (or morbidity). Interpretation is also complicated by the difficulty in determining whether the risk factor (e.g., malnutrition) is the
proximate cause of mortality or simply a complication of
the underlying disease (e.g., cancer). Finally, statistical associations must be interpreted cautiously since they do not
imply cause and effect.

444

Bradley J. Maroni

DIETARY REQUIREMENTS IN RENAL FAILURE

Protein (Table 2)
As discussed earlier, the benefit of dietary protein restriction in slowing progression of renal failure remains controversial. Patients with mild renal insufficiency (i.e., GFR
>60 to 70 mL/min) are usually asymptomatic, and there
is no information regarding the utility of dietary protein
restriction in this setting. Emphasis should focus on controlling hypertension and other coexisting problems such as
hyperlipidemia.
In patients with GFRs between 25 and 60 mL/min, controversy remains, but LPDs may retard progression of renal
failure and are safe when used properly. In motivated individuals, dietary therapy would begin with a conventional
LPD providing 0.6 g of protein per kg per day, with twothirds provided as high-biologic value protein, i.e., the protein
should have a high content of EAAs (usually animal proteins such as meat, fish, and eggs). In patients with advanced
CRF (GFR 5 to 25 mL/min), in addition to a diet providing
0.6 g of protein per kg per day, a very -low protein diet
containing ~0.3 g of protein per kg per day supplemented
with a mixture of EAA or their nitrogen-free ketoanalogues has been used. In contrast to Europe and Japan,
ketoacid supplements are not available in the US. Because
patients with a GFR <10 mL/min are at the greatest risk
for developing malnutrition, these individuals should be
closely monitored to ensure adequate intake and maintenance of nutritional status. Dialysis is typically initiated
when the GFR is <10 mL/min.
Because proteinuria is a risk factor for progressive renal
insufficiency and LPDs reduce urinary protein losses and
hypercholesterolemia in nephrosis, dietary protein restric-

tion could be used as adjunctive therapy in nephrotic patients. A diet providing 0.8 g of protein per kg per day (plus
1 g protein/g proteinuria) yields neutral nitrogen balance
in nephrotic patients. Moreover, there was no correlation
between nitrogen balance and GFR (range: 19 to 120 mL/
min). Finally, several long-term studies indicate that serum
albumin levels remained stable or increased when nephrotic patients were prescribed diets providing 0.45 to
0.80 g of protein per kg per day. Taken together these
studies suggest that dietary protein restriction is safe in the
nephrotic syndrome, even in patients with advanced renal
failure. However, the use of LPDs in nephrotic patients
with very high levels of proteinuria (i.e., >15 g/day), or in
patients with superimposed catabolic illnesses (e.g., systemic lupus) or receiving catabolic medications is not recommended.
The protein requirements of patients receiving hemodialysis or peritoneal dialysis are increased compared to
healthy subjects or predialysis CRF patients (vide supra).
It is recommended that maintenance hemodialysis patients
consume 1.1 to 1.2 g of protein per kg per day. Nitrogen
balance studies performed in CAPD patients indicate that
their requirements are at least 1.1 and preferably 1.2 to
1.4 g of protein per kg per day. It is commonly recommended that 50% of the protein be high-biologic value,
but with this level of protein intake it is unclear whether
this is necessary. If the patient's dietary intake is marginal
or the proposed diet severely limits food choices, it may
be preferabe to relax restrictions on protein quality.
Energy (Table 2)
In both predialysis and dialysis patients, energy expenditure (and hence energy requirements) at rest and during

TABLE I

Recommended Intakes of Protein, Energy, and Phosphorus for Patients with Chronic Renal Failure, Nephrotic Syndrome, and
End Stage Renal Disease

Protein (g kg * day x )
Chronic renal failure
GFR (mL/min)
>60 -70 25-60
5-25

Nephrotic syndrome
GFR <60 mL/min
End stage renal failure
Hemodialysis
Peritoneal dialysis

Energy" (kcal
kg-1 day"1)

Protein restriction not usually recommended

0.60 g kg' 1 day"1 including aO.35 g kg- 1
day"1 of HBV
(1) 0.60 g kg" 1 day- 1 including >0.35 g
kg' 1 day- 1 of HBV or
(2) 0.28 g kg-' day"1 supplemented with
EAA or KA

>35 >35

0.80 g kg- 1 day- 1 (plus 1 g protein/g

proteinuria)

>35 235

Phosphorus*
(mg kg- 1 day- 1)

No restriction

>35 >35
slO
>35

12

day 1
day- .
1
-1.2 g kg- 1 1.21.4 g kg- 1

" Energy intake may be cautiously decreased in obese individuals (>120% normal weight) or those who gain undesired adiposity with
the recommended intake.
b
Phosphate binders are often needed to maintain normal serum phosphorus.
Note: HBV, high biologic value protein; EAA, essential amino acid supplement; KA, amino acid-ketoacid supplement.

68. Nutrition and Renal Disease

exercise are similar to those in healthy subjects. An energy

intake of 35 kcal kg"1 day" 1 is recommended because
it maintains nitrogen balance, anthropometries, and serum
proteins in patients consuming the recommended intakes
of protein (Table 2). In the elderly and obese individuals,
or those gaining undesired weight, lower energy intakes
may be cautiously prescribed (30 kcal kg'1 day"1).
Unfortunately, ESRD patients frequently consume less
than the recommended intakes, e.g., daily average intakes
of 0.95 to 1.01 g of protein and 23 to 32 kcal/kg body
wt have been reported. Because malnutrition is common
among ESRD patients, this decrease in energy intake cannot be an adjustment to a lower energy requirement, but
must be considered maladaptive. To improve dietary intake, counseling sessions with the renal dietitian can be
useful. If attempts to improve intake using conventional
food sources are unsuccessful, a number of nutritional supplements are available which provide a high calorie or
protein density, while limiting the fluid and electrolyte content. Examples include adding glucose polymers (Polycose)
to beverages, or high-density oral supplements (Nepro;
Suplena) and candy bars (Regain).
HYPERLIPIDEMIA IN RENAL DISEASE

Hyperlipidemia is common in renal failure, with a prevalence of 20 to 70% in predialysis and dialysis patients. The
characteristic abnormality in nownephrotic predialysis as
well as hemodialysis patients is hypertriglyceridemia with
increased levels of very low-density lipoproteins (VLDL),
decreased high-density lipoprotein cholesterol (HDL-C)
and normal to low levels of low-density lipoprotein choles terol (LDL-C). Although impaired catabolism of triglyceride-rich lipoproteins is the major cause of hypertriglyceridemia, increased production also contributes since
hemodialysis patients with normal chylomicron clearance
rates also have higher fasting serum triglyceride levels. In
peritoneal dialysis patients, serum cholesterol frequently
rises and then stabilizes during the first year of therapy;
serum triglycerides are variable but hypertriglyceridemia
has been attributed to absorption of glucose from the dialysate, and possibly to lipoprotein and apolipoprotein losses
into the peritoneal fluid. In the nephrotic syndrome, plasma
LDL-C is usually increased and HDL-C is normal or decreased; hypertriglyceridemia may also be present. Initially
there is overproduction of VLDL, which are rapidly catabolized to form LDL. LDL clearance may also be impaired,
especially in patients with severe nephrosis. The most likely
explanation for hypertriglyceridemia in nephrosis is decreased clearance of triglyceride-rich VLDL particles.
In patients with renal disease, it is not known if hyperlipidemia increases the risk of atherosclerosis or if therapy
designed to reduce lipid levels is beneficial. Proponents of
therapy cite epidemiologic studies linking an elevated serum cholesterol with accelerated atherosclerosis and the
reduction in coronary artery disease found in primary and
secondary intervention trials of otherwise healthy subjects.
Despite the lack of conclusive clinical data demonstrating
a benefit of lipid-lowering therapy in patients with renal
failure, some general guidelines are available.

445

As recommended for the general population, the National Cholesterol Education Program (NCEP) guidelines
seem appropriate for hyperlipidemic renal patients. Specifically, adults with a total cholesterol >200 mg/dL on two
occasions should be instructed in a Step I American Heart
Association (AHA) diet providing <30% of total calories
from fat (with <10% of calories from saturated fat) and
<300 mg cholesterol daily. Attention should also be directed at eliminating other cardiovascular risk factors (e.g.,
smoking, obesity, and hypertension), avoiding excessive
carbohydrate and alcohol intake, and improving glycemic
control in diabetics. In patients with a total cholesterol
>240 mg/dL or 200 to 239 mg/dL with two risk factors
[i.e., documented coronary artery disease (CAD) or family
history of CAD before age 55, cigarette smoking, male
sex, hypertension, HDL-C <35 mg/dL, diabetes mellitus,
history of cerebrovascular or peripheral vascular disease, or
severe obesity] a lipoprotein profile should be performed. If
the LDL-C remains greater than 190 mg/dL (or 160 mg/
dL with two risk factors) following 6 months of dietary
modification, drug therapy should be considered. Despite
dietary modification, a number of patients (particularly
nephrotic and transplant patients) are likely to have cholesterol levels above the desirable range (i.e., LDL-C
<130 mg/dL). In view of their potency, once a day dosing,
and favorable side-effect profile, HMG-CoA reductase inhibitors should be considered first-line therapy. In patients
refractory to monotherapy, the combination of a HMGCoA reductase inhibitor with either nicotinic acid or a bile
acid sequestrant may be considered. Bile acid sequestrants
interfere with cyclosporine absorption and cannot be used
in transplant patients. The fibric acids, clofibrate and gemfibrozil, reduce serum triglycerides, but should be avoided
in patients with renal failure because of the increased risk
of myopathy.
MINERALS, TRACE ELEMENTS, AND VITAMINS IN
RENAL FAILURE
Minerals and Trace Elements

Sodium intake is usually restricted to 2 g/day to control

hypertension and edema in CRF patients (see Chapters 15
and 79). In dialysis patients, sodium restriction will also
reduce thirst, interdialytic weight gain, and the likelihood
of intradialytic hypotension due to increased ultrafiltration
requirements. Disorders of potassium metabolism are dis cussed in detail in Chapter 13. Hyperkalemia is uncommon
until the CrCl <15 mL/min (and usually <5 mL/min), and
it is usually not necessary to restrict potassium intake prior
to initiating dialysis. Generally, potassium is restricted to
2 g/day once dialysis is begun, although CAPD patients
may tolerate or even require a more liberal intake.
Iron metabolism and the use of erythropoietin are dis cussed in Chapter 72. The impact of renal disease on calcium, phosphorus, vitamin D, and aluminum metabolism
is reviewed in Chapters 14 and 69. However, several comments are warranted. First, because calcium absorption is
impaired in CRF and protein-restricted diets are low in
calcium, at least 1.0 to 1.5 g/day of elemental calcium should

446

Bradley j. Maroni

be provided to ensure neutral calcium balance; this is easily

achieved with calcium-containing phosphate binders. Second, in view of the efficacy of calcium carbonate and acetate
as phosphate binders, and the risk of augmenting aluminum
absorption with citrate, the use of calcium citrate as a phosphate binder should be avoided. Magnesium-containing
laxatives and antacids should be avoided in patients with
renal insufficiency because the kidney is primarily responsible for magnesium excretion, and thus hypermagnesemia
may result.
Although postmortem studies in uremic patients indicate
that the tissue distribution of trace elements is abnormal,
the clinical importance of these changes is unclear. Elevated serum nickel and chromium levels have been attributed to contaminated dialysate. Conversely, low plasma
and red blood cell bromide levels have been reported in
CAPD patients, perhaps due to dialysate losses. Plasma
zinc is usually decreased, whereas tissue levels are generally
increased. Zinc supplementation has been reported to increase B lymphocyte counts, granulocyte motility, and
taste, and sexual dysfunctionalthough these findings
have not been uniform. In conclusion, the dietary requirements for trace elements have not been well defined in
renal failure. It would seem prudent to monitor closely
dialysis water supplies and peritoneal dialysate for excessive concentrations of trace elements. Because the kidney
is the principal route of excretion, supplementation is
not recommended.
Vitamins

A water-soluble vitamin supplement is frequently prescribed for CRF patients based on studies indicating intakes
below the recommended daily requirements (RDA), decreased intestinal absorption, impaired enzyme activity and
the presence of circulating inhibitors (vitamins are cofactors), or losses into dialysate. Notably, diets which are restricted in protein, phosphorus, and potassium are often
low in water-soluble vitamins. In CAPD patients, the net
intake of folate, niacin, B I, Bg, and B12 were below the
RDA for healthy subjects. In contrast, one study found
that after discontinuing a vitamin supplement most hemodialysis patients were able to maintain normal blood and
erythrocyte water-soluble vitamin levels by diet alone; the
blood levels of a few vitamins initially decreased but then
stabilized within the normal range. Although further research is needed to quantify actual intakes and losses, renal
patients should be given a water-soluble vitamin supplement formulated to meet their estimated requirements
(e.g., Nephrocap, Nephrovite). However, "megavitamin"
therapy should be discouraged, because severe peripheral
neuropathy and hyperoxalemia have been reported with
high-dose pyridoxine and vitamin C supplementation, respectively.
Treatment of vitamin D deficiency is discussed in Chapter 69. Vitamin A supplements should be avoided because
plasma vitamin A (retinol) levels are increased in CRF
patients. Vitamin A toxicity has been reported in CRF
patients receiving parenteral nutrit ion containing a stan-

dard multivitamin supplement (providing 1500 /xg vitamin

A). Therefore, parenteral nutrition solutions should be
screened for vitamin A before use in patients with renal
failure. It has been suggested that oxidant stress may be
in part responsible for increased erythrocyte turnover in
uremia. However, Vitamin E supplementation is not recommended because plasma vitamin E levels are normal in
patients with renal failure and it has not been demonstrated
that vitamin E supplements increase erythrocyte survival
in uremic patients.
NUTRITIONAL THERAPY IN ACUTE RENAL FAILURE

The goals of nutritional therapy in acute renal failure

(ARF) are to: (1) avoid uremic symptoms and the need
for dialysis prior to renal recovery; (2) maintain or improve
functional and nutritional status, including host resistance
and wound healing; and (3) promote recovery of the kidneys. Patients with ARF are frequently catabolic and it
is likely that malnutrition is an important risk factor for
morbidity and mortality in these patients. Factors contributing to wasting in ARF include: (1) superimposed catabolic conditions such as surgery, trauma, acidosis, and sepsis, (2) alterations in metabolism resulting from the loss of
renal function; (3) preexisting malnutrition and inadequate
nutrient intake; and (4) intradialytic losses and catabolism
induced by the dialysis procedure.
Several caveats regarding nutritional therapy in ARF
should be recognized. First, intuitively it would seem that
the early institution of aggressive nutritional support in
ARF is warranted, yet its impact on renal recovery, morbid ity, and mortality remains unproved. Second, although nutritional intervention improves nitrogen balance in many
catabolic states, it is unusual for these patients to become
anabolic (i.e., positive nitrogen balance). Presumably, the
metabolic defects responsible for protein (and amino acid)
catabolism also impair utilization of dietary protein. Consequently, increasing protein intake above the daily requirement will not ameliorate wasting in patients with ARF;
the excess protein is simply catabolized, increasing waste
nitrogen production and the need for dialysis. Third, due
to the limitations of nutritional intervention trials in ARF,
the protein requirements of ARF patients are poorly defined. Consequently, protein requirements of ARF patients
have been derived from clinical experience using nutritional support in acutely ill individuals with or without
renal insufficiency and studies performed in CRF patients.
When ARF is self-limited (e.g., radiocontrast nephropathy)
and catabolism is modest, 0.6 g of protein per kg per day is
usually recommended in order to minimize waste nitrogen
accumulation and the need for dialysis while awaiting recovery of renal function. Patients requiring hemodialysis
or peritoneal dialysis should receive 1.0 to 1.2 and 1.2 to
1.4 g of protein per kg per day, respectively. The protein
requirements of patients receiving continuous renal replacement therapies (e.g., CAVHD, CVVHD, etc.) have
not been carefully examined, although dialysate amino acid
losses are increased compared to conventional hemodialysis. Intakes of 1.5 to 2.5 g of protein per kg per day have

68. Nutrition and Renal Disease

been recommended by some experts for patients receiving

continuous dialysis therapies.
In the absence of sepsis, trauma, or burns, ARF patients
are not hypermetabolic and their energy requirements are
similar to normal subjects; 35 to 40 kcal kg" 1 day" 1 is
commonly recommended. Alternatively, energy needs may
be based on estimates of basal energy expenditure (BEE)
derived from healthy populations using equations that take
into account body weight, height, age, and sex; this value
is then increased by 25% (i.e., 1.25 X BEE) to account
for biologic variability and physical activity. Adjustment
factors for patients with trauma, burns, and sepsis have
been proposed (e.g., sepsis; 1.2 to 1.6 X BEE), although
it should be appreciated that calculated and measured rates
of energy expenditure may vary widely. When available,
indirect calorimetry may help refine the energy prescription
by directly measuring energy expenditure.
In ARF, water and mineral intake (e.g., sodium, potassium, phosphorus) must be restricted. Typically, fluid intake is limited to ~1 to 1.5 L/day, sodium and potassium
is restricted to 2 g/day, and phosphorus is limited according
to the level of protein intake (Table 2). A number of enteral
formulas are available, including several products designed
specifically for patients with renal failure (vide supra).
These renal products are convenient and are also useful
for individuals who can eat, but who would benefit from
oral supplements. Tube-feeding mixtures taking into account the restrictions imposed by ARF can also be compounded by the pharmacy using other commercially available products. For individuals without a functioning gut,
nutrition is given parenterally. Parenteral nutrition can be
provided through a peripheral vein, but the solution must
be diluted to minimize the risk of phlebitis, thus limiting
its caloric content and increasing the volume administered.
Consequently, most patients who require intravenous feedings receive total parenteral nutrition (TPN) through a
central vein. The parenteral solution can be tailored according to the patient's dietary prescription and serum chemis tries and usually delivered in a volume of 1.0 to 1.5 L, thus
limiting the ultrafiltration requirement of patients receiving
dialysis. Although the plasma amino acid profile is abnormal in ARF, the optimal composition of amino acid solutions for ARF patients is unknown, and a benefit of EAA
solutions has not been proved. Therefore, conventional
TPN amino acid solutions are typically used in patients
with ARF. It should be emphasized that TPN is associated
with many potential complications including mineral and
electrolyte imbalances, acid-base disturbances, and
catheter-related complications including sepsis. Therefore,
whenever possible, the gut should be used for nutritional
support. Besides reducing cost and the complications associated with parenteral nutrition, enteral nutrition also promotes mucosal growth and bowel integrity.

447
BIBLIOGRAPHY
Ahmed KR, Kopple JD: Nutrition in maintenance hemodialysis
patients. In: Kopple JD, Massry SG, (eds) Nutritional Management of Renal Disease. William & Wilkins, Baltimore, MD, pp.
563-600, 1996.
Bergstrom J, Fiirst P, Alvestrand A, et al.: Protein and energy
intake, nitrogen balance and nitrogen losses in patients treated
with continuous ambulatory peritoneal dialysis. Kidney Int 44:
1048-1057, 1993.
Blumenkrantz MJ, Kopple JD, Gutman RA, et al.: Methods for
assessing nutritional status of patients with renal failure. Am J
Clin Nutr 33: 1567-1585, 1980.
Blumenkrantz MJ, Kopple JD, Moran JK, et al.: Metabolic balance
studies and dietary protein requirements in patients undergoing
continuous ambulatory peritoneal dialysis. Kidney Int 21: 849861, 1982.
Chertow GM, Lazarus JM: Malnutrition as a risk factor for morbidity and mortality in maintenance dialysis patients. In: Kopple
JD, Massry SG (eds) Nutritional Management of Renal Disease.
William & Wilkins, Baltimore, MD, pp. 257-276, 1996.
Coresh J, Walser M, Hill S:Survival on dialysis among chronic
renal failure patients treated with a supplemented low-protein
diet before dialysis. / Am Soc Nephrol 6: 1379-1385, 1996.
FAO/WHO/UNU: Energy and Protein Requirements. In Technical Report Series 724. World Health Organization, Geneva, pp.
206, 1985.
Ikizler TA, Greene JH, Wingard RL, et al.: Spontaneous dietary
protein intake during progression of chronic renal failure. J Am
Soc Nephrol 6:1386-1391, 1995.
Kaysen G A, Rathore V, Shearer GC, et al.: Mechanisms of hypoalbuminemia in hemodialysis patients. Kidney Int 48: 510-516,
1995.
Klahr S, Levey AS, Beck GJ, et al.: The effects of dietary protein
restriction and blood pressure control on the progression of
chronic renal disease. N Engt J Med 330: 877-884, 1994.
Kopple JD: Nutritional management of acute renal failure. In:
Kopple JD, Massry SG (eds) Nutritional Management of Renal
Disease. William & Wilkins, Baltimore, MD, pp. 713-753,1996.
Maroni BJ: Requirements for protein, calories, and fat in the
predialysis patient. In: Mitch WE, Klahr S (eds) Nutrition and
the Kidney. Little, Brown, Boston, pp. 185-212, 1993.
Maroni BJ, Staffeld C, Tom K, et al.: Mechanism permitting nephrotic patients to achieve nitrogen equilibrium with a proteinrestricted diet. J Clin Invest 99:2479-2487, 1997.
Maschio G, Alberti D, Janin G, et al.: Effect of the angiotensin converting-enzyme inhibitor benazepril on the progression of
chronic renal insufficiency. N Engl J Med 334:939-945, 1996.
Monteon FJ, Laidlaw SA, Shaib JK, et al.: Energy expenditure in
patients with chronic renal failure. Kidney Int 30:741-747,1986.
Owen WF Jr, Lew NL, Yan Liu SM, et al.: The urea reduction
ratio and serum albumin concentration as predictors of mortality
in patients undergoing hemodialysis. N Engl J Med 329: 1001 1006, 1993.
Tom K, Young V, Chapman T, et al.: Long-term adaptive responses to dietary protein restriction in chronic renal failure.
Am J Physiol 4: E668-E677, 1994.

RENAL OSTEODYSTROPHY AND OTHER

MUSCULOSKELETAL COMPLICATIONS OF
CHRONIC RENAL FAILURE
JAMES A. DELMEZ

and activity of osteoclasts, increased bone resorption, and

marrow fibrosis. In addition, osteoblastic activity is increased with an abnormally large amount of the bone surface involved in bone formation. This state of high bone
turnover is also characterized by an increased quantity of
unmineralized bone matrix (osteoid). It differs from normal
lamellar osteoid in that there is a haphazard arrangement
of collagen fibers giving the appearance of a woven straw
basket. Although woven osteoid can be mineralized, the
calcium is deposited as amorphous calcium-phosphate instead of hydroxyapatite. These features are shown in Fig. 1
and contrast markedly with normal bone histology (Fig. 2).

The term, renal osteodystrophy, is used in a generic sense

to include all skeletal disorders that occur in patients with
renal failure. These include osteitis fibrosa, osteomalacia,
mixed and adynamic bone lesions, and dialysis -related amy loidosis. Because the kidney plays a major role in mineral
homeostasis, by maintaining external balance for calcium,
phosphorus, magnesium and pH, the occurrence of metabolic bone disease in patients with renal failure is predictable. For example, the kidneys are responsible for the
excretion of phosphorus. When renal failure occurs, phosphorus accumulates causing a reciprocal fall in calcium
levels and, therefore, stimulation of parathyroid hormone
(PTH) secretion. Phosphorus retention also decreases the
renal la hydroxylation of 25(OH)vitamin D to give
1,25(OH)2 D (calcitriol), the most potent metabolite of vitamin D. Calcitriol stimulates intestinal calcium absorption
and mobilization of calcium from bone. It also directly
suppresses PTH release, independent of its calcemic effect.
Low levels of calcitriol may, therefore, cause malabsorption
of calcium, hypocalcemia and further stimulation of PTH
secretion. In addition, the kidney is the organ most responsible for the excretion of aluminum and /32 -microglobulin,
substances implicated in the induction of dialysis -related
osteomalacia and amyloidosis, respectively. The pathophysiology of renal osteodystrophy recently has been better
defined, making possible a rational approach to its prevention and treatment.
The types of bone diseases seen in patients with renal
failure are divided into those with accelerated rates of bone
turnover (i.e., osteitis fibrosa and mixed lesions) due to
persistently high levels of PTH and low bone turnover
states (i.e., osteomalacia and adynamic lesions) due to alu
minum toxicity or relatively low levels of PTH. Recent
studies show that high turnover bone disease is the predom
inant lesion in patients treated with hemodialysis (50 to
60% of total) whereas low turnover is present in 60 to 70%
of
patients
undergoing
peritoneal
dialysis.

Pathogenesis of Secondary Hyperparathyroidism

Although recent studies have shown that hyperparathyroidism may develop without hypocalcemia, it is well accepted that low levels of this cation stimulate PTH secretion. The factors that contribute to hypocalcemia and
secondary hyperparathyroidism include phosphate retention, impaired vitamin D metabolism, skeletal resistance
to the calcemic action of PTH, altered calcium-regulated
PTH secretion, and decreased rates of degradation of
PTH. The pathophysiological events are summarized in
Fig. 3.
Phosphate Retention and Calcitriol Deficiency

The original explanation for the development of secondary hyperparathyroidism in patients with renal failure was
the "trade-off hypothesis." As glomerular filtration rate
(GFR) fell, it was reasoned that there would be a rise in
serum phosphorus levels with a transient reciprocal fall in
calcium concentrations. The latter would stimulate PTH
secretion leading to a phosphaturia and a normalization of
calcium and phosphorus levels. The trade-off was a normalization of phosphate levels at the cost of sustained PTH
hypersecretion. Although hyperphosphatemia could not be
consistently demonstrated in early renal failure, the hypothesis was supported by studies in animals and humans
with mild renal failure showing that phosphate restriction

'
OSTEITIS FIBROSA AND SECONDARY
HYPERPARATHYROIDISM

High levels of PTH lead to the development of osteitis

fibrosa. The histological features include increased number
Primer on Kidney Diseases, Second Edition

448

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

69. Renal Osteodystrophy and Other Musculoskeletal Complications of Chronic Renal Failure

FIGURE I Histological features of severe osteitis fibrosa. Many

multinucleated osteoclasts (OC) are actively resorbing trabecular
bone (T), resulting in an irregular jagged surface. Osteoblasts (OB)
are seen forming new bone surfaces. The marrow is replaced by
fibrosis (F). (Courtesy of Steven L. Teitelbaum, M.D.; X40, modified Masson stain.)

in proportion to the decrease in GFR prevented or corrected the hyperparathyroid state.

Recent studies have shown that renal synthesis of calcitriol falls with worsening renal function, presumably because of declining renal mass. It is now appreciated that
the principal effect of phosphorus restriction in early renal
failure is to increase the renal production of calcitriol, resulting in near normal levels. However, as renal failure
becomes advanced (GFR less than 20% of normal), hyperphosphatemia develops, leading to hypocalcemia and wors ening hyperparathyroidism. Phosphate restriction no
longer stimulates calcitriol synthesis, and thus calcitriol levels remain low. Presumably, this is due to the limited
amount of residual renal mass available for synthesis. However, PTH levels may still decline with phosphorus restriction independent of changes in calcium or calcitriol levels.
This suggests that phosphorus may directly affect PTH
secretion. In vitro studies with parathyroid glands of normal
rats have shown that high phosphorus levels in the media

449

FIGURE 2 Normal bone histology. The trabecular bone (T) is

fairly smooth without a prominence of osteoclasts or osteoblasts.
No fibrosis or osteoid is seen. (Courtesy of Steven L. Teitelbaum,
M.D.; X40, modified Masson stain.)

stimulate PTH secretion in a process that requires protein synthesis.

Calcium Malabsorption
In patients with a moderate degree of renal failure, calcitriol levels are either normal or slightly low. Calcium absorption is usually normal. In advanced renal failure, calcitriol deficiency may lead to impaired gastrointestinal
absorption of calcium and negative calcium balance. Hypocalcemia is a potent stimulus for PTH secretion that, in
turn, raises calcium levels toward normal by increasing
bone resorption.
Skeletal Resistance to the Calcemic Action of PTH
The calcemic response to an infusion of PTH extract is
less in hypocalcemic patients with renal insufficiency than
in normal subjects or in patients with hypoparathyroidism.
In experimental animals, prior parathyroidectomy corrects
the defect. This suggests that some form of downregulation

lames A. Delmez

450

Chronic Renal Failure

Ca
Sensor

Decreased
1,25 (OHfe
D3
Receptors
FIGURE

Recently, a calcium receptor in parathyroid cell membranes has been cloned. By sensing extracellular calcium,
the receptor regulates PTH secretion by changes in phosphoinositide turnover and cytosolic calcium levels. Decreased number of calcium receptors has been shown in
glands of uremic patients. This may be another mechanism
affecting the sensitivity of the parathyroid gland to calcium
in renal failure.
Parathyroid Hyperplasia

Pathogenesis of hyperparathyroidism in renal

failure. From Slatopolsky E, Delmez J: Miner
Electrolyte Metab 21:91-96, 1995. With permission from S.
Karger AG Basel.

of the calcemic response to PTH occurs when the hormo ne

is in excess. This would lead to a vicious cycle in renal
failure where the impaired calcemic response leads to hypocalcemia, thus stimulating PTH secretion that would further impair the calcemic response of the bone. Phosphorus
restriction improves the calcemic response to PTH via unknown mechanisms.

Parathyroid hyperplasia is a prominent finding in uremic

patients with severe secondary hyperparathyroidism. Little
is known about the factors that lead to the cellular proliferation. Studies in vivo in uremic animals, however, suggest
that calcitriol administration retards the development of
parathyroid hyperplasia independent of changes in serum
calcium levels. Hyperplasia, once established, is not reversed by short-term calcitriol treatment. Dietary phosphorus restriction also prevents parathyroid hyperplasia in uremia rats without affecting serum calcium or calcitriol
concentrations. In uremic subjects with secondary hyperparathyroidism, monoclonal allelic losses on chromosome
11 have been demonstrated in the glands. This finding suggests that monoclonal transformation of previous hyperplastic parathyroid tissue often occurs.
Diagnosis of Uremic Hyperparathyroidism

The diagnosis of severe hyperparathyroidism is made by

the findings of high levels of PTH and radiographic changes
of osteitis fibrosa. Typically, these features include subperiosteal erosions of the phalanges (Fig. 4), and erosions at
the proximal end of the tibia, the neck of the femur or
humerus, and the inferior surface of the distal end of the
clavicle. In the skull, there is a mottled and granular (salt

Altered Calcium-Regulated PTH Secretion

An insensitivity to the suppressive effects of calcium on

PTH secretion has been shown in vitro in glands obtained
from patients with chronic renal failure. The resistance
to suppression by calcium may be overcome in a dosedependent manner if the glands are incubated with calcitriol. Subsequent studies have shown that calcitriol reduces
prepro-PTH messenger RNA levels by decreasing the rate
of gene transcription. It has also been shown that there
are a decreased number of calcitriol receptors in the uremic
parathyroid gland that may promote a relative resistance
to the steroid. Interestingly, the calcitriol receptor mRNA
increases when the parathyroid gland is exposed to calcitriol. This suggests that calcitriol may upregulate its own receptor.
The low levels of calcitriol observed in patients with
advanced renal failure could play a role in the abnormal
secretion of PTH. Several studies have evaluated the PTH
response to hypercalcemic suppression and hypocalcemic
stimulation before and after intravenous calcitriol. Some,
but not all investigators, have shown an increased sensitivity of the gland to ambient calcium levels following calcitriol.

FIGURE 4 Radiographic changes of osteitis fibrosa. There are

subperiosteal erosions of the phalanges and tuft (arrow).

69. Renal Osteodystrophy and Other Musculoskeletal Complications of Chronic Renal Failure

and pepper) appearance with areas of resorption commonly

associated with areas of osteosclerosis. Osteosclerosis is
due to an increase in the thickness and number of trabeculae in spongy bone.
The correct interpretation of PTH levels in renal failure
depends on an understanding of the metabolism of PTH
and the specificity of the assay used to measure the hormone. PTH is primarily secreted as the intact hormone
containing 84 amino acids. In the circulation, intact PTH
undergoes degradation by the liver and kidneys, yielding
biologically active amino-terminal and biologically inactive
mid- and carboxyl-terminal fragments . The removal of the
latter two depends primarily, if not exclusively, on glomerular filtration and subsequent reabsorption and degradation
by the renal tubules. Therefore, in renal failure these fragments accumulate. This results in multiple forms of circulating PTH in which the concentration of biologically inactive
fragments is about 100-fold greater than that of the biologically active amino-terminal fragments. To interpret PTH
levels in renal failure, it is critical to know if the assay
measures the intact, amino-terminal, mid-terminal, and/or
carboxyl-terminal portions of the molecule. For example,
if one uses an assay that measures the mid- and carboxylterminal regions, the values associated with severe hyperparathyroidism may be 100-fold greater than the upper
limits of normal determined in subjects with normal renal
function. By contrast, results of an assay specific for the
amino-terminal portion or intact PTH may suggest severe
osteitis fibrosa when the results are 10 times the upper
limits of normal. Most clinical centers have now adapted
the intact PTH assay, and its use is encouraged. Without
liver disease, the serum alkaline phosphatase levels often
correlate with PTH and may confirm the presence of hyperparathyroidism. A bone biopsy may be necessary when the
diagnosis is uncertain.
Prevention and Treatment of Uremic Hyperparathyroidism
Phosphorus Control

In early renal failure, phosphorus accumulation may be

avoided by restriction of a dietary phosphorus intake to
600 to 800 mg/day. This generally involves limiting meats
and dairy products. More strict dietary phosphorus restriction is usually impractical. When the GFR falls to less than
20 to 30 mL/min, hyperphosphatemia develops and agents
that bind phosphorus in the bowel are usually necessary.
In the past, phosphorus binders containing aluminum were
commonly used. It is now known that this metal may be
absorbed and can cause substantial toxicity (see "Osteomalacia"). Calcium-containing phosphorus binders are now
widely used. Agents such as calcium carbonate are most
effective in binding phosphorus when given with meals in
proportion to the usual phosphorus content of each meal.
When calcium carbonate is ingested in the fasted state, less
phosphorus is bound and more calcium is absorbed by
the gastrointestinal tract. Because there is both patient-topatient variability and variability from meal to meal in
the same patient, a thorough dietary history and ongoing
counseling are essential. Mismatching of the amount o r

451

timing of calcium carbonate and phosphorus ingested may

lead to inadequate phosphorus control or hypercalcemia.
The usual starting dose is 500 to 1000 mg/day of elemental
calcium (1250 to 2500 mg calcium carbonate, 1 to 2 pills/
day) with further titration based on the resultant changes
in calcium, phosphorus, and PTH levels. Recently, much
interest has focused on calcium acetate as a phosphorus
binder. Short-term studies have shown that this compound
binds about twice as much phosphorus per calcium absorbed compared with calcium carbonate. Compared with
calcium carbonate, one half of the amount of elemental
calcium is usually effective. Unfortunately the decreased
dose does not result in a lower incidence of hypercalcemia. In addition, because calcium acetate pills contain only
167 mg of elemental calcium, patients taking this preparation often require greater number of pills per day. Thus,
there may be no clinical advantage in using this preparation.
As patients begin dialysis therapy, the dietary protein
requirements increase to more than 1 g kg" 1 day" 1. This
results in a requisite increase in dietary phosphorus to 800
to 1200 mg/day. Although hemodialysis removes approxi mately 1000 mg/treatment three times a week and continuous ambulatory peritoneal dialysis 300 mg/day, these effects
are not sufficient to avoid the use of phosphorus binders
in well-nourished patients. The physician may control the
calcium fluxes during dialysis treatments by changing the
concentration of calcium in the dialysate. A dialysate calcium concentration of 3.25 to 3.50 mEq/L (6.5 to 7.0 mg/
dL, all ionized) causes a net influx of calcium to the patient.
The use of phosphorus binders containing calcium also
cavies a risk of developing hypercalcemia. If this occurs,
the physician has the flexibility of decreasing the concentration of calcium in the dialysate to 2.0 or 2.5 mEq/L.
Control of Calcium

As the GFR falls below 50 mL/min, there is impaired

absorption of calcium by the gastrointestinal tract. This is
due partly to the decreased renal synthesis of calcitriol.
Administration of sufficient amounts of calcium to bind
phosphorus may suffice in reversing the negative calcium
balance. In those patients with good control of phosphorus
levels who nonetheless are hypocalcemic, calcium supplements may be added at night or between meals on an empty
stomach. Alternatively, if the PTH levels are excessive
(>200 to 400 pg/mL), calcitriol may be started. Calcium
and phosphorus levels must be monitored closely because
there is a tendency for hypercalcemia to develop after
several months, often heralded by a fall in alkaline phosphatase to the normal range. A similar approach is appro priate for patients treated with dialysis. Increasing the dialysate calcium to 3.5 mEq/L is another option for increasing
calcium levels.
Use of Vitamin D Sterols

Treatment with oral vitamin D, dihydrotachysterol, 25hydroxyvitamin D, and calcitriol have all been shown to
lessen symptoms of bone pain, improve bone histology,
and lower PTH levels. All confer a risk of hypercalcemia.
Vitamin D-induced hypercalcemia is particularly pro -

452

James A. Delmez

longed and may require weeks for resolution. The duration

of hypercalcemia with calcitriol is usually 2 to 4 days. It is
also the most potent metabolite of vitamin D in directly
suppressing PTH secretion. Therefore, most investigator
consider it the vitamin D sterol of choice for suppressing
hyperparathyroidism in renal failure. Calcitriol should not
be used when hyperphosphatemia is present because a
calcium-phosphorus product of greater than 75 (with each
concentration expressed in mg/dL) may lead to the development of soft tissue calcifications. In addition, calcitriol
increases the gastrointestinal absorption of phosphorus and
may require an increase in the dose of phosphorus binders.
Analogues of calcitriol have been developed that suppress
PTH secretion, yet are less active in raising calcium levels.
Their efficacy is undergoing evaluation in clinical trials.
Many studies have demonstrated a marked suppression
of PTH by administering 1.0 to 2.0 /^g of calcitriol intravenously after each hemodialysis treatment three times a
week. This suppression, which is greater than that observed
with oral daily calcitriol, may be due to the very high serum
levels achieved when the drug is given via this route. Pulse
oral calcitriol (2 to 4 jug twice a week) also results in high
levels and PTH suppression. This regimen is particular
useful in patients treated with peritoneal dialysis wherein
ready access to the circulation is not available. Some have
advocated doses as high as 6 to 8 /ug per treatment in
refractory hyperparathyroidism. Whatever the route of administration and the dose, calcium and phosphorus levels
should be carefully monitored during treatment with calcitriol. If PTH levels do not fall into an acceptable range
after 1 year of treatment, a surgical parathyroidectomy
should be considered.
OSTEOMALACIA

Osteomalacia is characterized by histological findings of

impaired mineralization activity and an increase in the osteoid seam width (Fig. 5). Thus, in contrast to osteitis fibrosa,
the rate of bone turnover is low. Patients often complain
of bone and muscle pain, and spontaneous fractures occur
in approximately 15% of those afflicted.
Pathogenesis

Although abnormalities of vitamin D metabolism and

the presence of persistent metabolic acidosis may contribute to the development of osteomalacia, the most common
cause in patients on dialysis is aluminum intoxication. Aluminum may accumulate due to contamination of the dialysate with this metal. With proper pretreatment of the tap
water used to prepare dialysate, this route has diminished in
relative importance. The main current source of aluminum
accumulation is the ingestion of phosphorus binders containing aluminum. Because aluminum is normally excreted
via urinary excretion, patients on dialysis are particularly
prone to develop toxicity that, in its most severe state,
encompasses dementia and premature death. Those with
diabetes, prior parathyroidectomy, failed renal transplant,
complete anuria, or a long history of consumption of large

FIGURE 5 Histological features of osteomalacia. There are wide

osteoid seams (O) surrounding trabecular bone (T). There is a
paucity of osteoclasts and osteoblasts, thus reflecting a low rate
of bone turnover. (Courtesy Steven L. Teitelbaum, M.D.; X40,
modified Masson stain.)

amounts of aluminum-containing phosphorus binders are

at highest risk. Uremic patients prescribed medications
containing citrate (calcium citrate, Shohl's solution), along
with those medications containing aluminum (Basaljel,
Amphogel, Carafate) are particularly prone to aluminum
accumulation.
Diagnosis

The diagnosis of osteomalacia is often difficult. The clinical setting of a large exposure to aluminum and bone pain
suggests the diagnosis. Aluminum directly suppresses PTH
secretion, and a low PTH level is characteristic but not
universal. The radiographic findings of osteomalacia are
not distinctive. Some data suggest that a basal serum aluminum level of MOO /ug/L and an increment in aluminum
level of >150 /u-g/L following chelation with deferoxamine
(DFO) in combination with a low PTH level has a high

69. Renal Osteodystrophy and Other Musculoskeletal Complications of Chronic Renal Failure
predictive power for the presence of aluminum bone dis ease. For the DFO challenge test, 20 mg/kg is administered
intravenously at the end of dialysis and the serum aluminum is measured 24 to 48 hours later. The "gold standard"
however remains the bone biopsy with appropriate staining
for aluminum.

453

be seen in patients with previously established osteitis fibrosa who are developing aluminum-related bone disease.
The treatment is withdrawal of the aluminum exposure
and aggressive treatment of the hyperparathyroidism.
Dialysis- Related Amyloidosis

Prevention and Treatment of Osteomalacia Due

to Aluminum

Prevention of osteomalacia is critical with attention given

to dialysate purity and avoidance of phosphorus binders
containing aluminum, if possible. Chronic chelation therapy with DFO cures the osteomalacia, and follow-up biopsies usually show the development of osteitis fibrosa of
variable severity. The major mechanism for this is the removal of aluminum from the bone surface, associated with
marked increases in osteoblast number and rate of bone
formation. A major problem with DFO treatment is the
risk of fatal mucormycosis infections, which occur in up to
5% of treated patients. DFO also binds iron and the DFOiron chelate can function as a siderophore to stimulate the
growth of mucormycosis. Thus, the use of DFO should be
restricted to those who have severe symptoms of aluminum
intoxication and histological evidence of aluminum accumulation in the bone. Patients with mild symptoms may be
treated by eliminating all aluminum-containing phosphorus
binders and substituting ones containing calcium. Over 1
to 2 years, the aluminum burden will decrease as will the
symptoms.

It has long been known that there is an association between bone cysts, pathological fractures, arthritis, and
carpal tunnel syndrome in patients treated with longterm dialysis. It is now clear that these are often due to
the deposition of a form of amyloid unique to dialysis patients. This protein is composed of intact and modified j82 microglobulin. This constant light chain of HLA class I
antigens is excreted in the urine. Therefore, very high serum
levels are uniformly seen in patients on dialysis. The amy loid is most frequently deposited in the flexor retinaculum
of the wrist, entrapping the median nerve and causing carpal tunnel syndrome. The symptoms of hand pain can be
relieved by a surgical release of the median nerve. The
amyloid may also invade the synovium of joints causing
pain, effusions and erosive arthritis. Periarticular radiolucent bone cysts (Fig. 6), the result of the replacement of
bone by amyloid, may result in pathological fracture. Once
established, there is no proven treatment for the amyloid
arthropathy. High flux hemodialysis or hemodiafiltration
removes some /32 -microglobulin and may delay the onset

Adynamic Bone Lesions

Adynamic (also termed aplastic) bone lesions are characterized by decreased bone mineralization, but normal
amounts of osteoid. In approximately 50% of cases, the
cause is aluminum deposition. The treatment is the same
as osteomalacia due to aluminum. Little is known about
the etiology in the remaining half of patients. However,
this lesion may be more common in patients treated with
peritoneal dialysis, in the elderly, and in patients with diabetes. The PTH levels are generally <3 times the upper limits
of normal when measured with an intact PTH assay. It is
likely that the lesion in some cases is the result of overzealous suppression of PTH. If so, allowing the PTH level to
increase to four times the upper limits of normal may be
advisable. It should be noted however that patients with
adynamic bone lesions not due to aluminum overload report fewer symptoms of bone pain and myopathy and develop fewer fractures than those with either aluminumrelated or osteitis fibrosa bone disease diagnosed by
bone biopsy.
MIXED LESIONS

Some patients display histological evidence of both osteitis fibrosa and osteomalacia on bone biopsy. Such patients
frequently have high PTH levels and impaired bone formation and mineralization. Mixed renal Osteodystrophy may

FIGURE 6 Bone cysts in a hemodialysis patient due to amyloid

infiltration of bone.

454
of the clinical expression of the disease. Early renal transplantation completely prevents the disease.
Renal Osteodystrophy Following Renal Transplantation
Following a successful renal transplant, the histological
abnormalities of mild to moderately severe osteitis fibrosa
usually resolve within 1 year. However, bone density of
the lumbar vertebrae decreases due to the effect of steroids
in impairing bone formation. Severe hyperparathyroidism
may not completely resolve and elevated PTH levels may
persist for 5 to 10 years. In those patients with osteomalacia,
return of renal function allows the removal of aluminum
and, as a result, improvement in the bone histology. Unpublished data suggest that those with adynamic bone lesions
are prone to avascular necrosis of the femoral head following transplant. The symptoms of amyloid deposition often
improve dramatically following transplantation but the radiographic findings do not change. This suggests that the
amyloid may be irreversibly bound to the tissue.
OVERALL APPROACH TO THE MANAGEMENT OF
RENAL OSTEODYSTROPHY

Renal osteodystrophy is a dynamic process whose severity and even type of lesion may vary with time. In early
renal failure, phosphorus restriction alone may suffice.
With more severe renal failure (GFR less than 30 mL/
min), calcium malabsorption should be countered by the
administration of oral calcium with meals if the phosphorus
level is high or at night if the phosphorus is normal. In
the face of a calcium phosphorus product of greater than
75 mg 2/dL2, the phosphorus should initially be lowered by
use of aluminum-containing phosphorus binders to prevent
soft tissue calcifications. When the product has decreased
below this value, calcium carbonate with meals should be
substituted for the long-term binding of phosphorus. The
target calcium and phosphorus levels are 9.5 to 10.5 and
4.5 to 5.5 mg/dL, respectively. Frequent monitoring of these
values is necessary to avoid iatrogenic complications such
as hypercalcemia. In hemodialysis patients, calcium and
phosphorus levels should be determined monthly and PTH
concentrations quarterly. The target intact PTH levels
should be three to four times the upper limits of normal.
The dose of calcium carbonate should be adjusted in relation to the calcium, phosphorus, and PTH levels. For example, if the calcium is 8.5 mg/dL, the phosphorus level is
acceptable, and the intact PTH concentration is normal to
twice normal, one would consider doing nothing or lowering the dose of calcium carbonate. If, with the same levels
of calcium and phosphorus, the PTH level is five times
normal and progressively increasing, one would attempt a
more aggressive approach. This may be achieved by increasing the dose of calcium given at night or by starting
calcitriol therapy. When prescribing calcitriol, the calcium
and phosphorus levels must be assessed more frequently.
As with calcium carbonate, calcitriol should not be given
in the presence of an elevated calcium-phosphorus product.
This presents a problem in the patient with severe hyper-

James A. Delmez

parathyroidism. In this situation the patient's elevated serum calcium or phosphorus may be emanating from bone
due to enhanced bone resorption. A surgical parathyroidectomy is often required. The other main indication for surgery is persistently high levels of PTH, corresponding to
the presence of severe osteitis fibrosa and bone pain. A
bone biopsy is usually not required to make the diagnosis.
If, however, the patient is at risk for the development of
aluminum accumulation (see above), a bone biopsy may be
warranted to confirm the diagnosis and exclude aluminumrelated osteomalacia as a cause for the bone pain or hypercalcemia. At time of surgery, all four glands should be
identified. Some surgeons will do a 3i gland parathyroidectomy, whereas others will remove all glands and then implant small pieces of one gland into the forearm to avoid
future neck surgery if hyperparathyroidism recurs.
Bibliography
Andress DL, Keith MD, Norris C, et al.: Intravenous calcitriol in
the treatment of refractory osteitis fibrosa of chronic renal failure. N Engl J Med 321:274-279, 1989.
Boelaert JR, Fenves AZ, Coburn JW: Deferoxamine therapy and
mucormycosis in dialysis patients; report of an international
registry. Am J Kidney Dis 18:660-667, 1991.
Brown EM, Wilkson RE, Eastman RC, et al.: Abnormal regulat ion
of parathyroid hormone release by calcium in secondary hyperparathyroidism due to chronic renal failure. J Clin Endocrinol
Metab 54:172-179, 1982.
Delmez JA, Slatopolsky E: Hyperphosphatemia: Its consequences
and treatment in patients with chronic renal disease. Am J Kid ney Dis 19:303-317, 1992.
Delmez JA, Tindira C, Grooms P, et al.: Parathyroid hormone
suppression by intravenous 1,25-dihydroxyvitamin D: A role for
increased sensitivity to calcium. / Clin Invest 83:1349-1355,1989.
Dunlay R, Rodriguez M, Felsenfeld AJ, et al.: Direct inhibitory
effect of calcitriol on parathyroid function (sigmoidal curve) in
dialysis. Kidney Int 36:1093-1098, 1989.
Fukuda N, Tanaka H, Tominaga Y, et al.: Decreased 1,25dihydroxyvitamin D3 receptor density is associated with a more
severe form of parathyroid hyperplasia in chronic uremic patients. J Clin Invest 92:1436-1443, 1993.
Goodman WG, Belin T, Gales B, et al.: Calcium-regulated parathyroid hormone release in patients with mild or advanced secondary hyperparathyroidism. Kidney Int 48:1553-1558, 1995.
Hruska KA, Teitlebaum SL: Renal osteodystrophy. N Engl J Med
333:166-174, 1995.
Julian BA, Laskow DA, Dubovsky J, et al.: Rapid loss of vertebral
mineral density after renal transplantation. N Engl J Med
325:544-550, 1991.
Kifor O, Moore FD, Wang P, et al.: Reduced immunostaining for
the extracellular Ca2+ sensing receptor in primary and uremic
secondary hyperparathyroidism. / Clin Endocrinol Metab
81:1598-1606, 1996.
Llach F: Secondary hyperparathyroidism in renal failure: The
trade-off hypothesis revisited. Am J Kidney Dis 25:663-679,
1995.
Malluche H, Faugere M: Renal bone disease 1990: An unmet
challenge for nephrologist. Kidney Int 38:193-210, 1990.
Nebeker H, Andress D, Milliner D, et al.: Indirect methods for
the diagnosis of aluminum bone disease: plasma aluminum, the
desferrioxamine infusion test, and serum iPTH. Kidney Int
18:S96-S99, 1986.

70. Cardiac Function and Cardiac Disease in Renal Failure

Sheikh MS, Ramirez A, Emmett M, et al.: Role of vitamin Ddependent and vitamin D-independent mechanisms in absorption of food calcium. J Clin Invest 81:126-132, 1988.
Sherrard DJ, Hercz G, Pei Y, et al.: The spectrum of bone disease
in end-stage renal failureAn evolving disorder. Kidney Int
43:436-442, 1993.
Slatopolsky E, Weerts C, Lopez-Hilker S, et al.: Calcium carbonate
as a phosphat e binder in patients with chronic renal failure
undergoing dialysis. N Engl J Med 315:157-161, 1986.

455
Slatopolsky E, Weerts C, Norwood K, et al.: Long term effects of
calcium carbonate and 2.5 mEq/L calcium dialysate on mineral
metabolism. Kidney Int 36:897-903, 1989. Slatopolsky E,
Finch J, Denda M, et al.: Phosphorus restriction
prevents parathyroid gland growth. / Clin Invest 97:25342540, 1996. Szabo A, Merke J, Beier E, et al.: 1,25(OH)2
vitamin D3 inhibits
parathyroid cell proliferation in experimental uremia. Kidney
Int 35:1049-1056, 1989.

CARDIAC FUNCTION AND CARDIAC

DISEASE IN RENAL FAILURE
ROBERT N. FOLEY AND PATRICK S. PARFREY

INTRODUCTION

Uremic pericarditis was a frequent cause of death in the

predialysis era. Since the 1960s, cardiac disease has been
shown repeatedly to be the major cause of death in endstage renal disease (ESRD). This is still the case today,
with older and sicker patients being accepted for renal
replacement therapy. The excessive cardiac mortality of
ESRD crosses the divides of nationality, race, gender, and
cause of ESRD. Cardiac disease also accounts for a substantial degree of comorbidity. Most studies in industrialized
nations show that symptomatic ischemic heart disease and
cardiac failure are present prior to initiation of ESRD
therapy in more than one third of patients. About 10% of
patients will go on to develop each of these conditions
every year after starting dialysis therapy. This chapter dis cusses the natural history, risk factors, and treatment of
cardiomyopathy and ischemic heart disease in chronic uremia. The pericarditis seen in uremic patients will also be dis cussed.
EPIDEMIOLOGY
Prevalence

Echocardiographic left ventricular (LV) hypertrophy,

coronary artery disease, and cardiac failure are conservatively estimated to be between two and five times more

Primer on Kidney Diseases, Second Edition

prevalent in ESRD patients compared to an age-matched

general population.
Abnormalities of LV structure and function are very
common in patients starting renal replacement therapy.
Concentric LV hypertrophy (where wall thickening occurs
in response to pressure overload) is present in 39% of
patients starting dialysis programs for ESRD. Twentyseven percent of patients had left ventricular dilatation
(increase in cavity size in response to volume overload)
and 18% had systolic dysfunction (inadequate contractility,
defined on echocardiography as a fractional shortening less
than 25%). A recent study of patients with chronic renal
failure (not yet dialysis dependent) indicated that these
abnormalities develop early in chronic renal failure and
progress rapidly as renal function declines. Twenty-seven
percent of patients with creatinine clearance greater than
50 mL/min had LV hypertrophy; this figure rose to 31%
for clearances between 25 and 50 mL/min and 45% for
clearances less than 25 mL/min.
Almost half of all hemodialysis patients in the US have
had clinically defined ischemic heart disease. Silent coronary artery disease is also common in dialysis patients,
especially in those with diabetes mellitus. The situation is
confounded even more by the observation that over one
quarter of dialysis patients with typical angina have normal
coronary arteriograms.
Cardiac failure is also very common in dialysis patients;
a recent report from the United States Renal Data System
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

Robert N. Foley and Patrick S. Parfrey

456
Registry suggested that almost one half of all patients had
a history of cardiac failure.
Risk Factors
Many of the traditional, Framingham-type parameters
are also risk factors for cardiac disease in chronic uremia.
Older age and diabetes mellitus have been consistently
associated with mortality in ESRD patients.
In one study, even a moderate degree of hypertension
was associated with progressive cardiac enlargement and
subsequent cardiac failure. Hypertension was also associated with the development of new symptomatic ischemic
heart disease. Two thirds of all deaths were preceded by an
admission for cardiac failure. After development of cardiac
failure, low blood pressure was the single greatest predictor
of subsequent mortality. Low blood pressure and low serum
cholesterol have been associated with mortality in largescale epidemiological studies of dialysis patients as well.
The basis for this apparent paradox is that low serum cholesterol levels reflect malnutrition, which has been shown
to be the biggest predictor of mortality in most recent
studies. High levels of lipoprotein (a) and apoB have been
associated with an adverse outcome in patients with
chronic uremia.
These data suggest that the association between low
blood pressure and mortality reflects the very high frequency of advanced cardiomyopathy in dialysis patients.
They also suggest that aggressive management of hypertension is needed in patients without clinically apparent cardiac disease. In predialysis patients with diabetic nephropathy, very aggressive control of blood pressure slows the
progression of chronic renal failure; some have advocated
targeting the lowest blood pressure that does not lead to
symptoms a practical target level is 120/80 mm Hg. There
is emerging evidence that this may also be an appropriate
target level in nondiabetics with chronic renal impairment.
The target blood pressure that minimizes cardiac risk in
chronic uremia is not yet clear.
The data linking smoking to cardiac disease in ESRD
are inconsistent. Some studies suggest that smoking is an
independent mortality factor in ESRD. Smoking and diabetes appears to be a particularly adverse combination of
risk factors.
Recent epidemiological data suggest that many factors
related to the uremic state may be associated with cardiac
disease. As noted, hypertension contributes to development of concentric LV hypertrophy, LV dilatation, ischemic heart disease, and cardiac failure. Anemia promo tes
development of LV dilatation and cardiac failure. Hyperparathyroidism and inadequate dialysis may also play a
role. Most of these factors are amenable to intervention.
Prognosis
Systolic dysfunction, concentric LV hypertrophy, and
LV dilatation are all strong and independent predictors of
future cardiac failure, ischemic heart disease, and death in
dialysis patients. The adverse association between systolic

dysfunction and mortality is immediate. In contrast, the

adverse impact of concentric LV hypertrophy and LV dilatation on survival is only apparent after a lag phase of
approximately 2 years, suggesting a possible time window
to target interventions. It is very likely that these echocardiographic abnormalities lead to death via the intermediate
step of cardiac failure.
Coronary artery disease predicts mortality independently of age and diabetes. It predisposes to systolic dysfunction and cardiac failure, which seem to be the major mediators of its adverse impact in ESRD patients. Clinically
defined cardiac failure is a rapidly lethal condition in ESRD
patients, much the same as it is in the general population.
In a prospective cohort study, cardiac failure preceded two
thirds of all deaths of chronic dialysis patients with a median
time interval of 13 months between admission for cardiac
failure and death. It is often difficult to distinguish intrinsic
cardiac dysfunction from salt and water excess in dialysis
patients. It is thought that chronic salt and water excess
can lead to progressive cardiac failure, which adds to the
diagnostic confusion. Many dialysis patients can tolerate
vast amounts of extracellular fluid (ECF) volume expansion without developing pulmonary edema. Although dialysis patients with normal hearts can develop pulmonary
edema with the rapid accumulation of large amounts of
salt and water, it is safer to assume cardiac dysfunction as
opposed to simple ECF volume expansion in patients who
develop pulmonary edema. Concurrently, patients like this
should undergo further cardiac workup, initially by echocardiography. Noninvasive technology that can rapidly give
an accurate assessment of ECF volume status should be
routinely available in the near future. This technology
should assist in distinguishing simple ECF volume expansion from intrinsic cardiac dysfunction.
Valvular dysfunction is common in dialysis patients.
Some studies have noted abnormal mitral or aortic valvular
abnormalities in half of all patients with chronic renal failure. Mild to moderate mitral regurgitation, usually secondary to LV dilatation, and calcification of the mitral valve
are the most commonly reported abnormalities. The prevalence of hemodynamically important valvular dysfunction
is not well described in the literature. In one study, about
10% of patients had hemodynamically important abnormalities of the mitral or aortic valves; the prognosis of these
patients appears to be determined by the degree of associated cardiomyopathy. Figure 1 shows survival of ESRD
patients with the major clinical and structural manifestation
of cardiac disease.
DIAGNOSIS
Cardiomyopathy
Chest radiographs and electrocardiography are not sufficiently sensitive to detect cardiac enlargement. Echocardiography is noninvasive and relatively easy to perform. It
gives information about cardiac dimensions, valve function,
systolic function, and diastolic function. It is not clear how
often echocardiography should be performed in chronic

70. Cardiac Function and Cardiac Disease in Renal Failure

457

Baseline Echocardiography

80

Baseline Clinical Status

ence of coronary arterial narrowing. The role of noninvasive testing is still unclear. Exercise-based stress tests are
often uninformative in ESRD patients, because most patients are unable to achieve an adequate level of exercise
intensity. Of the other noninvasive screening tests available, adenosine 201thallium, dipyridamole 201thallium, and
dobutamine stress echocardiography appear useful. Coronary arteriography, however, remains the gold standard.
ESRD patients with symptomatic ischemic heart disease
should be investigated with coronary arteriography if their
physical state is such that revascularization would be seriously considered. The optimal approach to screening for
coronary artery disease in asymptomatic prospective renal
transplant patients is a matter of dispute (see Chapter 74)
Our current approach is to perform noninvasive screening
on the following subgroups: patients over 45 years, diabetics over 25 years, smokers, and those with ischemic changes
on electrocardiography. Those who do not clearly have
negative noninvasive screening tests should have arteriography.
T R E A T M E N T Risk
Factor Management

80

Months
FIGURE I (Top) Survival according to baseline echocardiography in patients starting ESRD therapy. (I) Normal left ventricle.
(II) Concentric LV hypertrophy. (Ill) LV dilatation. (IV) Systolic
dysfunction. Data from Foley et al.: J Am Soc Nephrol 5:20242031, 1995; Parfrey et al.: Nephrol Dial Transplant 11:1277-1285,
1996. (Bottom) Survival according to presence or absence of
ischemic heart disease and cardiac failure in patients starting
ESRD therapy. (A) No ischemic heart disease, no cardiac failure.
(B) Ischemic heart disease, no cardiac failure. (C) Cardiac failure,
no ischemic heart disease. (D) Both ischemic heart disease and
cardiac failure. Data from Harnett et al.: Kidney Int 47: 884-890,
1995, Parfrey et al.: Kidney Int 49:1428-1434, 1996. With per-

ESRD. Cardiomyopathy often progresses rapidly in

chronic uremia. Many of the causes of this rapid progres sion are treatable. There may be value in annual performance of echocardiography in chronic uremic patients.
There are, however, no comparative studies available to
tell us the optimum use of echocardiography in chronic
uremic patients. It is important that echocardiography be
carried out as close to dry weight as possible, preferably
within 1 kg.
Ischemic Heart Disease
ESRD patients with symptoms of ischemic heart disease,
diabetic patients, and older patients undergoing pretransplant assessment should be formally evaluated for the pres -

There are no randomized trials to assess the efficacy of

modification of Framingham-type risk factors in uremic
patients without coronary disease and cardiomyopathy.
The cardiac morbidity and mortality of these patients probably exceeds that of survivors of myocardial infarction in
the general population. In the absence of high-quality data
from randomized controlled trials, the following targets for
all uremic patients, should minimize the risk of new cardiac
disease and the progression of existing cardiac disease:
nonsmoking, euglycemia, blood pressure <120 systolic and
80 diastolic in predialysis patients, <140 mm Hg systolic
and 90 mm Hg diastolic in dialysis patients, total serum
cholesterol less than 200 mg/dL, LDL cholesterol less than
140 mg/dL, hemoglobin 10 0.5 g/dL, KtIV > 1.2 in hemo dialysis patients, weekly KtIV > 2 in peritoneal dialysis
patients, serum albumin >3.8 g/dL, serum calcium >2.2
mmol/L, PTH < 200 ng/L.
At present there is no good evidence pointing to the
superiority of any antihypertensive or lipid lowering strategy in this overall picture. However, in predialysis patients,
there is considerable evidence that angiotensin-converting
enzyme (ACE) inhibitors may be the vasoactive agent of
choice, because of their ability to retard the progression
to ESRD. Malnutrition is a major killer in dialysis patients;
as most dialysis patients are already on restricted diets, the
threshold for using pharmacological treatment for hyperlipidemia should be low. HMG-CoA reductase inhibitors
and fibrates are well tolerated and safe in patients with
chronic renal failure. There is, as yet, no evidence that full
correction of renal anemia is better than partial correction
of anemia to hemoglobin levels of 10 g/dL. Erythropoietin
(rHuEPO) is the agent of choice to treat renal anemia.
Using rHuEPO in this manner will lead to an increase in
blood pressure in about one third of patients; for most

458

Robert N. Foley and Patrick S. Parfrey

patients this increase in blood pressure is readily reversible

with standard antihypertensive treatment. Using rHuEPO
to increase hematocrit also increases the chance of arteriovenous graft loss; for most patients the benefits of partial
correction of renal anemia outweigh the potential risks.
The efficacy and risks of routine use of antiplatelet agents
and anticoagulants in this patient group have not been
satisfactorily resolved.
Management of Coronary Artery Disease
As in the general population, antiplatelet therapy should
be used unless otherwise contraindicated. Calcium channel
blockers, /3-blockers, and nitrates are the cornerstone of
symptomatic therapy. Risk factor management should be
aggressive in this group. Patients should be encouraged to
reach ideal body mass, stop smoking, and start a graded
exercise program. Hypertension and hyperlipidemia should
be aggressively managed. /3 blockers should be prescribed
unless contraindicated. It is important that anemia be
treated in this group; although full correction of anemia
may ameliorate symptoms, such an approach has not been
shown to increase longevity in this group, and has been
associated with a clear excess of arteriovenous access loss;
the target hemoglobin for these patients should be approximately 10 g/dL.
ESRD patients with symptoms of ischemic heart disease
who do not have a very large burden of noncardiac comorbidity should have coronary arteriography. Dialysis patients fulfilling the anatomical and functional criteria used
in the general population are likely to benefit from coronary revascularization. Generally accepted criteria are:
(1) one, two, or three vessel disease with angina refractory
to medical management, where the intent is symptom relief;
(2) left main coronary artery disease where the intent is
improved survival; and (3) triple-vessel disease associated
with ventricular dysfunction or easily inducible ischemia,
where the intent is also improved survival. There is some
evidence that recurrence rates after coronary angioplasty
are unacceptably high in dialysis patients. The perioperative morbidity and mortality rates associated with coronary
artery bypass surgery are higher in dialysis patients than
in the general population. Recent studies suggest that these
rates are acceptable and justifiable on the basis of good
subsequent survival rates.
Management of Cardiomyopathy
Eighty percent or more of dialysis patients have cardiomyopathy, which can progress rapidly. Because it is difficult
to predict progression, and because there are many potentially reversible etiological factors (such as coronary artery
disease, anemia, inadequate dialysis, uncontrolled hypertension, malnutrition, and hyperparathyroidism), we believe that all chronic renal failure patients should have
regular echocardiography; our current recommendation is
that they be done annually. All patients who develop cardiac failure should have echocardiography to try to distinguish systolic from diastolic dysfunction. ACE inhibitors,

calcium channel blockers, and /3-blockers form the cornerstone of treatment in concentric LV hypertrophy. Based
on data from the nonuremic population, we use ACE inhibitors as the first vasoactive agent in patients with LV dilatation, systolic dysfunction, and symptomatic cardiac failure.
Digoxin may benefit patients with cardiac failure in the
presence of systolic dysfunction, but is best avoided in
patients with diastolic dysfunction. Recent studies in the
general population suggest that ^-blockers and calcium
channel antagonists may have a role as primary therapy
for cardiac failure associated with both diastolic and systolic
dysfunction. Such an approach should be considered experimental at present, and is best done in consultation with a
cardiologist. Nitrates are useful for symptomatic relief in
frank cardiac failure, whether due to systolic or diastolic
dysfunction.
PERICARDITIS IN UREMIC PATIENTS

Clinically important pericarditis is now an infrequent

occurrence, although small pericardial effusions are very
common on echocardiography. Some authors distinguish
between uremic pericarditis, which occurs in uremic individuals who have never received dialysis therapy, and pericarditis in individuals already on dialysis. The uremic milieu
is responsible for the former condition, although the specific culprits are unknown. About half of all patients who
develop pericarditis while on dialysis therapy will respond
to intensified dialysis therapy, suggesting that uremia is
causative in only a proportion of these cases; it has been
speculated that conditions such as viral pericarditis are
responsible for the remainder.
Clinical Pres entation
The major clinical findings are due to inflammation (precordial pain, dyspnea, fever, generalized ST changes on
electrocardiography) and fluid in the pericardial sac leading
to a restriction to ventricular filling (fluid overload and
intolerance of ultrafiltration). With large effusions, frank
cardiac tamponade and cardiogenic shock may occur. It is
more likely that a given volume of pericardial fluid will
lead to cardiac tamponade when this accumulates quickly.
Echocardiography should be performed to estimate the
volume of fluid within the pericardium. Intensification of
dialysis therapy (with avoidance of heparin) and analgesia
form the cornerstone of management. The use of nonsteroidal antiinflammatory drugs is recommended by some but
not all authors. In many centers, pericardiocentesis is used
for management of large effusions (>250 mL), or where
cardiac tamponade is present. The fluid is typically hemorrhagic and the pericardial sac may contain fibrin and clot.
Surgical drainage with anterior pericardiectomy may thus
be required.
Bibliography

Bloembergen WE, Stannard DC, Port FK, et al.: Relationship of

dose of hemodialysis and cause- specific mortality. Kidney Int
50: 557-565, 1996.

71. Neurological Manifestations of Renal Failure

459

Churchill DN, Taylor DW, Cook RJ, et al.: Canadian hemodialysis

morbidity study. Am J Kidney Dis 19: 214-234, 1992.
Foley RN, Parfrey PS, Harnett JD, et al.: DC, Clinical and echocardiographic disease in end-stage renal disease: prevalence, associations and prognosis. Kidney Int 47: 186-192, 1995.
Foley RN, Parfrey PS, Harnett JD, Kent GM, Murray DC, Barre
PE: The progostic importance of left ventricular geometry in
uremic cardiomyopathy. / Am Soc Nephrol 5:2024-2031, 1995.
Harnett JD, Foley RN, Kent GM, Barre PE, Murray DC, Parfrey
PS: Congestive heart failure in dialysis patients: Prevalence,
incidence, prognosis and risk factors. Kidney Int 47: 884-890,
1995.
Levin A, Singer J, Thompson CR, Ross H, Lewis M: Prevalent
left ventricular hypertrophy in the predialysis population: Identi-

fying opportunities for intervention. Am J Kidney Dis 27: 347354, 1996.

Murphy SW, Parfrey PS: Screening for cardiovascular disease in
dialysis patients. Curr Opin Nephrol Hypertens 5:532-540,1996.
Owen SR, Lew NL, Yan Liu SM, Lowrie EG, Lazarus JM: The
urea reduction ratio and serum albumin concentrations as predictors of mortality in patients undergoing hemodialysis. N Engl
J Med 329: 1001-1006, 1993.
Parfrey PS, Foley RN, Harnett JD, Kent GM, Murray DC, Barre
PE: Outcome and risk factors for left ventricular disorders in
chronic uremia. Nephrol Dial Transplant 11: 1277-1285, 1996.
Parfrey PS, Foley RN, Harnett JD, Kent GM, Murray DC, Barre
PE: Outcome and risk factors of ischemic heart disease in chronic
uremia. Kidney Int 49: 1428 -1434, 1996.

NEUROLOGICAL MANIFESTATIONS OF
RENAL FAILURE
COSMO L FRASER

Nervous system dysfunction is common among renal failure patients and contributes significantly to their overall
morbidity and mortality. Patients with chronic renal failure
who have not yet received dialysis may develop symptoms
which could be as mild as sensorial clouding or as severe
as coma and death. Even after the initiation of adequate
maintenance dialysis, patients may continue to manifest
subtle nervous system dysfunction such as impaired mentation, generalized weakness, sexual dysfunction, and peripheral neuropathy. Hemodialysis itself has been associated
with distinct disorders of the central nervous system. Dialysis disequilibrium syndrome occurs in a small number of
patients and is a consequence of the initiation of dialysis.
Dialysis dementia is a progressive and generally fatal encephalopathy which can affect patients on chronic hemodialysis as well as children with chronic renal failure who have
not yet started dialysis. Progressive intellectual dysfunction
and peripheral neuropathy can occur in patients who are
being treated with maintenance dialysis therapy.
UREMIC ENCEPHALOPATHY

The term uremic encephalopathy is used to denote the

central neurological signs and symptoms that are mani-

Primer on Kidney Diseases, Second Edition

fested by patients with either an acute or chronic deterioration of their renal function. Symptoms are usually seen
when glomerular filtration rate falls to a level below 10%
of normal. In common with other organic brain syndromes,
these patients display variable disorders of consciousness
that can affect psychomotor behavior, thinking, memory,
speech, perception, and emotion. The severity and rate of
progression of symptoms varies directly with the rapidity
with which renal failure develops. In patients with acute
renal failure, uremic symptoms are generally more severe
and progress more rapidly than in patients with chronic
renal failure. In progressive chronic renal failure, the number and severity of symptoms in patients may be quite
variable, and symptoms may also occur in a cyclic manner
in that there are intervals of well-being in an otherwise
inexorable downhill course (Table 1).
Differential Diagnosis

Uremic encephalopathy should be suspected in patients

with renal failure if there are clinical signs and symptoms
consistent with central nervous system deterioration. However, the presenting symptoms of uremic encephalopathy
Copyright 1998 by the National Kidney Foundation
All rights of reproduction in any form reserved.

460

Cosmo L Fraser
TABLE I

Neurological Manifestations of Uremia

Predialysis
Uremic encephalopathy
Acute CNS signs and symptoms of
uremia
Asterixis usually present
Indication to initiate dialysis
Postdialysis
Dialysis disequilibrium Idiogenic osmoles result in brain -tosyndrome
plasma osmolar gradient
Caused by aggressive initial HD
treatments
Dialysis dementia Epidemic, endemic, and childhood forms
Aluminum associated with
epidemic form
Frequently fatal
Prevented by deionization of
dialysis water and restriction of
oral aluminum
Avoid citrate and aluminum
compounds

Deferoxamine chelation therapy

Intellectual function
Peripheral neuropathy Generally impaired
Autonomic dysfunction Motor and sensory impairment
Postural hypotension, impaired
sweating, gastrointestinal
disturbances, impotence

may be similar to those of other metabolic encephalopathies so there is always a risk of misdiagnosis. The diagnosis
may also be made difficult because patients with renal
failure may also have other intercurrent illnesses which can
induce encephalopathy. If a patient with renal insufficiency
is taking a drug with potential central nervous system toxicity that is excreted or metabolized primarily by the kidney,
the central nervous system symptoms could be due to
achievement of toxic drug levels at ordinary dosage or
uremia, or a combination of both factors. Also, in patients
with both advanced liver and renal diseases, it is often
difficult to determine whether the encephalopathy is due
to hepatic or renal causes or both. In such patients the
BUN and serum creatinine do not always adequately reflect
the degree of renal function impairment. Because most
alcoholics consume a diet inadequate in protein and calories, they are usually malnourished and have reduced muscle mass. Thus, their ability to generate creatinine from
muscle is greatly impaired and there usually is not enough
protein intake to produce amino acids and generate urea
by the urea cycle. Apparently, normal blood urea nitrogen
and creatinine levels may actually be indicative of a diminished capacity of the patient to generate urea and creatinine, rather than a normal glomerular filtration rate. Thus,
when alcoholic patients have "normal" BUN and creatinine values, their GFR may be far below 30 mL/min.
Clinical Manifestations

Abnormalities of the mental state are early and sensitive

indicators of the development of neurological disorders in

patients with acute renal failure. The initial presentation

of patients with acute renal failure may include signs of
toxic psychosis, abnormal mental status, lassitude and lethargy, with disorientation and confusion occurring later.
Physical findings may include cranial nerve signs, nystagmus, dysarthria, abnormal gait and abnormalities of skeletal muscles, manifested by weakness, fasciculations, and
asymmetrical variation in deep tendon reflexes. These findings may progress to asterixis and hyperreflexia with unsustained clonus at the ankle. If uremia is left untreated and
allowed to progress, seizures and coma often supervene.
Electroencephalograms (EEG) in patients with acute
renal failure are generally grossly abnormal when the diagnosis of acute renal failure is first made. In most instances,
the percentage of EEG frequency less than 7 Hz is 20 times
greater than normal. The abnormal percentages of EEG
frequencies both above 9 Hz and below 5 Hz are not usually
improved by dialysis within the first few weeks of treatment,
although they may return to normal with recovery of renal
function. If renal failure continues, the EEG may transiently worsen both during and after hemodialysis and this
pattern may continue for up to 6 months after starting
dialytic therapy. In spite of these fairly characteristic EEG
findings, patients who present with elevated BUN and creatinine and mental status changes do not require an EEG
to make the diagnosis. The EEG is a tool which is not
generally used in diagnosing uremic encephalopathy.
The neurological manifestations of chronic renal failure
are also quite numerous. The EEG findings are usually not
as severe as those reported in patients with acute renal
failure. In general, there is good correlation between the
percentage of EEG frequencies and power below 7 Hz
and the decline of renal function as estimated by serum
creatinine. After the initiation of dialysis, there may be an
initial period of clinical stabilization during which the EEG
continues to deteriorate. However, after approximately 6
months of dialysis treatment, the EEG tends to normalize.
However, normal values may not be reached unless the
patient receives a kidney transplant. In patients with
chronic renal failure cognitive functions such as attention
span, speed of decision making, short-term memory and
mental manipulation of symbols are impaired.
Pathogenesis

Although many factors may contribute to uremic encephalopathy, no precise correlation exists between the
degree of encephalopathy and any of the commonly measured blood chemistries associated with renal dysfunction
(BUN, creatinine, bicarbonate, or pH). However, there is
much evidence to suggest that parathyroid hormone (PTH)
may be a uremic toxin.
In uremic animals, both the EEG and observed brain
calcium abnormalities can be prevented by parathyroidectomy. Conversely, many of the CNS abnormalities observed in uremia can be reproduced by administration of
PTH to normal animals. Parathyroid hormone has been
shown to produce central nervous system effects in humans,
even in the absence of impaired renal function. Patients

71. Neurological Manifestations of R e n a l F a i l u r e

with either primary or secondary hyperparathyroidism

have EEG changes that are similar to those observed in
patients with acute renal failure. In uremic patients, both
the EEG changes and abnormalities of psychological testing are improved by either parathyroidectomy or medical
suppression of PTH.
In early studies, the sodium potassium adenosine triphosphate enzyme activity in crude brain preparations was said
to be normal to low in uremia. However, later studies with
metabolically active brain synaptosomes have shown that
both the sodium potassium adenosine triphosphate pump
and several of the calcium pumps are altered in uremia.
Alterations of the calcium pumps were thought to be due
at least in part to PTH acting through monophosphateindependent pathways. The uremic environment also
played a role as well. Since the calcium pumps at the nerve
terminals mediate neurotransmitter release and information transfer, these abnormalities most likely have the potential to affect information processing in the uremic state.

461
dying-back polyneuropathies or central peripheral axonopathies. Uremic neuropathy is also associated with a secondary demyelinating process in the posterior columns of
the spinal cord and the central nervous system. Motor and
sensory modalities are both generally affected and the
lower extremities are more severely involved than are the
upper extremities. Clinically, uremic neuropathy cannot
easily be distinguished from the neuropathies associated
with diabetes mellitus, chronic alcoholism, and other deficiency states. The occurrence of uremic neuropathy bears
no relationship to the type of the underlying kidney disease.
However, certain diseases which can lead to renal failure
may simultaneously affect peripheral nerve function separately from the manifestations of uremia. Such diseases
include amyloidosis, multiple myeloma, systemic lupus erythematosus, polyarteritis nodosa, diabetes mellitus, and hepatic failure.
Symptoms

PERIPHERAL NEUROPATHY

Two broad categories of peripheral neuropathy are associated with uremia. They are described in terms of their
pattern of neuronal involvement and include: (1) bilaterally
symmetrical disturbance of nerve function that can be des ignated as polyneuropathies (axonopathies). Polyneuropathy tends to be associated with toxic substances, metabolic
disorders (uremia, diabetes, deficiency states) and certain
immune reactions that diffusely affect the peripheral nervous system; and (2) isolated or multiple isolated lesions
of the peripheral nerves designated as mononeuropathies .
In severe symmetrical polyneuropathies, a generalized loss
of peripheral nerve function may occur, and the impairment
is usually maximal distally. This is characterized by a mixed
motor and sensory polyneuropathy with a distal distribution which often results in weakness and wasting in the
arms and legs. There may also be distal sensory changes
of "glove and stocking" distribution. The motor nerve conduction velocity is frequently used to assess peripheral neuropathy. However, the test is somewhat unreliable in uremic subjects, because the procedure itself has a normal
daily variation of up to 20%. Thus, motor nerve conduction
velocity has very limited utility in detecting moderately
impaired peripheral nerve function. Although sensory
nerve conduction velocity is more sensitive than motor
nerve conduction velocity, it is not widely employed because it is so painful.
Neuropathy of some degree is probably present in about
65% of patients with end stage renal disease who are undergoing dialysis. When the glomerular filtration rate is greater
than about 10% of normal, peripheral neuropathy is uncommon. Abnormal nerve conduction may be present in the
absence of symptoms or physical findings. However, many
patients who volunteer no complaints have impotence or
postural hypotension that may be detected with appropriate questioning or physical examination.
In summary, uremic neuropathy is a distal, symmetrical,
mixed polyneuropathy that belongs to a group known as

The restless-leg syndrome is a common early manifestation of chronic renal failure. Clinically, patients experience
sensations in their lower extremities such as crawling, prickling, and pruritus. The sensations are generally worse dis tally and are usually more prominent in the evening. The
burning-foot syndrome, which is present in less than 10%
of patients with chronic renal failure actually represents
swelling and tenderness of the distal lower extremities. The
physical signs of peripheral nerve dysfunction often begin
with loss of deep tendon reflexes, particularly in the knee
and ankle. There is also impaired vibratory sensation and
loss of sensation in the lower leg in the form of "stocking
glove" anesthesia. Sensory modalities which are lost include pain, light touch, vibration, and pressure.
Uremic Toxins as Causes of Neuropathy

A number of potential uremic toxins might lead to symp toms. Although any or all of these agents may play a role
in the development of uremic neuropathy, the evidence
that any of them are true "neurotoxins" is scant. For example, the "middle molecules" (MW 500 to 2500 Da), once
thought to be uremic toxins, have not with stood the test
of time. Several other potential "uremic toxins" have been
suggested. One of these is PTH, which has received much
of the attention. Suggestions that PTH may be a uremic
neurotoxin are based on correlation between plasma PTH
levels and motor nerve conduction velocity in patients and
animals with chronic renal failure. However, in patients
with hyperparathyroidism without uremia, there is no consistent observable effect of PTH on peripheral nerve function. Presently, no single uremic toxin has been shown
to affect peripheral nerve function. Most of the evidence
suggests that uremic neuropathy may be related to anatomical nerve damage of unknown etiology. It may also be
related to cumulative effects of multiple toxic agents, and
usually takes months to years to develop.

462

Cosmo L. Fraser
Autonomic and Cranial Nerve Dysfunction

Autonomic dysfunction is usually quite common in

chronic renal failure, and is usually associated with postural
hypotension, impaired sweating, impotence, gastrointestinal disturbances, abnormal Valsalva maneuver, and alterations in gastric motility. Dialysis -related hypotension is
often associated with autonomic insufficiency particularly
in patients with diabetes and amyloidosis. The autonomic
nervous system function can be evaluated by the following
means (1) the hand-grip dynamometer; (2) the heart rate
response following Valsalva maneuver to determine the
Valsalva ratio; and (3) the vascular response to norepinephrine infusion.
Cranial nerves are also often affected in uremic patients,
and the nerve involvement is most often manifested as
transient nystagmus, miosis, heterophoria, and facial asymmetry. Involvement of the eighth nerve affecting both auditory and vestibular divisions is most commonly seen. In
such cases, it is important to distinguish deafness due to
uremia from that associated with various hereditary interstitial nephropathies and ototoxicity of drugs such as the
aminoglycosides.
Treatment

A number of treatments have been tried to alleviate

symptoms of peripheral neuropathy. However, no one
treatment appears to be effective in everyone, possibly
because neuropathy in dialysis patients is of multifactorial
etiology. For instance, a number of toxins such as alcoholism, arsenic, lead, and PTH can produce neuropathy. Systemic diseases such as diabetes, amyloidosis, and multiple
myeloma can cause neuropathy independently of uremia.
Additionally, vitamin deficiency such as beriberi and B12
can also cause neuropathy. As a result, it is often quite
difficult to determine the precise etiology of the neuropathy. Nonetheless, a number of treatments have been tried,
including analgesics, anticonvulsants, antidepressants, and
antiarrythmics such as mexiletine which appears to have
some benefit in diabetic neuropathy. In very severe cases,
patients may only get relief from medical or surgical sympathectomy. There is presently no reliable evidence to suggest
that the symptoms of neuropathy can be ameliorated by
increasing the intensity of dialysis.
INTELLECTUAL FUNCTION IN DIALYSIS PATIENTS

Impairment of intellectual function is another frequently

recognized complication in patients with chronic renal failure who are being treated with dialysis. However, this abnormality is not well characterized and there is no anatomical lesion that characterizes intellectual deterioration.
Although some studies suggest that the overall intellectual
ability in patients with chronic renal failure may not differ
significantly from normal, others have suggested that the IQ
in dialysis patients is below that of the general population.
Based on psychological testing, the general consensus is
that over time, chronic renal failure is associated with
organic-like loss of intellectual function. Because patients

on dialysis are susceptible to other conditions that can cause

decline in intellectual function, it is often quite difficult to
establish a clear etiology for this problem.
COMPLICATIONS OF DIALYSIS THERAPY
Dialysis Disequilibrium Syndrome

Several central nervous system disorders may occur as

a consequence of dialytic therapy. One such disorder is
dialysis disequilibrium syndrome (DDS), which can occur
in patients who have recently initiated hemodialysis. The
symptom complex is quite variable and may include muscle
cramps, anorexia, restlessness, dizziness, headache, nausea,
emesis, blurred vision, muscular twitching, disorientation,
hypertension, tremors, and seizures. Although, it was originally thought that the EEG was abnormal in DDS, recent
studies have indicated that the EEG may indeed be normal.
DDS has been reported among patients of all age groups;
however, it occurs most often in the elderly and in children.
The syndrome is generally associated with rapid hemodialysis of patients who have just started hemodialysis, but it may
also occur following routine maintenance hemodialysis in
patients with chronic renal failure for as long as 6 months.
The pathogenesis of DDS has been extensively investigated in animal models. It was first postulated that DDS
was probably due to the "reverse urea effect," where it
was hypothesized that urea was cleared less rapidly from
the brain than from the blood, and this resulted in an
osmotic gradient between the blood and brain. It has since
been shown that during dialysis, urea is cleared from brain
at essentially the same rate as it is cleared from blood, so
there is no significant blood-to-brain osmotic gradient
based on the movement of urea alone. However, rapid
hemodialysis of animals with renal failure does result in an
increased brain-to-plasma osmolality often with resulting
cerebral edema. It is now thought that the increased brain
osmolality is due to the formation of organic acids, amino
acids, methylamines, and polyols which are produced in the
brain during the development of plasma hyperosmolality.
These compounds act as effective osmolar agents and are
called "idiogenic osmoles." Many of these solutes have
now been identified (myoinositol, sorbitol, betaine, glycerophosphorylcholine, phosphocreatine, glutamate, glutamine, taurine), and are found not only in uremia but also
in hyperglycemia and hypernatremia.
Idiogenic osmoles protect the brain from shrinkage during conditions of plasma hyperosmolality (uremia, hypernatremia, or hyperglycemia). In other words, the brain
increases its osmolality to match that of plasma without
losing water. A substantial decrease in brain volume could
tear the brain from the skull and produce intracerebral
hemorrhage. Thus, any treatment that acutely lowers
plasma osmolality without allowing adequate time for the
removal of idiogenic osmoles runs the risk of developing
substantial brain-to-plasma osmolar gradient and cerebral
edema. If the gradient is large enough (>25 mOsm/kg
H2O), the resulting cerebral edema may produce pressures
that could result in brain herniation (brain damage or

71. Neurological Manifestations of Renal Failure

death). The complications of DDS are usually avoided in

dialysis patients by decreasing both the length of dialysis
and the blood flow rate through the dialyzer. These maneuvers generally necessitate more frequent dialysis treatments. As the plasma osmolality is gradually brought back
closer to normal by this technique, idiogenic osmoles are
removed and brain osmolality gradually returns to normal
without any dramatic changes in brain volume.
Dialysis Dementia

Dialysis dementia is a progressive, frequently fatal neurologic disease which is seen almost exclusively in patients
who are being treated with chronic hemodialysis. Although
early studies focused on the distinctive neurological findings of this disorder, later reports have suggested that some
forms of dialysis dementia may be part of a multisystem
disease which includes dementia, osteomalacia, proximal
myopathy, and anemia.
Although the etiology of dialysis dementia remains controversial, it most likely represents a symptom complex
that is the final common pathway for a variety of proces ses. Dialysis dementia can be subdivided into three types:
(1) an epidemic form; (2) dementia associated with childhood renal disease; and (3) an endemic form. The initial
reports of dialysis dementia were all of the endemic form
that occurred in patients who had been on chronic hemodialysis for more than 2 years. Initial symptoms of this disorder are dysarthria, apraxia, slurring of speech with stuttering, and hesitancy. Later in the course of the disease,
symptoms progress to personality changes, psychosis, my oclonus, seizures, and eventually dementia and death
within 6 months after the onset of symptoms.
Early in dialysis dementia, the EEG shows multifocal
bursts of high amplitude delta activity with spikes and sharp
waves, intermixed with runs of more normal appearing
background activity. These EEG changes may precede
overt clinical symptoms by up to 6 months. However, as
the disease progresses, the normal background activity of
the EEG will deteriorate to a predominance of slow frequencies. When accompanied by the appropriate clinical
pictures, this EEG pattern is pathognomonic of dialysis
dementia; however, these patterns may also be seen in
other metabolic encephalopathies. Thus, the diagnosis of
dialysis dementia depends on the presence of the typical
clinical picture, the characteristic EEG findings, and, most
importantly, exclusion of other causes of central nervous
system dysfunction.
Finally, dialysis dementia has also been reported in children with renal failure, some of whom were neither on
dialysis nor exposed to aluminum-containing compounds.
Therefore, the cause of encephalopathy in such children
cannot be ascribed to aluminum alone, and may represent
developmental neurological defects resulting from exposure of the growing brain to uremia.
Role of Aluminum in Dialysis Dementia

Aluminum intoxication was probably first implicated in

this disorder when it was observed that it was markedly

463

elevated in brain gray matter in patients with dialysis dementia. Aluminum content in brain is also more than threefold greater in patients with dialysis dementia than in those
on chronic hemodialysis without dementia. The reasons
for the increase remain controversial although the actual
source of the aluminum is evident. Aluminum is the second
most common element in the earth's crust, and in the
United States the typical dietary intake of aluminum is 10
to 100 mg/day. Aluminum intake is substantially greater
in individuals taking aluminum-containing antacids. Normally, only a minimal amount of orally administered aluminum is absorbed, but for unknown reasons absorption appears to be increased in patients with renal failure. Prior to
the routine deionization of the water used in hemodialysis,
most of the aluminum in dialysis patients came from the
water. It is estimated that a dialysate aluminum concentration of only 70 /u-g/L results in a minimum positive aluminum balance of 300 ^g/day. Because aluminum is normally
excreted by the kidney, renal failure leads to aluminum
retention in such patients. Thus, the problem is not where
the aluminum comes from, but rather how it enters the
brain and how it causes dementia. In dialysis dementia,
abnormalities of the blood brain barrier most likely play
an important role.
Elevated brain aluminum levels are also present in several groups of patients who are not demented. These include patients with acute renal failure, hepatic encephalopathy, age greater than 60 years, and metastatic cancer.
However, the levels are generally lower than that observed
in dialysis dementia. In patients with dialysis dementia,
aluminum appears to be important in the development of
the epidemic form; however, whether aluminum plays an
important role in the other types of dialysis dementia (endemic, childhood) is still unresolved. Deionization of the
water used to prepare dialysate is now used as a preventive
measure to reduce aluminum concentration in dialysate
fluid. However, deionization not only removes aluminum
but also cadmium, mercury, lead, manganese, copper,
nickel, thallium, boron, and tin. Thus, not only aluminum,
but several other trace elements might be involved in the
pathogenesis of dialysis dementia.
Despite these unresolved questions, most outbreaks of
the epidemic form of dialysis encephalopathy have been
associated with high levels of aluminum in the dialysate.
Thus, therapy has been directed toward removing aluminum both from dialysate and from the patients. Lowering
the dialysate aluminum levels to below 20 /ug/L by deionization and reverse osmosis appears to be quite beneficial for
patients who are starting dialysis. In patients with overt
disease, eliminating the source of aluminum has only rarely
resulted in improvement in brain function. Diazepam or
clonazepam appear to be useful in controlling initial seizure
activity associated with the disease; however, they usually
become ineffective later on and do not appear to alter the
usually fatal outcome. Improvement in symptoms has been
reported in several patients treated with deferioxamine,
which chelates serum aluminum and reduces tissue burden
of aluminum. However, some studies suggest that deferioxamine without diazepam may lead to worsening of symp -

464
toms because it promotes aluminum removal from other
tissues but facilitates deposition in brain. Deferoxamine is
also associated with a number of side effects that markedly
limit it usefulness. These include fatal fungal mucormycosis
in approximately 10% of treated dialysis patients, cataracts,
retinal changes, and thrombocytopenia. Chelation therapy
only appears to be effective if started early in the course
of the disease. Thus, deferoxamine should be used early
in all patients suspected of aluminum intoxication.
The best dose and route of administration of deferoxa mine (DFO) is still in question. Symptomatic improvement
has occurred as rapidly in inpatients receiving 0.5 g as in
those receiving doses as high as 6 g/week. DFO can be
given once a week or three times per week with dialysis.
To avoid brain aluminum toxicity resulting from mobilization of aluminum from other sites; DFO should be given
intramuscularly the evening prior to dialysis, so that much
of the aluminum-DFO complex can be removed during
dialysis on the following day.
Acute neurotoxicity from aluminum intoxication can develop in uremic patients who have received oral preparations of citrate and aluminum compounds simultaneously.
It also appears that many nondialyzed children and adults
with renal failure who develop dialysis dementia do so as
a result of the oral administration of citrate solutions for
acidosis and aluminum hydroxide as phosphate binder. The
increased toxicity of aluminum under these circumstances
occurs because of enhanced gastrointestinal absorption of
aluminum when complexed with citrate. The elimination
of aluminum from dialysate water mainly prevents the occurrence of dialysis dementia (epidemic form); however,
removal of aluminum from patients with established dialysis dementia may not affect the course of their disease.
Treatment of sporadic cases in which the etiology of the
encephalopathy is unclear is more difficult. In this instance,
it is most important to differentiate dialysis dementia from

Cosmo L. Fraser

other metabolic encephalopathies and from structural le sions such as strokes and intracerebral hemorrhages.
Bibliography
Andreoli SP, Bergstein JM, Sherrard DJ: Aluminum intoxication
from aluminum containing phosphate binders in children with
azotemia not undergoing dialysis. N EngL J Med 310:10791084, 1984.
Arieff Al: Dialysis disequilibrium syndrome: Current concepts on
pathogenesis and prevention. Kidney Int 45:629-635, 1994.
Arieff Al, Massry SG, Barrientos A, Kleeman CR: Brain water
and electrolyte metabolism in uremia: Effects of slow and rapid
hemodialysis. Kidney Int 4:177-187, 1973.
Arnaud CD: Hyperparathyroidism and renal failure. Kidney Int
4:89-95, 1973.
Bakir AA, Hryhorczuk DO, Berman E, et al.: Acute fatal hyperaluminemic encephalopathy in undialyzed and recently dialyzed
uremin patients. Trans Am Soc Aetif Organs 32:171, 1986.
Fraser CL, Arieff Al: Nervous sy stem complications in uremia.
Ann Intern Med 109:143 -153, 1988.
Fraser CL, Arieff Al: Nervous system manifestations of renal
failure. In: Schrier RW, Gottschalk CW (eds) Diseases of the
Kidney, 5th. Little, Brown and Co., Boston, pp. 2625-2646,1997.
Fraser CL, Sarnacki P, Arieff Al: Abnormal sodium transport in
synaptosomes from brain of uremic rats. J Clin Invest 75:20142023, 1985.
Lien Y-HH, Shapiro JI, Chan L: Effects of hypernatremia on
organic brain osmoles. / Clin Invest 85:1427-1435, 1990.
Olivieri NF, Bungig RJ, Chew E, et al.: Visual and auditory neurotoxicity in patients receiving subcutaneous deferoxamine infusions. N EngL J Med 314:869-873, 1986.
Prior JC, Cameron EC, Knickerbocker WJ, Sweeney VP, Suchowersky O: Dialysis encephalopathy and osteomalacic bone disease. Am J Med 72:33-42, 1982.
Rotundo A, Nevins TE, Lipton M, Lockman LA, Mauer SM,
Michael AF. Progressive encephalopathy in children with
chronic renal insufficiency in infancy. Kidney Int 21:486-491,
1982.
Slatopolsky E, Martin K, Hruska K: Parathyroid hormone metabo lism and its potential as a uremic toxin. Am J Physiol 238;F1F12, 1980.

HEMATOLOGIC MANIFESTATIONS OF
RENAL FAILURE
JONATHAN HIMMELFARB

INTRODUCTION

The precise mechanisms by which hypoxia stimulates

renal erythropoietin secretion remain incompletely understood. Studies using a human hepatoma cell line suggest
that the oxygen-sensing mechanism involves the locking of
a heme protein in the deoxy conformation in response to
low oxygen tension.
In normal subjects, erythropoietin circulates in minute
quantities. In response to anemia, normal subjects have an
exponential increase in plasma erythropoietin levels. In
contrast, anemic patients with progressive renal failure
have circulating plasma levels of erythropoietin similar to
normal nonanemic subjects. In this patient population the
relationship between the degree of anemia and plasma
erythropoietin levels is also lost (Fig. 1).
The pathogenesis of anemia of chronic renal failure is
multifactorial. Erythropoietin deficiency, shortened erythrocyte survival, the presence of uremic inhibitors of erythropoiesis, bleeding, and iron deficiency have all been proposed as major factors. However, with the advent of
erythropoietin therapy, it has become clear that the principal cause of the anemia of chronic renal disease is a decrease in erythropoietin production due to the loss of functional renal tissue.
Support for the hypothesis that uremic toxins retained
in the plasma inhibit bone marrow erythroid production
comes from in vitro experiments where uremic human sera
has been shown to blunt the growth of erythroid colony
forming units (CFU-E). This in vivo erythroid inhibition
has been thought to be due to polyamines such as spermine
or putrescine as well as yet unidentified "middle molecules." Recent studies using human autologous cultured
bone marrow cells have not demonstrated inhibition of
erythroid progenitor cell growth by uremic serum. Furthermore, in vivo studies involving the infusion of recombinant
erythropoietin into normal subjects and hemodialysis patients demonstrate that erythropoiesis, as quantitated by
ferrokinetics and the reticulocyte response, are not blunted
in subjects with chronic renal failure. Almost all patients
with chronic renal failure, if iron replete and free of inflammation or infection, will have an appropriate erythropoietic response to erythropoietin. Thus, current evidence
suggests that uremic inhibitors of erythropoiesis play a
minimal role if any in the pathogenesis of the anemia of
chronic renal failure.

The development of progressive chronic renal failure

results in a number of adverse hematologic effects with
significant clinical consequences. Uremia almost invariably
results in profound changes in erythrocyte production and
destruction resulting in anemia. The development of recombinant human erythropoietin to treat the anemia of
renal disease has been one of the most significant advances
associated with renal replacement therapy in the past decade. While patients with chronic renal failure generally
have normal platelet and leukocyte counts, they often experience a qualitative change in platelet and granulocyte function. These changes are manifested clinically by an increased susceptibility to hemorrhage and infection. This
chapter reviews the pathogenesis of the anemia of chronic
renal failure, describe its clinical and laboratory manifestations, and provides an approach to its treatment. The pathogenesis, clinical manifestations, and treatment of platelet
and granulocyte dysfunction in uremia will also be described.
ANEMIA OF CHRONIC RENEL FAILURE
Pathogenesis

Normal erythropoiesis is primarily regulated by circulating erythropoietin. When erythropoietin binds to specific
receptors on bone marrow erythroid progenitor cells their
proliferation, differentiation and development into mature
erythrocytes is enhanced. Erythropoietin is now know to
be a heavily glycosylated protein with a MW 30,400 Da
comprised of 165 amino acids and four carbohydrate side
chains. The gene for erythropoietin encompasses about
3000 base pairs and is located on chromosome 7.
The major site of erythropoietin production is the renal
peritubular capillary endothelial cell. Renal interstitial fibroblasts may produce erythropoietin. The kidney produces up to 90% of circulating erythropoietin, accounting
for its pivotal role in erythropoiesis. Most extra renal erythropoietin is produced in the liver by centrilobular hepatocytes. However, extrarenal erythropoietin production is
rarely able to provide for sufficient erythropoiesis in an
anephric state, suggesting that the liver is less sensitive than
the kidney to hypoxic stimuli.

Primer on Kidney Diseases, Second Edition

465

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

466

Jonathan Himmelfarb
TABLE \ Causes of
Apparent Erythropoietin Resistance

Inadequate erythropoietin
Iron deficiency Infection or

'6

inflammation Osteitis
fibrosa cystica Aluminum

oi

intoxication Folate

Basal

deficiency Multiple
myeloma Severe
malnutrition Hemolysis or
blood loss

12

15

to4

10

101

Hemoglobin (g/dL)
FIGURE \ Erythropoietin levels in renal disease. The box entitled
"basal" refers to serum erythropoietin levels in normal individuals.
The upward arrow reflects the typical rise in serum erythropoietin
levels in patients with anemia not due to chronic renal failure.
From Hillman RS, Ault KA: Hematology in Clinical Practice.
McGraw-Hill, New York, 1995. With permission.

A decrease in erythrocyte survival has also been postulated to contribute to the anemia of chronic renal failure.
The etiology of decreased erythrocyte survival in uremic
patients is unclear but may be related to reduced resistance
to oxidant stress. Erythrocyte survival generally does not
improve with dialysis therapy. The degree of shortening of
erythrocyte survival in uremia is generally mild (erythrocyte life span 60 to 90 days in uremic patients vs 120 days
in normal individuals). There does not appear to be a correlation between the amount of erythropoietin required to
maintain a stable hematocrit and erythrocyte survival suggesting that decreased erythrocyte survival is only a minor
cause of the anemia of chronic renal disease. The use of
erythropoietin may also extend erythrocyte survival.
Iron deficiency limits the erythroid response to erythropoietin and can contribute to the anemia of chronic renal
failure. Patients may develop iron deficiency anemia because of occult gastrointestinal and other bleeding, the
effects of routine phlebotomy, or because of occult blood
losses in the dialysis tubing as part of the dialysis procedure.
Other factors which can contribute to the reduction of the
hematocrit level in some patients with chronic renal failure
are shown in Table 1.
Clinical and Laboratory Manifestations

The anemia of chronic renal disease is normocytic and

normochromic. Both the corrected reticulocyte count and
serum erythropoietin levels are inappropriately low when
compared to values seen in patients with similar degrees
of anemia without renal failure. The bone marrow examination is usually normal in appearance and is not usually
necessary for the hematologic evaluation. Measurement of
serum erythropoietin levels is generally not required. An
abnormal platelet count or white blood cell count may
reflect a more generalized disturbance of bone marrow

function. Erythrocyte indices, reticulocyte count, and iron

parameters are helpful in detecting many anemias which
are not due to erythropoietin deficiency. Microcytosis can
reflect iron deficiency, aluminum intoxication, and certain
hemogloblinopathies. Macrocytosis may be associated with
vitamin B12 or folate deficiency.
As renal failure progresses, there is an increased likelihood of developing anemia because diseased kidneys are
unable to produce sufficient quantities of erythropoietin.
Figure 2 indicates the relationship between the change in
hematocrit and the serum creatinine in more than 900 patients. While the mean hematocrit drops below 30 at a
serum creatinine of approximately 6 mg/dL, there is consid erable interpatient variation. There is even less correlation
between the BUN and hematocrit.
Many patients with polycystic kidney disease will have
a higher hemoglobin level and hematocrit than would otherwise be expected for a comparable level of renal dysfunction. In contrast, patients who have undergone bilateral
nephrectomies will generally have more severe anemia than
other patients on chronic dialysis therapy. Even when kidneys are devoid of excretory function, they usually retain
some capability for erythropoietin biosynthesis.
Patients with severe anemia display signs and symptoms
attributable to tissue hypoxia. However, many of the symp toms usually attributable to anemia may be clinically difficult to distinguish from symptoms related to uremia. For
most patients, symptoms develop when the hematocrit decreases to 30 or less. However, therapy is indicated at a
higher level of hematocrit if angina pectoris or other significant symptoms develop. Younger patients often will
adjust better to anemia. When anemia develops gradually,
an intraerythrocytic increase in 2,3-DPG occurs, which
shifts the oxygen dissociation curve to the right and allows
oxygen to be unloaded at the tissue level more readily.
Thus, many patients may not be as severely symptomatic
as might be expected from the degree of anemia. Additional
cardiovascular compensatory mechanisms may also develop over time, thereby limiting symptoms directly attributed to anemia.
The symptoms of anemia include weakness, fatigue, and
dyspnea, particularly with exertion. Decreased exercise tolerance is common and many patients experience difficulty
with concentration, attention span, and memory. Sexual

72. Hematologic Manifestations of Renal Failure

50

467

A2CA2 ABAC A A
AH 553A3HA A
A
AE2C3IAAA 3
A
AOCEI A =AA
A A
3LK?=:
33A
A

jtcr - : : A C E A 2 2 A
5vT--3a

40

30

1
20

A A

A B A

AAA
a
A
A
C
AESL5*1^3 AA A A
A
A AA
._
_ 3 3 AAAAAA3A
AA
A3
: 3 K C A 3 2 A A C r 3 AAAAA
B A A A A A A
A A33CC33E3A AACCA
A A B S A
AS
AA
J > A D A C A C S A C 3 C3ABAA 80 ABA
CAA3 3
A
A
3 E 3 3 D 3 A A C B ADBBA 3 A A3 3BA3AAABACA 3A
A A
C
A
OSS CE3A 3ACAAA 3 C B C 2 8 C A B S A A A
A AA A
SA
: & A B C A C A 3 3 D 3AC AEACA33 A AA AAB
A S C I =1: 3 S A 2 C > S a i t i 2 S ^ 3 C C p C E A H A D 3 A S C C C C A A C 3 2 A A A C E A 3 3
A
3EC0 A X-E 3 s S " ^ W r C f c B C -l A F C T A A i X 2 C A C r A H C 3 CCCCI3
3 AABC CA AA
S A 3
AACi=r3IC2AA A A > ^ C 3 5 S C F C A C E C i 3 8 r D 0 8 C i C S C C 3 3 A D 3 A A B B B A D B A A C
A
A
AS
rv r
~ - - --------"*"~
- V - ^ L _ r > ^ _ --------_ -------------------------------^ASADC C2ABCABEB A CC
A
A
- AE : J^AC2AC 3 A C A I I
'
l l ________ I
I I I ll
ll I I I !
CAC AAAA
B A AAAAAA
A
33 3 AAAAA A3 |
i \ \ ........Tiji
*> !. i II I
" i i l U I C T A A A C a A C ACCBCBAAABAB AAA
AA
C EC B ACAA3 Afl 3 SCSSBBBC
IN
I
,
..... II I I I M l M i l 3 B A B B A A A A A A
3CAA3 C
BAA A A33AB i "MM .......... I I i | " H i < i
T ^ * ^ B ^ ^ ' ^ ' 1 , ' B I 1 ^ 1 >.
A
AA
A 3 iAB
C A 3 A A A C A 3 S D A A C A A I F O C S F C C L - .,_, - -----------------------&AAA 3 ACAA
Afl S '
BCOAr^3CABOAaA^03E>fDD
C 2 3 F c 3 0 E A : 1. ________
A AA 3
A 3JAA A
ADBA DOCCC DA CCC06D3AC 03DEAErF3DCr D
AABBCSTfl
A A
A
D
A
BAAAABA A&A 3A C AfC3CACCACBBC3EABF AAAAACAfiBTaA C3A
?A A
AA
A ABAA A AAAAAA A A B B C BBAAOBC BA BACAC5CDBE AE AABACBB AC AA 3
AAAA
A
A A
A A BA A A ABAC' E 3 A C A 28BB S 3
AAAA A ABB 3 ABA B
A
A
f
A
BAAD BA
3 C AA B AS 0 CAB
AAAA
AAB
AAAA3A
A A
A AA A A
3AA AS
AA
AAABA
AA 3
AAAA*'
AA
A3BAAA
3AAA
B A A A B A
A
A
AAA3AAA
AA
AA
A
AAAAA
AA
A
A
A

20
CREATININE (mg/dL)
FIGURE 1 The change of hematocrit with creatinine in chronic renal failure. This figure
represents approximately 4000 data points obtained from 911 patients followed at the
Brigham & Women's outpatient renal clinic from 1977 to 1983. The 95% confidence limit
of the slope is shown around each line. Each letter represents the number of data points
at that value (e.g., A = 1 data point, B = 2 data points, etc.). From. Hakim RM, Lazarus
JM: Am J Kidney Dis XI(3):238-247, 1988. With permission.

dysfunction, cold intolerance, and anorexia may also result

from anemia.
The use of recombinant erythropoietin to treat the anemia of chronic renal failure has led to the recognition that
the contribution of anemia to the morbidity of chronic renal
failure is greater than formerly appreciated. The clinical
benefits of improved tissue oxygenation have been quantitated to include improvements in exercise capacity, central
nervous system function, and endocrine and cardiac function. Correction of the anemia of chronic renal failure has
resulted in marked clinical improvement in exercise capacity. Recent data from both the United States Medicare
ESRD Program as well as data from the Canadian Erythro poietin Study Group have demonstrated a reduction in
hospital readmission in patients with end stage renal dis ease treated with erythropoietin. Chronic anemia and consequent tissue hypoxia results in a compensatory increase
in cardiac contractility and heart rate in order to optimize
tissue oxygenation. Over time, the cardiac response to anemia results in the development of left ventricular hypertrophy. Ultimately the development of left ventricular hypertrophy results in both diastolic dysfunction and an
increased incidence of myocardial ischemia and cardiac
arrhythmias. An inverse relationship between the degree
of anemia and the extent of left ventricular end diastolic
diameter has been demonstrated. Correction of the anemia
of chronic renal failure with erythropoietin has generally
but not always resulted in an improvement in left ventricular hypertrophy. The use of erythropoietin has resulted in
a lower rate of hospitalization for myocardial infarction in

the United States Medicare population and a lower rate

of hospitalization for cardiac disease in general in the Canadian dialysis population. A recent study has identified the
degree of anemia as a risk factor for cardiovascular mortality in ESRD patients.
The use of erythropoietin to improve anemia in hemodialysis patients is also associated with improvement in cognitive function. This improvement has been quantitated both
by documenting normalization of abnormal EEGs as well
as improvement on neuropsychiatric testing. Normalization
of cerebral blood flow has also been demonstrated with
improvement in anemia. These improvements in cognitive
changes have translated into a lessening of depression and
improvements in quality of life indicators. Endocrine
changes may also occur. Sexual function has improved in
some men after an increase in hematocrit. Some but not
all studies have demonstrated an increase in serum testosterone levels after erythropoietin therapy. Serum prolactin
levels have decreas ed after erythropoietin in some but not
all male hemodialysis patients. Some amenorrheic women
experience a resumption in menstruation following the use
of erythropoietin. Anemia may also play a role in growth
retardation in pediatric patients. Impaired immune responsiveness and platelet function may also be partially related
to anemia in this patient population.
Treatment

The therapeutic modalities available today for treatment

of the anemia of renal disease include administration of

Jonathan Himmelfarb

468
recombinant erythropoietin, institution of renal replacement therapy including kidney transplantation, and packed
red blood cell transfusions. Prior to the advent of erythropoietin therapy, approximately 75% of chronic dialysis patients had hematocrit values less than 30 while 15 to 25%
required periodic red blood cell transfusions.
Phase I and II clinical trials demonstrated a clear-cut
relationship between the dose of erythropoietin and the
rate of rise in hematocrit (Fig. 3). Since these initial studies,
patient variability in response to erythropoietin has been
noted, and a bell-shaped distribution pattern of doses is
required to maintain a stable hematocrit in iron replete
stable dialysis patients. Erythropoietin response is dependent on dose, route, and frequency of parenteral adminis tration. Although intravenous erythropoietin has 100% bioavailability, most of the administered intravenous dose is
probably biologically ineffective once the relatively small
number of erythropoietin receptors on erythroid progenitor cells are saturated. Subcutaneous erythropoietin administration results in a slower increase in serum erythropoietin
levels with maintenance of a stable serum level for hours,
and despite incomplete absorption, most studies indicate
that there is on average a 20 to 40% reduction in erythropoietin requirements for subcutaneous compared to intravenous dosing. Erythropoietin can also be administered intraperitoneally, which may be preferable for children treated
with peritoneal dialysis.
It is currently recommended that the initial administration of erythropoietin for treatment of anemia in predialysis, peritoneal dialysis, and hemodialysis patients be subcutaneous at a dose of 80 to 120 U kg"1 week" 1 in two
to three divided doses t o achieve the target hematocrit.
4

10

12

14

16

Weeks of rHuEPO therapy

500 U/Kg
150U/Kg

15
FIGURE 3 The response in hematocrit to erythropoietin therapy.
Each line represents the response to the given dose of erythropoietin administered intravenously three times a week to hemodialysis
patients. From Eschbasch JW, Egrie JC, Downing MR, Brown
JK, Adamson JW: N EnglJ Med 316:73-78, 1987. With permission.

Children less than 5 years of age frequently require higher

doses than adults (300 U kg"1 week" 1). The site of
erythropoietin injection should be rotated with each administration.
The appropriate target hematocrit for patients with
chronic renal failure on recombinant erythropoietin therapy remains controversial. The target maintenance hematocrit for patients in the initial phase I to II clinical trial
was 35 to 40% and was 35% for the phase III multicenter
clinical trial. The target hematocrit recommended by the
FDA in 1989 was 30 to 33%, which was subsequently widened to a target hematocrit range of 30 to 36% in June
1994. The most recent data from the United States Renal
Data System (USRDS) 1996 Annual Data Report demonstrated that the mean hematocrit for erythropoietin treated
dialysis patients in the United States was 30.2%, and that
43% of patients had values less than 30%. Preliminary analysis of a recent study involving more than 1200 hemodialysis patients with documented heart disease found that in
the more than 600 patients randomized to have their hematocrit increased to normal (42 3%) there was an increase
in death or nonfatal myocardial infarctions compared to
patients maintained at a hematocrit of 30%. Until further
data are available, it seems prudent to exercise caution
before attempting to raise the hematocrit above 36%.
A crucial aspect of erythropoietin therapy is the maintenance of adequate iron stores for the production of new
erythrocytes. Normally three quarters of body iron is pres ent in circulating erythrocytes and one quarter exists as
storage iron primarily in the liver and bone marrow. Because the demands for iron by the erythroid marrow frequently exceed iron stores once erythropoietin therapy is
initiated, iron supplementation is essential to ensure adequate response. In the pre-erythropoietin era, advanced
renal failure and its associated hypoproliferative anemia
were frequently accompanied by excess storage iron. However, since the widespread use of erythropoietin, iron overload is now uncommon, whereas iron deficiency has become quite common.
Iron status in patients receiving erythropoietin is monitored by measuring both the serum ferritin and the percent
transferrin saturation. Serum ferritin and percent transferrin saturation are complementary tests because they measure different pools of body iron. The serum ferritin is
proportional to storage iron and is also an acute phase
reactant. It may be increas ed in the presence of acute or
chronic inflammation. Thus, a low serum ferritin is a specific
but not a sensitive marker for iron deficiency.
In contrast, the percent transferrin saturation (measured
as the serum iron divided by the total iron binding capacity
multiplied by 100) reflects iron that is readily available
for erythropoiesis. The transferrin molecule contains two
receptors for molecular iron transported from storage iron
sites to erythroid progenitor cells. The percent transferrin
saturation is decreased with either absolute or functional
iron deficiency. Transferrin saturation appears to be a better predictor of iron responsiveness than the serum ferritin
level. Other tests of iron status such as RBC ferritin or
zinc protoporphyrin do not appear to increase diagnostic

72. Hematologic Manifestations of Renal Failure

sensitivity or specificity. The definitive test for iron status,
bone marrow stainable iron, is invasive and expensive and
need not be used on a routine basis. For optimal responsiveness to erythropoietin, current recommendations are that
the serum ferritin should be maintained at greater than
300 ng/mL and the percent transferrin saturation at greater
than 30%.
Iron can be administered either as oral iron salts or
intravenously. When oral iron is used it should be given
as 200 mg of elemental iron per day in two to three divided
doses. For optimal absorption, oral iron should be given
more than 1 hour before meals or more than 2 hours after
meals or ingestion of phosphate binders. When oral iron
is insufficient to maintain adequate iron status, patients can
receive intravenous iron. As of 1996, iron dextran is the
only form of intravenous iron approved in the United States
although intravenous ferric gluconate and ferrous saccharate are available in other countries. Anaphylactic reactions
occur in 0.6% of intravenous iron dextran treated patients.
Prior to initiating intravenous iron dextran therapy, a one
time test dose of 25 mg must be given intravenously. However, anaphylactic reactions may occur in patients who have
received any number of previous iron dextran injections.
Iron dextran is frequently administered as a 100 mg dose
with each hemodialysis treatment, with a total administered
dose of 1000 mg. Some investigators now recommend that
intravenous iron dextran be administered in 50 to 100 mg
weekly doses continually in patients intolerant of oral iron
salts or in patients who cannot maintain adequate iron
status with oral iron.
True resistance to erythropoietin therapy for the treatment of chronic renal failure is rare (Table 1). In the largest
clinical trial with erythropoietin, more than 96% of ironreplete dialysis patients responded when erythropoietin
was given at 150 U/kg intravenously three times a week.
The median dose of erythropoietin needed to maintain the
target hematocrit was 75 U/kg intravenously three times
a week. The most common cause of a poor response to
erythropoietin is either an inadequate erythropoietin dose
or the presence of iron deficiency. Other causes of erythropoietin resistance include the presence of inflammation or
infection. Inflammation results in a deficient iron supply
by blocking the release of iron from the reticuloendothelial
system despite the presence of normal to increased iron
stores. Iron absorption is also reduced in acute inflammatory states. Bone marrow fibrosis from osteitis fibrosa cystica as a consequence of hyperparathyroidism can cause
true erythropoietin resistance. Hyperparathyroidism without osteitis fibrosa cystica does not blunt the effect of erythropoietin. The definitive diagnosis of osteitis fibrosa cystica
requires a bone biopsy. Aluminum intoxication may be
associated with a reduction in the response to erythropoietin. Aluminum intoxication can be suspected in the ironreplete patient who has a low erythrocyte mean corpuscular
volume and can be confirmed by measuring an increased
serum aluminum level after deferoxamine administration
or by positive staining for aluminum in a bone biopsy, (see
Chapter 69) Rarely, folate or vitamin B!2 deficiency can
be a cause erythropoietin resistance. Resistance to erythro -

469
poietin therapy has also been documented due to antiNform antibody-mediated hemolysis in patients with chronic
formaldehyde exposure. Patients with multiple myeloma
and those who are severely malnourished may also be resis tant to erythropoietin therapy. Occult bleeding needs to
be considered in erythropoietin-resistant patients.
Early experiences with erythropoietin therapy in hemo dialysis patients suggested that accelerated hypertension,
the development of seizures, and hyperkalemia were frequent accompaniments to erythropoietin therapy. However, recent clinical experience and the results of controlled
trials have demonstrated that the adverse effects of erythropoietin therapy are infrequent and generally manageable.
The most frequent adverse effect is aggravation of hypertension. Approximately 20 to 30% of erythropoietintreated predialysis or dialysis patients developed hypertension or an increase in blood pressure during treatment of
erythropoietin. An increase in blood pressure usually occurs in association with a rapid increase in hematocrit and/
or at higher doses of erythropoietin. The development or
worsening of hypertension is thought to be related to increased, vascular wall reactivity as well as hemodynamic
changes that occur as a result of increasing red cell mass.
Endothelial cells have now been demonstrated to possess
erythropoietin receptors, and infusion of erythropoietin in
vitro can result in release of endothelin and vasoconstriction. Worsening hypertension associated with erythro poietin therapy can be treated by initiating or increasing antihypertensive therapy, by intensifying ultrafiltration in dialysis
patients with evidence of volume expansion, or by reducing
erythropoietin dose.
The initial multicenter study with erythropoietin in the
United States noted a higher incidence of seizures during
the first 3 months of the study compared to historical controls. In subsequent studies, with the exception of patients
with hypertensive encephalopathy, there is no evidence for
an increased incidence of seizures in patients in whom
appropriate dose and titration recommendations for erythropoietin therapy are followed. A prior history of seizures
is not considered a contraindication to the use of erythro poietin. While serious hyperkalemia was observed during
the early clinical experience with erythropoietin, recent
studies have not confirmed this finding.
Whether the use of erythropoietin increases the likelihood of developing vascular access thrombosis in hemodialysis patients remains controversial. Most studies have not
demonstrated an increase in the rate of either native fistulae
or synthetic graft thrombosis when the hematocrit is kept
at or below 36%. An exception is the Canadian multicenter
trial which suggested that erythropoietin increased the rate
of thrombosis in PTFE grafts. Preliminary analysis of a
recent study in which hemodialysis patients with heart dis ease were randomized to a target hematocrit of 42 3%
did demonstrate a significant increase in thrombosis of both
native fistulae and synthetic grafts in the higher hematocrit group.
ERYTHROCYTOSIS IN RENAL DISORDERS

Erythrocytosis is occasionally seen in previously anemic

uremic patients who develop a renal cell carcinoma or

470

Jonathan Himmelfarb

large renal cyst, and is occasionally seen in patients with

polycystic kidney disease. Erythrocytosis has also been reported to occur in up to 17% of renal transplant recipients.
Recent observations indicate that angiotensin converting
enzyme inhibition corrects renal transplant erythrocytosis.
Treatment efficacy of angiotensin converting enzyme
(ACE) inhibitors for posttransplant erythrocytosis has
been demonstrated using captopril, enalapril, lisinopril, and
fosinopril, suggesting that angiotensin II production is
somehow related to the development of erythrocytosis.
Studies of erythropoiesis after withdrawal of enalapril in
posttransplant erythrocytosis have not demonstrated that
high plasma levels of erythro poietin contribute to the development of posttransplant erythrocytosis. Thus, the pathogenesis of posttransplant erythrocytosis and mechanism
underlying the beneficial effect of angiotensin converting
enzyme inhibition remain undetermined. Phosphodiesterase inhibitors such as theophylline can also correct posttransplant erythrocytosis.
Because erythrocytosis of any etiology enhances the risks
of thromboembolic disease, therapy for posttransplant
erythrocytosis is recommended. Standard therapy includes
discontinuation of cigarette smoking or diuretic use. While
serial phlebotomy has been the recognized standard of care,
most transplant centers now use low-dose ACE inhibitor
therapy to successfully manage posttransplant erythrocytosis.
PLATELET DYSFUNCTION IN UREMIA

In the uremia patient, a bleeding diathesis is a major

cause of morbidity and mortality. Both acute and chronic
renal failure are associated with an increased frequency of
bleeding complications, including gastrointestinal hemorrhage, bleeding from surgical sites, and mucocutaneous
bleeding. Clinical bleeding in patients with uremia is due
to an acquired qualitative platelet defect and has best been
correlated with a prolonged bleeding time. At present,
however, there is no single unifying pathogenetic mechanism to explain the acquired platelet dysfunction seen in
patients with uremia.
Uremic platelet dysfunction is considered to be a multifactorial disorder. Defects in in vitro platelet aggregability,
diminished thromboxane-A2 production, abnormal intracellular calcium mobilization, and increased intracellular
cAMP have all been described in uremic platelets. Several
studies have emphasized a defect in platelet adhesion to
vascular subendothelium in uremic patients. Platelet adhesion to vascular subendothelium is the initiating step in
forming a platelet plug and thus is critical to achieving
hemostasis. Platelet adhesiveness to subendothelium depends on the cooperative interactions of platelet glycoproteins with von Willebrand's factor in the vascular wall.
Several studies have identified abnormal platelet glycopro tein function in hemodialysis patients. There is conflicting
data as to whether patients with uremia have decreased
von Willebrand factor activity. Several recent studies have
emphasized that uremia may be associated with increased

release of both prostacyclin and nitric oxideboth of

which inhibit adhesion of platelets to endothelium.
The clinical bleeding tendency of patients with renal
failure is generally improved by dialysis. Therefore, uremic
toxins have been implicated as important mediators of the
bleeding diathesis. However, the search for specific uremic
toxins that interfere with primary hemostasis has not been
conclusive. Furthermore, the general experience has been
that dialysis only incompletely corrects the prolonged
bleeding time, and, in some cases, may actually transiently
worsen the bleeding diathesis.
The synthetic vasopressin derivative desmopressin
(DDAVP) has been shown to shorten the bleeding time
rapidly and improve clinical bleeding in uremia. The mechanism of action of DDAVP is thought to be related to
release of large multimeric von Willebrand's factor complexes from endothelial cells and platelets, although this
has not been conclusively demonstrated. A limitation to
DDAVP therapy is the development of tachyphylaxis after
two to three doses. Other therapeutic approaches have
included the administration of cryoprecipitate or conjugated estrogens (Table 2). As with DDAVP, the mechanism of action of these agents is thought to be related either
to release of large von Willebrand's factor multimers or to
endothelial functional changes. Unfortunately, all of these
agents have variable clinical efficacy in uremia.
In anemic patients, increasing the hematocrit to above
30, either via packed red blood cell transfusions or the use
of high dose erythropoietin, has improved the bleeding
time in most patients. While the primary pathogenetic
mechanism of increasing the hematocrit in improving the
bleeding time may be rheological, additional evidence suggests that increasing hemoglobin concentration may inactivate the platelet inhibiting effects of nitric oxide.
LEUKOCYTE DYSFUNCTION IN UREMIA

Infection remains a major cause of morbidity and mortality in the dialysis patient. Most infections in chronic hemo dialysis patients are due to common catalase-producing
bacteria rather than to opportunistic organisms. Vascular
access infections, particularly with staphylococcus species,
remain the leading source of serious infection in hemodialy sis patients. The pattern of infectious organisms in chronic
dialysis patients is similar to patients with chronic granulomatous disease who lack the phagocytic cell capability to
produce reactive oxygen species.
The number and morphology of granulocytes in patients
with uremia are normal except for a tendency toward hypersegmentation in some patients. Numerous studies in
patients with uremia or on chronic dialysis have documented alterations in granulocyte function, including defects in chemotaxis, granulocyte adherence, phagocytic capability, and reactive oxygen species production.
In addition to the changes in granulocyte function associated with uremia, it is now well documented that hemodialysis with unmodified cellulosic membranes results in
complement-mediated granulocyte activation and subsequent dysfunction. Changes in granulocyte function as a

7 2 . H e m a t o logic Manifestations of Renal Failure

471
TABLE 2
Treatment of Uremic Platelet Dysfunction

Therapy
Increase hematocrit (EPO,
RBC transfusion)
DDAVP
Cryoprecipitate

Conjugated estrogens

Dose or goal

Onset

Duration

>30

Immediate

Prolonged

Highly effective

0.3 /j,g/kg iv
10 U
0.6 mg kg" 1 day"1 x 5 days

Immediate

4 -8 h r

Rapid tachyphylaxis

1-4 hr

24 hr
14 days

Hepatitis, HIV risk

6hr

Comment

Hot flashes, HTN,

abnormal LFTs

Note: HTN, hypertension; LFT, liver function test.

consequence of dialysis with cellulosic membranes include

modulation of chemotactic receptors, changes in granulocyte reactive oxygen species production, alterations in granulocyte adherence and alterations in granulocyte expression of cell adhesion molecules. A recent cohort study
using the USRDS database has demonstrated a correlation
between the use of unmodified cellulosic dialysis membranes and the risk of mortality due to infection in chronic
hemodialysis patients.
Changes in monocyte function as well as a reduction in
number and function of lymphocytes in patients with uremia lead to altered immunity in this patient population.
Evidence of altered immunity includes alterations in T celldependent humoral responses such as reduced responsiveness to vaccinations, diminished delayed hypersensitivity
responses, and a dramatic attenuation of autoimmune dis ease activity with uremia. Recent data have suggested that
the chronic use of unmodified cellulosic dialysis membranes
may also exacerbate the underlying defect in immunity in
patients with uremia.
Bibliography
Beusterien KM, Nissenson AR, Port FK, Kelly M, Steinwald B,
Ware JE: The effects of recombinant human erythropoietin on
functional health and well- being in chronic dialysis patients.
J Am Soc Nephrol 7:763-773, 1996.
Cheung AK: Biocompatibility of hemodialysis membranes. J Am
Soc Nephrol 1:150-161, 1990.
Churchill DN, Muirhead N, Goldstein M, et al.: Effect of recombinant human erythropoietin on hospitalization of hemodialysis
patients. Clin Nephrol 43(3):184-188, 1995.
Erslev AJ: Erythropoietin. A/Eng//Merf324(19):1339-1343,1991.
Eschbach JW, Egrie JC, Downing MR, Browne JK, Adamson
JW: Correction of the anemia of end-stage renal disease with
recombinant human erythropoietin: Results of a combined
phase I and II clinical trial. N Engl J Med 316:73-78, 1987.

Eschbach JW, Kelly MR, Haley NR, Abels RI, Adam son JW:
Treatment of the anemia of progressive renal failure with recombinant human erythropoietin. N Engl J Med 321:158-163,1989.
Eschbach JW: The anemia of chronic renal failure: Pathophysiology and the effects of recombinant erythropoietin. Kidney Int
35:134-148, 1989.
Eschbach JW, Haley NR, Egrie JC, Adamson JW: A comparison
of the responses to recombinant human erythropoietin in normal
and uremic subjects. Kidney Int 42:407-416, 1992.
Fishbane S, Frei GL, Maesaka J: Reduction in recombinant human
erythropoietin doses by the use of chronic intravenous iron
supplementation. Am J Kidney Dis 26(l):41-46, 1995.
Fishbane S, Maesaka JK: Iron management in end-stage renal
disease. Am J Kidney Dis 29:319-333, 1997.
Foley RN, Parfrey PS, Harnett JD, Kent GM, Murray DC, Barre
PE: The impact of anemia on cardiomyopathy, morbidity, and
mortality in end-stage renal disease. Am J Kidney Dis 28(1 ):5361, 1996.
Goldblum SE, Reed WP: Host defenses and immunologic alterations associated with chronic hemodialysis. Ann Intern Med
93:597-613, 1980.
Hakim RM, Lazarus JM: Biochemical parameters in chronic renal
failure. Am J Kidney Dis XI(3):238-247, 1988.
Himmelfarb J, Hakim RM: Biocompatibility and risk of infection
in haemodialysis patients. Nephrol Dial Transplant 9(Suppl
2):138-144, 1994.
Lewis SL, Van Epps DE: Neutrophil and monocyte alterations in
chronic dialysis patients. Am J Kidney Dis 9:381-395, 1987.
Powe NR, Griffiths RI, Watson AJ, et al.: Effect of recombinant
erythropoietin on hospital admissions, readmissions, length of
stay, and costs of dialysis patients. / Am Soc Nephrol 4:14551465, 1994.
Ratcliffe PJ: Molecular biology of erythropoietin. Kidney Int
44:887-904, 1993.
Steiner RW, Coggins C, Carvalho CA: Bleeding time in uremia:
A useful test to assess clinical bleeding. Am J Hematol 7:107117, 1979
Vigano G, Remuzzi G: Bleeding time in uremia. Semin Dialysis
9(l):34-38, 1996.

ENDOCRINE MANIFESTATIONS OF
RENAL FAILURE
EUGENE C. KOVALIK

Chronic renal failure (CRF), end stage renal disease

(ESRD), and renal transplantation all have effects on the
endocrine system. The effects are related to alterations
in feedback mechanisms, hormone production, transport,
metabolism, and elimination, hormone binding, and drug
interactions (Tables 1 and 2). Diagnosis of endocrine dis ease in renal patients must take into account these effects
in order to ensure proper treatment, when needed. Alterations of bone and mineral metabolism and erythropoiesis
are discussed elsewhere and will not be covered in this
chapter. Through a mechanism that is not understood,
erythropoietin, ameliorates many of the endocrine abnormalities seen in CRF, as discussed below.
HYPOTHALAMIC-HYPOPHYSEAL AXES

an increase in T3 levels, but not in the expected increase

in T4. It has been postulated that two different problems
occur in patients with CRF. First, there is an inappropriate
response to decreased thyroid hormone levels due to a
reset in the normal feedback loop with a lower level of
TSH secretion for a given level of thyroid hormone. Second,
thyroid gland resistance to TSH also occurs. The use of
recombinant erythropoietin normalizes the TSH response
to TRH but not the thyroid response to TSH.
Studies have suggested that the impaired thyroid axis
may play a role in protecting the body by maintaining a
positive nitrogen balance in the setting of the uremic state.
After renal transplantation, thyroid function tests normalize, although some patients may still have an abnormal
TSH response to TRH due to glucocorticoid suppression
of TSH secretion.

Thyroid

Growth Hormone

Thyroid abnormalities have been well documented in

renal patients. Up to 58% of uremic patients may have
evidence of a goiter by palpation, thought to be due to some
unknown circulating goitrogen when glomerular filtration
rate (GFR) falls. Total thyroxine (T 4) is either normal or
decreased and triiodothyronine (T3) levels are depressed.
Free T4 levels may be low. The use of reverse T3 (rT3)
to differentiate between hypothyroid states and so-called
"euthyroid sick" state is not useful in patients with CRF.
Although binding globulin levels are normal, circulating
inhibitors result in decreased binding and interfere with the
older resin uptake-based tests of thyroid function. These
abnormalities tend to worsen the longer the patient is on
dialysis. Thyroid stimulating hormone (TSH) tends to be
normal despite the abnormalities in thyroid hormone levels, and is perhaps the best indicator of thyroid function,
especially with the availability of ultrasensitive TSH assays.
Basal metabolic rate is also normal. The differences in
thyroid function between hemodialysis and patients on continuous ambulatory hemodialysis are inconsistent.
Although TSH levels are normal in patients with CRF,
abnormalities exist in the hypothalamic-hypophyseal axes.
Exogenous thyrotropin stimulating hormone (TRH) stimu lation results in a blunted TSH response. The usual midnight TSH surge is absent. TSH administration results in

Plasma growth hormone (GH) levels are elevated in

patients with CRF. This is partly due to both increased
secretion and impaired clearance of the hormone, although
there seems to be no clinical significance. Insulin-like
growth factor (IGF) levels are normal. Dynamic testing of
the GH axis does show abnormalities. Oral glucose loading
does not suppress GH levels, whereas intravenous glucose
or glucagon or TRH paradoxically increase GH levels.
Insulin-induced hypoglycemia does not stimulate GH secretion. Growth hormone releasing hormone (GHRH) and
L-dopa infusions also cause a prolonged and exaggerated
response in GH secretion. Correction of anemia with erythropoietin corrects the paradoxical response to TRH and
insulin-induced hypoglycemia, although the prolonged
GHRH response remains.
In children, uremia results in growth retardation despite
normal or elevated GH and IGF levels. The cause is multifactorial and is thought to include protein malnutrition,
chronic acidosis, recurrent infections, hyperparathyroidism, and decreased bioactive IGF. The assurance of adequate nutrition, dialysis, and correction of acidosis and
hyperparathyroidism allows for improved growth. The use
of recombinant human growth hormone has greatly improved the well-being of children with CRF, restoring
growth velocity and increasing muscle mass without ad-

Primer on Kidney Diseases, Second Edition

472

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

473

73. Endocrine Manifestations of Renal Failure

TABLE I

TABLE 2

Pathogenetic Mechanisms of Endocrine Dysfunction in

Chronic Renal Failure"

Directional Changes of Hormones in CRF

Increased circulating hormone levels

Impaired renal or extrarenal clearance (e.g., insulin,
glucagon, PTH, calcitonin, prolactin) Increased secretion
(e.g., PTH, aldosterone?) Accumulation of immunoassayable
hormone fractions that
may lack bioactivity (e.g., glucagon, PTH, calcitonin,
prolactin)

Decreased circulating hormone levels

Decreased secretion by diseased kidney (e.g., erythropoietin,
renin, 1,25-dihydroxyvitamin D3) Decreased
secretion by other endocrine glands (e.g.,
testosterone, estrogen, progesterone)
Decreased sensitivity to hormones
Altered target tissue response (e.g., insulin, glucagon, 1,25dihydroxyvitamin D3, erythropoietin, PTH)
Note: PTH, parathyroid hormone.
" From Mooradian AD: In: Becker KL (ed) Principles and Practice of Endocrinology and Metabolism. Lippincott Company, Philadelphia, p. 1759, 1995. With permisssion

versely affecting bone age over chronological age. The fact

that renal transplantation in children does not by itself
reverse the abnormal growth patterns of the children is
likely due to the effects of glucocorticoids. European data
suggest return of growth after transplantation with the use
of recombinant human growth hormone, although its use
is not yet approved in the United States for pediatric patients post renal transplant.
Prolactin
Secretion normally is under inhibition via prolactin inhibitory factor (P1F), which in turn is controlled by catacholamines via dopaminergic fibers. Basal levels of prolactin
are elevated up to six times normal in patients with CRF.
Many medications that have antidopaminergic effects may
contribute to the increased levels. The major effects of
increased prolactin levels are reflected in the reproductive
system of CRF patients with gynecomastia, impotence, and
amenorrhea. Bromocriptine can reduce prolactin levels but
its effectiveness in relieving symptoms has not been well
proven, and side effects usually lead to discontinuation
of therapy in one third of patients. Again, erythropoietin
therapy can normalize prolactin levels and can improve
sexual dysfunction in men and menstrual regularity in
women, but it does not correct the abnormal hypophyseal
response to TRH administration.
Any medications that can affect prolactin levels should
be minimized or avoided if possible (i.e., a-methyldopa,
phenothiazines, neuroleptics, metaclopramide, cisapride
and H2-blockers, and especially cimetidine), particularly if
gynecomastia becomes painful or cosmetically displeasing
for male patients, even though the pathogenesis of gynecomastia may be more related to an increased estrogen to
androgen ratio as seen in pubertal gynecomastia in males.

Hypothalamopituitary axis
GH t prolactin f
Thyroid
T T 4 N or |, FT 4 N or | TT 3
|, FT 3 | , rT 3 N, TSH N

Gonads
Testosterone I, spermatogenesis |
Estrogen N or J, , progesterone J, L H
N or |, FSH N

Pancreas
Insulin 1, glucagon f

Adrenals
Aldosterone N or J,, cortisol N or f ,
ACTH N or t Catecholamine N or f
Kidney
EPO 1, renin J, , 1,25 Vit D3 J,
" From Lim VS: In: Greenberg A (ed) Primer on
Kidney Diseases. Academic Press, San Diego, p. 315,
1994. With permission.

Mammography should be performed since breast cancer

does also occur in men. Alternative therapies for gynecomastia include subcutaneous mastectomies or breast bud irradiation.
Glucocorticoids
Patients with CRF exhibit normal to elevated levels of
adrenocorticotropin hormone (ACTH). In various studies,
ACTH response to corticotropin releasing hormone
(CRH) has been observed to be blunted or normal. Correction of anemia with erythropoietin can lead to an exaggerated ACTH response to CRH. The standard ACTH stimu lation test for diagnosing hypocortisolism is not affected
by the uremic state.
Basal cortisol levels in CRF patients are normal although
some assays may show elevated results. The reason for
this discrepancy is that cortisol metabolites are primarily
removed by glomerular filtration; the accumulation of
glucoronidated metabolites can lead to cross-reactivity with
the antibody in some assay kits. Assays which extract metabolites prior to the radioimmunoassay or chromatography can overcome the problem and consultation
with the laboratory performing the test is important.
Circadian rhythm of cortisol secretion remains intact. Because of the elevated levels of metabolites (conjugated 17hydroxycorticosteroids), the negative feedback loop is
altered. The usual low dose 1 mg overnight or 2 day dexa methasone suppression tests used to evaluate hypercortisolism do not suppress patients with CRF due in part to
decreased oral absorption of the hormone and the altered
set point of the axis. The 1 mg intravenous or 8 mg high
dose overnight dexamethasone test will suppress cortisol
levels in CRF patients. Although insulin-induced hypogly-

474

Eugene C. Kovalik

cemia fails to raise serum cortisol levels, the response to

major stress, such as surgery, remains preserved.
Gonadotropins
Males

Loss of libido, impotence, testicular atrophy, gynecomastia, and infertility may occur in males with CRF. Testosterone is decreased, with elevated luteinizing hormone (LH)
and follicle stimulating hormone (FSH) levels. Testosterone-binding globulin levels are normal. Testicular biopsy
reveals abnormal sperm maturation. Prolonged stimulation
with human chorionic gonadotropin (HCG) can result in
increased testosterone levels, suggesting some testicular
reserve. Administration of luteinizing hormone releasing
hormone (LHRH) results in a normal, blunted, or an exaggerated, prolonged response. Thus it appears that both a
central hypothalamic and peripheral testicular problem exist in men with CRF. Treatment with erythropoietin can
improve symptomatology without actually affecting testosterone levels. Zinc deficiency has also been thought to
contribute to the problem, although replacement therapy
has yielded varying results. Transplantation reverses the
problems; however, in some men, the hypogonadism may
worsen due to the side effects of immunosuppressive
therapy.
Impotence may also be related to neuropathies or vasculopathies. Drugs that may contribute (such as /3-blockers)
should be discontinued or their dose reduced. Unfortunately, many patients may have to resort to penile implants,
cavernous injections (due to their vasoconstrictor effects,
these drugs should be used with caution in severely hypertensive patients), or vacuum erector devices to resolve their
problem. Table 3 outlines an approach to the male patient
once medications have been modified. If testosterone levels
are in the low normal range, a trial of replacement therapy
can be undertaken.
Female

Women with CRF may also have impaired libido or inability to achieve orgasms. Approximately half of women
on dialysis become amenorrheic, and those who still have
menses find their periods progressively irregular and anovulatory as renal failure progresses. Only 10% or less of
women on dialysis have regular menses. FSH levels are
normal with mildly elevated LH, resulting in an increased
LH: FSH ratio, similar to prepubertal patterns, with a defect in the positive hypothalamic feedback mechanism in
response to estrogen. This lack of positive feedback results
in the failure of the mid-cycle LH and FSH surge to occur
and hence, anovulation. Estradiol, estrone, progesterone,
and testosterone levels are normal to low. Unlike premenopausal women, postmenopausal women have the expected
increases in both LH and FSH. Hyperprolactinemia may
also contribute to some of the abnormalities. Despite these
problems, women who have some residual renal function
and who are well dialyzed can, on rare occasions, become
pregnant and carry to term, although the fetus tends to be
premature and small for gestational age (see Chapter 60).

TABLE 3

Management of Sexual Dysfunction in Patients with Chronic

Renal Insufficiency"

| Potency
I
I

J, Libido
I
Nocturnal penile tumescence
1

Normal
I

1. Psychological testing
2. Reassurance,
antidepressant, etc

Decreased
I

1. Neuropathy "|
2. Vascular
LPenile
insufficiency \ prosthesis
3. Zinc deficiency -zinc
supplementation
4. Assess hormonal profile
a. \ PRL-bromocriptine
b. Testosterone
c. ? Centrally acting
agents
Clomiphine
HCG
LHRH

Note: PRL, serum prolactin; HCG, human chorionic gonadotropin; LHRH, luteinizing hormone releasing hormone.
a
From Lim VS: Am J Kidney Dis 9(4):365, 1987. With permission.

The use of estrogen replacement therapy to decrease bone

loss and cardiovascular risk in premenopausal CRF women
has not been investigated.
Transplantation rapidly restores fertility to premenopausal women. Ovulation can start within 1 month of transplantation. Appropriate counseling should be undertaken
to stress the need for contraception. Current guidelines call
for women who wish to become pregnant to wait 2 years
with stable graft function (creatinine under 1.8 mg/dL), to
be on minimal immunosupressant therapy, and to have
easily controlled blood pressure. There are no good data
on the use of oral contraceptives in the transplant patient
group. Referral to a gynecologist should be made if this
mode of contraception is considered, due to the possibility
of thromboembolic disease. Blood pressure should be monitored carefully.
CARBOHYDRATE METABOLISM

Patients with CRF and uremia can develop what has

been termed "pseudodiabetes." This condition results from
a combination of peripheral resistance to insulin, circulating inhibitors, and abnormal islet cell release. The defect
can be corrected with dialysis therapy. In contrast, diabetics
with CRF often find that their need for oral hypoglycemic
agents or insulin decreases due to reduced insulin clearance
by the kidney. In the case of non-insulin-dependent diabetes mellitus (NIDDM), where the problem is mostly due
to peripheral insulin resistance, endogenous insulin half
life is prolonged, resulting in a decreased or eliminated
need for medication. In addition, the decreased clearance
of oral hypoglycemic agents (primarily first-generation sul-

73. Endocrine Manifestations of Renal Failure

fonylureas) can lead to prolonged hypoglycemia. Agents

that are primarily hepatically metabolized such as secondgeneration sulfonylureas or insulin, should be used to treat
NIDDM patients. Metformin is contraindicated in patients
with CRF due to the increased risk of lactic acidosis. Patients with insulin-dependent diabetes mellitus (IDDM)
may need their insulin dose reduced, but never discontinued since the underlying problem is a lack of endogenous
insulin production. Glycemic control can worsen in peritoneal dialysis patients due to the glucose load absorbed
from the peritoneal dialysis fluid. Intraperitoneal insulin
administration, based on a sliding scale, may facilitate management.
Spontaneous hypoglycemia can occur in CRF patients
due to malnutrition and impaired glycogenolysis. Although
many CRF patients have glucose intolerance, fasting hyperglycemia with glucose values above 140 mg/dL suggests
frank diabetes mellitus. Glycosylated hemoglobin values
may underestimate the degree of hyperglycemia due to the
shortened erythrocyte life span seen in uremia.
Transplantation often reveals an underlying abnormality
of glucose metabolism in many patients due to the high
doses of steroids used for immunosupression. Patients not
previously thought to be diabetic can develop diabetes
mellitus and require insulin therapy, whereas those already
diabetic may require conversion from an oral hypoglycemic
agent to insulin or large increases in their insulin doses due
to steroid-induced peripheral resistance to insulin action.
MINERALOCORTICOIDS

Due to the progressive loss of renal tissue with CRF,

renin production is decreased with either normal to low
aldosterone levels. The renin-angiotensin-aldosterone response to volume contraction or hypotension is blunted.
With low renin levels, hyperkalemia becomes the most
important stimulus for aldosterone secretion. Aldosterone
in turn stimulates colonic loss of potassium, an often overlooked means of potassium removal in CRF patients.
ADRENAL MEDULLA

Basal levels of catecholamines in CRF and dialysis patients are elevated. The cause is multi-factorial: decreased
degradation, decreased neuronal uptake, and impaired renal clearance of metabolites. Hemodialysis treatments do
remove catecholamines but this is not the reason for intradialytic hypotension. Because of the aforementioned problems, the diagnosis of pheochromocytoma is difficult in
patients with CRF. Suppression testing has not been validated in this setting. The combination of high levels of
catecholamines and appropriate radiologic studies should
be used together to make the diagnosis.
LIPID ABNORMALITIES

Patients with CRF infrequently have hypercholesterolemia or elevated levels of low-density lipoprotein (LDL),
but hypertriglyceridemia is observed in half of these pa-

475

tients. Low levels of high-density lipoprotein (HDL) are

also common. Nephrotic patients are an exception; they
have elevated levels of all lipid fractions. Heparin adminis tration during hemodialysis causes release of both hepatic
and lipoprotein lipase, resulting in a depletion of stores and
worsening of triglyceride levels, particularly of triglyceride
rich remnants that are highly atherogenic. Studies of lipoprotein a [Lp(a)j levels show variable results.
Until further data are available, treatment of lipid disorders in CRF and dialysis patients should follow general
guidelines used in non renal failure patients. Patients with
elevated LDL levels should be started on diet therapy and
then advanced to pharmacologic therapy. Bile acid resins
such as cholestyramine and cholestipol should be avoided
since they can increase TG levels. Fibric acid derivatives
such as gemfibrozil are relatively contraindicated in CRF
patients because they are cleared primarily by the kidney
and accumulation increases the risk of rhabdomyolysis. The
safest agents are the HMG-CoA reductase inhibitors. They
are particularly useful in nephrotic patients to decrease
LDL levels and reduce cardiovascular risk. High doses
may increase the risk of myalgias and rhabdomyolysis and
require close patient follow-up. Because of its many side
effects, nicotinic acid, is not a good lipid-lowering agent in
the CRF population, and experience with its use is limited.
Data on the use of probucol are lacking.
Recent studies have demonstrated the possible protective effects of vitamin E and folic acid in cardiac patients
with normal renal function. Studies also suggest that elevated LDL levels, particularly oxidized LDL, can hasten
the progression of CRF to ESRD. Since the mechanism of
action of these agents is either to serve as antioxidants (in
the case of the former) or to reduce homocysteine levels
(in the case of the latter), they should be safe to use in the
CRF population as a means to reduce cardiac risk and
theoretically to slow the progression of renal disease. Although vitamin C also acts as an antioxidant, its clearance
is reduced in CRF, because it can lead to hyperoxalemia or
hyperoxaluria and renal calcification, it should not be used.
Renal transplant patients should be considered separately from other CRF patients when it come to lipid abnormalities. Since cardiovascular disease is the major cause of
death in the transplant population (as is the case for dialysis
patients), every effort should be made to reduce risk factors. Total cholesterol, TG, LDL, oxidized LDL, and Lp(a)
levels are elevated, whereas HDL levels are variable. In
addition to the increased cardiac risk of post-transplant
lipid abnormalities, some evidence suggests that hyperlipidemia may also contribute to chronic graft rejection.
Treatment of lipid abnormalities in transplant patients
is similar to that of CRF patients. Bile resin binders should
be avoided because they interfere with cyclosporine absorption. Gemfibrozil use is associated with an increased
risk of rhabdomyolysis. HMG-CoA reductase inhibitors
(lovastatin, simvastatin, and pravastatin have been used
successfully in studies on transplant patients). But again,
higher doses should be used with caution. Simvastatin and
pravastatin have been employed in several studies, without
any major side effects. At least one study has shown a

476

reduction in acute rejection episodes using pravastatin. The

use of vitamin E, vitamin C, and folic acid has not been investigated.
Bibliography
Deck KA, Fischer B, Hillen H: Studies on cortisol metabolism
during hemodialysis. Eur J Clin Invest 9:203-207, 1979.
Ferraris JR, Domene HM, Escobar ME, et al.: Hormonal profile
in pubertal females with chronic renal failure: Before and under
haemodialysis and after renal transplantation. Acta Endocrinologica (Copenh) 115:289-296, 1987. Fine RN: Growth hormone
and the kidney: The use of recombinant
human growth hormone (rhGH) in growth-retarded children
with chronic renal insufficiency. Jo Am Soc Nephrol 1:1136
1145, 1991. Grundy SM: Management of hyperlipidemia of
kidney disease.
Kidney Int 37:847-853, 1990. Holdsworth S, Atkins RC,
Kretser DM: The pituitary-testicular
axis in men with chronic renal failure. N Engl J Med
296(22):1245-1249, 1977. Jones DC, Hayslett JP: Outcome of
pregnancy in women with
moderate or severe renal insufficiency. N Engl J Med
335(4):226-232, 1996. Katznelson S, Wilkinson AH,
Kobashigawa et al.: The effect of
pravastatin on acute rejection after kidney transplantationA
pilot study. Transplantation 61(10):1469-1474, 1996. Kokot F,
Wiecek A, Grzeszczak W, et al.: Influence of erythropoietin treatment on function of the pituitary-adrenal axis and somatotropin secretion in hemodialyzed patients. Clin Nephrol
33(5):241-246, 1990. Lim VS, Henriquez C, Sievertsen G, et
al.: Ovarian function in
chronic renal failure: Evidence suggesting hypothalamic anovulation. Ann Intern Med 93(Part l):21-27, 1980.

Eugene C. Kovalik
Lim VS, Flanigan MJ, Zavala DC, et al.: Protective adaptation of
low serum triiodothyronine in patients with chronic renal failure.
Kidney Int 28:541-549, 1985.
Lim VS: Reproductive function in patients with renal insufficiency.
Am J Kidney Dis 9(4):363-367, 1987.
Massey ZA, Kasiske BL: Post-transplant hyperlipidemia: Mechanisms and management. J Am Soc Nephrol 7:971-977, 1996.
Mooradian AD: Endocrine dysfunction due to renal disease. In:
Becker KL (ed) Principles and Practice of Endocrinology and
Metabolism. Lippincott, Philadelphia, pp. 1759-1762, 1995.
Ramirez G, Jubiz W, Gutch CF, et al.: Thyroid abnormalities in
renal failure: A study of 53 patients on chronic hemodialysis.
Ann Intern Med 79:500-504, 1973.
Ramirez G, O'Neill WM, Bloomer HA, et al.: Abnormalities in
the regulation of Growth Hormone in Chronic Renal Failure.
Arch Intern Med 138:267-271, 1978.
Ramirez G, Gomez- Sanchez C, Meikle WA, et al.: Evaluation of
the hypothalamic hypophyseal adrenal axis in patients receiving
long-term hemodialysis. Arch Intern Med 142:1448-1452, 1982.
Ramirez G, Butcher DE, Newton JL, et al.: Bromocriptine and
the hypothalamic hypophyseal function in patients with chronic
renal failure on chronic hemodialysis. Am J Kidne Dis 6(2):111 118, 1985.
Ramirez G: Abnormalities in the hypothalamic-hypophyseal axes
in patient s with chronic renal failure. Semin Dialysis 7(2):138 146, 1994.
Schaefer RM, Kokot F, Geiger H, et al.: Improved sexual function
in hemodialysis patients on recombinant erythropoietin: A possible role for prolactin. Clin Nephrol 31(l):l-5, 1989.
Spitalewitz S, Porush JG, Cattran D, et al.: Treatment of hyperlip idemia in the nephrotic syndrome: The effects of pravastatin
therapy. Am J Kidney Dis 22(1):143-150, 1993.

THE EVALUATION OF PROSPECTIVE RENAL

TRANSPLANT RECIPIENTS
BERTRAM L KASISKE

TIMING OF TRANSPLANTATION

showing SPK to reduce the long-term complications of

diabetes compared to kidney transplantation alone. Poor
blood glucose control remains the sole indication for SPK.
Patients should be informed that, compared to kidney
transplantation alone, there is increased surgical morbidity
and an increased risk of renal allograft rejection with SPK
compared to kidney transplantation alone. Most agree that
a diabetic with a potential living donor should undergo
kidney transplantation without a pancreas.

Once it is certain that a patient has end stage renal

disease (ESRD) and is likely to be a transplant candidate,
consideration should be given to the optimal timing of
transplantation. Although every attempt should be made
to maximize the benefit that patients receive from their
native kidneys, it is sometimes possible to perform preemp tive transplantation before initiating maintenance dialysis.
In such cases, the morbidity and expense of acquiring a
permanent dialysis access and undergoing dialysis initiation
may be avoided. Since the median waiting time for cadaveric kidneys in the United States is now 24 months, preemp tive transplantation is usually only possible for recipients
with living donors. Prior to acceptance for transplantation,
a careful evaluation should determine if the patient is prepared not only for immunosuppression and its consequences, but also for the surgical procedure itself.

SCREENING PROCEDURES
Cancer

It is generally accepted that immunosuppression may

favor the growth of malignant tumors, and an active malignancy is usually an absolute contraindication to transplantation. Exceptions may be locally invasive basal or squamous cell skin carcinomas. All patients should undergo
routine screening with a physical examination, chest radiograph, and stool hemoccult. Older individuals should also
have flexible sigmoidoscopy. Women should have had recent mammography, a pelvic examination, and a pap smear.
Although immunosuppression may increase the risk of
cancer recurrence, it is not justified to deny transplantation
to a patient who has had a tumor that was cured. Data
from registries, although imperfect, provide some guidance
on the chances of recurrence of different malignant tumors.
In general, registry data suggest that a patient who has
been free from all evidence of malignancy for 2 years has
about a 47% chance of recurrence after renal transplantation. Although a 5-year waiting period would reduce this
to 13%, it is probably unreasonable to demand that all
patients wait 5 years. Individual tumors behave differently,
and guidelines address how different tumors in potential
transplant candidates should be approached.

INITIAL ASSESSMENT

The evaluation should begin with a brief history and

physical examination so that obvious contraindications to
transplantation can obviate the need for other expensive
or invasive tests. The history and physical examination
should give the transplant team an indication of the overall
well-being of the recipient. There is no absolute age that
precludes transplantation, but physiologic age and overall
health status should be carefully considered in individuals
older than age 60. Patients should also be screened for
severe pulmonary disease, which might greatly increase the
risk of surgery. Obesity per se is rarely a contraindication
to transplantation, but obesity increases morbidity and may
increase the rate of graft failure. Therefore, weight reduction should be attempted prior to transplantation. Baseline
laboratory evaluation should include a complete blood
count with differential, routine serum chemistries, and a
lipoprotein profile.
Patients with diabetes require special consideration. Cardiovascular complications are particularly common among
diabetics, and may greatly increase the risk of transplant
surgery (see below). In addition, some diabetic patients
may be candidates for simultaneous pancreas -kidney (SPK)
transplantation. There are no controlled clinical trials

Primer on Kidney Diseases, Second Edition

Infections

Immunosuppression greatly increases the risk for lifethreatening infections. Immunizations for influenza, pneumococcus, and hepatitis B should be mandatory. Although
the effectiveness of these vaccinations is not well documented in large numbers of patients with ESRD, potential
477

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

478

Bertram L. Kasiske

benefits outweigh negligible risks. Patients should also be

carefully screened for infections that may become problematic with immunosuppression. Sites of occult infection include the lung, urinary tract, and dialysis access catheters.
Dialysis -related peritonitis within 3 to 4 weeks is a relative
contraindication to transplantation. Patients should be
screened for human immunodeficiency virus (HIV). Although occasional HIV patients may have a prolonged,
disease-free interval after transplantation, an even greater
number will die within 6 months of surgery. Therefore, the
current practice in the United States is to avoid transplantations in HIV-positive patients.
Tuberculosis is common in the ESRD population, and
may be asymptomatic. Screening should include a high
index of suspicion, a chest radiograph, and a purified protein derivative (PPD) skin test (unless there is already a
history of a positive skin test). High-risk individuals should
receive prophylactic therapy. High-risk individuals are
those: (1) with a past history of active disease; (2) from
a high-risk population, and (3) with an abnormal chest
radiograph consistent with active or inactive tuberculosis.
It is less clear whether any PPD-positive individuals who
have no evidence of tuberculosis or other risk factors should
receive prophylaxis.
Although cytomegalovirus (CMV) infection is common
and is commonly transmitted with the transplanted organ,
the presence or absence of CMV antibodies in donors and
recipients should not preclude transplantation. Some centers routinely use prophylactic therapy with hyperimmune
globulin, ganciclovir, or acyclovir for recipients of kidneys
from CMV-seropositive donors. Potential recipients who
are seronegative for the varicella-zoster virus are at risk
for disseminated infection, and should be identified before
transplantation. Patients from tropical regions should be
screened for Strongyloides stercolaris. Viral hepatitis
should also be considered in the pretransplant evaluation
(see below).
RECURRENT RENAL DISEASE RISK

Although renal disease frequently recurs in the transplant recipient, the threat of recurrent renal disease is rarely
a contraindication to transplantation. Estimated rates of
clinically evident renal disease recurrences are: systemic
lupus erythematosus (1%), hemolytic uremic syndrome (10
to 25%), Henoch-Schonlein purpura (10%), mixed essential
cryoglobulinemia (50%), Waldenstrom's macroglobulinemia (10 to 25%), light chain deposition disease and fibrillary glomerulonephritis (50%), amyloidosis (20 to 30%),
antiglomerular basement membrane disease (25%), Wegener's granulomatosis (rare), scleroderma (20%), IgA nephropathy (10%), membranoproliferative glomerulonephritis (MPGN) type 1 (20 to 30%), MPGN type 2 (80%),
membranous nephropathy (10 to 25%), and focal segmental
glomerulosclerosis (FSGS; 10 to 30%). Patients who have
already lost one allograft to recurrent FSGS have a 40 to
50% chance of losing another. Patients with primary oxalosis (once thought to be an absolute contraindication
to transplantation) should be considered for preemptive

transplantation with aggressive orthophosphate and pyridoxine therapy, and possibly liver transplantation to provide a source of the deficient enzyme. Although diabetic
nephropathy often recurs in the allograft, the rate of progression is generally slow enough to make this an unusual
cause of graft failure. One-year graft survival in patients
with sickle cell nephropathy has been reported to be as
low as 25% or as high as 67%. Finally, patients with Alport's
syndrome have a small chance of developing antiglomerular basement membrane disease after transplantation, but
this risk should not preclude transplantation.
LIVER DISEASE

Liver failure is a major cause of morbidity and mortality

after renal transplantation, and transplant candidates
should be carefully screened for liver disease. The A and
E viruses are not known to cause chronic liver disease in
transplant recipients, while hepatitis B virus (HBV) and
hepatitis C virus (HCV) are common causes of chronic
active hepatitis post-transplant. Transplant recipients who
are hepatitis B surface antigen (HBsAg) positive are at
increased risk of dying in the post-transplant period; however, HBsAg per se is not a contraindication to transplantation. Patients who are HBsAg positive and have serologic
evidence of viral replication should probably forego transplantation. Likewise, HBsAg positive patients who also
have hepatitis D (fortunately rare) develop severe liver
disease and should not receive a transplant. Otherwise,
HBsAg positive patients with elevated liver enzymes
should probably undergo biopsy, and those with chronic
active hepatitis may wish to stay on dialysis due to the
increased risk of disease progression with immunosuppres sion. Fortunately, the incidence of HBV is declining in the
ESRD population, mainly due to the use of an effective
vaccination. Certainly, patients who do not have liver dis ease and who are HBsAg negative should receive vaccination for HBV, if not already immunized.
Although the natural history of HCV liver disease is
less well defined, patients who are HCV antibody positive
should probably undergo liver biopsy if enzymes are elevated. HCV positive patients with evidence of viral replication and/or chro nic active hepatitis on biopsy are probably
at increased risk for progressive liver disease after transplantation. Antiviral therapy (interferon-ct) has been used
in an attempt to induce a remission in ESRD patients
with HBV or HCV viral hepatitis, and thereby allow renal
transplantation. Although the long-term results of antiviral
therapy in patients with ESRD are unclear, the number of
patients who remain in remission after such therapy appears to be small.
CARDIOVASCULAR DISEASE

Ischemic heart disease (IHD) is a major cause of death

after renal transplantation. All patients should be evaluated
for possible IHD as part of the routine history and physical
examination, chest radiography and electrocardiogram. Patients with a history of IHD, or with signs and symptoms

74. The Evaluation of Prospective Renal Transplant Recipients

suggestive of IHD, should undergo a more thorough evaluation with noninvasive stress testing and/or angiography.
Asymptomatic patients with multiple risk factors for IHD
should also have noninvasive stress testing. The best noninvasive stress test remains to be defined. However, thallium
scintigraphy (with exercise or dipyridamole) and echocardiography (with exercise or dobutamine) may have reasonably high positive and negative predictive values in ESRD
patientsa population which has a high prevalence of
IHD. Because many ESRD patients may be unable to
exercise to achieve the target heart rate needed for a valid
exercise stress test, pharmacologic stress testing may be a
better choice. Patients with critical coronary artery lesions
should undergo revascularization prior to transplantation.
All patients should be encouraged to reduce their risk of
cardiovascular disease by managing known risk factors. In
particular, patients should be encouraged to stop smoking.
CEREBROVASCULAR DISEASE

There is also an increased risk of atherosclerotic cerebral

vascular diseas e complications after renal transplantation.
Patients with a history of transient ischemic attacks or other
cerebrovascular disease events should be evaluated by a
neurologist for possible treatment and should be symptomfree for at least 6 rnonths before surgery. Whether asymp tomatic patients should undergo screening with a carotid
ultrasound examination is unclear. In the general population, controlled clinical trials have shown that the success
of prophylactic carotid endarterectomy is dependent on
the center and on the selection of patients. Similarly,
whether to treat patients who have asymptomatic carotid
bruits with aspirin and risk factor intervention, or to use
carotid ultrasound to select individuals for prophylactic
endarterectomy is a decision that must be tailored to the
patient and the center.
PSYCHOSOCIAL EVALUATION

Transplant candidates should be screened for cognitive

or psychological impairments that may interfere with their
ability to give informed consent. Failure to adhere to immu nosuppressive therapy is a major cause of renal allograft
failure, and the psychological assessment should also attempt to identify patients who are at risk. However, reliably
identifying patients who will not adhere to therapy is difficult at best, and care should be exercised to avoid unjustifiably refusing transplantation. Most centers require patients
with a history of substance abuse to undergo treatment
and to demonstrate a period of abstinence prior to transplantation. Major psychiatric disorders are usually apparent during the routine pretransplant evaluation, and appropriate psychiatric care can be sought.
UROLOGIC EVALUATION

Evidence for possible chronic infection and/or incomplete bladder emptying should be sought on history and
physical examination. In the absence of such evidence, a

479

routine voiding cystourethrogram is probably not warranted. For patients with a ureteral diversion, every effort
should be made to overcome the need for the diversion
posttransplant. This can often be achieved surgically,
but may sometimes require intermittent bladder selfcatheterization. Indications for pretransplant nephrectomy include: (1) reflux associated with chronic infection,
(2) polycystic kidneys that are too large to allow placement
of the allograft or have infected cysts, (3) severe nephrotic
syndrome, (4) nephrolithiasis associated with infection, and
(5) renal carcinoma.
GASTROINTESTINAL EVALUATION

Patients with a history of cholecystitis should undergo

cholecystectomy, because cholecystitis in an immunocompromised transplant recipient may be more severe, more
difficult to diagnose, and more difficult to treat. However,
it is less clear whether patients should be routinely screened
with ultrasound, and subjected to cholecystectomy for
asymptomatic cholelithiasis. Similarly, patients with symp tomatic diverticulitis may be considered for partial colectomy; however, whether it is justified to conduct screening
and surgery for asymptomatic diverticular disease is less
clear, and most centers do not. Peptic ulcer disease is common in the posttransplant period, but can usually be managed medically. Pretransplant evaluation for peptic ulcer
disease can probably be reserved for patients with appro priate signs and symptoms.
BLOOD AND TISSUE TYPING

The immunologic evaluation is designed to determine

the compatibility of a donor kidney with the recipient.
Three major immunologic barriers to transplantation are
addressed.
1. Transplants should not cross ABO blood group
barriers.
2. The degree of matching at the major
histocompatibility (MHC) loci A, B, and
particularly DR correlates with long-term graft
survival and is used in the United Network for
Organ Sharing (UNOS) point system for cadaveric
organ allocation (see below).
3. The presence of preformed antibodies and how
broadly they react to a random panel of antigens
from the general population is directly correlated to
the likelihood of a positive crossmatch when an
organ becomes available, and is also used in the
UNOS kidney allocation scheme.
As a final screen, the recipient's serum is tested against
donor tissue, since a positive cross-match indicates the pres ence of preformed antibodies that can cause hyperacute
rejection and is, therefore, a contraindication to transplantation.
ALLOCATION OF CADAVERIC ORGANS IN THE
UNITED STATES

Once a patient has been found to be suitable for transplantation and is ready for transplant surgery, he or she can

480

Bertram L Kasiske

be placed on the UNOS waiting list to receive a cadaveric

kidney. In general, a patient should not be placed on
the waiting list until the glomerular nitration rate (most
commonly estimated by creatinine clearance) is less than
15 mL/min. Kidneys are usually allocated based on the
UNOS point system (Table 1). Occasionally, UNOS may
allow a center to transplant kidneys outside of the UNOS
point system (usually for research purposes). The UNOS
point system is designed to balance equity with efficiency.
Equity demands that all patients be given the same access
to transplantable organs, regardless of race, ethnicity, gender, or socioeconomic status. Efficiency dictates that kidneys be given to the patients who are likely to benefit the
most, which is usually patients in whom the longest graft
survival can be expected. Unfortunately, the goals of equity
and efficiency often conflict. Although the UNOS point
system is designed to allocate organs, the final decision to
accept a particular organ once it is offered by UNOS rests
in the hands of the patient's physician.
Although the median waiting time for people listed in
1993 was 23.9 months (95% confidence interval 23.2 to 24.6
months), waiting time varies substantially according to a
patient's ABO blood group. For patients who were listed
in 1992 and were ABO blood group O, median waiting
time was 25.0 (24.1 to 25.9) months, for A it was 13.2 (12.8
to 13.8) months, for B it was 32.6 (30.6 to 35.3) months,
and for AB it was 6.2 (5.2 to 7.6) months. It would be
relatively disadvantageous for patients who are ABO blood
group O if A, B, and AB blood group patients could also
compete for O kidneys, so UNOS has established that all
ABO blood group O kidneys must go to O recipients (Table

TABLE I

Synopsis of the UNOS Point System for

Kidney Allocation
Criteria
Points
ABO blood group O kidneys to O recipients
NA
O antigen mismatch (mandatory sharing)
NA
Medical urgency (local arrangement only)
NA
Time waiting since listing
Relative (least to most)
0-1
Each year
1
Number of B or DR antigen mismatches
0
7
1
5
2
2
Panel reactive antibody level >80%
4
Pediatric age (yr):
<11
4
all and <18
3
Note: When information on a cadaveric kidney donor be comes available, all active patients on the UNOS list who have
an ABO blood type that is compatible with that of the donor
are assigned points as indicated in the table. Locally, the patient's physician can use "medical urgency" to give priority to
a recipient who has fewer points if there is only one transplant
center. If there is more than one center a cooperative medical
decision is required. NA, not applicable.

1). In the UNOS scheme, points awarded for waiting time

and points given for having preformed antibodies (preformed antibodies make it more difficult to find a crossmatch negative kidney) are designed to make the system
equitable (Table 1). On the other hand, points are also
assigned for how well matched a kidney is according to
MHC antigens, because grafts that are better matched are,
on average, more likely to survive longer.
Usually, a kidney is first offered locally, then if there is
no suitable donor, it is offered regionally, and thereafter
nationally. This scheme is designed to minimize the time
it takes to ship kidneys, because prolonged cold ischemia
time may delay graft function and possibly decrease graft
survival. On the other hand, the long -term survival of
0 antigen mismatched kidneys (there are 2 A, 2 B, and
2 DR MHC antigens) overrides concerns regarding increased cold ischemia time, and there is a mandatory policy
that a kidney must be shipped anywhere in the United
States if there is a cross-match negative, 0 antigen mis matched recipient. Thus, an 0 antigen mismatched recipient
from another region may receive a kidney even if a potential local recipient has more points. The only exception to
the 0 antigen mismatch rule is a recipient of a simultaneous
kidney and nonrenal organ transplantation. Under the
UNOS scheme, kidneys can voluntarily go to a recipient
of a simultaneous kidney and nonrenal organ transplant
with fewer points, although the receiving center must then
"pay the kidney back" at a future date. Thus patients receiving simultaneous organ transplants take precedent over
recipients of a kidney alone. Unfortunately, patients awaiting only a kidney may wait longer, not only because a
kidney that may have gone to them went instead to a
simultaneous nonrenal organ recipient, but also because
another kidney that may have gone to them was shipped
somewhere else as a payback.
MHC antigens influence waiting time. One reason why
the waiting time for African-Americans is longer than for
whites (median 29.6 vs 15.9 months) is that there are relatively fewer African-American donors combined with the
fact that African-Americans have a different frequency
distribution of MHC antigen types as compared with
whites. One solution to this particular problem, and to the
growing length of waiting times in general, is to increase
the rate of organ donation. Indeed, one of the medical
tragedies of our age is our increasingly frustrating inability
to offer transplantation to all suitab le recipients due to
the growing organ shortage. Although increasing organ
donation is one partial solution, this alone is probably not
enough, and many centers are placing a greater emphasis
on living donation.
LIVING DONORS

Kidneys from living donors generally survive longer than

cadaveric kidneys. The duration of graft survival based on
the source of the donor kidney is, on average, identical twin > 2-haplotype-matched sibling > 1-haplotypematched sibling or parent > 0-haplotype-matched sib ling = distantly related or unrelated (emotionally related)

74. The Evaluation of Prospective Renal Transplant Recipients

living donor > cadaveric kidney. Living donor kidneys have

the added advantages of allowing preemptive transplantation and making more kidneys available for individuals
who do not have suitable living donors.
Detailed guidelines have been developed for the evaluation of living donors. Potential living donors should be
counseled regarding both the short- and long-term risks of
donation. In a recent survey of transplant centers, mortality
from donation was estimated to be 0.03%, while major
morbidity was 0.23%. With regard to long-term risk, a recent meta-analysis of 48 studies including 3124 patients and
1703 controls found little evidence of progressive renal
dysfunction among normal individuals with only one kidney. Although there was a small increase in blood pressure,
this increase was not enough to raise the prevalence of
hypertension in patients who had a single kidney. Although
there was a statistically significant increase in proteinuria,
it was probably too small to be of clinical relevance.
In general, proteinuria greater than 200 mg/24 hr should
be considered a contraindication to donation. Microhematuria and pyuria should be investigated to rule out underlying renal disease that would preclude donation. Renal function should generally be normal, after adjusting for gender,
age, and possible dietary influences on glomerular filtration rate.
Blood typing and cross-matching are often the first steps
in evaluating a living donor. If a potential donor and recipient are blood group compatible and cross-match negative,
further evaluation can then be carried out. This should
include a psychological evaluation to ensure that the donation is truly voluntary and that the patient can give an
informed consent. A complete medical evaluation should
also be carried out to uncover conditions that would increase the risk of surgery. Potential donors should also be
screened for conditions such as diabetes or hypertension
that may worsen by having only one kidney. How diligently
to test for possible incipient diabetes in donors with a
positive family history or other risk factors for diabetes is
unclear. This is because the effect of having one kidney on
the rate of progression of diabetic nephropathy (if diabetes
occurred), is uncertain. Consideration should be given to
the risk of inherited renal diseases such as autosomal dominant polycystic kidney disease and hereditary nephritis.
Finally, the medical evaluation should ensure that the donor is free of diseases that could be transmitted with the
kidney, including malignancies, HIV, viral hepatitis, and tuberculosis.

481

If there is more than one potential living-related donor,

selection should be based on both medical and nonmedical
factors, and good matching need not be the only determinant of donor choice. Although the best donor is usually
a member of the recipient's immediate family, most centers
would consider an emotionally related donor. Once a potential donor has been selected and evaluated, the final
step is usually arteriography to define the renal vasculature
and to look for potential anatomic abnormalities.
TRANSPLANT PROCEDURE

The kidney is usually placed retroperitoneally in the iliac

fossa. The renal artery is usually anastomosed (end-toend) with the internal iliac artery. Different techniques
are available for dealing with multiple renal arteries and
atherosclerosis in the recipient's iliac artery. The renal vein
is usually anastomosed (end-to-end) with the external iliac
vein. The ureter is implanted via a long submucosal tunnel
into the bladder. In general, this surgical approach has
made the transplant procedure relatively routine; however
great care must be taken because immunosuppression delays wound healing and increases the risk of postoperative infection.
Bibliography
1995 Annual Report of the U.S. Scientific Registry for Transplant
Recipients and the Organ Procurement and Transplantation
Network - Transplant Data: 1988-1994. UNOS, Richmond, VA,
and the Division of Transplantation, Bureau of Health Resources Development, Health Resources and Services Administration. Rockville, MD, 1995.
American Diabetes Association. Pancreas transplantation for patients with diabetes mellitus. Diabetes Care 16 (Suppl 2):21,1993.
Bia MJ, Ramos EL, Danovitch GM, et al.: Evaluation of living
renal donors. The current practice of US transplant centers.
Transplantation 60:322-327, 1995
Kasiske BL, Ma JZ, Louis TA, Swan SK: Long-term effects of
reduced renal mass in humans. Kidney Int 48:814-819, 1995
Kasiske BL, Ramos EL, Gaston RS, et al.: The evaluation of renal
transplant candidates: Clinical practice guidelines. / Am Soc
Nephrol 6:1-34, 1995
Kasiske BL, Ravenscraft M, Ramos EL, Gaston RS, Bia MJ,
Danovitch GM: The evaluation of liv ing renal transplant donors:
Clinical practice guidelines. J Am Soc Nephrol 1996, 7:22882313.
Ramos EL, Kasiske BL, Alexander SR, et al.: The evaluation of
candidates for renal transplantation: The current practice of
U.S. transplant centers. Transplantation 57:490-497, 1994
United Network for Organ Sharing. UNOS Policies as of March
22, 1996, 1996 (UnPub).

RENAL TRANSPLANTATION:
IMMUNOSUPPRESSION AND
POSTOPERATIVE MANAGEMENT
DOUGLAS J.NORMAN

Renal transplantation is the preferred treatment for

failed native kidneys. In 1995, 11,807 kidney transplants
were performed in the United States. This is a small number
compared to the 31,045 patients on waiting lists at the 251
transplant centers. The success of kidney transplantation
has greatly lengthened the list of patients requesting to be
transplanted. Survival is improving steadily because of a
number of factors including the availability of potent drugs
that better suppress the immune response. Expected clinical outcomes for renal transplantation in 1997 are substantially better than those of 15 years ago, before cyclosporine,
OKT3, and globally improved patient care and donor management. Typical 1-year patient survivals are 97% and 93%
for living-related donor and cadaveric donor transplants,
respectively. One-year graft survivals are 91% for livingrelated donor transplants and 81% for cadaveric donor recipients.

the immune response. The longest graft survival, with a

half life of 22 years, is achieved with an HLA identical, or
two haplotype, match. HLA A, B, C, DP, DQ, and DR
alleles on both haplotypes of the donor match those of the
recipient. The next longest survival with a 12-year half life,
is with a one haplotype match. The worst survival, with a
5-year half life, is with an organ from a cadaveric, unrelated
donor with no matches for any of the six HLA A, B,
and DR antigens. Among unrelated donor-recipient pairs,
however, there is a step-wise increase in graft survival with
increased matching.
The goal of clinical immunogenetics testing is to ensure
that an adequate tissue match is achieved and that a pretransplant cross-match to rule out the presence of antidonor
antibodies in the serum of a prospective recipient is negative. While a negative cross-match is required, a good antigen match is not and, unfortunately, only 10 to 20% of
cadaver organs are transplanted into well-matched recipients.

RATIONALE FOR IMMUNOSUPPRESSION

CLASSIFICATION OF ALLOGRAFT REJECTION

To achieve these excellent survivals requires measures

to counter the immune response engendered against a
transplanted kidney. Antibodies produced by activated B
lymphocytes and cytolytic cells derived from T lymphocytes
are the key mediators of graft damage. B and T lymphocytes are dependent on accessory cells including tissue macrophages, monocytes, and endothelial cells, on cytokines,
including a variety of growth and differentiation factors,
and on adhesion molecules and receptors that promote
specific or nonspecific cell-to-cell contacts and interactions.
Immunosuppression strategies for organ transplantation
are increasingly being directed against specific elements of
the immune response cascade that develop in response to
an allograft.
A natural way to diminish the strength of the immune
reaction propagated against a kidney allograft is to match
the major histocompatibility antigens. Human lymphocyte
antigens, known as HLA, are coded by the short arm of
the sixth chromosome. Both class I (HLA A, B, C) and
class II (HLA DP, DQ, DR) antigens are major targets of

Primer on Kidney Diseases, Second Edition

Following transplantation of a kidney, episodes of renal

dysfunction can occur for a variety of reasons. The most
important of these is rejection. The three distinct syndromes of rejection have unique pathology, immunopathogenesis, and clinical presentations. Hyperacute rejection is
caused by preformed antidonor antibodies present in the
recipient's serum at the moment of transplantation. Hyperacute rejection leads to kidney failure almost immediately
after transplantation (within 24 hours). Allografts, which
must be removed from the recipient, demonstrate fibrin
thrombi in the glomerular capillaries and small vessels,
vasculitis, and necrosis. Acute rejections often occur in the
first 3 months after transplantation. They are cell mediated,
caused by T cells, via direct cytotoxicity, and by T cells and
macrophages, via local cytokine release. The pathological
appearance of acute rejection is a mononuclear interstitial
cell infiltrate and with tubulitis, lymphocytes invading renal
tubules, and sometimes with endothelial damage suggestive
of a vascular rejection. Glomeruli are spared. These rejec482

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

75. Renal Transplantation: Immunosuppression and Postoperative Management

tions are usually reversible by using a pulse of corticosteroids or anti-T cell antibody therapy. Chronic rejection,
the most common cause of late graft loss, is generally progressive and refractory to immunosuppressive therapy. Its
pathological appearance is a vasculopathy with intimal
thickening and occlusion leading to ischemia and subsequent interstitial fibrosis and glomerulosclerosis. Chronic
rejection is most likely triggered by endothelial cell damage
and is mediated by various components of the immune
system, that cause subsequent smooth muscle cell proliferation and fibrosis. Acute rejection episodes might begin this
process; their occurrence correlates directly with long-term
graft survival. A variant of chronic rejection, transplant
glomerulopathy, is characterized by glomerular capillary loop thickening and significant proteinuria. It too is
untreatable. However, when this condition is suspected,
it may still be useful to perform a renal biopsy to distinguish this entity from recurrent or de novo glomerulonephritis.
OTHER CAUSES OF ALLOGRAFT LOSS

Other causes besides true chronic rejections are now

known to contribute to the pathological process that is still
commonly termed chronic rejection. Thus, other factors
can contribute to chronic allograft failure. Hyperfiltration
damage occurs in the remnants of kidneys of laboratory
animals that have had one kidney and part of another
removed (see Chapter 66.) It is therefore assumed that
when the glomerular filtration rate of a renal allograft falls
below 20% of normal that hyperfiltration might be an important cause of further damage to the allograft and subsequent loss of kidney function. Any factor that might reduce
the number of functioning nephrons in a transplanted kidney can contribute to chronic allograft failure by shortening
the time to the onset of hyperfiltration-induced kidney
damage. Such factors include advanced age, female sex,
African-American ancestry, and smaH size of the donor.
Preexisting renal damage in the donor from hypertension,
lipids, drugs or other illness may also contribute, as can
complications occurring around the time of organ harvesting including the cause of death of the donor, hypotensive
episodes, cardiac arrest, and the use of nephrotoxic drugs
prior to the declaration of brain death. Still others include
any nonrejection damage that can supervene in an allograft
after transplantation due to hypertension, infections, nephrotoxic drugs, etc.
Some forms of glomerulonephritis, particularly focal glomerulosclerosis, may recur after retransplantation, and are
a relative contraindication to retransplantation. IgA nephropathy recurs in a high percentage of allografts but
rarely causes graft loss. Systemic lupus erythematosus may
also recur. Anti-GBM disease will recur in patients with a
high titer of anti-GBM antibodies. In general, glomerular
diseases that have a rapid and malignant course have the
highest potential for recurrence in an allograft.
GENERAL POSTTRANSPLANT CARE

Posttransplant in-hospital surgical management of a

transplant patient begins with care of the wound, drains,

483

urinary catheter, central venous line, gastrointestinal function, diet, and ambulation. Routine medical care includes
dialysis if the allograft does not function at once, fluidelectrolyte management, blood pressure control, infection
surveillance, renal function monitoring, and immunosuppression. The pharmacist and transplant coordinator play
important roles in teaching the patient about drug use, drug
interactions, and side effects, providing pivotal education
about the post transplant routine, and preparing the patient
for the outpatient follow-up. This team approach to transplant patient management combining physician, surgeon,
nurses, social worker and pharmacist, is key to the success
of transplantation.
Management of patients post transplantation requires
special attention to renal function, side effects of the immu nosuppressive drugs, and consequences of overimmunosuppression. Renal dysfunction episodes can occur due to
a variety of causes. A careful history and physical and
laboratory examination will often disclose the cause of
dysfunction, but more detailed testing, including a kidney
biopsy, renal scan, or sonogram may be required. The clas sic signs and symptoms of acute rejection include oliguria,
graft swelling and tenderness, fever, hypertension, graft
pain, anorexia, and progressive azotemia. With the the
recent use of stronger immunosuppressive drugs, these
finding seldom develop. When they do, only pyelonephritis
needs to be ruled out before treatment with corticosteroids
is begun. Rejection typically presents as only a minimal
rise in the serum creatinine, without accompanying signs
and symptoms. Other causes of renal dysfunction must be
considered, including cyclosporine or tacrolimus nephrotoxicity, urinary tract infection, hypersensitivity reaction to
a penicillin or other drug, volume depletion from diuretic
use or hyperglycemia, or other drug effect such as that of
trimethoprim or cimetidine to decrease creatinine secretion. If none of these can be documented by clinical or
routine laboratory findings, anatomical abnormalities such
as obstruction, urinary leak, and renal artery or vein thrombosis must be considered. The radionucleide renal scan is
probably the most useful single diagnostic test because it
can demonstrate each of these abnormalities. If other studies have not provided a diagnosis, a renal allograft biopsy
is the definitive diagnostic tool. As the biopsy is generally
performed under ultrasound guidance, additional information about obstruction and presence of an ext rarenal fluid
collection, such as a lymphocele, hematoma, or urinoma
due to a urine leak is also provided. Some centers use fine
needle aspiration of the allograft with cytologic examination in lieu of biopsies.
IMMUNOSUPPRESSIVE DRUGS
Immunosuppression is the cornerstone of a successful
kidney transplantation. The evolution of immunosuppres sion during the past decade has been extraordinarily brisk
and will likely continue at an accelerated pace during the
next 10 years. Immunosuppressive drugs can be divided
into a few broad categories. The monoclonal and polyclonal
antibodies and other biological agents are antitarget recog-

484

Douglas J. Norman

nition drugs that prevent cell-cell interaction and antigen

recognition. Cyclosporine, cyclosporine Neoral, and tacrolimus are anticytokine synthesis drugs. They block IL-2
synthesis by preventing transcription of the IL-2 gene. Key
steps are the binding of cyclosporine or tacrolimus to a
specific immunophilin (cyclophilin or FK binding protein
respectively), blockade of calcineurin (a phosphatase that
promotes the activation and movement of transcription
factor NFAT), and prevention of the transcription of cytokine genes, specifically IL-2. Rapamycin is the only known
example of an anticytokine action drug. It prevents IL-2
driven cell activation via an unknown mechanism. Rapamycin binds to FK binding protein and antagonizes the
effect of tacrolimus. Azathioprine, cyclophosphamide, my cophenolate mofetil, methotrexate, and brequinar sodium
are anti-cell proliferation drugs. They block enzymes vital
to purine or pyrimidine production and therefore to DNA
replication. Finally, corticosteroids are both antiinflammatory and immunosuppressive. As a rule, immunosuppres sive drugs are given in combination. Typically, an anticytokine synthesis drug (cyclosporine or tacrolimus), an
antiproliferation drug (azathioprine or mycophenolate mo fetil), and an antiinflammatory drug (corticosteroids) will
be employed together. Whether such triple therapy is better
than double therapy with a corticosteroid and anticytokine
drug alone is uncertain despite the theoretical benefit of
using agents with complementary modes of action. Different combinations of these drugs are currently being evaluated.
PHASES OF IMMUNOSUPPRESSION

Three distinct phases of immunosuppression follow kidney transplantation: (1) immunosuppression induction,
(2) immunosuppression maintenance, and (3) antirejection
treatment. Immunosuppression induction requires drugs
that are powerful yet specific for cells that initiate and
effect the allograft-directed immune response. The immune
response is strongest when it encounters the allograft initially. Passenger leukocytes from the graft flow to regional
lymph nodes and directly activate the immune response.
Strong immunosuppression is required during this period,
which is unfortunately also the time at which a patient is
at greatest risk for bacterial wound, lung, and urinary tract
infections. Therefore, drugs that promote delayed wound
healing or broadly affect all leukocyte function are potentially dangerous. Moreover, during this period, exposure
to potentially nephrotoxic drugs like cyclosporine or tacrolimus, which can exacerbate preservation-induced organ
damage, must be limited or avoided. The antilymphocyte
antibodies, in combination with corticosteroids, an antiproliferative drug, and the delayed use of an cytokine inhibitor
drug, are widely used for induction. Several drugs are available. OKT3 is a murine, monoclonal anti-CD3 (pan T cell)
antibody. Atgam is an equine, polyclonal, antithymocyte
antibody. Thymoglobulin is a rabbit, polyclonal, antithymo cyte globulin. All have been proven effective for induction,
if given for 7 to 14 days. They prevent early rejection
episodes that would otherwise require high doses of corti-

costeroids, and also improve long-term graft survival. Induction of immunosuppression is accomplished at some
centers with an intravenous cytokine inhibitor and corticosteroids, with or without an antiproliferative drug. Although this approach may be successful, it is associated
with a higher incidence of early rejection as compared with
antilymphocyte antibody induction.
Immunosuppression maintenance generally requires
lower doses of the same drugs, except that antilymphocyte
antibodies are never used for maintenance. The use of
lower doses of the drugs is possible because the body adapts
to the organ after the donor passenger leukocytes have
been destroyed or dispersed. The maintenance phase begins after 1 to 3 months and extends for life. Immunosuppressive drugs can never be discontinued because of a very
high risk of rejection without them, even years after transplant.
The third phase of immunosuppression is antirejection
treatment. Acute cell-mediated rejections occur most frequently during the first 3 months after transplantation and
after major sensitizing events such as infections, trauma,
surgery, or noncompliance with medications. First-line
treatment of acute rejections is usually with an increased
dose of corticosteroids given either intravenously or orally.
Treatment of steroid-resistant or steroid-dependent rejections requires one of the antilymphocyte antibody preparations. These are generally effective for 75 to 80% of steroidresistant rejections when administered for 7 to 14 days.
Some of the drugs used for maintenance have also been
used for treating both mild rejections and steroid-resistant
or recurrent rejections. Mild rejections have been treated
by increasing the doses of cyclosporine and or azathioprine
that a patient is already taking. Resistant rejections have
been treated by switching from azathioprine to mycophenolate mofetil or tacrolimus.
Standard doses of the most commonly used drugs are as
follows: azathioprine (1 to 2 mg kg" 1 day" 1 once a day);
mycophenolate mofetil (Ig bid); cyclosporine (4 to 8 mg
kg"1 day"1 in one or two divided doses); tacrolimus
(0.15 mg kg"1 day'1 in two divided doses), and prednisone
(0.10 mg kg"1 day" 1 or alternatively 0.2 mg/kg every
other day for maintenance). Cyclosporine and tacrolimus
therapy are monitored with drug levels. These desired levels vary according to assay used and frequency of dosing.
SIDE EFFECTS OF IMMUNOSUPPRESSIVE DRUGS

The drugs required for renal transplant patients have a

variety of side effects, as summarized in Table 1. Corticosteroids have the most deleterious long-term effects, including
osteopenia and osteonecrosis neither of which has a satis factory treatment. Steroids also cause a variety of cosmetic
problems, including hirsutism, acne, and fat deposits in the
face ("moon face"), neck (buffalo hump), and abdomen.
In children, growth retardation can be a serious problem.
Cyclosporine causes nephrotoxicity almost universally, although its severity is variable and dose related. Hepatotoxicity also occurs rarely and much more commonly with
the intravenous use of cyclosporine. Hirsutism can be a

75. Renal Transplantation: Immunosuppression and Postoperative Management

TABLE I

Individual Drug Toxicities

Corticosteroids Suppress the adrenal gland, decrease intestinal
absorption, and increase renal excretion of
calcium

Weight gain and redistribution of fat

to the abdomen, face, and neck
Osteoporosis, osteonecrosis
Myopathy
Delayed wound healing
Hyperlipidemia accelerated
atherosclerosis
Diabetes mellitus
Cataracts

Cyclosporine

Nephrotoxicity
Hypertension
Hirsutism
Gingival hyperplasia
Tremors, dysesthesias, seizures
Hepatotoxicity

Azathioprine Myelosuppression
Pancreatitis, gastrointestinal hypersensitivity
hepatotoxicity, venoocclusive
disease of the liver
Alopecia
Tacrolimus Nephrotoxicity Tremors, dysesthesias
Gastrointestinal hypersensitivity Diabetes
mellitus
Mycophenolate Mofetil Myelosuppression Gastrointestinal
hypersensitivity
OKT3 Cytokine release syndrome with fever, chills, nausea,
diarrhea encephalopathy, pulmonary edema and
nephropathy
Human antimouse antibody
production (HAMA)
ATG (polyclonal)

Serum sickness
Thrombocytopenia, leukopenia Mild
cytokine release syndrome

problem among dark-haired individuals, and gum hypertrophy is occasionally severe enough to require surgical resection. Neurotoxicity manifests as tremor or paresthesias, is
usually dose related, and can be controlled. Cyclosporineinduced hypertension and hyperuricemia with gout are
common. Tacrolimus also causes nephrotoxicity and neuro toxicity, perhaps more severe than that due to cyclosporine.
One of the most serious side effects of tacrolimus is insulindependent diabetes. Fortunately, this is dose related. However, in some cases it has not been reversible. Tacrolimus
causes diarrhea in some patients. In contrast to cyclosporine, tacrolimus does not cause hirsutism, gingival hyperplasia, or hypercholesterolemia. Therefore, it is an excellent alternative for patients who develop any of these
problems. Azathioprine causes myelosuppression, including leukopenia and anemia, which can also occur as an
isolated effect. Hepatotoxicity occurs rarely, as does pan-

485
creatitis. The latter carries a high mortality in immunosuppressed patients, whether azathioprine associated or not.
A severe form of hepatotoxicity, veno-occlusive disease, is
generally progressive even if azathioprine is discontinued.
Hair loss can occur, but it is reversible and does not preclude the reuse, or even continued use of azathioprine.
Mycophenolate mofetil causes myelosuppression and gas trointestial toxicity.
OKT3 causes the "cytokine release syndrome" which is
probably mediated by IL-2, tumor necrosis factor, interferon-y, IL-6, other cytokines, and possibly complement
activation products. Its features include fever, chills, nausea, diarrhea, headache, and photophobia. Pulmonary
edema, due in part to a "capillary leak syndrome" and
exacerbated by concurrent volume overload, hypotension,
and encephalopathy are rare. Human antimouse antibody
(KAMA) production occurs in approximately 75% of patients receiving OKT3, and occas ionally (10 to 20%) prevents subsequent use of OKT3 when present in a liter
greater than 1:1000. The polyclonal antilymphocyte antibody preparations are rarely associated with a cytokine
release syndrome. Atgam, however, causes thrombocytopenia, which occasionally requires dosage adjustment. Approximately 75% of patients make antibodies to horse (Atgam) or rabbit (Thymoglobulin) antibodies and these
rarely can cause a serum sickness syndrome with a petechial
skin rash and fever. Table 2 lists the organ-specific abnormalities that occur in some transplant patients as a result
of the drugs used for immunosuppression. When any of
these occur the transplant physician must consider a drug
toxicity as a cause.
POSTTRANSPLANT INFECTION

Infection may result from overimmunosuppression. In

the immediate postoperative period, the most common infections are bacterial urinary tract, pulmonary, wound, and
intravenous line-related infections. These can be avoided,
or mitigated by careful anticipation and use of preventive
measures. Opportunistic infections can occur at any time
after transplantation, but generally they develop when immunosuppression is most intense, such as during the first
few months after transplant and following treated episodes
of rejection. Herp es simplex is commonly activated by immunosuppression but can be prevented by acyclovir prophylaxis. Cytomegalovirus (CMV) is a particular problem
when a CMV seronegative recipient receives an organ from
a seropositive donor. Approximately 80% of such patients
will develop CMV disease ranging from fever alone to
pneumonitis, hepatitis, and enteritis. Reactivation CMV
disease also occasionally occurs in seropositive recipients
with prior CMV infection but is generally much less severe
than the primary form. It may be difficult to distinguish
CMV shedding in the urine, saliva, or bronchoalveolar
lavage fluid from true infection. CMV is most likely to
occur between 1 and 6 months posttransplant. Treatment
with ganciclovir is usually very effective and patients rarely
die of CMV disease, although significant morbidity and
expense can result. Active CMV may present during a

486

Douglas J. Norman

TABLE 2

General

Organ-Specific Drug Toxicity

Increased susceptibility to infections
and malignancies Weight gain

Integument

Cancer
Acne

Fungal infection
Warts
Bruising
Hirsutism
Alopecia
Musculoskeletal System Osteopenia
Osteonecrosis
Myopathy Pulmonary
Respiratory System
edema
Carciovascular System

Accelerated atherosclerosis
Hypertension

Hemodynamic System

Myelosuppression
Anemia
Thrombocytopc nia
Thrombocytosis
Hemolysis

Digestive System Peptic ulcer disease Diarrhea

Pancreatitis Hepatotoxicity
Urinary Tract

Sexual Function

Nephrotoxicity
Pyelonephritis Cystitis
Impotence Loss of
libido
Diabetes mellitus
Dysesthesias

Endocrine System
Nervous System

Euphoria
Depression
Encephalopathy
Seizures
Tremors
Cataracts

Eyes

rejection; reduction of immunosuppression during active

disease may promote rejection. Various preventive measures such as the use of high-dose acyclovir, CMV immune
globulin, and ganciclovir are available but the only treatment that has proven to be highly successful is ganciclovir
given for 3 months after transplant. The intravenous form
of ganciclovir is effective for preventing CMV. An oral
formulation is effective for both preventing and treating
CMV retinitis in HIV-infected patients, and trials with this
agent in transplant patients have begun. One of the most
effective ways to prevent a primary CMV infection, which
generally is more serious than reactivation disease, is by
seromatching: the avoidance of transplanting a seronegative recipient with a kidney from a seropositive donor.
However, most transplant centers view this as impractical.

The transplant physician must also be vigilant regarding

the development of other rarer diseases such as pneumocystis carinii, listeria, legionella, toxoplasmosis, and cryptococcal meningitis. The incidence of pneumocystis and legionella infection has been reduced by the routine use of
postoperative trimethoprim/sulfamethoxazole prophylaxis.
Community-acquired infections such as the common cold,
influenza, pneumococcal pneumonia, chlamydia, diarrheal
syndromes, and sexually transmitted diseases can also affect transplant patients. All prospective transplant patients
should be vaccinated against hepatitis A and B, tetanus,
diphtheria, and pneumococcal disease. Like any other infectious diseases, these are potentially much more serious
in immunosuppressed patients; the benefit of prevention
is obvious.
POSTTRANSPLANT MALIGNANCIES

Neoplasia may also result from excessive immunosuppression. The incidence rates of most malignancies are
probably increased among transplant patients, although
lymphomas and skin cancers are the most significantly increased. Approximately 1% of kidney transplant patients
develop lymphoma and approximately 5% develop skin
cancer. B cell lymphomas are the most common, are often
associated with Epstein-Barr virus and, if identified early,
can sometimes be successfully treated by a drastic reduction
in immunosuppression. An increased incidence of lymphoma has been linked to use of antilymphocyte antibodies
in some centers but this relationship is not consistent. Prolonged or repeated courses of antilymphocyte antibodies,
and in particular OKT3, as well as high doses of immunosuppression are generally thought to place a transplant
patient at greater risk for developing lymphoma. Both
squamous cell and basal cell carcinomas should be anticipated in transplant patients, and appropriate referrals made
to a dermatologist for evaluation of suspicious skin lesions.
Many solid tumors that become manifest clinically after a
transplant were actually present before hand. Therefore,
a careful evaluation for tumors in age-appropriate patients
should be conducted during the pretransplant recipient
evaluation (see Chapter 74). Rarely, tumors are inadvertently transplanted with the allograft. Their discovery
should prompt immediate discontinuation of immunosuppression.
POSTTRANSPLANT HYPERTENSION

Hypertension posttransplant is common and often requires multiple drug therapy. While essential hypertension
and cyclosporine are the usual causes, allograft renal artery
stenosis and excessive renin production by the native kidneys should always be considered. Calcium channel blockers are usually the first-line therapy for hypertension for
several reasons, including a demonstrated reversal of the
reduction in renal plasma flow and GFR induced experimentally by cyclosporine and a possible direct immunosuppressive effect. Verapamil and diltiazem decrease the metabolism of cyclosporine and raise its blood levels.

75. Renal Transplantation: Immunosuppression and Postoperative Management

Nifedipine can independently cause gingival hypertrophy.
Angiotensin converting enzyme inhibitors, alpha- and betaadrenergic blockers can also be used effectively in renal
transplant patients.

487
POSTTRANSPLANT DRUG PRESCRIPTIONS

As the result of using prednisone and cyclosporine, many

patients will develop hyperlipidemia following transplantation. Generally, transplant physicians delay the treatment
of hyperlipidemia until the prednisone dose has been tapered to maintenance levels, because the problem often
disappears or lessens. However, recent data from the heart
transplant literature indicating that HMG-CoA reductase
inhibitors improve outcome by reducing the incidence of
both acute and chronic rejection suggest that these drugs
should be used earlier. After the prednisone dose has been
tapered to low levels (e.g., 0.1 mg kg"1 day"1), further
reductions or discontinuation of prednisone altogether
have generally not had any additional effect on lowering
lipid levels. Since tacrolimus does not cause hyperlipidemia, one must consider conversion from cyclosporine to
this drug if posttransplant hyperlipidemia is severe. Finally,
dietary maneuver should always be a part of any attempt
to lower a patient's lipid levels.

Numerous drugs may have adverse effects in transplant

patients. Nonsteroidal antiflammatory drugs often cause
precipitous and protracted declines in GFR, presumably
because they block production of vasodilator prostaglandins that counter the renal vasoconstriction induced by
cyclosporine. Allopurinol, a xanthine oxidase inhibitor, inhibits breakdown of 6-mercaptopurine, the active metabolite of azathioprine and increases the latter's marrow suppressive effect. Together, these effects complicate
management of cyclosporine-induced gout. Intra-articular
steroid injections, temporary increases in oral prednisone
dosing, colchicine, and judicious coadministration of azathioprine and allopurinol may be necessary. Erythromycin,
ketoconazole, fluconazole, itraconazole, diltiazem, nicardipine, verapamil, and methylprednisolone may interfere
with cyclosporine degradation. Rifampin, phenytoin, phenobarbital, and carbamazepine accelerate cyclosporine metabolism. Cyclosporine levels should be closely monitored
and a need for dosage adjustment anticipated when these
agents are prescribed. Because transplant patients are typically receiving numerous medications, and because of the
propensity for adverse interactions among them, the physician should be especially vigilant when altering the medication regimen of transplant patients.

POSTTRANSPLANT HYPERCALCEMIA

MANAGEMENT OF A FAILED RENAL ALLOGRAFT

In patients who have had long-standing renal failure, the

parathyroid glands are often quite hyperplastic and it may
take months to years for them to regress to normal activity.
The usual practice is to wait for this to happen, unless a
patient is symptomatic from hyperparathyroidism. Occasionally a posttransplant patient will develop significant
and symptomatic hypercalcemia. In this case, early surgery
should be considered.

When a renal allograft fails and a patient returns to

dialysis, a decision must be made regarding discontinuation
of immunosuppression and removal of the allograft. A retained allograft that is uncovered by immunosuppression
can become inflamed causing local and even systemic symp toms. A more serious consequence of a retained allograft
is the development of anti-HLA antibodies. If a patient
desires retransplantation, these antibodies can be devastating. In general, most allografts that are lost acutely to
rejection are removed and immunosuppression is discontinued with a short taper of the prednisone. Allografts that
are lost to chronic rejection can be retained if immunosuppression is continued (usually azathioprine and prednisone
because they are inexpensive), if the patient is interested
in being retransplanted, and if anti-HLA antibodies are
absent so that it is likely that a kidney will be found within
months. For sensitized patients in whom the wait for a
subsequent transplant will be long, continuing the immunosuppression becomes too risky and it is wiser to remove
the kidney and discontinue the drugs. If the patient is not
a candidate for retransplantation and the allograft was lost
to a chronic process, it is reasonable to discontinue immu nosuppression and observe the patient. A spike in antiHLA antibodies occurs in about 25% of patients after the
removal of an allograft if immunosuppression is discontinued immediately. Therefore, if there is no major risk of
continuing immunosuppression, the drugs can be continued
for 1 month before tapering. It has not been proved that
this prevents antibody formation.

POSTTRANSPLANT HYPERLIPIDEMIA

OTHER POSTTRANSPLANT COMPLICATIONS

Some patients develop diabetes mellitus after transplantation because of steroid use or tacrolimus. Many of these
patients require insulin. Urinary tract infection and pneumocystis prophylaxis are generally prescribed for 3 to 4
months after transplant, and both goals can be accomplished with trimethoprim sulfamethoxazole, one double
strength tablet every Monday, Wednesday, and Friday. Hypophosphatemia and hypomagnesemia are both very common following renal transplantation because of renal tubular abnormalities induced by cyclosporine, as well as
hyperparathyroidism and other causes. These minerals
should be replaced as needed. Calcium, alendronate, vitamin D, and calcium-sparing diuretics might reduce the severity of osteopenia. Erythrocytosis may result from excessive erythropoietin production by the allograft or native
kidneys. Phlebotomy may be required; however, some patients respond to angiotensin converting enzyme inhibitors
or theophylline.

488

Douglas J. Norman
OUTCOME OF RETRANSPLANTATION

The survival of second transplants is clearly inferior to

that of first transplants. If a previous graft was lost to
rejection within 3 months of transplant, the outcome of a
retransplant is especially poor. However, patients who did
well with a first transplant can do well with a second one
and should not be denied that opportunity. The cause of
a poorer survival with a retransplant is usually undetected
sensitization. Patients receiving a second transplant are of
high risk for failure and are usually given more potent
immunosuppression. This makes them more susceptible to
the adverse consequences of immunosuppression.
Bibliography
Almond PS, Matas A, Gillingham K, et al.: Risk factors for chronic
rejection in renal allograft recipients. Transplantation 55:752757, 1993. Barbosa LM, Gauthier GJ, Davis CL: Bone pain
that responds
to calcium channel blockers: A retrospective and prospective
study of transplant recipients. Transplantation 59:541-544,1995.
Barret W, First MF, Orm BS, et al.: Clinical course of malignancies
in renal transplant recipients. Cancer 72(7):2186 -2189, 1993.
Bass NM, Ockner RK: Drug induced liver disease. In: Hepatology:
A Textbook of Liver Disease. Saunders, Philadelphia, 1990.
Braun WE: Long-term complications of renal transplantation (Nephrology Forum: Clinical Conference). Kidney Int 37:13631378, 1990.

Curtis JJ: Hypertension following kidney transplantation. Am J

Kidney Dis 23:471-475, 1994.
de Mattos AM, Olyaei AJ, Bennett WM: Pharmacology of immunosuppressive medications used in renal diseases and transplant ation. Am J Kidney Dis 28(5):631-667, 1996.
Hricik DE, Almawi WY, Strom TB. Trends in the use of glucocorticoids in renal transplantation. Transplantation 57:979-989,1994.
Kasiske BL: Risk factors for accelerated atherosclerosis in renal
transplant recipients. Am J Med 84:985-992, 1988.
Kuo PC, Dafoe DC, Alfrey EJ, et al.: Post-transplant lymphoproliferative disorders and Epstein -Barr virus prophylaxis. Transplantation 59:135-138, 1995.
Luke RG: Pathophysiology and treatment of post -transplant hypertension. / Am Soc Nephrol 2:537-544, 1991.
Mahony JF: Long-term results and complications of transplantation: The kidney. Transplant Proc 21:1433-1434, 1989.
Penn I: The effect of immunosuppression on pre-existing cancers.
Transplantation 55:742-747, 1993.
Pereira BJ: Hepatitis C in organ transplantation: Its significance
and influence on transplantation policies. Curr Opin Nephrol
Hypertens 2:912-922, 1993.
Rubin RH: Infection in transplantation. Infect Dis Clin North Am
9(December):811-1092, 1995.
Solez K, Axelsen RA, Benediktsson H, et al.: International standardization of criteria for the histologic diagnosis of renal allograft rejection: The Banft working classification of kidney transplant pathology. Kidney Int 44(2):411-422, 1993.
Watkins PB. Drug metabolism by cytochrome P450 in the liver and
small bowel. Gastroenteroly Clin North Am 21(3):511-526,1992.

SECTION I I
HYPERTENSION

PATHOGENESIS OF HYPERTENSION
STEPHEN C. TEXTOR

Blood pressure control and hypertension reflect forces

governing systemic hemodynamic variables, specifically
blood flow (cardiac output) and peripheral vascular resis tance. Understanding these components will enable the
reader to approach the complex and expanding list of proposed causes of the various forms of primary and secondary hypertension.
Hypertensive disorders are common and are of particular
relevance for the nephrologist. Most diseases which impair
kidney function ultimately lead to a rise in arterial pressure,
usually accompanied by impaired urinary sodium excretion
and volume regulation. This relationship is so uniform that
many authors believe that any sustained rise in blood pres sure requires the participation of the kidney. Conversely,
hypertension is itself an important mechanism of kidney
disease, which accelerates many forms of nephron injury
when transmitted as increased hydraulic pressure to the
glomerulus.

case in end stage liver failure, in which widespread systemic

vasodilation produces low arterial pressures despite expansion of blood volume.
When measured in most essential hypertensive patients,
cardiac filling volumes and cardiac output are normal. Although occasional instances of hypertension due to an increased cardiac output in patients with hyperkinetic circulatory states have been reported, these are unusual and often
temporary. Even experimental models of hypertension induced by volume expansion ultimately return cardiac output to normal with a secondary rise in peripheral resistance.
Understanding hypertension, therefore, should be
viewed as a problem of understanding the relative determinants of both blood volume and vascular resistance.
VOLUME CONTROL AND BLOOD
PRESSURE REGULATION

Some of the major regulators of sodium and volume

homeostasis are listed in Table 1. The importance of intravascular volume in blood pressure regulation is obvious
when blood pressure is restored after administration of
volume during hypotensive hemorrhage. At the other extreme, "refractory" hypertension in patients with end stage
renal disease becomes manageable after removal of extracellular fluid volume with dialysis. However, the role of
volume control is more subtle in patients with clinical hypertension and apparently normal renal function.
Guyton et al. have argued that sustained elevations in
blood pressure require a disturbance in the kidney. This
concept derives from well-established experimental observations of "pressure natriuresis," as depicted in Fig. 2.
This model indicates that the kidney functions as a precise
controller of arterial pressure. Changes in blood pressure,
and therefore renal perfusion pressure, shown in the left
curve, are translated into steep changes in sodium excretion. This scheme fits well with the clinical observation that
sodium excretion falls with a drop in blood pressure and
does not return to normal until pressure is restored. Conversely, a rise in arterial pressure leads to enhanced sodium
excretion until the pressure again falls. In principle, sodium
retention has no limit and can proceed until blood pressure
is fully restored. For that reason, it has been described as
a response system with "infinite gain."
Such a model would lead hypertensive subjects to undergo natriuresis and eventual volume depletion. Actually,

HEMODYNAMIC RELATIONSHIPS IN HYPERTENSION

The fundamental hemodynamic equation relating blood

pressure to the product of cardiac output and total systemic
vascular resistance is depicted in Fig. 1. Arterial pressure
is determined by the diverse combination of factors that
affects either (or both) of these variables. Hypertension
therefore results from a relative imbalance of cardiac output and systemic vascular resistance.
Cardiac output and blood flow are affected by blood
volume reaching the heart, often described as "cardiac
filling volume." This expression reflects the general relationship, albeit imprecise, between total blood volume and
arterial pressure. At the extremes of low volume, as produced by hypotensive hemorrhage or gastrointestinal fluid
losses, arterial pressure falls. Conversely, conditions producing a rise in blood volume without other major organ
abnormalities regularly lead to systemic hypertension. Sodium retention following acute glomerulonephritis is one
example. However, it should be emphasized that large portions of the total blood volume are located in capillary
networks , venous beds and the spleen, which are distensible
capacitance compartments (Fig. 1). These are not in direct
equilibrium with the arterial circuit.
Hence, blood volume alone does not directly determine
blood pressure. In some situations, blood volume is dissociated from systemic arterial pressure altogether. This is the

Primer on Kidney Diseases, Second Edition

491

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

Stephen C. Textor

492

Capacitance
beds
Veins, venules and
pulmonary (65%)

Arteries
(16- 18%)

50

Capillary beds
Venous Volume (65-70%)

Arterial Volume
(30-35%)

FIGURE I Systemic blood pressure is measured

on the "resis tance" side of the circulation. Organ perfusion is
determined both by the pump capacity [cardiac output (CO)] and
the systemic vascular resistance (SVR), which is the impedance to
blood flow, primarily determined by the lumen diameters of
resistance vessels. The blood volume within the arterial circuits is
less than that sequestered in the venous portions of the
circulation.

Systemic Sodium balance:

dietary
intake
Aldosterone
Glucocorticoids Atrial

Clinical examples "Saltsensitive" hypertension

Primary aldosteronism
Cushing's syndrome
TABLE I

Regulatory Mechanisms Affecting Cardiac Output and

"Filling Volumes"
natriuretic factors Thyroid
hormones
Local: Renal: "Pressure
natriuresis" Adrenergic
nerves Intrarenal hormones
Prostanoids: prostacyclin

Hyper- and hypothyroidism

Chronic renal failure

Loss of blood pressure

control during NSAID
Kallikrein
Nitric oxide
Renin-angiotensin system
Interstitial pressure Distal
tubular epithelial
sodium channel

Cyclosporine-induced
hypertension
Renovascular hypertension
Obstructive uropathy
Liddle's syndrome

NSAID, Nonsteroidal anti- inflammatory drug; HTN, Hypertension.

IOO
ISO
Arterial Pressure ( mm.Hg)

200

FIGURE 2. Schematic figure illustrating the ability of the normal

kidney to adjust sodium excretion depending on renal perfusion
pressure. Normally, a steep excretion curve provides precise control of sodium excretion around normal levels of pressure. Experimental data indicate that hypertensive models depend on a shift
of this curve to the right, requiring higher pressures to achieve
sodium balance (see text). From Guyton AC: In: Guyton AC (ed)
Textbook of Medical Physiology. W. B. Saunders, Philadelphia,
pp. 205-220, 1991.

measured intravascular volumes are reduced in essential

hypertension, although an occasional patient with severe
hypertension develops volume depletion, such that even
a minor vasodilation leads to hypotension. An important
corollary to this hypothesis is that any patient who maintains normal sodium and volume balance despite high
blood pressures must have an "abnormal" kidney. This
paradigm is illustrated in Fig. 2 as the curve on the right.
If progressive volume depletion does not occur, the argument follows that resetting of renal function must have
occurred to allow pressure natriuresis to occur at higher
renal perfusion pressures. While this model has some unproven elements, it nonetheless fits many clinical observations, some of which follow.
First, experimental and clinical studies have established
that hypertension follows the kidney. Experimental models
of "salt-sensitive" hypertension in which animals who are
genetically susceptible to develop high blood pressure during high sodium intake lose this susceptibility when transplanted with a "salt-resistant" kidney. In humans, transplantation of kidneys from normotensive organ donors to
patients that have lost their own kidneys from hypertensive
nephrosclerosis can lead to reversal of hypertension. Second, disorders of sodium transport, either acquired (such
as an aldosterone-producing adenoma) or genetic (such
as an abnormal epithelial sodium transporter in Liddle's
syndrome), can lead to hypertension by changing the
pressure-sodium relationship, effectively shifting from
point (A) to point (B) in Fig. 2. Third, the long-established
role of diuretics in treating hypertension becomes comprehensible without postulating "volume depletion": One im-

76. Pathogenesis of Hypertension

493

portant effect of diuretic administration is to "reset" pres sure natriuresis to lower pressures. Hence, hypertensive
patients can maintain sodium balance despite lower arterial
pressures than their "hypertensive" native kidneys would
normally tolerate, thereby moving from point (B) to point
(A). Fourth, efforts to lower blood pressure that do not
"reset" pressure natriuresis typically induce sodium retention and eventually fail. Hence, antihypertensive therapy
with alpha-methyldopa commonly was complicated by
"pseudoresistance" during which blood pressure rose after
a period of successful reduction. This was shown to reflect
occult volume retention and could be reversed with diuretics.
Taken together, blood pressure and the integrity of circulatory perfusion depend on maintaining sodium balance
and cardiac filling volumes. The other major element in
the equation is systemic vascular resistance.
DETERMINANTS OF VASCULAR RESISTANCE

The major hemodynamic abnormality underlying hypertension is unquestionably elevated vascular resistance.
With time, severely hypertensive patients develop hemodynamic profiles characterized by marked vasoconstriction
and reduced cardiac outputs. Vascular resistance itself cannot be explained in simple terms, but reflects an expanding
list of interrelated mechanisms that change vessel characteristics and lumen diameter.
Major mechanisms regulating the level of vasomotor
tone are depicted in Fig. 3 and listed in Table 2. They may
be viewed as representing different "levels" of cardiovascular control, depending than where they originate. They

TABLE 2

Regulatory Mechanisms Affecting Vascular Resistance

'Functional" Regulation
Long range Neurogenic
CNS adrenergic stimuli
e.g., stress, pain,
trauma
Regional adrenergic tone:
renal nerves Baroceptor
reflexes Hormonal
systems Renin angiotensin Aldosterone
Adrenal catecholamines
Thyroid hormone Insulin
Na,K-ATPase
inhibitors Short range
Paracrine systems
Prostaglandins, e.g.,
Prostacyclin PGI 2 /
Thromboxane Endothelin
Nitric oxide

Clinical examples

"Structural" regulation
Vascular tissue Medial
hypertrophy Calcium
compartmentalization

?Essential hypertension
NSAID use/pregnancy/steroid
administration
?Cyclosporine
?Cyclosporine

Posttraumatic hypertension

Autonomic failure
Renovascular hypertension
Primary aldosteronism
Pheochromocytoma Hyperand hypothyroidism
?Essential hypertension
?Essential hypertension

? Essential hypertension
?Essential hypertension
NSAID, nonsteroidal anti-inflammatory drug.

^Vessel wall

Lumen,

Hormonal Factors
Angiotensin Na+ K+ ATPase
Inhibition
Vasopressin'

.Interstitial Growth
^Factors

/
Growth Factors
Insulin PDGF
Angiotensin II
TGF -fi'

Neurogenic
Factors

/ ^/Vascular smooth
muscle
// \ Calcium (
), localization
'

Endothelial
substances
prostacyclin

endothelin'
EDRF
F.DCF

FIGURE 3 Schematic view of arterial resistance vessel, for which the lumen diameter and the resistance to blood
flow are regulated by many levels of control. Systemic factors such as neurogenic and hormonal stimuli are combined
with local systems including vascular-derived materials such as EDRF (nitric oxide) and medial hypertrophy. The
rise in systemic vascular resistance which characterizes most hypertensive states results from an imbalance of
these forces.

494

range from global mechanisms from distant sites, such as

adrenergic vasomotor stimuli from the central nervous system, to local factors limited to the vessel wall itself. An
example of the latter is nitric oxide (NO) [previously referred to as "endothelium-derived relaxing factor"
(EDRF)] from the surface layer of endothelial cells diffusing to adjacent vascular smooth muscle cells.
Some general comments are warranted to place these
mechanisms in a clinical perspective. Many of these systems
provide "redundancy" and are activated under specific
circumstances. The renin-angiotensin system, for example, is suppressed when patients ingest a surfeit of dietary sodium, i.e., more than 150 mEq/day. When dietary sodium
is restricted, however, renin is released from the kidney
as part of an integrated response that maintains blood
pressure and retains sodium. This characteristic applies
equally to normal subjects and patients with essential hypertension. It follows that pharmacologic blockade of the
renin-angiotensin system has different effects depending
on the pretreatment activation of the system.
Administering such an agent to patients depleted of sodium and therefore dependent on circulating angiotensin
for vascular tone produces a profound fall in vascular resis tance and blood pressure that would not occur if that individual is taking larger amounts of sodium. Renin is also
released when perfusion pressures are reduced to the kidney. Hence, any condition with reduced renal perfusion
pressure activates this potent pressor and sodium-retaining
mechanism. The ability of these systems to change under
various conditions cannot be overstated. Understanding
these interactions is an important component of treating
hypertensive diseases effectively.
Another reflex response activated during reduced aortic
pressure is the adrenergic system, which compensates with
increased sympathetically mediated vasoconstriction.
Thus, adrenergic vasoconstriction and the renin-angiotensin system reinforce each other to protect the organism
from hypotension and shock. Hypertension, therefore, appears as "inappropriate" activation of these mechanisms
which normally "defend" blood pressure.
Regulators of vascular tone interact with one another
at many levels, e.g., vasodilating prostaglandins, such as
prostacyclin, and other vasodilators including nitric oxide
"buffer" vasoconstricting forces, such as angiotensin II.
The latter itself stimulates local prostacyclin and nitric oxide production. Hence, a vasoconstrictor's effect is limited
by the vasodilators it induces. Such counteraction protects
tissue microcireulations from extreme changes. The importance of these relationships becomes apparent clinically
when one arm is pharmacologically disrupted. For example,
during inhibition of prostaglandin synthesis by the adminis tration of nonsteroidal antiinflammatory drugs, unopposed
vasoconstriction due to removal of prostacyclin leads to
deterioration of blood pressure control and renal function.
Excess activity of specific vasoconstrictor mechanisms
may lead to hypertension as noted in Table 2. Many experimental models and clinical forms of "secondary" human
hypertension reinforce this concept. Pheochromocytoma is
an example of autonomous production of catecholamines,

Stephen C. Textor

usually norepinephrine, which leads to either sustained or

paroxysmal hypertension. Both unilateral renovascular dis ease and renal infarction lead to overactivity of the reninangiotensin system and hypertension characterized by angiotensin II-mediated vasoconstriction. Verification of this
sequence depends, of course, on reversal of hypertension
by removal of the offending mechanism, which, in fact,
occurs after surgical removal of a pheochromocytoma or
revascularization of the kidney.
The fundamental abnormalities underlying vascular regulation in essential hypertension are uncertain. Rare forms
of hypertension appear to be the result of single gene mutations. An example is the familial form of Liddle's syndrome,
for which a mutation in the gene coding for an epithelial
sodium transporter leads to enhanced distal nephron sodium reabsorption. However, population studies indicate
that arterial blood pressure generally is an inherited phenotypic trait of polygenic origin. Hence, no single gene plays
a dominant role in essential hypertension. Rather, the interaction of many genes with each other and with environmental factors such as sodium intake, ultimately leads to a rise in
pressure. Provocative data suggest that insulin "resistance"
also is a trait commonly present in patients developing
essential hypertension. Such insulin resistance is present
in offspring of hypertensive parents prior to the onset of
clinical hypertension in the children. Whether this trait and
the potential for insulin to magnify vascular effects leading
to medial hypertrophy participate in the development of
essential hypertension are under study. Ouabain-like inhibitors of Na,K-dependent ATPase have long been suspected
of circulating in humans. It has been proposed that such
an inhibitor may offset a subtle defect in urinary sodium
excretion. Hence, inhibition of Na,K-ATPase can restore
sodium balance, but at the price of enhanced vasoconstriction in vascular smooth muscle. Taken together, insulin
resistance and sodium pump inhibitors offer potential examples of how circulating factors may participate in abnormal vascular regulation that ultimately leads to hypertension. It is likely that the effects of genetically determined
components of blood pressure in humans are modulated
heavily by interaction with environmental factors, such as
the level of dietary sodium intake and obesity.
Recent data suggest that vessel-derived vasoactive substances also modulate vascular resistance locally. The precise role of these substances in human disease states is only
partly understood. Recent studies using inhibitors of No
indicate that NO plays a tonic role in blood pressure control. Another local factor is endothelin, a potent vasoconstrictor peptide demonstrated first in endothelial cell cultures. This substance is released during conditions of
vascular wall injury, such as with atherosclerosis or
cyclosporine administration. Hence, control of vascular resistance and smooth muscle tone are subject to extensive
local and systemic controls. The force of vascular smooth
muscle contraction depends on intracellular calcium, and
perhaps other cations such as magnesium. These ions are
in continuous redistribution within cellular compartments
and across the cell outer membrane, moving via complex
transport mechanisms and ion "channels." Most vascular

495

76. Pathogenesis of Hypertension

regulatory mechanisms exert effects by changing the rate or

distribution sites of calcium, as do antihypertensive drugs.
Whether essential hypertension depends also primarily on
abnormalities of intracellular ion kinetics is not yet known.
Most of the aforementioned factors change vascular resistance by increasing smooth muscle contraction surrounding the blood vessel. It is likely also that structural alterations of the vessel wall contribute to changes in vascular
geometry and resistance. Classic studies have demonstrated
that hypertrophy of vascular tissue reduces lumen size.
Whether these "structural" changes represent a permanent
disease process or are in some way reversible is uncertain.
Recent studies indicate that cellular growth and medial
hypertrophy can be induced by altered local shear forces.
Continuous remodeling of the vessel can be demonstrated.
These changes in turn may be modified by growthpromoting factors including insulin, platelet-derived
growth factors, and probably many others.
INTEGRATION OF VOLUME AND
RESISTANCE ABNORMALITIES

Taken together, the determinants of circulatory volume

and vascular resistance are complex and interrelated. Examination of Tables 1 and 2 indicates that many of the
mechanisms that appear in both tables affect both parameters of the pressure equation. It should come as no surprise
that a biologic variable as critical as organ perfusion pres sure should be subject to multiple levels of regulation and
redundant systems . This has sometimes been described as
a "mosaic," from which elevated blood pressure derives
from many permutations of affected systems. Hence, it is
rare to encounter a "pure" form of hypertension in which
only one parameter is deranged. Even reversible, secondary forms of hypertension are regulated by multiple mechanisms, as suggested in the right-hand columns of Tables 1
and 2. This "plasticity" of blood pressure control is sometimes not fully appreciated. Changing one element leads
to other changes, which may finally return pressures to the
original level, albeit sustained by a different balance of
forces. A corollary of this observation is that effective treatment of hypertension requires intervening simultaneously
at multiple levels.

The complexity of blood pressure control necessarily

makes it difficult to identify the dominant factors in a specific patient. For the same reason, determining what primary pathogenic process (or processes) is responsible for
"essential" hypertension is hard. Nonetheless, understanding the determinants of volume and vascular resistance
provides a framework for approaching and intervening rationally in hypertensive disorders.
Bibliography
Cowley AW, Roman RJ: The role of the kidney in hypertension.
JAMA 275:1581-1587, 1996. Curtiss JJ, Luke RG, Dustan HP,
et al.: Remission of essential
hypertension after renal transplantation. N Engl J Med
309:1009-1015, 1983. Dominiczak AF, Bohr DF: Nitric oxide
and hypertension in 1995.
Curr Opin Nephrol Hypertens 5:174-180, 1996. Gavras H: The
place of angiotensin -converting enzyme inhibition
in the treatment of cardiovascular diseases. N Engl J Med
319:1541-1543, 1989. Goldstein DS: Plasma catecholamines in
essential hypertension:
An analytical review. Hypertensio n 5:86-99, 1983. Guyton AC:
Dominant role of the kidneys in long-term regulation
of arterial pressure and hypertension: The integrated system
for pressure control. In: Guyton AC (ed) Textbook of Medical
Physiology. W.B. Saunders, Philadelphia, pp. 205-220, 1991.
Heagerty AM, Bund SJ, Izzard AS: Long-term structural changes
in human hypertensive vessels. Basic Res Cardiol 86 (Suppl
l):19-23, 1991. Kuchel O: The autonomic nervous system
and blood pressure
regulation in human hypertension. In: Genest I, Kuchel O,
Hamet P, Cantin M (eds) Hypertension: Physiopathology and
Treatment. McGraw-Hill New York, pp. 140-160, 1983. Levin
ER: Mechanisms of disease: Endothelins. N Engl J Med
333:356-363, 1995. Reaven GM: Role of insulin resistance in
human disease. Diabetes
37:1595-1607, 1988. Romero JC, Lahera V, Salom MG, Biondi
ML: Role of the endo thelium-dependent relaxing factor nitric oxide on renal function.
/ Am Soc Nephrol 2:1371-1387, 1992. Tarazi RC: The
hemodynamics of hypertension. In: Genest K,
Kuchel O, Hamet P, Cantin M (eds) Hypertension: Physiopathology and Treatment. McGraw-Hill, New York, pp. 15 -42, 1983.
Vane JR, Anggard EE, Botting RM: Regulatory functions of the
vascular endothelium. N Engl J Med 323:27-36, 1990.

ESSENTIAL HYPERTENSION
JOSEPH V. NALLY, JR.

Hypertension is one of the most common problems in

clinical medicine in the United States and is the leading
indication for both physician office visits and prescription
drugs. Essential or primary hypertension is the sustained
elevation of systemic arterial pressure without a readily
definable etiology. It is by far the most common form of
hypertension, accounting for over 90% of the nearly 50
million Americans with hypertension. This increase in arterial pressure exists as a major rick factor for cardiovascular
disease. Consequently, there has been a tremendously increased effort over the past two decades to identify and
effectively treat patients with this "silent killer."

surveys regarding hypertension were performed during

four time periods (1960 to 1962, 1971 to 1975, 1976 to
1980,1988 to 1991). For these studies, three blood pressure
determinations were taken at a single visit and individuals
were questioned about their blood pressure and/or antihypertensive therapy. NHANES employed the WHO definition of hypertension, i.e., systolic blood pressure
>160 mm Hg, diastolic blood pressure >95 mm Hg, or
currently taking antihypertensive medications. Using these
criteria, 20 to 22% of the United States' population were
deemed hypertensive. As seen in Table 2, there has been
a gratifying increase in patient awareness, use of antihypertensive therapy, and control of blood pressure. Since the
inception of the National High Blood Pressure Education
Program (NHBPEP) in 1972, there has been a concomitant
decrease in age-adjusted cardiovascular mortality rates,
with a marked reduction in deaths from coronary heart
disease (about 50%) and stroke (about 57%) over the past
two decades. Since 1984, a lower threshold for defining
hypertension (140/90) has been adopted by the JNC such
that approximately 50 million Americans are deemed hypertensive.
Special subsets within a population have an increased
risk of hypertension. Both age and race are important variables. Blood pressure, especially systolic blood pressure,
increases with age such that hypertension is quite common
in the elderly. Over 50% of adults older than 65 have a
systolic blood pressure greater than 140 or diastolic blood
pressure greater than 90. Isolated systolic hypertension,
defined as systolic pressure >140 and diastolic pressure
< 90, is particularly problematic in the elderly population.
Isolated systolic hypertension confers the same cardiovas cular risk of stroke, congestive heart failure, and coronary
heart disease as does combined systolic/diastolic hypertension. During the Systolic Hypertension in the Elderly Program (SHEP) trial, active treatment with diuretics /Qblocker significantly reduced stroke and coronary heart
disease. Active treatment of systolic/diastolic hypertension
in the elderly in the Swedish Trial in Old Patients with
Hypertension (STOP-Hypertension) and the Medical Research Council (MRC) trial with a diuretic- and /3-blockerbased regimen significantly reduced all cardiovascular mortality. Overall, the elderly have a greater prevalence of
hypertension, yet they benefit from active antihypertensive therapy.

DEFINITION OF HYPERTENSION

There is no strict definition of hypertension, yet arbitrary

limits have been proposed to identify increasing cardiovas cular risk as a function of the elevation of systemic blood
pressure. Initial efforts by the World Health Organization
(WHO) suggested that blood pressure determinations
greater than 160/95 mm Hg defined hypertension and increasing risk. In the United States, the 1984 report of the
Joint National Committee on Detection, Evaluation, and
Treatment of High Blood Pressure ( JNC III) determined
that cardiovascular risk increases at a lower threshold and
established a limit
rneasurement^nthree separate visits. Traditionally, elevation of blood pressur^ia^Teeristratified based on the
diastolic blood pressure. The Fifth Report of the JNC ( JNC
V) recognized the importance of systolic hypertension correlating with increased cardiovascular risk and proposed a
definition of blood pressure determinations based upon
both the systolic and diastolic readings. Table 1 summarizes
these various levels of blood pressure determinations and
the JNC V recommendations for patient follow-up or referral.

PREVALENCE AND CARDIOVASCULAR RISK

The most exhaustive source of data for the prevalence

of hypertension, hypertension awareness and control in the
US population emanates from the National Health and
Nutrition Examination Survey (NHANES). Population

Primer on Kidney Diseases, Second Edition

496

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

497

77. Essential Hypertension

TABLE I
Classification of Blood Pressure and Recommendations for Follow -up for Adults Age 18 and Older"

Category
Normal

High normal

Systolic
(mm Hg)
< 130
130-139

Diastolic
(mm Hg)
< 85
85-89

140-159
160-179
180-209
> 210

90-99
100-109
110-119
> 120

Follow-up recommended
Recheck in 2 years

Recheck in 1 year

Hypertension

Stage 1 (Mild)
Stage 2 (Moderate)

Stage 3 (Severe)
Stage 4 (Very severe)

Confirm within 2 months

Evaluate or refer to source of care within 1 month
Evaluate or refer to source of care within 1 week
Evaluate or refer to source of care immediately

Not taking antihypertensive drugs and not acutely ill. When systolic and diastolic pressure fall into different categories,
the higher category should be selected to classify the individual's blood pressure status.
From The Fifth Report of the Joint National Committee on Detection, Evaluation, and Treatment of High Blood
Pressure.

Race is also an important variable regarding the prevalence and target organ effects of elevated systemic pressure.
Hypertension (defined as blood pressure > 160/95) is nearly
twice as common in African-Americans and rates of more
severe hypertension (diastolic blood pressure > 105) are
several-fold higher than in their white counterparts. Furthermore, for a given level of hypertension, AfricanAmericans tend to have more target organ damage with
left ventricular hypertrophy (LVH), strokes, and renal insufficiency.
Overall, cardiovascular risk in the entire population increases with elevation of systemic pressure above an arbitrary "normal" of 120/80. Simply stated, "the higher the
pressure. . . the higher the risk." For example, a sustained
elevation of diastolic pressure of 5 to 6 mm Hg increases
the risk of stroke by 34% and the risk of coronary heart
disease by 21%. Conversely, reduction of blood pressure by
a similar degree during large-scale antihypertensive studies
has lowered the rate of stroke by approximately 40%; although it only reduced coronary heart disease by about
14%. Additional cardiovascular risk factors such as hyperlipidemia, diabetes mellitus, smoking, and LVH may coexist with hypertension in any given patient. The goal of
antihypertensive therapy in the 1990s is both the reduction
of systemic blood pressure and the simultaneous modification of other cardiovascular risk factors. Increasingly, LVH
has been identified as an independent risk factor for coronary heart disease and congestive heart failure. A recent

meta-analysis suggests that different classes of antihypertensive agents may reduce blood pressure equally well, but
have a variable effect on the regression of LVH. Whether
regression of LVH in the short term translates into reduced
patient morbidity and mortality remains to be determined.
To date, the incidence of end stage renal disease (ESRD)
attributed to hypertension and nephrosclerosis has not decreased and remains a leading cause of renal failure in
the African-American community. The question of which
antihypertensive agents could have a renal protective effect
remains unclear. The Afro-American Study of Kidney Dis ease (AASKD) has been initiated to address this important issue.
CLASSIFICATION OF HYPERTENSION

Over 90% of all patients with high blood pressure pres enting to their primary care physician will have essential
hypertension as the cause of their hypertension is not
readily definable. A classification of hypertension is outlined in Table 3, with the estimates of prevalence of the
various forms of hypertension within the general population also listed. As seen in Table 3, approximately 5 to 10%
of patients may have secondary forms of hypertension.
Renal parenchymal disease is the most common medical
condition associated with hypertension. Renovascular hypertension is the most common potentially remediable
form of hypertension. Primary aldosteronism, pheochro-

TABLE I

Hypertension" Awareness, Treatment, and Control Rates

Aware: Percentage of hypertensives told by physician

Treated: Percentage of hypertensives on medication
Controlled: Percentage of hypertensives with blood pressure < 160/95 mm Hg on
one occasion measurement and reported currently taking antihypertensive

1971-1972

1974-1975

1976-1980

1988-1991

51

64
34
20

(54) 73
(33) 56
(11)34

(65) 84
(49) 73
(21) 55

36
16

medications
Note: Numbers in parentheses are percentage at 140/90 mm Hg.
" Denned as > 160/95 mm Hg on one occasion measurement or reported currently taking antihypertensive medication.

498

Joseph V.Nally, jr.

TABLE 3

Classification of Hypertension
Prevalence

Systolic and diastolic hypertension

Essential (primary) hypertension

90-95%

Secondary hypertension

Renal
Renal parenchymal disease
Polycystic kidney disease
Hydronephrosis
Renin-producing tumors
Renovascular hypertension or renal
infarction
Coarctation of the aorta
Endocrine
Oral contraceptives
Adrenal
Primary aldosteronism
Cushing's syndrome
Pheochromocytoma
Congenital adrenal hyperplasia
Liddle's syndrome
Thyroid (hypo- and hyper-)
Hypercalcemia
Hyperparathyroidism
Exogenous hormones glucocorticoids,
mineralocorticoids, sympathomimetics
Pregnancy-induced hypertension
"
Neurogenic
Alcohol and drugs (cyclosporine A,
erythropoietin)
Systolic hypertension
Isolated systolic hypertension (especially
elderly) Increased
cardiac output
Thyrotoxicosis
AV fistula
Aortic insufficiency
Paget's disease
Beriberi
a

5-10%

2.5-6.0%

or essential hypertension. The etiology of essential hypertension is a heterogeneous mixture of complex interactions
of heredity/environmental factors, sodium homeostasis, the
renin-angiotensin system, sympathetic nervous system, and
other factors which are incompletely understood. A detailed discussion of the pathophysiology of hypertension is
found in Chapter 76.
CLINICAL PRESENTATION

0.2-4.0%
1-2%

Hypertension is often discovered in an asymptomatic

individual as a result of a community screening program
or during an evaluation for another clinical problem. As
noted previously, the systemic pressure should be persis tently elevated on at least three different occasions before
labeling a patient as "hypertensive." The next important
decision is the extent of the diagnostic evaluation required.
Three key questions need to be addressed:
1. What is the cause of hypertension? Specifically, is
there a potentially reversible form of secondary
hypertension?
2. Is there evidence for target organ damage?
3. Are there coexistent cardiovascular risk factors?
All patients should undergo a thorough medical history
and physical examination and have a selected laboratory
evaluation to address these three important questions. The
following evaluation is recommended by JNC V.

Medical History
A detailed medical history should include:
1. Known duration and level of blood pressure
elevation

TABLE 4

Manifestations of Target Organ Disease

Organ system Manifestations

Special circumstances.
Cardiac

mocytoma, and Cushing's disease are seen less frequently.

Note that use of oral contraceptives and excesses of alcohol
present potentially reversible causes of hypertension.
PATHOPHYSIOLOGY

Simply stated, blood pressure (BP) is the product of

cardiac output (CO) and total peripheral resistance (TPR):
BP = CO X TPR
Nearly all forms of sustained clinical hypertension are due
to an elevation of TPR in the face of a normal or mildly
depressed cardiac output. This simple concept is important
to grasp in order to appreciate the intricacies of pathophysiology and implications for therapeutic intervention. In over
90% of cases, the precise cause of hypertension is not
readily definable and the patient is said to have idiopathic

Clinical, electrocardiographic, or
radiologic evidence of coronary artery
disease Left ventricular hypertrophy
or "strain"
by electrocardiography or left
ventricular hypertrophy by
echocardiography Left ventricular
dysfunction or cardiac
failure
Transient ischemic attack or stroke

Cerebrovascular
Peripheral vascular Absence of one or more major pulses in
the extremities (except for dorsalis
pedis) with or without intermittent
claudication; aneurysm
Renal
Serum creatinine (1.5 mg/dL)
Proteinuria (1+ or greater)
Microalbuminuria
Hemorrhages or exudates, with or
Retinopathy
without papilledema

77. Essential Hypertension

499

2. Antihypertensive therapy, results, and side effects

3. History of cardiovascular, cerebrovascular, or renal
disease and knowledge of diabetes, hyperlipidemia,
or gout.
4. Symptoms suggesting secondary hypertension
5. Family history of hypertension
6. Family history of premature atherosclerotic heart
disease, stroke, diabetes, or hyperlipidemia
7. History of weight gain, smoking, physical activity,
or mental stress
8. Dietary assessment including salt, fat, and
alcohol use
9. Thorough medication history, especially oral
contraceptives, steroids, NSAIDS, cold remedies,
appetite suppressants, tricyclic antidepressants,
MAO inhibitors, cyclosporine, erythropoietin.

6. Examination of the extremities for decreased or

absent pulses, bruits, or edema
7. Neurological assessment
Routine Laboratory Evaluation

Blood Pressure Measurements

The patient must refrain from smoking, caffeine use, and
exogenous adrenergic stimulants for 30 to 60 minutes. The
patient should be resting comfortably (seated or supine)
in a quiet room for at least 5 minutes. Using the appropriate
equipment and cuff size (15 cm wide cuff with bladder size
80% of arm circumference), two or more BP measurements
separated by 2 minutes should be obtained and then repeated after standing for at least 2 minutes. During the
initial exam, verify blood pressure in the contralateral arm
(if different, the higher value should be used).
Physical Examination
A thorough examination is required to search for evidence of target organ damage (see Table 4) or a cause of
secondary hypertension. The initial exam should include:
1. Height and weight
2. Fundoscopic exam (with pupil dilatation if
necessary) for arteriolar narrowing, arteriovenous
nicking, hemorrhages, exudates, or papilledema
(see Table 5)
3. Examination of the neck for carotid bruits, jugular
venous distension, thyroid gland enlargement
4. Examination of the heart for rhythm, displaced
point maximal impulse, murmur, or congestive
heart failure
5. Examination of the abdomen for bruits, large
kidneys, masses, or abnormal aortic pulsations

The JNC V has recommended that a few laboratory tests

should be performed routinely prior to initiating therapy.
These studies include: urinalysis, serum chemistry profile
(fasting blood sugar, potassium, creatinine, calcium, uric
acid), cholesterol/triglyceride profile, and electrocardiogram. The purpose of this testing is to screen for secondary
causes of hypertension, cardiovascular risk factors, and existent cardiovascular disease. Although the echocardiogram is a more sensitive test than the ECG for detecting
LVH, routine echocardiography is not recommended because of cost concerns. In the diabetic hypertensive population, the presence of microalbuminuria may be predictive
of the increased risk of renal or cardiovascular disease.
Additional laboratory testing for secondary forms of hypertension may be required if clinical suspicion exists.
Searching for Remediable Forms of Hypertension
In general, patients with essential hypertension present
during mid-life (30 to 55 years of age) with mild to moderate
hypertension, mild or no target organ damage, and often
have a family history of hypertension. However, a common
exception is the young African-American male with essential hypertension who may present with severe hypertension and significant renal impairment or LVH. Table 6
summarizes the clinical features that are "atypical" for
essential hypertension and should alert the physician to
possible secondary forms of hypertension. Five to ten percent of all hypertensive patients may have definable forms
of secondary hypertension, as outlined in Table 3. For
example, patients with edema and proteinuria may have
renal parenchymal disease requiring further evaluation. Patients with a family history of renal disease and hypertension who have a normal urinalysis should undergo renal
ultrasonography to screen for polycystic kidney disease.
Others with severe or resistant hypertension and/or an
abdominal bruit may be candidates to screen for renovascular hypertension (see Chapter 78). Coarctation of the aorta
should be excluded by careful examination with blood pres sures measured in the upper and lower extremities.

TABLE 5

Simplified Classification of Hypertensive Retinopathy

Retinal arterioles
Group
1

2
3
4

Sclerosis
<1
1 or more
0-4
0-4

Narrowing grade
Hemorrhages
0-4
0-4

0-4

0-4

Exudates
+

Papilledema
+

500

Joseph V.Nally, Jr.

TABLE 6

"Atypical" Features of Essential Hypertension

Age < 30 or > 55

Abrupt onset, severe hypertension (> stage 3)
Hypertension resistant to effective medical therapy
Target organ damage
Fundi with acute hemorrhages/exudates
Renal dysfunction Left
ventricular hypertrophy
Features indicative of secondary hypertension
Unprovoked hypokalemia Abdominal bruit or
diffuse atherosclerosis ACE inhibitor-induced
renal dysfunction Labile hypertension, sweats,
tremor, headache Family history of renal
disease

In selected cases, adrenal causes of hypertension should

be considered as potentially remediable forms of hypertension. Patients with a symptom complex suggestive of pheochromocytoma (episodic hypertension, sweats, headaches,
etc.) may be evaluated with plasma catecholamines, urinary
vanillylmandelic acid/metanephrines, or a clonidine suppression test. Clinical clues suggesting Cushing's syndrome
may be investigated with a dexamethasone suppression
test and ACTH measurement. Alternatively, patients with
unprovoked hypokalemia and metabolic alkalosis may be
screened biochemically for aldosterone overproduction. If
a biochemical abnormality is suggestive of adrenal hyperfunction in these three states, the adrenal glands should
be imaged with thin-cut CT.
THERAPY

The overall goal of antihypertensive therapy for patients

with essential hypertension is to reduce cardiovascular morbidity and mortality. The goal of therapy may be best met
by both lowering elevated systemic pressure and favorably
modifying other cardiovascular risk factors. Specifics of
nonpharmacologic and pharmacologic therapy in the management of the hypertensive patient are detailed in Chapter 35.

Bibliography
Bravo EL: Evolving concepts in the pathophysiology, diagnosis,
and treatment of pheochromocytoma. Endocrinol Rev 15(3):
356-368, 1994.
Bravo EL: Primary aldosteronism: Issues in diagnosis and management. Endocrinol Metab Clin North Am 23:271-283, 1994.
Breslin DJ, Gifford RW, Fairbairn JF, Kearns TP: Prognostic
importance of ophthalmoscopic findings in essential hypertension. JAMA 195:335-338, 1996.
Dahlof B, Lindholm LH, Hansson L, et al.: Morbidity and mortality
in the Swedish Trial in Old Patients with Hypertension (STOPHypertension). Lancet 338:1281-1284, 1991.
The Fifth Report of the Joint National Committee on Detection,
Evaluation, and Treatment of High Blood Pressure: Arch Intern
Med 153:154-183, 1993.
Gifford RW Jr: Treatment of patients with systemic arterial hypertension. In: Schlant RC, Alexander RW (eds) Hurst's The Heart
8th ed. McGraw-Hill, New York, pp. 1427-1448, 1993.
Hypertension Detection and Follow-up Program Cooperative
Group. Five-year findings of the Hypertension Detection and
Follow-up Program: I. Reduction in mortality of persons with
high blood pressure, including mild hypertension. JAMA
242:2562-2571, 1979.
Hypertension Detection and Follow-up Program Cooperative
Group. Five-year findings of the Hypertension Detection and
Follow-up Program. II. Mortality by race-sex and age. JAMA
242:2572-2577, 1979.
Hypertension Detection and Follow-up Program Cooperative
Group. Five-year findings of the Hypertension Detection and
Follow-up Program. III. Reduction in stroke incidence among
persons with high blood pressure. JAMA 247:633-638, 1982.
Kannel W. Hypertension and global cardiovascular risk. In: Black
H (ed) Primer on Hypertension. American Heart Association,
Chicago, 1993.
Kaplan NM: Hypertension in the population at large. In: Kap lan
NM, Lieberman E (eds). Clinical Hypertension, 5th ed. Williams
and Wilkins, Baltimore, pp. 1-25, 1990.
MRC Working Party. Medical Research Council trial of treatment
of hypertension in older adults: Principal results. Br Med J
304:405-412, 1992.
Multip le Risk Factor Intervention Trial Research Group: Risk
factor changes and mortality results. JAMA 248:1465-1477,
1982.
Schmieder RE, Martus P, Klingbeil A: Reversal of left ventricular
hypertrophy in essential hypertension. A meta-analysis of randomized double-blind studies. JAMA 275:1507-1513, 1996.
SHEP Cooperative Research Group: Prevention of stroke by antihypertensive drug treatment in older persons with isolated systolic hypertension. Final results of the Systolic Hypertension in
the Elderly Program (SHEP). JAMA 265:3255-3264, 1991.

RENOYASCULAR HYPERTENSION
LAURA P. SVETKEY

Renovascular hypertension is a potentially curable form

of high blood pressure. Its treatment, with angioplasty,
stenting, or surgery, provides the opportunity to prevent
cardiovascular morbidity and mortality due to hypertension and to avoid the expense and potential morbidity of
lifelong pharmacotherapy for hypertension. Therefore, this
condition should be identified and corrected whenever possible.

of reliable hemo dynamic information, a conservative approach would dictate that luminal narrowing of 50% or
greater should be considered significant, particularly if
there is poststenotic dilatation.
PATHOPHYSIOLOGY

When RAS is present, blood pressure increases because

of activation of the renin-angiotensin-aldosterone axis (Fig.
1). Decreased perfusion to the affected kidney leads to
renin release, which in turn promotes the conversion of
angiotensinogen to angiotensin I. In the rate-limiting step,
angiotensin converting enzyme converts angiotensin I to
angiotensin II (A-II), which is a potent vasoconstrictor. AII also leads to aldosterone release, which promotes salt
and water retention. Vasoconstriction without the ability
to compensate through increased salt and water excretion
leads to sustained hypertension.
An animal model of renovascular hypertension, the two kidney, one-clip Goldblatt rat, demonstrates the pathophysiology of RVH. In the Goldblatt rat, a metal clip is
applied to one renal artery, decreasing perfusion to that
kidney. Shortly thereafter, as indicated in Fig. 2, Phase I,
plasma renin activity rises, leading to the generation of
A-II, and blood pressure consequently rises. In Phase II,
hypertension persists, despite the return of plasma renin
levels toward normal. Although plasma renin activity is
within the normal range, it should actually be considered
elevated in the setting of elevated blood pressure. During
this phase, the animals are volume expanded and may have
increased vascular responsiveness to A-II. The persistence
of hypertension in these rats without elevated renin activity
may be analogous to RVH in humans with normal plasma
renin levels.
Relieving the Goldblatt rat's stenosis by removing the
clip helps us predict the response to intervention in humans
with RAS. As demonstrated in Fig. 2, if the clip is removed
during Phase I or II, and blood flow to the kidney is restored, blood pressure promptly returns to normal. In Phase
III, however, remo ving the clip does not lead to decreased
blood pressure. After a period of stenosis, hypertension is
fixed, presumably due to alterations in the renin-angiotensin system acting locally (i.e., in renal or vascular smooth
muscle cells), as well as irreversible vascular changes. One
can extrapolate to humans and assume that when blood
pressure does not improve after correction of a RAS either

DEFINITION

Renovascular hypertension (RVH) is defined as high

blood pressure secondary to renal artery stenosis. In clinical
practice the diagnosis is made retrospectively, i.e., one says
that a patient has RVH if hypertension improves after
repair of a renal artery stenosis. Improvement in blood
pressure, as defined by the Cooperative Study of Renovascular Hypertension, means that previously elevated
blood pressure becomes normal without antihypertensive
medication, or that less medication is required to control
blood pressure. Hypertension does not have to be cured
to make the diagnosis of RVH.
Any process that narrows a main renal artery can cause
RVH. In this discussion, renal artery stenosis (RAS) will
generally refer to narrowing of a main renal artery, although stenosis of a branch or accessory artery may occasionally be responsible for RVH. RAS is most commonly
due to either fibromuscular dysplasia (particularly but not
exclusively in young women) and atherosclerosis. Rarely,
stenosis may be due to neurofibromatosis, radiation fibrosis, extrinsic compression, embolism, congenital anomaly,
and Takayasu's disease.
It is unclear how severe a stenosis must be in order to
cause RVH. Some authors consider 50% luminal narrowing
to indicate significant stenosis of a renal artery, but perfusion pressure in large arteries may be preserved until the
lumen is reduced by at least 70%. In addition, the obstruction to blood flow depends on the length and configuration
of the stenosis, so that the two-dimensional view on arteriography may over- or underestimate the three-dimensional
process of luminal narrowing. This problem may be particularly severe in the case of fibromuscular dysplasia (FMD),
in which the lesion is often web-like across the lumen.
The relevant information is the degree to which a stenosis
interferes with perfusion and stimulates renin release, but
this information is not obtainable clinically. In the absence

Primer on Kidney Diseases, Second Edition

501

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

502

Laura P. Svetkey
have RAS, and 10 to 20% (about 65% of those with RAS)
will have reduction in blood pressure following correction
of the stenosis.
Many clinicians assume that RVH is exceptionally rare
in African-Americans. This idea persists because the prevalence of essential hypertension is so high among AfricanAmericans that the subset of all hypertensives with RVH
is small. However, in populations that are assessed carefully
(i.e., with arteriography performed in all subjects), it is
clear that clinically selected African-Americans have similar prevalence rates to clinically selected whites.

GREEN -78-1 -

PRORENIN

Trypsin-like
Activating Enzyme
ANGIOTENSINI

ANGIOTENSINOGEN

Angiotensin
Coverting
Enzyme

ANGIOTENSIN II

CLINICAL FEATURES

INCREASED BLOOD PRESSURE

ANGIOTENSIN II RECEPTORS

Aldosterone
Synthesis
Vasoconstriction

FIGURE I

Renin-angiotensin-aldosterone axis.

the stenosis was not hemodynamically significant (i.e., it

didn't cause the elevated blood pressure to begin with) or
the hypertension, originally due to the stenosis, is now irreversible.
EPIDEMIOLOGY

Estimates of the prevalence of RVH vary widely, depending primarily on the population studied. Overall, however, probably less than 5% of all hypertensives have this
secondary form. Its occurrence is associated with several
clinical features, discussed below. When clinically selected
populations are evaluated, approximately 25 to 35% will

Table 1 lists clinical features that are associated with

renovascular hypertension. RVH is found in up to 32% of
patients with a history of malignant hypertension (severely
elevated blood pressure with papilledema) or accelerated
hypertension (severely elevated pressure with fresh retinal
hemorrhage), and in almost 50% of patients with abdominal
or flank bruit. Some authors specify that the bruit must
persist into diastole, but similarly high prevalence of RVH
may be found associated with systolic bruits. In fact, bruit
may simply be a marker for severe and/or diffuse vascular disease.
The presence of either severe hypertension or bruit
makes RVH likely, but their absence does not rule it out.
RVH should also be considered in the other settings listed
in Table 1. In addition, although cigarette smoking is too
widely prevalent to be useful in distinguishing RVH from
essential hypertension, tobacco use is associated with increased prevalence of both atherosclerotic and fibromuscular RAS.
Among patients (with or without hypertension) undergoing cardiac catheterization, 15% have significant RAS.
Even when RAS exists without accompanying hypertension, it is associated with increased 5-year mortality. In a
large prospective study, RAS was associated with severe
coronary artery disease, congestive heart failure, and peripheral vascular disease. Therefore, the presence of these
conditions should increase the clinical suspicion of RVH.
DIAGNOSTIC TESTS

As stated above, the diagnosis of RVH can be made

with assurance only when blood pressure falls in response
Phase I
TABLE I

Clinical Characteristics Associated with

Renovascular Hypertension

BeninA BP
on removing clip
FIGURE 2 Phases of renovascular hypertension in the two kidney, one-clip Goldblatt rat.

Malignant or accelerated hypertension

Abdominal or flank bruit
Progression in the severity of chronic hypertension
Onset early (age < 25 yr) or late (age > 60 yr)
Hypotension or renal failure with ACE inhibitor
Severe or difficult to control hypertension Recent
onset of hypertension
Moderate hypertension and diffuse vascular disease
New onset of hypertension following renal
transplantation

78. Renovascular Hypertension

503

to interventional therapyan assessment that can only be

made retrospectively. In clinical practice, however, diagnostic tests are used to detect RAS, which is a necessary
precondition for the diagnosis of RVH. The gold standard
of diagnosis for RAS is conventional renal arteriography. Intravenous digital subtraction renal angiography
(IVDSA) will detect most stenoses and will be normal in
most patients without RAS. IVDSA has the advantage of
venous rather than arterial access, with a comparable dose
of contrast material. However, IVDSA requires subjective
judgment in interpretation, and it becomes uninterpretable
if there is overlying bowel gas or the patient is extremely
obese. In addition, in many institutions, there may be no
cost advantage to IVDSA over arteriography, and both
procedures can be performed on an outpatient basis. If
IVDSA is used instead of arteriography, a technically poor
result should never be considered adequate to rule out
RAS. In general, if the complication rate for arteriography
at a particular institution is low, arteriography offers advantages over IVDSA in accuracy and clarity, and is the preferred method of diagnosing RAS. The ordinary intravenous pyelogram is insufficiently sensitive and specific for
detection of RAS, and should not be used for this purpose.
It is frequently appropriate to proceed directly from clinical
suspicion to angiography (see below). But in other cases, the
expense and risk of this procedure may not be warranted.
Several noninvasive techniques have been developed
(Table 2), and may be useful in determining which
patients should be evaluated more definitively. These
screening tests are briefly summarized.

conditions are met, however, sensitivity and specificity may

be as low as 75%. In addition, ACE inhibitor-stimulated
plasma renin may be less useful in detecting RVH in
African-Americans.
ACE Inhibitor-Stimulated Renography
In the setting of RVH, both glomerular filtration rate
and renal plasma flow depend on angiotensin II-mediated
constriction of the glomerular efferent arteriole. Treatment
with an ACE inhibitor antagonizes this vasoconstriction
and decreases renal uptake and excretion of radiopharmaceuticals. A renogram is considered abnormal at baseline
if there is evidence of assymmetric function, or if uptake
or excretion of the marker is delayed. RAS is suspected if
the abnormalities are exacerbated by ACE inhibition (see
Fig. 3). False positive results (due to ACE inhibitorinduced oliguria and consequent "hang-up" of radiopharmaceutical in the renal cortex) can be avoided by simultaneous injection of furosemide. The test's utility may be limited
in bilateral disease since asymmetry may not occur and
because renal dysfunction may further confound interpretation. Estimates of the sensitivity and specificity of ACE
inhibitor-stimulated renography vary widely, but on average probably exceed 80% (somewhat lower in the setting
of renal insufficiency). Patients should be well hydrated
and must discontinue chronic ACE inhibitor therapy (they
may continue taking other blood pressure medications).
ACE inhibitor-stimulated renography is currently the most
useful noninvasive screening test for RVH. However, this
test must be interpreted cautiously since its interpretation

Peripheral Renin Activity

Many patients with RVH do not have elevated plasma
renin activity (PRA), and 16% of patients with essential
hypertension do. Therefore, PRA alone is not useful in
determining which patients should be evaluated further for
RVH. In order to improve the predictive value of PRA,
two tests take advantage of the expected excessive increase
with angiotensin converting enzyme (ACE) inhibition in
the setting of RVH: elevation of PRA after a single oral
dose of ACE inhibitor, or, alternatively, a measure of the
degree to which PRA increases from baseline after ACE
inhibitor (the Captopril Test). Unfortunately, these tests
are most useful in unmedicated patients (often difficult
to do safely in patients with severe hypertension), and
in patients with normal renal function. Even when these

32191
Left Kidney

DTPA Renogram
with captopril

16095
X

0
TABLE 2

Diagnostic and Screening Tests for

Renovascular Hypertension
Conventional renal arteriography

Intravenous digital subtraction renal angiography

ACE inhibitor-stimulated peripheral renin activity

ACE inhibitor-stimulated renography

Duplex ultrasound

Magnetic resonance angiography

Right Kidney

%*.

930

1860

FIGURE 3 Captopril renography in renal artery stenosis. The

figure shows accumulation of counts over the kidneys vs time after
injection of 99mT c-DTPAa GFR and renal perfusion marker.
Before administration of captopril (not shown), both kidneys
showed similar and normal prompt uptake and excretion. After
captopril, the right kidney continues to show rapid uptake and
excretion, but the left shows a marked delay in uptake and release
of labeled marker. At arteriography, the patient was found to
have an atherosclerotic lesion of the renal artery. Courtesy of
Dr. A. N. Shah.

504

Laura P. Svetkey

is somewhat subjective and its accuracy is reduced in the

setting of renal disease or bilateral RAS.
Duplex Ultrasound
The use of color-flow doppler in the diagnosis of RVH
is an attempt to combine anatomic information from ultrasound with hemodynamic information from doppler. This
technique seems to be seriously limited by the fact that the
proximal main renal artery is not visualized in 25% of cases.
In addition, small accessory and secondary renal arteries
are often not detected. Duplex ultrasound is highly
operator-dependent, and requires over 1 hour to perform
correctly. For these reasons, duplex ultrasound is not useful
in screening for RVH.
Magnetic Resonance Angiography (MRA) and Computerized
Tomographic Angiography
MRA is a noninvasive modality that can provide information about both anatomy and blood flow. Extensive data
are not yet available, but small studies suggest that MRA
detects RAS with a sensitivity of 70 to 90% and specificity
of 78 to 94%. If the more favorable of these results are
corroborated, MRA may offer a reasonable alternative
to conventional arteriography in patients at high risk for
contrast-related complications. Computerized tomo graphic angiography (CTA) require intravenous contrast
but avoids the need for an arterial puncture. At some centers, it has shown promise in detecting RAS (see Chapter
6, Fig. 3).

find RAS is extremely high, for instance if a patient has

severe hypertension (JNC-5 Stage 3 or 4) that is difficult
to control in the setting of progressive renal insufficiency,
or if the patient has moderate hypertension (Stage 2) and
is unable to tolerate any antihypertensive regimen, then it
is reasonable to proceed directly to angiography. At the
other end of the spectrum, if a patient has borderline or
mild hypertension (Stage 1) that is easily controlled, and
none of the clinical clues discussed above are present, then
no further workup is indicated.
Diagnostic difficulties arise in patients with clinical clues
that lead to an intermediate pretest probability of RVH.
This group comprises, for example, those with severe hypertension without history of a malignant or accelerated
phase, abrupt onset of hypertension very early or very late
in life, sudden increase in the severity of chronic hypertension, and unexplained renal insufficiency, especially if there
is evidence of vascular disease elsewhere. In this subgroup,
a noninvasive test may be helpful. In most circumstances,
the ACE inhibitor-stimulated renogram is the most useful
screening test. With an intermediate pretest probability of
RAS, renography is sufficiently sensitive and specific that
a negative test will effectively rule out RVH, and obviate
the need for further evaluation. A positive test will increase
the likelihood of RVH sufficiently to warrant angiography.
The decision to pursue this diagnosis in any individual
patient should include consideration of the potential risks
of arteriography and revascularization, risks which may be
magnified by diffuse atherosclerosis and renal insufficiency.
HEMODYNAMIC SIGNIFICANCE

DIAGNOSTIC STRATEGY

The utility of any given test for screening and diagnosis

of RVH will depend on the prevalence of RVH in the
population being evaluated (the pretest probability), and
on the sensitivity and specificity of the test. Even an excellent screening test, for example one with 90% sensitivity
and 90% specificity, will provide little help if the prior
probability of RVH is low. If one assumes that unselected
hypertensives have a 5% chance of having RVH, this hypothetical test will increase that likelihood to 32% if it is
positive, leaving 68% of those with an abnormal test with
a false positive result. A negative test will essentially rule
out RVH (negative predictive value = 99%), but the likelihood that RVH was absent was 95% before the test.
Because of these limitations of diagnostic tests that perform well (and not all of the procedures in Table 2 have
a sensitivity and specificity as high as 90%), Mann and
Pickering have suggested a strategy, somewhat modified
here, for selecting patients for further evaluation. This
strategy involves the definition of subgroups, based on the
clinical suspicion (pretest probability) of RVH.
If a patient has a history of malignant hypertension,
history of renal failure following ACE inhibitor, or the
presence of an abdominal or flank bruit, the likelihood of
finding RAS is greater than 30%, and one should proceed
directly to angiography. In addition, 'if the motivation to

After RAS has been detected, it would be useful to

predict blood pressure response to correction of the lesion.
Traditionally, measurement of the renal vein renin ratio is
used for this purpose. This test is based on the expectation
that a hemodynamically significant lesion will stimulate
renin release from the stenosed kidney and suppress its
release from the contralateral kidney. The predictive value
of this test is improved if renin release is stimulated with a
single oral dose of ACE inhibitor, but nonetheless remains
quite poor. In practice it is very difficult to predict blood
pressure response to intervention. In addition, relieving
a stenosis without lowering blood pressure may still be
beneficial if it protects that kidney from progressive ischemia Therefore, if RAS is detected, and it is possible to
intervene safely, this should be attempted, reserving conclusions about the hemodynamic significance of the stenosis
for retrospective consideration.
TREATMENT

Renal vascular disease is occasionally treated medically

unless blood pressure is difficult to control. However, the
implication of this approach is that many individuals will be
exposed to the expense and risk of lifelong antihypertensive
medication unnecessarily. In addition, the presence of RAS
may increase t he risk of chronic renal disease, and reperfu-

78. Renovascular Hypertension

sion may preserve renal function. Therefore, RAS should

be corrected whenever possible without undue risk.
Renal artery stenosis can be relieved by either surgical
revascularization, percutaneous transluminal angioplasty
(PTA), or renal artery stenting. When technically feasible,
angioplasty is preferred since it can be performed without
general anesthesia or lengthy hospitalization (see Kuhlmann et al, 1985). PTA is less likely to be technically
successful if an ostial stenosis is present (because these
lesions usually originate in the wall of the aorta and are
thus not affected by a balloon catheter inflated in the renal
artery). The recent introduction of stents has markedly
improved success rates in relieving ostial RAS.
The main limitations of PTA are primary failure and
subsequent restenosis. Primary failure occurs if the balloon
catheter can not be passed beyond the lesion. In addition,
even when the procedure can be performed, angiography
performed immediately after dilatation occasionally demonstrates persistent stenosis. These problems will be detected at the time of the procedure, and should lead to
alternative treatment (repeat PTA, stenting, or surgical
revascularization). After successful angioplasty, restenosis
will occur over the following year in approximately 20%
of patients, with slightly higher rates in patients with athero sclerotic than FMD lesions. Stents markedly reduce restenosis rates. When restenosis occurs following PRA, a second angioplasty is often successful, but stenting or surgical
revascularization may ultimately be necessary.
Bibliography
Davidson RA, Wilcox CS: Newer tests for the diagnosis of renovascular disease. JAMA 268:3353-3358, 1992.
Harding MB, Smith LR, Himmelstein SI, et al.: Renal artery steno sis: prevalence and associated risk factors in patients undergoing

505
routine cardiac catheterization. / Am Soc Nephrol 2:16081616, 1992.
Hudspeth DA, Hansen KJ, Reavis SW, Starr SM, Appel RG, Dean
RH: Renal duplex sonography after treatment of renovascular
disease. J Vase Surg 18:381-388, 1993.
Jacobson HR: Ischemic renal disease: an overlooked clinical entity? Kidney Int 34:729-743, 1988.
Kuhlmann, U, Greminger P, Gruntzig A, et al.: Long-term experience in percutaneous transluminal dilatation of renal artery
stenosis. Am J Med 79:692-698, 1985.
Mann SI, Pickering TG: Detection of renovascular hypertension:
state of the art, 1992. Ann Intern Med 117:845-853, 1992.
Miller GA, Ford KK. Braun SD, et al.: Percutaneous transluminal
angioplasty vs surgery for renovascular hypertension. AJR
144:447-450, 1985.
Martinez-Maldonado M: Pathophysiology of renovascular hypertension. Hypertension 17:707-719, 1991.
Olin IW, Piedmonte MR, Young IR, DeAnna S, Grubb M, Childs
MB: The utility of duplex ultrasound scanning of the renal arter ies for diagnosing sign ificant renal artery stenosis. Ann Intern
Med 122:833-838, 1995.
Plouin P-F, Darne B, Chatellier G, et al.: Restenosis after a first
percutaneous transluminal angioplasty. Hypertension 21:89
96, 1993.
Rimmer IM, Gennari FJ: Atherosclerotic renovascular disease
and progressive renal failure. Ann Intern Med 118:712-719,1993.
Simon N, Franklin SS, Bleifer KH, et al.: Clinical characteristics
of renovascular hypertension. JAMA 220:1209-1218, 1972.
Svetkey LP, Kadir S, Dunnick NR, et al.: Similar prevalence of
renovascular hypertension in selected blacks and whites. Hypertension 17:678-683, 1991.
Working Group on Renovascular Hypertension: Detection, evaluation and treatment of renovascular hypertension. Arch Intern
Med 147:820-829, 1987.
Ying CY, Tifft CP, Gavras H, Chobanian AV: Renal revascularization in the azotemic hypertensive patient resistant to therapy.
N EnglJ Med 311:1070-1075, 1984.

THERAPY OF HYPERTENSION
JOHN M. FLACK

By definition, hypertension is present when there is

a persistent elevation of either systolic blood pressure
(>140 mm Hg) or diastolic blood pressure (2:90 mm
Hg), or when a patient is taking antihypertensive medication (regardless of the blood pressure level). The blood
pressure level correlates directly with the magnitude of
risk for clinical sequelae such as premature death, stroke,
myocardial infarction, congestive heart failure, renal insufficiency, dementia, and peripheral vascular disease. Table
1 in Chapter 77 displays the blood pressure classification
from the fifth report of the Joint National Committee
on the Detection Treatment and Evaluation of High
Blood Pressure expert panel. This scheme determines
hypertension stage according to the highest pressure
either systolic or diastolic. The preponderance of epidemiological data suggests that systolic, not diastolic, blood
pressure is the primary determinant of pressure-related
risk.

pressure lowering. The evidence in support of this thesis

is compelling.
GENERAL THERAPEUTIC PRINCIPLES

Several important therapeutic principles should be consider in treating hypertensive patients. In most hypertensives there is little to be gained from the pursuit of rapid
blood pressure control. The most important therapeutic
goal for the vast majority is to prescribe a combination of
appropriate lifestyle modifications (weight loss, salt and
alcohol restriction, and increased physical activity) plus the
lowest doses of drug(s) that allow blood pressure normalization over the long term. The erroneous clinical perception that diastolic blood pressure is the predominant pres sure mediator of clinical risk contributes to physician
hesitancy to intensify treatment to normalize systolic blood
pressure once diastolic blood pressure has been lowered to
<90 mm less than Hg. Many antihypertensive medications
have dose-related side effects. Drug acquisition costs also
usually increase at higher dose levels. In certain instances,
cost is a major barrier to patient compliance with prescribed
drug therapies . The clinician and nursing staff should routinely ask patients whether they can afford the prescribed
medications. Drug acquisition costs are but one consideration, albeit an important one, in embarking on a hypertension disease management strategy.

RATIONALE FOR TREATMENT

A compelling case can be made for pharmacological

treatment of elevated blood pressure. Pharmacological
blood pressure lowering reduces the risk of premature cardiovascular morbid and fatal events as well as all-cause
mortality. Antihypertensive drug therapy has also been
shown to prevent the gradual progression of mild hypertension to more severe elevations of blood pressure. The prevalence of pressure-related target-organ damage (i.e., elevated serum creatinine, left ventricular hypertrophy) is also
greater at higher blood pressure levels. Hypertensives with
pressure-related target-organ damage manifest a severalfold higher risk for pressure-related clinical complications
at a given blood pressure level compared to hypertensives
with similar levels of pressure without target-organ damage. Early treatment of hypertension favorably impacts
long-term clinical risk, in part, by preventing the development of pressure-related target organ damage. It is not
well appreciated that drug-treated hypertensive patients
subjectively feel better (i.e., improved quality of life and
fewer pressure-related symptoms) after successful blood

Primer on Kidney Diseases, Second Edition

Es tablish a Goal Blood Pressure

Establishing a goal blood pressure is an extremely important consideration prior to the initiation of antihypertensive drug therapy. In most uncomplicated hypertensive
patients, the target blood pressure level will be <140/
90 mm Hg. Nevertheless, in selected high-risk subgroups
(i.e., renal insufficiency, diabetes), the target blood pressure
may justifiably be lower (<130/85 mm Hg). The Treatment
of Mild Hypertension Study [TOMHS] findings support a
more stringent goal blood pressure (< 130/85 mm Hg)
even for uncomplicated stage 1 diastolic hypertensives. In
TOMHS, drug-treated hypertensives (also prescribed a
multifactorial lifestyle intervention) had an average blood
pressure of 140/91 mm Hg at baseline. After 4 years
of treatment in this group, blood pressure averaged 127/

506

Copyright 1998 by the National Kidney Foundation

All rights of reproduction in any form reserved.

507

79. Therapy of Hypertension

79 mm Hg vs 133/82 mm Hg in TOMHS participants receiv ing only the multifactorial lifestyle intervention. The incidence of major and minor clinical events was 11.1% in
the drug-treatment group vs 16.2% in those receiving only
lifestyle modification.
Clinical Implications of Drug Pharmacokinetics

Drug pharmacokinetic profiles influence both the time

to onset of maximal blood pressure lowering and the degree
of protection against loss of blood pressure control during
the intervals between missed medication doses. The terminal drug half life in general, though not always, correlates
directly with both the duration of blood pressure lowering
as well as with the time to reach steady-state drug levels.
Drugs with a longer duration of action have a delayed onset
of maximal blood pressure lowering compared to drugs
with a shorter duration of action. However, once steadystate drug levels have been achieved, long duration of action drugs provide superior blood pressure control throughout the 24-hour dosing interval and also have a slow offset
of action which minimizes the loss of blood pressure control
when medication doses are missed. Pharmacokinetic considerations are important, as only 60% of hypertensives
who take antihypertensive medications over the long term
do so without missing scheduled doses.
Frequency of Monitoring/Drug Titration

A reasonable time frame for attainment of normal blood

pressure levels should be measured in months, not days or
weeks. In most patients, it matters little over the long term
if blood pressure control is achieved in 4 months as opposed
to 4 weeks. Long-acting antihypertensive drugs usually take
up to 3 to 6 weeks to fully manifest their maximal blood
pressure lowering effect. Thus, in most hypertensives, dose
increases or medication changes should be undertaken no
more frequently than every 6 weeks. This strategy will
facilitate attainment of goal blood pressure with a minimum
of office visits and also will minimize the likelihood of
precipitating symptoms attributable to rapid blood pres sure lowering. Nevertheless, if a patient manifests insignificant blood pressure lowering during the first few weeks
of therapy, the likelihood of a meaningful blood pressure
response over the ensuing weeks is low. A reasonable strategy at this point is to titrate the dose upward, add another
medication, or stop the current medication and switch to
another drug. Potential problems with rapid blood pressure
lowering are listed in Table 1.
Side Effects

Both the patient and physician usually attribute subjective side effects experienced by drug-treated hypertensives
to antihypertensive medication, because hypertension has
long been considered an asymptomatic condition. However, there is considerable evidence that hypertensive patients do indeed experience a constellation of symptoms
(Table 2) that correlate with the blood pressure level. Even

TABLE I

The Potential Hazards of a Hypertension

Disease Management Approach Utilizing Rapid
Upward Titration of Antihypertensive Agents
and Frequent Office Visits to Accomplish Rapid
Blood Pressure Lowering
Subjective side effects Dose-related
Correlate with rapidity of BP fall
Target organ ischemia Angina pectoris,
MI, TIA, or stroke
Higher total cost of care
Increased Drug doses
Unnecessary office visits
Excessive medication changes
Excessive patient monitoring

though antihypertensive drugs can cause symptoms, they

appear to alleviate more symptoms than they cause, as
drug-treated hypertensives report fewer subjective side effects than placebo-treated individuals with higher blood
pressure levels. Pressure-related symptoms such as weakness and fatigue not only relate to the blood pressure level,
but may also occur in a time-limited manner when the
blood pressure levels change rapidly in either direction.
Hypertensive patients are most at risk for time-limited
pressure-related symptoms during the transition from one
medication to another as a consequence of a transient loss
of blood pressure control. There is also an increased risk
for pressure-related symptoms of limited time duration
after rapid blood pressure lowering that correlates with the
height of the pretreatment blood pressure. The symptomatic drug-treated hypertensive is not uncommon in clinical
practice as fewer than 50% of drug-treated hypertensives
attain blood pressure levels < 140/90 mm Hg.
ANTIHYPERTENSIVE DRUG SELECTION
Monotherapy

The clinician must have reasonable expectations about

the likelihood of blood pressure normalization with mo notherapy. If expectations are unrealistic, then subsequent
therapeutic decisions will be misguided and therefore unlikely to result in blood pressure normalization. The lack
of therapeutic success will frustrate both the patient and
physician. The probability of achieving blood pressure normalization with monotherapy in the overall hypertensive
population ranges between 50 and 60%. Two or fewer drugs
will control about 85 to 90% of all hypertensives, whereas
three or more drugs will be required for blood pressure
normalization in the remainder. The probability of blood
pressure normalization with monotherapy is significantly
influenced by the magnitude of the pretreatment blood
pressure elevation. The Fifth Joint National Comittee

John M. Flack

508
TABLE 2

Constellation of Symptoms Reported More

Often in Hypertensives with Elevated Blood
Pressures Compared to Drug-Treated
Hypertensives with Lower Blood
Pressure Levels
Headache
Fatigue/weakness
Lack of stamina, poor exercise tolerance

Cardiac awareness
Dizziness

Nervousness
Sleep disturbance
Chest pain

blood pressure classification scheme can thus be quite useful in guiding therapeutic decisions (Table 3). A common
clinical mistake is to expect monotherapy to normalize
blood pressure levels in hypertensives with "high" Stage
2 or higher hypertension. If, however, an adequately dosed
therapeutic trial of monotherapy of sufficient duration is
undertaken, then it is a more effective strategy to add a
second drug when monotherapy fails. However, if a second
drug from an additional drug class is substituted, it is also
likely that this agent will fail to normalize blood pressure.
In this situation, drug substitution leads to extra office visits,
excessive patient monitoring, and unnecessary changes in
their therapeutic regimen, all of which contribute to a
higher total cost of care. Conversely, in "high" Stage 2 or
higher hypertension, initial monotherapy is a viable initial
treatment strategy even though the probability of ultimately achieving blood pressure control with a single drug
is low. Table 4 displays the advantages and disadvantages
of the major antihypertensive drug classes.
Diuretics
Monotherapy with thiazide diuretics effectively lowers
blood pressure with a similar degree of efficacy in both
African-American and white hypertensives. Thiazides
are, however, ineffective when the glomerular filtration
rate is < 40 mL/min; a clinical rule of thumb is that thiazides should be avoided when serum creatinine levels are
a 2.3 mg/dL. Current recommendations are to prescribe
25 mg/day or less. Doses as low as 6.25 mg/day have been
shown to effectively lower blood pressure. These agents
are cheap, as drug acquisition costs can be as low as pennies
per day. The absence of a diuretic is perhaps the most
common cause of resistant hypertension in patients treated
with complex [> 2] antihypertensive drug regimens.
The relatively long drug half lives of thiazides (chlorthalidone > hydrochlorothiazide) is a favorable pharmacokinetic characteristic for long-term hypertension management. Thiazides have dose-related side effects such as
hypokalemia (< 3.5 mEq/L), hypomagnesemia, and
hyperuricemia. Hypokalemia is the most common meta-

bolic side effect, occurring in 30 to 35% of diuretic-treated

hypertensives consuming an ad libitum sodium diet. The
likelihood of hypokalemia can be reduced by these nonmu tually exclusive strategies: (1) reduce the diuretic dose; (2)
lower intake of dietary sodium; or (3) add a potassiumsparing diuretic or potassium supplement. Thiazides can
impair free water clearance and thus their use occasionally
results in hyponatremia in older hypertensives. These
agents also cause modest lipid disturbances, as they raise
cholesterol levels 6% and triglyceride levels approxi mately 15%, while leaving high density lipoprotein levels
unaffected. Thiazides can adversely affect glycemic control;
however, this can be avoided by preventing potassium
depletion. These agents are the worst offending drug class
for causing sexual dysfunction in men. Loop diuretics, metolazone, and indapamide maintain their blood pressure
lowering efficacy in patients with compromised renal function. Though loop diuretics are more potent diuretics than
the thiazides, these agents lower blood pressure less effectively in individuals with normal renal function. Diuretics
are especially useful in hypertensives with congestive
heart failure.
/3 -Blockers
/3-blockers are a heterogeneous group of drugs which
can be subclassified according to their relative affinity for
beta receptor subtypes, the presence or absence of intrinsic
sympathomimetic activity, and their lipophilicity. /3blockers can be categorized as: (1) nonselective (/3i and
/32) (i.e., nadolol, propranolol, timolol), (2) cardioselective
(/3i) (i.e., atenolol, metoprolol), (3) cardioselective with
intrinsic sympathomimetic activity (acebutolol), (4) nonselective with intrinsic sympathomimetic activity (i.e., pindolol), or (5) a-/3 blocker (labetalol). /3i receptors are the
predominant /3 receptor subtype in the heart while /32 receptors predominate in the lung and peripheral blood vessels.
Nevertheless, both receptor subtypes are found in target
organs. Cardioselective (/3i) /3-blockers have theoretical
advantages in patients with bronchospastic lung disease as
nonselective /3 blockers (/3i and /32) are more prone to
incite bronchospasm. However, because target tissues are
not populated by distinct /3 receptor subtypes, even cardioselective /3 blockers can cause bronchospasm and should be
avoided in patients with lung disease. /3 blockers (without
intrinsic sympathomimetic activity) should be a favored
antihypertensive therapy in the postmyocardial infarction
patient because of their proven cardioprotection and their
ability to lower mortality. Likewise these nonintrinsic sympathomimetic activity /3 blockers are particularly useful
in the hypertensive patient with either angina pectoris or
migrane headaches. These agents are highly effective blood
pressure lowering agents when used alone or in combination with diuretics. In the calcium antagonist group, only
the dihydropyridines are suitable for combination therapy
with /3-blockers. /3-blockers reduce cardiac output by approximately 20 to 25%; thus, these agents have been associated with low cardiac output symptoms such as weakness,
fatigue, and poor exercise tolerance. /3-blockers can also
worsen symptomatic peripheral vascular disease, have been

79. Therapy of Hypertension

509
TABLE 3

An Adaptation of the JNC V Blood Pressure Classification Scheme That Describes Therapeutic Options When Monotherapy
Fails to Lower Blood Pressure to < 140/90 mm Hg According to Patients Initial Blood Pressure Level

Stage 1
Stage 2
"low"
"high"
Stage 3-4

SBP (mm Hg)

DBF (mm Hg)

140-159

90-99

160-169
170-179
>180

100-104
105-109

allO

linked to depression, and lower high density lipoprotein

while raising triglycerides. The adverse lipid effects are
less pronounced with intrinsic sympathomimetic activity /Sblockers. Less lipophilic /3 blockers may cause less depres sion because they poorly traverse the blood brain barrier
into the central nervous system. /3-blockers should be used
with caution in hypoglycemia-prone diabetics because they:
(1) blunt the tachycardia accompanying hypoglycemia, and
(2) delay the metabolic recovery from hypoglycemia. There
are new data suggesting the benefit of low-dose /3 blockade
in patients with congestive heart failure. Nevertheless, until
more data are available, the clinician should avoid /3blockers in congestive heart failure patients unless there are
contraindications to one or more of the standard congestive
heart failure treatments (digitalis, diuretics, and angiotensin-converting enzyme inhibitors).
Angiotensin-Converting Enzyme Inhibitors

Angiotensin-converting enzyme inhibitors are particularly useful in diabetic hypertensives with proteinuria as
well as in hypertensives with congestive heart failure and
in those who have survived myocardial infarction. These
agents are metabolically neutral, do not cause male erectile dysfunction, and have no effect on fasting glucose
even though they modestly improve insulin sensitivity.
The major side effect of this drug class is cough, which
occurs in ~ 9% of hypertensives taking angiotensinconverting enzyme inhibitors. Angioedema is a potentially
life -threatening side effect occuring in approximately
3/1000 angiotensin-converting enzyme inhibitor-treated
patients. Angioedema can occur months after the initiation
of angiotensin converting inhibitor therapy. The incidence
of hyperkalemia in all angiotensin-converting enzyme
inhibitor-treated hypertensives is less than 1%. However,
those with renal insufficiency as well as those treated with
nonsteroidal anti-inflammatory drugs, potassium-sparing
diuretics, potassium supplements, and/or heparin have a
higher risk. Angiotensin-converting enzyme inhibitors
should be avoided in patients with bilateral renal artery
stenosis and should be used cautiously in patients with
unilateral renal artery stenosis because of the risk of iatro genic renal failure. African-Americans, particularly those
with high intake of dietary sodium, may require higher
doses of these agents to achieve blood pressure control.
The combined use of angiotensin-converting inhibitors

When monotherapy is unsuccessful

Switch drugs

Switch drugs; however, consider adding a second drug

Add a second drug
Add a second drug; however, consider two drug treatment initially

with either a diuretic or a calcium antagonist appears

to be highly efficacious and well tolerated. Angiotensinconverting enzyme inhibitors also blunt thiazide-induced
hypokalemia. These agents are potentially teratogenic and
should be avoided in pregnancy. Women of child-bearing
potential should be counseled to use contraceptives while
taking angiotensin-converting enzyme inhibitors and
should be changed to another agent if they are attempting
to become pregnant.
Alpha, Antagonists

Alphai antagonists improve insulin sensitivity and cause

modest favorable changes in all blood lipoproteins making
them desirable in hypertensives with either diabetes mellitus or dyslipidemia. These are preferred agents in older
men with benign prostatic hyperplasia as they improve
maximum urine flow rates in a dose-dependent manner.
This drug class either does not affect or modestly improves
erectile function in men. Doses required to achieve blood
pressure control are slightly higher in African-American
hypertensives than in white hypertensives. These agents
can precipitate or worsen orthostatic hypotension. Thus the
clinican should routinely take orthostatic blood pressure
measurements in at-risk groups including older persons,
diabetics, sedentary individuals, and hypertensives concurrently treated with diuretics, central adrenergic inhibitors,
or combined a-/3 blockers.
Calcium Antagonists

Calcium antagonists are a heterogenous class of vasoactive compounds which effectively lower blood pressure
in all racial, ethnic, and demographic groups. These agents
have variable hemodynamic, inotropic, chronotropic, and
neurohumoral effects. There is essentially no role for shortacting calcium antagonists in the long-term therapy of hypertension. Short-acting nifedipine can be useful for acute
blood pressure lowering during legitimate hypertensive urgencies, although some recent evidence suggests caution
when using this agent for this indication. Calcium antagonists also can be subclassified as either rate-limiting (verapamil, diltiazem) or dihydropyridine (i.e., nifedipine, amlodipine, felodipine). Rate-limiting calcium antagonists, in
contrast to the dihydropyridines, have atrioventricular
nodal blocking properties and usually should not be used
in combination with /3-blockers, nor should they be used

510

John M. Flack
TABLE 4

Advantages and Disadvantages of Seven Commonly Used Antihypertensive Drug Classes

Advantages
Renoprotective in type 1 diabetes

Disadvantages
Angiotensin-Converting Enzyme (ACE) Inhibitors
Cough (~ 9%)

Profoundly antiproteinuric

Rare angioedema (~ 3/1000)

I Morbidity and mortality and symptomatic improvement in CHF Rare hyperkalemia

patients
Higher dose requirements in some African-Americans
J, Ventricular remodeling and mortality post -Mi
BP lowering effect very sensitive to level of dietary sodium intake
Leftward shift of pressure-natriuresis curve
Safely combined with other anti-hypertensives, especially diuretics or
calcium antagonists
ax Antagonists
Orthostatic hypotension (particularly in the elderly, in

Positive effect on all lipoprotein fractions

Positive effect on all lipoprotein fractions

combination with other vasodilators, in the setting of autonomic

Improved insulin sensitivity

dysfunction, and in volume depleted patients)

Unchanged to improved sexual function in men

Slightly higher doses needed in African-Americans

Blunts thiazide - induced rise in cholesterol

Improves maximal urine flow in men with BPH
Lesser peak BP lowering compared to ACEs

Angiotensin II (AT,) Receptor Antagonists

Relatively flat BP dose-response curve

Nearly complete blockage of A-II effect

No effect on bradykinin

No effect on bradykinin, enkephalins, or substance P

Limited data available in African-Americans

Uricosuric (losartan)
Profoundly antiproteinuric
Metabolically neutral
Rare angioedema
No cough

Absence of hard clinical endpoint (i.e., CHF, MI, renal) data

Rare hyperkalemia

j3 Blockers
Can worsen CHF; however, at low doses may improve CHF

Differential cardiac and hemodynamic effects

Proven to lower morbidity and mortality post-M i

Low cardiac output symptoms

Selected agents useful in patients with migraine or angina

Can worsen or precipitate depression

Can aggravate PVD symptoms
Lowers HDL and raises TGs
Delays recovery from and masks symptoms of hypoglycemia
Abrupt discontinuation can lead to rebound hypertension
Calcium

Unqualified efficacy profile (African-Americans, diabetics,

Antagonists

Immediatley post-Mi or during unstable angina (short-acting

elderly)

dihydropyridines) may increase CHD risk

Heterogeneous electrophysiologic, inotropic, hemodynamic, and

Rate-limiting CBs can worsen CHF and f mortality in patients

SNS effects

with systolic heart failure

Useful in diastolic dysfunction

Side effect profile varies (constipation-verapamil, pedal edema,

Metabolically neutral

and vasodilatory symptoms-dihydropyridines)

Minimal erectile dysfunction in men

BP lowering effect is robust in setting of high dietary sodium

Combination with /3-blockers can result in profound brady cardia

and depression of myocardial contractility

intake BP lowering effect not attenuated by

NSAIDs
Relatively cheap
Useful in hypertensive urgencies (clonidine)
Methyldopa is drug of choice for pregnant women

Central Adrenergic Inhibitors

Sedation
Depression (high dose reserpine)
Orthostatic hypotension
Decreased heart rate (clonidine) Frequent skin irritation
(clonidine patch), dry mouth Abrupt discontinuation can
lead to rebound hypertension Positive Coombs test in
~15% (methyldopa)
Direct Vasodilators

Relatively cheap

Best suited for adjunctive therapy

Effective in severe hypertension (minoxidil > hydralazine)

Salt and water retention necessitate use of loop diuretics

Reflex tachydardia necessitates use of rate-limiting CCB or /3blocker

Do not regress LVH

(continues)

511

79. Therapy of Hypertension

TABLE 4 (continued)
Advantages

Disadvantages

Lupus-like syndrome (>200 mg/day of hydralazine)

Hirsutism (minoxidil) Edema (minoxidil)
Cheap acquisition cost
Highly effective and lowers BP similarly in all demographic
groups
Proven to lower BP -related morbidity and mortality
Enhances the BP lowering effect of all other BP drugs
Thiazides and metolazone have relatively long half lives

Diuretics
Thiazides ineffective when GFR <40 mL/min
Dose-related hypokalemia

Raises cholesterol and triglycerides

Glucose intolerance especially in setting of K+ depletion
Increased uric acid
Erectile dysfunction in men

"Hidden" costs of metabolic monitoring

BP, blood pressure; BPH, benign prostatic hyperplasia; ACE, angiotensin-converting enzyme inhibitors; A-II, angiotensin II; CCBs,
calcium channel blockers; CHD, coronary heart disease; CHF, congestive heart failure; HDL, high-density lipoprotein; PVD, peripheral
vascular disease; TG, triglycerides, SNS, sympathetic nervous system; NSAIDs, nonsteroidal anti-inflammatory drugs.

in patients with impaired systolic heart function. Dihydropyridine calcium antagonists are more potent peripheral
vasodilators and therefore do not depress myocardial contractility. Lower extremity edema can occur with any calcium antagonist but has been more closely associated with
the dihydropyridines. Neither rate-limiting nor dihydropyridine calcium antagonists should be prescribed to patients
with sick sinus syndrome unless a functioning ventricular
pacemaker is in place.

sedation, dry mouth, and orthostatic hypotension limit patient acceptance of these agents. Abrupt discontinuation
of these drugs can lead to rebound hypertension making
these agents less than ideal therapeutic agents for patients
with known noncompliance. The use of these agents in
combination with /3-blockers should be discouraged because of the risk of bradycardia. Methyldopa is the antihypertensive agent of choice in hypertensive pregnant
women.

Angiotensin Receptor Antagonists

Angiotensin receptor antagonists are the first new antihypertensive drug class introduced into clinical practice
in over a decade. Losartan and valsartan are the only
currently available agents in this class; however, irbersartan, telmisartan, seprosartan, and candesartan will likely
soon be available. Angiotensin recep tor antagonists
are well tolerated as monotherapy and have been most
extensively studied in combination with diuretics. These
agents are metabolically neutral and do not cause erectile
dysfunction in men. Like the angiotensin-converting enzyme inhibitors, angiotensin receptor antagonists antagonize the renin-angiotensin-aldosterone system (though at
a different site). Unlike the angiotensin-converting enzyme inhibitors, angiotensin receptor antagonists do not
cause cough. Anecdotal reports of angioedema with angiotensin receptor antagonists have, however, now surfaced, though the incidence is probably lower than with
the angiotensin converting enzyme inhibitors. Similar to
the angiotensin-converting enzyme inhibitors, these agents
should be avoided in individuals with bilateral renal
artery stenosis and used cautiously in patients with unilateral renal artery stenosis. These agents should also be
used with caution in hyperkalemia-prone individuals.

Direct Vasodilators
Direct vasodilators (i.e., minoxidil, hydralazine) are
uncommonly used as monotherapy mostly because they
evoke marked arterial vasodilatation and thus cause reflexive activation of the sympathetic nervous system.
This leads to both salt and water retention as well as
tachycardia. The latter raises myocardial oxygen consump tion and therefore can precipitate new or worsen preexisting coronary ischemia in patients with coronary heart
disease. These agents are most effectively used in combination with diuretics and /3-blockers or a rate-limiting
calcium antagonist. Lower extremity edema may complicate minoxidil therapy even when potent diuretics are
simultaneously prescribed.

Central Adrenergic Inhibitors

Central adrenergic inhibitors (i.e., clonidine, methyldopa) are most useful for adjunctive antihypertensive therapy. These agents effectively lower blood pressure and also
cause regression of left ventricular hypertrophy. However,

Combination Therapy
Almost 50% of all hypertensives and most Stage 2 or
higher hypertensives will require more than a single antihypertensive agent to achieve blood pressure normalization.
Thus, physicians should become familiar with the clinical
scenarios when combination therapy will likely be needed.
In "high" Stage 2 or higher hypertensives, it is reasonable
for physicians to consider initiating antihypertensive therapy with two drugs. Dual drug therapy can be initiated
either by prescription of two separate pills or by prescribing
a single combination pill. The combined blood pressure
lowering effect of two drugs at low-to-moderate doses usually exceeds the magnitude of blood pressure, lowering
obtainable effect by upward titration of either drug alone

512

John M. Flack

2
o
most desirable potentially
useful limited usefulness
contraindicated

ACE Inhibitors
Alpha-, Antagonists
ATi Antagonists
Beta Blockers
Calcium Antagonist
Diuretics

Monotherapy with diuretics and calcium antagonists has,

however, been shown to result in greater blood pressure
lowering in African-Americans compared to other commonly used nondiuretic drug classes (i.e., angiotensinconverting enzyme inhibitors, a-antagonists, and /3blockers]. Nevertheless, the clinician should not generalize
these data to an individual and therefore presume a lack
of blood pressure lowering efficacy solely, or even mostly,
because of their race or ethnicity. When an adequately
dosed therapeutic trial of sufficient duration is undertaken,
most African-Americans will have a meaningful blood
pressure response to antihypertensive agents from any of
the commonly used drugs classes. It also seems prudent to
establish a lower therapeutic blood pressure goal [< 130/
85 mm Hg] because of the exceedingly high risk of morbid
and fatal pressure -related sequelae in hypertensive
African-Americans.

Central Adrenergic Inhibitors

Direct Vasodilators

FIGURE I Two agent drug

combinations for seven antihyperten-sive drug classes according
to their relative clinical utility.

to its maximum dose and with fewer side effects. Figure 1

shows useful therapeutic combinations of antihypertensive drugs.
THERAPEUTIC CONSIDERATIONS FOR
SPECIAL POPULATIONS
African-Americans

African-Americans have a premature onset of hypertension that is associated with a greater burden of pressurerelated target organ damage (i.e., microalbuminuria, left
ventricular hypertrophy, elevated serum creatinine) compared to whites. The clinician will not infrequently encounter severe hypertension (Stages 2 to 4) with concomitant
diabetes mellitus and other cardiovascular conditions (i.e.,
stroke) in the hypertensive African-American. Obesity,
particularly among African-American women, contributes
to the high frequency of salt sensitivity, an intermediate
blood pressure phenotype which correlates with higher antihypertensive drug requirements. Persuasive data indicate
that even modest reductions in dietary sodium intake
(<135 mmol/day] will augment the hypotensive effect of
virtually all antihypertensive drugs in all hypertensives,
though more so in African-Americans.
Diuretics will often be required to achieve blood pressure
normalization given the higher prevalence of renal insufficiency and reduced natriuretic capacity in AfricanAmericans. Nevertheless, African-Americans should be
viewed as individuals for whom the most appropriate antihypertensive drug therapy is chosen based on considerations (i.e., concomitant diseases, drug synergy/interactions,
renal function) unique to that individual. It is an outdated
concept that race should be the primary consideration when
selecting antihypertensive drug therapy for any individual.

Elderly

Although some clinicians have been hesitant to initiate

antihypertensive therapy in older individuals (s 60 years),
the totality of evidence from epidemiological studies and
clinical trials documents that older patients have a higher
risk for pressure-related clinical events than younger hypertensives at any given blood pressure level. Moreover,
drug-treated older hypertensives take their antihypertensive medication and comply with lifestyle modifications (i.e., weight loss, salt and alcohol restrictions) as well
or better than their younger counterparts. Elderly hypertensives actually manifest more impressive reductions
in pressure-related complications within a shorter time
frame compared to younger hypertensives. Concomitant
cardiovascular conditions (i.e., diabetes mellitus, vascular
disease, renal insufficiency), noncardiovascular chronic
conditions, and poverty are more common in older compared to younger hypertensives. In part because of impaired baroreceptor reflexes, older hypertensives are prone
to develop orthostatic hypotension and should therefore
be checked for postural hypotension. If present, then the
standing blood pressure should be used to guide therapeutic decisions. The clinician should have a heightened suspicion for the presence of orthostatic hypotension in older
hypertensives treated with alpha! antagonists, diuretics,
and/or central adrenergic inhibitors.
Renal Insufficiency

Hypertension is a risk factor for renal insufficiency, and

in turn, renal insufficiency irrespective of etiology causes
hypertension. Approximately 85% of individuals with renal
insufficiency also have hypertension. Diabetes mellitus and
hypertension account for over one half of all cases of endstage renal disease. The major therapeutic goal for individuals with hypertension associated with renal impairment
is to set and subsequently achieve an aggressive therapeutic
blood pressure goal [< 130/85 mm Hg] over many weeks
to months. The average number of drugs needed to achieve
blood pressure control in this hypertensive subgroup is 4.3;

513

79. Therapy of Hypertension

thus, therapeutic debates that focus on choices of a single

agent are mostly superfluous given that complex drug regimens are the rule. Limited renal natriuretic capacity, routine utilization of complex drug regimens, and the frequent
use of drugs (i.e., angiotensin-converting enzyme inhibitors), which can cause hyperkalemia, are important reasons
to use diuretics (loop diuretics or metolazone) in hypertensive patients with compromised renal function. Potassiumsparing diuretics and potassium supplements should be
used with considerable caution, if at all, in these patients,
particularly when angiotensin-converting enzyme inhibitors, angiotensin receptor antagonists, or nonsteroidal antiinflammatory drugs are concurrently prescribed. Angiotensin-converting enzyme inhibitors, and perhaps angiotensin
receptor antagonists may be particularly useful because in
addition to being antiproteinuric, they antagonize the renal
renin-angiotensin system which appears to be pathophysiologically linked to the characteristic structural and functional renal abnormalities found in experimental models
of renal injury (see Chapter 66). Because of the potentially
serious risk for hyperkalemia, these agents should, however, be used with caution in patients taking potassiumsparing diuretics or potassium supplements. Nonsteroidal
anti-inflammatory drugs also further augment the risk for
hyperkalemia. In hypertensives with renal impairment, simply lowering blood pressure can lead to a rise in creatinine
which, in most instances is followed by a fall in creatinine
to lower than pretreatment levels. Angiotensin-converting
enzyme inhibitors and perhaps angiotensin receptor antagonists can incite a precipitous decline in renal function as
evidenced by a rise in creatinine in patients with critical
bilateral renal stenosis or unilateral stenosis in a solitary
kidney.
Diabetics

Over 50% of diabetics are hypertensive. Diabetic hypertensives usually manifest disproportionate elevations in systolic compared to diastolic blood pressure and more often
than not will require complex multidrug regimens to
achieve blood pressure normalization. The long-term target
blood pressure should minimally be < 130/85 mm Hg in
this high risk group. Roughly one third of diabetics have
hyporeninemic hypoaldosteronsim (type IV renal tubular
acidosis), a condition which attenuates renal potassium secretion. Thus, diabetic hypertensives should be regularly
checked for hyperkalemia, particularly when potassiumsparing diuretics, potassium supplements, nonsteroidal
anti-inflammatory drugs, angiotensin converting enzyme
inhibitors, or angiotensin receptor antagonists are prescribed. If blood pressure normalization is to be achieved,
diuretics will often be required in complex drug regimens
[> 2 drugs]. Diuretics augment renal potassium secretion
because the reduction in plasma volume causes secondary
hyperaldosteronism. Several major hypertension drug
treatment trials (Hypertension Detection and Follow-up
Program and the Systolic Hypertension in the Elderly Program) have documented a similar reduction in risk for
pressure-related clinical events in diabetics and nondiabet-

ics when treated with diuretic-based regimens. Angiotensin-converting enzyme inhibitors have proved to be renoprotective in type I diabetics with proteinuria and thus
should be utilized in such patients even in the absence of
established hypertension. Though the risk of end-stage renal disease is much lower in type II diabetics (lifetime risk
8%) compared to type I diabetics, the former accounts
for the overwhelming majority of diabetes -related endstage renal disease cases. A reasonable clinical extrapolation is that angiotensin-converting enzyme inhibitors
should also be preferred antihypertensive agents in type
II diabetics with proteinuria. Angiotensin receptor antagonists, like the angiotensin-converting enzyme inhibitors,
are antiproteinuric, although long-term clinical outcome
studies are lacking with the former, ai antagonists, like the
angiotensin-converting enzyme inhibitors, improve insulin
sensitivity and additionally have a favorable impact on all
blood lipoprotein fractions. (See Chapter 29 for an in -depth
discusssion of treatment of the diabetic hypertensive.)
HYPERTENSIVE URGENCIES AND EMERGENCIES
Hypertensive Urgencies

A hypertensive urgency can be defined as a blood pres sure elevation in the absence of ongoing major pressurerelated symptoms that is high enough to engender concerns
regarding the new onset or worsening of pressure-related
target organ damage (i.e., congestive heart failure, new
or progressive renal insufficiency, neurological symptoms)
unless the blood pressure is lowered over the ensuing hours
to days. Hypertensive urgencies represent the most common indication for which acute blood pressure lowering
therapy is undertaken. In true hypertensive emergencies,
blood pressure levels usually exceed 210 mm Hg systolic or
130 mm Hg diastolic. A sizable proportion of hypertensives
treated acutely to lower blood pressure have only uncontrolled hypertension, not a legitimate hypertensive urgency.
In minimally symptomatic hypertensives with less severe
blood pressure elevations, the short-term risks of treatment
(Table 1) clearly exceed the potential benefits of immediate
blood pressure lowering. However, blood pressure levels
higher than the aforementioned cutpoints do not automatically warrant therapeutic intervention. Treatment when
indicated is usually via the oral route (Table 5). Oral therapy can be conveniently administered in a wide range of
clinical settings. However, the blood pressure response to
oral therapy is unpredictable. In special clinical situations
(i.e., postoperative period, intractable epistaxis), hypertensive urgencies may be most appropriately treated with intravenous medications.
Malignant/Accelerated Hypertension

The incidence of both malignant and accelerated hypertension has steadily declined, mainly because of increas ingly effective treatment of chronic hypertension. The clinical presentation of both entities can be quite dramatic,
leading to devastating clinical consequences including
death and serious long-term disability unless recognized

TABLE 5

Oral and Intravenous Antihypertensive Drugs Commonly Used to Treat Hypertensive Urgencies and Emergencies
Drug

Dose schedule

0.1 mg initially then 0.1 mg every

Oral
Clonidine 2 hr until a maximum dose of 0.7
mg reached

Duration
3-12 hr

Onset of action is within 30-45 min; peak BP lowering 3 to 4 hr postdose;

side effects include sedation, dry mouth, and orthostatic hypotension; can
dramatically lower BP in volume- depleted and older patients; CBF, CO,
pulse rate, and PVR all reduced; favor use when SNS tone is high (i.e.,
abrupt cessation of /3 -blockers, alcohol/drug withdrawal); avoid use in
CHF or in setting of greater than first degree heart block.

4-6 hr

Onset of action is 10-15 min; peak BP lowering occurs 1-1.5 hr postdose;

abrupt onset of action evokes SNS activation which can lead to reflex
tachycardia thus increasing myocardial oxygen demand with precipitation
or worsening of coronary ischemia; should be avoided in patients with
unstable angina pectoris, MI or aortic stenosis; only short -acting
formulation is useful in HTN urgencies.
Onset of BP lowering within 15-30 min after po dosing; peak BP lowering
occurs 1-2 hr postdose; BP lowering effect will likely be blunted in
setting of volume overload; however, rapid and excessive BP reductions
can occur in volume- depleted patients; preload and afterload are
decreased and CBF preserved; avoid use in bilateral renal artery stenosis
and in patients with a history of ACE inhibitor-induced angioedema.

<5 min

Rapid onset (seconds) and offset of action allow precise titration of BP

response; very potent; balanced preload and afterload reduction; increases
CO thus desirable in CHF; may, however, cause a "coronary steal"
syndrome and precipitate or worsen coronary ischemia; hypoxemia may
result from intrapulmonary shunting; activates SNS, thus avoid in aortic
dissection; raises TCP; thiocyanate and cyanide toxicity can occur.

<5 min

Onset and offset of action is within minutes; reduces preload and, at higher
doses, afterload; shunts blood flow into the subendocardium, an ischemiaprone area; intravenous agent of choice for treating severe hypertension
in setting of MI, unstable angina, or in individuals with known CHD; also
desirable in setting of pulmonary edema and CHF; occasionally causes
bradycardia; raises ICP; also may result in excessive BP lowering in
setting of volume depletion; causes methemoglobinemia rarely.

2-6 hr

BP response is variable thus precise titration not possible; onset of action is

rapid; however, offset of action takes hours; CBF is preserved; avoid use
in setting of bronchospastic lung disese or CHF; excessive BP lowering
can occur in setting of volume depletion; monitor for orthostatic
hypotension.

Nifedipine 5-10 po or -5.1., repeat after 1 hr

maximum total dose = 20 mg

6.25-25 mg po or chewed,
maximum total dose =
25 mg

Captopril

0.25 to 8 mg kg" min"

with titration every 5
min

Intravenous
Sodium
nitroprusside

Nitroglycerine

5-200 mg/min

>4 hr

Labetalol

Nicardipine

Enalapril

Furosemide

Bolus: 20-40 mg every

20-30 min continuous
infusion: 0.5-2.0 mg/
min max dose = 200300 mg/24 hr
2-4 mg/hr with titration
every 15 min; when goal
BP attained lower
infusion to 4 mg/hr
0.625-1.25 mg every 6 hr;
reduce initial dose by
50% in patients on
diuretics or if renally
impaired
May initiate treatment
with doses as low as 510 mg if renal function
is normal; 20-40 mg in
CHF; if renal function
abnormal may need to
dose much higher

Comment

Onset of action is within 15 min; (offset of action takes hours; avoid in

setting of ongoing coronary ischemia (MI, unstable angina pectoris) and in
aortic dissection, unless combined with /3 -blockade; maintains or
increases CO in CHF; potent dilatory effect on coronary arteries.

~6 hr

Onset of action is within 15 min; offset of action takes hours; reduces both
preload and afterload and dilates coronary vessels; thus is a desirable
agent in CHF and CHD patients with severe hypertension; CBF
maintained; avoid in bilateral renal artery stenosis and in patients with a
history of ACE inhibitor-induced angioedema; can result in excessive BP
lowering in volume- depleted pat ients.

~4.5 hr

Onset of action is within 15 min; initiates peripheral venous pooling

resulting in preload reduction followed by a diuresis, the latter being
maximal at 1.5 hr; favor when severe HTN is complicated by CHF,
pulmonary edema, and/or renal insufficiency; avoid in setting of
dehydration; use cautiously in conjunction with aminoglycosides because
of ototoxicity; need to monitor glucose and electrolytes, especially
potassium and magnesium. Can lower BP when used alone but mostly
used to potentiate other agents.

ACE, angiotensin converting enzyme; BP, blood pressure; CBF, cerebral blood flow; CHD, coronary heart disease; CHF, congestive
heart failure; CO, cardiac output; HTN, hypertensive; ICP, intracranial pressure; MI, myocardial infarction; PVR, peripheral vascular
resistance; SNS, sympathetic nervous system.
Note: A brief description of the pharmacokinetic, hemodynamic, and clinical indications and contraindications is provided for each drug.

79. Therapy of Hypertension

and treated appropriately. While these two hypertensive

emergencies can be distinguished on clinical grounds, they
portend a similar ominous prognosis. The 1-year survival
rate in untreated accelerated or malignant hypertension is
only 20%; however, 5-year survival rates are over 90% after
successful treatment. Blood pressure is usually higher than
200/130 mm Hg. Patients typically present with one or
more of the following: severe headaches, blurred vision,
nystagmus, focal neurological symptoms, nystagmus, positive plantar reflexes, congestive heart failure, and/or renal
failure. The fundoscopic examination routinely confirms
severe retinopathy (bilateral hemorrhages and exudates).
Papilledema is, however, required to make the diagnosis
of malignant hypertension. Abrupt cessation of either /8blockers or central adrenergic inhibitors can also lead to
a rebound in blood pressure to much higher than pretreatment levels and thus may present as a hypertensive emergency.
Patients with malignant or accelerated hypertension
should always be treated with intravenous medications
(Table 5) in a setting where close patient monitoring is
feasible (i.e., intensive care unit). Intravenous sodium
nitroprusside and nitroglycerin are the most commonly
used effective therapies for hypertensive emergencies.
The major advantage of these agents is that their onset
and offset of action occurs within seconds to minutes,
allowing for precise titration of the infusion according
to the blood pressure response. The therapeutic goal for
these patients is not blood pressure normalization but
rather a gradual reduction of mean arterial pressure [((2 X
diastolic blood pressure) + systolic blood pressure) /3] of
no more than 15 to 20%. Systolic and diastolic blood pres sure should not be lowered to < 170 and < 110 mm Hg,
respectively. Gradual and cautious blood pressure lowering will minimize the probability of iatrogenic target organ
ischemia as a consequence of vital organ hypoperfusion.
In special situations such as unstable angina pectoris, congestive heart failure, pulmonary edema, or aortic dis section a lower target blood pressure may be justified.
Secondary causes of hypertension such as critical renal
artery stenosis, glomerulonephritis, Cushing's syndrome
and pheochromocytoma should be sought in patients
presenting with malignant or accelerated hypertension.
Chronic Hypertension in the Hospital Setting

The clinician should have a relatively high threshold

for prescribing acute blood pressure lowering therapy,
even via the oral route, in asymptomatic to minimally
symptomatic hospitalized hypertensives with poorly controlled blood pressure. Potentially reversible causes for
elevated blood pressure should be considered such as
pain, hypoxia, hypercarbia, hypoglycemia, pulmonary
edema, and status epilepticus. Withdrawal from alcohol
or drugs or even rebound hypertension attributable to
prior abrupt discontinuation of either /3-blockers or central adrenergic inhibitors may underlie blood pressure
elevations in hospitalized patients. Infusion of saline
containing intravenous fluids should be minimized and

515

dietary sodium restricted to 2 (88 mmol) per day or less.

If no reversible cause of elevated blood pressure is
identified, and the blood pressure level is high enough,
then the approach to treatment is virtually identical
to that utilized in hypertensive urgencies. Hospitalized
patients are prone to orthostatic hypotension because of
prolonged bed rest. Thus, if feasible, blood pressure
should be periodically measured in both the seated or
supine and upright positions. Automated blood pressure
measurement devices should be regularly calibrated
against a mercury manometer to ensure their accuracy.
Most hospitalized patients will be optimally treated simply
with intensification of their ambulatory blood pressure
medication regimen. The tendency to titrate drugs every
day or so on hospital rounds can result in excessive
blood pressure lowering leading to side effects and/or
targetorgan ischemia. The ultimate therapeutic goal in
hospitalized patients with poor blood pressure control is
to keep blood pressure under levels that are likely to
incite target organ damage over the short term. Thus,
blood pressure normalization (< 140/90 mm Hg), per se, is
not the usual therapeutic goal. In minimally symptomatic
patients with elevated blood pressure, the pressure levels
triggering physician notification should be stated in their
admission orders. The blood pressure level should be
high enough (210/120 mm Hg) to minimize the likelihood
of unnecessary and potentially harmful therapy being
prescribed for acute blood pressure lowering.
Bibliography
Abdelwahab W, Frishman W, Landau A: Management of hypertensive urgencies and emergencies. Ther Rev 35:747-762,1995.
Calhoun DA, Oparil S: treatment of hypertensive crisis. N Engl
J Med 323(17): 1177 -1183, 1990
Flack JM, Neaton JD, Daniels B, et al.: Ethnicity and renal disease:
Lesson from multiple Risk factor intervention trial and the treat ment of mild hypertension study. Am J Kidney Dis 21 (4):
31-40, 1993.
Epstein M, Bakris G: Newer approaches to antihypertensive therapy. Use of fixed- dose combination therapy. Arch Intern Med
156:1969-1978, 1996.
Flack JM, McVeigh GE, Grimm RH Jr: Hypertension therapy in
the elderly. Curr Opin Nephrol 2: 386-394, 1993.
Flack JM, Mensah, McVeigh GE, et al.: Diagnosis, evaluation and
management of hypertension in an ambulatory clinic setting.
/ Clin Outcome Manage 2(5):1 -21, 1995.
Flack JM, Cushman W: Evidence of the efficacy of low-dose
diuretic monotherapy. Am J Med 101(Suppl. 3A):53S-60S,
1996.
Flack, JM, Neaton J, Grimm RH Jr, et al.: Blood pressure and
mortality among men with prior myocardial infarction. Circulation 92:2437-2445, 1995.
Flack JM, Novikov SV, Ferrario CM: Benefits of adherence to
anti-hypertensive drug therapy. Eur Heart J 17(a):16-20, 1996.
Grimm RH Jr, Flack JM, Grandits GA, et al.: Long-term Effects
on plasma lipids of diet and drugs to treat hypertension. JAMA
275:1549-1556, 1996.
Joint National Committee. The fifth report of the Joint National
Committee on Detection, Evaluation and Treatment of High
Blood Pressure (JNC V). Arch Intern Med 153:154-183,
1993.

516
Lewis CE, Grandits OA, Flack JM, et al.: Efficacy and tolerance
of antihypertensive treatment in men and women with stage 1

diastolic hypertension. Arch Intern Med 156:377-385, 1996.

Murphy C: Hypertensive emergencies. Emerg Med Clin North Am
13(4):973-1001, 1995.
Meredith PA. Therapeutic implications of drug holidays. Eur Heart
J 17 (Suppl A):21-24, 1996.

John M. Flack
National High Blood Pressure Education Program. 1995 Update
of the Working Group Reports on Chronic Renal Failure and

Renovascular Hypertension. Arch Intern Med 156:1928-1947,

1996.
Neaton JD, Grimm RH Jr, Prineas RJ, et al.: Treatment of mild
Hypertension Study (TOMHS) final results. Jo A M Ass
270:713-724, 1993.

INDEX

AA-amytoidosis, nephrotic syndrome

from, 139 ACDK, see Acquired
cystic disease of
the kidney
ACE inhibitors, see Angiotensinconverting enzyme inhibitors
Acetazolamide, 89 Acid-base balance in
elderly, 398 in infants and children, 386
Acid-base disorders acid-base map, 9394 acidemia, 92 acidosis, see Acidosis
alkalemia, 92 alkalosis, see Alkalosis
clinical conditions associated with, 95
compensatory responses, 93 diagnostic
approach anion gap calculations, 93, 96
arterial blood gas measurements, 96
laboratory values, 95-96 physical
examination, 95 steps involved in, 93
effect on extrarenal potassium
disposal, 99-100 in
HIV patients, 238-239
mixed
definition of, 93 diagnosis of, 96-97
respiratory disturbances, 92-93
simple, 93 Acidemia
description of, 92
diagnosis of, 96
Acidity, body fluid, regulation of
ammonium excretion, 18-19 basic
considerations, 17 bicarbonate
absorption, mechanisms of, 17-18
secretion, 18
excretion of hydrogen ions, 18 titratable
acidity, 18 Acidosis in acute renal
failure, management
approaches, 285
description of, 92
distal renal tubule, see Distal renal
tubular acidosis

metabolic, see Metabolic acidosis

renal tubular, see Renal tubular
acidosis
in uremic syndrome, 405 Acquired
cystic disease of the kidney,
372-373 Acquired
immunodeficiency syndrome,
renal manifestations of acidbase disorders, 238-239 acute
renal failure, 239 dialysis, 242
end stage renal disease, 242 fluid
and electrolyte abnormalities
hyperkalemia, 238 hypokalemia,
237-238 hyponatremia, 237
overview, 238
HIV-associated nephropathy
clinical course, 240-241 clinical
manifestations, 239
epidemiology, 239-240
pathogenesis, 240 pathology, 240
treatment, 240-241 immunemediated renal disease,
241-242 Acute azotemia
differential diagnosis, 191
hcptorenal syndrome, see
Hepatorenal syndrome
Acule glomerulonephritis
clinical manifestations, 128-129
pathology, 129 Acute interstitial
nephritis causes of
idiopathic, 281 immune disorders,
281 infections, 281 therapeutic
agents, 261-262 nonsteroidal antiinflammatory
drugs, 280, 292 overview,
279 penicillins, 279-280
rifampin, 280 sulfonamide
derivatives,
280-281
clinical course, 279 clinical
features, 277-279 description of,
278 laboratory findings, 278-279
517

pathogenesis, 277 treatment,

279 Acute lymphocytic
leukemia,
renal parenchymal infiltration
and, 228
Acute nephritic syndrome
definition of, 38-39
hematuria, see Hematuria
urine sediments in, 35 Acute
pyelonephritis, 51 Acute renal
failure causes
acute tubular necrosis, 254
endogenous toxins, 255
overview, 254
classification of, 247, 254-255
definition of, 247, 253 diagnostic
tests
fluids administration, 258
serum biochemistry, 257
urinalysis, 257 urine
electrolytes, 257 urine output
measurements,
257-258
differential diagnosis sodium and
water metabolism, 258 urinalysis,
258
using glomerular and tubular
function indices, 250-251 druginduced agents that cause acyclovir,
263 aminoglycosides, 262-263
amphotericin B, 263 antibiotics, 262264 cephalosporins, 263
immunosuppressants cyclosporine,
see Cyclosporine tacrolimus, 265
pentamidine, 263-264 radiocontrast,
264 vancomycin, 263 diagnosis, 260262 overview, 261 history -taking for,
255 in HIV patients, 239
inflammatory states, 248 intrinsic
description of, 254

518
Acute renal failure (continued)
therapeutic agents that cause,
261-262
lithium-induced, 333
management of acid- base
homeostasis, 285 dialysis, 286
drugs, 286-287
electrolyte homeostasis, 284-285
fluid intake, 283 guidelines, 287
nutritional approaches, 285-286
principles, 282 sodium intake, 283
uremia, 285
volume homeostasis, 282-283
metabolically induced
hypercalcemic nephropathy, 268
hyperphosphatemic nephropathy,
268-269
uric acid nephropathy, 231 clinical
outcome, 267 clinical
presentation, 266-267 definition
of, 266 diagnosis, 267
pathogenesis, 266 pathology, 266
treatment
for existing disease, 268 prophylactic
approaches, 267 nutritional therapy
energy requirements, 447
goals, 446
principles of, 446-447 water and
mineral intake, 447 obstructive
nephropathy, therapeutic
agents that cause, 262 oliguric, 247
pathophysiology cell morphologic
considerations cell injury,
pathophysiology of,
249-250
cellular fates
apoptosis, 247-248 necrosis, 247248 types of, 247 disruption of
cell cytoskeleton,
247
regeneration process, 248
stress response, 249
concentrating defect, 250
glomerular filtration rate
reductions, 250 peritoneal
dialysis, 421 physical examination
abdomen, 256 cardiovascular and
volume status,
256
extremities, 256-257
neuropsychiatric abnormalities, 257
ocular findings, 255-256 skin, 255
pigment nephropathy, see Pigment
nephropathy

Index
postrenal, 255
in pregnancy
etiology
acute fatty liver, 394 acute tubular
necrosis, 393-394 cortical necrosis,
393-394 hemolytic uremic syndrome,
394 obstruction, 393 preeclampsiaeclampsia, 393 risks associated with,
393 prerenal
description of, 254 drug-induced, 260261 renal biopsy, 259 renal imaging
Doppler scans, 258 magnetic
resonance imaging, 258 nuclear scans,
258 renal angiography, 258-259
ultrasound, 258 treatment, new
approaches derived
from pathophysiologic discoveries
inflammation modifications,
251-252
renal replacement therapy, 252
survival factors, 252 vasodilation, 251
vs. chronic renal failure, differential
diagnosis, 253-254 Acute
tubular necrosis, 191 acute renal
failure and, 254 during pregnancy,
393-394 therapeutic agents that
cause,
261-262 Acyclovir, acute renal
failure caused by,
263
Adenocarcinomas, of bladder, 375
ADH, see Antidiuretic hormone
ADPKD, see Autosomal dominant
polycystic kidney disease
Adrenocorticotropin hormone, elevated
levels in renal failure, 473-474
Afferent arterioles, 5 AfricanAmericans, hypertension treatment
considerations, 512 AG, see Anion
gap Aging
acid- base balance, 398 glomerular
filtration rate, 395-396 glucose
reabsorption via renal
tubules, 399 kidney
anatomy, 395 potassium
balance, 397 renal blood
flow, 395-396 renal changes,
396 sodium balance
low sodium diet, 397 volume
expansion, 397 tubular
function, 396 urinary
considerations concentration,
398 dilution, 398-399
/3 2 -Agonists, hypokalemia and, 102
AGT, see Alanine glyoxylate
aminotransferase

AIN, see Acute interstitial nephritis ALamyloidosis clinical features, 225

description of, 224 light-chain
deposition disease and,
comparison between, 225 nephrotic
syndrome from, 139 pathogenesis, 225
pathology, 224-225 prognosis, 225-226
treatment, 225-226 Alanine glyoxylate
aminotransferase glyoxylate conversion,
338 role in primary hyperoxaluria, 339
Albumin
elevated levels in diabetic
nephropathy patients description of,
217 treatment options, 219 mortality
indicator, in end stage renal
disease, 443
Alcoholic ketoacidosis, 78
Alcoholism, hypophosphatemia
associated with, 112
Aldosterone
drugs that inhibit, potassium disorders
associated with, 104 effect on
potassium excretion, 16, 98 excess
amounts, volume regulation
effects, 16 selective deficiency clinical
presentation, 75 description of, 74
spironolactone antagonism of, 117
Aldosteronism, hyperkalemia and, 101
Alkalemia, 92 Alkali therapy, renal
tubular acidosis,
74-75 Alkalosis description of, 92
hypophosphatemia and, 112 metabolic,
see Metabolic alkalosis respiratory, see
Respiratory alkalosis Alphaj
antagonists, for hypertension,
509-510
Alport's syndrome adult
form, 319 anti-GBM
nephritis, 319
biochemistry, 319
clinical features
nonrenal, 320-321 renal,
320
definition of, 318 -319 description
of, 128 diagnosis, 321 differential
diagnosis, 322 genetic etiology,
318-319 and Goodpasture's
disease, 177 juvenile form, 319
pathology, 319-320 progression
to end stage renal
disease, 320
related disorders
autosomal recessive, 322

Index
familial thin basement membrane
disease, 321-322
thrombocytopathy (Epstein's
syndrome), 322
treatment, 321 Aluminum,
110
clinical studies, 463-464 dialysis
dementia and, 463-464 dietary
sources, 463 osteomalacia, 453
Amikacin, 299, 305 Amiloride
for nephrogenic diabetes insipidus, 70
site of action, 117 Amino acids,
reabsorption, 19 e-Amino caproic acid,
310 Aminoglycosides, drug-induced
acute
renal failure and, 262-263
Amphotericin B, acute renal failure
caused by, 263
Amyloidosis
AA-amyloidosis, nephrotic syndrome
from, 139
AL-amyloidosis
clinical features, 225
description of, 224
light-chain deposition disease and,
comparison between, 225
pathogenesis, 225 pathology, 224-225
prognosis, 225-226 treatment, 225-226
dialysis-related, 406, 453-454 Analgesic
nephropathy caffeine and, 295
carcinoma rates, 297 description of, 293
diagnostic approach diagnostic criteria,
297 dosage ingested, 295 history-taking,
295-296 laboratory findings, 296-297
physical examination, 296 habitual use,
293-294 pathogenesis, 295 prevalence
of, 294-295 risk of, 294 treatment, 297
ANCA, see Antineutrophil cytoplasmic
autoantibodies
Anemia
in chronic renal failure, 437 clinical
manifestations, 466-467 laboratory
manifestations, 466-467
pathogenesis, 465-466 treatment
iron supplementation, 468-469
monitoring, 468 recombinant
erythropoietin,
467-469
in uremic syndrome, 403-404
Angiography
for acute renal failure, 258-259
computed tomographic, 50

519
diagnostic, 50 interventional, 50-51
Angiotensin, 3 Angiotensin converting enzyme
inhibitors
acute renal failure and, 261
clinical uses
cardiac anomalies, 457 congestive
heart failure, 186 diabetic
nephropathy, 219 focal segmental
glomerulosclerosis,
163
hypertension, 509
idiopathic membranous
nephropathy, 167-169 primary
IgA nephropathy, 174 progressive
renal disease, 435, 442 proteinuria,
45-46, 152 renal insufficiency, 431
contraindications during pregnancy,
390
Angiotensin I, 13
Angiotensin II
description of, 13
hyperkalemia and, 104
Anion gap
definition of, 93
normal values, 93
plasma
conditions that affect, 72 definition of,
71 mathematical expression, 71 ANP,
see Atrial natriuretic peptide
Antibiotics, see also specific drug druginduced acute renal failure and,
262-264 Antidiuretic
hormone, see also
Vasopressin description of, 14
lithium-induced effects, 332-333 role in
osmolarity regulation, 14 Antigenpresenting cells initiation of immune
response, 142 types of, 142 Antigens,
in in situ immune complex
disease
native, 144-145 planted,
145, 166
Antihypertensive drugs
advantages and disadvantages,
510-511
combination therapy, 511 -512
monotherapy
alpha] antagonists, 509-510
angiotensin -converting enzyme
inhibitors, 509-510
angiotensin receptor antagonists,
510-511
/3-blockers, 508-510 calcium
antagonists, 509-510 central
adrenergic inhibitors,
510-511
direct vasodilators, 510-511
diuretics, 508

Antineutrophil cytoplasmic
autoantibodies in pauci-immune
crescentic glomerulonephritis, 134
systemic vasculitides and diagnostic
testing, 205 pathogenesis, 202-204
Antineutrophil cytoplasmic autoantibody
glomerulonephritis, 134, 136-137
APC, see Antigen-presenting cells
Apparent volume of distribution, 299
Arcuate arteries, description of, 5
ARF, see Acute renal failure Arginine
vasopressin, effect on water
reabsorption, 185 ARPKD, see
Autosomal recessive
polycystic kidney disease Arterial
disease, 51 Arteries, age -related
changes, 396 Arteriovenous fistula, for
hemodialysis,
412-413, 439
Arteritis giant
cell
clinical features, 202, 204 renal
pathology, 200 treatment, 205
Takayasu's clinical features,
202 renal pathology, 200
treatment, 205 Ascites, 61, 189
Aspirin, habitual use effects, 294
Asymptomatic hematuria
definition of, 127 in IgA
nephropathy, 172 renal biopsy,
127-128 Atenolol, 305
Atheroembolic renal disease,
cholesterol clinical features,
271 clinical outcome, 272
definition of, 269 diagnosis, 271272 differential diagnosis, 271272 pathogenesis, 271 pathology,
269-271 risk factors, 271
treatment, 272
ATN, see Acute tubular necrosis Atrial
natriuretic peptide description of, 13
renal effects, 185-186 sodium
regulation, 13 Autoantibodies, 209
Autoimmunity, 143 Autosomal
dominant polycystic kidney
disease
clinical course, 316 clinical
manifestations, 315-317
diagnosis, 314-315 incidence, 314
intracranial aneurysms, 317
nephrolithiasis findings, 316
pathology, 314

520
Autosomal recessive polycystic kidney
disease
clinical course, 314
clinical features, 314
diagnosis, 313-314
differential diagnosis, 314
incidence, 313
Azathioprine
considerations during pregnancy, 393
toxicities, 485
Azlocillin, 299
Azotemia, prerenal, see also Uremic
syndrome
acute renal failure and, 254
fractional excretion of sodium
measurements, 257
physical examination, 256
Aztreonam, 299

B
Bartter's syndrome, 82, 102 Beer
potomania, 61 Behcet's syndrome,
214 Bence Jones proteins, 220 -221,
349 Bicarbonate
absorption, mechanisms of, 17-18
adaptive decreases for respiratory
alkalosis, 90 adaptive
increases for respiratory
acidosis, 86-87
clinical uses
diabetic kctoacidosis, 77 lactic
acidosis, 77 decreased levels, see
Metabolic
acidosis increased levels, see
Metabolic
alkalosis
regeneration of, 73 secretion, 18
Bilirubin, 29 Bioavailability, 299
Biological response modifiers, 230
Biopsy, renal
complications, 139 -140
contraindications, 139
indications
acute renal failure diagnosis, 259
focal segmental glomerulosclerosis,
162
lupus nephritis, 195-196 overview,
139, 208 poststreptococcal
glomerulonephritis,
195-196
methods, 140-141
Black water fever, 274
Bladder neoplasms
diagnosis, 375-376
etiology, 375
nonepithelial, 377
pathology, 375 staging,
376 treatment, 376

Index
j8-blockers
clinical uses
coronary artery disease, 458
hypertension, 508-510
description of, 508
hyperkalemia associated with, 104
receptors, 508
Blood flow, renal
age-related changes, 395-396
cardiac performance and, 184-185
effect of nonsteroidal antiinflammatory drugs, 291-292
during pregnancy, 388
Blood pressure
baseline for hypertension treatment,
506-507
classification of, 497
formulaic expression of, 498
high levels, see Hypertension
in infants and children, 386-387
regulatory mechanisms, 491-492
Blood urea nitrogen
acute renal failure diagnosis using, 257
creatinine ratio, 25
definition of, 23
glomerular filtration rate
measurements, 23 BNP, see Brain
natriuretic peptide Body fluid
composition, 3-4 Bone marrow
transplantation, renal
complications of, 230-231
Bowman's capsule
description of, 4
function of, 6
Brain, adaptations to hypernatremia, 69
Brain natriuretic peptide, renal effects,
185-186
Bumetanide, 116 BUN, see
Blood urea nitrogen

c
Calcitriol metabolism problems in
chronic renal
disease, 437 renal
synthesis of, 449 Calcium
antagonists, for hypertension, 509-510
distribution, 106
effect on parathyroid hormone
secretion, 450 excretion distal
convoluted tubule diuretics,
117
in infants and children, 385 renal
handling, 107 homeostasis
disorders
hypercalcemia, see
Hypercalcemia
hypocalcemia, see Hypocalcemia
method that regulate, 106-107 in
infants and children, 385

intestinal absorption, 107

malabsorption, in secondary
hyperparathyroidism, 449 serum
concentration description of, 106
regulation of, 107 Calcium channel
blockers, 458 Calculi, renal, see
Stones, renal cAMP, 14 Cancer
bladder, see Bladder neoplasm
chemotherapeutic agents, see
Chemotherapeutic agents
hypocalcemia and, 108 imaging
procedures, 51-52 pelvic, see
Pelvic tumors prostate, see
Prostate cancer renal and urologic
complications
associated with disease process
glomerular disease, 227 renal
parenchyma! infiltration,
228
urinary tract obstruction, 228
treatment
bone marrow transplantation,
230-231
genitourinary hemorrhage, 228
hemorrhagic cystitis, 228 high dose chemotherapy, 230-231
nephrotoxic agents
biological response modifiers,
230
cisplatin, 229 cyclophosphamide, 229
methotrexate, 230 mithramycin, 230
mitomycin C, 230 semustine, 230
streptozotocin, 229 radiation
nephropathy, 231 total body
irradiation, 230-231 tumor lysis, 231
renal transplantation
contraindications, 477 staging
of, see specific cancer ureteral, see
Ureteral tumors CAPD, see
Continuous ambulatory
peritoneal dialysis
Capillary beds
glomerular, 5-6
peritubular, 5-6
Captopril, 514
Carbenicillin, 305
Carbon dioxide
deficit, see Respiratory alkalosis
excess, see Respiratory acidosis
Carbonic anhydrase inhibitors
clinical uses, 116 mechanism of
action, 115-116 Cardiac disease, in
renal failure cardiomyopathy
description of, 456-457
management of, 458

Index
congestive heart failure, see
Congestive heart failure
coronary artery disease
management of, 458 prognosis,
456 ischemic heart disease, 457
overview, 455 pericarditis, 458
prevalence of, 455-456
prognosis, 456 risk factors
description of, 456 management of,
457-458 treatment, 457-458
Cardiac filling volume, 491
Cardiomyopathy, in renal failure
description of, 456-457
management, 458 Cast
nephropathy clinical features, 222
pathogenesis, 222 pathology, 221
prognosis, 222-223 treatment, 222223 Casts, in urine granular, 33
hyaline, 33 red blood cell, 33
tubular cell, 33, 35 waxy, 33
white blood cell, 33 Catecholamines,
effect on extrarenal
potassium disposal, 99
Catheters
for hemodialysis, 412 for peritoneal
dialysis, 417-418 CAVH, see
Continuous arteriovenous
hemofiltration C5b-9
membrane attack complex,
147-148 CCPD, see Continuous
cycle peritoneal
dialysis CD diuretics, see
Collecting duct,
diuretics Cefalexin, 305 Cefazolin, 299
Cefotaxime, 299 Cefoxitin, 299
Ceftazidime, 299, 305 Cefuroxime, 299,
305 Cell-mediated immunity
description of, 142-143 in
glomerulonephritis cellular sources,
147 complement activation, 147-148
mediator types, 146-147 Central
adrenergic inhibitors, for
hypertension, 510 -511
Cephalosporins, acute renal failure
caused by, 263
Chemotherapeutic agents,
nephrotoxicity of biological
response modifiers, 230 cisplatin, 229

521
cyclophosphamide, 229 methotrexate,
230 mithramycin, 230 mitomycin C, 230
semustine, 230 streptozotocin, 229
Chemotherapy for bladder neoplasms,
376 high -dose, renal complications of,
230-231
CHF, see Congestive heart failure
Children, renal considerations for
anatomical development, 383 functional
development acid- base balance, 386
blood pressure, 386-387 calcium
metabolism and excretion,
385
glomerular filtration rate, 383
oliguria, 385
phosphate concentrations, 385
potassium levels, 385-386
sodium metabolism, 384-385
urine concentration, 383 uremia,
472
Chlorambucil, clinical uses idiopathic
membranous nephropathy,
168
minimal change nephropathy, 151
Chloride
decreased levels, see Hypochloremia
urinary concentration levels, and
metabolic alkalosis, 85 Cholesterol
atheroembolic renal disease clinical
features, 271 clinical outcome, 272
definition of, 269 diagnosis, 271-272
differential diagnosis, 271-272
pathogenesis, 271 pathology, 269-271
risk factors, 271 treatment, 272 Chronic
glomerulonephritis, urine
sediments in, 35 Chronic
interstitial nephritis lead and, 335
lithium-induced, 334 Chronic lead
nephropathy, 336 Chronic renal
disease complications
acid- base metabolism, 437
anemia, 437 lipid metabolism,
438 metabolism, 437-438
potassium excretion, 437
psychological factors, 438 sleep
disorders, 438 sodium balance,
436-437 uremic coagulopathy,
438 water excretion, 437 drug
therapy, 438 effect on pregnancy,
390-391 stages of, 433

surgical considerations, 438

treatment, 433-435 Chronic
renal failure cardiac disease in
cardiomyopathy
description of, 456-457
management of, 458 coronary
artery disease management of,
458 prognosis, 456 ischemic
heart disease, 457 overview,
455 pericarditis, 458
prevalence of, 455-456
prognosis, 456 risk factors
description of, 456 management
of, 457-458 treatment, 457-458
dietary requirements energy,
444-445 protein, 444
endocrine manifestations adrenal
medulla, 475 carbohydrate
metabolism, 474-475
catecholamine levels, 475
glucocorticoids, 473-474
gonadotropins in females, 474 in
males, 474
growth hormone levels, 472-473
lipid abnormalities, 475-476
mineralocorticoids, 475 prolactin, 473
thyroid abnormalities, 472
extracorporeal therapeutic modalities
description of, 408 hemodialysis, see
Hemodialysis hemofiltration, see
Hemofiltration from habitual
analgesic use, 294 hematologic
manifestations anemia, 437
clinical manifestations, 466-467
laboratory manifestations,
466-467
pathogenesis, 465-466
treatment
iron supplementation, 468 -469
monitoring, 468 recombinant
erythropoietin,
467-469
erythrocytosis, 469-470 leukocyte
dysfunction, 470 -471 platelet
dysfunction, 470 -471 hemodialysis,
see Hemodialysis hemofiltration, see
Hemofiltration hyperlipidemia, 445
hyperparathyroidism, see
Hyperparathyroidism minerals and
trace element requirements, 445-446
neurological manifestations
peripheral neuropathy
autonomic and cranial nerve
dysfunction, 462

522
Chronic renal failure (continued)
classification of, 461 prevalence in
end stage renal
disease, 461
symptoms, 461
treatment, 462
uremic toxin etiology, 461-462 uremic
encephalopathy, 404 clinical
manifestations, 460 definition of, 459
differential diagnosis, 459-460
pathogenesis, 460-461 nutrition
assessment strategies, 442
goals of, 440 low-protein
diets
adequacy monitoring, 442-443
compliance monitoring, 442-443
effect on renal disease
progression, 441-442
spontaneous protein intake
decreases, 441
osteodystrophies associated with
adynamic bone lesions, 453
dialysis-related amyloidosis,
453-454
osteitis fibrosa, see Osteitis fibrosa
osteomalacia, see Osteomalacia
progression to end stage renal
disease, see End stage renal disease
protein
levels in healthy patients vs. chronic
renal failure patients, 440-441
monitoring of, 442-443 spontaneous
intake considerations, 441 therapeutic
agents that cause, 262 uremic
syndrome biochemical changes drug
protein binding, 405 enzymatic and
metabolic
dysfunction, 405 clinical
features acidosis, 405 anemia, 403404 cardiovascular anomalies, 403
coagulation disturbances, 404
endocrine dysfunction, 404 immune
deficiency, 404 malnutrition, 405
peripheral neuropathy, 403 pruritus,
405 uremic bone disease, 405
uremic encephalopathy, 403
definition of, 403 solute retention
factors that influence, 407
hippuric acid, 406 indoxyl sulfate,
406 methylguanidine, 406 organic
phosphates, 407 peptides, 406407 urea, 405-406 vitamin
requirements, 446

Index
Churg- Strauss syndrome clinical
features, 200, 202 treatment, 206
Chvostek's sign, 107 Cimetidine,
effect on creatinine
secretion, 24 Ciprofloxacin, 299
Circulating immune complexes in
IgA nephropathy, 173 in
membranoproliferative
glomerulonephritis, 156 in
renal disease, 145-146
Cirrhosis
ascites, management of, 189
renal sodium handling
abnormalities in clinical features, 188
efferent factors, 189 pathogenesis, 188189 renal water handling abnormalities
in clinical features, 189-190
hyponatremia treatment, 190
pathogenesis, 190 Cisplatin, 229, 305
CK, see Creatine phosphokinase
Clavulanic acid, 299 Clearance formula
creatinine, see Creatinine clearance
definition of, 7 Clindamycin, 299
Clonidine, 514 C3 nephritic factor, 148
Coagulation, uremic syndrome effects,
404
Collecting duct
antidiuretic hormone effects, 14
cell types, 10 description of, 4
diuretics
complications of, 119 mechanism
of action, 117 types of, 117
function of, 73 illustration of, 12
Colon, potassium excretion, 99
Complement activation
in idiopathic membranoproliferative
glomerulonephritis, 157 in
renal injury, 147-148
Computed tomography
clinical uses
arterial disease, 51
cysts, 52
inflammatory diseases, 51
renal vein lesions, 53
trauma, 52
ureteral obstruction, 52
urolithiasis, 52 indications, 48
Concentration, urine
in children and infants, 383
in chronic renal disease, 437
in dehydration, 14 in
elderly, 398

in normal functioning, 14 in obstructive

uropathy patients, 352 specific gravity
measurements, 27 in tubular dysfunction
patients, 310 and vasopressin secretion,
64 Congestive heart failure, see also
Cardiac disease with chronic renal
failure, treatment
implications, 187 incidence of, 183
medical therapy for, renal effects of,
186-187
sodium and water retention afferent
mechanisms, 183-184 efferent
mechanisms hormonal vasodilator
activation,
185-186 neurohormonal
vasoconstrictor
activation, 185 renal
hemodynamic changes,
184-185 Continuous
ambulatory peritoneal
dialysis
description of, 416 drug dosing
considerations for,
303-304
exchange test, 417
Continuous arteriovenous hemofiltration
description of, 409-410 indications,
413
Continuous cycle peritoneal dialysis, 416
Continuous renal replacement therapy,
drug dosing considerations in, 304-305
Continuous venovenous hemofiltration
description of, 409-410 indications, 413
Contrast nephropathy, in intravenous
urography, 47-48
Coronary artery disease, in renal failure
management of, 458 prognosis, 456
Cortical necrosis, during pregnancy,
393-394
Corticosteroids
clinical uses
idiopathic membranoproliferative
glomerulonephritis, 159
idiopathic membranous
nephropathy, 168 minimal
change nephropathy,
151-152
systemic lupus erythematosus, 3 systemic
vasculitides, 206 toxicities, 485
Countercurrent multiplier description of,
15 role in body fluid osmolarity
regulation, 15 urea
effects, 15-16 Creatine
phosphokinase, in
rhabdomyolysis, 273
Creatinine dietary sources,
24-25

523

Index

food restriction effect on, 25

properties, 24-25 renal handling
of, 24 secretion, drugs that inhibit,
24 Creatinine clearance age-related
changes, 396 in children and
infants, 383-384 in elderly, 396
estimation of, from serum
creatinine, 23
formula for, 22
glomerular filtrate clearance and, 7
glomerular filtration rate
measurements, 22-23, 301
lead effects, 335
Crescentic glomerulonephritis, 129
CRRT, see Continuous renal
replacement therapy
Cryoglobulinemia classification of,
156 clinical manifestations, 158
glomerular deposits, 156
progression of secondary
membranoproliferative
glomerulonephritis, 156-157
treatment, 159
Cryoglobulinemic vasculitis, 137
Cryoglobulins, 158 Crystals cystine,
35 in urine, 34-35
CT, see Computed tomogr aphy
Cutaneous leukocytoclastic angiitis, 202
Cyclic adenosine monophosphate, see
cAMP
Cyclophosphamide
clinical uses
Goodpasture's syndrome, 178
idiopathic membranous
nephropathy, 168 minimal
change nephropathy, 151 systemic
lupus erythematosus, 210
complications, 40 contraindications
during pregnancy,
392
nephrotoxicity, 229
Cyclosporine clinical
uses
idiopathic membranous
nephropathy, 168 minimal
change nephropathy, 152
considerations during pregnancy, 393
fetal effects, 393 hyperkalemia
induced by, 104 nephrotoxicity of
acute renal failure, 264-265 vs.
acute rejection in transplant
patients, 264-265
pharmacokinetics, 265
Cystic diseases acquired,
317-318 congenital, 313
hereditary
polycystic kidney disease

autosomal dominant clinical course, 316

clinical manifestations, 315-317
diagnosis, 314-315 incidence, 314
intracranial aneurysms, 317
nephrolithiasis findings, 316 pathology,
314 autosomal recessive clinical course,
314 clinical features, 314 diagnosis, 313314 differential diagnosis, 314 incidence,
313 tuberous sclerosis, 317 Von HippelLindau disease, 317 syndromic, 313
Cystitis, acute, see also Urinary tract
infection
antimicrobial prophylaxis, 369
bacterial etiology, 366 clinical
manifestations, 367 differential
diagnosis, 367 incidence, 366
prognosis, 369 treatment, 367
Cysts, imaging techniques, 52
Cytomegalovirus
contraindications for renal
transplantation, 478 in
immunosuppressed patient, 485

DCT, see Distal convoluted tubule

DDAVP, see Desmopressin DOS, see
Dialysis disequilibrium
syndrome Deferoxamine, 464
Dehydration testing, for polyuria in
diabetes insipidus patients, 68
Dementia, dialysis
aluminum and, 463-464
electroencephalographic findings, 463
etiology, 463
Desmopressin, 470
Diabetes insipidus
diagnosis of, 67
forms of, 66
nephrogenic, 66-67
pure water deficits associated with,
66-67
treatment, 70
Diabetes mellitus
hypertension treatment
considerations, 513
microalbuminia detection, 43
Diabetic ketoacidosis
bicarbonate use, 77
diagnosis of, 77
hormonal profile, 77
hypophosphatemia and, 112
potassium deficiencies, 100-101
treatment, 77

Diabetic nephropathy description of,

215 effect of pregnancy, 392
epidemiology, 215 natural history
functional changes, 216 illustration
of, 216 microalbuminia, 217
pathology, 217 pathogenesis, 215-216
progressive disease, treatment
clinical studies, 218-219 modalities,
219 screening, 217-218 transition to
end stage renal disease,
treatment options, 219-220
Dialysate, 411-412, 416 Dialysis,
see also Uremia amyloidosis, 406,
453-454 cardiac-related
complications cardiac failure, 455456 valvular dysfunction, 456
description of, 408 drug therapy
considerations
continuous ambulatory peritoneal
dialysis, 303-304 hemodialysis,
303 gonadotropic effects, 474
hemodialysis, see Hemodialysis for
hepatorenal syndrome, 192
hypocalcemia associated with, 110
indications
acute renal failure, 286 cast
nephropathy, 223 end stage renal
disease, 439 HIV patients, 242
malnutrition associated with, 443
morbidity, 425-426 neurological
complications dialysis dementia, 463
dialysis disequilibrium syndrome,
462-463 intellectual function
impairments,
462
nutritional assessments during, 443
peritoneal, see Peritoneal dialysis
during pregnancy, 392 Dialysis
disequilibrium syndrome,
462-463
Dialysis membranes
description of, 408-409 solute
clearance, 411 ultrafiltration
rate, 408-409 water transport,
410-411 Diet
low-protein
adequacy monitoring, 442-443
compliance monitoring, 442-443
effect on renal disease progression,
441-442 spontaneous
protein intake
decreases, 441
as treatment for progression to end
stage renal disease, 431-432

524
1,25-Dihydroxyvitamin D3 , 3
Dipstick strips
for determining chemical properties
of urine
bilirubin, 29
blood, 28-29
glucose, 29
ketones, 29
leukocytes, 29
nitrite, 29
pH, 28
protein, 28
specific gravity, 27-28
urobilinogen, 29 proteinuria
evaluation, 44 Distal
convoluted tubule description
of, 4 diuretics
absorption, 122
description of, 116-117
resistance, 122-123 function of, 9-10
Distal nephron, 9-11 Distal renal
tubular acidosis classification of, 73
collecting tubule acidification defects
that cause, 73-74
diagnostic approach, 75-76
hyperkalemic, 74 Diuretics
absorption, 121-122
adaptation to, 118-119
clinical uses
gross hematuria, 310
hypertension, 492 -493
collecting duct, 117
combination therapy
with angiotensin -converting enzyme
inhibitors, 46
with other diuretics, 122-123
complications of, 119-120 distal
convoluted tubule, 116-117 for
edema, 114, 120 effect on potassium
excretion, 17, 98 hypercalcemia and,
110 loop
absorption, 122
adaptation to, 118-119
complications of, 119
continuous infusion, 123
hemodynamic effects, 116
indications
ascites, 189
edema, 120
mechanism of action, 116
resistance, 122-123
types of, 115-116
mechanism of action, 114
osmotic, 117 -118 postdiuretic
sodium chloride
retention, 118 potassium
disorders caused by
hyperkalemia, 103
hypokalemia, 102

Index
proximal tubule, 114-116 resistance
causes of, 120-121 in congestive heart
failure patients,
186
in nephrotic syndrome patients, 122
treatment approaches, 122-123
thiazide clinical uses
hypertension, 508
hypothalamic diabetes
insipidus, 70 for hypercalciuria,
363 metabolic alkalosis and, 82 Donors,
for renal transplantation cadaveric, 479480 contraindications, 481 living, 480481 Dopamine, 187 Doppler scan, for
acute renal failure,
258
Drug elimination, effect of renal failure
hepatic metabolism, 300 renal excretion,
301 renal metabolism, 300-301 Drug
therapy, see also specific drug dialysis
considerations continuous ambulatory
peritoneal
dialysis, 303-304
hemodialysis, 303 renal failure,
considerations for absorption
effects, 299-300 classification of,
298-299 distribution effects, 300
dosage adjustments, 301-303
elimination effects
hepatic metabolism, 300 renal
excretion, 301 renal metabolism,
300-301 metabolite accumulation,
301-302 overview, 298
pharmacodynamics effects, 301
pharmacokinetic effects, 298-299

EACA, see e- Amino caproic acid

ECF, see Extracellular fluid Edema,
61 diuretic treatment
considerations before initiating
therapy, 120
diuretic selection and dosage, 120
resistance, 120-121 physiological basis,
114, 120 Efferent arteriole, 5 EHEC,
see Enterohemorrhagic
Escherichia coli
Elderly
acid- base balance, 398 glomerular
filtration rate, 395-396 glucose
reabsorption via renal tubules, 399

hypertension treatment
considerations, 512 kidney
anatomy, 395 potassium
balance, 397 renal blood
flow, 395-396 renal
changes, 396 sodium
balance
low sodium diet, 397 volume
expansion, 397 tubular function, 396
urinary considerations concentration,
398 dilution, 398-399 Electrolytes
calcium, see Calcium disorders, see
specific disorder magnesium, see
Magnesium obstructive uropathy
effects, 352 phosphorus, see
Phosphorus potassium, see Potassium
regulation of, 3 sodium, see Sodium
Electroneutrality, 71 Enalapril, 514
Endocrine manifestations, of chronic
renal failure adrenal medulla, 475
carbohydrate metabolism, 474-475
catecholamine levels, 475
glucocorticoids, 473-474
gonadotropins in females, 474 in
males, 474
growth hormone levels, 472-473 lipid
abnormalities, 475-476
mineralocorticoids, 475 prolactin, 473
thyroid abnormalities, 472 Endocrine
system, uremic syndrome
effects, 404-405 End stage renal
disease Alport's syndrome
progression, 320 cardiac disease
effects, see Cardiac
disease
costs associated with, 427-428
diabetic nephropathy progression
description of, 215 treatment
options, 219-220 focal segmental
glomerulosclerosis
progression, 162
glomerulonephritis progression, 139
from habitual analgesic use, 294 in
HIV patients, 242 IgA nephropathy
progression, 175 morbidity, 425426 mortality causes, 422
factors associated with
comorbidity, 422-425
demographics, 422 modality,
425 socioeconomic status,
425 treatment parameters,
425 nutritional status and,
443

Index
nutritional assessment, 443
peripheral neuropathy in, 461
progression from renal disease
algorithm, 434 course of, 429
mechanisms, 429-430 methods to
forestall, 435-436 treatment
blood pressure control, 431 dietary
approaches, 431-432 quality of life,
426-427 rehabilitation, 427 standardized
mortality ratio, 425 technique survival,
426 therapeutic preparations, 439
Enterohemorrhagic Escherichia coli, 232
Epinephrine, 99 Erythrocytes, 29, 32
Erythrocytosis, 469-470 Erythromycin,
299 Erythropoietin definition of, 3
hypoxia effects, 465 recombinant, for
anemia of chronic
renal failure, 467-469 site of
production, 465 ESRD, see End stage
renal disease Essential cryoglobulinemic
vasculitis characteristics of, 202
treatment, 206 Essential hypertension
atypical features, 500 cardiovascular
risk, 496-497 clinical presentation, 498
definition of, 496 pathophysiology, 498
prevalence, 496 screening for, 498-500
treatment, 500
vascular regulatio n abnormalities, 494
ESWL, see Extracorporeal shock wave
lithotripsy
Ethacrynic acid, 116 Ethylene
glycol, ingestion of, 79 Euvolemia
description of, 58 syndrome of
inappropriate antidiuresis etiology, see
Syndrome of inappropriate antidiuresis
Excretory urography, see Intravenous
urography
Extracellular fluid
decreased volume, see Hypovolemia
normal volume, see Euvolemia
volume
diuretic therapy for, see Diuretics
expansion of, 114 sodium effects, 13
Extracorporeal shock wave lithotripsy,
for renal stones, 365
Extracorporeal therapy, for chronic
renal failure
continuous arteriovenous
hemofiltration, 409-410, 413

525
continuous venovenous
hemofiltration, 409-410, 413
hemodialysis, see Hemodialysis
hemofiltration, see Hemofiltration

Familial hypocalciuric hypercalcemia,

110 Familial hypokalemic periodic
paralysis,
102
Familial Mediterranean fever, 214
Familial renal diseases Alport's
syndrome adult form, 319 anti-GBM
nephritis, 319 biochemistry, 319
clinical features nonrenal, 320-321
renal, 320
definition of, 318 -319 diagnosis,
321 differential diagnosis, 322
genetic etiology, 318-319
juvenile form, 319 pathology,
319-320 progression to end
stage renal
disease, 320
related disorders
autosomal recessive, 322 familial
thin basement membrane
disease, 321-322
thrombocytopathy (Epstein's
syndrome), 322 treatment, 321
polycystic kidney disease autosomal
dominant clinical course, 316 clinical
manifestations, 315-317 diagnosis, 314315 incidence, 314 intracranial
aneurysms, 317 nephrolithiasis findings,
316 pathology, 314 autosomal recessive
clinical course, 314 clinical features, 314
diagnosis, 313-314 differential diagnosis,
314 incidence, 313 tuberous sclerosis,
317 Von Hippel-Lindau disease, 317
Familial thin basement membrane
disease, in Alport's syndrome, 321-322
FEK, see Fractional excretion of
potassium
FENa, see Fractional excretion of sodium
Fibrillary glomerulonephritis, 226
Fibrogenesis, 331 Filtrate, glomerular
absorption, 7-8 formation, 6-7
secretion, 7-8

Filtration markers
glomerular filtration rate
measurements, 23-24 inulin, see Inulin
Filtration rate, glomerular age-related
changes, 395-396 bicarbonate
increases, 81-82 in children and
infants, 383 description of, 20-21
determinants, 7, 21, 49, 250 in diabetic
nephropathy patients, 217 effect of
nonsteroidal anti-inflammatory drugs,
291 estimation of, 301 factors that
affect, 7 formula, 6, 9
in juvenile nephronophthisis
medullary cystic disease
complex, 324 measurement
conditions that limit creatinine,
errors associated wit h
measurement, 25 metabolism, 2425 renal handling, 24 urea, errors
associated with metabolism, 26
renal handling, 25-26 methods
blood urea nitrogen levels, 23
creatinine clearance description
of, 22-23 from serum creatinine,
23 exogenous filtration markers,
23-24
renal clearance, 21 serum creatinine,
22 urea clearance, 23 in membranous
nephropathy, 167 nonsteroidal antiinflammatory drug
reduction of, 46 normal range,
21 obstructive uropathy effects, 352
plasma solute concentrations and,
21-22
reductions in acute renal failure, 250
sodium retention, 189 Fluconazole, 305
Flucytosine, 305 Focal segmental
glomerulonephritis, in
sickle cell disease, 312 Focal
segmental glomerulosclerosis clinical
features, 162 description of, 160-161
diagnosis, 162 etiologies of, 160-161
idiopathic
clinical features, 162 description
of, 161 pathogenesis, 162 in
renal transplantation, 163
treatment, 163 pathogenesis,
161-162 pathology, 161

526
Focal segmental glomerulosclerosis
(continued) progression to end
stage renal
disease, 162-163
proteinuria, 160, 162
treatment, 162-163
variants, 161
Fractional excretion of potassium
definition of, 100 mathematical
formula, 100 Fractional excretion
of sodium in acute interstitial
nephritis, 279 in infants and
children, 384-385 maintenance of,
114 prerenal vs. intrinsic acute
renal
failure diagnosis using, 251, 257
Fresh frozen plasma, for Goodpasture's
syndrome, 178 FSGS, see
Focal segmental
glomerulosclerosis Functional
proteinuria, 43 Furosemide, 514
absorption, 121-122 for congestive
heart failure, renal
effects, 186
description of, 116

G
Ganciclovir, 305
GBM, see Glomerular basement
membrane Gentamicin, 299, 305
GFR, see Glomerular filtration rate
GHP, see Glucoheptonate Giant cell
arteritis
clinical features, 202, 204
renal pathology, 200
treatment, 205
Gitelman's syndrome, 82, 102
Glomerular basement membrane
in Alport's syndrome, 319
description of, 42
diabetic nephropathy-associated
changes, 217
permeability, 42 Glomerular capillary
tuft, 4 Glomerular disease, from cancer,
227 Glomerular filtration barrier
description of, 6
permeability, 6
Glomerular filtration rate
age-related changes, 395-396
bicarbonate increases, 81-82
in children and infants, 383
description of, 20-21
determinants, 7, 21, 49, 250
in diabetic nephropathy patients, 217
effect of nonsteroidal antiinflammatory drugs, 291
estimation of, 301
factors that affect, 7
formula, 6, 9

Index
in juvenile nephronophthisis
medullary cystic disease
complex, 324 measurement
conditions that limit
creatinine, errors associated with
measurement, 25 metabolism, 2425 renal handling, 24 urea, errors
associated with metabolism, 26
renal handling, 25-26 methods
blood urea nitrogen levels, 23
creatinine clearance description
of, 22-23 from serum creatinine,
23 exogenous filtration markers,
23-24
renal clearance, 21 serum creatinine,
22 urea clearance, 23 in membranous
nephropathy, 167 nonsteroidal antiinflammatory drug
reduction of, 46 normal range,
21 obstructive uropathy effects, 352
plasma solute concentrations and,
21-22
reductions in acute renal failure, 250
sodium retention, 189 Glomerular
podocytes, 6 Glomerular proteinuria
description of, 43-44 laboratory
evaluations, 44 Glomerulonephritis, see
also Nephritis;
Nephropathy
acute, 193
clinical manifestations, 128-129
pathology, 129 systemic diseases
associated with,
137
antibody-mediated, 130
antineutrophil cytoplasmic
autoantibody, 137 associated with
infections bacterial endocarditis, 197198 hantavirus, 199 hepatitis B, 198
hepatitis C, 198-199 parvovirus B19,
199 sepsis, 197-198 shunt nephritis,
198 syphilis, 198
visceral abscesses, 197-198
characteristics of, 193 ch ronic,
35, 139 classification of, 133
immune mechanisms cellmediated
cellular sources, 147 complement
activation, 147 -148 mediator
types, 146-147 humoral

circulating immune complex

disease, 145-146 in situ
immune complex disease,
144-145
morphologic stages, 136
poststreptococcal
acute nephritic syndrome etiology
and, 38-39 clinical
course, 197 clinical
features, 194-195
description of, 194
laboratory findings, 195
pathogenesis, 196-197
pathology, 196 prevention,
197 prognosis, 197
renal biopsy indications, 195-196
treatment, 197 progression to end
stage renal
disease, 139
rapidly progressing
clinical manifestations, 128-129
pathology, 129
systemic diseases associated with,
137
renal biopsy, see Renal biopsy and
systemic diseases, 137 treatment,
134 Glomerulus abnormalities, 20
capillary wall, 42 description of, 4
filtration barrier, see Glomerular
filtration barrier
juxtaglomerular apparatus
cellular components, 7-9
macula densa, 7 structure of, 6
ultrafiltration, 6-7
Glucocorticoid-remediable
aldosteronism, 101
Glucocorticoids
elevated levels in renal failure,
473-474
hypokalemia and, 102
Glucoheptonate, 50
Glucose
dipstick testing, 29 elevated levels, see
Hyperglycemia proximal tubule
transport, 19 Glycyrrhizic acid, 84,
101-102 GN, see Glomerulonephritis
Gonadotropins, alterations in renal
failure patients
females, 474 males,
474
Goodpasture's syndrome autoantigens
and antibodies, 177-178 clinical
features, 176-177 concurrence with
other diseases, 177 course, 178-179
definition of, 175 diagnosis, 177
environmental influences, 178

Index

genetics, 178
pathology, 176-177

plasma exchange, 178

treatment, 178 GRA, see
Glucocorticoid-remediable
aldosteronism
Granular cells
casts, 33

sympathetic innervation, 7-9

Granulomatous disease, hypercalcemia

and, 110
Gross hematuria
clinical features, 309

in medullary sponge kidney, 343

pathogenesis, 309
treatment, 310 Growth hormone,
elevated levels in
renal failure, 472-473 Gut,
potassium excretion in, 98

H
Half-life, 299
Hantavirus, glomerulonephritis
associated with, 199 HCV, see
Hepatitis C virus Heart disease, see
Cardiac disease HELLP syndrome,
see Hemolysis, elevated liver
enzymes, low platelets syndrome
Hematuria
in Alport's syndrome, 320
asymptomatic, 127-128
causes of, 36-37 definition
of, 36, 127 epidemiology, 36
evaluation
common findings, 38 history, 37-38
physical examination, 38 urinalysis,
38 glomerular causes of, 37 evaluative
studies, 38-40 vs. nonglomerular
origin, diagnostic
methods, 127
glomerular capillary changes, 132
gross
clinical features, 309 in medullary
sponge kidney, 343 pathogenesis, 309
treatment, 310 macroscopic, 36
microscopic, 36 nonglomerular causes
of, 37 evaluative studies, 40-41 vs.
glomerular origin, diagnostic
methods, 127 proteinuria and, 38
racial predilections, 40 secondary to
vascular anomalies, 41 Heme pigments,
disposition in plasma,
274-275
Hemodiafiltration, 408-409

527
Hemodialysis
characteristics of, 409
complications
anaphylactoid reactions, 414
anemia
description of, 467
recombinant erythropoietin
treatment, 469 hypoxemia,
414-415 malnutrition, 443 from
rapid fluid removal, 414
technical-related, 415 dialysate,
411-412 drug considerations,
303, 415 high efficiency, 408
indications, 413 machine, 409
membranes
description of, 408-410 solute
clearance, 410-411 water transport,
410-411 neurological complications
dialysis dementia, 463 dialysis
disequilibrium syndrome,
462-463 intellectual function
impairments,
462
peritoneal dialysis and, 420 protein
requirements, 444 quantitation, 413414 vascular access anticoagulation,
413 arteriovenous fistula, 412-413,
439 indwelling catheter placement,
412 morbidity associated with, 422
thrombosis, erythropoietin use and,
469
Hemodialzyers choice of, 415
composition of, 408 reusability, 415
Hemofilters, 408 Hemofiltration
characteristics of, 409 continuous
arteriovenous, 409-410 continuous
venovenous, 409-410 description of,
408 dialysate, 411-412 indications, 413
machine, 409 membranes
description of, 408-410
solute clearance, 410-411
water transport, 410-411
vascular access
anticoagulation, 413 arteriovenous
fistula, 412-413, 439 indwelling catheter
placement, 412 Hemoglobinuria, 274
Hemolysis, elevated liver enzymes, low
platelets syndrome, 389-390 Hemolytic
uremic syndrome acute renal failure
etiology during pregnancy, 394

clinical features, 234 clinical

outcome, 235 epidemiology, 232233 laboratory findings, 234
management of, 234-235
pathogenesis, 233 postdiarrheal,
233-234 during pregnancy, 235
prodrome, 233-234 Hemoperfusion,
409 Hemorrhagic cystitis, 40
Henoch-Schonlein purpura clinical
features, 172, 200, 211 definition of,
211 treatment, 206, 212 Hepatitis B,
glomerulonephritis
associated with, 198 Hepatitis C
virus, glomerulonephritis
associated with, 158-159, 198-199
Hepatorenal syndrome clinical features,
190-191 description of, 190 differential
diagnosis, 191 orthotopic liver
transplantation, 192 prognosis, 191 renal
failure, 190 transjugular intrahepatic
portosystemic shunt, 192 treatment, 191192 Hereditary cystic diseases
autosomal dominant polycystic kidney
disease
clinical course, 316 clinical
manifestations, 315-317 diagnosis,
314-315 incidence, 314 intracranial
aneurysms, 317 nephrolithiasis
findings, 316 pathology, 314
autosomal recessive polycystic kidney
disease
clinical course, 314 clinical features, 314
diagnosis, 313-314 differential diagnosis,
314 incidence, 313 tuberous sclerosis,
317 Von Hippel-Lindau disease, 317
Heroin nephropathy, 162-163
Heymann's nephritis, 145, 147 Hippuric
acid, in uremic syndrome, 406 HIV, see
Human immunodeficiency
virus
HIV-associated nephropathy
clinical course, 240-241 clinical
manifestations, 239
epidemiology, 239-240
pathogenesis, 240 pathology,
240 treatment, 240-241
Hodgkin's disease, renal
complications of nephrotic
syndrome, 227 renal parenchymal
infiltration, 228

528
Homeostasis
calcium
disorders
hypercalcemia, see
Hypercalcemia
hypocalcemia, see Hypocalcemia
methods that regulate, 106-107
potassium
excretion, 98-99
extrarenal disposal
catecholamine effects on, 99
effect of acid-base disorders,
99-100
insulin effects on, 99 shifts between
extracellular and intracellular fluid
compartments, 99-100 systemic effects,
3 HRHA, see Hyporeninemic
hypoaldosteronism HRS, see
Hepatorenal syndrome Human
immunodeficiency virus, renal
manifestations of acid-base
disorders, 238-239 acute renal
failure, 239 dialysis, 242
end stage renal disease, 242 fluid
and electrolyte abnormalities
hyperkalemia, 238 hypokalemia,
237-238 hyponatremia, 237
overview, 238
HIV-associated nephropathy
clinical course, 240-241 clinical
manifestations, 239
epidemiology, 239-240
pathogenesis, 240 pathology,
240 treatment, 240-241
immune-mediated renal disease,
241-242
Humoral immunity description of,
143 role in glomerulonephritis
circulating immune complex
disease, 145-146 in situ
immune complex disease,
144-145
HUS, see Hemolytic uremic syndrome
Hydrogen ions characteristics of, 16
effect on potassium excretion, 16
excretion of, 18 secretion, regulation
of, 19 Hydrometer, for specific gravity
measurements, 27 Hydronephrosis,
353 11/3-Hydroxylase, 102 11 /3Hydroxysteroid dehydrogenase
deficiency description of,
101 hyperkalemia and, 101102 Hypercalcemia causes of,
109-110

Index
clinical presentation, 109
definition of, 109
effect on nephrotoxicity of light
chains, 223 nephropathy, 268
posttransplant, 487 Hypercalcemic
nephropathy, 268 Hypercalciuria
calcium oxalate stone formation and,
362-363
definition of, 362
idiopathic, 363
Hypercapnia, see Respiratory acidosis
Hyperchloremic metabolic acidosis
description of, 72 diagnostic approach,
75-76 in HIV patients, 238 renal causes
of, 72-73 Hyperglycemia, and
hypertonicity relationship between, 69
treatment, 70 Hyperkalemia causes of,
103-104 clinical manifestations, 104105 collecting duct diuretics use and,
119 description of, 103 differential
diagnosis, laboratory tests
to establish fractional excretion
of potassium,
100 transtubular potassium
gradient,
100
in disease-related states acute renal
failure, 284 obstructive uropathy, 352353 drug-induced, 103-104 in elderly,
397 in HIV patients, 238 treatment, 105
true, 103 Hyperlipidemia in chronic
renal failure, 445 posttransplant, 487
Hypernatremia in acute renal failure,
283 clinical classification hypertonicity
due to nonelectrolyte
solutes, 68-69 hypertonic sodium
gain, 68 hypotonic water deficits, 6768 overview, 65-66 pure water
deficits, 66-67 clinical
manifestations, 69 description of, 64
essential, 65
evaluation algorithm, 66 mortality rate,
69 pathophysiology, 65 prevalence of,
64 treatment, 69-70 Hyperoxaluria
drug-induced, 341 enteric, 341 in
idiopathic urolithiasis, 341-342

primary type I, 339-340 type II, 340

type III, 340-341 renal stone formation
and, 364 secondary forms, 337 toxininduced, 341 Hyperparathyroidism
hypercalcemia and, 109 secondary,
pathogenesis of altered calciumregulated parathyroid hormone
secretion, 450 calcemic action, skeletal
resistance
to, 449-450
calcitriol deficiency, 448-449
calcium malabsorption, 449
parathyroid hyperplasia, 450
phosphate retention, 448-449
uremic
diagnosis of, 450-451 prevention and
treatment calcium control, 451
phosphorus control, 451 vitamin D
sterol use, 451-452 Hyperphosphatemia
in acute renal failure, 284, 449 causes
of, 112-113 clinical presentation, 112
definition of, 112 treatment, 113
Hyperphosphatemic nephropathy,
268-269 Hypertension
accelerated, 513-515 blood
volume
conditions that affect, 491 control
of, 491-492 cardiac disease and,
relationship
between, 456
classification of, 497-498
diuretic treatment, 492-493
essential
atypical features, 500
cardiovascular risk, 496-497
clinical presentation, 498 definition
of, 496 pathophysiology, 498
prevalence, 496 screening for, 498500 treatment, 500 vascular
regulation abnormalities,
494
hemodynamics, 491 in
hospital setting, 515 in
infants and children, 387 lead
exposure and, 336 malignant,
513-515 in poststreptococcal
glomerulonephritis, 194
posttransplant, 486-487 during
pregnancy, 388-390 in primary IgA
nephropathy, 173 progression to end
stage renal disease

Index
pathophysiology, 429 -430
treatment, 431, 435-436
reflux nephropathy and, 357
renovascular clinical
features, 502 definition of,
50:1 diagnostic strategy, 504
diagnostic tests
angiotensin converting enzyme
inhibitor-stimulated renography,
503-504 computerized
tomographic
angiography, 504 duplex
ultrasound, 504 intravenous
digital subtraction
renal angiography, 503
magnetic resonance angiography,
504
overview, 503
renin activity measurements, 503
epidemiology, 502 hemodynamic
effects, 504 pathophysiology, 501-502
treatment, 504-505 secondary, 499
therapeutic approach
antihypertensive drugs advantages
and disadvantages,
510-511
combination therapy, 511 -512
monotherapy
alpha! antagonists, 509-510
angiotensin -converting enzyme
inhibitors, 509-510
angiotensin receptor
antagonists, 510-511 /3blockers, 508-510 calcium
antagonists, 509-510 central
adrenergic inhibitors,
510-511
direct vasodilators, 510-511
diuretics, 508
general principles, 506-507
rationale, 506 special
populations
African-Americans, 512
diabetics, 513
elderly, 512
renal insufficiency patients,
512-513 urgencies and
emergencies, treatment
for, 513-515 vascular
resistance determinants,
493-495 Hypertonicity description
of, 64 from nonelectrolyte solutes, 6869 physiologic responses
renal water conservation, 64 thirst
stimulation, 64-65
Hypoalbuminemia, 122
Hypocalcemia
in acute renal failure, 284
causes of, 108

529
in chronic renal disease, 437 clinical
presentation, 107 -108 definition of,
107 treatment, 108-109 Hypocapnia,
see Respiratory alkalosis
Hypochloremia, and maintenance of
metabolic alkalosis, 81-82
Hypocitraturia, renal stone formation
and, 364
Hypocomplementemic
glomerulonephritis, 157
Hypodipsia, 65 Hypokalemia
causes of, 101-102 clinical
manifestations, 102
differential diagnosis
fractio nal excretion of potassium,
100
potassium deficiency, 100-101
transtubular potassium gradient,
100
diuretics use and, 119 drug-induced,
102 in HIV patients, 237-238
treatment, 102-103 Hypotnagnesemia,
hypocalcemia and,
108
Hyponatremia in acute renal
failure, 283 in cirrhotic patients,
treatment
approaches, 190 clinical
manifestations, 61 definition of, 57
diuretics use and, 119 in HIV
patients, 237 incidence of, 57
therapy, 61-63 Hypoosmolality
definition of, 57 differential
diagnosis e uvolemia, 58, 60
hypovolemia, 58 overview, 57-58
pathogenesis, 57
Hypoparathyroidism
hyperphosphatemia and, 113
hypocalcemia and, 108
Hypophosphatemia causes of, 112
clinical presentation, 111-112
definition of, 111 treatment, 112
Hyporeninemic hypoaldosteronism, 74,
238
Hypotonic fluids, 68
Hypovolemia, 58

I
Idiopathic focal segmental
glomerulosclerosis
clinical features, 162
description of, 161
pathogenesis, 162

in renal transplantation, 163

treatment, 163 Idiopathic
membranoproliferative
glomerulonephritis age predilection,
157 clinical features, 157-158
outcome, 157-158 pathogenesis, 156
pathology of, 153 -155 recurrence
after renal transplantation,
159-160
treatment, 158 -159 Idiopathic
membranous nephropathy clinical
features, 164-165 course, 167 diagnosis,
166-167 gender predilection, 164
pathology, 165 treatment, 168-169
ultrastructural features, 165 Idiopathic
nephrotic syndrome, 149 IgA
nephropathy classification of, 170
clinical presentation, 172 epidemiology,
171-172 in HIV patients, 241-242
laboratory findings, 172 natural history,
172-173 pathogenesis, 173-174
pathology
electron microscopy, 170-171
immunohistology, 170 light
microscopy, 170 prevalence of,
170 primary
clinical presentation, 172
hypertension findings, 173
pathogenesis, 173-174 treatment,
174-175 prognostic markers, 173
progression to end stage renal
disease, 175 secondary
pathogenesis, 174 treatment,
174 worldwide frequency, 172
Imaging techniques
angiography
computed tomographic, 50
diagnostic, 50 interventional, 5051 computed tomography
angiography, 50 clinical uses
arterial disease, 51 cysts, 52
inflammatory diseases, 51
renal vein lesions, 53
trauma, 52
ureteral obstruction, 52
urolithiasis, 52 indications,
48 intravenous urography
contrast media, 47

530
Imaging techniques (continued)
contrast nephropathy risks, 47-48
indications, 47 magnetic resonance
imaging diagnostic uses
acute renal failure, 258 ureteral
obstruction, 52 principles of, 48
renal indications, 48 plain
abdominal radiography, 48
radionuclide imaging for acute
renal failure, 258 clinical uses, 49
radiopharmaceutical agents, 48-49
selection of, based on clinical
situation
arterial disease, 51
congenital disorders, 51
cystic disorders, 52
inflammatory disease, 51
neoplasia, 51-52
parenchyma! disorders, 52
renal transplantation, 52
renal vein lesions, 53
trauma, 52
ureteral obstruction, 52
urolithiasis, 52
ultrasonography, see Ultrasonography
voiding cystourethrography, 51
Imipenem, 299
Immobilization, hypercalcemia and, 110
Immune complex disease circulating,
145-146 in situ, 144-145, 148 Immune
response activation of, 143
autoimmunity, 143 cell-mediated, 143
humoral, 143 pathways of, 142 selftolerance causes of, 144 principles of,
143 Immunoglobulins, in idiopathic
membranoproliferative
glomerulonephritis, 155
Immunosuppression, posttransplant
drugs
acute renal failure caused by,
264-265
cancer recurrence and, 477
consideratio ns during pregnancy,
393
cyclosporine, see Cyclosporine
side effects, 484-485 tacrolimus,
265 types of, 483 -484 phases of,
484 rationale for, 482
Immunotactoid glomerulopathy, 226
Indoxyl sulfate, in uremic syndrome,
406
Infants, renal considerations for
anatomical development, 383

Index
functional development acid- base
balance, 386 blood pressure, 386387 calcium metabolism and
excretion,
385
glomerular filtration rate, 383
oliguria, 385
phosphate concentrations, 385
potassium levels, 385-386 sodium
metabolism, 384-385 urine
concentration, 383 Infections, see
specific infection Inflammatory
diseases, imaging
procedures, 51
In situ immune complex disease
experimental models of, 144-145
illustration of, 144 mechanisms
of, 144-145 in membranous
nephropathy,
165-166
Insufficiency, renal characteristics of,
433 definition of, 429 effect on
pregnancy, 391-392 hemodynamically
mediated, from
therapeutic agents, 260 -261
hypertension treatment
considerations, 512-513
progression to end stage renal disease
course, 429 mechanisms, 429-431
treatment
blood pressure control, 431 dietary
approaches, 431-432 Insulin, effect on
extrarenal potassium
disposal, 99 Intercalated cells, 10
Interlobular arteries, 5 Interstitial
nephritis, acute causes of, 261-262
idiopathic, 281 immune disorders, 281
infections, 281 therapeutic agents
nonsteroidal anti-inflammatory
drugs, 280
overview, 279
penicillins, 279-280
rifampin, 280
sulfonamide derivatives,
280-281
clinical course, 279 clinical features,
277-279 description of, 278 laboratory
findings, 278-279 pathogenesis, 277
treatment, 279 Intravenous pyelography,
for urinary
tract infection, 40 Intravenous
urography contrast media, 47 contrast
nephropathy risks, 47-48 indications, 47

Inulin
glomerular filtration rate
determinations using, 7, 23-24
properties, 7, 23-24 Ischemic heart
disease, in renal failure,
457
IVP, see Intravenous pyelography
IVU, see Intravenous urography

Juvenile nephronophthisis medullary

cystic disease complex
autosomal recessive, 324
clinical features, 324
components of, 323
description of, 323-324
diagnosis, 325
nonrenal findings, 324
prognosis, 325
treatment, 325
Juxtaglomerular apparatus, 4
Juxtaglomerular granular cells, 7
Juxtamedullary glomeruli, 6

K
Kawasaki disease clinical
features, 202, 204 diagnosis,
205 renal pathology, 200
Ketoacidosis alcoholic, 78
diabetic
bicarbonate use, 77 diagnosis of,
77 hormonal profile, 77
hypophosphatemia and, 112
potassium deficiencies, 100
treatment, 77 starvation, 77-78
Ketones, 29 Kidney
age-related changes, 395 anatomical
development, 383 calcium excretion,
107 chronic insufficiency,
hypocalcemia
and, 108
disease, see specific disease
filtration rate, see Glomerular
filtration rate functions of, 3
glucose transport, 19 homeostasis
regulation, 3 in infants and children,
functional
considerations acid- base
balance, 386 blood pressure, 386387 calcium metabolism and
excretion,
385
glomerular filtration rate, 383
oliguria, 385
phosphate concentrations, 385
potassium levels, 385-386

Index
sodium metabolism, 384-385
urine concentration, 383
phosphorus excretion, 111
potassium regulation, 98
structure of, 4
transplantation, see Renal
transplantation

Labetalol, 514
Lactate, formation of, 76
Lactic acidosis
classification of, 77
description of, 76-77
in HIV patients, 238-239
treatment, 77 Lamoxatam, 305
LCDD, see Light chain deposition
disease Lead
nephrotoxicity
clinical manifestations, 335-336
diagnosis, 336-337
pathology, 336
pathophysiology, 335
renal impairment associated with, 336
risk factors, 335
treatment, 337
Left ventricular hypertrophy, 455
Leiomyomatosis, in Alport's syndrome,
321
Leukocytes
dipstick testing, 29
dysfunction, in uremia, 470-471
in urine, 32 LeVeen shunt, for
hepatorenal
syndrome, see Peritoneal jugular shunt,
for hepatorenal syndrome LHRH, see
Luteinizing hormone
releasing hormone
Libido, effect of renal failure on, 474
Liddle's syndrome, 84, 98, 102 Lightchain deposition disease, 221
AL-amyloidosis and, comparison
between, 225
clinical features, 224
immune complex deposits, 156
pathogenesis, 224
pathology, 223-224
prognosis, 224
treatment, 224
Light chains
filtration of, 221
globular regions, 220-221
cast nephropathy, see Cast
nephropathy
urinary concentration, 221 Lipids,
abnormalities in chronic renal
failure, 475-476 Lipoid
nephrosis, 149 Lithium, 305
Lithium-induced renal disease
acute renal failure, 333

531
chronic interstitial nephritis, 334
nephrogenic diabetes insipidus,
332-333
nephrotic syndrome, 334 renal
tubular acidosis, 333 Lithium
intoxication, 333 Liver disease
ascites, management of, 189
cirrhosis renal sodium
handling
abnormalities in clinical features,
188 efferent factors, 189
pathogenesis, 188-189 renal water
handling abnormalities in clinical
features, 189-190 hyponatremia
treatment, 190 pathogenesis, 190
Loop diuretics absorption, 122
adaptation to, 118-119
complications of, 119 continuous
infusion, 123 hemodynamic effects,
116 indications ascites, 189 edema,
120
mechanism of action, 116
resistance, 122-123 types of,
115-116 Loop of Henle
ascending limbs, 9
countercurrent multiplier, 15
description of, 4 diuretics, see
Loop diuretics illustration of,
11
sodium chloride reabsorption, 116 Lowprotein diets adequacy monitoring, 442443 compliance monitoring, 442-443
effect on renal disease progression,
441-442 spontaneous protein
intake decreases,
441
Luminal transporters, 13 Lupus
nephritis description of, 129 effect
of pregnancy, 392 mesangial
nephropathy, 209 morphologic
classes of, 209-210 proliferative,
210 renal biopsy, 208 ultrastructural
features, 129 Luteinizing hormone
releasing
hormone, 474
LV hypertrophy, see Left ventricular
hypertrophy

M
Macula densa
definition of, 13

renal secretion, 7 role in renin

release, 13 Magnesium,
excretion, effect of
distal convoluted tubul e diuretics
on, 117
Magnetic resonance imaging
diagnostic uses acute renal failure, 258
ureteral obstruction, 52 principles of,
48 renal indications, 48 Major
histocompatibility complex
proteins, 142 Malignancies, see
also Neoplasms
hypercalcemia and, 109 Malnutrition,
see also Nutrition in dialysis patients,
443 in end stage renal disease, 443
Mannitol, 117-118, 275-276 MCP-1,
see Monocyte chemotactic
peptide-1 MCTD, see Mixed
connective tissue
disease
Medulla
blood supply, 6
description of, 4
hypertonicity, 14
Medullary cystic disease, see Juvenile
nephronophthisis medullary cystic
disease complex Medullary sponge
kidney clinical features
gross hematuria, 343
urinary tract infection,
343-344 definitions, 343
diagnosis, 344 epidemiology, 343
pathology, 343 pathophysiology, 343
prognosis, 344 treatment, 344
Membranoproliferative
glomerulonephritis characteristics of,
153 cryoglobulinemia, 156 definition
of, 153 hepatitis C virus incidence,
158 in HIV patients, 241-242
idiopathic (primary) age predilection,
157 clinical features, 157-158
outcome, 157-158 pathogenesis, 156
pathology of, 153-155 recurrence
after renal
transplantation, 159 -160
treatment, 158-159 nephrotic
features, 138-139 pathogenesis,
156-157 secondary
clinical features, 158
outcome, 158 overview,
154

532
Membranoproliferative
glomerulonephritis (continued)
pathogenesis, 156-157 pathology,
155-156 progression to
cryoglobulinemia,
158
treatment, 159 types of, 153-154
Membranous glomerulopathy age of
onset, 138 immunofluorescence staining
pattern,
135
incidence of, 138 and nephrotic
syndrome, 138 Membranous lupus
nephropathy, 211 Membranous
nephropathy clinical features, 164-165
definition of, 164 diagnosis, 166-167 in
HIV patients, 241-242 idiopathic
clinical features, 164-165 course,
167 diagnosis, 166-167 gender
predilection, 164 pathology, 165
treatment, 168-169
ultrastructural features, 165
nephrotic syndrome, 164-165
pathogenesis, 165-166 secondary
clinical entities associated with, 165
pathology, 165 treatment, 169
treatment, 167-169 Mesangial
hyperplasia, 129 Mesangial
nephropathy, 208 Mesna, genitourinary
hemorrhage from,
228 Metabolic
acidosis
in chronic renal disease, 437
definition of, 71
evaluation of, using plasma anion gap
findings, 71-72 hyperchloremic
description of, 72 diagnostic
approach, 75-76 renal causes of, 7273 types of, associated with high
anion
gap
alcoholic ketoacidosis, 78
diabetic ketoacidosis, 77
diagnostic approach, 78
ethylene glycol ingestion, 79
lactic acidosis, Id11
methanol ingestion, 79
renal failure, 76
salicylate intoxication, 78 -79
starvation ketoacidosis, 77-78
Metabolic alkalosis bicarbonate
levels
generation of high levels, 81
maintenance of, 81-82
characteristics of, 80

Index
clinical conditions that cause
antacid ingestion, 84 Bartter's
syndrome, 82 diuretics, 82, 119
hypertension
with low renin and
aldosterone, 84
with mineralocorticoid excess, 84
milk-alkali syndrome, 84 nonrenal
causes, 82-84 renal causes, 82 vomiting,
82-84 clinical presentations, 80
diagnosis of, 84-85 differential
diagnosis, 83 maintenance of, 81-82
pathophysiology, 81-82 treatment, 85
Metabolically induced acute renal
failure
hypercalcemic nephropathy, 268
hyperphosphatemic nephropathy,
268-269
uric acid nephropathy, 231
clinical outcome, 267 clinical
presentation, 266-267
definition of, 266 diagnosis,
267 pathogenesis, 266
pathology, 266 treatment
for existing disease, 268 prophylactic
approaches, 267 Methanol ingestion,
and metabolic
acidosis, 79
Methotrexate, 230, 305 Methyldopa, for
hypertension during
pregnancy, 390 Methylguanidine, in
uremic syndrome,
406
Metolazone, 122
Metronidazole, 299 MHC
proteins, see Major
histocompatibility complex proteins
Microalbuminia in diabetic nephropathy
patients,
217-218
diseases that manifest, 43 /32 Microglobulin, 330, 406-407
Microscopic polyangiitis, 202 Milkalkali syndrome, 84, 110
Mineralocorticoids excess states,
metabolic alkalosis
associated with, 84 hypokalemia and,
102 Minimal change nephropathy age
of onset, 150 clinical course, 151 clinical
presentation, 150 -151 complications,
152 histopathology, 149-150 long-term
outcome, 152 nephrotic syndrome
associated with, 150-151

relapses, 151-152
terminology associated with, 149
treatment protocols, 151 relapses, 151152 for steroid-resistant patients, 152
Mithramycin, nephrotoxicity, 230
Mitomycin C, nephrotoxicity, 230
Mixed connective tissue disease
autoantibodies, 209
renal involvement, 214
Mixed lesions
description of, 453
dialysis-related amyloidosis, 453-454
"mMAG3, 49
Monocyte chemotactic peptide -1, 277
MPGN, see Membranoproliferative
glomerulonephritis
MRI, see Magnetic resonance imaging
MSK, see Medullary sponge kidney
"mTc-DPTA, 49 Multiple myeloma
hypercalcemia and, 223
renal failure effects, 223
Mycophenolate mofetil, 485

N
Na+, see Sodium
Nafcillin, 299
Na,K-ATPase pump, 12-13, 114
Natriuresis, and diuretic use, 83
Natriuretic peptides, 185-186
Neoplasms
bladder, see Bladder neoplasms
imaging procedures, 51-52 pelvic, see
Pelvic tumors postrenal
transplantation, 486 renal, see Renal
neoplasms ureteral, see Ureteral
tumors Nephritis, see also
Glomerulonephritis differential
diagnosis, 322 in Goodpasture's
syndrome, 176-177 hereditary, see
Alport's syndrome Heymann's, 145,
147 interstitial, see Interstitial nephritis
lupus, see Lupus nephritis
tubulointerstitial, 148 clinical
manifestations, 329 diagnosis
enzyme analysis, 330-331
proteinuria, 329-330 tubular
analysis, 331 urinalysis, 329
histology, 329
in juvenile nephronophthisis medullary
cystic disease complex, 324
Nephrogenic diabetes insipidus, 67,
332-333
Nephrolithiasis
in autosomal dominant polycystic
kidney disease, 316
calcium, risk factors for
hypercalciuria, 362-363

Index
hyperoxaluria, 364
hypoeitraturia, 364 low
urinary volume, 362 uric
acid, 364 cystine, 365
gender predilection, 360
incidence, 360 infection, 364-365
metabolic evaluation, 361-362 in
obstructive uropathy, 354
pathogenesis, 361 stones
characteristics of, 360-361
formation, 361 metabolic
activity, 361-362 urological
management, 361,
365
treatment, 354 uric
acid, 364 Nephron
organization, 5
segments
chloride entry. 13 description of, 4,
9 sodium entry, 13 Nephropathy
analgesic caffeine and, 295
carcinoma rates, 297 description
of, 293 diagnostic approach
diagnostic criteria, 297 dosage
ingested, 295 history-taking, 295296 laboratory findings, 296-297
physical examination, 296 habitual
use, 293-294 pathogenesis, 295
prevalence of, 294-295 risk of, 294
treatment, 297 cast
clinical features, 222
pathogenesis, 222
pathology, 221
prognosis, 222-223
treatment, 222-223
diabetic
description of, 215 effect of
pregnancy, 392 epidemiology,
215 natural history functional
changes, 216 illustration of,
216 microalbuminia, 217
pathology, 217 pathogenesis,
215-216 progressive disease,
treatment clinical studies, 218219 modalities, 219 screening,
217-218 transition to end
stage renal disease, treatment
options, 219-220

533
HIV-associated
clinical course, 240-241 clinical
manifestations, 239
epidemiology, 239-240
pathogenesis, 240 pathology,
240 treatment, 240-241 IgA
classification of, 170 clinical
presentation, 172
epidemiology, 171-172 in
HIV patients, 241-242
laboratory findings, 172
natural history, 172-173
pathogenesis, 173-174
pathology
electron microscopy, 170-171
immunohistology, 170
light microscopy, 170
prevalence of, 170
primary
clinical presentation, 172
hypertension findings, 173
pathogenesis, 173-174
treatment, 174-175 prognostic
markers, 173 progression to end
stage renal
disease, 175
secondary
pathogenesis, 174
treatment, 174 worldwide
frequency, 172 membranous
clinical features, 164-165
definition of, 164
diagnosis, 166-167
in HIV patients, 241-242
idiopathic
clinical features, 164-165
course, 167
diagnosis, 166-167
gender predilection, 164
pathology, 165
treatment, 168-169
ultrastructural features, 165
nephrotic syndrome, 164-165
pathogenesis, 165-166 secondary
clinical entities associated with,
165
pathology, 165
treatment, 169 treatment, 167169 minimal change age of onset,
150 clinical course, 151 clinical
presentation, 150 -151
complications, 152 histopathology,
149-150 long-term outcome, 152
nephrotic syndrome associated
with, 150-151 relapses, 151152 terminology associated with,
149

treatment protocols, 151 relapses, 151152 for steroid-resistant patients, 152

pigment
heme pigments, disposition in
plasma, 274-275
hemolysis with hemoglobinuria, 274
mechanisms of, 275 rhabdomyolysis
causes of, 273 drug-induced, 273
exertional, 273
laboratory diagnosis, 273-274
traumatic, 273 treatment, 275-376
reflux
definition of, 356 diagnosis, 358
evaluation, 358-359
glomerulosclerosis, 357 progressive
renal insufficiency, 357 proteinuria,
357 scar formation, 357 sickle cell
disease focal segmental
glomerulonephritis,
312 gross
hematuria
clinical features, 309 pathogenesis,
309 treatment, 310 proteinuria
diagnosis, 312 pathogenesis, 312
treatment, 312 renal papillary
necrosis, 309 tubular dysfunction
pathogenesis acid excretion, 311
potassium excretion, 311 proximal
tubular reabsorption,
311
tubular secretion defects, 311 urine
concentration, 310-311 water
reabsorption, 310-311 treatment, 311
uric acid, 231
clinical outcome, 267 clinical
presentation, 266-267
definition of, 266 diagnosis, 267
pathogenesis, 266 pathology,
266 treatment
for existing disease, 268 prophylactic
approaches, 267 Nephrotic syndrome
characteristics of, 137 -138
complications of peritonitis, 152
proteinuria, 44, 137-139, 152 diuretic
resistance, 122 etiology, 138
focal segmental glomerulosclerosis,
163

534
Nephrotic syndrome (continued)
IgA nephropathy, 172
membranous nephropathy,
164-165, 167 minimal
change nephropathy,
150-151
glomerular capillary changes, 131
glomerular disease and, 137-138
Hodgkin's disease and, 227
hyperlipidemia, 165, 167-168
idiopathic, 149 lithium-induced, 334
during pregnancy, 391 therapeutic
agents that cause, 262 Nephrotoxic
agents aminoglycosides, 262-263
antibiotics, 262-264 biological
response modifiers, 230 cisplatin, 229
cyclophosphamide, 229 lead, see Lead
nephrotoxicity methotrexate, 230
mithramycin, 230 mitomycin C, 230
semustine, 230 streptozotocin, 229
Netilmicin, 305 Nicardipine, 514
Nifedipine, 514 Nil disease, 149
Nitrite, dipstick testing, 29
Nitroglycerin, 514 Nitrosoureas,
nephrotoxicity, 230 Nonelectrolyte
solutes glucose, see Glucose
hypertonicity due to, 68-69
Nonobstructive urinary tract dilatation
description of, 351-352 diagnosis, 354
Nonsteroidal anti-inflammatory drugs
clinical uses
membranous nephropathy, 167
proteinuria, 46, 152, 167 electrolyte
and water excretion
alterations
potassium, 292
sodium, 292
water, 292-293
hyperkalemia associated with, 104
renal syndromes associated with
acute interstitial nephritis, 280,
292
acute renal failure causative
agents, 292 epidemiology, 291
mechanism, 291-292 management
of, 293 nephrotic syndrome, 262
prerenal azotemia, 261 renal
insufficiency, 261 Nutrition, for
acute renal failure energy
requirements, 447 goals, 446

Index
principles of, 446-447 water and
mineral intake, 447 for chronic renal
failure assessment strategies, 442
goals of, 440 low-protein diets
adequacy monitoring, 442-443
compliance monitoring, 442-443
effect on renal disease
progression, 441-442
spontaneous protein intake
decreases, 441

Obstructive uropathy
classification, 348 clinical
features, 348-349 definition
of, 348 diagnosis, 353-354
etiology
acquired extrinsic, 350-351
acquired intrinsic, 349-350
congenital, 350-351
nonobstructive urinary tract
dilatation, 351-352
overview, 349 incidence, 348
intravenous pyelogram use, 353
laboratory findings, 349 natural history,
348 and pregnancy, 350-351
radionuclide studies, 49-50 renal
function effects electrolyte
abnormalities, 352 glomerular, 352
tubular, 352-353 renal pathology, 353
treatment, 354-355 ureteropelvic
junction obstruction description of, 351
diagnosis, 353-354 Ofloxacin, 299
1,25(OH) 2 D3 , 113 OKT3, 485
Oliguria, in infants and children, 385
OLTX, see Orthotopic liver
transplantation
Orthophosphate
for primary hyperoxaluria, 340 for
stone formation, 363 Orthotopic liver
transplantation, for
hepatorenal syndrome, 192
Osmolarity, body fluid, regulation of
mechanisms
countercurrent multiplier, 15
medullary hypertonicity, 14-15 role of
antidiuretic hormone, 14 tubular water
absorption, 14 overview, 14
Osmoregulation
antidiuretic hormone control of,
14, 16
countercurrent multiplication, 15-16
mechanisms that control, 16

medullary hypertonicity, 14-15

overview, 14
tubular water absorption, 14
volume regulation and,
comparison, 16 Osmotic
diuretics, 117-118 Osteitis fibrosa
histological features, 448
secondary hyperparathyroidism,
pathogenesis altered calcium regulated parathyroid hormone
secretion, 450 calcemic action,
skeletal resistance
to, 449 -450
calcitriol deficiency, 448-449 calcium
malabsorption, 449 parathyroid
hyperplasia, 450 phosphate retention,
448-449 Osteodystrophy, renal
adynamic lesions, 453 dialysis-related
amyloidosis, 453-454 management
approach, 454 mixed lesions, 453-454
osteitis fibrosa, see Osteitis fibrosa
osteomalacia, see Osteomalacia
Osteomalacia aluminum -related,
prevention and
treatment, 453
characteristics of, 452
diagnosis, 452-453
pathogenesis, 452
Overproduction proteinuria
differential diagnosis, 43
laboratory evaluation, 44
Over-the-counter analgesics
habitual use, 293-294
nephropathy, see Analgesic
nephropathy Oxalate absorption
description of, 338 dietary influences,
341 diseases that increase, 341 cellular
transport abnormalities, 342 crystal
formations, 337 dietary sources, 338
overproduction, see Hyperoxaluria
physiology of, 338-339 renal handling,
338 stone formation, see Stones, renal
urinary excretion of, normal values,
338

Pancreatitis, acute, hypocalcemia and,

108
Papillae, 4
Papillary necrosis clinical course,
346 clinical manifestations, 346
conditions associated with, 346
diagnosis, 346

Index
pathogenesis, 345-346
in sickle cell patients with hematuria,
309
treatment, 347 Paracentesis, for
management of ascites
in cirrhotic patient, 189
Parathormone, 406-407
Parathyroidectomy, 108
Parathyroid hormone
accumulation, see
Hyperparathyroidism
calcemic action of, skeletal resistance,
449-450
metabolism of, 451
peripheral neuropathy and, 461
Parathyroid hyperplasia, 450
Parenchymal disorders, 52
Parvovirus, glomerulo nephritis
associated with, 199 Fauci-immune
glomerulonephritis, 148 PCR, see
Protein catabolic rate PD, see Peritoneal
dialysis Pelvic tumors, renal
diagnosis, 374
etiology, 373
pathology, 373-374
staging, 374
treatment, 374-375 Penicillins, acute
interstitial nephritis
and, 279-280
Pentamidine
acute renal failure caused by, 263-264
hyperkalemia induced by, 104
Pericarditis, 458 Peripheral
neuropathy
autonomic and cranial nerve
dysfunction, 462
classification of, 461
prevalence in end stage renal disease,
461
symptoms, 461
treatment, 462
uremic toxin etiology, 461-462
Peritoneal dialysis
for acute renal failure, 421
adequacy determinations, 418-419
catheters
access site selection, 417
infections, 418 -419 leaks,
417
change to hemodialysis, 420
considerations for diabetics, 420
continuous cycle, 416
description of, 416
fluid composition, 417
forms of, 417
intermittent, 421
nutritional considerations, 420
outcome, 420-421
peritoneal membrane, 416
protein requirements, 444
ultrafiltration failure, 419-420
Peritoneal equilibration test, 416
Peritoneal jugular shunt, for
hepatorenal syndrome, 192

535
Peritoneal membrane, 416
Peritonitis, 418
Peritubular capillary network, 5
Permeability, glomerular, 42
PET, see Peritoneal equilibration test
pH
dipstick testing, 28 of
urine, 28
Pharmacokinetics, drug continuous renal
replacement therapy,
304-305
dialysis effects
continuous ambulatory peritoneal
dialysis, 303-304 hemodialysis,
303 intraperitoneally administered
drugs,
304
Phenobarbital, 305
Phosphate, in infants and children, 385
Phosphorus decreased levels, see
Hypophosphatemia
distribution, 111 elevated
levels, see
Hyperphosphatemia
homeostasis, 111 intestinal
absorption, 111 renal excretion,
111 serum concentrations, 111
Pigment nephropathy
heme pigments, disposition in plasma,
274-275
hemolysis with hemoglobinuria, 274
mechanisms of, 275 rhabdomyolysis
causes of, 273 drug-induced, 273
exertional, 273
laboratory diagnosis, 273-274 traumatic,
273 treatment, 275 -376 Piperacillin, 299
Plain abdominal radiography description
of, 48 for urolithiasis, 52 Plasma
exchange for Ooodpasture's syndrome,
178 for pigment nephropathy, 276
Platelet dysfunction, in uremia, 470-471
Polyarteritis nodosa angiography for, 50
clinical features, 202 diagnosis, 205 renal
pathology, 200 Polycystic kidney disease
anemia, 466 autosomal dominant clinical
course, 316 clinical manifestations, 315317 diagnosis, 314-315 incidence, 314
intracranial aneurysms, 317
nephrolithiasis findings, 316 pathology,
314

autosomal recessive clinical course,

314 clinical features, 314 diagnosis,
313-314 differential diagnosis, 314
incidence, 313 tuberous sclerosis,
317 Von Hippel-Lindau disease,
317 Poly neuropathy, 461 Polyuria
diabetes insipidus, 67 lithium-induced,
332 Poststreptococcal
glomerulonephritis clinical course, 197
clinical features, 194-195 description
of, 194 laboratory findings, 195
pathogenesis, 196-197 pathology, 196
prevention, 197 prognosis, 197
renal biopsy indications, 195-196
treatment, 197
Posttransplant considerations
complications diabetes
mellitus, 487 hypercalcemia,
487 hyperlipidemia, 487
hypertension, 486-487
infection, 485-486
malignancies, 486 drug
selection, 487 general care,
483 immunosuppression
drugs
acute renal failure caused by,
264-265
cancer recurrence and, 477
considerations during pregnancy,
393
cyclosporine, see Cyclosporine side
effects, 484-485 tacrolimus, 265 types
of, 483-484 phases of, 484 rationale for,
482 Potassium
alkalemia effects, 80 balance of,
in elderly, 397-398 body fluid
characteristics of, 16
distribution of, 16
excretion
diuretics and, 17 regulation of, 17, 98
renal handling of, 16, 98 secretion
mechanisms, 16-17 decreased levels,
see Hypokalemia deficiency, vs.
hypokalemia,
differential diagnosis, 100-101
elevated levels, see Hyperkalemia
excretion
effect of nonsteroidal antiinflammatory drugs, 292

536
Potassium (continued) fractional
definition of, 100 mathematical
formula, 100 in obstructive uropathy,
352-353 in tubular dysfunction, 311
homeostasis, mechanisms of
excretion, 98-99 extrarenal disposal
catecholamine effects on, 99 effect of
acid- base disorders,
99-100
insulin effects on, 99 shifts between
extracellular and intracellular fluid
compartments, 99-100 in infants and
children, 385-386 Prednisolone, 178
Preeclatnpsia
characteristics of, 388-389 definition of,
388 follow-up care, 394 tissue damage
associated with, 390 Pregnancy
acute renal failure during, causes of
acute fatty liver, 394 acute tubular
necrosis, 393-394 cortical necrosis,
393-394 hemolytic uremic syndrome,
394 obstruction, 393 preeclampsiaeclampsia, 393 chronic renal
insufficiency and,
391-392
diabetic nephropathy effects, 392 in
dialysis patients, 392 follow-up care,
394 hypertensive disorders, 388-390
lupus nephritis and exacerbations, 392
renal failure due to, indications for
terminating pregnancy, 392
and obstructive uropathy, 350-351
in renal allograft patients, 392-393
renal function during
glomerular filtration rate increases,
388
renal blood flow, 388 urinary tract
infection, 388 in women with chronic
renal disease,
factors that affect hypertension,
390-391 proteinuria, 391 in women
with renal disease but good
renal function, 391 Prerenal
azotemia, see also Uremic
syndrome
acute renal failure and, 254
fractional excretion of sodium
measurements, 257 physical
examination, 256 Prerenal proteinuria,
see Proteinuria,
overproduction Primary
amyloidosis, see ALamyloidosis

Index
Primary hypercapnia, see Respiratory
acidosis Primary hypocapnia, see
Respiratory
alkalosis
Principal cells, 10
Progressive renal disease
algorithm, 434 course of,
429
low-protein diet effects, 441 -442
mechanisms, 429-430
progression to end stage renal
disease, methods to forestall,
435-436 treatment angiotensin
converting enzyme
inhibitors, 435, 442 blood
pressure control, 431 dietary
approaches, 431-432 Prolactin,
elevated levels in renal
failure, 473
Proliferative lupus nephritis, 210-211
Prostaglandins
nonsteroidal anti-inflammatory drug
effect, 291 regulation of renal
function in sickle
cell disease, 311 renal
effects, 186 Prostate
cancer diagnosis, 378-379
etiology, 377-378
pathology, 378 staging,
379 treatment, 379
urinary tract obstruction and, 228
Prostate specific antigen, 378-379
P rotein chronic renal failure
requirements,
444
dialysis requirements, 443 dipstick
testing, 28 low-protein diets adequacy
monitoring, 442-443 compliance
monitoring, 442-443 effect on renal
disease progression,
441-442 spontaneous
protein intake
decreases, 441 renal
handling of, 42 restrictions,
as treatment for progression
to end stage renal disease,
431-432 turnover
in chronic renal failure patients, 441 in
healthy individuals, 440-441 Protein
catabolic rate, 443 Proteinuria
asymptomatic, 137 -139
clinical consequences, 44-45
components of, 42 definition
of, 42 effect on pregnancy,
391 functional, 43
glomerular, 43-44 vs.
tubular, 330

laboratory evaluation, 44
measurement of, 42-43 mechanisms
of, 43-44 overproduction description
of, 43 laboratory evaluation, 44 reflux
nephropathy and, 357 renal
involvement, tubulointerstitial
disease, 148, 329-330
renal presentation
diabetic nephropathy, 217
focal segmental glomerulosclerosis,
160-161
membranous nephropathy, 164 -165
minimal change nephropathy,
151-152
sickle cell nephropathy, 311-312
therapy for, 45-46 treatment options
angiotensin -converting enzyme
inhibitors, 152 nonsteroidal
anti-inflammatory
drugs, 152
tubular, 44
Proximal convoluted tubule
acidosis
description of, 73 renal
bicarbonate threshold, 73
description of, 4 diuretics
carbonic anhydrase inhibitors,
115-116
mechanism of action, 114-116
types of, 115
filtrate absorption and secretion, 9
glucose transport, 19 illustration of,
10
necrosis, in pigment nephropathy, 275
phosphate transport, 9 potassium
excretion, 98 protein reabsorption, 42
reabsorption defects, in sickle cell
disease, 311 Prune-belly
syndrome, and obstructive
uropathy, 351
PSA, see Prostate specific antigen
Pseudodiabetes, 474
Pseudohyperkalemia, 103-104
Pseudohyponatremia, 57 Pseudomonas
aeruginosa, peritoneal
catheter infections and, 418
PSGN, see Poststreptococcal
glomerulonephritis Pulmonary-renal
vasculitic syndrome clinical
presentation of
glomerulonephritis, 137
types of, 177 Pure water
deficits causes of, 66 in
diabetes, 66-67 Pyelonephritis,
acute diagnosis, 367 prognosis,
369 treatment, 367-368

Index
Pyramids, 4
Pyuria, urinary tract infections and, 40

RA, see Rheumatoid arthritis Radiation

nephropathy, 231 Radiocontrast agents,
acute renal failure
caused by, 264 Radionuclide
imaging for acute renal failure,
258 clinical uses, 49
radiopharmaceutical agents, 48-49
Rapidly progressing glomerulonephritis
clinical manifestations, 128-129
pathology, 129 RBF, see Renal blood
flow Red blood cells casts, 33
hematuria, see Hematuria
Reflux nephropathy, see also
Vesicoureteral reflex definition of,
356 diagnosis, 358 evaluation, 358-359
glomerulosclerosis, 357 progressive renal
insufficiency, 357 proteinuria, 357 scar
formation, 357 Refractometers, for
specific gravity
measurements, 27 Relapsing
polychondritis, 209, 214 Renal
artery, 4-5 Renal biopsy
complications, 139-140
contraindications, 139
indications
acute renal failure diagnosis, 259
focal segmental glomerulosclerosis,
162
lupus nephritis, 195-196
overview, 139, 208
poststreptococcal
glomerulonephritis, 195-196 methods,
140-141 Renal blood flow age-related
changes, 395-396 cardiac performance
and, 184-185 effect of nonsteroidal antiinflammatory drugs, 291-292 during
pregnancy, 388 Renal cell carcinoma
diagnosis, 372 incidence, 371 staging of,
372 treatment, 372-373 Renal
circulation anatomy of, 4-5 capillary
beds, 5-6 medullary blood supply, 6
Renal clearance formula for, 21
glomerular filtration rate measurements,
21

537
Renal disease
cholesterol atheroembolic, see
Cholesterol atheroembolic renal
disease chronic complications
acid-base metabolism, 437
anemia, 437 lipid metabolism, 438
metabolism, 437-438 potassium
excretion, 437 psychological
factors, 438 sleep disorders, 438
sodium balance, 436-437 uremic
coagulopathy, 438 water
excretion, 437 drug therapy, 438
stages of, 433
surgical considerations, 438 treatment,
433-435 end stage, see End stage
renal disease hyperlipidemia, 445
nutritional therapy, see Nutrition
progressive algorithm, 434 course of,
429
low-protein diet effects, 441-442
mechanisms, 429-430 progression
to end stage renal disease, methods
to forestall, 435-436 treatment
blood pressure control, 431
dietary approaches, 431-432
protein turnover
in chronic renal failure patients, 441 in
healthy individuals, 440-441 Renal
failure, see also Hepatorenal
syndrome
acute, see Acute renal failure
characteristics of, 433 chronic, see
Chronic renal failure drug therapy
considerations absorption effects,
299-300 dialysis continuous
ambulatory peritoneal
dialysis, 303-304 hemodialysis,
303 distribution effects, 300 dosage
adjustments, 301-303 elimination
effects, 300-301 metabolite
accumulation, 301-302
pharmacodynamics effects, 301
hypercalcemia associated with, 110
hyperphosphatemia associated with,
112-113
in juvenile nephronophthisis
medullary cystic disease
complex, 324
metabolic acidosis etiology and, 76 in
minimal change nephropathy, 152
potassium homeostasis during, 98-99
Renal function factors that affect, 435

indices of
creatinine clearance, see Creatinine
clearance
glomerular filtration rate, see
Glomerular filtration rate in infants
and children acid-base balance, 386
blood pressure, 386-387 calcium
metabolism and excretion, 385
glomerular filtration rate, 383 oliguria,
385
phosphate concentrations, 385
potassium levels, 385-386 sodium
metabolism, 384-385 urine
concentration, 383 Renal
insufficiency characteristics of, 433
definition of, 429 effect on
pregnancy, 391-392
hemodynamically mediated, from
therapeutic agents, 260-261
hypertension treatment
considerations, 512-513
progression to end stage renal disease
course, 429 mechanisms, 429-431
treatment
blood pressure control, 431 dietary
approaches, 431-432 Renal neoplasms
angiomyolipoma, 373 diagnosis, 372
etiology, 371-372 incidence, 371
metastatic tumors, 373 oncocytoma, 373
pathology, 372 renal cell carcinoma, see
Renal cell
carcinoma sarcomas, 373 staging,
372 treatment, 372-373 Renal
osteodystrophy adynamic lesions, 453
dialysis-related amyloidosis, 453-454
management approach, 454 mixed
lesions, 453-454 osteitis fibrosa, see
Osteitis fibrosa osteomalacia, see
Osteomalacia Renal papillary necrosis
clinical course, 346 clinical
manifestations, 346 conditions associated
with, 346 diagnosis, 346 pathogenesis,
345-346 in sickle cell patients with
hematuria,
309
treatment, 347
Renal parenchymal infiltration,
Hodgkin's disease and, 228
Renal replacement therapy
continuous, drug dosing
considerations in, 304-305
for end stage renal disease, 439

538
Renal transplantation
contraindications
cancer, 477
cytomegalovirus, 478
infections, 477 -478
tuberculosis, 478
description of, 482
donors
cadaveric, 479-480 living, 480-481
failure, management of, 487 focal
segmental glomerulosclerosis in,
163
gonadotropin levels, 474
hypocalcemia associated with, 110
hypophosphatemia associated with,
112
idiopathic membranoproliferative
glomerulonephritis after, 7-8 imaging
techniques, 52 mortality after,
diseases that cause cardiovascular
disease, 478-479 cerebrovascular
disease, 479 liver disease, 478
posttransplant complications
diabetes mellitus, 487
hypercalcemia, 487 hyperlipidemia,
487 hypertension, 486-487
infection, 485-486 malignancies,
486 drug selection, 487 general
care, 483 immunosuppression drugs
acute renal failure caused by,
264-265
cancer recurrence and, 477
considerations during
pregnancy, 393
cyclosporine, see Cyclosporine
side effects, 484-485 tacrolimus,
265 types of, 483-484 phases of,
484 rationale for, 482
pregnancy considerations, 392-393
procedure, 481 recipient evaluation
blood and tissue typing, 479
gastrointestinal, 479 initial
assessment, 477 placement on United
Organ
Sharing List, 479-480
psychological, 479 screening,
477-478 for specific diseases
cardiovascular, 478-479
cerebrovascular, 479 liver, 478
urologic, 479 recurrent renal
disease risk, 478

Index
rejection causes of, 483 classification of,
482-483 renal osteodystrophy after, 454
retransplantation, outcome of, 488
timing of, 477 Renal tubular acidosis
clinical features, 74-75 distal
classification of, 73 collecting tubule
acidification defects that cause, 7374 hyperkalemic, 74 hypokalemia
associated with, 102 lithiuminduced, 333 proximal, 73
treatment, 74-75 Renal tubular
epithelial cells, in
urine, 32
Renal tubules
acidosis, see Renal tubular acidosis
disease, see Tubulointerstitial disease
distal convoluted description of, 4
diuretics absorption, 122 description
of, 116-117 resistance, 122-123
function of, 9-10 dysfunction
clinical features, 311
pathogenesis
acid excretion, 311 potassium
excretion, 311 proximal tubular
reabsorption, 311 tubular secretion
defects, 311 urine concentration,
310-311 water reabsorption, 310311 treatment, 311
filtrate absorption and secretion, 9
flow rate
diuretic effects on, 17 effect on
potassium excretion, 17 loop of Henle
ascending limbs, 9
countercurrent multiplier, 15
description of, 4 diuretics, see
Loop diuretics illustration of,
11
sodium chloride reabsorption, 116
proximal convoluted description of, 4
filtrate absorption and secretion, 9
glucose transport, 19 illustration of,
10 phosphate transport, 9
potassium excretion, 98 protein
reabsorption, 42 reabsorption
defects, in sickle cell
disease, 311 sodium
absorption, 12 water
absorption, 14 Renin
description of, 3, 13

release, methods that control, 13

for renovascular hypertension
diagnosis, 503 Reninangiontensin system, sodium
regulation, 13 Renin-angiotensin aldosterone system,
185, 502
Renography, angiotensin converting
enzyme inhibitor-stimulated, for
renovascular hypertension diagnosis,
503-504 Renovascular hypertension
clinical features, 502 definition of, 501
diagnostic strategy, 504 diagnostic t ests
angiotensin converting enzyme
inhibitor-stimulated
renography, 503-504
computerized tomographic
angiography, 504 duplex
ultrasound, 504 intravenous digital
subtraction renal
angiography, 503 magnetic
resonance angiography,
504
overview, 503
renin activity measurements, 503
epidemiology, 502 hemodynamic effects,
504 pathophysiology, 501 -502
radionuclide studies for, 49 treatment,
504-505 Replacement therapy
continuous, drug dosing
considerations in, 304-305 for end
stage renal disease, 439 total parenteral,
447 Resistance, diuretic causes of, 120121 in congestive heart failure patients,
186
in nephrotic syndrome patients, 122
treatment approaches, 122-123
Respiratory acidosis acute
causes of, 87
treatment, 89
chronic
causes of, 88
treatment, 89
clinical manifestations, 87 -88
description of, 86 diagnosis of,
88-89 etiology, 87
hemodynamic effects, 88
pathophysiology, 86 secondary
physiological response,
86-87
treatment, 89 Respiratory
alkalosis clinical
manifestations, 90 -91
description of, 89 diagnosis, 91

Index
etiology, 90
pathophysiology, 89-90
secondary physiological response, 90
treatment, 91
Retinol binding protein, 330
Retroperitoneal tumors, urinary tract
obstruction and, 228
Rhabdomyolysis
causes of, 273
drug-induced, 273
exertional, 273
hypoealcemia and, 108
laboratory diagnosis, 273-274
traumatic, 273
Rhabdomyosarcomas, hematuria and, 36
Rheumatic diseases, with renal
involvement
Behcet's disease, 214
Familial Mediterranean fever, 214
Henoch-Schonlein purpura, see
Henoch-Schonlein purpura
mixed connective tissue disease, see
Mixed connective tissue disease
relapsing polychondritis, 214
Sjogren's syndrome, see Sjogren's
syndrome
systemic lupus erythematosus, see
Systemic lupus erythematosus
systemic sclerosis, see Systemic
sclerosis
systemic vasculitis, see Systemic
vasculitides
Rheumatoid arthritis
autoantibodies, 209
characteristics of, 214
renal involvement, 214
treatment, 214
Rifampin, 280
RPN, see Renal papillary necrosis
RTA, see Renal tubular acidosis

SAD, see Selective aldosterone

deficiency
Salicylate intoxication
acidosis associated with, 78
diagnosis of, 78
treatment, 78-79
Saturine gout, and lead exposure, 336
Scleroderma, see Systemic sclerosis
Secondary hyperparathyroidism,
pathogenesis
altered calcium -regulated parathyroid
hormone secretion, 450
calcemic action, skeletal resistance to,
449-450
calcitriol deficiency, 448-449
calcium malabsorption, 449
parathyroid hyperplasia, 450
phosphate retention, 448-449
Selective aldosterone deficiency
clinical presentation, 75
description of, 74

539
Self-tolerance causes of, 144
principles of, 143 Semustine, 230
Septicemia, from vascular access for
hemodialysis, 422
Septic shock, hypoealcemia and, 108
Serum creatinine creatinine
clearance estimations
using, 23 glomerular
filtration rate
measurements, 22 Sexual
functioning, effect of renal
failure on, 474 Shiga-like toxins,
hemolytic uremic
syndrome pathogenesis and, 233
Shunts, glomerulonephritis associated
with, 198 SIAD, see Syndrome of
inappropriate
antidiuresis
Sickle cell disease, renal pathology in
focal segmental glomerulonephritis,
312
gross hematuria clinical features, 309
pathogenesis, 309 treatment, 310
proteinuria diagnosis, 312
pathogenesis, 312 treatment, 312 renal
papillary necrosis, 309 tubular
dysfunction pathogenesis acid
excretion, 311 potassium excretion,
311 proximal tubular reabsorption,
311
tubular secretion defects, 311 urine
concentration, 310-311 water
reabsorption, 310-311 treatment,
311 Sjogren's syndrome
autoantibodies, 209 characteristics
of, 214 treatment, 214
SLE, see Systemic lupus erythematosus
Sleep apnea, in chronic renal failure,
438
Small intestine, potassium excretion, 99
Sodium absorption
calculations of amount absorbed, 12
mechanisms, 12 calcium oxalate stone
formation and,
363

considerations for elderly

conservation deficiencies, 397
excretion, 397
decreased levels, see Hyponatremia
elevated levels, see Hypernatremia
excretion
effect of nonsteroidal antiinflammatory drugs, 292

fractional, 12
in acute interstitial nephritis, 279
in infants and children, 384-385
maintenance of, 114 prerenal vs.
intrinsic acute renal failure
diagnosis using, 251, 257
regulatory mechanisms atrial
natriuretic factor, 13 renin-angiotensin
system, 13 sympathetic nervous
system, 13 vasopressin, 13 intracellular
regulation, 12 metabolism in children
and infants,
384-385 renal handling
abnormalities, in
cirrhosis
clinical features, 188
efferent factors, 189
pathogenesis, 188-189
restrictions
in chronic renal failure patients, 445 in
dialysis patients, 445 Sodium channel
blockers, 117 Sodium nitroprusside, 514
Solutes
peritoneal dialysis, 416 retention in
uremic syndrome factors that influence,
407 hippuric acid, 406 indoxyl sulfate,
406 methylguanidine, 406 organic
phosphates, 407 peptides, 406-407 urea,
405-406 Sotalol, 305 Specific gravity
definition of, 27 measurement methods
description of, 27 dipstick testing, 29
Spironolactone, 117 Squamous cell
carcinoma, of bladder,
375
Squamous cells, in urine, 32-33
Staphylococcus aureus complicated
urinary tract infections,
369-370 peritoneal catheter
infections,
418
Starvation ketoacidosis, 77-78
Steady state, 21 Stones, renal
characteristics of, 360 -361
formation, 361 metabolic
activity, 361-362 passage,
361
urological management, 365
Streptococcal pharyngitis, 194
Streptozotocin, 229
Sulfamethoxazole, 299
Sulfonamides, 280-281 Sympathetic
nervous system, 13

540
Syndrome of inappropriate antidiuresis
description of, 59-60
diagnosis of, 60
etiologies, 60
treatment, 63 Syphilis,
glomerulonephritis associated
with, 198 Systemic lupus
erythematosus

autoantibodies, 209
definition of, 208
diagnosis, 209
signs and symptoms, 208
treatment, 210
Systemic sclerosis

autoantibodies, 209
characteristics of, 212

renal involvement, 212-213

signs and symptoms, 212

treatment, 213
Systemic vasculitides, see also specific
disease
clinical features, 204-205
clinical outcome, 205-206
diagnosis, 205
immunologic causes, 203

pathogenesis, 203
renal involvement, 200, 202
treatment, 205-206

Tagamet, see Cimetidine Takayasu's

arteritis clinical features, 202 renal
pathology, 200 treatment, 205 TammHorsfall glycoprotein, 222-223,
330
TCC, see Transitional cell carcinoma T
cells, 142 Temporal arteritis, see Giant
cell
arteritis
Tetracycline, 305 Thiazide
diuretics clinical uses
hypercalciuria, 363
hypertension, 508
hypothalamic diabetes insipidus, 70
metabolic alkalosis and, 82 Thirst
defects, 65
role in hypernatremia pathogenesis, 65
stimulation of, 65 Thrombotic
microangiopathy, history of,
232
Thrombotic thrombocytopenic purpura
clinical features, 236 epidemiology,
235 management of, 236 pathogenesis,
235 prodrome, 236
Thyroid abnormalities, in renal failure,
472

Index
Thyrotoxicosis, hypercalcemia and, 110
Ticarcillin, 299
TIPS, see Transjugular intrahepatic
porto-systemic shunt Titratable
acidity, 18 TMA, see Thrombotic
microangiopathy Tobramycin, 299, 305
Torsemide, 116, 122 Total body
irradiation, renal
complications of, 230-231 Total
parenteral nutrition, for acute
renal failure, 447
Transitional cell carcinoma, 51, 375
Transjugular intrahepatic porto-systemic
shunt, for hepatorenal syndrome, 192
Transplantation, see Bone marrow
transplantation; Renal
transplantation
Transtubular potassium gradient, 100
Triamterene
complications of, 120 site of action,
117 Trimethoprim, 299 effect on
creatinine secretion, 24
hyperkalemia induced by, 104
Trousseau's sign, 107-108 TTKG,
see Transtubular potassium
gradient TTP, see Thrombotic
thrombocytopenic
purpura
Tuberous sclerosis, 317
Tubular cells casts, 33,
35 in urine, 32
Tubular proteinuria, 44 Tubular
transport maximum for
glucose, 19
Tubules, renal
disease, see Tubulointerstitial disease
distal convoluted description of, 4
diuretics absorption, 122 description
of, 116-117 resistance, 122-123
function of, 9-10 dysfunction
clinical features, 311 pathogenesis
acid excretion, 311 potassium
excretion, 311 proximal tub ular
reabsorption,
311
tubular secretion defects, 311
urine concentration, 310-311
water reabsorption, 310-311
treatment, 311
filtrate absorption and secretion, 9
flow rate
diuretic effects on, 17 effect on
potassium excretion, 17 loop of Henle
ascending limbs, 9

countercurrent multiplier, 15
description of, 4
diuretics, see Loop diuretics
illustration of, 11
sodium chloride reabsorption,
116
proximal convoluted
acidosis
description of, 73 renal
bicarbonate threshold, 73
description of, 4 diuretics
carbonic anhydrase inhibitors,
115-116
mechanism of action, 114-116
types of, 115
filtrate absorption and secretion, 9
glucose transport, 19 illustration
of, 10 necrosis, in pigment
nephropathy,
275
phosphate transport, 9 potassium
excretion, 98 protein reabsorption,
42 reabsorption defects, in sickle
cell
disease, 311 sodium
absorption, 12 water
absorption, 14
Tubuloglomerular feedback, 7
Tubulointerstitial disease cast
nephropathy, see Cast
nephropathy causes of, 330
definition of, 329 and
glomerular filtration rate,
331-332 and primary
glomerular disease,
331-332
Tubulointerstitial nephritis, 148
clinical manifestations, 329
diagnosis
enzyme analysis, 330-331
proteinuria, 329-330
tubular analysis, 331
urinalysis, 329 histology,
329
in juvenile nephronophthisis
medullary cystic disease
complex, 324
Tumor lysis, renal complications of,
231

u
UAG, see Unmeasured anion gap
UAN, see Uric acid nephropathy
Ultrafiltration rate, 408-409
Ultrasonography description of,
47 diagnostic uses, 47
acute renal failure, 258
analgesic nephropathy, 296
inflammatory diseases, 51
nonglomerular hematuria, 40

Index
obstructive uropathy, 353
urolithiasis, 52 Unmeasured
anion gap, 75 UPJ obstruction, see
Ureteropelvic
junction obstruction
Urea
blood urea nitrogen measurements, 23
clearance, 25 effect on countercurrent
multiplier,
15-16
metabolism, 26 renal
handling, 25-26
ultrafiltration rate, 411
Uremia
leukocyte dysfunction, 470-471
management in acute renal failure
patient, 285
neurological manifestations, 460
pericarditis in, 458 peripheral
neuropathy associated
with, see Peripheral neuropathy
platelet dysfunction, 470 Uremic
encephalopathy, 404 clinical
manifestations, 460 definition of, 459
differential diagnosis, 459-460
pathogenesis, 460-461 Uremic
hyperparathyroidism diagnosis of, 450451 prevention and treatment
calcium control, 451
phosphorus control, 451
vitamin D sterol use, 451-452
Uremic syndrome biochemical
changes
drug protein binding, 405
enzymatic and metabolic
dysfunction, 405
characteristics of, 433
clinical features
acidosis, 405
anemia, 403-404
cardiovascular anomalies, 403
coagulation disturbances, 404
endocrine dysfunction, 404
immune deficiency, 404
malnutrition, 405
peripheral neuropathy, 403
pruritus, 405
uremic bone disease, 405
uremic encephalopathy, 403
definition of, 403 hemolytic
acute renal failure etiology during
pregnancy, 394
clinical features, 234
clinical outcome, 235
epidemiology, 232-233
laboratory findings, 234
management of, 234-235
pathogenesis, 233
postdiarrheal, 233-234
during pregnancy, 235
prodrome, 233-234

541
solute retention factors that
influence, 407 hippuric acid, 406
indoxyl sulfate, 406
methylguanidine, 406 organic
phosphates, 407 peptides, 406-407
urea, 405-406 Ureteral tumors
diagnosis, 374 etiology, 373
pathology, 373-374 staging, 374
treatment, 374-375 Ureteropelvic
junction obstruction description of,
351 diagnosis, 353-354 imaging
techniques, 52 Ureterovesical
junction, 356 Uric acid
renal handling of, effect on
development of uric acid
nephropathy, 266 stones, 364
Uric acid nephropathy, 231
clinical outcome, 267 clinical
presentation, 266-267 definition
of, 266 diagnosis, 267
pathogenesis, 266 pathology,
266 treatment
for existing disease, 268
prophylactic approaches, 267
Urinalysis clinical uses acute renal
failure, 257 hematuria, 38
tubulointerstitial disease, 329
specimen collection and handling, 27
spun sediment, microscopic
examination of
bacteria, 35 casts
granular, 33
hyaline, 33 red
blood cell, 33
tubular cell, 33, 35
waxy, 33
white blood cell, 33
cellular elements
erythrocytes, 29, 32
leukocytes, 32
renal tubular epithelial cells, 32
squamous cells, 32-33
characteristic findings, 35 crystals,
34-35 lipiduria, 35 specimen
preparation and
viewing, 29 yeast
organisms, 35 Urinary
bladder, see Bladder Urinary
proteins jS2 -microglobulin, 330

normal values, 330

Tamm-Horsfall glycoprotein, 222-223,
330
Urinary tract
cancer-related obstruction, 228
nonobstructive dilatation description
of, 351-352 diagnosis, 354 obstructive
uropathy, see Obstructive
uropathy
upper, neoplasms of
diagnosis, 374 etiology, 373
pathology, 373-374
staging, 374 treatment,
374-375 Urinary tract
infection complicated
clinical manifestations, 370
diagnosis, 370 differential
diagnosis, 370 etiology, 369 -370
incidence, 369 pathogenesis, 370
prognosis, 370-371 treatment,
370 hematuria and, 36
in medullary sponge kidney, 343-344
nephrolithiasis, 364-365 during
pregnancy, 388 pyuria and, 40
uncomplicated cystitis
antimicrobial prophylaxis, 369 bacterial
etiology, 366 clinical manifestations,
367 differential diagnosis, 367 incidence,
366 prognosis, 369 treatment, 367 -368
definition of, 366 pyelonephritis
diagnosis, 367 prognosis, 369 treatment,
367-368 in vesicoureteral reflex, 358
Urine casts
granular, 33 hyaline,
33 red blood cell, 33
tubular cell, 33, 35
waxy, 33
white blood cell, 33
chemical properties dipstick
reagent strips for
determining
bilirubin, 29
blood, 28-29
glucose, 29
ketones, 29
leukocytes, 29
nitrite, 29 pH,
28

542
Urine (continued) protein,
28 specific gravity, 29
urobilinogen, 29 factors
that affect, 27
concentration
in children and infants, 383
in chronic renal disease, 437
in dehydration, 14
in elderly, 398
in normal functioning, 14
in obstructive uropathy patients,
352
specific gravity measurements, 27 in
tubular dysfunction patients, 310 and
vasopressin secretion, 64 output, for
diagnosis of acute renal
failure, 257-258 physical
properties appearance, 27 specific
gravity, 27 voiding of, 4 volume,
calcium oxalate stone
formation and, 362
Urobilinogen
description of, 29 dipstick testing, 29
Urolithiasis, idiopathic, hyperoxaluria
in, 341-342
US, see Ultrasonography UTI,
see Urinary tract infection

VAD, see Vincristine, with adriamycin

and dexamethasone VAMP, see
Vincristine, with adriamycin
and methylprednisolone
Vancomycin, 263, 299 Vasa
recta, 6

Index

Vasculitides, systemic, see also specific

disease
clinical features, 204-205 clinical
outcome, 205-206 diagnosis, 205
immunologic causes, 203
pathogenesis, 202-204 renal
involvement, 200, 202 treatment,
205-206 Vasculitis,
cryoglobulinemic, 137
Vasoconstrictor mechanisms,
neurohormonal activation of, for
congestive heart failure medical
therapy effects, 186 -187 sodium and
water retention effects,
185
Vasodilators, for hypertension, 510-511
Vasopressin, see also Antidiuretic
hormone
effect on hypertonicity, 64 regulation
of secretion, 64 sodium regulation, 14
VCU, see Voiding cystourethrography
Vesicoureteral reflex, see also Reflux
nephropathy in
children diagnosis,
358
management approaches, 358
definition of, 356 diagnosis, 358
evaluation of, 358-359
management of, 359 pathogenesis,
356-358 primary, 356 secondary,
356 surgical approaches, 359-360
Vincristine
with adriamycin and dexamethasone,
222 with
adriamycin and
methylprednisolone, 222 Vitamin
D deficiency, hypocalcemia and, 108

intoxication, hypercalcemia and, 110

sterols, for uremic
hyperparathyroidism, 451-452
Vitamins, for chronic renal failure
patients, 446
Voiding cystourethrography, 51
Vomiting, metabolic alkalosis and,
82-84
Von Hippel-Lindau disease, 317
VUR, see Vesicoureteral reflex

w
Waldenstrom's macroglobulinemia, 226
Water excretion, see also Urine,
concentration effect of nonsteroidal
anti-inflammatory drugs, 292-293
renal handling abnormalities, in
cirrhosis
clinical features, 189-190
hyponatremia treatment, 190
pathogenesis, 190
retention, in congestive heart failure
afferent mechanisms, 183-184
efferent mechanisms hormonal
vasodilator activation,
185-186 neurohormonal
vasoconstrictor
activation, 185 renal
hemodynamic changes,
184-185
Wegener's granulomatosis
clinical features, 200, 202, 204-205
treatment, 206 White blood cell casts,
33 Wilms tumors, hematuria and, 36