John H.

Stone
Editor

A Clinician’s Pearls and

Myths in Rheumatology

123
John H. Stone, M.D., M.P.H.
Director, Clinical Rheumatology
Massachusetts General Hospital
55 Fruit Street / Yawkey 2
Boston, MA. 02114

ISBN: 978-1-84800-933-2 e-ISBN: 978-1-84800-934-9

DOI: 10.1007/978-1-84800-934-9
Springer Dordrecht Heidelberg London New York

Library of Congress Control Number: 2009933269

© Springer Science+Business Media B.V. 2009

No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by any means,
electronic, mechanical, photocopying, microfilming, recording or otherwise, without written permission from the
Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a
computer system, for exclusive use by the purchaser of the work.

Printed on acid-free paper.

Springer is part of Springer Science+Business Media (www.springer.com)

 Preface

Once when I was an intern, an attending rheumatologist bemoaned the number of decisions he
had to make when caring for a single complex patient. Which dose of prednisone? When to
taper? Which steroid-sparing agent to add, or whether to add one at all? Was an ACE inhibitor
a good idea in a patient with a serum creatinine of 3.5 mg/dL? When to employ Pneumocystis
prophylaxis, and when to stop it? These struck me as highly interesting questions, but as an
intern more concerned that my beeper might sound again any moment to signal my next “hit,”
I took only passing note of the remark and evinced little sympathy for the beleaguered attend-
ing. At least he was going to get some sleep that night!
Some years later, having differentiated into a rheumatologist myself, I pursued training in
clinical investigation, wrote papers, conducted randomized clinical trials, and developed a
stable of complex patients of my own. Only then did I recall the attending’s remark with empa-
thy and observe just how few of the clinical decisions I made were based upon rigorous evi-
dence. Indeed, even if the budget at the National Institutes of Health were once again to double
within a short period of time and then to double again, the highly nuanced nature of rheumatic
disease would yield to “Grade A evidence” on only a minority of important clinical decision
points. In our discipline, there will always remain ample room for the keen clinical “Gestalt.”
This inevitably brings chagrin to advocates of comparative effectiveness studies, among
whom I count myself a member. The application of clinical evidence (when available) to major
treatment decisions is critical to conscientious and effective patient care. But the dozens of
smaller decisions that comprise the craft of medicine are still rooted in a clinician’s direct
experience; in clinical intuition; in nuggets of wisdom handed down from mentors; and in tips
imparted to practitioners by patients themselves. Rheumatology training and practice rely, in
short, on the understanding and application of clinical Pearls. Further, becoming a good clini-
cian and an effective teacher also involves the ability to recognize and debunk Myths: those
specious concepts and harebrained ideas that cling to the body of medical knowledge like gum
to a shoe, despite being fundamentally wrong.
Pearls and Myths are a substantial, ancient, and ever-renewed portion of the medical canon.
When Hippocrates sat beneath the shade of his plane tree on the island of Kos and formulated
aphorisms, he generated Pearls that continue to influence the fabric of medical practice down
to the present day. Rheumatology fellows on the wards today will do well to recall Hippocrates’
Pearl that “A woman does not take the gout, unless her menses be stopped”. And even great
clinicians are not immune to the unwitting propagation of Myths from time to time. No less an
authority than Sir William Osler advocated the use of arsenic for the treatment of pernicious
anemia.
The discipline of rheumatology is more conducive than most to teaching and learning by
Pearls and Myths. Rheumatologists pride themselves on the idea that no subspecialty relies so
heavily upon the history and physical examination – the laying on of hands – for rendering diag-
noses. Astute rheumatologists can leap broadly to speculations on prognosis after an examination
of only a patient’s fingers and hands. At the same time, rheumatology rivals any subspecialty for
its array of diagnostic tests that appear arcane to outsiders: What are the clinical implications of
a high-titer ANA with a speckled pattern? And the oligodot pattern?

v
vi Preface

Pearls and Myths in Rheumatology is a compilation of the wisdom of some of the most
experienced clinicians and insightful clinician-scientists whose work touches upon rheumatic
disease. No fewer than 126 contributors, experts all, have written the Pearls and Myths for this
book:
73 rheumatologists for adults
19 pediatric rheumatologists
2 dermatologists
1 dermatologist/rheumatologist
3 otolaryngologists
4 orthopedists
2 internists
3 pathologists
1 dentist
3 neurologists
1 endocrinologist
2 neuroophthalmologists
1 ophthalmologist
1 ocular immunologist
1 pediatric infectious disease specialist
1 nephrologist
4 pulmonologists
1 clinical immunologist
2 clinical geneticists
1 undifferentiated stem cell (medical student), but an expert nonetheless
Of course, nearly all of these contributors wear more than one hat. They are clinical investiga-
tors, educators, epidemiologists, basic scientists, or accomplished practitioners of other disci-
plines in which they engage when not doing clinical work. They hail from 13 countries: Austria,
Belgium, Canada, Denmark, Germany, Great Britain, Greece, Italy, Japan, the Netherlands,
Spain, Turkey, and the United States. Together, they have written 48 chapters of Rheumatology
Pearls and Myths. Their chapters comprise a total of more than 1,400 such packets of knowl-
edge (836 Pearls and 610 Myths, actually). Their points are driven home by 400 illustrative
elements, including 300 clinical photographs.
Finally, a word about the book’s dedication. Pearls & Myths in Rheumatology is dedicated
to my father, Dr. John Stone III (1936–2008). My father served as the Director of Admissions
at the Emory University School of Medicine for more than 20 years. He was a cardiologist,
poet, teacher – and an extraordinary clinician. Much of what I know about taking a history I
learned from hearing him recount the details of patients he had cared for. As a child, I was
thrilled by opportunities to accompany him to the hospital or on the occasional home visit.
From time to time as a young boy, some of my playmates were his pediatric patients from the
institution that Atlanta once called the “Crippled Children’s Clinic”: children with congenital
or acquired heart disease. As an undergraduate pursuing pre-medical studies, I occasionally
tagged along on Cardiology Consult Rounds, where I strained my ears to discern the pericar-
dial friction rub of a 22-year-old woman with end-stage renal disease caused by lupus. And as
a medical student performing a visiting clerkship at Grady Memorial Hospital and examining
a man alleged to have syphilitic aortitis, I endeavored unsuccessfully to distinguish the poten-
tial Austin Flint murmur of aortic regurgitation from the rumble of mitral stenosis caused by
rheumatic heart disease. (I still wonder if the underlying cause of the patient’s aortic regurgita-
tion was not giant cell arteritis….)
I regret that my father’s superb cardiac auscultation skills, acquired by dint of many years
of hard practice, were not bestowed upon me by right of primogeniture, but it is only fitting that
Pearls and Myths in Rheumatology is devoted to him: in memory of my finest role model and
in celebration of the many cases we enjoyed discussing together. How I wish that those discus-
sions could continue, and go on and on.
Preface vii

I hope you will enjoy, learn from, and pass along the Pearls you deem worthy, all in the
spirit of “See one, do one, teach one.” Moreover, I hope you will see fit to debunk loudly any
Myths described herein (some of which have been passed off as “fact” for generations).
Remember only that clinical medicine, described by Lewis Thomas as “The Youngest Science,”
is ever changing. This will bring, in short order, not only a whole new generation of Pearls and
Myths, but also the realization that some of today’s Pearls are to become tomorrow’s Myths.
The converse is also true. These are the principal purposes, perhaps, of second editions.

John H. Stone, MD, MPH

Massachusetts General Hospital
Boston, Massachusetts
[email protected]

References:

Hippocrates. The genuine works of Hippocrates, volumes 1 and 2. Translated and edited by
Adams F. New York: Wood; 1886.
Osler W, McCrae T. The Principles and Practice of Medicine. 8th Edition. D. Appleton and
Company; 1912.
Thomas L. The Youngest Science: Notes of a medicine-watcher. Viking; 1983.
 Contents

1 Rheumatoid Arthritis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
James R. O’Dell, Josef S. Smolen, Daniel Aletaha, Dwight R. Robinson,
and E. William St. Clair
2 Rheumatoid Vasculitis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
John H. Stone and Eric L. Matteson
3 Adult-Onset Still’s Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Michael H. Weisman
4 Juvenile Idiopathic Arthritis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Patience H. White, Patricia Woo, and Carol B. Lindsley
5 Monogenic Autoinflammatory Syndromes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Marco Gattorno, Alberto Martini, Raphaela Goldbach-Mansky,
Pamela Aubert, and Polly J. Ferguson
6 Juvenile Spondyloarthropathy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
David A. Cabral and Shirley M.L. Tse
7 Ankylosing Spondylitis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Jürgen Braun and David Tak Yan Yu
8 Psoriatic Arthritis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Christopher Ritchlin, Elinor Mody, Philip Mease, and Dafna D. Gladman
9 Reactive Arthritis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Robert D. Inman and Andrew Keat
10 Systemic Sclerosis (Scleroderma) and Raynaud’s Phenomenon . . . . . . . . . . . . . 77
Janet E. Pope, Philip J. Clements, Daniel E. Furst, Laura Hummers,
Dinesh Khanna, Maureen D. Mayes, Thomas Medsger, James Seibold,
and Virginia Steen
11 Nephrogenic Systemic Fibrosis and Other Scleroderma Mimickers . . . . . . . . . 97
Mark A. Perazella, Janet E. Pope, and Shawn E. Cowper
12 Sjögren’s Syndrome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
Manuel Ramos-Casals, Troy E. Daniels, Robert I. Fox, John P. Whitcher,
George E. Fragoulis, Fotini N. Skopouli, and Haralampos M. Moutsopoulos
13 Systemic Lupus Erythematosus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
Michelle Petri, Rachel Abuav, Dimitrios Boumpas, Fiona Goldblatt,
David Isenberg, Grant J. Anhalt, Victoria P. Werth, and R. John Looney
14 Pediatric Systemic Lupus Erythematosus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
Earl D. Silverman

ix
x Contents

15 Mixed Connective Tissue Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169

Robert W. Hoffman
16 The Antiphospholipid Syndrome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
David P. D’Cruz and Munther Khamashta
17 Pregnancy in the Rheumatic Diseases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
Megan E.B. Clowse and Rosalind Ramsey-Goldman
18 Inflammatory Myopathies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
Ira N. Targoff, Chester V. Oddis, Paul H. Plotz,
Frederick W. Miller, and Mark Gourley
19 Juvenile Dermatomyositis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
Ann M. Reed, Clarissa A. Pilkington, Brian M. Feldman,
Lauren M. Pachman, and Lisa G. Rider
20 Vasculitic Neuropathy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
David A. Chad and Peter Siao
21 Pediatric Vasculitis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
Anne H. Rowley, Seza Ozen, Robert P. Sundel, and Frank T. Saulsbury
22 Behçet’s Syndrome. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
Hasan Yazici and Yusuf Yazici
23 The Churg–Strauss Syndrome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
Andrew Churg, Ulrich Specks, and John H. Stone
24 ANCA-Associated Vasculitis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
John H. Stone, Shoichi Ozaki, Karina Keogh, Ulrich Specks,
Carol A. Langford, Niels Rasmussen, Cees G.M. Kallenberg,
and Ingeborg M. Bajema
25 Cryoglobulinemia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
Dimitrios Vassilopoulos
26 Polyarteritis Nodosa. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
John H. Stone
27 Giant Cell Arteritis and Polymyalgia Rheumatica . . . . . . . . . . . . . . . . . . . . . . . . 285
James R. Stone, Misha Pless, Carlo Salvarani, Nicolo Pipitone,
Simmons Lessell, and John H. Stone
28 Takayasu’s Arteritis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305
Eamonn S. Molloy, John H. Stone, Carol A. Langford
29 Primary Angiitis of the Central Nervous System
and Reversible Cerebral Vasoconstriction Syndromes. . . . . . . . . . . . . . . . . . . . . 311
Leonard H. Calabrese and Aneesh B. Singhal
30 Thromboangiitis Obliterans (Buerger’s Disease) . . . . . . . . . . . . . . . . . . . . . . . . . 317
Eric L. Matteson and John H. Stone
31 Less Common Forms of Vasculitis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
Eric L. Matteson and John H. Stone
32 Relapsing Polychondritis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335
Harvinder S. Luthra
33 Fibromyalgia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 341
Daniel Clauw and Don L. Goldenberg
Contents xi

34 The Clinical Features of Gout. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351

Bruce N. Cronstein and Michael H. Pillinger
35 Gout Epidemiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355
Hyon K. Choi
36 Gout Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 361
N. Lawrence Edwards, Terence Gibson, and Robert A. Terkeltaub
37 Calcium Pyrophosphate Dihydrate (CPPD) Crystal Deposition Disease . . . . . . 369
Ann K. Rosenthal and Robert A. Terkeltaub
38 Inflammatory Eye Disease. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377
James T. Rosenbaum and George N. Papaliodis
39 Immune-Mediated Inner Ear Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385
Yuri Agrawal and Howard W. Francis
40 Osteoporosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 393
Kenneth G. Saag, Sarah L. Morgan, and Amy H. Warriner
41 Paget’s Disease of Bone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405
Margaret Seton
42 Sarcoidosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409
Robert P. Baughman and Elyse E. Lower
43 Osteoarthritis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 421
Nancy E. Lane and Roy D. Altman
44 Regional Musculoskeletal Complaints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433
Joseph J. Biundo, W. Neal Roberts, and Atul Deodhar
45 Low Back and Neck Pain. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 447
Rajiv K. Dixit and Joseph H. Schwab
46 Amyloidosis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 461
George H. Sack Jr
47 The Ehlers–Danlos Syndrome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 467
Fransiska Malfait and Anne De Paepe
48 Osteonecrosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 477
Michael G. Zywiel, Mike S. McGrath, and Michael A. Mont
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 485
 Contributors

Rachel Abuav Johns Hopkins University, Department of Dermatology, 600 N. Caroline

Street, Baltimore, MD 21287, USA
Yuri Agrawal Johns Hopkins University, Department of Otolaryngology, 600 N. Caroline
Street, Baltimore, MD. 21287, USA
Daniel Aletaha Medical University of Vienna, Department of Rheumatology, Waehringer
Guertel 18-20, A-1090, Vienna, Austria
Roy D. Altman University of California, Los Angeles, Internal Medicine/Rheumatology,
9854 W. Bald Mountain Court, Agua Dulce, California 91390
Grant J. Anhalt Johns Hopkins University, Department of Dermatology, 600 N. Caroline
Street, Baltimore, MD 21287, USA
Pamela Aubert National Institute of Arthritis & Musculoskeletal Disease, National
Institutes of Health, 10 Center Drive, Bethesda, Maryland 20892-1828 USA
Ingeborg M. Bajema Department of Pathology, Leiden University Medical Center, Leiden,
The Netherlands
Robert P. Baughman University of Cincinnati, Internal Medicine, 1001 Homles,
Eden Avenue, Cincinnati, OH 45267, USA
Joseph J. Biundo Tulane University Health Science Center, 4315 Houma Blvd.,
Ste 303, Metairie, LA 70006, USA
Dimitrios Boumpas University of Crete. Department of Medicine, Greece
Jürgen Braun Ruhr University Bochum, Rheumazentrum Ruhrgebiet,
Landgrafenstrasse 15, Herne NRW 44652, Germany
David A. Cabral British Columbia Children's Hospital, Pediatrics, 4480 Oak Street, V6H
3V4 Canada
Leonard H. Calabrese Cleveland Clinic Foundation, Department of Rheumatology &
Immunology, 9500 Euclid Avenue, Cleveland, OH 44195 USA
David A. Chad Neurology Department, Massachusetts General Hospital,
165 Cambridge Street, Suite 820, Boston, MA 02114, USA
Hyon K. Choi Department of Medicine, Boston University School of Medicine,
650 Albany Street, Suite 200, Boston, MA 02118, USA
Andrew Churg Department of Pathology, University of British Columbia, 2211 Wesbrook
Mall Vancouver, BC, Canada V6T 2B5

xiii
xiv Contributors

William E. St. Clair Medicine Department, Duke University Medical Center,

Box 3874 DUMC, Durham, NC 27710, USA
Daniel Clauw Medicine and Psychiatry, Chronic Pain and Fatique Research Center,
University of Michigan, 24 Frank Lloyd Wright Dr., Domino Farms, Lobby M. Ann Arbor,
MI 48105, USA
Philip J. Clements UCLA Medical School, Box 951670, 1000 Veterans Avenue Rehab
Center, Los Angeles, California 90095-1670 USA
Megan E.B. Clowse Medicine, Division of Rheumatology and Immunology,
Duke University, Box 3535 Trent Dr., Durham, NC 27710, USA
Shawn E. Cowper Yale Nephrogenic Systemic Fibrosis Registry, Yale University School of
Medicine, Yale Dermatopathology Service, 15 York Street, LMP 5031, New Haven, CT
06520-8059, USA
Bruce N. Cronstein NYU Langone Medical Center, Clinical Pharmacology/Rheumatology/
Medicine, 550 First Avenue, NBV16NI, New York, NY 10016, USA
Troy E. Daniels University of California at San Franscisco School of Medicine and
Dentistry, Orofacial Sciences Box 0422, 521 Parnassus Avenue, San Francisco, CA. 94143-
2206
David P. D’Cruz St. Thomas Hospital, Lupus Research Unit, Rayne Institute, Lambath
Palace Road, London SE1 7EH England
Atul Deodhar Oregon Health Sciences University, Division of Rheumatology, Portland,
Oregon 97021
Rajiv K. Dixit Medicine Department, University of California, San Francisco,
120 La Casa Via, Suite 204, Walnut Creek, CA 94598, USA
Anne De Paepe Medical Genetics Department, Ghent University Hospital, De Pintelaan
185, 9000Gent, Belgium
N. Lawrence Edwards University of Florida, Division of Rheumatology 100-277,
Gainesville, Florida 32610 USA
Brian M. Feldman University of Toronto/SickKids, Rheumatology, 555 University
Avenue,Room 8253 Elm Wing, Toronto, Ontario M5G 1X8, Canada
Polly J. Ferguson Department of Pediatrics / Rheumatology, Children's Hospital of Iowa,
200 Hawkins Drive 2534JCP, Iowa City, Iowa 52242
Robert I. Fox Rheumatology Clinic, Scripps Memorial Hospital, 9850 Genesee Avenue,
Suite 910, La Jolla, CA 92037, USA
6565 Caminito Sinnecock, La Jolla, CA 92037, USA
George E. Fragoulis National University of Athens Medical School Department of
Pathophysiology 75 Mikras Asias Str Athens 115 27 Greece
Howard W. Francis Johns Hopkins University School of Medicine,
Otolaryngology – Head and Neck Surgery, 601 N. Caroline Street, JHOC 6251,
Baltimore, MD 21287, USA
6008 Lakehurst Dr., Baltimore, MD 21210, USA
Daniel E. Furst UCLA Medical School, Box 951670, 1000 Veterans Avenue Rehab Center,
Los Angeles, California 90095-1670 USA
Marco Gattorno University of Genova, Pediatria H, Instituto G. Gaslini,
Genova 16147, Italy
Contributors xv

Terence Gibson Medicine Department, Guy’s and St. Thomas’ Hospitals,

60B Copers Cope Road, Beckenham, Kent BR31RJ, UK
Department of Rheumatology, Guy’s Hospital, London SE1 9RT, UK

Dafna D. Gladman Toronto Western Hospital 399 Bathurst St. RM1E-4OB Toronto,
Ontario M5T 2S8 Canada

Fiona Goldblatt Royal Adelaide Hospital North Terrace, Adelaide SA 5000, AUSTRALIA

Raphaela Goldback-Mansky National Institute of Arthritis & Musculoskeletal Disease,

National Institutes of Health, 10 Center Drive, Bethesda, Maryland 20892-1828 USA
Don L. Goldenberg Newton-Wellesley Hospital Newton, Massachusetts 02462 USA
Mark Gourley National Institute of Arthritis and Musculoskeletal Diseases,
National Institutes of Health, 10 Center Drive MSC 1616 Bethesda, Maryland 20892-1616

Robert W. Hoffman Chief Division of Rheumatology and Immunology,

University of Miami Miller School of Medicine, Medicine, 1120 N.W. 14th Street,
Suite 968, Miami, FL 33101, USA

Laura K. Hummers Johns Hopkins Scleroderma Center, Johns Hopkins University,

5501 Hopkins Bayview Circle, Baltimore, MD 21224, USA

Robert D. Inman Rheumatology, University of Toronto and Toronto Western Hospital,

399 Bathurst Street, Toronto, Ontario M5T 2S8, Canada

David Isenberg University College of London Center for Rheumatology Research Windeyer
Building Room 331 London W1T 4JF England

Cees G.M. Kallenberg University Medical Groningen Clinical Immunology/Rheumatology

T3 242 Hanzeplein 1 Groningen 9713 GZ The Netherlands

Andrew Keat Northwick Park Hospital, Watford Rd., Harrow HA1 3UJ, England

Karina Keogh Division of Pulmonary and Critical Care Medicine, May Clinic,200 First
Street, SW, Rochester, MN 55905, USA

Munther Khamashta The Lupus Research Unit, The Rayne Institute,

Kings College School of Medicine, Street Thomas’ Hospital, London SE1 7EH, UK

Dinesh Khanna UCLA Medical School, Box 951670, 1000 Veterans Avenue Rehab Center,
Los Angeles, California 90095-1670 USA

Nancy E. Lane Medicine, University of California at Davis Medical School,

4800 Second Avenue, Suite 2600, Sacramento, CA 95817, USA

Carol A. Langford Cleveland Clinic Foundation, Department of Rheumatology &

Immunology, 9500 Euclid Avenue, Cleveland, OH 44195 USA

Simmons Lessell Massachusetts Eye & Ear Infirmary, Boston, Massachusetts, USA

Carol B. Lindsley Chief Pediatric Rheumatology, University of Kansas Medical Center,

Pediatrics, 3901 Rainbow, Kansas City, KS 66160, USA

John R. Looney University of Rochester Division of Rheumatology 601 Elmwood Avenue

Rochester, New York 14642 USA

Elyse E. Lower University of Cincinnati, Internal Medicine, 1001 Holmes, Eden Avenue,
Cincinnati, OH 45267, USA
xvi Contributors

Harvinder S. Luthra Rheumatology Department, Mayo Medical School,

200 First Street SW, Rochester, MN 55905, USA
Mike S. McGrath Sinai Hospital of Baltimore, Center for Join Preservation and
Reconstruction, Rubin for Advanced Orthopedics, 2401 W. Belvedere Avenue, Baltimore,
MD 21215, USA
Alberto Martini University of Genova, Pediatria H, Instituto G. Gaslini, Genova 16147,
Italy
Eric L. Matteson Mayo Clinic, Division of Rheumatology, 200 First Street SW,
Rochester, MN 55095, USA
Maureen D. Mayes University of Texas-Houston Health Science Center/Rheumatology
6431 Fannin MSB 5.270 Houston, Texas 77030, USA
Philip Mease Seattle Rheumatology Associates 1101 Madison Street, Suite 1000 Seattle,
Washington 98104
Thomas Medsger University of Pittsburgh Medicine/Rheumatology 3500 Terrace Street 7S
BST Pittsburgh, Pennsylvania 15261
Fransiska Malfait Medical Genetics Department, Ghent University Hospital,
De Pintelaan 185, 9000Gent, Belgium
Frederick W. Environmental Autoimmunity Group National Institutes of Health / NIEHS
Building 10-2330 9000 Rockville Pike MSC 1301Bethesda, Maryland 20892-1301
Elinor Mody Brigham & Women's Hospital Division of Rheumatology Boston,
Massachusetts 02115
Eamonn S. Molloy Cleveland Clinic, Rheumatic and Immunologic Diseases,
Desk A50, 9500 Euclid Avenue, Cleveland, OH 44195, USA
Michael A. Mont Sinai Hospital of Baltimore, Center for Join Preservation and
Reconstruction, Rubin for Advanced Orthopedics, 2401 W. Belvedere Avenue,
Baltimore, MD 21215, USA
Sarah L. Morgan Division of Clinical Nutrition Department of Nutrition Sciences1714 9th
Ave S # Lrc354 Birmingham, AL 35205-3606
Haralampos M. Moutsopoulos National University of Athens Medical School Department
of Pathophysiology 75 Mikras Asias Str Athens 115 27 Greece
Chester V. Oddis University of Pittsburgh Medicine/Rheumatology 3500 Terrace Street 7S
BST Pittsburgh, Pennsylvania 15261
James R. University of Nebraska Medical Center Department of Internal Medicine Omaha,
Nebraska 68198-3025 USA
Shoichi Ozaki St. Marianna University School of Medicine Department of Internal
Medicine 2-16-1 Sugao Miyamae-ku Kawasaki 216-8511 Japan
Seza Ozen Department of Pediatric Nephrology and Rheumatology, Hacettepe University
Faculty of Medicine, Sihhiye 06100 Ankara, Turkey
Lauren M. Pachman Department of Pediatrics, Division of Rheumatology,
Children’s Memorial Research Center, Children’s Memorial Hospital, Northwestern
University, Feinberg School of Medicine, 2340 N. Halsted Street, Box 212, Chicago,
IL 60614-4314, USA
George N. Papaliodis Massachusetts Eye & Ear Infirmary Ocular Immunology Clinic
Boston, Massachusetts 02114
Contributors xvii

Mark A. Perazella Yale University Medical Center Division of Nephrology New Haven,
Connecticut USA
Michelle Petri Medicine, Johns Hopkins University School of Medicine,
1830 East Monument Street, Suite 7500, Baltimore, MD 21205, USA
Clarissa A. Pilkington Great Ormond Street Rheumatology Department University College
London London England
Michael H. Pillinger NYU Medical Center, Rheumatology/Medicine, 550 First Avenue,
New York, NY 10016, USA
Nicolo Pipitone Arcispedale S Maria Nuova Viale Risorgimento 80 42100 Reggio Emilia
Italy
Misha Pless Massachusetts General Hospital Department of Neurology55 Fruit
StreetBoston, Massachusetts 02114
Paul H. Plotz Arthritis and Rheumatism Branch, National Institute of Arthritis and
Musculoskeletal and Skin Diseases, National Institutes of Health, 9000 Rockville Pike,
Bethesda, MD 20892, USA
Janet E. Pope St. Joseph Health Care London 268 Grosvenor St. London, Ontario N6A 4V2
Canada
Manuel Ramos-Casals Servei de Malalties Autoimmunes, Hospital Clínic, C/Villarroel,
170, 08036-Barcelona, Spain
Rosalind Ramsey-Goldman Northwestern University Division of Rheumatology FSM-300
240 E. Huron Street Chicago, Illinois 60611 USA
Niels Rasmussen Rigshospitalet, Otolaryngology – Head and Neck Surgery, Blegdamsvej 9,
Copenhagen 2100, Denmark
Ann M. Reed Mayo Clinic 200 First Street SW Rochester, Minnesota 55905 USA
Lisa G. Rider Environmental Autoimmunity Group National Institutes of Health / NIEHS
Building 10-2352 9000 Rockville Pike MSC 1301 Bethesda, Maryland 20892-1301
Christopher Ritchlin Rheumatology Department, University of Rochester,
601 Elmwood Avenue, Box 695, Rochester, NY 14642, USA
4459 Middle Cheshire Road, Canandaigua, NY 14424, USA
Neal W. Roberts Medical College of Virginia Department of Internal Medicine 11th &
Marshall - Box 9890647 Richmond, VA. 23298-0001
James T. Rosenbaum Department of Ophthalmology, Division of Arthritis and Rheumatic
Diseases, Oregon Health and Science University, Casey Eye Institute,
2275 SW Terwilliger Blvd, Portland, OR 97239, USA
Ann K. Rosenthal Medical College of Wisconsin Division of Rheumatology 9200 W.
Wisconsin Avenue Milwaukee, Wisconsin 53226
Anne H. Rowley Department of Microbiology/Immunology Ward 12-140, Pediatrics
W-140, Northwestern University Medical School, Chicago, IL 60611-3008, USA
Kenneth G. Saag Department of Medicine, Division of Clinical Immunology and
Rheumatology, University of Alabama at Birmingham, 510 Twentieth Street,
South, Suite 820, Birmingham, AL 35233, USA
FOT 820, 1530 Third Avenue, South Birmingham, AL 35294-3408, USA
George H. Sack Johns Hopkins University School of Medicine, Medicine and Biological
Chemistry, 725 N. Wolfe Street, Physiology G15, Baltimore, MD 21205, USA
xviii Contributors

Carlo Salvarani Arcispedale S Maria NuovaViale Risorgimento 80 42100 Reggio Emilia

Italy
Frank T. Saulsbury University of Virginia Medical Center Department of Pediatrics P.O.
Box 800386 Charlottesville, Virginia 22908 USA
Joseph H. Schwab Massachusetts General Hospital Department of Orthopedics 55 Fruit
Street Boston, Massachusetts 02114
James Seibold University of Michigan Medical Center Division of Rheumatology Ann
Arbor, Michigan USA
Margaret Seton Rheumatology, Allergy and Immunology, Harvard Medical School,
Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114, USA
Peter Siao Neurology Department, Massachusetts General Hospital,165 Cambridge Street,
Suite 820, Boston, MA 02114, USA
Earl D. Silverman Pediatrics, Division of Rheumatology, Hospital for Sick Children,
University of Toronto, 555 University Avenue, Toronto, Ontario M5G 1X8, Canada
Aneesh B. Singhal Neurology, MGH Stroke Research Center, Harvard Medical School,
Massachusetts General Hospital, 175 Cambridge Street, Suite 300, Boston, MA 02114, USA
Fotini N. Skopouli National University of Athens Medical School Department of
Pathophysiology 75 Mikras Asias Str Athens 115 27 Greece
Josef S. Smolen Medical University of Vienna, Department of Rheumatology, Waehringer
Guertel 18-20, A-1090, Vienna, Austria
Ulrich Specks Division of Pulmonary and Critical Care Medicine, May Clinic,
200 First Street, SW, Rochester, MN 55905, USA
Virginia Steen Georgetown University Medical Center Department of Rheumatology 3800
Reservoir Road Washington, D.C. 20007 USA
John H. Stone Massachusetts General Hospital, 55 Fruit Street / Yawkey 2, Rheumatology
Clinic, Boston, Massachusetts, 02114
James R. Stone Massachusetts General Hospital Department of Pathology 55 Fruit Street
Boston, Massachusetts 02114
Robert P. Sundel Children’s Hospital, Harvard Medical School, 300 Longwood Avenue,
Boston, MA 02115, USA
Ira N. Targoff Oklahoma Medical Research Foundation Arthritis & Immunology 825 NE
13th Street Oklahoma City, Oklahoma 73104 USA
Robert A. Terkeltaub VA Division of Rheumatology, VA Healthcare San Diego,
University of California, San Diego, 3350 La Jolla Village Drive 111k, San Diego,
CA 92161, USA
Shirley M.L. Tse University of Toronto/SickKids, Rheumatology, 555 University Avenue,
Room 8253 Elm Wing, Toronto, Ontario M5G 1X8, Canada
Dimitrios Vassilopoulos Athens University School of Medicine, Hippokration General
Hospital, 2nd Department of Medicine, 114 Vass. Sophias Avenue, 115 27 Athens, Greece
Amy H. Warriner Division of Endocrinology & Metabolism 2000 6th Ave S #
4Birmingham, AL 35233-2110
Contributors xix

Michael H. Weisman Division of Rheumatology, Cedars Sinai Medical Center,

8700 Beverly Blvd., Room B131, Los Angeles, CA 90201, USA
Victoria P. Werth University of Pennsylvania Dermatology 3600 Spruce Street / Rhoads
Pavilion Philadelphia, Pennsylvania 19104 USA
John P. Whitcher University of Californa-San Francisco Department of Ophthalmology
San Francisco, California
Patience H. White Arthritis Foundation Department of Public Health 2011 Pennsylvania
Avenue, Suite 610 Washington, D.C. 20006 USA
Patricia Woo Centre for Paediatric and Adolescent Rheumatology, University College
London, London England
Yusuf Yazici New York University Hospital for Joint Diseases Division of Rheumatology
246 East 20th Street New York, New York 10003 USA
Hasan Yazici University of Istanbul/Cerrahpasa Medical Faculty,
Division of Rheumatology, Department of Medicine, Askaray, Istanbul 34098, Turkey
Safa sok. 17/7, Kadikoy, Istanbul 34170, Turkey
David Tak Yan Yu UCLA Medical School, Box 951670, 1000 Veterans Avenue Rehab
Center, Los Angeles, California 90095-1670 USA
Michael G. Zywiel Sinai Hospital of Baltimore, Center for Join Preservation and
Reconstruction, Rubin for Advanced Orthopedics, 2401 W. Belvedere Avenue,
Baltimore, MD 21215, USA
xx Contributors
 Rheumatoid Arthritis
1
James R. O’Dell, Josef S. Smolen, Daniel Aletaha, Dwight R. Robinson,
and E. William St. Clair

1.1.1 Diagnosis
1.1 Overview of Rheumatoid Arthritis
Myth: A patient must fulfill at least four of the seven American
› Rheumatoid arthritis (RA) affects all ethnic groups.
Rheumatism Association (now called the American College
Women are nearly three times more likely than men to
of Rheumatology; ACR) classification criteria to be diag-
develop the disease.
nosed as having rheumatoid arthritis (RA).
› The pattern of arthritis typically favors distal and sym-
metrical involvement. Reality: Diagnostic criteria do not exist for RA. The ACR
› The most commonly involved joints are the wrists, classification criteria developed in 1987 were not intended
metacarpophalangeal, proximal interphalangeal, and for the purpose of diagnosing individual patients, but rather
metatarsophalangeal joints. However, many other joints for defining patient populations that are comparable across
can also be involved. Shoulder, elbow, hip, knee, or neck different clinical studies (Table 1.1) (Arnett 1988).
disease (particularly at the atlanto-axial joint, C1–C2) In addition, the RA classification criteria were developed
are frequently observed. in patients with established RA, not in patients with early
› Most presentations are subacute in nature, with the inflammatory arthritis. Thus, they have limited applicability
insidious onset of fatigue, morning stiffness, and arthri- for classifying patients with early disease, who may lack sev-
tis. More explosive onsets of disease are also described. eral of the relevant criteria (e.g., erosions or rheumatoid nod-
› If untreated, RA is a chronic, progressive disorder that ules) (Aletaha et al. 2005b). Among patients with early RA,
leads to joint damage, disability, and early mortality. many do not fulfill at least four of the seven ACR criteria.
› A variety of extraarticular features are typical of “sero- Conversely, many patients with early arthritis have diagnoses
positive” RA (RA associated with the presence of rheu- other than RA yet do fulfill the criteria, at least temporarily
matoid factor in the serum). These include rheumatoid (Harrison et al. 1998; Machold 2006). The gold standard for
nodules, secondary Sjögren’s syndrome, interstitial the diagnosis of RA continues to be the clinical pronounce-
lung disease, scleritis, and rheumatoid vasculitis. ment by a rheumatologist.
› Approximately 70% of patients with RA are rheuma-
toid factor positive. An approximately equal percent-
age has antibodies directed against cyclic citrullinated
1.1.2 Clinical Features
peptides (i.e., anti-CCP antibodies). There is substan-
tial but not complete overlap between groups of
patients who are rheumatoid factor positive and those Myth: Rheumatoid arthritis patients rarely get gout.
who have anti-CCP antibodies. Reality: Adequate data on this association are lacking, but
› Some patients have RA that appears in every way to be this myth is almost certainly an artifact of mid-twentieth cen-
typical disease yet do not have either rheumatoid fac- tury RA treatment regimens. High-dose aspirin, a uricosuric
tor or anti-CCP antibodies. These patients are said to agent, was used in majority of RA patients in that era. With
have “seronegative RA.” the development of more effective treatment regimens for
› Radiographic studies in RA reveal joint space narrow- RA that do not involve high-dose aspirin, it is not rare to see
ing, erosions, deformities, and periarticular osteopenia. the two diseases in the same patient.
› Treatment approaches now emphasize early interven-
Myth: RA burns out with time in many patients.
tions designed to suppress joint inflammation entirely
as soon as possible after the onset of clinical disease. Reality: Some clinicians refer to “burned out” RA. Patients to
whom this monicker is applied generally have long-standing

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 1

DOI:10.1007/978-1-84800-934-9_1, © Springer Science + Business Media B.V. 2009
2 J. R. O’Dell et al.

Table 1.1 1987 Revised ARA criteria for the classification of rheumatoid i.e., patients who show no radiologic evidence of damage can
arthritis have flagrant clinical signs of ongoing synovitis. As an exam-
For classification purposes, a patient is said to have RA if he or she ple, clinical trials of denosumab (an antiRANK-ligand mono-
has satisfied at least four of the following seven criteria. Criteria 1–4
must have been present for at least 6 weeks. Patients with two clonal antibody) have demonstrated that this medication halts
clinical diagnoses are not excluded. Designation as classic, definite, radiographic progression but exerts comparatively little effect
or probable RA is not to be made on clinical measures of disease activity (Cohen et al. 2008).
1. Morning stiffness
Myth: Rheumatoid nodules are an important finding in
Morning stiffness in and around the joints, lasting at least 1 h
before maximal improvement early RA.
2. Arthritis of three or more joint areas Reality: Rheumatoid nodules are one of the seven ACR crite-
At least 3 joint areas simultaneously have had soft tissue swelling ria for established RA. However, rheumatoid nodules are
or fluid (not bony overgrowth alone) observed by a physician; the
14 possible joint areas are right or left proximal interphalangeal usually not present within the first few months of the onset of
(PIP) joints, metacarpophalangeal (MCP) joints, wrist, elbow, synovitis. Thus, their absence at presentation does not reli-
knee, ankle, and metatarsophalangeal (MPT) joints ably exclude the diagnosis of RA.
3. Arthritis of hand joints
Pearl: Troublesome rheumatoid nodules can be injected with
At least 1 area swollen (as defined above) in a wrist, MCP or PIP
joint glucocorticocoids.
4. Symmetric arthritis Comment: Nodules that are painful or interfere with function
Simultaneous involvement of the same joint areas (see 2 above) require an intervention. Direct injection of the nodule with a
on both sides of the body (bilateral involvement of PIPs, MCPs,
or MTPs is acceptable without absolute symmetry)
mixture of local anesthetic and glucocorticoids may be ben-
eficial. Surgical excision is necessary for some nodules.
5. Rheumatoid nodules
Subcutaneous nodules, over bony prominences, or extensor Glucocorticoid injections of nodules are often effective in
surfaces, or in juxta-articular regions, observed by a physician decreasing the size of these lesions. This Pearl was illustrated
6. Serum rheumatoid factor in a study of 24 nodules in 11 patients with RA (Ching 1992).
Demonstration of abnormal amounts of serum rheumatoid factor Each injection was assigned randomly to be performed with
by any method for which the result has been positive in <5% of glucocorticoids or with placebo. Among the nodules injected
normal control subjects
with glucocorticoids, 9 of 12 shrank by more than 50%,
7. Radiographic changes
compared with only 1 of 12 injected with placebo.
Radiographic changes typical of RA on posteroanterior hand and
wrist radiographs, which must include erosions or unequivocal The procedure for injecting rheumatoid nodules is as follows.
bony decalcification localized to or most marked adjacent to the Up to 0.3 mL of methylprednisolone or triamcinolone hexac-
involved joints (osteoarthritis changes alone do not qualify) etonide 40 mg/mL is mixed in 1:1 ratio with 1% lidocaine.
Following local anesthesia of the overlying skin, the needle
is inserted directly into the center of the nodule. The medica-
disease, severe joint damage, and little evidence of synovitis.
tion is injected slowly as the needle is withdrawn.
Most are on minimal therapy for RA, albeit some have tried
Considerable force on the plunger of the syringe is needed in
multiple agents in the past.
some cases to express the mixture into the nodule.
Most patients with “burned out” RA have at least moder-
ately elevated acute phase reactants. More importantly, they Myth: Rheumatoid nodules occur in the heart and cause con-
often continue to deteriorate radiographically, despite having duction defects.
symptoms that are minimal when compared with patients Reality: Although mentioned in most textbooks as an extra-
with early disease. If patients with “burned out” RA are articular complication of RA, there is only one case reported
treated, their symptoms improve. Thus, the label of “burned in the world’s literature (Thery 1974). Don’t expect to see this
out” applies only on rare occasions. Strong consideration vanishingly rare complication in a clinic near you any time
should be given to treating patients who fit the operative defi- soon.
nition of this condition with DMARDs.
Pearl: Rheumatoid pleural effusions can mimic empyemas.
Myth: Clinically detectable synovitis and radiographic pro-
Comment: RA is one of the few causes of a low pH in pleural
gression in RA are linked tightly, such that the control of one
effusions. Two other major ones are empyema and esophageal
leads to control of the other.
rupture. Pleural effusions in RA are exudates, characterized
Reality: This impression is a classic example of dogma that by high protein and lactate dehydrogenase levels and very low
was believed widely for decades, but which is now recognized glucose. White blood cell counts are frequently elevated in
as false. Several trials have shown that excellent control of clin- RA effusions. In contrast to empyemas, lymphocytes tend to
ical features of disease activity may be achieved in the face of predominate over neutrophils unless the fluid collection is
continued radiographic deterioration. The converse is also true; tapped early (Avnon et al. 2007).
1 Rheumatoid Arthritis 3

Pearl: Amyloidosis and possibly rheumatoid vasculitis pose Jaccoud’s arthropathy can be differentiated from RA
less of a threat than they did before the availability of effec- radiographically. Bone erosions are absent in Jaccoud’s,
tive therapy. even in patients with longstanding disease and severe joint
deformities (Fernandez et al. 2006). In one study, MRI
Comment: Two potentially lethal complications of RA are
revealed images described as bony erosions, but these lesions
the development of secondary amyloidosis and the occur-
were not shown to progress to bone loss (Ostendorf et al.
rence of rheumatoid vasculitis. Secondary amyloidosis
2003). When patients with SLE do develop bony erosions,
caused by uncontrolled RA is now rarely seen in countries
their disorder is often termed “rhupus” (Van Vugt et al.
where modern therapies are available. Even before the age of
1998). Such patients have serological features of both RA
biologic therapies for RA, the adoption of methotrexate as
and SLE (Chan 2008).
the standard of care in the 1980s had led to improvements in
disease control and patient survival (Choi 2002). Pearl: Calcium pyrophosphate dihydrate (CPPD) arthropa-
Most RA experts believe that cases of rheumatoid vasculitis thy can mimic rheumatoid arthritis.
are also now less common than in the past. However, as
Comment: CPPD arthropathy often presents as pseudogout,
described in the chapter on rheumatoid vasculitis, the epide-
usually in a monoarticular pattern and most commonly in the
miologic data on this issue are mixed (see Chapter 2, Rheu-
knee. However, chronic polyarticular disease can also occur,
matoid Vasculitis).
with variable degrees of synovitis. The radiographic and clin-
ical examination may be similar to osteoarthritis, but polyar-
ticular CPPD disease involves joints that are not typically
affected in osteoarthritis, such as the wrist, shoulder, and
1.1.3 Rheumatoid Arthritis Mimickers metacarpophalangeal joints (Fig. 1.1). Involvement of these
joints is much more typical of RA.
Myth: Gout and RA are always easily differentiated from
each other.
Reality: Acute gout usually involves the joints of the lower
extremities and is episodic. In addition, it tends to involve
one joint or only a few (i.e., it is oligoarticular). These fea-
tures contrast clearly with typical RA. However, chronic
tophaceous gout can cause persistent inflammatory disease
of multiple joints of both upper and lower extremities, a con-
dition that mimics RA more closely. Tophi, which occur
typically over bony prominences, can be mistaken for rheu-
matoid nodules. More than one seasoned clinician has been
fooled by gout and mistaken this disorder for RA (Doherty
and Dieppe 1986; De Souza 2005).
Myth: Symmetrical polyarthritis at presentation is the rule
for RA.
Reality: Classic RA involves the joints in a symmetrical
manner, affecting the joint areas on both sides of the body in
a relatively equal manner. However, in early disease, sym-
metry is often not apparent (Rantapaa-Dahlqvist 2005; Masi
1983). Approximately one-third of patients with RA have
single joint disease as the initial manifestation.
Conversely, symmetrical swelling and deformity of periph-
eral joints including the metacarpophalangeal joints are not
always due to RA. As an example, Jaccoud’s arthropathy is
seen in patients with rheumatic fever and SLE. Jaccoud’s
arthropathy is characterized by swelling, ulnar deviation, and
volar subluxation of the MCP joints. Such changes may mimic Fig. 1.1 Involvement of the wrist and metacarpophalangeal joints in a
RA closely. The deformity in SLE results from ligamentous patient with calcium pyrophosphate dehydrate (CPPD) deposition
laxity rather than the destructive synovial pannus of RA. disease
4 J. R. O’Dell et al.

a b

Fig. 1.2 (a) Articular cartilage loss and subchondral cysts are common arthritis. Knee radiographs showed CPPD deposition within cartilage
in CPPD deposition disease. (b) This patient also had pronounced (Courtesy of Dr Greg Emkey)
swelling the ulnar styloid region, a finding common to rheumatoid

Articular cartilage loss and subchondral cysts are com- within 2 months. Virtually, all patients’ joint symptoms have
mon in CPPD deposition disease (Fig. 1.2), but the marginal subsided by 6–12 months.
erosions of RA are not found. The finding of CPPD deposi-
Pearl: Not all destructive polyarthritis is rheumatoid or pso-
tion disease in a young patient raises the spectre of hemo-
riatic arthritis.
chromatosis (Axford 1991).
Comment: Erosive osteoarthritis is a subset of osteoarthritis
Pearl: Parvovirus infection can mimic RA precisely, even
characterized by destructive disease of the proximal and dis-
down to the presence of autoantibodies.
tal interphalangeal joints. The joints develop osteophytes,
Comment: Before starting methotrexate for an acute, sym- often accompanied by inflammatory pain and swelling. There
metrical polyarthritis affecting small joints of the hands, are histologic changes of inflammation in the synovium
wrists, and feet, one is strongly advised to check serolo- (Bellhorn and Hess 1993).
gies for parvovirus B. Parvovirus B infection has the The distal interphalangeal joint involvement characteristic
uncanny ability to pose as RA. Acute infections with this of erosive osteoarthritis can cause particular confusion with
virus tend to cause the classic “slapped cheek” skin rash psoriatic arthritis. Bony erosions are evident characteristi-
(“fifth disease” Young and Brown 2004) in children. Fifth cally in the center of the joints on plain radiographs (Fig. 1.3).
disease is not associated with significant joint manifesta- This is accompanied by sclerosis of the remaining subchon-
tions in the majority of pediatric patients. However, the dral bone, giving a gull-winged appearance typical of osteoar-
opposite is true in adults, in whom cutaneous features are thritis. This appearance differs from the marginal erosions
unusual but the arthropathy is paramount (Young and that occur in rheumatoid arthritis (Fig. 1.4). Loss of motion
Brown 2004). and fusion of interphalangeal joints can occur in erosive
Parvovirus infection not only causes an arthropathy com- osteoarthritis, but metacarpophalangeal joint involvement is
patible with RA in adults, it can also be associated with a uncommon.
host of autoantibodies (e.g., rheumatoid factor) that provide Acute phase reactants, rheumatoid factor, and anti-CCP
further misleading evidence of RA or another connective tis- antibody assays are typically normal in erosive osteoarthri-
sue disease such as lupus (Lunardi et al. 2008). In this set- tis, further helping to differentiate this entity from RA and
ting, however, a positive IgM assay for antibodies to psoriatic arthritis (Morozzi et al. 2005). Finally, the symp-
parvovirus leads to the correct diagnosis. Fortunately, the toms of EO fail to respond to immunomodulatory agents
great majority of symptomatic parvovirus infections subside such as methotrexate.
1 Rheumatoid Arthritis 5

Fig. 1.5 Multicentric reticulohistiocytosis is usually accompanied by

erythematous, nodular swelling of the skin that has been referred to as
a “coral beads” (Courtesy of Dr Jeffrey Callen)

There appears to be a strong heritable component to

erosive osteoarthritis, with expression of the trait particularly
common in women in successive generations within fami-
lies. The same is true for “conventional” osteoarthritis of the
hands.
Fig. 1.3 Erosive osteoarthritis. Bony erosions are evident characteristi- Pearl: For a real zebra, consider the possibility of multi-
cally in the center of the joints, rather than at the margins. Note the centric reticulohistiocytosis in a case of aggressive, polyar-
sparing of the MCP joints, typical of osteoarthritis, but the involvement ticular joint disease.
of the distal and proximal interphalangeal joints and the degenerative
changes at the base of the thumb Comment: Reticulohistocytosis causes a symmetrical polyar-
thritis that focuses on the small joints of the hands and there-
fore often resembles aggressive RA (Kalajian and Callen
2008). A distinctive histiocytic infiltration of the synovium
that is associated with multinucleated giant cells leads to
destruction of the joints and juxta-articular bone. Multicentric
reticulohistiocytosis is usually accompanied by erythema-
tous, nodular swelling of the skin that has been referred to as
a “coral beads” (Fig. 1.5).
Multicentric reticulohistiocytosis is often accompanied
by an underlying malignancy. The disease is usually refrac-
tory to conventional anti-inflammatory medications, but spo-
radic reports describe responses to antiTNF agents.

1.1.4 Serological Features and Radiology

Pearl: High titers of rheumatoid factor (RF) are of diagnos-

tic and prognostic value.
Comment: RA is characterized by the presence of autoantibod-
ies, among which RF was the first described. The finding of a
serum RF is relatively non-specific in the sense that it may be
Fig. 1.4 Marginal joint erosions characteristic of rheumatoid arthritis associated with a variety of other conditions that can cause an
6 J. R. O’Dell et al.

inflammatory arthritis, particularly connective tissue diseases et al. 2002). Whether smoking independently worsens estab-
and chronic hepatitis. In addition, up to 15% of the healthy lished disease remains controversial, as retrospective case-
elderly population have RF, albeit usually in low titers. control studies have shown divergent results on radiographic
Several studies have shown that a high cut-off point for a progression (Finckh et al. 2007).
positive assay (e.g., 50 IU/mL) enhances the positive predic- Cigarette smoking also has been associated with an
tive value of RF for RA (Nell et al. 2005; Sinclair and Hull increased risk of anti-CCP antibody-positive RA among indi-
2003). One meta-analysis of 50 studies that evaluated the viduals who have the HLA-DRB1 shared epitope alleles
operating characteristics of RF pegged the sensitivity of this (Klareskog et al. 2006). These results suggest a model in
test at 69% and its specificity at 85% (Nishimura et al. 2007). which smoking triggers anti-CCP antibodies in patients with
High titers of RF are a marker for patients at risk for aggres- RA and the shared epitope. However, other studies have failed
sive, destructive joint disease and extra-articular complica- to confirm the links among smoking, the shared epitope, and
tions of RA, such as interstitial lung disease and rheumatoid anti-CCP antibodies (Lee et al. 2007; Mikuls et al. 2008).
vasculitis.
Assays for anti-CCP antibodies have a sensitivity that
rivals that of RF, but a higher specificity (Schellekens et al. 1.1.6 Disease Assessment
2000; Nell et al. 2005). The sensitivity and specificity of anti-
CCP antibodies in the meta-analysis mentioned above were
67% and 95%, respectively (Nishimura et al. 2007). Moreover, Pearl: A simple clinical index that does not depend on labo-
anti-CCP antibodies also have prognostic value. They iden- ratory data allows outcome measures to be employed pro-
tify a subset of patients more likely to suffer erosive disease spectively in the clinic.
and also indicate a greater likelihood of a treatment response Comment: Prospective data indicate that the adjustment of
to methotrexate or tumor necrosis factor inhibitors (Kroot RA therapies every 8–12 weeks with the goal of controlling
et al. 2000). Some data indicate that anti-CCP antibodies pre- specific indices of disease activity leads to improved patient
dict erosive disease even more accurately than do high titers outcomes (Grigor et al. 2004; Verstappen et al. 2007; Goekoop-
of RF. However, most patients with anti-CCP antibodies are Ruiterman et al. 2005; Smolen et al. 2003a). Thus, the assess-
also RF positive, and vice versa. Only a small percentage of ment of disease activity is important in both clinical trials
patients have one of these antibodies but not the other. and clinical practice.
Composite disease activity indices are superior to indi-
Pearl: In patients with early polyarthritis, anti-CCP anti-
vidual variables in the assessment of disease activity. Two
body positivity is an excellent predictor of an RA diagnosis.
major factors pose potential barriers to the routine use of vali-
Comment: A study from the Netherlands demonstrated con- dated disease activity indices in clinical practice. First, many
vincingly that in patients with early undifferentiated arthritis validated scores are so complex that they require a calculator
who are anti-CCP antibody positive, at least 93% have defi- for computation. Second, some composite indices require the
nite diagnoses of RA 3 years later (or earlier). Among the results of laboratory tests (e.g., an ESR or C-reactive protein
patients who were anti-CCP antibody negative, only 25% level) before they can be completed. Both of these issues dra-
had developed RA by 3 years (van Gaalen et al. 2004). matically reduce the utility of such scores.
The Clinical Disease Activity Index (CDAI) is a simple
numerical index that is calculated by summing the number of
tender and swollen joints using the 28-joint count and the patient
1.1.5 Risk Factors and physician global assessments on a 10-cm visual analog
scale (Fig. 1.6). The CDAI has proven to be a valid, reliable,
and sensitive measure to change as reflected by ACR responses,
Pearl: Cigarette smoking is bad if one is otherwise geneti-
the DAS28, and the Simplified Disease Activity Index, all of
cally predisposed to the development of RA.
which rely on laboratory tests (Smolen et al. 2003b).
Comment: This warning is in addition, of course, to the other The CDAI ranges from 0 to 76. The cutpoints for remis-
reasons that cigarette smoking is bad Several studies show a sion and various degrees of activity are as follows:
significant and independent association between cigarette
• Remission ≤ 2.8
smoking and increased susceptibility to RA (Silman et al.
• Low disease activity ≤ 10
1996). This environmental exposure interacts in RA with the
• Moderate disease activity ≤ 22
HLA-DRB1 shared epitope to increase susceptibility to RA
• High disease activity > 22
(Padyukov et al. 2004). Smoking is also associated with the
presence of extraarticular manifestations and serum rheuma- The CDAI provides the most stringent remission criteria
toid factor (Turesson et al. 2003; Saag et al. 1997; Mattey available and correlates with outcomes such as the Health
1 Rheumatoid Arthritis 7

Complete Disease Activity Index (CDAI) Wolfe 2007). Radiographic evidence of joint disease is detected
in a large percentage of seropositive RA patients within 2 years
Swollen joint count of diagnosis, and within several months in some patients (Plant
et al. 1998). Both joint damage and subsequent disability are
+
related directly to the activity of inflammation in many patients
Tender joint count (Drossaers-Bakker et al. 1999). These facts underscore the
importance of treating RA early, aggressively, and effectively.
CDAI = +
Pearl: Remission is the aim of therapy.
Patient global assessment
Comment: Only the complete and sustained clinical control
+ of RA activity is associated with the cessation of joint dam-
age (Smolen et al. 2008a). Even states of low disease activity
Evaluator global assessment
are associated with the progression of joint damage (Smolen
Fig. 1.6 The Complete Disease Activity Index (CDAI). The CDAI is et al. 2008a). Functional impairment is higher in states of
composed of the count of swollen joints (SJC), the number of tender low disease activity than in remission. Attaining remission is
joints (TJC), the patient global assessment (PGA), and the evaluator the ultimate therapeutic goal. The achievement of low dis-
global assessment (EGA)
ease activity constitutes an appropriate intermediate aim.
Pearl: Effective treatment in RA involves prompt changes of
Assessment Questionnaire (HAQ) and radiographic progres- therapy if clinical responses are not observed within several
sion (Aletaha et al. 2005b; Mierau et al. 2007; Smolen et al. months.
2003b, 2009a). Treatment modifications should be consid-
Comment: The long-term clinical outcome can be predicted
ered every 3 months if patients do not reach a CDAI ≤ 10
reliably in the majority of RA patients within 3–6 months of
(Aletaha et al. 2007).
starting therapy (Aletaha et al. 2007). The maximal efficacy of
Myth: Assessing joint counts and composite indices is too RA treatment is observed within the first few months. Patients
time-consuming to be practical in routine clinical care. who have not responded with significant improvement by 6
months are unlikely to do so if the same medications are
Reality: The 28-joint count is valid, reliable, and correlates
continued. These patients should be switched to other treat-
well with the total joint count. More importantly, the 28-joint
ment approaches (Aletaha et al. 2007; Goekoop-Ruiterman
count requires only a minute or two for a trained person to
et al. 2005; Smolen et al. 2003a).
calculate. The clinician’s 28-joint count is combined with the
physician’s disease assessment and the patient global assess- Myth: Evidence-based data from randomized controlled tri-
ment to form the CDAI. als confirm that a variety of conventional and biologic
DMARDs are highly effective for the treatment of an RA
patient who has five tender joints and five swollen joints.

1.1.7 Treatment Reality: Most major clinical trial has been performed to address
this patient population. Most major randomized controlled tri-
als published in the last 15 years have required at least six ten-
The Treatment section is divided into three subsections:
der joints and six swollen joints. All 3 of the phase III
• Treatment strategy golimumab trials required 4 tender and 4 swollen joints, so
• Glucocorticoids, Methotrexate, and conventional DMARDs "all" is no longer formally correct. Other studies have shown
• Biologic agents that the vast majority of patients in rheumatologists’ clinics do
not have disease that is sufficiently active to qualify for clinical
trials. There is, therefore, a gap in our knowledge regarding
1.1.7.1 Treatment Strategy these patients. Rightly or wrongly, the results from the pub-
lished clinical trials are extrapolated to this patient population.
Myth: RA is a benign disease for which therapy can be
delayed safely for a while.
1.1.7.2 Glucocorticoids, Methotrexate,
Reality: A cardinal feature of RA is its propensity to destroy
and Conventional DMARDs
joints. Joint destruction can lead to irreversible physical dis-
ability, additional comorbidities (e.g., complications of joint
surgery), economic losses, and a reduction in life expectancy Myth: Once an RA patient is started on glucocorticoids, he
(Yelin et al. 2002; Kobelt and Jönsson 2008; Michaud and or she will never be able to discontinue the medication.
8 J. R. O’Dell et al.

Reality: Clinical trials in the era of biologic agents have use, the prevailing maxim was to “start low and go slow.”
shown that this oft-heard refrain is a Myth. With use of effec- Most clinicians began methotrexate at 7.5 mg/week and
tive DMARD therapy – conventional, biological, or both – advanced the dose slowly over many months, never exceeding
RA patients can be tapered to low doses of glucocorticoids 15 mg/week for many patients.
and many discontinue them entirely. In an ironic twist, clinical trials of the biologic agents
Glucocorticoids are an effective therapy for RA, but 60 changed the way methotrexate is used in practice. A com-
years after their introduction we still struggle to understand paratively aggressive methotrexate regimen was used in the
the optimal way to use them. As discussed below, some comparison arm of a randomized, double-blind clinical
patients should be maintained on a low dose of prednisone trial that compared etanercept to methotrexate over a
(e.g., 5 mg/day). 6-month period (Bathon 2000). Patients in the methotrex-
ate arm began treatment with 15 mg/week and quickly
Pearl: Glucocorticoids are an important part of early RA
escalated their dose to a mean of approximately 20 mg/
therapy.
week. The results were dramatic: although the etanercept
Comment: Glucocorticoids provide important benefits in the group had a superior “area under the curve” disease activ-
treatment of early RA, whether they are administered in ity analysis at 6 months, patients in the methotrexate group
moderately high doses or low doses (Boers et al. 1997; had a similar ACR20 response rate at 12 months as the
Mottonen et al. 1999; Wassenberg et al. 2005; Svensson et al. etanercept group.
2005). The COBRA study employed an aggressive glucocor- Methotrexate therapy in most patients should begin at
ticoid regimen that began patients with early RA on 60 mg of 10–15 mg/week, and then be escalated to 20 mg/week over
prednisone/day (Boers et al. 1997). Other trials have the next 4 weeks, as tolerated. The dose can be increased to
employed substantially lower starting doses (Mottonen et al. 25 mg/week or even higher in some cases.
1999; Wassenberg et al. 2005; Svensson et al. 2005).
Myth: When administered weekly in low doses (10–25 mg/
Regardless of the starting dose used, the addition of gluco-
week), methotrexate often accumulates to dangerous levels
corticoids provides significant advantages when compared
in physiologic third spaces, such as peritoneal effusions,
with treatment approaches that employ methotrexate alone
pleural effusions, and pericardial effusions.
or combinations of synthetic DMARDs.
The upshot of these data is that the optimal therapeutic Reality: Methotrexate has been reported to accumulate in
approach for RA is a blend of the old and the new. Methotrexate physiologic third spaces, such as pleural and peritoneal fluid
is the cornerstone of initial RA treatment and should be used in during high-dose MTX therapy for cancer (Li and Gwilt
combination with glucocorticoids at the start of therapy. Most 2002). In these cases, the presence of a physiologic third
clinicians also employ low-dose glucocorticoids. If methotrex- space may prolong the half-life of MTX and cause unantici-
ate is insufficiently effective in this scheme, a biologic agent pated toxicity. Such toxicity has not been reported in patients
should be added within 3–6 months, possibly with another glu- with RA who are receiving standard weekly arthritis doses.
cocorticoid boost at the initiation of this new treatment. The issue of methotrexate “third-spacing” probably would
not be a significant concern in RA unless the patient has sig-
Pearl: Glucocorticoids act as DMARDs in RA.
nificant renal insufficiency.
Comment: The Empire rheumatism studies of the 1950s dem-
Pearl: Leflunomide remains in the body for years after
onstrated that glucocortiosteroid therapy significantly slowed
administration.
radiographic progression of disease. Many clinicians did not
believe these data and avoided prescribing glucocorticoids Comment: Because of its extensive enterohepatic
because of their well-known adverse effects. Well-designed re-circulation, leflunomide persists in the body for years after
randomized, controlled clinical trials have subsequently con- the medication has been discontinued. This is an important
firmed the findings from the Empire rheumatism studies fact when toxicities occur or if pregnancy is considered,
(Kirwan 1995). because leflunomide has significant teratogenic potential. In
Adjunct medications such as bisphosphonates and statins these situations, a washout with cholestyramine may be
now attenuate many of the potential drawbacks to glucocor- desired. If the persistence of leflunomide is a potential con-
ticoids. Many clinicians should probably be more liberal in cern, blood levels may be checked.
the use of these medications, which are remarkably cheap
Myth: Antibiotics are not effective in the treatment of RA.
and effective therapy.
Reality: There is little to no evidence that the antibacterial
Myth: When using methotrexate, start low and go slow.
effects of some DMARDs are important in the treatment of
Reality: Methotrexate has been the cornerstone of treatment RA. However, sulfasalazine and minocycline, both of
for RA since the mid-1980s. In the early years of methotrexate which have significant antibacterial effects, also have some
1 Rheumatoid Arthritis 9

efficacy in RA. Whether the antibacterial effects of these Table 1.2 Treatment arms in the BeSt study
drugs are important with regard to their activity against RA • Group 1 (sequential monotherapy) – Sequential monotherapy
is unclear. began with methotrexate. The following agents were added in
succession if the disease remained poorly controlled:
In the case of minocycline, a number of other mechanisms – Sulfasalazine (methotrexate discontinued)
of action may be important (O’Dell 1999). Minocycline has – Leflunomide (sulfasalazine discontinued)
substantial immunomodulatory actions, some of which con- – Methotrexate plus an anticytokine agent (leflunomide
tribute to its well-known ability to induce lupus. Minocycline discontinued)
is unique among antibiotics in terms of its ability to inhibit • Group 2 (step-up) – Step-up therapy began with methotrexate.
metalloproteinases and among DMARDs because of its abil- The following agents were added together (one at a time) if the
disease remained poorly controlled:
ity to inhibit peptidyl arginine deiminase. – Sulfasalazine
– Hydroxychloroquine
Myth: Hydroxychloroquine is not an effective DMARD.
– Prednisolone
Reality: There is a paucity of direct evidence that hydroxy- – An anticytokine agent (with discontinuation of sulfasalazine
chloroquine slows radiographic progression in RA. However, and hydroxychloroquine)
multiple studies have shown that if hydroxychloroquine is • Group 3 (initial combination DMARD plus prednisolone) – Initial
combination treatment consisted of the combination of metho-
started early in the course of disease, patients experience bet- trexate, sulfasalazine, and prednisolone. Persistent disease led to
ter outcomes, including less radiographic progression. The the addition of:
combination of methotrexate and hydroxychloroquine is the – Cyclosporine (sulfasalazine discontinued)
most common tandem DMARD regimen used for the treat- – Methotrexate (prednisolone discontinued)
ment of RA. In addition, some data suggest that the use of • Group 4 (combination methotrexate and infliximab) – Initial
hydroxychloroquine markedly reduces the incidence of dia- combination treatment consisted of methotrexate and infliximab.
Persistent disease led to the addition (one at a time) of:
betes among patients with RA (Wasko et al. 2007). – Sulfasalazine
The potency of the MTX-HCQ combination may be – Leflunomide
explained at least in part by a patient’s increased exposure to – Methotrexate, cyclosporine, and prednisolone
MTX (e.g., increased area under the curve) when MTX is
administered together with HCQ (Carmichael et al. 2002).
Myth: Combination therapy with conventional (i.e., non- 1.1.7.3 Biologic Agents
biologic) DMARDs is uniformly effective when compared
with MTX alone. Myth: Nearly all patients with RA respond to tumor necrosis
factor inhibitors.
Reality: Combination therapies of traditional DMARDs
were advocated strongly for the treatment of RA until well Reality: The BeST study randomly assigned 508 DMARD-
into the 1990s. One major trial that preceded the era of bio- naive patients with RA of less than 2 years duration to one of
logic treatments indicated that the combination of metho- four different treatment strategies. The four treatment groups
trexate, sulfasalazine, and hydroxychloroquine was superior included in this study are shown in Table 1.2. Approximately
to methotrexate alone (O’Dell 1996). However, most clinical 22% of patients randomized to the TNF inhibitor arm, a
trials comparing combination therapy to monotherapy have combination of infliximab and methotrexate, failed treatment
not shown superiority of the combination therapy arms (sum- (Goekoop-Ruiterman et al. 2005).
marized in Smolen et al. 2005).
Myth: When a TNF inhibitor fails, other TNF inhibitors are
Pearl: Gold shots are dramatically efficacious for the treat- unlikely to work.
ment of RA in one of every five patients.
Reality: Several open-label studies or analyses from regis-
Comment: No one knows for certain whether or not this state- tries have suggested that a second TNF inhibitor can be effi-
ment is a Pearl or a Myth, but it is echoed widely by practi- cacious following the failure of a first TNF inhibitor
tioners who treated RA in the heyday of chrysotherapy. (Gomez-Reino and Carmona 2006; van Vollenhoven 2004;
Because of unequivocal advances in RA treatment since the Hyrich et al. 2007). This appears to be true even if the first
1980s, the truth or falseness of this statement is likely to and second TNF inhibitors are pharmacologically similar;
remain uncertain. However, when injections of gold sodium for example, the case of adalimumab following infliximab
thiomalate were compared with weekly MTX therapy in a (both medications are monoclonal antibodies).
double-blind, randomized trial, gold produced twice as many These data from uncontrolled investigations have now
clinical remissions as the latter (24.1% vs. 11.5%) (Rau et al. been expanded in a double-blind, controlled clinical trial of
1997). Gold was also associated with more side effects and golimumab (Smolen et al. 2008b). These data reveal that a
withdrawals due to toxicity. third TNF inhibitor can be effective even after two have
10 J. R. O’Dell et al.

failed. However, if all three other TNF inhibitors (etanercept, RA. Gastrointestinal cancers, however, appear to be decreased
infliximab, and adalimumab) have failed before golimumab, among patients with RA, probably by a factor of 3 or 4.
then golimumab is unlikely to be effective. The decrease in colon cancers among patients with RA
led investigators to hypothesize that NSAIDs slow polyp
Pearl: TNF inhibitors work better in combination with meth-
formation. This led, in turn, to randomized trials of COX-2
otrexate than alone.
inhibitors designed to prove this hypothesis – and to the
Comment: All clinic trials that have addressed this question unexpected finding that certain COX-2 inhibitors are associ-
have found that the combination of methotrexate plus inflix- ated with an increased risk of myocardial infarction.
imab, etanercept, or adalimumab is more effective than meth-
otrexate alone or any of the TNF inhibitors by themselves. Pearl: Severe COPD and exposure to cyclophosphamide
This is true for both clinical response (ACR20 or DAS) and may be risk factors for malignancy in patients treated with
radiographic outcomes. In addition, observational studies TNF inhibition.
have strongly suggested that protection against cardiovascu- Comment: Most long-term observational data on TNF inhi-
lar morbidity and mortality by the TNF inhibitors is depen- bition are reassuring with regard to the risk of malignancy.
dent on the concomitant use of methotrexate. The overall incidence of cancer does not appear to be ele-
The synergistic effects of methotrexate extend beyond vated dramatically. However, it may be wise to avoid TNF
TNF inhibitors. Abatacept and rituximab also work better inhibition in some patient subsets.
when prescribed in combination with methotrexate. A randomized clinical trial in Wegener’s granulomatosis
Pearl: Neutralizing antibodies (human antichimeric antibod- found an increased risk of solid tumors in patients who were
ies [HACA]) may decrease the efficacy of infliximab over exposed to etanercept (Stone et al. 2006). All patients who
time. developed solid malignancies also had histories of exposure
to cyclophosphamide. These data suggest that it is wise to
Comment: Secondary failure, i.e., the loss of a response after avoid TNF inhibitors in patients who are on cyclophosph-
an initial improvement, may be due in part to the develop- amide at the same time or who have significant histories of
ment of neutralizing antibodies. Strategies for managing this cyclophosphamide use.
potential complication of infliximab include concomitant use A trial of infliximab therapy for severe COPD suggested
of an immunosuppressive agent such as MTX and avoidance an increase in early lung cancers in the patients treated with
of intermittent therapy with infliximab. Prolonged intervals infliximab (Rennard et al. 2007).
between stopping and re-starting infliximab are associated
with a greater risk of HACA development. Pearl: All TNF inhibitors are not alike when it comes to
toxicity.
Myth: B cells are depleted for life after therapy with rituximab.
Reality: B cells eventually return. This return often begins as Comment: Infliximab is associated much more often with the
early as 4–6 months after completion of the rituximab infu- reactivation of tuberculosis than is etanercept (Wallis et al.
sions but is not complete for years. B cell numbers do cor- 2004). Data from Europe suggest that the risk of tuberculosis
relate with response to rituximab in groups of patients, with adalimumab is similar to that of infliximab and higher
implying that more investigation is required to understand than etanercept. Infliximab is also more likely to be associ-
why rituximab works (when it works) for autoimmune con- ated with histoplasmosis, coccidioidomycosis, and listeriosis
ditions (Breedveld et al. 2007). (Wallis et al. 2004). Conversely, etanercept has been linked
more commonly to demyelinating diseases.

Pearl: The effects of TNF inhibition in patients with viral

hepatitis diverge widely depending on the specific form of
1.1.8 Co-Morbidities and Adverse Effects hepatitis.

Comment: Well-documented, life-threatening, or fatal flares

Myth/Pearl: RA is associated with an increased incidence of
of hepatitis B virus infection have occurred in patients treated
cancer.
with TNF inhibitors or rituximab (Calabrese et al. 2006).
Comment: This statement may be regarded as either a Myth or a Thus, both medications are relatively contraindicated in
Pearl, depending on the type of malignancy to which one refers. patients with a history of hepatitis B, but TNF inhibitors have
The incidence of lymphoma is increased in RA; most studies been used in patients with hepatitis B virus infection kept on
suggest at least a twofold increase compared with the general antiviral agents. In contrast, hepatitis C virus infection does
population. Lung cancers are also overrepresented in patients not appear to be a contraindication for therapy with TNF
with RA, perhaps because smoking is a potent risk factor for inhibition. Limited data suggest a possible beneficial effect
1 Rheumatoid Arthritis 11

of TNF inhibition on the liver disease during treatment for Myth: Cardiovascular risk factors confer the same risk for
hepatitis C. cardiovascular outcomes in patients with RA as they do in
the general population.
Myth: Biologic agents can be used safely in combination to
treat RA. Reality: One retrospective study has shown that male gender,
smoking, and personal cardiac history had weaker associa-
Reality: All available data suggest that when two biologic
tions with cardiovascular events among patients with RA
agents are used in tandem, toxicity (particularly from infec-
than they do among individuals who do not have RA
tions) rises significantly. Studies of biologic combinations of
(Gonzalez et al. 2006). In contrast, the risks imparted by
etanercept and anakinra and etanercept and abatacept have
other risk factors – hypertension, dyslipidemia, body mass
shown only marginal, if any, improvements in efficacy, but
index, diabetes mellitus, and family history – were no differ-
significantly increased rates of infections in the combination
ent between the two groups.
group when compared with either biologic alone (Furst in
press). Clinicians should avoid using two of the currently
available biologic agents at the same time.
1.2 Immunizations in RA
Myth: AntiTNF agents are safe in pregnancy.
Reality: Only anecdotal data about the impact of TNF inhibi- Myth: Methotrexate monotherapy has been generally associ-
tion on pregnancy are available. These seemingly reassuring ated with a profoundly reduced response to the influenza
outcomes have led some clinicians to believe “so far, so vaccine.
good” and allow pregnant women to use TNF inhibitors. It is
too early to judge whether TNF inhibitors are safe in preg- Reality: Therapy with methotrexate has been associated with
nancy. Two reports raise at least some concern. One abstract little (if any) reduction in the response to the influenza vaccine.
has reported a possible increase in a distinctive pattern of Myth: Influenza vaccination is NOT recommended for
birth defects called VACTERL (or Vater syndrome) where V patients receiving a TNF inhibitor.
= verterbral anomalies, A = anal atresia, C = cardivascular
anomalies, T = tracheoesophageal fistula, E = esoghageal Reality: Although the serological responses are less robust in
atresia, R = renal and/or radial abnormalities, and L = limb patients with RA receiving a TNF inhibitor when compared
anomalies (Carter et al., 2009). A second report suggested an with that of healthy individuals, their use does not raise any
increased risk of miscarriage among women exposed to TNF safety concerns and the titers are sufficiently high to warrant
agents early in pregnancy. influenza vaccination in all RA patients. However, a large
A prudent approach is to discontinue TNF inhibitors when proportion of patients with RA may not respond to pneumo-
conception is confirmed or, even better, before conception coccal vaccination while they are taking MTX and a TNF
(to the extent that pregnancy can be planned). Patients and inhibitor. The optimal approach to vaccination in these
clinicians alike hope for the temporary remission of RA that patients, therefore, is to administer the pneumococcal vacci-
frequently accompanies pregnancy once TNF inhibitor use nation before starting treatment with these agents.
has been suspended (see Chap 17).
Pearl: RA is an independent risk factor for cardiovascular
References
disease.
Comment: Several studies show that cardiovascular disease Aletaha D, Nell VPK, Stamm T, et al Acute phase reactants add little to
is the leading cause of death in patients with RA. In a recent composite disease activity indices for rheumatoid arthritis:
meta-analysis, the risk of cardiovascular death was 50% Validation of a clinical activity score. Arthritis Res 2005a;
7:R796–R806
higher in patients with RA when compared with the general Aletaha D, Ward MM, Machold KP, et al Remission and active disease
population (Aviña-Zubieta et al. 2008). in rheumatoid arthritis: Defining criteria for disease activity states.
Arthritis Rheum 2005b; 52:2625–2636
Pearl or Myth: DMARD and/or biologic therapy reduce the Aletaha D, Breedveld FC, Smolen JS. The need for new classification
occurrence of adverse cardiovascular events in RA. criteria for rheumatoid arthritis. Arthritis Rheum 2005c; 52:
3333–3336
Comment: Retrospective case-control studies suggest that Aletaha D, Funovits J, Keystone EC, Smolen JS. Disease activity early
MTX and TNF inhibitors decrease cardiovascular morbidity in the course of treatment predicts response to therapy after one year
and mortality in RA, but these positive outcomes have not in rheumatoid arthritis patients. Arthritis Rheum 2007; 56(10):
3226–3235
been substantiated by evidence from prospective studies or Arnett FC, Edworthy SM, Block DA, et al The American Rheumatism
randomized, controlled clinical trials. A verdict on this will Association 1987 revised criteria for the classification of rheuma-
require additional prospective work. toid arthritis. Arthritis Rheum 1988; 31:315–324
12 J. R. O’Dell et al.

Aviña-Zubieta JA, Choi HK, Sadatsafavi M, et al Risk of cardiovascular Gonzalez A, Kremers HM, Crowson CS, et al Do cardiovascular risk
mortality in patients with rheumatoid arthritis: A meta-analysis of factors confer the same risk for cardiovascular outcomes in rheuma-
observational studies. Arthritis Rheum 2008; 59:1690–1697 toid arthritis as in non-rheumatoid arthritis patients? Ann Rheum
Avnon LS, Abu-Shakra M, Flusser D, et al Pleural effusion associated Dis 2006; 67:64–69
with RA Rheum International 2007; 27:919–925 Grigor C, Capell H, Stirling A, et al Effect of a treatment strategy of
Axford JS. Rheumatic manifestations of haemochromatosis. Baillieres tight control for rheumatoid arthritis (the TICORA study): A single-
Clin Rheumatol 1991; 5:351–365 blind randomised controlled trial. Lancet 2004; 364:263–269
Bathon JM, Martin RW, Fleischmann RM, et al A comparison of etan- Harrison BJ, Symmons DP, Barret EM, Silman AS. The performance of
ercept and methotrexate in early rheumatoid arthritis. N Engl J Med the 1987 ARA classification criteria for rheumatoid arthritis in a
2000;343:1586–93 population based cohort of patients with early inflammatory polyar-
Bellhorn LR, Hess EV. Erosive osteoarthritis. Semin Arthritis Rheum thritis. J Rheumatol 1998; 25(2324):2330.
1993; 22:298–306 Hyrich KL, Lunt M, Watson KD, et al Outcomes after switching from
Boers M, Verhoeven AC, Markusse HM, et al Randomised comparison one anti-tumor necrosis factor alpha agent to a second anti-tumor
of combined step-down prednisolone, methotrexate and sulphasala- necrosis factor alpha agent in patients with rheumatoid arthritis:
zine with sulphasalazine alone in early rheumatoid arthritis: A ran- Results from a large UK national cohort study. Arthritis Rheum
domised trial. Lancet 1997; 350:309–318 2007; 56(1):13–20
Breedveld F, Agarwal S, Yin M, et al Rituximab pharmacokinetics in Kalajian A, Callen J. Multicentric reticulohistiocytosis successfully
patients with rheumatoid arthritis: B-cell levels do not correlate with treated with infliximab: An illustrative case and evaluation of
clinical response. J Clin Pharmacol 2007; 47:1119–1128 cytokine expression supporting anti-tumor necrosis factor therapy.
Calabrese LH, Zein NN, Vassilopoulos D. Hepatitis B virus (HBV) reacti- Arch Dermatol 2008; 144:1360–1366
vation with immunosuppressive therapy in rheumatic diseases: assess- Kirwan JR, The Arthritis and Rheumatism Council Low-Dose Gluco-
ment and preventive strategies. Ann Rheum Dis 2006; 65: 983–989 corticoid Study Group. The effect of glucocorticoids on joint destruc-
Carmichael SJ, Beal J, Day RO, Tett SE. Combination therapy with tion in rheumatoid arthritis. N Engl J Med 1995; 333: 142–146
methotrexate and hydroxychloroquine for rheumatoid arthritis Klareskog L, Stolt P, Lundberg K, et al A new model for an etiology of
increases exposure to methotrexate. J Rheumatol 2002; 29: rheumatoid arthritis: Smoking may trigger HLA-DR (shared
2031–2033 epitope)-restricted immune reactions to autoantigens modified by
Carter JD, Ladhani A, Ricca LR, et al A safety assessment of tumor citrullination. Arthritis Rheum 2006; 54:38–46
necrosis factor antagonists during pregnancy: A review of the Food Kobelt G, Jönsson B. The burden of rheumatoid arthritis and access to
and Drug Administration database. J Rheumatol 2009; 36:3 treatment: Outcome and cost-utility of treatments. Eur J Health
Chan MT. Owen P. Dunphy J, et al Associations of erosive arthritis with Econ 2008; 8 (Suppl 2):S95–S106
anti-cyclic citrullinated peptide antibodies and MHC Class II alleles Kroot EJ, de Jong BA, van Leeuwen MA, Swinkels H, et al The prog-
in systemic lupus erythematosus. J Rheumatol 2008; 35 (1):77–83 nostic value of anti-cyclic citrullinated peptide antibody in patients
Ching DW, Petrie JP, Klemp P, et al Injection therapy of superficial with recent-onset rheumatoid arthritis. Arthritis Rheum 2000;
rheumatoid nodules. Br J Rheumatol 1992; 31:775 43(8):1831–1835
Choi HK, Hernan MA, Seeger JD, et al Methotrexate and mortality in Lee H, Irigoyen P, Kern M, et al Interaction between smoking, the
patients with rheumatoid arthritis: A prospective study. Lancet shared eiptope, and anti-cyclic citrullinated peptide: A mixed pic-
2002; 359(9313):1173–1177 ture in three large North American rheumatoid arthritis cohorts.
Cohen SB, Dore RK, Lane NE, et al Denosumab treatment effects on Arthritis Rheum 2007; 56:1745–1753
structural damage, bone mineral density, and bone turnover in rheu- Li J, Gwilt P. The effect of malignant effusions on methotrexate dispo-
matoid arthritis: A twelve-month, multicenter, randomized, double- sition. Cancer Chemother Pharmacol 2002; 50:373–382
blind, placebo-controlled, phase II clinical trial. Arthritis Rheum Lunardi C, Tinazzi E, Bason C, et al Human parvovirus B19 infection
2008; 58(5):1299–1309 and autoimmunity. Autoimmun Rev 2008; 8:116–120
De Souza AW, Fernendes V, Ferrari AJ. Female gout: Clinical and labo- Machold KP, Stamm TA, Nell VP, et al Very recent onset rheumatoid
ratory features. J Rheumatol 2005; 32:2186–2188 arthritis: Clinical and serological patient characteristics associated
Doherty M, Dieppe PA. Crystal deposition disease in the elderly. Clin with radiographic progression over the first years of disease.
Rheum Dis 1986; 12:97–116 Rheumatology (Oxford) 2006 [Epub 2006 Aug 9 ahead of print]
Drossaers-Bakker KW, de Buck M, van Zeben D, et al Long-term Masi, AT. Articular patterns in the early course of rheumatoid arthritis.
course and outcome of functional capacity in rheumatoid arthritis: Am J Med 1983; 75:16
The effect of disease activity and radiologic damage over time. Mattey DL, Dawes PT, Clarke S, et al Relationship among the HLA-
Arthritis Rheum 1999; 42:1854–1860 DRB1 shared epitope, smoking, and rheumatoid factor production
Fernandez A, Quintana G, Rondon F, et al Lupus arthropathy: A case in rheumatoid arthritis. Arthritis Care Res 2002; 47:403–407
series of patients with rhupus. Clin Rheumatol 2006; 25:164–167 Michaud K, Wolfe F. Comorbidities in rheumatoid arthritis. Best Pract
Finckh A, Dehler S, Costenbader KH, et al Cigarette smoking and Res Clin Rheumatol 2007; 21(5):885–906
radiographic progression in rheumatoid arthritis. Ann Rheum Dis Mierau M, Schoels M, Gonda G, et al Assessing remission in clinical
2007; 66:1066–1071 practice. Rheumatology 2007; 46:975–979 [Epub 2007 Mar 6]
Furst DE. The risk of infections with biologic therapies for rheumatoid Mikuls TR, Huges LB, Westfall AO, et al Cigarette smoking, disease
arthritis. Semin Arthritis Rheum 2009 (in press) severity, and autoantibody expression in African Americans with
Genovese MC, Becker JC, Schiff M, et al Abatacept for rheumatoid recent-onset rheumatoid arthritis. Ann Rheum Dis 2008; 67:
arthritis refractory to tumor necrosis factor alpha inhibition. N Engl 1529–1534
J Med 2005; 353(11):1114–1123 Morozzi G, Bellisai F, Fioravanti A, Galeazzi M. Absence of anti-cyclic
Goekoop-Ruiterman YP, De Vries-Bouwstra JK, Allaart CF, et al Clinical citrullinated peptide antibodies in erosive osteoarthritis: Further
and radiographic outcomes of four different treatment strategies in serological evidence of the disease as a subset of osteoarthritis. Ann
patients with early rheumatoid arthritis (the BeSt study): A random- Rheum Dis 2005; 64:1095–1096
ized, controlled trial. Arthritis Rheum 2005; 52: 3381–3390 Mottonen T, Hannonen P, Leirisalo-Repo M, et al Comparison of com-
Gomez-Reino JJ, Carmona L. Switching TNF antagonists in patients bination therapy with single-drug therapy in early rheumatoid
with chronic arthritis: An observational study of 488 patients over a arthritis: A randomised trial. FIN-RACo trial group. Lancet 1999;
four-year period. Arthritis Res Ther 2006; 8(1):R29 353(9164):1568–1573
1 Rheumatoid Arthritis 13

Nell V, Machold KP, Stamm TA, et al Autoantibody profiling as early Smolen J, Kay J, Doyle MK, et al Golimumab, a new human anti-TNF
diagnostic and prognostic tool for rheumatoid arthritis. Ann Rheum alpha monoclonal antibody, subcutaneously administered every 4
Dis 2005; 64:1731–1736 [Epub 2005 May 5] weeks in patients with active rheumatoid arthritis who were previ-
Nishimura K, Sugiyama D, Kogata Y, et al Meta-analysis: Diagnostic ously treated with anti-TNF-alpha agent(s): Results of the random-
accuracy of anti-cyclic citrullinated peptide antibody and rheumatoid ized, double-blind, placebo-controlled trial. Ann Rheum Dis 2008b;
factor for rheumatoid arthritis. Ann Intern Med 2007; 146:797–808 67(Suppl II):50
O’Dell J. Treatment of rheumatoid arthritis with methotrexate alone, Stone JH, Holbrook JT, Marriott MA, et al, for the Wegener’s
sulfasalazine and hydroxychloroquine, or a combination of all three Granulomatosis Etanercept Trial Research Group. Solid malignan-
medications. N Engl J Med 1996; 334:1287–1291 cies among patients in the Wegener’s Granulomatosis Etanercept
O’Dell R. Is there a role for antibiotics in the treatment of patients with Trial. Arthritis Rheum 2006; 54:1608–1618
rheumatoid arthritis? Drugs 1999; 57:279–282 Svensson B, Boonen A, Albertsson K, et al Low-dose prednisolone in
Ostendorf B, Scherer A, Specker C, et al Jaccoud’s arthropathy in sys- addition to the initial disease-modifying antirheumatic drug in
temic lupus erythematosus: Differentiation of deforming and ero- patients with early active rheumatoid arthritis reduces joint destruc-
sive patterns by magnetic resonance imaging. Arthritis Rheum tion and increases the remission rate: A two-year randomized trial.
2003; 48:157–165 Arthritis Rheum 2005; 52:3360–3370
Padyukov L, Silva C, Stolt P, et al A gene-environment interaction Thery C, Lekieffre J, Gosselin B, et al Atrio-ventricular block in rheu-
between smoking and shared epitope genes in HLA-DR provides a matoid polyarthritis. Arch Mal Coeur Vaiss 1974; 67:1181–1191
high risk of seropositive rheumatoid arthritis. Arthritis Rheum 2004; Turesson C, O’Fallon WM, Crowson CS, et al Extra-articular disease
50:3085–3092 manifestation in rheumatoid arthritis: Incidence, trends, and risk
Plant MJ, Jones PW, Saklatvala J, et al Patterns of radiological progres- factors over 46 years. Ann Rheum Dis 2003; 62:722–727
sion in rheumatoid arthritis: Results of an 8 year prospective study. van der Heijde DM, van Riel PL, van Leeuwen MA, et al Prognostic
J Rheumatol 1998; 25:417–426 factors for radiographic damage and physical disability in early
Rantapaa-Dahlqvist S. Diagnostic and prognostic significance of rheumatoid arthritis. A prospective follow-up study of 147 patients.
autoantibodies in early rheumatoid arthritis. Scand J Rheumatol Br J Rheumatol 1992; 31(8):519–525
2005;34:83 van der Helm-van Mil AH, le Cessie S, van Dongen H, et al A predic-
Rau R, Herborn G, Menninger H, Blechschmidt J. Comparison of intra- tion rule for disease outcome in patients with recent-onset undif-
muscular methotrexate and gold sodium thiomalate in the treatment ferentiated arthritis: How to guide individual treatment decisions.
of early erosive rheumatoid arthritis: 12 month data of a double- Arthritis Rheum 2007; 56:433
blind parallel study of 174 patients. Br J Rheumatol 1997; van Gaalen FA, Linn-Rasker SP, van Venrooij WJ, de Jong BA,
36:345–352 Breedveld FC, Verweij CL, Toes RE, Huizinga TW. Autoantibodies
Rennard SI, Fogarty C, Kelsen S, et al The safety and efficacy of inflix- to cyclic citrullinated peptides predict progression to rheumatoid
imab in moderate to severe chronic obstructive pulmonary disease. arthritis in patients with undifferentiated arthritis: A prospective
Am J Respir Crit Care Med 2007; 175:926–934 cohort study. Arthritis Rheum 2004; 50:709–715
Saag KG, Cerhan JR, Kolluri S, et al Cigarette smoking and rheumatoid van Vollenhoven R. Switching between biological agents. Clin Exp
arthritis severity. Ann Rheum Dis 1997; 56:463–469 Rheumatol 2004; 22(Suppl 5):S115–S121
Schellekens GA, Visser H, de Jong BA, et al The diagnostic properties van Vugt RM, Derksen RH, Kater L, Bijlsma JW. Deforming arthropa-
of rheumatoid arthritis antibodies recognizing a cyclic citrullinated thy or lupus and rhupus hands in systemic lupus erythematosus.
peptide. Arthritis Rheum 2000; 43:155–163. Ann Rheum Dis 1998; 57: 540–544
Silman AJ, Newman J, MacGregor AJ. Cigarette smoking increases the Verstappen SM, Jacobs JW, Heurkens AH, et al Intensive treatment
risk of rheumatoid arthritis. Arthritis Rheum 1996; 39:732–735 with methotrexate in early rheumatoid arthritis: Aiming for remis-
Sinclair D, Hull RG. Why do general practitioners request rheumatoid sion. Computer Assisted Management in Early Rheumatoid Arthritis
factor? A study of symptoms, requesting patterns and patient out- (CAMERA, an open-label strategy trial). Ann Rheum Dis 2007;
come. Ann Clin Biochem 2003; 40(Pt 2):131–137 66(11):1443–1449
Smolen JS, Sokka T, Pincus T, Breedveld FC. A proposed treatment Wallis RS, Broder MS, Wong JY, Hanson ME, Beenhouwer DO.
algorithm for rheumatoid arthritis: Aggressive therapy, methotrex- Granulomatous infectious diseases associated with tumor necrosis
ate, and quantitative measures. Clin Exp Rheumatol 2003a; 21(5 factor antagonists. Clin Infect Dis 2004; 38(9):1261–1265
Suppl 31):S209–S210 Wasko MC, Hubert HB, Lingala VB, Elliott JR, Luggen ME, Fries JF,
Smolen JS, Breedveld FC, Schiff MH, et al A simplified disease activity Ward NM. Hydroxychloroquine and risk of diabetes in patients with
index for rheumatoid arthritis for use in clinical practice. Rheuma- rheumatoid arthritis. JAMA 2007; 298:187–193
tology 2003b; 42:244–257 Wassenberg S, Rau R, Steinfeld P, Zeidler H. Very low-dose predniso-
Smolen JS, Aletaha D, Keystone E. Superior efficacy of combination lone in early rheumatoid arthritis retards radiographic progression
therapy for rheumatoid arthritis. Fact or Fiction? Arthritis Rheum over two years: A multicenter, double-blind, placebo-controlled
2005; 52:2975–2983 trial. Arthritis Rheum 2005; 52:3371–3380
Smolen JS, Han C, Van der Heijde DM, et al Radiographic changes in Yelin E, Trupin L, Wong B, Rush S. The impact of functional status and
rheumatoid arthritis patients attaining different disease activity change in functional status on mortality over 18 years among persons
states with methotrexate monotherapy and infliximab plus metho- with rheumatoid arthritis. J Rheumatol 2002; 29(9):1851–1857
trexate: The impacts of remission and TNF blockade. Ann Rheum Young NS, Brown KE. Parvovirus B19. N Engl J Med 2004; 350:
Dis 2009; (6):823–7 586–597
 Rheumatoid Vasculitis
2
John H. Stone and Eric L. Matteson

2.1 Overview of Rheumatoid Vasculitis a

› Rheumatoid arthritis (RA) is associated with vasculitis

in approximately 1% of patients.
› Rheumatoid vasculitis (RV) refers specifically to a
destructive inflammatory process with protean clinical
features that center on the blood vessel wall. The dis-
ease strongly resembles polyarteritis nodosa in its pre-
dilection for particular organs: the skin, peripheral
nerves, gastrointestinal tract, and heart.
› RV can affect a wide range of blood vessel types, from
medium-sized muscular arteries to smaller arterioles b
even post-capillary venules (the size of blood vessels
involved in many forms of small-vessel vasculitis).
› RV usually occurs in patients with severe, longstand-
ing, nodular, destructive RA.
› Palpable purpura, cutaneous ulcers (particularly in the
malleolar region), digital infarctions, and peripheral
sensory neuropathy are common manifestations.
› Tissue biopsy establishes the diagnosis of RV. Deep
skin biopsies from the border of ulcers may capture
medium-sized vessels that are involved. Nerve con-
duction studies help identify involved nerves for
biopsy. Muscle biopsies should be performed simulta-
neously with nerve biopsies to increase the diagnostic
yield of the procedure.
Figs. 2.1 (a, b) RV tends to occur in patients with burnt-out joint dis-
ease (Figures courtesy of Dr. John Stone)

Pearl: RV tends to occur in patients with “burnt-out” RA.

unfortunately means that the patient now needs intensive
Comment: This very good rule of thumb, seldom violated, therapy – glucocorticoids and possibly cyclophosphamide –
remains one of the many mysteries surrounding rheumatoid more than ever.
disease. RV singles out patients who have had severe, destruc- In addition to RA duration, other risk factors for RV
tive RA whose joint disease is no longer active (Figs. 2.1a include the presence of other extraarticular manifestations of
and b). Thus, the classic patient with this complication is one RA, especially rheumatoid nodulosis, scleritis, amyloidosis,
with a longstanding history of nodular, seropositive, destruc- and the presence of rheumatoid factor and antiCCP antibody
tive RA who, after years of ineffectual disease-modifying (Turesson et al. 2003; Turesson and Matteson 2008;Van
antirheumatic drugs (DMARDs) and too much prednisone, Gaalen et al. 2004). There appears to be a genetic predisposi-
has minimal synovitis. The life-threatening nature of RV tion toward developing RV, as HLA-DRB1 shared epitope

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 15

DOI:10.1007/978-1-84800-934-9_2, © Springer Science + Business Media B.V. 2009
16 J. H. Stone and E. L. Matteson

Fig. 2.2 Vasculitis of the left great toe in a patient with RA of 5 months
duration treated with methotrexate and hydroxychloroquine. Minimal
synovitis is present. Biopsy of the skin was consistent with a leukocyto-
clastic vasculitis. The patient developed a foot drop from mononeuritis,
which responded to oral prednisone, at an initial dose of 1 mg/kg/daily
(Figure courtesy of Dr. Eric Matteson)

phenotypes are strongly associated with extraarticular dis-

ease manifestations including RV (Turesson et al. 2005).
Although the occurrence of RV in patients with longstand-
ing disease is the rule, cases developing early in the course of b
RA have been reported (Fig. 2.2) (Voskuyl et al. 1996).
Myth: Nailfold infarcts in RA are a harbinger of serious vas-
culitis and should trigger an intensification of treatment.
Reality: Bywaters’ lesions are cutaneous infarctions that
occur around the nailbeds in patients with RA (Bywaters
1949, 1957) (Fig. 2.3a and b). These lesions do not imply the
need for intensive immunosuppression. Although such
lesions do tend to occur in patients with seropositive RA,
they do not correlate with systemic vasculitis in other organs
(Watts et al. 1995; Price-Forbes et al. 2002).
Pearl: The development of pericarditis in a patient with RA
should not be considered an “intercurrent event”.
Comment: Pericarditis is the cardiac manifestation most
likely to present early in the setting of RV. Its development
should trigger an evaluation for other hallmarks of vasculitis:
subtle neurological symptoms of peripheral nerve disease,
skin lesions, higher than baseline elevations of acute phase
reactants, or hypocomplementemia. One series of 50 patients
documented cardiac manifestations in approximately one- Figs. 2.3 Bywaters’ Lesions. (a) Seventy-four-year-old woman with
third of patients (Scott et al. 1981). These complications osteoarthritis and RA for 15 years treated with methotrexate and
prednisone (5 mg daily). The patient developed recurrent periungual
included pericarditis, arrhythmia, and aortic incompetence, and terminal digit infarctions over an 8-month period. These healed
and myocardial infarction. without intervention. No active synovitis was present at the time the
infarctions developed (Figures courtesy of Dr. Eric Matteson). (b)
Myth: RV is high on the differential diagnosis when a patient Bywaters’ lesions in another patient with RA (Figure courtesy of
with RA has a myocardial infarction. Dr. John Stone)
2 Rheumatoid Vasculitis 17

Reality: Cases of coronary vasculitis are well-documented in a

the medical literature (Sokoloff 1953; Cruikschank 1954;
Johnson et al. 1969). However, myocardial infarction as a
direct result of coronary arteritis in RV is unusual (Van
Albada-Kuipers et al. 1986). Clinically manifest coronary
vasculitis is likely to occur only in RA patients with clearcut
evidence of vasculitis in other organ systems.
Despite the rarity of true coronary vasculitis in RV, one of
the most important determinants of the increased mortality
in RA patients is an elevated risk of cardiovascular disease.
This appears to represent a significant acceleration of com-
mon atherosclerosis (Wallberg-Jonsson et al. 2002; Park
et al. 2002; Kumeda et al. 2002; Hurlimann et al. 2002).
Myth: Leg ulcers in RA are almost certainly due to vasculitis.
Reality: Other clinical entities often cause leg ulcers in RV. b
Pyoderma gangrenosum occurs occasionally in RA, but is
significantly less common than ulcers due to RV. Indeed,
because of the absence of a diagnostic histopathology of
pyoderma gangrenosum and the frequency with which biop-
sies in RV are non-diagnostic, many cases of “pyoderma
gangrenosum” associated with RA in the past were probably
in fact actually a manifestation of RV.
Another factor that may contribute to the development of
leg ulcers in RA is the combination of venous stasis disease and
chronic glucocorticoid use, the latter of which leads to increased
skin fragility and a tendency to ulcerations with minor trauma.
Thus, not all skin ulcerations in RA represent RV.
Leg ulcers in patients with RA often occur due to minor
trauma. There may be an underlying vasculitis that promotes
the lesion. However, ulcer expansion and chronicity are
influenced much more by other factors, including concomi-
tant immunodeficiency, arterial insufficiency, trauma, and
dependent edema (Turesson 2004; Puechal et al. 2008).
Chronic glucocorticoid use and smoking also promote devel-
opment of these chronic ulcers. Approximately 30% of
patients with Felty’s syndrome (neutropenia, splenomegaly,
and RA) develop skin ulcers.
The development of leg ulcers should trigger a thorough
evaluation for the possibility of RV including, if appropriate,
a skin biopsy. Patients with leg ulcers must be monitored
closely, particularly those who have high titers of rheumatoid
factor, low complement, and cryoglobulins, as more ominous
manifestations of RV including digital infarcts and sensory
Figs. 2.4 Mononeuritis multiplex with bilateral foot drop on the right
motor neuropathy can occur in these patients (Fig. 2.4).
and extensive foot ulcerations in a 69 year-old man with a 20-year his-
Appropriate treatment of leg ulcers in patients with RA tory of RA. There had been multiple previous episodes of vasculitis;
includes aggressive wound management. This may include extensive tissue atrophic changes especially of the over the soft tissues
skin grafting. Aggressive immunosuppressive therapy and of the calf are evident (a) The patient also suffered upper extremity
ulcerations (b) Note, in these figures, the rheumatoid deformities of the
aggressive antiTNF therapy, including the use of biologics, is
feet, the olecranon nodules, left thumb deformity and left-hand muscle
usually not warranted (Turesson 2004). High-dose glucocor- atrophy (Figures courtesy of Kenneth Calamia, MD, Mayo Clinic
ticoids are not warranted; the chronic use of glucocorticoids Jacksonville)
18 J. H. Stone and E. L. Matteson

and requires aggressive management (Turesson 2004;

Puechal et al. 2008).
Effective control of RA is important in RV. However, once
RV is established, the role for antiTNF therapy for manage-
ment of the vasculitis of RA is uncertain. Some reports claim
benefits of this approach, others suggest no benefit or disease
worsening while patients are on TNF inhibitors (Puechal
et al. 2008).
Decisions about the treatment of RV must be made on a
case-by-case basis. However, high-dose glucocorticoids and
cyclophosphamide are the standard of care for disseminated
disease involving multiple organs.
Pearl: A rise in the white blood cell count, thrombocytosis,
and acute phase reactants may precede or accompany RV.
Comment: A marked increase in acute phase reactants may
Fig. 2.5 Palpable purpura occurring 4 months after the initiation of reflect active joint disease as well as extraarticular manifesta-
etanercept therapy for rheumatoid arthritis in a 61-year-old woman.
Biopsy of a lesion revealed leukocytoclastic vasculitis. The lesions
tions. There is no single test that predicts the development of
resolved with discontinuation of etanercept. Oral prednisone was initi- RV, but patients with RV often have high titers of rheumatoid
ated at 20 mg daily and tapered over 6 weeks (Figure courtesy of factor and/or antiCCP antibodies, anemia, elevated acute
Dr. Eric Matteson) phase reactants, thrombocytosis, and leukocytosis (Turesson
and Matteson 2008). Patients with antinuclear antibodies and
promotes arterial sclerosis and may promote occlusive vas- antineutrophil cytoplasmic antibodies that cause perinuclear
culopathy in patients with vasculitis. (P-ANCA) staining by immunofluorescence have higher fre-
quencies of extraarticular manifestations of RA, including
Pearl: TNF inhibitors can cause a drug-induced vasculitis,
vasculitis.
but this is not “RV”.
Myth: Distal, symmetric polyneuropathy occurring in a
Comment: Figure 2.5 illustrates a case of small-vessel, leu-
patient who has longstanding RA is likely to be a vasculitic
kocytoclastic vasculitis induced by etanercept. The form of
neuropathy.
vasculitis, limited to the skin and involving small blood ves-
sels only, is a different entity from RV. TNF inhibitor-induced Reality: The inclination is to lump most patients who have
vasculitis is usually self-limited, resolves with discontinua- symmetrical sensory symptoms into the category of distal
tion of the agent, and is unlikely to develop if other TNF peripheral neuropathy, along with the 400 or so other causes
inhibitors are used. of this disorder (including diabetic neuropathy and alcoholic
neuropathy). In patients who have RA who are the typical
Pearl: RV overlaps substantially with polyarteritis nodosa
substrate for RV, however, the cause of such symptoms is
and tends to affect the same types of organs.
often vasculitic neuropathy (Puechal et al. 1995). Careful
Comment: Parallels between RV and polyarteritis nodosa have evaluations of these patients and an increase in their RA ther-
been recognized for more than half a century (Ball 1954). apy may be required, even if no specific therapy for vasculi-
Both disorders tend to involve medium-sized arteries and to tis is initiated.
affect the skin, peripheral nerves, gastrointestinal tract, heart, In contrast to the myriad other causes of peripheral neu-
and other organs. RV is somewhat less likely to cause microa- ropathy (e.g., diabetes-associated and alcohol-related periph-
neurysms than is polyarteritis nodosa, but microaneurysms eral nerve dysfunction), the onset of vasculitic neuropathy in
complicating the vasculitis of rheumatoid disease have cer- rheumatoid vasculopathy can be quite rapid. An unusually
tainly been described (Achkar 1995; Pagnoux et al. 2005). swift tempo may be the telltale piece of information indicat-
ing that further evaluation is required.
Pearl: RV is associated with other extraarticular disease,
including scleromalacia perforans. Myth: The first symptom of mononeuritis multiplex is pain or
weakness.
Comment: Vasculitis of RA almost always occurs in patients
with other extraarticular features, including nodulosis and Reality: As a rule, pain is not striking at the onset of vascu-
inflammatory eye disease. The vasculitis may involve sys- litic neuropathy; rather, the predominant symptom is anes-
temic organs including the viscera and the cranial arteries. thesia. Within days or weeks of the start of sensory symptoms,
The involvement of parenchymal organs is often catastrophic muscle weakness caused by motor nerve infarction may
2 Rheumatoid Vasculitis 19

ensue (Said and Lacroix 2005). Patients often experience

their first symptoms – numbness, tingling, or other sensory
symptoms – after a period of sleep (Schmid et al. 1961).
Myth: Mononeuritis multiplex is always asymmetric.
Reality: Mononeuritis multiplex has three clinical hallmarks:
asymmetry, asynchrony, and a predilection for distal nerves.
Early in the process, nerve involvement is likely to affect one
side more than the other (e.g., the left peroneal nerve, leading
to a unilateral foot drop). Within days to weeks, however, the
process can develop a more symmetrical appearance, as dis-
tal peripheral nerves are affected in an additive fashion.
The full development of motor dysfunction within a given
nerve often occurs within the day of onset. There is maximal
damage at the time the condition is recognized, followed by Fig. 2.6 Healing, superficial infarctions of rheumatoid nodules. The
patient developed several ulcers and mononeuritis multiplex of the
deficits that persist for weeks, months, and sometimes lower extremities approximately 4 months after the onset of these elbow
forever. and finger infarctions (Figure courtesy of Dr. Eric Matteson)
Myth: Diagnosing vasculitic neuropathy in RA often means
recognizing fire through a lot of smoke. This is true even in
million individuals to less than 4 per million was reported
the interpretation of histopathology.
(Watts et al. 2004). In contrast, in a population-based study
Reality: Sural nerve biopsies sometimes show active arteri- from Olmsted County, Minnesota that examined data cover-
tis, particularly if the patient has been treated intensively ing a period of 40 years observed no decrease in either the
before the procedure. However, the finding of a proliferative incidence of RV or other extra-articular manifestations of
endarteritis of the epineurium reflects the healed stage of a RA (Turesson 2004). Data on the impact of biologic agents
previous acute arteritis (Conn et al. 1972). Clinicopathologic on RV is awaited eagerly.
correlation is essential in such cases. In the optimal situation,
Pearl: The development of scleritis in a patient with RA is
this means sitting down at the microscope with the neuro-
often the first indication of RV.
pathologist and correlating the histopathologic findings with
the patient’s clinical presentation, serologic features, and Comment: In a patient with longstanding, seropositive RA,
electrodiagnostic study results. scleritis equals vasculitis, even if histopathological proof is
difficult to obtain. In one series of nine patients with RA and
Pearl: Vasculitis is a crucial feature in the formation of rheu-
necrotizing scleritis, scleral biopsies confirmed vasculitis in
matoid nodules.
each case (Fong 1991).
Comment: Vascular inflammation has long been considered a Scleritis can occur in either anterior or posterior locations
primary event in the formation of rheumatoid nodules (Okhravi et al. 2005). Bilateral disease is quite common, though
(Sokoloff 1953). A small-vessel vasculitis that leads to fibrin- one eye can be affected more severely than the other. Anterior
oid necrosis forms the core of the lesion in rheumatoid nod- forms of scleritis are evident from the appearance of the eye. In
ules. This core is surrounded by a proliferation of fibroblasts. contrast to episcleritis (which may occur in the absence of an
Although RV is strongly associated with rheumatoid nod- underlying condition), scleritis is usually highly symptomatic
ules, only a minority of patients with nodules develop RV. and cannot be ignored by the patient.
The nodules of one patient who did are shown in Fig. 2.6. Anterior scleritis is subdivided further into three clinical
variants:
Myth: The incidence of RV is diminishing.
1. Diffuse (least severe) (Fig. 2.7a)
Reality: Although the assumption that more effective therapy
2. Nodular (intermediate in severity) (Fig. 2.7b)
will lead to the observation of less RV in the future, there
3. Necrotizing (most severe) (Fig. 2.7c)
remain few data on the impact of biologic agents on the inci-
dence of RV. Data from the era just preceding the common These variants are largely non-overlapping. Progression
availability of biologics are contradictory on the changing from one variant to another is unusual. The necrotizing form
incidence of RV (Watts et al. 2004; Turesson 2004). of scleritis may lead to scleromalacia perforans.
In a population-based study from Norfolkshire in the In contrast to anterior scleritis, obvious on clinical exami-
United Kingdom that covered the years from 1988 to 2002, a nation, posterior scleritis must be diagnosed by inference
decrease in the annual incidence of RV from about 12 per from the patient’s symptoms – primarily a deep-seated pain
20 J. H. Stone and E. L. Matteson

Fig. 2.8 Patch sewn over a perforated cornea that resulted from rheu-
matoid vasculitis and the corneal melt syndrome. The patient is blind
b in this eye, but may benefit in the future from a corneal transplant.
(Figure courtesy of Dr. John Stone)

Comment: Two terms rheumatologists must know in order to

converse with their ophthalmology colleagues and avoid ocu-
lar disasters are PUK (an acronym for peripheral ulcerative
keratitis, generally spelled out: “P – U – K”), and its potential
consequence, corneal melt. A corneal transplant in the eye of
a patient with PUK and corneal melt is shown in Fig. 2.8.
PUK can be thought of as necrotizing scleritis that is situ-
ated in a highly unfortunate portion of the eye: the outer rim
of the cornea or the kerato-scleral junction. PUK and necro-
tizing scleritis often occur together. In PUK, inflammatory
cells infiltrate the peripheral area of the cornea, leading to a
c crescent-shaped ulceration near the corneoscleral junction.
The principal complication of PUK is a “corneal melt” syn-
drome, in which corneal keratolysis, perforation of the globe,
and visual failure can result within days (Squirrell et al. 1999).
Following the occurrence of a corneal melt, patients often
lose all useful vision in the eye. Both necrotizing scleritis and
PUK require urgent therapy with high doses of immunosup-
pressive medications, generally both high-dose glucocorti-
coids and cyclophosphamide.
Myth: The finding of rheumatoid nodules in the lungs usually
coincides with RV.
Reality: Although rheumatoid nodules occur in the lungs and
Fig. 2.7 (a–c) Diffuse, nodular, and necrotizing scleritis in patients
pathologic evidence of vasculitis is an inherent feature of
with rheumatoid arthritis (Figures courtesy of Dr. John Stone)
rheumatoid nodules wherever they are found, most patients
with rheumatoid nodules in the lungs do not have evidence
and ocular tenderness, but also visual blurring. The posterior of systemic vasculitis (Sokoloff et al. 1953; Yousem et al.
coat of the eye is observed to be thickened on orbital ultra- 1985). The biggest problem that pulmonary rheumatoid nod-
sonography or magnetic resonance imaging (McCluskey ules cause is the concern that they may actually represent a
et al. 1999). Posterior scleritis should be considered in malignancy.
patients with RA who present with severe headaches. Pearl: Two types of renal disease occur in RV: a PAN-like
Pearl: Beware the “corneal melt” syndrome in the RA patient medium-vessel renal arteritis, and a pauci-immune glomeru-
with a painful red eye. lonephritis. Both of these complications are rare.
2 Rheumatoid Vasculitis 21

Comment: Both renal artery involvement similar to that Myth: RV can cause an immune complex-mediated glomeru-
which occurs in polyarteritis nodosa and glomerulonephritis lonephritis.
that is reminiscent of microscopic polyangiitis, or Wegener’s
Reality: The histological findings in RV may include: (1) leu-
granulomatosis are well described (Ball 1954; Johnson 1959;
kocytoclastic vasculitis associated with immune complex
Harper et al. 1997; Boers et al. 1987). Medium-vessel arteri-
deposition in venules, capillaries, and arterioles; and (2)
tis involving the renal vessels rarely leads to microaneu-
pauci-immune lesions (i.e., inflammation associated with
rysms, in contrast to PAN. The glomerular pathology caused
sparse deposition of immunoreactants) in medium-sized
by vasculitis in RA is a pauci-immune glomerulonephritis,
arteries and renal glomeruli. One paradoxical finding in RV is
sometimes associated with crescent formation.
that although glomeruli are considered small blood vessels –
This type of glomerular disease contrasts with common
the renal equivalent of capillaries – the well-documented
beliefs about the pathophysiology of RV in other organs,
cases of renal involvement in RV are generally associated
which is generally considered to result from an immune
with pauci-immune glomerulonephritis.
complex-mediated process characterized by the organ depo-
sition of rheumatoid factor, complement components, and
other immunoreactants (Epstein and Engleman 1959).
Antineutrophil cytoplasmic antibodies (ANCA) do not play References
a role in the glomerulonephritis that occasionally compli-
cates rheumatoid disease. If a patient has ANCA directed
Achkar AA, Stanson AW, Johnson CM et al Rheumatoid vasculitis
against either proteinase-3 or myeloperoxidase, then a sec- manifesting as intra-abdominal hemorrhage. Mayo Clin Proc 1995;
ond process is present. 70(6):565–9
Ball J. Rheumatoid arthritis and polyarteritis nodosa. Ann Rheum Dis
Pearl: The strongest risk factor for the development of RV is 1954; 13:277–290
a high serum titer of rheumatoid factor. Boers M, Croonen AM, Dijkmans BAC, et al Renal findings in rheuma-
toid arthritis: Clinical aspects of 132 necropsies. Ann Rheum Dis
Comment: Vasculitis in seronegative RA is rare (Mongan 1987; 46:658–663
et al. 1969; Geirsson et al. 1987). High titers of rheumatoid Bywaters EGL. A variant of rheumatoid arthritis characterized by recur-
factor are reported consistently to be the strongest predictor rent digital pad nodules and palmar fasciitis, closely resembling
palindromic rheumatism. Ann Rheum Dis 1949; 8:1–30
of the development of RV (Vollertsen et al. 1986 (Medicine); Bywaters EGL. Peripheral vascular obstruction in rheumatoid arthritis
Geirsson et al. 1987). Few data exist to date on the correla- and its relationship to other vascular lesions. Ann Rheum Dis 1957;
tion between antiCCP antibodies, but this correlation is 16:84–103
undoubtedly high. Serum levels of IgM rheumatoid factor – Conn DL, Schroeter AL, McDuffie FC. Immunopathologic study of
sural nerves in rheumatoid arthritis. Arthritis Rheum 1972; 15:135
the serotype usually measured – are poor markers of disease Cruikschank B. The arteritis of rheumatoid arthritis. Ann Rheum Dis
activity (Scott et al. 1981; 38). 1954; 13:136–145
Epstein WV, Engleman EP. The relation of the rheumatoid factor con-
Myth: Cyclophosphamide should not be used if the only tent of serum to clinical neurovascular manifestations of rheumatoid
manifestation of RV is skin ulceration. arthritis. Arthritis Rheum 1959; 2:250–258
Fong LP, Sainz de la Maza M, Rice BA et al Immunopathology of
Reality: Some baseball managers err by “going to the bull- scleritis. Ophthalmology 1991; 98(4):472–9
pen” – i.e., bringing on a relief pitcher – too late. Some clini- Geirsson AJ, Sturfelt G, Truedsson L, et al Clinical and serological fea-
cians make the same mistake by leaving cyclophosphamide tures of severe vasculitis in rheumatoid arthritis: Prognostic impli-
cations. Ann Rheum Dis 1987; 46:727–733
on the bench for too long. Harper L, Cockwell P, Howie AJ, et al Focal segmental necrotizing
Clinicians rightly resent having to contemplate cyclo- glomerulonephritis in rheumatoid arthritis. QJM 1997; 90:125
phosphamide for “skin-only” disease. However, in some Hazes JM. Management of Extraarticular Disease and Complications.
patients, cyclophosphamide is an essential drug, without In: Hochberg MC, Silman AJ, Smolen JS, Weinblatt ME, Weisman
MH (eds) Extraarticular Features of Rheumatoid Arthritis and
which, healing will not occur. The baleful direct effects (pain, Systemic Involvement. Rheumatology (4th edn). Philadelphia, PA,
reduced mobility, increasing incapacitation), and the sub- Mosby, 2008, pp. 897–914
stantial complications of leg ulcers (infection, sepsis), under- Hurlimann D, Forster A, Noll, G et al Anti-tumor necrosis factor-alpha
score the need to suppress and heal these lesions as quickly treatment improves endothelial function in patients with rheumatoid
arthritis. Circulation 2002; 106:2184
as possible. Johnson RL, Smyth CJ, Holt GW, et al Steroid therapy and vascular
High doses of prednisone, conventional DMARDs, and lesions in rheumatoid arthritis. Arthritis Rheum 1959; 2:224–249
biologic agents should be considered before cyclophosph- Kumeda, Y, Inaba, M, Goto, H, et al Increased thickness of the arterial
amide. However, for disease that is severe at the time of diag- intima-media detected by ultrasonography in patients with rheuma-
toid arthritis. Arthritis Rheum 2002; 46:1489
nosis and for vasculitis that does not continue to improve as McCluskey PJ, Watson PG, Lightman S, et al Posterior scleritis: Clinical
glucocorticoids are tapered, the use of cyclophosphamide features, systemic associations, and outcome in a large series of
earlier rather than later may be the correct choice. patients. Ophthalmology 1999; 106:2380–2386
22 J. H. Stone and E. L. Matteson

Mongan ES, Cass RM, Jacox RF, et al A study of the relation of serone- Turesson C, Matteson EL. Management of extraarticular disease mani-
gative and seropositive rheumatoid arthritis to each other and to festations in rheumatoid arthritis. Curr Opin Rheum 2004;16:206–211
necrotizing vasculitis. Am J Med 1969; 47:23–35 Turesson C, Matteson EL. Extraarticular features of rheumatoid arthri-
Okhravi N, Odufuwa B, McCluskey P, et al Scleritis. Surv Ophthalm tis and systemic involvement. In: Hochberg MC, Silman AJ, Smolen
2005; 50(4):351–363 JS, Weinblatt ME, Weisman MH (eds) Rheumatology (4th edn).
Pagnoux, C, Mahr, A, Cohen, P, Guillevin, L. Presentation and outcome of Philadelphia, PA, Mosby, 2008, pp. 773–783
gastrointestinal involvement in systemic necrotizing vasculitides: Turesson C, O’Fallon WM, Crowson C, et al Extraarticular disease mani-
Analysis of 62 patients with polyarteritis nodosa, microscopic poly- festations in rheumatoid arthritis: Incidence, trends, and risk factors
angiitis, Wegener granulomatosis, Churg-Strauss syndrome, or rheuma- over 46 years. Annals of Rheumatic Disease 2003; 62:722–727
toid arthritis-associated vasculitis. Medicine (Baltimore) 2005; 84:115 Turesson, C, McClelland, R, Christianson, TJ, Matteson, EL. No decrease
Park YB, Ahn CW, Choi HK, et al Atherosclerosis in rheumatoid arthri- over time in the incidence of vasculitis or other extra-artic-ular manifesta-
tis: Morphologic evidence obtained by carotid ultrasound. Arthritis tions in rheumatoid arthritis (abstract). Arthritis Rheum 2004; 50:S380
Rheum 2002; 46:1714 Turesson C, Schaid DJ, Weyand CM, et al The impact of HLA-DRB1
Price-Forbes AN, Watts RA, Lane SE, et al Do we need to treat isolated genes on extraarticular disease manifestations in rheumatoid arthri-
nailfold vasculitis (NFV) in rheumatoid arthritis (RA) more aggres- tis. Arthritis Res Ther 2005; 7:R1386–R1393
sively? Abstract 848, American College of Rheumatology annual Van Albada-Kuipers GA, Bruijn JA, Westedt ML, et al Coronary arteritis
scientific meeting, October 2002 complicating rheumatoid arthritis. Ann Rheum Dis 1986; 45:963
Puechal X, Said G, Hilliquin P, et al Peripheral neuropathy with necrotiz- Van Gaalen FA, Linn-Rasker SP, van Venrooij WJ, et al Autoantibodies
ing vasculitis in rheumatoid arthritis. A clinicopathologic and prognos- to cyclic citrullinated pepties predict progression to rheumatoid
tic study of thirty-two patients. Arthritis Rheum 1995; 38:1618–1629 arthritis in patients with undifferentiated arthritis. Arthritis Rheum
Puechal X, Miceli-Richard C, Mejjad O, et al Anti-tumour necrosis fac- 2004; 50(3):709–715
tor treatment in patients with refractory systemic vasculitis associ- Vollertsen RS, Conn DL, Ballard DJ, et al Rheumatoid vasculitis:
ated with rheumatoid arthritis. Ann Rheum Dis 2008; 67:880–884 Survival and associated risk factors. Medicine 1986; 65(6):365–375
Said G, Lacroix C. Primary and secondary vasculitis neuropathy. J Voskuyl AE, Zwinderman AH, Wested ML, et al The mortality of rheu-
Neurol 2005; 252:633–641 matoid vasculitis compared with rheumatoid arthritis. Arthritis
Schmid FR, Cooper NS, Ziff M, et al Arteritis in rheumatoid arthritis. Rheum 1996; 39:266–271
Am J Med 1961; 30:56–83 Wallberg-Jonsson, S, Cvetkovic, JT, Sundqvist, KG, et al Activation of
Scott DG, Bacon PA, Tribe CR. Systemic rheumatoid vasculitis: A the immune system and inflammatory activity in relation to markers
clinical and laboratory study of 50 cases. Medicine (Baltimore) of atherothrombotic disease and atherosclerosis in rheumatoid
1981; 60:288 arthritis. J Rheumatol 2002; 29:875
Sokoloff L. The heart in rheumatoid arthritis. Am Heart J 1953; 45:635 Watts RA, Carruthers DM, Scott DG. Isolated nail fold vasculitis in
Sokoloff L, McCluskey RT, Bunim JJ. Vascularity of the early subcutane- rheumatoid arthritis. Ann Rheum Dis 1995; 54:927
ous nodule in rheumatoid arthritis. Arch Pathol 1953; 55: 475–479 Watts RA, Mooney J, Lane SE, Scott DG. Rheumatoid vasculitis:
Squirrell DM, Winfield J, Amos RS. Peripheral ulcerative keratitis “cor- Becoming extinct?. Rheumatology (Oxford) 2004; 43:920
neal melt” and rheumatoid arthritis: A case series. Rheumatology Yousem SA, Colby TV, Carrington CB. Lung biopsy in rheumatoid
1999; 1245–1248 arthritis. Am Rev Respir Dis 1985; 131:770
 Adult-Onset Still’s Disease
3
Michael H. Weisman

that is separate from that of RA and JRA (JRA is now termed

3.1 Overview of Adult-Onset Still’s juvenile idiopathic arthritis (Chap. 4).
Disease
Pearl: Bywaters anticipated the relationship between AOSD
and the group of conditions now termed “autoinflammatory
› Adult-onset Still’s disease (AOSD) was described
syndromes” (Chapter 5).
originally by Eric Bywaters in 1971.
› AOSD closely resembles systemic-onset juvenile idio- Comment: Bywaters commented on the similarities between
pathic arthritis, a pediatric disorder known originally AOSD, familial Mediterranean fever, and an entity now
as Still’s disease. known as the Muckle–Wells syndrome (Chapter 5). These
› An intensely inflammatory, multiorgan system disorder other disorders, now grouped as autoinflammatory syndrome,
characterized by a polyarthritis, high spiking fevers, are characterized by an urticarial rash, “aguey bouts” (malar-
and an evanescent, salmon-colored rash (Figs. 3.1a–c). ial fever), and amyloidosis, all features typical of AOSD.
› The rash is described as small macules that disappear In truth, Bywaters’ original paper is almost all one needs
during the night but reappear the next day, usually to know about AOSD. Little of any major significance has
coincident with a fever spike. been written since those papers, with the important excep-
› Peripheral joint involvement can be fleeting but also tion being the identification of AOSD in many patients as an
may settle into a refractory, destructive arthritis marked IL-1 mediated disease (see below).
by a tendency in some joints for fusion.
Pearl: The fever curve is the most important diagnostic cri-
› AOSD is typically seronegative. Patients do not have
terion for AOSD.
rheumatoid factor or antibodies to cyclic citrullinated
peptides. Comment: Very few conditions mimic the fevers of AOSD,
› Other common features are leukocytosis, thrombocy- with their tendency to decline to normal temperatures at least
tosis, elevations of the serum hepatic aminotransferase once every 24-h period. Most other rheumatic conditions
concentrations, splenomegaly, and serositis. associated with fever (and most infectious or malignant
› Pharyngitis is often the initial symptom. causes) tend to be associated with more continuous patterns
of fever if left untreated, with temperature elevations sus-
tained over days.
Bywaters described the fever of AOSD as dramatic and
remittent, often mistaken for sepsis and treated energetically
but in vain with a succession of antibiotics. Patients’ tempera-
Pearl: Still’s disease should not be regarded as a category of
tures should be measured every 4 h to demonstrate the charac-
“juvenile rheumatoid arthritis”.
teristic quotidian (or “double-quotidian”) pattern, because
Comment: AOSD strongly resembles the pediatric condition fever spikes can occur at any time in the 24 h period.
known as “Still’s disease”. Bywaters wrote that Still’s disease
Myth: AOSD is a benign, self-limited condition that does not
was often referred to as “juvenile rheumatoid arthritis” (JRA),
produce joint damage or disability.
a name that (he believed) prejudged the issue of the disorder’s
appropriate classification (Bywaters 1971). Bywaters ques- Reality: This statement is a total myth. The original papers
tioned whether Still’s disease was really related directly to the by Bywaters and Aptekar in the early 1970s emphasized the
JRA/RA disease spectrum, or whether it comprised an entirely benign nature of AOSD (Bywaters 1971; Aptekar et al. 1973),
separate condition. Bywaters strongly favored the hypothesis but almost every large series since then has indicated that the
that AOSD and Still’s disease formed a spectrum of disease disease becomes chronic, if intermittent, in the majority of

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 23

DOI:10.1007/978-1-84800-934-9_3, © Springer Science + Business Media B.V. 2009
24 M. H. Weisman

Fig. 3.1 The evancescent,

a d
salmon-colored rash of adult-
onset Still’s disease (Figures
courtesy of Dr. John Stone)

b e

c f

patients. In addition, it can be notoriously difficult to treat. in multiple joints in a symmetric fashion, with a distribution
The joint manifestations and the systemic features may be similar to that of seropositive RA.
out of sync with each other: one can dominate the clinical The state of the literature on AOSD is that it is difficult to
picture while the other remains quiet. This creates challenges make anything other than general conclusions about its natu-
in selecting how aggressive to be with therapy. ral history, because of the many different definitions used to
Joint destruction can occur in AOSD patients. Hip or knee characterize the disease course, the varying length of follow-
replacements are not uncommon sequellae of this disease. up of different studies, and the selection bias inherent in many
The wrists are particularly prone to damage in AOSD and of the cohorts reported (Esdaile 2008). One reasonable state-
often develop joint fusion. Joint damage can occur in AOSD ment about AOSD is that 30–50% of patients will display a
3 Adult-Onset Still’s Disease 25

chronic course, and the remainder are divided equally between support the efficacy of traditional DMARDs in this disease,
those who have self-limited disease and those who experi- e.g., methotrexate, sulfasalazine, and hydroxychloroquine.
ence intermittent disease exacerbations. The initial enthusi- Finally, much against expectations, anti-TNF therapies for
asm that AOSD is a benign, self-limited condition is clearly AOSD have demonstrated only marginal (read largely disap-
not the case most of the time. pointing) results (Husni et al. 2002; Fautrel et al. 2005).
The second major myth uttered above pertains to gluco-
Pearl: Arthritis may be absent in the initial stages of AOSD.
corticoids. In fact, high doses of systemic glucocorticoids
Comment: Joint pain is common in the early stages, but frank are often required to gain control of the acute disease mani-
arthritis may not be evident. Many patients with AOSD present festations. If a relapse of symptoms occurs during the gluco-
as fevers of unknown origin. In this setting, joint involvement corticoid taper, then methotrexate should be started with a
can be masked or ignored by the overwhelming nature of the rapid titration of the dose toward 25 mg/week. Clinical expe-
systemic disease features such as fever, rash, anemia, leukocy- rience suggests that the combination of glucocorticoids and
tosis, and thrombocytosis. The absence of arthritis as a domi- methotrexate suffices to control AOSD in a sizeable propor-
nant symptom makes the diagnosis difficult in the early stages tion of patients, but not all.
and probably accounts for the oft-quoted diagnostic delays, The extrapolation of IL-1 inhibition strategies to AOSD
which frequently approach 6 months. made perfect sense in view of the similarities between AOSD
In some patients, mild joint involvement is not accompa- and the autoinflammatory syndromes, most of which are
nied by obvious swelling or synovitis. However, if joint mediated by abnormalities in the IL-1 pathway. Multiple
symptoms have not become more obvious within 6 months reports indicate a rapid and sustained reduction in signs and
of the patient’s presentation, then the diagnosis of AOSD symptoms along with parallel improvement of inflammatory
should be held in some doubt. markers (Fitzgerald et al. 2005; Bresnihan 2008). Some
patients with treatment-resistant AOSD have experienced
Pearl: The rash of AOSD does not itch, burn, or sting, and
striking responses to IL-1 inhibition. This author has even
may therefore be missed by patient and clinician alike.
used responsiveness to anakinra as a diagnostic test for
Comment: Aside from its salmon-colored appearance and AOSD in some patients.
characteristically evanescent nature, there are few other rules However, anakinra is a difficult drug to use, not only
about the AOSD rash except that is it generally asymptom- because of the requirement for daily parenteral injections but
atic. The rash has a tendency to affect the trunk but certainly also the frequent and sometimes dramatic skin rashes associ-
occurs on the extremities, as well. ated with this agent. Such reactions tend to resolve within a
It has been written that the rash of AOSD accompanies the few weeks of continued administration, but newer IL-1 block-
fever spike in the afternoon when the rounding physicians are ing agents with greater ease of administration are welcome.
not in attendance, and that clinicians who see the patient only
Myth: The presence of hepatic, pulmonary, and cardiac
in the early morning or late in the day will miss the rash
involvement decreases the likelihood of AOSD and adds new
entirely. This statement flies in the face of the fact that the
urgency to the search for vasculitis or an infectious etiology
fevers of AOSD can occur at any time. The more important
of the patient’s problems.
point is that repeated examinations and vigilance for obser-
vation of the rash may be required to detect the AOSD exan- Reality: Disease manifestations in the liver, lungs, and heart
thema. The poor lighting of most hospital rooms and failure were present but not emphasized clearly in the initial reports
to examine the patient’s skin thoroughly probably both con- of Bywaters, Aptekar, and colleagues (Bywaters 1971;
tribute to the elusiveness of the rash. Dermatographism, a Aptekar et al. 1973). In the decades since those original
common feature of AOSD, is non-specific. descriptions, a fuller picture of the disease has emerged.
AOSD patients can present with dramatic liver function
Myth: AOSD disease management is fairly straightforward
abnormalities. As a result, various forms of hepatitis must
and tends to follow the approach once used in RA.
be considered and excluded. A reactive hemophagocytic
Nonsteroidal anti-inflammatory drugs (NSAIDs) should be
syndrome must also be considered. Pleural effusions and
attempted first, followed by disease-modifying antirheumatic
even interstitial lung disease have been observed, as well.
drugs (DMARDs), and then anti-TNF agents. There is no
Pericarditis is evident in some patients.
role for systemic glucocorticoids.
In a disease for which there exists no single diagnostic
Reality: Unfortunately, these statements contain not one but test, one must always keep an open mind about other etiolo-
several myths. First, the management of AOSD patients is gies. However, the presence of liver, lung, or cardiac disease
anything but straightforward. Patients seldom have a com- does not necessarily exclude AOSD.
plete, sustained response to NSAIDs alone. The author some-
Pearl: Primary renal involvement does not occur in AOSD.
times uses a high dose up to 200 mg/day of indomethacin to
get the biggest bang for the buck, yet additional therapies are Comment: Despite the protean nature of AOSD, this Pearl
required in most cases. No clinical trial data are available to holds up. Primary renal disease is not characteristic of this
26 M. H. Weisman

disorder. Of course, one must account for the possibility that this has not been tested in a prospective study. Moreover, the
renal dysfunction might result from intercurrent problems, incremental improvement in terms of diagnosis offered by an
e.g., acute tubular necrosis caused by NSAIDs. But the guid- elevated serum ferritin level compared with measurements of
ing principle that primary renal involvement does not occur other acute phase reactants is not clear.
in AOSD can be very useful in differentiating AOSD from
various forms of vasculitis, lupus, or malignancies. Pearl: Advances in our understanding of innate immunity
mediated diseases and certain auto-inflammatory conditions
Pearl: The diagnosis of AOSD can be made without actually (especially in children) have enabled us to diagnose and
seeing the patient and with the aid of a cell-phone camera. treat AOSD more effectively.
Comment: This outrageous Pearl is actually true in some
cases. Few disease entities reproduce the fever curve of AOSD Comment: The observations that certain “autoinflammatory
and if a reliable nursing record of temperature recordings can syndromes” are under genetic control and that many of those
be obtained, the consultant can make the diagnosis from the conditions can be traced to mutations which cause increased
nurses’ station. Photodocumentation of the rash before meet- IL-1 production have opened our eyes to new therapeutic tar-
ing the patient in person can add further certainty to the gets for AOSD (Allantaz et al. 2007). The discovery of the
diagnosis. importance of the IL-1 pathway in the pathogenesis of AOSD
and related conditions has increased our understanding of
Pearl: Severe sore throat can be both a misleading clue as the inflammasome and opened the door for the successful
well as a helpful adjunct to the diagnosis of AOSD. treatment of many AOSD patients with IL-1 antagonists.
Comment: Some patients with AOSD present with severe sore
throats, even to the point of having difficulty swallowing. Despite
the severity of this symptom, patients with this complaint gener-
ally manifest no physical evidence of tonsillar, laryngeal, or tra- References
cheal inflammation. These patients usually undergo courses of
antibiotics, often have multiple visits to otolaryngologists, Allantaz F, Chaussabel D, Banchereau J, Pascual V. Microarray-based
identification of novel biomarkers in IL-1-medicated diseases. Curr
and even sometimes undergo endoscopic procedures in Opin Immunol 2007; 19:623–632
efforts to understand the source of this complaint. Aptekar RG, Decker JL, Bujak JS, Wolff SM. Adult onset juvenile
Sore throat is as common as rash and arthritis as the pre- rheumatoid arthritis. Arthritis Rheum 1973; 16:715–718
senting complaint of AOSD (Nguyen and Weisman 1997). Bresnihan B. Cytokine neutralizers: IL-1 inhibitors. In: Hochberg MC,
Silman AJ, Smolen JS, Weinblatt ME, Weisman MH (eds)
Thus, to the knowledgeable clinician, a severe sore throat Rheumatology (4th edn). Edinburgh, UK, Mosby, 2008, pp. 495–500
that defies the conventional explanation can be a clue to the Bywaters, EGL. Still’s disease in the adult. Ann Rheum Dis 1971;
diagnosis of AOSD. 30:121–132
Esdaile J. Adult Still’s disease. In: Hochberg MC, Silman AJ, Smolen JS,
Myth: An elevated ferritin level is diagnostic for AOSD. Weinblatt ME, Weisman MH (eds) Rheumatology (4th edn).
Edinburgh, UK, Mosby, 2008, pp. 785–792
Reality: There are no diagnostic tests for AOSD. The serum Fautrel B, Sibilia J, Mariettte X, et al Tumor necrosis factor alpha
ferritin is nearly always elevated in AOSD, but because of blocking agents in refractory adult Still’s disease: An observational
the low specificity of this test its positive predictive value is study of 20 cases. Ann Rheum Dis 2005; 64:262–266
Fitzgerald AA, LeClercq SA, Yan A, et al Rapid responses to anakinra
poor. Other acute phase reactants are also elevated, and often in patients with refractory adult-onset Still’s disease. Arthritis
strikingly so. As examples, an increased ESR is nearly uni- Rheum 2005; 52:1794–1803
versal; the platelet count can exceed one million per cubic Husni ME, Maier AL, Mease PJ, et al Etanercept in the treatment of adult
millimeter. A significant leukocytosis is also very common. patients with Still’s disease. Arthritis Rheum 2002; 46:1171–1176
Nguyen KH, Weisman MH. Severe sore throat as a presenting symptom
The negative predictive value of a serum ferritin level nor- of adult-onset still’s disease: A case series and review of the litera-
mal or only moderately elevated is probably quite high, but ture. J Rheumatol 1997; 24:592–597
 Juvenile Idiopathic Arthritis
4
Patience H. White, Patricia Woo, and Carol B. Lindsley

Myth: Systemic-onset juvenile idiopathic arthritis (sJIA)

4.1 Overview of Juvenile represents a homogeneous disease subset.
Idiopathic Arthritis
Reality: The classic clinical signs of quotidian fevers, macu-
lar rashes, and arthritis separate patients with sJIA (some-
› Juvenile idiopathic arthritis (JIA) is an umbrella term
times termed “Still’s disease”) from the oligoarticular and
for a group of persistent arthritides of unknown etiol-
polyarticular JIA subtypes. However, clinicians have long
ogy that affect children under the age of 16, and last
recognized that sJIA has a broad clinical spectrum that
for more than 6 weeks (Petty et al. 2004). The clini-
includes patients with mild to severe arthritis, transient to
cally defined categories are shown in Table 4.1.
persistent systemic features, variable disease duration, dif-
› The pathogenesis of these diseases involves both auto-
fering propensities to relapse, and disparate responses to
immune and genetic factors. Dysregulation of the
standard therapies.
immune and inflammatory systems are observed. In
Patients with sJIA tend to respond to methotrexate less
addition, hormonal, infectious, and other environmen-
well than do those with other forms of JIA (Woo 2006).
tal agents yet to be identified probably participate in
Fewer than 50% of patients with sJIA achieve good clinical
the disease-diagnosing process.
responses on methotrexate and even these are often not sus-
› The diagnosis of JIA is rendered from the combination
tained (Lovell 2006). Some patients show dramatic improve-
of data derived from the history, physical examination,
ment with IL-1 receptor antagonist therapy, but overall the
and laboratory testing.
percentage of patients who respond to this intervention is
› The prevalence of JIA varies depending on subtypes.
less than 50% (Lequerre and Quartier 2008). Finally, a sub-
As a group, the prevalence is generally agreed to be
stantial proportion of patients with sJIA appear to respond to
1:1,000 children.
IL-6 blockade (Yokota et al. 2008). These results suggest
› For the vast majority of patients with JIA, the immuno-
strongly that subtypes of sJIA exist.
genetic associations, clinical course and functional
Other systemic inflammatory disorders that mimic the
outcome are quite different from adult-onset rheuma-
clinical features of sJIA (e.g., the autoinflammatory/periodic
toid arthritis. In addition, certain specific disease sub-
fever syndromes) have specific single-gene mutations that
sets are not observed at all or are seen very rarely in
lead to an imbalance toward a proinflammatory state. These
adults. These subsets include the oligoarticular pattern
conditions include the chronic infantile neurological cutane-
of arthritis associated with uveitis and antinuclear anti-
ous arthropathy (CINCA) syndrome, the Muckle–Wells syn-
bodies, and sJIA (sJIA).
drome, the hyper-IgD syndrome, and the tumor-necrosis
› Most patients with JIA do not achieve a remission and
receptor associated periodic syndromes (TRAPSs). All of
require long-term treatment. However, the use of new
these conditions are associated with urticarial or macular
therapies such as methotrexate and the biologic agents
rashes, periodic fevers, constitutional disturbances, and arth-
has improved the outcome of JIA.
ralgia or arthritis – disease features that mimic those of sJIA
closely. Gene association studies have revealed many cytokine
genes that alter the homeostasis of the sJIA inflammatory
response (Woo 2000). Thus, although sJIA and the autoin-
flammatory syndromes can bear striking resemblance to each
other clinically, the genetic background of sJIA appears to be
far more complex.

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 27

DOI:10.1007/978-1-84800-934-9_4, © Springer Science + Business Media B.V. 2009
28 P. H. White et al.

Table 4.1 International league against rheumatism classification of • Cytopenias (leucopenia, thrombocytopenia, and anemia, in
JIA (Petty 2004) contrast to the leucocytosis and thrombocytosis of sJIA)
• Oligorticular – perisistent or extended • Extremely elevated serum hepatic transaminase levels
• Polyarticular – rheumatoid factor negative • The bone marrow aspirate characteristically shows
• Polyarticular – rheumatolod factor positive
hemophagocytosis
• Systemic
• Psoariatic Paradoxically, the arthritis of sJIA may improve as MAS
• Enthesitis related develops. Thus, an improvement in the patient’s arthritis, a
• Unclassified
decline in the ESR, and decreases in the white blood cell and
Note 1: Arthritis related to inflammatory bowel disease is excluded
platelet counts can all signal improvement in sJIA, and may
Note 2: sJIA is often now classified as an autoinflammatory disease
also herald the complication of MAS, a life-threatening con-
dition (Stephen et al. 1993).
Table 4.2 Blood abnormalities in patients with sJIA
Myth: JIA will “burn out” when the child grows up.
• Elevated erythrocyte sedimentation rate
• Leukocytosis Reality: Many children with JIA who have a persistent oli-
• Thrombocytosis
goarthritis, particularly those with a monoarthritis, eventu-
• Anemia
ally experience disease remission. However, some patients
• Hypoalbuminemia
• Mild increases in the hepatic transaminases experience the recurrence of arthritis in the same joints
• Elevated d-dimer levels involved earlier by JIA, or the extension of inflammatory
• Dramatic increases in serum ferritin levels arthritis to other joints much later in life (unpublished clini-
cal observation).
Recent presentations of gene expression profiles in polyar-
Pearl: Patients with sJIA are at risk for the macrophage acti- ticular JIA patients who are in remission showed abnormal
vation syndrome. expression patterns compared to controls, strengthening the
impression that the immune system is balanced in remission
Comment: More recently, the development of macrophage but not in the same way as those without JIA (Jarvis et al.
activation (or secondary hemophagocytic histiocytosis, 2006). This preliminary finding is consistent with the fact that
HLH) has been recognized to be more prevalent in sJIA there are genes associated with JIA that alter the innate and
patients. Genetic studies indicate that patients with the mac- adaptive immune responses in these patients (Prahalad and
rophage activation syndrome are a disease subset in which Glass 2008). Thus, triggers such as serious viral infections
the individuals share genetic risk factors with patients who can alter the balance to a persistent inflammatory state again,
have familial HLH (Zhang et al. 2008). leading in theory to the recurrence of arthritis. Therefore, the
Pearl: Certain clinical features may indicate the emergence of term “burnt out” should be avoided.
the macrophage activation syndrome in a patient with sJIA. Myth: IgM rheumatoid factor is usually positive in patients
Comment: Children with systemic JRA have a host of abnor- with polyarticular JIA. This assay is therefore helpful in dis-
mal findings in the blood (Table 4.2). All of these abnormali- tinguishing inflammatory disease from other forms of arthrop-
ties return to normal when the disease becomes less active. athy in children.
The macrophage activation syndrome (MAS) can occur in Reality: The polyarticular subset comprises about 40% of the
sJIA and shares many features of that disorder, but also children with JIA. However, only about 5% of JIA patients
occurs in patients with intracellular viral infections and in test positively for rheumatoid factor. JIA patients who are
patients who have no known risk factors. The following fea- seropositive tend to be adolescents with aggressive disease
tures are typical of MAS: and significant involvement of the small joints of the hands.
• Fever Thus, a negative rheumatoid factor assay in a child or adoles-
• Hepatosplenomegaly cent with arthritis does not exclude the diagnosis of JIA.
• Encephalopathy, manifested by dizziness, lethargy, and Rheumatoid nodules are also unusual in JIA, but when
disorientation present tend to occur in patients who have rheumatoid factor.
• An extraordinarily high serum ferritin level
Myth: Anti-CCP antibodies are helpful in identifying disease
• Elevated serum triglyceride levels
subtypes.
• Coagulopathy (increasing d-dimer levels and abnormal
PT and PTT) and bruising Reality: Anti-CCP antibodies have a low prevalence in JIA
• Decreasing ESR (as opposed to the elevation typical of but are increased in HLA-DR4 positive patients with polyar-
sJIA) ticular disease. They are associated with erosive disease and
4 Juvenile Idiopathic Arthritis 29

rheumatoid factor positivity, and therefore have little rele- long term, psoriatic arthritis in children demonstrates a pat-
vance beyond this small JIA subtype. tern of joint involvement that is asymmetric. The onset of
cutaneous psoriasis may follow the presentation of arthritis
Pearl: In oligoarticular JIA, the activity of the eye disease is
by several years. Therefore, queries about the patient’s fam-
independent of the activity of the joint disease.
ily history with regard to psoriasis and inspection of the nails
Comment: The activity of the uveitis that occurs in JIA is inde- for the characteristic pitting are important aspects of assem-
pendent of the disease activity in the joints. Uveitis precedes bling the entire picture and recognizing psoriatic arthritis
the onset of arthritis in about 10% of patients with olgioarticu- early (Fig. 4.3).
lar JIA. Uveitis occurs simultaneously with arthritis in 30%,
Pearl: Bone and muscle growth abnormalities are affected
and develops after arthritis in 60% (Rosenberg and Oen 1986).
substantially by joint inflammation and use (or disuse) of the
Even when both arthritis and uveitis are present in the same
extremities.
patient, the activity of these disease manifestations may be
incongruent. Thus, children with oligoarticular JIA must be Comment: Overgrowth, undergrowth, and maturation of the
screened regularly according to recommended guidelines (an bone growth plates are affected by variations in blood flow that
example of USA guidelines can be found in Table 4.3) (Cassidy occur during the course of JIA. The presence of inflammation
et al. 2006), even if the arthritis appears well controlled.
Uveitis can proceed insidiously and lead to significant ocular
damage in the absence of eye symptoms (Figs. 4.1 and 4.2).
Pearl: Single-digit arthritis is often an early sign of psoriatic
arthritis.
Comment: Psoriatic arthritis in children often begins with
diffuse, sausage-like swelling in one or more digits. Over the

Table 4.3 Screening guidelines for uveitis in patients with pauciarticular

JIA in the USA (adapted from Cassidy et al. 2006)
Type ANA Onset Duration Risk Examination
(years) (years)
Oligo/ + £6 £4 High q 3m
Poly + £6 >4 Moderate q 6m
+ £6 >7 Low q 12 m
Fig. 4.2 Uveitis – irregular pupil with synechiae
+ >6 £4 Moderate q 6m
+ >6 >4 Low q 12 m
− £6 £4 Moderate q 6m
− < or >6 >4 Low q 12 m
Systemic NA NA NA Low q 12 m
q: every; m: months

Fig. 4.3 Pitting of the nails in a pediatric patient with psoriatic

Fig. 4.1 Uveitis – band keratopathy arthritis
30 P. H. White et al.

can accelerate bone age and lead to increased bone length. Reality: Chronic arthritis in one to four joints over the initial
Alternatively, inflammation can lead to early closure of the epi- 6 months of disease meets the criteria for oligoarthritis.
physis, resulting in a shorter bone. If an extremity is underuti- However, 20–30% of children with oligoarthritis evolve to a
lized, stunting of the extremity’s growth may occur. A typical polyarticular pattern of disease (extended oligoarthritis) over
example in JIA is shortening of the forefoot resulting from the ensuing few years.
poorly-controlled ankle arthritis. Physical therapy and control
of the inflammation are essential to ensuring growth patterns Myth: Back pain is uncommon in children and most always
that are as normal as possible. means pathology.
Another functional complication of oligoarthritis is mus- Reality: Back pain is common in the general pediatric popu-
cle atrophy on both sides of the affected joint. This can often lation in developed countries and occurs in 11–36% of school
be observed strikingly in the quadriceps muscle. Muscle age children. Among pediatric patients referred to orthope-
atrophy occurs quickly and patients require lengthy periods dists for back pain, 50% have no explanation for their pain.
of time to recover muscle mass, even if remission is achieved Scheuermann’s disease, spondylolysis, disc prolapse, infec-
and physical therapy efforts are active. tion, and tumor are other causes of back pain in children
Pearl: Enthesitis can occur in sJIA. (Tumer 1989, Feldman 2000).

Comment: Gross swelling of tendon insertions can occur in Myth: The onset of sacroiliitis does not occur before the teen
sJIA in certain individuals. These swelling are sometimes years.
mistaken for arthritis or for muscle abscesses. Ultrasound
and MRI studies show amorphous material in the tendon Reality: Epidemiologic studies from areas in which pediatric
near the enthesis (Fig. 4.4) and aspiration is often difficult. spondylitis is prevalent indicate that inflammation of the
The common sites of enthesitis in sJIA are the bicipital ten- sacroiliac joints can occur in the pre-teen years (Huerta-Sil
dons near the shoulder and the gastrocnemius tendons behind et al. 2006). The use of contrast-enhanced MRI studies in
the knee. appropriate clinical settings has facilitated the diagnosis in
these cases (Fig. 4.5).
Myth: In a girl who has antinuclear antibodies and a knee
monoarthritis that has persisted for 6 months, polyarticular Myth: Children with the psoriatic arthritis subset of JIA are
disease will not occur. likely to develop axial arthritis.

a b

Fig. 4.4 (a) An 11-year-old boy with symptoms of left buttock pain T1-weighted sequence. (b) Repeat MRI after 1 year while on metho-
and bilateral hip (groin) pains. This sagittal view of an MRI study trexate treatment. There is bilateral involvement of the sacroiliac joints
shows enhancement of the left sacroiliac and hip joints on this
4 Juvenile Idiopathic Arthritis 31

In about half of all patients with leukemic arthritis, a sin-

gle joint – most often the knee – is involved. The finding of
an oligoarthritis is also common. The pain of leukemic arthri-
tis is more severe than JIA and tends to be focused over the
metaphysis rather than the joint itself. The ESR in leukemia
is normally elevated out of proportion to the small number of
effected joints. In contrast, oligoarticular JIA is usually asso-
ciated with a low or normal ESR.
In leukemia, low white blood cell and platelet counts are
the norm, in contrast to the leukocytosis and thrombocytosis
that are characteristic of JIA. A striking elevation of the
serum lactic dehydrogenase level should also make the clini-
cian suspect leukemia. A bone scan can distinguish between
malignancy and JIA when the musculoskeletal complaints
are generalized (Cabral and Tucker 1999).

References

Cabral DA, Tucker LB. Malignancies in children who initially present

with rheumatic complaints. J Pediatr 1999; 1345:53–57
Fig. 4.5 Enthesitis of the bicipital tendon and inflammation of the del- Cassidy J, Kivlin J, Lindsley C, Nocton J. Opthalmologic examinations
toid muscle in a patient with systemic-onset JIA in children with juvenile rheumatoid arthritis. Pediatrics 2006;
117:1843–1845
Ferucci ED, Majka DS, Parrish LA, Moroldo MB, Ryan M, Passo M,
Reality: In adults, sacroiliitis or spondylitis occurs in less Thompson SD, Deane KD, Rewers M, Arend WP, Glass DN, Norris
than 40% of patients with psoriatic arthritis (Lambert and JM, Holers VM. Arthritis Rheum. 2005 Jan;52(1):239–46
Huerta-Sil G, Casasola-Vargas JC, Londoño JD, et al Low grade radio-
Wright 1977). For children with psoriatic arthritis identified graphic sacroiliitis as prognostic factor in patients with undifferentiated
using the so-called “Vancouver criteria”, less than 5% develop spondyloarthritis fulfilling diagnostic criteria for ankylosing spon-
sacroiliac disease (Robertson et al. 1989; Stoll et al. 2006). dylitis throughout follow up. Ann Rheum Dis 2006; 65(5):642–646
Children identified by the JIA subcategory of psoriatic arthri- Jarvis JN, Petty HR, Tang Y, et al Evidence for chronic peripheral acti-
vation of neutrophils in polyarticular juvenile rheumatoid arthritis.
tis are even less likely to develop sacroiliitis and spondylitis, Arthritis Res Ther 2006; 8:R154
because of the associated exclusion criteria. Lambert JR, Wright V. Psoriatic spondylitis: a clinical and radiological
description of the spine in psoriatic arthritis. QJ Med 1977; 46(184):
Pearl: Pediatric psoriatic arthritis patients should be 411–425
screened for uveitis, just as patients with pauciarticular JIA Lequerre T, Quartier P, et al Interleukin-1 receptor anatagonist treat-
are screened for this complication. ment in patients with systemic-onset juvenile idiopathicarthiritis or
adult onset Still dieases: Preliminary experience in France. Ann
Comment: Many pediatric rheumatologists consider psori- Rheum Dis 2008; 67:281–282
atic arthritis to be an entity separate from the spondyloar- Lovell DJ. Update on treatment of arthriits in children: New treatment,
new goals. Bull NYU Hosp Jt Dis 2006; 64 (1–2):72–76
thropathies. This view is bolstered by the fact that children Petty RE, Southwood TR, Manners P, Baum J, Glass DN, Goldenberg
with psoriatic arthritis rarely develop painful iritis but are at J, He X, Maldonado-Cocco J, Orozco-Alcala J, Prieur AM, Suarez-
moderate risk of the development of asymptomatic eye Almazor ME, Woo P. International League of Associations for
inflammation. Pediatric patients with psoriatic arthritis Rheumatology. J Rheumatol 2004 Feb; 31(2):390–392
Prahalad S, Glass DN. A comprehensive review of the genetics of juve-
should be screened routinely for uveitis in the same manner nile idiopathic arthritis. Paedr Rheumatol Online J 2008; 6:11
in which patients with oligoarticular JIA are screened. Robertson DM, Cabral DA, Malleson PN, Petty RE. Juvenile psoriatic
arthritis: Follow-up and evaluation of diagnostic criteria. J Rheumatol
Myth: Leukemia is easily distinguished from JIA. 1989; 32:1007–1013
Rosenberg A, Oen K. The relationship between ocular and articular
Reality: Distinguishing leukemia from JIA and in particu- disease activity in children with JRA and associated uveitis.Arthritis
lar from sJIA can be a challenge. About 20–40% of chil- Rheum 1986; 29:797
dren with leukemia have arthritis at the presentation of their Stephen JL, Zeller J, Hubert PH, et al Macrophage Activation syndrome
disease. The hematologic findings can be normal in leuke- and rheumatic disease in childhood: A report of four cases. Clin Exp
Rheumatol 1993; 11:451–456
mia for weeks or months after onset of musculoskeletal Stoll ML, Zurakowski D, Nibrovic LE, et al Patients with juvenile pso-
symptoms, so repeated evaluations are needed and a bone riatic arthritis comprise two distinct populations. Arthritis Rheum
marrow examination is usually required. 2006; 54:3564–3572
32 P. H. White et al.

Woo, P. Systemic juvenile idiopathic arthritis. Nat Clin Pract Rheuma- Zhang K, Biroschak J, Glass DN, Thompson SD, Finkel T, Passo MH,
tology 2006; 2:28–34 Binstadt BA, Filipovich A, Grom AA. Macrophage activation syn-
Yokota S, Imagawa T, Mori M, et al Efficacy and safety of tocilizumab drome in patients with systemic juvenile idiopathic arthritis is asso-
in patients with systemic onset juvenile idiopathic arthritis: A ran- ciated with MUNC13-4 polymorphisms. Arthritis Rheum 2008;
domised double-blind, placebo-controlled, withdrawal phase III 58:2892–2896.
trial. Lancet 2008 Mar 22; 371(9617):998–1006
 Monogenic Autoinflammatory Syndromes
5
Marco Gattorno, Alberto Martini, Raphaela Goldbach-Mansky,
Pamela Aubert, and Polly J. Ferguson

5.1 Overview of the Monogenic › Most autoinflammatory syndromes can be differenti-

Autoinflammatory Syndromes ated on the basis of their clinical features (Table 5.2).

› “Autoinflammatory syndromes” is the broad name

given to a group of heritable conditions that initially
consisted of a group of diseases termed hereditary
periodic fever syndromes. Autoinflammatory syn-
drome is the preferred name for these conditions,
because not all of the disorders that have been added to 5.1.1 Familial Mediterranean Fever
this class of inflammatory disorde rs present with peri-
odic febrile episodes. • Familial Mediterranean fever (FMF) is an autosomal
› In contrast to autoimmune diseases such as systemic recessive disease.
lupus erythematosus and rheumatoid arthritis, autoanti- • Mutations in the MEFV gene, located on the short arm of
bodies and antigen-specific T cells do not play a role in chromosome 16, cause FMF.
the pathogenesis of the autoinflammatory syndromes. • MEFV gene encodes a protein named pyrin, which is
› Although the term autoinflammatory syndrome is now expressed in neutrophils, macrophages, and eosinophils
commonly used to include polygenic diseases such as but not lymphocytes.
Behcet’s syndrome, Still’s disease, and even Type 2 • More than 70 mutations in the MEFV gene have been
diabetes, this chapter addresses the monogenic autoin- described, but the six most prevalent mutations account
flammatory syndromes. for 80% of all FMF cases.
› These syndromes demonstrate a Mendelian inheritance • Painful serositis accompanied by fever is a hallmark of
pattern, and their genetic causes have been identified. the disease.
The disorders typically present in childhood. • The synovitis of FMF is intense but nondestructive, and
› The disorders are divided into different clinical syn- tends to involve only one joint.
dromes that include mixtures of clinical manifestations • Skin lesions mimic erysipelas or acute infectious celluli-
in the joints, the bones, the skin, the abdomen, and the tis and usually affect the lower extremities (Fig. 5.1).
lymph nodes. The seven currently identified monogenic • Secondary amyloidosis is a major long-term complication
autoinflammatory syndromes are shown in Table 5.1. of FMF that affects patients’ longevity. Colchicine use is
› Intermittent febrile episodes of noninfectious origin essential to prevent this complication.
can occur throughout life in patients with some autoin- • Proteinuria (>0.5 g/24 h) suggests amyloidosis.
flammatory syndromes. Febrile episodes, which last Myth: FMF is a disease of the Mediterranean people.
for variable periods according to the specific autoin-
flammatory syndrome, are accompanied by an intense Reality: The term Mediterranean is somewhat misleading.
acute phase response. Ethnic groups with the highest disease prevalence are of
› Most autoinflammatory syndromes are caused by Mediterranean ancestry, e.g., Armenians, Turks, Arabs, and
mutations that lead to a gain-of-function of the affected Jews (Sephardic > Ashkenazi). However, the disease is
innate immune pathway; for example, the genetic uncommon in Mediterranean populations such as Greeks,
mutations lead to inappropriately elevated levels of Italians, and Spanish. In fact, the clinical symptoms have
inflammatory cytokines, such as interleukin-1. been well documented to occur in a variety of ethnicities that
lack any Mediterranean background (Samuels et al. 1998).

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 33

DOI:10.1007/978-1-84800-934-9_5, © Springer Science + Business Media B.V. 2009
34 M. Gattorno et al.

Table 5.1 Epidemiology and genetic inheritance

Ethnic group predilection Mutation Protein Mode of inheritance
FMF Jewish, Armenian, Arab, Turkish, MEFV Pyrin/marenostrin Autosomal recessive
Italian (16p13)
TRAPS Any ethnic group TNFRS1A Tumor necrosis factor Autosomal dominant
(12p13) receptor 1
HIDS European (originally described MVK Mevalonate kinase Autosomal recessive
among Dutch children) (12q24)
CAPS
FCAS Mostly European CIAS1/NLRP3/NALP3/ Cryopyrin Autosomal dominant
PYPAF-1
(1q44)
Exon 3
MWS Northern European CIAS1/NLRP3/NALP3/ Cryopyrin Autosomal dominant
PYPAF-1
(1q44)
Exon 3
NOMID Any ethnic group CIAS1/NLRP3/NALP3/ Cryopyrin Autosomal dominant/de
PYPAF-1 novo
(1q44)
Exon 3
CRMO Any ethnic group (murine) PSTPIP2 18p (murine) PSTPIP2 Sporadic
Majeed’s syndrome Any ethnic group LPIN2 Lpin2 Autosomal recessive
(18p11)
PAPA Caucasian American CD2BP1 Pstpip1/CD2-binding Autosomal dominant
(15q24) protein 1
Cherubism

Pearl: FMF patients may have persistently elevated acute common (Kastner 2008). A sterile peritonitis is the most
phase reactants between attacks. common form of serositis in FMF.
Comment: Febrile attacks are associated invariably with Myth: A diagnosis of FMF requires the detection of two
marked elevations of acute phase proteins. Even during peri- mutations in the MEFV gene.
ods of clinical quiescence, however, many FMF patients Reality: Despite the elucidation of the genetic basis for FMF,
have persistent elevations of the erythrocyte sedimentation the diagnosis of FMF remains predicated upon clinical
rate (ESR) and C-reactive protein (CRP) and serum amyloid grounds (The International FMF Consortium 1997; The
A (SAA) levels (Lachmann et al. 2006). French FMF Consortium 1997). This does not minimize the
Myth: A negative family history of FMF excludes the importance of genetic testing but rather reflects the fact that
diagnosis. current mutation detection methods do not identify 100% of
the mutant alleles. The most striking example of this is in the
Reality: In countries with large families and high carrier rates, Arabic population, in which as many as half of the mutant
it is common to elicit a family history of FMF. However, only alleles are not identifiable in a cohort of well-characterized
50% presents a positive family history (Kastner 2008). FMF patients (El-Shanti et al. 2006).
Pearl: The arthritis of FMF can mimic a septic joint. Pearl: FMF may present as functional abdominal pain in
Comment: The arthritis of FMF often presents as an acute, childhood.
exquisitely painful, monoarticular arthritis associated with Comment: In the high risk ethnic populations, FMF should
fever, acute phase reactant elevations, and a neutrophil-rich be on the differential diagnosis of functional abdominal pain;
joint effusion (Kastner 2008). A sterile synovial fluid culture as many as twenty percent of Arabic and Jewish children
and the absence of crystals should lead one to consider FMF with functional abdominal pain are homozygous for MEFV
as a diagnostic possibility in this setting. mutations (Brik et al. 2001; El-Shanti et al. 2006).
Myth: Pericarditis is common form of serositis in FMF. Myth: Among Jewish people, FMF affects primarily non-
Ashkenazi Jews.
Reality: Although recurrent serositis is one of the hallmarks
of FMF, symptomatic pericarditis is relatively rare in FMF. Reality: FMF is an autosomal recessive disease with variable
However, asymptomatic pericardial effusions are relatively penetrance. The three major mutations in MEFV are M694V,
 5

Table 5.2 Cardinal clinical findings

Age at presentation Flare/fever Skin Ocular Musculo-skeletal Distinguishing features Treatment
pattern
FMF Eighty percent of the 1–3 days Erysipeloid Rare Arthralgia, oligoarthritis, Serosal inflammation, Colchicine, thalidomide, etaner-
cases occur before erythema of myalgia amyloidosis cept, anakinra
the age of 20 lower extremity
TRAPS Median age of onset 1–4 weeks Migratory rash Conjunctivitis Arthralgia, arthritis Serosal inflammation, Corticosteroids, anakinra,
3 years amyloidosis etanercept
Monogenic Autoinflammatory Syndromes

HIDS Median age of onset 3–7 days Rare Conjunctivitis, Arthralgia Some with elevated IgD, NSAIDs, anakinra, etanercept,
6 months episcleritis diarrhea, lymphadeno-pathy simvastatin
CAPS Interleukin-1 inhibition
FCAS First 6 months of life <24 h Cold-induced Conjunctivitis Arthralgia Cold-induced symptoms Cold avoidance, NSAIDs, anakinra,
urticaria-like rash rilonacept
MWS Infancy to 24–48 h Urticaria-like rash Conjunctivitis, Arthralgia Sensorineural hearing loss; anakinra, NSAIDs, rilonacept
adolescence episcleritis amyloidosis (30%)
NOMID Neonatum or Early Continuous Urticarial rash Conjunctivitis, Arthralgia, arthropathy Chronic aseptic meningitis, Anakinra, Corticosteroids for
infancy with flares episcleritis, secondary to osseous sensorineural hearing loss, severe CNS manifestations and
uveitis, overgrowth characteristic facies, acute hearing loss
papilledema amyloidosis (rare)
CRMO Childhood (as early as Uncommon Palmoplantar None Recurrent, multifocal Association with IBD NSAIDs, corticosteroids, bisphos-
<2 years as late as pustulosis, osteolytic sterile bone phonates, interferon a and g,
55years) psoriasis lesions azithromycin, etanercept
Majeed’s Early infancy Common Sweet’s syndrome, None Bone pain, osteitis, early Cultures negative, NSAIDs or corticosteroids. Other
syndrome (1–19 months) psoriasis in onset CRMO organomegaly, microcytic possible treatments include
carriers dyserythropoietic anemia bisphosphonates, interferon a and
g, sulfasalazine, azithromycin,
etanercept
PAPA Early childhood Weeks-months Cystic acne, None Sterile pyogenic arthritis Skin findings begin at Intra-articular and oral glucocorti-
pyoderma puberty coids, etanercept, anakinra
gangrenosum
Cherubism Early childhood Not applicable None Up turned eyes Marked expansion of the Typical facies with marked No effective treatment is known.
jaw bones enlargement of the jaw and Based on murine model and role
upturned eyes of osteoclasts, TNF inhibitors and
bisphosphonates may prove
useful
35
36 M. Gattorno et al.

Figs. 5.1 (a, b) The skin lesions

a
in familial Mediterranean fever
(FMF) can mimic erysipelas or
acute infectious cellulitis and
usually affect the lower
extremities

V726A, and E148Q. The M694V mutation, which is observed amyloidosis is proteinuria. A specific genotype, homozygos-
across a number of ethnic groups, is by far the predominant ity for the M694V mutation of the MEFV gene, is the most
mutation in North African (Sephardic) Jews. M694V relevant risk factor for the development of amyloidosis. This
homozygosity increases an individual’s risk for early age of genotype is found most commonly in Armenia, Israel, and
onset, more frequent attacks, and systemic amyloidosis. Arab countries (Touitou et al. 2007).
The carrier frequency of FMF in non-Ashkenazi (mostly Renal involvement is usually observed after a variable
Sephardic) Jews is estimated to be between 1:5 and 1:16. time from the onset of the disease (phenotype I). However, in
Several studies have found that approximately 1 in 5 some patients, renal amyloidosis is the first manifestation of
Ashkenazi Jews also carries mutations that can cause FMF, FMF (phenotype II). In an international, multicenter study of
the most common being E148Q and P369S (Aksentijevich FMF, amyloidosis was diagnosed in 73 of the 371 patients,
et al. 1999; Stoffman et al. 2000). The higher than expected and was the presenting clinical manifestation in seven
carrier frequency among the Ashkenazi Jewish population patients (Mimouni et al. 2000).
suggests that some mutations, particularly E148Q, are not
Myth: Long-term colchicine treatment affects fertility and
fully penetrant. Indeed, milder mutations with lower pene-
has mutagenic and teratogenic effects during pregnancy.
trance such as E148Q and V726A are more common than
M694V in Ashkenazi Jews compared with Sephardic Jews. Reality: Colchicine is believed to affect fertility because of
its potential to inhibit cell division. However, no clear rela-
Pearl: Renal amyloidosis may be the first manifestation of FMF.
tionship exists between female infertility and colchicine
Comment: Amyloid A (AA) type amyloidosis is the most therapy. Several reports have described women with FMF
severe long-term complication of FMF. This protein is a who were receiving colchicine but who were fertile (Kallinich
cleavage product of SAA, an acute phase reactant produced et al. 2007). With regard to males, although adverse effects
by the liver. The most common clinical manifestation of of colchicine on male fertility are alleged, sperm analyses in
5 Monogenic Autoinflammatory Syndromes 37

volunteers and patients with FMF who received therapeutic elimination pathways are potentially inhibited by
doses of colchicine were normal. cyclosporine. Other factors include reduced renal elimina-
Spontaneous abortions occurred in 12% of pregnant tion of colchicine caused by cyclosporine-induced reduc-
women with FMF who were treated with colchicine, com- tions in the glomerular filtration rate, but this appears to be of
pared with 20% of women in the same cohort who were not lesser importance when compared towith the effects of
treated (Kallinich et al. 2007). In theory, mutagenic and cyclosporine on the MDR-1 transport system. The net effect
teratogenic effects of colchicine are possible because the is an increase in the plasma concentration of colchicine and
drug crosses the placenta. However, there is no clear evi- a higher likelihood of toxic drug effects (Minetti et al.
dence for human teratogenicity or increased rate of aneu- 2003).
ploidy. In some reports, colchicine therapy during conception Most cases of gastrointestinal side effects and muscle
and the first trimester of pregnancy had no adverse effects on weakness caused by colchicine administration in the setting
the offspring (Kallinich et al. 2007). The current approach of cyclosporine use disappear within a few days of stopping
for most female patients with FMF is to continue colchicine cyclosporine (Cohen et al. 1989). However, the coadminis-
before conception and during pregnancy (Ben-Chetrit et al. tration of cyclosporine and colchicine should be avoided.
2003). This poses substantial challenges in the management of FMF
patients who undergo renal transplantation.
Pearl: FMF patients who are lactose intolerant may develop
dose-limiting side effects of colchicine. Pearl: FMF-associated amyloidosis rarely affects the heart,
the tongue, or the peripheral nerves.
Comment: Some FMF patients experience abdominal pain
and diarrhea as a side effect of chronic colchicine therapy. In Comment: Systemic AA amyloidosis is the most devastating
such patients, it is worth considering the possibility that the long-term complication of FMF. This complication results
symptoms are due to acquired lactose deficiency. A case– from the organ deposition of AA protein, a cleavage product
control study in FMF patients receiving colchicine confirmed of SAA that is produced by the liver. In systemic AA amyloi-
the experimental animal data that had suggested the gastroin- dosis, the kidneys, adrenal glands, intestine, and the spleen
testinal side effects of colchicine therapy to be often caused are most often involved. The AA protein is deposited only
by lactose intolerance induced by the medication (Fradkin rarely in the heart, the musculoskeletal system, and the ner-
et al. 1995). These patients should be managed with lactose vous system (Altiparmak et al. 2002; Yildiz et al. 2001;
avoidance or Lactaid in order to permit the continuation of Kavukçu et al. 1997).
colchicine therapy. The most common outcomes of systemic amyloidosis are
nephrotic syndrome and chronic renal failure. Most FMF
Pearl: Patients undergoing renal transplant who receive
patients who develop amyloidosis present such complica-
cyclosporine and colchicine can develop colchicine toxicity.
tions by the age of 40 years (Samuels et al. 1998). The preva-
Comment: FMF patients who receive renal allografts because lence of amyloidosis differs among various ethnic groups.
of amyloidosis-associated end-stage renal disease can Country of residence (foremost Armenia and Turkey),
develop serious colchicine toxicity if immunosuppressive homozygosity of the M694V mutation in MEFV, male gen-
therapy with cyclosporine is coadministered with colchicine. der, and the a/a genotype of serum amyloid A1 (SAA1) gene
The side effects are mild to moderate in many cases, consist- are the risk factors for the development of amyloidosis
ing of diarrhea and muscle weakness, but can result in mul- (Touitou et al. 2007).
tiple organ failure (Cohen et al. 1989; Minetti et al. 2003; Amyloidosis and chronic renal disease are less common
Eleftheriou et al. 2008). in Arabs with FMF than in other ethnic groups often affected
Colchicine is secreted into the bile as a major pathway of by the disease. The frequency of amyloidosis ranges from
elimination (Speeg et al. 1992). Renal excretion accounts for 0.4% in Jordanian patients to 2% in a mixed Arab population
10–20% of drug elimination. Colchicine toxicity is induced residing in Kuwait. In contrast, the frequency of amyloidosis
by cyclosporine by several mechanisms, the most important in Sephardic Jews, Armenians, and Turks has been estimated
of which is the modulation of the P-glycoprotein (also known to range from 24 to 60%. The low rate of occurrence of FMF-
as P-gp or gp-170), a 170-kDa cell surface glycoprotein related amyloidosis reported in Arabs is probably a result of
encoded by the MDR-1 (multidrug resistance) gene. the fact that these figures were obtained after the establish-
P-glycoprotein is expressed in enterocytes, renal tubular ment of colchicine as the standard of care (El-Shanti et al.
cells, hepatocytes, and some endothelial cells. Cyclosporine 2006).
inhibits P-glycoprotein expression, leading to increased Colchicine remains the mainstay of treatment for FMF. At
intracellular drug concentrations. low doses, colchicine is effective in controlling acute febrile
Both hepatic and renal colchicine secretion probably flares, while high doses are required for the prevention of
depend on the MDR-1 transport system. Thus, both the amyloidosis (Samuels et al. 1998).
38 M. Gattorno et al.

5.1.2 Tumor Necrosis Factor Receptor- pediatric age group. Attacks of intermittent fever with body
Associated Periodic Syndrome (TRAPS) temperatures above 38–39°C last from 1 to 3 weeks and are
interspersed by intervals of variable duration during which
the patient feels completely well. Febrile episodes are asso-
• TRAPS is inherited in an autosomal dominant fashion. ciated with acute phase reactant elevations, leukocytosis, and
• The median age of onset is 3. anemia.
• TRAPS is caused by mutations in TNFRFSF1A, located In adulthood, febrile episodes generally become less fre-
on chromosome 12p13. This gene encodes the main sur- quent and patients experience milder disease courses, char-
face receptor for TNF. acterized by intermittent abdominal pain, arthralgias,
• More than 50 mutations in this gene have been associated myalgias, ocular findings (e.g., conjunctivitis, uveitis), and
with TRAPS. The precise mechanism through which slight but persistent acute phase reactant elevations. SAA
these mutations lead to the clinical TRAPS syndrome is levels are usually elevated in TRAPS.
not known precisely. However, two hypotheses relate to Some patients carrying mutations in the TNFRSF1A gene
defective shedding of TNF receptors (TNFR) from the have been reported, who present with the typical clinical
cell membrane and misfolding of the TNFR. manifestations of TRAPS but without fever. These patients
• Febrile episodes in TRAPS are often of longer duration have had mutations that do not affect cysteine residues (Y20H
(up to several weeks) than those of other periodic fever and T50M). One patient with the T50M mutation developed
syndromes. severe renal AA amyloidosis in her second decade without
• Localized myalgia usually afflicts the limbs during peri- any preceding febrile episodes or other TRAPS-associated
ods of fever. This symptom, probably caused by a mono- clinical manifestations earlier in life (Kallinich et al. 2006).
cytic fasciitis, can be severe.
• Erythematous macules and patches are also found on the Myth: Low serum levels of the p55 TNFR1 during febrile
limbs during febrile episodes. attacks should be used as a screening test before proceeding
• Abdominal pain, usually less severe than that of FMF but with a molecular analysis of TNFRSF1A gene.
sometimes accompanied by vomiting, constipation, or Reality: Following cell activation, the extracellular portion
even bowel obstruction, occurs in almost all patients. of the p55 and p75 TNFR undergo metalloprotease-depen-
• Eye manifestations of TRAPS are diverse, and include: dent cleavage (shedding) from the cell membrane. Shedding
conjunctivitis, uveitis, iritis, and periorbital edema. of free TNFRs from the membrane produces a pool of solu-
ble receptors that may scavenge the circulating TNF by com-
Myth: TRAPS is a disorder of individuals of northern
peting with membrane-bound receptors. It has been suggested
European stock.
that some TNFRSF1A mutations may interfere with this pro-
Reality: TRAPS, formerly known as familial Hibernian fever, cess of TNFR shedding.
was described in 1982 in a kindred of Irish-Scottish descent Consistent with this hypothesis, is the finding that leuko-
(Williamson et al. 1982). TRAPS is caused by mutations in the cytes from TRAPS patients display a reduced cleavage rate
p55 TNF Receptor (TNFR1) and inherited in an autosomal of the p55 TNFR but not the p75 isoform following cell stim-
dominant fashion. TNFR1 is encoded by the TNF Super ulation (Aksentijevich et al. 2001; D’Osualdo et al. 2006).
Family Receptor 1A gene (TNFRSF1A) (McDermott et al. Impaired elevation of the soluble p55 TNFR during fever
1999). Fewer than 200 families have been reported worldwide, episodes has been adopted as a screening test in patients sus-
but this is likely an underestimate of the true number. pected of having TRAPS. Unfortunately, the potential utility
Till date, TRAPS has been reported mainly in the people of this assay is impaired substantially by the large variability
of northern European ancestry. However, the disease has in serum p55 TNFR levels during systemic inflammation,
been described in almost every ethnic group, including both in healthy individuals and in other inflammatory condi-
patients from Mediterranean countries (Italy, Spain), Africa tions. Moreover, recent observations have shown that some
(Mauritius), Central America (Puerto Rico, Mexico), and TNFRSF1A mutations are not associated with a defect in
Asia (India and Japan) (Aksentijevich et al. 2001; Aganna TNFR shedding (Huggins et al. 2004).
et al. 2003). The prevalence of TNFRSF1A mutations in
Pearl: Anti-IL-1 treatment controls fevers in TRAPS.
non-Ashkenazi Jews, Arabs, Turks and Armenian popula-
tions is yet to be tested. Comment: Fever episodes usually respond to glucocorti-
coids. However, due to the possible long duration of fever
Pearl: The absence of a clear history of recurrent fever
attacks and the tendency to a chronic course, patients may
attacks does not exclude the diagnosis of TRAPS.
become steroid-dependent.
Comment: TRAPS-associated mutations usually lead to Many other immunosuppressive medications appear to be
severe inflammatory episodes. Most patients present in the ineffective in reducing the frequency and the intensity of the
5 Monogenic Autoinflammatory Syndromes 39

inflammatory episodes and in preventing amyloidosis (Hull • Cutaneous findings in HIDS are common. These include ery-
et al. 2002). The observation of a defect in TNFRI shedding thematous maculopapular lesions, urticarial, and a variety of
led to the proposal of TNF inhibition as a potential treatment other rashes. Genital and oral ulcers occur in some patients.
strategy (Aksentijevich et al. 2001; Hull et al. 2002; Galon • Patients with HIDS have normal life expectancies, and
et al. 2000). However, incomplete treatment responses have both the frequency and the severity of the episodes tend to
been reported (Arostegui et al. 2005; Jacobelli et al. 2007). decrease after childhood.
An excellent short-term response to the treatment with the • Serum levels of immunoglobulin D are persistently ele-
recombinant receptor antagonist for IL-1 (IL-1Ra, Anakinra) vated in most patients with HIDS. However, levels of IgD
was observed in one TRAPS patient (Simon et al. 2004a) and do not correlate with disease activity.
supported by the findings, in a long-term, of the efficacy and
Pearl: HIDS is not limited to the people of Dutch ancestry.
safety of this approach in five patients with TRAPS (Gattorno
et al. 2008a). As described below, IL-1 blockade is also effi- Comment: Periodic fever associated with mevalonate kinase
cacious in some other monogenic autoinflammatory diseases. deficiency (MKD) was originally identified in 1984, in six
patients of Dutch ancestry who had long histories of recur-
Pearl: Low-penetrant TNFRFS1A mutations are usually
rent attacks of fever of unknown cause and high serum IgD
associated with mild disease courses.
levels (van der Meer et al. 1984). For this reason, this disor-
Comment: Mutations that result in cysteine substitutions der has also been named HIDS or Dutch fever.
demonstrate a higher penetrance of the clinical phenotype. After the identification of the molecular defect, it became
Patients with these mutations have severe disease courses and clear that the distribution of MKD is not limited to Dutch or
an increased probability of developing renal amyloidosis. In other northern European populations. Patients have been
contrast, low-penetrance mutations such as the R92Q and observed also among populations living around the
P46L mutations are usually associated with a more heteroge- Mediterranean basin (Italy) (D’Osualdo et al. 2001) and Asia
neous clinical presentation, a milder disease course, and a (Japan) (Drenth et al. 1994), and even in ethnic groups, in
lower incidence of amyloidosis (Aksentijevich et al. 2001). which there is a high prevalence of FMF (e.g., Turks and
In a recent study of a pediatric population, TRAPS patients Arabs) (Gattorno et al. 2008b; Demirkaya et al. 2007).
with missense substitutions of cysteine residues had more
Myth: HIDS and mevalonic aciduria have two completely
aggressive disease, compared with that of patients with other
distinctive phenotypes.
mutations (D’Osualdo et al. 2006). Cysteine residue muta-
tions were associated with febrile episodes of longer dura- Reality: HIDS is caused by mutations in the MVK gene,
tion (mean: 23 days), higher glucocorticoids requirement, encoded on chromosome 12q24. MVK is an essential enzyme
and a higher incidence of amyloidosis. Conversely, children in the isoprenoid biosynthesis pathway. The complete defi-
carrying a R92Q substitution had febrile episodes that lasted ciency of this enzyme causes a distinct syndrome called
for a mean of only 4.1 days and responded more quickly to mevalonic aciduria, which is characterized clinically by
glucocorticoids. severe mental retardation, ataxia, failure to thrive, myopathy,
cataracts, and recurrent fever attacks (Hoffman et al. 1993).
The most common mutation in MVK gene is the V377I vari-
ant, which is associated exclusively with the mild phenotype
5.1.3 Hyper Immunoglobulin D and Periodic
of HIDS and some residual MVK activity (Houten et al.
Fever Syndrome 1999). It is therefore clear that mevalonic aciduria and HIDS
associated with MKD represent the two extremities of a
• Hyper IgD syndrome (HIDS) is caused by mutations in broad clinical spectrum. Intermediate clinical scenarios are
the mevalonate kinase gene (MVK), which is located on sometimes observed (Simon et al. 2004b).
the long arm of chromosome 12 (12q24).
Pearl: Splenomegaly and gastrointestinal manifestations
• The mechanism by which reduced activity of MVK leads
during fever attacks are suggestive of HIDS.
to an autoinflammatory condition is not understood
completely. Comment: In a recent study, 228 consecutive children with a
• Patients with HIDS have recurrent fever attacks that last clinical history of periodic fever were screened for mutations
approximately 4 to 6 days. Attacks begin early in child- of the MVK, TNFRSF1A, and MEFV genes (Gattorno et al.
hood and often recur every 4 to 6 weeks. 2008b). Clinical variables such as a positive family history,
• Febrile attacks in HIDS generally begin with chills, fol- early age of onset, presence of abdominal and chest pain,
lowed by a swift and steep temperature elevation. diarrhea, and the absence of aphthosis correlated highly with
• Cervical adenopathy and abdominal pain, with vomiting the probability of relevant mutations in genes known to be
and diarrhea, occur during periods of HIDS activity. associated with periodic fever.
40 M. Gattorno et al.

Fever durations of less than 2 days were associated with Myth: High serum levels of IgD are the hallmark of HIDS.
FMF. In contrast, febrile periods longer than 7 days were
Reality: High IgD plasma levels (>100 UI/mL) during fever
characteristic of TRAPS. Among patients who presented
episodes and in basal conditions were used as a diagnostic
with fevers between 3 and 6 days duration, the differential
hallmark until mutations in the MVK gene were identified in
diagnosis usually focused on two entities: FMF and MKD.
1999. However, the sensitivity and the specificity of these
Splenomegaly and vomiting strongly suggest HIDS, even
findings are debated (D’Osualdo et al. 2005; Demirkaya
though these symptoms can be present in patients with FMF,
et al. 2007). In a recent series of 50 patients, the sensitivity of
as well. Figure 5.2 shows a flow chart to guide the genetic
a high IgD value for the diagnosis of HIDS was 79%
testing for children with recurrent fever of unknown origin
(Ammouri et al. 2007). In five patients with MVK mutations,
(Gattorno et al. 2008b). Additional information is available
IgD levels were found to be in the normal range. Elevated
at http://www.printo.it/periodicfever.
levels of IgD are also found in other inflammatory condi-
Myth: HIDS is a self-limited, benign disorder that has a good tions, including FMF.
long-term prognosis. An increased urine excretion of mevalonic acid is observed
during fever spikes, and decreased MVK activity may also
Reality: HIDS is essentially a pediatric disease. The onset of suggest the diagnosis. However, these determinations require
symptoms in HIDS occurs very early in life – usually in highly specialized laboratories, and their utility as screening
infancy. Almost all the patients become symptomatic during tests is therefore limited. Thus, the decision to undergo the
the first decade of life. Fever attacks have an abrupt onset and molecular analysis of MVK gene in a child with periodic
last for 4–6 days. The symptoms of HIDS persist for years fever is usually taken on clinical grounds.
but may become less prominent over time. Nevertheless, a
study from the international HIDS study group indicated that
more than 12 attacks per year still occurred in 18% of the
5.1.4 Cryopyrin-Associated Periodic
patients who were older than 20 years, compared with 24%
of those between the ages of 10 and 20 (van der Hilst et al. Syndromes
2008). Amyloidosis has been reported in a small number of
patients with MKD (D’Osualdo et al. 2005; Siewert et al. • The cryopyrin-associated periodic syndromes (CAPS)
2006). include three overlapping conditions.

Patient with periodic fever

(exclude other causes) Sensitivity to cold Exclude
exposure?
Urticarial rash?
NALPs-related
Hearing loss? diseases
Diagnostic score

High risk Low risk

Go for genetic testing

Follow-up
Ethnicity & positve No (6-12 months)
FMF Criteria
Duration of episodes

Yes 52 days 27 days

3-6 days Persistence Resolution or
(new symptoms?) improvement

No Vomiting TNFRSF1A
MEFV
Splenomegaly Yes
Follow-up or Follow-up
Fig. 5.2 Flow chart for the No Yes go for genetic (6-12 months)
genetic testing of children with MVK testing
recurrent fevers of unknown
origin If negative test
5 Monogenic Autoinflammatory Syndromes 41

• The familial cold autoinflammatory syndrome (FCAS). also occur (Aksentijevich et al. 2007; Goldbach-Mansky et al.
• The Muckle-Wells syndrome (MWS). 2008). The NOMID/CINCA syndrome occurs almost exclu-
• The neonatal onset multisystem inflammatory disorder, sively as a sporadic mutation, because most untreated patients
which is also known as the chronic infantile neurologic, with NOMID/CINCA are not able to reproduce.
cutaneous, and articular syndrome (NOMID/CINCA).
Pearl: Never stop treatment with anakinra in patients with
• These three disorders represent a spectrum of disease sever-
CAPS, even during infections or surgery.
ity with overlapping clinical features. The clinical features
found in each of these entities are fever, an urticaria-like Comment: The pivotal role of IL-1 oversecretion in CAPS is
skin rash, and varying degrees of arthralgia and arthritis. confirmed by the marked response to interleukin-1 (IL-1)
• The CAPS disorders also have similar laboratory profiles, blockade in patients with CAPS. Treatment with daily injec-
characterized by neutrophil-mediated inflammation and tions of anakinra, a recombinant antagonist of the human IL-1
elevated acute-phase reactants (Stojanov et al. 2005; receptor, results in markedly improved clinical and laboratory
Aksentijevich et al. 2007). manifestations in patients with FCAS, MWS, and NOMID/
• All patients present in childhood, usually in the perinatal CINCA. Withdrawal leads to recurrence of symptoms (Hawkins
period. et al. 2004; Hoffman et al. 2004; Goldbach-Mansky et al. 2006).
In a prospective withdrawal trial of anakinra in 18 patients with
Pearl: The determination of the precise clinical phenotype NOMID, the discontinuation of anakinra uniformly resulted in
has important prognostic implications in this spectrum of disease relapses, sometimes within hours of stopping the medi-
disease. cation (Goldbach-Mansky et al. 2006). Therefore, anakinra
should not be discontinued, even during an infection.
Comment: The FCAS has the mildest phenotype and the best Controlled data on the development of infections on IL-1
prognosis; MWS has an intermediate phenotype and progno- blockade are derived from a recent placebo-controlled trial
sis; and NOMID has the worst of both. In addition to the in patients with FCAS and MWS using the long acting IL-1
symptoms that are commonly shared by them, these three inhibitor called rilonacept. Rilonocept is an IL-1 trap, a sol-
disorders have distinguishing characteristics: uble receptor molecule that binds IL-1 and prevents its inter-
• As its name implies, the FCAS is characterized by attacks action with the cell surface IL-1 receptor. In this study, 18%
that are precipitated by exposure to cold and or drafts of of patients on IL-1 inhibition had infections, compared with
air. Hearing loss and amyloidosis are rare but are reported 22% in the placebo group (Hoffman et al. 2008). Unpublished
complications of FCAS (Thornton et al. 2007). data suggest that infections heal as expected in patients on
• Symptoms of the MWS seldom have any clear association continuous IL-1 inhibition, and that the risk of infection dur-
with cold temperatures, but most develop sensorineural hear- ing IL-1 blockade does not appear to be increased signifi-
ing loss during their teens or twenties. Deafness is a common cantly over the risk before treatment.
complication of MWS, and the prevalence of secondary AA Myth: The rash in CAPS is urticarial.
amyloidosis reaches 30% among untreated patients.
• In the NOMID/CINCA syndrome, inflammation-induced Reality: The rash in CAPS is often mistaken for “classic urti-
organ damage occurs early in life. Central nervous system caria.” In FCAS, the rash is induced by cold, usually within
inflammation with aseptic meningitis and cognitive the first 6 months of life. In the MWS and NOMID/CINCA
impairment are common in NOMID/CINCA. Increased syndromes, the rash is typically present at birth and does not
intracranial pressure leads to frequent headaches and require cold for induction. The rash of CAPS is difficult to
papilledema. Progressive optic nerve atrophy and sen- distinguish clinically from an urticarial, allergic rash.
sorineural hearing loss usually occurs in the first decade The histopathological features of CAPS are consistent with
of life. Joint and bone abnormalities include patellar bony a neutrophilic dermatosis (Fig. 5.3), in contrast to the typical
enlargement, joint deformity, and severe contractures lymphocytic or eosinophilic infiltrate of classic urticaria
(Hill et al. 2007). (Leslie et al. 2006). The epidermis is normal in CAPS, but
there is a superficial and a deep perivascular and interstitial
cellular infiltrate predominantly composed of neutrophils,
Pearl: FCAS, MWS, and NOMID/CINCA are all caused by
some lymphocytes, and rare eosinophils (Prieur 2001; Hoffman
mutations in CIAS1, which encodes the protein cryopyrin.
et al. 2001b; Shinkai et al. 2008; Kilcline et al. 2005).
Comment: These mutations are autosomal dominant. Confusion frequently exists in distinguishing the skin rash
Mutations in CIAS1 lead to the oversecretion of interleukin-1 of FCAS from that of cold urticaria, a common form of physi-
(IL-1) in vitro (Agostini et al. 2004). The FCAS and MWS cal urticaria that can be primary (idiopathic) or secondary
occur mostly as familial diseases with the presence of founder (acquired). True cold urticaria is nonfamilial and is not caused
mutations in FCAS (Hoffman et al. 2003). Sporadic cases by mutations in CIAS1. The diagnosis can be established
42 M. Gattorno et al.

a to 40% of the patients are negative for CIAS1 mutations

(Aksentijevich et al. 2002).
Patients without CIAS1 mutations detectable by routine
analysis may have somatic mutations (Saito et al. 2008).
Regardless of their mutation status, patients of the entire
CAPS spectrum respond equally well to IL-1 blocking ther-
apy (Hawkins et al. 2004; Hoffman et al. 2004; Goldbach-
Mansky et al. 2006; Gattorno et al. 2007).
Pearl: Bony deformities, which result from derangements
of the growth plate, are unique to the NOMID/CINCA
syndrome.
Comment: Joint pain and swelling occur occasionally during
febrile exacerbations of all three CAPS syndromes, but is
particularly characteristic of the NOMID/CINCA syndrome.
b In CAPS, synovitis can present with warm, swollen, stiff
joints and increased synovial fluid production, reminiscent
of juvenile idiopathic arthritis (Prieur et al. 1987; Kilcline
et al. 2005). However, the joint manifestations of the FCAS
and MWS are usually limited to arthralgias and limb pain
and synovitis, if present, is mild. In contrast, joint involve-
ment in NOMID can be much more severe and is among the
diagnostic criteria for the disease.
The onset of arthropathy in NOMID/CINCA is usually
within the first year of life. The knees are the most com-
monly affected joints. Other sites of involvement, by order of
decreasing frequency, include the ankles, the elbows, the
wrists, the hips, and the small joints. The axial joints are not
affected in NOMID/CINCA (Yarom et al. 1985; Prieur 2001;
Stankovic et al. 2007).
Fig. 5.3 (a, b) Neutrophilic dermatosis in a patient with a cryopyrin- The musculoskeletal changes in NOMID/CINCA are
associated periodic syndrome (CAPS) characterized by endochondral bony overgrowth and prema-
ture ossification that originate in the growth plate. These
lesions lead to deformities in the adjacent metaphysis and
with the use of an ice cube test or an ice water immersion test, epiphysis (Hill et al. 2007). Premature patellar and epiphy-
in which a wheal reaction appears on the exposed skin seal long bone ossification and resultant osseous overgrowth
10–20 min after exposure (Neittaanmäki et al. 1985). The ice often develop early in life and lead to severe contractures,
cube test is negative in patients with FCAS, MWS or NOMID/ deforming arthropathy, and long-term disability (Fig. 5.4)
CINCA (Delorme et al. 2002; Hoffman et al. 2001a). (Prieur et al. 1987; Yarom et al. 1985; Kilcline et al. 2005).
Pearl: There is a subset of patients with classical CAPS who Pearl: Hearing loss in CAPS, caused by cochlear inflamma-
are negative for mutations in CIAS1 but respond similarly to tion, responds to IL-1 inhibition.
IL-1 blockade as patients who are CIAS1 mutation positive.
Comment: Central nervous system manifestations are among
Comment: All the three CAPS syndromes are caused by the most devastating sequelae of CAPS. They tend to occur
autosomal dominant mutations in exon 3 of the CIAS1 gene in the more severe phenotypes. Sensorineural hearing loss
on chromosome 1q44 (Cuisset et al. 1999; Hoffman et al. develops in patients with MWS in the second to fourth decade
2001a, b; Feldman 2002; Aksentijevich et al. 2002). Although of life. NOMID patients show signs of hearing loss in the
there is some genotype/phenotype correlation observed first decade of life. The sensorineural hearing loss is progres-
within the spectrum of these diseases, there also appears to sive and can lead to complete hearing loss.
be the role of background genes and environmental factors in The presence of cochlear enhancement on FLAIR sequences
controlling disease expression (Hawkins et al. 2004). In of magnetic resonance imaging in almost all NOMID patients
addition, not all patients with unequivocal CAPS have detect- suggests that hearing loss is caused by cochlear inflammation.
able mutations in CIAS1 (Neven et al. 2004; Aksentijevich Treatment with anakinra for 6 months led to a decrease and
et al. 2007). This is especially true for NOMID, in which up even elimination of the cochlear enhancement within 3 months
5 Monogenic Autoinflammatory Syndromes 43

• Majeed’s syndrome is a syndromic form of CRMO caused

by an autosomal recessive alteration in the LPIN2 gene on
the short arm of chromosome 18.
• Majeed’s syndrome is characterized by an early onset of
CRMO, congenital dyserythropoietic anemia (CDA), and
inflammatory neutrophilic dermatosis.
• Although mutations in the PSTPIP2 gene mapped to chro-
mosome 18 cause chronic multifocal osteomyelitis in a
murine model, the genetic cause for the more common
presentations of CRMO in humans has not yet been
discovered.
Myth: CRMO is a benign, self-limited disorder with little
long-term morbidity.
Reality: CRMO has been described as a benign, self-limited
disorder, yet this conclusion is based upon data from short-
term follow-up. Estimates of the likelihood of active disease at
long-term follow-up have varied widely. The incidence of
permanent bone deformity leading to leg length discrepancies
is high (Duffy et al. 2002). CRMO has also been reported to
evolve into spondyloarthropathy. The outcome for children
with an early onset of CRMO (onset <2 years of age) is more
guarded; they tend to have severe disease often associated with
failure to thrive, secondary joint contractures, and the need for
long-term immune suppression.
Pearl: CDA that accompanies the Majeed syndrome is typi-
cally microcytic and may be mild.
Comment: The Majeed syndrome is an autosomal recessive
disorder caused by mutations in LPIN2. Classic cases of the
Fig. 5.4 The arthropathy of NOMID/CINCA Majeed syndrome present with the triad of early-onset
CRMO, CDA, and a neutrophilic dermatosis (Majeed et al.
1989, 2000, 2001; Al-Mosawi et al. 2007; Ferguson et al.
of treatment (Goldbach-Mansky et al. 2006). One third of the 2005). However, skin involvement may be absent and the
patients showed improved hearing on audiometry and the anemia is mild, posing substantial challenges in the diagno-
remainder showed stability in their hearing deficits. Additional sis of the Majeed syndrome. One helpful clue is that the ane-
case reports in NOMID and MWS patients confirmed improved mia of the Majeed syndrome is microcytic. This feature
hearing by audiometry after treatment with anakinra (Yamazaki contrasts sharply with other well-described CDA, which are
et al. 2008; Rynne et al. 2006). either macrocytic or normocytic (Majeed et al. 1989, 2000,
2001; Al-Mosawi et al. 2007).
Bone marrow aspiration should be included in the work-up
of all children who present with CRMO before the age of two.
5.1.5 Chronic Recurrent Multifocal The Majeed syndrome should be suspected if CDA is present.
Osteomyelitis This in turn, leads to the evaluation of mutations in LPIN2.
Pearl: Inflammatory disorders of the skin and the gut are
• Chronic recurrent multifocal osteomyelitis (CRMO) is
frequently present in individuals with CRMO.
characterized by chronic recurrent multifocal sterile
osteomyelitis that commonly affects the metaphysis of Comment: In the late 1970s, several reports noted the fre-
long bones, but that can be found in many other sites quent cooccurrence of CRMO and pustulosis palmoplanta-
including the clavicle, spine, and the pelvic bones. ris. The link between psoriasis and CRMO was recognized in
• Patients with CRMO can have both symptomatic and the late 1980s. Since then, multiple inflammatory skin disor-
asymptomatic bone lesions. ders have been associated with CRMO including generalized
• An association exists between inflammatory skin condi- pustulosis, pyoderma gangrenosum, severe acne, Sweet’s
tions, inflammatory bowel disease, and CRMO. syndrome, and Crohn's disease.
44 M. Gattorno et al.

Myth: In CRMO, tubular long bone involvement is confined 2000). One explanation is that CRMO is indeed an infectious
to the metaphysis. disease due to an indolent infectious agent and that azithro-
mycin is helping through its antimicrobial properties.
Reality: Involvement of the metaphyses of the long bones is
However, microbial signatures have not been identified in
the most common site for CRMO lesions. Diaphyseal involve-
bone biopsies taken from active CRMO lesions (Girschick
ment is typically in the area immediately adjacent to active
2007). Another plausible explanation is that azithromycin
metaphyseal lesions. The presence of isolated diaphyseal
exerts its benefit through its reported anti-inflammatory
lesions should prompt a second look for an alternative diagno-
properties (Labro 2004). Additional studies are needed to
sis. In addition, a substantial number of children with CRMO
determine the role of azithromycin in treating CRMO.
never have long bone involvement but have lesions in other
sites including the clavicles, vertebral bodies, pelvis, ribs, feet, Pearl: The clavicle is a common site of involvement in
sternum, jaw, hands, skull, and scapula. CRMO and SAPHO syndrome, but a rare site for infectious
osteomyelitis.
Myth: CRMO bone lesions are always symptomatic.
Comment: Clinically, CRMO and infectious osteomyelitis
Reality: One common mistake made in working up children
can be very difficult to differentiate. In many cases, culture
with a single symptomatic sterile osteolytic lesion is the fail-
results are unreliable because of empiric antibiotic use before
ure to look for asymptomatic disease at other sites. A reason-
a bone biopsy is obtained. However, clavicular involvement
able algorithm for imaging in this disorder is as follows:
may be a clue that it is inflammatory rather than infectious.
1. Plain radiographs of symptomatic lesions. Clavicular involvement, rare in infectious osteomyelitis, is
2. Bone scan to identify asymptomatic lesions. relatively common in CRMO (Girschick et al. 1998).
3. Radiography for all lesions seen on bone scan. Therefore, involvement of the clavicle should raise the suspi-
4. Magnetic resonance imaging for any lesions that need addi- cion for a noninfectious process such as CRMO or SAPHO.
tional documentation of the disease extent and for lesions
seen on bone scan that are not detectable on plain film.
5.1.6 PAPA
Pearl: Primary intraosseous lymphoma can mimic CRMO.
Comment: Given that glucocorticoids are a reasonable treat- • PAPA syndrome is caused by mutations in the PSTPIP1
ment option for CRMO, one must be very careful not to miss gene mapped to chromosome 15q24 and is inherited in an
an unusual presentation of lymphoma. Primary multifocal autosomal dominant pattern.
osseous lymphoma is challenging to diagnose because the • The most common symptom of PAPA is pyogenic sterile
lymphoma occurs in the bone without the evidence of vis- arthritis; other symptoms include pyoderma gangrenosum
ceral or lymph node involvement (Sato et al. 2008). The and cystic acne.
osseous lesions are osteolytic, multifocal, and often involve • The arthritis usually presents during childhood with recur-
the long bones. Common sites of bony involvement in pri- rent inflammatory episodes resembling septic arthritis but
mary osseous lymphoma include the femur, tibia, pelvis, and with sterile cultures.
the spine (Sato et al. 2008). These are also common sites of
Pearl: PAPA is an acronym for pyogenic arthritis, pyoderma
lesions in CRMO. Clues to a more sinister diagnosis of lym-
gangrenosum, and acne.
phoma include cortical erosion and the involvement of the
diaphysis of the long bones. Comment: PAPA syndrome, which demonstrates an auto-
Glucocorticoids should not be administered before lym- somal dominant pattern of inheritance, is caused by missense
phoma has been excluded with certainty. Inappropriate glu- mutations in the PSTPIP1 gene (chromosome 15q24). Four
cocorticoid use can affect survival in lymphoma. When in different mutations have been identified in four Caucasian
doubt, re-biopsy should be done. US families and one Spanish family. The syndrome is char-
acterized by skin lesions and polyarticular arthritis. Skin
Myth: Antibiotics have no role in treating CRMO.
manifestations include severe cystic acne, pyoderma gan-
Reality: Despite the current dogma that bone lesions in grenosum and sterile abscess formation at injection sites.
CRMO are sterile, symptomatic improvement associated The arthritis, which involves peripheral joints, presents with
with azithromycin therapy has been reported (Schilling et al. recurrent sterile effusions that contain high neutrophil counts.
5 Monogenic Autoinflammatory Syndromes 45

Periods of arthritis fluctuate throughout the patient’s lifetime.

Chronic recurrences can lead to joint erosion.
Mutations in PSTPIP1 lead to gains of function. Two in
vitro models link the mutations that cause PAPA to the exces-
sive production of IL-1-beta, thereby suggesting that IL-1
inhibition might be an effective treatment strategy (Shoham
et al. 2003; Yu et al. 2007). The in vivo function of PSTPIP1
requires further exploration. Treatment for PAPA includes
intra-articular and systemic glucocorticoids (Tallon et al.
2006). Case reports show clinical improvement and a
decrease in inflammatory markers with either tumor necrosis
factor inhibition or anakinra (Cortis et al. 2004; Stichweh
et al. 2005; Dierselhuis et al. 2005).

5.1.7 Cherubism

• Cherubism presents with marked, symmetric expansion

of the jaw bones resulting in a distinctive chubby cheeked
appearance occasionally accompanied by upturned eyes.
• Age of onset is usually between 3 and 5 years of age.
• Bone involvement in cherubism is nearly always confined
to the jaw.
• Biopsy of the involved bone reveals fibroblastic stroma
and abundant osteoclasts.
• Severe dental problems, including loss of teeth and mal-
occlusion, occur.
• Cherubism is an autosomal dominant disorder. Most cases
are associated with mutations in SH3BP2.
Fig. 5.5 Typical osteolytic lesions in a child with chronic recurrent
Myth: Cherubism is a form of fibro-osseous dysplasia. multifocal osteomyelitis (CRMO). Multiple rounded lacunar shaped
osteolytic lesions surrounded by a rim of sclerosis are seen immediately
Reality: Cherubism is an autosomal dominant disorder with adjacent to the growth plate in the distal fibula and tibia. Additional
incomplete penetrance that presents with painless jaw swell- osteolytic lesions are seen in the dome of the talus and calcaneus.
ing (Fig. 5.5) (OMIM reference number 118400). Multiple Diffuse osteopenia is present
large osteolytic lesions are seen on radiographs (Fig. 5.6).
Bone biopsy reveals dense fibrous connective tissue inter-
spersed with abundant osteoclasts. Mutations in exon 9 of
the SH3-binding protein 2 gene are present in majority of the Pearl: Individuals with cherubism may have extra-gnathic
cases (Ueki et al. 2001). bone lesions.
Until recently, cherubism was viewed as a form of bone Comment: The vast majority of patients with cherubism have
dysplasia with limited treatment options. However, Ueki lesions confined to the jaw. However, several reports have
et al. used a knock-in mouse model to demonstrate that detailed extra-gnathic bone involvement, including involve-
cherubism is an inflammatory disorder driven by the innate ment of the jaw with the rib, humerus, femur, or tibial dis-
immune system, which is dependent in the presence of tumor ease (Ozkan 2003).
necrosis factor (Ueki et al. 2007). Consequently, cherubism
is now considered as an autoinflammatory bone disorder Acknowledgement We would like to thank Dr. Daniel Kastner for
(Ferguson et al. 2007; Novack et al. 2007; Ueki et al. 2007). sharing his expertise and suggestions for the myths and pearls in the
TNF inhibitors are the main approach to treatment. FMF section.
46 M. Gattorno et al.

a c

Fig. 5.6 (a) Chubby cheek appearance due to mandibular enlarge- in the mandible of a female with cherubism (From Jain and Sharma
ment in a boy with cherubism (From Meng et al. (2005). Reprinted (2006). Reprinted with permission from Springer Science + Business
with permission from Elsevier). (b) Expansion of the mandible in Media). (d) Histologic findings in cherubism of multinucleated giant
cherubism with resultant dental abnormalities (From Meng et al. cells in collagen stromal tissue (From Meng et al. (2005). Reprinted
(2005). Reprinted with permission from Elsevier). (c) This plain film with permission from Elsevier)
reveals multiple large radiolucent lesions with surrounding sclerosis

References Aksentijevich I, Galon J, Soares M, et al The tumor-necrosis-factor

receptor-associated periodic syndrome: new mutations in
TNFRSF1A, ancestral origins, genotype-phenotype studies, and
Aganna E, Hammond L, Hawkins PN, et al Heterogeneity among evidence for further genetic heterogeneity of periodic fevers. Am J
patients with tumor necrosis factor receptor-associated periodic Hum Genet. 2001;69(2):301–14
syndrome phenotypes. Arthritis Rheum. 2003;48(9):2632–44 Aksentijevich I, Nowak M, Mallah M, et al De novo CIAS1 mutations,
Agostini L, Martinon F, Burns K, McDermott MF, Hawkins PN, cytokine activation, and evidence for geneticheterogeneity in patients
Tschopp J. NALP3 forms an IL-1beta-processing inflammasome with neonatal-onset multisystem inflammatory disease(NOMID): a
with increased activity in Muckle-Wells autoinflammatory disorder. new member of the expanding family of pyrin-associated autoin-
Immunity. 2004;20(3):319–25 flammatory diseases. Arthritis Rheum. 2002 46(12):3340–8
5 Monogenic Autoinflammatory Syndromes 47

Aksentijevich I, Putnam C, Remmers EF, et al The clinical continuum El-Shanti H, Majeed HA, El-Khateeb M. Familial mediterranean fever
of cryopyrinopathies: novel CIAS1 mutations in North American in Arabs. Lancet. 2006;367:1016–24
patients and a new cryopyrin model. Arthritis Rheum. 2007;56(4): Ferguson PJ, Chen S, Tayeh MK, et al Homozygous mutations in LPIN2
1273–85 are responsible for the syndrome of chronic recurrent multifocal
Aksentijevich I, Torosyan Y, Samuels J, et al Mutation and haplotype osteomyelitis and congenital dyserythropoietic anemia (Majeed
studies of familial Mediterranean fever reveal new ancestral rela- Syndrome). J Med Genet. 2005;42:551–7
tionships and evidence for a high carrier frequency with reduced Ferguson PJ, El-Shanti HI. Autoinflammatory bone disorders. Curr
penetrance in the Ashkenazi Jewish population. Am J Hum Genet. Opin Immunol. 2007;19:492–8
1999;64(4):949–62 Fradkin A, Yahav J, Zemer D, et al Colchicine-induced lactose malab-
Al-Mosawi ZS, Al-Saad KK, Ijadi-Maghsoodi R, et al A splice site sorption in patients with familial Mediterranean fever. Isr J Med Sci.
mutation confirms the role of LPIN2 in Majeed syndrome. Arthritis 1995;31:(10): 616–20
Rheum. 2007;56:960–4 Galon J, Aksentijevich I, McDermott MF, et al TNF receptor-associated
Altiparmak MR, Pamuk ON, Pamuk GE, et al Amyloid goitre in famil- periodic syndromes (TRAPS): Mutations in TNFR1 and early expe-
ial Mediterranean fever: report on three patients and review of the rience with Etanercept therapy. Faseb Journal. 2000;14(6): A1150
literature. Clin Rheumatol. 2002;21(6):497–500 Gattorno M, Tassi S, Carta S, et al Pattern of interleukin-1beta secretion
Ammouri W, Cuisset L, Rouaghe S, et al Diagnostic value of serum in response to lipopolysaccharide and ATP before and after interleu-
immunoglobulinaemia D level in patients with a clinical suspicion kin-1 blockade in patients with CIAS1 mutations. Arthritis Rheum.
of hyper IgD syndrome. Rheumatology. 2007;46(10):1597–600 2007;56(9):3138–48
Arostegui JI, Solis P, Aldea A, et al Etanercept plus colchicine treatment Gattorno M, Pelagatti MA, Meini A, et al Persistent efficacy of anakinra
in a child with tumour necrosis factor receptor-associated periodic in patients with tumor necrosis factor receptor-associated periodic
syndrome abolishes auto-inflammatory episodes without normalis- syndrome. Arthritis Rheum. 2008a;58(5):1516–20
ing the subclinical acute phase response. Eur J Pediatr. Gattorno M, Sormani MP, D’Osualdo A, et al A diagnostic score for
2005;164(1):13–6 molecular analysis of hereditary autoinflammatory syndromes with
Ben-Chetrit E, Levy M. Reproductive system in familial Mediterranean periodic fever in children. Arthritis Rheum. 2008b;58(6):1823–32
fever: an overview. Ann Rheum Dis. 2003;62(10):916–9 Girschick HJ, Krauspe R, Tschammler A, et al Chronic recurrent osteo-
Brik R, Litmanovitz D, Berkowitz D, et al Incidence of familial myelitis with clavicular involvement in children: diagnostic value of
Mediterranean fever (FMF) mutations among children of different imaging techniques and therapy with non-steroidal anti-
Mediterranean extraction with functional abdominal pain. J Pediatr. inflammatory drugs. Eur J Pediatr. 1998;157:28–33
2001;138:759–62 Goldbach-Mansky R, Dailey NJ, Canna SW, et al Neonatal-onset mul-
Cohen SL, Boner G, Shmueli D, et al Cyclosporin: poorly tolerated in tisystem inflammatory disease responsive to interleukin-1beta inhi-
familial Mediterranean fever. Nephrol Dial Transplant. 1989; bition. N Engl J Med. 2006;355(6):581–92
4(3):201–4 Goldbach-Mansky R, Shroff SD, Wilson M, et al A pilot study to evalu-
Cortis E, DeBenedetti F, Insalaco A, Cioschi S, Muratori F, ate the safety and efficacy of the long-acting interleukin-1 inhibitor
D’Urbano LE, Uqazio AG. Abnormal production of tumor necrosis rilonacept (interleukin-1 Trap) in patients with familial cold autoin-
factor (TNF) – alpha and clinical efficacy of the TNF inhibitor etan- flammatory syndrome. Arthritis Rheum. 2008;58(8):2432–42
ercept in a patient with PAPA syndrome [corrected]. J Pediatr. Hawkins PN, Lachmann HJ, Aganna E, et al Spectrum of clinical fea-
2004;145(6): 851–5 tures in Muckle-Wells syndrome and response to anakinra. Arthritis
Cuisset L, Drenth JP, Berthelot JM, et al Genetic linkage of the Muckle- Rheum. 2004;50(2):607–12
Wells syndrome to chromosome 1q44. Am J Hum Genet. 1999;65(4): Hill SC, Namde M, Dwyer A, Poznanski A, et al Arthropathy of neona-
1054–9 tal onset multisystem inflammatory disease (NOMID/CINCA).
Delorme N, Drouet M, Thibaudeau A, et al Urticaire au froid. Allergie Pediatr Radiol. 2007;37(2):145–52
Et Immunologie. 2002;34(7):255–8 Hoffman HM, Gregory SG, Mueller JL, Tresierras M, Broide DH,
Demirkaya E, Caglar MK, Waterham HR, et al A patient with hyper- Wanderer AA, Kolodner RD. Fine structure mapping of CIAS1:
IgD syndrome responding to anti-TNF treatment. Clin Rheumatol. identification of an ancestral haplotype and a common FCAS muta-
2007;26(10):1757–9 tion, L353P. Hum Genet. 2003;112(2):209–216
Dierselhuis MP, Frenkel J, Wulfraat NM, et al Anakinra for flares of Hoffman HM, Mueller JL, Broide DH, et al Mutation of a new gene
pyogenic arthritis in PAPA syndrome. Rheumatology. 2005;44: encoding a putative pyrin-like protein causes familial cold
406–8 autoinflammatory syndrome and Muckle-Wells syndrome. Nat
D’Osualdo A, Ferlito F, Prigione I, et al Neutrophils from patients with Genet. 2001a;29(3):301–5
TNFRSF1A mutations display resistance to tumor necrosis factor- Hoffman HM, Rosengren S, Boyle DL, et al Prevention of cold-associated
induced apoptosis – pathogenetic and clinical implications. Arthritis acute inflammation in familial cold autoinflammatory syndrome by
Rheum. 2006;54(3):998–1008 interleukin-1 receptor antagonist. Lancet. 2004;364(9447):
D’Osualdo A, Picco P, Caroli F, et al MVK mutations and associated 1779–85
clinical features in Italian patients affected with autoinflammatory Hoffman HM, Throne ML, Amar NJ, et al Efficacy and safety of
disorders and recurrent fever. Eur J Hum Genet. 2005;13(3): rilonacept (interleukin-1 Trap) in patients with cryopyrin-associated
314–20 periodic syndromes: results from two sequential placebo-controlled
Drenth JP, Haagsma CJ, van der Meer JW. Hyperimmunoglobulinemia studies. Arthritis Rheum. 2008;58(8):2443–52
D and periodic fever syndrome. The clinical spectrum in a series of Hoffman HM, Wanderer AA, Broide DH. Familial cold autoinflamma-
50 patients. International Hyper-IgD Study Group. Medicine tory syndrome: phenotype and genotype of an autosomal dominant
(Baltimore). 1994;73(3):133–44 periodic fever. J Allergy Clin Immunol. 2001b;108(4):615–20
Duffy CM, Lam PY, Ditchfield M, et al Chronic recurrent multifocal Hoffmann GF, Charpentier C, Mayatepek E, et al Clinical and biochem-
osteomyelitis: review of orthopaedic complications at maturity. J ical phenotype in 11 patients with mevalonic aciduria. Pediatrics.
Pediatr Orthop. 2002;22:501–5 1993;91(5):915–21
Eleftheriou G, Bacis G, Fiocchi R, et al Colchicine-induced toxicity in Houten SM, Kuis W, Duran M, et al Mutations in MVK, encoding
a heart transplant patient with chronic renal failure. Clin Toxicol mevalonate kinase, cause hyperimmunoglobulinaemia D and peri-
(Phila). 2008;17:1–4 odic fever syndrome. Nature Genetics. 1999;22(2):175–7
48 M. Gattorno et al.

Huggins ML, Radford PM, McIntosh RS, et al Shedding of mutant Novack DV, Faccio R. Jawing about TNF: new hope for cherubism.
tumor necrosis factor receptor superfamily 1A associated with Cell. 2007;128:15–7
tumor necrosis factor receptor-associated periodic syndrome: differ- Prieur AM, Griscelli C, Lampert F, et al A chronic, infantile, neurologi-
ences between cell types. Arthritis Rheum. 2004;50(8):2651–9 cal, cutaneous and articular (CINCA) syndrome. A specific entity
Hull KM, Drewe E, Aksentijevich I, et al The TNF receptor-associated analysed in 30 patients. Scand J Rheumatol. 1987;66 Suppl:57–68
periodic syndrome (TRAPS): emerging concepts of an autoinflam- Prieur AM. A recently recognised chronic inflammatory disease of early
matory disorder. Medicine (Baltimore). 2002;81(5):349–68 onset characterized by the triad of rash, central nervous system involve-
Jacobelli S, Andre M, Alexandra JF, et al Failure of anti-TNF therapy in ment and arthropathy. Clin Exp Rheumatol. 2001;19(1): 103–6
TNF Receptor 1-Associated Periodic Syndrome (TRAPS). Rynne M, Maclean C, Bybee A, et al Hearing improvement in a patient
Rheumatology (Oxford). 2007;46(7):1211–2 with variant Muckle-Wells syndrome in response to interleukin 1
Kallinich T, Haffner D, Niehues T, et al Colchicine use in children and receptor antagonism. Ann Rheum Dis. 2006;65:533–4
adolescents with familial Mediterranean fever: literature review and Saito M, Nishikomori R, Kambe N, Fujisawa A, Tanizaki H, et al
consensus statement. Pediatrics. 2007;119(2):e474–83 Disease-associated CIAS1 mutations induce monocyte death,
Kallinich T, Haffner D, Rudolph B, et al “Periodic fever” without fever: revealing low-level mosaicism in mutation-negative cryopyrin-asso-
two cases of non-febrile TRAPS with mutations in the TNFRSF1A ciated periodic syndrome patients. Blood. 2008;111(4):2132–41
gene presenting with episodes of inflammation or monosymptom- Samuels J, Aksentijevich I, Torosyan Y, et al Familial Mediterranean
atic amyloidosis. Ann Rheum Dis. 2006;65(7):958–60 fever at the millennium. Clinical spectrum, ancient mutations, and a
Kastner D. Familial Mediterranean fever. In: Fauci A, Braunwald E, survey of 100 American referrals to the National Institutes of Health.
Kasper D, Hauser SL, Longo DL, Jameson L, Loscalzo J, editors. Medicine (Baltimore). 1998;77(4):268–97
Harrison’s principle of internal medicine. 17th ed. USA: McGraw- Sato TS, Khanna G, Ferguson PJ. Primary multifocal osseous lym-
Hill; 2008 phoma in a child. Pediatr Radiol. 2008;38:1338–41
Kavukçu S, Türkmen M, Eroğlu Y, et al Renal, gastric and thyroidal Schilling F, Wagner AD. Azithromycin: an anti-inflammatory effect in
amyloidosis due to familial Mediterranean fever. Pediatr Nephrol. chronic recurrent multifcal osteomyelitis? A preliminary report. Z
1997;11(2):210–2 Rheumatol. 2000;59:352–3
Kilcline C, Shinkai K, Bree A, et al Neonatal-onset multisystem inflam- Shinkai K, McCalmont TH, Leslie KS. Cryopyrin-associated periodic
matory disorder: the emerging role of pyrin genes in autoinflamma- syndromes and autoinflammation. Clin Exp Dermatol. 2008;33(1):
tory diseases. Arch Dermatol. 2005;141(2):248–53 1–9
Labro MT. Macrolide antibiotics: current and future uses. Expert Opin Shoham NG, Centola M, Mansfield E, et al Pyrin binds the PSTPIP1/
Pharmacother. 2004;5:541–50 CD2BP1 protein, defining familial Mediterranean fever and PAPA
Lachmann HJ, Sengul B, Yavuzsen TU, et al Clinical and subclinical syndrome as disorders in the same pathway. Proc Natl Acad Sci
inflammation in patients with familial Mediterranean fever and in U S A.. 2003;100(23):13501–6
heterozygous carriers of MEFV mutations. Rheumatology (Oxford). Siewert R, Ferber J, Horstmann RD, et al Hereditary periodic fever with
2006;45:746–50 systemic amyloidosis: is hyper-IgD syndrome really a benign dis-
Leslie KS, Lachmann HJ, Bruning E, et al Phenotype, genotype, and ease? Am J Kidney Dis. 2006;48(3):e41–5
sustained response to anakinra in 22 patients with autoinflammatory Simon A, Bodar EJ, van der Hilst JC, et al Beneficial response to inter-
disease associated with CIAS-1/NALP3 mutations. Arch Dermatol. leukin 1 receptor antagonist in TRAPS. Am J Med. 2004a;117(3):
2006;142(12):1591–7 208–10
Majeed HA, Al-Tarawna M, El-Shanti H, et al The syndrome of chronic Simon A, Kremer HP, Wevers RA, et al Mevalonate kinase deficiency:
recurrent multifocal osteomyelitis and congenital dyserythropoietic Evidence for a phenotypic continuum. Neurol. 2004b;62(6):994–7
anaemia. Report of a new family and a review. Eur J Pediatr. Speeg KV, Maldonado AL, Liaci J, et al Effect of cyclosporine on
2001;160:705–10 colchicine secretion by the kidney multidrug transporter studied in
Majeed HA, El-Shanti H, Al-Rimawi H, et al On mice and men: an vivo. J Pharmacol Exp Ther. 1992;261(1):50–5
autosomal recessive syndrome of chronic recurrent multifocal Stankovic K, Grateau G. Auto inflammatory syndromes: Diagnosis and
osteomyelitis and congenital dyserythropoietic anemia. J Pediatr. treatment. Joint Bone Spine. 2007;74:544–50
2000;137:441–2 Stichweh DS, Punaro M, Pascual V. Dramtic improvement of pyoderma
Majeed HA, Kalaawi M, Mohanty D, et al Congenital dyserythropoi- gangrenosum with infliximab in a patient with PAPA syndrome.
etic anemia and chronic recurrent multifocal osteomyelitis in three Pediatr Dermatol. 2005;22(3):262–5
related children and the association with Sweet syndrome in two Stoffman N, Magal N, Shohat T, et al Higher than expected carrier rates
siblings. J Pediatr. 1989;115:730–4 for familial Mediterranean fever in various Jewish ethnic groups.
McDermott MF, Aksentijevich I, Galon J, et al ermline mutations in the Eur J Hum Genet. 2000;8(4):307–10
extracellular domains of the 55 kDa TNF receptor, TNFR1, define a Stojanov S, Kastner DL. Familial autoinflammatory diseases: Genetics,
family of dominantly inherited autoinflammatory syndromes. Cell. pathogenesis and treatment. Curr Opin Rheumatol. 2005;17(5):
1999;97(1):133–44 586–99
Mimouni A, Magal N, Stoffman N, et al Familial Mediterranean fever: Tallon B, Corkill M. Peculiarities of PAPA syndrome. Rheumatology.
effects of genotype and ethnicity on inflammatory attacks and amy- 2006;45:1140–3
loidosis. Pediatrics. 2000;105(5):E70 The French FMF Consortium. A candidate gene for familial
Minetti EE, Minetti L. Multiple organ failure in a kidney transplant Mediterranean fever. Nat Genet. 1997;17(1):25–31
patient receiving both colchicine and cyclosporine. J Nephrol. The International FMF Consortium. Ancient missense mutations in a
2003;16(3):421–5 new member of the RoRet gene family are likely to cause familial
Neittaanmäki H. Cold urticaria. Clinical findings in 220 patients. J Am Mediterranean fever. Cell. 1997;90(4):797–807
Acad Dermatol. 1985;13(4):636–44 Thornton BD, Hoffman HM, Bhat A, Don BR. Am J Kidney Dis.
Neven B, Callebaut I, Prieur AM, et al Molecular basis of the spectral 2007;49:477–81
expression of CIAS1 mutations associated with phagocytic cell- Touitou I, Sarkisian T, Medlej-Hashim M, et al Country as the primary
mediated autoinflammatory disorders CINCA/NOMID, MWS, and risk factor for renal amyloidosis in familial Mediterranean fever.
FCU. Blood. 2004;103(7):2809–15 Arthritis Rheum. 2007;56(5):1706–12
5 Monogenic Autoinflammatory Syndromes 49

Ueki Y, Lin CY, Senoo M, et al Increased myeloid cell responses to possibly by inhibiting the cryopyrin inflammasome. Arthritis
M-CSF and RANKL cause bone loss and inflammation in SH3BP2 Rheum. 2008;58(3):864–8
“cherubism” mice. Cell. 2007;128:71–83 Yarom A, Rennebohm RM, Levinson JE. Infantile multisystem inflam-
Ueki Y, Tiziani V, Santanna C, et al Mutations in the gene encoding matory disease: a specific syndrome? J Pediatr. 1985;106(3):
c-Abl-binding protein SH3BP2 cause cherubism. Nat Genet. 390–6
2001;28:125–6 Yildiz A, Akkaya V, Kiliçaslan I, et al Cardiac and intestinal amyloido-
van der Hilst J, Bodar E, Barron K, et al Follow-up, clinical features, sis in a renal transplant recipient with familial Mediterranean fever.
and quality of life in 10.3 patients with HyperImmunoglubulin D J Nephrol. 2001;14(2):125–7
syndrome. Clin Exp Rheumatol. 2008;26(2):180 Yu JW, Fernandes-Alnemri T, Datta P, et al Pyrin activates the ASC
van der Meer JW, Vossen JM, Radl J, et al Hyperimmunoglobulinaemia D pyroptosome in response to engagement by autoinflammatory
and periodic fever: a new syndrome. Lancet. 1984;1(8386): 1087–90 PSTPIP1 mutants. Mol Cell. 2007;28:214–227
Williamson LM, Hull D, Mehta R, et al Familial Hibernian fever. Q J Meng XM et al. Clinicopathologic study of 24 cases of cherubism. Int J
Med. 1982;1(204):469–80 Oral Maxillofac Surg. 2005;34:350–6
Yamazaki T, Masumoto J, Agematsu K, et al Anakinra improves sen- Jain V, Sharma R. Radiographic, CT and MRI features of cherubism.
sory deafness in a Japanese patient with Muckle-Wells syndrome, Pediatr Radiol. 2006;36(10):1099–104
 Juvenile Spondyloarthropathy
6
David A. Cabral and Shirley M. L. Tse

ultimately develop classical sacroiliitis and back disease,

6.1 Overview their initial manifestations are usually a peripheral joint
arthritis and either enthesitis or a constellation of other asso-
› Juvenile spondyloarthropathy includes both differenti- ciated features (e.g., acute anterior uveitis).
ated and undifferentiated forms. The differentiated Patients with juvenile idiopathic arthritis (JIA) who are at
forms include juvenile ankylosing spondylitis, reactive high risk of developing ankylosing spondylitis in the future
arthritis, psoriatic arthritis, and arthritis with inflam- are described by the JIA subcategory of “enthesitis-related
matory bowel disease. arthritis” (ERA) (Petty et al. 1998, 2004). The diagnostic
› Enthesitis-related arthritis (ERA) is now the preferred features of ERA are shown in Table 6.1.
term for undifferentiated forms of juvenile spondyloar- Adolescents with ERA may have back symptoms but
thropathy. often lack radiologic evidence of inflammatory disease. The
› Most juvenile spondyloarthropathy patients present evolution of ERA to back and sacroiliac disease that is radio-
with asymmetric oligoarthritis involving the knee, hip, logically evident generally occurs in the late teens or when
ankle, tarsal, metatarsophalangeal, or foot interphalan- the patient is in his/her 20s.
geal joints.
Pearl: Most cases of spondyloarthropathy that present in
› Enthesitis – pain, tenderness, and swelling at the site of
childhood do so in an “undifferentiated” manner.
attachment of tendon, ligament, or fascia to bone – is
the most common feature of juvenile spondyloarthrop- Comment: Many adult patients with non-rheumatoid inflam-
athy. The frequency of this finding has led to the term matory arthropathy are at risk of developing axial arthritis;
“enthesitis-related arthritis” (ERA). yet do not fulfill criteria for a traditional spondyloarthropathy
› Monoarthritis of one of these joints, particularly one of initially. The generic diagnosis of “spondyloarthropathy” has
the large ones, is also common disease presentation. been used to describe these adult patients with “early” or
› Axial involvement seldom begins less than 5 years after “undifferentiated” disease (Amor et al. 1990; Dougados et al.
the clinical onset of juvenile spondyloarthropathy. 1991).
› Some patients with ERA are at high risk of developing The same is true for children with inflammatory arthropa-
ankylosing spondylitis. These include males, patients thy, a large majority of who present with “undifferentiated”
who are HLA-B27 (human leukocyte antigen B27) pos- spondyloarthropathy syndromes, but ultimately develop
itive, and those with early arthritis of the hips, ankles, or axial arthritis (Cabral et al. 1992; Flato et al. 1998; Jacobs
sacroiliac joints. et al. 1982; Rosenberg and Petty 1982). A variety of terms
› Acute anterior uveitis and inflammatory bowel disease have been used to describe children with these pediatric
are common extraarticular manifestations of ERA. spondyloarthropathies who are recognized as being at high
risk of developing axial arthritis (Table 6.2).
The most recently proposed nomenclature for classifying
pediatric chronic arthritis has encapsulated all of these syn-
dromes in a subset of JIA described as ERA (Petty et al.
Myth: Children with juvenile spondyloarthropathy – by
1998, 2004).
definition – have inflammatory disease of the back and
sacroiliitis. Myth: HLA-B27 is a good screening test for juvenile
spondyloarthropathy.
Reality: Most children with a juvenile spondyloarthropathy
do not have the archetypal ankylosing spondylitis that occurs Reality: HLA-B27 (human leukocyte antigen B27) is a class
in adults.Although many children with a “spondyloarthropathy” I surface antigen encoded by the B locus in the major

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 51

DOI:10.1007/978-1-84800-934-9_6, © Springer Science + Business Media B.V. 2009
52 D. A. Cabral and S. M. L. Tse

Table 6.1 Diagnostic features of ERA (adapted from Petty et al. 2004) Pearl: Children rarely present with back pain from spondy-
1. Arthritis and enthesitis or loarthropathy in the first decade of life.
2. Arthritis or enthesitis with at least two of the following:
• Sacroiliac joint tenderness or inflammatory lumbosacral pain Comment: Orthopedic, traumatic, infectious, and malignant
• HLA-B27 antigen positivity disorders are much more likely in children than are idio-
• Onset of arthritis in a male after 6 years of age pathic inflammatory conditions such as ERA or ankylosing
• Acute (symptomatic) anterior uveitis spondylitis. In particular, pre-pubertal children with ERA
• Family history in a first degree relative of ankylosing spondyli- virtually never present with back arthritis. Bone infections,
tis, ERA, sacroiliitis with inflammatory bowel disease, reactive
discitis, and malignancies such as neuroblastoma are the
arthritis, or acute anterior uveitis
major diagnostic considerations in that setting. In the older
Diagnostic exclusion criteria:
child or youth, Scheuerman’s disease (juvenile osteochon-
• Rheumatoid factor
• Features of systemic arthritis drosis of the spine), spondylolisthesis, scoliosis, and trauma
• Psoriasis in the patient or in a first-degree relative are other important considerations.
Some data suggest that Hispanic children and pediatric
patients with inflammatory bowel disease have axial arthritis
Table 6.2 Diagnostic syndromes used to describe juvenile spondyloar-
of earlier onset than do pediatric patients with other forms of
thropathies rheumatic disease (Burgos-Vargas 1989).
• Seronegative enthesopathy and arthropathy (SEA syndrome) Pearl: Tarsitis is a characteristic feature of spondyloarthrop-
• Pauciarticular onset JRA* type II athy in children, but occurs in only a minority of patients.
• Late-onset pauciarticular juvenile chronic arthritis (LOPA)
• HLA-B27-associated arthropathy and enthesopathy syndrome Comment: Tarsitis presents as tenderness and swelling of the
• ERA mid-foot (Fig. 6.1). This clinical manifestation of pediatric
spondyloarthropathy is probably caused by the combination
of synovitis, enthesitis, tenosynovitis, and bursitis. It is a
histocompatibility complex (MHC). HLA-B27 occurs in characteristic, albeit uncommon, feature of juvenile spondy-
about 90% of patients with ankylosing spondylitis and per- loarthropathy. Tarsitis may occur early in the disease course
haps 75% of children with ERA. The frequency of HLA-B27 (Burgos-Vargas, 1991).
varies considerably in different populations. As examples, The finding of tarsitis is a clue to look for other features
the frequency of this antigen is as low as 0.1% in the Japanese that suggest either ERA or juvenile ankylosing spondylitis.
population but as high as 24% among Laplanders. Tarsitis can occur with other forms of JIA, especially the pol-
As with adults, HLA-B27 is not a good screening test in yarthritis variant, but usually develops later in the disease
pediatric patients who do not have complaints that are rea- course as an “extension” of ankle disease.
sonably specific for ERA. HLA-B27 occurs in about 8% of Pearl: ERA should be considered in the differential diagnosis
the Caucasian population. However, only about 10% of HLA- of Osgood–Schlatter’s disease.
B27-positive individuals develop ankylosing spondylitis, and
10% of patients with ankylosing spondylitis are HLA-B27 Comment: Children diagnosed with ERA may have a previ-
negative. ous history of Osgood–Schlatter’s disease (osteochondrosis
For children who have documented arthritis or enthesitis, of the tibial tuberosity). The insertion of the patellar tendon
the presence of HLA-B27 is useful in classifying the type of to the tibial tubercle is a common site of enthesitis in chil-
arthritis as ERA. dren with ERA, and may be difficult to distinguish clinically
from Osgood–Schlatter’s disease.
Pearl: Enthesitis is the most common defining clinical fea- Osgood–Schlatter’s disease is more common than ERA,
ture of juvenile spondyloarthropathy. but ERA should be considered carefully if there is concur-
Comment: Enthesitis refers to inflammation – pain, tender- rent arthritis, a family history of spondyloarthropathy, or
ness, and swelling – at the enthesis, the site of attachment of enthesitis at other sites typical of ERA: behind of the heel (at
tendon, ligament, or fascia to bone. Enthesitis, the most com- the insertion of the Achilles tendon to calcaneus), beneath
mon clinical feature of ERA, occurs more frequently in chil- the heel (at the insertion of the plantar fascia to calcaneus),
dren who have classically-defined ankylosing spondylitis and beneath the metatarsal heads.
than it does in adults with that disorder (Burgos-Vargas 1989; Myth: Enthesitis requires early, aggressive pharmacologic
Gensler et al. 2008). Enthesitis is not unique to ERA in chil- intervention.
dren, and is detected occasionally among children with other
disorders, particularly psoriatic arthritis but also systemic Reality: Untreated enthesitis does not usually lead to per-
lupus erythematosus (Cabral et al. 1992). manent functional bony destruction in the same way that
6 Juvenile Spondyloarthropathy 53

a Severe cases may be difficult to treat, but initial therapy

should include simple analgesics, nonsteroid antiinflamma-
tory drugs (NSAIDs), and physical interventions such as
improved footwear and custom-made, semi-rigid orthotics.
Insightful podiatric care can redistribute the weight from
sites of painful enthesitis and provide cushioning at the ten-
der sites. Heel lifts alleviate stress in the retrocalcaneal area
and the knee. Low-dose prednisone is effective in patients
when enthesitis persists despite these interventions.

Myth: Inflammatory enthesitis is the most common cause of

pain behind the heel and under the feet in children.

Reality: Children who present with foot pain usually have

activity-related issues, mechanical problems, or both, rather
than inflammatory enthesitis. Flat feet and non-supportive
footwear are common. Children with flat feet often complain
of pain in the summer, when their levels of physical activity
are increased and they wear flat sandals or “flip-flops”.
Sever’s disease, also known as calcaneal apophysitis, is
another diagnostic possibility in the older child who presents
with pain behind the heel. Sever’s disease is caused by inflam-
mation that arises from overuse and repetitive microtrauma
of growth plates of the calcaneus. (It is, essentially, Osgood–
Schlatter’s disease of the heel.)

Pearl: Buttock pain in a patient with childhood spondyloar-

thropathy is usually not caused by sacroiliitis.

Comment: When assessing the clinical features of juvenile

b
spondyloarthropathy, clinicians should inquire about the
presence of buttock pain. This pain can be exacerbated by
activity or noted when sitting on a hard bench or chair. In
adults with ankylosing spondylitis, this pain in this area is
usually attributed to sacroiliitis. In children, however, sacro-
iliitis is not an early feature of the disease. Rather, enthesitis
involving the ischial tuberosity is the usual cause of buttock
pain. Enthesitis of the ischial tuberosity can be particularly
painful when sitting.

Myth: Sacroiliitis can only be detected by ordering a radio-

graph of the pelvis.

Reality: Sacroiliitis can be suspected or confirmed by either

physical examination or imaging. Positive clinical signs
include pain elicited on direct palpation of the sacroiliac joint
Fig. 6.1 Tarsitis shown clinically in the right foot (a) and using MRI (b) or stress maneuvers of the sacroiliac joints (Figs. 6.1 and 6.2).
Imaging techniques available to detect sacroiliitis include
arthritis does. As a result, aggressive pharmacological thera- radiographs, bone scans, computed tomography (CT), and
pies with second- or third-line antirheumatic drugs may not magnetic resonance imaging (MRI). Radiographic sacroilii-
be necessary. The natural history of enthesitis involving the tis in children often presents as “pseudo-widening” of the
lower limbs in children is that it tends to improve or disap- sacroiliac joint. Subsequent changes include sclerosis, ero-
pear with age. Nevertheless, enthesitis can be quite debilitat- sions, narrowing, and ankylosis (Bennett 1967).
ing in some patients. As an example, some are unable to walk Radiographic evidence of sacroiliitis typically lags behind
because of inferior heel pain. the clinical signs and MRI findings. Bone scans may not be
54 D. A. Cabral and S. M. L. Tse

a IBD. These symptoms can also develop after the onset of

gastrointestinal complaints. The activity of peripheral arthri-
tis in these patients correlates well with the activity of bowel
inflammation in about half of pediatric patients with IBD.
However, the arthritis of IBD in this setting tends to be short-
lived, responsive to therapy, and non-erosive.
Sacroiliitis and spondylitis are less common than periph-
eral arthritis among pediatric patients with IBD, but both are
more likely to occur in the setting of HLA-B27 positivity. In
addition, sacroiliitis and spondylitis often follow courses that
b are independent of IBD flares.
The pattern of peripheral arthropathy in children with
IBD who ultimately develop a spondyloarthropathy is simi-
lar to that of patients with ERA. The most common features
are an asymmetric, large-joint, oligoarthritis involving lower
extremities, frequently associated with enthesitis.
Myth: Patients with juvenile spondyloarthropathy generally
Fig. 6.2 Physical examination maneuvers designed to detect sacroilii- “outgrow” their arthritis.
tis (a) Patrick or FABER test: The patient lies supine and the ankle is
placed on the contralateral straight leg. Passive pressure is exerted onto Reality: Remission in children with juvenile spondyloar-
the ipsilateral thigh/knee resulting in flexion, abduction, and external thropathy occurs in between 15 and 45% of cases, and appears
rotation of the hip. Ipsilateral sacroiliac joint region tenderness is elic- to have the highest likelihood among patients classified as
ited in the presence of sacroiliitis. (b) Mennell’s Sign: The patient is having “undifferentiated” spondyloarthropathies (Flato et al.
positioned on his or her side and asked to flex the hip on the lower side
to the maximum extent. The examiner then extends the patient’s upper 1998, 2006; Minden et al. 2002). However, the majority of
hip by drawing back patient’s upper knee with a slight jerk and presses patients with undifferentiated juvenile spondyloarthropathy
a hand against the patient’s sacrum. Pain in the upper side is consistent progress to ankylosing spondylitis if followed for sufficient
with sacroiliitis in the upper sacroiliac joint lengths of time (Burgos-Vargas 1989; Cabral et al. 1992;
Minden et al. 2002; Flato et al. 2006).
helpful in the detection of sacroiliitis in children because the The development of axial disease is most likely in patients
differentiation of sacroiliitis from the growth plate within the who are HLA-B27 positive, who demonstrate hip arthritis at
pelvis is difficult. CT scans have the significant drawback of an early stage of disease, and who are older than eight years
high radiation exposure to the pelvic region. MRI has the of age at presentation. Screening and early therapy for axial
advantages of high sensitivity and the absence of ionizing involvement are especially important in patients with these
radiation, but is (of course) expensive. characteristics.
Pearl: Extra-articular manifestations of juvenile spondy- Myth: Juvenile spondyloarthropathy only occurs in boys.
loarthropathy frequently include the gut and the eye.
Reality: The prevalence of spondyloarthropathy is indeed
Comment: The most common extra-articular manifestations higher among boys, particularly before puberty. However,
of juvenile spondyloarthropathy are inflammatory bowel the proportion of girls with juvenile spondyloarthropathy
disease (IBD) and acute, symptomatic uveitis. Up to 80% of increases with age. Over time, the percentage of cases that
children with juvenile spondyloarthropathy have nonspecific occur in females is approximately equal to that in males.
IBD (Mielants et al. 1991, 1993; Cabral et al. 1992). Moreover, some data suggest that girls with juvenile spondy-
Many extra-articular manifestations that are reported in loarthropathy have worse outcomes (Stone et al. 2005). Thus,
adult spondyloarthropathy are less common in children. As the notions that juvenile spondyloarthropathy is primarily a
examples, cardiac conduction abnormalities, aortic valvular disease of boys and that boys have more severe disease than
insufficiency, apical fibrosis of the lungs, urethritis, circinate girls both appear untrue.
balanitis, and atlanto-axial subluxation are only observed
rarely in juvenile spondyloarthropathy. Pearl: Regular eye examinations are required for some sub-
sets of patients with JIA, but not patients with juvenile
Myth: Pediatric patients with IBD eventually develop spondyloarthropathy.
spondylitis.
Comment: Patients with juvenile spondyloarthropathy are at
Reality: Peripheral arthritis, sacroiliitis, and spinal disease risk for an acute anterior uveitis (iritis). In contrast to the chronic,
can be present at the onset of gut symptoms in children with asymptomatic uveitis that is observed in pauciarticular JIA, the
6 Juvenile Spondyloarthropathy 55

References

Amor B, Dougados M, Mijiyawa M. Criteres de classification des spon-

dyloarthropathies. Rev Rhum Mal Osteoartic 1990; 57:85–89
Bennett PH. B. T. New York symposium on population studies in the
rheumatic diseases: New diagnostic criteria. Bull Rheum Dis 1967;
17:453–458
Burgos-Vargas R. Ankylosing tarsitis: Clinical features of a unique
form of tarsal disease in the juvenile-onset spondyloarthropathies.
Arthritis Rheum 1991; 34(Suppl):D196
Burgos-Vargas R, Clark P. Axial involvement in the seronegative enthesop-
athy and arthropathy syndrome and its progression to ankylosing
spondylitis. J Rheumatol 1989; 16(2):192–197
Burgos-Vargas R, Naranjo A, Castillo J, Katona G. Ankylosing spon-
dylitis in the Mexican Mestizo: Patterns of disease according to age
at onset. J Rheumatol 1989; 16:186–191
Burgos-Vargas RC, Gutierrez-Suarez R, Vazquez-Mellado J. Efficacy,
safety, and tolerability of Infliximab in Juvenile-onset Spondyloar-
Fig. 6.3 Patient with acute exacerbation of HLA-B27 associated iritis thropathies (JO-SpA): Results of the three-month, randomized,
and hypopyon (Figure courtesy of Dr. George Papaliodis) double-blind, placebo-controlled trial phase. Arthritis Rheum 2007;
56(9S):319
Cabral, DA, Oen KG, et al SEA syndrome revisited: A longterm fol-
lowup of children with a syndrome of seronegative enthesopathy
acute anterior uveitis that complicates juvenile spondyloar- and arthropathy. J Rheumatol 1992; 19(8):1282–1285
thropathy is associated with photophobia, blurred vision, and a Cassidy J, Kivlin J, et al Ophthalmologic examinations in children with
red eye (Fig. 6.3). juvenile rheumatoid arthritis. Pediatrics 2006; 117(5):1843–1845
Dougados M, van der Linden S, et al The European Spondylarthropathy
Because the ocular disease of juvenile spondyloarthropathy
Study Group preliminary criteria for the classification of spondylar-
is highly symptomatic, most patients do not need more than the thropathy. Arthritis Rheum 1991; 34(10):1218–1227
recommended ophthalmology follow-up for children unless Flato B, Aasland A, et al Outcome and predictive factors in juvenile
they have symptomatic eye disease. This normally means eye rheumatoid arthritis and juvenile spondyloarthropathy. J Rheumatol
1998; 25(2):366–375
examinations every 6 to 12 months (Cassidy et al. 2006).
Flato B, Hoffmann-Vold AM, et al Long-term outcome and prognostic
Pearl: Tumor necrosis factor (TNF) inhibitors are effective for factors in enthesitis-related arthritis: A case–control study. Arthritis
Rheum 2006; 54(11):3573–3582
both the arthritis and enthesitis of juvenile spondyloarthropathy. Gensler LS, Ward MM, Reveille JD, Learch TJ, Weisman MH, Davis Jr.
JC. Clinical, radiographic, and functional differences between juve-
Comment: Currently, there are no effective therapies for the axial
nile-onset and adult-onset ankylosing spondylitis: Results from the
disease (spine or sacroiliac) in juvenile spondyloarthropathy. PSOAS cohort. Ann Rheum Dis 2008; 67:233–237
Treatment with NSAIDs and glucocorticoids may provide symp- Henrickson M, Reiff A. Prolonged efficacy of etanercept in refractory
tomatic improvement, but do not alter axial disease progression. enthesitis-related arthritis. J Rheumatol 2004; 31(10):2055–2061
Jacobs JC, Berdon ED, Johnston WE. HLA-B27-associated spondy-
Disease-modifying antirheumatic drugs such as sulfasalazine
loarthritis and enthesopathy in childhood: Clinical, pathologic, and
and methotrexate are also ineffective for axial disease. radiographic observations in 58 patients. J Pediatr 1982; 100:
High levels of TNF are identified at the sites of synovitis 521–528
and enthesitis in adults and children with spondyloarthropa- Mielants H, Veys EM, et al Gut inflammation in the spondyloarthropa-
thies: Clinical, radiologic, biologic and genetic features in relation
thies. Treatment with TNF inhibitors such as etanercept or
to the type of histology. A prospective study. J Rheumatol 1991;
infliximab has led to clinical and radiologic improvement in 18(10):1542–1551
juvenile spondyloarthropathy in both short- and long-term Mielants H, Veys EM, et al Gut inflammation in children with late onset
studies (Henrickson and Reiff 2004; Schmeling and Horneff pauciarticular juvenile chronic arthritis and evolution to adult
spondyloarthropathy–a prospective study. J Rheumatol 1993; 20(9):
2004; Tse et al. 2005, 2006, 2007; Burgos-Vargas et al. 2007). 1567–1572
Specifically, the arthritis (peripheral and axial), enthesitis, and Minden K, Niewerth M, et al Long-term outcome in patients with juve-
inflammatory markers all improved and demonstrated sus- nile idiopathic arthritis. Arthritis Rheum 2002; 46(9):2392–2401
tained responses over time. Petty RE, Southwood TR, et al Revision of the proposed classification
criteria for juvenile idiopathic arthritis: Durban, 1997. J Rheumatol
AntiTNF agents in juvenile spondyloarthropathy are asso- 1998; 25(10):1991–1994
ciated with clinical and radiographic (MRI) improvement in Petty RE, Southwood TR, et al International league of associations for
the peripheral and axial joints. Because axial involvement is rheumatology classification of juvenile idiopathic arthritis: Second
rare early in the disease course and does not present usually revision, Edmonton, 2001. J Rheumatol 2004; 31(2):390–392
Rosenberg AM, Petty RE. A syndrome of seronegative enthesopathy and
until the second or third decade of life, it remains unclear if arthropathy in children. Arthritis Rheum 1982; 25(9):1041–1047
TNF inhibition will prevent syndesmophyte formation and Schmeling H, Horneff G. Infliximab in two patients with juvenile anky-
spinal fusion later in the disease course. losing spondylitis. Rheumatol Int 2004; 24(3):173–176
56 D. A. Cabral and S. M. L. Tse

Stone M, Warren RW, et al Juvenile-onset ankylosing spondylitis is Tse SM, Laxer RM, et al Radiologic improvement of juvenile idiopathic
associated with worse functional outcomes than adult-onset ankylo- arthritis-enthesitis-related arthritis following anti-tumor necrosis factor-
sing spondylitis. Arthritis Rheum 2005; 53(3):445–451 alpha blockade with etanercept. J Rheumatol 2006; 33(6):1186–1188
Tse SM, Burgos-Vargas R, et al Anti-tumor necrosis factor alpha block- Tse SML, Burgos-Vargas R, O’Shea F, Inman R, Laxer R. Long term
ade in the treatment of juvenile spondylarthropathy. Arthritis Rheum outcome of anti-TNF therapy in juvenile spondyloarthropathy.
2005; 52(7):2103–2108 Arthritis Rheum 2007; 56(9S):898
 Ankylosing Spondylitis
7
Juergen Braun and David Tak Yan Yu

with chronic back pain that has persisted for more than 3
7.1 Overview of Ankylosing Spondylitis months is on the order of 5% (Underwood and Dawes 1995).
Myth: AS is a benign disease with a relatively good prognosis.
› Ankylosing spondylitis (AS) is the prototypical form
of the spondyloarthritides, a group of disorders that Reality: An oft-quoted study that examined military veterans
involves inflammation of the sacroiliac joints, spine, with AS is partly responsible for this myth (Carette et al.,
joints, and entheses, as well as extraspinal lesions of 1983). However, that study was characterized by a high loss
the eye, bowel, and heart. to follow-up. Contrary to this Myth as stated, many patients
› Human leukocyte antigen (HLA) B27 is a strong who were tracked for a long period developed severe func-
genetic risk factor for AS. However, this gene is nei- tional restrictions. Other studies indicate that at least one
ther necessary nor sufficient to cause the disease. Other third of patients with AS have severe courses characterized
recently recognized genes are also involved in the by significant pain and disability (Zink et al. 2000). These
pathophysiology of this condition. morbidities have substantial implications for family life and
› The principal musculoskeletal lesions of AS are sacro- occupation (Ward et al. 2008).
iliitis, spondylitis, synovitis, and enthesitis. Enthesitis
Myth: Only males get AS.
consists of inflammation at the site of tendinous inser-
tions into bone. Reality: Older textbooks of rheumatology reported the male
› Sacroiliitis causes inflammatory back pain often local- predominance in AS to be as high as 16:1. However, a more
ized to the buttocks that can be uni- or bilateral and accurate estimation of the male:female ratio is on the order of
typically alternates between the left and right side. 2:1. Besides numbers, some true differences do appear between
› Spondylitis begins characteristically in the lumbosacral male and female patients with AS. For example, male patients
region and proceeds cephalad. are more likely to have syndesmophytes and bony ankylosis
› The most common extraspinal joints involved are the than females with the disease (Rudwaleit et al. 2009a).
hips, knees, ankles, and metatarsophalangeal joints. The
Myth: Inflammatory back pain is such a highly specific symp-
pattern of arthritis is typically an asymmetric oligoarthri-
tom that it is nearly diagnostic of AS or some form of spondy-
tis that involves the large joints of the lower extremities.
loarthritis.
› Acute anterior uveitis usually occurs in one eye at a
time, although either eye can be involved in any given Reality: Inflammatory back pain is a hallmark of AS.
episode. Patients usually present with a red, painful, However, it is present in only about 70% of patients with
photophobic eye. established disease. In addition, up to 30% of patients with
back pain due to other causes also report symptoms compat-
ible with an inflammatory etiology. Thus, additional history
that supplements symptoms compatible with inflammatory
back pain is required to make the diagnosis. Such additional
Myth: AS is a rare disease.
information might include a good response to NSAIDs, posi-
Reality: AS, the most common subtype of spondyloarthritis, tivity for HLA-B27, an elevated C-reactive protein, other
has a prevalence in the range of 0.3–0.5% (Braun and Sieper features of spondyloarthritis such as uveitis, psoriasis, and
2007). The group of spondyloarthritides as a whole has a colitis, and compatible findings on imaging studies of the
prevalence that is approximately as high as that of rheumatoid sacroiliac joints. New criteria for inflammatory back pain
arthritis (RA) (Saraux et al. 2005). The prevalence of spondy- (Rudwaleit et al. 2006) and axial SpA (Rudwaleit et al.
loarthritis among patients who present to general practitioners 2009b, c) have been proposed recently.

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 57

DOI:10.1007/978-1-84800-934-9_7, © Springer Science + Business Media B.V. 2009
58 J. Braun and D. T. Y. Yu

Myth: All patients with AS are HLA-B27 positive. 100

Reality: Although HLA-B27 is clearly the strongest genetic 90

factor involved in the pathogenesis of AS, it is neither neces-
sary nor sufficient to cause the disease. The risk contributed 80

by HLA B27 is about 50% of the total risk attributed to

70
genes. Other genetic factors such as the IL-23 receptor and

Post-test probability
ERAP-1 have been discovered recently (Wellcome Trust 60
Nature Genetics 2007). Environmental factors may also play
a role, albeit a less important one. Firmly linked environmen- 50
tal factors have been identified for only a subgroup of SpA,
namely, patients with reactive arthritis (see Chap. 9). 40

The relative frequency of HLA-B27 varies across differ-

30
ent populations. More than 40 HLA-B27 subtypes exist. At
least two subtypes are not associated with AS. These are 20
HLA-B2706 found in Thailand (Lopez-Larrea et al. 2002)
and HLA-B2709 found in Sardinia. In Western countries, it 10
is generally not useful to determine HLA B27 subtypes.
0
Caucasians usually carry the HLA B2705 subtype, which is
0 50 100
known to be associated with the disease. A substantial num-
Prior disease probability
ber of patients who have clinical features compatible with
AS, but who lack HLA-B27 actually have psoriasis or inflam- Fig. 7.1 The relationship between pretest probability to posttest prob-
matory bowel disease. ability for HLA-B27 determinations (Reproduced from Brown 2007,
with permission from Wiley-Blackwell)
Myth: HLA-B27 is not useful for the diagnosis of AS.
Reality: In certain populations, including the Caucasians of elevated CRP levels at baseline, these levels usually fall
North America and Western Europe and most of the popula- dramatically once TNF inhibition is begun (Braun et al. 2002).
tions in China and India, HLA-B27 is present in at least 90% A low CRP level does not necessarily predict that the
of patients with AS but among less than 10% of the general patient will respond poorly to TNF inhibition. If sufficient
population. doses of NSAIDs do not lead to an adequate clinical response,
The utility of HLA-B27 in diagnosing AS and SpA is a trial of a TNF inhibitor is indicated.
because, in a patient who presents with lower back pain of
Myth: Measurement of CRP levels is the only way to assess
greater than 3 months’ duration, the likelihood of an SpA
disease activity in AS.
increases tenfold if HLA-B27 is present (Braun et al. 1998).
In general, the pretest probability of having the disease mat- Reality: The clinical standard to assess disease activity in AS
ters. The higher the suspicion of AS before the test, the is the BASDAI (see below). The items “pain” and “morning
greater the posttest probability of a positive result (Fig. 7.1) stiffness” are considered especially relevant in this regard.
(Brown 2007). However, its contribution to clinical diagno- However, magnetic resonance imaging (MRI) has emerged
sis is greatest in early disease, the time at which patients as an important tool that has the capacity to localize spinal
often do not have definitive radiographic signs. Accordingly, inflammation rather exactly (Fig. 7.2) (Braun 2002). MRI is
a negative test for HLA-B27 largely decreases the likelihood also useful in quantifying the extent of spondylitis before
for AS and SpA. and after anti-TNF therapy (Braun 2006a).
Thus, if HLA-B27 genotyping is used in an appropriate
Myth: Function in AS is mainly determined by ankylosis.
clinical circumstance, a positive result contributes signifi-
cantly to the diagnostic evaluation. Therefore, HLA B27 Reality: As shown in a recent study (Landewé 2009), disease
testing has become an important item in the new classifica- activity and structural damage both contribute independently
tion criteria for axial SpA (Rudwaleit et al. in press (b)). to functional decline in patients with AS. This information is
highly relevant for determining patients’ prognosis for func-
Myth: All patients with AS have elevated C-reactive protein
tional recovery and informs management decisions.
(CRP) levels.
Pearl: Remember to exclude inflammatory bowel disease in
Reality: This unfortunate but widely espoused myth frequently
a patient with AS/SpA.
leads to diagnostic delays. Only one half of all patients with
AS have elevated CRP levels (Spoorenberg et al. 1999). Comment: The clinical features of AS overlap substantially
However, among the sizeable subset of patients who have with those of inflammatory bowel disease. Patients with
7 Ankylosing Spondylitis 59

Fig. 7.2 Patient with AS,

spondylitis, and structural
a b
changes (a) chronic structural
changes (b) active lesions

symptoms such as elevated stool frequency and unexplained Table 7.1 Preliminary criteria for the diagnosis of ankylosing spondylitis
abdominal pain should undergo endoscopy. A low threshold ASAS classification criteria for SpA
for endoscopy should also be maintained in patients who are
In patients with ≥3 months In patients with peripheral
HLA-B27 negative. This is because, among patients with back pain and age at onset symptoms ONLY
inflammatory bowel disease who have symptoms of SpA, < 45 years
only approximately 50% are HLA-B27 positive. This con-
trasts with the situation with AS, a population of patients in
Sacroiliitis on HLA-B27 plus Arthritis or enthesitis or
whom the great majority are HLA-B27 positive. imaging plus OR ≥2 other SpA dactylitis plus
A reorientation of the diagnosis to inflammatory bowel ≥1 SpA feature features
disease alters the entire therapeutic strategy for a patient with ≥1 SpA features
joint symptoms. TNF blockers such as infliximab and adali- SpA features − Uveitis
− IBP − psoriasis
mumab are also approved for Crohn’s disease, but additional
− arthritis − Crohn’s/colitis
therapeutic measures may also be important. − Preceding infection
− enthesitis heel
− uveitis − HLA-B27
Myth: Because all patients with AS have sacroiliitis, the − dactylitis OR
diagnosis is generally quite straightforward. − psoriasis ≥2 SpA features
− Crohn’s/colitis − Arthritis
Reality: The 1984 New York classification criteria for AS − good response to NSAIDs − enthesitis
(van der Linden et al. 1984) are useful for the purpose of − family history for SpA − Dactylitis
− HLA-B27 − amily history for SpA
classification and diagnosis in patients with long-standing − elevated CRP − Sacroiliitis on imaging
disease, but not in patients with early disease. This is
because the sensitivity of structural changes within the
sacroiliac joints is high among patients with established
Reality: Structural changes in the sacroiliac joints occur in
disease, but low (approximately 20%) in patients with
approximately 95% of patients with AS. In contrast, only
early SpA.
about 30% of patients with psoriatic SpA and an even smaller
New criteria under development for the diagnosis of AS
percentage of those with reactive arthritis develop sacroilii-
include HLA-B27 positivity and findings on MRI studies.
tis. Thus, AS is much more prevalent than other forms of
These new criteria are likely to resolve most issues related to
axial SpA.
an early diagnosis of axial SpA. Importantly, several param-
Whereas the sacroiliitis in AS is typically bilateral (at
eters are required in order for the diagnosis to be classified as
least in advanced stages of disease), sacroiliac joint inflam-
secure (Table 7.1) (Rudwaleit et al. b, c).
mation in PsA and ReA is often unilateral and asymmetric
In the discipline of rheumatology, criteria sets seldom
(Muche et al. 2003). However, the full clinical picture and an
obviate the need for astute clinical judgement.
accurate determination of symptoms and signs of inflamma-
Myth: The radiographic pattern of sacroiliac changes is tion in all joints (and skin, eyes, and gastrointestinal tract) is
clinically relevant because it helps differentiate among the crucial. In the early diagnosis of sacroiliitis, MRI is clearly
different forms of spondyloarthritis. the study of choice.
60 J. Braun and D. T. Y. Yu

Reality: Advanced cases of AS are difficult to distinguish

from diffuse idiopathic skeletal hyperostosis (DISH). One
study of patients with AS, who had a mean disease duration
of approximately 12 years, indicated that more than 10% of
all spinal osteophytes detected had the radiologic appearance
of degenerative osteophytes. The cutoff used in that study
was a 45° growth angle (Baraliakos et al. 2007b).
There are two potential causes of difficulty in parsing the
etiology of syndesmophytes. First, patients with AS may
develop degenerative spinal changes secondary to those
induced by inflammation. Second, not all syndesmophytes
follow the same rules, and they may have a less than classic
appearance. There is evidence that the detection of syndesmo-
phytes by MRI is difficult (Braun 2002). Radiography remains
the gold standard for syndesmophyte characterization.
Pearl: Remember to consider DISH before diagnosing
advanced AS.
Comment: Many patients with late-stage AS are recognizable
easily by their posture. They have a forward stooping of the
neck, a high dorsal kyphosis, a rounding of the shoulders,
loss of the normal lumbar lordosis, wasting of the buttocks
muscles, flattening of the chest, and ballooning of the abdo-
men (Olivieri et al. 2007). Before rendering the diagnosis of
AS on this posture alone, however, clinicians should consider
an alternative diagnosis that can cause the identical postural
changes: diffuse idiopathic skeletal hyperostosis, commonly
Fig. 7.3 Lateral view of the cervical spine in a patient with ankylosing
spondylitis. Note the slender and vertical characters of the bony briges known as DISH or Forestier’s disease (Mader 2008).
between the anterior aspects of the vertebrae (Figure courtesy of Physical examination maneuvers alone cannot distinguish
Dr. Thomas Learch, Cedar Sinai Medical Center, Los Angeles) DISH from AS. The range of motion in the cervical and lum-
bar spines can be limited to similar extents in these two con-
ditions. In addition, patients with DISH have similar decreases
A patient with cutaneous psoriasis and definite structural
in their chest expansions and in the ranges of motion at the
changes in the sacroiliac joints is usually diagnosed more
shoulders and hips. Patients with DISH may complain of spi-
appropriately as AS associated with psoriasis. Structural
nal pain, but their history of back pain starts usually later than
changes in the sacroiliac joints of patients with reactive
in AS, and the initial localization is different.
arthritis are relatively rare.
Patients with both DISH and AS develop bony bridges
Pearl and Myth: The nature of syndesmophytes differs across between their vertebral bodies. The key to differentiating
the subtypes of spondyloarthritis. these disorders is that these bony bridges have different
radiologic appearances:
Comment: The syndesmophytes of AS are usually symmet-
ric, delicate, and vertical— at least in about 90% of the • DISH leads to ossification of the anterior longitudinal lig-
patients (Fig. 7.3). In contrast, the syndesmophytes of ament of the spine. These changes are frequently described
patients with psoriasis are often bulky, asymmetric, and tend as “exuberant” or “flowing” mantles along the anterior and
to protrude laterally before progressing vertically. anterolateral aspects of the vertebrae (Fig. 7.4).
Although this fine point is correct from the standpoint of • In AS, the bony bridges between the vertebrae are vertical
diagnosis, it makes little difference now in most circumstances and much more slender, involving the outer margin of the
with regard to treatment. Regardless of the radiologic appear- annulus fibrosus (see Fig. 7.3).
ance of their syndesmophytes, the back pain symptoms of
The other radiological feature that distinguishes AS from
patients with AS, psoriatic arthritis, or inflammatory bowel
DISH, sometimes more reliably, is the presence of erosions
disease all respond well to TNF inhibitors.
within the sacroiliac joints of patients with AS. Sacroiliac
Myth: Syndesmophytes are distinguished easily from the osteo- joint erosions are present invariably among patients with AS
phytes associated with degenerative arthritis of the spine. who have significant radiological involvement of the spine.
7 Ankylosing Spondylitis 61

The hypothesis that Klebsiella infections cause some

cases of AS was put forth in the 1970s and 1980s on the basis
of two arguments:
• The potential for molecular mimicry between Klebsiella
species and HLA B27
• The reportedly high prevalence of antiKlebsiella antibod-
ies in AS.
The functional relevance of the molecular similarity between
Klebsiella species and HLA B27 has never been proven. In
addition, the serologic findings are probably attributable to
silent gut inflammation, possibly enhanced by NSAID use,
rather than a true pathogenetic mechanism.
Putative links between AS and Chlamydia species or other
bacteria that can cause clinically inapparent inflammation
within the urogenital tract have been proposed, but the data
for such associations are not convincing. Extensive poly-
merase chain reaction studies have provided evidence for
small amounts of bacterial DNA in the joints of patients with
reactive arthritis but not of those with AS (Braun et al. 2007a,
b). However, it is altogether much more probable that the cel-
lular immune reaction to bacterial antigens is of greater
pathophysiological importance than the presence of bacte-
ria—especially since they have also been detected in the joints
of healthy individuals.
Although several interesting ideas have been followed, the
pathogenesis of AS has still not been fully elucidated. The
Fig. 7.4 Lateral view of the cervical spine in a patient with DISH. Note
the exuberant appearance of the ossification along the anterior margins genetic contribution seems to be rather dominant, explaining
of the vertebrae (Figure courtesy of Dr. Thomas Learch, Cedar Sinai at least 80% of the pathogenesis.
Medical Center, Los Angeles)
Myth: AS is difficult to treat.
Reality: This was true–painfully true–until the era of TNF
Sacroiliac joint erosions tend to occur first on the iliac side of inhibitors (Braun 2002). Assuming that patients have access to
the joint. appropriate anti-TNF regimens, most respond very well to this
Sacroiliac joint erosions do not occur in DISH, but the intervention. Of note, there is some evidence that these agents
sacroiliac joints are obscured in some patients by sacroiliac work even better in SpA than in RA (Heiberg et al. 2008).
capsular bridging. In such cases, computed tomography may NSAIDs remain the standard first-line therapy for AS
be required to differentiate AS from DISH. However, this is (Zochling et al. 2006). TNF inhibition, on the other hand, is
a largely academic question that has usually no therapeutic the standard of care for patients who have persistently high
consequences. disease activity despite optimal NSAID regimens. In contrast to
HLA-B27 is present in a subtantially higher percentage of the situation with RA, in which methotrexate remains the cor-
patients with AS compared with patients with DISH. However, nerstone of therapy even if a biologic agent is employed simul-
the coexistence of not only HLA-B27 positivity and DISH but taneously (see Chap. 1), methotrexate and other synthetic
also AS and DISH has been reported (Maertens et al. 1992). DMARDs have extremely limited roles in the treatment of AS.
Myth: AS is caused by Klebsiella or some other species of Pearl: Doses of NSAIDs must be optimized.
bacteria.
Comment: NSAIDs should be employed in at the highest
Reality: The concept that bacterial infections in the urogeni- approved dose when treating AS. NSAIDs with longer half
tal tract, gastrointestinal tract, and other sites can trigger a lives may be more effective than those with shorter half lives,
noninfectious inflammatory arthritis stems from the strong particularly in providing pain relief through the night. If a
association that exists between certain pathogens and reac- particular, NSAID has not demonstrated sufficient therapeu-
tive arthritis (Chapter 9). Reactive arthritis shares some fea- tic effect within 1 week, another NSAID should be prescribed
tures with other SpA. (Song et al. 2008).
62 J. Braun and D. T. Y. Yu

The potential for gastrointestinal, renal, and cardiovascu-

lar adverse effects must always be considered when using
NSAIDs and weighed against the possible adverse effects of
TNF inhibitors.
Pearl: Elevated acute phase reactants in a patient with AS do
not imply that glucocorticoids will be an effective treatment.
Comment: The systemic use of glucocorticoids is not recom-
mended in AS, in contrast to RA. Clinical experience in AS
indicates that only high doses of glucocorticoids (too high) are
effective for the spinal disease in AS, and even then only in
some patients. Greater efficacy of glucocorticoids may be
anticipated for patients with a predominantly peripheral
arthritis, but even this has not been tested adequately. The clini-
cal judgement of AS experts is that comparable doses of gluco-
corticoids are far less likely to be effective in AS than in RA. Fig. 7.5 A patient with severe ankylosis of the spine from long-standing,
untreated ankylosing spondylitis. The patient had a serum C-reactive
Pearl: Injectable glucocorticoid preparations are effective in protein level of 62.4 mg/dL (normal < 8.0 mg/dL) and an erythrocyte
patients with isolated sacroiliitis, monoarthritis or oligoar- sedimentation rate of 132 mm/h (normal < 20 mm/h). He responded
thritis of peripheral joints, or enthesitis. dramatically to infliximab, with pronounced symptom relief and a swift
normalization of his acute phase reactants
Comment: Although oral glucocorticoids are generally inef-
fective in AS, clinical anecdotes suggest that intraarticular
injections provide clinical relief for up to 9 months in many A prevalent myth holds that as the spine becomes ankylosed,
patients. Computed tomography-guided injections may be the degree of inflammation also subsides, such that patients’
necessary for the sacroiliac joints, especially if sacroiliitis is symptoms become much less responsive to antiinflammatory
the predominant clinical problem (Braun et al. 1996). therapies. The response of the patient whose spine is shown
Fig. 7.5 to infliximab repudiates this myth strongly.
Pearl: Methotrexate definitely does not work for axial disease.
The only phase III TNF blocker study that included some
Comment: Although methotrexate has been used for this indi- patients with AS with complete spinal ankylosis (n = 11) was
cation for a long time, there is no evidence (not even a hint) a multicenter trial involving 315 patients that compared adal-
that this agent is effective for the spinal inflammation of AS imumab with placebo (Van der Heijde et al. 2006). Among
(Haibel et al. 2007). There is also no evidence that the addi- these patients, three of the six who were treated with adali-
tion of methotrexate to a TNF inhibitor regimen potentiates mumab demonstrated significant clinical responses. In con-
the effect of the biologic agent (Breban et al. 2008). In theory, trast, none of the five patients who received placebo were
methotrexate might prevent the formation of human antichi- responders. Hence, TNF blockers can be considered even in
meric antibodies in patients treated with infliximab, but this patients with very late stages of the AS, especially if NSAIDs
has not been demonstrated to be relevant in patients with AS. and other analgesics are poorly tolerated or not efficacious.
This is in agreement with immunohistological studies
Pearl: Consider sulfasalazine in patients with relatively mild
showing that even at a late stage of ankylosis, significant
disease, but don’t hesitate to prescribe anti-TNF therapy in
inflammatory changes still exist in the spine (Appel 2006;
patients with persistently active disease.
Neidhart et al. 2008).
Comment: Some studies indicate a limited effect of sulfasala-
Myth: TNF blockers can arrest the progression of AS.
zine for patients with active peripheral and axial disease,
especially in early stages (Braun 2006b). However, this does Reality: The reverse side of the myth discussed earlier is the
not imply that sulfasalazine must be prescribed before resort- misconception that TNF inhibitors should be used early in
ing to an anti-TNF agent. Patients with persistently active patients with AS, even those with mild disease, because they
disease should receive a TNF inhibitor early in their disease will arrest the progression of structural damage. One of the
course (Zochling et al. 2006). major problems in trying to test this hypothesis is that AS
progresses at a slower rate compared with that of RA. At
Myth: TNF blockers are useful only in patients with AS before
least 2 years are required in AS to detect significant radio-
their spines become fused.
logic changes using the best available scoring system, the
Reality: Most of the phase III studies on the use of TNF block- mSASSS. In this era, it is not ethical to put patients on pla-
ers have not included patients with total ankylosis of the spine. cebo for a period that long.
7 Ankylosing Spondylitis 63

Because of this limitation, radiological progression in fractures among patients with AS is increased several fold
patients on TNF blockers can be compared only with base- compared with healthy controls. The treatment of osteopenia
line observations of the same cohort or alternately to data of and osteoporosis has not been studied in AS, but reasonable
cohorts collected before use of biologics. Using either method approaches to this problem include calcium (1,500 mg of
of comparison, neither infliximab nor etanercept has shown elemental calcium per day) and vitamin D, as well as bispho-
significant effects on radiological progression (Baraliakos sphonates if appropriate. Some data indicate that bisphos-
et al. 2007a; Van der Heijde et al. 2008). This is not surpris- phonates have a favorable effect on other disease outcomes
ing, because the process of bone remodeling responsible for (Maksymowych et al. 2002).
ankylosis may be partly independent of that of inflammation Bone mineral density measurements at the lumbar spine
(Baraliakos et al. 2008). Etanercept, which suppresses the in patients with syndesmophytes are not useful because the
symptoms of AS effectively, does not suppress ankylosis in measurements will be falsely high (Karberg et al. 2005).
an animal model of enthesitis (Lories et al. 2007). The opti-
Pearl: Talk to the patient with a physically demanding job
mal approach to suppressing bone remodeling is a major cur-
about the future of his employment.
rent focus of AS research.
Comment: A physically demanding profession is associated
Myth: The concomitant use of methotrexate enhances the
with a poor AS outcome in terms of both structural and func-
utility of TNF blockers in patients with AS.
tional damage (Ward et al. 2008). Patients with AS should be
Reality: In a randomized trial, patients with AS received advised of the potential wisdom of selecting different types
either the combination of infliximab and methotrexate or inf- of work if their current occupations bring significant physi-
liximab and placebo (Breban et al. 2008). No additional cal demands.
effect of MTX was found. In another study, there was no
Pearl: Systematic assessments of disease activity are crucial
synergistic effect of methotrexate plus a TNF inhibitor on
to the longitudinal management of AS.
MRI progression of disease (Li et al. 2008).
Not all patients with AS respond dramatically to their first Comment: Standardized and validated assessments are
TNF inhibitor. In practice, if a patient with AS responds increasingly the standard of care for the management of
poorly to one TNF blocker, it is probably more useful to patients with RA. Similarly, measuring the degree of disease
switch to another TNF blocker than to add methotrexate activity is also crucial to the management of AS. It helps
(Coates et al. 2008). physicians to decide on what particular therapies to initiate.
Once a therapy is initiated, disease activity assessment is
Pearl: Remind the patient regularly that physiotherapy and
required to monitor response. However, assessing disease
exercise are as important as the use of medications designed
activity in AS is more difficult than that in RA because of
to inhibit inflammation.
several factors: there are very few if any swollen joints to
Comment: Patients (and clinicians) sometimes place too count; the correlation with acute phase reactants is often
much faith on the pharmacological side of therapy. Regular poor; and radiographic changes are comparatively slow.
physiotherapy and exercise are also critical features of the Partly because of our ignorance on the factors that cause
therapy for patients with AS. These measures improve and the symptoms and disability of AS, the instruments we use to
maintain muscle strength over time and contribute to the measure disease activity are rather artificial and not based on
preservation of proper posture. the pathophysiology of the disease. For drug trials and for
consideration of TNF blocker treatment, a frequently recom-
Pearl: Talk to the patient about the increased cardiovascular
mended method is the Bath AS Disease Activity Index
risk associated with systemic inflammatory disorders and
(BASDAI) (Garrett et al. 1994). The BASDAI is available at
explain the detrimental effects of smoking on the disease.
http://www.basdai.com/ and is illustrated in Table 7.2.
Comment: Patients with AS have an increased cardiovascu- For patients with established AS, the BASDAI scores are
lar risk because of the chronic inflammation. Moreover, rather stable when studied over a period as long as 5 years
smoking is a risk factor for poor clinical outcomes. Patients (Robertson and Davis 2004). BASDAI scores have been used
should understand the importance of pursuing healthful to assist in making decisions about whether or not TNF
lifestyles and modifying risk factors for cardiovascular inhibitors should be employed in a patient with AS. A
disease. BASDAI score >4.0 has been felt to justify TNF inhibitor
use (Braun et al. 2005; Henderson and Davis 2006). Most
Pearl: Check bone mineral density in time and measure it at
clinical trials of TNF blockers in AS have used this cutoff as
the right location.
one of the entry criteria.
Comment: Approximately half of all patients with AS have An obvious drawback of the BASDAI is that the parame-
abnormal bone mineral density levels. The risk of pathologic ters are dependent entirely upon subjective patient-based
64 J. Braun and D. T. Y. Yu

Table 7.2 Calculating the Bath Ankylosing Spondylitis Disease bacterial species in inflammatory joint diseases. I. Screening of syn-
Activity Index (BASDAI) ovial fluid samples of patients with spondyloarthropathies and other
• How would you describe the overall level of fatigue/tiredness you arthritides. J Rheumatol 1997b; 24(6):1092–1100
have experienced? Braun J, Bollow M, Remlinger G, et al Prevalence of spondylarthropa-
thies in HLA-B27 positive and negative blood donors. Arthritis
• How would you describe the overall level of AS neck, back or hip
Rheum 1998; 41(1):58–67
pain you have had?
Braun J, Brandt J, Listing J, et al Treatment of active ankylosing spon-
• How would you describe the overall level of pain/swelling in dylitis with infliximab - a double-blind placebo controlled multi-
joints other than neck, back or hips you have had? center trial. Lancet 2002; 359:1187–1193
• How would you describe the overall level of discomfort you have Braun J, Davis J, Dougados M, Sieper J, van der Linden S, van der
had from any areas tender to touch or pressure? Heijde D. First update of the international ASAS consensus state-
• How would you describe the overall level of morning stiffness ment for the use of anti-TNF agents in patients with ankylosing
you have had from the time you wake up? spondylitis. Ann Rheum Dis 2005 Aug 11 [Epub ahead of print]
• How long does your morning stiffness last from the time you Braun J, Landewe R, Hermann KG, Han J, Yan S, Williamson P, van
wake up? The scale of the duration of morning stiffness ranges der Heijde D. Major reduction in spinal inflammation in patients
from the minimum score of 0 h to the maximum score of 2 or with ankylosing spondylitis after treatment with infliximab:
more hours. Results of a multicenter, randomized, double-blind, placebo-con-
Calculating the mean BASDAI score: trolled magnetic resonance imaging study. Arthritis Rheum 2006a;
Patients are requested to provide a score for six questions regarding 54(5):1646–1652
their conditions in the preceding week. The scoring can be accom- Braun J, Zochling J, Baraliakos X, et al Efficacy of sulfasalazine in
plished either on a visual analogue scale or in a numerical scale ranging patients with inflammatory back pain due to undifferentiated spondy-
from 0 to 10. loarthritis and early ankylosing spondylitis: A multicentre random-
ized controlled trial. Ann Rheum Dis 2006b
Because questions five and six both address the symptom of morning Breban M, Ravaud P, Claudepierre P, et al Maintenance of infliximab
stiffness, their scores are averaged and then added to the sum of the treatment in ankylosing spondylitis: Results of a one-year random-
other four questions. The total of these five items are then divided by 5 ized controlled trial comparing systematic versus on-demand treat-
to derive the mean BASDAI score. ment. Arthritis Rheum 2008; 58(1):88–97
Brown MA. Human leukocyte antigen-B27 and ankylosing spondylitis.
Int Med J 2007; 37:739–740
evaluations. Alternative strategies for measuring disease Carette S, Graham D, Little H, Rubenstein J, Rosen P. The natural disease
activity in AS have been proposed (Spoorenberg et al. 2005; course of ankylosing spondylitis. Arthritis Rheum 1983; 26(2):186–190
Dougados et al. 2008; Maksymowych et al. 2007). Coates LC, Cawkwell LS, Ng NW, et al Real life experience confirms
sustained response to long-term biologics and switching in ankylos-
ing spondylitis. Rheumatology (Oxford) 2008; 47(6):897–900
Dougados M, Luo MP, Maksymowych WP, et al Evaluation of the
References patient acceptable symptom state as an outcome measure in patients
with ankylosing spondylitis: Data from a randomized controlled
trial. Arthritis Rheum 2008; 59(4):553–560
Appel H, Kuhne M, Spiekermann S, Ebhardt H, Grozdanovic Z, Garrett S, Jenkinson T, Kennedy LG, et al A new approach to defining
Kohler D, et al Immunohistologic analysis of zygapophyseal joints disease status in ankylosing spondylitis: The bath ankylosing spon-
in patients with ankylosing spondylitis. Arthritis Rheum 2006; dylitis disease activity index. J Rheumatol 1994; 21(12):2286–2291
54(9):2845–2851 Haibel H, Brandt HC, Song IH, et al No efficacy of subcutaneous meth-
Baraliakos X, Listing J, Brandt J, Haibel H, Rudwaleit M, Sieper J, et al otrexate in active ankylosing spondylitis: A 16-week open-label
Radiographic progression in patients with ankylosing spondylitis trial. Ann Rheum Dis 2007; 66(3):419–421
after 4 yrs of treatment with the anti-TNF-alpha antibody inflix- Heiberg MS, Koldingsnes W, Mikkelsen K, et al The comparative one-
imab. Rheumatol (Oxford) 2007a; 46(9):1450–1453 year performance of anti-tumor necrosis factor alpha drugs in
Baraliakos X, Listing J, Rudwaleit M, Haibel H, Brandt J, Sieper J, patients with rheumatoid arthritis, psoriatic arthritis, and ankylosing
Braun J. Progression of radiographic damage in patients with anky- spondylitis: Results from a longitudinal, observational, multicenter
losing spondylitis - Defining the central role of syndesmophytes. study. Arthritis Rheum 2008; 59(2):234–240
Ann Rheum Dis 2007; 66(7):910–5 Henderson C, Davis JC. Drug insight: Anti-tumor-necrosis-factor ther-
Baraliakos X, Listing J, Rudwaleit M, Sieper J, Braun J. The relation- apy for ankylosing spondylitis. Nat Clin Pract Rheumatol 2006;
ship between inflammation and new bone formation in patients with 2(4):211–218
ankylosing spondylitis. Arthritis Res Ther 2008 Sep 1; 10(5):R104 Karberg K, Zochling J, Sieper J, Felsenberg D, Braun J. Bone loss is
Braun J, Sieper J. Ankylosing spondylitis. Lancet 2007; 369:1379–1390 detected more frequently in patients with ankylosing spondylitis
Braun J, Bollow M, Seyrekbasan F, et al Computed tomography guided with syndesmophytes. J Rheumatol 2005; 32(7):1290–1298
corticosteroid injection of the sacroiliac joint in patients with spon- Landewé R, Dougados M, Mielants H et al Physical function in anky-
dyloarthropathy with sacroiliitis: Clinical outcome and followup by losing spondylitis is independently determined by both disease
dynamic magnetic resonance imaging. J Rheumatol 1996; 23(4): activity and radiographic damage of the spine. Ann Rheum Dis
659–664 2009; 68(6):863–7
Braun J, Tuszewski M, Ehlers S, et al Nested polymerase chain reaction Li EK, Griffith JF, Lee VW, et al Short-term efficacy of combination
strategy simultaneously targeting DNA sequences of multiple bac- methotrexate and infliximab in patients with ankylosing spondylitis:
terial species in inflammatory joint diseases. II. Examination of sac- A clinical and magnetic resonance imaging correlation. Rheumatology
roiliac and knee joint biopsies of patients with spondyloarthropathies (Oxford) 2008 Sept; 47(9):1358–1363 [Epub 2008 Jun 23]
and other arthritides. J Rheumatol 1997a; 24(6):1101–1105 Lopez-Larrea C, Sujirachato K, Mehra NK, et al HLA-B27 subtypes in
Braun J, Tuszewski M, Eggens U, et al Nested polymerase chain reac- Asian patients with ankylosing spondylitis. Evidence for new asso-
tion strategy simultaneously targeting DNA sequences of multiple ciations. Tissue Antigens 1995; 45(3):169–176
7 Ankylosing Spondylitis 65

Lories RJ, Derese I, de Bari C, Luyten FP. Evidence for uncoupling of paper patients by expert opinion including uncertainty appraisal.
inflammation and joint remodeling in a mouse model of spondylar- Ann Rheum Dis 2009b; 68(6):770–6
thritis. Arthritis Rheum 2007; 56(2):489–497 Rudwaleit M, van der Heijde D, Landewé R et al The development of
Mader R. Diffuse idiopathic skeletal hyperostosis: Time for a change. Assessment of SpondyloArthritis international Society classifica-
J Rheumatol 2008; 35(3):377–379 tion criteria for axial spondyloarthritis (part II): validation and final
Maertens M, Mielants H, Verstraete K, Veys EM. Simultaneous occur- selection. Ann Rheum Dis 2009c; 68(6):777–83
rence of diffuse idiopathic skeletal hyperostosis and ankylosing spon- Saraux A, Guillemin F, Guggenbuhl P, et al Prevalence of spondyloar-
dylitis in the same patient. J Rheumatol 1992; 19(12):1978–1983 thropathies in France: 2001. Ann Rheum Dis 2005; 64(10):1431–1435
Maksymowych WP, Jhangri GS, Fitzgerald AA, et al A six-month ran- Song IH, Poddubnyy DA, Rudwaleit M, Sieper J. Benefits and risks of
domized, controlled, double-blind, dose-response comparison of ankylosing spondylitis treatment with nonsteroidal antiinflamma-
intravenous pamidronate (60 mg versus 10 mg) in the treatment of tory drugs. Arthritis Rheum 2008; 58(4):929–938
nonsteroidal antiinflammatory drug-refractory ankylosing spondyli- Spoorenberg A, van der Heijde D, de Klerk E, et al Relative value of eryth-
tis. Arthritis Rheum 2002; 46(3):766–773 rocyte sedimentation rate and C-reactive protein in assessment of disease
Maksymowych WP, Richardson R, Mallon C, et al Evaluation and vali- activity in ankylosing spondylitis. J Rheumatol 1999; 26(4):980–984
dation of the patient acceptable symptom state (PASS) in patients Spoorenberg A, van Tubergen A, Landewe R, et al Measuring disease
with ankylosing spondylitis. Arthritis Rheum 2007; 57(1):133–139 activity in ankylosing spondylitis: Patient and physician have differ-
Muche B, Bollow M, François RJ, Sieper J, Hamm B, Braun J. Which ent perspectives. Rheumatology (Oxford) 2005; 44(6):789–795
anatomical structures are involved in early and late sacroiliitis in Underwood MR, Dawes P. Inflammatory back pain in primary care. Br
spondylo¬arthritis – a detailed analysis by contrast enhanced mag- J Rheumatol 1995; 34(11):1074–1077
netic resonance imaging. Arthritis Rheum 2003; 48(5):1374–1384 van der Heijde D, Kivitz A, Schiff MH, et al Efficacy and safety of
Neidhart M, Baraliakos X, Seemayer C, et al Expression of cathepsin K adalimumab in patients with ankylosing spondylitis: Results of a
and MMP-1indicate persistent osteodestructive activity in long- multicenter, randomized, double-blind, placebo-controlled trial.
standing ankylosing spondylitis. Ann Rheum Dis 2008 Aug 4 [Epub Arthritis Rheum 2006; 54(7):2136–2146
ahead of print] van der Heijde D, Landewe R, Einstein S, et al Radiographic progres-
Olivieri I, D’Angelo S, Cutro MS, et al Diffuse idiopathic skeletal hyper- sion of ankylosing spondylitis after up to two years of treatment
ostosis may give the typical postural abnormalities of advanced anky- with etanercept. Arthritis Rheum 2008; 58(5):1324–1331
losing spondylitis. Rheumatology (Oxford) 2007; 46(11):1709–1711 van der Linden S, Valkenburg HA, Cats A. Evaluation of diagnostic
Robertson LP, Davis MJ. A longitudinal study of disease activity and criteria for ankylosing spondylitis. A proposal for modification of
functional status in a hospital cohort of patients with ankylosing the New York criteria. Arthritis Rheum 1984; 27(4):361–368
spondylitis. Rheumatology (Oxford) 2004; 43(12):1565–1568 Ward MM, Reveille JD, Learch TJ, Davis JC Jr, Weisman MH.
Rudwaleit M, Metter A, Listing J, Sieper J, Braun J. Inflammatory back Occupational physical activities and long-term functional and radio-
pain in ankylosing spondylitis - a reassessment of the clinical history graphic outcomes in patients with ankylosing spondylitis. Arthritis
for classification and diagnosis. Arthritis Rheum 2006; 54(2): Rheum 2008; 59(6):822–832
569–578 Zink A, Braun J, Listing J, Wollenhaupt J. Disability and handicap in
Rudwaleit M, Haibel H, Baraliakos X et al. The early disease stage in rheumatoid arthritis and ankylosing spondylitis–results from the
axial spondylarthritis: results from the German Spondyloarthritis German rheumatological database. German Collaborative Arthritis
Inception Cohort. Arthritis Rheum 2009a; 60(3):717–27 Centers. J Rheumatol 2000; 27(3):613–622
Rudwaleit M, Landewé R, van der Heijde D et al The development of Zochling J, van der Heijde D, Burgos-Vargas R, et al ASAS/EULAR
Assessment of SpondyloArthritis international Society classifica- recommendations for the management of ankylosing spondylitis.
tion criteria for axial spondyloarthritis (part I): classification of Ann Rheum Dis 2006 Aug 26 [Epub ahead of print]
 Psoriatic Arthritis
8
Christopher Ritchlin, Elinor Mody, Philip Mease,
and Dafna D. Gladman

8.1 Overview of Psoriatic Arthritis

› Psoriatic arthritis (PsA) is a distinctive category of

inflammatory arthritis that complicates cutaneous
psoriasis.
› The prevalence of PsA in the general population is
estimated to be between 0.3% and 1%, only slightly
less common than that of rheumatoid arthritis (RA).
› PsA is one of the “seronegative spondyloarthropa-
thies”, along with ankylosing spondylitis (AS), reac-
tive arthritis, and the arthropathy of inflammatory
bowel disease. As a seronegative disorder, PsA is not
associated with autoantibodies such as rheumatoid fac-
a
tor or antibodies to cyclic citrullinated peptides.
› A variety of clinical patterns of PsA are recognized:
distal joint disease, oligoarthritis, a symmetrical pol-
yarthritis that resembles RA, and spondylitis. Some
patients present with combinations of these features:
e.g., oligoarthritis plus spondylitis.
› These clinical patterns are not necessarily stable over
time. For example, distal joint disease can evolve into
polyarthritis.
› The most common pattern of joint involvement at pre-
sentation is an asymmetric oligoarthritis. This pattern
accounts for about 70% of PsA cases.
b

Myth: PsA is RA with psoriasis.

Reality: The principal differences between PsA and RA are
summarized in Table 8.1. Cutaneous psoriasis occurs in up to
3% of the population, and RA in about 1%. Thus, the co-occur-
rence of psoriasis and RA is expected in about three people in
10,000. Nail lesions, particularly onycholysis and more than
20 nail pits per person, are associated with PsA (Fig. 8.1a and
8.1b) (Eastmond andWright 1979). The presence of nail pits
c
not only helps distinguish PsA from RA in some circum-
stances, it also identifies patients with psoriasis who are more Fig. 8.1 Nail changes in psoriasis (a) Nail pitting (b) Onycholysis.
likely to develop PsA, particularly arthritis of the distal inter- Note that some fingernails are affected and others are not (c) A close-up
phalangeal (DIP) joints (Fig. 8.2) (Gladman et al. 1986). of onycholysis (figures courtesy of Dr. John Stone)

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 67

DOI:10.1007/978-1-84800-934-9_8, © Springer Science + Business Media B.V. 2009
68 C. Ritchlin et al.

b
Fig. 8.2 DIP joint arthritis in PsA (note the DIP joint of the index fin-
ger). This patient also had nail pitting. DIP involvement and nail pitting
often occur together in PsA (figure courtesy of Dr. John Stone)

Table 8.1 Contrasts between PsA and RA

Disease feature PsA RA
Nail lesions Pitting Bywaters’ lesions
Onycholysis Splinter hemorrhages
Joint pattern in Asymmetry Symmetry
hands and DIP involvement MCPs/PIPs/DIPs
wrists common Ulnar styloid
“Ray” pattern involvement
Dactylitis (“sau-
sage” digits)
Spinal disease Syndesmophytes C1–C2 involvement
Eye disease Anterior and Episcleritis
posterior uveitis Keratoconjunctivitis
Scleritis
Peripheral ulcerative Fig. 8.3 Dactylitis: A “Sausage Digit” in PsA (a) A 21-year-old woman
keratitis presented with sudden onset of pain and swelling in the right fifth toe,
Other extra-artic- Cutaneous psoriasis Nodules but denied pain in other joints. She had a small patch of scalp psoriasis.
ular features Bluish discoloration Interstitial lung disease (b) A radiograph of the right fifth toe taken 3 months after presentation
over joints Pericarditis revealed fusion of the DIP joint (with kind permission from the ameri-
Enthesitis can college of Rheumatology)
Systemic vasculitis
Urethritis
Inflammatory bowel
Pearl: PsA tends to occur in a “ray” distribution.
disease
Comment: PsA has a tendency to affect all joints in a particu-
lar digit: e.g., the MCP (metacarpophalangeal), PIP (proximal
interphalangeal), and DIP joints of a finger (Fig. 8.3). This is
PsA is not associated with atlanto-axial subluxation (C1–
termed dactylitis (from the Greek word “daktyos” for finger),
C2), the classic form of spinal involvement in RA. About
but known more colloquially as a “sausage digit” (Brockbank
half of all PsA patients have some degree of spinal inflam-
et al. 2005). The sausage appearance of the affected digit likely
mation (Gladmann 2007), but PsA in the cervical spine usu-
results from the involvement not only of the joints by synovi-
ally manifests itself as syndesmophytes rather than erosions
tis, but also by tenosynovitis, particularly of flexor tendons.
of the dens (C2).
Even though all of the joints in one finger might be affected by
PsA affects males and females almost equally. This con-
PsA, the adjacent fingers can be entirely free of arthritis.
trasts with RA, in which a predilection for women exists.
In some patients, dactylitis becomes chronic such that the
Patients with PsA often have a bluish–purplish discoloration
finger remains swollen but is no longer painful or red. Chronic
over their inflamed joints (Jajic 2001), a finding that is not
dactylitis may respond less well to anti-inflammatory thera-
common in RA. Patients with PsA generally have less joint
pies, which may need to be modified accordingly.
tenderness than patients with RA and are also less likely to
have secondary fibromyalgia (Buskila et al. 1992). Pearl: Enthesitis can help differentiate PsA from RA.
8 Psoriatic Arthritis 69

Comment: Forty percent of patients with PsA have enthesitis affects the DIP joints of the hands and feet; i.e., the joints that
or inflammation at insertion of tendons into bones (Fig. 8.4) are not usually affected by RA. Furthermore, the distribution
(Fernández-Sueiro et al. 2007). This feature is atypical of of joint disease in PsA tends to be asymmetric, but RA is char-
RA and strongly suggests the presence of a seronegative acteristically symmetric, with a tendency to involve the MCP
spondyloarthropathy. The most common sites of enthesitis in and PIP joints bilaterally (Gladman 1987).
PsA are the plantar fascia, the Achilles tendon, the insertions
Myth: There is a direct relationship between the activity of
of tendons at the knees and shoulders, and in the pelvic
skin and joint disease in PsA.
bones. The finding of enthesitis in association with cutane-
ous psoriasis is sufficient for the diagnosis of PsA. Reality: Some investigators suggest that there is a relationship
between the severity of psoriasis and the prevalence of PsA;
Pearl: PsA and RA have distinctive extra-articular features. i.e., the more severe the skin disease, the higher the likelihood
Comment: Rheumatoid nodules do not occur in PsA. Indeed, of inflammatory joint disease (Gelfand et al. 2005). However,
if a patient with a diagnosis of PsA has a rheumatoid nodule, once a patient has developed PsA, there does not appear to be
the diagnosis should be called into question. Other extra-artic- a direct relationship between the flares of skin and joint disease
ular features of PsA and RA also contrast strikingly: urethritis (Jones et al. 1994; Cohen et al. 1999; Elkayam et al. 2000).
and inflammatory bowel disease symptoms are common in Some medications employed by rheumatologists exacer-
PsA, but interstitial lung disease, myocarditis, and pericarditis bate cutaneous psoriasis. These include non-steroidal anti-
are far more characteristic of RA. The extra-articular differ- inflammatory drugs and glucocorticoids. Moreover, not all
ences are highlighted most often in the eye, where anterior disease-modifying antirheumatic drugs (DMARDs) used
and less commonly posterior uveitis typifies PsA but episc- potentially by rheumatologists to treat joint disease are effec-
leritis, scleritis, keratitis, and keratoconjunctivitis are much tive for skin disease, nor are all therapies for cutaneous pso-
more likely to complicate RA. riasis efficacious in PsA. When cutaneous psoriasis is severe
and fails to respond to treatments directed primarily toward
Pearl: Inflammatory arthritis in PsA can occur before cuta- the arthritis, consultation and co-management with a derma-
neous psoriasis has become evident. tologist is appropriate.
Comment: About 25% of patients with cutaneous psoriasis Myth: PsA is AS with psoriasis.
develop PsA. The majority of patients with PsA have psoriasis
either simultaneously or before the onset of arthritis. However, Reality: PsA and AS are both seronegative spondyloarthrop-
about 15% of PsA patients have arthritis that precedes their athies. The seronegative spondyloarthropathies affect the
cutaneous disease. In those patients, the pattern of joint spine and share a number of extra-articular manifestations
involvement is critical to arriving at the correct diagnosis. PsA (Helliwell 2004). About 10% of patients with AS have pso-
riasis, but important clinical and radiologic differences exist
SC SF between these disorders (Gladman 1998).
SM EF
Moll and Wright described a number of different patterns of
ET joint disease in PsA, including an arthritis that resembles RA
(symmetrical, polyarticular, small joint), DIP-joint predomi-
nant, oligoarticular, spondylitis, and arthritis mutilans (Moll
and Wright 1973). These categories encompass both periph-
eral and axial joint involvement. The majority of PsA patients
have peripheral disease, and about 50% have spondylitis. Less
than 5% of patients with PsA have isolated spondylitis.
Although the types of extra-articular manifestations
observed in PsA and AS are similar, their frequencies differ
substantially. The most obvious example, of course, is the skin,
whereas essentially 100% of patients with PsA have cutaneous
psoriasis and more than 80% have nail lesions; only 10% of
AS patients have psoriasis. Iritis (anterior uveitis) occurs in
more than 30% of AS patients but in a far lower percentage of
Fig. 8.4 Schematic representation of the synovial entheseal complex PsA patients. Urethritis is more common in AS than PsA.
(SEC). An SEC of an extensor tendon (ET) in an interphalangeal joint,
as seen in a sagittal section of a finger. The synovium lines the deep Pearl: Spinal involvement presents later in life in PsA than it
surface of the tendon except in the region of the sesamoid fibrocartilage does in AS.
(SF) that, in a flexed finger, is compressed against articular cartilage.
SM = synovial membrane; SC=synovial cavity; EF=enthesis fibrocarti- Comment: Patients with AS present with back pain in their
lage (from McGonagle 2007. With kind permission from Wiley) late teens or early 20s. In contrast, those with psoriatic
70 C. Ritchlin et al.

spondylitis seldom present before their 30s or 40s. Moreover, Buskila D, Langevitz P, Gladman DD, Urowitz S, Smythe H. Patients
the presenting feature of spondylitis in PsA is limitation of with rheumatoid arthritis are more tender than those with psoriatic
arthritis. J Rheumatol 1992; 19:1115–1119
movement rather than back pain (Gladman 2009). Just as the Cohen, MR, Reda, DJ, Clegg, DO. Baseline relationships between pso-
peripheral arthritis of PsA is less symptomatic than that of riasis and psoriatic arthritis: Analysis of 221 patients with active
RA, so too is the spinal disease of PsA less symptomatic than psoriatic arthritis. J Rheumatol 1999; 26:1752–1756
that of AS (Gladman et al. 1993; Helliwell et al. 1998; Coulton BL, Thomson K, Symmons DP, Popert AJ. Outcome in patients
hospitalised for psoriatic arthritis. Clin Rheumatol 1989; 8:261–265
O’Shea et al. 2006). Patients with PsA demonstrate more Eastmond CJ, Wright V. The nail dystrophy of psoriatic arthritis. Ann
spinal mobility on average than do AS patients, perhaps Rheum Dis 1979; 38:226–228
because the syndesmophytes in PsA are asymmetric, allow- Eder L, Zisman D, Barzilai M, Laor A, Rahat M, Rozenbaum M, et al
ing for a greater preservation of spinal mobility. Subclinical atherosclerosis in psoriatic arthritis: A case-control
study. J Rheumatol 2008; 35:877–782
Myth: PsA is a mild disease that does not require aggressive Elkayam O, Ophir J, Yaron M, Caspi D. Psoriatic arthritis: Interre-
lationships between skin and joint manifestations related to onset,
intervention.
course and distribution. Clin Rheumatol 2000;19: 301–305
Reality: This notion was prevalent following the early Fernández-Sueiro JL, Willisch A, Pinto J, et al Prevalence and location
of enthesitis in ankylosing spondylitis and psoriatic arthritis. Ann
descriptions of PsA (Shbeeb et al. 2000). Wright observed Rheum Dis 2007; 66(Suppl II):99
that PsA tends to be less severe than RA (Wright 1959). A Gelfand JM, Gladman DD, Mease PJ, et al Epidemiology of psoriatic
subsequent study reported less mortality and greater func- arthritis in the United States population. J Amer Acad Dermatol
tional capacity among PsA patients compared with RA 2005; 53:573–577
Gladman DD. Clinical aspects of spondyloarthropathies. Am J Med Sci
(Coulton et al. 1989). 1998; 316:234–238
More recent studies have confirmed the high prevalence Gladman DD. Axial disease in psoriatic arthritis. Curr Rheumatol Rep
of severe, destructive, and disabling disease in PsA, and an 2007; 9:455–460
increased mortality risk for patients with this diagnosis Gladman DD, Anhorn KB, Schachter RK, Mervart H. HLA antigens in
psoriatic arthritis. J Rheumatol 1986; 13:586–592
(Gladman 2007; Sokoll and Helliwell 2001; McHugh et al. Gladman DD, Shuckett R, Russell ML, Thorne JC, Schachter RK.
2003; Kane et al. 2003; Queiro-Silva et al. 2003; Bond et al. Psoriatic arthritis - clinical and laboratory analysis of 220 patients.
2007; Ali et al. 2007). A large proportion of PsA patients Quart J Med 1987; 62:127–141
develop joint erosions and disability within the first 2 years Gladman DD, Brubacher B, Buskila D, Langevitz P, Farewell VT.
Differences in the expression of spondyloarthropathy: A compari-
of disease. In many patients, particularly those with the aptly son between ankylosing spondylitis and psoriatic arthritis. Genetic
named “arthritis mutilans,” PsA is a more severe disease and gender effects. Clin Invest Med 1993; 16:1–7
even than is RA. Early, aggressive treatment of PsA may pre- Gladman DD, Farewell VT, Husted J, Wong K. Mortality studies in
vent untoward outcomes (Kane et al. 2003; Ali et al. 2007). psoriatic arthritis. Results from a single centre. II. Prognostic indi-
cators for mortality. Arthritis Rheum 1998; 41:1103–1110
Myth: PsA is basically a disease of skin and joints. Gladman DD, Ang M, Su L, Tom BDM, Schentag CT, Farewell VT.
Cardiovascular morbidity in psoriatic arthritis (PsA). Ann Rheum
Reality: PsA patients are at an increased risk of both coro- Dis Online. Accessed December 8, 2008
nary artery disease and the metabolic syndrome, as are Gladman DD, Helliwell PS, Mease PJ. GRAPPA at the European
League Against Rheumatism (EULAR) 2008. J Rheumatol 2009;
patients with RA and systemic lupus erythematosus (Bruce 36:656–658
et al. 2000; Gladman et al. 2008; Eder et al. 2008; Tam et al. Helliwell PS. Relationship of psoriatic arthritis with the other spondy-
2008a, b). Some data suggest that the underlying inflamma- loarthropathies. Curr Opin Rheumatol 2004; 16:344–349
tory milieu in PsA is responsible for these co-morbidities Helliwell PS, Hickling P, Wright V. Do the radiological changes of clas-
sic ankylosing spondylitis differ from the changes found in the
(Tam et al. 2008a, b). spondylitis associated with inflammatory bowel disease, psoriasis,
and reactive arthritis? Ann Rheum Dis 1998; 57:135–140
Jajic J. Blue-coloured skin over involved joints in psoriatic arthritis.
References Clin Rheumatol 2001; 20:304–305
Jones SM, Armas JB, Cohen MG, et al Psoriatic arthritis: Outcome of
disease subsets and relationship of joint disease to nail and skin dis-
Ali Y, Tom BDM, Schentag CT, Farewell VT, Gladman DD. Improved ease. Br J Rheumatol 1994; 33:834–839
survival in psoriatic arthritis (PsA) with calendar time. Arthritis Kane D, Stafford L, Bresniham B, Fitzgerald O. A prospective, clinical
Rheum 2007; 56:2708–2714 and radiological study of early psoriatic arthritis: an early synovitis
Bond SJ, Farewell VT, Schentag CT, Gladman DD. Predictors for clinic experience Rheumatol 2003; 42:1460–1468
radiological damage in Psoriatic Arthritis: Results from a single McGonagle, D. The concept of a synovio-enthesial complex and its
centre. Ann Rheum Dis 2007; 66:370–376 implications for understanding joint inflammation and damage in
Brockbank J, Stein M, Schentag CT, Gladman DD. Dactylitis in psori- psoriatic arthritis and beyond. Arthritis Rheum 2007; 56:8
atic arthritis (PsA): A marker for disease severity? Ann Rheum Dis McHugh NJ, Balachrishnan C, Jones SM. Progression of peripheral
2005; 62:188–190 joint disease in psoriatic arthritis: A 5-yr prospective study.
Bruce IN, Schentag C, Gladman DD. Hyperuricemia in psoriatic arthri- Rheumatology (Oxford) 2003; 42:778–783
tis (PsA) does not reflect the extent of skin involvement. J Clin Moll JM, Wright V. Psoriatic arthritis. Semin Arthritis Rheum 1973;
Rheumatol 2000; 6:6–9 3:55–78
8 Psoriatic Arthritis 71

O’Shea FD, Chandran V, Toloza SMA, Schentag CT, Inman RD, Sokoll KB, Helliwell PS. Comparison of disability and quality of life in
Gladman DD. Does axial disease exhibit a different phenotype in rheumatoid and psoriatic arthritis. J Rheumatol 2001; 28: 1842–1846
primary ankylosing spondylitis versus psoriatic arthritis? Arthritis Tam LS, Shang Q, Li EK, Tomlinson B, Chu TT, Li M, et al Subclinical
Rheum 2006; 54:4091 carotid atherosclerosis in patients with psoriatic arthritis. Arthritis
Queiro-Silva R, Torre-Alonso JC, Tinture-Eguren T, et al A polyarticu- Rheum 2008a; 59:1322–1331
lar onset predicts erosive and deforming disease in psoriatic arthri- Tam LS, Tomlinson B, Chu TT, et al Cardiovascular risk profile of
tis. Ann Rheum Dis 2003; 62:68–70 patients with psoriatic arthritis compared to controls–the role of
Shbeeb M, Uramoto KM, Gibson LE, O’Fallon WM, Gabriel SE. The inflammation. Rheumatology (Oxford) 2008b; 47:718–723
epidemiology of psoriatic arthritis in Olmsted County, Minnesota, Wright V. Rheumatism and psoriasis a re-evaluation. Am J Med 1959;
USA, 1982–1991. J Rheumatol 2000; 27:1247–1250 27:454–462.
 Reactive Arthritis
9
Robert D. Inman and Andrew Keat

Pearl: Always consider ReA when confronted with monoar-

9.1 Overview of Reactive Arthritis thritis in a young adult.
Comment: ReA most often affects adults between 18 and 40
› In reactive arthritis (ReA), exposure to an infectious
years of age. This disorder presents as an oligoarthritis, with
agent leads to the development of an inflammatory
pain and swelling only a few joints (often only one). The knee,
arthritis and other characteristic clinical findings.
ankle, or one of the metatarsophalangeal joints is usually
However, this syndrome occurs in the absence of an
affected first (Fig. 9.1). When the patient is a child, ReA must
ongoing infectious process.
be distinguished from juvenile idiopathic arthritis, particularly
› About 50% of ReA and undifferentiated oligoarthritis
the oligoarticular form. Among adult patients, ReA must be
cases can be attributed to a specific pathogen by a com-
differentiated from microcrystalline disorders. Septic arthritis
bination of culture and serology. The predominant organ-
must always be considered and excluded in any patient who
isms are Chlamydia, Salmonella, Shigella, Yersinia, and
presents with a monoarthritis.
Campylobacter species.
Symptoms of ReA may develop before other features of
› The frequency of ReA following exposure to potential
the infection that triggered the inflammation become
etiologic agents is between 3% and 10%.
apparent. Thus, a clear history relating to other features of a
› ReA characteristically involves the joints of the lower
spondyloarthropathy (past or present) and exposure to and
extremities in an asymmetric, oligoarticular pattern.
symptoms of genitourinary (GU) or enteric infection should
› The presence of HLAB27 increases disease suscepti-
be sought.
bility, but is neither sufficient nor necessary for ReA to
occur. Individuals who are HLA-B27 positive tend to Pearl: Acute ReA may mimic septic arthritis.
have more severe and longer episodes of ReA.
Comment: Most often ReA is associated with relatively mild
› A dactylitis (“sausage digit”) pattern of arthritis in the
systemic symptoms and characteristic multi-site disease. In a
feet is typical of ReA, as it is of psoriatic arthritis. ReA
has a predilection for joints of the lower extremities,
usually in an asymmetric, oligoarticular pattern.
› Enthesopathy and anterior uveitis often occur in ReA.
› Cutaneous manifestations of ReA include:
- Oral ulcers, typically painless
- Nail dystrophy
- Keratoderma blenorrhagicum, a papulosquamous
rash that affects the palms and soles
- Circinate balanitis, characterized by shallow ulcers
on the glans or the shaft of the penis

Fig. 9.1 Monoarthritis of the left knee in a patient with post-streptococcal

ReA (figure courtesy of Dr. John Stone)

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 73

DOI:10.1007/978-1-84800-934-9_9, © Springer Science + Business Media B.V. 2009
74 R. D. Inman and A. Keat

few patients, however, monoarthritis is associated with high

fever and marked malaise. Yersiniosis and salmonellosis are
particularly likely to be associated with such systemic fea-
tures. Septic arthritis may occur in Campylobacter and
Salmonella infections. The culture of organisms from a joint
indicates treatment for a septic arthritis, even if other clinical
features are highly consistent with ReA.
Pearl: The only way to differentiate definitively between ReA
and septic arthritis is through culture of the synovial fluid.
Comment: Certain organisms, e.g., Yersinia and Salmonella,
can cause both septic arthritis and ReA. Thus, sampling and
culture of the synovial fluid is mandatory in the assessment
of a patient with possible ReA. Differentiating gonococcal
arthritis from post-GU ReA is a common challenge in this Fig. 9.2 Left Achilles’ tendon swelling, consistent with a ReA-related
regard. Recall that joint cultures for Neisseria gonococcus enthesopathy (figure courtesy of Dr. John Stone
are frequently negative, even in the setting of disseminated
gonococcal infection. This heightens the difficulty in dis- Reality: In adult patients, post-streptococcal ReA is not
criminating between gonococcal infections and ReA. accompanied by carditis, chorea, erythema marginatum, or
subcutaneous nodules (Barash et al. 2008). Post-streptococcal
Myth: ReA responds to antibiotic therapy. ReA and acute rheumatic fever are two different entities. The
implications of this are significant; for example, long-term
Reality: Despite the role of infections in triggering ReA, cul- antibiotic prophylaxis is not indicated in adults with post-
tures of synovial fluid are sterile. There is no evidence that streptococcal ReA.
antibiotics alter the long-term course of ReA that follows a
gastrointestinal infection, and no indication that long-term Pearl: No inflammation, no ReA!
treatment with antibiotics is effective for shortening episodes Comment: ReA is an inflammatory condition associated with
of ReA. There remains some uncertainty about the impact of genuine signs of joint or enthesial swelling. Tenderness or
antibiotics on the course of ReA after an infection of the GU aching in the joints or periarticular tissues without hard signs
tract caused by Chlamydia. However, no compelling data sup- of inflammation is not sufficient for the diagnosis. Similarly,
port the prolonged use of antibiotics beyond the course required non-specific urethral symptoms in men must not be attrib-
to eradicate the inciting infection (Hamdulay et al. 2006). uted to ReA if they are not associated with musculoskeletal
symptoms that are incompatible with ReA. No inflamma-
Pearl: Dysuria and a genital rash can follow gastrointestinal tion, no ReA.
infections as well as GU infections.
Myth: HLA-B27 typing is a helpful diagnostic test for ReA.
Comment: When GU complaints occur concurrently with an
inflammatory arthritis, clinicians may assume that a sexually- Reality: About 50% of people with ReA carry the HLA-B27
transmitted disease has been the inciting event. However, gene, with the figure being somewhat higher in certain popu-
urethritis is perfectly compatible with the syndrome of ReA lations. Thus, negative HLA-B27 genotyping does not exclude
that follows a gastrointestinal infection (Ahvonen et al. the diagnosis of ReA. The diagnosis must be founded upon
1969), and circinate balanitis may occur without an anteced- compatible clinical features and compelling evidence of an
ent infection of the urogenital tract. A sexually-transmitted infection.
disease need not be invoked to explain GU features. Pearl: The presence or absence of HLA-B27 affects clinical
Pearl: The enthesitis of ReA can be the dominant feature of disease phenotype.
the disease. Comment: HLA-B27 positivity is typical of patients who
Comment: Enthesitis, particularly Achilles tendonitis and have persistent disease and is nearly universal among patients
plantar fasciitis, can be profoundly disabling in the acute with ReA and uveitis. In addition, 90% of patients with
phase. Always examine patients with suspected ReA stand- radiographic evidence of sacroiliitis are HLA-B27 positive.
ing, to search for Achilles’ tendonitis (Fig. 9.2). Enthesitis In patients with ReA in sub-Saharan Africa, an aggressive
overshadows inflammatory joint symptoms in some patients. form of disease is observed. These patients are generally
HLA-B27 negative, but often are infected with the human
Myth: Post-streptococcal ReA is really acute rheumatic fever. immunodeficiency virus (Leirisalo-Repo 1998).
9 Reactive Arthritis 75

Myth: When ReA is suspected, examination of the joint fluid re-infection with enteric bacteria – even when of the same
by PCR may reveal the infectious cause. species as that which caused the initial episode – does not
always lead to a recurrence of ReA.
Reality: Inflamed joints contain a wide range of bacterial pro-
teins and some bacterial DNA. Some studies have revealed Myth: ReA is precipitated only by gastrointestinal and GU
specific bacterial DNA (Hannu et al. 2006) and even RNA, infections.
indicating probable viability of the bacteria presumed to have Reality: The commonly recognized arthritogenic pathogens are
caused the arthritis (Hannu et al. 2006). However, bacterial Salmonella, Shigella, Campylobacter, Yersinia, and Chlamydia.
DNA from many different genera have been identified within However, ReA may occur after infection by any portal of entry,
joint material (Kempsell et al. 2000), and it remains unclear and may be caused by a large number of other organisms, as
how (or if) to ascribe causality to any of these findings. This is well. As an example, cases of ReA following Clostridium dif-
an exciting space to be watched, but not a helpful clinical test. ficile infections are well described.
Pearl: The uveitis of ReA is anterior, unilateral, and highly In some patients whose clinical features are consistent with
symptomatic. ReA, there is no antecedent history of infection. This suggests
that this syndrome can result either from subclinical infec-
Comment: ReA is associated with an anterior uveitis that tions or from other environmental (non-infectious) triggers.
tends to afflict one eye at a time. The contralateral eye may
be involved in a subsequent disease flare. The finding of con- Myth: The interval between antecedent infection and ReA is
current bilateral disease or disease in the posterior pole of the 1 week or less.
eye strongly invokes other diagnoses, e.g., sarcoidosis or Reality: Although an interval of 1–2 weeks after the inciting
Behcet’s disease. infection is typical for the appearance of ReA, this time period
Patients with the anterior uveitis of ReA have pronounced extends up to 4 weeks. Patients with joint symptoms and other
photophobia, eye pain, ocular erythema, and tearing. Shining clinical features compatible with ReA must therefore be que-
a flashlight in the contralateral (uninvolved) eye leads to pain ried closely about more “remote” occurrences of infections.
in the involved eye because of the consensual light response,
which leads to pupillary narrowing in the inflamed eye and Myth: ReA is a chronic disease similar to rheumatoid arthritis.
an increase in ocular discomfort. In addition to topical gluco- Reality: ReA usually consists of either a single attack that
corticoid eye drops, a mydriatic agent is essential to dilate runs its course within a matter of months or recurrent epi-
the involved eye and prevent the formation of synechiae sodes of arthritis that last weeks to months between longer
between the pupil and the lens. periods of remission. A chronic, destructive, disabling arthri-
Pearl: An episode of ReA may be followed by a prolonged tis evolves in only a minority of patients.
period of non-inflammatory arthralgia. Pearl: Not all inflammatory back pain symptoms in ReA are
Comment: Most episodes of ReA resolve within 12 months caused by spondylitis, sacroiliitis, or both.
(Keat 1983), but the persistence of symptoms following an Comment: Some symptoms of low back pain in ReA are
episode of inflammatory disease is not uncommon in ReA. caused by enthesitis that involves the pelvic girdle. The
The explanation for this phenomenon is not clear though it pelvis, in addition to the Achilles tendon and the plantar fas-
may be a source of great anxiety for patients. Indeed, a psy- cia, is another common site of involvement by enthesopathy.
chological element may complicate the clinical picture, requir- Unrelated, non-inflammatory causes of back pain should not
ing great reassurance from the clinician. If all evidence of be overlooked even during an episode of ReA.
inflammation has resolved, there is no role for antibiotic treat-
ment or disease-modifying antirheumatic drugs (DMARDs). Myth: DMARDs should be avoided in ReA.
Reality: The typical course of ReA is a period of greatest
Pearl: Recurrence of ReA is common.
joint activity over 8–12 weeks. Most patients respond satis-
Comment: Recurrence of sexually-acquired ReA is common factorily to sustained doses of NSAIDs. If the symptoms are
(Colmegna et al. 2004). About 50% of such patients suffer not controlled adequately or if the course is more protracted,
subsequent episodes. In contrast, recurrence after enteric methotrexate or sulfasalazine can be added to the NSAID
infection is uncommon. Patients should be advised about the regimen. Intra-articular glucocorticoid injections can be of
risk of re-infection leading to recurrence of arthritis and symptomatic benefit.
about specific precautions for avoiding re-infection. There is little published experience with the use of bio-
The precise role of repeated infections in precipitating logic agents in chronic refractory ReA. However, given the
recurrences of ReA is not clear. Additional genital tract efficacy of tumor necrosis factor inhibition in the treatment
infections may be associated with recurrent ReA, but of several related disorders, e.g., ankylosing spondylitis,
76 R. D. Inman and A. Keat

psoriatic arthritis, and inflammatory bowel disease, this Comment: The term “gonococcal arthritis” refers to a septic
approach is reasonable for patients with highly symptomatic, arthritis, often part of a syndrome of persistent gonococce-
persistent ReA. mia in which primary infection with Neisseria gonorrhea
disseminates widely, leading to low-grade fevers, sparse skin
Pearl: The appearance of circinate balanitis differs accord-
lesions, and septic arthritis. The primary infection is usually
ing to whether or not the patient is circumcised.
but not always in the genital tract. The disease responds to
Comment: If the male is uncircumcised, the lesions of circi- appropriate antimicrobial therapy.
nate balanitis can appear as multiple, serpiginous, shallow The problem arises when aseptic arthritis arises in an indi-
ulcers on the glans or shaft of the penis. These lesions often vidual who also has gonococcal infection. It is unclear
have raised borders. In circumcised males, circinate balanitis whether Neisseria gonorrhea is a true initiator of ReA, but
can appear as dry, hyperkeratotic plaques that are reminis- the prevailing view is that in this circumstance the gonococ-
cent of psoriasis. cal infection is accompanied by an additional simultaneous
non-gonococcal infection and it is this that provokes the
Pearl: If the arthritis of ReA persists beyond the time usually
arthritis.
associated with this condition, consider whether the patient
might have psoriatic arthritis instead. Pearl: Attempting to identify the presumed causal infection
may be very helpful.
Comment: The clinical features of ReA overlap significantly
with those of many patients with psoriatic arthritis. As Comment: Frequently the diagnosis of ReA is made (accu-
examples: rately) on evidence of infection – e.g., diarrhea or urethritis
– without identification of a specific bacterial pathogen.
• Asymmetric, oligoarticular patterns of joint disease are
Because the precise causal link between the bacterium and
present in both ReA and psoriatic arthritis.
the arthritis is unknown, this is not a great problem. However,
• Keratoderma blenorrhagicum, a finding associated with
if there is an infection, knowing its cause or excluding some
ReA, cannot be distinguished histologically from pustular
potential etiologies is valuable. A patient who develops severe
psoriasis.
ReA after bacterial diarrhea associated with ingestion of par-
• The fingernails and toenails in ReA can become thick-
ticular food may be an “index case” that explains a pattern of
ened and develop subungual debris and onychodystrophy.
food-borne illness within a community. Moreover, the diag-
However, nail pitting in a patient with inflammatory joint
nosis of chlamydial urethritis offers the opportunity to treat
disease clearly favors psoriatic arthritis as the underlying
this infection in the patient’s sexual partners, who may or
diagnosis.
may not be symptomatic.
Pearl: Preventing a recurrence of ReA is influenced more by
advice than antibiotics.
Comment: Safe sex practice is sound advice for all sexually- References
active individuals. Among patients with a history of ReA fol-
lowing a GU infection, such advice should be reinforced
Ahvonen P, Sievers K, Aho K. Arthritis associated with Yersinia entero-
with particular vigor. Similarly, for patients who have had colitica infection. Acta Rheumatol Scand 1969; 15(3):232–253
episodes of ReA following enteric infections, extra precau- Barash J, Mashiach E, Navon-Elkan P, et al Differentiation of post-strep-
tion with regard to local food and hygiene should be exer- tococcal reactive arthritis from acute rheumatic fever. J Pediatrics
cised when traveling to locales where enteric pathogens are 2008; 153:696
Colmegna I, Cuchacovich R, Espinosa LR. HLA-associated reactive
common. arthritis: Pathogenic and clinical considerations. Clin Microbiol
Rev 2004; 17:348–369
Pearl: Post-diarrheal arthritis should always occasion a
Hamdulay SS, Glynne SJ, Keat A. When is arthritis reactive? Postgrad
search for gastrointestinal pathogens. Med J 2006; 82(969):446–453
Hannu T, Inman R, Granfors K, Leirisalo-Repo M. Reactive arthritis or
Comment: The new onset of joint pain in the setting of diar- post-infectious arthritis? Best Pract Res Clin Rheumatol 2006; 20
rhea may indeed herald a case of inflammatory bowel disease (3):419–433
with accompanying arthritis. However, infectious diarrhea Keat A. Reiter’s syndrome and reactive arthritis in perspective. N Engl
from bacteria, parasites, or toxins must be excluded defini- J Med 1983; 29;309(26):1606–1615
Kempsell KE, Cox CJ, Hurle M, et al Reverse transcriptase-PCR analy-
tively as the first priority. Sending stool specimens for fecal sis of bacterial rRNA for detection and characterization of bacterial
leukocytes, cultures, ova and parasites, and Clostridium dif- species in arthritis synovial tissue. Infect Immun 2000; 68(10):
ficile toxin assays are critical in this evaluation. 6012–6026
Leirisalo-Repo M. Prognosis, course of diseases and treatment of the
Pearl: Gonococcal arthritis and ReA are different. spondyloarthropathies. Rheum Dis Clin N Amer 1998; 24:737–751
 Systemic Sclerosis (Scleroderma)
and Raynaud’s Phenomenon 10
Janet E. Pope, Philip J. Clements, Daniel E. Furst, Laura K. Hummers,
Dinesh Khanna, Maureen D. Mayes, Thomas Medsger,
James Seibold, and Virginia Steen

10.1.1 Raynaud’s Phenomenon

10.1 Overview of Systemic Sclerosis
and Digital Ischemia
› Systemic sclerosis (scleroderma; SSc) is a chronic con-
nective tissue disease characterized by inflammation Pearl: Skin induration that spares the fingers and is not asso-
within, and fibrosis of, the skin, vascular abnormalities, ciated with RP is not SSc.
visceral damage, and the production of autoantibodies. Comment: The diagnosis of RP is made when a patient pro-
› SSc is divided generally into limited and diffuse forms, vides a history of acral skin color changes precipitated by
based on the extent of skin involvement. cold or emotional distress.
› A group of conditions known as “localized” sclero-
derma, which includes morphea, linear scleroderma, Pearl: The timing of the first development of Raynaud’s phe-
and en coup de sabre, is discussed in the chapter on nomenon and of the onset of other SSc features differs
Scleroderma Mimickers (Chap. 11). according to whether the patient has limited or diffuse SSc.
› Diffuse SSc has a greater extent of skin involvement Comment: RP and the skin changes of early diffuse SSc often
and is more likely to be associated with renal crisis, occur closely together in time. In contrast, RP may be evi-
pulmonary fibrosis, and cardiomyopathy. Patients with dent years before other features of limited SSc are manifest.
diffuse SSc have a higher mortality rate.
› Pulmonary disease remains a major source of morbid- Pearl: Those who develop RP later in life (age >40) are more
ity and mortality for patients with SSc. Patients with apt to have secondary RP.
limited SSc are predisposed to the development of pul- Comment: RP that begins in childhood, during the teen years,
monary arterial hypertension (PAH). Those with dif- or in the patient’s 20s, when it is most likely to be primary
fuse SSc are predisposed to interstitial fibrosis. RP in the absence of overt connective tissue disease sympto-
› Autoantibody profiles can predict to a certain extent the motology. When trying to determine if a patient has a sec-
types of organ involvement that can be expected in SSc. ondary cause of RP, essential points on the physical
› Raynaud’s phenomenon (RP) is characterized by examination are swollen fingers and the presence of dilated
vasospasm of the digital arteries that leads to well- capillaries in the finger nailbeds, either of which points to a
demarcated pallor with cyanosis and rubor. secondary cause of RP. A positive antinuclear antibody
› RP is essentially universal among patients with SSc and (ANA) assay is also instructive; particularly if an anticen-
is the first symptom in a large majority of patients. tromere pattern is present on immunofluorescence (Fig. 10.1)
› Primary RP is often mild enough not to require treatment; or if additional testing identifies a common connective tissue
however, with secondary RP from SSc, there is not only disease autoantibody such as those directed against the Ro,
vasospasm but also fixed blood vessel problems with La, Sm, or RNP antigens. However, most RP is primary and
luminal narrowing, so the ischemia can be more severe. not associated with an underlying condition (Pope 2007;
› Complications of RP in SSc can include digital ulcers Hirschl et al. 2006; Luggen et al. 1995).
often leading to infection and amputation causing pain
Myth: All calcium channel blockers are equally effective for
and functional disability.
the treatment of RP.
› Diffuse SSc often presents with puffiness of the hands
associated with pain, in a way that can mimic rheuma- Reality: Only the dihydropyridine class of calcium channel
toid arthritis. RP often occurs around the time of devel- blockers (nifedipine, amlodipine, felodipine) offers signifi-
opment of puffy hands in diffuse SSc. cant benefit in treating RP. These agents have a stronger
peripheral vasodilatory action when compared with other

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 77

DOI:10.1007/978-1-84800-934-9_10, © Springer Science + Business Media B.V. 2009
78 J. E. Pope et al.

Pearl: Nailfold capillary examination is easily performed in

a standard examination room.
Comment: The only tool that is required is a standard oph-
thalmoscope. Microscope immersion oil enhances the tech-
nique, but surgical lubricating jelly can be substituted if
immersion oil is not available. The oil is placed at the nail-
fold and the ophthalmoscope is set at +40 diopters, the high-
est magnification, which is equivalent to magnifying the skin
by a factor of 10. To bring the nailfold capillaries into focus,
the examiner does not adjust the ophthalmoscope but rather
moves his or her head toward or away from the finger under
examination. The presence of dilated capillary loops or areas
of capillary dropout are indicative of an underlying connec-
tive tissue disease (most often SSc). Some capillaries at the
nailbed are seen on the cuticle and can even be observed by
Fig. 10.1 Anti-centromere pattern of immunofluorescence on antinu- the naked eye.
clear antibody testing. If this pattern is present, a specific anticentrom-
ere antibody assay does not need to be performed Myth: Dilated capillary loops are observed only in patients
with SSc.
Reality: Nailfold capillary loop dilations are characteristic of
calcium channel blockers. Diltiazem and verapamil provide a number of connective tissue disorders, including lupus, der-
only limited benefit and should play no role in therapy matomyositis, and even (occasionally) rheumatoid arthritis
(Thompson et al. 2001). (Fig. 10.2) (Pope et al. 2008). However, this finding is most
common in SSc. Nailfold capillary abnormalities usually
Pearl: Calcium channel blockers in SSc reduce the number
occur early in the patient’s disease course and drop out or
of RP attacks.
disappear over time, perhaps an indication of the progression
Comment: The best studied treatment class for RP is the of an obliterative vasculopathy that is present in some patients
dihydropyridine calcium channel blockers. The usual benefit with SSc.
is on the order of a 30% reduction in attack frequency. Most
Myth: The disappearance of dilated capillaries is a good thing.
randomized controlled trials of these agents have shown no
improvement in attack severity, but the doses employed in Reality: The vasculature in SSc becomes more abnormal over
most studies have been low (Thompson et al. 2001). Calcium time and often eventually no dilated capillaries are seen at the
channel blockers can be prescribed on either a regular or an nailbeds. This is because of the “dropout” of vasculature that
as needed basis. Side effects of calcium channel blockers are
common and often limit the dose that can be used.
Pearl: Clinical indicators provide clues to the likelihood of
secondary RP.
Comment: Leading clinical indicators of secondary RP are
shown in Table 10.1. Longitudinal prospective studies of
patients presenting with RP suggest that abnormal nailfold
capillary at baseline is the strongest predictor of the develop-
ment of a defined connective tissue disease in the future
(Pavlov-Dolijanović et al. 2006).

Table 10.1 Leading clinical indicators of secondary Raynaud’s

phenomenon Fig. 10.2 A patient with a
Age of onset >40 years history of decades of seropositive
Males presenting at any age rheumatoid arthritis developed
Raynaud’s phenomenon, dilated
Presence of antinuclear antibodies (at a titer greater than 1:160)
capillaries at the nailbeds, and
(Pope 2003a)
sclerodactyly distal to the
Presence of abnormal nailfold capillaries (Pope 2003a; Herrick 2005)
proximal interphalangeal joints
10 Systemic Sclerosis (Scleroderma) and Raynaud’s Phenomenon 79

occurs over time with this disorder. An inverse correlation approach with topical antimicrobial therapy, protection of the
exists between the extent of nailfold changes detected by vid- area (bandaging) when minor trauma is more likely, vasodi-
eocapillaroscopy and the duration of both RP symptoms and lator therapy, and good skin care practice (moisturizer and
the time since diagnosis of SSc (Cutolo et al. 2004). Enlarged emollient therapy).
and giant capillaries, at times with hemorrhages, are the earli-
Pearl: SSc patients can lose padding on their feet, just as
est pattern observed. Loss of capillaries, ramified capillaries,
they do on their fingertips.
and vascular architectural disorganization are increased in
later disease (Cutolo et al. 2004; Meli et al. 2006). Comment: Digital pulp loss in the fingertips is characteristic
of SSc. SSc patients can also develop thinned padding on the
Myth: Cyanosis correlates with ischemia. soles of the feet, which makes it uncomfortable to walk in
Reality: Cyanosis occurs when blood is stagnant and deoxy- bare feet (Fig. 10.3). Loss of connective tissues on the sole of
genation occurs. An intact nutritional blood supply can exist the foot tends to occur in the distal portions.
even in the presence of cyanosis. This fact is evidenced by the Pearl: Extrusion of digital tip calcium deposits is a challeng-
lack of ischemic pain and the finding of intact capillary refill ing mimicker of infected ulcerations.
that generally accompanies cyanosis. Many patients have
completely asymptomatic episodes of cyanosis. Ischemic Comment: Subcutaneous calcinosis frequently occurs at
pain and diminished capillary refill with or without evidence sites of minor trauma and pressure, such as the extensor sur-
of tissue damage are signs of ongoing ischemia and require faces of the forearms and the fingertips. These deposits can
immediate action to try and improve flow. erupt through the skin, leading to a purulent-appearing dis-
charge. This lesion is often mistaken for a digital tip ulcer-
Myth: Ulcerations that occur overlying the dorsum of the ation that has become secondarily infected. A pocket of
PIP and MCP joints are ischemic in nature and should be subcutaneous calcinosis can often also be accompanied by
managed with aggressive vasodilator therapy. erythema, particularly as new deposit occurs, further confus-
ing the picture.
Reality: Digital tip ulcerations are related mostly to the absence
Calcinosis should be suspected when an SSc patient pres-
of blood flow to the distal finger. In contrast, the lesions that
ents with spontaneous development of a draining lesion of
occur over the dorsal surfaces of the PIP and MCP joints are
the fingertip in the absence of ischemic pain and any history
multifactorial in nature. Patients who develop these lesions
of previous digital ulcerations. Some patients describe a
typically have significant skin thickening in their fingers, usu-
chalky discharge. Plain radiographs can confirm the presence
ally accompanied by more widespread cutaneous changes
of calcium (Fig. 10.4a, b). Recognition of this complication
characteristic of diffuse SSc.
can help distinguish an infected digital ulcer from the drain-
The hand joints in such patients have a tendency to become
age of calcinosis. The extrusion of an area of calcinosis is
contracted and prone to minor trauma. The skin overlying the
accompanied normally by the relief of discomfort.
joint is atrophic and dry because of damage of sebaceous
glands that increases the sensitivity to minor trauma and irri- Pearl: Acro-osteolysis in SSc can mimic “clubbing” and may
tation. The overall blood flow in the fingers is reduced. be a manifestation of peripheral vascular disease rather than
Management of these lesions requires a multi-faceted pulmonary disease.

Fig. 10.3 Diminished fat pads on

the feet. Scleroderma patients can
lose padding on their feet, just as
they do on their fingertips. This
makes it uncomfortable to walk in
bare feet. Note loss of connective
tissues on the sole, especially
distally
80 J. E. Pope et al.

a disorders, in which the acute phase reactants are usually ele-

b
vated (often strikingly so).
Despite the often normal ESR in SSc, the ANA titer is
typically sky-high. Few if any disease entities lead to higher
ANA titers than does SSc.
Myth: Sclerodactyly distal to the metacarpophalangeal
(MCP) joints usually indicates a diagnosis of SSc.
Reality: Sclerodactyly distal to the MCP joints without any
proximal skin involvement can occur in many other condi-
Fig. 10.4 (a) Fingertips in a patient with scleroderma. The middle fin- tions such as in diabetes, other connective tissue diseases,
ger shown (the third digit) has a calcium deposit in the pulp of the fin- and rheumatoid arthritis. It is not specific for SSc.
ger. The patient presented with intense pain in the digit that was
exacerbated by dorsiflexion. A digital pit is also present in the second Pearl: Scleroderma can mimic rheumatoid arthritis in its
digit, positioned superiorly to the third digit in this photograph. (b) presentation.
Radiograph revealing calcinosis of the third digit (Figures courtesy of
Dr. John Stone) Comment: Scleroderma can present with arthralgias and dif-
fusely swollen hands, frequently leading to the diagnosis of an
undifferentiated polyarthropathy or even rheumatoid arthritis.

10.1.3 Autoantibodies
and Disease Phenotypes

Myth: Anti-centromere antibodies are the principal SSc

autoantibodies that are linked to pulmonary hypertension.
Reality: Anti-centromere antibodies are the classic marker
for limited SSc (formerly called the CREST syndrome, an
acronym for calcinosis, Raynaud’s phenomenon, esophageal
dysmotility, sclerodactyly, and telangiectasias). The limited
SSc disease subset is a population of patients generally con-
sidered to be at the highest risk for the development of PAH.
However, the presence of anti-centromere antibodies is not a
predictor of pulmonary hypertension within the limited SSc
Fig. 10.5 The second and third fingertips have lost pulp. There is a healed population (Steen et al. 1984).
ulcer on the thumb (volar side) (Figure courtesy of Dr. John Stone)
Limited SSc patients with a nucleolar staining pattern
on their ANA, typically caused by antibodies to either
Comment: Pseudo-clubbing can occur when a patient devel- the U3-RNP or the Th/To antigen, are at an increased risk
ops osteolysis of the distal digital tuft (Fig. 10.5). This pro- for PAH (Fig. 10.6) (Steen 2005). Patients with anti-topoi-
cess may be completely painless, or may be associated with somerase-3 antibodies are predisposed to the development of
chronic ischemia. pulmonary fibrosis. When their interstitial lung disease is
severe, patients with anti-topoisomerase-3 antibodies often
develop pulmonary hypertension that is secondary to fibro-
sis. Patients with diffuse SSc can also develop the primary
10.1.2 Diagnosis vasculopathy that leads to severe PAH, particularly if they
have U3RNP autoantibodies (Sacks et al. 1996). Because
Myth: As a “connective tissue disorder,” SSc is usually asso- pulmonary hypertension can occur in either major disease
ciated with an elevated erythrocyte sedimentation rate. subset of SSc, the risk of this complication should be strati-
fied by serologic phenotypes rather than clinical classifica-
Reality: On the contrary, even in the setting of significant end-
tion schemes.
organ involvement in SSc, the ESR is often entirely normal.
This stands in stark contrast to other connective tissue Myth: Autoantibodies in SSc are epiphenomena.
10 Systemic Sclerosis (Scleroderma) and Raynaud’s Phenomenon 81

through which these autoantibodies operate and interact with

other elements of the immune system requires further study.
Pearl: Antibodies to RNA polymerase are associated with
scleroderma renal crisis.
Comment: Autoantibodies are increasingly being recognized
to be associated with specific phenotypic manifestations of
rheumatic disease. Antibodies to RNA polymerase are a sig-
nificant risk factor for scleroderma renal crisis. Some patients
with scleroderma renal crisis (with or without anti-RNA
polymerase antibodies) present with microangiopathic hemo-
lytic anemia, mimicking thrombotic thrombocytopenic pur-
pura. Other examples of the associations between
autoantibodies and clinical manifestations of SSc are shown
in Table 10.2.
Pearl: Anticentromere antibody, an autoantibody associated
with good survival, virtually never occurs in diffuse SSc.
Comment: Patients with diffuse SSc may have antibodies to
topoisomerase I (Scl-70), RNA polymerase, and a number
Fig. 10.6 Immunofluorescence assay for antinuclear antibodies dem-
onstrating a nucleolar pattern of immunofluorescence of other antigens, but virtually never possess anticentrom-
ere antibodies. Ten-year survival rates for patients with
anticentromere antibodies are on the order of 93%, com-
pared with 66% for patients with anti-Scl70 antibodies
Reality: The pathways through which at least some SSc
(Kuwana et al. 1994).
autoantibodies exert direct deleterious effects remain to be
elucidated fully, but as described elsewhere in this chapter Pearl: SSc can overlap with other CTDs, so treat what is
there is abundant evidence that specific autoantibodies cor- treatable.
relate with particular clinical features. As an example, the
Comment: SSc is commonly associated with Sjögren’s syn-
topoisomerase I (Scl70) autoantibody promotes interstitial
drome and less often occurs in overlap combinations with
lung disease in a genetically predisposed animal model
clinical features of systemic lupus erythematosus, inflamma-
(Kuwana et al. 2001). These autoantibodies are also linked
tory myopathy, or rheumatoid arthritis. Remember to treat
strongly to the risk of developing interstitial lung disease.
what is treatable. Some clinicians believe that Sjögren’s syn-
Another autoantibody, directed against RNA polymerase III,
drome is associated with (as opposed to “secondary to”)
is associated with a high risk of the development of sclero-
SSc. There are reports of patients whose clinical features are
derma renal crisis. Finally, anticentromere antibodies are
similar to those of primary Sjögren’s syndrome who also
linked tightly to the limited SSc phenotype.
appear to have “incidentally” mild cases of SSc (Pope 2002;
Thus, at a minimum, autoantibody profiles in patients
Salliot et al. 2007).
with SSc are helpful in identifying the probably evolution of
clinical phenotypes. Mechanistic proof of the pathways Pearl: Men are more likely to have diffuse SSc if they get SSc.

Table 10.2 Systemic sclerosis: Associations of phenotypic features with specific autoantibodies (Steen 2005)
Autoantibody Staining pattern Disease Organ manifestations
ANA
Nucleolar Scleroderma
Centromere Limited scleroderma Pulmonary hypertension
Anti-topoisomerase 1 (Scl-70) Speckled or homogeneous Diffuse scleroderma Renal crisis; pulmonary fibrosis
Anti-U1-RNP Limited scleroderma Severe GI involvement
Anti-U3-RNP (fibrillarin) Nucleolar Scleroderma and mixed Pulmonary fibrosis; Severe GI involvement
connective tissue disease
Anti-PM/Scl Nucleolar Scleroderma/polymyositis Inflammatory myopathy; Pulmonary fibrosis;
overlap Severe GI involvement
Anti-RNA polymerase Scleroderma Renal crisis; Decreased frequency of lung
disease
82 J. E. Pope et al.

Comment: Although SSc is more common in women than in

men – up to 87% of SSc patients are women – SSc in males
is likely to be diffuse in nature. The ratio of limited to diffuse
disease among female SSc patients has been estimated to be
2:1. In contrast, the ratio of limited to severe disease among
male SSc patients is 1:2. Thus, if a man has SSc he is more
likely to have diffuse disease. A woman is more likely to have
limited disease (Walker et al. 2007; Al-Dhaher et al. 2008).
Pearl: Most major organ involvement in diffuse SSc also
occurs early in the disease course (within the first 5 years).
Comment: Interstitial lung disease (Fig. 10.7), scleroderma
renal crisis, and scleroderma cardiomyopathy, all of which
are features of diffuse SSc, usually begin within the first 5
years but can recur or worsen over time. Severe organ involve-
ment in SSc patients with diffuse SSc most often occurs early
in the course of the disease (Steen 2000; Al-Dhaher et al.
2008). This is not necessarily true for PAH, typically a com-
plication of limited SSc, which can develop at any time.
Pearl: There is more pigmentation in patients with diffuse
SSc as opposed to limited SSc.
Fig. 10.7 Posteroanterior chest radiograph in a patient with interstitial
Comment: SSc can be associated with either hyper- or hypop- lung disease secondary to scleroderma. The radiograph reveals severe
igmentation, and sometimes both are found in the same scarring and very small lung volumes
patient (Fig. 10.8). Hyperpigmentation can be caused by
either increased numbers of melanocytes within the skin of
patients with SSc, but can also be a function of postinflam-
matory hyperpigmentation. Pigmentary changes are more
common in patients with diffuse SSc than in those with lim-
ited SSc. Some patients with diffuse disease have a strikingly
tanned appearance and smooth skin, because skin affected by
SSc tends to lose its appendages. As the cutaneous features
of SSc improve over time as they often do, there may be a
regrowth of hair on the patient’s extremities (Pope et al.
1996). Pigmentary changes are slower to resolve.
Pearl: The Health Assessment Questionnaire Disability
Index (HAQ-DI) is useful in SSc.
Fig. 10.8 Photograph of a patient with both hyperpigmentation and
Comment: Because of its multi-organ system nature, SSc hypopigmentation secondary to scleroderma. The “tanned” skin is actu-
normally has a substantial impact on patients’ function and ally hyperpigmentation secondary to scleroderma. The hypopigmenta-
tion over the metacarpophalangeal joints is also a result of skin
quality of life. The Health Assessment Questionnaire
inflammation
Disability Index (HAQ-DI) is one of the most commonly
used health related quality of life (HRQoL) measures in SSc
(Khanna 2006). This index has been validated for use in SSc
patients (Merkel et al. 2002; Poole et al. 1995; Steen and Pearl: The strongest epidemiologic risk factor for SSc is a
Medsger 1997b). The HAQ-DI is usually higher in diffuse family history of the disease.
SSc than in limited disease.
Comment: The prevalence of SSc in the US population
Pearl: Most men with SSc have erectile dysfunction. is 0.026%. This compares with prevalence in the families
of patients with SSc that is on the order of 1.6% (Arnett
Comment: Erectile dysfunction can occur early in the disease
et al. 2001).
course of SSc. Male patients suffering from erectile dysfunc-
tion who also have RP should be queried and examined for Pearl: Many SSc patients have depression. Recognize and
other clinical features of SSc. treat it.
10 Systemic Sclerosis (Scleroderma) and Raynaud’s Phenomenon 83

Comment: Many chronic diseases associated with pain and

fatigue are accompanied by depression. However, in SSc,
patients also often suffer from significant changes in appear-
ance over time. This constitutes an additional “issue” for
patients, and for many their altered appearance has a substan-
tial impact on mood. The estimate of depression in SSc is
higher than that of comparative population estimates and may
be higher than many other chronic diseases such as heart and
lung diseases (Hudson et al. 2009).

Myth: Patients with limited SSc do not develop significant

internal organ disease.

Reality: Although limited SSc has a better overall prognosis

than does diffuse disease, some individuals with limited SSc Fig. 10.9 Pterygium inversum unguis. This finding gives the finger-
develop severe pulmonary hypertension or profound gastro- nails of patients with SSc an “unclean” appearance (Courtesy of
intestinal hypomotility. A subset of patients with limited SSc Dr. Fredrick Wigley)
also develops pulmonary fibrosis, a complication usually
associated with diffuse SSc.
Monitoring for pulmonary hypertension in limited SSc
Reality: Patients with SSc often develop a condition known
includes periodic echocardiography with confirmation by
as pterygium inversum unguis (Fig. 10.9), a distal extension
right heart catheterization in those individuals with elevated
of the hyponychial (subungual) tissue that is anchored to the
right ventricular systolic pressures. Exercise echocardiogra-
undersurface of the nail.
phy and exercise pulmonary function tests may also have a
role in detecting incipient pulmonary hypertension. Myth: Telangiectasias are uncommon in diffuse SSc.
Reality: In the 1970s, the term “CREST syndrome” was used
Pearl: The new onset of pruritus in a patient with SSc can be
to describe patients with SSc who had most or all the follow-
an early symptom of primary biliary cirrhosis (PBC).
ing constellation of symptoms: C – calcinosis, R – Raynaud
Comment: Intense pruritus is sometimes experienced by phenomenon, E – esophageal dysmotility, S – sclerodactyly,
patients with new-onset SSc, particularly those with diffuse and T – telangiectasias. CREST patients were considered to
disease. The pruritus, probably caused by cutaneous inflam- have skin thickening restricted to their distal extremities and
mation accompanied by mast cells and histamine elevations occasionally their faces, and distinctions were drawn between
in the skin (Pope et al. 1996), commonly subsides once the this disease subset and patients with diffuse SSc. Although
skin thickening has become established. differences clearly exist between patients who fit the CREST
In contrast, the new onset or recurrence of pruritus in an phenotype and those with diffuse SSc, it has been noted that
otherwise stable SSc patient might herald the presentation of many patients with diffuse SSc demonstrate many or all of
PBC. PBC occurs almost exclusively among patients with the features of CREST, particularly later in the course of their
limited SSc rather than diffuse SSc, and also develops in a illness when diffuse skin changes have softened (Furst et al.
minority of patients with Sjögren’s syndrome (Pope and 1984). For this reason, the term CREST syndrome has been
Thompson 1999). The first laboratory indication of PBC is replaced by the preferred “limited cutaneous involvement.”
usually an elevation of the alkaline phosphatase that occurs
Pearl: The rate of skin thickness progression is a useful clini-
in the absence of serum transaminase increases. Antimi-
cal tool.
tochondrial antibodies (AMA) are present in most but not all
patients with SSc who have PBC. The early diagnosis of Comment: A rapid rate of skin thickness progression in dif-
PBC is important, because ursodiol can delay or prevent the fuse SSc identifies patients at increased risk for new major
development of cirrhosis (Rigamonti et al. 2006). internal organ involvement and death. When calculated either
at the first visit from onset of skin thickening or between the
second and third visits using the modified Rodnan skin score,
a rapid rate is associated with an increased frequency and ear-
10.1.4 Skin Disease lier appearance of interstitial lung disease, cardiac dysfunc-
tion, renal involvement, and higher mortality (Perera et al.
2007; Shand et al. 2007).
Myth: Patients with SSc have a difficult time keeping their
nails clean. Myth: Skin of normal thickness in SSc is unaffected.
84 J. E. Pope et al.

Reality: Although the clinical assessment of skin thickness at Comment: Severe esophageal disease may be subclinical in
different anatomical sites is helpful in classifying SSc patients with SSc. Consequently, empiric treatment with a
patients, the Rodnan skin score is in reality a fairly blunt proton pump inhibitor should be employed in all patients
instrument. Many patients with limited cutaneous involve- with SSc because such therapy can prevent stricture forma-
ment have widespread cutaneous hyper- and hypopigmenta- tion, aspiration pneumonitis, and Barrett’s esophagus
tion, signifying that melanocytes are either over- or (Ebert 2006).
understimulated in areas without apparent skin thickening.
Pearl: SSc patients are prone to a host of dental problems.
In immunohistochemical studies of clinically involved and
uninvolved skin in diffuse SSc, both contained markers of Comment: Oral hygiene is compromised in SSc for several
endothelial and fibroblastic activation compared with con- reasons. The frequent findings of a small oral aperture from
trols (Claman et al. 1991). Thus, it is likely that all skin is mouth furrowing and decreased hand function conspire to
affected by the disease process, even if it appears to be nor- make holding a toothbrush a challenge. Xerostomia caused
mal clinically by palpation (Pope et al. 1996). by coexistent Sjögren’s syndrome or fibrosis of the salivary
glands, acid reflux (GERD) (Fig. 10.10), and the use of medi-
Pearl: After 2–5 years, skin tightening in diffuse SSc often cations that alter oral hygiene unfavorably (e.g., nonsteroidal
regresses somewhat but internal organs can continue to antiinflammatory drugs or glucocorticoids) also complicate
worsen. dental care. SSc patients should therefore have frequent, reg-
Comment: In patients with diffuse SSc who were participants ular visits to the dentist.
in a clinical trial, the mean skin score remained stable for the Myth: Gastric antral vascular ectasia is a rare complication
first 12 months but decreased significantly at both 24 and 36 of SSc.
months. Among patients with limited SSc, however, the skin
score did not change significantly over 3 years. Reality: More than 90% of SSc patients have gastrointestinal
Although the involved skin in diffuse SSc often displays dysfunction, with problems that range in severity and include
softening by 5 years of disease duration, progression of the dysphagia, reflux, small bowel overgrowth, gastroparesis,
disease in internal organs (e.g., the lung) can progress even rectal incontinence, pseudodiverticuli, and a dilated esopha-
as the skin softens over time. The rapid progression of skin gus (Fig. 10.11) (Marie 1996). Gastric antral vascular ectasia
disease in diffuse SSc is an ominous sign. However, improve- (GAVE, watermelon stomach) (Fig. 10.12) is a common
ment in the skin score does not equate to a lack of disease finding on endoscopy in patients with early diffuse SSc.
progression within internal organs (Clements et al. 1993; GAVE is an infrequent finding in the non-SSc population
Perera et al. 2007; Steen and Medsger 2001). and some gastroenterologists are unfamiliar with this entity,
particularly in its early or developing phase. The most com-
Myth: Scleroderma skin often normalizes over time. mon features of this disorder relate to the gradual
Reality: Skin thickness measured by the modified Rodnan
method at a given anatomic site, e.g., forearm, can regress
over time from 2+ (moderate thickening) to 0 (no thickening).
However, skin biopsies from later phases of the disease do not
show normal skin but rather thinning (atrophy) of the dermis
and “tethering,” i.e., increased attachment of the dermis to the
underlying subcutaneous tissue (Clements et al. 2004).
Pearl: Skin thickening that precedes RP predicts subsequent
events.
Comment: RP is the first clinical manifestation in more than
half of all SSc patients. However, skin thickening precedes
RP in a minority of SSc patients (25%), where such individu-
als were at higher risk of developing diffuse SSc (79 vs. 45%;
p = 0.0001) and renal crisis (9 vs. 3%, p = 0.0001) compared
with those whose RP came before thickening of the skin
(Medsger 2009).

10.1.5 Gastrointestinal Involvement

Pearl: All patients with SSc should be treated with therapy Fig. 10.10 Gastroesophageal reflux disease (GERD) with esophagitis
designed to suppress acid production by the stomach. near the lower esophageal sphincter in a patient with scleroderma
10 Systemic Sclerosis (Scleroderma) and Raynaud’s Phenomenon 85

Fig. 10.12 Gastric antral vascular ectasia (“watermelon stomach”) in a

patient with scleroderma

Comment: Malnutrition should be suspected if there is marked

Fig. 10.11 Dilated esophagus in scleroderma, as seen on a barium weight loss or a low serum albumin. If severe dysphagia
swallow necessitates consideration of tube feeding, a jejunal tube is
preferable to a gastric tube because of the high likelihood of
significant gastroesophageal reflux. Total parenteral nutrition
may be needed in some patients. Features associated with
development of iron deficiency anemia. Upper endoscopy
SSc malnutrition are weight loss, anorexia, diarrhea, history
may not show an obvious bleeding site as the ectatic blood
of bloating, a low serum albumin level, and a distended abdo-
vessels ooze blood on an intermittent basis. Therapy for
men on physical examination.
GAVE includes laser coagulation. More than one session
may be required (Calamia et al. 2000; Shibukawa et al. Pearl: Proton pump inhibitors should be taken before a meal
2007). in patients with severe gastroesophageal reflux disease
(GERD).
Pearl: Abdominal bloating, early satiety, gas, and constipa-
tion that alternates with intermittent diarrhea might be Comment: Proton pump inhibitors (PPIs) work better on an
related to small bowel overgrowth. empty stomach. The PPI dose may need to be doubled or
increased even further beyond approved doses. Raising the
Comment: Patients can be diagnosed with small bowel over-
head of the bed at night, no food after supper time, and the
growth by the history alone. No additional investigations
addition of other prokinetic agents may help to reduce
such as a hydrogen breath test are necessary in most cases.
GERD. Individual triggers of GERD such as alcohol, caf-
This underdiagnosed problem can have a substantial impact
feine, chocolate, and mint should be avoided if they worsen
on patients’ quality of life and should be treated if symptom-
symptoms.
atic. Empiric courses of therapy are reasonable to undertake
without diagnostic testing if the history is sufficiently com-
pelling. Intermittent antibiotics can be highly effective and
may need to be cycled. The most commonly used antibiotics
10.1.6 Interstitial Lung Disease
are amoxicillin, ciprofloxacin, tetracycline, erythromycin,
and metronidazole. Promotility agents such as metaclopro-
Pearl: Most declines in patients’ forced vital capacity (FVC)
mide, domperidone, and erythromycin can also be helpful.
occur during the first 4–6 years after the onset of symptoms
Pearl: Malnutrition may occur in SSc. attributable to SSc.
86 J. E. Pope et al.

Comment: Steen et al. assessed 890 patients with SSc, 13% in patients with SSc (White et al. 2000; Silver et al. 1990;
of whom (n = 116) developed severe restrictive lung disease Behr et al. 1996; Witt et al. 1999). However, more recent
with an FVC ≤ 50% predicted (Steen et al. 1994). Among reports suggest that the percentage of neutrophils and eosino-
those 116 patients, 55 (48%) with severe restrictive disease phils in BAL fluid does not predict FVC declines (Goh et al.
had two sets of pulmonary function tests during the first 5 2007; Strange et al. 2008). Goh et al. reported their analysis
years after diagnosis. The FVC% predicted declined by 32% of data drawn from a prospective longitudinal study. Strange
per year in the first 2 years of illness, 12% in years 2–4, and et al. analyzed findings from the Scleroderma Lung Study. In
only 3% during years 4–6 of disease. These observational that latter analysis, the presence of an abnormal BAL did not
data suggest that only a minority of SSc patients develop add to the information obtained from pulmonary function
severe restrictive disease and that the greatest decline occurs tests and high-resolution computed tomography of the chest
in the first 4–6 years after onset of SSc symptoms. (Fig. 10.13) in the prediction of treatment response.
Similar observational data from Greece also suggested
Myth: The use of glucocorticoids in interstitial lung disease
that the 10% of patients who are destined to have major
associated with SSc has a sound scientific basis.
declines, most of the change occurs early in disease (Plastiras
et al. 2006). Reality: The American Thoracic Society and European
Respiratory Society issued an International Consensus
Pearl: The interstitial lung disease in SSc usually occurs at
Statement that addressed the use of glucocorticoids in inter-
the bases.
stitial lung disease in 2000 (American Thoracic Society
Comment: Interstitial lung disease or pulmonary fibrosis in 2000). To quote the statement: “Despite their ubiquitous
SSc is found primarily in the lung bases bilaterally, but the use, no prospective, randomized, double-blind, placebo-
process can extend beyond this area to involve all regions of controlled trial has evaluated the efficacy of glucocorticoids
the lungs. In patients with atypical patterns of pulmonary in the treatment of idiopathic pulmonary fibrosis. In three
disease, aspiration caused by incompetence of the lower trials that compared glucocorticoids with no treatment, none
esophageal sphincter should be considered. of the untreated patients with IPF improved.” Thus, as in
Myth: Interstitial lung disease does not occur in patients
with limited SSc.
Reality: The community perception of interstitial lung dis-
ease in SSc is that this complication occurs almost exclusively
in patients with diffuse cutaneous SSc. A recent example of
how this bias could influence the care of SSc patients with
interstitial lung disease was a randomized controlled trial that
compared bosentan therapy to placebo in interstitial lung dis-
ease: patients with limited SSc were excluded from that study.
Yet approximately 40% of SSc patients who have an FVC
that is less than 75% of predicted – the definition of intersti-
tial lung disease in some studies – have limited SSc (Steen
et al. 1994).
In the Scleroderma Lung Study (Tashkin et al. 2006),
approximately 40% of the 158 patients enrolled had limited
SSc. The lung physiology of these patients was not different
from that of patients with diffuse SSc at baseline. Moreover,
the limited SSc subset actually had more extensive fibrotic
changes at baseline, as assessed by high-resolution computed
tomograms of the chest. Most importantly, the course of
interstitial lung disease in the two patient subsets did not dif-
fer over 24 months.
Fig. 10.13 Computed tomographic scan in scleroderma, showing
Myth: Data from bronchoalveolar lavage in patients with SSc interstitial lung disease (ILD). The ILD in SSc typically involves the
bases but can be widespread. If pulmonary changes spare the bases, one
predict the response of interstitial lung disease to treatment. should think of aspiration or other reasons. This high-resolution com-
puted tomography scan is from a woman with “scleroderma lung,”
Reality: The finding of increased neutrophils or eosinophils
characterized by marked honeycombing. She has only 30% lung capac-
in bronchoalveolar lavage (BAL) fluid has been considered ity and is on oxygen. The imaging study shows scarring, traction fibro-
to be predictive of a deteriorating course in lung physiology sis, and honeycombing
10 Systemic Sclerosis (Scleroderma) and Raynaud’s Phenomenon 87

many inflammatory conditions associated with autoimmu- finding of a “ground glass” appearance or “active alveolitis”
nity, the benefit of treatment with immunosuppression is on BAL. In addition, physical function, vitality, and dyspnea
inferred from the outcomes of patients whose natural history all improved in a clinically meaningful ways among the
unfolds without an intervention. cyclophosphamide-treated patients compared with the pla-
A more recent article stated: “Treatment of IPF has not cebo group, as did skin findings.
been shown to influence outcomes.” Therapeutic options are The other study examined 6 months of pulse intravenous
limited and usually ineffectual. Glucocorticoids and immuno- cyclophosphamide (600 mg per meters-squared) followed by
suppressive or cytotoxic agents have been used to treat idio- azathioprine (2.5 mg/kg/day, maximum 200 mg/day) for 6
pathic pulmonary fibrosis and usual interstitial pneumonitis, months versus placebo cyclophosphamide followed by pla-
but benefit has not been established” (Lynch et al. 2006). The cebo azathioprine (Hoyles et al. 2006). This study was sub-
International Consensus Statement of the American Thoracic stantially smaller than the Scleroderma Lung Study, with a
Society and European Respiratory Society concluded that “no total of only 46 patients. The cyclophosphamide group had a
data exist that adequately document that any of the current 4% improvement in FVC when compared with the placebo
treatment approaches improves survival or the quality of life group. Although the difference was not statistically signifi-
for patients with IPF. Nonetheless, given the poor prognosis cant – p = 0.08 – the results were consistent with those of the
associated with idiopathic pulmonary fibrosis/usual intersti- larger study.
tial pneumonitis, physicians may offer treatment in patients One understandable criticism of these studies is that the
with severe or progressive disease.” improvement in FVC was small, albeit statistically signifi-
In this context, the International Consensus Statement cant. However, given the clinically important differences in
advocated an empirical trial of either azathioprine or cyclo- patient-derived measures such as dyspnea and function,
phosphamide alone or combined with low-dose glucocorti- cyclophosphamide appears to have a modest but real effect in
coids. These recommendations reflect expert opinion, but have the interstitial lung disease associated with SSc. There remains
not been validated. In addition, the recommendations apply to ample room for improvement.
interstitial lung disease in general rather than to the specific
Pearl: One year of oral cyclophosphamide has effects that
entity of SSc. Thus, the use of glucocorticoids in SSc is still
last long after the medication has been discontinued.
based on “eminence” rather than “evidence.” In the Scleroderma
Lung Study, glucocorticoid use was limited to a maximum Comment: The perception is that cyclophosphamide has very
prednisone dose of 10 mg/day (Tashkin et al. 2006). dangerous long-term consequences and that it is therefore
In summary, there is no firm scientific evidence that glu- prudent to stop the medication after 6 months to a year.
cocorticoids have a place in the treatment of interstitial lung So what happened in the second year of the Scleroderma
disease, especially in the interstitial lung disease associated Lung Trial, when patients were followed on no medications
with SSc. after stopping cyclophosphamide? In fact, the FVC in the
group of patients treated previously with cyclophosphamide
Pearl: Cyclophosphamide has a modest effect on forced vital
continued to improve for 6 months, while the FVC in the
capacity, function, and dyspnea in interstitial lung disease
placebo treated group remained stable or decreased slightly.
associated with SSc.
By 18 months the differences in FVC between the two treat-
Comment: Two randomized, double-blind, placebo-con- ment groups actually increased to about 6.8% (Tashkin et al.
trolled trials of cyclophosphamide have shown modest but 2007). Thereafter, the FVC in the cyclophosphamide group
real efficacy in the treatment of SSc-related interstitial lung declined rapidly so that by 24 months, the two treatment
disease. One study compared oral cyclophosphamide (2 mg/ groups were equivalent by this measure. Other outcomes
kg, rounded to the nearest 25 mg) against placebo for 1 year, tended to remain stable after the year of cyclophosphamide
followed by a 1 year period of untreated observation (Tashkin therapy so that whatever difference occurred at 12 months
et al. 2006). Among the 158 patients, 79 were assigned to remained that way for 18 and even 24 months, including dys-
cyclophosphamide and 79 to placebo. At the end of the year pnea scores, function, and skin score.
of treatment, the mean FVC among the cyclophosphamide-
Pearl: Cyclophosphamide therapy is well-tolerated during
treated patients was only about 3% higher than that of the
1 year of treatment.
placebo-treated patients (p = 0.03). When stratified into
patients whose baseline FVCs were less than 70% of pre- Comment: When cyclophosphamide is used in large doses to
dicted, the differences at 1 year were somewhat greater treat cancer, there are numerous and important toxicities,
(6.8%). ranging from severe stomatitis through hair loss to bone mar-
It is worth noting that baseline fibrosis as demonstrated row suppression, bleeding, infection, and possible cancer.
by high-resolution computed tomography was associated The toxicities of relatively low-dose cyclophosphamide regi-
with a better response to cyclophosphamide than was the mens, as used in the oral cyclophosphamide study by Tashkin
88 J. E. Pope et al.

et al. and Hoyles et al. are significantly lower over the short • Patients who have at least moderate fibrosis on baseline
term (Tashkin et al. 2006; Hoyles et al. 2006). high-resolution computed tomography scans of the chest.
The adverse effects reported in the Scleroderma Lung “Moderate” had been defined variably across studies,
Trial were strikingly fewer than those observed in the e.g., more than 20% lung involvement in one study and at
Wegener’s Granulomatosis Etanercept Trial (WGET least 25% of six lobes in another.
Research Group 2005), a study of comparable size in which
Debatable points within clinical practice now are the wisdom
the majority of patients received oral cyclophosphamide in
of treating patients who have had more than 6 years of SSc
addition to the investigational treatment or placebo (WGET
and the appropriateness of treating patients who have no
Research Group 2002). The major difference in the experi-
dyspnea.
ence of these two clinical trials may be that the Scleroderma
Lung Trial used far lower doses of glucocorticoids. Myth: Patients with SSc and significant interstitial lung dis-
ease always have Velcro crackles.
Pearl: In a patient with SSc-interstitial lung disease, GERD
may be a contributing factor to the signs and symptoms of Reality: The physical and radiological examination is not
lung involvement. always concordant. Most but not all patients with significant
interstitial lung disease in SSc have crackles on auscultation.
Comment: GERD affects approximately 90% of patients with The breath sounds may be quiet or even normal in patients
SSc. Previous studies have shown an association between with interstitial lung disease. Thus, pulmonary function tests
GERD and SSc-related interstitial lung disease (Marie et al. and chest radiographs should be ordered in patients with SSc
2001; Lock et al. 1997). GERD has been implicated as one of who complain of dyspnea.
inciting factors for SSc-related interstitial lung disease due to
repeated microaspirations of acid gastric contents in the
lungs. In a recent study, severity of esophageal manometric
changes was associated with higher prevalence of interstitial 10.1.7 Pulmonary Arterial Hypertension
lung disease (57% with severe involvement versus 18% with-
out involvement) on high-resolution computed tomography
Myth: Echocardiography can diagnose PAH in SSc.
(Marie et al. 2001). At 2-year follow-up, patients with severe
esophageal motor impairment had a greater decline in DLCO Reality: Although following Doppler echocardiograms regu-
(−16% of the predicted value) when compared with those larly to estimate the pulmonary artery pressure using the tri-
without any manometric changes (+1% predicted). However, cuspid regurgitant jet, right heart catheterization is essential to
other authors have failed to show this association. In other confirm the diagnosis and exclude other pathological entities
respiratory disorders (e.g., asthma), treatment of GERD (Arcasoy 2003). Echocardiograms give both false-negative
improves symptoms and PFT parameters. and (more likely) false-positive results.
In summary, patients with SSc and early interstitial lung
Myth: PAH occurs late in SSc.
disease should be managed aggressively with antireflux mea-
sures (e.g., elevating the head of the bed, not eating for sev- Reality: PAH often occurs among limited SSc patients who
eral hours before bedtime), proton-pump inhibitors, antacids, are older, i.e., they have had longstanding disease or disease
and prokinetic therapy. that was subclinical for years before the diagnosis was made.
In contrast, among patients with diffuse, SSc, PAH can occur
Pearl: Certain baseline clinical, physiological, and radio- at any age or duration of disease. Thus, regular monitoring
logical variables inform the treatment of SSc-related intersti- for PAH by routine echocardiography is important (Al-Dhaher
tial lung disease. et al. 2008).
Comment: The performance of pulmonary function tests Pearl: An FVC%/DLCO% ratio of greater than 1.8 is a help-
with DLCO measurement every 6–12 months for the first 5 ful predictor for patients at increased risk for PAH, regard-
years after the onset of symptoms or signs attributable to SSc less of the mechanism of PAH.
is a reasonable approach to longitudinal management. Which
SSc patients should be offered treatment for interstitial lung Comment: A falling DLCO occurs well in advance of PAH in
disease? Several categories of patients are good candidates some patients. PAH causes impaired gas exchange, which is
for treatment: reflected by low DLCO measurements. Isolated PAH in SSc
is strongly associated with a marked increase in the ratio of
• Patients whose FVC has declined at least 7% within 6 the percentage predicted FVC to the percentage predicted
months of the last measurement. DLCO, i.e., the FVC%/DLCO% ratio (Fig. 10.14). In patients
• Patients whose FVC is less than 70% predicted at the time with classic PAH related to SSc, interstitial lung disease is
of presentation may also benefit from therapy. minimal but impaired gas exchange secondary to pulmonary
10 Systemic Sclerosis (Scleroderma) and Raynaud’s Phenomenon 89

1.0

0.9
% FVC / % DLCO < 1.8 n=337

Probability of Survival
Fig. 10.14 Those with a low
DLCO when compared with FVC
0.8
are more apt to die (i.e., FVC%
predicted/DLCO % predicted of
>1.8 (or in another study >2 have
increased mortality). The better 0.7
survival is curve contains both p = .007
those with interstitial lung disease
and those without any lung
involvement and the lower 0.6
survival curve is those with PAH
alone or PAH that is occurring in % FVC / % DLCO ≥ 1.8 n=169
patients with interstitial lung
disease (Courtesy of Dr. James 0.5
Seibold) 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

vascular injury is the rule. The significant intimal prolifera- women with SSc were studied by bicycle ergometer (Morelli
tion of the pulmonary vessels leads to a marked decrease in et al. 2000). The results indicated a marked limitation of
the DLCO. This typically develops over many years and exercise performance among SSc patients when compared
leads to a high FVC%/DLCO% ratio (Steen 2003a). with normal controls. Patients with baseline pulmonary
A high FVC%/DLCO% ratio has also been observed in hypertension had an even further decrease in exercise per-
patients who have a low FVC because of interstitial lung dis- formance compared with SSc patients who did not have
ease. A disproportionately low DLCO in this setting implies elevations in their baseline pulmonary pressures. More
injury to both the pulmonary vasculature as well as to the recently, patients with SSc judged to be free of both cardiac
alveolar diffusing surface. Thus, the FVC%/DLCO% ratio is and pulmonary disease were confirmed to have reduced V02
high in this setting. Many of these patients have only moder- peak but also a reduced metabolic equivalent at V02 peak
ate fibrosis and have not been hypoxic, so their pulmonary (De Oliviera et al. 2007). Although not directly measured,
hypertension is more likely to be vascular than to be related this infers that muscle perfusion is an unrecognized influ-
to parenchymal lung disease. However, even the patients ence on exercise capacity in SSc.
with severe fibrosis, hypoxia, and secondary pulmonary The 6MWT may be helpful to measure change in short
hypertension can have a severe decrease in the DLCO and a term intervention for a specific problem (i.e., pulmonary
high ratio (Steen et al. 2007). fibrosis, pulmonary hypertension, myopathy), but further
validation studies need to be done in SSc.
Myth: A reduced 6 min walk is reflective of pulmonary hyper-
tension in SSc. Myth: SSc patients with significant PAH will report dyspnea.
Reality: The 6 min walk test (6MWT) is a standardized mea- Reality: SSc patients adapt often to PAH, as this condition is
sure of submaximal exercise capacity that has served as the sometimes slow in onset. They complain of dyspnea less
primary measure of outcome in studies of PAH. The 6MWT often than do patients with idiopathic PAH. Questions such
has never been validated fully in SSc. However, a variety of as “Has your breathing changed over the last year?” might be
data suggest that many nonpulmonary aspects of SSc con- more useful than “Are you short of breath?” Other useful
tribute affect the 6MWT, thereby blunting the ability of the queries are “Can you carry in the groceries from the car?”;
6MWT to discriminate between pulmonary issues and other “If not, what stops you (Pain? Poor hand function?
SSc-associated problems as a cause of an abnormal 6MWT. Dyspnea?)”; “How many flights can you climb?”; and”Does
Many factors can influence the distance walked: sources breathing worsen if you go outside in the cold?” (Mathai
of variability in test conduct, training effect, technician et al. 2009).
experience, subject encouragement, medication, supplemen-
Myth: All patients with elevated pulmonary artery pressures
tal oxygen, other activities on day of testing, decondition-
demonstrated by echocardiography will progress to PAH.
ing, and the effect of musculoskeletal conditions.
Investigators have studied exercise capacity in SSc patients Reality: Echocardiography estimates the pulmonary arte-
with and without pulmonary involvement. In one study, 18 rial pressure and often overestimates it. In addition, upon
90 J. E. Pope et al.

follow-up of patients with an elevated pulmonary artery sys- Reality: Scleroderma renal crisis usually occurs in patients
tolic pressure >35 mmHg on echocardiogram, 65% did not with diffuse SSc and is almost never seen in patients who are
progress over the next 3 years. In addition, most studies of anticentromere antibody positive. However, scleroderma
SSc patients show that 25% have echocardiograms that sug- renal crisis may emerge at a time when the full disease phe-
gest an elevated pulmonary artery pressure, but only 10–15% notype of the patient is not yet clear and diffuse cutaneous
have PAH that is class III or greater PAH (Steen 2005; disease is absent. Some who develop scleroderma renal cri-
Arcasoy 2003; Pope 2005a). sis have very early disease with less than a year of symp-
toms, such as carpal tunnel syndrome, arthritis, fatigue,
Myth: A normal echocardiogram in SSc excludes PAH.
Raynaud’s phenomenon, and swollen fingers or legs, and
Reality: Although there are more false-positive echocardio- may have been considered to have undifferentiated connec-
grams in the estimation of pulmonary artery pressure than tive tissue disease (Steen 2003b). In short, the full disease
false-negative ones, there are some patients in whom an esti- phenotype may not have emerged at the time scleroderma
mate of the pulmonary artery pressure cannot be obtained renal crisis strikes.
because no tricuspid regurgitation is evident. Accurate Patients with the highest risk for renal crisis are those
echocardiographic measurements are operator-dependent with an anti-RNA polymerase III antibody (Steen 2005).
and can vary according to the precise placement of the probe. These patients often do not have Raynaud’s phenomenon at
Thus, if PAH is suspected but the echocardiogram does not the onset of their disease and the ANA may be negative if the
appear to support this diagnosis, other procedures need to be assay is performed by ELISA, since RNA polymerase III is
considered: right heart catheterization; a magnetic resonance not identified by the ELISA ANA assay. In addition, since
imaging study of the heart, or exercise echocardiography. In scleroderma renal crisis occurs (or recurs) late in the illness
addition, one should also consider the possibility that a fall- in almost 20% of patients, diffuse cutaneous skin changes
ing PA pressure in a patient with severe PAH may indicate a may have remitted. Such patients have a history of severe
failing right ventricle (Arcasoy 2003). skin disease and are likely to have significant hand contrac-
tures (Fig. 10.15).
Pearl: A pericardial effusion in SSc is often related to PAH.
Pearl: Scleroderma renal crisis may be the first manifesta-
Comment: Obstruction of drainage of the heart can cause a
tion of SSc.
pericardial effusion when the right-sided heart pressures are
high, such as in PAH. The presence of a moderate to large Comment: In some patients, nonspecific symptoms and signs
pericardial effusion in SSc is usually a poor prognostic sign such as puffy fingers (Fig. 10.16), Raynaud’s phenomenon,
(Thompson and Pope 1998). and GERD had not yet been recognized as manifestations of
SSc when scleroderma renal crisis develops (Steen 2003b).
Keen diagnostic acumen and a high degree of suspicion are
10.1.8 Cardiac Disease required to make the diagnosis in this setting.
Myth: Treatment with an ACE inhibitor prevents scleroderma
Pearl: Treat congestive heart failure from a cardiomyopathy in renal crisis.
SSc the same way you would treat other cardiomyopathies.
Reality: ACE inhibitors have changed patient outcomes in
Comment: Cardiac involvement in SSc is a major cause of scleroderma renal crisis dramatically (Steen 2003b).
death, but it can be subtle and is often under-recognized. Scleroderma renal crisis is driven by a rennin-mediated
Diastolic dysfunction is particularly common in SSc. The hypertension. Thus, it was anticipated prophylaxis with ACE
treatments for congestive heart failure in SSc are the same as inhibitors would prevent renal crisis. Unfortunately, this is
those employed for this condition in other settings: angio- not the case. Two reviews of patients with renal crisis both
tensin converting enzyme (ACE) inhibitors, diuretics, and demonstrated that patients who received ACE-inhibitors
beta blockers. Attention to traditional cardiac risk factors prior to the onset of scleroderma renal crisis actually had
such as hypertension, smoking, and hyperlipidemia is also worse outcomes when compared with those without previous
important. Patients with SSc are not immune to atherosclero- ACE inhibitor use (Penn et al. 2007; Teixeira et al. 2007).
sis or ischemic cardiomyopathy (Steen 2000). One explanation for this is that the dose of ACE inhibitors
used in patients who are not hypertensive is insufficient to
control the excessive outpouring of rennin that occurs acutely
10.1.9 Scleroderma Renal Crisis from the juxtaglomerular apparatus in scleroderma renal cri-
sis. The acute response may be blunted in such patients, but
Myth: Scleroderma renal crisis occurs only in patients with instead they develop a more chronic, less reversible process.
diffuse SSc. The best outcomes occurred in patients with the highest
10 Systemic Sclerosis (Scleroderma) and Raynaud’s Phenomenon 91

Reality: Normotensive cases have been described. A high

index of suspicion is required if a decline in renal function is
observed in a patient with SSc (Teixeira et al. 2007). SRC is
defined as new onset hypertension (occurs in >90%), ele-
vated creatinine (50% at onset of diagnosis of SRC) and
microangiopathic hemolytic anemia (50%).
Myth: Once SSc patients go on dialysis, the renal “ball-
game” is over.
Reality: ACE inhibition should be continued even in patients
who require dialysis because of scleroderma renal crisis. A
significant number of such patients are eventually able to dis-
continue dialysis, even after undergoing renal replacement
therapy for many months.
Pearl: If a patient has scleroderma renal crisis, use an ACE
inhibitor and ADD any other antihypertensive agent required
to control the blood pressure as rapidly as possible.
Fig. 10.15 Hand contractures in severe, long-standing diffuse systemic
sclerosis. This patient’s disease had its onset in childhood Comment: Before the introduction of ACE inhibitors, sclero-
derma renal crisis was the leading cause of death among
patients with diffuse SSc (Steen and Medsger 2001, 2003).
When prescribing ACE inhibitors in scleroderma renal crisis,
it is helpful to use captopril because of its three-times-a-day
dosing. However, likely any ACE inhibitor can be used and
the dose can and should be increased rapidly to the maximal
dose or until adequate blood pressure control is achieved.
The ACE inhibitor should never be stopped, even in the face
of a rising serum creatinine.
Other antihypertensive can also be added if necessary, but
maximizing the ACE inhibitor dose is the chief priority.
Myth: Angiotensin II inhibitors are also useful in the treat-
ment of scleroderma renal crisis.
Reality: Angiotensin II inhibitors do not lead to consistent
reductions of bradykinins, in contrast to ACE inhibitors. This
is believed to be the reason that angiotensin II inhibitors are
Fig. 10.16 A man with early diffuse systemic sclerosis, demonstrating less effective in the treatment of scleroderma renal crisis (Steen
erythema and puffiness of the hands. The patient also had significant 2000; Steen and Medsger 2001; Rhew and Barr 2004).
cutaneous pruritus
Myth: Low-dose glucocorticoids do not increase the risk of
scleroderma renal crisis.
blood pressure levels but the lowest serum creatinine mea- Reality: Glucocorticoids at high dose do appear to increase
surements. These variables may be a function of the acuity of the risk of scleroderma renal crisis and some data indicated
the process. that even low doses of prednisone (e.g., >10 mg/day) increase
In summary, although ACE inhibitors can be lifesaving the risk of scleroderma renal crisis. For this reason, it is wise
drugs in patients with scleroderma renal crisis, one cannot to avoid the use of glucocorticoids in patients with early dif-
assume that having the patient on such a medication as fuse SSc unless an absolute indication for these medications
prophylaxis or to treat baseline hypertension will prevent the is present (Steen and Medsger 2001).
development of this potentially lethal disease complication.
Patients with diffuse SSc should be instructed to take their
blood pressures regularly and report increases to their doctors. 10.1.10 Musculoskeletal Manifestations
Myth: Scleroderma renal crisis always presents with hyper-
tension. Myth: SSc associated arthritis is not erosive.
92 J. E. Pope et al.

Fig. 10.17 This SSc patient has

diffuse SSc, with inflammatory
arthritis. She had evidence of
ankylosis and erosive arthritis at
the wrists and a pencil and cup
type deformity of the R second
MCP. There is distal tuft
resportion at many fingertips and
a Z deformity (or Boutinneire) at
the thumbs. She has periarticular
osteopenia. Ulnar styloids have
eroded away and architecture of
the proximal carpel bones is
destroyed

Fig. 10.18 This left-hand

dominant SSc patient has
subluxed MCPs on dominant
hand with MCP inflammatory
arthritis. Note loss of distal finger
pulp, acro-osteolysis on L third
distal phalanx and calcinosis such
as distal tip of L thumb

Reality: The arthritis of SSc can be erosive and can mimic hands and pain that is difficult to localize to joints and
a destructive seronegative arthritis (Figs. 10.17 and 10.18) refractory to treatment. This discomfort generally improves
(Pope 2003b, 2005b). However, the more common scenario over several months as other disease features become
is that patients with early-onset diffuse SSc have puffy manifest.
10 Systemic Sclerosis (Scleroderma) and Raynaud’s Phenomenon 93

neuropathies such as the pure sensory neuropathy that some-

times occurs in Sjögren’s syndrome or the mononeuritis
multiplex of vasculitis are notable for their absence in SSc
(Lori et al. 1996; Steen et al. 1994; Tashkin et al. 2006;
Khanna et al. 2008).

References

Al-Dhaher FF, Pope JE, Ouimet JM. Determinants of morbidity and

mortality of systemic sclerosis in Canada. Semin Arthritis Rheum.
2009 (in press)
American Thoracic Society Statement. Guidelines for the six-minute
walk test. Am J Respir Care Med. 2002;166:111–7
Arcasoy SM. Echocardiographic assessment of pulmonary hyperten-
sion in patients with advanced lung disease. Am J Respir Crit Care
Med. 2003;167:735–40
Fig. 10.19 This radiograph shows thumb calcinosis distally. The third Arnett FC, Cho M, Chatterjee S, et al Familial occurrence frequen-
distal phalanx is missing and the dip is ankylosed on that finger. In addi- cies and relative risks for systemic sclerosis (scleroderma) in
tion the fifth DIP and first IP joints are beginning to ankylose three United States cohorts. Arthritis Rheum. 2001;44: 1359–62
Behr J, Vogelmeier C, Beinert T, et al Bronchoalveolar lavage for evalu-
ation and management of scleroderma disease of the lung. Am J
Pearl: Calcinosis on hand radiographs can be a tip-off to the Respir Crit Care Med. 1996;154:400–6
presence of SSc. Calamia KT, Scolapio JS, Viggiano TR. Endoscopic YAG laser treat-
ment of watermelon stomach (gastric antral vascular ectasia) in
patients with systemic sclerosis. Clin Exp Rheumatol. 2000;
Comment: Figure 10.19 shows distal calcinosis of the thumb. 18(5):605–8
The third distal phalanx is missing and the distal interphalan- Claman HN, Giorno RC, Seibold JR. Endothelial and fibroblastic acti-
geal (DIP) joint on that finger is ankylosed. In addition, the vation in scleroderma. The myth of the “uninvolved skin”. Arthritis
Rheum. 1991;34:1495–501
fifth DIP and the first interphalangeal joint are beginning to
Clements P, Lauchenbruch P, Furst D, Paulus H. The course of skin
undergo ankylosis. involvement in systemic sclerosis over three years in a trial
of chlorambucil versus placebo. Arthritis Rheum. 1993;36:1575–9
Pearl: Palpable tendon friction rubs herald an evolution to Clements PJ, Medsger TA Jr, Feghali CA. Cutaneous involvement in
diffuse SSc. systemic sclerosis. In: Clements PJ, Furst DE, editors. Chapter 7
Systemic sclerosis. 2nd ed. Philadelphia, PA: Lippincott Williams
Comment: Tendon friction rubs result from fibrous teno- and Wilkins; 2004. p. 129–50
Cutolo M, Pizzorni C, Tuccio M, et al Nailfold videocapillaroscopic pat-
synovitis or bursitis and are detected frequently in early terns and serum autoantibodies in systemic sclerosis. Rheumatology
diffuse SSc. In patients whose skin thickening is limited at (Oxford) 2004;43(6):719–26
the time of examination, patients with one or more friction De Oliviera NC, dos Santos Sabbag LM, et al Reduced exercise capac-
rubs have a greater risk of progression to diffuse skin ity in systemic sclerosis patients without pulmonary involvement.
Scand J Rheumatol. 2007;36:458–61
thickening (36 vs. 13%, p = 0.0007) (Steen and Medsger Ebert EC. Esophageal disease in scleroderma. J Clin Gastroenterol.
1997a). Tendon friction rubs are a marker for a poor prog- 2006;40:769–75
nosis and are associated with a increased mortality risk Furst DE, Saab M, Clements P. Clinical and serological comparison of
(Medsger 2009). 17 PSS and 17 CREST syndrome patients for sex, age and matched
disease duration. Ann Rheum Dis. 1984;43:794–801
Goh NS, Veeraraghavan S, Desai SR, et al Bronchoalveolar lavage cel-
lular profiles in patients with systemic sclerosis-associated intersti-
tial lung disease are not predictive of disease progression. Arthritis
10.1.11 Peripheral Neuropathy in SSc Rheum. 2007;56:2005–12
Herrick A. Pathogenesis of Raynaud’s phenomenon. Rheumatology
(Oxford) 2005;44(5):587–96
Pearl: Aside from unusual cases of compression neuropathies, Hirschl M, Hirschl K, Lenz, et al Transition from primary Raynaud’s
nervous system involvement is not characteristic of SSc. phenomenon to secondary Raynaud’s phenomenon identified by
diagnosis of an associated disease results of ten years of prospective
Comment: Carpal tunnel syndrome can develop in early SSC surveillance. Arthritis Rheum. 2006;54(6):1974–81
because of edema in the skin and subcutaneous tissues. Hoyles RK, Ellis RW, Wellsbury J, et al A multicenter, prospective, ran-
domized, double-blind, placebo-controlled trial of corticosteroids
Carpal tunnel syndrome and other compression neuropathies
and intravenous cyclophosphamide followed by oral azathioprine
such as trigeminal nerve compression may occur in SSc for the treatment of pulmonary fibrosis in scleroderma. Arthritis
because of fibrosis, but the latter is uncommon. Other overt Rheum. 2006;54(12):3962–70
94 J. E. Pope et al.

Hudson M, Thombs BD, Panopalis P, et al Quality of life in patients with Pope J. Scleroderma Overlap Syndromes. Curr Opin Rheumatol. 2002;
systemic sclerosis compared to the general population and patients 14(6):704–10
with other chronic conditions. J Rheumatol 2009;36:768–72 Pope J. The diagnosis and treatment of Raynaud’s phenomenon: A
Khanna D. Health-related quality of life: A primer with a focus on scle- practical approach. Drugs 2007;67(4):517–25
roderma. Scleroderma Care Res. 2006;3(2):3–13 Pope J. Raynaud’s phenomenon (primary). Clin Evidence. 2003a;9:
Khanna D, Furst DE, Clements PJ, et al Oral cyclophosphamide for 1339–48
active scleroderma lung disease: a decision analysis. Med Decis Pope J, Lee P, Baron M, Dunne J, Smith D, Docherty PS. The preva-
Making. 2008;28(6):926–37 lence of elevated pulmonary arterial pressures as measured by
Kuwana M, Kaburaki J, Okano Y, et al Clinical and prognostic associa- echocardiography in a large multi-centre cohort of systemic sclero-
tions based on serum antinuclear antibodies in Japanese patients sis subjects in Canada. J Rheumatol. 2005a;32(7):1273–8
with systemic sclerosis. Arthritis Rheum. 1994;37(1):75–83 Pope JE. Musculoskeletal involvement in scleroderma. Rheum Dis Clin
Kuwana M, Feghali CA, Medsger TA Jr, Wright TM. Autoreactive T North Am Scleroderma 2003b May 29:391–408
cells to topoisomerase I in monozygotic twins discordant for sys- Pope JE. Other manifestations of mixed connective tissue disease.
temic sclerosis. Arthritis Rheum. 2001;44:1654–59 Rheum Dis Clin North Am. 2005b;31(3):519–33
Lock G, Holstege A, Lang B, Scholmerich J. Gastrointestinal manifes- Pope JE, Thompson A. Antimitochondrial antibodies and their signifi-
tations of progressive systemic sclerosis. Am J Gastroenterol. cance in diffuse and limited scleroderma. J Clin Rheumatol. 1999;
1997;92(5):763–71 5(4):206–9
Lori S, Matucci-Cerinic M, Casale R, et al Peripheral nervous system Pope JE, Shum D, Gottschalk R, et al Increased pigmentation in sclero-
involvement in systemic sclerosis: The median nerve as target struc- derma. J Rheumatol. 1996;23:1912–16
ture. Clin Exp Rheumatol. 1996;14(6):601–05 Pope JE, Al-Bishri J, Al-Azem H, Ouimet JM. The temporal relationship
Luggen M, Belhorn L, Evans T, Fitzgerald O, Spencer-Green G. The of Raynaud’s phenomenon and features of connective tissue disease
evolution of Raynaud’s phenomenon: A longterm prospective study. in rheumatoid arthritis. J Rheumatol. 2008;35(12):2329–33
J Rheumatol. 1995;22(12):2226–32 Rhew EY, Barr, WG. Scleroderma renal crisis: new insights and devel-
Lynch JP 3rd, Saggar R, Weigt SS, et al Usual interstitial pneumonia. opments. Curr Rheumatol. Rep 2004;6(2):129–36
Semin Respir Crit Care Med. 2006;27(6):634–51 Rigamonti C, Shand LM, Feudjo M, et al Clinical features and progno-
Marie I. Gatrointestinal involvement in systemic sclerosis. Presse Med. sis of primary biliary cirrhosis associated with systemic sclerosis.
1996;35:1952–65 Gut 2006;55(3):388–94
Marie I, Dominique S, Levesque H, et al Esophageal involvement and Sacks DG, Okano Y, Steen VD, et al Isolated pulmonary hypertension in
pulmonary manifestations in systemic sclerosis. Arthritis Rheum. systemic sclerosis with diffuse cutaneous involvement: Association
2001;45(4):346–54 with serum anti-U3RNP antibody. J Rheumatol. 1996; 23(4):639–42
Mathai SC, Hummers LK, Champion HC, et al Survival in pulmonary Salliot C, Mouthon L, Ardizzone M, et al Sjögren’s syndrome is associ-
hypertension associated with the scleroderma spectrum of diseases: ated with and not secondary to systemic sclerosis. Rheumatology
Impact of interstitial lung disease. Arthritis Rheum. 2009;60(2): 569–77 (Oxford) 2007;46(2):321–6
Medsger TA Jr. Unpublished data from the University of Pittsburgh Shand L, Lung M, Nihtyanova S, et al Relationship between change in
Scleroderma Databank, 2009 skin score and disease outcome in diffuse cutaneous systemic scle-
Meli M, Gitzelmann G, Koppensteiner R, et al Predictive value of nail- rosis. Application of a latent linear trajectory model. Arthritis
fold capillaroscopy in patients with Raynaud’s phenomenon. Clin Rheum. 2007;56(7):2422–31
Rheumatol. 2006;25(2):153–8 Shibukawa G, Irisawa A, Sakamoto N, et al Gastric antral vascular ecta-
Merkel PA, Herlyn K, Martin RW, et al Measuring disease activity and sia (GAVE) associated with systemic sclerosis: Relapse after endo-
functional status in patients with scleroderma and Raynaud’s phe- scopic treatment by argon plasma coagulation. Intern Med. 2007;46
nomenon. Arthritis Rheum. 2002;46(9):2410–20 (6):279–83
Morelli S, Ferrante L, Sgreccia A, et al Pulmonary hypertension is asso- Silver RM, Miller KS, Kinsella, et al Evaluation and management of
ciated with impaired exercise capacity in patients with systemic scleroderma lung disease using bronchoalveolar lavage. Am J Med.
sclerosis. Scand J Rheumatol. 2000;29:236–42 1990;88:470–6
No authors: American Thoracic Society. Idiopathic pulmonary fibrosis: Steen V. Epidemiology of systemic sclerosis. In: Hochberg MC, Silman
diagnosis and treatment. International consensus statement. American AJ, Smolen JS, Weinblatt ME, Weisman MH, editors. Rheumatology.
Thoracic Society (ATS), and the European Respiratory Society Vol. 2. Toronto: Elsevier(Mosby); 2003. p. 1455–61
(ERS). Am J Respir Crit Care Med. 2000;161:646–64 Steen V. Advancements in diagnosis of pulmonary arterial hypertension
Pavlov-Dolijanovic S, Damjanov N, Ostojic P, et al The prognostic in scleroderma. Arthritis Rheum. 2005a;52(12):3698–700
value of nailfold capillary changes for the development of connective Steen V, Medsger T. Severe organ involvement in systemic sclerosis
tissue disease in children and adolescents with primary Raynaud’s with diffuse scleroderma. Arthritis Rheum. 2000;43(11):2437–44
phenomenon: A follow-up study of 250 patients. Pediatr Dermatol. Steen VD. Scleroderma renal crisis. Rheum Dis Clin North Am.
2006;23(5):437–42 2003;29(2):315–33
Penn H, Howie AJ, Kingdon EJ, Bunn CC, Stratton RJ, Black CM, et al Steen VD. Autoantibodies in systemic sclerosis. Semin Arthritis
Scleroderma renal crisis: Patient characteristics and long-term out- Rheum. 2005b;35(1):35–42
comes. QJM 2007;100(8):485–94 Steen VD, Medsger TA Jr. The palpable tendon friction rub: An impor-
Perera A, Fertig N, Lucas M, et al Clinical subsets, skin thickness pro- tant physical examination finding in patients with systemic sclero-
gression rate, and serum antibody levels in systemic sclerosis sis. Arthritis Rheum. 1997a;40:1146–51
patients with anti-topoisomerase I antibody. Arthritis Rheum. 2007; Steen VD, Medsger TA Jr. The value of the Health Assessment Questionnaire
56:2740–6 and special patient-generated scales to demonstrate change in systemic
Plastiras SC, Karadimitrakis SP, Ziakas PD, et al Scleroderma lung: sclerosis patients over time. Arthritis Rheum. 1997b;40:1984–91
Initial forced vital capacity as predictor of pulmonary function Steen VD, Medsger TA Jr. Long-term outcomes of scleroderma renal
decline. Arthritis Rheum. 2006;55(4):598–602 crisis. Ann Intern Med. 2000;133(8):600–3
Poole JL, Williams CA, Bloch DA, et al Concurrent validity of the Steen VD, Ziegler GL, Rodnan GP, Medsger TA Jr. Clinical and labora-
Health Assessment Questionnaire disability index in scleroderma. tory associations of anticentromere antibody in patients with pro-
Arthritis Care Res. 1995;8:189–93 gressive systemic sclerosis. Arthritis Rheum. 1984;27(2):125–31
10 Systemic Sclerosis (Scleroderma) and Raynaud’s Phenomenon 95

Steen VD, Conte C, Owens GR, Medsger TA Jr. Severe restrictive lung Thompson AE, Shea B, Welch V, et al Calcium Channel Blockers for
disease in systemic sclerosis. Arthritis Rheum. 1994;37(9): 1283–9 Raynaud’s Phenomenon in Progressive Systemic Sclerosis. Arthritis
Steen VD, Lucas M, Fertig N, Medsger TA Jr. Pulmonary arterial hyper- Rheum. 2001;44(8):1841–7
tension and severe pulmonary fibrosis in systemic sclerosis patients Walker UA, Tyndall A, Czirjak L, et al Clinical risk assessment of organ
with a nucleolar antibody. J Rheumatol. 2007;34(11):2230–5 manifestations in systemic sclerosis: A report from the EULAR
Strange C, Bolster MB, Roth MD, et al Bronchoalveolar lavage and scleroderma trials and research group database. Ann Rheum Dis.
response to cyclophosphamide in scleroderma interstitial lung dis- 2007;66(6):754–63
ease. Am J Respir Crit Care Med. 2008;177:91–8 WGET Research Group. Design of the Wegener’s Granulomatosis
Tashkin DP, Elashoff R, Clements PJ, et al Cyclophosphamide versus Etanercept Trial (WGET). Controlled Clinical Trials. 2002;23:
placebo in scleroderma lung disease. N Engl J Med. 2006;354(25): 450–68
2655–66 WGET Research Group. Etanercept plus standard therapy for Wegener’s
Tashkin DP, Elashoff R, Clements PJ, et al Effects of 1-year treat- granulomatosis. N Engl J Med. 2005;352(4):351–61
ment with cyclophosphamide on outcomes at 2 years in sclero- White B, Moore WC, Wigley FM, et al Cyclophosphamide is associ-
derma lung disease. Am J Respir Crit Care Med. 2007;176(10): ated with pulmonary function and survival benefit in patients with
1026–34 scleroderma and alveolitis. Ann Intern Med. 2000;132:947–54
Teixeira L, Mahr A, Berezne A, et al Scleroderma renal crisis, still a life- Witt C, Borges AC, John M, et al Pulmonary involvement in diffuse
threatening complication. Ann N Y Acad Sci. 2007;1108:249–58 cutaneous systemic sclerosis: Broncheoalveolar fluid granulocyto-
Thompson A, Pope JE. A study of the frequency of pericardial and pleu- sis predicts progression of fibrosing alveolitis. Ann Rheum Dis.
ral effusions in scleroderma. Br J Rheumatol. 1998;37: 1320–3 1999;58:635–40
 Nephrogenic Systemic Fibrosis
and Other Scleroderma Mimickers 11
Mark A. Perazella, Janet E. Pope, and Shawn E. Cowper

Myth: Nephrogenic systemic fibrosis is the “systemic” form

11.1 Overview of Nephrogenic Systemic of nephrogenic fibrosing dermopathy.
Fibrosis and Other Scleroderma
Mimickers Reality: “Nephrogenic fibrosing dermopathy” was the first
popular name given to a disorder that has carried multiple
monikers in its relatively brief lifetime. Other examples are
› Nephrogenic systemic fibrosis (NSF) is a systemic shown in Table 11.1. Nephrogenic systemic fibrosis (NSF) is
fibrosing disorder that mimics systemic sclerosis (scle- currently the preferred term, because it reflects the concept
roderma) and is strongly associated with exposure to that NSF is a systemic disorder and that dialysis is not
gadolinium in the setting of renal insufficiency. required for its development (Cowper et al. 2006b).
› In addition to the skin, fibrotic changes can occur in
the lungs, diaphragm, muscles, heart, and esophagus. Myth: NSF develops more commonly in certain subgroups of
› Both clinical and histopathological findings are people, such as women, Caucasians, and the elderly.
required to confirm the diagnosis of NSF. Reality: NSF afflicts patients of both genders and all ages
› No cases of NSF have been described in patients with and race. In one report from 2008, the age of the patients
normal kidney function or mild underlying renal who had developed NSF ranged from 8 to 87 years (mean
disease. 46.4 years). The male:female ratio approximates one. The
› The more severe the kidney dysfunction and the higher condition has been reported in Caucasians, Hispanics,
the dose of gadolinium, the greater is the risk of NSF. African–Americans, and Asians, and described in North and
› NSF develops subacutely in most cases. It usually South America, Europe, Australia, and Asia (Cowper 2008)
becomes apparent first in the lower extremities with
symmetrical edema that may involve the upper extrem- Myth: NSF is distinguished from other fibrosing disorders
ities. The edema may be associated with erythema, primarily by its tendency to involve the skin as opposed to
blisters, and bullae. Upon resolution of the edema, the internal organs.
indurated skin is nonpitting and “unpinchable.” Reality: NSF is a systemic fibrosing disorder in which the
› NSF tends to spare the skin of the face, in contrast to most readily recognized clinical manifestations are cutane-
both scleroderma and scleromyxedema. ous. The description of skin involvement as a primary dis-
› Yellow scleral plaques are described in many cases of ease feature led initially to the name of “nephrogenic
NSF. fibrosing dermopathy” (NFD) for this condition (Cowper
› The histopathological features of NSF are consistent et al. 2001). However, in 2003, fibrosis was detected in other
with a dermal reaction to injury: fibrocyte-like cells, organs in an “NFD” patient who died from complications of
histiocytes, dermal dendritic cells, scar-like fibrosis, the disorder and underwent autopsy (Ting et al. 2003;
mucin, edema, calcification, and ossification. Swaminathan et al. 2008). Fibrotic involvement was discov-
› Scleroderma mimickers such as morphea, scleromyxe- ered in the diaphragm, muscles, heart, and esophagus. In
dema, and eosinophilic fasciitis are generally distin- addition, the systemic nature of "NFD" was confirmed by
guished from scleroderma by their absence of identification of the spindle cells within NFD tissues as bone
extracutaneous disease, Raynaud’s phenomenon, and marrow-derived cells by their CD34/procollagen-1 coex-
antinuclear antibodies. pression with special train (Cowper et al. 2003). This cell,

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 97

DOI:10.1007/978-1-84800-934-9_11, © Springer Science + Business Media B.V. 2009
98 M. A. Perazella et al.

Table 11.1 Names given to the syndrome now known as nephrogenic bound state. This form of the disorder is highly morbid and
systemic fibrosis. associated with increased mortality due to complications
Nephrogenic fibrosing dermopathy such as hip fractures from falls and systemic involvement by
Scleromyxedema-like cutaneous disease in renal dialysis patients the fibrosing process.
Fibrosing dermopathy of dialysis
It is more common for NSF to develop subacutely. In most
Dialysis-associated systemic fibrosis
Scleromyxedema-like fibromucinosis
patients, NSF begins with symmetrical lower extremity
edema that subsequently involves the upper extremities
(Cowper 2008). The edema may be associated with erythema,
blisters, and bullae. Upon resolution of the edema, the indu-
termed the “circulating fibrocyte,” has been implicated as rated skin is nonpitting and “unpinchable” (Fig. 11.1).
the cell principally responsible for the fibrosing process that Resolution of the edema leads to the formation of firm
characterizes NSF. plaques that progress to brawny induration and thickening of
Numerous other reports have confirmed the occurrence of the dermis. Other dermal manifestations include superficial
fibrosis within systemic organs, including the lungs, kidneys, papules, nodules, and hyperpigmentation. Deeper skin involve-
and other tissues. Thus, NSF must be distinguished from ment is manifested as cobblestoning, peau d’orange skin
other fibrosing disorders by the combination of clinical fea-
tures and histopathological findings. Clinical features that
help differentiate NSF from other disorders are:
• Its tendency to spare the skin of the face, in contrast to
both scleroderma and scleromyxedema.
• Its lack of any association with autoantibodies, e.g., anti-
centromere or antitopoisomerase-3 antibodies.
• Its absence of a paraproteinemia, which occurs commonly
in scleromyxedema.
• Finally, dermatopathologists can also distinguish among
NSF, scleroderma, and scleromyxedema by the absence
of inflammatory cells in NSF (Table 11.2) (Cowper et al.
2001).
Myth: Rapid and fulminant NSF, which is associated with
severe pain, limb immobility, and joint contractures, occurs
in more than half of patients.
Reality: Only about 5% of patients present in a fulminant
manner, with disease courses that advance rapidly over as
Fig. 11.1 NSF is characterized by edema that is nonpitting and
little as 2 weeks (Cowper 2007, 2008). In such settings, NSF unpinchable. Note the blanching of the fingertips of the person attempt-
is marked by severe pain and accelerated loss of mobility ing to pinch the skin (From Girardi 2008. Reprinted with permission
with joint contractures, wheelchair dependence, and a bed- from Elsevier)

Table 11.2 Clinical and histopathological findings of fibrosing disorders considered in patients with nephrogenic systemic fibrosis (NSF)
(Adapted from Cowper et al. 2001)
NSF Scleroderma, morphea, and eosinophilic fasciitis Scleromyxedema
Distribution Trunk and extremities Trunk, extremities, and face (particularly in Trunk, extremities, and face
(sparing of face) scleroderma)
Paraproteinemia Absent Absent Present
Systemic involvement Sometimes (myocar- Present (fibrosis of viscera, renal disease) Present (internal mucin deposition)
dium, pericardium,
pleura, diaphragm)
Density of dermal Increased numbers Normal numbers or decreased Increased numbers
fibroblasts
Interstitial mucin Present (early), but not Present, large collections sometimes Present, large collections
in large collections sometimes
Inflammatory cells Absent Present, mixed infiltrates Present, especially plasma cells
11 Nephrogenic Systemic Fibrosis and Other Scleroderma Mimickers 99

excluded based on a combination clinical features and

histopathological findings (Cowper 2008, personal commu-
nication). Other fibrosing disorders – e.g., systemic sclerosis
(scleroderma), scleromyxedema, morphea, lipodermatoscle-
rosis, and eosinophilic fasciitis – have clinical features that
overlap substantially with NSF. These conditions may require
laboratory testing or a skin biopsy for definitive diagnosis.
Myth: If dermal mucin deposits are absent, the diagnosis of
NSF can be excluded.
Reality: Excess dermal mucin is a commonly encountered
histopathological feature in cases of NSF (Fig. 11.3a). This
observation led initially to comparisons of the disorder to
Fig. 11.2 Peau d’orange skin changes are noted in the arm of a patient with scleromyxedema, which is also characterized by dermal
nephrogenic systemic fibrosis (With permission, Cowper 2001–2008) mucinosis and fibrosis.
Although mucin deposition is characteristic of NSF, this
finding is sometimes indistinct or even absent (Fig. 11.3b).
changes (Fig. 11.2), and dermal tethering (Cowper 2008). The absence of mucin deposition within the skin should not
These lesions are often extremely painful and may be associ- exclude the diagnosis of NSF if the clinical and pathological
ated with symptoms such as tightening, stiffness, burning, findings otherwise favor it (Cowper et al. 2008).
pruritus, paresthesias, and causalgia. Finally, flexion and Pearl: The diagnosis of NSF cannot be rendered accurately
extension contractures of large joints such as the hips, knees, without clinical and histopathological correlation.
elbows, hands, and feet develop in some patients, in processes
that evolve over many months. Comment: Neither the clinical nor the histopathological fea-
tures of NSF are unique. The clinical features of NSF can be
Pearl: Both clinical and histopathological findings are
mimicked by scleromyxedema, scleroderma, morphea, lipo-
required to confirm the diagnosis of NSF.
dermatosclerosis, eosinophilic fasciitis, and other entities.
Comment: NSF is a clinicopathological diagnosis (Cowper The histopathological findings in NSF are virtually identical
2001). The diagnosis can be classified as certain, possible, or to those of scleromyxedema, but also overlap with other

Figs. 11.3 The cutaneous

histopathological features of
nephrogenic systemic fibrosis
(NSF) may (a) or may not (b)
reveal obvious mucin deposition.
Mucin deposition is not required
to make a histopathological
diagnosis of NSF (Figure
courtesy of Dr. Cowper)
100 M. A. Perazella et al.

entities. The conclusions of published reports purported to end-stage renal disease, hemodialysis reduces the T1/2 to 2.1 h
describe NSF in the absence of histopathological evaluation but requires up to four dialysis sessions to clear more than
should be viewed with skepticism, particularly if they diverge 95% of gadolinium from the plasma.
from the conclusions of studies that have employed rigorous Peritoneal dialysis is less efficient than hemodialysis in
clinicopathological correlation (Cowper et al. 2008). clearing gadolinium. The T1/2 of gadolinium-based contrast is
approximately 53 h in a patient on peritoneal dialysis
Myth: NSF can develop in patients with any degree of kidney
(Perazella and Rodby 2007).
dysfunction.
Myth: All gadolinium-containing contrast agents are equally
Reality: Although NSF develops only in patients who suffer
associated with NSF.
from kidney disease, it has not been described in patients with
mild degrees of renal impairment. To be more specific, NSF Reality: The majority of published cases of NSF originate in
has not been reported in any individual with an estimated the United States, where six gadolinium-containing contrast
glomerular filtration rate (eGFR) that exceeds 30 mL/ agents are FDA-approved for MRI. In cases of NSF in
min/1.73 m2 (Grobner and Prischl 2008). A survey of the pub- which a gadolinium exposure could be identified with cer-
lished literature (~264 cases) observed that NSF had been tainty, approximately 90% were exposed to gadodiamide
detected in patients with acute kidney injury, chronic kidney (Omniscan®, GE Healthcare) and 10% to gadopentetate
disease (stages IV and V), and end-stage renal disease on dial- dimeglumine (Magnevist®, Bayer-Schering). Unpublished
ysis. Approximately 80% of patients with NSF have end-stage data and published case reports support this impression.
renal disease requiring dialysis. The other 20% suffers from While marketing statistics are not readily available, it is
either acute kidney injury (with or without dialysis) or stage generally accepted that, during this period, Magnevist® doses
IV or V chronic kidney disease (Grobner and Prischl 2008). probably slightly exceeded Omniscan® doses in the US. This
information suggests a higher propensity for Omniscan® to be
Myth: NSF can develop in any patient exposed to a gadolinium-
associated with NSF, although recently European authorities
based contrast agent.
have specifically banned the use of both agents in the renally
Reality: In early 2006, Grobner published the observation impaired (MHRA 2007). A variety of lines of evidence sug-
that patients with end-stage renal disease on dialysis devel- gest that GCCA with linear chelate molecules are more apt to
oped NSF within weeks to months of exposure to gadolin- induce NSF than those with macrocyclic chelates, and that
ium-based contrast (Grobner 2006). Subsequently, numerous “nonionic” formulations are more risky than “ionic” ones.
case series and case reports have verified the epidemiologi- These tendencies spring, in part, from the hypothesis that
cal association of gadolinium-based contrast exposure and gadolinium dechelation is at the heart of NSF induction and,
NSF. Five case–control studies in patients with end-stage in part, from the observation of the relative reported frequen-
renal disease reported odds ratios between 22 and 46 for the cies of NSF to various GCCA (Cowper 2008).
development of NSF following exposure to gadolinium (Deo
Myth: Confirmation of a gadolinium exposure is required for
2007; Collidge et al. 2007; Othersen et al. 2007; Marckmann
the diagnosis of NSF.
et al. 2007; Broome et al. 2007).
As of 2008, all patients reported in the medical literature Reality: NSF was defined by its clinical and pathological
and the FDA MedWatch adverse events reporting system had features several years before any association with gadolin-
underlying advanced kidney disease (eGFR <30 mL/ ium-containing contrast agents was made. Although the
min/1.73 m2). No cases of NSF have been described in association between gadolinium exposure and NSF is strong,
patients with normal kidney function or mild underlying it is not clear that gadolinium is responsible for all cases of
renal disease. this syndrome. If a patient satisfies the clinicopathological
Differences in the pharmacokinetics of gadolinium-based criteria for NSF, then that diagnosis is appropriate, regard-
contrast in patients with advanced kidney disease probably less of whether or not there is a known exposure to gadolin-
explain much of the risk for NSF development. Most gado- ium. Detection of gadolinium within involved tissue is not
linium-based contrast agents are cleared completely by required to make the diagnosis (Cowper et al. 2008).
glomerular filtration. Thus, as the GFR declines, the terminal
Myth: Liver failure is an independent risk factor for NSF.
half life (T1/2) of gadolinium increases (Perazella and Rodby
2007). For example, the T1/2 in the setting of normal kidney Reality: The admonition from the U.S. Food and Drug
function is 1.3–1.9 h, with 95% excreted within 24 h. In con- Administration that “patients with acute renal insufficiency of
trast, the T1/2 of gadolinium in the setting of kidney disease is any severity due to the hepatorenal syndrome or in the periop-
approximately 5 h for stage III chronic kidney disease, erative liver transplantation period” should not receive gado-
approximately 9 h for stage IV, and approximately 27 h for linium-containing contrast agents has led some practitioners
stage V (Perazella and Rodby 2007). Among patients with to infer erroneously that liver disease is an independent risk
11 Nephrogenic Systemic Fibrosis and Other Scleroderma Mimickers 101

Table 11.3 Other postulated cofactors in the development of nephrogenic angiography, phase-contrast angiography, and arterial spin
systemic fibrosis labeling MR techniques all provide excellent information
Presence of an inflammatory disease state (infection, inflammation) about blood vessels and blood flow (Dawson and Punwani
Presence of a thrombotic or hypercoagulable state 2008). Ultrasound with various Doppler techniques or con-
Recent surgical procedure (especially vascular procedures)
trast-enhancement (intravenous microbubbles) is another
Metabolic disturbance (hyperphosphatemia, hypercalcemia)
Iron overload or recent intravenous iron therapy
option. Color, power, and spectral Doppler studies can visu-
High-dose erythropoietin exposure alize the direction and magnitude of blood flow through
blood vessels, but these techniques are highly operator depen-
dent (Dawson and Punwani 2008). Computed tomography
without iodinated radiocontrast agents can provide all the
factor for NSF (FDA 2007). Liver failure as an isolated find- information necessary in some settings. Radiocontrast agents
ing has never been associated with NSF. pose significant risks of renal toxicity in patients with mar-
The FDA recommendations regarding the use of gadolinium- ginal renal function.
containing contrast agents are based on a risk assessment
Pearl: NSF-like lesions can be induced in rats injected with
that includes calculation of the estimated glomerular filtra-
gadolinium-containing contrast agents.
tion rate (eGFR) by the MDRD method. As the calculated
value for eGFR is not validated for either acute kidney injury Comment: There have been published and unpublished
or the hepatorenal syndrome, the Food and Drug reports indicating that cutaneous lesions clinically and histo-
Administration highlighted these exceptions in their state- logically similar to NSF have been induced in rats subjected
ments (Gonwa et al. 2004). to high doses of certain gadolinium-containing contrast
agents (Sieber et al. 2008). As these animals had intact renal
Pearl: Advanced kidney disease and exposure to gadolinium-
function, doses of gadolinium were administered repeatedly
based contrast appear necessary for the development of NSF,
to model the expected tissue levels of gadolinium-containing
but other risk factors may also be required.
contrast agents in a human patient with renal dysfunction. In
Comment: NSF develops predominantly in patients with contrast to human disease, the animal model tended to mani-
advanced kidney disease (eGFR <30 mL/min/1.73 m2) who fest lesions on the dorsal trunk rather than on the extremities.
are exposed to a gadolinium-based contrast agent. The more In addition, the animal lesions tended to ulcerate or to be
severe the kidney dysfunction and the higher dose of gado- excoriated to the point of ulceration. Thus, although this rat
linium, the greater the risk of NSF. However, many patients model is not a perfect model of human NSF, this study lends
who fulfill these risk criteria do not develop NSF, suggesting further credibility to the hypothesis that NSF is caused
that other variables also impart risk (Perazella 2008). Other by gadolinium-containing contrast agents’ administration
potential contributing factors, still unproven, are shown in (Cowper 2008).
Table 11.3.
Pearl: Yellow scleral plaques are a useful diagnostic feature
Pearl: The best approach to prevent the development of NSF for NSF in patients younger than 45 years of age.
is to avoid exposure to gadolinium-based contrast agents in
Comment: Yellow scleral plaques have been described in
patients who are at high risk.
numerous case reports of NSF. In the experience of the Yale
Comment: The strong epidemiological links between NSF NSF Registry, scleral plaques are present in as many as 75%
and gadolinium exposure and the finding of significant of cases. They commonly have surrounding vascular ectasia
amounts of Godolinium in NSF tissues (35–150 times higher in their early stage (Fig. 11.4).
than in non-NSF tissues) provide compelling evidence for The yellow scleral plaques of NSF may be clinically
gadolinium-based contrast agents as the causative agent in indistinguishable from pinguecula in older patients. The
NSF (High et al. 2007). Further support is added by a rat younger the patient, the more specific scleral plaques are as
model in which lesions that resemble NSF macroscopically an indicator of NSF (Cowper et al. 2008).
and microscopically occur after gadolinium-based contrast
Pearl: NSF almost always presents on the extremities first.
agent exposure (Sieber et al. 2008). Because there remains
The lower extremities are involved before the upper
no effective therapy for NSF, prevention is the best approach
extremities.
to this condition.
In high-risk patients, alternate imaging modalities must be Comment: This distribution pattern nearly always holds true.
considered if there is a reasonable hope that other approaches In some cases, upper and lower extremity involvement devel-
might address the clinical question. Certain MRI techniques ops simultaneously. The finding of both upper and lower
that do not require gadolinium often provide sufficient infor- extremity involvement is more specific for NSF than is lower
mation. As examples, three-dimensional time-of-flight MR extremity involvement alone. Isolated truncal involvement is
102 M. A. Perazella et al.

Fig. 11.5 A medium power image of the reticular dermal changes of

nephrogenic systemic fibrosis. There is markedly increased cellularity,
with increased fibroblast-like cells, occasional multinucleated giant
cells (GC), and scant to absent mononuclear cells (MC). Vascularity (V)
is increased, as is dermal collagen and mucin. This combination of fea-
tures is highly suggestive of a dermal injury reaction (Figure courtesy
of Dr. Cowper)

general, the majority of cases of NSF follow a progressive

course. Extracorporeal photopheresis has been associated with
a slight improvement in skin thickening and joint mobility
Fig. 11.4 Yellow scleral plaques: a distinctive feature of nephrogenic (Knopp and Cowper 2008; Linfert et al. 2008). Glucocorticoids,
systemic fibrosis. In this relatively early phase of scleral plaque devel- immunosuppressive agents, intravenous immune globulin,
opment, ectatic vessels are prominent (From Streams et al. 2003. plasmapheresis, and thalidomide have all failed to improve the
Reprinted with kind permission from Elsevier)
skin or joint manifestations of NSF. Anecdotal reports suggest
uncharacteristic, and should prompt a search for an alterna- that pentoxifylline, sodium thiosulfate, and imatinib mesylate
tive explanation. In rare cases, NSF presents on the upper might improve skin fibrosis (Linfert et al. 2008; Case Records
extremity, along the course of a peripherally inserted central of MGH 2008). The lack of efficacy for many agents used to
catheter line (Cowper 2006a). treat NSF may relate to diagnosis of the process at a late stage,
when the degree of fibrosis is advanced, and to repeated expo-
Pearl: The histopathological features of NSF strongly sug- sures of the patient to gadolinium-based contrast agents before
gest a dermal tissue injury state. recognition of the relationship between these agents and NSF.
Comment: NSF is characterized by a panoply of histopatho- Early initiation of therapy and prevention of further gadolin-
logic components: fibrocyte-like cells, histiocytes, dermal ium exposure might enhance treatment responses, but this
dendritic cells, scar-like fibrosis, mucin, edema, calcifica- hypothesis remains unproven.
tion, and ossification (Fig. 11.5). All of these features are
Myth: Once NSF develops, reversal of the underlying fibros-
also encountered commonly in the setting of tissue repair.
ing process is not possible.
However, unlike tissue injury due to an infection or primary
inflammatory condition, NSF lesions are devoid of signifi- Reality: In general, NSF is considered a fibrosing disorder
cant numbers of lymphocytes, granulocytes, and plasma that progresses relentlessly and is associated with permanent
cells. This Pearl facilitates the histological diagnosis of NSF dermal and systemic end organ manifestations. However, the
and is also consistent with the concept that NSF is a dermal one circumstance in which NSF may stabilize or even
reaction to injury (Cowper et al. 2003). improve (with softening of the skin and improved joint
mobility) is recovery of the patient from acute kidney injury
Myth: Several therapeutic agents are available to success-
or the receipt of a renal allograft. Several patients with NSF
fully stabilize and/or reverse the course of NSF.
associated with acute kidney injury were noted to develop
Reality: Currently, no single therapeutic agent has been defini- spontaneous improvement in skin findings and joint mobility
tively proven to stabilize and reverse the course of NSF. In (Cowper et al. 2001; Mackay-Wiggan et al. 2003). Similarly,
11 Nephrogenic Systemic Fibrosis and Other Scleroderma Mimickers 103

both published and anecdotal cases of improvements in NSF

have been reported with kidney transplantation (Grobner
2006; Jan et al. 2003; Perazella 2007, personal observation).
However, this is not a universal finding, perhaps depending
on the chronicity of NSF at the time of renal transplantation
and the level of kidney function achieved.
Pearl: Patients on peritoneal dialysis are at a substantially
higher risk for the development of NSF compared with those
who undergo hemodialysis.
Comment: Assuming that gadolinium dechelation is an
important trigger for NSF, it is no surprise that peritoneal
dialysis patients are at higher risk for this complication.
Peritoneal dialysis is significantly less effective in removing
gadolinium-containing contrast agents than is hemodialysis.
A study from the Centers for Disease Control and Prevention Fig. 11.6 Linear scleroderma. Linear scleroderma in an adult patient
provided the statistical evidence to substantiate this clinical with left leg involvement. Notice the demarcation between normal and
impression (CDC 2007). abnormal on the thigh (Figure courtesy of Dr. John Stone)

11.2 Localized Scleroderma

Pearl: Localized scleroderma is a distinct entity and should

not be confused with limited cutaneous systemic sclerosis
(limited SSc).
Comment: Localized scleroderma should not be confused
with limited cutaneous systemic sclerosis. It is categorized
as morphea (patches of thick skin) or linear scleroderma
(an area of thickened skin demarcated clearly by a line)
(Fig. 11.6). The latter includes scleroderma en coup de
sabre, which affects the face or head (Fig. 11.7). Although
restrictive lung disease has been reported in some pediatric
patients with localized disease (Christen-Zaech et al. 2008),
localized scleroderma is not usually associated with internal Fig. 11.7 En coup de sabre (Figure courtesy of Dr. John Stone)
organ involvement.
Localized scleroderma is accompanied by a positive ANA
11.3 Morphea
in 30–50% of cases, but specific SSc-related autoantibodies
such as anticentromere, antitopoisomerase I (Scl 70), and
antiRNA polymerase III antibodies are absent (Takehara and Myth: Morphea is caused by Lyme disease.
Sato 2005). The absence of Raynaud phenomenon and scle- Reality: Uncontrolled studies have linked morphea with posi-
rodactyly, along with the presence of typical lesions of mor- tive Lyme serologies or Borrelia burgdorferi DNA within skin
phea or of linear scleroderma, are additional clues that this is lesions. However, the preponderance of data suggests that mor-
not systemic disease. phea is not associated with Lyme disease (Weide et al. 2000).
Pearl: Active morphea and linear SSc are warm to touch. Pearl: Morphea often burns out in 2–5 years.
Comment: The examiner’s hand can detect the activity of the Comment: The Olmsted County study of morphea showed it
lesion by feeling heat. Heat-sensitive Doppler equipment can was increasing in prevalence and that, for half the patients, there
be used to detect the activity of a lesion. As the lesion was resolution of the activity of the lesion by 3–4 years. Deeper
becomes inactive, the heat resolves. Skin affected by local- lesions may take longer to resolve and are often accompanied
ized scleroderma is often pruritic, just as the cutaneous fea- by more disability (such as joint contractures) and damage
tures of systemic sclerosis. depending on their location (Peterson et al. 1997).
104 M. A. Perazella et al.

Myth: Morphea is not an autoimmune disease. Fig. 11.8 A deep

biopsy of the skin in
Reality: Many patients with morphea have autoantibodies nephrogenic systemic
(ANA, RF), they have more relatives with autoimmune dis- fibrosis, revealing
eases, and there seems to be more autoimmune disease within involvement of both the
dermis (D) and subcutis
morphea patients than would be expected such as diabetes, (S). There is marked
SLE, ITP, primary biliary cirrhosis, and vitiligo (Hawk and widening of subcutane-
English 2001; Tuffanelli 1998). ous septa (asterisk).
The epidermis (E) is
Pearl: Morphea often has satellite lesions around the earlier unaffected (Figure
lesions. courtesy of Dr. Cowper)

Comment: Many patients will have lesions active for a few

years, but near to some of the original lesions, there may be
more lesions that appear or even areas of pinkish discolor-
ation and pigmentary changes slightly distant from the mor-
phea plaque with seemingly normal skin in between.
Pearl: Antimalarials benefit many patients with morphea.
Comment: There are no systematic data about efficacy of
antimalarials in morphea (Fig. 11.8), and it can be difficult to
demonstrate meaningful improvement in this disease.
However, anecdotal evidence suggests that when antimalari-
als are stopped in patients with morphea, they often develop
new skin lesions.
Myth: Morphea is purely a skin disease.
Reality: About one third of patients with morphea have eosino-
philic fasciitis, which has a clear systemic component. In addi-
tion, patients with morphea are more likely to have a second
autoimmune disease (Peterson et al. 1995; Zulian et al. 2005).

11.4 Scleromyxedema

Pearl: Scleromyxedema and NSF are often indistinguishable

histologically but can be differentiated by their clinical
features.
Comment: Two major features highlight the differences
between NSF and scleromyxedema:
• Pattern of skin involvement – scleromyxedema tends to
involve the head and neck rather than the extremities. In
contrast, NSF involves the extremities first but virtually
never involves the head and neck region.
Fig. 11.9 Morphea (Figure courtesy of Dr. John Stone)
• Paraproteinemia – the majority of patients with sclero-
myxedema have circulating paraproteins that are detectable
by serum protein electrophoresis or immunoelectrophore- deep biopsy with a generous sampling of fat is the best diag-
sis. Patients with NSF almost never have such paraproteins nostic test for NSF (Cowper 2008). In this way, the approach
unless their renal disease developed in the setting of multi- to diagnosing NSF via skin biopsy is similar to those for
ple myeloma. diagnosing medium-vessel vasculitis, eosinophilic fasciitis,
and other disorders with a predilection for the subcutaneous
Subtle histopathological differences also exist in some cases.
tissues.
Scleromyxedema typically involves the dermis only, but NSF
has a predilection for both the subcutis and dermis and some- Pearl: Patients with scleromyxedema should be evaluated for
times the subcutis exclusively (Fig. 11.9). For this reason, a the presence of a paraprotein.
11 Nephrogenic Systemic Fibrosis and Other Scleroderma Mimickers 105

Fig. 11.11 Eosinophilic fasciitis. Note the enlargement of the arm dis-
tal to the elbow (Figure courtesy of Dr. John Stone)

skin of scleroderma and NSF, the overlying skin in eosino-

philic fasciitis can be pinched. The easiest way for the exam-
iner to do this is to place his or her thumbs on the patient’s
skin and then slide the thumbs together, pinching some skin
gently between the opposing ends.
Eosinophilic fasciitis is associated with venous grooving
(Fig. 11.10). This is observed most effectively on the skin of
the medial upper arm as the patient externally rotates and
abducts the arm.
Pearl: Eosinophilic fasciitis usually burns out.
Comment: Eosinophilic fasciitis often becomes inactive
after several years. Early, aggressive treatment with gluco-
corticoids is important to minimize damage from this disor-
der. Limbs affected by eosinophilic fasciitis can have the
Fig. 11.10 Eosinophilic fasciitis. (a) Venous grooving in a patient with
appearance of pseudohypertrophied muscles (Fig. 11.11).
eosinophilic fasciitis. (b) Peaud' Orange skin on the back of the patient's
calf (Figure courtesy of Dr. John Stone)

Comment: There is no Raynaud phenomenon, digital involve- References

ment, and dilated nailbed capillaries in scleromyxedema.
However, the skin can look like SSc. If scleromyxedema is
Broome DR, Girguis MS, Baron PW, et al Gadodiamide-associated
present, it can be associated with a monoclonal gammopathy nephrogenic systemic fibrosis: why radiologists should be con-
of uncertain significance (MGUS) or myelodysplastic prob- cerned. Am J Roentgenol. 2007;188:586–92
lems. Thus serum and urine protein electrophoresis evalua- Case Records of MGH. Case 6-2008: A 46-year-old woman with renal
tions should be performed. failure and stiffness of the joints and skin. N Engl J Med.
2008;358:827–38
Centers for Disease Control (CDC). Nephrogenic fibrosing dermopathy
associated with exposure to gadolinium-containing contrast agents–
11.5 Eosinophilic Fasciitis St. Louis, Missouri, 2002–2006. Morb Mortal Wkly Rep. 2007;
56(7):137–41
Christen-Zaech S, Hakim MD, Afsar FS, Paller AS. Pediatric morphea
Pearl: Eosinophilic fasciitis has “pinchable” skin. (localized scleroderma): Review of 136 patients. J Am Acad
Dermatol. 2008;59(3):385–96
Comment: The pathologic disturbance lies below the skin in Collidge TA, Thompson PC, Mark PB, et al Gadolinium-enhanced MR
eosinophilic fasciitis. Therefore, in contrast to the hidebound imaging and nephrogenic systemic fibrosis: retrospective study of a
106 M. A. Perazella et al.

renal replacement therapy cohort. Radiology 2007;245:168–75 Linfert DR, Schell JO, Fine DM. Treatment of nephrogenic systemic fibro-
Cowper SE. Gadolinium: is it to blame? J Cutan Pathol. 2008; sis: limited options but hope for the future. Sem Dial. 2008;51:155–9
35(5):520–2 Mackay-Wiggan JM, Cohen DJ, Hardy MA, et al Nephrogenic fibros-
Cowper SE, Su L, Robin H, et al Nephrogenic fibrosing dermopathy. ing dermopathy (scleromyxedema-like illness of renal disease). J
Am J Dermatopathol. 2001;23(5):383–93 Am Acad Dermatol. 2003;48:55–60
Cowper SE, Bucala R. Nephrogenic fibrosing dermopathy: suspect iden- Marckmann P, Skov L, Rossen K, et al Case-control study of gadodi-
tified, motive unclear [letter]. Am J Dermatopathol. 2003;25(4):35 amide-related nephrogenic systemic fibrosis. Nephrol Dial
Cowper SE, Boyer PJ. Nephrogenic systemic fibrosis: an update. Curr Transplant. 2007;22:3174–8
Rheumatol Rep. 2006a;8(2):151–7 Medicines and Healthcare products Regulatory Agency (MHRA).
Cowper SE, Bucala R, Leboit PE. Nephrogenic fibrosing dermopathy/ Nephrogenic Systemic Fibrosis (NSF) and gadolinium-containing
nephrogenic systemic fibrosis: Setting the record straight. Semin MRI contrast agents: Public Assessment Report. 2007. http://www.
Arthritis Rheum. 2006b;35(4):208–10 mhra.gov.uk/home/idcplg?IdcService=GET_FILE&dDocName=C
Cowper SE, Rabach M, Girardi M. Clinical and histological findings in ON2030232&RevisionSelectionMethod=LatestReleased Accessed
nephrogenic systemic fibrosis. Eur J Radiol. 2008;66(2):191–9 15 Jul 2008
Cowper SE. Nephrogenic fibrosing dermopathy [NFD/NSF Website]. Othersen JB, Maize JC, Woolson RF, Budisavljevic MN. Nephrogenic
2001–2008. 2007. http://www.icnfdr.org. Accessed 16 Apr 2008 systemic fibrosis after exposure to gadolinium in patients with renal
Dawson P, Punwani S. Nephrogenic systemic fibrosis: non-gadolinium failure. Nephrol Dial Transplant. 2007;22:3179–85
options for the imaging of CKD/ESRD patients. Sem Dial. Perazella MA. Tissue deposition of gadolinium and development of
2008;51:160–5 NSF: a convergence of factors. Semin Dial. 2008;21:150–4
Deo A, Fogel M, Cowper SE. Nephrogenic systemic fibrosis: a popula- Perazella MA, Rodby RA. Gadolinium use in patients with kidney dis-
tion study examining the relationship of disease development to ease: a cause for concern. Semin Dial. 2007;20:179–85
gadolinium exposure. Clin J Am Soc Nephrol. 2007;2:264–7 Peterson LS, Nelson AM, Su WP. Classification of morphea (localized
FDA. Information for Healthcare Professionals Gadolinium-Based scleroderma). Mayo Clin Proc. 1995;70:1068–76
Contrast Agents for Magnetic Resonance Imaging (marketed as Peterson LS, Nelson AM, Su WP, Mason T, O’Fallon WM, Gabriel SE.
Magnevist, MultiHance, Omniscan, OptiMARK, ProHance). 2007. The epidemiology of morphea (localized scleroderma) in Olmsted
http://www.fda.gov/cder/drug/InfoSheets/HCP/gadolinium-con- County 1960–1993. J Rheumatol. 1997;24(1):73–80
taining contrast agents_200705.htm Sieber MA, Pietsch H, Walter J, et al A preclinical study to investigate
Gonwa TA, Jennings L, Mai ML, et al Estimation of glomerular filtra- the development of nephrogenic systemic fibrosis: a possible role for
tion rates before and after orthotopic liver transplantation: evalua- gadolinium-based contrast media. Invest Radiol. 2008;43(1):65–75
tion of current equations. Liver Transpl. 2004;10(2):301–9 Streams BN, Liu V, Liegeois N, Moschella SM. Clinical and pathologic
Girardi M. Nephrogenic systemic fibrosis: a dermatologist’s perspec- features of nephrogenic fibrosing dermopathy: a report of two cases.
tive. J Am Coll Radiol. 2008;5(1):40–4 J Am Acad Dermatol. 2003;48:42–7
Grobner T. GBC: a specific trigger for the development of nephrogenic Swaminathan S, High WA, Ranville J, et al Cardiac and vascular metal
fibrosing dermopathy and nephrogenic systemic fibrosis? Nephrol deposition with high mortality in nephrogenic systemic fibrosis.
Dial Transplant. 2006;21:1104–8 Kidney Int 2008; 73(12):1413–18
Grobner T, Prischl FC. Patient characteristics and risk factors for neph- Takehara K, Sato S. Localized scleroderma is an autoimmune disorder.
rogenic systemic fibrosis following gadolinium exposure. Sem Dial. Rheumatology (Oxford). 2005;44(3):274–9
2008;21:135–9 Ting WW, Stone MS, Madison KC, Kurtz K. Nephrogenic fibrosing der-
Hawk A, English JC III. Localized and systemic scleroderma. Semin mopathy with systemic involvement. Arch Dermatol. 2003;139: 903–6
Cutan Med Surg. 2001;20:27–37 Tuffanelli DL. Localized scleroderma. Semin Cutan Med Surg. 1998;
High WA, Eng M, Ayers RA, Cowper SE. Gadolinium is quantifiable 17:27–33
within the tissue of patients with nephrogenic systemic fibrosis. Weide B, Schittek B, Klyscz T, et al Morphoea is neither associated
J Am Acad Dermatol. 2007;56:1–2 with features of Borrelia burgdorferi infection, nor is this agent
Jan F, Segal JM, Dyer J, et al Nephrogenic fibrosing dermopathy: two detectable in lesional skin by polymerase chain reaction. Br J
pediatric cases. J Pediatr. 2003;143:678–81 Dermatol. 2000;143(4):780–5
Knopp EA, Cowper SE. Nephrogenic systemic fibrosis: early recogni- Zulian F, Vallongo C, Woo P, et al Localized scleroderma in childhood
tion and treatment. Sem Dial. 2008;21:123–8 is not just a skin disease. Arthritis Rheum. 2005;52: 2873–81
 Sjögren’s Syndrome
12
Manuel Ramos-Casals, Troy E. Daniels, Robert I. Fox,
John P. Whitcher, George E. Fragoulis, Fotini N. Skopouli,
and Haralampos M. Moutsopoulos

(Vitali et al. 2002) were created to update the 1993/1996

12.1 Overview of Sjögren’s Syndrome European Community Criteria for Classifying SjS (Vitali
et al. 1993, Vitali et al. 1996). These criteria sought to correct
› Primary Sjögren’s syndrome (SjS) is a systemic auto- problems with earlier criteria sets by requiring that evidence
immune disease that is associated with early and grad- of an autoimmune process characteristic of SjS be included.
ually progressive lacrimal and salivary dysfunction. The 2002 Criteria stipulate that at least one criterion must be a
› Secondary SjS occurs in association with other auto- positive anti-Ro/SSA or -La/SSB antibody assay or a positive
immune disorders, the most common of which is rheu- labial salivary gland biopsy (Table 12.1) (Vitali et al. 2002).
matoid arthritis. The 2002 criteria have been criticized for excessive emphasis
› About 90% of patients with SjS are women. on glandular disease and failure to capture the full spectrum of
› Minor salivary glands and lacrimal glands in SjS exhibit other organ system involvement in SjS, particularly extraglan-
a particular pattern of periductal focal lymphocytic dular disease features and the full spectrum of immunological
infiltration known as focal lymphocytic sialadenitis. abnormalities. Known prognostic factors in SjS, particularly the
› Primary SjS has a community prevalence that ranges occurrence of hypocomplementemia, cryoglobulinemia, and
from 0.1 to 0.6%. vasculitis, are not captured in the 2002 criteria.
› The major eye problem in SjS is keratoconjunctivitis
Pearl: Differences between primary SjS and secondary SjS
sicca, which leads to xerophthalmia. The principal oral
have major clinical relevance.
manifestation of SjS is decreased salivary gland pro-
duction, leading to xerostomia and a predilection for Comment: Primary SjS and secondary SjS are related sys-
dental caries. temic autoimmune diseases, but important differences exist
› Extraglandular manifestations of SjS include arthral- between these conditions (Pavlidis et al. 1982). Primary SjS
gias, thyroiditis, renal involvement (leading to renal is characterized by early and progressive salivary and lacri-
tubular acidosis (RTA) ), peripheral neuropathy, cuta- mal dysfunction. In addition, primary SjS also encompasses
neous vasculitis, and lymphoma. a host of extraglandular manifestations that can involve the
› The risk of lymphoma in SjS is approximately 5%. thyroid gland, kidneys, liver, skin, peripheral nerves, lungs,
› Most patients with SjS develop increased circulating and other organs. Patients with primary SjS rarely develop
polyclonal immunoglobulins and autoantibodies. These rheumatoid arthritis (RA), and secondary SjS often develops
autoantibodies include two fairly specific antibodies in the setting of RA. Approximately 30% of patients with RA
directed against the Ro (SS-A) and La (SS-B) antigens. develop secondary SjS, usually years after their RA diagno-
sis (Andonopoulos et al. 1987).
The three most common primary SjS symptoms are dry
eyes, dry mouth, and musculoskeletal pain. Diagnosing pri-
mary SjS is more difficult than diagnosing secondary SjS
12.2 Classification and Epidemiology because those with primary disease typically present with only
one of these complaints to different specialists. All three com-
plaints have their own differential diagnosis. Specialists who
Myth: It doesn’t matter which criteria are used for classify-
are addressing one complaint are often unfamiliar with the
ing or diagnosing patients with Sjögren’s syndrome (SjS).
breadth of possibilities related to the others. A survey of more
Reality: At least ten diagnostic/classification criteria for SjS than 3,000 SjS patients in 2005 reported that the average time
have been published since the 1960s. The 2002 American– between occurrence of their first symptoms and diagnosis was
European Consensus Group classification criteria for SjS more than 6 years (Sjögren’s Syndrome Foundation 2006).

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 107

DOI:10.1007/978-1-84800-934-9_12, © Springer Science + Business Media B.V. 2009
108 M. Ramos-Casals et al.

Table 12.1 American–European consensus group classification criteria for sjögren’s syndrome
I. Ocular symptoms: a positive response to at least one of the following questions:
1. Have you had daily, persistent, troublesome dry eyes for more than 3 months?
2. Do you have a recurrent sensation of sand or gravel in the eyes?
3. Do you use tear substitutes more than 3 times a day?
II. Oral symptoms: a positive response to at least one of the following questions:
1. Have you had a daily feeling of dry mouth for more than 3 months?
2. Have you had recurrently or persistently swollen salivary glands as an adult?
3. Do you frequently drink liquids to aid in swallowing dry food?
III. Ocular signs: a positive result for at least one of the following two tests:
1. Schirmer I test, performed without anesthesia (£5 mm in 5 minutes)
2. Rose Bengala score or other ocular dye score (³4 on the van Bijstervled scale)
IV. Histopathology: in minor salivary glands (obtained through normal-appearing mucosa) focal lymphocytic sialadenitis, evaluated by an
expert histopathologist, with a focus score ³1, defined as a number of lymphocytic foci (which are adjacent to normal-appearing mucous
acini and contain more than 50 lymphocytes) per 4 mm2 of glandular tissue.
V. Salivary gland involvement: a positive result for at least one of the following tests:
1. Unstimulated whole salivary flow (£1.5 mL in 15 minutes)
2. Parotid sialography showing the presence of diffuse sialectasis (punctuate, cavitary, or destructive pattern), without evidence of major
duct obstruction
3. Saolivary scintigraphy showing delayed uptake, reduced concentration, and/or delayed excretion of tracer
VI. Autoantibodies: presence in the serum of the following:
1. Antibodies to Ro(SS-A) or La(SS-B) antigens, or both
Rules for classification
For primary SS: in patients without any potentially associated disease
a. Presence of any 4 four of the 6 six items indicates primary SjS as long as either item IV (histopathology) or VI (serology) is positive
b. Presence of any 3 three of the 4 four objective criteria items (that is items III, IV, V, VI)
c. The c lassification tree procedure (best used in clinical-–epidemiological surveys)
For Secondary SjS: patients with a potentially associated disease (e.g., another well defined-connective tissue disease), the presence of item I
or item II plus any 2 two from among items III, IV, and V.
Exclusion criteria: past head and neck radiation treatment; hepatitis C infection; acquired immunodeficiency disease (AIDS); pre-existing
lymphoma; sarcoidosis; graft vs. host disease; use of anticholinergic drugs (since a time shorter than 4-fourfold the half life of the drug)
a
Lissamine green has now replaced rose bengal for this test.
Adapted from (Vitali et al. 2002)

Myth: A patient is diagnosed with definite primary SjS if she the prevalence of SjS depends on the classification criteria
or he fulfills at least four of the six American–European applied. Studies based on the American–European criteria
Classification Criteria. have reported a prevalence of primary SjS of 0.3% in the
female population in Greece and up to 0.4% of females in the
Reality: The classification criteria have been derived to clas-
United Kingdom (Trontzas and Andrianakos 2005; Bowman
sify patients as primary SjS for the purpose of research stud-
et al. 2004).
ies, not clinical diagnosis (Vitali et al. 2002). For example, a
patient who has recurrent bilateral parotid gland enlarge- Pearl: The expression of primary SjS in males is relatively
ment, autoantibodies to Ro/SSA and La/SSB, and a positive muted when compared with its expression in females.
minor salivary gland biopsy has unequivocal primary SjS.
Comment: Various studies have described a less pronounced
However, some patients with objective measures of lacrimal
clinical phenotype among male SjS patients compared with
or salivary gland dysfunction do not have clinical symptoms.
female patients, regardless of whether one considers clinical,
In the absence of ocular or oral symptoms, antibodies to the
histological, sialographic, or immunologic criteria. In a
Ro/SSA or La/SSB antigens, and a positive minor salivary
series of 1,010 patients, the 73 male patients had a lower
gland biopsy, such patients do not fulfill classification crite-
frequency of abnormal ocular tests and a lower prevalence of
ria for SjS. Most of these patients take medications that
positive antinuclear antibodies, Raynaud’s phenomenon, and
explain their abnormal gland function.
thyroiditis (Ramos-Casals et al. 2008). These findings are
Myth: SjS is an uncommon disease. consistent with the precept that most autoimmune diseases
are more common in women. The relatively muted presenta-
Reality: Epidemiological studies from Scandinavia and
tion in men can make the diagnosis more difficult.
Greece have shown that primary SjS is less prevalent than
RA (presuming a 1% prevalence for RA), but not by much Pearl: When SjS occurs in males, look for other clinical stig-
(Jacobsson et al. 1989; Dafni et al. 1997). To a large extent, mata of Klinefelter syndrome.
12 Sjögren’s Syndrome 109

Comment: In one study, up to 15% of the male SjS patients 12.4 Sicca Features
had symptoms of Klinefelter’s syndrome (lack of reproduc-
tive capacity, low testosterone, and abnormal XXY karyo-
type) (Aoki, 1999). These findings are interesting in view of 12.4.1 Ocular Manifestations
a finding in the BXSB mouse that involves the translocation
of a portion of the X chromosome to the Y chromosome, Pearl: The predominant presenting complaints in patients
because this is the only male mouse model to develop SjS- or with primary SjS are dry eyes and mouth.
SLE-like features (Krieg and Vollmer 2007).
Comment: More than ninety percent of patients with primary
SjS complain of dry eyes and/or dry mouth (Skopouli et al.
2000). Patients with this disorder also sometimes present with
12.3 Pathogenesis recurrent or persistent parotid gland enlargement, Raynaud’s
phenomenon, bronchitis sicca, hypergammaglobulinemic
Myth: The predominant cells comprising the focal lympho- purpura, peripheral neuropathy, or interstitial nephritis.
cytic infiltrates of labial minor salivary glands in primary
Myth: Concurrent symptoms of dry eyes and dry mouth (i.e.,
SjS are T-lymphocytes.
the “sicca complex”) are all that one needs to diagnose SjS.
Reality: The type of lymphocyte that predominates within
Reality: Dry mouth and dry eyes are both highly prevalent in
minor salivary gland biopsies in primary SjS depends on the
SjS, but they do not occur in all patients. However, they occur
degree of glandular inflammation. In heavy lymphocytic infil-
in many other types of patients, as well, and can often be
trates, B lymphocytes predominate and germinal centers are
induced by medications (Tables 12.2 and 12.3). In a large
formed (Gerli et al. 1997). In mild to moderate degrees of lym-
cohort of participants in the NIH-sponsored International
phocytic infiltration, T cells predominate.
Sjögren’s Syndrome registry, 91% of the patients complained
Pearl: SjS can be considered both a T cell and B cell disease. of a dry mouth and 85% complained of dry eyes. However,
the correlation between these symptoms and objective signs
Comment: The earliest lymphocytic infiltrates in salivary
glands are composed of CD20 + B cells and T cells, mostly
of the primed memory T helper phenotype. Later, other B Table 12.2 Dry mouth symptoms: differential diagnosis
cell phenotypes join the infiltrates (CD27 + memory and Chronically administered systemic drugs with anticholinergic
CD79a+). Clusters of plasma cells (CD38+) are present in effects (e.g., antidepressants, parasympatholytics, neuroleptics)
both normal salivary glands and at the periphery of T cell and Sjögren’s syndromea
Sarcoidosis, atuberculosis
B cell infiltrates in SjS (Larsson et al. 2005). These infiltrates
HIVa or hepatitis-C infection
may exhibit lymphoid follicle formation in various stages of Depression
development as they enlarge. Their cellular portfolio com- Uncontrolled diabetes; amyloidosis
prises primarily CD20 + B cells and follicular dendritic cells, Therapeutic radiation to head and neck that includes major salivary
with a few helper T cells (Prochorec-Sobieszek et al. 2004). glands in fields
The T cells have initiating roles in the pathogenesis of SjS Graft vs. host disease
a
and secrete many of the cytokines found within affected May also cause bilateral major salivary gland enlargement
organs. The T-helper (Th) infiltrates in SjS include both pro- (From Klippel et al. 2008. With kind permission from Springer Science +
Business Media.)
inflammatory Th1 cytokines such as interferon (IFN)-g and
interleukin (IL)-2, as well as antiinflammatory Th2 cytokines,
such as IL-4, IL-5, and IL-13. Over time, these profiles shift; Table 12.3 Dry eye symptoms: differential diagnosis
Th2 cytokines predominate early in SjS, but the balance shifts
Sjögren’s syndrome (keratoconjunctivitis sicca)
toward Th1 in more advanced disease (Mitsias et al. 2002). Chronically administered systemic drugs with anticholinergic
The B cells not only lead to the circulating autoantibodies effects
seen prominently in SjS, they also become the proliferating AIDS-associated keratoconjunctivitis sicca
component. A B cell activating factor known as BAFF or BLyS Conjunctival cicatrization
(B cell activating factor/B lymphocyte stimulator), which is Stevens–Johnson syndrome
Ocular cicatricial pemphigoid
regulated by IFN-g, promotes the survival and maturation of B
Drug-induced pseudopemphigoid
cells. BAFF/BLyS is elevated in SjS serum. This factor is impli- Trachcoma
cated in the polyclonal activation of B cells, correlates with the Graft vs. host disease
levels of circulating autoantibodies (Mariette et al. 2003), and Trigeminal or facial nerve paralysis
may have a long-term role in the development of lymphoma. Vitamin A deficiency (“xerophthalmia”)
SjS-related lymphomas are almost always of B cell origin. Adapted from (Whitcher et al. 1998)
110 M. Ramos-Casals et al.

of SjS was poor. In fact, neither symptom was associated

with positivity for antibodies directed against the Ro/SSA or
La/SSB antigents or focal lymphocytic sialadenitis in labial
salivary gland biopsies (Daniels et al. 2007). Moreover, the
symptom of dry eyes was only marginally associated with
the presence of keratoconjunctivitis sicca (Daniels et al.
2007).
In short, symptoms of dry eyes and dry mouth are often
caused by conditions other than SjS and are insufficient for
the diagnosis of SjS.
Pearl: “Dry eyes” associated with SjS should be termed
“keratoconjunctivitis sicca” rather than “xerophthalmia.”
Comment: Xerophthalmia refers specifically to the dry eyes
associated with hypovitaminosis A and should not be used in
reference to SjS.
Pearl: Symptoms of keratoconjunctivitis sicca are more prev- Fig. 12.1 Slit lamp examination of the cornea and conjunctiva after
rose Bengal staining
alent than xerostomia in patients who have SjS secondary to
RA.
foreign body” sensation. However, 61% of patients without
Comment: On close questioning, complaints consistent with keratoconjunctivitis sicca also offered that complaint, making
keratoconjunctivitis sicca are elicited in nearly 40% of RA its usefulness as a discriminator of true keratoconjunctivitis
patients, but dry mouth is reported in only 6% (Andonopoulos sicca rather limited.
et al. 1989). RA patients who have secondary SjS rarely Among the other seven ocular symptoms, “inability to tear,”
mention sicca symptoms spontaneously and must be asked had the strongest association with keratoconjunctivitis sicca,
about them. but was experienced by a smaller proportion of patients with
Patients with RA and secondary SjS rarely experience keratoconjunctivitis sicca (45%), compared with only 9% of
episodes of parotid gland swelling. In contrast, either inter- those who did not have keratoconjunctivitis sicca (p < 0.0001)
mittent or permanent parotid swelling is a hallmark of pri- (Whitcher et al. 1998).
mary SjS, occurring in 34 of 65 patients (56%) in one study
Pearl: Patients with keratoconjunctivitis sicca who need to
(Bloch et al. 1965).
use artificial tears chronically must use a product that is
Myth: The diagnosis of keratoconjunctivitis sicca is based on preservative-free.
an abnormal Schirmer’s test.
Comment: Many commonly used artificial tears contain pre-
Reality: Both the sensitivity and specificity of the Schirmer’s servatives which, when used chronically, can elicit an inflam-
test are low for keratoconjunctivitis sicca (Paschides et al. matory reaction in the cornea and conjunctiva that is painful.
1989). The Schirmer’s test measures lacrimal gland function, This preservative-induced effect is difficult to distinguish
and many factors can interfere with the physiology of these from those of true keratoconjunctivitis sicca. Prolonged use
glands: medications, older age, occupation, the time of day, of preservative-containing artificial tears can result in con-
the season of the year, and the patient’s hydration status. junctival scarring that further complicates the clinical
The test that provides unequivocal evidence of keratocon- presentation.
junctivitis sicca is a slit lamp examination of the cornea and SjS patients are particularly susceptible to ocular surface
conjunctiva after rose Bengal or lissamine green staining toxicity from these preservatives because of their aqueous tear
(Fig. 12.1). In keratoconjunctivitis sicca, the stain collects deficiency. Preservative-free artificial tears should be used by
within defects of the cornea caused by ocular drying and the all patients who instill them more than 4 times per day.
consequent minor trauma to the eye.
Myth: All patients diagnosed with keratoconjunctivitis sicca
Myth: The symptom of ocular foreign-body sensation is the have SjS.
best symptomatic indicator of keratoconjunctivitis sicca.
Reality: The term “keratoconjunctivitis sicca” was coined by
Reality: In comparing 449 patients referred to a SjS clinic, Henrik Sjögren to describe a particular type of dry eye dis-
“ocular foreign body” sensation was the most prevalent of ease in 1933 . His description later came to define the ocular
seven ocular symptoms. More than three fourths of the patients component of SjS. A few subsequent studies noted that not
with keratoconjunctivitis sicca complained of an “ocular all patients with keratoconjunctivitis sicca exhibit evidence
12 Sjögren’s Syndrome 111

of the oral/salivary or serological components of SjS. One Anesthesiologists must limit the quantity of anticholinger-
prospective study of 34 patients with keratoconjunctivitis gic agents administered during intubation for patients with
sicca found that 44% had objective evidence of the salivary SjS. SjS patients may be unduly sensitive to these drugs and
and serological components of SjS, but that 56% had no evi- develop inspissated secretions that are not cleared easily.
dence of salivary gland dysfunction (Forstot et al. 1982). This can be particularly problematic after chest or abdominal
Some patients in the latter group were ANA positive, but surgery, when the expectoration of tenacious secretions is
none had antibodies to Ro/SSA or La/SSB. difficult even under better circumstances.
A long-term study reexamined a group of 106 patients Finally, the use of oral saliva substitutes should be encour-
who had been diagnosed with keratoconjunctivitis sicca aged. It is expected that patients will be “NPO” prior to most
(Kruize et al. 1996). Repeat examinations up to 12 years surgeries. In the absence of normal saliva, patients with SjS
after their initial diagnoses of keratoconjunctivitis sicca experience unnecessary discomfort if they are not allowed to
determined that 29% had primary SjS and 18% had second- have their artificial saliva. This is particularly true when the
ary SjS, but 53% had isolated keratoconjunctivitis sicca. patient’s surgery is delated until later in the day.
Pearl: Ocular cosmetic procedures can exacerbate SjS. Myth: Topical ophthalmological glucorticoids should not be
used in SjS.
Comment: Three specific types of cosmetic surgery are con-
traindicated in patients with SjS: blepharoplasty (eyelid Reality: One randomized trial assigned SjS patients to either
“lift”), Lasik surgery, and Botox® injections. loteprednol 0.5% solution or placebo 4 times a day for 4 weeks
Blepharoplasty may interrupt the basal tearing that occurs (Pflugfelder et al. 2004). All patients had failed previous thera-
in the lower lid by the glands of Sherring. (These are the same pies with artificial tears. Patients in the topical glucocorticoid
glands that one stimulates by rubbing one’s eyes). Stretching group demonstrated significant improvement and ability to
of the eyelid during blepharoplasty appears to disrupt the deli- return to traditional artificial tears after loteprednol. Thus, topi-
cate neural interconnections within the network of glands. cal glucocorticoids may have a role in some patients, but they
Blepharoplasty can also lead to increased zones of the cornea must be employed with caution in patients who have glaucoma
that are susceptible to exposure keratitis. Particularly when or cataracts, especially if prolonged use is intended.
sleeping, the lower lid may not make adequate contact with
Pearl: The environment plays a key role in exacerbating
the upper lid, leading to a zone of increased evaporative loss
patients’ ocular symptoms.
and resulting dessicative injury.
SjS is a contraindication to Lasik surgery because of the Comment: Although SjS patients haves decreased rates of
increased dryness that occurs after the procedure (Liang aqueous tear formation and increased rates of evaporative loss
et al. 2008). This increased dryness presumably results from due to the inflammatory process, both of those processes are
the “flap” cut by the microtome across the cornea, which sev- exacerbated by environmental factors. As examples, factors
ers the nerve bodies from afferent sensory nerves that inner- such as low humidity can be partially helped by humidifiers.
vate the cornea. The resulting “neuropathic” eye is more The effect of dry winds are ameliorated by “wrap-around”
sensitive to abrasions as well as to the sensation of dryness sunglasses or side shields on glasses.
(friction as the upper lid traverses the globe). Additional factors such as the decreased “blink rate” asso-
Finally, a standard model for induction of keratoconjunc- ciated with the use of computer monitors are underappreci-
tivitis sicca is the injection of botulinum toxin (Lekhanont ated. The modern work place environment is typically an office
et al. 2007). The wisdom of avoiding this intervention in a with low humidity, where individuals spent large amounts of
patient with SjS is self-evident. time staring at computer screens (Wolkoff et al. 2006). People
concentrating on computer monitors have a 90% decrease in
Pearl: SjS patients have unique ocular needs at the time of
their baseline blink rate. Thus, concentration on “the screen”
anesthesia and surgery.
can override the normal corneal surface conditions that lead to
Comment: Operating rooms and postoperative recovery blinking and spreading of the available tears.
rooms are notorious for their low humidity. SjS patients are
at risk for exacerbations of their keratoconjunctivitis sicca or
even for corneal abrasions in such environments. The risk is 12.4.2 Oral Manifestations
perhaps greatest in the recovery room, where the patient who
is partially awake has fluttering eyelids and often receives
Myth: The most common cause of dry mouth symptoms is
unhumidified air directly to the face. These patients are
primary or secondary SjS.
unaware of ocular symptoms as they awaken from anesthe-
sia. Ocular lubricants should be employed during surgery Reality: The symptom of dry mouth (xerostomia) is highly
and during the recuperative period to prevent complications. prevalent in the general population. It is often associated
112 M. Ramos-Casals et al.

with salivary hypofunction, but occurs in some patients with

normal salivary flow rates. The most common cause of this
symptom is the effect of one or more prescription drugs.
Hundreds of drugs have the capacity to induce xerostomia. A
list of these medications can be viewed at http://www.dry-
mouth.info/practitioner/default.asp. In addition to SjS, other
causes of dry mouth symptoms are listed in Table 12.2.
Pearl: One pattern of lymphocytic infiltration in salivary
glands is diagnostic of the salivary component of SjS. Other
patterns are not.
Comment: Diagnoses of the salivary component of SjS by the
presence of a focal lymphocytic infiltrate in a labial salivary
gland biopsy have been rendered since the 1960s (Daniels
1991). The ocular component of SjS (keratoconjunctivitis
sicca) is significantly correlated with the presence of focal
lymphocytic sialadenitis in labial salivary gland biopsies, but
not with other patterns of chronic salivary inflammation com-
monly seen in those speicmens (Daniels and Whitcher 1994).
Diagnostic confusion among pathologists who review
labial salivary gland biopsies is rife. In a prospective review
of minor salivary gland biopsies from patients who had been
referred to a SjS Clinic, 53% needed diagnostic revision. In
23% of the patients, the misclassification of the biopsy led to
a diagnostic delay that ranged from a few months to more
than 7 years (Vivino et al. 2002).
Myth: Minor labial salivary gland biopsy is a painful proce-
dure often associated with unwanted side effects.
Reality: Minor labial salivary gland biopsy is a safe proce-
dure that is associated with few adverse effects (Caporali
et al. 2008). Hematoma, infection, and long-lasting numb-
ness at the incision site occur rarely. The procedure provides
valuable diagnostic confirmation of the salivary component
of SjS and is often essential in parsing the differential diag-
nosis in patients who present with sicca symptoms (e.g., sar-
coidosis, amyloidosis). Labial salivary gland biopsy also
offers unusually easy access to an end-organ affected by
autoimmune pathology, thus making it an important tool for
research on disease pathophysiology.
Pearl: Several frequently forgotten causes of the sicca syn-
drome are type IV and V lipoproteinemias, amyloidosis, and
sarcoidosis.
Comment: The patients with these conditions can present with
dryness of the mouth and eyes. Patients with these disorders Fig. 12.2 (a–c) Contrasts between Sjögren’s syndrome and sarcoidosis
normally are negative on testing for antibodies to the Ro/SSA in the labial salivary glands. (a) Lymphocytic infiltrates within the
labial minor salivary gland of a patient with primary Sjogren’s syn-
and La/SSB autoantigens. Labial minor salivary gland biopsy is drome. (b) Lymphocytic infi ltrates in within the minor salivary gland
of paramount diagnostic importance, because it can reveal fatty biopsy of a patient with SLE-SS overlap syndrome. The infiltrates are
infiltrates in the lipoproteinemias; amorphous material within perivascular (arrows). (c) Granuloma formation within the minor sali-
the blood vessel walls that stains positively for congo red in vary gland biopsy of a patient who also had erythema nodosum, arthri-
tis, and bilateral hilar lymphadenopathy. Both the clinical picture and
amyloidosis; and noncaseating granulomas in sarcoidosis the histopathologic features depicted in (c) are compatible with sarcoi-
(Fig. 12.2) (Reinertsen et al. 1980; Simon and Moutsopoulos dosis. [Figure courtesy of Dr. Haralampos M. Moutsopoulos]
1979; Drosos et al. 1989).
12 Sjögren’s Syndrome 113

Fig. 12.4 This patient with SjS exhibits signs of multiple dental caries
and chronic erythematous candidiasis that are characteristic of chronic
Fig. 12.3 Oral features of candidiasis, a common complication of salivary dysfunction. Caries are seen next to the gingival margins in all
Sjögren’s syndrome. Angular cheilitis is present, along with a red, beefy of the teeth illustrated (including those with gold crowns) and three
tongue consistent with candidiasis [Figure courtesy of Dr. Haralampos anterior teeth also have caries on their incisal edge. The diffuse atrophy
M. Moutsopoulos] of lingual filiform papillae is a clear sign of erythematous candidiasis
[Figure courtesy of Dr. Troy Daniels]

Pearl: Angular cheilitis and a red tongue with atrophic papil-

In a lip biopsy from an SjS patient with severe dryness,
lae strongly suggest Candida overgrowth in the mouth.
attention is usually focused on the dense lymphoid infiltrates
Comment: Oral candidiasis is common in patients with pri- (Fig. 12.5a). However, residual acinar units are still visible
mary SjS (Hernandez and Daniels 1989). This frequently (arrows). Indeed, morphometric analysis has shown that only
presents as diffuse, painful erythema of the oral mucosa, par- about 50% of the gland acinar or ductal tissue is replaced or
ticularly affecting the dorsal tongue and causing angular destroyed (Fox 2002). This may seem surprising, because
cheilitis (Fig. 12.3). A common symptom is intolerance to the kidneys and liver continue to function ably until their
acidic or spicy foods. Several weeks of therapy with flucon- functional units are more than 90% destroyed.
azole are often required to cure this problem. In addition, Residual salivary gland tissue raises the possibility that
patients with severe salivary hypofunction may require pro- substantial improvement is possible in SjS, presuming suc-
longed antifungal treatment with topical preparations that do cessful therapy against the glandular inflammation. The
not contain sugar (Daniels 2000). Recurrent candidiasis is glandular tissue beyond the lymphoid infiltrate may retain its
common in some patients. neural innervation. Studies in man and murine models have
indicated the presence of receptors for acetylcholine and
Myth: Periodontal disease is a very common problem among
other critical neurotransmitters. The release of and response
patients with primary SjS.
to neurotransmitters are strongly influenced by inflamma-
Reality: Periodontal disease is actually no more common in tory cytokines, including Tumor necrosis factor (TNF) and
primary SjS patients as it is in healthy, age-matched controls IL-1. Unfortunately, broad-spectrum immunosuppressive
(Boutsi et al. 2000). What is unique about the oral health of approaches such as glucocorticoids and conventional immu-
patients with SjS, however, is the patients’ propensity to nosuppressive agents are not effective in treating the sicca
develop a characteristic pattern of dental caries (cavities) symptoms of SjS. New treatment strategies are required.
located along the gingival margins and cusp tips of multiple
Pearl: Salivary flow rates can be evaluated by minimally
teeth (Fig. 12.4) (Daniels 2000). The treatment and preven-
invasive methods.
tion of these progressive lesions requires ongoing caries pre-
vention activities and interaction with the patient’s dentist. Comment: Many individuals who do not have SjS complain of
dry mouth. Moreover, patients’ symptoms of dry mouth corre-
Myth: Oral dryness in SjS results from the total destruction
late very poorly with actual salivary flow rates. Thus, it is
of the gland.
important to correlate patients’ symptoms with objective signs
Reality: In fact, the glandular secretory elements are gradually of dryness. Technetium scans of salivary function are performed
replaced or “destroyed” by lymphocytic infiltration and prolifer- after coating the tongue with a lemon concentrate (Hakansson
ation. Even in patients with severe dryness, some normal-appear- et al. 1994; Helman et al. 1987; Kohn et al. 1992). The uptake
ing ducts and acini remain but do not function well because of of contrast material by the gland and its rate of secretion into
the inflammatory process that disrupts the ability of the residual the mouth can be quantified reliably. Although the decreased
secretory units to release or respond to neurotransmitters. flow rate is not specific to SjS, a Technetium scan can be useful
114 M. Ramos-Casals et al.

Comment: Sialograms to assess the salivary status of SjS

patients or to visualize the ductal structures for punctal
sialadenitis are unnecessary. Sialograms have been largely
replaced by gadolinium-enhanced MR studies with “fat sup-
pression” views, which provide an excellent means of evalu-
ating glandular tissues (Jungehulsing et al. 1999; Makula
et al. 2000). This advance is fortunate, because few medical
centers had sufficient experience with retrograde sialography
to perform the procedure safely and well, anyway.
In any event, the role of ductal structure visualization in
the assessment of the salivary component of SjS (by either
injection sialography or MRI sialography) remains periph-
eral to other better established means such as minor gland
biopsy or measuring unstimulated flow rate.
Pearl: A dry mouth is not necessarily a painful mouth.
Comment: The physician should look for signs of oral can-
didiasis such as angular cheilitis, atrophy or loss of filiform
papillae on the dorsal tongue, or erythematous changes on
the hard pallate, as well as lichen planus-like changes in buc-
cal recess (see Fig. 12.3).
Many patients develop dry mouth as they age (Hershkovich
et al. 2007; Nagler 2004). This is not a “normal part of the
aging process,” but rather a consequence of chronically
administered prescription drugs. However, some event usu-
ally brings the patient to clinical attention. Frequently, a dry
mouth is converted to a painful mouth by the occurrence of
oral yeast infections, particularly in a patient who is on glu-
cocorticoids or has recently been taking antibiotics (Abraham
et al. 1998; Almstahl et al. 1999; Rhodus and Michalowicz
Figs. 12.5 (a, b) In a lip biopsy from a Sjögren’s syndrome patient
2005). Alterations in the oral microbial flora, as well as rela-
with severe dryness, attention is usually focused on the dense lymphoid
infiltrates. In this biopsy of a labial minor salivary gland from a patient tive decreases in the salivary flow of naturally occurring anti-
with primary SjS, follicular colonization of neoplastic B cell lympho- fungal agents such as transferrin or calprotectin, histatins,
cytes are evident (arrowheads). Remnants of follicular dentritic cells and other small molecules of the defensin family, further pre-
are also present (arrows) [Figure courtesy of Dr. H.M. Moutsopoulos]
dispose the SjS patient to oral candidiasis.
Oral candidiasis can present as reddish petechiae in the
in the evaluation of the patient who complains that “I don’t feel
mouth, generally found on the hard palate (see Fig. 12.3)
any saliva in my mouth,” yet has a normal salivary pool.
(Daniels and Fox 1992; Daniels 2000). Denture removal may
Pearl: The “sponge method” is more effective as a contra- be required to observe the lesions.
ceptive than as a measure of salivary production. Treatment of the oral candidiasis may require a rather
prolonged treatment with topical antifungal drugs (Wu and
Comment: A variety of methods of quantifying salivary pro-
Fox 1994), using mouth rinses similar to those employed by
duction have been devised through the years. Most of these
the radiation therapists and topical application of nystatins
are as ineffective as they are cumbersome. As an example, in
(Daniels 2000).
the Saxon test, saliva is collected on a preweighed sponge
placed under the tongue (Kohler and Winter 1985). Substantial Pearl: Complaints of symptoms of “mouth burning” are
intra-patient variability is observed over the course of the common in clinical practice and often have explanations
same day. The Saxon test can be affected by factors such as besides a systemic autoimmune condition such as SjS.
the time since the last meal, teethbrushing, history of smok-
Comment: Other causes of burning mouth syndrome must also
ing, and medications (Stevens et al. 1990).
be considered, including: nutritional deficiencies, hormonal
Pearl: Magnetic resonance imaging (MRI) sialography has changes associated with menopause, local oral infections,
virtually replaced invasive procedures for visualizing the denture-related lesions, hypersensitivity reactions, medi-
ductal structure of major salivary glands. cations, and systemic diseases including diabetes mellitus
12 Sjögren’s Syndrome 115

(Maltsman-Tseikhin et al. 2007; Patton et al. 2007). In many the drugs reduce oral symptoms in many patients, they do
cases, no clear cause can be found, and the burning mouth is attrib- not have any effect on the prevention of dental caries, which
uted to a local neuropathy or to a manifestation of depression. are progressive in most patients with significant hyposaliva-
In one study of 45 patients with the complaint of a burn- tion. They also seem to have no effect on reducing or pre-
ing mouth in whom no cause could be established (Patton venting oral candidiasis.
et al. 2007), either a localized neuropathy or psychogenic Managing the oral manifestations of chronic hyposaliva-
etiology was suggested as the cause. A therapeutic trial of tion varies with its severity, but in general requires: (1) that
topical clonazapam and antioxidants (alpha-lipoic acid) was patients are supervised adequately in a program of dental
employed in some patients. Systemic agents such as gabap- caries prevention and treatment; (2) recognition and treat-
entin, pregabalin, or antidepressants with benefit in neuropa- ment of oral candidiasis (usually of the chronic erythematous
thy (SjSRIs, SNSRIs, or NSRIs) have been employed in type, which occurs in about one-third of SjS patients); and
other patients. Agents with known anticholinergic side (3) reduction of their oral symptoms. Oral symptoms can be
effects such as tricyclic antidepressants are not tolerated reduced by the prescription of a sialogogue, the use of a
well. saliva substitute, and by the elimination of drugs with anti-
cholinergic effects.
Pearl: Decreased saliva volume can lead to complaints of
dysphagia.
Comment: Subjective difficulty swallowing is one of the
symptoms elicited in the diagnostic criteria for SjS. Some 12.5 Parotid and Submandibular
SjS patients have an underlying esophageal motility disorder Involvement
that is part of their connective tissue disease syndrome.
However, most patients with this complaint have adequate Myth: Isolated submandibular gland enlargement is typical
mechanical deglutition (Mandl et al. 2007a). The reported of SjS.
dysphagia results from the decreased volume and increased
viscosity of saliva that are characteristic of SjS, which do not Reality: Patients with SjS have a range of major salivary gland
provide adequate “bulk” for swallowing (Belafsky and involvement, but isolated submandibular gland enlargement is
Postma 2003). This imbalance due to decreased saliva vol- an atypical finding that should make one consider other
ume and content predisposes to dysfunction of the gastroe-
sophageal sphincter, leading to gastroesophageal and
laryngotracheal reflux. Laryngotracheal reflux should be
suspected if the patient repeatedly clears her throat during
conversation or has unexplained hoarseness.
Pearl: SjS patients have more difficulty swallowing certain
types of tablets or capsules than other patients.
Comment: SjS patients have deglutition problems, as noted.
As a result, they have difficulty with both swallowing and
esophageal transit of many medications. When available, the
use of smaller, “polished” tablets is preferred. An example is
“branded” Plaquenil (hydroxychloroquine), which is a pol-
ished tablet, compared with some generic forms of the drug
that are larger in size and contain a residue with bitter taste on
the unpolished surface. Other unpolished capsules, e.g., those
containing iron, can adhere to the dry esophageal mucosa,
where they cause erosion. For these reasons, “polished” Fig. 12.6 Isolated submandibular gland enlargement. This finding is
(coated) tablets are preferred to “sticky” capsules. atypical of Sjögren’s syndrome. This patient was sent for rheumato-
logic evaluation because of xerostomia and major salivary gland
Myth: Symptoms and signs of chronic dry mouth are man- enlargement, but she had no parotid enlargement. A left submandibular
aged adequately by prescribing pilocarpine or cevimeline. adenectomy yielded the diagnosis of chronic sclerosing sialadenitis
with intense IgG4-staining plasma cells within the gland. These find-
Reality: Sialogogues such as pilocarpine or cevimeline ings were consistent with IgG4-associated systemic disease, a spectrum
of conditions in which elevated levels of IgG4 are often found within
increase saliva production to some degree in all patients who
serum and tissue infiltration of IgG4-producing plasma cells is found
have remaining salivary tissue. However, normal levels of within the pancreas, biliary ducts, salivary glands, and other organs
secretion are not restored by these interventions. Although [Figure courtesy of Dr. John Stone]
116 M. Ramos-Casals et al.

disorders (Fig. 12.6). A variety of hematopoietic malignancies Gradually increasing swelling of parotid glands in SjS
and carcinomas can cause submandibular gland enlargement, may be associated with an enlarging benign lymphoepithe-
as can amyloidosis and certain chronic infections such as lial lesion or MALT lymphoma. An incisional biopsy is
tuberculosis. A more recently recognized cause of subman- required to distinguish between these two entities. A rapid
dibular gland enlargement is IgG4-associated systemic dis- increase in the size of an enlarged gland associated with
ease, a disorder in which IgG4-producing plasma cells symptoms and signs of acute inflammation suggests a super-
infiltrate certain exocrine glands, e.g., the salivary glands, imposed bacterial sialoadenitis and calls for systemic anti-
pancreas, and biliary tract (Kamisawa and Okamoto 2006). biotic treatment. However, rapid increase in glandular
In a classic paper on SjS, the clinical presentations of 62 swelling without signs of acute inflammation often heralds
SjS patients were reported (Bloch et al. 1965). Thirty-four of transformation to a high-grade lymphoma or other malig-
the patients (55%) had parotid gland enlargement and 10 nant neoplasm. Tissue examination is obviously indicated in
(16%) had submandibular gland enlargement, but not a single this setting.
patient had isolated submandibular gland disease (i.e., sub-
mandibular gland enlargement without parotid enlargement). Pearl: Glandular enlargement in SjS can wax and wane.

Myth: Parotid gland enlargement in patients with primary Comment: The course of glandular enlargement in SjS varies
SjS often responds to a course of glucocorticoids. from patient to patient. Some develop parotid enlargement
that persists largely unchanged for years. In others, the glan-
Reality: Parotid gland enlargement does not respond to glu- dular involvement waxes and wanes over periods of several
cocorticoids. The recommended therapy is the local applica- weeks or months. This is often asymmetric.
tion of heat. If a superimposed infection is suspected,
antibiotic therapy is mandatory.
Pearl: Salivary gland enlargement in SjS must be considered
carefully. 12.6 Extraglandular Involvement
Comment: Salivary gland enlargement is observed in approx-
imately one-third of patients with primary SjS or secondary Myth: Sicca symptoms are usually the first manifestation of SjS.
SjS. Enlarged salivary glands are usually bilateral, firm to
Reality: A chronic, asymmetric, purely sensory neuropathy
palpation, either symmetrical or asymmetrical in size, and
can be the first manifestation of SjS. Sensory symptoms pre-
minimally symptomatic. Enlargement can be episodic, with
cede the development of other clinical features by several
gradual waxing and waning, or chronic, with gradual pro-
years in some patients (Mori et al. 2005; Denislic and Meh
gression over months or years. The parotid glands, subman-
1997). Sensory deficits may be accompanied by an auto-
dibular glands, or both may be affected. Reports from two
nomic neuropathy (Dyck 2005).
well-documented cohorts found salivary gland enlargement
on examination in 39 and 43% of primary SjS patients and 31 Pearl: Dysphagia in primary SjS patients can be caused by
and 24% of secondary SjS patients, respectively (Bloch et al. esophageal dysmotility.
1965, Daniels et al. 1975). Other conditions that can cause
bilateral salivary gland enlargement are listed in Table 12.4. Comment: Patterns of esophageal dysmotility observed in pri-
mary SjS include aperistalsis, triphasic tertiary contractions,
frequent nonperistaltic contractions, and low contractions
Table 12.4 Bilateral salivary gland enlargement – differential diagnosis
(Tsianos et al. 1985). These esophageal abnormalities do not
Sjögren’s syndromea (lymphoepithelial lesion) correlate with the parotid flow rate, the degree of inflamma-
Viral infectionsb
tory infiltrate of the minor salivary glands, the extraglandular
Chronic granulomatous diseasesa (e.g., sarcoidosis)
Sialadenosisc (associated with: diabetes mellitus, acromegaly,
manifestations, or the presence of autoantibodies.
gonadal hypofunction, hyperlipoproteinemia, hepatic cirrhosis,
anorexia/bulimia, or pancreatitis) Myth: Cutaneous vasculitis in SjS is usually a lymphocytic
Recurrent parotitis of childhood vasculitis.
a
Associated with chronic salivary hypofunction
b Reality: In fact, vasculitis of the skin in SjS is usually leuko-
Mumps is usually acute but CMV, HIV, and Coxsackie may cause pro-
longed enlargement cytoclastic. The most common clinical feature of this is pal-
c
Sialadenosis (idiopathic acinar hypertrophy) affects parotid glands pable purpura (Ramos-Casals et al. 2004a). Cryoglobulinemia
only; always bilaterally symmetrical enlargement that is soft and non- is present in up to 30% of vasculitis cases associated with
tender to palpation; not associated with symptoms or signs of salivary
SjS. As discussed elsewhere in this chapter, this cryoglobu-
hypofunction; diagnosis by clinical presentation and disease associa-
tion; biopsy of affected glands is unnecessary (From Klippel et al. 2008. linemia is typically part and parcel of the SjS rather than a
With kind permission from Springer Science + Business Media.) complication of a hepatitis C infection.
12 Sjögren’s Syndrome 117

Other prominent features of skin vasculitis in SjS are: disease manifestation observed in patients with SLE. The
recurrent arthritic episodes in primary SjS and SLE patients
• Urticarial lesions in approximately 25%.
can lead to hand deformities called Jaccoud’s arthropathy.
• Medium-vessel disease that mimics polyarteritis nodosa
In primary SjS, this arthritis is not one of an erosive, joint-
(Fig. 12.7). Fortunately, this occurs in less than 5% of SjS
destructive nature (Tsampoulas et al. 1986). In contrast, a
patients with vasculitis.
patient who’s SjS is secondary to rheumatoid arthritis typi-
Myth: Patients with primary SjS suffer from erosive symmet- cally develops joint erosions.
rical arthritis.
Pearl: The clinical course of Raynaud’s phenomenon is
Reality: The arthritis of primary SjS is characterized by short- milder in primary SjS than in other autoimmune diseases
lived episodes of joint inflammation that remit spontaneously, such as systemic sclerosis.
usually within days. This arthropathy resembles a similar
Comment: Raynaud’s phenomenon is an early clinical fea-
ture of SjS in nearly 50% of patients and can appear before
a sicca symptomatology (Skopouli et al. 1990; Youinou et al.
1990; García-Carrasco et al. 2002b). The clinical course of
Raynaud’s phenomenon is milder in primary SjS than in
other systemic autoimmune diseases such as systemic sclero-
sis, in which this complication is often associated with digital
ulcers and ischemia. In primary SjS, Raynaud’s phenomenon
is rarely accompanied by vascular complications and only
40% of patients require pharmacological treatment.
Myth: Interstitial fibrosis is the predominant form of pulmo-
nary involvement in primary SjS.
Reality: Various studies have recently analyzed pulmonary
involvement in primary SjS (Papiris et al. 1999; Taouli et al.
b 2002; Franquet et al. 1999). Most investigators have reported
a predominance of bronchial and bronchiolar involvement
rather than interstitial disease. One study identified the
ground-glass pattern as the predominant pattern observed on
computed tomography, but this coexisted with bronchiectasis
in some cases (Fig. 12.8) (Wright et al. 2003). The typical
symptoms of patients with bronchial or bronchiolar disease
are cough, dyspnea, and recurrent respiratory infections.
Pearl: SjS patients are more likely to develop mucus plugs.
Comment: SjS patients suffer not only from dryness of the
eyes and mouth, but also from dryness of the skin, vagina,
and bronchi. Bronchial dryness becomes especially impor-
tant in two situations. First, in the presence of upper airway
infection, there is a predilection to develop inspissated mucus
plugs. This tendency can be exacerbated by over-the-counter
cold preparations, which usually contain anticholinergic
drugs. Second, mucus plugs may occur postoperatively as a
result of both the anticholinergic drugs used during anesthe-
sia and the dehydration sustained during surgery.
Pearl: Dry cough, a common manifestation of SjS, often indi-
Figs. 12.7 (a, b) Medium-vessel vasculitis (a) that mimics polyar- cates the presence of airway involvement and desiccation.
teritis nodosa is an unusual complication of Sjögren’s syndrome (SjS),
occurring in less than 5% of patients. The more common form of cuta- Comment: A chronic, nonproductive cough in a patient with
neous vasculitis in SjS is palpable purpura (b), which is usually asso-
primary SjS should alert the clinician to the possibility of
ciated with a leukocytoclastic as opposed to lymphocytic vasculitis.
[Figure 12.7a and b courtesy of Manuel Ramos-Casals and H.M. bronchitis sicca. The most common symptom of laryngeal,
Moutsopoulos] tracheal, and bronchial involvement is a dry, persistent cough.
118 M. Ramos-Casals et al.

Altered liver profile

No SjS-related SjS-related

Hepatotoxic drugs
Steatosis
Chronic ilness

Viral infection Autoimmune liver disease

-Mediterranean areas -Younger and female patients

-Older and male patients -Hypergammaglobulinemia
-Cryoglobulins+ -Autoantibodies+++
-Hypocomplementemia

HCV infection
ANA+
SMA+ AMA-M2+ AMA-

Fig. 12.8 Bronchiectasis and ground-glass pattern in a 62-year-old

Autoimmune cholangitis
woman with primary Sjögren’s syndrome and interstitial lung disease Type1 AIH PBC Sclerosing cholangitis
[Figure courtesy of Dr. Manuel Ramos-Casals]

If the patient does not have other features of SjS, the diagno- Fig. 12.9 Algorithm for the evaluation of abnormal liver function tests
in Sjögren’s syndrome. AIH = autoimmune hepatitis; PBC = primary
sis may be missed and the patient treated incorrectly for biliary cirrhosis; ANA = antinuclear antibody; SMA = anti-smooth
asthma or bronchitis. Despite the fact that upper airway muscle antibody; AMA = antimitochondrial antibody [Figure courtesy
symptoms afflict 50–70% of patients with SjS, only 20% of Dr. Manuel Ramos-Casals]
have abnormalities that can be identified by rhinoscopy or
indirect laryngoscopy (Freeman et al. 2005). prevalence of 13% (Ramos-Casals et al. 2006b). HCV infec-
The underlying pathology in these patients consists of tions were nearly 3 times more common as a cause of liver
peribronchial infiltrates that lead to small airway disease disease than was autoimmune hepatitis. Of course, the preva-
(Papiris et al. 1999). The classic pulmonary function test lence of HCV infection varies widely in different regions of
manifestation of this disease complication is a decreased the world, but this finding (from Spain) underscores the impor-
MEF25–75. Chest radiographs in these individuals are usu- tance of chronic HCV infection as a cause of liver disease in
ally normal or show an ill-defined pattern of infiltrates that SjS patients from regions with a high HCV prevalence.
suggests interstitial lung disease. High-resolution computed
Myth: “Liver function tests” differentiate cleanly between
tomography of the lungs reveals thickened bronchial walls.
autoimmune and viral hepatitis in patients with primary SjS.
Lymphocytic interstitial pneumonitis evolves only rarely.
Bronchitis sicca does not respond well to bronchodilators. Reality: The major forms of liver disease that are relevant to
However, the oral administration of pilocarpine hydrochlo- SjS are primary biliary cirrhosis, autoimmune liver disease,
ride (20 mg/day) may be helpful. and HCV infection. The differential diagnosis of liver disease
in patients with primary SjS is important, because the thera-
Pearl: The development of pleurisy in a patient with SjS sig-
peutic approaches and prognoses of the various forms of
nals the presence of an additional autoimmune condition.
hepatic dysfunction in this disease vary substantially (see
Comment: Pleurisy is an extremely rare manifestation of pri- Algorithm, Fig. 12.9). Unfortunately, routine laboratory tests
mary SjS. In contrast, in patients with secondary SjS, pleu- are not helpful in distinguishing among the common forms of
risy occurs in up to 30% (Moutsopoulos et al. 1979b; liver disease in SjS or in differentiating such hepatic complica-
Manoussakis et al. 2004). tions from HCV infection. Serum concentrations of hepatic
transaminases (aspartate and alanine aminotransferase),
Pearl: Hepatitis C virus (HCV) infection is an important
gamma glutaryl transferase, bilirubin, and alkaline phosphatase
cause of liver function test abnormalities in patients with SjS
are all elevated to a similar degree in patients with primary
in some geographic areas.
biliary cirrhosis, autoimmune hepatitis, and HCV infection.
Comment: Chronic HCV infection was the main cause of Immunological evaluations are also imperfect in differen-
liver dysfunction in a large series of patients with SjS, with a tiating among viral-associated and autoimmune causes of
12 Sjögren’s Syndrome 119

hepatic dysfunction (Ramos-Casals et al. 2006b). Patients (Yamamoto et al. 2005). The concept of IgG4-related sys-
with chronic HCV infection have a higher frequency of cryo- temic disease is discussed elsewhere in this chapter.
globulins and low complement levels, but patients with SjS
Pearl: Glomerulonephritis is a rare complication of primary
can have these abnormalities, too. Patients with autoimmune
SjS.
liver disease have a higher frequency of autoantibodies such
as antismooth muscle and antimitochondrial antibodies, but Comment: Tubulointerstitial disease is regarded widely as the
variability in the quality of autoantibody assays across dif- most common form of renal dysfunction in primary SjS.
ferent laboratories often makes these data less helpful than However, both tubular and glomerular diseases have important
one would wish. pathogenic, clinical, and prognostic implications in primary SjS.
Among 27 SjS patients with documented renal biopsy reported
Pearl: Elevated liver enzymes (2–3 times) or antimitochon-
in the literature (Bossini et al. 2001; Goules et al. 2000), 15 had
drial antibodies in the serum of a patient with primary SjS
tubulointerstitial nephritis, 11 had glomerulonephritis, and one
suggest that the patient has autoimmune cholangiitis.
had both tubulointerstitial disease and glomerulonephritis.
Comment: Approximately 5% of patients with primary SjS Among the patients with glomerulonephritis, the most
have asymptomatic elevation of liver enzymes or antimito- common glomerular lesions were membranoproliferative
chondrial antibodies (Skopouli et al. 1994a). Liver biopsy in (seven patients), mesangial proliferative (six patients), and
these patients shows lymphocytic infiltrates around bile ducts, membranous (two patients). Cryoglobulinemia was detected
reminiscent of early primary biliary cirrhosis (stage I-II). The in half of the patients with glomerulonephritis. Only two
progression of these lesions is very slow and usually does not patients ultimately developed end-stage renal disease.
lead to liver failure. Tubulointerstitial disease in SjS, which is usually found in
This clinical entity must be distinguished from the IgG4- younger patients, is characterized by an indolent course in
associated systemic disease described above, which can mimic which renal dysfunction is often subclinical. IgM and comple-
SjS through its involvement of submandibular glands and also ment proteins comprise the primary deposits in the glomerulo-
affect the biliary tree (Kamisawa and Okamoto 2006). nephritis of SjS. This contrasts with the immunopathologic
lesion of lupus nephritis, in which a “full house” of immunore-
Myth: An asymptomatic increase of serum amylase levels in
actant deposition (immunoglobulin and complement) is
a patient with primary SjS should alert the clinician to the
observed (Moutsopoulos et al. 1978). SjS glomerulonephritis
possibility of pancreatic cancer development.
usually responds to glucocorticoids at a starting dose of 0.5–
Reality: High serum amylase levels are detected in one- 1.0 mg/kg of body weight per day.
fourth of patients with primary SjS patients (Tsianos et al.
Pearl: Recurrent renal colic in a patient with primary SjS
1984). In the majority of these individuals, the amylase arises
suggests that the patient has interstitial nephritis.
from the inflamed salivary glands. In a small percentage of
primary SjS patients, the amylase originates from the pan- Comment: Interstitial nephritis can be an early manifestation
creas. The later group of individuals suffers from subclinical of SjS (Goules et al. 2000). This condition is usually sub-
pancreatitis. clinical, and is manifested (if sought) by a low urine specific
gravity (hyposthenuria) and an alkaline urine pH. An ele-
Myth: Pancreatitis is a common extraglandular feature of SjS.
vated serum creatinine seldom occurs as a complication of
Reality: Studies in the 1970s and 1980s found a high fre- interstitial nephritis.
quency of altered pancreatic function in primary SjS (>40%), Nephrocalcinosis that presents with renal colic is a com-
although no data were presented on the clinical significance mon clinical expression of distal renal tubular dysfunction in
of these altered tests. These studies led to the consideration of these patients. The classic renal manifestation of SjS is a dis-
pancreatic involvement as one of the typical extraglandular tal RTA caused by interstitial nephritis. The distal RTA can
features of primary SjS. However, the frequency of clinical lead to hypokalemia. Patients who develop distal RTAs may
pancreatitis is very low in large series of patients with primary require spironolactone to control hypokalemia, but the use of
SjS (<2%) (Ramos-Casals et al. 2008). In patients with pri- loop diuretics should be discouraged as this may exacerbate
mary SjS, pancreatic involvement is usually asymptomatic hypokalemia. Proximal RTAs, which can lead to osteomala-
and is demonstrated by altered pancreatic function tests. cia and the Fanconi syndrome, are rare in SjS (but reported)
Clinically significant pancreatitis in primary SjS is rare. (Goules et al. 2000).
Some of the early reports of “pancreatitis” occurring in
Pearl: A wide range of peripheral neuropathies can compli-
association with “SjS” may actually represent cases of IgG4-
cate SjS.
related systemic disease, an emerging spectrum of illness
that can affect multiple exocrine organs but which appears to Comment: Early attention to peripheral neuropathies is
be a different entity altogether compared with primary SjS extremely important. Sensory neuropathies and dorsal root
120 M. Ramos-Casals et al.

ganglionopathies are the most common forms that afflict SjS Cardiovascular reflex tests are more likely to be abnormal in
patients (Mellgren et al. 2007). The pathological findings in patients with SjS than among healthy controls.
cases of sensory ganglionopathy consist of loss of neuronal
Pearl: Trigeminal neuralgia is a common complication of
cell bodies and the infiltration of T cells.
primary SjS.
Peripheral motor neuropathies can include mononeuritis
multiplex (which stems from vasculitis) or CIDP (chronic Comment: Peripheral neuropathy has been described in about
idiopathic demyelinating polyneuropathy), the latter of 10–20% of patients with primary SjS. The major forms of
which is linked in some cases to antimyelin associated gly- neuropathy observed include sensory ataxic neuropathy, pain-
coprotein. SjS patients can also suffer trigeminal and other ful sensory neuropathy without sensory ataxia, trigeminal
cranial neuropathies, autonomic neuropathy, and mixed pat- neuropathy, multiple mononeuropathy, multiple cranial neu-
terns of neuropathy. ropathy, autonomic neuropathy, and radiculoneuropathy.
Sural nerve biopsy may show vascular or perivascular Trigeminal neuropathy is described in about 15% of patients
inflammation of small epineurial vessels (both arterioles and with any kind of neuropathy. It is usually unilateral. The pain
venules) and in some cases necrotizing vasculitis. The loss of is distributed in the regions that are innervated by the branches
myelinated nerve fibers is common and loss of small-diameter of the trigeminal nerve.
nerve fibers occurs. Peripheral neuropathy in primary SjS often
Myth: Central nervous system vasculitis is among the most
is refractory to treatment with currently available agents.
common extraglandular manifestations of primary SjS.
Pearl: Some patients with primary SjS present with a painful
Reality: This association, described first in the late-1980s, is
sensory neuropathy but normal nerve conduction studies.
a matter of considerable controversy (Alexander et al. 1988).
Comment: Small-fiber neuropathy occurs in patients with A wide variety of central nervous system (CNS) disease
primary SjS (Mori 2003; Gorson 2003). These patients often manifestations have been described in primary SjS patients.
present with burning pain in the feet. Small-fiber neuropathy This pathology extends from cognitive dysfunction to demen-
can develop either in isolation as the sole neurologic mani- tia. Seizures, aseptic meningitis, multiple sclerosis-like
festation of disease, or in combination with larger sensory lesions, and vasculitis have been also described. If the inves-
fiber involvement. The diagnosis often relies on quantitative tigators of these studies had applied strict criteria for the
sensory testing and sural nerve biopsy, but skin biopsy is an classification of the disease, then the individuals with CNS
increasingly useful technique for demonstrating small-fiber involvement probably would have been categorized as suf-
neuropathy (Chai et al. 2005). The pathological finding on fering from overlap syndromes with features of both SjS and
skin biopsy is a decrease in the density of epidermal nerve SLE (Ioannidis and Moutsopoulos 1999).
fibers (Gøransson et al. 2006). Patients with small-fiber neu- The frequency of true CNS disease among patients with
ropathy have normal nerve conduction studies, because the “pure” primary SjS has probably been overestimated. In one
size of nerve fibers involved is below the resolution of con- study of more than 1,000 patients, symptomatic CNS involve-
ventional electrodiagnostic studies. ment was observed in only 21 (<2%) (Ramos-Casals et al.
2008). Nevertheless, there is clearly a subset of patients who
Pearl: Sensory neuropathy in primary SjS evolves in a
develop important neurological illnesses involving the brain
chronic, insidious manner and usually demonstrates a poor
and spinal cord. These comprise a subset of the patients who
response to glucocorticoids and immunosuppressive agents.
have antibodies directed against the Ro/SSA antigens. Some
Comment: Sensory neuropathy becomes symptomatic before of these patients develop neurological features of a disease
the underlying disorder is recognized in nearly half of pri- that are extremely difficult to distinguish from multiple scle-
mary SjS cases. The majority of patients with sensory neu- rosis or from SLE associated with antiphospholipid antibod-
ropathies (>70%) have a long-term, insidious evolution of ies (Delalande et al. 2004). The precise nature of this patient
their symptoms (Font et al. 2003). The symptoms are gener- subset requires further study.
ally refractory to treatment with glucocorticoids and immu-
Pearl: SjS patients may develop myelopathy and optic neuri-
nosuppressive agents (Font et al. 2003). Recent reports have
tis, similar to “Devic’s disease” in multiple sclerosis patients
suggested a better therapeutic response to intravenous immu-
(MS).
noglobulins and to rituximab (Gorson 2007).
Comment: The therapeutic issue for the rheumatologist is
Pearl: Patients who describe “light-headedness” should have
whether this represents central nervous system vasculitis that
their blood pressure and pulse checked when both supine and
might require high-dose glucocorticoids or biologic agents,
erect.
or whether the findings actually represent MS.
Comment: Autonomic neuropathy is common among SjS Initial studies on the correlation of MS and SjS were com-
patients (Stojanovich et al. 2007; Mandl et al. 2007b). plicated by the “ascertainment” bias of patients with sicca
12 Sjögren’s Syndrome 121

symptoms that were referred to institutions specializing in Comment: Lymphomas that develop in primary SjS patients
MS. Other centers pointed out that MS patients often exhibit are extranodal in 80% of cases. The most common site of
dryness in the absence of positive salivary gland biopsies, extranodal lymphoma development is the salivary glands
leading to the suggestion that their dryness is due to central (Fig. 12.10). Ninety percent of primary SjS patients who
nervous involvement involving the cholinergic outflow tracts. develop lymphoma have histories of major salivary gland
enlargement during their disease course (Fig. 12.11). Nearly
Myth: An MS patient with a positive ANA has SjS.
30% of such patients have persistent as opposed to intermit-
Reality: A common clinical question is whether the finding tent glandular enlargement (Voulgarelis et al. 1999).
of ANA in a patient with an abnormal brain MRI means that Myth: The incidence of fibromyalgia in SjS patients is the
the patient has SjS with involvement of the central nervous same as in the general population.
system. Addressing this question has been challenging,
because positive ANA results are found in normal individu- Reality: An increased prevalence of fibromyalgia is found in
als and in up to 20% of MS patients who lack any other evi- both SjS and SLE. In fact, the chronic fatigue and myalgias
dence of a connective tissue disorder (Collard et al. 1997; are such prevalent factors that they have made clinical drug
Ferreira et al. 2005). development rather difficult, because fibromyalgia symptoms

Pearl: More than 25% of patients with primary SjS have sen-
sorineural hearing loss.
Comment: Sensorineural hearing loss was detected in 38 of 140
primary SjS patients (27%) whose results were pooled from four
studies (Ramos-Casals et al. 2006a, b). Associations with immu-
nologic parameters such as ANA, antiphospholipid antibodies,
and anti-Ro/SSA or anti-La/SSB antibodies have been postu-
lated but not proven. Sensorineural hearing loss in primary SjS
preferentially affects high-frequency hearing, but deficits often
remain subclinical (Boki et al. 2001). Retrocochlear disease and
symptoms of vestibular dysfunction are not typical of SjS.
Pearl: Patients with primary SjS are at increased risk for
lymphoma.
Fig. 12.10 The most common site of extranodal lymphoma develop-
Comment: Primary SjS patients are at higher risk of lym- ment in Sjögren’s syndrome is the salivary glands. This figure reveals
phoma than are healthy individuals and patients with other parotid gland enlargement in a patient with primary SjS who has
autoimmune diseases (Kassan et al. 1978). Different studies developed extranodal lymphoma during follow-up [Figure courtesy of
Dr. Haralampos M. Moutsopoulos]
have estimated the relative risk of lymphoma in patients with
primary SjS when compared with the general population to
range from 10- to 44-fold.
A meta-analysis of five studies in four different countries
that included a combined total of 1,200 primary SjS patients
confirmed the high risk of non-Hodgkin’s lymphoma and
calculated a standardized incidence rate (SIR) of 18.8
(Zintzaras et al. 2005). This SIR contrasts with those for SLE
and RA of 7.4 and 3.9, respectively, from the same study.
Lymphoma tends to occur in a subgroup of SjS patients
who express special risk factors early in their disease course.
These risk factors include palpable purpura and C4 hypoco-
mplementemia. This patient subgroup has increased mortal-
ity (Skopouli et al. 2000; Ioannidis et al. 2002). The long-term
risk of lymphoma for patients with primary SjS is often esti-
mated to be on the order of 5%.
Pearl: Persistently hard enlargement of the lacrimal or
Fig. 12.11 Ocular B cell lymphoma in a patient with Sjögren’s syn-
parotid glands in a patient with primary SjS should alert the drome and chronic hepatitis C virus (HCV) infection [Figure courtesy
clinician to the possibility of an extra-lymphoid lymphoma. of Dr. Manuel Ramos-Casals]
122 M. Ramos-Casals et al.

constitute a sizeable effect on patients’ quality-of-life evalu- Comment: Elevated ESRs in SjS correlate directly with the
ations (Bowman et al. 2004; Pillemer et al. 2005). degree of hypergammaglobulinemia (Ramos-Casals et al.
2002). Moreover, both these parameters are normally found
Pearl: Thyroid disease is more frequent in SjS patients.
in patients with primary SjS who are rheumatoid factor posi-
Comment: The most common thyroid disorder found in asso- tive or who have autoantibodies directed against either the
ciation with SjS is autoimmune thyroiditis (Jara et al. 2007). Ro/SSA or La/SSB antigens. Clinicians who observe high
Subclinical hypothyroidism is common among patients with ESRs in patients with SjS should therefore not leap to con-
primary SjS. In one study, primary SjS was 10 times more fre- clusions about occult infections, subclinical malignancies, or
quent in patients with autoimmune thyroid disease, and auto- the presence of systemic vasculitis; the abnormality may
immune thyroiditis was 9 times more frequent in primary SjS. simply be the SjS itself. In such patients, the C-reactive pro-
One dissenting case-control study reported no significant tein is usually normal.
differences in patients with primary SjS and a group of age-
Pearl: Highly elevated serum C-reactive protein levels in a
and sex-matched controls (Ramos-Casals et al. 2000). In that
patient with primary SjS should raise the suspicion of an
study, subclinical hypothyroidism affected 11% of the cases
infection.
and 8% of the controls.
Comment: Patients with primary SjS typically do not mount
Pearl: Dyspareunia is a common premenopausal complaint
an acute phase response related to their disease itself, at least
of primary SjS.
not one associated with elevated C-reactive protein levels
Comment: Almost half of premenopausal women with pri- (Moutsopoulos et al. 1983a). The finding of a strikingly ele-
mary SjS complain of dyspareunia (Skopouli et al. 1994b). vated serum C-reactive protein should trigger careful scru-
Such symptoms are uncommon in age-matched control women tiny for an infection. Systemic vasculitis occurring in the
and when present generally have an obvious etiology (trauma setting of primary SjS can also lead to an elevated C-reactive
or inflammation). Despite dyspareunia, primary SjS patients protein level.
appear to have similar frequency of sexual activity, fertility,
Pearl: Patients with anti-Ro/SSA or anti-La/SSB antibodies
and parity when compared with age-matched controls.
often develop leukopenia or thrombocytopenia.
Pearl: Salivary gland toxicity may accompany the treatment
Comment: The relationship between anti-Ro/SSA antibodies
of thyroiditis of 131-iodine.
and hematologic alterations was described in the early 1980s
Comment: Salivary gland toxicity is a potential adverse effect in a study of 75 patients (Alexander et al. 1983). This link
of high-dose radioiodine (131-I) (Hyer et al. 2007). One was confirmed in a multivariate analysis of data on 400 pri-
study of 20 patients revealed that 11 (15%) had symptoms of mary SjS patients, which found that leukopenia was associ-
xerostomia within the first 48 h of receiving such therapy. ated with anti-Ro/SSA antibodies and thrombocytopenia
These symptoms persisted for at least 12 months in seven with anti-La/SSB antibodies (García-Carrasco et al. 2002a).
patients. Medical or surgical interventions may be preferable In a series of 1,010 Spanish patients with primary SjS, the
to radioiodine administration in patients with SjS, whose odds ratio of leukopenia associated with anti-Ro/SSA
salivary production is already compromised. antibodies was 2.6. The odds ratio of thrombocytopenia
associated with anti-La/SSB antibodies was 2.3 (Ramos-
Pearl: Urinary tract symptoms and cystitis are underdiag-
Casals et al. 2008).
nosed in primary SjS.
Pearl: Neutropenia is a relevant hematologic finding in pri-
Comment: Two recent studies have investigated lower uri-
mary SjS.
nary tract symptoms in primary SjS. Severe urological symp-
toms (increased frequency, urgency, and nocturia) were Comment: The neutrophil count should be monitored in SjS
reported in 61% of patients in one study (Walker et al. 2003). patients, especially in those with recurrent infections. Nearly
Biopsy-proven interstitial cystitis was found in some cases. 30% of patients with primary SjS have autoimmune neutro-
Another study found that 5% of SjS patients fulfilled the cri- penia (Brito-Zerón et al. 2008). The percentage of patients
teria for interstitial cystitis (Leppilahti et al. 2003). with SjS who develop this hematological complication is
substantially higher than that of other cytopenias, e.g., leuko-
penia or thrombocytopenia. Neutropenia is associated with a
higher rate of hospital admission due to infection.
12.7 Laboratory Findings The occurrence of agranulocytosis is rare in primary SjS
(only about 2% of patients). Agranulocytosis is observed pri-
Pearl: The height of the erythrocyte sedimentation rate (ESR) is marily in patients with a hematopoietic malignancy (mainly
SjS correlates with the level of immunoglobulins in the serum. B cell lymphoma). The etiopathogenic role of antineutrophil
12 Sjögren’s Syndrome 123

antibodies in such patients, if any, is unclear. Two studies of autoantibodies directed against nonnuclear antigens,
have found no correlation between autoantibodies to surface such as antismooth muscle antibodies, has not been studied
neutrophil antigens among SjS patients who had agranulocy- thoroughly. In a series of 335 patients, antismooth muscle
tosis (Coppo et al. 2003; Lamour et al. 1995). antibodies were detected in 208 (62%) (Nardi et al. 2006).
However, no particular associations with any clinical feature
or laboratory abnormality are yet known.
Pearl: Clinical events associated with antiparietal cell anti-
12.8 Immunological Assays bodies and antiphospholipid antibodies are uncommon in
primary SjS.
Pearl: The most likely autoimmune disease to develop in a
Reality: In a study of 335 SjS patients, 208 (27%) had anti-
woman with Raynaud’s phenomenon and autoantibodies to
parietal cell antibodies but only two had either pernicious
Ro/SSA and La/SSB is primary SjS.
anemia or atrophic gastritis (Nardi et al. 2006). A literature
Comment: Raynaud’s phenomenon is a common manifesta- review revealed only four other reported cases of pernicious
tion of patients with primary SjS. This feature appears up to anemia in primary SjS. These data support the concept that
many years before the diagnosis of SjS in approximately the co-occurrence of SjS and pernicious anemia is uncom-
one-third of patients (Skopouli et al. 1990; Youinou et al. mon, despite the fact that both are associated with autoim-
1990; García-Carrasco et al. 2002b). munity. A similar study of 281 patients found that 36 (13%)
had antiphospholipid antibodies, but only four fulfilled the
Myth: Serum from patients with primary SjS who do not have
classification criteria for the antiphospholipid syndrome
autoantibodies to Ro/SSA and La/SSB should be examined
(Ramos-Casals et al. 2006a).
every 6–12 months to detect the appearance of these
autoantibodies. Pearl: The finding of a positive ANA with an anticentromere pat-
tern may have important clinical implications in primary SjS.
Reality: Anti-Ro and anti-La antibodies are present in the
sera of primary SjS patients at or before the time of diagno- Comment: Anticentromere antibodies are detectable by the
sis. The probability that a seronegative primary SjS patient proper interpretation of an immunofluorescence assay for
will become seropositive during follow-up is low. Thus, con- ANA. However, because of the time-intensive nature of
tinued monitoring of the sera of primary SjS patients for immunofluorescence studies, these assays have been replaced
autoantibodies to these antigens makes little sense (Skopouli in many laboratories by enzyme immunoassays. Thus, in
et al. 2000). some clinical settings, anticentromere antibodies must be
assayed by specific enzyme immunoassays.
Pearl: The presence of low serum C4 or palpable purpura in
The finding of anticentromere antibodies in patients pre-
a patient with SjS may predict the development of lymphoma
sumed to have primary SjS may be important, particularly if
in patients with primary SjS.
assays for antibodies to the Ro/SSA and La/SSB antigens are
Comment: Lymphomas developing in SjS may occur in the negative (Salliot et al. 2007). SjS patients who have anticen-
salivary glands, gastrointestinal tract, or lungs. They often tromere antibodies represent a specific clinical subset, which
begin as B cell MALT lymphomas or, in lymph nodes, as may be classified initially as having primary SjS but have a
marginal zone lymphomas. After years of slow progression, higher probability of developing limited systemic sclerosis.
these indolent tumors can progress to rapidly growing, high- In one study, one-fourth of such patients developed limited
grade, large B cell lymphomas. scleroderma (Ramos-Casals et al. 2006a). Prominent
Various studies have identified risk factors for lymphoma Raynaud’s phenomenon, sclerodactyly, and nailfold capilla-
development. These include cryoglobulinemia, hypocomple- roscopic changes are clues to the presence of this clinical
mentemia, extremely low C4 levels, and palpable purpura phenotype (Figs. 12.12a and 12.12b).
(Ioannidis et al. 2002). Additional risk factors for lymphoma The critical feature may be the presence or absence of
development include the onset of SjS at a young age and antibodies to the Ro/SSA and La/SSB antigens. Patients who
prolonged salivary gland enlargement. have such antibodies may be more likely to behave as SjS
rather than as limited scleroderma.
Pearl: Antismooth muscle antibodies have no clinical signifi-
cance in SjS. Myth: ANCA positivity strongly suggests a coexisting sys-
temic vasculitis in patients with primary SjS.
Comment: ANA play a central role in the immunologi-
cal expression of primary SjS, due to the fact that they are Reality: ANCA positivity in SjS patients is common when
usually caused by antibodies directed against extractable sera are tested by immunofluorescence. The preponder-
nuclear antigens (ENAs). However, the clinical significance ance of patients who are ANCA-positive have perinuclear
124 M. Ramos-Casals et al.

Figs. 12.12 (a, b) Sjögren’s

a b
syndrome patients who have
anticentromere antibodies often
have a clinical phenotype that
resembles limited scleroderma:
Raynaud’s phenomenon,
sclerodactyly, and nailfold
capillaroscopic changes. Some
would term this combination of
findings an “overlap” connective
tissue disorder. [Figure courtesy
of Dr. Manuel Ramos-Casals]

immunofluorescence pattern (Font et al. 1998). However, Comment: The clinical expression of SjS-HCV is similar to
these ANCA are epiphenomena that appear irrelevant to dis- primary SjS with respect to the prevalence of glandular fea-
ease pathogenesis or any particular clinical complication of tures and the fulfillment of the 2002 criteria, but differs in
the disease. When cutaneous or systemic vasculitis occurs having a higher prevalence of cryoglobulinemia, liver
in SjS, the usual underlying cause is cryoglobulinemia, not involvement, and neoplasia (mainly B cell lymphoma). The
ANCA (Terrier et al. 2007). immunological expression of SjS-HCV includes a higher
percentage of patients who are anti-Ro/SSA and anti-La/
Myth: A serum monoclonal gammopathy is associated with
SSB antibody negative. Patients with SjS-HCV are also more
an underlying hematological neoplasm in most patients with
likely to have cryoglobulins. This accounts for the higher
primary SjS.
prevalence of rheumatoid factor in their sera and hypocom-
Reality: Circulating monoclonal immunoglobulins or/and free plementemia (particularly C4 hypocomplementemia)
monoclonal light chains are detected in the serum of a consid- (Ramos-Casals et al. 2005).
erable number of extraglandular SjS patients (Moutsopoulos
Pearl: The etiopathologic and clinical significance of HTLV-I
et al. 1983b; Brito-Zerón et al. 2005). The monoclonal light
infection in patients with SjS varies depending on the geo-
chains are also detected in the urine of SjS patients
graphical area.
(Moutsopoulos et al. 1985). The monoclonal spike in the
serum usually consists of IgG, but other types of immunoglob- Comment: In Japan, where HTLV-I is endemic, nearly 25%
ulins have also been reported. Despite the high frequency of of patients with primary SjS have HTLV-I infection
monoclonal immunoglobulins in SjS, only around 5% of SjS (Nakamura et al. 2000). HTLV-I is now considered the viral
patients ultimately develop a B-lymphocyte malignancy. counterpart in Asian countries to HCV in the Mediterranean
region. Although genomic sequences of HTLV-I have been
Pearl: Serum monoclonal gammopathy often indicates the
detected in European patients with primary SjS, in one study
presence of an underlying type II mixed cryoglobulinemia.
the prevalence of HTLV-I was similar between the SjS and
Comment: In patients with primary SjS, detection of serum control groups (Mariette et al. 2000).
monoclonal immunoglobulins may indicate cryoglobuline-
Myth: Mikulicz disease is a specific clinical presentation of
mia. However, a significant percentage of patients with pri-
primary SjS.
mary SjS and cryoglobulinemia have insufficient amounts of
cryoprecipitate (<5%) for immunofixation testing (Brito- Reality: In the late-1800s, Johann von Mikulicz reported a
Zerón et al. 2005). The detection of an IgM kappa monoclonal patient with painless, bilateral, symmetrical swelling of the
spike on serum immunoelectrophoresis strongly suggests a lacrimal, parotid, and submandibular glands (Mikulicz,
type II mixed cryoglobulinemia. This consists most commonly 1892). In 1953, Morgan and Castleman suggested that most
of a monoclonal IgM kappa component and polyclonal IgG. cases classified as Mikulicz disease should actually be con-
sidered to be SjS, because the two conditions appear strik-
ingly on routine histopathological staining (Morgan 1953).
More recent studies and the application of new techniques
12.9 Differential Diagnosis suggest that patients with Mikulicz disease have a distinct clin-
ical, immunological, and histological profile. In contrast to pri-
Pearl: SjS-associated with HCV has a distinct clinical and mary SjS, patients with Mikulicz disease are predominantly
immunological profile but overlaps substantially with pri- male and have higher levels of serum IgG4 (Yamamoto et al.
mary SjS. 2005). They also have lower titers of ANA and are negative for
12 Sjögren’s Syndrome 125

antibodies to the Ro/SSA and La/SSB antigens. The close Comment: The sialotropism of HCV explains its close asso-
association of Mikulicz disease with IgG4-related alterations ciation with SjS and the sicca syndrome. Its lymphotropism,
suggests that its proper new classification is separate from in turn, links the presence of HCV with the synthesis of cryo-
SjS, part of a spectrum of “IgG4-related systemic disease.” globulins and lymphoma. This extrahepatic tropism suggests
The other components of this disease spectrum include the possible development of both SjS and lymphoma in
many cases of “autoimmune pancreatitis,” chronic sclerosing patients with chronic HCV infection. The following charac-
sialoadenitis (Küttner’s tumor), Riedel’s thyroiditis, some teristics apply to patients with SjS who are infected with
cases of cholangitis that mimic primary sclerosing cholangi- HCV (Ramos-Casals et al. 2007):
tis, some cases of retroperitoneal fibrosis, and some cases of
tubulointerstitial nephritis (Takeda et al. 2004). • Strongly positive rheumatoid factor serologies and type II
To date, most of the literature on this disease entity has mixed cryoglobulinemia
derived from Japan, but increasingly the disorder is recog- • A high frequency of parotid enlargement and vasculitis
nized worldwide. The principal importance of recognizing • A high risk for the development of MALT lymphomas.
IgG4-related systemic disease relates to the fact that many
cases of IgG4-related systemic disease respond briskly to Among patients with SjS and HCV who do develop lympho-
glucocorticoids. In addition, many of the disease entities mas, there is a high frequency of primary extranodal involve-
with which it can be confused do not (for example, adenocar- ment in organs in which HCV replicates, e.g., the exocrine
cinoma of the pancreas, primary sclerosing cholangitis, and glands, liver, and stomach.
most features of SjS).
Myth: An abrupt decline in serum immunoglobulin levels
and the loss of seropositivity for autoantibodies heralds the
appearance of lymphoma in primary SjS.
12.10 Prognosis and Outcome Reality: Studies in the 1970s suggested that an abrupt decline
in hypergammaglobulinemia precedes the development of
Myth: The course of primary SjS involves an evolution from lymphoma (Cummings et al. 1971). Other authors have
an organ-specific autoimmune disorder (autoimmune exo- described reductions in baseline serum IgM and IgM rheuma-
crinopathy) to a systemic inflammatory disease and con- toid factor levels. This has not been confirmed by subsequent
cludes with a B cell malignancy. studies (Voulgarelis et al. 1999). More recent data suggest
that the development of lymphoma is associated more closely
Reality: Long-term follow-up of large primary SjS conhorts
with ongoing immunological abnormalities than with their
indicate that the majority of the patients – approximately
disappearance (Pertovaara et al. 2001; Theander et al. 2004;
60% – maintain stable disease courses characterized by
Tzioufas et al. 1996; Brito-Zerón et al. 2007).
quantitative and qualitative changes in tear and saliva secre-
tion, sicca symptoms, and circulating autoantibodies. These Pearl: Purpura, hypocomplementemia, and cryoglobuline-
patients do not develop disease in other organ systems. mia are three key prognostic factors for adverse outcome in
On the other hand, approximately 20% of patients dem- primary SjS.
onstrated extraglandular involvement in addition to the exo-
Comment: Prospective studies of Spanish, Greek, and
crinopathy, early in their disease course. The extraglandular
Swedish patients have identified cutaneous vasculitis, hypo-
clinical manifestations in these patients include small airway
complementemia, and mixed cryoglobulinemia as factors
disease, interstitial nephritis, and autoimmune cholangiitis.
associated with adverse outcomes (development of systemic
This patient subset has parenchymal disease that seldom
vasculitis, B cell lymphoma, or death) (Brito-Zerón et al.
compromises the function of the involved organs signifi-
2007; Ioannidis et al. 2002; Theander et al. 2004). Serum
cantly. The mortality of these patients does not differ from
complement levels and cryoglobulins are key immunological
that of healthy age- and sex-matched controls.
parameters for long-term monitoring.
A third group of primary SjS patients express small-vessel
Cryoglobulinemia in SjS is associated with an increased
vasculitis (palpable purpura) and low C4 levels early in their
risk of both vasculitis and lymphoma. In one series, life-
disease course. These patients are at high risk for systemic
threatening vasculitis was related closely to cryoglobulinemia
vasculitis and for the development of lymphoma when
(Ramos-Casals et al. 2004a). Among the 52 patients with
compared with other SjS patients, and have a higher mortality
cutaneous vasculitis who were described, all six deaths
rate than healthy age- and sex-matched controls (Skopouli
occurred in patients with multisystemic cryoglobulinemic
et al. 2000).
vasculitis. Cryoglobulins and vasculitis were independently
Pearl: The association between SjS and HCV infection may associated with mortality in a multivariate analysis of data
increase the risk of B cell lymphoma. from 266 patients. Both these risk factors were associated
126 M. Ramos-Casals et al.

with hazard ratios (relative risks) of more than 5.0 for mortal- spondyloarthropathies, and inflammatory bowel disease. A
ity during the course of the study (Brito-Zerón et al. 2007). randomized, double-blind, placebo-controlled trial of inflix-
imab in primary SjS showed no evidence of efficacy (Mariette
Myth: Interstitial lung disease in SjS is associated with a
et al. 2004). Two small studies of etanercept came to similar
poor prognosis and should be treated aggressively as in sys-
conclusions (Zandbelt et al. 2004; Sankar et al. 2004).
temic sclerosis.
Pearl: Aggressive diffuse B cell lymphomas in patients with
Reality: Interstitial lung disease can occur early in the course
primary SjS should be treated with chemotherapy (cyclo-
of SjS (Davidson et al. 2000). It tends to afflict patients who
phosphamide, doxorubicin, vincristine, and prednisone) in
have anti-Ro/SSA antibodies, but rarely worsens over follow-
combination with rituximab, a B cell depleting agent.
up. Thus, a conservative approach that does NOT involve
high-dose glucocorticoids and cyclophosphamide is advised. Comment: Combination therapy of the diffuse large-cell
lymphomas that sometimes complicate primary SjS has a
dramatic impact on patient survival (Voulgarelis et al. 2006).
12.11 Systemic Treatment This treatment strategy also appears to be effective in many
cases of SjS complicated by palpable purpura or peripheral
neuropathy. The combination chemotherapy regimen results
Pearl: Hydroxychloroquine is an excellent therapeutic option
in a decrease of circulating cryoglobulins, rheumatoid factor
for treating general SjS symptomatology and musculoskele-
titers, and an increase in serum C4 levels.
tal features.
Pearl: Ask patients about their use of Chinese or other herbal
Comment: Patients with primary SjS often present with con-
medications.
stitutional symptoms, including fever, generalized pain, and
fatigue. Antimalarial drugs have a beneficial effect in many Comment: Many patients do not inform their physicians
such patients, similar to the effects they exert in SLE. about herbal drugs, as they consider them “nutritional” sup-
Hydroxychloroquine (200 mg/day) has been reported to plements. However, the agents may have significant direct
reduce markers of inflammation within saliva (Tishler et al. toxicities on the SjS patient. As an example, some supple-
1999). Hydroxychloroquine may mediate enhanced salivary ments have been reported to cause profound hypokalemia in
secretion through the inhibition of glandular cholinesterase SjS patients with interstitial nephritis (Atalar et al. 2007).
activity (Dawson et al. 2005). In our experience, the “herbal” medicines come in the
One dissenting report came from a small double blind form of “Chinese” herbs or “Indian ayurvedic medicine.” In
crossover trial of 400 mg/day that showed significant decrease addition to the adverse effect of the herb itself, the prepara-
in serum IgG and IgM in the treatment group. However, there tions may be contaminated with heavy metals (especially
was no beneficial clinical effect as expressed in preference common in ayurvedic medications) or pesticides that were
for the active or placebo treatment with regard to symptoms used at the time of crop harvesting. Because there is no regu-
and signs of primary SjS, nor was there any relevant change lation of the manufacture or sale of these “health supple-
in tear gland activity or salivary gland scintigraphy (Kruize ments” by the Food and Drug Administration or any other
et al. 1993). regulatory body, patients use them at their own risk.
Myth: Hydroxychloroquine can alleviate sicca symptoms in
patients with SjS.
Reality: Hydroxychloroquine is beneficial for the nonerosive References
arthritis and skin rashes that occur in primary SjS. Some
experts suggest that this medication has a special role in Abraham CM, al-Hashimi I, Haghighat N. Evaluation of the levels of
hypergammaglobulinemic purpura because it fosters the low- oral Candida in patients with Sjögren’s syndrome. Oral Surg Oral
Med Oral Pathol Oral Radiol Endod. 1998;86:65–8
ering of immunoglobulin levels in serum (Kruize et al. 1993; Alexander E, Arnett F, Provost T, Stevens MB. Sjögren’s syndrome:
Mavragani et al. 2006). However, as noted above, hydroxy- association of anti-Ro(SS-A) antibodies with vasculitis, hemato-
chloroquine is not likely to have any impact on a patient’s logic abnormalities, and serologic hyperactivity. Ann Intern Med.
sicca symptoms. 1983;98:155–9
Alexander EL, Beall SjS, Gordon B, et al Magnetic resonance imaging
Pearl: TNF inhibition is of no value in the treatment of pri- of cerebral lesions in patients with the Sjögren syndrome. Ann
Intern Med. 1988;108:815–23
mary SjS. Almstahl A, Kroneld U, Tarkowski A, Wikstrom M. Oral microbial
Comment: TNF inhibition in primary SjS falls short of the flora in Sjögren’s syndrome. J Rheumatol. 1999;26:110–4
Andonopoulos AP, Drosos AA, Skopouli FN, Acritidis NC,
expectations raised by its efficacy in other autoimmune Moutsopoulos HM. Secondary Sjögren’s syndrome in rheumatoid
conditions, including rheumatoid arthritis, the seronegative arthritis. J Rheumatol. 1987;14:1098–103
12 Sjögren’s Syndrome 127

Andonopoulos AP, Drosos AA, Skopouli FN, Moutsopoulos HM. Daniels TE, Silverman S, Michalski JP, Greenspan JS, Sylvester RA,
Sjögren’s syndrome in rheumatoid arthritis and progressive sys- Talal N. The oral component of Sjögren’s syndrome. Oral Surg Oral
temic sclerosis. A comparative study. Clin Exp Rheumatol. 1989;7: Med Oral Pathol. 1975;39:875–85
203–5 Daniels TE. Evaluation, differential diagnosis, and treatment of xeros-
Aoki N. Klinefelter’s syndrome, autoimmunity, and associated endo- tomia. J Rheumatol. 2000;27(Suppl 61):6–10
crinopathies. Intern Med. 1999;38:838–9 Davidson BK, Kelly CA, Griffiths ID. Ten year follow up of pulmonary
Atalar K, Afzali B, Lams B, et al Falls, hypokalaemia, and a dry mouth. function in patients with primary Sjögren’s syndrome. Ann Rheum
Lancet. 2007;370(9582):192 Dis. 2000;59:709–12
Belafsky PC, Postma GN. The laryngeal and esophageal manifestations Dawson LJ, Caulfield VL, Stanbury JB, et al Hydroxychloroquine ther-
of Sjögren’s syndrome. Curr Rheumatol Rep. 2003;5:297–303 apy in patients with primary Sjögren’s syndrome may improve sali-
Bloch KJ, Buchanan WW, Wohl MJ, Bunim JJ. Sjögren’s syndrome. vary gland hypofunction by inhibition of glandular cholinesterase.
A clinical, pathological, and serological study of sixty-two cases. Rheumatology (Oxford). 2005;44:449–55
Medicine (Baltimore). 1965;44:187–231 Delalande S, de Seze J, Fauchais AL. Neurologic manifestations in pri-
Boki KA, Ioannidis JP, Segas JV, et al How significant is sensorineural mary Sjögren syndrome: a study of 82 patients. Medicine (Baltimore).
hearing loss in primary Sjögren’s syndrome? J Rheumatol. 2001;28: 2004;83:280–91
798–801 Denislic M, Meh D. Early asymmetric neuropathy in primary Sjögren’s
Bossini N, Savoldi S, Franceschini F, et al Clinical and morphological syndrome. J Neurol. 1997;244(6):383–7
features of kidney involvement in primary Sjögren’s syndrome. Drosos AA, Voulgari PV, Psychos DN, Tsifetaki N, Bai M. Sicca syn-
Nephrol Dial Transplant. 2001;16:2328–36 drome in patients with sarcoidosis. Rheumatol Int. 1999;18: 177–80
Boutsi EA, Paikos S, Dafni UG, et al Dental and periodontal status of Dyck PJ. The clinical heterogeneity of immune sensory and autonomic
Sjögren’s syndrome. J Clin Periodontol. 2000;27:231–5 neuropathies with (or without) sicca. Brain. 2005;128(Pt 11): 2480–2
Bowman SK, Ibrahim GH, Holmes G, et al Estimating the prevalence Ferreira S, D’Cruz DP, Hughes GR. Multiple sclerosis, neuropsychiat-
among Caucasian women of primary Sjögren’s syndrome in two ric lupus and antiphospholipid syndrome: where do we stand?
general practices in Birmingham, UK Scand J Rheumatol. 2004;33: Rheumatology (Oxford). 2005;44:434–42
39–43 Font J, Ramos-Casals M, Cervera R, et al Antineutrophil cytoplasmic
Brito-Zerón P, Ramos-Casals M, Bove A, et al Predicting adverse out- antibodies in primary Sjögren’s syndrome: prevalence and clinical
comes in primary Sjögren’s syndrome: identification of prognostic significance. Br J Rheumatol. 1998;37(12):1287–91
factors. Rheumatology (Oxford). 2007;46(8):1359–62 Font J, Ramos-Casals M, de la Red G, et al Pure sensory neuropathy in
Brito-Zerón P, Ramos-Casals M, Nardi N, et al Circulating monoclonal primary Sjögren’s syndrome. Longterm prospective followup and
immunoglobulins in Sjögren syndrome: prevalence and clinical review of the literature. J Rheumatol. 2003;30(7):1552–7
significance in 237 patients. Medicine (Baltimore). 2005;84(2): Forstot JZ, Forstot SL, Greer RO, Tan EM. The incidence of Sjögren’s
90–7 sicca complex in a population of patients with keratoconjunctivitis
Brito-Zerón P, Soria N, Muñoz S, et al Prevalence and clinical relevance sicca. Arthritis Rheumat. 1982;25:156–60
of autoimmune neutropenia in patients with primary Sjögren’s syn- Fox RI, Stern M. Sjögren's syndrome: mechanisms of pathogenesis
drome. Semin Arthritis Rheum. 2008;38(5):389–95 involve interaction of immune and neurosecretory systems. Scand
Caporali R, Bonacci E, Epis O, et al Safety and usefulness of minor J Rheumatol Suppl 2002;(116):3-13
salivary gland biopsy: retrospective analysis of 502 procedures per- Franquet T, Diaz C, Domingo P, et al Air trapping in primary Sjögren
formed at a single center. Arthritis Rheum. 2008;59:714–20 syndrome: correlation of expiratory CT with pulmonary function
Chai J, Herrmann DN, Stanton M, et al Painful small-fiber neuropathy tests. J Comput Assist Tomogr. 1999;23:169–73
in Sjögren’s syndrome. Neurology. 2005;65:925–7 Freeman SR, Sheehan PZ, Thorpe MA, Rutka JA. Ear, nose, and throat
Collard RC, Koehler RP, Mattson DH. Frequency and significance of manifestations of Sjögren’s syndrome: retrospective review of a
antinuclear antibodies in multiple sclerosis. Neurology. 1997;49: multidisciplinary clinic. J Otolaryngol. 2005;34(1):20–4
857–61 García-Carrasco M, Ramos-Casals M, et al Primary Sjögren syndrome:
Coppo P, Sibilia J, Maloisel F, et al Primary Sjögren’s syndrome associ- clinical and immunologic disease patterns in a cohort of 400 patients.
ated agranulocytosis: a benign disorder? Ann Rheum Dis. 2003;62: Medicine (Baltimore). 2002a;81(4):270–80
476–8 García-Carrasco M, Sisó A, Ramos-Casals M, et al Raynaud’s phenom-
Cummings NA, Schall GL, Asofski R. Sjögren’s syndrome. New enon in primary Sjögren’s syndrome. Prevalence and clinical charac-
aspects of research, diagnosis and therapy. Ann Intern Med. 1971;75: teristics in a series of 320 patients. J Rheumatol. 2002b;29: 726–30
937 Gerli R, Muscat C, Giansanti M, et al Quantitative assessment of sali-
Dafni UG, Tzioufas AG, Staikos P, et al Prevalence of Sjögren’s syndrome vary gland inflammatory infiltration in primary Sjögren’s syndrome:
in a closed rural community. Ann Rheum Dis. 1997;56: 521–5 its relationship to different demographic, clinical and serological
Daniels T, Greenspan JS, Cox D, Criswell LA, DeSouza Y, Dong Y, features of the disorder. Br J Rheumatol. 1997;36:969–75
et al Objective measures in Sjögren’s syndrome are strongly associ- Gøransson LG, Herigstad A, Tjensvoll AB, et al Peripheral neuropathy
ated with each other but not with sicca symptoms: analysis of 564 in primary Sjögren syndrome: a population-based study. Arch
enrollees in the SICCA international registry and repository. Neurol. 2006;63:1612–5
Arthritis Rheum. 2007;56 Suppl:S446 Gorson KC, Ropper AH. Muscle Nerve 2003; 28(5):553–60. Positive
Daniels T, Whitcher J. Association of patterns of labial salivary gland salivary gland biopsy, Sjögren syndrome, and neuropathy: clinical
inflammation with keratoconjunctivitis sicca. Analysis of 618 implications
patients with suspected Sjögren’s syndrome. Arthritis Rheumatol. Gorson KC, Natarajan N, Ropper AH, Weinstein R. Rituximab treat-
1994;37:869–77 ment in patients with IVIg-dependent immune polyneuropathy: a
Daniels TE. Benign lymphoepithelial lesion and Sjögren’s syndrome. prospective pilot trial. Muscle Nerve 2007; 35(1):66–9
Ch 6. In Ellis GL, Auclair PL, Gnepp DR, editors. Surgical pathol- Goules A, Masouridi S, Tzioufas AG, et al Clinically significant and
ogy of the salivary glands. Philadelphia: W.B.Saunders; 1991. p. biopsy-documented renal involvement in primary Sjögren’s syn-
83–106 drome. Medicine (Baltimore). 2000;79:241–9
Daniels TE, Fox PC. Salivary and oral components of Sjögren’s syn- Hakansson U, Jacobsson L, Lilja B, et al Salivary gland scintigraphy in
drome. Rheum Dis Clin North Am. 1992;18(3):571–89 subjects with and without symptoms of dry mouth and/or eyes, and
128 M. Ramos-Casals et al.

in patients with primary Sjögren’s syndrome. Scand J Rheumatol. Mandl T, Ekberg O, Wollmer P, et al Dysphagia and dysmotility of the
1994;23:326–33 pharynx and oesophagus in patients with primary Sjögren’s syn-
Helman J, Turner RJ, Fox PC, Baum BJ. 99mTc-pertechnetate uptake drome. Scand J Rheumatol. 2007a;36:394–401
in parotid acinar cells by the Na + /K + /Cl- co- transport system. Mandl T, Wollmer P, Manthorpe R, Jacobsson LT. Autonomic and
J Clin Invest. 1987;79:1310–3 orthostatic dysfunction in primary Sjögren’s syndrome. J Rheumatol.
Hernandez YL, Daniels TE. Oral candidiasis in Sjögren’s syndrome: 2007b;34:1869–74
prevalence, clinical correlations, and treatment. Oral Surg Oral Med Manoussakis MN, Georgopoulou C, Zintzaras E, et al Sjögren’s syn-
Oral Pathol. 1989;3:324–9 drome associated with systemic lupus erythematosus: clinical and
Hershkovich O, Shafat I, Nagler RM. Age-related changes in salivary laboratory profiles and comparison with primary Sjögren’s syn-
antioxidant profile: possible implications for oral cancer. J Gerontol drome. Arthritis Rheum. 2004;50:882–91
A Biol Sci Med Sci. 2007;62:361–6 Mariette X, Agbalika F, Zucker-Franklin D, et al Detection of the tax
Hyer S, Kong A, Pratt B, et al Salivary gland toxicity after radioiodine gene of HTLV-I in labial salivary glands from patients with Sjögren’s
therapy for thyroid cancer. Clin Oncol (R Coll Radiol). 2007;19: 83–6 syndrome and other diseases of the oral cavity. Clin Exp Rheumatol.
Ioannidis JP, Moutsopoulos HM. Sjögren’s syndrome: too many asso- 2000;18:341–7
ciations, too limited evidence. The enigmatic example of CNS Mariette X, Ravaud P, Steinfeld S, et al Inefficacy of infliximab in pri-
involvement. Semin Arthritis Rheum. 1999;29:1–3 mary Sjögren’s syndrome: results of the randomized, controlled
Ioannidis JP, Vassiliou VA, Moutsopoulos HM. Long-term risk of mor- trial of remicade in primary Sjögren’s syndrome (TRIPSjS). Arthritis
tality and lymphoproliferative disease and predictive classification Rheum. 2004;50:1270–6
of primary Sjögren’s syndrome. Arthritis Rheum. 2002;46:741–7 Mariette X, Roux S, Zhang J, et al The level of BlyS (BAFF) correlates
Jacobsson LT, Axell TE, Hansen BU, et al Dry eyes or mouth: an epide- with the titer of autoantibodies in human Sjögren’s syndrome. Ann
miological study in Swedish adults, with special reference to pri- Rheum Dis. 2003;62:168–71
mary Sjögren’s syndrome. J Autoimmun. 1989;2:521–7 Mavragani CP, Moutsopoulos NM, Moutsopoulos HM. The management
Jara LJ, Navarro C, Brito-Zeron P, et al Thyroid disease in Sjögren’s of Sjögren’s syndrome. Nat Clin Pract Rheumatol. 2006;2: 252–61
syndrome. Clin Rheumatol. 2007;26:1601–6 Mellgren SI, Goransson LG, Omdal R. Primary Sjögren’s syndrome
Jungehulsing M, Fischbach R, Schroder U, et al Magnetic resonance associated neuropathy. Can J Neurol Sci. 2007;34:280–7
sialography. Otolaryngol Head Neck Surg. 1999;121:488–94 Mikulicz J. Über eine eigenartige symmetrische Erkrankung der Tranen
Kamisawa T, Okamoto A. Autoimmune pancreatitis: proposal of IgG4- und Mundspeicheldrusen. Stuttgart: Beitr Chir Fortsch Gewidmet
related sclerosing disease. J Gastroenterol. 2006;41:613–25 Theodor Billroth. 1892;610–30
Kassan SS, Thomas TL, Moutsopoulos HM, Hoover R, Kimberly RP, Mitsias D, Tzioufas A, Veiopoulou C, et al The Th1/Th2 cytokine bal-
Budman DR, Costa J, Decker JL, Chused TM. Increased risk of ance changes with the progress of the immunopathological lesion of
lymphoma in sicca syndrome. Ann Intern Med. 1978;89:888–92 Sjögren’s syndrome. Clin Exp Immunol. 2002;128:562–8
Kohler PF, Winter ME. A quantitative test for xerostomia. The Saxon Mori K, Iijima M, Sugiura M et al Sjögren's syndrome associated pain-
test, an oral equivalent of the Schirmer test. Arthritis Rheum 1985; ful sensory neuropathy without sensory ataxia. J Neurol Neurosurg
28:1128–32 Psychiatry 2003; 74(9):1320–2
Kohn WG, Ship JA, Atkinson JC, et al Salivary gland 99mTc-scintigra- Mori K, Iijima M, Koike H, et al The wide spectrum of clinical mani-
phy: a grading scale and correlation with major salivary gland flow festations in Sjögren’s syndrome-associated neuropathy. Brain.
rates. J Oral Pathol Med. 1992;21:70–4 2005;128:2518–34
Krieg AM, Vollmer J. Toll-like receptors 7, 8, and 9: linking innate Moutsopoulos HM, Balow JE, Lawley TJ, et al Immune complex glom-
immunity to autoimmunity. Immunol Rev. 2007;220:251–69 erulonephritis in sicca syndrome. Am J Med. 1978;64:955–60
Kruize AA, Hené RJ, Kallenberg CG, et al Hydroxychloroquine treat- Moutsopoulos HM, Elkon KB, Mavridis AK, et al Serum C-reactive
ment for primary Sjögren’s syndrome: a two year, double-blind protein in primary Sjögren’s syndrome. Clin Exp Rheumatol.
crossover trial. Ann Rheum Dis. 1993;52:360–4 1983a;1:57–8
Kruize AA, Hene RJ, van der Heide A, Bodeutsch C, deWilde PCM, Moutsopoulos HM, Webber BL, Vlagopoulos TP, et al Differences in
vanBijsterveld OP, et al Long-term followup of patients with the clinical manifestations of sicca syndrome in the presence and
Sjögren’s syndrome. Arthritis Rheum. 1996;297–303 absence of rheumatoid arthritis. Am J Med. 1979;66:733–6
Lamour A, Le Corre R, Pennec YL, Cartron J, Youinou P. Heterogeneity Moutsopoulos HM, Steinberg AD, Fauci AS, et al High incidence of
of neutrophil antibodies in patients with primary Sjögren’s syn- free monoclonal lambda light chains in the sera of patients with
drome. Blood. 1995;86:3553–9 Sjögren’s syndrome. J Immunol. 1983b;130:2663–5
Larsson X, Bredberg A, Henriksson G, Manthorpe R, Sallmyr A. Moutsopoulos HM, Costello R, Drosos AA, et al Demonstration and
Immunohistochemistry of the B cell component in lower lip salivary identification of monoclonal proteins in the urine of patients with
glands of Sjögren’s syndrome and healthy subjects. Scand J Immunol. Sjögren’s syndrome. Ann Rheum Dis. 1985;44:109–12
2005;61:98–107 Nagler RM. Salivary glands and the aging process: mechanistic aspects,
Lekhanont K, Leyngold IM, Suwan-Apichon O, et al Comparison of topi- health-status and medicinal-efficacy monitoring. Biogerontology.
cal dry eye medications for the treatment of keratoconjunctivitis sicca 2004;5:223–33
in a botulinum toxin B-induced mouse model. Cornea. 2007;26: 84–9 Nakamura H, Kawakami A, Tominaga M, et al Relationship between
Leppilahti M, Tammela TL, Huhtala H, et al Interstitial cystitis-like uri- Sjögren’s syndrome and human T-lymphotropic virus type I infec-
nary symptoms among patients with Sjögren’s syndrome: a popula- tion: follow-up study of 83 patients. J Lab Clin Med. 2000;135:
tion-based study in Finland. Am J Med. 2003;115:62–5 139–44
Liang L, Zhang M, Zou W, Liu Z. Aggravated dry eye after laser in situ Nardi N, Brito-Zerón P, Ramos-Casals M, et al Circulating auto-anti-
keratomileusis in patients with Sjögren syndrome. Cornea. 2008;27: bodies against nuclear and non-nuclear antigens in primary Sjögren’s
120–3 syndrome: prevalence and clinical significance in 335 patients. Clin
Makula E, Pokorny G, Kiss M, et al The place of magnetic resonance Rheumatol. 2006;25(3):341–6
and ultrasonographic examinations of the parotid gland in the diag- Papiris SA, Maniati M, Constantopoulos SH, et al Lung involvement in
nosis and follow-up of primary Sjögren’s syndrome. Rheumatology primary Sjögren’s syndrome is mainly related to the small airway
(Oxford). 2000;39:97–104 disease. Ann Rheum Dis. 1999;58:61–4
Maltsman-Tseikhin A, Moricca P, Niv D. Burning mouth syndrome: Paschides CA, Kitsios G, Karakostas KX, et al Evaluation of tear break-up
will better understanding yield better management? Pain Pract. time, Schirmer’s-I test and rose bengal staining as confirmatory tests
2007; 7:151–62 for keratoconjunctivitis sicca. Clin Exp Rheumatol. 1989;7:155–7
12 Sjögren’s Syndrome 129

Patton LL, Siegel MA, Benoliel R, De Laat A. Management of burning Skopouli FN, Dafni U, Ioannidis JP, Moutsopoulos HM. Clinical evolu-
mouth syndrome: systematic review and management recommenda- tion, and morbidity and mortality of primary Sjögren’s syndrome.
tions. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2007;103 Semin Arthritis Rheum. 2000;29:296–304
Suppl:S39.e1–13 Skopouli FN, Papanikolaou S, Malamou-Mitsi V, et al Obstetric and
Pavlidis NA, Karsh J, Moutsopoulos HM. The clinical picture of primary gynaecological profile in patients with primary Sjögren’s syndrome.
Sjögren’s syndrome: a retrospective study. J Rheumatol. 1982;9: Ann Rheum Dis. 1994b;53:569–73
685–90 Skopouli FN, Talal A, Galanopoulou V, et al Raynaud’s phenomenon in
Pertovaara M, Pukkala E, Laippala P, et al A longitudinal cohort study primary Sjögren’s syndrome. J Rheumatol. 1990;17:618–20
of Finnish patients with primary Sjögren’s syndrome: clinical, Stevens WJ, Swartele FE, Empsten FA, De Clerck LS. Use of the Saxon
immunological, and epidemiological aspects. Ann Rheum Dis. 2001; test as a measure of saliva production in a reference population of
60:467–72 schoolchildren. Am J Dis Child. 1990;144:570–1
Pflugfelder SC, Maskin SL, Anderson B, et al A randomized, double- Stojanovich L, Milovanovich B, de Luka SR, et al Cardiovascular auto-
masked, placebo-controlled, multicenter comparison of loteprednol nomic dysfunction in systemic lupus, rheumatoid arthritis, primary
etabonate ophthalmic suspension, 0.5%, and placebo for treatment Sjögren syndrome and other autoimmune diseases. Lupus. 2007;16:
of keratoconjunctivitis sicca in patients with delayed tear clearance. 181–5
Am J Ophthalmol. 2004;138:444–57 Takeda S, Haratake J, Kasai T, et al IgG4-associated idiopathic tubu-
Pillemer SR, Smith J, Fox PC, Bowman SJ. Outcome measures for lointerstitial nephritis complicating autoimmune pancreatitis.
Sjögren’s syndrome. J Rheumatol. 2005;32:143–9 Nephrol Dial Transplant. 2004;19:474–6
Prochorec-Sobieszek M, Wagner T, Loukas M, Chwali ska-Sadowska Taouli B, Brauner MW, Mourey I, et al Thin-section chest CT findings
H, Olesi ska M. Histopathological and immunohistochemical analy- of primary Sjögren’s syndrome: correlation with pulmonary func-
sis of lymphoid follicles in labial salivary glands in primary and sec- tion. Eur Radiol. 2002;12:1504–11
ondary Sjögren’s syndrome. Med Sci Monit. 2004;10: BR115–21 Terrier B, Lacroix C, Guillevin L, et al Diagnostic and prognostic rele-
Ramos-Casals M, Anaya JM, García-Carrasco M, et al Cutaneous vascu- vance of neuromuscular biopsy in primary Sjögren’s syndrome-
litis in primary Sjögren syndrome: classification and clinical signifi- related neuropathy. Arthritis Rheum. 2007;57(8):1520–9
cance of 52 patients. Medicine (Baltimore). 2004a;83(2): 96–106 Theander E, Manthorpe R, Jacobsson L. Mortality and causes of death
Ramos-Casals M, Font J, Garcia-Carrasco M, et al Primary Sjögren in primary Sjögren’s syndrome: a prospective cohort study. Arthritis
syndrome: hematologic patterns of disease expression. Medicine Rheum. 2004;50:1262–9
(Baltimore). 2002;81(4):281–92 Tishler M, Yaron I, Shirazi I, Yaron M. Hydroxychloroquine treatment
Ramos-Casals M, García-Carrasco M, Cervera R, et al Thyroid disease for primary Sjögren’s syndrome: its effect on salivary and serum
in primary Sjögren syndrome: study in a series of 160 patients. inflammatory markers. Ann Rheum Dis. 1999;58:253–6
Medicine (Baltimore). 2000;79(2):103–8 Trontzas PI, Andrianakos AA. Sjögren’s syndrome: a population based
Ramos-Casals M, la Civita L, De Vita S, et al For the SjS-HCV Study study of prevalence in Greece. Ann Rheum Dis. 2005;64:1240–1
Group. Characterization of B cell lymphoma in patients with Tsampoulas CG, Skopouli FN, Sartoris DJ, et al Hand radiographic
Sjögren’s syndrome and hepatitis C virus infection. Arthritis Rheum. changes in patients with primary and secondary Sjögren’s syn-
2007;57(1):161–70 drome. Scand J Rheumatol. 1986;15:333–9
Ramos-Casals M, Loustaud-Ratti V, De Vita S, et al For the Sjögren Tsianos EB, Chiras CD, Drosos AA, Moutsopoulos HM. Oesophageal
syndrome associated with hepatitis C virus: a multicenter analysis dysfunction in patients with primary Sjögren’s syndrome. Ann
of 137 cases. Medicine (Baltimore). 2005;84(2):81–9 Rheum Dis. 1985;44:610–3
Ramos-Casals M, Nardi N, Brito-Zerón P, et al Atypical autoantibodies Tsianos EB, Tzioufas AG, Kita MD, et al Serum isoamylases in patients
in patients with primary Sjögren syndrome: clinical characteristics with autoimmune rheumatic diseases. Clin Exp Rheumatol. 1984;2:
and follow-up of 82 cases. Semin Arthritis Rheum. 2006a; 35(5): 235–8
312–21 Tzioufas AG, Boumba DS, Skopouli FN, Moutsopoulos HM. Mixed mono-
Ramos-Casals M, Sánchez-Tapias JM, Parés A, et al Characterization and clonal cryoglobulinemia and monoclonal rheumatoid factor cross-reac-
differentiation of autoimmune versus viral liver involvement in patients tive idiotypes as predictive factors for the development of lymphoma in
with Sjögren’s syndrome. J Rheumatol. 2006b;33(8): 1593–9 primary Sjögren’s syndrome. Arthritis Rheum. 1996;39:767–72
Ramos-Casals M, Solans R, Rosas J, et al Primary Sjögren syndrome in Vitali C, Bombardieri S, Moutsopoulos HM, Balestrieri G, Bencivelli W,
Spain: clinical and immunologic expression of 1010 patients. Bernstein RM, et al Preliminary criteria for the classification of Sjögren’s
Medicine (Baltimore). 2008;87(4):210–9 syndrome. Results of a prospective concerted action supported by the
Reinertsen JL, Schaefer EJ, Brewer HB, Moutsopoulos HM. Sicca-like European Community. Arthritis Rheum. 1993;36: 340–7
syndrome in type V hyperlipoproteinemia. Arthritis Rheum. 1980;23: Vitali C, Bombardieri S, Moutsopoulos HM, Coll J, Gerli R, Hatron Y,
114–8 et al Assessment of the European classification criteria for Sjogren’s
Rhodus NL, Michalowicz BS. Periodontal status and sulcular Candida syndrome in a series of clinically defined cases: results of a prospec-
albicans colonization in patients with primary Sjögren’s syndrome. tive multicentre study. Ann Rheum Dis. 1996;55:116–21
Quintessence Int. 2005;36(3):228–33 Vitali C, Bombardieri Jonsson R, Moutsopoulos HM, Alexander EL,
Salliot C, Gottenberg JE, Bengoufa D, et al Anticentromere antibodies Carsons SE, et al Classification criteria for Sjögren’s syndrome: a
identify patients with Sjögren’s syndrome and autoimmune overlap revised version of the European criteria proposed by the American-
syndrome. J Rheumatol. 2007;34(11):2253–8 European consensus group. Ann Rheum Dis. 2002;61:544–58
Sankar V, Brennan MT, Kok MR, et al Etanercept in Sjögren’s syn- Vivino FB, Gala I, Hermann GA. Change in final diagnosis on second
drome: a twelve-week randomized, double-blind, placebo-controlled evaluation of labial minor salivary gland biopsies. J Rheumatol.
pilot clinical trial. Arthritis Rheum. 2004;50:2240–5 2002;29:938–44
Simon BG, Moutsopoulos HM. Primary amyloidosis resembling sicca Voulgarelis M, Dafni UG, Isenberg DA, Mousopoulos HM. Malignant
syndrome. Arthritis Rheum. 1979;22:932–4 lymphoma in primary Sjögren’s syndrome. A multicenter, retro-
Sjögren’s Syndrome Foundation. And the survey says. The Moisture spective, clinical study by the European concerted action on
Seekers. 2006;24:1–3 Sjögren’s syndrome. Arthritis Rheum. 1999;42:1765–72
Skopouli FN, Barbatis C, Moutsopoulos HM. Liver involvement in pri- Voulgarelis M, Giannouli S, Tzioufas AG, Moutsopoulos HM. Long
mary Sjögren’s syndrome. Br J Rheumatol. 1994a;33:745–8 term remission of Sjögren’s syndrome associated aggressive B cell
130 M. Ramos-Casals et al.

non-Hodgkin’s lymphomas following combined B cell depletion Wu AJ, Fox PC. Sjögren’s syndrome. Semin Dermatol. 1994;13:138–43
therapy and CHOP (cyclophosphamide, doxorubicin, vincristine, Yamamoto M, Takahashi H, Sugai S, Imai K. Clinical and pathological
prednisone). Ann Rheum Dis. 2006;65:1033–7 characteristics of Mikulicz’s disease (IgG4-related plasmacytic exo-
Walker J, Gordon T, Lester S, et al Increased severity of lower urinary crinopathy). Autoimmun Rev. 2005;4:195–200
tract symptoms and daytime somnolence in primary Sjögren’s syn- Zandbelt MM, de Wilde P, van Damme P, et al Etanercept in the treat-
drome. J Rheumatol. 2003;30:2406–12 ment of patients with primary Sjögren’s syndrome: a pilot study.
Whitcher J, Gritz D, Daniels T. The dry eye: a diagnostic dilemma. Int J Rheumatol. 2004;31:96–101
Ophthal Clin. 1998;38:23–37 Zintzaras E, Voulgarelis M, Moutsopoulos H. The risk of lymphoma
Wolkoff P, Nojgaard JK, Franck C, Skov P. The modern office environ- development in autoimmune diseases. Arch Intern Med. 2005;
ment desiccates the eyes? Indoor Air. 2006;16:258–65 165:2337–44
Wright SA, Convery RP, Liggett N. Pulmonary involvement in Sjögren’s Klippel JH, Stone JS, Crofford L, White P. Primer on the Rheumatic
syndrome. Rheumatology (Oxford). 2003;42:697–8 Diseases. 13th Edition. 2008. Springer. London
 Systemic Lupus Erythematosus
13
Michelle Petri, Rachel Abuav, Dimitrios Boumpas, Fiona Goldblatt,
David Isenberg, Grant J. Anhalt, Victoria P. Werth, and R. John Looney

13.1 Overview of Systemic Lupus 13.2 Diagnosis

Erythematosus
Myth: A patient must have at least 4 of the 11 American
College of Rheumatology (ACR) classification criteria to be
› Systemic lupus erythematosus (SLE) is the prototypi- diagnosed with SLE.
cal autoimmune disease, associated with autoantibody
production and evidence for immune complex deposi- Reality: The ACR classification criteria were devised to clas-
tion among the many aspects of its pathophysiology. sify patients as having SLE for the purpose of research stud-
› SLE occurs most commonly in women during their ies, not for clinical diagnosis (Table 13.1) (Hochberg 1997;
reproductive years. The disease is approximately nine Tan et al. 1982). These criteria are relatively easy to apply
times more common among women than men. and help underscore the multisystem nature of SLE. Thus, it
› Virtually any organ can be affected by SLE. However, is understandable that they are frequently referred to and
constitutional symptoms, mucocutaneous features, mus- even used as “diagnostic” criteria.
culoskeletal involvement, and renal and central nervous However, because the criteria are historical and cumula-
system disease are the most common manifestations. tive (i.e., a malar rash counts even if the patient had it last
› Many aspects of SLE appear to be triggered by classic year but does not have it now), time is often required before
inflammatory mechanisms, accompanied by elevated a patient accrues four criteria. Some patients have only two
acute phase reactants, and responsive to immunosuppres- criteria (e.g., a positive ANA and lupus nephritis demon-
sive therapies. Other aspects, such as those associated strated by renal biopsy), yet have unequivocal SLE. When it
with antiphospholipid antibodies (aPL), are associated comes to making the diagnosis of SLE, clinical judgment
with hypercoagulability. Although inflammation contrib- trumps the ACR classification criteria.
utes to these aspects as well, the approach to the treat- Pearl: “ANA-negative lupus” still exists.
ment of aPL-mediated disease manifestations is usually
Comment: The sensitivity of ANA assays is suboptimal, par-
centered on anticoagulation.
ticularly early in the disease course. Depending on which
› Autoantibodies can occur in the absence of clinical
type of assay is used and the quality of the laboratory in
features of SLE, but strong evidence implicates certain
which it is performed, the sensitivity of ANA testing can be
autoantibodies in the mediation of tissue damage in the
less than 70% in patients who have SLE. This is particularly
kidney and other organs.
true in laboratories that employ ELISA or bead assays rather
› Genetics plays a significant role in SLE, but the disease
than immunofluorescence tests. Thus, a significant number
is clearly polygenic, with multiple genetic risk factors
of SLE cases will be missed if the diagnosis is discarded
contributing incrementally to the overall genetic risk.
because the ANA assay is negative.
› Environmental risk factors are also critical. Established
In patients who have features that are typical of lupus but
environmental risk factors for the disease include hormones,
no history of a positive ANA assay, the ANA must be repeated
ultraviolet light, the exposure to certain medications,
in an experienced laboratory that employs an indirect immu-
and possibly dietary exposures and infectious agents.
nofluorescence assay, using Hep2 cells as the test substrate.
In addition, assays for other autoantibodies such as those
directed against Ro, La, Sm, RNP, and cardiolipin are impor-
tant, and the measurement of complement levels (C3, C4,
Issues related to pregnancy in patients with SLE are addressed in
Chapter 17. The majority of Pearls and Myths related to the antiphos- CH50) can also be informative. Biopsies of skin and kidney
pholipid syndrome are covered in Chapter 16. lesions are required in selected cases.

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 131

DOI:10.1007/978-1-84800-934-9_13, © Springer Science + Business Media B.V. 2009
132 M. Petri et al.

Table 13.1 The 1997 update of the 1982 American College of Rheumatology revised criteria for the classification of systemic lupus erythematosusa
Criterion Definition
1. Malar Rash Fixed erythema, flat or raised, over the malar eminences, tending to spare the nasolabial folds
2. Discoid rash Erythematous raised patches with adherent keratotic scaling and follicular plugging; atrophic scarring
may occur in older lesions
3. Photosensitivity Skin rash as a result of unusual reaction to sunlight, by patient history or physician observation
4. Oral ulcers Oral or nasopharyngeal ulceration, usually painless, observed by physician
5. Nonerosive Arthritis Involving 2 or more peripheral joints, characterized by tenderness, swelling, or effusion
6. Pleuritis or Pericarditis (a) Pleuritis – convincing history of pleuritic pain or rubbing heard by a physician or evidence of pleural
effusion
OR
(b) Pericarditis – documented by electrocardigram or rub or evidence of pericardial effusion
7. Renal Disorder (a) Persistent proteinuria > 0.5 g/day or > than 3+ if quantitation not performed
OR
(b) Cellular casts – may be red cell, hemoglobin, granular, tubular, or mixed
8. Neurologic Disorder (a) Seizures – in the absence of offending drugs or known metabolic derangements; e.g., uremia,
ketoacidosis, or electrolyte imbalance
OR
(b) Psychosis – in the absence of offending drugs or known metabolic derangements, e.g., uremia,
ketoacidosis, or electrolyte imbalance
9. Hematologic Disorder (a) Hemolytic anemia – with reticulocytosis
OR
(b) Leukopenia – < 4,000/mm3 on ³ 2 occasions
OR
(c) Lymphopenia – < 1,500/ mm3 on ³ 2 occasions
OR
(d) Thrombocytopenia – <100,000/ mm3 in the absence of offending drugs
10. Immunologic Disorder (a) AntiDNA: antibody to native DNA in abnormal titer
OR
(b) AntiSm: presence of antibody to Sm nuclear antigen
OR
(c) Positive finding of antiphospholipid antibodies on:
an abnormal serum level of IgG or IgM anticardiolipin antibodies,
a positive test result for lupus anticoagulant using a standard method, or
a false-positive test result for at least 6 months confirmed by Treponema pallidum immobilization or
fluorescent treponemal antibody absorption test
11. Positive Antinuclear Antibody An abnormal titer of antinuclear antibody by immunofluorescence or an equivalent assay at any point in
time and in the absence of drugs
a
American College of Rheumatology webpage: http://www.rheumatology.org/publications/classification/SLE/1997UpdateOf1982RevisedCriteri
aClassificationSLE.asp?aud=mem. Accessed May 26, 2009

Although the suboptimal sensitivity of ANA assays has looking (nonlesional) skin. These lesions, when present, can
come into clearer focus with the move of laboratories away be visualized by direct immunofluorescence studies of skin
from Hep2 cell-based immunofluorescence studies, the poor biopsies, and their detection was taken as presumptive evi-
specificity for ANA assays has been known for a long time. dence of SLE (Provost et al. 1980).
As an example, a young woman with a positive ANA and The term “positive lupus band test” has been applied inap-
arthralgias is more likely to have autoimmune thyroiditis propriately to the finding of immunoreactants in the basement
than lupus. membrane zone of lesional skin in SLE. The true concept of
the lupus band test refers to findings in nonlesional skin.
Myth: The “lupus band test” is useful in diagnosis of prob-
Patients who have such deposits in nonlesional skin have
lematic cases with possible SLE.
a high frequency of antibodies against native DNA, depressed
Reality: There was once a great deal of interest in using the complement levels, nephritis, and poor survival rates (Gilliam
so-called “lupus band test” to aid in the diagnosis of prob- et al. 1974). Thus, the lupus band test is really just a surro-
lematic cases with possible SLE. This test was based on the gate marker for patients who have circulating native DNA-
observation that some patients with SLE have deposition of containing immune complexes. Most SLE experts consider
immunoreactants in the basement membranes on normal- the lupus band test to be obsolete.
13 Systemic Lupus Erythematosus 133

predictive of eventual evolution into SLE were young age,

alopecia, serositis, discoid lupus, a positive Coombs test, and
the presence of antiSm and antiDNA antibodies.
Latent or incomplete lupus describes patients who present
with a constellation of symptoms suggestive of SLE, but do
not qualify by clinical intuition or classification criteria as
having classical SLE (Ganczarczyk et al. 1989; Graninger
and Smolen 2001). These patients usually present with one
or two of the ACR criteria and other features that are not
included in the criteria. Most of these patients do not develop
SLE ultimately or, if they do, their disease usually remains
mild (Calvo-Alen et al. 1996).
Pearl: Not all lupus is diagnosed in patients before the age
of 50.

Fig. 13.1 An “in vivo ANA,” demonstrated in a skin biopsy. This Comment: This myth is perpetuated by clinicians who are
patient had a high-titer serum ANA associated with limited sclero- reluctant to consider the diagnosis of SLE in older patients.
derma. However, the finding of an in vivo ANA is an occasional artifact SLE can present in older patients: up to 20% of patients with
of performing direct immunofluorescence on a skin biopsy. An in vivo SLE develop their disease after the age of 50 (Font et al.
ANA does not always correlate with clinically relevant serum antibod-
ies and a clinical diagnosis of rheumatic disease, but the phenomenon is 1991; Boddaert et al. 2004). The interval between symptom
more likely to occur if the serum ANA is strongly positive (Figure cour- onset and diagnosis is often longer for older patients, because
tesy of Dr. Grant Anhalt) the clinical manifestations for patients in this subgroup can
be insidious and atypical (Font et al. 1991).
The overwhelming female predominance that is typical of
Myth: The presence of an in vivo ANA on a skin biopsy is the peak incidence years is less evident among patients diag-
equivalent to finding a positive serum ANA. nosed later in life. In addition, older patients are less likely
Reality: Dermatopathologists sometimes report the finding of than are younger ones to have malar rashes, other cutaneous
an “in vivo ANA” on skin biopsies submitted for direct immu- manifestations, clinically significant renal disease, and hypo-
nofluorescence (Fig. 13.1). Is this finding equivalent to a posi- complementemia (Font et al. 1991; Domenech et al. 1992;
tive serum assay for ANA? (Wells et al. 1979). Not necessarily. Cervera et al. 1993). Serositis and sicca features are typical
Here is why…. of late-onset SLE (Wilson et al. 1981; Costallat and Coimbra
Tissue processed for immunofluorescence is not fixed. 1994; Ho et al. 1998; Formiga et al. 1999). The 5- and 10-year
Thus, immunoglobulins are present within the interstitial survival rates are lower in patients with late-onset disease,
spaces of the specimen. As the specimen is sectioned and but mortality in the late-onset subgroup might reflect only
incubated with antibody probes, ANA can diffuse into the the consequences of aging rather than disease-related differ-
open nuclei and bind to antigens within. This process is ences (Boddaert et al. 2004).
sometimes efficient enough that one observes a positive in
vivo ANA even in the absence of a significant serum ANA
titer. Thus, the finding of an incidental in vivo ANA on skin
biopsy does not necessarily unmask an occult connective tis-
13.3 Clinical Features
sue disorder. As with any test, laboratory data must be cor-
related with the clinical context.
Myth: The major cause of death in SLE patients is: (a) active
Pearl: Undifferentiated connective tissue diseases and “incom- lupus; (b) complications of renal failure; or c) infection.
plete lupus” represent up to 10–20% of patients referred to
Reality: The correct answer: none of the above! In the 1970s,
rheumatologists.
Urowitz and Gladman reported on the bimodal pattern of
Comment: The recognition that systemic rheumatic diseases mortality in SLE. Whereas “early” deaths were caused pre-
have several common features has led to the concept of the dominantly by active SLE, “later” deaths were linked pri-
undifferentiated connective tissue syndromes. These patients marily to cardiovascular disease (Urowitz et al. 1976). Thus,
account for 10–20% of patients referred to tertiary care cen- although SLE patients certainly die of complications related
ters. Among patients presenting with symptoms suggestive directly to a, b, and c, the most common cause of death in
of a connective tissue disease, only a small fraction (10–15%) many studies is accelerated atherosclerosis. This is even truer
fulfill classification criteria for SLE 5 years later. Factors today than it was 3 decades ago.
134 M. Petri et al.

The risk of myocardial infarction in SLE women aged Reality: Lymphadenopathy occurs in approximately 40%
35–44 years is 50-fold higher when compared with age- and of lupus patients. This finding is found typically at disease pre-
sex-matched controls (Manzi et al. 1997). This excess risk is sentation and during disease flares. SLE patients with lymph-
not explained by traditional cardiovascular risk factors adenopathy are more likely to have constitutional symptoms
(Esdaile et al. 2001). Case-control studies have shown a two- such as fever and malaise. Lymphadenopathy in SLE is char-
fold increase in subclinical atherosclerosis, as measured by acterized by soft, discrete, slightly tender lymph nodes. The
carotid duplex (Roman et al. 2003). involved lymph node chains are found usually in the cervical,
Myth: SLE causes pain. axillary, and inguinal areas. Biopsies reveal areas of follicular
hyperplasia and necrosis.
Reality: The most frequent cause of a positive ANA and Features that suggest a pathological cause other than SLE
chronic pain is fibromyalgia! The failure to recognize that include the absence of tenderness in the nodes; enlargement
pain is NOT because of active SLE leads to much unnecessary greater than three centimeters; induration; age older than 40,
glucocorticoid use and even the inappropriate use of narcot- and a location above the clavicle (Vassilakopoulos and Pangalis
ics. Unfortunately, SLE patients frequently have concomitant 2000). Additional indicators of concern are splenomegaly and
fibromyalgia (Akkasilpa et al. 2005; Middleton et al. 1994; a monoclonal expansion of CD19+/CD22+ B lymphocytes in
Taylor et al. 2000). Estimates of the prevalence of fibromyal- the peripheral blood (Table 13.2).
gia among patients with SLE range from 10 to 30%. Lupus patients have a slightly increased risk for both
Arthralgias are the principal musculoskeletal feature of SLE, Hodgkin’s and non-Hodgkin’s lymphoma. The standardized
but synovitis also occurs. Patients describe their joint symptoms incidence ratios for these two disorders are 3.6 and 3.1,
as “stiffness” and note an accentuation of the problem in the respectively, for patients with SLE. These risks persist even
morning (the “gel phenomenon”). The PIPs, MCPs, wrists, and when the use of alkylating agents is taken into account.
knees are involved in typical cases. SLE does not affect the
neck, shoulder girdle, back, or trochanteric region. Pain in those Pearl: Up to one-third of lupus patients have one or more
areas in an SLE patient usually reflects fibromyalgia. additional autoimmune diseases.
SLE patients also have pain from musculoskeletal damage Comment: The association between SLE and other autoim-
(e.g., osteonecrosis) or from processes unrelated to SLE such mune diseases is well documented. Other autoimmune disor-
as osteoarthritis, mechanical low back pain, bursitis, com- ders often linked to SLE are Sjögren’s syndrome, autoimmune
pression neuropathy, and a host of other conditions to which thrombocytopenia, hypothyroidism, and polymyositis (Foote
humans are prone. But beware the “lupus” patient who hurts et al. 1982; Karpatkin 1985; Pyne and Isenberg 2002;
all over: the correct diagnosis is probably fibromyalgia. Manoussakis et al. 2004). Approximately 30% of SLE patients
Myth: Fatigue is a sign of active SLE. manifest other autoimmune disorders; a smaller percentage
develops more than one such condition (Table 13.3) (McDonagh
Reality: Fatigue is a major cause of the reduced quality of life
and Isenberg 2000; Chambers et al. 2007).
reported by SLE patients and a significant contributor to
Thus, the diagnosis of lupus should not be regarded as the
disability. Acute fatigue often accompanies SLE flares, but
end of the story. Vigilance is essential to detect the one-third
this usually resolves along with the other features of active
of these patients who eventually develop additional autoim-
disease. However, for many patients with lupus, fatigue is a
mune disease manifestations.
chronic symptom that is unresponsive to NSAIDs, glucocorti-
coids, and immunosuppressive medications. In these SLE Myth: Autoimmune liver disease is common in SLE.
patients, chronic fatigue is probably due to fibromyalgia. Thus,
Reality: Autoimmune liver disease is relatively uncom-
treating all chronic fatigue with glucocorticoids exposes many
mon in SLE. It affects only about 2% of patients. Autoimmune
SLE patients to unnecessary toxicity.
Every SLE patient with chronic fatigue deserves an evalu-
ation for comorbid processes, including anemia, hypothy- Table 13.2 When is lymphadenopathy in an SLE patient a cause for
roidism, adrenal insufficiency, sleep disorders, fibromyalgia, concern? lymphadenopathy out of proportion to SLE activity
and depression. However, if no other cause is identified, then Patient age > 40 years
sensible advice is a brief daily nap. This nap should be suffi- Hardness of the affected nodes to palpation
ciently short that it does not interfere with the normal sleep– An absence of tenderness but the presence of hardness
wake cycle. SLE patients with fatigue should also exercise Enlargement of any node > 3 cm in diameter
Supraclavicular location
daily (as should patients with fibromyalgia) (see Chapter 33,
Splenomegaly
Fibromyalgia) (Iaboni et al. 2006; Omdal et al. 2002). Monoclonal expansion of CD19+/CD22+ B lymphocytes in the
Myth: In a lupus patient, lymphadenopathy must be investi- peripheral blood
gated aggressively to rule out lymphoma. Modified from (Vassilakopoulos 2000)
13 Systemic Lupus Erythematosus 135

Table 13.3 Prevalence of autoimmune diseases in patients with SLE compared to the UK population (From Chambers et al. 2007. Reprinted with
permission from BMJ Publishing Group)
Autoimmune disease Prevalence in UK (%) Frequency in UCLH SLE cohort (total cohort
n = 401), prevalence (%)
Sjögren’s syndrome 3.3 45 (11.22)
Hashimoto’s hypothyroidism 0.80 26 (6.48)
Antiphosholipid syndrome No data 23 (5.7)
Myositis 0.002–0.01 14 (3.49)
Rheumatoid arthritis 0.44–1.16 14 (3.49)
ITP No data 13 (3.24)
Graves disease/hyperthyroidism 0.65 8 (1.99)
Autoimmune hepatitis 0.014 5 (1.25)
AHA No data 4 (0.99)
Fibrosing alveolitis No data 3 (0.75)
Type 1 Diabetes Mellitus 0.34 3 (0.75)
Pernicious anaemia 0.13 3 (0.75)
Myasthenia gravis 0.015 3 (0.75)
Vitiligo 0.38 2 (0.50)
Coeliac disease 0.820 1 (0.25)
ITP Idiopathic thrombocytopaenia purpura; AHA Autoimmune haemolytic anaemia

Table 13.4 Autoimmune versus lupus-associated hepatitis

Feature Autoimmune hepatitis Lupus-associated hepatitis
Autoantibodies Type I (classic): ANA, ASMA ASMA, AMA only in 30% of the cases and
usually in low titers; antiribosomal P
Type II: ALKM-1, ALC-1
Serum chemistry (AST, ALT, LDH) Variable elevations in serum liver biochemistry Mild elevation (usually x3–4 normal)
Histology Variable rates of periportal hepatitis with Rarely seen
piece-meal necrosis
*
ANA antinuclear antibodies, ASMA antismooth muscle antibodies, AMA antimitochondrial antibodies, ALKM-1 antibodies to liver kidney
microsomes, ALC-1 antibodies to liver cytosolic antigen

hepatitis, a separate disorder once known as “lupoid hepatitis” in a patient with SLE or possible SLE, histopathology may
(Mackay et al. 1959), also has a predilection for young resolve the question. Autoimmune hepatitis is associated with
women. ALKM-1 (type 2) autoimmune hepatitis is generally a characteristic periportal hepatitis and piecemeal necrosis.
a disease of girls and young women. This disorder is associ- Liver dysfunction can also exist in SLE as a result of issues
ated with antibodies to liver/kidney microsomes (ALKM-1) that are not immunological in nature (Runyon et al. 1980). A
or to a liver cytosol antigen (ALC-1). wide variety of pathological lesions may be observed. Hepatic
Autoantibodies help to distinguish between autoimmune steatosis, a common finding, may occur as part of the disease
hepatitis and liver disease associated with lupus (Irving et al. process, as a complication of glucocorticoid use, or intercur-
2007). Positive ANA assays are the norm in both conditions. rent factors such as alcohol use. Abnormalities of hepatic ami-
However, antismooth muscle and antimitochondrial antibod- notransferases, lactate dehydrogenase, and alkaline phosphatase
ies occur in fewer than 30% of SLE patients and, when pres- can also occur as a complication of NSAID or salicylate use.
ent, are usually found in low titers (Table 13.4). The absence
Pearl: Primary biliary cirrhosis rarely occurs in SLE.
of these antibodies suggests that any apparent liver inflam-
mation might be due to SLE itself rather than the distinct Comment: Primary biliary cirrhosis, an autoimmune disease
condition of autoimmune hepatitis. of the liver that predominantly affects women over the age of
Elevations of the serum hepatic aminotransferases tend to 20, is unlikely to complicate SLE. In a female patient whose
be lower in lupus-associated hepatitis when compared with serum tests reflect a cholestatic pattern of liver injury, e.g., an
autoimmune liver disease. The alanine and aspartate amin- unexplained elevation in alkaline phosphatase but normal
otransferases are elevated generally to only three or four times bile ducts on ultrasound, a positive ANA, and a positive
the upper limit of normal in lupus-associated liver inflamma- antimitochondrial antibody (AMA); Sjögren’s syndrome and
tion. Much higher elevations occur in autoimmune liver dis- systemic sclerosis are both more likely concurrent disorders
ease. When doubt persists about the cause of liver dysfunction than is SLE.
136 M. Petri et al.

Myth: The myositis of SLE resembles that of idiopathic

inflammatory myopathies.
Reality: In lupus, generalized myalgia and muscle tender-
ness are common, especially during disease exacerbations.
Inflammatory myositis involving the proximal muscles
occurs in up to 10% of patients with SLE and may develop at
any time during the course of the disease. However, this
myositis is usually mild and responds well to moderate doses
of glucocorticoids.
Electromyographic studies and elevations of the serum
creatine phosphokinase (CK) or aldolase levels do not help
differentiate idiopathic inflammatory myopathies from drug-
induced conditions (both show myopathic patterns of tissue
injury). Similarly, a low serum CK value even in the face of
Fig. 13.2 Jaccoud’s arthropathy. A deforming, non-erosive arthropathy
obvious muscle dysfunction can be found in patients with in a patient with SLE. A hallmark of Jaccoud’s arthropathy is the pres-
connective tissue disease, including SLE. Thus, a normal CK ence of marked but reducible flexion deformities (Figure courtesy of
value in the presence of other symptoms and signs of myosi- Dr. John Stone)
tis should not dissuade the clinician from pursuing that diag-
nosis through electrodiagnostic testing and muscle biopsy. The erosive arthritis, sometimes termed “rhupus,” comprises
a clinical and serological overlap syndrome between RA and
Myth: Autoimmune hemolytic anemia is the most common
SLE.
cause of anemia in lupus.
In SLE patients with synovitis, antiCCP IgG antibodies
Reality: Autoimmune hemolytic anemia (AIHA) is found in are strongly associated with erosions (odds ratio 29; 95%
approximately 10% of patients with SLE. AIHA is recog- confidence interval 5–174) (Chan et al. 2008). Furthermore,
nized by the finding of spherocytes on the peripheral blood nearly all antiCCP antibody positive lupus patients fulfill cri-
smear, a sharp fall in hemoglobin (>3g/dL), an elevated retic- teria for both SLE and RA and thus have “rhupus” (Damian-
ulocyte count (>5%), a rise in the serum bilirubin level, and Abrego et al. 2008). The HLA alleles associated with antiCCP
a positive Coombs test. In contrast to idiopathic AIHA, sple- antibodies in RA patients are also associated with antiCCP
nectomy is rarely indicated in SLE. antibodies in SLE patients (Chan et al. 2008). Thus, from an
The most common causes of anemia in SLE are chronic immunological point of view, rhupus patients really appear to
disease and iron deficiency. In the absence of symptoms such have both RA and SLE.
as easy fatigability or dyspnea on exertion, the anemia of Lupus is associated with the production of type I inter-
chronic inflammation does not require specific treatment. feron and a peripheral blood interferon “signature.” The fact
However, for lupus patients with stage 4 chronic kidney dis- that type I and type II interferons are powerful inhibitors of
ease – which corresponds to a glomerular filtration rate of osteoclasts is one likely explanation for the fact that only a
less than 30 mL/min – the Kidney Disease Outcomes Quality small subset of lupus patients has bony erosions.
Initiative guidelines recommend maintaining a target hemo-
globin of 11–12 mg/dL through the use of erythropoietin
(http://www.kdoqi.org).
13.4 Cutaneous Lupus
Myth: Erosive arthritis does not occur in patients with SLE.
Reality: Joint involvement is a common, early manifestation Pearl: The malar rash of SLE is not transient.
of SLE. Many clinicians believe that a critical discriminator
Comment: A lupus rash should be substantiated by an experi-
between rheumatoid arthritis (RA) and SLE is the lack of
enced clinician. A “malar rash” that is transient may not be
erosive changes in lupus. As a general rule, this is true.
due to lupus at all: flushing, blushing, polymorphous light
However, up to 5% of patients with SLE develop an erosive
eruption, solar urticaria, glucocorticoid-induced acne, sebor-
arthropathy (Isenberg and Horsfall 1998).
rheic keratosis, and rosacea can all lead to facial erythema.
Joint involvement in SLE can be categorized as follows:
Lupus rashes are inflammatory and therefore have texture:
• A non-deforming arthropathy they are raised. Moreover, because inflammation takes time to
• A deforming but non-erosive arthropathy (Jaccoud’s resolve, lupus rashes do not resolve within hours. The malar
arthropathy) (Fig. 13.2) rash of SLE spares areas shaded from ultraviolet light, such as
• An erosive arthritis the nasolabial folds and below the nares (Fig. 13.3a).
13 Systemic Lupus Erythematosus 137

Pearl: The malar rash of lupus is often confused with other a

entities, even by dermatologists.

Comment: Dermatologists are often asked to evaluate patients

who have a central facial eruption. The question is whether
this represents rosacea, some other entity, or if it is indeed
the butterfly rash of SLE.
In practice, the vast majority of such patients have rosa-
cea, seborrheic dermatitis, or some other papulosquamous
eruption. The classic butterfly eruption of SLE is observed
typically only in patients with acute, fulminant SLE. Patients
with chronic disease or those under treatment rarely develop
malar rashes. When it does occur, it is a continuous lesion
that goes from one cheek to the other and involves the nasal
bridge. It is also scaly, red, and papular, and pustules are
absent (Fig. 13.3b).
In contrast, the most commonly confused facial eruption is
rosacea. Rosacea has discrete areas of involvement isolated to
the cheeks, chin, and occasionally the central forehead. The
nasal bridge is not involved in rosacea, but the bulb of the nose
is often edematous and red. In active rosacea, one can see
pustules on examination, although these may be quite small.
Seborrheic dermatitis usually involves the nasolabial
folds, eyebrows, ears, and scalp. The scale often has a greasy
appearance. Flaking of the scalp is common. b
Finally, the cutaneous features of dermatomyositis can
mimic SLE. Atypical skin rashes in dermatomyositis are
sometimes confused with SLE (Fig. 13.4). Distinguishing
features of dermatomyositis are the presence of violaceous,
edematous plaques on the upper eyelids, forehead, and sides
of the neck, with sparing of the underside of the chin.
Dermatomyositis rashes often develop on a background of
telangiectasia. Patients with dermatomyositis also often have
scalp involvement that is highly pruritic (Fig. 13.5).

Pearl: Cutaneous LE is photosensitive.

Comment: The ACR definition of photosensitivity in the

SLE classification criteria is paraphrased as follows: “A skin
rash either observed by a physician or reported by the patient
that occurs as a result of an unusual reaction to sunlight.”
Unfortunately, there are many photosensitive skin condi-
tions that can fulfill this ACR criterion for SLE. Such condi- Figs. 13.3 (a, b) Malar Rashes. (a) One of the key features of a malar
rash is that is spares the nasolabial folds, as illustrated in this figure. (b)
tions include polymorphous light eruption, photoallergic The malar rash of SLE is a continuous plaque that extends over the
contact dermatitis, rosacea, dermatomyositis, and medica- bridge of the nose. There is distinct sparing of the nasolabial folds.
tion-induced photosensitivity. There is substance to the plaque, which is not transient (Figure courtesy
The concept of photosensitivity in SLE also includes a of Dr. Grant Anhalt)
phenomenon that does not manifest itself by a skin rash, yet
still constitutes an abnormal reaction to sunlight. As an
example, ultraviolet light can serve as the stimulus for flares demonstrated in Fig. 13.6, the underside of the chin, a sun-
of systemic signs and symptoms of SLE, not just those con- protected site, is spared. In contrast, widespread cutaneous
fined to the skin (Lin 2007). lesions occur over areas of the head, neck, chest, and arms,
Clinical evidence of photosensitivity is demonstrated all of which have greater sun exposure (Bijl and Kallenberg
most easily by the distribution of skin lesions. In the example 2006).
138 M. Petri et al.

Fig. 13.4 Dermatomyositis masquerading as lupus. This patient was Fig. 13.6 Photosensitivity. In this patient with systemic lupus erythe-
misdiagnosed as having systemic lupus erythematosus for years. matosus, there is a distinct sparing of the underside of the chin, a sun-
Despite the atypical features of the facial eruption, clinical clues to the protected area (Figure courtesy of Dr. Rachel Abuav)
correct diagnosis of dermatomyositis are the heliotrope rash in the peri-
ocular area and the involvement of the scalp (Figure courtesy of
Dr. Rachel Abuav)

or the patient because of the delay in lesion formation that

often follows UV light exposure (Tsokos 2004).
Myth: Patients with SLE who develop skin lesions most likely
have a cutaneous manifestation of SLE.
Reality: Patients with SLE often develop non-lupus skin
lesions that mimic cutaneous lupus. These can include scaly,
erythematous macules or pustules on their skin or annular
lesions with peripheral scale (Modi et al. 2008). It is impor-
tant to consider the possibility of non-lupus diagnoses in
such patients. As an example, tinea corporis is an excellent
mimicker of both acute SLE and subacute cutaneous lupus
erythematosus (SCLE).
The diagnosis of tinea can be made in SLE patients with
Fig. 13.5 Scalp involvement in dermatomyositis (Figure courtesy of scale or pustules by performing a potassium hydroxide (KOH)
Dr. John Stone)
preparation of the scale or the contents of pustules to look for
fungi. KOH digests the keratin, thereby facilitating the micro-
scopic examination of the preparation for septated hyphae.
Pearl: The clinical effects of UV light exposure can be
Cutaneous scales can also be cultured in a microbiology labo-
delayed by months.
ratory. SLE patients who receive systemic immunosuppres-
Comment: In the great majority of patients with SLE, UV sion for a cutaneous manifestation that is attributed to lupus
light exposure results in an abnormally prolonged erythema but experience expansion of the cutaneous lesions should be
of the skin. In many patients, this reaction ensues several evaluated carefully for a possible fungal infection of the skin
weeks after the UV exposure. Furthermore, it appears that (Fig. 13.7). Furthermore, treatment with topical glucocorti-
UV light can cause systemic disease flares up to 6 months coids can alter the clinical appearance of tinea, rendering it
after a prolonged UV light exposure (Lin 2007; Scheinfeld “tinea incognito” (Fig. 13.8).
and Deleo 2004). In a photoprovocation study, UVB light Other skin diseases, such as erythema annulare cen-
triggered lesions in 33% of patients and UVA triggered trifugum, cutaneous T-cell lymphoma, and granuloma annu-
lesions in 14%. In 53% of patients, however, both UVA and lare are annular inflammatory skin diseases that can recur and
UVB were required to induce lesions of cutaneous LE. have an appearance similar to SCLE. The possibility of these
Most types of cutaneous LE are photosensitive, but this diagnoses underscores the importance of obtaining a skin
relationship may not be appreciated by either the physician biopsy on patients at the time they first develop skin disease.
13 Systemic Lupus Erythematosus 139

addition to the “usual” lupus autoantibodies against nuclear

proteins and ribonuclear proteins. When this happens, such
patients develop mucosal erosions in addition to large, tense
blisters, usually on the head and neck or in the axillae and
groin (Fig. 13.9a–c). This condition is known as bullous LE.
Patients with bullous LE tend to have very aggressive disease
courses and frequently have multiple other autoantibodies,
e.g., those directed against double-stranded DNA and the Sm
or RNP antigens. Hypocomplementemia, glomerulonephri-
tis, arthritis, and serositis are other common disease compli-
cations in the setting of bullous LE.
When bullous LE lesions occur outside the context of
SLE, the disorder is termed epidermolysis bullosa acquisita.
Fig. 13.7 Tinea in lupus. Extensive annular scaly plaques on the arm In both conditions, patients with true bullous LE or epider-
despite adequate control of systemic disease. Scraping and KOH exam- molysis bullosa acquisita have circulating autoantibodies
ination revealed numerous hyphae. Fungal culture grew Trichophyton that are directed against a dermal collagen protein called col-
rubrum, the most common of pathogens that cause tinea corporis. The
patient was treated successfully with a topical antifungal cream (Figure
lagen type VII.
courtesy of Dr. Rachel Abuav) Patients with bullous LE and epidermolysis bullosa acquisita
have tense blisters of the skin and mucosa that lead to scarring,
because type VII collagen is present in the dermis below the
basement membrane zone. Bullous LE and epidermolysis
bullosa acquisita are both serious disorders that can involve the
eyes, esophagus, or larynx. If left untreated, these disorders
result in corneal blindness or laryngoesophageal stenosis.
The diagnosis of bullous LE depends on the demonstra-
tion of collagen type VII antibodies in skin or in the serum.
Direct immunofluorescence of skin in perilesional tissue
reveals thick, homogeneous deposition of IgG along the
basement membrane zone (Fig. 13.10). Only skilled immu-
nofluorescence laboratories will be able to distinguish bullous
LE or EBA from other blistering disorders of the skin, such
as bullous pemphigoid.
Pearl: All bullous lesions in SLE do not constitute bullous LE.
Comment: Bullous LE is a highly distinctive condition, with
specific serologic markers. The medical literature is often
confusing on this point. In cutaneous lupus, injury to the
basal epithelial cells can cause what is called a “vacuolar
degenerative change” at the interface of the dermis and
epidermis. If this vacuolar interface change is severe enough,
the epidermis separates from the underlying dermis at focal
Fig. 13.8 Tinea incognito. Application of topical glucocorticoids often sites and a limited degree of blistering can ensue. However,
eliminates the surface scale associated with tinea faciei, making diag-
nosis tricky. A scraping and KOH examination of the plaque revealed this condition is separate from true bullous LE, which is
hyphae. The patient’s rash cleared with topical antifungal cream (Figure associated with autoantibodies to type VII collagen (Schmidt
courtesy of Dr. Grant Anhalt) et al. 2008).
Myth: Hair loss in SLE is usually due to systemic flares of the
lupus.
Pearl: Bullous lupus erythematosus (LE) and epidermolysis
Reality: Diffuse thinning of scalp hair can herald an exacer-
bullosa acquisita are actually the same disease from the
bation of SLE. However, there are many causes of diffuse,
standpoint of the skin.
non-scarring alopecia in SLE. First, medications such
Comment: A small percentage of patients with acute SLE as prednisone, methotrexate, mycophenolate mofetil, and
develop autoantibodies directed against type VII collagen in hydroxychloroquine can be associated with significant hair
140 M. Petri et al.

a c

Fig. 13.9a–c Bullous lupus erythematosus. (a) Tense blisters on the tesy of Dr. Grant Anhalt) (c) Extensive scarring of the inframammary
neck of a young woman with acute SLE. Autoantibodies to type VII col- region due to chronic blistering in this poorly compliant patient with
lagen were present (Figure courtesy of Dr. Grant Anhalt) (b) Extensive bullous LE. The patient also had esophageal stenosis secondary to bullous
erosions and tense blisters in the axilla. This is a common site of involve- LE. Note the complete and permanent destruction of the fingernail from
ment in bullous lupus erythematosus. Scarring is evident (Figure cour- scarring of the nail matrix (Figure courtesy of Dr. Rachel Abuav)

thinning. In addition, some patients who have lost hair areas of scalp involvement, although more typically there are
because of a previous cause of diffuse loss, e.g., telogen circular areas of nonscarring hair loss (Werth et al. 1992).
effluvium associated with a lupus flare, do not regrow their
Myth: Most nodules in patients with lupus are from lupus
hair to its pre-illness density because of pre-existing (geneti-
panniculitis.
cally predetermined) androgenetic alopecia (Fig. 13.11).
Eliciting a family history of androgenetic hair loss is helpful Reality: Many cutaneous nodules in SLE patients are not due
in identifying this phenomenon. to lupus panniculitis. The incidence of erythema nodosum is
Patients with SLE are also at an increased risk of alopecia increased in patients with lupus. Other possible causes of
areata, another form of nonscarring hair loss that is mediated nodules in SLE are vasculitis, livedoid vasculopathy, pan-
by the immune system. Some of these patients have large niculitic lymphoma, lipodermatosclerosis, and infections,
13 Systemic Lupus Erythematosus 141

Fig. 13.10 Direct immunofluorescence of a skin biopsy that reveals a

thick homogeneous band of IgG deposition along the basement mem-
brane zone. The epidermis is above this band, and the dermis below.
This finding is diagnostic of bullous lupus erythematosus or epidermo-
lysis bullosa acquisita (Figure courtesy of Dr. Grant Anhalt)

Fig. 13.12 Lupus panniculitis. Lupus panniculitis is manifest in this

patient as depressions in the skin on her cheeks. Overlying skin changes
are also prominent (Figure courtesy of Dr. Rachel Abuav)

mately half of all patients with lupus panniculitis have over-

lying skin changes that are evident such as scaling, erythema,
and ulceration. Topical therapy is generally not an effective
Fig. 13.11 Persistent alopecia in lupus. Despite prolonged remission first-line therapy for lupus panniculitis. Intralesional gluco-
following a severe lupus flare in which the patient developed alopecia,
this patient’s hair density has never returned to its pre-illness density. corticoids are employed on occasion (Callen 2006).
The lupus flare unmasked the presence of androgenetic alopecia
Pearl: The sudden onset of multiple nodules in SLE may be
(Figure courtesy of Dr. Rachel Abuav)
caused by eruptive dermatofibromas.
particularly bacterial, fungal, or atypical mycobacterial Comment: Dermatofibromas are common benign skin tumors
infections (Aguilera et al. 2007; Kroshinsky 2008). A skin of spindle cells. In general, they form on the arms and legs,
biopsy that captures a sizeable amount of subcutaneous fat is but can be eruptive in SLE, leukemia, HIV infection, and
critical to making the proper diagnosis in lupus patients with other diseases of immune dysregulation. The sudden onset of
nodules. numerous nodules in a patient with SLE may be caused by
Lupus panniculitis, a rare condition, has a characteristic eruptive dermatofibromas (Fig. 13.13).
morphology. Lupus panniculitis, which has a predilection The patient shown in Fig. 13.13 developed numerous ten-
for the malar region and the lateral deltoid fat pads, leads der nodules at the time of a lupus flare. With prednisone, these
eventually to depressions in the skin (Fig. 13.12). Approxi- lesions diminished in size but did not resolve completely.
142 M. Petri et al.

Fig. 13.13 Eruptive dermatofibromas in lupus. Numerous subcutane- Fig. 13.14 Erythema over the phalanges in systemic lupus erythemato-
ous nodules appeared suddenly in the setting of an SLE flare. Skin sus. The location of erythema in lupus contrasts with the Gottron’s pap-
biopsy of a nodule was consistent with a dermatofibroma. Prednisone ules typical of dermatomyositis (Figure courtesy of Dr. John Stone)
diminished the size of the nodules but did not eliminate them com-
pletely (Figure courtesy of Dr. Rachel Abuav)

Treatment is not necessary but can include surgery, which is

curative (Niiyama et al. 2002).
Pearl: Lupus can “Koebnerize.”
Comment: Heinrich Koebner described the phenomenon that
bears his name in 1876, when he observed new psoriatic
lesions appearing in normal skin after a patient with psoriasis
had been bitten by a horse. Since this description, the Koebner
phenomenon has been witnessed in numerous other skin con-
ditions, including SLE. Scratching induces discoid lesions in
some patients. This has been confirmed by biopsying the
lesions and finding histologic features that are characteristic
of chronic cutaneous lupus. In addition, old scars can be a
focus of new lesional lupus. These include burns, surgical Fig. 13.15 Discoid lupus in the mouth. A discoid lupus erythematosus
scars, and vaccination sites. Oral lesions of lupus can be initi- plaque on the buccal mucosa of a man with SLE (Figure courtesy of Dr.
ated by bite trauma (Ueki 2005). Rachel Abuav)

Pearl: The location of erythema on the hands can distinguish

SLE from dermatomyositis. necrotizing vasculitis and alleged to predict severe systemic
disease flares (Ropes 1976). This concept is cited widely
Comment: Although lupus may cause erythematous rashes
today, despite subsequent studies that refute the notion. In one
that resemble those of dermatomyositis, involvement of the
study of oral lesions in ten patients with SLE, biopsy of the
knuckles of the fingers (Gottron’s papules) is atypical of
ulcerations, erythema, and discoid lesions all showed inter-
SLE. SLE usually affects the areas between the small joints
face mucositis (Jorizzo et al. 1992). None showed vasculitis.
of the fingers and spares the knuckles (Fig. 13.14). Although
Oral lesions occur in 20–25% of patients with cutaneous
there is significant overlap between SLE and dermatomyosi-
lupus and in 40% of patients with SLE. They range from
tis with regard to their cutaneous manifestations over the
erythematous patches to true discoid plaques (Fig. 13.15) to
hands, this Pearl is a good “rule of thumb.”
frank ulceration. The histopathological appearance of these
Myth: Oral ulcerations in SLE are due to necrotizing lesions is one of interface mucositis with vacuolar degenera-
vasculitis. tion of basal keratinocytes.
Reality: One of the ACR criteria for the classification of SLE Myth: It is impossible to distinguish lip lesions of SLE from
is oral ulceration. Oral ulcerations have been attributed to herpes simplex in an immunocompromised patient.
13 Systemic Lupus Erythematosus 143

a a

b b

c
Fig. 13.17a–b Chronic lip lesions in patients with systemic lupus
erythematosus (SLE). These lesions are consistent with chronic cutane-
ous LE (discoid LE) and are centered on the “wet line” of the lip,
extending onto the vermillion, but sparing the vermillion border (Figure
courtesy of Dr. Rachel Abuav)

Although lesions of orolabial HSV among immunocom-

promised patients can be atypical, these general principles
often hold true:
• HSV on the lip erupts on the vermillion border with vesi-
Fig. 13.16a–c Lesions of acute systemic lupus erythematosus (SLE)
on the lips. Lip lesions in three patients with SLE flares. All lesions are cles and crusting on an edematous base.
centered on the “wet line” of the lip and extend onto the vermillion. • In immunocompetent patients, intraoral lesions of HSV
There is hemorrhagic and serum crusting. The ulceration is shallow are restricted to mucosa affixed to bone: i.e., to the gingival
(Figure courtesy of Dr. Rachel Abuav) and hard palate.
• Ulcerations due to HSV are typically punched out, edema-
tous, and often grouped. In the lupus patient, ulcerations
Reality: Lip lesions due to SLE are centered on the wet line of
may occur more extensively and on areas not affixed to
the lip and extend onto the vermillion border (Fig. 13.16a–c).
bone, but if one looks carefully, there will be lesions that
In contrast, lip lesions due to the herpes simplex virus classi-
are typical of orolabial HSV present. Along with viral cul-
cally erupt on the vermillion border and extend onto the ver-
tures, these clinical clues make the diagnosis possible.
million of the lip. Moreover, lip ulcers of SLE tend to be oval
in shape and have ill-defined borders. SLE lesions tend to have Of course, both lupus ulceration and reactivated HSV infec-
an overlying hemorrhagic and serum crust. The edges are not tions can occur concomitantly in patients with SLE. Because
heaped up; the center is typically rather shallow. Chronic SLE there can be morphologic overlap and reactivation of HSV
lesions have characteristic “discoid” borders with depigmen- within ulcerations caused by SLE, one should always obtain
tation and scar in the center (Figs. 13.17a and b). viral cultures of oral ulcerations in the lupus patient.
144 M. Petri et al.

Pearl: Patients with a skin biopsy read as lupus actually may months after stopping the culprit drug, but short-term treat-
have dermatomyositis. ment with prednisone and hydroxychloroquine leads to quick
resolution. AntiRo/SSA antibodies can persist even after reso-
Comment: The skin biopsy findings in SLE, discoid lupus,
lution of the skin disease (Sontheimer et al. 2008).
subacute cutaneous lupus erythematosus, and dermatomyosi-
tis can be identical. Because cutaneous LE is more common Pearl: Skin disease is uncommon in drug-induced lupus.
than dermatomyositis, biopsies are frequently read as consis-
tent with or diagnostic of lupus. Careful clinicopathologic cor- Comment: One must distinguish between drug-induced lupus
relation is required to determine a patient’s true diagnosis. If a and lupus-like drug eruptions. In patients with drug-induced
patient has a biopsy read as showing changes of lupus but has lupus, the clinical manifestations are typically serositis and
Gottron’s papules, then that patient has dermatomyositis. arthritis, but skin disease is rare (Tsokos 2004).
The most common forms of drug-induced lupus are due
Pearl: Patients with subacute cutaneous lupus frequently to drugs such as hydralazine, procainamide, and minocy-
have drug-induced disease. cline. These patients have a monomorphic serologic profile
Comment: Many medications are associated with the develop- with antibodies directed only against histones, producing
ment of cutaneous lupus. Nearly all of these triggering medi- high titers of ANA that stain with a homogeneous pattern on
cations are associated with a specific subset of lupus, namely immunofluorescence testing.
subacute cutaneous lupus erythematosus, which is most often
Pearl: The presence of discoid lupus lesions below the head
photosensitive (Fig. 13.18). If the patient is receiving a thiaz-
and neck region is associated with an increased risk of devel-
ide, calcium channel blocker, terbenifine, interferon, or an
oping systemic lupus.
inhibitor of tumor necrosis factor, strong consideration should
be given to stopping the drug. Many of these patients develop Comment: Discoid lupus erythematosus (DLE) is the most
antiRo/SSA antibodies. Improvement may not be apparent for common type of chronic cutaneous lupus erythematosus. It

a b

Fig. 13.18 (a, b) Subacute cutaneous lupus erythematosus. This annular, papulosquamous rash is often associated with extreme photosensitivity and
antibodies directed against the Ro antigen (Fig. 13.18a courtesy of Dr. Victoria Werth, Fig. 13.18b courtesy of Dr. Rachel Abuav)
13 Systemic Lupus Erythematosus 145

Fig. 13.20 Late discoid lupus erythematosus (DLE). In this patient,

atrophy and pigmentary alteration denote scarring, which is irrevers-
ible. This is characteristic of established plaques of DLE (Figure cour-
tesy of Dr. Rachel Abuav)

Among all patients with DLE, the risk of developing SLE

is only about 5%. Among those with DLE lesions localized
to the head and neck, the probability of a complete remission
is on the order of 50% (Callen 2006).
Myth: Patients with cutaneous lupus should get some sun
exposure to increase their vitamin D levels.
Reality: Vitamin D levels in cutaneous lupus patients are fre-
quently low. This is probably related to sun avoidance and
the use of sunscreens (Cusack et al. 2008). However, it is
easy to provide oral supplementation of vitamin D to these
patients, and the risks of vitamin D replacement are minimal
relative to the potential for causing exacerbations of the
underlying lupus through sun exposure.

Fig. 13.19 Early discoid lupus erythematosus (DLE). In this patient

with early plaques of DLE, early treatment will avoid scarring (Figure
courtesy of Dr. Grant Anhalt) 13.5 Lupus Nephritis

can be seen as an isolated finding; however, 15–20% of SLE Myth: Serum creatinine is a reliable indicator of renal
patients have DLE lesions. Clinically, they are sharply demar- function.
cated, round (hence, “discoid”) plaques that initially appear
Reality: Measurement of the serum creatinine level is a prac-
erythematous or violaceous (Fig. 13.19). With time, they
tical but relatively insensitive indicator of abnormalities in
become scaly and demonstrate follicular plugging. In the end
the glomerular filtration rate. The serum creatinine level is
stages, the rims are hyperpigmented and the centers are
affected by variables with little direct relationship to renal
depigmented, scarred, and often atrophic (Fig. 13.20).
function, including sex, muscle mass, and age. Nevertheless,
When DLE is found in isolation, the lesions are usually
reproducible elevations in the serum creatinine level
confined to the scalp, face, and neck, and demonstrate a pho-
(e.g., ³20–30% increase) are of concern, even if they fall
toaccentuated distribution. When lesions are extensive and
within the normal range.
present above and below the waist, the risk of developing
SLE (if not already evident) appears high. Such patients usu- Pearl: In a pregnant lupus patient, substantial renal dysfunc-
ally have positive serologies and should be monitored closely tion can be present despite the finding of a normal serum
for internal organ involvement. creatinine level.
146 M. Petri et al.

Comment: Normal pregnancies increase the glomerular In the SELENA studies, the investigators amended the
filtration rate through expansion of the plasma volume. This SLEDAI to require that hematuria only be used in the pres-
leads to a decrease in serum creatinine levels in pregnancy. ence of proteinuria as a manifestation of lupus (Buyon et al.
Thus, a “normal” serum creatinine level in a pregnant lupus 2005; Petri et al. 2005).
patient might actually signal a substantial alteration of renal
Myth: The urinalysis is the poor man’s renal biopsy.
function (see Chapter 17).
Reality: If this were true, then he would be a poor man, indeed
Pearl: An experienced clinician recognizes the point of
(and she a poor patient). Ample evidence suggests that the
diminishing returns with the use of immunosuppression for
findings in a spun urine sample correlate poorly with the his-
glomerulonephritis.
topathologic features on biopsy (Christopher-Stine et al.
Comment: It’s not worth saving the kidneys if you’ve lost the 2007; Huong et al. 1999). Even in the setting of diffuse pro-
patient. The efficacy of renal replacement therapy – both liferative glomerulonephritis, minimal hematuria may be evi-
dialysis and transplantation – means that end-stage renal dis- dent on urinalysis.
ease no longer marks the end of the line for a patient with Yet the importance of a careful urinalysis in SLE should
SLE. Sometimes, it is prudent to let severely damaged kid- not be discounted. An abnormal urinalysis provides critical
neys go, rather than lose the patient to complications of information. Urinalysis with microscopic examination of the
immunosuppression. urine sediment should be performed at every patient visit.
As an example, end-stage renal disease at some point in the However, in the setting of renal dysfunction of unclear etiol-
future is probably inevitable for a patient who has arrived at a ogy, the fundamental issue is tissue.
serum creatinine of 4.0 mg/dL slowly. When in doubt, a renal
biopsy showing Class VI (renal sclerosis) can provide suffi-
cient evidence that the need for renal replacement therapy is
unavoidable and that additional cyclophosphamide or myco- 13.6 Central Nervous System Lupus
phenolate mofetil is more likely to cause harm than good.
Myth: “Lupus headaches” are part and parcel of the disease.
Pearl: Proteinuria can be diminished by 50% without pre-
scribing a single milligram of prednisone. Reality: Although “lupus headache” is one of the neurologic
manifestations listed on the SLE Disease Activity Index
Comment: Only a minority of patients with lupus nephritis
(SLEDAI) (Bombardier et al. 1992), headaches are actually
achieve complete remission, even with modern treatment
rare in lupus. Most women with SLE who have headaches
regimens. As an example, only 23% of patients treated with
have migraine physiology. Case-control studies have not
mycophenolate mofetil in one randomized trial achieved a
shown any increase in headaches among SLE patients com-
complete remission (Ginzler et al. 2005). Our preoccupation
pared with controls (Fernandez-Nebro et al. 1999). Although
with selecting and honing the proper regimen of immuno-
headache is listed under the ACR neuropsychiatric case defi-
suppression can lead us to overlook other measures that play
nitions, this symptom has poor specificity for SLE in popula-
important roles in the long-term salvage of kidneys.
tion-based studies (ACR ad hoc committee on neuropsychiatric
Renal-sparing protocols are underutilized in lupus nephri-
lupus 1999; Ainiala et al. 2001).
tis. The reduction of proteinuria through ACE inhibition delays
An SLE patient with a severe headache should have an
and prevents renal sclerosis, hyperlipidemia, hypercoagulabil-
evaluation designed to exclude brain hemorrhage or dural
ity, and their concomitant complications. The use of ACE
sinus thrombosis. Lumbar puncture is also appropriate to help
inhibitors and angiotensive receptor blockers can reduce pro-
exclude infection, malignancy, and pseudotumor cerebri.
teinuria by up to 50%. Spironolactone also helps to reduce pro-
Lupus meningitis, manifested by a pleocytosis and elevated
teinuria (Chrysostomou and Becker 2001). (In men, eplerenone
CSF protein, is a diagnosis of exclusion.
should be used, because of its lower risk of gynecomastia.)
Pearl: When managing patients with neuropsychiatric lupus,
Pearl: Proteinuria is a better indicator of lupus nephritis
improvement is likely to take several months. Patience is
than is hematuria.
required.
Comment: Hematuria is one of the renal manifestations of
Comment: A standardized nomenclature for neuropsychiatric
lupus listed on the SLE Disease Activity Index (SLEDAI)
lupus was developed by the American College of Rheumatology.
(Bombardier et al. 1992). However, sound clinical judgment
This distinguishes three subsets of syndromes:
is essential before ascribing hematuria to active lupus. SLE
patients may have hematuria from a variety of causes, includ- • Psychiatric, cognitive deficits, and acute confusional states
ing menses, trauma, renal calculi, and thin basement mem- • Neurological syndromes of the central nervous system
brane disease. • Neurological syndromes of the peripheral nervous system
13 Systemic Lupus Erythematosus 147

In one single-center study, only 10 of 485 patients with SLE with eye movements. In contrast, optic neuropathy usually
developed major neuropsychiatric disease manifestations presents with acute unilateral vision loss that is painless,
(Pego-Reigosa and Isenberg 2008). In eight of these ten caused by occlusion of the small vessels to the optic nerve.
cases, the outcome was ultimately excellent but the time to Unilateral optic neuropathy appears to reflect a focal throm-
recovery was long, often up to 12 months or longer for com- botic event associated with the antiphospholipid syndrome.
plete symptom resolution. In contrast to idiopathic optic neuritis or to that associated
A critical feature in the management of some of these with multiple sclerosis, the visual prognosis in SLE is not
patients is to understand that improvement in the psychosis always good, particularly in the setting of optic neuropathy.
may require several months. Many patients attain significant Recurrence is not unusual and further worsens the prognosis.
and sustained resolution of symptoms following intensive In optic neuritis, glucocorticoids alone or in combination
immunosuppressive therapy, but in some the improvement with immunosuppressive therapy are effective in almost 50%
does not occur quickly. of the cases. In optic neuropathy, the outcome is worse. Some
With regard to the treatment of lupus psychosis, there are patients appear to respond to immunosuppressive therapy
no published controlled trials of any treatment strategy. alone. For patients in whom antiphospholipid antibodies are
Moreover, the long-term outcome of lupus psychosis is not identified, anticoagulation is appropriate.
well established. Several retrospective studies have reported Posterior reversible encephalopathy syndrome (PRES) is
the benefit using glucocorticoids and intravenous cyclophos- a rare neurologic condition associated with renal insuffi-
phamide in patients with neuropsychiatric lupus in addition ciency, hypertension, acute intermittent porphyria, calcineu-
with antipsychotic, antidepressant, or anticonvulsant therapy rin inhibitors, and a variety of other conditions, including
(Boumpas et al. 1991; Neuwelt et al. 1995; Ramos et al. lupus. Patients present with headache, seizures, loss of vision
1996; Baca et al. 1999; Takada et al. 2001). from cortical blindness, and altered mental function. The
pattern on magnetic resonance imaging studies is one of
Pearl: Practically speaking, it is often difficult to separate
transient posterior cerebral hyperintensities on T2-weighted
central nervous system disease manifestations that are due
images (Fig. 13.21). Patients may respond to glucocorticoids
to inflammation from those that are caused by thrombosis.
and immunosuppressive therapy.
Comment: Anticoagulation with warfarin is a cornerstone of
treatment for lupus patients who present with neuropsychiat-
ric events that are characteristic of the antiphospholipid syn-
drome (APS), e.g., stroke, intracerebral venous thrombosis, or
chorea. However, the cause of a particular clinical syndrome
is often not clearcut in lupus. When the relative contributions
of “inflammatory” mechanisms as opposed to “thrombotic”
pathways are unclear, the patient may require combination
therapy with glucocorticoids, cytotoxic drugs, and warfarin.
Myth: Loss of vision in lupus is probably due to optic neuritis.
Reality: Significant ocular pain or reduction in vision requires
urgent assessment by an ophthamologist (Sivaraj et al. 2007).
Sight-threatening ocular disease in lupus may be due to a
variety of processes at a number of ocular sites:
• Lens (cataract formation)
• Vitreous humor (hemorrhage)
• Retina (occlusion of retinal vein or artery, retinal detach-
ment, toxic maculopathy from antimalarial agents or
glucocorticoids)
• Choroid (choroidopathy, infarction, effusion)
• Neuro-ophthalmic tract (optic nerve, optic chiasm, or the
occipital cortex).
Optic nerve disease occurs in fewer than 1% of patients with
SLE and may be due to optic neuritis or ischemic optic neu-
Fig. 13.21 Magnetic resonance imaging study in a lupus patient with
ropathy (anterior or posterior). Optic neuritis typically presents the posterior reversible encephalopathy syndrome (PRES) (Figure
acutely with unilateral loss of vision and pain that is worse courtesy of Dr. Arezou Khosroshahi)
148 M. Petri et al.

13.7 General Management Points to increase in the setting of severe bacterial and fungal
infections. Data on this point have been inconsistent. The
test is currently not widely available, and is not recommended
Myth: A high erythrocyte sedimentation rate (ESR) indicates
for general use.
active clinical SLE.
Pearl: Lupus patients are immunosuppressed, even without
Reality: Many SLE patients have dramatic elevations in their
our help.
ESRs in the absence of apparent clinical disease activity.
Similarly, hypocomplementemia and elevated titers of dsDNA Comment: Patients with SLE have an increased risk of cer-
antibodies also occur frequently in the absence of clinical dis- tain viral, fungal, and bacterial infections even before their
ease. Such patients fall into the category of “serologically introduction to prednisone, glucocorticoid-sparing agents,
active but clinically quiescent” SLE (Walz LeBlanc et al. and other treatments. SLE patients have difficulty with mul-
1994). These patients should not be treated for active SLE, but tiple viral infections, especially herpes zoster, cytomegalovi-
rather be watched carefully: they are more likely to experience rus, and papillomaviruses. They are also subject to fungal
disease flares within the next year. Changes in serum comple- infections, especially oral candidiasis, but also onychomyco-
ment levels and antidsDNA antibody titers do not predict dis- sis. Perhaps because of splenic dysfunction, SLE patients
ease flares over the following month (Ho et al. 2001a and b). have difficulty with encapsulated organisms such as the
Serum levels of C-reactive protein (CRP) can also be ele- pneumococcus and meningococcus. Don’t forget to vacci-
vated in SLE that is clinically quiescent. These patients also nate all SLE patients, especially the teenager with lupus who
bear watching closely, but some data indicate that obesity, a is going off to college (Petri 1998).
problem in many patients with longstanding SLE, can lead to
Pearl: Most infections in SLE are preventable.
CRP elevations (Petri 2008).
Comment: Infection remains a leading cause of morbidity
Myth: In a febrile lupus patient, both C-reactive protein and
and mortality in patients with SLE (Zandman-Goddard and
procalcitonin levels are useful in excluding infection.
Shoenfeld 2003). This susceptibility is due to a combination
Reality: Infections account for approximately one-fourth of of factors, including underlying immune dysregulation
all deaths in lupus. These disease complications are attribut- (manifested as defects in neutrophil chemotaxis), impaired
able primarily to therapy, particularly glucocorticoids and clearance of opsonized bacteria, and functional hyposplenia
cytotoxic drugs. When it comes to excluding infections in (Iliopoulos and Tsokos 1996).
SLE, the stakes of being wrong are so high that one can never In addition, certain disease manifestations, principally
rely entirely on a single laboratory test. One study from the renal involvement and treatment, also contribute to the occur-
National Institutes of Health in the 1970s indicated that 60% rence, type, and severity of infections (Noel et al. 2001;
of febrile episodes among SLE patients were due to active Fessler 2002; Gladman et al. 2002). Serious but non-fatal
lupus, but 23% were caused by infections (Stahl et al. 1979). infections are an independent risk factor for death 10 years
The oft-quoted myth is that SLE flares do not increase after the diagnosis of SLE (Noel et al. 2001). Judicious use
either C-reactive protein or procalcitonin. This statement is of glucocorticoids and the conscientious administration of
wrong often enough to make it dangerous. In a study from immunizations decrease the frequency of infections (Gilland
the Netherlands, serum CRP levels were normal (less than and Tsokos 2002).
6 mg/L) in only 34% of patients with SLE flares (13 of 38
Myth: Vaccines cause SLE flares.
cases), and were also normal in 10% of those with a systemic
infection (4 of 36 cases) (terBorg et al. 1990). Thus, even if Reality: Vaccines and lupus are a complicated story. There is
the CRP level is normal, infections must be excluded rigor- no evidence that inactivated vaccines cause SLE flares. This
ously in all SLE patients with fever. This is particularly true, has been studied rigorously for both the influenza vaccine and
of course, if the patient has a history of substantial immuno- the pneumococcal vaccines, but the risk posed to lupus patients
suppressive drug use (Roy and Tan 2001). by the hepatitis B virus vaccination also appears negligible
Two features of active lupus, namely arthritis and pleuro- (Abu-Shakra et al. 2000; Klippel et al. 1979; Elkayam et al.
pericarditis, are often associated with elevated CRP levels 2002, 2005; Turner-Stokes et al. 1988). Because SLE patients
(Spronk et al. 1992; Hesselink et al. 2003). A 2-year prospec- may not mount vigorous vaccine responses, the pneumococcal
tive study reported that SLE flares associated with serositis vaccine should be administered every 5 years. The pneumo-
were substantially more likely to be accompanied by an ele- coccal vaccine should never be omitted because SLE patients
vated CRP than were those without serositis (mean CRP level have particular problems with encapsulated organisms.
76 mg/L versus 16 mg/L; P < 0.02). In short, clinicians should have no reticence about admin-
Similar considerations apply to procalcitonin. Procalci- istering killed vaccines to patients with SLE. Common vac-
tonin levels were alleged to remain low during SLE flares but cine practice recommendations are shown in Table 13.5.
13 Systemic Lupus Erythematosus 149

Table 13.5 Use of vaccinations in patients with systemic lupus erythematosus rated according to evidence (Adapted O’Neill 2006)
Vaccine Type(s) Evidence of efficacy Evidence of safety Concerns/comments
in SLEa in SLEa
BCG Live attenuated No evidence available No evidence available CI if immunosuppressed
MMR Live attenuated No evidence available Some evidence for use CI if immunosuppressed
Varicella Live attenuated No evidence available No evidence available CI if immunosuppressed
Yellow Fever Live attenuated No evidence available No evidence available CI if immunosuppressed
Polio Live attenuated (oral) No evidence available Some evidence for use CI if immunosuppressed
Available in some countries
Inactivated (parenteral)
No evidence available No evidence available
Hepatitis B Component (recombinant Some evidence against Some evidence for use ? Decreased efficacy
DNA)
Hepatitis A Inactivated No evidence available No evidence available
Influenza Inactivated component Good evidence for use Good evidence for use
Meningococcus Component polysaccharide No evidence available No evidence available
Conjugate
No evidence available No evidence available
Pertussis Inactivated whole cell No evidence available No evidence available
Component
No evidence available No evidence available
Pneumococcus Component polysaccharide Good evidence for use Strong evidence for use Significant minority do not
respond or response
short lived
Conjugate
No evidence available No evidence available
Haemophilus influenzae B Conjugate Good evidence for use Good evidence for use
Tetanus Toxoid Strong evidence for Strong evidence for use
use
Diptheria Toxoid No evidence available No evidence available CI if immunosuppressed
MMR = Measles, mumps, rubella
a
Represents overall impression from weighing up published series of the degree of evidence for or against the vaccines use in SLE. CI: Contra-
indicated

Pearl: The situation with live vaccines is more complex. patients on <20 mg/day prednisone, methotrexate (<0.4 mg/
kg/week), azathioprine (<3.0 mg/kg/day), or 6-mercaptopu-
Comment: Although influenza and pneumococcal vaccines rine (<1.5 mg/kg/day) do not constitute contraindications to
appear to be safe and effective in SLE, the story is not so immunization against Varicella zoster. However, this is clearly
clear-cut for live vaccines. Vaccines such as Varicella zoster an area where additional research is essential.
and measles, mumps, and rubella (MMR) have been consid- The panel did not comment on the additive effects of gluco-
ered contraindicated in patients with SLE and any patient corticoids plus immunosuppressive drugs. AntiTNF agents
receiving more than 10 mg of prednisone a day or other were felt to be too new to be sure of their impact on immuniza-
immunosuppressive agents (O’Neill and Isenberg 2006). It tion against Varicella zoster. One reasonable approach is to dis-
has also been suggested that live vaccines should not be continue TNF inhibitors for 1 month before administering the
given for 3 months after cessation of immunosuppressive Varicella zoster vaccine and to wait 1 month before resuming
drugs. However, in June 2008, the Advisory Committee on TNF inhibition.
Immunization Practices (ACIP) provided some new recom- The situation in lupus is complex. Rheumatoid arthritis,
mendations and raised several new questions. psoriasis, polymyositis, sarcoidosis, and inflammatory bowel
The ACIP now recommends routine vaccination of all per- disease were all listed specifically as conditions in which
sons older than 60 years of age who have no contraindication vaccination might be allowed under appropriate circum-
(ACIP 2008). The recommendation includes individuals with stances. Lupus was not. Not only are lupus patients immuno-
chronic medical conditions. However, the panel felt that mod- suppressed by medications, they may also be significantly
erately or severely immunosuppressed (e.g., those on ³ 20 mg/ immunodeficient because of active lupus and to some degree
day prednisone for >2 weeks) or immunodeficient patients because of the genetic defects associated with lupus.
(e.g., AIDS or HIV with CD4+ T-lymphocyte values < 200 Perhaps, for these reasons the ACIP did not make specific
per mm3 or <15% of total lymphocytes) should not receive recommendations about Varicella zoster immunization in
the vaccine. Lower levels of immunosuppression, e.g., SLE patients. At this time, firm recommendations about
150 M. Petri et al.

Varicella zoster immunization in SLE are not possible. Two a

points appear reasonable:
• Lupus patients whose disease is inactive and who are on
stable doses of prednisone (<15 mg/day) with or without
hydroxychloroquine may be appropriate candidates for
Varicella zoster vaccination.
• Lupus patients with recently active disease should not
receive this vaccination.
Additional studies to document the safety and efficacy of the
zoster vaccine in SLE are needed.
b
Pearl: Pneumocystis carinii (jiroveci) prophylaxis should be
employed in selected patients with SLE.
Comment: Patients with SLE are at increased risk for
Pneumocystis carinii (jiroveci) pneumonia (PCP) (Porges
et al. 1992). In a study of hospitalized patients, there were
approximately 12 cases of PCP for every 10,000 SLE admis-
sions, compared with only 2 cases of PCP for every 10,000
rheumatoid arthritis admissions (odds ratio = 2.5; 95% con-
fidence interval 1.7–3.8) (Ward and Donald 1999).
SLE patients who develop PCP are usually (but not always)
on high doses of glucocorticoids and immunosuppressive
medications such as cyclophosphamide. (Godeau et al. 1994).
Some patients who develop PCP are not on intensive immuno-
suppression at the time of their infection but have been
on intensive immunosuppression in the preceding months
(Suryaprasad 2008). PCP prophylaxis should be considered in
an SLE patient on moderate to high doses of prednisone, par-
ticularly if another immunosuppressive drug is part of the
treatment regimen or if the patient is severely lymphopenic.
Pearl: Lupus patients tolerate trimethoprim-sulfamethoxazole
medications poorly.
Comment: Hypersensitivity to sulfonamides is more common
among patients with SLE when compared with the general
population. In addition, some investigations have indicated
that sulfonamide use in SLE patients is associated with dis- Fig. 13.22 (a, b) Eczema induced by trimethoprim/sulfamethoxazole
ease flares (Petri and Allbritton 1992). Maculopapular skin (TMP/SMX). This lupus patient had been on TMP/SMX for approxi-
mately 2 months before he developed a diffuse cutaneous eruption
eruptions with or without fever are the most common adverse
(nonpalpable purpura on the lower extremities, blanching erythema on
effect of sulfonamides in SLE (Fig. 13.22). Such reactions the back, abdomen, and trunk). The rash resolved following the discon-
are probably type IV (cell-mediated) hypersensitivity tinuation of TMP/SMX (Figure courtesy of Dr. John Stone)
responses (Pichler 2003). The rate of type I hypersensitivity
(IgE-mediated) hypersensitivity responses does not appear to
be markedly increased. These reactive intermediates are more likely to occur in slow
A high frequency of maculopapular rash and fever in acetylators and in the settings in which oxidative stress leads
response to sulfonamides has also been seen in patients to low levels of glutathione, as seen in HIV and SLE (Perl
infected with the human immunodeficiency virus (HIV). The et al. 2004).
pathophysiology for the increase in sulfonamide hypersensi- Despite the increase in risk of hypersensitivity reactions
tivity has not been established. However, metabolism to in HIV, TMP-SMX is still used for the prophylaxis and treat-
reactive intermediates such as hydroxylamines and nitroso- ment of PCP. Indeed, the potential benefit from TMP-SMX
sulfonamides, which leads to either direct cytotoxicity or to has led to the development of desensitization protocols for
haptenization of host proteins, seems likely to play a role. patients with a history of TMP-SMX hypersensitivity. These
13 Systemic Lupus Erythematosus 151

protocols are usually (but not always) successful. Of course, than those prescribed placebo (Buyon et al. 2005). However,
desensitization should not be attempted in patients with severe flares were not more common in the HRT group.
severe reactions such as Stevens–Johnson syndrome, toxic Another study focused specifically on the influence of
epidermal necrolysis, or serum sickness. Among patients HRT on the occurrence of arterial and venous thrombotic
with HIV, the gradual introduction of prophylactic TMP- events in post-menopausal women with SLE (Fernandez et al.
SMX leads to fewer adverse reactions (Para et al. 2000). 2007). That study reported no increase in vascular events
Many similarities exist between patients with HIV and among patients who were antiphospholipid antibody negative
SLE with regard to sulfonamide hypersensitivity. One impor- and had no history of such events. However, data from the
tant difference, however, is the potential association of TMP- Women’s Health Study indicate that HRT increases the throm-
SMX use with disease exacerbation in SLE. Thus, the botic risk in the general female population.
avoidance of sulfonamides for the treatment of routine bacte- Decisions on the use of HRT in patients with SLE must be
rial infections in SLE patients seems a reasonable precau- individualized according to each patient’s risk profile. Data
tion. Whether to use TMP-SMX as opposed to dapsone, from the studies cited above cannot be extrapolated to
atovaquone, or aerosolized pentamidine for PCP prophylaxis patients with high titers of antiphospholipid antibodies or
in SLE remains a controversial issue. If the decision is made previous thrombotic events.
in favor of TMX-SMX, the medication should be introduced
Myth: Statins should be used with caution in SLE patients
gradually over a 2-week period.
because they heighten patients’ risk of muscle injury.
Pearl: The majority of patients with serologically active but
Reality: Lupus patients have a substantially increased risk
clinically quiescent lupus are likely to flare at some point in
for accelerated atherosclerosis. Thus, cardiovascular risk
the future, but the disease flare may not occur for many months
factors, including dyslipidemia, should be sought and man-
or even several years.
aged aggressively. Statins are both effective and generally
Comment: A minority of patients with SLE enter a period in safe in SLE. Severe myopathy affects only about 0.1% of all
which their disease is serologically active but clinically individuals who take statins. Thus, the issue with regard to
quiescent. In the University College London Hospital Cohort, statin use for many patients with SLE is not whether or not to
this state has been defined as an antidsDNA antibody titer use these drugs, but rather how to use them wisely.
greater than 50 units/mL on two occasions but a global Patients with significant renal insufficiency or hypothy-
BILAG score of less than 6 for at least 6 months. roidism are at increased risk of skeletal muscle toxicity from
In one cohort of patients with serologically active but statin use. Because SLE patients have a higher than normal
clinically quiescent disease, 81% suffered a disease flare likelihood of both renal and thyroid dysfunction, these factors
within 5 years (WalzLeBlanc et al. 1994; Ng et al. 2006). should be considered before prescribing statins. Because some
Approximately half of those patients had multiple flares dur- patients with connective tissue disease have low CK values, a
ing that time. The mean duration to first disease flare was 15 normal serum CK in the presence of symptoms and signs of
months (range: 2–46 months). Shorter times to disease flare myositis does not exclude the possibility of muscle injury.
correlated with high titers of antinucleosome antibodies and In the absence of clinical symptoms and weakness, a CK
antidsDNA antibody titers that were five times above the level more than three times the upper limit of normal that is
normal limit (Ng et al. 2006). attributed to statin use is an indication for discontinuing the
medication. Patients should drink large quantities of fluids to
Myth: Hormone replacement therapy (HRT) causes flares in
facilitate the renal excretion of CK. After the CK has returned
patients with SLE.
to baseline, patients may be tried on a statin less likely to
Reality: The safety of exogenous estrogens in SLE patients cause muscle toxicity. Pravastatin and fluvastatin are both
remains a controversial topic. Estrogen use in SLE has spe- less likely to cause muscle injury than is lovastatin.
cial relevance because of the premature ovarian failure that
Pearl: Diffuse or focal myocarditis can mimic myocardial
these patients often experience as a result of treatment, the
infarction.
elevated risk of osteoporosis, and the rapid acceleration of
atherosclerosis to which lupus patients are prone. Symptoms Comment: The first order of business in a lupus patient with
of estrogen deficiency are often particularly severe in younger chest pain, regardless of the patient’s age, is to exclude active
women and frequently require long-term HRT. coronary ischemia. Patients with SLE can and do develop
Clinicians often avoid HRT in patients with SLE because coronary arteritis, leading to myocardial infarction.
the fear of inducing disease flares. The Safety of Estrogens in However, in young patients with lupus who present with
Lupus Erythematosus National Assessment (SELENA) trial symptoms, signs, and laboratory findings consistent with acute
found that patients who are prescribed HRT were more likely coronary ischemia (e.g., angina, ST segment elevations on
to experience mild to moderate disease flares more often electrocardiography, and CK-MB or troponin concentration
152 M. Petri et al.

elevation), the clinician must bear in mind the possibility of • Hydroxychloroquine is the drug of first choice for cutaneous
myocarditis. Cardiac scintigraphy or magnetic resonance imag- SLE. For skin lupus refractory to hydroxychloroquine
ing can reveal diffuse or focal myocarditis in such patients. alone, the addition of quinacrine can be helpful.
Pearl: Fifty percent of SLE patients who have a lupus antico- Methotrexate and mycophenolate mofetil are useful second-
agulant at diagnosis will suffer a thrombotic event. line agents for cutaneous lupus. However, mycophenolate
mofetil may require many months or even more than 1 year
Comment: This is true, and the lupus anticoagulant is a more
before its benefits in chronic cutaneous lupus are evident.
powerful predictor of thrombotic risk than is anticardiolipin
Patients should be counseled about this.
antibody (Somers et al. 2002; Wahl et al. 1997). However, the
50% risk of thrombosis associated with lupus anticoagulants • Thalidomide can be added for recalcitrant discoid lupus, but
is cumulative, occurring over 20 years. Thus, although empiric its toxicity profile (teratogenicity, neuropathy, premature
prophylactic therapy is worth considering, the choice of medi- gonodal failure, and thrombosis) makes its use unusual.
cation is determined by long-term safety concerns. The use of
Pearl: Hydroxychloroquine should be put in the water
aspirin makes intuitive sense and has a better side-effect pro-
supply.
file for most patients than does warfarin, but two clinical trials
failed to demonstrate benefit of aspirin in the prevention of Comment: At least in the water imbibed by lupus patients.
thrombotic events (Erkan et al. 2007; Ginsburg et al. 1992). Hydroxychloroquine has gained new respect as a preventive
Despite the absence of evidence supporting its use, the agent in SLE. Experienced lupus clinicians refer to this
absence of evidence is not evidence of absence: many clini- medication as “lupus health insurance”. Hydroxychloroquine
cians use a daily baby aspirin to treat patients with antiphos- reduces the frequency of SLE flares and probably also reduces
pholipid antibodies who have clinical manifestations of the the “spread” of lupus to renal disease (Tsakonas et al. 1998;
antiphospholipid syndrome. Some data also indicate that Fessler et al. 2005). Furthermore, it diminishes the likelihood
hydroxychloroquine prevents antiphospholipid antibody- of thrombosis caused by antiphospholipid antibodies (Petri
mediated thrombosis (Petri 1996; Pierangeli and Harris 1996; Pierangeli and Harris 1996). Finally, the drug may have
1996). effects that are synergistic with those of other SLE medica-
The avoidance of medications that foster hypercoagula- tions. In one study of patients with lupus nephritis, patients
bility is also important in managing patients at risk for throm- who remained on hydroxychloroquine doubled their chance of
bosis. Such medications include oral contraceptives, hormone responding to mycophenolate mofetil (Kasitanon et al. 2006).
replacement therapy, selective estrogen receptor modulators, Concern about hydroxychloroquine retinopathy is over-
thalidomide, and doses of erythropoietin that lead to overcor- blown. This retinal complication of this medication occurs
rection of anemia. quite rarely (on the order of 1 in 5,000 patients on long-term
therapy), develops slowly, and is reversible if the medication
is stopped promptly. A safe recommendation is to insist that
13.8 Treatment patients treated with hydroxychloroquine have an annual
ophthalmology examination.
Pearl: Immunosuppressive drug regimens in SLE are selected The dose of hydroxychloroquine should be reduced in
according to the pattern of organ involvement. patients with significant renal dysfunction, in the elderly, and
in patients with a low body mass index (e.g., children). The
Comment: Clinicians who treat many patients with lupus
safety profile of chloroquine is less clear. Patients taking
tend to have “favorite” regimens that are prescribed accord-
chloroquine should have more regular ophthalmic examina-
ing to the specific organs affected. Some generalizations are
tions (Marmor et al. 2002).
possible:
Pearl: Cigarette smoking is associated with an increased
• Mycophenolate mofetil is the preferred medication for
risk of cutaneous lupus and decreased responsiveness to
lupus nephritis because of its therapeutic equivalency
antimalarial therapy.
with cyclophosphamide and its lower rate of adverse
effects (Chan et al. 2000; Contreras et al. 2004; Ginzler Comment: Patients who smoke are more likely to have cuta-
et al. 2005). neous lupus. In addition, there is good evidence that patients
• Azathioprine is probably equivalent to mycophenolate who are therapeutically resistant to the usual therapies, includ-
mofetil as a remission maintenance agent in lupus nephri- ing antimalarials, immunosuppressive agents, and thalido-
tis (Contreras et al. 2004). mide are frequently smokers (Moghadam-Kia et al., in press;
• Methotrexate and leflunomide are the preferred agents for Jewell and McCauliffe 2000). Smoking cessation should be
the treatment of arthritis in SLE (Sato 2001; Tam et al. urged for a variety of reasons in SLE. Better control of skin
2004). disease is one of them (Miot et al. 2005).
13 Systemic Lupus Erythematosus 153

Pearl: Quinacrine is a helpful adjunctive therapy when com- Myth: Ultrapotent topical glucocorticoids should never be
bined with hydroxychloroquine or chloroquine for the treat- used on the face for cutaneous lupus.
ment of cutaneous lupus.
Reality: The truth is that dermatologists almost always use
Comment: Quinacrine (100 mg/day) has an effect on cutane- ultrapotent topical glucocorticoids for chronic cutaneous lupus,
ous lupus that is synergistic with that of hydroxychloroquine even on the face. Topical glucocorticoids generally work very
(Feldman et al. 1994). The usual approach to cutaneous lupus well in early lesions. If this strategy is employed early enough,
is start hydroxychloroquine at a dose of <7.5 mg/kg/day. If one can clear DLE lesions completely, leaving neither a scar
there is insufficient improvement after 2 months, then quina- nor a pigment abnormality. Patients should be monitored
crine is added. Two additional months may be required for closely by a dermatologist to avoid adverse reactions.
improvement to occur after the addition of quinacrine. Potential side effects of chronic application of topical glu-
Recent studies suggest that hydroxychloroquine works by cocorticoids are classified into the categories of systemic and
blocking Toll-like receptor (TLR) 9. Quinacrine appears to local. Systemic side effects include suppression of the hypo-
achieve its effects at least in part by the blockade of other thalamic–pituitary–adrenal axis, iatrogenic Cushing’s syn-
TLRs, e.g., TLR7, and inhibition of dendritic cell migration drome, and growth retardation (infants and children). Side
(Kalia and Dutz 2007; Gorbache et al. 2007). effects at the site of application include skin atrophy and
Quinacrine must be obtained in the United States now striae, hypopigmentation, steroid-induced acne or rosacea,
through compounding pharmacies. The main side effects of glaucoma, and cataracts.
this medication are reversible discoloration of the skin and Acne and slight hypopigmentation are the most common
sclerae, occasional drug eruptions, and hyperpigmentation side effects of topical glucocorticoid therapy. Acne is gener-
caused by the deposition of drug in the skin. ally managed easily with topical creams or a short course of
If patients do not respond to the combination of hydroxy- systemic tetracycline antibiotics. Hypopigmentation is revers-
chloroquine and quinacrine, then hydroxychloroquine is fre- ible on discontinuation of the topical glucocorticoid.
quently discontinued in favor of chloroquine. The higher
likelihood of adverse effects with chloroquine (retinopathy, Pearl: The topical calcineurin inhibitors, tacrolimus and
myopathy, and cardiomyopathy) compared with hydroxy- pimecrolimus, are effective in cutaneous lupus.
chloroquine is the major reason for reserving chloroquine for Comment: These topical agents are used for a variety of inflam-
patients who have failed the combination of hydroxychloro- matory dermatoses, including atopic dermatitis. They also
quine and quinacrine. appear to be efficacious in the treatment of cutaneous lupus.
Pearl: Patients who develop a drug exanthem to hydroxy- Tacrolimus and pimecrolimus are superior to cyclosporine for
chloroquine frequently are able to tolerate chloroquine well. topical use because they are smaller molecules that have sig-
nificantly greater cutaneous penetration.
Comment: Possible drug reactions to the antimalarial agents In one clinical trial of 20 patients with cutaneous SLE that
include exanthems, urticaria, hyperpigmentation, alopecia, compared topical tacrolimus ointment (0.1%) to clobetasol
and lichenoid drug eruptions. Patients who develop an exan- proprionate ointment (0.05%), no significant difference in
them with hydroxychloroquine should stop the medication. efficacy was observed between the two treatments (Tzung
Once the eruption resolves, it is often possible to treat the et al. 2007). However, 61% of the patients treated with clo-
patient with chloroquine without triggering a recurrence of betasol developed telangiectasia when compared with none
the skin rash. in the tacrolimus group.
In contrast, patients who develop urticarial reactions to
hydroxychloroquine frequently demonstrate cross-reactions Pearl: Many SLE flares are due to noncompliance.
with chloroquine treatment. Therefore, such patients should Comment: No one wants to have a chronic disease. Denial
not be switched routinely from hydroxychloroquine to chlo- and medical non-compliance are greater problems with SLE
roquine. In addition, patients with cutaneous lupus or der- than with chronic diseases that afflict middle-aged and
matomyositis can develop a lichenoid drug eruption when elderly individuals, because SLE affects teenagers and young
treated with antimalarials. Biopsy of this reaction can be dif- adults. One study showed that 51% of SLE patients prescribed
ficult to distinguish from the underlying lupus or dermato- hydroxychloroquine were not taking it (Koneru et al. 2007)!
myositis, but eosinophils in the infiltrate of a skin biopsy Noncompliance with prednisone might be even worse,
point toward a lichenoid drug eruption rather than a rash because of the acne, moon facies, and weight associated with
related to the underlying disease (Geraminejad et al. 2004). that medication.
An empiric trial of discontinuing a medication is often
required to determine if a rash has been a result of the drug Pearl: The development of end-stage renal disease in SLE is
or the disease. linked to poor medical compliance in many patients.
154 M. Petri et al.

Table 13.6 Prognostic factors in lupus nephritis (Adapted from triamcinolone should be administered only in the buttocks
Faurschou 2006) because it can cause focal lipodystrophy.
• Duration of glomerulonephritis signs for more than 6 months
before biopsy Pearl: Don’t be in too much of a hurry to stop the steroids
• Serum creatinine greater than XXXX mg/dL (140 mmol/L) altogether.
• Histopathological findings:
– Diffuse proliferative glomerulonephritis (WHO class IV) Comment: When lupus has gone into clinical remission,
– Tubular atrophy there is a tendency on the part of both the patient and the
• Treatment regimens physician to want to stop the glucocorticoids. However,
• Ethnicity stopping the patient’s prednisone too quickly often triggers
a disease flare. One approach to preventing this is to con-
tinue glucorticoids at a stable low dose for at least 1 year.
Comment: Up to 60% of patients with SLE in some centers Prednisolone 5 mg/day is a reasonable lowest dose for remis-
develop renal involvement, and up to 20% eventually develop sion maintenance. Once the decision is undertaken to taper
end-stage renal disease (Faurschou et al. 2006). The course glucocorticoids, the rate of decrease should not exceed 1 mg/
of lupus nephritis is affected by multiple clinical, demo- day per month.
graphic, serological, and histopathological factors. Some
Pearl: If glucocorticoids are not working, consider potential
major predictors of renal outcome in lupus nephritis are
pharmacologic issues.
shown in Table 13.6.
When a patient does not respond as predicted, it is worth Comment: Approximately 25% of all cases of glucocorticoid-
considering non-adherence to treatment. Treatment non- resistant asthma are explained by either poor absorption or
compliance is underrecognized as a negative prognostic fac- rapid clearance (Nimmagadda et al. 1996). Poor absorption is
tor in lupus nephritis. Convincing some patients with unusual and can sometimes be explained by interaction with
glomerulonephritis to take prednisone is difficult because the other medications, e.g. antacids. Among patients with asthma,
organ manifestation causes no symptoms that are obvious to abnormal clearance of glucocorticoids is a more common
the patient, yet the adverse effects of therapy are highly pre- explanation for low glucocorticoid levels in the blood. If no
dictable (Bruce et al. 2000; Mok 2005; Adler et al. 2006; explanation for poor absorption is found, then switching from
Petri et al 1991a). Between 30% and 50% of patients with prednisone to methylprednisone or to liquid prednisone or
lupus nephritis do not adhere to their prescribed treatment liquid prednisolone can be helpful.
regimens (Bruce et al. 2000; Petri 1991a). Glucocorticoid metabolism has not been evaluated exten-
sively in SLE patients, perhaps because non-glucocorticoid
Pearl: Consider a short “burst” of glucocorticoids to man-
immunosuppressive agents are usually effective in treating
age mild to moderate SLE flares.
patients who do not respond to glucocorticoids alone. However,
Comment: A common approach to the management of SLE metabolic problems with glucocorticoids can be caused by
patients is to maintain a low dose of glucocorticoids over some of the medications used commonly in SLE, particularly
extended periods of time. This is the “maintenance steroids” anticonvulsant agents and medications that upregulate hepatic
approach. In contrast, in other autoimmune diseases such as p450 enzymes.
multiple sclerosis, which is also often characterized by a The effects of these drugs in SLE have not been studied,
“relapsing-remitting” course, glucocorticoids are usually not but flares or treatment failures have been found in other dis-
maintained between flares (Barr et al. 1999). Because the orders such as asthma, giant cell arteritis, and renal trans-
average SLE patient has one flare per year, maintenance of the plantation (Brooks et al. 1972; Carrie et al. 1994; Wassner
patient of prednisone during these interval periods may be et al. 1976). The effect of drugs that induce p450 enzymes
unnecessary in some patients (Petri et al. 1991b). on glucocorticoid metabolism can be profound, with a
The FLOAT study (Flares in Lupus: Outcomes Assessment reduction in area under the curve of blood levels by approx-
Trial) evaluated the use of glucocorticoid “bursts” in the imately 60%. Thus, the co-administration of such medica-
management of mild to moderate disease flares (Danowski tions with glucocorticoids could require a doubling of the
et al. 2006). Patients were assigned randomly to receive glucocorticoid dose in order to maintain effective blood
either oral methylprednisolone dose packs or a single intra- levels.
muscular injection of triamcinolone. Neither group received
Myth: Osteonecrosis is caused by doses of prednisone higher
maintenance glucocorticoids between flares. The majority of
than 40–60 mg/day.
patients in both arms achieved rapid resolution of their SLE
activity. Reality: The threshold dose of prednisone for the risk of
Intramuscular triamcinolone ensures 100% compliance osteonecrosis is only 20 mg/day (Petri 1995). Other poten-
with the glucocorticoid regimen. However, intramuscular tial risk factors are ethnicity (African–Americans are at
13 Systemic Lupus Erythematosus 155

greater risk), vasculitis, Raynaud’s phenomenon, and the mild to moderate proliferative and/or membranous lupus
presence of a hypercoagulable state such as the antiphos- nephritis. Failure to achieve complete remission within the
pholipid syndrome. first 4–6 months should evoke discussions about switching
to the combination of pulse therapy of glucocorticoids with
Myth: In proliferative lupus nephritis, a “wait and see” cyclophoshamide (Ioannidis et al. 2000; Boumpas et al.
approach can save the patient from the toxicity of immuno- 2005).
suppressive drugs.
Myth: Medications that cause drug-induced lupus cannot be
Reality: Some clinicians, arguing that cyclophosphamide used in idiopathic SLE.
saves kidneys but not lives, delay the use of cytotoxic therapy
Reality: The “big three” medications linked to drug-induced
in patients with proliferative lupus nephritis. Unfortunately,
lupus are isoniazid, procainamide, and hydralazine. Contrary
delay in cytotoxic therapy for such patients diminishes the
to popular opinion, these medications can be used in idio-
likelihood of remission and increases the risk of subsesquent
pathic SLE. There is no evidence that any of them exacer-
relapses and end-stage renal disease. Moreover, short-term
bates SLE.
induction courses (e.g., 6 months) without follow-up mainte-
TNF inhibitors have also been implicated in causing
nance therapy increases the risk of relapse.
drug-induced lupus. The story of TNF inhibition and lupus
Serious lupus nephritis needs to be identified quickly and
remains to be told fully. Although TNF inhibitors can induce
treated appropriately for a sufficient duration of time. For the
ANA, antidsDNA, and anticardiolipin antibodies, these
majority of patients, this means either mycophenolate mofetil
medications may have a role in the treatment of some lupus
or intravenous cyclophosphamide.
patients (Aringer et al. 2004). Great caution is urged when
Myth: Patients with lupus nephritis should receive the full using TNF inhibitors in patients with SLE, because the full
NIH regimen of intravenous cyclophosphamide. implications of elevations in anticardiolipin and antidsDNA
antibody titers remain unclear (Fusconi et al. 2007; Jonsdottir
Reality: Few patients with lupus nephritis require the full
et al. 2004).
duration of the NIH cyclophosphamide regimen, which
called for treating patients for a minimum of 2 years (Austin Myth: The management of lupus is empiric and varies widely
et al. 1986). Patients with moderate lupus nephritis can be among practicing physicians.
treated with glucocorticoids and mycophenolate mofetil Reality: Although there are several areas in the management
(Appel et al., in press). of lupus that require further investigation, reasonable recom-
mendations exist based on a combination of evidence and
Pearl: A sizeable subset of patients with lupus nephritis can
expert opinion (Bertsias et al. 2008). The EULAR recom-
be treated with azathioprine.
mendations for the management of SLE cover general man-
Comment: Not all SLE patients with glomerulonephritis need agement as well as selected aspects of lupus such as the
cyclophosphamide (MacGowan et al. 2002; Grootscholten antiphospholipid syndrome, pregnancy, neuropsychiatric
et al. 2006). Patients with mild renal disease can be treated lupus, and nephritis. The recommendations provide a general
with the combination of glucocorticoids and azathioprine framework for the management of lupus yet allow latitude
(Glas-Vos et al. 1995; Nossent and Koldingsnes 2000). for physician autonomy and patient preferences.

Myth: With the introduction of mycophenolate mofetil, cyclo-

phosphamide has become an “obsolete” therapy for SLE.
References
Reality: The LUNAR trial of rituximab for lupus nephritis
showed no added benefit over mycophenolate mofetil.
Abu-Shakra M, Zalmanson S, Neumann L, Flusser D, Sukenik S,
In spite of earlier trials purporting to show that mycopheno- Buskila D. Influenza virus vaccination of patients with systemic
late mofetil is superior to cyclophoshamide as induction lupus erythematosus: Effects on disease activity. J Rheumatol 2000;
therapy for proliferative lupus nephritis, this superiority has 27(7):1681–1685
not been confirmed in subsequent studies. No long-term fol- ACR ad hoc committee on neuropsychiatric lupus. The American College
of Rheumatology nomenclature and case definitions for neuropsychi-
low-up studies are available on patients treated with myco- atric lupus syndrome. Arthritis Rheum 1999; 42:599–608
phenolate mofetil, and the efficacy of mycophenolate mofetil Adler M, Chambers S, Edwards C, Neild G, Isenberg D. An assessment
in patients with severe lupus nephritis (i.e., patients with cre- of renal failure in an SLE cohort with special reference to ethnicity,
sents, fibrinoid necrosis, or substantial impairment of renal over a 25-year period. Rheumatology 2006; 45:1144–1147
Advisory Committee on Immunization Practices (ACIP). Recommenda-
fuction) remains to be demonstrated. tions for the Prevention of Herpes Zoster Morbidity and Mortality
Mycophenolate mofetil is a good alternative to pulse Weekly Report Recommendations and Reports June 6, 2008/Vol. 57/
cyclophosphamide as induction therapy for patients with RR-5(www.cdc.gov/mmwr)
156 M. Petri et al.

Aguilera P, Mascaró JM Jr, Martinez A, et al Cutaneous gamma/delta Cervera R, Khamashta M, Font J, et al Systemic lupus erythematosus:
T-cell lymphoma: A histopathologic mimicker of lupus erythemato- Clinical and immunologic patterns of disease expression in a cohort
susprofundus (lupus panniculitis). J Am AcadDermatol 2007; 56: of 1,000 patients. The European Working Party on Systemic Lupus
643–647 Erythematosus. Medicine 1993; 72:113–124
Ainiala H, Loukkola J, Peltola J, Korpela M, Hietaharju A. The preva- Chambers S, Charman S, Rahman A, Isenberg D. Development of addi-
lence of neuropsychiatric syndromes in systemic lupus erythemato- tional autoimmune diseases in a multiethnic cohort of patients with
sus. Neurology 2001; 57(3):496–500 systemic lupus erythematosus with reference to damage and mortality.
Akkasilpa S, Goldman D, Magder LS, Petri M. Number of fibromyalgia Ann Rheum Dis 2007; 66:1173–1177
tender points is associated with health status in patients with sys- Chan MT. Owen P. Dunphy J, et al Associations of erosive arthritis with
temic lupus erythematosus. J Rheumatol 2005; 32(1):48–50 anti-cyclic citrullinated peptide antibodies and MHC Class II alleles
Appel G, Contreras G, MA D, et al Mycophenolate Mofetil versus in systemic lupus erythematosus. J Rheumatol 2008; 35(1): 77–83
Cyclophosphamide as Lupus Nephritis Induction Treatment. J Am Chan TM, Li FK, Tang CS, et al Efficacy of mycophenolate mofetil in
Soc Nephrol 2009; 20(5):1103–12 patients with diffuse proliferative lupus nephritis. Hong Kong-
Aringer M, Graninger WB, Steiner G, Smolen JS. Safety and efficacy Guangzhou Nephrology Study Group. N Engl J Med 2000; 343
of tumor necrosis factor alpha blockade in systemic lupus ery- (16):1156–1162
thematosus: An open-label study. Arthritis Rheum 2004; 50(10): Christopher-Stine L, Siedner M, Lin H, Haas M, Parekh H, Petri M,
3161–3169 et al Renal biopsy in lupus patients with low levels of proteinuria.
Austin H, Klippel J, Balow J, et al Therapy of lupus nephritis. Controlled J Rheumatol 2007; 34: 332–335
trial of prednisone and cytotoxic drugs. NEJM 1986; 314:614–619 Chrysostomou A, Becker G. Spironolactone in addition to ACE inhibi-
Baca V, Lavalle C, Garcia R, et al Favorable response to intravenous tion to reduce proteinuria in patients with chronic renal disease.
methylprednisolone and cyclophosphamide in children with severe N Engl J Med 2001; 345(12):925–926
neuropsychiatric lupus. J Rheumatol 1999; 26:432–439 Contreras G, Pardo V, Leclercq B, Lenz O, Tozman E, O’Nan P, et al
Barr S, Zonana-Nacach A, Magder L, Petri M. Patterns of disease activ- Sequential therapies for proliferative lupus nephritis. N Engl J Med
ity in systemic lupus erythematosus. Arthritis Rheum 1999; 42: 2004; 350(10):971–980
2682–2688 Costallat L, Coimbra A. Systemic lupus erythematosus: Clinical and
Bertsias G, Ioannidis JP, Boletis J, et al Task Force of the EULAR laboratory aspects related to age at disease onset. Clin Exp Rheumatol
Standing Committee for International Clinical Studies Including 1994; 12:603–607
Therapeutics. EULAR recommendations for the management of Cusack C, Danby C, Fallon JC, et al Photo protective behaviour and
systemic lupus erythematosus. Report of a Task Force of the EULAR sunscreen use: Impact on vitamin D levels in cutaneous lupus erythe-
Standing Committee for International Clinical Studies Including matosus. Photodermatol Photoimmunol Photomed 2008; 24:
Therapeutics. Ann Rheum Dis 2008; 67(2):195–205 260–267
Bijl M, Kallenberg CG. Ultraviolet light and cutaneous lupus. Lupus Damian-Abrego G, Cabiedes J, Cabral A. Anti-citrullinated peptide
2006; 15(11):724–727 antibodies in lupus patients with or without deforming arthropathy.
Boddaert J, Huong D, Zahir A, Wechsler B, Godeau P, Piette J. Late- Lupus 2008; 17:300–304
onset systemic lupus erythematosus: A personal series of 47 patients Danowski A, Magder L, Petri M. Flares in lupus: Outcome Assessment
and pooled analysis of 714 cases in the literature. Medicine 2004; Trial (FLOAT), a comparison between oral methylprednisolone and
83:348–359 intramuscular triamcinolone. J Rheumatol 2006; 33(1):57–60
Bombardier C, Gladman DD, Urowitz M, Caron D, Chang CH, the Domenech I, Aydintug O, Cervera R, et al Systemic lupus erythemato-
Committee on Prognosis Studies in SLE. Derivation of the SLEDAI: sus in 50 year olds. Postgrad med J 1992; 68:440–444
A disease activity index for lupus patients. Arthritis Rheum 1992; Elkayam O, Paran D, Caspi D, et al Immunogenicity and safety of
35:630–640 pneumococcal vaccination in patients with rheumatoid arthritis or
Boumpas D, Yamada H, Patronas N, Scott D, Klippel J, Balow J. Pulse systemic lupus erythematosus. Clin Infect Dis 2002; 34:147–153
cyclophosphamide for severe neuropsychiatric lupus. Q J Med Elkayam O, Paran D, Burke M, et al Pneumococcal vaccination of
1991; 81:975–984 patients with systemic lupus erythematosus: Effects on generation
Boumpas DT, Sidiropoulos P, Bertsias G. Optimum therapeutic of autoantibodies. Autoimmunity 2005; 38:493–496
approaches for lupus nephritis: what therapy and for whom? Nat Erkan D, Harrison MJ, Levy R, Peterson M, Petri M, Sammaritano L,
Clin Pract Rheumatol 2005; 1(1):22–30 et al Aspirin for primary thrombosis prevention in the antiphospho-
Brooks SM, Werk EE, Ackerman SJ, et al Adverse effects of phenobar- lipid syndrome: A randomized, double-blind, placebo-controlled
bital on corticosteroid metabolism in patients with bronchial asthma. trial in asymptomatic antiphospholipid antibody-positive individu-
N Engl J Med 1972; 286:1125–1128 als. Arthritis Rheum 2007; 56(7):2382–2391
Bruce I, Gladmann D, Urowitz M. Factors associated with refractory Esdaile JM, Abrahamowicz M, Grodzicky T, Li Y, Panaritis C, du
renal disease in patients with systemic lupus erythematosus: The role Berger R, et al Traditional Framingham risk factors fail to fully
of patient non-adherence. Arthritis Care Res 2000; 13:406–408 account for accelerated atherosclerosis in systemic lupus erythema-
Buyon J, Petri MA, Kim MY, et al The effect of combined estrogen and tosus. Arthritis Rheum 2001; 44(10):2331–2337
progesterone hormone replacement therapy on disease activity in Faurschou M, Starklint H, Halberg P, Jacobsen S. Prognostic factors in
systemic lupus erythematosus: A randomized trial. Ann Intern Med lupus nephritis: Diagnostic and therapeutic delay increases the risk
2005; 142:953–962 of terminal renal failure. J Rheumatol 2006; 33:1563–1569
Callen JP. Cutaneous lupus erythematosus: A personal approach to Feldman R, Salmon D, Saurat JH. The association of the two antimalari-
management. Australas J Dermatol 2006; 47(1):13–27 alschloroquine and quinacrine for treatment-resistantchronic and sub-
Calvo-Alen J, Alarcon GS, Burgard SL, et al Systemic lupus erythema- acutecutaneous lupus erythematosus. Dermatology 1994;186:425–427
tosus: Predictors of its occurrence among a cohort of patients with Fernandez M, Calvo-Alen J, Bertoli A, et al Systemic lupus erythemato-
early undifferentiated connective tissue disease: Multivariate analy- sus in a multiethnic US cohort (LUMINA L II): Relationship between
ses and identification of risk factors. J Rheumatol 1996; 23:469–475 vascular events and the use of hormone replacement therapy in post-
Carrie F, Roblot P, Bouquet S, et al: Rifampin-induced nonresponsive- menopausal women. J Clin Rheumatol 2007; 13:261–265
ness of giant cell arteritis to prednisone treatment. Arch Intern Med Fernandez-Nebro A, Palacios-Munoz R, Gordillo J, Abarca-Costalago
1994; 154:1521–1524 M, De Haro-Liger M, Rodriguez-Andreu J, et al Chronic or recur-
13 Systemic Lupus Erythematosus 157

rent headache in patients with systemic lupus erythematosus: A case Ho A, Magder LS, Barr SG, Petri MA. A decrease in complement is
control study. Lupus 1999; 8(2):151–156 associated with increased renal and hematologic flares in patients
Fessler B. Infectious diseases in systemic lupus erythematosus:risk fac- with systemic lupus erythematosus. Arthritis Rheum 2001b; 44:
tors, management and prophylaxis. Best Pract Res Clin Rheumatol 2350–2357
2002; 16:281–291 Hochberg MC. Updating the American College of Rheumatology
Fessler BJ, Alarcon GS, McGwin G, Jr., Roseman J, Bastian HM, revised criteria for the classification of systemic lupus erythemato-
Friedman AW, et al Systemic lupus erythematosus in three ethnic sus [letter]. Arthritis Rheum 1997; 40:1725
groups: XVI. Association of hydroxychloroquine use with reduced Huong DL, Papo T, Beaufuls H, Wechsler B. Renal involvement in sys-
risk of damage accrual. Arthritis Rheum 2005; 52(5):1473–1480 temic lupus erythematosus: A study of 180 patients from a single
Font J, Pallares L, Cervera R, et al Systemic lupus erythematosus in the center. Medicine (Baltimore) 1999; 78:148
elderly: Clinical and immunological charracteristics. Ann Rheum Iaboni A, Ibanez D, Gladman DD, Urowitz MB, Moldofsky H. Fatigue
Dis 1991; 50:702–705 in systemic lupus erythematosus: Contributions of disordered sleep,
Foote R, Kimbrough S, Stevens J. Lupus myositis. Muscle Nerve 1982; sleepiness, and depression. J Rheumatol 2006; 33(12):2453–2457
5:65–68 Iliopoulos A, Tsokos G. Immunopathogenesis and spectrum of infec-
Formiga F, Moga I, Pac M, Mitjavila F, Rivera A, Pujol R. Mild presen- tions in systemic lupus erythematosus. Semin Arthritis Rheum
tation of systemic lupus erythematosus in elderly patients assessed 1996; 25:318–336
by SLEDAI. Lupus 1999; 8:462–465 Ioannidis JP, Boki KA, Katsorida ME, et al Remission, relapse, and re-
Fusconi M, Vannini A, Dall’aglio AC, Pappas G, Bianchi FB, Zauli D. remission of proliferative lupus nephritis treated with cyclophosph-
Etanercept and infliximab induce the same serological autoimmune amide. Kidney Int 2000; 57(1):258–264
modifications in patients with rheumatoid arthritis. Rheumatol Int Irving KS, Sen D, Tahir H, et al A comparison of autoimmune liver
2007; 28(1):47–49 disease in juvenile and adult populations with systemic lupus ery-
Ganczarczyk L, Urowitz MB, Gladman DD. Latent lupus. J Rheumatol thematosus-a retrospective review of cases. Rheumatology (Oxford)
1989; 16:475–478 2007; 46(7):1171–1173
Geraminejad P, Stone MS, Sontheimer RD. Antimalarial lichenoid tis- Isenberg D, Horsfall A. Systemic lupus erythematosus-adult onset. In:
sue reactions in patients with pre-existing lupus erythematosus. Maddison P, Isenberg D, Woo P, Glass D (eds) Oxford textbook of
Lupus 2004; 13:473–477 rheumatology. Oxford, Oxford University Press, 1998, pp. 1145–1180
Gilland WR, Tsokos GC. Prophylactic use of antibiotics and immunisa- Jewell ML, McCauliffe DP. Patients with cutaneous lupus erythemato-
tions in patients with SLE. Ann Rheum Dis 2002 Mar; 61(3):191–192 sus who smoke are less responsive to antimalarial treatment. J Am
Gilliam JN, Cheatum DE, Hurd ER, Stastny P Ziff M. Immunoglobulin AcadDermatol 2000; 42:983–987
in clinically uninvolved skin in systemic lupus erythematosus: Jonsdottir T, Forslid J, van Vollenhoven A, Harju A, Brannemark S,
Association with renal disease. J clin invest 1974; 53(5): Klareskog L, et al Treatment with tumour necrosis factor alpha
1434–1440 antagonists in patients with rheumatoid arthritis induces anticardio-
Ginsburg KS, Liang MH, Newcomer L, et al Anticardiolipin antibodies lipin antibodies. Ann Rheum Dis 2004; 63(9):1075–1078
and the risk for ischemic stroke and venous thrombosis. Ann Intern Jorizzo JL, Salisbury PL, Rogers RS 3rd, Goldsmith SM, Shar GG,
Med 1992; 117:997–1002 Callen JP, Wise CM, Semble EL White WL. Oral lesions in systemic
Ginzler EM, Dooley MA, Aranow C, Kim MY, Buyon J, Merrill JT, lupus erythematosus. Do ulcerative lesions represent a necrotizing
et al Mycophenolate mofetil or intravenous cyclophosphamide for vasculitis?. J Am Acad Dermatolo 1992; 27(3):389–394
lupus nephritis. N Engl J Med 2005; 353(21):2219–2228 Kalia S, Dutz JP. New concepts in antimalarial use and mode of action
Gladman D, Hussain F, Ibanez D, Urowitz M. The nature and outcome of in dermatology. DermatolTher 2007; 20:160–174
infection in systemic lupus erythematosus. Lupus 2002; 11:234–239 Karpatkin S. Autoimmune thrombocytopaenic purpura. Semin Haematol
Glas-Vos Jd, Krediet R, Weening J, Arisz L. Treatment of proliferative 1985; 22:260–288
lupus nephritis with methylprednisolone pulse therapy and oral aza- Kasitanon N, Fine DM, Haas M, Magder LS, Petri M. Hydroxychloroquine
thioprine. Neth J Med 1995; 46:4–14 use predicts complete renal remission within 12 months among
Godeau B, Coutant-Perronne V, Le Thi Huong D, et al Pneumocystis patients treated with mycophenolate mofetil therapy for membra-
carinii pneumonia in the course of connective tissue disease: Report nous lupus nephritis. Lupus 2006; 15(6):366–370
of 34 cases. J Rheumatol 1994; 21(2):246–251 Klippel JH, Karsh J, Stahl NI, et al A controlled study of pneumococcal
Gorbache AV, Gasparian AV, Gurova KV, et al Quinacrine inhibits the polysaccharide vaccine in systemic lupus erythematosus. Arthritis
epidermal dendritic cell migration initiating T cell-mediated skin Rheum 1979; 22:1321–1325
inflammation. Eur J Immunol 2007; 37:2257–2267 Koneru S, Shishov M, Ware A, Farhey Y, Mongey AB, Graham TB,
Graninger, Smolen JS. Incomplete lupus erythematosus: Results of a et al Effectively measuring adherence to medications for systemic
multicentre study under the supervision of the EULAR Standing lupus erythematosus in a clinical setting. Arthritis Rheum 2007;
Committee on International Clinical Studies Including Therapeutic 57(6):1000–1006
Trials (ESCISIT). Rheumatology (Oxford) 2001; 40:89–94 Kroshinsky D, Stone JH, Bloch D, Stone JR. A 47-year-old woman
Grootscholten C, Ligtenberg G, Hagen E, et al Azathioprine/methyl- with a nodular rash and numbness and pain in the legs. New England
prednisolone versus cyclophosphamide in proliferative lupus nephri- Journal of Medicine 2009; 360(7):711–20
tis. A randomized controlled trial. Kidney Int 2006; 70:732–742 Lee SS, Singh S, Magder LS, Petri M. Predictors of high sensitivity
Hesselink D, Aarden L, Swaak A. Profiles of the acute-phase reactants C-reactive protein levels in patients with systemic lupus erythema-
C-reactive protein and ferritin related to the disease course of tosus. Lupus 2008; 17:114–23
patients with systemic lupus erythematosus. Scand J Rheumatol Lin A. Chapter 36: Topical calcineurin inhibitors. In: Wolverton SE (ed)
2003; 32:151–155 Comprehensive dermatologic drug therapy (2nd edn). Philadelphia,
Ho C, Mok C, Lau C, Wong R. Late onset systemic lupus erythemato- PA, Saunders Elsevier, 2007, pp. 671–689
sus in southern Chinese. Ann Rheum Dis 1998; 57:437–444 MacGowan J, Ellis S, Griffiths M, Isenberg D. Retrospective analysis
Ho A, Magder L, Barr S, Petri M. Decreases in anti-double stranded of outcome in a cohort of patients with lupus nephritis treated
DNA levels are associated with concurrent flares in patients between 1977 and 1999. Rheumatology 2002; 41:981–987
with systemic lupus erythematosus. Arthritis Rheum 2001a; 44: Mackay IR et al Lupoid hepatitis and the hepatic lesions of systemic
2342–2349 lupus erythematosus. Lancet 1959; 1:65
158 M. Petri et al.

Manoussakis M, Georgopoulou C, Zintzaras E, et al Sjogren’s syn- Petri M. Thrombosis and systemic lupus erythematosus: The Hopkins
drome associated with systemic lupus erythematosus: Clinical and Lupus Cohort perspective. Scand J Rheumatol 1996; 25:191–193
laboratory profiles and comparison with primary Sjogren’s syn- Petri M. Infection in systemic lupus erythematosus. Rheumatic Dis
drome. Arthritis Rheum 2004; 50:882–891 Clin N Am 1998; 24:423–456
Manzi S, Meilahn E, Rairie J, et al Age-specific incidence rates of myo- Petri M, Allbritton J. Antibiotic allergy in systemic lupus erythemato-
cardial infarction and angina in women with systemic lupus erythe- sus: A case-control study. J Rheumatol 1992; 19(2):265–269
matosus: Comparison with the Framingham Study. Am J Epidemiol Petri M, Perez-Gutthann S, Longenecker J, Hochberg M. Morbidity of
1997; 145:408–415 systemic lupus erythematosus: Role of race and socioeconomic sta-
Marmor MF, Carr RE, Easterbrook M, et al Recommendations on tus. Am J Med 1991a; 91:345–353
screening for chloroquine and hydroxychloroquine retinopathy. A Petri M, Genovese M, Engle E, Hochberg M. Definition, incidence and
report by the American Academy of Ophthalmology. Ophthalmology clinical description of flare in systemic lupus erythematosus: A pro-
2002; 109:1377–1382 spective cohort study. Arthritis Rheum 1991b; 34:937–944
McDonagh J, Isenberg D. Development of additional autoimmune dis- Petri M, Kim MY, Kalunian KC, et al Combined oral contraceptives in
ease in a population of patients with systemic lupus erythematosus. women with systemic lupus erythematosus. N Engl J Med 2005;
Ann Rheum Dis 2000; 59:230–232 353:2550–2558
Middleton GD, McFarlin JE, Lipsky PE. The prevalence and clinical Pichler WJ. Delayed drug hypersensitivity reactions. Ann Intern Med
impact of fibromyalgia in systemic lupus erythematosus. Arthritis 2003; 139(8):683–693
Rheum 1994; 37(8):1181–1188 Pierangeli SS, Harris EN. In vivo models of thrombosis for the antiphos-
Miot HA, Bartoli MLD, Haddad GR. Association between discoid lupus pholipid syndrome. Lupus 1996; 5(5):451–455
erythematosus and cigarette smoking. Dermatology 2005; 211:118–122 Porges AJ, Beattie SL, Ritchlin C, et al Patients with systemic lupus
Modi GM, Maender JL, Coleman N, et al. Tineacorporis masquerading erythematosus at risk for Pneumocystis carinii pneumonia.
as subacutecutaneous lupus erythematosus. Dermatol Online J 2008; J Rheumatol 1992; 19(8):1191–1194
15; 14(4):8 Provost TT, Andres G, Maddison PJ Reichlin M. Lupus band test in
Moghadam-Kia S, Chilek K, Gaines E, et al. Cross-sectional analysis of untreated SLE patients: Correlation of immunoglobulin deposi-
a collaborative web-based database for lupus erythematosus associ- tion in the skin of the extensor forearm with clinical renal disease
ated skin lesions: 114 prospectively enrolled patients. Arch Dermatol and serological abnormalities. J Invest Dermatol 1980; 74(6):
2009; 145(3):255–60 407–412
Mok C. Prognostic factors in lupus nephritis. Lupus 2005; 14:39–44 Pyne D, Isenberg D. Autoimmune thyroid diseases in systemic lupus
Neuwelt C, Lacks S, Kaye B, Ellman J, Borenstein D. Role of intrave- erythematosus. Ann Rheum Dis 2002; 61:70–72
nous cyclophosphamide in the treatment of severe neuropsychiatric Ramos P, Mendez M, Ames P, Khamashta M, Hughes G. Pulse cyclo-
systemic lupus erythematosus. Am J Med 1995; 98:32–41 phosphamide in the treatment of neuropsychiatric systemic lupus
Ng K, Manson J, Rahman A, Isenberg D. Association of antinucleosome erythematosus. Clin Exp Rheumatol 1996; 14:295–299
antibodies with disease flare in serologically active clinically quies- Roman MJ, Shanker BA, Davis A, Lockshin MD, Sammaritano L,
cent patients with systemic lupus erythematosus. Arthritis Care Res Simantov R, et al Prevalence and correlates of accelerated athero-
2006; 55:900–904 sclerosis in systemic lupus erythematosus. N Engl J Med 2003;
Niiyama S, Katsuoka K, Happle R Hoffmann R. Multiple eruptive 349(25):2399–2406
dermatofibromas: A review of the literature. Acta Derm Venereol Ropes MW. Systemic lupus erythematosus. Cambridge, MA, Harvard
2002; 82(4):241–244 University Press, 1976, p. 28
Nimmagadda SR, Spahn JD, Leung DY, Szefler SJ. Steroid-resistant Roy S, Tan K. Pyrexia and normal C-reactive protein (CRP) in patients
asthma: Evaluation and management. [Review] [76 refs] [Journal with systemic lupus erythematosus: Always consider the possibility
Article. Research Support, Non-U.S. Gov’t. Review] Ann Allergy of infection in febrile patients with systemic lupus erythematosus
Asthma Immunol 1996 Nov; 77(5):345–355; quiz 355–356 regardless of CRP levels. Rheumatology 2001; 40:349–350
Noel V, Lortholary O, Casassus P, et al Risk factors and prognostic Runyon BA, LaBrecque DR, Anuras S. The spectrum of liver disease in
influence of infection in a single cohort of 87 adults with systemic systemic lupus erythematosus. Report of 33 histologically-proved
lupus erythematosus. Ann Rheum Dis 2001; 60:1141–1144 cases and review of the literature. Am J Med 1980; 69:187–194
Nossent H, Koldingsnes W. Long-term efficacy of azathioprine treatment Sato EI. Methotrexate therapy in systemic lupus erythematosus. Lupus
for proliferative lupus nephritis. Rheumatology 2000; 39:969–974 2001; 10(3):162–164
Omdal R, Mellgren SI, Koldingsnes W, Jacobsen EA, Husby G. Fatigue Scheinfeld N, Deleo VA. Photosensitivity in lupus erythematosus.
in patients with systemic lupus erythematosus: Lack of associations Photodermatol Photoimmunol Photomed 2004; 20(5):272–279
to serum cytokines, antiphospholipid antibodies, or other disease Schmidt E, Brocker EB, Goebeler M. Rituximab in treatment-resistant
characteristics. J Rheumatol 2002; 29(3):482–486 autoimmune blistering skin disorders. Clin Rev Allergy Immunol
O’Neill S, Isenberg D. Immunizing patients with systemic lupus erythe- 2008; 34(1):56–64
matosus: A review of effectiveness and safety. Lupus 2006; 15: Sivaraj RR, Durrani OM, Denniston AK, et al Ocular manifestations of
778–783 systemic lupus erythematosus. Rheumatology (Oxford) 2007;
Para MF, Finkelstein D, Becker S, et al Reduced toxicity with gradual 46(12):1757–1762
initiation of trimethoprim-sulfamethoxazole as primary prophylaxis Somers E, Magder LS, Petri M. Antiphospholipid antibodies and inci-
for Pneumocystis carinii pneumonia: AIDS Clinical Trials Group dence of venous thrombosis in a cohort of patients with systemic
268. JAIDS 2000; 24(4):337–343 lupus erythematosus. J Rheumatol 2002; 29(12):2531–2536
Pego-Reigosa JM, Isenberg DA. Psychosis due to systemic lupus ery- Sontheimer RD, Henderson CL, Grau RH. Drug-induced subacutecuta-
thematosus: Characteristics and long term outcome of this rare neous lupus erythematosus: A paradigm for bedside-to-bench patient-
manifestation of the disease. Rheumatology 2008; 47:1498–1502 oriented translational clinical investigation Arch Dermatol Res 2009;
Perl A, Nagy G, Gergely P, et al Apoptosis and mitochondrial dysfunc- 301(1):65–70
tion in lymphocytes of patients with systemic lupus erythematosus. Spronk P, terBorg E, Kallenberg C. Patients with systemic lupus erythema-
Methods Mol Med 2004; 102:87–114 tosus and Jaccoud’s arthropathy: A clinical subset with an increased C
Petri M. Musculoskeletal complications of systemic lupus erythemato- reactive protein response? Ann Rheum Dis 1992; 51:358–361
sus in the Hopkins Lupus Cohort: an update. Arthritis Care Res Stahl NI, Klippel JH, Decker JL. Fever in systemic lupus erythemato-
1995; 8:137–145 sus. Am J Med 1979; 67:935–940
13 Systemic Lupus Erythematosus 159

Suryaprasad A, Stone JH. When is it safe to stop Pneumocystis carinii Urowitz MB, Bookman AAM, Koehler BE, Gordon DA, Smythe HA,
(jiroveci) prophylaxis? Insights from cases complicating autoim- Ogryzlo MA. The bimodal mortality pattern of systemic lupus ery-
mune diseases. Arthritis Care & Research 2008; 59(7):1034–9 thematosus. Am J Med 1976; 60:221–225
Takada K, Illei G, Boumpas D. Cyclophosphamide for the treatment of Vassilakopoulos TP, Pangalis GA. Application of a prediction rule to
systemic lupus erythematosus. Lupus 2001; 10:154–161 select which patients presenting with lymphadenopathy should
Tam LS, Li EK, Wong CK, Lam CW, Szeto CC. Double-blind, random- undergo a lymph node biopsy. Medicine (Baltimore) 2000; 79(5):
ized, placebo-controlled pilot study of leflunomide in systemic 338–347
lupus erythematosus. Lupus 2004; 13(8):601–604 Wahl DG, Guillemin F, de Maistre E, Perret C, Lecompte T, Thibaut G.
Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, Rothfeld NF, et al Risk for venous thrombosis related to antiphospholipid antibodies
The 1982 revised criteria for the classification of systemic lupus in systemic lupus erythematosus: A meta-analysis. Lupus 1997;6:
erythematosus. Arthritis Rheum 1982; 25:1271–1277 467–473
Taylor J, Skan J, Erb N, Carruthers D. Lupus patients with fatigue: Is Walz LeBlanc BAE, Gladman DD, Urowitz MB. Serologically active
there a link with fibromyalgia syndrome? Rheumatology 2000; 39: clinically quiescent systemic lupus erythematosus: Predictors of
620–623 clinical flares. J Rheumatol 1994; 21:2239–2241
terBorg E, Horst G, Limburg P, vanRijswijk M, Kallenberg C. C-reactive Ward MM, Donald F. Pneumocystis carinii pneumonia in patients with
protein levels during disease exacerbations and infections in sys- connective tissue diseases: The role of hospital experience in diag-
temic lupus erythematosus: A prospective longitudinal study. nosis and mortality. [Journal Article] Arthritis Rheum 1999 Apr;
J Rheumatol 1990; 17:1642–1648 42(4):780–789
Tsakonas E, Joseph L, Esdaile JM, Choquette D, Senecal JL, Cividino Wassner SJ, Pennisi AJ, Malekzadeh MH, et al The adverse effect of
A, et al A long-term study of hydroxychloroquine withdrawal on anticonvulsant therapy on renal allograft survival. A preliminary
exacerbations in systemic lupus erythematosus. The Canadian report. J Pediatr 1976; 88:134–137
Hydroxychloroquine Study Group. Lupus 1998; 7:80–85 Wells JV, Webb J, Van Deventer M, et al In vivo anti-nuclear antibodies
Tsokos GC. Drugs, sun and T cells in lupus. Clin Exp Immunol 2004; in epithelial biopsies in SLE and other connective tissue diseases.
136(2):191–193 Clin Exp Immunol 1979; 38(3):424–435
Turner-Stokes L, Cambridge G, Corcoran T, et al In vitro response to Werth VP, Sanchez M, White W, et al Incidence of alopecia areata in
influenza immunisation by peripheral blood mononuclear cells from lupus erythematosus. Arch Derm 1992; 128:368–371
patients with systemic lupus erythematosus and other autoimmune Wilson H, Hamilton M, Spyker D, et al Age influences the clinical and
disease. Ann Rheum Dis 1988; 47:532–535 serologic expression of systemic lupus erythematosus. Arthritis
Tzung TY, Liu YS Chang HW. Tacrolimus vs. clobetasol propionate in Rheum 1981; 24:1230–1235
the treatment of facial cutaneous lupus erythematosus: A random- Zandman-Goddard G, Shoenfeld Y. SLE and infections. Clin Rev
ized, double-blind, bilateral comparison study. Br J Dermatol 2007; Allergy Immunol 2003; 25:29–40
156(1):191–192
Ueki H. Koebner phenomenon in lupus erythematosus with special con-
sideration of clinical findings. Autoimmunity reviews 2005; 4(4):
219–223
 Pediatric Systemic Lupus Erythematosus
14
Earl D. Silverman

Pearl: A prolonged prothrombin time is likely the result of a Comment: At least, this is true in developed nations. The sig-
prothrombin (factor II) deficiency as well as a lupus anti- nificant decline in the incidence of rheumatic fever in devel-
coagulant. oped countries has meant that cases of Sydenham’s chorea
are increasingly rare. Cases of new-onset chorea in Western
Comment: SLE is often associated with a prolonged partial
countries are now more likely to be caused by antiphospho-
thromboplastin time (PTT) because of the presence of a lupus
lipid antibodies. However, there has been no formal study of
anticoagulant (Chapter 16, Antiphospholipid Syndrome).
the relative incidence of the two disorders in Western coun-
However, in approximately 5% of lupus patients, there is also
tries (or anywhere else).
a prolonged prothrombin time (PT). The implications of a
Distinguishing between Sydenham’s chorea and the
prolonged PT in lupus are altogether different. A prolonged
antiphospholipid syndrome can be challenging, as patients
PT in lupus is generally the result of an acquired deficiency
with rheumatic fever can have anticardiolipin antibodies
of factor II or an acquired inhibitor to factor II. Patients with
(Figueroa et al. 1992). The additional finding of a lupus anti-
these defects generally present with hemorrhage rather than
coagulant favors the antiphospholipid syndrome (Avcin et al.
thrombosis, the hematological event normally linked with
2008). Among pediatric patients, chorea is almost always
lupus anticoagulants.
associated with SLE rather than the primary antiphospho-
The treatment approach to an acquired factor II deficiency
lipid syndrome (Sanna et al. 2003; Olfat et al. 2004). In SLE,
or inhibitor is immunosuppression. Prednisone is the only
the APS (and, for that matter, acute rheumatic fever) chorea
medication required in most circumstances, and high doses
is more common in children than in adults and may be either
of this medication (e.g., 1–2 mg/kg/day) usually lead to a
unilateral or bilateral (Cervera et al. 1997). Other movement
rapid normalization of the PT (Eberhard et al. 1992; Hudson
disorders associated with aPL are hemidystonia, hemiballis-
et al. 1997; Massengill et al. 1997; Taddio et al. 2007).
mus, myoclonus, and Parkinsonism (Martino et al. 2006).
Pearl: Headaches in lupus can be caused by cerebral vein
Pearl: Children and adolescents require higher doses of glu-
thrombosis.
cocorticoids than adults with SLE.
Comment: Cerebral vein thrombosis is generally seen in the
Comment: The current treatment standard for severe SLE in
presence of a lupus anticoagulant. Venograms performed
the pediatric age group is 2 mg/kg/day prednisone as an ini-
noninvasively by either computed tomography or magnetic
tial dose. Most pediatric rheumatologists do not use more
resonance imaging are usually adequate to make the diagno-
than a total of 60 mg of prednisone per day, but starting doses
sis, and the therapy (obviously) is anticoagulation with cou-
of 80 mg/day are sometimes employed for short periods. This
madin or low molecular weight heparin. Pediatric patients
approach results in a standard dose of 2 mg/kg in patients
may be at higher risk of cerebral vein thrombosis than are
who weigh 30 kg or less, but somewhat less than 2 mg/kg in
adults (Brik et al. 1995; Carhuapoma et al. 1997).
patients who weigh more.
As with many cases of pathological thrombosis, more
The requirement for these relatively higher doses of pred-
than one defect in the coagulation cascade may be present
nisone in smaller children probably results from increased
and a thorough search for all known clotting lesions is impor-
metabolism and a larger volume of distribution of the drug in
tant. Factor V Leiden mutation in combination with a lupus
smaller children. Younger children do not appear to have
anticoagulant has been reported on one patient with cerebral
worse glucocorticoid side-effects, despite the higher dose
vein thrombosis (Uthman et al. 2004).
per kg when compared with adolescents and adults. This is
Pearl: New-onset chorea is more likely to be associated with not to say that adverse effects of glucocorticoids do not occur
antiphospholipid antibodies than with rheumatic fever. in young children: they do.

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 161

DOI:10.1007/978-1-84800-934-9_14, © Springer Science + Business Media B.V. 2009
162 E. D. Silverman

Table 14.1 Manifestations of neonatal lupusa (Frequency data derived Reality: One in 14,000 live births is complicated by com-
from Cimaz et al. 2003 and Boros et al. 2007) plete congenital heart block. Approximately 90% of these
Clinical feature Frequency among consecutive 124 cases are secondary to neonatal lupus erythematosus. The
pregnancies in women with anti-Ro/
SSA or –La/SSB antibodies (%) most serious complication of neonatal lupus erythematosus
is complete congenital heart block or complete atrioventricu-
Complete congenital heart 1.6
block lar block. However, conduction defects are not the most
Rash 16 common manifestation of this disorder. The clinical manifes-
Hepatitis 26 tations of neonatal lupus are shown in Table 14.1.
Hematologic abnormalities 27 Only about one-third of babies born to women with the rel-
(usually neutropenia) evant antibodies have no clinical or laboratory abnormalities
Neurologic abnormalities/ 8
macrocephaly*
(Cimaz et al. 2003). The rash can be subtle and difficult to diag-
Chondrodysplasia punctata 0 nose if it is in a non-sun exposed area. The rash may be present
a
Data from 87 consecutive pregnancies at birth but is most commonly seen at age 6–7 weeks and usu-
ally resolves without treatment by age 4 months. Both liver test
abnormalities and hematologic abnormalities are almost always
Myth: Children with congenital heart block frequently do asymptomatic and will usually resolve with 4–6 months
have mothers with known SLE (or Sjögren’s syndrome). without sequelae (Cimaz et al. 2003). Fulminant liver failure
is a rare complication of neonatal lupus (Lee et al. 2002).
Reality: Neonatal lupus erythematosus (NLE) is associated
Pearl: Neonatal lupus erythematosus can be caused by anti-
with the transplacental passage of maternal anti-Ro/SSA and
RNP antibodies in the absence of anti-Ro/SSA or anti-La/
anti-La/SSB antibodies. Large, prospective studies of moth-
SSB antibodies.
ers with SLE and anti-Ro/SSA or anti-La/SSB antibodies
indicate that the incidence of congenital heart block is between Comment: In 5% of the cases of cutaneous neonatal lupus ery-
1% and 2%. Many mothers who deliver children with con- thematosus, serum from both the mother and baby are negative
genital heart block (CHB) are well or have subclinical fea- for anti-Ro/SSA and–La/SSB antibodies but positive for anti-
tures more consistent with Sjögren’s syndrome than with bodies to ribonucleoprotein (anti-RNP antibodies) (Provost
SLE. Because Sjögren’s syndrome (SjS) tends to develop or et al. 1987; Sheth et al. 1995; Su et al. 2001). Thus, the find-
become clinically manifest later in life than does SLE, many ing of a cutaneous rash or other findings consistent with neo-
women with SjS and anti-Ro/SSA or -La/SSB antibodies are natal lupus should trigger screening for anti-RNP antibodies
asymptomatic at the time of delivery of a child with CHB. in addition to anti-Ro/SSA and–La/SSB antibodies.
For most babies, unfortunately, the fact of these maternal
autoantibodies is identified first as a consequence of the eval- Pearl: Cutaneous neonatal lupus may be associated with
uation for congenital heart block rather than from routine fetal long-term sequelae.
echocardiographic screening of pregnancies in women whose Comment: Cutaneous neonatal lupus presents with a charac-
anti-Ro/SSA or –La/SSB antibodies are known (Waltuck and teristic annular rash that often appears around the eyes in a
Buyon 1994; Press et al. 1996; Julkunen et al. 1998; Lawrence so-called “raccoon distribution.” The rash also occurs com-
et al. 2000; Brucato et al. 2001; Jaeggi et al. 2002; Costedoat- monly on the trunk and back. Although the rash of neonatal
Chalumeau 2004). lupus is often photosensitive, it can be present at birth and in
All pregnant women with anti-Ro/SSA antibodies should areas not exposed to the sun.
be screened with serial fetal echocardiograms beginning at Ninety percent of cases resolve without scarring within
gestational week 18. Currently, it is recommended that at the several months, after the maternal antibodies have disap-
time of diagnosis of autoantibody-associated CHB, the peared from the infant’s serum (Neiman et al. 2000). Systemic
mother of the fetus should be treated with dexamethasone or topical therapy is seldom required. However, in approxi-
(a steroid that crosses the placenta) to try to minimize the mately 10% of patients, the rash heals with puckering or
damage to the developing fetal heart. This therapy appears to telangiectasiae (Thornton et al. 1995; High and Costner
decrease the incidence of intrauterine and neonatal death but 2003). These telangiectasiae are classically in the area of the
does not permanently reverse the heart block. temples, and may develop in that area even if the active rash
Among women who have had one child with congenital did not affect that region clinically (Fig. 14.1). The telangi-
heart block, the likelihood of recurrence in a second preg- ectasiae usually heal slowly with time but may be removed
nancy is on the order of 15–20% (Gladman et al. 2002). by laser surgery if they persist.

Myth: Congenital heart block is the most common manifesta- Pearl: Hydrocephalus should be excluded in all children born
tion of neonatal lupus. to mothers with anti-Ro/SSA and anti-La/SSB autoantibodies.
14 Pediatric Systemic Lupus Erythematosus 163

significant urinalysis abnormalities. Laboratory features

include a pancytopenia with elevation of the serum hepatic
aminotransferases, lactate dehydrogenase, ferritin, triglycer-
ide, and C-reactive protein levels. Elevated levels of d-dimers
are typical of the MAS, but serum fibrinogen levels tend to
be low relative to the degree of inflammation.
MAS is not restricted to pediatric patients but is well
described in adults with rheumatic conditions and in particu-
lar SLE (Lambotte et al. 2006; Ruperto et al. 2006; Pringe
et al. 2007; Qian and Yang 2007).
Pearl: Herpes virus infections can cause significant leukope-
nia and lymphopenia.
Comment: SLE is characterized by a leukopenia that may
include both a lymphopenia and a neutropenia. (Many clini-
cians focus on the tendency to lymphopenia, because this is
one of the American College of Rheumatology criteria for
the classification of SLE.)
Among SLE patients who are leukopenic, however, the
total white blood cell count usually remains above 3,000/dL.
When counts are decreased to a level significantly below this
figure, systemic herpes virus infections must be considered.
These include cytomegalovirus, varicella-zoster virus, and
Epstein–Barr virus (EBV) infections.
Herpes virus infections can be present at the diagnosis of
SLE or, more likely, develop over the course of the illness.
The potential for the reactivation of EBV is particularly a
Fig. 14.1 Cutaneous neonatal lupus
problem in patients who are immunosuppressed (Chung and
Fas 1998; Dror et al. 1998; Rappaport and Tang 2000;
Comment: Neurologic abnormalities have been described in Nakahara et al. 2001; Sekigawa et al. 2002; Verdolini et al.
patients with neonatal lupus (Cabanas et al. 1996; Inoue et al. 2002; Hirose et al. 2006; Kasapcopur et al. 2006; Pluchinotta
2002; Prendiville et al. 2003; Boros et al. 2007). The most et al. 2007; Takei et al. 2007).
important neurological manifestation is macrocephaly, which
Myth: The arthritis of SLE is painful.
can occur with or without hydrocephalus (Nakayama-
Furukawa et al. 1994; Boros et al. 2007). The head circumfer- Reality: It can be, but is not always. The arthritis of SLE is
ence of infants born to mothers with anti-Ro/SSA antibodies regarded generally as a painful, symmetrical polyarthritis.
is larger on average than that of infants whose mothers do not Arthritis in lupus usually does not cause erosive disease
have these antibodies. These large heads tend to return to nor- unless the patient also has a positive serum rheumatoid fac-
mal as the child by age two. However, in rare cases, hydro- tor. In such cases, the overall clinical designation of “rhupus”
cephalus develops and can require surgical intervention. is often used to indicate the clinical and radiologic overlap
between rheumatoid arthritis and lupus.
Pearl: Patients with culture-negative sepsis should be con-
Although the arthritis of lupus is not associated typically
sidered to have the macrophage activation syndrome.
with joint erosions (except in rhupus), tendon and ligament
Comment: Patients with systemic juvenile idiopathic arthri- abnormalities can occur secondary to SLE and lead to joint
tis are at risk for the macrophage activation syndrome (MAS). deformities. More commonly in patients with pediatric SLE,
Patients with SLE are also at risk for this condition. Early as many as 50% of patients have an arthritis that is relatively
recognition of the MAS is the key to its successful manage- painful although there may be some mild discomfort and
ment. Patients with MAS are acutely ill, toxic-appearing, morning stiffness and does not usually lead to permanent
and hypotensive. They appear to have overwhelming sepsis, joint damage. This is not to be confused with the less com-
but all cultures are negative. monly seen painless arthritis, which may lead to joint defor-
One of the keys to the diagnosis of MAS is the relative mities that resemble Jaccoud’s arthropathy. These forms of
preservation of urine output in the face of hypotension. arthritis may cause few symptoms but must be considered as
However, patients with MAS can have azotemia and an additional activity measure in SLE.
164 E. D. Silverman

Myth: African–American patients with proliferative lupus MRI is not sufficiently specific to distinguish SLE patients
nephritis require cyclophosphamide. with diffuse CNS dysfunction but not CNS involvement nor
specific CNS syndromes from patients with SLE without
Reality: African–American patients with proliferative lupus CNS involvement. However, currently the diagnosis of CNS
nephritis have poor disease outcomes when compared with involvement remains a clinical diagnosis.
many other SLE subsets. Many investigators have suggested,
consequently, that African–American patients should be Pearl: Patients with chronic ITP should be monitored for the
treated with pulse cyclophosphamide. However, large series development of lupus.
of pediatric and adult African–American patients with pro- Comment: Idiopathic thrombocytopenic purpura (ITP) is a
liferative nephritis have reported poor outcomes despite the common disorder, even among individuals who do not have
use of pulse cyclophosphamide (or perhaps, in the case of SLE. When this illness becomes chronic and/or resistant to
some patients, because of it) (Dooley et al. 1997; Vyas et al. therapy, screening of these patients with an ANA is indi-
2002; Flower et al. 2006). cated. In many patients, repeat screening is required; the
One trial that compared intravenous pulse cyclophosph- ANA may become positive only months to years after the
amide to mycophenolate mofetil for the treatment of lupus diagnosis of ITP.
nephritis reported that African–Americans had a poorer out- Between 10% and 25% of patients with chronic ITP and a
come than predicted by the NIH studies in the cyclophosph- positive ANA develop overt SLE over time. This percentage
amide arm. One of the major differences was that the is even higher if a Coombs’-positive hemolytic anemia is
majority of patients in the NIH study were white (Contreras present in addition to the thrombocytopenia (the so-called
et al. 2004). “Evans’ syndrome”).
The proper conclusion from this experience may be that
African–American patients with proliferative lupus nephritis Pearl: Some patients with Evans’ don’t have classic lupus,
have a poorer outcome than other patients with other racial/ but rather a defect in the Fas–Fas ligand pathway known as
ethnic backgrounds and therefore should not be exposed to “ALPS”: the autoimmune lymphproliferative syndrome.
more toxic therapy, which does not appear to improve long- Comment: ALPS is a heritable disorder of the Fas (CD95)-
term outcome. Fas ligand pathway that leads to a defect in lymphocyte apop-
Myth: Magnetic resonance imaging studies are useful in the tosis. The disorder usually presents between the ages of 1 and
evaluation of SLE patients with diffuse central nervous sys- 18 with a median age of presentation of 2 years. ALPS is
tem involvement. associated with mutations in the following genes: TNFRSF6,
Fas ligand (TNFSF6), caspase 8, and 10, but these known
Reality: Sorting out the cause of diffuse central nervous sys- mutations do not account for all recognized cases. ALPS is
tem (CNS) dysfunction, i.e., acute confusional states, mood categorized into three recognized subtypes: (1) Type 1a due to
disorders, headaches, psychosis, or cognitive impairment is mutations in TNFRSF6; (2) Type 1b due to mutations in
challenging. Laboratory investigations in many such patients TNFSF6 and (3) Types IIa and IIb due to defects in caspase-10
indicate that the usual parameters measured in the hope of and caspase-8 respectively; and (4) Type III without a known
gleaning insight on disease activity – serum complement lev- mutation. The diagnostic laboratory features are the presence
els, anti-dsDNA antibody levels, and acute phase reactants – of T lymphocytes, which express the alpha/beta T-cell recep-
are normal or unchanged from the patients’ baseline. tor but lack expression of either CD4 or CD8 (alpha/beta
Much hope has been placed in MRI investigations of these double negative T cell) and defective in vitro Fas-mediated
patients, particularly in the potential specificity of FLAIR apoptosis.
and diffusion imaging in delineating active neuropsychiatric As may be expected by the name, this syndrome is clini-
SLE from other conditions. Unfortunately, studies of this cally characterized by lymphoproliferation, which includes
imaging modality indicate that many patients with diffuse “massive” lymphadenopathy, hepatomegaly, splenomegaly,
CNS disease have only non-specific changes or even normal and hypersplenism. Less common clinic findings include urti-
examinations on MRI (Sabbadini et al. 1999; Jennings et al. caria, uveitis, vasculitis, arthritis, oral ulcers, panniculitis, and
2004; Sundgren et al. 2005; Castellino et al. 2008). One of occasionally glomerulonephritis. Many of these features may
the most common findings in this setting, volume loss, can be seen in patients with SLE. Other autoimmune features
be secondary to acute SLE but also to prior glucocorticoid include autoimmune hemolytic anemia, autoimmune throm-
treatment (Jennings et al. 2004). This helps neither the clini- bocytopenia, and autoimmune neutropenia. Many patients are
cian nor the patient in most cases. ANA positive and may have specific autoantibodies.
Many of these imaging tests distinguish normal controls The diagnosis of ALPS can be confirmed by either genetic
from SLE patients, but no one needs a $1,500 test to do that. studies, the persistence of elevated levels of double negative
14 Pediatric Systemic Lupus Erythematosus 165

alpha/beta T-cells, or the characteristic defect in Fas-mediated

apoptosis. Lymph node biopsies show marked expansion of
paracortical T cells, the majority of which are double nega-
tive alpha/beta T cells. These cells show active expansion but
very little apoptosis. Probably, as a result of abnormal sur-
vival of lymphocytes and the potential role of Fas as a tumor
suppressor gene, these patients are at an increased risk for
the development of malignancy and approximately 10% of
patients will develop a malignancy. These patients are at par-
ticular risk for hematologic malignancies with very high
relative risks for lymphomas but thyroid cancer, hepatocel-
lular carcinoma, colon cancer, and skin cancers have also
been described. The major clinical problems seen in patients
with ALPS are related to the persistent cytopenias, which
may be life-threatening. They generally respond well to cor- Fig. 14.3 Discoid lupus erythematosus lesion. When this cutaneous
finding is present in a child, the patient usually evolves features of sys-
ticosteroid therapy but frequently these patients become ste- temic lupus erythematosus over time. In contrast, in adults, discoid
roid-dependent and may require immunosuppressive agents. lupus lesions are associated with a low likelihood of the development of
Recently, there have been reports of good responses to myco- systemic lupus erythematosus (Courtesy of Dr. Rachel Abuav)
phenolate mofetil.
Pearl: Discoid lesions in children and adolescents are likely Comment: Discoid lupus lesions comprise a scarring der-
secondary to SLE, not to isolated discoid Lupus Erythe- mopathy that is characterized by lesions involving the skin
matosus. of the head and neck (classically the ears, face, or scalp)
(Fig. 14.2). Many adults with discoid lesions have no fea-
tures of systemic involvement and are believed to have a
low likelihood of evolution to SLE. In contrast, discoid
lesions are rarely observed in children. When discoid lesions
are present in children, however, they indicate a high likeli-
hood that SLE will develop eventually, albeit many months
or years may elapse before the onset of systemic disease
(Fig. 14.3).

References

Avcin T, Benseler SM, et al A followup study of antiphospholipid anti-

bodies and associated neuropsychiatric manifestations in 137 chil-
dren with systemic lupus erythematosus. Arthritis Rheum 2008;
59(2):206–213
Boros CA, Spence D, et al Hydrocephalus and macrocephaly: New
manifestations of neonatal lupus erythematosus. Arthritis Rheum
2007; 57(2):261–266
Brik R, Padeh S, et al Systemic lupus erythematosus in children in
Israel. Harefuah 1995; 129(7–8):233–235, 296, 295
Brucato A, Frassi M, et al Risk of congenital complete heart block in
newborns of mothers with anti-Ro/SSA antibodies detected by
counterimmunoelectrophoresis: A prospective study of 100 women.
Arthritis Rheum 2001; 44(8):1832–5
Cabanas F, Pellicer A, et al Central nervous system vasculopathy in neo-
natal lupus erythematosus. Pediatr Neurol 1996; 15(2):124–126
Carhuapoma JR, Mitsias P, et al Cerebral venous thrombosis and anti-
cardiolipin antibodies. Stroke 1997; 28(12):2363–2369
Castellino G, Padovan M, et al Single photon emission computed tomog-
raphy and magnetic resonance imaging evaluation in SLE patients
Fig. 14.2 Telangiectasias occurring over the temple region in a child with and without neuropsychiatric involvement. Rheumatology
with a history of neonatal lupus (Oxford) 2008; 47(3):319–323
166 E. D. Silverman

Cervera R, Asherson RA, et al Chorea in the antiphospholipid syndrome. Martino D, Chew NK, et al Atypical movement disorders in antiphos-
Clinical, radiologic, and immunologic characteristics of 50 patients pholipid syndrome. Mov Disord 2006; 21(7):944–949
from our clinics and the recent literature. Medicine (Baltimore) 1997; Massengill SF, Hedrick C, et al Antiphospholipid antibodies in pediat-
(3):203–212 ric lupus nephritis. Am J Kidney Dis 1997; 29(3):355–361
Chung AB Fas N. Successful acyclovir treatment of herpes simplex type Nakahara C, Hayashi D, et al Delayed onset of systemic lupus erythe-
2 hepatitis in a patient with systemic lupus erythematosus: A case matosus in a child with endothelial tubuloreticular inclusion. Clin
report and meta analysis. Am J Med Sci 1998; 316(6):404–407 Nephrol 2001; 56(4):332–335
Cimaz R, Spence DL, et al Incidence and spectrum of neonatal lupus Nakayama-Furukawa F, Takigawa M, et al Hydrocephalus in two
erythematosus: A prospective study of infants born to mothers with female siblings with neonatal lupus erythematosus. Arch Dermatol
anti-Ro autoantibodies. J Pediatr 2003; 142(6):678–683 1994; 130(9):1210–1212
Contreras G, Pardo V, Baudouin L, Lenz O, Tozman E, O’Nan P, Roth D. Neiman AR, Lee LA, et al Cutaneous manifestations of neonatal lupus
Sequential therapies for proliferative lupus nephritis. N Engl J Med without heart block: Characteristics of mothers and children enrolled
2004; 350(10):971–980 in a national registry. J Pediatr 2000; 137(5):674–680
Costedoat-Chalumeau N, Amoura Z, et al Outcome of pregnancies in Olfat MO, Al-Mayouf SM, et al Pattern of neuropsychiatric manifesta-
patients with anti-SSA/Ro antibodies: a study of 165 pregnancies, with tions and outcome in juvenile systemic lupus erythematosus. Clin
special focus on electrocardiographic variations in the children and com- Rheumatol 2004; 23(5):395–399
parison with a control group. Arthritis Rheum 2004; 50(10):3187–3194 Pluchinotta FR, Schiavo B, et al Distinctive clinical features of pediat-
Dooley MA, Hogan S, et al Cyclophosphamide therapy for lupus nephri- ric systemic lupus erythematosus in three different age classes.
tis: Poor renal survival in black Americans. Glomerular Disease Lupus 2007; 16(8):550–555
Collaborative Network. Kidney Int 1997; 51(4):1188–1195 Prendiville JS, Cabral DA, et al Central nervous system involvement in
Dror Y, Blachar Y, et al Systemic lupus erythematosus associated with neonatal lupus erythematosus. Pediatr Dermatol 2003; 20(1):60–67
acute Epstein-Barr virus infection. Am J Kidney Dis 1998; 32(5): Press J, Palayew K, et al Antiribosomal P antibodies in pediatric patients
825–828 with systemic lupus erythematosus and psychosis. Arthritis Rheum
Eberhard A, Couper R, et al Exocrine pancreatic function in children with 1996; 39(4):671–676
systemic lupus erythematosus. J Rheumatol 1992; 19(6):964–967 Pringe A, Trail L, et al Macrophage activation syndrome in juvenile
Figueroa F, Berrios X, et al Anticardiolipin antibodies in acute rheu- systemic lupus erythematosus: An under-recognized complication?
matic fever. J Rheumatol 1992; 19(8):1175–1180 Lupus 2007; 16(8):587–592
Flower C, Hennis A, et al Lupus nephritis in an Afro-Caribbean population: Provost TT, Watson R, et al The neonatal lupus syndrome associated with
Renal indices and clinical outcomes. Lupus 2006; 15(10):689–694 U1RNP (nRNP) antibodies. N Engl J Med 1987; 316(18):1135–1138
Gladman G, Silverman ED, et al Fetal echocardiographic screening of Qian J, Yang CD. Hemophagocytic syndrome as one of main manifes-
pregnancies of mothers with anti-Ro and/or anti-La antibodies. Am tations in untreated systemic lupus erythematosus: Two case reports
J Perinatol 2002; 19(2):73–80 and literature review. Clin Rheumatol 2007; 26(5):807–810
High WA Costner MI. Persistent scarring, atrophy, and dyspigmenta- Rappaport KD, Tang WM. Herpes simplex virus type 2 acute retinal
tion in a preteen girl with neonatal lupus erythematosus. J Am Acad necrosis in a patient with systemic lupus erythematosus. Retina
Dermatol 2003; 48(4):626–628 2000; 20(5):545–546
Hirose I, Ymamaguchi H, et al Fatal varicella infection in a girl with Ruperto N, Ravelli A, et al The pediatric rheumatology international
systemic lupus erythematosus after oral acyclovir prophylaxis. Eur trials organization/American College of Rheumatology provisional
J Pediatr 2006; 165(4):280–281 criteria for the evaluation of response to therapy in juvenile systemic
Hudson N, Duffy CM, et al Catastrophic haemorrhage in a case of pae- lupus erythematosus: prospective validation of the definition of
diatric primary antiphospholipid syndrome and factor II deficiency. improvement. Arthritis Rheum 2006; 55(3):355–363
Lupus 1997; 6(1):68–71 Sabbadini MG, Manfredi AA, et al Central nervous system involvement
Inoue K, Fukushige J, et al Central nervous system vasculopathy asso- in systemic lupus erythematosus patients without overt neuropsy-
ciated with neonatal lupus. Pediatr Neurol 2002; 26(1):68–70 chiatric manifestations. Lupus 1999; 8(1):11–19
Jaeggi ET, Hamilton RM, et al Outcome of children with fetal, neonatal Sanna G, Bertolaccini ML, et al Neuropsychiatric manifestations in
or childhood diagnosis of isolated congenital atrioventricular block. systemic lupus erythematosus: Prevalence and association with
A single institution’s experience of 30 years. J Am Coll Cardiol antiphospholipid antibodies. J Rheumatol 2003; 30(5):985–992
2002; 39(1):130–137 Sekigawa I, Nawata M, et al Cytomegalovirus infection in patients with
Jennings JE, Sundgren PC, et al Value of MRI of the brain in patients systemic lupus erythematosus. Clin Exp Rheumatol 2002;
with systemic lupus erythematosus and neurologic disturbance. 20(4):559–564
Neuroradiology 2004; 46(1):15–21 Sheth AP, Esterly NB, et al U1RNP positive neonatal lupus erythema-
Julkunen H, Kaaja R, et al Immune-mediated congenital heart block tosus: Association with anti-La antibodies? Br J Dermatol 1995;
(CHB): Identifying and counseling patients at risk for having chil- 132(4):520–526
dren with CHB. Semin Arthritis Rheum 1998; 28(2):97–106 Su CT, Huang CB, et al Neonatal lupus erythematosus in association
Kasapcopur O, Ergul Y, et al Systemic lupus erythematosus due to with anti-RNP antibody: A case report. Am J Perinatol 2001;
Epstein-Barr virus or Epstein-Barr virus infection provocating acute 18(8):421–426
exacerbation of systemic lupus erythematosus? Rheumatol Int 2006; Sundgren PC, Jennings J, et al MRI and 2D-CSI MR spectroscopy of
26(8):765–767 the brain in the evaluation of patients with acute onset of neuropsy-
Lambotte O, Khellaf M, et al Characteristics and long-term outcome of chiatric systemic lupus erythematosus. Neuroradiology 2005;
15 episodes of systemic lupus erythematosus-associated hemophago- 47(8):576–585
cytic syndrome. Medicine (Baltimore) 2006; 85(3):169–182 Taddio A, Brescia AC, et al Steady improvement of prothrombin levels
Lawrence S, Luy L, et al The health of mothers of children with cutane- after cyclophosphamide therapy in pediatric lupus anticoagulant
ous neonatal lupus erythematosus differs from that of mothers of chil- hypoprothrombinemia syndrome (LAHPS). Clin Rheumatol 2007;
dren with congenital heart block. Am J Med 2000; 108(9):705–709 (12):2167–2169
Lee LA, Sokol RJ, et al Hepatobiliary disease in neonatal lupus: preva- Takei M, Yamakami K, et al A case of systemic lupus erythematosus
lence and clinical characteristics in cases enrolled in a national reg- complicated by alveolar hemorrhage and cytomegalovirus colitis.
istry. Pediatrics 2002; 109(1):E11 Clin Rheumatol 2007; 26(2):274–277
14 Pediatric Systemic Lupus Erythematosus 167

Thornton CM, Eichenfield LF, et al Cutaneous telangiectases in neona- Vyas S, Hidalgo G, et al Outcome in African-American children of
tal lupus erythematosus. J Am Acad Dermatol 1995; 33(1):19–25 neuropsychiatric lupus and lupus nephritis. Pediatr Nephrol 2002;
Uthman I, Khalil I, et al Lupus anticoagulant, factor V Leiden, and meth- 17(1):45–49
ylenetetrahydrofolate reductase gene mutation in a lupus patient with Waltuck J, Buyon JP. Autoantibody-associated congenital heart block:
cerebral venous thrombosis. Clin Rheumatol 2004; 23(4):362–363 Outcome in mothers and children. Ann Intern Med 1994; 120(7):
Verdolini R, Bugatti L, et al Systemic lupus erythematosus induced by 544–551
Epstein-Barr virus infection. Br J Dermatol 2002; 146(5):877–881
 Mixed Connective Tissue Disease
15
Robert W. Hoffman

States, Mexico, and Japan indicate that MCTD is less com-

Overview of Mixed Connective mon than lupus but more common than scleroderma or der-
Tissue Disease matomyositis (Kotajima et al. 1999). Although “rare” is
relative when it comes to rheumatic diseases, MCTD is actu-
› Mixed connective tissue disease (MCTD) is often ally more common than some other connective tissue diseases
viewed as a “cousin” of systemic lupus erythematosus that have a higher profile among both medical practitioners
(SLE), with which it shares many disease features. and the general public.
› Some experts regard MCTD as a variant of SLE
Myth: MCTD, undifferentiated connective tissue disease
because of the considerable overlap in clinical and
(UCTD), and “overlap syndrome” are all synonymous
serological manifestations. Others consider it to be a
terms.
unique disorder because of the frequency with which
certain phenotypic features occur in patients classified Reality: These three terms are frequently confused and often
as MCTD. used interchangeably, particularly by non-rheumatologists.
› Patients with MCTD by definition have antibodies The term “UCTD” was coined to describe patients who fail to
directed against ribonuceoprotein, i.e., anti-RNP meet classification criteria for certain distinct rheumatic disor-
antibodies. ders, particularly scleroderma, SLE, or Sjögren’s syndrome.
› Common clinical manifestations of MCTD are arthritis, UCTD patients typically have disease of recent onset.
Raynaud’s phenomenon, digital ischemia progressing Over time, their clinical features often evolve to resemble a
to fingertip ulceration, pulmonary hypertension, inter- “clear cut” rheumatic disease (or perhaps more than one)
stitial lung disease, and gastrointestinal hypomotility. more closely. Many such rheumatic conditions are “difficult
› Some patients with MCTD evolve with time to a clini- to pigeonhole” early in their course, but in many such cases,
cal phenotype that is more easily classified as SLE. the patients ultimately develop features that are recognizable
Others, perhaps a majority, evolve over time to clinical as a disease sanctioned by ACR classification criteria. As an
syndromes most compatible with systemic sclerosis. example, the importance of making the diagnosis of SLE in
› MCTD must be distinguished from “undifferentiated” some patients with fewer than four ACR criteria for the clas-
connective tissue diseases (in which a clear-cut rheuma- sification of that disorder is discussed elsewhere (see Systemic
tological process may or may not emerge with time) and Lupus Erythematosus, chapter 13).
also from “overlap” connective tissue diseases, in which All that said, within 5 years of the onset of symptoms,
the disease features of two or more well-defined connec- long-term follow-up studies have demonstrated that nearly
tive tissues diseases are present in the same patient. As all patients with possible MCTD can be classified confidently
example of an overlap connective tissue disease is the as having MCTD or another distinct rheumatic disease within
occurrence of SLE and polymyositis in the same patient. 5 years of their symptom onset (Burdt et al. 1999). A subset
Such a patient would not be regarded as a “mixed” con- of patients fails to meet classification criteria for MCTD or
nective tissue disease (MCTD), but rather as the co- any other rheumatic disease at that time. These patients
occurrence of two diseases in the same patient. should be considered truly “undifferentiated,” i.e., UTCD.
Long-term follow-up studies of these true UCTD patients
suggest a generally good outcome (Mosca et al. 1999).
Myth: Mixed Connective Tissue Disease (MCTD) is rare.
Finally, an overlap CTD is one in which there are unequiv-
Reality: No definitive epidemiologic studies related to MCTD ocal features of two autoimmune diseases recognized in text-
have been published, but clinical studies from the United books as being separate. One classic example of an overlap

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 169

DOI:10.1007/978-1-84800-934-9_15, © Springer Science + Business Media B.V. 2009
170 R. W. Hoffman

CTD is the occurrence of symptoms of polymyositis and sys- esophageal dysmotility, Raynaud’s phenomenon, serositis, and
temic sclerosis in the same patient. In such cases, there is skin rash (Nimelstein et al. 1980; Burdt et al. 1999). However,
nothing “undifferentiated” about the manifestations of either both persistent and new clinical features, including sclerodac-
disease: a muscle biopsy confirms the histopathologic fea- tyly, pulmonary hypertension, and peripheral neuropathy,
tures of polymyositis, and the patient has skin thickening and became the dominant features of the illness later in the course of
other disease manifestations of scleroderma. This overlap of the disease. After some years of disease, some patients meet the
polymyositis and systemic sclerosis actually has a unique ACR classification criteria for scleroderma.
autoantibody: PM/Scl autoantibodies. These antibodies are Despite its challenges with regard to disease classifica-
directed against a nuclear antigen complex that consists of tion, the concept of MCTD has been recognized widely since
11–16 proteins, the precise cellular function of which remains its formulation in the early 1970s (Sharp et al. 1972). MCTD
unknown (Treadwell et al. 1984). has been identified in a variety of different populations across
the world. Although some iconoclasts continue to grouse
Pearl: Sicca symptoms are common in MCTD. about “defining a disease by the presence of an autoanti-
Comment: Keratoconjunctivitis sicca and xerostomia are body” (a practice that no true MCTD expert endorses), the
detected often whenever MCTD patients are asked in a sys- notion of a distinct MCTD phenotype rings true with experi-
tematic manner about these symptoms. Symptoms and signs enced clinicians, and most consider it a valid entity.
of Sjögren’s syndrome have been identified in more than half Myth: MCTD is no longer a clinically relevant term.
of all MTCD patients in some studies. These findings, con-
sidered to represent secondary Sjögren’s syndrome rather Reality: The primary disease-related cause of death in MCTD
than an overlap CTD (see above), are often accompanied by is pulmonary hypertension. This feature marks a striking
the development of anti-SSA/Ro autoantibodies. divergence of this disease from SLE, a disorder in which
clinically significant pulmonary hypertension rarely becomes
Pearl: MCTD can present as a unilateral or bilateral trigem- an issue. Advances in our ability to diagnose and treat pul-
inal neuropathy. monary hypertension effectively in recent years strengthen
the argument for early recognition of the MCTD clinical
Comment: A well-recognized complication of MCTD, early
phenotype.
in the disease course, is a fifth cranial nerve palsy. A similar
or identical finding (of equally obscure pathogenesis) has Pearl: Periodic screening of patients using with pulmonary
been reported in systemic sclerosis (Farrell and Medsger function tests (PFTs) and two-dimensional echocardiogra-
1982). The common link between the occurrence of this neu- phy helps identify incipient pulmonary hypertension.
rological lesion and these two rheumatic diseases might be
Comment: Annual PFTs with DLco measurement as well as
the presence of anti-RNP antibodies.
two-dimensional echocardiograms are recommended for
The development of trigeminal nerve dysfunction in
patients with MCTD. Substantial pulmonary hypertension
MCTD or systemic sclerosis is bilateral as often as it is uni-
can exist in MCTD even in the setting of mild clinical symp-
lateral. The clinical features of this problem consist of the
tomatology and a normal chest radiograph. A declining DLco
sudden onset of either unilateral or bilateral facial numbness,
or elevated right-sided cardiac pressures should lead to fur-
with or without pain or paresthesias. In one series of 442
ther evaluations, including a high-resolution chest CT or
consecutive scleroderma patients from the University of
right heart catheterization.
Pittsburgh, 6 of the 16 patients (38%) with trigeminal neu-
ropathies were strongly positive for antibodies to RNP. In Pearl: The emergence of anti-RNP antibodies is linked
contrast, only 8% of those who did not have trigeminal neu- closely to the onset of clinical manifestations of disease.
ropathy had antibodies to RNP.
Comment: The development of anti-RNP autoantibodies
Myth: MCTD is simply a subset of either lupus or bears a close temporal relationship to the onset of clinical
scleroderma. signs and symptoms in MCTD. A retrospective analysis of
serum samples in patients subsequently diagnosed with a
Reality: The etiology and pathogenesis of all three of these con-
rheumatic disease found that most patients develop clinical
ditions remain unknown. Consequently, any classification
signs and symptoms within 6 months of the appearance of
scheme for such diseases is potentially flawed and subject to
anti-RNP antibodies (Arbuckle et al. 2003). Moreover, the
refinement in the light of future scientific advances. In a follow-
presence of anti-RNP antibodies is rare among individuals
up study of the original group studied by Sharp et al. and in
who are asymptomatic at testing.
longitudinal long-term outcome studies, investigators found that
with treatment there was a reduction over time in the inflamma- Myth: Mixed connective tissue disease (MCTD) can be diag-
tory features of the disease, including arthralgias/arthritis, nosed by the presence of anti-RNP antibodies.
15 Mixed Connective Tissue Disease 171

Reality: Although anti-RNP antibodies are not diagnostic of Pearl: Gastroesophageal reflux responds well to proton-
MCTD in and of themselves, the diagnosis should be viewed pump inhibitors but less well to older classes of agents for
with great skepticism if these antibodies are absent. Anti- this problem.
RNP reactivity is extremely rare in the absence of an autoim-
Comment: Proton pump inhibitors offer an effective form of
mune disease but does occur with some frequency in overlap
therapy for what was once one of the more refractory fea-
connective tissue disorders, systemic lupus erythematosus
tures of MCTD, reflux esophagitis. MCTD patients bear a
(SLE), and scleroderma. Thus, anti-RNP antibodies are nec-
striking resemblance to patients with systemic sclerosis in
essary but not sufficient for the diagnosis of MCTD.
their tendency to develop gastroesophageal reflux, as well as
Anti-RNP antibodies are required for the classification of
in the frequency with which Raynaud’s phenomenon, digital
an individual as having MCTD (Sharp et al. 1972; Burdt
ulceration (Fig. 15.2), and pulmonary hypertension compli-
et al. 1999). However, there are no validated diagnostic cri-
cate the disease.
teria for MCTD. (The same is true for many, if not most,
other rheumatic diseases.) Classification criteria developed Myth: Myositis is common in MCTD.
for the purpose of research cannot be applied reliably to the
Reality: Although an inflammatory myopathy was included as
diagnosis of individual patients, even though clinicians are
part of the initial disease description (Sharp et al. 1972), myo-
often tempted to do so.
sitis actually tends to be quite mild in most patients with
Pearl: Diffuse swelling of the fingers that extends to the dor- MCTD. Myositis remains clinically silent in many patients
sum of the hand may be a persistent feature in MCTD. This unless sought specifically. Often a mild serum creatine kinase
can lead to the misdiagnosis of rheumatoid arthritis. (CK) elevation provides the only clue to muscle inflammation.
The typical MCTD patient with muscle involvement has a
Comment: Diffuse swelling of the fingers is a hallmark of
serum CK level elevated not higher than two or three times the
MCTD. This is sometimes related to synovitis, but more often
corresponds to a soft tissue swelling not arthritis. Non-arthrititic
soft tissue swelling can extend to the dorsum of the hands, per-
a
sist for years, and progress to sclerodactyly in some patients. It
also resolves with treatment in other patients. In addition to dif-
fuse swelling, tapering of the fingers is also characteristic (but
not pathognomonic) of MCTD in many patients (Fig. 15.1).
Arthralgias or arthritis can also be a primary MCTD fea-
ture. Rheumatoid factor is detectable in the sera of 50–75%
of MCTD patients. Antibodies to cyclic citrullinated peptide
have also been reported in MCTD. The presence of the major
histocompatibility gene encoding HLA-DR4 appears to be a
risk factor for this disease (Hoffman et al. 1990).

Fig. 15.1 Sclerodactyly often leads to pronounced finger tapering in

patients with mixed connective tissue disease. This feature is not Fig. 15.2 Sclerodactyly and digital ischemia in patients with mixed con-
pathognomonic of mixed connective tissue disease but is a marker for a nective tissue disease. (a) shows digital pitting, a long-term consequence
variety of disorders in the “connective tissue disease” spectrum (Figure of digital ischemia. (b) shows gangrenous changes in the distal part of
courtesy of John H. Stone) one finger (Figures courtesy of John H. Stone)
172 R. W. Hoffman

upper limit of normal. Even more important than low serum References
CK levels, however, is the fact that few MCTD patients develop
clinical muscle weakness. The inflammatory myopathy may Arbuckle MR, McClain MT, Rubertone MV, et al Development
be so mild that it does not affect treatment considerations. of autoantibodies before the clinical onset of systemic lupus erythe-
matosus. N Engl J Med 2003; 349:1526–1533
Pearl: Raynaud’s phenomenon is nearly universal among Burdt MA, Hoffman RW, Deutscher SL, et al Long-term outcome in
patients with MCTD. mixed connective tissue disease: Longitudinal clinical and serologic
findings. Arthritis Rheum 1999; 42:899–909
Comment: Pointed questioning unearths this symptom in Farrell DA, Medsger TA Jr. Trigeminal neuropathy in progressive sys-
almost every case of MCTD. This finding heightens the sus- temic sclerosis. Am J Med 1982; 73:57–62
picion that an autoimmune disorder might be present in a Kotajima L, Aotsuka S, Sumiya M, et al Clinical features of patients
with juvenile onset mixed connective tissue disease: Analysis of
patient with otherwise nondescript symptoms. data collected in a nationwide collaborative study in Japan.
J Rheumatol 1999; 23:1088–1094
Myth: MCTD is more severe in children than in adults.
Michels H. Course of mixed connective tissue disease in children. Ann
Reality: Initial studies on MCTD from the Los Angeles Med 1997; 29:359–364
Mosca M, Neri R, Bombardieri S. Undifferentiated connective tissue
County Children’s Hospital suggested that the disease was diseases (UCTD): A review of the literature and a proposal for pre-
more severe on average in children than in adults (Singsen liminary classification criteria. Clin Exp Rheumatol 1999; 17:
et al. 1977). As the recognition of MCTD has grown more 615–620
widespread, it has become apparent that a range of manifes- Nimelstein SH, Brody S, McShane D, Holman HR. Mixed connective
tissue disease: A subsequent evaluation of the original 25 patients.
tations exists among pediatric patients and the long-term out- Medicine 1980; 59:239–248
comes in this population are variable. Studies from Missouri, Sharp GC, Irvin WS, Tan EM, et al Mixed connective tissue disease: An
Minnesota, Germany, and Japan have demonstrated that the apparently distinct rheumatic disease syndrome associated with a
course and prognosis of MCTD in children is similar to that specific antibody to extractable nuclear antigen (ENA). Am J Med
1972; 52:148–159
in adults (Hoffman 1993; Burdt et al. 1999; Michels 1997). Singsen BH, Kornreich HK, Koster-King K, et al Mixed connective tis-
sue disease in children. Arthritis Rheum 1977; 20(2 Suppl):
Pearl: MCTD can cause a nasal septal perforation.
355–360
Comment: The etiology is not clear, but MCTD joins Wegener’s Treadwell EL, Alspaugh MA, Wolfe JF, Sharp GC. Clinical relevance
of PM-1 antibody and physiochemical characterization of PM-1
granulomatosis, relapsing polychondritis, sarcoidosis, and a antigen. J Rheumatol 1984; 11(5):658–662
smattering of other conditions that sometimes lead to a nasal
septal perforation.
 The Antiphospholipid Syndrome
16
David P. D’Cruz and Munther Khamashta

16.1 Overview of the Antiphospholipid dinal study of SLE patients, aPL-related thromboses
Syndrome accounted for 27% of all deaths in the cohort (Cervera
et al. 1993).
› The term “catastrophic APS” refers to a rare but poten-
› The antiphospholipid syndrome (APS) is a multi-system tially lethal disorder characterized by accelerated
autoimmune disorder characterized by recurrent arte- thrombosis, multiple organ involvement that includes
rial and venous thromboses and pregnancy morbidity renal thrombotic microangiopathy and death in a high
(Hughes 1983). percentage of patients (Bucciarelli et al. 2006).
› The most common clinical manifestations of the APS
› The combination of anticoagulation, glucocorticoids,
are shown in Table 16.1. and plasma exchange is a useful treatment approach to
› Cutaneous manifestations of the APS include livedo the catastrophic APS (Bucciarelli et al. 2006). Immuno-
reticularis (livedo racemosa), leg ulcers, skin necrosis, suppression, particularly with cyclophosphamide,
superficial thrombophlebitis, splinter hemorrhages, should be avoided.
digital ischemia, and gangrene. Many of these features
must be distinguished from vasculitis, which can also
› Arterial and venous events associated with the APS are
treated initially with unfractionated or low molecular
be associated with these findings. weight heparins, followed by oral anticoagulation with
› The APS is linked to multiple central nervous system
warfarin.
manifestations. These are listed in Table 16.2.
› Women with APS are at high risk of complications
› The APS can occur in a primary form which is not
during pregnancy and in the postpartum period. In the
associated with another underlying condition. In addi-
postnatal period, women with Thrombotic APS should
tion, it can develop concurrently with disorders such as
be given thromboprophylaxis with subcutaneous hepa-
systemic lupus erythematosus (SLE).
rin followed by the resumption of oral anticoagulation
› Antiphospholipid antibodies (aPL) are a heterogeneous
as soon as possible. Controversies in the management
group of immunoglobulins directed at phospholipid
of patients during pregnancy are discussed.
binding proteins. Three antibodies are central to the
diagnosis of APS: the lupus anticoagulant (LA), anti-
cardiolipin antibodies (aCL), and b-2-glycoprotein 1.
› aCL and b-2-glycoprotein 1 are detected by enzyme-
linked immunosorbent assays. The LA is commonly Myth: The lupus anticoagulant indicates that the patient has
detected by the dilute Russell viper venom time, but systemic lupus erythematosus.
other assays are also available (Brandt et al. 1995).
› aPL are found in less than 1% of healthy populations Reality: This is a common misconception among patients
of all ages. However, the figure is substantially higher (and some physicians). The term “lupus anticoagulant” is
among healthy older populations – as high as 5%. The among the more confusing terms in medicine, for several rea-
prevalence in SLE is substantially higher, on the order sons. The lupus anticoagulant (LA) is in fact a pro-thrombotic
of 24% for IgG aCL, 13% for IgM aCL, and 15% for autoantibody directed against phospholipid-binding proteins.
LA (Cervera et al. 1993). (This is true for all antiphospholipid antibodies.) In addition,
› APS has a significant impact on morbidity and mortal- despite the fact that the updated classification criteria for sys-
ity (Shah et al. 1998; Jouhikainen et al. 1993; Schulman temic lupus erythematosus include the presence of a lupus
et al. 1998; Ruiz-Irastorza et al. 2004). In one longitu- anticoagulant, many patients with lupus anticoagulants do
not have and will not develop lupus (Hochberg 1997).

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 173

DOI:10.1007/978-1-84800-934-9_16, © Springer Science + Business Media B.V. 2009
174 D. P. D’Cruz et al.

Table 16.1 Most common cumulative clinical manifestations of the Myth: A normal activated partial thromboplastin time (aPTT)
antiphospholipid syndrome in 1,000 patients followed prospectively excludes an LA.
(Adapted from Cervera et al. 2002)
Deep vein thrombosis (38.9%) Reality: A prolonged aPTT suggests the possibility of an LA,
Pulmonary embolism (14.1%) but only about half of all patients with documented LAs have
Cutaneous manifestations, especially livedo reticularis/racemosa an abnormal aPTT. The dilute Russell viper venom time
(24.1%) (dRVVT) test is more sensitive and specific for the detection
Neurological manifestations including migraine (20.2%) (see of an LA than is prolongation of the aPTT.
Table 16.2)
Stroke (19.8%) Myth: The LA affects the aPTT, not the prothrombin time (PT).
Transient ischemic attacks (11.1%)
Epilepsy (7.0%) Reality: On the contrary, an LA does affect the PT in some
Cardiac valve disease (11.6%) (Fig. 16.5) cases. A prolonged PT might indicate an extremely high level
Myocardial infarction (5.5%) of LA. Two tests that are not commonly affected by LA are
Hematological manifestations the chromogenic factor X and the prothrombin–proconvertin
Thrombocytopenia (29.6%) time. However, these assays are not available routinely in
Hemolytic anemia (9.7%)
most laboratories (Moll and Ortel 1997).
Obsteric complications
Early and late fetal loss (52.5%) Pearl: Many medications may be associated with the produc-
Pre-eclampsia (9.5%) tion of aPL.
Premature delivery (10.6%)
Renal disease (2.7%) Comment: Many conditions other than the primary APS and
SLE can be associated with the finding of aPL in serum.
However, in many other such instances, the finding of aPL
does not indicate an individual at heightened risk for throm-
Table 16.2 Central nervous system manifestations of the antiphos-
pholipid syndrome (Adapted from Sanna et al. 2006) bosis or other APS complications. As examples, aPL occur
frequently in infections such as HIV and syphilis, as well as
Classic cerebrovascular disease complications: in association with some types of malignancy or following
• Ischemic strokes exposure to certain drugs. However, unless these antibodies
• Transient ischemic attacks are directed against b-2-glycoprotein-1, the likelihood that
• Acute ischemic encephalopathy they are clinically significant is very low.
• Cerebral venous sinus thrombosis
Other well-recognized (but often unproven) associations: Pearl: Many clinical features widely acknowledged to be
• Headaches, including migraine associated with the APS are not included in classification
• Cognitive dysfunction, including dementia criteria for definite cases of this condition.
• Seizures, abnormal electroencephalography
• Chorea Comment: Classification criteria for definite APS have been
• Transverse myelopathy validated (Table 16.3) (Wilson et al. 1999; Lockshin et al.
• Demyelinating syndromes 2000) and updated (Miyakis et al. 2006). These classification
• Idiopathic intracranial hypertension criteria were developed for use in research studies rather than
• Sensorineural hearing loss as diagnostic criteria, which as yet do not exist. Many clini-
• Guillain–Barre syndrome
cal manifestations generally considered to be present in some
• Transient global amnesia
• Ocular ischemic syndromes
patients with the APS, e.g., thrombocytopenia, hemolytic
• Dystonia–Parkinsonism anemia, transient ischemic attacks, seizures transverse myeli-
• Progressive supranuclear palsy tis (Fig. 16.1), livedo reticularis, valvular heart disease,
demyelinating syndromes (Fig. 16.2), chorea, and migraine
were not included in the final criteria established for “defi-
Pearl: Multiple different autoantibodies can cause the lupus nite” APS (Miyakis et al. 2006). The exclusion of such crite-
anticoagulant phenomenon. ria from the classification criteria results in higher specificity
but lower sensitivity.
Comment: A broad spectrum of autoantibodies that are directed
In clinical practice, sound judgment often dictates that a
against phospholipid-binding proteins can be associated with
clinician consider a patient with such clinical features in the
LA activity. Such autoantibodies include those directed against
context of persistently positive aPL to have the APS and treat
prothrombin, phosphatidylserine, phosphatidylethanolamine,
the patient accordingly.
and b2-glycoprotein 1. However, clinical testing for such other
specific antibodies seldom adds much to the standard labora- Myth: A patient with thrombosis and/or pregnancy morbidity
tory aPL assays for aCL antibodies and LA. can only be diagnosed with APS with moderate to high titers
16 The Antiphospholipid Syndrome 175

Fig. 16.1 Transverse myelitis in

a patient with SLE and aPL. A
large high signal lesion is seen in
the upper cervical cord with cord
edema (D'Cruz et al. 2004)

Fig. 16.2 White matter hyper-intense lesions on MRI in a patient with

aPL resembling demyelination

of aCL antibodies and/or a positive LA and/or anti-b-2-gly-

Pearl: The only types of aCL that are associated with an
coprotein 1 antibodies tested at least 12 weeks apart.
increased risk of thrombosis are those that are dependent on
Reality: This myth gets at another common point of confu- b2-glycoprotein 1.
sion between classification and diagnostic criteria. There are
Comment: This point was mentioned above but is worth
no diagnostic criteria for the APS. The current classification
re-emphasizing. Patients with positive aCL tests but negative
criteria were designed to be highly specific so as to avoid
assays for anti-b2glycoprotein 1 are not at increased risk for
bias and to foster homogeneity within research study popula-
clotting or other APS manifestations. These b2glycoprotein 1
tions. High specificity inevitably comes at a cost of lower
independent antibodies are commonly found in infections
sensitivity and means that many patients whose clinical diag-
such as HIV, hepatitis C, leprosy, syphilis, and other infec-
nosis is the APS do not meet classification criteria.
tions, as well as in association with certain drugs.
A 35-year-old woman with extensive livedo reticularis
and a previous pregnancy resulting in intrauterine death of
the fetus who presents with a pulmonary embolism in the
context of persistent low-titer aCL antibodies would be con- 16.2 Risk Factors for Thrombosis in APS
sidered by most experts to have APS. However, the patient
would not meet the APS classification criteria.
Pearl: Previous thrombosis is the strongest predictor of
Once anticoagulation with warfarin has begun, it is often
future events.
difficult to interpret LA assays. This fact complicates the
point within the classification criteria that requires retesting Comment: History of a clinical event compatible with the APS
a minimum of 12 weeks later. Strict classification of the in the context of persistent moderate to high aCL levels or LA
patient then becomes difficult, particularly if antibodies to is the most powerful predictor of a future APS
aCL and b2-glycoprotein 1 are negative and the patient’s event (Akimoto et al. 2005). The importance of previous throm-
sole aPL antibody has been the LA. bosis was highlighted by the recurrence rate in a retrospective
176 D. P. D’Cruz et al.

Table 16.3 Classification criteria for the antiphospholipid syndrome time, mean dose of glucocorticoids, and shorter disease dura-
(From Miyakis et al. 2006. Reprinted from J Thromb Haemost, with tion (Calvo-Alen et al. 2005).
kind permission from Wiley Interscience)
Antiphospholipid antibody syndrome (APS) is present if at least one Pearl: aPL are found in approximately 30–40% of patients
of the clinical criteria and one of the laboratory criteria that with “idiopathic” immune thrombocytopenia (ITP).
follow are met
Clinical criteria Comment: The recognition of aPL in ITP is important
1. Vascular thrombosis because conventional thinking associates ITP with an
One or more clinical episodes of arterial, venous, or small vessel increased risk of bleeding. In contrast, patients with aPL may
thrombosis, in any tissue or organ. Thrombosis must be
confirmed by objective validated criteria (i.e. unequivocal suffer thromboses even in the setting of thrombocytopenia
findings of appropriate imaging studies or histopathology). For (Diz-Küçükkaya et al. 2001).
histopathologic confirmation, thrombosis should be present
without significant evidence of inflammation in the vessel wall Pearl: Livedo reticularis is strongly associated with the
2. Pregnancy morbidity occurrence of arterial ischemic events in the APS.
(a) One or more unexplained deaths of a morphologically normal
fetus at or beyond the 10th week of gestation, with normal fetal
Comment: Livedo reticularis is strongly associated with arte-
morphology documented by ultrasound or by direct examination rial lesions in APS. In one study that involved 200 consecu-
of the fetus, or tively evaluated patients, livedo reticularis was associated
(b) One or more premature births of a morphologically normal with cerebral or ocular ischemic events with an odds ratio of
neonate before the 34th week of gestation because of: 10.8 (95% CI 5.2 – 22.5) (Francès et al. 2005). In contrast,
(i) eclampsia or severe preeclampsia defined according to
standard definitions, or (ii) recognized features of placental livedo reticularis was observed with decreased frequency in
insufficiency, or patients who only experienced venous thromboses (odds
(c) Three or more unexplained consecutive spontaneous abortions ratio 0.2; 9% CI 0.1 – 0.5).
before the 10th week of gestation, with maternal anatomic or
hormonal abnormalities and paternal and maternal chromosomal Myth: Livedo reticularis is the most common cutaneous
causes excluded manifestation of APS.
In studies of populations of patients who have more than one type of
pregnancy morbidity, investigators are strongly encouraged to Reality: The literature is remarkably muddy here. This issue
stratify groups of subjects according to a, b, or c above. is whether the proper term for the lacy discoloration of the
Laboratory criteria
skin in this condition is livedo reticularis as opposed to livedo
1. Lupus anticoagulant (LA) present in plasma, on two or more
occasions at least 12 weeks apart, detected according to the racemosa. Since the earliest descriptions of the APS, the term
guidelines of the International Society on Thrombosis and “livedo reticularis” has been used rather indiscriminately.
Haemostasis (Scientific Subcommittee on LAs/phospholipid- Clinicians have used livedo reticularis in the context of the
dependent antibodies)
APS when, in many instances, livedo racemosa would have
2. Anticardiolipin (delete extra 'c') (aCL) antibody of IgG and/or IgM
isotype in serum or plasma, present in medium or high titer (i.e. >40 been the more accurate term. Livedo racemosa, a term coined
GPL or MPL, or >the 99th percentile), on two or more occasions, at by Ehrmann in 1907 (Ehrmann 1907), is a much more strik-
least 12 weeks apart, measured by a standardized ELISA ing cutaneous finding than is livedo reticularis, and livedo
3. Anti-beta2 glycoprotein-I antibody of IgG and/or IgM isotype in racemosa is far more likely to indicate a serious medical con-
serum or plasma (in titer > the 99th percentile), present on two
dition, be it the APS, polyarteritis nodosa, or something else.
or more occasions, at least 12 weeks apart, measured by a
standardized ELISA, according to recommended procedures Livedo racemosa is the most common dermatologic man-
ifestation of APS and is frequently associated with cerebro-
vascular events, arterial thrombosis, and pregnancy morbidity.
study in which patients with the APS and previous thrombosis The racemosa pattern may be distinguished by the broken
had a recurrence rate of 20% per patient-year of follow-up pattern, widespread extent (especially on the trunk), and its
(Khamashta et al. 1995). minimal alteration during cool temperatures (Fig. 16.3). In
Other risk factors also contribute to the development of a contrast to livedo racemosa, livedo reticularis often occurs in
first thrombotic event in the presence of aPL. For example, at physiologic settings such as cold weather and is not always
the time of the initial thrombosis, 50% of aPL-positive associated with an underlying disease.
patients have coincident risk factors (Giron-Gonzalez et al. Livedo racemosa versus livedo reticularis is also dis-
2004). These include recent surgery and prolonged immobi- cussed in the chapter on polyarteritis nodosa (Chapter 26).
lization (associated with venous thrombosis), and hypercho-
Pearl: APS nephropathy may complicate lupus nephritis and
lesterolemia and arterial hypertension (linked to arterial
should be considered in all patients with SLE, renal disease,
thrombosis). SLE itself is a risk factor for thrombosis, regard-
and positive assays for aPL.
less of the patient’s aPL status. Finally, multivariate models
in studies of APS patients have demonstrated independent Comment: APS nephropathy is a distinct entity with charac-
associations of smoking, older age, disease activity over teristic clinical and histological features (Tektonidou 2004).
16 The Antiphospholipid Syndrome 177

Fig. 16.3 Livedo racemosa. The extensive “broken” pattern indicates

pathologic rather than physiological livedo reticularis

All renal vessels may be affected by the thrombotic microan-

giopathy, but glomerular lesions comprised of basement
membrane wrinkling and reduplications appear to be pathog-
nomonic. As acute lesions heal, focal areas of fibrosis and
atrophy may develop. Renal impairment may ensue, depend-
ing on the extent and severity of these processes.
Recent studies have also confirmed the importance of
livedo reticularis (racemosa) as a clinical marker of this syn-
drome in association with hypertension. In addition, asymp-
tomatic renal artery stenosis may be more common than
suggested by early case reports (Sangle 2003). Clinicians
should thus assess aPL positive patients very carefully for
Fig. 16.4 High signal lesions on MRI compatible with small vessel
the development of renal disease and patients with lupus ischemia in a patient with APS
nephritis who develop an additional APS nephropathy should
be considered for long-term anticoagulation as well as immu-
nosuppression (D’Cruz 2005).
Anticoagulation is the mainstay of treatment for APS
nephropathy, but this strategy remains imperfect. Close
attention to other risk factors, especially hypertension and
the traditional Framingham risk factors, may reduce the risk
of end-stage renal failure in these patients.
Myth: High signal lesions on brain MRI indicate cerebral
vasculitis.
Reality: This problem arises especially in patients with SLE in
whom the radiologist reports high signal lesions as being con-
sistent with “cerebral vasculitis.” In fact, true cerebral vasculi-
tis is extremely rare in SLE. Moreover, vasculitis is not part of
the APS. High signal white matter lesions correlate with older
age, but also with aPL (Fig. 16.4) (Sanna et al. 2003). Fig. 16.5 Aortic valve vegetation in a patient with APS presenting
with a stroke
Myth: APS is a bland disorder characterized by thrombosis
in the absence of signs of systemic inflammation.
Reality: “Straightforward” APS is usually not associated with catastrophic APS, which is often accompanied by the systemic
signs of systemic inflammation. However, inflammation, partic- inflammatory response syndrome (SIRS). The catastrophic APS
ularly complement activation, does appear to play a central role is difficult to distinguish it from thrombotic thrombocytopenic
in the pathophysiology of this condition (Girardi et al. 2006). purpura/hemolytic uremic syndrome (Asherson et al. 2007).
There is one important exception to the “bland inflamma- Pearl: Patients with the APS should not receive estrogen-
tion” rule in the APS. This occurs in the setting of the containing medications.
178 D. P. D’Cruz et al.

Comment: No discussion here. Both the combined oral con- hazard. Major bleeds occur in up to 40% of patients in some
traceptive pill and hormone replacement therapy are abso- long-term studies of non-APS patients (Fihn et al. 1993).
lutely contraindicated in APS patients because of their Data on APS patients from both retrospective cohort reviews
association with thromboembolic disease. and randomized controlled trials suggest that the risk of
recurrent thrombotic events exceeds the risk of bleeding,
Myth: A baby aspirin a day is useful in primary prophylaxis
which is relatively low (Finazzi 2005; Crowther et al. 2003;
for patients who have aPL but no history of clinical events
Cervera et al. 2008).
compatible with the APS.
The intensity of anticoagulation has been debated.
Reality: Many clinicians recommend low-dose aspirin for Randomized controlled trials have suggested that INR levels
primary prevention in patients without previous thrombosis, between 2.0 and 3.0 are sufficient in uncomplicated APS
particularly if titers of aPL are high. However, there is little patients with only venous thrombotic events (Finazzi et al.
in the way of solid data to support this. Indeed, one placebo- 2005; Crowther et al. 2003). However, in patients with arterial
controlled trial did not show any protective benefit against events or with recurrent thrombotic events, an INR between
thrombosis, albeit follow-up in that trial was limited (Erkan 3.0 and 4.0 may be required (Ruiz-Irastorza et al. 2007).
et al. 2007). Although not treating asymptomatic aPL-posi- Once patients are established on warfarin, aspirin does
tive individuals with aspirin remains a reasonable option not confer any added advantage and can be discontinued.
(Lim et al. 2006), many expert clinicians give aspirin anyway
Myth: A young woman who has suffered a stroke and has
in the interest of doing something that might help and is
strongly positive aPL assays can be managed safely with
unlikely to cause harm.
aspirin alone and secondary prevention measures.
For SLE patients, some evidence supports the use of
hydroxychloroquine in this setting. In the LUMINA study, Reality: Among patients with the APS, strokes tend to occur
hydroxychloroquine was protective against thrombosis in a mean of 2 decades earlier when compared with the general
univariate analysis but not in multivariate analysis (Ho et al. population (Verro et al. 1998; Brey 2004). Indeed, one sys-
2005). A recent systematic review endorsed antimalarial tematic review suggested that aPL were an independent risk
therapy for both its disease-modifying (i.e., “anti-inflamma- factor for incidental ischemic stroke, although data on their
tory”) effects as well as its protective effects against throm- role in recurrent ischemic stroke was less clear (Verro et al.
bosis (Ruiz-Irastorza et al. 2008). 1998). Nonetheless, once ischemic stroke occurs, secondary
prevention becomes a major issue.
Pearl: Patients with aPL who undertake long airplane flights
Stroke management in APS patients has been extremely
are increased risk of venous thrombosis.
controversial. Older retrospective studies have shown that
Comment: Prolonged journeys in which passengers are seated recurrent thrombotic events are significantly less likely on
in cramped conditions for lengthy periods of time may be warfarin therapy with a target INR > 3.0 with or without low-
associated with “traveller’s thrombosis” (Hughes et al. 2003). dose aspirin (Khamashta et al. 1995; Rosove and Brewer
No studies have addressed the possible relevance of aPL in 1992). One retrospective case series that examined the out-
this syndrome. In the absence of evidence, it seems reasonable comes of 66 APS patients with previous thrombosis reported
to recommend simple measures such as support stockings, an alarming rate of recurrent thrombosis: 9.1 cases per 100
hydration, exercises, and, for patients on treatment, ensuring patients-years (95% CI, 3.3 – 19.6) (Ruiz-Irastorza et al.
therapeutic levels of anticoagulation for APS patients who 2002). Most recurrent events developed at INRs between 2.1
undertake long journeys by any mode of transportation. and 2.6, a level of anticoagulation assumed by many clini-
In high-risk aPL-positive individuals who have not suf- cians to be therapeutic.
fered thromboses but who are deemed to be at high risk for Results and the recommendations stemming from the
thrombosis for one reason or another (e.g., high aPL titers or Antiphospholipid Antibodies and Stroke Study (APASS) proved
the presence of both aCL and an LA), some clinicians rec- to be very controversial. In that study, 1770 aPL-positive patients
ommend a single injection of low molecular weight heparin with ischemic stroke were allocated randomly to receive either
prior to commencing the journey, provided that the patient’s aspirin (325 mg/day) or warfarin (target INR: 1.4–2.8) (The
renal function is normal. Other high-risk periods for patients APASS Investigators 2004). Over a 2-year follow-up period, the
with aPL include the post-operative period, especially after presence of aPL among patients with ischemic strokes did not
abdominal, gynecological, and orthopedic surgery, and the predict subsequent thromboembolic events. Recurrent throm-
post-partum period. botic events developed in 24% of patients with or without aPL.
The main study conclusions of the APASS were: (a) LA
Pearl: Anticoagulation in a patient with the APS should be
or aCL testing was not important for prognosis or treatment
lifelong.
of unselected patients with recent ischemic stroke; (b) in
Comment: Several studies support this consensus view (Lim patients with stroke who tested positive for aPL, warfarin
et al. 2006). However, lifelong anticoagulation is not without therapy with a target INR of 1.4–2.8 was not associated with
16 The Antiphospholipid Syndrome 179

fewer recurrent events than aspirin. The small group of 120 Pearl: Women with aPL and previous pregnancy morbidity
patients who tested positive for both aCL and LA had a are at increased of future thrombosis.
higher risk of thromboembolic events or death regardless of
Comment: Many women with recurrent fetal losses as the
the type of treatment. Unfortunately, several important study
sole manifestation of APS attend recurrent miscarriage clin-
flaws hamper generalization of the results to the APS popula-
ics. However, once these women deliver successfully, they
tion (Ruiz-Irastorza et al. 2007).
are often discharged from these clinics without counseling to
Pearl: Anticoagulation with warfarin may be continued in reduce the risk of a future thrombotic event. There is a con-
patients with mild stable thrombocytopenia. sensus that these women should be considered to be at
increased risk of a future thrombotic event and management
Comment: Thrombocytopenia is common among APS
should aim to reduce potential risk factors. However, there is
patients. Paradoxically, however, such patients remain at risk
no consensus on the best primary prevention strategy. Many
for thrombosis despite their thrombocytopenia. Warfarin
experts prescribe low-dose aspirin therapy, but there are no
should be continued at therapeutic levels unless the degree of
robust data to support this practice.
thrombocytopenia is severe.
In patients with severe thrombocytopenia, glucocorticoids
and intravenous immunoglobulins are usually effective. Other
agents such as danazol, dapsone, and even aspirin have been
tried. There is some debate as to the effectiveness of splenec-
tomy in APS patients, but some studies suggest that splenec- References
tomy contributes to more effective control of thrombocytopenia
over the long term (Hakim et al. 1998; Galindo et al. 1999). Akimoto T, Kobayashi S, Tamura N, et al Risk factors for recurrent
thrombosis: Prospective study of a cohort of Japanese systemic
Pearl: Women with menorrhagia may be managed with hor- lupus erythematosus. Angiology 2005; 56:601–609
monal intrauterine devices. Asherson RA, Pierangeli SS, Cervera R. Is there a microangiopathic
antiphospholipid syndrome? Ann Rheum Dis 2007; 66:429–342
Comment: A frequent problem in the management of female Baglin T, Barrowcliffe TW, Cohen A, et al Guidelines on the use and
patients is menorrhagia, particularly when INRs are main- monitoring of heparin. Br J Haematol 2006; 133:19–34
tained at high levels. In this situation, the “Mirena” coil, an Brandt JT, Triplett DA, Alving B, et al Criteria for the diagnosis of
lupus anticoagulants: An update. On behalf of the Subcommittee on
intrauterine coil with a silastic capsule containing levonorg- Lupus Anticoagulant/Antiphospholipid Antibody of the Scientific
estrel, is helpful in decreasing the volume of menstrual flow and Standardisation Committee of the ISTH. Thromb Haemost
(Pisoni et al. 2006). 1995; 74:1185–1190
Brey RL. Management of the neurological manifestations of APS –
Pearl: Data on the optimal management of women with the what do the trials tell us? Thromb Research 2004; 114:489–499
APS during pregnancy are conflicting. Bucciarelli S, Espinosa G, Cervera R, et al Mortality in the catastrophic
antiphospholipid syndrome: Causes of death and prognostic factors
Comment: The current choices are aspirin, heparin, or both in a series of 250 patients. Arthritis Rheum 2006; 54:2568–2576
(Petri and Qazi 2006). This area remains controversial, espe- Calvo-Alén J, Toloza SM, Fernández M, et al Systemic lupus erythema-
tosus in a multiethnic US cohort (LUMINA). XXV. Smoking, older
cially since a randomized controlled trial demonstrated that age, disease activity, lupus anticoagulant, and glucocorticoid dose
good pregnancy success rates were achievable with low-dose as risk factors for the occurrence of venous thrombosis in lupus
aspirin alone and there appeared to be no advantage to add- patients. Arthritis Rheum 2005; 52:2060–2068
ing low molecular weight heparin (Farquharson et al. 2002). Carlin AJ, Farquharson RG, Quenby SM, et al Prospective observational
study of bone mineral density during pregnancy: Low molecular
These data contrasted with previous controlled trials that weight heparin versus control. Hum Reprod 2004; 19:1211–1214
showed a significant benefit of heparin with aspirin over Cervera R, Khamashta MA, Font J, et al Systemic lupus erythematosus:
aspirin alone (Rai et al. 1997; Kutteh 1996). Clinical and immunologic patterns of disease expression in a cohort
Heparin does not cross the placenta and is not known to of 1,000 patients. The European Working Party on Systemic Lupus
Erythematosus. Medicine (Baltimore) 1993; 72:113–124
cause any adverse fetal effects (Baglin et al. 2006). In con- Cervera R, Piette JC, Font J, et al Antiphospholipid syndrome: Clinical
trast, warfarin exposure between 6 and 12 weeks of gestation and immunologic manifestations and patterns of disease expression
can be associated with an embryopathy characterized by in a cohort of 1,000 patients. Arthritis Rheum 2002; 46:1019–1027
stippled epiphyses and nasal hypoplasia. Thus, if heparin is Cervera R, Khamashta MA, Shoenfeld Y, et al Morbidity and mortality in
the antiphospholipid syndrome during a 5-year period: A multicenter
selected, the change from warfarin to heparin should be prospective study of 1,000 patients. Ann Rheum Dis 2008 Sep 18
achieved prior to 6 weeks gestation. [Epub ahead of print]
The risk of osteoporosis in the mother from long-term Crowther MA, Ginsberg JS, Julian J, et al A comparison of two intensi-
heparin has probably been overestimated. A controlled study ties of warfarin for the prevention of recurrent thrombosis in patients
with the antiphospholipid antibody syndrome. N Engl J Med 2003;
showed that bone loss is a feature of normal pregnancy and 349:1133–1138
that low molecular weight heparin does not exacerbate this D’Cruz DP. Renal manifestations of the antiphospholipid syndrome.
loss (Carlin et al. 2004). Lupus 2005; 14:45–48
180 D. P. D’Cruz et al.

Diz-Küçükkaya R, Hacihanefioğlu A, Yenerel M, et al Antiphospholipid Lockshin MD, Sammaritano LR, Schwartzman S. Validation of the
antibodies and antiphospholipid syndrome in patients presenting Sapporo criteria for antiphospholipid syndrome. Arthritis Rheum
with immune thrombocytopenic purpura: A prospective cohort 2000; 43:440–443
study. Blood 2001; 98:1760–1764 Miyakis S, Lockshin MD, Atsumi T, et al International consensus state-
Ehrmann S. Ein Gefaessprozess Bei Lues. Wien Med Wochenschr. ment on an update of the classification criteria for definite antiphos-
1907;57:777 pholipid syndrome (APS). J Thromb Haemost 2006; 4:295–306
Erkan D, Harrison MJ, Levy R, et al Aspirin for primary thrombosis Moll S, Ortel TL. Monitoring warfarin therapy in patients with lupus
prevention in the antiphospholipid syndrome: A randomized, anticoagulants. Ann Intern Med 1997; 127:177–185
double-blind, placebo-controlled trial in asymptomatic Petri M, Qazi U. Management of antiphospholipid syndrome in preg-
antiphospholipid antibody-positive individuals. Arthritis Rheum nancy. Rheum Dis Clin North Am 2006; 32:591–607
2007; 56: 2382–2391 Pisoni CN, Cuadrado MJ, Khamashta MA, Hunt BJ. Treatment of men-
Farquharson RG, Quenby S, Greaves M. Antiphospholipid syndrome in orrhagia associated with oral anticoagulation: Efficacy and safety of
pregnancy: A randomized, controlled trial of treatment. Obstet the levonorgestrel releasing intrauterine device (Mirena coil). Lupus
Gynecol 2002; 100:408–413 2006; 15:877–880
Fihn SD, McDonell M, Martin D, et al Risk factors for complications of Rai R, Cohen H, Dave M, Regan L. Randomised controlled trial of
chronic anticoagulation. A multicenter study. Warfarin Optimized aspirin and aspirin plus heparin in pregnant women with recurrent
Outpatient Follow-up Study Group. Ann Intern Med 1993; 118: miscarriage associated with phospholipid antibodies (or antiphos-
511–520 pholipid antibodies). BMJ 1997; 314:253–257
Finazzi G, Marchioli R, Brancaccio V, et al A randomized clinical trial Rosove MH, Brewer PM. Antiphospholipid thrombosis: clinical course
of high-intensity warfarin vs. conventional antithrombotic therapy after the first thrombotic event in 70 patients. Ann Intern Med 1992;
for the prevention of recurrent thrombosis in patients with the 117:303–308
antiphospholipid syndrome (WAPS). J Thromb Haemost 2005; Ruiz-Irastorza G, Khamashta MA, Hunt BJ, et al Bleeding and recur-
3:848–853 rent thrombosis in definite antiphospholipid syndrome. Arch Intern
Francès C, Niang S, Laffitte E, et al Dermatologic manifestations of the Med 2002; 162:1164–1169
antiphospholipid syndrome: Two hundred consecutive cases. Ruiz-Irastorza G, Egurbide MV, Ugalde J, Aguirre C. High impact of
Arthritis Rheum 2005; 52:1785–1793 antiphospholipid syndrome on irreversible organ damage and sur-
Galindo M, Khamashta MA, Hughes GR. Splenectomy for refractory vival of patients with systemic lupus erythematosus. Arch Intern
thrombocytopenia in the antiphospholipid syndrome. Rheumatology Med 2004; 164:77–82
(Oxford) 1999; 38:848–853 Ruiz-Irastorza G, Hunt BJ, Khamashta MA. A systematic review of sec-
Girardi G, Yarilin D, Thurman JM, et al Complement activation induces ondary thromboprophylaxis in patients with antiphospholipid anti-
dysregulation of angiogenic factors and causes fetal rejection and bodies. Arthritis Rheum (Arthritis Care Res) 2007; 57:1487–1495
growth restriction. J Exp Med 2006; 203:2165–2175 Ruiz-Irastorza G, Ramos-Casals M, Brito-Zeron P, Khamashta MA.
Giron-Gonzalez JA, Garcia del Rio E, Rodriguez C, et al Antiphospholipid Clinical efficacy and side effects of antimalarials in systemic lupus
syndrome and asymptomatic carriers of antiphospholipid antibody: erythematosus: A systematic review. Ann Rheum Dis 2008 Dec 22
Prospective analysis of 404 individuals. J Rheumatol 2004; 31: [Epub ahead of print]
1560–1567 Sangle SR, Jan W, Lau IS, et al Coeliac artery stenosis and antiphos-
Hakim AJ, Machin SJ, Isenberg DA. Autoimmune thrombocytopenia in pholipid (Hughes) syndrome/antiphospholipid antibodies. Clin Exp
primary antiphospholipid syndrome and systemic lupus erythema- Rheumatol 2006; 24:349
tosus: The response to splenectomy. Semin Arthritis Rheum 1998; Sanna G, Bertolaccini ML, Cuadrado MJ, et al Neuropsychiatric mani-
28:20–25 festations in systemic lupus erythematosus: Prevalence and associa-
Ho KT, Ahn CW, Alarcon GS, et al Systemic lupus erythematosus in a tion with antiphospholipid antibodies. J Rheumatol 2003;
multiethnic cohort (LUMINA): XXVIII. Factors predictive of 30:985–992
thrombotic events. Rheumatology (Oxford) 2005; 44:1303–1307 Sanna G, D’Cruz D, Cuadrado MJ. Cerebral manifestations in the
Hochberg MC. Updating the American College of Rheumatology antiphospholipid (Hughes) syndrome. Rheum Dis Clin North Am
revised criteria for the classification of systemic lupus erythemato- 2006; 32:465–490
sus. Arthritis Rheum 1997; 40:1725 Schulman S, Svenungsson E, Granqvist S, The Duration of Antico-
Hughes GRV. Thrombosis, abortion, cerebral disease, and the lupus agulation Study Group. Anticardiolipin antibodies predict early
anticoagulant. Br Med J 1983; 287:1088–1089 recurrence of thromboembolism and death among patients with
Hughes RJ, Hopkins RJ, Hill S, et al Frequency of venous thromboem- venous thromboembolism following anticoagulant therapy. Am J
bolism in low to moderate risk long distance air travellers: The New Med 1998; 104:332–338
Zealand Air Traveller’s Thrombosis (NZATT) study. Lancet 2003; Shah NM, Khamashta MA, Atsumi T, Hughes GRV. Outcome of
362:2039–2044 patients with anticardiolipin anti-bodies: A 10 year follow up of 52
Jouhikainen T, Stephansson E, Leirisalo-Repo M. Lupus anticoagulant patients. Lupus 1998; 7:3–6
as a prognostic marker in systemic lupus erythematosus. Tektonidou MG, Sotsiou F, Nakopoulou L, et al Antiphospholipid
Br J Rheumatol 1993; 32:568–573 syndrome nephropathy in patients with systemic lupus erythema-
Khamashta MA, Cuadrado MJ, Mujic F, et al The management of tosus and antiphospholipid antibodies: Prevalence, clinical asso-
thrombosis in the antiphospholipid antibody syndrome. N Engl ciations, and long-term outcome. Arthritis Rheum 2004; 50:
J Med 1995; 332:993–997 2569–2579
Kutteh WH. Antiphospholipid antibody-associated recurrent pregnancy The APASS Investigators. Antiphospholipid antibodies and subsequent
loss: Treatment with heparin and low-dose aspirin is superior to low- thrombo-occlusive events in patients with ischemic stroke. JAMA
dose aspirin alone. Am J Obstet Gynecol 1996; 174:1584–1589 2004; 291:576–584
Lampropoulos CE, Koutroumanidis M, Reynolds PP, et al Electroen- Verro P, Levine SR, Tietjen GE. Cerebrovascular ischemic events with
cephalography in the assessment of neuropsychiatric manifestations high positive anticardiolipin antibodies. Stroke 1998; 29:2245–2253
in antiphospholipid syndrome and systemic lupus erythematosus. Wilson W, Gharavi A, Koike T, et al International consensus statement
Arthritis Rheum 2005; 52:841–846 on preliminary classification for definite antiphospholipid syn-
Lim W, Crowther MA, Eikelboom JW. Management of antiphospholipid drome. Report of an International Workshop. Arthritis Rheum 1999;
antibody syndrome: A systematic review. JAMA 2006; 295:1050–1057 42:1309–1311
 Pregnancy in the Rheumatic Diseases
17
Megan E.B. Clowse and Rosalind Ramsey-Goldman

17.2 Introduction
17.1 Overview of Pregnancy in the
Rheumatic Diseases
Few conditions in rheumatology are fraught with greater
concern and more unknowns than those of patients’ fertility
› Few conditions in rheumatology are fraught with and pregnancy. Women with rheumatic disease pose special
greater concern and more unknowns than those of fer- problems during the periconception period and puerperium
tility and pregnancy. due to the effects of pregnancy on their underlying disease,
› Patients with rheumatologic disease are not at higher the impact of their underlying disease on pregnancy, and the
risk for infertility than the general population, unless influence of medications on both. In this chapter, we begin
they have received cyclophosphamide. with General Considerations, then discuss the use of specific
› In general, the inflammation from active rheumato- medications in pregnancy, and conclude with Pearls and
logic disease is more dangerous to a pregnancy than Myths related to individual rheumatic disorders and their
immunosuppressant medications. interactions with pregnancy.
› Medications to continue if needed to control disease:
hydroxychloroquine, azathioprine, sulfasalazine, and
prednisone.
› Medications to discontinue prior to pregnancy: metho- 17.3 General Considerations
trexate, leflunamide, mycophenolate mofetil, cyclophos-
phamide. Myth: Women with rheumatic diseases cannot become preg-
› Normal pregnancy can alter laboratory findings. In nant because of infertility.
particular, the creatinine should fall and proteinuria
Reality: Infertility is not more common among women with
may increase mildly. The sedimentation rate and com-
rheumatologic diseases in general. Women with systemic
plement may increase.
lupus erythematosus (SLE) have rates of pregnancy similar
› Flares in lupus during pregnancy tend to be mild.
to those of sisters and friends, but lupus pregnancies are more
Patients at highest risk for a severe flare have active
likely to result in gestational loss (Petri and Allbritton 1993).
lupus at the time of conception.
In addition, fertility may be impaired in individual patients
› Even treated with low molecular weight heparin and
by anovulation during episodes of active disease or chronic
aspirin 81mg a day, up to 25% of pregnancies in
renal failure, or administration of high dose of glucocorti-
women with antiphospholipid syndrome will result in
coids and cyclophosphamide. Patients requiring cyclophos-
a pregnancy loss.
phamide may be protected from infertility by the use of
› Maternal SSA/Ro and SSB/La antibodies put a baby at
godanotrophins releasing analogs (Somers et al. 2005;
risk for neonatal lupus. Up to 50% of these mothers do
Raptopoulou et al. 2004).
not fit criteria for a rheumatologic disease.
There is a debate about whether antinuclear antibodies
› Rheumatoid arthritis will improve in many women
(ANAs) and antiphospholipid antibodies (aPL) impair
during pregnancy, but often flares post-partum.
implantation, leading to infertility (Kikuchi et al. 2003).
Medications may be tapered during pregnancy but
Positive ANAs and aPL are more common among women
should be restarted following delivery to avoid a severe
evaluated in infertility clinics. Treatment directed at the
flare.
underlying rheumatic disease, e.g., with anticoagulation or

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 181

DOI:10.1007/978-1-84800-934-9_17, © Springer Science + Business Media B.V. 2009
182 M. E. B. Clowse et al.

low-dose glucocorticoids, may improve pregnancy rates in Reality: The ESR is an unreliable marker of inflammation in
patients suffering from infertility. pregnancy. Pregnancy itself commonly causes an elevation
Women with rheumatoid arthritis (RA) have (on average) of the ESR, to greater than 100 mm/h in some cases. Thus,
smaller families than do women without RA, but this is more measurements of the ESR should be avoided in pregnant
likely the result of personal choice and difficulty with inter- women; it rarely provides useful information but may cause
course than of intrinsic infertility (Katz 2006). undue concern. The C-reactive protein (CRP) might be a bet-
ter measure of inflammation in pregnancy, but this has not
Myth: Ovarian stimulation can induce or exacerbate SLE.
been confirmed in prospective investigations.
Reality: In theory, ovulation stimulation as a fertility treatment
Myth: Dyspnea during pregnancy means that the woman has
could induce SLE, but studies have not confirmed this potential
pleurisy or a pulmonary embolism.
side effect. On the other hand, patients with SLE or primary
APS, who are undergoing infertility treatment, could be at risk Reality: Dyspnea is a common finding during pregnancy in
of flare or thrombosis. When carefully monitored in selected healthy women. In the first trimester, increased levels of pro-
patients, assisted reproductive technologies do not appear to gesterone lead to an increased respiratory rate, but this is
exacerbate SLE in patients whose disease is inactive. rarely associated with the sensation of shortness of breath. In
Stable autoimmune diseases that are not associated with the third trimester, the increasing size of the uterus leads to
major organ damage probably do not affect the outcomes of compression of the lungs, which can promote dyspnea. This
pregnancies conceived through assisted reproductive tech- dyspnea may be accompanied by pain in the lower chest or
nologies (Lockshin 2004). The children born from pregnan- abdomen from the baby, but this pain should be distinguish-
cies conceived through assisted reproductive technologies able from pleurisy.
are apparently normal at birth, whether or not the genetic or Pregnancy is a hypercoagulable state, probably because
birth mother has autoimmune disease. Few long-term follow- the body has evolved to prevent bleeding from the highly
up data are available on the children from these pregnancies. vascular placenta. For this reason, healthy women are at a
For patients with antiphospholipid syndrome (or, in selected threefold increased risk for deep venous thrombosis (DVT)
cases with antiphospholipid antibodies without clinical events), or pulmonary embolism (PE) during pregnancy and the puer-
most experts recommend prophylaxis to prevent thrombosis perium. Women with SLE or the antiphospholipid syndrome
during the conception period. The combination of unfraction- (APS) are at higher risk for these complications when not
ated or low molecular weight heparin and aspirin is the regi- pregnant, and the risk is even greater during pregnancy
men employed most often. Anticoagulation should be (Clowse et al. 2008). Low-dose aspirin may attenuate this
suspended 24 h before oocyte retrieval and reinstituted 24 h risk to some extent, but this intervention is not an evidence-
after to reduce the risk of hemorrhage (Costa and Colia 2008). based therapy. Vigilant evaluation of symptoms, such as
Anticoagulation is extended if pregnancy results. unequal swelling of the legs or new dyspnea, is important.
Myth: There is no way to decrease the risk of preeclampsia. Pearl: A creatinine value of 1.0 mg/dL is likely to be abnor-
mal in a woman who is pregnant.
Reality: Preeclampsia, the combination of hypertension and
proteinuria induced by pregnancy, typically develops in the Comment: During pregnancy, a woman’s blood volume
third trimester but can start as early as the 20th week of gesta- increases by 50%, leading to increased renal perfusion and a
tion. Eclampsia, a more severe form of preeclampsia, includes physiologic lowering of the serum creatinine level. One
grand mal seizures. Up to a quarter of women with SLE should anticipate a decline in the creatinine level for all
develop significant hypertension or preeclampsia during preg- women during pregnancy, with a normal creatinine consid-
nancy (Clowse et al. 2008). Preeclampsia places both mother ered to be between 0.4 and 0.7 mg/dL. A serum creatinine of
and child at increased risk for death. The best treatment for 1.0 mg/dL is therefore often a cause for concern.
preeclampsia is delivery. The mother’s symptoms usually Pearl: An increase in proteinuria during pregnancy does not
resolve completely within a few days of delivery. necessarily indicate a flare of renal lupus.
Daily low-dose aspirin taken throughout pregnancy
decreases the risk of preeclampsia by about 20% among Comment: In normal pregnancies, the maternal blood vol-
patients who do not have SLE (Duley et al. 2007). This inter- ume increases by 50%. This leads to an increase in renal
vention has not been tested in women with rheumatologic dis- blood flow that causes, in turn, an increase in proteinuria
ease, but daily treatment with low-dose aspirin may decrease from previously damaged kidneys. A 24-h urine protein
the risk of preeclampsia in this high-risk population. measurement of up to 300 mg is considered within the nor-
mal range for a healthy pregnant woman. For women with a
Myth: An elevated erythrocyte sedimentation rate (ESR) dur- history of lupus nephritis who have some proteinuria at base-
ing pregnancy is a bad prognostic sign. line, the amount of proteinuria can double simply as a
17 Pregnancy in the Rheumatic Diseases 183

function of the increase in renal perfusion. For this reason, an low molecular weight heparin. “Adjunct” medications must
assessment of baseline proteinuria early in pregnancy helps be reviewed carefully along with medications use to treat
enormously in the interpretation of subsequent questions of SLE directly. As examples, both angiotensin converting
renal problems suggested by rising proteinuria. enzyme inhibitors and statins are associated with congenital
malformations with exposure during the first trimester
(Taguchi et al. 2008).
Similar strategies are also utilized when adjusting medi-
17.4 Medications cations prior to conception in women with RA. For example,
methotrexate should be stopped 3–6 months prior to preg-
Pearl: Some rheumatic disease medications should be con- nancy. Leflunomide levels in serum should be assessed and
tinued during pregnancy. cholestyramine administered if appropriate for women with
a history of leflunomide use. However, NSAIDs, sulfasala-
Comment: A healthy mother is the best environment for the
zine, azathioprine, hydroxychloroquine, and glucocorticoids
fetus. Thus, medications should be used before and during
can be continued during pregnancy.
pregnancy if they promote maternal health. Balancing the
risks and benefits of rheumatologic medications to the mother Pearl: A steroid is not a steroid during pregnancy.
and fetus is essential to the decision to continue or discon-
tinue medications during pregnancy. A comprehensive review Comment: Prednisone is metabolized by placental 11-
of the safety of rheumatic disease medications during preg- hydroxygenase before it reaches the fetal circulation. Thus,
nancy was published in 2006 and updated in 2008 (Østensen the fetus is exposed to only 10% of the maternal dose. In
et al. 2006, 2008). contrast, fluorinated glucocorticoids such as betamethasone
For SLE, active disease (particularly renal disease) within and dexamethasone are not metabolized by placental
6 months of conception and the discontinuation of SLE med- 11-hydroxygenase. Much larger doses of those medications
ications are both associated with active disease during preg- reach the fetal circulation. (This is why betamethasone or
nancy and poor pregnancy outcomes (Clowse et al. 2005, dexamethasone is administered to foster lung maturity in
2006b ; Clowse 2007(a,b); Ramsey-Goldman et al. 1993). fetuses at risk for preterm delivery and to fetuses that appear
Approximately 50% of women with SLE have some mea- to have early congenital heart block.) Short-acting prepara-
surable disease activity during pregnancy. Most SLE activity tions, including hydrocortisone, prednisone, prednisolone,
in these situations represents mild to moderate disease, but and methylprednisolone are recommended for maternal dis-
treatment is often necessary to minimize the potential effects ease indications.
of active disease on the mother and fetus. Up to 30% of All glucocorticoid preparations, regardless of their fluori-
women with SLE have pregnancies that are complicated by nation state, have the potential for causing the maternal
the occurrence of very active disease. Under these circum- complications of hypertension, osteopenia, osteonecrosis, dia-
stances, increases in the doses of the patient’s SLE medica- betes, increased susceptibility to infection, cataracts, weight
tions or the addition of new treatments may be required. gain, acne, and proximal muscle weakness in both the non-
The opposite course can be anticipated in RA, which pregnant and pregnant state. The lowest dose needed to treat
tends to improve during pregnancy. Some women with RA either maternal or fetal indications is recommended (Ramsey-
are able to taper their medications altogether during preg- Goldman and Schilling 1996; Lin et al. 2007).
nancy. Unfortunately, the risk of a postpartum flare is high Pearl: Men may have an even higher risk of infertility follow-
among women with RA. Most RA patients are compelled to ing cyclophosphamide therapy than do women.
resume their prepregnancy medication regimen shortly after
delivery (Barrett et al. 1999). This is discussed in greater Comment: Most of the data on fertility risk in men exposed to
detail below. cyclophosphamide comes from the oncology literature, where it
is emphasized that just one exposure can compromise fertility,
Myth: Clinicians should wait until pregnancy is an estab-
even in younger men. This contrasts with the data on women
lished fact before making any changes in a patient’s medical
exposed to cyclophosphamide, whose older age and cumulative
regimen.
dose are important risk factors for premature menopause.
Reality: Preconception counseling is important to discuss One study of male patients with SLE indicated that the
medication options, to adjust medications according to pub- testes are highly susceptible to the toxic effects of cyclophos-
lished recommendations, and to monitor the mother for phamide (Soares et al. 2007). Persistent damage from this
approximately 6 months before attempting conception (Clowse drug is reflected in the primordial sperm cells, leading to sig-
et al., 2005; Clowse 2007). Plans must be in place to switch nificant sperm alterations. Abnormal testicular function,
patients with the antiphospholipid syndrome from warfarin noted even 5 years after drug exposure, was associated with
184 M. E. B. Clowse et al.

elevated FSH levels and lower testicular volumes (Soares Comment: Both MMF and azathioprine are categorized as
et al. 2007). class D medications by the U.S. Food and Drug Administration,
Thus, the story with regard to cyclophosphamide and indicative of positive evidence of risk to a human fetus (FDA
infertility may be at least as glum for men as it is for women. 2007). However, the practical implications for use in preg-
Sperm cyroprecipitation should be discussed early in the dis- nant patients are quite different. MMF is associated with an
ease course, especially if the use of cyclophosphamide is increased risk of spontaneous abortion and congenital mal-
anticipated. It is usually wise for a male patient to bank at formations (FDA 2007). The reported malformations include
least six samples, because a higher number of samples avail- bilateral microtia or anotia (malrotation of the ears), abnor-
able provides greater options for conception. Samples can be malities of the external auditory canals, oral clefts, anomalies
collected every other day. of the limbs, heart, esophagus, and kidney (Østensen et al.
2008). Data from the U.S. National Transplantation Pregnancy
Myth: All fetuses exposed to methotrexate or leflunomide will Registry indicated that among 33 MMF-exposed pregnancies
be born with birth defects. in 24 transplant patients, 15 (45%) resulted in spontaneous
Reality: The congenital anomalies described in animals and abortions. In addition, 4 of the 18 live born infants (22%) had
humans exposed to methotrexate in utero during the first tri- congenital malformations.
mester usually involve the central nervous system, skeletal Azathioprine is also a category D drug in pregnancy.
system, and palate (Wilson et al. 1979; Milunsky et al. 1968). However, this medication is used routinely in pregnant
Intrauterine growth restriction has also been observed. women who have received organ transplants as well as in
Methotrexate is used as an abortifacient for women with those with immune-mediated conditions such as inflamma-
tubal pregnancies. When used for this purpose, the metho- tory bowel disease and SLE who need the drug to control
trexate dose is approximately three times higher than that their underlying disease. No recurrent patterns of malforma-
used normally to treat arthritis. A recent study of 38 pregnan- tions have been identified among infants exposed to azathio-
cies exposed to methotrexate reported 21 live births (55%), prine, despite isolated reports of malformations, intrauterine
eight elective abortions (21%), seven spontaneous abortions growth restriction, and transient chromosomal anomalies in
(18%), and three babies with congenital anomalies (14% of clinically normal infants, severe immune deficiency, cyto-
live births) (Chakravarty et al. 2003). megalovirus infections, temporary neonatal lymphopenia,
Leflunomide causes skeletal and central nervous system and depressed hematopoesis (Østensen et al. 2006).
malformations in rats and rabbits exposed in utero to this The upshot: even though azathioprine bears a category D
drug. Premature birth, blindness, and cerebral palsy were label, this medication is recommended when indicated for
reported in one human child exposed to leflunomide during control of inflammatory disease in the mother. Azathioprine
the first 16 weeks of pregnancy (Neville and McNally 2007). should be used in place of MMF during pregnancy.
However, preliminary results from a study of 63 pregnancies
exposed to leflunomide during the first trimester showed no Pearl: Stopping hydroxychloroquine before or after conception
increased risk of miscarriage or malformation (Østensen leads to an increased risk of a lupus flare during pregnancy.
et al. 2008). Comment: Early reports of in utero chloroquine toxicity led to
Based on the data currently available, we do not recom- reluctance to use this medication during pregnancy (Paufique
mend routine pregnancy termination for fetuses exposed to and Magnard 1969). However, these concerns have been
methotrexate or leflunomide. Instead, careful obstetrical care diminished by more recent reports on pregnancies in mothers
and evaluation should be implemented to identify abnormali- with connective tissue disorders who were exposed to daily
ties as early as possible. hydroxychloroquine or chloroquine during the first trimester
(Levy et al. 1991; Parke and West 1996; Costedoat-Chalumeau
Pearl: Leflunomide must be discontinued long in advance of
et al. 2003; Clowse et al. 2006b). No increases in congenital
any planned pregnancy.
malformations or cardiac conduction disturbances were
Comment: The active metabolite of leflunomide is detectable detected in these studies. A consensus workshop concluded
in plasma up to 3 years after discontinuation of the drug. that antimalarials are recommended to manage disease activ-
Thus, cholestyramine or active charcoal should be given to ity in SLE during pregnancy and its use did not increase the
enhance elimination from the body when conception is risk of fetal malformations (Østensen et al. 2006). Women
desired. Conception should only be attempted after the who continue hydroxychloroquine require lower doses of
plasma level is undetectable (Østensen et al. 2006, 2008). prednisone during pregnancy when compared to women who
do not take this medication (Clowse et al. 2006b).
Pearl: Mycophenolate mofetil (MMF) and azathioprine are
used in comparable clinical scenarios, but considerations in the Myth: Biologic agents are generally regarded as safe during
pregnant patient differ substantially for these two medications. pregnancy.
17 Pregnancy in the Rheumatic Diseases 185

Reality: Experience with TNF-inhibitors (infliximab, etaner- assuming a significant risk of a disease flare, but most flares
cept, adalimumab) and other biologic agents, remains too that occur can be managed effectively. Furthermore, some
limited to claim safety during pregnancy. There are also insuf- can be avoided through the timing of conception to coincide
ficient data to infer anything about the potential effects of with a period of relative disease quiescence. We recommend
these medications on spermatogenesis. Experts recommend that women avoid pregnancy for at least 6 months after an
stopping rituximab prior to pregnancy (Østensen et al. 2008). SLE flare.
Up to one-fourth of women with SLE have significant
Pearl: Breastfeeding does not increase the risk of developing
hypertension or preeclampsia (Clowse et al. 2008). These
a rheumatologic illness.
complications normally resolve following delivery. The inci-
Comment: If there is a relationship between breastfeeding dence of end-stage renal disease following pregnancy is low,
and the rheumatic diseases, it may be a protective one. In with only a few cases reported in the literature. Women at
four cohort studies, breastfeeding for 13–24 months appeared greatest risk for these complications are those with active
to reduce the likelihood that the mother would develop RA SLE, particularly lupus nephritis, at the time of conception.
by approximately 50% (Brun 1995; Merlino et al. 2003; Some circumstances related to SLE do elevate pregnancy-
Karlson et al. 2004; Pikwer et al. 2008). Two cohort studies associated risks to a level above that acceptable to most
that utilized data from the Nurses’ Health Study found no women. For example, women with a history of stroke sec-
increase in the risk of incident SLE among mothers who ondary to the antiphospholipid syndrome are at high risk for
breastfed their infants (Costenbader et al. 2007; Simaard a recurrent stroke or other arterial thrombosis during preg-
et al. 2008). nancy. Women whose disease has recently required cyclo-
phosphamide or MMF for SLE control may not tolerate 9
Pearl: Ibuprofen is the NSAID of choice for women who are
months without these drugs. Women with a prior myocardial
breastfeeding.
infarction, significant pulmonary hypertension, or interstitial
Comment: NSAIDs, prednisone, antimalarial agents, and lung disease may not tolerate the added cardiopulmonary
sulfasalazine can be used safely by lactating mothers. strain of pregnancy poorly. These patients usually are served
Ibuprofen is the NSAID of choice for women who breastfeed best by physicians who discourage them from considering
because of its extremely low levels in breast milk and its pregnancy.
short half-life. In addition, the therapeutic dose for an infant
Pearl: Oral contraceptives are not inappropriate for all
is significantly higher than that found in breast milk.
women with SLE.
Peak levels of prednisone in breast milk are found about
2 h after the mother takes a dose, but levels decline rapidly Comment: For most SLE patients, the avoidance of oral con-
thereafter. Breastfeeding women may attempt to time their traceptives that contain estrogen appears prudent. However,
feedings accordingly. Several studies have confirmed the the dogma that oral contraceptives are contraindicated in all
absence of adverse effects in infants exposed to prednisone patients with SLE was challenged by the SELENA study,
through breast milk. which showed that oral contraceptives do not increase the
Drug levels of hydroxychloroquine also peak in the breast rate of SLE flares in women with mild SLE (Petri et al.
milk 2 h after a dose and decline over the next 9 h. In a study 2005). The SELENA study did not include women with
of 19 children treated with daily exposure to hydroxychloro- moderate to severely active SLE, so no statements about the
quine through breast milk for periods of up to 30 months, no appropriateness of oral contraceptive use in that setting are
retinal, motor, or growth abnormalities were observed during possible. For patients known to have risk factors for hyper-
a 12-month follow-up period. coagulability, the avoidance of estrogen-containing contra-
Sulfasalazine is compatible with breastfeeding in healthy, ceptives is prudent.
full-term infants, but caution is indicated in preterm infants Reliable alternatives to estrogen-based contraceptives are
or in babies with hyperbilirubinemia. Sulfasalazine metabo- available to women with SLE. Progestin-only contraceptives,
lites may contribute to neonatal jaundice. either in the form of an injection every 3 months or proges-
terone-only pills that are taken daily, do not increase SLE
activity or the risk for thrombosis. The depot form is particu-
larly effective in preventing pregnancy. An intrauterine
17.5 Systemic Lupus Erythematosus device, also an excellent long-term contraceptive option, is
not contraindicated in women with rheumatologic disease.
In the event that a woman has unprotected intercourse, the
Myth: Women with SLE should never get pregnant.
over-the-counter emergency contraceptive pill (“Plan B”)
Reality: The majority of women with SLE have safe pregnan- contains only progestin. Thus, it is safe for women with
cies and deliver healthy, full-term babies. Pregnancy involves SLE.
186 M. E. B. Clowse et al.

Table 17.1 Clinical features helpful in distinguishing a lupus flare from preeclampsia
Flare of systemic lupus erythematosus Preeclampsia/eclampsia
• Can develop at any time during the pregnancy • Never develops before the 20th week of gestation
• Rash • Pathologically brisk reflexes
• Arthritis • Clonus
• Serositis • Elevated serum levels of hepatic transaminases
• Increase in uric acid level less common than in preeclampsia/eclampsia • Increased serum uric acid level
• Fever • Decreased urine calcium concentration
• Active urine sediment

Pearl: In some patients, it is impossible to distinguish preec- with moderately active lupus and low complement had far
lampsia from lupus nephritis. worse pregnancy outcomes than women with similarly active
lupus but normal complement (Clowse et al. 2004).
Comment: Preeclampsia presents with hypertension (or a ris-
Although the trend of the complement levels is more
ing blood pressure when compared with the patient’s preg-
helpful than any single measurement, complement levels
nancy baseline) and new-onset proteinuria. Preeclampsia
seldom provide definitive direction for clinical decision-
never occurs before 20 weeks of gestation and usually not
making lupus pregnancies. Information about serial comple-
before the third trimester. Thus, the timing of proteinuria
ment measurements must be correlated with the clinical
onset helps distinguish preeclampsia from active SLE.
picture. There remains a large margin for good clinical
Unfortunately, in the third trimester, preeclampsia can
judgment.
mimic a lupus flare closely. Clinical features that help distin-
guish a lupus flare from preeclampsia are shown in Table 17.1. Pearl: Thrombocytopenia during pregnancy is not always
Several laboratory tests are helpful in identifying preeclamp- due to active lupus.
sia, including an elevated serum uric acid and a decreased Comment: Up to 8% of healthy women develop moderate
urine calcium concentration (Meads et al. 2008). In some situ- thrombocytopenia during pregnancy, with platelet counts in
ations, treatment for both conditions may be required: delivery the 80–120,000/mm3 range. This moderate drop in platelets
for the preeclampsia and immunosuppression for the lupus. is not pathologic and recovers following delivery. Thrombo-
Pearl: Lupus can complicate a pregnancy, even if the disease cytopenia may also be a sign of the HELLP syndrome
remains quiescent during the pregnancy itself. (Hemolysis, Elevated Liver tests, and Low Platelets), a severe
form of preeclampsia that is typically seen in the third
Comment: Quiet lupus promotes a healthy pregnancy. How- trimester.
ever, the lingering effects of previously active SLE (i.e.,
damage) can still impact a pregnancy’s outcome, even if no Myth: Pregnancy is contraindicated in patients with renal
flare occurs during gestation. Up to 30% of pregnancies in allografts.
women who’s SLE remains quiescent are delivered preterm. Reality: More than 14,000 pregnancies in women with renal
Damage to the kidney incurred before pregnancy promotes allografts have been documented over the past 50 years.
hypertension and preeclampsia. In addition, the elevated The majority of these pregnancies resulted in a live baby
thrombotic risk shared by many patients with SLE is and a functional renal transplant in the mother. An esti-
enhanced several times over by pregnancy. Many SLE experts mated 75% of such pregnancies result in live births, half of
prescribe low-dose aspirin throughout pregnancy in the hope which are preterm (Armenti et al. 2004). Fewer than 5% of
of obviating thrombotic complications (Duley et al. 2007). women experience an episode of allograft rejection during
pregnancy.
Myth: Normal complement levels during pregnancy make the
On the other hand, approximately 10% of patients lose
likelihood of a disease flare low.
kidney function within 2 years of delivery, a rate that is
Reality: Complement levels increase between 10 and 50% higher than expected (Armenti et al. 2004). Thus, pregnancy
during normal pregnancies. Thus, complement level mea- does pose some risk the health of renal allografts. It is gener-
surements during lupus pregnancies are unpredictable and ally recommended that women wait at least 1 year after
have poor prognostic value for either SLE flares or pregnancy undergoing renal transplantation before attempting to con-
outcomes. In the Johns Hopkins Lupus Cohort, 50% of ceive and that the patient’s blood pressure and allograft func-
women had a low level of complement at some point during tion be stable before conception (McKay et al. 2005). Vaginal
pregnancy. Those with low complement had a slightly higher deliveries are possible in half of patients with renal allografts,
risk of preterm birth. Complement levels were most helpful despite the pelvic location of the transplanted kidney
when evaluating clinically evident lupus activity: women (Armenti et al. 2004a).
17 Pregnancy in the Rheumatic Diseases 187

Myth: The great majority of women with lupus suffer disease of gestation (Izmirly et al. 2007). Organ involvement in this
flares during or after pregnancy. disorder includes cardiac, cutaneous, hepatic, hematologic,
and other manifestations. Organ involvement in neonatal
Reality: The estrogen-driven theory of SLE does not extend
lupus is associated with antibodies in both the maternal and
to many women in pregnancy. Estrogen levels in the third
fetal serum directed against the Ro or La antigens (or, in a
trimester increase to levels 100-fold higher than prepregnancy
minority of cases, against RNP).
levels, yet the majority of women remain relatively symptom-
Most of the organ manifestations of neonatal lupus are self-
free. More than 85% of women whose SLE was quiescent
limited and benign. However, congenital heart block can result
during the 6 months prior to conception have no more than
in significant morbidity and even fetal death. Approximately
mild SLE activity during pregnancy (Clowse et al. 2005).
50% of newborns with complete heart block require cardiac
The majority of SLE flares in pregnancy involve the skin
pacing early in life. Such babies require lifelong cardiac pac-
and joints. These flares can occur at any time during preg-
ing. One-year mortality among infants with congenital heart
nancy or in the months following delivery. Flares during
block ranges from 12 to 41%. Most deaths occur in utero or
pregnancy are best prevented by continuing SLE medica-
within the first 3 months of life. The risk of death continues
tions, particularly hydroxychloroquine, during the gesta-
through age 4, with mortality figures on the order of 3–4%
tional period (Clowse et al. 2006b).
after the first year of life (Gordon 2007).
In rare cases, death attributed to congenital heart block
has occurred in adulthood (Moak et al. 2001). The main
17.6 Antiphospholipid Syndrome (APS) cause of early and late death in those affected with congeni-
tal heart block is cardiac failure due to dilated cardiomyopa-
thy. Predictors of early mortality include a fetal heart rate
Pearl: Pregnancy losses occur despite low-molecular weight less than 55/min, delivery before 34 weeks of gestation, and
heparin and aspirin treatment in 25% of women with the hydrops fetalis.
APS.
Pearl: Many mothers who deliver babies with neonatal lupus
Comment: This is true. Although a 25% loss rate is far from have no established connective tissue disorder at the time of
optimal, it is better than the estimated 90% loss rate without their child’s birth.
any treatment and 40% loss rate with aspirin alone (Empson
Comment: Most mothers who give birth to a baby with the
et al. 2005). For a woman who suffers a pregnancy loss
neonatal lupus syndrome have a connective tissue disorder
despite treatment, a standard approach is to attempt preg-
but do not know it. Up to 66% of mothers who deliver a baby
nancy again with low molecular-weight heparin and aspirin.
with neonatal lupus have no recognized symptoms of a con-
For patients with repeat pregnancy losses, intravenous
nective tissue disorder at the birth of the affected child. SLE,
immune globulin is endorsed by some reproductive immu-
Sjögren’s syndrome, and undifferentiated connective tissue
nologists. However, evidence-based support for this approach
disorders can all be associated with the birth of babies with
is lacking.
neonatal lupus.
Pearl: Heparin and aspirin should be continued for 6 weeks In a study of 229 mothers of babies with neonatal lupus
after delivery in women with the APS. who were followed for a mean of 6 years, approximately half
of the mothers who were initialy asymptomatic developed
Comment: Mothers are at increased risk for thrombosis dur-
some symptoms of a rheumatic disorder (Gordon 2007;
ing and immediately after pregnancy. Low-molecular weight
Izmirly 2007). However, progression to a recognized con-
heparin and aspirin should be continued for 6 weeks follow-
nective tissue disorder occurred in less than 15% of patients.
ing delivery to prevent venous thrombotic events during this
The underlying diagnoses in 321 mothers of babies with the
period of relative immobility (Empson et al. 2005).
neonatal lupus syndrome are shown in Table 17.2.

17.7 Neonatal Lupus Syndromes Table 17.2 Diagnoses of connective tissue disease in 321 mothers of
babies with the neonatal lupus syndrome (Adapted from Izmirly et al.
2007)
Myth: Congenital heart block in neonatal lupus always leads 23% asymptomatic (no connective tissue disorder diagnosis)
to fetal demise. 24% undifferentiated connective tissue disorder
15% systemic lupus erythematosus
Reality: Neonatal lupus is an example of acquired autoim- 22% had Sjögren’s syndrome
munity. This syndrome is caused by the passage across the 17% systemic lupus erythematosus with secondary Sjögren’s
placenta of maternal autoantibodies following the 16th week syndrome
188 M. E. B. Clowse et al.

Myth: Babies with the neonatal lupus syndrome are at high 17.9 Scleroderma
risk of developing rheumatic disorders as adults.

Reality: Limited data are available on the long-term out- 1. Women with scleroderma should never get pregnant.
comes of infants with the neonatal lupus syndrome. One
In the older literature, patients with scleroderma were thought
study investigated 55 mothers of babies with neonatal lupus,
to be at high risk for poor fetal and maternal outcome.
including 49 affected babies and 45 unaffected siblings
However, in several more recent retrospective case-control
(Martin et al. 2002). Six children (12%) who had had neona-
studies, fewer adverse pregnancy outcomes were observed.
tal lupus as babies developed definite autoimmune condi-
Careful planning, close monitoring and appropriate therapy
tions between the ages of 2 and 13. Their diagnoses included
allows many of these patients to have a successful pregnancy.
juvenile rheumatoid arthritis (n = 2), Hashimoto’s thyroiditis
Retrospective case-control studies clearly show an increased
(1), psoriasis and iritis (1), diabetes mellitus and psoriasis
frequency of pre-term births and small full-term infants but
(1), and congenital hypothyroidism and nephrotic syndrome
the frequency of miscarriage and neonatal survival rate did not
(1). None of the unaffected siblings developed rheumato-
differ from healthy controls. (Steen, Conte et al. 1989; Miniati,
logic disease. These data suggest that most children’s risk of
Guiducci et al. 2008)
autoimmune disease expression disappears along with the
Women with scleroderma usually do not have a worsening
maternally derived autoantibodies at approximately 6 months
of their disease during pregnancy provided that the disease is
of life. During adolescence and young adulthood, neither
stable at conception. However, new onset scleroderma has
individuals with neonatal lupus nor their unaffected siblings
been reported during pregnancy and postpartum. The later
appear to have increased risks of developing autoimmune
stages of pregnancy are associated with an increase in cardiac
conditions (Martin et al. 2002).
output, and, as a consequence symptoms related to Raynauds
phenomenon improve or even abate completely. Skin changes
may improve during pregnancy and may then worsen post-
partum. Gastroesophageal reflux and dyspnea are common in
17.8 Rheumatoid Arthritis both healthy pregnant women and those women with sclero-
derma. The development of hypertension during pregnancy
in a woman with scleroderma must be evaluated for pre-
Pearl: Women with rheumatoid arthritis generally improve eclampsia, hemolytic anemia, elevated liver enzymes and
during pregnancy but flare in the postpartum phase. low platelets (HELLP) syndrome, and scleroderma renal cri-
Comment: Up to 75% of women with RA have significant sis. Daily increases in serum creatinine and the absence of
improvement in their arthritis during pregnancy. Some proteinuria characterize scleroderma renal crisis from pre-
women enjoy complete remissions during this time. The eclampsia or HELLP. Recurrent renal crisis is more likely to
improvement is more modest in others, but still sufficient to occur within 5 years of the initial event, therefore women are
permit a decrease in RA medications during pregnancy. advised to wait 3-5 years after an episode of renal crisis
Approximately 20% of women with RA actually experi- before considering a pregnancy. (Steen 1999)
ence a worsening of their inflammatory arthritis during preg- Since ACE inhibitors are life saving in scleroderma renal
nancy (Barrett and Brennan 1999). Moreover, within 1 month crisis, they may need to be continued during pregnancy
of delivery, 66% of women note exacerbations of their RA. despite their teratogenic potential in the first trimester and
Within 6 months, 77% of patients have flared. renal effects in the third trimester if nifedipine or other cal-
cium channel blockers fail to control blood pressure. (Barr
Myth: If a pregnant woman has swollen joints, it means that 1994; Cooper, Hernandez-Diaz et al. 2006) Appropriate
her rheumatoid arthritis is getting worse. counseling is necessary to discuss the disease course and
medications in women with scleroderma.
Reality: This is a pretty good rule if the joint swelling is
2. Pulmonary hypertension is an absolute contraindication
accompanied by pain. However, pregnant women often
to pregnancy.
develop swollen joints independent of rheumatoid arthritis.
This is sometimes secondary to the generalized swelling One of the most lethal complications of scleroderma is pulmo-
caused by changes in sodium excretion from the kidney in nary hypertension. The prevalence of pulmonary hyperten-
the third trimester. In these cases, the swelling includes the sion in scleroderma is estimated to occur in 10-20% of
soft tissues of the hands and feet as well as the joints. Some individuals with this disorder. (Note to the editor: could refer
women also develop bland knee effusions during to the scleroderma section for more details on this topic).
pregnancy. Women with scleroderma or primary pulmonary hypertension
17 Pregnancy in the Rheumatic Diseases 189

(without systemic sclerosis) have significantly increased rates Costa M, Colia D. Treating infertility in autoimmune patients.
of maternal complications including hypertensive disorders, Rheumatology (Oxford) 2008; 47(Suppl 3):38–41
Costedoat-Chalumeau N, Amoura Z, Duhaut P, et al Safety of dyroxy-
pre-eclampsia and longer hospital stays. (Chakravarty, Khanna chlroroquine in pregnant patients with connective tissue diseases: A
et al. 2008) Even though there is a case report of a successful study of one hundred thirty-three cases compared with a control
pregnancy in a woman with both systemic sclerosis and pul- group. Arthritis Rheum 2003; 48:3207–3211
monary hypertension, pregnancy in the setting of systemic Costenbader KH, Feskanich D, Stampfer MJ, Karlson EW. Reproductive
and menopausal factors and risk of systemic lupus erythematosus in
sclerosis and pulmonary hypertension is not recommended. women. Arthritis Rheum 2007; 56:1251–1262
(Baethge and Wolf 1989) Furthermore, we would agree with Duley L, Henderson-Smart DJ, et al Antiplatelet agents for preventing
the recommendation to perform an evaluate for pulmonary pre-eclampsia and its complications. Cochrane Database Syst Rev
hypertension as part of the pre-conception counseling workup 2007; (2):CD004659
Empson M, Lassere M, et al Prevention of recurrent miscarriage for
in women with scleroderma, and if it is detected, would rec- women with antiphospholipid antibody or lupus anticoagulant.
ommend not conceiving due to the high mortality rate associ- Cochrane Database Syst Rev 2005; (2):CD002859
ated with this disease manifestation. (Gayed and Gordon FDA, 2007.www.fda.gov/cder/drug/InfoSheets/HCP/mycophelolate-
2007) HCP.htm
Gayed, M. and C. Gordon (2007). "Pregnancy and rheumatic diseases."
(Cooper, Hernandez-Diaz et al. 2006; Schwarz and Manzi Rheumatology (Oxford) 46(11): 1634–40
Gordon PA. Congenital heart block: Clinical features and therapeutic
2008)
options. Lupus 2007; 16:642–646
Izmirly PM, Rivera TL, Buyon JP. Neonatal lupus syndromes. Rheum
Dis Clin N Amer 2007; 33:267–285
Karlson EW, Mandl LA, Hankinson SE, Grodstein F. Do breast-feeding
References and other reproductive factors influence future risk of rheumatoid
arthritis. Arthritis Rheum 2004; 50:3458–3467
Armenti VT, Radomski JS, et al Report from the National Transpl- Katz PP. Childbearing decisions and family size among women with
antation Pregnancy Registry (NTPR): outcomes of pregnancy after rheumatoid arthritis. Arthritis Rheum 2006; 55:217–223
transplantation. Clin Transpl 2004;103–114 Kikuchi K, Shibahara H, et al Antinuclear antibody reduces the preg-
Baethge, B. A. and R. E. Wolf (1989). "Successful pregnancy with scle- nancy rate in the first IVF-ET treatment cycle but not the cumulative
roderma renal disease and pulmonary hypertension in a patient pregnancy rate without specific medication. Am J Reprod Immunol
using angiotensin converting enzyme inhibitors." Ann Rheum Dis 2003; 50:363–367
48(9): 776–8 Levy M, Buskila D, Gladman DD, Urowitz MB, Koren G. Prengnancy
Barr, M., Jr. (1994). "Teratogen update: angiotensin-converting enzyme outcome following first trimester exposure to chloroquine. Am
inhibitors." Teratology 50(6): 399–409 J Perinatol 1991; 8:174–178
Barrett JH, Brennan P, et al Does rheumatoid arthritis remit during Lin P, Bonaminio P, Ramsey-Goldman R. Pregnancy & rheumatic diseases.
pregnancy and relapse postpartum? Results from a nationwide study In: Imboden J, Hellmann D, Stone J (eds) Current diagnosis & treatment.
in the United Kingdom performed prospectively from late preg- Rheumatology (2nd edn). New York, McGraw-Hill, 2007, pp. 146–159
nancy. Arthritis Rheum 1999; 42:1219–1227 Lockshin MD. Autoimmunity, infertility and assisted reproductive
Brun JG, Nilssen S, Kvale G. Breast feeding, other reproductive factors technologies. Lupus 2004; 13(9):669–672
and rheumatoid arthritis: A prospective study. Br J Rheumatol 1995; Martin V, Lee LA, Askanase AD, Katholi M, Buyon JP. Long-term fol-
34:542–546 lowup of children with neonatal lupus and their unaffected siblings.
Chakravarty EF, Sanchez-Yamamoto D, Bush TM. The use of disease Arthritis Rheum 2002; 46:2377–2383
modifying antirheumatic drugs in women with rheumatoid arthritis McKay DB, Josephson MA, et al Reproduction and transplantation:
of childbearing age: A survey of practice patterns and pregnancy Report on the AST Consensus Conference on reproductive
outcomes. J Rheumatol 2003; 30:241–246 issues and transplantation. Am J Transplant 2005; 5:1592–1599
Chakravarty, E. F., D. Khanna, et al (2008). "Pregnancy outcomes in Meads CA, Cnossen, et al Methods of prediction and prevention of pre-
systemic sclerosis, primary pulmonary hypertension, and sickle cell eclampsia: Systematic reviews of accuracy and effectiveness literature with
disease." Obstet Gynecol 111(4): 927–34 economic modelling. Health Technol Assess 2008; 12(6):iii–iv, 1–270
Cooper, W. O., S. Hernandez-Diaz, et al (2006). "Major congenital mal- Merlino LA, Cerhan JR, Criswell LA, Mikuls TR, Saag KG. Estrogen
formations after first-trimester exposure to ACE inhibitors." N Engl and other female reproductive risk factors are not strongly associ-
J Med 354(23): 2443–51 ated with the development of rheumatoid arthritis in elderly women.
Clowse M. Lupus activity in pregnancy. Rheum Dis Clin N Amer 2007; Semin Arthritis Rheum 2003; 33:72–82
33:237–252 Milunsky A, Graef JW, Gaynor MF. Methotrexate-induced congenital
Clowse MEB, Witter F, et al Complement and anti-double stranded malformations. Med J Aust 1968; 2:1076–1077
DNA antibodies predict pregnancy outcomes in lupus patients. Miniati, I., S. Guiducci, et al (2008). "Pregnancy in systemic sclerosis."
Arthritis Rheum 2004; 50(suppl):S689 Rheumatology (Oxford) 47 Suppl 3: iii16–8
Clowse ME, Magder LS, et al The impact of increased lupus activity on Moak JP, Barron KS, Hougen TJ, et al Congenital heart block:
obstetric outcomes. Arthritis Rheum 2005; 52:514–521 Development of late-onset cardiomyopathy, a previously underap-
Clowse ME, Magder L, et al Hydroxychloroquine in lupus pregnancy. preciated sequela. J Am Coll Cardiol 2001; 37:238–242
Arthritis Rheum 2006a; 54:3640–3647 Neville CE, McNally J. Maternal exposure to leflunomide associated
Clowse ME, Magder LS, et al Early risk factors for pregnancy loss in with blindness and cerebral palsy (letter). Rheumatol 2007; 46:1506
lupus. Obstet Gynecol 2006b; 107(2 Pt 1):293–299 Østensen M, Khmashta M, Lockshin M, et al. Anti-inflammatory and
Clowse ME, Jamison M, et al A national study of the complications of immunosuppressive drugs in reproduction. Arthritis Res Ther 2006;
lupus in pregnancy. Am J Obstet Gynecol 2008; 199:127.e1–e6 8:1–19
190 M. E. B. Clowse et al.

Østensen M, Lockshin M, Doria A, et al Update on safety during preg- Schwarz, E. B. and S. Manzi (2008). "Risk of unintended pregnancy
nancy of biological agents and some immunosuppressive anti- among women with systemic lupus erythematosus." Arthritis
rheumatic drugs. Rheumatol 2008; 47:iii28–iii31 Rheum 59(6): 863–6
Parke AL, West B. Hydroxychloroquine in pregnant patients with sys- Simaard JF, Karlson EW, Costenbader KH, Hernán MA, Stampfer
temic lupus erythematosus. J Rheumatol 1996; 23:1715–1718 MJ, Liang MH, Mittlemen MA. Perinatal factors and adult-onset
Paufique L, Magnard P. Retinal degeneration in 2 children following lupus. Arthritis Rheum (Arthritis Care Res) 2008; 59:
preventive anitmalarail treatment of the mother during pregnancy. 1155–1161
Bull Soc Ophtalmol Fr 1969; 69:466–467 Soares PMF, Borba EF, Bonfa E, Hallak J, Corrêa AL, Silva CAA.
Petri M, Allbritton J. Fetal outcome of lupus pregnancy: A retrospective Gondal evalution in male systemic lupus erythematosus. Arthritis
case-control study of the Hopkins Lupus Cohort. J Rheumatol 1993; Rheum 2007; 56:2352–2361
20:650–656 Somers EC, Marder W, Christman GM, et al Use of a gonadotro-
Petri M, Kim MY, et al Combined oral contraceptives in women with pin-releasing hormone analog for protection against premature
systemic lupus erythematosus. N Engl J Med 2005; 353(24): ovarian failure failure during cyclophosphamide therapy in
2550–2558 women with severe lupus. Arthritis Rheum 2005; 52(9):
Pikwer M, Bergströom U, Nilsson J-Å, Jacobsson L, Berglund G, 2761–2767
Turesson C. Breast feeding, but not use of oral contraceptives, is Steen, V. D. (1999). "Pregnancy in women with systemic sclerosis."
associated with a reduced risk of rheumatoid arthritis. Ann Rheum Obstet Gynecol 94(1): 15–20
Dis online May 14, 2008 as 10.1136/aard.2007.084707 Steen, V. D., C. Conte, et al (1989). "Pregnancy in women with sys-
Ramsey-Goldman R, Schilling E. Optimum use of disease-modifying temic sclerosis." Arthritis Rheum 32(2): 151–7
and immunosuppressive antirheumatic agents during pregnancy and Taguchi N, Rubin ET, Hosokawa A, et al Prenatal exposure to HMG-
lactation. Clin Immunother 1996; 5:40–58 CoA reductase inhibitors: Effects on fetal and neonatal outcomes.
Ramsey-Goldman R, Mientus JM, Kutzer JE, Mulvihill JJ, Medsger TA Reproductive Toxicology 2008 [Epub]
Jr. Pregnancy outcome in women with systemic lupus erythemato- Wilson JG, Scott WJ, Ritter EJ, Fradkin R. Comparative distribution
sus (SLE) treated with immunosuppressive drugs (ISD). J Rheumatol and embryo toxicity of methotrexate in pregnant rats and rhesus
1993; 20:1152–1157 monkeys. Teratology 1979; 19:71–98
Raptopoulou A, Sidiropoulos P, Boumpas D. Ovarian failure and strategies
for fertility preservation in patients with systemic lupus erythematosus.
Lupus 2004; 13(12):887–890
 Inflammatory Myopathies
18
Ira N. Targoff, Chester V. Oddis, Paul H. Plotz,
Frederick W. Miller, and Mark Gourley

18.2 Dermatomyositis Versus Polymyositis

18.1 Overview of the Inflammatory
Myopathies
Myth: Dermatomyositis is polymyositis with a rash.

› The idiopathic inflammatory myopathies (IIM) are a Reality: Although dermatomyositis (DM) and polymyositis
heterogeneous group of disorders characterized by (PM) share the term “myositis,” these disease entities should
varied patterns of inflammation within striated not be considered the same for many reasons. Pathologic dif-
muscle. ferences suggest that DM is a humorally mediated vascul-
› The major disease categories among the IIM are der- opathy. In contrast, PM results from T cell-directed injury to
matomyositis (DM), polymyositis (PM), inclusion the myocyte. Histologic examination of inflamed muscle in
body myositis (IBM), myositis associated with con- DM reveals more B cells, complement deposition, and CD4-
nective tissue diseases, and myositis associated with positive T cells in perivascular regions. In contrast, PM is
malignancy. characterized by a predominantly CD8+ T cell infiltrate, with
› The skin, joints, lungs, heart, and gastrointestinal fewer B cells. Although the same autoantibodies can be
tract are also involved in different forms of these found in DM and PM, some are associated with specific IIM
disorders. subsets such as antisignal recognition particle (SRP) in PM
› Muscle weakness that is proximal, symmetrical, and and anti-Mi-2 in DM. Other differences include a substan-
painless is a hallmark feature of the inflammatory tially greater interferon-a/b-induced gene expression in DM
myopathies. Patients with IBM are also prone to distal, when compared to PM.
asymmetric muscle involvement. Response to therapy also differentiates the myopathies.
› Measurement of serum concentrations of muscle DM typically responds readily to glucocorticoids, but PM
enzymes, skin and muscle biopsy, electromyography, responds more slowly and often requires the addition of a
and magnetic resonance imaging can assist in the second-line immunosuppressive agent.
diagnosis.
› Testing for serum autoantibodies is helpful both in
diagnosis and in predicting the clinical phenotype and
response to therapy. 18.3 Skin Disease
› A minority of patients with DM and PM have myositis
that is associated with an underlying malignancy. The Pearl: Biopsy of any of the skin lesions in DM shows the
risk is greatest for middle-aged to elderly patients with same histology.
DM. Approximately 15% of DM patients have malig-
nancy-associated disease. Comment: This is almost entirely correct! With the excep-
› Most malignancies present within 1 year before or tion of the uncommon subcutaneous cystic lesions, biopsy
after the diagnosis of inflammatory myopathy. of any skin lesion in DM will show focal epidermal atro-
phy, liquefaction, and degeneration of the basal cell layer

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 191

DOI:10.1007/978-1-84800-934-9_18, © Springer Science + Business Media B.V. 2009
192 I. N. Targoff et al.

a c

Fig. 18.1 Cutaneous findings of dermatomyositis. Skin biopsy of any of these lesions shows the same histopathologic features (see Fig. 18.2)
(a) Gottron’s papules (b) Heliotrope erythema (c) Shawl sign (d) Mechanic’s hands (Figures Courtesy of Dr. John Stone)

and a perivascular or upper dermal mononuclear or lympho-

cytic infiltrate. Immunofluorescence studies confirm the
presence of an interface dermatitis, with positive staining for
immunoglobulin or complement at the dermal–epidermal
junction. Biopsies of Gottron’s papules (Fig. 18.1a), helio-
trope erythema (Fig. 18.1b), the shawl sign (Fig. 18.1c), and
mechanic’s hands (Fig. 18.1d) all reveal the same histo-
pathological features. The dermatopathology of these lesions
is depicted in Fig. 18.2.

Myth: Patients must have muscle disease to make a diagnosis

of DM.

Reality: The original Bohan and Peter criteria required that

patients have muscle disease in order to receive a diagnosis
of DM (Bohan and Peter 1975). However, the pathogno-
monic skin findings in patients with DM permit the diagnosis Fig. 18.2 Histopathology of the skin in dermatomyositis. A hematoxy-
of “amyopathic” DM in the absence of muscle findings. The lin and eosin stain of a skin biopsy (× 100 magnification) reveals an
atrophic epidermis, extensive dermal mucin deposition (pink-colored
hallmark skin findings of DM include Gottron’s papules and material), and only a mild dermal inflammatory infiltrate are seen
erythema over the extensor surfaces of joints, particularly the (Figure courtesy of Dr. Mai Hoang, Massachusetts General Hospital)
18 Inflammatory Myopathies 193

Patients with amyopathic DM are also at an increased risk

for cancer and the frequency of malignancy is likely similar
in both classic DM and amyopathic DM (Gerami et al. 2006);
thus, patients should be screened accordingly (Callen 2001).
The cancers that are particularly increased in DM include
lung, ovary, pancreas, stomach, colon, and rectum, as well as
non-Hodgkin’s lymphoma (Hill et al. 2001).

18.4 Muscle Enzymes

Myth: All patients with active myositis have an elevated cre-

Fig. 18.3 Periungual erythema in a patient with dermatomyositis atine kinase concentration.
(Figure courtesy of Dr. John Stone)
Reality: Serum creatine kinase (CK) activity levels do not
correlate perfectly with muscle inflammation or weakness in
metacarpophalangeal and interphalangeal joints, the elbow, myositis as muscle inflammation and weakness can be pres-
the malleoli, and the knee along with a heliotrope rash. ent without CK elevation. This is most often seen in DM, in
Other characteristic but not pathognomonic findings include both the adult and juvenile subsets (see below). The mecha-
periungual erythema (Fig. 18.3) or capillary changes, dys- nism by which this occurs is not understood completely.
trophic cuticles, and a violaceous erythema found in a Some patients may have an inhibitor of enzyme function that
photodistribution. leads to an underestimation of the CK activity (Kagen and
Patients with amyopathic DM have undiminished muscle Aram 1987). The serum CK activity level may correlate bet-
strength, no muscle pain, and normal muscle enzymes ter with the degree of muscle inflammation early in the
(Gerami et al. 2006). Approximately one-third of DM patients course of the disease as opposed to later when muscle atro-
who present to dermatologists have amyopathic DM. Another phy is more likely.
third has minimal muscle symptoms, with only subclinical In patients with a normal CK, one or more of the other
evidence of myositis on laboratory and electrophysiologic muscle enzymes (e.g., aldolase, lactic dehydrogenase (LDH),
evaluation and radiographic imaging (Quain et al. 2007). aspartate aminotransferase (AST), alanine aminotransferase
(ALT) ), or another muscle factor (e.g., myoglobin) may be
Myth: Patients with amyopathic DM do not have the same
elevated. However, as discussed below, serum enzyme con-
risk of lung disease as those with DM who have muscle
centrations should not exclusively guide treatment.
inflammation as well as skin disease.
Even if the serum CK activity level remains within the
Reality: Patients with amyopathic DM can have interstitial range of normal, changes in activity levels within this range
lung disease in the absence of muscle inflammation. In fact, may be significant as elevations of the CK above the patient’s
one study reported that 25% of patients with amyopathic DM baseline activity level may be important. Magnetic resonance
had interstitial lung disease (Quain et al. 2007). As a result, imaging of the muscle is helpful in the assessment of disease
patients with amyopathic DM must be screened with pulmo- activity when muscle weakness is present in the absence of
nary function tests (PFTs) that include measurement of dif- muscle enzyme elevations.
fusion capacity. Further evaluation of the lungs is required if
Pearl: Muscle inflammation in the setting of a normal serum
this screen detects any abnormalities. Annual PFTs are
CK activity level is more likely in DM than in PM.
appropriate assessments in patients with DM and PM.
Some patients with interstitial lung disease have erythema Comment: Active myositis with a normal CK is more com-
and/or roughness of the palmar surface and sides of the fin- mon in DM than in PM, but must be distinguished from
gers and palm, otherwise known as “mechanic’s hands.” This amyopathic DM (in which there is no clinical or laboratory
skin finding is associated with the antisynthetase syndrome evidence of muscle involvement). Even if cases of amyo-
(Quain et al. 2007). pathic DM are excluded, serum CK activity levels correlate
less well with the severity of muscle involvement in DM than
Pearl: Patients with amyopathic DM should be screened for
they do in PM.
underlying malignancies.
PM is unlikely in the setting of a normal CK and is more
Comment: Patients with DM have a higher risk of malig- difficult to diagnose than DM when the CK is not elevated.
nancy than do patients with PM (Buchbinder et al. 2001). In fact, “definite” PM cannot be diagnosed according to the
194 I. N. Targoff et al.

Bohan and Peter criteria without an elevation of serum mus- Myth: An elevated CK-MB activity level is a sign of cardiac
cle enzymes (Bohan and Peter 1975). Some reports indicate damage in patients with myositis.
that normal CK activity levels in the setting of active
Reality: The CK-MB activity level and its proportional
myositis are likely to occur in patients with malignancy-
activity level when compared to that of the total CK can
associated myositis or inflammatory myopathy associated
increase in myositis even in the absence of detectable car-
with a connective tissue disease, but reports of series of
diac muscle inflammation or injury. This is usually attrib-
such patients are too small to permit definitive conclusions
uted to an increased proportion of CK-MB production by
on this point.
myoblasts in regenerating muscle. Cardiac muscle inflam-
Myth: The most important parameter in assessing disease mation does occur in myositis and is sometimes clinically
activity in myositis patients is the concentration of muscle significant, but the more frequent occurrence of CK-MB
enzymes. elevation from noncardiac sources in patients with myositis
makes it difficult to use CK-MB as a sign of cardiac
Reality: The main clinical manifestation of myositis is mus-
damage.
cle weakness that translates into functional impairment and
Serum troponin-T levels have also been proposed as spe-
disability. Since the principal objective of treatment is the
cific markers of myocardial disease, but they too have been
recovery and maintenance of normal strength through the
associated with myositis disease activity unrelated to cardiac
suppression of muscle inflammation, the most important
involvement (unpublished observations and Erlacher et al.
parameter to monitor is muscle strength, not serum enzyme
2001). On the other hand, serum concentrations of cardiac
concentrations.
troponin-I are usually normal in patients with PM or DM
This concept was incorporated into the consensus defini-
who do not have cardiac involvement (Erlacher et al. 2001).
tion of improvement for myositis established by a panel of
Consequently, elevations of cardiac troponin-I support an
experts (Rider et al. 2004). Muscle strength is most often
inflammatory process in the myocardium.
assessed by manual muscle testing, but function can be
ascertained by other approaches, including patient question- Myth: Aldolase is a muscle-specific enzyme.
naires. The serum CK and other muscle enzymes are indirect
Reality: Elevations of the serum CK are usually derived
measures of muscle inflammation and can be misleading so
from muscle, particularly when the isotype is CK-MM.
the notion that myositis treatment should focus on CK nor-
Skeletal muscle injury also results in the elevation of the
malization is a Myth.
serum aldolase activity, because skeletal muscle is rich in
However, despite the shortcomings of serum muscle
aldolase A. The CK activity level is generally a more sensi-
enzyme measurements as a biomarker of disease activity,
tive and specific marker of muscle injury than is aldolase.
these determinations still play a role in the assessment of
The aldolase may be elevated in hemolytic states, with liver
most myositis patients. Indeed, CK activity levels correlate
damage and in CNS disorders as aldolase B is found in the
well with other measures of disease activity in many patients
liver and other tissues. However, in some myositis patients,
and an increase in the serum CK in an otherwise stable
the serum CK may not correlate with disease activity and
patient serves as a potential warning of a disease exacerba-
the serum aldolase is more useful in such unusual and rarely
tion, assuming there is no intercurrent process such as mus-
reported cases.
cle injury, hypothyroidism, or medication use to explain the
abnormality.
Myth: Serum concentrations of ALT do not rise during peri-
ods of active inflammatory muscle disease because this
enzyme is specific to the liver, not skeletal muscle. 18.5 Myositis-Specific Autoantibodies
Reality: Elevations in the serum ALT concentration can
Myth: The “myositis-specific autoantibodies” are specific
occur in muscle injury as well as with liver damage. Although
for myositis, as patients with one of these antibodies always
there is more aspartate aminotransferase (AST) than ALT in
have myositis.
muscle, serum ALT can rise when injury to skeletal muscle
is significant. Reality: The term “myositis-specific autoantibodies” (MSA)
The fact that ALT elevation can occur in muscle disease was first used to identify a group of autoantibodies that is
complicates the clinician’s ability to detect medication- associated primarily with myositis. MSAs have been distin-
induced liver toxicity. The failure of the AST and ALT con- guished from “myositis-associated autoantibodies” (MAAs),
centrations to fall in conjunction with the CK activity level as which may occur in myositis as well as other connective tis-
muscle strength improves may unmask the presence of drug- sue diseases (CTD). The MSAs originally included the
induced hepatotoxicity. following:
18 Inflammatory Myopathies 195

• Antisynthetase antibodies (anti-Jo1 antibodies being the if clinical features such as arthritis, Raynaud’s phenomenon,
most common, but now also including at least seven other mechanic’s hands, or fevers are present.
antibodies directed against aminoacyl-tRNA synthetases)
• Anti-SRP antibodies Pearl: Patients with antisynthetase antibodies often present
• Anti-Mi-2 antibodies with features other than myositis but may develop myositis
later.
These antibodies, when assessed by immunoprecipitation
Comment: Patients with antisynthetase antibodies do not
methods, generally have a specificity for myositis on the
always develop myositis, but in those who do, myositis may
order of 90% or more, but previous studies have reported
neither be the presenting or the symptom that brings the
antisynthetase antibody-positive patients with ILD who do
patient to medical attention (Schmidt et al. 2000). Such
not have any clinical evidence of muscle inflammation
patients may present with interstitial lung disease, a syn-
(Friedman et al. 1996; Tillie-Leblond et al. 2008), or late
drome that resembles rheumatoid arthritis, fever or even the
onset myositis in the course of other CTDs (Tillie-Leblond et
rash of dermatomyositis.
al. 2008; Schmidt et al. 2000). Anti-SRP autoantibodies may
also not be as specific for myositis as originally reported Pearl: Certain antisynthetase antibodies are associated with
(Kao et al. 2004), and although anti-Mi-2 autoantibodies are a higher frequency of myositis than are others.
relatively specific for myositis by immunoprecipitation
Comment: Although anti-Jo1 antibodies are detected in some
assays, antibodies to Mi-2 peptides may be seen in other con-
patients with interstitial lung disease who do not have myositis
ditions when measured by ELISA or immunoblotting
(see above), this clinical picture is more common in patients
(Hengstman et al. 2005).
with the non-Jo-1 antisynthetase autoantibodies. The fre-
The greatest utility of the MSAs occurs when the diagno-
quency of myositis is lower in patients who are anti-PL-12
sis is unclear, such as in the setting of possible myositis or
(anti-alanyl-tRNA synthetase), anti-OJ (anti-isoleucyl-tRNA
another immunologic disorder where autoimmunity is pre-
synthetase), or anti-KS (anti-asparaginyl-tRNA synthetase)
sumed, such as in patients with ILD of unclear etiology (see
antibody positive (Targoff and Arnett 1990; Targoff et al.
below).
1993; Sato et al. 2007; Hirakata et al. 2007). The frequency of
Pearl: Not all of the autoantibodies relevant to the IIMs have myositis associated with particular antisynthetases may vary
been identified. with the ethnic population under study (Hirakata et al. 2007).
Comment: The list grows longer every year. In addition to Myth: Patients who have MSAs should be antinuclear anti-
the antisynthetase, anti-SRP, and anti-Mi2 autoantibodies, body (ANA) positive if an indirect immunofluorescent assay
several additional autoantibodies appear to have a primary is used.
association with PM or DM. These include the anti-hPMS-1
Reality: Antisynthetases are directed at aminoacyl-tRNA
antibody, the anti-p155/140 antibody, an antibody against
synthetases, which are cytoplasmic in location. Thus, these
small ubiquitin-like modifier activating enzyme (anti-SAE
autoantibodies usually cause cytoplasmic (not nuclear) pat-
antibody), the anti-caDM140 antibody, and others. The spec-
terns of fluorescence on indirect immunofluorescent ANA
ificities of these autoantibodies have not been studied as
testing. Cytoplasmic staining may or may not be reported by
extensively as those of traditional MSAs, but they appear to
commercial laboratories; thus, patients with myositis and
have diagnostic utility. These tests are not yet commercially
antisynthetase antibodies are often ANA negative. In patients
available.
with antisynthetase antibodies who are ANA positive, co-
Myth: Patients with antisynthetase antibodies always have existent autoantibodies may be present in the serum. Patients
myositis. with anti-SRP autoantibodies also usually demonstrate cyto-
plasmic staining upon immunofluorescence testing for simi-
Reality: The antisynthetase antibodies are more specific for
lar reasons as patients with an antisynthetases antibody.
an underlying connective tissue disease or an autoimmune
Conversely, patients with anti-Mi-2 and anti-PM-Scl
syndrome rather than for myositis itself. For each of the anti-
autoantibodies should have a positive ANA test. The anti-
synthetase antibodies, patients have been reported who have
Mi-2 antibody yields a nuclear speckled pattern, and anti-PM-
the antibody but do not have myositis. However, these
Scl antibodies cause a combined nuclear and nucleolar pattern.
patients have had other features of the antisynthetase syn-
Anti-p155/140 antibodies usually cause a nuclear speckled
drome such as interstitial lung disease or an inflammatory
pattern, but is less consistent and usually lower in titer.
arthropathy. Antisynthetase antibody testing should be con-
sidered in patients whose interstitial lung disease may have Pearl: The finding of diffuse cytoplasmic immunofluorescence
an autoimmune basis, even in the absence of evidence of on ANA testing can be a clue to the presence of antisynthetase
myositis (Tillie-Leblond et al. 2008). This is particularly true or anti-SRP autoantibodies.
196 I. N. Targoff et al.

Comment: In a patient who has clinical manifestations of an Not all roughened hands are mechanic’s hands. This find-
inflammatory myopathy, the finding of a cytoplasmic pattern ing must be distinguished from psoriasis, dyshydrotic
by indirect immunofluorescence is an indication for obtain- eczema, and simple calluses.
ing a more thorough analysis for potential MSAs and MAAs.
Myth: All patients with anti-SRP antibodies have a severe
Antiribosomal P antibodies can also give this pattern.
myopathy and a disease course characterized by treatment
Pearl: Many patients with myositis do not have a myositis- refractoriness.
specific autoantibody (MSA). Reality: Most studies note that anti-SRP patients have more
Comment: Only about 30–40% of patients with DM or PM severe weakness than myositis patients without this antibody
have antisynthetase, anti-SRP, or anti-Mi2 autoantibodies. (Kao et al. 2004; Miller et al. 2002) and some have a rapidly
By testing for more recently described antibody specificities progressive and relatively treatment-resistant course (Love
(see Table 18.1), the frequency of autoantibody identification et al. 1991). However, some anti-SRP antibody positive
in these patients increases to greater than 50%. Although a patients are more treatment-responsive and do well with glu-
percentage of myositis patients have MAAs, some patients cocorticoids and other immunosuppressive agents (Hengstman
with myositis have no currently identifiable autoantibodies et al. 2006; Kao et al. 2004).
(MSA or MAA) by either commercial or investigational Pearl: Patients with anti-SRP autoantibody frequently have
testing. less lymphocytic inflammation and relatively greater myofi-
One study that employed multiple assay techniques in a ber necrosis on their muscle biopsy.
group of 100 patients detected MSAs or MAAs in 80% of the
cohort (Koenig et al. 2007). Although numerous antibodies Comment: The muscle biopsy in patients whose myositis is
and combinations of antibodies were found, the study did not associated with antibodies to SRP often demonstrates a
include detection of recently described antibodies known to necrotizing myopathy with little or no inflammation (Dimitri
be associated with myositis so the 80% figure may be some- et al. 2007; Hengstman et al. 2006; Kao et al. 2004; Miller et
what low. al. 2002). The lack of the classic inflammatory changes of
myositis may lead to difficulty with diagnosis. The presence
Myth: Mechanic’s hands are specific for the antisynthetase of a necrotizing myopathy and severe muscle weakness
syndrome. should suggest the diagnosis of SRP-associated PM, particu-
larly in the presence of a cytoplasmic pattern on ANA testing
Reality: Mechanic’s hands, the cracking and/or fissuring of
(see above). Patients with anti-SRP antibodies rarely have
the lateral and palmar aspects of the fingers, are associated
the cutaneous findings of DM.
with the antisynthetase syndrome (see Fig. 18.1d). However,
mechanic’s hands can occur with other autoantibodies, e.g., Pearl: Patients with antisynthetase or anti-PM-Scl autoanti-
anti-PM-Scl antibodies (Oddis et al. 1992). bodies can have a significant arthropathy.

Table 18.1 Myositis-specific autoantibodies

1. Antibodies to aminoacyl-tRNA synthetases (“anti-synthetase antibodies”):
• Anti-Jo1 antibodies, directed against histidyl-tRNA synthetase
Other antisynthetase antibodies are directed against other specific synthetases:
• The OJ, EJ, PL-7, PL-12, KS, and Zo antigens
– The antisynthetase antibodies are associated with interstitial lung disease, Raynaud’s phenomenon, inflammatory arthritis, and
mechanic’s hands
2. Antibodies to signal recognition particle (anti-SRP antibodies)
• Associated with necrotizing myopathy and treatment-refractory disease
3. Antibodies to Mi-2
• Associated with dermatomyositis
4. Other:
• Antibodies to hPMS-1
• Antibodies to a 155-kiloDalton antigen
• Anti-Ku and anti-PM-Scl antibodies
– associated with overlapping features of polymyositis and scleroderma
• Anti-CADM-140 antibodies
– associated with amyopathic dermatomyositis
• Anti-155/140 antibody
– Distinct from the anti-155 and anti-CADM-140 antibodies. Associated with flagellate erythema and malignancies.
18 Inflammatory Myopathies 197

Comment: Arthritis is found in up to 90% of patients with the Pearl: The presence of anti-p155/140 antibodies and the
antisynthetase syndrome (Love et al. 1991) and is the pre- absence of other MSAs or MAAs are associated with an
senting feature in some patients, leading to a diagnosis of increased risk of malignancy in adult DM.
rheumatoid arthritis before other features of the syndrome
Comment: Three studies have found an increased frequency
become evident (see above). The arthritis is not usually ero-
of cancer in adult DM patients who have an autoantibody
sive, but it can be deforming (Oddis et al. 1990). Patients
that immunoprecipitates 155 and 140 kD proteins (Chinoy
with anti-PM-Scl can also have arthritis as part of an overlap
et al. 2007; Kaji et al. 2007; Targoff et al. 2006). The group
syndrome that may also demonstrate features of myositis
with cancer-associated myositis also has a low frequency of
(either PM or DM) and/or scleroderma.
other MSAs and MAAs. When used together, these features
Pearl: The myositis associated with anti-PM-Scl and anti- are helpful in assessing the likelihood of cancer in patients
U1RNP tends to be milder and easier to treat than the myo- with DM (Chinoy et al. 2007). Unfortunately, testing for this
sitis associated with antisynthetase and anti-SRP “doublet” is not commercially available and can only be
autoantibody subsets. done in research laboratories specializing in the testing of
MSAs and MAAs.
Comment: The myositis associated with anti-PM-Scl anti-
bodies often occurs in overlap with scleroderma. These
patients have milder and more responsive myositis than other
autoantibody subsets (Jablonska and Blaszczyk 1999) and 18.6 Imaging
respond to lower doses of glucocorticoids than are usually
used in myositis. Similarly, patients with myositis and the Pearl: Magnetic resonance imaging (MRI) findings may dis-
U1RNP autoantibody generally respond well to lower doses tinguish PM from IBM.
of glucocorticoids and have less long-term morbidity related
to their myopathy when compared to the generally poor out- Comment: MRI of the thighs with T1 and short tau inversion
comes seen in myositis patients with antisynthetase and anti- recovery (STIR) sequences can be helpful in distinguishing PM
SRP autoantibodies. from sporadic IBM (sIBM) (Fig. 18.4). PM is characterized by

a b

c d

Fig. 18.4 Axial STIR and T1-weighted MRI images through the mid- and atrophy in all musculature in inclusion body myositis, particularly
section of the thighs of two patients – one with polymyositis (a) (top intense in the anterior muscle group. Polymyositis tends to show inflam-
left and (b) top right), the other with inclusion body myositis (c) bottom mation and atrophy in a fascial distribution (Figure courtesy of Dr.
left and (Fig. 18.4d bottom right). Note the widespread inflammation Mark Gourley)
198 I. N. Targoff et al.

isolated inflammation that is symmetrical and generally favors PM cases are in fact mimickers of PM. Common mimickers
the posterior muscle group. IBM is more likely to be asymmet- of PM are IBM, muscular dystrophies, mitochondrial or met-
ric with fatty replacement and to involve muscles of the anterior abolic myopathies, and drug-induced syndromes, the prime
thigh and more distal musculature (Dion et al. 2002). example of which is a statin myopathy.
Myth: Exercise is contraindicated in patients with inflamma-
tory myopathy during the initial phase of treatment.
18.7 Inclusion Body Myositis
Reality: A supervised exercise program is a necessary pre-
scription when treating patients with an inflammatory myo-
Myth: The presence of rimmed vacuoles on muscle biopsy is
pathy. The long-held belief that exercise harms patients who
diagnostic of IBM.
have active myositis has been disproved by studies demon-
Reality: Although rimmed vacuoles are a characteristic strating benefits in strength, improvements in quality of life,
pathological feature in IBM, they may also occur in other and the absence of muscle disease exacerbations when an
neuromuscular disorders, such as distal myopathies, ocu- appropriately structured exercise program is initiated
lopharyngeal myopathy, PM, rigid spine syndrome, congen- (Alexanderson et al. 2007). The program should be super-
ital myopathies, limb girdle muscular dystrophies, and vised and instituted in a graded manner so as not to cause
various neurogenic diseases. The patient’s clinical features, muscle pain or worsening weakness.
the full pathologic picture, the response to therapy, and the
Pearl: Combination therapy with methotrexate and azathio-
overall disease course are all critical components in estab-
prine may be effective in the treatment of refractory myositis
lishing the diagnosis of IBM. Similarly, the classic finding
when either agent alone is ineffective.
of rimmed vacuoles in the muscle biopsy of patients with
IBM may require repeated biopsies over time. Comment: Although methotrexate or azathioprine in con-
junction with glucocorticoids is often at least partially
Myth: Dysphagia is less common in patients with IBM than
effective in the treatment of myositis, some patients are
DM and PM.
refractory to either methotrexate or azathioprine alone.
Reality: When dysphagia is found in a patient with an inflam- The combination of oral methotrexate and azathioprine
matory myopathy, IBM is the most likely diagnosis. In a has evidence-based support for its use in refractory myo-
study of 62 patients with IIM-associated dysphagia, 42% had sitis, showing benefit in some patients who previously
IBM, 29% had DM, 15% had PM, and 15% had overlap syn- demonstrated an inadequate response to either methotrex-
dromes (Oh et al. 2007). The dysphagia of IBM is reported ate or azathioprine alone (Villalba et al. 1998). Clinicians
by the patient as feeling like a “blocking sensation” with should consider combination therapy in myositis in a fash-
swallowing. It is often associated with the findings of cri- ion similar to that utilized in other connective tissue
copharyngeal spasm or achalasia on cinesophagram studies. diseases.
Pearl: The presence of a foot drop in a patient with “refrac- Pearl: Clinical improvement in muscle weakness lags behind
tory PM” is IBM until proven otherwise. the response in the serum CK and an increase of the serum
CK often predates a flare of myositis.
Comment: Although distal weakness may occur in the IIMs,
this is often in the setting of severe and/or chronic PM or Comment: Although the serum CK should not be used as the
DM. However, IBM is associated with both distal and asym- only determinant of treatment in patients with myositis (see
metric muscle weakness as well as muscle atrophy. The find- above), it can provide useful clues in the management of
ing of a foot drop (which may be subtle), manifested by myositis. Many times patients will continue to complain of
weakness of the ankle dorsiflexors should lead the clinician some degree of muscle weakness even though their CK has
to consider IBM. This may be unilateral but may be bilateral normalized. They should be reassured that the clinical
and variable in severity between both lower extremities. response to therapy lags behind the biochemical response
and that with continued glucocorticoid tapering (and the
maintenance of other immunosuppressive therapy, as appro-
18.8 Treatment priate), their muscle strength will likely improve. Similarly,
patients who are clinically stable but demonstrate a rising
CK from the normal range to elevated values may be headed
Myth: Many cases of PM are refractory to immunosuppres-
for a disease flare. Although the clinician would not want to
sive therapy.
simply increase immunosuppressive therapy in such
Reality: This statement is true in cases associated with anti- instances, he/she should carefully monitor such patients for
SRP autoantibodies. However, many “steroid-refractory” an impending disease flare.
18 Inflammatory Myopathies 199

Myth: The presence or absence of peripheral eosinophilia Kagen LJ, Aram S. Creatine kinase activity inhibitor in sera
helps distinguish methotrexate lung from other causes of from patients with muscle disease. Arthritis Rheum 1987; 30:
213–217
interstitial lung disease. Kaji K, Fujimoto M, Hasegawa M, et al Identification of a novel autoanti-
body reactive with 155 and 140 kDa nuclear proteins in patients with
Reality: The initial cases reported of methotrexate lung
dermatomyositis: An association with malignancy. Rheumatology
described peripheral eosinophilia (Whitcomb et al. 1972). 2007; 46:25–28
More comprehensive studies performed subsequently, how- Kao AH, Lacomis D, Lucas M, et al Anti-signal recognition particle
ever, reveal that eosinophilia is a nonspecific finding found in autoantibody in patients with and patients without idiopathic inflam-
matory myopathy. Arthritis Rheum 2004; 50:209–215
only about one-third of the cases of methotrexate lung.
Koenig M, Fritzler MJ, Targoff IN, et al Heterogeneity of autoantibod-
ies in 100 patients with autoimmune myositis: Insights into clinical
features and outcomes. Arthritis Res Therapy 2007; 9:R78
References Love LA, Leff RL, Fraser DD, et al A new approach to the classification
of idiopathic inflammatory myopathy: Myositis-specific autoanti-
bodies define useful homogeneous patient groups. Medicine 1991;
Alexanderson H, Dastalmachi M, Esbjornsson-Liljedahl M, et al 70:360–374
Benefits of intensive resistance training in patients with chronic Miller T, Al Lozi MT, Lopate G, et al Myopathy with antibodies to the
polymyositis or dermatomyositis. Arthritis Care Res 2007; 57:768. signal recognition particle: Clinical and pathological features.
Bohan A, Peter JB. Polymyositis and dermatomyositis. Parts 1 and 2 N. J Neurol Neurosurg Psychiatry 2002; 73:420–428
Engl J Med 1975; 292:344–3447,403–407 Oddis CV, Medsger TA Jr., Cooperstein LA. A subluxing arthropathy
Buchbinder R, Forbes A, Hall S, Dennett X, Giles G. Incidence of associated with the anti-Jo-1 antibody in polymyositis/dermatomy-
malignant disease in biopsy-proven inflammatory myopathy. A pop- ositis. Arthritis Rheum 1990; 33:1640–1645
ulation-based cohort study. Ann Intern Med 2001; 134:1087–1095. Oddis CV, Okano Y, Rudert WA, et al Serum autoantibody to the nucle-
Callen JP. Relation between dermatomyositis and polymyositis and olar antigen PM-Scl: clinical and immunogenetic associations.
cancer. Lancet 2001; 357:85–86 Arthritis Rheum 1992; 35:1211–1217
Chinoy H, Fertig N, Oddis CV, et al The diagnostic utility of myositis Oh TH, Brumfield KA, Hoskin TL, et al Dysphagia in inflammatory
autoantibody testing for predicting the risk of cancer-associated myopathy: Clinical characteristics, treatment strategies, and out-
myositis. Ann Rheum Dis 2007; 66:1345–1349 come in 62 patients. Mayo Clin Proc 2007; 82:441–447
Dimitri D, Andre C, Roucoules J, et al Myopathy associated with anti- Quain RD, Teal V, Taylor L, et al Number, characteristics, & classifica-
signal recognition peptide antibodies: Alinical heterogeneity contrasts tion of dermatomyositis patients seen by dermatology & rheumatol-
with stereotyped histopathology Muscle Nerve 2007; 35:389–395 ogy at a tertiary medical center. J Am Acad Dermatol 2007;
Dion E, Cherin P, Payan C, et al Magnetic resonance imaging criteria 57:937–943
for distinguishing between inclusion body myositis and polymyosi- Rider LG, Giannini EH, Brunner HI, et al International consensus on
tis. J Rheumatol. 2002; 29:1897 preliminary definitions of improvement in adult and juvenile myosi-
Erlacher P, Lercher A, Falkensammer J, et al Cardiac troponin and beta- tis. Arthritis Rheum 2004; 50:2281–2290
type myosin heavy chain concentrations in patients with polymyosi- Sato S, Kuwana M, Hirakata M. Clinical characteristics of Japanese
tis or dermatomyositis. Clin Chim Acta 2001; 306:27–33 patients with anti-OJ (anti-isoleucyl-tRNA synthetase) autoantibod-
Friedman AW, Targoff IN, Arnett FC. Interstitial lung disease with autoan- ies. Rheumatology 2007; 46:842–845
tibodies against aminoacyl- tRNA synthetases in the absence of clini- Schmidt WA, Wetzel W, Friedlander R, et al Clinical and serological
cally apparent myositis. Semin Arthritis Rheum 1996; 26:459–467 aspects of patients with anti-Jo-1 antibodies–an evolving spectrum
Gerami P, Schope JM, McDonald L, et al A systematic review of adult- of disease manifestations. Clin Rheumatol 2000; 19:371–377
onset clinically amyopathic dermatomyositis (dermatomyositis sine Targoff IN, Arnett FC. Clinical manifestations in patients with antibody
myositis): A missing link within the spectrum of the idiopathic to PL-12 antigen (alanyl-tRNA synthetase). Am J Med 1990; 88:
inflammatory myopathies. J Am Acad Dermatol 2006; 54:597–613 241–251
Hengstman GJ, van Brenk L, Vree Egberts WT, et al High specificity of Targoff IN, Trieu EP, Miller FW. Reaction of anti-OJ autoantibodies
myositis specific autoantibodies for myositis compared with other with components of the multi-enzyme complex of aminoacyl-tRNA
neuromuscular disorders. J Neurol 2005; 252:534–537 synthetases in addition to isoleucyl-tRNA synthetase. J Clin Invest
Hengstman GJ, ter Laak HJ, Vree Egberts WT, et al Anti-signal recog- 1993; 91:2556–2564
nition particle autoantibodies: Marker of a necrotising myopathy. Targoff IN, Mamyrova G, Trieu EP, et al A novel autoantibody to a 155-
Ann Rheum Dis 2006; 65:1635–1638 kd protein is associated with dermatomyositis Arthritis Rheum
Hill CL, Zhang Y, Sigurgeirsson B, Pukkala E, Mellemkjaer L, Airio A, 2006; 54:3682–3689
Evans SR, Felson DT. Frequency of specific cancer types in der- Tillie-Leblond I, Wislez M, Valeyre D, et al Interstitial lung disease and
matomyositis and polymyositis: a population-based study. Lancet anti-Jo-1 antibodies: Difference between acute and gradual onset.
2001; 357:96–100 Thorax 2008; 63:53–59
Hirakata M, Suwa A, Takada T, et al Clinical and immunogenetic fea- Villalba L, Hicks J, Adams E, et al Treatment of refractory myositis: A
tures of patients with autoantibodies to asparaginyl-transfer RNA randomized crossover study of two new cytoxic regimens. Arthritis
synthetase. Arthritis Rheum 2007; 56:1295–1303 Rheum 1998; 41:392–399
Jablonska S, Blaszczyk M. Scleroderma overlap syndromes. Adv Exp Whitcomb ME, Schwarz MI, Tormey DC. Methotrexate pneumonitis:
Med Biol 1999; 455:85–92 Case report and review of the literature. Thorax 1972; 27(5):636–639
 Juvenile Dermatomyositis
19
Ann M. Reed, Clarissa A. Pilkington, Brian M. Feldman,
Lauren M. Pachman, and Lisa G. Rider

Gottron’s papules (Fig. 19.2), for the classification of a

19.1 Overview of Juvenile Myositis patient as having JDM. In addition, a patient must have at
least two of the following clinical or laboratory criteria:
› Children with inflammatory myopathies have clinical
• Symmetric, proximal muscle weakness
and laboratory features, as well as risk factors and out-
• Elevation of one or more “muscle” enzyme in the blood
comes that overlap with their adult counterparts.
(CK, aldolase, alanine aminotransferase (ALT), AST, or
However, a number of important differences exist
LDH)
between pediatric and adult disease.
• Abnormal electromyogram, with polyphasic small motor
› Juvenile dermatomyositis (JDM), the most common of
unit potentials, high-frequency discharges, spontaneous
this group of illnesses in children, has an incidence of
fibrillations, positive sharp waves, or increased insertional
approximately 3.2 cases/million children/year.
irritability
› JDM has a gender ratio of 2 girls:1 boy.
• An abnormal muscle biopsy
› JDM has a mean age of 6.7 years at disease onset.
› The duration of active disease before the time of diag- In the absence of characteristic rashes, a diagnosis of juve-
nosis alters the patients’ features at presentation as nile polymyositis would require at least three of the clinical
well as their clinical outcomes. or laboratory features.
› Environmental and genetic factors affect the child’s Muscle biopsies are extremely helpful in supporting the
susceptibility to these conditions and also alter the diagnosis of myositis, in confirming the specific type if the
inflammatory response, adding heterogeneity to disease clinical differentiation is unclear, and in excluding a number of
pathophysiology. inflammatory myopathy mimickers (Table 19.2). Typical biopsy
findings in JDM include perifascicular atrophy, perivascular
inflammatory infiltrate, degenerating and regenerating myofi-
bers, and centralization of nuclei. However, many patients with
JDM fulfill three clinical and laboratory features, including the
presence of weakness, characteristic rashes, and elevation of
Pearl: JDM is not just a muscle condition, but rather a sys- serum muscle enzyme levels. Therefore, a muscle biopsy is
temic disease. not necessary in every case of JDM.
Fewer than 40% of the surveyed pediatric rheumatologists
Comment: The pathological basis of JDM is systemic vascul-
in North America and Europe rely on a muscle biopsy for
opathy, which affects the smallest blood vessels. Consequently,
diagnosis. Most rely upon magnetic resonance imaging (MRI)
a variety of organ systems can be affected (Table 19.1).
findings (Fig. 19.3) to confirm the diagnosis (Brown et al.
Interstitial lung disease, dyspnea on exertion, dysphonia, dys-
2006). However, muscle histopathologic features provide
phagia, vasculopathy (vasculitis) of the gastrointestinal tract,
important prognostic information in many cases (Wedderburn
and bowel dysmotility are relatively common in JDM.
et al. 2007b; Miles et al. 2007; Lopez de Padilla et al. 2007).
Cerebral, ocular, and cardiac manifestations are rare in JDM,
A muscle biopsy should be performed in patients who do
but can be life threatening.
not have a rash characteristic of JDM (i.e., heliotrope ery-
Myth: The diagnosis of JDM requires a muscle biopsy. thema or Gottron’s papules). A muscle biopsy is also indi-
cated if atypical features are present.
Reality: The classical criteria for inflammatory myopathies
(Bohan and Peter 1975 a,b) require the presence of charac- Myth: Muscle weakness and inflammation in a child always
teristic rashes, including the heliotrope rash (Fig. 19.1) or indicates a diagnosis of idiopathic juvenile myositis.

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 201

DOI:10.1007/978-1-84800-934-9_19, © Springer Science + Business Media B.V. 2009
202 A. M. Reed et al.

Table 19.1 Clinical features typically seen in patients with juvenile

dermatomyositis (From Feldman et al. 2008. Reprinted with permission
from Elsevier)
Constitutional
• Fever: 16–65%
• Adenopathy: 20%
• Lethargy: 10%
Pulmonary
• Dyspnoea: 7–43%
Gastrointestinal
• Dysphonia or dysphagia: 18–44%
• Gastrointestinal symptoms: 22–37%
Musculoskeletal
• Weakness: 95%
• Myalgia or arthralgia: 25–73%
• Arthritis: 23–58% Fig. 19.2 Gottron’s papules. Raised erythematous patches overlying
• Contractures: 26–27% the knuckles, which are characteristic lesions of dermatomyositis
• Raynaud’s disease: 9–14%
Cutaneous
• Gottron’s papules: 57–100%
Patients with myositis have certain clinical and laboratory
• Heliotrope rash: 66–100% clues, which suggest the diagnosis of myositis and distinguishing
• Nailfold capillary changes: 91% myositis from other causes of muscle dysfunction (Miller 2005).
• Malar or facial rash: 42–73% Patients with myositis often have features of systemic rheumatic
• Mouth ulcers: 35% diseases, including fever, photosensitive rashes, periungual capil-
• Skin ulcers: 23–30% lary changes, and arthritis. Approximately 70% of children with
• Limb oedema: 11–32%
myositis have a positive anti-nuclear antibody. They may also
• Calcinosis: 6–30%
have a family history of other forms of autoimmune disease.
• Lipodystrophy: 10–14%
Patients with myositis have a characteristic pattern of
weakness that is symmetric and involves primarily proximal
and axial muscles. In other forms of muscle dysfunction in
children, the weakness can involve both proximal and distal
muscles; affect the facial, ocular, or scapular muscles; be
asymmetric; or be worsened by fasting or exercise. The level
of creatine kinase (CK) elevation can rise to several thousand
units per liter (U/L) (a normal value is in the order of
250 U/L). However, a significant percentage of patients with
JDM have normal CK levels. Finally, in contrast to patients
with other pediatric muscle disorders, patients with myositis
usually respond well to glucocorticoids and other immuno-
suppressive therapy.
Myth: The serum CK level is a good indicator of disease
activity in childhood myositis.
Reality: The serum concentration of enzymes derived from
Fig. 19.1 Heliotrope discoloration of the eyelids and malar/facial
erythema muscle has been used traditionally as an indicator of JDM
disease activity and to discriminate active disease (Guzman
et al. 1994). However, serum levels of CK and other muscle-
Reality: Muscle weakness or fatigue may result from pri- derived enzymes often correlate poorly with measures of
mary muscle tissue abnormalities or from factors that affect muscle strength in JDM; CK concentrations may improve
muscle tissue only indirectly. Chronic muscle inflammation without a corresponding improvement in muscle strength.
may result from infection, medications, local trauma, or be Serum CK levels sometimes return to normal even when the
idiopathic, as in juvenile dermatomyositis (JDM). However, illness remains active (Rider 2002). The longer the delay in
many other myopathies other than juvenile myositis can lead the diagnosis of JDM, the greater the likelihood that serum
to muscle weakness or fatigue, and some may have associ- levels of CK and other muscle enzymes will be normal upon
ated muscle inflammation (see Table 19.2). measurement (Pachman et al. 2006). The points apply equally
19 Juvenile Dermatomyositis 203

Table 19.2 Differential diagnosis of childhood idiopathic inflammatory

myopathies. In many of these conditions, a diagnosis is facilitated by
muscle biopsy; muscle biopsy should be strongly considered in the
absence of typical rashes of juvenile dermatomyositis, including
heliotrope or Gottron’s papules. For further information, see http://
www.neuro.wustl.edu/neuromuscular/index.html (From Feldman et al.
2008. Reprinted with permission from Elsevier)
Group Condition
Weakness alone
Muscular Dystrophies (MD) Limb-girdle dystrophies
Dystrophinopathies
Fascioscapulohumeral dystrophy
Other dystrophies
Metabolic myopathies Muscle glycogenocses
(glycogen storage diseases)
Lipid storage disorders
Mitochondrial myopathies
Endocrine myopathies Hypothyroidism
Hyperthyroidism
Cushing syndrome/exogenous Fig. 19.3 Magnetic resonance image of thigh muscles, axial view. This
steroid myopathy short tau inversion recovery image demonstrates muscle edema, seen in
Diabetes Mellitus active myositis
Drug-induced myopathy Consider in patients taking any of
the following drugs or biologic
therapies: statins, interferon a,
glucocorticoids, hydroxychloro-
quine, diuretics, amphotericin b, (IFN)-regulated proteins. However, abnormalities of a mea-
“caine” anesthetics, growth sure in isolation may not be accurate (Rider 2002; Baechler
hormone, cimetidine, and et al. 2007; Griffin and Reed 2007). MRI of the upper leg or
vincristine.
shoulder muscles can be helpful. Short tau inversion recovery
Neuromuscular Myasthenia gravis
transmission disorders (STIR) or fat-suppressed images on T2-weighted sequences
Spinal muscular atrophy assess the degree of muscle edema or inflammation.
Infectious myopathy T1-weighted images are useful in assessing the degree of
Viral Enterovirus, influenza, coxsackievi- muscle atrophy and fibrosis (see Fig. 19.3). However, the reli-
rus, echovirus, parvovirus, able application of these imaging techniques across medical
poliovirus, hepatitis B, HTLVI
centers is currently lacking.
Bacterial and parasitic Staphylococcus, streptococcus,
organisms toxoplasmosis, trichinosis, Myth: The rash of JDM is relatively benign. Treatment strat-
Lyme borreliosis
egies should be directed toward the underlying muscle
Other rheumatic Systemic lupus erythematosus
conditions (SLE), scleroderma, juvenile disease.
idiopathic arthritis, mixed
connective tissue disease, Reality: Many different types of rashes occur in JDM. The
idiopathic vasculitis heliotrope rash, marked by red or purple discoloration of the
Other inflammatory Inflammatory bowel disease, upper eyelids, often associated with edema, occurs in 80% of
conditions celiac disease patients (see Fig. 19.1). Gottron’s papules, raised erythema-
Rash without weakness Psoriasis, eczema, allergy tous patches that appear over the extensor surfaces of joints are
each seen in about 80% of patients with JDM (see Fig. 19.2).
well to aldolase, lactate dehydrogenase (LDH), and aspartate Cutaneous ulcerations, which are present in fewer than
aminotransferase (AST) as to CK. 10% of all JDM patients, portend a difficult illness course.
Current data suggest that serum LDH may be the most The periungual nailfold capillaries typically demonstrate ves-
clinically useful serum muscle-derived enzyme, as it best sel dilation, avascularity, and morphologic changes (Fig. 19.4).
predicts the global disease activity in JDM patients with These correlate with active skin disease. An erythematous
longstanding disease (Rider 2002). The combination of LDH appearance of the skin at either the sites of classic lesions
in combination with AST may be best in predicting flares of (the eyelids, knuckles) or elsewhere (the face, shoulders,
JDM activity (Guzman et al. 1994). etcetera) often indicates active cutaneous disease. However,
Other proposed biomarkers in JDM are von Willebrand this must be distinguished from poikiloderma atrophicans,
factor VIII-related antigen, flow cytometry of peripheral post-inflammatory dyspigmentation, and cutaneous atrophy,
blood lymphocyte subsets, neopterin, and interferon all of which are consistent with skin damage.
204 A. M. Reed et al.

disease. In this regard, photoprotective measures may be

helpful, including the use of sunscreen or sunblock that
decreases exposure to ultraviolet A and B (UVA and UVB)
light, use of wide brimmed hats and photoprotective cloth-
ing, and avoidance of sun exposure during peak day-light
hours.
Pearl: A rash that persists despite therapy heralds a chronic
course of illness.
Comment: Rash can be a persistent manifestation of JDM. In
our experience, the presence of persistent rash and periun-
gual nailfold capillary changes after 3 months of systemic
therapy predicts a chronic course of disease.
Myth: Autoantibodies in JDM are phenomenology that has
little bearing on patients’ clinical features and prognosis.
Fig. 19.4 Periungual nailfold capillary changes. Dilated periungual
capillaries, with bushy loop formation, capillary dropout and subungual Reality: Myositis-specific autoantibodies (MSA) are autoanti-
hemorrhage also evident, under magnification bodies that are found in many adult patients who have specific
idiopathic inflammatory myopathies. These autoantibodies
are also detected in children and adolescents with these disor-
ders, but less often than in adults. Anti-Jo1 autoantibodies,
detected in 5–8% of patients with juvenile myositis, are asso-
ciated with anti-synthetase syndrome. The anti-synthetase
syndrome consists of arthritis, Raynaud’s phenomenon,
mechanic’s hands, fevers, and interstitial lung disease in the
setting of moderate to severe muscle weakness and high serum
CK levels (Rider and Miller 1997; Wedderburn et al. 2007a).
Other autoantibodies, though less common than anti-syn-
thetase antibodies, help predict disease subset and treatment
response (Rouster Stevens 2008; Rider et al. 1994).
Patients with juvenile myositis often have anti-nuclear
antibodies of unknown specificity. A novel autoantibody that
immunoprecipitates 155- and 140-kD proteins has been
found in the sera of up to 30% of JDM patients (Targoff et al.
Fig. 19.5 Calcinosis. Small subcutaneous nodules extruding from the 2006; Gunawardena et al. 2008). JDM patients with anti-
skin over the elbows in the region of overlying Gottron’s papules
p155 autoantibodies have more cutaneous involvement and a
prolonged illness course (Gunawardena et al. 2008). This
Active, severe rash even in a JDM patient who does not autoantibody has also been associated with generalized
have muscle weakness is an indication for systemic treat- rather than localized lipodystrophy (Bingham et al. 2008).
ment. Agents that may be effective are enteral or parenteral
Myth: Dystrophic calcifications are associated with a good
glucocorticoids, methotrexate, hydroxychloroquine, mycophe-
outcome.
nolate mofetil, intravenous immunoglobulin, and cyclosporine.
These agents may be prescribed individually, but are often Reality: Dystrophic calcification can develop in either the
used in combination. Active rashes also frequently benefit muscle or the skin of a patient with JDM. One of the first
from the use of topical therapies, including topical glucocor- publications concerning the outcome of JDM in the era
ticoids or topical tacrolimus/pimecrolimus. before the advent of glucocorticoids presented the concept
that if the child with JDM lived to develop calcifications, the
Pearl: The rash of JDM is photosensitive.
outcome was “good”. Approximately 30% of patients fell
Comment: The purpose of treatment for rash is to prevent the into that category (Bitnum 1964). In the glucocorticoid era,
scarring damage of JDM, including the chronic rashes men- 20–40% of JDM patients still develop dystrophic calcifica-
tioned earlier and, most importantly, calcinosis (Fig. 19.5). tion, although the extent and severity of calcification may be
Protection from ultraviolet A and B (UVA and UVB) light is less than those observed before the availability of current
hypothesized to help control skin and potentially muscle therapies (Huber et al. 2000; Sheshadri et al. 2008).
19 Juvenile Dermatomyositis 205

Calcifications are often accompanied by a significant loss • Bisphosphonates, which inhibit calcium hydroxyapatite
of range of motion, physical dysfunction, as well as infection. formation, macrophage function, and the mobilization of
They are a major contributor to the morbidity and mortality of bone calcium.
JDM (Huber et al. 2000). The development of calcifications is
For some patients with tumorous deposits, surgical removal is
associated with a young age at diagnosis (Pachman et al. 2006),
indicated. This includes patients with chronic or severe pain,
delay in the institution of effective therapy, and a genetic pre-
loss of function, recurrent infections, or non-healing ulcers.
disposition to increased production of TNF-a(alpha). The use
The optimal timing of surgery is when myositis is quiescent.
of high-dose intravenous glucocorticoids at diagnosis may
decrease the development of calcifications (Fisler et al. 2002), Myth: Treatment of children with JDM can be standardized.
but the deposits may appear later in the disease course, if
Reality: The variability of JDM mitigates efforts to propose
unsuppressed inflammation persists or recurs.
single therapeutic strategies that might apply to all patients.
A variety of treatments have been reported anecdotally,
Patients can present with a variety of rashes, arthritis, joint
including the use of calcium channel blockers, probenecid,
contractures, esophageal disease leading to dysphagia, mes-
aluminum hydroxide, and bisphosphonates, but none appears
enteric vasculitis, fatigue, dystrophic calcification, and vary-
consistently effective in the dissolution of calcifications once
ing degrees of weakness and functional disability (Rider
they have formed. Calcinosis can regress, sometimes without
2007; Feldman et al. 2008). The duration of the inflamma-
treatment, but not in a predictable manner. It can also develop
tory process, the age of the child at disease onset, autoanti-
even as medical treatments are being administered (McCann
body status, and genetic factors all contribute to the disease’s
et al. 2006). In short, contrary to the earliest publications on
heterogeneity (Pachman et al. 2006; Rider 2007).
outcomes in JDM, calcifications are not associated with good
Genetic markers such as certain major histocompati-
clinical outcomes.
bility complex class II alleles, tumor necrosis factor-
Pearl: There is no definitive treatment for calcinosis. alpha-308 and interleukin-1-alpha polymorphisms all
appear to contribute to JDM (Wedderburn et al. 2007a;
Comment: Current approaches to the treatment of calcinosis Pachman et al. 2000; Mamyrova et al. in press). Preliminary
associated with JDM are largely based on anecdotal reports. results of novel immune activation markers suggest molecu-
Regression of calcinosis or retardation of progression has lar biomarkers from the IFN and IL-17 pathways predict dis-
been observed in case reports following treatment with ease resistant and drug responsive phenotypes (Baechler
colchicine, hydroxychloroquine, IVIG, cyclosporine, intra- et al. 2007).
lesional steroid and most recently with infliximab (Rider and
Miller 1997; Riley et al. 2008; Hazen et al. 1982). One prob- Myth: Most children with JDM do not have long-term com-
lem with these uncontrolled studies is that the natural history plications of their illness.
of calcinosis is unpredictable: spontaneous regression may
Reality: Disease damage, defined as irreversible manifesta-
occur, making positive successes difficult to interpret.
tions of illness present at least 6 months and attributed to
The first aim of therapy in patients with JDM is to prevent
previously active disease, medication side effects, or other
calcinosis. Several reports suggest that the early initiation of
concurrent illnesses, occurs in many patients with JDM.
intensive anti-inflammatory therapy for JDM is effective in
Cutaneous scarring or atrophy, joint contractures, persistent
preventing calcinosis. This strategy includes the early admin-
muscle dysfunction or weakness, dysphagia, and dysphonia
istration of daily oral glucocorticoid therapy in adequate
are frequently reported in long-term outcome studies.
doses, as well as the early introduction of methotrexate or
Endocrine complications are also observed, including growth
intravenous methylprednisolone (Fisler et al. 2002).
failure (10% of JDM patients) and osteoporosis with frac-
Other therapeutic approaches have aimed to inhibit the
tures (Huber et al. 2000; Ramanan et al. 2005).
deposition of calcinosis through mechanisms intended to
Metabolic abnormalities, including insulin resistance and
disrupt calcium-phosphorous homeostasis. The following
hyperlipidemia, are frequent sequelae of JDM. Metabolic
medications have been attempted for this purpose:
abnormalities reported in JDM are:
• Diltiazem, a calcium channel-blocker
• Lipodystrophy, loss of subcutaneous fat
• Aluminum hydroxide, which decreases the intestinal
• Insulin resistance, with a visible clue being acanthosis
absorption of phosphate and lowers the serum calcium-
nigricans (Fig. 19.6)
phosphate product, thereby decreasing the likelihood of
• Diabetes mellitus
calcium deposition in tissues
• Probenecid, which increases the renal excretion of phos- Patients with partial and generalized lipodystrophy have a
phate, thereby diminishing the calcium-phosphorus prod- high prevalence of metabolic abnormalities, including glu-
uct in the serum cose intolerance, insulin resistance, and hypertriglyceridemia
206 A. M. Reed et al.

Gottron’s papules (see Fig. 19.2) may be misdiagnosed as flat

warts, psoriasis, or be missed altogether. When Gottron’s
lesions are not raised, they are termed “Gottron’s sign”.
In fair-skinned children, mild heliotrope or malar rashes
can be confused with their normal complexion, facial flush-
ing, or allergies. Up to 70% of patients with JDM have a
photosensitive rash; only a minority of these are malar rashes.
In dark-skinned children, mildly erythematous rashes with
hyperpigmentation may be difficult to discern from sur-
rounding skin.
Periungual capillary changes, most easily visible at the
nailfold using oil or water-soluble gel and an otoscope or
other magnified light source, are almost always present in
patients with JDM (see Fig. 19.4). The most sensitive find-
ing, however, is a reduction in the capillary density, which
can be difficult to appreciate without a measuring micro-
scope. Nonetheless, this is an important sign to look for and
may be a diagnostic clue in the presence of otherwise subtle
signs and symptoms.
Pearl: Nailfold capillary end row loops are important mark-
ers of JDM disease activity.
Comment: JDM patients frequently have periungual capil-
lary drop-out. The remaining capillaries are often dilated and
bushy, as seen under magnification using an ophthalmoscope
or DermLite™. The quantitative decrease in the number of
end row capillary loops in untreated children with JDM at
diagnosis correlates with skin inflammation rather than mus-
cle involvement. In a 36-month longitudinal study of JDM,
40% of JDM patients regenerated their end row capillary
Fig. 19.6 Acanthosis nigricans. Increased pigmented skin in the axilla,
loops as they responded to therapy. Persistent capillary loss
which is a sign of insulin resistance was highly associated with a delay in starting effective medi-
cal therapy and was related to chronically active cutaneous
inflammation rather than muscle Christian-Zäch et al., 2008.
(Bingham et al. 2008). Non-alcoholic steatohepatitis (NASH), Decreased nailfold end row capillary loops may also corre-
testosterone elevation, and menstrual irregularities are also late with impaired intestinal absorption.
frequently present in patients with total or generalized lip-
Pearl: Oral glucocorticoid therapy alone is not always suf-
odystrophy (Bingham et al. 2008).
ficient to control active disease in patients with JDM.
Pearl: The presenting signs and symptoms of JDM can be
Comment: Our understanding of effective treatments for
subtle.
JDM comes from observational studies and clinical experi-
Comment: Many patients with JDM are symptomatic for many ence. Standard treatment for JDM has been high-dose daily
months before a diagnosis is made. The duration of untreated oral glucocorticoids (e.g., up to 2 mg/kg/day of prednisone,
disease is a poor prognostic factor: longer duration is associ- at times in divided doses), which is continued until clinical
ated with increased long-term damage and a greater likelihood and laboratory improvements are evident and then reduced
of developing calcinosis (Pachman 1998). Physicians may slowly over a 2-year period (at least). With this regimen,
miss the early symptoms and signs of JDM, as they are often however, most patients develop treatment-related side-effects.
subtle and slow to evolve. Many patients are now treated with adjunctive immunosup-
Muscle weakness may be mild and often presents as fatigue. pressive medications in an attempt to spare high doses of
In younger children, clumsiness, more frequent falls, and daily glucocorticoids.
inability to keep up in play activities may become evident. The Some children with more severe myositis or those unable to
serum CK level may be normal in up to 40% of JDM patients swallow safely are treated with pulse intravenous methylpred-
at the time of diagnosis, even those with evident weakness. nisolone (usually 30 mg/kg/day, to a maximum of 1 g daily).
19 Juvenile Dermatomyositis 207

Intravenous methylprednisolone is proposed to minimize the worsening of their rash and possibly their muscle disease fol-
daily oral exposure. However, studies of mild to moderate JDM lowing exposure to the sun. Many of the rashes of DM are
patients have not shown that long-term outcome is altered by also photodistributed, suggesting a role for sun exposure in
intravenous glucocorticoid treatment (Sheshadri et al. 2008). their activity. Sun protection decreases the extent and sever-
The introduction of steroid-sparing agents, such as meth- ity of skin rash in these patients.
otrexate, cyclosporine, azathioprine, or intravenous immune These findings of a role for UV radiation, although pre-
globulin may limit the total glucocorticoid exposure and liminary, suggest a potential benefit in using photoprotective
results in fewer steroid-related side effects. The introduction measures as an adjunct to therapy. This includes avoiding
of methotrexate at the start of treatment does not impact sun exposure during peak hours of the day; using sunscreen
muscle strength outcomes or JDM activity, but does limit the or sun blocks (containing titanium dioxide and zinc oxide)
duration and total amount of glucocorticoid exposure and with a high sun protective factor (SPF) daily; reapplying
attenuates problems related to weight gain and growth fail- them several times a day; and using wide-brimmed hats and
ure (Ramanan et al. 2005). Methotrexate is now widely used photoprotective clothing.
by pediatric rheumatologists at the start of the illness.
Pearl: Bone health is important in JDM.
Standard therapy for the initial treatment of children with
JDM now ranges from oral glucocorticoids alone to oral glu- Comment: Children with active JDM symptoms are at risk
cocorticoids preceded by a 3-day pulse of methylpredniso- for bone fracture. They often have markedly diminished
lone, with or without methotrexate, cyclosporine, or another bone mineral density on dual x-ray absorptiometry studies
agent. Hydroxychloroquine is often employed specifically and decreased serum levels of osteocalcin (reflecting reduced
for the treatment of skin disease (Feldman et al. 2008). bone synthesis). Several factors contribute to decreased bone
density in JDM:
Pearl: Children with JDM may require the use of several
medications to achieve disease control. • A history of prolonged inflammation
• Decreased calcium absorption because of glucocorticoid
Comment: The aim of treatment in JDM patients is to control
use
the disease completely and achieve remission off therapy, with
• An intake of calcium and vitamin D that is insufficient for
few or no sequelae. Responses to treatment vary widely:
proper bone maintenance and repair
• Disease that is relatively easy to control in all respects
Children with JDM often have less UV exposure and there-
• Myositis that is difficult to control, with residual unac-
fore less natural vitamin D, because photoprotective mea-
ceptable weakness or functional impairment
sures are often recommended. Consequently, JDM patients
• Persistently active skin disease
should be provided with supplemental vitamin D and cal-
• Recalcitrant systemic manifestations such as dysphagia,
cium. Supplemental dosing needs to be individualized, but
interstitial lung disease, or gastrointestinal ulceration
general guidelines suggest 1,200 mg/day of calcium and
• Smoldering, mildly active disease complicated by the
800 IU/day of vitamin D for older children and 600 mg cal-
development of calcinosis.
cium/400 IU vitamin D for younger ones.
Most treatment-refractory patients do not respond simply to
increasing dosage of disease-modifying drugs. Their symptoms Pearl: Delays in skeletal growth and the onset of puberty
may improve somewhat with glucocorticoids, but usually only suggest uncontrolled disease or the effects of chronic gluco-
on a short-term basis accompanied by unacceptable side- corticoid administration.
effects. These patients inevitably require a combination of Comment: JDM patients whose disease is controlled in a sus-
immunosuppressive or immunomodulatory drugs. Disease- tained, effective manner grow and develop normally. A patient
modifying anti-rheumatic drugs (DMARDS), including cyclo- with growth delay should be investigated for subtle, uncon-
phosphamide, mycophenolate mofetil, cyclosporine, tacrolimus, trolled JDM disease activity or the presence of another autoim-
TNFa(alpha) inhibitors, and rituximab, have been employed mune condition, e.g., thyroid disease or inflammatory bowel
in treatment-resistant myositis or potentially life-threatening disease. Chronic glucocorticoid use also leads to growth delay.
organ involvement (Pilkington and Wedderburn 2005; Reed Successful DMARD use permits the control of JDM
2002). Anecdotal treatment successes have been reported, but and glucocorticoid tapering. In one study of methotrexate,
controlled studies of these agents have not been reported. children who received 15 mg/m2/week either orally or sub-
cutaneously from the time of diagnosis had half the cumu-
Pearl: There may be a role for sun protection in the manage-
lative glucocorticoid exposure and weight gain, improved
ment of JDM.
height velocity, and equivalent disease control when com-
Comment: Exposure to UVA and UVB light is of concern pared to patients treated with glucocorticoids alone
with JDM patients. A subset of patients report significant (Ramanan et al. 2005).
208 A. M. Reed et al.

Pearl: Acutely ill patients with JDM require aggressive Comment: DM and PM in adult patients have a strong asso-
treatment. ciation with malignancy (see Chapter 18). In contrast, the
association of JDM and malignancy is much weaker than in
Comment: Serious complications of JDM include acute gas-
adult dermatomyositis (Rider and Miller 1997). Because of
trointestinal ulceration or other acute bowel involvement
the rarity of events in which JDM and malignancy occur
pneumatosis intestinalis; aggressive interstitial lung disease;
together, patients with JDM whose diagnoses appear straight-
acute respiratory insufficiency caused by severe respiratory
forward do not require malignancy evaluation.
muscle weakness or air leaks such as pneumomediastinum or
pneumothorax; myocarditis; severe muscle weakness with an
inability to rise from bed; profound dysphagia leading to aspi-
ration risk; and (rarely) central nervous system vasculitis.
The management of such complications requires the ini- References
tiation of aggressive therapies, often in combination, to
reverse these manifestations quickly. Medications of poten- Baechler EC, Bauer JW, Slattery CA, et al An interferon signature in
tial utility in such scenarios include methylprednisolone the peripheral blood of dermatomyositis patients is associated with
30 mg/kg/day thrice (maximum dose: 1,000 mg), intravenous disease activity. Mol Med 2007; 13(1–2):59–68
immune globulin, and intravenous cyclophosphamide. The Bingham A, Mamyrova G, Rother KI, et al Predictors of acquired lip-
odystrophy in juvenile-onset dermatomyositis and a gradient of
monthly dose of cyclophosphamide may be divided into a severity. Medicine (Baltimore) 2008; 87(2):70–86
weekly dose (10–15 mg/kg) for more rapid effect. Bitnum S, Daeschner CW, Jr., Travis LB et al Dermatomyositis.
J Pediatr 1964; 64:101-31
Pearl: JDM patients may have an overlap syndrome or Bohan A, Peter JB. Polymyositis and dermatomyositis (first of two
another autoimmune condition. parts). N Engl J Med 1975a 292(7):344–347
Bohan A, Peter JB. Polymyositis and dermatomyositis (second of two
Comment: Up to 25% of patients with JDM meet classification parts). N Engl J Med 1975b; 292(8):403–407
criteria for another autoimmune disease, e.g., systemic sclero- Brown VE, Pilkington CA, Feldman BM, Davidson JE. An international
sis (scleroderma), systemic lupus erythematosus, juvenile consensus survey of the diagnostic criteria for juvenile dermatomyo-
sitis (JDM). Rheumatology (Oxford) 2006; 45(8):990–993
idiopathic arthritis, inflammatory bowel disease, type I diabe- Christen-Zäch S, Seshadri R, Sundberg J et al Pathological nailfold cap-
tes mellitus, or celiac disease (Pilkington and Reed 2007). illaroscopy changes in juvenile dermatomyositis: association with
Patients with JDM, in association with scleroderma features, duration of untreated disease, skin involvement over 36 months and
are (on average) older than those without scleroderma features; a non-unicyclic disease course. Arthritis Rheum 2008; 58:571-576
Feldman BM, Ravelli A. Maintaining flexibility: Joints and other con-
more likely to be girls; have a lengthy diagnostic delay; and siderations. In: LG Rider, LM Pachman, FW Miller, H Bollar (eds)
manifest a higher frequency of myositis-specific or myositis- Myositis and you: A guide to juvenile dermatomyositis for patients,
associated autoantibodies (Wedderburn et al. 2007a). families and health care providers. The Myositis Association;
Mixed connective tissue disease (MCTD) is characterized 2007:269–274
Feldman BM, Rider LG, Reed AM, Pachman LM. Juvenile dermato-
by the presence of a high-positive titer of anti-U1-RNP myositis and other idiopathic inflammatory myopathies of child-
autoantibody in the setting of symptoms and signs of rheu- hood. Lancet 2008; 371(9631):2201–2212
matic conditions such as Raynaud’s phenomenon and arthri- Fisler RE, Liang MG, Fuhlbrigge RC, Yalcindag A, Sundel RP.
tis. Children with MCTD often present with features of JDM, Aggressive management of juvenile dermatomyositis results in
improved outcome and decreased incidence of calcinosis. J Am
including Gottron’s papules or heliotrope rashes, proximal Acad Dermatol 2002; 47(4):505–511
weakness, and elevated serum muscle enzyme levels. Griffin TA, Reed AM. Pathogenesis of myositis in children. Curr Opin
Juvenile idiopathic arthritis can be associated with muscle Rheumatol 2007; 19(5):487–491
inflammation. However, the histopathological features of such Gunawardena H, Wedderburn LR, North J, et al Clinical associations of
autoantibodies to a p155/140 kDa doublet protein in juvenile der-
cases differ from those of classic JDM. Conversely, 60% of matomyositis. Rheumatology (Oxford) 2008; 47(3):324–328
JDM subjects have arthritis, usually as a non-erosive joint Guzman J, Petty RE, Malleson PN. Monitoring disease activity in juve-
condition that responds well to immunosuppressive therapy nile dermatomyositis: The role of von Willebrand factor and muscle
(Feldman and Ravelli 2007; Lindehammar and Lindvall 2004). enzymes. J Rheumatol 1994; 21:739–743
Hazen PG, Walker AE, Carney JF, Stewart JJ. Cutaneous calcinosis of
Other gastrointestinal conditions such as active hepatitis, scleroderma. Successful treatment with intralesional adrenal ste-
chronic biliary cirrhosis, inflammatory bowel disease, and roids. Arch Dermatol 1982; 118(5):366–367
celiac disease have been described in association with juve- Huber AM, Lang B, LeBlanc CM, et al Medium- and long-term func-
nile myositis. Endocrine conditions such as type-1 diabetes tional outcomes in a multicenter cohort of children with juvenile
dermatomyositis. Arthritis Rheum 2000; 43(3):541–549
and autoimmune thyroid disease are seen at an increased fre- Lindehammar H, Lindvall B. Muscle involvement in juvenile idiopathic
quency in association with JDM. arthritis Rheumatology (Oxford) 2004; 43(12):1546–1554
Lopez de Padilla CM, Vallejo AN, McNallan KT, et al Plasmacytoid
Pearl: Children with JDM do not require a routine evalua- dendritic cells in inflamed muscle of patients with juvenile dermato-
tion for malignancy. myositis. Arthritis Rheum 2007; 56(5):1658–1668
19 Juvenile Dermatomyositis 209

Mamyrova G., O’Hanlon TP, Sillers L, et al, for the Childhood Myositis Reed AM, Lopez M. Juvenile dermatomyositis: Recognition and treat-
Heterogeneity Collaborative Study Group. Cytokine gene polymor- ment. Paediatr Drugs 2002; 4(5):315–321
phisms as risk and severity factors for Juvenile Dermatomyositis. Rider L, eed A, James-Newton L, Feldman B, Ravelli A, Cabalar I,
Arthritis Rheum 2008; 58(12):3941-50 Villalba M, Rider LG, Miller FW. New perspectives on the idio-
McCann LJ, Juggins AD, Maillard SM, et al The Juvenile pathic inflammatory myopathies of childhood. Curr Opin Rheumatol
Dermatomyositis National Registry and Repository (UK and 1994; 6(6):575–582
Ireland)–clinical characteristics of children recruited within the first Rider LG. Outcome assessment in the adult and juvenile idiopathic inflam-
5 yr. Rheumatology (Oxford) 2006; 45(10):1255–1260 matory myopathies. Rheum Dis Clin North Am 2002; 28(4):935–977
Miles L, Bove KE, Lovell D, et al Predictability of the clinical course of Rider LG. The heterogeneity of juvenile myositis. Autoimmun Rev
juvenile dermatomyositis based on initial muscle biopsy: A retrospec- 2007; 6(4):241–247
tive study of 72 patients. Arthritis Rheum 2007; 57(7):1183–1191 Rider LG, Miller FW. Classification and treatment of the juvenile idio-
Miller, FW. Inflammatory myopathies: Polymyositis, dermatomyositis, pathic inflammatory myopathies. Rheum Dis Clin North Am 1997;
and related conditions. In: Koopman W, Moreland L (eds) Arthritis 23(3):619–655
and allied conditions - A textbook of rheumatology (15th edn, vol- Riley P, McCann LJ, Maillard SM, Woo P, Murray KJ, Pilkington CA.
ume 2). Philadelphia, PA, Lippincott Williams & Wilkins, 2005, pp. Effectiveness of infliximab in the treatment of refractory juvenile der-
1593–1620 matomyositis with calcinosis. Rheumatology 2008; 47(6): 877–880
Pachman LM, Hayford JR, Chung A et al Juvenile dermatomyositis at Rouster-Stevens KA, Pachman LM. Autoantibody to signal recognition
diagnosis: clinical characteristics of 79 children. J Rheumatol 1998; particle in African-American girls. J Rheumatol 35:927-929, 2008
25:1198-1204 Sheshadri R, Feldman BM, Ilowite N, Cawkwell G, Pachman LM. The
Pachman LM, Abbott K, Sinacore JM, et al Duration of illness is an role of aggressive corticosteroid therapy in patients with juvenile
important variable for untreated children with juvenile dermatomy- dermatomyositis: A propensity score analysis. Arthritis Rheum
ositis. J Pediatr 2006; 148(2):247–253 2008; 59(7):989–995
Pilkington C, Reed AM. When your child has more than one autoim- Targoff IN, Mamyrova G, Trieu EP, et al A novel autoantibody to a 155-
mune disease. In: Rider LG, Pachman LM, Miller FW, Bollar H kd protein is associated with dermatomyositis. Arthritis Rheum
(eds) Myositis and you (1st edn), Addition. Washington DC, The 2006; 54(11):3682–3689
Myositis Association, 2007, pp. 321–326 Wedderburn LR, McHugh NJ, Chinoy H, et al HLA class II haplotype
Pilkington CA, Wedderburn LR. Paediatric idiopathic inflammatory and autoantibody associations in children with juvenile dermato-
muscle disease: Recognition and management. Drugs 2005; 65(10): myositis and juvenile dermatomyositis-scleroderma overlap.
1355–1365 Rheumatology (Oxford) 2007a; 46(12):1786–1791
Ramanan AV, Campbell-Webster N, Ota S, et al The effectiveness Wedderburn LR, Varsani H, Li CK, et al International consensus on a
of treating juvenile dermatomyositis with methotrexate and proposed score system for muscle biopsy evaluation in patients with
aggressively tapered corticosteroids. Arthritis Rheum 2005; 52(11): juvenile dermatomyositis: A tool for potential use in clinical trials.
3570–3578 Arthritis Rheum 2007b; 57(7):1192–1201
 Vasculitic Neuropathy
20
David A. Chad and Peter Siao

20.2 Presentation
20.1 Vasculitic Neuropathy Overview
Pearl: Although mononeuritis multiplex is the best known
› Vasculitic neuropathy results from inflammation within
pattern of presentation of vasculitic neuropathy, other clini-
the blood vessel walls of the vasa nervorum, the blood
cal expressions of neuropathy are in fact more common.
vessels that supply peripheral nerves.
› The forms of systemic vasculitis most likely to be Comment: Mononeuropathy multiplex is defined as some
associated with vasculitic neuropathy are polyarteritis combination of sensory loss, weakness, or both in the terri-
nodosa, microscopic polyangiitis, the Churg–Strauss tory of two or more specific, clearly defined peripheral
syndrome, Wegener’s granulomatosis, and mixed nerves. This pattern of disease is the most widely recognized
cryoglobulinemia. presentation for vasculitic neuropathy. However, mononeu-
› Large-vessel vasculitides such as giant cell arteritis ropathy multiplex occurs in only a minority of patients with
and Takayasu’s arteritis seldom cause vasculitic neu- vasculitic neuropathy (Kissel and Mendell 1992). In fact,
ropathy. Similarly, vasculitic neuropathy is seldom (if this condition can present with a wide range of clinically
ever) a complication of the types of vasculitis that defined neurological patterns. Two patterns are relatively
involve the smallest blood vessels, e.g., cutaneous leu- common:
kocytoclastic angiitis (hypersensitivity vasculitis) and
• The most common pattern of presentation is a distal, sym-
Henoch–Schönlein purpura. Rather, vasculitic neurop-
metrical polyneuropathy. This is a length-dependent dis-
athy tends to be caused by systemic vasculitides that
order, characterized by sensory loss, motor weakness, or
have the ability to target medium-sized blood vessels.
both. It begins in the feet and extends proximally.
› Vasculitic neuropathy can also occur as a disease pro-
• The second most common pattern is an overlapping
cess that is limited to the peripheral nerves, sparing
mononeuropathy multiplex, defined as an asymmetrical
other organs. Such cases are sometimes referred to as
polyneuropathy that blends neurological deficits of simi-
“isolated vasculitis of the peripheral nervous system”
lar character in adjacent nerve territories.
or “nonsystemic vasculitic neuropathy.”
› Vasculitic neuropathy is generally a “length-dependent” Two additional manifestations of polyneuropathy are observed
disease process. Thus, the distal extremities tend to be less often:
involved more extensively than are the proximal
• A pure sensory neuropathy, predominantly involving
extremities and trunk importantly, vasculitic neuropa-
large nerve fibers (Seo et al. 2004).
thy can also present initally as a multifocal, nonlength
• A painful, small-fiber sensory polyneuropathy that targets
dependent process (arms affected earlier than legs).
the thinly myelinated and unmyelinated sensory nerve
› Vasculitic neuropathy has the potential to cause seri-
fibers (Lacomis et al. 1997).
ous morbidity (persisting weakness) through the resul-
tant infarctions of peripheral nerves. Many patients have a pattern of distal symmetrical polyneu-
› The combination of cyclophosphamide and high doses ropathy when they are first evaluated. This pattern might be
of glucocorticoids is an appropriate treatment regimen secondary to widespread, fulminant vasculitic involvement
for most patients with vasculitic neuropathy. of many peripheral nerves (Schaumburg et al. 1992). It may
also be caused by the involvement of individual peripheral
nerves and nerve trunks in a step-wise, asymmetrical fash-
ion, such that the neuropathy has evolved through a pattern
of overlapping multiple mononeuropathies by the time the

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 211

DOI:10.1007/978-1-84800-934-9_20, © Springer Science + Business Media B.V. 2009
212 D. A. Chad and P. Siao

patient is evaluated. In such cases, it is only by careful review “symmetrical,” “stocking-glove,” “axonal,” or the so-called
of the history and a painstaking neurological examination “length-dependent” sensorimotor polyneuropathy (Fig. 20.1).
that subtle asymmetries can be detected. These asymmetries At this stage, even a careful neurologic examination may not
suggest a stuttering, multifocal process, underlying what reveal any asymmetry of the sensorimotor deficits. Only a
appears to be a typical distal symmetrical polyneuropathy. detailed history can alert the clinician about the possibility
of vasculitic neuropathy.
Pearl: A careful, detailed history (often followed by elec-
The timing of the neurological examination and elec-
trodiagnostic studies) of the onset and progression of poly-
trodiagnostic studies (nerve conduction studies and needle
neuropathy is the are important first steps in the evaluation
electromyographic examination) in relationship to the onset
of a patient with possible vasculitic neuropathy.
and progression of the neurologic symptoms is critical. The
Comment: Mononeuropathy multiplex is the most distinc- importance of bilateral nerve conduction studies with side-
tive feature of vasculitic neuropathy. For patients with to-side comparison of sensory and motor responses cannot
known systemic vasculitides, such as polyarteritis nodosa, be overemphasized. When electrodiagnostic studies are
Churg–Strauss syndrome, Wegener’s granulomatosis, micro- performed at a stage when asymmetry is still demonstrable,
scopic polyangiitis, or cryoglobulinemic vasculitis, any neu- it supports the clinical suspicion of vasculitic neuropathy
ropathic complaints alert the clinician to consider the and helps the clinician choose the appropriate nerve for
possibility of vasculitic neuropathy. However, when evaluat- biopsy.
ing patients with no known history of systemic vasculitis or
Pearl: Vasculitic neuropathy has a predilection for some
any condition associated with vasculitic neuropathy, e.g.,
nerves or nerve distributions over others. Lower extremity
systemic lupus erythematosus, Sjögren’s syndrome, hepati-
involvement is more common than upper extremity involve-
tis C, or HIV infection, the situation is different. In this cir-
ment, and cranial nerve involvement is rare.
cumstance, a stepwise progression and accumulation of
neurologic deficits of multiple mononeuropathies may be Comment: In a series of 200 patients with necrotizing vascu-
the only clue for the clinician to suspect the presence of a litis, the peroneal division of the sciatic nerve was the most
vasculitic neuropathy. commonly affected distribution (62.5% unilaterally, 33%
The accumulation of multiple individual peripheral nerve bilaterally) (Said 1997). The posterior tibial nerve was the
deficits may coalesce into a picture of a relatively next most commonly involved nerve (27.5% unilaterally, 5%

Fig. 20.1 The figure on the left

shows the sensory deficits in the
right ulnar and left sural nerve
distributions. The middle figure
shows additional sensory deficits
in the right saphenous and left
radial nerves distributions. The
figure on the right shows the
merging or summation of sensory
deficits of different nerves giving
the picture of a distal, symmetri-
cal, “length-dependent” axonal
polyneuropathy
20 Vasculitic Neuropathy 213

bilaterally). In the upper limbs, the ulnar nerve was most Pearl: Patients diagnosed with nonsystemic vasculitic neu-
commonly involved (25.5% unilaterally and 8% bilaterally), ropathy are unlikely to experience systemic spread of the
followed by the median nerve (21.5% bilaterally, 3% unilat- vasculitic process.
erally) and the radial nerve (8% unilaterally and 2% bilater-
Comment: When patients with suspected nonsystemic vasc-
ally). Involvement of more proximal nerves was less common.
ulitic neuropathy are evaluated thoroughly and not found to
Unilateral disease of the femoral and sciatic nerves was
have symptoms, signs, or serological features of a systemic
found in only 6 and 3% of patients, respectively.
vasculitis and when immunosuppressive treatment is insti-
A review of four case series of vasculitic neuropathy
tuted at an appropriate time, there is a low risk of systemic
noted that the sural nerve was involved in 84% of patients
dissemination of the vasculitic pathology. One study of
and that the radial nerve was involved in up to 44% of cases
patients followed for more than 5 years indicated that disease
(Lacomis and Zivkovic 2007). Thus, the precise percentages
beyond the peripheral nerves and muscles evolved in only
vary from series to series but the general principles of lower
6% of patients (Collins and Periquet 2004).
extremity involvement being more common than upper
Other investigators have reported a higher conversion rate
extremity involvement and distal disease more common than
from nonsystemic to systemic disease (34%) (Said et al.
proximal are consistent.
1988). The likelihood of the emergence of systemic symp-
Cranial vasculitic neuropathies were rare, with the trigem-
toms undoubtedly varies from center to center and depends
inal and facial nerves involved in only one patient each.
to a large extent on how thorough the initial examination has
However, cranial nerve involvement may be more common
been. The principal point, however, is that a large percentage
in Wegener’s granulomatosis than in other forms of systemic
of patients (indeed, a majority or even a large majority) of
vasculitis. Nearly 20% of the patients with Wegener’s granu-
patients who have nonsystemic vasculitic neuropathy at
lomatosis had evidence of cranial neuropathies (Said 1997).
diagnosis do not develop clinical disease in other organs,
This finding may relate to the extension of inflammation
despite lengthy periods of follow-up.
from the cavernous sinus or meninges (Said 1997).
Myth: Vasculitic neuropathy is a manifestation of systemic Pearl: In the absence of trauma, the abrupt onset of pain and
necrotizing vasculitis or connective tissue disease. Thus, it paresthesias followed by weakness in the territory of a spe-
always presents in the context of clinical signs of systemic cific peripheral nerve is a clinical presentation that should
illness with constitutional symptoms (fever, anorexia, weight raise the suspicion of an emerging vasculitic neuropathy.
loss) along with laboratory abnormalities that indicate
Comment: When the involvement of one nerve is followed
immune activation.
by an evolution of several peripheral neurological deficits in
Reality: The most common presentation of a vasculitic neu- a relatively short time frame that ranges from days to months,
ropathy is in the context of a known systemic disorder such a vasculitic etiology must be considered. As described above,
as polyarteritis nodosa, an ANCA-associated disease, or a vasculitic neuropathy may evolve over a period of months
paraneoplastic syndrome (Oh 1997). In most cases, a variety and present as a polyneuropathy, but it is not unusual for it to
of laboratory abnormalities occur, including elevated acute come on abruptly and evolve rapidly, with deficits that accrue
phase reactants; the presence of antineutrophil cytoplasmic over days to weeks. This rapid onset can be associated with
antibodies, rheumatoid factor, or other autoantibodies; and multiple ischemic lesions in different nerve trunks and nerve
the finding of active hepatitis B or C infections. However, roots, leading to pronounced asymmetrical weakness. Distal
30–40% of patients presenting with a vasculitic neuropathy disease, e.g., of the feet or hands, is generally prominent, as
have isolated involvement of the peripheral nervous system is sensory loss and hyporeflexia in affected limbs.
and no sign of disease beyond peripheral nerve and muscle. Electrodiagnostic studies are consistent with axon loss
This vasculitic syndrome is designated nonsystemic vascu- and a sensorimotor polyneuropathy, characterized by asym-
litic neuropathy (Dyck et al. 1987; Burns et al. 2007). metric motor and sensory responses and normal or low-normal
Nonsystemic vasculitic neuropathy tends to run a pro- nerve conduction velocities. A given nerve in one limb might
tracted, indolent course. Its prognosis is considerably better be affected while its fellow nerve on the opposite side is rela-
than that of vasculitic neuropathy occurring in the context of tively spared.
a systemic disease. Patients with nonsystemic vasculitic neu- Other rapidly progressive neuropathies that must be con-
ropathy may have arthralgias, constitutional signs, and eleva- sidered in the differential diagnosis include immune-mediated,
tions of the acute phase reactants, but major organ involvement demyelinating neuropathies classified as Guillain–Barre syn-
of the kidneys, lungs, skin, and bowel does not occur (Kissel drome (GBS), toxic neuropathies, infectious neuropathies
et al. 1992). Hence, establishing the diagnosis of nonsystemic (e.g., West Nile virus infection). In GBS, the time-course to
vasculitic neuropathy poses a challenge to even the most the nadir of weakness is typically only days to weeks. In con-
experienced clinicians. trast to vasculitic neuropathy, the weakness is usually
214 D. A. Chad and P. Siao

symmetrical, present in distal and proximal muscle groups, abnormality and by electrodiagnostic features that point more
involves the legs and arms, and is accompanied by cranial to a lumbosacral polyradiculoplexopathy rather than an asym-
mononeuropathies. Bilateral facial nerve involvement is typi- metrical polyneuropathy. Whether this condition responds to
cal of GBS. Respiratory compromise occurs in 25% of patients immunosuppressive therapy is uncertain.
with GBS. An elevated cerebrospinal fluid protein in the
absence of a pleocytosis is a hallmark of GBS, and electrodi-
agnostic features underscore demyelination rather than axonal
20.3 Diagnosis: Electrodiagnostic Studies
loss as the predominant pathologic process.
Patients with toxic neuropathies generally have a clear cut and Biopsy
history of exposure to a potential culprit toxin, e.g., saxitoxin
from the ingestion of shell fish in red tide season or cigua- Myth: Normal sensory and motor nerve conduction stud-
toxin from the ingestion of predatory fish. Infectious neu- ies exclude vasculitic neuropathy with a high degree of
ropathies are often revealed by accompanying systemic confidence.
features of infection (for example, headache and fever in
Reality: “Normal” nerve conduction studies reduce the likeli-
West Nile virus infection).
hood of vasculitic neuropathy but certainly do not rule it out.
Myth: The patient presenting with multiple mononeuropa- The normal range of sensory and motor nerve conduction val-
thies should be considered to have a vasculitic neuropathy ues is broad. As an example, the amplitude of a normal sural
until proven otherwise. sensory potential ranges from 4 to 50 mV or higher. If one
finds a sural sensory potential amplitude of 10 mV, the value is
Reality: There is a kernel of truth to this statement. Certainly,
“normal” but should not be interpreted as such if it was 30 mV
one cannot afford to miss the diagnosis of vasculitic neu-
only 1 year before. The biopsy of a nerve that was electro-
ropathy because of its potential to progress to further nerve
physiologically normal is shown in Figs. 20.2 and 20.3.
damage or systemic disease if left untreated. The point of
Of course, most patients do not have prior nerve conduc-
this Myth, however, is that other neuropathic conditions can
tion studies for comparison, and the electromyographer must
present as multifocal peripheral nerve disorders and mimic
therefore regard the sural sensory potential of 10 mV to be
closely the clinical phenotype of a vasculitic neuropathy.
normal or, more accurately, within normal limits. In a series
A major mimicker of vasculitic neuropathy is a variant of
of 40 patients with biopsy-proven vasculitic neuropathy, 5
chronic inflammatory demyelinating polyneuropathy known
patients (12.5%) had normal nerve conduction studies of the
as “multifocal acquired demyelinating sensory and motor
nerve that was biopsied (Živković et al. 2007).
neuropathy” (Lewis et al. 1982). This disorder initially affects
Follow-up nerve conduction studies often yield impor-
the arms and later spreads to distal nerves in the legs in a man-
tant information. The first study may reveal “normal” sural
ner that conforms to discrete peripheral nerve distributions.
sensory potentials, but subsequent studies may demonstrate
The onset of this condition is insidious and slow in its pro-
significant reductions or even absence of a sensory nerve
gression (over several months). It often presents with pain and
potential.
paresthesias. The demyelinating character of this neuropathy
is revealed by electrodiagnostic studies. A nerve biopsy is Pearl: Bilateral nerve conduction studies for side-to-side
generally not necessary. The treatment response to intrave- comparison are often helpful.
nous gamma globulin and corticosteroids is generally good.
Comment: A sural sensory potential amplitude of 10 mV is
A second major mimicker of vasculitic neuropathy is
“within normal limits” but should be considered abnormal if
multifocal motor neuropathy (Pestronk 1998). This condi-
the contralateral side shows an amplitude of 20 or 30 mV. In
tion typically has only motor manifestations and tends to
general, an amplitude asymmetry of more than 50% is con-
affect the upper limb nerves (radial, median, ulnar) in an
sidered significant.
asymmetric manner. The condition evolves slowly over
The nerve conduction studies must be directed by the cli-
months to years. Electrodiagnostic testing discloses multiple
nician so that the appropriate nerves are studied. As an exam-
regions of motor nerve conduction block and points to the
ple, if the patient presents with radial sensory symptoms, the
demyelinating nature of this condition. It too has an often
finding of normal median, ulnar, superficial peroneal, and
gratifying response to intravenous immune globulin.
sural sensory potentials does not exclude a radial sensory
Finally, patients with glucose intolerance or frank diabetes
neuropathy.
can present with an asymmetric distribution of sensory loss,
pain, and weakness in the territories of the lumbosacral roots, Pearl: Examination of the flexor hallucis brevis muscle with
plexuses, and large nerve trunks of the lower extremities electromyography and nerve conduction studies is critical in
(Bradley et al. 1984). This presentation simulates vasculitic the evaluation of length-related axonopathies.
overlapping multiple mononeuropathies. The character of the Comment: A study of the flexor hallucis brevis may establish
illness is typically revealed by identifying the metabolic the presence of motor involvement that is not otherwise
20 Vasculitic Neuropathy 215

apparent. The finding of motor nerve involvement may alter Although nerve conduction studies alone may suggest a
the differential diagnosis substantially. pure sensory axonopathy (with normal motor conduction
velocities and absent sensory nerve action potentials in the
lower limbs), fibrillation potentials in both flexor hallucis
brevis muscles indicate subclinical involvement of motor
fibers. This points to a sensorimotor multiple mononeuropa-
thy, thereby broadening the differential diagnosis and plac-
ing vasculitic neuropathy front and center as a cause of the
patient’s symptoms (Vallat et al. 2007).
Pearl: It is important to understand the difference between
clinical and neurophysiologic presentations of vasculitic
neuropathy.
Comment: Multiple mononeuropathies are the most distinc-
tive clinical and neurophysiologic pattern of abnormality in
vasculitic neuropathy. However, the most common neuro-
physiologic finding is a diffuse polyneuropathy (symmetric
or asymmetric). This pattern is observed in more than half of
all cases of vasculitic neuropathy (Živković et al. 2007). In
contrast, neurophysiologic evidence of multiple mononeu-
ropathies is detected in approximately 30% of cases. The dif-
ferent neurophysiologic patterns of injury that occur in
vasculitic neuropathy are shown in Table 20.1. With time,
multiple mononeuropathies often merge into a picture of dis-
tal symmetric or asymmetric sensorimotor polyneuropathy,
as illustrated in Fig. 20.1.
Most patients with vasculitic neuropathy (90%) have sen-
sorimotor involvement, but 8% have a pure sensory involve-
ment. Two percent of patients have neurophysiologic
evidence of a demyelinating polyneuropathy, even though
the nerve biopsy confirms an axonal pattern of injury
Fig. 20.2 Low power magnification of the sural nerve. The nerve fibers
(Živković et al. 2007).
are seen in the lower right corner. The epineurial blood vessel in the
upper left corner showed marked lymphocytic infiltrates and fibrinoid Pearl: The performance of a muscle biopsy simultaneously
necrosis of the vessel wall, characteristic of vasculitic neuropathy.
Hematoxylin and eosin with a sensory nerve biopsy increases the diagnostic yield in
patients with vasculitic neuropathy.
Comment: In a group of 83 patients with vasculitic neuropathy
with or without multisystem disorders who underwent both
muscle and nerve biopsies, the muscle biopsy was diagnostic
for necrotizing arteritis in 80% and the nerve biopsy was diag-
nostic in 55% (Said et al. 1988). Even among the 32 patients
with clinical features that suggested an isolated vasculitic neu-
ropathy, 26 (81%) had necrotizing arteritis demonstrated within
their muscle biopsy. Only 13 patients (41%) showed

Table 20.1 Neurophysiologic patterns of injury in vasculitic neuropathy

Diffuse polyneuropathy (symmetrical or asymmetrical)
Mononeuritis multiplex
Mononeuropathies
Brachial plexopathy
Multilevel lumbosacral radiculopathy
Fig. 20.3 High power magnification of an epineurial blood vessel with Abnormalities of the upper extremity nerves that are out of
fibrinoid necrosis of the vessel wall and intense inflammatory infiltrates proportion to the lower extremity nerves
characteristic of vasculitic neuropathy. Hematoxylin and eosin Disproportionate abnormality of two neighboring nerves
216 D. A. Chad and P. Siao

necrotizing arteritis in the nerve specimen. Necrotizing arteritis The blood supply of the peripheral nerves is profuse, derived
in both muscle and nerve occurred in seven patients (22%). from a plethora of nutrient arteries that penetrate the nerve
Biopsy of the superficial peroneal sensory nerve has the from nearby arterial trunks. Vascular branches divide and sub-
advantage of its proximity with the peroneus brevis muscle. divide to establish an anastamosing vascular network in the
The surgeon can biopsy the nerve and muscle with a single epineurium and perineurium – a reticular endoneurial vascular
incision. Otherwise, the patient may need one incision for plexus with frequent interconnections. In the face of this richly
the sural nerve and another incision for the gastrocnemius collateralized blood supply, nerve ischemia is difficult to pro-
muscle biopsy. duce. Thus, a key prerequisite for nerve ischemia is a disease
process such as a vasculitis that targets and damages (in an
Myth: In the evaluation of a patient with possible vasculitic
extensive manner) blood vessels that range in diameter from
neuropathy, the finding of an absent sural sensory potential
50 to 300 mm. It is blood vessels of this caliber that constitute
should direct the clinician to biopsy another nerve.
the vasa nervorum.
Reality: When choosing the nerve for biopsy, the finding of The most vulnerable portion of the peripheral nerves is at
an absent sural sensory potential does not mean that sural proximal locations – the mid-humerus and mid-thigh – where
sensory nerve biopsy would not reveal any useful informa- the nerve trunks lie between the distributions of major nutri-
tion, especially if the symptoms are recent or subacute. ent arteries (i.e., in watershed areas). The nerve fascicles that
Pathologic confirmation of vasculitic neuropathy is very comprise the nerve trunks and contain motor and sensory
important prior to embarking on long-term immunosuppres- fibers are most vulnerable in their central regions (Dyck et al.
sant or immunomodulating treatments. Selection of the opti- 1972). Hence, axon loss is most notable in what has been
mal sensory nerve or sensory nerve/muscle combination to called a central-fascicular location.
target is therefore critical to establishing the diagnosis. Motor Because sural nerve biopsies are far downstream from the
nerves are not biopsied because muscle weakness would be site of the axonal degeneration lesion and because of branch-
the inevitable consequence of the biopsy. Biopsies of nerve ing and inter-mixing of nerve fibers that occurs between fas-
fascicles rather than the entire nerve (so as to avoid signifi- cicles as they descend through the nerves of the limbs, a
cant weakness or sensory loss from the procedure) are pos- central-fascicular pattern of degeneration is seldom observed
sible, but this technique is generally available only in a few in sural nerve biopsies. Rather, the appearance of the fasci-
academic centers. cles at the level of the sural nerve that often emerges is one
In practice, one has relatively few candidate sensory of preferential fascicular involvement (see below).
nerves from which to choose. The patient will have residual
Pearl: A multifocal pattern of axon loss in a nerve biopsy,
sensory loss in the distribution of the biopsied sensory nerve.
characterized by the selective depletion of myelinated nerve
The usual target selected in the lower extremities is either the
fibers in some nerve fascicles more than others, is strongly
sural or the superficial peroneal sensory nerve. In the upper
suggestive of a vasculitic neuropathic process.
extremities, one may choose the superficial radial sensory
nerve or the lateral and medial antebrachial cutaneous sen- Comment: The nerve fiber pathology in vasculitic neuropa-
sory nerves. The residual sensory loss in the upper extremi- thy is characterized by axonal degeneration that appears as
ties may be more bothersome to patients than the residual myelin ovoids in longitudinal sections (representing the dis-
sensory loss in the lower extremity. Patients should be warned integration of the axons and their myelin sheaths). The loss
about the possibility of pain syndromes or neuroma forma- of myelinated fibers sometimes involves all of the nerve fas-
tion after nerve biopsy. cicles within a biopsy specimen. More often, however, a pro-
portion of the total number of nerve fascicles is much more
depopulated of myelinated fibers than others, leading to a
20.4 Pathophysiology pattern of a patchy, asymmetrical nerve fiber loss among the
fascicles (Kissel and Mendell 1992; Said 1997; Lacomis and
Zivkovic 2007). This type of pattern, termed preferential fas-
Pearl: The tissue targeted by the pathological process of vas-
cicular involvement, is typical of vasculitic neuropathy.
culitic neuropathy is the neural vasculature. Nerve damage
and axonal degeneration are mediated by ischemia. Myth: The diagnosis of vasculitic neuropathy is established
by the finding of perivascular inflammatory infiltrates involv-
Comment: The vasa nervorum provide conduits for nourishing
ing epineurial blood vessels.
peripheral nerve tissue. When its elements are damaged, the
stage is set for nerve ischemia and ultimately the clinical mani- Reality: The brunt of damage in vasculitic neuropathy stems
festations of peripheral neuropathy. In vasculitic neuropathy, from the involvement of muscular-walled arteries within the
the predominant nerve pathology is axonal degeneration, with epineurium that constitute the vasa nervorum. However,
the process of segmental demyelination much less prevalent. perivascular inflammation alone is not sufficient to support a
20 Vasculitic Neuropathy 217

definitive diagnosis of vasculitis. Rather, the characteristic

and diagnostic findings include necrosis of all components
of the vessel wall (transmural necrosis), often with the depo-
sition of fibrin (fibrinoid change), thrombotic occlusion of
the lumen, and infiltration of the vessel wall by inflammatory
cells. The inflammatory cells are principally polymorphonu-
clear leukocytes, but eosinophils and mononuclear cells also
occur occasionally.
Pearl: Electrodiagnostic studies (nerve conduction and elec-
tromyographic studies) play an important role in the diagno-
sis of vasculitic neuropathy.
Comment: The clinical presentation, physical examination
findings, and laboratory results often suggest the possibility
of a vasculitic neuropathy, and the analysis of a nerve and
muscle biopsy specimens can provide confirmation of that Fig. 20.4 Vasculitis within skeletal muscle. Intramuscular, medium-
diagnosis. Electrodiagnostic studies also have a major place sized artery from a patient with a overlapping multiple mononeuroapa-
thies, showing severe transmural mononuclear cell inflammation and
in the evaluation of patients with possible vasculitic neuropa-
fibrinoid necrosis, indicative of vasculitis. Surrounding muscle fibers,
thy, and fulfill two main functions. First, electrodiagnostic seen in cross section, are angulated and atrophic consistent with the
studies are a critical intermediate step between the clinical process of neurogenic atrophy. [Courtesy of Professor Thomas W.
evaluation and tissue biopsy, serving to support the diagnosis Smith, University of Massachusetts, Worcester, MA]
of vasculitic neuropathy or to cast doubt on that clinical
impression before a nerve/muscle biopsy is undertaken. conditions are met, the likelihood that a nerve biopsy speci-
Second, electrodiagnostic studies help select the most appro- men will reveal diagnostic histopathology is only approxi-
priate nerve for biopsy. mately 60% (Burns et al. 2007).
Electrodiagnostic studies can confirm the clinical suspi- Several studies have shown that the addition of a muscle
cion that the patient’s weakness is secondary to axon loss biopsy (Fig. 20.4) increases the chance of visualizing charac-
rather than to demyelination. (Demyelination is a relatively teristic arterial lesions (Said et al. 1988; Vital et al. 2006).
minor component of vasculitic neuropathy.) In addition, these When the superficial peroneal nerve is deemed an acceptable
investigations define the pattern of nerve involvement – for candidate for biopsy, the subjacent peroneus brevis muscle
example, discerning a multifocal, asymmetric process by should also be sampled because no additional incision is
identifying examining side-to-side differences in the motor required. When the sural nerve is selected, a second incision
and sensory responses. should be performed to obtain a muscle biopsy from either
The most commonly biopsied nerves are the sural and the gastrocnemius or the vastus lateralis.
superficial peroneal nerves. The diagnostic yield of nerve
biopsy is increased enormously when electrodiagnostic stud-
ies demonstrate abnormalities of the target nerve (Wees et al. 20.5 Treatment
1981; Kissel and Mendell 1992). The cutaneous branch of
the radial nerve is also a candidate nerve to biopsy for histo-
Pearl: The combination of cyclophosphamide and glucocor-
logical confirmation of the diagnosis, particularly if the neu-
ticoids is more effective in the treatment of vasculitic neu-
ropathic manifestations occur prominently in the upper limbs
ropathy that occurs in the context of systemic vasculitis than
(Kissel and Mendell 1992).
is treatment with glucocorticoids alone.
Pearl: The combination of nerve and muscle biopsy increases
Comment: There are no controlled treatment trials for the
the chances of identifying vascular histopathology diagnos-
management of vasculitic neuropathy. However, published
tic of a vasculitic neuropathy.
clinical experience suggests strongly that more than two-thirds
Comment: In most cases of suspected vasculitic neuropathy, of patients with systemic vasculitic syndromes treated early in
either the sural nerve or the superficial peroneal nerve is their clinical course with cyclophosphamide in addition to glu-
deemed suitable for biopsy. A positive biopsy is more likely if cocorticoids have favorable outcomes. Patients treated with
the nerve biopsied is the one from the more symptomatic side, cyclophosphamide and glucocorticoids have a greater degree
if the area of skin supplied by the nerve in question is of improvement in their neuropathic symptoms and a reduced
hypesthetic, and if the nerve is found to be compromised by rate of relapse compared with those who receive glucocorti-
electrodiagnostic testing. However, even when these coids alone (Kissel and Mendell 1992; Mathew et al. 2007).
218 D. A. Chad and P. Siao

For patients with nonsystemic vasculitic neuropathy, no Dyck PJ, Benstead TJ, Conn DL, et al Nonsystemic vasculitic neuropa-
adequate clinical studies exist to serve as a guide to phar- thy. Brain 1987;110:843–54
Dyck PJ, Conn DL, Okazaki H. Necrotizing angiopathic neuropathy:
macotherapy (Vrancken et al. 2007). Consequently, the man- three dimensional morphology of fiber degeneration related to sites
agement of these patients is controversial (Collins and of occluded vessels. Mayo Clin Proc. 1972;47:461–75
Periquet 2004; Kissel and Mendell 1992; Burns et al. 2007). Kissel JT, Mendell JR. Vasculitic neuropathy. Neurol Clin. 1992;10:
Methotrexate and azathioprine have both been proposed as 761–81
Lacomis D, Giuliani MJ, Steen V, et al Small fiber neuropathy and vas-
alternatives to cyclophosphamide, but there are insufficient culitis. Arthritis Rheum. 1997;40:1173–7
data on which to judge this recommendation. Lacomis D, Zivkovic SA. Approach to vasculitic neuropathies. J Clin
For patients with hepatitis C-associated nonsystemic vas- Neuromusc Dis. 2007;9:265–76
culitic neuropathy, antiviral therapy might increase the likeli- Lewis RA, Sumner AJ, Brown MJ, Asbury AK. Multifocal demyelinat-
ing neuropathy with persistent conduction block. Neurology 1982;
hood of a favorable outcome for peripheral nerve recovery 32:958–64
and lessens the risk of cirrhosis (David et al. 1996). Mathew L, Talbot K, Love S, et al Treatment of vasculitic peripheral
neuropathy: a retrospective analysis of outcome. QJM. 2007; 100:41
Myth: The development of a new nerve infarction dictates the Oh SJ. Paraneoplastic vasculitis of the peripheral nervous system.
need for intensification of therapy in the treatment of vascu- Neurol Clin. 1997;15:849–63
litic neuropathy. Pestronk A. Multifocal motor neuropathy: diagnosis and treatment.
Neurology 1998;51:S22–24
Reality: The response to a new nerve deficit all depends on Said G, Lacroix-Ciaudo C, Fujimura H, Blas C, Faux N. The peripheral
the context. New nerve infarctions can occur in distributions neuropathy of necrotizing arteritis: a clinicopathological study. Ann
Neurol. 1988;23:461–5
of extensive vascular compromise even up to several weeks Said G. Necrotizing peripheral nerve vasculitis. Neurol Clin. 1997; 15:
after the institution of high-dose glucocorticoids and cyclo- 835–48
phosphamide. The development of a new nerve infarction Seo J-H, Ryan HF, Claussen GC, Thomas TD, Oh SJ. Sensory neuropa-
shortly after starting immunosuppression does not necessar- thy in vasculitis: a clinical, pathological, and electrophysiologic
study. Neurology 2004;63:874–8
ily signal the need to intensify treatment, particularly if other Schaumburg HH, Berger AR, Thomas PK. Disorders of peripheral
indicators of disease activity suggest disease control. nerves (Contemporary Neurology Series 36). 2nd ed. Philadelphia:
FA Davis; 1992. p. 130–5
Vallat J-M, Cros DP, Hedley-White ET. Case 9-2007: a 27 year-old
References woman with pain and swelling of the legs. N Engl J Med. 2007;
356:1252–9
Vital C, Vital A, Canron MH, Jaffre A, Viallard JF, Ragnaud JM,
Bradley WG, Chad D, Verghese JP, Liu HC, Good P, Gabbai AA, Brechenmacher C, Lagueny A. Combined nerve and muscle
Adelman LS. Painful lumbosacral plexopathy with elevated eryth- biopsy in the diagnosis of vasculitic neuropathy. A 16-year retro-
rocyte sedimentation rate: a treatable inflammatory syndrome. Ann spective study of 202 cases. J Peripher Nerv Syst. 2006;11 :20–9
Neurol. 1984;15:457–64 Vrancken AF, Hughes RA, Said G, et al Immunosuppressive treatment
Burns TM, Schaublin GA, Dyck PJ. Vasculitic neuropathies. Neurol for non-systemic vasculitic neuropathy. Cochrane Database Syst
Clin. 2007;25:89–113 Rev. 2007;24(1):CD006050
Collins MP, Periquet MI. Non-systemic vasculitic neuropathy. Curr Wees SJ, Sunwoo IN, Oh SJ. Sural nerve biopsy in systemic necrotizing
Opin Neurol. 2004;17:587–98 vasculitis. Am J Med. 1981;71:525–32
David WS, Peine C, Schlesinger P, Smith SA. Nonsystemic vasculitic Živković SA, Ascherman D, Lacomis D. Vasculitic neuropathy: elec-
mononeuropathy multiplex, cryoglobulinemia, and hepatitis C. trodiagnostic findings and association with malignancies. Acta
Muscle Nerve. 1996;19:1596–602 Neurol Scand. 2007:115;432–6
 Pediatric Vasculitis
21
Anne H. Rowley, Seza Ozen, Robert P. Sundel, and Frank T. Saulsbury

Overview › One-third or more of children with PAN have the cuta-

neous form, largely limited to skin and joint involve-
ment. While mortality is rare, aggressive control of the
› Based on surverys of providers, vasculitis in children
systemic inflammation and end-organ involvement is
appears to have an incidence of about 50 cases per
nonetheless needed if significant long-term morbidity
100,000 children per year. Only HSP and KD are more
is to be avoided.
common in children than in adults, with each one
affecting at least 10-20/100,000 pediatric patients per
year. Takayasu arteritis and ANCA-positive vasculiti-
des, on the other hand, strike approximately one child
in a million each year.
› Differences between adult and pediatric vasculitides 21.1 Kawasaki Disease
are generally subtle, though they may have important
implications. Since these conditions are so rare in the
Myth: Kawasaki’s disease is a rare illness.
pediatric age group, diagnosis may be significantly
delayed, leading to more disease-related damage at the Reality: Kawasaki’s disease is a very common illness in Japan.
time therapy is initiated. In addition, unlike adults, Approximately 1 in 150 Japanese children develop Kawasaki’s
children may suffer irreversible effects on their physi- disease by the age of 5 years incidence of 175,100,00 children
cal and mental development both from the disease and (Yanagawa et al. 2001). The incidence is about tenfold lower
from the medications used to treat it. among Caucasian children (15–20/100,000 children <5 years)
› On the other hand, children tend to have fewer comor- (Holman et al. 2003) and is intermediate among African-
bidities and they are therefore more tolerant of poten- Americans. The incidence also varies by age and gender
tially toxic therapies. This often results in better (Table 21.1). Thus, the incidence of Kawasaki’s disease in
outcomes for children than adults, especially in condi- U.S. children is slightly higher than the incidence of invasive
tions like Takayasu arteritis and PAN. Haemophilus influenzae type b (Hib) in that population prior
› Kawasaki disease (KD) has replaced acute rheumatic to the introduction of Hib conjugate vaccine (13.9/100,000
fever as the leading cause of acquired heart disease children <5 years) (Centers for Disease Control 1998). No one
among children in developed nations. Since IVIG considered Hib to be a “rare” condition in those days.
became the standard treatment for KD in 1986, the
Pearl: Neck swelling is the most dramatic clinical symptom
mortality rate of the disease has decreased from 1-2%
in some children with Kawasaki’s disease.
to <0.1%, and the incidence of coronary artery aneu-
rysms has decreased by 75%. Comment: A diagnosis of Kawasaki’s disease should be con-
› Henoch-Schonlein purpura, also known as anaphylac- sidered in any child with prolonged fever and neck swelling
tic purpura, is a benign condition in the vast majority that does not respond to antibiotics. Other Kawasaki’s dis-
of affected children. Nonetheless, in rare cases, chil- ease features can remain unrecognized for days after neck
dren affected by HSP may develop hypertension (espe- swelling develops, or begin afterward. Neck swelling in
cially during pregnancy) or renal failure long after the Kawasaki’s disease can represent superficial adenitis or,
acute syndrome has resolved. more rarely, inflammation within the deeper tissue planes of
the neck, including the retropharyngeal or parapharyngeal
spaces (Stamos et al. 1994). Despite the intensity of inflam-
mation in these spaces and the decreased attenuation that can

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 219

DOI:10.1007/978-1-84800-934-9_21, © Springer Science + Business Media B.V. 2009
220 A. H. Rowley et al.

Table 21.1 Approximate incidence of Kawasaki’s disease in U.S. Infections must be considered first in any febrile infant.
and Japanese population by agea (adapted from Holman (2003) and However, if no organism has been identified after a week of
Yanagawa (2001) )
fever, Kawasaki’s disease should receive serious consider-
Age U.S. Japan
ation. The disease in infants tends to be both less typical,
Male Female Male Female
often without other signs of mucocutaneous inflammation,
<1 year 25 14 225 140 and more morbid, causing aneurysms in up to 35% of patients
1–4 years 20 13 100 75
despite treatment with IVIG.
>4 years 1–4 1–4 10 10
a
The need for a heightened concern for Kawasaki’s disease in
Per 100,000 children
young babies is reflected in recommendations that children under
6 months of age with at least 7 days of fever receive an echocar-
be observed on CT scan (suggesting an abscess), true diogram, even if other classic manifestations of Kawasaki’s dis-
abscesses do not develop in these areas. Surgical drainage ease are absent (Newburger et al. 2004a).
does not yield purulent material and is generally not indi-
cated, as most patients improve dramatically with intrave- Myth: Any patient who develops Kawasaki’s disease should
nous gammaglobulin (IVIG) therapy. have periodic echocardiograms performed for life.
Reality: Kawasaki’s disease is an acute vasculitis; coronary
Myth: A positive rapid test for respiratory syncytial virus
artery abnormalities typically are visible by echocardiogra-
(RSV) or parainfluenza virus (PIV) from the nasopharynx
phy at 3–4 weeks after the onset of fever. There is no con-
excludes Kawasaki’s disease.
vincing evidence that coronary artery disease will develop in
Reality: A statement included in classic diagnostic criteria a Kawasaki’s disease patient if it has not developed within
for Kawasaki’s disease is that “the illness cannot be explained the first 2 months after the onset of Kawasaki’s disease.
by a known disease process.” Some clinicians have taken this A patient who develops coronary artery abnormalities
to mean that a positive test for a known pathogen from any needs to be followed with periodic echocardiograms. Other
site rules out the diagnosis of Kawasaki’s disease. This is studies may also be indicated, depending upon the severity
only true if that pathogen is known to produce clinical find- of the coronary artery disease (Newburger et al. 2004a).
ings that truly mimic Kawasaki’s disease. However, the child who has normal echocardiograms at
Because RSV and PIV produce respiratory symptoms but diagnosis, 2–3 weeks after onset, and 6–8 weeks after onset
not the other clinical features of Kawasaki’s disease, they are generally does not require additional echocardiograms.
unlikely to be responsible for an illness compatible with
Kawasaki’s disease. Because Kawasaki’s disease cases occur Pearl: In a toddler with some clinical features of Kawasaki’s
with greater frequency in winter–spring in nontropical cli- disease, refusal to walk should heighten suspicion of the
mates, the same time that RSV and some strains of PIV are diagnosis.
circulating, Kawasaki’s disease patients will occasionally be Comment: Conditions that mimic Kawasaki’s disease, such
coinfected with RSV, PIV, or other respiratory pathogens. Such as adenovirus or enterovirus infections or scarlet fever, gen-
patients should be promptly treated for Kawasaki’s disease. erally do not manifest this sign. Children with Kawasaki’s
disease frequently develop small-joint arthritis in the hands
Pearl: Conjunctival injection can be the most persistent
and feet, which is painful and limits their willingness to walk
Kawasaki’s disease manifestation, persisting long after IVIG
or to pick up small objects. Swelling of the hands and feet is
therapy has led to the improvement of other symptoms and
also common in Kawasaki’s disease but rarely observed in
signs.
other illnesses that comprise the usual differential diagnosis
Comment: Persistent conjunctival injection can be observed (excluding measles).
in Kawasaki’s disease patients who have experienced resolu-
Myth: A normal ESR rules out the diagnosis of acute
tion of fever and other clinical signs following the institution
Kawasaki’s disease.
of IVIG and aspirin therapy. If the patient is otherwise doing
well, persistent conjunctival injection does not necessarily Reality: The ESR is influenced by numerous factors, and can
indicate treatment failure. be falsely lowered by a very high white blood cell count,
hypofibrinogenemia, or other factors (Jurado 2001). The
Pearl: Kawasaki’s disease is a particularly difficult diagno-
CRP is not affected by these factors, and should therefore be
sis to make in young babies.
determined in a patient presenting with possible acute
Comment: Young babies seldom get autoimmune or idio- Kawasaki’s disease. Discrepancy between the ESR and CRP
pathic inflammatory disorders, perhaps because their immune in acute Kawasaki’s disease has been reported (Anderson et
systems remain quite immature for the first several months of al. 2001). CRP normalizes faster than the ESR following
life. Kawasaki’s disease is an exception to this rule. acute Kawasaki’s disease.
21 Pediatric Vasculitis 221

The ESR can be particularly useful in evaluating a child A child with Kawasaki’s disease and chickenpox or a vari-
for possible “missed” Kawasaki’s disease between the sec- cella-susceptible Kawasaki’s disease patient with a significant
ond and fourth weeks after fever onset. During that time exposure should discontinue aspirin, and another antiplatelet
period following acute Kawasaki’s disease, the ESR would agent be administered until varicella has resolved or the incu-
likely remain elevated while the CRP may have normalized. bation period has passed. Clopidogrel offers an option for
antiplatelet therapy, especially in a child deemed to be at par-
Pearl: The ESR is not useful in monitoring clinical progress
ticularly high risk for thrombosis (Newburger et al. 2004a).
for at least 7–10 days after IVIG therapy.
Myth: If a child with suspected Kawasaki’s disease is not
Comment: The ESR is dependent upon the levels of several
cranky, he probably does not have Kawasaki’s disease.
plasma proteins, including IgG (van Leeuwen and van
Rijswijk 1994). Thus, the ESR rises transiently following Reality: The cause of irritability in children with Kawasaki’s
IVIG infusion even as the disease begins to come under con- disease is not known. Many clinicians suspect subclinical
trol in most patients. In contrast to the ESR, C-reactive pro- meningoencephalitis and, indeed, many children with Kawasaki’s
tein (CRP) levels are useful in monitoring disease responses disease are found to have abnormalities of the CSF when a lumbar
to IVIG, because the CRP level is not altered directly by the puncture is performed (Dengler et al. 1998).
constituents of this therapy. Crankiness is typical of children with Kawasaki’s disease but
is not present invariably. Older children, for example, are more
Myth: A child who develops an illness with some features of likely to be somnolent or lethargic than irritable, but even young
Kawasaki’s disease in association with a finding not compatible children are not always cranky. Thus, if Kawasaki’s disease is
with Kawasaki’s disease has “atypical Kawasaki’s disease.” suspected because of certain clinical features, the presence or
Reality: A patient with prolonged fever, red lips, rash, and a the absence of crankiness should not influence clinical thinking
lobar pneumonia is much more likely to have bacterial infec- significantly one way or the other.
tion than Kawasaki’s disease, because lobar pneumonia is Pearl: A diagnosis of Kawasaki’s disease is unlikely in chil-
not characteristic of acute Kawasaki’s disease. The terms dren with splenomegaly, discrete intraoral lesions, or a lym-
“atypical” and “incomplete” Kawasaki’s disease refer to ill- phocytic predominance in the peripheral blood.
nesses in which the patient has some but not all features of
classic Kawasaki’s disease. They were not intended to Comment: Differentiation of Kawasaki’s disease from viral
describe patients whose clinical features are inconsistent conditions is often difficult. Indeed, the manifestations of
with that diagnosis. In such patients, additional testing to atypical measles (i.e., measles following vaccination against
determine the correct diagnosis should be performed. measles) can be clinically indistinguishable from Kawasaki’s
disease. Similarly, EBV, adenovirus, parvovirus, and other
Myth: Dilation of coronary arteries in a febrile child is pre- conditions can mimic Kawasaki’s disease. Splenomegaly,
sumptive evidence of a diagnosis of Kawasaki’s disease. discrete oral lesions (as opposed to diffuse oral erythema),
Reality: Coronary artery ectasia and aneurysmal dilation and a lymphocytic predominance in the peripheral blood all
can be caused by a variety of illnesses. Kawasaki’s disease suggest one of the viral mimics of Kawasaki’s disease rather
is the most widely recognized of these conditions and the than Kawasaki’s (Burns et al. 1991).
one that is most important to diagnose acutely to improve Pearl: Kawasaki’s disease should be considered in the dif-
outcomes. However, systemic-onset JRA (Still’s disease), ferential diagnosis of prolonged fever and aseptic meningitis
systemic lupus erythematosus, polyarteritis nodosa, and of unknown etiology in infants and in young children.
EBV infection should also be considered in the differen-
Comment: Infants present with incomplete Kawasaki’s dis-
tial diagnosis of a child with dilated coronary arteries
ease more commonly than do older children. In an infant or
(Binstadt et al. 2005).
young child with prolonged fever and aseptic meningitis, the
Pearl: A child with Kawasaki’s disease who develops influ- diagnosis of Kawasaki’s disease should be considered if
enza or chickenpox should discontinue aspirin. enterovirus and herpes simplex virus infections have been
Comment: The discontinuation of aspirin in the setting of an excluded (Dengler et al. 1998). Aseptic meningitis is occa-
influenza or varicella infection is essential to reduce the pos- sionally the most significant clinical manifestation of the ill-
sibility of Reye syndrome. Children with Kawasaki’s disease ness, but other findings potentially related to Kawasaki’s
on aspirin therapy should be immunized with the inactivated disease must be sought carefully.
influenza vaccine. There is considerable overlap between Pearl: Peribronchial infiltrates and mild interstitial infil-
influenza season and the peak of acute Kawasaki’s disease trates can be observed on chest radiographs in children with
incidence in nontropical climates (winter–spring). As a result, acute Kawasaki’s disease and should not dissuade the clini-
patients with both conditions are observed on occasion. cian from that possible diagnosis.
222 A. H. Rowley et al.

Comment: Peribronchial and interstitial infiltrates are more Reality: Kawasaki’s disease is uncommon but does occur in
common in Kawasaki’s disease than generally appreciated, this age group (Rosenfeld et al. 1995; Burns et al. 1986).
because many Kawasaki’s disease patients do not have chest Infants with acute Kawasaki’s disease have prolonged fever,
radiographs performed. One study reported radiographic but other clinical findings of Kawasaki’s disease can be absent
abnormalities in 15% of acute Kawasaki’s disease patients or subtle. Unfortunately, young infants can have severe coro-
(Umezawa et al. 1989). These consisted primarily of reticu- nary artery sequelae following acute Kawasaki’s disease.
logranular infiltrates, peribronchial cuffing, pleural effusion,
Myth: Kawasaki Disease is not observed in older children
and atelectasis. Autopsy studies of acute Kawasaki’s disease
and adolescents.
patients reveal that pneumonitis and bronchial inflammation
are observed in the majority of fatal acute Kawasaki’s dis- Reality: Kawasaki’s disease can occur in older children and
ease cases (Amano et al. 1980). teenagers, and these patients can be at risk for developing
coronary artery disease. Older children are likely to experi-
Pearl: In a child with prolonged fever, the presence of tachy-
ence a delay in diagnosis (Stockheim et al. 2000; Momenah
cardia even when fever has temporarily abated should
et al. 1998; Muta et al. 2004). A high index of suspicion is
heighten suspicion of acute Kawasaki’s disease.
needed to make the diagnosis.
Comment: Because myocarditis is present to some degree in
the majority of acute Kawasaki’s disease patients, the pres- Pearl: The clinical features of Kawasaki’s disease can
ence of tachycardia even when fever is not present is com- appear discontinuously (i.e., all may not be present at the
mon, and may provide a clue to the diagnosis. Myocarditis same time).
and its associated cardiac dysfunction improve following Comment: Obtaining a complete history of all physical find-
IVIG therapy (Newburger et al. 1989). ings from the family is often critical in establishing the diag-
Myth: If the platelet count is not elevated and there is no nosis of Kawasaki’s disease. A history of a maculopapular
peeling of the fingers and toes, then acute Kawasaki’s dis- rash that lasted for more than a day or of swollen hands and
ease is unlikely. feet that lasted several days but have since resolved could be
helpful clues to establishing a diagnosis of Kawasaki’s dis-
Reality: Thrombocytosis and peeling of the hands and feet ease. Consensus criteria for the classification of Kawasaki’s
are characteristic features of the subacute phase of Kawasaki’s disease are shown in Table 21.2.
disease. They usually occur 2–3 weeks after the onset of
fever, and are therefore not useful for the diagnosis of acute Myth: Kawasaki’s disease and infantile periarteritis nodosa
Kawasaki’s disease. The optimal approach to caring for a are two different illnesses.
patient with Kawasaki’s disease is to make the diagnosis and Reality: Children in the U.S. and throughout the world die of
begin treatment with IVIG before either thrombocytosis or unrecognized acute Kawasaki’s disease. In the years before
peeling has developed. Treatment does not affect the occur- Dr. Kawasaki reported the characteristic clinical features of
rence of either finding. the illness, the only cases of Kawasaki’s disease reported in
Thrombocytopenia during the acute phase of Kawasaki’s the literature were the fatal ones. Pathologists used the term
disease, although uncommon, is a well-recognized risk fac-
“infantile periarteritis nodosa” to describe these cases.
tor for the development of coronary artery disease as a com- Studies comparing the pathologic features of acute fatal
plication of the illness (Niwa et al. 1995). Kawasaki’s disease cases and cases of infantile periarteritis
Pearl: Erythema in the groin area might be the result of a
Candida infection – but also possibly a manifestation of
Kawasaki’s disease. Table 21.2 Criteria for the classification of Kawasaki’s disease
(modified from Ozen 2006)
Comment: When prolonged fever accompanies groin ery-
thema and peeling, a simple cutaneous fungal infection may Mandatory criterion: fever persisting for at least 5 days
not be the true diagnosis. Clinicians sometimes miss a case Additional criteria (four of five features required)
of incomplete Kawasaki’s disease in an infant or young child Skin changes in the peripheral extremities or perineal region
Polymorphous exanthema
when a groin rash is mistakenly diagnosed as candidal der-
Bilateral conjunctival injection
matitis. Knowledge of the many clinical presentations of Changes of lips and oral cavity, with injection of oral and
acute Kawasaki’s disease, combined with a high index of pharyngeal mucosa
suspicion, is needed to make the diagnosis and prevent the Cervical lymphadenopathy
potentially severe coronary artery sequelae of the illness. In the presence of coronary artery involvement (detected onechocar-
diography) and fever, fewer than four of the remaining five criteria are
Myth: Kawasaki’s disease does not occur in infants younger sufficient (the exact number of criteria required is to be defined in an
than 6 months of age. ongoing study)
21 Pediatric Vasculitis 223

hyper-IgD syndrome, another hereditary periodic fever syn-

drome (Wickiser et al. 2005).
Myth: A skin biopsy is necessary for the diagnosis of HSP.
Reality: The vast majority of patients with HSP can be diag-
nosed on clinical grounds alone (Ozen et al. 2007). The typi-
cal rash of HSP consists of palpable purpuric lesions 2–10 mm
in diameter. The lesions are concentrated on the buttocks and
lower extremities, but not confined exclusively to those areas
(Fig. 21.1). A skin biopsy is helpful if the rash is atypical.
For example, only 2% of children with HSP have bullous
lesions; thus, a biopsy should be considered if bullous lesions
are present. A biopsy should be considered if there is evi-
dence of tissue necrosis, since this complication is extremely
uncommon in children with HSP.
A skin biopsy should also be considered if the patient has
Fig. 21.1 Typical rash of HSP concentrated on buttocks and lower unusual or atypical symptoms, such as nasal or sinus inflam-
extremities (Figure courtesy of Dr. Frank Saulsbury)
mation or pulmonary disease, all of which suggest other dis-
orders. Of course, if a biopsy is performed, the point is not to
demonstrate leukocytoclastic vasculitis (this could be
nodosa indicate that these two conditions are, in fact, one inferred from the finding of palpable purpura). Rather, the
(Landing and Larson 1977; Tanaka et al. 1976). goal of the biopsy is to either demonstrate IgA deposition in
small blood vessels or to exclude the finding. Although HSP
Myth: Kawasaki’s disease only strikes children who were
is just one cause of small vessel leukocytoclastic vasculitis, it
previously healthy.
is the only form of vasculitis that is always associated with
Reality: The vast majority of children with acute Kawasaki’s granular IgA deposition in vessel walls.
disease are previously healthy. However, the illness can occur Because 80% of HSP cases occur in children younger
in children with underlying medical conditions. Kawasaki’s than 10 years of age, a skin biopsy should be considered in
disease may not be recognized in these children, because the teenagers, particularly those in their late teen years. Finally,
condition is not even considered in the differential diagnosis. a skin biopsy should be considered in patients with high
One example is a patient who developed fever, conjunctival fever or prolonged fever.
injection, and rash 1 year following an autologous stem cell
Pearl: Joint symptoms precede the onset of purpura in up to
transplantation for another illness. Adenovirus infection was
one quarter of patients with HSP.
the leading diagnosis, but the virus was not identified by cul-
ture from any site. Fevers eventually resolved, but the patient Comment: Among HSP patients whose presentation consists of
presented 3 months later with a myocardial infarction, and joint symptoms alone, purpura – a sine qua non of HSP – almost
had aneurysms of multiple coronary arteries. always appears within 1 week (Saulsbury 1999; Trapani et al.
2005). In general, cutaneous purpura develops before the results
of the laboratory studies that were ordered when the patient pre-
21.2 Henoch–Schönlein Purpura sented with joint complaints. The joint swelling associated with
HSP may not represent a true synovitis, but rather periarticular
soft tissue edema. The swelling often extends beyond the joint
Myth: Henoch–Schönlein purpura (HSP) is the only small
boundaries (e.g., to include the dorsum of the feet in addition to
vessel vasculitis that affects children.
the ankles, or the dorsum of the hand in addition to the wrists)
Reality: HSP is the most common form of small-vessel vas- (Fig. 21.2). The joint involvement of HSP always involves the
culitis in children, but not the only one. Other vasculitides, lower extremities (knees, ankles, or feet), but one half of the
including infection-induced vasculitis, hypersensitivity vas- patients also have upper extremity involvement, usually the
culitis, microscopic polyangiitis, and Wegener’s granuloma- wrists and hands (Saulsbury 1999; Trapani et al. 2005).
tosis, can cause a similar clinical picture (Ozen et al. 2007).
Pearl: Gastrointestinal symptoms precede the onset of pur-
In addition, HSP or HSP-like vasculitis can occur as a part of
pura in up to 20% of patients.
other conditions. For example, HSP occurs in 5–10% of
patients with familial Mediterranean fever (Ozdogan et al. Comment: Colicky abdominal pain, vomiting, and gastroin-
1997). It has also been reported in some children with the testinal bleeding precede the onset of purpura by up to 2 weeks
224 A. H. Rowley et al.

drugs with gastrointestinal and renal side effects to treat a

disease in which the major complications are gastrointestinal
bleeding and nephritis.
Myth: Early administration of glucorticoids prevents the
development of nephritis.
Reality: Approximately 40% of children with HSP develop
nephritis. The majority of patients develop nephritis within
the first 4 weeks of the illness, but the onset of nephritis is
delayed for up to 3 months in 25% of patients. There is little
evidence that initiation of glucocorticoid therapy at the onset
of symptoms prevents the acute or delayed onset of nephritis.
Indeed, there is substantial evidence to the contrary
(Saulsbury 1993; Ronkainen et al. 2006).
Pearl: Glucocorticoids are beneficial in the treatment of
Fig. 21.2 Swelling of the ankles and dorsum of the foot in a patient some aspects of HSP.
with HSP (Figure courtesy of Dr. Frank Saulsbury)
Comment: A recent randomized, double-blind, placebo-con-
trolled study showed that prednisone in a dose of 1 mg/kg/
in a sizeable subset of patients: nearly one-fifth (Saulsbury day for 2 weeks, then tapered over the subsequent 2 weeks,
1999; Trapani et al. 2005). Gastrointestinal symptoms in the is effective in treating the joint and abdominal pain associ-
absence of the characteristic rash can mimic a surgical abdo- ated with HSP (Ronkainen et al. 2006). These results con-
men, an infectious enteritis, or inflammatory bowel disease. firmed previous retrospective studies, anecdotal reports, and
More than one surgeon has been fooled by HSP. many unpublished clinical observations. However, glucocor-
Pearl: Most intussusceptions in patients with HSP are con- ticoids do not shorten the duration of purpura, prevent nephri-
fined to the small bowel. tis, or forestall disease recurrences (Ronkainen et al. 2006).
Pearl: Renal function may deteriorate years after an episode
Comment: Only 1–5% of patients with HSP develop intus-
of HSP nephritis in a child.
susception. Intussusceptions are confined to the small bowel
in 60% of patients with HSP. This contrasts with the usual Comment: Nephritis is the one feature of HSP that may have
ileo-colic location in children without HSP who develop chronic consequences, and the long-term prognosis is heav-
idiopathic intussusception (Choong and Beasley 1998). ily dependent upon the severity of nephritis at presentation.
Thus, abdominal computerized tomography or ultrasound, Approximately 30–50% of patients with nephritis have per-
rather than air contrast enema, is the preferred diagnostic sistent urinary abnormalities for months or years, but only
modalities if intussusception is suspected in a child with 1–3% of patients progress to end-stage renal disease
HSP. Surgical reduction of the intussusception is often (Saulsbury 1999; Trapani et al. 2005).
necessary. In general, patients with microscopic hematuria and trivial
proteinuria have an excellent prognosis (Narchi 2005). By
Pearl: Nephritis rarely precedes the appearance of cutane-
contrast, nephritis complicated by nephrotic syndrome has a
ous purpura.
poor prognosis (Ronkainen et al. 2003). In addition, the histo-
Comment: Unlike arthritis and gastrointestinal symptoms, pathologic finding of crescents in more than 50% of glomeruli
nephritis rarely develops before the appearance of the char- on renal biopsy is associated with a poor prognosis. However,
acteristic rash. In fact, the onset of nephritis may be delayed neither the clinical presentation nor the biopsy findings serve
for up to 3 months in a substantial minority of patients. as precise predictors of the ultimate outcome in all patients.
Two long-term follow-up studies of children with HSP
Myth: Nonsteroidal anti-inflammatory agents (NSAIDs) are
nephritis showed worsening renal disease as adults. Renal
an effective treatment for the joint pain and abdominal pain
deterioration was seen in some patients who had relatively
associated with HSP.
mild nephritis as children (Goldstein et al. 1992; Ronkainen
Reality: There has never been a properly conducted study et al. 2002). Both studies also showed that pregnancy was
examining the efficacy of NSAIDs in the treatment of HSP. complicated by proteinuria and hypertension in 40–60% of
Moreover, there is little anecdotal evidence that NSAIDs are women who had HSP nephritis as children. One half of the
effective. Nevertheless, NSAIDs are often recommended for women had mild nephritis that had completely resolved
the treatment of HSP. One should be circumspect about using when they were children. Thus, long-term monitoring of
21 Pediatric Vasculitis 225

patients is necessary, and particular vigilance is warranted in

pregnant women.
Pearl: The vast majority of patients have a disease duration
of less than 12 weeks.
Comment: One should not classify patients as having pro-
longed disease until they have experienced persistent symp-
toms lasting greater than 12 weeks (Saulsbury 1999; Trapani
et al. 2005). A variety of infections appear able to precipitate
HSP, and it is possible that others are also capable of prolong-
ing the disease process, as well. If an infection is present, a
course of antibiotic therapy is often successful in bringing the
condition to a halt. Thus, a careful search for underlying infec-
tion is warranted. A few anecdotal reports suggest that dapsone
and colchicine are effective in treating prolonged symptoms,
but neither drug has been studied in a systematic fashion.
Myth: HSP is always a self-limited disorder. Fig. 21.3 Palpable purpura: the sine qua non of HSP. This patient is a
30-year-old man who experienced months of recurrent, diffuse purpura,
Reality: HSP can appear with a frightening, multi-organ sys- refractory to conventional doses of prednisone (e.g., 1 mg/kg/day) and
tem presentation in previously healthy patients. Fortunately, other therapies. He responded dramatically to 3 days of pulse methyl-
in the large majority of cases, the process is self-limited and prednisolone (1 g/day three times) and entered a year-long remission
before a recurrence, which responded again to pulse methylpredniso-
does not require immunosuppressive therapy. (Indeed, it is lone (Figure courtesy of Dr. John Stone)
not entirely clear that immunosuppressive therapy helps
some of the clinical manifestations.)
In a small minority of cases – particularly those in adults –
HSP appears to be a more chronic condition or at least one
that requires months or years to resolve. These cases can be
characterized by recurrent or even virtually continuous cuta-
neous purpura, with intermittent arthritis or other disease
manifestations.
Pearl: If HSP is not self-limited, “mega” doses of prednisone
may work.
Comment: In a small minority of patients, palpable purpura
(Fig. 21.3) and sometimes other major disease manifestations
(joint complaints and abdominal pain) recur regularly and
respond poorly to series of conventional treatment approaches:
glucocorticoids at doses of up to 60 mg/day, colchicine, dap-
sone, azathioprine, and others. Limited anecdotal experience
suggests that rituximab is also ineffective in this setting.
For reasons that remain unclear, “mega” doses of gluco-
corticoids appear to work in a significant number of these Fig. 21.4 IgA vasculitis associated with an IgA paraproteinemia. There
cases. A 3-day methylprednisolone pulse (1 g/day) can lead are cutaneous findings of both small- and medium-vessel vasculitis.
to disappearance of the purpura for a year or more. In some Small-vessel: purpura; medium-vessel: livedo reticularis (racemosa)
cases, HSP returns after that time, but typically responds (Figure courtesy of Dr. John Stone)
again to mega doses of glucocorticoids. Because glucocorti-
coids have 100% oral bioavailability, anecdotal experience Comment: Patients with IgA paraproteinemia and vasculitis
has also shown that oral pulses of prednisone can be effec- have involvement of larger caliber vessels than do patients
tive: prednisone 960 mg (sixteen 60 mg tablets) once daily with typical HSP, and often present with livedo reticularis
for 3 days. This is appropriate, because most cases of recur- (indicative of medium-vessel vasculitis) as well as purpura
rent purpura related to HSP do not require hospitalization. (Fig. 21.4). The natural history of this disorder and its opti-
Pearl: In adult cases of refractory HSP, patients should be mal treatment remain unclear. Evolution to multiple myeloma
evaluated for the presence of an IgA paraproteinemia. has been reported (Birchmore et al. 1996).
226 A. H. Rowley et al.

a c

Fig. 21.5 Direct immunofluorescence studies in four types of cutane- ary to hepatitis C (from Stone and Nousari (2001); reprinted with kind
ous vasculitis: (a) HSP; (b) hypocomplementemic urticarial vasculitis; permission from Current Science)
(c) connective tissue disease; and (d) type II cryoglobulinemia second-

Myth: Performing direct immunofluorescence (DIF) studies vasculitis. Not only is DIF the only way of confirming the
on skin biopsies from patients with cutaneous vasculitis is diagnosis of HSP (Fig. 21.5a), proper interpretation of the
usually unnecessary. pattern of immunoreactant deposition may also yield clues to
the diagnosis and underlying pathophysiology of other con-
Reality: Failure to perform DIF squanders the opportunity to ditions. For example, DIF studies in hypocomplementemic
collect important data about the etiology of cutaneous urticarial vasculitis show striking IgG deposition around
21 Pediatric Vasculitis 227

blood vessels and at the basement membrane zone 21.3 Polyarteritis Nodosa
(Fig. 21.5b). In cutaneous vasculitis associated with connec-
tive tissue disorders, the phenomenon of the “in vivo ANA”
Myth: The outcome of polyarteritis nodosa (PAN) in child-
is also observed in keratinocytes and dermal cells in DIF
hood is poor.
studies (Fig. 21.5c). In cryoglobulinemia, DIF reveals IgM,
IgG, and complement deposition (Fig. 21.5d). Reality: Childhood PAN is a curable disease (as in PAN in
adults [see polyarteritis nodosa]). In fact, a recent international
Pearl: Prolonged standing sometimes triggers a flare of cuta-
study of 110 children with PAN showed that the prognosis in
neous vasculitis in patients whose disease is not entirely
children compared favorably with that of adults reported in
controlled.
the literature. Relapses were rare following the induction of
Comment: Purpura resolves more quickly with bed rest. This remission in children, and the overall mortality was only 1.1%
pressure-related phenomenon partly explains the most com- (Ozen et al. 2004). Like March, PAN can come in like a lion,
mon distribution of purpura: over the dependent lower but if treated with prudent aggression it is subject to taming.
extremities and buttocks. Patients who have recently begun
therapy for cutaneous vasculitis should be advised to avoid Pearl: Hepatitis B-related PAN is now vanishingly rare.
prolonged periods of standing, as this may cause new purpu- Comment: Thanks to the hepatitis B vaccine, this type of
ric eruptions. PAN has almost become an extinct disease in childhood. In
Pearl: Immunofluorescence studies of skin biopsies may yield an international registry of 110 PAN patients (Ozen et al.
a diagnosis of vasculitis even when H&E stains are 2004), only 4.6% of the children had hepatitis B-related
nondiagnostic. cases. In all of those patients, their diagnosis antedated the
time when vaccination became mandatory.
Comment: This is a sampling issue that is obvious to patholo- Because hepatitis C virus (HCV) is usually mentioned in
gists but not appreciated by many clinicians. The skin biopsy the same breath as hepatitis B, it is worth noting here that
is divided and sent to separate laboratories for processing. HCV-associated cryoglobulinemic vasculitis is also almost
One sample may contain diagnostic information that is absent never encountered in children. This is probably because of the
in the other. Even if the H&E stain is nondiagnostic, the long latency observed between the acquisition of HCV infec-
immunofluorescence study may demonstrate the characteris- tion and the appearance of cryoglobulinemic vasculitis in the
tic immunoreactants within blood vessel walls (i.e., IgA). minority of patients who develop that complication. By the
Incidentally, the same phenomenon is true for other kinds time a pediatric patient has had HCV long enough to develop
of tissue biopsies. Kidney biopsy samples sent for H&E and cryoglobulinemic vasculitis; the child has become an adult.
immunofluorescence studies in renal disease may lack the
all-important glomeruli, yet the electron microscopy study Myth: The American College of Rheumatology (ACR) clas-
can reveal crescentic and necrotizing changes, findings that sification criteria for Takayasu’s arteritis (TA) cover the
are critical to rendering the diagnosis. pediatric manifestations of this disease adequately.
Pearl: The immunopathogenesis of HSP involves IgA1, not Reality: An important hole in the ACR classification criteria
IgA2. is the failure to acknowledge that most childhood cases of
Comment: IgA plays a pivotal role in the pathogenesis of HSP, TA present with hypertension (Golding et al. 1977; Ozen et
even though the precise details surrounding the etiology of this al. 2007; Cakar et al. 2008). Hypertension, usually caused by
disease remain unclear. There are two subclasses of IgA: IgA1 the development of intra- or extra-renal arterial or arteriolar
and IgA2. Only IgA1 is involved in the pathogenesis of HSP. stenoses (Fig. 21.6), is not numbered among the ACR crite-
The clinical manifestations of HSP are a consequence of depo- ria, which were based principally upon adult cases. In con-
sition of IgA1 in vessel walls and the renal mesangium. trast, the European League Against Rheumatism/Paediatric
The reasons for the exclusive involvement of IgA1 in HSP Rheumatology European Society criteria include hyperten-
are not known for certain. However, IgA1, unlike IgA2, con- sion as a feature (Ozen et al. 2006).
tains a hinge region with multiple O-linked glycosylation sites.
Pearl: Subglottic stenosis is more common among children
Two studies have demonstrated aberrant glycosylation of the
with Wegener’s granulomatosis (WG) than it is in adult
hinge region of IgA1 in patients with HSP (Saulsbury 1997;
patients.
Allen et al. 1998). IgA1 molecules with diminished glycosyla-
tion of the hinge region are prone to aggregate into macromo- Comment: Pediatric case series highlights the greater fre-
lecular complexes have delayed clearance from the circulation, quency of subglottic stenosis among children with WG
activate the alternative pathway of complement, and deposit in (Rottem et al. 1993; Frosch and Foell 2004). In addition to
vessel walls and renal mesangium (Saulsbury 2001). subglottic stenoses, tracheal and endobronchial lesions are
Understanding the mechanisms of aberrant glycosylation of probably also more common in pediatric cases of WG.
IgA1 may shed light on the fundamental cause of HSP. Subglottic, tracheal, and endobronchial stenoses comprise a
228 A. H. Rowley et al.

ulcers caused by medium-vessel vasculitis there are distinct

disease entities. One of these is cutaneous PAN, a disorder
that affects primarily the medium-sized arteries and muscu-
lar arterioles of the skin.
The progression of cutaneous PAN to a severe systemic
form is unusual, and it is likely that cutaneous PAN is differ-
ent from classic PAN from the start (Daoud et al. 1997;
Kumar et al. 1995). Cutaneous PAN is characterized by a
relapsing and remitting course that responds to glucocorti-
coids and other immunosuppressive agents, but tends to flare
when the prednisone is discontinued or the dose tapered to a
sufficiently low level. Cutaneous PAN has also been associ-
ated with streptococcal infections, albeit the precise relation-
ship between this organism and at least some cases of the
disease is not clear (Kumar et al. 1995; Ozen et al. 2004).
Patients with cutaneous PAN occasionally have musculo-
skeletal complaints in addition to skin ulcers. Some also have
a sensory peripheral neuropathy, but the size and type of
nerve fibers involved in this entity remain poorly defined.
The potentially devastating sensorimotor mononeuritis mul-
tiplex that often complicates classic PAN is not characteristic
of cutaneous PAN, and if present this finding puts the diag-
nosis squarely into the classic PAN basket.
Myth: PAN is more common in boys than in girls.
Fig. 21.6 Renal angiogram in a pediatric patient with Takayasu’s arteri-
tis. Several arteriolar stenosis in an intra-renal vessel has led to dramatic Reality: The origin of this myth is not clear, but it is a myth.
collateral circulation and high serum renin levels. Efforts to revascularize
this patient’s kidney through stent placements were not successful. The Girls are equally likely to be affected by childhood PAN as
blood pressure was control effectively with an angiotensin converting are boys (1997; Ozen et al. 2004).
enzyme inhibitor (Figure courtesy of Dr. John Stone)
Pearl: Genetic associations have been defined in systemic
vasculitis.
Table 21.3 Consensus classification criteria for Wegener’s granuloma-
tosis in childrena Comment: A number of genetic associations have been
Three of the following six features should be present: reported in systemic vasculitides, suggesting leads for fur-
Abnormal urinalysisb ther understanding of these complex genetic traits that cer-
Granulomatous inflammation on biopsyc tainly involve other factors (particularly environmental) in
Nasal sinus inflammation their pathogenesis. A study in ANCA associated vasculitis
Subglottic, tracheal, or endobronchial stenosis
found an association between an IL-10 polymorphism
Abnormal chest radiograph or CT
PR3-ANCAd or C-ANCA staining
(−1082) and ANCA-associated vasculitis (Bártfai et al.
a
2003). Disease susceptibility and clinical course in WG may
European League Against Rheumatism/Paediatric Rheumatology
European Society (adapted from Ozen 2006) be associated with polymorphisms in genes of certain cytok-
b
Hematuria and/or significant proteinuria ines and CTLA4 genes (Spriewald et al. 2005). In children
c
Kidney biopsy typically shows necrotizing pauci-immune with vasculitis and PAN, mutations in the MEFV gene have
glomerulonephritis been reported as risk factors for the disease (Ozen et al. 2003;
d
ANCA: antineutrophil cytoplasmic antibodies Yalçinkaya et al. 2007). In Kawasaki’s disease, genetic varia-
tion in the gene of IL4 appears to play a role in disease sus-
separate criterion in the EULAR/PReS classification criteria ceptibility (Burns et al. 2005). The genetic epidemiology of
for WG (Table 21.3). A high index of suspicion is required the vasculitides remains a wide open field.
for the diagnosis of all three of these types of lesions.
Myth: The vasculitis damage index can be applied readily to
Myth: Cutaneous PAN is always at risk for progression to the pediatric cases.
systemic form of the disease.
Reality: The vasculitis damage index has significant short-
Reality: Classic PAN can cause painful skin nodules and hor- comings, even in adult patients (Seo et al. 2007). Its use in
rific cutaneous ulcerations, but within the universe of skin children highlights additional shortcomings. For example,
21 Pediatric Vasculitis 229

“growth failure” due to either the underlying disease or the References

effects of long-term glucocorticoids use is a major source of
damage in children with vasculitis. Other items that are Allen AC, Willis FR, Beattie TJ, et al Abnormal IgA glycosylation in
scored commonly in adults (e.g., myocardial infarction) Henoch–Schönlein purpura restricted to patients with clinical
almost never apply to children with vasculitis. nephritis. Nephrol Dial Transplant. 1998;13:930–4
Amano S, Hazama F, Kubagawa H, et al General pathology of Kawasaki
In short, children are not “just little adults.” Their cases
disease. Acta Pathol Jpn. 1980;30(5):681–94
have distinct nuances that require special expertise in the Anderson MS, Burns J, Treadwell TA, et al Erythrocyte sedimentation
interest of proper management. A new classification of child- rate and C-reactive protein discrepancy and high prevalence of coro-
hood vasculitis is shown in Table 21.4. nary artery abnormalities in Kawasaki disease. Pediatr Infect Dis J.
2001;20(7):698–702
Pearl: The von Willebrand factor antigen is a useful bio- Bártfai Z, Gaede KI, Russell KA, et al Different gender-associated
marker in small-vessel vasculitis in children. genotype risks of Wegener’s granulomatosis and microscopic poly-
angiitis. Clin Immunol. 2003;109(3):330–7
Comment: Assays for the von Willebrand factor antigen are Binstadt BA, Levine JC, Nigrovic PA, et al Coronary artery dilation
among patients presenting with systemic-onset juvenile idiopathic
used most commonly for evaluating boys with a bleeding diath-
arthritis. Pediatrics 2005;116:e89–93
esis. This antigen is also an acute phase reactant that is released Birchmore D, Sweeney C, Choudhury D, et al IgA multiple myeloma
by both platelets and endothelial cells following cellular presenting as HSP/Polyarteritis nodosa overlap. Arthritis Rheum.
injury or death. Conditions such as small-vessel vasculitides 1996;39:698–703
Burns JC, Wiggins JW, Toews WH, et al Clinical spectrum of Kawasaki
(e.g., HSP, PAN, and microscopic polyangiitis) and microan-
disease in infants younger than 6 months of age. J Pediatr.
giopathic disorders (dermatomyositis and systemic sclerosis) 1986;109:759–63
can be associated with von Willebrand levels that are elevated Burns JC, Mason WH, Glode MP, et al Clinical and epidemiologic
up to four times the upper limit of normal (Tse et al. 1998). characteristics of patients referred for evaluation of possible
Kawasaki disease. United States Multicenter Kawasaki Disease
Elevated levels of von Willebrand factor antigen do not
Study Group. J Pediatr. 1991;118:680–6
distinguish primary vasculitic disorders from those associ- Burns JC, Shimizu C, Shike H, et al Family-based association analysis
ated with vasculopathy, but serial measurements of the levels implicates IL-4 in susceptibility to Kawasaki disease. Genes Immun.
of this antigen may be helpful in tracking disease activity in 2005;6(5):438–44
individual patients. Cakar N, Yalcinkaya F, Duzova A, et al Takayasu arteritis in children.
J Rheumatol. 2008;35(5):913–19
Centers for Disease Control. Haemophilus influenzae invasive disease
among children age <5 years–California, 1990–1996. MMWR.
1998;47:737–40
Choong CK, Beasley SW. Intra-abdominal manifestations of Henoch-
Schönlein purpura. J Paediatr Child Health. 1998;34:405–9
Table 21.4 Classification scheme for pediatric vasculitis (adapted Daoud MS, Hutton KP, Gibson LE. Cutaneous periarteritis nodosa: a clini-
from Ozen (2006) ) copathological study of 79 cases. Br J Dermatol. 1997;136:706–13
Dengler LD, Capparelli EV, Bastian JF, et al Cerebrospinal fluid profile
Predominantly large-vessel vasculitis in patients with acute Kawasaki disease. Pediatr Infect Dis J.
Takayasu’s arteritis 1998;17:478–81
Predominantly medium-vessel vasculitis Frosch M, Foell D. Wegener granulomatosis in childhood and adoles-
Childhood polyarteritis nodosa cence. Eur J Pediatr. 2004;163:425–34
Cutaneous polyarteritis Golding RL, Perri G, Cremin BJ. The arteriographic manifestations of
Kawasaki’s disease Takayasu’s arteritis in children. Pediatr Radiol. 1977;5:224–30
Goldstein AR, White RHR, Akuse R, et al Long-term follow-up of
Predominantly small-vessel vasculitis
childhood Henoch-Schönlein nephritis. Lancet 1992;339:280–2
Granulomatous
Holman RC, Belay ED, Curns AT, et al Kawasaki syndrome hospital-
Wegener’s granulomatosis izations among children in the United States, 1988–1997. Pediatrics
Churg–Strauss syndrome 2003;111:e448
Non-granulomatous Jurado RL. Why shouldn’t we determine the erythrocyte sedimentation
Microscopic polyangiitis rate? Clin Infect Dis. 2001;33:548–9
HSP Kumar L, Thapa BR, Sarkar B, Walia BN. Benign cutaneous PAN in
Isolated cutaneous leucocytoclastic vasculitis children below 10 years of age: a clinical experience. Ann Rheum
Hypocomplementemic urticarial vasculitis Dis. 1995;54:134–6
Other vasculitides Landing BH, Larson EJ. Are infantile periarteritis nodosa with coronary
artery involvement and fatal mucocutaneous lymph node syndrome
Behcet disease
the same? Comparison of 20 patients from North America with
Vasculitis secondary to infection (including hepatitis B-associated
patients from Hawaii and Japan. Pediatrics 1977;59:651–62
polyarteritis nodosa), malignancies, and drugs, including
Momenah T, Sanatani S, Potts J, et al Kawasaki disease in the older
hypersensitivity vasculitis
child. Pediatrics 1998;e102(1):e7
Vasculitis associated with connective tissue diseases Muta H, Ishii M, Sakaue T, et al Older age is a risk factor for the devel-
Isolated vasculitis of the central nervous system opment of cardiovascular sequelae in Kawasaki disease. Pediatrics
Cogan syndrome 2004;114:751–4
230 A. H. Rowley et al.

Narchi H. Risk of long term renal impairment and duration of follow up Saulsbury FT. Corticosteroid therapy does not prevent nephritis in
recommended for Henoch–Schönlein purpura with normal or mini- Henoch–Schönlein purpura. Pediatr Nephrol. 1993;7:69–71
mal urinary findings: a systematic review. Arch Dis Child. Saulsbury FT. Alterations in the O-linked glycosylation of IgA1 in chil-
2005;90:916–20 dren with Henoch–Schönlein purpura. J Rheumatol. 1997;24:2246–9
Newburger J, et al Diagnosis, treatment, and long-term management of Saulsbury FT. Henoch–Schönlein purpura in children: report of 100
Kawasaki disease: a statement for health professionals from the patients and review of the literature. Medicine 1999;78:395–409
committee on rheumatic fever, endocarditis, and Kawasaki disease, Saulsbury FT. Henoch–Schönlein purpura. Curr Opin Rheumatol.
council on cardiovascular disease in the young, Am Heart Assoc 2001;13:35–40
Pediatr. 2004a;114:1708–33 Seo P, Luqmani R, Flossmann O, et al The future damage assessment in
Newburger JW, Sanders SP, Burns JC, et al Left ventricular contractility vasculitis. J Rheumatol. 2007;34:1357–71
and function in Kawasaki syndrome. Circulation 1989;79:1237–46 Spriewald BM, Witzke O, Wassmuth R, et al Distinct tumour necrosis
Newburger JW, Takahashi M, Gerber MA, et al Diagnosis, treatment, factor alpha, interferon gamma, interleukin 10, and cytotoxic T cell
and long-term management of Kawasaki disease. Circulation antigen 4 gene polymorphisms in disease occurrence and end stage
2004a;110:2747–71 renal disease in Wegener’s granulomatosis. Ann Rheum Dis.
Niwa K, Aotsuka H, Hamada H, et al Thrombocytopenia: a risk factor 2005;64:457–61
for acute myocardial infarction during the acute phase of Kawasaki Stamos JK, Corydon K, Donaldson J, Shulman ST. Lymphadenitis as
Disease. Coron Artery Dis. 1995;6(11):857–64 the dominant manifestation of Kawasaki disease. Pediatrics
Ozdogan H, Arisoy N, Kasapcapur O, et al Vasculitis in familial 1994;93(3):525–8
Mediterranean fever. J Rheumatol. 1997;24:323–7 Stockheim JA, Innocentini N, Shulman ST. Kawasaki disease in older
Ozen S, Bakkaloglu A, Yilmaz E, Duzova A, Balci B, Topaloglu R, children and adolescents. J Pediatr. 2000;137:250–2
Besbas N. Mutations in the gene for familial Mediterranean fever: Stone JH, Nousari HC. Essential cutaneous vasculitis: everything a
do they predispose to inflammation? J Rheumatol. rheumatologist needs to know about vasculitis of the skin. Curr
2003;30(9):2014–8 Opin Rheumatol. 2001;13:23–34
Ozen S, Anton J, Arisoy N, et al Juvenile polyarteritis: results of a mul- Tanaka N, Sekimoto K, Naoe S. Kawasaki disease. Relationship with
ticenter survey of 110 children. J Pediatr. 2004;145:517–22 infantile periarteritis nodosa. Arch Pathol Lab Med. 1976;100:81–6
Ozen S, Ruperto N, Dillon MJ, et al EULAR/PReS endorsed consensus Trapani S, Micheli A, Grisolia F, et al Henoch–Schonlein purpura in
criteria for the classification of childhood vasculitides. Ann Rheum childhood: epidemiological and clinical analysis of 150 cases over a
Dis. 2006;65:936–41 5-year period and review of the literature. Semin Arthritis Rheum.
Ozen S, Duzova A, Bakkaloglu A, et al Takayasu arteritis in children: 2005;35:143–53
preliminary experience with cyclophosphamide induction and corti- Tse WY, Cockwell P, Savage CO. Assessment of disease activity in sys-
costeroids followed by methotrexate. J Pediatr. 2007;150:72–6 temic vasculitis. Postgrad Med J. 1998;74:1–6
Ronkainen J, Nuutinen M, Koskimies O. The adult kidney 24 years Umezawa T, Saji T, Matsuo N, Odagiri K. Chest X-ray findings in the acute
after childhood Henoch–Schönlein purpura: a retrospective cohort phase of Kawasaki disease. Pediatr Radiol. 1989;20(1–2):48–51
study. Lancet 2002;360:666–70 van Leeuwen MA, van Rijswijk MH. Acute phase proteins in the moni-
Ronkainen J, Ala-Houhala M, Huttunen NP, et al Outcome of Henoch- toring of inflammatory disorders. Baillieres Clin Rheumatol.
Schoenlein nephritis with nephrotic-range proteinuria. Clin Nephrol. 1994;8(3):531–52
2003;60:80–4 Wickiser JE, Saulsbury FT. Henoch–Schönlein purpura in a child with
Ronkainen J, Koskimies O, Ala-Houhala M, et al Early prednisone hyperimmunoglobulinemia D and periodic fever syndrome. Pediatr
therapy in Henoch–Schönlein purpura: a randomized, double-blind, Dermatol. 2005;22:138–41
placebo-controlled trial. J Pediatr. 2006;149:241–7 Yalçinkaya F, Ozçakar ZB, Kasapçopur O, et al Prevalence of the
Rosenfeld EA, Corydon KE, Shulman ST. Kawasaki disease in infants MEFV gene mutations in childhood PAN. J Pediatr.
less than one year of age. J Pediatr. 1995;126:524–49 2007;151:675–8
Rottem M, Fauci AS, Hallahan CW, et al Wegener granulomatosis in chil- Yanagawa H, Nakamura Y, Yashiro M, et al Incidence survey of
dren and adolescents: clinical presentation and outcome. J Pediatr. Kawasaki disease in 1997 and 1998 in Japan. Pediatrics 2001;
1993;122(1):26–31 107(3):e33
 Behçet’s Syndrome
22
Hasan Yazici and Yusuf Yazici

rule among a cohort of patients reported from northern Italy

22.1 Disease Overview (Pipitone et al. 2004).
Myth: Neutrophil hypermotility is a cardinal characteristic
› The prevalence of Behçet’s syndrome is highest in
of Behçet’s syndrome.
countries of the eastern Mediterranean, the Middle
East, and East Asia. Reality: Neutrophil hypermotility is often referred to as a cen-
› Aphthous oral ulcers are usually the first and most per- tral feature in the pathogenesis of Behçet’s syndrome.
sistent clinical feature of Behçet’s syndrome. Aphthous Statements about the importance of neutrophil hypermotility
ulcers also occur frequently on the genitals (e.g., the are usually accompanied by assertions that colchicine achieves
scrotum or vulva). its effectiveness through the inhibition of neutrophil motility.
› Uveitis – most frequently a bilateral panuveitis is com- In fact, there is little evidence that any of this is true. We
mon in Behçet’s syndrome. Eye disease constitutes a are not aware of any studies that demonstrate greater neutro-
major source of morbidity. phil motility in Behçet’s syndrome when compared with other
› Many forms of central nervous system disease may inflammatory conditions. On the contrary, one controlled
occur in Behçet’s syndrome. These include dural sinus study revealed no such differences in a study of Behçet’s syn-
thrombi and white matter lesions in the midbrain and drome and other disorders (Tüzün et al. 1999). Moreover,
brainstem. Unlike other vasculitides, however, periph- colchicine has rather limited utility in the treatment of Behçet’s
eral nervous system disease is rare. syndrome.
› HLA-B51 is a risk factor for Behçet’s syndrome.
Myth: Colchicine is an effective agent for the treatment of
oral ulcers in Behçet’s disease.
Reality: Colchicine remains one of the most widely used
Myth: Behçet’s syndrome is very rare.
medications for the treatment of Behçet’s syndrome, yet,
Reality: This certainly is not so in Turkey, where the preva- objective data do not suggest tremendous efficacy for this
lence of the disorder is one in every 250 adults (Azizlerli drug. Neither of the two randomized, controlled trials of
et al. 2003). A recent survey from France showed that this agent in Behçet’s syndrome demonstrated efficacy for
Behçet’s syndrome is several times more common than the treatment of oral ulcers (Aktulga et al. 1980; Yurdakul
Wegener’s granulomatosis, microscopic polyangiitis, the et al. 2001). Although one trial reported efficacy for colchi-
Churg-Strauss syndrome, and polyarteritis nodosa (Mahr cine in the management of genital ulcers, this effect was
et al. 2004, 2008). observed only in female patients (Yurdakul et al. 2001).
Although the diagnosis of Behçet’s syndrome is probably In short, the putative efficacy of colchicine in Behçet’s
more common in many geographic areas than generally syndrome is significantly overblown. Patients tend to tolerate
believed, there appear to be important differences in the clin- the drug well, but clinicians should not delay long before
ical expression of Behçet’s syndrome across different regions. moving on to another strategy if the positive effects of colchi-
In general, the severity of Behçet’s syndrome is lower in cine are slow in appearing.
regions where the disease is least prevalent (e.g., in the
Myth: Behçet’s syndrome is a form of spondyloarthropathy.
United States and Western Europe). For example, among
300 patients collected in New York City, a region of low Reality: The origin of this Myth stems from the early 1970s,
prevalence when compared to Turkey, no patients with com- when the newly described HLA-B27 antigen was believed to
plete visual loss secondary to Behçet’s syndrome were be the seat of the soul of many rheumatic conditions. Behçet’s
observed (Yazici and Moses 2007). Mild disease was also the syndrome was classified as a seronegative spondyloarthropathy

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 231

DOI:10.1007/978-1-84800-934-9_22, © Springer Science + Business Media B.V. 2009
232 H. Yazici and Y. Yazici

because of certain characteristics that were reminiscent of

ankylosing spondylitis, reactive arthritis, and other HLA-B27-
associated conditions: e.g., uveitis, oral and genital ulcers, and
oligoarthritis. Although additional investigations demonstrated
that sacroiliitis is not a part of the spectrum of Behçet’s syn-
drome and that the strong HLA association for this disorder is
with B51 rather than B27, the notion that Behçet’s syndrome is
a form of spondyloarthropathy has somehow persisted in the
minds of some.
Behçet’s syndrome and seronegative spondyloarthropa-
thies belong to separate kettles, but there is an additional dis-
ease manifestation that these conditions sometimes share:
enthesopathy. The finding of enthesopathy in Behçet’s syn-
drome tends to cluster in patients who have both acne and
arthritis (Tunc et al. 2002; Hatemi et al. 2008).
Pearl: Behçet’s syndrome is not a “connective tissue disorder”.
Reality: Although non-rheumatologists frequently tar
Behçet’s syndrome with the same brush as they do systemic
lupus erythematosus, rheumatoid arthritis, and scleroderma,
the fundamental differences exist between Behçet’s syn-
drome and the other rheumatic conditions. They share little,
other than the need for the involvement of a rheumatologist
in their care.
Behçet’s syndrome differs from the classic “connective
tissue disorders” in several respects. For example, it is often Fig. 22.1 Papulopustular lesions on the buttocks of a patient with
Behçet’s syndrome. Acneiform lesions tend to occur in unusual areas in
especially helpful to remember that: patients with Behçet’s syndrome
• Behçet’s syndrome is not associated with either Raynaud’s
phenomenon or digital ulceration.
• Behçet’s syndrome is not associated with secondary Table 22.1 Differences between autoinflammatory diseases and
Sjögren’s syndrome (Günaydin et al. 1994). Behçet’s syndrome
• Behçet’s syndrome is often complicated by unique eye Autoinflammatory Behçet’s syndrome
diseases
findings, genital ulcerations, and papulopustular lesions
Age group affected Pediatric Adult
that occur beyond the sites where acne is usually found
Areas of highest Mostly Europe and Silk Route
(e.g., the buttocks) (Fig. 22.1) (Hatemi et al. 2004). prevalence North America
• Behçet’s syndrome tends to be more severe in men. This Monogenic heritability Yes No
marks a noticeable departure from the pattern of disease Relapsing fever Very common Very rare
generally observed in connective tissue diseases such as Disease course Relentless unless Abates in time
lupus. treated
• Behçet’s syndrome does not appear to carry a long-term
risk of atherosclerosis development. The frequency of coro-
differences between Behçet’s syndrome and autoinflamma-
nary calcifications, atherosclerotic plaques, angina pectoris, and
tory disorders are shown in Table 22.1.
myocardial infarction are not increased in Behçet’s syndrome
Particular links have been drawn between Behçet’s syn-
(Kural-Seyahi et al. 2003; Seyahi 2004; Urgurlu 2008).
drome and the hyper IgD syndrome (Sakane et al. 1999). In
fact, IgD levels are not elevated in Behçet’s syndrome
Myth: Behçet’s syndrome is an “autoinflammatory disorder
(Brezniak et al. 1998).
(Yazici and Fresko 2005)”.
Myth: All clinical manifestations of Behçet’s syndrome are
Reality: Behçet’s syndrome is an idiopathic inflammatory
caused by vasculitis.
condition, but differs substantially from the emerging group
of diseases with which it is sometimes lumped, known as the Reality: Many Behçet’s syndrome lesions are associated
autoinflammatory disorders (see Chap. 45). The major with perivascular inflammation in the absence of true
22 Behçet’s Syndrome 233

vascular wall destruction. Examples of such manifestations conditions with tender, nodular lesions. However, the EN of
are the oral ulcers, genital lesions, and brain findings Behçet’s syndrome bears some important differences to that
(Hirohata 2008). Acne lesions and their associated pustules observed, for example, in sarcoidosis and inflammatory
have also been considered to represent cutaneous vasculitis, bowel disease. The following major distinctions can be
but it is now known that they are indistinguishable histo- drawn:
pathologically from acne vulgaris (Ergun et al. 1998).
• Many EN lesions in Behçet’s syndrome are accompanied
Tissue biopsies in Behçet’s syndrome generally yield no
by a medium-vessel vasculitis. Vasculitis is not a compo-
pathological findings that are pathognomonic of that condi-
nent of the EN found in sarcoidosis and inflammatory
tion. Clinicians must bear in mind that unless there are com-
bowel disease.
pelling reasons for biopsy to exclude other disorders, the
• Permanent pigmentation of the overlying skin can com-
diagnosis of Behçet’s syndrome usually rests upon its clini-
plicate the EN associated with Behçet’s syndrome. This
cal features alone.
finding is reminiscent of the hyperpigmentation that
Myth: The pustules of Behçet’s syndrome are sterile. occurs frequently following the resolution of medium-
vessel vasculitides involving the skin (e.g., polyarteritis
Reality: Bacteria are present in the cultures of pustules in
nodosa; Fig. 22.2).
Behçet’s syndrome patients, just as they are in acne vulgaris
(Hatemi et al. 2004). However, the bacteriology of Behçet’s
syndrome differs from that of acne vulgaris. Staphylococcus
aureus, common in the lesions of Behçet’s syndrome, is
unusual in acne vulgaris.
Knowledge of the bacteriology of non-pustular lesions in
Behçet’s syndrome remains poorly described. Impairments
of innate immunity may explain some of the findings within
the acne pustules associated with Behçet’s syndrome, includ-
ing Staphylococcus aureus growth. One piece of evidence
supporting defects in innate immunity is the fact that
decreased mannose binding lectin levels (MBL) correlate
with more severe Behçet’s syndrome (Inanc et al. 2005).
MBL plays an important role in complement activation and
the opsonization of pathogens. Deficiency in this molecule
heightens a person’s susceptibility to infection.
Myth: The diagnosis of Behçet’s syndrome is untenable with-
out a history of oral ulcers.
Reality: Oral ulcers are often viewed as a sine qua non of
Behçet’s syndrome. Although oral ulcers occur in the over-
whelming majority of Behçet’s patients, there is a small sub-
set that has classic disease features in every other way but do
not manifest oral ulcers.
During the development of the International Study Group’s
1990 classification criteria (International Study Group 1990),
28 such patients were identified among the 914 patients
included in the study. Thus, 3% of patients diagnosed by
Behçet’s syndrome experts as having that disorder did not have
oral ulcers. At times, some of those patients developed the clas-
sic oral lesions, but others remained ulcer-free even as other
manifestations of Behçet’s syndrome persisted or recurred.
Myth: Tender red lesions that occur on the lower extremities Fig. 22.2 Cutaneous hyperpigmentation following the resolution of a
in Behçet’s syndrome are inevitably erythema nodosum. medium-vessel vasculitis (in this case, polyarteritis nodosa). The same
phenomenon can be observed in cases of erythema nodosum that occur
Reality: Erythema nodosum (EN), a septal panniculitis, does in Behçet’s syndrome, because these lesions are often associated with a
occur in Behçet’s syndrome and presents as it does in other medium-vessel vasculitis
234 H. Yazici and Y. Yazici

Fig. 22.3 Superficial thrombophlebitis in a patient with Behçet’s syn-

drome. Biopsy may be required to distinguish this lesion from erythema
nodosum
b
Superficial thrombophlebitis (STP) is another important
cause of painful red lesions in Behçet’s syndrome (Fig. 22.3).
STP is observed much more commonly among men with
Behçet’s syndrome. In contrast, EN lesions are more com-
mon among female patients. Making the distinction between
EN and STP is important, because STP is associated with a
different subset of severe disease manifestations, e.g., inflam-
mation within large veins that can lead to intraluminal throm-
bosis (Tunc 2002). In many instances, a biopsy is required
for differential diagnosis (Fig. 22.4a, b).
Myth: Behçet’s syndrome is an unrelenting disorder that
generally leads to blindness among patients with ocular
involvement.
Reality: Several decades ago, uveitis of Behçet’s syndrome
had an extremely poor prognosis, particularly if the process
was advanced at the time of diagnosis. Although posterior
uveitis still poses significant challenges in therapy, its prog-
nosis is not inevitably dire. Tumor necrosis factor inhibitors
and interferon-alpha have both found important niches in the
treatment of ocular Behçet’s syndrome.
Pearl: Males with Behçet’s syndrome tend to have a more
severe disorder than females with the disease.
Comment: Disease-related mortality is significantly higher
among males than females. Ocular disease is particularly severe
in male patients. Differences in the expression of Behçet’s syn-
drome according to sex are summarized in Table 22.2.
In contrast to the typically more severe disease in male Fig. 22.4 Histology of superficial thrombophlebitis (a) and erythema
patients, Behçet’s syndrome is often a mild condition among nodosum lesions (b) (a). An intraluminal thrombus (b). Septal panniculi-
women and the elderly. Moreover, in many patients, the dis- tis. Also see Fig. 22.2
ease goes into complete remission with the passage of time
Comment: This feature helps distinguish the ulcers of Behçet’s
(Kural-Seyahi et al. 2003).
syndrome from those of reactive arthritis and venereal dis-
Pearl: Genital ulcers in male patients with Behçet’s syn- eases, which tend to occur on the penis. In female patients
drome are usually found on the scrotum. with Behçet’s syndrome, genital ulcers usually occur on the
22 Behçet’s Syndrome 235

Table 22.2 Sex differences in the disease phenotype of Behçet’s

syndrome
Same frequency More More
in males and common common
females among men among
women
Oral ulceration, +
genital ulceration
and arthritis
Erthema nodosum +
Superficial ++
thrombophlebitis
Eye disease ++
Unilateral eye +
disease
Pulmonary artery ++++
aneurysms
Venous disease ++
Peripheral arterial +++
disease
Central nervous ++
system disease

labia – the embryological equivalent of the scrotum in men.

Another helpful point in sorting through the differential diag-
nosis of genital lesions is that the scrotal ulcers of Behçet’s
syndrome usually leave scars (Fig. 22.5). Urethritis is also
uncommon in Behçet’s syndrome and suggests either a sexually-
transmitted disease or a complication of reactive arthritis.
Pearl: Intermittent claudication can be a feature of Behçet’s
Fig. 22.5 Scars on the scrotum of a patient with Behçet’s syndrome,
syndrome, but the mechanism appears to be independent of occurring at the sites of previous genital ulcers
atherosclerosis.
Comment: Intermittent claudication is observed in patients Comment: Some Behçet’s syndrome patients develop nodu-
with peripheral venous disease rather than arterial involve- lar pulmonary lesions that have a tendency to cavitate. These
ment (Ugurlu et al. 2008). It is thought to be due to severe nodules may be seen independently of pulmonary artery
venous outflow impairment, but the precise mechanism aneurysms. Such lesions, often seen in association with pul-
remains unclear. monary arterial aneurysms, are often mistaken for opportu-
Pearl: If a Behçet’s syndrome patient presents with hemopty- nistic infections (see Fig. 22.6). They can also create
sis, aneurysms of the pulmonary artery must be excluded. diagnostic confusion vis-à-vis Wegener’s granulomatosis.

Comment: Pulmonary arterial aneurysms comprise the most Myth: Behçet’s syndrome patients who suffer deep venous
deadly complication of this condition (Fig. 22.6). They normally thromboses should be anti-coagulated for life.
present with hemoptysis, and in the absence of aggressive Reality: The optimal management of deep venous thromboses
therapy (typically involving cyclophosphamide), the 1-year in Behçet’s syndrome remains a source of debate. Indeed, it is
patient mortality associated with this complication is in the not clear whether patients who develop deep venous thrombo-
order of 50% (Hamuryudan et al. 2004). sis require anti-coagulation at all. Deep venous thromboses in
Pulmonary arterial aneurysms are associated with throm- Behçet’s syndrome are believed to result from vascular inflam-
bosis of the big veins, i.e. vena cavae, in 80% of the cases. mation rather than from a hypercoagulable tendency per se. In
The so-called “Hughes-Stovin syndrome”, which consists of addition, deep venous clots that occur in Behçet’s syndrome
aneurysms of the pulmonary artery in the absence of other are believed to adhere tightly to the blood vessel wall. Thus,
stigmata of Behçet’s syndrome, probably represents an embolization is rare. Although there remains little data to sup-
incomplete form of Behçet’s syndrome (Erkan et al. 2004). port this practice, the general approach to the management of
Pearl: Parenchymal lung disease can also occur in Behçet’s deep venous thrombosis in Behçet’s syndrome is to intensify
syndrome. immunosuppression rather than to begin anti-coagulation.
236 H. Yazici and Y. Yazici

ineffective (Yazici et al. 1990; Hamuryudan et al. 1997).

Guidelines on the use of azathioprine and the importance of
testing for thiopurine methyltransferase alleles are discussed
elsewhere (see ANCA-associated vasculitis Chap. 24).
Pearl: Azathioprine should not be employed together with
interferon-alpha.
Comment: Interferon-alpha has a number of potential uses in
Behçet’s syndrome. However, the use of interferon-alpha is
frequently associated with leukopenia. Because of the pos-
sibility of a synergistic effect on bone marrow suppression,
interferon-alpha and azathioprine should not be employed
together.
Pearl: Although multiple sclerosis shares the tendency of
Behçet’s syndrome to involve the central nervous system
white matter, distinguishing these two conditions is usually
straightforward.
Comment: The following points are useful in distinguishing
between multiple sclerosis and Behçet’s syndrome:
Fig. 22.6 Cavitating pulmonary artery aneurysms
• Sensory symptoms are common in multiple sclerosis but
extremely rare in Behçet’s syndrome.
• Optic neuritis and spinal involvement are also uncommon
Myth: Behçet’s syndrome patients with thrombophlebitis
in Behçet’s syndrome, but occur frequently in multiple
should be anti-coagulated until the acute event is resolved.
sclerosis.
• On neuroimaging, multiple sclerosis lesions tend to occur
Reality: There is also a debate about the optimal management
in a periventricular distribution (Fig. 22.7a). In contrast,
of thrombophlebitis in Behçet’s syndrome. Formal data show-
the typical brain lesions of Behçet’s syndrome have a
ing that anti-coagulation is not useful are sparse (Kahraman
subcortical location and extend into the midbrain and
et al. 2003; Ahn et al. 2008). Superficial thrombophlebitis in
brainstem (Fig. 22.7b).
Behçet’s syndrome is often associated with pulmonary arte-
• Brainstem lesions in multiple sclerosis are usually small.
rial aneurysms, and anti-coagulation in this setting can lead
In contrast, both brainstem and cerebellar atrophy are
to catastrophic bleeding events. If anti-coagulation is contem-
often prominent in chronic Behçet’s syndrome.
plated for a patient with superficial thrombophlebitis, it seems
• An inflammatory cerebrospinal fluid strongly favors
prudent to perform computed tomography scans of the chest
Behçet’s syndrome (Siva and Yazici 2004).
to exclude pulmonary artery aneurysms first. Most Behçet’s
syndrome experts do not anti-coagulate patients with superfi-
Pearl: If vasculitic neuropathy is present, the diagnosis is
cial thrombophlebitis, but rather employ immunosuppressive
probably not Behçet’s syndrome.
agents such as azathioprine along with brief courses of gluco-
corticoids for treatment. Comment: Although Behçet’s syndrome has a predilection
for the nervous system, it rarely involves the peripheral ner-
Pearl: Azathioprine is frequently underdosed in the treat-
vous system. This helps considerably in differentiating
ment of Behçet’s syndrome.
Behçet’s syndrome from other primary forms of vasculitis,
Comment: Azathioprine has salutary effects on the course of e.g., polyarteritis nodosa, microscopic polyangiitis, or
Behçet’s syndrome, but is frequently administered in doses Wegener’s granulomatosis.
that are too low. Azathioprine decreases the frequency of
Pearl: Clusters of clinical manifestations are observed to
hypopyon uveitis attacks; the development of new eye dis-
occur among individual patients in Behçet’s syndrome.
ease; the occurrence of new genital ulcers; arthritis; and the
likelihood of thrombophlebitis (Hatemi et al. 2008). Comment: Three studies have confirmed the high frequency
Despite its potential efficacy in Behçet’s syndrome, azathio- with which acne, arthritis, and enthesopathy occur together
prine is frequently dosed in regimens that are too low to be in Behçet’s syndrome (Tunc 2002; Diri et al. 2001 Hetami
effective. Doses approaching 2.5 mg/kg/day should be employed 2008). A second recognized clinical cluster is the association
in Behçet’s syndrome before concluding that the medication is of superficial and deep vein thrombosis, with the
22 Behçet’s Syndrome 237

Figs. 22.7 Contrasting magnetic

resonance imaging features in
a
multiple sclerosis (a) and
Behçet’s syndrome (b).
(a). Multiple periventricular, T2
hyperintense white matter lesions
in a patient with multiple
sclerosis. (b). A single lesion
starting at basal ganglia and
extending into the midbrain in a
patient with Behçet’s syndrome.
Widespread central nervous
system lesions can also occur
(Figures courtesy of Professor S.
Saip)

development of dural sinus thrombi (Tunc et al. 2004). A Azizlerli G, Kose AA, Sarica R, et al Prevalence of Behçet’s disease in
third possible cluster is that of the presence of antibodies to Istanbul, Turkey. Int J Dermatol 2003; 42:803–806
Brezniak N, Shtrasburg S, Langevitz P, et al Serum IgD as a discriminator
anti-saccharomyces, a hallmark of Crohn’s disease, only between the two periodic febrile syndromes hyperimmunoglobulinaemia
among those patients with gut disease (Fresko et al. 2005). D syndrome and Behçet’s disease. Ann Rheum Dis 1998; 57:255–256
Apart from their obvious clinical implications, the pres- Diri E, Mat C, Hamuryudan V, et al Papulopustular skin lesions are seen
ence of such clusters suggests that there might be more than more frequently in patients with Behçet’s syndrome who have
arthritis: A controlled and masked study. Ann Rheum Dis 2001;
one disease mechanism in Behçet’s syndrome and hence, the 60:1074–1076
designation “syndrome” is more appropriate than “disease”. Ergun T, Gurbuz O, Dogusoy G, et al Histopathologic features of the
spontaneous pustular lesions of Behçet’s syndrome. Int J Dermatol
1998; 37:194–196
Erkan D, Yazici Y, Sanders A, et al Is Hughes-Stovin syndrome Behçet’s
References disease? Clin Exp Rheumatol 2004; 22(4 Suppl 34):S64–S68
Fresko I, Ugurlu S, Ozbakir F, et al Anti-Saccharomyces cerevisiae
antibodies (ASCA) in Behçet’s syndrome. Clin Exp Rheumatol
Ahn JK, Lee YS, Jeon CH, et al. Treatment of venous thrombosis 2005; 23(4 Suppl 38):S67–S70
associated with Behçet’s disease: Immunosuppressive therapy alone Günaydin I, Ustündağ C, Kaner G, et al The prevalence of Sjögren’s
versus immunosuppressive therapy plus anticoagulation. Clin syndrome in Behçet’s syndrome. J Rheumatol 1994; 21:1662–1664
Rheumatol 2008; 27:201–205 Hamuryudan V, Ozyazgan Y, Hizli N, et al Azathioprine in Behçet’s
Aktulga E, Altac M, Muftuoglu A, et al A double blind study of colchi- syndrome: Effects on long-term prognosis. Arthritis Rheum 1997;
cine in Behçet’s disease. Haematologica 1980; 65:399–402 40:769–774
238 H. Yazici and Y. Yazici

Hamuryudan V, Er T, Seyahi E, et al Pulmonary artery aneurysms in Seyahi E, Memisoglu E, Hamuryudan, et al Coronary atherosclero-
Behçet syndrome. Am J Med 2004; 117:867–870 sis in Behçet’s syndrome: A pilot study using electron-beam
Hatemi G, Bahar H, Uysal S, et al The pustular skin lesions in Behçet’s computed tomography. Rheumatology (Oxford) 2004; 43:
syndrome are not sterile. Ann Rheum Dis 2004; 63:1450–1452 1448–1450
Hatemi G, Fresko I, Tascilar K, et al Increased enthesopathy among Sakane T, Takeno M, Suzuki N, et al Behçet’s disease. N Engl J Med
Behçet’s syndrome patients with acne and arthritis: An ultrasonog- 1999; 341:1284–1291. Arthritis Rheum 2008 (in press)
raphy study. Arthritis Rheum 2008; 58:1539–1545 Siva A, Yazici H. Behcet’s Disease. In: Levine S, Doruk E (eds)
Hirohata S. Histopathology of central nervous system lesions in Handbook of systemic autoimmune diseases, neurology volume
Behçet’s disease. J Neurol Sci 2008; 267:1–2 [The Neurologic Involvement in Systemic Autoimmune Disorders].
Inanc N, Mumcu G, Birtas E, et al Serum mannose-binding lectin levels New York, Elsevier Science, 2004, pp. 193–216
are decreased in Behçet’s disease and associated with disease sever- Tunc R, Keyman E, Melikoglu M, et al Target organ associations in
ity. J Rheumatol 2005; 32:287–291 Turkish patients with Behçet’s disease: A cross sectional study
International Study Group for Behcet’s disease. Criteria for diagnosis by exploratory factor analysis. J Rheumatol 2002; 29:
of Behcet’s disease. Lancet 1990; 335:1078–1080 2393–2396
Kahraman O, Celebi-Onder S, Kamali S, et al Long-term course of Tunc R, Saip S, Siva A, et al Cerebral venous thrombosis is associated
deep venous thrombosis in patients with Behçet’s disease. Arthritis with major vessel disease in Behçet’s syndrome. Ann Rheum Dis
Rheum 2003; 48 (Suppl):S385 2004; 63:1693–1694
Kural-Seyahi E, Fresko I, Seyahi, et al The long-term mortality and Tüzün B, Tüzün Y, Yurdakul, et al Neutrophil chemotaxis in Behçet’s
morbidity of Behçet syndrome: AN 2-decade outcome survey of syndrome. Ann Rheum Dis 1999; 58:658
387 patients followed at a dedicated center. Medicine (Baltimore) Ugurlu S, Seyahi E, Yazici H. Prevalence of angina, myocardial infarc-
2003; 82:60–76 tion and intermittent claudication assessed by Rose Questionnaire
Mahr A, Guillevin L, Poissonnet M, et al Prevalences of polyarteritis among patients with Behçet’s syndrome. Rheumatology (Oxford)
nodosa, microscopic polyangiitis, Wegener’s granulomatosis, and 2008; 47(4):472–475
Churg-Strauss syndrome in a French urban multiethnic population in Yazici Y, Moses N. Clinical manifestations and ethnic background of
2000: A capture-recapture estimate. Arthritis Rheum 2004; 51:92–99 patients with Behçet’s syndrome in a US cohort. Arthritis Rheum
Mahr A, Belarbi L, Wechsler W, et al Population-based study of 2007; 56:S502
Behçet’s disease: Differences by ethnic origin and low variation by Yazici H, Pazarli H, Barnes CG, et al A controlled trial of azathio-
age at migration. Arthritis Rheum 2008; 58:3951-9 (in press) prine in Behçet’s syndrome. N Engl J Med 1990; 322:
Pipitone N, Boiardi L, Olivieri, et al Clinical manifestations of Behçet’s 281–285
disease in 137 Italian patients: Results of a multicenter study. Clin Yurdakul S, Mat C, Tuzun Y, et al. Double-blind trial of colchicine in
Exp Rheumatol 2004; 22(Suppl 36):S46–S51 Behçet’s syndrome. Arthritis Rheum 2001; 44:2686–2692
 The Churg–Strauss Syndrome
23
Andrew Churg, Ulrich Specks, and John H. Stone

Pearl: CSS tends to unfold in three stages.

23.1 Overview of the Churg–Strauss
Syndrome Comment: It is often possible to recognize three distinct
phases of CSS (Lanham et al. 1984):

› The Churg–Strauss Syndrome (CSS) is a form of • A prodromal phase characterized by allergic disease (typ-
necrotizing vasculitis characterized by eosinophilic ically asthma or allergic rhinitis). This phase may last
infiltration of small- and medium-sized blood vessels. from months to many years.
› The initial disease manifestation is usually the devel- • An eosinophilia/tissue infiltration phase, in which remark-
opment of asthma in an individual who did not suffer ably high peripheral eosinophilia may occur and tissue
previously from reactive airway disease. Allergic rhin- infiltration by eosinophils is observed in the lung, gastro-
itis is another manifestation of atopy, often identified intestinal tract, and other tissues.
in retrospect as the first sign of CSS. • A vasculitic phase, in which systemic vasculitis afflicts a
› The extra-pulmonary organs commonly involved in wide range of organs, ranging from the heart and lungs to
the CSS are the peripheral nerves, skin, joints, and the peripheral nerves and skin.
heart. The brain, eyes, gastrointestinal tract, and kid- With the onset of the vasculitic phase of CSS, patients’
neys are involved less often. asthma may improve substantially, even before therapy for
› Vasculitic neuropathy occurs in 80% of patients with vasculitis is begun.
CSS and can lead to devastating complications through
paralysis of the distal extremities. Nerve infarctions Myth: CSS requires the combination of three histopathologic
generally lead to an asymmetric, axonal sensorimotor findings for confirmation of the diagnosis: vasculitis, granu-
neuropathy. lomas, and eosinophils in the tissue.
› The clinical features of CSS overlap in many ways on Reality: CSS granulomas were present in all patients reported
those of two other forms of necrotizing vasculitis, in the original case series published in 1951 (Churg and
Wegener’s granulomatosis, and microscopic poly- Strauss 1951). However, this publication preceded the wide-
angiitis. However, patients with CSS are less likely to spread availability of glucocorticoids. Glucocorticoid ther-
have antineutrophil cytoplasmic antibodies (ANCA) apy has altered both the clinical and histopathological
than those with other conditions. presentation of CSS: fewer than half of the cases in which
› When patients with CSS are ANCA-positive, the antigen the diagnosis is confirmed by either tissue biopsy or autopsy
specificity is directed more often against myeloperoxidase have granulomas (Lanham et al. 1984).
than against serine proteinase-3. Anti-myeloperoxidase In the early phase of CSS, tissue biopsies often reveal
ANCA produces a perinuclear (“P-ANCA”) pattern of eosinophilic infiltration but neither granulomas nor vasculi-
staining in immunofluorescence studies of serum. tis (Churg 2005). The absence of histopathological evidence
› CSS must be distinguished from a group of non-vasc- of vasculitis correlates with a good response to treatment.
ulitic conditions associated with hypereosinophilia.
Myth: Necrotizing vasculitis on biopsy is the sine qua non of
Among these are the hypereosinophil syndrome,
CSS.
eosinophilic leukemia, eosinophilic fasciitis, and para-
sitic infections. Reality: The morphologic appearance of vasculitis in CSS is
highly variable. Some cases demonstrate necrotizing vasculitis

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 239

DOI:10.1007/978-1-84800-934-9_23, © Springer Science + Business Media B.V. 2009
240 A. Churg et al.

initially as having simple reactive airway disease. Leukotriene

inhibitors often permit the tapering of prednisone and, in a
minority of patients with “asthma”, to the unmasking of CSS
(Wechsler et al. 2000).
CSS-related vasculitis recurs in only about 25% of all
patients. Unfortunately, after the vasculitic phase of CSS has
been treated effectively, most CSS patients are left with glu-
cocorticoid-dependent asthma and are unable to taper their
prednisone below 8 mg/day (Pagnoux et al. 2007). Managing
this refractory reactive airway disease is a major challenge in
caring for CSS patients over the long-term. Some clinicians
who are also experienced in the care of CSS patients are suf-
ficiently comfortable with the fact that leukotriene inhibitors
do not cause the disease, and they actually use these agents
to treat glucocorticoid-dependent asthma of CSS (Keogh and
Fig. 23.1 Classic necrotizing vasculitis in the heart in a fatal case of the Specks 2003).
Churg–Strauss syndrome. Note the eosinophils infiltrating the tissue
(Figure courtesy of Dr. Andrew Churg)
Pearl: CSS often involves lymphadenopathy.
Comment: Godman and Churg noted that CSS, Wegener’s
granulomatosis, and microscopic polyangiitis are linked
(Godman and Churg 1954). These three diseases, observed
those investigators, “group themselves into a compass, (rang-
ing from) necrotizing and granulomatous processes with
angiitis to vasculitis without granulomata” (with mixed
forms in between). These disorders share intriguing similari-
ties, but also have features that mark them individually.
One interesting difference between CSS and the other
forms of ANCA-associated vasculitis is the frequency with
which CSS is associated with lymphadenopathy (Churg
2001), which is rarely a feature of either WG or MPA. CSS
can also cause striking salivary gland enlargement (Boin et
al. 2006).
Myth: CSS is an “ANCA-associated vasculitis”.
Reality: Despite the undeniable connections between CSS,
Fig. 23.2 Non-necrotizing vasculitis in a small vessel in the lung in the
Churg–Strauss syndrome. Collections of eosinophils are also present in Wegener’s granulomatosis, and microscopic polyangiitis, a
the airspaces (the biopsy has caught the edge of an eosinophilic pneu- substantial percentage of CSS patients – more than 50% in
monia-like area) (Figure courtesy of Dr. Andrew Churg) some studies – are ANCA negative (Guillevin et al. 1999).
To some extent, the ANCA status of CSS patients may
(Fig. 23.1), but others document only the infiltration of eosino- depend on when the testing is performed during the course of
phils into the vascular walls (Fig. 23.2). In a large series of therapy (Keogh and Specks 2003), but in general, it seems
patients from France, only half of the cases proved histopatho- that CSS patients are significantly less likely to be ANCA
logically had necrotizing vasculitis (Guillevin et al. 1999). positive than those with WG or MPA. ANCA negativity
Insistence on the presence of necrotizing vasculitis for the should not dissuade one from the diagnosis of CSS if other
label of CSS leads to under-diagnosis. clinical findings fit.
The presence or absence of ANCA may also influence clin-
Myth: CSS is caused by the use of leukotriene inhibitors.
ical phenotype in CSS, but the literature is mixed on this point.
Reality: This Myth, approximately as old as leukotriene inhib- In one study (Keogh and Specks 2003), central nervous sys-
itors themselves, is a classic case of confounding by indica- tem involvement was more common among ANCA-positive
tion. Leukotrienes are highly effective asthma medications patients (20 vs. 5%). In another (Sablé-Fourtassou et al. 2005),
and permit the taper of prednisone in glucocorticoid-depen- ANCA-positive patients were significantly more likely to have
dent asthma. The majority of patients with CSS have asthma glomerulonephritis and neurologic disease, but ANCA-negative
as their presenting feature and are usually misdiagnosed patients were more likely to have cardiac involvement. ANCA
The Churg–Strauss Syndrome 241

Fig. 23.4 Capillaritis in the Churg–Strauss syndrome, associated with

hemorrhage. In this example, there is not only infiltration of the alveolar
walls by eosinophils, but considerable eosinophil degranulation as well
Fig. 23.3 Computed tomographic scan of the lung revealing diffuse
(Figure courtesy of Dr. Andrew Churg)
pulmonary infiltrates caused by alveolar hemorrhage in a patient with
the Churg–Strauss syndrome (Figure courtesy of Dr. John H. Stone)

status did not appear to affect either mortality or the likeli- Myth: The finding of neutrophils on tissue biopsy is not a fea-
hood of relapse. ture of CSS vasculitis and should suggest another diagnosis.
Reality: Neutrophils can be observed in some cases of
Myth: Pulmonary capillaritis and diffuse alveolar hemor- CSS, particularly those complicated by leukocytoclastic
rhage do not occur in CSS. vasculitis of the skin. However, such neutrophils are usu-
Reality: The frequency with which pulmonary capillaritis ally mixed with a liberal sprinkling or even an abundance
and lung hemorrhage occur in CSS is a point of some conten- of eosinophils.
tion. These complications have been documented to occur in Myth: CSS always involves the lung parenchyma.
CSS (Fig. 23.3), but they are substantially more common in
Wegener’s granulomatosis and microscopic polyangiitis than Reality: Although almost all CSS patients have asthma, a
in CSS. In one series of 96 patients, alveolar hemorrhage significant fraction of patients have no other indication of
occurred in three (Guillevin et al. 1999). pulmonary disease. Approximately 30% of the patients
Some data suggest that ANCA positivity is a risk factor reported by Lanham had no pulmonary disease apart from
for alveolar hemorrhage in CSS: In a series of 93 patients, reactive airway disease (Lanham et al. 1984). These findings
seven of 35 patients with CSS (20%) who were ANCA- were confirmed in a larger study from France, in which 27%
positive had alveolar hemorrhage, but none of the 58 patients of the cohort had no lung disease other than asthma (Pagnoux
who were ANCA-negative experienced this complication et al. 2007). Thus, the absence of lung involvement does not
(Sinico et al. 2005). However, in contrast to these results, a detract from the diagnosis.
series of 99 patients reported no cases of diffuse alveolar
Pearl: Glucocorticoid therapy can alter the histopathologic
hemorrhage despite the fact that 70% were ANCA positive if
findings in CSS lung substantially.
tested before initiation of immunosuppressive therapy (Keogh
and Specks 2003). Comment: Glucocorticoid therapy suppresses the tissue infil-
Thus, the largest series of CSS patients to date indicate tration of eosinophils and often renders the diagnosis of CSS
that alveolar hemorrhage is less common in this disorder difficult on the basis of pathologic features alone. In this situ-
than in Wegener’s granulomatosis and microscopic poly- ation, vasculitis without eosinophils and with or without
angiitis. The reason for this must relate in part to the differ- necrosis can be observed. These findings are difficult to dis-
ing cellular nature of these diseases’ inflammatory infiltrates. tinguish from Wegener’s granulomatosis, microscopic poly-
The inflammatory infiltrate that invades small blood-vessel angiitis, vasculitis associated with a collagen vascular disease,
walls in CSS can be purely eosinophilic (Fig. 23.4). This is or other forms of vasculitis. Close clinico-pathological cor-
not observed in Wegener’s granulomatosis and microscopic relation, with questions about the patient’s history of asthma,
polyangiitis, although substantial eosinophilic infiltration allergic rhinitis, or blood eosinophilia, points the way to the
may accompany either of those disorders. correct diagnosis.
242 A. Churg et al.

Pearl: Patients with CSS may develop “rheumatoid nodules”.

Comment: Early in the disease course, CSS (as well as
Wegener’s granulomatosis) can be confused with rheumatoid
arthritis. There are several reasons for this:
• Inflammatory arthritis, particularly involving large joints
in a migratory fashion, is a common presentation of both
CSS and WG.
• CSS and WG are often associated with positive serologi-
cal assays for rheumatoid factor.
• Patients with CSS and WG often develop lesions over the
extensor surfaces of the elbows that strongly resemble
rheumatoid nodules (Fig. 23.5).
The occurrence of these lesions, sometimes termed “Churg–
Strauss granulomas” was described by Dr. Jacob Churg and
Dr. Lotte Strauss themselves in the earliest description of
CSS (Churg and Strauss 1951). These lesions, also called
cutaneous extravascular necrotizing granulomas, are histo-
pathologically identical to the extravascular nodules reported
by Churg and Strauss within other tissues in CSS (e.g., the
Fig. 23.5 “Churg–Strauss granulomas”, also known as cutaneous
heart, lungs, or spleen). They can be mistaken clinically and extravascular necrotizing granulomas, in the Churg–Strauss syndrome
pathologically for rheumatoid nodules. Over the extensor surface of the elbow (Figures courtesy of
The misclassification of Churg–Strauss granulomas as Dr. John Stone)
rheumatoid nodules sometimes leads to severe disease mani-
festations such as mononeuritis multiplex or rapidly progres-
sive GN before the correct diagnosis is established. (Mathew et al. 2007). In 49 of these cases (58%), the diagno-
Pearl: CSS has a striking predilection for the peripheral sis of vasculitic neuropathy was confirmed. Thirty-one of the
nerves and myocardium. remaining biopsies demonstrated pathological findings “con-
sistent with” but not diagnostic of vasculitic neuropathy; i.e.,
Comment: Among all of the systemic vasculitides, none has acute asymmetric axonal loss with differential fascicular
a greater tendency than CSS to cause mononeuritis multi- involvement, with or without active Wallerian degeneration
plex. The vasculitic neuropathy in CSS can be devastating. or regenerating fibers. Thus, 80 of 84 procedures (95%) led
Recovery from severe cases, if recovery occurs, requires to findings compatible with or diagnostic of vasculitic neu-
more than a year and is usually incomplete. Milder cases ropathy. In the proper clinical context, this is all the proof a
may improve more quickly, but the process is unpredictable clinician needs.
and many patients have persistent distal sensory deficits that
are troubling. These represent damage, not active disease, Myth: Nerve/muscle biopsies performed at the site of previ-
but sometimes lead to excessive treatment. ous electromyography/nerve conduction studies (EMG/NCS)
CSS is also distinguished (in a notorious sense) for its may be difficult to interpret because of artifact.
tendency to damage the myocardium, often fatally. In one Reality: This is a Myth, as it relates to vasculitis and to NCS.
study from France, congestive heart failure was a leading The statement is closer to being correct with regard to inflam-
cause of death among patients with CSS (Guillevin et al. matory myopathies and EMG studies.
1999). Compared with its cousins, Wegener’s granulomato- NCS are critical to identifying nerves involved in vascu-
sis and microscopic polyangiitis, CSS is far more likely to litic neuropathy; subsequent biopsy may confirm the diagno-
cause clinically-evident cardiac disease. sis. NCS studies do not involve needles and do not produce
any artifacts that confound the diagnosis of vasculitis. In
Pearl: Although peripheral nerve biopsy is not always diag-
contrast, EMG needles cause injury to muscle and lead to
nostic in peripheral neuropathy, it nearly always tells you
inflammation, which may complicate the identification of a
the information you need to know.
primary inflammatory myopathy. However, because of the
Comment: In one study of 106 patients with vasculitic neu- starkly contrasting pathologies between vasculitis and mus-
ropathy, 84 peripheral nerve biopsies were performed (82 cle inflammation, previous EMGs should not pose a problem
sural nerve biopsies, 2 superficial radial nerve biopsies) to the histopathological identification of vasculitis.
The Churg–Strauss Syndrome 243

(Rothenberg et al. 2008; Wechsler 2008). Others respond

well to imitinab (Dahabreh et al. 2007). A third subset
requires hydroxyurea to lower the eosinophil count.
Myth: CSS never requires treatment with cytotoxic agents.
Reality: Although CSS is more likely than WG or MPA to
respond adequately to glucocorticoids alone, a substantial
number of patients with CSS do require cyclophosphamide.
The complications of CSS most likely to require cyclophos-
phamide are vasculitic neuropathy, glucocorticoid-refractory
glomerulonephritis, myocardial disease, severe gastrointesti-
nal ischemia, and central nervous system disease. Confronted
with any of these complications, it is probably better to treat
aggressively with cytotoxic therapy from the outset.
Fig. 23.6 Tongue wasting in a patient with CSS (Figure courtesy of
Dr. John Stone) Myth: Renal disease in CSS, when present, is mild.
Reality: Renal disease in CSS is typically less severe than
Sural nerve biopsies are tolerated well by most patients. that encountered in WG, but segmental necrotizing glomeru-
However, some patients experience several months of numb- lonephritis leading to a rapid decline in renal function is well-
ness or paresthesias. If the sensory symptoms of vasculitic neu- recognized (Clutterbuck et al. 1990). On renal biopsy, most
ropathy are severe, patients scarcely notice the biopsy afterwards. of these patients have crescentic lesions and other findings
The incidence of permanent discomfort following a sural nerve indistinguishable from the renal pathology of its cousins,
biopsy is probably about 5%. WG and MPA. Some CSS patients have marked eosinophilic
interstitial nephritis in addition to the glomerular lesions.
Myth: Nerve infarction in CSS signals the need to intensify
treatment. Myth: Persistent asthma represents persistent CSS disease
activity.
Reality: Active vasculitic neuropathy is one of the strongest
indications for the use of cyclophosphamide in CSS. However, Comment: In the absence of helpful data, this remains a topic
the development of a new nerve infarction shortly after start- of contentious debate. The vasculitic disease manifestations
ing immunosuppression does not necessarily signal the need of CSS usually respond readily to remission induction therapy
to intensify treatment, particularly if other indicators of dis- with immunosuppressive regimens used for AAV. However,
ease activity suggest disease control (Fig. 23.6). For example, once these symptoms are controlled and glucocorticoid doses
nerve infarctions may manifest themselves even several have been reduced, patients are frequently left with difficult to
weeks after the start of appropriate treatment (high-dose glu- manage asthma, and they remain in need of significant sys-
cocorticoids and cyclophosphamide), with the abrupt occur- temic glucocorticoid doses to control the asthma. It remains
rence of a foot or wrist drop or other nerve lesion. unclear whether this truly represents persistent CSS as opposed
In addition, the appearance of muscle wasting secondary to “simple” asthma. Long-term use of azathioprine or metho-
to nerve infarctions may continue for weeks after the correct trexate may provide some glucocorticoid-sparing benefit in
therapy is initiated. This is also not a sign of treatment failure patients with severe residual reactive airway disease.
and should not trigger an intensification of therapy.
Pearl: Compared with CSS, the eosinophil count in the
hypereosinophil syndrome (HES) is more likely to be refrac- References
tory to glucocorticoid therapy.
Comment: CSS and HES are difficult to differentiate. The Boin F, Sciubba JJ, Stone JH. Churg-Strauss syndrome presenting with
two disorders have similar organ tropisms (heart, skin, periph- salivary gland enlargement and respiratory distress. Arthritis Rheum
2006; 55:167–170
eral nerves), and both are associated with striking eosino-
Churg A. Recent advances in the diagnosis of Churg-Strauss syndrome.
philia. Making the distinction is crucial, however, because the Mod Pathol 2001; 14:1284–1293
treatments and prognoses of these two disorders differ sub- Churg A. Churg-Strauss syndrome, microscopic polyangiitis, and other
stantially. While the eosinophil count in CSS usually drops forms of pulmonary vasculitis. In: Churg A, Myers J, Tazelaar H,
Wright JL (eds) Thurlbeck’s pathology of the lung (3rd edn). New
dramatically upon the institution of glucocorticoids, HES can
York, Thieme Medical Publishers, 2005, pp. 391–412
be refractory to this approach. A subset of patients with HES Churg J, Strauss L. Allergic granulomatosis, allergic angiitis, and peri-
responds well to an anti-interleukin-5 monoclonal antibody arteritis nodosa. Am J Pathol 1951; 27:277–301
244 A. Churg et al.

Clutterbuck E, Evans D, Pusey C. Renal involvement in Churg-Strauss Pagnoux C, Guilpain P, Guillevin L. Churg-Strauss syndrome. Curr
syndrome. Nephrol Dial Transplant 1990; 5:161–167 Opin Rheumatol 2007; 19:25–32
Dahabreh IJ, Giannouli S, Zoi C, et al Management of hypereosino- Rothenberg ME, Klion AD, Roufosse FE, et al Treatment of patients
philic syndrome: A prospective study in the era of molecular genet- with the hypereosinophilic syndrome with mepolizumab. N Engl J
ics. Medicine (Baltimore) 2007; 86(6):344–354 Med 2008; 358(12):1215–1228
Godman G, Churg J. Wegener’s granulomatosis: Pathology and review Sablé-Fourtassou R, Cohen P, Mahr A, et al Antineutrophil cytoplasmic
of the literature. Arch Pathol Lab Med 1954; 58:533–551 antibodies and the Churg-Strauss syndrome. Ann Intern Med 2005;
Guillevin L, Cohen P, Gayraud M, et al Churg-Strauss syndrome: 143:632–638
Clinical study and long-term follow-up of 96 patients. Medicine Sinico RA, Di Toma L, Maggiore U, et al Prevalence and clinical sig-
(Baltimore) 1999; 78:26–37 nificance of antineutrophil cytoplasmic antibodies in Churg-Strauss
Keogh KA, Specks U. Churg-Strauss syndrome: Clinical presentation, syndrome. Arthritis Rheum 2005; 52: 2926–2935
antineutrophil cytoplasmic antibodies, and leukotriene receptor Wechsler ME. Combating the eosinophil with anti-interkeukin-5 ther-
antagonists. Am J Med 2003; 115:284–290 apy. N Engl J Med 2008; 358(12):1293–1294
Lanham JG, Elkon KB, Pusey CD, Hughes GRV. Systemic vasculitis Wechsler ME, Finn D, Gunawardena D, et al Churg-Strauss syndrome
with asthma and eosinophilia: A clinical approach to the Churg- in patients receiving montelukast as treatment for asthma. Chest
Strauss syndrome. Medicine (Baltimore) 1984; 63:65–81 2000; 117:708–713
Mathew L, Talbot K, Love S, et al Treatment of vasculitic peripheral neu-
ropathy: A retrospective analysis of outcome. Q J Med 2007; 100:41
 ANCA-Associated Vasculitis
24
John H. Stone, Shoichi Ozaki, Karina Keogh, Ulrich Specks,
Carol A. Langford, Niels Rasmussen, Cees G.M. Kallenberg,
and Ingeborg M. Bajema

The frequencies of clinical manifestations of WG in a clinical

24.1 Overview of ANCA-Associated trial cohort of 180 patients are shown in Table 24.1.
Vasculitis

› The majority of patients with Wegener’s granulomatosis

24.2 Epidemiology
(WG) or microscopic polyangiitis (MPA) have antineu-
trophil cytoplasmic antibodies (ANCA) in their serum.
This is particularly true of patients with “disseminated”
Pearl: WG is believed to be two or three times more common
disease, the great majority of whom are ANCA positive.
than MPA in western countries. This ratio is inverted in
› WG and MPA are often termed “ANCA-associated
Japan.
vasculitides” (AAV), even though not all patients with
these conditions have ANCA. The Churg–Strauss syn- Comment: Epidemiologic studies from Europe indicate that
drome, another disorder classified as an AAV, is dis- the annual incidence of WG is approximately three times
cussed in Chap. 23. higher than that of MPA. These figures are concordant with
› Multiple antibodies may lead to positive immunofluores- the experience of clinicians in the United States. For reasons
cence testing for ANCA in either perinuclear (P-ANCA) that remain unclear, the ratio of these two diseases is flipped
or cytoplasmic (C-ANCA) patterns. However, only anti- in Japan, where MPA appears to be substantially more com-
bodies to myeloperoxidase (MPO) and proteinase-3 mon than WG. Japanese clinicians are far more likely to
(PR3) are associated with the AAV. evaluate patients with MPO-ANCA than PR3-ANCA.
› Antibodies directed against PR3 and MPO are termed
PR3-ANCA and MPO-ANCA, respectively.
› WG may be associated with destructive upper respira-
tory tract disease, including saddle-nose deformity, 24.3 Nose, Sinuses, and Ears
erosive sinusitis, and subglottic stenosis.
› A host of ocular lesions may occur in the AAV, includ- Myth: The saddle nose is a diagnostic sign of active WG.
ing episcleritis, scleritis, peripheral ulcerative keratitis,
and orbital pseudotumor. Most of these lesions are Reality: Yes and no. Active WG may lead to the well-known
more common in WG than in MPA. saddle-nose deformity (Fig. 24.1a, b) due to destruction of
› Lung disease in the AAV ranges from nodular lesions the nasal cartilages, especially the nasal septum. The nasal
with a tendency to cavitate (in WG), to interstitial lung septum is particularly vulnerable to vasculitis as its blood
disease (MPA), and to alveolar hemorrhage (both WG supply is derived from terminal arteries.
and MPA). Ironically, the partially effective therapies available for
› Segmental, necrotizing glomerulonephritis commonly WG over the past few decades have led to an alternative route
accompanies the AAV. Because of the paucity of by which patients can develop saddle nose deformities. This
immunoreactants such as immunoglobulins and com- is observed in patients who have constant pain localized to
plement components in kidney biopsies relative to the the nose and midfacial area. The examination of these
biopsies from patients with immune complex-mediated patients is remarkable for crust formation and possibly super-
glomerulonephritis, the glomerulonephritis of the AAV infection, but a septal perforation is absent.
is termed “pauci-immune.” The “treatment-induced” saddle-nose deformity comes
about in the following manner. Once it becomes clear that

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 245

DOI:10.1007/978-1-84800-934-9_24, © Springer Science + Business Media B.V. 2009
246 J. H. Stone et al.

Table 24.1 Clinical features of limited and severe Wegener’s granulomatosis in a clinical trial cohort of 180 patientsa (Modified from Stone and
the WGET Research Group 2003)
Limited (n = 52) Severe (n = 128) Overall (n = 180) P-valueb
General (%) 76.9 70.3 72.2 0.370
Arthralgias/Arthritis 69.2 64.8 66.1
Fevers 7.7 21.1 17.2
Cutaneous (%) 15.4 21.9 20.0 0.324
Purpura 13.5 18.8 17.2
Ulcers 3.9 3.1 3.3
Gangrene 0.0 0.8 0.6
Mucous membranes/eyes (%) 25.0 27.3 26.7 0.747
Mouth ulcers 7.7 11.7 10.6
Conjunctivitis/episcleritis 13.5 14.8 14.4
Retro-orbital mass/proptosis 1.9 4.7 3.9
Uveitis 0.0 0.0 0.0
Scleritis 1.9 2.3 2.2
Ear/Nose/Throat (%) 88.5 71.9 76.7 0.017
Bloody nasal discharge/nasal rusting/ulcer 67.3 56.3 59.4
Sinus involvement 59.6 41.4 46.7
Swollen salivary gland 1.9 3.1 2.8
Subglottic inflammation 11.5 7.0 8.3
Conductive deafness 25.0 22.7 23.3
Sensorineural deafness 1.9 3.1 2.8
Cardiovascular (%) 1.9 1.6 1.7 0.864
Pericarditis 0.0 0.8 0.6
Gastrointestinal (%) 1.9 2.3 1.7 0.862
Mesenteric ischemia 0.0 0.8 0.6
Pulmonary (%) 51.9 63.3 60.0 0.159
Pleurisy 7.7 7.0 7.2
Nodules or cavities 38.5 28.1 31.1
Other infiltrates 23.1 32.0 29.4
Endobronchial involvement 3.9 3.1 3.3
Alveolar hemorrhage 0.0 25.0 17.8
Respiratory failure 0.0 2.3 1.7
Renal (%) 28.9 64.1 53.9 <0.001
Hematuria 25.0 28.9 27.8
RBC casts 3.9 35.9 26.7
Elevated serum creatinine 3.9 28.1 21.1
Nervous system (%) 0.0 12.5 8.9 0.017
Meningitis 0.0 0.0 0.0
Cord lesion 0.0 0.0 0.0
Stroke 0.0 0.0 0.0
Cranial nerve palsy 0.0 2.3 1.7
Sensory peripheral neuropathy 0.0 10.2 7.2
Motor mononeuritis multiplex 0.0 2.3 1.7
Other (%)c 46.2 39.1 41.1 0.244
a
Organ involvement at enrollment, as recorded on the Birmingham Vasculitis Activity Score for Wegener’s Granulomatosis (BVAS/WG) Form
b
Chi-square or Fisher’s Exact Test for categorical data
c
Include lacrimal gland involvement (3 patients), supraglottic airway inflammation (2), aural chondritis (2), breast mass (2), peripheral 7th cranial
nerve lesion (2), keratitis (2), optic neuropathy mimicking optic neuritis (1), splenic involvement - asymptomatic infarctions found on computed
tomography (1), central nervous system involvement with parenchymal brain lesions and seizures (1), skull-based lesion in the infra-temporal
fossa, misdiagnosed as a tumor (1), diabetes insipidus (1), pituitary involvement (1), hilar adenopathy (1), tongue ulcer (1), nodular vocal cord
lesion (1), and hepatic transaminase elevation (1)a

the patient has active nasal disease and antiinflammatory unfortunate, because the patients are generally encouraged
treatment is commenced, the nasal septum can disappear by this point that their disease is treatable and that the
rapidly—within days to weeks. This is probably due much-dreaded complication of a saddle-nose deformity can
to the absorption of inflamed tissue and is profoundly be avoided.
24 ANCA-Associated Vasculitis 247

Fig. 24.2 Nasal crusting in Wegener’s granulomatosis (Figure courtesy

a of Dr. Robert Lebovics)

often complicated by serous otitis media due to Eustachian

tube involvement, and an endoscope is necessary to make the
diagnosis.
Pearl: For patients with nasal crusting, irrigation of the nose
with saline is a cornerstone of therapy.
Comment: There is a spectrum of responses to treatment in
nasal disease. Some patients’ noses clear up completely with
immunosuppressive therapy. Others settle into a pattern of
chronic nasal crusting that responds poorly to immunosup-
pressive therapy and is often difficult to parse according to
active disease or damage (Fig. 24.2). Such patients often find
that irrigation of the nose, performed by themselves once or
twice daily, is the most effective intervention available for
nasal crusting.
Isotonic saline is the preferred solution for nasal irriga-
tion. WG patients devise ingenious solutions for irrigating
b their noses effectively, but generally an eyedropper or water
pick is sufficient (Fig. 24.3). In addition to irrigation, antimi-
Fig. 24.1 Saddle-nose deformity repair, before (a) and after (b) crobial therapy with trimethoprim-sulfamethoxazole or the
(Figures courtesy of Dr. John Stone) local application of antimicrobial substances, such as mupi-
rocin, has dramatic effects on crusting and rhinorhea in some
patients.
Pearl: To detect a nasal septal perforation, light is shone into
one of the patient’s nostrils through a nasal speculum while Myth: Surgery on the nose or upper respiratory tract in WG
peering into the other nostril. may trigger a disease flare.
Comment: This is a simple test if the nose is relatively free of Reality: This contention is based entirely on anecdote. If sur-
crusts. The light streams through the nasal septal defect if a gery on the upper respiratory tract actually does increase the
perforation is present allowing the diagnosis to be made eas- likelihood of a disease flare, the probable reason could be the
ily. If there are substantial crusts, bleeding, or purulent secre- preoperative tapering of patients’ glucocorticoids that occurs
tions, the nose must be cleaned (generally by an due to wound healing concerns. Thus, in patients whose WG
otolaryngologist) before a full examination is possible. has been quiescent, particularly those who have been in remis-
Inflammatory changes that are localized to the posterosu- sion on little or no immunosuppression, there is no reason to
perior part of the nasal cavity can cause a septal perforation forgo elective repairs of saddle-nose deformities or subglottic
in the absence of a saddle nose deformity. Such cases are stenoses.
248 J. H. Stone et al.

a Cases of WG “smoldering” in the upper respiratory tract,

undiagnosed for years, are documented well in the literature.
In many cases, the explanation for the chronic sinus dysfunc-
tion, nasal pain, bleeding, recurrent “otitis media,” and other
symptoms is understood only in retrospect, upon the occur-
rence of disease manifestions that threaten vital organs. The
detection of ANCA can shorten the time to diagnosis by
many months once the clinician thinks of ordering this test.
Pearl: Grommets (tubes) are helpful in WG-associated serous
otitis media.
Comment: Yes—in the correct setting. In the initial disease
phase, serous otitis media often develops because of inflam-
matory involvement of the Eustachian tube, leading to
“Eustachian tube dysfunction.” These acute symptoms often
b
resolve without a mechanical intervention and are associated
with a return of the hearing to normal after treatment with
systemic immunosuppressive medications. Thus, the grom-
met should be spared in this phase.
Chronic serous otitis media, which occurs in later phases
of WG, may benefit from the placement of grommets.
Adequate audiometry testing before grommet placement is
essential to ensure that the patient’s hearing loss, as antici-
pated, is conductive rather than sensorineural in nature. (Tubes
will not ameliorate sensorineural hearing loss.)
In cases of chronic serous otitis media, only a permanent
T-tube can provide substantial benefit. Such tubes might have
to be removed in cases of chronic infections. When that hap-
pens, permanent perforations of the eardrum may ensue.
Bone-anchored hearing aids may be the ultimate solution.

Fig. 24.3 (a, b) One patient’s device for irrigating his nose (Figures
courtesy of Dr. John Stone)
24.4 Subglottic Stenosis
Ensuring that the disease is quiet both locally and sys-
temically before surgery is important. Poor surgical out-
Pearl: Dyspnea on exertion in a WG patient whose disease
comes, including severe fistulas, have occurred when nasal
has been quiescent suggests subglottic stenosis.
reconstruction is performed during a period of active disease.
The cosmetic results of saddle-nose repairs can be spectacu- Comment: When WG patients present with decreased exer-
lar (see Fig. 24.1a, b). Saddle-nose deformities may recur in cise tolerance, dyspnea on exertion, “asthma,” or other simi-
the setting of a disease reactivation, but secondary and even lar complaints, clinicians must remember the propensity of
tertiary repairs can be performed once disease control is WG to cause subglottic stenosis (see below).
restored. Subglottic stenosis can develop even in patients who have
been in apparent remission (Fig. 24.4). The reason for this
Pearl: WG that is limited to the upper respiratory tract can
can be either subclinical disease activity, progression to sub-
remain subclinical or be misdiagnosed as a series of innocu-
glottic stenosis because of scarring rather than active disease,
ous ailments for many months or even years.
or both.
Comment: WG should be suspected when mundane prob-
Pearl: Granulomatous subglottic stenosis responds easily to
lems recur with excessive frequency or when they persist too
immunosuppressive treatment when it is an early manifesta-
long. Friedrich Wegener reported three patients in their 30s
tion of WG.
who presented with upper respiratory tract symptoms that
had appeared harmless at first (Wegener 1936). Within 6 Comment: Subglottic stenosis is one of the more mysteri-
months, all three patients had died of renal failure. ous manifestations of WG. As illustrated in Fig. 24.5a, b,
24 ANCA-Associated Vasculitis 249

Fig. 24.4 Subglottic Confirming the suspicion of subglottic stenosis can be

stenosis in Wegener’s challenging. The suspicion of a subglottic stenosis can be
granulomatosis
(Figure courtesy of confirmed by endoscopic laryngoscopy using the flexible lar-
Dr. John Stone) yngoscope or a telescope as used for videostroboscopy. Thin-
cut computed tomography scans of the trachea can be helpful
to evaluate the extent of the stenosis, particularly if a surgical
correction is planned.
Pearl: Pulmonary function testing with an inspiratory flow
volume curve is helpful in the diagnosis and monitoring of
large airway involvement in WG.
Comment: Subglottic stenosis classically causes a fixed
obstruction with airflow limitation during both inspiration
and expiration (Polychronopoulos et al. 2007). This results
a in plateaued or flattened inspiratory and expiratory flow vol-
ume curves (Fig. 24.6a, b).
Myth: Tracheostomies are usually good solutions for WG
patients with difficult subglottic stenosis.
Reality: Tracheostomies should be avoided at all costs in
WG. Even otolaryngologists who perform the procedure
well in patients with WG concede that a tracheostomy gener-
ally makes the patient sicker than he or she was before the
procedure. Avoiding a tracheostomy usually necessitates fre-
b quent endoscopic procedures to dilate stenotic segments of
the trachea manually and to maintain the larger airway with
glucocorticoids and mitomycin C injections.
Fig. 24.5 (a, b) When subglottic stenosis is associated with a significant
Prompt resolution of degree of active inflammation, the narrowing may respond to
an early subglottic an inhaled glucocorticoid powder. One regimen employed
stenosis lesion
following the commonly in this setting is fluticasone (220 mg). This is
institution of administered initially as four puffs twice daily, then reduced
treatment (Figures to two puffs twice daily after several weeks. The success of
courtesy of Dr. Niels this approach probably stems from the turbulence created by
Rasmussen)
the stenosis, which makes the inhaled glucocorticoid settle
locally and deliver the maximum medication dose at the site
of the stenosis. After endoscopic dilatation of subglottic or
subglottic stenosis may vanish completely within 2 weeks endobronchial lesions, inhaled glucocorticoids may help to
on standard immunosuppressive treatment if it is an early limit further scarring secondary to inflammation generated
manifestation of inflammatory disease. Unfortunately, this by the procedure itself. Regular nebulized saline is also
favorable experience contrasts markedly with the course of recommended.
subglottic stenoses later in the course of many patients, when
Pearl: Metal tracheal stents should not be used in WG.
the airway narrowing is sclerotic in nature and more refrac-
tory to medical interventions. Comment: Metal tracheal stents should not be used in patients
with benign airway disorders, including WG. This was the
Pearl: Subglottic stenosis can be diagnosed over the telephone.
subject of a Public Health Notification by the U.S. Food and
Comment: Subglottic stenosis can usually be detected by lis- Drug Administration (FDA Public Health Notification 2005).
tening to patients as they speak. Because of their upper air- The Notification was based upon the high risk of complica-
way narrowing, they are forced to suck in air before beginning tions from metal stents, including airway wall perforation
a sentence. This can be detected over the telephone as well as and stent rupture.
in the examining room. On chest auscultation, loud inspira- When stenting of a large airway obstruction in WG is
tory sounds (sometimes audible without the stethoscope) are required, silicone stents are the preferred devise. The stent-
a finding consistent with subglottic stenosis. ing procedure is performed with a rigid bronchoscope and
250 J. H. Stone et al.

Fig. 24.6 Flow volume loops in

the setting of subglottic stenosis
in a 48 year-old woman with
Wegener’s granulomatosis (a).
The inspiratory flow volume
curve, graphed above the
expiratory flow volume curve, is
mildly flattened. This suggests a
limitation of inspiratory flow and
variable extrathoracic airway
obstruction. (b). One year later,
there was progression of her
subglottic scarring. The flow
volume curve demonstrates a
fixed upper airway obstruction,
evidenced by the more severe
flattening of the inspiratory flow
volume curve and the flattened
expiratory flow volume curve
(Figures courtesy of Dr. Karina
Keogh)

requires a general anesthetic. Silicone stents are generally around the stent and may require careful laser therapy to facil-
tolerated well and are removable. They may migrate distally itate stent removal.
and require repositioning, which involves a repeat broncho-
Myth: Laser therapy is a tissue-sparing method of interven-
scopic procedure.
tion for subglottic stenosis in WG.
Stents may become obstructed by bronchial secretions.
This can be prevented usually by the routine use of nebulized Reality: Medical therapy alone, especially in the chronic
saline treatments. Granulation tissue sometimes develops phase of the disease, is usually insufficient to control
24 ANCA-Associated Vasculitis 251

subglottic stenosis. This complication occurs to at least some 24.5 Mouth

degree in approximately 15% of WG patients, but the clini-
cal features often reflect the development of scar tissue rather
Pearl: Strawberry gums are a classic (if uncommon) sign of
than active inflammation. Collaboration with an experienced
WG.
otolaryngologist is required to manage this problem
effectively. Comment: WG is unique among the vasculitides in its ability
Laser surgeries provide short-term relief in many cases to cause extensive gum inflammation (Fig. 24.7). This com-
but eventually become a self-defeating proposition: the resul- plication responds well to WG treatment, but patients who
tant scar formation exacerbates the situation over the long have advanced gum inflammation at the time of diagnosis
term. This exuberant scarring that occurs in WG tissues fol- can be left with some recession of the gums.
lowing laser procedures makes laser interventions contrain-
dicated for subglottic stenosis in this disorder (Langford Pearl: WG can cause painful oral lesions.
et al. 1996).
Comment: The prominence of oral ulcers in WG is often
Combined endoscopic dilatations and local glucocorti-
underappreciated (Fig. 24.8). Yet, oral ulcers (along with
coid injections are the most effective approach to this prob-
nasal ulcers) comprise one of the four American College of
lem. The injection of glucocorticoids decreases inflammation,
Rheumatology criteria for the classification of WG (Leavitt
impairs scar formation, and enhances resorption of the scar.
et al. 1990). The oral ulcers of WG typically involve the buc-
Some patients need to undergo this procedure several times a
cal mucosa or the tongue and heal quickly following the start
year before stability of the lesion is achieved. The real key in
addressing this problem effectively is having an alliance with
an otolaryngologist who is willing to take on the problem
energetically and provide close surveillance, with interven-
tions as appropriate.

Pearl: Beware the “Boy Who Cried Wolf” effect, which cli-
nicians can cause by submitting too many upper airway
biopsies accompanied only by the question “Wegener’s
granulomatosis?”

Comment: Submitting biopsy specimens without detailed

clinical information is a mistake. Clinicians sometimes
believe they should not tell the pathologist too much, lest the
pathologist be “led” to the diagnosis inappropriately. Failure
to provide sufficient clinical information is particularly harm-
ful in the case of upper-respiratory biopsies. Upper airway Fig. 24.7 Strawberry gums in Wegener’s granulomatosis (Figure cour-
biopsies have a very low specificity for lesions that are char- tesy of Dr. John Stone)
acteristic of WG (i.e., granulomas, vasculitis, or both)
(Devaney et al. 1990).
Most upper airway biopsies from patients who are diag-
nosed eventually with WG show nonspecific inflammation
(D’Cruz et al. 1989). The type of inflammation present in
combination with certain subtle features such as focal ulcer-
ation or a spot of necrosis may implicate WG. Understanding
the full clinical history is important information for the
pathologist.
The use of endoscopes to guide the bioptic procedures in
the nasal cavity greatly enhances the diagnostic yield of
biopsies, which should be obtained from inflamed but non-
necrotic areas. This calls for experience with obtaining these
biopsies and at the same a restrictive policy on biopsying too
often. Tissue damage may be increased by the biopsy, so
biopsies should be performed only if their results have the Fig. 24.8 Tongue ulcer in Wegener’s granulomatosis (From Stone and
potential to alter management. Hoffman 2006)
252 J. H. Stone et al.

of treatment, without scarring. In the WG Etanercept Trial Glucocorticoid use is associated with edema at a variety of
(WGET), 14% of the patients had oral ulcerations that were sites, e.g., the legs and the retina (cystoid macular edema).
attributed by the investigators to active disease over a mean High-dose glucocorticoid use probably explains chemosis, as
follow-up of 2 years (Mahr et al. 2008). well, though the precise mechanism remains obscure.

Pearl: WG rarely erodes the roof of the mouth. Pearl: Orbital pseudotumor can be refractory to medical
therapy.
Comment: In contrast to the upper respiratory tract disease, Comment: Retrobulbar lesions are recalcitrant to conven-
oral lesions in WG are not destructive. When a patient has an tional immunosuppression in some patients with WG, par-
oral lesion that destroys the roof of the mouth, a variety ticularly those with longstanding, fibrotic lesions. Medical
of other disorders must be considered, particularly cocaine therapy is most likely to be effective for early inflammatory
use and T-cell lymphomas (once termed “lethal midline masses behind the eye. Radiographic persistence of orbital
granuloma”). lesions is the rule, unfortunately, even if the masses appear to
ANCA testing may be useful in differentiating cocaine- become dormant for a while after therapy.
associated oral (and nasal) lesions from true WG (Wiesner et For refractory orbital pseudotumors, surgical debulking
al. 2004). Cocaine-induced nasal destructive lesions have can be an effective approach but is associated with a poten-
been associated with ANCA directed against human neutro- tial risk to vision. The biggest challenge is finding an oph-
phil elastase, which can result in a positive ANCA by com- thalmologist who is willing to undertake the procedure.
mercial assays.
Pearl: The eyes of a patient with WG are much more likely to
be wet than dry.
Comment: Nasal inflammation in WG can lead to nasolacri-
24.6 Eyes mal duct dysfunction, insufficient tear drainage, and watery
eyes. In the setting of unilateral conjunctivitis in a patient
who has WG, the underlying cause is probably nasolacrimal
Pearl: Chemosis, a common eye finding in AAV, is a compli-
duct dysfunction.
cation of glucocorticoid therapy.
Nasolacrimal duct obstruction (and accompanying
Comment: Patients are often concerned and clinicians puz- dacryocystitis) is caused by active inflammation or damage
zled by chemosis, the development of edema under the con- to the nose. A dacrocystorhinostomy may ameliorate the
junctivae (Fig. 24.9). One might postulate many reasons for a problem, but the larger tear evacuation route often becomes
patient with AAV to have chemosis: fluid overload, persis- occluded over time by inflamed tissue or scar. This is par-
tence of tears in the eye because of a blocked nasolacrimal ticularly likely when chronic Staphylococcus aureus infec-
duct, low-grade inflammation, and others. However, the com- tion is present and associated with nasal crusts. In some
mon denominator is probably therapy with glucocorticoids. cases, a glass or plastic tube placed in the lacrimal duct may
be beneficial.
Nasal disease in WG has other potential implications for
extra-nasal symptoms. Serous otitis in WG is caused by
problems that prevent ventilation of the middle ear through
the Eustachian tube. As discussed earlier (see “Ears”), tym-
panostomy tubes (grommets) may provide prompt relief of
this problem, but are used only in the context of chronic ear
problems, not acute ones.
Pearl: There is a tarsal structure in the eye, and it can be a
site of trouble in WG.
Comment: Most rheumatologists think of the foot when they
hear the word “tarsus”; as in: “The first metatarsal joint is the
most common initial site of gout.” However, there is a tarsus
in the eye, and it can cause a problem in WG….
The tarsus (plural tarsi) is the thick connective tissue plate
within each of the eyelids that gives the eyelids their firm struc-
Fig. 24.9 Chemosis in a patient with microscopic polyangiitis (Figure ture. When one describes the conjunctiva, the terms tarsal and
courtesy of Dr. John Stone) bulbar conjunctivae are used to describe the regions of
24 ANCA-Associated Vasculitis 253

involvement of inflammatory processes. The tarsal conjunctiva other forms may respond well to milder interventions (gluco-
refers to the conjunctiva that covers the interior surface of the corticoids alone, or even NSAIDs).
eyelids. In contrast, the bulbar conjunctiva covers the globe. Although it is unusual for patients to morph from one type
Patients with WG can develop tarsal conjunctival inflam- of scleritis to another, it is not uncommon for them to need to
mation and scarring. Tarsal conjunctival involvement in WG switch therapies. NSAID therapy for diffuse anterior scleritis
is characterized by inflammation of the palpebral conjunc- may require ratcheting up to glucocorticoids if the initial
tiva and tarsus, followed by a fibrovascular proliferation and therapy does not suppress the inflammation.
scar formation. The histopathological features of tarsal-
conjunctival involvement in WG are granulomatous inflam-
mation, focal necrosis, and areas of occlusive vasculitis in 24.7 Lungs
the tarsus and conjunctiva. WG patients who develop this
complication also appear to be at risk for subglottic stenosis Pearl: Interstitial fibrosis is frequently reported in MPA.
and nasolacrimal duct obstruction (Robinson et al. 2003).
Comment: Interstitial lung disease is an important and under-
Myth: Episcleritis is the most common eye lesion in WG. recognized occurrence in MPA (Fig. 24.11) (Eschun et al.
Reality: Among the many ocular lesions of WG are scleritis, 2003; Homma et al. 2004; Birnbaum et al. 2007). This pul-
peripheral ulcerative keratitis, orbital pseudotumor, dacrocys- monary complication is highly unusual among patients with
titis, and (more rarely) uveitis. However, the most common AAV whose clinical phenotype is most consistent with WG
eye problem in WG is episcleritis (Fig. 24.10), a painless ocu- (Table 24.2).
lar erythema that can occur in one eye or both. Episcleritis can A study comprising 44 patients comparing two separate
cause diffuse eye redness or involve the eye in a more sectoral cohorts from Brescia, Italy, and the Mayo Clinic, respectively,
fashion. It can usually be managed by topical therapy, but revealed the same clinical, radiographic, and serologic mani-
may signal the presence of active disease in other organs that festations of both MPA and the interstitial lung disease in both
requires additional treatment. cohorts. The lung disease is radiographically and physiologi-
cally indistinguishable from usual interstitial pneumonia. It can
Myth: The diagnosis of diffuse scleritis mandates the use of precede other features of MPA and is usually present already at
cyclophosphamide. the time of MPA diagnosis. Additional studies are required to
Reality: Scleritis is not a monolithic disorder. “Splitters” understand the response to treatment and long-term prognosis
divide this form of ocular inflammation into at least five types: of patients with MPA who also have interstitial fibrosis.
diffuse anterior scleritis, nodular scleritis, necrotizing anterior Pearl: Patients with WG are at increased risk of venous
scleritis, posterior scleritis, and scleromalacia perforans (see thrombotic events (VTE).
Chapter 39). These forms of ocular inflammation are quite
distinct clinically, and the individual types of scleritis rarely Comment: During the Wegener’s Granulomatosis Etanercept
change categories. Whereas necrotizing anterior scleritis and Trial (WGET), 16 of the 180 patients enrolled developed
posterior scleritis both require intensive immunosuppression VTEs (Merkel et al. 2005). Moreover, 13 additional patients
with cyclophosphamide and high-dose glucocorticoids, the

Fig. 24.10 Episcleritis in a patient with Wegener’s granulomatosis Fig. 24.11 Interstitial lung disease in microscopic polyangiitis (Figure
(Figure courtesy of Dr. John Stone) courtesy of Dr. John Stone)
254 J. H. Stone et al.

Table 24.2 Clinical phenotypes of Wegener’s granulomatosis and microscopic polyangiitis

WG MPA
ANCA positive (%) 80–90% 75%
Typical ANCA pattern and antigen specificity Cytoplasmic (C-ANCA) proteinase 3 (PR3) Perinuclear (P-ANCA)
Myeloperoxidase (MPO)
Upper respiratory tract Nasal septal perforation Absent or mild
Saddle-nose deformity
Subglottic stenosis
Lung Nodules, infiltrates, or cavitary lesions Alveolar hemorrhage
Interstitial lung disease
Stenotic bronchial disease
Kidney Necrotizing, crescentic glomerulonephritis Necrotizing, crescentic glomerulonephritis
Distinguishing features Destructive upper airway disease, No granulomatous inflammation
granulomatous inflammation

(7%) had histories of VTE prior to enrollment. Thus, approx- some other forms of vasculitis (e.g., PAN), WG involves both
imately one patient in six had a VTE either before or after the venous and arterial sides of the circulation (Wegener
trial entry. In a comparison of patients in the WGET cohort 1939). Venous inflammation, combined with the low flow
to patients in the Johns Hopkins Lupus Cohort, the incidence state in veins and possibly other more disease-specific fac-
of VTE among patients in the WGET cohort was sevenfold tors, may explain this finding. Other ANCA-associated disor-
higher. Moreover, the data on the timing of VTE occurrence ders such as MPA and the Churg-Strauss syndrome probably
were consistent with the concept that VTEs are a manifesta- share this predisposition to venous thrombotic events.
tion of active WG (Fig. 24.12). The median time from enroll-
Myth: Pulmonary-renal syndromes are caused by antiglomeru-
ment in WGET to VTE was 2 months.
lar basement membrane (anti-GBM) disease most of the time.
The explanation for this finding probably relates to the dis-
tribution of blood vessel involvement in WG. In contrast to Reality: AAV is substantially more common than anti-GBM
disease as a cause of pulmonary-renal syndromes (Niles
et al. 1996; Saxena et al. 1995; Gallagher et al. 2002). In a
Time to First VTE
0.15 series of 88 patients with lung hemorrhage and glomerulone-
phritis, 48 (55%) had either MPO- or PR3-ANCA, 6 (7%)
had antiglomerular basement membrane (GBM) antibodies,
and 7 (8%) had both ANCA and anti-GBM antibodies (Niles
et al. 1996). Among the 48 patients with ANCA only, 19 had
antibodies to PR-3, and 29 had antibodies to MPO.
WG
Cumulative Probability of VTE

0.10 Myth: Most patients with MPA have pulmonary hemorrhage.

Reality: Pulmonary hemorrhage is the most dramatic mani-
festation of MPA, but only 12% of patients develop this com-
plication (Fig. 24.13) (Guillevin et al. 1999). Far more
common than pulmonary hemorrhage are glomerulonephritis
(79%), weight loss (73%), mononeuritis multiplex (58%),
0.05 p<0.0001 and fever (55%). Thus, if one views MPA only as a pulmo-
nary-renal syndrome and requires lung involvement before
considering the diagnosis, detection of MPA will be delayed
substantially or missed altogether.

SLE
Myth: Patients with alveolar hemorrhage always have
hemoptysis.
0.00
0 2 4 6 8 10 12 14 16 18 20 22 24 Reality: Patients with MPA or WG can have substantial
Time of Follow-Up (months) amounts of blood in their lungs without manifesting hemop-
tysis. Clinicians must have a low threshold for suspecting
Fig. 24.12 Time to first venous thrombotic event (VTE). The incidence
of venous thrombotic events in Wegener’s granulomatosis patients is
alveolar hemorrhage. Alveolar infiltrates on chest roentgeno-
increased compared with that in patients with systemic lupus erythema- gram (typically in multiple lobes) that have a “ground glass”
tosus (Merkel et al. 2005) appearance on CT scans should raise suspicion of alveolar
24 ANCA-Associated Vasculitis 255

that are undetected on chest radiography, including modest

hilar adenopathy.
Pearl: Tracheobronchial involvement of WG can be clini-
cally silent.
Comment: Symptoms associated with tracheobronchial dis-
ease activity may be very minimal. As a result, this compli-
cation of WG is among the most difficult to diagnose and
treat. The most common symptom of tracheobronchial dis-
ease is a cough, which is usually nonproductive. However,
streaky hemoptysis may also occur. Dyspnea caused by tra-
cheobronchial involvement usually requires significant
mainstem bronchus narrowing or tracheal narrowing (usu-
ally in a subglottic location). Minor or localized (lobar or
segmental) bronchial stenosis is usually asymptomatic but
can lead to abnormalities of the shape of the expiratory flow
volume loop on pulmonary function testing.
Symptoms and pulmonary function test abnormalities may
progress during immunosuppressive therapy that successfully
controls disease activity elsewhere. This may be because of
Fig. 24.13 Alveolar hemorrhage in microscopic polyangiitis (Figure persistent smoldering inflammatory activity of the tracheo-
courtesy of Dr. John Stone) bronchial tree, bacterial (or rarely fungal) superinfection, or
cicatricial narrowing of the lumen of affected bronchi.
Pearl: Inhaled glucocorticoids should be used in patients
hemorrhage, particularly in patients with reduced or falling
with documented endobronchial disease.
hemoglobin levels.
Bronchoscopy with bronchoalveolar lavage is the best Comment: Tracheobronchial disease activity often is less
approach to excluding alveolar hemorrhage. In the case of responsive to methotrexate and at times even to cyclophos-
active alveolar hemorrhage, bronchoalveolar lavage yields phamide than is necrotizing granulomatous inflammation in
progressively bloodier return with each successive aliquot. other locations. Glucocorticoids seem to be the most effec-
This is contrast to an endobronchial bleeding source, where tive agent for the control of endobronchial disease activity in
successive aliquots clear. The presence of a high percentage such patients. In order to minimize exposure to systemic glu-
(>30%) of hemosiderin-laden macrophages on BAL fluid cocorticoids, high doses of inhaled glucocorticoids can pro-
cytology also indicative of blood in the alveoli that has been vide substantial benefit for patients with endobronchial
present for at least 24 h. disease when used in conjunction with other therapies.
Hemosiderin-laden macrophages are an indication of
Pearl: Normal saline nebulizer treatment twice daily may be
hemorrhage even in the absence of bloody return on bron-
beneficial in WG patients who complain of cough and diffi-
choscopy. However, this finding does not identify the source
culties with the expectoration of thick secretions.
of the blood as alveolar in origin. Active alveolar hemor-
rhage, even if mild at the time of presentation, should always Comment: The bronchial epithelium is morphologically and
be considered potentially life-threatening. Consequently, it functionally almost identical to the nasal mucosa. Consequently,
needs to be treated promptly and aggressively, usually with inflammatory insults result in similar damage, which causes
high (pulse) doses of glucocorticoids and either cyclophos- difficulties moisturizing and raising secretions (damaged cili-
phamide or rituximab. ary mucous elevator function). Normal saline inhalations may
partially compensate and provide symptomatic relief. Nasal
Pearl: In WG, chest radiography may reveal almost any
mucosal damage also frequently requires the chronic adminis-
abnormality except hilar adenopathy.
tration of saline nasal spray.
Comment: Hilar adenopathy is an unusual finding in WG.
Pearl: Progressive large airway obstruction does not neces-
Evidence of hilar adenopathy on chest radiography vaults
sarily represent disease activity.
other diagnoses, particularly sarcoidosis, lymphoma, and
tuberculosis, ahead of WG on the differential diagnosis. Comment: Only bronchoscopy can allow differentiation of
Computed tomography scans of the chest are more sensitive active inflammation from circumferential scarring as a cause
than chest radiography and often reveal pulmonary lesions of progressive bronchial narrowing (Polychronopoulos et al.
256 J. H. Stone et al.

2007). In the case of circumferential scarring, additional sys-

temic immunosuppression is not necessary. Bronchoscopic
balloon dilation or silastic airway stent placement may be
beneficial, depending on the specific location.

24.8 Kidneys

Myth: Granulomas in the renal biopsy of a patient with

ANCA are pathognomonic of WG.
Reality: Granulomas occur in many different conditions and
their histopathological features usually fail to reveal their spe-
cific cause. Granulomas in renal biopsies of patients with WG
are a rare finding, but may occur in different forms. They can
Fig. 24.14 Glomerulus with fibrinoid necrosis nearby the glomerular
contain multinucleated giant cells, histiocytes, macrophages, vascular pole, highlighted in this trichrome staining as carmine red
fibroblasts, and areas of necrosis. They can occur around spots (400 ×) (Figure courtesy of Dr. Ingeborg Bajema)
destroyed glomeruli, blood vessels, and tubes.
Granulomas were detected in 40 renal biopsies performed Comment: No one knows the minimum number of glomeruli
as part of a consecutive series between 1991 and 2004 that a renal biopsy should contain in order to be considered
(Javaud et al. 2007). Among these biopsies, 20 (50%) were representative of overall disease extent within the kidneys.
from patient with sarcoidosis; seven (17.5%) were attributed The issue has been debated for many years.
to drug reactions; and three (7.5%) were related to tubercu- To some extent, the answer depends on the disease in
losis (Javaud et al. 2007). Only two (5%) of the 40 biopsies question. In IgA nephropathy, for example, all glomeruli
were from patients with WG. generally reveal mesangial IgA deposits by immunofluores-
Even among patients with ANCA, the differential diagno- cence. Thus, a limited number of glomeruli suffice to estab-
sis for granulomas found within kidneys should include the lish the diagnosis. (In the same way, the palpable purpura of
full range of conditions in which granulomas occur, particu- Henoch-Schönlein purpura is generally more diffuse than is
larly if the glomerular lesions are not consistent with a small- this finding in pauci-immune conditions such as WG.)
vessel vasculitis (i.e., pauci-immune glomerulonephritis). The pauci-immune immunofluoresence pattern that is
characteristic of ANCA-associated vasculitis does not help to
Myth: Wainright and Davson were the first to propose that
distinguish this disease in biopsies that lack the characteristic
fibrinoid necrosis in the glomerulus represents small-vessel
necrotic lesions, crescentic lesions, or both. In renal biopsies
vasculitis of the glomerular tuft.
that show interstitial inflammation, the possibility of a glom-
Reality: Although the 1950 article by Wainright and Davson erular disease such as pauci-immune crescentic glomerulo-
is often cited as the first to claim that fibrinoid necrosis rep- nephritis should be considered. Deeper cuts through the
resents small-vessel vasculitis in the glomerulus, it does not biopsy block may be helpful. As is true with many disorders
actually make this claim (Wainright et al. 1950). Fibrinoid that require close clinicopathologic correlation for diagnosis,
necrosis of the glomerulus is a segmental lesion that consists a dialog between the clinician and pathologist (often con-
largely of amorphic eosinophilic material, which is high- ducted most effectively over a microscope) is essential to get-
lighted in trichrome stains as carmine red spots (Fig. 24.14). ting the diagnosis right.
The term “fibrinoid” was coined because of the fibrin-like Myth: Fibrinoid necrosis must be present within a glomeru-
structure of the lesion, but it is likely that the lesion is com- lus if a crescent is evident.
posed of constituents other than fibrin and fibrinogen (Bajema
and Bruijn 2000). The relationship between ANCA in the Reality: This Myth stems from the notion that glomerular
serum and fibrinoid necrosis in the kidneys and other tissues crescent formation is caused by fibrin leakage through dis-
is not clear. rupted glomerular basement membranes. It is not clear that
this notion is correct. In a study of renal biopsies from
Pearl: Crescentic necrotizing glomerulonephritis is often a patients with WG, 87 glomeruli were reconstructed in three
focal, segmental disease. It can therefore be missed in renal dimensions and scored for the presence of crescents and
biopsies that contain few glomeruli. fibrinoid necrosis (Hauer et al. 2000). Extracapillary
24 ANCA-Associated Vasculitis 257

proliferation (i.e., crescent formation) occurred in the absence Hematuria in a cyclophosphamide treated patient can
of fibrinoid necrosis in 22% of the glomeruli. also occur as a result of drug-induced injury or a bladder malig-
These data do not disprove a relationship between cres- nancy, even if treatment was stopped many years ago. The
cent formation and fibrinoid necrosis. It is possible, for presence of nonglomerular hematuria in a cyclophosphamide-
example, that the necrotic lesions resolved before the cres- treated patient should prompt cystocopy to examine for
cents came to “full bloom.” Nevertheless, pathologists should urothelial lesions or neoplasia (see “Bladder” below).
use the term “necrotizing crescentic glomerulonephritis”
Pearl: ANCA positivity may correspond with a renal biopsy
only if both fibrinoid necrosis and crescents are actually
that reveals more than pauci-immune staining for immuno-
detected in the biopsy.
globulins (i.e., >2 + for IgG, IgA, or IgM).
Myth: Persistent hematuria in a patient with AAV under treat-
Comment: “Pauci-immunity” has been defined as 2 + stain-
ment is a signal to intensify immunosuppressive therapy.
ing or less (on a scale of 0 to 4+) for any immunoglobulin or
Reality: Persistent hematuria in a patient with AAV is a com- complement component, as well as the absence of immune
mon cause of clinician anxiety. Distinction between active complex–type electron-dense deposits on electron micros-
disease and damage is essential in these patients. Disease- copy. However, some renal biopsies from patients with
related hematuria results from rents in the glomerular base- ANCA-associated vasculitis patients stain > 2 + for immuno-
ment membrane. As a result, red blood cells may persist in reactants. In one study of renal biopsies from 213 patients
the urine for many months after the institution of appropriate with ANCA-associated vasculitis, 30 (14%) had staining
treatment and the establishment of disease control. Red blood patterns for immunoglobulins that were more than pauci-
cell casts (Fig. 24.15) may also persist in the urine for many immune (Harris et al. 1998).
weeks and even months after arrest of the renal inflammation Synergy between ANCA and the immune complexes could
and systemic vasculitis. Thus, persistent red blood casts do lead to more severe glomerular injury in some patients. This
not necessarily imply the failure of therapy, even if they are could also be the case for overlapping conditions in which
found still several months into treatment. (But they do make AAV accompanies another form of glomerulonephritis. This
one worry, and look carefully for evidence of active disease phenomenon was underlined by the finding of extensive
in other organs.) necrosis and crescent formation in a group of ten patients with
Patients with persistent hematuria and red blood cell casts lupus nephritis and ANCA positivity (Nasr et al. 2008).
must be observed very closely. They require other investiga- The notion that electron-dense deposits are absent in
tions, such as more frequent serial measurements of the ANCA-associated glomerulonephritis has also been called into
glomerular filtration rate, the urine protein: creatinine ratio question. In a study of 126 renal biopsies from patients with
and even cystoscopy. However, more immunosuppression is ANCA-associated vasculitis, immune complex deposits were
an inappropriate knee-jerk response in this clinical scenario. detected via electron microscopy in 54% (Haas et al. 2004).
Failure to recognize that hematuria and even red cell casts
Pearl: Once renal involvement by WG begins, organ- or life-
can persist in the urine even the achievement of clinical dis-
threatening disease may ensue swiftly.
ease remission leads to misguided and excessive therapy.
Comment: Few diseases can cause renal deterioration as rap-
idly as WG. Once the serum creatinine begins to rise in WG,
the disease often appears to accelerate, with rapidly progressive
GN, swift progression to renal failure, and the appearance of
“disseminated” involvement (alveolar hemorrhage, mesenteric
vasculitis, and so on). In a patient with possible or known WG,
a rising creatinine can’t wait until Monday. Time is nephrons.
Pearl: The term “limited” WG should be used advisedly.
Comment: Contrary to the connotation of “limited,” the
prognosis of patients with limited WG is not benign in many
patients to whom that appellation is applied. Patients with
WG who did not have renal involvement were first reported
in 1966 (Carrington and Liebow 1966). Many reports and
studies in the literature now refer to such patients as having
“limited” WG. This condition was once believed to correlate
Fig. 24.15 Red blood cell casts in a patient with glomerulonephritis
associated with microscopic polyangiitis (Figure courtesy of Dr. John with a more favorable clinical outcome compared with that
Stone) of the generalized form (Luqmani et al. 1994).
258 J. H. Stone et al.

However, in a study of 180 patients with WG enrolled in Comment: Patients with end-stage or near end-stage renal
a clinical trial, patients with limited disease tended to have disease are at increased risk for complications. In a random-
longer disease duration, a greater likelihood of experiencing ized trial of plasma exchange in ANCA-associated vasculitis
exacerbation of previous disease following a period of remis- in which all patients had a serum creatinine of at least 5.7 mg/
sion, and a higher prevalence of destructive upper respiratory dL, 19 of the 137 patients died of infection within 1 year
tract disorders (WGET Research Group 2003). A careful (Jayne et al. 2007). The overall mortality in that trial was
examination of the patients described by Carrington and 26%, underscoring the poor prognostic implications of pre-
Liebow indicated that a number of those patients in fact had sentation with end-stage renal disease.
mild signs of renal involvement. The increased morbidity and mortality among patients
Clinicians and pathologists alike should bear in mind that with advanced renal dysfunction has multiple causes and
renal biopsies from patients with only slightly elevated serum several implications for therapy. The utmost precautions are
creatinine levels often reveal severe histological lesions. This required to prevent dialysis-related infections.
is another reason that the term “limited” WG should be used
Pearl: Renal transplantation is an excellent option for treat-
with caution.
ing patients with AAV who reach end-stage renal disease.
It should also be stressed that isolated alveolar hemorrhage
in an ANCA-positive patient should never be labeled as “lim- Comment: The patient and graft survival rates following
ited” disease. Even though the disease is “limited” to the lung renal transplantation in patients with ANCA-associated vas-
at the time of presentation, this complication warrants remis- culitis are similar to those in patients with end-stage renal
sion induction therapy for life-threatening disease. disease from other causes (Geetha and Seo 2007). In a recent
study of 35 patients with ANCA-associated glomerulone-
Myth: WG patients who present with end-stage renal disease
phritis who received kidney transplants, the 5-year patient
who do not have severe pulmonary disease can be managed
and death-censored allograft survival rates were 94 % and
with prednisone alone.
100%, respectively (Gera et al. 2007).
Reality: WG, in contrast to lupus, does not “burn out” in Standard transplant immunosuppression does not preclude
patients whose kidneys fail. Therapy with a cytotoxic agent a relapse of AAV, but not all relapses affect the transplanted
in addition to glucocorticoids is important to treat or pre- kidney. In one study, 40% of relapses were extrarenal (Nachman
vent serious nonrenal disease in WG, e.g., alveolar hemor- et al. 1999). If a renal relapse does occur, it is characterized by
rhage. A second agent is usually required to control WG, a crescentic necrotizing glomerulonephritis, which can be dis-
even if salvaging useful renal function is no longer a tinguished easily from a transplant glomerulitis.
possibility.
Pearl: Prognostication about renal function early in the
Among patients with disseminated WG treated with glu-
treatment course of WG is risky.
cocorticoids alone, survival is approximately 50% at 12
months (Hollander 1967). Among WG patients in the longi- Comment: The kidneys are often the last organ to respond to
tudinal NIH series who were only treated with glucocorti- therapy in WG. Even if aggressive treatment begins when
coids, approximately 50% experienced some degree of patients appear to have relatively good renal function (e.g., a
clinical improvement, but only 4% (all of whom had mild serum Cr of 1.4 mg/dL), a small minority of patients march
disease) achieved remissions. directly to end-stage renal disease despite the use of intensive
The second reason for aggressive (but judicious) treat- immunosuppression. The reasons for the delay in renal response
ment in WG patients who require dialysis is that aggressive to treatment in some cases are probably multifactorial and may
therapy may lead to renal recovery. This point has also been include intercurrent processes such as acute tubular necrosis.
emphasized in patients with MPA. Among 107 patients with However, a significant proportion of patients who require dial-
MPA and GN in one study, 23 (21%) required dialysis at ysis in this setting are able to discontinue renal replacement
presentation. Of the 17 patients who received aggressive therapy within several weeks to a few months.
treatment, 9 no longer required dialysis after 3 months It is not clear that any types of treatment alter the renal
(Nachman 1996). prognosis in such patients. Some centers would add plasma-
The caveat to treating AAV aggressively in patients with pheresis in this setting but data on the utility of that therapy
renal failure is that treatment must be administered with for this particular indication remain unclear. “Sitting tight,”
appropriate caution and adjustments of immunosuppression removing other potential renal insults, and waiting for the
doses for renal dysfunction. These issues are discussed fur- immunosuppressive therapy to take effect may be the best
ther below. course of action.

Pearl: The time around the start of dialysis is a hazardous Pearl: Co-occurrence of ANCA and anti-GBM antibodies is
one for patients with vasculitis. associated with a worse renal prognosis.
24 ANCA-Associated Vasculitis 259

Comment: Up to 30% of patients diagnosed with anti-GBM 24.10 Skin

disease also have ANCA, most often MPO-ANCA (Bosch et
al. 2006). Conversely, up to 10% of patients with AAV also
Myth: Vasculitic lesions in AAV are always “pauci-immune.”
have anti-GBM antibodies. Both types of autoantibodies
should be sought in pulmonary-renal syndromes. The pres- Reality: Most vasculitic skin lesions in AAV have a histology
ence of anti-GBM antibodies is an indication for plasma of leukocytoclastic vasculitis. However, “leukocytoclastic
exchange in addition to immunosuppressive agents. vasculitis” does not distinguish AAV from Henoch-Schönlein
purpura, mixed cryoglobulinemia, erythema elevatum diuti-
Myth: Patients with AAV on the renal transplant list must be
num, and a host of other conditions that cause vasculitis of
ANCA negative in order to be suitable candidates for a renal
the skin. Immunofluorescence studies of skin biopsies are
allograft.
frequently used to narrow the diagnostic possibilities further.
Reality: The currently available evidence does not support Positive findings, especially for IgA and complement, are
this practice (Rostaing 1997; Geetha and Seo 2007). Patients often used to exclude AAV. However, a substantial number of
with AAV and end-stage renal disease can undergo renal skin biopsies in AAV that reveal leukocytoclasis are not
transplantation even if they remain ANCA positive, provided pauci-immune in nature. These can reveal substantial IgA,
their disease has been quiescent clinically for a reasonable IgM, and complement staining, and may create diagnostic
length of time (e.g., 6 months). confusion (Brons et al. 2001).
Pearl: Scurvy can mimic AAV, particularly MPA.
Comment: Scurvy can cause perifollicular petechiae that
24.9 Bladder mimic purpura, extensive gum inflammation, and alveolar
hemorrhage, all of which can be confused with MPA (Duggan
et al. 2007). Think of scurvy when a severely malnourished
Pearl: Drug-induced cystitis caused by cyclophosphamide
patient is suspected of having an AAV.
can occur within several weeks of treatment initiation.
Comment: One important cause of hematuria in patients with
WG that is not related to either active disease or damage is
drug-induced cystitis caused by cyclophosphamide. The 24.11 Joints
degree of symptomatology in cyclophosphamide-induced
cystitis is highly variable. Some patients, a minority, present Pearl: Arthritis is a common presentation of a flare in AAV.
with profound symptoms of dysuria. On the other hand, many
Comment: Arthritis and arthralgias are a frequently over-
patients with this complication of therapy do not have urinary
looked manifestation in AAV, perhaps because they improve
symptoms (urinary frequency, dysuria), but present rather
so quickly after the institution of therapy. The arthritis com-
with asymptomatic microscopic hematuria. Gross hematuria
monly involves large joints in a migratory, asymmetric, oli-
is less common.
goarticular pattern, 1 day involving a knee and an ankle, the
Pearl: Self monitoring by patients for recurrent hematuria by next day a shoulder or wrist (Fig. 24.16). Small joints may
regular urine dipsticks is an excellent early warning system also be involved. Arthritis and arthralgias are a very common
for the detection of recurrent disease. tip-off to a disease flare in AAV.
Comment: This Pearl is useful for patients whose hematuria
has resolved after treatment. Patients can learn easily to dip-
stick their urine every 1–2 weeks. Patients whose urine dip- 24.12 Peripheral and Cranial Nerves
stick shows evidence of hematuria should immediately
undergo a full urinalysis with microscopy on a fresh urine
Pearl: Even a frail, elderly woman should not have “break-
specimen to exclude the presence of red blood cell casts.
able” foot dorsiflexion strength.
Pearl: WG can involve the genitourinary tract.
Comment: Vasculitic neuropathy can become lost in the
Comment: In this protean disease, even the genitourinary tract shuffle of other organ system involvement and remain sub-
is not spared. WG involvement of the cervix, vulva, uterus, clinical (i.e., undiagnosed) if the clinician does not ask about
penis, prostate gland, and other parts of this organ system symptoms of numbness, pain, and burning in the feet and
have been described (Stone 1996; Bean and Conner 2007). If hands. In addition, a careful physical examination for foot
a patient with WG develops peculiar problems in these organs, dorsiflexion weakness must be performed, because this is
WG must be excluded with appropriate biopsies. the most common peripheral nerve lesion in vasculitic
260 J. H. Stone et al.

Pearl: Vasculitic neuropathy is more common in MPA than in

WG.
Comment: This statement is correct: MPA has a greater pro-
pensity to involve peripheral nerves than does WG. However,
devastating vasculitic neuropathy can occur in either disor-
der and is associated with either PR3-ANCA or MPO-ANCA
positivity.
Pearl: A facial droop may result from granulomatous inflam-
mation within the middle ear in WG.
Comment: The seventh cranial nerve courses through the
middle ear cavity (see Fig. 39.1 in Chap. 39, I). If middle ear
inflammation is sufficiently severe in WG and leads to an
Fig. 24.16 Wrist arthritis in a patient with Wegener’s granulomatosis expanding, compressive lesion, both conductive hearing loss
(WG). When frank arthritis occurs in WG, involvement of one major and a seventh nerve palsy can result (Fig. 24.18). Such lesions
joint (often with an effusion, as in this case) is common. Diffuse, migra-
may heal completely through medical treatment alone.
tory arthralgias are often the presenting feature of a disease flare (Figure
courtesy of Dr. John Stone)

24.13 Central Nervous System

Pearl: WG can present with headaches that mimic giant cell

arteritis (GCA).
Comment: The principal characteristic of a GCA headache is
that it is new; i.e., the patient has not experienced such

Fig. 24.17 Examination of the foot for dorsiflexion weakness. In mild

cases of foot dorsiflexion weakness, the most common manifestation of
vasculitic neuropathy, the lesion is often unrecognized by patients with
more severe manifestation of disease in other organs and is also often
overlooked by clinicians. This patient had substantial bilateral foot dor-
siflexion weakness, left greater than right, which even she had not
appreciated. She had, however, noted pain from sensory neuropathy,
which almost always accompanies motor neuropathy in vasculitic neu- Fig. 24.18 Peripheral seventh cranial nerve palsy in Wegener’s granu-
ropathy (Figure courtesy of Dr. John Stone) lomatosis. The patient has a facial droop on the affected side and is
unable to wrinkle the ipsilateral forehead. This lesion results from gran-
ulomatous inflammation within the middle ear, leading to compression
of the portion of the seventh cranial nerve that runs through that cavity.
neuropathy (Fig. 24.17). Even a strong examiner should not Middle ear surgery places the patient at risk for irreversible damage to
be able to overcome the foot dorsiflexion strength of a patient the nerve if granulomatous tissue it. Thus, if the diagnosis can be estab-
whose distal muscle strength is intact. lished in other ways, the middle ear should be left alone
24 ANCA-Associated Vasculitis 261

24.15 ANCA

Myth: A rise in ANCA titer predicts a disease flare.

Reality: This has been one of the most hotly debated issues in
AAV since the mid-1980s. Several prospective studies have
shown that the temporal association between ANCA titer ele-
vations and the occurrence of disease flares is poor (Table 24.3)
(Finkielman et al. 2007; Boomsma et al. 2000; Jayne et al.
1995). Many patients flare after elevations in ANCA titer, but
these events are seldom linked tightly enough in time for clini-
cal decisions to be based upon ANCA serologies.
Data from prospective studies on whether or not the
achievement of a negative ANCA protects against future dis-
ease flares are mixed. Two studies from the Netherlands
reported that patients who became ANCA-negative were less
likely to flare than those who remained ANCA positive
(Boomsma et al. 2000; Stegeman et al. 1994). However, in
the Wegener’s Granulomatosis Etanercept Trial, the achieve-
ment of ANCA-negative status did not protect patients against
subsequent disease flares (Finkielman et al. 2007).
Fig. 24.19 Meningeal involvement in Wegener’s granulomatosis (Figure Pearl: There is considerably more overlap than dissimilarity
courtesy of Dr. John Stone)
between patients who have MPO- as opposed to PR3-ANCA.
Comment: Various studies have investigated whether the
clinical and pathological parameters differ between patients
headaches before. WG can cause intense, unrelenting headaches with anti-MPO and anti-PR3 ANCA. One investigation of 46
that are new for the patient and mimic GCA. There are several anti-MPO positive patients and 46 anti-PR3 positive patients
potential sources of headache in WG: sinusitis, chronic osteitis concluded that extrarenal organ manifestations and respira-
in the mid-facial bones, meningeal inflammation (Fig. 24.19), tory tract granulomas were more common in patients with
ocular disease (posterior scleritis, which may require ocular anti-PR3 ANCA than in those with anti-MPO antibodies,
ultrasound to diagnose (Sadiq et al. 2003) ), and temporal artery and that anti-PR3–positive patients are predisposed to more
inflammation (Small and Brisson 1991; Genereau et al. 1999). dramatic deteriorations in renal function (Franssen et al.
1998). Earlier studies had found that these patient groups
have similar prognoses (Geffriaud-Ricouard et al. 1993).
In renal biopsies, there is a tendency toward more chronic
24.14 Spleen lesions in anti-MPO positive patients (Hauer et al. 2002).

Myth: Splenic involvement in WG is rare. Table 24.3 ANCA titers correlate poorly with the timing of disease
flares
Reality: Splenic involvement was actually recognized by
Study Patients PR3- Serum ANCA Relapses
Friedrich Wegener in his first paper on this disease (Wegener
ANCA collection increases (no.)
1936). Wegener’s paper was illustrated with a pathologic fig- positive (no.)
ure of splenic infarctions, which he termed Fleckmilz (“spot- (no.)
ted spleen”). Boomsmab 100 85 2 38 22a
Because of the decrease in the number of WG patients who 30 15a
die from their illness and the declining frequency of deaths Jaynec 60 39 1 27 13a
studied by autopsy, splenic involvement is underrecognized in Kyndtd 43 17 1–3 14 4
WG (and its clinical implications, if any, are unclear). Cross- 17 2
a
sectional imaging studies have reminded us that splenic Within the following year after the increase in ANCA
b
Boomsma et al 2000
involvement can occur in WG, and that it usually presents with c
Jayne et al. 1995
splenic infarctions. d
Kyndt et al. 1999
262 J. H. Stone et al.

This may be related to the timepoint at which the biopsy is 24.16 Treatment and Course
performed: patients with PR3-ANCA are likely to have ear-
lier renal biopsies because upper respiratory tract disease
Pearl: “Refractory WG” is an opportunistic infection until
calls the patient to medical attention.
proven otherwise.
Pearl: Seek, and ye shall find ANCA. Comment: During the first 3–4 months of therapy, patients
who improve and enter remission on cyclophosphamide and
Comment: All large studies of “ANCA-associated vasculitis”
glucocorticoids are more likely to suffer serious complica-
have reported small numbers of patients who are ANCA-
tions of this treatment than an exacerbation of WG. Patients
negative despite the presence of generalized disease. These
with “disseminated” or generalized WG who are refractory
patients, who normally comprise up to 10% of most studies,
to high doses of conventional immunosuppression are truly
stand as a compelling argument against the pathogenicity of
rare. Thus, a WG patient who “worsens” after being on
ANCA.
cyclophosphamide and prednisone for a while probably has
By combining all ANCA testing methods (indirect immu-
an opportunistic infection or another complication of treat-
nofluorescence, direct ELISA, and capture ELISA), the
ment. The most common site of an opportunistic infection in
number of patients with severe clinical disease who remain
AAV is the lung. In critically ill patients, bronchoscopy with
ANCA negative can be reduced to a minimum of 4%. This
bronchoalveolar lavage should be performed immediately.
was demonstrated in a study of 180 patients with WG, of
The upper respiratory tract is another challenging site. A
whom 128 had severe disease and 52 had limited disease
sensible approach to distinguishing chronic sinusitis from
(Finkielman et al. 2007).
smoldering WG is to treat vigorously with antimicrobial
There are very few data regarding the histomorphology of
agents for an appropriate length of time. If this strategy does
ANCA-negative patients. Eisenberger et al. reported a group
not achieve control of the problem, it may be assumed that
of 20 patients with pauci-immune, crescentic, necrotizing
the disease is active. If neither antimicrobial agents nor an
renal vasculitis who were ANCA negative (Eisenberger et al.
increased dose of prednisone helps, then the patient may
2005). The occurrence of crescents, fibrinoid necrosis, and
need a surgical intervention to clean damaged sinuses.
interstitial infiltrates was similar to that found in PR3- and
MPO-positive patients. Renal biopsies of ANCA-negative Pearl: A little bit of prednisone goes a long way toward sus-
patients tended to show more interstitial fibrosis and glom- taining disease remissions in AAV.
erulosclerosis, probably a consequence of later diagnosis
Comment: An important lesson from the major clinical trials
because of their ANCA negativity (Hauer et al. 2002).
in AAV published since 2003 is that keeping patients on a
Pearl: Antithyroid medications are the most common cause small glucocorticoid dose has a dramatic impact on the flare
of drug-induced AAV. rate. This is highlighted in Table 24.4. Patients enrolled in
protocols that called for tapering glucocorticoid doses to
Comment: Drug-induced AAV provides a compelling model zero in the months after remission had dramatically higher
for studying the role of ANCA in disease pathogenesis. The rates of disease flare compared with patients whose treat-
strongest associations between medications and ANCA pro- ment protocols permitted ongoing doses equivalent to
duction involve antithyroid drugs such as proplthiouracil between 5 and 10 mg of prednisone (Goek et al. 2005).
(Gunton et al. 2000). ANCA positivity develops in up to 25%
Myth: Cyclophosphamide should be continued until the
of patients treated with PTU, but clinically evident vasculitis
patient has been in a clinical remission for 1 year.
is found in only a small minority (Gunton et al. 2000). Far less
robust relationships have been reported between the develop-
Table 24.4 Relationship between maintenance glucocorticoid dose
ment of ANCA and the use of medications such as pen- and disease flare in three randomized clinical trials
icillamine, alloprinol, procainamide, phenytoin, cefotaxime, CYCAZAREMb NORAMc WGETd
isoniazid, and indomethacin (Sakai et al. 1999; Feriozzi et al. (n = 144) (n = 100) (n = 180)
2000; Bienaime et al. 2007). Prednisolone >10 mg 7.5 mg Off
In a study of 209 patients with hyperthyroidism, 13 (6%) dose
became positive for MPO-ANCA, PR3-ANCA, or antie- 6 months
lastase ANCA (Slot et al. 2005). Among the ANCA-positive Prednisolone 7.5 mg Off Off
dose
patients, the odds ratio for previous treatment with antithy-
12 months
roid medications was 11.8. Some data suggest that the likeli- Flare rate 15 % 70 % (Methotrexate 51 %
hood of developing clinically significant AAV (e.g., (%) arm)
glomerulonephritis) correlates with length of time on anti- 47 % Cyclophosph-
thyroid medications. amide arm)
24 ANCA-Associated Vasculitis 263

Reality: Cyclophosphamide revolutionized the treatment Pearl: The effect of cyclophosphamide on lymphocytes per-
of WG. The use of this medication (first employed regu- sists for months after the drug has been discontinued.
larly at the National Institutes of Health [NIH] in the late
Comment: The effects of cyclophosphamide on lymphocyte
1960s) changed an illness that was almost universally fatal
counts persist for many months after the drug has been dis-
within 1 year to one in which remission is achieved, at least
continued (Fauci 1974). This has direct implications for the
temporarily, in the great majority of cases. Patients and
duration of Pneumocystis jiroveci pneumonia prophylaxis
physicians alike were reluctant to discontinue cyclophos-
(Suryaprasad 2008). No one knows precisely how long
phamide following early successes with this medication. It
Pneumocystis prophylaxis should be continued in patients after
became standard practice to continue cyclophosphamide
cyclophosphamide (or other immunosuppressive medications)
for at least 1 year after the patient had achieved disease
has been stopped. However, it seems prudent to monitor CD4
remission.
lymphocyte counts and to continue Pneumocystis prophylaxis
The downside of using cyclophosphamide in this fashion
until the total lymphocyte count is greater than 200/mm3.
became apparent in the 1990s. In a report of the longitudinal
The take home lesson is that Pneumocystis prophylaxis
NIH series of patients with WG, the mean duration of cyclo-
may need to continue in some patients for some months, even
phosphamide use was approximately 2 years. More than
after cyclophosphamide and other intensive immunosuppres-
40% of the patients suffered permanent treatment-related
sive agents have been discontinued. Fatal Pneumocystis
morbidity, including drug-induced cystitis, bladder cancer,
infections have been reported months after the discontinua-
an increased risk of leukemia/lymphoma, opportunistic
tion of cyclophosphamide.
infection, and others (Hoffman et al. 1992).
Subsequent studies have demonstrated that the acceptable Myth: The induction of neutropenia is essential to the suc-
risk: benefit ratio for the continuation of cyclophosphamide cessful treatment of WG.
is exceeded long before the 2-year mark. Shorter courses of
Reality: Neutropenia (and lymphopenia) is precisely what
cyclophosphamide are equally effective in inducing remis-
leads to many of the adverse events associated with the con-
sion. The standard of care for remission induction in patients
ventional treatment of WG. Nearly, all patients’ disease can
with severe AAV now is to use cyclophosphamide for 3–6
be controlled without the induction of neutropenia (white
months along with high-dose glucocorticoids, and then to
blood counts below 4,000/mm3). This concept was promul-
discontinue cyclophosphamide in favor of either azathio-
gated in the 1970s in one of the early papers from the NIH on
prine or methotrexate. This standard of care may be altered
WG: “Granulocyte kinetics were measured… we found that
by conclusions from the ongoing Rituximab in ANCA-
marked suppression of granulocytopoiesis was not necessary
associated Vasculitis (RAVE) trial (Specks et al. 2009).
for a favorable response…” (Fauci 1974). Despite this obser-
Pearl: Patients tolerate less and less cyclophosphamide as vation, too many practitioners target neutropenia or lym-
time goes by. phopenia as a therapeutic target, believing that this is
necessary for the induction of disease remission.
Comment: Regular monitoring for toxicity is required for
patients on cyclophosphamide. The need for dose adjust- Pearl: Watch not only the absolute neutrophil count but also
ments may be apparent after only 1 or 2 weeks in some the trend in values in cyclophosphamide-treated patients.
patients, but more often signs of bone marrow toxicity become Comment: Although a total white blood cell count of 5.0 ×
apparent after several months on the drug. It is a rare patient 103/mm3 sounds fine for a patient on cyclophosphamide, it
who can tolerate full doses of cyclophosphamide without a signals a worrisome trend if the previous week’s count was
decline in bone marrow function for more than 1 year. 9.0 × 103/mm3 and measurement the week before that was
The bone marrow never forgets that it has seen cyclophosph- 12.4 × 103/mm3. Appropriate steps in the management of this
amide. Patients tolerate second courses of cyclophosphamide patient would be to hold the cyclophosphamide for a few
less well than their first. Moreover, the use of azathioprine or days and repeat the white blood cell count, or at a minimum
methotrexate after an induction course of cyclophosphamide to reduce the daily dose (and repeat the count).
can be associated with precipitous declines in cell counts Blood counts drawn on 1 day reflect the dose of cyclophos-
shortly after the start of the remission maintenance regimen. phamide taken roughly 1 week earlier. This makes following
Vigilance must remain high for cytopenias even after the trends as important as following absolute numbers. In a patient
patient is off cyclophosphamide. with a declining white blood cell count, consideration should
The white blood cell line is usually affected first by cyclo- be made for dosage reduction or even drug holding in certain
phosphamide toxicity. A minority of patients become throm- instances.
bocytopenic quickly, before an effect is evident on leukocytes.
The red blood cell line is typically affected last. Myth: Cyclophosphamide is more effective if given twice a
day.
264 J. H. Stone et al.

Reality: Splitting the dose of cyclophosphamide offers no for drug interaction will be caught. Thus, patients themselves
pharmacokinetic advantages and increases the medication should be aware of this potential interaction.
risks. Acrolein, a metabolite of cyclophosphamide, is toxic to On the other hand, Pneumocystis prophylaxis is crucial for
the urothelium and can lead to drug-induced cystitis. To mini- all patients receiving immunosuppressive regimens comprised
mize this risk, daily cyclophosphamide should always be taken of high-dose glucocorticoids and another immunosuppressive
all at once and in the morning, with a large amount of fluid to regimen, including methotrexate. A daily single-strength TMP/
maintain frequently voided, dilute urine. Cyclophosphamide SMZ tablet (or a double-strength tablet every other day) given
should never be taken before bedtime because this increases for that purpose appears to be well-tolerated by patients on
the potential for acrolein to be in contact with the urothelium. methotrexate, provided that folic acid supplementation (1 mg/
day orally) is given at the same time.
Pearl: Disease-associated morbidity usually occurs within
the first 3 months. Myth: Vaccination is contraindicated in patients with AAV
Comment: The CYCAZAREM trial compared the abilities of because it can trigger disease flares.
cyclophosphamide and azathioprine to maintain remission in
Reality: The results of a large cohort study of individuals
155 patients with ANCA-associated vasculitis (WG = 95
vaccinated with the influenza vaccine indicated this vaccina-
patients; MPA = 60 patients) (Jayne et al. 2003). Eighteen
tion did not lead to disease flares in patients with AAV
severe or life-threatening adverse events occurred during the
(Stassen et al. 2008). In addition, contrary to popular percep-
first 3 months of therapy. In contrast, only 27 such events
tion, patients with these diseases are able to mount adequate
occurred during the remission-maintenance phase, which lasted
antibody responses to influenza vaccination despite active or
five times longer. Seven patients died during the remission-
previous immunosuppressive therapy (Holvast et al. 2008).
induction phase (two from pneumonia, three from pulmonary
vasculitis and infection, and two from stroke), but only one Pearl: The dose of cyclophosphamide needs to be modified in
died during the remission-maintenance phase (from a stroke). the setting of renal dysfunction.
These data were echoed also in findings from the Wegener’s
Granulomatosis Etanercept Trial, in which a disproportionate Comment: Cyclophosphamide should be dosed for renal
number of adverse events occurred in the first 6 months failure (Haubitz et al. 2002). The dose should be reduced by
(remission induction) as opposed to the subsequent remission 50% for those on dialysis (i.e., 1 mg/kg/day should be
maintenance phase (WGET 2005). administered rather than 2 mg/kg/day). Cyclophosphamide
There are several explanations for the high incidence of dose reductions are also advisable in the elderly, even those
adverse events shortly after the initiation of treatment: with “normal” renal function as reflected by their serum
creatinine.
• The patients are generally sickest at the start of remission
induction. Pearl: Cyclophosphamide is dialyzed.
• A sizeable percentage of patients experience significant Comment: This is another way that renal failure complicates
renal dysfunction (and even end-stage renal disease) at this the management of WG. In addition to the importance of
time. dosing cyclophosphamide for renal failure, as described
• The doses of immunosuppression are highest just after above, cyclophosphamide must be given back to the patient
therapy has begun. after dialysis on the day when dialysis occurs. An approved
Pearl: Make sure your patient also knows that methotrexate and practice is to administer the patient’s morning dose as usual
daily double-strength TMP/SMZ should not be used together. and then to give a half-dose later in the day after the dialysis
session (Haubitz et al. 2002).
Comment: Because both methotrexate and TMP/SMZ are
folate antagonists, daily administration of double-strength Pearl: Methotrexate is contraindicated in patients with a
TMP/SMZ should be avoided. Prolonged courses of double- serum creatinine > 2.0 mg/dL.
strength TMP/SMZ and methotrexate can be disastrous,
Comment: Methotrexate is cleared in part through renal metab-
leading to bone marrow aplasia.
olism. The use of this medication in the setting of significant
Primary-care doctors who do not prescribe methotrexate
renal dysfunction can lead to severe bone marrow toxicity.
regularly may not recall this detail but are inclined to pre-
Although most clinicians get this right when prescribing meth-
scribe double-strength TMP/SMZ for a variety of complaints:
otrexate initially, methotrexate doses must also be adjusted in
urinary tract infections, upper respiratory disturbances, and
the setting of subsequent renal function declines. The case of
other problems that might not come to the attention of the
herpes zoster depicted in Fig. 24.20 illustrates this point.
patient’s rheumatologist. Moreover, patients may also get
one medicine from one pharmacy and the other from another Pearl: Be extra careful about monitoring for cytopenias dur-
pharmacy, thereby reducing the likelihood that the potential ing times of glucocorticoid dose reductions or during
24 ANCA-Associated Vasculitis 265

case series of 85 patients with MPA with longer follow-up,

disease flares occurred in approximately 33% (Guillevin et
al. 1999). In contrast, up to 80% of patients with WG flare
over time (WGET Research Group 2005; Reinhold-Keller et
al. 2000).
As WG is more frequently associated with antibodies
against PR3 than to MPO, while the opposite is true for MPA,
it may be the ANCA specificity that is associated with the
risk of flares. In addition to the data from the CYCAZAREM
trial, some retrospective data suggest that this is the case, as
within a group of patients with WG the relapse risk was
found to be significantly higher in PR3-ANCA than in MPO-
ANCA associated patients (Stegeman 2002).
Pearl: Check a patient’s thiopurine methyltransferase status
before starting azathioprine.
Comment: Azathioprine is commonly used as a remission
maintenance drug in ANCA-associated vasculitis. Although
the medication has been around for half a century, only
recently has it been realized that people are born with (or
without) genes necessary to metabolize the drug properly.
Azathioprine metabolism is governed largely by thiopurine
methyltransferase (TPMT) genes, which are inherited in an
Fig. 24.20 Zoster that occurred in a neutropenic patient with Wegener’s autosomal fashion. Approximately 90% of individuals are
granulomatosis on methotrexate (MTX). The patient’s serum creatinine born with two TPMT alleles and have no problem with aza-
was 1.7 mg/dL when MTX therapy began. However, his renal function thioprine metabolism. Approximately 9% of patients are
declined slowly during follow-up and his serum creatinine rose to
2.4 mg/dL. He presented with disseminated herpes zoster in the setting heterozygous for the TPMT mutation. In these patients, a “start
of a peripheral white blood cell count of 1.2 × 103/mm3 low and go slow” is reasonable, even though many heterozy-
gotes are able to tolerate large doses of azathioprine (Stassen et
al. 2008). Fifty milligrams per day is a reasonable starting dose
periods of transitioning from cyclophosphamide to another for TPMT heterozygotes. Less than 1% of individuals are
immunosuppressive agent. homozygous negative for TPMT and are at risk for disastrous
Comment: When cyclophosphamide is administered orally, bone marrow toxicity if they receive azathioprine.
the WBC has a tendency to fall once the prednisone dose is Pearl: The syndrome of azathioprine hypersensitivity has
reduced below 20 mg/day. nothing to do with TPMT alleles.
Because the action of cyclophosphamide on bone mar-
row function lasts for many months beyond the last dose, Comment: TPMT alleles go a long way toward determining
the initial period when another agent is begun (e.g., aza- patients’ susceptibility to bone marrow toxicity but do not
thioprine or methotrexate) is a time when the patient may affect the likelihood of certain other AZA side effects. For
suffer bone marrow suppression from the effects of both example, azathioprine hypersensitivity reactions can occur
cyclophosphamide and the new agent. quite severely even in patients with two TPMT alleles.
Avoid the tendency to “loosen up” on blood count moni- Similarly, gastrointestinal tract toxicity – severe nausea that
toring once the patient is off cyclophosphamide. The usual commences shortly after taking the medication – is not affected
starting doses of azathioprine (2 mg/kg/day), methotrexate by the patient’s number of TPMT alleles.
(15–25 mg/week), and mycophenolate (2,000 mg/day) are
sufficient to cause cytopenias by themselves and require espe-
cially close monitoring when these medications are given
after cyclophosphamide. References
Pearl: MPA is less likely than WG to flare.
Antineutrophil cytoplasmic autoantibody-associated pauci-immune
Comment: In the CYCAZAREM trial (Jayne et al. 2003), glomerulonephritis and vasculitis, and other vasculitides. In: D’Agati
disease flares occurred in 17 of the 92 patients with WG VD, Jennette JC, Silva FG (eds) Non-neoplastic kidney diseases (1st
(18%), compared with 4 of 52 (8%) of those with MPA. In a edn). Silver Spring, MD, ARP, 2005, pp. 385–424
266 J. H. Stone et al.

Bajema IM, Bruijn JA. What stuff is this! A historical perspective on Gera M, Griffin MD, Specks U, et al Recurrence of ANCA-associated
fibrinoid necrosis. J Pathol 2000; 191:235–238 vasculitis following renal transplantation in the modern era of
Bean SM, Conner MG. Wegener’s granulomatosis of the uterine cervix: immunosupression. Kidney Int 2007; 71:1296–1301
A case report and review of the literature. Int J Gynecol Pathol 2007; Goek ON, Stone JH. Randomized controlled trials in ANCA-associated
26:95–98 vasculitis. Curr Opin Rheumatol 2005; 17:257–264
Bienaime F, Clerbaux G, et al D-Penicillamine-induced ANCA-associated Guillevin L, Durand-Gasselin B, Cevallos R. Microscopic polyangiitis:
crescentic glomerulonephritis in Wilson disease. Am J Kidney Dis Clinical and laboratory findings in eighty-five patients. Arthritis
2007; 50(5):821–825 Rheum 1999; 42:421–430
Birnbaum J, Danoff S, Askin FB, Stone JH. Microscopic polyangiitis Gunton JE, Stiel J, et al Prevalence of positive anti-neutrophil cytoplas-
presenting as a ‘pulmonary-muscle’ syndrome: Does alveolar hem- mic antibody (ANCA) in patients receiving anti-thyroid medication.
orrhage contribute to the mechanism of pulmonary fibrosis?. Eur J Endocrinol 2000; 142(6):587
Arthritis Rheum 2007; 56(6):20–71 Haas M, Eustace JA. Immune complex deposits in ANCA-associated
Boomsma MM, Stegeman CA, van der Leij MJ, et al Prediction of crescentic glomerulonephritis: A study of 126 cases. Kidney Int
relapses in Wegener’s granulomatosis by measurement of antineu- 2004; 65:2145–2152
trophil cytoplasmic antibody levels. A prospective study. Arthritis Harris AA, Falk RJ, Jennette JC. Crescentic glomerulonephritis with a
Rheum 2000; 43:2025–2033 paucity of glomerular immunoglobulin localization. Am J Kidney
Bosch X, Guilabert A, Font J. Antineutrophil cytoplasmic antibodies. Dis 1998; 32:179–184
Lancet 2006; 368:404–418 Haubitz M et al Cyclophosphamide phramacokinetics and dose require-
Brons RH, de Jong MC, de Boer NK, et al Detection of immune depos- ments in patients with renal insufficiency. Kidney Int 2002; 61:
its in skin lesions of patients with Wegener’s granulomatosis. Ann 1495–1501
Rheum Dis 2001; 60:1097–1102 Hauer HA, Bajema IM, de Heer E, et al Distribution of renal lesions in
Carrington CB, Liebow A. Limited forms of angiitis and granulomato- idiopathic systemic vasculitis: A three-dimensional analysis of 87
sis of Wegener’s type. Am J Med 1966; 41:497–527 glomeruli. Am J Kidney Dis 2000; 36:257–265
D’Cruz DP, Baguley E, Asherson RA, et al Ear, nose, and throat Hauer HA, Bajema IM, van Houwelingen HC, et al Renal histology in
symptoms in subacute Wegener’s granulomatosis. BMJ 1989; 299: ANCA-associated vasculitis: Differences between diagnostic and
419–422 serologic subgroups. Kidney Int 2002; 61:80–89
Devaney KO, Travis WD, Hoffman G, et al Interpretation of head and Hoffman G, Kerr G, Leavitt R, et al Wegener’s granulomatosis: An
neck biopsies in Wegener’s granulomatosis. A pathologic study of analysis of 158 patients. Ann Intern Med 1992; 116:488–498
126 biopsies in 70 patients. Am J Surg Pathol 1990; 14:555–564 Holvast A, Stegeman CA, Benne CA, et al Wegener’s granulomatosis
Duggan CP, Westra SJ, Rosenberg AE. Case records of the patients show an adequate antibody response to influenza vaccina-
Massachusetts General Hospital. Case 23–2007. A 9-year-old boy tion. Ann Rheum Dis 2008 July 14. [Epub ahead of print]
with bone pain, rash, and gingival hypertrophy. N Engl J Med 2007; Hollander D, Manning RT. The use of alkylating agents in the treatment
357:392–400 of Wegener's granulomatosis. Ann Intern Med 1967; 67:393-8
Eisenberger U, Fakhouri F, Vanhille P, et al ANCA-negative pauci- Homma S, Matsushita H, Nakata K. Pulmonary fibrosis in myeloper-
immune renal vasculitis: histology and outcome. Nephrol Dial oxidase antineutrophil cytoplasmic antibody-associated vasculiti-
Transplant 2005; 20:1392–1399 des. Respirology 2004; 9(2):190–196
Eschun GM, Mink SN, Sharma S. Pulmonary interstitial fibrosis as a Javaud N, Belenfant X, Stirnemann J, et al Renal granulomatoses: A
presenting manifestation in perinuclear antineutrophilic cytoplas- retrospective study of 40 cases and review of the literature. Medicine
mic antibody microscopic polyangiitis. Chest 2003; 123:297–301 (Baltimore) 2007; 86:170–180
FDA Public Health Notification: Complications from metallic tracheal Jayne D, Rasmussen N, Bacon, et al Randomised trial of cyclo-
stents in patients with benign airway disorders. http://www.fda.gov/ phosphamide versus azathioprine during remission in ANCA-
cdrh/safety/072905-tracheal.html associated systemic vasculitis. N Engl J Med 2003; 349: 36–44
Feriozzi S, Muda AO, et al Cephotaxime-associated allergic interstitial Jayne DR, Gaskin G, Pusey CD, Lockwood CM. ANCA and predicting
nephritis and MPO-ANCA positive vasculitis. Ren Fail 2000; relapse in systemic vasculitis. QJM 1995; 88:127–133
22(2):245–251 Jayne DR, Gaskin G, Rasmussen N, et al Randomized trial of plasma
Finkielman JD, Lee AS, Hummel AM, et al ANCA are detectable in exchange or high-dosage methylprednisolone as adjunctive therapy
nearly all patients with active severe Wegener’s granulomatosis. Am for severe renal vasculitis. J Am Soc Nephrol 2007; 18:2180–2188
J Med 2007; 120:643.e9–e14 Kyndt et al Serial measurements of antineutrophil cytoplasmic autoan-
Finkielman JD, Merkel PA, Schroeder D, et al for the WGET Research tibodies in patients with systemic vasculitis. Am J Med 1999;
Group. Antineutrophil cytoplasmic antibodies against proteinase 3 106:527–533
do not predict disease relapses in Wegener’s granulomatosis. Ann Langford C, Talar-Williams C, Barron K, et al A staged approach to the
Intern Med 2007 147(9):611–9 treatment of Wegener’s granulomatosis: Induction of remission
Franssen C, Gans R, Kallenberg C, et al Disease spectrum of patients with glucocorticoids and daily cyclophosphamide switching to
with antineutrophil cytoplasmic autoantibodies of defined specific- methotrexate for remission maintenance. Arthritis Rheum 1999; 42:
ity: Distinct differences between patients with anti-proteinase 3 2666–2673
and anti-myeloperoxidase autoantibodies. J Intern Med 1998; 244: Leavitt RY, Fauci AS, Bloch DA, et al The American College of
209–216 Rheumatology 1990 criteria for the classification of Wegener’s
Gallagher H, Kwan JT, Jayne DR. Pulmonary renal syndrome: A 4-year, granulomatosis. Arthritis Rheum 1990; 33:1101–1107
single-center experience. Am J Kidney Dis 2002; 39:42–47 Luqmani RA, Bacon PA, Beaman, et al Classical versus non-renal
Geetha D, Seo P. Renal transplantation in the ANCA-associated vascu- Wegener’s granulomatosis. Q J Med 1994; 87:161–167
litides. Am J Transplant 2007; 7:2657–2662 Mahr AD, Neogi T, LaValley MP, et al Assessment of the item selection
Geffriaud-Ricouard C, Noël LH, Chauveau D. Clinical significance of and weighting in the Birmingham Vasculitis Activity Score for
ANCA in 98 patients. Adv Exp Med Biol 1993; 336:273–279 Wegener’s Granulomatosis (BVAS/WG) (submitted)
Genereau T, Lortholary O, Pottier M, et al Temporal artery biopsy. A Merkel PA, Lo GH, Holbrook JT, et al for the WGET Research Group.
diagnostic tool for systemic necrotizing vasculitis. Arthritis Rheum Incidence of venous thrombotic events among patients with
1999; 42:2674–2681 Wegener’s granulomatosis. Ann Intern Med 2005; 142:620–626
24 ANCA-Associated Vasculitis 267

Nachman PH, Hogan SL, Jennette JC, Falk RJ. Treatment response and Stassen PM, Sanders JS, Kallenberg CG, Stegeman CA. Influenza vacci-
relapse in antineutrophil cytoplasmic autoantibody-associated nation does not result in an increase in relapses in patients with ANCA-
microscopic polyangiitis and glomerulonephritis. J Am Soc Nephrol associated vasculitis. Nephrol Dial Transplant 2008; 23:654–658
1996; 7(1):33-9 Stassen PM, Derks RPH, Kallenberg CGM, Stegeman CA.
Nachman PH, Segelmark M, Westman K, et al Recurrent ANCA- Thiopurinemethyltransferase (TPMT) genotype and TPMT activity
associated small vessel vasculitis after transplantation: A pooled in patients with anti-neutrophil cytoplasmic antibody-associated vas-
analysis. Kidney Int 1999; 56:1544–1550 culitis: relation to azathioprine maintenance treatment and adverse
Nasr SH, D’Agati VD, Park, et al Necrotizing and crescentic lupus effects. Ann Rheum Dis Ann Rheum Dis 2009; 68(5):758–9
nephritis with antineutrophil cytoplasmic antibody seropositivity. Stegeman CA, Cohen Tervaert JW, Sluiter WJ, Manson WL, de Jong
Clin J Am Soc Nephrol 2008; 3:682–690 PE, Kallenberg CGM. Association of nasal carriage of Staphylococcus
Niles JL, Böttinger EP, Saurina GR, et al The syndrome of lung hemor- aureus and higher relapse in Wegener’s Granulomatosis. Ann Intern
rhage and nephritis is usually an ANCA-associated condition. Arch Med 1994; 120:12–17
Intern Med 1996; 156:440–445 Stegeman CA. Anti-neutrophil cytoplasmic antibody (ANCA) levels
Polychronopoulos VS, Prakash UB, Golbin JM, et al Airway involve- directed against proteinase-3 and myeloperoxidase are helpful in
ment in Wegener’s granulomatosis. Rheum Dis Clinics N America predicting disease relapse in ANCA-associated small-vessel vasculi-
2007; 33:755–775 tis. Nephrol Dial Transplant 2002; 17(12):2077-80
Reinhold-Keller E, Beuge N, Latza U, et al An interdisciplinary approach Stone J, Millward C, Criswell L. Two Unusual Genitourinary
to the care of patients with Wegener’s granulomatosis. Long term Manifestations Of Wegener’s Granulomatosis. The Journal of
outcome in 155 patients. Arthritis Rheum 2000; 43(5):1021–1032 Rheumatology 1997; 24(11): 1846-1848
Robinson MR, Lee SS, Sneller MC, et al Tarsal-conjunctival disease Stone JH: and the Wegener’s Granulomatosis Etanercept Trial Research
associated with Wegener’s granulomatosis. Ophthalmology 2003; Group. Limited versus severe Wegener’s granulomatosis: Baseline
110:1770–1780 data on patients in the Wegener’s granulomatosis etanercept trial.
Rostaing L, Modesto A, Oksman F et al Outcome of patients with anti- Arthritis Rheum 2003; 48:2299–2309
neutrophil cytoplasmic autoantibody-associated vasculitis follow- Stone JH, Hoffman GS, Wegener’s Granulomatosis. In: Hochberg MC,
ing cadaveric kidney transplantation. Am J Kidney Dis 1997; Smolen JS, Weinblatt ME, Weisman MH (eds) Rheumatology (4th
29(1):96-102 edn). London, Elsevier, 2006, pp. 72.1–72.25
Sadiq Z, Shepherd R, Vardy S. Posterior scleritis: An unusual presenta- Suryaprasad A, Stone JH. When is it safe to stop Pneumocystis carinii
tion of facial pain. Br J Oral Maxillofac Surg 2003; 41:62–63 (jiroveci) prophylaxis? Insights from cases complicating autoim-
Sakai N, Wada T, et al Tubulointerstitial nephritis with anti-neutrophil mune diseases. Arthritis Care & Research 2008; 59(7):1034-9
cytoplasmic antibody following indomethacin treatment. Nephrol The WGET Research Group. Design of the Wegener’s Granulomatosis
Dial Transplant 1999; 14(11):2774 Etanercept Trial (WGET). Controlled Clinical Trials 2002; 23:
Saxena R, Bygren P, Arvastson B, Wieslander J. Circulating autoanti- 450–468
bodies as serological markers in the differential diagnosis of pulmo- The WGET Research Group. Etanercept plus standard therapy in patients
nary renal syndrome. J Intern Med 1995; 238:43–52 with wegener’s granulomatosis. N Engl J Med 2005; 352:351–361
Slot MC, Links TP, et al Occurrence of antineutrophil cytoplasmic anti- Wainright J, Davson J. The renal appearances in the microscopic form
bodies and associated vasculitis in patients with hyperthyroidism of periarteritis nodosa. J Pathol Bacteriol 1950; 62:189–196
treated with antithyroid drugs: A long-term followup study. Arthritis Wegener F. Uber generalisierte, septische Gefasserkrankungen. Verh
Rheum 2005; 53(1):108–113 Dtsch Ges Pathol 1936; 29:202
Small P, Brisson M-L. Wegener’s granulomatosis presenting as tempo- Wegener F. Über eine eigenartige rhinogene Granulomatose mit beson-
ral arteritis. Arthritis Rheum 1991; 34:220–223 derer Beteiligung des Arteriensystems und der Nieren. Beitr Pathol
Sneller MC, Hoffmann GS, Talar-Williams C, et al An analysis of forty- 1939; 102:36
two Wegener’s granulomatosis patients treated with methotrexate Wiesner O, Russell KA, Lee AS, et al Antineutrophil cytoplasmic anti-
and prednisone. Arthritis Rheum 1995; 38:608–613 bodies reacting with human neutrophil elastase as a diagnostic
Specks U et al Design of the rituximab in ANCA-associated vasculitis marker for cocaine-induced midline destructive lesions but not auto-
(RAVE) trial. 2009 (in press) immune vasculitis. Arthritis Rheum 2004; 50:2954–2965
 Cryoglobulinemia
25
Dimitrios Vassilopoulos

Myth: The majority of patients with mixed cryoglobulinemia

25.1 Overview of Cryoglobulinemia develop clinical manifestations of cryoglobulinemic vasculitis.
Reality: Although circulating mixed cryoglobulins (type II or
› Cryoglobulins are immunoglobulins that precipitate
III) are detected in a number of infectious, autoimmune, or
from serum under laboratory conditions of cold. The
malignant diseases, the development of cryoglobulinemic
usual laboratory temperature used to precipitate cryo-
vasculitis occurs in only a small minority. For example, in
globulins is 4°C.
patients who have chronic HCV infections, cryoglobulins are
› Cryoglobulinemia is divided into three clinical subsets
detected in approximately 40% using traditional methods of
(types I, II, and III). This classification is based on two
detection (Vassilopoulos et al. 2002). Despite the high preva-
features: (1) the clonality of the IgM component and
lence of cryoglobulins, the syndrome of cryoglobulinemic
(2) the presence of rheumatoid factor activity.
vasculitis occurs in less than 5% of cases (Vassilopoulos et al.
› In its clinical manifestations, type I cryoglobulinemia
2002).
is usually quite distinct from types II and III. In con-
trast, substantial clinical overlap exists between types Pearl: The likelihood of cryoglobulinemic vasculitis increases
II and III. in proportion to the length of time the patient has been
› Type I cryoglobulinemia is associated with a monoclo- infected with HCV.
nal component and is typically associated with a
Comment: Patients have usually been infected with HCV for
hematopoietic malignancy.
many years by the time cryoglobulinemic vasculitis appears.
› Type II and III cryoglobulinemia are often termed
A common scenario is for a patient who received a blood
“mixed” cryoglobulinemias, as they are comprised of
transfusion 25 years ago to present with palpable purpura of
both IgG and IgM components.
the lower extremities, heralding the diagnosis of cryoglobu-
› Vasculitis associated with mixed cryoglobulinemia
linemia (and HCV).
involves both small- and medium-sized blood vessels.
The relationship between the length of time a patient has
Small-vessel disease is more common than medium-
been infected with HCV and the likelihood of cryoglobuline-
vessel disease.
mic vasculitis is reflected in the peak age at presentation with
› Vasculitis associated with mixed cryoglobulinemia is
vasculitis, which is generally between the ages 50 and 70
caused by hepatitis C virus (HCV) infections in approx-
years (Vassilopoulos et al. 2002). In this regard, HCV and
imately 90% of the cases. In such patients, the antigens
cryoglobulinemia differ substantially from hepatitis B viral
involved in cryoglobulinemic vasculitis are portions of
(HBV) infections and polyarteritis nodosa. With HBV, pol-
the (HCV) virion. The relevant antibodies are both IgG
yarteritis nodosa typically develops either during the acute
and IgM, leading to the designation “mixed”
HBV infection or in the early stages of chronic HBV infec-
cryoglobulinemia.
tion. Thus, most cases of HBV-associated polyarteritis
› Virtually, all patients with mixed cryoglobulinemia are
nodosa present within 1 year of HBV infection.
rheumatoid factor positive.
› The treatment of cryoglobulinemia of any of the three Pearl: False-negative results in testing for cryoglobulins are
types is directed whenever possible at the underlying common.
cause. In some cases, the broad-spectrum immunosup-
Comment: Sensitive testing for cryoglobulins requires an
pression and other measures must be employed with
experienced laboratory that is set up to perform the collection
glucocorticoids, cytotoxic therapies, and plasma
in the proper conditions. While the patient is fasting (lipids
exchange.
can interfere with the assay), at least 20 mL of blood should

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 269

DOI:10.1007/978-1-84800-934-9_25, © Springer Science + Business Media B.V. 2009
270 D. Vassilopoulos

Fig. 25.1 Characteristic purpuric lesions in a patient with chronic HCV

infection and associated type II cryoglobulinemic vasculitis. New ery-
thematous lesions and old hyperpigmented lesions are evident in the
lower extremities of this patient (Figure courtesy of Dr. Vassilopoulos)

Fig. 25.2 Severe digital ischemia in a patient with type I cryoglobu-

be drawn into a tube that has not been treated with anticoagu- linemia. Although most patients with symptomatic type I cryoglobu-
lant. The specimen should be transported and centrifuged at linemia manifest features of hyperviscosity, digital ischemia can occur
in some. The precise basis of this digital ischemia is not always clear.
37°C, and then kept for more than 72 h at 4°C, to allow for This patient had multiple myeloma and a cryocrit of 48%. She responded
the precipitation of cryoglobulins (Musset et al. 1992). well to melphalan and prednisone, and entered a remission sustained by
Persistence (i.e., repeat testing and coordination with the thalidomide for several years (Figure courtesy of Dr. John Stone)
laboratory) may be required to obtain a good assay. Clues
that suggest a harder look for cryoglobulins is in order
include positive HCV serologies and a serum C4 level that is a condition that does not occur under physiologic conditions
depressed out of proportion to the decrease in C3. because of the maintenance of body temperature at approxi-
mately 35–37°C, even in the acral extremities.
Myth: Purpura is a persistent manifestation of patients with
In contrast, type I cryoglobulins can precipitate at tem-
type II or III cryoglobulinemic vasculitis.
peratures achieved in the distal extremities, nose, ears, and
Reality: In most patients with cryoglobulinemic vasculitis, elsewhere. Type I cryoglobulins often present with cold-
purpura is the presenting symptom. It is usually located on the induced symptoms, and are then termed “cryoprecipitating
lower extremities and is not associated with pruritus (Gorevic monoclonal gammopathies.”
et al. 1980; Vassilopoulos et al. 2002). In the majority of cases
Pearl: A disproportionate decrease in C4 levels provides an
of HCV-related cutaneous vasculitis, purpura has an intermit-
important clue to the diagnosis of cryoglobulinemia.
tent pattern of appearance – even without specific treatment.
In patients, with chronic purpura, residual hyperpigmentation Comment: The presence of significant levels of cryoglobu-
of the lower extremities is a common finding (Fig. 25.1). lins is often associated with very low or undetectable levels
of C4 and normal levels of C3. This is most common in type
Pearl: Type I cryoglobulinemia is more likely to be associ-
II cryoglobulinemia. Low complement levels may also be
ated with syndromes of hyperviscosity than with vasculitis.
seen in association with type II cryoglobulinemia complicat-
Comment: Symptoms of sludging within the central nervous ing Sjögren’s syndrome or, rarely, in type III cryoglobuline-
system are typical in type I cryoglobulinemia because of the mia in patients with HCV infection, rheumatoid arthritis,
burden of cryoproteins in that disorder. (Type I cryoglobu- vasculitis, or SLE (Kallemuchikkal and Gorevic 1999).
linemia is often associated with an extremely high cryocrit). Interestingly, the C4 level may be persistently low, even
In contrast, diffuse, confluent purpura is atypical of type I when cryoglobulins have become undetectable following
cryoglobulinemia. Type I cryoglobulinemia can cause intense treatment. Presumably there are sufficient immune com-
digital ischemia, however, as shown in Fig. 25.2. plexes to activate complement, but not enough to be detected
as cryoglobulins.
Myth: Cryoglobulinemic symptoms are exacerbated by win-
ter and other conditions of cold. Myth: Renal disease is often detectable in the early stages of
cryoglobulinemia.
Reality: The cryoglobulins associated with HCV (types II and
III) are immune complexes that precipitate in cold laboratory Reality: In contrast to other forms of immune-complex medi-
conditions (e.g., refrigeration for several days at 4°C). This is ated glomerulonephritis such as lupus and Henoch–Schönlein
25 Cryoglobulinemia 271

purpura, renal involvement in the form of membranoprolif- Reality: It is generally accepted that chronic HCV viremia is
erative glomerulonephritis (MPGN) is usually a late mani- responsible for the appearance of serum cryoglobulins and the
festation of cryoglobulinemia. MPGN typically appears 3–5 clinical manifestations of cryoglobulinemia in patients who
years after the onset of purpura (D’Amico 1998; Roccatello have HCV-associated disease. Most patients with HCV-
et al. 2007). associated cryoglobulinemia who clear the virus after antiviral
The presentation of MPGN in cryoglobulinemia is usu- treatment demonstrate disappearance of the cryoglobulins from
ally associated with asymptomatic hematuria or proteinuria the serum and clinical remission (Vassilopoulos et al. 2002).
(Roccatello et al. 2007). In approximately 20% of cases, pro- However, a number of recent studies have indicated that
teinuria in cryoglobulinemia-associated MPGN reaches the some patients with HCV-associated cryoglobulinemia con-
nephrotic range. Flares of glomerulonephritis during long- tinue to have vasculitic symptoms and to maintain cryoglob-
term follow-up are common. However, the overall prognosis ulins in their serum, despite the absence of detectable virus
of renal disease is good, with only about 15% of patients (HCV RNA) in the blood or bone marrow (Landau et al.
developing end-stage renal disease over long-term follow-up 2008; Levine et al. 2005). An underlying non-Hodgkin lym-
(D’Amico 1998). Nevertheless, renal involvement at presen- phoma (NHL) could be responsible for the persistent cryo-
tation is an independent variable associated with poor sur- globulinemia in some of these cases. Thus, a thorough
vival (Ferri et al. 2004; Roccatello et al. 2007). evaluation to exclude NHL is recommended in this setting
(Landau et al. 2008).
Myth: Arteritis is a common finding in renal biopsies of
patients with HCV-associated cryoglobulinemia. Myth: The cryocrit correlates well with disease activity in
cryoglobulinemic vasculitis.
Reality: Renal involvement occurs in approximately 30% of
patients with HCV-associated cryoglobulinemia, usually (as Reality: This is particularly incorrect in the setting of type I
noted) in the form of MPGN (Ferri et al. 2004). Renal biopsy cryoglobulinemia, where clinical responses (e.g., improve-
also shows mesangial proliferative or focal membranoprolif- ment in digital ischemia) may occur in the face of cryocrits
erative glomerulonephritis in some patients (Roccatello et al. that continue to be frighteningly high. When forced to use
2007). Arteritis involving small- or medium-sized vessels is immunosuppression to treat cryoglobulinemia, the cryocrit
an uncommon finding, occurring in less than 10% of patients is just like almost every biomarker ever described: treat the
(Roccatello et al. 2007). In this way, cryoglobulinemic glom- patient and not the cryocrit.
erulonephritis resembles lupus nephritis, another form of Pearl: NHL is a known complication of cryoglobulinemia.
immune-complex-mediated renal disease in which the find-
ing of arteritis in the kidney is rare. Comment: Patients with cryoglobulinemia with or without
HCV infection have a higher than normal risk of developing
Pearl: True inflammatory arthritis is a rare manifestation of hematologic malignancies, particularly NHLs. NHL in this
cryoglobulinemia. setting, reported in approximately 5% of patients with cryo-
Comment: Many patients with cryoglobulinemia are misdi- globulinemia, usually belong to the lymphoplasmacytic or
agnosed early in their course as having rheumatoid arthritis. diffuse large cell type (Ferri et al. 2004; Trejo et al. 2003).
There are two reasons for this: (1) the high prevalence of NHL presents with extra-nodal involvement in 70% of
joint complaints in cryoglobulinemia and (2) the rheumatoid cases associated with cryoglobulinemia (Trejo et al. 2003).
factor activity demonstrated by the IgM component of the The usual extra-nodal sites are the liver, salivary glands, and
cryoglobulins. stomach (Trejo et al. 2003). NHL accounts for approximately
Although symmetric polyarthralgias are common in 15% of the deaths in patients with cryoglobulinemia (Ferri
patients with cryoglobulinemia (70–80%), frank synovitis is et al. 2004).
unusual (2–5%) (Rosner et al. 2004; Vassilopoulos and Myth: Mononeuritis multiplex is the most common form of
Calabrese 2003). The small joints of the hands are the joints neuropathy in cryoglobulinemic vasculitis.
most commonly affected by arthralgias, followed by the
knees. The absence of true synovitis, the absence of erosive Reality: Peripheral nerve involvement is very common in
changes on joint radiographs, and the absence of antibodies cryoglobulinemic vasculitis, occurring eventually in up to
to cyclic citrullinated peptides (anti-ccp) are all useful in dis- 80% of cases. The most common type of nerve dysfunction is
tinguishing cryoglobulinemia from rheumatoid arthritis a distal symmetric sensory polyneuropathy that has a predilec-
(Vassilopoulos and Calabrese 2003). tion for the lower extremities. Mononeuritis multiplex is the
second most common form of peripheral nerve involvement.
Myth: Disappearance of HCV from the serum of patients A number of pathologic lesions are found in nerve biop-
with HCV-associated cryoglobulinemia is always associated sies from affected patients, including vasculitis of the small-
with clinical remission. and/or medium-sized vessels or perivascular infiltrates
272 D. Vassilopoulos

(Authier et al. 2003). The neuropathy of cryoglobulinemic Pearl: Interferon-alpha can exacerbate HCV-associated
vasculitis is less responsive to antiviral and immunosuppres- cryoglobulinemic vasculitis.
sive therapy than are some other manifestations of cryoglob-
Comment: Interferon-alpha (IFN-a) in either its standard or
ulinemic vasculitis (e.g., purpura and nephropathy). This is
pegylated form remains the cornerstone of treatment for
probably a consequence of the fact that peripheral nerves,
chronic hepatitis C. In patients with HCV-associated cryo-
once injured, require many months to recover and continue
globulinemia, IFN-based regimens have been associated
to be symptomatic for long periods of time.
with favorable virologic and clinical outcomes in patients
Some data suggest that mononeuritis multiplex responds
when administered either with ribavirin or in combination
better than the distal polyneuropathy to therapy (Saadoun
with immunosuppressive agents (Saadoun et al. 2006, 2008;
et al. 2006). Encouraging results in the treatment of cryo-
Vassilopoulos et al. 2002).
globulinemic neuropathy have been reported using rituximab
However, clinicians should be aware that IFN-a can exac-
either alone or in combination with antiviral therapy in
erbate every single manifestation of cryoglobulinemia includ-
patients with HCV-associated cryoglobulinemic neuropathy
ing purpura, glomerulonephritis, or peripheral neuropathy
(Cacoub et al. 2008; Saadoun et al. 2008).
(Vassilopoulos et al. 2002). This is of particular clinical
Pearl: HCV-associated cryoglobulinemia can present with a importance when treatment decisions are made for patients
pulmonary–renal syndrome. with severe disease manifestations such as severe renal dis-
ease, gastrointestinal vasculitis, or digital necrosis. In such
Comment: A proliferative glomerulonephritis is a well- cases, immunosuppressive therapy (glucocorticoids, cyclo-
known complication of cryoglobulinemic vasculitis. Less phosphamide, rituximab) should be administered first to sup-
common complications of HCV infection can involve the press the immune response of the host. After control of the
lung. Both pulmonary hemorrhage and interstitial fibrosis immune response, antiviral therapy can be initiated with less
have been reported in patients with HCV infection (Prasad concern about a worsening of clinical manifestations.
et al. 2006). The potential association between HCV and
lung dysfunction is worth keeping in mind when the pieces Myth: HBV is a common etiologic factor for cryoglobu-
of a puzzling pulmonary case do not quite add up. linemia.

Myth: Immunosuppressive therapy is frequently associated Reality: Although studies in the early 1980s suggested that
with flares of the underlying hepatitis in patients with HCV- HBV causes a substantial proportion of cryoglobulinemic vas-
associated cryoglobulinemia. culitis cases, these data have been refuted in subsequent inves-
tigations. One must wonder if the cases of cryoglobulinemic
Reality: In patients with severe manifestations of cryoglobu- vasculitis linked to HBV in those early studies were not exam-
linemic vasculitis, immunosuppressive therapy with high-dose ples of patients coinfected with both HBV and “non-A/non-B
glucocorticoids and cyclophosphamide is employed frequently hepatitis virus,” as HCV was known before its identification in
(Vassilopoulos et al. 2002). This type of treatment was used the late 1980s. HBV now accounts for less than 5% of cryo-
for years, even prior to the discovery of HCV as the principal globulinemic vasculitis cases (Ferri et al. 2004).
etiology of cryoglobulinemia (Vassilopoulos et al. 2002). Nevertheless, a search for HBV infection is obligatory for
In contrast to HBV infection, in which immunosuppres- patients who present with cryoglobulinemia, because HBV
sives (especially glucocorticoids) frequently lead to rebound often coincides with HCV (and human immunodeficiency
viremia and hepatitis flares, a number of observational stud- virus) infections in high risk groups. An early diagnosis of
ies over the years have shown that such complications of HBV infection has significant clinical implications because
immunosuppression are uncommon events in patients with its treatment can prevent the detrimental effects of immuno-
HCV-associated cryoglobulinemia (Vassilopoulos et al. suppression in that patient population.
2002). This is true also for the newer modalities of treatment
Pearl: Ribavirin can cause significant hemolysis in patients
(e.g., rituximab), which can cause a transient increase in
with HCV-associated cryoglobulinemia and renal dysfunction.
HCV RNA levels without immediate harmful effects on liver
function (Saadoun et al. 2008). Comment: The combination of ribavirin and IFN-a (in either
Although these observations on the short-term safety of its standard or pegylated form) is the standard of care for the
immunosuppression are reassuring, the full impact of long- treatment of chronic hepatitis C. This regimen is also effective
term immunosuppression on liver function in HCV-infected for patients with HCV-associated cryoglobulinemia (Saadoun
cryoglobulinemic patients is not known. It is always impor- et al. 2006). However, extreme caution is required when using
tant to employ immunosuppressive therapy only when nec- ribavirin in patients with renal dysfunction. In its usual dose
essary and to minimize the length of time that patients are (800–1,200 mg/day), ribavirin can cause severe hemolytic
exposed to these potentially toxic agents. anemia during the first 2 weeks of its administration.
25 Cryoglobulinemia 273

Ribavirin is contraindicated in patients whose creatinine D’Amico G. Renal involvement in hepatitis C infection: Cryoglobulinemic
clearance is less than 50 mL/min. In patients with borderline glomerulonephritis. Kidney Int 1998; 54:650–671
Ferri C, Sebastiani M, Giuggioli D, et al Mixed cryoglobulinemia:
renal function, a slow upward titration of the dose accompa- Demographic, clinical, and serologic features and survival in 231
nied by close monitoring of the hematocrit is imperative. In patients. Semin Arthritis Rheum 2004; 33:355–374
many cases, patients cannot tolerate even low doses of riba- Gorevic PD, Kassab HJ, Levo Y, et al Mixed cryoglobulinemia: Clinical
virin and the medication must be stopped. aspects and long-term follow-up of 40 patients. Am J Med 1980;
69:287–308
Pearl: Use plasmapheresis when patients present with severe Kallemuchikkal U, Gorevic P. Evaluation of cryoglobulins. Arch Pathol
Lab Med 1999; 123:119–125
systemic vasculitis.
Landau DA, Saadoun D, Halfon P, et al Relapse of hepatitis C virus-
Comment: Physicians vary in their threshold for resorting to associated mixed cryoglobulinemia vasculitis in patients with sus-
tained viral response. Arthritis Rheum 2008; 58:604–611
plasmapheresis, an intervention that has potential side-effects Levine JW, Gota C, Fessler BJ, et al Persistent cryoglobulinemic vascu-
(particularly infection) and is unproven for the treatment of litis following successful treatment of hepatitis C virus. J Rheumatol
most of the disorders in which it is employed. However, the 2005; 32:1164–1167
absence of evidence does not constitute evidence of absence, Musset L, Diemert MC, Taibi F, et al Characterization of cryoglobulins
by immunoblotting. Clin Chem 1992; 38:798–802
and there are certain clinical situations in which the applica- Prasad M, Buller GK, Mena CI, Sofair AN. Sum of the parts. N Engl J
tion of plasmapheresis makes good sense. Overwhelming Med 2006; 355:2468–2473
systemic vasculitis associated with cryoglobulinemia is one Roccatello D, Fornasieri A, Giachino O, et al Multicenter study on
such instance. In these cases, the removal of cryoglobulins hepatitis C virus-related cryoglobulinemic glomerulonephritis. Am
J Kidney Dis 2007; 49:69–82
and their associated immune complexes, as quickly as pos- Rosner I, Rozenbaum M, Toubi E, et al The case for hepatitis C arthri-
sible, is important. There is no quicker way to accomplish tis. Semin Arthritis Rheum 2004; 33:375–387
this than by plasmapheresis. Saadoun D, Resche-Rigon M, Thibault V, et al Antiviral therapy for
hepatitis C virus-associated mixed cryoglobulinemia vasculitis:
A long-term followup study. Arthritis Rheum 2006; 54:3696–3706
Saadoun D, Resche-Rigon M, Sene D, et al Rituximab combined with
peg-interferon-ribavirin in refractory HCV-associated cryoglobu-
linemia vasculitis. Ann Rheum Dis 2008;67;1431–1436
References Trejo O, Ramos-Casals M, Lopez-Guillermo A, et al Hematologic
malignancies in patients with cryoglobulinemia: Association with
autoimmune and chronic viral diseases. Semin Arthritis Rheum
Authier FJ, Bassez G, Payan C, et al Detection of genomic viral RNA 2003; 33:19–28
in nerve and muscle of patients with HCV neuropathy. Neurology Vassilopoulos D, Calabrese LH. Rheumatic manifestations of hepatitis
2003; 60:808–812 C infection. Curr Rheumatol Rep 2003; 5:200–204
Cacoub P, Delluc A, Saadoun D, et al Anti-CD20 monoclonal antibody Vassilopoulos D, Calabrese LH. Hepatitis C virus infection and vascu-
(rituximab) treatment for cryoglobulinemic vasculitis: Where do we litis. Implications of antiviral and immunosuppressive therapies.
stand? Ann Rheum Dis 2008; 67:283–287 Arthritis Rheum 2002; 46:585–597
 Polyarteritis Nodosa
26
John H. Stone

are not caused by antibodies directed against PR3 or MPO,

13.1 Overview of Polyarteritis Nodosa but rather by antibodies to antigens such as lactoferrin, bac-
tericidal permeability inhibitor, or others (Talor et al. 2007).
› Polyarteritis nodosa (PAN) primarily affects medium- If enzyme immunoassays for PR3- and MPO-ANCA are
sized arteries that supply the skin, gut, nerve, and kid- negative, then positive immunofluorescence assays are usu-
ney. However, multiple other organs can be involved ally “false-positive” results, at least with regard to diagnoses
either clinically or subclinically. of systemic vasculitis.
› Microaneurysms of arteries to or within the kidneys,
Pearl: The seronegative nature of PAN makes the diagnosis
liver, or gastrointestinal tract are highly characteristic
elusive.
of PAN.
› Mononeuritis multiplex, an asymmetric sensory and Comment: Most cases of systemic lupus erythematosus
motor neuropathy due to ischemia and infarction of (SLE) are not difficult to diagnose these days. Once the clini-
peripheral nerves, occurs frequently in PAN. cian considers the possibility of an autoimmune disease, the
› In mononeuritis multiplex, nerve conduction studies of vast majority of patients with true SLE are found to be anti-
peripheral nerves reveal a distal, asymmetric, axonal nuclear antibody (ANA) positive and to have more specific
neuropathy involving both motor and sensory nerves. serologic tests that clarify the diagnosis. (Unfortunately, the
› PAN is characterized pathologically by patchy, trans- use of enzyme immunoassays with poor sensitivity has
mural inflammation in medium- and small-sized mus- caused a resurgence of “ANA-negative lupus” in recent
cular arteries, sparing large arteries, capillaries, and years.)
the venous system. The inflammation leads to fibrinoid Similarly, even if clinicians overlook subtle signs of scle-
necrosis, but is not associated with granulomatous rodactyly, they may eventually get to the right diagnostic
features. ballpark and think of scleroderma if they observe a strongly
› High-dose glucocorticoids are the mainstay of therapy positive ANA assay and recognize that additional history,
in PAN. In cases that are rapidly progressive or life or physical examination, and more detailed autoantibody test-
organ threatening, however, cyclophosphamide is ing are in order.
added to glucocorticoid treatment. In contrast, many patients with PAN go undiagnosed for
weeks or months because routine tests for “collagen vascular
diseases” or “connective tissue disorders” are negative. PAN
Myth: PAN is sometimes associated with antineutrophil cyto-
is not associated with ANA, rheumatoid factor, or other
plasmic antibodies (ANCA).
known autoantibodies. The fact is the task of teasing out the
Reality: This Myth is a holdover from the early days of frequently subtle presentation of a subacute, multiorgan sys-
ANCA testing, when reliable enzyme immunoassays for tem disease or the detection of foot dorsiflexion weakness is
antibodies to proteinase-3 (PR3) and myeloperoxidase much more difficult than checking off lists of assays on a
(MPO) were not available. Classic PAN is an ANCA-negative laboratory slip.
disease, meaning that it is neither associated with PR3- nor
Pearl: Classic PAN is often a curable illness.
with MPO-ANCA.
Many types of disorders, even many types of nonvascu- Comment: In contrast to Wegener’s granulomatosis (WG),
litic diseases, are associated with ANCA when only immu- giant cell arteritis (GCA), and some other forms of vasculi-
nofluorescence testing is used (Stone et al. 2000). However, tis, “classic” PAN – the disorder described by Kussmaul and
positive ANCA assays in patients with PAN, inflammatory Maier – is usually a curable disease. Approximately 50% of
bowel disease, sarcoidosis, and liver or thyroid dysfunction PAN cases can be cured with glucocorticoid treatment alone,

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 275

DOI:10.1007/978-1-84800-934-9_26, © Springer Science + Business Media B.V. 2009
276 J. H. Stone

usually in courses lasting a total of 9–12 months (Guillevin Table 26.1 Forms of vasculitis most likely to cause vasculitic
and Lhote 1998). The remainder of PAN patients requires neuropathy
cyclophosphamide to induce disease remission. Once classic Polyarteritis nodosa
PAN is in remission, however, it usually stays there and does Microscopic polyangiitis
Churg-–Strauss syndrome
not require long-term maintenance therapy.
Wegener’s granulomatosis
The propensity of classic PAN to be cured was illustrated Cryoglobulinemia
by a study from the French Vasculitis Study Group that eval-
uated the outcomes of 351 patients with classic PAN
(Pagnoux et al. 2006). More than two-thirds of the patients Pearl: When selecting a peripheral nerve for biopsy, target
were cured by their initial treatment courses. an affected nerve but not a dead one.
Pearl: The development of mononeuritis multiplex in the Comment: Several forms of systemic vasculitis have par-
absence of diabetes mellitus or multiple compression inju- ticular predilections for causing vasculitic neuropathy
ries strongly suggests the presence of vasculitis. (Table 26.1). Which nerve offers the highest yield for mak-
ing the diagnosis through biopsy? Biopsies of peripheral
Comment: Mononeuritis multiplex is defined as a peripheral
nerves that are electrophysiologically normal are rarely
neuropathy in which named nerves are injured or infarcted
useful: such nerves are histologically normal, too. The sur-
one a time (Fig. 26.1). This syndrome is one of the great
geon should biopsy a nerve that is affected but not “dead.”
“eureka” signs in medicine, as its differential diagnosis is
If the sural nerves demonstrate no action potentials, the
relatively brief and its identification also points the way to a
superficial peroneal nerve and peroneus brevis muscle are
potential diagnostic test: a nerve biopsy.
other potential targets (Collins et al. 2000). In addition,
Peripheral nervous system vasculitis is a form of nerve
upper extremity nerves (e.g., the lateral antebrachial cuta-
dysfunction in which vascular inflammation is confined to
neous nerve or other nerves in the upper extremities) are
the vasa nervorum. The disorder can occur in an isolated
also candidates for biopsy. A contiguous muscle should be
fashion, involving only the peripheral nerves, or in the con-
biopsied during the same procedure.
text of a multiorgan system disorder such as several primary
vasculitides. Inflammatory disorders associated with autoim- Myth: Biopsy of a peripheral nerve that is demonstrably
munity such as rheumatoid arthritis, SLE, Sjögren’s syn- abnormal by EMG/NCV has a high likelihood of yielding
drome, and a variety of hematopoietic malignancies can also diagnostic tissue.
cause mononeuritis multiplex (Gorson 2001). However,
Reality: Don’t be overly optimistic about this. Useful infor-
these other disorders usually cause vasculitic neuropathy
mation is gained in the great majority of peripheral nerve
only months or years after the underlying disease has already
biopsies performed under such circumstances, and the pro-
been diagnosed. With PAN, in contrast, mononeuritis multi-
cedure is very worthwhile under the proper clinical circum-
plex typically occurs early in the disease course, and may be
stances. Nevertheless, even when a given nerve is
a primary feature of the disease by the time the patient gets
demonstrably abnormal on electrophysiologic testing, the
to medical attention.
chance that a nerve biopsy will provide incontrovertible
histologic proof of the diagnosis is only about 50%. Careful
clinicopathological correlation is required.
Nevertheless, a peripheral nerve biopsy usually provides
important clues to the presence of a vasculitic neuropathy.
This was illustrated by a study of 93 cases of vasculitic neu-
ropathy (Mathew et al. 2007). The following observations
were made:
• Eighty-four peripheral nerve biopsies were performed.
Eighty-two of these were sural nerve biopsies. Two were
biopsies of superficial radial nerves.
• The diagnosis of vasculitic neuropathy was confirmed in
49 of the 84 biopsies (58%).
• In an additional 31 cases (37%), the histopathological
results were “consistent with” but not diagnostic of vascu-
litic neuropathy. Most of those cases demonstrated an
acute, asymmetric loss of axons and differential involve-
Fig. 26.1 The effects of vasculitic neuropathy ment of nerve fascicles.
26 Polyarteritis Nodosa 277

When evaluating a patient for vasculitic neuropathy, the

electromyographer must perform EMGs of the flexor hallu-
cis brevis. Requesting this part of the examination specifi-
cally will also probably buy you a more careful procedure
and a more thoughtful interpretation, as the request implies
greater than average insight of the test orderer.
Pearl: The greatest challenge in vasculitic neuropathy lies in
managing the phase after the start of therapy.
Comment: Management of the long-term sequelae of nerve
infarctions in PAN – painful sensory neuropathies and residual
muscle weakness – poses major challenges. This is true for
vasculitic neuropathy caused by any form of vasculitis. Pain
Fig. 26.2 Muscle biopsy showing necrotizing vasculitis
control and physical and occupational therapy modalities
assume the highest priority once remission has been induced.
Thus, 95% of nerve biopsies (58 plus 37%) contributed A variety of newer agents have been employed to treat the
information critical to the diagnosis of vasculitic neuropathy. pain associated with peripheral nerve injury in recent years.
In the absence of another clear site to biopsy, it is usually For example, gabapentin and related medications have
worth biopsying a peripheral nerve that has been demon- become popular and widely used, particularly by neurolo-
strated to be abnormal by NVCs. gists. Tricyclic antidepressants (TCAs) should not be dis-
In many cases, the diagnosis can be inferred from a highly carded too quickly in the rush to embrace the new. These
consistent clinical history, NCVs that reveal an asymmetric, drugs remain the most effective medication for the treatment
length-dependent process, and compatible pathological find- of painful vasculitic neuropathy. TCAs should be regarded as
ings, even if the histology is not diagnostic. the cornerstone of therapy for this type of pain, the treatment
to which other interventions are added.
Pearl: Gastrocnemius muscle biopsies augment the yield of
sural nerve biopsies. Pearl: Observing a patient’s recovery from severe vasculitic
neuropathy is like watching grass grow, only much slower.
Comment: Biopsies of skeletal muscle increase the yield of
tissue sampling substantially, even in the absence of symp- Comment: Although the vasculitic process in PAN responds
toms, signs, or laboratory abnormalities that suggest skeletal readily to therapy, damage from the disease may be slower to
muscle involvement. Serum creatine kinase levels, for exam- resolve. Indeed, most patients do not recover full function of
ple, can be entirely normal even in the face of a raging vascu- both sensory and motor nerves that have been affected
litis that involves the arteries of muscles. Skeletal muscle is a severely. To the extent that recovery occurs, it takes many
highly vascular tissue, and biopsies in the search for vasculi- months and often continues over a year or more (thus, con-
tis often “hit paydirt” (Fig. 26.2). siderably slower than grass).
The pain of chronic nerve injury and the painfully slow
Pearl: Examination of the flexor hallucis brevis muscle on time to recovery underscore the importance of preventing this
electrophysiologic testing is important in the evaluation of complication whenever possible through early diagnosis and
length-dependent axonopathies. the appropriate application of aggressive immunosuppres-
Comment: Length-dependent axonopathies are those diseases sion. In this day and age, the presence of an active vasculitic
of nerve that affect first the longest nerves in the body, i.e., neuropathy related to PAN is an indication for cyclophosph-
usually those of the feet. Vasculitic neuropathy is one type of amide. It is not clear whether any biologic agent on the hori-
a length-dependent axonopathy. Others include diabetes mel- zon will change that any time soon.
litus, sarcoidosis, leprosy, and connective tissue diseases such
Myth: Punch biopsies of the skin are an easy way to make the
as SLE or Sjögren’s syndrome.
diagnosis of PAN.
EMG/NCVs of the flexor hallucis brevis can establish the
presence of motor involvement that is not otherwise appar- Reality: Classic PAN involves muscular, medium-sized arteri-
ent. This finding can alter the differential diagnosis substan- oles and arteries. The cutaneous location of such blood vessels
tially. Fibrillation potentials of the flexor hallucis brevis is in the deep dermis or subcutaneous fat (Fig. 26.3). No punch
muscle indicates subclinical involvement of motor fibers – biopsy extends deeply enough to reach the culprit vessels and
thereby pointing to a sensorimotor multiple mononeuropathy confirm the diagnosis of PAN (Stone and Nousari 2001).
if sensory nerve action potentials are also absent in the lower In contrast, if the cutaneous lesion is one of a small-vessel
extremities (Vallat and Cros 2007). vasculitides (e.g., palpable or nonpalpable purpura), a simple
278 J. H. Stone

Fig. 26.3 Depth of blood vessels

involved in different forms of
vasculitis

Fig. 26.4 Nodule from a patient with polyarteritis nodosa Fig. 26.5 Cutaneous ulcer in a patient with polyarteritis nodosa

punch biopsy should suffice for diagnostic purposes. In addi-

Comment: Medium-vessel vasculitis is associated with four pri-
tion to the standard hematoxylyn and eosin stains, immuno-
mary skin lesions: nodules, ulcers, livedo reticularis, and digital
fluorescence studies are critical in parsing the different forms
ischemia. Nodules (Fig. 26.4) are the lesion most likely to yield
of small-vessel vasculitis that involve the skin (Stone and
a diagnosis of vasculitis. These should be biopsied in their cen-
Nousari 2001).
ter, deeply enough to capture subcutaneous fat in the sample.
Pearl: Among the cutaneous lesions of medium-vessel vasculi- Unfortunately, nodules in cutaneous vasculitis are usually
tis, nodular lesions are most likely to yield a diagnosis of short-lived; they quickly break down to form ulcers (Fig. 26.5).
vasculitis. The best place to biopsy an ulcer is not in the center, but rather
26 Polyarteritis Nodosa 279

Table 26.2 Differential diagnosis of ulcers on the legs and feet

Atherosclerosis
• Peripheral vascular disease
Ulcers due to arterial insufficiency typically occur on the toes, the heels, the anterior shin, and extend over the malleoli
Diabetes-related issues
• Neuropathic ulcers
• Necrobiosis lipoidica diabeticorum
Longstanding, poorly controlled diabetes mellitus is the rule
Vasculitis
• Polyarteritis nodosa
• ANCA-associated disorders
• Cryoglobulinemia (history of recurrent palpable purpura is usually strong. Lower extremities are characteristically hyperpigmented)
• Rheumatoid vasculitis (occurs in patients with longstanding, severe rheumatoid arthritis)
• Buerger’s disease (predilection for digital ischemic lesions – necrotic fingers and toes – rather than leg ulcers)
• Erythema nodosum (rarely ulcerates – only in Behcet’s disease)
Vasculitis mimickers
• Atrophie blanche (livedoid vasculopathy)
• Cholesterol embolization (history of vascular instrumentation within days to weeks is usually present)
• Pyoderma gangrenosum
• Sweet’s syndrome
• Calciphylaxis (lesions tend to localize in fatty tissues of the buttocks and thighs rather than distal ones)
Connective tissue disorders
• Systemic sclerosis (digital ischemia and physical findings of sclerodactyly, telangiectasias, calcinoses, and dilated nailbed capillaries)
• Systemic lupus erythematosus
• Antiphospholipid syndrome
Stasis disease
• Venous disease with stasis
(venous ulcers typically occur above the lateral or medial malleoli)
• Lymphedema
Clotting problems
• Disseminated intravascular coagulation
• Heparin-induced thrombocytopenia (HIT)
HIT is associated with both venous and arterial thrombi
• Warfarin-induced skin necrosis
Hematologic disorders
• Polycythemia vera
• Essential thrombocythemia
• Therapy with hydroxyurea
Infections
• Osteomyelitis
• Necrotizing fasciitis
• Mycobacterium marinum
• Tuberculosis (erythema induratum)
• Fungal
• Cryptococcus
• Cocciodes
• Sporothrix
(London and Donnelly 2000; Arepally and Ortel 2006; Bazari et al. 2007).

at the edge. Again, a biopsy of sufficient depth must be per- examination of multiple serial sections (Frances et al. 2005;
formed. The differential diagnosis of leg ulcers is shown in Uthman and Khamashta 2006). When biopsies of livedo retic-
Table 26.2. The clinical and laboratory approach to the evalu- ularis are attempted, the surgeon should take more than one
ation of these lesions is shown in Table 26.3. deep punch biopsy (of a depth of 4 mm) from different areas
Compared with nodules or ulcers, livedo reticularis is asso- of both the white and red areas. The sensitivity of three biop-
ciated with a significantly lower likelihood of yielding a diag- sies taken from white areas has been estimated to be as high as
nosis. In 19 of 26 patients with livedo reticularis from whom 80% (Stockhammer et al. 1993).
two biopsy specimens were available, no obvious pathologic Lesions of digital ischemia are rarely biopsied because of
change was detected in either biopsy, despite careful their low yield and concerns about wound healing.
280 J. H. Stone

Table 26.3 Diagnostic evaluation of leg ulcers (Ha and Nousari 2003; Mimouni et al. 2003)
History and physical examination
Skin biopsy
• Three deep skin biopsies performed using a double trephine punch biopsy technique for sampling subcutaneous tissue. In short, after per-
forming a 6 mm punch biopsy and before closing the surgical site, a 4 mm punch in introduced through the surgical defect in order to reach
deeper subcutaneous tissue
• Both cylinders, 6 and 4 mm, are sent for
• From the center of the ulcer (one sample): The specimen is split, with one half going for routine histology and special stains and the other
for DIF
• From the edge of the ulcer (two samples): One specimen is evaluated for routine histology and special stains and another is evaluated for
bacterial (including mycobacterial) and fungal stains and cultures
Laboratory studies
• Complete blood count with differential
• Serum chemistries, including hepatic transaminases
• Acute phase reactants (erythrocyte sedimentation rate, C-reactive protein)
• Autoantibodies: ANCA, RF, cryoglobulins, C3/C4, ANA, dsDNA antibodies, antibodies to the Ro, La, Sm, and RNP antigens
• Clotting risk factor analysis, including anticardiolipin antibodies (IgG and IgM), Russell viper venom time, PT/PTT, antithrombin III, pro-
tein C & S, factor V Leiden, prothrombin gene mutation, homocysteine levels
• Paraproteinemia screen: SPEP, serum protein immunofixation, total serum immune globulins, UPEP, and urine protein immunofixation
• Assays for antibodies to hepatitis B, C, and the human immunodeficiency virus
Ancillary investigations
• Nerve conduction studies, compression duplex ultrasonography

Pearl: “Nodular vasculitis” is a confusing term that means Fig. 26.6 “Livedo reticularis”
medium-vessel arteritis. from a patient with polyarteritis
nodosa. The correct name of this
cutaneous lesion is really livedo
Comment: The subject of cutaneous vasculitis is difficult
racemosa (Uthman and
enough by itself without the use of nonspecific nomencla- Khamashta 2006). This patient
ture. “Nodular vasculitis” usually refers to biopsy-proven had vasculitic neuropathy with
vasculitis of medium-sized muscular arteries and arterioles motor mononeuritis multiplex.
She was cured with the
that causes a nodular skin lesion. Many forms of systemic
combination of high-dose
vasculitis – e.g., PAN, the ANCA-associated disorders, and glucocorticoids (tapered off over
mixed cryoglobulinemia – can be associated with nodular 6 months) and 6 months of daily
cutaneous findings. Thus, “nodular vasculitis” is a syndrome, cyclophosphamide, followed by
6 months of azathioprine
not a specific diagnosis. This term should be avoided by cli-
nicians and pathologists alike.

Myth: Classic PAN involvement of the skin can be distin-

guished from important mimickers by its clinical appearance.

Reality: Diagnosing the cause of skin ulcers without histopa-

thology is fraught with hazard. Cholesterol emboli, calciphylaxis
lesions, the antiphospholipid syndrome, livedoid vasculopathy,
pyoderma gangrenosum, and a host of other disorders can
mimic PAN closely (see Table 26.2). A well-performed skin
biopsy reviewed in person with an experienced dermatopathol-
ogist will shed light on many a confusing case – and often yield
surprising results.
The common objection to biopsying a skin ulcer is con-
cern about wound healing, but diagnostic confirmation and
the ability to employ the correct therapy usually justify that
risk, even if poor wound healing or extension of the lesion
through pathergy are possible concerns.
Myth: The proper name for the lacy, purplish discoloration
illustrated in Fig. 26.6 is livedo reticularis. Fig. 26.7 Livedo racemosa
26 Polyarteritis Nodosa 281

Reality: Purists note correctly the consistent sloppiness in the Myth: All patients with PAN should be treated with a cyto-
literature with regard to the terms “livedo reticularis” and “livedo toxic agent.
racemosa.” The literature is remarkably muddy here, and,
Reality: Not true. Approximately half of all patients with
frankly, at this point, it is probably not worth trying to change.
PAN can be treated successfully with prolonged courses of
Livedo racemosa, named by Ehrmann (1907), is a more
glucocorticoids alone. However, certain disease manifesta-
striking cutaneous finding than livedo reticularis (Fig. 26.7).
tions strongly suggest the use of concomitant cyclophosph-
Ehrmann described livedo racemosa as a “tendril-like, bluish
amide. These manifestations include mesenteric vasculitis,
pattern reminiscent of forked lightening.”
cardiac disease, and nervous system involvement (either
Livedo racemosa differs from livedo reticularis in the fol-
central nervous system disease or vasculitic neuropathy of
lowing ways:
the peripheral nerves). The French Vasculitis Study Group
• Livedo racemosa is more generalized. Trunk or buttocks has developed a “Five Factor Score” to specify which
involvement is present in nearly all patients. Even the face patients with PAN (and other forms of systemic vasculitis)
can be involved in some patients. are likely to benefit from the addition of cyclophosphamide
• Livedo racemosa is bad news. It is nearly always associ- from the outset of treatment (Guillevin et al. 2003).
ated with an underlying disease. The short- and long-term toxicities of high-dose glucocorti-
• In contrast, livedo reticularis often occurs in physiologic coid treatment are often underemphasized. Sometimes, physi-
settings, e.g., a response to cold, rather than in the setting cians continue excessive doses of glucocorticoids for too long,
of a disease. much to the detriment of the patient. Using a cytotoxic agent
concomitantly may decrease the quantity of glucocorticoid
Disorders associated with livedo racemosa are not invariably
therapy required. In patients who are likely to tolerate gluco-
vasculitic in nature. For example, in the antiphospholipid
corticoids poorly because of weight gain, glucose intolerance,
syndrome, livedo racemosa is frequently associated with
osteopenia, or other reasons, the simultaneous use of a cytoxic
cerebrovascular events, arterial thrombosis, and pregnancy
agent from the outset may be the wisest course.
morbidity (Frances et al. 1999). The constellation of cerebro-
If one waits too long to initiate a cytotoxic agent, then all
vascular events, antiphospholipid antibodies, and livedo race-
too often the patient is “beaten up” already from heavy-
mosa has been termed Sneddon’s syndrome (Frances and
handed glucocorticoid treatment. This only heightens the
Piette 2000). Other causes of livedo racemosa include essential
likelihood of complications from cyclophosphamide and
thrombocythemia, polycythemia vera, pernicious anemia, dis-
concerns about using the medication in the first place.
seminated intravascular coagulation, and cryofibrinogemia.
Inability to taper prednisone below 15 mg/day is a clear
Pearl: If leukopenia and thrombocytopenia are important
components of the clinical presentation, a primary form of
vasculitis is unlikely.
Comment: Leukopenia and thrombocytopenia are not usually
associated with primary vasculitic syndromes. Their presence
suggests other disorders, such as a connective tissue disorder
in the spectrum of SLE, Felty’s syndrome, or a malignancy.
Pearl: Classic PAN spares the lungs.
Comment: This observation goes back to the report of Kussmaul
and Maier, who noted that in contrast to the “peculiar, mostly
nodular thickening (periarteritis nodosa) of countless arteries
…in the bowel, stomach, kidneys, spleen, heart, and voluntary
muscles…The lung artery remains unaffected, while the bron-
chial arteries, even if only in the mediastinum, were affected to
a minor degree” (Kussmaul and Maier 1866).
Even today, reports of pulmonary parenchymal involve-
ment in classic PAN are virtually nonexistent. The diagnosis
of microscopic polyangiitis is more likely if nongranuloma-
tous, small- and/or medium-sized blood vessel involvement
in the lung is detected. In the rare cases when pulmonary
involvement is evident in PAN, it is the bronchial arteries Fig. 26.8 Microaneurysms within blood vessels of the gastrointestinal
that are involved (Matsumoto et al. 1993). tract in a patient with polyarteritis nodosa
282 J. H. Stone

signal that it is time to get another agent on board. The trick Comment: Figure 26.9a illustrates this point well. Although
is not to use cyclophosphamide too long… there is a rounding of the blood vessel and an intensification
of color density that could represent a microaneurysm, this
Myth: Visceral abdominal angiography is both specific and
finding actually represents only a natural bending of the
sensitive for PAN.
blood vessel back into the sagittal plane of the angiogram.
Reality: In the proper clinical setting, an angiogram that reveals This twist leads to an accentuation of the dye image, but is
multiple saccular aneurysms in more than one organ is quite not a true outpouching of the blood vessel wall. The image in
specific for PAN (Fig. 26.8). For the purposes of roundsman- Fig. 26.9b is far more convincing.
ship, it is worth noting that microaneurysms can also be seen
Myth: If no microaneuryms are demonstrated in the renal or
in conditions such as atrial myxoma, neurofibromatosis, the
mesenteric vasculature by magnetic resonance angiography
Ehlers–Danlos syndrome, endocarditis, and others. Thus, these
or computed tomographic angiography, then further study of
findings are strongly suggestive but not diagnostic of PAN.
those vascular beds is not indicated.
More common than multiple saccular aneurysms, unfortu-
nately, are the vaguer luminal irregularities, vascular cut-offs, Reality: Despite the dramatic advances in imaging that have
and fusiform changes — features that are not specific for PAN occurred, the resolution of magnetic resonance angiography
or any other type of vasculitis. These changes can be observed and computed tomographic angiography still do not match
in a variety of vascular disorders, including atherosclerosis. that of conventional angiography. When the highest sensitivity
Most data on the test characteristics of angiography result from for microaneurysms is required, conventional angiography
relatively few case series, many of which are outdated and suf- remains essential. Clinicians should not draw false reassur-
fer from selection bias. The sensitivity of mesenteric angiogra- ances from less sensitive but noninvasive studies.
phy for the findings of PAN is not known with any reliability.
Pearl: A highly elevated serum renin level provides a hint that the
The upshot is that abdominal angiography should be per-
answer may lie in further evaluation of the renal vasculature.
formed in patients suspected of having classic PAN but in
whom no symptomatic peripheral target organ is available Comment: Renin is produced by the juxtaglomerular apparatus
for biopsy. Careful interpretation of these studies by both of the kidney in response to perceived hypotension (Coffman
experienced clinicians and angiographers is essential. and Crowley 2008). In the setting of PAN that involves the kid-
ney, lesions of the intrarenal arteries and muscular arterioles lead
Pearl: Don’t be fooled by the turning of vessels “en fas” into
to renal hyperperfusion and elevated levels of renin. The hyper-
thinking that a microaneurysm is present.
tension characteristic of PAN is a renin-mediated process.
Although elevated serum renin levels are not diagnostic of
PAN, this finding in the proper clinical context urges a more
a in-depth evaluation to exclude a medium-vessel vasculitis.
Pearl: PAN favors medium-sized muscular arteries, but spares
veins.
Comment: This Pearl dates back to the original observations
of Kussmaul and Maier: “The most widespread and conspicu-
ous change is that of the arteries. It is present….only in arter-
ies whose media preferably have muscular elements and which
do not exceed the caliber of the liver artery …and nowhere in
the capillaries. A similar change like that of the arteries is not
detectable in the veins…” (Kussmaul and Maier 1866).
Many other forms of vasculitis do involve the venous circu-
b lation, however. Most forms of cutaneous vasculitis target the
postcapillary venules. The lesions of the vasculitides that are
commonly associated with ANCA often involve small- and
medium-sized veins. Behçet’s disease may affect veins of any
size (including ones as large as the pulmonary veins and the
vena cavae), and is associated with an increased risk of deep
venous thrombosis. The posterior uveitis (retinal vasculitis) in
Behcet’s disease also consists primarily of a venulitis.
The reasons for this striking selectivity for different por-
Fig. 26.9 Angiogram findings in a patient without (a) and with (b) tions of the vascular bed remain unclear remain one of the
polyarteritis nodosa great mysteries of vasculitis.
26 Polyarteritis Nodosa 283

Myth: Lower extremity ulcers that heal with the scars depicted
in Fig. 26.10 are pathognomonic of atrophie blanche.
Reality: Atrophie blanche is the hallmark cutaneous lesion
of livedoid vasculopathy, a bland vascular disorder of small
blood vessels characterized by thrombosis (Kroshinsky et al.
2008). Livedoid vasculopathy is a thrombotic microangiopa-
thy characterized clinically by lower extremity ulcerations
and the stellate-shaped, porcelain-colored scars known as
atrophie blanche (Fig. 26.11) (Maessen-Visch et al. 1999;
Callen, 2006). Healed lesions of cutaneous PAN can mimic
this clinical finding perfectly.
Livedoid vasculopathy is far less likely than is PAN to
Fig. 26.10 Healed skin ulcers in a patient with polyarteritis nodosa cause mononeuritis multiplex. However, mononeuritis multi-
plex has been reported in livedoid vasculopathy (Kroshinsky
et al. 2008; Hairston et al. 2006). Thus, this feature does not
differentiate reliably between these two conditions, and histo-
pathological distinction must be sought. A good skin biopsy
is usually sufficient to settle the question (Fig. 26.12).

References

Arepally GM, Ortel TL. Heparin-induced thrombocytopenia. N Engl

J Med. 2006;355:809–17
Bazari H, Jaff MR, Mannstadt M, Yan S. Case 7-2007: a 59 year-old
woman with diabetic renal disease and nonhealing skin ulcers.
N Engl J Med. 2007; 356:1049–57
Callen JP. Livedoid vasculopathy: what it is and how the patient should
be evaluated and treated. Arch Dermatol. 2006;142:1481–2
Coffman TM, Crowley SD. Kidney in hypertension: Guyton redux.
Hypertension 2008;51:811–16
Collins MP, Mendell JR, Periquet MI, et al Superficial peroneal nerve/
peroneus brevis muscle biopsy in vasculitic neuropathy. Neurology
Fig. 26.11 Cutaneous lesions of livedoid vasculopathy, known as 2000;55:636–43
“atrophie blanche” Ehrmann S. Ein Gefaessprozess Bei Lues. Wien Med Wochenschr.
1907;57:777
Frances C, Niang S, Laffitte E, et al Dermatologic manifestations of the
antiphospholipid syndrome. Arthritis Rheum. 2005;52: 1785–93
Frances C, Piette JC. The mystery of Sneddon’s syndrome: the relation-
ship with antiphospholipid syndrome and systemic lupus erythema-
tosus. J Autoimmun. 2000;15:139–43
Frances C, Papo T, Wechser B, et al Sneddon’s syndrome with and
without aPL. Medicine 1999;78:209–19
Gorson KC, Hedley-White ET, Skinner MM. Case records of
the Massachusetts General Hospital. N Eng J Med. 2001;344: 917–23
Guillevin L, Lhote F. Treatment of polyarteritis nodosa and microscopic
polyangiitis. Arthritis Rheum. 1998;41:2100–5
Guillevin L, Cohen P, Mahr A, et al Treatment of polyarteritis nodosa
and microscopic polyangiitis with poor prognosis factors: a pro-
spective trial comparing glucocorticoids and six or twelve cyclo-
phosphamide pulses in sixty-five patients. Arthritis Rheum. 2003;
49:93–100
Ha C, Nousari, HC. Double trephine punch biopsy technique for sam-
pling subcutaneous tissue. J Am Acad Dermatol. 2003;48:609–10
Hairston BR, Davis MD, Pittelkow MR, Ahmed I. Livedoid vasculopa-
Fig. 26.12 Segmental hyalinizing vasculopathy. Livedoid vasculopa- thy: further evidence for procoagulant pathogenesis. Arch Dermatol.
thy is not a vasculitis per se but rather a vasculopathy associated with 2006;142:1413–8
small-vessel thrombi and hyalinizing changes within the blood vessel Kroshinsky D, Stone JH, Bloch D, Sepehr A. A 47 year-old woman
walls. Note hyaline changes in the walls of these vessels that might be with numbness and pain in the lower extremities. N Engl J Med.
mistaken (by an amateur pathologist!) for fibrinoid necrosis. 2009; 360:711–20
284 J. H. Stone

Kussmaul A, Maier R. Über eine bisher nicht beschriebene eigenthum- Stockhammer GJ, Felber SR, Zelger B, et al Sneddon’s syndrome: diag-
liche arterienerkrankung (periarteritis nodosa), die mit morbus nosis by skin biopsy and MRI in 17 patients. Stroke 1993;24 :685–90
brightii und rapid fortschreitender allgemeiner muskellahmung ein- Stone JH, Talor M, Stebbing J, et al Test characteristics of immunofluo-
hergeht. Dtsch Arch Klin Med. 1866;1:484–518 rescence and ELISA tests in 856 consecutive patients with possible
London NJ, Donnelly R. ABC of arterial and venous disease: ulcerated ANCA-associated conditions. Arthritis Care Res. 2000;13: 424–34
lower limb. BMJ 2000;320:1589–91 Stone JH, Nousari HC. “Essential” cutaneous vasculitis: what every
Maessen-Visch M, Koedam M, Hamulyak K, et al Atrophie blanche. Int rheumatologist should know about vasculitis of the skin. Curr Opin
J Dermatol. 1999;38:161–72 Rheumatol. 2001;13:23–34
Mathew L, Talbot K, Love S, et al Treatment of vasculitic peripheral Talor MV, Stone JH, Stebbing JJ, et al Antibodies to selected minor
neuropathy: a retrospective analysis of outcome. QJM 2007; target antigens in patients with anti-neutrophil cytoplasmic antibod-
100:41–57 ies (ANCA) by immunofluorescence antibodies to minor target anti-
Matsumoto T, Homma S, Okada M, et al The lung in polyarteritis nodosa: gens in patients with anti-neutrophil cytoplasmic antibodies. Clin
a pathologic study of 10 cases. Hum Pathol. 1993;24:717–24 Exp Immunol. 2007;150:42–8
Mimouni D, Ng PP, Rencic A, et al Cutaneous PAN in patients present- Uthman IW, Khamashta MA. Livedo racemosa: a striking dermatological
ing with atrophie blanche. Br J Dermatol. 2003;148:789–94 sign for the antiphospholipid syndrome. J Rheumatol. 2006;33: 2379–82
Pagnoux C, Seror R, Mahr A, et al Polyarteritis nodosa: clinical descrip- Vallat J-M, Cros DP, Hedley-White ET. Case 9-2007: a 27 year-old
tion and long-term follow-up of the 351 patients from the French woman with pain and swelling of the legs. N Engl J Med. 2007;356:
Vasculitis Study Group. Arthritis Rheum. 2006;49:S420 1252–9
 Giant Cell Arteritis and Polymyalgia Rheumatica
27
James R. Stone, Misha Pless, Carlo Salvarani, Nicolo Pipitone,
Simmons Lessell, and John H. Stone

Comment: Biological and chronological age often correlate

Overview of Giant Cell Arteritis poorly with each other, but the statement that GCA does not
occur in patients younger than 50 is remarkably robust. Of the
› Giant cell arteritis (GCA) is a form a large- to medium- 1,435 GCA patients with biopsy-proven disease included in a
vessel vasculitis that involves the aorta and its branches. metaanalysis of 26 studies, only two patients were younger
› Polymyalgia rheumatica (PMR), a syndrome of mus- than 50 years of age (Smetana and Shmerling 2002).
cle pain and stiffness in the neck, shoulders, and hips, One suspects that even those two patients might have had
often occurs with GCA but can occur independently. some form of systemic vasculitis other than GCA, as several
› GCA predominantly affects the second- to fifth-order other vasculitic conditions can involve the temporal arteries
aortic branches, often in the extracranial arteries of the and mimic GCA. The ANCA-associated vasculitides are
head. notorious for this (Small and Brisson 1991).
› The mean age at diagnosis of GCA is approximately
Pearl: GCA is rare among African-Americans, particularly
72.
African-American men.
› The diagnosis of GCA is made usually by biopsy of
the temporal artery. Comment: This point is difficult to prove in the absence of a
› GCA associated with granulomatous inflammation large, population-based study conducted in an area comprised
within the blood vessel wall, but granulomatous of a substantial number of African-Americans. Olmsted
changes are not always evident early in the disease. County, Minnesota, and the Scandinavian countries, sites of
Giant cells occur in a slight majority of cases of posi- much GCA research to date, are not such places.
tive temporal artery biopsies. Estimates of the fre- A study from Baltimore, a city that is 65% African-
quency of giant cells in temporal artery biopsy tissues American, evaluated 90 cases of biopsy-proven temporal
are on the order of 50-80%. arteritis and compared them with 90 controls (Regan et al.
› In GCA, clinical symptoms of local vascular inflam- 2000). The cases included every single temporal artery
mation and vascular insufficiency are usually accom- biopsy-proven case of GCA at the Johns Hopkins Hospital
panied or preceded by a systemic inflammatory process. between 1968 and 2000. The cases were matched to the con-
› Visual loss is the most feared complication of GCA. trols on age, gender, and year of biopsy, but not race.
Visual loss may occur through the syndrome of ante- Among the controls, 26 of the 90 patients (28%) were
rior ischemic optic neuropathy, caused by narrowing African-Americans. However, among the cases, only five
of the posterior ciliary artery, as well as through other patients (6%) were African-Americans (p < 0.001). All five
arterial insufficiency syndromes. GCA patients who were African-American were female.
› Glucocorticoids are the cornerstone of treatment for Thus, at the Johns Hopkins Hospital, there was not a singe
GCA and PMR. PMR occurring in the absence of case of biopsy-proven GCA in an African-American man
GCA requires a lower dose of prednisone for disease over a period spanning 32 years.
control. The upshot of these data is that, when considering the cause
of an inflammatory illness in an African-American patient,
diagnoses other than GCA should be weighed more heavily.
27.1 Symptoms and Signs Myth: GCA is characterized by temporal arteries that are
abnormal on physical examination.
Pearl: GCA rarely (if ever) occurs among patients less than Reality: Many patients who turn out to have biopsy-proven
50 years of age. GCA do not have temporal arteries that are obviously

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 285

DOI:10.1007/978-1-84800-934-9_27, © Springer Science + Business Media B.V. 2009
286 J. R. Stone et al.

Table 27.1. Estimated likelihood ratios for the prediction of a positive temporal artery biopsy for findings on the history and physical
examination
Symptom/sign Number of patients with data on variable Positive LR (95% CI) Negative LR (95% CI)
Symptoms
Anorexia 674 1.2 (0.96–1.4) 0.87 (0.75–1.0)
Weight Loss 1,417 1.3 (1.1–1.5) 0.89 (0.79–1.0)
Arthralgia 582 1.1 (0.86–1.4) 1.0 (0.92–1.1)
Diplopia 703 3.4 (1.3–8.6) 0.95 (0.91–0.99)
Fatigue 1,095 1.2 (0.98–1.4) 0.94 (0.86–1.0)
Fever 1,708 1.2 (0.98–1.4) 0.92 (0.85–0.99)
Temporal headache 386 1.5 (0.78–3.0) 0.82 (0.64–1.0)
Any headache 2,475 1.2 (1.1–1.4) 0.7 (0.57–0.85)
Jaw claudication 2,314 4.2 (2.8–6.2) 0.72 (0.65–0.81)
Myalgia 681 0.93 (0.81–1.1) 1.1 (0.87–1.3)
Polymyalgia rheumatica 1,383 0.97 (0.76–1.2) 0.99 (0.83–1.2)
Unilateral visual loss 341 0.85 (0.58–1.2) 1.2 (1.0–1.3)
Any visual symptoms 2,083 1.1 (0.93–1.3) 0.97 (0.9–1.0)
Vertigo 212 0.71 (0.38–1.3) 1.1 (0.93–1.2)
Signs
Optic atrophy or ischemic 142 1.6 (1.0–2.5) 0.8 (0.58–1.1)
optic neuropathy
Scalp tenderness 923 1.6 (1.2–2.1) 0.93 (0.86–1.0)
Synovitis 734 0.41 (0.23–0.72) 1.1 (1.0–1.2)
Beaded temporal artery 323 4.6 (1.1–18.4) 0.93 (0.88–0.99)
Prominent/enlarged temporal 508 4.3 (2.1–8.9) 0.67 (0.5–0.89)
artery
Tender temporal artery 755 2.6 (1.9–3.7) 0.82 (0.74–0.92)
Absent temporal artery pulse 68 2.7 (0.55–13.4) 0.71 (0.38–0.75)
Adapted from Smetana and Shmerling (2002)

abnormal on physical examination. In approximately one brainstem, can also be a feature of GCA. Diplopia occurs in
third of cases, in fact, the physical examination of the tempo- up to 15% of patients with this disease (Gordon and Levin
ral arteries is completely normal (Hall et al. 1983). 1998). Diplopia was reported in 5 of 46 patients (11%) with
The classic physical examination finding of nodular, biopsy-proven GCA from Olmsted County, Minnesota (Hall
thickened temporal arteries is extremely helpful when pres- et al. 1983). Along with jaw claudication, diplopia is one of
ent but, alas, is quite rare. (This is probably a good thing, as the most specific symptoms of GCA.
nodular temporal arteries suggest longstanding, undiagnosed A variety of mechanisms can lead to diplopia in GCA.
disease.) Nodular thickening of the temporal artery on physi- • Branches of the ophthalmic artery supply ocular muscles
cal examination is associated with an odds ratio of 4.5 for within the orbit. Ischemia of the extraocular muscles is
biopsy-proven GCA (Smetana and Shmerling 2002). This believed to be the etiology of diplopia in some cases.
makes it one of the most useful physical findings for the dif- • More proximal branches of the ophthalmic artery consti-
ferentiation of patients who will have a positive biopsy from tute the vasa nervorum of the cranial nerves, which direct
those who will not (Table 27.1). ocular motor function. Involvement of the vasa nervorum
Pearl: Patients with PMR have normal passive range of by GCA can lead to discrete patterns of ophthalmoplegia,
motion of the shoulder. e.g., trochlear, oculomotor, or abducens nerve palsies.
• Finally, brainstem strokes caused by infarctions within
Comment: Patients with PMR can be moved to tears because the territories of the vertebral or posterior cerebral arteries
of pain when attempting to raise their arms above their heads are possible in GCA. Such lesions can affect the struc-
or abduct their shoulders through active movement. However, tures within the brainstem that serve binocular motor
these same patients have full range of motion of the affected function, thereby leading to diplopia.
extremities if the examiner takes them through these motions
passively. Pearl: A persistent dry cough can be a manifestation of occult
GCA.
Myth: Diplopia is a rare manifestation of GCA.
Comment: Cough is one of the many “occult” manifestations
Reality: Many clinicians fail to recognize that diplopia, of GCA – disease features that are not considered classic for
caused by ischemia of extraocular muscles, cranial nerves, or the diagnosis but which, when present, unveil the diagnosis
27 Giant Cell Arteritis and Polymyalgia Rheumatica 287

to the astute clinician. The upper respiratory tract contains Table 27.2 Comparison of features of arteritic and nonarteritic anterior
numerous “cough receptors.” The etiology of cough in GCA ischemic optic neuropathy (AION)
is presumed to be irritation of the cough receptors by isch- Feature Arteritic Nonarteritic
emia to surrounding tissues (Olopade et al. 1997). Age (mean years) 70 60
Gender Female: male = 3 Female: male = 1
Pearl: The most specific symptom of GCA is jaw claudication. distribution
Systemic Often Uncommon
Comment: A 1983 study from the Mayo Clinic demonstrated complaints
this point nicely (Hall et al. 1983). Between 1965 and 1980, Visual acuity Impaired (<20/20) in up Preserved in the
134 Olmsted County residents underwent temporal artery to three-fourths majority
biopsies, of which 46 (34%) were positive for GCA and 88 Disc Pale > hyperemic edema Hyperemic > pale
edema
(66%) were negative. The symptoms of PMR, fever, headache
Cup size Normal Small
of recent onset, scalp tenderness, and visual loss were all Mean ESR 70 20–40
similar between the positive and negative biopsy groups (after (mm/h)
all, the patients had to have symptoms compatible with GCA Natural history Improvement rare Improvement in up
to prompt a biopsy). Jaw claudication, however, was present to 43%
in 54% of the patients with positive biopsies, compared with Contralateral Fellow eye in 95% Fellow eye in
involvement <30%
only 3% of those whose biopsies were negative (p < 0.001).
Treatment Glucocorticoids None proved
When the patient tells you that she has jaw claudication, the Adapted from Seo and Stone (2002)
diagnosis can be relatively straightforward. Of course, it never
happens that way. Patients never complain of that buzzword –
“jaw claudication” – but rather might note facial pressure while among patients older than 65 years of age (Esposito and
talking, the reluctance to eat meat because of difficulty chew- Gleckman 1978). In one study of fever in GCA, 15 of 100
ing, the inability to open their mouths (reminiscent of trismus), consecutive cases of biopsy-proven GCA had fevers before
difficulty speaking (Lee et al. 1999), pain in the tongue (Larson diagnosis. The patients with fever all had normal white blood
et al. 1984), and other similarly oblique confessions. cell counts, but as a group they had significantly higher ESRs,
platelet counts, and alkaline phosphatase levels than the other
Pearl: Jaw claudication comes on fast. 85 patients, as well as lower hemoglobin concentrations. In
four patients with no other symptoms or physical findings sug-
Comment: It usually does not require a tough piece of steak to
gestive of GCA, temporal artery biopsies revealed arteritis.
elicit jaw claudication. Patients with this symptom develop it
quickly following the initiation of chewing, even of relatively Pearl: The most common cause of acute visual loss in GCA
soft foods. Jaw claudicators are seized characteristically with is anterior ischemic optic neuropathy.
an angina-like pain in the cheek or jaw that gets their attention
Comment: Five known mechanisms of visual loss occur in
abruptly, usually after only a few seconds of chewing.
GCA (Table 27.2).
Pearl: New, unexplained pain located above the neck in a
• Arteritic anterior ischemic optic neuropathy (A-AION) is
person older than 60 should suggest the possibility of GCA.
the most common mechanism of visual loss. A-AION
Comment: Headache and jaw claudication are classic mani- involves vasculitic occlusion of the posterior ciliary arteries,
festations of GCA. However, patients with GCA can present which perfuse the vulnerable neuro-ocular junction at the
with many other less well-described pains above the neck, optic nerve head (Fig. 27.1). The neuro-ocular junction con-
including pain over the nose, in the ears, along the jaw (not tains 1.1 million axons that exit the globe posteriorly through
necessarily related to chewing), in the throat, in the mouth the optic nerve. Approximately 80% of all cases of loss of
(e.g., tongue or “dental” like pain), or over the occiput. Indeed, vision from GCA occur through the mechanism of A-AION.
many patients with these symptoms are first misdiagnosed as The typical funduscopic finding 24 h after A-AION is swell-
having “sinusitis,” “otitis media,” or “dental infections,” even ing of the optic disc (Fig. 27.2).
though physical signs of these disorders are absent. • A subset of AION – posterior ischemic optic neuropathy
Other facial manifestations described in GCA include (PION) – can also occur in GCA. In PION, the ischemic
odontogenic pain, chin numbness, glossitis, submandibular insult occurs in the retrobulbar portion of the optic nerve,
swelling, and necrosis of the scalp, tongue, and lips (Paraskevas through occlusion of the proximal posterior ciliary arteries.
et al. 2007; Ruiz-Masera et al. 1995; Porter, 1990). As a result, no disc edema is visible to the clinician upon
funduscopy.
Pearl: GCA is a common cause of fever of unknown origin
• Central or branch retinal artery occlusion (CRAO or
(FUO) in the elderly.
BRAO) is the cause of visual loss in approximately 20%
Comment: Although GCA causes less than 2% of all cases of of GCA cases. The central and branch retinal arteries are
FUO, GCA is the culprit in 16% of FUO cases occurring both somewhat larger than the posterior ciliary artery.
288 J. R. Stone et al.

Fig. 27.1 Arteritic anterior 1 2

ischemic optic neuropathy
(A-AION) is the most common
DN
mechanism of visual loss. ST DN
A-AION involves vasculitic ST
occlusion of the posterior ciliary MP
MP
arteries (LPCA and MPCA),
which perfuse the vulnerable
neuro-ocular junction at the optic
MPCA
nerve head. The neuro-ocular MPCA
junction contains 1.1 million AE
LacA AE
axons that exit the globe LacA
posteriorly through the optic
nerve. Approximately 80% of all LPS LR MR MR
cases of loss of vision from GCA SO LPS LR SO
SR SR
occur through the mechanism of MSO MSO
A-AION. ICA internal carotid SOF PE MM
artery; OA ophthalmic artery; SOF
CAR MM PE
CAR central retinal artery; LPCA LPCA CAR
lateral posterior ciliary artery; LPCA
MPCA medial posterior ciliary OA
artery

Width 25 pi
ICA
Height 24 pi

a Myth: In patients with GCA who present with acute visual

loss, funduscopy inevitably reveals abnormal findings.

Reality: Funduscopy yields findings consistent with GCA in

most cases. However, the findings on funduscopy are never
pathognomonic of GCA, and this examination can be normal
even in patients with GCA who have lost vision. This is dic-
tated by the precise site of the vascular lesion that has led to
visual loss.
PION is the cause of visual loss in only a minority of
GCA cases (Hayreh et al. 1998a, b). In contrast to AION,
PION is associated with a normal optic disk in the acute
Fig. 27.2 Arteritic b
anterior ischemic
phase of visual loss. Optic disk pallor ensues 4–6 weeks later
optic neuropathy (Rucker et al. 2004).
(A-AION). (a) The In extremely rare cases, blindness in GCA results from
typical finding 24 h cortical ischemia. This is also associated with a normal optic
after the onset of
vision loss within the
disk.
eye is swelling of the
optic nerve. (b) A Myth: It is possible to diagnose GCA by funduscopy.
normal optic nerve is
shown for Reality: It is impossible to distinguish arteritic from nonar-
comparison teritic AION based on direct visualization of the optic nerve
and/or ocular fundus. The optic nerve in AION of any type
appears swollen and is often associated with papillary hem-
• A rare mechanism of visual loss in GCA is ophthalmic
orrhages. In PION, the optic nerve appears normal in the con-
artery occlusion. In such cases, the entire eye is ischemic,
text of acute loss of vision until atrophy sets in over a period
the orbit edematous, and there is ophthalmoplegia.
of weeks. The optic nerve may look partially swollen within
• Finally, a posterior circulation stroke involving the calcar-
the background of retinal infarction in CRAO or BRAO.
ine cortices can lead to visual field defects or cortical
blindness by virtue of bilateral hemispheric involvement. Pearl: Pallid disc edema is helpful in diagnosing GCA.
27 Giant Cell Arteritis and Polymyalgia Rheumatica 289

Comment: When the optic nerve has pallor in addition to edema Table 27.3 Vascular lesions that can lead to vision loss in giant cell
at acute onset of visual loss, this implies there has been axonal arteritis
loss before a more massive optic nerve head infarction. This is Posterior ciliary artery occlusion
Anterior ischemic optic neuropathy (AION)
presumed to occur in GCA-associated AION more often than
Posterior ischemic optic neuropathy (PION)
in a nonarteritic ischemic disorder of the optic nerve. Retinal artery occlusion
Pearl: GCA can cause visual hallucinations. Central retinal artery occlusion (CRAO)
Branch retinal artery occlusion (BRAO)
Comment: One highly intriguing ocular finding in GCA is the Ophthalmic artery occlusion
Charles Bonnet (pronounced “Bō nay’ ”) syndrome. The Posterior circulation stroke
Charles Bonnet syndrome is the occurrence of visual halluci-
nations – “release hallucinations” – in individuals who are psy- With the exception of glaucoma, NAAION is the most
chologically normal (Razavi et al. 2004). Reversibility of these common cause of optic neuropathy (Lessell 1999). NAAION
symptoms with glucocorticoid therapy has been reported. tends to affect Caucasians older than 60 years of age, as is
The precise mechanism of these hallucinations is not the case with GCA. Unlike GCA, however, NAAION is not
defined, but the Charles Bonnet syndrome is believed to be a an inflammatory disorder: constitutional symptoms and ele-
consequence of ischemia to parts of the visual pathways. vated acute phase reactants are usually absent.
This includes of the posterior circulation, i.e., the posterior NAAION is associated with funduscopic findings that help
cerebral arteries, leading to reversible ischemia in the calcar- differentiate it from GCA: hyperemic disc swelling and a
ine cortex or the parietal subcortical white matter. morphologically small optic disc with no cup in NAAION,
Pearl: GCA may present as noises in the head. compared with disc pallor, disc swelling, cotton-wool spots,
and a normal optic cup size in GCA. Table 27.3 compares
Comment: This disease feature is rarely mentioned in the litera- these two entities.
ture, and its etiology is obscure. However, patients have reported
a diversity of sounds appearing to come from within their heads. Pearl: GCA can cause facial swelling.
Examples of such sounds are turbine engines and tree frogs. Comment: Facial swelling has been described as a complica-
Myth: GCA does not affect the central nervous system (CNS). tion of GCA (separate from prednisone-induced Cushingoid
features) (Cohen et al. 1982). The pathogenesis of facial
Reality: If this Myth were reworded to read that “GCA sel- swelling in GCA is poorly understood. In one study, facial
dom involves the intracerebral circulation yet can still affect swelling was found to be associated with the risk of develop-
the brain through causing extensive extracranial disease,” ing visual disturbances (Friedman and Friedman 1986),
then this statement would be correct. When transient isch- whereas in another study facial and neck swelling were found
emic attacks, vertigo, hearing loss, and stroke occur in GCA, not to be linked to visual disturbances or to ischemic stroke
they are usually the result of vertebral artery or internal (Liozon et al. 2006).
carotid artery lesions that are extradural.
The rarity of intracerebral disease in GCA is a longstand- Pearl: Audiovestibular dysfunction may be a manifestation
ing mystery. One frequently proffered explanation is the fact of GCA.
that once the extracranial arteries penetrate the dura they lose Comment: A syndrome of sensorineural hearing loss, often
their external elastic laminae (Wilkinson and Russell 1972; accompanied by vertigo and tinnitus, has been described in
Caselli et al. 1988). The full answer for the rarity of CNS several immune-mediated disorders. These include vasculiti-
vascular disease in GCA is probably more complicated. des such as polyarteritis nodosa and Wegener’s granulomato-
Regardless of the reason, however, intracerebral GCA is sis (WG) (Stone and Francis 2000). This syndrome,
very rare but well-documented in a small number of patients. sometimes referred to as immune-mediated inner ear disease
In a study of 463 patients from the Mayo Clinic with clinical (IMIED), is frequently bilateral and is characterized by a
diagnoses of CNS vasculitis, only two had convincing evi- clinical course that is more rapid than Meniere’s syndrome, a
dence of intracranial GCA (Salvarani et al. 2006). disorder in which fluctuating hearing loss occurs over months
to years (Papadimitraki et al. 2004).
Myth: Acute unilateral vision loss in an elderly individual is
A retrospective review of 271 GCA patients disclosed four
likely to be GCA.
who presented with sensineural hearing loss (Hausch and
Reality: Only a minority of acute unilateral vision loss events Harrington 1998). Audiovestibular dysfunction was found to
in this setting are caused by GCA. Another entity with a ten- be even more frequent in a prospective study on 44 patients
dency to occur in old age – nonarteritic anterior ischemic with GCA (Amor-Dorado et al. 2003). This study investigated
optic neuropathy (NAAION) – is approximately ten times patients systematically for evidence of such dysfunction both
more common than GCA as a cause of unilateral vision loss. by clinical interview and a battery of specific tests. More than
290 J. R. Stone et al.

one half of the patients complained of hearing loss, tinnitus, early involvement of these arteries may have neither local
vertigo, or dizziness. Moreover, nearly 90% of the patients symptoms (e.g., arm claudication) nor stroke, the diagnosis
had objective evidence of vestibular dysfunction. may be suggested first by discrepancies detected between the
These symptoms were generally reversible after several blood pressures in the right and left arms, or auscultation over
days of glucocorticoid treatment. After 3 months of treat- the brachial, subclavian, and axillary arteries.
ment, the percentage with vestibular dysfunction had declined Many GCA patients with aortic arch syndromes and find-
to 30%, and after 6 months only one patient had persistently ings of occlusive disease in the arms have few of the classic
abnormal vestibular tests. “temporal arteritis” symptoms such as headache, scalp ten-
These data suggest that audiovestibular dysfunction – derness, and jaw claudication. Thirty-one of the 74 GCA
both sensorineural hearing loss and vestibular problems – is cases (42%) associated with aortic arch syndromes had nega-
a feature of GCA. The audiovestibular manifestations of tive temporal artery biopsies (Brack et al. 1999).
GCA appear to respond well to conventional GCA therapy,
Pearl: If the patient has headaches, PMR symptoms, and calf
but their pathophysiologic basis remains obscure.
pain, do not forget to check an ANCA.
Pearl: GCA frequently affects large vessels, i.e., the aorta and
Comment: Absent histopathologic confirmation of GCA by
its branches to the extremities.
biopsy or unequivocal findings on imaging consistent with
Comment: Although the diagnosis of GCA is usually con- large-vessel vasculitis, checking an ANCA is a good idea any-
firmed by biopsy of the temporal artery, this is only because way, calf pain or not. ANCA-associated disorders can present
of the convenience of sampling this vessel. In fact, the dis- with headaches and PMR symptoms that are nearly perfect
ease involves a wide range of other vessels, and sometimes mimics of GCA. Calf pain is much more typical of vasculiti-
appears to spare the temporal arteries altogether. des that involve small- to medium-sized vessels than it is of
The frequency of aortic involvement in GCA was confirmed GCA. Thus, the ANCA-associated disorders, polyarteritis
in 1995 by a population-based study conducted in Olmsted nodosa, and cryoglobulinemia must all be kept in mind.
County (Evans et al. 1995). Sixteen of the 96 patients (17%) Among these, the ANCA-associated disorders are by far the
diagnosed with GCA between 1950 and 1985 developed either most common to masquerade as GCA.
thoracic (n = 11) or abdominal (n = 5) aneurysms. Compared
Myth: All GCA patients – even those without ocular
with all persons of the same age and gender living in the same
symptoms – should be referred to an ophthalmologist.
county during that time period, patients with GCA were 17
times more likely to develop thoracic aneurysms, and 2.4 times Reality: When GCA affects the eyes, the symptoms are dra-
more likely to develop abdominal aortic aneurysms. matic and not ignorable by the patient. CRAO and AION typi-
Most aneurysms were detected several years after the cally cause acute, severe vision loss. Patients who have transient
diagnosis of GCA (median: 5.8 years). Less extensive inflam- amaurosis as a prelude to one of these ophthalmic catastrophes
matory involvement of the aorta not sufficient to cause an are usually aware that they have temporarily lost their vision.
aneurysm is likely to be even more common. Myocardial Patients with ophthalmoplegia from GCA (usually from sixth
infarction, thoracoabdominal aortic dissections, and aortic or third nerve palsy) have sudden binocular diplopia.
regurgitation are all potential complications of GCA. If a sentient GCA suspect has none of these symptoms, it
Patients who develop thoracic aortic aneurysms usually is highly unlikely that the ophthalmologist will detect any
do not have histories of involvement of the subclavian arter- abnormality that contributes to the diagnosis of GCA.
ies or other primary branches of the thoracic aorta (Evans
Myth: Ophthalmologists and rheumatologists should manage
et al. 1995). Thus, the group of patients with thoracic aortic
patients with GCA, who have ocular symptoms together.
aneurysms and the group with aortic arch syndromes may
constitute two different GCA subsets. Reality: In the absence of new eye symptoms, ophthalmic
Annual chest radiographs are recommended for patients surveillance will not detect new disease activity in patients
with GCA, to screen for the development of thoracic aortic who do not have ocular complaints. The patient who is stable
aneurysms (Bongartz and Metteson 2006). Patients with known from the standpoint of visual health can be managed by the
aneurysms of the thoracic aorta require closer follow-up with rheumatologist alone once the diagnosis has been established
other imaging modalities (e.g., echocardiography, MRI). and treatment initiated. The occurrence of new or recurrent
ocular symptoms is an indication to involve an ophthalmolo-
Pearl: Check the blood pressure in both arms in patients sus-
gist in the case again.
pected of having GCA.
Myth: The eye is an important source of headache.
Comment: The primary branches of the thoracic aorta (carotid,
vertebral, subclavian, and arteries) are involved clinically in Reality: Popular lore holds that the eyes are an important
about 15% of cases (Brack et al. 1999). Because patients with “headache generator.” This is not true. With regard to pain
27 Giant Cell Arteritis and Polymyalgia Rheumatica 291

and common ocular syndromes, the following points are the primrose path. PMR is even more likely than GCA to be
important: associated with normal acute phase reactants.
• Open-angle glaucoma, a prevalent disease in the elderly, Pearl: A normal ESR is more useful than an abnormal one
is not associated with either headaches or eye pain. when evaluating a patient for GCA.
Because open-angle glaucoma is asymptomatic, the diag-
Comment: A normal ESR is more useful in excluding GCA
nosis often goes unrecognized even during the evolution
than a high result is in diagnosing it (Smetana and Shmerling
of irreversible peripheral vision loss. Thus, if headache is
2002). In a metaanalysis that included 941 biopsy-proven
a major symptom, clinicians should not waste time refer-
GCA patients with sufficient information about the ESR, only
ring patients for ocular pressure measurements.
4% of the patients had normal values. A normal ESR dimin-
• Primary angle-closure glaucoma, much rarer than open-
ished the probability of a positive temporal artery biopsy by a
angle glaucoma, is painful. However, acute angle closure
factor of five.
is associated with red eye, severe ocular pain, and impaired
Even a sky-high ESR adds comparatively little to the likeli-
vision. In such cases, the source of pain – the eye – is
hood of GCA. For example, an ESR > 100 mm/h is a less pow-
obvious to any physician, even if the mechanism remains
erful predictor of a positive biopsy than are many elements of
obscure.
the history and physical examination, particularly jaw claudi-
• Anterior uveitis and keratitis can occur at any age and are
cation, diplopia, and an abnormal (beaded or enlarged) tempo-
painful, but present acutely with prominent photophobia
ral artery (Smetana and Shmerling 2002).
and external signs of inflammation. The pain in these
patients is localized readily to the eye. Myth: GCA headaches tend to focus over the temples.
• Finally, miscellaneous disorders such as refractive error,
Reality: There is no classic GCA headache. The chief distin-
cataracts, and eye muscle imbalance do not cause head-
guishing feature of headaches in GCA is that they mark a
aches.
significant departure from normal for the patient. Thus, for
Searches for “ocular causes” of headaches in patients with- patients who are not “headache people,” the very presence of
out eye symptoms or signs are nearly always fruitless. They any headache at all is abnormal. For patients given to head-
divert the clinician from etiologies that demand more urgent aches in general, the headache associated with GCA consti-
attention, such as GCA. tutes a change from the usual pattern: a different location,
greater or lesser severity, or a quality that is discrepant from
Myth: GCA does not cause vasculitic neuropathy.
the patient’s baseline intermittent headaches.
Reality: Mononeuritis multiplex is most often associated GCA headaches can be of a boring, unrelenting quality that
with medium- to small-vessel vasculitis (medium much involves the entire head. They can be focused over the temples
more so than small). The Churg–Strauss syndrome (CSS) or the eyes. They can also be so trifling as to have the patient
and polyarteritis nodosa, for example, cause a potentially unaware of them unless asked specifically. But their cardinal
crippling vasculitic neuropathy in up to 80% of cases. GCA feature is that they are different for that specific patient.
causes vasculitic neuropathy much less often, but well-docu-
Pearl: Leukocytosis does not occur in GCA before the insti-
mented cases are described (Caselli et al. 1988).
tution of glucocorticoid treatment.
Comment: Despite the often dramatic other indications of
27.2 Laboratory Testing systemic inflammation, the white blood cell count is usually
normal before the patient starts prednisone. This is true even
in patients with fevers of unknown origin. Nonspecific labo-
Myth: GCA never presents with a low ESR.
ratory abnormalities that can accompany GCA include a nor-
Reality: A population-based study from Olmsted County, mochromic, normocytic anemia, thrombocytosis, polyclonal
Minnesota, soundly repudiated this notion (Salvarani and hypergammaglobulinemia, and hypoalbuminemia.
Hunder 2001). In that report, eighteen of 167 patients (11%)
Myth: A rising ESR indicates the return of GCA.
with GCA had ESRs of less than 50 mm/h before receiving
glucocorticoids. Nine of the 167 (5%) had ESRs less that Reality: This Myth is probably the most common error in the
40 mm/h. All of the latter nine patients had positive temporal management of GCA. “Treating the sed rate” leads to sub-
artery biopsies. stantial morbidity from excessive glucocorticoid therapy. An
The C-reactive protein (CRP) level can also be normal in increase in the ESR should not trigger retreatment in the
a minority of patients with biopsy-proven GCA. Relying absence of clinical symptoms that recall the patient’s presen-
upon acute phase reactants for the diagnosis of this disorder tation or which clearly signal the possibility of recurrent
is merely one way that these tests can lead the clinician down GCA. Remember: the ESR normally rises with each decade
292 J. R. Stone et al.

of life. In and of itself, an ESR of 48 mm/h in an 80-year-old management of GCA is a distinct possibility. However,
is not worth spilling any more prednisone over unless there that modality is more likely to be MRI than ultrasound.
is compelling evidence of active disease.
Pearl: Ultrasound is not useful for the detection of disease
Pearl: The serum alkaline phosphatase level can be elevated flares in GCA.
in GCA.
Comment: Ultrasonography, notably color Doppler ultra-
Comment: Approximately one quarter of patients with GCA sonography, is able to delineate both the vessel wall and the
with active disease have elevated levels of serum alkaline lumen of many arteries, including the temporal arteries. In
phosphatase (Kyle et al. 1991). The alkaline phosphatase level GCA, the most specific and sensitive sign of temporal artery
normalizes rapidly following the institution of glucocorticoid inflammation is a concentric hypoechogenic mural thicken-
treatment. The cause for this abnormality is unknown, but it ing (“halo” sign) (Schmidt et al. 1997). This probably repre-
has been hypothesized that hepatic arteritis might be the sents vessel wall edema caused by inflammation. In the hands
underlying lesion. of a skilled and experienced ultrasonographer, the finding of
Myth: A positive test for anticardiolipin antibodies is associ- a halo sign significantly increases the likelihood of a GCA
ated with an increased risk of ischemic complications in GCA. diagnosis. Whether or not it precludes, the need for a tempo-
ral artery biopsy is a matter of debate.
Reality: Patients with GCA test positive for anticardiolipin However, even the most ardent devotees agree that ultra-
antibodies more often than age- and gender-matched controls, sound is not reliable as a gauge of disease activity or as a
but anticardiolipin antibodies do not increase the risk of throm- diagnostic technique for recurrent disease once glucocorti-
bosis in these patients (Duhaut et al. 1998; Manna et al. 1998; coids have been initiated. The utility of ultrasound is limited
Meyer et al. 1996). In a multicenter study, anticardiolipin anti- to the time of initial diagnosis.
bodies were found in 21% of patients with GCA, compared
with 3% of controls (Duhaut et al. 1998). However, no correla- Myth: Angiography is the most sensitive imaging study for
tion was observed between positivity for anticardiolipin anti- GCA in its early stages.
bodies and thrombotic events in any of these studies.
Reality: Angiography is an excellent technique for the docu-
mentation of luminal irregularities such as stenoses or aneu-
rysms in large-vessel vasculitis. However, angiography does
27.3 Imaging not detect vessel wall thickening and mural edema, which
represent vasculitic lesions in an earlier stage of the disease.
Thus, angiography is not useful in diagnosing early large-
Myth: Ultrasonography is a useful procedure for the diagno-
vessel GCA.
sis of GCA in the community.
Pearl: Among patients in whom GCA is suspected despite
Reality: Ultrasonography made a splash when the idea of
negative temporal artery biopsies, large-vessel imaging is an
using this technology to diagnose GCA was introduced in
important adjunct to diagnosis.
the late 1990s. However, subsequent experience as revealed
it to be a tool that is impractical for broad use in the com- Comment: MRI assists in the diagnosis of GCA in some set-
munity. In short, in the current state of its technology, ultra- tings (Fig. 27.3). In GCA with large-vessel involvement,
sound is unlikely to have a significant impact on the diagnosis MRI can demonstrate increased vessel wall thickness in the
or management of GCA. aorta, its primary branches, and even smaller blood vessels
Early studies of this imaging procedure were flawed by (the resolution of the technique improves continually)
the failure to blind ultrasonographers to the clinical history (Pipitone et al. 2008). Vessel wall thickening can suggest the
and physical examination data of the patients. Subsequent diagnosis of GCA and, in the proper clinical setting, is virtu-
studies have indicated that ultrasound of the temporal arter- ally diagnostic of that entity. To date, however, no data from
ies adds no information to the evaluation of a patient with prospective studies have confirmed the utility of specific
possible GCA, beyond that which can be gleaned from a MRI findings (e.g., vessel wall edema or enhancement) for
thorough history and physical examination (Karahaliou et al. distinguishing active disease in large vessels from vascular
2006; Salvarani et al. 2002b). The technique remains highly remodeling, which can be associated with similar changes
operator-dependent, and it seems unlikely that radiologists (Tso et al. 2002).
or other practitioners beyond specialty centers can be trained Preliminary data suggests that MRI at high field strength
to use this technology in the diagnosis and management of (1.5–3 T) is also be sensitive and specific for demonstrating
GCA responsibly and well. inflammation within temporal arteries (Bley et al. 2005a, b).
In the next few years, the emergence of a clinically use- Thus, MRI studies of the temporal arteries might be useful
ful imaging modality for the diagnosis and longitudinal one day in guiding the site of temporal artery biopsy. These
27 Giant Cell Arteritis and Polymyalgia Rheumatica 293

Fig. 27.4 Positron

a
emission tomography
(PET) scanning in giant
cell arteritis. PET scan
showing marked (>grade
1+) tracer uptake in the
ascending aorta as well as
in the supra-aortic
vessels. The high tracer
uptake, the smooth
appearance of the lesions,
and the vessel sites
involved suggest the
vasculitic nature of the
alterations shown by PET

vasculitis can also involve the temporal arteries. Although they

do so less often than giant cell arteritis, their precise diagnosis
is important because of their divergent implications for treat-
ment, prognosis, and the involvement of other organ systems.

Myth: Positron emission tomography (PET) scans of the great

vessels are superior to conventional acute phase reactants in
the assessment of disease activity in large-vessel vasculitis.

Fig. 27.3 Large-vessel involvement in giant cell arteritis. (a) Axial Reality: Fluorodeoxyglucose PET (FDG-PET) scans make
STIR MRI of the thoracic aorta showing vessel wall thickness with very pretty pictures when vasculitis involves large arteries
increased signal intensity corresponding to mural edema. (b) Sagittal
STIR MRI of the thoracic aorta showing vessel wall thickness with
(Fig. 27.4). This technology is helpful in making the diagno-
increased signal intensity corresponding to mural edema, more marked sis of GCA when temporal artery biopsies are negative but the
in the ascending aorta and in the aortic arch focus of disease is within the great vessels (Meller et al. 2003).
(No study to date has compared PET and MRI technologies
preliminary results must be confirmed in rigorous, prospec- head-to-head for the purpose of diagnosis in this setting). In a
tive studies before examination of the temporal arteries by study of 35 patients with new-onset, untreated GCA, increased
MRI becomes part of standard clinical practice. vascular 18fluorodeoxyglucose uptake was observed in 83% of
The potential shortcoming of all currently available imag- patients, mainly in the subclavian arteries (74%), in the aorta
ing modalities with regard to “temporal arteritis” is their failure (>50%), and in the femoral arteries (37%) (Blockmans et al.
to provide histopathologic specificity. Microscopic polyangii- 2006). FDG-PEt also sometimes reveals that patients believed
tis, WG, polyarteritis nodosa, and other forms of systemic to have PMR alone actually have GCA.
294 J. R. Stone et al.

The utility of PET for following disease activity after the general, the entire pathological specimen submitted is not
initiation of treatment remains unclear (Pipitone et al. 2008). It sectioned, therefore, leaving substantial tissue within the
is appealing to believe that FDG-PET studies detect smoldering paraffin block that could harbor hidden GCA.
disease activity within the great vessels and their branches, but The second potential explanation for false-negative tempo-
some data indicate that PET is no better than conventional acute ral artery biopsies is that GCA does not involve the temporal
phase reactants (i.e., the ESR and CRP) for this purpose – which arteries in all patients. This is discussed earlier. Third, inade-
is to say not very good (Blockmans 2008). Part of the reason for quate sampling might be the source of the problem. The skill
this is that the vascular remodeling process that the vessel of the operator who performs the biopsy is important. It is
undergoes following treatment is expected to be associated with easier to biopsy a normal temporal artery than an abnormal
inflammatory components and edema, as is active large-vessel one, but the abnormal ones are the ones one wants! More than
vasculitis. Thus, distinguishing active disease from the healing one pathology report from a temporal artery biopsy procedure
process is not straightforward once treatment with glucocorti- has been signed out as “Nerve,” indicating that the surgical
coids has begun. Additional investigations of FDG-PET as a intern assigned to do the case missed the mark.
mirror of disease activity in large-vessel vasculitis are required, Finally, insufficient length of the surgical specimen can
and the use of serial PET studies as a gauge of disease activity also cause the diagnosis of GCA to be overlooked. Although
is premature. there is no definitive evidence to guide the optimal length of
One limitation of FDG-PET is its inability to visualize the a temporal artery biopsy, a strong consensus of vasculitis
temporal arteries, which is due both to the high background experts – rheumatologists, ophthalmologists, and patholo-
signal of the neighboring brain and to the small diameter of gists alike – supports the excision of at least two centimeters
the temporal arteries, which is below the resolution threshold of temporal artery.
of the PET scanners currently available (Blockmans, 2003;
Myth: There are many contraindications to temporal artery
Bley et al. 2006).
biopsy.
Reality: There are actually very few contraindications to per-
forming a temporal artery biopsy. The contraindications to
27.4 Temporal Artery Biopsy this procedure are shown in Table 27.4. If necessary, tempo-
ral artery biopsy can be performed even while the patient is
Myth: Temporal artery biopsy is an excellent way to make on warfarin therapy, although it is preferable to convert such
the diagnosis of giant cell arteritis. patients to heparin before the biopsy.
Reality: Temporal artery biopsy is the best way to make the One absolute contraindication to temporal artery biopsy
diagnosis of GCA. Unfortunately, it is far from perfect. Even is the potential need for pial synangiosis surgery (external-
when performed under rigorous circumstances, the false- to-internal circulation bypass) in patients with severe intrac-
negative rate of temporal artery biopsies is approximately ranial or extracranial carotid or vertebral artery stenosis. In
10% (Hall et al. 1983). The false-negative rate is probably such cases, the temporal artery branches can be used to sal-
considerably higher in many centers that do not perform as vage severe intracranial arterial stenosis.
many temporal artery biopsies. Myth: Temporal artery biopsies are not necessary in some
One reason is that “temporal arteritis” does not always cases because they will not alter the patient’s management.
clearly involve the temporal arteries (Horton et al. 1932,
1937). We merely use these vessels whenever possible to Reality: This rationalization for not performing the biopsy
make the diagnosis because of their convenient location just usually stems from one of two reasons: (1) reluctance on the
under the skin. Sometimes, the diagnostic tissue is that of the part of the physician to put an elderly patient through a “pro-
occipital artery, not the temporal artery, but the occipital cedure”; or, (2) anticipated difficulties in arranging a timely
artery is less accessible (without shaving the patient’s hair) biopsy.
and therefore is biopsied less often. Many clinicians have misconceptions about how “inva-
sive” a temporal artery biopsy is. The procedure is performed
Myth: Bilateral negative temporal artery biopsies effectively in an outpatient setting, under local anesthesia. The patient
exclude giant cell arteritis.
Reality: There are several potential explanations for false- Table 27.4 Contraindications to temporal artery biopsy
negative temporal artery biopsies. First, skip lesions are Scalp cellulites
known to occur in GCA such that the inflammatory changes Sepsis
do not affect the entire temporal artery. Clinicians should Potential for nonhealing in patients with severe hematologic
realize that the pathologists’ procedure for the evaluation of abnormalities
temporal artery biopsies varies from center to center. In Severe skin conditions that preclude instrumentation along the scalp
27 Giant Cell Arteritis and Polymyalgia Rheumatica 295

leaves with dissolving sutures covered by a Band-aid. Within since amyloidosis may mimic GCA closely. In some cases,
2 weeks, it is often difficult to tell that the patient had the large-vessel imaging (MRI or PET) is helpful.
procedure done. Serious complications are vanishingly rare Finally, if the biopsies are unequivocally normal yet there
when the biopsy is performed by an experienced surgeon. remains a strong case for the disease on clinical grounds,
Thus, confronted with a patient with possible GCA at 4 then the patient must be treated for GCA, fully assured that
p.m. on a Friday afternoon, the proper course is not to decide everything possible to confirm the diagnosis has been done.
that a temporal artery biopsy is unnecessary – but rather to
Myth: Treatment of a patient suspected of GCA with gluco-
begin glucocorticoids at once and to arrange for a biopsy the
corticoids before the temporal artery biopsy renders the pro-
following week. Starting glucocorticoids without making
cedure worthless.
every effort to confirm the diagnosis by biopsy is a sin that
often comes home to roost only months later, when the Reality: This myth was debunked by a paper from the Mayo
patient suffers side effects of glucocorticoids and the next Clinic in 1994 (Achkar et al. 1994). The investigators con-
physician to evaluate her is left to wonder whether or not she firmed that diagnostic histopathological findings could be
ever really had GCA at all. In such cases, the value of a posi- demonstrated readily by experienced pathologists, even 2
tive biopsy is immense. weeks (and as long as 11 months) after the institution of
prednisone. Thus, concern about failing to establish the diag-
Myth: Unilateral temporal artery biopsies are generally suf- nosis because of prebiopsy treatment should never interfere
ficient for the diagnosis of GCA. with the institution of appropriate treatment.
Reality: Unilateral biopsies in GCA are O.K., as long as one Although it is still possible to make the diagnosis well
is content to miss between 20 and 40% of cases (Seo 2005; after starting glucocorticoids, treatment does alter the histo-
Boyev et al. 1999; Pless et al. 2000; Brass et al. 2008). pathological features of GCA. “Atypical” temporal arteritis
Although a unilateral temporal artery biopsy can establish (e.g., the absence of giant cells or confinement of inflamma-
the diagnosis of GCA in many cases, at some institutions tion to the adventitia) was more common in the subset of
both temporal arteries are biopsied routinely. patients treated with more than 14 days of glucocorticoids
Bilateral temporal artery biopsies were performed on 186 before undergoing biopsies (Achkar et al. 1994). Conse-
patients in one study (Boyev et al. 1999). Although only six quently, the temporal artery biopsy should be obtained as
patients had arteritis on only one side, this number repre- promptly as possible following the institution of glucocorti-
sented 20% of the total number of GCA cases diagnosed coids. Failing to do so within a few days, however, is not an
through biopsy. In another study that involved 60 patients opportunity irretrievably lost.
(Pless et al. 2000), eight (13%) had temporal arteritis on only Myth: The surgeon can tell at the time of the biopsy what the
one side. Those eight patients comprised 40% of the total pathologist will find.
number of patients diagnosed with GCA by biopsy.
Considering the potential consequences of a missed diag- Reality: No study has ever validated the notion that the sur-
nosis and the very low morbidity of the procedure, the rou- geon can discern by the appearance and feel of the vessel
tine performance of bilateral temporal artery biopsies is whether or not the procedure will be diagnostic of GCA. In
prudent, especially at institutions where frozen sections can- the case of a nodular, indurated vessel, the positive predictive
not be examined intraoperatively to determine if the first value of such an examination while the skin is open is prob-
biopsy is diagnostic. ably high. However, more subtle cases of GCA exist as well,
and arteries that appear normal at the time of surgery fre-
Pearl: The diagnosis of GCA remains a clinical one in a signifi- quently reveal inflammation consistent with GCA under the
cant number of cases, despite the best efforts of pathologists. microscope.
Comment: As noted elsewhere, even bilateral temporal artery Myth: Using lidocaine with epinephrine is contraindicated
biopsies performed in the most rigorous manner yield false- in temporal artery biopsy.
negative results in some cases – at least 10% but perhaps sub-
Reality: Epinephrine can cause a previously palpable tempo-
stantially higher than this in the community (Hall et al. 1983).
ral artery to constrict, thereby making it more difficult to iden-
What does one do if bilateral biopsies are negative yet
tify during the dissection phase of the procedure. However,
GCA still seems likely? The biopsies should be reviewed
the benefit of diminishing blood loss is preferable in some
with the pathologist to confirm the adequacy of the speci-
cases, and therefore epinephrine should be used according to
mens and the accuracy of the negative reading. Additional
the discretion of the surgeon. If the artery is no longer readily
cuts of the available specimens might be informative, depend-
apparent at dissection, a Doppler device is helpful.
ing on the pathologist’s procedure for reviewing biopsies ini-
tially. It is also important to ensure that a Congo red stain of Myth: Temporal artery biopsies should always be performed
the tissues was performed to exclude amyloid deposition, in the operating room.
296 J. R. Stone et al.

Reality: If sterility can be ensured, a treatment room or a the other hand, if the surgeon is inexperienced, untoward
procedure room is a perfectly acceptable place to perform a results can occur. A facial nerve palsy is one example.
temporal artery biopsy. Lack of timely access to a standard
Pearl: Failure to locate the temporal artery is the biggest
operating room should not delay this important procedure.
potential challenge faced in performing a biopsy.
The procedure is sufficiently simple that an experienced sur-
geon does not require a fully equipped operating room. Comment: Finding normal arteries, which are pulsatile, is
much easier than identifying arteries that are narrowed or
Pearl: Temporal artery biopsies should be performed in the
occluded by GCA. Patients should be informed about this
scalp.
possible difficulty. Before making an incision, the surgeon
Comment: Some surgeons tend to perform biopsies in an should be meticulous about locating and marking the vessel.
area just anterior to the tragus of the ear. This vessel, the A Doppler instrument should be available to help locate the
common (superficial) temporal artery, is not the right one to artery if it is not readily identified by palpation. Owing to the
biopsy, even if a palpable pulse is present at that site. Branches scaphoid shape of surgical incisions, the skin incision should
of the facial nerve are vulnerable at a site that low, and facial be “generous,” as the breadth of the opening will gradually
paralysis can result from a misdirected scalpel. become smaller as the surgeon deepens the incision.
The scalp is a much safer site at which to perform the
biopsies. In the scalp, a rich blood supply promotes rapid
healing of the surgical incision. The resulting scar is minimal 27.5 Pathology
and ultimately hard to detect, even in patients who are bald.
Performance of a temporal artery biopsy in the scalp region
Myth: Pathologists use standardized criteria when assessing
requires shaving of a small region of hair, which grows back
temporal artery biopsies for GCA.
quickly.
The preferred site for the biopsy is the anterior temporal Reality: Some pathologists have outlined their individual
artery, also known as the frontal temporal artery. The poste- opinions concerning minimal diagnostic criteria for GCA.
rior temporal artery, also termed the parietal temporal artery, However, there has never been a consensus statement by vas-
is sometimes selected when the anterior branch is not readily cular pathologists that addresses this issue formally. Significant
found (Fig. 27.5a, b). variability characterizes the criteria employed by different
pathologists. As an example, some pathologists consider any
Pearl: The invasiveness of temporal artery biopsy procedure
adventitial inflammation sufficient for the diagnosis of GCA,
is overestimated.
but others demur.
Comment: Complications of temporal artery biopsies per- This dilemma of variability among the opinions of pathol-
formed in the scalp region are very rare. As with any surgery, ogists is illustrated by the skip lesion shown in Fig. 27.6.
bleeding or infection may occur, but these are is unusual. On Most pathologists would consider the right side of the image

a b

Fig. 27.5 Optimal and risky sites

for temporal artery biopsies. The
biopsy in (a) was taken from the
anterior temporal artery and is
generally the optimal site for
biopsy. The biopsy in (b) was
taken from the common temporal
artery. Biopsy at this site runs the
risk of facial nerve injury
27 Giant Cell Arteritis and Polymyalgia Rheumatica 297

Fig. 27.6 Longitudinal section of a temporal artery biopsy with a skip

lesion. On the right side of the sample, there is transmural inflammation
composed of lymphocytes and macrophages, accompanied by intimal
hyperplasia. Moving from right to left, there is progressively less
inflammation. At the far left is a noninflamed segment of the artery.
(Hematoxylin and eosin stained slide at 100× magnification) Fig. 27.7 Wegener’s granulomatosis involving a temporal artery. There
is extensive necrosis of the arterial wall with a mixed inflammatory infil-
to be diagnostic of GCA and the left side not. However, upon trate that contains lymphocytes, macrophages, and frequent neutrophils
(Hematoxylin and eosin stained slide of a temporal artery biopsy at 200×
moving from right to left in the biopsy, the precise site at
magnification). This patient presented with recurrent amaurosis fugax
which criteria sufficient for a GCA diagnosis are met varies but was found to be positive for antineutrophil cytoplasmic antibodies
from pathologist to pathologist. directed against proteinase-3 (i.e., PR3-ANCA). She also had pulmonary
Clinicians must be conversant with the histopathological nodules that revealed granulomatous inflammation on biopsy
features of GCA themselves, conscious of the variability
among pathologists, and willing to engage pathologists in a
discussion about the final histopathologic diagnosis. Pearl: In the temporal artery, the presence of extensive necro-
sis suggests involvement by a vasculitis distinct from GCA.
Myth: Transmural inflammation is required for a pathologic
diagnosis of GCA. Comment: In the aorta, GCA can display large areas of
necrosis. However, when GCA is present in the temporal
Reality: GCA is a segmental disease that has “skip lesions.”
arteries, the amount of necrosis is typically limited. Extensive
Figure 27.6 also demonstrates how GCA can cause a spec-
necrosis in a temporal artery biopsy, particularly fibrinoid
trum of severity within the same vessel: transmural inflam-
necrosis, suggests the presence of a different vasculitis such
mation is evident on the right side of the Figure. In contrast,
as WG, the CSS, microscopic polyangiitis, polyarteritis
the adventitia and internal elastic lamina are the sites of pri-
nodosa, or mixed cryoglobulinemia (Fig. 27.7).
mary involvement in the center.
Some pathologists would classify the findings in the cen- Pearl: Amyloid deposition in the temporal artery can stimu-
ter portion of Fig. 27.6 as “atypical GCA” because of the late reactive inflammatory infiltrates that mimic GCA.
restricted nature of the inflammatory infiltrates. In many
Comment: Amyloid deposits, particularly those caused by AL
such cases, the adventitial component is predominantly lym-
amyloidosis, can stimulate reactive inflammatory responses in
phocytic and the component located at the internal elastic
multiple tissues, including the temporal artery (Fig. 27.8). The
lamina primarily contains macrophages.
amyloid may not be appreciated without Congo red staining,
Although transmural inflammation is not required for the
and in such cases the inflammation could be misdiagnosed as
diagnosis of active GCA, some degree of involvement of the
GCA (Churchill et al. 2003).
arterial media or intima by macrophages and lymphocytes is
Amyloidosis can also cause jaw claudication. One case
typically present (Allsop and Gallagher 1981).
series reported 22 patients with primary systemic amyloido-
Myth: Giant cells are always present in lesions of active GCA. sis who complained of typical jaw claudication. Some of these
patients also had claudication of the upper and lower extrem-
Reality: Giant cells are neither required for a diagnosis of
ities. Appropriate staining of temporal artery biopsies often
GCA, nor are they entirely specific for that condition. In fact,
revealed amyloid deposits in the temporal arteries in patients
giant cells are present in approximately only one half of tem-
who were clinically misdiagnosed as having GCA (Gertz
poral artery biopsies that demonstrate active GCA (Lie 1990).
et al. 1986).
Giant cells can also be detected in other conditions that
involve temporal artery inflammation, such as WG and sys- Myth: A temporal artery biopsy revealing lymphocytic inflam-
temic amyloidosis. mation is diagnostic of GCA.
298 J. R. Stone et al.

Fig. 27.8 Systemic amyloidosis involving a temporal artery. The patient Fig. 27.9 Giant cell arteritis found incidentally in a hysterectomy spec-
was a 76 year-old woman who presented with jaw claudication and scalp imen. The patient was a 76 year-old woman who underwent the proce-
tenderness. The biopsy shows numerous large, nodular deposits of amy- dure because of uterine prolapse. She had no history, symptoms, or
loid within the wall of the vessel with a secondary reactive inflammatory signs to suggest vasculitis. (Hematoxylin and eosin stained section at
infiltrate. Subsequent evaluation, including bone marrow biopsy, 200× magnification)
revealed the presence of multiple myeloma. (Hematoxylin and eosin
stained section of a temporal artery biopsy at 200× magnification)
lymphocytic infiltrates is only present in 4–5% of biopsies
(Allsop and Gallaghar 1981).
Reality: The classic picture of granulomatous inflammation
Myth: The pathologic lesions of active GCA are always asso-
is detected in only approximately 50% of temporal artery
ciated with a clinically apparent systemic vasculitis.
biopsy specimens from symptomatic GCA patients. Most of
the other 50% of cases demonstrate a panarteritis, with an Reality: Large autopsy studies indicate that the GCA is ten
infiltrate comprised primarily of lymphocytes, monocytes, times more prevalent pathologically than clinically. Thus, in
and occasional other cells, e.g., neutrophils and eosinophils, 90% of patients who have pathologic lesions of GCA, the
but no giant cells. disease remains subclinical.
Increasing numbers of reports have documented such path- Pathologic lesions of GCA are observed unexpectedly on
ological findings in patients with PAN, WG, CSS, and sarcoi- occasion in surgical specimens, particularly specimens from
dosis (Small and Brisson 1991; Morgan and Harris 1978; the ascending aorta and the female genital tract (Fig. 27.9). In
Genereau et al. 1999). Headache, musculoskeletal pains, and a series of 1,204 patients from the Cleveland Clinic who
a temporal artery biopsy demonstrating vasculitis are highly underwent aortic surgery, 4% were found to have aortitis that
suggestive but not diagnostic of GCA. In the setting of upper was unsuspected prior to surgery (Rojo-Leyva et al. 2000).
airway disease, asthma, polyneuropathy, hilar adenopathy, or Some of these patients have no signs or symptoms of a sys-
a positive serum assay for ANCA, one must remain suspi- temic vasculitis, and many do well without systemic immuno-
cious of other entities. suppression once the “nidus” of vasculitis has been removed.
Pearl: In temporal artery biopsies, atherosclerosis is less com- Pearl: Immunohistochemistry can assist in the pathologic
mon than GCA. diagnosis of GCA.
Comment: Atherosclerosis can be associated with adventitial Comment: Although lymphocytes are usually readily
lymphocytic infiltrates within the temporal arteries. However, apparent on routine hematoxylin and eosin stained slides,
caution should be exercised before attributing these findings subtle macrophage infiltrates can be overlooked (Fig. 27.10).
to atherosclerosis. GCA is more likely to be the cause. The Immunohistochemical staining for the macrophage marker
superficial temporal arteries are relatively resistant to athero- CD68 assists in the diagnosis of GCA when these infiltrates
sclerosis. are subtle (Gooi et al. 2008; Font and Prabhakaran 2007).
Arteriosclerotic changes such as intimal hyperplasia and
Myth: “Healed arteritis” – the presence of scarring in a tem-
fragmentation of the internal elastic lamina are quite common
poral artery biopsy – is diagnostic of previously active GCA.
in temporal artery biopsy specimens. (Disruption of the inter-
nal elastic lamina is not synonymous with GCA!). However, Reality: Scarring is a nonspecific response to injury. Scarring
true atherosclerosis with lipid deposition and adventitial in a temporal artery can represent healed vasculitis, healed
27 Giant Cell Arteritis and Polymyalgia Rheumatica 299

Fig. 27.10 Immunohistochemistry as an aid in evaluating temporal rophage marker CD68 reveals that numerous macrophages are also
artery biopsies. In this temporal artery biopsy specimen, a lymphocytic present in the adventitia, and that this macrophage infiltrate extends
infiltrate is clearly present in the adventitia of the hematoxylin and into the media (b). (Magnification for both images is 400×)
eosin stained slide (a). An immunohistochemistry stain for the mac-

trauma, or the effects of some other injury. The superficial Comment: Diagnostic information can be present in the
temporal arteries are particularly prone to traumatic injury small branches of the temporal artery. As examples, sarcoi-
because of their proximity to the skin, and scarring from dosis and WG have been diagnosed from histopathological
traumatic injury cannot be distinguished from scarring from findings within small blood vessels or connective tissue that
healed vasculitis. Moreover, even assuming that a systemic is adherent to a normal temporal artery. In addition, GCA
vasculitis caused the “healed arteritis,” there is no pathologic itself is sometimes present in branches smaller than the
means of pinning the precise diagnosis on GCA as opposed temporal artery (Esteban et al. 2001). The surgeon should
to some other form of systemic vasculitis. not strive to remove tissue that is attached to the vessel.
“Healed arteritis” in a temporal artery is not synonymous The specimen should be placed in fixative and then sec-
with GCA. The term creates confusion for both pathologists tioned and stained expeditiously. The technician should not
and clinicians alike, and should be discarded. strip or clean the specimen prior to embedding, lest one be
deprived of potentially useful information from examination
Pearl: Frozen section analysis of a temporal artery biopsy
of the adherent tissues. Owing to the possibility of skip areas,
can hinder or prevent a pathologic diagnosis of GCA.
sub-serial sectioning should be requested.
Comment: The preferred method of examining temporal
artery biopsies is through the assessment of cross sections of
formalin-fixed, paraffin-embedded samples. The procedure
of snap-freezing arteries significantly distorts the histology. 27.6 Treatment
In addition, significant amounts of tissue are lost during
the processes of obtaining the frozen section, thawing and Pearl: Alternate-day glucocorticoid tapers are ineffective in
reprocessing it, and cutting new sections from the paraffin- GCA.
embedded material. Both of these factors hinder the ability
Comment: For most patients, the worst part of GCA is the
of the pathologist to diagnose GCA, albeit the diagnosis is
prednisone (Hellmann et al. 2003). But clinicians should
still possible in many cases.
resist the temptation to try alternate-day glucocorticoid ther-
Diagnosing GCA is difficult enough without suboptimal
apy in GCA; it is less effective than daily prednisone. A ran-
processing of the sample. Communicate with the surgeon
domized, controlled trial demonstrated this convincingly in
before the procedure and instruct him or her not to perform a
the 1970s (Hunder et al. 1975). Moreover, there is little evi-
frozen section unless the identity of the tissue at the time of
dence that alternate-day glucocorticoids tapers lead to less
the operation is in question.
toxicity. In the long run, patients and clinicians are better off
Pearl: GCA can sometimes be identified in small branches of controlling the disease with the lowest possible daily dose of
the temporal artery. prednisone.
300 J. R. Stone et al.

Pearl: High doses of glucocorticoids can cause patients to medication an important adjunct to glucocorticoids. A daily
hear voices. baby aspirin should be part of the standard therapy for GCA
(Stone, 2007).
Comment: Glucocorticoids can cause auditory hallucinations When GCA is suspected strongly, the patient can begin a
(Stone, 2006). Don’t let the Psychiatry Liaison Consult Service baby aspirin even before the temporal artery biopsy is done.
be the ones to make the diagnosis of a glucocorticoids-induced (The performance of a temporal artery biopsy in a patient on
psychosis. aspirin is safe.)
Pearl: High doses of glucocorticoids can trigger major Myth: Judicious tapering of glucocorticoids can prevent dis-
depressions. ease flares in GCA.
Comment: Clinicians who are experienced in the use of glu- Reality: Rapid tapering of glucocorticoids is associated with
cocorticoids do not use this treatment cavalierly. High doses an increased risk of disease flares in both GCA and PMR
of prednisone can trigger life-threatening depression, partic- (Pipitone et al. 2005). However, the converse is not necessar-
ularly in individuals with histories of a depressed mood. ily true. In a sizeable number of patients, disease flares appear
Doses may need to be decreased more quickly than most cli- to have little relationship to the speed of the glucocorticoid
nicians are comfortable with in this setting. tapering regimen. The issue for these patients relates more to
Myth: PMR always improve within 24–48 h of the adminis- a threshold dose of prednisone below which they cannot
tration of prednisone. taper, no matter how slowly the prednisone is decreased. If a
patient suffers repeated flares on 3 mg of prednisone a day,
Reality: Classic “PMR responses” to prednisone do occur then she should remain on 4 mg of prednisone a day.
and are enormously gratifying to the clinician. However, they
do not always occur. In some patients with PMR, real clinical Myth: GCA always goes into remission if sufficient doses of
responses can require up to a week. Because the clinical glucocorticoids are given.
response to prednisone can be less than dramatic over the first Reality: Patients with GCA usually demonstrate swift clini-
few days in patients who really have PMR, clinicians must do cal responses following the institution of glucocorticoids,
everything possible to design a clean clinical experiment: and the majority achieves disease remissions while on this
treatment. However, in a small number of patients, GCA is
• Start with 15 mg/day of prednisone. Ten mg/day may be
not suppressed clinically despite large doses of glucocorti-
too little to treat PMR, and more than 20 mg/day might be
coids administered over prolonged periods (Caselli et al.
too much (i.e., it might treat other diseases entities effec-
1988). In such cases, histopathological analysis of clinical
tively, too).
samples (e.g., limb amputation specimens) shows chronic
• Do not change more than one variable at once. Do not add
arteritis. The best treatment options for such patients are not
an nonsteroidal antiinflammatory drug at the same time,
clear. Fortunately, such patients are very rare, and nearly all
do not give additional medications designed to relieve
patients respond clinically to glucocorticoids.
pain, and do not add an antidepressant or sleep medica-
tion, too. Let the prednisone act alone. Myth: All patients with GCA can be weaned off glucocorti-
• See the patient again in 1 week. A longer period allows coids within 2 years of the start of treatment.
too many potential confounders to complicate the response
Reality: Many patients with GCA are able to discontinue
to treatment.
glucocorticoid treatment within 6 months to 2 years. For
Within 1 week of starting prednisone, a patient with PMR instance, in a population-based study from Olmsted County,
should feel like a new person. If not, you have got the wrong Minnesota, the median duration of glucocorticoid treatment
diagnosis. was 7 months (Huston et al. 1978).
Unfortunately, ample other published experience con-
Pearl: Aspirin is an important adjunct to glucocorticoids in firms that this is by no means always the rule. In a study from
the therapy of GCA. Scandinavia of ninety patients with GCA, PMR, or both,
25% of those followed for 9 years were still on prednisone at
Comment: The results of clinical trials designed to identify an
an average dose of 5 mg/day (Andersson et al. 1986). Another
effective steroid-sparing agent in GCA have been disappoint-
study on ninety patients with biopsy-verified GCA reported
ing. However, two retrospective cohort studies demonstrated
that one third of the patients required low-dose glucocorti-
that a baby aspirin along with prednisone is more effective at
coid treatment indefinitely (Graham et al. 1981).
preventing visual loss and cerebral ischemic events than is
prednisone alone (Nesher et al. 2004; Lee et al. 2006). Pearl: Revascularization is rarely needed in large-vessel
The consistency of these two studies and the strong reduc- vasculitis, even if the artery appears to be narrowed beyond
tion in the odds ratio for clinical events by aspirin make this the point that blood can pass.
27 Giant Cell Arteritis and Polymyalgia Rheumatica 301

Fig. 27.11 Abundant collateral

circulation in GCA. Exuberant
collateral blood vessels that have
formed to compensate for a
severe stenosis at the junction of
the subclavian and axillary
arteries. An attempt at stent
placement for a tight subclavian
lesion failed (the stent clotted
before the patient even left the
radiology suite). The abundant
collateral vessels make the point
that the stent placement was
probably ill-advised, anyway

Comment: One of the striking angiographic features of GCA Myth: The prompt use of pulse glucocorticoids may restore
(and Takayasu’s arteritis) is the exuberant neovasculariza- vision lost in GCA.
tion and collateral circulation that develops in patients with
Reality: This, unfortunately, is wishful thinking. Stuttering
narrowing of the subclavian arteries and other large vessels
losses of vision may occur in which sight is gone for a sec-
(Fig. 27.11). Many patients with large-vessel GCA lose pal-
ond to minutes, but dense vision loss that has persisted for
pable pulses in one or both upper extremities. Despite this
hours in GCA is usually permanent.
potentially frightening finding, such patients inevitably have
A carefully performed study that examined visual out-
sufficient blood flow to the extremity to maintain its viabil-
comes in GCA patients who had lost vision concluded that
ity. Many such patients, in fact, are asymptomatic.
essentially all improvement in patients’ ability to read eye
Attempts at revascularization in patients with large-vessel
charts was related to their accommodation of new visual dif-
GCA and stenotic lesions of primary branches of the aorta are
ficulty rather than to qualitative improvement in ocular func-
frequently misguided. The collateral circulation that develops
tion (Hayreh et al. 2002).
is not pulsatile – patients remain “pulseless” – but still have
In a patient who presents with the vision gone in one eye,
sufficient blood flow with stents, angioplasty, or bypass. In
a reasonable approach is to initiate treatment with methyl-
this sense, the purpose of prednisone is to suppress inflamma-
prednisolone 1 g/day times three (plus a baby aspirin, if the
tion sufficiently that neovascularization can occur.
patient is not already on it). The reason? There is no guarantee
that even this will preserve vision in the other eye. If therapy
fails, no clinician wants to be left wondering, “What if…?”

27.7 Outcomes Myth: Reversible loss of vision is rare in GCA.

Reality: It is often stated that visual loss in GCA is perma-
Myth: Prompt treatment with glucocorticoids prevents new nent (see above). However, transient, fully reversible, visual
visual loss in GCA. obscurations and temporary monocular blindness occur fre-
quently in GCA. These episodes may last up to hours, yet
Reality: Would that this were true! Although prompt treatment
still be reversible.
usually prevents further loss of vision and protects the fellow
Funduscopic examination in this setting may demonstrate
eye, there are numerous published examples in which there
a completely normal optic nerve and fundus, mild disc
was not only further loss of vision in the eye originally affected,
edema, hemorrhagic disc edema, branch retinal artery infarc-
but also involvement of the second eye, normal at baseline.
tion, and/or cotton wool spots. The full reversibility of the
The prevalence of vision loss in one randomized clinical
visual disturbances underscores the intermittent disruptions
trial of 98 patients with GCA was 18% at study entry
in the blood supply of affected vasculitic territories.
(Hoffman et al. 2002). The incidence of new vision loss at 1
Unfortunately, once visual loss has persisted for more
year was 14%, despite the fact that patients received high-
than several hours, no recovery can be expected. The visual
dose daily prednisone for 3 months before transitioning to an
deficit will be permanent, indeed.
alternate-day tapering regimen. There is wisdom in inform-
ing patients that although they are being properly treated, the Pearl: Transient visual obscurations triggered by a Valsalva
threat of visual loss (or further visual loss) remains and may maneuver or change in body position indicate a GCA patient
be higher than generally regarded by clinicians. at risk for visual loss.
302 J. R. Stone et al.

require longer treatment courses compared with those with

isolated GCA or PMR (Lundberg and Hedfors 1990; Kyle
and Hazleman 1993).
One common error in treating GCA is the use of high
doses of glucocorticoids for too long at the start of therapy.
The general approach should be high doses of prednisone
(e.g., 60 mg/day) for 1 month, followed by tapering to mod-
erate daily doses by the end of month two (e.g., 20 mg/day).
After that, the prednisone taper should be designed with a
long “tail,” such that patients are decreasing by 1 mg/month
as the taper finishes (5 mg/day for 1 month, then 4 mg/day for
1 month, then 3 mg/day, etcetera).
Pearl: GCA is not associated with an increased mortality.
Fig. 27.12 Fluorescein Angiogram of the right eye in a patient with arter- Comment: GCA is fraught with significant disease- and
itic ischemic optic neuropathy. Note the areas of fluorescein hypoperfu-
sion of the deep choroid along the superior and inferior retinal arterioles treatment-related morbidity. However, studies reveal no dif-
ferences between GCA patients and age- and gender-matched
Comment: A swollen optic nerve may experience transient controls in terms of mortality (Pipitone et al. 2006). The only
disturbances of blood flow associated with Valsalva maneu- disease subset known to have an increased risk of mortality
vers (e.g., coughing). This can lead to a fully reversible loss is the group of patients with thoracic aorta aneurysms
of vision that is fleeting or persistent for up to several min- (Nuenninghoff et al. 2003). This complication carries a sub-
utes. It is critical to recognize that transient visual loss with stantial risk of death from aortic dissection.
or without systemic symptoms might be a clue to the pres- Myth: Orbital ischemia and proptosis are not seen in GCA.
ence of GCA.
Reality: Orbital ischemia may result in GCA from vasculitic
Pearl: Very dense loss of vision tends to occur more fre- involvement of the proximal ophthalmic artery branches that
quently in arteritic AION than in nonarteritic AION spares the posterior ciliary arteries. This pattern of vascular
(NAAION). involvement can cause orbital pain, proptosis, ptosis, and
Comment: Once in place, the vision loss that results from diplopia, with or without vision loss.
GCA tends to be dense and more often than not is associated Pearl: Distinguishing PMR from RA is often impossible at
with complete monocular blindness in the affected eye. the patient’s initial presentation.
Small, sectoral visual field defects or well-preserved visual
acuity are more typical of NAAION, which is typically Comment: Patients with PMR can have synovitis in periph-
caused by an embolic lesion from an atheromatous source in eral joints. Sometimes only a period of observation deter-
the aorta or its primary branches. mines whether PMR is the correct diagnosis or if the true
Fluorescein angiography can be helpful in discriminating cause of the patient’s symptoms is RA. The availability of
GCA from nonarteritic AION (Fig. 27.12). Large areas of serologic tests for antibodies to anti-cyclic citrullinated pep-
nonperfusion in the posterior pole, choroidal nonperfusion, tides (anti-CCP antibodies) has facilitated this distinction in
and circumferential optic nerve nonperfusion (as opposed to some cases, but some patients with RA are negative for both
segmental nonperfusion) favor GCA. rheumatoid factor and anti-CCP antibodies at presentation.

Pearl: Patients with GCA and PMR have a longer disease In general, patients with RA improve on smaller doses of
course compared with those with GCA or PMR alone. glucocorticoids than do patients with PMR, but they are less
likely to enter complete remissions with prednisone alone.
Comment: No reliable predictors of glucocorticoid treatment
Pearl: If a patient with PMR consistently requires more than
duration exist. Treatment regimens must be tailored to each
20 mg of prednisone a day to control the disease symptoms,
individual patient and adjusted as needed on the basis of
the diagnosis is wrong.
clinical course. Most attempts to discontinue prednisone
altogether in less than 1 year, unfortunately, are met with Comment: A dose of 15 mg/day of prednisone is sufficient to
treatment failure and disease flares at the lower doses. control most cases of PMR. Failure of 20 mg of prednisone
The mean duration of therapy for patients with GCA is administered in divided doses to control the patient’s symp-
approximately 2 years, taking into consideration repeat toms should certainly make one rethink the diagnosis and
courses of treatment necessitated by disease flares. There is consider such entities as GCA (which may require higher
some suggestion that patients with both GCA and PMR initial glucocorticoids doses), cancer (renal cell carcinoma,
27 Giant Cell Arteritis and Polymyalgia Rheumatica 303

sarcoma, multiple myeloma and other hematopoietic malig- Boyev LR, Miller NR, Green WR. Efficacy of unilateral versus bilateral
nancies), amyloidosis, fibromyalgia, and bursitis. temporal artery biopsies for the diagnosis of giant cell arteritis. Am
J Ophthalmol 1999;128:211–5
Pearl: Methotrexate is effective as steroid-sparing agent in Brack A, Martinez-Taboada V, Stanson A, Goronzy JJ, Weyand CM.
GCA. Disease pattern in cranial and large-vessel giant cell arteritis.
Arthritis Rheum 1999;42(2):311–7
Comment: Truly, no one knows for certain whether this is a Brass SD, Durand ML, Stone JH, et al A 59 year-old man with chronic
Pearl or a Myth. Three randomized controlled trials have daily headaches. N Engl J Med 2008; 359(21):2267–78.
Caselli RJ, Daube JR, Hunder GG, et al Peripheral neuropathic syn-
assessed the efficacy of methotrexate (MTX) in recent-onset dromes in giant cell (temporal) arteritis. Neurology 1988;38: 685–9
GCA, producing discordant results (Jover et al. 2001; Caselli RJ, Hunder GG, Whisnant JP. Neurologic disease in biopsy-
Hoffman et al. 2002; Spiera et al. 2001). The results from proven giant cell (temporal) arteritis. Neurology 1988; 38:352–9.
these trials were pooled and analyzed (Mahr et al. 2007). Churchill CH, Abril A, Krishna M, et al Jaw claudication in primary
amyloidosis: unusual presentation of a rare disease. J Rheumatol
According to this meta-analysis, MTX used as adjunctive 2003;30:2283–6
treatment in GCA at a mean dosage of 11.1 mg per week led Cohen MD, Ginsburg WW, Allen GL. Facial swelling in giant cell
to a reduction of both first and second relapses of GCA, with arteritis. J Rheumatol 1982;9:325–7
hazard ratios of 0.65 and 0.49, respectively. However, despite Duhaut P, Berruyer M, Pinede L, et al Anticardiolipin antibodies and
giant cell arteritis: a prospective, multicenter case-control study.
a significant between-group difference in the cumulative GC Groupe de Recherche sur l’Arterite a Cellules Geantes. Arthritis
dose by 842 mg at week 48, MTX use did not lead to a Rheum 1998;41(4):701–9
decreased incidence of glucocorticoid-related complications. Esposito AL, Gleckman RA. Fever of unknown origin in the elderly. J
In addition, the onset of action of MTX appeared to be slow. Am Geriatr Soc 1978;26:498–505
Esteban MJ, Font C, Hernandex-Rodriguez J, et al Small-vessel vascu-
At this point, one conclusion appears safe: If MTX has a litis surrounding a spared temporal artery: clinical and pathological
beneficial effect in GCA, it is a small one relative to that of glu- findings in a series of twenty-eight patients. Arthritis Rheum
cocorticoids. It is conceivable that a dose of MTX higher than 2001;44:1387–95
that used in any of the three clinical trials to date might lead to Evans J, O’Fallon W, Hunder G. Increased incidence of aortic aneu-
rysm and dissection in giant cell (temporal) arteritis. Ann Int Med
greater efficacy than that demonstrated to date. The maximum 1995;122:502–7
dose used in any trial to date has been only 15 mg/week. Font RL, Prabhakaran VC. Histological parameters helpful in recogniz-
ing steroid-treated temporal arteritis: an analysis of 35 cases. Br J
Ophthalmol 2007;91:204–9
References Friedman G, Friedman B. The sensation of facial swelling in temporal
arteritis: a predictor for the development of visual disturbance.
Postgrad Med J 1986;62(733):1019–20
Achkar A, Lie J, Hunder G, et al How does previous corticosteroid Genereau T, Lortholary O, Pottier M, et al Temporal artery biopsy. A
treatment affect the biopsy findings in giant cell (temporal) arteritis? diagnostic tool for systemic necrotizing vasculitis. Arthritis Rheum
Ann Intern Med 1994;120:987–92 1999;42:2674–81
Allsop CJ, Gallagher PJ. Temporal artery biopsy in giant-cell arteritis: Gertz MA, Kyle RA, Griffing WL, et al Jaw claudication in primary
a reappraisal. Am J Surg Pathol 1981;5:317–23 systemic amyloidosis. Medicine 1986;65:173–9
Amor-Dorado JC, Llorca J, Garcia-Porrua C, et al Audiovestibular Gooi P, Brownstein S, Rawlings N. Temporal arteritis: a dilemma in clini-
manifestations in giant cell arteritis: a prospective study. Medicine cal and pathological diagnosis. Can J Ophthalmol 2008;43: 119–20
(Baltimore) 2003;82(1):13–26 Gordon L, Levin L. Visual loss in giant cell arteritis. JAMA 1998; 280:
Andersson R, Malmvall BE, Bengtsson BA. Long-term corticosteroid 385–6
treatment in giant cell arteritis. Acta Med Scand 1986;220(5):465–9 Graham E, Holland A, Avery A, Russell RW. Prognosis in giant-cell
Bley TA, Brink I, Reinhard M. [Imaging procedures for giant cell arteri- arteritis. Br Med J (Clin Res Ed) 1981;282(6260):269–71
tis (Horton’s disease)]. Ophthalmologe 2006;103(4):308–16 Hall S, Lie J, Kurland L, et al The therapeutic impact of temporal artery
Bley TA, Weiben O, Uhl M, et al Assessment of the cranial involvement biopsy. Lancet 1983;2:1217–20
pattern of giant cell arteritis with 3 T magnetic resonance imaging. Hausch RC, Harrington T. Temporal arteritis and sensorineural hearing
Arthritis Rheum 2005a;52(8):2470–7 loss. Semin Arthritis Rheum 1998;28(3):206–9
Bley TA, Wieben O, Uhl M, et al High-resolution MRI in giant cell Hayreh SS, Podhajsky PA, Zimmerman B. Occult giant cell arteritis:
arteritis: imaging of the wall of the superficial temporal artery. AJR ocular manifestations. Am J Ophthalmol 1998a;125(4):521–6
Am J Roentgenol 2005b;184(1):283–7 Hayreh SS, Podhajsky PA, Zimmerman B. Ocular manifestations of
Blockmans D. The use of (18F)fluoro-deoxyglucose positron emission giant cell arteritis. Am J Ophthalmol 1998b;125(4):509–20
tomography in the assessment of large vessel vasculitis. Clin Exp Hayreh SS, Zimmerman B, Kardon RH. Visual improvement with cor-
Rheumatol 2003;21(6 Suppl 32):S15–22 ticosteroid therapy in giant cell arteritis. Report of a large study and
Blockmans D. PET in large-vessel vasculitis and polymyalgia rheu- review of literature. Acta Ophthalmol Scand 2002;80:355–67
matica. In: Fourth International Conference on Giant Cell Arteritis Hellmann DB, Uhlfelder ML, Stone JH, et al Domains of quality of life
and Polymyalgia Rheumatica, New York; 2008 important to patients with giant cell arteritis. Arthritis Rheum
Blockmans D, De Ceuninck L, Vanderschueren S, et al Repetitive 2003;49(6):819–25
18F-fluorodeoxyglucose positron emission tomography in giant cell Hoffman GS, Cid MC, Hellmann DB, et al A multicenter, randomized,
arteritis: a prospective study of 35 patients. Arthritis Rheum 2006;55 double-blind, placebo-controlled trial of adjuvant methotrexate treat-
(1):131–7 ment for giant cell arteritis. Arthritis Rheum 2002;46(5):1309–18
Bongartz T, Matteson EL. Large-vessel involvement in giant cell arteri- Horton BT, Magath TB, Brown GE. Undescribed forms of arteritis of
tis. Curr Opin Rheumatol 2006;18(1):10–7 temporal vessels. Proc Staff Meet Mayo Clin 1932;7:700–1
304 J. R. Stone et al.

Horton BT, Magath TB, Brown GE. Arteritis of temporal vessels: report Paraskevas KI, Boumpas DT, Vrentzos GE, Mikhailidis DP. Oral and
of 7 cases. Proc Staff Meet Mayo Clin 1937;12:548–53 ocular/orbital manifestations of temporal arteritis: a disease with
Hunder GG, Sheps SG, Allen GL, Joyce JW. Daily and alternate-day deceptive clinical symptoms and devastating consequences. Clin
corticosteroid regimens in treatment of giant cell arteritis: compari- Rheumatol 2007;26(7):1044–8
son in a prospective study. Ann Intern Med 1975;82:613–8 Pipitone N, Boiardi L, Bajocchi G, Salvarani C. Long-term outcome of
Huston KA, Hunder GG, Lie JT, et al Temporal arteritis: a 25-year epi- giant cell arteritis. Clin Exp Rheumatol 2006;24(2 Suppl 41):
demiologic, clinical, and pathologic study. Ann Intern Med 1978; S65–70
88(2):162–7 Pipitone N, Boiardi L, Salvarani C. Are steroids alone sufficient for the
Jover JA, Hernandez-Garcia C, Morado IC, et al Combined treatment of treatment of giant cell arteritis? Best Pract Res Clin Rheumatol
giant-cell arteritis with methotrexate and prednisone. a randomized, 2005;19(2):277–92
double-blind, placebo-controlled trial. Ann Intern Med 2001;134 Pipitone N, Versari A, Salvarani C. Role of imaging studies in the diagno-
(2):106–14 sis and follow-up of large-vessel vasculitis: an update. Rheumatology
Karahaliou M, et al Colour duplex sonography of temporal arteries (Oxford) 2008;47(4):403–8
before decision for biopsy: a prospective study in 55 patients with Pless M, Rizzo JF III, Lamkin JC, Lessell S. Concordance of bilateral
suspected giant cell arteritis. Arthritis Res Ther 2006;8:R116 temporal artery biopsy in giant cell arteritis. J Neuroophthalmol
Kyle V, Hazleman B. Treatment of polymyalgia rheumatica and giant 2000;20:216–8
cell arteritis. II. Relation between corticosteroid dose and corticos- Porter MJ. Temporal arteritis presenting as a submandibular swelling.
teroid-associated diseases. Ann Rheum Dis 1989;48:662–6 J Laryngol Otol 1990;104(10):819–20
Kyle V, Hazleman BL. The clinical and laboratory course of polymyal- Razavi M, Jones, RD, Manzel, et al Steroid-responsive Charles Bonnet
gia rheumatica/giant cell arteritis after the first two months of treat- syndrome in temporal arteritis. J Neuropsychiatry Clin Neurosci
ment. Ann Rheum Dis 1993;52(12):847–50 2004;16:505
Kyle V, Wraight EP, Hazleman BL. Liver scan abnormalities in polymyal- Regan MJ, Green WR, Stone JH. Ethnic disparity in the incidence of
gia rheumatica/giant cell arteritis. Clin Rheumatol 1991;10(3):294–7 temporal arteritis: a 32-year expereince at an urban medical center.
Larson T, Hall S, Hepper N, et al Respiratory tract symptoms as a clue Arthritis Rheum 2000;43:S128
to giant cell arteritis. Ann Intern Med 1984;101:594–7 Rojo-Leyva F, Ratliff N, Cosgrove D, et al Study of 52 patients with
Lee CC, Su WW, Hunder GG. Dysarthria associated with giant cell idiopathic aortitis from a cohort of 1204 surgical cases. Arthritis
arteritis. J Rheumatol 1999;26:931–2 Rheum 2000;43:901–7
Lee MS, Smith SD, Gabor A. Antiplatelet and anticoagulant therapy in Rucker JC, Biousse V, Newman NJ. Ischemic optic neuropathies. Curr
patients with giant cell arteritis. Arthritis Rheum 2006;54:3306–9 Opin Neurol 2004;17(1):27–35
Lessell S. Nonarteritic anterior ischemic optic neuropathy: enigma Ruiz-Masera JJ, Alamillos-Granados FJ, Dean-Ferrer A, et al Sub-
variations. Arch Ophthalmol 1999;117:386–8 mandibular swelling as the first manifestation of giant cell arteritis.
Lie JT. Illustrated histopathologic classification criteria for selected Report of a case. J Craniomaxillofac Surg 1995;23(2):119–21
vasculitis syndromes. Arthritis Rheum 1990;33:1074–87 Salvarani C, Cantini F, Boiardi L, Hunder GG. Polymyalgia rheumatica
Liozon E, Roblot P, Paire D, et al Anticardiolipin antibody levels pre- and giant-cell arteritis. N Engl J Med 2002a;347(4):261–71
dict flares and relapses in patients with giant-cell (temporal) arteri- Salvarani C, Giannini C, Miller DV, et al Giant cell arteritis: involve-
tis. A longitudinal study of 58 biopsy-proven cases. Rheumatology ment of intracranial arteries. Arthritis Rheum 2006;55:985–9
(Oxford) 2000;39(10):1089–94 Salvarani C, Hunder GG. Giant cell arteritis with a low erythrocyte
Lundberg I, Hedfors E. Restricted dose and duration of corticosteroid sedimentation rate: Frequency of occurence in a population-based
treatment in patients with polymyalgia rheumatica and temporal study. Arthritis Rheum 2001;45:140–5
arteritis. J Rheumatol 1990;17(10):1340–5 Salvarani C, Silingardi M, Ghirarduzzi A, et al Is duplex ultrasonogra-
Mahr AD, Jover JA, Spiera RF, et al Adjunctive methotrexate to treat phy useful for the diagnosis of giant-cell arteritis? Ann Intern Med
giant cell arteritis: an individual patient data meta-analysis. Arthritis 2002b;137:232
Rheum 2007;56(8):2789–97 Schmidt WA, Kraft HE, Vorpahl K, Volker L, Gromnica-Ihle EJ. Color
Manna R, Latteri M, Cristiano G, et al Anticardiolipin antibodies in duplex ultrasonography in the diagnosis of temporal arteritis. N
giant cell arteritis and polymyalgia rheumatica: a study of 40 cases. Engl J Med 1997;337(19):1336–42
Br J Rheumatol 1998;37(2):208–10 Seo P, Stone JH. Large-vessel vasculitis. Arthritis Care Res 2002;51:
Meller J, Strutz F, Siefker U, et al Early diagnosis and follow-up of 128–39
aortitis with [(18)F]FDG PET and MRI. Eur J Nucl Med Mol Small P, Brisson M-L. Wegener’s granulomatosis presenting as tempo-
Imaging 2003;30(5):730–6 ral arteritis. Arthritis Rheum 1991;34:220–3
Meyer O, Nicaise P, Moreau S, et al Antibodies to cardiolipin and beta Smetana GW, Shmerling RH. Does this patient have temporal arteritis?
2 glycoprotein I in patients with polymyalgia rheumatica and giant JAMA 2002;287:92–101
cell arteritis. Rev Rhum Engl Ed 1996;63(4):241–7 Spiera RF, Mitnick HJ, Kupersmith M, et al A prospective, double-
Morgan G, Harris E. Non-giant cell temporal arteritis. Arthritis Rheum blind, randomized, placebo controlled trial of methotrexate in the
1978;21:362–6 treatment of giant cell arteritis (GCA). Clin Exp Rheumatol 2001;
Nesher G, Berkun Y, Mates M. Low-dose aspirin and prevention of cra- 19(5):495–501
nial ischemic complications in giant cell arteritis. Arthritis Rheum Stone JH. Hearing voices. Pharos 2006;69:31–2
2004;50:1332–7 Stone JH. Antiplatelet versus anticoagulant therapy in patients with
Nuenninghoff DM, Hunder GG, Christianson TJ, et al Mortality of large- giant cell arteritis: which is best? Nat Clin Pract Rheumatol 2007;3:
artery complication (aortic aneurysm, aortic dissection, and/or large- 136–7
artery stenosis) in patients with giant cell arteritis: a population-based Stone JH, Francis HW. Immune-mediated inner ear disease. Curr Opin
study over 50 years. Arthritis Rheum 2003;48(12):3532–7 Rheumatol 2000;12(1):32–40
Olopade, CO, Sekosan, M, Schraufnagel, DE. Giant cell arteritis mani- Tso E, Flamm SD, White RD, et al Takayasu arteritis: utility and limita-
festing as chronic cough and fever of unknown origin. Mayo Clin tions of magnetic resonance imaging in diagnosis and treatment.
Proc 1997;72:1048 Arthritis Rheum 2002;46(6):1634–42
Papadimitraki ED, Kyrmizakis DE, Kritikos I, Boumpas DT. Ear-nose- Wilkinson IM, Russell RW. Arteries of the head and neck in giant cell
throat manifestations of autoimmune rheumatic diseases. Clin Exp arteritis: a pathological study to show the pattern of arterial involve-
Rheumatol 2004;22(4):485–94 ment. Arch Neurol 1972;27:378
 Takayasu’s Arteritis
28
Eamonn S. Molloy, John H. Stone, Carol A. Langford

Pearl: Both upper back pain or neck pain and tenderness in

Overview of Takayasu’s Arteritis a young woman both suggest TA.
Comment: Just as tearing of the aorta in aortic dissection can
› The aorta and its major branches are the prime disease result in thoracic back pain, inflammation within the aortic
targets in Takayasu’s arteritis (TA). wall can lead to similar, though generally less severe, symp-
› Clinical symptoms of vascular inflammation and vas- toms. In some TA patients, thoracic back pain is the chief
cular insufficiency are usually accompanied or pre- complaint. Pain and tenderness may also be noted over periph-
ceded by a systemic inflammatory process. eral arteries, especially the carotid artery (carotidynia). Most
› TA disease tends to affect women much more often patients also have other, more common manifestations of the
than men (female:male ratio 9:1), and has a predilec- disease (such as fever of unknown origin, loss of a pulse, new
tion for Asian women. However, the disease has been onset hypertension, aortic regurgitation, or claudication in an
described in individuals of all racial and ethnic arm or leg). Still, unfamiliarity with thoracic back pain or
backgrounds. carotidynia as a manifestation of TA can lead to delays in the
› TA is associated with granulomatous inflammation diagnosis.
within the blood vessel wall. The histopathologic fea-
tures of the disease closely resemble those of giant cell Pearl: Digital gangrene is not an expected feature of TA.
arteritis. Comment: Pulseless limbs are common in TA, hence the name
› TA leads to vascular narrowing in most of the blood “pulseless disease,” which was bestowed by Japanese clini-
vessels it affects. However, in the aorta, the disease can cians. The pulselessness of TA classically develops in a “reverse
cause aneurysms. These often lead to aortic regurgita- coarctation” form: the upper extremity pulses are absent, while
tion of the ascending aorta. the lower extremity pulses are present (and even bounding, if
› A lesion of great concern in TA is renal artery stenosis, the disease has led to aortic regurgitation). Coarctation, in con-
which can lead to a rennin-mediated hypertension. If trast, because it usually involves the descending aorta, is asso-
the patient has narrowings in all of the blood vessels to ciated with diminished pulses in the feet. The contrast between
the four extremities, then peripheral blood pressure these two diseases is not always so stark, though: TA also fre-
measurements may not be accurate assessments of the quently involves the large vessels to the lower extremities,
true central aortic pressures. Catheterization may be leading to absent pedal pulses, as well (Fig. 28.1).
required to determine the patient’s blood pressure In any event, despite the frequent occurrence of proximal
accurately. limb arterial occlusion in TA, distal gangrene is extremely
› Eye lesions in TA are of two major types: hypertensive uncommon. The development of this complication should
retinopathy (particularly in the setting of renal artery prompt an evaluation for contributing cofactors, e.g., throm-
stenosis) and proliferative retinopathy (caused by boembolism and hypercoagulable states. The relative immu-
extreme narrowing of the blood vessels supplying the nity from distal gangrene in TA relates to the fact that vascular
head, which leads to neovascularization of the retinal occlusions in this disorder develop insidiously, thereby
vascular supply). Anterior ischemic optic neuropathy, allowing the development of dramatic collateral circulation.
the most common ocular lesion of giant cell arteritis, The spiderweb of collateral vessels in Takayasu’s is gener-
occurs very rarely in TA. ally not associated with a palpable pulse, yet provides suffi-
cient blood flow to the extremities.

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 305

DOI:10.1007/978-1-84800-934-9_28, © Springer Science + Business Media B.V. 2009
306 E. S. Molloy et al.

Fig. 28.1 Involvement of the

femoral artery by Takayasu’s
arteritis (TA). The Figure shows
a long, smooth tapering of the
femoral artery of a 19-year-old
woman with TA. The lesion was
not amenable to angioplasty or
stent placement because of its
length, but responded with the
resolution of leg claudication
when the patient was treated with
glucocorticoids and an exercise
regimen of walking for 1 h each
day. The development of
collateral circulation is assumed
to have mediated the symptom-
atic improvement. (Figure
courtesy of Dr. John Stone)

Pearl: TA is in the differential diagnosis of a young woman festation of that condition. However, a variety of other sys-
presenting with cardiomyopathy. temic inflammatory disorders may be associated with
large-vessel vasculitis. These include giant cell arteritis,
Comment: Although best known for its large vessel involve-
Behçet’s disease, Cogan’s syndrome, relapsing polychondri-
ment, TA can also involve the myocardium with histological
tis, and sarcoidosis (See appropriate chapters). Aortitis may
features of an inflammatory myocarditis. Patients with TA
also be associated with systemic autoimmune diseases such as
can present with variable degrees of dilated cardiomyopathy
systemic lupus erythematosus, rheumatoid arthritis, and
and congestive heart failure. Heart failure may also occur in
spondyloarthropathies.
TA patients as a consequence of uncontrolled hypertension,
Up to 20% of patients with giant cell arteritis have large
severe aortic regurgitation, or coronary artery disease.
artery disease detectable by careful auscultation. Moreover,
Myth: TA is a monophasic illness in the majority of patients. the majority of giant cell arteritis patients have subclinical
Reality: Large series emphasize that TA is a monophasic ill- large artery involvement. Behçet’s disease may involve blood
ness in only a minority of patients (less than 20%) vessels of all types and sizes, and large artery lesions may
(Maksimowicz-McKinnon et al. 2007; Kerr et al. 1994). The lead to bruits in the chest or abdomen; it is associated with
large majority of patients follow a chronic, relapsing course inflammatory venous thromboses and large artery aneu-
over time, with significant accrual of disease- and medication- rysms. Behçet’s disease should particularly be suspected in
related morbidity. young males with features of large-vessel vasculitis. These
differential diagnoses can usually be distinguished from TA
Pearl: Large-vessel vasculitis is associated with a number of on the basis of a thorough history and physical examination.
systemic inflammatory diseases besides TA. Infectious etiologies such as syphilis, tuberculosis, and
Comment: TA is the prototypical large-vessel vasculitis, and mycotic aneurysms must also be considered in the differen-
inflammation within the great vessels is the cardinal mani- tial diagnosis of large-vessel vasculitis.
28 Takayasu’s Arteritis 307

Finally, aortitis that is not related to any known infection

may occur in the absence of disease of the primary aortic
branches and other large vessels. This entity is termed “idio-
pathic aortitis.”
Myth: TA can be distinguished from giant cell arteritis by its
greater tendency to involve the primary branches of the aorta.
Reality: There is much overlap between TA and giant cell
arteritis with regard to vascular distribution. This overlap
renders any conclusions about diagnosis that are based on
the distribution of vascular involvement unreliable. Both TA
and giant cell arteritis can involve the proximal aorta, leading
to dilatation of the aortic root and aortic regurgitation. In
addition, both disorders can involve vessels supplying the
upper extremity involving regions as distal as (and perhaps
even beyond) the axillary-brachial junction.
The same is true with regard to lower extremity vessels.
Fig. 28.2 Fibromuscular dysplasia. This disorder is frequently con-
Both TA and giant cell arteritis can be associated with large- fused clinically with Takayasu’s arteritis, but the “stack of coins” lesion
vessel inflammation and vascular narrowing that extends to evident in this figure is classic for fibromuscular dysplasia. (Figure
the knee and possibly beyond. courtesy of Dr. John Stone)

Pearl: A number of noninflammatory conditions may mimic

large-vessel vasculitis.
features of the disease (Pernicario et al. 1987). Failure to
Comment: Aneurysms of large arteries can occur with genetic include TA in the differential diagnosis may lead to serious
disorders such as Marfan’s syndrome, Ehlers–Danlos syn- delays in diagnosis. However, given the rarity of this associa-
drome (type IV), and Loeys–Dietz syndrome. Stenotic lesions tion, other disorders associated with large-vessel vasculitis
are also evident in some patients with neurofibromatosis, that have prominent skin manifestations, including Behcet’s
Grange syndrome, and congenital disorders such as aortic disease and sarcoidosis, also warrant careful consideration in
coarctation and congenital mid-aortic syndrome. Congenital this setting.
aortic coarctation may be associated with Turner’s syndrome
Pearl: Peripheral blood pressure readings should be inter-
and Williams’ syndrome.
preted with caution in patients with TA.
Fibromuscular dysplasia may cause stenotic and/or aneu-
rysmal lesions, but is distinguished generally by the typical Comment: TA frequently causes bilateral subclavian artery
“stack of coins” appearance on angiography (Fig. 28.2. In stenoses and may also lead to stenosis of the infra-renal aorta
some cases, there may be other clinical features to point or narrowing of the large vessels to both lower extremities.
toward the diagnoses listed earlier. The presence of constitu- Unfortunately, it can also cause renal artery stenosis, leading
tional symptoms, abnormal acute phase reactants, or evidence to renin-mediated hypertension. The combined effects of
of arteritis on pathologic examination of surgical specimens these various stenoses over time are that blood pressure mea-
distinguishes TA from these genetic and congenital disorders. surements in the arms and even the legs do not accurately
However, these features are absent in a significant proportion reflect the central aortic pressure, which may become dan-
of patients with TA. In such cases, consultation with a medi- gerously elevated.
cal geneticist may help resolve the underlying diagnosis. Patients with TA and symptoms or signs referable to
severe hypertension (e.g., intractable headaches) require a
Myth: TA does not involve the skin.
serum renin measurement, a thorough investigation of the
Reality: This Myth illustrates the fallacy of placing too much circulation to all extremities, precise imaging of the renal
emphasis on classification schemes that focus on the size of arteries, and an accurate determination of the central aortic
the blood vessels typically affected by a given disorder. pressure. This usually means a catheter-directed angiogram
Although such systems are useful, they must be regarded as and measurement of central aortic pressures directly.
guidelines rather than rules.
Myth: Normal acute phase reactants effectively exclude the
TA is the prototypical large vessel vasculitis, and the cuta-
diagnosis of TA.
neous blood supply is generally confined to vessels smaller
than 150 mm. Nevertheless, nodular and ulcerative skin lesions Reality: Up to 11% of TA patients have normal acute phase
can occur in patients with TA and may be the presenting reactants at their initial diagnosis (Maksimowicz-McKinnon
308 E. S. Molloy et al.

et al. 2007; Kerr et al. 1994). Furthermore, approximately MRA can detect anatomic abnormalities such as luminal
25% of patients have normal acute phase reactants during stenosis, aneurysm formation, vessel wall thickening, and
periods of overtly active disease. Therefore, the erythrocyte collateral vessel formation (Tso et al. 2002). MRA can also
sedimentation rate and C-reactive protein level cannot be detect signal changes or brightness in the vessel wall, which
relied upon to exclude the diagnosis of TA. is purported to represent neovascularization of the vasa vaso-
rum, edema, or inflammation. The precise significance of
Pearl: TA can be associated with a markedly abnormal acute
these findings in the vessel wall and their correlation with
phase response.
disease activity in TA patients has yet to be defined.
Comment: A young woman with a markedly abnormal acute Positron emission tomography (PET) scanning is a promis-
phase response, for example, thrombocytosis, a hematocrit ing modality for the assessment of disease activity in TA (Gotway
compatible with anemia of chronic disease, and a bone mar- et al. 2005; Moreno et al. 2005). However, the performance char-
row examination that shows “reactive hyperplasia,” should acteristics of this technique in TA are still being defined. PET
be evaluated for TA. Platelet counts greater than 1 million/ also requires concomitant performance of computed tomog-
mm3 have been observed in TA, and the erythrocyte sedi- raphy imaging to detect vascular stenoses and aneurysms.
mentation rate may exceed 100 mm/h. In short, although imaging modalities such as MRA, CTA,
and PET now play a major role in establishing the diagnosis
Myth: The ESR and CRP are accurate gauges of disease
of large-vessel vasculitis, much remains to be learned about
activity in TA.
the utility and pitfalls of these techniques in gauging disease
Reality: Assessing disease activity is one of the greatest chal- activity.
lenges in caring for patients with TA. Acute phase reactants
Pearl: The finding of one vascular lesion in a patient with
correlate with disease activity in only 70% of patients
suspected large vessel vasculitis mandates screening for
(Maksimowicz-McKinnon et al. 2007; Kerr et al. 1994).
additional asymptomatic lesions.
Furthermore, longitudinal series of treated TA patients under-
going vascular surgery indicate that active arteritis is found Comment: Lesions in TA may be asymptomatic, especially
in up to 44% of patients, often in the absence of systemic when not associated with hemodynamically significant
symptoms or elevations of acute phase reactants and, in some stenosis or occlusion of the vessel or when associated with
cases, without abnormal findings on vascular imaging stud- aneurysm. Imaging of the entire aorta and major branch ves-
ies (Hoffman 1997; Kerr et al. 1994; Lagneau et al. 1987). sels should be performed in all patients with suspected large
vessel vasculitis. Identification of additional vascular lesions
Myth: The angiogram is always abnormal in TA.
upon the initial assessment of patients with suspected large
Reality: Catheter-directed angiography has been a key tool for vessel vasculitis may help establish the diagnosis of TA. A
diagnosis of TA and assessment of disease activity. However, comprehensive noninvasive imaging evaluation is a key com-
because angiography only detects luminal abnormalities ponent of the baseline patient evaluation. Without a full
(stenoses, dilatations), which may be absent in the early stages assessment of the extent of disease at baseline, any abnor-
of disease, patients with TA may have a normal angiogram. mality discovered subsequently may be difficult to interpret.
Although not invariably abnormal in TA, catheter-directed
Myth: Radiographic worsening of a preexisting stenotic or
angiography remains important for certain patients even if the
aneurysmal lesion in TA indicates active disease and war-
diagnosis has been established by magnetic resonance angiog-
rants intensification of immunosuppressive therapy.
raphy or computed tomography angiography, as it offers the
opportunity to measure directly the central aortic pressures. Reality: Preexisting stenotic or aneurysmal lesions are not
generally considered to be reversible, even if immunosup-
Pearl: Noninvasive imaging techniques can detect evidence
pression is effective. Moreover, such lesions may worsen
of large vessel vasculitis, but their role in monitoring disease
because of hemodynamic factors even if the disease is con-
activity remains poorly defined.
trolled effectively. Turbulent flow due to preexisting stenosis
Comment: While catheter-directed angiography has tradi- may lead to a greater degree of stenosis or occlusion as a
tionally been the gold standard for vessel imaging in TA result of secondary atherosclerosis, especially in the presence
patients, it is impractical for repeated monitoring because of of cofactors such as hypertension, dyslipidemia, and smok-
the associated risks, including hemorrhage, arterial injury, ing. Worsening stenosis may also occur as a result of tighten-
infection, contrast-induced nephropathy, and radiation expo- ing of scar tissue within the vessel wall over time. Aneurysms
sure. Consequently, it is increasingly being supplanted by may enlarge or rupture because of mechanical factors, par-
noninvasive modalities, particularly magnetic resonance ticularly hypertension. Therefore, only a new lesion in a pre-
angiography (MRA) (Atalay and Bluemky 2001; Flamm viously unaffected vascular territory should be interpreted as
et al. 1998) and computed tomography angiography (CTA). unequivocal evidence of active arteritis.
28 Takayasu’s Arteritis 309

Myth: Prednisone monotherapy is sufficient for the treatment be candidates for bypass procedures. In such cases, the sub-
of TA in the large majority of patients throughout the course clavian or innominate arteries should not serve as the origin
of their illness. of these grafts because of the high frequency with which
these vessels are affected in TA.
Reality: Prednisone remains the foundation of treatment for
The subclavian and innominate vessels become stenotic
TA. Although other therapies are often used to treat this dis-
in more than 90% of patients over time, threatening the util-
ease, no randomized clinical trial of any TA treatment has
ity of the bypasses that originate from these vessels. In con-
ever been published.
trast, even if the aortic arch were involved in TA, occlusive
Prednisone monotherapy is usually the initial treatment of
disease at that site is almost unknown (Giordano et al. 1991;
choice for TA patients who do not exhibit severe manifesta-
Kerr et al. 1994). In addition, there is a much higher flow rate
tions. However, most patients are unable to taper glucocorti-
in the aorta, which also lessens the risk of occlusion.
coids entirely without suffering disease relapses. Consequently,
additional therapies must be considered in order to spare Myth: Patients with the unexpected finding of an inflamma-
excessive toxicity from high-dose glucocorticoids. tory infiltrate within the aortic root at surgery require treat-
Open-label studies have suggested potential benefit with ment with large doses of glucocorticoids.
methotrexate, anti-TNF therapy, and a handful of other immu-
Reality: The unexpected histologic finding of aortic root inflam-
nosuppressive agents (Kerr et al 1994; Hoffman et al. 1994;
mation at the time of aortic surgery is more common than gen-
Liang and Hoffman 2005; Maksimowicz-McKinnon et al.
erally appreciated. In one study of 1,204 patients who underwent
2007; Molloy et al. 2008). It must be emphasized that all these
aortic surgery at one institution, 52 of the patients (4%) were
studies involve small numbers of patients, and none have been
found to have idiopathic aortitis (Rojo-Leyva et al. 2000).
randomized, controlled trials. As a general rule, cyclophosph-
Among those 52 patients, 51 did not have current features of a
amide should be avoided in TA unless there are critical lesions
systemic illness at the time of their surgery. Despite the absence
that affect the coronary arteries or multiple cerebral blood
of systemic symptoms and signs in patients with idiopathic aor-
vessels. A randomized controlled trial of abatacept for the
titis, the degree of inflammatory change within the aortic wall is
treatment of TA began its enrollment in the late 2008.
often striking. In 16 of the 52 patients (31%), aortitis was associ-
Pearl: Many attempts at revascularization in large-vessel ated with a remote history of vasculitis or other systemic disor-
vasculitis are unnecessary or even ill-advised. der that was not considered to be active at the time of surgery.
This observation has profound implications on our percep-
Comment: In TA, everything narrowed does not need to be tion and assessment of apparent “remission” in patients with
“fixed.” The risks and benefits of potential interventions large-vessel vasculitis. Among the 52 patients with idiopathic
should be considered carefully and with the full knowledge aortitis, 36 were followed for a mean period of 41 months,
of the extensive collateral blood vessel involvement that over which time new aneurysms were identified in only 6
characterizes this disease. patients (16% of those followed longitudinally). This experi-
Stenotic lesions in large blood vessels such as the subcla- ence suggests that most patients with isolated focal aortitis
vian artery require many months or years to form. The slow identified at aortic surgery have no indication for immuno-
development of these lesions permits dramatic neovascular- suppressive therapy. However, such patients require careful
ization to occur naturally. The network of collateral blood imaging of the entire aorta and its primary branches to ensure
vessels that form in large-vessel vasculitis are not usually that disease was unifocal, and careful follow-up to detect the
associated with palpable pulses at the brachial or radial sites; development of other features of large vessel vasculitis.
hence, the term “pulseless disease.” However, this collateral
circulation is sufficient to provide blood flow to the extremi- Pearl: IgG4-related systemic disease can involve the aorta.
ties that prevents ischemia quite effectively. Comment: IgG4-related systemic disease is an emerging dis-
Critical stenoses to vital organs in which the collateral order associated with lymphoplasmacytic infiltration into a
circulation is poor may be candidates for revascularization variety of organs, plasma cells that stain intensely for IgG4,
with bypass surgery, angioplasty, or stent placement. The and elevated serum IgG4 levels (Terumi and Atsutake; Stone
limited outcomes data available for these procedures suggest et al., 2009) (See Chap. 12). IgG4-related systemic disease,
that the failure rates are high. first described in the pancreas and termed “autoimmune pan-
Pearl: In TA, carotid bypass grafts for central nervous sys- creatitis” (Yoshida et al. 1995), is now known to involve the
tem symptoms should originate from the aortic root or arch. biliary tree, salivary glands, kidneys, lungs, and other organs.
Cases of aortitis caused by IgG4-related systemic disease
Comment: Stenoses of the carotid or vertebral artery occur in have been reported (Fig. 28.3) (Satomi et al. 2008). The mea-
about 60% of all patients with TA. Depending on the severity surement of serum IgG4 levels is an appropriate screening
of symptoms and anatomic abnormalities, these patients may test if this condition is suspected.
310 E. S. Molloy et al.

Atalay M, Bluemke D. Magnetic resonance imaging of large vessel vas-

culitis. Curr Opin Rheum. 2001;13:41–47
Flamm S, White R, Hoffman G. The clinical application of “edema-
weighted” magnetic resonance imaging in Takayasu’s arteritis. Int J
Cardiol. 1998;66:151–9
Gotway MB, Araoz PA, Macedo TA, et al Imaging findings in
Takayasu’s arteritis. Am J Roentgenol. 2005;184:1945–50
Giordano J, Leavitt R, Hoffman G, et al Experience with surgical treat-
ment for Takayasu’s disease. Surgery 1991;109:252–8
Hoffman G. Takayasu’s arteritis: lessons from the American National
Institutes of Health experience. Int J Cardiol. 1997;54:S99–102
Hoffman GS, Leavitt RY, Kerr GS, et al Treatment of glucocorticoid-
resistant or relapsing Takayasu arteritis with methotrexate. Arthritis
Rheum. 1994;37:578–82
Hoffman GS, Merkel PA, Brasington RD, et al Anti-tumor necrosis fac-
tor therapy in patients with difficult to treat Takayasu arteritis.
Arthritis Rheum. 2004b;50:2296–304
Kerr GS, Hallahan CW, Giordano J, et al Takayasu arteritis. Ann Intern
Med. 1994;20:919–29
Lagneau P, Michel J-B, Vuong P. Surgical treatment of Takayasu’s dis-
ease. Ann Surg. 1987;205:157–66
Fig. 28.3 IgG4-related systemic disease involving the aorta. This Liang P, Hoffman GS. Advances in the medical and surgical treatment
patient suffered an aortic dissection, and the diagnosis of IgG4-related of Takayasu’s arteritis. Curr Opin Rheumatol. 2005;17:16–24
systemic disease was made upon histopathological examination of the Maksimowicz-McKinnon K, Clark TM, Hoffman GS. Takayasu’s
replaced aorta. (Figure courtesy of Dr. Arezou Khosroshahi) arteritis: limitations of therapy and guarded prognosis in an
American cohort. Arthritis Rheum. 2007;56:1000–9
Matsumura A, Moriwaki R, Numano F. Pregnancy in Takayasu arteritis
from the view of internal medicine. Heart Vessels. 1992;7(S):120–4
Myth: Pregnancy is contraindicated in patients with TA. Molloy ES, Langford CA, Clark TM, et al Durable remission in patients
with refractory Takayasu’s arteritis treated with infliximab and etan-
Reality: The striking (9:1) female predominance in TA and ercept. Ann Rheum Dis. 2008;67:1567–9
the inference that the increased estrogen concentrations of Moreno D, Yuste JR, Rodriguez M, et al Positron emission tomography
pregnancy are likely to aggravate the illness have raised con- use in the diagnosis and follow up of Takayasu’s arteritis. Ann
Rheum Dis. 2005;64:1091–3
cerns about the safety of pregnancy in such patients. However,
Pernicario CV, Winkelmann RK, Hunder GG. Cutaneous manifestations
the data available to address this question do not suggest that of Takayasu’s arteritis. J Am Acad Dermatol. 1987;17:998–1005
pregnancy exacerbates TA. Rojo-Leyva F, Ratliff N, Cosgrove D, et al Study of 52 patients with
If complications of TA already include uncontrolled idiopathic aortitis from a cohort of 1,204 surgical cases. Arthritis
Rheum. 2000;43:901–7
hypertension, renal insufficiency, or aortic regurgitation,
Satomi K, Yoh Z, Atsuhiro K, et al Inflammatory abdominal aortic
pregnancy-associated increases in intravascular volume may aneurysm: close relationship to IgG4-related periaortitis. Am J Surg
make pregnancy risky and unadvised. Several small series Pathol. 2008;32:197–204
have demonstrated that in the absence of these risk factors, Sharma B, Jain S, Vasishta K. Outcome of pregnancy in Takayasu
arteritis. Int J Cardiol. 2000;75:S159–62
however, patients with TA may have successful pregnancies,
Stone JH, Khosroshahi A, Hilgenberg AD et al IgG4-related systemic
uncomplicated deliveries, and healthy children (Matsumura disease and lymphoplasmacytic aortitis. Arthritis Rheum 2009 (in
et al. 1992; Aso et al. 1992; Sharma et al. 2000). press)
Terumi K, Atsutake O. IgG4-related sclerosing disease. World J
Gastroenterol. 2008;14:3948–55
Tso E, Flamm SD, White RD, et al Takayasu arteritis: utility and limita-
References tions of magnetic resonance imaging in diagnosis and treatment.
Arthritis Rheum. 2002;46:1634–42
Yoshida K, Toki F, Takeuchi T, et al Chronic pancreatitis caused by an
Aso T, Abe S, Yaguchi T. Clinical gynecologic features of pregnancy in autoimmune abnormality. Proposal of the concept of autoimmune
Takayasu’s arteritis. Heart Vessels. 1992;7:S125–132 pancreatitis. Dig Dis Sci. 1995;40:1561–8
 Primary Angiitis of the Central Nervous System
and Reversible Cerebral Vasoconstriction 29
Syndromes

Leonard H. Calabrese and Aneesh B. Singhal

Comment: A lumbar puncture is essential for the diagnostic

29.1 Overview of Primary Angiitis evaluation of patients with potential PACNS. The cerebro-
of the Central Nervous System spinal fluid (CSF) must be examined for signs of inflamma-
tion, infection, and malignancy.
› Vasculitis that exclusively involves the central nervous The CSF is abnormal in more than 95% of cases of granu-
system (CNS) is among the most complex and poorly lomatous angiitis of the CNS (GACNS). Thus, a normal CSF
understood forms of vascular inflammatory disease. analysis has a high negative predictive value for this condi-
› The variable terminology for CNS vasculitis continues tion. The mean CSF protein in patients with PACNS is 177 mg/
to pose significant challenges. CNS vasculitis that is dL, and the mean cell count is 77/mm3 (Calabrese et al. 1997).
not associated with a known underlying cause or some An equally compelling reason to analyze CSF is to exclude
form of systemic vasculitis is termed primary angiitis mimics such as infection and malignancy, which can yield
of the CNS (PACNS). angiographic findings identical to those of PACNS.
› Some cases of PACNS confirmed by histopathology
Pearl: PACNS almost never presents as a simple stroke.
have granulomatous features of inflammation. The
term “granulomatous inflammation of the nervous sys- Comment: Although CNS vasculitis remains in the differen-
tem” has been used for such cases. For practical pur- tial diagnosis of an unexplained ischemic event, patients with
poses, now GANS and PACNS are considered to be this rare disorder almost never present with a simple stroke.
largely synonymous terms. In patients with GACNS as well as other pathologically docu-
› PACNS presents in a subacute manner, with an elapse mented cases, the usual clinical manifestations include head-
of many months between the time of first symptom aches, mental changes, and multiple small-vessel strokes that
onset and diagnostic confirmation. occur over weeks to many months (Moore 1998). Additional
› The clinical hallmarks of PACNS are headache, slowly- clinical settings that raise the likelihood of PACNS include
evolving encephalopathy, and multi-focal strokes. chronic meningitis and multifocal strokes that occur in a sub-
Fever, other constitutional symptoms, and extra-CNS acute fashion (see Fig. 29.1a–c).
manifestations are absent.
Pearl: PACNS can present as a mass lesion.
› Acute phase reactants tend to be normal in PACNS.
› Lumbar puncture typically reveals a lymphocytic Comment: Patients presenting with a mass lesion on a neu-
pleocytosis. roimaging study, regardless of the associated signs or symp-
› Magnetic resonance imaging (MRI) usually reveals toms, must be considered to have a malignancy or an infection
multiple foci of strokes, some of which may be asymp- until proven otherwise. However, approximately 5% of patients
tomatic. Imaging findings typical of PACNS are shown with PACNS also present with mass lesions (see Fig. 29.1f).
in Fig. 29.1. CNS angiography is normal in many Biopsy confirmation is always required.
patients with PACNS, but when abnormal, can reveal The optimal management of this disorder has not been
arterial beading (see Fig. 29.1d, e), focal narrowing, defined clearly, but CNS vasculitis that presents as a mass
apparent vascular cutoffs, and even microaneurysms. lesion occasionally appears to have been cured by surgical
resection. Other cases have proved to require immunosup-
pressive therapy (Molloy et al. 2008).
Pearl: A thorough evaluation of a patient for possible PACNS Pearl: PACNS is usually not associated with constitutional
always must include a lumbar puncture. symptoms or signs of systemic inflammation.

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 311

DOI:10.1007/978-1-84800-934-9_29, © Springer Science + Business Media B.V. 2009
312 L. H. Calabrese and A. B. Singhal

Fig. 29.1 Brain imaging in primary angiitis of the CNS (PACNS). the same patient show areas of irregularity, “beading,” and ectasia in
Brain MRI, axial FLAIR images (a–c) from a 21-year man with PACNS branches of the left (d) and right (e) middle cerebral arteries. Brain
show hyperintense lesions (subacute and chronic ischemic strokes) in MRI, axial FLAIR image (f) from a 26-year man shows a mass lesion
multiple deep locations. Digital subtraction angiographic images from that proved to be PACNS on brain biopsy

Comment: This concept flies in the face of most clinicians’ pre- intensive immunosuppression. For these reasons, diagnostic
conceived notions of vasculitis and frequently confounds the certainty is invaluable whenever it can be established. Further-
diagnostic evaluation of PACNS. However, it is quite true. The more, numerous PACNS mimickers, including infections and
complete blood count is usually normal in PACNS, as are acute malignancies, can be diagnosed only by obtaining histopathol-
phase reactants such as the erythrocyte sedimentation rate and ogy. Although the sensitivity of tissue biopsy is only in the
C-reactive protein. When the diagnosis is suspected on clinical range of 75–80%, the morbidity of brain biopsy is less than 2%
neurologic grounds, the clinician should not be dissuaded by when biopsies are performed under appropriate circumstances
the absence of features normally associated with systemic in the setting of nonmalignant conditions (Parisi and Moore
vasculitis. The lack of symptoms and signs to suggest an 1994). The optimal approach to brain biopsy is to go directly
inflammatory process is actually a point in favor of PACNS. to accessible lesions by a stereotactic technique. In the absence
of a focal lesion, an open biopsy of the tip of the nondominant
Myth: Biopsy of the CNS or brain biopsy is not worth the
temporal lobe with overlying leptomeningies is reasonable.
risk.
Reality: This is probably one of the greatest myths and pitfalls Myth: The high rate of false-negative results on brain biopsy
in the diagnosis of PACNS. Pathologically-documented even in patients who really have PACNS makes the procedure
PACNS has serious prognostic implications and requires relatively useless.
29 Primary Angiitis of the Central Nervous System and Reversible Cerebral Vasoconstriction Syndromes 313

Reality: The most reliable data available suggest that the sen- Table 29.1 RCVS: key elements for diagnosis
sitivity of brain biopsy in PACNS is as low as 75%. Thus, in Digital subtraction angiography or a CT or MR angiogram that
the presence of PACNS, the procedure will miss the patho- documents multifocal cerebral artery vasoconstriction
logical area of tissue in up to a quarter of cases. It is regret- No evidence for ruptured brain aneurysm or ruptured cerebral
vascular malformation
table that this fact often discourages clinicians from
Normal or near-normal cerebrospinal fluid analysis (protein <
performing brain biopsies in cases in which PACNS is sus- 80 mg/dL, white blood cells < 10/mm3)
pected. However, an equally important justification for the Onset with severe headache, with or without additional neurologic
performance of brain biopsy is the efficiency of this proce- signs or symptoms
dure in excluding nonvasculitic conditions. The diagnosis cannot be confirmed until (a) reversibility of the
angiographic abnormalities is documented, usually within 12
In a series of 30 consecutive brain biopsies from the Duke weeks after onset; or (b) if death occurs before the follow-up
University Medical Center, the predictive value of brain neurovascular imaging is obtained, autopsy rules out conditions
biopsy was greater than 90% in patients suspected of PACNS such as cerebral vasculitis, intracranial atherosclerosis, and
(Chu et al. 1998). In a series of 61 consecutive patients from aneurysmal subarachnoid hemorrhage that can also manifest
with headache, stroke, and cerebral angiographic abnormalities.
the University of Michigan, brain biopsy suggested alterna-
Adapted from Calabrese et al. (2007)
tive diagnoses in 22 cases (36%) (Alrawi et al. 1999). In our
estimation, brain biopsy is underutilized in patients for whom
RCVS is usually benign. Nearly 90% of patients have few
PACNS is a true possibility.
or no residual neurological deficits (Hajj-Ali et al., 2002).
Pearl: Specific clinical features can distinguish RCVS from
PACNS.
29.2 Overview of Reversible Cerebral
Vasoconstriction Syndromes Comment: The confusion between RCVS and PACNS is
attributable to overlapping clinical-angiographic features,
low specificity of diagnostic tests, and lack of validated diag-
• Reversible cerebral vasoconstriction syndromes (RCVS)
nostic criteria for either condition. Nevertheless, these disor-
include a diverse array of conditions, such as eclampsia, the
ders have major differences in their clinical presentations,
postpartum state, hypertensive encephalopathy, migraine,
and clinicians ignore them at their peril.
and exposure to vasoconstrictive drugs.
Patients with PACNS usually come to medical attention
• The angiographic phenomenon of reversible “beading” has
because of smoldering, low-grade headaches, subacute
been reported using a variety of terms, each reflecting a pre-
encephalopathy, or progressive neurological deficits. In con-
sumed mechanism or associated underlying condition:
trast, the onset of RCVS is typically dramatic. RCVS patients
migrainous vasospasm, migraine angiitis, thunderclap
report the explosive onset of the worst headache of their
headache with reversible vasospasm, drug-induced vasos-
lives. Most patients have three to five headaches over a period
pasm, drug-induced arteritis, eclamptic vasospasm, post-
of days to a couple of weeks. These headaches are exacer-
partum angiopathy, benign angiopathy (Hajj-Ali et al.,
bated by the Valsalva maneuver (Ducros et al. 2007). A sig-
2002), cerebral pseudovasculitis, and Call’s syndrome.
nificant minority of patients develop generalized seizures or
• Patients with reversible cerebral arterial narrowing appear
focal neurological deficits.
to have similar clinical, laboratory, imaging, and prognos-
Cortical blindness (Balint syndrome) and visual scotomas
tic features, regardless of the associated underlying
occur in approximately 30% of patients. Bilateral hyperre-
condition.
flexia is common. The headaches resolve spontaneously over
• Provisional diagnostic criteria for RCVS have been pro-
days to weeks, and unlike PACNS, most patients recover
posed (Table 29.1) (Calabrese et al. 2007).
completely within weeks.
• Patients with RCVS are typically young (age 20–60 years).
Two critical clues in the presentation distinguish RCVS
More than 80% of patients present with “thunderclap”
from PACNS with great reliability. First, RCVS begin quickly
headaches, which occur suddenly, have great intensity,
(explosively) with the onset of a thunderclap headache, in
and can recur over the next few days (Chen et al. 2006).
contrast to the subacute development characteristic of
• Imaging features typical of RCVS are shown in Fig. 29.2.
PACNS. Second, the clinical setting of RCVS is highly typi-
with RCVS, by definition, have segmental cerebral arterial
cal: onset after childbirth or after exposure to vasoconstric-
vasoconstriction that resolves over a period of weeks.
tive drugs such as sumatriptan, nasal decongestants, certain
• RCVS has many features that overlap with PACNS: head-
antidepressants, and illicit drugs of abuse such as cocaine or
ache, potential for strokes and seizures, and “beading” of
ecstasy favors the diagnosis of RCVS.
the cerebral arteries on angiographic studies.
• Despite the ominous presentation and impressive angio- Pearl: Cross-sectional neuroimaging studies help distinguish
graphic and imaging abnormalities, the prognosis of RCVS from PACNS.
314 L. H. Calabrese and A. B. Singhal

Comment: Computed tomography (CT) or MRI of the brain RCVS patients who do have brain imaging abnormalities
should be performed immediately in all patients with “thun- usually have a combination of parenchymal hemorrhage (see
derclap” headaches. These studies are essential for the exclu- Fig. 29.2b), small subarachnoid hemorrhages on the cortical
sion of explanations such as cerebral venous sinus thrombosis, surface (see Fig. 29.2d), brain infarctions, and transient
pituitary apoplexy, carotid or vertebral artery dissection, and vasogenic brain edema. RCVS-associated brain infarctions
CNS hemorrhage (Schwedt et al. 2006). Once these condi- tend to be bilateral, symmetric, and located in cortical-sub-
tions have been excluded, consideration can be given to cortical regions. They are surrounded by areas of cerebral
RCVS and PACNS as potential diagnoses. hypoperfusion. Brain edema (see Fig. 29.2c) is similarly
Analysis of the type and distribution of brain lesions on bilateral, symmetric, and located in the cerebral or cerebellar
neuroimaging studies is helpful. Patients with PACNS virtu- hemispheres. The edema associated with RCVS is similar to
ally always have small, disseminated, punctuate infarcts that that of the edematous brain lesions found in the reversible
are of variable age (acute, subacute, and even chronic). posterior leukoencephalopathy syndrome (RPLS), also
Diffused white matter lesions are also often present, suggest- known as the posterior reversible encephalopathy syndrome
ing a chronic microangiopathy. Brain hemorrhages seldom (PRES) (Singhal 2004).
occur in bona fide cases of PACNS.
On the other hand, approximately 70% of patients with Myth: RCVS and PACNS can be differentiated by their angio-
RCVS have no parenchymal brain lesions on CNS imaging. graphic features if classic findings of PACNS are present.

Fig. 29.2 Brain imaging in the reversible cerebral vasoconstriction hemisphere (d), and dilated cortical surface vessels (arrows) overlying
syndrome (RCVS). Brain MRI, axial FLAIR images (a–e) show the the left hemisphere (e). However, it is important to note that most
typical brain imaging abnormalities in RCVS, including an ischemic patients with RCVS have no parenchymal lesion on brain MRI. Digital
stroke in the watershed distribution of the left middle and posterior cere- subtraction angiogram (f) from a patient with RCVS shows the typical
bral artery (a), parenchymal brain hemorrhage (b), reversible hyperin- segmental narrowing (“sausaging”) of multiple cerebral arteries that
tense lesions (brain edema) in subcortical regions (c), non-aneurysmal usually reverse within weeks
cortical surface subarachnoid hemorrhage (arrow) overlying the left
29 Primary Angiitis of the Central Nervous System and Reversible Cerebral Vasoconstriction Syndromes 315

Reality: The single greatest pitfall in evaluating patients with Moreover, the presence of vasoconstriction at the onset of
possible CNS vasculitis is over-reliance upon cerebral arteriog- RCVS is contrary to what is observed normally in subarach-
raphy as a diagnostic test. There is not a single angiographic noid hemorrhage, a setting in which vasospasm ensues after
feature that is pathognomonic for PACNS; RCVS can mimic a period of 5–10 days.
every last one. Thus, a CNS angiogram read as “consistent Despite the important differences between the vasocon-
with vasculitis” (see Fig. 29.2f) merely indicates that insightful striction of RCVS and subarachnoid hemorrhage, many
clinical correlation is required and additional work-up (e.g., patients with RCVS are subjected to serial angiograms
lumbar puncture and an MRI) are probably necessary. designed to detect occult sources of CNS bleeding or exclude
Although the cerebral angiographic abnormalities can be unruptured aneurysms as the cause of their recurrent thun-
identical in RCVS and PACNS, the evolution over time of derclap headaches. Such serial studies are unnecessary. The
angiographic lesions is very helpful in distinguishing these role of follow-up angiography in RCVS is to confirm the
two entities. Cerebral angiography is abnormal by definition diagnosis by documenting the resolution of vasospasm 4–6
in RCVS, but shows reversibility within 10–12 weeks in weeks after the initial study. In practice, few patients undergo
nearly all cases. It must be acknowledged that in a minority such second studies.
of RCVS cases, the intense vasoconstriction characteristic of Myth: RCVS is simply a bad migraine attack.
RCVS invokes a secondary inflammatory response that leads
to some permanent changes on angiography. Reality: Patients with RCVS present with headaches and 25%
Clinicians must bear in mind that PACNS patients can have have histories of migraine headaches. A significant proportion
normal cerebral angiograms. This is because the caliber of the of RCVS patients develop “migraine-like strokes” in water-
arteries affected is sometimes below the resolution of angiog- shed arterial territories (see Fig. 29.2a). These features, com-
raphy. Arterial irregularities, if present, seldom get reversed in bined with the rare observation of angiographic vasoconstriction
PACNS without chronic immunosuppressive treatment. during acute migraine attacks, have made some authors ques-
tion whether RCVS is simply a severe migraine attack.
Myth: Blood and CSF test results generally distinguish RCVS A true relationship between RCVS and migraines is
from PACNS quite reliably. unlikely, for several reasons. The character of thunderclap
headache is quite different from migraines. In addition, RCVS
Reality: Neither RCVS nor PACNS has a characteristic sero-
rarely recur, in contrast to migraines. However, the spectrum
logic marker of the disease. Peripheral blood markers of
of RCVS probably does include primary headache disorders
inflammation such as the erythrocyte sedimentation rate and
such as benign sexual, cough, and exertional headaches, all of
C-reactive protein are usually normal in both conditions, but
which have been associated with cerebral vasoconstriction. In
are helpful (if markedly abnormal) in suggesting diagnoses
a prospective study, MR angiography showed beading of the
other than RCVS or PACNS. Perhaps, the only clue to a
cerebral arteries in 39% of patients with unexplained thun-
cause of RCVS is the finding of elevated norepinephrine
derclap headache. The clinical features of those with and
metabolites in the blood and urine of patients with cerebral
without vasoconstriction were similar (chen et al., 2006).
vasoconstriction associated with a pheochromocytoma.
CSF analysis is more helpful but remains imperfect. A Pearl: Transcranial Doppler studies are useful in RCVS.
“benign” CSF favors the diagnosis of RCVS. However, the Comment: Conventional angiography is the modality of
CSF is normal in at least 20% of patients with PACNS and also choice for making the diagnosis of RCVS. MR and CT angio-
has mild abnormalities in up to 20% of those with RCVS. graphic studies are also helpful if the blood vessels involved
are not below the resolution of these instruments (Remember,
Myth: Angiographic “beading” in RCVS results from suba-
even conventional angiography is not sufficiently sensitive to
rachnoid blood or transient inflammation.
detect disease in the smallest blood vessels). Transcranial
Reality: The precise mechanism of vasoconstriction in RCVS Doppler studies, noninvasive investigations that measure
remains unknown. At the molecular level, the pathophysiol- blood flow velocities in multiple intracranial arteries, can
ogy may be related to the immunological and biochemical also provide support for the diagnosis of RCVS.
factors such as endothelin-1, serotonin, nitric oxide, and Transcranial Doppler studies have potential shortcomings.
prostaglandins that are known to be involved in subarach- Hyperemia and other factors can raise blood flow velocities,
noid hemorrhage-related vasospasm. One-third of patients leading to false impressions of vessel narrowing. In addition,
with RCVS have minor subarachnoid hemorrhages localized some patients with vasoconstriction demonstrated on angiog-
to the cortical surface, raising the question of hemorrhage- raphy have normal blood flow velocities. Transcranial Doppler
induced cerebral vasospasm, similar to that which occurs in studies are probably most useful when employed in a serial
patients following the rupture of berry aneurysms. However, fashion in individual patients. These procedures permit simple,
the widespread distribution of vasoconstriction in RCVS noninvasive, contrast-free, bedside monitoring of the progres-
cannot be explained by the small surface hemorrhages. sion and resolution of vasoconstriction (Chen et al. 2008).
316 L. H. Calabrese and A. B. Singhal

Pearl: Vasoconstrictive medications should be avoided in cerebral vasculature, the widespread use and abuse of vaso-
patients with RCVS. constrictive drugs and medications, and use of the term
“RCVS,” which has brought together a group of related condi-
Comment: Numerous vasoconstrictive drugs and medica-
tions that previously seemed disparate.
tions such as amphetamine derivatives, sumatriptan, ergot
The recognition that cerebral vasoconstriction is often pres-
derivatives, diet pills, nasal decongestants, serotonergic anti-
ent in isolated thunderclap headache and RPLS has increased
depressants, cocaine, ecstasy, and marijuana have been iden-
the use of neurovascular testing in these conditions. It is likely
tified as triggers of RCVS (Ducros et al. 2007; Singhal 2002;
that primary care physicians, rheumatologists, obstetricians,
Singhal et al. 2002). These data are anecdotal. Confirmatory
neurologists, neurosurgeons, intensivists, emergency depart-
prospective studies are warranted, but avoidance of further
ment physicians, and other specialists will encounter and
exposure to such agents in patients with evidence for cere-
report more cases of RCVS in the years ahead.
bral vasoconstriction is reasonable.
This issue is particularly relevant because the majority of
patients present with the sudden onset of intense headaches
and often receive medications such as sumatriptan before
undergoing neurovascular imaging tests. References
Pearl: Pharmacological blood pressure modulation should
Alrawi A, Trobe JD, Blaivas M, Musch DC. Brain biopsy in primary
be avoided in patients with RCVS. Simple observation is an angiitis of the central nervous system. Neurology 1999;53:858–60
appropriate management option. Calabrese LH, Dodick DW, Schwedt TJ, Singhal AB. Narrative review:
reversible cerebral vasoconstriction syndromes. Ann Intern Med.
Comment: It is tempting to use pharmacological maneuvers 2007;146:34–44
to raise blood pressure so as to improve cerebral perfusion, Calabrese LH, Duna GF, Lie JT. Vasculitis in the central nervous sys-
or to lower the blood pressure in hypertension in the interest tem. Arthritis Rheum. 1997;40:1189–201
Chen SP, Fuh JL, Chang FC, Lirng JF, Shia BC, Wang SJ. Transcranial
of minimizing brain hemorrhage. Clinicians should under-
color Doppler study for reversible cerebral vasoconstriction syn-
stand, however, that blood pressure fluctuations in RCVS dromes. Ann Neurol. 2008;63:751–7
may represent autoregulatory phenomena. Therefore, blood Chen SP, Fuh JL, Lirng JF, Chang FC, Wang SJ. Recurrent primary
pressure modulation should be avoided. Drug-induced hyper- thunderclap headache and benign CNS angiopathy: spectra of the
same disorder? Neurology 2006;67:2164–9
tension can exacerbate vasoconstriction, and relative
Chu CT, Gray L, Goldstein LB, Hulette CM. Diagnosis of intracranial
hypotension carries the risk of precipitating infarction within vasculitis: a multi-disciplinary approach. J Neuropathol Exp Neurol.
hypoperfused brain tissue. 1998;57:30–8
Angioplasty and direct infusion of vasodilators have been Ducros A, Boukobza M, Porcher R, Sarov M, Valade D, Bousser MG.
The clinical and radiological spectrum of reversible cerebral vaso-
attempted as measures to relieve severe arterial narrowing.
constriction syndrome. A prospective series of 67 patients. Brain
These intra-arterial strategies carry the risk of reperfusion 2007;130:3091–101
injury (edema and hemorrhage). They should be reserved for Hajj-Ali RA, Furlan A, Abou-Chebel A, Calabrese LH. Benign angiopa-
patients who have declined clearly as a result of progressive thy of the central nervous system: cohort of 16 patients with clinical
course and long-term followup. Arthritis Rheum. 2002;47: 662–9
vasoconstriction. Calcium-channel blockers such as vera-
Molloy ES, Singhal AB, Calabrese LH. Tumor-like mass lesion – an
pamil and nimodipine may be useful in mitigating headaches, under-recognized presentation of primary angiitis of the central ner-
but to date, have not been found to influence the course of vous system. Ann Rheum Dis. 2008;67:1732–5
vasoconstriction (Chen et al. 2008; Ducros et al. 2007). They Moore PM. Central nervous system vasculitis. Curr Opin Neurol.
1998;11:241–6
may also lead to vascular steal phenomena in which vaso-
Parisi JE, Moore PM. The role of biopsy in vasculitis of the central
constricted vessels are further deprived of blood. nervous system. Semin Neurol. 1994;14:341–8
Schwedt TJ, Matharu MS, Dodick DW. Thunderclap headache. Lancet
Myth: RCVS is an extremely rare condition. Neurol. 2006;5:621–31
Singhal AB. Cerebral vasoconstriction without subarachnoid blood:
Reality: Reports of RCVS have increased steadily since the
associated conditions, clinical and neuroimaging characteristics.
turn of the current century. Our two centers have encountered Ann Neurol. 2002:S:59–60
more than 140 RCVS cases, and a single-center prospective Singhal AB. Postpartum angiopathy with reversible posterior leukoen-
study reported 67 cases occurring over a period of only 5 years cephalopathy. Arch Neurol. 2004;61:411–16
Singhal AB, Caviness VS, Begleiter AF, Mark EJ, Rordorf G, Koroshetz
(Ducros et al. 2007). Several factors account for this – the
WJ. Cerebral vasoconstriction and stroke after use of serotonergic
advent of CT and MR technologies for the assessments of the drugs. Neurology 2002;58:130–3
 Thromboangiitis Obliterans (Buerger’s Disease)
30
Eric L. Matteson and John H. Stone

Myth: The risk of amputation in patients with thromboangii-

30.1 Overview of Thromboangiitis tis obliterans is low in the first years of disease but increases
Obliterans (Buerger’s Disease) markedly over time.
Comment: This statement is too optimistic. It boils down to
› A form of vasculitis associated with a highly inflam-
this: the risk of amputation in patients with thromboangiitis
matory clot within the lumen as well as inflammation
obliterans is high in the first years of disease and even higher
within the vascular wall.
in the long run, particularly among patients who continue to
› Associated with active cigarette smoking, typically
smoke.
moderate to heavy.
The probability of having an amputation of either a termi-
› Tends to involve medium-sized arteries and veins, with
nal digit or a larger part of a limb is 50% at 15 years of dis-
the most striking clinical pathology related to arterial
ease (Fig. 30.1) (Cooper et al. 2004). As shown in Fig. 30.1,
disease.
the risk of amputation is already high in the first 5 years of
› Digital ischemia in all four extremities is a hallmark of
disease. The risk of amputation persists up to at least 18
the disorder, with the involvement of some extremities
years following the diagnosis in thromboangiitis obliterans
more characteristic than others.
patients.
› Severe digital ischemia without the evidence of inter-
nal organ involvement. Pearl: Thromboangiitis obliterans never involves only one
› Angiography reveals segmental involvement of medium- limb.
sized arteries, with abrupt vascular cut-offs and cork-
Comment: Thromboangiitis obliterans is always present in
screw collaterals. The major vessel involvement occurs
more than arm or leg, and often in both arms and both legs,
at the levels of the ankle and wrist, with major conse-
even if clinical evidence of multi-limb involvement is not
quences to the distal extremities.
› Smoking cessation is the only therapy known to be
effective for Buerger’s disease.
100

80
Myth: Patients with thromboangiitis obliterans are more
addicted to nicotine than are those with atherosclerotic coro- 60
nary artery disease. %
Reality: The “classic” patient with thromboangiitis obliter- 40
ans is a young male who smokes heavily. As discussed in this
chapter, several parts of this stereotype are misleading. For 20
example, in a prospective study, patients with thromboangii-
tis obliterans actually smoked significantly fewer cigarettes 0
per day than did patients with coronary atherosclerosis 0 2 4 6 8 10 12 14 16 18 20
(Cooper et al. 2006). Patients with thromboangiitis obliter- Years
ans are also more likely to have made serious attempts to quit
Fig. 30.1 Survival free of amputation in patients with thromboangiitis
smoking. Their degree of tobacco dependence was judged to
obliterans. The risk of major or minor amputation at 20 years was 56%.
be comparable to that of patients with atherosclerotic coro- From Dr. Leslie T. Cooper, Mayo Clinic. Reprinted with kind permis-
nary artery disease, not greater. sion from Elsevier Limited

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 317

DOI:10.1007/978-1-84800-934-9_30, © Springer Science + Business Media B.V. 2009
318 E. L. Matteson and J. H. Stone

Fig. 30.2 Clinical and angio-

graphic findings in throm-
boangiitis obliterans. (a) Left
foot of a patient with throm-
boangiitis obliterans and critical
limb ischemia, illustrating
dependent rubor and hair loss.
(b) The contrast angiogram of
the same foot illustrates diffuse
thrombosis of the distal pedal
arteries. Numerous small
collateral blood vessels,
including adventitial
“corkscrew”-type vessels, can be
seen in the region of the
occluded distal peroneal and
dorsalis pedis arteries. (Courtesy
of Cooper, Mayo Clinic.)

apparent immediately (Hirai and Shionoya 1979) (Fig. 30.2a, b).

If called to see a patient who has intense ischemia in only
one limb and no angiographic evidence of disease in other
extremities, thromboangiitis obliterans is not the diagnosis
(Fig. 30.3).
Myth: Thromboangiitis obliterans is a systemic vasculitis.
Reality: Although often classified as a systemic vasculitis,
there are several important distinctions between throm-
boangiitis obliterans and other forms of vasculitis (Puechal
and Fiessinger 2007). Thromboangiitis obliterans is charac-
terized pathologically by a highly cellular thrombus with
relative sparing of the blood vessel walls (Fig. 30.4).
Multinucleated giant cells can even be observed within the
clot in thromboangiitis obliterans. The architecture of the
vascular wall is preserved, and fibrinoid necrosis is absent. Fig. 30.3 Early lesions in thromboangiitis obliterans (hematoxylin and
eosin staining). Thrombosed artery with highly cellular thrombus and
The occlusive clot found within the lumen of blood vessels intact internal elastic lamina, illustrating the typical early lesion of
involved by thromboangiitis obliterans explains why immuno- thromboangiitis obliterans (hematoxylin and eosin, medium power).
suppressive therapy is ineffective in this disorder. Unfortunately, (Courtesy of William Edwards, Mayo Clinic.)
30 Thromboangiitis Obliterans (Buerger’s Disease) 319

anticoagulation and thrombolytic treatment approaches are as 23% (Lie 1987; Olin 2000). Thromboangiitis obliterans has
also either unhelpful or unproven. Smoking cessation is the a lot more to do with whether or not one smokes than whether
critical component of thromboangiitis obliterans therapy. or not a Y chromosome is present.
Myth: Thromboangiitis obliterans is a disease of young men. Myth: Smoking marijuana can cause a disease identical to
Buerger’s disease.
Reality: In a Mayo Clinic series from the 1930s, less than 1% of
the patients were women (Horton 1938). As the proportion of Reality: Digital ischemia has been reported as a result of mar-
smokers who are women has grown, the percentage of throm- ijuana use (Fig. 30.5) (Sterne and Ducastaing 1960). Scrutiny
boangiitis obliterans patients who are women has expanded of these results indicates that “cannabis arteritis” bears a
proportionally. More recent estimates of the percentage of remarkable resemblance to thromboangiitis obliterans clini-
thromboangiitis obliterans patients who are women are as high cally, that most of the patients reported smoked cigarettes as
well as marijuana, and that the reported pathological differ-
ences between these allegedly separate entities are uncon-
vincing (Combemale et al. 2005; Disdier et al. 1999). Cannabis
arteritis is probably really just thromboangiitis obliterans.
Pearl: Superficial thrombophlebitis may be the first manifes-
tation of thromboangiitis obliterans.
Comment: Buerger himself noted that fleeting, painful, erythema-
tous lesions—symptomatic of a superficial thrombophlebitis—
are frequently the initial manifestation of this disorder (Buerger
1908). Buerger called attention to two distinct disease phases:
(1) a migratory thrombophlebitis; (2) an occlusive disease of
deeper vessels. Describing the first phase, Buerger wrote:
“Most interesting is the acute stage or earliest lesion, which
occurs simultaneously with or shortly after the onset of throm-
Fig. 30.4 Highly cellular thrombus with relative sparing of the blood bosis.” Histologic examination of these lesions reveals an acute
vessel walls (Courtesy of Dr. John Stone) thrombophlebitis with marked perivascular infiltration.

a b

Fig. 30.5 (a, b) “Cannabis

arteritis” is probably caused by
concomitant cigarette smoking
and consequent thromboangiitis
obliterans and is therefore not a
true vasculitic entity itself.
However, this cannabis user did
develop a vasculitis: cryoglobu-
linemia secondary to hepatitis
C, acquired through injection
drug use (Courtesy of Dr. John
Stone)
320 E. L. Matteson and J. H. Stone

The superficial thrombophlebitis of thromboangiitis oblit- Fig. 30.6 Angiogram in

erans is migratory, recurrent, and may affect the arms and thromboangiitis obliterans
showing an area of normal-
legs. Migrating phlebitis (phlebitis saltans) in a young smoker appearing artery distal to severely
is therefore highly suggestive of thromboangiitis obliterans. diseases vessels (Courtesy of Dr.
Biopsy of a subcutaneous nodule can be used to confirm the John Stone)
presence of superficial thrombophlebitis at an early phase of
thromboangiitis obliterans, before the onset of larger vessel
thrombosis (Hill 1974).
Patients rarely consult physicians at the time of this early
complaint, and therefore these lesions are usually overlooked
or disregarded. This “herald lesion,” however, is generally
followed by progressive occlusive disease in the deeper arter-
ies and veins, leading the patient to medical attention.
Pearl: Thromboangiitis obliterans targets medium-sized ves-
sels at the level of the wrists and ankles.
Comment: Again, Buerger’s elegant description: “The begin-
nings of the obliteration are not to be sought in the capillar-
ies nor in the finest branches. If we follow the vessels upward,
we frequently see a sudden cessation of the process, and in a
number of instances we find that some 5 or 10 cm of a ves-
sel’s length is closed, and that the portions above and below
are apparently normal” (Buerger 1908).
Radiologic findings often correlate well with this patho-
logic observation (Fig. 30.6). An angiogram of the extremities
often demonstrates areas of flagrantly abnormal arteries, fol-
lowed by stretches of radiographically normal blood vessels.

Pearl: Thromboangiitis obliterans may begin with joint

manifestations.

Comment: Recurrent episodes of large-joint arthritis and tran-

sient, migratory single joint episodes accompanied by local
signs of inflammation have been described in thromboangiitis
obliterans (Puechal et al. 1999). When present, joint symp-
toms usually precede the onset of digital ischemia. In this way,
thromboangiitis obliterans resembles many other forms of
vasculitis, which are often heralded by nonspecific musculo-
skeletal complaints, including a migratory oligoarthritis.
In thromboangiitis obliterans, the arthritis disappears with Raynaud’s phenomenon, so typical of “true” connective tis-
the appearance of ischemic signs, much as asthma disappears sue disorders (i.e., systemic sclerosis, systemic lupus erythe-
once the vasculitic phase of the Churg–Strauss syndrome matosus, and dermatomyositis), is caused by vasospasm. This
begins. is usually induced by cold, emotional upset, and other stimuli
recognized by the patient. Conversely, thromboangiitis obliter-
Myth: Thromboangiitis obliterans may present with Raynaud’s
ans is not the result of vasospasm but rather a highly inflamma-
phenomenon.
tory clot in the medium-sized arteries at the level of the wrists
Reality: A century and a half after Maurice Raynaud (1862) and ankles. The term “Raynaud’s phenomenon” should not be
first described De l’asphyxie locale et de la gangrène symé- applied to patients who have thromboangiitis obliterans, nor to
trique des extrémités (On the local asphyxia and symmetric any other form of vasculitis that causes digital ischemia.
gangrene of the digits), there remains a tendency to lump
Pearl: Patients who continue to smoke are the ones who lose
many forms of digital ischemia under the rubric of “Raynaud’s
limbs.
disease.” Thromboangiitis obliterans causes digital ischemia
equaled by few other disorders, but it is not because of Comment: An analysis of 850 patients identified by a national
Raynaud’s phenomenon. study in Japan showed that the risk for amputation was 2.7
30 Thromboangiitis Obliterans (Buerger’s Disease) 321

Comment: Cases of extra-extremity involvement are suffi-

ciently rare as to constitute exceptions to the rule that paren-
chymal organ involvement by this disorder does not happen.
Only a handful of well-documented cases of exist in which
thromboangiitis obliterans has involved the vessels of the
mesentery or central nervous system (Puechal and Fiessinger
2007). If a patient presents with intense digital ischemia to
the virtual exclusion of ischemia in other vascular beds, think
of thromboangiitis obliterans.
Pearl: The three forms of vasculitis most likely to lead to the
loss of a pulse are Takayasu’s arteritis, giant cell arteritis,
and thromboangiitis obliterans.
Comment: Takayasu’s arteritis has been termed “pulseless dis-
ease” because of the frequency with which it causes the extinc-
tion of pulses in the arms (and less often the legs) (Fig. 30.9a).
Subclavian and axillary artery disease in giant cell arteritis
Fig. 30.7 Patients who have thromboangiitis obliterans who continue also leads to the loss of pulses in the upper extremity. In com-
to smoke continue to lose digits and other portions of their extremities.
This patient lost every finger on both hands and had bilateral below- parison to thromboangiitis obliterans, the vascular occlusions
the-knee amputations (Courtesy of Dr. John Stone) are more proximal in Takayasu’s arteritis and giant cell arteri-
tis. Thus, the entire arm or leg is affected, and extremity clau-
dication is a common symptom. Because of the effectiveness
of collateral circulation in Takayasu’s and giant cell arteritis,
ischemia sufficient to cause gangrene of the extremities is sel-
dom seen in those disorders today.
Through its involvement of medium-sized arteries at the
levels of the wrist and ankles, thromboangiitis obliterans
also obliterates pulses in the upper and lower extremities
(Fig. 30.9b). In thromboangiitis obliterans, severe ischemic
changes tend to develop in the distal extremities, usually in
one or more digits of the hands or feet, often accompanied by
gangrene.
Myth: Smoking must be heavy for thromboangiitis obliterans
to occur.
Fig. 30.8 Splinter hemorrhages at presentation in a patient who had
thromboangiitis obliterans (Courtesy of Dr. John Stone) Comment: The definitions of “heavy” vary among patients
and between physicians and patients, but thromboangiitis
obliterans has been reported in patients who smoke less than
times higher (95% confidence interval: 1.9–4.0) in patients
a pack of cigarettes a day (Fig. 30.10). Furthermore, even
who continued to smoke (Sasaki et al. 2000). Patients who
second-hand smoke has been implicated in thromboangiitis
have thromboangiitis obliterans must choose between their
obliterans. Although a compelling case for thromboangiitis
cigarettes and their fingers (Fig. 30.7).
obliterans cannot be made without a history of cigarette expo-
Pearl: Thromboangiitis obliterans may present with splinter sure, there is no two-pack per day minimum requirement.
hemorrhages.
Myth: Necrotic fingertips should be amputated for pain relief
Comment: Endocarditis is not the only disorder that causes and avoidance of other complications, such as infection.
splinter hemorrhages. A variety of systemic vasculitides,
Reality: Although there is nothing like timely surgical inter-
including thromboangiitis obliterans, also cause this physical
vention for many conditions, it is usually a mistake to have a
finding (Fig. 30.8) (Quenneville et al. 1981). Other vasculiti-
patient with digital ischemia due to thromboangiitis obliter-
des known to cause these subungual lesions are the ANCA-
ans (or any form of vasculitis) evaluated first by a surgeon.
associated vasculitides and cryoglobulinemia.
Such patients often wind up having amputations prematurely
Pearl: Extra-extremity disease occurs in thromboangiitis and lose more tissue than they would have had Nature been
obliterans very rarely. allowed to take its course—partly redirected by smoking
322 E. L. Matteson and J. H. Stone

a a

Fig. 30.9 Imaging findings in Takayasu’s arteritis (a) as opposed to

Buerger’s disease (thromboangiitis obliterans) (b). (a) Magnetic reso-
nance angiogram in Takayasu’s arteritis. There is an abrupt vascular
narrowing right at the take-off of the right subclavian artery, the branch
of the aorta most commonly involved in Takayasu’s arteritis (arrow).
(b) Conventional angiogram in a patient with Buerger’s disease (throm-
boangiitis obliterans). The image shows the radial and ulnar arteries
near the level of the patient’s wrist, far more distal that the typical
involvement in Takayasu’s. The findings include “corkscrew” collater- Fig. 30.10 Clinical picture and angiogram of thromboangiitis obliterans
als and obliteration of the distal vessels. The proximal arteries (e.g., the in an 18-year-old woman who smoked only one pack of cigarettes a day.
subclavians) were normal in this patient (Courtesy of Dr. John Stone) Women can get thromboangiitis obliterans, too, and their smoking does
not have to be “heavy.” (a) Bilateral ischemia with small areas of skin
necrosis in the toes. (b) Angiography of both lower extremities revealing
dramatically damaged blood vessels, with an extent of lesions far more
cessation (in thromboangiitis obliterans) or immunosuppres- proximal and far worse than one would suspect from the clinical picture
of the toes alone. Clinical picture and angiogram of thromboangiitis
sion (in other vasculitides).
obliterans in an 18-year-old woman who smoked only one pack of ciga-
Despite the sometimes hopeless appearance of digital rettes a day. Women can get thromboangiitis obliterans, too, and their
gangrene, fingers and toes can make astonishing recoveries if smoking does not have to be “heavy” (Courtesy of Dr. John Stone)
30 Thromboangiitis Obliterans (Buerger’s Disease) 323

the underlying inflammatory process is arrested. After a Hill GL. A rational basis for management of patients with the Buerger
period of some months, traces of vasculitis can be difficult to syndrome. Br J Surg 1974;61:476–81
Hirai M, Shionoya S. Arterial obstruction of the upper limb in throm-
find even in patients who had gangrene. Patients usually boangiitis obliterans: its incidence and primary lesion. Br J Surg
require ample support with pain medications during the reso- 1979;66:124–8
lution of these events. Horton BT. The outlook in thromboangiitis obliterans. JAMA 1938;
111:2184–9
Lie JT. Thromboangiitis obliterans (Buerger’s disease) in women.
Medicine (Baltimore) 1987;66:65–72
References O'Dell J, Linder R, Markin RS, Moore GF. Thromboangiitis obliterans
(Buerger's disease) and smokeless tobacco. Arthritis Rheum 1987;
30(9):1054–6
Buerger L. Thromboangiitis obliterans: a study of the vascular lesions Olin J. Thromboangiitis obliterans (Buerger’s disease). N Eng J Med
leading to presenile spontaneous gangrene. Am J Med Sci 1908; 2000;343:864–70
136:567–80 Puechal X, Fiessinger JN, Kahan A, et al Rheumatic manifestations in
Combemale P, Consort T, Denis-Thelis L, et al Cannabis arteritis. Br patients with Thromboangiitis obliterans (Buerger’s disease).
J Dermatol 2005;152:166–9 J Rheumatol 1999;26:1764–8
Cooper LT, Henderson SS, Ballman KV, et al A prospective, case con- Puechal X, Fiessinger JN. Thromboangiitis obliterans or Buerger’s disease:
trol study of tobacco dependence in thromboangiitis obliterans challenges for the rheumatologist. Rheumatology 2007;46:192–9
(thromboangiitis obliterans). Angiology 2006;57:73–8 Quenneville JG, Prat A, Gossard D. Subungual-splinter hemorrhage: an
Cooper LT, Tse PS, Mikhail MA, et al Long-term survival and amputa- early sign of thromboangiitis obliterans. Angiology 1981;32:424–32
tion risk in thromboangiitis obliterans (thromboangiitis obliterans). Sasaki S, Sakuma M, Yasuda K. Current status of thromboangiitis oblit-
J Am Coll Cardiol 2004;44:2410–11 erans (Buerger’s disease) in Japan. Int J Cardiol 2000;75 Suppl
Disdier P, Swiader L, Jouglard J, et al Arterite du cannabis versus mala- 1:S175–81
die de leo buerger. Discussion nosologique a propos de deux nou- Sterne J, Ducastaing C. [Arteritis caused by Cannabis indica]. Arch
veaux cas. Presse Med 1999;28:71–4 Mal Coeur Vaiss 1960;53:143–7
 Less Common Forms of Vasculitis
31
Eric L. Matteson and John H. Stone

A variety of adjunct therapies have been employed, including

31.1 Cogan’s Syndrome azathioprine, cyclophosphamide, methotrexate, and tumor
necrosis factor (TNF) inhibitors, in efforts to improve out-
31.1.1 Overview of Cogan’s Syndrome comes and reduce glucocorticoid-related side effects.
Cogan’s syndrome patients who suffer profound hearing
loss are good candidates for cochlear implantation (Pafanisi
› The hallmark of Cogan’s syndrome is the presence of
et al. 2003). Hence, a protracted course of immunosuppres-
ocular inflammation and audiovestibular dysfunction.
sive and glucocorticoids should not be pursued, especially if
These findings may be accompanied by the evidence of
there is only marginal response to treatment, and hearing loss
a systemic vasculitis that involves large blood vessels.
is the major disease manifestation. Unfortunately, in some
› Interstitial keratitis is the most common form of ocular
patients, refractory eye disease necessitates ongoing immuno-
involvement.
suppression.
› Audiovestibular dysfunction may lead to the acute
onset of vertigo, tinnitus, nausea, and vomiting. Pearl: If disease remission is achieved in Cogan’s syndrome,
› Vasculitis in Cogan’s syndrome may take the form of treatment should never be stopped completely.
aortitis, renal artery stenosis, or occlusion of the great
Comment: Clinicians’ efforts to control most forms ocular
vessels.
inflammation with immunosuppression are usually rewarded.
In contrast, the challenge of preventing deafness in Cogan’s
syndrome can be humbling (Mazlumzadeh and Matteson
Myth: Hearing loss in Cogan’s syndrome is preventable by
2007; Riente et al. 1996). The two largest series of Cogan’s
aggressive use of glucocorticoid therapy and additional
syndrome patients reported to date paint discouraging pictures
immunosuppressive agents.
of the likelihood of retaining useful hearing over time. In a
Reality: It is important to recognize the limitations of current study from the Mayo Clinic, 42 of 60 patients became deaf in
therapies for the auditory component of Cogan’s syndrome. at least one ear, and 31 of 60 became completely deaf over
Prompt implementation of high-dose glucocorticoid treat- time (Gluth 2006). In a study from France, 30 of 32 patients
ment (1 mg/kg/day) initially results in the improvement in developed significant auditory dysfunction; 11 became com-
hearing for many patients, but severe hearing loss and deaf- pletely deaf (Grasland et al. 2004).
ness eventually occur in at least 50% of patients with Cogan’s These reports and anecdotal experiences inform the bias
syndrome (Gluth et al. 2006; Mazlumzadeh and Matteson that patients who achieve disease control should never dis-
2007). The deafness often occurs gradually, through incre- continue their treatment entirely. Useful hearing loss is usu-
mental hearing losses resulting from disease “flares.” Patients ally not restored entirely by treatment. The optimal remission
often recover some but not all of the hearing they possessed maintenance regimen is far from clear (to be more precise, it
before an individual disease exacerbation. is entirely empirical). Remission maintenance regimens may
After an initial 2 weeks of 1 mg/kg prednisone, it is impor- include immunomodulating agents such as methotrexate,
tant to taper in decrements of 5–10 mg every 1–2 weeks. In the azathioprine, or mycophenolate mofetil, as well as low-dose
absence of other systemic manifestations, prednisone should glucocorticoids.
be reduced even more quickly (e.g., 5–10 mg/day) to avoid
Myth: Cogan’s syndrome has a specific autoantibody profile.
excessive morbidity from glucocorticoids (Fig. 31.1a, b).
There is little evidence that protracted use of high-dose Reality: To date, no immune abnormality unique to Cogan’s
glucocorticoids is effective in preserving hearing over the long syndrome has been identified. Autoantibody testing may be
term (Gluth et al. 2006; Mazlumzadeh and Matteson 2007). performed to rule out other causes of sensorineural hearing

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 325

DOI:10.1007/978-1-84800-934-9_31, © Springer Science + Business Media B.V. 2009
326 E. L. Matteson and J. H. Stone

Fig. 31.1 Audiograms of a patient with

a
Cogan’s syndrome before and after therapy.
There is typical bilateral hearing loss (a),
and some improvement on glucocorticoid
therapy 2 months later (b). HL hearing level
(this is the decibel (dB) level that corre-
sponds to the numbers on the ordinate of
the audiogram); masked AC: masked air
conduction threshold; masked BC: masked
bone conduction threshold; SL sensation
level (this is the dB level above the SRT,
usually 25 or 40 dB SL); speech reception
threshold is the lowest dB level at which a
patient accurately repeats common
two-syllable words, such as “bottle” or
“marriage” 50% of the time (Courtesy of
Dr. Eric Matteson)

loss, dizziness, and eye disease in patients with suspected Pearl: Any kind of ocular inflammation can occur in Cogan’s
Cogan’s syndrome, but there are no specific autoantibody syndrome.
tests which are present in patients with Cogan’s syndrome
(Mazlumzadeh and Matteson 2007). Antibodies to heat shock Comment: Nonsyphilitic interstitial keratitis is the classic
protein-70 have been posited as a marker of active inflamma- ocular manifestation of Cogan’s syndrome, but just about any
tion in various forms of autoimmune inner ear disease, but the type of ocular inflammation can occur in this disorder. In addi-
experience with this antibody testing has been disappointing. tion to interstitial keratitis and scleritis, other ocular lesions in
The work-up for patients with clinical presentations com- Cogan’s syndrome include conjunctivitis, episcleritis, uveitis,
patible with Cogan’s syndrome includes a slit-lamp exami- retinal vasculitis, orbital pseudotumor, and a pan-ophthalmitis
nation, audiogram, vestibular studies, chest radiography, and that resembles orbital pseudotumor (Vollertsen et al. 1986).
a magnetic resonance imaging study to exclude acoustic neu- Patients who do not have the classic interstitial keratitis at
roma and other intracranial pathology (Mazlumzadeh and diagnosis are sometimes referred to as having “atypical”
Matteson 2007). Cogan’s syndrome, but it is not clear that the courses of such
patients differ significantly from those of patients who have
Pearl: Interstitial keratitis is the most common ocular mani- textbook presentations. If an atypical eye lesion develops
festation of Cogan’s syndrome. before the onset of sensorineural hearing loss, the diagnosis
Comment: In most cases, interstitial keratitis occurs early in of Cogan’s syndrome is impossible to make until both ocular
the disease course and responds well to glucocorticoid eye- and auditory findings are present.
drops. In some cases, atropine or cyclosporine eyedrops are In some patients, the diagnosis of Cogan’s syndrome
required (Mazlumzadeh and Matteson 2007) (Fig. 31.2a, b). becomes clear only when interstitial keratitis occurs following
31 Less Common Forms of Vasculitis 327

a b

Fig. 31.2 (a, b) The eye in Cogan’s syndrome. A 54-year-old man with (arrows). The lesion responded to glucocorticoid therapy, but the
Cogan’s syndrome. There is interstitial keratitis that appears as an area patient suffered hearing loss and eventually required cochlear implanta-
of oval-shaped haziness in the upper inner quadrant of the cornea tion (note cochlear wire behind patient’s left ear)

the treatment of another ocular manifestation of this disorder

such as scleritis.
Pearl: Cogan’s syndrome can cause meningitis.
Comment: Cogan’s syndrome can be complicated by intense
headaches, photophobia, and neck stiffness. Lumbar puncture
in such cases often reveals a lymphocytic pleocytosis. The
presentation raises obvious concerns about infections (oppor-
tunistic or otherwise), but following the appropriate exclusion
of microbial pathogens the meningitis responds promptly to
glucocorticoids.

31.2 Erythema Elevatum Diutinum

31.2.1 Overview of Erythema Elevatum

Diutinum

• Erythema elevatum diutinum (EED) is an extremely rare

form of small-vessel vasculitis that leads to the formation
of distinctive nodules or papules, usually over the exten-
sor surfaces of joints.
• New lesions come in the form of tender papules that are
associated with pruritus or a burning sensation.
• Lesions develop into red, reddish-brown, or purple pap- Fig. 31.3 Erythematous, firm, papular, and nodular lesions over the
elbow of a 54-year-old male with EED. (Courtesy of Dr. L.E. Gibson,
ules or nodules, and may coalesce to form large plaques.
Mayo Clinic)
Pearl: The skin lesions of EED usually appear on the exten-
sor surfaces, especially near joints of the hands, elbows,
patients. The skin lesions of EED can be reddish, purplish,
knees, and ankles.
and yellowish cutaneous papules, plaques, and nonblanching
Comment: EED lesions are usually symptomatic at onset, nodules, occurring preferentially over the extensor surfaces,
with itching and burning in affected areas (Gibson and especially near joints of the hands, elbows, knees, and ankles
El-Azhary 2000). They are initially tender to touch in many (Fig. 31.3). They may also occur on the trunk and scalp.
328 E. L. Matteson and J. H. Stone

Myth: EED is a systemic vasculitis. syndrome (polyneuropathy, organomegaly, endocrinopathy,

monoclonal gammopathy, and skin changes). Autoimmune
Reality: EED is a rare form of chronic, recurring cutaneous
conditions with which EED is associated include Crohn’s
leukocytoclastic vasculitis. It is frequently associated with a
disease, celiac disease, relapsing polychondritis, Wegener’s
variety of underlying infectious, malignant, autoimmune
granulomatosis, rheumatoid arthritis, pyoderma gangreno-
conditions, but it is not a systemic vasculitis sui generis. EED
sum, and type 1 diabetes mellitus.
is believed to be secondary to immune complex deposition,
which can occur in the context of a number of underlying Pearl: The optimal therapy for EED is treatment of the under-
disorders (Gibson and El-Azhary 2000; Wahl et al. 2005). lying condition.
Pathogens associated with EED include the human immu- Comment: Amelioration of the underlying condition improves
nodeficiency virus (HIV), hepatitis B and C, Treponema pal- EED (presumably) by decreasing the formation of immune
lidum, Mycobacterium tuberculosis, and Streptococcus complexes related to the underlying disease (Gibson and
species. EED had also been reported in hematopoietic malig- El-Azhary 2000; Wahl et al. 2005). This explanation is com-
nancies such as lymphoma, myelodysplasia, and the POEMS patible with the best current hypothesis about disease patho-
genesis, but is likely incomplete.
Dapsone is also effective in treating many cases of EED.
Other therapies have included antibiotics, colchicine, gluco-
corticoids (intralesional, topical, or oral), niacinamide, and
chloroquine. In patients who have advanced fibrotic skin dis-
ease, these interventions are less effective than when used in
patients with early skin lesions.
Pearl: EED lesions are most likely to respond to therapy if
they are treated early.
Comment: EED is a chronic disease that can present as an
acute leukocytoclastic vasculitis. The histopathologic fea-
tures are a diffuse, mixed cellular infiltrate with polymorpho-
nuclear cell infiltration of the vessel wall, nuclear dust, and
spindle cells (histiocytes, fibroblasts, and myofibroblasts)
(Figs. 31.4 and 31.5) (Gibson and El-Azhary 2000). In the
later stages of disease evolution, the lesions become fibrotic
Fig. 31.4 Photomicrograph (hematotoxin and eosin staining) demon-
and contain lipid deposits (Fig. 31.6a, b). At this advanced
strating leukocytoclastic vasculitis with a diffuse dermal mixed cellular
infiltrate in EED. (Courtesy of Dr. L.E. Gibson, Mayo Clinic) stage, the skin lesions respond poorly to treatment.

31.3 Urticarial Vasculitis

31.3.1 Overview of Urticarial Vasculitis

• Urticarial vasculitis can be divided into a form associated

with normal complement levels and a form associated
with hypocomplementemia.
• Normocomplementemic patients are likely to have a self-
limited course that resembles hypersensitivity vasculitis.
In contrast, hypocomplementemic patients are more likely
to have chronic disease.
• The lesions of urticarial vasculitis are frequently associ-
ated with burning or pain rather than pruritus and require
more than 24 h to resolve.
Fig. 31.5 Photomicrograph (hematotoxin and eosin staining) demon-
• Hypocomplementemic urticarial vasculitis has many sim-
strating more mature changes as vascular damage and polymorphonu-
clear cell infiltrate is replaced by a fibrosing response in EED. (Courtesy ilarities to systemic lupus erythematosus (SLE), and is
of Dr. L.E. Gibson, Mayo Clinic) considered by some to be part of the SLE spectrum.
31 Less Common Forms of Vasculitis 329

Fig. 31.7 Photomicrograph (hematoxylin and eosin staining) demon-

b strating leukocytoclastic vasculitis in a skin biopsy from a patient with
normocomplementemic urticarial vasculitis. Red blood cell extravasation,
leukocyte fragmentation, and perivascular infiltrate composed mostly of
neutrophils in the background of dermal edema are seen. These findings
can be present in any type of leukocytoclastic vasculitis that involves
small blood vessels. (Courtesy of Dr. L.E. Gibson, Mayo Clinic)

• Hypocomplementemic urticarial vasculitis is associated

with heritable deficiencies of complement in some patients.
• Direct immunofluorescence of a skin biopsy specimen is
the critical test. Intense staining for immunoreactants (i.e.,
IgG, IgM, C3, C4, and C1q) is found not only in and
around small blood vessel walls but also in a ribbon along
the dermal/epidermal junction. In the proper clinical set-
ting, these findings are diagnostic of hypocomplement-
emic urticarial vasculitis.
Myth: All forms of urticarial vasculitis are alike.
Reality: The critical distinction between the two major forms
of urticarial vasculitis is whether or not the patient is hypo-
complementemic, based on the measurements of C3, C4, and
total complement (CH50). Normocomplementemic urticarial
vasculitis is viewed most appropriately as a subset of cutane-
ous leukocytoclastic angiitis (hypersensitivity vasculitis) in
which the skin manifestation of small-vessel vasculitis takes
the form of urticaria rather than the more familiar palpable
purpura. The histopathologic features of normocomplement-
emic urticarial vasculitis are shown in Fig. 31.7.
Normocomplementemic urticarial vasculitis normally as a
self-limited course whether or not the underlying stimulus –
often a medication – can be identified and removed. In
Fig. 31.6 Advanced lesions of EED over the knuckles (a) and elbow
(b). Lesions that have become indurated are typically refractory to ther- chronic cases, normocomplementemic ultraviolet must be
apy (Courtesy of Dr. John Stone) distinguished carefully from neutrophilic urticaria, a persis-
tent form of urticaria not associated with vasculitis (Mehregan
et al. 1992; Davis et al. 1998; Davis and Brewer 2004; Toppe
• Patients with hypocomplementemic urticarial vasculitis et al. 1998).
can develop a number of features that are atypical of SLE, In contrast to the normocomplementemic variant, hypo-
including scleritis and chronic obstructive pulmonary complementemic urticarial vasculitis represents a unique
disease. form of small-vessel vasculitis that is characterized by a high
330 E. L. Matteson and J. H. Stone

likelihood of extracutaneous manifestations. Hypocomple- urticarial lesions resolve without skin discoloration, hyper-
mentemic urticarial vasculitis often occurs with other disease pigmentation is often observed following the resolution of
features that suggest an SLE-like illness. In such cases, the the lesions of urticarial vasculitis, particularly if the lesions
term hypocomplementemic urticarial vasculitis syndrome is recur multiple times (more likely among patients with hypo-
sometimes used (Wisnieski et al. 1995). complementemia) (Fig. 31.8a–c).
Myth: Patients with urticarial vasculitis must be treated with Pearl: Urticarial lesions persisting for more than 48 h are
prednisone. more likely to be related to an underlying vasculitis than
such lesions of shorter duration.
Reality: Urticarial vasculitis patients have urticarial lesions
that usually persist more than 24 h can be painful, and fre- Comment: Common urticaria is encountered far more com-
quently resolve with residual pigmentation in the skin. The monly in clinical practice than urticarial vasculitis. Urticaria
infiltrate is neutrophilic, as opposed to the lymphocytes seen can be troubling and even disabling, but does not require (or
in most patients with cutaneous lupus. Patients who have low respond to) the type of therapy used in urticarial vasculitis.
complement levels frequently have SLE as the underlying Thus, it is critical to distinguish these two conditions. Aside
disease. For patients with skin-predominant disease, colchi- from the duration of skin lesions, indications that urticaria
cine and dapsone can be an effective therapy and patients may be associated with vasculitis include a stinging or burn-
may not always need treatment with prednisone (Sunderkötter ing component to the symptoms, rather than mere pruritus.
et al. 2005). In addition, visual inspection or diascopy reveals a purpuric
A G6PD should be checked before starting dapsone to component to the urticarial lesions in vasculitis (Fig. 31.9),
make sure that the patient can metabolize the drug. Dapsone and laboratory studies may reveal hypocomplementemia.
is usually begun at a dose of 50 mg/day if baseline blood One other word of caution with regard to urticarial vasculi-
count and liver function tests are normal. The dose is titrated tis is in order. Neutrophilic urticaria, a nonvasculitic condition,
up by 25 mg/day each week, as long as the weekly hemoglo- is readily confused histopathologically with the leukocyto-
bin and liver functions tests are fine. A target dose is nor- clastic vasculitis associated with urticarial vasculitis. Having
mally 150 mg/day, but may be less if control is achieved at a the skin biopsy reviewed by an experienced dermatopatholo-
lower dose. After attaining the maximum stable dose, labora- gist is usually worthwhile.
tories are checked monthly for 3 months, and thereafter every
Pearl: To establish the diagnosis of urticarial vasculitis, skin
3 months.
biopsy should be performed of a recent lesion.
Pearl: Patients with urticarial vasculitis who have low comple-
Comment: It is helpful to obtain two punch biopsies (Davis
ments are more likely to have severe, systemic, and treatment-
et al. 1998). Because the lesions of urticarial vasculitis are in
resistant disease.
the superficial dermis, punch biopsies usually suffice for
Comment: Patients with normal complementemic urticarial diagnosis. Each punch biopsy should be cut in half. One half
vasculitis have good prognoses with waxing and waning dis- should be sent for formalin fixation and for hematoxylin and
ease courses that usually resolve within 1–4 years. Patients eosin staining. The second half should be placed into liquid
with hypocomplementemic urticarial vasculitis are more nitrogen without formalin fixation and sent for direct immu-
likely to have a systemic underlying disease and to have clin- nofluorescence. The immunoreactants worth testing for on
ical manifestations that require ongoing therapy. DIF are C3, C4, C1q, IgG, IgA, and IgM.
Repeat biopsies of new lesions are occasionally required
Pearl: Many patients with urticarial vasculitis (with or with-
to confirm the diagnosis of urticarial vasculitis.
out hypocomplementemia) experience tenderness, burning,
and pruritus. This contrasts with common urticaria that is Pearl: The major entities from which hypocomplementemic
neither tender nor painful. urticarial vasculitis must be distinguished are Henoch–
Schönlein purpura, cryoglobulinemia, and hypersensitivity
Comment: Common chronic urticaria is associated with itch-
vasculitis.
ing but not pain (Mehregan et al. 1992; Davis et al. 1998). On
physical examination, the lesions of urticarial vasculitis can be Comment: The legs in Fig. 31.10a, b could belong to a patient
differentiated from chronic urticaria by applying a magnifying with any of those other three disorders. These are the legs of
glass or glass slide to the skin (diascopy). Diascopy reveals a a 77-year-old man who had extensive palpable purpura on
central, dark brown or red macule, indicative of the purpuric the lower extremities, trunk, and upper extremities. A direct
nature of urticaria. This finding is absent in chronic urticaria. immunofluorescence study of his skin biopsy showed granu-
The lesions of urticarial vasculitis typically last for 1–3 lar deposition of C3, C4, IgG, and IgM along the dermal–
days. In contrast, the lesions of chronic urticaria generally epidermal junction and around small blood vessels in the
resolve within 24 h of their first appearance. Whereas chronic superficial dermis.
31 Less Common Forms of Vasculitis 331

b c

Fig. 31.8 Urticarial vasculitic lesions of 3 days’ duration in a 52-year- sation has occurred, they are nonblanching. (b) Trunk of same patient
old male with an 8-year history of rheumatoid arthritis. (a) The lesions with scratch efflorescences. (c) Purpuric lesions on the soft palate.
on the leg are moderately pruritic and painful, raised, erythematous, and (Courtesy of Dr. Eric Matteson)
develop central clearing with resolution. Where red blood cell extrava-

Pearl: The level of complement C1q is low in most patients Myth: Hypocomplementemic urticarial vasculitis is almost
with hypocomplementemic urticarial vasculitis. always associated with a systemic underlying disease.
Comment: In hypocomplementemic urticarial vasculitis, any Reality: Most cases of hypocomplementemic urticarial vas-
one or all of the complements involved in the classic comple- culitis are idiopathic, just as SLE is idiopathic (Mehregan
ment pathway may be reduced (Davis et al. 1998; Davis and et al. 1992; Davis et al. 1998). Thorough histories, physical
Brewer 2004). This includes total complement, C1, C1q, C2, examinations, and extensive laboratory evaluations are con-
C3, and C4. ducted in most patients with this disorder, but at the end of it
332 E. L. Matteson and J. H. Stone

Fig. 31.9 This urticarial lesion has a central purpuric component, sug-
gestive of an underlying vasculitic nature (Courtesy of Dr. John Stone)

all the proper conclusion in most patients is that an indepen-

dent disorder is present and there lurks no occult malignancy,
infection, or other idiopathic inflammatory disorder as a
rational explanation.
In contrast, normocomplementemic urticarial vasculitis
has multiple potential causes: the lengthy list of disorders
that can cause hypersensitivity vasculitis.
Pearl: Eye findings can help distinguish between hypocom-
plementemic urticarial vasculitis and SLE.
Comment: Scleritis is a common ocular complication of
Fig. 31.10 Palpable purpura in a patient with hypocomplementemic
hypocomplementemic urticarial vasculitis (Fig. 31.11). This urticarial vasculitis involving (a) the lower extremities, (b) trunk, and
complication is highly atypical of lupus. upper extremities (not shown) (Courtesy of Dr. John Stone)
31 Less Common Forms of Vasculitis 333

References

Davis MD, Brewer JD. Urticarial vasculitis and hypocomplementemic

urticarial vasculitis syndrome. Immunol Allergy Clin North Am
2004;24:183
Davis MDP, Daoud MS, Kirby B, et al Clinicopathologic correlation of
hypocomplementemic and normocomplementemic urticarial vascu-
litis. J Am Acad Dermatol 1998;38:899
Gibson LE, El-Azhary RA. Erythema elevatum diutinum. Clin Dermatol
2000;18:295–99
Gluth MB, Baratz KH, Matteson EL, et al Cogan’s syndrome: a retro-
spective review of 60 patients throughout a half-century. Mayo Clin
Proc 2006;81:483–8
Grasland A, Pouchot P, Hachulla E, et al Typical and atypical Cogan’s
syndrome: 32 cases and review of the literature. Rheumatology
2004;43:1007–15
Mazlumzadeh M, Matteson EL. Cogan’s syndrome: an audiovestibular,
ocular, and systemic autoimmune disease. Rheum Dis Clin North
Am 2007;33:855–74
Mehregan DR, Hall MJ, Gibson LE. Urticarial vasculitis: a histopatho-
logic and clinical review of 72 cases. J Am Acad Dermatol 1992;26(3
Pt 2):441–8
Pafanisi, Pafanisi E, Vincenti V, et al Cochlear implantation and Cogan’s
syndrome. Otol Neurotol 2003;24:601–4
Riente L, Taglione E, Berrettini S. Efficacy of methotrexate in Cogan’s
syndrome. J Rheumatol 1996;23:1830–1
Sunderkötter C, Bonsmann G, Sindrilaru A, et al Management of leuko-
cytoclasticvasculitis. J Dermatol Treat 2005;16:193–206
Toppe E, Haas N, Henz BM. Neutrophilic urticaria: clinical features,
histologic changes, and possible mechanisms. Br J Dermatol
1998;138:248–53
Vollertsen R, McDonald T, Younge B, et al Cogan’s syndrome: 18
cases and a review of the literature. Mayo Clin Proc 1986;61:
Fig. 31.11 Scleritis in a patient with hypocomplementemic urticarial 344–61
vasculitis. (Courtesy of Dr. John Stone.) Wahl CE, Bouldin MB, Gibson LE. Erythema elevatum diutinum: clini-
cal, histopathologic, and immunohistochemical characteristics of
six patients. Am J Dermatopathol 2005;27:397–400
Wisnieski JJ, Baer AN, Christensen J, et al Hypocomplementemic urti-
carial vasculitis syndrome: clinical and serologic findings in 18
patients. Medicine 1995;74:24–41
 Relapsing Polychondritis
32
Harvinder S. Luthra

in the pinna bony hard changes that are a reflection of previ-

32.1 Overview of Relapsing Polychondritis ously active disease, some of which may have been subclinical
(Fig. 32.1b) (Kent et al. 2004).
› Relapsing polychondritis (RP) is a multisystem inflam-
Myth: Hearing loss in RP is irreversible.
matory disease of unknown etiology in which the car-
tilaginous structures of the ears, nose, trachea, and Reality: The auditory canal comprises the cartilaginous struc-
joints are the main sites of damage. ture in the outer third of the ear. This portion can become
› A variety of other organs and tissues, including the eyes, inflamed in RP (see Fig. 32.1a). Similarly, the Eustachian
heart, and blood vessels, can also be affected by RP. tube is cartilaginous in its medial two-thirds of the canal.
› Clues from human and animal studies suggest that car- Inflammation in the auditory canal or in the Eustachian tube
tilage components may be the antigens driving the can lead to conductive hearing loss. Nerve deafness caused by
immune response, leading to damage from the result- dysfunction of the auditory nerve has been described in RP
ing inflammation. and is presumed to be secondary to “vasculitis” of the internal
› First described by Jaksch–Wartenhorst (1923), more auditory artery. Histologic proof of vasculitis in such cases is
than 500 patients with RP have been reported in the lacking because of the general inaccessibility of the inner ear
medical literature. (see Chap. 39).
› RP has a worldwide distribution. Both conductive and sensorineural hearing loss in RP gen-
› Some RP cases are associated with other conditions, erally responds well to therapy, particularly if treatment is
particularly myelodysplastic syndromes and connec- initiated early.
tive tissue disorders such as rheumatoid arthritis or
Pearl: Occasionally, patients with RP develop hearing diffi-
systemic lupus erythematosus.
culties along with vertigo.
Comment: This combination of symptoms usually reflects
inner ear disease. These symptoms may be mild and may be
32.2 Ear Involvement ignored by the patient. On the other hand, these may be severe
enough to cause nausea, vomiting, difficulty sitting up in a
chair or bed, standing, and walking. As noted above, the occur-
Pearl: Inflammatory changes in RP are localized to the car-
rence of inner ear symptoms in RP is presumed to be related to
tilaginous part of the external ear, the pinna, and do not
a vasculopathy (and possibly vasculitis) of the cochlear and/or
involve the noncartilaginous area: the lobule (Fig. 32.1a).
vestibular branch of the internal auditory artery. Endolymphatic
Comment: The ear is the most common site of inflammation hydrops may occur in this setting (Murata et al. 2006).
in RP. Inflammation can involve the outer, middle, and inner Such complications require urgent intervention with
ear. The pinna is the most commonly involved part of the ear; immunosuppressive therapy, typically high doses of gluco-
the inflammation in RP localizes to the cartilagenous portion corticoids.
of the ear, sparing the lobule. If the lobule is inflamed, one
Myth: Nasal bridge collapse cannot be surgically repaired
should consider other diagnosis such as cellulitis. The acutely
because of the inherently fragile tissue and the risk of induc-
inflamed pinna is warm to touch, exquisitely tender, and
ing a disease flare.
painful to move in any direction.
The pinna may return to its normal shape between attacks Reality: Nasal involvement in RP is very common and may
or remain thick because of the fibrosis that has occurred during affect as many as 54% of patients. Nasal disease can present in
the period of acute inflammation. In some cases, one may find a subacute manner as mild tenderness or redness of the nasal

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 335

DOI:10.1007/978-1-84800-934-9_32, © Springer Science + Business Media B.V. 2009
336 H. S. Luthra

a b

Fig. 32.1 (a) Involvement of the pinna is the most common manifestation of RP. Note involvement of the cartilagenous portions of the pinna and
the external auditory canal while sparing of the lobule. (b) Ossification of the cartilage in a patient with recurrent attacks of RP

bridge and tip of the nose. Alternatively, it may appear rapidly A routine chest radiograph can show narrowing of the tra-
in a day or so. In many patients with RP, nasal disease resolves cheal outline (Fig. 32.3a). Appreciation of this abnormality
either spontaneously or with modest degrees of treatment, with can be accentuated if a tomogram (i.e., computed tomogra-
no residual damage. In others, however, nasal disease leads to phy (CT) ) is performed (Fig. 32.3b). All RP patients should
collapse of the nasal bridge due to inflammatory changes in the have flow-volume loops performed on their first visit, with
cartilage, resulting in a saddle nose deformity (Fig. 32.2a>). follow-up comparisons every 6–12 months. Flow-volume
Because of its cosmetic implications, there is generally a loops help determine if the patient has an obstructive pattern
strong desire on the part of the patient to have a surgical to the air flow, consistent with the presence of a stricture or
repair of the nose. Surgery should be undertaken only when fixed narrowing.
the disease is under excellent control and the patient is main- In every patient who has an abnormal flow-volume loop
tained on minimal immunosuppression (Fig. 32.2b, c) (Cody study, in patients in whom the flow-volume loop study wors-
and Sones 1971). ens on follow-up, and in all patients who have respiratory
symptoms, further investigations of the airway are very
Pearl: Respiratory symptoms should always be taken
important. CT scans of the airway can show thickening of the
seriously.
airway wall, which leads eventually to airway narrowing.
Comment: Patients with RP often develop hoarseness, non- The thickening is considered a reflection of the inflammatory
productive cough, wheezing, dyspnea, or stridor. There is process (Fig. 32.4a).
frequently little correlation between the severity of symp- Cartilage calcification, if present, reflects sequelae of old
toms and the extent of laryngotrachael involvement. Most of inflammation. Dynamic CT scanning at end-inspiration and
these patients have inflammation of the cartilaginous struc- end-expiration can help evaluate the collapsibility of the air-
tures and the soft tissues surrounding them. way. Direct laryngoscopy (Fig. 32.4b) can be helpful in the
32 Relapsing Polychondritis 337

a b c

Fig. 32.2 (a) Patient with RP at the age of 16 years with collapse of the a year at this time. (c) Same patient at the age of 47 years. He had had
bridge of the nose due to destruction of the cartilage. (b) Same patient no further attacks of nasal chondritis
at the age of 19 years, after plastic repair. He was off all therapy for over

a b

Fig. 32.3 (a) Chest radiograph of a patient with RP. Note the narrowing of the trachea and main bronchi. (b) Tomogram of a patient with subglottic
stenosis
338 H. S. Luthra

a b c

Fig. 32.4 (a) CT scan of a patient with tracheal involvement in RP. Note the thickness of the tracheal wall from inflammation. (b) Direct laryn-
goscopy of a patient with RP showing subglottic stenosis. (c) Patient with RP with traschestomy in place

assessment of the upper airway, but bronchoscopy is used for

more detailed examination. Bronchoscopy should be performed
with great caution in patients with airway narrowing. Both the
patient and family must be prepared for the possibility of tra-
cheostomy in the event of a complicated procedure, because
complete or near-complete obstruction of the airway can occur
very rapidly. (Fig. 32.4c) (Staats et al. 2002; Ernst et al. 2009).
Pearl: Recurrent pneumonias or bronchiectasis may reflect a
narrowed or collapsed bronchus.
Comment: In RP, the pulmonary parenchyma is spared and
does not become involved by the disease process. However,
there are circumstances in which a patient may experience
recurrent episodes of pneumonias or give a history of a
chronic, productive cough. In such circumstances, one should
suspect the presence of a stenotic or collapsed bronchus (Fig.
32.5). In such circumstances a balloon dilation, stenting, or
even resection of that segment need to be considered.
Significant respiratory involvement should be managed
by the rheumatologist in partnership with a pulmonologist.
Pearl: Any surgical repairs of cardiovascular lesions should Fig. 32.5 3D reconstructed image of the bronchus showing the severity
of narrowing
be undertaken when disease is under control.
Comment: RP can involve the heart and great vessels. In the
heart, the aortic root dilation resulting in aortic regurgitation because of the inflammatory process that involves the media,
is the most common manifestation. Other manifestations of leading to vascular destruction.
the heart include conduction abnormalities and mitral valve In cases of significant cardiac valve disease (or dilation of
regurgitation due to mitral valve prolapse. This occurs the aorta leading to secondary valvular dysfunction), many
32 Relapsing Polychondritis 339

patients require valve replacements as well as vascular grafts. Cody DTR, Sones DA. Relapsing polychondritis: audiovestibular man-
Valve-sparing surgeries can be performed in some patients if ifestations. Laryngoscope. 1971;81:1208–22
Dib C, Moustafa SE, Mookadam M, Zehr K, Michet CJ, Mookadam F.
the process is diagnosed in sufficient time, before valvular Surgical treatment of the cardiac manifestations of relapsing poly-
damage ensues. Whenever possible, surgical therapy should chondritis. Mayo Clin Proc. 2006;81:772–6
be deferred until such time that the vascular inflammation is Ernst A, Rafeq S, Boiselle P, Sung A, Reddy C, et al Relapsing poly-
controlled, usually with glucocorticoids and often with other chondritis and airway involvement. Chest 2009;135:1024–30
Esdaile J, Hawkins D, Gold P, Freedman SO, Duguid WP. Vascular
immunosuppressive medications (Esdaile et al. 1977; Dib involvement in relapsing polychondritis. Can Med Assoc J.
et al. 2006; Barretto et al. 2002). 1977;1019–22
Jaksch-Wartenhorst R. Polychondropathia. Wien Arch F Inn Med.
1923;6:93–100
Kent PD, Michet CJ, Luthra HS. Relapsing polychondritis. Curr Opin
References Rheumatol. 2004;16:56–61
Murata J, Horii A, Tamura M, et al Endolymphatic hydrops as a cause
of audio-vestibular manifestations of relapsing polychondritis. Acta
Barretto S, Oliveira GH, Michet CJ, Nyman M, Edwards WD, Kullo IJ. Otolaryngol. 2006;126:548–52
Multiple cardiovascular complications in a patient with relapsing Staats BA, Utz JP, Michet CJ. Relapsing polychondritis. Sem Resp Crit
polychondritis. Mayo Clin Proc. 2002;77:971–4 Care Med. 2002;23:145–54
 Fibromyalgia
33
Daniel Clauw and Don L. Goldenberg

Table 33.1 FM and mood disorders

33.1 Overview of Fibromyalgia • At the time of FM diagnosis, mood disorders (primarily depres-
sion) are present in 30–50% of cases.
› Fibromyalgia (FM) is a soft tissue pain syndrome that • The increased prevalence of mood disorders is observed primarily
in tertiary-referral patients.
has been estimated to affect 4% of the U.S.
• There is an increased lifetime risk or family history of mood disorders
population. in FM compared with rheumatoid arthritis (odds ratio = 2.0).
› Precursor terms such as “neurasthenia” and “fibrosi- • FM co-aggregates with major mood disorders in families (odds
tis” were used as early as the nineteenth century. The ratio = 1.8).
controversy about whether FM is a disorder of the Adapted from Arnold et al. (2004b)
mind or of the body still rages.
› The American College of Rheumatology (ACR) crite- Psychiatric Disorders in FM,FM Non-patients, Controls
ria for the classification of FM include a history of
chronic, widespread body pain and at least 11 of 18 3.5
designated tender points on physical examination.
FM Patients
› Patients with FM often have a variety of nonspecific 3
FM Non-patients
complaints. As examples, fatigue, paresthesias, irrita-
Mean Number of Diagnoses

2.5 Controls
ble bowel complaints, subjective swelling of the hands
and feet, sleep disturbances, migraine headaches, and 2
deficits of attention and memory are reported com-
monly in FM. 1.5
› Exercise is a critical part of therapy for FM.
1

0.5

Myth: FM is a psychiatric disorder. 0

N = 64 N = 28 N = 23
Reality: There is a higher rate of psychiatric disorders in FM
patients compared with the general population. Depression Fig. 33.1 Psychiatric disorders in FM, non-FM patients, controls.
and anxiety are the disorders most commonly associated with From Aaron et al. (1996)
FM. However, FM 0is not a psychiatric illness. At the time of
the diagnosis of FM, mood disturbances are present in 30–50%
Pearl: Dividing chronic illnesses such as FM into mutually
of patients with FM (Table 33.1) (Arnold et al. 2004a, b).
exclusive mind or body disorders is harmful to patients and
Estimates of the prevalence of psychiatric disturbances
unproductive for clinicians.
among patients with FM are probably elevated artifactually
because of the fact that most studies of the disorder have Comment: There is ample evidence to suspect shared bio-
been performed in tertiary care centers. Community-dwelling logic and genetic factors in FM and depression. Furthermore,
individuals who fulfill the ACR criteria for the classification FM and mood disturbances often occur together. Higher
of FM have a much lower rate of identifiable psychiatric con- rates of psychiatric disorders are reported in all chronic pain
ditions (White et al. 2002). conditions, including headaches and idiopathic low back
In short, rheumatologists should not get off the hook sim- pain. The nature of the cause and effect relationship between
ply by regarding FM as a psychiatric disorder (Fig. 33.1). chronic pain psychiatric disorders, especially depression, is

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 341

DOI:10.1007/978-1-84800-934-9_33, © Springer Science + Business Media B.V. 2009
342 D. Clauw and D. L. Goldenberg

not clear. FM patients often become defensive if they believe remarkably consistent in different countries and cultures,
that another doctor is “telling me it’s all in my head.” about 2–4% of most populations.
Myth: FM is not the legitimate cause of chronic pain, in con- Myth: FM only occurs in women.
tradistinction to diseases such as rheumatoid arthritis.
Reality: Gender differences in pain sensitivity and tolerance
Patients with FM exaggerate their symptoms.
extend across the human species (Kim et al. 2004). These
Reality: Such statements, often muttered in frustration even gender differences influence tender point counts. Women
by rheumatologists, arise from the notion that illnesses are not are only 1.5 times more likely than men to experience chronic
real if they have no defined organ pathology. In such a sce- widespread pain, but are 10 times more likely than men to
nario, objective findings are considered to be hard, scientific have 11 or more tender points (Wolfe et al. 1995). Because of
proof of a person’s suffering. In contrast, subjective symp- this, women are about 10 times more likely to meet ACR
toms are regarded as “soft” and prone to “interpretation” by criteria for FM than are men.
the patient and physician. Assigning such a disease hierarchy Most men who have chronic widespread pain but are not
creates a second-class status for many common chronic ill- tender enough to meet criteria for FM probably have the
nesses, including FM, but also for headaches, low back pain, same underlying problem as the women who do. Such male
chronic neck and shoulder pain, chronic fatigue, and mood FM patients are often misdiagnosed as having regional mus-
disorders (Table 33.1). culoskeletal disorders (e.g., chronic low back pain or cervi-
cal osteoarthritis) and therefore are treated with inappropriate
Pearl: Physicians should accept FM symptoms as real.
medications, injections, or surgical procedures. Although
Comment: Health care professionals no longer question the men are less likely to be diagnosed with FM, the clinical
legitimacy of conditions such as depression or headaches, features and disease outcomes do not differ significantly
even though they cannot see evidence that these conditions between the sexes (Yunus et al. 2000).
are “real.” FM should be regarded in the same context as
Myth: A patient must have at least 11 tender points to be
these conditions. A biopsychologic framework fits current
diagnosed with FM.
concepts of FM and lends itself to a comprehensive, multi-
disciplinary treatment program. This model takes into Reality: The ACR criteria for FM were intended for use in
account the complex interaction of pain and stress. When patient classification for research studies, not clinical diag-
pain causes distress, individuals function less well in their nosis. Tender points are representative of heightened pain
various roles. They may have difficulties with spouses, chil- perception rather than anatomic sites of true inflammation or
dren, and work inside or outside the home. These stressors tissue pathology. Thus, they are proxies for abnormal pain
exacerbate symptoms and lead to maladaptive illness behav- processing. This number of 11 tender points that is required
iors such as social isolation, cessation of pleasurable activi- to fulfill research diagnostic criteria is totally arbitrary.
ties, and reductions in physical activity and exercise. At the time the ACR criteria were formulated, it was
In the worst-case scenario, the patient becomes entangled thought that the location of tender points might have diagnos-
with disability and compensation systems. These programs tic significance. In fact, the concept of “control points” was
almost guarantee that the patient will not improve (Hadler coined to describe areas of the body that should not be tender,
1996). If a patient senses that his/her doctor or family “does even in patients with FM. Individuals who were tender at such
not believe” him/her – a common complaint among patients control points were assumed to have a psychologic cause of
with FM – he/she may unconsciously exaggerate his/her ill- their pain. Since the ACR criteria were published, it has
ness behavior. However, there is no evidence that malinger- become clear that tenderness in FM patients extends through-
ing is any more common in FM than in other chronic medical out the entire body and is not confined inevitably to specific
disorders. body regions (albeit tenderness is more common in some
areas than in others). We now know that, relative to a non-FM
Myth: FM occurs more commonly in industrialized countries
patient, these regions such as the thumb nail and the forehead
and Western cultures.
are just as tender as areas of tender points – it is just that every-
Reality: The prevalence of FM is just as high in rural or non- one is more tender at tender point regions (Petzke et al. 2001;
industrialized societies as is in “developed” countries (Raspe Granges and Littlejohn 1993; Cohen and Quintner 1993).
1992; Peleg et al. 2008; White and Thompson 2003).
Pearl: The laying on of hands is important. Formal tender
Individuals who have greater access to health care are more
point examinations are not.
likely to present for care with symptoms of FM than are
those whose access in inferior, but FM occurs as commonly Comment: The recognition that an individual has tenderness
in Amish communities and Bedouin tribes as it does in the that is diffuse and not confined to the joints is helpful in diag-
general US population. In fact, the prevalence of FM is nosing FM. Clinicians should incorporate this approach into
33 Fibromyalgia 343

A. Widespread pain Table 33.2 Laboratory evaluation in FM

(axial + upper and In most patients:
lower + L and R
sides) Erythrocyte sedimentation rate or C-reactive protein
Complete blood count
Thyroid function tests
C. At least 4 of : In some patients: serum hepatic aminotransferases, creatine kinase
Generalized Avoid:
fatigue, Headache,
B. 11 of 18
Sleep disturbance,
Serologic screening tests (e.g., antinuclear antibodies and rheuma-
reproducible OR toid factor) clinical suspicion for a systemic connective tissue
Neuropsych,
tender points
complaints, disease exists
Numbness/tingling, Extensive neurologic testing, radiographs, or other imaging unless
IBS
clinical picture warrants
Explained by no other
condition

Because FM is not an inflammatory condition, normal acute

Fibromyalgia
phase reactants immediately provide confidence that an
Fig. 33.2 Structured interview with or without a tender point occult inflammatory disorder is unlikely.
examination for FM
Myth: All patients should have routine autoantibody testing
before the diagnosis of FM is considered “final” or
“confirmed.”
their physical examination routines, using the same degree of
pressure on all patients to get a sense of any individual Reality: Serologic studies such as antinuclear antibody (ANA)
patient’s pain threshold. This combines a general assessment and rheumatoid factor assays should generally be avoided,
of a patient’s pain threshold with a specific examination of the unless there are historic features or findings on physical
joints affected in inflammatory or noninflammatory arthritis. examination that suggest diagnoses other than FM (see Table
Some investigators have suggested guidelines to diagnose 33.2). The promiscuous use of autoantibody testing poses a
FM using a structured interview, without performing a tender major problem in clinical practice because of the high sensi-
point examination (Fig. 33.2) (Pope et al. 1991). Even if no tivity of many autoantibody assays. “False-positive” results
formal tender point examination is performed, in our experi- lead to incorrect diagnostic labels (“lupus,” “rheumatoid
ence, patients find “the laying on of hands” an essential part arthritis,” and “undifferentiated connective tissue disease”),
of the evaluation process. further unnecessary testing and, perhaps most damaging,
heightened patient anxiety (Tan et al. 1997).
Pearl: The most critical parts of the FM evaluation are the
The scope of the diagnostic evaluation for patients with
history and physical examination.
possible FM depends upon the duration of pain and other
Comment: Diagnosing FM efficiently and confidently somatic complaints, as well as the presence or the absence of
requires considerable clinical acumen and breadth of expe- specific symptoms that suggest alternative diagnoses. The
rience. FM patients often have symptoms that mimic those only patients who require serologic studies are those with
of inflammatory disease. These include fatigue, arthral- symptoms of relatively short duration (e.g., <1 year) who
gias, myalgias, stiffness, and the subjective swelling of the also have symptoms or signs that suggest an inflammatory or
hands and feet. Certain dermatologic features commonly autoimmune disorder.
seen among patients with FM (e.g., flushing over the malar
Myth: FM is due to a sleep disturbance.
region, livedo reticularis, and erythema over the digits) are
reminiscent of a malar rash or Raynaud’s phenomenon. Reality: One of the first biologic findings in FM was that
Endocrine disorders that cause unexplained fatigue and selective sleep deprivation leads to symptoms of FM in
muscles aches, most notably hypothyroidism, may mimic healthy individuals (Moldofsky et al. 1975). These findings
FM. It is reasonable to obtain thyroid function tests in the have been replicated by several groups (Older et al. 1998).
work-up for possible FM. Tests such as electromyography However, the EEG abnormalities that were noted in this
and nerve conduction velocities should be limited to patients first study and initially thought to be a marker for FM, the
with abnormal neurologic findings. so-called “alpha intrusions,” have subsequently been found to
occur in normal individuals and among patients with other
Pearl: Acute phase reactants should be ordered in patients
conditions (Branco et al. 1994; Drewes and Svendsen 1994).
suspected of FM.
Polysomnography evaluations in FM do find a variety of
Comment: The only screening tests generally recommended nonspecific findings including fewer sleep “spindles,” an
for FM are erythrocyte sedimentation rate, C-reactive pro- increase in cyclic alternating pattern rate, upper airway resis-
tein, and thyroid-stimulating hormone (TSH) (Table 33.2). tance syndrome, and poor sleep efficiency. However, it is not
344 D. Clauw and D. L. Goldenberg

clear that correction of any sleep problems identified (e.g., their symptoms. Furthermore, studies to date in FM do not
obstructive sleep apnea, upper airway resistance, and restless suggest that health care utilization is increased following the
leg or periodic limb movement syndromes) result in the diagnosis of FM (White et al. 2002). On the contrary, a recent
improvement of patients’ core FM symptoms. study in the U.K. showed significantly reduced health care
costs in the year following the diagnosis of FM, most likely
Myth: Patients given the “label” of FM begin to think they
because individuals stopped going from doctor to doctor and
have an illness. This will only make them worse.
from test to test (Annemans et al. 2008) (Fig. 33.3).
Reality: “Labeling” an individual with an illness always has Management of any illness begins with an accurate diag-
the potential to increase illness behavior. Some physicians nosis. Most FM patients report that a diagnosis, even if based
contend that the diagnosis of FM “medicalizes” everyday upon “soft” findings such as the history and physical exami-
symptoms. However, this scenario is unlikely if the FM diag- nation, reassures them that a degenerative structural disease
nosis is coupled with appropriate patient education, and a is not present. It also validates their sense that the problem is
treatment plan that emphasizes the patient’s role in improving not in their head.

a Tests and imaging b Referrals

Resource use Observed Costs Resource use Observed Costs

3,000.00 Predicted 90.00 140.00 Predicted 70.00

Observed Predicted
80.00
120.00 60.00
2,500.00 Predicted Observed
70.00
100.00 50.00
2,000.00 60.00

80.00 40.00
50.00
1,500.00
40.00 60.00 30.00

1,000.00 30.00
40.00 20.00
20.00
500.00 20.00 10.00
10.00

0.00 0.00 0.00 0.00

0.5 year 1 year 1.5 year 2 year 2.5 year 3 year 3.5 year 4 year 0.5 year 1 year 1.5 year 2 year 2.5 year 3 year 3.5 year 4 year

c GP visits d Drugs
Resource use Observed Costs Resource use Observed Costs

8,000.00 Predicted 300.00 3,000.00 Predicted 100.00

Predicted Observed 90.00

7,000.00
Observed 250.00 2,500.00 Predicted 80.00
6,000.00
70.00
200.00 2,000.00
5,000.00
60.00

4,000.00 150.00 1,500.00 50.00

40.00
3,000.00
100.00 1,000.00
30.00
2,000.00
20.00
50.00 500.00
1,000.00 10.00

0.00 0.00 0.00 0.00

0.5 year 1 year 1.5 year 2 year 2.5 year 3 year 3.5 year 4 year 0.5 year 1 year 1.5 year 2 year 2.5 year 3 year 3.5 year 4 year

Fig. 33.3 Impact on National Health Service resources and expenses (in British pounds) of making a diagnosis of FM syndrome. (a) Tests and
imaging. (b) Referrals. (c) General practitioner (GP) visits. (d) Drugs. From Annemans et al. (2008)
33 Fibromyalgia 345

Pearl: Attempting to establish links between the diagnosis of 0.4

FM and possible “etiologies” can be counterproductive.

Less Glu (post)

change in Glu/cr (post-pre)
0.2
Comment: Clinicians should avoid linking the diagnosis of
FM to putative causes, such as an injury or exposure to an 0.0
environmental toxin. In the overwhelming majority of cases,
causal associations are speculative. All too often, the affir- –0.2
mation of such an association leads to blame, inactivity, and
a sense of being victimized and litigation.
–0.4
Patients establish tentative (or firm) links in their own
Less Pain Sensitivity (post)
minds about previous exposures and the development of FM,
–0.6
usually long before they see a rheumatologist. –1.4 –1.2 –1.0 –0.8 –0.6 –0.4 –0.2 0.0 0.2 0.4 0.6
change in pressure (kg; pre-post)
Myth: Patients with FM require specialist care.
Fig. 33.4 Higher glutamate levels associated with increased FM pain.
Reality: The diagnosis and management of FM should be
From Harris et al. (2008)
approached like that of chronic headaches or chronic mood
disturbances. FM is a chronic illness and needs to be man-
aged as such. As with any chronic illness, management region in patients with FM (Fig. 33.4) (Harris 2008). The
requires the imparting of appropriate information about the insula plays a critical role in sensory integration. The ante-
diagnosis to the patient, as well as both medications and non- rior insula is linked with the emotional processing of sensa-
drug therapies. Patients must exercise, follow a good diet, tions (Tracey and Mantyh 2007). The posterior insula has a
and become educated about self-management techniques, more purely sensory role.
just as they would for the management of other disorders FM should be thought of as part of functional pain disor-
such as diabetes or high blood pressure. These illnesses are ders such as chronic fatigue syndrome, irritable bowel syn-
largely diagnosed and treated by primary care physicians. drome, chronic headaches, and chronic pelvic and bladder
Specialists should be consulted when the diagnosis is uncer- pain. Research indicates a strong familial component to the
tain or when the initial management has been unsuccessful. development of FM. First-degree relatives of individuals
More than half of all patient visits for FM are to primary with FM display an eightfold greater risk of developing this
care physicians. Specialists are involved in about 25% of condition compared with the general population (Arnold et
cases, and 16% of FM visits are to rheumatologists. The al. 2004b). Family members of individuals with FM are also
ACR suggests that the role of rheumatologists should be pri- much more likely to complain of tenderness on examination
marily as consultants to primary care physicians. With an than are the family members of healthy controls, regardless
estimated 6–10 million FM patients in the U.S., rheumatolo- of whether they have pain. Family members of FM patients
gists cannot possibly accommodate the longitudinal care of are also much more likely to have irritable bowel syndrome,
all such patients. Rheumatologists should advise primary temporomandibular dysfunction, headaches, and a host of
care physicians with regard to team management, which other regional pain syndromes (Buskila et al. 1996; Hudson
often will include other specialists such as mental health pro- et al. 1985; Kato et al. 2006a, b).
fessionals, physiatrists, and pain management experts. This familial and personal co-aggregation of painful con-
The close association of FM with mood disturbances man- ditions was originally termed affective spectrum disorder
dates that every new FM patient be carefully evaluated for (Hudson et al. 1993). More recently, the term central sensi-
comorbidity of psychiatric illness. Even in patients with no tivity syndromes and chronic multisymptom illnesses have
psychiatric diagnosis, levels of psychosocial stressors, coping been employed (Yunus, 2008; Fukuda et al. 1998). In popu-
ability, and locus of control should be ascertained. Often this lation-based studies, the key symptoms (in addition to pain)
requires formal referral to a mental health professional. that often co-aggregate are fatigue, memory difficulties, and
Myth: There is no reasonable pathophysiologic explanation mood disturbances (Fukuda et al. 1997, 1998). Twin studies
that accounts for the multitude of symptoms in FM. suggest that about half of the risk of developing chronic
widespread pain can be attributed to genetic factors, and the
Reality: There is substantial evidence that FM is associated other half to environmental influences (Kato et al. 2006b).
with abnormalities in neural pain regulation. These data are Certain genetic polymorphisms are associated with a
consistent with the notion that FM and related syndromes higher risk of developing FM. To date, the serotonin 5-HT2A
represent biologic amplification of all sensory stimuli receptor polymorphism T/T phenotype, serotonin trans-
(Gracely et al. 2002). Functional imaging studies indicate porter, dopamine 4 receptor, and catecholamine o-methyl
that the insula is the most consistently hyperactive brain transferase (COMT) polymorphisms have all been detected
346 D. Clauw and D. L. Goldenberg

in a higher frequency among patients with FM (Bondy et al. Table 33.3 Typical outcome in FM
1999; Offenbaecher et al. 1999; Buskila et al. 2004; Buskila, FM does not herald the onset of a systemic disease
2007). All of the polymorphisms identified to date involve There is no progressive, structural or organ damage
the metabolism or transport of monoamines, compounds that Most patients in specialty practice have chronic, persistent
symptoms (Felson)
play a critical role in activity of the human stress response.
Primary care patients more commonly report complete remission of
Multiple genetic polymorphisms probably help determine symptoms and 50% no longer had FM after 1 year (Granges)
individuals’ “set point” for pain and sensory processing. Most patients continue to work, but 10–15% are disabled (Felson)
There is often adverse impact on work and leisure activities
Pearl: When considering the diagnosis of FM, ask patients if
Most patients’ quality of life improves with medical management
they are sensitive to noises, odors, and other sensory
Summarized from Felson and Goldenberg (1986); Granges et al. (1994)
stimuli.
Comment: FM might be related to a generalized disturbance diagnosed with FM in the community often have spontaneous
in the processing of sensory information. FM patients are remission of their symptoms (Table 33.3). Nearly 50% of FM
hyperresponsive not only to painful stimuli, but also to other patients in primary care no longer had symptoms of FM 1
sensory stimuli. For example, FM patients display a low nox- year after the initial diagnosis (Granges et al. 1994). Even in
ious threshold to auditory tones (Gerster and Hadj-Djilani tertiary referral clinics dedicated to FM, most patients con-
1984; McDermid et al. 1996). Yunus coined the term “cen- tinue to work full time and nearly two-thirds report feeling
tral sensitivity syndrome” to describe illnesses such as FM, well or very well (Felson 1986). The notion that FM carries a
irritable bowel syndrome, tension headache, and other disor- poor prognosis whether diagnosed in the community or in
ders than tend to cluster in the same patients (Yunus 2008). specialty clinics contributes to pessimism on the part of both
One study used an identical random staircase paradigm to patients and clinicians. In the majority of patients, FM
test FM patients’ thresholds to the loudness of auditory tones responds favorably to a combination of pharmacologic and
and to pressure (Geisser et al. 2007). In that study, FM nonpharmacologic therapy (Table 33.3).
patients displayed low thresholds to both types of stimuli.
Pearl: Emphasize a “disease management” approach to
Moreover, the correlation between the results of auditory and
FM.
pressure pain threshold testing suggested that some of these
differences were due to shared variance to the different stim- Comment: Optimal management of FM requires multidisci-
uli and that other portions were unique to one stimulus or the plinary health care team. It is useful in discussions with
other. patients to compare their problem with diabetes. If they had
been diagnosed with diabetes rather than FM, they would
Pearl: Patients with FM can understand their disorder as
understand that they would not simply be given a prescrip-
one of “increased volume control” in which their central
tion for insulin and syringes and expect that this would man-
nervous system processes pain and other sensory stimuli in
age their illness. Rather, they would receive education about
an altered (do not say aberrant!) manner.
their diagnosis, they would learn of the importance of exer-
Comment: Pain and sensory processing are physiologic pro- cise and following a diet, and other information. In FM, the
cesses in which large differences exist among individuals patient must also take an active role in their overall disease
across the population. In this way, pain and sensory process- management. The patient and primary care provider should
ing are similar to blood pressure, glucose metabolism, or any be at the helm of that team. Specialists, including rheuma-
other physiologic process. Ample data suggest that the fun- tologists, physiatrists, mental health professionals, pain man-
damental neurobiologic problem in FM and related syn- agement specialists, physical therapists, and occupational
dromes is that patients perceive more pain from a certain therapists also often participate in the care of these patients.
amount of pressure or heat applied to their skin than do indi- Specialists can provide important education to FM
viduals without FM. As noted above, this sensitivity extends patients. This teaching can be accomplished in either small
to other sensory processes. Patients should be informed that or large group settings. Rheumatologists or physiatrists are
pain and other sensory symptoms do not relate necessarily to ideal specialists to develop such educational sessions, either
something “wrong” with the symptomatic part of their body, as part of office practice or as a stand-alone education pro-
but rather to their underlying sensitivity. gram. The patient’s spouse and immediate family members
should be encouraged to attend.
Myth: FM is an intractable illness and a carries a poor prog-
nosis. There is no effective therapy. Myth: Antidepressants work in FM because most FM patients
are depressed.
Reality: The medical literature (especially older literature)
often depicts FM as an intractable illness for which very lim- Reality: Two new serotonin-norepinephrine reuptake inhibitors
ited therapeutic options exist. On the contrary, patients (SNRIs), milnacipran and duloxetine, have undergone recent
33 Fibromyalgia 347

multicenter trials and were shown to be effective in a number of 35

outcome variables (Vitton et al. 2004; Arnold et al. 2004a). In 30 Exercise

these trials, significant differences were reported in such param- 25 Control

eters as overall improvement, physical functioning, fatigue lev- 20

%Change
els, degree of reported physical impairment, self-reported pain 15
10
and stiffness, and the number of tender points. With both of
5
these drugs and more broadly in the pain field, the analgesic
0
and other benefits of the drugs were equal in depressed and
–5
nondepressed patients. These findings suggest that the anal-
–10
gesic and other positive effects of tricyclics and SNRIs in FM –15
are not simply due to their antidepressant effects. Aerobic Mean Tender Point
Pain Intensity
Perfromance PainThreshold
Pearl: Cognitive behavioral therapy is effective in the treat- statistically significant
ment of FM.
Fig. 33.5 Improvements in aerobic exercise vs. nonexercise controls
Comment: The two forms of nonpharmacologic therapy stud-
ies most thoroughly to date are cognitive behavioral therapy Women with FM gain the same physiologic benefits from
and exercise. Both of these therapies are efficacious in the exercise, including cardiovascular fitness and strength, as do
treatment of FM, as well as a plethora of other medical con- healthy women. In most patients, it is best to initiate exercise
ditions (Williams et al. 2000; Goldenberg et al. 2004). Both a few months after the start of drug therapy so that the patient
of these treatments can lead to sustained improvements when is less symptomatic and able to tolerate the exercise program
an individual complies with therapy. better. Any cardiovascular exercise should be initiated slowly
and incrementally, with a target goal of 30–40 min of aerobic
Pearl: Start low, go slow when initiating new therapy for
cardiovascular fitness training 3 or 4 times weekly. Exercise
FM.
programs of moderate intensity (55–75% of age-adjusted
Comment: Perhaps because of their generalized sensory maximal heart rate) are usually tolerated well, but high-
hypersensitivity, patients with FM are much more likely to intensity (heart rate >150/min) aerobic exercise is not.
experience adverse effects of drugs than patients without
FM. Thus, it is often quite helpful to begin at very low doses Pearl: Describe exercise as a “drug” to patients.
of drugs, and to escalate the dose very slowly. Comment: Exercise likely works in part in FM by raising
Pearl: Patients on selective serotonin reuptake inhibitors levels of serotonin and norepinephrine. This intervention is
may require higher doses than those necessary to treat most effective if administered in low, frequent doses several
depression. times a week rather than high doses taken intermittently.
Clinicians caring for patients with FM need to mount the
Comment: Studies that have demonstrated the efficacy of exercise soapbox and ask patients continually how much
SSRIs in FM have either combined the drug with a low dose of activity and exercise they undertake. Patients often resist
a tricyclic medication (e.g., amitriptyline) or used much higher beginning exercise programs for years, but perseverance on
doses of the SSRI than is often used for depression (Goldenberg the part of the clinician pays off. Once patients become “sick
et al. 1996; Arnold et al. 2002). For example, the dose of flu- and tired of being sick and tired” and begin to exercise, they
oxetine necessary to treat FM in one of the above studies was usually continue this as part of their treatment program.
45 mg/day, much higher than the average dose used as an anti- Exercise should be considered similar to any medication
depressant. These higher doses of SSRIs might achieve their and explained to the patient in detail. Make a show of writing
outcomes through the recruitment of noradrenergic effects. the patient a prescription for exercise. Referral of the patient
Thus, although there is some evidence that SNRIs may be to a physical therapist can be helpful, especially for patients
preferable to SSRIs in the treatment of FM, if SSRIs are used who are reluctant to begin an exercise program.
they may need to be used at higher doses.
Myth: Exercise makes people with FM worse, such that they Pearl: Ask patients if they had been were active before devel-
are never able to comply. oping FM.

Reality: Several studies have demonstrated that cardiovascu- Comment: Several studies suggest that FM patients have
lar exercise is helpful in FM (Fig. 33.5). A large meta-analy- higher levels of premorbid activity and exercise than do con-
sis of about 40 clinical trials demonstrated that exercise trols. These data suggest that individuals learned early in life
improves physical status, daily function, and FM symptoms that regular exercise made them feel better. Pointing this out
(Busch et al. 2007). to patients often assists in getting individuals to begin an
348 D. Clauw and D. L. Goldenberg

exercise program. Exercise has a salutatory effect on pain, Fukuda K, et al An epidemiologic study of fatigue with relevance for
fatigue, and other symptoms, even in healthy individuals. the chronic fatigue syndrome. J Psychiatr Res. 1997;31:19–29
Geisser ME, et al The association between experimental and clinical
Myth: When patients experience a flare of their FM symp- pain measures among persons with fibromyalgia and chronic fatigue
syndrome. Eur J Pain. 2007;11:202–07
toms, switch them to a new therapy. Gerster JC, Hadj-Djilani A. Hearing and vestibular abnormalities in pri-
Reality: FM flares occur most often because of stressors such mary fibrositis syndrome. J Rheumatol. 1984;11:678–80
Goldenberg DL, Burckhardt C, Crofford L. Management of fibromyal-
as psychologic pressure, infections, trauma, and other events. gia syndrome. JAMA 2004;292:2388–95
In these settings, the discontinuation of current therapy can Goldenberg D, Mayskiy M, Mossey C, et al A randomized, double-
be counterproductive. However, when patients are consis- blind crossover trial of fluoxetine and amitriptyline in the treatment
tently doing poorly, it is reasonable to consider switching of fibromyalgia. Arthritis Rheum. 1996;39:1852–59
Gracely RH, Petzke F, Wolf JM, Clauw DJ. Functional magnetic reso-
medications or adding a different drug. Whatever initial drug nance imaging evidence of augmented pain processing in fibromy-
is chosen should be titrated to an adequate therapeutic dos- algia. Arthritis Rheum. 2002;46:1333–43
age for an adequate duration before deciding on response. If Granges G, Littlejohn G. Pressure pain threshold in pain-free subjects,
no response occurs consider switching to a different antide- in patients with chronic regional pain syndromes, and in patients
with fibromyalgia syndrome. Arthritis Rheum. 1993;36:642–6
pressant or using combinations, such as an antidepressant in Granges G, Zilko P, Littlejohn GO. Fibromyalgia syndrome: assess-
the morning with either gabapentin or pregabalin at night. ment of the severity of the condition 2 years after diagnosis.
J Rheumatol. 1994;21:523–29
Hadler NM. If you have to prove you are ill, you can’t get well. The
object lesson of fibromyalgia. Spine 1996;21:2397–400
Harris R, Sundgren P, Pang Y, et al Dynamic levels of glutamate within
References the insula are associated with improvements in multiple pain
domains in fibromyalgia. Arthritis Rheum. 2008;58:903–7
Hudson JI, Hudson MS, Pliner LF, et al Fibromyalgia and major affec-
Aaron LA, et al Psychiatric diagnoses in patients with fibromyalgia are tive disorder: a controlled phenomenology and family history study.
related to health care-seeking behavior rather than to illness [see Am J Psychiatr. 1985;142:441–6
comments]. Arthritis Rheum. 1996;39:436–45 Hudson JI, Goldenberg DL, Pope HGJ, et al Comorbidity of fibromyalgia
Annemans L, Wessely S, Spaepen E, et al Health economic conse- with medical and psychiatric disorders. Am J Med. 1993;92: 363–7
quences related to the diagnosis of fibromyalgia syndrome. Arthritis Kato K, Sullivan PF, Evengard B, Pedersen NL. Chronic widespread
Rheum. 2008;58:895–902 pain and its comorbidities: a population-based study. Arch Intern
Arnold LM, et al A double-blind, multicenter trial comparing dulox- Med. 2006a;166:1649–54
etine with placebo in the treatment of fibromyalgia patients with or Kato K, Sullivan PF, Evengard B, Pedersen NL. Importance of genetic
without major depressive disorder. Arthritis Rheum. 2004a;50: influences on chronic widespread pain. Arthritis Rheum. 2006b;
2974–84 54:1682–6
Arnold LM, et al A randomized, placebo-controlled, double-blind, Kim H, Neubert JK, San Miguel A, et al Genetic influence on variabil-
flexible-dose study of fluoxetine in the treatment of women with ity in human experimental pain sensitivity associated with gender,
fibromyalgia. Am J Med. 2002;112:191–7 ethnicity and psychological temperament. Pain 2004;109:488–96
Arnold LM, et al Family study of fibromyalgia. Arthritis Rheum. McDermid AJ, Rollman GB, McCain GA. Generalized hypervigilance
2004b;50:944–52 in fibromyalgia: evidence of perceptual amplification. Pain 1996;66:
Bondy B, et al The T102C polymorphism of the 5-HT2A-receptor gene 133–44
in fibromyalgia. Neurobiol Dis 1999; 6: 433–439 Moldofsky H, Scarisbrick P, England R, Smythe H. Musculoskeletal
Branco J, Atalaia A, Paiva T. Sleep cycles and alpha-delta sleep in symptoms and non-REM sleep disturbance in patients with “fibrosi-
fibromyalgia syndrome. J Rheumatol. 1994;21:1113–7 tis syndrome” and healthy subjects. Psychosomatic Med. 1975;
Busch AJ, Barber KA, Overend TJ, et al Exercise for treating fibromy- 37:341–51
algia syndrome. Cochrane Database Syst Rev. 2007;4 Offenbaecher M, et al Possible association of fibromyalgia with a poly-
Buskila D, Neumann L, Hazanov I, Carmi R. Familial aggregation morphism in the serotonin transporter gene regulatory region.
in the fibromyalgia syndrome. Semin Arthr Rheum. 1996;26: Arthritis Rheum. 1999;42:2482–8
605–11 Older SA, et al The effects of delta wave sleep interruption on pain
Buskila D, Cohen H, Neumann L, Ebstein RP. An association between thresholds and fibromyalgia-like symptoms in healthy subjects;
fibromyalgia and the dopamine D4 receptor exon III repeat poly- correlations with insulin-like growth factor I. J Rheumatol 1998;25:
morphism and relationship to novelty seeking personality traits. 1180–6
Mol Psychiatry. 2004;9:730–1 Peleg R, et al Characteristics of fibromyalgia in Muslim Bedouin
Buskila D. Genetics of chronic pain states. Best Pract Res Clin women in a primary care clinic. Semin Arthritis Rheum. 2008;37(6):
Rheumatol. 2007;21:535–47 398–402
Cohen ML, Quintner J. Fibromyalgia syndrome: a problem of tautol- Petzke F, et al Dolorimetry performed at 3 paired tender points highly
ogy. Lancet 1993;342:906–9 predicts overall tenderness. J Rheumatol 2001;28:2568–69
Drewes AM, Svendsen L. Quantification of alpha-EEG activity during Pope HGJ, Hudson JI. A Supplemental interview for forms of "affective
sleep in fibromyalgia: a study based on ambulatory sleep monitor- spectrum disorder". International Journal of Psychiatry in Medicine
ing. J Musculoskel Pain. 1994;2:33–53 1991; 21(3): 205–32
Felson DT, Goldenberg DL. The natural history of fibromyalgia. Raspe H. [Rheumatism epidemiology in Europe]. Soz Praventivmed.
Arthritis Rheum. 1986;29:1522–6 1992;37:168–78
Fukuda K, et al Chronic multisymptom illness affecting Air Force vet- Tan EM, et al Range of antinuclear antibodies in “healthy” individuals.
erans of the Gulf War. JAMA 1998;280:981–8 Arthritis Rheum. 1997;40:1601–11
33 Fibromyalgia 349

Tracey I, Mantyh PW. The cerebral signature for pain perception and its Williams DA, Cary MA, Glazer LJ, et al Randomized controlled trial of
modulation. Neuron. 2007;55:377–91 CBT to improve functional status in fibromyalgia. Arthritis Rheum.
Vitton O, Gendreau M, Gendreau J, et al A double-blind placebo-con- 2000;43:S210
trolled trial of milnacipran in the treatment of fibromyalgia. Hum Wolfe F, Ross K, Anderson J, et al Aspects of fibromyalgia in the gen-
Psychopharmacol. 2004;19(Suppl 1):S27–35 eral population: sex, pain threshold, and fibromyalgia symptoms.
White KP, Nielson WR, Harth M, et al Does the label “fibromyalgia” J Rheumatol. 1995;22:151–6
alter health status, function, and health service utilization? A pro- Yunus MB Central sensitivity syndromes: a new paradigm and group
spective, within-group comparison in a community cohort of adults nosology for fibromyalgia and overlapping conditions, and the
with chronic widespread pain. Arthritis Rheum. 2002;47:260–5 related issue of disease versus illness. Semin Arthritis Rheum.
White KP, Thompson J. Fibromyalgia syndrome in an Amish commu- 2008;37(6):339–52
nity: a controlled study to determine disease and symptom preva- Yunus MB, Inanici F, Aldag JC, et al Fibromyalgia in men: comparison
lence. J Rheumatol. 2003;30:1835–40 of clinical features with women. J Rheumatol. 2000;27:485–90
 The Clinical Features of Gout
34
Bruce N. Cronstein and Michael H. Pillinger

crystals do not induce inflammatory responses, for reasons

34.1 Overview of the Clinical that are not fully understood (Bomalaski et al. 1986). Thus,
Features of Gout the finding of only extracellular crystals in a joint aspirate
raises the possibility that extracellular crystals are not the
› Gout is caused by the crystal deposition of monoso- “primum mobile” of the inflammatory response. Other poten-
dium urate in and around joint tissues. tial etiologies for the presence of PMNs (e.g., infection) must
› The first joint affected in the majority of cases is the be excluded.
first metatarsophalangeal joint, a condition known as Myth: A therapeutic response to colchicine confirms that the
podagra. cause of acute monoarticular arthritis was gout.
› Acute gout is characterized by the rapid development
of warmth, swelling, erythema, and pain in the affected Reality: The major mechanism of action of colchicine is the
joint. disruption of microtubules. The consequence of this is the
› The natural history of gout is to worsen with time, pro- impairment of the ability of PMNs to infiltrate into joints.
gressing through the stages of asymptomatic hyperuri- Other potential mechanisms of action for colchicine include
cemia and acute intermittent gouty attacks to tophaceous effects on adhesion molecules and interleukin-I (Molad et al.
gout. 1998; Cronstein and Terkeltaub 2006). These effects make
› The time period between the first development of acute colchicine a useful drug for acute gouty attacks. In the setting
intermittent gouty attacks and the appearance of chronic of a high suspicion of acute gout, a response to colchicine sup-
tophaceous gout is approximately a decade in most ports the diagnosis.
cases. However, colchicine also inhibits inflammatory responses
› The development of tophaceous deposits of monoso- in other conditions, including pseudogout, familial Mediterra-
dium urate is a function of the degree and duration of nean fever, and leukocytoclastic vasculitis (Famaey 1988;
hyperuricemia. Ben-Chetrit and Levy 1998; Chen and Carlson 2008). Pseudog-
› However, most people with hyperuricemia never develop out is the disorder most likely to mimic gout, and the two can-
clinical gout. not be rigorously distinguished by a therapeutic response to
colchicine. Because the long-term management strategies of
these two crystal diseases differ, establishing the definitive
diagnosis of one or the other by joint aspiration and crystal
examination is important.
Myth: The presence of polymorphonuclear neutrophils
(PMNs) and EXTRACELLULAR urate crystals in a synovial Myth: A normal serum uric acid level at the time of acute
fluid aspirate is pathognomonic of acute gout. monoarticular arthritis excludes the possibility of gout.
Reality: The sine qua non of an acute gouty attack is the pres- Reality: Hyperuricemia, defined as chronic elevation of the
ence in synovial fluid of PMNs whose response is directed serum uric acid level above 6.8 mg/dL, is the sine qua non for
against urate crystals. Examination of the joint fluid in a true the development of the condition of gout. However, serum uric
gout attack usually demonstrates urate crystals both within acid levels fluctuate over time and are not always elevated at
PMNs and in the extracellular fluid. Only the presence of the time of an acute attack. Acute attacks of gout can be caused
intracellular crystals provides unequivocal evidence that the by the liberation of preformed urate crystals from tophaceous
PMN attack is directed toward the crystals. or microtophaceous deposits within joints, bursae, skin, and
Individuals with a history of gout can have “residual” extra- other organs. In such cases, an elevated uric acid is not required
cellular crystals in the joint between gouty attacks. These “old” at the time of the attack.

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 351

DOI:10.1007/978-1-84800-934-9_34, © Springer Science + Business Media B.V. 2009
352 B. N. Cronstein and M. H. Pillinger

A patient with a negative Gram stain should not be

presumed to be infection-free based on that result alone.
Indeed, in patients whose joint fluid cell count approaches or
exceeds 100,000 cells/mm3, it is often wise to assume the
patient has an infection until the results of synovial fluid
cultures are available.
Pearl: A 24-h urine collection for uric acid should not be
performed during or immediately after an acute gout attack.
Comment: The quantity of uric acid present in a 24-h collec-
tion can provide insight into the etiology of the patient’s
hyperuricemia:
• In a patient with relatively normal glomerular filtration
rate and hyperuricemia, the excretion of more than 800 mg
of uric acid in a 24-h period indicates metabolic overpro-
Fig. 34.1 Tophaceous deposits within synovial tissue. This patient had a duction of urate. In other words, the excretion of urate is
clinical course complicated by both fungal arthritis (Candida tropicalis)
and chronic gouty arthropathy in the knee. He underwent an orthopedic insufficient to keep pace with production. Such a patient
procedure to help eradicate the fungal infection. The synovial pathology is said to be an “overproducer.”
revealed extensive tophaceous deposits. The patient’s fungal infection • On the other hand, hypoexcretion of uric acid (e.g., less
was believed to have led to continuous “crystal stripping” from the syn- than 600 mg in a 24-h period) suggests a failure of the renal
ovium (Figure courtesy of Dr. John Stone)
tubular system to excrete a normal load of urate. Such
patients are termed “underexcreters.”
A common scenario on the Inpatient Rheumatology Con-
sult Service is as follows: (1) Elderly patient with or without Classification of a patient as an “overproducer” or “underex-
a recognized history of gout is admitted for elective surgery; creter” can be useful in selecting the appropriate urate-lowering
(2) The patient is made NPO in anticipation of surgery; (3) therapy. In the absence of contraindications to one drug or
The serum uric acid level declines acutely; and (4) Uric acid the other, allopurinol would be an appropriate choice for an
crystals are leeched from tophaceous deposits within joint overproducer and probenicid a reasonable selection for an
synovium, triggering an acute attack of gout (Fig. 34.1). underexcreter.
Individuals who begin a xanthine oxidase inhibitor (e.g., Many aspects of urate handling are in flux around the time
allopurinol) or uricosuric agent (e.g., probenicid) are actually of an acute gout attack. Cytokines associated with the attack
at higher risk of an acute gout flare during the first 6 months can increase the degree of urate excretion. Conversely, the
of therapy (Becker et al. 2005) (for this reason, colchicine presence of acidosis or the recent institution of a diuretic can
should be administered along with the urate-lowering agent lead to decreased urate excretion. Under these circumstances,
for the first 6 months). In addition, some data suggest that the a 24-h collection is unlikely to reflect accurately the patient’s
inflammatory response that accompanies acute gout induces baseline urate metabolism and excretion. Clinicians wishing
a urate diuresis, transiently leading to a lowering of the serum to assess urate excretion should not obtain a 24-h urine col-
urate level (Urano et al. 2002). Thus, a normal serum urate lection until the intercritical period is reached, several weeks
level during a flare of acute joint disease does not exclude after an acute gout flare.
gout as the cause. Pearl: Subchondral defects with “overhanging edges” are
Myth: A negative Gram stain excludes a joint infection and virtually pathognomonic of tophaceous deposits on radio-
implicates a microcrystalline disorder as the cause of joint graphs of the hands or feet.
inflammation.
Comment: In assessing a patient for the diagnosis of chronic
Reality: Gram stain should be performed on any joint fluid in gout, one radiographic finding is highly characteristic: the
which infection is considered as an alternative to acute gout punched-out lesion with an overhanging edge (Fig. 34.2)
or another microcrystalline disorder (this, of course, includes (Choi et al. 2006). Subchondral defects with classic over-
essentially all cases). A positive Gram stain strongly sup- hanging edges confirm the presence of small tophaceous
ports the diagnosis of infection and frequently provides deposits consistent with chronic gout.
important clues to the nature of the offending organism. These lesions typically are located in a juxta-articular
However, the sensitivity of Gram stain is limited. As many as region, at either the distal or proximal ends of bones (e.g.,
55% of patients with infected joints have negative Gram near the metacarpophalangeal or interphalangeal joints).
stains of synovial fluid (Faraj et al. 2002). Gouty erosions often have a cystic quality to them and can be
34 The Clinical Features of Gout 353

the knee) is easily “tappable,” and if the result of the synovial

fluid analysis would alter the decision about whether or not
to begin longitudinal treatment.
Pearl: The occurrence of acute monoarticular arthritis in a
premenopausal woman who is otherwise well is almost never
gout.
Comment: Clinicians who are consulted to “rule out gouty
attack” in a premenopausal woman with joint pain should
recall the wisdom of Hippocrates, who observed that “A
woman does not take the gout, unless her menses be stopped”
(Hippocrates 1886).
Premenopausal women normally have persistently low
serum urate levels because of a high rate of renal excretion.
Thus, they lack the conditions necessary to develop gout.
Indeed, the prevalence of gout in premenopausal women is
on the order of only 2 cases per 10,000 (Centers for Disease
Control 1996). Premenopausal women who develop gout
almost certainly have one or more well-defined risk factors
(e.g., diuretic use or renal failure).
Pearl: Gout is a hereditary disease.
Fig. 34.2 Radiograph of a chronic gouty arthropathy, characterized by Comment: Most clinicians pay little attention to the family
joint erosions with overhanging edges (Figure courtesy of Dr. John Stone)
history when evaluating a patient with gout. However, it has
been known for at least a century that the risk of gout can be
confused with the kind of bone cysts found in osteoarthritis. conveyed in families. A family history of gout is identified
However, they differ from such cysts in that their cortical approximately 40% of patients with this disease. For the pur-
“roof” has been broken, leading to the overhanging edge. poses of initial diagnosis, the specific gene in question is not
Gouty lesions therefore often appear to have an open lid. important, but a positive family history of gout can provide
Many other radiologic features of gout are nonspecific. As strong supporting evidence for the diagnosis of gout as the
examples, gouty erosions may be indistinguishable from course of an evaluation is underway. The precise genetic
those of rheumatoid arthritis or osteomyelitis. In addition, bases for the risk of gout are understood only partly. Pertinent
although tophi are not usually radio-opaque, they may become known genes include hypoxanthine–guanine phosphoribosyl
so if they undergo secondary calcification. transferase, phosphoribosyl pyrophosphate synthase, fruc-
tose-1-phosphatase aldolase, and the URAT-1 renal trans-
Pearl: The diagnosis of gout can often be confirmed in an
porter, among others.
asymptomatic patient by aspirating a previously involved
joint and examining the synovial fluid for uric acid crystals. Pearl: Gout and hyperuricemia are strongly associated with
other chronic diseases, including obesity, hypertension, and
Comment: Urate crystals persist in the joint during the inter-
coronary artery disease.
critical phase in the great majority of patients with gout (per-
haps all patients with gout) (Bomalaski et al. 1986; Pascual Comment: Gout or hyperuricemia often occur as part of a
et al. 1999). For reasons that remain unclear, these “old” constellation of other problems. The high purine intake asso-
crystals are no longer inflammatory. ciated with hyperuricemia in most cases contributes to obe-
A common scenario in the rheumatology clinic is the sity, a condition common among patients with gout (Annemans
arrival of a patient alleged to have gout who is no longer et al. 2008). In addition, the prevalence of hypertension
symptomatic. In the usual situation, the patient has been among patients with gout may be as high as 40%. Some data
managed empirically 2 weeks before the clinic appointment from both animal and human studies suggest that hyperurice-
during an acute presentation, without the performance of an mia is a direct cause of high blood pressure (Edwards 2008).
arthrocentesis to confirm the presence of monosodium urate Finally, myocardial infarction is a common cause of morbid-
crystals. ity in patients with gout.
Aspiration of the previously affected but no longer symp- The true relationship between hyperuricemia and clinical
tomatic joint can be considered in this setting. Moreover, this entities that are associated clearly with the metabolic syn-
procedure is strongly indicated if the joint in question (e.g., drome requires further study. However, from the clinical
354 B. N. Cronstein and M. H. Pillinger

standpoint, the diagnosis of gout is a marker for a patient who Becker MA, Schumacher HR Jr, Wortmann RL, MacDonald PA,
may benefit from a number of interventions in addition to the Eustace D, Palo WA, Streit J, Joseph-Ridge N. Febuxostat compared
with allopurinol in patients with hyperuricemia and gout. N Engl
direct treatment of hyperuricemia complications. Weight loss, J Med. 2005;353:2450–61
blood pressure treatment, and other strategies designed to Ben-Chetrit E, Levy M. Colchicine: 1998 update. Semin Arthritis
reduce the risk of cardiovascular disease may be appropriate. Rheum. 1998;28(1):48–59
Bomalaski JS, Lluberas GR, Schumacher HR Jr. Monosodium urate
Pearl: Gout often complicates a joint that is infected, and crystals in the knee joints of patients with asymptomatic nontopha-
vice versa. ceous gout. Arthritis Rheum. 1986;29(12):1480–4
Centers for Disease Control and Prevention, national Center for Health
Comment: Gout is often a sufficient explanation for inflam- Statistics. Vital and health statistics: current estimates from the
mation in a joint, particularly if intracellular urate crystals national Health Interview Survey, 1996. Series 10, No. 200. Atlanta:
Department of Health and Human Services (US). Available at http://
are demonstrated within synovial fluid neutrophils. However, www.cdc.gov/nchs/data/series/sr_10/sr10_200.pdf
the positive finding of crystals does not exclude a concurrent Chen KR, Carlson JA. Clinical approach to cutaneous vasculitis. Am
problem. Gout probably leads to an increased risk of joint J Clin Dermatol. 2008;9(2):71–92
infections in the same way that rheumatoid arthritis does: the Choi MH, MacKenzie JD, Dalinka MK. Imaging features of crystal-
induced arthropathy. Rheum Dis Clin N Am. 2006;32:427–46
existence of baseline structural damage, the presence of an Cronstein BN, Terkeltaub R. The inflammatory process of gout and its
effusion, and alterations in the local immune response. treatment. Arthritis Res Ther. 2006;8(Suppl 1):S3
Conversely, joint infections are proposed to result in the Edwards NL. The role of hyperuricemia and gout in kidney and cardio-
liberation of monosodium urate crystals sites of synovial vascular disease. Cleve Clin J Med. 2008;75(Suppl 5):S13–6
Famaey JP. Colchicine in therapy. State of the art and new perspectives
deposition – a phenomenon known as “crystal stripping” – for an old drug. Clin Exp Rheum. 1988;6(3):305–17
and secondary gout attacks. One study revealed the preva- Faraj AA, Omonbude OD, Godwin P. Gram staining in the diagnosis of
lence of infection to be low in the setting of known crystal acute septic arthritis. Acta orthop Belg. 2002;68(4):388–91
disease, but reported a substantial increase (from 1.5% to 22%) Hippocrates. The genuine works of Hippocrates, volumes 1 and 2.
Translated and edited by Adams F. New York: Wood; 1886
when only joints with synovial fluid counts ≥100,000/mm3 Molad Y, Cronstein BN, Malawista SE. Colchicine. In: Smyth CJ,
were considered (Shah et al. 2007). Given the potentially Holers VM, editors. Gout, hyperuricemia, and other crystal-associ-
disastrous consequences of an untreated septic joint, even the ated arthropathies. New York: Marcel Dekker; 1998. p. 193–204
lower value is concerning. Thus, the diagnosis of acute gout Pascual E, Batlle-Gualda E, Martinez A, Rosas J, Vela P. Synovial fluid
analysis for diagnosis of intercritical gout. Ann Intern Med.
attack should not cause the clinician to disregard the possi- 1999;131:756–9
bility of infection. Shah K, Spear J, Nathanson LA, McCauley J, Edlow JA. J Emerg Med.
2007;32(1):23–26
Urano W, Yamanaka H, Tsutani H, Nakajima H, Matsuda Y, Taniguchi
A, Hara M, Kamatani N. The inflammatory process in the mecha-
References nism of decreased serum uric acid concentrations during acute gouty
arthritis. J Rheumatol. 2002;29:1950–3

Annemans L, Spaepen E, Gaskin M, Bonnemaire M, Malier V, Gilbert

T, Nuki G. Gout in the UK and Germany: prevalence, comorbidities
and management in general practice 2000–2005. Ann Rheum Dis.
2008;67:960–6
 Gout Epidemiology
35
Hyon K. Choi

of physician-diagnosed gout. However, even if the true age-

35.1 Overview of Gout Epidemiology specific prevalence of gout are 50% lower than these estimates,
they still rival or exceed that of rheumatoid arthritis among
› The prevalence of gout rises with advancing age. women (2.3% in the NHANES III) (Rasch et al. 2003).
Among men and women older than 80 years of age, the Moreover, data from the Rochester Epidemiology project indi-
prevalence of the disorder is 9% and 6%, respectively. cate a doubling in the incidence of primary gout among women
› Gout is more common among individuals with lower between the years 1977 and 1996 (Arromdee et al. 2002).
family income levels. This is probably attributable to a
Myth: Gout occurs more often among African-Americans
disproportionate occurrence within those income strata
than Caucasians.
of other risk factors for gout, including hypertension,
obesity, and dietary patterns. Reality: A prospective study of 352 Black men in the Meharry
› Ethanol or fructose consumption increases uric acid Cohort Study and 571 White men in the Johns Hopkins
production by net adenosine triphosphate (ATP) degra- Precursors Study investigated this issue. Based upon self-
dation to adenosine monophosphate (AMP). AMP is reported cases of gout, the study found that the risk of gout was
degraded rapidly to uric acid. 70% higher among the Black men compared with that among
› Two major genetic mutations lead to gout, urolithiasis, the White men ( p = 0.04) (Hochberg et al. 1995). However,
and related problems: this excess risk disappeared after adjustment for hypertension
› Mutations in the gene that encodes hypoxanthine- (adjusted relative risk = 1.30 (95% CI 0.77–2.19).
guanine phosphoribosyl transferase are associated with These data demonstrate that Black men have a greater
a spectrum of diseases in children, which ranges from incidence of gout than White men. However, this excess risk
the Lesch–Nyhan syndrome at the severe end of the is explained primarily by a greater incidence of hypertension
spectrum to simple hyperuricemia at the mild end. among Black men.
› Mutations in the 5’-phosphoribosyl-1-pyrophosphate
Myth: Gout is painful but does not affect longevity.
(PRPP) synthetase genes can result in overactivity of
the pathway, leading ultimately to increased urate Reality: A large prospective study among men found that
production. those with gout have a higher risk of death from all causes
(Choi and Curhan 2007b). Among men without known coro-
nary heart disease at baseline, the increased mortality risk is
due primarily to an elevated risk of cardiovascular death. In
Myth: Gout is common among men but rare among women.
addition, at least two other major studies confirm the rela-
Reality: Both the prevalence and incidence of gout are con- tionship between gout and shortened longevity, primarily
siderably higher in men than in women before the age of through the increased risk of coronary heart disease:
menopause, but the burden of gout in women increases sub-
stantially after menopause and rises with age. In the National • One study of men with above-average risk for coronary
Health and Nutrition Examination Survey (NHANES), the heart disease found that hyperuricemia and a diagnosis of
prevalence of gout among women between the ages 60 and gout were associated with increased long-term all-cause
69 years was 3.5%, but increased to 4.6% and 5.6% among mortality (Krishnan et al. 2008). The excess risk was
those aged 70–79 and 80 years or older, respectively (Kramer attributable principally to an increased risk of death from
and Curhan 2002). cardiovascular disease.
NHANES data may overestimate the prevalence of some • Gout also increased the risk of overall mortality among
disorders, as the survey is based upon individuals’ self-reports patient with end-stage renal disease by 49%, even after

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 355

DOI:10.1007/978-1-84800-934-9_35, © Springer Science + Business Media B.V. 2009
356 H. K. Choi

adjustments for gender, race, chronic pulmonary disease, Pearl: Diabetes is associated with lower serum uric acid
peripheral vascular disease, diabetes, ischemic heart dis- levels compared to patients with glucose intolerance.
ease, congestive failure, albumin levels, and smoking sta-
Comment: The relation between serum glucose levels and uric
tus (Cohen et al. 2008).
acid levels is not linear. In fact, it is bell shaped, rising to a
These data suggest that men with gout have a higher risk of vertex and then declining. Several studies have shown that a
death from all causes and that the increased mortality risk is moderate degree of hyperglycemia is associated with higher
due mainly to an elevated risk of death from cardiovascular serum uric acid levels, but marked hyperglycemia is associated
disease. These findings indicate that aggressive management with lower serum uric acid levels (Cook et al. 1986; Tuomilehto
of cardiovascular risk factors is appropriate for individuals et al. 1988; Herman and Goldbourt 1982; Whitehead et al.
with gout. 1992). The biologic explanation for this is the uricosuric effect
Pearl: The risk of the metabolic syndrome is remarkably high of glycosuria, which occurs when the blood glucose level
among patients with gout or a high level of hyperuricemia. is greater than approximately 10 mmol/L (180 mg/dL) (Cook
et al. 1986).
Comment: The metabolic syndrome is a constellation of Serum uric acid levels rise with moderately increasing
interrelated atherosclerotic risk factors. Abdominal obesity levels of serum glucose HbA1c (6–6.9%), but then decrease
and insulin resistance are two major features. Other meta- with further HbA1c elevations (again, a bell-shaped curve).
bolic components include atherogenic dyslipidemia, elevated Serum uric acid levels also rise in parallel with increases in
blood pressure, and elevated plasma glucose (Grundy et al. the level of fasting C-peptide levels (Choi and Ford 2008).
2005). The metabolic syndrome affects more than 50 million These data suggest divergent risks for hyperuricemia and
Americans and increases the risk for atherosclerotic cardio- gout among prediabetic and diabetic individuals. Individuals
vascular disease, type 2 diabetes, and death (Ford et al. 2002; with moderately elevated HbA1c levels (i.e., prediabetes) are
Ford 2005). at higher risk of hyperuricemia and gout. Those with highly
Close associations between serum urate levels and individual elevated HbA1c levels (i.e., diabetes) are at lower risk for
components of the metabolic syndrome exist (Emmerson 1998; these conditions, all other things being equal.
Lee et al. 1995; Rathmann et al. 1998). The prevalence of the
metabolic syndrome is high among hospitalized patients with Pearl: With regard to the prevention of gout, Osler got at
gout (82% in Mexican men (Vazquez-Mellado et al. 2004) and least part of the equation right: “The sugar should be reduced
44% in Korean men) (Rho et al. 2005). These data suggest that to a minimum. The sweeter fruits should not be taken.”
hyperuricemia should be regarded as an intrinsic part or surro- Comment: A number of foods and beverages are known to
gate marker for the metabolic syndrome (Emmerson 1998). affect the risk of hyperuricemia and gout. These are summa-
The high insulin levels known to be associated with the rized in Table 35.1.
metabolic syndrome reduce the renal excretion of uric acid Writing in his The Principles and Practice of Medicine
(Ter Maaten et al. 1997; Muscelli et al. 1996; Facchini et al. (1893), William Osler prescribed a diet low in fructose as a
1991; Dessein et al. 2000). Insulin may enhance renal urate means of preventing gout (Osler and Gout 1893; Nakagawa
reabsorption via stimulation of urate–anion exchanger URAT1 et al. 2005). Yet conventional dietary recommendations for
(Enomoto et al. 2002) and/or the Na+-dependent anion co- the prevention of gout have focused on the restriction of
transporter in brush border membranes of the renal proximal purine intake (Fam 2002). Was Osler wrong?
tubule (Choi et al. 2005b). The conventional low-purine diet No (Of course not). Both mechanistic data and prospective
recommended for gout patients that permits unlimited fruc- evidence support Osler’s concept that a high intake of fructose
tose ingestion could also contribute to the high prevalence of poses a substantial risk for gout. Fructose induces uric acid
the metabolic syndrome among these patients because fruc- production by increasing ATP degradation to AMP, a uric acid
tose intake is associated with elevated serum insulin levels,
insulin resistance, and adiposity.
A diagnosis of hyperuricemia or gout should trigger a high Table 35.1 Foods and beverages that affect the risk of hyperuricemia
and gout
clinical suspicion for coexistent metabolic syndrome (Ford
Increase risk Decrease risk Risk neutral
2005). Lifestyle interventions in this setting (e.g., enhanced
physical activity and dietary modifications) delay or prevent Beer Vitamin C Vegetables rich in
purines: peas, beans,
the transition from impaired glucose tolerance to type 2 diabe- spinach, mushrooms
tes (Ford et al. 2002; Eriksson and Lindgarde 1991; Pan et al. Whiskey Low-fat diary products Wine
1997; Tuomilehto et al. 2001). Daily exercise and weight con- Fructose Coffee (regular or Tea
trol are the foundations of the new healthy eating pyramid and decaffeinated)
an important recommendation for patients with gout (Choi Red meat Overall protein intake
et al. 2005, b; Willett and Stampfer 2003). Seafood
35 Gout Epidemiology 357

precursor (Choi et al. 2005b;Gibson et al. 1983; Fox and is somewhat lower with that beverage. In contrast, tea intake
Kelley 1972; Puig and Fox 1984; Faller and Fox 1982). and total caffeine ingestion from all sources do not alter the
Fructose phosphorylation in the liver uses ATP, and the accom- risk of gout. Thus, it seems to be something about the coffee.
panying phosphate depletion limits the regeneration of ATP Serum uric acid levels in the NHANES decreased as cof-
from ADP, which in turn serves as substrate for the catabolic fee intake increased, but tea consumption and total caffeine
pathway to uric acid formation (Fox et al. 1987). (Ethanol, ingestion were not linked to serum uric acid levels (Choi and
another known risk factor for gout, acts in the same manner). Curhan 2007a). These findings are consistent with a cross-
Thus, diets high in fructose heighten the risk of gout. sectional study from Japan in which coffee but not tea con-
Within minutes of a fructose infusion, the plasma uric acid sumption was associated inversely with serum uric acid
concentrations are increased (Fox and Kelley 1972). In con- levels (Kiyohara et al. 1999).
junction with purine nucleotide depletion, rates of purine The effect of coffee on insulin sensitivity may mediate its
synthesis de novo are accelerated, thus potentiating uric acid decreased risk of gout. Higher long-term coffee intake is asso-
production (Raivio et al. 1975). In contrast, glucose and other ciated with lower insulin levels (Wu et al. 2005) and improved
simple sugars do not have the same effect (Nakagawa et al. insulin sensitivity (Arnlov et al. 2004). Greater insulin sensi-
2005). Moreover, fructose intake correlates with the levels of tivity appears to foster lower serum levels of uric acid.
serum insulin as well as with insulin resistance (Wu et al.
2004), and therefore could increase indirectly both the serum Pearl: Purine-rich foods of vegetable origin do not lead to
uric acid level and the risk of gout (Choi et al. 2007; Lee et al. gout.
2004; Vasquez-Mellado et al. 2004). An increase in fructose Comment: The relationships between purine-rich foods, high
consumption contributes to a positive energy balance, and protein intake, and incident gout were examined prospectively
therefore to insulin resistance and adiposity (Tordoff and over a 12-year period in 47,150 male participants with no
Alleva 1990; Anderson et al. 1989; Gross et al. 2004; Bray baseline history of gout (Choi et al. 2004b). The investigators
et al. 2004; Thorburn et al. 1989). found no increased risk associated with the consumption of
A prospective study of male professionals confirmed that purine-rich vegetables (e.g., peas, beans, spinach, and mush-
consumption of sugar-sweetened soft drinks and fructose was rooms) or with overall protein intake. In fact, this study found
associated strongly with an increased risk of gout (Choi and a protective effect from vegetable and dairy proteins. However,
Curhan 2008). The risk of incident gout was 85% higher the consumption of meat, particularly red meat, increased the
among men who consumed two or more servings of sugar- risk of gout significantly. Seafood consumption was associ-
sweetened soft drinks per day compared with those who con- ated with a still higher risk (Choi et al. 2004b).
sumed less than one per month. In contrast, diet drinks were
not associated with the risk of gout. The consumption of fruit Myth: All alcoholic beverages are equally likely to lead to
juice or fructose-rich fruits such as apples and oranges is also gout.
associated with a modestly increased risk of gout. Reality: Beer consumption showed the strongest indepen-
Data from the NHANES indicate that serum uric acid dent association with the risk of new gout in the Health
concentrations increase in parallel with intake of sugar- Professionals Follow-up Study (Choi et al. 2004a, b). The
sweetened soft drinks (Choi et al. 2008). Diet soft drinks do risk of gout increased by 50% for every 12-ounce serving/
not demonstrate this same association. In men, added sugar day. Liquor consumption also increased the risk of new gout
intake was also associated with serum uric acid concentra- by approximately 15% for each drink. Moderate levels of
tion (Gao et al. 2007). wine intake did not increase the risk of gout in this study.
These findings indicate that consumption of sugar-sweetened In the NHANES III study, the association between beer
soft drinks and fructose is strongly associated with an increased consumption and serum uric acid level was strong for beer
risk of hyperuricemia and gout. Dietary recommendations that (an increase of 0.46 mg/dL for each additional serving) and
emphasize modest consumption of fructose are appropriate in for liquor (0.29 mg/dL for each additional serving), but no
individuals with these conditions. independent association existed for wine consumption (Choi
Pearl: Moderate- to high-level coffee consumption is associ- and Curhan 2004) It remains unclear whether there are
ated with a lower risk of gout. offending factors unrelated to alcohol in beer or liquor or per-
haps instead protective factors in that wine mitigate the effect
Comment: The more coffee one drinks, the lower one’s risk of of alcohol upon the risk of gout.
gout (Choi et al. 2007b). People who imbibe four or five cups
Pearl: Moderate- to high-dose vitamin C intake is associated
of java per day have a 40% lower risk of gout compared to
with a lower risk of gout.
those who do not drink coffee. Individuals who consume more
than six cups/day have a 61% lower risk. Decaffeinated coffee Comment: The suspicion for a potential protective effect of
is associated with a lowering of the risk of gout, albeit the effect vitamin C intake against gout originated from metabolic
358 H. K. Choi

experiments that examined the impact of short-term loading Arromdee E, Michet CJ, Crowson CS, O’Fallon WM, Gabriel SE.
of high-dose vitamin C on the serum uric acid levels. For Epidemiology of gout: is the incidence rising? J Rheumatol.
2002;29:2403–6
example, ingestion of a single dose of 4 g vitamin C doubled Bray GA, Nielsen SJ, Popkin BM. Consumption of high-fructose corn
the fractional excretion of uric acid, and a daily ingestion of syrup in beverages may play a role in the epidemic of obesity. Am
8 g of vitamin C for 3–7 days reduced serum uric acid by J Clin Nutr. 2004;79:537–43
2.0–3.1 mg/dL as a result of uricosuria (Stein et al. 1976). Choi HK, Atkinson K, Karlson EW, Willett WC, Curhan G. Alcohol
intake and risk of incident gout in men – a prospective study. Lancet
A randomized, double-blind, placebo-controlled trial 2004a;363:1277–81
(n = 184) showed that dietary supplementation with as little Choi HK, Atkinson K, Karlson EW, Willett WC, Curhan G. Purine-rich
as 500 mg of vitamin C a day for 2 months reduced serum foods, dairy and protein intake, and the risk of gout in men. N Engl
uric acid by 0.5 mg/dL, compared to no change in the pla- J Med. 2004b;350:1093–103
Choi HK, Curhan G. Beer, liquor, wine, and serum uric acid level – the
cebo group (Huang et al. 2005). A retrospective case–control Third National Health and Nutrition Examination Survey. Arthritis
study (91 gout cases and 91 controls) reported an inverse Rheum. 2004:1023–9
association between vitamin C intake and the presence of Choi HK, Curhan G. Coffee, tea, and caffeine consumption and serum
gout (Lyu et al. 2003). uric acid level: the Third National Health and Nutrition Examination
Survey. Arthritis Rheum. 2007a;57:816–21
An updated analysis of the Health Professionals Follow-up Choi HK, Curhan G. Independent impact of gout on mortality and risk
Study over a 20-year period found that the risk of gout decreased for coronary heart disease. Circulation 2007b;116:894–900
with increasing vitamin C intake (Choi et al. 2009). Individuals Choi HK, Curhan G. Soft drinks, fructose consumption, and the risk of
who ingested 1,500 mg or more of vitamin C daily had a risk of gout in men: prospective cohort study. BMJ 2008;336:309–12
Choi HK, Ford ES. Haemoglobin A1c, fasting glucose, serum C-peptide
gout that was 45% lower than those in the lowest category. This and insulin resistance in relation to serum uric acid levels – the Third
finding was independent of all known dietary factors and other National Health and Nutrition Examination Survey. Rheumatology
risk factors for gout such as body mass index, age, hyperten- (Oxford) 2008;47:713–7
sion, diuretic use, alcohol, and chronic renal failure. These data Choi HK, Ford ES, Li C, Curhan G. Prevalence of the metabolic syn-
drome in patients with gout: the Third National Health and Nutrition
suggest that vitamin C intake may be beneficial in the preven- Examination Survey. Arthritis Rheum. 2007a;57:109–15
tion and management of gout. Choi HK, Liu S, Curhan G. Intake of purine-rich foods, protein, dairy
products, and serum uric acid level – the Third National Health and
Pearl: Low-fat dairy product consumption is associated with Nutrition Examination Survey. Arthritis Rheum. 2005a;52:283–9
a lower risk of gout. Choi HK, Mount DB, Reginato AM. Pathogenesis of gout. Ann Intern
Med. 2005b;143:499–516
Comment: The potentially protective role of dairy consump- Choi HK, Willett W, Curhan G. Coffee consumption and risk of inci-
tion on gout was suggested by several studies that measured dent gout in men: a prospective study. Arthritis Rheum. 2007b;56:
uric acid as a surrogate outcome (Loenen et al. 1990; Garrel 2049–55
Choi HK Gao X, Curhan G. Vitamin C intake and the risk of gout in
et al. 1991; Ghadirian et al. 1995). The ingestion of the milk men: a prospective study. Arch Intern Med 2009;169:502–7
proteins casein and lactalbumin decreases serum uric acid lev- Choi JW, Ford ES, Gao X, Choi HK. Sugar-sweetened soft drinks, diet
els in healthy subjects via a uricosuric effect (Garrel et al. soft drinks, and serum uric acid level: the Third National Health and
1991). Conversely, a significant increase in uric acid level was Nutrition Examination Survey. Arthritis Rheum. 2008;59:109–16
Cohen SD, Kimmel PL, Neff R, Agodoa L, Abbott KC. Association of
induced by a dairy-free diet in a 4-week randomized clinical incident gout and mortality in dialysis patients. J Am Soc Nephrol.
trial (Ghadirian et al. 1995). Those who consumed milk one or 2008;19:2204–10
more times per day had a lower serum uric acid level than those Cook DG, Shaper AG, Thelle DS, Whitehead TP. Serum uric acid,
who did not drink milk (p for trend <0.0001) (Choi et al. serum glucose and diabetes: relationships in a population study.
Postgrad Med J. 1986;62:1001–6
2005a). Similarly, those who consumed yogurt at least once Dessein PH, Shipton EA, Stanwix AE, Joffe BI, Ramokgadi J. Beneficial
every other day had a lower serum uric acid level than those effects of weight loss associated with moderate calorie/carbohy-
who did not consume yogurt (p for trend <0.0001). The actual drate restriction, and increased proportional intake of protein and
risk of gout was 44% lower in the highest quintile of dairy con- unsaturated fat on serum urate and lipoprotein levels in gout: a pilot
study. Ann Rheum Dis. 2000;59:539–43
sumption as compared with the lowest quintile. This inverse Emmerson B. Hyperlipidaemia in hyperuricaemia and gout. Ann Rheum
association was evident with low-fat dairy consumption, but Dis. 1998;57:509–10
not with high-fat dairy consumption (Choi et al. 2004b). Enomoto A, Kimura H, Chairoungdua A, et al Molecular identification
of a renal urate anion exchanger that regulates blood urate levels.
Nature 2002;417:447–52
Eriksson KF, Lindgarde F. Prevention of type 2 (non-insulin-dependent)
References diabetes mellitus by diet and physical exercise. The 6-year Malmo
feasibility study. Diabetologia 1991;34:891–8
Facchini F, Chen YD, Hollenbeck CB, Reaven GM. Relationship between
Anderson JW, Story LJ, Zettwoch NC, Gustafson NJ, Jefferson BS. resistance to insulin-mediated glucose uptake, urinary uric acid clear-
Metabolic effects of fructose supplementation in diabetic individu- ance, and plasma uric acid concentration. JAMA 1991;266:3008–11
als. Diabetes Care. 1989;12:337–44 Faller J, Fox IH. Ethanol-induced hyperuricemia: evidence for increased
Arnlov J, Vessby B, Riserus U. Coffee consumption and insulin sensi- urate production by activation of adenine nucleotide turnover. N
tivity. JAMA 2004;291:1199–201 Engl J Med. 1982;307:1598–602
35 Gout Epidemiology 359

Fam AG. Gout, diet, and the insulin resistance syndrome. J Rheumatol. Nakagawa T, Tuttle KR, Short RA, Johnson RJ. Hypothesis: fructose-induced
2002;29:1350–5 hyperuricemia as a causal mechanism for the epidemic of the metabolic
Ford ES. Risks for all-cause mortality, cardiovascular disease, and dia- syndrome. Nat Clin Pract Nephrol. 2005; 1: 80–6
betes associated with the metabolic syndrome: a summary of the Osler W, Gout. The principles and practice of medicine. New York:
evidence. Diabetes Care 2005;28:1769–78 Appleton; 1893. p. 287–95
Ford ES, Giles WH, Dietz WH. Prevalence of the metabolic syndrome Pan XR, Li GW, Hu YH, et al Effects of diet and exercise in preventing
among US adults: findings from the third National Health and NIDDM in people with impaired glucose tolerance. The Da Qing
Nutrition Examination Survey. JAMA 2002;287:356–9 IGT and Diabetes Study. Diabetes Care. 1997;20:537–44
Fox IH, Kelley WN. Studies on the mechanism of fructose-induced Puig JG, Fox IH. Ethanol-induced activation of adenine nucleotide turn-
hyperuricemia in man. Metabolism 1972;21:713–21 over. Evidence for a role of acetate. J Clin Invest. 1984;74: 936–41
Fox IH, Palella TD, Kelley WN. Hyperuricemia: a marker for cell Raivio KO, Becker A, Meyer LJ, Greene ML, Nuki G, Seegmiller JE.
energy crisis. N Engl J Med. 1987;317:111–2 Stimulation of human purine synthesis de novo by fructose infusion.
Gao X, Qi L, Qiao N, et al Intake of added sugar and sugar-sweetened Metabolism 1975;24:861–9
drink and serum uric acid concentration in US men and women. Rasch EK, Hirsch R, Paulose-Ram R, Hochberg MC. Prevalence of
Hypertension 2007;50:306–12 rheumatoid arthritis in persons 60 years of age and older in the
Garrel DR, Verdy M, PetitClerc C, Martin C, Brule D, Hamet P. Milk- United States: effect of different methods of case classification.
and soy-protein ingestion: acute effect on serum uric acid concen- Arthritis Rheum. 2003;48:917–26
tration. Am J Clin Nutr. 1991;53:665–9 Rathmann W, Funkhouser E, Dyer AR, Roseman JM. Relations of
Ghadirian P, Shatenstein B, Verdy M, Hamet P. The influence of dairy prod- hyperuricemia with the various components of the insulin resistance
ucts on plasma uric acid in women. Eur J Epidemiol. 1995; 11:275–81 syndrome in young black and white adults: the CARDIA study.
Gibson T, Rodgers AV, Simmonds HA, Court-Brown F, Todd E, Meilton Coronary Artery Risk Development in Young Adults. Ann Epidemiol.
V. A controlled study of diet in patients with gout. Ann Rheum Dis. 1998;8:250–61
1983;42:123–7 Rho YH, Choi SJ, Lee YH, et al The prevalence of metabolic syndrome
Gross LS, Li L, Ford ES, Liu S. Increased consumption of refined car- in patients with gout: a multicenter study. J Korean Med Sci. 2005;
bohydrates and the epidemic of type 2 diabetes in the United States: 20:1029–33
an ecologic assessment. Am J Clin Nutr. 2004;79:774–9 Stein HB, Hasan A, Fox IH. Ascorbic acid-induced uricosuria. A conse-
Grundy SM, Cleeman JI, Daniels SR, et al Diagnosis and management quence of megavitamin therapy. Ann Intern Med. 1976;84:385–8
of the metabolic syndrome: an American Heart Association/National Ter Maaten JC, Voorburg A, Heine RJ, Ter Wee PM, Donker AJ, Gans
Heart, Lung, and Blood Institute Scientific Statement. Circulation RO. Renal handling of urate and sodium during acute physiological
2005;112:2735–52 hyperinsulinaemia in healthy subjects. Clin Sci (Lond). 1997;92:
Herman JB, Goldbourt U. Uric acid and diabetes: observations in a 51–8
population study. Lancet 1982;2:240–3 Thorburn AW, Storlien LH, Jenkins AB, Khouri S, Kraegen EW.
Hochberg MC, Thomas J, Thomas DJ, Mead L, Levine DM, Klag MJ. Fructose-induced in vivo insulin resistance and elevated plasma
Racial differences in the incidence of gout. The role of hyperten- triglyceride levels in rats. Am J Clin Nutr. 1989;49:1155–63
sion. Arthritis Rheum. 1995;38:628–32 Tordoff MG, Alleva AM. Effect of drinking soda sweetened with aspar-
Huang HY, Appel LJ, Choi MJ, et al The effects of vitamin C supple- tame or high-fructose corn syrup on food intake and body weight.
mentation on serum concentrations of uric acid: results of a random- Am J Clin Nutr. 1990;51:963–9
ized controlled trial. Arthritis Rheum. 2005;52:1843–7 Tuomilehto J, Lindstrom J, Eriksson JG, et al Prevention of type 2 dia-
Kiyohara C, Kono S, Honjo S, et al Inverse association between coffee betes mellitus by changes in lifestyle among subjects with impaired
drinking and serum uric acid concentrations in middle-aged Japanese glucose tolerance. N Engl J Med. 2001;344:1343–50
males. Br J Nutr. 1999;82:125–30 Tuomilehto J, Zimmet P, Wolf E, Taylor R, Ram P, King H. Plasma uric
Kramer HM, Curhan G. The association between gout and nephrolithia- acid level and its association with diabetes mellitus and some bio-
sis: the National Health and Nutrition Examination Survey III, logic parameters in a biracial population of Fiji. Am J Epidemiol.
1988–1994. Am J Kidney Dis. 2002;40:37–42 1988;127:321–36
Krishnan E, Svendsen K, Neaton JD, Grandits G, Kuller LH. Long- Vazquez-Mellado J, Conrado GG, Vazquez SG, et al Metabolic syn-
term cardiovascular mortality among middle-aged men with gout. drome and ischemic heart disease in gout. J Clin Rheumatol. 2004;
Arch Intern Med. 2008;168:1104–10 10:105–9
Lee J, Sparrow D, Vokonas PS, Landsberg L, Weiss ST. Uric acid and Wallace SL, Robinson H, Masi AT, Decker JL, McCarty DJ, Yu TF.
coronary heart disease risk: evidence for a role of uric acid in the Preliminary criteria for the classification of the acute arthritis of pri-
obesity–insulin resistance syndrome. The Normative Aging Study. mary gout. Arthritis Rheum. 1977;20:895–900
Am J Epidemiol. 1995;142:288–94 Whitehead TP, Jungner I, Robinson D, Kolar W, Pearl A, Hale A. Serum
Lee WY, Park JS, Noh SY, Rhee EJ, Kim SW, Zimmet PZ. Prevalence urate, serum glucose and diabetes. Ann Clin Biochem. 1992;29
of the metabolic syndrome among 40,698 Korean metropolitan sub- (Pt 2):159–61
jects. Diabetes Res Clin Pract. 2004;65:143–9 Willett WC, Stampfer MJ. Rebuilding the food pyramid. Sci Am.
Loenen HM, Eshuis H, Lowik MR, et al Serum uric acid correlates in 2003;288:64–71
elderly men and women with special reference to body composition Wu T, Giovannucci E, Pischon T, et al Fructose, glycemic load, and
and dietary intake (Dutch Nutrition Surveillance System). J Clin quantity and quality of carbohydrate in relation to plasma C-peptide
Epidemiol. 1990;43:1297–303 concentrations in US women. Am J Clin Nutr. 2004;80:1043–9
Lyu LC, Hsu CY, Yeh CY, Lee MS, Huang SH, Chen CL. A case–control Wu T, Willett WC, Hankinson SE, Giovannucci E. Caffeinated coffee,
study of the association of diet and obesity with gout in Taiwan. Am decaffeinated coffee, and caffeine in relation to plasma C-peptide
J Clin Nutr. 2003;78:690–701 levels, a marker of insulin secretion, in U.S. women. Diabetes Care.
Muscelli E, Natali A, Bianchi S, et al Effect of insulin on renal sodium 2005;28:1390–6
and uric acid handling in essential hypertension. Am J Hypertens.
1996;9:746–52
 Gout Treatment
36
N. Lawrence Edwards, Terence Gibson, and Robert A. Terkeltaub

36.1 Overview of the Treatment of Gout

› Treatment of gouty inflammation, hyperuricemia, and

comorbidities are distinct but closely linked core ele-
ments of therapeutic strategy for patients with gout.
› All gout patients should be evaluated and treated for
comorbid medical problems such as hypertension,
metabolic syndrome, and coronary artery disease.
› Lifestyle modifications, including weight loss, are
important adjuncts to pharmacologic urate lowering.
› The target serum urate level for urate-lowering therapy
in most patients is less than 6.0 mg/dL.
› Serial monitoring of serum uric acid levels is essential, Fig. 36.1 Glucocorticoids are an effective therapy for acute gout flares.
following the initiation of urate-lowering therapy. The patient whose feet are shown had a known history of gout when he
› High-dose oral colchicine for the treatment of acute developed intense erythema over the midfoot. Tenderness was most
intense over the first metatarsophalangeal joint. Because of existing
gout has an unacceptable risk:benefit ratio. renal dysfunction (serum creatinine 2.5 mg/dL), he was treated with
› Maintenance colchicine should be continued for at prednisone 20 twice daily for 5 days, followed by a 1-week prednisone
least the first 6 months of urate-lowering therapy. taper. Daily colchicine (0.6 mg) was initiated after the start of the pred-
› Allopurinol should not be initiated or discontinued nisone taper and continued for 6 months after the flare. Allopurinol
(100 mg/day) was begun after the acute gout flare had subsided (Figure
during an acute gout flare. Courtesy of Dr. John Stone)
› In most cases, allopurinol and other urate-lowering
therapies should be continued for life.
20 mg twice daily for 3–5 days and then taper the glucocorti-
coids off over the following week. Using parenteral glucocor-
ticoids has no particular advantage over oral administration,
Pearl: Oral glucocorticoids are an excellent treatment option
unless the subject is NPO or has upper GI symptoms. In these
for acute gout.
circumstances, one dosing option is triamcinolone acetonide
Comment: Prednisone and prednisolone are excellent thera- 60 mg IM or its equivalent (Siegel et al. 1994).
peutic choices for acute gout in clinical settings in which Although glucocorticoids are highly effective for the
NSAIDs, colchicine, or intraarticular glucocorticoids are con- treatment of acute gout, using low-dose oral glucocorticoids
traindicated or have not been adequately effective for treat- for the prevention of gout flares is not evidence based.
ment of acute gout. A randomized clinical trial that compared Glucocorticoids should be withdrawn as quickly as possible
prednisolone (35 mg/day for 5 days) to naproxen (500 mg after the acute gouty inflammation resolves.
twice daily for 5 days) for the treatment of acute gouty arthri-
Pearl: “Less is more” when using colchicine in the treatment
tis demonstrated comparable efficacy and tolerance (Janssens
of acute gout.
et al. 2008).
Acute gouty arthritis (Fig. 36.1) requires moderate-to-high Comment: In the past, the typical colchicine regimen for the
doses of glucocorticoids. A typical initial dosing regimen is treatment of acute gout was the administration of one 0.6 mg
equivalent to a minimum of 0.5 mg/kg of prednisone each tablet each hour until the patient improved substantially, devel-
day. As an example, the clinician might prescribe prednisone oped intolerable side effects (usually nausea and diarrhea), or

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 361

DOI:10.1007/978-1-84800-934-9_36, © Springer Science + Business Media B.V. 2009
362 N. L. Edwards et al.

reached a predetermined maximum dose (typically 8–12 tab- Table 36.1 Dosing guidelines for colchicine in patients with renal
lets within 24 h). Nearly all patients’ gout attacks benefit from dysfunction
such treatment. Unfortunately, every patient treated in this Creatinine clearance (mL/min) Colchicine dose
manner experiences adverse gastrointestinal (GI) effects, ≥ 50 0.6 mg twice daily
yielding a benefit-to-toxicity ratio for colchicine of 1 (Ahern 35–49 0.6 mg once daily
10–34 0.6 mg every two 2 or
et al. 1987). Moreover, the diarrhea usually begins before
three 3 days
relief of the gout flare, with the result that the patient limps off <10, or on hemodialysis, or combined Do not use colchicine
to the bathroom on joints that are still acutely inflamed. moderate-to-severe renal and
There are better ways than colchicine to treat gout flares in hepatobiliary disease
most patients (namely, glucocorticoids or nonsteroidal anti-
inflammatory drugs). However, if a patient’s comorbidities
indicate that colchicine is the preferred therapy, there is a less clarithromycin), certain statins, some calcium channel block-
toxic approach to administering the drug. The European League ers (nifedipine and verapamil), several antineoplastic agents,
Against Rheumatism (EULAR) guidelines advocate a regimen and HIV protease inhibitors (Niel and Scherrmann 2006;
of three 0.6 mg colchicine tablets in the first 24 h for treatment Terkeltaub 2008b). Statins are particularly associated with
of acute gout (Zhang et al. 2006a). Moreover, a recent, large, rhabdomyolysis in the presence of colchicine. The guidelines
randomized controlled trial that compared three 0.6 mg colchi- for the dosing of colchicine in patients with renal dysfunction
cine tablets to eight tablets administered within the first 24 h of are shown in Table 36.1.
an acute gout flare demonstrated nearly equal efficacy between Patients with combined significant hepatic (and particu-
the two treatment arms and significantly fewer GI side effects larly hepatobiliary) disease and renal disease are at particular
in the lower dose group (Terkeltaub 2008a). risk for colchicine toxicity. For patients with severe renal
dysfunction, colchicine should not be used more frequently
Pearl: Colchicine works best when started in the first day or
than 0.6 mg every 2 or 3 days. Finally, the impaired colchi-
two of the acute gouty flare.
cine pharmacokinetics in the elderly make it advisable to
Comment: If you are going to use colchicine to treat acute reduce the daily dosage regimens further by up to 50% in
gout, then it needs to be started early. The optimal use of patients above the age of 70.
colchicine is to instruct the patient to take up to three 0.6 mg Only rarely is it necessary for a patient to be treated with
tablets orally within 3 h of the first sign of a new gouty flare. prophylactic colchicine for more than 12 months! Colchicine
However, it is not prudent to superimpose this therapeutic should not be viewed as a chronic medication for gout treat-
strategy on a patient already maintained on daily low-dose ment but rather as a bridge for preventing gout flares until
prophylactic colchicine, as this may enhance colchicine tox- sufficiently low-urate levels are achieved.
icity (Terkeltaub 2003). In short, if the patient has been on
Pearl: Keep your eyes on the target level of serum urate.
colchicine prophylaxis and has been compliant, select another
medication for treating an acute gout flare. Comment: Serum urate levels of 6.8 mg/dL (0.41 mmol/L)
and higher exceed the saturation limit of urate in solution.
Myth: Oral low-dose colchicine for prophylaxis of gout flares
Levels above 6.8 mg/dL promote deposition of monosodium
is contraindicated in all patients with moderate to severe
urate crystals in connective tissues. To provide some margin
renal insufficiency.
for error in treating gout, clinicians should target a serum
Reality: Colchicine is eliminated, in part, via renal excretion. urate level less than 6.0 mg/dL (0.357 mmol/L). Achieving
Thus, colchicine is contraindicated in patients undergoing such a level will eventually lead to fewer gout flares.
dialysis (colchicine is not dialyzable) and in patients with a Patients with bulky tophi (Fig. 36.2) benefit from having
creatinine clearance less than 10 mL/min. Colchicine neuro- their serum urate levels lowered to the 3.0–4.0 mg/dL to has-
myotoxicity, which can be acute or subacute in onset, is a ten tophus resorption (Perez-Ruiz et al. 2002).
potentially devastating and slowly reversible consequence of
Myth: Serum urate levels do not need to be monitored if the
this misstep (Wilbur 2004). Colchicine neuromyotoxicity
patient is on a standard dose of allopurinol.
occurs most frequently in the setting of chronic kidney dis-
ease but has also been reported in patients with normal renal Reality: The majority of patients do not achieve target serum
function. urate levels on 300 mg or less of allopurinol daily. The
Prophylactic low-dose colchicine can be used safely in serum urate should be monitored 3–4 weeks after the initia-
most patients and even those with chronic kidney disease tion of urate-lowering therapy and 3–4 weeks after each
if there is particular attention to the potential for drug–drug dose adjustment. Once the serum urate level has been sup-
interactions. Medications of particular concern are cyclosporine, pressed adequately (<6.0 mg/dL), the urate level should
some macrolide antibiotics (including erythromycin and continue to be monitored every 6–12 months.
36 Gout Treatment 363

Reality: Allopurinol 300 mg daily lowers serum urate by an

average of ~33% in subjects with intact renal function
(Becker et al. 2005). However, titration of the allopurinol
dose is essential to optimize its urate-lowering efficacy. The
uricosuric agent benzbromarone (not available in the United
States) works at least as well as allopurinol in lowering serum
urate levels and shrinking tophi, provided the patient has
intact renal function (Perez-Ruiz et al. 2002).
Uricosuric agents are not effective in patients with sub-
stantial renal impairment. As an example, probenecid is rela-
tively ineffective in lowering serum urate levels in patients
with a creatinine clearance of less than 60 mL/min. In con-
trast, allopurinol is a reasonable first choice agent for all
patients with recurrent gout, regardless of renal function or
Fig. 36.2 Patients with extensive tophi should have a lower serum the patients’ status as “overproducers” or “underexcreters”
urate target level. For most patients with gout, the serum urate target (Terkeltaub 2003).
level is 6.0 mg/dL. For those with bulky tophi, resorption will occur
more quickly if a serum level of 3.0–4.0 mg/dL can be achieved (Figure Pearl: The use of azathioprine and allopurinol together can
Courtesy of Dr. John Stone)
lead to a severe drug–drug interaction. It is usually better not
to use this combination.
Pearl: Allopurinol is often underdosed.
Comment: This combination is particularly relevant to the
Comment: Allopurinol is approved by the FDA for the treat- population of patients with organ transplants, many of whom
ment of hyperuricemia in gout at doses up to 800 mg daily. are on azathioprine to prevent graft rejection and many of
However, more than 98% of all allopurinol prescriptions are whom are at risk for gout because of calcineurin inhibitors.
written for 300 mg daily or less. This reticence to use higher Both cyclosporine and tacrolimus alter urate handling, lead-
doses is, in part, due to the fear of precipitating a major ing to hyperuricemia and gout in a significant number of
allopurinol hypersensitivity syndrome. patients. Xanthine oxidase is involved in the metabolism of
The safest clinical practice is the “start low, go slow” azathioprine. As an inhibitor of xanthine oxidase, allopurinol
approach recommended in the EULAR guidelines (Zhang interferes with the metabolism of azathioprine (and its
et al. 2006a). In this model, all patients are started on 100 mg/ metabolite, 6-mercaptopurine). This can lead to marked ele-
day, and the serum urate level is monitored 3–4 weeks later. vation of azathioprine levels and the potential for life-threat-
If the target urate level of <6.0 mg/dL has not been achieved ening bone-marrow toxicity (Duley et al. 2005). Moreover,
by the current dose, the dose is increased by 100 mg every the inhibition of xanthine oxidase by allopurinol alters aza-
3–4 weeks until the target serum urate is reached. In subjects thioprine levels in a variable and unpredictable way. Serious
with stage 4 or 5 chronic kidney disease, the initial dose may toxicity can occur, despite careful monitoring of hematology
be dropped to 50 mg/day and the subsequent increases made and liver function.
in 50 mg increments every 3–4 weeks. The need to consider the combined use of allopurinol and
Remember that while the patient’s allopurinol dose is azathioprine is receding because mycophenolate mofetil
being titrated upwards, the patient should also be on colchi- (MMF) is replacing azathioprine as the immunosuppressive
cine to prevent disease flares triggered by declining levels of agent of choice. For patients who require both allopurinol
serum urate and associated tophus remodeling. and azathioprine and for whom no substitute for either is per-
Myth: Severe allopurinol hypersensitivity syndrome is common. mitted, the dose of azathioprine should be decreased by
Reality: Full-blown allopurinol hypersensitivity syndrome is a 50–75%. Complete blood counts must be followed at fre-
serious medical illness with a mortality rate of up to 20%. quent intervals until stable doses of the medications and urate
However, it is rare. In contrast, skin rashes develop in approxi- levels have been achieved. A second (and lesser) option that
mately 2% of patients who receive allopurinol. Hypersensitivity is available in patients who do not have extensive tophi is to
reactions to allopurinol are hereditary in some patients. Genetic continue azathioprine but prescribe a uricosuric drug such as
linkage to HLA-B5801 has been described in patients of either probenecid, sulfinpyrazone, or benzbromarone.
European or Han Chinese descent. In other patients, hypersen- Pearl: Renal and heart transplant recipients experience
sitivity reactions are dose dependent. hyperuricemia and gout more frequently than liver trans-
Myth: Allopurinol lowers serum urate and shrinks tophi plant patients, but these problems occur in all of these trans-
more effectively than do all uricosuric agents. plant populations.
364 N. L. Edwards et al.

The development of visible tophi in the absence of typical

gout has also been reported in long-term transplant patients
who have not been received a calcineurin inhibitor (Braun
1999). These data support for an effect of prednisone and
other transplant medications (MMF, azathioprine) on the
suppression of clinical gout.
Pearl: Colchicine is best avoided in transplant patients
receiving cyclosporine because of the risk of colchicine
neuromyotoxicity.

Comment: Most clinicians prudently avoid the use of NSAIDs

in renal transplant patients because of their deleterious effect
on kidney function. Colchicine might appear to be a safer
Fig. 36.3 Tophi in patients who develop gout in the setting of organ
transplantation often occur over the distal interphalangeal joints (at the alternative with regard to potential nephrotoxic effects.
site of Heberden’s nodes) or as subcutaneous deposits that resemble a However, when colchicine is used for acute gout or in regular
local infection (Figure courtesy of Dr. Terence Gibson) low doses as prophylaxis in patients with renal impairment, it
carries the risk of proximal myopathy. This adverse effect has
been observed most often among transplant recipients who
are also on cyclosporine, because cyclosporine interferes
Comment: This is true. However, hyperuricemia and gout cer-
with colchicine metabolism (Dupont 2002; Simkin and
tainly occur in liver transplant patients, too. The calcineurin
Gardner 2002).
inhibitors cyclosporine and tacrolimus are the principal cause
Colchicine myopathy usually presents as muscle weak-
of hyperuricemia in all organ transplant patients. These medi-
ness and is associated with an elevated creatine kinase level
cations cause hyperuricemia in any clinical situation in which
(Kuncl 1987). Characteristic electromyographic and muscle
they are used. Cyclosporine and tacrolimus alter urate han-
biopsy abnormalities are present in this condition, and the
dling within the kidney and also cause renal dysfunction and
myopathy can be accompanied by a mild peripheral neuropa-
hypertension. The hyperuricemic effect is mediated either by
thy. Symptoms regress over 3–4 weeks after stopping colchi-
a reduction of glomerular filtration or, more likely, by the sup-
cine, but may be permanent. The preferred treatment of acute
pression of uric acid secretion by the proximal tubule (Hansen
gout in renal transplant patients is intraarticular or intramus-
1998; Laine 1996).
cular injection of a depot preparation of prednisolone, a 7–10
The renal and heart transplant patients often have additional
day course of an oral glucocorticoid dose such as predni-
risk factors for gout. These include the existence of hyperuri-
sone, or an increase in the patient’s baseline prednisone
cemia before renal transplantation (because of end-stage renal
dose.
disease), the high frequency of renal allograft dysfunction, and
a greater likelihood of concomitant diuretic use. Pearl: Back pain in the transplant patient treated with
cyclosporine might be due to gout.
Pearl: Transplant patients are likely to develop tophi without
a history of acute gout. Comment: Back pain due to tophaceous deposits or acute gout
can mimic infective diskitis when fever, leukocytosis, and
Comment: The accumulation of tophi in the absence of acute
elevated serum inflammatory markers are present (Suk et al.
arthritis was first emphasized as a manifestation of hyperurice-
2007). This is a particularly difficult diagnosis in transplant
mia in older persons on diuretics (Macfarlane and Dieppe
patients because of their underlying immunosuppression.
1985). The tophi in such cases often occur over the distal inter-
Spinal gout is not common, but transplant patients are
phalangeal joints (at the site of Heberden’s nodes) or as subcu-
disproportionately represented among all documented cases
taneous deposits that resemble a local infection (Fig. 36.3).
(Clive 2000; Hou et al. 2007). Spinal tophi can cause pain,
Both cyclosporine and diuretics have an intense inhibitory
compress emerging nerve roots, and lead to spinal stenosis
impact on the secretion of urate by the proximal tubule. Both
(Draganescu and Leventhal 2004). The possibility of gout
these drugs are often used in transplant patients. As a conse-
should be considered in patients with a history of hyperuri-
quence, hyperuricemia develops and total body urate accu-
cemia and peripheral articular gout. Spinal infection must
mulates very quickly. This rapidity of uric acid accumulation
be excluded in these cases by image-guided aspiration and
may account for the relative paucity of acute gouty arthritis in
biopsy.
transplant patients. The anti-inflammatory effects of immu-
nosuppressive medications used in transplant patients may Myth: Acute gout is less common than anticipated among
also contribute. hyperuricemic patients in renal failure.
36 Gout Treatment 365

Reality: The clinical impression of less frequent gout in renal urate secretion in the proximal tubules. All diuretics, with the
failure has been attributed to an antiinflammatory effect of exception of spironolactone, reduce uric acid excretion by
uremia (Hollingsworth 1988; Buchanan et al. 1965). Neither the kidneys, and can interfere with the ability of allopurinol
of these assertions is supported by strong clinical evidence. to lower serum urate levels.
The only comparative study of the frequency of gout among Having said that, patient noncompliance is a major issue
patients with equivalent levels of hyperuricemia but with or in the treatment of gout. Noncompliance with allopurinol
without renal failure failed to demonstrate any difference was reported to be ~50% in one study (Riedel et al. 2004).
between the two groups in the incidence of acute gout (Youssef The question of whether or not a patient is taking his medi-
et al. 1995). cine can be settled by the measurement of a serum oxypurinol
level. Oxypurinol is the major active metabolite of allopurinol,
Myth: A high-protein diet causes hyperuricemia and gout.
and its presence in a plasma measurement means the patient
Reality: A high-protein meal actually increases uric acid is taking at least some of his medication. Therapeutic levels
clearance and can reduce serum urate (Matzkies et al. 1980). of oxypurinol are between 5 and 15 mg/L.
Low-protein diets have no effect on gout. Allopurinol compliance can also be monitored by the
Most circulating uric acid derives from the catabolism of quantification of allopurinol metabolites in the urine. These
endogenous purines that are generated in the liver. Purines estimations rely on one or more 24-h urine collections—not
that are ingested, in contrast, contribute variably to the total always easy to obtain in patients who are inclined to non-
uric-acid pool. The food items that are richest in dietary compliance.
purines are shellfish, caviar, and organ meats such as liver,
Myth: There are no therapeutic options for patients who fail
kidney, brain, pancreas (“sweetbreads”), and intestines
or are intolerant to allopurinol.
(“chitlins”). Some vegetables, such as legumes, contain more
purines than others. Reality: Many patients fail to achieve the target suppression of
A formula diet that is entirely devoid of purines for sev- serum urate to <6.0 mg/dL on “standard doses” of allopurinol.
eral days can reduce the serum urate of a healthy man by A person should not be considered a treatment failure until the
30% (Griebsch and Zollner 1974), but such diets are rarely dose of allopurinol has been escalated slowly to 800 mg/day.
sustainable over the long term. A general rule of thumb is For patients who are intolerant of allopurinol, febuxostat
that lowering the serum urate level by more than 1 mg/dL is (a daily oral xanthine oxidase inhibitor) and pegloticase (an
not possible by diet alone. Patients with substantial hyperuri- intravenous pegylated urate oxidase protein given every 2–4
cemia and frequent symptoms of gout require medication. weeks) may both be accessible therapeutic options shortly.
Both medications are currently under review by the FDA.
Myth: Patients with gout ingest more purines than healthy
Febuxostat use has been approved in Europe (Edwards 2008).
people.
Adjunctive medications that are readily available now
Reality: There is only one irrefutable dietary association of include the use of uricosuric agents such as probenecid,
hyperuricemia and gout, namely, the strong link with alco- losartan, and fenofibrate. Rasburicase is a nonpegylated urate
hol consumption (Rodnan 1980; Choi et al. 2004a, b). There oxidase used primarily for short-term prophylaxis of the
are also broad correlations between measures of obesity and tumor lysis syndrome. Its short half-life (~24 h in serum) and
serum uric-acid levels (Acheson and Chan 1969). One high degree of immunogenicity (including anaphylaxis)
dietary study did not find higher levels of total purine con- makes it a poor choice for the treatment of chronic hyperuri-
sumption among patients with gout compared with controls cemia (Terkeltaub 2007).
(Gibson et al. 1983). However, an excess of alcohol was
Pearl: All forms of intensive urate-lowering therapy actually
confirmed in the gout population. All forms of alcohol
promote gout flares.
heighten the risk of gout, but beer appears to be a stronger
risk factor in this respect than are wine and whiskey (Choi Comment: Any condition that dramatically raises or lowers
et al. 2004a, b). the serum urate level is likely to induce a flare of gout. Thus,
the risk of a gout flare increases when a patient is begun on
Pearl: If a patient’s serum urate fails to reach the target
urate-lowering therapy alone. The increased risk is particu-
level, despite reasonable allopurinol doses, the clinician
larly high within the first 6 months of urate-lowering therapy.
should be suspicious of noncompliance.
Tophus remodeling is probably responsible for this. The
Comment: Hyperuricemia and recurrent gout can be encoun- breakup of quiescent, walled-off tophaceous aggregates of
tered despite a dose of allopurinol that exceeds 600 mg daily. urate crystals within the synovium may initially release more
Only 25–50% of patients on allopurinol 300 mg/day achieve inflammatory forms of crystals. These mobilized crystals
a target urate level of <6.0 mg/dL. Lactic acidosis or ketoaci- interact readily with resident synovial cells and initiate the
dosis can cause hyperuricemia by competitive inhibition of “attack.”
366 N. L. Edwards et al.

As an example, in one study of patients who started However, the overall effect on urate levels can be substantial
allopurinol 300 mg/day, at least a quarter of subjects experi- if HCTZ is eliminated.
enced gout flares following the discontinuation of prophylac- Fenofibrate (micronized fenofibrate 200 mg/day), a medi-
tic colchicine (Becker et al. 2005). Prophylactic low-dose cation used for the treatment of hyperlipidemia, is also urico-
colchicine therapy should be continued for 6–9 months after suric. This drug decreases serum urate levels by up to 27%
the target serum urate is achieved or until the visible subcu- (Liamis et al 1999). The combination of fenofibrate and
taneous tophi have resolved (Borstad et al. 2004). losartan can have an even greater impact on urate lowering
(Elisaf et al. 1999). Atorvastatin also is uricosuric.
Myth: Urate-lowering therapy should be halted temporarily
Caution should be exercised, including attention to oral
during acute gout flares.
hydration, baseline renal function, and possible uric acid
Reality: Both rising and falling serum urate levels are associ- overproducer status, when considering treatment of any
ated with increased gouty flares, likely due to changes in the hyperuricemic patient with drugs with known substantial uri-
structure of microscopic tophi in the joint space. As a result, cosuric effects.
urate-lowering therapy should not be initiated during an
acute attack, and ongoing urate-lowering therapy should not Pearl: A baby aspirin is the ideal dose of salicylates in gout
be halted during acute gout flares. patients.
If the patient is not on urate-lowering therapy at the time Comment: Low-dose aspirin therapy promotes renal urate
of a gout flare, the clinician should wait until at least 1 week reabsorption and hyperuricemia and can therefore exacerbate
after the acute gouty inflammation has resolved before begin- gout. Decreasing the daily aspirin dose from 325 to 81 mg is
ning a urate-lowering agent. Daily low-dose colchicine, with advisable in the many gout patients who stand to benefit
doses adjusted for renal dysfunction (see Table 36.1), should from low-dose daily aspirin for other reasons.
begin as the acute flare is resolving. The patient should be on
colchicine for at least 1 week before beginning a urate-low- Myth: Wine intake is beneficial for patients with established
ering agent. gout.
Pearl: Show me a gout patient and I’ll show you a person in Reality: Incident gout is significantly lower among drinkers
need of a good internist. of wine than among those who prefer beer or other alcoholic
Comment: Gout is associated with hypertension, metabolic beverages (Choi 2004b). However, this does not mean that
syndrome, diabetes, hyperlipidemia, chronic kidney disease, wine is good for patients with gout. All forms of alcohol con-
diuretic use, and heavy alcohol consumption. Gout and sumption not only elevate serum urate but can also precipi-
serum urate elevations have been linked with increased coro- tate acute gout flares (Zhang 2006b). Wine and other forms
nary artery disease and risk of death due to vascular disease. of alcohol should be used at most in moderation in patients
Several studies have implicated hyperuricemia as an inde- with gout.
pendent risk factor for hypertension, coronary artery disease,
Myth: The institution of a diet that is low in purines is essen-
and death due to vascular disease (Fang and Alderman 2000).
tial to the effective management of gout.
All of these issues require close longitudinal management
and patient education. Lifestyle management that includes Reality: Even severely purine-restrictive diets result in a rel-
weight loss should be an early and forceful recommendation atively small decrease (1–2 mg/dL) in serum urate levels.
for treatment. Furthermore, such diets are unpalatable to many, and there is
no evidence that dietary restriction of vegetables with high
Pearl: Losartan and other commonly used drugs can have a
purine content protects against incident gout (Choi 2004b).
beneficial effect on lowering serum urate.
Dietary modifications are in order for patients who consume
Comment: Thiazide diuretics elevate serum urate levels sig- large amounts of beer, shellfish, and organ meats, because
nificantly and can exacerbate gout. In contrast, losartan, an indiscretions with these foodstuffs can exacerbate gout.
angiotensin II receptor blocker, is uricosuric. In addition, However, a patient with significant hyperuricemia and fre-
losartan raises the urinary pH and thereby reduces the risk of quent gouty episodes, tophi, or uric acid overproduction
uric acid urolithiasis promoted by uricosuric therapy needs a medication designed to reduce the serum urate level
(Shahinfar et al. 2000). Changing the drug from hydrochlo- to below the level of urate solubility (target serum in most
rothiazide to losartan is a reasonable decision if hypertension patients: <6.0 mg/dL).
is controlled adequately with losartan. Weight-reduction diets that stress smaller portion sizes
The uricosuric effect of losartan is only moderate, typi- and counteracting insulin resistance appear promising for
cally leading to only a 10% reduction of serum urate levels. overweight patients with gout (Dessein et al. 2000).
36 Gout Treatment 367

References Kuncl RW, Duncan G, Watson D et al Colchicine myopathy and neu-

ropathy. N Engl J Med 1987; 18; 316(25): 1562–8
Liamis G, Bairaktari ET, Eliasf M. Effects of fenofibrate on serumuric
Acheson R, Chan Y. New Haven survey of joint diseases. The predic- acid levels. Am J Kidney Dis. 1999;34:594–8
tion of serum uric acid in a general population. J Chronic Dis. 1969; Macfarlane D, Dieppe PA. Diuretic induced gout in elderly women. Br
21:543–5 J Rheumatol. 1985;24:155–7
Ahern MJ, Reid C, Gordon TP, et al Does colchicine work? The results of Matzkies F, Berg G, Madl H. The uricosuric action of protein in man
the first controlled study in acute gout. Aust NZJ Med. 1987; 17:301–4 Adv Exp Med Biol. 1980;122A:227–31
Becker MA, Schumacher HR Jr, Wortmann RL, et al Febuxostat com- Niel E, Scherrmann JM. Colchicine today. Joint Bone Spine.
pared with allopurinol in patients with hyperuricemia and gout. N 2006;73:672–8
Engl J Med. 2005;353:2450–61 Perez-Ruiz F, Calabozo M, Pijoan JI, et al Effect of urate-lowering
Borstad GC, Bryant LR, Abel MP, et al Colchicine for prophylaxis of therapy on the velocity of size reduction of tophi in chronic gout.
acute flares when initiating allopurinol for chronic gouty arthritis. J Arthritis Rheum. 2002;47:356–60
Rheumatol. 2004;31:2429–32 Riedel AA, Nelson M, Joseph-Ridge N, et al Compliance with
Buchanan WW, Klinenberg JR, Seegmiller JE. The inflammatory response allopurinol therapy among managed care enrollees with gout: a ret-
to injected microcrystalline monosodium urate in normal, hyperuri- rospective analysis of administrative claims. J Rheumatol. 2004;31:
caemic, gouty and uremic subjects. Arthritis Rheum. 1965;8:361–7 1575–81
Choi HK, Atkinson K, Karlson EW, et al Alcohol intake and risk of inci- Rodnan G. The pathogenesis of aldermanic gout: procatarctic role of
dent gout in men: a prospective study. Lancet 2004a;363:1277–81 fluctuations in serum urate concentration in gouty arthritis provoked
Choi HK, Atkinson K, Karlson EW, et al Purine-rich foods, dairy and by feast and alcohol. Arthritis Rheum. 1980;23:737
protein intake, and the risk of gout in men. N Engl J Med. 2004b;350: Shahinfar S, Simpson RL, Carides AD, et al Safety of losartan in
1093–103 hypertensive patients with thiazide-induced hyperuricemia. Kidney
Clive D. Renal transplant-associated hyperuricemia and gout. J Am Soc Int. 1999;56:1879–85. Published correction appears in 2000;
Nephrol. 2000;11:974–9 57:370
Dessein PH, Shipton EA, Stanwix AE, et al Beneficial effects of weight Siegel LB, Alloway JA, Nashel DJ. Comparison of adrenocorticotropic
loss associated with moderate calorie/carbohydrate restriction, and hormone and triamcinolone acetonide in the treatment of gouty
increased proportional intake of protein and unsaturated fat on arthritis. J Rheumatol. 1994;21:1325–27
serum urate and lipoprotein levels in gout: a pilot study. Ann Rheum Simkin P, Gardner G. Colchicine use in cyclosporine treated trans-
Dis. 2000;59:539–43 plant recipients: how little is too much? J Rheumatol. 2000;27:
Draganescu M, Leventhal LJ. Spinal gout: case report and review of the 1334–7
literature. J Clin Rheumatol. 2004;10:74–9 Suk KS, Kim KT, Lee SH, et al Tophaceous gout of the lumbar spine
Duley JA, Chocair PR, Florin TH. Observations on the use of allopurinol mimicking pyogenic discitis. Spine J. 2007;7:94–9
in combination with azathioprine or mercaptopurine. Aliment Terkeltaub 2008. Semin Arthritis Rheum 2009; 38(6):411–9
Pharmacol Ther. 2005;22:1161–2 Terkeltaub R. Learning how and when to employ uricase as bridge
Edwards NL. Treatment-failure gout: a moving target. Arthritis Rheum. therapy in refractory gout. J Rheumatol. 2007;34:1955–8
2008;58:2587–90 Terkeltaub RA. Clinical practice. Gout. N Engl J Med. 2003;349:
Elisaf M, Tsimilhodimos V, Bairaktari E, et al Effect of micronized 1647–55
fenofibrate and losartan combination on uric acid metabolism in Wilbur K, Makowsky M. Colchicine myotoxicity: case reports and lit-
hypertensive patients with hyperuricemia. J Cardiovasc Pharmacol. erature review. Pharmacotherapy 2004;24:1784–92
1999;34:60–3 Youssef P, Brama T, York H, et al Does renal impairment protect from
Fang J, Alderman MH. Serum uric acid and cardiovascular mortality: gout? J Rheumatol. 1995;22:494–6
The NHANES I epidemiologic follow-up study, 1971–92. JAMA Zhang W, Doherty M, Bardin T, et al EULAR Standing Committee for
2000;283:2404–10 International Clinical Studies Including Therapeutics. EULAR evi-
Gibson T, Rodgers A, Simmonds H, et al A controlled study of diet in dence based recommendations for gout. Part II: Management.
patients with gout. Ann Rheum Dis. 1983a;42:123–7 Report of a task force of the EULAR Standing Committee for
Griebsch A, Zollner N. Effect of ribomononucleotides given orally on International Clinical Studies Including Therapeutics (ESCISIT).
uric acid production in man. Adv Exp Med Biol. 1974;41B:435–42 Ann Rheum Dis. 2006a;65:1312–1324
Hou LC, Hsu AR, Veeravagu A, Boakye M. Spinal gout in a renal trans- Zhang Y, Woods R, Chaisson CE, et al Alcohol consumption
plant patient: a case report and literature review. Surg Neurol. as a trigger of recurrent gout attacks. Am J Med. 2006b;119:
2007;67:65–73 800e13–8
Janssens HJ, Janssen M, van de Lisdonk EH, et al Use of oral predniso-
lone or naproxen for the treatment of gout arthritis: a double-blind,
randomised equivalence trial. Lancet 2008;371:1854–60
 Calcium Pyrophosphate Dihydrate (CPPD)
Crystal Deposition Disease 37
Ann K. Rosenthal and Robert A. Terkeltaub

CPPD arthropathy overlap substantially with those of rheu-

37.1 Overview of CPPD matoid arthritis (RA). CPPD deposition disease can mimic
RA in its inflammatory signs and symptoms as well as in its
› Calcium pyrophosphate dihydrate (CPPD) crystal ability to involve small and large joints in a symmetric man-
deposition disease is a clinically heterogeneous disor- ner (McCarty 1975).
der that is characterized by the presence of intraarticu- A positive rheumatoid factor (RF) occurs in 10% of
lar CPPD crystals. patients with CPPD deposition disease (McCarty 1975). This
› CPPD crystals deposit primarily in cartilage but can reflects the high frequency of false-positive RF assays in
traffic within the joint space. aging. Elevated markers of systemic inflammation are also
› CPPD crystals form in the normally unmineralized peri- common in symptomatic CPPD patients, particularly those
cellular matrix of articular hyaline and fibrocartilage. with polyarticular CPPD deposition disease.
The most common sites of involvement are the knee The other side of the coin is this: patients with longstand-
meniscus, the triangular fibrocartilage of the wrist, and ing RA sometimes develop articular CPPD crystal deposition.
the glenohumeral joint. Asymptomatic disease is com- This is attributable to the occurrence of secondary osteoarthri-
mon in these sites as well as in others (e.g., the symphy- tis, as well as the effects of aging. In fact, in a study of syn-
sis pubis). ovial fluid samples from 93 patients who fulfilled the American
› Pathologic cartilage calcification is promoted by College of Rheumatology (ACR) classification criteria for
changes in inorganic pyrophosphate (PPi) metabolism, RA, 25% contained CPPD crystals (Gerster et al. 2006).
extracellular matrix, and chondrocyte differentiation. The response to therapy may not differentiate CPPD crys-
› CPPD deposition disease has a close and complex rela- tal deposition disease from RA in all cases. Some studies
tionship with osteoarthritis. suggest that patients with refractory, polyarticular CPPD
› CPPD most commonly afflicts the elderly but can occur deposition disease also responds to methotrexate, but this
in younger patients who have a familial variant of the finding is not consistent (Chollet-Janin et al. 2007; Doan
condition or who are afflicted with any of several meta- et al. 2008).
bolic diseases that predispose to CPPD deposition.
Pearl: CPPD deposition disease can mimic a variety of rheu-
› No specific evidence-based therapies for CPPD exist.
matic disorders.
However, the inflammation promoted by CPPD crystal
deposition can be managed successfully in most Comment: CPPD deposition disease can also present as
patients through strategies typically employed for the pseudoseptic arthritis, neuropathic arthritis, polymyalgia
treatment and prophylaxis of gout. rheumatica, masses of the soft tissue and bone, recurrent
hemarthrosis, and of course, pseudogout (Terkeltaub 2008).
Pearl: The “crowned dens syndrome” must be excluded in
an elderly patient presenting with neurological disturbances
37.2 Symptoms and Signs
and a painful cervical mass.

Pearl: CPPD deposition disease can present not only as Comment: Large CPPD crystal deposits can develop in the
pseudogout but also as a polyarticular inflammatory ligamentum flavum or the transverse ligament at C1, and
arthritis. crystal-induced inflammation can occur at this site (Fig. 37.1).
This collection of disease features is termed the “crowned
Comment: In a small but significant fraction of subjects with dens syndrome” (Goto et al. 2007). Patients may present with
CPPD deposition disease, the manifestations of polyarticular meningismus, cervical canal stenosis, myelopathy, and the

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 369

DOI:10.1007/978-1-84800-934-9_37, © Springer Science + Business Media B.V. 2009
370 A. K. Rosenthal and R. A. Terkeltaub

Fig. 37.1 The crowned dens

syndrome. Computed tomogra-
phy scan of the C1 region,
showing calcium deposition
around the odontoid process of
the axis

foramen magnum syndrome. Fracture of the odontoid peg True infectious arthritis is a particular consideration in
precipitated by calcification at C1–C2 has been reported. patients with established CPPD crystal deposition disease.
CPPD crystal deposits in the joint, like those of monosodium
Pearl: CPPD deposition disease may be present but unrelated
urate, are detected in some infected joint fluids. This is because
to the cause of the patient’s joint pain.
CPPD crystals within articular cartilage are “stripmined” by
Comment: The prevalence of CPPD deposition disease is the enzymes associated with joint sepsis.
remarkably high in the elderly patients. However, CPPD
Pearl: CPPD crystals can be hard to find. Keep looking!
deposition disease is asymptomatic or subclinical in many
cases. This disorder is detected often as an incidental finding Comment: CPPD crystals vary in size and shape more than
on radiographs, with calcification of the knee menisci or tri- do urate crystals. They are generally smaller than urate crys-
angular fibrocartilage of the wrist (Figs. 37.2a, b). tals and typically are rods or rhomboids (Fig. 37.3). CPPD
Arthritis in aged patients with preexisting CPPD deposi- crystals demonstrate weakly positive birefringence or some-
tion should not be automatically attributed to CPPD deposi- times no birefringence, in contrast to the consistent bright,
tion disease. Other disorders must be excluded judiciously, as negative birefringence of monosodium urate crystals.
RA and polymyalgia rheumatica can both coexist with CPPD CPPD crystals, notoriously difficult to detect in some syn-
deposition disease and be mimicked by that condition. ovial fluids, are missed or misidentified routinely in clinical
37 Calcium Pyrophosphate Dihydrate (CPPD) Crystal Deposition Disease 371

Fig. 37.2 Calcium pyrophos-

a b
phate dehydrate (CPPD)
deposition within the meniscal
cartilage of the knee (a) and the
triangular fibrocartilage of the
wrist (b) (Courtesy of Dr. Richard
Chou, Dartmouth-Hitchcock
Medical Center)

In one study, two separate examinations of the same synovial

fluid 24 h apart significantly increased the number of fluids in
which CPPD crystals were detected (Yuan et al. 2003). This
phenomenon is probably a function of the time spent search-
ing for crystals (Hearn et al. 1978).
Pearl: Clinical clues help differentiate gout and pseudogout.
Comment: Using the physical examination criteria alone, it
is impossible to differentiate among the three major causes
of monoarticular arthritis: trauma, infection, and crystal
arthritis. Thus, all patients with acute inflammatory monoar-
ticular arthritis should have a joint aspiration. However, cer-
tain clinical clues raise the index of suspicion for
pseudogout.
Fig. 37.3 Calcium pyrophosphate dehydrate (CPPD) crystals in syn-
ovial fluid. CPPD crystals are often weakly positively birefringent and • Pseudogout is more likely to present in a large joint
have a characteristic rectangular shape with blunted ends (Courtesy of (e.g., the knee) or an upper extremity joint than is gout.
Laureen Daft, BS) The classic location for an initial attack of gout of course
is the first metatarsophalangeal joint.
specimens (Gordon et al. 1989; Schumacher et al. 1986; • Blood-tinged synovial fluid is more often seen during
Hasselbacher 1987). Several factors and techniques enhance pseudogout attacks than in acute gout.
the likelihood of finding such crystals, if present: • Compared with gout attacks, attacks of pseudogout take
• the microscopist’s experience and the length of time spent longer to peak in intensity, respond more slowly to therapy
searching for crystals improve the ability to identify them in general, and often demonstrate less complete treatment
correctly responses.
• phase contrast and high magnification enhance the sensi- • When gout and CPPD crystals coexist, which is not
tivity of the procedure uncommon, lowering uric-acid levels may not completely
• using the oil objective permits examination for crystals ameliorate the arthritis (Fig. 37.4).
within synovial fluid leukocytes Pearl: Pseudogout attacks can be provoked reliably by certain
• second looks can yield the diagnosis. well-defined medical procedures.
372 A. K. Rosenthal and R. A. Terkeltaub

Fig. 37.4 Coexistence of monosodium urate and calcium pyrophos- Fig. 37.5 Glucocorticoid crystals in synovial fluid. These brightly
phate dehydrate (CPPD) crystals in a single joint fluid. In this fluid, birefringent crystals are from an intraarticular glucorticoid injection
both monosodium urate and CPPD crystals are seen (Courtesy of performed several weeks before this joint aspiration (Courtesy of
Laureen Daft, BS) Laureen Daft, BS)

Table 37.1 Medical and surgical procedures known to trigger CPPD

flares
Arthroscopy
Intra-articular administration of hyaluronan
Parathyroid surgery for hyperparathyroidism
Parenteral administration of granulocyte colony stimulating factor
(G-CSF)
Bisphosphonates

Comment: Pseudogout can be provoked by minor trauma. In

theory, this is the result of a “shaking loose” of CPPD crys-
tals from joint cartilage. However, the precise pathophysiol-
ogy remains unclear. A variety of medical and surgical
procedures can also trigger CPPD flares (Table 37.1).
Fig. 37.6 Talc crystals in synovial fluid. These crystals are seen less
Pearl: Chronic degenerative arthropathy in CPPD crystal frequently as less talc is used in the manufacture of gloves. They typi-
deposition disease commonly affects several joints that are cally appear as round “beach-ball”-like structures (Courtesy of Laureen
spared in classic cases of primary osteoarthritis. Daft, BS)

Comment: The metacarpophalangeal joints, wrists, elbows,

and glenohumeral joints are not typical targets for primary 37.3 Diagnosis
osteoarthritis. However, these joints are affected commonly
in CPPD crystal deposition disease. If an elderly patient with Pearl: Beware CPPD mimics in synovial fluid.
an inflammatory arthritis of these joints does not have RA,
Comment: All positively birefringent particulate matter
CPPD deposition disease is a good bet.
within synovial fluid is not CPPD. Be particularly wary of
Pearl: Tumoral deposits of CPPD crystal can present with crystals that are described as very “brightly” birefringent or
acute attacks of arthritis and behave as a locally aggressive fluids described as having large quantities of crystals. The
but benign chondroid tumor. most common mimics of CPPD crystals include contami-
nants such as glucocorticoids crystals (introduced at earlier
Comment: Tumoral or pseudotophaceous CPPD crystal depo-
joint injections (Fig. 37.5) or talc crystals (from gloves
sition has been reported in periarticular structures such as ten-
(Fig. 37.6) ).
dons, ligaments, bursae, and also in bones such as the temporal
bone, femur, and tibia (close to the joint) (Terkeltaub 2008). Myth: Synovial fluid examinations are worthless unless per-
The clinical presentation can mimic osteonecrosis. Tumoral formed immediately. This is because stored synovial fluid
CPPD crystal deposition sometimes necessitates surgical generates artifactual calcium-containing crystals, including
removal. CPPD crystals.
37 Calcium Pyrophosphate Dihydrate (CPPD) Crystal Deposition Disease 373

Reality: It is a widely held belief that the storage of synovial Table 37.2 The five H club: metabolic causes of CPPD crystal depo-
fluid (e.g., overnight) results in the de novo formation of sition
crystals that are of no clinical significance. Several older Hyperparathyroidism
Hemochromatosis
studies suggested that stored synovial fluids produce brush-
Hypomagnesemia
ite, a calcium-containing crystal (CaHPO4~2H2O), that is Hypophosphatasia
described often as having a star-shaped morphology (Dieppe (Benign Familial) Hypocalciuric hypercalcemia
et al. 1979).
More recent studies do not support the phenomenon of
brushite formation. Galvez et al. (2002) showed that, for syn- Table 37.3 Laboratory evaluation for the younger patient presenting
ovial fluid samples deemed crystal-free on immediate exami- with calcium pyrophosphate dehydrate (CPPD) deposition disease
nation, no new crystals appeared following storage in either Calcium
Phosphate
a refrigerator or freezer. In addition, clinically important
Magnesium
crystals were retained, despite refrigeration for at least 72 h,
Alkaline phosphatase
with only minor losses observed with freezing. Ferritin
These findings support the theory that, based on the Serum iron and total iron binding capacity
known levels of calcium and pyrophosphate in synovial flu- Thyroid stimulating hormone
ids, CPPD crystals are extremely unlikely to form in solution Parathyroid hormone levels (drawn only if the serum calcium level
(Hearn et al. 1978). is abnormal)

Myth: Intracellular CPPD crystals are more significant than

are extracellular ones. detected in patients younger than 60 years, unless the disease
is familial, caused by trauma to a single joint, or associated
Reality: Many clinicians believe that, unless CPPD crystals with one of a handful of metabolic abnormalities. A 90-year-
are intracellular, they are clinically irrelevant (Fig. 37.7). old patient with chondrocalcinosis on a knee film is unlikely
Few studies support this belief. On the contrary, in one study, to have an occult metabolic condition.
neither the clinical presentation nor the presence or absence Young patients should certainly be screened for the
of joint inflammation correlated with the location of CPPD metabolic associations of CPPD crystal deposition, of
crystals (intracellular versus extracellular) (Martinez-Sanchis which hyperparathyroidism, hemochromatosis, and hypo-
and Pascual 2005). magnesemia (Table 37.2) are most common. Gout is also a
Myth: All patients with CPPD deposition disease should well-characterized risk factor for CPPD deposition dis-
undergo an extensive metabolic evaluation. ease. The appropriate laboratory evaluation for these disor-
ders is shown in Table 37.3.
Reality: The three most common risk factors for articular
CPPD deposition are: (1) age; (2) age; and, (3) age. Fifty Myth: Findings on plain radiographs correlate reliably with
percent of nonagenarians have articular CPPD crystals pathological and clinical manifestations in CPPD crystal
(Mitrovic et al. 1988). It is rare for CPPD crystals to be deposition disease.
Reality: Conventional radiography is most often the initial
approach to the evaluation of patients with suspected CPPD
deposition disease. However, in one study that employed
knee arthroscopy, the correlation between radiographic and
pathological findings was only 39% (Fisseler-Eckhoff and
Muller 1992). Improved applications to CPPD deposition
disease of ultrasonography, advanced imaging, and quantita-
tive physicochemical analyses of joint fluid and tissue crys-
tals are needed to understand better the relationships between
the amounts and articular foci of crystal deposition, symp-
toms, and disease outcome.

37.4 Epidemiology
Fig. 37.7 Intra- and extracellular calcium pyrophosphate dehydrate
(CPPD) crystals. Typically, CPPD crystals are seen both intra- and Myth: CPPD deposition disease is equally prevalent in all
extra-cellularly (Courtesy of Laureen Daft, BS) ethnic groups.
374 A. K. Rosenthal and R. A. Terkeltaub

Reality: CPPD deposition disease of the wrist and knee is meniscal cartilage calcification detected this abnormality in
markedly less common among Chinese in Beijing than only 16% of women between the ages of 80 and 89, but in 30%
among whites in Framingham, MA (Zhang et al. 2006). This of women older than 89 (Ellman and Levin 1975). Moreover,
disparity is evident, despite the higher prevalence of knee in a radiographic survey of the hands, wrists, pelvis, and knees,
osteoarthritis in the Chinese population. Relatively high cal- the prevalence of chondrocalcinosis was 44% among individu-
cium in drinking water in Beijing, which leads to long-term als older than 84 years, 36% among 75–84-year-olds, and
suppression of parathyroid hormone production, might 15% among 65–74-year-olds (Wilkins et al. 1983).
explain this observation, at least in part.
Myth: Hypothyroidism is a common metabolic cause of sec-
ondary CPPD crystal deposition disease. 37.5 Pathophysiology
Reality: In case studies, CPPD crystals have been detected in
Pearl: The term “pyrophosphate arthropathy” appropriately
joint effusions of some patients with severe hypothyroidism,
describes the chronic degenerative changes in cartilage that
a condition that by itself can cause arthropathy. However,
are associated with CPPD crystal deposition.
both hypothyroidism and CPPD crystal deposition are highly
prevalent among older individuals. This fact alone probably Comment: Altered concentrations of calcium, PPi, and the
explains the frequent coexistence of these two conditions. solubility products of these ions clearly promote CPPD crys-
Indeed, several cross-sectional studies of patients with radio- tal formation. The levels of magnesium in cartilage and the
graphically detected CPPD deposition disease have failed to composition of the extracellular matrix also influence the
detect a significant association of hypothyroidism with chon- dynamics of CPPD crystal formation. These factors deter-
drocalcinosis (Visinoni et al. 1993; Job-Deslandre et al. mine the ratio of monoclinic to triclinic CPPD crystals. This
1993; Komatireddy et al. 1989). ratio is important, because monoclinic CPPD crystals are
more inflammatory.
Myth: The clinical presentation of familial CPPD crystal
Both calcium and PPi also exert important effects on gene
deposition disease is homogeneous.
expression, cell differentiation, and viability in chondrocytes.
Reality: Familial CPPD deposition disease is best character- Excess quantities of PPi leads to the induction of matrix met-
ized in Chile and Spain (Zaka et al. 2006; Terkeltaub 2008). alloproteinase-13 and the promotion of apoptosis (Terkeltaub
Familial CPPD crystal deposition disease usually presents in 2008). Hence, “pyrophosphate arthropathy” appropriately
the third and fourth decades of life, but familial disease can describes some of the chronic degenerative changes in carti-
also be detected before the age of 20 or into late middle age. lage in CPPD crystal deposition disease. These changes are
The ANKH gene, which encodes a transmembrane pro- not simply due to the toxic effects on chondrocytes and proin-
tein that channels PPi into and out of the chondrocyte, appears flammatory effects of CPPD crystals.
to be important in the pathophysiology of chondrocalcinosis
Pearl: Altered PPi transport by ANKH is central to the patho-
(Zaka 2006; Terkeltaub 2008). Some kindreds with linkages
genesis of both idiopathic and familial CPPD crystal deposi-
to ANKH on chromosome 5p manifest early-onset polyar-
tion disease.
thritis. The knees and wrists are affected most commonly,
but ankylosing intervertebral disease and involvement of the Comment: As noted earlier, ANKH encodes a transmem-
symphysis pubis and sacroiliac joints are also described. brane protein that facilitates the transport of PPi in and out of
Chondrocalcinosis of late onset has also been reported in the chondrocyte (Rosenthal 2006; Zaka et al. 2006; Terkeltaub
some kindreds. In an English kindred with CPPD disease 2008). “Gain of function” alterations in intrinsic ANKH PPi
linked to ANKH mutation on chromosome 5p, recurrent channeling activity could lead to chronic, low-grade “PPi
childhood seizures were associated with development of leak” from chondrocytes.
CPPD deposition disease in later life. Familial CPPD deposi- ANKH is increased in osteoarthritic and chondrocalcino-
tion disease can involve more than one joint and manifests a tic cartilage chondrocytes. In addition, homozygosity for a
level of clinical intensity comparable with that seen in idio- single nucleotide substitution (-4 G to A) in the ANKH
pathic CPPD deposition disease. Premature osteoarthritis is 5’-untranslated region that promotes increased ANKH expres-
common but not universal. sion was present in approximately 4% of British subjects with
the diagnosis of idiopathic chondrocalcinosis of aging (Zhang
Pearl: Most elderly patients with chondrocalcinosis of the
et al. 2005). Hence, some patients with late-onset CPPD crys-
knee also have radiographic evidence of this disorder in
tal deposition disease appear to have a familial component.
other joints.
Two major chromosomal linkages, 8q and 5p, have been
Comment: The prevalence of chondrocalcinosis increases identified in studies of familial CPPD deposition disease.
progressively with aging. As an example, one study of knee Chromosome 5p-linked chondrocalcinosis, which is more
37 Calcium Pyrophosphate Dihydrate (CPPD) Crystal Deposition Disease 375

widely distributed than 8q-linked disease, is associated with containing crystals in their joint fluids, have been reported to
ANKH mutations. However, ANKH is not the only modula- have more severe radiographic scores and require knee-replace-
tor of PPi metabolism that is linked with idiopathic chondro- ment surgery more often than those with primary osteoarthritis
calcinosis. The disorder also is characterized by excessive who do not have crystals (Rosenthal 2006). However, this con-
NPP1 activity, which leads to the generation of PPi by chon- clusion contrasts with the results of two magnetic-resonance
drocytes (Terkeltaub 2008). Finally, the chondrocyte in idio- imaging studies of knee osteoarthritis in which the relationship
pathic chondrocalcinosis responds abnormally to growth between chondrocalcinosis and the course of osteoarthritis was
factors insulin-like growth factor-1 and transforming growth evaluated longitudinally (Neogi et al. 2006). Knees that dem-
factor, with increased PPi production (Rosenthal 2006). onstrated chondrocalcinosis had a decreased or unchanged risk
of cartilage loss, compared with those without chondrocalcino-
sis (Neogi et al. 2006). When interpreting these studies, it is
37.6 Treatment and Outcomes important to remember that MRI is neither sensitive nor spe-
cific for the imaging of CPPD deposition.
Similarly, in a study of 102 consecutive patients sent for
Pearl: Systemic glucocorticoids are useful for treatment of
total knee replacement, CPPD crystal deposition disease was
acute pseudogout.
identified in 53%. No associations were observed between
Comment: Several studies support use of systemic glucocor- chondrocalcinosis and overall joint function or age at arthro-
ticoids for both gout and pseudogout in the short term (Roane plasty (Viriyavejkul et al. 2006). The dysregulated cartilage
et al. 1997; Alloway et al. 1993). Most studies to date have matrix repair that generates CPPD crystal deposition may be
employed either one or two 60 mg intramuscular injections a biomarker for processes that retard the progression of car-
of triamcinolone or oral prednisone for up to 10 days. tilage tissue loss in some patients with osteoarthritis.
On the other hand, prolonged courses or using high doses In summary, there is no established direct correlation
of prednisone have not been studied thoroughly in acute between extent of cartilage calcification and progression of
pseudogout. Such approaches may create more problems CPPD deposition arthropathy.
than they solve, particularly in the elderly population that is
at greatest risk for CPPD deposition disease. Thus, glucocor-
ticoid bursts are helpful in treating acute flares of microcrys-
talline disease, but “maintenance” prednisone is a bad idea. References
Pearl: Intraarticular glucocorticoids are the most effective Alloway J, Moriarty M, Hoogland Y, et al Comparison of triamcinolone
and often the safest therapy for acute CPPD deposition dis- acetonide with indomethacin in the treatment of acute gouty arthri-
ease-associated arthritis. tis J Rheumatol. 1993;20:111–3
Chollet-Janin A, Finckh A, Dudler J, et al Methotrexate as an alterna-
Comment: A variety of therapeutic options for the treatment tive therapy for chronic calcium pyrophosphate deposition disease:
of acute gout exist. Many of these same therapies are useful an exploratory analysis. Arthritis Rheum. 2007;56:688–92
for pseudogout, as well, but tend to be less effective in Dieppe P, Crocker P, Corke C, et al Synovial fluid crystals. Q J Med.
1979;48:533–55
pseudogout. The risks of colchicine and NSAIDs are not Doan TH, Chevalier X, Leparc JM, et al Premature enthusiasm for the
trivial in elderly patients. Few controlled trials of any therapy use of methotrexate for refractory chondrocalcinosis: comment on
exist, but intraarticular glucocorticoids appear to be more the article by Chollet-Janin et al Arthritis Rheum. 2008;58:2210–1
effective than the alternatives and are certainly safer than Ellman MH, Levin B. Chondrocalcinosis in elderly persons. Arthritis
Rheum. 1975;18:43–7
most systemic therapies (O’Duffy 1976). Glucocorticoids Fisseler-Eckhoff A, Muller KM. Arthroscopy and chondrocalcinosis.
should be used as the treatment of choice unless there is a Arthroscopy 1992;8:98–104
concurrent infection or, in rare instances, when the involved Galvez J, Saiz E, Linares L, et al Delayed examination of synovial fluid
joint is inaccessible to injection. by ordinary and polarized light microscopy to detect and identify
crystals. Ann Rheum Dis. 2002;61:444–7
Despite using intraarticular glucocorticoids, joint aspira- Gerster J, Varisco P, Kern J, et al CPPD crystal deposition disease in
tion, and nonsteroidal antiinflammatory drugs (NSAIDs), the patients with rheumatoid arthritis. Clin Rheumatol. 2006;25:468–9
average duration of a pseudogout attack in one study was 18 Gordon C, Swan A, Dieppe P. Detection of crystals in synovial fluid by
days (Masuda and Ishikawa 1987). Colchicine is useful for light microscopy: sensitivity and reliability. Ann Rheum Dis.
1989;48:737–42
prophylaxis against recurrent pseudogout. Goto S, Umehara J, Aizawa T, et al Crowned dens syndrome. J Bone
Joint Surg Am. 2007;89:2732–6
Myth: Secondary CPPD crystal deposition in cartilage Hasselbacher P. Variation in synovial fluid analysis by hospital labora-
always worsens the outcome of primary osteoarthritis. tories. Arthritis Rheum. 1987;30:637–42
Hearn P, Russell R, Elliott J, et al Formation products of calcium pyro-
Reality: This appears to be true only in a subset of patients with phosphate crystals in vitro and the effect of iron salts. Clin Sci.
osteoarthritis. Patients with knee osteoarthritis, with calcium- 1978;54:29P
376 A. K. Rosenthal and R. A. Terkeltaub

Job-Deslandre C, Menkes CJ, Guinot M, et al Does hypothyroidism Terkeltaub R. Firestein GS, Budd RC, Harris ED, Jr., Mclnnes IB,
increase the prevalence of chondrocalcinosis? Br J Rheumatol. Ruddy S, Sergent JS. Diseases associated with articular deposition
1993;32:177–9 of calcium pyrophosphate dihydrate and basic calcium phosphate
Komatireddy GR, Ellman MH, Brown NL. Lack of association between crystals. In: Firestein G et al, editor. Kelley’s textbook of rheumatol-
hypothyroidism and chondrocalcinosis. J Rheumatol. 1989;16:807–8 ogy, 8th ed, WB Saunders, Philadelphia, 2008
Martinez-Sanchis A, Pascual E. Intracellular and extracellular CPPD Viriyavejkul P, Wilairatana V, Tanavalee A, et al Comparison of charac-
crystals are a regular feature in synovial fluid from uninflamed joints teristics of patients with and without calcium pyrophosphate dihy-
of patients with CPPD related arthropathy. Ann Rheum Dis. 2005; drate crystal deposition disease who underwent total knee
64:1769–72 replacement surgery for osteoarthritis. Osteoarthr Cartil 2006;13:
Masuda I, Ishikawa K. Clinical features of pseudogout attack: a survey 232–5
of 50 cases. Clin Orth Rel Res. 1987;229:173–81 Visinoni RA, Ferraz MB, Furlanetto RP, et al Hypothyroidism and
McCarty DJ. Diagnostic mimicry in arthritis: patterns of joint involve- chondrocalcinosis: new evidence for lack of association between the
ment associated with calcium pyrophosphate dihydrate crystal 2 pathologies. J Rheumatol. 1993;20:1991–2
deposits. Ann Rheum Dis. 1975;25:804–9 Wilkins E, Dieppe P, Maddison P, et al Osteoarthritis and articular
Mitrovic D, Stankovic A, Iriarte-Borda O, et al The prevalence of chon- chondrocalcinosis in the elderly. Ann Rheum Dis. 1983;42:
drocalcinosis in the human knee joint. An autopsy survey. J 280–4
Rheumatol. 1988;15:633–41 Yuan S, Bien C, Wener M, et al Repeat examination of synovial fluid for
Neogi T, Nevitt M, Niu J, et al Lack of association between chondrocal- crystals: Is it useful? Clin Chem. 2003;49:1562–3
cinosis and increased risk of cartilage loss in knees with osteoarthri- Zaka R, Williams CJ. Role of the progressive ankylosis gene in carti-
tis: results of two prospective longitudinal magnetic resonance lage mineralization. Curr Opin Rheumatol. 2006;18:181–6
imaging studies. Arthritis Rheum. 2006;54:1822–8 Zhang Y, Johnson K, Russell RG, et al Association of sporadic chondro-
O’Duffy J. Clinical studies of acute pseudogout attacks. Comments on calcinosis with a -4-basepair G-to-A transition in the 5’-untranslated
prevalence, predisposition and treatment. Arthritis Rheum. 1976;19 region of ANKH that promotes enhanced expression of ANKH pro-
(Suppl):349–52 tein and excess generation of extracellular inorganic pyrophosphate.
Roane D, Harris M, Carpenter M, et al Prospective use of intramuscular Arthritis Rheum. 2005;52:1110–7
triamcinolone in pseudogout. J Rheumatol. 1997;24:1168–70 Zhang Y, Terkeltaub R, Nevitt M, et al Lower prevalence of chondrocal-
Rosenthal AK. Calcium crystal deposition and osteoarthritis. Rheum cinosis in Chinese subjects in Beijing than in white subjects in the
Dis Clin North Am. 2006;32:401–12 United States: the Beijing Osteoarthritis Study. Arthritis Rheum.
Schumacher H, Sieck M, Rothfuss S, et al Reproducibility of synovial 2006;54:3508–12
fluid analyses. Arthritis Rheum. 1986;29:770–4
 Inflammatory Eye Disease
38
James T. Rosenbaum and George N. Papaliodis

or Wegener’s granulomatosis (WG). In the context of the

38.1 Overview of Inflammatory term used by ophthalmologists, retinal vasculitis can:
Eye Disease
• Involve arteries, veins, or both
• Be categorized as either occlusive or nonocclusive
› The eye is frequently the target of immune-mediated
• Be limited to the peripheral retina or a focal area of the retina.
inflammation, either as the sole organ involved or as
part of a multiorgan system disease. The systemic diseases that are most commonly associated
› Systemic inflammatory disorders tend to involve the with retinal vasculitis are sarcoidosis, Behcet’s disease, and
eye in characteristic manners. As examples, disorders multiple sclerosis (MS).
associated with HLA-B27 often cause a highly symp-
Pearl: All patients with new diagnoses of scleritis should
tomatic anterior uveitis; Behcet’s disease can cause a
have a serum assay for antineutrophil cytoplasmic antibod-
subclinical but ultimately vision-threatening posterior
ies (ANCA).
uveitis; and Wegener’s granulomatosis is associated
with scleritis, peripheral ulcerative keratitis, and orbital Comment: Autoantibodies play a relatively small role in the
pseudotumor. evaluation of patients with scleritis (Fig. 38.2). A detailed
› The diagnosis of systemic disease can therefore be history is by far the most useful tool in establishing a sys-
inferred from the specific types of ocular involvement. temic disease as the underlying cause of the eye inflamma-
› The uveal tract is divided into anterior, intermediate, tion (Smith et al. 2007). Antinuclear antibody (ANA) assays
and posterior segments. seldom yield anything other than “false-positive” results. In
› “Anterior uveitis” and “iritis” are synonyms. a patient with scleritis, a positive ANA assay would not
establish the diagnosis of lupus in the absence of other evi-
dence for that disease, because scleritis is not a typical fea-
ture of lupus and multiple other conditions can be associated
Myth: Patients with retinal vasculitis should be evaluated with a positive ANA. Rheumatoid arthritis certainly causes
thoroughly for systemic vasculitis. scleritis, but severe ocular inflammation is seldom the initial
disease manifestation.
Reality: “Retinal vasculitis” is a widely misunderstood term.
Ophthalmologists use vasulitis in a manner that differs sig-
nificantly from usage by rheumatologists (Rosenbaum et al.
1991). Rheumatologists generally rely on histopathological
confirmation or the finding of a classic angiographic appear-
ance in medium-sized arteries to make the diagnosis of vas-
culitis. In contrast, ophthalmologists rarely biopsy the retina
owing to the morbidity of this procedure. Ophthalmologists
diagnose retinal vasculitis either by the appearance of the eye
on fundoscopy or by fluorescein angiography. Leakage of
fluorescein dye is sufficient evidence for an ophthalmologist
to diagnose “vasculitis” (Fig. 38.1).
Fig. 38.1 Retinal
Endothelial cells are highly diverse. The microvascula- vasculitis (Courtesy
ture of the retina differs substantially from the blood vessels of Dr. James
typically involved in disorders such as polyarteritis nodosa Rosenbaum)

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 377

DOI:10.1007/978-1-84800-934-9_38, © Springer Science + Business Media B.V. 2009
378 J. T. Rosenbaum and G. N. Papaliodis

Fig. 38.2 Scleritis (Courtesy of Dr. John Stone)

On the other hand, ANCA assays can be extremely useful

in the evaluation of patients with scleritis. Among the
“ANCA-associated” vasculitides, WG is by far the most
likely to cause scleritis (Hoang et al. 2008). (Microscopic
polyangiitis and the Churg–Strauss syndrome rarely cause
Fig. 38.3 Episcleritis (Courtesy of Dr. John Stone)
this ocular condition). In the WG Etanercept Trial, 8 of the
180 patients (5%) had scleritis (Mahr et al. 2008). WG often
causes scleritis early in the disease course, when manifesta-
tions of “full-blown” disease are absent.

Pearl: The development of peripheral ulcerative keratitis is

associated with the presence of an occult systemic disease
that may be lethal if not identified and treated aggressively.
Comment: Peripheral ulcerative keratitis (PUK) is a devas-
tating disease in which progressive destruction of the periph-
eral cornea leads to perforation of the globe and predisposes
the patient to ocular infections. The most common disease
entities that cause PUK are rheumatoid arthritis and WG.
Pearl: Approximately 50% of patients with WG have oph-
Fig. 38.4 CMV vasculitis (Courtesy of Dr. James Rosenbaum)
thalmologic manifestations at some point in their course.
Comment: WG can involve essentially any ocular tissue Herpes viruses can infect the retina and typically cause
(orbital inflammatory pseudotumor, conjunctivitis, episc- both intraretinal hemorrhages and blood vessels that appear
leritis (Fig. 38.3), scleritis, keratitis, and uveitis). However, abnormal on fluorescein angiography. Thus, these infections
retinal involvement by vasculitis in WG is relatively are often mislabeled as “retinal vasculitis.” Fluorescein
uncommon. angiography usually demonstrates an occlusive vasculopathy
in patients with herpes virus infections. Ironically, the treat-
Pearl: If an ophthalmologist reports that your patient with
ment for “retinal vasculitis” in a patient with WG is usually
WG has retinal vasculitis, the correct diagnosis is probably
to reduce the immunosuppression medications—and treat
an infection.
with an appropriate antimicrobial agent.
Comment: WG very seldom causes a true retinal vasculitis.
Myth: Patients with giant cell arteritis often recover from
The disease has been associated with retinal vasculitis in only
their visual loss with glucocorticoid therapy.
a handful of anecdotal reports. On the other hand, patients
with WG who are being treated actively with immunosuppres- Reality: In a New England Journal of Medicine clinicopatho-
sion—particularly cyclophosphamide and glucocorticoids— logical case conference, the clinician suspected giant-cell
are at substantial risk for an infectious retinitis caused by arteritis in a patient who had suffered monocular visual loss
herpes simplex, herpes zoster, or cytomegalovirus (Fig. 38.4). (Seton et al. 2008). High-dose glucocorticoids were
38 Inflammatory Eye Disease 379

administered and the vision improved. This should have Pearl: The most common cause of intraocular inflammation
alerted the clinician to the fact that the underlying disease throughout the world is toxoplasmosis.
was not giant-cell arteritis (in fact, on autopsy, a dissemi-
Comment: Toxoplasma gondii is an obligate intracellular par-
nated fungal infection was found to be the cause of the
asite found throughout the world that produces a characteristic
patient’s temporary monocular blindness).
retinochoroiditis. This organism is responsible for 30–50% of
Patients with giant-cell arteritis frequently experience
all cases of posterior uveitis in the United States. The primary
episodes of amaurosis fugax, which is (by definition) tran-
site of infection in the eye is the retina or optic nerve, but clini-
sient visual loss. This symptom may improve without
cal findings can include vitritis, retinal vasculitis, and iritis.
therapy—for a while. Once the visual loss secondary to the
ischemia of giant-cell arteritis has been present for 1 day or Myth: Patients who have experienced a single, brief episode
more, however, the ship has sailed: no intervention will lead of uveitis do not require further evaluation.
consistently to any substantial restoration of vision (Aiello
et al. 1993). In fact, marked improvement in acuity follow- Reality: Many ophthalmologists regard a self-limited case of
ing the institution of glucocorticoids therapy should call uveitis the way that rheumatologists regard one episode of
into question the accuracy of the underlying diagnosis. podagra: generally not worthy of further evaluation or treat-
ment. On the contrary, even a single episode of uveitis that
Myth: A normal erythrocyte sedimentation rate effectively resolves on its own requires a thorough medical evaluation.
excludes giant-cell arteritis. The most common systemic disease associated with
uveitis in North America and Europe is either ankylosing
Reality: The great majority of patients with giant-cell arteri-
spondylitis or a related spondyloarthropathy. The uveitis that
tis have elevated sedimentation rates (ESRs), but biopsy-
accompanies this family of systemic conditions is anterior,
proven disease can occur in patients whose ESRs are normal
unilateral, and self-limited; i.e., the eye inflammation resolves
(Salvarani and Hunder 2001). Moreover, at least, one group
completely within several months even without treatment.
has found that patients who experience visual loss tend to
Two thirds of the patients with the acute onset of unilat-
have lower (but not necessarily normal) ESRs relative to
eral anterior uveitis in the setting of a spondyloarthropathy
patients with GCA who do not suffer visual loss (Lopez-Diaz
are not aware that their chronic low-back pain is caused by
et al. 2008). One explanation for this is that clinicians are
sacroiliitis (Rosenbaum 1989). Thus, the first episode of iri-
less likely to suspect GCA in a patient who does not have a
tis presents an opportunity to recognize a treatable cause of
strikingly high ESR. Anterior ischemic optic neuropathy and
chronic low-back pain. HLA-B27 testing and sacroiliac
its permanent blindness in the affected eye may be the symp-
imaging are appropriate parts of the evaluation of a patient
tom that announces the presence of GCA.
with unilateral anterior uveitis.
A normal ESR in the setting of a convincing clinical his-
tory for giant-cell arteritis should not dissuade the clinician Pearl: Intraocular CNS lymphoma is an uncommon malig-
from obtaining a temporal artery biopsy (and even treating nancy but should be considered in older patients who present
the patient empirically until the biopsy is obtained). with new onset vitritis.
Pearl: Hoarseness is a possible initial manifestation of giant- Comment: Documentation of intense intraocular inflamma-
cell arteritis. tion (in the absence of pain or photophobia), subretinal infil-
trates, and lack of a sustained response to glucocorticoid
Comment: A rare but established manifestation of giant-cell treatment should raise concern about the possibility of
arteritis is hoarseness. Other less well-appreciated manifes- intraocular lymphoma. A high index of suspicion is neces-
tations of GCA include sore throat, nonproductive cough, or sary to avoid a delay in diagnosis. The average time between
an ulceration of the tongue. onset of ocular symptoms and definitive diagnosis is approx-
Pearl: The aggressive use of topical glucocorticoids early in imately 21 months (Whitcup et al. 1993).
the course of an iritis exacerbation minimizes the duration of Pearl: The treatment of infectious uveitis requires the appro-
the event and potential for ocular injury. priate antimicrobial agent but often also necessitates the use
of systemic glucocorticoids.
Comment: Increasing the dosing frequency of a topical
glucocorticoid enhances the bioavailability in the anterior Comment: Multiple infectious diseases can involve ocular
chamber. The package insert for prednisolone acetate and structures. The cornerstone of therapy for eye infections is
prednisolone sodium phosphate suggests BID to QID dosing the appropriate antimicrobial agent. The host inflammatory
for ocular inflammation. Most ocular inflammatory disease response to the infectious agent and the reaction after initia-
specialists prescribe the use of topical glucocorticoids every tion of antimicrobial treatment can induce significant ocular
1 or 2 h at the onset of an acute exacerbation. injury, including cystoid macular edema, vitritis, retinitis,
380 J. T. Rosenbaum and G. N. Papaliodis

papillitis, and choroiditis. The use of systemic glucocorti- Comment: There are many reports in the literature implicat-
coids, although counterintuitive given the etiology of the ing drugs as causal agents for uveitis. These medications
ocular inflammation, is often required. include topical agents, oral medications, intravenous prepa-
rations, and vaccinations. The agents most commonly asso-
Pearl: A normal serum level of angiotensin converting
ciated with uveitis are the bisphosphonates, rifampin,
enzyme (ACE) does not exclude sarcoidosis as the cause of
rifabutin, and cidofovir.
uveitis.
Pearl: Thyroid disease is the most common cause of orbital
Comment: Ocular disease is nearly as common as pulmonary
inflammatory disease.
disease as the initial manifestation of sarcoidosis. ACE lev-
els are elevated in approximately 60% of patients with pul- Comment: Several systemic immune-mediated diseases
monary sarcoidosis. Early sarcoidosis and disease localized can cause orbital inflammation (Lutt et al. 2008). Orbital
to the eye are associated with a relatively small “burden” of inflammatory disease results in pain, exophthalmos,
granulomas, the site of ACE production. Thus, serum ACE restricted eye motility, diplopia, and visual loss caused by
levels are often normal even when sarcoidosis is the cause of compression of the optic nerve (Fig. 38.5). The most com-
ocular inflammatory disease. mon cause of orbital inflammation is Graves’ disease. The
specific diagnosis of thyroid-associated orbital inflamma-
Pearl: A computed tomography (CT) scan of the chest is tory disease is sometimes strongly supported by imaging
indicated in the evaluation of patients with uveitis. (e.g., MRI) and by testing for thyroid autoantibodies. In
Comment: Sarcoidosis is a common cause of uveitis. In fact, some instances, an orbital biopsy is required to establish
ocular disease is nearly as common as pulmonary disease as the diagnosis.
the initial manifestation of sarcoidosis. Many patients with Pearl: Children with uveitis often have asymptomatic disease.
sarcoid-associated uveitis have no pulmonary symptoms,
and chest radiographs often fail to capture hilar or mediasti- Comment: In adults with uveitis, patients’ symptoms include
nal lymphadenopathy. eye pain, eye redness, light sensitivity, and blurry vision.
A study from the Cleveland Clinic found that 57% of older Children with juvenile idiopathic arthritis (JIA) may have
women with uveitis who were otherwise asymptomatic had significant intraocular inflammation without complaints.
adenopathy detectable on chest CT (Kaiser et al. 2002). Ninety percent of patients who have uveitis associated with
Consequently, CT scans of the hilum and mediastinum are con- JIA have no ocular symptoms (Grassi et al. 2007).
sidered appropriate for the evaluation of patients with uveitis Pearl: Children with a form of uveitis known as pars planitis
who do not fit neatly into a specific diagnostic category. often present with vitreous hemorrhage.
Pearl: Penetrating trauma to one eye can induce ocular Comment: The most common form of uveitis in childhood is
inflammation in both eyes. associated with JIA. The uveitis occurs in the oligoarticular
Comment: Sympathetic ophthalmia is a granulomatous
uveitis that occurs after penetrating trauma or after surgical
procedures. The disease is uncommon but devastating due to
its potentially blinding effects in both the inciting and the
sympathizing eye. The presumed mechanism of inducing
ocular inflammation is due to exposure of uveal antigens
(previously sequestered in immunologically privileged sites
in the eye) to the host immune system, with the resultant loss
of immune tolerance.
Pearl: Despite advances in diagnostic capabilities, many
cases of uveitis remain idiopathic.
Comment: Careful ophthalmologic examination with appro-
priate serologic and radiologic studies often lead to a defini-
tive diagnosis of a specific disease that is causing intraocular
inflammation. However, up to 50% of cases of uveitis remain
“idiopathic”; i.e., they cannot be attributed to a particular
underlying disease.
Pearl: Certain medications can cause uveitis. Fig. 38.5 Orbital pseudotumor (Courtesy of Dr. John Stone)
38 Inflammatory Eye Disease 381

Fig. 38.6 Pars planitis Fig. 38.7 Birdshot

(Courtesy of choroidopathy
Dr. George Papaliodis) (Courtesy of
Dr. George Papaliodis)

form of JIA and usually affects young girls whose serum

tests positively for ANA.
Children can also develop a condition called pars planitis, multiple evanescent white-dot syndrome, punctate inner
a form of uveitis associated with inflammatory exudates choroidopathy, multifocal choroiditis, and acute multifocal
overlying the peripheral retina which is known as the pars placoid pigmentary epitheliopathy.
plana. Neovascularization is a common complication of pars “Birdshot choroidopathy” (Fig. 38.7) another white-dot
planitis, particularly in children. These new vessels are fria- syndrome, is typically bilateral. The lesions in birdshot chor-
ble and may bleed. This results in a sudden “shower” obscur- oidopathy are thought to represent focal granulomas. Visual
ing vision. A careful evaluation of a child who suffers this loss with birdshot choroidopathy can be severe, but the pro-
type of bleeding known as a vitreous hemorrhage may reveal cess is responsive to immunosuppression. Approximately
pars planitis (Lauer et al. 2002). Since the blood usually will 95% of patients with birdshot choroidopathy are HLA-A29
not allow an adequate examination of the eye, the first clue to positive (LeHoang et al. 1992), making this disorder one of
the diagnosis may be a careful examination with indirect the best examples of a link between an HLA class I antigen
ophthalmoscopy of the opposite eye, which is free of hemor- and disease susceptibility.
rhage (Fig. 38.6).
Pearl: HLA-B27 is associated with uveitis secondary to
Pearl: Pars planitis is a risk factor for the development of either psoriatic arthritis or inflammatory bowel disease.
MS, just as optic neuritis is a risk factor for that diagnosis. However, the association is not as strong as that between
HLA-B27 and uveitis secondary to ankylosing spondylitis.
Comment: Uveitis has a rare but well-defined association
with MS. Approximately 1% of patients with MS have Comment: Just as HLA-B27 is associated with the subset of
uveitis, and MS patients comprise about 1.5% of all patients patients with psoriatic arthritis who develop sacroiliitis and
with uveitis in tertiary referral clinics. The uveitis of MS can the subset of inflammatory bowel disease (IBD) patients who
take several forms, the most common of which is pars develop sacroiliitis, so HLA-B27 is associated with the sub-
planitis. set of either patients with psoriatic arthritis or IBD who
Patients with MS and pars planitis share common HLA develop uveitis (Paiva et al. 2000; Lyons and Rosenabum
haplotypes. One study found a 20% risk of developing either 1997). Furthermore, just as approximately 50% of patients
MS or optic neuritis within 5 years of the diagnosis of pars with sacroiliitis and either psoriasis or IBD are HLA-B27
planitis (Malinowski 1993). This risk may be even higher if positive, about 50% of patients with either psoriatic arthritis
the patient is female and has the type of perivascular changes or IBD who develop uveitis are HLA-B27 positive.
in the retina that are termed peripheral retinal periphlebitis.
Pearl: Pain is a typical manifestation of scleritis, but pain is
In experimental autoimmune encephalomyelitis, an animal
not typical of episcleritis.
model of demyelinating disease, the animals often develop
uveitis. Comment: The episclera is a superficial layer of tissue that
overlies sclera. Patients who have scleritis often have a con-
Pearl: One of the strongest HLA links to disease susceptibil-
comitant component of episcleritis, but the converse is not
ity is the association of HLA-A29 with birdshot retino-
true; i.e., patients can have episcleritis without scleritis.
choroidopathy.
Intense, boring pain is nearly universal in patients with
Comment: Ophthalmologists recognize a variety of “white- scleritis, but absent in episcleritis. Patients with episcleritis
dot syndromes” that are caused by inflammation in the retina often complain of a vague discomfort in the eye, but it does
and the adjacent choroid. These entities, which lead to focal not rise to the level of “pain.” Both scleritis and episcleritis
whitening of the retina, include serpiginous choroiditis, the cause ocular redness, which may be nodular or elevated and
382 J. T. Rosenbaum and G. N. Papaliodis

may assume the shape of a wedge (see Fig. 38.3). Thus, Laboratory studies indicate a mild anemia, slightly ele-
distinguishing scleritis from episcleritis on the clinical vated hepatic transaminases, and a markedly elevated ESR.
appearance of the eye redness alone (and without the help of Serum creatinine elevations are present in many but not all
a slit-lamp) can be challenging. The presence or absence of patients. However, a sterile pyuria and elevation of the uri-
pain is useful in discriminating the two conditions. nary b-2 microglobulin level are usually detected. The sys-
Episcleritis and scleritis generally have entirely different temic symptoms of TINU usually respond dramatically to
implications. Only a small minority of cases of episcleritis are modest doses of oral glucocorticoids such as 20 mg of pred-
associated with a systemic inflammatory disease, and episc- nisone per day.
leritis does not pose a risk to vision. In contrast, roughly half
Pearl: The differential diagnosis of cotton wool spots includes
of patients with scleritis have an associated systemic disease,
hypertension, diabetes, lupus, and infection with the human
the most common of which is rheumatoid arthritis or WG.
immunodeficiency virus (HIV).
Scleritis can cause intraocular complications such as
glaucoma, uveitis, or optic neuropathy. It can also cause a Comment: Cotton wool spots are retinal exudates that indi-
“corneal melt,” an intensely painful condition that leads to cate local retinal ischemia. In addition to diabetes mellitus
the loss of useful vision in the eye (Fig. 38.8). and diabetic retinopathy, cotton spots are seen commonly in
patients with hypertension, lupus, or HIV. One study of lupus
Pearl: The syndrome of fever, arthralgias, renal dysfunction,
patients followed for as long as 16 years reported that 7%
and the sudden onset of bilateral anterior uveitis is often due
had cotton wool spots at some time (Stafford-Brady et al.
to tubulointerstitial nephritis with uveitis (TINU), a signifi-
1988). Patients with SLE and cotton wool spots were more
cantly underappreciated disorder.
likely to have a lupus anticoagulant, central nervous system
Comment: The differential diagnosis for uveitis in associa- disease, and a poor prognosis.
tion with systemic symptoms includes a broad variety of
Myth: Hydroxychloroquine is a dangerous medication and
infections and idiopathic inflammatory conditions. One dis-
should be avoided due to ocular toxicity.
ease that is arguably the least recognized, but perhaps, the
most common explanation of systemic symptoms and uveitis Reality: The retinal toxicity of hydroxychloroquine is depen-
is TINU (Mackensen et al. 2007). This disease usually occurs dent on dose and duration of treatment. In a prospective
in patients who are younger than 20 years of age. TINU is the cohort study of patients with SLE and RA who were treated
second most common cause of uveitis among children in with hydroxychloroquine at doses less than or equal to
Japan (Goda et al. 2005). 6.5 mg/kg/day, no retinal toxicity was reported within the
The uveitis of TINU is sudden in onset and typically first 6 years of treatment (Mavrikakis et al. 2003). Among
involves both anterior chambers simultaneously. However, patients treated with more than 6 years of hydroxychloro-
inflammation may begin in one eye several days before the quine, 3.4% of Patients with maculopathy caused by the
other is involved clinically. TINU patients experience substan- medication, which stabilized when hydroxychloroquine was
tial systemic symptoms, including fever, malaise, and arthral- stopped.
gias. The systemic symptoms may precede the uveitis or begin The principal risk factors for hydroxychloroquine toxicity
simultaneously with the eye disease. Rarely, the systemic appear to be the use of high doses (i.e., >6.5 mg/kg/day), long
symptoms follow the eye disease. durations of treatment, and concomitant renal dysfunction.
Myth: Cataract extraction via phacoemulsification (the most
successful surgical technique in modern medicine) is as success-
ful in patients with uveitis as it is in patients without uveitis.
Reality: Advances in cataract surgery and intraocular lens
technology have improved the quality of life for patients
tremendously. Patients with uveitis pose a special chal-
lenge from both a technical and disease-management per-
spective. Uveitic patients often have small pupillary
apertures from episodes of ocular inflammatory disease,
unstable zonules and capsular support. Intraocular inflam-
mation, a normal response to surgical trauma, can be sig-
nificantly exaggerated in uveitis patients. This leads to a
higher incidence of macular edema, posterior capsular
opacification, or intraocular lens intolerance among patients
Fig. 38.8 Corneal melt (Courtesy of Dr. George Papaliodis) with histories of uveitis.
38 Inflammatory Eye Disease 383

In one series of 242 uveitic eyes undergoing cataract sur- have reported that etanercept is not as consistently effective
gery, 10.7% had vision loss compared with preoperative as infliximab for the uveitis associated with Behcet’s disease
visual acuity (Yamane 2007). (Galor et al. 2006). The experience with adalimumab is
promising but remains limited at the time of this writing.
Myth: Syphilis rarely involves the eye, and serologic studies
Although infliximab is often highly effective for uveitis,
to exclude syphilis are unnecessary.
one prospective study found a surprisingly high morbidity
Reality: Syphilis is called the “great imitator” for good rea- from the medication, including drug-induced lupus and
sons. Its ability to involve the eye is no exception. Syphilis thrombosis (Suhler et al. 2005). One hypothesis is that
can cause episcleritis, scleritis, keratitis, uveitis, retinal vas- patients with localized inflammation are more prone to com-
culitis, papillitis, and iris gummas. In addition, syphilis is plications from TNF inhibition than are patients with sys-
enjoying a resurgence. Between 2005 and 2006, the inci- temic inflammation, who in theory have higher serum levels
dence of primary and secondary syphilis increased by nearly of TNF. No TNF inhibitor is approved for the treatment of
12%. Ocular disease occurs in approximately 10% of syphi- uveitis in the United States, but infliximab has this approval
lis cases. in Japan for patients with Behcet’s disease.
Pearl: Although included in the differential diagnosis as a
cause for uveitis, ocular tuberculosis is extremely rare.
Comment: In the 1940s, tuberculosis was estimated to be References
responsible for up to 80% of granulomatous uveitis world-
wide. Today, ocular tuberculosis is extremely rare, even in Aiello P, Trautmann J, McPhee T, et al Visual prognosis in giant cell
regions characterized by a high incidence of tuberculosis. arterits. Ophthalmology 1993;100:550–5
One study of 1,273 patients who presented with uveitis in Biswas J, Narain S, Das D, Ganesh SK. Pattern of uveitis in a referral
uveitis clinic in India. Int Ophthalmol. 1996–1997;20(4) 223–8
India found that tuberculosis was the culprit in only five
Galor A, Perez V, Hammel J, et al Differential effectiveness of etaner-
patient (0.4%) (Biswas et al. 1996). cept and infliximab in the treatment of ocular inflammation.
Ophthalmology 2006;113:2137–2323
Myth: Roth spots are diagnostic of bacterial endocarditis. Goda C, Kotake S, Ichiishi A, et al Clinical features in tubulointerstitial
nephritis and uveitis (TINU) syndrome. Am J Ophthalmol.
Reality: A Roth spot is an intraretinal hemorrhage with a
2005;140:637–41
white center (Fig. 38.9). The white center is a cotton wool Grassi A, Corona F, Casellato A, et al Prevalence and outcome of juve-
spot, which indicates the presence of local ischemia. Septic nile idiopathic arthritis associated uveitis and relation to articular
emboli secondary to subacute bacterial endocarditis are the disease. J Rheumatol. 2007;34(5):1139–45
Hoang L, Lim L, Vaillant B, et al Antineutrophil cytoplasmic antibody-
cause of Roth spots most familiar to medical students.
associated active scleritis. Arch Ophthalmol. 2008;126:651–5
However, lupus can also cause this finding. Kaiser PK, Lowder CY, Sullivan P, et al Chest computerized tomogra-
phy in the evaluation of uveitis in elderly women. Am J Ophthalmol.
Myth: Any tumor necrosis factor (TNF) inhibitor is a good 2002;133:499–505
therapeutic choice for a patient with uveitis who has failed Lauer AK, Smith JR, Robertson JE, et al Vitreous hemorrhage is a com-
other immunosuppressive therapies. mon complication of pediatric pars planitis. Ophthalmology
2002;109:95–8
Reality: Systemic disorders associated with uveitis respond LeHoang P, Ozdemir N, Benhamou A, et al HLA-A29.2 subtype asso-
differently to TNF inhibitors. Ankylosing spondylitis appears ciated with birdshot retinochoroidopathy. Am J Ophthalmol.
1992;113:33–5
to respond rather uniformly to TNF inhibitors, regardless of
Lopez-Diaz M, Liorca J, Gonzalez-Juanatey C, et al The erythrocyte
which one is used. In contrast, several retrospective studies sedimentation rate is associated with the development of visual
complications in biopsy-proven giant cell arteritis. Semin Arthritis
Rheum. 2008;38:116–23
Lutt J, Lim L, Phal P et al Orbital inflammatory disease. Semin Arthritis
Rheum. 2008;37:207–22
Lyons JL, Rosenbaum JT. Uveitis associated with inflammatory bowel
disease compared with uveitis associated with spondyloarthropathy.
Arch Ophthalmol. 1997;115:61–4
Mackensen F, Smith J, Rosenbaum JT. Enhanced recognition, treat-
ment, and prognosis of tubulointerstitial nephritis and uveitis syn-
drome. Ophthalmology 2007;114:995–9
Mahr AD, Neogi T, LaValley MP et al A re-evaluation of item selec-
Fig. 38.9 Roth spot tion and weighting in the Birmingham Vasculitis Activity Score for
(Courtesy of Wegener’s granulomatosis. Arthritis Care Res. 2008;59: 884–91
Dr. George Malinowski SM, Pulido JS, Folk JC. Long-term visual outcome and
Papaliodis) complications associated with pars planitis. Ophthalmology
1993;100:818–24
384 J. T. Rosenbaum and G. N. Papaliodis

Mavrikakis I, Sfikakis PP, Mavrikakis E, et al The incidence of irrevers- and visual changes after liver transplantation. N Engl J Med. 2008;
ible retinal toxicity in patients treated with hydroxychloroquine: a 358:2619–28
reappraisal. Ophthalmology 2003;110(7):1321–6 Smith J, Mackensen F, Rosenbaum JT. Therapy insight: scleritis and its
McCannel CA, Holland GN, Helm CJ, et al Causes of uveitis in the gen- relationship to systemic autoimmune disease. Nat Clin Pract 2007;
eral practice of ophthalmology. Am J Ophthalmol. 1996: 121:35 3:219–26
Paiva ES, Macaluso DC, Edwards A, et al Characterisation of uveitis in Stafford-Brady FJ, Urowitz MB, Gladman DD, et al Lupus retinopathy.
patients with psoriatic arthritis. Ann Rheum Dis. 2000;59: 67–70 Patterns, associations, and prognosis. Arthritis Rheum. 1988; 31:
Rosenbaum JT. Characterization of uveitis associated with spondyloar- 1105–10
thritis. J Rheumatol. 1989;16:792–96 Suhler EB, Smith JR, Wertheim MS, et al A prospective trial of inflix-
Rosenbaum JT, Robertson JE, Watzke RC. Retinal vasculitis: a primer. imab therapy for refractory uveitis: preliminary safety and efficacy
West J Med. 1991;154:182–5 outcomes. Arch Ophthalmol. 2005;123:903–12
Salvarani C, Hunder GG. Giant cell arteritis with low erythrocyte sedi- Whitcup SM, deSmet MD, et al Intraocular lymphoma. Clinical and
mentation rate: frequency of occurrence in a population-based study. histopathological diagnosis. Ophthalmology 1993;100:1399–406
Arthritis Rheum. 2001;45:140–5
Seton M, Pless M, Fishman J, et al Case records of the Massachusetts
General Hospital. Case 18–2008. A 68 year old man with headache
 Immune-Mediated Inner Ear Disease
39
Yuri Agrawal and Howard W. Francis

39.1 Overview of Immune-Mediated Inner › For patients with severe or recurrent audiovestibular
Ear Disease dysfunction secondary to IMIED, it is not clear whether
long-term courses of high-dose glucocorticoids and
additional immunosuppressive agents are effective in
› Immune-mediated inner ear disease (IMIED) is the restoring inner ear function. Such treatment approaches
preferred term for the syndrome involving cochlear should be used cautiously.
dysfunction, leading to degrees of sensorineural hear-
ing loss and vertigo.
› Although vasculitis has been inferred to be the cause of
IMIED in some cases, histopathological proof of vascu-
litis involving the inner ear in this condition is lacking. 39.2 Definition
The condition does appear to be mediated in some fash-
ion by the immune system; hence, the name IMIED. Pearl: IMIED is the preferred term for the syndrome of sen-
› IMIED can occur in association with a recognized sorineural hearing loss, tinnitus, and vertigo.
autoimmune condition such as Sjögren’s syndrome,
Wegener’s granulomatosis, or giant cell arteritis, or as Comment: In 1979, McCabe described a case series of 18
an isolated entity in which no other organ dysfunction patients with rapidly progressive, bilateral, asymmetric hear-
is evident. ing loss (McCabe 1979). These patients’ hearing loss was
› The onset of hearing loss in IMIED is rapid compared often accompanied by tinnitus and vertigo. In many cases,
with other entities that lead to auditory dysfunction. the patients’ symptoms appeared to respond to the combina-
Hearing loss in IMIED occurs over the course of weeks tion of glucocorticoids and cyclophosphamide.
to months. McCabe termed this syndrome “autoimmune sensorineu-
› Patients with IMIED-associated hearing loss have prob- ral hearing loss,” leading to the recognition of this syndrome
lems with both hearing acuity and word discrimination. as a distinct clinical entity. Because hearing loss is often not
Following conversations in noisy environments (e.g., in the sole feature of this disorder and evidence of a true auto-
crowded restaurants) may be particularly challenging. immune basis for this disease is lacking, the term IMIED is
› For patients with IMIED, vestibular problems that a more appropriate designation (Stone and Francis 2000).
result from cochlear damage can be as challenging as Pearl: “Sympathetic otitis” is an antiquated name for IMIED.
the auditory dysfunction, if not more so.
› Up to 70% of patients diagnosed with IMIED demon- Comment: Anticochlear antibodies were hypothesized to
strate clinical improvement following glucocorticoid contribute to a subset of cases of bilateral sensorineural hear-
treatment. Prednisone (1 mg/kg/day) is a reasonable ing loss as early as in the 1950s. Further support for immune
starting dose for the treatment of rapidly progressive processes in the inner ear came from the phenomenon of
sensorineural hearing loss secondary to IMIED. “sympathetic otitis,” whereby patients who had sustained an
› If the patient regains significant auditory and vestibu- injury in one ear (either traumatic or iatrogenic) later experi-
lar function within 2 weeks, tapering of the prednisone enced hearing loss in the contralateral ear (Kikuchi 1959).
dose can begin in a taper designed to discontinue the The presumed mechanism of this was the exposure of previ-
medication over 2 months. ously cryptic inner ear antigens as a result of injury, leading
to activation of the immune system against these antigens

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 385

DOI:10.1007/978-1-84800-934-9_39, © Springer Science + Business Media B.V. 2009
386 Y. Agrawal and H. W. Francis

Fig. 39.1 Anatomy of the

temporal bone and audiovestibu-
lar apparatus. AN auditory nerve;
C cochlea; ES endolymphatic sac;
OC ossicular chain; SCC
semicircular canals; V vestibule.
From (Stone and Francis 2000).
Reprinted from Current Opinion
in Rheumatology with kind
permission from Wolters Kluwer
Health

and autoimmune destruction of the remaining healthy inner syndrome, relapsing polychondritis, and sarcoidosis (Stone
ear tissue (ten Cate and Bachor 2005; Schindler and Niparko and Francis 2000). Systemic vasculitides such as Wegener’s
1998). granulomatosis, polyarteritis nodosa, and Behcet’s disease
can be associated with sensorineural hearing loss, but the true
Pearl: The main blood supply to the inner ear, the labyrin-
prevalence of IMIED in these disorders is not known with
thine artery, is an end-artery that has minimal collateral
certainty.
anastomoses.
Comment: The inner ear, housed within the temporal bone of Myth: Wegener’s granulomatosis is a common cause of
the skull, consists of an audiovestibular apparatus collectively IMIED.
referred to as the labyrinth (Fig. 39.1). The principal vascular Reality: Ear involvement occurs in a significant percentage
supply to the inner ear is the labyrinthine artery, which is a of patients with Wegener’s granulomatosis. However, the
branch of the anterior inferior cerebellar artery. The labyrin- usual focus of ear dysfunction in Wegener’s granulomatosis
thine artery, which terminates within the structures of the is the middle ear rather than the inner ear (Kornblut et al.
labyrinth, lacks collateral anastomoses. This renders the inner 1982a, b). Sinonasal or nasopharyngeal disease in Wegener’s
ear particularly vulnerable to vascular pathologies such as granulomatosis can lead to Eustachian tube dysfunction, serous
thromboembolism, ischemia, and vasculitis. otitis media, and conductive hearing loss (see Chap. 24).
Myth: IMIED always occurs in the context of a systemic vas- Destructive granulation tissue located in the middle ear itself
culitis or autoimmune disease. can also lead to chronic otitis media and conductive hearing
loss in Wegener’s granulomatosis. These complications are
Reality: IMIED is associated with systemic autoimmune much more common in Wegener’s granulomatosis than is
conditions in only about 15–30% of cases (Bovo et al. 2006). sensorineural hearing loss.
Thus, IMIED is more likely than not to be a primary inner
Myth: The temporal bone histopathologic changes observed
ear disease process, without any clear link to an associated
in IMIED reflect a form of vasculitis that is localized to the
systemic disorder.
inner ear.
A large handful of systemic disorders can be associated
with sensorineural hearing loss that closely resembles those Reality: Very few human temporal bones from patients with
found in IMIED. Sensorineural hearing loss is an essential IMIED have been examined early in the disease course.
feature of Cogan’s syndrome (see Chap. 31) and also occurs This is because it is impossible to obtain tissue from the
in patients with systemic lupus erythematosus (SLE), Sjögren’s inner ear at surgery without destroying viable labyrinthine
39 Immune-Mediated Inner Ear Disease 387

function, thereby defeating the purpose of establishing an the right and left vestibular organs have a baseline firing rate.
early diagnosis. If the head is moved towards the right, the hair cells on the
The few temporal bone specimens from patients with right increase their firing rate as those on the left decrease
IMIED that have been investigated are generally from patients theirs. The opposite changes occur when the head moves to
with longstanding disease treated for lengthy periods with the left. The brain compares the input from right and left
glucocorticoids and other immunosuppressive medications. vestibular organs in order to gauge the head position.
Features of the original disease process are difficult to discern In patients with vertigo at rest, there is a baseline imbalance
in these specimens. The available specimens have demon- in vestibular input between the right and left sides, producing
strated four patterns of temporal bone pathology: a false perception of motion. This type of vertigo usually
results from acute unilateral vestibular hypofunction, which
• Endolymphatic hydrops (or fluid-filled distension of the
characteristically undergoes central compensation over time.
scala media). This is similar to the appearance of the inner
In motion-induced vertigo, the brain fails to encode
ear in Meniere’s disease.
changes in head position. This leads to destabilization of gaze
• Acute labyrinthitis resulting in atrophy of hair cells and
and oscillopsia, the sense that stationary objects are moving
their supporting structures.
(Schubert and Minor 2004). Chronic oscillopsia, which typi-
• Proliferation of fibrous tissue or bone (“neo-osteogenesis”).
cally occurs in the setting of bilateral vestibular hypofunc-
• Neuronal atrophy.
tion, has a devastating impact on the patients’ quality of life.
Features characteristic of active vasculitis, specifically infil- Vestibular symptoms in IMIED may fluctuate over time,
tration of the blood vessel walls with inflammatory cells, leu- often in concert with changes in hearing. Imbalance associ-
kocytoclasis, and fibrinoid necrosis, have not been observed. ated with dizziness or vertigo should be distinguished from
that resulting from impaired motor coordination due to pare-
sis, decreased proprioception, or cerebellar deficits.
39.3 The Clinical History

Pearl: The time course of symptoms is a key clinical feature

39.4 Physical Findings
of IMIED.
Comment: The onset of hearing loss in IMIED is relatively
Pearl: The vestibular system can be evaluated at the bedside
rapid, occurring over a course of weeks to months. The time
as well as in the otolaryngologist’s office.
course of IMIED distinguishes it from sudden sensorineural
Comment: The function of the vestibular end-organs cannot
hearing loss (e.g., that associated with acute acoustic trauma),
be measured directly either at the bedside or in the labora-
which occurs over hours to days. The time course also distin-
tory. However, routine vestibular tests that evaluate directly
guishes IMIED from slowly progressive hearing loss
observable phenomena such as eye movements are helpful in
(e.g., presbycusis), which evolves over years.
gauging the degree of vestibular dysfunction.
Myth: IMIED affects both ears to the same degree. One of the principal functions of the vestibular system is
Reality: The hearing loss in IMIED is typically bilateral, but to maintain gaze stability during head movements. This pro-
the magnitude of the hearing loss may be asymmetric. cess is known as the vestibulo-ocular reflex. In order to main-
Although one ear can be affected before the other, unilateral tain gaze stability, the vestibular apparatus tracks changes in
disease is followed in most cases by decreased hearing in the head position and communicates this information to the ocul-
contralateral ear within weeks to months. Symptoms may omotor system, which produces compensatory eye move-
also fluctuate over time. ments in the direction opposite to that of the head movement.
The vestibulo-ocular reflex allows the retinal fovea to remain
Pearl: Vestibular symptoms include the illusory sense of fixed upon on an object, even during head motion. Dysfunction
motion at rest or with head movements. of the vestibular organs leads to unstable gaze fixation and
Comment: Patients with dysfunction of their peripheral ves- corrective eye movements, which can be observed on physi-
tibular apparatus typically complain of dizziness. Their cal examination.
symptoms of feeling “dizzy” include vertigo, which is an The principal part of the bedside vestibular examination
illusory sense of motion. Vertigo can occur at rest or with is assessment for spontaneous or gaze-evoked nystagmus
movements of the head. (Minor and Zee 2003). Nystagmus refers to spontaneous and
The vestibular apparatus tracks information about the repetitive eye movements; these can be physiologic in some
head position. It also contributes to the maintenance of gaze cases, pathologic in others (Fig. 39.2). Nystagmus consists of
and postural stability via the vestibulo-ocular and vestibulo- both a slow phase and a fast phase. Spontaneous nystagmus,
spinal reflexes (Schubert and Minor 2004). The hair cells in the correlate of spontaneous vertigo (discussed above), is
388 Y. Agrawal and H. W. Francis

Fig. 39.3 Sample audiogram of a patient with right sensorineural hear-

ing loss. Bone conduction thresholds (R = [, L =]) are measures of audi-
tory function of the cochlea and proximal neural pathway, whereas air
conduction thresholds (R = circle, L = X) measure function of the entire
auditory system. dB decibel; PB Score: percentage correct on speech
discrimination test using phonetically-balanced (PB) words

50% of a list of phonetically-balanced words. The speech

Fig. 39.2 Vestibular nystagmus in the horizontal plane showing slow reception threshold serves as a valuable crosscheck of pure
phase (a–c) and fast phase (d) eye movements (Figure courtesy of
tone audiometry for the frequencies of normal speech, which
David Dickman, Washington University from http://vestibular.wustl.
edu/primer5.html accessed 8/16/2008) range from 500 to 2,000 Hz (Roeser 1996). The speech dis-
crimination test reflects the percentage of words presented
above the speech reception threshold that a listener can repeat
typically pathologic. It is observed usually in acute, unilateral, correctly. An example audiogram is provided in Fig. 39.3.
uncompensated vestibular hypofunction, and suggests asym- Changes in pure-tone audiometry over 5 years in a patient
metry in the firing rates of the right and left vestibular nerves. with IMIED are presented in Table 39.1. Several features of
Asymmetric firing of the vestibular nerves leads the brain progressive sensorineural hearing loss are illustrated: the ini-
to think that the head is moving towards the side with the tial involvement of one ear only, the evolution to bilateral
higher discharge rate – away from the side of the lesion. disease, and the progression to profound bilateral sensorineu-
Compensatory eye movements are made in the opposite ral hearing loss.
direction of the presumed head movement towards the side
Myth: The ability to hear soft environmental sounds excludes
of the lesion (slow phase of nystagmus), followed by rapid
IMIED as a cause of perceived hearing dysfunction.
refixation of gaze towards the intact side once the brain real-
izes that the head is not moving (fast phase of nystagmus). Reality: Hearing loss in IMIED can take two forms. First,
patients may experience a decline in the ability to perceive
sound (hearing acuity). Second, patients may note a decreased
39.5 Auditory and Vestibular Tests ability to identify individual words (word discrimination).
A decline in either type of auditory function constitutes a
hearing deficit. Assessments of a patient’s auditory function
Pearl: The audiogram is a formal test of hearing acuity and
must consider both.
speech perception.
Formal audiological assessment uses frequency-specific
Comment: The audiogram is a graphical representation of bone and air conduction thresholds to assess hearing acuity,
the lowest volume at which sounds of specific frequencies and speech discrimination testing to assess the clarity of
can be heard by the listener. Audiometry assesses both the speech perception. An individual whose ability to hear envi-
speech reception threshold and the discrimination of speech ronmental sounds is stable but who experiences a decline in
presented at suprathreshold levels. The speech reception speech clarity has a hearing loss that can be both isolating
threshold is the lowest volume at which the patient can repeat and disabling. For example, patients with decreased word
39 Immune-Mediated Inner Ear Disease 389

Table 39.1 Five-year temporal progression of audiometric thresholds (dB) in a patient with Cogan’s syndromea
250 Hz 500 Hz 1,000 Hz 2,000 Hz 4,000 Hz 8,000 Hz
Right Left Right Left Right Left Right Left Right Left Right Left
25 5 35 5 40 5 40 5 50 5 65 5
30 55 35 55 40 50 50 55 70 60 70 65
35 105 35 105 35 105 45 110 65 110 75 95
105 105 105 105 105 105 105 105 105 105 105 105
dB decibels hearing level; Hz Hertz
a
Four audiograms obtained in 5 years

discrimination can have profoundly impaired abilities to loss or vertigo can be sudden (evolving over hours to days); of
follow conversations in settings with significant ambient intermediate duration (evolving over days to several months);
noise, such as in restaurants. or slowly progressive (evolving over several months to years).
The time course of IMIED is typically of intermediate
Pearl: Electronystagmography provides an objective mea-
duration, occurring over a period of days to several months.
sure of peripheral vestibular function.
A careful chronology of the inner ear symptoms helps
Comment: Electronystagmography, an assessment of eye exclude sudden-onset inner ear disease and its associated eti-
movements in response to a variety of stimuli, is a central ologies (e.g., viral labyrinthitis, acoustic trauma, barotrauma,
part of the assessment of peripheral vestibular dysfunction. perilymph fistula) as well as slowly progressive inner ear dis-
In electronystagmography, surface electrodes placed around ease and its associated causes (e.g., presbycusis). Once an
the orbit are used to detect and record eye movements both at intermediate time course of symptoms has been established,
rest and in response to various stimuli of the vestibular sys- the laterality of disease, the relative effects on hearing and
tem. Standard stimulatory test include caloric stimulation of balance, and serologic testing serve to distinguish further
the ears and changes in head position. between the relevant clinical entities.
Caloric testing consists of warming and then cooling of Major causes of inner ear dysfunction of intermediate
the ears. Either air or water irrigation can be applied to the onset are acoustic neuromas, spirochetal etiologies, and oto-
external auditory canals for this purpose. Caloric testing toxic medications. Acoustic neuromas typically are associ-
causes a convective movement of endolymph within the hor- ated with unilateral symptoms. The diagnosis is established
izontal semicircular canals, which are located in proximity to by the finding on MRI of an enhancing lesion at the cerebel-
the external auditory canals. The movement of fluid within lopontine angle. A diagnosis of syphilis or Lyme disease can
the horizontal canal alters the firing rate within the vestibular be made based on positive FTA-ABS (fluorescent trepone-
nerve. This elicits, in turn, an eye movement, corresponding mal antibody absorption test) or Borrelia burgdorferi serol-
to the slow phase of nystagmus (see Fig. 39.2), followed by ogy, respectively (see below for further discussion of otologic
a rapid corrective saccade that corresponds to the fast phase. manifestations of syphilis).
The slow phase eye movement normally is towards the Ototoxic medications such as gentamicin usually have a
cooled ear and away from the warmed ear. The fast phase is preferential effect on the vestibular system as opposed to
away from the cooled ear and towards the warmed ear auditory function. Symptoms of chronic dysequilibrium and
(A simple mnemonic: COWS, cold opposite, warm same). signs of bilateral vestibular hypofunction are characteristic
Decreased eye movements (<10/s) following caloric stimula- of drug-induced ototoxicity.
tion suggest vestibular hypofunction. Unilateral vestibular Pearl: Behcet’s disease, Cogan’s syndrome, relapsing poly-
hypofunction is considered to produce a 20% or greater chondritis, and sarcoidosis must all be considered in patients
reduction in eye movement velocity on the affected side with a history of ocular disease and findings consistent with
when compared with the nonaffected side. IMIED.
Comment: Ear findings and eye findings often go together.
Recognition of characteristic patterns of audiovestibular and
39.6 Differential Diagnosis ocular inflammatory disease can crack the case. Behcet’s dis-
ease (Chap. 22), Cogan’s syndrome (Chap. 31), relapsing
polychondritis (Chap. 32), and sarcoidosis (Chap. 42) must
Pearl: Several categories of disease must be excluded in the
all be considered when ocular inflammation and auditory
evaluation of a patient with possible IMIED.
dysfunction are present. The diagnosis of these and other
Comment: The differential diagnosis of IMIED can be parsed conditions can be complicated by the fact that ocular find-
primarily according to the time course of the patient’s symp- ings, audiovestibular symptoms, and disease manifestations
toms (Table 39.2) (Minor et al. 2003). The onset of hearing in other organ systems do not always occur simultaneously.
390 Y. Agrawal and H. W. Francis

Table 39.2 Differential diagnosis of immune-mediated inner ear disease (IMIED)

Time course Associated disorders Vestibular symptoms Distinguishing features
Slowly-progressive Meniere’s syndrome + Episodic vertigo, unilateral hearing loss, tinnitus,
(>3 months to aural fullness
years) −
Presbycusis Symmetrical high frequency hearing loss

Latent or tertiary syphilis ± +FTA-ABS, ±RPR

Acoustic neuroma ± Unilateral hearing loss, tinnitus; enhancing lesion

on MRI
Intermediate IMIED ±
(days – 3 months) Primary See text
Secondary (vasculitis, connective Signs and symptoms of systemic inflammatory
tissue disorder) disorders

Drugs
Aminoglycosides + Chronic dysequilibrium; signs of bilateral
Antimalarials vestibular hypofunction (e.g., oscillopsia)
Loop diuretics
NSAIDs ± Exposure risk; +B. burgdorferei serology
Lyme disease ± +FTA-ABS, ±RPR
Latent or tertiary syphilis ± Unilateral hearing loss, tinnitus; enhancing lesion on
Acoustic neuroma MRI
Sudden (hours to Acoustic trauma − Recent intense noise exposure
days) Barotrauma ± Recent deep sea diving, barotrauma
Perilymph fistula + Otolaryngology evaluation
Viral/bacterial labyrinthitis + Acute vertigo and/or hearing loss
Early or secondary syphilis + +FTA-ABS, +RPR
Acoustic neuroma ± Unilateral hearing loss, tinnitus; enhancing lesion
on MRI
FTA-ABS fluorescent treponemal antibody absorption test; IMIED immune-mediated inner ear disease; MRI magnetic resonance imaging study;
NSAIDS nonsteroidal anti-inflammatory drugs; RPR rapid plasma reagin

Myth: The loss of balance accompanied by hearing loss, tin- Comment: The inner ear complications of syphilis take many
nitus, and ear fullness is Meniere’s disease until proven forms. The spirochete employs numerous pathologic mecha-
otherwise. nisms, each of which causes a distinct pattern of injury.
Syphilitic infection is divided into three stages, primary, sec-
Reality: Vertigo accompanies hearing loss in more than 50%
ondary, and tertiary. The second and third stages both have the
of patients with IMIED (Bovo et al. 2006). In addition,
potential to affect the inner ear (Pletcher and Cheung 2003). In
25–50% of IMIED patients experience tinnitus and aural
the secondary stage of syphilis, meningeal invasion by the
fullness associated with their hearing loss (Bovo et al. 2006).
organism can lead to multiple associated cranial neuropathies,
These features are also characteristic of Meniere’s disease.
including an acute disturbance of the cochleovestibular nerve.
Thus, distinguishing between IMIED and Meniere’s disease
The sudden onset of hearing loss and vertigo may evolve over
is challenging in patients who have sensorineural hearing
hours to days. The rapid plasma reagin (RPR) assay is invari-
loss, vertigo, tinnitus, and aural fullness.
ably positive in the secondary stage of syphilis.
The principal method for distinguishing between IMIED
In latent or tertiary syphilis, the spread of gummatous
and Meniere’s disease is a consideration of the time course of
lesions and associated arteritis within the temporal bone is
symptoms. Meniere’s disease causes hearing loss over a period
responsible for the subacute to slowly progressive pattern of
of years rather than weeks to months that are characteristic of
hearing loss. The time course of hearing loss in the tertiary
IMIED. In addition, Meniere’s disease is more likely to be
stage of syphilis can also be similar to the time course of hear-
unilateral, with progression to the contralateral ear occurring
ing loss in IMIED. Obtaining the FTA-ABS test is critical to
in fewer than 50% of cases (Sajjadi and Paparella 2008).
distinguish between secondary and tertiary syphilis, because
Distinction between IMIED and Meniere’s disease is
the RPR is negative in many cases of tertiary disease.
important because IMIED is more likely to respond to the
early initiation of immunosuppressive therapy. Myth: IMIED can be excluded with relative certainty if spe-
Pearl: Both secondary and tertiary syphilis can mimic IMIED. cialized serologic assays are negative.
39 Immune-Mediated Inner Ear Disease 391

Reality: There is no pathognomonic serologic test for IMIED. progressive sensorineural hearing loss, prednisone of 1 mg/kg/
Nevertheless, numerous candidate molecules have been day is a reasonable starting dose. If the patient regains signifi-
advanced as being involved in the pathophysiology of cant auditory and vestibular function within 2 weeks, tapering
IMIED. Most data available are on antibodies that are of the prednisone dose can begin in a taper designed to discon-
directed against a 68 kilodalton (kDa) antigen. In a cohort of tinue the medication over 2 months. Methotrexate is a reason-
72 patients with idiopathic, progressive, bilateral sensorineu- able steroid-sparing agent to use if patients demonstrate a
ral hearing loss, higher levels of an antibody to a 68-kDa convincing response to glucocorticoids (Harris et al. 2003).
inner ear protein were observed in patients whose hearing Pearl: IMIED patients are good candidates for hearing aids
loss responded to glucocorticoid treatment and who had or cochlear implants.
more active disease (Moscicki et al. 1994). This inner ear
protein was believed initially to be heat-shock protein 70 Comment: All is not lost for patients with IMIED who fail or
(HSP70). Although this hypothesis appeared biologically cannot tolerate immunosuppressive therapy. Therapeutic
plausible because of the role of heat shock proteins in inflam- options beyond prednisone, cyclophosphamide, and the like
mation, the findings were not substantiated by subsequent are available. Patients who develop moderate unilateral or
studies (Bloch et al. 1995; Billings et al. 1995; Yeom et al. bilateral hearing loss can derive significant benefit from pro-
2003; Trune et al. 1998). grammable hearing aids. Those with severe to profound
More recent work has suggested that the antibody to the bilateral sensorineural hearing loss and poor word discrimi-
68 kDa antigen is the Kresge Hearing Research Institute-3 nation despite the use of hearing aids are candidates for
(KHRI-3) antibody, which binds to an antigen on the support- cochlear implantation.
ing cells of the organ of Corti, leading to hair cell depletion and Hearing aids deliver amplified sound through the ear canal.
hearing loss (Disher et al. 1997; Nair et al. 1995). Moreover, a In contrast, a surgically-placed cochlear prosthesis stimulates
protein known as the choline transporter-like-2 (CTL-2) has the auditory nerve directly, generating electrical representa-
been identified as the likely 68 kDa antigen (Nair et al. 2004). tions of speech sounds that patients learn to decode in the audi-
CTL-2 regulates the function of both choline, a constituent of tory portions of their central nervous system. In many patients,
the neurotransmitter acetylcholine, and phosphatidylcholine, a cochlear implants provide a return to the land of the hearing.
cell membrane component that is integral to hair cell survival. Myth: Benzodiazepines or meclizine are an effective long-
Knowledge of a patient’s antibody status with regard to term approach to patients with dizziness.
the 68 kDa antigen may turn out to be of clinical value. Some
Reality: The long-term use of vestibular suppressant medica-
data suggest that glucocorticoids are more likely to be effec-
tions should be limited to short-term management of acute
tive in patients who are positive for such antibodies (Zeitoun
vertigo. Agents such as meclizine impede the development
et al. 2005). Commercial assays for antibodies to the 68 kDa
of central vestibular compensation.
protein remain poorly-validated or not widely available
A more effective long-term approach to patients with sig-
(Ruckenstein 2004).
nificant peripheral vestibular dysfunction is vestibular reha-
bilitation. Peripheral vestibular dysfunction can be overcome
by compensation of the central nervous system. This process is
39.7 Treatment aided by a program of vestibular rehabilitation, whereby
trained physical therapists teach habituation exercises that pro-
Myth: The symptoms of IMIED are irreversible. mote central compensation to peripheral deficits. Physical
therapists who focus on vestibular rehabilitation can also pro-
Reality: IMIED is an important diagnosis to make because it
vide strategies that minimize fall risk. Patients are challenged
is one of the few causes of hearing loss (or vestibulopathy)
to perform the head movements and postural changes that pro-
that is potentially reversible. The natural history of IMIED
duce vertigo and then coached about compensatory strategies.
bears some similarity to that of rapidly progressive glomeru-
lonephritis, in which irreversible renal damage frequently
occurs within weeks to several months of onset. As such,
IMIED is worth at least one stab at aggressive therapy. References
Prolonged courses of immunosuppression may not be worth-
while in the absence of a clear treatment response, but few Billings PB, Keithley EM, Harris JP. Evidence linking the 68 kilodalton
patients are willing to forgo the ability to hear without a trial antigen identified in progressive sensorineural hearing loss patient
of immunosuppressive therapy. sera with heat shock protein 70. Ann Otol Rhinol Laryngol.
1995;104(3):181–8
Up to 70% of patients diagnosed with IMIED demonstrate Bloch DB, San Martin JE, Rauch SD, Moscicki RA, Bloch KJ. Serum
clinical improvement following steroid therapy (Harris et al. antibodies to heat shock protein 70 in sensorineural hearing loss.
2003; Matteson et al. 2001). In the setting of a rapidly Arch Otolaryngol Head Neck Surg. 1995;121(10):1167–71
392 Y. Agrawal and H. W. Francis

Bovo R, Aimoni C, Martini A. Immune-mediated inner ear disease. of 68 and 72 kDa that is the target of antibody-induced hearing loss.
Acta Otolaryngol. 2006;126(10):1012–21 J Neurosci. 2004;24(7):1772–9
Disher MJ, Ramakrishnan A, Nair TS, et al Human autoantibodies and mono- Nair TS, Raphael Y, Dolan DF, et al Monoclonal antibody induced
clonal antibody KHRI-3 bind to a phylogenetically conserved inner-ear- hearing loss. Hear Res. 1995;83(1–2):101–13
supporting cell antigen. Ann N Y Acad Sci. 1997;830: 253–65 Pletcher SD, Cheung SW. Syphilis and otolaryngology. Otolaryngol
Harris JP, Weisman MH, Derebery JM, et al Treatment of corticoster- Clin N Am. 2003;36(4):595–605, vi
oid-responsive autoimmune inner ear disease with methotrexate: a Roeser R. Audiology desk reference: a guide to the practice of audiol-
randomized controlled trial. JAMA. 2003;290(14):1875–83 ogy. 1 ed. New York: Thieme; 1996
Kikuchi M. On the “sympathetic otitis”. Zibi Rinsyo Kyoto. 1959; 52:600 Ruckenstein MJ. Autoimmune inner ear disease. Curr Opin Otolaryngol
Kornblut AD, Wolff SM, deFries HO, Fauci AS. Wegener’s granuloma- Head Neck Surg. 2004;12(5):426–30
tosis. Otolaryngol Clin N Am. 1982a;15(3):673–83 Sajjadi H, Paparella MM. Meniere’s disease. Lancet 2008;372
Kornblut AD, Wolff SM, Fauci AS. Ear disease in patients with (9636):406–14
Wegener’s granulomatosis. Laryngoscope 1982b;92(7 Pt 1): 713–17 Schindler JS, Niparko JK. Imaging quiz case 1. Transverse temporal
Matteson EL, Fabry DA, Facer GW, et al Open trial of methotrexate as bone fractures (left) with subsequent progressive SNHL, consistent
treatment for autoimmune hearing loss. Arthritis Rheum. 2001;45(2): with sympathetic cochleolabyrinthitis. Arch Otolaryngol Head
146–50 Neck Surg. 1998;124(7):814, 816–8
Mccabe BF Autoimmune sensorineural hearing loss. Ann otol Rhinol Schubert MC, Minor LB. Vestibulo-ocular physiology underlying ves-
Laryngol. Sep-Oct 1979;88 (5pt1):585–589 tibular hypofunction. Phys Ther. 2004;84(4):373–85
Minor L, Carey J, Lustig LR. Disorders of balance. In: Lustig LR, Stone JH, Francis HW. Immune-mediated inner ear disease. Curr Opin
Niparko JK, editors. Clinical neurotology: diagnosing and manag- Rheumatol. 2000;12(1):32–40
ing disorders of hearing, balance and the facial nerve. New York: ten Cate WJ, Bachor E. Autoimmune-mediated sympathetic hearing
Martin Dunitz; 2003. p. 225–44 loss: a case report. Otol Neurotol. 2005;26(2):161–5
Minor L, Zee DS. Clinical evaluation of the patient with dizziness. In: Trune DR, Kempton JB, Mitchell CR, Hefeneider SH. Failure of ele-
Lustig LR, Niparko JK, editors. Clinical neurotology: diagnosing vated heat shock protein 70 antibodies to alter cochlear function in
and managing disorders of hearing, balance and the facial nerve. mice. Hear Res. 1998;116(1–2):65–70
New York: Martin Dunitz; 2003. p. 81–110 Yeom K, Gray J, Nair TS, et al Antibodies to HSP-70 in normal donors
Moscicki RA, San Martin JE, Quintero CH, Rauch SD, Nadol JB Jr, and autoimmune hearing loss patients. Laryngoscope. 2003;113(10):
Bloch KJ. Serum antibody to inner ear proteins in patients with pro- 1770–6
gressive hearing loss. Correlation with disease activity and response Zeitoun H, Beckman JG, Arts HA, et al Corticosteroid response and
to corticosteroid treatment. JAMA. 1994;272(8):611–16 supporting cell antibody in autoimmune hearing loss. Arch
Nair TS, Kozma KE, Hoefling NL, et al Identification and characteriza- Otolaryngol Head Neck Surg. 2005;131(8):665–72
tion of choline transporter-like protein 2, an inner ear glycoprotein
 Osteoporosis
40
Kenneth G. Saag, Sarah L. Morgan, and Amy H. Warriner

fractures in a fashion that is dose and time dependent. As

40.1 Overview of Osteoporosis many as half of all persons taking glucocorticoids on a long-
term basis will experience fracture in their lifetime.
› Bone continually undergoes a process of renewal Glucocorticoids affect the bone independently of their effects
called remodeling. In the normal adult skeleton, bone on BMD due to their direct toxicity to osteoblasts and osteo-
formation and bone resorption are closely coupled: cytes (Manolagas, Weinstein 1999; van Staa et al. 2003). The
new bone laid down by osteoblasts exactly matches large General Practice Research Database of the United
osteoclastic bone resorption. Kingdom has shown that even at doses of glucocorticoid
› The bone mass of an individual in later life is a conse- below the physiologic replacement dose (estimated at 7.5 mg/
quence of the peak bone mass accrued in utero and day prednisolone), glucocorticoid use is associated with
during childhood and puberty, as well as the subse- greater fracture rates at both the hip and spine (van Staa et al.
quent rate of bone loss. 2000). In normal volunteers, oral glucocorticoids at physio-
› Genetic factors strongly contribute in determining the logic doses result in a prompt reduction in osteocalcin as a
peak bone mass. However, hormonal, nutritional, and measure of bone formation (Ton et al. 2005). Similarly, tri-
environmental influences modulate the genetically amcinolone administered intra-articularly also was associ-
determined pattern of skeletal growth. ated with a reduction in bone formation (Emkey et al. 1996).
› More than 1.5 million fractures related to osteoporosis Finally, inhaled glucocorticoids may also adversely impact
occur in the United States every year. bone density and lead to an increased rate of fractures (Israel
› Ninety percent of all hip and spine fractures are related et al. 2001; Wong et al. 2000).
to osteoporosis. Pearl: African Americans get osteoporosis, too.
› Most osteoporotic fractures involve the femoral neck,
the vertebral bodies, or the wrist. Comment: Although young, healthy African Americans have
› Two types of scores have been used traditionally to a 10% greater BMD on an average when compared with
quantify bone mineral density (BMD): Caucasians, osteoporosis and fracture risk are under-appreciated
- The T score is the number of standard deviations of among African. Among older African American women,
the patient’s BMD above or below the young-normal osteoporosis is a common problem that is less frequently
mean BMD. diagnosed and treated than in Caucasian women of similar
- The Z score is the number of standard deviations of ages (Looker et al. 1997; Mikuls et al. 2005; Miller et al.
the measurement above or below the age-matched 2005; Mudano et al. 2003; Siris et al. 2001). African
mean BMD. American women who have already experienced a fracture
› The World Health Organization (WHO) defines osteo- are 50% less likely to receive subsequent testing or treatment
porosis with a T score ≤ −2.5. than Caucasian women who have fractured (Mikuls et al.
2005; Miller et al. 2005; Mudano et al. 2003). Furthermore,
when African American men and women fracture, they have
worse outcomes including longer hospital stays, increased
40.2 Osteoporosis Epidemiology disability, and greater mortality (Furstenberg, Mezey 1987;
Jacobsen et al. 1992; Kellie, Brody 1990). Black women
older than 65 years of age have a mortality rate from hip
Myth: There is a safe dose of glucocorticoids for bone.
fractures that is 1.5 times higher than that of white women.
Reality: Glucocorticoids are the most common cause of Antiosteoporotic therapies appear to have equal efficacy
drug-induced osteoporosis and lead to an increased risk of across races.

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 393

DOI:10.1007/978-1-84800-934-9_40, © Springer Science + Business Media B.V. 2009
394 K. G. Saag et al.

Pearl: Men also get osteoporosis. scarring (“cigarette papering”), blue-tinged sclera, mitral
valve prolapse/other cardiac anomalies, short stature, and
Comment: Approximately 30% of all fractures in the aging dental problems are among a number of the historical or
population occur in men. These fractures account for up to physical examination clues that may heighten suspicion for
25% of the total osteoporosis costs (Burge et al. 2007; Jones such a condition (See Chap. 47, “Heritable Disorders of
et al. 1994; van Staa et al. 2001), and the risk of recurrent Connective Tissue”).
fracture is nearly threefold higher in men than women (Center Forme frustes or incomplete penetrance of hereditary
et al. 2007; Sanders et al. 1999). Fracture-related death rates connective tissue disorders may escape clinical attention for
have been increasing in both men and women, but are noted many years and may be concurrently missed by the existing
to be consistently higher in men. The age-related rise in frac- genetic testing evaluating for mutations in COL1A and other
tures in men tends to occur approximately 5–10 years later, related genes affecting collagen productions. In children and
than the increase seen in women (Kudlacek et al. 2000). The adults with osteogenesis imperfecta, bisphosphonate medi-
use of BMD for fracture-risk prediction also differs between cations have proven effectiveness in lowering fracture rates
men and women: men develop vertebral fractures at a higher (Rauch, Glorieux 2004).
BMD than women (Kudlacek et al. 2000; Orwoll 2000). The
difference in body size between men and women partially Pearl: In older adults, any fracture, regardless of the degree
explains the difference in BMD, but does not fully explain of trauma, may represent osteoporosis.
the lower fracture risk that men have when compared to Comment: Low trauma fractures have historically been
women (Looker et al. 2001; Melton et al. 2005). linked to osteoporosis due to the subsequent increased risk of
Physicians’ perception that osteoporosis does not affect a recurrent fracture. In one study, investigators found an
men adds a further barrier to successful diagnostic and thera- increased risk of subsequent fracture following a low-trauma
peutic intervention. Similar to women, current guidelines set fracture at all fracture sites with only two exceptions: rib
by the International Society for Clinical Densitometry fractures in men and ankle fractures in women (Center et al.
(ISCD) recommend using a T score of 2.5 standard devia- 2007).
tions below the normal male reference levels to diagnose Low trauma falls are the most common cause of fractures
osteoporosis in men over the age of 50 years, unless other in persons over the age of 65 years, and their predominance
secondary causes of low BMD exist (ISCD 2007). However, increases with rising age to nearly 90% in persons over the
uncertainty exists on when to start screening for osteoporosis age of 90 years (Bergstrom et al. 2008). The site of fracture
in men. ISCD recommends screening all men over the age of of most of these low trauma falls is the hip or pelvis
70, based only on expert opinion (ISCD 2007). In contrast, (Bergstrom et al. 2008). In an earlier study on pelvic frac-
one analysis suggests that it is only cost-effective to screen tures, the majority of fractures that occurred as the result of
and begin treatment in men over the age of 65 with a diagno- moderate trauma (defined as trauma not usually expected to
sis of osteoporosis based on BMD and a history of a fracture, result in fracture) were in older persons with a mean age of
or men over the age of 80 with osteoporosis based on BMD 69 (Melton et al. 1981).
and no history of a fracture (Schousboe et al. 2007). Further Several studies have shown that the association with
investigation into the proper timing for osteoporosis screen- osteoporosis is similar for high and low trauma fractures.
ing and treatment for men is currently ongoing. Regardless of trauma type, individuals with fractures are
Until evidence-based recommendations become avail- three times more likely to have osteoporosis at one or more
able, assessment of clinical risk factors (e.g., prior fracture, sites (Karlsson et al. 1993; Melton et al. 1993; Sanders et al.
use of androgen deprivation therapy for prostate cancer) is 1998). A case-control study found that Australian women
important in order to identify men most at risk of a fracture. older than 50 years of age with high trauma fractures had
In terms of treatment options, both bisphosphonates and teri- lower BMD at hip, spine, forearm, and total body sites than
paratide have been shown to reduce fracture risk in men. population-based samples of older women without fracture
(Sanders et al. 1998).
Pearl: In young adults with new fractures, hereditary disor-
A Swedish study that evaluated hip fractures in men
ders of collagen should be considered.
reported that high energy fractures were associated with age-
Comment: Osteogenesis imperfecta and the Ehlers–Danlos related changes in bone strength (Hedlund, Lindgren 1987). A
syndrome are two of the more common hereditary disorders retrospective cohort study of 1288 individuals of age 35 and
of collagen synthesis that can lead to fractures in young older with distal forearm fractures found no difference in the
adults. Severe variants of these conditions are detected by risk of subsequent fracture when stratified based on severity
fractures at birth or in childhood, but persons with less severe of trauma (i.e., severe vs. low trauma) (Cuddihy et al. 1999).
phenotypes may present with their first of successive frac- A prospective study of two large osteoporosis cohorts of
tures in later years. Excessive joint mobility, abnormal wound patients 65 years or older, the Study of Osteoporotic Fractures
40 Osteoporosis 395

(SOF) and Osteoporotic Fractures in Men Study (MrOS), the hip or 20% or greater of any major osteoporosis-related
found that each standard deviation reduction in total hip BMD fracture be treated with an antiosteoporotic therapy.
was similarly associated with an increased risk of high trauma
Myth: DXA scans from different manufacturers and facilities
fracture and low trauma fracture. The risk of future fracture
give comparable results.
was 34% higher in women with a history of high trauma frac-
ture and 31% higher in women with an initial low trauma frac- Reality: There are numerous manufacturers of machines to
ture, when compared with women having no history of measure bone density by DXA. DXA systems and the cor-
fracture. Risk of subsequent fracture was not examined for responding data generated from DXA machines by different
men (Mackey et al. 2007). These findings suggest that any manufacturers are not interchangeable. Some of the reasons
prior fracture, regardless of the degree of trauma, may indi- that they are not interchangeable include different methods
cate underlying skeletal fragility and an increased risk of for dual energy X-ray production (voltage switching vs.
future fracture. K-edge filter), different X-ray detectors, different X-ray
acquisition methods (pencil vs. fan beams), different calibra-
tion methods (continuous vs. intermittent calibration), differ-
ent edge detection software, different regions of interest, and
40.3 Osteoporosis Diagnosis: different young normal databases.
Bone Mass Measurement The ISCD stated that “it is not possible to quantitatively
compare BMD or to calculate a least significant change
between facilities without a cross-calibration study” (ISCD
Pearl: Both the absolute and relative risks of fractures are
2007). Therefore, in order to make quantitative comparisons
important in osteoporosis prevention.
between successive DXA tests, it is important to scan patients
Comment: BMD measurement provides a valid and reliable on the same DXA system at each study.
way to define fracture risk. The “T-score,” which represents
Myth: Low bone density in younger women means they have
the number of standard deviations above or below peak bone
osteoporosis.
mass of sex-matched comparisons, has become a standard to
characterize the relative risk of fracture and to define the Reality: The fracture risk in women before menopause dif-
presence or absence of osteoporosis in the absence of frac- fers from that of postmenopausal women. The criteria set by
tures. More recently, attention has shifted from the relative the WHO to diagnose postmenopausal osteoporosis – namely,
risk to absolute risk of fracture, using the WHO FRAX cal- a T score of <−2.5 – cannot be applied to premenopausal
culator (FRAX 2008; Kanis et al. 2001). FRAX is based on women. The occurrence of low bone mass in women between
risk-factor modeling of data from numerous international the ages of 30 and 40 is unusual: only one such woman at the
datasets and provides a 10-year absolute fracture risk of the age of 40 had a T score ≤−2.5 (Khan 2006). In most cases,
hip and spine. low BMD in premenopausal women represents low peak
Absolute fracture risk is a valuable construct that incorpo- bone mass, likely due to genetic factors but with possible
rates BMD, but is also derived from other significant risk environmental contributions as well. Despite their low peak
factors including age, sex, body mass index, previous frac- bone mass, the 5- and 10-year fracture risk is very low in
ture history, parental history of hip fracture, tobacco smok- these women (Lewiecki 2005).
ing, glucocorticoid use, history of rheumatoid arthritis, and Estrogen deficiency and glucocorticoid use are the most
other secondary osteoporosis risk factors. Important second- common identifiable primary and secondary causes of low
ary risk factors include type 1 diabetes mellitus, osteogenesis bone mass in younger women (Moreira Kulak et al. 2000).
imperfecta, hyperthyroidism, hypogonadism, premature Estrogen deficiency is generally either due to premature
menopause, chronic malnutrition, malabsorption, and chronic ovarian failure (e.g., because of chemotherapy) or anorexia
liver disease. The FRAX model can calculate fracture risk in nervosa. If menstrual irregularities are present, the estro-
the absence of data related to BMD. gen status of the patient should be determined. A serum
Following development of the FRAX model, the National follicle stimulating hormone (FSH) level greater than
Osteoporosis Foundation issued new guidelines for osteopo- 20 mIU/L is related to increased bone loss (Khan, Syed
rosis treatment that now incorporate FRAX along with 2004). Among premenopausal women with low bone mass,
revised T-score criteria (NOE 2008). These guidelines were there is also a significant correlation with a family history
developed based on cost effectiveness considerations. It is of osteoporosis (Moreira Kulak et al. 2000), supporting a
recommended that a person with low bone mass by dual- genetic predisposition.
energy X-ray absorptiometry (DXA) (formerly called In addition to consideration of hypogonadism, initial
osteopenia: a T score between −1.0 and −2.5) and a FRAX evaluation of low bone mass should include investigation for
10-year absolute probability of fracture of 3% or greater at the underlying causes of bone loss, such as hypovitaminosis
396 K. G. Saag et al.

D, medications, thyroid disease, parathyroid disease, dietary (Gankam Kengne et al. 2008). Monitoring selective sero-
causes, and renal or hepatic disease. If fragility fractures tonin reuptake inhibitors and diuretic use is important in pre-
have occurred and no secondary causes of bone loss are venting hyponatremia.
noted, a bone biopsy should be considered (Khan 2006).
Myth: If you are elderly or taking a medicine that decreases
Even if an etiology is identified, the appropriate treatment
gastric acid, you must take calcium citrate as a calcium
of premenopausal women found to have low bone mass is
supplement.
unclear. Pharmacologic therapies have not been studied in
sufficient detail in this population, and such studies that exist Reality: Changes in the gastrointestinal tract have been doc-
have occurred primarily on patients treated with glucocorti- umented with aging (Bhutto, Morley 2008). It is often sug-
coids (Adachi et al. 2001; Reid et al. 2000; Saag et al. 1998; gested that the majority of the elderly have gastric
Wallach et al. 2000). In a case study of bisphosphonate use achlorhydria. However, in one study, approximately 90% of
during pregnancy, no congenital anomalies were noted in the elderly subjects had the ability to acidify stomach contents,
fetuses exposed (Ornoy et al. 2006). However, there contin- even in the basal (unstimulated) state (Hurwitz et al. 1997).
ues to be a concern regarding long-term side effects of cer- Among persons with achlorhydria, calcium absorption was
tain osteoporosis therapies among younger women, especially superior with calcium citrate when fasting. However, when
bisphosphonates, due to the long half-life of these medica- calcium was administered with a normal breakfast, calcium
tions. Animal models have shown that bisphosphonates do carbonate absorption was normalized (Recker 1985). The
cross the placenta and skeletal effects have been seen (Patlas effect of proton pump inhibitors on calcium absorption has
et al. 1999). The Federal Drug Administration (FDA) contin- been documented to be variable with some studies, suggest-
ues to consider bisphosphonates as a class D medication dur- ing that omeprazole may decrease absorption (O’Connell
ing pregnancy because of these concerns (Novartis 2004). et al. 2005; Serfaty-Lacrosniere et al. 1995).
Information regarding the use of other osteoporosis treat- Studies of calcium carbonate vs. calcium citrate absorb-
ments in premenopausal women is lacking. At the least, ability from commercial supplements have yielded conflict-
younger women with low bone mass should be encouraged to ing results depending upon the methodology used to assess
receive adequate calcium, vitamin D, and exercise. Risk fac- calcium absorbability and whether the calcium salts were
tors, such as tobacco use and moderate to heavy ethanol use taken fasting or with a test meal (Heaney et al. 1999; Heller
(i.e., more than one serving of alcohol per day) should be mini- et al. 2000; Heller et al. 1999). In a randomized, crossover
mized. Such patients should be followed-up around the time of trial that compared commercial calcium carbonate and cal-
menopause by physicians knowledgeable about osteoporosis. cium citrate, there were no significant differences between
preparations with regard to urinary calcium or intact PTH
levels. A cost-benefit analysis was also completed between
the preparations. The authors concluded that calcium car-
40.4 Osteoporosis Prevention: bonate was a good choice based upon bioavailability, cost,
Calcium and Sodium Effects and clinical efficacy when given with a meal. The relative
bioavailability of calcium from orange juice fortified with
calcium citrate malate, skim milk, and a calcium carbonate
Pearl: Sodium plays a role in bone health.
supplement as a portion of a test meal appeared to be equiva-
Comment: A positive relationship exists between dietary lent in a sample of elderly patients (Martini, Wood 2002).
sodium intake and calciuria (Heaney 2006a, b). On an aver- Many other calcium salts besides calcium carbonate and
age, urinary calcium increases by approximately 1 mmol calcium citrate for calcium supplementation are available
(40 mg) for each 100 mmol (2300 mg) of sodium ingested. (Hanzlik et al. 2005). In a four-way crossover study of
Increased calcium absorptive efficiency is also associated healthy women treated with 1200 mg of calcium as calcium
with calciuria (Breslau et al. 1982; Meyer et al. 1976). One carbonate, calcium citrate, calcium formate, or placebo,
of the mechanisms of the relationship between sodium intake serum calcium rose by 15% with calcium formate and the
and calciuria is the increase in parathyroid hormone (PTH) intact PTH level fell by 70% within 60 min of administration.
secretion that occurs in response to a sodium load (50). The The authors concluded that calcium formate may be a useful
amount of calcium consumed in the diet probably have an calcium supplement (Hanzlik et al. 2005). When bone mark-
effect on whether calciuria caused by a sodium load affects ers are followed as an outcome measure, calcium citrate may
calcium balance. Adjustment of a person’s daily sodium decrease markers of bone resorption more than calcium car-
intake may be beneficial for bone health as well as for hyper- bonate in postmenopausal women (Kenny et al. 2004).
tension (Teucher et al. 2008). Therefore, in patients with documented achlorydria, if
Mild hyponatremia, a common electrolyte abnormality in calcium is to be taken while fasting, calcium citrate is an
the elderly, has been associated with a higher fracture rate appropriate choice. However, as not all elderly individuals
40 Osteoporosis 397

have achlorhydria, calcium citrate is not necessary in all such A randomized trial in men evaluated a “normal” amount of
patients. Calcium carbonate, which is inexpensive and read- calcium, reduced animal protein (52 g/day), and 50 mmol of
ily available, appears to be absorbed as along with calcium salt per day vs. a low-calcium diet in individuals with hyper-
citrate when taken with a meal. Future investigations should calciuria and recurrent calcium oxalate stones (Borghi et al.
center on other calcium salts for supplementation. 2002). At 5 years, urinary oxalate excretion increased in the
low-calcium diet group and decreased on the normal cal-
Myth: Calcium supplementation as monotherapy is a suffi-
cium, low protein, and low sodium diet. The authors con-
cient therapy following a fracture.
cluded that a diet that was restricted in salt and protein and
Reality: Randomized controlled trials comparing calcium normal in calcium content was more protective than a low
therapy alone to placebo have shown no benefit in reduction calcium diet.
of vertebral and nonvertebral fractures (Grant et al. 2005; It has been hypothesized that calcium supplementation
Lyons et al. 2007; Prince et al. 2006; Reid et al. 2006). binds oxalate in the gut, leading to a relative reduction in the
However, these findings may be partially confounded by proportion of oxalate absorbed in the gut and lowering cal-
nonadherence. Patients with adherence of 80% or better did cium oxalate stone formation (Williams et al. 2001). Many
have a reduced risk of fracture in one study (hazard ratio 0.7; other nutritional factors, besides calcium, likely contribute to
95% CI, 0.4–1.0) (Prince et al. 2006). the pathogenesis of calcium-containing stones (Curhan et al.
Most studies on calcium have evaluated the combination 2004). A prospective cohort study of 45,619 men without a
of calcium and vitamin D. Although there is evidence that history of kidney stones had self-administered food fre-
calcium and vitamin D reduce the risk of fracture, the benefit quency questionnaires every 4 years (Taylor et al. 2004).
is not observed in all populations (Chapuy et al. 1992; After adjusting for age, a higher intake of dietary calcium
Dawson-Hughes et al. 1997; Larsen et al. 2004; Bischoff- was associated with a reduced risk of nephrolithiasis. For
Ferrari et al. 2005; Heikinheimo et al. 1992; Jackson et al. men older than 60 years of age, no association between
2006; Lips et al. 1996; Papadimitropoulos et al. 2002). Both dietary calcium and stone formation was observed.
the type and the amount of vitamin D supplementation appear Magnesium intake decreased and vitamin C intake increased
relevant to the reduction of fracture risk (Bischoff-Ferrari the risk of symptomatic nephrolithiasis. Dietary phytate
et al. 2005). Owing to these multiple confounding issues sur- intake was associated with a higher risk of kidney stone for-
rounding the absolute benefit of calcium, it is recommended mation. Higher fluid intake was associated with a lower risk
that all persons take calcium and vitamin D supplements for of kidney stones.
general bone health. Calcium alone is probably not sufficient The precise recommendations for the prevention of kid-
to reduce the risk of fracture. ney stones depend upon patient’s characteristics as well as
the composition of the kidney stone (Taylor and Curhan
Myth: Taking calcium supplements leads to calcium-con-
2006). Therefore, the dictum that all patients who have had
taining renal stones.
renal stones should not consume dietary or supplemental cal-
Reality: Nephroureterolithiasis is a common problem with a cium is usually incorrect.
rising incidence. Kidney stones arise from a variety of medi-
Myth: Taking calcium supplements for osteoporosis increases
cal conditions and approximately 80% of stones are esti-
the risk of cardiovascular disease.
mated to contain calcium (Taylor and Curhan 2004). In the
Women’s Health Initiative (WHI) of 36,282 postmenopausal Reality: In a secondary analysis of vascular events from a
women who received 1,000 mg of calcium as calcium car- randomized double-blind, placebo-controlled trial of cal-
bonate and 400 IU of vitamin D3 (cholecalciferol) or pla- cium supplementation (n = 1,471), myocardial infarction as
cebo each day, the hazard ratio for kidney stones was 1.17 well as a composite of myocardial infarction, stroke, and
with 95% percent confidence interval of 1.02–1.34 (Jackson sudden death were more common in the calcium supplement
et al. 2006). This report raised concern related to nephro- vs. the placebo group. A search of a New Zealand data base
lithiasis and calcium consumption. However, the preponder- for hospital admissions attenuated the calcium effect to bor-
ance of the evidence does not support a relationship of higher derline significance. The authors concluded that “this poten-
calcium intake to a greater risk of calcium-containing tially detrimental effect should be balanced against the likely
stones. benefits of calcium on bone” (Bolland et al. 2008).
The relationship of dietary risk factors and kidney stones Evaluation of data from the WHI of 36,282 postmenopausal
was evaluated prospectively in 96,245 females in the Nurses’ women who were randomized to 500 mg calcium and 200 IU
Health Study II (Curhan et al. 2004). A higher dietary intake of vitamin D twice daily found that women in the calcium and
of calcium decreased the risk of symptomatic stones, and vitamin D arm were not at increased risk for transient ischemic
calcium supplement intake was associated with a “slight and attack, myocardial infarction, stroke, coronary revasculariza-
nonsignificant increase in risk” (Curhan et al. 2004). tion, angina requiring hospitalization, or transient ischemic
398 K. G. Saag et al.

attacks (Hsia et al. 2007). In addition, in a cohort of 1,179 deficiency is present in up to 90% of the elderly population
women followed for 4 years with a calcium supplement dos- (Mezquita-Raya et al. 2001), even in southern latitudes (Lips
age of 1400–1500 mg/day (Lappe and Heaney 2008), the vas- 2001; Visser et al. 2006). Even younger persons have low
cular event rate was less than half of the rate seen in an earlier vitamin D levels with 1/3 of premenopausal women with low
large randomized controlled trial (Bolland et al. 2008). levels by the end of winter (Tangpricha et al. 2002). Current
Thus, the data at this point do not seem to support a detri- recommendations aim to maintain 25(OH) vitamin D at or
mental effect of calcium supplementation on vascular events above 32 ng/mL (80 nmol/dL), the level at which PTH levels
in postmenopausal women. It has been suggested that dete- plateau.
riorating renal function, with vascular calcification related to Obtaining an accurate assessment of vitamin D levels in
chronic renal failure, may have contributed to the findings in individuals is also an area of concern. Owing to its extreme
the study in New Zealand (Lappe and Heaney 2008). Future hydrophobic characteristics and the presence of two forms of
research is planned to continue to evaluate the hypothesized 25(OH) vitamin D {25(OH) D2 and 25(OH) D3}, the preci-
link between calcium supplementation and vascular disease. sion and accuracy of measuring vitamin D levels has been a
historical challenge (Hollis 2008). 25(OH) D3 (cholecalcif-
Myth: All calcium-fortified products are good sources of
erol) is formed mainly through ultraviolet radiation of the
calcium.
skin, with only a small amount coming from dietary sources.
Reality: Studies evaluating calcium from calcium-fortified 25(OH) D2, ergocalciferol, is produced by the irradiation of
orange juice showed that 48% greater absorption was present ergosterol. Some assays provide separate analyses of D2 and
with calcium citrate malate fortified juice than tricalcium D3, but total 25(OH) vitamin D is the level analyzed in most
phosphate/calcium lactate (Heaney et al. 2005). Calcium for- population based studies (Zerwekh 2008).
tified-soy milk was shown to be absorbed at 75% of the effi- Currently, great variation can be seen when evaluating
ciency of cow’s milk (Heaney et al. 2000). In addition, it has 25(OH) vitamin D levels by the assorted analysis methods
been shown that in some soy beverages, the added calcium is (Binkley et al. 2004). Concurrent analyses revealed that 15%
present as sediment on the bottom of the container (Heaney of tested patients were vitamin D insufficient when tested in
et al. 2000; Heaney and Rafferty 2006). In the evaluation of one laboratory, whereas 90% were insufficient based on
eight national brands, the unshaken samples averaged a cal- another lab’s analysis (Binkley et al. 2004). This variation
cium concentration that was 31% of the labeled calcium con- probably stems from differing methods for evaluating
tent. Shaking the beverages increased the measurable calcium 25(OH) vitamin D status as vitamin D in plasma and serum
to 59% of the labeled content (Heaney and Rafferty 2006). appears to be very stable (Lissner et al. 1981).
Therefore, soy beverages are not equivalent to milk in cal- Earlier methods used for evaluating 25(OH) vitamin D
cium content. High-calcium mineral waters have absorption levels used a vitamin D binding protein, which not only mea-
that is equal to calcium from milk or slightly higher absorp- sured 25(OH)D2 and 25(OH)D3, but also other metabolites,
tion (Heaney 2006a) and can serve as an additional supple- and the process was cumbersome for technicians (Hollis
ment in patients needing calcium supplementation. 2008; Haddad, Chyu 1971). In the 1980s, a radioimmunoas-
In summary, not all calcium-fortified products are good say (RIA) technique was developed to simultaneously iden-
sources of calcium and there is variability in calcium bio- tify 25(OH) D2 and 25(OH) D3, revealing a total 25(OH)
availability. Depending upon the calcium salt selected and vitamin D level. More recently, direct methods using high
interaction of the supplement with the product, calcium performance liquid chromatography (HPLC) and liquid
absorption can be greatly affected (Rafferty et al. 2007). chromatography–mass spectrometry (LC–MS) have largely
replaced older methods. HPLC and LC–MS methods evalu-
ate 25(OH)D2 and 25(OH)D3 levels separately (Eisman
et al. 1977; Maunsell et al. 2005) and require dedicated tech-
40.5 Osteoporosis Prevention: Vitamin D nicians and internal standards for the measured compounds
(Hollis 2008). HPLC is considered the gold standard by
many experts in the field (Hollis 2008). Currently, an updated
Pearl: Vitamin D insufficiency is common, but vitamin D
chemiluminescence RIA method (Diasorin Coorporation) is
level is difficult to measure accurately.
also available for use and measures the total 25(OH) vitamin
Comment: Vitamin D is important in bone health due to its D levels (Hollis 2008). LCMS, HPLC, and the more recent
ability to counter-regulate PTH, a promoter of bone loss, and RIA methods appear to have improved precision in deter-
its stimulation of intestinal and renal calcium absorption. In mining total 25(OH) vitamin D levels, when compared to
the presence of inadequate calcium intake or vitamin D defi- prior methods, and attempts are being made to ensure contin-
ciency, PTH levels typically rise, resulting in secondary ued quality assessment of vitamin D measurements. These
hyperparathyroidism. Despite this knowledge, vitamin D variations in vitamin D assays need to be accounted for when
40 Osteoporosis 399

evaluating patients in practice and when comparing results The best gauge of vitamin D status is a 25(OH) vitamin D
from clinical trials. level (Holick 2008). Owing to many factors detailed earlier,
which can affect the production of vitamin D in the skin,
Myth: Routine exposure to sunshine exposure leads to an
sunlight exposure cannot always be counted upon to produce
adequate 25(OH) vitamin D level.
a 25(OH) vitamin D level above 30 ng/mL. However, a com-
Reality: Vitamin D forms in the skin after exposure to solar bination of supplements and dietary sources and sun expo-
ultraviolet B (UVB) radiation (Holick 2006a). This UVB sure can be useful to assure adequate vitamin D status.
radiation converts 7-dehydrocholesterol to previtamin D3
and ultimately vitamin D3. Excessive exposure to sunlight
does not cause vitamin D intoxication because excess vita-
min D3 and previtamin D3 are also destroyed by sunlight. 40.6 Osteoporosis Treatment:
Many factors decrease the ability to form vitamin D in the General Considerations
skin, including increasing skin melanin, sunscreens, clothing
that prevents the sun from reaching the skin, and a reduction
Myth: All prescription osteoporosis medications are equal.
in 7-dehydrocholesterol in the skin with aging (Holick 2006a;
Holick 2006b; Holick 2007). At some latitudes, in the winter Reality: There is a growing number of prescription therapies
months, the sunlight is not sufficient to produce vitamin D available that are approved for the treatment of postmeno-
(Holick 2006a). Exposure to 5–10 min of sunlight on the legs pausal osteoporosis and other forms of bone loss. Each ther-
between 10 a.m and 3 p.m in the spring, summer, and fall has apy has been proven to be effective in reducing fractures in
been shown to be adequate to prevent vitamin D insufficiency some manner. Each treatment option is also associated with
(Holick 2006a). unique risks and side effects. Comparisons of efficacy
With experimental exposure to UVB light, individuals with between osteoporosis medications are generally based on
a lighter skin tone had greater increases in their 25(OH) vita- nonhead-to-head evaluation of individual clinical trials and
min D level than darker skinned individuals (Ton et al. 2005; the few trials comparing intermediate outcomes. There are
Armas et al. 2007). Another study in Honolulu, Hawaii, evalu- currently no studies available that adequately provide a
ated 25(OH) vitamin D levels in 93 individuals who were direct head-to-head comparison of fracture risk reduction
under the sun for an average of 28.9 h/week without sunscreen (Table 40.1).
(Binkley et al. 2007). There was a substantial variability of As a class, bisphosphonates have been found to be effec-
25(OH) vitamin D levels in the population, with no level tive in reducing the risk of fractures at the spine, hip, and
exceeding 60 ng/mL. Fifty-one percent of the subjects had a nonvertebral sites with the exception of ibandronate, which
25(OH) vitamin D level <30 ng/mL. The authors postulated is lacking evidence of nonvertebral risk reduction (Black
that the individuals with low vitamin D status had inadequate et al. 1996; Chesnut et al. 2004; Delmas et al. 2006; Recker
cutaneous production of vitamin D3, enhanced destruction of et al. 2004; Stakkestad et al. 2003; Wells et al. 2008a; Wells
previtamin D3 or vitamin D3 from sunlight, down-regulation et al. 2008b). However, newer secondary analysis data on
of vitamin D production by sun-induced melanin production, prior ibandronate studies suggest that the currently pre-
or abnormalities of transport of vitamin D from the skin to the scribed doses (in comparison to lower doses used in some
circulation (Binkley et al. 2007). There have been reports of clinical trials) do result in a nonvertebral fracture-risk reduc-
treatment of patients with malabsorption with ultraviolet light tion (Cranney et al. 2009). The selective estrogen receptor
to improve vitamin D status (Chandra et al. 2007). modulator (SERMs), raloxifene, and calcitonin, reduce the

Table 40.1 Relative risk of a future fracture with use of osteoporosis treatments in a person with a prior fracture; RR (95%CI) (111, 115, 116, 121)
Drug class Vertebral fracture RRa Nonvertebral fracture (e.g., hip) RRa
Bisphosphonates (BPs) 0.39–0.75 0.20–0.41
Alendronate (116) 0.55 (0.43–0.69) 0.77 (0.64–0.92)
Ibandronate 0.62 (0.41–0.75) Not significant
Risedronate (115) 0.61 (0.50–0.76) 0.80 (0.72–0.90)
Zoledronic acid 0.54 (0.32–0.92) 0.73 (0.55–0.98)
Parathyroid hormone (PTH) 0.65 0.53
Teriparatide
Selective estrogen-receptor modulators (SERMs) 0.30–0.55% Not significant
Raloxifene
Calcitonin 0.23 Not significant
a
Unless noted, data were obtained from product labeling and are not from head to head studies
400 K. G. Saag et al.

risk of vertebral fractures, but also have no significant risk 0.7; 95% CI 0.7–1.0; other fracture HR, 0.8; 95% CI 0.7–
reduction for nonvertebral fractures (Miacalcin® Insert 2003; 0.8), as was colorectal cancer. The risk of CHD in the treat-
Evista® Insert 2007). PTH is the only available anabolic ment group was increased but did not reach statistical
agent, whereas the other current therapies are antiresorptive significance (HR, 1.2; 95% CI, 1.0–1.5). Estrogen only ther-
in effect. PTH effectively reduces vertebral and nonvertebral apy, on the other hand, was associated with an increased risk
fracture risk (Neer et al. 2001). of stroke and thromboembolus, a decreased risk of hip and
Owing to these many considerations, the choice of osteo- other fractures (Hip fracture HR, 0.6; 95% CI 0.4–0.9; other
porosis treatment should be determined for each individual fracture HR, 0.7; 95% CI 0.6–0.8), and a nonsignificantly
patient, taking into account the risks and benefits of all the reduced risk of CHD and breast cancer. The reduced risk of
therapies. As a result of the strong fracture-risk reduction fracture was seen shortly after the initiation of therapy in
evidence with bisphosphonate therapy, these medications both the treatment groups (Rossouw et al. 2002).
are currently considered first-line therapy. Gastrointestinal There is a debate regarding the applicability of the WHI
side effects are the most common problems associated with data to the women seen in practice, since the women partici-
oral bisphosphonates, which can be overcome with the use pating in the treatment groups were on an average 63 years
of intravenous zoledronic acid or ibandronate. There are old and had been postmenopausal for approximately 12–16
also reported cases of osteonecrosis of the jaw and atrial years when the therapy was initiated. A meta-analysis found
fibrillation with the use of bisphosphonates (see other that HRT appeared to be cardioprotective when started in
Myths). A full understanding of all putative adverse events younger women, but no risk reduction was seen when used
is not available. Raloxifene is associated with an increased in older women (Salpeter et al. 2006). Similarly, re-analysis
risk of thromboembolic events. Teriparatide appears safe of the WHI data, stratified by the time since menopause,
with rare adverse events, but due to cost and inconvenience revealed a direct correlation between increasing CHD risk
of a daily injection, it should be reserved for those with and time since menopause (Rossouw et al. 2007).
severe osteoporosis or persons who fracture despite bispho- Further randomized, controlled trials evaluating the risks
sphonate therapy. and benefits of HRT in younger postmenopausal women are
pending. In the interim, the use of HRT should be individual-
ized for each patient, based on age, symptoms, and pertinent
risk factors. Although HRT is known to be beneficial in
40.7 Osteoporosis Treatment:
reducing fracture risk and treating menopausal symptoms,
Hormone Replacement Therapy there are other treatment options that should be considered if
the patient or physician has concerns regarding the use of
Myth: Older postmenopausal women should be on long-term HRT.
hormone replacement for bone health.
Reality: Since the publication of the Women's Health
Initiative (WHI) results in 2002, there has been considerable 40.8 Osteoporosis Treatment:
controversy surrounding the risks and benefits of hormone Bisphosphonates
replacement therapy (HRT) (Rossouw et al. 2002). The WHI
proved that estrogen is beneficial in preventing bone loss and
Pearl: Osteonecrosis of the jaw is uncommon when bisphos-
reducing fracture risk. However, it heightened concerns
phonates are used for osteoporosis.
about the association of HRT on the risk of invasive breast
cancer, thromboembolism, and cardiovascular disease. Comment: Osteonecrosis of the jaw (ONJ), manifested by a
The WHI evaluated the effect of estrogen only (for women nonhealing area of exposed bone in the jaw, has been associ-
without a uterus) or estrogen plus progesterone (for women ated with bisphosphonate therapy (Woo et al. 2006). Most
with a history of a hysterectomy) in 16,608 women, with pri- cases reported have occurred in the setting of bisphospho-
mary endpoints of prevention of coronary heart disease nate use for metastatic breast cancer or myeloma, have fol-
(CHD), and secondary effects of other cardiovascular dis- lowed intravenous bisphosphonate, and have commonly been
ease prevention and reduction of hip and other fractures predated by dental procedures including tooth extraction and
(Rossouw et al. 2002). Women treated with combination dental implants. A large observational experience at the MD
estrogen plus progesterone were found to have an increased Anderson medical center suggests an incidence of 1–2% for
risk of invasive breast cancer (hazard ratio {HR}, 1.2; 95% patients receiving high dose intravenous bisphosphonates for
confidence interval {CI}, 1.0–1.5), as well as stroke, throm- malignancies (Hoff et al. 2008).
boembolism, and pulmonary thromboembolus. The risk of The current incidence of osteonecrosis of the jaw when
all fractures was reduced in treated women (Hip fracture HR, these bisphosphonates are used for osteoporosis is estimated
40 Osteoporosis 401

at just below one per hundred thousand exposed. There have Hu H, Caminis J, Tong K, Rosario-Jansen T, Krasnow J, Hue TF,
been few controlled studies to adequately address this ques- Sellmeyer D, Eriksen EF, Cummings SR. Once-yearly zoledronic
acid for treatment of postmenopausal osteoporosis. N Engl J Med.
tion. Large clinical trials of intravenous zoledronic acid for 2007;356(18):1809–22
treatment of osteoporosis have demonstrated only one poten- Bolland MJ, Barber PA, Doughty RN, Mason B, Horne A, Ames R,
tial case in persons exposed to zoledronic acid (and one in Gamble GD, Grey A, Reid IR. Vascular events in healthy older
those exposed to the control) among over eight thousand per- women receiving calcium supplementation: randomised controlled
trial. BMJ. 2008;336(7638):262–6
sons treated (Lyles et al. 2007; Black et al. 2007). What Borghi L, Schianchi T, Meschi T, Guerra A, Allegri F, Maggiore U,
remains unknown is whether the experience in persons Novarini A. Comparison of two diets for the prevention of recurrent
receiving high dose intravenous bisphosphonates (for malig- stones in idiopathic hypercalciuria. N Engl J Med. 2002;346(2):
nancy) is a precursor to what may become a more common 77–84
Breslau NA, McGuire JL, Zerwekh JE, Pak CY. The role of dietary
problem as more osteoporosis patients are treated with oral sodium on renal excretion and intestinal absorption of calcium and
and intravenous bisphosphonates at lower doses but for lon- on vitamin D metabolism. J Clin Endocrinol Metab. 1982;55(2):
ger periods of time. With further data pending, the clinician 369–73
and patient must weigh the risk and benefits with respect to Burge R, Dawson-Hughes B, Solomon DH, Wong JB, King A, Tosteson A.
Incidence and economic burden of osteoporosis-related fractures in the
the clear fracture-risk reduction and the putative risk of United States, 2005–2025. J Bone Miner Res. 2007;22(3): 465–75
osteonecrosis of the jaw. Center JR, Bliuc D, Nguyen TV, Eisman JA. Risk of subsequent frac-
ture after low-trauma fracture in men and women. JAMA.
2007;297(4):387–94
Chandra P, Wolfenden LL, Ziegler TR, Tian J, Luo M, Stecenko AA,
References Chen TC, Holick MF, Tangpricha V. Treatment of vitamin D defi-
ciency with UV light in patients with malabsorption syndromes: a case
series. Photodermatol Photoimmunol Photomed. 2007;23(5): 179–85
Adachi JD, Saag KG, Delmas PD, Liberman UA, Emkey RD, Seeman E, Chapuy MC, Arlot ME, Duboeuf F, Brun J, Crouzet B, Arnaud S,
Lane NE, Kaufman JM, Poubelle PE, Hawkins F, Correa-Rotter R, Delmas PD, Meunier PJ. Vitamin D3 and calcium to prevent hip
Menkes CJ, Rodriguez-Portales JA, Schnitzer TJ, Block JA, Wing J, fractures in the elderly women. N Engl J Med. 1992;327(23):
McIlwain HH, Westhovens R, Brown J, Melo-Gomes JA, Gruber BL, 1637–42
Yanover MJ, Leite MO, Siminoski KG, Nevitt MC, Sharp JT, Malice Chesnut IC, Skag A, Christiansen C, Recker R, Stakkestad JA,
MP, Dumortier T, Czachur M, Carofano W, Daifotis A. Two-year Hoiseth A, Felsenberg D, Huss H, Gilbride J, Schimmer RC, Delmas
effects of alendronate on bone mineral density and vertebral fracture in PD. Effects of oral ibandronate administered daily or intermittently
patients receiving glucocorticoids: a randomized, double-blind, pla- on fracture risk in postmenopausal osteoporosis. J Bone Miner Res.
cebo-controlled extension trial. Arthritis Rheum. 2001; 44(1):202–11 2004;19(8):1241–9
Armas LA, Dowell S, Akhter M, Duthuluru S, Huerter C, Hollis BW, Cranney A, Wells GA, Yetisir E, Adami S, Cooper C, Delmas PD,
Lund R, Heaney RP. Ultraviolet-B radiation increases serum Miller PD, Papapoulos S, Reginster JY, Sambrook PN, Silverman S,
25-hydroxyvitamin D levels: the effect of UVB dose and skin color. Siris E, Adachi JD. Ibandronate for the prevention of nonvertebral
J Am Acad Dermatol. 2007;57(4):588–93 fractures: a pooled analysis of individual patient data. Osteoporos
Bergstrom U, Bjornstig U, Stenlund H, Jonsson H, Svensson O. Fracture Int. 2009;20(2):291–7
mechanisms and fracture pattern in men and women aged 50 years Cuddihy MT, Gabriel SE, Crowson CS, O’Fallon WM, Melton LJ III.
and older: a study of a 12-year population-based injury register, Forearm fractures as predictors of subsequent osteoporotic frac-
Umea, Sweden. Osteoporos Int. 2008;19(9):1267–73 tures. Osteoporos Int. 1999;9(6):469–75
Bhutto A, Morley JE. The clinical significance of gastrointestinal Curhan GC, Willett WC, Knight EL, Stampfer MJ. Dietary factors and
changes with aging. Curr Opin Clin Nutr Metab Care. 2008;11(5): the risk of incident kidney stones in younger women: Nurses’ Health
651–60 Study II. Arch Intern Med. 2004;164(8):885–91
Binkley N, Krueger D, Cowgill CS, Plum L, Lake E, Hansen KE, Dawson-Hughes B, Harris SS, Krall EA, Dallal GE. Effect of calcium
DeLuca HF, Drezner MK. Assay variation confounds the diagnosis and vitamin D supplementation on bone density in men and women
of hypovitaminosis D: a call for standardization. J Clin Endocrinol 65 years of age or older. N Engl J Med. 1997;337(10):670–6
Metab. 2004;89(7):3152–7 Delmas PD, Adami S, Strugala C, Stakkestad JA, Reginster JY,
Binkley N, Novotny R, Krueger D, Kawahara T, Daida YG, Lensmeyer G, Felsenberg D, Christiansen C, Civitelli R, Drezner MK, Recker RR,
Hollis BW, Drezner MK. Low vitamin D status despite abundant sun Bolognese M, Hughes C, Masanauskaite D, Ward P, Sambrook P,
exposure. J Clin Endocrinol Metab. 2007;92(6):2130–5 Reid DM. Intravenous ibandronate injections in postmenopausal
Bischoff-Ferrari HA, Willett WC, Wong JB, Giovannucci E, Dietrich T, women with osteoporosis: one-year results from the dosing intrave-
Dawson-Hughes B. Fracture prevention with vitamin D supplemen- nous administration study. Arthritis Rheum. 2006;54(6): 1838–46
tation: a meta-analysis of randomized controlled trials. JAMA. Eisman JA, Shepard RM, DeLuca HF. Determination of 25-hydroxyvita-
2005;293(18):2257–64 min D2 and 25-hydroxyvitamin D3 in human plasma using high-pres-
Black DM, Cummings SR, Karpf DB, Cauley JA, Thompson DE, sure liquid chromatography. Anal Biochem. 1977;80(1): 298–305
Nevitt MC, Bauer DC, Genant HK, Haskell WL, Marcus R, Ott SM, Emkey RD, Lindsay R, Lyssy J, Weisberg JS, Dempster DW, Shen V.
Torner JC, Quandt SA, Reiss TF, Ensrud KE. Randomised trial of The systemic effect of intraarticular administration of corticosteroid
effect of alendronate on risk of fracture in women with existing ver- on markers of bone formation and bone resorption in patients with
tebral fractures. Fracture Intervention Trial Research Group. Lancet rheumatoid arthritis. Arthritis Rheum. 1996;39(2):277–82
1996;348(9041):1535–41 Evista (raloxifene) [package insert]. Indianapolis, IN: Eli Lilly; 2007
Black DM, Delmas PD, Eastell R, Reid IR, Boonen S, Cauley JA, FRAX Who Fracture Risk Assessment Tool. http://www.shef.ac.uk/
Cosman F, Lakatos P, Leung PC, Man Z, Mautalen C, Mesenbrink P, FRAX/ (2008)
402 K. G. Saag et al.

Furstenberg AL, Mezey MD. Differences in outcome between black Hurwitz A, Brady DA, Schaal SE, Samloff IM, Dedon J, Ruhl CE.
and white elderly hip fracture patients. J Chronic Dis. 1987; 40(10): Gastric acidity in older adults. JAMA. 1997;278(8):659–62
931–8 ISCD. The International Society for Clinical Densitometry:2007 ISCD
Gankam Kengne F, Andres C, Sattar L, Melot C, Decaux G. Mild Official Positions Brochure. http://www.iscd.org/Visitors/pdfs/
hyponatremia and risk of fracture in the ambulatory elderly. QJM. ISCD2007OfficialPositions-Combined-AdultandPediatric.pdf
2008;101(7):583–8 (2007)
Grant AM, Avenell A, Campbell MK, McDonald AM, MacLennan GS, Israel E, Banerjee TR, Fitzmaurice GM, Kotlov TV, LaHive K,
McPherson GC, Anderson FH, Cooper C, Francis RM, Donaldson C, LeBoff MS. Effects of inhaled glucocorticoids on bone density in
Gillespie WJ, Robinson CM, Torgerson DJ, Wallace WA. Oral vita- premenopausal women. N Engl J Med. 2001;345(13):941–7
min D3 and calcium for secondary prevention of low-trauma frac- Jackson RD, LaCroix AZ, Gass M, Wallace RB, Robbins J, Lewis CE,
tures in elderly people (Randomised Evaluation of Calcium or Bassford T, Beresford SA, Black HR, Blanchette P, Bonds DE,
vitamin D, RECORD): a randomised placebo-controlled trial. Brunner RL, Brzyski RG, Caan B, Cauley JA, Chlebowski RT,
Lancet 2005;365(9471):1621–8 Cummings SR, Granek I, Hays J, Heiss G, Hendrix SL, Howard BV,
Haddad JG, Chyu KJ. Competitive protein-binding radioassay for Hsia J, Hubbell FA, Johnson KC, Judd H, Kotchen JM, Kuller LH,
25-hydroxycholecalciferol. J Clin Endocrinol Metab. 1971;33(6): Langer RD, Lasser NL, Limacher MC, Ludlam S, Manson JE,
992–5 Margolis KL, McGowan J, Ockene JK, O’Sullivan MJ, Phillips L,
Hanzlik RP, Fowler SC, Fisher DH. Relative bioavailability of calcium Prentice RL, Sarto GE, Stefanick ML, Van Horn L, Wactawski-
from calcium formate, calcium citrate, and calcium carbonate. Wende J, Whitlock E, Anderson GL, Assaf AR, Barad D. Calcium
J Pharmacol Exp Ther. 2005;313(3):1217–22 plus vitamin D supplementation and the risk of fractures. N Engl J
Heaney RP, Dowell MS, Barger-Lux MJ. Absorption of calcium as the Med. 2006;354(7):669–83
carbonate and citrate salts, with some observations on method. Jacobsen SJ, Goldberg J, Miles TP, Brody JA, Stiers W, Rimm AA.
Osteoporos Int. 1999;9(1):19–23 Race and sex differences in mortality following fracture of the hip.
Heaney RP, Dowell MS, Rafferty K, Bierman J. Bioavailability of the Am J Public Health. 1992;82(8):1147–50
calcium in fortified soy imitation milk, with some observations on Jones G, Nguyen T, Sambrook PN, Kelly PJ, Gilbert C, Eisman JA.
method. Am J Clin Nutr. 2000;71(5):1166–9 Symptomatic fracture incidence in elderly men and women: the
Heaney RP, Rafferty K, Dowell MS, Bierman J. Calcium fortification sys- Dubbo Osteoporosis Epidemiology Study (DOES). Osteoporos Int.
tems differ in bioavailability. J Am Diet Assoc. 2005;105(5): 807–9 1994;4(5):277–82
Heaney RP, Rafferty K. The settling problem in calcium-fortified soy- Kanis JA, Johnell O, Oden A, Dawson A, De Laet C, Jonsson B. Ten
bean drinks. J Am Diet Assoc. 2006;106(11):1753; author reply year probabilities of osteoporotic fractures according to BMD and
1755 diagnostic thresholds. Osteoporos Int. 2001;12(12):989–95
Heaney RP. Absorbability and utility of calcium in mineral waters. Am Karlsson MK, Hasserius R, Obrant KJ. Individuals who sustain nonos-
J Clin Nutr. 2006a;84(2):371–4 teoporotic fractures continue to also sustain fragility fractures.
Heaney RP. Role of dietary sodium in osteoporosis. J Am Coll Nutr. Calcif Tissue Int. 1993;53(4):229–31
2006b;25(3 Suppl):271S6S Kellie SE, Brody JA. Sex-specific and race-specific hip fracture rates.
Hedlund R, Lindgren U. Trauma type, age, and gender as determinants Am J Public Health. 1990;80(3):326–8
of hip fracture. J Orthop Res. 1987;5(2):242–6 Kenny AM, Prestwood KM, Biskup B, Robbins B, Zayas E, Kleppinger
Heikinheimo RJ, Inkovaara JA, Harju EJ, Haavisto MV, Kaarela RH, A, Burleson JA, Raisz LG. Comparison of the effects of calcium
Kataja JM, Kokko AM, Kolho LA, Rajala SA. Annual injection of loading with calcium citrate or calcium carbonate on bone turnover
vitamin D and fractures of aged bones. Calcif Tissue Int. 1992;51(2): in postmenopausal women. Osteoporos Int. 2004; 15(4):290–4
105–10 Khan A. Premenopausal women and low bone density. Can Fam
Heller HJ, Greer LG, Haynes SD, Poindexter JR, Pak CY. Pharma- Physician. 2006;52:743–7
cokinetic and pharmacodynamic comparison of two calcium sup- Khan AA, Syed Z. Bone densitometry in premenopausal women: syn-
plements in postmenopausal women. J Clin Pharmacol. 2000; thesis and review. J Clin Densitom. 2004;7(1):85–92
40(11):1237–44 Kudlacek S, Schneider B, Resch H, Freudenthaler O, Willvonseder R.
Heller HJ, Stewart A, Haynes S, Pak CY. Pharmacokinetics of calcium Gender differences in fracture risk and bone mineral density.
absorption from two commercial calcium supplements. J Clin Maturitas 2000;36(3):173–80
Pharmacol. 1999;39(11):1151–4 Lappe JM, Heaney RP. Calcium supplementation: results may not be
Hoff AO, Toth BB, Altundag K, Johnson MM, Warneke CL, Hu M, generalisable. BMJ. 2008;336(7641):403; author reply 404
Nooka A, Sayegh G, Guarneri V, Desrouleaux K, Cui J, Adamus Larsen ER, Mosekilde L, Foldspang A. Vitamin D and calcium supple-
A, Gagel RF, Hortobagyi GN. Frequency and risk factors associ- mentation prevents osteoporotic fractures in elderly community
ated with osteonecrosis of the jaw in cancer patients treated with dwelling residents: a pragmatic population-based 3-year interven-
intravenous bisphosphonates. J Bone Miner Res. 2008;23(6): tion study. J Bone Miner Res. 2004;19(3):370–8
826–36 Lewiecki EM. Premenopausal bone health assessment. Curr Rheumatol
Holick MF. Deficiency of sunlight and vitamin D. BMJ. 2008;336(7657): Rep. 2005;7(1):46–52
1318–9 Lips P, Graafmans WC, Ooms ME, Bezemer PD, Bouter LM Vitamin D.
Holick MF. High prevalence of vitamin D inadequacy and implications supplementation and fracture incidence in elderly persons. A ran-
for health. Mayo Clin Proc. 2006a;81(3):353–73 domized, placebo-controlled clinical trial. Ann Intern Med.
Holick MF. Resurrection of vitamin D deficiency and rickets. J Clin 1996;124(4):400–6
Invest. 2006b;116(8):2062–72 Lips P. Vitamin D deficiency and secondary hyperparathyroidism in the
Holick MF. Vitamin D deficiency. N Engl J Med. 2007;357(3):266–81 elderly: consequences for bone loss and fractures and therapeutic
Hollis BW. Measuring 25-hydroxyvitamin D in a clinical environment: implications. Endocr Rev. 2001;22(4):477–501
challenges and needs. Am J Clin Nutr. 2008;88(2):507S-–10S Lissner D, Mason RS, Posen S. Stability of vitamin D metabolites in
Hsia J, Heiss G, Ren H, Allison M, Dolan NC, Greenland P, Heckbert SR, human blood serum and plasma. Clin Chem. 1981;27(5):773–4
Johnson KC, Manson JE, Sidney S, Trevisan M. Calcium/vitamin D Lyles KW, Colon-Emeric CS, Magaziner JS, Adachi JD, Pieper CF,
supplementation and cardiovascular events. Circulation 2007; Mautalen C, Hyldstrup L, Recknor C, Nordsletten L, Moore KA,
115(7):846–54 Lavecchia C, Zhang J, Mesenbrink P, Hodgson PK, Abrams K,
40 Osteoporosis 403

Orloff JJ, Horowitz Z, Eriksen EF, Boonen S. Zoledronic acid and Ornoy A, Wajnberg R, Diav-Citrin O. The outcome of pregnancy fol-
clinical fractures and mortality after hip fracture. N Engl J Med. lowing pre-pregnancy or early pregnancy alendronate treatment.
2007;357(18):1799–809 Reprod Toxicol. 2006;22(4):578–9
Lyons RA, Johansen A, Brophy S, Newcombe RG, Phillips CJ, Lervy B, Orwoll E. Assessing bone density in men. J Bone Miner Res.
Evans R, Wareham K, Stone MD. Preventing fractures among older 2000;15(10):1867–70
people living in institutional care: a pragmatic randomised double Papadimitropoulos E, Wells G, Shea B, Gillespie W, Weaver B,
blind placebo controlled trial of vitamin D supplementation. Zytaruk N, Cranney A, Adachi J, Tugwell P, Josse R, Greenwood C,
Osteoporos Int. 2007;18(6):811–8 Guyatt G. Meta-analyses of therapies for postmenopausal osteopo-
Mackey DC, Lui LY, Cawthon PM, Bauer DC, Nevitt MC, Cauley JA, rosis. VIII: meta-analysis of the efficacy of vitamin D treatment in
Hillier TA, Lewis CE, Barrett-Connor E, Cummings SR. High- preventing osteoporosis in postmenopausal women. Endocr Rev.
trauma fractures and low bone mineral density in older women and 2002;23(4):560–9
men. JAMA. 2007;298(20):2381–8 Patlas N, Golomb G, Yaffe P, Pinto T, Breuer E, Ornoy A. Transplacental
Manolagas SC, Weinstein RS. New developments in the pathogenesis effects of bisphosphonates on fetal skeletal ossification and mineral-
and treatment of steroid-induced osteoporosis. J Bone Miner Res. ization in rats. Teratology. 1999;60(2):68–73
1999;14(7):1061–6 Prince RL, Devine A, Dhaliwal SS, Dick IM. Effects of calcium supple-
Martini L, Wood RJ. Relative bioavailability of calcium-rich dietary mentation on clinical fracture and bone structure: results of a 5-year,
sources in the elderly. Am J Clin Nutr. 2002;76(6):1345–50 double-blind, placebo-controlled trial in elderly women. Arch Intern
Maunsell Z, Wright DJ, Rainbow SJ. Routine isotope-dilution liquid Med. 2006;166(8):869–75
chromatography-tandem mass spectrometry assay for simultaneous Rafferty K, Walters G, Heaney RP. Calcium fortificants: overview and
measurement of the 25-hydroxy metabolites of vitamins D2 and D3. strategies for improving calcium nutriture of the U.S. population. J
Clin Chem. 2005;51(9):1683–90 Food Sci. 2007;72(9):R152–8
Melton LJ III, Atkinson EJ, O’Fallon WM, Wahner HW, Riggs BL. Rauch F, Glorieux FH. Osteogenesis imperfecta. Lancet 2004;363(9418):
Long-term fracture prediction by bone mineral assessed at different 1377–85
skeletal sites. J Bone Miner Res. 1993;8(10):1227–33 Recker RR, Weinstein RS, Chesnut CH III, Schimmer RC, Mahoney P,
Melton LJ III, Beck TJ, Amin S, Khosla S, Achenbach SJ, Oberg AL, Hughes C, Bonvoisin B, Meunier PJ. Histomorphometric evaluation
Riggs BL. Contributions of bone density and structure to fracture risk of daily and intermittent oral ibandronate in women with postmeno-
assessment in men and women. Osteoporos Int. 2005;16(5): 460–7 pausal osteoporosis: results from the BONE study. Osteoporos Int.
Melton LJ III, Sampson JM, Morrey BF, Ilstrup DM. Epidemiologic 2004;15(3):231–7
features of pelvic fractures. Clin Orthop Relat Res. 1981;155:43–7 Recker RR. Calcium absorption and achlorhydria. N Engl J Med.
Meyer WJ III, Transbol I, Bartter FC, Delea C. Control of calcium 1985;313(2):70–3
absorption: effect of sodium chloride loading and depletion. Reid DM, Hughes RA, Laan RF, Sacco-Gibson NA, Wenderoth DH,
Metabolism 1976;25(9):989–93 Adami S, Eusebio RA, Devogelaer JP. Efficacy and safety of daily
Mezquita-Raya P, Munoz-Torres M, Luna JD, Luna V, Lopez- risedronate in the treatment of corticosteroid-induced osteoporosis
Rodriguez F, Torres-Vela E, Escobar-Jimenez F. Relation between in men and women: a randomized trial. European Corticosteroid-
vitamin D insufficiency, bone density, and bone metabolism in Induced Osteoporosis Treatment Study. J Bone Miner Res. 2000;
healthy postmenopausal women. J Bone Miner Res. 2001;16(8): 15(6):1006–13
1408–15 Reid IR, Mason B, Horne A, Ames R, Reid HE, Bava U, Bolland MJ,
Miacalcin® Nasal Spray [Package Insert], East Hanover, NJ: Novartis Gamble GD. Randomized controlled trial of calcium in healthy
Pharmaceuticals; 2003 older women. Am J Med. 2006;119(9):777–85
Mikuls TR, Saag KG, George V, Mudano AS, Banerjee S. Racial dis- Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C,
parities in the receipt of osteoporosis related healthcare among Stefanick ML, Jackson RD, Beresford SA, Howard BV, Johnson KC,
community-dwelling older women with arthritis and previous frac- Kotchen JM, Ockene J. Risks and benefits of estrogen plus proges-
ture. J Rheumatol. 2005;32(5):870–5 tin in healthy postmenopausal women: principal results from the
Miller RG, Ashar BH, Cohen J, Camp M, Coombs C, Johnson E, Women’s Health Initiative randomized controlled trial. JAMA.
Schneyer CR. Disparities in osteoporosis screening between at-risk 2002;288(3):321–33
African-American and white women. J Gen Intern Med. 2005;20(9): Rossouw JE, Prentice RL, Manson JE, Wu L, Barad D, Barnabei VM,
847–51 Ko M, LaCroix AZ, Margolis KL, Stefanick ML. Postmenopausal
Moreira Kulak CA, Schussheim DH, McMahon DJ, Kurland E, hormone therapy and risk of cardiovascular disease by age and years
Silverberg SJ, Siris ES, Bilezikian JP, Shane E. Osteoporosis and since menopause. JAMA. 2007;297(13):1465–77
low bone mass in premenopausal and perimenopausal women. Saag KG, Emkey R, Schnitzer TJ, Brown JP, Hawkins F, Goemaere S,
Endocr Pract. 2000;6(4):296–304 Thamsborg G, Liberman UA, Delmas PD, Malice MP, Czachur M,
Mudano AS, Casebeer L, Patino F, Allison JJ, Weissman NW, Kiefe CI, Daifotis AG. Alendronate for the prevention and treatment of gluco-
Person S, Gilbert D, Saag KG. Racial disparities in osteoporosis preven- corticoid-induced osteoporosis. Glucocorticoid-Induced Osteoporosis
tion in a managed care population. South Med J. 2003;96(5): 445–51 Intervention Study Group. N Engl J Med. 1998;339(5): 292–9
National Osteoporosis Foundation. Physician’s guide to prevention and Salpeter SR, Walsh JM, Greyber E, Salpeter EE. Brief report: coronary
treatment of osteoporosis. Belle Mead, NJ: Excerpta Medica; 2008 heart disease events associated with hormone therapy in younger and
Neer RM, Arnaud CD, Zanchetta JR, Prince R, Gaich GA, Reginster JY, older women. A meta-analysis. J Gen Intern Med. 2006;21(4): 363–6
Hodsman AB, Eriksen EF, Ish-Shalom S, Genant HK, Wang O, Sanders KM, Pasco JA, Ugoni AM, Nicholson GC, Seeman E,
Mitlak BH. Effect of parathyroid hormone (1–34) on fractures and Martin TJ, Skoric B, Panahi S, Kotowicz MA. The exclusion of high
bone mineral density in postmenopausal women with osteoporosis. trauma fractures may underestimate the prevalence of bone fragility
N Engl J Med. 2001;344(19):1434–41 fractures in the community: the Geelong Osteoporosis Study. J Bone
Novartis API. http://www.fda.gov/cder/foi/label/2005/20036s030lbl. Miner Res. 1998;13(8):1337–42
pdf (2004). Accessed 1 Oct 2008 Sanders KM, Seeman E, Ugoni AM, Pasco JA, Martin TJ, Skoric B,
O’Connell MB, Madden DM, Murray AM, Heaney RP, Kerzner LJ. Effects Nicholson GC, Kotowicz MA. Age- and gender-specific rate of
of proton pump inhibitors on calcium carbonate absorption in women: fractures in Australia: a population-based study. Osteoporos Int.
a randomized crossover trial. Am J Med. 2005;118(7): 778–81 1999;10(3):240–7
404 K. G. Saag et al.

Schousboe JT, Taylor BC, Fink HA, Kane RL, Cummings SR, van Staa TP, Laan RF, Barton IP, Cohen S, Reid DM, Cooper C. Bone
Orwoll ES, Melton LJ III, Bauer DC, Ensrud KE. Cost-effectiveness density threshold and other predictors of vertebral fracture in
of bone densitometry followed by treatment of osteoporosis in older patients receiving oral glucocorticoid therapy. Arthritis Rheum.
men. JAMA. 2007;298(6):629–37 2003;48(11):3224–9
Serfaty-Lacrosniere C, Wood RJ, Voytko D, Saltzman JR, Pedrosa M, van Staa TP, Leufkens HG, Abenhaim L, Zhang B, Cooper C. Use of
Sepe TE, Russell RR. Hypochlorhydria from short-term omepra- oral corticosteroids and risk of fractures. J Bone Miner Res.
zole treatment does not inhibit intestinal absorption of calcium, 2000;15(6):993–1000
phosphorus, magnesium or zinc from food in humans. J Am Coll Visser M, Deeg DJ, Puts MT, Seidell JC, Lips P. Low serum concentra-
Nutr. 1995;14(4):364–8 tions of 25-hydroxyvitamin D in older persons and the risk of nurs-
Siris ES, Miller PD, Barrett-Connor E, Faulkner KG, Wehren LE, ing home admission. Am J Clin Nutr. 2006;84(3):616–22; quiz
Abbott TA, Berger ML, Santora AC, Sherwood LM. Identification 671–2
and fracture outcomes of undiagnosed low bone mineral density in Wallach S, Cohen S, Reid DM, Hughes RA, Hosking DJ, Laan RF,
postmenopausal women: results from the National Osteoporosis Doherty SM, Maricic M, Rosen C, Brown J, Barton I, Chines AA.
Risk Assessment. JAMA. 2001;286(22):2815–22 Effects of risedronate treatment on bone density and vertebral frac-
Stakkestad JA, Benevolenskaya LI, Stepan JJ, Skag A, Nordby A, ture in patients on corticosteroid therapy. Calcif Tissue Int.
Oefjord E, Burdeska A, Jonkanski I, Mahoney P. Intravenous iban- 2000;67(4):277–85
dronate injections given every three months: a new treatment option Wells G, Cranney A, Peterson J, Boucher M, Shea B, Robinson V,
to prevent bone loss in postmenopausal women. Ann Rheum Dis. Coyle D, Tugwell P. Risedronate for the primary and secondary pre-
2003;62(10):969–75 vention of osteoporotic fractures in postmenopausal women.
Tangpricha V, Pearce EN, Chen TC, Holick MF. Vitamin D insuffi- Cochrane Database Syst Rev. 2008a;(1):CD004523
ciency among free-living healthy young adults. Am J Med. Wells GA, Cranney A, Peterson J, Boucher M, Shea B, Robinson V,
2002;112(8):659–62 Coyle D, Tugwell P. Alendronate for the primary and secondary pre-
Taylor EN, Curhan GC. Diet and fluid prescription in stone disease. vention of osteoporotic fractures in postmenopausal women.
Kidney Int. 2006;70(5):835–9 Cochrane Database Syst Rev. 2008b;(1):CD001155
Taylor EN, Curhan GC. Role of nutrition in the formation of calcium- Williams CP, Child DF, Hudson PR, Davies GK, Davies MG, John R,
containing kidney stones. Nephron Physiol. 2004;98(2):p55–63 Anandaram PS, De Bolla AR. Why oral calcium supplements may
Taylor EN, Stampfer MJ, Curhan GC. Dietary factors and the risk of reduce renal stone disease: report of a clinical pilot study. J Clin
incident kidney stones in men: new insights after 14 years of follow- Pathol. 2001;54(1):54–62
up. J Am Soc Nephrol. 2004;15(12):3225–32 Wong CA, Walsh LJ, Smith CJ, Wisniewski AF, Lewis SA, Hubbard R,
Teucher B, Dainty JR, Spinks CA, Majsak-Newman G, Berry DJ, Cawte S, Green DJ, Pringle M, Tattersfield AE. Inhaled corticoster-
Hoogewerff JA, Foxall RJ, Jakobsen J, Cashman KD, Flynn A, oid use and bone-mineral density in patients with asthma. Lancet
Fairweather-Tait SJ. Sodium and bone health: impact of moderately 2000;355(9213):1399–403
high and low salt intakes on calcium metabolism in postmenopausal Woo SB, Hellstein JW, Kalmar JR. Narrative [corrected] review: bis-
women. J Bone Miner Res. 2008;23(9):1477–85 phosphonates and osteonecrosis of the jaws. Ann Intern Med.
Ton FN, Gunawardene SC, Lee H, Neer RM. Effects of low-dose pred- 2006;144(10):753–61
nisone on bone metabolism. J Bone Miner Res. 2005; 20(3): Zerwekh JE. Blood biomarkers of vitamin D status. Am J Clin Nutr.
464–70 2008;87(4):1087S–91S
van Staa TP, Dennison EM, Leufkens HG, Cooper C. Epidemiology
of fractures in England and Wales. Bone. 2001;29(6): 517–22
 Paget’s Disease of Bone
41
Margaret Seton

41.1 Overview of Paget’s Disease › Bisphosphonates are the treatment of choice. However,
calcitonin has a role in the elderly patient with renal
insufficiency in whom relief of pain rather than remis-
› Sir James Paget described the disease as follows: It
sion is a reasonable outcome.
begins in middle age or later, is very slow in progress,
may continue for many years without influence on the › Resistance to the effects of one bisphosphonate does
not predict resistance to another.
general health, and may give no other trouble than
those which are due to the changes of shape, size, and
direction of the diseased bones (Paget 1877).
› Paget’s disease is a focal disorder of bone remodeling
that occurs in an aging skeleton. Aberrant osteoclast Pearl: Most patients with Paget’s disease are asymptomatic.
activation is the critical lesion leading to the disease
phenotype. Comment: Despite its remarkable appearance on plain radio-
› Paget’s disease leads to impaired structural integrity graphs and bone scintigraphy, most patients with Paget’s dis-
and distortion of bone, the consequences of which ease of bone are asymptomatic. The diagnosis typically is
include localized bone pain, deformity, fracture, and made incidentally through findings on radiographs or blood
early arthritis. tests ordered for other reasons. Treatment is indicated when
› Neurological sequelae such as hearing loss, spinal pagetic bone affects weight-bearing limbs, approximates a
stenosis, and neuropathies can occur, depending on the joint, or involves the skull or spine. Treatment is also indi-
sites of affected bones. cated in patients planning elective total hip or knee replace-
› Osteosarcomas arise in pagetic bone in a small number ments that involve pagetic bone, to reduce blood loss from
of cases (Seton 2008; Hansen et al. 2006) these sites. Treatment rarely restores hearing in patients with
› Both familial and sporadic cases of Paget’s disease are Paget’s disease of the skull, straighten bowed limbs, or nor-
caused often by a mutation in SQSTM1. This mutation malize radiographs, but it is expected to ease pain and pre-
results in a C→ T transition mutation (P392L) that in vent disease progression. Early treatment is recommended in
turn results in aberrant osteoclast activation (Laurin et an effort to avoid complications of long-standing disease.
al. 2002). However, not all cases are associated with Pearl: The bones affected by Paget’s disease in a given indi-
this disease mutation (Kurihara et al. 2007; Morissette vidual remain unchanged over the lifetime of that person.
et al. 2006).
› The serum alkaline phosphatase, a marker of osteo- Comment: Paget’s disease tends to present in middle-aged or
blastic activity, is a good indicator of skeletal extent older individuals. The disorder has a predilection for the skull,
and activity of Paget’s disease. axial skeleton, and long bones of the lower extremity. The
› Spot urine assays for N-telopeptide levels have replaced number of bones involved in Paget’s disease tends to be fixed
24-h urine collections for hydroxyproline as a measure in a given individual, sometimes presenting as monostotic
of antiresorptive activity. disease and other times as a polyostotic disorder (Fig. 41.1).
› The goal of treating Paget’s disease with antiresorptive The persistence of the disease in certain bones in an indi-
agents such as the bisphosphonates and calcitonin is to vidual patient has implications for the initial evaluation of a
normalize the serum alkaline phosphatase and urine patient with possible Paget’s disease. This is because Paget’s
N-telopeptide levels. disease must be distinguished from cancers metastatic to
bone. Paget’s disease does not “metastasize” or extend across

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 405

DOI:10.1007/978-1-84800-934-9_41, © Springer Science + Business Media B.V. 2009
406 M. Seton

joints to involve contiguous bone. The involvement of new

bones strongly suggests an underlying malignancy.
A bone scan is indicated at diagnosis to identify the dis-
tribution of pagetic bone in the skeleton (Fig. 41.2). Bone
scintigraphy typically reveals intense uptake beginning at
one end of a bone and extending a variable distance through
that bone. Radiographs of the affected sites should be
obtained to confirm the diagnosis, define the anatomy of
the problem, and establish the need for treatment. The sub-
sequent involvement of new bones should be considered
metastatic cancer until proven otherwise by biopsy (Cooper
et al. 1999; Griffiths 1992; Haddaway et al. 2007).
Myth: Paget’s disease occurs in children.
Reality: This is not true. Familial idiopathic hyperphosphata-
sia, sometimes termed “juvenile Paget’s disease,” is a sepa-
rate entity. Familial idiopathic hyperphosphatasia is a rare
autosomal-recessive disorder that is ascribed in most cases to
a deficiency in osteoprotegerin. The disorder results in
ungoverned bone remodeling, deformity, and fractures in
children (Whyte et al. 2002).
Pearl: Paget’s disease begins in subchondral bone, moving
in one direction up or down the affected bone.
Comment: Early Paget’s disease is often described as purely
Fig. 41.1 Monostotic Paget’s disease, L tibia. Paget’s disease can
lytic. In fact, purely lytic lesions are an uncommon finding
involve one bone only or it can involve multiple bones. However, once
its pattern of bone involvement is established in an individual patient, it on plain radiographs at the time of diagnosis. Patients usu-
does not “metastasize” to new bones. Evidence of new bony abnormali- ally have radiographic evidence of both lysis (osteoclast
ties suggests the possibility of metastatic cancer until proven otherwise. activity) and thickened bone (osteoblast activity).
(the other site of intense uptake is the bladder)

Fig. 41.2 Bone scintigraphy in

Paget’s disease. Polyostotic
Paget’s disease evident on bone
scan, with the classic bowing of
the femur
41 Paget’s Disease of Bone 407

bone resorption, the drug also causes impaired mineralization

(Altman 1985; Eyres et al. 1992; Gibbs et al. 1986).
Myth: Treatment of patients with bisphosphonates invariably
eases their pain.
Reality: Paget’s disease is a focal disorder of an aging skel-
eton. Treatment is aimed at preventing the complications of
the existing disease, inhibiting disease progression, and eas-
ing the symptoms of bone pain at pagetic sites. Treatment
will alleviate pain if the pain stems from pagetic bone.
However, pain in Paget’s patients, many of whom are elderly,
can be multifactorial. Symptoms that result from degenera-
tive changes in the spine or hips do not respond well to bis-
phosphonates or calcitonin. Nonsteroidal antiinflammatory
drugs, physical therapy, or surgery may also be required.
Pearl: Teriparatide is contraindicated in patients with
Paget’s disease.
Fig. 41.3 Radiographic findings in late Paget’s disease of bone. There
is marked bone remodeling. The cortices become thickened. The trabe- Comment: Many elderly patients with Paget’s disease have
culae are coarsened, and the bone deformed osteoporosis, too. There may be some temptation to treat the
patient’s osteoporosis with teraparatide, a recombinant form
of human parathyroid hormone that stimulates bone forma-
Late Paget’s disease manifests itself as dense overgrowth tion. However, Paget’s disease is a contraindication to teri-
of bone. The characteristic radiographic findings in late dis- paratide. This is because of the rare but real risk of
ease are thickened cortices striped by tunneling osteoclasts, osteosarcoma arising from pagetic bone. Thus, bisphospho-
coarsened trabeculae, and marked remodeling of the affected nates and calcitonin are the current mainstays of Paget’s dis-
bone (Fig. 41.3). The bone enlarges, softens, and becomes ease treatment. Both treat osteoporosis as well as Paget’s
deformed. disease.
Myth: There is no effective treatment for Paget’s disease of
bone.
Reality: This is one of the most distressing myths of all related
References
to Paget’s disease. Bisphosphonates ease pain from pagetic
bone effectively, restore bone turnover markers to normal lev-
Altman RD. Long-term follow-up of therapy with intermittent etidronate
els, and result in the deposition of normal lamellar bone. disodium in Paget’s disease of bone. Am J Med. 1985;79(5):
Treatment may also result in the radiographic healing of lytic 583–90
lesions or a reduction in bone scan activity in some patients. Brown JP, Chines AA, Myers WR, et al Improvement of pagetic bone
No prospective clinical trials have yet been conducted to con- lesions with risedronate treatment: a radiologic study. Bone 2000;
26(3):263–7
firm the clinical suspicion that bisphosphonate therapy pre- Cooper C, Dennison E, Schafheutle K, et al Epidemiology of Paget’s
vents long-term complications of Paget’s disease. disease of bone. Bone 1999;24(5 Suppl):3S–5S.
Alendronate, risedronate, and zoledronic acid all lead to a Eyres KS, Marshall P, McCloskey E, et al Spontaneous fractures in a
biochemical remission: i.e., to normalization of serum alka- patient treated with low doses of etidronic acid (disodium etidronate).
Drug Saf. 1992;7(2):162–5
line phosphatase levels (Reid et al. 1996; Siris et al. 1996; Gibbs CJ, Aaron JE, Peacock M. Osteomalacia in Paget’s disease
Brown et al. 2000; Miller et al. 1999; Seton and Krane 2007). treated with short term, high dose sodium etidronate. Br Med J (Clin
This is incorrect. Retreatment is often required, prompted by Res Ed). 1986;292(6530):1227–9
rising serum alkaline phosphatase levels or recurrent bone Griffiths HJ. Radiology of Paget’s disease. Curr Opin Radiol.
1992;4(6):124–8
pain (Hosking et al. 2007; Reid et al. 2005). Haddaway MJ, Davie MW, McCall IW, et al Effect of age and gender
Pearl: Etidronate is contraindicated in patients who have on the number and distribution of sites in Paget’s disease of bone.
Br J Radiol. 2007;80(955):532–6
lytic lesions in weight-bearing bones. Hansen MF, Seton M, Merchant A. Osteosarcoma in Paget’s disease of
bone. J Bone Miner Res. 2006;21(Suppl 2):P58–63
Comment: Etidronate can worsen bone pain in pagetic limbs Hosking D, Lyles K, Brown JP, et al Long-term control of bone turn-
and heighten the risk of fracture. This stems from etidronate’s over in Paget’s disease with zoledronic acid and risedronate. J Bone
narrow therapeutic margin: at the doses required to suppress Miner Res. 2007;22(1):142–8
408 M. Seton

Kurihara N, Hiruma Y, Zhou H, et al Mutation of the sequestosome 1 Reid IR, Nicholson GC, Weinstein RS, et al Biochemical and radiologic
(p62) gene increases osteoclastogenesis but does not induce Paget improvement in Paget’s disease of bone treated with alendronate: a
disease. J Clin Invest. 2007;117(1):133–42 randomized, placebo-controlled trial. Am J Med. 1996;101(4):
Laurin N, Brown JP, Morissette J, et al Recurrent mutation of the gene 341–8
encoding sequestosome 1 (SQSTM1/p62) in Paget disease of bone. Seton M. Paget’s disease of bone.. In: Hochberg MC, Silman AJ,
Am J Hum Genet. 2002;70(6):1582–8 Smolen JS, Weinblatt ME, Weisman MH, editors. Rheumatology. 4th
Miller PD, Brown JP, Siris ES, et al A randomized, double-blind com- ed. Philadelphia: Mosby. 2008
parison of risedronate and etidronate in the treatment of Paget’s dis- Seton M, Krane SM. Use of zoledronic acid in the treatment of Paget’s
ease of bone. Paget’s Risedronate/Etidronate Study Group. Am J disease. Ther Clin Risk Manag. 2007;3(5):913–8
Med. 1999;106(5):513–20 Siris E, Weinstein RS, Altman R, et al Comparative study of alendronate
Morissette J, Laurin N, Brown JP. Sequestosome 1: mutation frequen- versus etidronate for the treatment of Paget’s disease of bone. J Clin
cies, haplotypes, and phenotypes in familial Paget’s disease of bone. Endocrinol Metab. 1996;81(3):961–7
J Bone Miner Res. 2006;21(Suppl 2):P38–44 Whyte MP, Obrecht SE, Finnegan PM, et al Osteoprotegerin deficiency
Paget SJ. On a form of chronic inflammation of bones (osteitis defor- and juvenile Paget’s disease. N Engl J Med. 2002;347(3):175–84
mans). Med Chir Trans. 1877;LX:29–53
Reid IR, Miller P, Lyles K, et al Comparison of a single infusion of
zoledronic acid with risedronate for Paget’s disease. N Engl J Med.
2005;353(9):898–908
 Sarcoidosis
42
Robert P. Baughman and Elyse E. Lower

f
roo
Myth: Sarcoidosis occurs only in black patients.
42.1 Overview of Sarcoidosis
Reality: Sarcoidosis patients in the United States often
› Sarcoidosis involves systemic inflammatory disorder encounter this Myth: “Hmmm. You’re white. I thought only
with noncaseating granulomatous inflammation in African–Americans got sarcoidosis.” On the contrary, sar-
affected organs. coidosis occurs in all ethnic groups. Data shown in Table 42.1

dP
› It commonly involves the lungs, eyes, skin, joints, confirm that black Americans have a higher rate of sarcoido-
lymph nodes, and upper respiratory tract. sis than white Americans do, but Scandinavians (an over-
› In the United States, prevalence of sarcoidosis is between whelmingly Caucasian population, obviously) have an
10 and 40 cases/100,000 population. In Sweden, the incidence rate of sarcoidosis that is higher than any group of
prevalence is higher, estimated to be between 60 and 80 Americans—black, white, Hispanic, or Asian.
cases/100,000 population. Although the source of this particular myth remains
cte
› In the United States, the highest incidence of sarcoido- unknown, it probably relates to the diversity of sarcoidosis
sis is observed in young African–American women. across various ethnic backgrounds. As an example, in a sin-
› Löfgren’s syndrome, the combination of fever, ery- gle clinic in London, West Indian black patients experienced
thema nodosum, bilateral hilar adenopathy, symmetric more advanced pulmonary disease and were more likely to
polyarthritis, and uveitis, is one mode of acute presen- develop chronic skin lesions (e.g., lupus pernio) than were
rre

tation of sarcoidosis. patients from other ethnic groups (Honeybourne 1980). In

› The Heerfordt syndrome, another acute presentation, the United States, an epidemiologic study ACCESS (A Case
consists of fever, granulomatous inflammation of the Controlled Etiologic Study of Sarcoidosis) examined more
lacrimal and parotid glands, uveitis, bilateral hilar ade- than 700 newly diagnosed sarcoidosis patients. That study
nopathy, and cranial neuropathies. found that black patients were more likely than other groups
co

› Ninety percent of patients with sarcoidosis have evi- to experience eye, splenic, and bone marrow disease
dence of pulmonary disease on chest radiography. The (Baughman et al. 2001). Caucasian patients were more likely
most common finding is hilar adenopathy. to have erythema nodosum, but blacks were more likely to
› Ocular involvement is a common complication of sar- develop other types of cutaneous sarcoidosis. Blacks were
Un

coidosis. The usual ocular manifestations are anterior also more likely to develop new organ involvement and to
uveitis and pars planitis. However, posterior uveitis experience chronic disease over the 2-year follow-up period
and optic neuritis may also occur. (Judson et al. 2003).
› Posterior uveitis can remain asymptomatic until it has Previous epidemiologic studies reported a disease inci-
reached the advanced stages. Thus, screening for oph- dence for sarcoidosis that was ten times higher among blacks
thalmologic disease is important for sarcoidosis compared with whites (Gorham et al. 2004; Keller 1971).
patients. However, these studies were based on a select group of males
› The diagnosis of sarcoidosis is established by correla- in military service. One population-based study reported the
tion of the clinical presentation, clinical course, histo- risk of sarcoidosis to be only three times higher in blacks
pathological findings, and response to therapy. compared with whites (Rybicki et al. 1997a, b).
› Glucocorticoids remain the cornerstone of therapy for
active sarcoidosis. Pearl: A positive family history of sarcoidosis makes the
diagnosis more likely.

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 409

DOI:10.1007/978-1-84800-934-9_42, © Springer Science + Business Media B.V. 2009
410 R. P. Baughman and E. E. Lower

Table 42.1 Prevalence of sarcoidosis per 100,000

>50 40–50 30–40 20–30 10–20 <10
b
Denmark Eire New York Cape Canada Algeria
Town
New Yorka Germany Holland France Brazil
Sweden Norway Japan China
Uruguay London India
Poland Spain
New Yorkc
Based on reports by Hosada (1994) and James (1992)
a
African–American

f
b
Puerto Rican
c
Caucasian

roo
Comment: No single gene mutation causes sarcoidosis, but it
is clear that a genetic predisposition exists (Rybicki and
Iannuzzi 2007). Five to 10% of Irish–American and African–
American patients in the United States report family histo-

dP
ries of individuals with the disorder (Rybicki et al. 2001b; Fig. 42.1 Patient with previously diagnosed sarcoidosis who devel-
Brennan et al. 1984). Thus, when evaluating a patient with oped acute onset of left facial weakness. Photograph shows patient
possible sarcoidosis, the presence of a positive family history closing right eye, with left eye remaining open
is supportive of the diagnosis.
In an analysis of the first- and second-degree relatives of In a retrospective study of sarcoidosis patients, the diag-
the 736 newly diagnosed cases of sarcoidosis compared with nosis of sarcoidosis was not considered in most cases when
cte
controls, the siblings of sarcoidosis patients had a relative risk facial nerve paralysis appeared before other findings (Winget
for the disease of 5.8 (Rybicki et al. 2001a, b). Caucasian et al. 1994). Moreover, patients often did not link the two
patients had a markedly higher familial relative risk compared problems even after the diagnosis of sarcoidosis. In a patient
with African–American patients (odds ratio 18.0 vs. 2.8), but suspected of having sarcoidosis, a history of facial nerve
this comparison did not reach statistical significance. palsy supports that diagnosis.
rre

For most cases of neurosarcoidosis, treatment with other

Pearl: Bell’s palsy sometimes heralds the diagnosis of
agents is often needed. These agents include methotrexate,
sarcoidosis.
cyclophosphamide, and infliximab (Lower et al. 1997;
Comment: The differential diagnosis of seventh cranial nerve Sollberger et al. 2004; Doty et al. 2003). Figure 42.2 sum-
palsy is lengthy (Fig. 42.1). Idiopathic occurrences of this marizes the clinical outcome of 71 neurosarcoidosis patients
co

problem are termed Bell’s palsy (Rahman and Sadiq 2007). from a group of 542 sarcoidosis patients followed in one sar-
Seventh cranial nerve lesions are among the most frequently coidosis clinic (Lower et al. 1997).
described abnormalities in neurosarcoidosis (Lower et al.
Pearl: The presence of bilateral hilar adenopathy in an
1997; Zajicek et al. 1999). Glucocorticoids are effective in
asymptomatic individual is likely to be sarcoidosis.
the treatment of both Bell’s palsy and sarcoid-associated sev-
Un

enth nerve lesions (Lower et al. 1997; Zajicek et al. 1999; Comment: Sarcoidosis presents as bilateral hilar and right para-
Stern 2004; Sullivan et al. 2007). tracheal adenopathy and clear lung fields in approximately one

Fig. 42.2 The clinical outcome Any neurologic

of 71 neurosarcoidosis patients Disease
treated at one institution. Of the N = 71
23 patients with seventh cranial
nerve involvement, all had
improvement whether treated 7th nerve paralysis 7th nerve paralysis Other Neurologic
with corticosteroids or not. Only No Therapy Corticosteroids Symptoms
one case had a small residual N = 10 N = 10 N = 48
defect, despite corticosteroid
therapy. Those with other forms Prednisone Methotrexate Cyclophosphamide
of neurosarcoidosis had a low N = 48 N = 28 N = 10
response to corticosteroids alone. 27% Response 63% Response 90% Response
(Adapted from Lower et al. 1997)
42 Sarcoidosis 411

f
roo
dP
Fig. 42.4 High-resolution CT scan of a sarcoidosis patient. This image
demonstrates the fine nodular opacities in a perilymphatic distribution,
which is characteristic for sarcoidosis
cte
of patients who have only hilar adenopathy on chest radio-
Fig. 42.3 Chest roentgenogram demonstrating bilateral hilar adenopa-
graphs (Koonitz et al. 1976). A transbronchoscopic or ultra-
thy due to sarcoidosis
sound-guided needle aspirate is also useful in the identification
of granulomas in patients with mediastinal adenopathy
third of patients (Fig. 42.3) (Baughman et al. 2001; Baughman (Garwood et al. 2007; Wong et al. 2007). However, granulo-
rre

et al. 2003a). This radiographic pattern is classified as Stage 1 mas can be found also in association with malignancies
in the scoring system developed by Scadding (1961). (Kennedy et al. 2008). Video-assisted mediastinoscopy has
Winterbauer et al. (1973) reported that the finding of bilateral become a rapid, safe, outpatient procedure (Karfis et al.
adenopathy in an individual who is asymptomatic is highly 2008).
predictive of sarcoidosis.
Myth: Computer tomography is not useful in assessing
co

Hilar adenopathy can also be associated with lymphoma,

sarcoidosis.
metastatic cancer, tuberculosis, and fungal infections. Symmet-
rical adenopathy is atypical for both tuberculosis and malig- Reality: The classic radiologic assessment of sarcoidosis
nancies. Morever, patients with tuberculosis and cancer are remains the routine chest roentgenogram (Hunninghake
usually symptomatic once adenopathy is evident on chest et al. 1999). However, computed tomography (CT) also has
Un

radiography (Sakowitz and Sakowitz 1977). Mediastinal and an important role (Vagal et al. 2007). Several radiologic fea-
hilar adenopathy are common among patients infected with tures have been reported in sarcoidosis patients including
HIV but usually less prominent than that associated with sar- adenopathy. The hilar lymph nodes in patients with sarcoido-
coidosis (Pastores et al. 1993). sis are usually larger than those of patients with a variety of
Pearl: The finding of an increased CD4:CD8 ratio in bron- interstitial lung diseases in whom this finding is encountered.
choalveolar lavage (BAL) fluid strongly supports the diagno- The lymph nodes larger than 2 cm in diameter are most likely
sis of sarcoidosis. caused by sarcoidosis (Niimi et al. 1996). The presence of a
fine nodularity with a perilymphatic distribution in the lung
Comment: Bronchoscopy with BAL can help confirm the parenchyma is a characteristic finding (Fig. 42.4).
diagnosis of sarcoidosis. An increase in lymphocytes, espe- Peribronchial thickening is another common CT feature
cially an increased CD4:CD8 ratio, is highly supportive of of sarcoidosis (Fig. 42.5). An upper lobe predominance of
the diagnosis of sarcoidosis (Drent et al. 2001; Welker et al. parenchymal disease is typical of sarcoidosis, but silicosis,
2004; Hunninghake et al. 1999). In addition, transbronchial hypersensitivity pneumonitis, tuberculosis, and Langerhans
biopsy reveals noncaseating granulomas in more than 50% cell lung disease also favor the upper lung zones.
412 R. P. Baughman and E. E. Lower

f
Fig. 42.7 Lupus pernio as a result of sarcoidosis

roo
This case reminds us that sarcoidosis patients may have
significant extra-pulmonary disease. Most series suggest that
5–10% of patients with sarcoidosis have no radiographic evi-
dence of lung disease (Baughman et al. 2001; Loddenkemper
et al. 1998). Such patients usually have chronic sarcoidosis

dP
of the skin, eyes, central nervous system, or heart. In some
patients, lung disease present at diagnosis and evident on
chest radiography may have resolved over time; yet, signifi-
Fig. 42.5 The image demonstrates peribronchial thickening from sar- cant disease remains in other organs (Judson et al. 2003).
coidosis (arrow)
Severe eye disease secondary to sarcoidosis can occur in
patients with normal chest roentgenograms. In a study of
Fibrotic sarcoidosis often leads to traction bronchiectasis
cte
patients with refractory ocular sarcoidosis, more than half of
in the upper lobes (Fig. 42.6a). The traction bronchiectasis of the patients had normal chest roentgenograms when their eye
sarcoidosis is created by the bronchi themselves rather than disease had worsened (Baughman et al. 2005; Bradley et al.
the interface between the lung and pleural surface. Hence, 2006). Neurosarcoidosis can exist without any evidence of
these cavities can be quite large and frequently associated active pulmonary disease (Bradley et al. 2006). Finally, a ret-
with aspergillomas or other mycetomas (Baughman et al. rospective review of 41 patients with cardiac sarcoidosis
rre

1991; Tomlinson and Sahn 1987) (Fig. 42.6b). revealed that only one third had abnormal pulmonary func-
tion studies (Chapelon-Abric et al. 2004).
Myth: If the patient has a normal chest roentgenogram, she
cannot have active sarcoidosis. Pearl: Initial disease manifestations can predict chronic
sarcoidosis.
Reality: This statement was made to the patient in Fig. 42.7,
co

who had active cutaneous sarcoidosis (lupus pernio). The Comment: Several sarcoidosis phenotypes are associated
patient had a normal chest radiograph, but she coughed fre- with disease resolution, whereas others portend chronic dis-
quently and remained dyspneic due to sarcoidosis in her ease (Table 42.2) (Neville et al. 1983; Rizzato et al. 1995).
sinuses. Significant sinus disease is commonly associated Among the predictive features of chronic disease are lupus
Un

with lupus pernio (Baughman et al. 2002). pernio (see Fig. 42.7) and fibrotic lung disease (see Fig. 42.6).

a b

Fig. 42.6 (a): Traction

bronchiectasis due to upper lobe
involvement from sarcoidosis
(circle). (b): Image higher in
chest demonstrates one of the
cavities is opacified by a
mycetoma (arrow)
42 Sarcoidosis 413

Table 42.2 Features predictive of prognosis in sarcoidosis been expanded to include periarthritis (Grunewald and
Resolution within 2 years Persistent disease beyond 2 years Eklund 2007).
Erythema nodosum Lupus pernio Several additional groups have also noted a high rate of
Hilar adenopathy Upper lobe traction bronchiectasis disease resolution with erythema nodosum and other signs of
Cranial seventh nerve paralysis Central nervous system sarcoidosis (Neville et al. 1983; Mana and Marcoval 2007).
sarcoidosis
However, 10–20% of patients continue to have active disease
Anterior uveitis Posterior uveitis
Cardiac disease
at 2 years, and some patients with Löfgren’s syndrome
Nephrolithiasis develop cardiac and neurologic disease.
A group from Sweden published a 2-year outcomes study
of patients who presented with Löfgren’s syndrome

f
Lupus pernio tends to be a chronic skin lesion that is unlikely (Grunewald and Eklund 2007). Disease resolution was expe-
to spontaneously resolve over time (Baughman et al. 2008). rienced within 2 years of diagnosis by 85% of their patients.

roo
Nonfacial skin lesions can also be chronic but seldom com- These investigators confirmed a strong association between
prise a significant cause of morbidity because they can be Löfgren’s syndrome and human leukocyte antigen (HLA)
treated with topical glucocorticoids (Marchell, Judson DRB1*0301/DQB1*0201(Sato et al. 2002). In their study,
2007). only one of 85 patients with the DRB1*0301/DQB1*0201
Some data indicate that glucocorticoid treatment is asso- allele developed persistent disease. Of the 40 patients nega-
ciated with the development of chronic disease (Fig. 42.8) tive for DRB1*0301/DQB1*0201 allele, only 22 patients

dP
(Gottlieb et al. 1997; Baughman et al. 2006a, c). However, experienced disease resolution within 2 years. Additional
this purported relationship may represent “confounding by investigations are needed to determine whether this Myth
indication”: the need for systemic therapy may be a surro- can be amended to illustrate a Pearl; namely, that HLA-
gate marker for patients who are inherently sicker and there- DRB1*0301/DQB1*0201 is almost always associated with a
fore more likely to require prolonged treatment. good prognosis.
cte
Myth: Löfgren’s syndrome is always associated with a good Pearl: Fatigue is a common and treatable symptom of
prognosis. sarcoidosis.
Reality: Professor Sven Löfgren originally reported that most Comment: Fatigue is a symptom that frequently afflicts sar-
sarcoidosis patients with erythema nodosum (Fig. 42.9) and coidosis patients of all ethnic groups and geographic loca-
hilar adenopathy experience spontaneous remission. This tions (de Vries et al. 2004a, b; Baughman et al. 2007). The
rre

report contained a truly untreated population since his obser- etiology of fatigue in sarcoidosis is likely multifactorial, but
vations preceded the availability of glucocorticoids. Since one potential explanation is the excessive levels of a variety
the original report, the definition of Löfgren’s syndrome has of cytokines released by patients with active disease. These
co
Un

Fig. 42.8 The percentage of patients on systemic therapy for two dif-
ferent studies: one performed in Philadelphia (Gottlieb et al. 1997) and
the other a ten-center study done across the United States (ACCESS)
(Baughman et al. 2006c). The outcome of patients after at least 2 years
is also shown. For those patients given initial therapy, a higher percent-
age required chronic therapy Fig. 42.9 Erythema nodosum on the leg of a sarcoidosis patient
414 R. P. Baughman and E. E. Lower

40 Alkaline Phosphatase Asparate transaminase

400 120
35
30 350
100
FAS Score

25 300
20 80
250
15

IU/L

IU/L
200 60
10
150
5 40
0 100
Baseline 1 2 3 4 5 6 7 8 20
50
Weeks of Therapy
0 0

f
Granulomas No Granulomas Granulomas No Granulomas
Placebo d-MPH

roo
Fig. 42.11 The serum levels of alkaline phophatase and aspirate
Fig. 42.10 The weekly Fatigue Assessment Score (FAS) during 8 weeks
transaminase before liver biopsy in 84 sarcoidosis patients. The upper
of treatment with either placebo or d-methylphenidate of ten patients
limit of normal for each test is indicated by the heavy dashed line. As a
with sarcoidosis associated fatigue. The higher the FAS, the more the
group, patients with granulomas in their liver biopsy had higher liver
fatigue is. There was a significant improvement in the FAS score during
function test results. However, there was significant overlap. (Adapted
treatment with d-MPH (p < 0.02). Adapted from (Lower et al. 2008)
from Baughman et al. 2003b)

dP
cytokines include tumor necrosis factor (TNF), gamma inter-
feron, and interleukin-2 (Baughman et al. 2003a).
The Fatigue Assessment Scale (FAS) was developed to
measure fatigue in sarcoidosis (de Vries et al. 2004a). For
this 50-point scale, the higher the number, the more fatigue
experienced. Patients with a score of 22 or greater are con-
cte
sidered fatigued. Depression contributes to fatigue in some
patients (de Vries et al. 2004b). The level of fatigue has also
been associated with pulmonary function: more severe
fatigue correlates with shorter 6-min walk distance and lower
forced vital capacity (Baughman et al. 2007).
rre

Effective treatment of sarcoidosis may relieve fatigue

(Fouchier and Moller 2004). However, many patients con-
tinue to complain of fatigue, despite using systemic therapy
for their sarcoidosis (de Vries et al. 2004b). Methylphenidate
may be useful in treating sarcoidosis-related fatigue
co

(Fig. 42.10) (Wagner et al. 2005; Lower et al. 2008).

Pearl: Liver involvement is common but usually asymptom-
atic in sarcoidosis patients. Fig. 42.12 CT scan with contrast of patient with extensive sarcoidosis,
including lung, brain, and liver involvement. Massive hepatomegaly
Comment: Noncaseating granulomas compatible with sarcoi-
Un

led to early satiety and nausea. Hypodense nodules are seen throughout
dosis frequently affect the liver (Rose et al. 2008; Warshauer the liver
et al. 1994; Klatskin and Yesner 1950). Serum liver function
tests are less sensitive than liver biopsy. The results of blood
Myth: Cardiac disease in sarcoidosis is totally unpredictable.
tests do not predict the severity of hepatic disease. Fig. 42.11
demonstrates the levels of alkaline phosphatase and aspirate Reality: Cardiac involvement by sarcoidosis is a well-recog-
transaminase from 84 sarcoidosis patients, compared to the nized cause of sudden death (Baughman et al. 1997). Congestive
findings of their liver biopsies (Baughman et al. 2003b). heart failure due to cardiomyopathy can be a significant prob-
Sarcoidosis patients with liver disease can have symptom- lem, but death is usually due to arrythmias (Chapelon-Abric
atic discomfort from hepatomegaly (Fig. 42.12) or splenom- et al. 2004). Ventricular tachycardia is the most life-threatening
egaly and pancytopenia related to splenic sequestration. arrhythmia in this disorder. Electrocardiography (ECG), 24-h
Most patients with sarcoid-related liver disease have an Holter monitoring, and echocardiography are used as screen-
asymptomatic condition that never requires therapy (Judson ing procedures for cardiac disease (Judson 2005). However,
2002). However, a small minority of patients develops cir- patients with normal baseline ECGs can develop significant
rhosis and portal hypertension. arrhythmias (Fig. 42.13).
42 Sarcoidosis 415

f
roo
Fig. 42.14 The prevalence of history of cardiac symptoms or abnormal
electrocardiography (ECG), 24 Holter monitoring (Holter), or echocar-
diography (Echo) in 62 sarcoidosis patients. One third of these patients
were subsequently determined to the have cardiac sarcoidosis. The sen-
sitivity and specificity of these individual tests are shown. The presence

dP
of one or more abnormal test had 100% sensitivity, but only 87% speci-
Fig. 42.13 The ECG monitoring a sarcoidosis patient with chest flut- ficity. Presence of three abnormalities was 100% specific but only 4%
tering. The top panel is normal sinus rhythm. The lower two panels sensitive. No patient had all four abnormalities. Adapted from (Mehta
demonstrate nonsustained ventricular tachycardia et al. 2008)

multiple sclerosis can be difficult (Lower and Weiss 2008).

cte
Twenty-four-hour Holter monitoring is more sensitive Although roentgenographic imaging may be useful, early
and specific than other forms of testing (Suzuki et al. 1994). lesions are nonspecific for both diseases. Disease beyond the
Reports suggest that electrophysiology testing is not as reli- central nervous system and eyes does not occur in multiple
able as continuous monitoring in detecting significant car- sclerosis, so the identification of pulmonary, cutaneous, or
diac sarcoidosis (Mezaki et al. 2001). Magnetic resonance hepatic disease supports the diagnosis of sarcoidosis.
rre

imaging (MRI) and positive emission tomography (PET)

Myth: A normal angiotensin converting enzyme (ACE) level
have emerging roles in the evaluation of cardiac sarcoidosis
excludes active sarcoidosis.
and are used routinely now in the clinical assessment of this
disease complication (Skold et al. 2002; Ohira et al. 2008). Reality: One of the greatest Myths surrounding sarcoidosis
Findings from MRI and PET may complement each other is that serum levels of ACE are useful in diagnosing the dis-
co

(Fig. 42.14). ease and predicting clinical outcomes (Lieberman et al. 1979;
DeRemee and Rohrbach 1980; Buaghman et al. 1983;
Pearl: Pars planitis and optic neuritis are features shared by
Gronhagen-Riska et al. 1980). Several factors limit the util-
both multiple sclerosis and sarcoidosis.
ity of ACE level in monitoring sarcoidosis.
Un

Comment: Ocular disease is a common manifestation of sar- First, the test is not elevated in all patients with sarcoido-
coidosis. Eye findings are often helpful in differentiating sar- sis. Up to 40% of patients with sarcoidosis have normal val-
coidosis from conditions such as lymphoma or tuberculosis ues at the time of diagnosis (Loddenkemper et al. 1998;
(Ohara et al. 2005). The eye is divided into three chambers— Lieberman 1975; Pietinalho et al. 2000). ACE levels are
anterior, posterior, and intermediate—based on blood flow influenced by polymorphisms of the ACE gene. The inser-
(Bradley et al. 2002). Intermediate uveitis occurs in the tion (I) and deletion (D) polymorphisms lead to three geno-
watershed area between the anterior and posterior circula- types. The highest ACE levels in normal controls are seen
tion. Sarcoidosis and multiple sclerosis are commonly asso- with those with the DD genotype (Tomita et al. 1997).
ciated with inflammation in this pars planis area (Bradley Second, glucocorticoids suppress ACE levels (Baughman
et al. 2002; Biousse et al. 1999; Prieto et al. 2001). et al. 1983; Gronhagen-Riska et al. 1980). Glucocorticoid
Multiple sclerosis is frequently associated with ocular withdrawal can lead to the elevation of ACE levels, which
disease (Chen and Gordon 2005). In addition to pars planitis, does not correlate with a worsening of disease. Figure 42.15
either multiple sclerosis or sarcoidosis can cause optic neuri- demonstrates the change in ACE levels measured serially in
tis. For the patient who presents with both neurologic and sarcoidosis patients followed in one clinic (Baughman et al.
ocular diseases, the differentiation between sarcoidosis and 1983). No relationship was identified between subsequent
416 R. P. Baughman and E. E. Lower

patients followed at one institution. In that series, serum ACE

levels were elevated in seven of the 11 patients (64%) in
whom they were measured (Mechanic et al. 1997).
The best treatment for CVID remains unclear. In some
patients, the granulomas resolve with immunoglobulin
replacement therapy (Pujol et al. 1999). However, many
patients do not respond to intravenous immunoglobulin infu-
sions alone (Mechanic et al. 1997). For refractory cases, suc-
cessful treatment has been reported with agents used to treat
sarcoidosis. In our experience, agents such as azathioprine

f
and methotrexate can be beneficial. There have also been
reports of successful treatment approaches that employed a

roo
Fig. 42.15 Changes in ACE level over time in sarcoidosis patients fol-
targeted inhibitor of TNF (Thatayatikom et al. 2005; Smith,
lowed at one clinic. During the time of observation, prednisone dose
was adjusted. There was no relation between the change in ACE level Skelton 2001).
and the change in clinical condition of the patient (a). However, there
was a significant negative correlation between the change in prednisone Pearl: “Irreversible” pulmonary sarcoidosis may still be
dose and change in ACE level (R = −0.76, p < 0.0001) (b). Adapted reversible.
from (Baughman et al. 1983)
Comment: Chronic pulmonary sarcoidosis can lead to pul-

dP
monary fibrosis. The finding of pulmonary fibrosis corre-
clinical course and ACE level. However, a significant inverse sponds to Stage 4 of the classification system devised by
correlation existed between change in prednisone dose and Scadding (Scadding 1961). The BAL cell population of
ACE level. patients with active pulmonary sarcoidosis in Stages 1–3
Pearl: Hypogammaglobulinemia can cause a granulomatous usually shows increased numbers of lymphocytes with an
disease similar to sarcoidosis. increased CD4:CD8 ratio (Baughman and Drent 2001). In
cte
contrast, BAL studies from patients with Stage 4 disease
Comment: Sarcoidosis can be associated with a polyclonal demonstrate elevated neutrophil counts and levels of inter-
hypergammaglobulinemia (Baughman and Hurtubise 1985), leukin-8 (Drent et al. 1999; Baughman et al. 1994). These
but many patients have normal immunoglobulin levels. latter findings correlate with pulmonary fibrosis (Baughman
Glucocorticoids reduce serum levels of immunoglobulin. and Drent 2001; Lynch et al. 1992).
rre

Therefore, a low immunoglobulin level in a patient with sar- The finding of fibrosis on chest roentgenogram or CT scan
coidosis may be a reflection of treatment. is considered irreversible. However, even lungs explanted at
Approximately 10% of patients with common variable lung transplantation from sarcoidosis patients often still dem-
immunodeficiency (CVID) experience a granulomatous dis- onstrate some granulomatous inflammation (Padilla et al.
ease that mimics sarcoidosis (Mechanic et al. 1997; Park 1997), indicative of disease that retains some potential respon-
co

et al. 2005). Bronchiectasis is encountered frequently in siveness to treatment. The insensitivity of our current meth-
CVID patients because of their increased rate of bacterial ods for detecting reversible inflammation within fibrotic areas
infections. The radiographic features of CVID include pul- underscores the importance of considering a course of treat-
monary nodules and increased reticular markings (Park et al. ment in any sarcoidosis patient with lung disease, no matter
2005). In addition, the granulomas that occur in CVID can
Un

how severe the degree of pulmonary fibrosis.

affect not just the lung but multiple organs. Table 42.3 shows In a single-center study of nearly 500 sarcoidosis patients,
the granulomatous organ involvement documented in 17 nine patients had at least one vital capacity measurement of
less than one liter. The chest roentgenogram findings and low
Table 42.3 Organ with granulomas identified by biopsy among 17 vital capacities in these patients were compatible with “end-
patients with common variable immunodeficiency syndrome stage” lung disease. However, five of those patients (56%)
Organ Number of patients with responded substantially to treatment (Baughman et al. 1997).
documented involvement
Figure 42.16 reveals the significant improvement experi-
Thorax 8 enced by one patient who had been felt to have irreversible
Liver 7
pulmonary fibrosis (Baughman et al. 2006b).
Extra thoracic lymph nodes 6
Spleen 1 Myth: Sarcoidosis patients with interstitial lung disease
Bone marrow 1 should always be treated.
Spleen 1
(Adapted from Mechanic et al. 1997). Patients could have more than Reality: This Myth has led to the treatment of many sarcoi-
one organ with documented granulomatous disease. dosis patients who have only an abnormal imaging study of
42 Sarcoidosis 417

Fig. 42.16 The CT scan of

patient with advanced pulmonary
sarcoidosis. The image before
treatment was felt to show
fibrotic changes. The follow-up
image was obtained at the same
level after 2 months of treatment
with infliximab

f
roo
dP
cte
rre

Fig. 42.17 Two levels of the chest

of a sarcoidosis patient with no
pulmonary symptoms but an
abnormal chest roentgenogram. As
the prednisone dosage is reduced,
the repeat CT scan demonstrates
co

increased nodularity and confluent

infiltrates in the parenchyma.
However, the patient had no change
in pulmonary function or respiratory
symptoms. The prednisone was
subsequently withdrawn with no
Un

deleterious effect to the patient 20 mg prednisone 10 mg prednisone

the chest. Figure 42.17 demonstrates the CT scans from a disease (Gibson et al. 1996). Only 22% of the patients were
sarcoidosis patient who was treated initially with 20 mg/day symptomatic enough to require therapy in the first 6 months.
of prednisone, followed by a reduction to 10 mg/day for 3 A high percentage of patients experienced spontaneous
months. Because the patient had no pulmonary symptoms, remission during the 6 months of observation. However,
the prednisone therapy was tapered slowly and discontinued patients who failed to improve or became symptomatic were
over an additional 6 months. The patient has remained off all randomized subsequently to receive either 18 months of sys-
glucocorticoids for 2 years with no evidence of pulmonary temic glucocorticoids or observation. Those treated with glu-
symptoms, despite the persistence of radiologic evidence of cocorticoids experienced a mild but significant improvement
interstitial lung disease. in lung function. Twenty percent of the observation group
Figure 42.18 demonstrates the clinical outcome of 149 eventually required systemic therapy for their pulmonary
sarcoidosis patients who presented with either stage 2 or 3 symptoms, but the remaining 80% treated with observation
418 R. P. Baughman and E. E. Lower

Fig. 42.18 The clinical outcome Initial Evaluation

of 149 sarcoidosis patients who N = 149 patients
presented with either Stage 2 or 3
disease (Gibson et al. 1996).
Only 22% of the patients were Observe for
symptomatic enough to require Six Months
therapy in the first 6 months. A
high percentage of patients Symptomatic Disease Persistent Asymptomatic Disease Radiographic Improvement
experienced spontaneous Treated with Corticosteroids No treatment considered
remission during the 6 months of N = 33 N = 58 N = 58
observation
Randomize

f
roo
Table 42.4 Incidence of pulmonary hypertension in sarcoidosis patients
First author Handa et al. Bourbonnais and Rizzato et al. Sulica et al. Baughman et al. Shorr et al. (2005)
(2006) Samavati (2008) (1983) (2005) (2006a, b, c)
Source of patients Clinic Clinic Clinic Clinic Clinic UNOS
Entry All patients All patients Dyspnea Dyspnea Dyspnea Transplant evaluation
Study to detect pulmonary Echo Echo/Cath Cath Echo Cath Cath

dP
hypertension
Number studied 212 162 62 106 53 363
% with pulmonary hypertension 6 15 55 52 60 72
UNOS United Network of Organ Sharing ; Echo echocardiography; Cath right heart catherterization
cte
alone never developed pulmonary symptoms. Another study References
demonstrated that sarcoidosis patients who were asymptom-
atic at diagnosis had only a 10% chance of developing symp- Baughman RP, Drent M, Kavuru M, et al Infliximab therapy in patients
tomatic disease during observation (Pietinalho et al. 1999). with chronic sarcoidosis and pulmonary involvement. Am J Respir
These studies support the principal of treating only symp- Crit Care Med. 2006a;174(7):795–802
rre

Baughman RP, Drent M. Role of bronchoalveolar lavage in interstitial

tomatic patients (Hunninghake et al. 1999).
lung disease. Clin Chest Med. 2001;22(2):331–41
Pearl: In persistently dyspneic sarcoidosis patient, think of Baughman RP, du Bois RM, Lower EE. Sarcoidosis. Lancet 2003a;
361:1111–8
pulmonary hypertension. Baughman RP, Engel PJ, Meyer CA, et al Pulmonary hypertension in
sarcoidosis. Sarcoidosis Vasc Diffuse Lung Dis. 2006b;23:108–16
Comment: Interstitial lung disease of sarcoidosis often leads
Baughman RP, Hurtubise PE. Systemic immune response of patients
co

to dyspnea, but sarcoidosis also affects the pulmonary vascu- with active pulmonary sarcoidosis. Clin Exp Immunol. 1985;
lature and can cause pulmonary hypertension (Hunninghake 61:535–41
et al. 1994). Table 42.4 summarizes several studies of pulmo- Baughman RP, Judson MA, Teirstein A, et al Chronic facial sarcoidosis
including lupus pernio: clinical description and proposed scoring
nary hypertension in sarcoidosis. The incidence of pulmo-
systems. Am J Clin Dermatol. 2008;9(3):155–61
Un

nary hypertension in unselected patients appears to be 5–15% Baughman RP, Judson MA, Teirstein A, et al Presenting characteristics
(Rizzato et al. 1983; Sulica et al. 2005; Baughman et al. as predictors of duration of treatment in sarcoidosis. QJM.
2006b). However, more than 70% of sarcoidosis patients 2006c;99(5):307–15
Baughman RP, Judson MA, Teirstein AS, et al Thalidomide for chronic
listed for lung transplant have pulmonary hypertension.
sarcoidosis. Chest 2002;122:227–32
Clinical features that suggest pulmonary hypertension are a Baughman RP, Keeton D, Lower EE. Relationship between interleu-
Stage 4 chest roentgenogram (Rizzato et al. 1983; Sulica kin-8 and neutrophils in the BAL fluid of sarcoidosis. Sarcoidosis
et al. 2005), diminished vital capacity (Rizzato et al. 1983; 1994;11:S217–20S
Baughman RP, Koehler A, Bejarano PA, et al Role of liver function tests
Sulica et al. 2005; Baughman et al. 2006b), reduced 6-min
in detecting methotrexate-induced liver damage in sarcoidosis. Arch
walk distance (Baughman et al. 2007; Bourbonnais and Intern Med. 2003b;163(5):615–20
Samavati 2008), and desaturation with exercise (Bourbonnais Baughman RP, Lower EE, Bradley DA, et al Etanercept for refractory
and Samavati 2008). Pulmonary hypertension in sarcoidosis ocular sarcoidosis: results of a double-blind randomized trial. Chest
patients can improve with treatment with a variety of pulmo- 2005;128(2):1062–67
Baughman RP, Ploysongsang Y, Roberts RD, et al Effects of sarcoid
nary vasodilators (Baughman et al. 2006b; Fisher et al. 2006; and steroids on angiotensin-converting enzyme. Am Rev Respir
Milman et al. 2008). Dis. 1983;128:631–3
42 Sarcoidosis 419

Baughman RP, Shipley RT, Loudon RG, et al Crackles in interstitial Handa T, Nagai S, Miki S, et al Incidence of pulmonary hypertension
lung disease. Comparison of sarcoidosis and fibrosing alveolitis. and its clinical relevance in patients with sarcoidosis. Chest
Chest 1991;100:96–101 2006;129(5):1246–52
Baughman RP, Sparkman BK, Lower EE. Six-minute walk test and Honeybourne D. Ethnic differences in the clinical features of sarcoido-
health status assessment in sarcoidosis. Chest 2007;132(1):207–13 sis in South-East London. Br J Dis Chest. 1980;74:63–9
Baughman RP, Teirstein AS, Judson MA, et al Clinical characteristics Hosada Y. Consensus conference: epidemiology of sarcoidosis.
of patients in a case control study of sarcoidosis. Am J Respir Crit Sarcoidosis 1994;11:17–21
Care Med. 2001;164:1885–9 Hunninghake GW, Costabel U, Ando M, et al ATS/ERS/WASOG state-
Baughman RP, Winget DB, Bowen EH, et al Predicting respiratory fail- ment on sarcoidosis. American Thoracic Society/European Respiratory
ure in sarcoidosis patients. Sarcoidosis 1997;14:154–8 Society/World Association of Sarcoidosis and other Granulomatous
Biousse V, Trichet C, Bloch-Michel E, et al Multiple sclerosis associ- Disorders. Sarcoidosis Vasc Diffuse Lung Dis. 1999;16:149–73
ated with uveitis in two large clinic-based series. Neurology Hunninghake GW, Gilbert S, Pueringer R, et al Outcome of the treatment

f
1999;52(1):179–81 for sarcoidosis. Am J Respir Crit Care Med. 1994;149(4 Pt 1):893–8
Bourbonnais JM, Samavati L. Clinical predictors of pulmonary hyper- James DG. Epidemiology of sarcoidosis. Sarcoidosis 1992;9:79–87

roo
tension in sarcoidosis. Eur Respir J. 2008;32(2):296–302 Judson MA, Baughman RP, Thompson BW, et al Two year prognosis of
Bradley DA, Baughman RP, Raymond L, et al Ocular manifestations of sarcoidosis: the ACCESS experience. Sarcoidosis Vasc Diffuse
sarcoidosis. Sem Resp Crit Care Med. 2002;23:543–8 Lung Dis. 2003;20(3):204–11
Bradley DA, Lower EE, Baughman RP. Diagnosis and management of Judson MA. Hepatc, splenic, and gastrointestinal involvement with sar-
spinal cord sarcoidosis. Sarcoidosis Vasc Diffuse Lung Dis. 2006; coidosis. Sem Resp Crit Care Med. 2002;23:529–43
23(1):58–65 Karfis EA, Roustanis E, Beis J, et al Video-assisted cervical mediasti-
Brennan NJ, Crean P, Long JP, et al High prevalence of familial sarcoi- noscopy: our seven-year experience. Interact Cardiovasc Thorac
dosis in an Irish population. Thorax 1984;39:14–8 Surg. 2008;7:1015–8

dP
Chapelon-Abric C, de ZD, Duhaut P, et al Cardiac sarcoidosis: a retro- Keller AZ. Hospital, age, racial, occupational, geographical, clinical
spective study of 41 cases. Medicine (Baltimore) 2004;83(6):315–34 and survivorship characteristics in the epidemiology of sarcoidosis.
Chen L, Gordon LK. Ocular manifestations of multiple sclerosis. Curr Am J Epidemiol. 1971;94:222–30
Opin Ophthalmol. 2005;16(5):315–20 Kennedy MP, Jimenez CA, Mhatre AD, et al Clinical implications of
de Vries J, Michielsen H, van Heck GL, et al Measuring fatigue in sar- granulomatous inflammation detected by endobronchial ultrasound
coidosis: the fatigue assessment scale (FAS). Br J Health Psychol. transbronchial needle aspiration in patients with suspected cancer
2004a;9(Pt 3):279–91 recurrence in the mediastinum. J Cardiothorac Surg. 2008;3(8):8
de Vries J, Rothkrantz-Kos S, Dieijen-Visser MP, et al The relationship Klatskin G, Yesner R. Hepatic manifestations of sarcoidosis and other
cte
between fatigue and clinical parameters in pulmonary sarcoidosis. granulomatous diseases. Yale J Biol Med. 1950;51:207–48
Sarcoidosis Vasc Diffuse Lung Dis. 2004b;21(2):127–36 Koonitz CH, Joyner LR, Nelson RA. Transbronchial lung biopsy via
DeRemee RA, Rohrbach MS. Serum angiotensin-converting enzyme the fiberoptic bronchoscope in sarcoidosis. Ann Intern Med. 1976;
activity in evaluating the clinical course of sarcoidosis. Ann Intern 85(1):64–6
Med. 1980;92(3):361–5 Lieberman J, Nosal A, Schlessner A, et al Serum angiotensin-convert-
Doty JD, Mazur JE, Judson MA. Treatment of corticosteroid-resistant ing enzyme for diagnosis and therapeutic evaluation of sarcoidosis.
rre

neurosarcoidosis with a short-course cyclophosphamide regimen. Am Rev Respir Dis. 1979;120(2):329–35

Chest 2003;124(5):2023–6 Lieberman J. Elevation of serum angiotensin converting enzyme (ACE)
Drent M, Jacobs JA, Cobben NA, et al Computer program supporting level in sarcoidosis. Am J Med. 1975;59:365–72
the diagnostic accuracy of cellular BALF analysis: a new release. Loddenkemper R, Kloppenborg A, Schoenfeld N, et al Clinical findings
Respir Med. 2001;95(10):781–6 in 715 patients with newly detected pulmonary sarcoidosis – results
Drent M, Jacobs JA, de Vries J, et al Does the cellular bronchoalveolar of a cooperative study in former West Germany and Switzerland.
lavage fluid profile reflect the severity of sarcoidosis? Eur Respir J. WATL Study Group. Wissenschaftliche Arbeitsgemeinschaft fur die
co

1999;13(6):1338–44 Therapie von Lungenkrankheitan. Sarcoidosis Vasc Diffuse Lung

Fisher KA, Serlin DM, Wilson KC, et al Sarcoidosis-associated pulmo- Dis. 1998;15(2):178–82
nary hypertension: outcome with long-term epoprostenol treatment. Lower EE, Broderick JP, Brott TG, et al Diagnosis and management of
Chest 2006;130(5):1481–8 neurologic sarcoidosis. Arch Intern Med. 1997;157:1864–8
Fouchier SM, Moller GM, Van Santen-Hoeufft M, et al [Successful Lower EE, Harman S, Baughman RP. Double-blind, randomized trial of
Un

treatment with infliximab of a patient with refractory sarcoidosis]. dexmethylphenidate hydrochloride for the treatment of sarcoidosis-
Ned Tijdschr Geneeskd. 2004;148(49):2446–50 associated fatigue. Chest 2008;133(5):1189–95
Garwood S, Judson MA, Silvestri G, et al Endobronchial ultrasound for the Lower EE, Weiss KL. Neurosarcoidosis. Clin Chest Med. 2008;
diagnosis of pulmonary sarcoidosis. Chest 2007;132(4):1298–304 29(3):475–92
Gibson GJ, Prescott RJ, Muers MF, et al British Thoracic Society Lynch JP, Standiford TJ, Rolfe MW, et al Neutrophilic alveolitis in idio-
Sarcoidosis study: effects of long term corticosteroid treatment. pathic pulmonary fibrosis. The role of interleukin-8. Am Rev Respir
Thorax 1996;51(3):238–47 Dis. 1992;145(6):1433–9
Gorham ED, Garland CF, Garland FC, et al Trends and occupational Mana J, Marcoval J. Erythema nodosum. Clin Dermatol. 2007;25(3):
associations in incidence of hospitalized pulmonary sarcoidosis and 288–94
other lung diseases in Navy personnel: a 27-year historical prospec- Marchell RM, Judson MA. Chronic cutaneous lesions of sarcoidosis.
tive study, 1975–2001. Chest 2004;126(5):1431–8 Clin Dermatol. 2007;25(3):295–302
Gottlieb JE, Israel HL, Steiner RM, et al Outcome in sarcoidosis. The Mechanic LJ, Dikman S, Cunningham RC. Granulomatous disease in
relationship of relapse to corticosteroid therapy. Chest 1997;111(3): common variable immunodeficiency. Ann Intern Med. 1997;127(8
623–31 Pt 1):613–7
Gronhagen-Riska C, Selroos O, Niemisto M. Angiotensin converting Mehta D, Lubitz SA, Frankel Z, et al Cardiac involvement in patients
enzyme. V. Serum levels as monitors of disease activity in corticos- with sarcoidosis: diagnostic and prognostic value of outpatient test-
teroid-treated sarcoidosis. Eur J Respir Dis. 1980;61(2):113–22 ing. Chest 2008;133(6):1426–35
Grunewald J, Eklund A. Sex-specific manifestations of Löfgren’s Mezaki T, Chinushi M, Washizuka T, et al Discrepancy between induc-
syndrome. Am J Respir Crit Care Med. 2007;175(1):40–4 ibility of ventricular tachycardia and activity of cardiac sarcoidosis.
420 R. P. Baughman and E. E. Lower

Requirement of defibrillator implantation for the inactive stage of Sakowitz AJ, Sakowitz BH. Bilateral hilar lymphadenopathy: an
cardiac sarcoidosis. Intern Med. 2001;40(8):731–5 uncommon manifestation of adult tuberculosis. Chest 1977;
Milman N, Burton CM, Iversen M, et al Pulmonary hypertension in 71(3):421–3
end-stage pulmonary sarcoidosis: therapeutic effect of sildenafil? Sato H, Grutters JC, Pantelidis P, et al HLA-DQB1*0201: a marker for
J Heart Lung Transplant. 2008;27(3):329–34 good prognosis in British and Dutch patients with sarcoidosis. Am J
Neville E, Walker AN, James DG. Prognostic factors predicting the out- Respir Cell Mol Biol. 2002;27(4):406–12
come of sarcoidosis: an analysis of 818 patients. Q J Med. 1983; Scadding JG. Prognosis of intrathoracic sarcoidosis in England. Br
208:525–33 Med J. 1961;4:1165–72
Niimi H, Kang EY, Kwong JS, et al CT of chronic infiltrative lung dis- Shorr AF, Helman DL, Davies DB, et al Pulmonary hypertension in
ease: prevalence of mediastinal lymphadenopathy. J Comput Assist advanced sarcoidosis: epidemiology and clinical characteristics.
Tomogr. 1996;20(2):305–8 Eur Respir J. 2005;25(5):783–8
Ohara K, Judson MA, Baughman RP. Clinical aspects of ocular sarcoi- Skold CM, Larsen FF, Rasmussen E, et al Determination of car-

f
dosis. In: Drent M, Costabel U, editors. Sarcoidosis. Wakefiled, UK: diac involvement in sarcoidosis by magnetic resonance imag-
The Charlesworth Group; 2005. p.188–209 ing and Doppler echocardiography. J Intern Med. 2002;252(5):

roo
Ohira H, Tsujino I, Ishimaru S, et al Myocardial imaging with 465–71
18F-fluoro-2-deoxyglucose positron emission tomography and Smith KJ, Skelton H. Common variable immunodeficiency treated with
magnetic resonance imaging in sarcoidosis. Eur J Nucl Med Mol a recombinant human IgG, tumour necrosis factor-alpha receptor
Imaging. 2008;35(5):933–41 fusion protein. Br J Dermatol. 2001;144(3):597–600
Padilla ML, Schilero GJ, Teirstein AS. Sarcoidosis and transplantation. Sollberger M, Fluri F, Baumann T, et al Successful treatment of steroid-
Sarcoidosis 1997;14:16–22 refractory neurosarcoidosis with infliximab. J Neurol. 2004;
Park JE, Beal I, Dilworth JP, et al The HRCT appearances of granu- 251(6):760–1
lomatous pulmonary disease in common variable immune defi- Sulica R, Teirstein AS, Kakarla S, et al Distinctive clinical, radio-

dP
ciency. Eur J Radiol. 2005;54(3):359–64 graphic, and functional characteristics of patients with sarcoidosis-
Pastores SM, Naidich DP, Aranda CP, et al Intrathoracic adenopathy related pulmonary hypertension. Chest 2005;128(3):1483–9
associated with pulmonary tuberculosis in patients with human Sullivan FM, Swan IR, Donnan PT, et al Early treatment with predniso-
immunodeficiency virus infection. Chest 1993;103(5):1433–7 lone or acyclovir in Bell’s palsy. N Engl J Med. 2007;357(16):
Pietinalho A, Ohmichi M, Lofroos AB, et al The prognosis of sarcoido- 1598–607
sis in Finland and Hokkaido, Japan. A comparative five-year study Suzuki T, Kanda T, Kubota S, et al Holter monitoring as a noninvasive
of biopsy-proven cases. Sarcoidosis Vasc Diffuse Lung Dis. 2000; indicator of cardiac involvement in sarcoidosis. Chest 1994;106(4):
17:158–66 1021–4
cte
Pietinalho A, Tukiainen P, Haahtela T, et al Oral prednisolone followed Thatayatikom A, Thatayatikom S, White AJ. Infliximab treatment for
by inhaled budesonide in newly diagnosed pulmonary sarcoidosis: a severe granulomatous disease in common variable immunodefi-
double-blind, placebo-controlled, multicenter study. Chest 1999; ciency: a case report and review of the literature. Ann Allergy
116:424–31 Asthma Immunol. 2005;95(3):293–300
Prieto JF, Dios E, Gutierrez JM, et al Pars planitis: epidemiology, treat- Tomita H, Ina Y, Sugiura Y, et al Polymorphism in the angiotensin-
ment, and association with multiple sclerosis. Ocul Immunol converting enzyme (ACE) gene and sarcoidosis. Am J Respir Crit
rre

Inflamm. 2001;9(2):93–102 Care Med. 1997;156(1):255–9

Pujol RM, Nadal C, Taberner R, et al Cutaneous granulomatous lesions Tomlinson JR, Sahn SA. Aspergilloma in sarcoid and tuberculosis.
in common variable immunodeficiency: complete resolution after Chest 1987;92(3):505–8
intravenous immunoglobulins. Dermatology 1999;198(2):156–8 Vagal AS, Shipley R, Meyer CA. Radiological manifestations of sarcoi-
Rahman I, Sadiq SA. Ophthalmic management of facial nerve palsy: a dosis. Clin Dermatol. 2007;25(3):312–25
review. Surv Ophthalmol. 2007;52(2):121–44 Wagner MT, Marion SD, Judson MA. The effects of fatigue and treat-
Rizzato G, Fraioli P, Montemurro L. Nephrolithiasis as a presenting ment with methylphenidate on sustained attention in sarcoidosis.
co

feature of chronic sarcoidosis. Thorax 1995;50(5):555–9 Sarcoidosis Vasc Diffuse Lung Dis. 2005;22(3):235
Rizzato G, Pezzano A, Sala G, et al Right heart impairment in sarcoido- Warshauer DM, Dumbleton SA, Molina PL, et al Abdominal CT find-
sis: haemodynamic and echocardiographic study. Eur J Respir Dis. ings in sarcoidosis: radiologic and clinical correlation. Radiology
1983;64(2):121–8 1994;192(1):93–8
Rose AS, Tielker MA, Knox KS. Hepatic, ocular, and cutaneous sarcoi- Welker L, Jorres RA, Costabel U, et al Predictive value of BAL cell
Un

dosis. Clin Chest Med. 2008;29(3):509–24 differentials in the diagnosis of interstitial lung diseases. Eur
Rybicki BA, Iannuzzi MC, Frederick MM, et al Familial aggregation of Respir J. 2004;24(6):1000–6
sarcoidosis: a case-control etiologic study of sarcoidosis (ACCESS). Winget D, O’Brien GM, Lower EE, et al Bell’s palsy as an unrecog-
Am J Respir Crit Care Med. 2001a;164:2085–91 nized presentation for sarcoidosis. Sarcoidosis 1994;11:S368–70
Rybicki BA, Iannuzzi MC. Epidemiology of sarcoidosis: recent advances Winterbauer RH, Belic N, Moores KD. A clinical intepretation of bilat-
and future prospects. Semin Respir Crit Care Med. 2007;28(1):22–35 eral hilar adenopathy. Ann Intern Med. 1973;78:65–71
Rybicki BA, Kirkey KL, Major M, et al Familial risk ratio of sarcoido- Wong M, Yasufuku K, Nakajima T, et al Endobronchial ultrasound: new
sis in African–American sibs and parents. Am J Epidemiol. insight for the diagnosis of sarcoidosis. Eur Respir J. 2007;
2001b;153(2):188–93 29(6):1182–6
Rybicki BA, Major M, Popovich J Jr, et al Racial differences in sarcoi- Zajicek JP, Scolding NJ, Foster O, et al Central nervous system sarcoi-
dosis incidence: a five year study in a health maintenance organiza- dosis – diagnosis and management. QJM. 1999;92(2):103–17
tion. Am J Epid. 1997;145:234–41
 Osteoarthritis
43
Nancy E. Lane and Roy D. Altman

increasing frequency due to the broader use of magnetic res-

43.1 Overview for Osteoarthritis onance imaging and ultrasound. The classic description of
inflammation includes heat, redness, pain, swelling, and loss
› Osteoarthritis is a common disease. This chapter will of function. The more modern understanding of inflamma-
review a number of myths and pearls that will assist tion is that it occurs at a cellular and subcellular level. Many
clinicians in treating this disease. of the same mechanisms present in rheumatoid arthritis and
other types of arthritis also occur in OA, albeit to a lesser
degree (Bonnet and Walsh 2005).
Myth: Osteoarthritis (OA) is an unavoidable outcome of Myth: Recreational running increases the risks of developing
aging. lower extremity OA.
Reality: The prevalence of OA in all joints increases with Reality: Research on weight-bearing activities that involve
age. However, more than 30% of people never develop radio- both low- and high-impact loading has been reported from
graphic evidence of OA as they age. In addition, only one retrospective, cross-sectional studies, as well as from longi-
third of those who have radiographic evidence of OA develop tudinal investigations. One study of middle-aged men and
joint pain. Thus, aging is only one factor in the development women runners who had jogged approximately 20 miles a
of OA (Lawrence et al. 2008). week for nine years found no increased risk of radiographic
knee OA compared with age-matched controls who did run
Myth: The more one uses one’s joints, the faster they wear
regularly (Lane et al. 1986).
out.
A follow-up study of that cohort performed ten years later
Reality: OA is not a disease of “wear and tear.” Regular (but showed no statistical difference between the runners and
not high impact) joint use is beneficial and necessary for the controls in incident knee OA Lane 1993. However, the run-
health of joints lined with synovium. The cartilage needs ners who did develop radiographic knee OA reported run-
motion to maintain nutrition. Weight-bearing exercise improves ning at a faster pace and were both heavier and older than
the circulation of blood to the joints because it allows fluid to were runners who did not develop OA. The preponderance of
pass through the synovium. Exercise also facilitates lymphatic other studies on this question indicate that individuals who
drainage, thereby expediting the clearance of oxidative prod- run recreationally in the absence of existing injuries are not
ucts. In contrast, the lack of joint motion leads to cartilage atro- at increased risks for knee and hip OA. On the other hand,
phy, a predecessor to OA (Sutton et al. 2001; Lane et al. 1986; repetitive bending does appear to be a risk factor for knee
Roos and Dahlberg 2005). OA (Schouten et al. 2002; Jensen 2008).
Myth: Inflammation has little role in the pathophysiology of Myth: Cracking one’s knuckles leads to hand OA.
OA.
Reality: The metacarpophalangeal joint is surrounded by
Reality: The classic findings of joint inflammation are not a car- synovial fluid, which contains approximately 15% carbon
dinal sign of OA. However, as the disease progresses, synovitis dioxide in solution. When one cracks one’s knuckles, a low-
develops. This is manifested as the joint warmth that is detected pressure zone (a vacuum) is created within the synovial cav-
in many patients who have advanced OA of the knee. Synovial ity. The source of the cracking sound is believed to be the
effusions are intermittent in OA, but often develop after overuse drawing in of carbon dioxide and water vapor from the joint
of the joint in a patient with advanced degenerative changes. fluid. This creates a bubble or a cavity of air that collapses
Patients with OA do not always notice or complain of instantly. The influx of fluid from all sides of the bubble gen-
joint swelling, but joint effusions are now recognized with erates the noise (Unsworth et al. 1971).

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 421

DOI:10.1007/978-1-84800-934-9_43, © Springer Science + Business Media B.V. 2009
422 N. E. Lane and R. D. Altman

The initial popping of the big bubble leaves small ones The progression of OA in most people is quite slow.
that remain for approximately 15 min before the carbon diox- Intermittent parinful episodes can usually be treated effec-
ide is totally redissolved. During this time, continued tug- tively with symptomatic treatment and there is little data to
ging on the finger can cause an expansion of the microbubble support the notion that NSAIDs accelerate joint degenera-
into a tiny shock absorber. This probably explains why crack- tion ( Haskinson et al. 1992; Rashad et al. 1989)
ing of the same knuckle in rapid succession is not possible.
Myth: Chondrocytes cannot divide in response to injury.
There are no data to confirm the old Myth that knuckle-
Thus, once articular cartilage is damaged, it inevitably
cracking causes OA. The sound of knuckle cracking must be
“fails.”
separated from the finding of crepitus that occurs with rou-
tine joint activity. Crepitus, caused by uneven joint surfaces, Reality: The end result of OA is loss of the articular cartilage
is a symptom of OA (Simkin 1990; Watson et al. 1989; and changes in the subchondral bone. The structure of carti-
Swezey and Swezey 1975). lage, which is comprised principally of water, is maintained
by aggrecan and collagen (Fig. 43.1). The aggrecan is mostly
Myth: Once joint pain is present with OA, progression of the
composed of proteoglycans: large, negatively charged groups
underlying process is inevitable.
of molecules that occupy space by imbibing water and elec-
Reality: The joint pain caused by OA is often intermittent trolytes. This confers elasticity and shear modulus to healthy
and variable. Many patients have prolonged periods of time cartilage. The primary collagen is a special form, collagen
in which they are free of pain. The cyclical nature of OA type II, which helps contain and bind to the aggrecan, pro-
relates to one of the etiologies of pain in this condition. OA viding tensile strength to the tissue.
results from damage to and then loss of articular cartilage. At When the surface of the cartilage is disrupted, chondro-
times, small or even microscopic pieces of cartilage break off cyte metabolic activity is upregulated with enhanced but
and float inside the joint. These fragments cause pain when defective proteoglycan and collagen production. The chon-
pressure is exerted on the joint. drocytes proliferate into clones but are eventually lost through
The breakdown or removal of these cartilage fragments apoptosis. It is not clear why the chondrocytes fail. However,
through natural processes or surgery leads to the recession of it has been postulated that their inability to reform the three-
joint pain. In addition, overuse of an osteoarthritic joint can dimensional structure of the cartilage leads to defects, par-
create synovitis that resolves spontaneously or with an anti- ticularly in older individuals. Some data suggest that
inflammatory agent (e.g., a glucocorticoid injection). chondrocytes can repair the articular cartilage if the defects

Fig. 43.1 Diagram of the articular cartilage. Chondrocytes, the only cell within the cartilage, are responsible for synthesizing the proteins that
make up the articular cartilage and the enzymes that degrade the articular cartilage
43 Osteoarthritis 423

occur in a young patient and are small (less than 3 mm) concentration of these compounds is reduced in OA. The
(Goldring and Goldring 2007; Lozado 2005). theory supporting the use of these medications holds that
these compounds are absorbed from the gastrointestinal tract
Myth: OA is caused by excessive cold.
and transported to the joint, where they reinforce articular
Reality: Limited clinical data indicate that symptoms of OA cartilage and prevent further joint deterioration.
worsen with changes in temperature, barometric pressure, or Other investigations suggest that there might be an intrac-
humidity. However, there is no evidence that OA is caused ellular regulatory effect associated with the downregulation
by excesses of either cold or heat in temperature. of inflammatory mediators such as nuclear factor-kappa B
(NFk(kappa)B). Multiple small clinical trials supported by
Myth: OA is an old person’s disease.
pharmaceutical companies reported that these compounds,
Reality: Although OA increases with age, the incidence of either alone or together, were efficacious for pain reduction
this problem begins to rise in the mid-40s (hardly the time of in OA (McAlindon et al. 2000; MIchel et al.2005; Bellamy
dotage!). Thus, OA is a common cause of joint symptoms 2006b; Towneed et al.2005). Moreover, two other studies
and loss of time from work long before someone is consid- (also sponsored by industry) reported reductions in joint
ered elderly. OA and back pain (which is often caused by space loss and the need for joint arthroplasty at five years
OA; see the Chap. 45) are the largest reasons for loss of time among patients treated with crystalline glucosamine sulfate
from work (Lawrence et al. 2008). (Reginster et al. 2001; Bruyere et al. 2008). However, a large
randomized trial sponsored by the National Institutes of
Myth: Once OA is present, nothing can be done.
Health showed that neither glucosamine hydrochloride nor
Reality: There remains no cure for OA. Current therapy com- chondroitin sulfate improved either pain or function com-
bines physical measures, education, medications, and sur- pared to celecoxib (Clegg et al. 2006). A two-year follow-up
gery. With appropriate therapy, most people with OA live study found no protection against joint space loss in indi-
normal or near normal lives. Pain from OA is a symptom that viduals on glucosamine or chondroitin (Sawitzke et al.
is addressable and should be brought to the attention of clini- 2008).
cians (Lozado 2005).
Myth: Dimethylsulfoxide (DMSO) applied to the joint is an
Myth: A highly regulated diet will cure OA. effective treatment for OA.
Reality: The word “diet” translates literally to controlled eat- Reality: DMSO is a topical solvent that penetrates the skin
ing. There are too many proposed diets to summarize. Among rapidly. The compound is available in many states by pre-
them all, only a few have been tested formal studies. None scription. There have been no formal studies of DMSO. No
has ever been shown to benefit patients with OA. Numerous scientific evidence indicates that it has any beneficial effect
dietary supplements are marketed, frequently in health food in OA (Brien et al. 2008).
stores, with unsupported claims of reducing or curing arthri-
Myth: Surgery is the only therapy likely to be effective once
tis (Lozado 2005).
advanced radiographic changes are evident.
The only relevant dietary fact at this time is that a bal-
anced diet that fosters weight reduction can reduce symp- Reality: Only half of all patients with radiographic evidence
toms in patients who are overweight. of OA have symptoms of OA. Although more severe radio-
graphic changes are more likely to be associated with symp-
Myth: Copper bracelets alleviate the pain of OA.
toms, such changes per se are not the primary consideration
Reality: There is no deficiency of copper in the body in OA. for surgery. The major determinants of the appropriateness
Indeed, the copper from a bracelet or other amulet does not of surgery are the severity of the patient’s pain and the restric-
penetrate the skin. Many patients develop a cutaneous reac- tion of function imposed by the OA (Felson 1993).
tion to copper bracelets that results from perspiration and
Myth: Weight-bearing exercise in subjects with knee OA
leads to a blue or black discoloration of the skin. Although
accelerates disease progression.
unseemly in some cases, this skin discoloration is self-limited
and has no clinical consequences (Walker and Keats 1976). Reality: On the contrary, numerous studies in subjects with
moderately severe knee OA have found that low-impact
Myth: Glucosamine hydrochloride or sulfate and chondroi-
exercise such as walking and bicycling improves patients’
tin sufate compounds slow the progression of knee OA.
pain and function (Halbert et al. 2001). Aquatic exercise is
Reality: The use of glucosamine and chondroitin compounds also helpful for both of these parameters in hip and knee OA
has become very popular (Theodosakis 2003). The initial (Bartels et al. 2007). Exercise programs for individuals with
theory to support treatment with these compounds was based OA should be monitored by a physician, nurse practitioner,
on two facts: (1) articular cartilage is composed of proteo- or physical therapist, with the intensity and duration titrated
glycans, including glucosamine and chondroitins and (2) the to pain and function. A balanced exercise program is an
424 N. E. Lane and R. D. Altman

effective therapeutic modality in OA. In addition, the exer- Approximately 10% of patients with OA have rapid pro-
cise program assists in weight control. gression of their disease. Bone marrow lesions may be a
marker for this disease subset (Hunter et al. 2006; Felson et al.
Pearl: The course of joint degeneration varies substantially
2007). Additional investigations of such patients are ongoing.
among patients with OA.
Pearl: Risk factors for OA progression can be modified.
Comment: Many years can elapse between the time clinical
symptoms or radiographic signs begin and the presence of Comment: Individuals with knee OA who lose weight expe-
both concurrently. OA probably begins with a change in the rience a reduction in joint pain and an improvement in joint
articular cartilage or subchondral bone. However, because function. How much weight is a patient required to lose
cartilage has no nerve supply, the inciting event (e.g., an before an impact on the symptoms of OA will be observed?
injury) may be asymptomatic. Over time, bone remodeling Some studies indicate that a minimum of five kilograms (12
occurs followed by hypertrophy and changes in the synovium pounds) is necessary (Felson and Chaisson 1997; Felson
with inflammation. When both articular cartilage and bone et al. 2004; Felson et al. 2007; Bierma-Zeinstra and Koes
changes have occurred, symptomatic pain is more consistent. 2007). The optimal place for finishing with weight reduction
The reason for joint pain remains difficult to identify in most depends on where one starts. For many patients, the answer
individuals with OA. Magnetic resonance imaging studies have is “the more, the merrier.”
revealed that transient bone marrow lesions around the joint
Pearl: Leg length discrepancy increases the progression of
affected by OA come and go and are associated with pain when
knee and hip OA.
they are present (Figs. 43.2a, b) (Hunter et al. 2006). These
bone marrow lesions represent areas of increased vascularity Comment: Some patients have leg length discrepancies from
and cellular congestion caused by a lymphatic drainage system birth. Others develop them as a consequence of arthritis in
that is inadequate to clear the accumulated lymphatic fluid. The the knee or hip or after hip arthroplasty. Correction of the leg
increased fluid coming into the lesion increases the pressure. length discrepancy through the use of orthotic, such as a heel

Fig. 43.2 Magnetic resonance a

imaging study from a patient
with osteoarthritis (OA) of the
knee. Note the bone marrow
lesions that are adjacent to the
joint space. These are known to
cause pain in OA patients. (a).
Right Knee: Normal X-Ray
where MRI with degenerative
changes (from OAI) Sagittal
MRI is Intermediate Weighted
Turbo Spin Echo with Fat
Suppression (IW TSE FS)
through medical compartment
showing: Bone Marrow Edema
(BME), menscal tear (black
arrow) and femoral cartilage loss
(white arrow). (b). Left Knee:
Large BMEs at baseline (left)
almost vanished by 24-month
follow up (right) Sagittal MRI is
IW TSE FS through medial b
compartment showing large
femoral BME (white arrow) that
has become diminished going to
tiny BME at 24 month follow up.
There is a tibial BME (white
arrow) at 24 months. This knee
also shows full thickness
cartilage loss of femur and tibia,
a severely damaged meniscus,
and osteophytes on femoral
condyle
43 Osteoarthritis 425

lift, can reduce low back pain and improve joint function for diagnosis. Magnetic resonance imaging should be reserved
(Hofmann et al. 2008; Golightly et al. 2007). for patients whose pain is persistent and unexplained.
Pearl: Abnormal biomechanical forces across the knee from Pearl: The method for obtaining knee radiographs is crucial
abnormal heel strike or pes planus can be modified with for the determination of OA severity. Each method provides
orthotics. These normalize the biomechanical forces across different information about the disease process.
the joint, slow the pace of cartilage degeneration, and reduce
Comment: The standard radiographic method for the assess-
joint pain.
ment of knee OA has the patient standing with approximately
Comment: Pes planus and other biomechanical abnormali- 5° of knee hyperextension (Fig. 43.3). This provides infor-
ties create abnormal mechanical forces across the ankle and mation about the presence of osteophytes but often does not
knee joints and accelerate the development of OA. Orthotics assess the severity of joint space narrowing accurately.
that provide an external longitudinal arch support across the If the patient is positioned with 15–20° of knee flexion, the
midfoot mitigate these adverse biomechanical forces and tibial plateau becomes flattened, permitting a reasonable esti-
improve exercise tolerance (Lee and Shmerling 2008; Janisse mate of the width of the tibiofemoral joint space (Fig. 43.4).
and Janisse 2000). Narrowing of this joint space is a surrogate measure for status
of the articular cartilage. Loss of joint space at the knee cor-
Pearl: Radiographic findings often correlate poorly with the
relates highly with articular pain and need for joint replace-
clinical severity of OA.
ment (Wolfe et al. 2002). Sunrise views of the knees to evaluate
Comment: Radiographic evidence of OA usually precedes the patellofemoral joint space have a limited clinical role.
the onset of clinical symptoms. However, radiographic find- Similar radiographic findings are found with hip OA.
ings generally lag behind tissue destruction and may under- Pelvic radiographs taken with the patient standing in 15° of
estimate the severity of the disease. In a small number of toe internal rotation are helpful in assessing the presence and
people with knee OA, the disease may be detected by mag- severity of hip OA (Fig. 43.5).
netic resonance imaging before changes on the radiograph.
Pearl: Intraarticular injections of glucocorticoids up to four
Magnetic resonance imaging is employed frequently to
times a year reduce joint pain and improve function.
determine the cause of knee pain and the severity of the under-
lying lesion. In patients with mild to moderate evidence of OA Comment: Depot glucocorticoid injections are quite effective
on plain radiographs, an magnetic resonance imaging study can in the treatment of a knee OA flare. They reduce joint swelling
reveal significant articular cartilage loss, synovitis, and bone rapidly and lead to improvements in pain and joint function.
marrow lesions, implying disease that is substantially more One randomized, placebo-controlled trial of glucocorticoid
severe than appreciated on radiographs (Felson 1993). However, injections into the knee joint reported rapid reductions in joint
in the majority of people, radiography is the only study needed pain (Dieppe et al. 1980 Jones and Doherty 1996). The

Fig. 43.3 The standard method

of assessing knee OA by
radiographs. The knee is
hyperextended by approximately
five degrees. This method is the
optimal one for detecting
osteophytes
426 N. E. Lane and R. D. Altman

Fig. 43.4 Radiograph of the

knee with the patient in 15–20°
of flexion. This view flattens the
tibial plateau and permits the
most accurate assessment of joint
space narrowing. (a). Normal
knee. (b). Mild joint space
narrowing. (c). Moderate joint
space narrowing. (d). Severe
joint space narrowing

three months (Raynauld et al. 2003). The American College

of Rheumatology recommends intraarticular glucocorticoid
injections not more than four times a year (Zhang et al. 2008;
Bellamy et al. 2006a).
Pearl: Activities of daily living can be increased with assis-
tive devices.
Comment: The following assistive devices are particularly
helpful to some individuals:

• Individuals with hand OA who have poor grip strength

can use devices to help open jars.
• Widened spoon, fork, and knife holders can help
patients eat.
• Sticks with a pincher at the end can help pick up things
from the floor and alleviate bending.
• A cane, held in the hand opposite the knee or hip affected
by OA, can help with balance for those with lower extrem-
ity OA (Lozado 2005).
• The most common error in the use of a cane is that they
are not the proper length. The elbow should be flexed to
12–20° (Lee and Shmerling 2008). Recommendations for
choosing and using a case in patients with hip and knee
OA are reviewed in Table 43.1.
Fig. 43.5 Radiographic view of hip OA. Superolateral joint space nar-
rowing, lateral and medial femoral osteophytes, and subchondral scle-
Pearl: The initial stages of joint degeneration in OA are clin-
rosis of the proximal femur are all present
ically silent.
improvement in overall joint pain persisted through a follow- Comment: This partially explains the discrepancy between
up period of 14 days. A randomized, controlled trial of intraar- the high prevalence of radiographic disease and lower pro-
ticular glucocorticoids for the treatment of hip OA was also portion of patients who eventually develop symptomatic OA.
effective in reducing joint pain (Lambert et al. 2007). However, even in the later stages of OA, there is a discrep-
One Myth about frequent intraarticular glucocorticoid use ancy between radiographic and clinical symptoms. Articular
is that this procedure accelerates the progression of cartilage cartilage has no nerve supply. As a result, once articular car-
damage. A two-year study showed no progression of radio- tilage is damaged, there is a delay until the development of
graphic damage following triamcinolone injections every symptomatic OA.
43 Osteoarthritis 427

Table 43.1 Recommendations for the selection and use of a cane

Consider the functionality of the cane, not only its appearance
Select from the various styles of canes by considering the stability
each offers
Choose a cane that is light
To select the proper length for a cane, stand up straight with your
shoes on and arms at your sides. The top of the cane should
reach the crease on the underside of your wrist
If the cane is a proper fit, your elbow will be flexed 15–20°when
you hold the cane while standing
Choose an adjustable cane if you plan to wear different styles of shoes
Make sure you have a good grip of the cane and that the fingers and
thumb do not overlap
Shift as much weight to the cane as necessary
Make sure that the tip of the cane is in good condition and that it is
replaceable
Hold the cane with the opposite hand of the side that needs extra
support Fig. 43.6 Patellofemoral OA (radiograph)
When ascending stairs, step first with the cane and good foot
following with the bad side
When descending stairs, step first with the bad foot and follow with
the cane and good foot ofemoral arthritis frequently accompanies the degeneration
Carol & Richard Eustice, About.com: Arthritis. observed in the other two knee compartments.
Pain arising from the patellofemoral joint degenerative pro-
cess is exacerbated by walking up and down stairs or inclines.
As patients with patellofemoral arthritis ascend stairs or walk
The time between the initial damage to the cartilage and
on inclines, the patella becomes pushed against the femoral
joint pain depends on a number of factors:
condyle. In the absence of cartilage, this interaction between
• the age of the patient the patella and femur produces pain. A common complaint
• the extent of the injury with patellofemoral pain is that the symptoms worsen follow-
• the location of the damaged cartilage within or outside of ing prolonged sitting in a movie theater (Lozado 2005).
a weight-bearing area
Pearl: There are reversible causes of knee pain in OA.
• the involvement of other joint structures such as the medial
menisci or lateral collateral ligaments. Injuries to these Comment: Not all knee pain in the patient with OA is caused
structures along with damage to the articular cartilage can by OA. The differential diagnosis for pain in the knee includes
render the joint unstable and result in a shorter time to meniscal and ligament tears, Baker’s cysts, anserine bursitis,
painful disease. the patellofemoral syndrome, gout, and calcium pyrophos-
phate deposition disease. Maintaining a mind that is open to
Pearl: Joint pain in OA does not originate from the articular
the possibility of other disorders is important for clinicians,
cartilage.
because the treatment of many of these other ailments can
Comment: Articular cartilage has no neural innervation. reduce knee pain and improve function in individuals with
Thus, joint pain does not arise directly from cartilage inju- OA. These disorders are reviewed in Table 43.2.
ries. Pain in OA frequently consists of a dull ache and joint
Pearl: A major risk factor for hip OA is mild dysplasia of the
stiffness that follows a period of sitting or sleeping. The pain
femoral head and acetabulum.
can be aggravated by joint use and can radiate to surrounding
joints and other structures. For example, pain that results Comment: Hip dysplasia (Fig. 43.7), a condition with which
from hip OA may radiate to the knee or to the lower lumbar many are born, results in an abnormal weight-bearing surface
spine (Lozado 2005). and an accelerated pace of hip degeneration with joint use.
Other abnormalities from childhood to consider in the adult
Pearl: Pain that originates from the patellofemoral joint is
patient who develops early hip OA are a slipped capital epi-
common and often accompanies tibiofemoral joint OA.
physis and Legg–Calve–Perthes disease (Figs. 43.8 and 43.9)
Comment: The knee joint is composed of three compartments: (Reijman et al. 2005b; Lane et al. 2000).
the medial and lateral tibiofemoral joints and the patellofemo-
Pearl: Joint pain from hip OA is usually referred to the
ral compartment. Degeneration of the medial and lateral com-
groin.
partments usually results in a significant amount of knee pain.
Disease in either of these compartments is used as a criterion Comment: Pain with internal rotation of the hip that is felt in
for joint replacement. OA in the patellofemoral compartment the groin probably represents hip OA. Patients who perceive
can also be a cause of significant pain (Fig. 43.6). Patell- pain laterally (over the trochanteric region) during internal or
428 N. E. Lane and R. D. Altman

Table 43.2 Common causes of knee pain that are not osteoarthritis
Condition History Physical examination Laboratory findings
Inflammatory arthritis (RA Morning stiffness and joint pain, Evidence of synovitis and warmth Can be associated with an
or PSA) worse in the a.m. on palpation elevated ESR of CRP,
Inflammatory joint fluid
Crystaline arthritis (gout or The first MTP or instep of the Significant pain on palpation and Inflammatory Joint fluid and
pseudogout) midfoot may be involved with erythema present during the crystals
gout other joints may not be acute episode
involved with pseudogout
Anserine brusitis Point tenderness with walking Pain on palpation of the anserine None
bursa below the knee at the
medial side of the tibia
Iliotibial band syndrome Pain that radiates down the lateral Pain at the insertion of the
thigh to the superior aspect of the iliotibial band
tibial/fibular junction. Frequently
observed in runners
Meniscal tear History that the knee may “give out” Tenderness on palpation of the MRI can identify tears
or buckle tibial femur joint and pain
with twisting of the knee with
patient lying prone on their
abdomen
Chondromalacia patellae Frequent knee pain in young women, Pain over patellofemoral joint and Usually physical examination or
and pain increases going up and with pressing the patella into lateral knee radiograph
down stairs or steps toward the femur
Adapted from Felson (2006)

Fig. 43.7 Mild hip dysplasia in a young adult. This condition predis-
poses the patient to the development of OA of the hip at a young age

external rotation probably have trochanteric bursitis rather

than significant hip OA.
Pain generated by abnormalities of the hip can also be
present in the anterior or lateral thigh, the knee, or the lumbar Fig. 43.8 Slipped capital epiphysis (radiograph)
spine. Patients who have symptoms simultaneously lower
back and anterior thigh pain are likely to have degenerative
disc disease of the lumbar spine.
Comment: The muscles around a joint reduce cartilage load-
Pearl: Muscle strength and tone are significantly reduced ing and help protect against joint damage. Maintaining mus-
around a joint affected by OA. cle strength and function around an osteoarthritic joint are
43 Osteoarthritis 429

Pearl: Prescribed periods of joint rest throughout the day are

useful in the management of OA.
Comment: Reduction of joint loading by either rest or use of
a cane improves activity levels. Continued or prolonged use
of OA joints, especially weight-bearing joints, increases joint
pain. Rest time reduces joint pain and prolongs the patient’s
ability to continue activities.
Pearl: The efficacy of “viscosupplementation” through injec-
tions of hyaluronic acid is controversial.
Comment: Hyaluronic acid is a component of synovial fluid
and articular cartilage, both of which are reduced in OA. As a
result of the decreases in synovial fluid and cartilage, the joint
loses much of the lubrication function of hyaluronic acid.
Commercial preparations of hyaluronic acid have been devel-
oped with the goal of improving joint pain and function.
Injections of hyaluronic acid alter synovial function,
reduce the sensitivity of nerve endings, increase the endoge-
Fig. 43.9 Legg–Calve–Perthes disease (radiograph). Plain radiograph nous production of hyaluronate, and have a modest antiin-
in an adult patient with residual coxa magna and plana deformity with flammatory effect. The FDA has approved a number of
superimposed joint changes hyaluronic acid preparations for patients whose knee pain is
not controlled by oral analgesics, NSAIDs, or intraarticular
important in reducing joint symptoms. Both muscle strength
glucocorticoid injections.
and range of motion around the OA joints can be improved
All approved hyaluronic acid preparations show that they
with a well-defined program of physical therapy that includes
are significantly more effective than placebo injections in
isometric and aerobic exercises.
reducing joint pain (Bellamy et al. 2006 b). Studies indicate
Pearl: Many nonpharmacologic interventions help treat OA that approximately two thirds of patients receive benefits
joint pain. from these injections. At the present time, there are no data
available to predict patients who are likely to respond to
Comment: Transcutaneous nerve stimulation, acupuncture,
hyaluronic acid injections. Studies are underway to access the
heat, and cold have all been shown to reduce the severity of
efficacy of hyaluronic acid preparations in nonweight-bearing
pain in patients with OA. Although these interventions have
joints, e.g., the shoulder and the elbow (Zhang et al. 2008; Jan
only modest effects on the reduction of pain, they provide an
et al. 2008; Bellamy et al. 2006 b; Blaine et al. 2008).
alternative to pharmacologic interventions. The use of cush-
ioned footwear such as running or walking shoes can also Pearl: Joint arthroplasty for OA of the first carpometacarpal
reduce lower extremity joint pains (Golightly et al. 2007). joint is not yet ready for prime time.

Fig. 43.10 OA of the first

carpometacarpal joint. (a)
Clinical photograph showing
squaring of the first carpometa-
carpal joint at the base of the
thumb. (b) Radiograph revealing
OA of the first
carpometacarpal joint
430 N. E. Lane and R. D. Altman

Comment: Degenerative joint disease of the first carpometa- Hunter DJ, Zhang Y, Niu J, et al Increase in bone marrow lesions associ-
carpal joint is extremely common, particularly among women ated with cartilage loss: a longitudinal magnetic resonance imaging
study of knee osteoarthritis. Arthritis Rheum. 2006;54: 1529–35
with hand OA (Fig. 43.10). However, currently there is very Jan MH, Lin JJ, Liau JJ, et al Investigation of clinical effects of high-
little if anything to gain in function with a joint replacement and low-resistance training for patients with knee osteoarthritis: a
at this site. randomized controlled trial. Phys Ther. 2008;88:427–36
Janisse DJ, Janisse E. Shoe modification and the use of orthoses in the
treatment of foot and ankle pathology. J Am Acad Orthop Surg.
References 2000;16(3):152–8
Jensen LK. Hip osteoarthritis: influence of work with heavy lifting,
climbing stairs or ladders, or combining kneeling/squatting with
Bartels EM, Lund H, Hagen KB, et al Aquatic exercise for the treatment heavy lifting. Occup Environ Med. 2008;65:6–19
of knee and hip osteoarthritis. Cochrane Database Syst Rev. Jones A, Doherty M. Intra-articular corticosteroids are effective in
2007;17(4):CD005523 osteoarthritis but there are no clinical predictors of response. Ann
Bellamy N, Campbell J, Robinson V, et al Intraarticular corticosteroid Rheum Dis. 1996;55:829–32
for treatment of osteoarthritis of the knee. Cochrane Database Syst Lambert RG, Hutchings EJ, Grace MG, et al Steroid injection for
Rev. 2006a;19:CD005328 osteoarthritis of the hip: a randomized, double-blind, placebo-con-
Bellamy N, Campbell J, Robinson V, et al Viscosupplementation for the trolled trial. Arthritis Rheum. 2007;56:2278–87
treatment of osteoarthritis of the knee. Cochrane Database Syst Rev. Lane NE, Bloch DA, Jones HH, et al Long-distance running, bone den-
2006b;19(2):CD005321 sity, and osteoarthritis. JAMA. 1986;7(255):1147–51
Blaine T, Moskowitz R, Udell J, et al Treatment of persistent shoulder Lane NE, Lin P, Christiansen L, et al Association of mild acetabular
pain with sodium hyaluronate: a randomized, controlled trial. J Bone dysplasia with an increased risk of incident hip osteoarthritis in
Joint Surg Am. 2008;90:970–9 elderly white women: the study of osteoporotic fractures. Arthritis
Bierma-Zeinstra SM, Koes BW. Risk factors and prognostic factors of hip Rheum. 2000;43(2):400–4
and knee osteoarthritis. Nat Clin Pract Rheumatol. 2007;3(2):78–85 Lawrence RC, Felson DT, Helmick CG, et al Estimates of the preva-
Bonnet CS, Walsh DA. Osteoarthritis, angiogenesis and inflammation. lence of arthritis and other rheumatic conditions in the United States.
Rheumatology (Oxford) 2005;44:(1):7–16 Part II. Arthritis Rheum. 2008;58:26–35
Brien S, Prescott P, Bashir N, et al Systematic review of the nutritional Lee YC, Shmerling RH. The benefit of nonpharmacologic therapy to treat
supplements dimethyl sulfoxide (DMSO) and methylsulfonyl- symptomatic osteoarthritis. Curr Rheumatol Rep. 2008; 10(1):5–10
methane (MSM) in the treatment of osteoarthritis. Osteoarthritis Lozado CJ. Management of osteoarthritis. In: Harris ED, Budd RC,
Cartilage. 2008;16(11):1277–88 Firestein GS, Genovese MC, Seargent JS, Ruddy S, Sledge CB edi-
Bruyere O, Pavelka K, Rovati LC, et al Total joint replacement after tors Kelley’s textboook of rheumatology. 7th edition, Vol II. PA,
glucosamine sulphate treatment in knee osteoarthritis: results of a Elsevier; 2005. p. 1528–40
mean 8-year observation of patients from two previous 3-year, ran- McAlindon TE, LaValley MP, Gulin JP, Felson DT. Glucosamine and
domised, placebo-controlled trials. Osteoarthritis Cartilage. 2008; chondroitin for treatment of osteoarthritis: a systematic quality
16(2):254–60 assessment and meta-analysis. JAMA. 2000;283(11):1469–75
Clegg DO, Reda DJ, Harris CL, et al Glucosamine, chondroitin sulfate, Raynauld JP, Buckland-Wright C, Ward R, et al Safety and efficacy of
and the two in combination for painful knee osteoarthritis. N Engl J long-term intraarticular steroid injections in osteoarthritis of the
Med. 2006;354(8):795–808 knee: a randomized, double-blind, placebo-controlled trial. Arthritis
Dieppe PA, Sathapatayavongs B, Jones HE, et al Intra-articular steroids Rheum. 2003;48:370–7
in osteoarthritis. Rheumatol Rehabil. 1980;19:212–7 Reginster JY, Deroisy R, Rovati LC, et al Long-term effects of glu-
Felson DT, Chaisson CE. Understanding the relationship between body cosamine sulphate on osteoarthritis progression: a randomised,
weight and osteoarthritis. Baillieres Clin Rheumatol. 1997;11(4): placebo-controlled clinical trial. Lancet 27;2001;357(9252):251–6
671–81 Reijman M, Hazes JM, Pols HA, et al Acetabular dysplasia predicts
Felson DT, Goggins J, Niu J, et al The effect of body weight on progres- incident osteoarthritis of the hip: the Rotterdam study. Arthritis
sion of knee osteoarthritis is dependent on alignment. Arthritis Rheum. 2005b;52:787–93
Rheum. 2004;50(12):3904–9 Roos EM, Dahlberg L. Positive effects of moderate exercise on gly-
Felson DT, Niu J, Guermazi A, et al Correlation of the development of cosaminoglycan content in knee cartilage: a four-month, random-
knee pain with enlarging bone marrow lesions on magnetic reso- ized, controlled trial in patients at risk of osteoarthritis. Arthritis
nance imaging. Arthritis Rheum. 2007;56(9):2986–92 Rheum. 2005;52:3507–14
Felson DT. The course of osteoarthritis and factors that affect it. Rheum Sawitzke AD, Allen D, Shi H, et al The effect of glucosamine and/or
Dis Clin North Am. 1993;19(3):607–15 chondroitin sulfate on the progression of knee osteoarthritis.
Felson DT. Clinical practice. Osteoarthritis of the knee. N Engl J Med. Arthritis Rheum. 2008;58(10):3183–91
2006;354(8):841–8 Schouten JS, de Bie RA, Swaen G. An update on the relationship
Goldring MB, Goldring SR. Osteoarthritis. J Cell Physiol. 2007; between occupational factors and osteoarthritis of the hip and knee.
213(3):626–34 Curr Opin Rheumatol. 2002;14:89–92.
Golightly YM, Tate JJ, Burns CB, Gross MT. Changes in pain and dis- Simkin PA. Habitual knuckle cracking and hand function. Ann Rheum
ability secondary to shoe lift intervention in subjects with limb Dis. 1990;9(11):957
length inequality and chronic low back pain: a preliminary report. Sutton AJ, Muir KR, Mockett S, Fentem P. A case-control study to inves-
J Orthop Sports Phys Ther. 2007;37:380–8 tigate the relation between low and moderate levels of physical activity
Halbert J, Crotty M, Weller D, et al Primary care-based physical activ- and osteoarthritis of the knee using data collected as part of the Allied
ity programs: effectiveness in sedentary older patients with osteoar- Dunbar National Fitness Survey. Ann Rheum Dis. 2001;60:756–64
thritis symptoms. Arthritis Rheum. 2001;45(3):228–34 Swezey RL, Swezey SE. The consequences of habitual knuckle crack-
Hofmann AA, Bolognesi M, Lahav A, Kurtin S. Minimizing leg-length ing. West J Med. 1975;122(5):377–9
inequality in total hip arthroplasty: use of preoperative templating Theodosakis J. The arthritis cure. London: MacMillan Publishers;
and an intraoperative X-ray. Am J Orthop. 2008;37(1):18–23 2003
43 Osteoarthritis 431

Unsworth A Dowson D Wright V. “Cracking joints”. A bioengineering Wolfe F, Lane NE, Buckland-Wright C. Radiographic methods in knee
study of cavitation in the metacarpophalangeal joint. Ann Rheum. osteoarthritis: a further comparison of semiflexed (MTP), schuss-
1971;30:348–58 tunnel, and weight-bearing anteroposterior views for joint space
Walker WR, Keats DM. An investigation of the therapeutic value of the narrowing and osteophytes. J Rheumatol. 2002;29: 2597–601
‘copper bracelet’-dermal assimilation of copper in arthritic/rheuma- Zhang W, Moskowitz RW, Nuki G, et al OARSI recommendations for
toid conditions. Agents Actions. 1976;6(4): 454–9 the management of hip and knee osteoarthritis, part II: OARSI evi-
Watson P, Hamilton A, Mollan R. Habitual joint cracking and radiologi- dence-based, expert consensus guidelines. Osteoarthritis Cartilage.
cal damage. BMJ. 1989;299(6715):1566 2008;16(2):137–62
 Regional Musculoskeletal Complaints
44
Joseph J. Biundo, W. Neal Roberts, and Atul Deodhar

44.1 Introduction for missing a septic joint is much higher than the chance of
introducing infection by tapping into the joint through an
overlying cellulitis. In addition, cutaneous erythema overly-
This chapter is divided into sections based upon the anatomic
ing a joint may not be secondary to an infection; gout, for
sites: shoulder, elbow, wrist, hand, coccyx, hip, knee, ankle,
example, can also be associated with intense cutaneous ery-
and foot. Before focusing on specific musculoskeletal
thema. It is critical to distinguish between microcrystalline
regions, however, we provide one introductory Pearl and
disease and infection. Never let the skin stand between you
then some general comments on the performance of arthro-
and the diagnosis!
centesis and joint injection. Some specific pointers on arthro-
centesis and the injection of individual joints are found Myth: Coagulopathy (INR > 3) and thrombocytopenia are
within the appropriate section. contraindications to arthrocentesis.
Introductory Pearl: “The number of rheumatologists doing Reality: Coagulopathy and thrombocytopenia are only rela-
procedures equals the number of different ways of perform- tive contraindications to arthrocentesis. One would prefer
ing them…” (Gardner 2007). not to have to tap a joint under such circumstances, but
potential coagulation issues should not stand as impediments
Comment: It is true: there are many ways to skin a cat, and
to the timely diagnosis of septic arthritis, which is a poten-
there are also many ways to perform most of the procedures
tially life-threatening condition. The statement that “No one
described in this chapter. The preferences of the authors are
has ever exsanguinated into a joint” is macho but true.
emphasized herein, but other approaches may also work.
Arthrocentesis is actually critical to perform in the setting
of a suspected hemarthrosis. Bleeding into a joint (e.g., as a
44.2 General Points on Performing complication of hemophilia) can cause chemical synovitis
and lead to irreversible joint damage. Evacuation of the blood
Arthrocentesis and Joint Injection via a surgical approach may be necessary to preserve joint
function. A diagnostic arthrocentesis is therefore essential
Pearl: The contraindications to performing an intra-articu- for management in such cases.
lar glucocorticoid injection are different from those for an Pearl: When assessing synovial fluid joint counts, the “Rule
arthrocentesis. of Twos” applies.
Comment: No joint should be tapped through an open, weep- Comment: The Rule of Twos refers to the number of white
ing, purulent ulcer. Apart from that proviso, there is virtually blood cells (WBCs) contained within the synovial fluid
no absolute contraindication to tapping a joint if the informa- (Table 44.2):
tion is needed for the purpose of diagnosis.
• Normal synovial fluid contains <200 WBCs/mm3.
Conversely, there are several important contraindica-
• Noninflammatory fluid contains between 200 and
tions to the injection of glucocorticoids. These are shown in
2000 WBCs/mm3.
Table 44.1.
• Inflammatory fluid contains >2,000 (but <50,000) WBCs/
Myth: A joint should not be tapped if there is an overlying mm3.
cellulitis.
Septic arthritis, a condition that does not fit neatly into the
Reality: The concern, of course, is the introduction of the Rule of Twos, is generally associated with synovial WBC
cutaneous infection into the joint. However, the price to pay counts in excess of 75,000/mm3.

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 433

DOI:10.1007/978-1-84800-934-9_44, © Springer Science + Business Media B.V. 2009
434 J. J. Biundo et al.

Table 44.1 Contraindications to the injection of glucocorticoids into a Large wheals are generally unnecessary and are discour-
joint aged because they are more painful than small ones. In addi-
Infection: septic joint, periarticular infection/cellulitis, bacteremia tion to the local infiltration of lidocaine, a vapocoolant spray
Unstable joint (fluorimethane) can be applied to the injection site to provide
Intra-articular fracture
topical anesthesia. After the skin is anesthetized, a small
Hypersensitivity to the agent or agents to be injected
Joint prosthesis amount of lidocaine is injected continuously as the needle is
Poor response to previous injection directed to the site of the lesion. When the needle reaches the
Significant coagulopathy desired site, additional lidocaine is injected.
The quantity of lidocaine used to anesthetize the lesional
area ranges from 0.2 to 1 mL, depending on the problem. The
clinician should pause for a few seconds after the lidocaine is
Pearl: Culture of the synovial fluid remains the only way to injected to ensure a full anesthetic effect. The lidocaine
diagnose septic bursitis. syringe is then changed to the glucocorticoid syringe (leaving
Comment: The “Rule of Twos” for the synovial fluid of joints the needle in place) and the glucocorticoid solution is injected.
does not apply to bursal fluids. Septic bursa fluids can have A Kelly clamp can facilitate the changing of syringes.
fewer than 2000 WBCs/mm3 and a lower percentage of neu- Pearl: No method of sterile skin preparation can eliminate
trophils when compared to infected joint fluids. Moreover, the the small risk of joint or soft tissue infection entirely.
systemic symptoms and signs of an infectious process are
often absent in patients with septic bursitis. The characteristics Comment: The rate of iatrogenic infection from arthrocentesis
of septic and nonseptic bursal fluids are shown in Table 44.3. is low, but clinicians should not become cavalier when per-
forming this procedure. The lowest estimate – 1 in 10,000 pro-
Pearl: Glucocorticoid-lidocaine injections should be pain-
cedures – is derived from the original description of the use of
free procedures, apart from the “bee sting” associated with
arthrocentesis as a route for the delivery of glucocorticoids
local anesthesia.
(Hollander 1961). The highest, 1 in 2,500, is an unpublished
Comment: The immediate pain relief that results from a description of anecdotal experience (Roberts WN). A full-time
properly directed injection into a tendon sheath, bursa, enthe- practitioner might be expected to cause three or four joint
sis, or nerve area helps to validate the clinician’s diagnosis infections in his or her career.
(Petri et al. 1987). The injection, if performed properly, The best evidence about sterile skin preparation is extrap-
should not cause discomfort to the patient. olated from studies of various surgical preparation tech-
Basic principles of intralesional injections include aseptic niques. Most such studies rely upon surrogate measures of
technique and the use of small needles (25-gauge 5/8", infection, e.g., colony counts from swabs of skin prepared in
25-gauge 1½", or a 22-gauge 1½"). (Larger needles may a sterile fashion. Trial data indicate that a single chlorhexi-
facilitate joint aspiration, when synovial fluid sampling is dine scrub, the most expensive technique, is probably the
necessary). A 2% lidocaine solution is preferred over a 1% best (Mimoz et al. 1999). A triple betadine preparation, simi-
solution because it is more effective in decreasing discomfort lar to the technique for obtaining blood cultures, is probably
related to the injection. The use of separate syringes for lido- the second most effective. The least effective approach is the
caine and glucocorticoid, hereafter termed the “two-syringe use of alcohol swabs alone.
technique,” avoids mixing of the substances and permits No technique can eliminate the small risk of joint infection
immediate infiltration of lidocaine, beginning with a small entirely (Charalambous et al. 2003). Every procedure, regard-
intracutaneous wheal (Kessell 1986). less of the details of sterile technique, instills some inoculum of

Table 44.2 Assessing synovial fluid joint counts

Normal Noninflammatory Inflammatory Septic
Color Transparent Transparent Translucent to opaque Opaque
Viscosity Very high High Low Variable
WBC count (#/mm3) <200 200–2,000 >2,000 but <50,000 >50,000
Polys (%) <75 <75 >75 >75
Examples OA RA Bacterial infection
Gout
AVN CPPD
HPOA Spondyloarthropathy
Lupus
Serum sickness
CPPD = calcium pyrophosphate dihydrate deposition; AVN = avascular necrosis; HPOA = hypertrophic osteoarthropathy
44 Regional Musculoskeletal Complaints 435

Table 44.3 Bursal fluid characteristics (Nonseptic vs. septic) needle track into the dermis, thereby avoiding the potential
Nonseptic Septic for skin depigmentation.
WBC/cm3 90–11,000 350–395,000
Myth: The glucocorticoid preparation and lidocaine solution
%PMN 0–90 50–100
Gram stain + 0 5–100% should be mixed for joint and soft tissue injections.
(Zimmermann et al. 1995 and Raddatz et al. 1987) Reality: Clinical practice on this point varies widely, but the
only sound rationale for mixing the two compounds is con-
microorganisms into the joint. The larger cores of skin result- venience. The two-syringe technique described earlier has
ing from the use of larger needles presumably are associated several advantages:
with larger inocula. Diligent attention to preparation of the skin
• Less pain with injection
is important, regardless of which cleansing agent is employed.
• Lower risk of injecting the glucocorticoid into the soft
Pearl: For optimal interpretations, synovial fluid should be tissue rather than the joint, which can be associated with
evaluated within a couple of hours of arthrocentesis. skin and subcutaneous atrophy
• Removal of concern about the potential interaction or pre-
Comment: Delay in fluid evaluation can be associated with a
cipitation of the glucocorticoid and anesthetic.
fall in the synovial fluid WBC count (Kerolus et al. 1989).
The quantity of CPPD and monosodium urate crystals may
also fall somewhat with time, but the crystals can still be
observed for days after the arthrocentesis, particularly if the 44.3 Tendinitis: General Points
fluid is refrigerated.
If synovial fluid is allowed to stand for hours, the fall in
Myth: Tendinitis is an inflammatory condition.
the number of CPPD crystals tends to be greater than the
decline in monosodium urate crystals. As discussed else- Reality: The various terms used to refer to tendon injuries or
where, new crystals do not develop in synovial fluid that has overuse syndromes are confusing. Tendinitis has been the
been aspirated and allowed to stand for hours before exami- main term for such conditions. However, the histology of this
nation (See Chap. 37). condition reveals degenerative changes and hyaline features
in the tendon but few inflammatory cells (Khan et al. 2002).
Pearl: Use of a more water-soluble glucocorticoid preparation
Thus, “tendinosis” may be more appropriate. Nevertheless,
decreases the possibility of skin atrophy if some of the injected
the beneficial response to glucocorticoid injections suggests
glucocorticoid remains within the dermis or epidermis.
a role for cytokines in the pathophysiology of this disorder.
Comment: If excessive dose of glucocorticoid is injected or “Tendinopathy” defers judgment on the contribution of
if the injection is placed into the subcutaneous tissue or inflammation to the problem. “Tendon insufficiency” refers
within the skin itself, then skin atrophy, subcutaneous fat to a stretching or tearing of the tendon. Finally, “tenosynovi-
atrophy, and skin depigmentation can occur (Gottleib et al. tis” and “peritendinitis” refer to an inflammatory response
1978). Atrophy and depigmentation are observed in particu- within the tenosynovium or peritenon.
lar following injections for lateral epicondylitis and Apart from these musings, we suspect that both the term
DeQuervain’s tenosynovitis. “tendonitis” and its incorrect spelling (What’s a tendin?) are
The use of a water-soluble glucocorticoid preparation here to stay.
diminishes the risk of these complications as well as the risk
Pearl: When injecting a tendon sheath, the needle should be
of glucocorticoid-induced tendon weakness or a postinjec-
placed parallel to the tendon fibers, and not into the tendon
tion flare of inflammation. Two good options for water-
itself.
soluble glucocorticoid preparations are betamethasone and
triamcinolone (Drendorf et al. 1986). Comment: The clinician should not inject glucocorticoids
directly into the tendon. This error heightens the risk of ten-
Pearl: Postglucocorticoid hypopigmentation of the skin can
don rupture. Injection of the glucocorticoids parallel to the
be avoided by the injection of air following the medication.
tendon sheath generally leads to a good therapeutic effect.
Comment: Cutaneous hypopigmentation can result from
“backleak” of glucocorticoids through the injection track
following the injection of a superficial joint or tendon such as
the ulnar styloid. After completion of the glucocorticoid 44.4 Shoulder
injection, the syringe can be unscrewed and replaced with a
tuberculin syringe containing 0.3 cc of air. The injection of Pearl: While injecting a joint or a bursa close to a tendon,
an “air seal” prevents the leakage of glucocorticoids from the the use of a small bore needle (25 gauge) can make the
436 J. J. Biundo et al.

difference between injecting glucocorticoids correctly into a

joint or a bursa and injecting them in a tendon.
Comment: While injecting the subacromial bursa, the needle
tip may enter the top of the rotator cuff. When this occurs,
intra-tendinous pressure will not allow the plunger of the
syringe to be depressed if a small-bore needle (25-gauge) is
used. The small-bore needle thus provides a feedback mech-
anism. If the clinician encounters excessive resistance, the
needle should be pulled back 2–3 mm before injection is
attempted again.
Myth: Bursitis is the most common problem of the shoulder.
Reality: The term “bursitis” is often used incorrectly by cli-
nicians and patients to describe the most common cause of
shoulder pain. Rotator cuff tendinitis is actually the most
common cause of this complaint (Chard et al. 1991). The
Fig. 44.1 Superior migration of the humeral head in the setting of a
tendons of the rotator cuff attach at the humeral tuberosities. complete rotator cuff tear. (Figure courtesy of Dr. Joseph Biundo)
The subacromial bursa, which lies inferior to the acromion,
is contiguous with the subdeltoid bursa that covers the
humeral head. Although rotator cuff tendinitis is the primary The rotator cuff normally holds the humeral head within the
cause of shoulder pain, secondary involvement of the subac- glenoid fossa. When the rotator cuff is torn, the deltoid exerts
romial bursa occurs in some cases. relatively unopposed upward pull on the humeral head, lead-
ing to superior migration. This finding may be noted inciden-
Myth: A complete tear of the rotator cuff means that the
tally on a chest radiograph.
entire rotator cuff is torn.
Pearl: Brachial plexopathy is a neurogenic cause of acute,
Reality: Rotator cuff pathology is another confusing area of
severe shoulder pain.
nomenclature. A partial or incomplete tear is a nonfull thick-
ness tear. Partial tendon tears can involve the glenohumeral Comment: Most causes of acute, severe shoulder pain are
side of the rotator cuff, the bursal side, or the intratendinous musculoskeletal in nature. However, severe shoulder pain can
portion of the cuff. A complete tear of the rotator cuff, in con- also be caused by a neurological lesion. Brachial plexopathy,
trast, refers to a tear that is a “through and through” or full also known as the Parsonage–Turner syndrome, is such an
thickness tear that involves both the rotator cuff and the sub- entity (Misamore and Lehman 1996). Brachial plexopathies
acromial bursa. Such a rupture creates a direct communication can result from trauma, tumor, radiation, vaccinations (“inoc-
between the glenohumeral joint and the subacromial bursa. ulation neuritis”), diabetes, infection, or median sternotomy
Complete rotator cuff tears vary in size and clinical done for cardiac surgery. They can also be idiopathic.
impact. Some are small and relatively asymptomatic. The A brachial plexopathy presents with a deep, sharp shoul-
patient’s ability to abduct the arm to some degree can persist der pain that begins rapidly. The pain is exacerbated by
despite small rotator cuff tears. Symptomatic tears, however, abduction and rotation. The severity of the pain is generally
result in a range of complaints that extends from severe pain quite perplexing at presentation. Its onset is followed by
and mild weakness to no pain and marked weakness, or some weakness of the shoulder girdle. Muscle atrophy of the del-
combination in between. The complete tears are classified as toid can ensue, and sensory findings can also be present.
small (1 cm or less), medium (1–3 cm), large (3–5 cm), or The entire brachial plexus or merely individual compo-
massive (>5 cm). Thus, a complete tear of the rotator cuff nents can be affected. Plain radiographs and MRI of the
does not necessarily mean a total avulsion of the rotator off shoulder are normal in patients with brachial plexopathy.
the humeral head. However, an electromyogram can help confirm the diagnosis
by demonstrating positive sharp waves and fibrillations in
Pearl: Superior migration of the humeral head detected by
the involved muscles. Ultrasonography can also be used to
plain radiography indicates a significant rotator cuff tear
evaluate brachial plexus pathology (Graif et al. 2004). The
(Fig. 44.1).
treatment of brachial plexopathy consists mainly of avoiding
Comment: Plain radiography of the shoulder is not the usual reinjury or exposure to a metabolic cause of nerve damage.
method of diagnosing a large tear of the rotator cuff, but such Recovery requires from 1 month to several years, and resid-
a finding points to a significant rotator cuff lesion. ual muscle weakness sometimes persists.
44 Regional Musculoskeletal Complaints 437

Pearl: Microcrystalline disease can cause acute, severe of the number of compartments and tendon insertions. These
shoulder pain. compartments communicate poorly or not at all. The shoul-
der is technically more than one joint, in the sense that the
Comment: Hydroxyapatite, calcium pyrophosphate, and acromio-clavicular joint is entirely separate from both the
monosodium urate crystals are all potential causes of acute subdeltoid space and the true glenohumeral joint. The latter
shoulder dysfunction. In such cases, the rotator cuff and sub- two are also separate from one another unless a connection is
acromial bursa are apt to become inflamed. Hydroxyapatite established in the form of a full-thickness rotator cuff tear.
crystal deposition leads to a tendinitis that presents with In a randomized trial conducted at Guys Hospital in
abrupt, excruciating pain. Such cases tend to occur in younger London, patients were examined by one orthopedic surgeon.
patients and are likely to be associated with calcific deposits Anatomic diagnoses such as rotator cuff pain, impingement
within the supraspinatus tendon (Fig. 44.2). A plain radio- syndrome, biceps tendinitis, and acromion-clavicular osteoar-
graph demonstrates these deposits most effectively when the thritis were rendered. The patients were then randomized to
arm is held in external rotation. The deposits appear round or have either the anterior glenohumeral approach for injection
oval and several centimeters in length. A true subacromial marked regardless of diagnosis, or to have the approach cor-
bursitis can also result from the rupture of calcific material responding to the actual anatomic diagnosis marked. A second
into the bursa. The hydroxyapatite deposits may resolve blinded surgeon in another room then injected the shoulder
spontaneously over a period as short as weeks, but some- according to the approach marked on the skin. Finally, a third
times require longer. The acute episode is best treated with a blinded surgeon later followed up to evaluate the results of the
glucocorticoid injection of the subacromial bursa. injection. The group with the injections directed by their ana-
Acute shoulder arthritis due to CPPD deposition also tomically specific diagnosis fared about 15% better in terms of
occurs (Hughes et al. 1990). In addition, CPPD deposition the magnitude and duration of their clinical improvement.
may be found in chronic rotator cuff tears. Gout in the shoul- Thus, from the subspecialty point of view, the “horse-
der is quite rare, and when present, it is most likely to be part shoes and hand grenades” dogma of glucocorticoid injection
of a polyarticular attack. fails. Learning to differentiate among the major causes of
Myth: A variety of approaches to injecting the shoulder work shoulder complaints and mastering the individual approaches
equally well. to shoulder injection are worthwhile.
On the other hand, in general practice, the 15% advantage
Reality: This Myth might also be called the “horseshoes and may not be sufficient to justify abandoning the “one size fits
hand-grenades” approach to shoulder injections. There are a all” approach to shoulder injections. It is impossible to dam-
couple of reasons why it is not quite true. The fundamental age the pleura or axillary when approaching the shoulder
one is that the shoulder is the most complex of joints in terms from the poster-lateral angle with a 1.5 inch needle into the
subacromial bursa (Fig. 44.3). For generalists, therefore, it is

Fig. 44.2 Calcific tendonitis secondary to hydroxyapatite deposition Fig. 44.3 Posterolateral approach to injecting the shoulder. (Figure
within the supraspinatus tendon. (Figure courtesy of Dr. Richard Chou) courtesy of Dr. John Stone)
438 J. J. Biundo et al.

safer to learn the posterolateral approach to shoulder injec- 44.5 Elbow

tions rather than the anterior glenohumeral approach. A
remote possibility of pneumothorax or injury to the axillary
Pearl: Be alert to the possibility of infection as the cause of
neurovascular bundle exists with the latter approach.
olecranon bursitis.
Pearl: Any palpable effusion in the shoulder should be evalu-
Comment: In septic olecranon bursitis, the major clues are
ated carefully.
localized erythema and warmth over the bursa, in addition to
Comment: Palpable effusions of the shoulder are rare and, bursal swelling (Fig. 44.5) (Smith et al. 1989). Some nonsep-
when present, should be given careful consideration. The tic inflammatory conditions can also cause erythema over the
glenohumeral joint lies deep inside many soft tissue struc- bursa, but this is atypical in the absence of an active infec-
tures. Thus, small or even moderate-sized effusions are usu- tion. Bursal swelling is also more likely to extend beyond the
ally not detected by palpation. Likewise, the presence of normal bursal area if an infection is present, and pain is pres-
warmth or erythema is less frequently detected in this joint, ent more frequently when the bursa is infected. However, in
even with acute inflammation or infection. contrast to infection within a more confined space (such as a
The anatomy of the normal shoulder capsule includes a joint), infection of the olecranon bursa does not always cause
dependent, pouch-like projection that extends into the axilla, significant pain.
which remains inaccessible to physical examination. This pouch Additional clues to a septic etiology are the presence of a
is apparent only on arthrography or another imaging study. Any skin abrasion, puncture, or other superficial trauma to the
effusion of the shoulder fills this pouch first before becoming olecranon as a possible entry site of infection. Bursal fluid
palpable on the antero-superior aspect of the shoulder. should be aspirated, examined for crystals, and submitted for
Palpable swelling of the shoulder might be due to an effu- Gram stain and culture. Glucocorticoids should not be
sion of only the glenohumeral joint or only of the subacro- administered until an infection has been excluded. WBC
mial bursa, or more likely a combination of the two with a counts in bursae do not reach the levels as high as in similar
communication via a torn rotator cuff. In fact, shoulder effu- problems of joints (see Table 44.3) (Canoso 1981).
sions are seen most commonly in association with a torn
Pearl: One common complication of tapping the olecranon
rotator cuff (Fig. 44.4) (Hollister et al. 1995). These cuff tears
bursa is the creation of a fistula.
are often associated with osteoarthritis. This combination,
when severe, is known as “cuff tear arthropathy” (Neer et al. Comment: The risk of fistula development can be minimized
1983). When associated with hydroxyapatite crystal deposi- by taking the “zigzag approach.” With this approach to the
tion, it has been termed “Milwaukee shoulder” (Halverson bursa, the needle is inserted at an angle that will not enter
1981). Shoulder effusions are also found in RA, but less com- the bursa directly. When the tip of the needle is within the
monly now than before the era of biologic therapies. When
shoulder effusions are present, they are frequently associated
with torn rotator cuffs (Weiss et al. 1975; Kim et al. 2007).

Fig. 44.4 Shoulder effusions are seen most commonly with complete Fig. 44.5 A septic olecranon bursitis. (Figure courtesy of Dr. John
tears of the rotator cuff. (Figure courtesy of Dr. Joseph Biundo) Stone)
44 Regional Musculoskeletal Complaints 439

subcutaneous tissue, it is redirected toward and into the a

bursa.
Pearl: Swelling of the antecubital area of the elbow is
unusual, but may be due to cubital bursitis.
Comment: Swelling of the antecubital area of the elbow is not
common. Several known entities may account for the swell-
ing. The first is cubital bursitis, also known as bicipito-radial
bursitis. This condition is manifested by swelling in the ante-
cubital fossa and tenderness, with some restriction of prona-
tion. Rheumatoid arthritis or other inflammatory arthritic
conditions are the most likely causes of cubital bursitis, but
the disorder can also be secondary to trauma or overuse.
Clinicians must also bear in mind the possibility of a par- b
tial or complete tear of the distal biceps tendon. An MRI or
ultrasound can confirm the diagnosis (Sofka and Adler 2004).
Arterial aneurysm from illicit drug injections, accidentally
but repeatedly injected arterially instead of intravenously,
can also give antecubital swelling.

44.6 Wrist

Myth: Carpal tunnel syndrome (CTS) is caused by overuse

among computer and keyboard operators.
Reality: This concept has received much interest in the fields
Fig. 44.6 (a, b) Site, orientation, and angle for a carpal tunnel injec-
of work-related injuries, occupational medicine, workers’ tion. (Figure courtesy of Dr. Joseph Biundo)
compensation, and disability. However, multiple studies have
failed to demonstrate correlations between CTS and the use
or overuse of computers and other keyboard devices (Atroshi Glucocorticoid injections can also be the treatment of first
et al. 2007; Palmer et al. 2007). Work-related CTS, however, choice to provide immediate relief for patients with severe
has been associated with occupations that require a high force neurogenic pain from median nerve compression. In general,
or vibration of the hand. Some nonillness and noninjury fac- milder CTS cases tend to respond more readily to an injection
tors are also associated with CTS. For example, women are than do chronic cases, in which the patients have already
three times more likely than are men to develop CTS. developed substantial prolongation of distal latencies on
Wrist shape also appears to be a risk factor for CTS. Persons nerve conduction velocity studies.
with “square-shaped” wrists are at greater risk of this condition The approach to a carpal tunnel injection is via the palmar
when compared with patients whose wrists are not “square- surface of the hand (Fig. 44.6). The site of injection is about
shaped.” A squared wrist results from an increase in the depth one centimeter distal to the distal wrist crease and about
of the wrist in relation to the width. When the depth:width ratio 0.75 cm toward the ulnar side of midline. Caution should be
is ³0.7, the likelihood of CTS increases. Finally, a number of employed not to inject too far to the ulnar side of midline,
systemic disorders (e.g., amyloidosis) are known to be associ- lest one hit the ulnar artery. The skin should be numbed down
ated with CTS. The traditional four-part differential diagnosis to the transverse carpal ligament, using only about 0.2 mL of
of simultaneous bilateral CTS is amyloidosis, rheumatoid lidocaine. The clinician can feel the transverse carpal liga-
arthritis, hypothyroidism, and pregnancy. However, nerve con- ment with the needle tip. The 5/8-inch needle should be
duction studies have shown that idiopathic CTS is often bilat- pushed through the ligament and inserted to the hilt.
eral, even though symptoms may be unilateral. Lidocaine that is injected directly next to the median
nerve can anesthetize the hand. After reaching the desired
Pearl: The carpal tunnel can be injected safely, easily, and
area of the flexor tendons beneath the flexor retinaculum, the
effectively.
needle is kept in place, and the syringe with lidocaine is
Comment: A glucocorticoid injection can be an effective changed to a tuberculin syringe containing 1 mL of 40 mg of
therapy for CTS if splints worn at night have been ineffective. methylprednisolone. The median nerve should not be injected
440 J. J. Biundo et al.

directly. If the patient feels tingling when the needle is

inserted or when the injection begins, the needle is slightly
withdrawn and redirected in a more ulnar orientation.
While injecting the carpal tunnel, ask the patient to flex
and extend the fingers. If the needle has lodged within the
tendon, then the needle will move along with the finger. If
this occurs, simply pull the needle back a millimeter or so
and redirect it. Repeat the finger movements to confirm that
you are not in the tendon.
Myth: A positive Finkelstein’s test is the best maneuver for
the diagnosis of deQuervain’s tenosynovitis.
Reality: The Finkelstein’s test is a venerable maneuver for
the diagnosis of deQuervain’s tenosynovitis. However, the
test has limitations and can lead to false-positive results. For
example, osteoarthritis of the first carpal-metacarpal joint Fig. 44.7 Palpation of the tendon on the palmar surface of the hands, the
often produces local pain on flexion of the thumb and abduc- key to diagnosing tendinitis of the flexor digitorum superficialis and
tion of the wrist, simulating a positive Finkelstein’s test. The flexor digitorum profundus tendons. Tendinitis of these tendons is a com-
mon cause of intense hand pain and is often underdiagnosed, in patients
Intersection Syndrome is another condition that can result in with underlying inflammatory disorders such as rheumatoid arthritis as
a positive Finkelstein’s test (Pantukosit et al. 2001). well as health individuals. (Figure courtesy of Dr. John Stone)
The method of performing the Finkelstein’s test has often
been described erroneously in the literature (Elliott 1992).
Finkelstein described the test as follows:“On grasping the volunteer a clear description of the pain’s location. In some
patient’s thumb and quickly abducting the hand ulnarward, cases, the pain is reported as radiating to the PIP and MCP
the pain over the styloid tip is excruciating” (Finklelstein joints on the dorsal side, thus further misleading the
1930). This account does not describe the thumb to be flexed examiner.
in the palm with a fist being made over the thumb, as often The only way to identify this condition properly is for the
misstated in the literature. In the correct maneuver, the fingers clinician to supinate the hand and palpate the palmar tendons
of the involved side are not flexed. The proper performance of directly (Fig. 44.7). The diagnosis is confirmed by the find-
this test helps decrease the number of false positives. ing of localized tenderness that reproduces the patient’s
symptoms and usually swelling of the palmar tendon sheaths.
Tenderness is usually present directly over the metacarpal
heads or (more likely) just proximal to them. The middle and
44.7 Hand index fingers are affected most often, but the thumb, ring,
and little fingers can also be involved.
Nodules composed of fibrous tissue can often be palpated
Pearl: Volar (palmar) flexor tenosynovitis is a common but
in the palm just proximal to the MCP joint on the volar side.
often overlooked cause of hand pain that can be quite
A nodule can interfere with the normal gliding of the tendon,
severe.
leading to a triggering or locking (trigger finger). The symp-
Comment: Inflammation of the flexor tendon sheaths in the toms of trigger finger are often intermittent and can produce
palm is extremely common, but often goes unrecognized. an uncomfortable or even painful sensation. The presenting
The most common cause of volar flexor tenosynovitis is complaint may be a trigger finger, which can vary from mild
overuse trauma of the hands. Excessive gripping leads to triggering to severe. Trigger finger may emerge as a conse-
increased pull on the flexor tendons. However, this tenosyno- quence of an earlier or evolving tenosynovitis, but volar
vitis can also be part of inflammatory conditions such as flexor tenosynovitis can occur in the absence of triggering.
rheumatoid arthritis, psoriatic arthritis, and apatite crystal The technique for injection for volar flexor tenosynovitis
deposition disease. is the following. The injection site is determined by careful
Flexor tenosynovitis of the palm is easily overlooked in palpation for tenderness, thickening, and possibly a nodule
the setting of PIP or MCP synovitis caused by an underlying in one of more of the flexor tendons on the palmar surface.
arthritic condition, yet the tenosynovitis can be the dominant The site is generally about 0.5–2.0 cm proximal to the palmar
cause of pain and loss of motion in some patients. Pain in the digital crease. About 0.3 mL of methylprednisolone acetate
palm generally occurs upon flexion of the involved digits, and 0.3 mL of 2% lidocaine are drawn into two separate 1 mL
but is felt even at rest in severe cases. Patients seldom tuberculin syringes, each of which has a 5/8 inch, 25-gauge
44 Regional Musculoskeletal Complaints 441

needle. The needle with the lidocaine syringe is inserted into

the skin at about a 30° angle, going from distal to proximal.
About 0.1 mL of lidocaine is injected during the insertion of
the needle, with the remainder injected at the lesion site, and
then followed by the steroid injection from the separate
syringe. The injection is performed parallel to the tendon
sheath, avoiding penetration of the tendon itself.
Nodules cannot be injected because of too much back
pressure, and the syringe may be knocked off the needle if
attempted. A nodule may be bathed with the glucocorticoid
in an attempt to decrease the size. When done properly, the
procedure is relatively painless. Moreover, an immediate
reduction in pain is effected in most cases, validating the
diagnosis and the accuracy of the injection.

Fig. 44.8 The runner’s stretch, for preventing nocturnal leg cramps.
(Figure courtesy of Dr. John Stone)
44.8 Coccyx

Pearl: Coccydynia can be treated successfully by a local containing compounds and various muscle relaxants are
injection. employed by some practitioners as therapy, but the benefits
are mixed at best. Calf cramps are often attributed errone-
Comment: Many physicians are reluctant to inject this area ously to hypokalemia, hypocalcemia, or other electrolyte
even in persistent cases of coccydynia, because of the deli- disturbances. Such disturbances are rarely the true cause.
cate nature of the site involved. Yet the response to a gluco- The best therapy for calf cramps is prophylactic stretch-
corticoid and lidocaine injection is often dramatic (Ramsey ing of the calf muscles. Many older people have unrecog-
et al. 2003). Tenderness is typically present at the lower part nized tightness of the gastrocnemius and soleus muscles.
of the coccyx. This bony area is the proper site to inject. The This muscle tightness can cause vague aching of the lower
precise injection site is identified by careful palpation. leg, in addition to being the major predisposing cause of
The patient lies prone and is draped appropriately. The cramps. The tight muscles, which may be acquired over a life
area is cleaned well with alcohol wipes and providone-iodine. time of not stretching, are easily palpable and observed to
Topical and local forms of anesthesia are employed as limit the degree of stretch of the calf.
described elsewhere in this chapter, and the two-syringe tech- The proper procedure for stretching the calves is to per-
nique is used. The injection of 1–2 mL of lidocaine (2%) must form the “runner’s stretch”: i.e., to stand close to a wall and
be directed down to the coccyx, with the injection needle place one’s outstretched hands on the wall. One foot is placed
actually reaching bone. The periosteum of the coccyx is very with the toe touching the wall and the other approximately
sensitive and painful when stuck with a needle. Therefore, the 12–24 inches behind the heel of the front foot. The knee of the
approach to the bone is gradual and gentle. The location of the front leg is flexed inducing a stretch of the calf of the back leg,
bony target should be maintained during the injection by con- which is not flexed at all. Both feet can remain flat on the floor
tinued palpation of the site with the index finger of the nonin- or the heel of the rear foot may be slightly raised (Fig. 44.8).
jecting hand. The needle remains at the site as the syringe is The stretch should be gentle and prolonged, lasting for about
changed to one containing 40 mg of methylprednisolone, 30 sec and repeated for a total of five repetitions on each leg.
which is then distributed over the bony area. If the diagnosis The stretching should be done nightly before bedtime.
of coccydynia is correct and the injection directed properly, Patients should also be instructed to dorsiflex the foot
lidocaine mediates immediate pain relief. when a calf cramp occurs. This movement stretches the calf
muscles and helps abort the cramp by countering the con-
traction of the gastrocnemius muscle.
44.9 Leg Cramps

Pearl: Prophylactic stretching is the most effective treatment 44.10 Hip

for calf cramps.
Comment: Nocturnal calf cramps, a frequent complaint of Myth: The main pathology in trochanteric bursitis is in the
elderly patients, do not respond well to medications. Quinine- bursa.
442 J. J. Biundo et al.

Reality: Like many syndromes that are misnamed, trochant- Pearl: If a trochanteric bursa injection does not help the
eric bursitis is actually a tendinosis of the gluteal medius and/ patient’s trochanteric area pain, it may be referred as pain
or gluteal minimus tendons. Several bursae are in the area, from piriformis syndrome.
including the subgluteus medius and the sub-gluteus mini-
Comment: A simple piriformis muscle injection will take
mus bursae. These bursae can be involved secondarily, but
care of it. Piriformis muscle is a fan-shaped muscle lying
tendinosis is the main problem. In some chronic cases, a par-
deep to the gluteal muscles. It originates from the lateral bor-
tial or complete tear of these tendons may be present and can
der of the sacrum and inserts on the greater trochanter. To
be demonstrated by MRI of the hip.
inject the piriformis muscle, draw a straight line from the
Myth: Trochanteric bursitis is an easy diagnosis to make. posterior superior iliac spine (the “dimple of Venus”) to the
greater trochanter. Palpate just lateral to the midpoint of this
Reality: Trochanteric bursitis is usually considered to be a
line. If the patient is very tender on this area, inject it with a
straightforward diagnosis, characterized by lateral “hip” pain
3-inch spinal needle using 3–6 cc of lidocaine.
that is exacerbated by walking, lying on the involved side,
and moving the proximal leg in various ways. The diagnosis
is generally confirmed by the presence of tenderness on pal-
44.11 Knee
pation over the trochanteric area. However, it is important to
remember a few other problems that can cause pain in the
trochanteric region. Pearl: The knee is easiest and best joint on which joint injec-
Pain and point tenderness localized over the trochanteric tion techniques can be learned.
region is not equivalent to bursitis in all cases. It may be the Comment: The knee joint is large and its surface anatomy
result of symptoms referred from degenerative arthritis in the landmarks are prominent. The neurovascular bundles, which
lower back, particularly in the L4–5 and L5–S1 facet joints are located on the lateral aspect of the popliteal fossa, are
(Traycoff 1991). This type of referred pain is termed somatic, well removed from potential injection points (If you remem-
nonradicular referred pain. True radicular pain may be ber anything about embryology and the rotation of the limbs,
referred to the lateral hip, but is less likely to be localized then this makes sense; otherwise just accept it as fact).
only to the lateral hip. Three different approaches – anterior, medial, and lateral
An important reason to consider other possible diagnoses – are used to inject the knee:
is that in 60% of patients with true hip joint pathology, pain
emanating from the hip joint radiates not only to the anterior • The anterior approach places the patient in a seated position,
groin region but also to the greater trochanter area. True pain with the knee flexed to 90° (Fig. 44.9). This approach is use-
emanating from the hip joint can also be referred to the tro- ful when the patient is unable to get onto the examining
chanteric region and can produce referred tenderness, as table easily, because the procedure can be performed with
well. the patient seated in a chair. The anterior approach is also
Other conditions that can mimic the symptoms of tro- useful if no arthrocentesis is required for diagnostic purposes,
chanteric “bursitis” are fibromyalgia and unrecognized stress but rather only an injection is needed for treatment. This is
fractures, or even the rare impacted hip fracture, which is
sometimes poorly seen on plain radiographs. If the diagnosis
remains in doubt, the selective use of imaging studies (e.g.,
MRI or ultrasound) increases the accuracy of the diagnosis
in this area, helping to identify tendinitis of the gluteus
medius and minimus tendons, tendon tears, and bursal
swelling.
None of these caveats and potential alternative diagnoses
makes it unreasonable to inject trochanteric tenderness after
an appropriate physical examination of the hip region. The
injection can serve as both a diagnostic and therapeutic trial.
However, not every case of putative trochanteric tenderness
responds to the first injection. When it does not, the differen-
tial diagnosis must be revisited. In one series of 65 patients
diagnosed with trochanteric bursitis, only two-thirds of the
cases improved after the first injection (Schapira and Nahir
1986). The remaining one-third took more time and addi- Fig. 44.9 The best way to inject (but not aspirate) a knee. (Figure
tional glucocorticoid injections. courtesy of Dr. Ken Fye)
44 Regional Musculoskeletal Complaints 443

because with the patient positioned for an anterior approach, Pearl: A dry tap medially is insufficient to exclude a septic
synovial fluid is forced posteriorly by knee flexion into the knee.
popliteal fossa, diminishing the likelihood of a successful
Comment: If sepsis is suspected, a dry tap through the medial
joint aspiration. The anterior approach is a poor choice if the
approach is insufficient to exclude infection. Some congeni-
diagnosis is unclear and the acquisition of synovial fluid
tal and acquired variations in anatomy can block aspiration.
is essential. However, it is the simplest approach to inject-
If no synovial fluid is obtained via the medial approach, then
ing the knee with glucocorticoids or viscosupplementation
a second attempt via the lateral approach is appropriate. The
treatments.
joint cavity is closer to the skin surface on the lateral side.
The ease and simplicity of the anterior approach for most Thus, any effusion present might be easier to aspirate. A true
knee joint injections cannot be overemphasized. This approach dry tap virtually rules out septic arthritis.
is valuable when there is only a minimal joint effusion (e.g., A study of knee injections and aspirations in patients who
in osteoarthritis); when the knee is deformed by osteoarthritis, did not have a large effusion to guide the operator showed that
rendering the anatomy difficult; when the patient is morbidly even experienced operators missed the knee joint 20% of the
obese and the landmarks around the knee are blunted; and time (Jackson and Evans 2002). Aspects of chronic synovitis
when the patient comes to the clinic in a wheelchair and does frequently conspire to create “dry tap” situations as shown in
not want to get onto the examination table. the last three images of the figure. The lateral approach for
Make the patient sit up on the edge of the chair or the knee aspiration is more successful when one gets an assistant
examination table with the leg dangling down – foot not rest- to displace the patella slightly (Anderson 2005).
ing on ground – so as to open the joint space because of
Pearl: If at first one does not succeed in tapping a knee, think
gravitational pull on the leg. Palpate the patellar border and
“anatomy.”
the patellar tendon at the lower end of the patella (this is eas-
ily palpable even in morbidly obese patients when they are Comment: When attempting to tap or inject a knee via either
sitting up). Feel the two “soft spots” on either side of the the medial or lateral approach, the clinician often strikes
patellar tendon, half an inch inferior to the lower end of bone immediately after entering the joint. In this setting, take
patella. Enter the needle through either of these spots, per- a step back and remember the anatomy of the patella. The
pendicular to the skin pointing to the center of the knee. inferior aspect of this bone is V-shaped (remember its shape
on sunrise radiographic view). Thus, the clinician’s needle is
• The medial approach (Fig. 44.10) has the best landmarks
probably striking the side of the patella, confounding attempts
and is easily accessed when the patient is in bed with the
to withdraw fluid. The needle should be partly withdrawn
hips externally rotated. This approach is preferred when
(but not removed entirely) and redirected at a steeper angle
obtaining synovial fluid is essential for the purposes of
in order to attain the retropatellar space.
diagnosis.
• The lateral approach is the essential back-up approach if Myth: In anserine bursitis, the main pathology is bursal
the medial approach fails to produce fluid. inflammation.
Reality: The pes anserinus (Latin for “goose foot”) is com-
posed of the conjoined tendons of the sartorius, gracilis, and
semitendinosus muscles. The bursa extends between the ten-
dons and the tibial collateral ligament. Although some cases
of bursal involvement with swelling have been documented,
the pain from this region is more likely to emanate from the
entheses in this region, the sites of attachment of the three
tendons to the bone.
At the insertion of the pes anserinus on the anterior tibia,
the sartorius is the most proximal and anterior tendon
(Fig. 44.11). The gracilis lies slightly distal and posterior,
and the semitendinosus lies most posterior and distal. One
possible mechanism leading to anserine bursitis is the
repeated stress on the pes anserinus that occurs in going from
a sitting to a standing position. This stress is particularly
severe in a person with osteoarthritis of the knees and weak-
ness of the quadriceps. Pain occurs in the pes anserinus
Fig. 44.10 The medial approaches to aspirating the knee. (Figure region when a patient with poor flexibility of the hips and
courtesy of Dr. Ken Fye)
444 J. J. Biundo et al.

Comment: If the needle hits the bone immediately after enter-

ing the ankle joint, it is most likely to touch the dome of the
talus. Ankle joint is formed by the concave lower end of the
tibia and the convex dome of the talus. The commonest mis-
take that happens is to direct the needle toward the heel,
which would touch the talus. Redirecting the needle cephalad
(away from the heel, upwards, toward tibia) will enable you
to enter the space under the concave inferior surface of the
tibia.
Pearl: Heel pain that occurs upon the first step out of bed in
the morning is plantar fasciitis.
Comment: Plantar fasciitis is characterized by pain on the
plantar surface of the heel. The history is the key to accuracy
in diagnosis: clear identification of the specific site of the
pain must be extracted from the patient. The pain character-
istically occurs in the morning upon arising and is most
Fig. 44.11 The site of the anserine bursa. Inflammation of this area is severe for the first few steps. Thus, it is known as “first step
often misdiagnosed as “degenerative arthritis” of the knee. (Figure
courtesy of Dr. John Stone) pain.” The heel pain then generally lessens after the patient
has been on his or her feet for 30 s or so. After this initial
improvement, the pain may return later in the day, particu-
knees climbs stairs. Failure of these joints to flex fully places larly after prolonged standing or walking.
greater strain on the tendons of the pes anserinus. The symptoms of plantar fasciitis can cause bitter com-
A glucocorticoid injection helps anserine bursitis acutely, plaints. Physical examination often reveals anteromedial ten-
but the longer term solution lies in physical therapy exercises derness on the sole of the foot, over the medial calcaneal
designed to improve muscle strength and joint flexibility. tubercle at the origin of the plantar fascia (Fig. 44.12).
Pearl: Anserine bursitis is the major cause of knee pain in Pearl: Plantar fasciitis can be treated effectively with a glu-
some patients with osteoarthritis of the knee. cocorticoid injection.
Comment: Anserine bursitis frequently occurs concomitantly Comment: Many clinicians are reluctant to inject plantar fas-
with knee osteoarthritis but is frequently overlooked. Once ciitis. This is usually because of uncertainty with regard to
osteoarthritis of the knee has been diagnosed, it is often the proper approach or concern about causing too much pain.
blamed for any symptomatology in this region. In many Other clinicians worry about the potential for inducing atro-
cases, however, anserine bursitis is the principal source of phy of the subcutaneous heel fat pad. The proper site for
pain. The diagnosis of anserine bursitis is made by eliciting injection is several millimeters proximal to the distal edge of
exquisite tenderness to palpation over the pes anserinus. the calcaneus and several millimeters toward the medial
This area is normally somewhat sensitive to pressure, and aspect. The target is often tender on deep palpation.
care must be exerted to compare the tenderness on one side The area is cleansed and anesthetized with lidocaine (2%).
with that of the opposite leg. The patients are asked the ques- A 1.5 inch, 25-gauge needle generally is required to ensure
tion that is central to all musculoskeletal evaluations: “Is this that the periosteal region of the calcaneus is reached. The
the same pain as the one that is bothering you?” This diagno- bony tissue of the calcaneus must be felt with the needle tip
sis may be validated by the remission of “knee” pain follow- to ensure that the injection is performed beneath the subcuta-
ing the infiltration of approximately 1 mL of lidocaine (2%) neous tissues. In addition, the fascial fibers involved are
into the pes anserinus area. Treatment consists of a glucocor- those that are attached to the bone itself. The patient can
ticoid injection, quadriceps strengthening, and stretching of expect immediate relief of pain if the diagnosis is correct and
the muscles of the involved tendons. the injection is accurate.
Pearl: A unilateral flat foot may be due to a rupture of the
posterior tibialis tendon.
44.12 Ankle Comment: A progressive flat foot, especially a unilateral one,
is often due to the rupture of the posterior tibialis tendon
Pearl: If at first you do not succeed at ankle arthrocentesis, (Churchhill and Sferra 1998). This diagnosis is often over-
think “anatomy.” looked. Trauma or chronic tendon degeneration can cause
44 Regional Musculoskeletal Complaints 445

Charalambous CP, Tryfonidis M, Sadiq S, et al Septic arthritis follow-

ing intra-articular glucocorticoid injection of the knee – a survey of
current practice regarding antiseptic technique used during intra-
articular glucocorticoid injection of the knee. Clin Rheumatol.
2003;22:386–90
Chard MD, Hazleman R, Hazleman BL, et al Shoulder disorders in the
elderly: a community survey. Arthritis Rheum. 1991;34:766–9
Churchhill RS, Sferra JJ. Posterior tibial tendon insufficiency: its diag-
nosis, management, and treatment. Am J Orthop. 1998;27:339–47
Drendorf H, Mollmann H, Gruner A, et al Pharmacokinetics and phar-
macodynamics of glucocorticoid suspensions after intra-articular
administration. Clin Pharmacol Ther. 1986;39:313–7
Elliott BG. Finkelstein’s test: a descriptive error that can produce a false
positive. J Hand Surg Eur Vol. 1992;17:481–2
Finklelstein H. Stenosing tenovaginitis at the radial styloid process.
J Bone Joint Surg. 1930;12:509–40
Gardner GC. Teaching arthrocentesis and injection techniques: what is
the best way to get our point across? J Rheumatol. 2007;34:1448–50
Gottleib NL, Penneys NS, Brown HEJR. Periarticular perilymphatic
skin atrophy after intra-articular corticosteroid injections. JAMA.
1978;240:559–60
Graif M, Martinoli C, Rochkind S, et al Sonographic evaluation of bra-
Fig. 44.12 Site of injection of plantar fasciitis. (Figure courtesy of chial plexus pathology. Eur. Radiol. 2004;14:193–200
Dr. John Stone) Halverson PB, Cheung HS, McCarty DJ et al “Milwaukee shoulder”:
Association of microspheroids containing hydroxyapatite crystals,
active collagenase, and neutral protease with rotator cuff defects. II.
this tendon to tear. Rheumatoid arthritis is another common Synovial fluid studies. Arthritis Rheum 1981;24:474–83
culprit. Hollander JL. Intrasynovial corticosteroid therapy: a decade of use.
With rupture of the posterior tibialis tendon, insidious Bull Rheum Dis. 1961;11:239–40
pain and tenderness is noted along the course of the tendon Hollister MS, Mack LA, Patten RM, et al Association of sonographj-
cally detected subacromial/subdeltoid bursal effusion and intraar-
just distal to the medial malleolus. Medial swelling of the ticular fluid with rotator cuff tear. Am J. Roent. 1995;165:605–8
hindfoot is also evident. In the setting of a ruptured posterior Hughes GM, Biundo JJ, Scheib JS, et al Pseudogout and Pseudosepsis
tibialis tendon, the patient cannot rise onto the ball of the of the shoulder. Orthopedics 1990;13:1169–72
affected foot if the contralateral foot is off the floor (single Jackson DW, Evans NA. Thomas: accuracy of needle placement into
the intra-articular space of the knee. J Bone Joint Surg Am. 2002;
heel lift test). 84A:1522–7
Pearl: Aspirating any fluid from the 1st MTP joint in a patient Kerolus G, Clayburne G, Schumacher HR. Is it mandatory to examine
synovial fluids promptly after arthrocentesis? Arthritis Rheum.
with gout can be difficult. Here are two strategies to increase 1989;32:271–8
your success rate of getting a sample out of the MTP joint to Kessell L. A colour atlas of rupture of the rotator cuff. Netherlands:
look for crystals. Wolfe Medical; 1986. p.22
Khan KM, Cook JL, Kannus P, et al Time to abandon the tendinitis
Comment: Use intra-articular anesthetic injection first. This “myth” (Editorial). BMJ. 2002;324:626–7
will allow for better positioning, landmark identification, Kim HA, Kim SH, Seo Y. Ultrasonographic findings of the shoulder in
patients with rheumatoid arthritis and comparison with physical
and sometimes one is able to aspirate back the injected anes- examination. J Korean Med Sci. 2007;22:660–6
thetic and identify crystals in that drop of fluid. Mimoz O, Karim A, Mercat A, Cosseron M, Falissard B, Parker F,
Even if no frank fluid is aspirated, the answer may already Richard C, Samii K, Nordmann P. Chlorhexidine compared with
lie in the needle tip or hub. Tap the contents of the needle, povidone-iodine as skin preparation before blood culture. A ran-
domized, controlled trial. Ann Intern Med. 1999 ;131(11):834–7
small though they may be, onto a glass slide. This may reveal Misamore GW, Lehman DE. Parsonage–Turner syndrome (acute bra-
enough fluid for a diagnostic analysis of crystals. chial neuritis). J Bone Joint Surg. 1996;78-A:1405–8
Neer CS, Craig EV, Fukuda H. Cuff-tear arthropathy. J Bone Joint Surg.
1983;65:1232–44
Palmer KT, Harris EC, Coggon D. Carpal tunnel syndrome and its rela-
References tion to occupation: a systematic literature review. Occup Med.
2007;57:57–66
Pantukosit S, Petchkrua W, Stiens SA. Intersection syndrome in Buriram
Anderson BC. Guide to arthrocentesis and soft tissue injection. hospital: a 4-yr. prospective study. Am J Phys Med Rehabil. 2001;
Philadephia: Elsevier Saunders; 2005. p. 132 80:656–61
Atroshi E, Gummesson C, Ornstein E, et al Carpal tunnel syndrome and Petri M, Dombrow R, Brower R, et al Randomized double-blind pla-
keyboard use at work: a population study. Arthritis Rheum. cebo controlled study of the treatment of the painful shoulder.
2007;56:3620–5 Arthritis Rheum. 1987;30:1040–5
Canoso JJ. Bursae, tendons and ligaments. Clin Rheum Dis. 1981;7: Raddatz DA Hoffman GS, Franck WA. Septic bursitis: presentation,
189–221 treatment and prognosis. J Rheumatol. 1987;14:1160–3
446 J. J. Biundo et al.

Ramsey ML, Toohey JS, Neidre A, et al Coccygodynia: treatment. Sofka CM, Adler RS. Sonography of cubital bursitis. AJR. 2004;183: 51–3
Orthopedics 2003;26:403–5 Traycoff RB. Pseudotrochanteric bursitis. The differential diagnosis of
Smith DL, McAfee JH, Lucas LM, et al Septic and Nonseptic olecranon lateral hip pain. J Rheumatol. 1991;18:1810–2
bursitis: utility of the surface temperature probe in the early differentia- Weiss JJ, Thompson GR, Doust V, et al Rotator Cuff Tears in Rheumatoid
tion of septic and nonseptic cases. Arch Intern Med. 1989;149:1581–5 Arthritis. Arch Intern Med. 1975;135:521–525
Schapira D, Nahir M, Scharf Y. Trochanteric bursitis: a common clini- Zimmermann B, Mikolich DJ, Ho G Jr. Septic bursitis. Semin Arthritis
cal problem. Arch Phys Med Rehabil. 1986;67(11):815–7 Rheum. 1995;24:391–10
 Low Back and Neck Pain
45
Rajiv K. Dixit and Joseph H. Schwab

45.2 General Considerations

45.1 Overview
Myth: Patients with low back pain never get better.
› Low back pain is the most common musculoskeletal
complaint. An estimated 80% of the population will Nucleus pulposus
a
experience low back pain over the course of their
lifetime.
› The surgical treatment of common lumbar disorders
Annulus fibrosus
accounts for over $20 billion as hospital fees each year
in the United States.
› The rate of procedures directed at the lumbar spine is
Spinal canal
increasing faster than the aging of the population. Yet, Pedicle
Cauda equina
the indications for and the efficacy of surgical inter-
ventions for common lumbar disorders have been a
source of controversy.
› Low back pain affects the area between the lower rib Lamina Facet joint

cage and gluteal folds, and frequently radiates into the Ligamentum
flavum Spinous Process
thighs.
› Neck pain is also a common complaint, affecting Vertebral

approximately 70% of the population at some time in b Transverse body

foramen
Traverse
their lives. Neck pain typically affects the posterior process
aspect of the neck and frequently radiates into the
shoulder blades and deltoid muscles, even in the
Nerve root
absence of nerve root or spinal cord involvement.
› An understanding of the basic anatomy of the spine is
essential for the diagnosis and management of lower
back and neck conditions (Fig. 45.1).
Pedicle
› Randomized trials have evaluated the surgical man- Superior articular
facet
agement of lumbar herniated discs, lumbar spinal Spinal Cord
stenosis, and lumbar spondylolisthesis with stenosis. Lamina
Spinal canal
These trials have improved our ability to advise patients Spinous Process
regarding the expectations of nonoperative vs. opera- c
tive management.
› Spinal stenosis is caused by encroachment upon the vertebral body
spinal nerves by bulging intervertebral discs, hypertro-
Intervertebral
phied facet capsules, osteophytes, and bulging liga- Facet Joint disc
mentum flavum. Spinous
Process
› Some individuals are predisposed to spinal stenosis by Intervertebral Foramen
virtue of congenitally narrowed spinal canals.
Fig. 45.1 Schematic representation of the superior aspect of a lumbar
vertebra (a), mid-cervical vertebra (b), and a lateral view of the spine (c)

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 447

DOI:10.1007/978-1-84800-934-9_45, © Springer Science + Business Media B.V. 2009
448 R. K. Dixit and J. H. Schwab

Reality: Approximately 90% of patients with acute low back In contrast to the pain caused by sciatica, sclerotomal pain
pain improve spontaneously within 4 weeks. The problem is the is nondermatomal in distribution and dull in quality. It does
5–7% of patients who develop a chronic low back pain syndrome. not radiate below the knee and is not associated with paraes-
thesias. Sciatica results from nerve root compression and is
Myth: Surgery is the final common pathway in the manage-
generally caused by a herniated disc. This produces lancinat-
ment of low back pain.
ing pain that has a dermatomal distribution. Sciatica typi-
Reality: The vast majority (>85%) of patients with low back cally radiates below the knee and may be accompanied by
pain have transient symptoms. In fact, most will not seek sensory and motor deficits.
medical attention at all. Those patients who do seek medical
attention often respond to activity modification, over-the- Pearl: The vast majority of patients with low back pain have
counter pain medications, and time. a mechanical cause.
The surgical treatment of low back pain without neurologic Comment: More than 95% of low back pain is mechanical
symptoms is controversial. Two prospective, randomized clini- (Deyo and Weinstein 2001). Degenerative changes, some-
cal trials have evaluated the efficacy of lumbar fusion surgery times termed as lumbar spondylosis, occur in both the inter-
for low back pain (Fritzell et al. 2001; Brox et al. 2003). The first vertebral disc and facet joint. These are the most important
study included 294 patients from 19 medical centers in Sweden, causes of mechanical low back pain.
enrolled over a 6-year period (Fritzell et al. 2001). The patients Systemic disorders, which include malignancies, infec-
enrolled had low back pain without evidence of nerve root com- tions, and the seronegative spondyloarthropathies, account
pression. All had tried conservative management before trial for only 1% of patients with low back pain. Nonsystemic
entry. Patients were randomized into one of four groups, includ- visceral disease accounts for only 2%. A patient with normal
ing three groups that received spinal fusion by different methods blood cell counts, erythrocyte sedimentation rate, and radio-
and one nonoperative group. At 2 years, an independent observer graphs of the lumbar spine is unlikely to have an underlying
found that the surgical group had “excellent or good” results in systemic disease as the cause of low back pain.
46% of cases, compared with 18% for the nonoperative group.
The surgical group was statistically superior to the nonoperative Myth: A careful evaluation yields a definitive pathoanatomic
group with regard to pain, disability, and return to work status. diagnosis in most patients with low back pain.
The major criticism of this study was the method by which
Reality: Eighty percent of patients who present with low back
the conservative group was treated. The patients in that group
pain defy a clear-cut pathoanatomic diagnosis. This is largely
had already failed conservative treatment yet were random-
because of the weak association between symptoms and the
ized to receive additional conservative therapy. This flaw
findings on imaging studies. Imaging abnormalities, often the
biases the trial results in favor of surgery. However, propo-
result of age-related degenerative changes, are frequently
nents of surgery cite the trial as evidence that fusion surgery
present even in healthy individuals who are entirely asymp-
is a viable option for patients whose clinical features fit that
tomatic. Conversely, some patients with severe low back pain
of patients enrolled in this study: low back pain without evi-
have minimal findings on imaging studies. Furthermore, even
dence of nerve root compression.
when one is certain that the cause of low back pain is lumbar
A second prospective, randomized trial compared lumbar
spondylosis, identification of the “pain generator” is chal-
fusion followed by postoperative physical therapy with a
lenging. The precise disc or discs or facet joint or joints
“modern” rehabilitation program that featured intensive
responsible for the patient’s pain is difficult to localize.
exercise sessions guided by principles of cognitive/behav-
ioral therapy (Brox et al. 2003). This trial revealed no differ- Myth: Strain and sprain are common, well-defined causes of
ences between the two treatment modalities with regard to low back pain.
pain, disability, or return to work.
At this point, broad conclusions based on the current lit- Reality: The diagnostic uncertainty with regard to low back
erature are not possible. A well-designed trial is needed to pain has resulted in the use of nonspecific terms such as “strain,”
shed more light on the optimal management of this common “sprain,” and “lumbago.” These terms have never been charac-
clinical problem. terized anatomically or histologically. Therefore, idiopathic
low back pain is perhaps a more accurate label for these patients
Myth: Low back pain that radiates into the lower extremities who have a mostly self-limited syndrome of back pain.
probably represents sciatica.
Reality: In the vast majority of patients who have low back
pain that radiates into the lower extremities, the pain arises 45.3 History and Physical Examination
from pathology within the disc, facet joints, or paraspinal
muscles and ligaments (Dixit and Dickson 2008). This pain is Pearl: Careful history taking helps differentiate mechanical
said to be “sclerotomal,” and it is not neurogenic in nature. causes of low back pain from inflammatory etiologies.
45 Low Back and Neck Pain 449

Comment: Mechanical low back pain is due to an anatomic or

functional abnormality that is not associated with inflamma-
tory or neoplastic disease. Mechanical low back pain typi-
cally increases with physical activity and upright posture, and
is relieved by rest and recumbency. Nonmechanical low back
pain, especially when accompanied by nocturnal pain, sug-
gests the possibility of an underlying infection or neoplasm.
Inflammatory causes of low back pain such as ankylosing
spondylitis are associated with marked morning stiffness that
lasts for more than 30 min. The pain frequently improves
with exercise but not with rest. Pain is often worse during the
second half of the night. Some patients complain of alternat-
ing buttock pain (Rudwaleit et al. 2006).
Pearl: Physical examination can often differentiate struc-
tural from functional scoliosis.
Comment: Scoliosis is associated with structural changes of
the vertebral column and sometimes the rib cage, as well. In
adults, structural scoliosis is usually secondary to degenera- Fig. 45.2 Posterolateral disc herniation resulting in nerve root
impingement
tive changes, although some individuals have a history of
adolescent idiopathic scoliosis. With forward flexion, struc-
tural scoliosis persists. In contrast, in functional scoliosis,
which usually results from a paravertebral muscle spasm or
leg length discrepancy, usually disappears.
Pearl: Neurologic evaluation of the lower extremities in a patient
with sciatica can identify the specific nerve root involved.
Comment: The nucleus pulposus in a degenerated disc may
prolapse and push out the weakened annulus. This usually
occurs in a posterolateral direction. Evidence of disc herniation
is commonly observed in many asymptomatic adults, but
sometimes, this herniation leads to nerve root impingement
(Fig. 45.2). In general, the more caudal nerve root is impinged
by the herniated disc; that is, the L5 nerve root is impinged by
an L4–5 herniation, and the S1 nerve root by an L5–S1 hernia-
tion. Neurologic evaluation of the lower extremities including
motor testing, reflex testing, and tests for dermatomal sensory
loss help to identify the specific nerve root involved (Fig. 45.3). Fig. 45.3 Neurologic evaluation of the lower extremities including
motor testing, reflex testing, and tests for dermatomal sensory loss help
Pearl: Organs that share segmental innervation with the to identify the specific nerve root involved
spine may refer pain to the spine.
Comment: Pelvic diseases can refer pain to the sacral area; vertebral osteomyelitis or surgical decompression for the
lower abdominal diseases to the lumbar area; and upper cauda equina syndrome.
abdominal diseases to the lower thoracic spine. However, in In the rest of the patients with low back pain (>95% of all
such cases, local signs of disease such as tenderness to palpa- comers with low back pain), the precise pathoanatomic cause
tion, paravertebral muscle spasm, and increased pain on spi- cannot be determined or when it is determined, no specific
nal motion are absent. treatment is available. These patients are managed conserva-
tively along similar lines, at least initially.
Pearl: The focus of the initial evaluation is to identify the few
patients with systemic disease or significant neurologic Pearl: Pseudoclaudication can be distinguished from claudi-
involvement. cation by elements of the history.
Comment: Patients with underlying systemic disorders or Comment: “Pseudoclaudication” refers to the nonvascular
incipient neurological involvement may require specific origin of pain that occurs in the legs during walking but not
interventions on an urgent basis; e.g., antibiotics for at rest. Pseudoclaudication is also known as neurogenic
450 R. K. Dixit and J. H. Schwab

claudication. This condition is caused by spinal stenosis. lumbar spine is equivalent to that of a daily chest radiograph
Lumbar MRI studies can reveal spinal stenosis clearly, and for several years (Jarvik and Deyo 2002).
arterial Doppler studies are a reliable way of documenting
Myth: Imaging evaluation usually helps to identify the pre-
peripheral arterial insufficiency or confirming adequate
cise cause of low back pain.
blood flow. Nevertheless, astute clinical judgment is required
to interpret the results of such studies in light of the patient’s Reality: A major problem with all imaging modalities is that
symptoms. Radiologic abnormalities often translate into many of the anatomic abnormalities identified are common
findings of trivial clinical importance. in asymptomatic individuals. These abnormalities, usually
The pain from neurogenic claudication is usually located the result of age-related degenerative changes, may have no
in the buttocks and hamstrings. Vascular claudication, in relationship whatsoever to the patient’s chief complaint.
contrast, is more likely to be reported as calf symptoms and Degenerative changes in the spine are among the earliest
may be relieved by a 5 min period of rest that restores the degenerative changes observed in the body. They frequently
tissue oxygen balance. This makes the time course more begin appearing in early adult life. Abnormalities such as
similar to that of coronary angina than to neurogenic claudi- single disc degeneration, facet joint degeneration, Schmorl’s
cation. Neurogenic claudication is also relieved by rest and nodes, spondylolysis, mild spondylolisthesis, transitional
diminished by forward flexion, but complete relief may take vertebrae (the “lumbarization” of S1 or “sacralization” of
longer (10–20 min) because the symptoms are allegedly L5), spina bifida occulta, and mild scoliosis are equally prev-
related to physical compression of small vessels on the sur- alent in persons without low back pain as they are in patients
face of the cord. The classic patient report in neurogenic who have low back symptoms. Thus, the exercise of making
claudication is of improvement that accompanies bending causal inferences based on imaging abnormalities in the
over to push a grocery cart. absence of corresponding clinical findings is hazardous, and
The presence of hair on the toes, which indicates good leads to unnecessary and costly interventions.
local blood supply, makes a vascular etiology of the patient’s
Myth: An MRI study is appropriate for all patients with sig-
complaint less likely. However, the simplest physical exami-
nificant acute back pain.
nation maneuver to distinguish between neurogenic and vas-
cular claudication is to feel the dorsalis pedis pulse. If the Reality: An MRI is not indicated in the evaluation of acute
pulse is present, then vascular claudication is unlikely. low back pain. Several reasons justify restraint in the order-
Unfortunately, expensive imaging studies are ordered even ing of these studies. First of all, the vast majority of patients
before the peripheral pulses are palpated. with this complaint have self-limiting episodes that require
no intervention. Moreover, an MRI is likely to detect multi-
ple “abnormalities” that have no relationship to the patient’s
complaints. Knowledge of these “findings” is not helpful in
45.4 Imaging managing the patient and indeed can be counterproductive

Pearl: Diagnostic testing is rarely indicated in patients with Table 45.1 Red flags
low back pain unless symptoms persist beyond 6 weeks. Symptom or condition Potential diagnosis

Comment: Early imaging studies are required in the small Severe morning stiffness Spondyloarthropathy
Pain improves with exercise, not rest
proportion of patients who have evidence of a significant or
Pain during second half of night
progressive neurologic deficit and those in whom an under-
Alternating buttock pain
lying systemic disease is suspected as the cause of low back Urinary incontinence Cauda equina syndrome
pain (Chou et al. 2007). Table 45.1 lists the features (“red Overflow incontinence
flags”) that suggest the underlying systemic disease or seri- Fecal incontinence
ous neurological involvement. Bilateral or progressive motor deficit
However, the great majority of patients with low back Saddle anesthesia
pain do not need early imaging. Ninety percent of patients Age > 50 years Cancer or infection
History of cancer
with low back pain recover spontaneously within 4 weeks
Unexplained weight loss
and will do so irrespective of whether or not they have had Immunosuppression
lumbosacral radiographs or an MRI. The careful selection of Injection drug use
patients for imaging studies avoids unnecessary testing and Nocturnal pain
the cascade of follow-up studies that can result from the find- Elderly Fracture
ing of “incidentalomas.” It is worth noting that radiation Prolonged glucocorticoid use
exposure to the female gonads from standard views of the Recent trauma
45 Low Back and Neck Pain 451

for a variety of reasons, not the least of which is that some A bulge is a symmetrical, circumferential extension of the
patients fixate on such “abnormalities.” disc material beyond the interspace. A herniation is a focal or
If a patient’s symptoms and signs suggest the possibility asymmetric extension of the disc bulge. Herniations are clas-
of a systemic disease, then plain radiographs and simple sified according to whether they are protrusions or extru-
laboratory tests are more helpful as first steps in defining the sions. Protrusions are broad-based. In contrast, extrusions
etiology (Jarvik and Deyo 2002). have a “neck,” so that the base is narrower than the extruded
material. Bulges (52%) and protrusions (27%) are common
Pearl: Imaging is warranted in a patient with low back pain
in asymptomatic adults but extrusions (1%) are rare (Jarvik
without radiculopathy when the patient’s clinical profile is wor-
and Deyo 2002).
risome for an inflammatory, infectious, or malignant process.
Comment: Patients older than 50 years of age who have low
back pain that persists for more than 6 weeks should have
plain radiographs performed (Jarvik and Deyo 2002). In that 45.5 Disc Herniation
setting, infections and malignancies must be excluded as
causes of the patient’s symptoms. A radiograph should be
Pearl: A negative straight leg raise (SLR) test makes a clini-
the first study, as it provides a significant amount of informa-
cally significant lumbar disc herniation unlikely.
tion at relatively low cost (Chou et al. 2007). By viewing a
plain radiograph first, an experienced clinician can gain a Comment: SLR places tension on the sciatic nerve roots (L4,
sense of the patient’s bone stock, evaluate the spinal align- L5, S1, S2, S3). If any of these nerve roots is irritated such as
ment, detect gross bony destruction, and assess for spinal by impingement from a herniated disc (see Fig. 45.2), further
instability. Additional imaging may be appropriate based on tension on the nerve root by SLR will result in radicular pain
the findings on plain radiography. that extends below the knee.
The great majority of clinically significant disc hernia-
Pearl: Imaging is warranted in the case of low back pain
tions occur at L4–5 or L5–S1. The SLR test is very sensitive
with radiculopathy when the symptoms have persisted for
(95%) but not specific (40%) for clinically significant disc
more than 6 weeks and the patient is a candidate for an inter-
herniations. False-negative tests can be observed at hernia-
vention such as an injection or surgery.
tions above the L4–5 level. The crossed SLR test, with sci-
Comment: Low back pain with radiculopathy is caused com- atica reproduced when the opposite leg is raised, is insensitive
monly by herniated lumbar intervertebral discs. If a patient’s (25%) but highly specific (90%) for disc herniation (Dixit
symptoms persist and the patient is a candidate for an epidu- 2007). The SLR is usually negative in patients with spinal
ral glucocorticoid injection, then an MRI should be per- stenosis.
formed. The clinician must carefully correlate any imaging
Pearl: Bladder incontinence in a patient with sciatica makes
findings with the patient’s history and physical examination
the exclusion of a cauda equina syndrome imperative.
before recommending an intervention.
Comment: Cauda equina syndrome usually occurs from a
Pearl: Cross-sectional imaging studies of the spine are vastly
centrally herniated disc (usually L4–5) that compresses the
overused.
cauda equina. It can also develop secondary to compression
Comment: MRI and computed tomographic (CT) studies of by neoplasia, abscess, or hematoma. Cauda equina syndrome
the spine should be reserved for patients in whom there is a is a true surgical emergency because the neurological results
clinical suspicion of underlying infection or cancer, and for are affected by the time to decompression. Patients usually
patients who are surgical candidates on clinical grounds present with unilateral or bilateral sciatica and motor defi-
(e.g., the presence of a significant or progressive neurologic cits. Sensory loss in the perineum (saddle anesthesia) is com-
deficit). MRI is generally the preferred modality for low back mon. Urinary retention with overflow incontinence is usually
imaging because it is more sensitive for infections and present.
metastatic cancer. When bony anatomy is critical, CT is
Myth: If a patient has clinical and radiographic evidence of
preferable.
radiculopathy from herniated disk, surgery should be per-
Pearl: Disc bulges and herniations are commonly identified formed to relieve the pain.
by CT and MRI studies, even in asymptomatic adults.
Reality: Several prospective, randomized trials have com-
Comment: Disc abnormalities, such as bulges or herniations, pared surgery with nonoperative management for the treat-
protrusions or extrusions, often have little to no relationship ment of herniated lumbar discs (Weinstein et al. 2006a;
with the patient’s symptoms. One should interpret these Weinstein et al. 2006b; Peul et al. 2007). These trials have
“abnormalities” cautiously. shown that surgery improves leg pain and function more
452 R. K. Dixit and J. H. Schwab

quickly than does nonoperative management. However, over Table 45.2 Acquired and congenital causes of lumbar spinal stenosis
time, the differences between the two treatment approaches Acquired causes
disappear. Degenerative
Hypertrophy of the facet joints
These data have been taken by some to mean that surgery
Hypertrophy of the ligamentum flavum
is never indicated for disc herniation leading to radiculopa- Disc herniation
thy. Such a conclusion, however, is not correct in all cases. Spondylolisthesis
The value of more rapid improvement in pain and function Iatrogenic
must be balanced against the risks of surgery. For some Postlaminectomy
patients, depending on a variety of individual circumstances, Postsurgical fusion
the decision to undergo surgery is justified. For others, non- Miscellaneous
operative management is more appropriate. Paget’s disease
Diffuse idiopathic skeletal hyperostosis
Myth: There is never an indication for acute surgical man- Fluorosis
agement of a herniated lumbar disc. Congenital causes
Achondroplastic
Comment: Compression of sacral nerve roots by a herniated Idiopathic
disc can cause a cauda equina syndrome. The most sensitive
symptom of cauda equina syndrome is urinary retention,
which is often followed by overflow urinary incontinence. Deyo 2002). However, the prevalence of symptomatic steno-
Bowel dysfunction and saddle anesthesia are also hallmarks sis is unknown.
of cauda equina syndrome. A cauda equina syndrome is a
Pearl: Degenerative changes in the spine are the cause of lum-
surgical emergency.
bar spinal stenosis in the vast majority of cases (Table 45.2).
A relative indication for surgical decompression is a sig-
nificant loss of motor function in the distribution of the com- Comment: Any herniation of a degenerated disc will narrow
pressed nerve. For instance, if a patient is unable to dorsiflex the anterior part of the spinal canal and a hypertrophied
the foot against gravity, surgery may be considered more degenerated facet joint may result in a narrowing of the lat-
acutely. The word “considered” is important in that last sen- eral recess or intervertebral foramen. Intervertebral discs lose
tence, because no good studies have examined the effect of vertical height as they degenerate. This can result in a buck-
surgery in patients with significant motor dysfunction. Such ling of the now redundant, and often hypertrophied, ligamen-
patients have been excluded deliberately from clinical trials tum flavum resulting in a narrowing of the posterior aspect of
because of ethical concerns. the spinal canal. Thus, a series of degenerative changes in the
spinal column lead to spinal stenosis.
Pearl: The hallmark of spinal stenosis is pseudoclaudication
(neurogenic claudication).
45.6 Spinal Stenosis
Comment: Patients with pseudoclaudication often have bilat-
eral symptoms of pain, weakness, and sometimes paresthe-
Myth: Radicular symptoms indicate a herniated disc.
sias in the buttocks, thighs, and legs. These symptoms are
Reality: Patients with spinal stenosis can also have radicular induced by standing or walking, but relieved by sitting or
leg pain. Severe neurological deficits are rarely seen in spinal flexing forward. Forward flexion increases the canal diame-
stenosis, and the physical examination is usually unimpres- ter and may lead to the adoption of a simian stance. Factors
sive. Unsteadiness of gait is a frequent complaint in patients that favor a diagnosis of pseudoclaudication over vascular
with spinal stenosis. The lumbar component of pain is fre- claudication include the preservation of pedal pulses, provo-
quently mild. Lumbar range of motion may be normal. cation of symptoms by standing just as readily as by walk-
Making the diagnosis is all about taking a good history: ing, and localization of the maximal discomfort to the thighs
listening to the patient and asking the right questions. The rather than to the calves.
diagnosis is confirmed most easily by MRI.
Myth: Spinal stenosis is purely a radiographic diagnosis.
Myth: Lumbar spinal stenosis is rare.
Reality: Clinical spinal stenosis is manifested as neurogenic
Reality: Lumbar spinal stenosis is defined as a narrowing of claudication. Pain radiates into the buttocks and often down
the spinal canal, its lateral recesses, and neural foramina that into the lower leg. Pain is worsened by spinal extension and
may result in compression of neural structures. relieved by flexion. The so-called “shopping cart sign” is a
As many as 20% of asymptomatic adults aged 60 years or useful means by which to assess for neurogenic claudication:
older have imaging evidence of spinal stenosis (Jarvik and many people with spinal stenosis find that they can walk
45 Low Back and Neck Pain 453

much further while leaning forward, as when pushing a shop-

ping cart.
The reason for this improvement is not entirely clear. One
theory is that the spinal canal dimensions are increased when
the spine is flexed forward. Another theory is that venous
congestion along the spinal nerves is relieved by bending
forward and worsened by extending the spine, leading to
pain. Pain caused by neurogenic claudication is sometimes
but not always accompanied by paresthesia or weakness
(Katz and Harris 2008).
The clinical components of spinal stenosis as described
earlier are generally accompanied by radiologic evidence of
this condition. However, the converse is not true: radiologic
evidence of spinal stenosis may be largely or entirely asymp-
tomatic. For example, one study reported an incidence of
asymptomatic spinal stenosis in 20% in individuals over the
age of 60 years (Boden et al. 1990) Distinguishing the clini-
cal components of spinal stenosis from radiologic evidence Fig. 45.4 This standing, lateral, plain radiograph demonstrates slip-
of this condition that is asymptomatic is critical in astute page of L4 onto L5. This is the most common location for degenerative
spondylolisthesis to occur. This particular patient has had a lumbar
management of these patients. laminectomy which provided short-term relief of her leg pain. After
several months, her symptoms of neurologic claudication returned
Pearl: Surgical management for lumbar stenosis is a reason-
able option for those patients who have failed conservative
management. Pearl: Degenerative spondylolisthesis may be missed if
Comment: Two randomized trials have compared nonopera- standing lumbar radiographs are not obtained.
tive management with lumbar laminectomy for the treatment Comment: An unstable spinal segment may be missed if one
of symptomatic lumbar stenosis (Malmivaara et al. 2007; relies upon supine plain radiographs when evaluating a
Weinstein et al. 2008). The first trial showed clinical and sta- patient with neurogenic claudication (Fig. 45.5a, b). Films
tistical improvement at 1 year with regard to back and leg pain obtained while the patient is standing are essential. Similarly,
among patients randomized to receive surgery (Malmivaara if one relies upon an MRI or CT study, both of which obtained
et al. 2007). The difference began to narrow after 2 years. The while the patient is supine, then the dynamic nature of their
second trial also demonstrated statistically significant improve- condition can be overlooked. Neurogenic claudication caused
ment among patients in the surgical arm with regard to leg by spinal stenosis and that caused by spondylolisthesis can-
pain, function, and disability scores. This study was limited by not be distinguished by details of the history and physical
significant crossover, with the result that the as-treated analy- examination. Weight-bearing images are essential to identify
sis probably biased the study in favor of surgery. However, the spondylolisthesis as the cause of neurogenic claudication in
results of an observational cohort study performed in parallel a patient with spinal stenosis.
with the clinical trial support the trial’s conclusions (Weinstein
Pearl: Surgical decompression and fusion for patients with
et al. 2008).
lumbar degenerative spondylolisthesis who present with
neurogenic claudication secondary to spinal stenosis is a
reasonable alternative to nonoperative management.
45.7 Spondylolisthesis
Comment: A prospective, randomized trial compared nonop-
erative management with decompression and fusion for
Myth: Degenerative spondylolisthesis occurs equally in all
patients with lumbar degenerative spondylolisthesis who pre-
ethnic groups and genders.
sented with neurogenic claudication secondary to spinal steno-
Reality: Degenerative spondylolisthesis occurs more com- sis (Weinstein et al. 2007). The authors performed three
monly in women than in men. In fact, this condition is approx- separate analyses. The intention-to-treat analysis failed to
imately four times more common in women. Degenerative show any benefit from surgery. However, because of signifi-
spondylolisthesis nearly always between L4 and L5 is diag- cant crossover that occurred between the two treatment arms –
nosed most easily by a standing lateral radiograph of the lum- with many patients in the nonoperative arm eventually opting
bar spine. The forward slippage of L4 on L5 narrows the for surgery – the authors also performed an “as treated” analy-
spinal canal and can cause spinal stenosis (Fig. 45.4). sis that demonstrated significant improvement in pain,
454 R. K. Dixit and J. H. Schwab

Fig. 45.5 (a). This supine MRI

a b
demonstrates slippage of L4 onto
L5. The degree of slippage is just
over 5 mm. When one compares
the supine MRI with the standing
image in (b), one can appreciate
the dynamic quality of this
problem. The L4 vertebra has
slipped to over 9 mm. The patient
subsequently underwent
decompression and posterolateral
spinal fusion with complete relief
of her symptoms

function, and disability for the patients treated surgically. A Pearl: The thoracic spine is the most common site of involve-
third analysis, performed in an observational cohort, also ment in diffuse idiopathic skeletal hyperostosis (DISH).
showed significant benefits from surgical management in pain,
Comment: DISH is characterized by florid hyperostosis of
function, and disability. This trial is the best study to date in
the spine. Marginal bony proliferation leads to the formation
evaluating the role of surgery in the management of this prob-
of osseous ridges that fuse, giving the appearance of flowing
lem. One can conclude that surgery remains a viable alterna-
wax. Lesions are most prominent anteriorly and along the
tive to continued nonoperative management in patients who
right lateral aspect of the spine. Involvement of the left lat-
have neurogenic claudication as a result of spondylolisthesis.
eral aspect in patients with situs inversus has lead to specula-
tion that the descending thoracic aorta plays a role in the
location of calcification.
The thoracic spine is invariably involved in DISH, but the
45.8 Miscellaneous but Important cervical and lumbar regions can be involved, as well.
Causes of Low Back Pain Intervertebral disc spaces are preserved and the facet joints
appear normal unless there is co-existing lumbar spondylo-
sis. The absence of sacroiliac joint involvement helps to dif-
Pearl: Severe, acute mechanical low back pain in a post-
ferentiate DISH from spondyloarthropathies.
menopausal woman should trigger suspicion of an osteo-
porotic vertebral compression fracture.
Comment: This is particularly true if there are additional risk
factors for osteoporotic fractures such as a history of a previ-
ous fracture, a family history of osteoporosis, or frailty. The
45.9 Other Approaches to Evaluation
pain associated with osteoporotic fractures is generally self- and Treatment
limited, with resolution over a few weeks. Kyphoplasty and
vertebroplasty, two technically demanding procedures with Myth: Provocative discography is a reliable test for identify-
the potential for complications, are rarely needed for the ing “pain generators” in the low back prior to the correction
treatment of osteoporotic compression fractures. of such problems through surgery or other means.
45 Low Back and Neck Pain 455

Reality: Provocative discography involves the injection of Table 45.3 Unproven, controversial, or downright ineffective therapies
dye into an intervertebral disc and then assessing the charac- for low back pain
ter of the induced pain. The assumption is that if the injection Trigger point injections
Sacroiliac joint injections
reproduces a patient’s usual low back pain, then that disc
Facet joint injections
must be the cause of the problem. These patients are given a Ultrasound
diagnosis of “internal disc disruption” or labeled as having Shortwave diathermy
“discogenic low back pain.” Transcutaneous electrical nerve stimulation (TENS)
The validity of this technique is controversial. The injec- Lumbar braces
tion of dye into a disc can simulate the quality and location Traction
of pain known not to originate from that disc. Furthermore, Acupuncture
Biofeedback
disc injections are painful 30–80% of the time for patients
Prolotherapy
who do not have symptomatic disc disease but who have had Radiofrequency denervation of nerves to the facet joints
previous disc surgery or who have psychological distress, Spinal cord stimulation
remote chronic pain, or disputed compensation claims Intraspinal drug infusion systems for intrathecal delivery
(Carragee 2005). The interpretation of provocative discogra- of morphine
phy procedures should be viewed with caution. Intradiscal electrothermal annuloplasty (IDET)

Pearl: Laboratory studies play a limited role in the investiga-

tion of low back pain. Limited data suggest that epidural injections in patients
with lumbar spinal stenosis temporarily relieve leg pain.
Comment: Laboratory studies are used primarily to identify
They do not, however, influence functional status or subse-
patients with systemic causes of low back pain. For patients
quent need for surgery (Katz and Harris 2008). There is no
whose findings suggest systemic disease or who have back
evidence that lumbar epidural injections are effective for
pain that has not improved after 6 weeks of conservative care,
acute or chronic low back pain in the absence of radicular
a normal complete blood cell count, erythrocyte sedimenta-
symptoms.
tion rate, and lumbar spine radiographs exclude systemic dis-
For a problem in which effective interventions are few,
ease with a high degree of likelihood (Jarvik and Deyo 2002).
many patients are willing to consider a broad range of treat-
Myth: In patients with acute low back pain, bed rest is an ment options. In some cases, the potential for the temporary
important part of the treatment program. relief of pain may justify an epidural glucocorticoid injec-
tion. Expectations about the long-term benefit of the proce-
Reality: Bed rest should be discouraged. Continuing ordi-
dure, however, should be realistic.
nary activities within the limits permitted by pain leads to
more rapid recovery than bed rest (Malmivaara et al. 1995). Pearl: The goals for the treatment of patients with chronic
low back pain are relief of the pain and restoration of
Myth: A large number of therapies, including physical ther-
function.
apy modalities, injection techniques, and invasive proce-
dures, are effective in the treatment of low back pain. Comment: The majority of patients with chronic low back
pain continue working. However, the results of treatment are
Reality: Multiple modalities are now employed for the man-
often unsatisfactory, and complete relief of pain is an unreal-
agement of low back pain. Most have not been tested rigorously.
istic goal for most.
Uncontrolled studies can produce a misleading impression of
Many patients respond to a program that includes analge-
efficacy in a condition with fluctuating symptoms and a favor-
sic agents, education, back exercises, aerobic conditioning,
able natural history. A list of therapies that are unproven,
and the loss of excess weight. Low-dose tricyclic antidepres-
controversial, or downright ineffective (but still commonly
sants (e.g., amitriptyline 10–75 mg at bedtime) may help
employed) is shown in Table 45.3.
some patients, but anticholinergic side effects are common.
Myth: Lumbar epidural glucocorticoid injections are helpful Some data support intensive rehabilitation programs that
for a broad spectrum of patients with low back pain. emphasize cognitive behavioral therapy.
Reality: The use of epidural glucocorticoid injections is
increasing remarkably rapidly despite the lack of consistent
evidence regarding the efficacy of this procedure. Epidural
glucocorticoid injections may offer temporary symptomatic 45.10 Surgery
relief in some patients with radicular pain, but appear to offer
no significant functional benefit and do not reduce the need Myth: Patients with chronic low back pain who fail conser-
for surgery. vative care are good candidates for surgery.
456 R. K. Dixit and J. H. Schwab

Table 45.4 Indications for surgery in spinal stenosis, spondylolisthesis, various parts of the vertebra including the lamina, ligamen-
and disc herniation tum flavum, enlarged facets, osteophytes, and bulging disc
Spinal stenosis material.
Severe or progressive neurologic deficit
Disabling pseudoclaudication or sciatica (elective)
Spondylolisthesis
Severe or progressive neurologic deficit
Disabling pseudoclaudication or sciatica (elective) 45.11 Neck Pain
Disc herniation
Cauda equina syndrome (emergency)
Severe or progressive neurologic deficit Pearl: There are a remarkable number of similarities in the
Greater than 6 weeks of sciatica (elective) clinical features and management of patients with neck pain
and low back pain.

Reality: The pathophysiology of pain in degenerative disease Comment: These include the following statements regarding
of the lumbar spine remains elusive, and a definitive patho- neck pain:
anatomic cause of back pain cannot be identified in most
• The natural history of neck pain is favorable in the major-
patients. Thus, it is not surprising that surgical procedures
ity of patients.
such as spinal fusion or the insertion of artificial discs, aimed
• In most patients with radiant neck pain (to the back of the
at relieving isolated back pain, often yield disappointing
head, shoulder, scapula, or arm), the cause is nonneuro-
results. Spinal fusion surgery is effective for some conditions
genic, sclerotomal pain. This usually arises from pathol-
in some patients, but its efficacy for the most common condi-
ogy within the disc, facet joint, or paraspinal muscles and
tions (e.g., isolated low back pain secondary to degenerative
ligaments. It is not radicular pain secondary to nerve root
disc disease) remains unclear and controversial (Carragee
compression.
2005; Deyo et al. 2004).
• Degenerative changes in the cervical spine (sometimes
Pearl: The major indications for surgery in patients with low termed as cervical spondylosis) are the most commonly
back pain are the presence of a severe neurological deficit or identified cause of neck pain. The pain is usually exacer-
a progressive neurological deficit (Table 45.4). bated by movement and relieved by rest.
• A definitive pathoanatomic diagnosis cannot be made in
Comment: Nerve root compression syndromes usually result
most patients with neck pain.
from acute disc herniation, and less often from spinal stenosis
• Early diagnostic testing is rarely indicated unless there is
or spondylolisthesis. The natural history of acute disc hernia-
a significant or progressive neurologic deficit or some
tion is favorable. Sequential MRI testing reveals that the her-
indication that an infection or malignancy might be caus-
niation tends to regress with time and radicular pain resolves.
ing the pain. The major focus of the initial evaluation is
In the absence of a severe or progressive neurologic deficit,
the identification of these few patients, as they need
patients with disc herniations should be managed conserva-
urgent, specific interventions.
tively. Fewer than 10% of these patients require surgical
• Imaging abnormalities identified in patients with neck
decompression. When necessary, decompression is usually
pain are common in asymptomatic individuals. These
accomplished by a laminotomy with limited discectomy.
abnormalities are often the result of age-related degenera-
Elective surgery is often performed in patients who have sci-
tive changes.
atica that does not resolve within 6 weeks (Peul et al. 2007).
• Laboratory tests are useful only in the small minority of
The symptoms of spinal stenosis remain stable for years
patients with a systemic disorder causing neck pain.
in many patients. In the absence of a severe or progressive
• Most patients with neck pain respond to a conservative
neurological deficit, patients are treated conservatively with
treatment program that may include NSAIDS, analgesics,
a program that includes analgesia, loss of excess weight,
muscle relaxants, and range-of-motion and strengthening
physical conditioning, and back exercises designed to reduce
exercises.
lumbar lordosis. Epidural glucocorticoids may provide tem-
• The major surgical indication in patients with neck pain is the
porary symptomatic relief. Patients with disabling pseudo-
presence of a significant or progressive neurologic deficit.
claudication or radicular pain may benefit from decompressive
surgery. The goal of the surgery is to relieve pressure on neu-
Pearl: Most neck pain is mechanical in nature.
ral elements by increasing the area of the spinal canal and
neural foramen. The most common surgical procedure for Comment: Degenerative change in the spine is the most impor-
spinal stenosis is a decompressive laminectomy, sometimes tant contributor to mechanical neck pain. The prevalence of
accompanied by fusion. This procedure involves removing neck pain and degenerative change increases with age.
45 Low Back and Neck Pain 457

Myth: Electrodiagnostic testing should be prescribed for Most patients with whiplash injuries improve within
most patients with neck pain and neurologic symptoms in the 4 weeks of treatment with a combination of analgesics,
upper extremities. NSAIDS, and gentle neck mobilization exercises. However,
in some patients – as many as 30% in some estimates – the
Reality: Electrodiagnostic testing (electromyography and
pain persists for many months or years. The reason for this is
nerve conduction studies) is overused. In most patients with
unclear, particularly because the response to treatment is
neck pain (or low back pain) who have neurological symp-
often unrelated to the severity of the whiplash injury. Patients
toms, the diagnosis is clear to the astute clinician following a
with persistent pain may benefit from a noninvasive multi-
careful history and physical examination. These patients do
disciplinary pain management program. In some patients,
not require electrodiagnostic testing.
the persistence of neck pain is associated with depression,
Electrodiagnostic testing can be helpful when the rela-
the pursuit of disability compensation, or litigation.
tionship between symptoms, physical findings, and imaging
results is unclear. For example, a number of diagnoses could Pearl: An older patient with chronic mechanical neck pain
explain the patient with neck pain, paresthesias of the fifth almost invariably has evidence of cervical spondylosis on
digit, and wasting of the intrinsic muscles of the hand. neck imaging.
Among these are a spinal cord lesion, a C8 radiculopathy, a
Comment: On the other hand, many older patients without
lower trunk brachial plexopathy, and an ulnar neuropathy.
chronic mechanical neck pain also have evidence of cervical
Electrophysiologic findings in this scenario might be essen-
spondylosis on neck imaging! Chronic mechanical neck pain
tial in guiding the clinician to the correct diagnosis.
is common in older people. The prevalence of cervical spon-
Electromyographic abnormalities depend upon the devel-
dylosis, i.e., degenerative changes in the disco-vertebral,
opment of muscle denervation following nerve injury. This
facet, and uncovertebral joints, increases with age. Although
can require 4–6 weeks in the distal musculature.
the neck pain in such patients is often attributed to cervical
Pearl: Acute nonspecific neck pain is a common, benign, and spondylosis, no consistent relationship exists between radio-
self-limited condition. graphic changes – either their presence or their degree – and
the complaint of neck pain, the response to treatment for
Comment: The etiology of pain and spasm in acute cervical
neck pain, or the prognosis of neck pain.
strains, also termed as “myofascial pain,” is unknown. There
Patients with neck pain secondary to cervical spondylosis
is usually no history of significant trauma. Pain is often pre-
often present with a background of chronic discomfort with
dominantly on one side of the neck and may radiate to the
intermittent acute attacks that last a few days. The neck pain
top of shoulder along the superior portion of the trapezius, as
often radiates along the superior portion of the trapezius to
well as inferiorly to the periscapular region. Paracervical
the shoulder, to the scapular region, and to suboccipital areas.
muscle spasm with secondary loss of cervical lordosis is
Neck movement is restricted during most such exacerbations
common. Tender points and trigger points may be present.
of neck pain. Conservative treatment, which includes reas-
Most episodes of acute cervical strain resolve spontane-
surance and education, analgesics, NSAIDS, and neck exer-
ously within 2 weeks, but may recur. No investigation other
cises, is usually effective.
than a careful history and physical examination is required.
Although a small number of patients with cervical spon-
Treatment involves reassurance, analgesics, NSAIDS, gentle
dylosis develop cervical radiculopathy caused by disc herni-
mobilization with the help of local heat, and time.
ation or cervical myelopathy secondary to spinal stenosis,
Myth: A whiplash injury has the same excellent prognosis as most causes of acute neck pain are caused by mechanisms
cervical strain. that remain obscure but have self-limited courses.
Reality: A whiplash is a rapid flexion/hyperextension (accel- Pearl: In a patient with radicular symptoms in the upper
eration/deceleration) injury to the neck. Whiplash usually extremity, a careful neurologic evaluation helps identify the
results from a rear or side-impact car accident. The paracer- specific nerve root involved.
vical muscles, ligaments, and facet joint capsules become
stretched and may sustain injury. In severe cases, the patient Comment: Cervical root irritation generally results from either
develops vertebral fractures or damage to the intervertebral posterolateral disc herniation or osteophyte pressure. The C5–C6
discs. Radiographic evaluation should always be obtained. level is the most common site of disc herniation, and results in a
Neck pain and stiffness develops within a few hours but C6 radiculopathy (As with low back pain, the more caudal nerve
may be delayed until the next day. Patients may also com- root is impinged). A C7 radiculopathy caused by a C6–C7 disc
plain of headaches, dizziness, anxiety, paresthesias, and herniation is the second most common such lesion in the neck.
insomnia. Physical examination usually reveals paracervical Radiculopathy symptoms can be exacerbated by Valsalva
spasm and a decreased range of motion of the cervical spine. maneuvers, which lead to increased intrathecal pressure. The
458 R. K. Dixit and J. H. Schwab

Fig. 45.6 Neurological

evaluation of the upper Upper Extremity Dermatomes Disc Nerve Motor Loss Sensory Reflex
extremities including motor Root Loss Loss
testing, reflex testing, and tests
for dermatomal sensory loss help
to identify the specific nerve root
involved C4–5 C5 Deltoid Lateral Biceps
deltoid

T2
C5–6 C6 Biceps Index finger Supinator,

C5
and thumb Biceps

T1
C6
C6–7 C7 Triceps Middle finger Triceps
(sometimes
index)
C8
C6
C7

C7–T1 C8 Hand* Ring and None

little finger

*Intrinsic muscles of the hand – the lumbricals and the interossei. Weakness of finger flexion,
abduction, and adduction. There may be weakness of grip strength.

Spurling test, a combination of lateral flexion and rotation of Table 45.5 Symptoms and signs of cervical myelopathy
the neck to the affected side, also exacerbates symptoms of Slowly progressive spastic paraparesis
cervical radiculopathy. Brisk deep tendon reflexes in the lower extremities
MRI is the most effective technique for localizing the site of Extensor plantar responses
disc herniation and nerve root impingement. Most patients Diminution of vibratory sense in the legs
respond to conservative care. Surgery is indicated only in patients Gait difficulty
Urinary incontinence
with a significant or progressive neurologic deficit (Fig. 45.6).
Radicular arm pain and other neurologic features
Myth: Cervical myelopathy causes severe neck pain. of cervical radiculopathy

Reality: Cervical myelopathy in and of itself is a painless

condition. Only one-third of patients with cervical myelopa- The natural history of spondylitic cervical myelopathy is
thy complain of neck pain (Borenstein 2007). However, one of gradual progression. Mild cases of myelopathy that
patients with cervical myelopathy can have neck pain or involve only mild arm and hand symptoms may improve
radicular symptoms caused by coexistent cervical spondylo- with conservative care. However, progressive myelopathy
sis or cervical nerve root compression. requires surgical decompression before severe neurologic
Myelopathy is a potentially devastating sequela of cervi- deficits appear.
cal canal stenosis. This condition usually results from cervi-
cal spondylosis. Osteophytes from the discovertebral and
facet joints encroach on the spinal canal. The canal is nar-
References
rowed further by disc herniation and hypertrophy and buck-
ling of the ligamentum flavum.
Cervical canal stenosis leads to compression of the spinal Boden SD, Davis DO, et al Abnormal magnetic-resonance scans of the
lumbar spine in asymptomatic subjects. A prospective investigation.
cord or nerve roots. Individual patients present with symp- J Bone Joint Surg Am. 1990;72(3):403–8
toms that can be predominantly myelopathic, predominantly Borenstein D. Approach to the patient with neck pain. In: Imboden J,
radicular, or both. Patients with cervical myelopathy com- Hellman D, Stone J editors. Current rheumatology diagnosis and
plain of symptoms such as paresthesias in all extremities. treatment. 2nd ed. NY: McGraw Hill; 2007
Brox JI, Sorensen R, et al Randomized clinical trial of lumbar instru-
Symptoms of cervical myelopathy are shown in Table 45.5. mented fusion and cognitive intervention and exercises in patients
The diagnosis of cervical myelopathy should be considered with chronic low back pain and disc degeneration. Spine 2003;28(17):
in a patient with progressive spastic paraparesis, paresthesias 1913–21
of hands and feet, or wasting of the intrinsic hand muscles. Carragee EJ. Persistent low back pain. N Engl J Med. 2005;352:
1891–8
The most efficient means of confirming the diagnosis is Chou R, Qaseem A, et al Diagnosis and treatment of low back pain: a
through an MRI. joint clinical practice guideline from the American College of
45 Low Back and Neck Pain 459

Physicians and the American Pain Society. Ann Intern Med. Malmivaara A, Slatis P, et al Surgical or nonoperative treatment for
2007;147:478–514 lumbar spinal stenosis? A randomized controlled trial. Spine 2007;
Deyo RA, Nachemson A, et al Spinal fusion surgery – the case for 32(1):1–8
restraint. N Engl J Med. 2004;350:722–6 Peul WC, et al Surgery versus prolonged conservative treatment for
Deyo RA, Weinstein JN. Low back pain. N Engl J Med. 2001;344: sciatica. N Engl J Med. 2007;356:2245–56
363–70 Rudwaleit M, Metter A, et al Inflammatory back pain in ankylosing
Dixit RK, Dickson J. Low back pain. In: Adebajo AO, editor. ABC of spondylitis: a reassessment of the clinical history for application as
rheumatology. 4th ed. London: Blackwell Publishing; 2008 classification and diagnostic criteria. Arthritis Rheum. 2006;54:
Dixit RK. Approach to the patient with low back pain. In: Imboden J, 569–78
Hellmann D, Stone J, editors. Current rheumatology diagnosis and Weinstein JN, Lurie JD, et al Surgical versus nonsurgical treatment for
treatment. 2nd ed. NY: McGraw Hill; 2007 lumbar degenerative spondylolisthesis. N Engl J Med. 2007;356(22):
Fritzell P, Hagg O, et al 2001 Volvo Award Winner in Clinical Studies: 2257–70
lumbar fusion versus nonsurgical treatment for chronic low back Weinstein JN, Lurie JD, et al Surgical vs nonoperative treatment for
pain: a multicenter randomized controlled trial from the Swedish lumbar disk herniation: the Spine Patient Outcomes Research
Lumbar Spine Study Group. Spine 2001;26(23):2521–32; discus- Trial (SPORT) observational cohort. JAMA. 2006a;296(20):
sion 2532–4 2451–9
Jarvik JG, Deyo RA. Diagnostic evaluation of low back pain with Weinstein JN, Tosteson TD, et al Surgical versus nonsurgical therapy
emphasis on imaging. Ann Intern Med. 2002;137:586–97 for lumbar spinal stenosis. N Engl J Med. 2008;358(8):
Katz JN, Harris MB. Lumbar spinal stenosis. N Engl J Med. 2008;358: 794–810
818–23 Weinstein JN, Tosteson TD, et al Surgical vs nonoperative treatment
Malmivaara A, Hakkinen U, et al The treatment of acute low back for lumbar disk herniation: the Spine Patient Outcomes Research
pain – bed rest, exercises, or ordinary activity? N Engl J Med. 1995; Trial (SPORT): a randomized trial. JAMA. 2006b;296(20):
332: 351–5 2441–50
 Amyloidosis
46
George H. Sack Jr

46.1 Overview of Amyloidosis › In tropical regions, “Secondary” amyloidosis can be

associated with chronic or episodic inflammatory or
infectious diseases. In such cases, the parent protein is
› The diagnosis of amyloidosis depends on finding
the acute phase reactant serum amyloid A (SAA).
deposits with congo red birefringence in affected tis-
sue.These deposits comprise fibrils of highly ordered › Amyloid deposits in the brain in Alzheimer disease
comprise a small fragment (Called Ab) derived by
proteins or protein fragments that generally contain
cleavage of a much larger precursor molecule called
prominent b-pleated sheet domains.
the b-protein.
› Many proteins have regions that can aggregate into
› Successful treatment of any form of amyloidosis
amyloid fibrils.Often only a small fragment of a larger
depends on eliminating or reducing the available par-
molecule is compatible with fibril geometry and thus
ent protein. For “primary” disease this generally
fibril formation occurs after cleaving the parent
involves chemotherapy; for “secondary” disease the
protein.
underlying infection or inflammatory process must be
› Amyloid deposits can be found in any tissue.However,
controlled.
some depostis have no pathologic consequence while
others can cause serious organ dysfunction by interrupt- › Amyloid deposits are relatively inert chemically and
intrinsically stable. Thus, successful treatment often
ing individual cellular physiology, cell-cell interactions
must be given for an extended time to permit gradual
and/or tissue integrity.
elimination of the fibrils and restoration of organ
› When specific organ dysfunction has been identified,
function.
biopsy of that tissue often is the simplest diagnostic
method(e.g.heart,kidney,liver). An alternative diagnostic
approach for systermic amyloidosis is to examine tissue
derived by aspiration of the abdominal fat pad.
Myth: Amyloid tissue is largely composed of carbohydrate.
› Both prognosis and treatment are based on identifying
the protein in the fibril.Often, this protein can be iden- Reality: The term “amyloid,” introduced by Virchow in 1858,
tified by immunofluorescence. is etymologically incorrect. Amyloid is derived from the Latin
› Amyloid proteins vary widely and include immuno- word for starch, but there is virtually no carbohydrate in amy-
globulins,inflammatory molecules, hormone precur- loid deposits, which are composed of protein. Nevertheless,
sors and inherited variants of normal serum the term has persisted.
proteins.
Myth: Amyloidosis is rare.
› Inherited forms of amyloidosis are uncommon but
well-recognized. They often involve specific organs– Reality: Amyloid deposits are encountered frequently.
e.g. heart,peripheral nerves,or blood vessels. Most Autopsy studies reveal a frequency as high as 25%. However,
have autosomal dominant inheritance. all deposits of amyloid are not necessarily associated with
› In America, the most common form of amyloidosis clinical manifestations.
(called “Primary” amyloidosis) is derived from immuno-
Myth: Amyloid is amorphous.
globulin light chains. This is usually due to overproduc-
tion from a plasma cell clone. The light chain often can be Reality: Although light microscopy is generally unrevealing,
detected by protein electrophoresis of serum or urine; l electron microscopy reveals that all amyloids are composed of
light chains are more frequently involved than k. masses of highly ordered microfibrillar protein assemblies.
b-pleated sheet protein domains are prominent in amyloid

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 461

DOI:10.1007/978-1-84800-934-9_46, © Springer Science + Business Media B.V. 2009
462 G. H. Sack Jr

fibrils. Amyloidosis is a disorder of protein folding. However, deposition of fragments of the serum amyloid A (SAA) pro-
different types of amyloid fibrils differ in their detailed three- tein, the levels of which are elevated as part of an inflammatory
dimensional folding patterns (Dobson 2003). response (Sack et al. 1989).

Myth: Amyloid is a disease of the liver and kidneys. Pearl: The diagnosis of amyloid requires tissue study with
specific staining.
Reality: Amyloid deposits can be found in any organ or sys-
tem. The disease can be both widespread throughout multiple Comment: The definition of amyloid is that it shows apple-
organs or involve individual organs in a localized fashion green birefringence when stained with Congo red dye and
(Biewend et al. 2006; Rocken and Sletten 2003). viewed under cross-polarized light (Fig. 46.1). This is the
case of all types of amyloid, regardless of etiology.
Myth: Amyloid reflects similar pathophysiology wherever it
is found. Pearl: The most important diagnostic and prognostic feature
of amyloid is the type of protein present in the fibrils. This is
Reality: Many different molecular types of amyloid are known
the basis for classifying amyloid disease.
(Table 46.1). These include products of neoplasia (Kyle et al.
2002), inflammatory molecules (Cunnane and Whitehead Comment: Identification of the type of protein within amy-
1999), genetic protein variants (Connors et al. 2003), and local loid fibrils shows the nature of the underlying condition.
deposits. In the U.S., about 90% of the cases of amyloidosis are Moreover, this identification is essential for determining the
due to fragments of immunoglobulin light chains (AL amy- patient’s prognosis and for recommending treatment. The
loid: the “L” stands for light chain). AL amyloid is generally type of protein present within amyloid fibrils can often be
secondary to a monoclonal gammopathy. In parts of the world identified by immunofluorescence, but sequence analysis of
where chronic infections and poorly treated inflammatory dis- the protein or gene is required in some cases to distinguish
eases are common, much of the amyloid is “secondary” to an among inherited forms. Table 46.1 presents the main catego-
underlying process. In such cases, amyloidosis is caused by the ries and their associated nomenclature.

Table 46.1 Amyloid disease nomenclature

Constituent amyloid protein Protein precursor Variant(s) Clinical
AA Serum amyloid A (SAA) Secondary (reactive)
Familial Mediterranean fever (FMF)
Muckle–Wells syndrome
AL k, l light chains Idiopathic (primary), MGUS
Myeloma-associated
Macroglobulin-associated
AH IgG 1 (g1) heavy chain Ag1 Idiopathic (primary)
Myeloma-associated
ATTR Transthyretin >80 mutants Familial amyloid polyneuropathy
Familial amyloid cardiomyopathy
Tenosynovial deposition
AApoA1 apoA1 Arg 26 Familial amyloid polyneuropathy
Aortic deposition
AGel Gelsolin Asn 187 Familial amyloidosis (Finnish)
Ab b protein precursor Gln 618 Alzheimer dementia
Down syndrome
Hereditary cerebral hemorrhage
Aging
Ab2M b2-microglobulin Complication of chronic dialysis
APrP PrP – cellular prion PrPSc, PrPCJD Spongiform encephalopathies
ACal (Pro)calcitonin (Pro)calcitonin Medullary thyroid carcinoma
AANF Atrial natriuretic factor Isolated atrial amyloid
AIAPP Islet amyloid polypeptide Type II diabetes, insulinoma
APro Prolactin Aging pituitary
Prolactinomas
ALys Lysozyme Familial
ACys Cystatin C Familial
AFib Fibrinogen a-chain Familial
46 Amyloidosis 463

a nerves, and heart, all involved commonly, may be appropri-

ate sites for biopsy.
Pearl: Aspiration of the abdominal fat pad offers a minimally
invasive source of tissue for analysis in the absence of a fla-
grantly involved organ.
Comment: Fat pad aspiration, a simple and safe procedure, is
often the first approach to the diagnosis of amyloid disease
(Duston et al. 1989; Olsen et al. 1999). Proper interpretation of
the aspirated material requires an experienced pathologist,
because the tissue planes can become distorted as an artifact of
the procedure. Figure 46.2 shows material from such a study.
Pearl: Several otherwise unusual clinical features make the
b diagnosis of amyloidosis more likely.
Comment: Macroglossia may be prominent, leading to stiff-
ening of the tongue and indentation by the teeth (Fig. 46.3).
This may interfere with eating and cause hoarseness.
Easy bruising with capillary fragility may be prominent.
Purpura in the periorbital tissues is virtually diagnostic of AL
amyloidosis (Fig. 46.4).

Fig. 46.1 (a–c) Serial tissue sections showing deposition of amyloid in

the wall of a blood vessel. (a). conventional hematoxylin and eosin
staining. Note the amorphous quality of the depositions with efface-
ment of normal tissue details. (b). Congo red staining with regular illu-
mination. The amyloid appears pink without distinct features. (c).
Viewing tissue stained with congo red using polarized light shows the
diagnostic “apple-green” birefringence. This is the basis for the diagno-
sis of all amyloid deposits

Pearl: Tissue from the affected organ offers an ideal source

for amyloid diagnosis. Fig. 46.2 Tissue from abdominal fat pad aspiration stained with hema-
toxylin and eosin (above) and congo red (below). Note the prominent
Comment: Clinically evident abnormalities generally guide birefringence on the lower image despite the poorly-defined anatomy
the selection of biopsy site. The liver, kidney, peripheral characteristic of adipose tissue.
464 G. H. Sack Jr

2. A restrictive left ventricular filling pattern with progres-

sive diastolic dysfunction.
3. Prominent valves, consistent with amyloid infiltration.
Myth: The echocardiogram in patients with cardiac amyloi-
dosis has a “sparkling” or “ground-glass” appearance.
Reality: Although often described with earlier technology,
improved image processing has made the “ground-glass”
appearance of the myocardium is less prominent on the new gen-
eration of echocardiography technology. However, increased
echogenicity of both valves and myocardium is notable as
amyloid deposition progresses.
Pearl: In AL amyloidosis, the protein constituent of the amy-
loid fibril is unique for each patient.
Fig. 46.3 Macroglossia is a relatively common finding in AL amyloi- Comment: At least 90% of the cases of amyloidosis diag-
dosis. Chewing, swallowing, breathing, and speech all can be affected
by this change
nosed in the U.S. result from light chain deposition. Lambda
light chains are twice as common as kappa light chains in
cases of amyloidosis, and there is a predominance of the lVI
subgroup. The tight fibril packing characteristic of amyloi-
dosis imposes constraints on the light chains; usually, only a
short portion of the light chain can fit within the b-pleated
sheet. The polypeptide incorporated into the b-pleated sheet
generally includes at least part of the variable region of the
light chain. As a result, only a small fraction of light chains
contribute to the formation of amyloid fibrils. Because the
variable region is commonly involved, the protein constitu-
ent of the fibril is likely unique for each patient. Hence, the
clinical distribution and consequences of AL amyloid are
unpredictable.
In effect, at the molecular level, each individual has
a novel disease. The long lists of “complications” described
conventionally in AL amyloid are unlikely to be encountered
in every patient. The clinician must be prepared for
Fig. 46.4 Periorbital purpura is virtually diagnostic of amyloid infiltra- surprises.
tion. In this case, the changes developed following a colonoscopy:
“postproctocopic purpura” Pearl: Overproduction of the light chain in AL disease may
be due to aggressive plasma cell growth, such as in myeloma.
Alternatively, it may arise from a single plasma cell clone
Pearl: Several clinical clues betray cardiac involvement in
that proliferates without causing the other known complica-
amyloid disease.
tions of myeloma.
Comment: The heart is involved in up to 50% of patients
Comment: The common feature in each case is overproduc-
with AL amyloidosis (Falk 2005). Weight loss, dyspnea, and
tion of a specific light chain that can be cleaved to yield a
edema are frequent presentations. Electrocardiographic
fragment capable of meeting the stringent constraints of fibril
changes include:
formation. This fragment usually includes at least part of the
1. Low QRS voltage (limb leads < 5 mm in height). variable region.
2. Otherwise nonspecific ST- and T-wave changes including
Pearl: The overproduced immunoglobulin in AL amyloidosis
a so-called “pseudoinfarction pattern.”
is generally detectable through immunoelectrophoresis of
3. Conduction disturbances (various forms of block).
serum or urine (i.e., an SPEP or UPEP).
Echocardiography can also help, by revealing:
Comment: Because the overproduction of immunoglobulins
1. An increased estimated left ventricular mass with wall in amyloidosis usually leads to light chain excess, the light
thickening. chains pass into the urine where they become detectable by
46 Amyloidosis 465

electrophoresis. About 80% of AL patients have proteinuria. predisposed to FMF with colchicine minimizes the frequency
Some of the light chains have the solubility characteristics of and severity of attacks, thereby reducing the likelihood of
Bence–Jones proteins. Consequently, any patient suspected renal amyloidosis. The usual dose of colchicine in the treat-
of having amyloidosis should be evaluated for the presence ment of FMF is 0.6 mg taken three times daily.
of Bence–Jones proteinuria.
Pearl: Treating individuals with amyloid disease generally
Pearl: Monoclonal gammopathies are often detected in requires eliminating the source of the aberrant (or excess)
asymptomatic individuals. protein.
Comment: This phenomenon is termed monoclonal gammo- Comment: Amyloid deposits themselves are largely inert and
pathy of unknown significance, or “MGUS.” Individuals with resolve very slowly. Thus, the goals of therapy are to mini-
MGUS represent one end of a broad spectrum of plasma cell mize the expansion or appearance of these deposits elsewhere
proliferation. At the other end of the spectrum is multiple and to attenuate the local consequences of tissue deposition
myeloma (Kyle et al. 2002). A small fraction of people on (Cohen and Kelly 2003). Specific treatments vary according
this spectrum produce an immunoglobulin that can cause to the responsible protein. The approaches to therapy for the
amyloid fibril formation. The ability of these immunoglobu- major types of amyloidosis are shown in Table 46.2.
lins to cause amyloidosis is unrelated to the aggressiveness of
Pearl: Transthyretin (TTR), a small serum protein of 127
the plasma cell biology; neither myeloma nor high quantities
amino acids, has more than 80 known mutations, most of
of aberrant immunoglobulins is required for amyloidosis to
which cause the amyloid disease termed ATTR.
develop. The “amyloidogenicity” is related entirely (it seems)
to the structural features of the individual patient’s protein. Comment: Some TTR mutations (e.g., Val30Met) cause a
Long-term follow-up of individuals with MGUS indicates progressive, debilitating neuropathy. Others (e.g., Thr60Ala)
that progression from a clinical curiosity to more aggressive lead to cardiac failure (Connors et al. 2003). Small molecules
changes occurs at a rate of ~1% per year. Such progression may stabilize the native TTR tetramer to minimize fibril for-
may or may not be associated with amyloid deposition. mation. Identifying the mutation in individuals at risk can be
accomplished by tracing the autosomal dominant transmis-
Pearl: “Secondary” amyloid (AA) can accompany a wide vari-
sion in the patient’s kindred and sequencing the responsible
ety of chronic conditions: rheumatologic, inflammatory, infec-
protein or gene.
tious, and inherited.
Most TTR, including the aberrant forms, is produced by
Comment: The distribution of AA amyloid varies across differ- the liver. Complications of TTR neuropathies have been mit-
ent regions of the world. Still’s disease is a very rare cause of igated by performing orthotopic liver transplantation before
AA amyloid in the U.S., but apparently more common in the patient becomes symptomatic. Cells of the transplanted
England and Europe. AA amyloid secondary to tuberculosis, liver do not have the TTR mutation, and therefore produce
leishmaniasis, and other chronic infections is more frequent in normal TTR protein.
Africa and Southeast Asia. Familial Mediterranean fever
Pearl: The other major site of TTR production is the choroid
(FMF), an autosomal recessive disorder, is encountered primar-
plexus.
ily in individuals from the eastern Mediterranean, from Turkey
to Egypt. Comment: Despite the presence of TTR mutations in chor-
The feature common to all of these conditions is overpro- oid plexus cells, symptoms referable to central nervous sys-
duction of the small serum protein, SAA (Sack et al. 1989). tem involvement are rare. Several isolated kindreds with the
The production of this protein is caused by high levels of
interleukin-6 and tumor necrosis factor. These cytokines are Table 46.2 Treatment approaches to major variants of amyloidosis
the proximal inducers of SAA synthesis in the liver and other Variant Treatment approach
cells.
AL Diminish the production of light chains by plasma
Pearl: Colchicine prevents the development of amyloidosis cells
in FMF. AA Reduce source of chronic inflammation by
treating the underlying infection or rheumato-
Comment: FMF is notable for recurrent episodes of pyrexia, logic disorder
serositis, and serum changes characteristic of the acute phase Inherited Orthotopic liver transplantation when the liver
forms is the source of the mutant protein (e.g., liver
response. The disorder is caused by mutations in the gene for in ATTR)
pyrin, an inflammatory protein expressed in granulocytes. Inhibit synthesis of mutant allele (RNAi)
Untreated homozygous individuals are at risk for renal fail- Colchicine treatment for FMF
ure secondary to cumulative renal insults with deposition of Organ-limited Local excision but removal may not be necessary
AA amyloid. Fortunately, long-term treatment of patients if there is no dysfunction
466 G. H. Sack Jr

so-called “oculoleptomeningeal amyloidosis” have been encephalopathies.” These conditions reflect alternative folding
described. patterns for normal cellular proteins. The transmission of such
alternative structures can lead to fibril propagation within
Pearl: Without a change in the production of the amyloid
neighboring cells. However, some of these disorders develop
constituent protein, clinical progression is likely.
in the presence of specific mutations that enhance the transi-
Comment: Once fibril formation has begun, the physiologic tion to monomer structures compatible with fibril formation
conditions and supplies of monomers are adequate to cause (Caughey and Baron 2006).
sustained tissue deposition. It can be assumed under those cir-
cumstances that fibril formation will be unimpeded. In vitro
studies show that existing fibrils hasten formation of addi-
tional fibrils. This means that clinical signs and symptoms, References
including any evident organ dysfunction, will progress.
Biewend ML, Menke DM, Calamia KT. The spectrum of localized
Pearl: Alzheimer’s dementia is associated with two charac- amyloidosis: a case series of 20 patients and review of the literature.
teristic changes in brain tissue: neurofibrillary tangles and Amyloid 2006;13:135–42
amyloid plaques. Caughey B, Baron GS. Prions and their partners in crime. Nature
2006;443:803–10
Comment: The amyloid plaques are formed from fibrils Cohen FE, Kelly JW. Therapeutic approaches to protein-misfolding
that comprise a small fragment of the so-called “amyloid b diseases. Nature 2003;426:905–9
Connors LH, Lim A, Prokaeva T, Roskens VA, Costello CE. Tabulation
protein.” Hence, this type of amyloid is called Ab (Selkoe
of human transthyretin (TTR) variants, 2003. Amyloid 2003;10:
1989). The small fragment that gets deposited within the 160–84
brain, which is usually a polypeptide 40–42 amino acids Cunnane G, Whitehead AS. Amyloid precursors and amyloidosis in
long, is derived from a much larger protein by specific rheumatoid arthritis. Baill Clin Rheumatol. 1999;13(4):615–28
Dobson CM. Protein folding and misfolding. Nature 2003;426:884–90
cleavages. There are three forms of the precursor molecule,
Duston MA, Skinner M, Meenan RF, Cohen AS. Sensitivity, specificity,
each of which is about 700 amino acids long. and predictive value of abdominal fat aspiration for the diagnosis of
Mutations in the Ab protein itself as well as in the enzymes amyloidosis. Arthritis Rheum. 1989;32:82–5
responsible for the cleavages can be associated with familial Falk HR. Diagnosis and management of the cardiac amyloidoses.
Circulation 2005;112:2047–60
forms of Alzheimer’s dementia, but such mutations are rare.
Kyle RA, Therneau TM, Rajkumar V, et al A long-term study of prog-
Histopathologic changes consistent with Alzheimer’s disease nosis in monoclonal gammopathy of undetermined significance.
are also common among individuals with Down’s syndrome. N Engl J Med. 2002;346:564–9
This is consistent with the fact that the gene for the Ab pro- Olsen KE, Sletten K, Westermark P. The use of subcutaneous fat tissue
for amyloid typing by enzyme-linked immunosorbent assay. Am J
tein is located on chromosome 21; patients with Down’s syn-
Clin Pathol. 1999;111:355–62
drome have three copies of this chromosome (trisomy 21). Rocken C, Sletten K. Amyloid in surgical pathology. Virchows Archiv.
2003;443:3–16
Pearl: Prion disorders involve the formation of amyloid-like Sack GH Jr, Talbot CC Jr, Seuanez H, O’Brien SJ. Molecular analysis of
fibrillar arrays. the human serum amyloid A (SAA) gene family. Scand J Immunol.
1989;29:113–9
Comment: The prion diseases (e.g., Creutzfeldt–Jakob disease) Selkoe DJ. Aging, amyloid, and Alzheimer’s disease. N Engl J Med.
are categorized frequently as “transmissible spongiform 1989;320(22):1484–7
 The Ehlers–Danlos Syndrome
47
Fransiska Malfait and Anne De Paepe

in at least ten different genes have been identified to date:

47.1 Overview of the Ehlers–Danlos COL1A1, COL1A2, COL3A1, COL5A1, COL5A2, PLOD1,
Syndrome ADAMTS2, TNX-B, b4-Gal-T7, and SLC39A13. Almost every
family affected by EDS harbors a unique mutation in one of
› The Ehlers–Danlos syndrome (EDS) comprises a clin- these genes. In addition, there remain many unclassified EDS
ically heterogeneous group of connective tissue dis- variants, the molecular defects of which are unknown.
eases, the principal features of which result from The upshot of this variation is that EDS is even more het-
generalized but variable tissue fragility. erogeneous than the current classification system indicates.
› EDS is a congenital disorder, but the age of onset of The 1997 Villefranche Nosology recognizes only six sub-
symptoms varies from birth to the fifth or even sixth types, a substantial under-representation of the variability of
decade of life. this condition (Beighton et al. 1998) (Table 47.1).
› The major characteristics of EDS are, to a varying
Pearl: EDS is more than a collagenopathy.
degree, present in all EDS subtypes. These include skin
hyperextensibility (see Fig. 47.1), poor wound healing Comment: The recognition of ultrastructural abnormalities of
causing wide, thin, atrophic scars (“cigarette paper collagen fibrils in EDS patients led to the concept that EDS is
scars”) (see Fig. 47.2), easy bruising, joint hypermobil- a disorder of fibrillar collagen metabolism. A number of EDS
ity (see Fig. 47.3) with recurrent dislocations, and gen- subtypes display abnormalities in the expression or structure
eralized tissue fragility. of the fibrillar collagen types I, III, and V. Others result from
› During childhood, patients may present to the pediatri- enzymatic abnormalities in the posttranslational modifica-
cian with complaints as diverse as hypotonia or delayed tion and processing of these collagens (such as lysyl hydrox-
neuromotor development, easy bruising, or abnormal ylases I and procollagen aminoproteinase or ADAMTS2)
wound healing. (see Table 47.1) (Beighton et al. 1998).
› Later in life, patients may present to the hematologist The molecular basis of the common hypermobile variant
due to unexplained bleeding diathesis; to the rheuma- of EDS is unresolved, as is the basis for some other EDS phe-
tologist with complaints of joint hyperlaxity and arth- notypes. Following the identification in EDS of mutations in a
ralgias; to the orthopedic surgeon due to recurrent noncollagenous protein, tenascin-X (TNX), it became clear
dislocations; or to the (vascular) surgeon with arterial that the search for EDS candidate genes should extend beyond
or bowel rupture. collagens and their modifying enzymes (Burch et al. 1997;
Zweers et al. 2003). TNX is regarded widely as a regulator of
in vivo collagen deposition (Mao et al. 2002; Bristow et al.
2005). Other noncollagenous extracellular matrix proteins
Pearl: EDS is a disease with many faces.
have been suggested as candidate molecules for EDS by trans-
Comment: The severity of EDS and the impact of this disorder genic mice studies, especially the group of small, leucine-rich
on the patient’s quality of life differ substantially according proteoglycans (SLRPs) such as decorin, fibromodulin, and
to the different disease subtypes. Inter and intrafamilial vari- lumican (Ameye and Young 2002). These SLRPs are involved
ability in severity of disease expression is encountered in in regulation of collagen fibril diameter and fibril–fibril inter-
many of the EDS subtypes. Moreover, each subtype displays actions. SLRP-deficient mice show clinical and ultrastructural
specific additional clinical manifestations that help to deter- characteristics reminiscent of EDS, but no defects in these
mine the EDS subtype. SLRPs have been associated with human EDS to date.
The EDS is extremely heterogeneous, not only at the clin- Mutations in SLC39A13, a gene that encodes a membrane-
ical level but also at the molecular basis of disease. Mutations bound zinc transporter, were identified in a new EDS subtype,

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 467

DOI:10.1007/978-1-84800-934-9_47, © Springer Science + Business Media B.V. 2009
468 F. Malfait and A. De Paepe

Table 47.1. Proposal for updated EDS-nosology

Clinical subtype Clinical manifestations IP Gene
Cardinal features Additional features
Classic Skin hyperextensibility Easy bruising AD COL5A1
Widened atrophic scarring Molluscoid pseudotumors COL5A2 (~50%)
Smooth and velvety skin Subcutaneous spheroids (Malfait et al. 2005)
Joint hypermobility Muscular hypotonia Unknown (~50%)
Complications of joint hypermobility
Surgical complications
Positive family history
“Classic –like” Skin hyperextensibility Smooth and velvety skin AR TNX-B (Burch et al. 1997)
Easy bruising Complications of joint hypermobility
Joint hypermobility
Hypermobility Generalized joint hypermobility Recurring joint dislocations AD TNX-B (Zweers et al. 2003)
Mild skin involvement (mild skin Chronic joint pain COL5A1 (Malfait et al. 2006a)
hyperextensibility, mild atrophic Positive family history Unknown (majority)
scarring, soft skin)
Vascular Excessive bruising Acrogeria AD COL3A1
Thin, translucent skin Early-onset varicose veins
Arterial/intestinal/ Hypermobility of small joints
Uterine fragility or rupture Tendon and muscle rupture
Characteristic facial appearance Arteriovenous or
Carotid-cavernous sinus fistula
Pneumo (hemo)thorax
Positive family history, sudden death in
close relative(s)
“Vascular-like” Skin hyperextensibility Thin, translucent skin AD COL1A1 (Malfait et al. 2007)
Atrophic scarring Joint hypermobility
Easy bruising Complications of joint hypermobility
Arterial rupture Positive family history
“Cardiac- Severe cardiac valvular disease Atrophic scarring AR COL1A2 (Malfait et al. 2006b)
Valvular” Skin hyperextensibility Easy bruising
Joint hypermobility Bluish sclerae
Increased bone fragility
Kyphoscoliotic Severe muscular hypotonia at birth Tissue fragility, including atrophic scars AR PLOD1
Generalized joint laxity Easy bruising Unknown
Kyphoscoliosis at birth Arterial rupture
Scleral fragility and rupture of the Marfanoid habitus
globe Microcornea
Bluish sclerae
Osteopenia
EDS/OI overlap Joint hypermobility Easy bruising AD COL1A1
syndrome Mild skin involvement Recurring joint dislocations COL1A2 (Cabral et al. 2005)
Blue sclera Kyphoscoliosis
Increased bone fragility Chronic joint pain
Arterial rupture
Positive family history
Arthrochalasis Severe generalized joint hypermobility Easy bruising AD COL1A1
with recurrent subluxations Muscular hypotonia COL1A2
Congenital bilateral hip dislocation Kyphoscoliosis
Skin hyperextensibility Mild osteopenia
Widened, atrophic scars
Dermatosparaxis Severe skin fragility Soft, doughy skin texture AR ADAMTS−2
Sagging, redundant skin Premature rupture of membranes
Excessive bruising Progressive
Characteristic facies Paperscar-formation
Postnatal growth deficiency, short Progressive generalized joint
hands and feet hypermobility
Umbilical hernia Delayed gross motor development
Delayed closure of fontanels Palmar wrinkling
Bladder rupture
Rupture of diaphragm
47 The Ehlers–Danlos Syndrome 469

Table 47.1. (Continued)

Clinical subtype Clinical manifestations IP Gene
Cardinal features Additional features
Progeroid Mental retardation Variable skin involvement AR b4Gal-T7
Wrinkled, loose skin on the face Easy bruising
Curly fine hair Joint hypermobility
Scanty eyebrows and eyelashes
Periodontotic Severe early-onset periodontal disease Variable skin involvement AD Unknown
Discrete, chronically inflamed pretibial Easy bruising
plaques Joint hypermobility
Spondylo-cheiro Postnatal growth retardation Tissue fragility, including atrophic scars AR SLC39A13 (Giunta et al. 2008)
dysplastic Hypermobility of small joints, later Bluish sclerae
form contractures Osteopenia
Slender, tapering fingers, palmar
wrinkling, (hypo)thenar atrophy
Platyspondyly
Widened metaphyses
The presence of more than 1 cardinal feature is highly suggestive for the diagnosis of a specific subtype of EDS, while the additional features
contribute to the diagnosis of the specific subtype
IP inheritance pattern; AD autosomal dominant; AR autosomal recessive

characterized by EDS-like features and a mild skeletal dys-

plasia (Giunta et al. 2008). This shows that genes other than
the fibrillar collagens and their known modifying enzymes
participate in the complex process of collagen fibrillogen-
esis and the organization of extracellular matrix architecture.
Given that many EDS subtypes are still unexplained, still
other undiscovered noncollagen genes probably account for
the clinical manifestations in some EDS patients.
Myth: EDS is a rare disorder.
Reality: The prevalence of EDS is not known with any preci-
sion. Beighton (1970) estimated a prevalence of 1 in 156,000
in southern England. Others, however, have only claimed that
EDS is the most common heritable disorder of connective tis-
sue. The perceived incidence is likely to increase as awareness
of the phenotypic variability increases among clinicians. Fig. 47.1 Skin hyperextensibility in the Ehlers–Danlos syndrome
Some of the severe subtypes, such as the arthrochalasis,
kyphoscoliosis, and dermatosparaxis subtype, are indeed
rare. The prevalence of “classic” EDS has been estimated at Comment: Not every double-jointed individual has EDS.
1:20,000 (Byers 2001). However, some individuals with mild The cutaneous features of a patient’s disease are often the
disease manifestations do not come to medical attention and key to the diagnosis. A careful physical examination with
therefore remain subclinical (or at least undiagnosed). The particular attention to the skin is critical to making the cor-
minimum prevalence of vascular EDS is on the order of 1 in rect diagnosis. Patients with EDS usually have soft, smooth
250,000 (Pepin and Byers 2008). However, because many skin that has a velvety, doughy feel. Some degree of skin
families with the vascular type of EDS are identified only hyperextensibility is typical (Fig. 47.1). Cutaneous scars in
after a severe complication or death, proportion of individu- EDS have an atrophic appearance.
als or families with a milder phenotype who do not come to In the vascular subtype of EDS, the skin is not hyperex-
medical attention is probably considerably higher. tensible but rather thin and transparent. A venous pattern is
The prevalence of the hypermobility syndrome is unknown. evident over the chest, abdomen, and extremities. Other clas-
Estimates range from 1:5,000–1:20,000 (Levy 2008), but this sic physical examination features of the vascular EDS sub-
is likely to be an underestimate because of the substantial clin- type are widespread ecchymoses and varicose veins.
ical variability of this disorder.
Myth: Ultrastructural abnormalities of the skin are charac-
Pearl: The skin is the window to the diagnosis of EDS. teristic for a specific EDS subtype.
470 F. Malfait and A. De Paepe

Reality: A common Myth surrounding EDS is that the ultra-

structural findings within the dermis can confirm or exclude
a diagnosis of specific EDS subtypes. Ultrastructural exami-
nation of the skin, performed by electron microscopy, usu-
ally reveals abnormalities of collagen fibrillogenesis. These
include irregular, disrupted collagen fibrils (“collagen-flow-
ers”) and variability within the diameter of the individual
collagen fibrils. However, these abnormalities are found in
several EDS variants and are usually insufficient to discrimi-
nate between individual EDS subtypes.
There is one exception to this rule. In the dermatosparaxis
subtype of EDS, a pathognomonic ultrastructural aspect of
the collagen fibril architecture is observed. The finding of
this architecture confirms the clinical diagnosis. Collagen
fibrils in the dermatosparaxis subtype lose their normal
cross-sectional circular aspect and have instead an irregular,
branched, “hieroglyphic” appearance (Pierard and Lapiere
1976; Malfait et al. 2004). Clinical diagnosis remains the pri-
mary means of identifying all EDS subtypes.
Myth: EDS and cutis laxa are related disorders.
Reality: EDS and the cutis laxa syndromes are different disor-
ders clinically and genetically. From the clinical point of view,
the “hyperextensible” skin in EDS must be distinguished from
that of “cutis laxa” syndromes. Hyperextensibility, a charac-
teristic of EDS, refers to skin that extends easily and snaps
back after release. This should be tested at a “neutral site”;
i.e., one not subjected to mechanical forces or scarring. The
volar surface of the forearm is one such neutral site.
Hyperextensibility is quantified by pulling the skin until
resistance is felt (Fig. 47.2). Cutaneous hyperextensibility is
observed to a varying degree in all subtypes of EDS with the
exception of the vascular subtype. (In the vascular subtype,
the skin is thin and transparent.) In the cutis laxa syndromes,
Fig. 47.2 “Cigarette paper” scarring and hemosiderin deposition as a
the skin is redundant: it hangs in loose folds and returns very
result of repetitive trauma on knees and shins of a patient with the clas-
slowly to its former position. The skin in cutis laxa is not sic Ehlers-Danlos syndrome
fragile, and wound healing is normal.
The EDS subtypes are caused by defects in within the
fibrillar collagens (types I, III, or V) or in other proteins necessary to distinguish this disorder from other causes of
involved in collagen fibrillogenesis. In contrast, the cutis bruising. Laboratory investigation of clotting factors, platelet
laxa syndromes result from the loss or fragmentation of the aggregation, and bleeding time is usually normal in patients
elastic fiber network. This is caused by mutations in the elas- with an EDS or related connective tissue disorders (De Paepe
tin gene or other genes that encode proteins related to elastic and Malfait et al. 2004).
fiber development or homeostasis (e.g., the fibulins).
Myth: Patients with EDS can be recognized by their blue
Pearl: EDS should be considered in the differential diagnosis sclerae.
of child abuse.
Reality: Most patients with EDS, especially with the more
Comment: In children with EDS, excessive bruising follow- common subtypes, such as the hypermobile, classic, or vascu-
ing minor trauma is often the presenting complaint to the lar variants of EDS, do not present blue sclerae. Blue sclerae
pediatrician. If pronounced, it sometimes raises suspicion of are commonly present in osteogenesis imperfecta, caused
child abuse. Careful evaluation of medical and family his- by genetic defects in type I collagen. However, blue–grey
tory and rigorous clinical examination with special atten- sclerae are observed in some of the rare EDS subtypes that
tion to subtle skin features that are characteristic for EDS is are associated with defects in the structure or the processing
47 The Ehlers–Danlos Syndrome 471

of type I collagen. As examples, the dermatosparaxis and Pearl: It is unknown whether hypermobile EDS and (benign)
arthrochalasis subtypes have much phenotypic overlap with joint hypermobility syndrome represent the same entity.
osteogenesis imperfecta, and patients with those EDS sub-
Comment: The hypermobility type of EDS is characterized
types can have blue–grey sclerae.
by generalized joint hypermobility and joint instability. This
Pearl: Collagen studies allow confirmation of the diagnosis causes recurrent subluxations and dislocations, and chronic
of certain EDS subtypes. musculoskeletal pain. Skin abnormalities are usually mild
(see Table 47.1). Considerable clinical overlap exists between
Comment: In certain EDS subtypes, biochemical and molec-
the EDS hypermobility subtype and two related disorders,
ular analysis can be very helpful to confirm the diagnosis. To
the (benign) joint hypermobility syndrome, and familial
this purpose, a skin biopsy is required to obtain cultured skin
articular hypermobility syndrome.
fibroblasts. Biochemical study of the collagen types I, III,
The familial articular hypermobility syndrome, formerly
and V includes SDS-polyacrylamide gel electrophoresis of
known as EDS type XI, is an autosomal dominant connective
radiolabeled collagens, extracted from the cultured fibro-
tissue disorder characterized by severe joint hypermobility
blasts. In the vascular subtype of EDS, biochemical analysis
and occasional congenital hip dislocation, but no skin
of type III procollagen identifies more than 95% of all
involvement (Beighton et al. 1988). The joint hypermobility
patients. Molecular screening of the COL3A1 gene identifies
syndrome is also an autosomal dominant connective tissue
virtually all mutations.
disorder, which was first recognized as a distinct condition in
Characterization of a type III collagen defect enables
1967 (Kirk et al. 1967). In the original description of the
asymptomatic, but affected family members to be aware of
disorder, the joint hypermobility syndrome was defined as
their condition, to learn about appropriate treatments for pos-
“the occurrence of musculoskeletal problems in hypermobile
sible complications of the disorder, to restriction high-impact
but otherwise healthy persons.”
sports, and to undergo elective surgery, arteriograms, and
A variety of nonarticular symptoms and signs, including
other possibly risky procedures in an informed manner.
fragility of the skin and other connective tissues, can be asso-
Biochemical analysis can also be helpful in the diagnosis
ciated with the joint hypermobility syndrome. In 1998,
of the arthrochalasis, kyphoscoliosis, dermatosparaxis, and
Grahame and coworkers presented a set of major and minor
some other rare subtypes of EDS, where distinct abnormal
diagnostic criteria for this disorder known as the Brighton
migration patterns of type I (pro)-collagen chains are observed.
criteria (Grahame et al. 2000) (Table 47.2). With the recogni-
These findings guide further molecular analysis of the correct
tion of the presence of mild fragility of connective tissues
genes. When a disease-causing mutation is identified, prenatal
other than the joints, it was suggested that JHS is itself an
and preimplantation genetic diagnosis becomes possible.
Pearl: Genetic counseling is an essential part of EDS
management. Table 47.2 1998 Revised Brighton criteria for benign joint
Comment: Provision of genetic counseling is an important hypermobility syndrome (BJHS)
issue in the management of patients with EDS. The different Major Criteria
subtypes of EDS are associated with differences in modes of Beighton score of ³ 4/9 (either historically or currently)
inheritance, subtype-specific risks, and variable long-term arthralgia for > 3 months in 4 or more joints
prognoses. Unnecessary or unrealistic fear for certain risks Minor criteria
in patients with specific EDS subtypes, misunderstandings Beighton score of 1–3 (0–3 if aged ³ 50+)
Arthralgia (>3 months) in 1–3 joints or back pain (>3 months),
and wrong perceptions, are sometimes the result of wrong spondylosis, or spondylolysis/spondylolisthesis
information, or the wrong interpretation of it, as supplied by Dislocation/subluxation in more than one joint, or in one joint on
various Internet sites or by the media. more than one occasion
Families should be provided with detailed information Three or more soft tissue lesions (e.g., epicondylitis, bursitis,
regarding the inheritance pattern, risk of recurrence, and identi- tenosynovitis)
Marfanoid habitus (tall, slim, span/height ratio > 1.03, upper/lower
fication of at-risk family members. Screening of other individu- segment ratio < 0.89, arachnodactyly)
als in the family for subtle signs and symptoms of the condition Skin striae, hyperextensibility, thin skin or abnormal scarring
should be offered, regardless of whether signs or symptoms are Eye signs: drooping eyelids, myopia, or antimongoloid slant
suggested by the family history. The prognosis and natural his- Varicose veins, hernia uterine or uterine/rectal prolapsed
tory of the particular EDS type, as well as the possibility of pre- BJHS is diagnosed in the presence of two major criteria, or one major
natal of preimplantation genetic diagnosis, should be discussed and two minor criteria, or four minor criteria. Two minor criteria will
suffice where there is an unequivocally affected first-degree relative.
with the family. Another key element in the genetic counseling BJHS is excluded in the presence of Marfan syndrome or EDS other
process includes triage toward appropriate medical services and than hypermobility type. From Grahame et al. (2000). Reprinted with
provision of information about national patient support groups. permission from The Journal of Rheumatology
472 F. Malfait and A. De Paepe

fatigue and sleep disturbance; autonomic dysfunction, causing

symptoms such as atypical chest pain, palpitations (at rest
and/or upon exercise), dizziness, faintness, syncope or pre-
syncope, postural orthostatic tachycardia syndrome, orthos-
tatic hypotension, and gastrointestinal disturbance (Gazit
et al. 2003; Hakim and Grahame 2004).
This apparent autonomic dysfunction has several potential
explanations, such as exaggerated blood pooling in the lower
extremities due to weakened vascular tissue elasticity; impaired
peripheral vasoregulation due to neuronal or adrenoreceptor
abnormalities; impaired central sympathetic control; or decon-
ditioning due to muscle disuse through pain or fear of pain
Fig. 47.3 Joint hypermobility in the EDS
(Rowe et al. 1999; Gazit et al. 2003; Hakim and Grahame
2004).
under-recognized heritable disorder of connective tissue. Autonomic dysfunction is also frequently encountered in
Some authorities consider the joint hypermobility syndrome fibromyalgia and chronic fatigue syndrome, two conditions
and hypermobile EDS to be essentially indistinguishable if that show many overlapping symptoms with EDS hypermo-
not identical disorders (Grahame 1999). bility type and JHS. Many patients with EDS hypermobility
have been diagnosed with fibromyalgia or chronic fatigue
Pearl: The hypermobility subtype of EDS is more than a cir-
syndrome (or both) prior to the recognition of their joint
cus act.
hypermobility. It remains currently unclear whether presence
Comment: Because the hypermobility subtype of EDS is not of chronic pain, autonomic dysfunction, and other overlap-
associated with major skin manifestations or vascular fragil- ping symptoms in these populations are the result of a com-
ity, it is generally considered the least severe of all EDS sub- mon pathway, or merely reflects the co-occurrence of these
types. Joint hypermobility is common in the general three relatively common syndromes.
population, but in the majority of individuals it does not The impact of this wide spectrum of symptoms can have
cause major complaints. Joint hypermobility is often consid- a devastating impact on sleep, work, physical activities, and
ered a curiosity rather than a medical problem, by clinicians social and sexual relationships of affected patients. This
as well as the general public (Fig. 47.3). leads to the patients being misunderstood or misdiagnosed as
Joint hypermobility continues to be under-recognized, having psychosocial problems, hypochondriasis, or depres-
poorly understood, and managed inadequately. However, sion. Hypermobility is also associated with anxiety and panic
significant complications of this disorder can occur. attack (Bulbena et al. 1993; Martin-Santos et al. 1998; Hakim
Subluxations and dislocations are common and can be very and Grahame 2003).
painful. The predominant presenting complaint is pain. This
Myth: Mitral valve prolapse is encountered frequently in
is often widespread and chronic, leading to major physical
EDS.
and psychosocial distress (Sacheti et al. 1997). It is variable
in age of onset, number of sites, duration, quality, severity, Reality: Mitral valve prolapse was considered at one time to
and response to therapy. There is often a discrepancy between be a manifestation common to all subtypes of EDS. However,
the extent of pain and the paucity of objective signs on physi- this has not been confirmed in rigorous studies (Dolan et al.
cal or radiological examination. In addition to the pain, there 1997). Mitral valve prolapse (and, less frequently, tricuspid
are many other joint symptoms such as stiffness, clicking valve prolapse) appears to be limited to the more severe
and popping of the joints, the feeling of joint instability or forms of classic EDS (McDonnell et al. 2006).
weakness, and many individuals describe the condition as
Pearl: Patients with vascular EDS require a special approach.
being imprisoned in an “old” or “dysfunctional” body.
Comment: Some prophylactic measures are critical for
Pearl: Hypermobile EDS: it is not all about the joints!
patients with the vascular subtype of EDS. The vascular sub-
Comment: In addition to the articular problems, patients with type is dominated by generalized vascular fragility that
EDS hypermobility type may suffer from a wide variety of affects large vessels on both sides of the circulatory system,
extraarticular manifestations. Complications of soft connec- as well as small arteries, veins, and capillaries.
tive tissue fragility are shown in Tables 47.1 and 47.2. Other Vascular fragility and bleeding may occur at any site in the
problems include proprioception difficulties (Hall et al. body and can present in various ways: acute abdominal pain,
1995; Ferrell et al. 2004); poor responses to local anesthet- stroke, hemoptysis, hematemesis, renal colic and hematuria,
ics, e.g., during dental care (Arendt-Nielsen et al. 1991); retroperitoneal bleeding, muscular swelling, shock, and sudden
47 The Ehlers–Danlos Syndrome 473

death. The most common locations of life-threatening arterial • Dermal wounds should be closed without tension, preferably
bleeding are in the abdominal cavity, where they involve in two layers. Deep stitches should be applied generously.
medium-sized arteries such as the renal or splenic arteries. The • Cutaneous stitches should be left in place twice as long as
aorta itself is seldom involved. In some individuals, there is evi- usual.
dence of aneurysmal degeneration and dissection or of arterio- • Fixation of skin adjacent to stitches with adhesive tape
venous fistulae, but in other patients ruptures occur at locations can help prevent stretching of the scar.
that appear completely normal by angiography (Steinmann
et al. 2002). The necessity for periodic arterial screening is
therefore uncertain and controversial. Some authors advice not
to pursue any additional investigation or surveillance, because 47.3 Bruising
a conservative approach will be used irrespective of the find-
ings. Others recommend noninvasive surveillance by means of • Patients with pronounced bruising should avoid contact
echocardiogram, carotid ultrasound, and noninvasive imag- sports and heavy exercise (Pepin and Byers 2008).
ing of chest and abdomen, because knowledge of incidental • Protective pads and bandages are also useful in the pre-
lesions may favorably affect survival (Oderich et al. 2005). vention of bruises and hematomas.
Invasive vascular procedures such as arteriography and • Supplementation of ascorbic acid, a cofactor for cross-
catheterization in these patients should be avoided because linking of collagen fibrils, ameliorates the tendency
of the risk of vascular ruptures (Cikrit et al. 1987; Freeman toward bruising in some patients (Steinmann et al. 2002).
et al. 1996). Whenever possible, invasive vascular procedures • DDAVP (desamino-D-arginin-vasopressin) may be useful
should be replaced by ultrasonography, computed tomogra- in EDS patients with chronic bruising or epistaxis, or
phy, or magnetic resonance angiography. perioperatively (e.g., for tooth extraction), in whom bleed-
Surgical interventions are discouraged in the vascular ing time is normalized by DDAVP (Stine and Becton
subtype of EDS. If surgery is unavoidable, the surgeon 1997).
should be aware of the diagnosis. Manipulation of vascular
and other tissues should be done with extreme care. Finally,
patients with vascular EDS should refrain from anticoagula-
tion therapy and from drugs that interfere with platelet func- 47.4 Joint Protection
tion (Steinmann et al. 2002).
• In children with hypotonia and delayed motor develop-
Myth: Patients with vascular EDS should not get pregnant.
ment, a physiotherapeutic program is important. Nonweight-
Reality: Pregnancy for women with the vascular type of EDS is bearing muscular exercise, such as swimming, is useful to
a high-risk venture. The risk of maternal death is as high as 12% promote muscular development, strength, and coordination.
from uterine rupture or peripartum arterial rupture (Pepin et al. • Patients should avoid excessive or repetitive heavy lifting
2000). On the other hand, some women with vascular EDS and other movements that produce undue strain on the
have had one or several successful pregnancies prior to the rec- already hypermobile joints.
ognition of their underlying condition. It is not known whether • They should not “show off” by demonstrating their joint
elective cesarean sections are preferred to vaginal deliveries. laxity to others. Excessive stretching of the joints further
exacerbates the underlying disorder.
Myth: There is no treatment for EDS.
Reality: No treatment for the underlying defect is available
presently for EDS. However, a series of “preventive” guide-
lines are applicable to all subtypes of this disorder. These are 47.5 Emotional Support
listed below in a series of Pearls related to skin care, bruis-
ing, joint protection, and emotional support. • Emotional support and behavioral and psychological
therapy may be indicated in all subtypes of EDS to accept
and cope with the handicap. Patient support groups are
available and can be very beneficial.
47.2 Skin Care
Pearl: Some but not all patients with EDS should have an
annual echocardiogram.
• Patients with EDS should be instructed to avoid undue trauma
to the skin and other organ systems. Children with pronounced Comment: A baseline echocardiogram should be performed
skin fragility should wear protective pads or bandages over in all patients with EDS to evaluate aortic root diameter, as
the forehead, knees, and shins from an early age. adjusted for age and body surface area, and to evaluate
474 F. Malfait and A. De Paepe

cardiac valvular abnormalities. Because longitudinal data on Beighton P. The Ehlers–Danlos syndrome. London: William Heinemann
progression of aortic dilation are not available, specific rec- Medical Books; 1970
Beighton P, De Paepe A, Danks D, et al International nosology of heri-
ommendations for follow up in individuals with a normal table disorders of connective tissue, Berlin, 1986. Am J Med Genet.
aortic diameter do not exist. However, if no abnormalities are 1988;29:581–94
found on echocardiogram in an adult, a follow-up echocar- Beighton P, De Paepe A, Steinmann B, et al Ehlers–Danlos syndromes:
diogram is probably not necessary. For children and adoles- revised nosology, Villefranche, 1997. Ehlers–Danlos National
Foundation (USA) and Ehlers–Danlos Support Group (UK). Am J
cents, it is reasonable to repeat the echocardiogram, Med Genet. 1998;77:31–7
approximately every 3 years until adulthood. Annual echocar- Bristow J, Carey W, Egging D, et al Tenascin-X, collagen, elastin, and
diograms are warranted only if an abnormality such as aortic the Ehlers–Danlos syndrome. Am J Med Genet C Semin Med Genet.
dilatation or mitral valve prolapse is present. 2005;139:24–30
Bulbena A, Duro JC, Porta M, et al Anxiety disorders in the joint hyper-
Pearl: Patients with EDS require a multidisciplinary mobility syndrome. Psychiatry Res. 1993;46:59–68
Burch GH, Gong Y, Liu W, et al Tenascin-X deficiency is associated
approach.
with Ehlers–Danlos syndrome. Nat Genet. 1997;17:104–8
Comment: As EDS affects many systems in the body, a num- Byers P. Disorders of collagen biosynthesis and structure. In: Scriver CR,
Beaudet AR, Sly WS and Valle D, editors. The metabolic and molec-
ber of specialists may need to be involved, depending on the ular bases of inherited disease. 2nd ed. Edinbugrh: Churchill
severity of the symptoms of an individual. Each patient will Livingstone; 2001. p. 1065–81
have an individual set of clinical problems, requiring a treat- Cabral WA, Makareeva E, Colige A, et al Mutations near amino end of
ment plan that is tailored to his or her particular needs. One alpha 1(I) collagen cause combined OI/EDS by interference with
N-propeptide processing. J Biol Chem. 2005;280:19259–69
doctor should have a coordinating role ensuring that all areas Cikrit DF, Miles JH, Silver D. Spontaneous arterial perforation: the
of the syndrome are covered, preferably the clinical geneticist, Ehlers–Danlos specter. J Vasc Surg. 1987;5:248–55
who will also provide genetic counseling. The key aspects of De Paepe A, Malfait F. Bleeding and bruising in patients with Ehlers–
management include cardiovascular work-up, physiotherapy, Danlos syndrome and other collagen vascular disorders. Br J
Haematol. 2004;127:491–500
pain management, and psychological follow up. Dolan AL, Mishra MB, Chambers JB, et al Clinical and echocardio-
Besides prescription of physiotherapy, the rheumatologist graphic survey of the Ehlers–Danlos syndrome. Br J Rheumatol.
and/or physical therapist may also provide assisting devices 1997;36:459–62
such as braces, ring splints, soft collar necks, while the occu- Ferrell WR, Tennant N, Sturrock RD, et al Amelioration of symptoms
by enhancement of proprioception in patients with joint hypermo-
pational therapist may provide tools that make the living and bility syndrome. Arthritis Rheum. 2004;50:3323–8
working environment more user-friendly for the patient. Freeman RK, Swegle J, Sise MJ. The surgical complications of Ehlers–
Many patients will have undergone one or more orthopedic Danlos syndrome. Am Surg. 1996;62:869–73
procedures prior to the diagnosis of EDS, such as arthro- Gazit Y, Nahir AM, Grahame R, et al Dysautonomia in the joint hyper-
mobility syndrome. Am J Med. 2003;115:33–40
plasty and capsulorraphy. The degree of joint stabilization, Giunta C, Elcioglu NH, Albrecht B, et al Spondylocheiro dysplastic
pain reduction, and duration of improvement are usually far form of the Ehlers–Danlos syndrome – an autosomal-recessive
less in patients with EDS than in those without this disorder. entity caused by mutations in the zinc transporter gene SLC39A13.
It is often appropriate to delay surgery in EDS patients in Am J Hum Genet. 2008;82:1290–305
Grahame R. Joint hypermobility and genetic collagen disorders: are
favor of physical therapy and bracing. they related? Arch Dis Child. 1999;80:188–91
Pain management is also very important. Pain medication Grahame R, Bird HA, Child A. The revised (Brighton 1998) criteria for
should be tailored to the individual’s subjective symptoms. the diagnosis of benign joint hypermobility syndrome (BJHS).
Cognitive behavioral therapy can be beneficial in patients J Rheumatol. 2000;27:1777–9
Hakim A, Grahame R. Joint hypermobility. Best Pract Res Clin
with hypermobility and chronic pain. Finally, psychological Rheumatol. 2003;17:989–1004
follow up designed to explore coping strategies and to recog- Hakim AJ, Grahame R. Non-musculoskeletal symptoms in joint hyper-
nize and treat depression is of utmost importance. mobility syndrome. Indirect evidence for autonomic dysfunction.
Rheumatology (Oxford) 2004;43:1194–5
Hall MG, Ferrell WR, Sturrock RD, et al The effect of the hypermobil-
ity syndrome on knee joint proprioception. Br J Rheumatol. 1995;
34:121–5
References Kirk JA, Ansell BM, Bywaters EG. The hypermobility syndrome.
Musculoskeletal complaints associated with generalized joint
hypermobility. Ann Rheum Dis. 1967;26:419–25
Ameye L, Young MF. Mice deficient in small leucine-rich proteogly- Levy HP. Ehlers–Danlos syndrome, hypermobility type. http://www.
cans: novel in vivo models for osteoporosis, osteoarthritis, Ehlers– genetests.org. Accessed September 16, 2008
Danlos syndrome, muscular dystrophy, and corneal diseases. Malfait F, Coucke P, Symoens S, et al The molecular basis of classic
Glycobiology 2002;12:107R–16R Ehlers–Danlos syndrome: a comprehensive study of biochemical and
Arendt-Nielsen L, Kaalund S, Hogsaa B, et al The response to local anaes- molecular findings in 48 unrelated patients. Hum Mutat. 2005;25:
thetics (EMLA-cream) as a clinical test to diagnose between hypermo- 28–37
bility and Ehlers Danlos type III syndrome. Scand J Rheumatol. Malfait F, De Coster P, Hausser I, et al The natural history, including
1991;20:190–5 orofacial features of three patients with Ehlers–Danlos syndrome,
47 The Ehlers–Danlos Syndrome 475

dermatosparaxis type (EDS type VIIC). Am J Med Genet A. Oderich GS, Panneton JM, Bower TC, et al The spectrum, management
2004;131:18–28 and clinical outcome of Ehlers–Danlos syndrome type IV: a 30-year
Malfait F, Hakim AJ, A De Paepe A, et al The genetic basis of the joint experience. J Vasc Surg. 2005;42:98–106
hypermobility syndromes. Rheumatology (Oxford). 2006a;45: Pepin M, Byers PH. Ehlers–Danlos syndrome, vascular type. . http://
502–7 www.genetests.org. Accessed September 16, 2008
Malfait F, Symoens S, Coucke P, et al Total absence of the alpha2(I) Pepin M, Schwarze U, Superti-Furga A, et al Clinical and genetic fea-
chain of collagen type I causes a rare form of Ehlers–Danlos syn- tures of Ehlers–Danlos syndrome type IV, the vascular type. N Engl
drome with hypermobility and propensity to cardiac valvular prob- J Med. 2000;342:673–80
lems. J Med Genet. 2006b;43:e36 Pierard GE, Lapiere M. Skin in dermatosparaxis. Dermal microarchitec-
Malfait F, Symoens S, De Backer J, et al Three arginine to cysteine ture and biomechanical properties. J Invest Dermatol. 1976;66: 2–7
substitutions in the pro-alpha (I)-collagen chain cause Ehlers– Rowe PC, Barron DF, Calkins H, et al Orthostatic intolerance and
Danlos syndrome with a propensity to arterial rupture in early adult- chronic fatigue syndrome associated with Ehlers–Danlos syndrome.
hood. Hum Mutat. 2007;28:396–405 J Pediatr. 1999;135:494–9
Mao JR, Taylor G, Dean WB, et al Tenascin-X deficiency mimics Sacheti A, Szemere J, Bernstein B, et al Chronic pain is a manifestation
Ehlers–Danlos syndrome in mice through alteration of collagen of the Ehlers–Danlos syndrome. J Pain Symptom Manage. 1997;14:
deposition. Nat Genet. 2002;30:421–5 88–93
Martin-Santos R, Bulbena A, Porta M, et al Association between joint Steinmann B, Royce P, Superti-Furga A. The Ehlers–Danlos syndrome.
hypermobility syndrome and panic disorder. Am J Psychiatry. In: Royce P and Steinmann B, editors. Connective tissue and its heri-
1998;155:1578–83 table disorders. 2nd ed. New York: Wiley-Liss; 2002. p. 431–523
McDonnell NB, Gorman BL, Mandel KW, et al Echocardiographic Stine KC, Becton DL. DDAVP therapy controls bleeding in Ehlers–
findings in classical and hypermobile Ehlers–Danlos syndromes. Danlos syndrome. J Pediatr Hematol Oncol. 1997;19:156–8
Am J Med Genet A. 2006;140:129–36 Zweers MC, Bristow J, Steijlen PM, et al Haploinsufficiency of TNXB
McKusick VA In: McKusick VA, editor. Heritable disorders of connec- is associated with hypermobility type of Ehlers–Danlos syndrome.
tive tissue. 1st ed. Saint Louis: Mosby; 1972. p. 292–371 Am J Hum Genet. 2003;73:214–7
 Osteonecrosis
48
Michael G. Zywiel, Mike S. McGrath, and Michael A. Mont

of osteonecrosis. Thus, osteonecrosis is in the differential

48.1 Overview of Osteonecrosis diagnosis for any patient who has a substantial history of glu-
cocorticoid use and the new onset of joint pain.
› Osteonecrosis, literally “bone death,” is the loss of via- Plain radiographs do not detect osteonecrosis at an early
ble bone tissue stage, the time at which treatment is most effective. Thus, if
› Up to 10% of all hip replacement procedures in the a significant question of osteonecrosis exists, MRI should be
U.S. are performed for osteonecrosis. performed to exclude or diagnose the condition definitively
› Major risk factors for osteonecrosis include glucocor- (Fig. 48.1).
ticoid use, alcohol, cigarette smoking, systemic lupus
Pearl: Smoking tobacco rivals alcohol consumption as a risk
erythematosus (SLE), scuba diving, and major hip
factor for secondary osteonecrosis.
trauma, leading to disruption of the vascular supply to
the femoral head. Comment: Glucocorticoid use, excessive alcohol consump-
› Patients who have identifiable risk factors for osteone- tion, and SLE are cited most frequently as the top risk factors
crosis are said to have secondary osteonecrosis. for secondary osteonecrosis. However, cigarette smoking is
› Idiopathic osteonecrosis occurs in some individuals also a significant risk factor, and a history of smoking increases
without overt risk factors for the disorder. suspicion for the diagnosis of secondary osteonecrosis in
› Clinical outcomes are better in patients who are treated patients with a compatible clinical presentation.
at early stages of the disease.
Myth: Osteonecrosis can develop after minor trauma to the hip.
› Patients who have large osteonecrotic lesions or femoral
head collapse at the time of presentation will typically Reality: Minor trauma can lead to subchondral fracture and
progress to end-stage degenerative disease requiring spontaneous focal bone necrosis in patients who have mark-
joint arthroplasty. edly compromised bone strength, but it is not a causative fac-
› Magnetic resonance imaging (MRI) is the study of tor for true osteonecrosis. On the other hand, the femoral head
choice for patients in whom there is a high index of is susceptible to circulatory disruption after major trauma
suspicion of osteonecrosis. because its blood flow originates in the relatively narrow neck
› Core decompression is a relatively noninvasive joint of the femur. Dislocation of the hip can compromise the
preservation technique that may be effective if applied medial femoral circumflex artery, which traverses the poste-
to patients with early disease. rior superior aspect of the femoral neck. Disruption of the
› Hip resurfacing is a joint arthroplasty technique that medial femoral circumflex artery reduces or eliminates perfu-
requires less resection of native bone compared with sion to the femoral head and leads to osteonecrosis. Subcapital
traditional joint arthroplasty procedures. femoral fractures, particularly when displaced, also carry a
high risk of disruption of the blood supply to the femoral head
and the development of osteonecrosis (Fig. 48.2).
Pearl: A patient treated with high-dose glucocorticoids who Pearl: Minor trauma actually helps some patients by high-
develops pain in the area of a single major joint should lighting the presence of early, preexisting osteonecrosis.
undergo screening of the affected area for osteonecrosis.
Comment: Early stages of osteonecrosis are marked often by
Comment: Glucocorticoid use is implicated in almost 70% of the gradual, insidious onset of pain in the affected joint.
advanced cases of secondary osteonecrosis (Zywiel et al. Because it increases slowly and often seems more irritating
2009). Some data are also consistent with a correlation than worrying, patients frequently delay seeking medical
between cumulative glucocorticoid use and the development care. Minor trauma sometimes brings patients fortuitously to

J. H. Stone (ed.), A Clinician's Pearls and Myths in Rheumatology, 477

DOI:10.1007/978-1-84800-934-9_48, © Springer Science + Business Media B.V. 2009
478 M. G. Zywiel et al.

Fig. 48.2 Blood is primarily supplied to the femoral head in retrograde

flow from the posterior (P) and anterior (A) femoral circumflex arteries.
Subcapital fractures and/or dislocations of the femoral head can easily
disrupt this supply

osteonecrosis of large joints share common pathologic feature

of dead necrotic bone, they are otherwise unrelated. There is
no need to screen for jaw lesions in patients with osteonecrosis
of large joints.
Myth: Patients who have secondary osteonecrosis should
undergo bone scintigraphy to detect occult disease foci in
other joints.
Reality: Bone scintigraphy has poor sensitivity for osteone-
Fig. 48.1 Comparison of the appearance of a stage I osteonecrotic crotic lesions and is therefore of limited value. One study of
lesion on plain radiograph (a) and T1-weighted MRI (b) bone scintigraphy reported that this modality identified
only 50% of symptomatic foci (Mont et al. 2008). Bone
scans also not effective for detecting lesions at an early
medical attention and provides an opportunity to make the stage, rendering them particularly unsuitable for use as a
diagnosis of osteonecrosis at an early stage. Thus, rather than screening tool.
cause osteonecrosis, minor trauma can actually lead to an A screening strategy that is more effective than the perfor-
earlier diagnosis of the condition. mance of bone scintigraphy is to remain vigilant for any new
onset of pain or ache in the large joints of patients at risk for
Myth: Patients with osteonecrosis of large joints should be
osteonecrosis. Such patients should be screened with MRI of
screened for osteonecrosis of the jaw.
the symptomatic joint upon the first appearance of symptoms.
Reality: Osteonecrosis of the jaw was described as early as the
Pearl: All patients who present with secondary osteonecrosis
mid-19th century. “Phossy jaw” – necrosis of the mandible
should be referred for MRI scans of the hips.
induced by phosphorus exposure in match-factory workers – is
a memorable historic example (Anonymous 1867). A tremen- Comment: Up to 90% of patients who present with symp-
dous resurgence of interest in jaw osteonecrosis has accompa- tomatic osteonecrosis of one or more joints other than the hip
nied reports of this disorder in association with bisphosphonate have osteonecrotic lesions of the femoral head. Thus, MRI
treatment (Marx 2003). Although osteonecrosis of the jaw and studies of the hips should be ordered for any patient who has
48 Osteonecrosis 479

secondary osteonecrosis in a nonhip joint (e.g., the knee or Similar clinical scenarios have been reported in the femo-
shoulder) (Mitchell et al. 1986). ral head (Legroux Gerot et al. 2004). Stress fractures of the
femoral head have also been reported in younger patients
Myth: All patients with secondary osteonecrosis should be
who have experienced sudden, dramatic increases in physical
followed with serial MRIs.
activity levels (e.g., military conscripts) (Lee et al. 2009).
Reality: Once a patient is diagnosed with secondary osteone-
Pearl: The bone marrow edema syndrome is not true oste-
crosis, MRI adds little to the management of the condition.
onecrosis and has a more benign course generally than
Two primary treatment concerns exist in osteonecrosis. First,
osteonecrosis.
the patient’s clinical symptoms of pain and any associated
functional limitations must be managed. Second, progres- Comment: The bone marrow edema syndrome is a disorder
sion of the disease to subchondral collapse must be delayed distinct from secondary osteonecrosis. Intra-osseous edema
or prevented whenever possible. Radiographs or computed occurs in both syndromes, but the elevated intra-osseous
tomography scans are both excellent modalities for the mon- pressure in the bone marrow edema syndrome does not lead
itoring of bony collapse and progressive joint degeneration. to bone death. Mild analgesia and protective weight bearing
They are also substantially cheaper than MRI. as required for pain is generally sufficient treatment, with
edema and pain typically resolving within a few months of
Pearl: Insufficiency fractures of the knee can mimic the clini-
the onset of symptoms.
cal and radiographic signs of secondary osteonecrosis.
Pearl: Early, appropriate therapy is the key to achieving
Comment: Subchondral stress fractures of a femoral condyle
optimal clinical outcomes in osteonecrosis.
sometimes present with pain and a lesion that suggests bone
necrosis on plain films or MRI. Although the initial presenta- Comment: Treating osteonecrosis in early stages of the disease,
tion can be similar to secondary osteonecrosis, this is a con- before subchondral collapse, can result in good clinical out-
dition with a different etiology marked by the absence of true comes. Nonoperative treatment does not prevent disease pro-
bone necrosis and a relatively benign course. Symptoms gression in most cases. One study demonstrated, for example,
resolve in the majority of patients who are treated nonopera- that nearly 80% of patients developed femoral head collapse
tively with protected weight bearing. Known as spontaneous within 4 years of their osteonecrosis diagnosis (Mont et al.
osteonecrosis of the knee (SPONK), this condition is charac- 1996). A proposed treatment algorithm is shown in Fig. 48.3.
terized by the sudden onset of pain in older individuals who Core decompression is an excellent treatment option for
do not have classic risk factors for secondary osteonecrosis osteonecrosis in which the femoral head has not collapsed.
(Zywiel et al. 2009). Radiographic imaging shows a single The contemporary percutaneous approach employs a small-
epiphyseal lesion in most cases. diameter drilling technique through the skin (Figs. 48.4a–c).

Symptoms?

Asymptomatic Symptomatic

Radiographs? Radiographs?

Normal Positive MRI Normal Positive MRI

Observation Pharmacologic Reconsider Collapse?

Treatment or Diagnosis
Core Decompression

Postcollapse Precollapse

Depression Size? Core Decompression and

Optional
-Bone Grafting
> 2mm < 2mm -Pharmacologic Treatment
-Ancillary use of growth
factors
Acetabular Involvement Bone Grafting or
Osteotomy
Yes No
Fig. 48.3 Proposed treatment algorithm
for secondary osteonecrosis (from Seyler
MOM Resurfacing or Limited Femoral
et al. 2008). Reprinted with kind permission Total Joint Arthroplasty Resurfacing
from Springer Science + Business Media
480 M. G. Zywiel et al.

Fig. 48.4 Core decompression of the a b

femoral head using modern percutane-
ous techniques (a) under fluoroscopic
control (b) is a quick and relatively
simple procedure than can be
performed on an outpatient basis. The
resulting skin wound (c); indicated by
arrowhead can be closed with an
adhesive bandage

One study reported good clinical outcomes in three-fourths

of patient who had a mean follow-up of 11 years (Mont et al.
1997). In contrast, patients who present with collapse of the
femoral head invariably progress to total joint replacement.
Pearl: Core decompression is a minimally invasive proce-
dure that can be performed on an outpatient basis.
Comment: Core decompression for osteonecrosis is not a
completely risk-free procedure. It does carry the hazards of
surgical anesthesia and the remote possibility of inducing
joint damage. The early iterations of this technique involved
significant surgical disruption of the bony architecture with a
large-diameter trephine (Fig. 48.5). In contrast, modern core
decompression is performed percutaneously using a small-
diameter drill bit (usually 3.2 mm). The procedure requires
only minutes to complete and is usually conducted in the Fig. 48.5 Traditional core decompression used a 6–10-mm trephine
outpatient setting. The wound is closed with a self-adhesive (8 mm shown, right), removing more of the patient’s bone than modern
percutaneous techniques, which are often performed using a 3.2-mm
bandage, and patients can bear weight and ambulate as toler- drill bit (left)
ated as soon as they have recuperated from the effects of the
surgical anesthesia. Many patients report almost immediate
pain reduction. decompression are small, but the absence of any potential
benefit makes core decompression an inappropriate treat-
Myth: Core decompression should be attempted in all
ment choice for patients who have already suffered femoral
patients who have osteonecrosis.
head collapse or those with large lesions.
Reality: Once osteonecrosis has progressed to the point of
Myth: The cessation of glucocorticoid use is a necessary pre-
femoral head collapse, core decompression will not halt fur-
requisite for successful treatment of secondary osteonecrosis.
ther disease progression. In addition, patients with very large
lesions are at extremely high risk of subchondral collapse, Reality: Glucocorticoid use is the strongest risk factor for
regardless of therapy. The risks associated with core osteonecrosis, but is neither a prerequisite for the condition
48 Osteonecrosis 481

a b

Fig. 48.6 Antero-posterior radiograph of stage IV osteonecrosis (a), with collapse of the femoral head, and degenerative changes of the acetabu-
lum. This patient was treated with total hip arthroplasty (b)

nor an absolute impediment to the successful treatment of the

problem. Tapering and even discontinuation of glucocorti-
coids are worthy goals (under most circumstances), but effec- B – Femoral
tive control of some disorders may require their continued component
use. Core decompression promotes healing of the necrotic
lesion regardless of the patient’s glucocorticoid dose.
Pearl: Patients with hip osteonecrosis who have degenerative
changes in the acetabulum have only one treatment option:
arthroplasty.
Comment: Degenerative changes in the acetabular surface in
a patient who has osteonecrosis lead to joint destruction.
They are therefore the acetabular equivalent of femoral head
collapse. The only option for restoring the patient’s joint
function and mobility in this setting is total joint arthroplasty
(Figs. 48.6a, b). A – Acetabular
component
Pearl: Hip resurfacing is an effective intervention for treat-
ing femoral head collapse.
Fig. 48.7 Modern hip resurfacing implant, showing the metal-on-metal
Comment: Secondary osteonecrosis tends to affect younger articulation of the acetabular (a) and femoral (b) components
patients compared to spontaneous disease. Thus, joint arthro-
plasty for those who require it is usually performed at a rela-
tively young age (e.g., 45 years of age (Johansson, Zywiel, comparable to that of total hip arthroplasty for both osteone-
Marker et al. unpublished data) ). Many patients eventually crosis and osteoarthritis (Mont et al. 2006; Revell et al. 2006)
require revision surgery – a more challenging procedure than
Myth: Hip resurfacing is associated with high rates of
the initial arthroplasty.
implant failure.
Hip resurfacing, a femoral bone-preserving hip arthroplasty
technique, is particularly desirable in this setting. Hip resur- Reality: Modern hip resurfacing implants combine a thin-
facing replaces only the affected portion of the femoral head shelled, cementless metal acetabular component that articu-
and acetabulum, retaining as much of the patient’s native bone lates directly with a large-diameter, stemmed, cemented
stock as possible. The efficacy of hip resurfacing appears metal bearing (Fig. 48.7). This technology is fundamentally
482 M. G. Zywiel et al.

Greater differences are observed among patients with dif-

ferent risk factors for osteonecrosis. Patients with sickle cell
disease or end-stage renal disease appear to have higher rates
of failure and revision compared to those without such risk fac-
tors (Johansson, Zywiel, Marker et al. Unpublished Data).
Pearl: Patients with multi-focal osteonecrosis that involves
six or more joints are less likely to progress to articular sur-
face collapse than are those with less widespread disease.
Comment: Secondary osteonecrosis presents with a wide
range of clinical and functional severity. Some patients pres-
ent with a single joint lesions that cause only mild discomfort
and never experience another symptomatic disease recur-
rence. Others develop painful lesions that limit function pro-
foundly in several large joints at once. Patients in the latter
Fig. 48.8 Antero-posterior radiograph of a patient with total hip arthro- group undoubtedly experience more profound effects on their
plasty of the left side, which necessitated removal of significantly more daily lives as a result of the disease, in the authors’ experience
of the patient’s bone as compared to the total hip resurfacing of the right
side
such patients progress to joint collapse less frequently than
patients who have less disseminated disease.
Pearl: Patients who have osteonecrosis can have a normal life.
different from the older metal-on-polyethylene implants that
were associated with high failure rates. Joint survival rates in Comment: For some patients, the diagnosis of secondary
hip resurfacing procedures are similar or even superior to osteonecrosis represents the start of profound life changes,
those of standard total hip arthroplasty over 10 years (Buergi marking the beginning of a relentless process that progresses
and Walter 2007). to pain and decreased function of large joints and requiring
Hip resurfacing requires markedly less resection of native multiple primary and revision arthroplasties. However, the
bone compared to traditional hip arthroplasty procedures large majority of patients who have secondary osteonecrosis
(Fig. 48.8). In addition, some data suggest higher postopera- lead very normal lives. Modern joint-preserving surgical treat-
tive range-of-motion and activity levels in patients treated ments such as core decompression have good long-term
with hip resurfacing (Pollard et al. 2006; Vail et al. 2006;
Table 48.1 Spontaneous vs. secondary osteonecrosis
Mont et al. 2009; Zywiel et al. 2009).
Characteristic Spontaneous Secondary
Pearl: Small, precollapse lesions have a good clinical prog- osteonecrosis osteonecrosis
nosis. Large, postcollapse lesions are likely to require Age 55 and greater Below 45
arthroplasty. Gender 3:1 ♀:♂ ♀:♂with SLE as
associated factor;
Comment: Several different staging and lesions sizing sys- ♂:♀with alcohol as
tems have been reported for describing osteonecrotic lesions. associated factor
The clinically relevant radiographic factors are: (1) whether Onset of pain Sudden Gradual
Bilaterality <5% >80%
the lesions are pre or postcollapse and (2) their size.
Number of lesions One Multiple
Regardless of what system is used, small precollapse lesions
Location on bone Epiphysis Epiphysis, metaphysis,
are amenable to nonoperative or joint-preserving surgical and diaphysis
treatment and generally have good outcomes. In contrast, Number of joints Single hip or knee Can simultaneouly
large postcollapse lesions almost always require one joint involved joint affect multiple
arthroplasty technique or another. knees, shoulders,
ankles
Myth: Secondary osteonecrosis is a predictor of poor out- Associated factors None Corticosteroids,
comes in total hip arthroplasty. alcohol, tobacco,
other
Reality: Arthroplasty techniques for osteonecrosis have been Associated None SLE, sickle cell,
diseases caisson disease,
associated historically with less favorable results compared
Gaucher’s disease,
to patients who underwent the procedure for osteoarthritis. thrombophilia,
More recent data indicate that the outcomes for these two hypofibrinolysis
groups are similar if modern implant designs and second- Pathology Fibrotic bone, Necrotic bone
generation cementing techniques are employed. healing fracture
48 Osteonecrosis 483

outcomes when used during early disease stages. Even patients Mont MA, Carbone JJ, Fairbank AC. Core decompression versus non-
who have late-stage disease of the hips or knees can expect operative management for osteonecrosis of the hip. Clin Orthop
Relat Res. 1996;324:169–78
excellent results with modern total joint arthroplasty tech- Mont MA, Marker DR, Smith JM, et al Resurfacing is comparable to
niques (Table 48.1). total hip arthroplasty at short-term follow-up. Clin Orthop Relat
Res. 2009;467:66–71
Mont MA, Seyler TM, Marker DR, et al Use of metal-on-metal total hip
References resurfacing for the treatment of osteonecrosis of the femoral head.
J Bone Joint Surg Am. 2006;88(Suppl 3):90–7
Mont MA, Tomek IM, Hungerford DS. Core decompression for avascu-
Anonymous. Necrosis of the lower jaw in makers of Lucifer matches. lar necrosis of the distal femur: long term followup. Clin Orthop
Am J Dent Sci. 1867;1(series 3):96–7 Relat Res. 1997;334:124–30
Buergi ML, Walter WL. Hip resurfacing arthroplasty: the Australian Mont MA, Ulrich SD, Seyler TM, et al Bone scanning of limited value
experience. J Arthroplasty 2007;22(7 Suppl 3):61–5 for diagnosis of symptomatic oligofocal and multifocal osteonecro-
Lee YK, Yoo JJ, Koo KH, et al Collapsed subchondral fatigue fracture sis. J Rheumatol. 2008;8(35):1629–34
of the femoral head. Orthop Clin N Am 2009;40:259–65 Pollard TC, Baker RP, Eastaugh-Waring SJ, Bannister GC. Treatment
Legroux Gerot I, Demondion X, Louville AB, et al Subchondral frac- of the young active patient with osteoarthritis of the hip. A five- to
tures of the femoral head: a review of seven cases. Joint Bone Spine seven-year comparison of hybrid total hip arthroplasty and metal-
2004;2(71):131–5 on-metal resurfacing. J Bone Joint Surg Br. 2006;5(88): 592–600
Seyler TM, Marker D, Mont MA. Osteonecrosis. In: Klippel JH, Revell MP, McBryde CW, Bhatnagar S, et al Metal-on-metal hip resur-
Stone JH, Crofford LJ, White PH, eds. Primer on the rheumatic facing in osteonecrosis of the femoral head. J Bone Joint Surg Am.
diseases. 13th ed. New York: Springer Science+Business Media; 2006;88(Suppl 3):98–103
2008:565–572 Vail TP, Mina CA, Yergler JD, Pietrobon R. Metal-on-metal hip resur-
Seyler TM, Marker D, Mont MA. Osteonecrosis. In: Primer on the facing compares favorably with THA at 2 years followup. Clin
Rheumatic Diseases. Klippel JH, Stone JH, Crofford LJ, White Orthop Relat Res. 2006;453:123–31
PH. Springer 2008 Zywiel MG, McGrath MS, Seyler TM, et al Osteonecrosis of the knee:
Marx RE. Pamidronate (Aredia) and zoledronate (Zometa) induced a review of three disorders. Orthop Clin North Am. 2009;40:
avascular necrosis of the jaws: a growing epidemic. J Oral Maxillofac 193–211
Surg. 2003;9(61):1115–7 Zywiel MG, Marker DR, McGrath MS, Delanois RE, Mont MA.
Mitchell MD, Kundel HL, Steinberg ME, et al Avascular necrosis of the Resurfacing matched to standard total hip arthroplasty by preopera-
hip: comparison of MR, CT, and scintigraphy. AJR Am J Roentgenol. tive activity levels - a comparison of postoperative outcomes. Bull
1986;1(147):67–71 NYU Hosp Jt Dis 2009;67(2):116–9
 Index

A TNF blockers, 62
Adult-onset still’s disease (AOSD) Anterior ischemic optic neuropathy (AION)
autoinflammatory syndromes, 23 arteritic-AION, 290
etiology infections, 25 fluorescein angiogram, 304
juvenile rheumatoid arthritis (JRA), 23 vasculitic occlusion, 289
nonsteriodal anti-inflammatory drugs (NSAIDs), 25 Antineutrophil cytoplasmic antibodies (ANCA),
serum ferritin level, 26 18, 21, 263–264
Alzheimer’s dementia, 470 Antinuclear antibody assay (ANA), 77, 80, 131–132, 197, 277, 381
American College of Rheumatology (ACR), 1, 131, 132, 146, 163, Antiphospholipid syndrome (APS)
229, 253, 373, 430 antiphospholipid antibodies (aPL)
Amyloidosis immune thrombocytopenia (ITP), 176
Alzheimer’s dementia, 470 transverse myelitis, 175
apple-green birefringence, 467 central nervous system manifestations, 174
cardiac involvement, 468 cerebral vasculitis, 177
diagnostic and prognostic feature, 466 classification criteria, 176
Familial Mediterranean fever (FMF), 469 clinical manifestations, 174
fat pad aspiration, 467 cutaneous manifestation (see Livedo reticularis)
macroglossia, 468 dermatologic manifestation (see Livedo racemosa)
monoclonal gammopathies, 469 menorrhagia, 179
nomenclature and tissue study, 466 nephropathy, 176–177
periorbital purpura, 468 pregnancy management, 179
treatment approaches, 469 thrombosis
ANCA-associated vasculitis (AAV) anticoagulation, 178, 179
antithyroid medications, 264 and pregnancy morbidity, 174–175
arthritis and arthralgias, 261 venous thrombosis, 178
chemosis, 254 Arthralgias, 134
epidemiology, 247 Arthritogenic pathogens, 75
kidneys Arthrocentesis
persistent hematuria, 259 bursal fluid characteristics, 439
renal transplantation, 260, 261 coagulopathy and thrombocytopenia, 437
nasal crusting, 249 glucocorticoids injections, 437, 438
nose irrigating device, 250 Rule of Twos, 437
saddle nose, 247–249 synovial fluid count, 438
subglottic stenosis, 250–253 Atrophic papillae, 113
treatment and course, 264–267 Autoimmune lymphoproliferative syndrome (ALPS), 164
vasculitic skin lesion, 261
Ankylosing spondylitis (AS) B
anti-TNF therapy, 62 Bath AS disease activity index (BASDAI), 58, 63, 64
C-reactive protein (CRP) levels, 58 B cell lymphomas, 121, 123, 125, 126
Forestier’s disease, 60 Behçet’s syndrome, 284
HLA-B27 positive, 58 vs. autoinflammatory disorder, 234
injectable glucocorticoid preparations, 62 azathioprine and interferon-alpha treatment, 238
Klebsiella infections, 61 cardinal characteristic, 233
physiotherapy, 63 clinical manifestations, 238–239
prognosis, 57 colchicine treatment, 233
radiographic pattern, 59 vs. connective tissue disorder, 234
radiological progression, 63 cutaneous hyperpigmentation, 235
sacroiliitis, 59 genital ulcer, scrotum, 236, 237
spondyloarthritis, 57 intermittent claudication, 237
syndesmophytes, 60 MRI, 239
systemic inflammatory disorders, 63 vs. multiple sclerosis, 239

485
486 Index

papulopustular lesions, 234 Colchicine therapy, 37, 366, 370

parenchymal lung disease, 237 Connective tissue disease syndrome, 80, 115, 133
sex differences, disease phenotype, 237 C-reactive protein (CRP), 58, 122, 148
spondyloarthropathy, 233–234 Cryoglobulinemia
sterile pustules, 235 clinical manifestations, 271
superficial thrombophlebitis, 236 cryocrit correlation, 273
Bell’s palsy, 414 etiologic factor, 274
Benign joint hypermobility syndrome (BJHS), 475 hepatitis C virus (HCV) infections, 271
Bilateral hilar adenopathy, 414–415 mononeuritis multiplex, 273
Birdshot retinochoroidopathy, 385 plasmapheresis, 275
Bone marrow edema syndrome, 483 pulmonary–renal syndrome, 274
Bone scintigraphy, Paget’s disease, 410 renal biopsies, 273
Buerger’s disease. See Thromboangiitis obliterans renal disease, 272
Bullous lupus erythematosus, 139, 140 ribavirin, 274
type II/III cryoglobulinemic vasculitis, 272
C Cutaneous systemic sclerosis, 103
Calcium pyrophosphate dihydrate (CPPD) crystal deposition disease, Cutaneous vasculitis, 116, 228, 280
3, 4
diagnosis D
intracellular vs. extracelluar crystals, 377 Dermatomyositis (DM)
metabolic causes, 377 amyopathic DM, 195
talc and glucocorticoids crystals, 376 cutaneous finding, 194
epidemiology, 377–378 muscle disease, diagnosis, 194
pathophysiology, 378–379 periungual erythema, 195
symptoms and signs vs. polymyositis (PM), 193
clinical clues, 375 skin histopathology, 194
crowned dens syndrome, 373–374 Diffuse idiopathic skeletal hyperostosis (DISH) disease, 60,
joints pain, 374 61, 458
medical and surgical procedures, 376 Diplopia, 288
meniscal cartilage of knee and fibrocartilage of wrist, 375 Disc herniation, 455–456, 460
monosodium urate coexistence, 376 Discoid lupus erythematosus (DLE), 144, 145
polyarticular inflamatory arthritis, 373 Disease-modifying antirheumatic drugs (DMARDs), 8, 9,
tumoral deposits, 376 69, 75
treatment and outcomes, 379 Distal inter-phalangeal (DIP) joint arthritis, 68
Cannabis arteritis, 323 Dual energy X-ray absorptiometry (DXA), 399
Carpal tunnel syndrome (CTS), 93, 443
Central nervous system (CNS) disease, 120, 147 E
Cerebral vein thrombosis, 161 Ehlers-Danlos syndrome (EDS)
Cervical myelopathy, 462 bruising, 477
Chrysotherapy, 9 “cigarette paper” scarring and hemosiderin deposition, 474
Churg–Strauss syndrome (CSS) collagen fibril abnormalities, 471
ANCA-associated vasculitis, 242 collagen studies, 475
capillaritis, 243 EDS-nosology proposal, 472–473
CT scan, lung, 243 emotional support, 477–478
cytotoxic agent treatment, 245 genetic counseling, 475
histopathologic findings, 241 hypermobility type of, 475–476
leukotriene inhibitors, 242 joint protection, 477
lung parenchyma involvement, 243 mitral valve prolapse, 476
lymphadenopathy, 242 skin care, 477
necrotizing and non-necrotizing vasculitis, 242 skin hyperextensibility, 473
nerve infarction, 245 vascular EDS, 477–478
peripheral nerve and myocardium predilection, 244 Enthesitis, 30, 31, 52, 68–69, 74
persistent asthma, 245 Episcleritis, 255
phases of, 241 Erythema elevatum diutinum (EED)
rheumatoid nodules, 244 photomicrographs, 332
tongue wasting, 245 skin lesions of, 331
Clinical disease activity index (CDAI), 6, 7 systemic vasculitis, 332
Cogan’s syndrome Esophageal disease, 84
audiograms, 330 Etanercept therapy, 18
autoantibody profile, 329–330
causing meningitis, 331 F
eye in, 331 Familial cold autoinflammatory syndrome (FCAS), 41
hearing loss, 329 Familial Mediterranean fever (FMF), 469
interstitial keratitis, 330 cardinal clinical findings, 35
ocular inflammation, 330–331 colchicine treatment, 36
Cognitive behavioral therapy, FM, 351 MEFV gene mutation, 34
Index 487

renal amyloidosis, 36 coffee consumption, 361

renal transplantation, 37 diabetes associated, 360
septic joint, 34 low-fat dairy product consumption, 362
systemic AA amyloidosis, 37 metabolic syndrome risk, 360
Fibromyalgia (FM) prevalence and incidence of, 359
aerobic exercise vs. nonexercise controls, 351 purine-rich foods, 361
cognitive behavioral therapy, 351 vitamin C intake, 361–362
diagnosis and etiologies, 349 joint erosions with overhanging edges, 357
disease management, 350 joint inflammation, 358
high glutamate level, 349 synovial fluid examination, 355, 357
history and physical examination, 347 therapeutic response to colchicines, 355
impact on National Health Service, 348 tophaceous deposits, 356
laboratory evaluation, 347 treatment
mood and psychiatric disorders, 345 allopurinol and azathioprine, 367
outcomes in, 350 baby aspirin, 370
sensory stimuli sensitivity, 350 colchicine treatment, 365–366
and sleep disturbance, 347 cyclosporine treatment, 368
specialist care requirement, 349 intensive urate-lowering therapy, 369–370
tender point examinations, 346–347 losartan treatment, 370
Fibro-osseous dysplasia, 45 oral glucocorticoids, 365
Finkelstein’s test, 444 renal and heart transplant recipients, 367–368
Flexor hallucis brevis muscle, 216, 279 serum urate levels, 366–367
Forestier’s disease. See Diffuse idiopathic skeletal hyperostosis tophi in patients, 368
(DISH) disease uric acid in 24hr collection, 356

G H
Giant cell arteritis (GCA), 277 Hemophagocytic syndrome, 25
abundant collateral circulation, 303 Henoch-Schönlein purpura (HSP)
diagnosis and imaging ankle and dorsum swelling, 226
angiography and ultrasonography, 294 gastrointestinal symptoms, 225
fluorodeoxyglucose PET (FDG-PET), 295 glucocorticoid treatment, 226
fundoscopy, 290 immunofluorescence studies, 228
MRI, 294–295 immunopathogenesis, 229
pallid disc edema, 290–291 nonsteroidal anti-inflammatory agents (NSAIDs),
temporal artery biopsy, 296–298 treatment, 226
laboratory testing, 293–294 palpable purpura, 227
methotrexate (MTX), 305 rash of, 225
pathology self-limited disorder, 227
amyloid deposition, 299 skin biopsy, 225
atherosclerosis, 300 Hip resurfacing implant, 485
frozen section analysis, 301 Hormone replacement therapy, osteoporosis, 404
immunohistochemistry, 300 HTLV-I infection, 124
necrosis, 299 Human leukocyte antigen B27 (HLA) screening test, 51
transmural inflammation, 299 Hypogammaglobulinemia, 420
and PMR, 304
symptoms and signs
abnormal temporal arteries, 287–288 I
acute unilateral vision loss, 291 Idiopathic inflammatory myopathies (IIM)
anterior ischemic optic neuropathy (AION), 289 azathioprine and methotrexate therapy, 200
audiovestibular dysfunction, 291–292 dermatomyositis (DM) vs. polymyositis (PM), 193
diplopia, 288 immunosuppressive therapy, 200
facial swelling and noises in head, 291 inclusion body myositis (IBM), 200
fever of unknown origin (FUO), 289 MRI, 199–200
jaw claudication, 289 muscle enzymes
ocular symptoms, 292 aldolase activity, 196
persistent dry cough, 288–289 creatine kinase (CK) activity, 195
unexplained neck pain, 289 myositis patients, 196
vasculitic neuropathy, 293 serum ALT concentration, 196
visual hallucinations, 291 myositis-specific autoantibodies
treatment, 302 anti-SRP antibodies, 198
Gastroesophageal reflux disease (GERD), 84, 85, 171 skin disease, 193–195
Glucocortiosteriod therapy, 8 Idiopathic thrombocytopenic purpura (ITP), 164
Gottron’s papules, 142, 144, 194, 204, 208 IgG4-associated systemic disease, 116, 313, 314
Gout IgM rheumatoid factor, 28
acute monoarticular arthritis, 355, 357 Immune-mediated inner ear disease (IMIED)
epidemiology auditory and vestibular tests, 392–393
488 Index

clinical history, 391 groin area erythema, 224

definition neck swelling symptom, 221
sensorineural hearing loss tinnitus and vertigo, 389 and periarteritis nodosa, 224
sympathetic otitis, 389–390 peribronchial infiltrates and mild interstitial infiltrates, 223
temporal bone and audiovestibular apparatus, 390 periodic echocardiograms, 222
differential diagnosis, 393–395 RSV or PIV test, 222
treatment, 395 Koebner phenomenon, 142
vestibular nystagmus, 391–392
Immune thrombocytopenia (ITP), 176 L
Inclusion body myositis (IBM), 200 Legg-Calve-Perthes disease, 433
Inflammatory arthritis, 69, 92, 244, 273, 373 Leukocytoclastic vasculitis, 18, 225, 261, 332, 333
Inflammatory bowel disease (IBD), 54, 58, 126 Leukopenia, 122, 163, 283
Inflammatory eye disease Livedo racemosa, 176, 177, 282, 283
birdshot retinochoroidopathy, 385 Livedo reticularis, 176, 177, 227, 282, 283
CMV vasculitis, 382 Löfgren's syndrome, 417
cotton wool spot diagnosis, 386 Low back and neck pain
episcleritis, 382, 385–386 cervical myelopathy, 462
glucocorticoids treatment, 382–383 degenerative spondylolisthesis, 457–458
infectious uveitis, 383–384 disc herniation, 455–456
intraocular CNS lymphoma, 383 evaluation and treatment, 458–459
ocular tuberculosis, 387 history and physical examination, 452–454
orbital pseudotumors, 384 imaging
pars planitis, 385 MRI and CT scan, 455
retinal vasculitis, 381 red flags, 454
Roth spots, 387 important causes, 458
scleritis diagnosis, 381–382 mechanical cause, 452, 460
tubulointerstitial nephritis with uveitis (TINU), 386 spinal stenosis, 456–457
Inflammatory myopathies, 197–198 spine basic anatomy, 451
surgery, 459–460
J whiplash injury, 461
Jaccoud’s arthropathy, 3, 117, 136 Lupus anticoagulant (LA) phenomenon, 174
Juvenile dermatomyositis (JDM) Lupus pernio, 416
acanthosis nigricans, 208 Lupus psychosis, 147
bone health, 209 Lyme disease, 103
calcinosis, 206
clinical features, 204 M
disease control, 209 Macrophage activation syndrome (MAS), 28, 163
dystrophic calcifications, 206–207 MALT lymphoma, 116, 123, 125
Gottron’s papule, 204 Membranoproliferative glomerulonephritis (MPGN), 273
heliotrope discoloration and erythema, 204 Meniere’s disease, 394
IIM differential diagnosis, 205 Metacarpophalangeal (MCP) joints, 3, 80, 82, 425
metabolic abnormalities, 207 Mevalonic aciduria, 39
muscle biopsy, 203 Microscopic polyangiitis (MPA), 283
oral glucocorticoid therapy, 208–209 alveolar hemorrhage, 256–257
overlap syndrome, 210 clinical phenotypes, 256
periungual nailfold capillary change, 206 interstitial fibrosis, 255
photosensitive rash, 206 pulmonary hemorrhage, 256
signs and symptoms, 208 RBC with glomerulonephritis, 259
sun protection, 209 treatment and course, 264–267
Juvenile idiopathic arthritis (JIA), 27–31, 51 vasculitic neuropathy, 262
Juvenile rheumatoid arthritis (JRA), 23 Mikulicz disease, 124–125
Juvenile spondyloarthropathy Mixed connective tissue disease (MCTD)
enthesitis clinical feature, 52 anti-RNP antibody emergence, 170
enthesitis-related arthritis (ERA), 51 diffuse swelling, fingers, 171
inflammatory bowel disease (IBD), 54 gastroesophageal reflux, 171
inflammatory enthesitis, 53 myositis, 171
Osgood–Schlatter’s disease, 52 nasal septal perforation, 172
Patrick/FABER test, 54 sclerodactyly and digital ischemia, 171
sacroiliitis, 53 sicca symptoms and pulmonary hypertension, 170
tumor necrosis factor (TNF) inhibitors, 55 unilateral or bilateral trigeminal neuropathy, 169
Monogenic autoinflammatory syndromes
K cherubism, 45–46
Kawasaki’s disease chronic recurrent multifocal osteomyelitis (CRMO)
classification criteria, 224 congenital dyserythropoietic anemia (CDA), 43
conjunctival injection, 222 primary intraosseous lymphoma, 44
coronary arteries dilation, 223 cryopyrin-associated periodic syndromes (CAPS)
Index 489

bony deformities, 42 localized scleroderma, 103

CIAS1 mutation, 41 morphea, 103–104
cochlear inflammation, 42 nephrogenic fibrosing dermopathy (NFD), 97
familial Mediterranean fever (FMF) risk factor, 100
cardinal clinical findings, 35 scleromyxedema, 104–105
colchicine treatment, 36 therapeutic agents, 102
MEFV gene mutation, 34 yellow scleral plaques, 101, 102
renal amyloidosis, 36 Neutropenia, 122
renal transplantation, 37 Nonsteroid antiinflammatory drugs (NSAIDs), 53, 61
septic joint, 34
systemic AA amyloidosis, 37 O
hereditary periodic fever syndromes, 33 Oral glucocorticoid therapy, 208–209
hyper immunoglobulin D and periodic fever syndrome Osteoarthritis (OA)
hyper IgD syndrome (HIDS), 39 aging outcome, 425
prognosis, 40 articular cartilage damage, 426
splenomegaly and gastrointestinal manifestations, 39 carpometacarpal joint, 433
PAPA syndrome, 44–45 dimethylsulfoxide treatment, 427
tumor necrosis factor receptor-associated periodic syndrome glucocorticoid injections, 429
(TRAPS) groin pain, 431–432
anti-IL-1 treatment, 38 joint degeneration
TNFRFS1A mutations, 39 course, 428
Muckle-Wells syndrome (MWS), 23, 41 initial stages, 430–431
Multicentric reticulohistiocytosis, 5 joint pain, 426
Muscle atrophy, 30, 205, 440 knuckle cracking, 425
Musculoskeleton complaints Legg-Calve-Perthes disease, 433
ankle, 448–449 leg length discrepancy, 428
coccyx, 445 mild hip dysplasia, 431, 432
elbow pain MRI, 428
antecubital area swelling, 443 patellofemoral OA, 431
septic olecranon bursitis, 442 radiographs assessment, 429, 430
hand, 444–445 recreational running risk, 425
hip, 445–446 risk factors, 428
knee slipped capital epiphysis, 432
anserine bursa, 448 viscosupplementation efficacy, 433
aspiration to, 447 Osteonecrosis
best injecting way, 446 antero-posterior radiograph, stage IV, 485
leg cramps, 445 arthroplasty treatment, 485
shoulder pain bone marrow edema syndrome, 483
brachial plexopathy, 440 early stages, 481–482
humeral head superior migration, 440 femoral head
microcrystalline disease, 441 blood supply disruption, 482
palpable effusion, 442 core decompression, 484
posterolateral approach, injections, 441 hip resurfacing, for collapse of, 485
subacromial bursa injections, 440 hip MRI, 482–483
wrist plain radiograph, 482
carpal tunnel syndrome (CTS), 443 risk factors, 481
Finkelstein’s test, 444 secondary osteonecrosis
Mycophenolate mofetil (MMF), 152, 155, 186, 367 vs. spontaneous, 486
Myocardial infarction, 16, 134, 151 treatment algorithm, 483
Osteonecrosis of the jaw (ONJ), 404–405
N Osteoporosis
Neisseria gonorrhea, 76 diagnosis, bone mass measurement
Neonatal lupus erythematosus (NLE), 162 DXA scans, 399
Neonatal onset multisystem inflammatory disorder (NOMID) fracture risks, 399
syndrome, 41, 42 low bone density, 399–400
Nephrocalcinosis, 119 epidemiology
Nephrogenic systemic fibrosis (NSF) and scleroderma mimickers among African Americans, 397
clinicopathological diagnosis, 99 among men, 398
dermal tissue injury, 102 hereditary disorder of collagen, 398
Doppler techniques, 101 low trauma fractures, 398–399
eosinophilic fasciitis, 105 prevention
fibrosing disorders, 98 calcium and sodium effects, 400–402
gadolinium-based contrast agent, 100 vitamin D, 402–403
hemodialysis, 103 treatment
histopathological findings, 99 bisphosphonates, 404–405
kidney dysfunction, 100 general considerations, 403–404
490 Index

hormone replacement therapy, 404 and GCA, 304

treatment WHO definition, 397 prednisone treatment, 302, 304
vs. RA, 304
P shoulder pain symptom, 288
Paget’s disease of bone Polymyositis (PM)
asymptomatic disorder, 409 vs. dermatomyositis (DM), 193
etidronate and teraparatide treatment, 411 vs. IBM, 199–200
monostotic and polyostotic disorder, 410 immunosuppressive therapy, 200
radiographic findings, 411 Posterior reversible encephalopathy syndrome (PRES), 147
Parainfluenza virus (PIV) test, 222 Prednisone monotherapy, 313
Pars planitis, 384, 385, 419 Pregnancy, rheumatic diesease
Pauciarticular JIA, 29, 31 antiphospholipid syndrome (APS), 189
Pauci-immune glomerulonephritis, 20 breastfeeding, 187
Pediatric systemic lupus erythematosus dyspnea, 184
autoimmune lymphoproliferative syndrome (ALPS), 164–165 elevated erythrocyte sedimentation rate (ESR), 184
cerebral vein thrombosis, 161 infertility, 183
cutaneous neonatal lupus, 162, 163 medications
discoid lesions, 165 hydroxychloroquine, 186
glucocorticoids dose, 161 ibuprofen, 187
herpes virus infection, 163 methotrexate or leflunomide, 186
MAS and macrocephaly, 163 mycophenolate mofetil (MMF) and azathioprine, 186
neonatal lupus erythematosus (NLE), 162 steroid, 185
proliferative lupus nephritis, 164 sulfasalazine, 187
telangiectasias, 165 neonatal lupus syndromes, 189–190
Pediatric vasculitis ovarian stimulation, 184
classification scheme, 231 proteinuria and creatinine level, 184
Henoch-Schönlein purpura (HSP) rheumatoid arthritis, 190
gastrointestinal symptoms, 225 systemic lupus erythematosus (SLE)
glucocorticoid treatment, 226 lupus flare and preeclampsia, 188
immunofluorescence studies, 228 oral contraceptives, 187
immunopathogenesis, 229 renal allografts, 188
nonsteroidal anti-inflammatory agents (NSAIDs), treatment, thrombocytopenia, 188
226 Primary angiitis of the CNS (PACNS)
palpable purpura, 227 brain imaging, 316
rash of, 225 lumbar puncture, diagnosis, 315
self-limited disorder, 227 with mass lesion, 315
skin biopsy, 225 vs. RCVS
Kawasaki’s disease, 221–224 angiographic features, 317
polyarteritis nodosa (PAN), 229–231 blood and CSF test results, 319
von Willebrand factor antigen, 231 clinical features, 317
Peritoneal dialysis, 103 cross-sectional neuroimaging, 317–318
Plantar fasciitis, 448, 449 Primary biliary cirrhosis (PBC) symptom, 83
Pneumocystis carinii (jiroveci) pneumonia (PCP), 150 Proliferative lupus nephritis, 164
Polyarteritis nodosa (PAN) Psoriatic arthritis (PsA), 29
ACR classification criteria, 229 coronary artery disease, 70
antinuclear antibody (ANA), 277 cutaneous psoriasis, 69
atrophie blanche, 285 dactylitis, 68
cutaneous PAN, 230 inflammatory arthritis, 69
diabetes mellitus, 278 vs. RA, 68
diagnosis, 279 rheumatoid arthritis (RA), 67
hepatitis B-related PAN, 229 skin and joint disease, 70
leg ulcers diagnosis, 281 Pulmonary arterial hypertension (PAH), 77, 88–90
livedoid vasculopathy, 285 Pulmonary fibrosis, 86
livedo racemosa, 283
livedo reticularis, 282 R
magnetic resonance angiography, 284 Raynaud’s phenomenon (RP), 117, 123, 172, 324
microaneurysm, 283, 284 acro-osteolysis, 79
nodular vasculitis, 282 antinuclear antibody assay (ANA), 77
peripheral nervous system vasculitis, 278 calcium channel blockers, 78
renal vasculature, 284 cyanosis ischemia correlation, 79
skin biopsy, 285 dilated capillary loops, 78
skin ulcer diagnosis, 282 skin induration, 77
thrombocytopenia, 283 vasodilator therapy, 79
vasculitic neuropathy, 278, 279 Reactive arthritis (ReA)
visceral abdominal angiography, 284 circinate balanitis, 76
Polymyalgia rheumatica (PMR) dactylitis pattern, 73
Index 491

diagnostic test, 74 Rheumatoid vasculitis

disease-modifying antirheumatic drugs (DMARDs), 75 anti-TNF therapy, 17
enthesitis, 74 corneal melt syndrome, 20
gastrointestinal infections, 75 cyclophosphamide, 15
gonococcal arthritis, 76 glomerular pathology, 21
monoarthritis, 73 leg ulcers, 17
psoriatic arthritis, 76 mononeuritis multiplex symptom, 18
rheumatic fever, 74 myocardial infarction, 16
rheumatoid arthritis, 75 pulmonary rheumatoid nodules, 20
septic arthritis, 73 pyoderma gangrenosum, 17
synovial fluid, 74 rheumatoid factor, 21
Relapsing polychondritis (RP) symmetric polyneuropathy, 18
cartilaginous part inflammation, 339 vasculitic neuropathy diagnosis, 19
chest radiograph, 341 Rodnan skin score method, 84
direct laryngoscopy, 342 Roth spots, 387
irreversible hearing loss, 339
nose bridge collapse, 341 S
patient with traschestomy, 342 SAPHO syndrome, 44
pinna involvement, 340 Sarcoidosis
recurrent pneumonias or bronchiectasis, 342 BAL, CD4:CD8, 415
respiratory symptoms, 340 Bell’s palsy, 414
subglottic stenosis tomogram, 341 bilateral hilar adenopathy, 414–415
vertigo and hearing difficulties, 339 in black patients, 413
Renal dysfunction, 274 clinical outcomes, 414, 422
Respiratory syncytial virus (RSV) test, 222 ECG monitoring, 419
Reversible cerebral vasoconstriction syndromes (RCVS) erythema nodosum, leg, 417
bad migraine attack, 319 Fatigue Assessment Scale (FAS), 418
brain imaging, 318 granulomatous disease, 420
key diagnosis elements, 317 high resolution CT scan, 415
vs. PACNS liver involvement, 418
angiographic features, 317 Löfgren’s syndrome, 417
blood and CSF test results, 319 lupus pernio, 416
clinical features, 317 pars planitis and optic neuritis, 419
cross-sectional neuroimaging, 317–318 peribronchial thickening, 416
pharmacological blood pressure observation, 320 predictive features, 417
transcranial Doppler studies, 319 prevalence of, 414
vasoconstrictive medications avoidance, 320 pulmonary hypertension incidence, 422
Rheumatoid arthritis (RA), 107 pulmonary sarcoidosis, 420–421
adverse effects traction bronchiectasis, 416
cardiovascular disease, 11 Seronegative spondyloarthropathies, 67, 69, 452
hepatitis B virus infection, 10 Seropositive rheumatoid arthritis, 78
clinical features Serum monoclonal gammopathy, 124
amyloidosis, 3 Sjögren’s syndrome (SjS), 81, 162
synovitis and radiographic progression, 2 classification and epidemiology
diagnosis, 1 American–European classification criteria, 108
disease assessment Klinefelter syndrome, 108–109
patient global assessment (PGA), 7 patient classification/diagnosis, 107
simplified disease activity index, 6 differential diagnosis, 124–125
immunizations, 11 extraglandular involvement
mimickers Devic’s disease, 120
calcium pyrophosphate dehydrate (CPPD) deposition disease, dyspareunia, 122
3, 4 dysphagia, 116
chronictophaceous gout, 3 erosive symmetrical arthritis, 117
erosive osteoarthritis, 5 glomerulonephritis, 119
parvovirus infection, 4 Hepatitis C virus (HCV) infection, 118
symmetrical polyarthritis, 3 interstitial fibrosis, 117
serological features and radiology, 5–6 lacrimal/parotid glands, 121
serum rheumatoid factor, 6 liver function tests, 118
treatment lymphocytic vasculitis, 116
antibacterial effects, 9 pancreatic cancer, 119
biologic DMARDs, 7 peripheral neuropathies, 119–120
cholestyramine, 8 sensorineural hearing loss, 121
human antichimeric antibodies [HACA], 10 sensory neuropathy, 120
methotrexate (MTX) therapy, 8, 9 thyroid disease, 122
tumor necrosis, 9 trigeminal neuralgia, 120
Rheumatoid nodules, 2 immunological assays
492 Index

hematological neoplasm, 124 proliferative lupus nephritis, 164

Raynaud’s phenomenon, 123 treatment
keratoconjunctivitis sicca, 107 antimalarial therapy, 152
laboratory findings, 122–123 azathioprine, 155
parotid and submandibular involvement, 115–116 chloroquine treatment, 153
pathogenesis, 109 immunosuppressive drug regimens, 152
prognosis, 125–126 osteonecrosis, 154
sicca Features quinacrine therapy, 153
ocular manifestations, 109–111 renal disease, 153–154
oral manifestations, 111–115 Varicella zoster immunization, 150
systemic treatment, 126 Systemic-onset juvenile idiopathic arthritis (sJIA)
SLE disease activity index (SLEDAI), 146 antiCCP antibodies, 28
Spinal stenosis, 456–457 knee monoarthritis, 30
Spondylolisthesis, 457–458 leukemic arthritis, 31
Subacute cutaneous lupus erythematosus oligoarticular JIA, 29
(SCLE), 138, 144 pauciarticular JIA, 31
Sydenham’s chorea, 161 sacroiliitis, 30
Synovial entheseal complex (SEC), 69 thrombocytosis, 31
Systemic lupus erythematosus (SLE) Systemic rheumatic diseases, 133
arthritis of, 163 Systemic sclerosis (scleroderma)
autoimmune lymph proliferative syndrome antibodies and disease phenotypes
(ALPS), 164 anti-centromere antibodies, 80
central nervous system lupus epidemiologic risk factor, 82
neuropsychiatric lupus, 146 the health assessment questionnaire disability index (HAQ-DI),
optic neuritis, 147 82
cerebral vein thrombosis, 161 pulmonary fibrosis, 83
clinical features RNA polymerase, 81
autoimmune hemolytic anemia (AIHA), 136 cardiac disease, 90
autoimmune liver disease, 134 diagnosis, 80
autoimmune vs. lupus-associated hepatitis, 135 gastrointestinal involvement
cardiovascular disease, 133 gastric antral vascular ectasia, 84
idiopathic inflammatory myopathies, 136 proton pump inhibitors (PPIs), 85
primary biliary cirrhosis, 135 xerostomia, 84
cutaneous lupus interstitial lung disease (ILD)
alopecia areata, 140 bronchoalveolar lavage (BAL), 86
clinical effects, 138 cyclophosphamide therapy, 87
discoid lupus lesions, 144 forced vital capacity (FVC), 85
epidermolysis bullosa acquisita, 139 glucocorticoids, 86
eruptive dermatofibromas, 141 velcro crackles, 88
koebnerize, 142 Wegener’s granulomatosis etanercept trial
lupus panniculitis, 140 (WGET), 88
malar rash, 136 musculoskeletal manifestations, 91–92
mycobacterial infections, 141 peripheral neuropathy, 92
nasolabial folds, 137 pulmonary arterial hypertension (PAH)
oral ulcerations, 142 echocardiography, 88
vermillion border, 143 parenchymal lung disease, 89
vitamin D levels, 145 pulmonary artery pressure, 90
diagnosis Raynaud’s phenomenon (RP) and digital ischemia
American College of Rheumatology acro-osteolysis, 79
(ACR), 131, 132 antinuclear antibody assay (ANA), 77
autoimmune thyroiditis, 132 calcium channel blockers, 78
skin biopsy, 133 cyanosis ischemia correlation, 79
diffuse central nervous system dysfuntion, 164 dilated capillary loops, 78
diffuse/focal myocarditis, 151 skin induration, 77
discoid lesions, 165 vasodilator therapy, 79
erythrocyte sedimentation rate (ESR), 148 scleroderma renal crisis
idiopathic thrombocytopenic purpura (ITP), 164 ACE inhibitor, 90
lupus nephritis angiotensin II inhibitors, 91
glomerulonephritis, 146 skin disease
renal dysfunction, 145–146 pterygium inversum unguis condition, 83
serum creatinine, 145 scleroderma, 84
partial thromboplastin time (PTT), 161 telangiectasias, 83
pregnancy
lupus flare and preeclampsia, 188 T
oral contraceptives, 187 Takayasu’s arteritis (TA)
renal allografts, 188 abnormal angiogram, 312
thrombocytopenia, 188 cardiomyopathy differential diagnosis, 310
Index 493

carotid bypass grafts, CNS symptoms, 313 electrodiagnosis and biopsy

contraindicated pregnancy, 314 bilateral nerve conduction studies, 216
ESR and CRP gauges, 312 flexor hallucis brevis muscle examination, 216–217
femoral artery involvement, 310 muscle biopsy, 217–218
fibromuscular dysplasia, 311 multiple mononeuropathies, 216
vs. GCA, 311 neurological pattern, 213
IgG4-related systemic disease, 313, 314 neurophysiologic patterns, 217
markedly abnormal acute phase response, 312 pain and paresthesias, 215
monophasic illness, 310 pathophysiology
prednisone monotherapy, 313 nerve fiber pathology, 218
Temporal artery biopsy, GCA neural vasculature, 218
contraindications to, 296 perivascular inflammatory infiltrates, 218–219
frozen section analysis, 301 vascular histopathology diagnosis, 219
GCA diagnosis, 296 polyneuropathy
immunohistochemistry aid, 300 manifestations, 213
likelihood estimation ratio, 288 onset and progression, 214
longitudinal section of, 299 sensory deficits, 214
optimal and risky sites, 298 systemic and nonsystemic vasculitis, 215
in scalp, 298 Vestibular nystagmus, 391–392
Tendinitis, 439
Thromboangiitis obliterans W
amputation risk, 321 Wegener’s granulomatosis (WG), 10
clinical and angiographic findings, 322, 326 bladder, 261
early lesions in, 322 clinical features, 248
extra-extremity disease, 325 clinical phenotypes, 256
necrotic fingertips amputation, 325–327 CNS, 262–263
nicotine addiction, 321 consensus classification criteria, 230
Raynaud’s phenomenon, 324 eye
splinter hemorrhages, 325 episcleritis, 255
superficial thrombophlebitis manifestation, 323–324 orbital pseudotumor, 254
systemic vasculitis, 322 tarsal structure in, 254
young men disease, 323 kidneys
Tubulointerstitial disease, 119 granulomas in renal biopsy, 258
Tubulointerstitial nephritis with uveitis (TINU), 386 prednisone treatment, 260
prognostication, 260
U lungs
Undifferentiated connective tissue disease (UCTD), 169. See also chest radiography, 257
Mixed connective tissue disease (MCTD) normal saline nebulizer treatment, 257
U3RNP autoantibodies, 80 venous thrombotic events (VTE), 255–256
Urticarial vasculitis misdiagnosis, 250
hypocomplementemic vasculitis mouth
low complement C1q level, 335 roof erosion, 254
vs. normocomplementemic, 333 strawberry gums, 253
palpable purpura, 336 tongue ulcer, 253
scleritis, 337 nasal crusting, 249
lesions of 3 day’s duration, 335 peripheral seventh cranial nerve palsy, 262
prednisone treatment, 334 splenic involvement, 263
skin biopsy, 334 subglottic stenosis, 229
Uveitis-band keratopathy, 29 diagnosis over phone, 251
dyspnea on exertion, 250
V flow volume loops, 252
Vasculitic neuropathy metal tracheal stents avoidance, 251
cyclophosphamide and glucocorticoids treatment and course, 264–267
treatment, 219–220 wrist arthritis, 262
diagnosis, 19 Whiplash injury, 461