CP

Clinical Perspectives in Obstetrics and Gynecology

OB/GYN

Series Editor:

Isaac Schiff, M.D.

 CP
Clinical Perspectives in Obstetrics and Gynecology
OB/GYN

Series Editor: Isaac Schiff, M.D.

Published Volumes:
Shoupe and Haseltine (eds.): Contraception (1993)
Lorrain (ed.): Comprehensive Management of
Menopause (1993)
Gonik (ed.): Viral Diseases in Pregnancy (1994)
Flamm and Quilligan (eds.): Cesarean Section:
Guidelines for Appropriate Utilization (1995)

Forthcoming Volumes:
Reindollar and Gray (eds.): Molecular Biology for the
Obstetrician- Gynecologist

Published Volumes (Series Editor: The Late

Herbert J. Buchsbaum, M.D.):
Buchsbaum (ed.): The Menopause (1983)
Aiman (ed.): Infertility (1984)
Futterweit: Polycystic Ovarian Disease (1984)
Lavery and Sanfilippo (eds.): Pediatric and Adolescent
Obstetrics and Gynecology (1985)
Galask and Larson (eds.): Infectious Diseases in the
Female Patient (1986)
Buchsbaum and Walton (eds.): Strategies in Gynecologic
Surgery (1986)
Szulman and Buchsbaum (eds.): Gestational Trophoblastic
Disease (1987)
Cibils (ed.): Surgical Diseases in Pregnancy (1990)
Collins (ed.): Ovulation Induction (1990)
Sanfilippo and Levine (eds.): Operative Gynecologic
Endoscopy (1990)
Altchek and Deligdisch (eds.): The Uterus (1991)
Cesarean Section
Guidelines for
Appropriate Utilization

Bruce L. Flamm
Edward J. Quilligan
Editors
With 24 Illustrations

Springer-Verlag
New York Berlin Heidelberg London Paris
Tokyo Hong Kong Barcelona Budapest
Editors:
Bruce L. Flamm, M.D., Research Chairman, Kaiser Permanente,
Southern California Permanente Medical Group, Riverside, CA 92505,
USA
Edward J. Quilligan, M.D., Professor Emeritus, Department of
Obstetrics and Gynecology, University of California-Irvine Medical
Center, Orange, CA 92668, USA
Series Editor:
Isaac Schiff, M.D., Chief of Vincent Memorial Gynecology Service,
Women's Care Division of the Massachusetts General Hospital and
the Joe Vincent Meigs Professor of Gynecology, Harvard Medical
School, Boston, MA 02114, USA

Library of Congress Cataloging-in-Publication Data

Cesarean section: guidelines for appropriate utilization / [edited
by] Bruce L. Flamm, Edward J. Quilligan.
p. cm.-(Clinical perspectives in obstetrics and
gynecology)
Includes bibliographical references and index.
ISBN-13: 978-1-4612-7556-5 e-ISBN-13: 978-1-4612-2482-2
DOT: 10.1007/978-1-4612-2482-2
1. Cesarean section. 2. Surgical indications. 3. Cesarean
section - Prevention. I. Flamm, Bruce L. II. Quilligan, Edward J.,
1925- . III. Series.
[DNLM: 1. Cesarean Section. WQ 430 C4216 1995]
RG761.C486 1995
618.8'6-dc20 94-29533

Printed on acid-free paper.

© 1995 Springer-Verlag New York, Inc.

Softcover reprint of the hardcover 1st edition 1995
All rights reserved. This work may not be translated or copied in whole or in
part without the written permission of the publisher (Springer-Verlag New
York, Inc., 175 Fifth Avenue, New York, NY 10010, USA), except for brief
excerpts in connection with reviews or scholarly analysis. Use in connection
with any form of information storage and retrieval, electronic adaptation,
computer software, or by similar or dissimilar methodology now known or
hereafter developed is forbidden.
The use of general descriptive names, trade names, trademarks, etc., in
this publication, even if the former are not especially identified, is not to be
taken as a sign that such names, as understood by the Trade Marks and
Merchandise Marks Act, may accordingly be used freely by anyone.
While the advice and information in this book are believed to be true and
accurate at the date of going to press, neither the authors nor the editors nor
the publisher can accept any legal responsibility for any errors or omissions
that may be made. The publisher makes no warranty, express or implied,
with respect to the material contained herein.
Production managed by Bill Imbornoni; manufacturing supervised by Jeff
Taub.
Typeset by Best-set Typesetter Ltd., Hong Kong.

987 654 3 2 1
Preface:
Defining the "Appropriate"
Cesarean Section Rate

I wrote an article for Contemporary Obstetrics and Gynecology

in 1983 entitled "Making inroads against the C-section rate."!
In this article, I examined the various factors responsible for
the relative rapid increase in the cesarean section rate in the
United States between 1970 and 1983, when it rose from 5.5%
to 18%. I suggested that it was possible to achieve a cesarean
section rate between 7.8% and 17.5%, and here I would like to
quote directly from the article: "If a hospital were to pursue a
conservative policy on section, for the usual obstetric popu-
lation, what section rate should it expect?" This frequently
asked question is hard to answer because hospital obstetric
populations differ so markedly. However, hospital staffs
should examine their cesarean rates frequently, indication by
indication. Purely as a guide (emphasis mine), the table on
page vi lists some high and low values. These values were
2%-4% for failure to progress, 2%-6% for repeat cesarean
section, 1.3%-3.5% for breech and abnormal lie, 1.5%-3% for
fetal distress, and 1% for third-trimester bleeding.
In 1992, the cesarean section rate in the United States was
22.6%, a somewhat slower increase in the past decade than the
very rapid rise of the 1970s. The rate seems to vary widely
between states, with a low of 16.3% and a high of 28.4%,
between hospitals with lows below 15% and highs above 40%,2
and even between doctors in the same hospital with similar
practices, from 10% to 17%.3 This wide variability depends
primarily on the section rate for failure to progress in labor,
fetal distress, and repeat cesarean sections. Frequently those
physicians who have a very high cesarean section rate do not
encourage vaginal births after a previous cesarean section, are
not active in their management of desultory labors, and are too
ready to label fetal distress when they see a pattern that is
only mildly abnormal. To some, these physicians are guilty of
greed, ignorance, and fear. In my personal opinion, greed plays
a very small role in the cesarean rate. While it is true that
most physicians charge more for a cesarean section than a

v
VI Preface

Cesarean section rates

Percentage
Indication Low High
Failure to progress 2.0 4.0
Repeat cesarean section 2.0 6.0
Breech and abnormal lie 1.3 3.5
Fetal distress 1.5 3.0
Third-trimester bleeding 1.0 1.0
Totals 7.8 17.5
From Quilligan, by permission of Contemporary
l

Obstetrics and Gynecology.

vaginal delivery, I have yet to meet a physician who would do

something they believed would harm their patient even if they
were paid ten times as much for a section. On the other hand,
there are fears and misconceptions. I have heard many doctors
say "I have never been sued for a section I did, but I have been
sued for the section I did not do." The fear of not having
performed a section in my opinion is real, although difficult to
prove, and until the public can be educated that cesarean
section delivery cannot eradicate fetal death and damage, this
fear will remain and will be responsible for some unnecessary
cesarean sections. Bruce Flamm and I hope this book will
correct misconceptions that have been responsible for many
unnecessary cesarean sections.
I am still frequently asked the same old question: What is an
ideal cesarean section rate? I still give an answer similar to
the 1983 answer, perhaps somewhat modified. Every hospital
that has an obstetric service should have some committee that
examines every cesarean section that is performed in that
hospital and determines whether it was indicated or not. If it
was not indicated, then the physician who performed the
section should be educated as to why it was not indicated.
Only through repeated educational efforts will individual
physicians lower their section rates. If the hospital rate
remains high, then it seems reasonable that the licensing body
in the state has an obligation to examine the records of that
hospital. Perhaps the rate is justified; if not, education of the
entire staff is in order. How do I define "high" rate? If the
hospital rate is persistently above the level statewide, it would
certainly seem reasonable to look at the r«::cords.
Another way to look at hospital performance is to look at a
segment of the cesarean section rate that depends primarily on
the philosophy of the physicians and their education as well
as their patients' education, the elective cesarean rate or
its reverse, the rate of vaginal birth after cesarean section
(VBAC). The VBAC rate should be on an upward trend toward
50% and certainly at the national level of 25%.
Preface vii

What about the other major factors in fetal distress and

failure to progress in labor? Electronic fetal heartrate moni-
toring has been blamed for the rapid increase in cesarean
section rate for fetal distress; however, it is not the monitoring
but the interpretation of the data that is at fault. Unfor-
tunately, many of the tracings that have resulted in cesarean
section show a temporary decrease in fetal oxygenation and
are not indicative offetal distress. It must be kept in mind that
the heartrate pattern shown by the monitor has a high false-
positive rate for fetal distress; therefore, other measures such
as fetal scalp stimulation, fetal acoustic stimulation, and fetal
scalp sampling must be employed to rule out false-positive
cases. Committees should determine whether these measures
are being used if the section rate for fetal distress exceeds 2%.
Further, let us not forget our educational efforts not only to
get physicians and nurses to correctly interpret patterns and
use ancillary diagnostic tests but also to educate the public
that not all abnormal patterns are indicative of ongoing brain
damage but may be the result of damage that has occurred
days or months before the onset oflabor. Failure to progress in
labor would seem to be the most difficult figure to determine;
however, some important concepts can help guide a committee
examining individual cases. It is paramount that accurate
records be kept of uterine activity, cervical change, and fetal
presenting part descent.
The active management of labor, described later in this
volume, has been very successful in keeping a very low cesarean
section rate at the National Maternity Hospital in Dublin,
Ireland. In my opinion, the two most important parts of that
program are ensuring that the patient is in labor on hospital
admission and the prompt use of oxytocin when the patient
falls off the labor curve. Both these factors require close ob-
servation of the patient, and the concept of one-on-one nursing
or midwife care is obviously important.
The key I would like to stress is education, education at
every level-the patient, the nurse, and the physician. It
is the only effective method to lower the cesarean section
rate, but to be effective education must be repetitious. I
wish I could say that simply reading this book will lower the
cesarean section rate; it will not. However, this volume will
give you a series of suggestions and the scientific rationale
behind those suggestions. If followed, these ideas will help
you to decrease your cesarean section rate.

E.J. Quilligan, M.D.

References
1. Quilligan EJ. Making inroads against the C-section rate.
Contemp Obstet Gynecol1983;Jan:221-225.
viii Preface

2. Gabbay M, Wolfe SM. Unnecessary cesarean sections: curing a

national epidemic. Washington, DC: Public Citizens Health
Research Group, May 1994.
3. Dermott RK, Sandmire HF. The South Bay cesarean section
study. II: The physician factor as a determinant of cesarean birth
rates for failed labor. Am J Obstet Gyneco11992;166:1799-1806.
Contents

Preface: Defining the "Appropriate" Cesarean Section

Rate............................................. V
EDWARD J. QUILLIGAN
Contributors ............ . . . . . . . . . . . . . . . . . . . . . . . . . . . Xl

Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv
BRUCE L. FLAMM

1 Cesarean Delivery in the United States:

A Summary ofthe Past 20 Years. . . . . . . . . . . . . . . . . 1
BRUCE L. FLAMM

2 Worldwide Utilization of Cesarean Section. . . . . . . . 9

T.J. BROADHEAD AND D.K. JAMES

3 Dystocia and "Failure to Progress" in Labor . . . . . . . 23

EMANUEL A. FRIEDMAN

4 Active Management of Labor 43

PETER BOYLAN

5 Vaginal Birth After Cesarean Section 51

BRUCE L. FLAMM

6 Breech Presentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
LUIS A. CIBILS

7 Cesarean Section for Fetal Distress. . . . . . . . . . . . . . . 95

DAVID A. MILLER AND RICHARD H. PAUL

8 Fetal Macrosomia .............................. 115

HUNG N. WINN AND JOHN C. HOBBINS

9 Twin Gestation and Multiple Births. . . . . . . . . . . . . . 125

CORNELIA R. GRAVES AND FRANK H. BOEHM

IX
x Contents

10 Genital Herpes: Contemporary Management ...... 131

ZANE A. BROWN

11 Methods for Safe Reduction of Cesarean

Section Rates .................................. 141
NORBERT GLEICHER, RICHARD H. DEMIR, JEANNE
B. NOVAS, AND STEPHEN A. MYERS

12 Cesarean Delivery: A Medical-Legal Perspective 163

JEFFREY P. PHELAN

13 Economic Considerations in Cesarean Section Use 173

LAURA B. GARDNER

14 Ethical Issues in the Utilization of Cesarean

Section ........................................ 191
THOMAS E. ELKINS AND DOUGLAS BROWN

15 The Patient Who Demands Cesarean Delivery ..... 207

BRUCE L. FLAMM

16 The Impact of Midwifery Care, Childbirth

Preparation, and Labor Support on Cesarean
Section Rates .................................. 223
JANICE R. GOINGS

17 Cesarean Projects at the State and National Level 247

BRUCE L. FLAMM

18 Guidelines for Appropriate Utilization of Cesarean

Operations ..................................... 255
BRUCE L. FLAMM

Index .............................................. 263

Contributors

FRANK H. BOEHM, M.D.

Maternal-Fetal Division, Department of Obstetrics and Gyn-
ecology, Vanderbilt University Medical Center, Nashville,
TN 37232, USA

PETER BOYLAN, M.D.

National Maternity Hospital, Hones Street, Dublin 2, Ireland

T.J. BROADHEAD, M.B., CH.B.

Department of Obstetrics, Queen's Medical Centre, Not-
tingham NG7 2UH, UK

DOUGLAS BROWN, PH.D.

Department of Obstetrics and Gynecology, Louisiana State
University School of Medicine in New Orleans, New Orleans,
LA 70112, USA

ZANE A. BROWN, M.D.

Professor of Obstetrics and Gynecology, University of Wash-
ington School of Medicine, Seattle, WA 98195, USA

LUIS A. CIBILS, M.D.

Mary Campau Ryerson Professor of Obstetrics and Gynecology,
University of Chicago, Chicago Lying-In Hospital, Chicago,
IL 60637, USA

RICHARD H. DEMIR, M.D.

The Center for Human Reproduction and The Foundation for
Reproductive Medicine, Inc., The Department of Obstetrics
and Gynecology, Mount Sinai Hospital Medical Center,
Chicago, IL 60610, USA

THOMAS E. ELKINS, M.D.

Department of Obstetrics and Gynecology, Louisiana State
University School of Medicine in New Orleans, New Orleans,
LA 70112, USA

Xl
Xll Contributors

BRUCE L. FLAMM, M.D., F.A.C.O.G.

Research Chairman, Kaiser Permanente, Southern California
Permanente Medical Group, Riverside, CA 92505, USA

EMANUEL A. FRIEDMAN, M.D., Sc.D.

Professor Emeritus of Obstetrics and Gynecology, Harvard
Medical School, One Lincoln Plaza, 20K, New York, NY 10023,
USA

LAURA B. GARDNER, M.D., M.P.H., PH.D.

President, Axiomedics Research, Inc., 300 Third Street,
Suite 4, Los Altos, CA 94022, USA

NORBERT GLEICHER, M.D.

The Center for Human Reproduction and The Foundation for
Reproductive Medicine, Inc., The Department of Obstetrics
and Gynecology, Mount Sinai Hospital Medical Center,
Chicago, IL 60610, USA

JANICE R GOINGS, RN., C.N.M.

Coordinator of Nurse-Midwifery Services, Kaiser Permanente
Medical Center, Riverside, CA 92505, USA

CORNELIA R GRAVES, M.D.

Maternal-Fetal Division, Department of Obstetrics and
Gynecology, Vanderbilt University Medical Center, Nashville,
TN 37232, USA

JOHN C. HOBBINS, M.D.

Professor of Obstetrics and Gynecology, Chief of Obstetrics,
University of Colorado Health Sciences Center, Denver,
CO 80262, USA

D.K. JAMES, M.A., M.D., F.RC.O.G., D.C.H.

Department of Obstetrics, Queen's Medical Centre, Notting-
ham NG7 2UH, UK

STEPHEN A. MYERS, D.O.

The Center for Human Reproduction and The Foundation for
Reproductive Medicine, Inc., The Department of Obstetrics
and Gynecology, Mount Sinai Hospital Medical Center,
Chicago, IL 60610, USA

DAVID A. MILLER, M.D.

Department of Obstetrics and Gynecology, LAC/USC Women's
Hospital, Los Angeles, CA 90033, USA

JEANNE B. NOVAS, M.D.

The Center for Human Reproduction and The Foundation for
Reproductive Medicine, Inc., The Department of Obstetrics
and Gynecology, Mount Sinai Hospital Medical Center,
Chicago, IL 60610, USA
Contributors XUl

RICHARD H. PAUL, M.D.

Department of Obstetrics and Gynecology, LAC/USC Women's
Hospital, Los Angeles, CA 90033, USA

JEFFREY P. PHELAN, M.D., J.D.

Co-Director, Maternal-Fetal Medicine, 1030 South Arroyo
Parkway, Suite 110, Pasadena, CA 91105, USA

EDWARD J. QUILLIGAN, M.D.

Professor Emeritus, Obstetrics and Gynecology, University of
California Irvine Medical Center, College of Medicine, Orange,
CA 92668, USA

HUNG N. WINN, M.D.

Associate Professor of Obstetrics and Gynecology, Director,
Division of Maternal-Fetal Medicine, St. Louis University
School of Medicine, Chief of Obstetrics, St. Mary's Health
Center, St. Louis, MO 63117, USA
Introduction

In 1993, President Bill Clinton rode into Washington on the

crest of a massive "health care reform" wave. The reform
did not occur in the way the Clinton administration had
envisioned, but major changes have indeed been taking place.
In what has been termed an "orgy of measurement," every-
thing we do in medicine is now being observed and evaluated.
In such an environment it would be naive to expect America's
most common operation to escape careful scrutiny. In a recent
Journal of the American Medical Association article, Donald
Berwick, M.D., points out, 1 "Physicians can themselves achieve
the purposes of health system reform. In fact, we should; all of
the visible alternatives are worse." He lists eleven worthy
aims of health care reform, including the reduction of cesarean
rates to pre-1980 levels.
The one million cesareans performed each year in the United
States represent a fivefold increase in 20 years. Recent reports
claiming that fully half these operations are unnecessary2,3
have fanned the flames of the cesarean controversy. Surely
this estimate is exaggerated, but just as surely we could be
doing better. Quality improvement experts know that variation
is often a warning sign. When one hospital has a 15% cesarean
rate while a similar hospital in the same town has a 45%
rate, it is only natural for someone to ask, why? When 2,000
babies are delivered annually at each of these hospitals,
but only 300 cesareans are performed at one while 900 are
"required" at the other, sooner or later someone is going to
ask why?
There is no easy answer to this question. Patient risk-
factors and demographics may explain part of the variation
but not all of it. Some would point an accusing finger at us
obstetricians and claim that it's our fault. Naturally I'm a bit
biased, but I don't think we deserve all the blame. Women
continue to request or even demand cesarean operations. If
we don't comply, in the current legal climate we clearly run
the risk of being sued for medical malpractice if there is

xv
xvi Introduction

any problem with a delivery. Experts remind us that physi-

cians generally prevail in such cases; this, however is of little
solace to the obstetrician caught in the clutches of a grueling
five year legal battle.
Some would claim that the escalating cesarean rate is a
financial issue and that paying obstetricians less for cesarean
deliveries would quickly turn things around. This is faulty
logic. Cesareans cost more than vaginal births, but only a
small fraction of the additional cost goes to doctors. Hospitals
do in fact charge much more for cesarean births but this does
not explain rising cesarean rates. If cesareans could be done
more cheaply, it is doubtful that their popularity would wane.
In fact, the opposite might prove to be true.
Legal and financial issues aside, parents and obstetricians
really share a common goal. Both want every pregnancy to
culminate with the birth of a perfect baby to a healthy mother.
Many patients and some physicians have begun to believe
that ever-increasing cesarean rates will lead us closer to
this goal, but statistics do not support this contention. Many
nations with the lowest perinatal mortality rates also have
the lowest cesarean rates. Likewise, many U.S. hospitals have
very low cesarean rates as well as low perinatal mortality
rates.
It is often said that "where there's a will there's a way."
This book shows the way in which cesarean rates can be
safely lowered. Several of our authors have found methods to
lower their cesarean rates to 15% (or less) while maintaining
excellent maternal and perinatal outcomes. However, the tools
provided in this book can only work effectively if there is first
a will to use them. Willingness to change is a prerequisite for
improvement.
Finally, it should be stressed that cesarean delivery is not
an evil that needs to be eliminated. As we strive to attain
reasonable cesarean rates we must never forget that cesarean
operations, when appropriately utilized, have saved the lives
of many mothers and countless babies.

Bruce L. Flamm, M.D., F.A.C.O.G.

References
1. Berwick, DM. Eleven worthy aims for clinical leadership of
health system reform. JAMA 1994;272:797 -802.
2. Silver L, Wolfe SM. Unnecessary cesarean sections: how to cure
a national epidemic. Washington, DC: Public Citizens Health
Research Group, 1989, 1994.
3. Steinbrook, R. Half the cesarean operations in U.S. called un-
necessary. Los Angeles Times. January 27,1989, Part 1. p. 1.
1
Cesarean Delivery in the
United States: A Summary of the Past
20 Years
BRUCE L. FLAMM

This chapter summarizes what has happened the national rate reached 20%.6,7 The re-
to cesarean rates in the United States during vised national birth certificate data for
the past 20 years and outlines some of the 1990, first reported in 1993 because of the
factors responsible for the changes that have lag time in processing millions of records,
taken place. indicates that the rate seems to have sta-
bilized at about 23%.8,9 A 23% national rate
means that approximately 1 million ce-
United States Cesarean sarean operations are now performed in the
Statistics: 1970 to Present United States each year (Figure 1.2). Pre-
liminary data from the National Hospital
A dramatic change in the U.S. national Discharge Survey (NHDS) also indicate that
cesarean rate has taken place during the the national cesarean rate is stabilizing at
past 20 years (Figure 1.1). Obstetricians about 23.5%.10 The NHDS report estimated
trained recently, accustomed to 20%-25% that 977,000 cesareans were performed in
cesarean section rates, may find it difficult 1990, compared to 966,000 in 1991. As is ex-
to comprehend that during most of this cen- plained on page 3, these numbers are so
tury the U.S. cesarean delivery rate remained similar that they may not represent a sig-
consistently below 4%. In 1970 there were nificant difference. The NHDS report also
3.7 million births and the national cesarean points out that 8.5% of women who had a
rate reached 5%. By 1975, the rate reached previous cesarean delivered vaginally in
10%.1 In a span of 5 years, the overall rate 1986 compared with 24.2% in 1991. This
had doubled. By 1978 the rate had reached dramatic increase in the vaginal birth after
15%, three times what it had been just 8 cesarean (VBAC) rate has clearly been a
years earlier. The number of cesareans was major factor in the recent leveling-off of the
growing by 50,000 additional operations overall cesarean rate.
each year, and 500,000 cesareans were being
performed annually.2 In 1979 the National
Institutes of Health (NIH) assembled a 19- How Is the National Cesarean
member Task Force on Cesarean Childbirth. Rate Determined?
In 1981, the results of the task force investi-
gation and their recommendations were This question appears to be trivial, but it is
published in a 537-page document entitled not. It would seem to be a simple matter of
"Cesarean Childbirth.,,3 The report received dividing the total number of cesareans by
widespread publicity but had little apparent the total number of births. However, at least
effect on the rising cesarean rate. 4 ,5 By 1983 until the past few years, the total number of

1
2 B.L.Flamm

FIGURE 1.1. Cesarean rate

in the United States, 1968-
1988.

20

10

o
196870 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 1988

1000000 rate was estimated by extrapolation from a

carefully selected sample population. The
NHDS is a survey that has been conducted
800000
each year since 1965 by the National Center
for Health Statistics (NCHS)Y Each year
the charts of several hundred thousand
patients discharged from hundreds of short-
600000
stay hospitals are abstracted. Some of these
patients have had admissions for obstetric
care. International Classification of Disease
400000
(lCD) codes indicating cesarean delivery are
tabulated and extrapolated to apply to the
entire national population. Before 1989, this
200000 was the only method of determining national
cesarean statistics. In 1989, the U.S. Stan-
dard Certificate of Live Birth was revised to
o include specific details on the method of
1965 1970 1975 1980 1985 1990 delivery.s The revised birth certificate now
FIGURE 1.2. Number of cesarean operations in the indicates whether the birth was by cesarean
United States, 1965-1990. and, if so, specifically whether it was a
primary or repeat operation. If the birth is
vaginal, the revised birth certificate also
cesarean deliveries was unknown and hence indicates if it was a vaginal birth after prior
the national rate could not be calculated cesarean delivery. The new birth certificate
directly. Until 1989 the national cesarean is a major step forward in obtaining ac-
1. Cesarean Delivery in the United States: A Summary of the Past 20 Years 3

curate national statistics. For 1989, data statistically significant. In other words, we
on method of delivery were available for 45 are not absolutely certain if the 1989 na-
states and the District of Columbia but tional rate is closer to 22.8% (birth certificate
were not available for Louisiana, Maryland, data) or 23.8% (NHDS data).
Nebraska, Nevada, and Oklahoma. Although
95% of the birth certificates from reporting
areas indicated the method of birth, 5% did The Cesarean Controversy
not. Hence, at least for 1989, calculation of
national rates from the revised birth cer- With the possible exception of abortion, no
tificate data still involved some degree of other issue in women's health care has
extrapolation. created more controversy in the past decade
The two methods (NHDS and revised birth than cesarean operations. A sampling of the
certificate) for determining the U.S. national dozens of cesarean-related articles that have
cesarean delivery rate have been compared. appeared in newspapers and magazines
Analysis of data from the revised birth cer- across the United States in the past few
tificate for 1989 yielded a national cesarean years are listed in Table 1.1.
rate of 22.8%. The NHDS data for 1989 A glance at the titles of these articles
indicated a 23.8% rate. Because about demonstrates that rising cesarean rates have
4,000,000 births occurred in 1989 in the generated a great deal of concern. While the
United States, the NHDS method indicated fact that the U.S. cesarean rate has increased
that about 40,000 more cesareans were per- dramatically is indisputable, the question of
formed than were detected by the revised whether this change is favorable or deleteri-
birth certificate. Alternatively, the NHDS ous is not so straightforward. During the
data may have overestimated the number of past several years people all over America
cesareans by 40,000 procedures. In any case, have expressed greatly different opinions.
considering the huge number of births Some women have openly stated that they
involved, the correlation between these two believe current cesarean rates are ridicul-
methods is remarkably good. It should be ously high and that this rise has occurred
emphasized that, because we do not know because obstetricians will seek out any
the exact cesarean rate, small changes in the excuse to perform the operation. But many
rate from one year to the next may not be other women have stated that they believe

TABLE 1.1. Selected media coverage of cesarean rates

Year Source Title of article
1987 LA Times C-Sections: Are There Too Many?
1988 LA Times "Runaway" C-Section Rates Reflect "Crisis"
1988 LA Times Hospital Plan Reduces Use of Cesareans
1989 LA Times Half the Cesarean Operations in the U.S. Called Unnecessary
1990 Physicians Weekly Cesareans: A Million a Year
1990 USA Today Doctor's Age Key in Decision on Cesareans
1990 OBC Management C-Sections Under Fire
1991 Parenting The Cesarean Rage
1991 Press Enterprise Corona Hospitals Rank High in Number of C-Section Births
1991 Wall Street J C-Sections Tied to Economic Factors in Study
1991 Consumer Reports Too Many Cesareans
1991 Associated Press Private Hospitals Record High Repeat Cesarean Rate
1992 Expecting What's Behind the Rise in Cesareans?
1992 American Baby Can You Avoid a Cesarean?
1992 -Parade Magazine When Is a Cesarean Really Necessary?
1993 Associated Press Feds Say Doctors Perform Too Many Cesareans
1993 Good Housekeeping C-Sections: Of the Million Done Each Year, Fully Half Aren't
Needed
4 B.L.Flamm

cesarean delivery is wonderful and that Prominent obstetricians have also taken a
vaginal birth is far more painful and probably strong stand against rising cesarean rates.
outmoded. The editor of an obstetric journal stated,13
Even among obstetricians there is no con-
In our specialty, critical scrutiny has lately
sensus. Some doctors believe that cesarean focused on one area: the runaway cesarean rate.
rates greater than 10% reflect unnecessary The problem has now caught the attention of the
surgery and others honestly believe that news media, and public criticism has become
even a 25% rate is too low. There are actually increasingly severe. This is not to argue that the
some physicians who believe that all babies low cesarean rates of the 1960s would be appro-
should be delivered by cesarean. In the words priate in 1986. But for almost one in four babies
of one such doctor,12 to be delivered operatively is a national disgrace.

Why is the trend toward more and more cesarean The cesarean controversy has been fueled
sections? I submit that, in this day of advanced
by a report listing individual cesarean rates
anesthesiological and surgical techniques, for more than 2400 hospitals across the
cesarean section is both easier and safer thanUnited States and claiming that more than
vaginal delivery. Why should the modern woman half the operations were unnecessary.14
undergo the sweaty, gut-wrenching ordeal oflabor
Clearly, the vast majority of obstetricians in
that may last twelve to twenty-four hours or
the United States would adamantly disagree
more? Why must she face the frantic rush to the
with this conclusion. However, even those
operating room for a stat section when the fetal
heart rate drops? In summary, I think the goalobstetricians who believe that current rates
are reasonable may be surprised by the
should be for all women to give birth by cesarean
report's finding that, at some American
section. I will bet that in 50 years, 90% will.
hospitals, more than 50% of all births are
At the opposite end of the spectrum from
now by cesarean. A book coauthored by Dr.
the doctor just quoted are the members oftwo
Sidney Wolfe, director of the Washington-
national organizations who strongly oppose
based Health Research Group, contends that
rising cesarean rates. C/SEC (Cesarean Sup-
3 million American women have already
port Education and Concern) was founded
had unnecessary cesareans and that the
in 1973 after the U.S. cesarean rate had
number is growing by almost half a million
increased from 5.2% to 7.8% in the previous
cases each year. 15 Again, many experts would
3 years. The Cesarean Prevention Move-
strongly disagree with this contention. But
ment (CPM), which later became the Inter-
it would be difficult to argue with the con-
national Cesarean Awareness Network
(lCAN), was formed in 1982 as the national clusion that some unnecessary operations
are being performed.
cesarean rate increased to 18%. The use of
the word "movement" in the group's original
name conjured up visions of hostile confron-
tations. This was unfortunate, because all Does Cesarean Section Prevent
the members I have met are caring indi- Cerebral Palsy?
viduals who volunteer hundreds of hours
each year to help women develop realistic Those who defend current cesarean rates
expectations about childbirth. The CPM point out that perinatal mortality rates have
eventually grew to 75 chapters in 35 states fallen as cesarean rates have increased. Their
and claimed to be counseling more than opponents observe that decreased perinatal
35,000 women each year. Although most mortality rates can be readily explained by
obstetricians were probably unaware oftheir improved antepartum care and great ad-
existence, these two organizations probably vances in neonatal intensive care. More
contributed significantly to the current than a decade ago, this issue was already
leveling-off of the national cesarean rate at being debated on an international level. A
about 23% to 24%. report from Parkland Hospital in Dallas,
1. Cesarean Delivery in the United States: A Summary ofthe Past 20 Years 5

Texas (U.S.A.) claimed that their liberal use pite the threefold difference in cesarean
of cesarean sections, 18% for 1983, was asso- rates, perinatal mortality and morbidity
ciated with decreased intrapartum deaths rates were not significantly different on the
and neonatal seizures when compared to two services.
National Maternity Hospital in Dublin, A recent publication from the National
Ireland. 16 However, the Dublin group pointed Institutes of Health reviewed the literature
out that for a valid comparison of uncommon of the past 25 years in an attempt to deter-
events, larger numbers of births needed to mine if cesarean delivery had been found to
be studied. When data for 1982 and 1984 prevent cerebral palsy.20 The conclusion of
were included, almost four times as many the review was that children born by cesarean
cesareans were performed in Dallas but no have no documented reduced risk ofchildhood
significant difference in perinatal mortality neurologic problems or cerebral palsy. The
was demonstrated between the two hos- authors also compared the rates of cerebral
pitalsP Moreover, when the study interval palsy in nations with very different cesarean
was expanded to include more than 20,000 rates. In spite of cesarean rates that varied
births at each hospital, there was actually a from 7% to 22%, all areas reported cerebral
higher incidence of neonatal seizures in palsy rates of 1.1 to 1.3 per 1000 neonatal
Dallas. The cesarean rates in Dublin during survivors born weighing more than 2500 g.
the study interval ranged from 4.2% to 6.0%
while the rates at Dallas ranged from 17.3%
to 18.0%. The Dublin group concluded that, What Caused the Cesarean
"The dilemma facing contemporary obstetrics Rate to Quintuple?
is how to continue to justify the massive
increase in cesarean births when there is not While there is widespread agreement that
convincing evidence of benefit to the child." the U.S. cesarean rate has increased drama-
The authors went on to say that "the high tically during the past 20 years, there is no
incidence of cesarean section in the United consensus as to why it happened. It is clear
States is not justified by results and that by that no single factor can explain why 200,000
analogy with Dublin the same perinatal cesareans were deemed to be adequate in
mortality rate could be achieved with one 1970 while almost 1,000,000 were performed
third the number of operations performed." in 1990. An increase in the overall birthrate
The controversy over whether the in- does not provide an explanation because
creased use of cesarean section has improved approximately 3.7 million births occurred in
fetal outcome has not been resolved. As one each of these 2 years. Some of the reasons
author pointed out, "The contention over the put forth to explain the rising cesarean rate
years has been that cesarean sections are outlined in Table 1.2.
decreased cerebral palsy and other neurologic The exact contribution of each of these
problems; however, this has not been proved. factors to the overall rise in the cesarean
Nevertheless, the reason for the sharp in- rate would probably be impossible to deter-
crease in cesarean births appears to be in mine. However, it is clear that some factors
many cases for protection of the fetus.,,18 In have had more impact than others. This
a prospective study designed specifically to point will be of more than academic interest
address this issue, unselected patients were to physicians and hospitals interested in
divided into two groupS.19 One group had lowering their overall cesarean rates; those
obstetric management directed at mini- factors that have had the most impact on
mizing the use of cesarean section, and the raising cesarean rates may also have the
other was a routine management group. greatest potential to reduce them.
After 2 years the cesarean rate on the first What individual factors seem to have con-
service was 5.7% while the rate on the rou- tributed most to rising cesarean rates? Data
tine management service was 17.6%. Des- from the NHDS showed that the overall U.S.
6 B.L.Flamm

TABLE 1.2. Some factors that have contributed to complete without consideration of medical-
increased cesarean rates legal ramifications. Cesarean-related dis-
Medical-legal factors cussions almost always end up focusing on
Higher malpractice premiums lawyers and lawsuits. Murphy's law states
Higher malpractice awards that if anything can go wrong, it will. The
Perceived increased risk of litigation obstetrical corollary is that if anything goes
Advertising by malpractice attorneys
wrong, you will be sued. To this corollary
Social factors
Childbearing by older women it is often added that if you did not do a
The "premium" baby concept cesarean you will lose the case. This is com-
The rise and fall of "natural" childbirth monly rephrased as "the only cesarean I've
Medical factors ever been sued for is the one I didn't do."
Management of breech presentation
Management of multiplt gestations
Many American physicians believe that the
Management of genital herpes single most effective way to reduce unne-
Management of postdates pregnancy cessary cesareans would be tort reform that
Management of the extremely small fetus eliminates frivolous lawsuits.
Concerns about macrosomia and shoulder dystocia
Decreased use of forceps and vacuum
Epildural analgesia
Fetal monitoring References
Repetitive factors
Cycle of one cesarean leading to repeat operations 1. Hibbard CT. Changing trends in cesarean
Convenience factors section. Am J Obstet Gynecol 1976;125:798.
Physician related 2. Petitti D, Olson R, Williams RL. Cesarean
Patient related section in California-1960 through 1975.
Monetary factors Am J Obstet Gynecol 1979;133:391.
Hospital compensation for cesarean versus vaginal 3. National Institutes of Health. Report of a
birth
consensus development conference. Bethesda,
Physician compensation for cesarean versus vaginal
birth
Maryland: National Institutes of Health,
Miscellaneous factors Department of Health and Human Services,
Perceived increased safety of surgery 1981. (DHHS publication No. 82-2067.)
Physician experience during training 4. Winkler JD, Kanouse DE, Brodsley L, Brook
RH. Popular press coverage of eight National
Institutes of Health consensus development
cesarean rate increased from 16.5% in 1980 topics. JAMA 1986;255:1323-1327.
to 22.7% in 1985. 6 More than 90% of this 5- 5. Gleicher H. Cesarean section rates in the
United States: the short-term failure of the
year increase appeared to be related to three
National Consensus Development Conference
factors. Repeat cesareans accounted for 48%, in 1980. JAMA 1984;252:3273-3276.
dystocia for 29%, and fetal distress for 16% 6. Taffel S, Placek P, Liss T. Trends in the
of the increase. Analysis of these three in- United States cesarean section rate and
dications will provide a good starting point reasons for the 1980-85 rise. Am J Public
for anyone interested in the evaluation of Health 1987;77:955-959.
cesarean section rates. However, many of 7. Taffel SM, Placek PJ. One-fifth of 1983 US
the issues presented in Table 1.2 are highly births by cesarean section. Am J Public
intertwined. For example, a discussion of Health 1985;75(2):190-192.
repeat cesarean versus trial oflabor certainly 8. Ventura S, Heuser R, Arbetha J, Dunn T.
involves social and convenience factors as Advance report of new data from the 1989
birth certificate. US Dep Health Hum Serv
well as medical factors. Likewise, a discus-
Mthly Vital Stat Rep 1992;40(12):6-7.
sion of fetal distress would not be complete 9. Ventura S, Taffel S, Mathews T. Advance
without considering potential medical-legal report of maternal and infant health data
issues. Unfortunately, in contemporary from the birth certificate, 1990. US Dep
obstetrics a discussion of almost any poss- Health Hum Serv Mthly Vit Stat Report
ible indication for cesarean would not be 1993;42(2):1-31.
1. Cesarean Delivery in the United States: A Summary of the Past 20 Years 7

10. Health objectives for the nation. Rates of 16. Leveno KJ, Cunningham FG, Prithchard JA.
cesarean delivery-United States, 1991. Cesarean section: an answer to the House of
Morb Mortal Wkly 42(15):285-289. Horne. Am J Obstet Gynecol 1984;63:485-
11. Graves JG. National hospital discharge 490.
survey annual summary. Rockville, Mary- 17. O'Driscoll K, Foley M, MacDonald D, Stronge
land: National Center for Health Statistics, J. Cesarean section and perinatal outcome:
1989. (NCHS Publication 89-1760.) response from the House ofHorne. AmJ Obstet
12. Overhulse PR. The cesarean section rate. GynecoI1988;158:449-452.
JAMA 1990; 264:971. 18. Harvey GA. Current status of cesarean births.
13. Queenan JT. The cesarean section rate: out Obstet Gynecol Annu 1984;13:71-82.
of sight, but not out of mind. Contemp Obstet 19. Porreco RP. High cesarean section rate: a new
Gynecol 1986;10:7-8. perspective. Obstet Gynecol 1985;65:307-
14. Silver L, Wolfe SM. Unnecessary cesarean 311.
sections: how to cure a national epidemic. 20. Scheller J, Nelson K. Does cesarean delivery
Washington, DC: Health Research Group, prevent cerebral palsy or other neurologic
1989. problems of childhood? Obstet Gynecol 1994;
15. Wolfe SM, Jones RD. Women's health alert. 83:624-630.
Reading, Massachusetls: Addison-Wesley,
1990.
2
Worldwide Utilization of
Cesarean Section
T.J. BROADHEAD AND D.K. JAMES

Cesarean section (CS) is one of the most on the use of CS in third-world countries. 3
common major operations performed around Figure 2.1 shows the CS rates per 100 hos-
the world. In this chapter we attempt to pital deliveries in 13 selected countries for
answer the following questions: 1983, the most recent year in which rates
were available for comparison.
What is the variation in cesarean section
The highest reported CS rate was in
rates worldwide?
Brazil, at 32% in 1986. 4 More recent figures
Have there been any changes in these rates?
available for the United States show a rate
What are the possible explanations for any
of 25% in 1988. 5 ,6 These high levels are in
differences?
marked contrast to those of Jamaica (5.5%
Are differences in cesarean section rates
in 1989)3 and Czechoslovakia (7% in 1986).4
reflected in differences in outcome?
Within Europe, rates also differ widely.
In England and Wales, the rate was 12.1%
in 19897 whereas in Italy it was 17.5% in
Worldwide Cesarean 1987. 7 Some of the lowest rates in Europe
Section Rates have been reported in the Netherlands (see
Table 2.1), which favors midwife and gen-
There are considerable variations in CS eral practitioner care of low-risk women. In
rates both within and between different addition there is a high proportion of home
countries. However, at a national level, deliveries, with a third of births occurring
relatively few countries routinely provide outside hospitals, and 42% of women being
popUlation-based data on the frequencies of cared for solely by midwives. 7 In general,
CS. 1 ,2 In some countries, such as the United lower rates are found in Eastern Europe
States, where most deliveries occur in hos- countries. For example, Czechoslovakia had
pitals, regional statistics probably provide a a rate of 7% in 19864 and Hungary, a rate of
reasonable estimate of national perform- 10% in 1985.4
ance, but it would be unwise to make this It is not possible to give a national figure
assumption for every country. From the for Australia because not all the states pro-
information that is available on national vide information. However, the most recent
rates, we have compiled an overview that is figure for Western Australia was 16.9% in
summarized in Table 2.1. The difficulties 1987. 8 In comparison, New Zealand had a
in obtaining that information are as just CS rate of 10% in 1985. 4
stated, and although it has been possible to In developing countries, surprisingly high
obtain statistics from a large number of rates have been reported, including 32% for
developed countries, few data are available Brazil in 1986,4 27% for the Mexico City

9
 10 T.J. Broadhead and D.K. James

TABLE 2.1. Overview of worldwide cesarean section rates (%)

Year
Country 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 Reference source

U.S.A. 5.5 5.8 7.0 8.0 9.2 lQ.4 12.1 13.7 15.2 16.4 16.5 17.9 18.5 20.3 23.0 24.0 25.0 1,4,5,6,10, 11, 14, 19, 41, 42, 42
Canada 5.7 6.4 7.2 8.0 9.0 9.6 10.8 12.1 13.9 15.1 15.9 19.0 20.0 1,4,5,10
England 5.0 5.2 5.3 5.3 5.6 6.0 6.5 7.4 7.5 8.5 9.0 9.3 10.1 10.1 10.0 15.0 12.1 13.0 1, 4, 5, 10, 37
Scotland 5.9 5.8 7.0 8.0 9.2 lOA 12.1 13.7 15.2 16.4 16.5 17.9 18.5 20.3 13.5 1,4,38
Puerto Rico 29.0 4
Bavaria 11.3 11.2 11.7 12.2 13.2 15.0 1.4
Denmark 5.7 5.9 7.2 7.5 8.0 8.8 9.0 11.7 11.7 12.8 13.0 12.1 1,4,39
Italy 11.2 12.1 13.0 14.5 15.7 15.8 17.5 4,7,18,19
Portugal 6.1 7.2 8.3 7.9 8.7 8.9 8.4 9.1 9.1 10.3 9.5 9.4 9.8 12.9 13.0 1,4
Greece 13.0 16.7 4,39
Sweden 5.5 6 .8 7.8 9.6 10.9 11.6 11.7 12.1 12.4 12.0 1,4
Spain 12.0 4
Norway 2.2 2.4 2.5 3.0 3.6 4.1 5.0 6.4 7.3 8.0 8.3 8.7 9.0 9.4 12.5 1,4, 14, 38
Switzerland 11.0 4
New Zealand 3.9 3.9 5.2 5.0 5.2 9.7 9.8 10.0 1, 4
Jamaica 4.3 5.8 8.1 4.6 4.5 5.5 3
W. Australia 4.2 4.1 4.6 5.1 7.3 8.2 8.7 9.9 11.7 13.0 11.2 11.8 12.5 13.3 13.9 15.1 15.6 16.9 1.8
S. Australia 16.9 19.0 40
Victoria 14.5 16.4 40
Tasmania 6.0 12.9 40
Brazil 15.0 31.0 27.9 32.0 4,17
East Africa 0.02
Netherlands 2.1 2.1 2.2 2.5 2.6 2.7 2.9 3.5 3.8 4.3 4.7 4.9 5.3 10.0 7.9 1,4,37
Hungary 6.2 6.7 6.9 7.0 6.7 6.5 6.6 6.9 7.4 7.6 8.0 8.6 9.2 9.5 10.0 10.2 1,4,39
Japan 6.4 6.5 6.8 8.0 7.0 1, 4
Czechoslovakia 2.3 2.3 2.4 2.5 2.6 2.8 3.1 3.4 3.7 4.0 4.4 4.7 5.2 6.0 7.0 1,4
Austria 6.5 7.0 7.5
Belgium 7.2 7.4 8.0 8.1
Finland 6.0 6.9 7.2 7.9 8.1 8.2 8.8 10.0 10.9 1.39
France 6.1 8.5 11.9 10.9 10.2 14.4 1
Saudi Arabia 5.4 6.0 5.4 7.5 9.9 15
Israel 3.9 39

Czech
Austria

Belgium

Norway

Hungary

England
Denmark
Portugal

Greece

Bavaria

Australia

Italy

USA

o 5 10 15 20 25
Section rate/100 deliveries

FIGURE 2.1. Cesarean section rate per 100 deliveries in 13 selected countries, 1983 (extracted from
Table 2.1).

metropolitan area,3 and 27% for Puerto have speculated that because the rates are
Rico. 3 Few data are available for African so low, a number of necessary operations are
countries but some reports have suggested not being performed. 9
extremely low rates. Indeed some authors
2. Worldwide Utilization of Cesarean Section 11

Trends in Worldwide able, the United States has maintained the

Cesarean Section Rates highest CS rate for the period shown. The
Canadian trend is similar, as one might
Although Table 2.1 and Figure 2.1 de- expect given the similarity in obstetric
monstrate that worldwide rates in CS dif- practice between these countries. These two
fer, Table 2.1 also illustrates that its use countries have been consistently the "league
throughout the western world has risen leaders" during the past two decades.
consistently in almost all countries during
the past 20 years. 7 For example, it is the Reasons for National
most commonly performed operation in
the United States, where the rate was 16.5%
Differences
in 1980 1 and rose to 25% by 1988. 5 ,6 This is
Why are there such wide differences in the
perceived as a major public health problem
utilization of CS among different countries?
and a cause for increasing concern because
The reasons can be classified as medical and
of the associated higher mortality, higher
nonmedical (Table 2.2).
morbidity, and hence higher cost in com-
parison to vaginal deliveries. Increasing CS
rates could place a serious burden on the Medical Reasons
health care programs of developing coun-
In the United States between 1971 and
tries in which resources tend to be scarce. 3
1982, four indications were responsible for
Figure 2.2 shows the trends in CS rates
most deliveries by CS: previous CS, breech
per 100 deliveries in seven selected coun-
presentation, fetal distress, and dystocia. 1o
tries from 1970 to 1992. The rates within
these countries for anyone year differ, but it
can be seen that there is an upward trend in
Previous Cesarean Section
delivery by CS. Apart from Brazil, from In the 1970s, dystocia and previous CS were
which only patchy information is avail- the main factors responsible for the overall

25

'"
.~ 20
.~ ---0-- USA
a;
"0
g ----- Canada
.,
!:: 15 --0-- Austral ia
"§
- :1:- England
c
o
~ --Norway
., 10

.,'"'"c
---{:po- Neth'lands

<Ii -Czech
'"
QI
U

O +-~-+--r-~~-+--r-~~-+--r-~~-+--r-~~-+--r-+-~~
70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92
Year

FIGURE 2.2. Trends in cesarean section rate per 100 deliveries in seven selected countries, 1970-1992
(see Table 2.1).
12 T.J. Broadhead and D.K. James

TABLE 2.2. Reasons for differences in rates of TABLE 2.4. The number of repeat cesareans as a
cesarean sections percentage of the total number of operations
Medical Nonmedical (adapted from Notzon et al. l )
Country Rate
Previous cesarean Maternal age
Breech presentation Socioeconomic factors Australia 37
Fetal distress Cultural reasons Bavaria 20
Cephalopelvic disproportion Hospital resources France 28
Dystocia Geographical location Hungary 8
Low birthweight and preterm Private practice New Zealand 11
delivery Litigation Norway 23
Physician factor Scotland 43
U.S.A. 35

rate increases.lO However, by the early vaginal delivery after CS are possibleY
1980s previous CS had become the principal Given the concern that is being expressed
factor behind the continued rise. Some about the clinical and financial impact of
reports have suggested that the single most rising CS rates, 3 a critical review of policies
important indication in determining the regarding management of women with a
overall rate is previous CS.lO Nevertheless, previous CS seems advisable.
there are differences between countries in
the number of women who have repeat CS
Breech Presentation
(Table 2.3). Large differences also exist in
the number of repeat CS performed as a The management of breech presentation,
percentage of the total CS rate (Table 2.4). both preterm and term, continues to be
These differences among countries in the controversial. 12 It is not surprising that
number of women undergoing repeat CS countries differ widely in their use of CS for
reflect different obstetric practices. The breech presentation (Table 2.5). Indeed
lowest proportion of vaginal births after there is probably as much variation again
previous CS is found in North America, within countries. All countries illustrated
where the figures are 6.6% for the United show a rise in CS rate for breech presentation.
States and 5.7% for Canada. 4 Cragin's dic- The highest was 93.3% in Sweden, and the
tum of 1916 tends to be adopted there: lowest was in the Netherlands at 34.8%. In
"once a CS, always a CS.,,5 However, vari- the United States, the rate has risen from
ous studies show that rates of 40%-80% for 67.2% in 1980 to 80.4% in 1985, despite the

TABLE 2.3. Percentage of women with a previous

cesarean section undergoing repeat procedure
(adapted from Notzon4) TABLE 2.5. Cesarean section rates per 100
hospital deliveries for breech presentation
Year
(adapted from Notzon4)
Country 1980 1985
Year
Australia 86.0 Country 1980 1985
Brazil 80.0
Bavaria 66.7 61.0 Bavaria 67.1 79.8
Canada 96.2 94.3 Canada 55.0 69.2
Hungary 67.4 68.3 Denmark 55.8 75.4
Netherlands 41.1 44.6 Hungary 32.7 38.7
New Zealand 87.7 81.8 Netherlands 24.5 34.8
Norway 57.4 53.5 New Zealand 41.5 45.9
Scotland 61.3 61.7 Sweden 91.9 93.3
U.S.A. 96.6 93.4 U.S.A. 67.2 80.4
2. Worldwide Utilization of Cesarean Section 13

recommendations of the U.S. Cesarean indications, with the incidence being 20%-
Birth Task Force in 1980. 13 This group 30% (Table 2.6). Further, there has not been
issued guidelines for selecting patients in a consistent rise with time.
whom a vaginal delivery of a full-term Fetal heart rate (FHR) monitoring began
breech presentation could be anticipated: in the early 1970s. By 1980, almost 48% of
mothers of live births were electronically
1. Estimated fetal weight less than 3.58 kg
monitored. However, this produced weIl-
2. Normal pelvic dimensions
documented increased rates of CS for fetal
3. Frank breech without an extended head
distress. 14,15 Admittedly, as experience
4. Delivery by an experienced obstetrician
increased the CS rate became less. Further,
In contrast to these recommendations, the addition of fetal blood sampling as an
others have suggested that all breeches be adjunct to FHR monitoring increases the
delivered by CS. For example, in one study specifity of diagnosis of hypoxia and limits
the argument of the protagonists is that the the CS rate. 13
risks of vaginal delivery to the baby are Obviously, the availability of these
greater than the risks to the mother with methods for monitoring the fetus during
elective CS. The incidence of intrapartum labor differs among countries, which may
and neonatal death associated with vaginal explain some of the small differences. Where
birth was 0.83% compared with 0.03% in there is greater use of fetal monitoring,
those born by CS. In addition, the numbers more frequent diagnosis of fetal distress is
of low Apgar scores and neonatal intubation made. However, it may be that this compli-
were doubled in babies born vaginally or by cation is being overestimated. For example,
emergency CS compared to elective CS. 12 25% of all CS in France in 1982 were for
Others have calculated that the excess risk fetal distress but 8%-10% of Apgar scores
of neonatal death attributable to vaginal were normal at delivery.13 The difficulties
delivery of the term breech was about 4 per in the identification of intrapartum hy-
1000. 12 poxia have been well discussed,13 and until
The reasons for the rise in rates of CS for methods of fetal monitoring with greater
breech presentation are complex. Part of the accuracy are available it is unreasonable to
explanation may be a reflection of reduced expect a fall in the number of CS performed
training opportunities. New obstetricians for fetal distress.
are less often exposed to vaginal delivery
and are unlikely to feel competent to attempt Cephalopelvic Disproportion
such a delivery, thus a spiral of falling
The diagnosis of cephalopelvic disproportion
vaginal delivery rates ensues. This in itself
(CPD) appears to be uniform for most
leads to more CS, with a previous section
countries. With the exception of the Nether-
being the indication, as outlined earlier. In
addition, medical-legal considerations may
contribute to the rising CS rate for breech TABLE 2.6. Cesarean section rates per 100

presentation. hospital deliveries for fetal distress (adapted from

Notzon4 )
Year
Fetal Distress
Country 1980 1985
Given the wide variation in criteria used for
Bavaria 27.3 37.9
the diagnosis of "fetal distress," one might Canada 24.3 24.2
expect a wide variation in the rates of CS for Denmark 21.6
this indication. Surprisingly therefore, with Hungary 25.5 27.0
the exception of the United States, the Netherlands 18.8 21.5
New Zealand 29.5 29.8
variation among countries for this com-
U.S.A. 66.9 55.7
plication was not so great as with other
14 T.J. Broadhead and D.K. James

lands (79%) and Denmark (71%), CS rates of this form of management will significantly
are in excess of 85% for all countries (Table reduce CS rates. 13
2.7). The only concern about drawing too
many conclusions about these figures is that Low Birthweight and Preterm Delivery
the accuracy of diagnosis of CPD is likely to Improvement in neonatal outcome, largely
be inconsistent among the various reports. resulting from advances in neonatal inten-
Indeed, some of the CS performed for CPD sive care, especially in the areas of ven-
could possibly result from dystocia. tilatory support and surfactant replacement,
has altered obstetric practice. 13 However, it
is tempting to speculate that some of the
Dystocia improvement is caused by the greater use of
Dystocia as a complication leading to CS steroids antenatally where preterm delivery
delivery is a significant contributor. How- is contemplated. 13 Few would have performed
ever, the term is vague and does not neces- CS for fetal reasons before 32 weeks gestation
sarily represent a single pathology. For 10-15 years ago, but now CS for severe fetal
example, inefficient uterine activity and compromise earlier than that gestation time
CPD can both lead to the diagnosis. While is commonplace. Once again, however, there
there are inconsistencies in the way the appears to be wide variation between
term is used and hence the diagnosis made, countries in their use ofCS for this indication
there are two additional reasons why CS (Table 2.8).
rates for dystocia may vary. The differing rates presumably reflect the
First, the diagnosis is more common standards of neonatal facilities that are
following induced labor. 16 Indeed, some available in various countries. However,
authors have argued that CS following within individual countries, rates are highest
induction with such an outcome is more among women with very low birthweight
correctly termed failed induction. 17 Induction infants «1500 g) and become lower as
rates vary widely worldwide, and this may birthweight increases. In this context, it is of
contribute to some of the variation in CS note that at the other end of the scale, CS
rates from dystocia. rates are again increased for birthweights
Second, the use of augmentation of labor over 4kg. 18
is not uniformly practiced by all obstetric
units. The protagonists argue that their Nonmedical Reasons
lower CS rates result in large part from this
approach. 13 While the size of the effect is
Maternal Age
debatable, there is sound evidence that use Childbearing patterns also differ among
countries. In western society, there is a
trend for women to have fewer children and
TABLE 2.7. Cesarean section rates per 100 hospital begin childbearing at an older age. l l
deliveries resulting from cephalopelvic dis-
proportion (CPD) (adapted from Notzon4)
TABLE 2.8. Cesarean section rates per 100 hospital
Year
deliveries for low birthweight babies «2500 g)
Country 1980 1985 (adapted from Notzon4)
Bavaria 75.5 86.3 Year
Canada 89.2 94.0 Country 1980 1985
Denmark 77.8 70.5
Hungary 96.5 Bavaria 21.2 35.1
Netherlands 75.6 79.2 Denmark 29.7 41.2
New Zealand 87.6 89.6 Hungary 10.8 15.8
Sweden 93.0 92.4 Sweden 42.5 44.2
U.S.A. 95.7 98.6 U.S.A. 27.3
2. Worldwide Utilization of Cesarean Section 15

In a 20-year U.S. study, CS rates were under $11,000 (Figure 2.3).22 It was con-
consistently highest in women of 35 years cluded that rates of primary CS vary direct-
and older (Table 2.9).11 However, while the ly with socioeconomic status and that this
overall rate over two decades increased association cannot be accounted for by dif-
fivefold, the lowest increase (by a factor of ferences in maternal age, parity, birth-
3.8) was seen in the oldest age group. weight, race, ethnic group, or complications
The finding of an increased risk of CS of pregnancy or childbirth. However, this
delivery with increasing maternal age observation may not solely be the respon-
appears to be a relatively consistent con- sibility of the obstetrician. The phenomenon
clusion of several studies. l l ,18-20 The reasons of women requesting or even demanding CS
proposed for this are not clear, but include is well recognized. 23
an increased risk of malpresentation and a
decreasing proportion of occipitoanterior Cultural Factors
positions in older mothers.20 Others have
Cultural factors can influence CS rates. For
suggested the increased risk results from
example, Brazil has one of the highest rates
increased dysfunctional labor patterns in
of CS in the world. Economic factors are
older women. 20 However, these obstetric/
important, with rates increasing with
medical factors are unlikely to be the sole
family income. In addition, there is strict
explanation. Undoubtedly in some cases,
adherence to "once a CS always a CS."
obstetricians may have a lower threshold for
However, arguably the most important
performing CS on these older women because
reason for Brazil's high rate is that many CS
their pregnancies are considered to be
are performed to carry out an intraoperative
"precious,,,18 that is, the likelihood of having
sterilization. In Brazil, for religious reasons,
further pregnancies is low.
sterilizations are only allowed under very
special circumstances, for example, when
Socioeconomic Status two independent physicians certify that fu-
Several reports from Brazil,21 Italy, and the ture pregnancies would be contraindicated.
United States have suggested there is a In addition, there is no reimbursement by
positive correlation between socioeconomic the national health insurance program. To
status and the use of CS, with women of get around these factors, doctors and fami-
higher social class having higher rates. For lies usually agree that the sterilization
example, a survey conducted between 1982 should be carried out during a CS. The
and 1983 in Los Angeles County, California, doctor may then charge an extra fee and
showed that women with a median family give other reasons for the operation in the
income of more than $30,000 had a primary hospital case notes. 21
CS rate of 23%, as compared with 13% One study of 5960 mothers in southern
among those with a median family income Brazil from 1982 to 1986 reported that 9.4%

TABLE 2.9. Cesarean section rates by age of mother in the United States, selected years 1965-1985
(data from Taffel et alP)
Age of mother (years)
Year All ages <20 20-24 25-29 30-34 ;;:35

1985 22.7 16.1 21.2 22.9 26.6 30.7

1980 16.5 14.5 15.8 16.7 18.0 20.6
1975 10.4 8.4 9.0 11.1 13.6 15.0
1970. 5.5 3.9 4.9 5.9 7.5 8.3
1965 4.5 3.1 3.5 4.3 6.4 7.9
Factor of increase 5.0 5.2 6.1 5.3 4.2 3.8
16 T.J. Broadhead and D.K. James

25

Ul
(I)

:~ 20
'iii
"0
8
';ii
c
15
o
~c
m
<\I
10
Ul
(I)
<>
~
~ 5
if

O+------------+----------~~----------+_----------_+----------~
<11,000 11,000-14,999 15,000-19,999 20,000-24,999 25,000-30,000 >30,000
Median family income ($)

FIGURE2.3. Rates of primary cesarean section in Los Angeles County (California, U.S.A.) according to
median family income. (Adapted from information appearing in The New England Journal of
Medicine. See Gould et a1. 22)

of all women undergoing CS were sterilized. munity or a teaching hospital may have a
However, among women who had had their bearing on CS rates. Some reports have
first child by a vaginal delivery and had shown that fetal distress is three times more
undergone a CS for the second delivery, likely to be diagnosed at a teaching hospital
almost one-third (31%) were sterilized. For than in a community hospita1. 24 This in part
those with two previous normal births and a reflects the fact that teaching hospitals
third by CS, 68% were sterilized. This in- handle more women with high-risk preg-
creased to 80% when a fourth child was de- nancies and may have greater availability
livered by CS after three normal deliveries. 21 of fetal monitoring. Further, community
In contrast, the rate of CS in Saudi Arabia hospitals may lack the resources for an
is much lower. Figures from the Maternity emergency CS that are thought to be re-
and Childrens Hospital, Riyadh, show a rate quired for management of a trial of labor in
of3.9% in 1979, which had increased to 9.9% those women who have had a previous (S.24
by 1984. Cultural reasons have been sug- Although these valid reasons might explain
gested to be responsible in part for this low why teaching hospitals have higher CS
rate. Vaginal delivery is particularly im- rates, there is still a suspicion that those
portant in Saudi Arabia because of the social practicing in such institutions may have an
preference for a large family: the average inherent tendency to prefer CS when faced
size is six or seven children. Because re- with a choice.
peated CS have the potential to limit family Other factors include the availability of
size, use of CS may have serious social re- neonatal facilities and an epidural service.
percussions such as marital breakdown or The effect of epidural analgesia on labor
remarriage by the husband. 15 remains controversial. Some have reported
that an on-demand epidural service does not
affect the primary CS rate25 while others
Hospital Resources and Status have claimed that epidurals increase it. 26
The resources that are available at a par- Clearly, these types of resources are likely
ticular hospital and whether it is a com- to be extremely limited in developing coun-
2. Worldwide Utilization of Cesarean Section 17

tries, and as such it is easy to see how these in terms of greater financial reward and less
factors would result in a lower CS rate. inconvenience.

Geographical Location Litigation

The location of hospitals and their proximity The practice of "defensive" medicine is
to the community is unlikely to be signifi- increasing, particularly in the United
cant in developed countries, but in third- States where litigation is high. Rising mal-
world areas it can have a significant influence practice claims have been documented by
on the rate of CS. For example, extremely the American College of Obstetricians and
low rates have been reported at rural hos- Gynecologists (ACOG). In 1985, 73% of
pitals in East Africa. In fact, rates are so low ACOG Fellows, in response to a survey,
(25 per 100,000) as to imply that numerous reported that one or more professional
obstetric disasters are occurring in remote liability claims had been filed against them,
areas because of difficulty of access to medi- compared with 67% in 1983. 11 Undoubtedly
cal facilities and lack of availability of this litigation is an important factor in CS rates.
surgical procedure. 9 Vesicovaginal fistulas In a survey of 306 consultants asking for
and childhood disability from cerebral their reaSOnS for the rise in CS rates, 7
trauma or anoxia at birth are just two likely litigation was the leading reason given by
examples of such problems. 9 42% of British obstetricians for the rise
in U.K. CS rates and by 84% of American
Private Practice obstetricians for the rise in the United
States. 7 Despite its importance in these
Source of payment for obstetrical care is
countries, litigation has little relevance
related to CS use. This is particularly the
in other countries, such as Italy, where
case in the United States, where women who
malpractice claims are a relatively Un-
are privately insured have the highest CS
commOn phenomenon. 18 If litigation COn-
rates. 26- 28 Studies have shown similar
tinues to rise, its impact on CS rates will
patterns in Brazil and Italy.18,29 However,
increase, because obstetricians are more
this strong relationship has not been dem-
likely to deliver by this method rather than
onstrated for all countries with their differing
attempt even a slightly difficult vaginal
health care systems.
delivery at the risk of a malpractice claim.
A Californian study in 1986 reported on
461,066 deliveries. 29 Women with private
insurance had the highest rate (29.1%), and
Physician Factor
successively lower rates were observed for Apart from the factors already described,
women covered under different schemes. 29 arguably the most important influence is
Similar reports have come from Brazil,30 that of the obstetricians themselves, who
one of which showed that 75% of deliveries make the decisions to deliver a baby by CS.
of private patients were by CS compared to Many medical and nonmedical factors
less than 25% for indigent patients. Medical influence the decisions to perform a CS
indications were unable to explain these birth. Obstetricians will differ in their
variations. In an attempt to decrease this interpretation of these factors according to
rate, the government insurance program their training and experience. In any given
subsequently equalized the reimbursement case, some may decide that CS is necessary,
fees for vaginal delivery and CS. while others may not.
Financial considerations certainly play The Green Bay study of 1986-1988
an important role. It may be that the analyzed 1030 CS deliveries performed
obstetrician is subjected to pressure from On singleton pregnancies by 11 different
their private patient to perform a CS. obstetricians practicing during this time.
Alternatively, it may be that the performance Rates varied from 5.6% to 19.7% for the
of CS provides obstetricians with incentives different physicians and were not attrib-
18 T.J. Broadhead and D.K. James

utable to patient obstetric risk factors, Sweden, which had a lower CS rate of 12%
socioeconomic status, service status, or but also a lower PNM ratio of 7.3 (Figure
duration of the physician practice. 31 Other 2.4). Clearly, perinatal mortality must be
such studies have shown similar results. 6 more closely linked with other characteris-
tics that vary among the nations. 2 This was
well demonstrated in 1983 by O'Driscoll and
Cesarean Section Rates and Foley, who compared the PNM and CS rates
Outcome between Dublin and the United States from
1965 to 1980. Perinatal mortality improved
It has been clearly demonstrated that CS along parallel lines for this period, but while
rates have continued to rise in recent years. the CS rate was increasing from 5% to 15%
What effect has this had on the outcome of in the United States, that in Dublin remained
pregnancy? remarkedly constant at less than 5%.32
The traditional measure used to assess Many studies have confirmed this and failed
outcome is perinatal mortality (PNM). to show a strong correlation between CS and
Perinatal mortality rates have decreased in PNM rates. 18 ,31-34
recent years, and it seems logical to apply a Factors other than increased CS rates
cause-and-effect relationship between these clearly are responsible for the improved
and increasing CS rates. The relationship PNM rates. It is possible that the region-
between CS rates and PNM rates for 14 alization of perinatar care, better techniques
countries is illustrated in Figure 2.4. The for fetal monitoring, and improving neonatal
data suggest that frequencies of delivery do care have a greater impact. In addition, it
not contribute much to variation in peri- has also been suggested that a reduction in
natal mortality rates. For example, in 1985 neonatal deaths from respiratory distress
the United States had a CS rate of23%, with syndrome has had a larger affect on PNM in
a PNM ratio of 10.8. This is in contrast to recent years than has the mode of delivery.35

30
011
r = 0.52, P not significant
25
~ 014
>.
u
y = 0.729 x + 5.89
c 20
<ll
::::>
0'
~
c 02 06
0 15 01
U 012
013 03
OJ
U) 010
C
C1l
~
10 09 04,8 05
C1l
U)
OJ 07
U
5

0+-------------+-------------1-------------~------------_r------------~
o 5 10 15 20 25
Perinatal mortality rate (%)

FIGURE2.4. Perinatal mortality rates compared to cesarean rate frequency in 14 countries, 1985. Key:
1, Australia; 2, Canada; 3, Denmark; 4, England; 5, Hungary; 6, Italy; 7, Japan; 8, Netherlands; 9,
New Zealand; 10, Norway; 11, Puerto Rico; 12, Scotland; 13, Sweden; 14, United States. (Adapted from
Notzon. 4 )
2. Worldwide Utilization of Cesarean Section 19

To assess the effectiveness ofCS with respect nosed by FHR abnormality will be lower
to outcome, perhaps it would be useful to if we use fetal scalp pH sampling as an
look more specifically at outcome (mortality adjunct.
and morbidity) for specific indications rather Finally, our CS rates for "dystocia" will be
than overall PNM rates. Thus, for example lower if we avoid unnecessary inductions and
we should study the benefit of CS for fetal adopt a policy of augmentation of labour.
distress and outcome measures relating to
fetal asphyxial insult, or we should study
the benefit of CS for repeat CS and outcome Acknowledgments. We are grateful to Mrs.
measures relating to fetal risk and maternal L. Straw and Dr. D. Sahota for their help in
complications. the preparation of this chapter.

Conclusions References
1. Notzon FC, Placek PJ, Taffel SM. Com-
Worldwide CS rates differ greatly. The only parisons of national CS rates. N Engl J Med
consistent finding is that they are increasing 1987;316:386-389.
in every country. The reason for the dif- 2. Bergsjo P, Schmidt E, Pusch D. Differences
ferences and the rise in CS rates are many in the reported frequencies of some obstetrical
and complex, reflecting medical and non- interventions in Europe. Br J Obstet
medical factors. Gynaecol 1983;90:628-632.
What should be the ideal rate? "How long 3. Webster LA, Daling JR, McFarlane C,
Ashley D, Warren CWo Prevalence and
is a piece of string?,,36 There is no answer to
determinants of CS in Jamaica. J Biosoc Sci
that question; however, that is no reason for 1992;24:515-525.
complacency. When looking at CS rates 4. Notzon FC. International differences in
for our own institution or discussing the the use of obstetric interventions. JAMA
optimum mode of delivery for a specific 1990;264:3286-3291.
mother, there are certain guidelines that 5. Lomas J. Holding back the tide of caesareans.
will avoid excessively high figures. Br Med J 1988;297:569-570.
We should be vigilant to avoid socio- 6. Goyert GL, Bottoms SF, Treadwell MC,
economic, emotional, and age-related factors Nehra PC. The physician factor in caesarean
that might influence our decisions, although birth rates. N Engl J Med 1989;320:706-709.
it is difficult to argue with a mother who 7. Savage W, Francome C. British CS rates:
have we reached a plateau? Br J Obstet
demands a CS. We should guard against
Gynaecol 1993;100:493-496.
using different standards with private 8. ReadAW, Waddell VP, Prediville WJ, Stanley
patients. We should resist attempts to make FJ. Trends in CS in Western Australia 1980-
our decisions litigation driven. In developing 1987. Med J Aust 1990;153:318-323.
countries, mechanisms should be in place for 9. Nordberg EM. Incidence and estimated need
transfer of women in labor who might ofCS, inguinal hernia repair, and operations
benefit from CS where no such facility exists for strangulated hernia in rural Africa. Br
locally. Med J 1984;289:92-93.
In women with previous single CS, we 10. Anderson GM, Lomas J. Recent trends in CS
should consider vaginal delivery as an rates in Ontario. Can Med Assoc J 1989;141:
option in subsequent pregnancies. In a 1049-1053.
11. Taffel SM, Placek PJ, Liss T. Trends in the
woman with a breech presentation we
United States CS rate and reasons for the
should not automatically think of a CS as 1980-85 rise. Am J Public Health 1987;77:
the only delivery option. We should seek 955-959.
improved and more accurate methods or 12. Thorpe-Beeston JG, Banfield PJ, St. G
diagnosing intrapartum fetal hypoxia. Cer- Saunders NJ. Outcome of breech delivery at
tainly our CS rates for fetal distress diag- term. Br Med J 1992;305:746-747.
20 T.J. Broadhead and D.K. James

13. Derom R, Patel NB, Thiery M. Implications 28. Bertollini R, DiLallo D, Spaden T, Perucci C.
of increasing rates of CS. In: Studd J, ed. CS rates in Italy by hospital payment mode:
Progress in obstetrics and gynaecology, Vol. an analysis based on birth certificates. Am J
6. New York: Churchill Livingstone, 1987: Public Health 1992;82:257-261.
175-194. 29. Stafford RS. CS use and source of payment:
14. Borthen I, Lossius P, Skjaerven R, Bergsjo P. an analysis of California hospital discharge
Changes in frequency and indications for CS abstracts. Am J Public Health 1990;80:313-
in Norway 1967-1984. Acta Obstet Gynecol 315.
Scand 1989;68:589-593. 30. Janowitz B, Nakamura M, Lins FE, Brown
15. Chattopadhyay SK, Sengupta PB, Edress ML, Clopton D. CS in Brazil. Soc Sci & Med
YB, Lambourne A. CS; changing patterns 1982;16:19-25.
in Saudi Arabia. Int J Gynaecol Obstet 31. DeMott RK, Sandmire HF. The Green Bay
1987;25:387 -394. CS study: 1. The physician factor as a deter-
16. Gibb DMF, Cardozo LD, Studd JWW, Cooper minant of caesarean birth rates. Am J Obstet
DJ. Prolonged pregnancy: is induction of Gynecol 1990;162:1593-1602.
labour indicated? A prospective study. Br J 32. O'Driscoll K, Foley M. Correlation of decrease
Obstet Gynaecol1982;89:292-295. in perinatal mortality and increase in CS
17. Bergsjo P, Bakketeig LS, Eikhom SN. Case- rates. Obstet Gynecol 1983;61:1-5.
control analysis of post-term induction of 33. Sanchez-Ramos L, Kaunitz AM, Peterson
labour. Acta Obstet Gynecol Scand 1982;61: HB, Martinez-Schnell B, Thompson RJ.
317-324. Reducing CS at a teaching hospital. Am J
18. Signorelli C, Elliott P, Cattaruzza MS, Obstet Gynecol1990;163:1081-1088.
Osborn J. Trend ofCS in Italy: an examination 34. Taylor U, Zentay Z, Ganesh V, Apuzzio J,
of national data 1980-1985. Int J Epidemiol Murphy G, Iffy L. Rates of CS and neonatal
1991;20:712-716. mortality. Aust NZ J Obstet Gynaecol1992;
19. Parazzini F, Pirotta N, La Vecchia C, Fedele 32:203-205.
L. Determinants of CS rates in. Italy. Br J 35. Rydhstrom H, Ingemarsson K, Ohrlander S.
Obstet Gynaecol1992;99:203-206. Lack of correlation between a high CS rate
20. Peipert JF, Bracken MB. Maternal age: and improved prognosis for low-birthweight
an independent risk factor for caesarean twins «2500 g). Br J Obstet Gynaecol
delivery. Obstet Gynecol1993;81:200-205. 1990;97:229-233.
21. Barros FC, Vaughan JP, Victoria CG, Huttly 36. Chamberlain G. What is the correct CS
SRA. Epidemic of CS in Brazil. Lancet 1991; rate? Br J Obstet Gynaecol 1993;100:403-
338:167-169. 404.
22. Gould JB, Davey B, Stafford RS. Socioeco- 37. Treffers P, Pel M. The rising trend for
nomic differences in rates of CS. N Engl J caesarean birth. Br Med J 1993;307:1017-
Med 1989;321:233-239. 1018 ..
23. Hall MM. When a woman asks for a CS. Br 38. Notzon FC, Bergsjo P, Cole S, Irgens LM,
Med J 1987;294:201-202. Daltveit AK. International collaborative
24. Anderson GM, Lomas J. Explaining varia- effort (ICE) on birth weight, plurality,
tions in CS rates: patients, facilities or perinatal and infant mortality: IV. Dif-
policies? Can Med Assoc J 1985;132:253- ferences in obstetrical delivery practice:
256. Norway, Scotland and the United States.
25. Gribble RK, Meier PR. Effect of epidural Acta Obstet Gynecol Scand 1991;70:451-
analgesia on the primary caesarean rate. 460.
Obstet Gynecol 1991;78:231-234. 39. Stephenson PA, Bakoula C, Hemminki E, et
26. Thorp JA, Parisi VM, Brylan PC, Johnstone al. Patterns of use of obstetrical interventions
DA. The effect of continuous epidural in 12 countries. Paediatr Perinatol Epidemiol
analgesia on CS for dystocia in nulliparous 1993;7:45-54.
women. Am J Obstet Gynecol 1989;161:670- 40. Foote AJ, Giles WB. Review of obstetric
675. operative intervention rates. Asia-Oceania J
27. Stafford RS. Alternative strategies for con- Obstet Gynaecol1992;3:195-198.
trolling rising CS rates. JAMA 1990;263: 41. Anderson GM, Lomas J. Determinants of the
683-687. increasing caesarean birth rate. Ontario
2. Worldwide Utilization of Cesarean Section 21

1979 to 1982. N Engl J Med 1984;311:887- rates. Obstet Gynecol 1987;69:696-

892. 700.
42. Shiono PH, McNellis D, Rhoads GG. 43. Halpin GJ, Rose E, Shapiro E. Trends in CS
Reasons for the rising caesarean delivery rates. N J Med 1989;86:867-873.
3
Dystocia and "Failure to Progress"
in Labor
EMANUEL A. FRIEDMAN

Too often invoked as the rationale for to reverse the trend, still more primary
cesarean section, dystocia is a major (if not cesarean sections are being done annually,
the principal) contributor to the phenomenal not fewer.
growth in the cesarean section rate during The national cost of these unnecessary
the past several decades in the United cesarean sections, as calculated by the CDC
States and, to a greater or lesser extent, in solely from the differential hospital costs of
every other industrialized nation in the cesarean versus vaginal delivery, is more
world. There is no denying that our current than 1 billion dollars annually. It deserves
cesarean section rate is excessively and to be emphasized, moreover, that the 1-2
unacceptably high, and there is no longer billion-dollar estimate for unnecessary
any logical argument to support it. Based on cesareans does not begin to take into
the most recent available national data account the economic toll in lost wages, pro-
(1991), the Centers for Disease Control longed convalescence, augmented home care
(CDC) reports that the national cesarean needs, and the well-documented increase in
rate has apparently leveled off at 23.5%.1 the number and severity of major and minor
The United States and Canada previously complications so commonly associated with
led the world in the frequency of cesarean cesarean delivery, to say nothing about the
deliveries,2 and we can take little solace in medical, physical, psychologic, and societal
the news that our current rates are exceeded repercussions that are likely to affect the
only by Brazil and Puerto Rico. patient and her family and which cannot be
In more specific and telling terms, the assigned a dollar value.
number of unnecessary cesarean sections, This information comes at a time when
according to the recent CDC statement, l the economic impact of medical care is
was 349,000, 36.1%, of the 966,000 done in foremost in the minds of the public and the
1991. There was some good news in this government. The pervading concern has
announcement to the effect that women who come about as a consequence of the percep-
had previously had a cesarean section were tion that our current health care system is
being delivered vaginally (vaginal birth too expensive and inefficient, and that it
after cesarean, VBAC) more often; the should be feasible to provide the same or
frequency of successful VBACs rose to 24.2% better care at equal or less cost to more
from 20.2% in 1990, a small but nonetheless people (conceivably, to everyone) by some
gratifying advance. However, the falling form of governmentally controlled health
rate of repeat cesareans coupled with the care system. The perception is bolstered by
unchanging overall rate has to reflect the the fact· that more than 12% of our gross
bad news that, notwithstanding all efforts national product goes for health care, far in

23
24 E.A. Friedman

excess of other nations in which the health for just these indications alone, so that these
care is at least comparable in quality and two factors would be used by obstetricians to
distribution. Given the general dissatisfac- justify abdominal delivery only when truly
tion with our current economic environment, indicated, the resulting reduction in the
it is clear that increasingly intense interest cesarean section rate would be substantial.
will be focused on modifying and amelio-
rating health care practices in those areas of
medicine that have already been defined
What Is Dystocia?
as potentially problematic. Obstetrician- Dystocia is etymologically derived from the
gynecologists, having been identified as Greek root tokos, childbirth, and the prefix
practicing a discipline in which "unneces- modifier dys, meaning hard, bad, or difficult.
sary" surgical procedures such as cesarean Thus, the term is merely a nonspecific
section and hysterectomy are being done at descriptor of a difficult parturition. Its use
what is widely regarded as an excessive rate without further definitional characteriza-
(regrettably, with some justification), are tion, therefore, is at best suspect. The
therefore especially vulnerable. suspicion that the term may be inappro-
Given this situation, there is an obvious priately used to describe a given labor, and
urgent need for us to address the problem thereby justify an unnecessary operative
objectively and unemotionally to help intervention, is magnified by the not-
understand the issues, determine their root infrequent application of the term on the
causes, consider potential solutions, and basis of a strictly subjective (if not overtly
develop means for correcting them. This erroneous) interpretation of the events of
chapter addresses the single condition labor.
recognized to be among those leading the list As the shadow of doubt looming over the
of factors contributing the current high merit of a diagnosis of dystocia has become
cesarean section rate, namely, dystocia. increasingly appreciated and publicized,
The only other competing factor in terms admonitions have been forthcoming to
of incidence is repeat cesareans, which is abandon it or to define it in clinically
variously encountered in about the same objective terms so as to yield diagnostic
frequency or slightly more often. relevance. 13 - 16 As a consequence, substitute
The quantitative impact of dystocia on the language has been introduced. The term
increase in the frequency of cesarean sec- now most widely used in lieu of dystocia is
tions has been put at between 28% and its clinical euphemism, "failure to progress."
43%.3-9 In perspective, these relative in- Substitution of this new term for dystocia
cidence figures take on greater relevance has apparently occurred widely, but it has
when it is realized that many surgical occurred without concomitant introduction
interventions for dystocia yield babies of no of measures to ensure that it would be
more than normal weight and size. 10 Still clearly defined in specific terms, documented
more impressively, more than half (56.5% when used, and applied meaningfully to any
in one report ll and 66% in another 12 ) of given case and only in those in which it truly
cesarean sections in nulliparas are done for reflects the existence of a labor process that
dystocia. is not only nonprogressive, but one that
Thus, one-third to one-half or more of really warrants aggressive intervention by
the growth in the cesarean rate can be cesarean section.
attributable to this factor alone, and the In a few institutions where compliance
combination of repeat cesareans and dystocia with these stringent constraints occurred,
accounts for two-thirds to three-quarters of the intended effect of reducing the number
the increase. It should be clear from these of unnecessary abdominal deliveries from
data that if it were possible to reduce the the inappropriate use of the diagnostic
frequency of cesarean sections undertaken term dystocia did materialize. 17 Programs
3. Dystocia and "Failure to Progress" in Labor 25

to monitor the quality of labor care, intro- adult with abdominal pain without suitable
ducing use of labor curves (see following) or clinical and laboratory evaluation to both
equivalent labor definitions to describe the confirm the diagnosis and rule out nonsurgi-
specific type of labor aberration, plus inter- cal conditions.
nal surveillance of management practices To evaluate labor progression in such a
supplemented with external reviews, proved woman, to determine whether she actually
capable of reducing the cesarean section rate has a labor abnormality, and to identify its
by a considerable degree without adversely specific type if present, one must be familiar
affecting outcome results. Where educa- with the normal course of labor, have a
tional programs for doctors, nurses, and the working knowledge of the limits of the
lay public were tried alone, no impact was normal labor pattern, and be able to re-
discernible, and the cesarean section rate cognize the kinds of possible disorders.
continued to rise. IS Overall, however, very Moreover, after a specific disorder is diag-
little has occurred nationwide to reduce the nosed, it becomes necessary for the health
cesarean section rate. 4 Needless to say, care provider to undertake a timely assess-
merely changing the terminology, which ment to determine what is the likely un-
appears to be all that took place in most derlying cause of the disorder.
communities, could not have been realisti- It is only with this type of information in
cally expected to have had any material hand, namely, a defined disorder with an
impact. identifiable probable etiology, that one can
proceed with an appropriate management
program tailored to meet the needs of that
Nonspecific Nature of particular patient and her fetus. To achieve
Failure to Progress this worthwhile goal, the attendant must
know about the kinds of conditions that can
A gravida who fails to progress in labor may cause each of the different types of labor
be exhibiting one of a number of different aberrations and thus investigate for them.
conditions ranging from patently innocuous He or she must know what therapeutic
to potentially very serious. 19- 21 It is neces- options are available and appropriately
sary to recognize that failure to progress applicable to the specific labor disorder and
comprises several dissimilar and quite the concurrent conditions prevailing in that
unrelated disorders, each of which must be patient. Moreover, the practitioner must
differentiated from the others and managed know what can be expected prognostically
in rather specific ways according to manage- from each condition in regard to delivery
ment principles which are appropriate for outcome and result to mother and fetus,
each abnormality and may prove inappro- thereby to be able to choose among them
priate, if not outright counterproductive or intelligently and with the best interests of
even harmful, if applied to the treatment of both mother and fetus in mind.
the others.22
To diagnose a given laboring patient as
having dystocia or failure to progress, and to Graphing the Labor Course
proceed forthwith to undertake the major
surgical procedure of cesarean section To distinguish among the several conditions
without first differentiating which one of responsible for the situation in which a
these several conditions she may actually laboring patient appears not to be making
be experiencing, cannot be considered ac- progress, the simple expedient of graphing
ceptable. Conducting one's practice in this the labor course should prove very useful.
manner would be no more correct than, It has long been common knowledge that
for example, undertaking an exploratory normal labor almost always follows a
laparotomy for appendectomy in every definable pattern over time. 22 Aside from
26 E.A. Friedman

consideration of the requirements for careful pattern can be done by the use of a simple
assessment of maternal and fetal status square-ruled grid drawn on paper or, as is
(matters of great importance not considered the case in many obstetric units around the
in this chapter, which is limited to the diag- world, a preprinted blank graphic form
nosis and management of dystocia and incorporated into the patient's hospital
therefore addresses the labor course only), chart for this purpose.
the complexity of the clinical observations The form consists of a graph with the
used to try to interpret the dynamic changes horizontal axis divided into hours. The
that occur during labor encompass those vertical axis is usually constituted into
that focus on the contractility pattern, the 10 equal divisions, scaled in two ways
changes that can be expected to occur in simultaneously: on the left border by cervical
cervical dilatation and effacement, and dilatation, in centimeters, from 0 to 10; and
those anticipated for fetal descent through on the right border by fetal station from - 5
the birth canal. to +5 according to the convention that
It is obvious that uterine contractions are designates station by the level in cen-
the sine qua non of labor; little will occur timeters to which the forward leading edge
over the course of labor without them. of the fetal presenting part has descended
Nonetheless, the obstetrician's ability to relative to the referent plane ("zero" station)
utilize the pattern of contractions, that is, of the ischial spines at the midplane of the
their intensity, frequency, and duration or maternal pelvis (negative numbers signify
the wave pattern generated by continuous stations cephalad to the ischial spines and
intrauterine pressure measurements, as positive numbers, caudad).
clinically relevant data for purposes of If the older but less preferable system
assessing labor progress has thus far proved of dividing the pelvis into thirds is used
to be of essentially no clinical value in as- instead, so that the range of station designa-
sessing whether a given labor is normal. 23 ,24 tions is -3 to +3, it is recommended that
Similarly, while information about efface- suitable adjustments be made in the scale to
ment of the cervix is sometimes helpful in ensure that the recordings are comparable
other ways, such as in determining whether to those made using the centimeter scale
a patient is in true labor or perhaps is likely and the evolving descent curves can be
to be a suitable candidate for labor induc- evaluated accordingly by the same criteria.
tion, it does not help solve the clinically The adjustment can be made graphically or
important concerns over whether a given by simple arithmetical calculation (one
labor is normal or not. station on the thirds scale being equivalent
The clinician's role can be greatly sim- to 1.67 cm on the centimeter scale).
plified by concentrating on the two re- For greatest utility, one should ordinarily
maining factors, namely cervical dilatation begin the timing scale on the horizontal
and fetal descent, both of which can be time axis (at the lower left corner ofthe grid)
considered to represent the effective end for a given patient with the onset of her
result of those labor forces acting to cause labor, by convention that time when she
the labor to progress along normal lines so reports that the contractions were perceived
that the mouth of the womb is opened suf- by her and became regular, acknowledging
ficiently to allow the fetus to exit from the that in due course it may become clear (by
uterus and to be eventually delivered to the hindsight only) that she may not actually
outside world. Plotting the course of cervical have been in labor at all at that time. This
dilatation against the time that elapses will mean that there will often be a period of
after labor begins (or is assumed to have time, extending from the onset of labor until
begun) and fetal station on a similar time the patient is first seen and examined,
basis permits one to evolve a graphic repre- during which the graph will necessarily
sentation of the labor course. Tracing the remain blank; while this might seem an ill-
3. Dystocia and "Failure to Progress" in Labor 27

advised waste of space in the chart, it may stage of labor and the beginning of the active
prove useful to help determine the duration phase, which continues until the end of the
of the initial portion of the labor process and first stage at full cervical dilatation. The
to remind all attendant personnel that that dilatation continues to speed up (the
portion of the labor occurred outside their acceleration phase, the first of the three
surveillance. While events during the pro- identifiable components of the active phase)
dromal or latent phase labor may seldom until it reaches a fairly rapid speed. Once its
prove to be clinically important, recent data maximum rate has been achieved, dilatation
suggest that the disorder of prolonged usually continues in this fashion at that
latent phase, heretofore deemed innocuous, constant rate (that is, the curve assumes
may actually contribute to adverse fetal linearity, now in the phase of maximum
outcome. 24 slope) until the cervix is nearly fully di1ated.
The results of each successive vaginal The rate of dilatation then appears to slow
examination for observing the degree of down (the deceleration phase) until the
cervical dilatation and fetal station from the cervix is fully retracted and fully dilated at
time of hospital admission until delivery are the onset of the second stage.
entered into the grid at the appropriate The deceleration phase is produced arti-
points on the graph according to when they factually in that the cervix is being pulled
were obtained. For example, a determina- centrifugally at the same rate as earlier in
tion of 3 cm of cervical dilatation found at the phase of maximum slope, but instead of
the examination done just 7 hafter labor dilating, it is being drawn upward by the
began would be entered on the seventh line action of the upper uterine segment. Thus,
from the left margin of the graph, three continuing dilatation appears to slow down
divisions up from the bottom. As each sub- or even cease in due course as the cervix is
sequent dilatation entry is recorded in progressively retracted around the fetal
sequence and joined to the preceding one by head. The interval encompassing the ac-
a straight line, it can be expected to yield an celeration phase, phase of maximum slope,
evolving pattern of cervical dilatation. The and deceleration phase is the active phase of
same is done to trace the descent pattern. the first stage. The combination of latent
Graphing a labor as it progresses has the and active phases comprises the first stage.
advantage of allowing attendant personnel At the same time, the descent curve
(as well as the patient) to see the pattern remains fiat until the dilatation pattern is
of her labor as it develops. This will show well advanced into the active phase. Descent
whether or not she is following a pathway usually begins during the phase of maximum
consistent with normal in regard to its slope of the dilatation curve, and reaches its
general shape and falling within the range maximum rate by the beginning of the
of normal component durations and slopes. deceleration phase or, at latest, at the onset
Under ordinary circumstances, for example, of second stage. Once a maximum rate of
all normal gravidas can be expected to trace descent is achieved, descent continues at
very typically shaped patterns. this rate, proceeding linearly, until the
The normal dilatation curve is S shaped. presenting part reaches the level of the
It begins as a fiat, essentially unchanging perineum.
line at the onset of labor. This is the latent
phase of the first stage of labor. After
a period of time, dilatation accelerates, Defining Abnormal Labor
although the contractility pattern or the Patterns Objectively
patient's subjective perception of her uterine
contractions is unlikely to change to any The characteristic pattern of the normal
degree. The upswing of the dilatation curve dilatation curve thus is sigmoid in shape
marks the end of the latent phase of the first and the normal descent curve is hyperbolic.
28 E.A. Friedman

When a given patient's labor pattern de- and slopes of the several component parts of
viates qualitatively from these shapes in the curves. Just as the aforementioned qua-
certain recognizable ways, that labor pattern litative variations in shape were able to
can be diagnosed as being abnormal. The identify several abnormalities of labor pro-
kinds of abnormalities that can be identified gression, so the quantitative changes serve
on the basis of such qualitative deviation in like manner as important indicators of
from normal are those in which the expected the presence of others. The qualitative ab-
linear dilatation in the active phase stops normalities thus defined include prolonged
(for at least 2 h) before the the second stage latent phase, protracted dilatation, pro-
is reached; this is called secondary arrest of longed deceleration phase, and protracted
dilatation. Another is the parallel cessation descent, as detailed in Table 3.1.
of descent (for at least 1 h) after the maxi-
mum slope of descent has begun; this is
arrest of descent in the second stage. A third Potential Errors in Making the
disorder recognized by its shape is that of
failure of descent, readily diagnosable if the Diagnosis of Failure to
fetal descent fails to materialize at or after Progress
the time descent should already be at its
maximum, that is, when the deceleration As should now be evident from our review
phase of dilatation begins or, at the latest, of the kinds of labor disorders that are so
when the second stage has begun. clearly diagnosable on the basis of the devia-
In addition to abnormalities based on the tions from the normal cervical dilatation
characteristic shapes of the dilatation and and fetal descent curves, failure to progress,
descent curves, other aberrations can be or the apparent lack of progress in either
identified on the basis of the clinically cervical dilatation or fetal descent, for any
valuable, quantitative attributes they ex- finite period of time in labor may represent
hibit. The latter are based on the statistical any of several quite different entities. The
distribution of the normal range of durations range of possibilities is listed in Table 3.2,

TABLE3.1. Labor disorders objectively identifiable by graphic analysis with diagnostic criteria and
managment options
Diagnostic criteria
Labor pattern Nulliparas Multiparas Recommended therapy
Prolonged latent phase >20h >14h Therapeutic rest
Exceptionally, oxytocin if delay
of 6-10 h is unacceptable
Protracted dilatation <1.2cm/h <1.5cm/h Expectancy and support
CS if CPD disclosed
Protracted descent <l.Ocm/h <2.0cm/h Expectancy and support
CS if CPD disclosed
Secondary arrest >2 h arrest in active phase Oxytocin only if CPD ruled out;
of dilatation CS if CPD probably present
Prolonged deceleration >3h >lh Oxytocin only if CPD ruled out;
CS if CPD probably present
Arrest of descent > 1 h arrest in second stage Oxytocin only if CPD ruled out;
CS if CPD probably present
Failure of descent No descent in second stage Oxytocin only if CPD ruled out;
CS if CPD probably present

CS, cesarean section; CPD, cephalopelvic disproportion.

3. Dystocia and "Failure to Progress" in Labor 29

TABLE 3.2. Types of labor conditions that failure to progress may represent
Labor condition Probability of diagnostic error Source of diagnostic error
Normal latent phase 100% Not an abnormality because it is
found in all labors without
exception
Prolonged latent phase 100% Not an indicator of obstructed labor
Arrest disorder 50% Unassociated with CPD in half the
Secondary arrest of dilatation cases
Prolonged deceleration phase
Arrest of descent
Failure of descent
Protraction disorder 70% Unassociated with CPD in more than
Protracted dilatation two-thirds of cases
Protracted descent
Prolonged second stage associated with ? (Unknown) Not all long second stages are the
Epidural anesthesia result of CPD; a prolonged second
Fetal malposition stage is not a sufficient indication
Maternal exhaustion per se for a cesarean
Ineffective uterine contractions
Poor expulsive efforts

CS, cesarean section; CPD, cephalopelvic disproportion.

along with an estimate of the probability of as high as 0.6 cm/h has been observed, but
making a clinical error if a diagnosis of this is rare. 22
failure to progress is invoked without However, it may be all too easy to be
further clarification and a cesarean section deluded into believing that the patient
is done forthwith on that basis alone. whose labor has not yet advanced beyond
Depending on the type of labor condition the normal latent phase is exhibiting the
that actually exists in a given case, this disorder of failure to progress for which
practice will frequently prove to have been aggressive intervention by cesarean section
the wrong choice. is indicated. This applies especially if the
observer is either unaware that that is what
is expected during early normal labor or,
Normal Latent Phase perhaps more often, does not fully appreciate
The condition of least significance insofar as that this patient is still in her latent phase.
it reflects a normal state of affairs is one This latter is the kind of misinterpretation
that is found in the course of every labor, that can arise when the cervix of a patient,
whether normal or not, namely the normal particularly a multipara, has already been
latent phase. During this initial part of the dilated somewhat in the days or weeks
first stage, nothing by way of active dilata- before the onset of her labor. This is the
tion or descent is supposed to be happening. gravida who is typically admitted to the
Careful observation will sometimes demon- labor unit some hours after labor begins and
strate some softening in the consistency of at that time is found to have a well-dilated
the cervix and even some minor degrees of cervix, say 4 cm. The impression is that her
effacement, especially in multiparas. Efface- labor must be in the active phase because
ment is usually complete in nulliparas long her cervix is so dilated. This not uncommon
before labor begins. In some gravidas, again false impression is based on the belief held
primarily multiparas, even some dilatation by many that the degree of dilatation defines
can be detected before the active phase the onset of active phase, rather than the
begins; ifthis occurs at all, it takes place to a upswing of the dilatation curve. The active
very limited extent and very slowly. A rate phase can begin at almost any cervical
30 E.A. Friedman

dilatation, although it is admittedly most a careful, astute vaginal examination, one

likely to be in progress by the time the might even consider uterotonic stimulation
cervix reaches 4cm; however, just because if the contractile pattern warranted it.
the cervix is so far dilated does not perforce However, to act upon the misdiagnosis by
means that the patient is in the active intervening aggressively with a cesarean
phase. section under these circumstances would
Given this mindset, if serial observations be clearly unacceptable. To do so, in most
over the next several hours should show that instances, would mean that the cesarean
the cervix has not continued to dilate, the delivery was being done for nothing more
patient may become a source of consterna- than a normal latent phase.
tion to the attendant personnel who mis-
takenly conclude that a labor abnormality
exists in the form of failure to progress.
Prolonged Latent Phase
Their misdiagnosis stems from not having A less common error is diagnosing failure to
recognized that, contrary to expectations, progress in the presence of a prolonged
this woman was still in the latent phase of latent phase. Patients who begin labor and
labor; indeed, she may not yet even have do not make any progress in the latent phase
been in labor at all. The difficult differential for many hours, passing the time limit
diagnosis in such an instance involves beyond which nearly all normal gravidas
distinguishing between normal latent phase will have entered the active phase, are easily
and secondary arrest of dilatation; it must recognized as meeting all the diagnostic
be addressed in this scenario. More often criteria for this labor aberration. This condi-
than not, the situation is easily resolved; tion, when determined to exist, responds
that is, the correct diagnosis is made, by most favorably to a program of therapeutic
taking into consideration the relatively rest, usually in the form of some narcotic-
short interval that has elapsed from the analgesic agent administered for purposes of
onset of labor, a review of the known pre- providing a period of 6-8 h of sleep. During
labor dilatation documented by a recent out- this interval, uterine quiescence is usually
patient or office examination, and perhaps but not always achieved. When gravidas
the relatively inconsequential contractile managed in this way awaken, most will be
pattern that reflects the antecedent pro- found to have advanced well into the active
dromal or early labor. In most cases, the phase; they can be expected to proceed to
diagnosis of the normal latent phase should delivery uneventfully, although they are not
be obvious merely from the history alone; necessarily immune from the kinds of com-
the patient reports that the onset of regular plications that may affect any woman in
uterine contractions occurred only a rela- labor. In fact, as stated earlier, there may
tively short time ago. Knowledge about some cause for concern based upon a recent
what the limits of normal for a latent phase report suggesting that the prognosis, both
should be (see Table 3.1) and about the fact maternal and fetal, is not entirely optimistic,
that advanced dilatation does not necessarily especially as regards the increased fre-
mean that the patient must be in the active quency of subsequent labor disorders,
phase should provide the necessary insight. cesarean delivery, neonatal depression, and
If, in any given case, there is still some need for resuscitation. 25
remaining concern about whether the Most of the remaining patients who are
patient's condition represents true secondary rested and fail to advance into the active
arrest of dilatation in the active phase, as phase of labor after they awaken are found
opposed to a normal latent phase, there to be out of labor altogether; that is, their
would then be ample justification for under- uterine contractions have subsided com-
taking an evaluation for cephalopelvic dis- pletely, signifying that the foregoing event
proportion. Having ruled it out by means of represented nothing more than a long
3. Dystocia and "Failure to Progress" in Labor 31

interval of false labor. Thus, the narcotic it does not represent a serious labor aberra-
agent has served the very useful purpose of tion in the same sense as a major labor
offering a means for making the differential aberration, such as true arrest of dilatation
diagnosis between true labor and false labor. or of descent, does in terms of the kind of
It must follow from this that if cesarean associated conditions that may place mother
section had been undertaken for a patient in and fetus in jeopardy.
this subgroup, it would have been done in a There should be little if any clinical dif-
gravida who had been in false labor. It ficulty in differentiating prolonged latent
seems superfluous to add that this would be phase from any other condition in the course
unconscionable. oflabor. The diagnosis takes little diagnostic
In the past, morphine was advocated as acumen; if anything, it tends to be over-
the sedative of choice for patients with pro- diagnosed by obstetricians who appear to be
longed latent phase, and it has proved clini- impatient to medicate patients so as to give
cally effective. 22 However, a recent as-yet them a therapeutic trial of rest to ascertain
unpublished investigation indicates tht if they are in true or false labor. They thus
morphine may not be entirely safe when invoke the diagnosis before the latent phase
used for this objective (Oriol N). Oximetry has actually exceeded normal limits, and
studies conducted under the aegis of an proceed with the therapeutic rest regimen.
obstetric anesthesia service showed that The end result is the same, but the gravida's
maternal oxygen saturation frequently fell labor has been unduly prolonged as a con-
after morphine was given to patients in sequence of the unnecessary use of the
labor, even without perceptible changes in inhibitory drug.
respiratory rate or the appearance of clini-
cally discernible cyanosis; it even fell below
an acceptable 80% level in 2 of 13 women.
Arrest Disorders
While no adverse effects have as yet been The four identifiable arrest disorders (see
detected among fetuses in these cases, Table 3.1) constitute the most serious of the
the observation of resulting maternal labor aberrations that can be diagnosed
hypoxemia, however mild, serves to warn objectively by means ofthe graphic analysis.
against use of depressing narcotic agents in They are serious because they occur so often
these patients, particularly if there is some in gravidas whose labor is obstructed, that
concurrently ongoing or potential problem is, who have insurmountable cephalopelvic
with regard to fetal oxygenation. Some disproportion (CPD). The incidence of docu-
consideration should therefore be given to mentable CPD among women who develop
substituting other, potentially less haz- one or more of the arrest disorders is about
ardous agents for morphine, such as bar- 50%; it is much higher (clinical experience
biturates or tocolytics. Limited experience suggests it may be as high as 90% but more
with the latter, not as yet fully evaluated or precise data are as yet unavailable) if
formally reported, suggest they may prove the arrest occurs while oxytocin is being
worthy for this purpose. administered or continues uncorrected for
As to the false designation of patients any length of time despite uterotonic
with prolonged latent phase as having augmentation by oxytocin.
dystocia or failure to progress and therefore As a basic management principle, it is
warranting cesarean section, the latter mandatory that gravidas who demonstrate
diagnoses are not supportable and the inter- an arrest disorder be evaluated for CPD
vention assuredly cannot be condoned. before any decisions are made about man-
Cesarean section done merely for prolonged agement. If CPD is present, or cannot be
latent phase is both inappropriate and ruled out, cesarean section would appear to
unnecessary. Lack of progression is charac- be entirely justified. The risks of permitting
teristic of this labor disorder, of course, but such patients to continue in their labors or
32 E.A. Friedman

to administer oxytocin to offer a further trial helps to focus on a common fault in man-
of labor are great. 22 Nonetheless, questions aging patients whose labors are poorly pro-
are now being raised as to whether the gressive or not progressive at all strictly on
ostensible risks are sufficiently real to the basis of ineffective contraction. This
warrant such a recommendation for man- fault consists of the failure to pay sufficient
agement. 26 The American College of Obste- heed to the effort to correct the contractions
tricians and Gynecologists (ACOG) has even for a long enough period of time to give the
issued a statement advocating the use of gravida an opportunity to effect progress if
oxytocin for use in cases of arrest of dilata- she is capable of doing so.
tion (but not arrest of descent) if the contrac- lt should be appreciated, however, that
tions are deemed inadequate. 27 some of these patients have developed poor
The criteria suggested for determining contractility patterns because the uterus
adequacy of uterine contractions include has evolved this pattern over time in re-
consideration of an objective determination sponse (assigning a teleologic purpose to it)
of the contractions by use of Montevideo to a physical impediment to progress, that
units, which are derived by the summation is, to CPD. In other words, just as a hyper-
of the amplitude, minus the basal tonus, of contractility pattern (frequent contractions
all contractions that occur in an interval of high amplitude, often with coupling of
of 10 min. Specifically, an adequate con- contractions) may serve as an index of uterine
tractility pattern is defined as one that overreaction to obstructed labor, so may
generates at least 200 Montevideo units (per a hypocontractile pattern. In fact, both
10-min window) over a period of2h in labor. may occur seriatim in a given labor. One
However, for this measurement of con- generally sees hypocontractility, reflected in
tractility to be at all meaningful, it re- what is clinically perceived as inadequate
quires the use of an indwelling intrauterine contractions, later in labor than hypercon-
pressure catheter that will yield a properly tractility; this perhaps reflects the exhaus-
calibrated continuous strain gauge re- tion of some myokinetic mechanism over the
cording. The use of external tokodynamo- course of time. To assume all instances of
metry is valueless in this regard and must poor contractility represent some intrinsic
not be relied upon for purposes of evaluating myometrial dysfunction is just as wrong as
the contractility. Assessment by palpation of assuming all are the result of disproportion.
the contractions with the examining hand Evaluation of the individual patient is
is far better, even though subjective and clearly in order to determine which she
not capable of providing a quantifiable represents.
measurement. One can applaud the admonition implied
The lack of an adequate scientific founda- in the ACOG pronouncement to the effect
tion for this recommendation as to what that cesarean section is not indicated in
constitutes a minimally acceptable con- all such cases; at the same time, however,
tractility pattern needs to be addressed be- cesarean section may actually prove to be
cause it is based on data pertaining to the appropriate in some of them. Individualiza-
level of measured contractility (in Monte- tion is imperative. To advocate routine use
video units) achieved by 90% of parturients of oxytocin for enhancing the contractions
who respond favorably to oxytocin by suc- under those circumstances would expose
cessful cervical dilatation and subsequent those gravidas in whom the poor uterine
fetal descent and vaginal delivery, not on contraction pattern signals CPD to poten-
what is minimally necessary to effect dilata- tially serious hazard. The practice, if com-
tion and descent in patients with an arrest plied with widely, would in effect turn the
or protraction disorder. 29 Nevertheless, it clock back to a time when oxytocin was used
does have the advantage of at least being almost indiscriminately, and without clear
objective and reproducible. Moreover, it understanding about what constituted its
3. Dystocia and "Failure to Progress" in Labor 33

contraindications, with rather unfortunate and identifying those women whose pelvic
results. capacity is limited.
With respect to the catch-basket inclusion Data on fetal growth over pregnancy,
of all arrest disorders within the diagnostic uterine size by abdominal palpation and
designation of failure to progress, aside from fundal measurement, and fetal size by
the nomenclatural confusion it fosters there ultrasonographic imaging, if indicated (by
is no strong clinical argument against suspicion of macrosomia, small pelvis, small
its use, with the proviso that it be strictly maternal stature, maternal diabetes, large
limited to true arrest disorders. There is an uterus, and other relevant indicators of
academic argument that can be brought to potential problems), will be supplemented in
bear, however. Before the several graphic the course of labor by additional examina-
labor disorders were formally described tion to evaluate the relationship between
and named, the obstetrical literature was the maternal pelvis and the fetal head. The
marked by a profusion of terms (regrettably, fetal head is obviously the "pelvimeter" of
a few are still in use in some areas around most critical relevance. How the head fits
the world) to designate disorders of labor into the mother's pelvis and whether there
progress. A total of more than 60 were is sufficient room to accommodate safe pass-
collected at one time in just the English age are questions that can be answered by
language literature alone. 22 This Babel timely, skillful, intelligent, knowledgeable,
came about largely because of the very and detailed vaginal examination.
subjective nature of clinical evaluation of Of special importance in this examina-
the labor phenomenon then in vogue. With tion is a dynamic assessment of the ce-
the introduction of greater objectivity in phalopelvic relationships; that is, one that
labor assessment, it was no longer necess- determines the ability of the fetal head to
ary to cloak our collective ignorance in be thrust into the pelvis in response to the
terminologic obfuscation. Introducing anticipated forces of labor. This examina-
newer terms at this time, without apparent tion, called the Mueller-Hillis maneuver
need and without clear-cut and well-defined (MHM),29 is achieved by undertaking a
criteria does not seem justified, and should vaginal examination during a uterine con-
be resisted. traction with an assistant applying firm,
The question of how one determines the but gentle, downward pressure against the
presence or absence of CPD deserves further uterine fundus (Kristellar or fundal pres-
consideration because it is crucial to the sure). If cervical dilatation is well advanced
management of patients with arrest dis- so that the lower uterine segment offers
orders, and to an equal degree of those with little resistance to caudad thrust of the fetal
protraction disorders (see following). The head, the head will ordinarily be felt to
only currently acceptable technique is clini- descend in the pelvis by 1-2cm; flexion and
cal pelvimetry by means of vaginal ex- internal rotation may also be perceived
amination for assessing cephalopelvic simultaneously. These are indicators of
relations. Prenatal clinical pelvimetry is ample room for future descent and clearly
useful for identifying a woman with a small demonstrate that there can be no CPD, at
or anatomically misshapen pelvis, an admit- least down to the level to which the head
tedly rare finding today. More pertinently, descended.
pelvic examination should be done on a If the head is fixed at its current station
routine basis for every prenatal patient at and does not descend at all when this
the first antepartum visit and again late in examination is done while simultaneous
the course of pregnancy for determining the endogenous and exogenous forces are being
architectural characteristics and capacity of applied, CPD is likely to exist. A recent
the pelvis. This information is important question has been raised as to the validity
for anticipating the mechanism of labor of this approach by Thorp et al.,30 who con-
34 E.A. Friedman

cluded that this valuable clinical tool should preciously little else available in our diagnos-
be summarily abandoned because they were tic armamentarium to replace it. Moreover,
unable to show that it could predict future its use has been recommended for these
problems in the course of labor. They tested purposes by the ACOG. 27
its predictive value well in advance of the
time in late active-phase labor when the
pelvic capacity is actually being tested
Protraction Disorders
by the fetal head, in effect using it as a It may seem inappropriate to include the two
screening device for CPD. Although the protraction disorders here in our discussion
original descriptions of the method did of the kinds of conditions that may be mis-
suggest it for this purpose, that is not the takenly called failure to progress because
manner in which it is used in clinical prac- continuous progress is being made (by defini-
tice, as emphasized as long ago as 1938 by tion) as these labor aberrations evolve, albeit
Hillis. 29 More recently, Greenhill and 131 abnormally slow. However, by virtue of the
recommended using this maneuver in late very slow progress that occurs in cervical
active phase and especially in second stage, dilatation or fetal descent when these con-
when pelvic architecture plays a major role ditions exist, attendant personnel may
in shaping the course of fetal descent. unwittingly believe that no real progress is
The primary value of the MHM is to actually occurring. The error in diagnosis
diagnose an obstruction to labor progress means that these patients will be managed
when the fetal presenting part has reached much more aggressively, and therefore
and impacted against the pelvic impediment, inappropriately, than is warranted by the
specifically when labor is well-advanced and actual labor pattern they exhibit.
particularly after the dilatation or descent Given the recognized degree of inter-
pattern has signaled the problem by evolv- observer error in estimating cervical dilata-
ing an arrest or protraction disorder. At tion and fetal station, there may not be any'
minimum, 5 cm cervical dilatation is required recorded changes over time; alternatively,
to ensure that the cervix itself offers negli- small recorded advances may be interpreted
gible resistance to the thrust of the fetal head to represent no change at all. Therefore, a
when fundal pressure is applied. Further, the protraction disorder may exist in a gravida
downward pressure should be applied during and remain undiagnosed. More to the point,
a uterine contraction (not between contrac- failure to progress may be diagnosed instead.
tions, as reported by Thorp et al. 30) to provide If this occurs and it leads to a management
a fair test of what the combination of contrac- option in which intervention by cesarean
tile force and firm fundal pressure is capable of section occurs, it would be most unfortunate
achieving. If the head is fixed under these because about 70% of such women can be
circumstances, disproportion is very likely; if delivered safely by the vaginal route.
it descends and rotates, bony dystocia is effec- Although little is known about what
tively ruled out, at least at that fetal station. causes the typical dilatory progress seen in
As undertaken by Thorpe et al., 30 the testing gravidas with protraction disorders, it
of the predictive value of the MHM in early is recognized that about one-third have
labor under conditions not designed to assess associated CPD. This relationship alone
its true diagnostic capability almost guar- makes the protraction disorders clinically
anteed that the study would show no effect. important, but because so many affected
Therefore, their recommendation to abandon women do not have CPD, it is worth empha-
the use of the MHM in clinical practice must sizing that uniform management by cesarean
be considered inappropriate. It is probably too section is clearly not warranted. Knowing
soon to forsake a tried-and-true method that this, the obstetrician confronted with such a
has been shown to be so clinically valuable patient wisely undertakes to determine
for so long, particularly because there is whether or not it is safe to allow vaginal
3. Dystocia and "Failure to Progress" in Labor 35

delivery. If CPD is encountered, cesarean delivery is accomplished, provided no pro-

section is indeed appropriate. This does not blems arise with regard to fetal or maternal
intend to suggest that every woman with a wellbeing. Deserving of special mention is
protraction disorder should be subjected to the requirement to avoid the temptation to
operative delivery, however, because if undertake forceps delivery in these case. It
careful and thorough assessment of the is understandable that one should wish to
cephalopelvic relations shows that CPD is effect delivery as the second stage drags on,
unlikely, a program of expectancy and but that temptation is best resisted in the
support can result in a safe vaginal delivery. interest of ensuring against the fetal injury
The aforementioned ACOG statement to which these infants may be exposed even
recommends the use of oxytocin for these by what may seem to be an easy outlet-type
patients if the contractility pattern is in- forceps delivery.32,33 It is as yet unclear
adequate, using the Montevideo unitage why the morbidity and mortality for these
criteria detailed previously.27 Although ad- offspring should be increased in association
ministration of oxytocin is intuitively a with any form of instrumental delivery, but
reasonable option, there is no documented given the observation that such is the case,
basis for this recommendation, particularly one should attempt instead to seek a safer
in view of the fact that the course of dilata- spontaneous delivery, if at all possible.
tion and descent has been shown to be Although those parturients with obvious
unaffected by oxytocin in cases of protrac- CPD are not permitted to undergo this
tion disorder. 22 There is an apparent and as management option, having been winnowed
yet unexplained dissociation between the out and delivered by cesarean section earlier,
contractile pattern and the effect that the those few gravidas with undetected CPD
uterine contractions has with regard to will have taken this path. Accordingly, one
the work they are able to accomplish in di- can expect some of the women who were
lating the cervix and causing the fetus to treated expectantly to fail along the way.
descend. Moreover, while oxytocin does no The most important indicator signaling
apparent good when used for stimulating such failure is the development of an ar-
the uterus to contract in these cases, it has rest disorder, superimposed on the pre-
the potential to cause harm, especially if it existing protraction pattern. When the labor
results in fetal hypoxia from hypercon- of a patient with a protraction disorder be-
tractility or physical trauma from excessive comes arrested, it serves as a strong portent
head deformation in the presence of CPD. of obstructed labor. The prognosis for a
The conservative program of expectancy safe vaginal delivery among them is very
and support consists of detailed discussion guarded; cesarean section is an option that
with the patient and her significant other to deserves consideration at this point without
educate them about what is in store for further trial of labor.
them. The labor that follows will be uncom-
monly long; there will be need to mobilize
and provide psychologic supports; careful
Prolonged Second Stage
attention must be paid to fluid needs and For many years, it was believed that a
electrolyte balance; analgesia and anes- second stage exceeding 2 h exposed the fetus
thesia must of necessity be stringently to considerable hazard. This firmly held
limited because the slow labor progress may concept, probably originating early this
be worsened by use of inhibitory agents to century by intuitive logic rather than objec-
which the uterus is especially sensitive; tive substantiation,34 appears to have been
careful maternal and fetal monitoring has to given major impetus and authoritative
be maintained. support from a report by Hellman and
Labors conducted in this way can be Prystowsky35 showing that the longer the
expected to continue to advance until vaginal second stage lasted, the greater the risk of
36 E.A. Friedman

infant mortality and maternal hemorrhage. the specific aberration, if it does exist, is
Additional buttressing came from the important under these circumstances. Even
observation that the probability of the fetus more important is the evaluation that should
developing acidosis increased with progres- flow from this recognition, with emphasis on
sive prolongation of the second-stage dura- disclosing any CPD that may account for the
tion. 36 In a more recent and larger series of abnormal labor pattern.
nulliparas, some with greatly prolonged Again, if one is unable to ensure that
second stage, Cohen37 was able to demon- there is ample room for safe vaginal delivery,
strate a contrary view to the effect that cesarean section will ordinarily be quite
there was no discernible increase in peri- justified. For the rest, it is clear that vaginal
natal mortality with progressive lengthen- delivery should be achievable. To accomplish
ing of the second stage. A number of other this, careful maternal and fetal surveillance
investigations have confirmed these obser- is needed to ensure that continued progress
vations as weI1. 33 ,38-41 is being made, albeit slowly, and that both
Similarly, while there is good documenta- mother and fetus remain in good condition.
tion that increasing levels of lactate appear If necessary and called for on the basis of an
in fetal umbilical arterial blood with in- inadequate contractility pattern, oxytocin
creasing duration of second stage, inversely augmentation of the uterine contractions
related to the falling pH levels, it has been may prove beneficial as well. As long as
demonstrated that this partly reflects things continue to go well, and one is rea-
maternal lactate levels plus the accumula- sonably certain that there is no anatomic
tion and excess production of lactate in the obstruction that would make fetal descent
fetus, but not necessarily from fetal hypox- hazardous, allowing continued labor is not
emia. 42 The safe limits of this process are only acceptable but entirely appropriate.
unknown, but continuous monitoring of the Therefore, expectancy and careful surveil-
acid-base balance in the human fetus lance provide a useful and apparently safe
carried through the delivery process has alternative to both cesarean section and
determined that recovery from the resulting midforceps delivery, the latter having been
mild respiratory acidosis occurs very quickly, done heretofore for no better reason than
indicating the likelihood that its impact of that the second-stage duration exceeded an
the fetus is negligible or nonexistent. 43 arbitrary 2-h limit.
The recommendations for management of
the prolonged second stage derived from the
observations of no ill effect from long second Quantifying the Potential
stages include admonitions to avoid reflex-
ively considering cesarean section when a Impact on the Cesarean Section
patient's second stage extends beyond any Rate
arbitrary duration. 37 Thus, allowing pa-
tients to proceed into a long second stage is A plea can be made that all those cases in
quite acceptable, even if descent is occurring which failure to progress or dystocia is
at an abnormally slow rate, with the import- invoked as a diagnosis leading to cesarean
ant proviso that appropriate electronic fetal delivery should instead be designated
heart rate monitoring is done on a con- according to a more specific diagnosis based
tinuous basis and CPD has been definitively on a more objective graphic assessment of
ruled out beforehand. This is not intended to the labor pattern. If this were done, suitable
give license to ignore such gravidas. On the evaluation would then be undertaken to
contrary, prolongation of the second stage determine the associated problems and the
may reflect the existence of one or another of ensuing management would address the
the arrest or protraction disorders detailed individual needs of each patient. Were this
earlier in this chapter. Correct diagnosis of to be done in all cases, a substantial propor-
3. Dystocia and "Failure to Progress" in Labor 37

tion of cases now subjected to cesarean sec- potential VBAC patients vaginally in larger
tion will be permitted to deliver vaginally. numbers. Furthermore, if this program were
Given the figures for the errors inherent to be applied at those institutions at which
in a diagnosis of failure to progress or dys- the cesarean section rate is now far in excess
tocia tallied in Table 3.2 for cases in which of the national average, in some approaching
specific labor aberrations are present, coup- a startling 50%, the benefit would be even
led with general population data for the greater in relative terms.
incidence of these several conditions,22 it This in fact comes very close to the
should be possible to approximate the reduc- national objective of 15 percent that has
tion in cesarean section rate attainable if been stated by federal health officials as the
obstetrical practices were modified accord- goal for the year 2000. This worthy goal has
ingly. To quantitate the rate reduction, such been deemed unlikely to be achievable by
a calculation would be based on the assump- the ACOG and has already been effectively
tion that vaginal delivery will take place abandoned by the CDC. 1 It can be seen from
for all those patients who are now being this brief analysis that not only is this goal
delivered by cesarean section in the presence within the realm of feasibility, but it should
of a labor pattern characterizable as (1) the be readily reachable and, in concert with
latent phase, whether normal or prolonged efforts to enhance the frequency of success-
in duration, (2) an arrest disorder unas- ful VBACs, easily exceeded.
sociated with CPD, (3) a protraction disorder The impact such a program would have on
in a gravida with ample cephalopelvic rela- the cesarean section rate for nulliparas is
tions, and (4) a prolonged second stage even more impressive. We can apply the
associated with a poor contractile pattern same simple calculations to the current
unrelated to anatomic obstruction. This cesarean frequency of 30% or more,l1 about
would result in definitively reducing the half or more of which are being done for
cesarean rate for women delivered in the failure to progress or dystocia. We can
latent phase by 100%, for those with arrest derive a conservative estimate of 9.8% for
disorders by 50%, for those with protraction the reduction attainable by the focused
disorders by 70%, and for those with long management principles detailed here. The
second stages by 60% (because of the relative result of this calculation is that the overall
impact of the arrest and protraction dis- cesarean rate for nulliparas would fall to
orders that make up this group). In total, no 20.2%. If the true relative frequency of
fewer than 65% of the cesarean sections now cesarean sections done for dystocia or failure
undertaken for failure to progress would not to progress is as high as 66%, as has been
be done. reported in at least one study, 12 then the
What might this mean in terms of its absolute overall reduction would total
impact on the actual cesarean section rate? 12.9%, leaving a section rate among nul-
If we accept the current national cesarean liparas of 17.1%. If reached, this would
rate as 23.5%, a figure just announced by clearly constitute a major step in the right
the CDC for 1991 data, 1 and embrace the direction.
generally agreed-upon concept that as many In an attempt to achieve this some states
as 40% are being done for failure to progress have introduced public health care legisla-
or dystocia, then not doing 65% of the tion to test and, more recently, to mandate
cesarean sections now being done for these such obstetric practices based on the
diagnoses would yield an absolute reduction recommendations of the NIH Consensus
in the overall rate of 6.1 %. This would drop Development statement on cesarean child-
the overall cesarean section rate to 17.4%, birth, first formulated in 1980.13 These
and this does not even begin to take into recommendations included standardizing
account what might be attainable simul- nomenclature by replacing the term "failure
taneously by concerted efforts to deliver to progress" with diagnoses based on the
38 E.A. Friedman

graphic labor curve pattern or equivalent second stage), plus encouragement to use
labor definitions, and utilizing a hospital oxytocin to augment nonprogressive labor
quality control system composed of internal where indicated. The results of this program
institutional reviews of management prac- in reducing the cesarean section rate over a
tices supplemented by external reviews to 4-year period, 1985-1989, were even better
ensure compliance. In addition to requiring than expected. Overall, it fell from 23% to
good chart documentation, use of the labor 12%, with an increase in the frequency of
curves or some form of equivalent docu- successful VBACs and a remarkable reduc-
mentation, and more complete chart notes tion in the rate of cesarean sections for
and operative notes, such an implemented dystocia among nulliparas by half, from 16%
program would insist on updating, modify- to 8%. Indeed, the latter accounted for most
ing, or developing protocols for obstetric of the overall decline in the cesarean section
patient care; education of the staff and rate in this series.
the affected patients, especially as regards In New York state, the Department of
VBAC and fetal monitoring; quality assur- Health has since taken the unusually
ance activities with increased participation enlightened step of promulgating rules
by physicians and nurses; and appropriate making the use of failure to progress and
use of oxytocin for labor augmentation. dystocia unacceptable as diagnoses for
A program designed precisely along these which cesarean section could be done, and
lines and aimed at monitoring quality of requiring all formal diagnoses of labor dis-
obstetric care in a sampling of one state's orders to be based on the graphic patterns
obstetric units, including 176 hospitals, has discussed earlier. It is too soon to know what
actually had a clinical demonstration trial, the impact of a statewide program such as
as reported by Dillon et a1. 17 The preliminary this will be on the cesarean section rate, but
results are indeed very gratifying. The state if it proves successful- and it is reasonable
cesarean section rate fell from 25% in 1987, to predict that it will from the aforemen-
just before the program was introduced, to tioned pilot study results - it is likely to
23.6% in 1990 and 23.4% in the first seven serve as a template for other states to follow.
months of 1991. Even greater rate reduc- It is regrettable only in that it has taken
tions were seen in the hospitals in which the such governmental action to force physicians
program had been implemented, and still to do what is clearly indicated and in the
stronger declines among those 24 units that best interests of their patients.
had been additionally subjected to external
review.
Lest this be considered just the result of
imposing strong rules in cosmopolitan or Alternative and
academic institutions, it should be noted Supplementary Programs
that another similar clinical trial was 'con-
ducted in a rural community hospital in The most striking development in operative
Canada. 12 Criteria were introduced for the obstetrics during the last 25 years has been the
diagnosis of dystocia based upon the 1986 extension ofthe operation of caesarean section, as
National Consensus Conference on Aspects a means of combating the complications of
of Cesarean Birth.15 These specified, for parturition ... It would be a sad day for prospec-
example, that the diagnosis could not be tive mothers if caesarean section were to run riot
made unless the labor was documented to and were to be indiscriminately employed for
complications which can be quite as well treated
have entered the active phase (designated
by ordinary vaginal procedures. 44
by cervical dilatation of at least 3 cm); there
was acceptance of delayed labor to allow This insightful statement by Munro Kerr,
vaginal delivery (that is, no mindset to effect the obstetric doyen of a previous generation,
delivery after an arbitrary duration of was written 30 years ago and is interesting
3. Dystocia and "Failure to Progress" in Labor 39

because it shows how accurately today's There is ample documentation to the

climate in which cesarean section has "run effect that there are often subtle (and some-
riot" could be predicted. It was cited by times not so subtle) factors at play in the
Seiler45 in a wistful proposal advocating a decision of doctors to undertake cesarean
retrogressive reintroduction of such aban- section for dystocia. The higher cesarean
doned practices as vaginal breech delivery rates among private patients, whose section
and midforceps for "correctable dystocia" for rates ought to be lower than those of non-
purposes of reducing the cesarean section private gravidas by virtue ofthe differential
rate. Although one cannot support or even incidence of maternal and fetal complica-
take seriously Seiler's proposal to turn the tions, raises the specter of the physician's
clock back and encourage the resurgence of role in influencing the choice of delivery
procedures that fell from favor because of mode based on hidden agenda items, includ-
their adverse consequences,46 it does at least ing such unacceptable reasons as con-
recognize that many of today's cesarean venience. 50- 56 Taylor57 attributed the lower
sections are actually done for dystocia that rates on nonprivate services of community
is clinically correctable by means other than hospitals and in patients delivered at
cesarean section so that vaginal delivery county, university, and military hospitals to
can be accomplished. The solution that is the more objective approaches to manage-
recommended, however, is anachronistic. ment at those units and the shared patient
Are there more acceptable approaches? care responsibility that exists there, imply-
Aside from the one just outlined and cham- ing oversight. He attributed the higher
pioned, a program with growing popularity cesarean section rates for dystocia among
is that of active management of labor as private patients to the fact that prolonged
advocated by O'Driscoll et a1. 47 ,48Its details labor is not tolerated by patients, that
and merits are beyond the scope of this oxytocin augmentation is used less, and
review (see Chapter 4). It is valid, neverthe- when used, may alarm the obstetrician
less, to point out that much of the benefit because of associated fetal heart rate
accrued from the active management pro- changes, and that labor progress may
gram appears to come primarily from the be interfered with by the epidural anes-
total commitment of the both physician and thesia that is used more often among these
nursing staffs, the use of strict criteria for gravidas. 58 Patient tolerance of labor is
diagnosing that a patient is truly in labor directly related to their prelabor education
(no small task), standardization of care, and intrapartum support, an intrinsic feature
organization and efficient functioning of the of active management. Use of oxytocin has
labor unit, a strong patient enducation not been shown to be a real factor in deter-
program, and one-on-one attendance and mining the type of delivery in cases of "func-
supervision of each gravida by a committed tional dystocia" as defined by Seitchik et
and knowledgeable nurse labor monitor (not al.,59,60 although its benefits are widely
the uninformed and emotionally involved extolled nevertheless.
significant other). While routine administra- The use of oxytocin for active manage-
tion of oxytocin to women who are docu- ment of the patient who is not progressing
mented to be in labor but who do not show adequately does apparently serve to ensure
the requisite minimally acceptable rate against being deluded into considering
of dilatation of 1 cmlh is undoubtedly impor- a patient who is actually still in the latent
tant in the successes reported with this phase as if she were exhibiting some
approach, it is perhaps of less relevance abnormal state of dystocia for which cesarean
because, if the uterine stimulation is really section should be done. It does not guard
limited only to those fulfilling the stated completely against other abuses of latent
criteria, it should apply to relatively few phase management, however, specifically
patients. 49 because by advocating routine administra-
40 E.A. Friedman

tion of oxytocin, the occasional patient who 3. Anderson GM, Lomas J. Recent trends in
may instead actually be in false labor will cesarean section in Ontario. Can Med Assoc
receive it and her failure to respond (to the J 1989;141:1049-1053.
inadvertant induction of labor) may be per- 4. Gleicher N. Cesarean section rates in the
ceived as a labor aberration worthy of United States: the short-term failure of the
National Consensus Development Con-
aggressive intervention by cesarean section,
ference in 1980. JAMA 1984;252:3273-3276.
the very problem the use of oxytocin was 5. Quilligan EJ. Cesarean section, 1988: To
trying to prevent. Moreover, if the patient have or have not! West J Med 1988;149:700-
actually has a protraction or an arrest dis- 703.
order to account for the dilatory progress, 6. Marieskind HI. An evaluation of cesarean
the labor augmentation will be providing a section in the United States: final report.
form of therapy many will later need. Thus, Washington, DC: U.S. Department of Health
at least one-half of the affected patients with and Human Services, Office of Assistant
arrest disorders and more than two-thirds of Secretary for Planning and Evaluation/
those with protraction disorders will per- Health,1979:16-18.
force get the therapy that is actually in- 7. Hibbard LT. Changing trends in cesarean
section. Am J Obstet Gynecol 1976;125:798-
dicated for them. This is not necessarily an
804.
unqualified blessing. 8. Mann LI, Gallant J. Modern indications for
The cases referred to here are those who, cesarean section. Am J Obstet Gynecol 1979;
under the program described, would in due 135:437 -441.
course have been diagnosed as having the 9. National Center for Health Statistics.
specific labor aberration; the diagnosis Detailed diagnosis and procedures, National
made, they would then have been evaluated Hospital Discharge Survey, 1987. Wash-
for its cause with dynamic clinical pel- ington, DC: Vital Health Statistics 1989;13:
vimetry and shown not to have demon- 1-304.
strable CPD. Because active management 10. Sokol RJ, Rosen MG, Bottoms SF, et al. Risks
does not take the precaution of addressing preceding increased primary cesarean birth
the issue of undetected CPD, however, rates. Obstet Gynecol 1982;59:340-346.
11. Boylan PC, Frankowski R. Dystocia, parity
administration of oxytocin would have the
and the cesarean problem. Am J Obstet
singular disadvantage of subjecting. some Gynecol 1986;155:455-456.
gravidas and their fetuses to uterotonic 12. Iglesias S, Burn R, Saunders LD. Reducing
stimulation contrary to currently acceptable the cesarean section rate in a rural com-
standards, which interdict such use in the munity hospital. Can Med Assoc J 1991;145:
presence of CPD because of the known 1459-1461.
hazards it may involve. In two studies in 13. NIH Consensus Development, 1980: state-
which active management was studied with ment on cesarean childbirth. NIH Publica-
these concerns in mind, there was indeed a tion No. 82-2067. Bethesda, Maryland:
reduction in the cesarean rate for dystocia National Institutes of Health, 1980.
among nulliparas, but concomitantly there 14. NIH Consensus Development statement on
cesarean childbirth: the Cesarean Birth Task
was an almost equal and counterbalancing
Force. Obstet GynecoI1981;57:537-545.
increase in the frequency of cesarean sec- 15. Indications for cesarean section: final state-
tions for fetal distress. 49,61 ment of the panel of the National Consensus
Conference on Aspects of Cesarean Birth.
Can Med Assoc J 1986;134:1348-1352.
References 16. Rosen MG, Alper MH, Bloomfield R, et al.
1. u.s. says 349,000 cesareans in 1991 were not NIH Consensus Development Task Force
necessary. New York Times, April 23, 1993. statement on cesarean childbirth. Am J
2. Notzon FC, Placek PJ, Taffel SM. Com- Obstet Gynecol 1981;248:1082-1084.
parison of national cesarean section rates. N 17. Dillon WP, Choate JW, Nussbaum ML, et al.
Engl J Med 1987;316:386-389. Obstetrical care and cesarean birth rates: a
3. Dystocia and "Failure to Progress" in Labor 41

program to monitor quality of care. Obstet rics. Philadl;)lphia: WB Saunders, 1974:651.

Gynecol1992;80:731-737. 32. Friedman EA. Effects of labor and delivery
18. Porreco RP. Meeting the challenge of the on the fetus. In: Cohen MR, Acker DB,
rising cesarean birth rate. Obstet Gynecol Friedman EA, eds. Management of labor,
1990;75:133-136. 2nd Ed. Rockville, Maryland: Aspen, 1989:
19. Friedman EA. Failure of labor progression. 461-477.
In: Queenan JT, ed. Management of high risk 33. Friedman EA, Neff RK. Labor and deliv-
pregnancy, 2nd Ed. Oradell, New Jersey: ery: impact on offspring. Littleton, Massa-
Medical Economics, 1985:584-594. chusetts: PSG Publishing, 1987:130-145.
20. Friedman EA. Failure to progress. In: 34. Cohen WR. The pelvic division of labor. In:
Friedman EA, Acker DB, Sachs BP, eds. Cohen WR, Acker DB, Friedman EA, eds.
Obstetrical decision making, 2nd Ed. Toronto: Management of labor, 2nd Ed. Rockville,
BC Decker/St Louis: CV Mosby, 1987:242- Maryland: Aspen, 1989:19-60.
243. 35. Hellman LM, Prystowsky H. The duration of
21. Friedman EA. Failure to progress during the second stage of labor. Am J Obstet
labor. Contemp Ob/Gyn 1989;34(6):42-47. Gynecol 1952;63:1223-1233.
22. Friedman EA. Labor: clinical evaluation and 36. Jacobson L, Rooth G. Interpretative aspects
management, 2nd Ed. New York: Appleton- on the acid-base composition and its varia-
Century-Crofts, 1978:61-135. tion in fetal scalp blood and maternal blood
23. Krapohl AJ, Myers GG, Caldeyro-Barcia R. during labor. J Obstet Gynaecol Br Commonw
Uterine contractions in spontaneous labor: a 1971;78:971-980.
quantitative study. Am J Obstet Gynecol 37. Cohen MR. Influence of the duration of
1970;106:378-389. second stage labor on perinatal outcome and
24. Seitchik J, Chatkoff ML. Intrauterine pres- puerperal morbidity. Obstet Gynecol 1977;
sure wave form characteristics in hypo- 49:266-269.
contractile labor before and after oxytocin 38. Kadar N. The second stage. In: Studd J, ed.
administration. Am J Obstet Gynecol 1975; The management of labour. Oxford: Black-
123:426-432. well, 1985:271.
25. Chelmow D, Kilpatrick SJ, Laros RK. 39. Saunders NSG, Paterson CM, Wadsworth J.
Maternal and neonatal outcomes after Neonatal and maternal morbidity in relation
prolonged latent phase. Obstet Gynecol 1993; to the length of second stage in labour. Br J
81:486-491. Obstet Gynaecol1992;99:381-385.
26. Rosen MG, Debanne SM, Thompson K. 40. Paterson CM, Saunders NSG, Wadsworth J.
Arrest disorders and infant brain damage. The characteristics of the second stage of
Obstet Gynecol 1989;74:321-324. labour in 25,069 singleton deliveries in
27. American College of Obstetricians and North West Thames Health Region, 1988. Br
Gynecologists. Dystocia. ACOG Technical J Obstet Gynaecol 1992;99:377 -380.
Bulletin 137. Washington, DC: American 41. Kadar N, Cruddas M, Campbell S. Esti-
College of Obstetricians and Gynecologists, mating the probability of spontaneous
1989. delivery conditional on time spent in the
28. Hauth JC, Hankins GDV, Gilstrap LC, et al. second stage. Br J Obstet Gynaecol 1986;93:
Uterine contraction pressures with oxytocin 568-576.
induction/augmentation. Obstet Gynecol 42. Katz M, Lunenfeld E, Meizner I, et al. The
1986;68:305-309. effect of the duration of the second stage of
29. Hillis DS. Diagnosis of contracted pelvis. III labour on the acid-base state of the fetus. Br
Med J 1938;74:131-134. J Obstet Gynaecol1987;94:425-430.
30. Thorp JM. Study raises serious doubt about 43. Nickelsen C, Thomsen SG, Keber T. Con-
value of Mueller-Hillis method. Ob-Gyn tinuous acid-base assessment of the human
News, April 1, 1993; Thorp JM, Pahel-Short fetus during labour by tissue pH and trans-
L, Bowes WA. The Mueller-Hillis maneuver: ient carbon dioxide monitoring. Br J Obstet
can it be used to predict dystocia? Obstet Gynaecol1985;92:220-225.
Gynecol 1993;82:519-522. 44. Moir JC, ed. Munro Kerr's operative
31. Greenhill JP, Friedman EA. Biological obstetrics, 7th Ed. London: Balliere Tindall
principles and modern practice of obstet- and Cox, 1964:3.
42 E.A. Friedman

45. Seiler JS. The demise of vaginal operative 54. Fraser W, Usher RH, McLean FH, et al.
obstetrics: a suggested plan for its revival. Temporal variation in rates of cesarean
Obstet Gynecol 1990;75:710-715. section for dystocia: does "convenience" play
46. Friedman EA. Midforceps? No. Clin Obstet a role? Am J Obstet Gynecol 1987;156: 300-
Gynecol 1987; 30:93. 304.
47. O'Driscoll K, Meagher D. Active manage- 55. Phillips RN, Thornton J, Gleicher N. Phy-
ment of labor. London: WB Saunders, 1980: sician bias in cesarean sections. JAMA 1982;
23-31, 130-138. 248:1082-1084.
48. O'Driscoll K, Foley M, MacDonald D. Active 56. Evans MI, Richardson DA, Sholl JS, et al.
management of labor as an alternative to Cesarean section assessment of the con-
cesarean section for dystocia. Obstet Gynecol venience factor. J Reprod Med 1984;29:670-
1984;63:485-490. 675.
49. Boylan P, Frankowski R, Rountree R, et al. 57. Taylor ES. Editorial comment. Obstet Gyne-
Effect of active management of labor on the col Surv 1990;45:55-56.
incidence of cesarean section for dystocia in 58. Taylor ES. Editorial comment. Obstet Gyne-
nulliparas. Am J Perinatol 1991;8:373-379. col Surv 1990;45:379-380.
50. Goyert Gl, Bottoms SF, Treadwell MC, et al. 59. Seitchik J, Holden AEC, Castillo M.
The physician factor in cesarean birth rates. Amniotomy and oxytocin treatment of func-
N Engl J Med 1989;320:706-709. tional dystocia and route of delivery. Am J
51. Gould JB, Davey B, Stafford RS. Socioecon- Obstet Gynecol 1986;155:585-592.
omic differences in rates of cesarean section. 60. Seitchik J, Holden AEC, Castillo M.
N Engl.J Med 198;321:233-239. Spontaneous rupture of membranes, func-
52. DeMott RG, Sandmire HF. The Green Bay tional dystocia, oytocin treatment, and route
cesarean section study: II. The physician of delivery. Am J Obstet Gynecol 1987;156:
factor as a determinant of cesarean birth for 125-130.
failed labor. Am J Obstet Gynecol 1992;166: 61. Satin AJ, Leveno KJ, Sherman ML, et al.
1799-1804. High- versus low-dose oxytocin for labor
53. Guillemette J, Fraser WD. Differences be- stimulation. Obstet Gynecol 1992;80:
tween obstetricians in cesarean section rates 111-116.
and the management of labour. Br J Obstet
Gynaecol 1992;99:105-108.
4
Active Management of Labor
PETER BOYLAN

Active management of labor (AML) is an Background

approach to the supervision of primigravid
labor developed during the past 30 years at In 1963 the originator of the concept, Kieran
the National Maternity Hospital, Dublin, O'Driscoll, was elected Master at the
Ireland. 1 The approach represents a philo- National Maternity Hospital (NMH). He
sophy that is aimed at ensuring that as observed a worrying high incidence of pro-
many women as possible achieve spon- longed labor, at that time defined as more
taneous vaginal birth with a minimum of than 36 h, and set about trying to decrease
appropriate intervention. The original in- the incidence. In many cases the labors
tention of the approach was to decrease were associated with instrumental vaginal
the incidence of prolonged labor. 2 In recent delivery with its associated trauma for both
times the approach has become associated mother and infant. It was felt that this
with low cesarean rates,3,4 and this has negative experience produced a permanent
stimulated considerable interest as a result revulsion to childbirth in a substantial
of the rapid escalation in cesarean birth number of women. The development of the
rates in developing countries over the past approach resulted in a rapid decline in the
decade. 5 ,6 incidence of prolonged labor to the extent
The basic principles of AML may be divided that a redefinition was possible, to 12 h, by
into two broad categories, organizational the early 1970s. 7 Over the years there have
and medical. The organizational category been many refinements to the approach, and
includes a regular series of antenatal educa- this chapter concerns the current implemen-
tion classes, the provision of a personal nurse tation of AML.
for each woman in labor, the involvement of
senior medical staff, responsibility delegated
to midwifery staff, and strict audit of out-
come. The specific medical components in-
Organizational Aspects
clude early diagnosis of labor, early rupture
of intact forewaters, and infusion of oxytocin
Antenatal Preparation
under strict guidelines. A series of six classes is held for all women
The organizational, or management, prin- expecting their first baby. Every primigra-
ciples are applicable to all women in labor, vida is encouraged to attend these classes,
but the medical interventions are applied with their partners if they so wish. The
only to primigravidae at term with a single purpose of the classes is to educate the
fetus, cephalic presentation, and no fetal attending women about what is going to
distress. happen in labor so that anxiety levels can be

43
44 P. Boylan

reduced and nothing unexpected will hap- senior resident on duty, equivalent in North
pen. Every aspect of the forthcoming birth is American terms to a junior faculty profes-
discussed; a copy of the partogram is given sional. Junior doctors are regarded purely as
to participants, and a tour of the labor ward postgraduate students and are there to learn
is also arranged. but not to make decisions about individual
cases, unless under the closest supervision
by senior medical staff.
Personal Nurse
Each woman is assigned her own personal
nurse when she is admitted to the labor
Midwifery Involvement
floor, and this nurse stays with the woman The labor ward is run on a day-to-day basis
throughout her labor, within the confines of by the senior midwife on duty, and she has
duty rosters and standard breaks. The nurse at least two other midwives reporting directly
assigned is a student midwife who has to her. There are in addition six student
already completed her basic nursing training midwives on anyone shift. The senior mid-
over a period of 3 years. These nurses are wife is directly responsible for the running
undergoing a 2-year course leading to a of the labor ward but liaises directly, and
degree in Midwifery. Medical students closely, with the senior medical staff. The
attending the hospital function in the same approach is regarded as a team approach
way as the student nurses when involved on with everthing the midwife does under-
the labor floor. The function of the personal, written by the Master of the Hospital at the
professional attendant in labor is to provide time. Midwives' responsibilities are exten-
emotional support at a professional level, sive but within clearly defined limits.
thereby relieving the partner of any respon-
sibility for the safe conduct of the labor.
Audit of Results
Partners are welcome throughout, but are
not expected in any way to replace the pro- All labor outcomes are reviewed by a senior
fessional support provided by the midwifery obstetrician on a daily basis, and results are
staff. Because the labors are relatively short audited at a weekly labor ward audit meeting
and the only women in the labor ward are attended by the senior midwife on duty and
those who are in labor or having their labors by the senior resident staff. All operative
induced, it is possible to assign an individual deliveries and abnormal outcomes are
nurse to each woman in labor. reviewed with a view to training of the re-
sidency staff as well as reviewing procedures
to fine-tune and maximize beneficial out-
Senior Doctor Involvement comes. The audit meeting is also attended by
A senior obstetrician is responsible for the antenatal education midwives so that
the labor ward on a daily basis and is in their input is available from the classes.
attendance at regular intervals throughout
the day. The function of the senior obstetri-
cian is to provide support for the midwifery
staff as well as being available for consulta-
Medical Components
tion on a constant basis. Regular rounds are
Diagnosis of Labor
done in the labor ward by the consultant.
The senior resident staff function in a similar The diagnosis of labor depends, as all diag-
way and also perform regular rounds on the noses in medicine depend, on a history and
labor floor. Residents function at a much physical examination. Important points in
lower level and have no decision-making the history are regular painful uterine con-
authority whatsoever. The senior midwife tractions occurring at least once every 8 min.
on duty communicates directly with the most Strong supporting evidence in the history is
4. Active Management of Labor 45

a bloody show, the release of the mucus plug, always corrected in AML by infusion of oxy-
or spontaneous rupture of membranes. tocin. This is not done, however, if the mem-
Without regular painful contractions, the branes are still intact. Rupture of intact
dignosis of labor is not entertained. membranes of itself will often be sufficient
On physical examination, the most im- to ensure progressive cervical dilatation.
portant feature is effacement of the cervix. The color and volume of amniotic fluid are
The primigravid cervix is a different organ observed because the fluid may be an indi-
from the multigravid cervix, which has cator of prior "placental insufficiency."
dilated on a previous occasion. The pri- Absence of amniotic fluid, or thick meconium
migravid cervix tends to efface before it staining, suggests preexisting oligohydram-
dilates, and in AML the diagnosis of labor is nios and raises the specter of placental
accepted with full cervical effacement, so insufficiency. In these circumstances the
long as it is accompanied by regular painful fetus is regarded as being at increased risk
uterine contractions. In cases of doubt, the of developing hypoxia in labor.
midwife has a grace period of 1 h after which
the condition of the cervix is reassessed and
a firm decision is made as to whether or not
Oxytocin
the woman is in labor. The diagnosis of labor Oxytocin is infused at a standard rate in
is therefore accepted if the cervix is fully cases in which progressive cervical dilatation
effaced and there are painful uterine does not occur. Cervical dilatation is assessed
contractions. by vaginal examination 2 h after admission
If the diagnosis of labor is rejected, the to the labor ward and rupture of membranes.
woman is discharged from the labor ward If the cervix has not dilated by 2 cm in the
but retained in the hospital because of the intervening period, oxytocin is commenced.
possibility that labor may commence within Approximately 50% of primigravidae require
the next several hours. If the woman ex- oxytocin to accelerate labor in the first stage.
presses a firm preference for discharge home, The proportion requiring oxytocin is largely
her wishes are accepted. Before discharge determined by the cervical dilatation on
from the labor unit, a nonstress test (NST) is admission. Oxytocin is infused at a standard
performed. The woman is advised that labor rate commencing at 4 mu (milliunits) per
may commence within the next 24 h or so minute and increased every 15 min, by 4 mu,
and is asked to return should she believe to a maximum of 32 mu/min. Safety of oxy-
that labor is commencing. Prior attendance tocin infusion is discussed later.
at the antenatal classes facilitates acceptance In the second stage of labor, oxytocin may
of the rejection of the diagnosis of labor. be commenced for the first time if the head
does not descend within 1 h following diag-
nosis of full dilatation. Management of
Artificial Rupture of Membranes second-stage labor is discussed separately
Artificial rupture of membranes is performed here because of the influence of epidural
in that proportion of women in whom spon- anesthesia on progress in the second stage.
taneous rupture has not already occurred. In
current practice, aproximately 30% of women
have had spontaneous rupture of membranes
Progress in Labor
before admission to the labor ward. The pur- Progress in labor is assessed on a standard
pose of rupturing the membranes is twofold: labor graph/partogram (Figure 4.1). The
(1) to ensure efficient uterine action, and (2) essence of the partogram is simplicity, and
to observe the volume and color of amniotic a minimum amount of relevant information
fluid. is included. The graph represents progress
Inefficient uterine action expressed as judged by cervical dilatation against time.
failure of progressive cervical dilatation is There is no reference to descent of the head
46 P.Boylan

DELIVERED

FULL=lOem

gem ~
D 8em V
V
I
L
A
7em

V
T
A
T

L
I 6em
0
N
Scm

/
0
F
4em

V
C
E
R 3em

/
V
I
X 2em

lem /
/
o 1 2 3456789 10 11 12
HOURS AFTER ADMISSION

FIGURE 4.1. Standard labor graph, or partogram, used at the National Maternity Hospital, Dublin,
Ireland.

in the first stage of labor. A conscious deci- anesthesia because descent does not appear
sion was made during the development of to occur with the same degree of efficiency as
the partogram to keep it simple so that in the absence of epidural anaesthesia.
clarity would be enhanced. Progress in the
first stage is judged by vaginal examination
and the rate of cervical dilatation. The
Safety
minimum accepted norm is 1 cm/h. The first The central philosophy behind active man-
vaginal examination takes place on ad- agement is to achieve spontaneous vaginal
mission to the labor ward, the second 2 h delivery in as high a proportion of cases as
after admission, and generally speaking the is compatible with good outcome. There is
third at the diagnosis of full dilatation of no question that, from the mother's point
the cervix. The average number of vaginal of view, spontaneous vaginal delivery is
examinations is three. almost always the safer option. Cesarean
Progress in the second stage is judged by section is of course preferable to a traumatic
descent of the head. Failure of descent of forceps delivery; this goes without saying.
the head in the second stage is treated by Safety issues in AML generally focus on the
oxytocin infusion if this has not already use of what appears to be high-dose oxytocin
commenced. The incidence of oxytocin in- infusion. From the mother's point of view,
fusion increases in the presence of epidural the risk of oxytocin infusion is rupture of the
4. Active Management of Labor 47

uterus. In the world literature, however, nity Hospital, where the epidural policy was
there are few cases of rupture of the primi- liberalized very significantly at the begin-
gravid uterus. 8 In most of those cases re- ning of 1991. 12 The reason why this increase
ported, either the management was clearly has not occurred in the NMH is not entirely
out of line with standard practice or the clear, but it may be that efficient uterine
birth occurred in circumstances in which action is ensured early in labor by the appro-
there was minimal medical supervision. priate exhibition of oxytocin and artificial
Often the latter occurrences have been in rupture of intact membranes.
third world countries after extremely pro- The major problem with epidural anes-
longed, obstructed labors. These circum- thesia in the management of primigravid
stances have no relationship to standard labor arises in the second stage. Normally
practice in the current Western world. The second-stage labor is diagnosed when the
major risk associated with oxytocin infusion woman has an urge to push. At this stage
from a mother's point of view therefore is the head is on the pelvic floor, and pushing,
that of water intoxication. This only occurs, or the active phase of the second stage, may
however, when very large doses of oxy- rightfully commence. With epidural anaes-
tocin, far in excess of those recommended thesia, however, the second stage of labor
in AML, are infused in association with may be diagnosed before the head has de-
large volumes, in excess of 31, of salt-free scended to a level where pushing is appro-
fluid. The most likely occasion for this com- priate. This is a clinical judgment decision,
bination to occur is when induction of labor which under AML is made by the senior
is accompanied by epidural anesthesia over midwife in attendance. If the head has not
a prolonged period of time. Water intoxica- descended to a level where the midwife feels
tion should not occur in standard practice. it is appropriate to commence pushing, oxy-
The principal danger to the fetus asso- tocin is infused if it is not already in place.
ciated with oxytocin infusion is the precipita- One hour later, vaginal examination is
tion of fetal hypoxia through strengthening repeated and pushing is commenced. After
uterine contractions. There is no question 1 h of pushing, if delivery is not imminent
that oxytocin infusion strengthens uterine the clinical situation is reviewed by the
contractions, as this is the very purpose for senior doctor on duty.
which it is given. Clearly the treatment of
fetal hypoxia is delivery of the fetus and not
infusion of oxytocin, which will only aggra-
Use of AML in Other Countries
vate the situation in these circumstances. Active management of labor has become
There is no evidence thaL oxytocin infusion associated with low rates of cesarean section,
of itself increases the incidence of fetal and this has generated much interest in
hypoxia, provided it is given under ade- recent years. Attempts have been made,
quate fetal supervision. Evidence from the with varying degrees of success, to imple-
National Maternity Hospital9 and else- ment various parts of AML in an attempt to
where lO clearly indicates that oxytocin lower the incidence of cesarean section.
infusion does not increase the incidence of Turner et al. 13 and Robson et al. 14 success-
fetal death in labor, birth asphyxia, or fully implemented AML in different units in
hypoxic acid-base values at birth. British hospitals, resulting in significant
declines in operative delivery and lower rates
of cesarean section. Akoury et al. 15 imple-
AML and Epidural Anesthesia mented AML in Canada and had similarly
There is evidence that epidural anesthesia successful results. Boylan et al.,16 imple-
may increase the incidence of cesarean menting AML in Texas, met with mixed
section and forceps deliveryY This has not success as a result of several factors including
been the experience at the National Mater- failure to comply with protocol. Outside
48 P. Boylan

North America, AML has also been success- incidence of cesarean birth, however, is one
fully implemented in Chile.17 that can only be answered on a local level.
The reasons why AML has not achieved
the same results as in Dublin, at the NMH,
may be misunderstanding of the importance
References
of the organizational components of the 1. O'Driscoll K, Meagher D, Boylan P. Active
approach and failure to implement the management oflabor, 3rd Ed. London: Mosby
specifically medical components faithfully. Year Book Europe, 1993.
A major difficulty with implementing AML 2. O'Driscoll K, Jackson RJA, Gallagher JT.
has been fear of the effect of oxytocin infusion The prevention of prolonged labor. Br Med J
in the doses outlined. This fear is rational 1969;ii:477 -480.
3. O'Driscoll K, Foley M, MacDonald D. Active
and is compounded by the medical-legal
management of labor as an alternative to
climate, particularly in North America. As cesarean section for dystocia. Obstet Gynecol
pointed out, however, there is no evidence 1984;63:485-490.
that oxytocin infusion of itself results in a 4. Socol ML, Garcia PM, Peaceman AM, Dooley
poorer outcome, and indeed at the NMH it is SL. Reducing cesarean births at a primarily
associated with significantly better outcome private university hospital. Am J Obstet
than in other units. Gynecol 1993;168:1748-1758.
5. Taffel DM, Placek PJ, Moien M. 1988
Cesarean section rate at 24.7 per 100 births-
Future Developments a plateau? N Engl J Med 1990;323:199-200.
6. Notzen FC, Cnattingius S, Bergsjo P, et al.
Cesarean section delivery in the 1980's:
Childbirth is unique in medicine insofar as international comparison by indication. Am
it is so strongly influenced by the cultural J Obstet Gynecol 1994;170:495-504.
environment in which it occurs. Women's 7. O'Driscoll K, Stronge JM, Minogue M. Active
expectations of childbirth vary enormously management of labour. Br Med J 1973;iii:
from country to country, and in such a mul- 135-137.
ticultural environment as North America 8. Boylan P. Labor in the primigravid patient.
there is of course a wide variety of expecta- Curr Probl Obstet Gynecol Fertil, 1991:14(1).
tions. There is no "correct" incidence of 9. Cahill DJ, O'Herlihy C, Boylan P. Does oxy-
cesarean section that is applicable to all tocin augmentation increase perinatal risk
societies. There are many subtle influences, in primigravid labor? Am J Obstet Gynecol
1992;166:847 -850.
which tend to increase the incidence or indeed
10. Thorp J, Boylan P, Parisi VM. The influence
keep it low. These factors are more related to of high dose oxytocin augmentation on
differences in cultural background, not only umbilical acid base balance. Am J Obstet
of the laboring women but especially of the Gynecol 1988;159:670-675.
physicians attending those women. It is clear 11. Thorp J, Hu DH, Albin RM, et al: The effect
from attempts to implement AML in dif- of intrapartum epidural analgesia on nulli-
ferent racial groups that almost identical parous labor: a randomized, controlled,
results can be achieved in a variety of dif- prospective trial. Am J Obstet Gynecol 1993;
ferent settings when the various components 169:851-858.
of AML are implemented most closely to 12. Robson MS, Boylan P, McParland P,
those at the NMH. Mac Quillan K, O'Neill M. Epidural analgesia
need not influence the spontaneous vaginal
It is also clear from experience both in
delivery rate. In: Proceedings of the 13th
Dublin and elsewhere that AML may have a Annual Meeting of the Society of Perinatal
significant role to play in reducing the Obstetricians, San Francisco, 1993 (abstract
incidence of cesarean section where this is & poster).
believed to be inappropriately high. The 13. Turner MJ, Brassil M, Gordon H. Active
question of what is an inappropriately high management of labor associated with a
4. Active Management of Labor 49

decrease in the caesarean section rate in 16. Boylan P, Frankowski R, Rountree R, Selywn
nulliparae. Obstet GynecoI1988;71:150-154. B, Parrish K. Effect of active management of
14. Robson MS, Scudamore I, Walsh S. Can labor on the incidence of caesarean section
labour ward audit improve obstetric outcome? for dystocia in nulliparas. Am J Perinatol
Oral presentation at XIII World Congress of 1991;8:373-379.
Gynaecology and Obstetrics, 15-20 Septem- 17. Masoli P, Pico V, Pellerano I. Manejo activo
ber, 1991, Singapore. del parto. Experiencia en el Hospital Gustavo
15. Akoury HA, Brodie G, Caddick R. Active Fricke. Rev Chil Obstet Ginecol 1986;51:
management of labor and operative delivery 223-230.
in nulliparous women. Am J Obstet Gynecol
1988;158:255-258.
5
Vaginal Birth After Cesarean Section
BRUCE L. FLAMM

Vaginal Birth After Cesarean: ally takes less than an hour and can be
scheduled for a convenient time of day. A
The Doctor's Dilemma trial of labor will take many hours, and the
birth may occur at any hour of the day or
Dr. Roy Pitkin, editor of Obstetrics and night. A second issue is the fear oflitigation.
Gynecology, recently stated that "without Some doctors point out that, although the
question, the most remarkable change in risk is low, if a uterine rupture or any sig-
obstetric practice over the last decade in- nificant problem does occur during a trial of
volves management of the woman with a labor an attorney will always be able to
prior cesarean delivery."llt would also prob- argue that the doctor should have "just"
ably be safe to say that few changes in done an elective cesarean. The third common
the past decade have made obstetricians so complaint that doctors voice about VBAC is
angry. Why is this true? that women often are not interested. Cer-
Some consumer advocates have argued tainly many women who have had difficult
that it is a monetary issue and that doctors first labors will not be excited about going
have resisted vaginal birth after cesarean through the whole process again. These
(VBAC) because they are paid more for three points are difficult to refute. Repeat
cesareans than for vaginal births. I do not cesareans are convenient, fairly low risk,
believe this. Although the average cost for and generally well accepted by patients. So
having a cesarean birth in the United States why rock the boat? Twenty-five years ago
in 1991 was $3106 higher than for vaginal that would have been a valid question. Only
birth, only 20% of the difference ($611) was 5% of all births were by cesarean, and only a
for physician fees. 2 In some areas, insurance fraction ofthose were repeat operations. But
companies are now paying doctors the same today some hospitals in the United States
fee regardless of the mode of birth. I have are reporting cesarean rates greater than
talked to some of these doctors and many of 50%.3 Because increasing numbers of
them are still not excited about VBAC even women presenting for prenatal care have
though they are no longer paid more for already had at least one cesarean, any
cesareans. attempt at reducing the overall cesarean
These doctors point out several reasons rate will be doomed to failure unless we can
why they prefer routine repeat cesareans. break the cycle of one cesarean leading to
The first is convenience. A cesarean gener- subsequent "routine" repeat operations.

51
52 B.L.Flamm

The Impact of Repeat of labors after previous cesarean will result

Cesareans on the Overall in vaginal deliveries. 6 - 8
The potential impact of VBAC on overall
Cesarean Rate cesarean rates can be illustrated with the
following example. Suppose doctors at a
In the United States, the single most com- hospital doing 4000 deliveries per year
mon indication for performing a cesarean decided to attempt to lower their 30%
operation is "repeat cesarean.,,4,5 It follows cesarean rate (currently 1200 cesareans per
that if a physician or hospital is interested year). Assume they are currently delivering
in reducing their cesarean rate, evaluating almost all breeches by cesarean and almost
the repeat cesarean rate is a logical place all patients with previous cesarean by re-
to begin. Repeat cesarean has become so peat cesarean. The doctors might be able to
common that an individual physician could utilize the technique of external cephalic
make a concerted effort to avoid cesarean version to avoid 36 cesarean operations
section for breech or twin gestations and yet (assuming a 3% rate of breech, 75% early
end up doing little to change his or her total detection rate, 50% version success rate,
cesarean rate. This is because only about 3% and that 90% of successful versions go on
of infants are breech at term and some of to deliver vaginally). If the same doctors
these cases are not detected until the onset decided to encourage VBAC, they might be
of labor. Even if detected early, the success able to avoid 250 cesareans (this assumes
rate for external cephalic version is only 11 % prior cesarean rate, 75% trial of labor
about 50%. Furthermore, many cases of rate, and 75% VBAC rate among those who
failed external version are not suitable can- attempt trial of labor). The breech version
didates for attempted vaginal delivery, and program would result in a reduction of the
even a successful version does not guarantee overall cesarean rate from 30% to 29.1%.
a successful vaginal birth. Similarly, an The VBAC program would result in a reduc-
effort to decrease the overall cesarean rate tion of the overall cesarean rate from 30% to
by attempting to avoid cesareans for twins 23.8%.
and higher order multiple births will not be This is a hypothetical example, but we
effective. Only about 2% of pregnancies have confirmed similar results with actual
involve multiple gestations and, depending data at the hospital where I practice. At
on the fetal presentations, many of these the Kaiser Riverside hospital (Riverside,
cases are not suitable candidates for at- California), the cesarean rate for 1992 was
tempted vaginal delivery. 12.7% (327 cesareans/2565 births). Some
In contrast, about 11 % of patients now physicians find this number difficult to
presenting for prenatal care have a history believe but it can be explained almost ex-
of previous cesarean delivery. Data from clusively on the basis of an active VBAC
the National Hospital Discharge Survey program. There were 193 vaginal ~irths
revealed that patients with prior cesareans after cesarean at our hospital in 1992. If we
accounted for approximately 441,000 of did not allow trial oflabor, our 1992 overall
the 4,158,000 births that occurred in 1990. cesarean rate would have been 20.3% (520
This represents almost 11% of all prenatal cesareans/2565 births), a figure very similar
patients. Data from the same source also to the national average. As shown by the
indicate that repeat operations accounted data in Table 5.1, this was not an unusual
for 36% of all cesareans in 1990 (626,000 year. For each of these years the total
primary, 351,000 repeat, 977,000 total). cesarean rate would have been at least 60%
Approximately 99% of these women have higher if there had been no trial of labor
had prior low-transverse incisions and most (TOL) program (Table 5.2).
are suitable candidates for trial of labor. This discussion is not meant to discourage
Several studies have shown that about 75% external cephalic version (or selective
5. Vaginal Birth After Cesarean Section 53

TABLE5.1. Cesarean statistics (actual), Kaiser plete or symptomatic). I will attempt to

Permanente Medical Center, Riverside, CA briefly summarize the vast amount of data
1990 1991 1992 on this subject in a few paragraphs.
In a review of 25 papers on trial of labor
Total births 2408 2548 2565
Primary cesareans 193 237 212 published between 1950 and 1980, Lavin et
Repeat cesareans 108 107 115 a1. 9 found 5,325 VBACs and no maternal
Total cesareans 301 344 317 deaths from uterine rupture. The review
Total cesarean rate 12.5% 13.5% 12.7% listed 14 fetal deaths but 12 of these were
TOLs 247 280 230 caused by rupture of prior classical incisions.
Successful VBACs 196 236 193 In 1985 I published a review of all 21 trial-
Successful TOLs 79% 84% 84% of-labor papers published between 1980 and
TOL, trial oflabor; VBAC, vaginal birth after cesarean. 1984.10 The papers included more than 6000
trials of labor and more than 5000 VBACs
without a single' maternal death from
TABLE 5.2. Cesarean statistics (hypothetical), uterine rupture. In 1987, a report from
Kaiser Permanente Medical Center, Riverside, Dublin revealed a 0.45% uterine rupture
CA rate in 1781 trials of laborY There was
1990 1991 1992 one intrapartum fetal death in this series.
Total births 2408 2548 2565
Interestingly, 45% of these patients did not
Primary cesareans 193 237 212 have electronic fetal monitoring during their
Repeat cesareansa 304 343 308 trial oflabor. In 1989, a group from Sweden
Total cesareansa 497 580 520 reported 1006 trials of labor with a 0.6%
Total cesarean ratea 20.6% 22.8% 20.3% rupture rate. 12 There were no serious com-
Increase in cesearan ratea 65% 69% 60%
plications in either the mothers or the
a If no trials oflabor had been allowed. infants. A recent review of more than 10,961
trials of labor at the Los Angeles County
medical center found a 0.7% rate of uterine
vaginal breech delivery). The analogy is rupture. 13
presented only to emphasize the relative im- A meta-analysis published in 1991 that
portance of vaginal birth after cesarean in included many ofthe papers just listed found
any program aimed at safely reducing the a combined rupture-dehiscence rate of
overall cesarean rate. 1.8% but did not differentiate between true
uterine rupture and asymptomatic dehis-
cence.14 At Kaiser Permanente hospitals in
Uterine Rupture During Trial California, we found a 0.2% rate of uterine
rupture (10 of 5733) during the years 1984-
of Labor After Previous 1988. 7 The incidence increased to 0.8% (39
Cesarean Section of 5022) during 1990-1992. We recently re-
viewed our data on more than 11,000 trials
Trial of labor after previous cesarean has of labor and found the overall incidence of
been a controversial subject for decades. No uterine rupture during the past decade to be
aspect of trial oflabor has created more con- 0.5%.15 With tens of thousands of patients
troversy than the subject of uterine rupture. now reported, it is seems safe to say that
More than 50 publications have addressed the incidence of true uterine rupture during
this issue. Many of the early publications on planned trial oflabor is less than 1%.
trial of labor after previous cesarean section Although most reports have concluded
did not differentiate between thin areas of that maternal and perinatal outcomes are
the lower uterine segment found at the time often good even when uterine rupture does
of repeat cesarean (dehiscence or uterine occur, two recent reports from Colorado and
window) and true uterine ruptures (com- Utah emphasized that these ruptures can
54 B.L.Flamm

have very serious consequences. 16 ,17 These distress. 2o The monitoring record was kept
two reports include a total of 20 uterine on as the uterus was incised (simulated com-
ruptures from an unknown number of plete uterine rupture) and even as the baby
trials oflabor. One of the papers estimates a was delivered through the incision (simu-
uterine rupture incidence of 0.7%, which is lated complete fetal expulsion). Neither the
consistent with published data. There were uterine tone nor the peak uterine pressures
no maternal deaths in either of these re- (contraction amplitude) were changed by
ports, but there were four perinatal deaths the simulated uterine ruptures.
and four cases of neurologic impairment. All these studies seem to indicate that the
Fortunately, in 70% of their catastrophic IUPC is of little or no value in the detection
uterine rupture cases (14/20), the infants of uterine rupture. Of course the IUPC has
had good outcomes. Interestingly, 3 of the 4 been shown to be of value during oxytocin
perinatal deaths in these reports apparently augmentation of labor. But for women
involved women laboring at home with- undergoing trial of labor without oxytocin,
out fetal monitoring. These reports serve there seems to be no reason to insert an
to point out that trial of labor is not risk IUPC. Although it would have been useful if
free and that careful fetal monitoring is the IUPC had been a harbinger of uterine
mandatory. rupture, there may be a good side to this
story as it means that there is no need to
perform artificial rupture of membranes at
Intrauterine Pressure the earliest possible moment just to insert a
catheter. In fact, if the fetal heart is being
Catheters: Can They Diagnose monitored externally without difficulty,
Uterine Rupture? there may be no need to intervene at all.

Because an intrauterine pressure catheter

(IUPC) rests inside the uterine cavity, one Oxytocin Use During
might expect the amplitude of contractions
detected by internal monitoring to decrease Trial of Labor After
with a uterine dehiscence and perhaps fall to Previous Cesarean Section
zero with a complete uterine rupture. This
would be a wonderful diagnostic aid, but I recently attended a malpractice trial to
unfortunately things are not so simple. testify in defense of an obstetrician accused
In 7 of the 8 complete uterine ruptures of negligence in a uterine rupture case. As I
in the Colorado series previously described, sat under oath on the witness stand, the
IUPCs were used but apparently did not aid plaintiff's attorney read a sentence aloud
in the diagnosis. 16 Likewise, Rodriguez et from the defendant doctor's pretrial deposi-
al. 18 reviewed a series of 39 uterine ruptures tion. When asked if he had ever heard of
(some patients had prior cesareans) in which Williams Obstetrics, the doctor responded,
IUPCs had been used and found no loss of "Oh yes, it's the bible of obstetrics."
uterine tone or cessation of labor. They con- The attorney then placed a copy of the 18th
cluded that the usefulness of the IUPC in edition of Williams Obstetrics in my hand
making the diagnosis of uterine rupture was and instructed me to "read aloud from the
not supported. 18 A group in England reached bible of obstetrics." The highlighted sen-
the same conclusion. 19 In an interesting ex- tence on p. 345 said, "Oxytocin is a powerful
periment (one which I wish I had thought drug, and it has killed or maimed mothers
of), a group in Georgia recently created through rupture ofthe uterus and even more
"artificial catastrophic uterine ruptures" by babies through hypoxia from markedly
simply leaving the IUPC in place while per- hypertonic uterine contractions." (If you
forming cesarean sections for CPD or fetal doubt that such a statement is in Williams,
5. Vaginal Birth Mter Cesarean Section 55

you may wish to open a copy of the 1989 21 % of all labors were either induced or
edition and tum to p. 345. 21 ) As the jury augmented in 1990 alone (863,000 cases).22
mumbled in astonishment I tried to explain Clearly, there are many times when the
that the sentence was taken out of context small potential risks of oxytocin are offset by
and was apparently meant to apply to the the potential benefits. Indications for oxy-
outmoded oral (buccal) administration of tocin use such as postdates pregnancy, pre-
the medication. However, the plaintiff's mature rupture of the membranes, and slow
attorney shouted that I was not asked to progress in labor occur about as frequently
give a medical lecture and cautioned me to in women planning trial of labor after pre-
just answer his questions. The judge sus- vious cesarean as in the general obstetric
tained his admonishment and warned me to population. If oxytocin could not be used
be silent. The plaintiff's attorney then in- during trial of labor, many (perhaps most)
structed me to tum to p. 321 of the "bible of planned trials of labor would end up as re-
obstetrics" and read aloud the sentences peat cesarean operations. Oxytocin is cer-
highlighted in yellow. "Before delivery, the tainly not a panacea, and many readers are
spontaneously laboring uterus is very likely well aware of the fact that there are alter-
to be exquisitely sensitive to oxytocin. Even natives such as expectant management after
with an intravenous dose of a few milliunits spontaneous rupture of the membranes and
per minute, the pregnant uterus may con- ambulation during desultory labor. Never-
tract so violently as to kill the fetus, rupture theless, there are many cases in which the
itself, or both.,,21 The plaintiff's attorney decision boils down to whether to use oxy-
then took the book from my hands and tocin or to proceed with repeat cesarean
slammed it down on the table as if to say, operation. In this situation the key issue
"this case is closed." I started to explain that becomes "is oxytocin safe to use during labor
oxytocin is one of the most commonly used after previous cesarean section?"
drugs in obstetrics and probably one of Oxytocin can cause uterine hyper-
the safest but I was again silenced by the stimulation, often accompanied by fetal
plaintiff's attorney and cautioned by the bradycardia, regardless of whether a patient
judge. has had prior uterine surgery. In other
This scenario points out a couple of things. words, oxytocin use carries some risk regard-
First, it is probably not wise for a defendant less of the presence or absence of a previous
to identify any book as the "bible of ob- uterine scar. The real issue is whether oxy-
stetrics"! Second, the next edition of tocin use substantially increases the risk of
Williams Obstetrics might consider editing rupture of a previous uterine scar. I stress
out these misleading and potentially damag- the word "substantially" because, as pointed
ing statements. The most recent edition of out earlier, the use of oxytocin is almost
Williams Obstetrics, published in 1993, still certainly associated with some increased
presents oxytocin as a dangerous drug that risk of uterine rupture during trial oflabor.
has killed or maimed both mothers and In multiparous patients, oxytocin utilization
babies. Third, it serves to remind ob- has been associated with an increased risk of
stetricians that even though oxytocin is a rupture of the unscarred uterus. It would be
safe and efficacious medication, its use (even surprising if the same relationship did not
its proper use) does appear to result in a exist for patients with prior uterine scars.
small but significant increase in the risk of Before 1980, almost no data existed on
uterine rupture. oxytocin utilization during trial of labor
In reality, oxytocin has been used to induce after previous cesarean section. In 1985 I
or augment labor in literally millions of reviewed 21 papers on trial oflabor that had
women, and most of these cases have re- been published during 1980-1984 and found
sulted in neither uterine rupture nor fetal more than 600 cases in which oxytocin had
damage. National statistics indicate that been used. 10 Two cases involved uterine rup-
56 B.L.Flamm

ture, yielding a rupture rate of less than (57%) of these women were treated with
0.5%, not significantly different than that oxytocin. Between 1982 and 1984 in Kaiser
expected without the use of oxytocin. In 1987 Permanente medical centers in California,
we reported on 485 patients treated with we collected data on 245 patients under-
oxytocin during trial of labor after previous going trial of labor with two or more prior
cesarean, and noted a small but not statisti- cesareans. Sixty-eight percent of these
cally significant increase in the incidence of women delivered vaginally, and the in-
uterine rupture. 23 Neilsen et alP found no cidence of uterine rupture did not differ sig-
increased risk of uterine rupture in 406 nificantly from that found in women with
trial-of-Iabor patients treated with oxytocin. one prior cesarean section. 7 Since that paper
Molloy et alP found no increased risk of was published our group has collected data
uterine rupture in 271 trial-of-Iabor patients on more than 500 additional cases involving
treated with oxytocin but noted an ap- two prior cesareans, and preliminary ana-
parently increased risk in 30 patients who lysis indicates no increased risk of uterine
also had epidural analgesia. A meta- rupture.
analysis of the papers reporting on oxytocin With more than 1000 cases reported in the
use during trial of labor noted that the literature it seems quite reasonable to allow
numbers were still small but that use trial of labor in the patient with two pre-
of oxytocin appeared to be unrelated to vious cesarean operations. However, there
perinatal mortality rates. 14 are very little data available on the outcome
of trial of labor with three or more previous
cesareans. The 1988 revision of the ACOG
Trial of Labor with More Than Guidelines for Vaginal Delivery After a Pre-
vious Cesarean Birth recommends that, "A
One Previous Cesarean woman with two or more previous cesarean
Operation deliveries with low transverse incisions who
wishes to attempt vaginal birth should not
A review of trial-of-Iabor papers published be discouraged from doing so in the absence
from 1950 to 1980 only identified about 100 of contraindiations." Because data to sup-
cases involving more than one previous port the "or more" portion of that statement
cesarean operation. 9 I reviewed the litera- are minimal, it would probably be prudent
ture from 1980 to 1984 and was only able to to inform women with more than two pre-
find another 100 cases. lO More recently, Far- vious cesarean deliveries that they may be
makides and associates24 reported on 57 at greater risk for uterine rupture if they opt
cases (39 two-prior, 18 three-prior) and noted for trial oflabor.
no uterine ruptures. Pruett et al. 25 reported
on 55 cases (51 two-prior, 3 three-prior)
and noted no complete uterine ruptures. Suspected Fetal Macrosomia
Interestingly, 42 of these patients had
previous uterine incisions of unknown type One problem that often develops before a
and two had known low vertical incisions. planned trial of labor is "suspected fetal
Novas and Gleicher (Novas et al. 26 ) reported macrosomia." This is a great concern to both
1 uterine rupture of 36 trials of labor (27 the patient and her physician. No woman
two-prior, 9 three-prior) but later learned wants to go through many hours of labor and
that this patient had two prior classical end up with a repeat cesarean because of
cesareans. Phelan et al. 27 reported data on failure to progress. This is a particular con-
trial of labor in 501 patients with two prior cern for women whose initial cesarean had
cesareans. Sixty-nine percent of the patients been performed following failure to pro-
in this large series delivered vaginally and gress. Doctors share this concern. Perhaps
there were no true uterine ruptures; 248 this explains why (with the exception of dia-
5. Vaginal Birth After Cesarean Section 57

betic pregnancies) suspected fetal macro- answer this question we investigated the
somia is rarely listed as an indication for outcomes of hundreds of trials oflabor where
primary cesarean yet is frequently given as the birthweight was more than 4000 g.33
an indication for elective repeat cesarean. There was no diagnostic dilemma here
The issues of shoulder dystocia and perineal because we considered actual birth weights,
trauma must also be addressed in any dis- not at estimates made before birth. Of 1776
cussion of macrosomia. patients who underwent TOL during 1984-
The original edition of the ACOG Guide- 1985 at eight California hospitals, 301
lines for Vaginal Delivery After a Previous delivered infants weighing 4000 g or more (8
Cesarean Birth and both revisions of these pounds, 13 ounces). In the birthweight
guidelines have included precautionary sta- range of 4000-4500 grams, 58% delivered
tements regarding estimated fetal weight. vaginally. In the birthweight range more
The 1984 revision stated that "the esti- than 4500 grams, (4500 g = 9 pounds, 15
mated fetal weight should be less than 4000 ounces) 43% delivered vaginally. When
grams." The 1988 revision stated that compared to the 1475 TOLs with actual
"the effects of labor on the patient with an birthweights less than 4000 grams, there
estimated fetal weight of more than 4000 were no significant differences in perinatal
grams have not been substantiated." or maternal morbidity or mortality. Other
There are two important questions relat- investigators have reported vaginal bir-
ing to suspected fetal macrosomia and trial thrates greater than 65% in TOLs with
of labor; the first concerns the difficulty in birthweights more than 4000 g.l1,34
predicting fetal weight and the second the The patient with an apparently large fetus
risk of trial oflabor, assuming that the fetal is a problem that obstetricians face almost
weight could be predicted with certainty. daily. It is tempting to recommend cesareans
Can fetal macrosomia be accurately diag- in all such cases and avoid the potential
nosed before birth? A study conducted in the risks. This is especially true if the patient
1970s found that only 35% of macrosomic has had a prior cesarean. Unfortunately,
infants could be correctly diagnosed within this policy would result in many operations
1 pound of actual birth weight using ab- with little or no improvement in outcomes.
dominal palpation. 28 This was the famous Abdominal delivery of all infants with sus-
study which revealed that estimates made pected fetal macrosomia would necessitate
by experienced obstetricians were no better huge numbers of cesarean operations, many
than those made by nurses and medical of which would yield normal-sized infants.
students. Although dozens of recent studies Even if the fetal weight could be predicted
have reported on sonographic estimation of with no doubt, the foregoing data show that
fetal weight, many have cautioned that the most macrosomic infants will deliver vagin-
predictive accuracy declines as fetal weight ally. It is also well known that many cases of
increases toward 4000 g.29-31 In some cases shoulder dystocia occur with normal-sized
these errors approached 450 g (1 pound), and infants.
large numbers of infants were incorrectly The reassuring data presented here will
labeled as macrosomic. A review of the not completely solve the dilemma presented
literature concluded that "Clearly, the cur- by the patient who has the combined prob-
rent ultrasonic weight estimation pro- lems of prior cesarean delivery and current
cedures are not accurate enough to be used suspected fetal macrosomia. Because the
effectively in the detection of macrosomia rate of vaginal birth is similar to that of
defined by weight criteria.,,32 patients with nonmacrosomic infants, it
If technology advances to the point that seems that the real issue pertains to the risk
fetal macrosomia can be accurately diag- of shoulder dystocia (and perineal trauma)
nosed, should trial of labor (TOL) be avoided rather than the risk of failed trial of labor.
with fetal weight over 4000 grams? To Although our study was reassuring in this
58 B.L.Flamm

regard, the numbers may not have been of information on this subject comes from
large enough to detect small differences in studies at the University of Southern
outcome. This is particularly true in the California/Los Angeles County Medical
birthweight range of 4500 g and more, where Center; of 751 patients having a trial of
we had only 61 cases. labor, 592 (79%) had "unknown" scar types
One thing does seem clear with regard and 82% of these patients were delivered
to suspected fetal macrosomia: It must be vaginally.8 A more recent paper from the
understood that shoulder dystocia and its same institution reports several thousand
potential sequelae (both maternal and trials of labor and indicates that the pre-
perinatal) are often not preventable and will vious uterine scar type was unknown in
continue to occur in spite of the very best 80%-90% of cases. 13
obstetrical care. Tort reform, which clearly Several studies have reported preliminary
acknowledges this simple fact, would go a results with the use of ultrasound in an
long way toward eliminating many frivolous attempt to detect potential scar defects
lawsuits and many unnecessary cesarean before labor or to determine whether the
operations. previous scar is vertical or low transverse.
The results have been been disappointing.
One study of 56 patients preceding repeat
Unknown Scar Type cesareans found what appeared to be some
type of "defect" of the scar in 21% of cases. 36
The inability to determine what type of in- The authors noted that their incidence of
cision was made in the uterus at the time of defects was 4- to 10 fold that of previous
a previous cesarean should be differentiated reports. Because the risk of uterine rupture
from the failure to make any attempt at with trial of labor is of the order of 1% or
obtaining prior records. The importance of less, they correctly concluded that "It is pos-
documenting that an attempt was made to sible that ultrasound technology overdiag-
obtain this information is discussed next noses anatomic variations that would not
(see VBAC: Avoiding the Pitfalls, later in cause abnormal outcomes."
this chapter). However, in many cases it is When prior operative records cannot be
simply not possible to get this information. obtained, it is reassuring to recall that
The records of the previous operation may approximately 99% of cesareans performed
have been destroyed, or the operation may in the past decade involve a low transverse
have been performed in another part of uterine incision. Thus only about 1% of "un-
the world. In other cases the patient may known" incisions will actually be vertical or
present for prenatal care late in the preg- classical. We can probably reduce the risk
nancy or even in labor, allowing little or no even further by thinking about why these
time to confirm the previous scar type. This types of incisions are made. If records are
was the case in the studies described next. unavailable and the patient explains that
A study from Baylor College of Medicine her first cesarean was performed because
compared outcomes of 300 patients with un- her baby was transverse or because the
known uterine scar type to 88 patients with placenta was blocking the birth canal, a
documented low transverse scars.35 They classical incision may have been performed.
found that the incidence of uterine rupture Likewise, if the patient explains that her
did not differ statistically between the baby was breech and weighed 4 pounds,
two groups and concluded that "allowing a there is reason for concern. In most cases,
trial of labor in a patient with an unknown however, the patient will explain that her
uterine scar does not place the mother or first cesarean was performed because of slow
fetus at greater risk than the risk for a progress in labor, suspicious fetal monitor-
patient with a known low cervical trans- ing, term breech, genital herpes, or other
verse incision." An even larger amount common conditions for which classical or
5. Vaginal Birth After Cesarean Section 59

vertical uterine incisions would rarely be 1963 a group in Ireland reported 22 breech
used. deliveries after previous cesarean and noted
no complications. 39 Two more recent studies
also addressed this issue. A group in Israel
Vaginal Delivery of Twins allowed trial of labor in 47 patients with
After Previous Cesarean breech presentation, and 37 delivered vagin-
ally.40 A group in the United States recently
Section reported 27 cases with 13 vaginal breech
deliveries. 41
Although most obstetricians recommend Yet another option exists for this group
elective repeat cesarean for women who pre- of patients. We recently reported on 56
sent with the combination of prior cesarean attempts at external cephalic version in
delivery and twin gestation in the current patient with prior low transverse cesareans;
pregnancy, trial of labor appears to be a 46 of these attempts were successful and
reasonable alternative. A review of the there was no morbidity.42 When I presented
literature found 30 cases of trial of labor the initial results of this study at the ACOG
with twins, all of which resulted in vaginal combined district VIII/IX meeting in Seattle
deliveries. 37 The review mentioned 1 mater-
in 1990, I asked for a show of hands to see if
nal death from elective repeat cesarean sec-
anyone in the audience had any expe~ience
tion for twins. The same group also reported
with external cephalic version in patients
15 of their own cases, all of whom delivered
with prior cesareans. Much to my surprise,
vaginally. Another study reported 25 trials
about half the attendees were already doing
oflabor, and in 18 ofthese cases both infants
the procedure in these patients. Since that
were delivered vaginally.38 There were no
paper was published we have attempted
uterine ruptures. They pointed out that con-
version in at least 50 more patients without
cerns about rupture caused by uterine over-
complications. Although the numbers are
distension with twins did not seem to be
still relatively low, it appears that the risk
substantiated in the literature. With fewer
of uterine rupture during external cephalic
than 100 cases reported in the literature, version is low.
this question has not been answered
In summary, few experts would argue
definitively but for the woman who strongly
with a doctor who recommends elective
desires vaginal birth, trial of labor in a twin
repeat cesarean for a patient with breech
gestation seems to be a reasonable option.
presentation and a history of a previous
cesarean. However, for patients who
strongly wish to avoid repetitive opera-
Vaginal Breech Delivery After tions, both the options discussed here
Previous Cesarean Section appear to be reasonable alternatives.

Many physicians in the United States no

longer perform any vaginal breech de- Trial of Labor After a
liveries. However, there is ample evidence Single-Layer Closure of a Low
in the literature to support selected vaginal
breech delivery. This is discussed in detail
Transverse Uterine Incision
in Chapter 6 of this book. Even for those In the past several years the technique of
physicians who are comfortable with vaginal single-layer closure of the uterine incision
breech delivery, the patient who presents has been rapidly gaining in popularity. For
with a prior breech presentation and a his- many physicians it has become routine. As
tory of a previous cesarean presents a early as 1986 an obstetric textbook was
dilemma. There have been a few reports on recommending the single-layer closure. 43
such patients in the medical literature. In The authors pointed out that
60 B.L. Flamm

In view of the increasing acceptance of a trial are available. Prostaglandin was used for
of labor following cesarean section, the uterine cervical ripening in 143 trials oflabor in an
incision should be closed in a fashion calculated English study. Of these women, 108 went
to result in optimal healing. The popular method on to deliver vaginally, and there were no
of repairing a transverse incision with a con- complications. 45 In another study, 16 of 19
tinuous locking stitch reinforced by a continuous
patients treated with 1.5 mg vaginal PGE2
inverting layer produces good hemostasis, but
can also leave a poorly vascularized mass of pessaries went on to deliver vaginally with-
tissue that interferes with healing. A second out complications. 46 The most recent study
inverted layer created by using a continuous reported 25 patients with unfavorable
Lembert's or Cushing's stitch is customary but is cervix who were treated with 1 mg PGE 2 gel
really needed only when opposition is unsatis- intracervically at the University of Texas. 47
factory after application of the first layer. 18 patients went on to deliver vaginally and
there no major complications. However,
This is an interesting and reasonable uterine scar rupture has been reported with
point of view. Two layers do not necessarily the use of PGE 2 and hence it must be used
heal better or stronger than one. In fact, the with caution. We are now evaluating the
opposite may be true. I attempted to study use of PGE 2 in patients with prior cesarean
this at our hospitals but found that most delivery and will soon be reporting the re-
cesarean operative reports contained the sults of more than 300 cases.
statement, "the uterine incision was closed
in the usual fashion."
Other investigators have had more luck.
Pruett et al. at Baylor College of Medicine
Elective Repeat Cesarean
studied 57 trials of labor after documented Versus Trial of Labor
single-layer uterine closure and found no
complete uterine ruptures. 35 They did note Most studies on trial of labor have focused
a 2.8% incidence of dehiscence (incom- on the results of women who actually
plete, asymptomatic separation) in this attempted vaginal birth, not on a com-
group and cautioned that more patients parison to women having elective repeat
would have to be studied before a definitive cesareanS. The best way to compare the
statement could be made. More recently, two alternatives would of course be a
researchers at the University of Alabama randomized controlled trial. However, such
compared 149 women with single-layer a trial would be a logistic nightmare. Once
closures to 143 women with the traditional randomized, women would have to open a
two-layer closure and found no complete sealed envelope to determine whether or
uterine ruptures in any of the 292 trials of not they would proceed to the operating
labor. 44 An incidence of partial scar separa- room. Many women would be required to
tion of approximately 2% was found in proceed with an operation that they really
each group. did not want and that dozens of studies
have shown to be unnecessary. Other
women would be required to have a trial of
Prostaglandin E2 Gel labor that they did not desire and some of
these women would eventually demand a
Induction of Patients with a cesarean, thus creating postrandomization
Prior Cesarean Section selection bias. Furthermore, because of the
uncommon nature of major complications
The recent obstetric literature has been in either group, such a randomized trial
filled with reports on the use of prostag- would have to be very large. Also, because
landin E2 (PGE 2) for cervical ripening. Al- of the presence of the telltale abdominal
though most of these studies have excluded scar, it would be difficult if not impossible
patients with prior cesareans, some data to blind attending obstetricians to the fact
5. Vaginal Birth After Cesarean Section 61

that they were dealing with a patient with a without the need for repeat cesarean opera-
scarred uterus. tions in the majority of cases. This had been
Having given up on the idea of conducting predicted by using hypothetical data and a
a randomized controlled trial, we decided decision analysis process a decade before the
to do the next best thing. We had already study began. 48
published the results of a multicenter study
involving more than 5000 trials of labor. 6 ,7
Patients who opted for elective cesarean VBAC: Avoiding the Pitfalls
were not evaluated in that study. In 1990 we
began a prospective study of all pregnant During the past 10 years, I have reviewed
women at all 10 Southern California Kaiser the medical records of several cases in which
Hospitals. This study would include all trial of labor resulted in a bad outcome.
patients who opted for elective repeat Some of these cases involved superior phy-
cesarean and would thus allow for a com- sician and nursing care and therefore reflect
parison to those who had trial of labor. the fact that childbirth, regardless of mode
Because this study was prospective but not of delivery, is not risk free. However, other
randomized there was a possibility that we cases offered information from which we can
would be comparing apples to oranges. For all learn. 49 Some of these cases resulted in
example, it would be likely that the elective lawsuits. Analysis ofthe records has allowed
repeat cesarean group would include signifi- me to come up with a few clinical "pearls of
cantly more women with multiple prior wisdom" that may make trial of labor a safer
cesareans and other risk factors. On the experience.
other hand, certain predetermined bad out-
comes would be preferentially shifted to the Attempt to Confirm the Type of
trial-of-Iabor group. For example, patients Uterine Incision Made at the Time of
with unexplained intrauterine fetal deaths
the Initial Cesarean
or preterm labor involving an extremely
premature infant are generally allowed Although this may seem to be a trivial state-
to deliver vaginally rather than by repeat ment, it is not. As trial of labor becomes
cesarean operation. Having recognized the more common, doctors and nurses seem to
possibility of nonequivalence of the two worry less about previous incisions. This can
groups, we used multivariate regression be a serious mistake. Many obstetricians
methods to adjust for differences in baseline who cover the labor and delivery area for
factors to evaluate the independent contri- colleagues may have had the experience, as I
bution of trial of labor to perinatal outcome. have, of trying to confirm a prior incision
During a 3-year interval, we studied 7229 type in the middle of the night. On many
patients; 5022 (70%) had a trial oflabor and occasions I found myself reviewing a clinic
2207 had elective repeat cesarean opera- chart that documented that the patient had
tions. 15 Of those patients opting for labor, 10 or 12 prenatal visits with absolutely no
3747 (75%) went on to deliver vaginally. mention of any attempt to obtain the old
When the two groups were compared, the records. This is not a pleasant experience at
length of hospital stay, the incidence of post- 3 A.M. I recently heard of a malpractice case
partum transfusion, and the incidence of that illustrates the hazards of this type of
postpartum fever were all significantly oversight. A catastrophic uterine rupture
higher (p :s; .0001) in the elective repeat occurred during a trial oflabor. No attempt
cesarean group. The incidence of uterine had been made to document to prior incision
rupture was less than 1 %. This study did not type. After the poor outcome, a simple call
prove that trial of labor is safer or better to the hospital where the primary cesarean
than elective repeat cesarean. The study had been performed confirmed that the init-
did demonstrate, on a large scale, that ial operation had been a classical cesarean.
equivalent outcomes could be achieved Earlier knowledge of this fact could have
62 B.L. Flamm

saved the patient and her doctor a lot of excellent outcomes in spite of complete
grief. uterine ruptures. Of course, the policy has
From a medical-legal perspective it is im- also led to emergency operations in which
portant to document in the patient's chart the uterus was found to be totally intact. We
that an attempt has been made to try to believe that this is not an error in manage-
determine the previous uterine incision ment. Even with this aggressive policy, only
type. However, there is a big difference be- 1 % or 2% of all trials of labor will involve a
tween failure to attempt to get the records "stat" cesarean. 50
and not being able to do so after a reasonable
effort. The inability to obtain prior medical Use Oxytocin with Care
records is certainly not a contraindication to
A malpractice attorney once told me that
trial of labor. As was discussed earlier in
oxytocin put all his children through college.
this chapter, several papers have shown that
I do not think he was joking. Although the
trial of labor with "unknown" scar type is
studies listed earlier reveal that it is reason-
relatively safe.
able to use oxytocin during a trial of labor,
the admonition for caution is worth heeding.
Act Quickly for Suspected Fetal Distress If the cervix is not closed, consideration
Malpractice cases involving the manage- should be given to the use of an intrauterine
ment of patients with prior uterine incisions pressure catheter. This will not necessarily
generally do not focus on whether the patient yield any valid information about uterine
was a reasonable VBAC candidate. ACOG rupture (as has been discussed), but it will
guidelines and the medical literature make allow a better analysis of the contractions
it clear that, with the exception of known that are being generated. Because oxytocin
prior classical or vertical uterine incisions, has been associated with rupture of even the
there are few contraindications to trial of unscarred uterus, we must not be cavalier
labor. The focus of litigation, as is true in with its use during trial of labor. A recent
most obstetric cases, will be on the fetal malpractice case involved uterine rupture
monitor recordings. Although proponents of during a trial of labor in which oxytocin was
universal fetal monitoring have long held being administered at almost 60 mu/min.
that the mountains of fetal monitor strips Although oxytocin protocols differ from one
accumulating across America will be used to hospital to the next, it might be wise to
defend doctors, I am not so sure. I have met limit oxytocin dosages in patients with prior
"expert" plaintiff's witnesses who can find cesareans to 20 or 30 mu/min. A recent re-
subtle "late" decelerations at some time view pointed out that, despite widespread
during the course of almost any labor. But use and many research investigations, no
years of experience with trial of labor leads consensus exists with respect to the appro-
to the conclusion that in many cases you do priate dosing of oxytocin. 51 However, active
not need a magnifying glass to detect signs management regimens that use a maximum
of complete uterine rupture. The rupture dose of 40 mu/min or more have often51
often presents with the sudden onset of although not always 52 involved nulliparous
severe variable decelerations or a prolonged patients who are at lower risk for uterine
bradycardia. Such findings during the course rupture.
of a trial of labor should lead to an im-
mediate trip to the operating room. There is
no point in wasting time performing ultra- References
sound examinations or fetal scalp samples. 1. Pitkin R. Once a cesarean? Obstet Gynecol
These cases should be treated with the same 1991;77(6):939.
urgency as suspected abruption or umbilical 2. 1992 Sourcebook of health insurance data.
cord prolapse. By following this policy we Washington, DC: Health Insurance As-
have delivered many vigorous infants with sociation of America, 1992.
5. Vaginal Birth After Cesarean Section 63

3. Steinbrook R. Half the cesarean operations catheters useful in the diagnosis? Am J Ob-
in U.S. called unnecessary. Los Angeles stet Gynecol 1989;161:666-669.
Times, January 27, 1989 (Part 1, p 14). 19. Beckley S, Gee H, Newton JR. Scar rup-
4. Taffel S, Placek P, Liss T. Trends in the ture in labor after previous cesarean
United States cesarean section rate and section: the role of uterine activity measure-
reasons for the 1980-85 rise. Am J Public ment. Br J Obstet Gynaecol 1991;98:265-
Health 1987;77:955-959. 269.
5. Shiono P, McNellis D,Rhoads G. Reasons for 20. Devoe L, Croom C, Youssef A, Murray C.
the rising cesarean delivery rates: 1978- The prediction of "controlled" uterine
1984. Obstet GynecoI1987;69:696-700. rupture by the use of intrauterine pressure
6. Flamm B, Lim 0, Jones C, Fallon D, New- catheters. Obtet Gynecol 1992;80:626-
man L, Mantis J. Vaginal birth after cesarean 629.
section: results of a multicenter study. Am J 21. Conningham FG, MacDonald PC, Gant NE,
Obstet Gynecol 1988;158:1079-1084. eds. Williams Obstetrics, 18th Ed. East Nor-
7. Flamm B, Neuman L, Fallon D, Thomas S, walk, Connection: Appleton Lange, 1989:
Yoshida M. Vaginal birth after cesarean sec- 321,345.
tion: results of a five-year multicenter study. 22. Ventura S, Taffel S, Mathews T. Advance
Obstet Gynecol 1990;76:750-754. report of maternal and infant health data
8. Paul R, Phelan J, Yeh S. Trial oflabor in the from the birth certificate, 1990. Month Vital
patient with a prior cesarean birth. Am J Stat Rep 1993;42(2):1-31.
Obstet Gynecol 1985;151:297-303. 23. Flamm B, Goings J, Fuelberth N, Fischer-
9. Lavin J, Stephens R, Miodovnik M, Barden mann E, Jones C, Hersh E. Oxytocin during
T. Vaginal delivery in patients with a prior labor after previous cesarean section: results
cesarean section. Am J Obstet Gynecol 1982; of a multicenter study. Obstet Gynecol 1987;
59:135-148. 70:709-712.
10. Flamm B. Vaginal birth after cesarean sec- 24. Farmakides G, Duvivier R, Schulman H,
tion: controversies old and new. Clin Obstet Schneider E, Biordi J. Vaginal birth after
GynecoI1985;28:735-745. two or more previous cesarean sections. Am J
11. Molloy B, Sheil 0, Duignan N. Delivery after Obstet Gynecol 1987;156:565-566.
cesarean section: review of 2176 consecutive 25. Pruett K, Kirshon B, Cotton D, Poindexter A.
cases. Br Med J 1987;294:1645-1647. Is vaginal birth after two or more cesarean
12. Nielsen T, Ljungblad U, Hagberg H. Rupture sections safe? Obstet Gynecol 1988;72:163-
and dehiscence of cesarean section scar dur- 165.
ing pregnancy and delivery. Am J Obstet 26. Novas J, Myers S, Gleicher N. Obstetric out-
Gynecol 1989;160(3):569-573. come of patients with more than one previous
13. Miller D, Paul R, Diaz F. Vaginal birth after cesarean section. Am J Obstet Gynecol 1989;
cesarean delivery: a ten-year experience. 160:364-367.
Obstet Gynecol 1994;84:255-258. 27. Phelan J, Ahn M, Diaz F, Brar H, Rodriguez
14. Rosen M, Dickinson J, Westhoff C. Vaginal M. Twice a cesarean, always a cesarean? Ob-
birth after cesarean: a meta-analysis of mor- stet GynecoI1989;73(2):161-165.
bidity and mortality. Obstet Gynecol 1991; 28. Ong HC, Sen DK. Clinical estimation of fetal
77(3):465-470. weight. Am J Obstet Gynecol 1972;112:877-
15. Flamm B, Goings J, Yunbao L, Wolde-Tsadik 880.
G. Elective repeat cesarean delivery versus 29. Miller JM, Brown HL, Khawli OF, Pastorek
trial of labor: a prospective multicenter JG, Galbert HA. Ultrasonic identification of
study. Obstet Gynecol 1994;83:927-932. the macrosomic fetus. Am J Ostet Gynecol
16. Jones R, Nagashima A, Hartnett-Goodman 1988;159:1110-1114.
M, Goodlin R. Rupture of low transverse 30. Miller JM, Kissling GA, Brown HL, Gabert
cesarean scars during trial of labor. Obstet HA. Estimated fetal weight: applicability
Gynecol 1991;77(6):815-817. to small and large-for-gestational-age fetus.
17. Scott J. Mandatory trial of labor after JCU 1988;16:95-97.
cesarean delivery: an alternative viewpoint. 31. Benacerraf BR, Gelman R, Frigoletto FD.
Obstet Gynecol 1991;77(6):811-814. Sonographically estimated fetal weights:
18. Rodriguez M, Masaki D, Phelan J, Diaz F. Accuracy and limitation. Am J Obstet
Uterine rupture: are intrauterine pressure Gynecol 1988;159:1118-1121.
64 B.L.Flamm

32. Deter RL, Hadlock FP. Use of ultrasound in vious cesarean section. Am J Obstet Gynecol
the detection of macrosomia: A review. J Clin 1991;165:370-372.
Ultrasound 1985;13:519-524. 43. Hibbard L. Cesarean section and other surgi-
33. Flamm B, Goings J. Vaginal birth after cal procedures. In: Gabbe S, Niebyl J, Simp-
cesarean section: is suspected fetal macro- son J, eds. Obstetrics, normal and problem
somia a contraindication? Obstet Gynecol pregnancies, New York: Churchill Living-
1989;74:694-697. stone, 1986:522-523.
34. Phelan J. Trial of labor in women with 44. Tucker M, Hauth J, Hodgkins P, Owen J,
macrosomic infants. J Reprod Med 1984;29: Winkler C. Trial oflabor after a one-layer or
36-40. two-layer closure of a low transverse uterine
35. Pruett K, Kirshon B, Cotton D. Unknown incision. Obstet Gynecol 1993;168:545-546.
uterine scar and trial of labor. Am J Obstet 45. MacKenzie I, Bradley S. Vaginal prostag-
Gynecol 1988;159:807-810. landins and labor induction for patients pre-
36. Michaels W, Thompson H, Boutt A, viously delivered by cesarean section. Br J
Schreibner B, Michaels S, Karo J. Ultra- Obstet GynecoI1984;91:7.
sound diagnosis of defects in the scarred 46. Goldberger S, Rosen D, Michaeli G, Markov
lower uterine segment during pregnancy. S, Ben-Nun I, Fejgin M. The use of PGE 2
Obstet GynecoI1988;71:112-120. for induction of labor in parturients with a
37. Gilbert L, Saunders N, Sharp F. The man- previous cesarean section scar. Acta Obstet
agement of multiple pregnancy in women Gynecol Scand 1989;68(2):523-526.
with a lower-segment cesarean scar. Is 47. Blanco J, Collins M, Willis D, Prien S. Pros-
repeat cesarean section really the safe taglandin E2 gel induction of patients with a
option? Br J Obstet Gynecol 1988;95:1312- prior low transverse cesarean section. Am J
1316. Perinatol 1992;9(2):80-83.
38. Strong T, Phelan J, Ahn M, Sarno A. Vaginal 48. Shy K, LoGerfo J, Karp L. Evaluation
birth after cesarean delivery in the twin of elective repeat cesarean section as a
gestation. Am J Obstet Gynecol 1989;161: standard of care: an application of decision
29-32. analysis. Am J Obstet Gynecol 1981;139:
39. Allahbadia N. Vaginal delivery following 123-129.
cesarean section. Am J Obstet Gynecol 1963; 49. Flamm B. VBAC: low risk, not no risk. Con-
85(2):241-249. temp Ob/Gyn 1991;October(10):24-32.
40. Ophir EOM, Yagoda A, Markovits Y, 50. Gibbs C. Planned vaginal delivery following
Rojansky N, Shapiro H. Breech presentation cesarean section. Clin Obstet Gynecol 1980;
after cesarean section: always a section? Am 23(2):508-515.
J Obstet Gynecol 1989;161:25-28. 51. Baker ER, D'Alton ME. Management oflabor
41. Sarno AP, PhelaiI JP, Ahn MO, Strong TH. in the nullipara. Clin Consult Ob/Gyn 1992;
Vaginal birth after cesaran delivery: trial of 4:218-230.
labor in women with breech presentation. 52. O'Driscoll K, Foley M, MacDonald D. Active
J Reprod Med 1989;34:831-833. management of labor as an alternative to
42. Flamm B, Fried M, Lonky N, Saurenman- cesarean section for dystocia. Obstet Gynecol
Giles W. External cephalic version after pre- 1984;63:485-490.
6
Breech Presentation
LUIS A. CIBILS

It is generally accepted that the normal fetal malformations, intrauterine growth retar-
lie at term is longitudinal, which indicates dation, abnormal implantation of the pla-
that the maternal and fetal spines are par- centa, prolapse of the umbilical cord, and
allel. Furthermore, the normal presentation uterine malformations.
at term is cephalic, meaning that the cephalic
pole of the fetus presents to the pelvic inlet
and the birth canal. The physical relation- Incidence
ship between fetus and birth canal is the
most favorable for delivery when these two The relatively excessive volume of amniotic
conditions are present. It follows that any fluid in proportion to fetal volume, normally
deviation from this physical relationship occurring in mid- and preterm gestation,
is an abnormality of variable importance. facilitates the instability of the fetal lie
Breech presentation is such an abnormality, and presentation. As a consequence the
one in which the lie is longitudinal but the prevalence of abnormal lie and presentation
presenting part is the lower pole of the fetus. is as high as 45% in late second trimester.1
There are several variants in attitude of Likewise, the prevalence of breech presen-
breech presentation, depending on the re- tation ranges from 24% to 33% in late second
lationship of the fetal body and the lower and early third trimester. 1,2 With the ad-
limbs, and also the lower limbs position: vance of pregnancy, these numbers decrease
they are known as complete, frank, footling steadily because there is proportionally less
(single or double), and kneeling. Each one of amniotic fluid, and by 33 weeks gestation
these attitudes has specific clinical charac- only 10% to 14% are still in breech presen-
teristics and types of complications which tation. About one-half of these fetuses will
make the importance of accurate diagnosis convert spontaneously to cephalic presen-
of more than academic interest. tation by the time they reach full term,1-3
Breech presentation in any attitude is when the prevalence of breech presentation
associated with a number of complications ranges from 3% to 7% according to standard
that contribute heavily to increase the textbooks. It is therefore understandable
perinatal mortality (PNM) and perinatal that when labor ensues prematurely the
morbidity of infants born by the breech. incidence of breech presentation is inversely
More detailed discussion follows in various related to the age of gestation. In other
sections of this chapter; suffice to say here words, premature labors carry a high in-
that among the more prevalent conditions cidence of breech presentations, and the
associated with breech presentation are earlier the labor, the higher the incidence of
premature labor and delivery, congenital breech presentations.

65
66 L.A. Cibils

Importance presentation and delivery should make us

pause and be very cautious when confronted
Many very interesting studies have been with one such case. Likewise, one must take
carried out in the past several years regard- these factors into consideration when re-
ing the morbidity and mortality of infants viewing published material, or when con-
presenting by the breech. Those observations sidering management recommendations
were essentially promoted by the need to based on dramatic differences with cephalic
determine the potential causes of their sub- presentations. Some long-term follow-up
stantially greater PNM and morbidity.4-8 studies of survivors delivered at various
Breech presentations seem to have in- gestational ages and according to route of
trinsic characteristics, such as relatively delivery are reviewed later in this chapter.
lower placental weights,S an extremely high One of the aspects of breech presentation
incidence of congenital anomalies,s-9 and receiving even more attention than the
premature deliveries. s- 8,lo Breech-delivered problems mentioned in the previous para-
infants appear to have an inherently lower graphs is the mode of delivery advised for
weight at birth even when corrections are those fetuses. In the United States, about
made for gestational age, and this deficit 85% of all breech presentations are currently
seems to persist in early infancyY On the delivered by cesarean section, regardless of
other hand, the observation that fetuses attitude or gestational age. 17 This represents
with a severe congenital handicap appear to around 10% of all Cesarean sections, close to
present by the breech in an inordinately 4% of all deliveries or approximately 100,000
high proportion prompted Braun to ask "why operations done every year. The importance
the fetus failed to assume the vertex posi- of these numbers with regard to morbidity,
tion.,,12 The average length of the umbilical mortality, and health care expenditures are
cord has been shown to be longer in cephalic thoroughly discussed in other chapters of
presentations, the observers suggesting this book. The perusal of those chapters
that this may be caused by increased fetal would suggest that a decision to perform a
activity.13 This postulated variation in ac- cesarean section should not be taken lightly
tivity may be related to other anatomic and without considering every aspect of
differences such as congenital hip disloca- obstetric management, including maternal
tion, found by Clausen to be nearly four morbidity and mortality.
times more frequent among infants born in
breech presentation. 14 Whether this ab-
normality is the cause or consequence of the Perinatal Mortality
fetal position in utero is not clear; however,
Luterkort found that there is an extremely Perinatal mortality traditionally has been
high proportion of hip joint instability among quoted as being 5- to 10-fold higher for breech
breech deliveries, particularly prevalent deliveries compared to cephalic presenta-
among those in frank attitude. IS The prob- tions. This fact is still currently observed.
ability that inherently anatomic or phy- However, it is important to attempt to study
siologic anomalies predispose to breech separately the various pathologic factors
presentation has been reinforced by the ob- that cluster around breech presentations.
servation by Nelson and Ellenberg16 that The PNM ranges from nearly 10%8 to 25%,s
about one-third of breech presentations with depending on the population studied or the
cerebral palsy had major malformations years when the material was collected. In
unrelated to the central nervous system our material of nearly 1150 consecutive
(CNS).16 These observations confirmed a cases observed over a 16-year period, there
prior report by Kauppila 6 about CNS and was a 24% uncorrected PNM, which is sever-
hip malformations. alfold higher than for cephalic presentation.
This partial review of the numerous ab- Prematurity occurs with extreme fre-
normal conditions associated with breech quency among all series, probably for the
6. Breech Presentation 67

reasons just discussed, and ranges from 22% dropping between the legs does not predis-
to 35%,5-7 which is three- to fourfold higher pose to significant cord compression and also
than the 6%-8% usually quoted as the because of the usual rapid termination of
prevalence for the general population in this pregnancy when the condition is recognized.
country. In our material the prematurity In our total population (term and preterm),
rate was 53% by weight, that is, fetuses it occurred in 7% of cases.
weighing less than 2500 g, and 51% if clas- Uterine malformations are classically
sified by gestational age (36 weeks or less).18 mentioned as causing breech presentation,
This extremely high rate of prematurity is but actually are mentioned as contributing
the direct cause of nearly one-half of all to only a small percentage of cases in the
perinatal (PN) deaths (11% of all births publications discussing the point.
in our population) and almost two-thirds of Tables 6.1, 6.2, and 6.3 summarize the
neonatal deaths (NND). There is an intrinsic perinatal results observed by authors who
trap with this condition because the more published the results of materials studied
premature a labor, the higher the chances of with different methods and collected under
a breech presentation,I,2 and thus more various circumstances. There is an improve-
cases of extreme prematurity are born III ment as the years have passed, but the
breech. mortality and morbidity are still very high
Premature rupture of membranes is, of compared to known figures for cephalic
course, the triggering of many premature presentations.
labors but at the same time predisposes to
prolapse ofthe umbilical cord, which may be
a life-threatening complication when pro-
ducing cord compression. It occurs with
Mode of Deli very
variable frequency depending on the attitude
of the presentation or age of gestation. It is
Neonatal Effects
variably quoted as about 5% for term-size From the foregoing discussion it is clear that
infants,6 while for others of the order of breech presentation at the time of delivery is
3%,19 or even less than 1%.20 It is fortunate associated with a number of high-risk con-
that the majority of these fetuses are de- ditions. The prevailing concept, during the
livered in excellent condition, perhaps be- early to middle part ofthis century, was that
cause the mechanical condition of the cord much of the PNM and morbidity could be

TABLE 6.1. Perinatal results (all ages)a

Corrected perinatal
Reference Number of cases Mortality Prematurity Malformations mortality (PNM)

Hall et al., 1965 21 6,044 12.3% 24.4% 4.7% 6.5%

Brenner et al., 19745 1,016 25.4% 36.0% 6.0% 11.0%
Kauppila, 19756 2,227 12.5% 20.0% 10.0%
Fianu, 1976 7 1,878 7.2% 15.1%
Lewis et al., 197922 579 10.4% 24.2%
Mann and Gallant, 197923 457 19.9% >24.0% >1.5% 5.9%
Sachs et al., 198324 9,626 2.5%
White and Cibils, 198410 302 7.9% 32.1% 12.0% 6.9%
Schutte et al., 1985 8 4,628 9.7% 17.1% 4.4% 3.5%
Fortney et al., 198625 10,749 5.2%
Kiely, 1991 26 17,587 9.7% 4.9%
Brown et al., 199418 843 24.0% 52.7% 9.4% 14.9%

aThe cases published until 1983 include only infants weighing more than 1000 g. Only the publications after 1983
include fetuses weighing 500 g and more. The calculations were made from information and figures given in the
respective publications.
68 L.A. Cibils

TABLE 6.2. Perinatal results (term, 2500 g or more)a

Corrected perinatal
Number of cases Mortality Malformations mortality (PNM)
Hall et aI., 196521 4,568 4.0% 1.1% 2.9%
Rovinsky et aI., 19734 2,145 3.2% 2.1% 0.6%
Kauppila, 19756 1,779 2.8% 9.0%
Fianu, 19767 1,595 2.8%
Sachs et aI., 198324 7,028 0.4%
White and Cibils, 198410 205 1.5% 6.0% 1.5%
Tatum et aI., 198520 580 1.1% 1.1% 0%
Schutte et aI., 19858 3,837 2.5% 3.5% 1.4%
Fortney et aI., 198625 8,546 1.6%
Kiely, 1991 26 13,694 2.5% 1.2%
Brown et aI., 199418 309 1.8% 0%

a The cases published until 1983 include only infants weighing more than 1000 g. Only the publications after 1983
include fetuses weighing 500 g and more. The calculations were made from information and figures given in the
respective publications.

TABLE 6.3. Perinatal results (prematures, less than 2500 g)a

Number of Corrected Corrected
cases PNM NNM PNM Malformations PNM,VLBW PNM,VLBW
Hall et aI., 196521 1476 34.3% 18.6% 22.3% 6.4% 70.7% 50.0%
Kauppila, 19756 448 51.1% 32.8% 32.4% 10.9% 90.1% 41.3%
Fianu, 19767 283 32.2% 17.0% 66.2%
Sachs et aI., 198324 1264 11.7% 38.5%
White and Cibils, 198410 97 21.6% 18.6% 25.0% 57.0% 52.0%
Schutte et aI., 1985 8 791 44.6% 20.5% 18.9% 6.2% 67.5% 54.0%
Fortny et aI., 198625 2203 19.2% 56.7%
Kiely, 1991 26 3893 17.8% 40.5%
Cibils et aI., 199427 444 42.5% 28.3% 22.5% 64.5% 53.3%

PNM, perinatal mortality; NNM, neonatal mortality; VLBW, very low birthweight.
aThe cases published until 1983 include only infants weighing more than 1000 g. Only the publications after 1983
include fetuses weighing 500 g and more. The calculations were made from information and figures given in the
respective publications.

prevented by improvements in conducting delivery." This, at the time radical, approach

the delivery and strict adherence to appro- did not gain immediate wide acceptance but
priate techniques. 28 However, significantly it was slowly introduced in the obstetric
higher complication rates than in cephalic practice, first in this country and then
presentations were still observed. As a con- abroad. By the early 1980s in the city of
sequence, in 1956 Goethals29 suggested that New York nearly 65% of all breeches were
cesarean section should be liberally indicated delivered abdominally26 while in some hos-
to deliver breech presentations at term pitals in Montreal (Canada) this incidence
when occurring in primiparas. 29 Shortly was only 22% in the early 1970s but rapidly
thereafter, in 1959, R.C. Wright30 postulated reached 94% by 1979. 31 This dramatic
that "any patient over 35 weeks gestation change in practice adopted in the North
who enters labor with a living baby in breech American continent spilled over other areas
presentation should be delivered by Cesar- of the world with more or less massive effect.
ean section, providing there is no maternal In some South American countries the
disease that contraindicates abdominal figures are similar to these for Canada, while
6. Breech Presentation 69

in Europe the incidence of cesareans in conclusions reached by these reviewers were

breech presentation was much lower, reach- contrary to that of some of the authors they
ing as much as 30% on the average for a quote in the table,38,44 most likely because
large number of departments surveyed by they disregarded the extremely important
Kubli et al. in the mid-1970s,32 or as in the variable of changing weight-specific mor-
Netherlands where in 1982 it was "only" tality from 500 to 1500 g birthweight by
26.5%.8 pooling the material in a single group.
After several years of this practice some Myers and Gleicher45 reviewed and re-
authors decided to review the outcomes, and analyzed six of those publications, in addition
compared cephalic and breech deliveries as to two others, and reached the opposite con-
well as abdominal and vaginal deliveries in clusion, that the route of delivery does not
both groups. For some it was not very con- influence neonatal outcome in these small
vincing that morbidity and mortality were infants, as they divided the cases into birth-
better among cesarean-delivered fetuses, in weights of less than 1000 g and more than
particular for term-size fetuses. 23 ,25 Others 1000 g. In our own institution, the population
strongly urged the systematic abdominal analyzed when adjusted for fetal weights
delivery of fetuses less than 1500 g, even and gestational age revealed no difference
acknowledging that there were no data to in neonatal outcome among those VLBW
substantiate that recommendation. 33 The infants delivered vaginally or by cesarean
same advice was given by others, who col- section. 27 This observation corroborates
lected their material several years later, what was found in 1987 in a small European
because their PNM was significantly better collaborative study reported by Thiery,
for those prematures delivered by Cesarean who concluded that "we are no longer con-
section. 34- 36 However, a number of authors vinced ... nor does our study support Cesar-
have been unable to observe a difference in ean delivery of all tiny neonates.,,46
PNM ascribable to mode of delivery for these Some authors have evaluated the long-
small infants. 37 - 41 From a recent large term association of mode of delivery with
multicenter study, Malloy et al. 42 concluded neurologic or intelligence quotient (lQ)
that "there were no statistical significant development in prematures. Effer and co-
associations in any of the birth weight strata workers 38 found no difference up to 3 years
between neonatal death or IVH and the of age, an observation replicated by Faber-
method of delivery." Nijholt et al. 47 among either "optimally
It is extremely important to very carefully matched" at 3-10 years of age or "paired" at
evaluate the potential confounding factors 18 months. From the foregoing it appears
in those reports. Two review publications that those who advocate routine cesarean
analyzing the same materials arrived at section for the VLBW fetus presenting by
completely different conclusions for a variety the breech have not been able to objectively
of reasons. Westgren and Paul43 tabulated demonstrate the validity of their recOm-
14 publications ostensibly comparing out- mendation on the basis of medical facts. In
comes of vaginal and cesarean deliveries in spite of this, a current textbook states that
very low birthweight (VLBW) breech pre- "the evidence presented above" (meaning
sentations. However, they included a study references 37-39, 43, 46) "is sufficient to
in which breech deliveries were compared support this practice: to perform a Cesarean
to cephalic vaginal deliveries and undeter- delivery for any live fetus presenting breech
mined presentations born by cesarean who weighs less than 2000 grams and more
sections. 33 They tabulated more than 8000 than 700 grams (26 weeks' gestation).,,48 It
breech presentations when, in fact, in the seems that there is a misinterpretation of
quoted publications only 823 and 523 cases "the evidence."
were VLBW breeches, respectively delivered The management of the term breech fetus
vaginally and abdominally. Furthermore, the does not significantly vary from that ob-
70 L.A. Cibils

served for the prematures or VLBW fetuses. are enforced in single institutions, these
The recommendation made by Goethals in results can be achieved. The latter authors
195629 for primiparas and later extended by concluded "the adjusted relative risk esti-
Wright30 that all breech presentations mate for the combined outcome category of
should be delivered by cesarean section be- head trauma, neonatal seizures, cerebral
came the standard of practice in this country palsy, mental retardation, or spasticity was
in the late 1970s and early 1980s. This dictum 0.5 in vaginal delivered infants (95% CI
spread through Canada31 and Europe. 46 In 0.1-3.2)."62 Likewise, the observations
view of the alarming rise of cesarean section carried out in our hospital failed to show any
rates in this country, which reached nearly difference whatsoever among fetuses de-
25% in 1988,17 efforts were made to evaluate livered either way; in fact, all the deaths
some of the indications with the aim of con- among the cases with fetuses weighing more
trolling its progressive escalation. Among than 1500 g were in the group receiving an
the contributing conditions, breech presen- elective cesarean section, while none oc-
tation appeared to deserve an evaluation. curred among vaginal deliveries/ 8 the dif-
Two randomized studies were conducted in ference was not significant.
this country for frank 49 and nonfrank 19 Relatively few studies have focused on the
presentations, and both failed to demonstrate long-term developmental differences among
any immediate perinatal benefit for the these neonates. Luterkort and collaborators
fetus but showed a significantly increased followed prospectively more than 130 breech
morbidity for the mother among those who presentations in late gestation from which
had an abdominal delivery. slightly more than one-half turned to
Numerous observational studies made in cephalic. When discussing mode of delivery
various areas of the world were also aimed they concluded that "according to multiple
at assessing the advantages, if any, of linear regression analysis the breech pre-
abdominal delivery over the vaginal route sentation and the vaginal delivery per se
for these infants. Barlow and Larsson50 were not factors in influencing the neuro-
identified patients "who could give birth logical score.,,3 Westgren and Ingemarsson57
vaginally without any mortality or per- stated that "vaginal delivery for full term
sistent morbidity." Oian and coworkers51 breech presentations show no significant dif-
did not observe any change in perinatal ferences in outcome brought about by birth
outcome when the cesarean section rate route." Otamiri et al. 63 later confirmed those
increased fourfold among their breech pre- observations when, from evaluation of their
sentations, thus corroborating what had been own material, they wrote "it may be con-
seen by Green et al. 31 a few years earlier in cluded that infants born by elective Cesarean
Montreal when their cesarean section rate section and by vaginal breech delivery
for breeches increased from 22% to 94% showed deviations in neurological adapta-
without any appreciable change in neonatal tion during the first 5 days of age ... the
asphyxia, trauma, or death. Several other neurologic dysfunction was transient ... the
relatively small series, gathered in this infants showed normal growth, neurology
country and abroad, likewise failed to re- and psychomotor development at 6 months
port any immediate beneficial effect for follow-up.,,63 Nevertheless, several authors
the fetuses delivered by cesarean sec- have been able to observe at the time of
tion. 10,46,52-60 Two recent large reviews of delivery among infants in breech presen-
material collected in single institutions in tation (regardless of route) what appears to
this country by Flanagan et al. 61 and by be defective development when compared to
Croughan-Minihane and coworkers62 found their cephalic-presenting peers. Luterkort et
no effect of route of delivery on neonatal al. 3 observed, when comparing cephalic with
morbidity or mortality in their populations. breech presentations who were small for
It appears that when strict technical criteria gestational age (SGA) that the latter had
6. Breech Presentation 71

lower neurologic scores; however, "children to 89%, is much higher than that observed
in the breech group born vaginally or by (1 %-6%) among vaginal deliveries. 20 ,49,65,66
elective Cesarean section had similar neuro- Clearly, vaginal delivery offers more imme-
logical scores." Likewise, Westgren and diate safety to the mothers. This is true even
Ingemarsson57 concluded that "prospective when considering the frequent need to do
follow-up studies and carefully matched con- a mediolateral episiotomy4S that may ex-
trolled studies with sophisticated neuro- tend to third- or fourth-degree laceration. 2o
logical evaluations indicate that breech Appropriate repair should avoid further
infants, regardless of mode of delivery, problems.
will score slightly less favorably that infants An important aspect to remember about
born in vertex presentation". From similar cesarean delivery is the frequent necessity
observations, Hytten64 thought that "we of a repeat procedure in a subsequent preg-
have come to assume that the baby may be nancy, in spite of the current vogue and
brain damaged by the trauma of breech relative success in attempting vaginal
delivery, but it may often be that the baby delivery under that circumstance. 17 Many
is presenting by the breech because it is times, in particular when the operation is
already brain damaged." After reviewing performed for VLBW fetuses requiring a
their own material, Cox and collaborators37 vertical incision with extension into the
concluded that "the value of routine Cesar- corpus, the subsequent cesarean is inevitable
ean section in the absence of other obstetric because of the risk of dehiscence or rupture
pathology for low birth weight breeches in with those scars.
unproven and may be deleterious to mother
and baby."
Characteristics of Breech Labor
Maternal Effects From the foregoing it appears that, barring
It is likely, because almost all normal medical or obstetric contraindications to
mothers will accept any sacrifice for the sake labor, a trial of labor should be permissible
of their infants, that little mention is made in most cases of breech presentation. How-
in publications and discussions about mater- ever, to properly conduct such labor the
nal morbidity related to mode of delivery obstetrician must be well informed of the
of the breech-presenting fetus. The few pub- particular aspects affecting the fetus and
lications in which this point is reviewed mother during labor.
uniformly report a significant incidence The well-known differences due to parity
of maternal febrile morbidity among the that are observed in the cervix apply equally
patients receiving a cesarean section.* The to breech presentation. Likewise, the uterine
need of blood transfusion because of excessive contractility pattern is indistinguishable
bleeding19,20,49 or even the need to perform from that observed in cephalic presentations
hysterectomy as the result of intraoperatory (Fig. 6.1), and its effectiveness to dilate the
complications is periodically reported. 2o ,49 cervix is the same. 67 Also, the labor curve
The occurrence of postoperatory pelvic ab- pattern against time is not different from
scesses or evisceration following cesarean cephalic presentations. 6s Monitoring of the
section has been observed by several au- fetus, either biochemical or electronic, has
thors. 20 ,49,65,66 Maternal mortality caused by been applied to the breech-presenting fetus.
postoperatory misadventures is more com- In this case, it has a characteristic pattern
monly observed among cesarean deliveries. 46 different in some aspects from the average
The commonly quoted incidence of post- cephalic presentation. Few authors studied
operative complications, ranging from 20% these points specifically. The first publication
discussing specifically the breech-presenting
* References 19, 20, 49, 53, 65. fetus was by Teteris et al.,69 but it described
72 L.A. Cibils
DELIVERED 20 ml nu' .. loter ASSISTED, MAURICEAU, EPIDURAL APGAR 4-8
."1-.0-"" FETUS 2430 gm Placenta 500 gm

-HI--
~
-
-:-
1:- ;
-:-
~~
--
iN I\ ~ II

1:: - - :
I I- h - -
14- ff
.. r==-

FIGURE 6.1. Premature rupture of membranes (PRM). Induced labor, primipara, 36 weeks gestation.
Top half of chart, 50 min of direct monitoring show oxytocin-driven early first stage (3 cm dilatation)
with normal contraction pattern. The fetal heart rate (FHR) shows large accelerations blunted by
decelerations (cord compression) on a baseline of 135 beats per minute. Bottom part, 140 min after the
end oftop part of recording, under epidural anesthesia, shows advanced first-stage uterine contractility
(UC) pattern (9 cm to complete). The FHR baseline has risen to about 160; the accelerations are
blunted earlier by deep variable decelerations. Twenty minutes after the end of the tracing, a fetus
in good condition was delivered by assisted technique with Mauriceau maneuver. Recording paper,
lcm/min.

only the characteristics and differences meaningful tracings in second stage and
from cephalics in the second stage. They thus advised the concomitant use of bio-
emphasized the predictable sustained decel- chemical monitoring.
eration during breech delivery and im- White and Cibils lO studied specifically
plicated as a possible mechanism the "chest FHR alterations in this presentation and
compression" during the process of delivery. reported a 63% incidence of variable dece-
Wheeler and Greene 70 described the frequent lerations ("cord compression pattern") with
occurrence of variable decelerations and one-third of them having the associated
baseline changes, and correlated them with "late" component indicating transient hy-
neonatal depression and pH changes. They, poxemia. They further reported the impor-
rightfully, attributed the fetal heartrate tance of assessing changes of the baseline
(FHR) alterations to cord compressions, (tachycardia, fixed or saltatory) occurring in
and ascribed the accelerations observed to association with the decelerations as means
"abdominal compression of the fetus during of assessing fetal well-being. In their ex-
contractions." Hill and collaborators 71 con- perience these associations predicted a high
firmed the observations made by Wheeler incidence of depressed I-min Apgar scores
but reported much difficulty in obtaining (Figures 6.1 and 6.2) compared to those who
6. Breech Presentation 73

Shirodkar CUT p.e.B. Anesthesio ASSISTED DELIVERY APGAR 5-9

pH pO, pCO, HCO. BE
Fetus 1270 9ft!
CORD BLOOD UfIIb. At!. 7.21 16 56 22 -5
Placenta 2~0 9m

FIGURE 6.2. Induction oflabor, multipara PRM, 27 weeks. Frank breech. Top tracing and left third of
lower tracing show 54 min of advanced first-stage DC and FHR pattern under paracervical block
anesthesia. Note mild tachycardia and variable decelerations. Lower right two-thirds shows the
second-stage FHR tracing with deep sustained variable decelerations followed by tachycardia and
fixed baseline. Assisted breech delivery occurred at the end of the tracing. Shown on top Apgar scores,
fetal weight, and umbilical cord gases: mild respiratory acidosis. In this and subsequent figures the
time marks below the tracings indicate minutes; note in the last 6 min the paper speed was increased
to 3 em/min. (From Cibils LA,72 by permission of Parthenon Publishing.)

had only simple variable decelerations with ably represents almost complete interruption
a normal baseline (Figure 6.3). On the other of umbilical cord circulation and sets the
hand, they observed accelerations in a very stage for early respiratory acidosis.
high percentages of cases (73%), either iso- Repetitive decelerations caused by cord
lated or preceding decelerations, and cor- compression of course may induce alterations
related them with good outcome (Figures 6.3 in fetal homeostasis when they either occur
and 6.4). Their presence was attributed to too frequently, are of very long duration, or
partial compression of the cord (umbilical when the fetal condition is already borderline
vein) and, when observed in late first stage compromised by high-risk factors or insuf-
or second stage, predicted a neonate in very ficient intervillous space circulation. The
good condition. In a later report, Cibils72 understanding of this pathophysiology 73 is
observed the almost invariable occurrence essential for proper management of those
the significant sustained decelerations (cord cases; the management could be either con-
compression pattern) during late second tinued observation or rapid intervention
stage and the process of delivery, as illus- (see technique of vaginal delivery) depending
trated in Figures 6.2 through 6.4. This prob- on the characteristics of the tracing. The
74 L.A. Cibils

ASSISTED DELIVERY, PIPERS F.tul 3340 1m Placenta 700 1m APGAR 1-9

. 11, 'I' 4'1 31 10 lit

, ffi
g: I.1;i
,
101 I -
.~
1. 30 II. e em Ollalallen
EPIDUfUl
ANESTHESIA -

.. .
. CORO BLOOO

pH ,Co. P

FIGURE 6.3. Induction of labor, secundipara, 41 weeks. Frank breech. Top tracing shows 60min of
advanced first-stage direct tracings under effective epidural. Normal oscillating baseline shows
accelerations triggered by almost every contraction. Lower tracing shows 17.5 min (paper speed, 3 cm/
min) of last part of second stage: accelerations were blunted by variable decelerations when the
patient pushed. Assisted delivery with Piper forceps occurred at the last contraction. Fetal data and
cord gases are shown. (From Cibils LA,72 by permission of Parthenon Publishing.)

changes in fetal blood homeostasis, when excellent correlation between buttocks blood
they occur, will be manifested by the fore- samples and umbilical arterial blood pH was
mentioned baseline alterations, and deter- documented by Brady and coworkers. 75 Sub-
ioration of aicd - base balance. Thus the fetal sequent observations by others compared
blood pH will fall, first because of accu- vaginal and abdominal breech deliveries
mulation of CO2 (respiratory acidosis) and found no differences in umbilical artery
(Figure 6.2) and, if the situation continues, pHs. 76 ,77 However, others found lower um-
the added loss of buffer bases (metabolic bilical artery pHs among breech vaginal de-
acidosis) (Figure 6.5). liveries when compared to cesarean breech
The rapid induction of respiratory acidosis deliveries,78 but did not find to be of clinical
with falling blood pH was shown in 1972 by significance. 6o ,77 This observation prompted
Eliot and Hill in an excellent study that also Socol et al. 77 to conclude that "the liberal
documented the rapid recovery in the first use of Cesarean section may improve Apgar
5min of life. 74 Kubli and coworkers32 re- scores but will probably not appreciably im-
ported the high proportion of umbilical artery prove the acid-base status of the breech
acidosis, which they attributed to cord com- neonate." This finding should, in fact, be
pression, and elaborated on the unpredic- expected if one remembers that during de-
tability of that outcome. At about the same livery there is a sustained cord compression,
time, Hill and collaborators71 suggested the the duration of which dictates the degree of
routine use of fetal blood sampling in the respiratory acidosis. The difference in pH
management of second-stage breech de- and base buffers between umbilical ar-
liveries in view of their difficulties to obtain terial and venous blood documents this
good FHR tracings. The fact that there is an point,72 and should therefore be taken into
6. Breech Presentation 75

ASSISTED BREECH DELIVERY under EPIDURAL ANESTHESIA

FETUS 4130 gm Ploclnto 810 gm APGAR 8·10 y ..... 1.11
• .t'·,,-In Cotd BloOd pH • ... az
7 .• "

" 0 . 10

~
l±
l±

."'.. ..
'-. .... ......
~

. "'.
~

W I
> •
::;
Q

I " . 10 , .

..... .:.:..' .... .. .. .. .-.. . . -- .... .... , .. '.'" !.-

..-.
FIGURE 6.4. Enhanced labor, multipara, at term. Frank breech. Top tracing shows 50 min of advanced
first stage. Normal DC pattern triggered some FHR accelerations. Bottom tracing, 80 min later, shows
the last 45 min of second stage: pushing with contractions and accelerations, tachycardia in latter
part. Two sustained variable decelerations preceded the assisted delivery of good neonate with cord
around neck. Fetal information and cord gases are shown on top. (From Cibils LA,72 by permission of
Parthenon Publishing.)

serious consideration for the technique of One of the most frequently mentioned
delivery as illustrated in Figures 6.2, 6.4, indications for cesarean sections in cases of
and 6.5. trial of labor is the so-called dystocia or
From the foregoing it seems clear that "poor progress," the implication being that
labor in breech presentations presents the fetal head will be too large for the pelvis
characteristics predisposing, if unchecked, and thus that either the uterine contractility
to fetal deterioration on short notice; thus, is abnormal or the cervix does not dilate as a
continuous observation and adequate moni- result of the disproportion. It is hard to
toring are essential for correct management. understand how this theoretical mechanical
Another well-known intrapartum risk is the maladaptation can play any part in the pro-
possibility of cord prolapse, which occurs in cess of the first stage oflabor, but it is duti-
1% to 20% of cases depending on the attitude fully mentioned in every textbook available.
of the breech presentation and the length
of the pregnancy. Fortunately, severe cord
compression is not frequently observed, and
The Use of Oxytocin
there is almost always time for successful As a corollary of the concept just discussed,
intervention. In our material there was a 7% it is also stated that oxytocin should not be
cord prolapse. 1s This accident may occur in a used to attempt to correct the labor abnor-
quite unexpected way even when the FHR mality and that instead a cesarean section
tracing is completely normal (Figure 6.6). should be carried out. 19,4S, 79 The intravenous
76 L.A. Cibils

EPIDURAL, PROLONGED DECELERATION, EXTRACTION PIPERS

Fetu. 211110 om Plocuta 675 om APGAR 7-9 Cord Around BODY
Po. pCO. HCO, BE
.100 - 01 - '7
12 74 16 - 12

I, " I I , I •
..
FIGURE 6.5. Induced labor, grand multipara, at term for PRM. Complete breech. Continuous 72 min of
late first stage and second stage under effective epidural anesthesia. Top: average variable decelera-
tions and mild tachycardia, occasional accelerations. Bottom: deep variable decelerations blunt the
accelerations on an oscillating baseline. As descent progressed with pushes, the decelerations did not
recover (the last 10min of recording at 2cm/min) and prompted extraction of the fetus with the cord
around the body, thus explaining the sustained deceleration as this part of the fetus passed through
the birth canal. Likewise, this mechanical condition explains the large arteriovenous blood gases
differences and the mixed acidosis. (From Cibils LA,72 by permission of Parthenon Publishing.)

infusion of oxytocin in physiologic doses has tocin does not increase the incidence of
been, for the past 30 years, the standard undesirable outcomes. *
therapeutic means to attempt to correct Similarly, the use of physiologic doses of
labor abnormalities or induce labor when oxytocin to induce labor is taught to be con-
indicated. Strangely, even when some clini- traindicated48 while the review of published
cians would allow spontaneous labor to series indicate that this technique may
proceed with a breech presentation, the phy- benefit the patient without increasing the
siologic stimulation of uterine contractions risk for the fetus. t The rate of induced labors
(enhancement) would be absolutely con- ranges in those series from 4% to 60%. Most
traindicated in the same patients. 48 ,79 This interesting, in the majority of those series
widespread practice, which inflates the of induced labors there is an important com-
number of unnecessary cesarean sections, ponent of primiparas, an absolute contrain-
has never been shown to have a rational
practical basis. On the contrary, the majority
of publications in which this subject is dis- * References 4-6, 10, 18, 20, 27, 49, 61, 70, 80.
cussed clearly indicate that the use of oxy- tReferences 10, 18, 19,27,46,49,61,70,80.
6. Breech Presentation 77

NORMAL BASELINE, VARIABLE DECELERATIONS, BRADYCARDIA - CORD PROLAPSE

C -SECTION DE LIVE REO 10 ",Inu' .. later Umbil i cal Artery pH 7.24
E ME RGENCY
FE T US 22 5 0 g m Pla c .nta DOO 'Om APGAR 1- '
99 4~ , 99C ~ I 99463

- - .... t . . ....... . . - .
. CLl
, ~9~'~ , ~g ~ II I 99471
It I , 1:1 : III II I 11111-1 I I t III I II I, ' t t ill U ell; ~'II . rU rtl
,
I..t: I ~ . 11
II I I
~ , IIIt1
' 11
I
.
11 Ii III
' :..1i
-I I t. 11 j. t:r
I
rI

.
:
I=t IJj r ~. i

,., , 11m II ~ ·n
m
~ ' 11 I 1)'\"'1- """'1 -1'-""'\ ~ .11. .... ~ ~ ' !.T
,
•
,- It' RI4" J. ,W "'- ..11 JJ 1\ I ~

... 1l1li
" ~~.
: 1J1
!
• J: • II III 1
.. : II t; tI 1':::- -I-i t- Il .. I !I I 1 ':- .JJl
I";L; h:; . ... -- I '; I , , + j "l If

."" .,..
.
. ~I ..
, . . .J.- i - - - ~
, . Ii" >;' : - -07\.-: I. :
rtUl I n It j .J".~- i I'" . ' 1)., "ani ell.
~ 'ii. . V .\f... '~L " ':l. ;oJ" f-.,. ~i J... !":":.. '\. ..J\. ..n:. i J I ).r Jj.{ , ! l "i.
'ti· i iriL' ..i..4.1- t ij"f·'<-+ ' "4 ; hi I I i . Ii. .H .\ delt r-id Ii ' ;.f~ -':fH Eft

FIGURE 6.6. Spontaneous labor, secundipara, 34 weeks gestation. Frank breech. Continuous 78 min of
mid first stage: normal DC and FHR baseline, variable decelerations triggered by most contractions.
The last deceleration did not recover; an exam documented prolapse of the cord. This was followed by
delivery in lOmin, by cesarean section under general anesthesia, of good fetus with Apgars 1 and 5
and umbilical artery pH of 7.24.

dication in the lore of breech presentation. It The Fetus

seems clear from the published evidence
that the appropriate use of oxytocin to The risk of prolapse of the umbilical cord
enhance labor may safely avoid many un- during labor has already been mentioned; of
necessary operations, and that induction of course, this is an unavoidable accident that
labor, within the realm of correct indications, has been observed as occurring many times
is as safe as for cephalic presentations. In during second stage and requiring prompt
our material, enhancement was given in intervention. 4,5 Most of the other potential
14% of cases (Figure 6.4), while induction of mechanical problems should be avoidable
labor was started in 10% of the patients 18 when the obstetrician follows closely the
(Figures 6.1, 6.2, 6.3, and 6.5). various technical steps during the delivery
process. These are described in the standard
obstetrics textbooks,48 but some personal
Risks During Deli very variations by the author are based on the
observations about FHR changes previously
Vaginal described. The variations in technique
practiced in our institution consist of con-
There are very characteristic risks during tinuous FHR monitoring during second
the delivery process of the breech presenta- stage. If the decelerations trigger baseline
tion because ofthe mechanical relationships abnormalities and the delivery does not
of the fetus and birth canal. occur promptly (see Figures 6.2 and 6.4), or
78 L.A. Cibils

the decelerations are very prolonged, extrac- evitable occurrence during the passage of
tion should be accomplished (see Figure 6.5). the fetus through the birth canal. However,
The fundamental principle, to avoid iatro- the other difficulties or traumas, as described
genic complications, is that once the delivery in textbooks and publications, are more
process starts (whether assisted or by ex- likely preventable with adequate assess-
traction), it has to be a continuous, gentle, ment of fetus-pelvis relationship and skill-
smooth traction following all the mechanical ful handling of the delivery process. In breech
adaptations of the mechanism of breech presentation there is no room for a trial
delivery. First and foremost, it has to be of labor, as is permissible in cephalic presen-
started during a contraction and maternal tations with "borderline" pelvises. When
bearing-down effort, preferably aided by there is the slightest doubt that the fetus
gentle fundal pressure by an assistant. may be too large or "tight" for a given pelvic
What appears to be a similar technique has canal, a cesarean section should be carried
been described by Arulkumaran and co- out.
workers,81 who with it obtained neonates The classic method to evaluate the pelvis
in better condition than with the classical has been x-ray pelvimetry.19,20,48,79 This
"assisted" technique of waiting to assist until test has the added advantage of defining the
the scapulas are visible. fetal attitude and demonstrating the re-
A further variation in technique from lationship of fetal head and spine (ruling out
classical textbook teaching,48 at our institu- extension or hyperextension of the head), as
tion, is that the anterior shoulder is delivered well as excluding malformations, all very
first, followed by gentle rotation toward the important conditions that should be weighed
fetal back to bring the posterior shoulder when deciding about route of delivery. Others
below the pubis and deliver it as anterior. have proposed the use of computed tomo-
Piper's forceps are applied when the graphy, which would give information
shoulders have been delivered unless spon- similar to x-ray pelvimetry but with lesser
taneous delivery of the head occurred while amount of irradiation for the fetus and
getting ready for the application. It has been mother. 61 ,84 The concept currently more
shown by Milner82 that application offorceps widely accepted is that only frank attitudes
to the after-coming head is associated with should be considered for vaginal delivery.
significant reduction in neonatal mortality However, in a randomized study Gimovsky
among fetuses weighing between 1 and 3 kg. and coworkers19 reported no difference in
Pulling from the limbs or body without a neonatal outcomes among nonfrank breeches
contraction or appropriately applied gentle delivered either way. Likewise, others have
fundal pressure facilitates the extension of had no complications20 ,61 when allowing
the fetal arms, and thus this becomes an vaginal deliveries in these cases, an experi-
iatrogenic complication. A further compli- ence confirmed in our own institution. 18,27
cation of the extension of the arms occurs Hyperextension of the head, on the other
when the extended arm slides behind the hand, is a rather dangerous condition with
fetal neck, thus creating the "nuchal arm," a varied frequency, as high as 7%_14%,20,35
more complicated and mechanical dangerous which predisposes to severe spinal cord in-
problem requiring skill and gentleness for juries when vaginal delivery is accom-
an atraumatic solution. Sherer and colla- plished. 86 In an excellent review Abroms
borators83 reported a 4% incidence of nuchal and coworkers87 reported the extremely high
arm during first stage, a figure much too incidence of spinal cord transection (21%)
high for our experience. As stated, this is among those delivered vaginally, and no
essentially an iatrogenic and thus avoidable severe sequelae among those delivered
complication. 49 abdominally. It seems that the safest route
It was mentioned earlier that compression of delivery is the latter.20 On occasions the
of the umbilical cord is a mechanical, in- hyperextended position of the fetus near
6. Breech Presentation 79

term may already represent a permanent clavicles and humerus are also occasionally
scar of earlier damage to the spinal cord. 88 observed. 31 ,90
The very high incidence of congenital mal- Intracranial hemorrhage is particularly
formations is shown in Tables 6.1 through damaging because its sequelae are often in-
6.3; many of these are incompatible with capacitating.1t occurs more often in smaller
life. 85 This incidence is even higher among fetuses,85,91 but larger ones are not exempt
VLBW fetuses (see Table 6.3), making a from this risk. 31 The worst possible com-
diagnosis very desirable before deciding plication from mechanical difficulties with
on route of delivery. Less severe congenital the head is severe injury or severance of the
malformations are also very common, spinal cord92 with quadriplegia. Skull frac-
in particular among fetuses of higher tures have also been observed. 31
weight. 6 - 8,14,15,49 All these traumatic complications are
There are numerous reports of various probably the result either of hasty man-
types of mechanical difficulties or neonatal euvers to deliver the infant or of misjudg-
trauma following breech vaginal deliveries. ment of fetal size in allowing too large a
One of the mechanical accidents most fre- fetus to go through the birth canal. On rare
quently mentioned as occurring during occasions the progress of labor may be so
vaginal delivery is the so-called head advanced that vaginal delivery may be
entrapment,48,49 either resulting from a the only feasible alternative. 2o From the
small pelvis,85 or, more often, described foregoing one may surmise that the mech-
in small fetuses and ascribed to "cervical anics of breech delivery are more complicated
dystocia.,,38,85,89 This is a mislabeled diag- than for cephalic, at least for the obste-
nosis, because the cervix is made of con- trician. Furthermore, it seems reasonable to
nective tissue and thus is always completely state that the great majority of traumatic
thinned out, having minimal resistance even complications should be preventable if the
wheri dilating to allow passage of a pre- fetopelvic relationship is adequately diag-
mature fetus. What "traps" the fetal head is nosed and the delivery process conducted
the lowermost part of the corpus where it with the right technique. There are four
joins with the lower segment (the same area cases reported unofficially,93 and one com-
that makes the Bandl's ring), when it is pletely described,94 of abdominal deliveries
allowed to clamp on the fetal neck. When it of breech-presentation infants partially
happens, because the corpus is a powerful, exteriorized through the vagina with
thick muscle contracting, only good general minimal deleterious fetal effects. A good
anesthesia can relax it quickly to facilitate estimation of fetal size and pelvic capacity
an atraumatic delivery. There was only one should make unnecessary this very dramatic
"trapped head" in our material of more than situation.
370 cases of vaginal deliveries; that was
a VLBW fetus admitted already dead ante-
The Mother
partum. 18 It is probable that the combination
of good anesthesia and the coordinated con- The risks for the mother are those related to
tinuous action of obstetrician and patient (or potential lacerations and infections of the
assistant) contributed to the prevention of birth canal and uterus. For term-sized fetuses
this dreaded complication. it is usually advised to perform a medio-
Injury of the brachial plexus is usually the lateral episiotorriy which, if improperly
aftermath of difficulties to deliver the head done, may extend through the sphincter and
and subsequent strong pulling. It is reported rectum. 20 Others report only on overall
to have occurred in the majority of series morbidity of about 7%.49 This would include
discussing complications. * Fractures of postpartum endometritis, which occurred in
4% of patients in our obstetric population.
* References 20, 31, 49, 85, 89, 90. Occasionally the episiotomy wounds may
80 L.A. Cibils

break down. Uterine atony has also been abdominal deliveries, general anesthesia
described. 49 often may be necessary because of either the
urgency of the case, the patient's request, or
Cesarean Section the necessity of uterine relaxation for an
atraumatic delivery. The relative risk ofthis
The understanding of the mechanism of type of anesthesia in obstetrics, which may
breech delivery through an abdominal range from "severe difficulties,,20,49 and
incision is a mandatory requirement for any aspiration85 to maternal death,46 is well
obstetrician doing a cesarean section, as it is known. Among our cases, 54% received
if the birth would be accomplished by the general anesthesia and the remainder re-
vaginal route. This is necessary because, to ceived conduction. The most common com-
accomplish the delivery as atraumatically plication intrinsic to the technique of
as possible, the fetus has to be pulled out fol- cesarean section is the extension of the in-
lowing the appropriate maneuvers through cision to the uterine pedicles49 when the
a tight uterine incision, often made through transverse incision is used. This is the most
a thick muscular uterine wall. popular type of incision for term or near-
The Fetus term pregnancies, and when it extends
triggers important blood loss that may re-
The indication for an abdominal delivery quire transfusion. 20 ,49,61
is that, ostensibly, it will be much less Among all cesarean sections, in our
traumatic and potentially less damaging for material, 55% were given a transverse in-
the fetus than vaginal delivery. Unfortun- cision, nearly all for fetuses weighing more
ately that is not always the case, as perio- than 1500 g; there were 15 extensions but
dically there are reports of accidental none required transfusion. In certain cir-
lacerations made to the fetus during incision cumstances the tears may be so large or the
of the uterus. 61 Likewise, during the extrac- bleeding so difficult to control that hyster-
tion of the fetus, if the extraction maneuvers ectomy may be necessary. 49 It has also been
are not gentle and smooth fractures of lower reported that while attempting hemostasis
limbs may occur,4 and have been reported to ureteral injuries occurred. 49 Wound infec-
be either metaphyseal of femur and tibia tions are a known complication of cesarean
near the joint95 or in the middle of the section65 and have been reported in the
thigh20 ,96,97 (Figure 6.7). One such fracture majority of breech series. 20 ,49,61 The high
occurred in our series during an emergency incidence of endomyometritis and para-
operation of a term fetus admitted with metritis is a well-recognized post-cesarean
a prolapsed cord. The humerus may be frac- complication, ranging from 15% to 70% even
tured when the proper maneuvers to deliver when using prophylactic antibiotics. 20,49,61, 79
the arms are not made, that is, flexion of the A more dramatic infectious complication is
forearm at the elbow and not in the middle dehiscence of the wound. 2o,79 The severity or
of the arm. 90 Difficulties to deliver the head extension of some infections may require a
create the condition for brachial plexus in- hysterectomy for adequate treatment. 20 ,49
juries because of uncontrolled pull from the Secondary morbidity following blood trans-
body.2o,90 This may be particularly likely fusion has been observed,49 a complication
to occur during delivery of small fetses much feared now because of the potential
through transverse incisions on poorly de- transmission of viral infections by blood
veloped lower segments. products.
An often-overlooked long-term risk of
The Mother
cesarean section for the mother is the neces-
Unlike vaginal deliveries, with cesarian sity of a repeat operation in a subsequent
section there are greater chances of com- pregnancy. It is true that the recommenda-
plications for the mother. Although conduc- tion of a trial of labor is more and more
tion anesthesia is the method of choice for accepted if breech presentation has been the
6. Breech Presentation 81

FIGURE 6.7. Flat plate of newborn

at term, 3700 g, delivered by elec-
tive cesarean section. Midshaft
femoral fracture. Not evident, but
later observed on healing, was
another fracture, of distal meta-
physeal femur. Good healing
and function were noted at 9
months. (From Vasa R and Kim
MR,96 reprinted with permission
from The American Journal of
Perinatology 7:46-48, 1990,
Thieme Medical Publishers, Inc.)

indication. However, because of the trend to to breech presentation17 is not based on

operate on almost all breeches of less than clinical facts. It is, rather, the conclusion
2500 g, which constitute a very high percen- based on biased or incomplete analyses of
tage of them (53% in our population)18 and information passed from previous genera-
the requirement of a vertical incision for tions and the refusal to objectively review
safe delivery, the need of a repeat operation more current observations. A further, widely
in the future is already determined. Thus, stated, reason given for recommending
the potential morbidity of one or more future cesarean section for almost all breech presen-
pregnancies with a scarred uterus and sub- tations48 is that not enough cases occur to
sequent cesarean section should be con- give adequate training to house officers. 48 ,79
sidered in the equation. The same is said about forceps deliveries but
both are in reality fallacious arguments
because the cases that can be used to trans-
Management mit knowledge are, instead, managed with a
rapid cesarean section. At least, the im-
From the foregoing review of published pressively higher morbidity of this route
experience it seems clear that the gener- of delivery compared to vaginal is not
ally applied approach of cesarean delivery denied but, in spite of that, is rarely taken
82 L.A. Cibils

into consideration when undertaking the be engaged; and the patient must be co-
operation. operative and able to tolerate a ~-mimetic
The obstetrician managing a patient with substance to relax the uterus. In terms of
a fetus in breech presentation must consider equipment, it is extremely useful to have
several options. Can the abnormal presen- available a real-time ultrasound scanner to
tation be safely corrected? If the answer follow the fetal position (and its heartrate)
is in the affirmative, and the maneuver is during the maneuvers. In addition, con-
successful, that part of the obstetric problem tinuous FHR monitoring before and after
was solved. When the correction is un- the procedure should provide timely infor-
successful, then the decision must be made mation to either abort the version, if Ilb-
between abdominal and vaginal delivery. normal before it is completed, or to begin
These various options are now discussed. prompt intervention if the fetal heart rate
becomes and stays abnormal after com-
pletion. Often transient abnormalities will
Prophylaxis be observed. lOl ,102 Success has been re-
The ideal approach to offset the effect of. ported variously as between 48% and nearly
abnormal presentation on perinatal morbi- 100% for those attempting the version.*
dity (mother and fetus) is its prevention. Others report lowering of cesarean section
That premise being an unattainable goal, rate either among their total population, or
the next best thing to try is to minimize among those who were successfully con-
the number of abnormal presentations. In verted to cephalic as compared to those not
reference to breech presentation, attempts converted. lOl ,103-l05,107-l09 In any case, a
have been made for many years utilizing an substantial number of breech presentations
"external cephalic version." could be made cephalic if the procedure is
This maneuver, well described in all ob- done at the right moment as correlating
stetric textbooks, fell into disrepute in the with the appropriate technique. 99 Some
mid_1960s,98,99 when the popularity of factors have been mentioned as correlating
elective cesarean section was on the rise, with the rate of success, such as parity and
and continued in disfavor until relatively amniotic fluid volume l05 ,106,110,l1l or earlier
recently. A few isolated crusaders main- gestational age. 99 ,103 Conversely, failure
tained the art and continued to practice it rate has also been reported to be promoted
with great benefit to their patients. Among by placental location,103 specifically the
them, Ranney99 reported his experience cornual location for some. llO ,112 In our ex-
of several years with excellent results. Its perience relative oligohydramnios has been
popularity was slowly regained after a report a frequent reason to discontinue attempts at
in Europe, where the use of uterine relaxants version.
was predicated, by Saling and Mueller- There are some general requirements, as
Holve. lOo By the middle of the 1980s the have been indicated. However, there are few
version became a popular maneuver in the reports of series in which the version was
major perinatal centers but was seldom safely and successfully accomplished despite
used in so-called private practice settings. relative contraindications. Some have done
Actually, it is a relatively simple procedure versions with a high rate of success on uteri
requiring gentleness and patience from the with cesarean section scars,113 which is
obstetrician and a number of clinical pre- generally considered a contraindication.
requisites. If successful, the chances of a Others have accomplished versions without
return to breech presentation decrease with the use of tocolytics (uterine relaxant
the fetus closer to term. For the best chance drugS),99,114,115 an important element in the
of success, the following are required: the currently recommended technique.
membranes must be intact with normal
amniotic fluid volume; the breech must not * References 61, 98, 99,102-107.
6. Breech Presentation 83

Complications have been observed fol- than 2000 g, and that maternal morbidity is
lowing external version and adscribed to much higher following cesarean section. In
its execution. Compound presentation and spite of that, about 85% of all breeches in
dystocia 105 could be considered relatively this country are still delivered by cesarean
benign. Others of more serious consequence section. 17
have been observed in moderate-sized series From the perusal of the most recent pub-
or as isolated cases. Fetomaternal passage of lications comparing vaginal and abdominal
significant amounts of blood has been docu- deliveries referred in the preceding pages, it
mentedl02.116.117 and seems to be likely in seems that vaginal delivery of the breech-
view of the manipulations necessary over presenting fetus, regardless of weight, should
the placental implantation. In our institution be the first contemplated option. However,
a placental implantation over a large area of there are criticisms that attempt to discredit
the anterior aspect of the uterus is a con- the validity ofthose reports. In particular, it
traindication to version. Unexplained fetal is universally stated that those studies are
deaths 98 ,lo2 or placental abruptio with sub- not randomized and prospective. That is
sequent perinatal deaths have been observed quite correct, and the option should be to
shortly after external versions. 98 ,1l8,1l9 Di- undertake such an epidemiologically more
rect physical trauma to the fetus has also refined study. The problem arises when one
been reported, this in spite of the "protec- tries to execute such undertaking because
tion" provided by the many layers of tissues most obstetricians in this country would
and the amniotic fluid separating it from the change their current approach of systematic
hands of the operator. Bruising and brachial cesarean section for fetuses of less than
palsy have been produced 120 as well as 2000 g122 if others do a satisfactory study.
severe spinal cord injury.121 It is hard to Likewise, in the United Kingdom the conduct
escape the conclusion that in these cases the is very similar. 123 Numerous reasons (ex-
overzealous determination of the operator cuses) are given to justify the continuous
must have played a major role in the com- practice of cesarean sections in breech pre-
plication. A maternal fatality from amniotic sentations and, as seen in the previous pages,
fluid embolism has been reported. 102 none of these are of very solid medical value.
To recapitulate, the safe prophylaxis of The so-called medical-legal argument is dis-
breech presentation is possible in nearly cussed in Chapter 12 of this book. Inter-
two-thirds of cases provided the operator re- estingly, the conduct to subject the mothers
spects the general contraindications, follows to a cesarean section has never been proven
the proper technique, and acts with gentle- to be justified, but the majority of obste-
ness. 99 Furthermore, it is important to ob- tricians responsible for making the decision
serve the patients with FHR monitoring to are reluctant to carry out the prospec-
detect alterations, and to obtain Kleihauer- tive study they claim should clarify the
Betke tests to investigate possible feto- problem. 124 Penn 125 attempted to carry out
maternal bleeds and act accordingly. In the a multicenter randomized prospective study
case of an Rh-negative mother, prophy- of VLBW breech fetuses (25 large separate
lactic anti -D gamma globulin should be departments agreed with the protocol and
administered. were enrolled) but had to discontinue that
effort after 15 months because only 13 cases
had been recruited in the study.
Choice of Route of Delivery
It has been accepted by the majority of V 1D r
authors who reviewed their own material agina e Ivery
that there is no difference in the corrected When one evaluates a breech-presenting
mortality or morbidity for fetuses delivered pregnancy in labor, or near labor, it is es-
vaginally or abdominally ifthey weigh more sential to remember that there are absolute
84 L.A. Cibils

contraindications to vaginal delivery and randomized study of nonfrank breech pre-

when they are present it should not be tried. sentation, prolapse of the cord occurred in
In previous pages it was discussed that sus- 4% of trials of labor, and all cases were
pected or clear cephalopelvic disproportion delivered by cesarean section in good con-
is one such contraindication. Many authors dition. 19 In our own material, of more than
continue to rely on x-ray pelvimetry, as part 1140 cases, prolapse occurred in 7% of cases
oftheir protocols, to permit or contraindicate while frank presentations occurred in 47% of
a trial oflabor,* while others suggest the use that total. Among the last 840 cases there
of computed tomography to minimize ir- were 48% nonfrank presentations delivered
radiation exposure. 59,126 Magnetic resonance vaginally with corrected perinatal mortality
imaging (MRI) gives comparable measure- and morbidity not different from those deliv-
ments without ionizing radiation. 127 In our ered by cesarean section,18,27 even though
institution x-ray pelvimetry is not routinely there were 6% prolapses among completes
done. Instead, a thorough assessment offetal and 14% among footlings.
size and clinical pelvimetry are the deciding The use of physiologic doses of oxytocin to
factors. Hyperextension of the fetal head correct desultory labor resulting from ab-
should preclude labor and vaginal delivery. normal uterine contractility was discussed
An abdominal radiograph, which at the previously. Suffice here to say that there is
same time may rule out severe congenital no evidence that oxytocin has a deleterious
malformations, is necessary to correctly effect on the fetus, and clearly it is the
make this diagnosis. A prolapsed cord with method of choice to improve unsatisfactory
incompletely dilated cervix cannot be safely contractions pattern (see Figure 6.4). Pre-
delivered vaginally, and as in the other mature rupture of the membranes with
situations mentioned, mandates a cesarean unripe (unfavorable) cervix is also often
section. Insufficient skills of the person in mentioned as a contraindication to labor. If
charge to conduct labor and delivery is also one uses oxytocin to induce or enhance labor
an absolute contraindication to attempt a in cephalic presentation with this syndrome
vaginal delivery. The first three are not con- there is no rational basis for not doing the
trollable by the obstetrician and therefore same with breech presentation (see Figures
not remediable, and about the last one there 6.1 and 6.2). This should be understandable
are some further comments. because the uterine contractility pattern
There are also relative contraindications and efficiency are the same as in cephalic
to vaginal breech delivery. The presence of a presentations. 67 ,68
uterine scar has been one of them, but with As discussed previously, labor in breech
the current trend of encouraging vaginal presentation presents special characteristics
deliveries after previous cesarean section with regard to the fetus. Nearly 70% of these
this is a dwindling contraindication, and it fetuses will develop variable decelerations
has been shown to be safe for breech presen- during first stage,1O,72 and almost all during
tations. 53 ,113,128,129 The nonfrank breech second stage (Figures 6.1 through 6.5). It is
presentation has been considered as a critical to understand and distinguish cor-
mandatory indication for cesarean section rectly the decelerations that are benign from
even in centers where vaginal birth is at- those which represent progressive alteration
tempted for breech presentations. The of fetal homeostasis. Those will be, essen-
ostensible reason for this conduct is the tially, associated changes of the baseline
known higher incidence of cord prolapse expressed as tachycardia and fixed baseline
among them (4%-12% compared to 1 %-2%). (diminished or absent variability).73 When
However, in the only published prospective these alterations are observed, an assess-
ment of the stage of labor should be under-
taken by the clinician to evaluate the
* References 19, 20, 49, 50, 61. likelihood of rapid progress toward complete
6. Breech Presentation 85

dilatation and vaginal delivery (see Figure nique of delivery has been briefly described
6.2). If this progress is not prognosticated under the heading "Risks During Delivery,
as occurring before fetal deterioration, an Vaginal." Crawford129 observed that there
abdominal delivery should be indicated at were fewer cases of depression among the
once and carried out expeditiously. neonates born from labors under epidural
Intrinsic to the mechanics of delivery, anesthesia, and Darby and coworkers130
which entails compression of the umbilical observed better 5-min Apgar scores among
cord during passage of the body and head, is the same group. In our population of vaginal
the transient curtailment of oxygen supply deliveries, conduction anesthesia was usd
to the fetus for a variable period of time. only in 35%, but it is important to clarify
There is a concurrent accumulation of carbon that 18% of these deliveries were stillborns,
dioxide and fall of pH. In our material, the majority of the latter done under sys-
49.5% of breech cases showed an umbilical temic analgesia and local anesthesia.
artery pH of 7.20 or less compared to 36.5%
among cesarean deliveries, but at 5 min
there were no differences in Apgar scores. 18
Cesarean Section
This particular characteristic of the breech Only a few additional pertinent points are
delivery requires that continuous FHR needed to be made to complete the discussion
monitoring be maintained until delivery is given in previous pages under the heading
complete because unexpected sustained "Risks During delivery. Cesarean Section."
compression of the cord may occur at any It was stated there that an atraumatic
time during second stage, as the fetus des- delivery is the objective, and to accomplish
cends in the birth canal (see Figure 6.5). that it is essential to use a combination of
This situation constitutes an indication to appropriate anesthesia and gentle extrac-
proceed with extraction which, if properly tion. The age of gestation and the size of the
done, should not adversely affect the fetus. fetus are important determinants of how
An extraction was carried out in 15% of these are conducted.
vaginal deliveries in our cases. The method of anesthesia controls the pre-
A very important factor contributing to sence or absence of contraction during the
good labor and smooth delivery is adequate process of extracting the fetus. On the other
control of pain. Classically, pain control of hand, the presence and size, or absence, of a
labor in breech was carried out by means of lower segment determines the type of incision
systemic analgesia (natural or synthetic more favorable for an easy delivery, and this
opiates), in first stage, and by local anes- is directly related to gestational age as well
thesia, preferably pudendal block, for second as prior labor. In view of these different
stage. situations it is necessary to discuss the tech-
With the widespread use of conduction nique to apply under the circumstances.
analgesia-anesthesia in obstetrics in the
past 20 years, it was tried with excellent
Term Fetus
results for breech presentation. It was shown
that, although second stage may be moder- When one makes the decision to proceed
ately lengthened, the expulsive forces were with a cesarean section on a fetus at term,
not obliterated,129,13o and that the delivery but the mother is not in labor, it is necessary
maneuvers could be executed without haste to evaluate the development of the lower
and with the patient's cooperation under segment by a thorough pelvic examination.
controlled conditions. Furthermore, when Deep fornices and a nearly effaced cervix
there is good anesthesia during first stage indicate a good lower segment and the
one can proceed almost immediately with an possibility of an easy delivery under con-
abdominal delivery if the need arises because duction anesthesia, with a low transverse
of an unanticipated emergency. The tech- incision. The same could be said for patients
86 L.A. Cibils

in labor operated for fetal distress. However, uterus continues to contract at its own
it may be necessary to do a vertical incision rhythm with caudal, spinal, or epidural
to avoid fetal trauma to limbs if the pre- blockade.131-133 The substances used to
senting part is too deeply engaged. 96 ,97 The induce general anesthesia by "rapid se-
abdominal incision is part of the route to be quence," that is, thiopental, succinylcholine,
negotiated by the fetus and therefore it has or drugs of similar effect, do not relax the
as much importance as the uterine incision; human uterus at doses utilized for cesarean
its inadequacy may contribute to neonatal section. 134 Only the volatile anesthetics
trauma. S5 In our total population, transverse (nitrous oxide, halothane, fiuothane, pen-
incision of the uterus was used only in 54% thrane, etc.) will trigger uterine relaxation
of cases, but among the fetuses weighing after they reach the appropriate blood levels
more than 1500 g this incision was done in in about 3-5 min after their administration
65% of cases. Conversely, a classical incision starts. 134 The surgeon should wait until
or a vertical with extension into the corpus these drugs produce their relaxing action on
was done in only 12% of the group of fetuses the uterus before attempting to extract the
more than 1500 g, of which 29% (of the 554 fetus. In 54% of our cesarean sections (all
cases) were premature by weight. IS It is fetal weights), general anesthesia was used
highly likely that all these incisions were but it was given to 70% ofthe very low birth
made on the cases weighing less than 2000 g. weight (VLBW) group. The remaining 30%
Low vertical incisions were made on the of these received conduction anesthesia.
remainder, 23%. It appears that there is an intrinsic higher
risk of neonatal depression among breech-
presenting fetuses to have lower Apgar
The Very Low Birthweight Fetus
scores and more often need of intubation
These fetuses, according to the widely held than among cephalics, as it ws shown by
view, are those that most benefit from an Calvert for healthy term-size neonates. 135
abdominal "atraumatic" delivery. As shown In addition to the method of anesthesia
elsewhere in this chapter, from the review of used one must decide on the best type of
more recently published series,27,37-42,46 uterine incision to accomplish an easy and
that concept has not proven to be correct. smooth extraction. As was stated, this step
Every step should be taken to fulfil the pre- is crucial in handling these small infants,
mise of an atraumatic delivery. These are particularly if the presenting part is deep in
very fragile fetuses and therefore the the pelvis or the uterus is not well relaxed.
anesthesia as well as the uterine incision Westgren and Paul43 reported that for their
must be appropriate. When labor is con- VLBW population of less than 30 weeks
ducted under regional anesthesia and there gestation about 60% were low vertical, and
is a reasonably developed lower segment, it the remaining 40% equally divided between
is safe to proceed with cesarean section, classical and low transverse; for the same
if indicated, under the same type of anes- weights but 31 weeks gestation and older
thesia. However, for emergency situations about 50% were low transverse, 30% low
or an inadequately formed lower segment it vertical, and 20% still classical. In our
is necessary to operate under general material ofVLBW infants, regardless of ges-
anesthesia with good uterine relaxation. In tational age, only 18% were delivered by
other words, the obstetrician must wait a transverse incision, 42% a low vertical
until the uterus is well relaxed before pro- incision, and 40% by classical or low vertical
ceeding with the maneuvers to extract the with extension into the corpus.
fetus, as it is mandatory when conducting The combination of a relaxed uterus and
any intrauterine manipulation. an appropriate vertical incision is, in our
Conduction anesthesia is not satisfactory judgment, the only technique that permits a
under these circumstances because the truly atraumatic delivery. A contracting
6. Breech Presentation 87

uterus ("tight") and a low transverse incision cularly those receiving a classical or a low
are an invitation to a "head entrapment" as vertical incision with extension into the
has been described by some to occur even corpus, is the necessity to be subjected to
with term-sized breeches. 85 "Entrapment" of repeated cesarean sections in subsequent
the whole fetus has been described to occur pregnancies. Of course this entails facing
in VLBW infants delivered by low (vertical again the same intraoperatory risks just
and transverse) incisions, necessitating ex- described. In addition, scar dehiscences occur
tension into the cOrpUS. 43 For this to occur, in a number of post-cesarean pregnancies,
the mothers must not have been given anes- some of which may become complete ruptures
thesia with uterine relaxant substances as is if not operated by chance or diagnosed by
our recommendation. Likewise, a tight very astute clinicians. Another also very
uterus (contracting) or an insufficient tran- serious complication observed in pregnancies
sverse incision are direct contributors to following cesarean section, only recently
fractures observed in low birthweight (LBW) recognized, is the significantly increased
as well as term-size neonates. 95 - 97 It is im- risk of placenta previa, proportionally grow-
portant to reemphasize that the extraction ing very rapidly with each subsequent
maneuvers must be carried out with great operation.137-139 A relative rarity in ob-
gentleness and following the mechanics of stetrics, placenta accreta (even increta or
breech delivery. The described traumatic percreta), occurs with alarming frequency in
complications to the fetus may occur unless pregnancies evolving in uteri with a scar
the proper technique is followed. from previous cesarean section.138-142 The
As far as intraoperative maternal com- association of placenta previa with a scar
plications are concerned, they may be related from a previous cesarean section carries a
to the type of incision used on the uterus. risk as great as 67% of being a placenta
Extension and lacerations of the uterine accreta. Of course, this entity almost always
artery or vein and their branches, causing requires emergent hysterectomy.
severe blood loss and often the need of red
cell transfusion, have been reported with
transverse incisions. 49 ,136 Ureteral, bladder, Concl usions
and small bowel wounds have also been ob-
served. 49 ,136 Complications labeled "severe" The detailed review of the available liter-
have been observed significantly more often ature based on comparable cases of breech
when the operation is performed on preg- presentations managed either by cesarean
nancies of less than 32 weeks. 136 This is the section or vaginal delivery indicates that
group of patients for whom the current trend there is no clear benefit for the fetus to
recommends systematic cesarean section. be delivered abdominally. However, the
Significant blood loss, even without exten- morbidity facing the mother is manyfold
sions of the incision or lacerations, but re- higher and often of significant severity.
quiring blood replacement have also been These two statements appear to be valid for
observed with regularity.19,20,49,136 Non- all breech presentations, regardless of gesta-
fatal severe complications of anesthesia49 tional age. Thus labor, either spontaneous,
or even maternal death from complications enhanced, or induced, appears to offer the
of general anesthesia have been also re- most sensible medical approach to deliver a
ported. 46 The postoperative complications, breech presentation that could not be verted
particularly infections, have already been to cephalic.
mentioned (see pages 69 and 80). There is no Of course there should be stringent re-
need to repeat the risks involved in this quirements if the obstetrician wants to avoid
area. the potenial complications of an operation
One of the long-term risks for the mother for the mother and, at the same time, offer
who undergoes a cesarean section, parti- the infant a safe delivery. The contraindica-
88 L.A. Cibils

tions must be observed. Continuous FHR References

and uterine contractions (UC) monitoring
should be applied when labor is permitted. 1. Scheer K, Nubar J. Variation of fetal pre-
The tracings should be artifact free and con- sentation with gestational age. Am J Obstet
tinued until delivery of the head. Further- Gynecol 1976;125:269-270.
more, an obstetrician competent in the 2. Hill LM. Prevalence of breech presentation
interpretation of the tracings should follow by gestational age. Am J PerinatoI1990;7:
the case. Finally, the person who will conduct 92-93.
the delivery should be skilled in the con- 3. Luterkort M, Polberger S, Persson PH, et al.
duction of breech deliveries (either assisted Role of asphyxia and slow intrauterine
or complete extraction). Adequate pain relief growth in morbidity among breech delivered
infants. Early Hum Dev 1986;14:19-31.
by a competent anesthesiologist (or anes- 4. Rovinsky J, Miller JA, Kaplan S. Manage-
thetist) is very important for a smooth second ment of breech presentation at term. Am J
stage. Only with these prerequisites should Obstet Gynecol 1973;115:497-513.
labor and delivery be planned. 5. Brenner WE, Bruce RD, Hendricks CH. The
The argument that not enough trained characteristics and perils of breech presen-
obstetricians are available nowadays is in- tation. Am J Obstet GynecoI1974;118:700-
consistent with the conduct applied to other 712.
high-risk conditions. These cases (hyperten- 6. Kauppila O. The perinatal mortality in
sives, diabetics, premature labors, placenta breech deliveries and observations on af-
previas, intrauterine growth retardations, fecting factors. Acta Obstet Gynaecol Scand
etc.) are referred to Level III centers staffed, [suppI39],1975:1-78.
7. Fianu S. Fetal mortality and morbidity
ostensibly, by obstetricians "possessed of following breech delivery. Acta Obstet
special knowledge and ... special competence Gynaecol Scand [suppl 56], 1976:1-85.
in maternal-fetal medicine," who manage 8. Schutte MF, van Hemel OJS, van de Berg
them. It seems uncongruous that these C, et al. Perinatal mortality in breech pre-
specialists will not have the necessary sentations as compared to vertex presenta-
training to conduct a safe vaginal breech tions in singleton pregnancies: an analysis
delivery when they have been certified to based upon 57,819 computer-registered
possess such special knowledge. Persistent pregnancies in the Netherlands. Eur J
breech presentation should, therefore, be Obstet Gynecol Reprod BioI 1985;19:391-
referred to these centers as is any other 400.
high-risk pregnancy. If these obstetricians 9. Nissell P, Bistoletti P, Palme C. Preterm
breech delivery. Acta Obstet Gynaecol
have not acquired the necessary skills to Scand 1981;60:(363-366.
conduct a breech delivery, they owe to their 10. White PC, Cibils LA. Clinical significance
patients to undertake the extratraining of fetal heart rate patterns during labor:
needed to become proficient in it and thus VIII. Breech presentations. J Reprod Med
avoid for them the immediate and long-term 1984;29:45-51.
risks inherent to cesarean section, without 11. Luterkort M, Polberger S, WeIdner BM, et
benefit to their infants and done solely al. Growth in breech presentation. Ultra-
because of their insufficient skills. sound and post-partal assessment of growth
Not all breech presentations are amenable in 225 fetuses presenting by the breech in
to vaginal delivery, as are not all cephalic the 33rd gestational week. Acta Obstet
presentations. Cesarean section should be Gynecol Scand 1986;65:157-160.
12. Braun FHT, Jones KL, Smith DW. Breech
done only for the standard obstetric (fetal presentation as an indicator of fetal ab-
or maternal) indications. As for vaginal normality. J Pediatr 1975;86:419-421.
deliveries, skill in the mechanics of smooth 13. Soemes T, Bakke T. The length of the
atraumatic extraction is required, and the human umbilical cord in vertex and breech
judgment to give the most convenient in- presentations. Am J Obstet Gynecol 1986;
cision is essential. 154:1086-1087.
6. Breech Presentation 89

14. Clausen I, Nielsen TK. Breech position, 28. Piper EB, Bachman C. The prevention of
delivery route, and congenital hip disloca- fetal injuries in breech delivery. JAMA
tion. Acta Obstet Gynaecol Scand 1988;67: 1929;92:217 -221.
595-597. 29. Goethals TR. Cesarean section as the
15. Luterkort M, Persson PH, Polberger S, method of choice in management of breech
et al. Hip joint instability in breech preg- delivery. Am J Obstet Gynecol 1956;71:
nancy. Acta Paediatr Scand 1986;75:860- 536-549.
863. 30. Wright RC. Reduction of perinatal mortality
16. Nelson KB, Ellenberg JH. Antecedents of and morbidity in breech delivery through
cerebral palsy: multivariate analysis of routine use of Cesarean section. Obstet
risk. N Engl J Med 1986;315:81-86. Gynecol 1959;14:758-763.
17. Taffel SM, Placek PJ, Moien M, et al. 1989 31. Green JE, McLean F, Smith LP, et al. Has
U.S. cesarean section rate studies- VBAC an increased cesarean section rate for term
rate rises to nearly one in five. Birth 1991; breech delivery reduced the incidence of
18:73-77. birth asphyxia, trauma and death? Am J
18. Brown L, Karrison T, Cibils LA. Mode of Obstet Gynecol 1982;142:643-648.
delivery and perinatal results in breech 32. Kubli F, Boos W, Ruuttgers H. Cesarean
presentation. Am J Obstet Gynecol 1994; section in the management of singleton
171:28-34. breech presentation. 5th Eur Congr Perinat
19. Gimovsky MI, Wallace RL, Schifrin BS, Med Uppsala (Sweden), 1976:69-75.
Paul RH. Randomized management of the 33. Goldenberg RL, Nelson KG. The premature
nonfrank breech presentation at term: a breech. Am J Obstet Gynecol 1977;127:
preliminary report. Am J Obstet Gynecol 240-244.
1983;146:34-40. 34. Main DM, Main EK, Maurer MM. Cesarean
20. Tatum RK, Orr JW, Soong SJ, et al. Vaginal section versus vaginal deli very for the breech
breech delivery of selected infants weighing fetus weighing less than 1500 grams. Am J
more than 2000 grams. Am J Obstet Gynecol Obstet Gynecol 1983;146:580-584.
1985;152:145-155. 35. Tejani N, Verma U, Shiffman R, Chayen B.
21. Hall JE, Kohl SG, O'Brien F, et al. Breech Effect of route of delivery on periventricularl
presentation and perinatal mortality. Am J intraventricular hemorrhage in the low-
Obstet Gynecol 1965;91:665-677. birth-weight fetus with a breech presen-
22. Lewis BV, Seneviratne HR. Vaginal breech tation. J Reprod Med 1987;32:911-914.
delivery or cesarean section? Am J Obstet 36. Weissman A, Blazer S, Zimmer EZ, et al.
Gynecol 1979;134:615-618. Low birth weight breech infant: short term
23. Mann LI, Gallant JM. Modern management and long term outcome by method of de-
of the breech delivery. Am J Obstet Gynecol livery. Am J Perinatol1988;5:289-292.
1979;134:611-614. 37. Cox C, Kendall AC, Hommers M. Changed
24. Sachs BP, McCarthy BJ, Rubin G, et al. prognosis of breech - presenting low birth
Cesarean section: risks and benefits for weight infants. Br J Obstet Gynaecol 1982;
mother and fetus. JAMA 1983;250:2157- 89:881-886.
2159. 38. Effer SB, Saigal S, Rand C, et al. Effect of
25. Fortney JA, Higgins JA, Kennedy KI, et al. delivery method on outcomes in the VLBW
Delivery type and neonatal mortality breech infant: is the improved survival re-
among 10,749 breeches. Am J Public Health lated to C-section or other perinatal care
1986;76:980-986. maneuvers? Am J Obstet Gynecol 1983;
26. Kiely JL. Mode of delivery and neonatal 145:123-128.
death in 17,587 infants presenting by the 39. Bodmer B, Benjamin A, McLean FH, Usher
breech. Br J Obstet Gynaecol 1991;98:898- RH. Has use of Cesarean section reduced
904. the risks of delivery in the preterm breech
27. Cibils LA, Karrison T, Brown L. Factors presentation? Am J Obstet Gynecol 1986;
influencing neonatal outcomes in the very- 154:244-250.
low-birth-weight fetus «1500 grams) with 40. Davidson S, Cohen WR. Influence of presen-
a breech presentation. Am J Obstet Gynecol tation on neonatal outcome of vaginally
1994;171:35-42. delivered low birth weight infants. A
90 L.A. Cibils

matched pair analysis. J Perinatol 1990;10: term. An analysis of 304 breech deliveries.
38-42. J Perinat Med 1989;17:121-126.
41. Westgren LMZ, Songster G, Paul RH. Pre- 56. Bingham P, Hird V, Lilford RJ. Manage-
term breech delivery: another retrospective ment of the mature selected breech presen-
study. Obstet Gynecol1985;66:481-484. tation: an analysis based on the intended
42. Malloy MH, Onstad L, Wright E, et al. The method of delivery. Br J Obstet Gynaecol
effect of Cesarean delivery on birth outcome 1987;94:746-752.
in very low birth weight infants. Obstet 57. Westgren LM, Ingemarsson I. Breech de-
Gynecol1991;77:498-503. livery and mental handicap. Bailliere's Clin
43. Westgren M, Paul RH. Delivery of the low Obstet Gynaecol 1988;2:187-194.
birth weight infant by cesarean section. 58. Wisestanakora W, Herabutya Y,
Clin Obstet Gynecol1985;28:752-762. Prasertsawat P, Thanantaseth C. Fetal out-
44. Rosen M, Chik L. The effect of delivery come in term frank breech primapara de-
route on outcome in breech presentation. livered vaginally and by elective cesarean
Am J Obstet Gynecol 1984;148:909-913. section. J Med Assoc Thail 1990;73[suppl
45. Myers SA, Gleicher N. Breech delivery: why 1]:47-51.
the dilemma? Am J Obstet Gynecol 1987; 59. Christian SS, Brady K, Read JA, Kopelman
156:6-10. IN. Vaginal breech delivery: a five-year
46. Thiery M. Management of breech delivery. prospective evaluation of a protocol using
Eur J Obstet Gynecol Reprod Biol1987;24: computer tomographic pelvimetry. Am J
93-103. Obstet Gynecol 1990;163:848-855.
47. Faber-Nijholt R, Huisjes HJ, Towen BCL, 60. Roumen FJ, Luyben AG. Safety of term
Fidler VJ. Neurological followup of 281 vaginal breech delivery. Eur J Obstet
children born in breech presentation: a con- Gynecol Reprod Biol1991;40:171-177.
trolled study. Br Med J 1983;286:9-12. 61. Flanagan TA, Mulchahey KM, Korenbrot
48. Cunningham FG, MacDonald PC, Gant NF, CC, et al. Management of term breech pre-
Leveno KJ, Gilstrap LC III. Williams ob- sentation. Am J Obstet Gynecol 1987;156:
stetrics, 19th Ed. Norwalk: Appleton and 1492-1502.
Lange, 1993:493-520. 62. Croughan-Minihane MS, Petitti DB, Gordis
49. Collea JV, Chein C, Quilligan EJ. The L, Golditch I. Morbidity among breech
randomized management of term frank infants according to method of delivery.
breech presentation. Am J Obstet Gynecol Obstet Gynecol1990;75:821-825.
1980;137:235-244. 63. Otamiri G, Berg G, Ledin T, et al. Influence
50. Barlow K, Larsson G. Results of a five-year of elective cesarean section and breech de-
prospective study using a fetopelvic scoring livery on neonatal neurological condition.
system for term singleton breech delivery Early Hum Dev 1990;23:53-66.
after uncomplicated pregnancy. Acta Obstet 64. Hytten FE. Breech presentation: is it a bad
Gynecol Scand 1986;65:315-319. omen? Br J Obstet Gynaecol 1982;89:879-
51. Oian P, Skramm I, Hannisdal E, Bjoro K. 880.
Breech delivery. An obstetrical analysis. 65. DePalma RT, Cunningham FG, Leveno KJ,
Acta Obstet Gynecol Scand 1988;67:75-79. Roark ML. Continuing investigation of
52. Jaffa AJ, Peyser MR, Ballas S, Toaff R. women at high risk for infection following
Management of term breech presentation in cesarean delivery. Obstet Gynecol 1982;60:
primigravidae. Br J Obstet Gynaecol 1981; 53-59.
88:721-724. 66. Duff P. Pathophysiology and management
53. Ophir E, Oettinger M, Yagoda A, et al. of post-cesarean endometritis. Obstet
Breech presentation after cesarean section: Gynecol 1986;67:269-276.
always a section? Am J Obstet Gynecol 67. Arulkumaran S, Ingemarsson I, Gibb DM,
1989;161:25-28. Ratman SS. Uterine activity in spontaneous
54. Watson WJ, Benson WL. Vaginal delivery labor with breech presentation. Aust NZ J
for the selected frank breech infant at term. Obstet Gynaecol 1988;28:275-278.
Obstet Gynecol 1984;64:638-640. 68. Friedman EA. Clinical management of
55. Mecke H, Weisner D, Freys I, Semm K. labor, 2nd Ed. New York: Appleton-
Delivery of breech presentation infant at Century-Crofts, 1978:168-175.
6. Breech Presentation 91

69. Teteris NJ, Botschner A W, Ullery JC, Essig 83. Sherer DM, Menashe M, Palti Z, et al.
GF. Fetal heart rate during breech de- Radiologic evidence of nucal arm in the
livery. Am J Obstet Gynecol 1970;107: breech-presenting fetus at the onset of
762-765. labor: an indication for abdominal delivery.
70. Wheeler T, Greene K. Fetal heart rate Am J Perinatol 1989;6:353-355.
monitoring during breech labour. Br J 84. Kopelman IN, DuffP, Karl RT, et al. Com-
Obstet Gynaecol 1975;82:208-214. puted tomographic pelvimetry III the
71. Hill JG, Eliot BW, Campbell AJ, Pickett- evaluation of breech presentation. Obstet
Heaps AA. Intensive care of the fetus in Gynecol 1986;68:455-458.
breech labour. Br J Obstet Gynaecol 1976; 85. Gimovsky ML, Paul RH. Singleton breech
83:271-275. presentation in labor: experience in 1980.
72. Cibils LA. Fetal heart rate patterns during Am J Obstet Gynecol 1982;143:733-739.
labor in breech presentation. In: Proceedings 86. Bhagwanani SG, Price HV, Laurence KM,
of the XII World Congress of Gynecology Giuz B. Risks and prevention of cervical
and Obstetrics (Rio de Janeiro), Vol. 5. cord injury in the management of breech
London: Parthenon, 1989:Chap. 26. presentation with hyperextension of the
73. Cibils LA. Electronic fetal-maternal moni- fetal head. Am J Obstet Gynecol 1973;115:
toring. Antepartum, Intrapartum. Boston: 1159-1161.
Wright-PSG, 1981:341-368. 87. Abroms IF, Bresnan MJ, Zuckerman JE,
74. Eliot BW, Hill JG. Method of breech man- et al. Cervical cord injuries secondary to
agement incorporating use of fetal blood hyperextension of the head in breech pre-
sampling. Br Med J 1972;4:703-706. sentation. Obstet Gynecol 1973;41:369-
75. Brady K, Duff P, Read AJ, Harlass FE. 378.
Reliability of fetal buttock blood sampling 88. Cattamanchi GR, Tamaskar V, Egel RT,
in assessing the acid-base balance of the et al. Intrauterine quadriplegia associated
breech fetus. Obstet Gynecol 1989;74:886- with breech presentation and hyperexten-
888. sion offetal head: a case report. Am J Obstet
76. Luterkort M, Marsal K. Umbilical cord Gynecol 1981;140:831-833.
acid-base state and Apgar score in term 89. Gimovsky ML, Petrie RH. The intrapartum
breech neonates. Acta Obstet Gynecol and neonatal performance of the low birth
Scand 1987;66:57-60. weight vaginal breech delivery. J Reprod
77. Socol ML, Cohen L, Depp R, et al. Apgar Med 1982;27:451-454.
scores and umbilical cord arterial pH in the 90. Bistoletti P, Nisell H, Palme C, Lagercrantz
breech neonate. Int J Gynaecol Obstet 1988; H. Term breech delivery. Acta Obstet
27:37-43. Gynecol Scand 1981;60: 165-171.
78. Christian SS, Brady K. Cord-blood acid base 91. Cahill DJ, Turner MJ, Stronge JM. Breech
values in breech presenting infants born presentation: is a reduction in traumatic
vaginally. Obstet Gynecol 1991;78:778- intracranial hemorrhage feasible? J Obstet
781. Gynaecol 1991;11:417 -419.
79. Pauerstein CJ. Clinical obstetrics. New 92. Castillo M, Quencer RM, Green BA.
York: Wiley, 1987:511-524. Cervical spinal cord injury after traumatic
80. Gimovsky M, Petrie RH, Todd WD. Neonatal breech delivery. AJNR 1989;10:S99.
performance of the select term vaginal 93. Sandberg EC. The Zavanelli maneuver
breech delivery. Obstet Gynecol 1980;56: extended: progression of a revolutionary
687-691. concept. Am J Obstet Gynecol 1988;158:
81. Arulkumaran S, Thavarasah HS, Inge- 1347-1352.
marsson I, Ratnam SS. An alternative 94. Iffy L, Apuzzio JJ, Cohen-Addad N, et al.
approach to assisted vaginal breech de- Abdominal rescue after entrapment of the
livery. Asia Oceania J Obstet Gynaecol aftercoming head. Am J Obstet Gynecol
1989;15:47-51. 1986;154:623-624.
82. Milner RDG. Neonatal mortality of breech 95. Alexander J, Gregg JEM, Quinn MW.
deliveries with and without forceps to the Femoral fractures at cesarean section. Case
aftercoming head. Br J Obstet Gynaecol reports. Br J Obstet Gynaecol 1987;94:
1975;82:783-785. 273.
92 L.A. Cibils

96. Vasa R, Kim MR. Fracture of the femur at 111. Tan GW, Jen SW, Tan SL, Salmon YM. A
cesarean section: case report and review of prospective randomized controlled trial of
literature. Am J PerinatoI1990;7:46-48. external cephalic version comparing two
97. Barnes AD, Van Geem TA. Fractured femur methods of uterine tocolysis with a non-
of the newborn at Cesarean section. A case tocolytic group. Singapore Med J 1989;30:
report. J Reprod Med 1985; 30:203-205. 155-158.
98. Bradley-Watson PJ. The decreasing value 112. Hofmeyr GJ, Sadan 0, Myer IG, et al.
of external cephalic version in modern External cephalic version and spontaneous
obstetric practice. Am J Obstet Gynecol version rates: ethnic and other deter-
1975;123:237 -240. minants. Br J Obstet Gynaecol 1986;93:13-
99. Ranney B. The gentle art of external 16.
cephalic version. Am J Obstet Gynecol 113. Flamm BL, Fried MW, Lonky NM, Giles
1973;116:239-251. WS. External cephalic version after pre-
100. Saling E, Mueller-Holve W. External ce- vious cesarean section. Am J Obstet Gynecol
phalic version under tocolysis. J Perinat 1991;165:370-372.
Med 1975;3:115-122. 114. Robertson AW, Kopelman IN, Read JA, et
101. Rabinovici J, Barkai G, Shalev J, Majshiach al. External cephalic version at term: is a
S. Fetal heart rate changes following ex- tocolytic necessary? Obstet Gynecol 1987;
ternal cephalic version under tocolysis near 70:896-899.
term. lnt J Gynaecol Obstet 1987;25:277- 115. Van-Veelen AJ, Van-Cappellen HW, Flu
281. PK, et al. Effect of external cephalic version
102. Stine LE, Phelan JP, Wallace R, et al. Up- in late pregnancy on presentation at de-
date on external cephalic version performed livery: a randomized controlled trial. Br J
at term. Obstet Gynecol 1985;65:642-646. Obstet Gynaecol 1989;96:916-921.
103. Morrison JC, Myatt RE, Martin IN, et al. 116. Marcus RG, Crewe-Brown H, Kawitz S,
External cephalic version of the breech pre- Katz J. Feto-maternal hemorrhage follow-
sentation under tocolysis. Am J Obstet ing successful and unsuccessful attempts at
Gynecol 1986;154:900-903. external cephalic version. Br J Obstet
104. Dyson DC, Ferguson JE, Hensleigh P. Gynaecol 1975;82: 578-580.
Antepartum external cephalic version 117. Gjode P, Rasmussen K, Jorgensen J. Feto-
under tocolysis. Obstet Gynecol 1986;67: maternal bleeding during attempts at ex-
63-68. ternal version. Br J Obstet Gynaecol1980;
105. O'Grady JP, Veille JC, Holland RL, Burry 87:571-573.
KA. External cephalic version: a clinical 118. Kasule J, Chimbira THK, Brown I McL.
experience. J Perinat Med 1986;14:189- Controlled trial of external cephalic version.
196. Br J Obstet GynaecoI1985;92:14-18.
106. Hellstrom AC, Nilsson B, Stange L, Nylund 119. DeRosa J, Anderle LJ. External cephalic
L. When does external cephalic version version of term singleton breech presen-
succeed? Acta Obstet Gynecol Scand 1990; tation with tocolysis: a retrospective study
69:281-285. in a community hospital. J Am Osteopath
107. Hanss JW. The efficacy of external cephalic Assoc 1991;91:351-357.
version and its impact on the breech ex- 120. Petrikovsky BM, DeSilva HN, Fumia FD.
perience. Am J Obstet Gynecol 1990;162: Erb's palsy and fetal bruising after external
1459-1463. cephalic version: Case report. Am J Obstet
108. Donald WL, Barton JJ. Ultrasonography GynecoI1987;157:258.
and external cephalic version at term. Am J 121. Chapman GP, Weller RO, Normand ICS,
Obstet GynecoI1990;162:1542-1545. Gibbens D. Spinal cord transection in utero.
109. Mohamed K, Seeras R, Coulson R. External Br Med J 1978;2:398.
cephalic version at term. A randomized con- 122. Amon E, Sabai BM, Anderson GD. How
trolled trial using tocolysis. Br J Obstet perinatologists manage the problem of the
Gynaecol 1991;67:63-68. presenting breech. Am J Perinatol 1988;5:
110. Ferguson JE, Armstrong MA, Dyson DC. 247-250.
Maternal and fetal factors affecting success 123. Penn ZJ, Steer PJ. How obstetricians
of antepartum external cephalic versions. manage the problem of preterm delivery
Obstet GynecoI1987;70:722-725. with special reference to the preterm
6. Breech Presentation 93

breech. Br J Obstet Gynaecol 1991;98: Am J Obstet Gynecol 1965;93:907 -

531-534. 916.
124. Penn ZJ, Steer PJ. Reasons for declining 133. Matadial L, Cibils LA. The effect of epidural
participation in a prospective randomized anesthesia on uterine activity and blood
trial to determine the optimum mode of pressure. Am J Obstet Gynecol 1976; 125:
delivery of the preterm breech. Controlled 846-854.
Clin Trials 1990;11:226-231. 134. Cibils LA. Electronic fetal-maternal moni-
125. Penn ZJ. Personal communication, 1993. toring - antepartum, intrapartum. Boston:
126. Aronson D, Kier R. CT pelvimetry: the Wright-PSG, 1981:411-471.
foveae are not an accurate landmark for the 135. Calvert JP. Intrinsic hazard of breech pre-
level of the ischial spines. AJR 1991;156: sentation. Br Med J 1980;281:1319-1320.
527-530. 136. Nielsen TF, Hokegard KH. Cesarean section
127. van Loon AJ, Mantingh A, Thijn CJ, and intraoperative surgical complications.
Mooyaart EL. Pelvimetry by magnetic re- Acta Obstet Gynecol Scand 1984;63:103-
sonance imaging in breech presentation. 108.
Am J Obstet Gynecol 1990;163:1256- 137. Singh PM, Rodrigues C, Gupta AN.
1260. Placenta previa and previous cesarean
128. Sarno AP, Phelan JP, Ahn MD, Strong TH. section. Acta Obstet Gynecol Scand 1981;
Vaginal birth after cesarean delivery. Trial 60:367 -368.
oflabor in women with breech presentation. 138. Clark SL, Koonings PP, Phelan JP. Placenta
J Reprod Med 1989;34:831-833. previa/accreta and prior cesarean section.
129. Crawford JS. An appraisal of lumbar epi- Obstet Gynecol 1985;66:89-92.
dural blockade in patients with a singleton 139. Nielsen TF, Hagberg H, Ljunblad U.
fetus presenting by the breech. J Obstet Placenta previa and antepartum hemor-
Gynaecol Br Commonw 1974;81:867-872. rhage after previous cesarean section.
130. Darby S, Thornton CA, Hunter DJ. Ex- Gynecol Obstet Invest 1989;27:88-90.
tradural analgesia in labour when the breech 140. Read JA, Cotton DB, Miller Fe. Placenta
presents. Br J Obstet Gynaecol 1976;83:35- accreta: changing clinical aspects and out-
38. come. Obstet Gynecol 1980;56:31-34.
131. Cibils LA, Spackman TJ. Caudal analgesia 141. Fox H. Placenta accreta, 1945-1969. Obstet
in first stage of labor: effect on uterine Gynecol Surv 1972;27:475-490.
activity and cardiovascular system. Am J 142. Breen JL, Neubecker R, Gregori CA,
Obstet Gynecol1962;84:1042-1051. Franklin JE. Placenta accreta, increta,
132. Hopkins EL, Hendricks CH, Cibils LA. and percreta. Obstet Gynecol 1977;49:
Cerebrospinal fluid pressure in labor. 43-47.
7
Cesarean Section for Fetal Distress
DAVID A. MILLER AND RICHARD H. PAUL

Cesarean section is the major surgical pro- rate of the procedure had fallen to 10%.4
cedure most frequently performed in the Throughout this century, improvements in
United States, accounting for 966,000 anesthetic and blood transfusion techniques,
(23.5%) of the estimated 4,111,000 live births the development of antibiotics, and advances
in 1991. 1 Nearly 89,000 (9.2%) of these in medical therapy for maternal illnesses
operations were performed for the indication have further contributed to reduction in
of fetal distress. Thus, cesarean deliveries maternal mortality. By 1970, the maternal
for fetal distress represented 2.16% of all mortality rate for cesarean birth was 113.8
live births in 1991. In the history of cesarean per 100,000 deliveries,2 and estimates place
delivery, performance of the procedure in a the current cesarean-related maternal mor-
living patient solely in the interest of fetal tality rate as low as 22.3 per 100,000. 5
safety is a relatively recent phenomenon. The dramatic decline in the maternal risk
Delivery of a fetus through an incision in the of cesarean delivery has led to more liberal
maternal abdominal wall has been reported use of the procedure, including increased
throughout history. As early as 3000 B.C., acceptance of operative intervention for
Egyptian law required that a postmortem purely fetal indications. In fact, as maternal
cesarean be performed in all cases of mortality has continued to fall, the per-
maternal death late in pregnancy so that the centage of cesareans performed for the diag-
stillborn infant might be buried separately.2 nosis offetal distress has risen more than 10
Rarely, the operation was performed in an fold, from 0.8% in the mid-1930s6 to 9.2% in
effort to salvage a fetus following the death 1991. 1 Despite significant advances in fetal
of the mother. assessment, however, the ability to accur-
Before the late nineteenth century, stag- ately diagnose fetal distress remains
gering maternal mortality rates of 75%- controversial.
100% limited the use of cesarean in living
patients to rare cases in which all other
means of delivery had failed. Although iso- Fetal Heart Rate Auscultation
lated cases of maternal and infant survival
were reported as early as 1500 A.D.,3 it was Auscultation of the fetal heart, described as
not until the late 1800s that advances in early as the seventeenth century in Le
aseptic technique, routine employment of Goust's "Humani Foetus Historia,,,7 was
uterine sutures,' and transition from the first reported in Western medical literature
classical to the low transverse uterine inci- by MayorS in 1818. In 1822, Jean Alexandre
sion led to significant reduction in maternal Le Jumeau, Vicomte de Kergaradec,9 re-
mortality. By 1900, the maternal mortality ported his observations of fetal heart sounds

95
96 D.A. Miller and R.H. Paul

using Laennec's stethoscope, and proposed corporated into obstetric management. By

that auscultation of the fetal heart could be 1978, more than half of all labors were
useful in confirming pregnancy, diagnosing electronically monitored. 21 The National
multiple pregnancy, determining fetal posi- Center for Health Statistics reported that, in
tion, and judging the state of fetal health or 1990, EFM was used in 73% of the 4,110,563
disease by changes in strength and frequency total births in the United States. 22
of the heart tones. Later, Kennedy,lO
Schwartz,l1 Winckel,12 and others used
auscultation to describe fetal heartrate Fetal Asphyxia
(FHR) changes associated with umbilical
cord compression, head compression, and The objective of EFM is to identify the fetus
fetal distress. Kilian,13 in 1849, and in distress so that measures might be taken
Winckel,12 in 1893, proposed indications for in time to avert permanent damage or death.
forceps delivery based upon FHR abnor- The term fetal distress, although commonly
malities such as tachycardia, bradycardia, used in the modern practice of obstetrics,
"irregularity," and "impurity of tone." remains poorly defined. It has been described
Schwartz11 and Seitz14 speculated that FHR as "a condition in which fetal physiology is
changes were associated with changes in so altered as to make death or permanent
fetal oxygenation. Remarkably, these obser- injury a probability within a relatively short
vations were made using only the stethoscope period of time.,,23 Some authors base the
(mediate auscultation), or the ear of the definition upon FHR abnormalities,24,25
examiner placed directly upon the maternal while others require abnormal fetal blood
abdomen (immediate auscultation). It was gas values or low Apgar scores. Neverthe-
not until 1917 that Hillis described the less, the term is generally understood to
modified stethoscope known today as the denote disruption of normal fetal oxygena-
DeLee-Hillis fetoscope. 15,16 tion, ranging from mild hypoxia to profound
asphyxia. Hypoxia refers to the reduction of
oxygen supply to tissues below physiologic
Electronic Fetal Monitoring levels. Asphyxia, derived from the Greek
word asphuxia meaning "a stopping of the
In 1906, Cremer17 recorded the first fetal pulse," is defined as "a condition in which an
electrocardiograph (ECG), marking the extreme decrease in the amount of oxygen in
beginning of the era of electronic fetal moni- the body accompanied by an increase of
toring (EFM). By placing one electrode on carbon dioxide leads to loss of consciousness
the maternal abdomen above the fundus and or death,,,26 or as a "lack of oxygen in respired
another in the vagina, he was able to demon- air, resulting in impending or actual
strate small fetal electrical impulses among cessation of apparent life.,,27 Parer and
the higher voltage maternal signals. Despite Livingston28 defined fetal asphyxia as the
technologic improvements, the quality of "insufficiency or absence of exchange of the
abdominal fetal ECG tracings remains respiratory gases." The term is usually con-
unreliable, and the clinical usefulness of the sidered to imply the presence of metabolic
technique is limited. acidosis as well.
The concept of direct application of the Historically, the clinical diagnosis of birth
ECG electrode to the fetus in-utero was asphyxia has been based upon a variety of
introduced in the 1950s,18-20 with results findings, including meconium passage,
clearly superior to those obtained abdo- abnormal FHR patterns, low Apgar scores,
minally. The first practical clinical electronic abnormal blood gases, and neonatal neuro-
fetal monitor became available in the United logic abnormalities. When present together
States in 1968 and, throughout the 1970s, these findings are highly suggestive of a
fetal monitoring became increasingly in- birth-related asphyxial insult. Isolated
7. Cesarean Section for Fetal Distress 97

abnormal findings, however, correlate poorly intestmal (GI) tract include hypoxic-ischemic
with birth asphyxia and subsequent neuro- mucosal injury, stress ulcers, hepatic injury,
logic impairment. In 1989, Gilstrap et al. 29 and necrotizing enterocolitis. In the lungs,
recommended that the diagnosis of birth asphyxial sequelae include meconium
asphyxia be reserved for infants who are aspiration, persistent pulmonary hyperten-
severely depressed (5 min Apgar ::::;3) and sion, impaired surfactant production, and
acidotic (pH < 7.00) at birth, require resus- respiratory distress syndrome. Asphyxial
citation, and have seizures in the first day of renal injury may lead to renal insufficiency
life. or acute renal failure. Hematologic mani-
In 1991, the American College of Obste- festations include thrombocytopenia,
tricians and Gynecologists (ACOG) Com- neutropenia, and disseminated intravascular
mittee on Obstetrics, Maternal and Fetal coagulation. Clinical manifestations of as-
Medicine stated that "a neonate who has phyxia are frequently seen in the central
had severe hypoxia close to delivery that nervous system. Impaired neuronal water-
is severe enough to result in hypoxic en- regulatory mechanisms and disruption of
cephalopathy will show other evidence of the blood-brain barrier may cause cerebral
hypoxic damage including all of the fol- edema and neuronal necrosis. Resultant dis-
lowing: (1) a profound umbilical artery ruption of normal membrane depolarization,
metabolic or mixed acidemia (pH < 7.00), (2) neurotransmission, and receptor stimulation
persistence of an Apgar score of 0-3 for may lead to seizures and respiratory de-
longer than 5 min, (3) neonatal neurologic pression. Finally, the most widely publicized
sequelae, e.g., seizures, coma, hypotonia, and unequivocally feared consequence of
and (4) multiorgan system dysfunction, e.g., fetal asphyxia is cerebral palsy.
cardiovascular, gastrointestinal, hemato-
logic, pulmonary, or renal." They further
declared that "the term birth asphyxia is Asphyxia and Cerebral Palsy
imprecise and should not be used.,,30 How-
ever the old terminology is well entrenched, Cerebral palsy (CP) is "a chronic disability,
and continues to generate confusion and characterized by aberrant control of move-
debate. ment and posture, appearing early in life
and not the result of recognized progressive
disease.,,32 It may be accompanied by mental
Asphyxia at the Cellular Level retardation (41%) and seizures (23%)33 as
well as cortical visual impairment. Three
At the cellular level, asphyxia is thought to forms of the disorder are recognized: the
trigger a cascade of events, including mem- spastic syndromes (quadriplegia, hemiple-
brane depolarization, disruption of energy gia, and diplegia), ataxic syndromes (simple
metabolism, altered release and reuptake of ataxia and ataxic diplegia), and the dys-
neurotransmitters, ion shifts, protease kinetic syndromes (choreoathetosis, dys-
activation, phospholipid degradation, and tonic/tonus changing). Spastic diplegia and
free radical production. 31 It is clear that quadriplegia are associated with periven-
profound and prolonged asphyxia may lead tricular leukomalacia, seen most commonly
to cell death and, eventually, death of the in premature, low birthweight infants
organism. Levels of asphyxia that are sub- Choreoathetoid cerebral palsy is associated
lethal to the organism, however, may result with neonatal hyperbilirubinemia and
in clinical evidence of cellular dysfunction. kernicterus as well as hypoxic-ischemic
Complications of asphyxial myocardial in- injury to the basal ganglia. The insult
jury include cardiac conduction defects, responsible for the later development of CP
myocardial dysfunction, and congestive may occur at any time during the prenatal,
heart failure. Manifestations in the gastro perinatal, or postnatal periods. 34
98 D.A. Miller and R.H. Paul

Cerebral palsy is classified, along with maturity, multiple gestation, malpresenta-

mental retardation, learning disorders, tion, resuscitation requirement, hemolytic
autism, and epilepsy, as a major disorder of disease, and hypoxia caused by cord prolapse,
neurodevelopment. Unlike most other abruption, and preeclampsia. In the same
developmental disorders, the relationship year, Steer and Bonney41 studied the histo-
between CP and abnormal or difficult birth ries of 317 patients with CP and found that
has long been recognized. The degree of cor- 41 (13%) were attributable to kernicterus
relation, however, is controversial and is the and other neurologic disease. Of the remain-
subject of evolving understanding. ing 276 (87%) cases, there were 116 (43%)
In 1862, William John Little,35 an English with no historical findings suggestive of
orthopedic surgeon, presented to the Obste- anoxia, and 160 (57%) in whom the possibi-
trical Society of London his treatise "On the lity of anoxia existed. In 92 of the latter,
influence of abnormal parturition, difficult anoxia was presumed to have been present
labours, premature birth and asphyxia solely on the basis of neonatal incubator
neonatorum, on the mental and physical requirement. "Possible anoxia" was diag-
condition of the child, especially in relation nosed in the remaining 68 cases by a range
to deformities." The report reviewed the of criteria including severe toxemia, tight
birth histories of children with spastic nuchal cord, cord prolapse, and intrapartum
rigidity, and found a high incidence of pre- maternal death. In 1985, using data from
term delivery, breech presentation, pro- the Collaborative Perinatal Project, Nelson
longed labor, late onset of crying and and Ellenberg42 reported an increased in-
respiration, and neonatal convulsions and cidence of CP in low birthweight infants. In
stupor. Little concluded that infantile spastic term neonates, prolonged depression of
palsies could be caused by virtually nothing Apgar scores was also significantly associated
other than abnormalities ofthe birth process. with CPo
Schreiber,36 in 1938, reviewed the birth Data in animals further implicated fetal
records of 500 patients with cerebral sym- asphyxia as a cause of neurologic injury. In
ptoms, and noted a 70% incidence of birth 1943, Windle and Becker43 demonstrated
apnea. In 1951, Lilienfeld et a1. 37 ,38 reported clinical and histopathologic evidence of
higher incidences of placenta previa, mal- neural damage in experimentally asphy-
presentation, prematurity, and abruptio xiated fetal guinea pigs. Windle later studied
placenta in children with CP than in con- fetal rhesus monkeys44,45 and reported
trols. In 1955, Eastman and DeLeon39 clinical and histopathologic changes asso-
reviewed the obstetric records of 96 patients ciated with prolonged total anoxia, hyper-
with CP and noted that the immediate capnia, severe acidosis, and hypotension.
neonatal condition was described as "poor" Total anoxia for less than 8 min did not
(abnormal respiratory behavior, flaccidity, consistently result in neuronal injury,
cyanosis) in 41% of the cases compared to whereas anoxia for more than 10min
only 2% of controls. In addition, they reported invariably produced neuropathology. There
higher incidences of third-trimester bleed- were no survivors beyond 20-25 min of
ing, prematurity, breech delivery, midforceps anoxia. In his animal models, total anoxia
deli very, shoulder dystocia, prolonged second produced a pattern of neural necrosis in the
stage, fetal distress, prolonged neonatal brainstem, thalamus, and basal ganglia,
apnea, intrapartum maternal fever, and pro- with relative sparing of the cerebral cortex.
longed neonatal fever in infants later diag- These injuries manifested clinically as
nosed with CPo Of note, there were also seizures, ataxia, and athetosis. They could
significantly more congenital anomalies not account for the more common subtypes
(polydactyly, facial clefts) in the CP group. ofCP, which involve cerebral injury, mental
In 1962, Eastman et a1. 40 reviewed 753 retardation, and spasticity. In the late 1960s
cases of CP and found high rates of pre- and early 1970s, Myers 46 ,47 demonstrated
7. Cesarean Section for Fetal Distress 99

that, unlike total anoxia, prolonged partial oping CP (odds ratio, 2.84; 95% confidence
asphyxia in monkeys produced acidosis, late interval, 1.85-4.37). However, most infants
FHR decelerations, and neuropathologic with birth asphyxia did not develop CPo
defects consistent with the more common Furthermore, of 183 cases of CP, birth
clinical and histopathologic findings in CPo asphyxia was considered to be present in
In addition to lesions in the thalamus and only 13. They concluded that only 8.2% of
basal ganglia, prolonged partial asphyxia CP was potentially attributable to birth
resulted in generalized cerebral necrosis or asphyxia.
focal necrosis in the parasagittal regions The prevalence ofCP in school-age children
and the border zones between the parietal is approximately 2-2.5 per 1000, and has
and occipital lobes. not decreased appreciably over several
Although the conclusions of some of the decades. 53 Recently, in fact, the prevalence
early epidemiologic studies have been called ofCP has increased in Japan,54 Australia,55
into question,48-51 they created and fostered Finland,56 and the United Kingdom,57
the long-held assumption that birth-related primarily as a result of improved survival of
asphyxial insults were the primary cause of low birthweight infants at increased risk for
CPo In reality, these studies demonstrated CP.58 In the face of increasing perinatal
that perinatal asphyxia is one cause of CPo survival and improvements in prenatal and
More recently, attention has been focused intrapartum care, the lack of a significant
on the relative contributions of possible decline in the incidence of CP provides
prenatal factors, including congenital central additional evidence that the association
nervous system (CNS) abnormalities, infec- with adverse intrapartum events is weaker
tions, mercury toxicity, in-utero strokes, than originally believed. Although the cause
maternal hyperthyroidism, and maternal in most cases of CP is unknown, adverse
proteinuria. 51 ,52 Nelson and Ellenberg51 prenatal events appear to playa greater role
performed a multivariate analysis of risk in than previously recognized.
189 cases of cerebral palsy from the Col- Imaging modalities such as ultrasound,
laborative Perinatal Project. After correcting computed tomography, magnetic resonance
for major non-CNS congenital malfor- imaging, and technetium scanning provide
mations, birthweight of 2000 g or less, new insights into the prenatal origins of
microcephaly, and alternative explanations neurologic injury. Following an asphyxial
for CP, they reported that only 9% of all insult, neuronal necrosis produces charac-
cases were assoCiated with birth asphyxia teristic changes that evolve over the course
(defined as one or more of the following: of days to weeks. These changes are fre-
lowest FHR ::::;: 60 bpm, 5-min Apgar score ::::;: quently detectable with imaging techniques,
3, time to first cry ~ 5 min). and can aid in establishing the timing of the
In 1988, Blair and Stanley50 reported their injury. The location of the abnormality may
findings in 183 CP cases and 549 matched also playa role in the timing of the injury.
controls, reaching very similar conclusions. Peri ventricular leukomalacia is usually seen
The diagnosis of birth asphyxia was assigned with injuries between 28 and 34 weeks of
to all infants with "fetal distress" and a 1- gestation. On the other hand, parasagittal
min Apgar less than 7 or a spontaneous neuronal damage is generally a phenomenon
respiration time of more than 2 min. Fetal of term infants. Greater understanding of
distress was defined as any of the following: the etiologic factors involved in the develop-
(1) meconium, (2) FHR greater than 160 bpm ment of CP should help to rectify the pre-
or less than 120 bpm, (3) "abnormal" FHR vailing misconception of birth asphyxia as
tracing, or (4) documentation of "fetal dis- the sole cause of the disorder. This, in turn,
tress" not otherwise specified. Using these should lead to more realistic expectations of
criteria, they demonstrated that birth the possible benefits of intrapartum fetal
asphyxia nearly tripled the odds of devel- monitoring.
100 D.A. Miller and R.H. Paul

Asphyxia and FHR Apgar scores than those with decreased

variability.
Abnormalities In 1982, Clark et a1. 64 reported that FHR
accelerations of 15 bpm for 15 s in response
As early as the nineteenth century, resear- to fetal scalp stimulation predicted a scalp
chers using auscultation recognized that pH of 7.19 at least. Smith et a1. 65 in 1986
certain FHR patterns were associated with reported a similar relationship between fetal
poor perinatal outcome. Kennedy,lO in 1833, scalp pH and the FHR response to vibro-
related Bodson's description of fetal distress acoustic stimulation with an artificial larynx
in association with a FHR that exhibited applied to the maternal abdomen over the
"slowness of its return when a contraction is fetal head for 1 to 3 s. Of 30 fetuses with
passing on." In 1838, Schwartz59 recom- FHR accelerations in response to this sti-
mended frequent counting of the fetal heart mulus, all had scalp pH values of 7.25 or
tones in labor, and implicated "asphyxic higher. This and other work has helped to
intoxication" as a cause of alterations in shed some light on the complex relationship
their "individual normal frequency." Refer- between fetal biochemistry and the neuro-
ring to Schwartz's description ofthe relation- logic regulation of FHR.
ship between FHR decelerations and uterine
contractions, Gultekin-Zootzmann6o noted
that "in those cases in which the heart Electronic Fetal Monitoring
sounds returned slowly to their earlier
rhythm, or when the attenuations persisted Versus Traditional
or deteriorated during the pauses, the result Auscultation
would be a weak, moribund or dead fetus."
Seitz14 in 1903 described three progres- With EFM rapidly replacing the traditional
sively ominous stages of FHR deceleration. practice of intermittent intrapartum FHR
He attributed the first two stages to irritation auscultation, a series of nonrandomized
and paralysis of the vagal centers, and the studies66- 76 in the mid-1970s reported sig-
third to paralysis of all extracardiac nerve nificantly decreased perinatal mortality in
centers, concluding that it was possible to electronically monitored patients. These
detect early signs of compromise before the studies are often criticized on the basis of
fetus was actually in danger. nonrandomized design and nonconcurrent
The introduction of EFM and fetal scalp controls, citing rapidly improving neonatal
blood sampling in the 1960s provided addi- care and falling perinatal mortality rates as
tional tools for evaluating the fetus. In 1967, possible sources of bias. MacDonald and
Ron and Quilligan61 proposed a system for Grant 77 pointed out that, over the time period
classification of FHR decelerations, and in of these studies, hospitals not using EFM
1969 Kubli et a1. 62 demonstrated the rela- experienced rates of improvement in perina-
tionship between the fetal scalp pH and the tal outcome similar to those seen in hospitals
type and severity of FHR deceleration. They that were using EFM. Nevertheless, these
reported that fetuses with no decelerations, studies had the effect of validating the use of
early decelerations, or mild variable de- EFM.
celerations had average scalp pH values of In 1976, the first of a series of randomized
7.29 or higher, while those with severe controlled trials was published, comparing
variable or late decelerations had pH values EFM to intermittent auscultation of the
lower than 7.16. In 1975, Paul et a1. 63 found FRR during labor. To date, there have been
that, despite the presence of late decelera- nine such studies; five in high-risk popula-
tions, neonates with average FHR variability tions,24,78-81 two in low-risk populations,82,83
had significantly higher scalp pH values and and two in combined low- and high-risk
7. Cesarean Section for Fetal Distress 101

populations. 84 ,85 These trials are sum- the control group was diagnosed by the
marized in Table 7.1. presence of bradycardia to 100 bpm after
three or more consecutive contractions.
Delivery was effected if fetal distress was
Randomized Controlled Trials not relieved within 15 min. There were
of EFM Versus Auscultation no significant differences in outcome as
measured by perinatal mortality, Apgar
In 1976, Haverkamp and associates,78 in scores, cord blood pH values, neurologic signs
Denver, reported the first prospective, in the neonate, or neonatal nursery morbidity
randomized study of 483 high-risk obstetric between the EFM and control groups. The
patients, comparing EFM with intermittent monitored group, however, had significantly
FHR auscultation in labor. A point-rating higher rates of cesarean delivery overall
system86 was used to assess risk status, with (16.5% versus 6.8%) and of cesarean for fetal
a score of 6 or less indicating high risk. distress (7.4% versus 1.2%).
Patients in labor who had meconium-stained Questions have been raised concerning
amniotic fluid, required oxytocin, or had FHR the comparability of the two groups. For
abnormalities on admission were also in- instance, review of the monitor tracings
cluded. In the EFM group, a scalp electrode revealed a higher incidence of abnormal
was placed as soon as possible. Auscultation FHR patterns early in labor in the study
in the control group was performed every group. Furthermore, the study group had a
15 min in the first stage oflabor, and every higher incidence of maternal postpartum
5 min in the second stage, for 30 s after infectious morbidity (13.2% versus 4.6%),
uterine contractions. Electronic monitoring which was not explained by the increased
was employed in both groups, but was blinded rate of cesarean birth. These findings suggest
in the control group. that the study group may have represented
In the EFM group, FHR patterns were a higher risk population than did the control
evaluated using the criteria of Kubli and group and that effective randomization was
Hon. 62 In patients with late decelerations or not achieved.
severe variable decelerations that persisted The second study, by Renou et al. 79 in
after 15 min of corrective measures (oxygen, Melbourne, Australia, in 1976, randomized
positional changes, correction of hypoten- 350 high-risk patients into EFM and auscul-
sion), delivery was effected. Fetal distress in tation groups. High-risk patients were

TABLE 7.1. Prospective randomized clinical trials ofEFM versus intermittent FHR auscultation
N Risk Perinatal Neonatal Cesarean
Authors Year (total) status mortality neurologic signs section rate
Haverkamp et al. 78 1976 483 High oa 0
Renou et al. 79 1976 350 High 0 f
Kelso et al. 82 1978 504 Low 0 0
Haverkamp et al. 24 1979 690 High 0 0
Wood et al. 83 1981 989 Low 0 0 o
MacDonald et al. 84 1985 12,964 Combined 0 f o
Neldam et al. 85 1986 969 Combined 0 0 o
Luthy et al. 8O 1987 246 High (PTL) 0 0 o
Vintzileos et al. 81 1993 1,428 High f 0 o
aSymbols:
0= No difference.
f = Lower in EFM group.
e = Higher in EFM group.
102 D.A. Miller and R.H. Paul

defined as those with a poor obstetric history, tion was not utilized. The dip area88 was
a medical or obstetric complication, an used as a measure of fetal distress in the
abnormal FHR detected by auscultation, or EFM group; however, criteria for interven-
meconium in the amniotic fluid. Continuous tion were not specified. In the control group,
EFM was performed in the study group, and a FHR higher than 160 bpm or lower than
scalp pH was measured if the FHR tracing 120 bpm was considered indicative of fetal
was judged to be abnormal. Abnormalities distress. There were no significant differ-
were defined as a slowing of the FHR in ences between the groups with respect to
relation to the contraction cycle, a baseline perinatal mortality, low Apgar scores, cord
FHR less than 100bpm, or loss of normal blood pH values, NICU admissions or length
beat-to-beat variability.87 The protocol for of stay, neonatal or maternal infections, or
auscultation in the control group was not abnormal neonatal neurologic findings.
reported. Criteria for obstetric intervention There was a significantly higher cesarean
were not specified in either group. In this rate in the monitored group (9.5% versus
study, there were no significant differences 4.4%); however there was no difference in
between the groups with respect to perinatal the incidence of cesarean for fetal distress
mortality, Apgar scores, or maternal or (EFM, 1.6%; control, 1.2%).
neonatal infection. Patients in the monitored In 1979, Haverkamp and associates 24 pub-
group, however, had significantly higher lished a second randomized controlled trial
cord blood pH values and significantly fewer in high-risk patients that was similar in
neonatal intensive care unit (NICU) admis- design to the first. It included additional
sions and abnormal neonatal neurologic measures of infant status as well as the
findings. The cesarean rate was significantly option to perform fetal scalp pH determina-
higher in the monitored group than in the tion during labor. Blinded EFM in the con-
control group (22.3% versus 13.7%); however, trol group was not performed in this trial. A
the indications for intervention were not total of 690 high-risk patients were rando-
specified, making this difference difficult to mized into three groups. In the first group,
interpret. The authors commented that the fetal assessment during labor was ac-
difference in cesarean rates was not statisti- complished by intermittent auscultation;
cally significant after removal of 6 patients the second group received continuous EFM
in the monitored group who had had a alone, and the third group received con-
previous cesarean birth. The rationale for tinuous EFM with the option to measure
removing these patients on the basis oftheir scalp blood pH as needed. Risk assessment
previous operations is unclear. The rates of guidelines, auscultation protocols, and cri-
cesarean for fetal distress were not reported. teria for the diagnosis of fetal distress were
In 1978, Kelso et al. 82 in Sheffield, the same as in their previous study. Among
England, published the first randomized the three groups, there were no significant
controlled trial comparing EFM and inter- differences in perinatal mortality, Apgar
mittent auscultation in 504 low-risk patients. scores, cord blood pH values, maternal or
Women with risk factors such as multiple neonatal infectious morbidity, NICU admis-
gestation, breech presentation, hyperten- sions, or neonatal neurologic abnormalities.
sion, diabetes, and other medical and obste- A significant increase in the incidence of
tric complications were excluded from the cesarean birth was demonstrated in the group
study. Continuous EFM was employed in with EFM alone (EFM alone, 18%; EFM
study patients; a fetal scalp electrode was with the option to scalp sample, 11%;
placed as early as possible. Auscultation in auscultation, 6%). The option to perform scalp
the control group was performed at least sampling resulted in an intermediate
every 15 min for 1 min during and im- cesarean rate that was not significantly dif-
mediately following a contraction. Crossover ferent from either of the other groups. When
was not permitted, and scalp pH determina- analyzed together, electronically monitored
7. Cesarean Section for Fetal Distress 103

patients had a significantly higher rate of patients would be needed to demonstrate a

cesarean for fetal distress than did controls 50% reduction in the combined incidence of
(5.2% versus 0.43%). intrapartum stillbirths, neonatal deaths,
The fifth trial was published in 1981 by and neonatal seizures in survivors (power,
Wood et al. 83 in Melbourne, Australia. A 75%; p ~ 0.05). A trial of that size would
total of 989 low-risk patients (890 at one have a 50% chance of detecting a 50% reduc-
hospital and 99 at another) were randomized tion in the rate of neonatal seizures, alone.
to receive EFM or intermittent auscultation. Of the study participants, 22.5% were iden-
Women with previous preterm births, tified as high risk.
meconium-stained amniotic fluid, fetal Amniotomy was performed within 1 h of
tachycardia or bradycardia, renal disease, admission in all patients, and those with no
hypertension, diabetes, or other medical or amniotic fluid or with moderate to dense
obstetric complications were excluded from meconium were excluded from the study. It
participation. Monitored patients had is unclear whether later passage of meconi urn
placement of a fetal scalp electrode as early resulted in removal from the study of a
as possible. The protocol for auscultation in patient once she had been included.
the control group was not described. Scalp In the EFM group, a fetal scalp electrode
pH measurements were obtained as needed. was applied as early as possible, and scalp
Fetal distress was diagnosed as in the pre- pH measurements were used as needed.
vious study by Renou et al. 79,87 The criteria Criteria for evaluation of the FHR tracings
for operative intervention were not specified. were similar to those of Kubli and Hon. 62
There were no significant differences between Suspicious or ominous tracings were those
the groups with respect to perinatal mor- with marked tachycardia or bradycardia,
tality, Apgar scores, cord blood pH values, moderate tachycardia or bradycardia with
NICU admissions, or neonatal neurologic decreased variability, absent or minimal
abnormalities. In this study, cesarean rates variability, late decelerations, moderate to
were not significantly different between the severe variable decelerations, and FHR
groups (4% in the monitored group and 2% patterns that were difficult to interpret. In
in the auscultated group), although the the first stage of labor, a scalp pH measure-
overall rate of operative intervention (in- ment was performed if a suspicious or
cluding forceps) was significantly higher in ominous FHR pattern persisted for at least
the monitored group. Rates of cesarean 10 min. A scalp pH less than 7.20 was an
delivery for fetal distress were not reported. indication for delivery, regardless of the
It should be noted that the randomization FHR pattern. If the pH was between 7.20
process was compromised at the larger and 7.25 with a persistently suspicious or
study hospital, requiring subsequent data ominous FHR pattern, delivery was effected.
manipulation. If the scalp pH was higher than 7.25, but
In 1985, MacDonald et al.,84 in Dublin the tracing remained suspicious or ominous,
and Oxford, published a randomized con- the pH was repeated within 30-60min.
trolled trial comparing EFM with intermit- In the second stage of loabor, delivery was
tent FHR auscultation in 12,964 pregnancies. effected if FHR abnormalities persisted for
It was the first study to prospectively cal- at least 10 min.
culate the sample size needed to demonstrate In the control group, FHR auscultation
statistically significant differences between was performed every 15 min for 60 s in the
the groups. Before initiation of the study, first stage of labor, and between each con-
estimates were made of the anticipated fre- traction during the second stage. If the FHR
quencies of intrapartum stillbirths, neonatal was less than 100 bpm or more than 160 bpm
deaths, neonatal seizures in survivors, and during three contractions and could not be
other severe abnormal neurologic charac- corrected with conservative measures, the
teristics. They calculated that 13,000 scalp pH was measured and the patient was
104 D.A. Miller and R.H. Paul

managed as described previously. Blood mellitus. In the EFM group, monitoring

sampling was also performed at unspecified was initiated when the patients no longer
intervals in the control group when labor desired to ambulate. A scalp electrode was
exceeded 8 h. There were no significant dif- placed as soon as possible thereafter.
ferences between the groups with respect to In the control group, fetal heart tones were
perinatal mortality, low Apgar scores, auscultated twice an hour for at least 15 s at
neonatal trauma, resuscitation requirement, a cervical dilatation of 5 cm or less, every
NICU admissions, or infectious morbidity. 15 min from 5 cm until the second stage of
Among the 28 perinatal deaths, asphyxia labor, and for 30 s after each contraction or
was considered to be the primary cause in 7 at least every 5 min during the second stage.
cases in each group. There were significantly Scalp pH sampling was optional, and was
more cases of neonatal seizures and persis- performed only five times (EFM, three; con-
tent neurologic abnormalities (> 1 week) in trol, two). In the EFM group, intervention
the control group; however, no differences was considered if FHR abnormalities re-
with respect to neurologic abnormality mained unresolved after 15 min of corrective
remained at I-year and 4-year follow-up (3 measures. Abnormalities included brady-
cases in each group). Labor was significantly cardia «120 bpm) , tachycardia (> 160 bpm),
shorter in the EFM group, and analgesia late decelerations, variable decelerations
(meperidine) was required less often. Scalp (not further specified), silent FHR pattern
sampling was used more frequently in the (beat-to-beat variability ::::;5 bpm) , and
EFM group (4.4% versus 3.5%), and twice as saltatory pattern (variability ;;::.25 bpm).
many fetuses with low scalp pH «7.20) were Intervention was considered in the control
identified. group if the FHR was less than 100bpm
The cesarean rate in the EFM group (2.4%) following three or more consecutive contrac-
was not significantly different from that in tions. No statistical differences were detected
the auscultated group (2.2%). Overall rates between the groups with respect to perinatal
of operative delivery were higher in the mortality, low Apgar scores, seizures, NICU
EFM group (10.6% versus 8.5%), as a result admissions or length of stay. Significantly
of a higher incidence of forceps delivery more pathologic FHR patterns were detected
(8.2% versus 6.3%). Rates of cesarean in the EFM group; however, there was no
delivery for fetal distress were not signifi- difference in the incidence of cesarean
cantly different (EFM, 0.4%; control, 0.2%). delivery between the groups.
The overall frequency of seizures in the The eighth study, by Luthy et al. so in
patients who were classified as high risk was Seattle (Washington) and Vancouver (British
4.3 per 1000. This was significantly higher Columbia) in 1987, compared EFM and
than in the low-risk group (2.6 per 1000). auscultation in 246 high-risk patients with
The incidence of seizures in surviving preterm labor. Inclusion criteria were pre-
neonates was the same in both groups (2.3/ term labor, singleton gestation, cephalic
1000). Electronic monitoring did not reduce presentation, estimated gestational age of
the seizure incidence to a greater extent in 26-32 weeks, and estimated fetal weight of
high-risk patients than in low-risk patients. 700-1750 g. Patients with preterm pre-
In this study, the largest to date, EFM was mature rupture of the membranes were not
associated with no increase in maternal excluded. In the EFM group, external moni-
morbidity. toring was used until advanced cervical
In 1986, Neldam and associatesS5 in dilatation (7 cm), at which time amniotomy
Copenhagen, Denmark, reported a ran- was performed and a scalp electrode was
domized controlled trial of EFM versus placed. In those with ruptured membranes,
intermittent auscultation in 969 combined a scalp electrode was placed once delivery
low- and high-risk patients. The study ex- was inevitable. Ominous FHR patterns were
cluded those with non-gestational diabetes those with persistent late decelerations with
7. Cesarean Section for Fetal Distress 105

at least three successive contractions in the was ended after the third review in light of a
absence of correctable cause, FHR greater statistically significant fivefold decrease in
than 180 bpm with total loss of variability perinatal mortality in the EFM group.
persisting more than 15 min, FHR less than All subjects had a singleton living fetus
100 bpm for more than 3 min, or severe with an estimated gestational age of 26
variable decelerations persisting for more weeks at least. Fetuses with known con-
than 30 min. An ominous FHR pattern genital or chromosomal anomalies were
lasting more than 30 min or a scalp pH less excluded.
than 7.20 was an indication for delivery. In the EFM group, external monitoring
In the control group, auscultation was per- was used as long as satisfactory tracings
formed for at least 30 s, at least every 15 min were obtained. Scalp electrodes were placed
in the first stage of labor, and at least every as needed. In the control group, FHR auscul-
5 min in the second stage. Ominous patterns tation was performed every 15 min during
were those with a FHR less than 100 bpm for the first stage oflabor and every 5 min during
more than 30 s after three or more conse- the second stage. The FHR was counted
cutive contractions, or baseline FHR greater during contractions and for at least 30 s
than 180 bpm for more than 15 min or less immediately afterward. N onreassuring
than 100 bpm for more than 60 s. Scalp pH patterns in the EFM group included late
was used as clinically indicated in both decelerations, prolonged decelerations to
groups. Fetal scalp pH values below 7.20 or 80 bpm or lower for more than 2 min, severe
ominous FHR patterns in the absence of a variables to 70 bpm or lower for 60 s or more,
correctable cause were considered indications variable decelerations with a rising baseline
for delivery. The groups did not differ with and loss of variability, tachycardia with
respect to the use oftocolytics, corticosteroids, decreased variability «5 bpm), persistent
oxytocin, or regional anesthesia. There were decreased variability, or a sinusoidal pat-
no differences in perinatal mortality, low tern. In the auscultated group, nonreassuring
Apgar scores, cord pH values, neonatal patterns included a FHR less than 100bpm
seizures, respiratory distress syndrome, or during and immediately after a contraction,
intracranial hemorrhage. Cesarean rates and a persistent FHR less than 100bpm or
were similar (EFM, 15.6%; controls, 15.2%). more than 160bpm. Scalp sampling was not
There was no difference in the incidence of used in either group, and crossover was not
cesarean for fetal distress (EFM, 8.2%; permitted.
controls, 5.6%). In both groups, delivery was effected if
The most recent randomized trial, pub- nonreassuring FHR patterns failed to resolve
lished in 1993 by Vintzileos et al.,S1 was after 20 min of conservative measures.
conducted in Athens, Greece and compared There were significantly fewer perinatal
EFM and intermittent auscultation in 1428 deaths in the EFM group (2.6/1000 versus
patients in a population with a high baseline 13/1000). Furthermore, no cases of hypoxia-
perinatal mortality rate (20.4-22.6 per related perinatal death occurred in the EFM
1000). The relatively high incidence of the group, whereas six cases occurred in the
outcome measure to be studied (perinatal auscultated group (0.9%). This difference
death) markedly improved the likelihood of was also statistically significant. The groups
detecting a statistically significant effect of did not differ significantly with respect to
EFM. Using an average incidence of 21 low Apgar scores, NICU admissions or length
perinatal deaths per 1000, they prospectively of stay, ventilator requirements, neonatal
calculated that a sample of 2210 patients hypoxic-ischemic encephalopathy, intraven-
would have an 80% chance of detecting a tricular hemorrhage, seizures, hypotonia,
67% reduction in perinatal mortality at the necrotizing enterocolitis, or respiratory dis-
5% level of significance. Reviews were tress syndrome. Although the incidence of
conducted every 3 months, and the study cesarean for fetal distress was significantly
106 D.A. Miller and R.H. Paul

higher in the EFM group (5.3% versus 2.3%), approximately 10% of CP is attributable to
the overall incidence of cesarean birth was birth asphyxia, the anticipated incidence of
not significantly different between the EFM asphyxia-related CP is roughly 0.2 per 1000.
and control groups (9.5% versus 8.6%). A study large enough to detect a 50% reduc-
tion in the incidence of asphyxia-related CP
(power, 80%; p ~ 0.05) would require more
Potential Benefits of Electronic than 500,000 patients. 89 Thus, it is not sur-
Fetal Monitoring prising that the randomized trials to date
have failed to detect a statistically sig-
When EFM was introduced in the 1960s, nificant reduction in CP with the use of
proponents anticipated marked reductions EFM.
in perinatal mortality and neonatal neuro-
logic injury. Regarding the former, eight of
nine randomized clinical trials conducted
during the past 20 years have failed to detect Potential Risks of Electronic
a statistically significant reduction in Fetal Monitoring
perinatal mortality with EFM, compared to
traditional intermittent FHR auscultation Early concerns regarding the potential for
(see Table 7.1). However, it is crucial to maternal or neonatal infections in electroni-
point out that only one randomized trial81 cally monitored patients have proven to be
had sufficient statistical power to demon- unfounded. Only one study78 demonstrated
strate such a difference. In that trial, an increased risk of maternal infectious
electronically monitored patients had a morbidity in patients randomized to EFM.
statistically significant fivefold improvement These results are very difficult to interpret
in perinatal mortality compared to those in light of the fact that fetal scalp electrodes
monitored with intermittent auscultation. were used in both the EFM and control
In comparison to the high perinatal mortality groups (FHR tracings were recorded in the
rates (20.4-22.6/1000) in the study by control group, but clinicians did not have
Vintzileos,81 MacDonald et al. 84 calculated access to them). The largest randomized trial
the combined anticipated frequencies of to date84 revealed no increased infectious
intrapartum stillbirths and neonatal deaths morbidity in electronically monitored
in their population to be 3 per 1000. In such patients. Current evidence does not support
a population, a study with an 80% likelihood an association between EFM and increased
of detecting a 50% reduction in perinatal infectious morbidity.
mortality (p < 0.05), would require more Current data suggest that the effect of
than 33,000 patients. 89 The total number of EFM on cesarean delivery rates is minimal.
patients in all nine studies, combined, was While four early randomized trials24 ,78,79,82
18,623. reported significantly more cesarean de-
With regard to neonatal neurologic injury, liveries in electronically monitored patients,
the problem of study size was the same. the five most recent studies8o- 85 have shown
Seven of the nine trials24,78,8o-83,85 showed no such difference. Furthermore, as the use
no beneficial influence of EFM (see Table ofEFM in the United States has expanded to
7.1). Two trials79 ,84 reported fewer neonatal include nearly three-fourths of all births,
seizures in the EFM groups. The only the incidence of cesarean delivery for fetal
study to examine long-term neurodevelop- distress has increased by only 1.8%, from
ment,84,90 however, found no difference 0.36% in 19742 to 2.16% in 1991.1 This
between the groups in the incidence of increase reflects less than 15% of the total
neurologic abnormality at 1 or 4 years of rise in the rate of cesarean birth. Since the
age. Assuming that the school-age incidence introduction of EFM, 20 years of research
of CP is approximately 2 per 1000, and that and clinical experience have refined the
7. Cesarean Section for Fetal Distress 107

interpretation of FHR patterns. Additional the FHR, and (3) periodic patterns that are
tools such as fetal scalp pH determination, classified according to their temporal
fetal scalp stimulation, and vibroacoustic relationship to uterine contractions.
stimulation help to confirm fetal well-being
in the presence of nonreassuring tracings
that previously may have prompted operative Baseline Fetal Heart Rate
intervention. Logic dictates that better
understanding of the capabilities and The normal FHR baseline ranges from 120
limitations of EFM should lead to better to 160 bpm. It is commonly closer to 160 bpm
perinatal outcome and fewer unnecessary in early pregnancy, declining as gestational
cesareans. Although this contention appears age advances. Likewise, the FHR may
to be borne out in the literature, the debate gradually decrease toward 120 bpm during
will no doubt continue. the course of labor. In general, a FHR base-
Despite the lack of consensus on many line below 120 bpm is termed bradycardia,
points, EFM has been clearly shown to be at and a rate in excess of 160 bpm is termed
least as effective in identifying fetal com- tachycardia. Abnormalities in the FHR
promise as is the practice of frequent FHR baseline may have very different causes and
auscultation with intensive, one-on-one consequences. It is important, therefore, to
nursing. While this level of individualized characterize the underlying etiology as
nursing care may be available in some set- accurately as possible and to institute
tings, most delivery units will find the per- appropriate therapy at the earliest possible
sonnel requirements to be impractical and time.
cost prohibitive.
Bradycardia
Bradycardia is defined as an abnormally low
Management of Intrapartum baseline FHR «120 bpm) , and must be
Fetal Distress differentiated from the abrupt FHR changes
characteristic of decelerations. Although
Timely diagnosis and appropriate manage- FHR decelerations are very common, true
ment of intrapartum fetal distress remain fetal bradycardia is not. A bradycardic FHR
among the most challenging tasks facing the baseline between 100 bpm and 120 bpm in
obstetrician. During the intrapartum period, association with otherwise reassuring FHR
uterine contractions subject the fetus and patterns most likely represents a normal
the uteroplacental exchange unit to inter- variant. Rarely, fetal bradycardia may be
mittent episodes of diminished maternal seen in association with maternal beta-
blood flow. These brief interruptions in blocker therapy, hypothermia, hypogly-
oxygen delivery are usually well tolerated cemia, hypothyroidism, or fetal cardiac
by the healthy fetus; however, repetitive or conduction defects (Le., congenital atrioven-
prolonged hypoxic stress may overwhelm tricular block).
the fetal compensatory mechanisms and Documentation of fetal heart block should
lead to acidosis and asphyxia. In such cases, prompt a search for structural fetal cardiac
the fetus will usually exhibit FHR patterns abnormalities, which are present in 20% of
indicative of stress or distress. Clear under- cases. Other possible causes of heart block
standing of the pathophysiology of these include viral infections (i.e., cytomegalo-
patterns is essential to optimize fetal out- virus) and maternal systemic lupus erythe-
come and minimize unnecessary operative matosis with anti-Ro (SSA) antibodies. Most
deliveries. congenital causes of fetal bradycardia do not
Interpretation of the FHR tracing is based present as acute changes in the FHR and
upon (1) the baseline FHR, (2) variability in rarely require emergency intervention. Any
108 D.A. Miller and R.H. Paul

abrupt decline in the FHR to less than secondary to severe fetal anemia. It may
120 bpm more likely represents a decelera- also occur in association with amnionitis,91
tion than a change in the baseline, and fetal sepsis, or administration of narcotic
should be considered pathologic until proven analgesics (Stadol, Demerol, Nisentil). A
otherwise. persistent sinusoidal pattern that is not
attributable to medications is a concerning
finding. Labor should be allowed to continue
Tachycardia only in the presence of a normal fetal scalp
Fetal tachycardia has many possible etiol- pH or other FHR evidence of fetal well-
ogies. Most commonly, it is the result of being.
decreased vagal or increased sympathetic
outflow, associated with fever, infection, fetal
anemia, or fetal hypoxia. Other possible Fetal Heart Rate Variability
causes include maternal hyperthyroidism,
sympathomimetic medications (ritodrine, Variability in the FHR results from constant
terbutaline), parasympatholytic medications interplay between the sympathetic and
(atropine, phenothiazines), and fetal cardiac parasympathetic arms of the fetal autonomic
arrhythmias. The underlying etiology should nervous system. Modulation of vagal tone
be identified and treated, when possible. occurs in response to blood pressure changes
The source of any maternal fever must be detected by aortic arch baroreceptors. Oxygen
aggressively sought. Specifically, intra- and carbon dioxide fluctuations, detected by
amniotic infection must be ruled out. The chemoreceptors, similarly influence vagal
diagnosis of chorioamnionitis requires outflow. Continual adjustments in vagal
intrapartum antibiotic therapy. Possible tone are manifested in the FHR tracing as
causative medications should be discon- "short-term" ("beat-to-beat") variability
tinued, and maternal hyperthyroidism superimposed upon broader, cyclical (3-5
should be excluded. Fetal cardiac arrhy- cycles/min) "long-term" variability. In clini-
thmias may require ultrasonographic cal use, the term "FHR variability" refers to
evaluation to rule out structural lesions and a composite of the two.
cardiac failure. Antiarrhythmic therapy Average FHR variability (6-25bpm)
may be instituted if deemed necessary. represents a normally oxygenated vagal
Although tachycardia alone does not necess- connection between the fetal central nervous
arily indicate fetal ·distress, it is commonly system (CNS) and the cardiac conduction
seen in association with other FHR patterns system. Increased variability (>25 bpm) , or
suggestive of hypoxia, including loss of "saltatory" FHR pattern, is uncommon and
variability and repetitive late decelerations. most often represents an exuberant autono-
In such cases, consideration should be given mic response of a normal fetus. On occasion,
to fetal scalp pH determination or delivery. it may reflect increased catecholamine
release in the early stages of fetal hypoxia.
Careful evaluation of the associated FHR
Sinusoidal Pattern findings should help to clarify such cases.
The sinusoidal FHR pattern is an uncommon Decreased variability (0-5 bpm) most often
FHR baseline abnormality. As the name reflects decreased fetal CNS activity asso-
implies, it is sinusoidal in appearance, with ciated with fetal sleep states, fetal ano-
an amplitude of 5-15 bpm and a frequency malies, medications (analgesics, magnesium
of 2-5 cycles per minute. Variability is sulfate, benzodiazepines, phenothiazines), or
usually decreased, ·and accelerations are fetal acidosis. Parasympatholytic medica-
absent. Although the pathophysiologic tions (atropine) may decrease FHR variabi-
mechanism is not known, this pattern is lity by blocking vagal influence on the
classically associated with fetal hypoxia sinoatrial node. Persistent decreased vari-
7. Cesarean Section for Fetal Distress 109

ability in a nonanomalous fetus (not attri- oxygenated, nonacidotic eNS-cardiac axis.

butable to medications or fetal sleep state) is The persistent absence of spontaneous
an ominous sign, particularly when asso- accelerations, on the other hand, is abnormal
ciated with other FHR patterns suggestive and may reflect fetal compromise. A non-
offetal hypoxia. In such cases, demonstration reactive FHR tracing must be carefully
of fetal well-being is desirable if labor is to interpreted in the context of the clinical
be permitted to continue. presentation and other FHR characteristics.
Reassurance may be obtained by several During labor, the frequency and amplitude
different means, including fetal scalp pH of FHR accelerations may be diminished by
determination. A fetal scalp pH of 7.25 or fetal sleep states, commonly used medica-
higher provides evidence that the fetus is tions (narcotics, MgS0 4 , atropine), or fetal
not acidotic. In the presence of a persistently acidosis. In the absence of spontaneous
nonreassuring FHR tracing, however, the accelerations, fetal scalp stimulation or
pH should be repeated every 30-60min. A vibroacoustic stimulation often provokes
scalp pH of 7.20-7.25 is suspicious, and fetal movement and FHR accelerations. If
should be repeated within 30-60 min, these measure fail to induce FHR accelera-
regardless of other FHR findings. A scalp pH tions, and if other FHR characteristics are
less than 7.20 more, when confirmed, is an concerning, consideration should be given to
indication for delivery. A more practical and fetal scalp pH determination or delivery.
less invasive approach employs fetal scalp
stimulation or vibroacoustic stimulation. Decelerations
Frequently, these stimuli will provoke FHR
accelerations and improve FHR variability, Decelerations in the FHR are most commonly
thereby providing reassurance that the fetus encountered during the intrapartum period.
is not acidotic. If reassurance cannot be They are generally divided into three
obtained, delivery should be considered. categories ("early, variable, and late") on
the basis of their temporal relationship to
the onset of a uterine contraction.
Periodic Patterns
Early Decelerations
The FHR baseline is commonly interrupted Early decelerations are typically uniform,
by accelerations or decelerations in rate. shallow dips in the FHR (rarely below
These periodic patterns have important 100 bpm) that start when a contraction starts
clinical implications regarding the well- and end when the contraction ends. They
being of the fetus. probably result from fetal head compression
and a reflex augmentation of vagal tone.
Accelerations Perinatal outcome is not adversely affected
by these decelerations, and they are con-
Accelerations in FHR occur in association sidered to be clinically benign.
with fetal movement, probably as a result of
increased catecholamine release and de-
Variable Decelerations
creased vagal stimulation of the heart.
Starting at approximately 30-32 weeks Variable decelerations result from umbilical
gestation, they normally occur during the cord compression, and have a variable tem-
fetal wake state at a rate of 15-20 per hour. poral relationship to uterine contractions.
The occurrence of at least two qualifying Initially, umbilical vein compression de-
FHR accelerations (15 bpm for 15 s) in a creases fetal venous return and causes reflex
period of 10 min constitutes a "reactive" FHR elevation ("shoulder"). Subsequent
FHR pattern. A reactive FHR pattern is a umbilical arterial compression dramatically
reassuring finding, reflecting a normally increases fetal peripheral resistance and
110 D.A. Miller and R.H. Paul

produces a rapid-onset baroreceptor-medi- contraction, uterine rupture, maternal

ated slowing of the heart rate. Maximum hypotension, maternal apnea, or placental
vagal tone may result in a junctional or abruption). Umbilical cord prolapse must be
idioventricular escape rhythm which appears excluded. Maternal positional changes or
as a relatively stable rate of 60-70 bpm. As manual elevation the fetal head may relieve
the cord is decompressed, this sequence of cord compression, if present. Documentation
events occurs in reverse. Isolated variable of separate maternal and fetal heart rates is
decelerations are usually not clinically sig- necessary. Heart block may be ruled out by
nificant. Repetitive severe variables «70 ultrasound confirmation of identical fetal
bpm for >60 seconds), however, may not atrial and ventricular rates. Acute maternal
allow sufficient fetal recovery between hypotension (i.e., epidural anesthesia) may
decelerations, resulting in persistent hypo- respond to positional changes, fluids, and
xemia, hypercapnia and respiratory acidosis. ephedrine, if necessary. Tetanic uterine con-
Prolonged tissue hypoperfusion may lead to tractions are often relieved by discontinuing
metabolic acidosis and, ultimately, fetal oxytocin; however, uterine relaxants
death. In animal models, Clapp et a1. 92 (MgS0 4 , terbutaline) are occasionally re-
reported that frequent episodes of hypoxemic quired. Oxygen is administered to the mother
stress, produced by intermittent umbilical by face mask. Ifthese measures fail to result
cord occlusion over a period of hours, pro- in resolution of the prolonged deceleration,
duced fetal injury even in the absence of rapid delivery is indicated.
acidosis.
When repetitive, severe variable decelera-
Late Decelerations
tions are present, prolapse of the umbilical
ctlrd must be ruled out by digital examina- Late decelerations reflect inadequate utero-
tion. Maternal positional changes may placental transfer of oxygen during uterine
relieve cord compression. Uterine hypertonus contractions. Typically, these are uniform
or tachysystole may be relieved by discon- decelerations that start after the onset of a
tinuing oxytocin, administering uterine contraction and end after the contraction
relaxants (MgS0 4 , terbutaline), or both. ends. During uterine contractions, disruption
Restoration of normal amniotic fluid volume of maternal uteroplacental perfusion causes
in patients with oligohydramnios can de- a decline in fetal Po 2 . When the fetal P0 2
crease the frequency and severity of variable falls below a critical threshold (17 -18
decelerations as well as the incidence of fetal' mm Hg), a complex chemoreceptor- and
distress as diagnosed by the FHR tracing. 93 - baroreceptor-mediated reflex is initiated.
95 Transcervical amnioinfusion with 300- Centralization of blood volume (favoring
500 ml of warmed saline should be instituted perfusion of the brain, heart, and adrenals)
as needed, and ideally an amniotic fluid index occurs via vasoconstriction in the vascular
(AFI) of at least 10 cm should be maintained. beds of the limbs and gut. The resulting
Repetitive severe variable decelerations increase in peripheral resistance provokes a
that persist despite these maneuvers must reflex deceleration in the FHR. Isolated late
be evaluated in the context of the associated decelerations within an otherwise normal
FHR patterns. In the absence of reassuring tracing are usually of minimal clinical sig-
FHR findings, delivery should be expedited. nificance. However, continuing hypoxic
Occasionally, variable decelerations fail stress, as evidenced by repetitive late de-
to promptly return to baseline, and may be celerations, may lead to metabolic acidosis,
more accurately termed "prolonged decelera- asphyxia, and eventually fetal death.
tions." Prolonged decelerations are usually The goal in treating late decelerations is
the result of cord compression (cord prolapse, to improve uteroplacental perfusion and
tight nuchal cord) or other acute interruption oxygen delivery to the fetus. Usual measures
of utero placental transfer of oxygen (tetanic include (1) the left lateral decubitus position,
7. Cesarean Section for Fetal Distress 111

to improve maternal venous return and 2. Rosen MG, Chairman. Consensus Task
cardiac output, (2) oxygen by face mask, (3) Force on Cesarean Childbirth. National
an intravenous fluid bolus of 250-500 ml of Institutes of Health Publication No. 82-2067.
crystalloid to restore the maternal intravas- Washington, DC: US Dept of Health and
cular volume and improve cardiac output, Human Services, 1981:49, 125, 275.
and (4) discontinuation of oxytocin. In the 3. Speert H. A pictorial history of gynecology
and obstetrics. Philadelphia: FA Davis,
presence oflate decelerations with decreased
1973:297.
FHR variability, reassurance of fetal well- 4. Morris RC, ed. Obstetrics. Philadelphia: WE
being is imperative to justify continuation of Saunders, 1895:917.
labor. A normal fetal scalp pH (>7.25), or an 5. Sachs BP, Yeh J, Acker D, Driscoll S, Brown
acceleratory FHR response to fetal scalp or DAJ, Jewett JF. Cesarean section-related
vibroacoustic stimulation provides reassur- maternal mortality in Massachusetts, 1954-
ance that acidosis is not present. A pattern 1985. Obstet GynecoI1988;71:385-388.
of repetitive late decelerations without such 6. Douglas RG, Landesman R. Recent trends in
reassurance strongly suggests fetal distress cesarean section. Am J obstet Gynecol 1950;
and is an indication for delivery. 59:96-107.
7. Phelippeaux. Notice biographique et biblio-
graphique sur Philippe Le Goust. Archives
de Tocologie des Maladies des Femmes, Paris
Conclusion 1879;6:304, quoted by Gultekin-Zootzmann.6o
8. Mayor H. Biblioth Univ. De Geneve, Novem-
Dramatic changes in intrapartum manage- ber 9, 1818. Quoted by Thomas H. Classical
ment have taken place during the past two Contributions to obstetrics and gynecology.
decades, highlighted by the tremendous pro- Springfield, Illinois: CC Thomos, 1935.
liferation of electronic fetal monitoring, the 9. Le Jumeau JA (de Kergaradec). Memoire sur
decline in maternal and perinatal mortality, l'Auscultation appliquee a 1'Etude de la
and the soaring utilization of cesarean birth. Grossesse ou Recherches sur deux nouveaux
Many widely held beliefs have been chal- signes propres a faire reconnaitre plusieurs
lenged. Large case-control studies have circonstances de 1'Etat de Gestation; lu a
l'Academie royale de medecine, dans sa
demonstrated the limited contribution of
seance generale du 26 decembre 1821. Paris,
"birth asphyxia" to the overall incidence of 1822, quoted by Gultekin-Zootzmann. 6o
CPo Randomized trials have dampened some 10. Kennedy E. Observations of obstetrical
of the early enthusiasm regarding the auscultation. Dublin: Hodges and Smith,
potential benefits of EFM (although it may 1833.
be rekindled by the most recent study). 11. Schwartz H. Arch Gynaekol 1870;1:361.
Nevertheless, the objective of intrapartum 12. Winckel F. Lehrbuch der Geburtshilfe.
management remains the same: to optimize Leipzig: 1893, Veit.
the outcome for both mother and fetus by (1) 13. ·Kilian. Quoted by Jaggard WW, in Hirst BC,
preventing intrapartum fetal asphyxia and ed. A system of obstetrics. Philadelphia: Lea
possible long-term sequelae, and (2) avoiding Broth,1888.
unnecessary operative deliveries. To that 14. Seitz L. Die fetalen Herztone unter der
Geburt. Munchen: Habil-Schrift, 1903.
end, the most effective resource available to
15. Hillis DS. Attachment for the. stethoscope.
the obstetrician is a thorough understanding JAMA 1917;68:910.
of the FHR patterns associated with normal 16. DeLee JB. Ein nues stethoskopf fur die
and abnormal fetal physiologic states. Geburtshilfe besonders geeignet. Zentralbl
Gynaekol 1922;46:1688.
17. Cremer MV. Ueber die direckte ableitung
References
der aktionsstrome des menchlichen herzens
1. National Center for Health Statistics. Rates vom oesophagus und ueber das elektrokardio-
of Cesarean delivery-United States, 1991. gramm des fetus. Munch Med Wochenschr
Morb Mortal Wkly Rep 1993;42:285-289. 1906;53:811.
112 D.A. Miller and R.H. Paul

18. Smyth CN. Lancet 1953;2:1124. 33. Alberman E. Main causes of major mental
19. Sureau CL. Recherches electrocardiographi- handicap: prevalence and epidemiology. In:
ques foetal au cours de la gestation et du DIBA Foundation symposium 59, major
travail: premiers resultal d'une nouvelle mental handicap: methods and costs of pre-
technique d'enregistrement par electrodes vention. Amsterdam, Elsevier-Excerpta
endouterines. Gynecol Obstet 1956;55:21. Medica, 1978.
20. Kaplan S, Toyama S. Fetal electrocardio- 34. National Institutes of Health. On causes of
graphy: utilizing abdominal and intrauterine mental retardation and cerebral palsy: task
leads. Obstet Gynecol 1958;11:391. force on joint assessment of prenatal and
21. Williams RL, Hawes WE. Cesarean section, perinatal factors associated with brain dis-
fetal monitoring and perinatal mortality in orders. Pediatrics 1985;76:457.
California. Am J Public Health 1979;69:864. 35. Little WJ. On the influence of abnormal
22. National Center for Health Statistics. parturition, difficult labours, premature birth
Monthly Vital Statisties Report: Advance and asphyxia neonatorum, on the mental and
Report of Maternal and Infant Health Data physical condition of the child, especially in
from the Birth Certificate, 1990. Vol. 42, No. relation to deformities. Trans Obstet Soc Lond
2 (suppl), July 8, 1993. 1862;3:293.
23. Kirschbaum TH. Editorial: diagnosis of fetal 36. Schreiber F. Apnea ofthe newborn and asso-
distress. Obstet Gynecol 1969;34:721-728. ciated cerebral injury: a clinical and statisti-
24. Haverkamp AD, Orleans M, Langendoerfer cal study. JAMA 1938;111:1263.
S,McFee J, Murphy J, Thompson HE. A 37. Lilienfeld AM, Parkhurst E. Study of asso-
controlled trial of the differential effects of ciation offactors of pregnancy and parturition
intrapartum fetal monitoring. Am J Obstet with development of cerebral palsy: preli-
Gynecol 1979;134:399-408. minary report. Am J Hyg 1951;53:262-270.
25. Haesslein HC, Niswander KR. Fetal distress 38. Lilienfeld AM, Pasamanick BP. The associa-
in term pregnancies. Am J Obstet Gynecol tion of maternal and fetal factors with the
1980;137:245-253. development of cerebral palsy and epilepsy.
26. The American Heritage Dictionionary of the Am J Obstet Gynecol1955;70:93-101.
English Language, Third Edition Boston, 39. Eastman NJ, DeLeon M. The etiology of
Massachusetts, Houghton Mifflin, 1992. cerebral palsy. Am J Obstet Gynecol 1955;
27. Dorland's illustrated medical dictionary, 69:950-961.
26th Ed. Philadelphia: WB Saunders 1985; 40. Eastman NJ, Kohl SG, Maisel JE, Kavaler F.
129:990. The obstetrical background of 753 cases of
28. Parer JT, Livingston EG. What is fetal dis- cerebral palsy. Obstet Gynecol Surv 1962;
tress? Am J Obstet Gynecol 1990;162:1421- 17:459-500.
1427. 41. Steer CW, Bonney W. Obstetric factors in
29. Gilstrap LC, Leveno KJ, Burris J, Williams cerebral palsy. Am J Obstet Gynecol 1962;
ML, Bertis BL. Diagnosis of birth asphyxia 83:526-531.
on the basis of fetal pH, Apgar score, and 42. Nelson KB, Ellenberg JH. Antecedents of
newborn cerebral dysfunction. Am J obstet cerebral palsy. I. Univariate analysis of risks .
Gynecol 1989;161:825-830. Am J Dis Child 1985;139:1031-1038.
30. Committee on Obstetrics, Maternal and Fetal 43. Windle WF, Becker RF. Asphyxia neon-
Medicine of the American College of Obste- atorum: an experimental study in the guinea
trics. Utility of umbilical cord blood acid-base pig. Am J Obstet Gynecol1943;45:183-199.
assessment (ACOG Committee Opinion #91, 44. Ranck JB, Windle WF. Brain damage in the
February 1991). Washington, DC: American monkey, Maccaca mulatta, by asphyxia
College of Obsktricians and Gynecologists, neonatorum. Exp Neurol1959;1:130-154.
1991. 45. Faro MD, Windle WF. Transneuronal
31. Johnson MV.~ Cellular alterations associated degeneration in brains of monkeys asphy-
with perinatal asphyxia. Clin Invest Med xiated at birth. Exp Neurol 1969;24:38-53.
1993;16(2):122-132. 46. Myers RE. Fetal asphyxia and perinatal brain
32. Nelson KB, Ellenberg JH. Epidemiology of damage affecting human development.
cerebral palsy. In: Schoenberg BS, ed. Publication No. 185. Washington, DC: Pan
Advances in neurology, Vol. 19. New York: American Health Organization, 1969:205-
Raven, 1978:421-435. 214.
7. Cesarean Section for Fetal Distress 113

47. Myers RE. Two patterns of perinatal brain baseline FHR variability. Am J Obstet
damage and their conditions of occurrence. Gynecol 1975;123:206-210.
Am J Obstet Gynecol 1972;122:246- 64. Clark SL, Gimovsky ML, Miller FC. The scalp
276. stimulation test: a clinical alternative to fetal
48. Freud S. Die Infantile CerebralHihmung. scalp blood sampling. Am J Obstet Gynecol
Nothnagels Spez Path u Therapie (Wien) 1984;148:274-277.
1957;IX:2. 65. Smith CV, Nquyen HN, Phelan JP, Paul RH.
49. Haddow KM, Gage RP. Neurologic lesions in Intrapartum assessment of fetal well-being:
relation to asphyxia of the newborn and a comparison of fetal acoustic stimulation
focators of pregnancy: long-term follow-up. with acid-base determinations. Am J Obstet
Pediatrics 1960;26:616-622. Gynecol 1986;155:726-728.
50. Blair E, Stanley FJ. Intrapartum asphyxia: a 66. Chan WH, Paul RH, Toews J. Intrapartum
rare cause of cerebral palsy. J Pediatr 1988; fetal monitoring: maternal and fetal mor-
112:515-519. bidity and perinatal mortality. Obstet
51. Nelson KB, Ellenberg JH. Antecedents of Gynecol 1973;41:7-13.
cerebral palsy: multivariate analysis of risk. 67. Kelly VC, Kulkarni D. Experiences with fetal
N Engl J Med 1986;315:81-86. monitoring in a community hospital. Obstet
52. Paneth N. The causes of cerebral palsy. Gynecol 1973;41:818-824.
Recent evidence. CIin Invest Med 1993;16: 68. Tutera G, Newman RL. Fetal monitoring:
95-102. its effect on the perinatal mortality and
53. Stanley FJ, Blair E. Why have we failed to caesarean section rates and its complications.
reduce the frequency of cerebral palsy. Med J Am J Obstet Gynecol 1975;122:750-754.
Aust 1991;154:623. 69. Edington PT, Sibanda J, Beard RW. Influence
54. Takesita K, Ando Y, Ohtani K, et al. Cerebral on clinical practice of routine intra-partum
palsy in Tottori, Japan. Neuroepidemiology fetal monitoring. Br Med J 1975;3:341-343.
1989;8:184-192. 70. Shenker L, Post RC, Seiler JS. Routine
55. Stanley F. Survival and cerebral palsy in electronic monitoring of fetal heart rate and
low birthweight infants: implications for uterine activity during labor. Obstet Gynecol
perinatal care. Paediatr Perinatol Epidemiol 1975;46:185-189.
1992;6:298-310. 71. Koh KS, Greves D, Yung S, et al. Experience
56. Riikonen R, Raumavirta S, Sinivuori E, et al. with fetal monitoring in a university teaching
Changing pattern of cerebral palsy in the hospital. Can Med Assoc J 1975;112:455-
south-west region of Finland. Acta Paediatr 462.
Scand 1989;78:581-587. 72. Lee WK, Baggish MS. The effect of unselected
57. Pharoah POD, Cooke T, Cooke RWI, et al. intrapartum fetal monitoring. Obstet Gynecol
Birthweight specific trends in cerebral palsy. 1976;47:516-520.
Arch Dis Child 1990;65:602-606. 73. Paul RH, Huey JR, Yaeger CF. Clinical fetal
58. Ellenberg JH, Nelson KB. Birth weight and monitoring-its effect on cesarean section
gestational age in children with cerebral palsy rate and perinatal mortality: five-year trends.
or seizure disorders. Am J Dis Child 1979; Postgrad Med 1977;61:160-164.
133:1044. 74. Amato JL. Fetal monitoring in a community
59. Schwartz H. Die vorzeitgen Athembewegun- hospital: a statistical analysis. Obstet
gen. Leipzig, 1838, quoted by Giiltekin- GynecoI1977;50:269-274.
Zootzmann. 60 75. Johnstone FD, Campbell DM, Hughes GJ.
60. Giiltekin-Zootzmann B. The history of Antenatal care: has continuous intrapartum
monitoring the human fetus. J Perinat Med monitoring made any impact on fetal out-
1975;3:135-144. come? Lancet 1978;1:1298-1300.
61. Hon EH, Quilligan EJ. The classification of 76. Hamilton LA, Gottschalk W, Vidyasagar D,
fetal heart rate. Conn Med 1967;31:779. et al. Effects of monitoring on perinates. Int J
62. Kubli FW, Hon EH, Khazin AF, Takemura Gynaecol Obstet 1978;15:483-490.
H. Observations on heart rate and pH in the 77. MacDonald D, Grant A. Fetal surveillance in
human fetus during labor. Am J Obstet labour-the present position. In: Bonnar
GynecoI1969;104:1190-1206. J, ed. Recent advances in obstetrics and
63. Paul RH, Suidan AK, Yeh S, Hon EH. The gynaecology, Vol. 15. London: Churchill
evaluation and significance of intrapartum Livingstone, 1987:83-100.
114 D.A. Miller and R.H. Paul

78. Haverkamp AD, Thompson HE, McFee JG, 86. Goodwin JW, Dunne JT, Thomas RW.
Cetrullo C. The evaluation of continuous Antepartum identification of the fetus at risk.
fetal heart rate monitoring in high-risk Can Med Assoc J 1969;101:458.
pregnancy. Am J Obstet Gynecol 1976;125: 87. Renou P, Wood C. Interpretation of the con-
310-320. tinuous fetal heart rate record. Clin Obstet
79. Renou P, Chang A, Anderson I, Wood C. GynaecoI1974;1:191-215.
Controlled trial of fetal intensive care. Am J 88. Shelley T, Tipton RH. Dip area: A quantita-
Obstet GynecoI1976;126:470-476. tive measure of fetal heart rate patterns.
80. Luthy DA, Kirkwood KS, van Belle G, et al. J Obstet Gynecol Br Commonw 1971;78:
A randomized trial of electronic fetal moni- 694-701.
toring in preterm labor. Obstet Gynecol 89. Fleiss. Statistical methods for rates and
1987;69:687 -695. proportions, 2nd Ed. New York: Wiley, 1981:
81. Vintzileos AM, Antsaklis A, Varvarigos I, 38-45.
Papas C, Sofatzis I, Montgomery JT. A 90. Grant A, O'Brien N, Joy M, Hennessy E,
randomized trial of intrapartum electronic MacDonald D. Cerebral palsy among children
fetal heart rate montoring versus intermit- born during the Dublin randomized trial of
tent auscultation. Obstet Gynecol 1993;81: intrapartum monitoring. Lancet 1989;2:
899-907. . 1233-1236.
82. Kelso 1M, Parsons R.I, Lawrence GF, Arora 91. Gleicher H, Runowicz C, Brown B. Sinusoidal
SS, Edmonds DK, Cooke ID. An assessment fetal heart rate patterns in association with
of continuous fetal heart rate monitoring in amnionitis. Obstet GynecoI1980;56:109.
labor: a randomized trial. Am J Obstet 92. Clapp JF, Peress NS, Wesley M, Mann LI.
Gynecol 1978;131:526-532. Brain damage after intermittent partial cord
83. Wood C, Renou P, Oats J, Farrell E, Bleischer occlusion in the chronically instrumented
N, Anderson 1. A controlled trial of fetal fetal lamb. Am J Obstet Gynecol 1988;159:
heart rate monitoring in a low-risk obstetric 504-509.
population. Am J Obstet Gynecol 1981;141: 93. Miyazaki FS, Taylor NA. Saline amnioinfu-
527-534. sion for relief of variable prolonged decelera-
84. MacDonald D, Grant A, Sheridan-Pereira M, tions. Am J Obstet Gynecol 1983;146:
Boylan P, Chalmers 1. The Dublin randomized 670-674.
controlled trial of intrapartum fetal heart 94. Miyazaki FS, Nevarez F. Saline amnioinfu-
rate monitoring. Am J Obstet Gynecol 1985; sion for relief of repetitive variable decelera-
152:524-539. tions: a prospective randomized study. Am J
85. Neldam S, Osler M, Hansen PK, Nim J, Obstet Gynecol 1985;153:301-306.
Smith SF, Hertel J. Intrapartum fetal heart 95. Strong TH, Hetzler G, Sarno AP, Paul RH.
rate monitoring in a combined low- and high- Prophylactic intrapartum amnioinfusion: a
risk population: a controlled clinical trial. randomized clinical trial. Am J Obstet
Eur J Obstet Gynecol Reprod BioI 1986;23: GynecoI1990;162:1370-1375.
1-11.
8
Fetal Macrosomia
HUNG N. WINN AND JOHN C. HOBBINS

Macrosomia, arbitrarily defined as a birth- diabetes, obesity, advanced age, multi-

weight of more than 4000 g at term, com- parity, postterm pregnancy, and prior
plicates about 10% and 25% of term and maternal and sibling birth weights of more
postterm pregnancies, respectively.1-4 It than 4000 g are at risk of having fetal
remains an important cause of perinatal macrosomia. 2.5- 8
morbidity and mortality and maternal In general, maternal diabetes mellitus
morbidity, which arise mainly from birth doubles the risk of having neonatal macro-
injury and asphyxia and increased rate of somia compared to that of the nondiabetic
cesarean section, respectively.5 In addition population. Maternal hyperglycemia, espe-
to the significant immediate complications, cially during the period of 29 to 32 weeks of
the long-term impact ofless severe perinatal gestation, has been associated with fetal
asphyxia on the mental development of the macrosomia. 9- 14 Interestingly, it has been
affected children is unknown. The major demonstrated that the risk of fetal macro-
obstacle in reducing the perinatal morbidity somia also rises in untreated patients with
and mortality associated with fetal macro- one abnormal value on a glucose tolerance
somia has been the inability to predict with test (GTT),15-17 even though these patients
certainly which fetuses will sustain birth may be classified as normal if the criterion of
injury before delivery. To minimize the having two abnormal values is used for
adverse perinatal and maternal outcomes diagnosing gestational diabetes mellitus.
associated with this problem, it is important The mechanism of fetal macrosomia
to have a well-planned management scheme in the case of maternal diabetes mellitus as
for patients with fetal macrosomia, recog- originally proposed by Pederson et al. 18
nizing that there is currently no perfect is the acceleration of fetal growth from
solution. In the following discussion, the fetal hyperinsulinemia in response to fetal
current understanding of fetal macrosomia hyperglycemia, which is in turn affected by
with regard to risk factors, diagnosis, and maternal hyperglycemia. This hypothesis is
intrapartum management is presented. supported by a high degree of correlation
between cord serum levels of C-peptide and
fetal macrosomia in diabetic patients. 19
Risk Factors Fetal macrosomia and organomegaly of
the placenta, liver, and heart have been
The first step in the management of fetal observed in the chronically maintained
macrosomia is the identification of patients euglycemic and hyperinsulinemic rhesus
at risk for this condition. It has been shown monkey.20 The direct impact of maternal
that patients with such characteristics as and fetal hyperglycemia per se on fetal

115
116 R.N. Winn and J.e. Robbins

macrosomia remains unclear. 19,21,22 It is nation. Estimation offetal weight by clinical

possible that macrosomic fetuses maintain examination of fundal height is usually
euglycemia by increased production of inaccurate, especially at the extremes of
insulin. This could explain the similar levels fetal sizes. A fundal height that is larger
of maternal glycosylated hemoglobin but than expected may point to the potential for
higher levels of C-peptide in cord serum of fetal macrosomia. Ultrasonography provides
macrosomic fetuses compared with non- a more accurate means of predicting fetal
macrosomic fetuses in diabetic patients. macrosomia. One approach utilizes the
It has also been demonstrated that the macrosomic index, which is the difference
exogenous animal insulin (bovine or por- between the fetal chest circumference and
cine) III combination with the anti- the biparietal diameter (BPD) , to identify
insulin antibody can cross the placenta as an macrosomic fetuses in pregnancies compli-
insulin-antibody complex and that there is cated by diabetes mellitus. In one study,
a significant correlation between fetal 87% of macrosomic infants had an average
macrosomia and elevated cord serum levels chest diameter that exceeded the BPD by
of animal insulin. 23 Thus, the exogenous 1.4cm or greater. 29
insulin administered to pregnant patients If the gestational age is known, ul-
may independently accelerate fetal growth. trasound examination . during the third
Metabolic factors other than maternal trimester can be used to identify large-for-
hyperglycemia may also play a role in gestational-age (LGA) fetuses who are at
causing fetal macrosomia. risk for being macrosomic at delivery. If the
Because the fetus may continue to grow estimated fetal weight (EFW) or the fetal
beyond 37 weeks of gestation, the incidence abdominal circumference (AC), the latter
of fetal macrosomia increases with advan- reflecting insulin-sensitive tissue, are above
cing gestational age with an overall rate of the 90th percentile, macrosomia can be
about 25%-30% at 41 weeks gestation or correctly diagnosed in 74% and 88.8% of
greater. 1,4 Maternal obesity (prepregnancy infants of diabetic mothers, respectively. 3D
weight of at least 90 kg) is a strong indicator In pregnancies complicated by diabetes
of fetal macrosomia and is associated with mellitus, accelerated growth of abdominal
an increased risk of at least 1.5 fold.24~27 circumference (above the 90th percentile),
The increased incidence of fetal macrosomia which may become apparent as early as 24
in obese patients could be caused by mul- weeks of gestation, is associated with in-
tiple factors, such as altered metabolic creased risk of fetal macrosomia at term. 31
homeostasis with an inherent risk of ab- A growth of abdominal circumference of
normal glucose metabolism and heredity. 1.2 cm or more per week during the gesta-
Advanced maternal age (more than 35 years tional interval of 32-39 weeks is noted in
old) is also a risk factor for fetal macrosomia about 80% of macrosomic infants of diabetic
even after correction for parity and gesta- pregnancies. 32 Optimal clinical manage-
tional diabetes. 28 Genetic influence may ment of fetal macrosomia requires more
also account for the high correlation be- than just identification of fetuses at risk for
tween maternal and sibling birth weights macrosomia; it requires an accurate estima-
and the current neonate's birthweight. tion of fetal weight. In general, by using the
fetal AC and the BPD, one can estimate fetal
weights on the basis of the published for-
Diagnosing Fetal Macrosomia mulas to within 10% and 5% of the actual
weight in 80% and 50% of normal fetuses,
The next step in the management of patients respectively.33 It appears that the formula of
at risk for fetal macrosomia at term is Hadlock et al.,34 which uses the AC and
estimating fetal weight. This can be done femur length (FL) [loglO(weight) = 1.304 +
by clinical assessment or ultrasound exami- 0.05281 AC + 0.1938 FL - 0.004 AC X FL]
8. Fetal Macrosomia 117

provides the best estimation of birth weight following situations: (1) uncertain gesta-
of macrosomic fetuses. 35 At present, ac- tional age, (2) gestational age less than 38
curacy in predicting birthweight in macro- weeks, (3) and gestational age of less than
somic fetuses remains limited. In fact, one 40.0 weeks in diabetic pregnancies.
study demonstrated a positive predictive
value for fetal macrosomia of only 67%
when the birthweight was greater than 4100
Induction of Labor
grams. 36 One of the major concerns about induction of
Failure to take into account the increased labor in the presence of an unfavorable
fetal body fat in macrosomic fetuses, at least cervix, that is, a Bishop score of 5 or less, is
among diabetic pregnancies, may contribute failed induction. To facilitate the induction,
to the overestimation of fetal weight by cervical ripening can be accompli sed by the
ultrasound utilizing available formulas. 37 use of intracervical artificial laminaria
An innovative model taking into account (Lumicil) and low-dose continuous oxytocin
many variables such as maternal height, infusion overnight. The oxytocin infusion is
gravity, parity, gestational age, fundal started at 0.5 mu/min and doubled hourly to
height, amniotic fluid index, and fetal sub- a maximum dose of 4 mu/min. Mild uterine
cutaneous tissue in addition to the com- contractions usually are obtained at this low
monly used parameters such as BPD, AC, dose. This combination of laminaria and
head circumference, and FL, has reduced the low-dose oxytocin infusion is usually suffi-
error of predicting EFW in macrosomic cient to make the cervix favorable for a full
fetuses to only 4.7%.38 In pregnancies at risk induction of labor the next morning. Some
for fetal macrosomia, serial ultrasound advocate that amniotomy by preformed as
examinations every 2-3 weeks for EFW and soon as possible and that this represents the
AC starting at about 36.0 weeks of gestation critical point of induction by allowing the
may be helpful in detecting fetal macro- release of arachidonic acid from the mem-
somia. It is important to recognize that the branes for conversion to prostaglandins. All
error of estimating fetal weight by ultra- our patients reach the active phase of labor,
sound utilizing the available formulas could and the majority do so within a normal
be as great as 10%. length of time, when this protocol is followed.
Accurate assessment of uterine contrac-
tions and optimal fetal heart rate monitor-
Intrapartum Management ing are important whenever induction of
labor is attempted. This becomes even more
Once fetal macrosomia is suspected, de- important when induction of labor is car-
livery should be considered after fetal lung ried out for impending fetal macrosomia
maturity is documented. In most situations, in view of the increased risk of CPD and
a trial oflabor can be attempted unless there birth injury. Labor course should be close-
are contraindications such as absolute ly monitored for arrest or protraction
cephalopelvic disproportion (CPD), placenta disorders. 22 ,39,4o Thus, intrauterine pres-
previa, vasa previa, malpresentation of the sure and direct fetal electronic monitoring
fetus, previous classical cesarean section, an should be initiated as soon as it is feasible
EFW greater than 5000 g, or documentation and safe to do so.
of substantial body-to-head disproportion in In the United States, intravenous oxyto-
a fetus of more than 4500 g. Fetal lung cin is the drug of choice for induction of
maturity should be documented before labor. Recently there has been a shift to a
induction of labor. Amniocentesis for the lower total dose by starting at a lower
lecithin/sphingomyelin (L/S) ratio and concentration and increasing the concen-
phosphatidylglycerol (PG) is generally uti- tration at longer intervals and in lesser
lized to document fetal lung maturity in the increments. This change in the usage of
118 R.N. WinnandJ.C. Robbins

oxytocin for induction of labor has been there is a strong correlation between birth-
made because (1) the steady-state oxytocin weight and brachial plexus injury, the latter
concentration is not reached until approxi- occurs as a consequence of shoulder dystocia
mately 40 min from the last dose,41,42 (2) in most patients. 51 Asphyxia and brachial
effective uterine contractions can occur with palsy occur in as many as 42% of infants
an oxytocin infusion rate between 4.0 and with true shoulder dystocia. 52
8.0 mu/min in the majority of patients (in Shoulder dystocia is a dreadful experience
fact, approximately 2 SD of patients will for many obstetricians and their patients
develop progressive cervical dilatation with because it is associated with significant
an oxytocin infusion rate of no more than perinatal morbidity and mortality and
8.0mu/min),42-44 and (3) the incidence of usually occurs unexpectedly. It would be
uterine hypercontractility and the resulting ideal if shoulder dystocia could be antici-
abnormal heartrate are reduced. pated and thus could be avoided in pregnan-
The following is a suggested protocol for cies at risk for this condition. In general,
the continuous infusion of oxytocin for fetal weight or pelvimetry is a poor indicator
induction of labor: The oxytocin infusion is of CPD or birth trauma. The fetal-pelvic
started at 0.5 mu/min and then doubled index (FPI) was recently introduced to
every hour up to the rate of 8 mu/min. address this issue. 7 First, the ante-
Thereafter, the dose is increased 4 mu/ roposterior and transverse diameters of the
min every hour until the maximum dose of fetal head and abdomen and the maternal
20 mu/min is reached. The dose of oxytocin inlet and midpelvis are determined. Fetal
should not be raised further when adequate and maternal measurements are obtained
uterine contractions, that is, a frequency of by ultrasonography and x-ray pelvimetry,
2-3 min, durations of 40-90 s, and inten- respectively. The corresponding circum-
sities of 40-90 mmHg, or progressive cervi- ferences are calculated from the two per-
cal dilatation and descent, are obtained pendicular diameters using the formula C =
regardless of how low the oxytocin dose may (TD + APD) x 3.14/2, where C is the cir-
be. A lower starting dose of oxytocin reduces cumference, APD the anteroposterior dia-
the risk of uterine hyperstimulation, which meter, and TD the transverse diameter. The
can occur at very low doses. 45 ,46 FPI is the sum of the two most positive
circumference differences (fetopelvic dif-
ferences: HC - IC and AC - MC). A positive
Birth Injury and Shoulder Dystocia or negative FPI indicates the presence or
The major complication associated with absence of CPD/shoulder dystocia, respec-
delivery of a macrosomic fetus is birth tively. The sensitivity and specificity of the
trauma, such as clavicle fracture and sub- FPI in the diagnosis ofCPD/shoulder dystocia
sequent brachial plexus injury from a diffi- are 94%. Further study is needed to deter-
cult vaginal delivery.39,47-5o The overall mine the predictive value of the FPI in
risk of shoulder dystocia rises sharply from detecting CPD/shoulder dystocia in pre-
3% for birthweights less than 4000 g to gnancies complicated by fetal macrosomia.
10.3% and 23.9% for birthweights ranging Another approach to the detection of
between 4000 and 4500 g and those greater shoulder dystocia has utilized fetal biome-
than 4500 g, respectively. When birth- tric evaluation. Studies of neonates have
weights are greater than 3500 g, the inci- demonstrated significant differences in
dence of shoulder dystocia generally doubles many neonatal anthropometric measure-
in diabetic patients compared to nondiabetic ments between newborn infants with and
mothers for similar birthweights. Thus, the without dystocia. The mean shoulder
combination of macrosomia and diabetes circumference was significantly larger
places the patient at a very high risk for when shoulder dystocia had complicated
neonatal shoulder dystocia. 47 Although delivery. 53 A recent study demonstrated
8. Fetal Macrosomia 119

that the neonatal bisacromial diameter is McRobert's Maneuver

well correlated with the ultrasonographi-
This easily performed maneuver involves a
cally measured circumferences of the fetal
sharp flexion of the patient's thighs onto the
chest and arm.54 Thus, the relationship
abdomen. It accomplishes the following: (1)
between fetal chest circumference and other
straightening of the maternal sacrum rela-
fetal parameters such as head circumference
tive to the lumbar spine, thus eliminating
or abdominal circumference may bepoten-
the sacral promontory as an obstruction; and
tially useful in predicting shoulder dystocia
(2) anterior rotation of the pubic symphysis,
in pregnancies at risk for this condition.
thus assisting in bringing the fetal posterior
Once a trial of labor is under way, the
shoulder through the pelvic inlet by simul-
course of labor should be closely followed by
taneously displacing the fetal anterior
a cervicographic analysis. Protraction and
shoulder cephalad. 59
arrest disorders of labor may be associated
with an increased risk for shoulder dystocia.
VVoods' Maneuver
Operative vaginal delivery by itself or in
combination with an abnormal course of Woods was the first author to describe the
labor may predispose to shoulder dystocia in principle of rotation of the shoulders rather
the case of fetal macrosomia. 45 ,52,55,56 Thus, than direct traction on the fetal head in
one should be very cautious when selecting a resolving shoulder dystocia. In this classical
midforceps delivery to correct disorders of approach, delivery of the posterior shoulder
descent in the presence of fetal macrosomia, is accomplished by pressing on the anterior
especially in a diabetic mother. aspect of the posterior shoulder toward the
To minimize the perinatal morbidity and fetal back while a gentle pressure is placed
mortality associated with shoulder dystocia on the uterus fundus. The posterior shoulder
once it occurs, one should be ready to initiate can usually be delivered under the pubic
the appropriate steps and maneuvers to symphysis after a rotation of 180°. The
effect the delivery in a timely fashion. Time remaining shoulder is similarly delivered by
is a critical factor with regard to perinatal rotating an arc of 180°, but in the op-
mortality and morbidity. The cardinal posite direction with pressure on its anterior
diagnostic sign of shoulder dystocia is re- aspect. 58
traction of the delivered head against the
maternal perineum caused by the obstruc- Rubin's Maneuver
tion to the passage of both shoulders at the
Rubin's maneuver applies the same prin-
pelvic inlet. This situation more likely
ciple of rotation but differs from the Woods'
arises when both shoulders present to the
maneuver in that adduction instead of
pelvic inlet from the anterior-posterior
abduction of the shoulders is carried out.
position instead of an oblique one and
Adduction is accomplished by pushing the
when the shoulder circumferences are
large. 53 ,56,57 Direct traction on the fetal shoulder toward the fetal chest. It should be
reemphasized that the adduction is per-
head may be ineffective unless therapeutic
formed with direct pressure on the fetal
fracture of the clavicles is associated. Be-
shoulder, not by rotating the fetal head.
cause the fetus naturally passes through the
This maneuver results in reduction of
birth canal with a screwlike motion, rota-
the shoulder transverse diameter and
tion of the shoulder to reduce the shoulder
circumference as well as rotation of the
circumference, or rotation to an oblique
shoulder to the more oblique position. 50
position, are the underlying mechanisms of
most effective maneuvers. These are des-
cribed next. 56 ,58 Delivery of the Posterior Arm
Performance of this maneuver requires
locating the fetal elbow with the operator's
120 H.N. Winn and J.e. Hobbins

hand in the vagina. The fetal forearm is respectively.47 One study showed that 76%
then flexed, drawn across the fetal chest, of shoulder dystocia occurs in infants of
and delivered through the vaginal opening. diabetic mothers with birthweights of 4250 g
It would be easier for the operator to use his or greater. 63 Thus, it would not be un-
or her left or right hand, depending on reasonable to perform an elective cesarean
whether the fetal back is facing the mater- section for delivery of the macrosomic fetus
nalleft or right side, respectively. in this setting. Of course, vaginal delivery
should be attempted even in this weight
range if the patient has a history of previous
Zavanelli Maneuver
vaginal delivery of other infants of similar
The Zavanelli maneuver should be used as size or larger.
the last resort when all other maneuvers to Another situation involves breech presen-
effect a vaginal delivery have failed. The tation of a macrosomic fetus, for which elec-
maneuver involves returning the fetal head tive cesarean section delivery after failed
to the vagina by reversing the sequence of external version may be indicated. The
the normal birthing process. The fetus is American College of Obstetricians and
subsequently delivered by cesarean section Gynecologists places the upper limit for fetal
from the cephalic presentation. Excellent size of 4000 g when vaginal delivery of a
neonatal outcomes have been reported with breech fetus is attempted. 64 The patient's
this maneuver.60 prior obstetric history and the obstetrician's
experience with breech delivery certainly
play a significant role in the decision.
Elective Cesarean Section It should be emphasized that individuali-
Although macrosomia is associated with an zation is key to the management of macro-
increased incidence of cesarean section de- somic fetuses in these clinical settings,
livery as a result of abnormal labor such as because good data are lacking. Regardless of
arrest of descent or cervical dilatation, an the method of delivery, patients should be
estimated fetal weight in the range of 4000- well informed of the risks for shoulder
4900 g may not be an absolute indication for dystocia as well as the available options
elective cesarean section. In fact, a trial of when macrosomia is suspected.
labor in patients with ultrasonographic
diagnosis of fetal macrosomia (estimated
fetal weight above 90th percentile) may Suggested Management
result in vaginal delivery in about 60% of Protocol
cases. 61 Furthermore, vaginal birth after
previous cesarean section occurs in 58% and The best approach to the management of
43% of patients with infant birthweight fetal macrosomia remains to be determined
ranges of 4000-4499 g and 4500 g or greater, because the available data on this subject
respectively, without a higher risk of are retrospective and the error of estimating
uterine rupture. 62 However, elective cesa- fetal weight by ultrasonography and pre-
rean delivery may have a role in the man- dicting shoulder dystocia is still large. These
agement of pregnancies with fetal macroso- limitations should be considered in the clini-
mia in certain situations. For example, cal management of fetal macrosomia. The
when the EFW is more than 5000 g, the patient should also be so informed. An ultra-
birthweight is likely to be at least 4500 g, sound determination of fetal weight and
assuming 10% error is the estimation. On fetal presentation should be made as early
the other hand, it could be more than 5500 g. as 37 weeks of gestation in patients at risk
In this range of birthweights (>4500 g), the for fetal macrosomia. If there has been a
risks for shoulder dystocia are 22.6% and question of gestational age, an ultrasound
50% for nondiabetic and diabetic patients, examination should be made prior to 20
8. Fetal Macrosomia 121

weeks of gestation to confirm the date of pressure, and delivery of the posterior arm
confinement. in decreasing order of preference. The
Once fetal macrosomia is suspected, fetal Zavanelli maneuver should be attempted if
lung maturity should be documented before these procedures to effect vaginal delivery
delivery is attempted. Elective cesarean have failed and sufficient time has elapsed
section may be considered in cases in which since the occurrence of shoulder dystocia to
the estimated fetal weight is greater than raise concern about neonatal asphyxia or
5000 g is normal pregnancies, greater than death.
4500 g in diabetic patients, or greater
than 4000 g in breech presentation with the
exact cutoff level being influenced by the Conclusion
perceived accuracy of sonographic estimates
of fetal weight. External version of the The management of fetal macrosomia with
breech fetus may be attempted unless there an increased risk of shoulder dystocia and
are contraindications such as bleeding, pla- its attendant morbidity and mortality
centa previa or abruption, certain fetal remains a challenge for the modern clini-
anomalies, oligohydramnios, or premature cian. In spite of many attempts to identify
rupture of membranes. 65 ,66 The alternative the patient at risk for fetal macrosomia, the
approach would be to obtain the FPI and perfect predictor has not yet emerged. The
select the mode of delivery depending on physician must therefore rely on clinical
whether the FPI is positive or negative. It skills in raising his or her index of suspicion
should be noted that the FPI has not been for fetal macrosomia; once it is considered,
tested in a large population of patients to ancillary diagnostic testing should help to
determine its predictive value. confirm the diagnosis. The summation of
If the fetus has an estimated weight be- this information should then be weighed
tween 4000 and 4900 g in nondiabetic pa- carefully as the plan for delivery is con-
tients and presents as vertex, a trial of labor structed. A well-orchestrated attempt at
may be conducted provided there are no vaginal delivery or a timely operative de-
contraindications. Internal monitoring of livery can be carried out with a high pro-
fetal heartrate and uterine pressure should bability of excellent maternal and fetal
be initiated as soon as it is feasible. The outcomes. In cases in which macrosomia
course of labor should be closely followed, with shoulder dystocia is encountered
keeping in mind that an abnormal course without antecedent warning, foreknowledge
of labor may herald a potential CPD or of the serial maneuvers that will enable the
shoulder dystocia. Operative vaginal de- delivery of the impacted shoulders can
livery is not recommended because it may be minimize intrapartum trauma.
associated with an unusually high
risk of shoulder dystocia and subsequent
birth injury. References
If shoulder dystocia occurs in spite of the
foregoing precautions, timely application of 1. Arias F. Predictability of complications
a preplanned sequence of steps and effective associated with prolongation of pregnancy.
maneuvers should be carried out. Time is Obstet Gynecol 1987;70:101.
2. Boyd ME, Usher RH, McLean FH. Fetal
critical because the risk of fetal brain injury
macrosomia: prediction, risks, proposed
and death increases as the duration of management. Obstet Gynecol 1983;61:715.
shoulder dystocia advances. Arbitrarily, 3. Golditch 1M, Kirkman K. The large fetus.
we recommend trying the McRobert's Management and outcome. Obstet Gynecol
maneuver first because of its simplicity 1978;52:26.
and effectiveness, followed by Rubin's 4. Chervenak JL, Divon MY, Hirsch J, et al.
maneuver, Woods' maneuver, suprapubic Macrosomia in the postdate pregnancy: Is
122 H.N. Winn and J.C. Hobbins

routine ultrasonographic screening indi- 18. Pederson J, Bojsen-Moller B, Paulson H.

cated? Am J Obstet Gynecol 1989;161:753. Blood sugar in newborn infants of diabetic
5. Spellacy WN, Miller S, Winegar A, et al. mothers. Acta Endocrinol 1954;15:33.
Macrosomia-maternal characteristics 19. Sosenko IR, Kitzmiller JL, Loo SW, et al. The
and infant complications. Obstet Gynecol infant of the diabetic mother. N Engl J Med
1985;66:158. 1979;301:859.
6. Klebanoff MA, Mills JL, Berendes HW. 20. Susa JB, McCormick KL, Widness JA, et al.
Mother's birth weight as a predictor of mac- Chronic hyperinsulinemia in the fetal rhesus
rosomia. Am J Obstet Gynecol 1985;153:253. monkey. Diabetes 1979;28:1058.
7. Morgan MA, Thurnau GR. Efficacy of 21. Widness JA, Schwartz HC, Thompson D, et
the fetal-pelvic index for delivery of neo- al. Glycohemoglobin (HbA1J: a predictor of
nates weighing 4000 grams or greater: a birth weight in infants of diabetic mothers. J
preliminary report. Am J Obstet Gynecol Pediatr 1978;92:8.
1988;158:1133. 22. Yatscoff RW, Mehta A, Dean H. Cord blood
8. Larsen CE, Serdula MK, Sullivan KM. glycosylated (glycated) hemoglobin: corre-
Macrosomia: Influence of maternal over- lation with maternal glycosylated (glycated)
weight among a low-income population. Am hemoglobin and birth weight. Am J Obstet
J Obstet Gynecol 1990;162:490. Gynecol 1985;152:861.
9. Coustan DR, Imarah J. Prophylactic insulin 23. Menon RK, Cohen RM, Sperling MA, et al.
treatment of gestational diabetes reduces Transplacental passage of insulin in preg-
the incidence of macrosomia, operative nant women with insulin-dependent diabetes
delivery, and birth trauma. Am J Obstet mellitus. N Engl J Med 1990;323:309.
Gynecol 1984;150:836. 24. Edwards LE, Dickes WF, Alton IR, et al.
10. Goldberg JD, Franklin B, Lasser D, et al. Pregnancy in the massively obese. Course,
Gestational diabetes: impact of home glucose outcome, and obesity prognosis of the infant.
monitoring on neonatal birth weight. Am J Am J Obstet Gynecol 1978;131:479.
Obstet Gynecol 1986;154:546. 25. Gross T, Sokol RJ, King K. Obesity inpreg-
11. Lin C-C, River J, River P, et al. Good diabetic nancy. Risks and outcome. Obstet Gynecol
control early in pregnancy and favorable 1980;56:4.
fetal outcome. Obstet Gynecol 1986;67:51. 26. Philipson EH, Kalhan SC, Edelbert SC, et al.
12. Morris MA, Grandis AS, Litton JC. Glycosy- Maternal obesity as a risk factor in gesta-
lated hemoglobin concentration III early tional diabetes. Am J Perinatol 1985;2:4.
gestation associated with neonatal outcome. 27. Wolfe HM, Zador IE, Gross TL, et al. The
Am J Obstet Gynecol 1985;153:651. clinical utility of maternal body mass in-
13. Willman SP, Leveno KJ, Guzick DS, et al. dex in pregnancy. Am J Obstet Gynecol
Glucose threshold for macrosomia in preg- 1991;164:1306.
nancy complicated by diabetes. Am J Obstet 28. Kirz DS, Dorchester W, Freeman RK.
Gynecol 1986;154:470. Advanced maternal age. The mature gravi-
14. Combs CA, Gunderson E, Kitzmiller JL, et da. Am J Obstet GynecoI1985;152:7.
al. Relationship of fetal macrosomia to 29. Elliott JP, Garite TJ, Freeman RK, et al.
maternal postprandial glucose control during Ultrasonic prediction of fetal macro-
pregnancy. Diabetes Care 1992;15:1251. somia in diabetic patients. Obstet Gynecol
15. Langer 0, Brustman L, Anyaegbunam A, 1982;60:159.
et al. The significance of one abnormal 30. Tamura RK, Sabbagha RE, Depp R, et al.
glucose tolerance test value on adverse Diabetic macrosomia-accuracy of third
outcome in pregnancy. Am J Obstet Gynecol trimester ultrasound. Obstet Gynecol
1987;157:758. 1986;67:828.
16. Lindsay MK, Graves W, Klein L. The rela- 31. Keller JD, Metzger BE, Dooley SL, et al.
tionship of one abnormal glucose tolerance Infants of diabetic mothers with accelerated
test value and pregnancy complications. fetal growth by ultrasonography: are they all
Obstet Gynecol 1989;73:103. alike? Am J Obstet Gynecol 1990;163:893.
17. Tallarigo L, Giampietro 0, Penno G, et al. 32. Landon MB, Mintz MC, Gabbe SG. Sono-
Relation of glucose tolerance to compli- graphic evaluation offetal abdominal growth:
cations of pregnancy in nondiabetic women. Predictor of large-for-gestational-age infant
N Engl J Med 1986;315:989. in pregnancies complicated by diabetes
8. Fetal Macrosomia 123

mellitus. Am J Obstet Gynecol 1989;160: sion: a randomized trial. Obstet Gynecol

115. 1991;77:659-663.
33. Shepard MJ, Richards VA, Berkowitz RL, et 47. Acker DB, Benjamin PS, Friedman EA. Risk
al. An evaluation of two equations for pre- factors for shoulder dystocia. Obstet Gynecol
dicting fetal weight by ultrasound. Am J 1985;66:762.
Obstet GynecoI1982;142:47. 48. Acker DB, Gregory KD, Sachs BP, et al. Risk
34. Hadlock FB, Harrist RB, Sharman RS, et al. factors for Erb-Duchenne palsy. Obstet
Estimation of fetal weight with the use of Gynecol 1988;71:389.
head, body, and femur measurements-a 49. Gordon M, Rich H, Deutschberger J, et al.
prospective study. Am J Obstet Gynecol The immediate and long-term outcome of
1985;151:333. obstetric birth trauma. I. Brachial plexus
35. Hirata GI, Medearis AL, Horenstein J, et al. paralysis. Am J Obstet Gynecol 1973;117:1.
Ultrasonographic estimation of fetal weight 50. Rubin A. Management of shoulder dystocia.
in the clinically macrosomic fetus. Am J JAMA 1964;189:11.
Obstet Gynecol 1990;162:238. 51. McFarland LV, Raskin M, Daling JR, et al.
36. Miller JM, Haywood LB, Oscar FK, et al. ErblDuchenne's palsy: a consequence of fetal
Ultrasonographic identification of the macrosomia and method of delivery. Obstet
macrosomic fetus. Am J Obstet Gynecol Gynecol 1986;68:784.
1988;158:1110. 52. Gross SJ, Shime J, Farine D. Shoulder
37. Bernstein 1M, Catalano PM. Influence of dystocia: predictors and outcome. A five-year
fetal fat on the ultrasound estimation offetal review. Am J Obstet GynecoI1987;156:334.
weight in diabetic mothers. Obstet Gynecol 53. Modanlou HD, Komatsu G, Dorchester W, et
1992;79:561. al. Large-for-gestational-age neonates:
38. Farmer RM, Medearis AL, Hirata GI, et al. anthropometric reasons for shoulder dysto-
The use of a neural network for the ultraso- cia. Obstet Gynecol 1982;60:417.
nographic estimation of fetal weight in 54. Winn HN, Grasso J, Holcomb W, et al. The
the macrosomic fetus. Am J Obstet Gynecol potential of fetal chest circumference in the
1992;166:1467. prenatal diagnosis of shoulder dystocia.
39. Oppenheim WL, Davis A, Growdon WA, et Proceedings of the 9th Annual Society of
al. Clavicle fractures in the newborn. Clin Perinatal Obstetricians Meeting, New
Orthop Relat Res 1990;250:176. Orleans, 1989.
40. Turner MJ, Rasmussen MJ, Turner JE, et al. 55. Benedetti TJ, Gabbe SG. Shoulder dystocia: a
The influence of birth weight on labor in complication of fetal macrosomia and pro-
nulliparas. Obstet Gynecol 1990;76:159. longed second stage of labor with midpelvic
41. Amico JA, Seitch J, Robinson AG. Studies delivery. Obstet Gynecol 1978;52:526.
of oxytocin in plasma of women during 56. Hopwood HG. Shoulder dystocia: fifteen
hypocontractile labor. J Clin Endocrinol & years' experience in a community hospital.
Metab 1984;58:274. Am J Obstet Gynecol 1982;144:162.
42. Seitchik J, Castillo M. Oxytocin augmenta- 57. Smeltzer JS. Prevention and management
tion of dysfunctional labor. I. Clinical data. of shoulder dystocia. Clin Obstet Gynecol
Am J Obstet Gynecol 1982;144:899. 1986;29:2.
43. Baxi LV, Petrie RH, Caritis SN. Induction of 58. Woods CE. A principle of physics as appli-
labor with low-dose prostaglandin F 2a and cable to shoulder delivery. Am J Obstet
oxytocin. Am J Obstet Gynecol 1980;136:28. Gynecol 1943;45:796.
44. Seitchik J, Castillo M. Oxytocin augmen- 59. Gonik B, Stringer CA, Held B. An alternate
tation of dysfunctional labor. III. Mul- maneuver for management of shoulder
tiparous patients. Am J Obstet Gynecol dystocia. Am J Obstet Gynecol 1983;145:7.
1983;145:777. 60. Sandberg ED. The Zavanelli maneuver: a
45. Blakemore KJ, Qin NG, Petrie RH, et al. potentially revolutionary method for the
A prospective comparison of hourly and resolution of shoulder dystocia. Am J Obstet
quarter-hourly oxytocin dose increase inter- Gynecol 1985;152:479.
vals for the induction of labor at term. Obstet 61. Delpapa EH, Mueller-Heubach E. Preg-
Gynecol 1990;75:757-761. nancy outcome following ultrasound diag-
46. Mercer B, Pilgrim P, Sibai B. Labor induc- nosis of macrosomia. Obstet Gynecol 1991;
tion with continuous low-dose oxytocin infu- 78:340.
124 H.N. Winn and J.C. Hobbins

62. Flamm BL, Goings JR. Vaginal birth after 95. Washington, DC: American College of
cesarean section: is suspected fetal macro- Obstetricians and Gynecologists, 1986.
somia a contraindication? Obstet Gynecol 65. Marchick R. Antepartum external cephalic
1989;74:694. version with tocolysis: a study of term single-
63. Langer 0, Berkus MD, Huff RW, et al. ton breech presentations. Am J Obstet
Shoulder dystocia: Should the fetus weighing Gynecol 1988;158:1339.
>4000 grams be delivered by cesarean 66. Morrison JC, Myatt RE, Martin IN, et al.
section? Am J Obstet Gynecol 1991;165:83l. External cephalic version of the breech
64. American College of Obstetricians and presentation under tocolysis. Am J Obstet
Gynecologists. Management of the breech Gynecol 1986;154:900.
presentation. ACOG Technical Bulletin No.
9
Twin Gestation and Multiple Births
CORNELIA R. GRAVES and FRANK H. BOEHM

The incidence of twins, triplets, and high- preeclampsia, placental abnormalities, and
order multifetal pregnancies has increased preterm labor. 6
significantly since the introduction of arti- There is consensus regarding the safety of
ficial stimulation of ovulation and in vitro vaginal delivery of twins when both twins
fertilization. 1 These pregnancies are at a are vertex, regardless of gestational age. 7
higher risk than singletons for perinatal Cesarean section should be reserved for
morbidity and mortality. Although much usual indications, that is, fetal stress or dis-
has been written about the mode of delivery tress and cephalopelvic disproportion.
for these gestations, there is little consensus
over whether cesarean section improves the
outcome. Despite little evidence to support Management of the Delivery of
cesarean section for multiple gestations, the Vertex/Nonvertex Twins
cesarean section rate has increased from 3%
in 1972 to 51% in 1985. 2 The cesarean rate Cesarean section has been recommended by
for triplets increased from 16% in 1977 some investigators as the mode of delivery
to almost 94% in 1988. 3 The purpose of in vertex/nonvertex presentations to avoid
this chapter is to discuss the management hazardous breech extraction of the second
of multiple gestations and to make re-. twin. 8 During the past several years numer-
commendations concerning possible modes ous studies have suggested, however, that
of delivery. the second, nonvertex twin weighing more
than 1500 g can be delivered vaginally with
as much safety as by cesarean section. 9 In
1986, Rabinovici et al. 1O managed 60 twin
Management of Delivery of deliveries after the 35th gestational week
Vertex/Vertex Twins with vertex-breech and vertex-transverse
presentations according to a randomized
Twin gestations make up approximately 1 % protocol; 33 patients were allowed to under-
of all pregnancies. 4 Although dizygotic go vaginal delivery, and 27 were delivered
twinning varies with race, age, parity, and by cesarean section. There were no signi-
,other factors, the monozygotic twinning rate ficant differences between 1- and 5-min
is constant, occurring at a rate between Apgars or the incidence of neonatal mor-
3 and 4 per 1000 births. 5 Morbidity and bidity between the second twin in both study
mortality are increased remote from de- groups. As expected, maternal febrile mor-
livery. Complications include spontaneous bidity was significantly higher in the cesa-
abortion, intrauterine growth retardation, rean section group.

125
126 C.R. Graves and F.R. Boehm

Because vaginal breech delivery is not 2. If the patient fails to progress in labor,
stressed in many training programs, exter- an intrauterine pressure catheter is inserted
nal cephalic version can be attempted after and oxytocin augmentation is begun.
delivery of the first twin to convert the 3. Epidural anesthesia is advised. This
second twin to a vertex presentation so as to allows for a comfortable labor and delivery
allow for vaginal delivery. Tchabo and as well as being an effective method of
Tomai 11 reported 11 of 12 conversions from pain control should fetal manipulation or
transverse to vertex and 16 of 18 successful cesarean section be needed.
breech conversions. No complications were 4. Ultrasound is used to note the position
noted in this study. Other authors, however, of the second twin, as well as for guidance
have found increased maternal and neonatal and monitoring of the second twin.
morbidity using external version. 12 Wells et 5. After the first twin delivers, an
al. 13 noted that patients undergoing exter- attempt is made to convert a transverse
nal version were more likely to undergo lie or breech presentation into a vertex
abdominal deli very for the second twin than presentation.
those in which a breech extraction was per- 6. If this cannot be accomplished, the
formed. The external version candidates second twin is then delivered by complete
also had a higher incidence of emergent breech extraction.
anesthesia for cord prolapse and persistent In a review of 50 vertex/nonvertex de-
fetal bradycardia. iveries of twins more than 1500 g at
Recent studies have also supported that Vanderbilt University, 55% were delivered
vaginal delivery of the nonvertex second vaginally versus 45% by cesarean. Out-
twin is safe. Fisherman et al. 14 in 1993 comes of both first and second twins, as
published a retrospective study that re- defined by Apgar scores at 1 and 5 min
viewed 781 twins at 20 weeks or more. Of and NICU admissions, were not statis-
the 309 liveborn vaginal second twins, 207 tically different.
were vertex and 183 were breech. There A dilemma in obstetrics is how to manage
were no significant differences in any of the the second twin when the birthweight is less
neonatal outcome measures, including than 1500 g. In the Fisherman study,14 no
Apgar scores at 5 min and neonatal inten- conclusion could be reached for twins under
sive care unit (NICU) admissions when , 1500 g. Adam et al. 15 concluded that there
stratified by birthweight. was increased perinatal morbidity for the
Ifvaginal delivery of the nonvertex second second twin with a birthweight of less than
twin is to be undertaken, there must be a 1500 g delivered vaginally, with morbidity
skilled obstertrician available who has been being measured in terms of intraventricular
trained in the techniques of internal and hemorrhage (lVH) , seizures, respiratory
external version, an anesthesiologist know- distress syndrome (RDS), and Apgar scores
ledgeable about uterine relaxation tech- less than 7. ,.,In addition, Barrett et al. 16
niques, and the ability to convert to an retrospectively evaluated the relationship
operative delivery mode within minutes. between type of delivery and perinatal
There should also be a second assistant morbidity and mortality in twins of birth-
skilled in ultrasonography and able to assist weight less than 2000 g, and found that
with delivery if needed. vaginally delivered second twins weighing
The Vanderbilt U ni versity Hospital Labor between 601 and 999 g had an increased risk
and Delivery guidelines for vertex/non- of neonatal mortality when compared to
ertex twins weighing more than 1500 g are their siblings. The study also found that,
as follow: among twins who weighed 1000-1499 g,
1. Amniotomy is performed when it can vaginally delivered second twins had signi-
be accomplished and internal electrode ficantly lower Apgar scores and increased
devices are placed. risks of neonatal morbidity in comparison to
9. Twin Gestation and Multiple Births 127

their siblings, whereas second twins who overlie each other. Some believe that in
were delivered by cesarean section had no carefully selected cases and in circum-
difference in Apgar scores or morbidity from stances in which the first twin is nonvertex,
those of their siblings. The authors con- vaginal delivery may be safely undertaken.
cluded that cesarean section was the optimal Many authors agree, however, that cesarean
route of delivery of all twins expected to section is recommended if the first twin
have a birth weight ofless than 1500g. 16 is nonvertex, because there are no studies
Rydstrom and Ingemarrson,17 however, in documenting the safety of vaginal delivery
1991 failed to reveal any significant impact in this groUp.21
of abdominal birth on fetal outcome for low
birthweight twins, even when fetal presen-
tation was taken into consideration. Greig et Management of Monoamniotic
a1. 4 in 1992 also failed to find support for Twin Gestations
routine cesarean section of the nonvertex
twin, regardless of birthweight. It has been assumed in the foregoing discus-
In 1985, Chervenak et a1. 9 outlined a sion that the pregnancies being managed
protocol for the management of twin de- were diamniotic. Monoamniotic twinning is
liveries. During the intrapartum period, an uncommon event widely recognized to
fetal weight is assessed by ultrasound. have increased antenatal and perinatal
If the estimated fetal weight is less than morbidity and mortality. Double survival is
2000 g, external version after delivery of rare. 22
the first twin is attempted if the second While most authors are of the opinion that
twin is nonvertex; if this is unsuccessful, these pregnancies should undergo cesarean
cesarean section is performed. If the second delivery to prevent cord entanglement and
twin weighs more than 2000 g, external ver- the risk offetal interlocking,22,23 Tessen and
sion is attempted; however, if unsuccessful, Zlatnick24 reported no fetal deaths and only
a vaginal breech delivery is attempted. 1 case of fetal distress requiring emergent
While there are few data to support cesarean section in 20 cases of labor and
cesarean section for the nonvertex second delivery of monoamniotic twins. Prophy-
twin weighing less than 1500 g, it is believed lactic preterm delivery has been advo-
by many perinatologists that because of cated by 32 weeks to prevent cord-related
possible entrapment of the aftercoming deaths;22 however, recent data reveal little
head, and a possible increased incidence of support for early intervention. 24
IVH and traumatic delivery, cesarean sec-
tion should be performed. 18,19 It should be
emphasized that if cesarean section is per-
formed, an adequate uterine incision, pre- Triplets and Other Multiple
ferably a low vertical incision that can be Gestations
extended into a classical incision, is advised
to avoid birth trauma. 9 Naturally occurring multiple births of more
than two infants are rare;25 however, the
discovery of ovulation induction agents has
Management of the Nonvertex- significantly increased their incidence.
Presenting Twin Virtually no guidelines exist in the United
States for optimal intrapartum care. Be-
The incidence of interlocking twins is ap- cause preterm labor in these patients is a
proximately 1 in every 1000 twin deliveries, frequent complication, cesarean section is
with a fetal mortality rate of 31%.20 This undertaken based on the gestational age of
condition occurs exclusively in breech- the infants.26 Loucopoulos et a1. 27 in 1982
vertex presentations when the chins directly reviewed their experience' wIth multifetal
128 C.R. Graves and F.H. Boehm

pregnancies. Cesarean section was per- 3. Creinin M, Katz M, Laros R. Triplet preg-
formed in 42% while the vaginal route was nancy: changes in morbidity and mortality.
the mode of delivery in 58%. Most infants J Perinatol 1991;11:207.
were delivered by breech extraction (32%) or 4. Greig P, Veille J-C, Morgan T, Henderson L.
spontaneously (27%). Cesarean section The effect of presentation and mode of de-
improved only the I-min Apgar for the third livery on neonatal outcome in the second
twin. Am J Obstet 1992;167:90l.
neonate. Ron-El et a1. 28 noted that fetal
5. Marivate M, Norman RJ. Twins. Clin Obstet
outcome was similar in the vaginal and Gynaecol 1982;9:783.
cesarean section deliveries among each 6. Kohl SG, Casey G. Twin gestation. Mt Sinai
gestational age group in their review of J Med 1975;42:523.
triplet and quadruplet pregnancies. 7. Chervenak FA. The controversy of mode of
More recently, Creinin et al} after re- delivery in twins: the intrapartum manage-
viewing the outcome of 13 sets of triplets ment of twin gestation (Part II). Semin
in a retrospective analysis, recommend Perinatol 1986;10:44.
cesarean section as the optimal mode of 8. Cetrulo CL. The controversy of mode of de-
delivery. Because of potential problems with livery in twins: The intrapartum manage-
malpresentation, changes in presentation ment of twin gestation (Part I). Semin
Perinatol 1986;10:39.
that can occur in active labor and delivery,
9. Chervenak F, Johnson R, Voucha S, Hobbins
the increased risk of cord prolapse, and J, Berkowitz R. Intrapartum management of
complications with fetal monitoring, most twin gestation. Obstet Gynecol 1985;65:119.
authors recommend cesarean section. In- 10. Rabinovici J, Barkai G, Reichman B, Serr
deed, since 1981 the trend has been toward DM, Mashiach S. Randomized management
cesarean section. However, some authors of the second nonvertex twin: vaginal de-
still advocate vaginal delivery as an option. livery or cesarean section. Am J Obstet
Itzkowic 29 and Deale and Cronje30 observed Gynecol 1987;156:52.
that a gestational age of more than 34 weeks 11. Tchabo JG, Tomai T. Selected intrapartum
and an estimated fetal weight of 2000 g were external cephalic version of the second twin.
Obstet Gynecol 1992;79:42l.
favorable for vaginal delivery.
12. Trofalter KF. Management of delivery. Clin
As there are no randomized trials for Perinatol 1988;15:93.
cesarean section versus vaginal delivery in 13. Wells S, Thorp J, Bowes W. Management of
triplet and high-order multiple deliveries, it the second nonvertex twin. Surgery 1991:172.
is the author's opinion that these preg- 14. Fisherman A, Grubb D, Kovacs B. Vaginal
nancies should be delivered by cesarean delivery of the nonvertex second twin. Am J
section, except for special circumstances. Obstet Gynecol 1993;168:86l.
If vaginal delivery is undertaken, the ob- 15. Adam C, Allen AC, Baskett TF. Twin de-
stetrician should be skilled in version livery: influence of presentation and method
maneuvers, and the capacity to proceed of delivery on the second twin. Am J Obstet
with immediate cesarean section should be Gynecol 1991;165:23.
16. Barrett J, Staggs SM, Van Houydonk JE,
available.
Growden JH, Killam AP, Boehm F. The
effect of type of delivery upon neonatal out-
come in premature twins. Am J Obstet
References Gynecol 1982;143:360.
17. Rydstrom H, Ingemarrson I. A case control
1. Holberg G, Biale Y, Lewenthal H, lnsler V. study of the effects of birth by cesarean sec-
Outcome of pregnancy in 31 triplet gesta- tion on intrapartum and neonatal mortality
tions. Obstet Gynecol 1982;59:472. among twins weighing 1500-2400 gm. Br J
2. Bell D, Johansson D, McLean FH, Usher RH. Obstet Gynaecol 1991;98:249.
Birth asphyxia, trauma and mortality in 18. Cetrulo CL, Ingardia CL, Sbarra AJ. Man-
twins: has cesarean improved outcome? Am J agement of multiple gestation. Clin Obstet
Obstet Gynecol 1986;167:901. Gynecol 1990;23:533.
9. Twin Gestation and Multiple Births 129

19. Kelsick F, Minkoff H. Management of the 25. James IN, Berbos BF, King A. Quintuplet
breech second twin. Am J Obstet Gynecol pregnancy. JAMA 1964;188:113.
1982;144:873. 26. Caspi E, Ronen J, Schreyer P. The outcome of
20. Khanda S. Locked twins. Obstet Gynecol pregnancy after gonadotropin therapy. Br J
1972;39:453. Obstet Gynaecol 1976;83:967.
21. Gabbe S, Neibyl J, Simpson J. Obstetrics, 27. Loucopoulos A, Jewelewic Z. Management
normal and abnormal pregnancies, 2nd Ed. of multifetal pregnancies: sixteen years'
New York: Churchill Livington, 1991. experience at the Sloan Hospital for Women.
22. Rodis JF, Vintzileos AM, Campbell DJ. Am J Obstet Gynecol 1982;143:902.
Antenatal diagnosis and management of 28. Ron-El R, Eliahu C, Schreyer P, Weinraub Z,
monoamniotic twins. Am J Obstet Gynecol Arleli S, Goldberg M. Triplet and quad-
1987;157:1255-1257. ruplet pregnancies and management. Obstet
23. Carr SR, Aronson MP, Coustan DR. Survival Gynecol 1981;57:458.
rates of monoamniotic twins do not increase 29. Itzkowic D. A survey of 59 triplet preg-
after 30 weeks' gestation. Am J Obstet nancies. Br J Obstet Gynaecol 1979;86:
Gynecol 1990;163:719-722. 23.
24. Tessen J, Zlatnick F. Monoamniotic twins: 30. Deale CJC, Cronje HS. A review of 367
a retrospective controlled study. Obstet triplet pregnancies. S Afr Med J 1984;66:
Gynecol 1991;77:832. 92.
10
Genital Herpes: Contemporary
Management
ZANE A. BROWN

To discuss the consequences of genital Recurrent or Reactivation Infection

herpes during pregnancy, particularly as it
relates to the route of delivery, it is necess- Patients in whom the HSV viral type that is
ary to review the classification of this recovered from the lesion or genital tract is
infection. the same as the serotype of the HSV anti-
bodies present in their sera are said to have
recurrent or reactivation infection.
Infectious Stages
Primary First Episode Asymptomatic Shedding
A primary first episode is the initial gen- Asymptomatic shedding is defined as the
ital herpes virus infection in an individ- presence of infectious virus in the genital
ual who lacks antibodies to both HSV -1 and tract in the absence of either symptoms
HSV-2. perceived by the patient or lesions observed
by a trained observer.
It is important to point out that recent
N onprimary First Episode studies have shown that as many as 85%
A nonprimary first episode is the initial of first episodes, whether primary or non-
genital herpes virus infection in an indi- primary, are asymptomatic or at least
vidual who has antibodies to the heterologous have symptoms that are unrecognized and
strain of the virus. The most common therefore unreported by the patient. 4 - 6
example is an individual who has antibodies The remaining 15% show bilateral genital
to HSV -1 as a result of oro labial herpes lesions, moderate to severe local pain, dy-
earlier in life and is infected with HSV-2 suria, sacral paresthesia, tender regional
for the first time. Less commonly, an indi- lymph node enlargement, central nervous
vidual may have preexisting HSV-2 anti- system (eNS) symptoms, and the systemic
bodies and is infected with HSV-1 (either signs and symptoms of a viremia such as
orally or genitally). The designation non- fever and malaise. Similarly, approximately
primary first episode is justified by data 75% of HSV-2 antibody-positive individuals
suggesting that heterologous antibody not do not report genital lesions or a history
only reduces the risk of acquiring a herpes compatible with recurrent genital herpes. 7
infection, but modifies the severity of the However, with education, more than 50% of
initial episode. 1 - 3 these "asymptomatic" seropositive individ-

131
132 Z.A.Brown

uals are able to identify and report genital enzyme immunoassay and indirect immuno-
lesions. 8 The frequency, duration, and fluorescence assay unreliable and therefore
severity of symptomatic recurrences in- unusable in distinguishing HSV-1 from
creases as pregnancy progresses. 6 Recur- HSV-2 antibodies, even though most pro-
rences when symptomatic are usually much prietary laboratories report the results as
milder than either primary or nonprimary "titers" to HSV-1 and HSV-2. Assays such as
first episodes. The lesions last several days the Western blot, which use HSV type-
instead of weeks, are usually unilateral, and specific antigens, are 98% sensitive and
are infrequently associated with systemic 100% specific for detecting the presence of
signs and symptoms. HSV-2 antibodies even in the presence of
A first-ever recurrence occurring during HSV -1 antibodies, and would enable the
pregnancy (in an individual with long- practitioner to detect HSV-2 seroconversion
standing HSV-2 seropositivity), however, during pregnancy and identify the HSV-2
can have a spectrum of involvement ranging seropositive but asymptomatic or unaware
from very mild to a severe illness suggesting patient. Unfortunately, these type-specific
first-episode disease. Because of its broad assays are currently available only through
spectrum of presentation, it is unreliable, regional research facilities. With increased
particularly in pregnancy, to classify genital demand, it is hoped that these assays will be
herpes infection using clinical presentation offered by commercial laboratories.
alone. In a recent study of 29 women pre- In a recent study at the University of
senting in pregnancy with a first clinical Washington in which cultures of the labia
episode of genital herpes, 50% had bilateral and cervix were taken from almost 24,000
lesions, 40% had bilateral inguinal lym- women at the onset of labor, the rate of
phadenopathy and reported malaise, 15% asymptomatic shedding was 0.4% or about
reported a headache, and 10% reported 1.2% ofthe HSV-2-seropositive population. 12
myalgias. 9 However, after serotyping them When the genital secretions of a subset of
with a type-specific HSV serologic assay, these HSV culture-negative, asymptomatic
only 4 ofthe 29 (14%) were true first episodes HSV-2-seropositive women were examined
(2 primary and 2 nonprimary). The re- by a polymerase chain reaction (peR), 20%
mainder were first-ever recurrences among showed evidence of HSV-2 DNA. 13 None of
seropositive individuals. In addition, a the newborns were infected with the HSV.
detailed survey of recurrent genital sym- The significance of this finding is at present
ptoms is as inadequate and does not improve unclear. It may represent noninfectious viral
upon the simple question "Have you ever fragments or it may represent infectious
had genital herpes?" in eliciting a history virus in a titer that is below the level of
of genital herpes in HSV-2 seropositive pa- sensitivity of viral isolation techniques.
tients. 1o Therefore, the appropriate classi- Whatever the explanation for this obser-
fication of genital herpes depends on the vation, it is apparent that asymptomatic
accurate serotyping of the patient using shedding of the HSV at the onset of labor
type-specific serologic assays. probably occurs much more frequently than
has been appreciated in the past using viral
isolation techniques. It also implies that
Detection of Antibodies neonatal transmission of HSV-2 from a
seropositive mother during labor and deli very
Type-specific HSV serologic assays have is probably an exceedingly rare event.
been developed that detect antibodies di- Therefore, HSV is more likely to be present
rected against viral protein epitopes which in the genital tract at the onset oflabor as a
are specific for each viral type.l1 This avoids result of asymptomatic rather than symp-
the extensive cross-reactivity between HSV tomatic disease (whether first episode or
serotypes that renders the commonly used recurrent). It is a result of this asymptomatic
10. Genital Herpes: Contemporary Management 133

shedding of virus that almost all neonatal than that found in most other areas of the
infections occur. 12,14-17 It is difficult to country. Nationally, the reported rates of
understand the rationale for the cesarean neonatal herpes range from 1 case per 2000-
delivery of women with symptomatic recur- 10,000 live births. 12,21,23,24 Although the
rent genital herpes at the onset of labor mortality and morbidity of neonatal herpes
when most neonatal infections occur as a has decreased with the advent of antiviral
result of maternal asymptomatic shedding. chemotherapy, more than 40% of neonates
with the infection still die or are impaired. 25
While neonatal HSV is a relatively in-
Neonatal Herpes frequent occurrence, its impact on the de-
livery of obstetric care is extensive. Concern
Genital herpes is the most common ulcerative for exposing the newborn to infected genital
sexually transmitted disease in North secretions during labor and vaginal delivery
America. 18,19 The major concerns related to has led to very high rates of cesarean section
genital herpes infection during pregnancy with its attendant increase in costs, mor-
are HSV transmission to the infant and the bidity, and mortality. In a recent study
cesarean section rate among women who are in the three most populous counties in
beginning pregnancy with symptomatic Washington State, the obstetric records of
recurrent genital herpes. The incidence of all pregnant women identified from birth
neonatal herpes appears to increase at least certificates as having genital herpes were
in some populations probably as a result reviewed and matched to pregnant women
of parallel increases of the infection in without genital herpes delivering at the
the adult population. 2o ,21 In SeattlelKing same time and in the same hospital as the
County, Washington between 1966 and index case. 22 Women with recurrent symp-
1985, the incidence of neonatal herpes rose tomatic genital herpes (n = 909) had five
from 3.2 to approximately 15 cases per times as many cesarean sections as women
100,000 live births.14 This occurred in spite without genital herpes (75% versus 15%)
of a cesarean section rate over the same (Table 10.1). Of the women with a history of
period of time approaching 70% for women genital herpes, 80% were without lesions at
entering pregnancy with symptomatic delivery and more than 50% were without
recurrent genital HSV-2.22 Since 1985, the lesions within 7 days of delivery; only 1.3%
rates have decreased to about 11 cases per were culture positive at the time of delivery.
100,000 live births and have remained at Thus, most did not have active genital herpes
that level to the present. Rates of neonatal at the onset of labor but were still delivered
herpes, however, vary widely throughout by cesarean section. Associated with the
the United Sates and Europe. For example, increased cesarean section rate was a 2.7-
the current rate of neonatal HSV at the fold increase in postpartum endometritis.
University of Washington is approximately We can make a rough and conservative
1 in 1800 live births, a rate that is higher estimate ofthe impact of this excess cesarean

TABLE 10.1. Route of delivery among women with and without genital herpesa
Lesion at less than
First episode History of recurrence 1 week from delivery No Hx HSV
(n = 71) (n = 650) (n = 290) (n = 917)

Primary cesarean section

HSV indication 46.5 45.4 78.6 o
Other indication 14.1 13.1 9.7 10.3
Repeat cesarian section 1.4 4.0 4.8 7.6
Vaginal delivery 38.0 37.5 6.9 82.1

aFrom Wolf. 22 Data are percentages.

134 Z.A.Brown

section rate nationally by making several as determined by weekly HSV cultures does
assumptions. If we assume a delivery rate in not predict whether the HSV will be present
the United States of approximately 3 million in the genital tract at the onset oflabor. 27 In
births per year and an incidence of symp- addition, the duration of asymptomatic
tomatic genital herpes of 5% in the repro- shedding is brief, seldom more than 3 days. 6
ductive-aged population, then 150,000 Recent studies have demonstrated that
women will deliver annually with a history most of the pregnancy morbidity and the
of symptomatic recurrent genital herpes. majority of cases of the neonatal herpes are
If we further assume a baseline cesarean not the result of reactivation of genital
section rate of 20%, then the excess cesarean herpes in women at the onset of labor, but
section rate of 50% attributable to genital result from women acquiring genital herpes,
herpes will result in about 75,000 cesarean frequently asymptomatically, in late preg-
sections for the indication of genital herpes. nancy.12,28,29 This is consistent with the
If we assume that a cesarean section costs observations that women with first-episode
approximately $3000 more than a vaginal disease are devoid of type-specific homo-
delivery, then the national costs for these logous antibody with which to transplacent-
excess cesarean sections resulting from ally protect their offspring and that they
recurrent genital herpes may approximate excrete HSV from the cervix in higher titers
$225 million annually. To this must be added and for longer periods of time than do
the indirect costs, such as the increased women with recurrent genital herpes. In a
recovery time and increased time lost from recent study performed at the University of
work, the cost of repeat cesarean sections, Washington, 12 herpes cultures were obtained
and the myriad of social consequences to the from the cervix and labia of 24,887 asym-
immediate family and the community. Thus, ptomatic women within 48 h of delivery
current obstetric practice has not only failed (Figure 10.1). Of these, 97 patients (0.4%)
to reduce the incidence of neonatal herpes, demonstrated asymptomatic shedding as a
but has significantly increased the rate of result of either first-episode (n = 24) or
cesarean section and complications among recurrent disease (n = 73). Of the 24 patients
women entering pregnancy with a history of with first-episode disease, 6 had primary
recurrent genital herpes. 26 while 18 had nonprimary first-episode infec-
Until recently, strategies to prevent tions. Three of the 6 mothers (50%) with
maternal fetal transmission centered about asymptomatic shedding as a result of primary
pregnant women with recurrent HSV-2 with disease and 5 of the 18 mothers (24%) with
the assumption that it was this population of nonprimary first-episode disease transmitted
women who by reactivating their genital the infection to their newborns. Therefore, 8
herpes near parturition were at greatest risk of24 mothers (33%) with first-episode disease
of transmitting the infection to their infants. transmitted the infection to their infants,
It was thought that women whose disease with two neonatal deaths. In contrast, ofthe
was reactivated in the last several weeks 73 women with asymptomatic shedding
before the onset of labor were at risk of secondary to recurrent disease, only 2 (3%)
having infectious virus persist in their infected their infants. Of interest, of the
genital tracts at the onset of labor and were 73 women with asymptomatic shedding
therefore delivered by cesarean section. This secondary to recurrent genital herpes, 67
thinking led to the practice of weekly HSV were HSV-2 and 6 were HSV-l. None ofthe
antepartum cultures from 34 weeks to term, 67 infants exposed to HSV-2 were infected
and was in large part responsible for the whereas 2 of the 6 infants (30%) exposed to
high rates of cesarean section in women HSV-1 became infected. Therefore, it would
with symptomatic recurrent genital herpes. appear that HSV-1 transmits significantly
A recent study has demonstrated that ante- more readily to the newborn than HSV-2.
partum asymptomatic shedding of the HSV However, recent studies have demonstrated
10. Genital Herpes: Contemporary Management 135

24,887 Cultures Obtained Within 48 Hours of Delivery

I
97 (0.4%) with Asymptomatic Shedding
I
1 1
24 (25%) First Episode HSV 73 (75%) Reactivation
1 I
I. 1 1
PrUhary Non-Primary History of HSV No history HSV
6 (25%) 18 (75%) 35 (48%) 38 (52%)
I I I I
3 Cases 5 Cases 2 Cases No Cases
Neonatal HSV Neonatal HSV Neonatal HSV Neonatal HSV
(1 death) (1 death)

FIGURE 10.1. Asymptomatic shedding of genital HSV during labor. (From Brown et al.,12 by per-
mission of the New England Journal of Medicine.)

that when infection with HSV-l occurs, the in the genital secretions of asymptomatic
neurodevelopmental consequences are sig- mothers at the onset oflabor; most commonly
nificantly less severe than those from infec- as a result of maternal first-episode disease
tions caused by HSV-2.25 One study12 also acquired in the latter half of pregnancy.
demonstrated that the use of fetal scalp elec- Even if every patient entering pregnancy
trodes in laboring mothers with a history of with symptomatic recurrent genital herpes
recurrent genital herpes was a risk factor was delivered by cesarean section, only a
for neonatal herpes. very small number of the cases of neonatal
In a companion study of 5742 patients, herpes would be prevented. It would not
sera for HSV antibodies were obtained at prevent those cases caused by asymptomatic
the first prenatal visit and again at the onset reactivation at the onset of labor among
of labor. 29 Among women who were HSV HSV-2 seropositive women without know-
seronegative at the first prenatal visit, the ledge of their disease or maternal first-
rate of HSV-2 seroconversion was 2.5% episode disease in late pregnancy with
during the mean observation period of 28 asymptomatic shedding at the onset oflabor.
weeks. When the mean interval of observa- Cesarean section may even decrease the
tion was adjusted for the entire pregnancy, transplacental transmission of protective
the seroconversion rate was 3.7%. In a small anti-HSV antibodies. 32 In addition, approxi-
subset of these patients in which partner mately 20%-30% of newborns who develop
sera were also obtained as part of a pilot neonatal HSV are delivered by cesarean
study, the rate of seroconversion for HSV section; ofthese, about one-fourth had intact
seronegative women with HSV -2 seropositive fetal membranes?1 Furthermore, about 5%
partners was 33%.30 In addition to the sig- of infants with neonatal HSV are born with
nificantly increased risk of neonatal herpes, the disease, suggesting that the infection
as has been described, women who HSV-2 was acquired in utero.33
seroconvert during pregnancy (whether
symptomatically or asmptomatically) have
an increased risk oflate fetal loss, premature Possible Strategies to Reduce
labor, and intrauterine growth retardation. 28
It is important to point out that many the Rate of Excess Cesarean
cases of neonatal herpes cannot be pre- Sections for Genital Herpes
vented, even with the best care. 31 From the
foregoing, it is evident that infants acquire First, the most important strategy of re-
their infection by contacting infectious virus ducing cesarean delivery would be to restrict
136 Z.A.Brown

the use of cesarean section to women with 8 h are comparable to those expected for
lesions or symptoms of a recurrence during nonpregnant woment. 37 Surprisingly, the
labor. Currently, most of the cesarean increased volume of distribution of pregnant
sections performed for genital herpes are for women and the increased renal blood flow
recurrences "close to" but not at the onset in late pregnancy do not alter the steady-
of labor. Cesarean delivery should be dis- state pharmacokinetics as would have been
couraged as a prophylactic measure in the expected. Acyclovir is highly concentrated
absence oflesions or symptoms. HSV cultures in the amniotic fluid and milk 38 but does
obtained during the antepartum period have not accumulate in the fetus. 38 The mean
not been shown to be predictive of the status maternal: infant plasma ratio at delivery
of the cervix at the onset of labor, and HSV is 1.3.
cultures obtained from the maternal geni- A number of studies have demonstrated
talia on admission in early labor have a poor the efficacy of acyclovir in suppressing
sensitivity for predicting infants who will symptomatic recurrences in nonimmuno-
subsequently develop neonatal herpes. 12 suppressed, nonpregnant adults. 39 However,
Therefore, HSV cultures either during preg- asymptomatic shedding of the virus may
nancy or at the onset of labor are of little continue in spite of clinically effective sup-
clinical use and are labor intensive to obtain pression of symptomatic recurrences. 40 In
as well as expensive and uncomfortable for pregnancy, preliminary data from several
the patient. Patients with a history of re- prospective trials currently in progress
current genital herpes admitted in early suggest that acyclovir is effective in sup-
labor should have their external genitalia pressing symptomatic recurrences. 41 - 42 As
thoroughly examined with a bright light on in the nonpregnant woman, asymptomatic
an examination table. All too often, a pelvic shedding may continue with a risk of ne-
exam, if it takes place, occurs on a labor bed onatal infection. 4o Until the prospective
in the muted light of a modern labor room. studies of prophylactic acyclovir in late
In addition, patients and partners should be pregnancy that are now in progress have
questioned about the presence of symptoms been completed and evaluated for safety and
suggesting recurrent genital herpes even if efficacy, routine prophylactic use of acyclovir
there is no history of genital herpes. in late pregnancy cannot be recommended.
Second, as a result of studies in progress, Third, a rapid test for HSV antigen could
acyclovir may be shown to suppress symp- be used at the onset of labor to identify those
tomatic reactivation of genital herpes at patients with HSV in their genital tracts.
term and therefore obviate the need for ce- Theoretically, this would permit a more
sarean section. selective and specific use of cesarean section
Acyclovir is a nucleoside analog whose to prevent contact between genital secretions
selective activity inhibits the replication of containing infectious virus and the fetus
HSV at concentrations as much as 3000 fold during parturition.
less than those that inhibit mammalian cel- Such a rapid test is not currently available
lular functions. Because of its remarkable and will not become available in the foresee-
specificity for cells already infected with able future for several reasons. It would
HSV, it has proven remarkably safe in have to be available within several hours
mammalian fetal test systems. 34 Several and have a specificity of greater than 99.5%
large retrospective series have been un- for the presence of infectious virus to avoid a
able to demonstrate any evidence of fetal substantial increase in unnecessary cesarean
teratogenicity or toxicity.35,36 Even though sections. This type of specificity may be
acyclovir is cleared by the kidney, recent impossible to achieve, because antigen
pharmacokinetic studies demonstrated that detection systems will detect some nonin-
the steady-state acyclovir levels of pregnant fectious cases in which viral fragments
women for doses of 200 mg or 400 mg every or noninfectious virus are present in the
10. Genital Herpes: Contemporary Management 137

specimen. 31 The "rapid" tests currently from symptomatic and asymptomatic source
available use monoclonal antibodies that contacts. Sex Transm Dis 1985;12:33-39.
bind to HSV proteins. They were designed to 5. Boucher FD, Yasukawa LL, Bronzan RN,
be used with material from lesions, and lack Hensleigh PA, Arvin AM, Prober CG. A pro-
the sensitivity to separate low levels of virus spective evaluation of primary genital herpes
from background activity. simplex virus type 2 infections acquired
during pregnancy. Pediatr Infect Dis J 1990;
9:499-504.
6. Brown ZA, Vontver LA, Benedetti J, et al.
Summary Genital herpes in pregnancy: risk factors
associated with recurrences and asymp-
The major consequence ofentering pregnancy tomatic viral shedding. Am J Obstet Gynecol
with a diagnosis of recurrent genital herpes 1985;153:24-30.
is the increased risk of cesarean section. 7. Koutsky LA, Stevens CE, Holmes KK, et al.
Most cases of neonatal herpes are caused by Underdiagnosis of genital herpes by current
first episodes of genital herpes acquired late clinical and viral-isolation procedures. N
in pregnancy. The most promising means of Engl J Med 1992;326:1533-1539.
reducing the extraordinary rates of cesarean 8. Langenberg A, Benedetti J, Jenkins J, Ashley
R, Winter C, Corey L. Development of clini-
section for genital herpes remains a careful
cally recognizable genital lesions among
history and examination on admission to the women previously identified as having
labor unit, with cesarean section being "asymptomatic" herpes simplex virus type 2
reserved for only those individuals who have infection. Ann Intern Med 1989;110:882-887.
lesions or symptoms suggestive of a recur- 9. Hensleigh P, Andrews W, Brown Z, et al.
rence. For women with frequent recurrences, Medical history and clinical presentation
the use of prophylactic acyclovir in late does not identify primary genital herpes
pregnancy is promising. However, safety infections in pregnant women. Presented
and efficacy have not as yet been proven. at the 15th Annual Meeting, Society of Per-
Most importantly, given the technology and inatal Obstetricians. January 23-28, 1995.
resources available in clinical practice today, Atlanta Hilton, Atlanta, Georgia.
10. Brown ZA, Benedetti J, Watts DH, et al.
most cases of neonatal herpes cannot be
A comparison between detailed and simple
prevented. histories in the diagnosis of genital herpes
complicating pregnancy. Am J Obstet
Gynecol 1994 (in press).
References 11. Ashley RA, Militonia J, Lee F, Nahmias
A, Corey L. Comparison of Western blot
1. Mertz GJ, Benedetti J, Ashley R, Selke S, (immunoblot) and glycoprotein G-specific
Corey L. Risk factors for the sexual trans- immunodot enzyme assay for detecting anti-
mission of genital herpes. Ann Intern Med bodies to herpes simplex virus types 1 and 2
1992;116:197 -202. in human sera. J Clin Microbiol 1988;26:
2. Bryson Y, Dillon M, Bernstein DI, Randolf J, 662-667.
Zakowski R, Garratty E. Risk of acquisition 12. Brown ZA, Benedetti J, Ashley R, Burchett
of genital herpes simplex virus type 2 in sex S, et al. Neonatal herpes simplex virus infec-
partners of persons with genital herpes: a tion in relation to asymptomatic maternal
prospective couple study. J Infect Dis 1993; infection at the time of labor. N Engl J Med
167:942-946. 1991;324:1247-1252.
3. Corey L, Holmes KK. Genital herpes simplex 13. Cone RW, Hobson AC, Brown ZA, et al.
virus infections: current concepts in diag- Frequent detection of genital herpes simplex
nosis, therapy and prevention. Ann Intern virus DNA by polymerase chain reaction
Med 1983;98:973-983. among pregnant women. JAMA 1994;272:
4. Mertz GJ, Schmidt 0, Jourden JL, et al. 797-796.
Frequency of acquisition of first-episode 14. Sullivan-Bolyai J, Hull HF, Wilson C, Corey
genital infection with herpes simplex virus L. Neonatal herpes simplex virus infection
138 Z.A. Brown

in King County, Washington: increasing in- simplex VIrus at delivery. N Engl J Med
cidence and epidemiologic correlates. JAMA 1991;324:1247-1252.
1983;250:3059-3062. 28. Brown ZA, Vontver LA, Benedetti J, et al.
15. Prober CG, Hensleigh P A, Boucher FD, Effects on infants of a first episode of genital
Yasukawa LL, Au DS, Arvin AM. Use of herpes during pregnancy. N Engl J Med
routine viral cultures at delivery to identify 1987;317:1246-1251.
neonates exposed to herpes simplex virus. N 29. Brown Z, Selke S, Zeh J, et al. HSV serocon-
Engl J Med 1988;318:887-891. version during pregnancy and its impact on
16. Whitley RJ, Corey L, Arvin A, et al. Changing pregnancy outcome and infant development.
presentation of herpes simplex virus infection In: Program and abstracts of the 32nd Inter-
in neonates. J Infect Dis 1988;158:109-116. science Conference on Antimicrobial Agents
17. Arvin AM, Yeager AS, Bruhn FW, Grossman and Chemotherapy, October 11-14, 1992,
M. Neonatal herpes simplex infection in the Anaheim, California.
absence of mucocutaneous lesions. J Pediatr 30. Brown ZA, Annholm A, Ashley R, et al.
1982;100:715-721. HSV serological discordancy among sexual
18. Genital herpes infection-United States, partners and rates of seroconversion during
1966-1984. Morbid Mortal Wkly Rep 1986; pregnancy. In: Program and abstracts of the
35:402-403. Infectious Disease Society for Obstetrics and
19. Quinn TC, Cates W. Epidemiology of sexually Gynecology, August 1993, Stowe, Vermont.
transmitted diseases in the 1990s. In: Quinn 31. Prober CG, Corey L, Brown ZA, et al. The
TC, ed. Sexually transmitted diseases. Raven, management of pregnancies complicated by
New York, 1992:1-37. genital infections with herpes simplex virus.
20. Nahmias AJ, Lee FK, Beckman-Nahmias S. Clin Infect Dis 1992;15:1031-1038.
Sero-epidemiological and -sociological pat- 32. Bujko M, Sulovic V, Sbutega-Milosevic G,
terns of herpes simplex virus infection in Zwanovic V. Mode of delivery and level of
the world. Scand J Infect Dis [suppl] 1990;69: passive immunity against herpes simplex
19-36. VIrus. Clin Exp Obstet Gynecol 1989;16:
21. Stone KM, Brooks CA, Guinan ME, Alex- 6-8.
ander ER. National surveillance for neonatal 33. Nahmias AJ, Keyserling HH, Kerrick GM.
herpes simplex virus infections. Sex Transm Herpes simplex. In: Remington JS, Klein
Dis 1989;16:152-156. JO, eds. Infectious diseases of the fetus and
22. Wolf M. Unpublished thesis. Obstetrical newborn infant, 2nd Ed. Philadelphia: WE
course and complications of women with Saunders, 1983:636-678.
symptomatic genital herpes: a population- 34. Gnann JW, Barton NH, Whitley RJ.
based analysis, 1990. Acyclovir: mechanism of action, pharmaco-
23. Garland SM. Neonatal herpes simplex: Royal kinetics, safety and clinical applications.
Women's Hospital 10 year experience with Pharmacotherapy 1983;3:275-283.
management guidelines for herpes in preg- 35. Kingsley S. Fetal and neonatal exposure to
nancy. Aust NZ J Obstet Gynecol 1992;32: acyclovir. Paris, 1986: unpublished data.
331-334. 36. Andrews EB, Yankaskas BC, Cordero JF,
24. Chuang T. Neonatal herpes incidence, pre- Schoeffler K, Hampp S, and the Acyclovir in
vention and consequences. Am J Prev Med Pregnancy Registry Advisory Committee.
1988;4:47 -53. Acyclovir in pregnancy registry: six years
25. Whitley R, Arvin A, Prober C, et al. Predictors experience. Obstet Gynecol 1992;79:7-13.
of morbidity and mortality in neonates with 37. Frenkel LM, Brown ZA, Bryson YJ, et al.
herpes simplex virus infections. N Engl J Pharmacokinetics of acyclovir in the term
Med 1991;324:450-454. human pregnancy and neonate. Am J Obstet
26. Randolph AG, Washington AE, Prober CG. Gynecol 1991;164:569-576.
Cesarean delivery for women presenting 38. LauRJ,EmeryMG,GalinskyRE. Unexpected
with genital herpes lesions. JAMA 1993;270: accumulation of acyclovir in breast milk.
77-82. Obstet Gynecol 1987;69:468-471.
27. Arvin AM, Hensleigh PA, Prober CG, et al. 39. Kaplowitz LG, Baker D, Gelb L, et al. Pro-
Failure of antepartum maternal cultures to longed continuous acyclovir treatment of
predict the infant's risk of exposure to herpes normal adults with frequently recurring
10. Genital Herpes: Contemporary Management 139

genital herpes simplex virus infection. JAMA Hall M, Wendel G. Prevention of cesarean
1991;265:747 -751. section for recurrent genital herpes simplex
40. Strauss SE, Seidlin M, TakiffHE, et al. Effect virus (HSV) using acyclovir suppressive
of oral acyclovir treatment on symptomatic therapy. Presented at the Society of Gyneco-
and asymptomatic virus shedding in recur- logic Investigation, Abstract #S223, 1993,
rent genital herpes. Sex Transm Dis 1989; Orlando, Florida.
16:107-113. 43. Haddad J, Langer B, Astruc D, Messer
41. Stray-Pedersen B. Acyclovir in late pregnancy J, Lokiec F. Oral acyclovir and recurrent
to prevent neonatal herpes simplex. Lancet genital herpes in late pregnancy. Obstet
1990;336:756. GynecoI1993;82:102-104.
42. Scott L, Jackson G, Sanchez P, Castaneda Y,
11
Methods for Safe Reduction of
Cesarean Section Rates
NORBERT GLEICHER, RICHARD H. DEMIR, JEANNE B. NOVAS,
and STEPHEN A. MYERS

This chapter is presented under the assump- led to an initial increase in popularity of
tion that cesarean section rates in the this surgery in the 1970s. A very dramatic
United States should be reduced. An un- improvement in the safety of the procedure,
derstanding of the reasons that lead obste- as a consequence of better anesthetic man-
tricians to perform cesarean sections should agement, blood banking, and antimicrobial
provide the necessary information needed to therapies of pregnant women, reduced many
effect such a change in practice patterns. of the most significant risks, lowering
Such a recommendation was already made the risk-benefit ratio in favor of surgical
by a National Consensus Development deliveries. When cesarean deliveries were
Conference in 1980, when the national rate reported to result in improvement in peri-
was still below 17%.1 Since then, it has natal outcome, the national cesarean section
further increased at an annual rate of rate was, however, still in the low single
approximately 1 % until a plateau of approx- digits. In 1970 only 5.5% of deliveries were
imately 24% was reached in the early 1990s by cesarean section. By 1987, this rate had
(Figure 11.1).2,3 Some physicians have increased to 24.4% (see Figure 11.1).2 At an
nevertheless challenged the effort toward a utilization level of 5.5%, cesarean delivery,
reduction in cesarean section rates,4,5 and a in selected high-risk situations, could be
level of general hesitance within the profes- expected to improve perinatal outcome. The
sion seems to remain. Efforts by a number of error consequently perpetrated was the
authoritative bodies6 - 8 seem to have had no assumption that a continuous benefit on
significant effect on reducing the rate. This perinatal outcome could be achieved if more
chapter thus needs to be seen against a and more indications for cesarean section
background of at least subliminal resistance deliveries were added. In the later 1970s it
toward efforts directed at the reduction in almost became a dictum in U.S. teaching
cesarean section rates. Any such reduction departments that cesarean section rates had
must therefore first and foremost consider to approach 20%. If that was not achieved,
the causes of this resistance. departmental quality assurance processes
Cesarean sections, like most other inva- kicked in and often led to a further expan-
sive procedures, are principally intended sion in cesarean section indications.
to improve medical outcome. Because It was during that decade of practice
pregnancy outcome relates to both mother that a mind-set was put into place which
and fetus, cesarean sections can improve principally suggested that one can never be
either maternal or fetal outcome or both. It wrong in performing a cesarean section. In
was the belief that cesarean section could contrast, the nonperformance of cesareans
beneficially affect perinatal outcome that was constantly subject to quality assurance

141
142 N. Gleicher et al.

30~----------------------------------------------------------~

25~---------------------------------------------

20 ~------------------------

15r-------------------

10 r-----------

1970 1975 1980 1985 1986 1987 1988 1989 1990

FIGURE 11.1. Cesarean section rates, 1970-1990, in the United States. Hatched bars, repeat; shaded
bars, primary; solid bars, total.

reviews. The trend toward increased cesar- psychologic impact of such a widely held
ean utilization was enhanced when a belief on attempts to reduce the national
presumption was made by the legal com- cesarean section rate. It is simply not
munity that the liberal performance of enough to tell the profession that all reason-
cesarean sections had become a new stan- able scientific evidence suggests that
dard of care. The average practitioner was medical-legal considerations do not sig-
now virtually forced to abandon the tradi- nificantly impact on physician behavior in
tional practice of obstetrics in favor of a regards to cesarean sections. Instead, one
cesarean section-driven approach. After all, has to provide the practicing physician with
"one never got sued for performing a cesar- the necessary reassurances that the non-
ean." However, even this dictum has since performance of a nonindicated cesarean
been refuted because even a cesarean sec- section will not increase their liability
tion cannot protect from legal exposure. An exposure.
attitude of practicing defensive medicine This reassurance can be provided at the
has in increasing fashion perpetrated the local level by establishing departmental
practice of obstetrics during the past two practice guidelines, which, if properly fol-
decades. Surveys of obstetricians have per- lowed, will represent an excellent defensive
sistently indicated that the profession con- tool should the need for one arise. On the
siders the current medical-legal situation national level one would welcome a similar
the principal contributing factor to the approach. Unfortunately, national organiza-
national cesarean section crisis. 1,g,lO tions have been slow and rather timid
Attempts to document such an impact in responding to the increase in national
through scientific investigation have largely cesarean section rates. While the previously
failed, however,11-13 although one recent mentioned Consensus Development Con-
study supported a possible positive associa- ference of 1980 already considered the
tion between malpractice claims risk and national cesarean section rates at that time
the rate of cesarean delivery.14 It is there- to be excessive, 1 the American College did
fore questionable whether cesarean section not publish a first policy change in regards
rates would, in fact, significantly decrease if to cesarean sections until 1984,6 and more
the profession's liability situation changed. up-to-date vaginal birth after cesarean
One can nevertheless not overlook the (VBAC) guidelines till 1988. 7 What or-
11. Methods for Safe Reduction of Cesarean Section Rates 143

ganized medicine apparently fails to alternative delivery methods. Cesarean

understand is the psychologic need of the sections then rather automatically became a
profession to be actively supported in the self-fulfilling prophecy. Either the practi-
attempt to reduce the national cesarean tioner has no choice but to perform a
section rate. This can best be done by giving cesarean (for lack of training in alternative
the practicing obstetrician clearly outlined approaches) or untrained individuals per-
recommendations for the performance of form alternative procedures with (expected)
cesarean sections. bad outcome. They then only falsely re-
The cesarean section is the most fre- emphasize the alleged benefit of cesarean
quently performed surgical procedure in the sections through their inexperience.
United States. 15 Like any other surgical One cannot expect individuals who are
intervention, it should be only performed to untrained or inexperienced with operative
improve (fetal or maternal) outcome. In the vaginal or vaginal breech deliveries to
age of meta-analyses, it appears relatively embrace such delivery modalities without
simple to define indications for cesarean further training. However, not even one
sections. 16 Almost 15 years after the Con- postgraduate course has addressed this lack
sensus Development Conference, 1 the only of training over the last few years. One can·
existing cesarean section guidelines refer to contrast this to the plethora of postgraduate
vaginal delivery after previous cesarean activities surrounding the concept of opera-
(VBACf and even those are timid and tive laparoscopy. A rather disturbing con-
do not reflect, for example, published ex- clusion is then reached as to where the
perience with more than one previous emphasis is placed within the profession in
cesarean section. 17- 20 There are still no regards to the training of manual skills.
clear guidelines for breech delivery, cesarean As a consequence, patients undergo
section for dystocia/cephalopelvic dispropor- cesarean sections because physicians lack
tion, or even fetal distress. How can one then the theoretical and practical skills to perform
be surprised if many obstetricians choose alternative procedures. Paradoxically, lack
the path of least resistance-cesarean of such skills is generally accepted as an
delivery. . indication for a cesarean section delivery.
The shift toward increased cesarean This stands in contrast to other obstetric
utilization also resulted in yet another situations, where an apparent lack of skills
(probably unexpected) consequence of con- usually results in a mandated referral to
siderable significance. A whole generation of a perinatologist. In most institutions a
physicians have departed their training vaginal breech at term or the need for a
programs not only indoctrinated with the possible forceps delivery are, however, not
wrong dogmas but also untrained in alterna- considered indications for perinatal con-
tive delivery methods. If the chairman of a sultation. This should not surprise because
major teaching department in 1992 still can there is considerable evidence that the
pride himself of "never having allowed the es~ablishment of perinatology as a sub-
vaginal delivery of a breech In his depart- specialty by itself contributed significantly
ment," then one can easily calculate the to the mind set that has given us the current
number of graduating residents who have cesarean section rates. 21
entered practice unable to deliver a vaginal The national cesarean section crisis can
breech, even if they wished to do so. The then be seen as the consequence of a highly
same is true for forceps deliveries and, successful education system which, during
probably, vacuum extractions. a span of approximately 20 years, has
A philosophy that sees a cesarean section indoctrinated a whole generation of obste-
delivery as the solution to basically every tricians in the use (and misuse) of cesarean
obstetric problem has thus also given us sections and, concomitantly, has created a
a generation of physicians untrained in generation of subspecialists who often
144 N. Gleicher et al.

reemphasize this practice pattern through organizations are already touting the cost
their own behavior. Because even many sub- savings from just stabilizing the national
specialists are not trained in alternative cesarean section rate. Because a reduction
delivery modalities, a vicious cycle has been in rate can impact on national health
created that tends to inhibit attempts to care costs quite significantly,3 considerable
affect the national cesarean section rate. efforts can be expected from both govern-
For example, newly published guidelines ment and the insurance industry in this
for the education of residents mandate area. We are well advised to prepare as a
exposure and participation in alternative profession and try to be leaders rather than
delivery methods to cesarean sections. 22 followers.
This effort should be applauded. Mandated In the remainder of this chapter we
educational efforts, however, do not work attempt to demonstrate how cesarean sec-
without proper faculty support. This has tion rates can be safely reduced at the local,
been demonstrated when a mandate to teach departmental level.
breast disease was introduced into ob/gyn
residencies. 23 The profession faces a similar
failure in regards to vaginal breech and
forceps deliveries unless a concentrated
Background
effort is made to enlist faculty support
for such approaches. Lomas et al. 24 demon- The assumption that the national cesarean
strated convincingly the importance of section rate can be safely reduced stems
opinion leaders in implementing cesarean from a number of observations. Perhaps the
section guidelines. most important one is that despite a con-
Insecurities about cesarean section stant increase in the national cesarean
utilization are not restricted to the United section rate since the 1960s, maternal and
States. A recent British study, the con- child health status has not improved in
sequence of rising cesarean section rates in parallel. 3 Cesarean section rates in other
an English teaching hospital, revealed a not developed nations are dramatically lower
entirely unexpected pattern. On retrospec- than in the United States, with perinatal
tive analysis, 30% of all cesarean sections outcome statistics that frequently are
performed were found to be unnecessary. superior to those in this country.26 The U.S.
More interesting, however, was the con- rate is the third highest among 21 reporting
siderable disagreement among auditors on countries, exceeded only by Brazil (where
when to perform a cesarean. The auditors cesareans are often performed for cosmetic
disagreed with themselves in 25% of cases reasons) and Puerto Rico. 3 Within the
when shown identical information a second United States, significant regional dif-
time. 25 One might expect similar findings in ferences can be found in the incidence in
U.S. teaching departments. cesarean section utilization, with the South
Well-trained opinion leaders are therefore being the clear leader (27.6% cesarean sec-
of crucial importance at the departmental tion rate), followed by the Northeast (22.6%),
and the national level. Only if a committed the Midwest (21.8%), and the West (19.8%).3
leadership actively supports the effort to Cesarean section rates are uniformly higher
reduce cesarean section rates will an impact in older women, in proprietary hospitals
be observed. What still remains to be (than nonprofit or government hospitals), in
determined is whether such an effort will be smaller hospitals (fewer than 300 beds),
directed by physicians or by bureaucrats for patients with private insurance (than
from government and insurance agencies. charity patients and women with public
The upcoming revolution in U.S. health health support), and for private services
care cannot overlook the most frequently (in comparison to teaching services, even
performed surgical procedure. Government within the same institutions).3,27
11. Methods for Safe Reduction of Cesarean Section Rates 145

Because neither geographic variation nor section patients had a vaginal delivery.3
most of the patient characteristics just noted While these numbers indicate an increase in
also define an increased risk for adverse VBAC utilization (from 8.5% in 1986), they
outcome, these variations in cesarean sec- still clearly lag behind a reasonable level
tion rates cannot be explained on the basis of of expected national utilization. One can
medical necessity. This is further supported therefore conclude that a further increase in
by the observation that institutions with VBAC deliveries is possible, is safe, and
a disproportionate number of high-risk should be the goal of each department. 32 On
patients, so-called perinatal centers (level a national level this means that of approx-
III institutions), usually demonstrate lower imately 338,100 repeat cesarean sections
cesarean section rates than providers of performed annually (35% of 966,000 total
lower levels of care. 28 All these observations cesareans), only 112,700 are really needed.
suggest that at least some cesarean section A number of studies have demonstrated
rates throughout the country can be reduced that the cesarean section indication of
without adverse impact on either mother or dystocia can be limited to no more than 2%
child. of all deliveries. 33 - 35 At present, slightly
Further evidence in support of this con- fewer than 9% of pregnant women undergo
tention comes from a careful review of cesarean sections with a primary diagnosis
published obstetric experience. Table 11.1 of dystocia/cephalopelvic disproportion,
demonstrates that approximately two-thirds more than four times the recommended
of cesarean sections are presently performed rate. 3
as either repeat cesareans or because of a Similar calculations can also be made
diagnosis of dystocia/cephalopelvic dispro- for less frequent indications for cesarean
portion. Any attempt to reduce cesarean delivery. For example, convincing data in
section rates has therefore to contend first the literature suggest that in carefully
with those two indications because they selected breech presentations delivery by
represent almost two-thirds of all indica- cesarean does not improve outcome over
tions for cesarean section deliveries. 3 vaginal delivery.36,37 In fact, these data
Evidence in the literature suggests that at refer not only to the classical singleton frank
most only one-third of previous cesarean breech38 ,39 at term but also to premature
sections require a repeat cesarean if patients breeches,40 nonfrank breeches,41,42 and
are allowed a proper trial of labor. 29 - 31 In breeches as part of a multiple birth.43 If one
1991, the most recently available data from summarizes these data, the conclusion
the Office of Vital and Health Statistics is that a safe vaginal delivery of breech
Systems, National Center for Health presentations can be accomplished in as
Statistics, Center for Disease Control, sug- many as two-thirds of cases. Considering the
gest that only 24.2% of all previous cesarean recently reported success with external ver-

TABLE 11.1. Current national and recommended cesarean section rates, according to indication (%)
Current national Deliveries
Percentage rates as a by cesarean
of all cesarean percentage of all section in Quilligan's low Mt. Sinai
Indication sections' deliveries" Porreco's study 36 recommendation45 target rate46
Previous cesarean section 30 6.8 1.5 2.0 2.2
Dystocia 30 6.8 1.5 2.0 2.2
Breech presentation of fetus 15 3.4 1.1 1.3 1.1
Fetal distress 5 1.1 0.6 1.5 1.1
Other 20 4.5 1.0 1.0 4.5
Total 100 22.7 5.7 7.8 11.1
"Calculated at a national cesarean section rate of 22.7% in 1985 (with a rounding error of 0.1 in the total).
146 N. Gleicher et al.

sions of breeches, one can further conclude cesarean section and vaginal delivery is
that a cesarean section because of breech $6193. 46 A safe 11% national cesarean sec-
delivery should be a rare event. tion rate would reduce the number of these
In 1991, 11.7% of all cesarean sections surgical procedures (based on 1991 data)
were performed because of breech presenta- from 966,000 to approximately 440,000,
tion. 3 This represents 113,022 women or ap- with a cost savings of $3.26 billion. The
proximately 2.8% of all deliveries. Because political significance of this fact is obvious.
breech presentations represent approx- A review of published experience thus
imately 3.5% of term pregnancies,3 one can confirms that a safe reduction in cesarean
calculate an 80% national cesarean section section rates can be achieved. This led US44
rate for breeches. This is clearly still reflec- and others 47 ,48 to implement specific cesar-
tive of an almost universally held opinion ean section reduction programs at a depart-
that breeches should be delivered by cesar- mental level.
ean section, even though valid studies
statistically supporting this practice are
rare. 37
Cesarean sections for a fetal distress
The Mount Sinai-Chicago
indication reflect only a small part of the Experience
total spectrum of cesarean section indica-
tions. The most recent data (1991), in fact, Even though the departmental cesarean sec-
suggest that at 9.2% of all cesarean section tion rate had been considerably below
indications, fetal distress is a precipitating the national average, the leadership of the
diagnosis even less frequent than breech department in 1985 reached the conclusion
presentation. 3 Because most medical-legal that only a well-defined program would
liability situations involve the allegation affect the departmental practice pattern as a
of fetal brain damage in utero, the proper whole. Our national research at that point
response to a presumption of fetal distress had confirmed the departmental impression
(possibly by cesarean section) has obviously that the well-thought-out recommendations
great potential legal ramifications. The fact of the National Consensus Development
that cesarean sections for fetal distress Conference were not effective in changing
represent only such a small fraction of practice patterns. 32 ,49 We therefore decided
all cesarean section indications further to design a cesarean section reduction initia-
supports the contention that medical-legal tive that appropriated the universally
considerations affect national cesarean sec- accepted recommendations of the Consensus
tion rates only marginally. Development Conference but, in addition,
Even if cesarean section rates for fetal took the step from "recommendation" to
distress and all other (less frequent) indica- peer-reviewed "implementation."
tions remained at their current level of Peer review was critically important. Not
utilization, one could still calculate a safe only did the Health Care Quality Improve-
target rate for a national cesarean section ment Act of 1986 attempt to bolster the
rate of approximately 11%.44 We did this concept of peer review, 50 but we believed
calculation in consideration of published that a well-designed peer review process
experience by Porreco 34 and based on a could be the key factor in controlling cesar-
detailed evaluation of published literature ean section rates. When extramural funding
by Quilligan (see Table 11.1),45 both of for a computer system was secured, we were
whom actually suggested an even lower goal able to put into place a computer-assisted
(6% and 8%, respectively). data-collection and analysis system that
Based on a recent survey by the New allowed us to implement a comprehensive
York-based Metropolitan Life Insurance and fair-minded peer review system which,
Co., the average cost differential between a in fact, became the cornerstone of our ce-
11. Methods for Safe Reduction of Cesarean Section Rates 147

sarean section reduction initiative. Data deliveries, have repeatedly resulted in

collection was not based on maximal com- blatantly false statistical outcome evalua-
pleteness but on the concept of minimal tions in the literature. 37
essential data. This turned out to be a wise As noted before, the program was vol-
decision because maintenance of the system untary. It was principally based on the
required online data input. A requirement establishment of clinical practice guidelines
for too many data would have reduced com- (CPGs) for the most frequent cesarean sec-
pliance and therefore reduced the reliability tion indications. While CPGs have since
of our data. Appendices I and II demonstrate become part of the medical vernacular, in
neonatologic and obstetric outcome data- 1986 the concept of CPGs required careful
collection summaries. explanation to all departmental physicians.
We then analyzed the literature (as pre- Appendix III summarizes the departmental
viously described) to determine a target CPGs for repeat cesarean section, dystocial
cesarean section rate that we as the depart- cephalopelvic disproportion, breech presen-
ment wanted to achieve (see Table 11.1). tation, and fetal distress. The CPGs basically
Our departmental rate in 1985 was approx- followed recommendations already made by
imately 17%. We felt confident that a rate of the National Consensus Development Con-
11%-12% could be safely achieved. 44 The ference 6 years earlier,1 and were therefore
Mount Sinai cesarean section reduction neither experimental nor department
initiative was started on a departmental specific.
level in January of 1986 and involved on a Departmental CPGs were reviewed on
voluntary basis all members of the depart- an ongoing basis as part of the quality
ment, whether geographic fulltime faculty assurance process. Obstetric management
or voluntary attending staff. Participation of patterns that did not follow departmental
the voluntary staff was crucial because CPGs were discussed in weekly conferences,
approximately 50% of all departmental preferably in the presence of the responsible
deliveries were under their control. attending physician. In addition, the depart-
Cooperation from the neonatal service mental data bank allowed the establishment
also was crucial. Any attempt to change a of physician-specific practice patterns. Each
cesarean section practice pattern has to attending and resident physician thus
carefully control for neonatal outcome. received at regular intervals a personal
Because the failure to do so had in our practice profile (Figure 11.2). These practice
opinion led to an unwarranted increase in profiles showed an individual's cesarean sec-
cesarean section utilization, it seemed of tion rate. in relationship to the rest of the
critical importance not to make the same department. If an individual fell outside
mistake in attempting to influence cesarean 2 SD from the departmental mean, their
section rates downward. We therefore chose practice pattern was reviewed in a private
to tabulate 35 different neonatal morbidity conference with either the department
factors in addition to standard perinatal chairman or chief of maternal-fetal me-
mortality tabulations (see Appendix I). dicine. Not a single disciplinary action was
Because birthweight and gestational age taken in the first 2 years of the program. 44,51
are the most important determinants of The department's cesarean section reduc-
neonatal outcome, we also ensured that our tion initiative also mandated a second
neonatal outcome system evaluation opinion from a nonaffiliated board-certified
permitted us to differentiate the effects member of the department in all nonemer-
of route of delivery from, for example, gency cesarean sections. This mandate was
prematurity. Such a capability is crucial initially actively reviewed. It became,
because disproportionate weight distribu- however, increasingly apparent that this
tion in allegedly comparable patient po- second opinion requirement was unneces-
pulations, especially in regards to breech sary in view of other program components.
148 N. Gleicher et al.

Cesarean Rates (0/0) Cesarean Rates (0/0)

25 25
* 1988 Cesarean Section Rates 1989 Cesarean Section Rates
by Practitioner 20 by Practitioner
20

* * *
15
*
15
**t *
* *
10 * * * 10 * *
5
**
*
*
*
*
**
*
-----",- 5
** *
-+-
*
*
.
4-
*

* *
I I
0 0
Total Primary Repeat Total Primary Repeat
*=Your Rate * = Your Rate
FIGURE 11.2. Scattergram of cesarean section outlier physicians than the 1989 statistics. Each
rates for selected Mount Sinai-Chicago physi- report is individualized (by star) to a physician.
cians in comparison to departmental mean rate (From Myers and Gleicher,51 by permission of
(horizontal line). The 1988 data reflect more The Quality Letter for Healthcare Leaders.)

Enforcement of the requirement was there- can therefore be expected. Our charge was to
fore abandoned after the first 2 years. determine whether that (prior) level was
The first 2 years of experience with the adversely affected by the cesarean section
program were reported in 1988. 44 The reduction initiative. We conclusively
departmental cesarean section rate fell in demonstrated that this was not the case,
that time period from 17.5% to 11.5% and although adverse outcomes obviously con-
neither fetal nor neonatal mortality or tinued to occur.
morbidity were affected. This report received The Mount Sinai -Chicago experience was
considerable attention, not the least because thus the first departmental cesarean section
of an accompanying editorial that implied reduction initiative that conclusively
that a reduction of cesarean section rates demonstrated the feasibility of such pro-
may after all affect neonatal outcome grams. Up to that point, efforts to reduce
adversely. 52 Specifically, the editorial noted cesarean section rates were either within the
the occurrence of adverse outcomes with realm of theoretical considerations,1,45,49
some vaginal breech deliveries. The editorial involved only single indications,29,38,4o
writers apparently did not know, however, or represented the effort of single practi-
that the New England Journal of Medicine tioners. 34 Moreover, the program also
had required much more detailed outcome demonstrated that a significant impact on
data submission during the review process cesarean section rates could be made by
than they were willing to publish in the final establishing rather simple program com-
manuscript. Moreover, a reduction in cesar- ponents as CPGs, with an accompanying
ean section utilization is not necessarily in quality assurance program.
itself a tool to improve neonatal outcome. Its It was probably exactly that simplicity
purpose is primarily to maintain neonatal that caused a considerable degree of skep-
outcome while achieving a less invasive and ticism in the profession after the initial
less costly method of delivery for the mother. Mount Sinai experience was published.
A level of neonatal mortality and morbidity Arguments were made that the experience
11. Methods for Safe Reduction of Cesarean Section Rates 149

was not duplicable in the "real" world had reached 27.3%. Their experience has
because such a program could only be imple- been reported in two publications. 48 ,54 Total,
mented in a teaching hospital with residents.primary, and repeat cesarean section rates,
Others argued that an inner-city population respectively, were reduced from 27.3%,
was not representative, and, finally, the 18.2%, and 9.1% in 1986 to 16.9%, 10.6%,
argument was repeatedly made that private and 6.4% in 1991 by "encouraging" VBAC
practitioners in a private hospital setting deliveries, distributing physicians' cesarean
just did not have an adequate emergency section rate information annually to all staff
backup to instigate a successful cesarean (as part of a peer review process) and by
section reduction initiative. As is demon- implementing an active management pro-
strated in the following section, none of gram for labor. While thus successful in
these arguments holds up to scrutiny. In their overall goal to diminish the depart-
fact, every serious attempt ever made to mental cesarean section rate, their impact
affect cesarean section rates within an on private versus teaching services varied
institution has been successful, whether in greatly. As Sandmire55 in a discussion of
inner-city teaching hospitals or private these data correctly asked, can a private
or public institutions, whether driven by service cesarean section rate of almost twice
faculty physicians or private practitioners. the clinic rate be really regarded as success,
or does it have to be seen as an acknowledg-
ment of failure to influence one's peers?
Cesarean Section Reduction The feasibility to affect cesarean section
Programs patterns in a strictly private practice setting
was first shown by Gellman and associates 31
While the Mount Sinai experience demon- who demonstrated quite early during the
strated the feasibility of a successful ce- national debate on the topic that VBAC
sarean section program within a teaching deliveries could be safely performed in a
institution, the argument was made that private setting. Even though the number of
experiences gained from a disproportionally studies from the private sector has (not
large service population were not applicable surprisingly) remained small, enough
to more private settings. This argument was evidence has, in fact, been presented to sup-
further supported by the repeatedly made port the contention that the private practice
observation that even within the same of obstetrics is just as receptive to properly
teaching institutions, the teaching service guided change as the academic world.
routinely achieved lower cesarean section Rayner 56 demonstrated this by obtaining a
rates than a private service. 12,48,53 very respectable VBAC rate in a small rural
Porreco's pioneering efforts demon- community (Gellman and associates 31 re-
strated this fact preceding the Mount Sinai ported their effort from New York City).
experience,34 although the Mount Sinai In a Canadian rural community hospital,
Program achieved a greater reduction in the Iglesias et al. 57 were able to reduce their
private cesarean section rate (from 20% to overall cesarean section rate from 23% to
12.4%) than the faculty rate (from 15% to 12% over 4 years. This was accomplished by
11.7%) and thus statistically equalized the implementing VBAC and breech delivery
cesarean section utilization between the two guidelines and criteria for the diagnosis
services. 44 of dystocia, following a report from the
In the same city, Northwestern Memorial National Consensus Conference on Aspects
Hospital, a primarily private tertiary of Cesarean Birth (NCCACB), released in
university facility with a much smaller Canada in 1986.8 Those were very similar
service population than Mount Sinai, to recommendations by the U.S. National
initiated a cesarean section reduction pro- Consensus Development Conference In
gram in 1986 when the departmental rate 1980. 1
150 N. Gleicher et al.

Peck58 recently reported the experience of mother and fetus are in good condition
Lutheran Hospital in La Crosse, Wisconsin, and progress in labor is observed. Epidural
a community-based tertiary-care center anesthesia was discouraged. A "homey"
with a wide perinatal referral area. Their delivery environment and, finally, a com-
cesarean section rate during 4 years had mitted physician staff were also considered
remained in the 10%-11% range with key.
a perinatal mortality of 11 in 1000. Peck Table 11.2 summarizes the cesarean sec-
attributed their success to 12 specific factors: tion experience of a suburban Chicago com-
an active midwifery program delivered 30% munity hospital for the years 1988-1992.
of their newborns with a cesarean section These data are presented to demonstrate the
rate of less than 5 percent for this very low importance of the training process that then
risk population. (Butler et al. 59 recently once creates the mindset for practicing obstetri-
again quite convincingly demonstrated that cians. Sherman Hospital in Elgin is a small,
nurse-widwife care is associated with a nonteaching community hospital, which in
reduced incidence of cesarean sections.) the Chicago area has gained a reputation for
Eighty percent of women with previous having the lowest cesarean section rate
cesarean section underwent a trial of labor, among comparable institutions.
and among those 80 percent delivered The institution was joined in 1988 by one
vaginally (almost the same experience as of the authors (RH.D.) as a private practi-
reported by Mount Sinai-Chicago).44 Two- tioner after he graduated from the residency
thirds of persistent breeches at 37 weeks program at Mount Sinai Hospital Medical
gestational age were successfully converted Center in Chicago. The departmental cesar-
to a vertex presentation; approximately ean section rate at that point was 19.9%,
50% of remaining breeches were vaginally with RH.D. having a 10% and the re-
delivered. Strict diagnostic criteria for the mainder of the department a 20.2% rate.
cesarean section indication of "fetal distress" During the ensuing 5 years, the RH.D.
resulted in only a 1% cesarean section rate practice grew from 53 (in 1988) to 1481
for this diagnosis. A physician on-call system (in 1992) deliveries and from one to five
placed a board-certified obstetrician on call physicians (plus two midwives). The cesar-
for hospital (labor floor) coverage at all ean section rate for this fully private group
times. The use of intrauterine pressure remained stable in a range between 6.3%
transducers whenever labor is augmented and 11.8%, while the remainder of the de-
was encouraged. No time limits are set for partment stayed in a range of 17.3%-20.2%
the length of the second stage so long as Because the RH.D. practice increased their

TABLE 11.2. Cesarean section utilization at Sherman Hospital, Elgin, Illinois

Year 1988 1989 1990 1991 1992
Sherman Hospital
Total births 1788 1741 1999 2372 2877
Cesarean section rate 19.9% 17.4% 15.5% 14.5% 15.0%
Primary 11.5 10.2 10.3 8.8 9.8
Repeat 8.4 7.2 5.2 5.7 5.2
Epidural rate <1% 1% 10% 15% 16%
RHD & Associates
Total births 53 167 436 917 1081
Cesarean section rate 10% 8.1% 6.3% 10.2% 11.8%
All other physicians
Total births 1735 1574 1563 1499 1396
Cesarean section rate 20.2% 18.4% 18.1% 17.3% 19.4%
11. Methods for Safe Reduction of Cesarean Section Rates 151

contribution to the departmental patient Selective Contributing Factors

pool from 3% to 52%, their practice pattern
significantly impacted the departmental
to Cesarean Section Deliveries
cesarean section rate, which decreased from
almost 20% to 15%.
Previous Cesarean Section
These data are presented to demonstrate Guidelines from a variety of authoritative
that low cesarean section rates can be national and international bodies strongly
achieved in a variety of settings. R.H.D. support the concept of VBAC,1,6-8 and
successfully transferred the basics of the generally suggest a trial of labor in women
Mount Sinai cesarean section reduction with one previous cesarean section, a history
initiative from an inner-city teaching of transverse uterine scar (with uncom-
hospital to a private, suburban community plicated healing process), and no other
hospital. Moreover, he transferred those complicating factor. In fact, one could sum-
principles from a large teaching department, marize those recommendations as an
which included a residency program, to endorsement of VBAC delivery with the
what was initially a one-person private caveat that a repeat cesarean section may
practice. This practice grew in the second still be indicated whenever "in doubt." We
year to two physicians, when R.H.D. was disagree with such an approach in that
joined by his wife J.B.N., another Mount we believe that a well thought out medical
Sinai graduate. As the practice continued to approach should always demand that the
expand these two Mount Sinai graduates more invasive, more costly procedure must
had to transmit their philosophy of practice establish its benefit over less invasive and
to other physicians who came from training less costly alternatives. Consequently, when
programs with very different approaches. In "in doubt" a trial of vaginal delivery should
doing so successfully, they demonstrated be attempted.
not only that a Mount Sinai-like program Under such a philosophy a trial ofVBAC
approach is feasible in private practice, but can and should be performed not only after
that it can be taught to any obstetrician one previous cesarean but independent of
with an open mind, wherever he or she may how many have been performed,17-19 if a
practice. They also demonstrated that a breech presentation is present (assuming
frequently mentioned "consumer demand" vaginal breech delivery criteria are met)62
for cesarean section delivery cannot be too with multiple gestations,63 if macrosomia of
pervasive if they succeeded in growing their the fetus is suspected,64 or even if a scar is
practice 20 fold over 4 years. unknown or classical. 65 Based on a meta-
That cesarean section review programs analysis of morbidity and mortality, Rosen
will affect cesarean section rates has been et al. 16 argued in favor of more trials of labor
demonstrated in a variety of different in women with previous cesarean sections.
settings. For example, Kaiser Permanente, Our own experience44 (and Table 11.2)
the nation's largest managed care provider, suggest that 80%-90% of women should
has reduced cesarean section rates very undergo a trial oflabor. Among those, 80%-
favorably in comparison to national and 90% can be expected to deliver successfully
regional figures. 6o New York State and the vaginally.
local ACOG District developed a program to
enhance in-house cesarean section review
processes for participating hospitals. 61 The
Previous Myomectomy
state's cesarean rate was reversed, and A previous myomectomy, which resulted
reviewed hospitals demonstrated a stronger in entry into the endometrial cavity, is
downward trend than nonreviewed institu- universally considered an indication for a
tions. Similar effects have also been re- cesarean section. We were unable to find
ported from outside the United States. 47 ,57 even a single valid study in support of such a
152 N. Gleicher et al.

practice pattern in the literature and can fetuses can be accomplished in approx-
therefore not support the contention of an imately 50% of cases. 38,40,44
automatic cesarean section. If one were to
extrapolate from the cesarean section scar
experience (and why should one scar differ
Preterm Delivery
from another), then at least with single entry A number of reports have suggested an
into the cavity, the experience should not increased utilization of cesarean sections
differ from that in patients with unknown or for preterm deliveries for vertex as well as
classical scars. Because uterine rupture in breech presentations. 1 ,67,68 As noted pre-
those appears to be a rare event,65 a trial of viously, there appears to be no benefit from
labor after at least single entry into the performing cesarean section deliveries in
cavity appears warranted. breech presentation. 40 ,41,66 Malloy et al.,69
As in patients after previous cesarean in a study from the National Institute of
section, labor should be monitored carefully Child Health and Human Development
and the longitudinal observation of in- Neonatal Research Network, recently con-
trauterine pressures may be of benefit. cluded (after accounting for potential other
contributing maternal and fetal factors) that
cesarean delivery is not associated with a
Dystocia/Cephalopelvic lower risk of either mortality or intraven-
Disproportion tricular hemorrhage in very low birthweight
The Mount Sinai-Chicag044 and North- infants, independent of presentation.
western54 experiences have clearly confirmed Similar results regarding the influence
that an active obstetric management ap- of route of delivery on intraventricular
proach can reduce cesarean section rates hemorrhage in preterm infants were also
for this indication dramatically. As noted reported by Anderson et a1. 70 Prematurity
earlier, not more than 2% of deliveries per se does therefore not constitute an in-
should undergo a cesarean section for dication for cesarean section delivery.
dystocia. 33 - 35 Both programs were based on
simple treatment protocols that can be Fetal Distress
easily repeated in different settings.
The use of fetal monitoring has been im-
plicated as a cause of increased cesarean
Breech Presentation section rates.71 Because this probably is the
We reviewed in 1987 the indications for result of a misinterpretation of fetal heart
cesarean sections for breech presentations in rate tracings, it is only reasonable to cor-
great detai1. 36 ,37 Our conclusions then, roborate the true presence of fetal distress
based on an extensive review of the litera- situation by other means whenever possible.
ture, was that statistically valid studies The Mount Sinai-Chicago program sug-
which demonstrate the benefit of a cesarean gested a fetal scalp pH, whenever technically
section over a vaginal delivery are lacking feasible, in confirmation of fetal acidosis
if minimal patient selection criteria were before a cesarean section was performed
followed. 38 ,42 In fact, vaginal breech delivery under the presumed diagnosis of fetal dis-
is indicated not only for term frank breech tress. 44 Others have recommended fetal
presentation but for nonfrank breech pre- scalp stimulation as an alternative.72
sentation and premature fetuses as well. Overally, cesarean sections for fetal
The same conclusion can be applied to breech distress should probably occur in not more
presentation in multiple gestations. 43 ,66 than 0.6% (Porrec034) to 1.5% (Quilligan45 )
Based on published data and our own of all deliveries. 44
experience, we therefore conclude that a safe
vaginal delivery of term and preterm breech
11. Methods for Safe Reduction of Cesarean Section Rates 153

Advanced Maternal Age herpetic lesions at time oflabor. 76 Randolph

et al. 77 recently suggested, however, that
One recent study suggests that advanced the practice to perform cesarean section
maternal age alone may influence phy- deliveries indiscriminately whether the
sicians in their decision on the mode of patient suffers from a first or a repeat episode
delivery and thereby place some older should be reexamined. They claimed that
women at an unnecessary risk of cesarean the automatic performance of a cesarean
section. 73 While older women may, in fact, section in women with repeat outbreaks
demonstrate an increased incidence of high of herpes simplex type 2 virus infections
risk factors, which by themselves can pre- results in more than 1580 excess cesarean
dispose to cesarean sections, age alone deliveries performed for every poor neonatal
should not. There is also no supportive outcome prevented, a cost of $2.5 million for
evidence in the literature for the frequently every case of prevented neonatal herpes, and
heard argument that older women have less a cost of $203,000 per quality-adjusted life-
tolerance for (prolonged) labor. year gained. These calculations were made
based on the fact that neonatal infection in
The Valuable Pregnancy recurrent herpes infection is, as a result of
protective antibodies, rare (1%78) in com-
One also frequently encounters in clinical parison to primary disease.
practice the concept of the so-called very In an accompanying editorial, Gibbs et
valuable pregnancy. This may be a first al. 79 supported the practice to perform
pregnancy in an older woman, a pregnancy cesarean section deliveries only in the pre-
after prolonged infertility, or a pregnancy sence of visible herpetic lesions. They noted,
that has reached viability after multiple however, that 38% of randomly chosen
earlier losses. While such pregnancies obstetricians were still performing screening
certainly are "valuable" or even "very cultures for herpes in pregnant women with
valuable," it is difficult to perceive of a treat- a history of genital herpes, an observation
ment approach that should differ from any that probably suggests that many cesarean
other standard of care. Because the treat- sections are still performed in only culture-
ment of any pregnancy should be maximal, positive women with no active disease. They
valuable or very valuable pregnancies editorial, however, also suggested that for
should have no "better" options available to the present not enough data are available to
them than any other conception. A concept recommend vaginal delivery in the presence
of preferred cesarean section delivery for of active lesions even in the women with a
such pregnancies therefore does not make repeat episode of genital herpes. A cesarean
sense unless one believes in the (incorrect) section delivery thus remains indicated
concept that cesarean section deliveries are whether an active lesions during labor is
in general safer to the fetus than vaginal primary or secondary in nature.
deliveries. 5 One therefore has to wonder
why, for example, conceptions achieved
through in vitro fertilization almost univer- Congenital Anomalies
sally are reported to demonstrate cesarean There are very limited data in the literature
section rates in the 50% range?4.75 on whether the performance of a cesarean
section impacts on outcome of fetuses with
known congenital anomalies. Luthy et al. 80
Genital Herpes investigated this question in regards to
The current recommendation of the In- prenatally diagnosed meningomyelocele.
fectious Disease Society for Obstetrics- Delivery by cesarean section before the
Gynecology is the performance of a cesarean onset of labor may result in better subse-
section delivery in the presence of active quent motor function than vaginal delivery
154 N. Gleicher et al.

or delivery by cesarean section after a period section utilization. In a well-designed

of labor. randomized study, Kennell et al. 87 demon-
Fetuses with congenital anomalies strated that the presence of a supportive
demonstrate an increased incidence of companion (doula) during labor and delivery
abnormal fetal heartrate patterns,81 which. significantly reduced the cesarean section
often lead to a false assumption of fetal dis- rate. Rosen88 in an accompanying editorial
tress and subsequent cesarean section. suggested that we may benefit from some
new, nontechnological approaches to
Physician Factors modern obstetric care. It is tempting to
speculate that it is at least in part this doula
Physicians do not perform cesarean sections effect that results in lower cesarean sec-
so as to make it to the golf course in time. 82 tion rates in birthing center settings and
Physician factors do, however, affect cesar- with midwife deliveries. 59 Any national
ean section utilization rates. Berkowitz et health care reform may benefit from this
al.,83 for example, have demonstrated that recognition.
older, more experienced physicians per-
formed fewer cesarean sections for dystocia
and a higher percentage of forceps deliveries Conclusions
and breech extractions. These results
reemphasize the earlier discussed (in) The reduction of the overall cesarean section
experience factor, which by itself has rate to 15 or fewer per 100 deliveries and of
become an indication for the performance the primary cesarean section rate to 12
of cesarean sections. In our opinion in- or fewer per 100 deliveries has become a
experience in the performance of alternative national health objective. 3,89 A national
delivery methods to cesarean sections does consensus seems finally to emerge that ce-
not represent an indication for cesarean sec- sarean sections are excessively utilized. In
tion so long as experience is available by response, multiple alternative strategies
consultation or referral. As the pendulum have been proposed to combat rising rates
swings back, it is increasingly recognized and possibly even reverse them. Stafford2
that even midforceps deliveries represent probably summarized them best by grouping
safe and acceptable alternatives to cesarean them into six distinct strategies (Table 11.3).
sections if patients are properly selected andMost of this chapter has been concerned
operators are experienced. 84 with his first strategy, which he called
Education and Peer Evaluation and which
Social Factors he described as strategies aimed at phy-
The rates of cesarean section utilization sicians reviewing the literature, the issuance
vary greatly with socioeconomic status. 85 of CPGs, and formal hospital programs. As
This is a potentially very disturbing finding he notes, these rely on implementation by
because it suggests that different levels of obstetricians and tend to preserve physician
obstetric care may be provided dependent autonomy. As physicians we are therefore
on the socioeconomic status of patients. well advised to support these strategies.
Interestingly, the provision of health in- The alternatives will undoubtedly in-
surance to previously uninsured low-income creasingly restrict physician independence.
pregnant woment may not by itself improve The External Audit and Review strategy,
maternal health but it certainly increases as Stafford claims, makes the assumption
cesarean section rates. 86 that without external input (or restraint)
physicians are incapable of making correct,
or economically sound, decisions. They
Stress therefore require external "guidance," a
The stress of labor appears to be a con- suggestion clearly not very popular among
tributing factor to the rate of cesarean physicians. The Public Dissemination of
11. Methods for Safe Reduction of Cesarean Section Rates 155

TABLE 11.3. Strategies for reducing cesarean section rates

Strategy Methods of implantation Impact on cesarean section
Education and peer evaluation Individual obstetricians reading Not effective
the medical literature
Clinical guidelines issued by Have failed to modifY cesarean
specialty society (ACOG) or section use
expert panels
Formal programs to reduce In teaching hospitals: reductions
cesarean section use in of as much as 34%
individual hospitals
External review of practices Audit of obstetric practices by In Scotland: single-year
outside agency reductions of 13% and 27%
Review of practices by health Impact has not been evaluated
care payers either before or
after services are provided
Public dissemination of cesarean section Making data accessible to Impact has not been evaluated
rates consumers for use in selection
of providers
Modification of physician reimbursement Equalize physician payments Reductions of 0%-50% but
for cesarean and vaginal limited implementation and
deliveries evaluation
Modification of hospital reimbursement Remove incentive for cesarean Limited implementation has
section through prospective been evaluated for impact
payment for deliveries
Medical malpractice reform Several approaches aimed at Implemention and assessment
reducing the volume and cost have been limited
of malpractice claims

Information strategy is based on the strong financial incentives for vaginal and
assumption that the public will make correct against cesarean section deliveries.
choices if only properly informed. In that Such an approach was recently also sup-
vein, the publication of cesarean section ported by Spellacy.90 A Change in Hospital
rates by physicians or hospitals has been Reimbursement strategy is then the next
proposed as a mandate. Whether such data logical consequence. At present hospitals
by themselve scan or will be correctly inter- have no incentive to reduce cesarean section
preted by the public is at least questionable. rates. In fact, payment incentives by third-
A Change in Physician Reimbursement party payers are probably in favor of cesar-
strategy has been proposed by these authors ean section deliveries because they reward
for almost a decade. 32 ,49 This is not based hospitals on a fee-for-service basis for all
on frequently cited (though totally unsup- additional services provided. A single rate
ported) allegations that physicians perform reimbursement for "delivery" would create
cesarean section deliveries to benefit from a strong incentive to hospitals to reduce ce-
increased reimbursement rates (in compari- sarean section rates because hospital costs
son to vaginal deliveries). In fact, nothing for cesarean sections are probably con-
supports such a contention. However, there siderably higher than for vaginal deliveries.
can also be no doubt that the average time Lastly, a Medical Malpractice Reform
effort for a vaginal birth is longer than for strategy has to be seen as an important
a cesarean section delivery, especially in component of any national cesarean section
regards to VBAC deliveries versus repeat effort.
cesarean sections. A financial incentive As we are entering the twenty-first
structure in favor of cesarean delivery there- century, the practice of medicine is under-
fore does not make sense, and the insurance going revolutionary changes. Outcome
industry would be well advised to create assessment for any form of medical in-
156 N. Gleicher et al.

tervention w~ll become an increasingly Cesarean Birth. Can Med Assoc J 1986;134:
important tool in an ongoing process of 1348-1352.
continuous improvement in the quality of 9. Marieskind HI. An evaluation of cesarean
patient care. At the same time, we should section in the United States. Washington,
learn from past mistakes and assess treat- DC: U.S. Department of Health, Education,
and Welfare, Office ofthe Assistant Secretary
ment strategies in their respective benefits
for Planning and Evaluation, 1980 ..
before they are widely implemented into
10. Rostow VP, Osterweis M, Bulger RJ. Medical
daily clinical practice. One of the lessons of professional liability and the delivery of
our national experience with cesarean sec- obstetrical care. N Engl J Med 1989;321:
tions is that practice patterns develop a life 1057-1060.
of their own. They not only determine how 11. Carpenter MW, Soule D, Yates WT, Meeker
we practice but, unfortunately, also how we CL. Practice environment in association
think and how we train. As a consequence, it with obstetric decision making regarding
is much more difficult to reverse a national abnormal labor. Obstet Gynecol 1987;70:
practice pattern than it is to implement a 657-662.
new one. In that sense the national cesarean 12. Goyert GL, Bottoms SF, Treadwell MC,
Nehra PC. The physician factor in cesarean
section experience teaches a very important
birth rates. N Engl J Med 1989;320:706-
lesson to everybody who is concerned about 709.
the future of medical care in this country. 13. Newton ER, Higgins CS. Factors associated
with hospital-specific cesarean birth rates. J
References Reprod Med 1989;34:407-411.
14. Localio AR, Lawthers AG, Bengtson JA, et
1. National Institutes of Health. Cesarean al. Relationship between malpractice claims
Childbirth: Report of a Consensus Develop- and cesarean delivery. JAMA 1993;269:366-
ment Conference. Bethesda, Maryland: 373.
Department of Health and Human Services, 15. Rutkow 1M. Obstetrics and gynecologic
1981. (Publication no. DHHS (NIH) 82-2067.) operations in the United States, 1979 to
2. Stafford RS. Alternative strategies for con- 1984. Obstet GynecoI1986;67:755-759.
trolling rising cesarean section rates. JAMA 16. Rosen MG, Dickinson JC, Westhoff CL.
1990;263:683-687. Vaginal birth after cesarean: a meta-
3. Anonymous. Rates of cesarean delivery- analysis of morbidity and mortality. Obstet
United States 1991. Morbid Mortal Wkly Rep Gynecol 1991;77:465-470.
1993;42:285-289. JAMA 1993;269:2360. 17. Porreco R, Meier P. Trial of labor in patients
4. Abraham L. Cesarean on demand. Am with multiple cesarean sections. J Reprod
Medical News 1988 Nov 4:1830. Med 1983;28:770-776.
5. Feldman GB, Friedman JA. Prophylatic 18. Novas J, Myers SA, Gleicher N. Obstetric
cesarean section at term? N Engl J Med outcome of patients with more than one
1985;312:1264-1267. previous cesarean section. Am J Obstet
6. Guidelines for vaginal delivery after a pre- GynecoI1989;160:364-367.
vious cesarean birth. statement of the Com- 19. Asakura H, Myers SA. Multiple cesarean
mittee on Obstetrics: Maternal and Fetal section: A five year experience with 434
Medicine. Washington, DC: American Col- patients. Am J Obstet Gynecol 1993;168:437
lege of Obstetricians and Gynecologists, (abstr).
1984. 20. Ophir E, Yagoda A, Rojansky N, Oettinger
7. New guidelines for VBAC. statement of the M. Trial of labor following cesarean section:
Committee on Obstetrics: Maternal and dilemma. Obstet Gynecol Surv 1988;44:19-
Fetal Medicine. Washington, DC: American 24.
College of Obstetricians and Gynecologists, 21. Gleicher N, Vermesh M, Rotmensch Z,
1988. Thornton J, Elrad H. Cesarean section pat-
8. Hannah WJ, Baskett TF, Chance GW, et al. terns: influence of a perinatology service.
Indications for cesarean section: final state- Mount Sinai J Med 1985;52:100-105.
ment of the panel of the National Consensus 22. Laube DW, ed. Core curriculum for residents
Development Conference on Aspects of of obstetrics and gynecology. Washington,
11. Methods for Safe Reduction of Cesarean Section Rates 157

DC: Council on Resident Education, 1992: tion: a study of 208 cases. Am J Obstet
103. Gynecol 1980;137:235-244.
23. Gleicher N. Breast disease programs in 39. Green JE, McLean F, Smith LP, Usher R.
obstetrics and gynecology: a plea for training Has an increased cesarean section rate for
in mammography. Am J Obstet Gynecol term breech delivery reduced the incidence of
1989;161:267 -270. birth asphyxia, trauma and death? Am J
24. Lomas J, Enkin M, Anderson GM, Hannah Obstet GynecolI982;142:643-648.
WJ, Vayda E, Singer J. Opinion leaders vs. 40. Kitchen W, Ford GW, Doyle LW, et al.
audit and feedback to implement practice Cesarean section or vaginal delivery at 24 to
guidelines. Delivery after previous cesarean 28 weeks gestation: comparison of survival
section. JAMA 1991;265:2202-2207. and neonatal and two-year morbidity. Obstet
25. Barrett JFR, Jarvis GJ, MacDonald HN, Gynecol 1985;66:149-157.
Buchan PC, Tyrrell SN, Lilford RJ. Incon- 41. Woods JR. Effect of low birth weight breech
sistencies in clinical decisions in obstetrics. delivery on neonatal mortality. Obstet
Lancet 1990;336:549-551. Gynecol 1979;53:735-740.
26. Notzen FC, Placek PJ, Taffel SM. Compari- 42. Gimovsky ML, Wallace RL, Shifrin BS, et al.
son of national cesarean-section rates. N Randomized management of the nonfrank
Engl J Med 1987;316:386-389. breech presentation at term. a preliminary
27. Stafford RS. The impact of nonclinical factors report. Am J Obstet Gynecol 1983;146:34-
on repeat cesarean section. JAMA 1991; 40.
265:59-63. 43. Fishman A, Grubb DK, Kovacs BW. Vaginal
28. Gleicher N. In reply. JAMA 1985;253:3248- delivery of nonvertex second twin. Am J
3249. Obstet Gynecol 1993;168:861-864.
29. Gibbs CEo Planned vaginal delivery following 44. Myers SA, Gleicher N. A successful program
cesarean section. Clin Obstet Gynecol 1980; to lower cesarean-section rates. N Engl J
23:507 -515. Med 1988;319:1511-1516.
30. Martin IN Jr, Harris BA Jr, Huddelston JF, 45. Quilligan E. Cesarean section: modern per-
et al. Vaginal delivery following previous spective. In: Queenan JT, ed. Management of
cesarean birth. Am J Obstet Gynecol 1983; high-risk pregnancy, 2nd Ed. Oradell, NJ:
146:255-263. Medical Economics, 1985:594-600.
31. Gellman E, Goldstein MS, Kaplan S, Shapiro 46. Anonymous. Childbirth. Chicago Tribune
WJ. Vaginal delivery after cesarean section 1993. Jan (sect 7) 17:2.
experience in private practice. JAMA 1983; 47. Tay SK, Tsakok FHM, Ng CSA. The use of
249:2935-2937. intradepartmental audit to contain cesarean
32. Gleicher N. The cesarean section epidemic. section rate. Int J Gynecol Obstet 1992;39:
Mt Sinai J Med 1986;53:563-565. 99-103.
33. Placek PJ, Taffel SM, Molen M. Cesarean 48. Socol ML, Garcia PM, Peaceman AM, Dooley
rates increases in 1985. Am J Public Health SL. Reducing cesarean section births at a
1987;77:241-242. primarily private university hospital. Am J
34. Porreco RP. High cesarean section rates: a Obstet Gynecol 1993;168:1748-1758.
new perspective. Obstet Gynecol 1985;65: 49. Gleicher N. Cesarean section rates in the
307 -311. United States. The short term failure of the
35. MacDonald D, Grant A, Sheridan-Penetra National Consensus Development Con-
M, Boylan P, Chalmers I. The Dublin ran- ference in 1980. JAMA 1984;252:3273-3276.
domized control trial of intrapartum fetal 50. Inglehart JK. Health policy report. N Engl J
heart rate monitoring. Am J Obstet Gynecol Med 1987;316:960-965.
1985;152:524-539. 51. Myers S, Gleicher N. The Mount Sinai
36. Confino E, Gleicher N, Elrad H, Ismajovich cesarean section reduction program: auto-
B, David MP. The breech dilemma: a review. mated data selection and peer review. Qual
Obstet Gynecol Surv 1985;156:6-10. Lett 1991;3:10-17.
37. Myers SA, Gleicher N. Breech delivery: why 52. Battaglia FC. Reducing the cesarean section
the dilemma? Am J Obstet Gynecol 1987; rate safely. N Engl J Med 1988;319:1540-
156:6-10. 1541.
38. Collea JV, Chein C, Quilligan EJ. The ran- 53. Heynes de Regt R, Minkoff HL, Feldman J,
domized management term breech presenta- Schwarz RH. Relation of private or clinical
158 N. Gleicher et al.

care to the cesarean birth rate. N Engl J Med 67. Taffel SM. Cesarean section in America:
1986;315:6619-6624. dramatic trends, 1970-1987. Stat Bull
54. Lopez-Zeno JA, Peaceman AM, Adasher JA, Metrop Life Insur Co 1989;70:2-11.
Socol ML. A controlled trial of a program for 68. Williams RL, Chen PM. Identifying the
the active management of labor. N Engl J sources of the recent decline in perinatal
Med 1992;326:450-454. mortality rates in California. N Engl J Med
55. Sandmire HF. In discussion of reference 48. 1982;306:207 -214.
Am J Obstet Gynecol 1993;168:1754- 69. Malloy MH, Onstad L, Wright E, et al. The
1756. effect of cesarean delivery on birth outcome
56. Raynor BD. The experience with vaginal in very low birth weight infants. Obstet
birth after cesarean delivery in a small rural Gynecol 1991;77:498-499.
community practice. Am J Obstet Gynecol 70. Anderson GD, Bada HS, Shaver DC, et al.
1993;168:60-62. The effect of cesarean section on intraven-
57. Iglesias S, Burn R, Saunders LD. Reducing tricular hemorrhage in the preterm infant.
the cesarean rate in a rural community Am J Obstet Gynecol1992;166:1091-1101.
hospital. Can Med Assoc J 1991;145:1456- 71. Haverkamp AD, Orleans M, Langendoerfer
1459. S, et al. A controlled trial of the differential
58. Peck T. Low cesarean rates possible. ObGyn effects of intrapartum fetal monitoring. Am J
News 1992;27:9. Obstet Gynecol 1979;134:399-404.
59. Butler J, Abrams B, Parker J, Roberts JM, 72. Clark SL, Gimovsky ML, Miller FC. The
Laros RK Jr. Supportive nurse-midwife care scalp stimulation test: a clinical alternative
is associated with a reduced incidence of ce- to fetal scalp blood sampling. Am J Obstet
sarean section. Am J Obstet Gynecol 1993; Gynecol1983;148:274-277.
168:1407 -1413. 73. Gordon D, Milberg J, Daling J, Hickok D.
60. Piturro M. Managed care obstetrical services: Advanced maternal age and risk factor for
an option where everyone wins. Group cesarean delivery. Obstet Gynecol 1991;77:
Practice Managed Health Care News 1992 493-497.
Feb: 16. 74. Wagner MG, St Clair PA. Are in-vitro fer-
61. Dillon WP, Choate JW, Nusbaum ML, Amao- tilization and embryo transfer of benefit to
McCarthy M, McCall M, Rosen MG. Obstetric all? Lancet 1989;2:1027-1030.
care and cesarean birth rates: a program to 75. Sauer MD, Paulson RJ, Lobo RA. Reversing
monitor quality of care. Obstet Gynecol1992; the natural decline in human fertility! An
80:731-737. extended clinical trial of oocyte donation to
62. Ophir E, Oettinger M, Yagoda A, Makovits women of advanced reproductive age. JAMA
Y, Rojansky N, Shapiro H. Breech presenta- 1992;268:1275-1279.
tion after cesarean section. Always a section? 76. Gibbs RS, Amstey MS, Sweek RL, Mead PB,
Am J Obstet Gynecol 1989;161:25-28. Sever JL. Management of genital herpes
63. Strong TH Jr, Phelan JP, Ahn MO, Sarno AP infection in pregnancy. Obstet Gynecol 1988;
Jr. Vaginal birth after cesarean delivery in 71:779-780.
the twin gestation. Am J Obstet Gynecol 77. Randolph AG, Washington E, Prober CG.
1989;161:29-32. Cesarean delivery for women presenting
64. Flamm BL, Goings JR. Vaginal birth after with genital herpes lesions. JAMA 1993;
cesarean section: is supected fetal macro- 270:77-82.
somia a contraindication? Obstet Gynecol 78. Chuang T. Neonatal herpes: incidence,
1989;74:694-698. prevention and consequences. Am J Prev
65. Beall M, Eglinton GS, Clark SL, Phelan JP. Med 1988;4:47-53.
Vaginal delivery after cesarean section: the 79. Gibbs RS, Amstey MS, Lezotte DC. Role of
unknown uterine scar. J Reprod Med 1984; cesarean delivery in preventing neonatal
29:31-36. herpes infection. JAMA 1993;170:94-95.
66. Davison L, Easterling TR, Jackson JC, 80. Luthy DA, Wardinsky T, Shurtleff DB, et al.
Benedetti TJ. Breech extraction oflow-birth- Cesarean section before the onset of labor
weight second twins: can cesarean section be and subsequent motor function in infants
justified? Am J Obstet Gynecol 1992;166: with meningomyelocele diagnosed ante-
497-502. natally. N Engl J Med 1991;324:662-666.
11. Methods for Safe Reduction of Cesarean Section Rates 159

81. Garite TJ, Linzey EM, Freeman RK, et al. pregnant women on maternal health and the
Fetal heart rate patterns and fetal distress in use of cesarean section. JAMA 1993;270:61-
fetuses with congenital anomalies. Obstet 64.
GynecolI979;53:716-721. 87. Kennell J, Klaus M, McGrath S, Robertson S,
82. Phillips RN, Thornton J, Gleicher N. Phy- Hinkley C. Continuous emotional support
sician bias in cesarean sections. JAMA 1982; during labor in a U.S. hospital. a randomized
248:1082-1084. controlled trial. JAMA 1991;265:2197-
83. Berkowitz GS, Fiarman GS, Mojica MA, 2201.
Bauman J, Haynes de Regt R. Effect of 88. Rosen NG. Doula at the bedside of the patient
physician characteristics on the cesarean in labor. JAMA 1991;265:2236-2237.
birth rate. Am J Obstet Gynecol 1989;161: 89. Public Health Service. Healthy people 2000:
146-149. National health promotion and disease pre-
84. Bashore RA, Phillips WH Jr, Brinkman CR vention objectives-full report with com-
III. A comparison of morbidity of midforceps mentary. Washington, DC: U.S. Department
and cesarean delivery. Am J Obstet Gynecol of Health and Human Services, Public Health
1990;162:1428-1432. Service, 1991. (DHHS publication no. (PHS)
85. Gould JB, Davey B, Stafford RS. Socioeco- 91-60212.)
nomic differences in rates of cesarean section. 90. Spellacy WN. Vaginal birth after cesarean: a
N Engl J Med 1989;321:233-239. reward/penalty system for national imple-
86. Haas JS, Udvarhelyi S, Epstein AM. The mentation. Obstet GynecolI991;78:316-317.
effect of health coverage for uninsured

Appendix I. Mount Sinai-Chicago Perinatal Database/

Neonatal Outcome

Birthweight >2500g 1500-2499g <1500g

Total % Total % Total %
A. Babies delivered 2001 86.0 245 10.9 71 3.1
1. 1 min apgar < 7 171 8.5 42 16.5 40 56
2. 1 min apgar < 3 18 0.8 9 3.5 14 19
3. 5 min apgar < 7 41 2.0 16 6.2 19 26
4. 5 min apgar < 3 3 0.1 1 0.3 2 2
Babies and Morbidity Data 1808 90.3 210 82.6 35 4.9
B. Cardiorespiratory
1. Endotracheal tube 22 1.2 22 lOA 21 60
2. Chest tube 7 0.3 7 3.3 10 28
3. Ventilator 14 0.7 22 lOA 19 54
4. Maximum F102 > 60% 02 10 0.5 7 3.3 11 31
5. Acidosis pH < 7.2 14 0.7 10 4.7 10 28
6. Pneumo th/med 2 0.1 0.0 0
7. Meconium aspiration 14 0.7 0.0 0
8. Pneumonia 1 0.0 0.0 0
C. Hematologic
1. Bilirubin> 10mg/dL 71 3.9 47 22.3 15 42
2. Phototherapy 73 4.0 51 24.2 22 62
3. EXC. transfusion 0.0 0.0 1 2
4. RBC transfusion 5 0.2 5 2.3 13 37
160 N. Gleicher et al.

Birthweight >2500g 1500-2499g <1500g

Total % Total % Total %

D. Neurologic
1. Seizures 3 0.1 0.0 0
2. Paralysis 0.0 0.0 0
3. CNS hemorrhage 2 0.1 2 0.9 3 8
4. Hydrocephalus 2 0.1 0.0 0
5. Other 14 0.7 2 0.9 0
E. Infection
1. Sepsis 7 0.3 3 1.4 2 5
2. Scalp 0.0 2 0.9 0
3. Lung 2 0.1 1 0.4 3 8
4. Other 12 0.6 3 1.4 2 5
F. Miscellaneous
1. Genetic abnormality 6 0.3 1 0.4 0
2. GI abnormality 35 1.9 15 7.1 7 20
3. Fracture 1 0.0 0.0 1 2
4. Peripheral IV 111 6.1 49 23.3 10 28
5. Central IV 113 6.2 50 2.38 13 37

Appendix II. Mount Sinai-Chicago Perinatal Database/

Obstetrical Outcome

Total % 1988 1987

A. Mothers delivered 1320 2430 2301

Babies delivered 1341 2472 2345
1. Single pregnancy 1300 99.4 2390 98.3 2257 98.0
2. Multiple pregnancy 20 1.6 40 1.6 44 1.9
3. Nulliparas 369 27.9 761 31.3 666 28.9
4. Multiparas 951 72.9 1669 68.8 1625 71.0
5. Private 871 50.8 1471 80.5 1418 61.2
6. Faculty 649 49.1 959 39.4 851 36.7
7. Walk-ins 39 2.9 104 4.2 69 2.9
8. Home births 23 1.7 40 1.6 35 1.5
B. Vaginal deliveries 1175 89.0 2132 87.7 2035 86.4
1. NSVD 1107 83.8 1997 82.1 1895 82.3
2. Outlet forceps 28 2.1 63 2.5 79 3.0
3. Breech 34 2.5 55 2.2 43 1.6
4. Mildpelvic delivery 5 0.3 15 0.6 27 1.1
A. Low or mild forceps 3 0.2 2 0.0 0.0
B. Mild forceps 2 0.1 10 0.4 13 0.5
C. Vacuum extractions 0.0 3 0.1 14 0.6
C. Abdominal deliveries 145 10.9 298 12.2 265 11.5
1. Private 76 5.7 193 7.9 168 7.3
2. Faculty 69 5.2 105 4.3 87 4.2
3. Low transverse 131 9.9 293 12.0 205 11.5
11. Methods for Safe Reduction of Cesarean Section Rates 161

Total % 1988 1987

4. Low vertical 11 0.8 4 0.1 0.0
5. Classical 3 0.2 1 0.0 0.0
6. Second opinion 24 20.1 18 7.5 0.0
7. Primary 99 7.5 191 7.8 169 0.9
A. Dystocia 44 44.4 76 39.7 54 33.7
B. Breech 13 13.1 20 10.4 27 18.1
C. Fatal distress 23 23.2 51 26.7 46 28.7
D. Others 19 13.1 44 23.0 31 19.3
8. Repeat 48 3.4 107 4.4 165 4.5
9. Hysterectomy 1 0.0 0.0 0.0
D. Previous cesarean section 125 9.4 275 11.3 285 11.5
1. Elective 12 9.6 32 11.6 37 13.9
2. Trial of labor 113 90.4 243 88.3 228 86.0
A. Vaginal 79 69.9 167 68.7 169 70.1
B. Failed trial 34 30.0 76 31.2 68 29.8
E. Breech deliveries 54 4.0 109 4.4 113 4.8
1. Spontaneous breech 3 5.5 16 14.6 18 15.9
2. Assisted breech 27 50.0 40 36.6 23 20.3
3. Breech extraction 3 5.5 3 2.7 7 6.1
4. Cesarean section 21 38.9 49 44.9 65 57.5
A. Breech primary 15 71.4 29 59.1 38 56.4
B. Breech repeat 6 28.5 20 40.8 27 41.5

F. Multigestational deliveries
1. Mothers 20 1.5 40 1.8 44 1.8
2. Babies 41 3.0 82 3.3 88 3.7
3. Babies, vaginal 25 1.8 52 2.1 53 2.2
4. Babies, cesarean section 16 1.1 30 1.2 35 1.4
A. Primaries 16 1.1 20 0.0 27 1.1
B. Repeats 0.0 10 0.4 8 0.3
G. Gestational hypertension 59 4.4 155 8.3 174 7.5
1. Mild preeclampsia 30 2.2 88 3.6 97 4.2
2. Severe preeclampsia 11 0.6 14 0.5 86 0.4
3. Eclampsia 1 0.0 3 0.1 2 0.0
4. Chronic hypertension 17 1.2 46 1.6 58 2.5
5. Mild preeclampsia and hyper-
tension 0.0 2 0.0 5 0.2
6. Severe preeclampsia and hyper-
tension 0.0 2 0.0 2 0.0
7. Eclampsia hypertension 0.0 0.0 0.0
H. Episiotomy
1. No episiotomy 565 48.0 1240 56.0 1183 56.9
A. No episiotomy or laceration 550 97.3 1211 97.6 1147 96.6
B. No episiotomy-third degree 12 2.1 17 1.3 12 1.0
C. No episiotomy-fourth degree 3 0.5 12 0.9 4 0.3
2. Episiotomy 811 61.9 897 41.9 879 43.0
A. Midline 256 41.8 441 49.1 504 57.5
1. Midline no laceration 237 92.5 374 94.5 438 86.7
2. Midline third degree 14 5.4 42 9.5 35 0.9
3. Midline fourth degree 5 1.9 25 5.6 32 0.3
B. Mediolateral 355 58.1 456 50.6 373 42.4
1. Mediolateral laceration 346 97.4 433 94.9 362 97.0
2. Mediolateral-third degree 7 1.9 23 5.0 7 1.8
3. Mediolateral-fourth degree 2 0.5 0.0 4 1.0
3. Cervical laceration 10 0.0 7 0.3 9.0
4. Vaginal laceration 8 0.0 4 0.1 8.0
162 N. Gleicher et al.

Total % 1988 1987

I. Miscellaneous complications
1. Premature labor 268 20.3 474 19.5 464 20.1
2. Induced labor 113 8.5 325 13.3 333 14.4
3. Previa abruption 15 1.1 27 1.1 22 0.9
4. Srom before labor 210 15.9 480 19.7 412 17.9
5. Drug abuse total 96 7.2 123 5.0 126 5.2
A. Suspected 15 1.1 11 0.4 12 0.5
B. Admitted 81 8.1 112 4.6 106 4.6
6. 1 min apgar <7 99 7.3 257 10.3 253 10.7
7. 1 min apgar <3 7 0.5 44 1.7 41 1.7
8. 1 min apgar <7 16 1.1 70 2.0 76 3.2
9. 1 min apgar <3 2 0.1 12 0.4 6 8.2
10. Stillbirths >500 g 26 1.9 35 1.4 21 0.0

ApPENDIX III. Mount Sinai- in early labor and 70 torr in late labor were
expected. Oxytocin was a:dministered freely to
Chicago CPGs for the Most achieve this goal.
Frequent Cesarean Section 3. Fetal Distress
Indications A diagnosis of fetal distress, based on moni-
toring of the fetal heartrate, had to be cor-
1. Previous Cesarean Section roborated by sampling of blood from the fetal
The department recognized in principle that scalp whenever technically feasible. A maternal-
vaginal delivery was preferred for all patients fetal pH difference of 0.20 to 0.28 was required for
who had previously undergone a cesarean section. confirmation, depending on the maternal pH. The
This included patients with vertical scars in the pH of the unbilical cord blood had to be obtained
lower segment of the uterus, those who had immediately after the birth of every infant
undergone classic cesarean sections or myo- delivered by cesarean section because of fetal
mectomies in which the uterine cavity had been distress. The result was used in the subsequent
entered, and those with previous uterine inci- case review.
sions of unknown type. Patients were usually 4. Breech ~resentation
advised during a counseling session at their first' Vaginal delivery was recommended for all
prenatal visit that a trial oflabor was anticipated. breech fetuses with the exception of those with
2. Dystocia/Cephalopevic Disproportion true hyperextension of the cervical spine or
The diagnosis of dystocia was accepted as an macrosomia. Macrosomia was defined as an
indication for a cesarean delivery only after no estimated fetal weight of more than 4300 g,
progress of labor was observed for more than 2 h determined by ultrasound with use of standard
of regular uterine contractions of appropriate nomograms for fetal biparietal diameter and
strength, as evaluated by intrauterine pressure abdominal circumference. The parity of the
catheters. A contraction frequency of once every 2 patient and type of breech presentation were not
to 3 min and a contraction intensity of 50 torr considered in the decision-making process.
12
Cesarean Delivery:
A Medical-Legal Perspective
JEFFREY P. PHELAN

Since the NIH consensus conference on simultaneously balance the interests of two
cesarean birth in 1980,1 the cesarean de- and sometimes more individuals. When it
livery rate has continued to rise. 2,3 Today, comes to cesarean delivery, the balancing
cesarean delivery continues to be the num- continues. Should a cesarean delivery be
ber one hospital-based operative procedure performed and thus expose the mother to a
in the United States. This rise has con- greater risk of morbidity and mortality? Or
tinued despite the increased availability of should vaginal delivery be allowed because
cesarean alternatives such as vaginal birth it reduces these maternal risks, yet exposes
after cesarean4 - 6 and external cephalic the fetus to a greater one? It may be that the
version. 7 - 10 circumstances of the case make cesarean
In fact, the rise in cesarean births appears delivery the only apparent option. Under
to be caused by factors that do not readily different circumstances, vaginal rather than
lend themselves to statistical analysis. For cesarean delivery would have been the
example, the incidence of cesarean births best choice.
has paralleled the rise in malpractice claims. It is these complexities that have fre-
During the 1980s, the average number of quently contributed to the rise in malpractice
paid claims rose 4-50%Y Currently, more claims and lawsuits. For these reasons, this
than 80% of the fellows of the American chapter is devoted to the medical-legal issues
College of Obstetricians and Gynecologists associated with cesarean birth, and reviews
(ACOG) have been sued at least once 12 and the basic concepts of the duty to refer, the
many members more than once. Of greater doctrine of informed consent, timely cesarean
concern to the practicing obstetrician is the delivery, and foreign bodies left in the
fact that the rise in malpractice claims has abdomen.
been accompanied by an increase in the A detailed analysis of each of these issues
number of awards in excess of $1 million. is not within the scope ofthis chapter. First,
Since 1975, these awards have risen 232 the laws vary from state to state. Second, the
foldP By 1994, the midpoint brain-injured individual circumstances of each case vary.
baby jury award was estimated to be Third, a purpose of this chapter is to provide
$2.5 million. the reader with an overview of the medical-
In obstetrics, the cesarean problem is legal issues that could arise from the perfor-
complicated by parental expectations of a mance of a cesarean. If a more detailed
"perfect child" and the belief that cesarean analysis with applicable state law is desired,
delivery can provide that child. Notwith- the reader is referred to an attorney in the
standing, and unlike other medical special- reader's state or jurisdiction. Finally, this
ists, the obstetric health care provider must chapter is intended to be used for educational

163
164 J.P. Phelan

purposes and is not designed to provide legal TABLE 12.2. Clinical situations in which the
advice. failure to consult a specialist at the time of
cesarean could arise
Intraoperative complications, e.g., bladder, bowel, or
Referral to a Specialist ureteral injury
Intensive care unit admissions
The technological advances and the infor- Postpartum hemorrhage with or without hysterectomy
Hypogastric or uterine artery ligation
mation explosion of the past 20 years have
made it harder for the practicing physician
to keep pace with the changes in medicine.
Sub specialization has been a necessary could be referred to these subspecialists,
outgrowth of these advances. The net effect not necessarily for primary prenatal care,
of sub specialization has been to make Amer- but for consultation. If, for example, an
ica the number one health care system in obstetrician-gynecologist fails to refer a
the world. Obstetrics and gynecology has high-risk obstetrical patient to a perinato-
also followed the subspecialization pathway. logist, that physician could be held to the
During the past 20 years, for example, fetal standard of care of the perinatologist (Table
monitoring and ultrasonography have been 12.1). This means a higher standard of care
introduced into the contemporary practice could be applied to the general ob-gyn.
of obstetrics. In many respects, the three For example (Table 12.2), assume a bladder
obstetrics and gynecology subspecialty laceration occurs during the performance of
areas-gynecologic oncology, reproductive a cesarean. Under most circumstances, a
endocrinology, and maternal-fetal medi- fully trained specialist in obstetrics and
cine-are a necessary outgrowth of the gynecology can repair this laceration without
technological and information explosion of calling in a consultant. But, a more extensive
the 1970s and 1980s. Initially, these spe- laceration of the bladder, bowel, or ureter,
cializing physicians limited their practices might, depending on the learning, skill, and
to university hospitals. It is estimated, experience of the obstetrician and the cir-
however, that, in the past 5 years the number . cumstances of the case, may require a phy-
of maternal-fetal subspecialists prac- sician more highly skilled in the repair of
ticing in community hospitals has risen these complications. This could include, but
exponentially. is not limited to, a urologist, a gynecologic
Nevertheless, a nonspecialist, general oncologist, or a general surgeon. If the repair
obstetrician-gynecologist, appears to have a is undertaken, however, by a nonspecialist
duty to refer a patient to a specialist or and the patient sustains a causally related
to request the assistance of a specialist injury, the nonspecialist could be held to the
whenever circumstances warrant it. For standard of care of one of these specialists.
example, with the increased availability of Thus, the underlying rationale for this
maternal-fetal medicine subspecialists, an principle is to encourage frequent consulta-
increasing number of high-risk patients tion in an effort to improve the quality of
care. After all, more frequent consultation
provides a direct benefit to the patient and
TABLE 12.1. Commonly used bases for the pos- an indirect one to her primary obstetrician.
sibility of a failure to refer the patient to a The primary issue in today's obstetric
specialist practice, however, is whether the specialist
A careful obstetrician/gynecologist would have referral will be authorized? During the past
recognized a problem existed decade, we have witnessed an expansion in
The obstetrician/gynecologist was not capable of health maintenance organizations with their
handling the problem
A timely referral was not made
nonphysician gate-keeper control of medical
decision making. Not infrequently, referrals
12. Cesarean Delivery: A Medical-Legal Perspective 165

to specialists are not authorized and the sufficient information to make an in-
burden of care will ultimately rest with the formed choice.
general ob-gyn. How does one in this practice The underlying rationale for the doctrine
setting reconcile the catch-22 of pericesarean of informed consent is illustrated in Cobbs v.
complications? The cornerstone of referrals Grant, 8 Cal 3d 229, 104 Cal Rptr 505, 502
is an awareness of one's limitations. This P2d 1 (1972). In that case, the patient agreed
does not mean that the general ob-gyn is not to undergo surgery for a duodenal ulcer.
competent to repair a bladder laceration. However, the inherent risks associated with
The issue is twofold: one, our limited ex- the surgical procedure were not discussed
perience in handling selected complications; with the patient. Following the surgery, the
and two, our goal is to provide quality care patient sustained internal bleeding and
to our obstetric patients. This necessarily required a splenectomy. A month after the
means that frequent consultation benefits second surgery, the patient was readmitted
the patient and improves patient care. for a gastric ulcer and underwent a partial
gastrectomy. Subsequently, the patient was
readmitted for abdominal bleeding because
Informed Consent of an abdominal suture dissolving pre-
maturely. At issue in this case was whether
As patients have sought greater control of the patient had been adequately informed of
their health care, informed consent has the risks of surgery.
assumed a prominent role in the practice of What must be disclosed? In many respects,
medicine today. One concern among practi- the information to be disclosed is that which
tioners is not whether informed consent was you or I would want to know if we were
given but whether sufficient information considering a proposed medical treatment
has been provided to the patient to permit or surgical procedure. This would include
an informed choice. 14 In obstetrics, the sufficient information for the patient to
adequacy of informed consent becomes com- make an informed decision. Frequently, the
plicated by the pain oflabor, analgesics that difficult part is deciding what is considered
can affect the patient's mental state, and the material. This decision often relates to the
expanding rights of the fetus. 15 It is these informed consent requirements in your state.
complexities that make each patient de- In a patient-oriented standard (minority
pendent on her physician to provide sufficient rule), the obstetrician would have to consider
information before undertaking the proposed what a reasonable patient would consider
operation, such as a cesarean. significant. 14 In contrast, the professional
What constitutes adequate patient in- standard (majority rule) is directed at what
formed consent varies from state to state a reasonable obstetrician would disclose
and according to the circumstances of the under the same or similar circumstances. 14
case. 16 But, there appears to be a common Under a patient-oriented standard, the
thread. For example, the doctrine of informed patient is entitled to know about any known
consent focuses on the characteristics of the risk of death or serious illness. 14 At the
physician - patient relationship. 17 First, same time, the disclosure or explanation to
knowledge ofmedicine between the physician the patient should be given in lay terms or
and the patient is not the same. Second, a at the patient's level of understanding. What
competent adult has the right to decide if there are lesser risks associated with a
whether to agree to lawful medical treatment. proposed procedure or treatment? Under
Third, before submitting to the treatment, these circumstances, disclosure of lesser
the patient has the right to be informed. risks may depend on whether physicians
Fourth, the patient is unlearned in the practicing under the same or similar cir-
medical sciences and, as a result, is de- cumstances would be required to disclose
pendent on her physician to provide her with those risks. But where the procedure is
166 J.P. Phelan

simple and the danger is remote and com- either no treatment or a different one. Fi-
monly appreciated to be remote, such as nally, the patient would have to show that
drawing blood or the patient's request not to
she was injured as a result of submitting to
be informed, there may be no duty to disclose
treatment. What consequences will follow if
the risks of that procedure. the patient has consented to a proposed
With respect to cesarean delivery, the treatment or operation and the physician
patient is generally entitled to be informed
has failed to inform the patient adequately?
of the reason for the cesarean, the risks and
The failure to obtain "such consent is negli-
potential complications of the procedure, gence and renders the physician or surgeon
alternative treatment approaches such as subject to liability for any injury resulting
external cephalic version or vaginal birth from the [treatment or operation] if a rea-
after cesarean, and the consequences if thesonably prudent person in the patient's
procedure is not performed and the problemsposition would not have consented to the
of recuperation. Equally important is the [treatment or operation] if he had been
fact that the performance of a cesarean adequately informed of all the significant
perils.,,2o
without the patient's consent 18 or where the
consent is deemed invalid 19 could be con- Informed refusal or the consequences of
sidered battery. nontreatment was established in a California
Even though informed consent appears to case, Truman v. Thomas. 21 In that case,
be an infrequently litigated issue today,16 it
Mrs. Truman saw her physician on several
is important to understand how these cases occasions between 1964 and 1969. Her
might be approached. As noted in Table 12.3,
physician, however, never obtained a Pap
the basic elements of negligence are also smear, nor did he specifically inform her of
involved. But, once a doctor-patient re- the risks involved in a failure to perform
lationship is established, the question be-this test. He said "You should have a Pap
comes whether the physician breached her smear ," but this was not supported by the
duty by failing to give sufficient information
medical records. Subsequently, she was
so that a proposed treatment could be chosen
diagnosed as having inoperable carcinoma
intelligently. This again will depend on of the cervix and died at age 30. Suit was
the applicable informed consent standard, brought by her dependent children. The
whether patient oriented or professional. finding of the court was that the patient
This will frequently focus on whether the should have been given adequate information
physician informed the patient of alternative
regarding the consequences of refusing to
treatments, the reasonably foreseeable risks
undergo a Pap smear.
of each alternative, the consequences of no The failure to inform the patient of alter-
treatment, and items of a personal interest.
native treatment approaches may also serve
Additionally, if the material risk of an as a basis for medical malpractice claims.
alternative treatment has been disclosed, Common clinical examples are demonstrated
the patient, it seems, could have chosen in 'Table 12.4. To illustrate, the recent re-
surgenc~ of vaginal birth after cesarean
where it has been shown that maternal
morbidity and mortality can be significantly
TABLE 12.3. Basic elements of adequate in- reduced has made the vaginal birth option
formed consent for a cesarean birth available to prior cesarean patients. For
Indications for procedure example, the performance of an elective
Alternative treatments repeat cesarean delivery without first dis-
Material risks of each option closing to the patient the trial-of-Iabor
Problems of recuperation alternative could be an example of inade-
Consequences of nontreatment
quate informed consent. This means that the
Items of personal interest
physician could be held liable even if a re-
12. Cesarean Delivery: A Medical-Legal Perspective 167

TABLE 12.4. Selected clinical examples and the Failure to Perform a Cesarean
options available in each one
Example Options
in a Timely Manner
Breech Selected vaginal delivery
"Cases frequently are generated by the
presentation External cephalic version failure of physicians or staff to recognize
Cesarean delivery and manage immediate or imminent emer-
Prior cesarean Vaginal birth after cesarean gencies. Failure to properly assess and
Repeat cesarean react is fertile ground for a negligence
Postdates Induction of labor with or
without cervical ripening
claim.,,24 But, when is it a failure to perform
Fetal surveillance testing a cesarean in a timely manner? According to
Cesarean delivery the most recent guidelines of the ACOG,25
Transverse lie External version with attempted any facility that provides obstetric care
vaginal delivery should have, as a minimum, a "cesarean
Cesarean delivery
delivery capability within 30 minutes." This
would imply, therefore, that a decision-
incision interval in excess of 30 min could
constitute a breach in the standard of care. If
peat cesarean is performed nonnegligently. the decision - incision interval does exceed
During the last decade, informed consent 30 min and the infant is born with a neuro-
requirements have changed dramatically. logic handicap, should the physician be held
For example, consent guarantees have liable for those injuries? For this reason,
recently been recommended in a North there may be a "fertile ground for a negli-
Carolina case. 22 In this case, the hospital gence claim," but it does not necessarily
was found liable for failure to assure a pa- follow that the obstetrician-gynecologist
tient with a breech presentation that her proximately caused the injuries. Circum-
obstetrician had properly informed her of . stances may be such that cesarean delivery
the risks of vaginal birth. In that case, Mrs. may not have been technically feasible
Campbell was permitted to deliver her known within the "allotted time."
breech infant vaginally. During the course Moreover, there is a popular misconception
of delivery, a prolapsed cord occurred, and that neonatal brain damage is proximately
the infant became severely asphyxiated and related to the events of labor and is caused
is now neurologically impaired. In holding by not the natural course of events of preg-
against the nurse-hospital, the Campbell nany but the negligent conduct of the
court stated that " ... explaining the risk of attendant physician, nursing staff, or hos-
alternative procedures [is] the responsibility pital. As sugested by Perkins,26 "the number
of the physician ... assuring that the patient of infants injured before labor is highly
has had an explanation [is] the responsibility underestimated [and those infants injured]
of the nurse.,,22 during labor is highly overestimated." In
Recently, items unrelated to a patient's support of Perkins' assertion, Paul and as-
health but of a personal interest to a phy- sociates27 have demonstrated that fetal
sician must be disclosed. In a 1990 California injuries such as intracranial hemorrhage,
case,23 a physician must disclose " ... per- myocardial infarction, and meconium as-
sonal interests unrelated to a patient's health piration can and do occur before the onset
whether research or economic, that may oflabor.
affect the physician's judgment." In that What then is the role of electronic fetal
case, a University of California at Los monitoring? Although "fetal heart rate
Angeles (UCLA) physician had substantial changes can provide an indicator of hypoxia
propriety interest in the selling of a patient's that precedes neurologic damage,,,28 the fetal
blood without first disclosing that interest to heart rate pattern at the onset of labor can
the patient. be indicative of fetal prelabor injury.29 This
168 J.P. Phelan

suggests, therefore, that "intrapartum as- or death. With in-house anesthesia, cesarean
phyxia may be the result rather than the delivery can readily be achieved. But in-
cause of neurologic abnormality.,,30 More- house anesthesia is not always encountered
over, neonatal compromise may represent a where obstetric services are provided. Under
continuum rather than a single isolated these circumstances, cesarean delivery would
event that "is not ... influenced markedly by be necessarily delayed while waiting for
the circumstances of labor and delivery.,,26 anesthesia personnel to come to the hospital.
This is not to say that an adverse fetal out- If anesthesia personnel are not in the hos-
come cannot be related to the events oflabor pital, several options are available. Two
or to the conduct of the obstetrician or nurse, options are available to correct the fetal
but rather that the failure to perform a ce- heartrate pattern. One is to attempt to arrest
sarean within a specified period of time is uterine activity with a beta-mimetic such
not necessarily the cause of the neonate's as terbutaline. 32 The other option is to re-
injuries. establish the amniotic fluid volume with
Nevertheless, the physician should make saline amnioinfusion. 33 This would allow
reasonable efforts either to evaluate a pa- intrauterine resuscitation of the fetus and
tient [Thomas v. Ellis, 329 Mass 93, 106 NE provide additional time for anesthesia
2d 687 (1952)] or to begin a cesarean in a support personnel to arrive. If the beta-
timely manner.31 For example, in Thomas v. mimetic is contraindicated, technically not
Ellis, the pregnant woman's husband first feasible, or unsuccessful, cesarean delivery
notified her physician around 11:30 P.M. that under local anesthesia is a reasonable con-
she was bleeding heavily from her vagina. sideration. 34
At that time, the husband was advised to In summary, reasonable efforts to evaluate
keep her at home and observe her for ap- a patient or to perform a cesarean in a timely
proximately 1 h. About a half-hour later, the manner would appear prudent. Whether
husband called the physician again because this will have an effect on the incidence
there was "more blood all over the blanket, a of neonatal injury or death remains to be
lot of blood" and the patient had pain. The proven.
physician told the husband that it sou¢ded
like an abruption. He advised them to go
directly to the hospital and stated that he Failure to Perform a Cesarean
would meet them there. The couple arrived Properly
at around 12:30 A.M. The record was unclear
as to when the physician arrived, but a ce- The failure to perform a procedure, such
sarean was done about 11:00 A.M. At the as the administration of rhogam to an
time of delivery, an abruption was confirmed Rh-negative gravida, may give rise to a
and the fetus was dead. claim against the physician for maternal
This case illustrates the importance of a or neonatal injuries in a subsequent preg-
timely evaluation and delivery. An earlier nancy.35 This concept has been extended to
evaluation and delivery might have provided include a host of cases in which physicians
a more favorable outcome for the fetus. But have been held liable for a child's injuries
because that evaluation and delivery were that resulted from negligence toward the
not undertaken earlier, we will never know child's mother in a prior pregnancy.36 This
whether it would have made a difference in extension includes the failure to perform a
this case. However, a timely evaluation is cesarean properly during an earlier preg-
the first step in minimizing the risk of harm. nancy. In Bergstreser v. Mitchell, 577 F2d22
In the event of acute fetal distress such (CA8 MO, 1978), the plaintiff had undergone
as an acute, prolonged fetal heartrate a cesarean in a previous pregnancy. The
deceleration, the timeliness of the cesarean prior incision, however, was not described in
may protect the fetus from neurologic injury the legal summary. At approximately 30
12. Cesarean Delivery: A Medical-Legal Perspective 169

TABLE 12.5. Documentation considerations in nurse. Once completed, the results of the
the patient with a classical or upper-segment count are given aubibly. Before the end of
cesarean incision the cesarean, none of the sponges, needles,
The type of incision or instruments should leave the operating
The patient has been told of the incision room. Before and after the closure of the
The patient has been advised that attempted vaginal peritoneum, the count is done again and the
delivery in a future pregnancy is not recommended
results are given audibly to the surgeons
The risk of uterine rupture in a subsequent pregnancy
and scrub nurse. The numbers must match.
If the count indicates a missing sponge,
needle, or instrument, a search for the
weeks gestation in a subsequent pregnancy, missing item is appropriate. The failure to
she sustained an uterine rupture that re- search for a missing item could be evidence
sulted in an exploratory laparotomy and of negligence [Smedra v. Stanek, 187, F2d
delivery of the infant. The child suffered 892 (CA 10 COLO 1951)]. Fortunately, most
brain damage. The plaintiff alleged that the items are radiopaque or, in the case of
injuries sustained by the infant resulted sponges, contain a radiopaque tape. Thus, if
from the negligent performance of the prior manual and visual exploration is unsuc-
cesarean and the failure to inform her of the cessful, an abdominal x-ray may help locate
prior uterine incision. the missing item.
In this era, where vaginal birth after ce- Reliance on the needle and sponge count
sarean is an acceptable alternative,4-6 it is alone may not support the claim that a phy-
important to inform the patient of the type sician exercised due care. 37 For example, in
of uterine incision performed and its con- the case of Key v. Caldwell [139 Cal App 2d
sequences (Table 12.5). In addition, reason- 698,104 P2d 87 (1940)], a laparotomy sponge
able efforts to avoid a cesarean birth or to was left in a patient's abdomen following a
reduce the necessity for an upper-segment cesarean delivery. The needle and sponge
incision with external cephalic version 7- 10 count was done and was said to be correct.
would appear prudent. If this is done, the Testimony demonstrated that the surgeon
likelihood of liability for a subsequent preg- in this case had the practice of placing a
nancy might be lessened. laparotomy sponge in the same location in
the abdominal cavity at each cesarean.
Althogh the physician attempted to rebut
Foreign Bodies Left in the the allegations, the sponge had nevertheless
Patient Following a Cesarean not been removed.
The Key case suggests that the surgeon
Throughout surgical history, numerous has an independent duty to determine
objects have been left in the abdomen fol- whether a sponge, for example, has been left
lowing surgery. These have included beetles, behind. Thus, in addition to an accurate
needles, drains, sponges, surgical instru- needle, sponge, and instrument count,
ments, forceps, and a cloth sac. 37 Although manual and visual exploration of the
foreign bodies may be more frequently left abdomen would also appear prudent.
behind during extensive surgery, a needle, When an instrument has been broken38
sponge, and instrument count is designed to during a cesarean, reasonable efforts to locate
protect against this event. The failure to the missing object would also be appropriate.
follow such a procedure could be evidence of These efforts may be sufficient if the item is
negligence [Bowers v. Dlck, 120 Cal App 2d not located [Roark v. Peters, 162 La 111, 110
108,260 P2d 997 (1953)]. So 106 (1926)], but the patient is entitled to
Before the performance of a cesarean, a be informed of the missing object unless
needle and sponge count is done by the there are medical reasons for not informing
surgical scrub nurse and the circulating her.37
170 J.P. Phelan

TABLE 12.6. At the time of a cesarean, the 6. Flamm BL, Newman LA, Thomas SJ, Fallon
count: what you should include 39 ,40 D, Yashida MM. Vaginal birth after cesarean
delivery: results of a 5-year multicenter col-
Needle
Sponge laborative study. Obstet Gynecol Surv 1991;
Instrument 46:360-365.
Fetal scalp electrode 7. Phelan JP, Stine LE, Edwards NB, et al. The
role of external cephalic version in the intra-
partum management of the transverse lie
presentation. Am J Obstet Gynecol 1985;
If an item is discovered to be missing after 151:724-726.
a cesarean, is consent required to remove it? 8. Dyson DC, Ferguson JE, Hensleigh P. Ante-
According to Jones,37 the operating surgeon partum external cephalic version under
can perform a second operative procedure for tocolysis. Obstet Gynecol 1986;67:63-67.
the purpose of removing a foreign body 9. Morrison JC, Myatt RE, Martin IN, et al.
immediately following the primary procedure External cephalic version of the breech pre-
without the consent of the patient. If, how- sentatoin under tocolysis. Am J Obstet
ever, a sponge is found to be missing some Gynecol 1986;154:900-905.
time after the cesarean, full disclosure and 10. Stine LE, Phelan JP, Wallace RL, et al.
consent would be appropriate 37 [Delahunt v. Update on external cephalic versoin per-
formed at term. Obstet Gynecol
Fenton, 244 Mich 226, 221 NW 168 (1928)].
1985;65:642-646.
In summary, needle, sponge, and instru- 11. Jacobson PD. Medical malpractice and the
ment counts are an important part of a ce- tort system. JAMA 1989;262:3320-3327.
sarean delivery. With a cesarean, the fetal 12. Swan JE. Your views on the crisis. Contemp
scalp electrode should also be counted (Table Ob/Gyn 1989;33:13-24.
12.6). In addition, a manual and visual 13. Harrison TF. The top ten jury awards of
exploration of the abdomen before closure 1990. Lawyers Alert 1991;11:4-16.
would appear prudent. Consideration should 14. Appelbaum PS, Lidz CW, Meisel A. The legal
be given to document the manual and visual requirements for disclosure and consent:
exploration in the medical records. If it is history and current status. In: Informed
Consent: Legal Theory and Clinical Practice.
subsequently determined that an item is
New York: Oxford University Press, 1987:
missing, the physician should tell the patient 35-65.
and obtain her consent to remove it. 15. Phelan JP. The maternal abdominal wall: a
fortress against fetal health care? South Cal
References Law Rev 1991;65:461-490.
16. Appelbaum PS, Lidz CW, Meisel A. Informed
1. Cesarean Childbirth Report of a Consensus Consent: Legal Theory and Clinical Practice.
Development. (DHHS publication no. 82- New York: Oxford University Press, 1987.
2076.) Conference sponsored by the National 17. Cobbs v. Grant, 8 Cal 3d 229, 104 Cal Rptr
Institute of Child and Human Development, 505, 502 P2d 1 (1972).
Washington, DC, 1980. 18. Mohr v. Williams, 95 Minn 261, 104 NW 2d
2. Taffell SM, Placek PJ, Liss T. Trends in the 523 (1905).
USA cesarean section rate and reasons for 19. Rainer v. Community Memorial Hosp, 18 CA
the 1980-1985 rise. Am J Public Health 3d 240, 95 Cal Rptr 901 (1971).
1987;77(8):955-959. 20. Book of Approved Jury Instructions. St. Paul:
3. Barr R. C-section under fire. OBG Manage- West, 1986.
ment 1990;(October):18-25. 21. Truman v. Thomas, 27 Cal 3d 285, 165 Cal
4. Phelan JP, Clark SL, Diaz F, et al. Vaginal Rptr 308, 611 P2d 902 (1980).
birth after cesarean. Am J Obstet Gynecol 22. Campbell v. Pitt County Hosp, 84 NC App
1987;157:1510-1516. 314,352 SE 2d 902 (1987) [decertified in later
5. Phelan JP, Ahn MO, Diaz F, et al. Twice a decision].
cesarean, always a cesarean? Obstet Gynecol 23. Moore v. Regents of California, 90 Daily
1989;73:161-165. Journal D.A.R. 8010 (1990).
12. Cesarean Delivery: A Medical-Legal Perspective 171

24. Olender JH. Obstetric negligence. Trial 33. Strong TJ, Phelan JP. Amnioinfusion: an
1984;May:52. intrapartum technique for the '90's. Contemp
25. Standards for Obstetric-Gynecologic Services, Ob/Gyn 1991;36(5):15-24.
7th Ed. Washington, DC: American College 34. Gilstrap LC, Hankins GDV. The uncom-
of Obstetricians and Gynecologists, 1989: plicated patient. In: Phelan JP, Clark SL,
33-39. eds. Cesarean Delivery. New York: Elsevier,
26. Perkins RP. Perspective on perinatal brain 1987:139-154.
damage. Obstet Gynecol 1987;69:807. 35. Renslow u. Mennonite Hasp, 67111 2d 348, 10
27. Paul RH, Yonekura ML, Cantrell CJ, et al. III Dec 484, 367 NE 2d 1250 (1977).
Fetal injury prior to labor: does it happen? 36. Liability for child's personal injuries or death
Am J Obstet Gynecol 1986;154:1187. resulting from tort committed against child's
28. Adamson SK, Myers RE. Late decelerations mother before child was conceived. 91 ALR
and brain tolerance of the fetal monkey 3d 316 (1979).
to asphyxia. Am J Obstet Gynecol 1977; 37. Jones JL. Malpractice: liability of physician,
128:893. surgeon, anesthetist, or dentist for injury
29. Phelan JP, Ahn MO. Perinatal observations resulting from foreign object left in patient.
in 48 neurologically impaired term infants. 10 ALR 3d 9 (1966).
Am J Obstet Gynecol 1994;171:424-431. 38. Zitter JM. Medical malpractice: instruments
30. Niswander K. Asphyxia in the fetus and breaking in course of surgery or treatment.
cerebral palsy. In: Pitkin RM, Zlatnik 20 ALR 45th 1179 (1986).
FJ, eds. 1983 Year Book of Obstetrics and 39. Thoulon JM, Gonnet C. Les accidents de
Gynecology. Chicago: Year Book Medical, l'electrocardiographic foetale directe (a propos
1983:107 -125. d'une de l'electrode de scalp). J Gynecol
31. Gee DR. Physician's failure to perform timely Obstet BioI Reprod 1978;7:1257.
cesarean. 19 POF 2d 285 (1979). 40. Nieburg P, Gross SJ. Breakage of a fetal
32. Patriarco MS, Viechnicki BM, Hutchinson scalp blade with retention of fragments
TA, et al. A study on intrauterine fetal re- in the infant's scalp. Am J Obstet Gynecol
suscitation with terbutaline. Am J Obstet 1987;157:441.
Gynecol 1987;154:384.
13
Economic Considerations in
Cesarean Section Use
LAURA B. GARDNER

Cesarean section was origillally indicated to $3386. 2 Total hospital charges for the
preserve the life of a mother with obstructed 934,000 cesarean sections performed in the
labor or to deliver a viable infant from a United States in 1987, which comprised
mother who was dying.! Although now in 24.4% of all deliveries,3 amounted therefore
widespread use in less dire situations, the to more than $3.1 billion in 1987 dollars. In
operative approach to delivery is associated comparison, the average charge for all other
with significantly greater economic costs hospital discharges with an obstetric proce-
compared with vaginal delivery. The in- dure as the principal procedure, exclusive of
creased cost of cesarean section is related normal deliveries, was $1587, or 53% lower
both to higher delivery charges and the costs than the charges for cesarean section. 2 For
of maternal and neonatal morbidity and discharges with a normal delivery, the aver-
mortality. In a health care policy environ- age hospital charge was $1396, almost 59%
ment that increasingly emphasizes cost- lower than for cesarean section.
effectiveness and cost containment, high These data suggest that decreasing the
rates of use of expensive health care tech- overall cesarean section rate would lead to a
nologies such as cesarean section must be substantial reduction in expenditures for
justified by improvements in outcome that obstetric hospitalizations nationwide.
are worth their cost. Assuming conservatively that for each de-
This chapter describes the nationwide livery in which a cesarean section was
economic cost of current levels of cesarean avoided there was instead a vaginal delivery,
section use, outlines the economic factors even one with one or more reportable com-
that influence and contribute to cesarean plications, the annual savings associated
section use, and reviews the clinical and with each percentage point reduction in the
economic issues to be considered in attempt- cesarean section rate would amount to more
ing to model the optimal choice of delivery than $68 million (1987 dollars). This estimate
modality using decision analysis theory. of savings does not include the additional
savings that would result from lower phy-
sician fees and avoided cesarean section-
Economic Impact of related morbidity and mortality.
Cesarean Section Use Experts have estimated that the ideal
cesarean section rate lies in the range of
In a 1987 nationwide sample of U.S. hospi- 10% to 18%.4-6 On the basis of these esti-
tals, the average total hospital charge across mates, if the cesarean section rate were
all payers for hospital discharges with cesa- reduced nationwide from 24% to 18% (the
rean section as the primary procedure was upper end of the estimated ideal range), the

173
174 L.B. Gardner

expected reduction in expenditures for embolism, postpartum infection, thrombo-

hospital services alone would surpass $435 phlebitis, wound dehiscence or infection,
million per year in 1987 dollars. and pyrexia) was more than 6 fold higher for
Cesarean section-related morbidity and patients receiving cesarean section than for
mortality has significant social and econo- those who had a nonoperative delivery. The
mic implications. The risk of maternal death presence of one or more of these postpartum
after cesarean section has been reported complications increased the mean inpatient
to be approximately 2 deaths per 1000 cost of hospital discharges with cesarean
operations, 1 or between 2 and 26 times that section as the principal procedure between
of vaginal deli very. 7-11 38% and 250%.
Maternal cesarean-related morbidity, Although use of cesarean section in ap-
which includes cerebrovascular events, endo- propriate clinical situations (e.g., clearcut
metritis, hemorrhage, thrombophlebitis, fetal distress) can dramatically reduce
wound and urinary tract infections, pulmon- neonatal morbidity, it has been shown that
ary embolism, and intestinal obstruction, higher cesarean section rates do not ex-
occurs in 25%-50% of patients. 1 Postpartum plain the trend toward decreased perinatal
infection occurs in as many as 20% of women mortality.4,S,17 A recent study of very low
delivering by cesarean section. 12 Overall, birthweight infants reported that there was
maternal cesarean section morbidity is asso- no improvement in the survival of such
ciated both with higher direct costs of care infants despite an increasing rate of both
and with indirect costs such as lost pro- cesarean section and active resuscitation. IS
ductivity, delayed recovery, and impaired If cesarean section rates are higher than
maternal-infant bonding. 13 clinically necessary, decreasing the number
Elective cesarean section occurring before of cesarean sections performed could be cost-
labor begins also carries the risk of iatro- saving without adversely affecting quality
genic (physician-induced) prematurity of care.
and respiratory distress syndrome in the
infant.14 Pulmonary complications are more
likely to occur with cesarean section than Economic Factors Influencing
with vaginal delivery, because optimal in- Use of Cesarean Section
fant lung function depends on the physical
pressures exerted on the fetal thorax during There has been much research to indicate
vaginal delivery. 15 A study of neonatal that economic and other nonclinical factors
outcomes reported by Martin et al. 16 con- contribute to the treatment decisions that
firmed that, controlling for the mother's per- physicians make in addition to clinical con-
sonal and clinical characteristics, babies siderations. In the case of obstetric decision
delivered by cesarean section were much making, the choice of cesarean section fre-
more likely to have health problems, many quently has been shown to be associated
of which could be attributed to the use of with factors other than the objective clinical
cesarean section. risks and benefits of cesarean delivery. Non-
Analysis of California hospital discharge clinical and economic factors that have been
abstract data (California Office of Statewide shown to be associated with the rate of use of
Health Planning and Development) con- cesarean section include patient prefer-
firms that in 1991 the risk of sudden mater- ences, payment incentives, physician and
nal death was 2.2 fold higher for patients hospital characteristics, and the organiza-
receiving cesarean section than for those tion of medical care delivery. 19-22
who had a nonoperative delivery. The com- Economic theory holds that self-interested
bined risk of the five most serious post- suppliers of services, in furtherance of their
partum complications other than maternal own economic interests, will seek to maxi-
death (cerebrovascular disorder, pulmonary mize both the profitability of their services
13. Economic Considerations in Cesarean Section Use 175

and their personal satisfaction. 23 - 25 In dis- physician faces is the time spent to provide
cussing the influence of economic incen- patient care. The economic value of time,
tives in health care, Luft26 , suggested that known as the "opportunity cost," is defined
"the rapid growth in the use of medical as the value of the next best use that could
technologies may stem not just from clinical have been made of that time. Luft26 noted
efficacy but also from the high returns phy- that technology can often substitute for phy-
sicians receive by using such technology" (p. sicians' time and effort, allowing physicians
505). The following discussion focuses on to use that time elsewhere to their net
physician- and hospital-level economic advantage.
factors that have been shown to influence Cesarean section deliveries require much
the use of cesarean section. less total physician time than do vaginal
deliveries. Physicians acknowledge that
they often feel pressured to make "efficient"
Physician-Level Economic use of their time when attending a delivery.
Incentives Thus, in a fee-for-service setting the incen-
tive to minimize time spent attending a
The most direct economic incentive that prolonged delivery is twofold: (1) the physi-
may influence a physician's decision making cian's current income is directly dependent
in a fee-for-service system is the relative on the total number of patients seen, and (2)
reimbursement amount. In particular, the viability of the physician's practice and
the difference in reimbursement between thus its future income are affected by the
cesarean section and vaginal delivery has attractiveness of the practice to new pa-
been cited as a contributor to the high tients, who do not generally appreciate hav-
rate of cesarean deliveries. Tussing and ing to wait in the office while the physician
Wojtowycz27 reported a positive and signi- is in the hospital. Tussing and WojtOwycz 27
ficant association between the ratio of reported that weekend deliveries (a proxy
physician 'charges for cesarean section to for deliveries less apt to be affected by a
physician charges for vaginal delivery physician's need to return to patients in the
and the probability of a cesarean section. office) were significantly less likely to be by
The reimbursement disparity between cesarean section.
cesarean section and vaginal delivery is con- Conversely, in systems in which phy-
siderable. Data from a private payer in sicians are prepaid, capitated, or salaried,
Northern California indicate that physician economic factors may be a disincentive
fees for deliveries in 1989 differed by 42% for to the use of higher cost treatment
the two procedures, with an average fee of alternatives. 22 ,30 In numerous analyses,
$1377 for a vaginal delivery and $1951 for a health maintenance organization (HMO)
cesarean section, both including prenatal patients have been shown to have lower
care28 (and Robinson et aI., unpublished cesarean section rates than fee-for-service
data). A recent survey of doctor charges patients. For example, Wilner et al. 31
indicated that nationwide the median physi- observed higher primary cesarean section
cian fee for a cesarean section in 1993 rates for physicians paid on a, fee-for-service
was $2400, or 20% higher than the $2000 (FFS) basis (14%) than for HMO physicians
median physician fee for vaginal delivery.29 in the same hospital (11 %), with apparently
In addition to the influence of reimburse- equal quality of outcome after risk adjust-
ment differentials, there are indirect eco- ment. Wright et al. 32 found that although
nomic incentives that influence physician patients were being delivered by the same
behavior. When the supplier of a service is physicians, FFS patients had a primary
able to decrease the costs of production, he is cesarean rate of 15%, compared with 8% for
able to provide the service at a greater HMO patients, again with no apparent dif-
profit. One of the main production costs a ference in outcome.
176 L.B. Gardner

McCloskey et al. 33 reported that a woman reported to be a factor in the historical in-
with an HMO physician was 78% as likely crease in cesarean section rates, especially
to have a cesarean as a woman delivered in the rate of repeat cesarean section,36,37
by a physician in private practice. Stafford and to the risk of malpractice liability.38
et al. 34 reported that in California, Kaiser Thus, physician coverage, convenience,
HMO patients treated in Kaiser hospitals, and clinical uncertainty are factors with
which pay physicians by salary, were ob- economic implications that contribute, along
served to have a lower primary cesarean with the reimbursement differential effect,
section rate than patients of non-Kaiser to increasing the use of cesarean section.
HMOs, which pay physicians on a nego- The theoretical basis for the reimburse-
tiated fee-for-service basis (18.1% compared ment differential effect is grounded in the
with 26.5% in 1989). The cesarean section principal agent economics concept of "sup-
rate in non-Kaiser HMO patients was closer plier-induced demand." This term generally
to the 28.7% rate seen in patients with pri- has negative connotations because it implies
vate fee-for-service insurance. The authors that physicians are attempting to provide a
noted that the difference in cesarean section level of care that is beyond what is in the
rate between members of Kaiser and non- best interests of patients. However, as
Kaiser HMOs lends support to the hypothe- Evans23 (p. 164) has pointed out, "Such a
sis that both the method of physician reaction [i.e., attempting to increase the
payment and the organization of medical level of care provided] by physicians should
care delivery affects cesarean section not be interpreted as the deliberate provi-
rates. sion of unnecessary care. If physicians be-
The presence of a reimbursement differen- lieve that the public 'needs' more care [for
tial effect on the cesarean section rate is a clinical reasons] than it now receives ...
characteristic of a fee-for-service system. then they may well [attempt] to take better
However, other economic incentives that care of their patients through more frequent
favor cesarean section, such as physician recalls and follow-ups, more extensive
convenience, are relevant in both fee-for- testing, consultations, and more services
service and capitated or salary-based sys- generally."
tems. Fraser et al. 21 reported a small but With regard to physician convenience,
significant trend toward higher rates of McCullough39 argued that physicians have
cesarean section during evening hours, con- a legitimate self-interest that is compatible
sistent with the hypothesis of physician with their ethical obligations to their pa-
convenience. Also consistent is the finding tients. McCullough's concept of legitimate
that physicians who shared night call with self-interest suggests that physician sa-
other colleagues had lower cesarean section tisfaction depends not just on economic
rates than those who were responsible for rewards but also on having both the oppor-
their own night coverage. 20 tunity for activities that result in improved
Finally, another key factor in the degree patient care, such as "the time to study,
of a physician's susceptibility to nonclinical reflect, and learn and the time to rest and to
influences on decision making is the degree maintain an alert mind" (p. 12), and to
of clinical uncertainty. Hillman35 has noted spend the time necessary to fulfill obliga-
that physicians will generally act in the tions to persons and responsibilities outside
patient's best interest whenever the medical the medical network, such as family and
decision is clear cut, but that financial con- community.
siderations may come into play when the Physicians who are responsible for attend-
"correct" decision is uncertain. Uncertainty ing their patients' deliveries during the
also predisposes the physician to be more same hours that they are scheduled to see
sensitive to factors such as patient prefer- patients in the office feel the greatest pres-
ence for cesarean section, which has been sure to speed a delivery and return to the
13. Economic Considerations in Cesarean Section Use 177

office. It is reasonable to suggest that the and physicians. 22 Not uncommonly, physi-
organization of medical care delivery could cians and hospitals face conflicting economic
be improved so as to address issues such as incentives. 40 For example, HMOs and pre-
coverage and the conflicting demands On ferred provider organizations (PPOs) often
physicians' time. In fact, many HMOs are use different methods to reimburse physi-
arranged so as to provide continuous cover- cians and hospitals; that is, hospitals may be
age, which allows physicians to leave at paid a fixed amount per admission while
the end of their "shift." Obstetricians, for physicians are paid on a fee-for-service basis.
example, would be relieved from being re- This dichotomy places a hospital's fixed
sponsible for both delivering patients at the reimbursement incentive (to manage the
hospital and seeing patients in the office cost of the admission) in conflict with the
during the same time period. This type of physician's fee-for-service incentive (to en-
arrangement neutralizes the economic pres- hance reimbursement by increasing the
sure On obstetricians to choose cesarean number or the intensity of services pro-
section when faced with a prolonged de- vided). Conversely, if the hospital is paid
livery. The strong organizational culture in fee-for-service or per diem, the hospital's
certain HMOs such as Kaiser promotes peer incentive favors clinical choices that prolong
interaction and standard setting as well, the patient's length of stay. However, physi-
thus providing a measure of peer support in cians would benefit more from higher charges
situations of clinical uncertainty. than from longer patient stays because rela-
tively fewer physician services are needed
during the later part (recovery portion) of
Hospital-Level Economic Incentives
the patient's admission.
Various hospital characteristics have been Hospitals that respond successfully to
found to be associated with a hospital's economic incentives are more likely than
cesarean section rate. These include hospital those that do not so respond to have a process
ownership and economic organization, by which hospital administrators monitor
teaching status, size, location, and techno- and attempt to influence physician deci-
logical sophistication (such as having a sion making. Gardner22 reported that the
neonatal intensive care unit). For example, hospital-physician relationship in hospitals
a lower hospital occupancy rate has been of different ownership types can be charac-
shown to be associated with a higher cesarean terized with respect to two specific features:
section rate. 22 This is consistent with the (1) the degree to which the hospital and the
expectation that the lower the hospital's physicians share a common set of values and
occupancy rate, the greater the incentive to goals in response to existing economic incen-
use a treatment choice that entails a longer tives, and (2) the presence of salaried, ver-
average length of stay. tically integrated, or exclusive arrangements
As with physician fees, hospital inpatient between the hospital and its physicians by
charges for cesarean section deliveries which the hospital monitors, controls, or co-
are significantly greater than charges for ordinates physician behavior. Together
vaginal deliveries. The average hospital these features comprise a concept termed
charge for uncomplicated cesarean sections "close linkage." Gardner reported that
in California in 1991 ($5877) was 2.4 times the degree of close linkage in the hospital-
that for uncomplicated vaginal deliveries physician relationship, as represented by
($2404), according to hospital discharge data the hospital's ownership category, appeared
for all payers. to be significantly related to the hospital's
Elaborating the influence of hospital-level cesarean section rate.
economic incentives on use of medical care Why is the rate of cesarean section use as
services is made difficult by the increasingly high as it is across all hospital ownership
complex interdependence between hospitals categories? A pertinent explanation for this
178 L.B. Gardner

phenomenon was offered recently in a dis- 1. The clinical course, expression, and out-
cussion of the discrepancy between physi- comes of the condition being treated
cians' decisions for individual patients versus 2. The baseline economic impact of the
those for groups. Redelmeier and Tversky41 condition
observed that physicians gave more weight 3. The relevant incremental costs and
to patients' personal concerns when they benefits that derive from existing clinical
viewed the patient as an individual, and management alternatives
more weight to general criteria of effective- 4. The expected costs and benefits of the
ness when they were making decisions for a treatment alternative being studied
group of patients (e.g., establishing clinical
policies). The authors further observed that A rigorous prospective .economic analysis
physicians dealing with individual patients enables the body of knowledge on which
were more likely to recommend a therapy subsequent clinical decision making is
with a high probability of success but the based to be more complete, less driven
chance of an adverse outcome. Both these by implicit value judgements, and more
situations pertain to the physician's decision responsive to consideration of limited
to perform a cesarean section. Although the economic resources.
authors stated that the observed differences When prospective economic analyses are
in physician behavior were not related to infeasible, as is the case with medical tech-
economic incentives, it may be that physi- nologies that are already accepted as the
cians are more sensitive to economic incen- standard of care, decision analysis theory
tives when dealing with patients individually can be used to model the expected costs and
than when they are asked to form general benefits of the treatment and its relevant
opinions as to the optimal treatment ap- comparators; thus the economic and clinical
proach in a particular clinical situation. consequences can be explored explicitly and
systematically. Decision analysis is also
U sing Decision Analysis particularly appropriate for analyses of
treatment choices made under conditions of
Theory to Model the Choice of clinical uncertainty.
Delivery Modality Decision analysis theory has been applied
previously in obstetrics to model the appro-
Analysis that explore the influence of priate rate of repeat cesarean section,42,43
economic factors on medical care decisions and the economic effects of alternative
are increasingly common, although clini- methods of inducing labor. 44 However,
cians still may be generally unfamiliar or to date there have been no modeling anal-
uncomfortable with the economic perspec- yses focusing on the use of primary cesarean
tive. However, economic analyses are not section.
independent of clinical considerations; Lowering the cesarean section rate among
ratQer, they require and build on existing patients with a previous cesarean section is
clinical knowledge about the nature of a a high priority because repeat cesarean sec-
particular disease, its management, and its tions account for 35% of all U.S. cesarean
outcome. Understanding the full economic sections,3 and 9.5% of deliveries involve
value of a treatment entails analyzing the women with previous cesarean section. The
clinical effectiveness of the treatment in the use of routine repeat cesarean section evolved
context of both the costs of the treatment as a response to the need to minimize the
and the clinical, economic, and psychosocial risk of uterine rupture. However, changes in
outcomes that result. A valid and mean- operative procedure decreased the incidence
ingful economic analysis of a particular of uterine rupture and brought into question
treatment option thus will reflect a thorough the assumption that repeat cesarean section
understanding of the following: should be performed routinely.
13. Economic Considerations in Cesarean Section Use 179

The professional organization of obstetri- performing a cesarean section rather than a

cians, the American College of Obstetrics vaginal delivery. Decision analysis is a use-
and Gynecology, now recommends a trial of ful tool for modeling the costs and benefits
labor in most patients with a previous cesa- of alternative delivery options for such
rean section. 45 However, uncertainty exists conditions.
as to which patients will successfully de-
liver vaginally after having had a previous Constructing the Decision Model
cesarean section. In 1987, only 9.8% of
women with previous cesareans delivered The first step in the analysis of a decision
vaginally.3 Physicians report that they problem is to define the events in the clinical
remain concerned about the potential risks pathway and their sequence. Events in the
to mother and child posed by the trial of clinical pathway include (1) conditions that
labor and by the subsequent need, ifthe trial are present by chance, such as the occur-
is unsuccessful, to perform an emergency rence of specific indications for intervention;
cesarean section, which is not as safe (or as (2) the set of clinical decisions that must be
convenient) as a planned cesarean section. made, such as the choice between a vaginal
In the two studies addressing repeat or cesarean delivery; and (3) the set of all
cesarean section,42,43 the decision that was possible outcomes and consequences that
analyzed was whether to schedule a routine might result from each intervention. The
operative delivery or a trial of labor in probability of occurrence of chance events,
women with previous cesarean section. Be- the interventions, and their sequence make
cause the clinical issues were so closely up the decision flow diagram, or decision
focused on one single outcome, uterine rup- tree. A decision tree should include all rele-
ture, the decision analytic models in the vant alternative courses of action and out-
studies of repeat cesarean section could be comes for a specific clinical situation.
limited to the incidence and expected out-
comes associated with uterine rupture com-
Determining Probabilities
pared with the expected outcomes associated
with a trial of labor. Modeling the costs and The second step in a decision analysis is to
benefits for primary cesarean section is ascertain or estimate the probabilities of
much more complicated than for repeat occurrence of the chance events. In medical
cesarean section because of the wide range decision analyses, the probabilities are
of clinical entities that comprise the indica- generally estimated from· epidemiologic
tions for primary cesarean section. The studies or clinical trials. In the absence of
model must incorporate the epidemiology of rigorous published data, the decision theor-
these clinical entities, the ambiguities ist may have to rely on expert opinion or
inherent in their clinical presentation, the case studies.
complex assortment of clinical management When a decision analysis is used to
options, and the full range of possible out- determine the preferred intervention for an
comes in each case given the clinical individual patient, each treatment decision
management pathway. point is evaluated as a yes/no choice by the
Currently, three of the clinical entities model. However, decision trees can also be
that are associated with high primary cesa- used to evaluate aggregate costs and bene-
rean section rates-breech presentation, fits using the experience of a group of people.
fetal distress, and dystocia-are conditions For this purpose, each decision point is
for which there is a great deal of clinical modeled as a chance event, and the treat-
uncertainty. This uncertainty relates to ment alternatives are each assigned a pro-
when and whether to use an operative bability equal to the observed frequency
approach and to the extent of the improve- of use of that treatment option in that
ment in clinical results to be expected from population.
180 L.B. Gardner

Assigning Values to Outcomes presentation. This decision is evaluated

from two perspectives: (1) a cost analysis of
The third step is to assign relative values to
the decision to attempt external cephalic
each of the possible outcomes. One can
version in patients who are eligible, and
evaluate single outcomes such as costs or
(2) the sensitivity of the cesarean section
mortality, or dual outcomes such as cost per
rate among all patients with breech presen-
year of life saved or cost per survivor. The
assigned value of an outcome multiplied by tation to the rate at which version is
the probability of that outcome results in a attempted, which allows us to draw conclu-
score, known as the expected value, for each sions about how changes in the management
choice option. Calculating expected values of breech presentation might be expected to
allows the treatment options to be quanti- affect the overall cesarean section rate.
tatively compared to eath other. This pro- Second, we explore the impact of various
cess is known as "folding back" the decision cost-effectiveness scenarios on the yes/no
tree. decision regarding cesarean section for any
individual patient. We also test the sensi-
tivity of this decision to certain assump-
Sensitivity Analysis tions. Either of these applications of decision
theory similarly could be applied to patients
The results of a decision analysis are only as with other presenting diagnoses.
valid as the information on which the analy-
sis is based. Ifthere is uncertainty regarding
the probabilities or the values that are as- The Decision to Attempt Version
signed to outcomes, it is advisable to per- Patients with breech presentation comprise
form a sensitivity analysis using a range of approximately 4% of all patients at term. In
estimates. Sensitivity analysis indicates 1987, 84.4% of all women who had breech
how robust the results of the decision analy- presentation were delivered by cesarean
sis are to changes in a key variable. 46 section. 3 However, studies of breech de-
liveries have reported that the increase in
cesarean section rate for term breech de-
Applying the Decision Model
liveries to nearly universal use did not
Patients with no prior history of cesarean appear to significantly reduce unfavorable
section can be categorized into five groups outcomes. 18 ,47,48 For this reason it is now
based on indication for cesarean section: (1) recommended that external cephalic version
patients with breech presentation, (2) pa- be attempted in patients with term breech
tients with fetal distress, (3) patients with presentation before the onset oflabor, except
dystocia, (4) patients with other indications in women with ruptured membranes, a prev-
for cesarean section, and (5) patients with- ious cesarean section, an engaged present-
out an indication for cesarean section. The ing part, or an estimated fetal weight of
following discussion uses the group of pa- 4000 g or greater. 49
tients with breech presentation to illustrate Using data from the published literature,
two ways that decision analysis theory can we first construct the decision model by out-
be applied to analyze the choice of delivery lining the clinical pathway for breech
modality. presentation and inserting reported proba-
First, using reported rates of cesarean bilities for each chance event. Descriptions
section for each patient subgroup, we evalu- of the clinical pathway indicate that pa-
ate the decision to attempt an external tients with breech presentation mayor may
cephalic version, one of the first clinical de- not be eligible for external cephalic version.
cisions physicians face when discovering, Of those who are eligible, a subset will
before delivery, that a patient has a breech undergo a version attempt and the remaind-
13. Economic Considerations in Cesarean Section Use 181

er will not.* Among those who undergo an TABLE 13.1. Cost and mortality statistics used in
attempt, a percentage will be successful. the decision analyses
There will also be a spontaneous version Maternal Neonatal
rate among women who are not eligible for mortality mortality
assisted version or who are eligible but who Clinical presentation Cost rate (%) rate (%)
do not undergo a version attempt. The fre- Vaginal delivery after $3186 0.04 2
quency of use of each delivery approach successful version
must then be determined for each subgroup Vaginal breech delivery $3519 0.04 2
of patients. Cesarean delivery after $5840 0.2 1
successful version
Data on total delivery costs were obtained Cesarean delivery $5876 0.2 1
from 1991 California hospital discharge abs- without version
tracts (California Office of Statewide Health
Planning and Development). The California
hospital data indicate that among patients
with breech presentation, the average total version is preferred on the basis of the cost
hospital cost is $3186 for a vaginal delivery analysis. Figure 13.2 also allows us to see
after a successful version attempt, $3519 for the cost per live outcome and percentage of
a vaginal delivery without version, $5840 patients in each of the patient subgroups.
for a cesarean delivery after a successful We can evaluate the sensitivity of the
version attempt, and $5876 for a cesarean overall cesarean section rate to the rate of
delivery without version. attempted version by converting the deci-
Data regarding maternal and neonatal sion node at which version is attempted into
survival rates were obtained from published a chance node, inserting various rates of
clinical literature. Based on numerous attempted version, and assessing the ex-
reports, it is reasonable to use 0.2% for the pected rate of cesarean section summed over
maternal mortality rate associated with each patient subgroup. The impact on the
cesarean section and 0.04% for the maternal cesarean section rate of three alternative
mortality rate associated with vaginal scenarios for the rate of attempted version is
delivery.8,43,5o For neonatal mortality, we assessed: a 50% attempt rate, a 75% attempt
initially use rates of 1% for cesarean section rate, and a 95% attempt rate. Table 13.2
deliveries and 2% for vaginal deliveries. presents the results, which indicate that
Subsequently we evaluate a range of neona- simply by increasing the rate of attempted
tal mortality rates as part of the sensitivity version among eligible patients from 50% to
analysis. Mortality statistics and costs are 95%, the cesarean section rate among all
summarized in Table, 13.1. patients with breech presentation decreases
from 62.0% to 42.4%, a relative decrease of
Figure 13.1 depicts the initial decision
32%. The impact on the overall cesarean
tree using expected probabilities and rates
section rate (all presentations) is an ab-
of cesarean section for each patient sub-
solute decrease of more than 0.75%.
group from various sources. Figure 13.2
demonstrates that the decision to attempt
The Impact of Cost-Effectiveness
Analysis on the Decision to Use
* Patients who are eligible for a version attempt Cesarean Section
but do not receive one are either: (1) patients in
whom the breech is not detected before the onset The preceding analysis indicates how
oflabor, or (2) patients who do not have access to changes in the management of patients with
a physician trained in external version or whose breech presentation before delivery can
physician chooses not to attempt version for have a substantial impact on the overall
reasons other than a clinical contraindication. cesarean section rate, regardless of the rela-
 f-'
00
t:-:l

Cesarean deliver'
Version successful . 10
.68

Cesarean deliver
Version unsuccessful .95
.32 Vaainal breech deliver
.05
Cesarean deliver
ontaneous version
.15
Pt. w/breech
Cesarean deliver
Remains breech .95
.85 ina! breech deliver
. 05
Cesarean deliver'
ontaneous version
.15

Cesarean deliver
1.00
.85 breech deliver
o
FIGURE 13.1. Initial decision tree for patients with breech presentation.
t""
to
p
e;
g
'"
 ......
c.o
t.:rj
8i:J
o
f-'.
,.,S
Cesarean delivery (1
Version successful L-----.. 10 IIII INS CS CPLO • $5.875; P • 0651
0. o
.~ . ~
~ .68 ........ va9,"al delivery s.:
(1)
_ ;,.r:,si.o_n.,;a
.:, ;,.lt;,;;e,;,;
m;,.::p..:,te:.,:d:....-4. IIII INS V CPLO ~ $3 .219; P ,, 0.5811
r Ve . 1$4 .2231 ,90 "'
~
Cesarean delivery o·
. IIII ls CS CPLO " $5,911 ; P • 0.
2891 i:J
(fl
Version unsuccess II
u ~
.~ .95 I
S·
,32 ......... vagmal breech delivery III Is V CPLO • $3.555; P • 0.G15 (1
(1)
Eli9ible lor version .05 (fl

.95 Cesarean delivery <l INS CS CPLO E $5.8751 ~

Spontaneous version
. $3 484
0~ . 10 §
. 15
; ::!:::.:..:..::.;;.;;.;..;....:---.. .......... vaginal delivery <l INS V CPLO • $3.219\ (1)
/ ,.,r:n
M-
Pt. w/breech presentation Version not altempted .90 o·
i:J
11$5.4471 Cesarean delivery <l Is CS CPLO" $5.911 1
c:::
(fl
Remains breech 0 $5 794 .95 . (1)

.85
~" vaginal breech delivery <lIs V CPLO _ $3.555\
'"
.05
Cesarean delivery
. ~ "INS CS CPLO - $5,875; P • 0.
0071
Spontaneous version • 90
r....:.....------t-I $5 .610 .
/ . 15 ......... vagmal delivery ~I NS V CPLO. $3.219; p" 0.0011
I Not eligible
.~ . 10
~·~~~---t-I ~
.05 '" Cesarean delivery ~
. ~ ~ IS CS CPLO • $5.911 ; P - 0, 0431
Remains breech . 1 .0 0
......------~-. $5.911
.85 ......... vilglnal breech delivery A =-:-:-::-::-:-';:---:-::-=:::1
.....--:;-----......;.------------<II~.l s V CPLO _ 53.5551
o
FIGURE 13.2. Expected costs and probabilities for decision analysis of attempted version (NB = not breech; B = breech; CS = cesarean section;
V = vaginal delivery; CPLO = cost per live outcome; P = percentage of total patients in selected pathway given use of preferred alternative at the
decision node).

......
00
c.o
184 L.B. Gardner

TABLE 13.2. Sensitivity of the overall cesarean cost data presented in Table 13.1. Because
section rate to the rate of attempted version we are now considering the optimum deci-
Cesarean section Overall sion for anyone patient individually, only
Rate of rate among patients breech-related one of the delivery decision nodes will be
attempted with breech cesarean section relevant. Each patient subgroup should be
version (%) presentation (%) rate (%)
considered independently of the others. The
50 62.0 2.47 preferred delivery method is determined
75 51.0 2.04 by comparing the outcome measure we
95 42.4 1.70 have chosen (the cost per live outcome) for
the two delivery modalities for that patient
subgroup.
tive costs or outcomes of cesarean section As can be seen in Figure 13.3, the ex-
compared with vaginal delivery. Another pected value calculation indicates that the
important application of decision analysis cost per live outcome for cesarean section
theory is to explore which method of de- exceeds the cost per live outcome of vaginal
livery is preferred if we consider both costs delivery in every patient subgroup, and that
and clinical outcomes together. therefore the preferred delivery modality for
Cost alone is not a useful outcome to COn- all patients is vaginal delivery. From this
sider in this decision analysis because, in we conclude that, given the previous as-
current practice, cesarean section delivery sumptions about cost, morbidity, and mor-
always costs more than vaginal delivery and tality rates in our simplified model, the
therefore a decision based on cost alone will difference in neonatal outcome between
always favor vaginal delivery. Clearly, the cesarean and vaginal delivery is not of suf-
key issue is the trade-off between the in- ficient magnitude to counterbalance the
creased costs of cesarean section and the incremental cost and maternal morbidity of
impact on maternal and neonatal mortality cesarean delivery.
and morbidity. For this reason, we need to
evaluate both costs and clinical outcomes.
Using hospital data to evaluate delivery
Threshold Analyses
charges, we can simplify the analysis by It is reasonable to suggest that, with a
focusing on costs and mortality, under the sufficiently great disparity in neonatal
assumption that actual hospital charges in- mortality between cesarean and vaginal
clude the impact of morbidity occurring delivery, the improvement in outcome as-
within the hospital stay. For this analysis, sociated with cesarean delivery would be-
the cost-effectiveness measure that is used is come worth the increase in cost. Threshold
the ratio of average total delivery costs to analysis is a type of sensitivity analysis that
average combined maternal and neonatal allows us to determine the level of a par-
survival rate. ticular probability at which the deci-
It is relevant to consider whether mater- sion maker would consider the alternative
nal and neonatal survival should be valued strategies to be equivalent. Generally, only
equally in calculating the cost-effectiveness one probability is allowed to vary at a time.
ratio. This issue can be addressed by per- All other probabilities and assumptions
forming a separate sensitivity analysis on remain constant.
the relative utility, or value, that is asso- We first perform a threshold analysis to
ciated with maternal survival compared determine the neonatal survival rate at
with neonatal survival. We will assume ini- which a cesarean delivery would become
tially that maternal and neonatal survival worth its cost. Table 13.3 shows the cost per
are given equal utility. live outcome of a vaginal delivery over a
Figure 13.3 depicts the decision tree for range of neonatal mortality rates. This table
breech presentation using the mortality and was generated by substituting vaginal de-
 .....
w
trj
8
g
~.
Cesarean delivery <I [N![CS:CPi.O ~J5ft75J o'"'
o
i:j
UJ
Vers,on successful
. ~~~~~~==~~~
aJvaginal delivery : $3.2t91
s:
I .68 _ u~n,n~, ~,,"u'u (l)
'1
~
Version attempted . 1$3 ,3261 M-

Cesarean delivery
o·
~
Version unsuccessful . A MA l S·
.32 " ~~, ~ ~A"~ n
o
(l)
UJ
Eligible for version
.95
/ "
1$3 ,3351
, Cesarean delivery (J INS CS CPlO a $5.8751
~
§
Spontaneous version
UJ
(l)
.15 g.
PI. w/breech presentation Version not attempted o·
i:j
.05 Cesarean delivery ~
UJ
Remains breech (l)
Vacinal breech delivery $3,555
.85
" ' vagina, OUIllen oellvery ~I B V CPlO _ $3,555 ; P = 0.0401

Cesarean delivery <IINS CS CPLO a $5,8751

Spontaneous version
........
. 15

\
Not eligible . 1$3,5051
.
.05 Cesarean delivery <118 cs CPlO _ $5,9111
Remains breech
.........
.85

FIGURE 13.3. Cost-effectiveness decision tree for choice of delivery modality in breech presentation (NB = not breech; B = breech; CS = cesarean
section; V = vaginal delivery; CPLO = cost per live outcome; P = percentage of total patients in selected pathway given use of preferred
alternative at the decision nodes).

.....
00
Cl1
186 L.B. Gardner

TABLE 13.3. Cost per live outcome of vaginal TABLE 13.4. Threshold analysis of neonatal
delivery over a range of values for neonatal mortality rate and relative utility
mortality rate Ratio of
Neonatal mortality rate Cost per live outcome ($) maternal
to neonatal
0.00 3187 utility 0.001 0.01 0.1 1.0 10.0
0.05 3268
0.10 3354 Neonatal
0.15 3445 mortality Cost per live outcome of cesarean
0.20 3541 rate section ($)
0.25 3642 0.01 5899 5900 5904 5929 5954
0.30 3749
0.35 3863 Cost per live outcome of vaginal
0.40 3983 delivery ($)
0.45 4112
0.00 3,186 3,186 3,186 3,187 3,187
0.50 4249
0.10 3,540 3,536 3,505 3,354 3,216
0.55 4396
0.15 3,748 3,742 3,689 3,445 3,231
0.60 4553
0.20 3,982 3,973 3,894 3,541 3,246
0.65 4721
0.25 4,247 4,234 4,123 3,642 3,261
0.70 4903
0.30 4,549 4,532 4,381 3,749 3,277
0.75 5099
0.35 4,899 4,876 4,673 3,863 3,292
0.80 5312
0.40 5,306 5,275 5,007 3,983 3,307
0.85 5543
0.45 5,788 5,746 5,392 4,112 3,323
0.90 5795
0.50 6,366 6,310 5,841 4,249 3,339
0.95 6071
0.55 7,071 6,995 6,372 4,396 3,355
1.00 6375
0.60 7,953 7,849 7,010 4,553 3,371
0.65 9,086 8,939 7,789 4,721 3,387
0.70 10,595 10,380 8,762 4,903 3,404
0.75 12,706 12,377 10,014 5,099 3,420
livery neonatal mortality rates ranging 0.80 15,867 15,323 11,684 5,312 3,437
0.85 21,120 20,112 14,021 5,543 3,454
from 0 to 100% into the cost-effectiveness 0.90 31,576 29,254 17,527 5,795 3,471
ratio used in the attempted version arm of 0.95 62,534 53,635 23,370 6,071 3,489
the decision tree depicted in Figure 13.3, 1.00 3,190,462 321,915 35,060 6,375 3,506
then recomputing the cost per live outcome.
Portion oftable not in boldface, cost per live outcome for
Given that the cost per live outcome of a cesarean section exceeds that for vaginal delivery; bold-
cesarean section in this case is $5929, the faced portion, cost per live outcome for cesarean section
results indicate that, in this simplified is less than or equal to that for vaginal delivery.
model, cesarean delivery would not be the
preferred method unless the vaginal delivery
neonatal mortality rate was greater than live outcome of cesarean section delivery at
90%. a neonatal mortality rate of 1%. It can be
This conclusion is based on the assump- seen from Table 13.4 that if the utility of
tion that maternal aml neonatal survival maternal survival were one-tenth that of
should be considered to have equal value neonatal survival, cesarean section de-
when calculating the cost-effectiveness livery would be worth its cost at vaginal
ratio. The next threshold analysis explores delivery neonatal mortality rates above
the two-way impact of changes in both the 50%, compared with the threshold of greater
vaginal delivery neonatal mortality rate than 90% that is seen at equal utilities.
and the relative utility of maternal survival In this analysis, the threshold value for
compared with neonatal survival. neonatal mortality does not appear to drop
Table 13.4 presents the cost per live below 45%, even when the relative utility
outcome of vaginal delivery over a range of maternal survival is as little as one-
of values for neonatal mortality rate and thousandth that of neonatal survival. Fur-
relative utility, compared with the cost per thermore, if the relative utility of maternal
13. Economic Considerations in Cesarean Section Use 187

survival is increased to 10 fold that of neon a- TABLE 13.5. Threshold analysis of neonatal
tal survival, then cesarean section would not mortality rate and relative utility with equal
be worth its cost even at a neonatal mortality charges for cesarean section and vaginal delivery
rate of 100%. Ratio of
In truth, actual vaginal delivery neonatal maternal
mortality rates are likely to be far smaller to neonatal
utility 0.001 0.01 0.1 1.0 10.0
than the threshold values calculated by
these sensitivity analyses, and actual utility Neonatal
for maternal survival is likely to at least mortality Cost per live outcome of cesarean
rate section ($)
equal if not surpass that for neonatal sur-
vival. We must conclude that if hospital 0.01 3218 3219 3221 3235 3248
charges and mortality rates were indeed the Cost per live outcome of vaginal
only outcomes to be considered, cesarean delivery ($)
section would never be the preferred de- 0.00 3,186 3,186 3,186 3,187 3,187
livery modality for breech presentation. 0.01 3,218 3,218 3,215 3,203 3,190
0.02 3.251 3,250 3,245 3,219 3,193
0.03 3,284 3,284 3,275 3,235 3,196
0.04 3,319 3,317 3,306 3,252 3,199
Discussion and Conclusions 0.05 3,354 3,352 3,338 3,268 3,202
0.10 3,540 3,536 3,505 3,354 3,216
The importance of economic analyses of 0.15 3,748 3,742 3,689 3,445 3,231
medical interventions is that they enable 0.20 3,982 3,973 3,894 3,541 3,246
0.25 4,247 4,234 4,123 3,642 3,261
those paying for an intervention to under-
0.30 4,549 4,532 4,381 3,749 3,277
stand the conditions under which the inter- 0.35 4,899 4,876 4,673 3,863 3,292
vention is worth the cost. Including economic 0.40 5,306 5,275 5,007 3,983 3,307
considerations explicitly in studies of choice 0.50 6,366 6,310 5,841 4,249 3,339
among alternative therapeutic options 0.60 7,953 7,849 7,010 4,553 3,371
0.70 10,595 10,380 8,762 4,903 3,404
enables the economic issues to be anal-
0.80 15,867 15,323 11,684 5,312 3,437
yzed in conjunction with the clinical 0.90 31,576 29,254 17,527 5,795 3,471
considerations. 1.00 3,190,462 321,915 35,060 6,375 3,506
Our decision analysis of cesarean section
versus vaginal delivery for women with Portion of table not in boldface, cost per live outcome
for cesarean section exceeds that for vaginal delivery;
breech presentation indicates that under the boldfaced portion, cost per live outcome for cesarean
given assumptions, vaginal delivery is section is less than or equal to that for vaginal delivery.
always the preferred method of delivery.
However, this conclusion is subject to
change as a function of variations in any of ing the same charge for both cesarean and
the underlying probabilities or assumptions vaginal delivery (set for this illustration at
that comprise the model. Particularly im- $3186), then cesarean section delivery
portant to consider is the way the results of would be worth its cost at much more realis-
the decision analysis would change if out- tic values for neonatal mortality rate (e.g.,
comes other than maternal and neonatal more than 2% at equal utility for maternal
mortality, and costs other than delivery and neonatal survival).
charges, were included. Furthermore, including the probability
In many areas of medical care, prices are and costs of maternal and neonatal mor-
not an accurate guide to the true cost of a bidity will dramatically influence the deci-
procedure. Finkler and Wirtschafter,17 in a sion in favor of cesarean section, because
study of one HMO, reported that cesarean neonatal morbidity related to vaginal
section rates were unrelated to direct costs. breech delivery (e.g., asphyxia and trauma),
Table 13.5 shows how, if the sensitivity are undesirable outcomes that also can be
analysis in Table 13.4 were repeated us- extremely expensive. Including the proba-
188 L.B. Gardner

bilities and costs of malpractice liability partum management. Philadelphia: Lea &
following each of the two methods of de- Febiger, 1989:207-215.
livery similarly is likely to exert an effect 7. Evrard JR, Gold EM. Cesarean section and
that would favor the decision to choose cesa- maternal mortality in Rhode Island. Obstet
rean section. Gynecol 1977;50:594.
Decision analysis is generally a normative 8. Minkoff HL, Schwarz RH. The rising cesa-
rean section rate: can it safely be reversed? J
tool when used in ongoing decision mak- Am CoIl Obstet GynecoI1980;56(2):135-143.
ing regarding the Care of a particular pa- 9. Rubin GL, Peterson HB, Rochat RW, et al.
tient. However, the clinical decision that Maternal death after cesarean section in
feels "right" may not always mirror the Georgia. Am J Obstet Gynecol 1981;139:681.
results of the decision analysis. In such cir- 10. Petitti DB, Cefalo RC, Shapiro S, et al. In-
cumstances, comparing the results of the hospital maternal mortality in the United
decision analysis with the clinical decision States: time trends and relation to method of
that is actually made can make it clear that delivery. Obstet GynecoI1982;59:6-12.
there are additional values and judgments 11. Rochat RW, Koonin LM, Atrash HK, Jewett
that are implicitly influencing the clinical JF. Maternal mortality in the United States:
decision. Whether these values and judg- report from the Maternal Mortality Col-
laborative. Obstet GynecoI1988;72(1):91-97.
ments ought to have a. place in the deci- 12. Guldholt I, Espersen T. Maternal febrile
sion making process is the purview, not of morbidity after cesarean section. Acta Obstet
decision theorists, but of clinical leaders, Gynecol Scand 1987;66:675-679.
policy makers, and society. 13. Tulman L, Fawcett J. Return of functional
ability after childbirth. Nurs Res 1988;
References 37:77-81.
14. NIH. Cesarean childbirth: report of the
1. Dunn LJ. Cesarean section and other obste- NICHD task force on cesarean childbirth.
tric operations. In: Scott JR, DiSaia PJ, Bethesda, MD: National Institntes of Health,
Hammond CB, Spellacy WN, eds. Danforth's 1981. (DHHS publication no. (NIH) 82-2067.)
Obstetrics and Gynecology, 6th Ed. Phil- 15. Bland R. Pathogenesis of pulmonary edema
adelphia: Lippincott, 1990:639-658. after premature birth. Adv Pediatr 1987;
2. Elixhauser A, Andrews RM, Fox S. Clinical 34:175-221.
classifications for health policy research: 16. Martin DP, Fadden MK, Holt VL, et al. The
discharge statistics by principal diagnosis relationship among delivery method, quality
and procedure. AHCPR Publication No. 93- and outcomes of obstetrical care. Presented
0043. Division of Provider Studies Research at the annual meeting of the Association for
Note 17, Agency for Health Care Policy Health Services Research, 1992, Chicago,
and Research. Rockville, Maryland: Public lllinois.
Health Service, 1993. 17. Finkler MD, Wirtschafter DD. Why ·pay
3. Taffel SM, Placek PJ. An overview of recent extra for cesarean deliveries? Inquiry 1993;
patterns in cesarean delivery and where we 30(2):208-215.
stand today. Presented at the 116th Annual 18. Hack M, Fanaroff AA. Outcomes of extre-
Meeting of the American Public Health mely-Iow-birth-weight infants between 1982
Association, November 13-17, 1988, Boston, and 1988. N Engl J Med 1989;321(24):1642-
Massachusetts. 1647.
4. Porreco RP. High cesarean section rate: a 19. Anderson GM, Lomas J. Explaining varia-
new perspective. Obstet Gynecol 1985;65(3): tions in cesarean section rates: patients,
307-311. facilities or policies? Can Med Assoc J 1985;
5. de Regt RH, Minkoff HL, Feldman J, 132:253-259.
Schwarz RH. Relation of private or clinical 20. Carpenter MW, Soule D, Yates WT, Meeker
care to cesarean birth rate. N Engl J Med CI. Practice environment is associated with
1986;315:619-624. obstetric decision making regarding abnor-
6. Petitti DB. The ideal cesarean section rate. mal labor. Obstet Gynecol 1987;70:657-
In: Parer JT, ed. Antepartum and intra- 662.
13. Economic Considerations in Cesarean Section Use 189

21. Fraser W, Usher RH, McLean FH, et al. 34. Stafford RS, Sullivan SD, Gardner LB.
Temporal variation in rates of cesarean Trends in cesarean section use in California,
section for dystocia: does "convenience" 1983-90. Am J Obstet Gynecol 1993;168(4):
playa role? Am J Obstet Gynecol 1987; 1297-1302.
156:300-304. 35. Hillman AL. Health maintenance organi-
22. Gardner LB. Hospital-physician relation- zations, financial incentives, and physicians'
ships and cesarean section rate variations. judgments. (Editorial.) Ann Intern Med
Doctoral dissertation, U ni versi ty ofCalifornia 1990;112(12):891-893.
at Berkeley. 36. Feldman GB, Freiman JA. Prophylactic
23. Evans R. Supplier-induced demand: some cesarean section at term? N Engl J Med
empirical evidence and implications. In: 1985;312(19):1264-1267.
Perlman M, ed. The economics of health and 37. Cepicky P, Stembra Z, Zeman J, Lomickova
medical care. New York: Wiley, 1974. T, Mandys F. When is it possible to meet the
24. Anderson R, House D, Ormiston M. A theory wish of a woman to terminate her labour by
of physician behavior with supplier-induced cesarean section? Eur J Obstet Gynecol
demand. South Econ J 1981;July:124- Reprod BioI 1990;38:109-112.
133. 38. Rock SM. Malpractice premiums and pri-
25. Wilensky GR, Rossiter LF. The relative mary cesarean section rates in New York
importance of physician-induced demand in and Illinois. Public Health Rep 1988;103:
the demand for medical care. Milbank Mem 459-463.
Fund Q 1983;61(2):252-277. 39. McCullough LB. The physician's virtues
26. Luft HS. Economic incentives and cons- and legitimate self-interest in the patient-
traints in clinical practice. In: Aiken LH, physician contract. Mt Sinai J Med 1993;
Mechanic D, eds. Applications of social 60(1):11-14.
science to clinical medicine and social policy. 40. Feldstein PJ. Health care economics, 3rd Ed.
New Brunswick: Rutgers University Press, Albany: Delmar, 1988.
1986:500-518. 41. Redelmeier DA, Tversky A. Discrepancy
27. Tussing AD, Wojtowycz MA. The cesarean between medical decisions- for individual
decision in New York State, 1986. Med Care patients and for groups. N Engl J Med 1990;
1992;30(6):529-540. 322(16):1162-1164.
28. Robinson JC, Gardner LB, Luft HS. Health 42. Shy KK, LoGerfo JP, Karp LE. Evaluation
plan switching in anticipation of increased of elective repeat cesarean section as a stan-
medical care utilization. Med Care 1992; dard of care: an application of decision
31(1):43-51. analysis. Am J Obstet Gynecol 1981;139(2):
29. Murray D. Do you know how much to 123-129.
charge? Medical Economics 1993 October 43. Silver RK, Minogue J. When does a statis-
11:106-133. tical fact become an ethical imperative?
30. Pauly MV, Hillman AL, Kerstein J. Am J Obstet Gynecol 1987;157(2):229-
Managing physician incentives in managed 233.
care. Med Care 1990;28:1013-1024. 44. Davies LM, Drummond MF. Management
31. Wilner S, Schoenbaum SC, Monson RR, of labour: Consumer choice and cost im-
WinickoffRN. A comparison ofthe quality of plications. Journal of Obstetrics and
maternity care between a health mainten- Gynecology 1991;11[Suppl. 1]:S28-S33.
ance organization and fee for service prac- 45. American College of Obstetrics and
tices. N Engl J Med 1981;304:784-787. Gynecology. Guidelines for Vaginal De-
32. Wright CH, Gardin TH, Wright CL. Obste- livery after a Previous Cesarean Birth.
tric care in a health maintenance organiza- Washington, DC: Committee on Obstetrics:
tion and a private fee-for-service practice: a Maternal and Fetal Medicine, 1988.
comparative analysis. Am J Obstet Gynecol 46. Luce BR, Elixhauser A. Standards for the
1984;149:848-856. socioeconomic evaluation of health care ser-
33. McCloskey L, Petitti DB, Hobel CJ. Varia- vices. Berlin: Springer-Verlag, 1990.
tions in the use of cesarean delivery for 47. Green JE, McLean F, Smith LP, Usher R.
dystocia. Lessons about the source of care. Has an increased cesarean section rate for
Med Care 1992;30:126-135. term breech delivery reduced the incidence of
190 L.B. Gardner

birth asphyxia, trauma, and death? Am J JR, DiSaia PJ, Hammond CB, Spellacy WN,
Obstet Gynecol 1982;142:643-649. eds. Danforth's Obstetrics and Gynecology,
48. Myers SA, Gleicher N. Breech delivery: why 6th Ed. Philadelphia: Lippincott, 1990:
the dilemma? Am J Obstet Gynecol 1987; 567-584.
156(1):6-10. 50. Danforth DN. Cesarean section. JAMA
49. Cruikshank DP. Malpresentations and 1985;253(6):811-818.
umbilical cord complications. In: Scott
14
Ethical Issues in the Utilization
of Cesarean Section
THOMAS E. ELKINS AND DOUGLAS BROWN

This chapter explores the ethical issues sur- course of a disease or the consequences of a
rounding the utilization of cesarean section serious accident. Minorities did not yet fully
for delivery. A few definitions are in order. benefit from the democratic promotion of
Ethics has to do with the decisions individual rights. Physicians were meas-
about what ought to be done, all things ured by the standard of compassion and
considered. 1 - 5 This definition distinguishes were trusted to "do no harm" when they
what could be or is done from what decision paternalistically made beneficent decisions
makers decide ought to be done. This defini- "in the patient's best interests."
tion calls for thorough consideration of non- Technological and cultural revolutions
clinical as well as clinical factors. Ethical eventually combined to restructure radi-
discourse concentrates on who decides and cally the discussion of medical ethics. Pa-
the means by which decisions are made. tient expectations shifted from care to cure,
Participants attempt to interpret immediate from acquiescence to self-determination.
experiences against the backdrop of societal Ethical principles (with autonomy and
wisdom about human behavior. justice added to nonmaleficence and bene-
Medical ethics focuses on decision making ficence) formed a core vocabulary for physi-
in situations involving health care. From cians, philosophers, and theologians. 8 The
a societal perspective, ethical reflection application ofthese principles in clinical set-
on health care issues concentrates on indi- tings was seen as the "end" by some (i.e.,
vidual human and civil rights and on a fair deontologists), and as the "means" by others
distribution of resources. From a profes- (i.e., utilitarians).9,lo (Their ethical conclu-
sional perspective, ethical reflection orients sions were not invariably at odds.)
clinical decisions toward the goal of enhanc- Each of these pillar principles initially
jng a patient's well-being or easing a pa- held prima facie importance. However,
tient's decline. From a patient perspective, given the affirmation of cultural diversity
ethical reflection measures clinical choices and individual rights, the principle of pa-
by the impact on one's life story, which has tient autonomy rose to "trump" status in
been disrupted by circumstances such as societal, and eventually in medical-ethical,
accidents or illnesses. discourse. 11 ,12 Case-by-case decisions were
Until the 1960s, the Hippocratic tradition thought to originate from within the privacy
provided historical context for ethical reflec- of individual or familial traditions. Atten-
tion about medicine in the United States. 6 •7 tion on patient values increased. 13 ,14
Technological advances had not yet given Most recently, perceived deficiencies
patients sufficient cause to presume that have been identified amid the strengths of
physicians could significantly alter the autonomy-based ethics, giving impetus

191
192 T.E. Elkins and D. Brown

to revised affirmations of physicians' au- confidentiality, to maternal and fetal health,

tonomy, accountability, and character rela- to mutual trust, and to the naturalness
tive to the decision-making process. 15- 17 of procreation flow without ambiguity
The patient-physician relationship is now or conflict.
being framed more as an overlapping of pa- It should be noted that obstetricians and
tient and physician "stories."18 Also, the other professionals comprising perinatal
range of participants in medical-ethical teams are hardly neutral participants in
deliberations, frequently organized as ethics cases in which cesarean section delivery is
committees, has expanded to include law- under consideration. Clinicians hold no uni-
yers, economists, minority advocates, politi- form value system. For example, one may
cians, and "lay" persons. The American link the value of an endangered fetus to
College of Obstetricians and Gynecologists racial, economic, or cognitive development
(ACOG) was among the first medical as- status. Another may view every fetus as
sociations to establish a committee on inherently valuable.
ethics with the responsibility to issue guid- Clinicians are likely to see situations
ance papers. from the perspective(s) established by their
Reproductive ethics narrows the focus to specialties. Clinicians have unique patch-
decision making associated with procreative works of experience that shape their reac-
physiology. The clinical situations cover a tion to each situation. For example, one
wide range, for example, gynecologic care, professional in the midst of a malpractice
infertility, birth control, pregnancy ter- suit may intend to avoid the risk of another
mination, prenatal care, fetal tissue re- suit. That professional may have a col-
search, labor and delivery serivice, high-risk league who recently became the mother of a
pregnancy management, and fetal therapy. healthy baby after fetal therapy. That col-
Promotion of a woman's procreative well- league may at least temporarily fail to grasp
being and control provides the primary the grounds for indecision about pursuing
ethical common ground for these varied intervention.
interests. Given advances in fetal diagnosis Clinicians have opinions about numerous
and therapy as well as in neonatal intensive related issues (e.g., elective abortion, selec-
care, it should be noted that physicians in- tive nontreatment of critically ill newborns,
volved in decisions to utilize cesarean sec- the right of a fetus to the status of "person,"
tion are trained as surgeons19 and have a patient's right to refuse treatment, and
distinguishable specialties within the field wrongful life/wrongful death suits). These
of maternal-fetal health care. opinions clarify their responses to clinical
Obstetric ethics (as distinguished from conflicts over recommended but refused
gynecologic ethics) concentrates on preg- interventions.
nancy management decisions. 2o- 23 The pre-
sence of two mutually dependent beings-
the pregnant woman and the fetus-is the Anatomy of a Recommendation
most obvious justification for isolating the
issues of this area of health care. The goal of Many issues that surround the use of cesa-
obstetric ethics is the delivery of a healthy rean section have broad ethical implica-
term baby without any compromise to ma- tions. In developing countries where the
ternal well-being. number of trained physicians is extremely
Ethical discussions and the associated low, the question is debated as to who should
literature often leave the impression that be trained to perform cesarean sections-
obstetric ethics has to do with dilemmas and nurses, midwives, traditional birth atten-
tragic choices at the commencement of life. dants. Cesarean section is the pivotal pro-
Instead, in the majority of pregnancies, cedure that influences funding decisions for
commitments to respect for new life, to vast projects aimed at reducing maternal
14. Ethical Issues in the Utilization of Cesarean Section 193

mortality rates (which reach 10 deaths per medical factors and is strikingly influenced
1000 births). by physician "style" in handling nonclinical
The soaring cesarean section rate in the considerations. 24- 27 We doubt that even
Unites States has raised a separate set of physicians for whom a pregnant woman's
ethical concerns. Issues of increased physi- self-determination is the trump ethical con-
cian fees for operative deliveries, the use of sideration would often acquiesce to such
cesarean sections to avoid long hours on requests. 28 It should, nonetheless, be noted
weekends of busy physicians, and a concern that refusing such patient requests does
for medical decisions being made as a result imply that patient autonomy is a condi-
of legal fears have all been explored in tional, rather than absolute, right.
recent years. Responses to requests for cesarean sec-
If put in international perspective, an tions when compelling medical need is
ethics chapter on this relatively simple lacking are value-laden responses. 29 Still,
operative procedure could obviously become an argument based on the absence of com-
quite broad. Our discussion, accordingly, pelling medical need max satisfy physi-
puts primary attention on decisions about cians wary of expanding ethical reflection
the utilization of cesarean section in Ameri- to non-clinical considerations. However, a
can settings. However, the significance of more thorough exploration of nonclinical
international considerations should not be considerations is necessary in situations
forgotten. complicated by possible or evident risk to
Questions about cesarean section delivery the pregnant woman or to the fetus.
are at times raised in situations that lack
Case Three
compelling medical need.
A 23-year-old primigravida woman pre-
Case One sents at 29 weeks gestation in active
A 15-year-old primigravida woman, labor with twins in A breech, vertex
progressing in labor and having re- presentation. A cesarean section is
ceived an epidural anesthetic as well as recommended.
narcotics, screams for a cesarean section.. Case Four
to end her labor pain. A 37-year-old primigravida woman pre-
Case Two sents in labor at 41 weeks gestation
A 39-year-old woman requests a cesa- with a footling breech presentation.
rean section at 37 weeks so as to be able Delivery by cesarean section IS
to make a trip 1 month later. recommended.
Case Five
Every resident physician in obstetrics has
A 24-year-old woman undergoes ultra-
been confronted by the distraught patient in
sonography at 28 weeks gestation and
labor who is demanding as loudly as possible
fetal hydrocephaly is identified. The pa-
for someone to "Take my baby!" The anguish
tient asks about routes of delivery and
of labor does preoccupy decision-making
her obstetrician recommends cesarean
considerations on occasion. A cesarean sec-
delivery.
tion is a welcome relief for many patients.
Case Six
Much less frequently, a woman may request
A 19-year-old woman presents at 27
delivery by cesarean section to facilitate the
weeks gestation in active labor. Fetal
timing of events relative to career plans or
monitor tracings reveal severe brady-
even vacation plans. Such extreme appeals
cardia with late decelerations. The
to autonomy are, fortunately, uncommon.
obstetrician recommends delivery by
Cesarean sections in these situations
cesarean section for fetal distress.
would be disturbing evidence that the cesa-
rean section rate in the United States is not In some developing countries where cesa-
simply a result of decision making based on rean section is rarely done for fetal well-
194 T.E. Elkins and D. Brown

being, all these cases might "properly" be information about the values ofthe hospital
managed by vaginal delivery. In situations and the obstetrician sufficient in detail to
in which antenatal assessment, fetal moni- establish confidence. The obstetrician in-
toring in labor, neonatal intensive care, and volved needs information about the mother
even routine pediatric follow-up (and mal- that is adequate to plan an initial strategy
practice attorneys) do not exist, vaginal that integrates his or her values and me-
deliveries are preferred. But is this ethically dical skills with the mother's values and
acceptable? aspirations. 3o
Ethical arguments take on new foci in We propose that the relationship formed
such countries. For example, in rural Africa, between the obstetrician and the pregnant
should a cesarean section be done to deliver woman necessitates compatibility on at
a fetus barely maintaining a heartbeat after least four matters:
3 days of obstructed labor and sepsis? Or
should a destructive procedure with vaginal 1. Aim. Is there agreement about the
delivery be done for the mother? Geography, place of fetal outcome in the envisioned
technology, and culture may alter tradi- pregnancy?
tional decisions normally made so easily in a 2. Perception of childbearing. 31 Is child-
technologically advanced country. However, bearing seen as a natural process? Has
to ignore the viable fetus as a patient today, the pregnant woman assumed responsi-
in America, is to be truly noncontemporary. bilities related to fetal outcome? What
The dilemma in the foregoing cases, in role does the obstetrician, trained to
which only residue of a noneventful preg- intervene to correct deviations in the pro-
nancy remain, has to do with the task of cess of childbearing, have?
making choices that benefit one patient 3. Decision-making process. Has the preg-
at some expense to the other. Multiple nant woman clarified her interest to be
experiences-for example, with pregnancy, involved in decision making? Has she
medical education, litigation, choice of spe- introduced other participants (e.g., father
cialty, and indigent patients-significantly of the fetus, parents) to the physician? Is
influence the sort of obstetrician a profes- the physician's role clear?
sional becomes. 4. Perception of the fetus. 31 Is there compa-
A pregnant woman in America today who tibility in the moral status assigned to
invites an obstetrician into a chapter of her the fetus?
"story," has, in turn, made the decision to
become or remain pregnant, has adopted a So long as the prenatal period is unevent-
perception of childbearing, has formed an ful, the obstetrician should remain inconspi-
opinion about the place of prenatal care, and cuous but alert as the pregnant woman
has some opinion about the relation of her focuses on the mystery of bringing a new life
interests to the well-being of her fetus. to birth. A threshold is crossed when the
Because our society remains grounded in obstetrician perceives a risk or risks to the
the presumption that its members can be pregnant woman or the fetus. The obstetri-
trusted to exercise their freedom of choice in cian, reacting with a heightened sense of
such a way that others will not be harmed responsibility, becomes more conspicuous in
and the ideals of a just society will be appro- the case. A "pause" to consider the ethical
ximated, the fundamental role of medicine is implications in this shift is warranted,
to avoid harming patients in the effort to because drawing the pregnant woman's
maximize the health they seek so as to con- attention to risks in the pregnancy changes
tinue their pursuit of a complete life. For a her childbearing experience. The obste-
fiduciary relationship to unite mother and trician's role becomes more aggressive.
obstetrician in the task of responding to any The privacy of the physician-patient
circumstances that may disrupt the flow of relationship 32 is diminished when consul-
the mother's childbearing, the mother needs tation is required.
14. Ethical Issues in the Utilization of Cesarean Section 195

The obstetrician faces another threshold if tinguish recommendation, persuasion, and

the pregnant woman resists or rejects at- coercion are blurred. We agree with the
tempts to reframe her situation as having observation by Brody40 that power pervades
significant risk. She may dispute the inter- the physician - patient relationship and is
pretation. She may refuse recommended disproportionately weighted toward the
interventions designed to stand between physician. Brody's assessment means that
("obstetrics" is a Latin combination, mean- every communication from a physician, in
ing "to stand between") her and the proposed both what is said and how it is said, puts
threat to her well-being or that of her fetus. leverage on a patient.
Another "pause" is critical here, in view of What can be documented are cases that
the ethical decisions created by such refusal. have passed from recommendation and per-
The obstetrician, whose clinical instinct to suasion to court-ordered intervention.
intervene often jumps ahead of reflection, Kolder et al. 41 conducted a review of cases,
may find this pause most difficult. The pre- through 1986, in which court orders had
vailing assumption is that recom- been sought to sanction interventions over
mended medical or surgical treatment is maternal refusal. They found 21 cases in-
beneficial. 33 Rhoden34 has argued, on the volving cesarean sections, hospital deten-
basis of extensive clinical exposure, that an tions, or intrauterine transfusions. Fifteen
obstetrician's clinical intuition to intervene court orders for cesarean sections had been
is so ingrained that a pregnant woman's sought in 11 states. Thirteen orders had
refusal is often virtually unthinkable. The been obtained (2 of which were not used
ethical questions facing the obstetrician, when maternal agreement subsequently
which must be resolved with "exteme occurred). The women were all Afro-
caution,,,35 have been variously worded: American or Asian; English was not the
When are the risks of fetal intervention to primary language of 1 in 4 of these women.
the woman sufficiently great to allow her Seven of the 15 court orders were sought
to maintain autonomy and when are the earlier than 37 weeks of gestation. The most
benefits to the fetus sufficiently great to common diagnoses were fetal distress, pre-
justify overriding her autonomy?36 vious cesarean section, and placenta previa.
Where risk of serious injury or death to the All cases had occurred within teaching hos-
mother is minimal and the possibility of pitals. No case had resulted in maternal
benefit to the fetus is considerable, whim morbidity or motality. In 1 case, a Nigerian
should the physician intervene on behalf father, strongly opposed to the ordered
of the unborn child?37 intervention, committed suicide a few
When, if ever, is it ethical to provide medical months after his wife's cesarean section.
treatment against a patient's will?33 Two infants experienced significant mor-
Which patient-the pregnant woman or the bidity. No fetal deaths occurred.
fetus-is the primary patient?36 Kolder et al. also sought opinions from the
Which patient choices ought to be re- heads of maternal-fetal medicine programs.
spected?38 Nearly half the respondents could conceive
Does force ever have a place in medicine?39 of situations in which women whose refusal
endangered thel life of the fetus should be
detained. A similar number thought sanc-
tioned interventions should include proce-
What Has Been Done dures other than cesarean sections. A
The Kolder et al. Review quarter of the respondents advocated state
surveillance.
Exact determination of the number of cesa- These investigators regarded the central
rean sections that have been done without question to be whether a patient's decision-
the agreement of the pregnant woman is not making rights should ever be usurped by
possible, in part because the lines that dis- a physician or by the government to ad-
196 T.E. Elkins and D. Brown

vance the therapeutic interests of a second against pregnant women who refuse their
patient. 41 physicians' recommendations, the study
Acceptance of forced cesarean sections, hospital posed nine situations not involving cesarean
detentions, and intrauterine transfusions may section. The respondents' willingness to
trigger demands for court-ordered prenatal force intervention in circumstances not
screening, fetal surgery, and restrictions on the involving delivery increased as fetal risk at
diet, work, athletic activity, and sexual activity term increased. The circumstances for
of pregnant women. (p. 1195) which the respondents were most willing to
An accompanying editorial argued that, force intervention-insulin treatment and
without the pregnant woman's informed excessive use of alcohol-would requie
consent, fetal access is only possible "by monitoring the pregnant woman's life-style,
treating her as a fetal container, a nonper- beginning in the early weeks of pregnancy.
son without rights to bodily integrity.,,42 To determine the respondents' support of
The article and editorial drew several criti- court-ordered cesarean sections, the study
cal letters to the editor. posed 10 situations. The respondents indi-
Elkins et al. 43 studied one subject group cated little compulsion to seek a court order
related to the Kolder et al. investigation. if fetal complications had not yet occurred.
This study reported the perceptions of 31 Their readiness increased as fetal risk grew
resident and faculty physicians in obstetrics to the point at which fetal health was at
and gynecology in two residency programs grave risk without cesarean delivery (as in
in Michigan, a state to which the Kolder et cases of well-documented fetal distess near
al. article referred several times. The res- term) or the future health of the newborn
pondents showed no consensus of opinion was in jeopardy (as in the refusal to deliver
on the management of the following case by cesarean section a breech presentation
presentation: with hyperextension of the head), or when
A competent 27-year-old white female presented
the health of both mother and fetus were at
to Planned Parenthood at 34 weeks gestation, grave risk (as in refusal of cesarean section
requesting an abortion. She was advised against delivery in spite of documented placenta
pregnancy termination. Because her blood pres- previa at term). The majority rooted their
sure was 180/110, she was transferred to a ter- clinical responses in medical reasonableness
tiary care hospital, where a physical examination (62%), ethical principles (62%), and mal-
revealed that she had severe preeclampsia. practice concerns (54%).
Monitoring, which the patient did not refuse, These data confirm the identification by
revealed serious fetal distress. Physicians in- Kolder et al. 41 of a willingness on the part of
formed the patient of the fetal distress and physicians practicing in a teaching hospital
advised her that she should be delivered imme-
setting to employ court-ordered interven-
diately by cesarean section. She refused the
tions. However, these data did not support
operation, stating that she did not want the baby.
Various labor and delivery personnel, as well as the impression that physicians are deter-
family members, tried to persuade her to have mined to maximize control of pregnant
the surgical delivery. Following the patient's women or the conclusion that court-ordered
initial refusal, the hospital attorney, at the re- interventions must be categorically rejected
quest of the delivery team, spoke with three local so as to avoid abuse.
judges, each of whom refused to hear the case.
Eventually, a stillborn female fetus weighing In Re: Angela Carder44 ,45
2140 g was delivered spontaneously. Although
her hypertension and proteinuria had not re- Reaction to the decision by a District of
solved, on her third postpartum day the woman Columbia judge on June 16, 1987, to sanc-
signed out of the hospital against medical advice. tion a cesarean delivery has catapulted a
In an effort to assess the willingness of the tragic story into ."a paradigm case that
respondents to appeal for judicial leverage has a secure niche in the annals of law and
14. Ethical Issues in the Utilization of Cesarean Section 197

ethics.,,46 An explanation and defense of the tion terminal, the couple's desire to leave a child
decision was filed November 10, 1987. Two behind as a symbol of their relationship was
weeks later, 39 organizations, including the underscored when they reaffirmed their June 12
American Medical Association (AMA) and decision to administer palliative treatment with
the American College of Obstetricians and the goal of extending her pregnancy to 28 weeks
(thus significantly improving the baby's chances)
Gynecologists (ACOG), submitted requests
and then delivering the baby by cesarean section.
for the Court of Appeals to reverse the rul-
Discussion of potential infant morbidity made
ing. The Court of Appeals decided on April Angela's mother, herself confined to a wheel-
26, 1990, to reverse the judge's decision, chair, extremely anxious. She feared that, with
with one judge dissenting in part. Angela's death, she would be responsible for
Attempts to reconstruct any case, includ- raising a child that might be severely handi-
ing cases devoid of controversy, fall short of capped after birth. The father of the fetus, emo-
the reality. The privacy, intensity, spon- tionally overwhelmed, added little to subsequent
taneity, and emotion unique to every case discussion. Angela's mother became increasingly
cannot be recovered. Looking back intro- the dominant figure. The attending physician, a
duces self-examination and reflection into maternal-fetal medicine fellow, supported the
plan, but apparently failed to explore with
the recollection. This limitation was cer-
Angela her preferences should her condition dete
tainly true with the reports of the Angela iorate more rapidly. The oncology resident on call
Carder case. However, initial descriptions of on June 15 asked, during a chart review, what
the Angela Carder case were further com- was to be done should Angela die during the
promised by the selection of information night. The attending physician, convinced that
and the interspersed biases of authors who Angela would not have chosen to deliver a poten-
appearred to be drawn to the case for its tially severely impaired child, communicated
political capita1. 47- 49 that no cesarean delivery would be attempted.
In spite of these limitations, the un- Thus, the initial question had to do with a post-
disputed prominence of this case makes an mortem procedure.
attempt to tell the story necessary. One By early morning, Angela's condition had
deteriorated more rapidly than expected. The
of the authors of this chapter (T .E.) was
family had been called. They had last rites ad-
asked by defense attorneys, apparently
ministered by 8:00 A.M. The attending physician
looking for a witness to support the court met with the family. The family opposed a post-
order, to review the testimony and related mortem delivery. Angela's husband expressed
literature. 46 ,50,51 The review led to this doubt that he could face the child without
reconstruction of the Angela Carder case. Angela. Her mother argued that Angela had
gotten pregnant with the confidence that she
Angela Carder was a 27-year-old high-risk pa- would be able to raise the baby. The attending
tient of the obstetric service of George Washington physician was convinced that the family had
University Hospital. Diagnosed at age 13 with Angela's wishes sufficiently in mind to make
cancer, she had a long history of bone cancer and their guidance reasonable. This management
had undergone the amputation of one leg. After plan was communicated on June 16 to the de-
years of courtship, she had married a man who partment head, who happened to be in a meeting
accepted her even with her major disabilities. with a hospital administrator at the time. The
This couple shared a deep desire to have a child administrator, made aware of the case, asked
together. When she became pregnant, she had whether the baby was viable. He was told via-
not undergone chemotherapy for more than a bility was probable. Without further inquiry into
year. She was approximately 25 weeks pregnant the details of the case, he called the case to the
when a large tumor in one of her lungs, detected attention of the hospital's senior legal counsel,
during her June 9 prenatal visit, confirmed that who advised that the case should be treated as an
the cancer was no longer in remission and had in emergency and that physician opinion should be
fact spread markedly. She was admitted to the followed.
hospital on June 11. She was informed on June 15 Pediatric opinion suggested a high probability
that the tumor was inoperable. With her condi- of neonatal survival if the fetus was delivered.
198 T.E. Elkins and D. Brown

Hospital administrators or attorneys called a competence. A local appeals court chose not to
Superior Court judge. Legal representation for issue a stay. The cesarean section was performed
Angela, the baby, the hospital, and the District of at 6:30 P.M. The baby, born at approximately 26
Columbia was arranged. weeks gestation, survived only a couple of hours.
Angela was no longer alert or capable of Angela's husband did not want her informed
communication by the time the judge and the about the death. However, the attending physi-
various lawyers met near Angela's room with the cian informed Angela. Angela died 2 days later.
medical personnel and the family. The initial The surgery was listed on the death certificate as
postmortem question by now had evolved into a a contributing cause. Lindsay Marie was buried
premortem question; that is, given that Angela with her mother.
would soon die, should a cesarean section be
The Court of Appeals reheard the case in
immediately performed. Angela's lawyer argued
against such a surgery, because consent could not September 1988 and issued a reversal in
be obtained and discussion of a cesarean section April 1990. 45 The Court of Appeals:
earlier than 28 weeks gestation had not been 1. Rejected the judge's attempt to balance
previously conducted with Angela. The baby's
interests
lawyer argued for surgery as the baby's only
2. Argued that the right of bodily integrity
hope, maternal interests being greatly dimin-
ished by Angela's imminent death. Angela's
protected competent and incompetent
mother strenuously opposed the surgical de- individuals from having to submit
livery, expressing her opposition in terms of to any invasive procedure for the benefit
her interests and the potential burden to her. of another
Angela's husband contributed little to the discus- 3. Stressed that cesarean section is major
sion. The attending physician opposed the inter- surgery, thus necessitating "an extra-
vention. However, other physicians present, ordinary case" before court-ordered
including the pediatrician, were not of one opin- intervention could be justified
ion. The hospital lawyers backed the attending 4. Concluded that the state's interest in
physician. preserving the Carder baby's life was not
On the basis of on Angela's evident willingness
truly compelling
to put herself at risk to have a child and on the
lack of a surrogate decision maker, the judge
5. Criticized the judge for not seeking a
concluded that the balance was weighted in favor substituted judgment as to Angela's
of the baby and that the state had sufficiently preferences
compelling interest to order cesarean section de- 6. Noted the inability of the judicial
livery. Entering Angela's room to prepare her for system to properly function. in such
surgery, the attending physician and a colleague circumstances
found the patient roused. He presented the
judge's decision as a fiat accompli. She agreed. The Angela Carder case has been classi-
When Angela's alertness and response were re- fied as a paradigm case for management of
ported to the judge, her tearful husband and her pregnancies complicated by disagreement
mother, along with her attending physician and a over the utilization of cesarean delivery or
senior colleague, went to her room for confirma- other prenatal interventions. 46 (A paradigm
tion. The mother urged her daughter not to at- is a way of seeing and interpreting data in
tempt delivery and instead carry the baby with the effort to make sense of life experiences.
her to heaven. Angela wanted to know who would A paradigm case would be a precedent-set-
be doing the surgery. Her attending physician ting case which, as a frame or point of refer-
told her he would if she authorized the surgery.
ence, provides clarification and guidance for
Finally she mouthed the words, "I don't want
it done." It was not clear to the physicians in
decision making in a present case. 52 ) We
the room whether Angela's response could be counterpropose that the Angela Carder case
considered an informed response. The reported is a paradigm case for mismanaging preg-
equivocation did not change the mind of the nancies complicated by disagreement over
judge, who had made his decision on grounds the utilization of cesarean delivery or other
other than the presence or absence of maternal prenatal interventions. Every facet of
14. Ethical Issues in the Utilization of Cesarean Section 199

the decision-making process seems to have administrators, lawyers, or judges faced

failed. with these difficult cases.,,54
1. The patient failed to prepare a func- We view this case as one tragedy (i.e., the
tional process of substituted judgment. patient story) overlaid by another tragedy
2. The patient's husband failed to inter- (i.e., the social and political response to the
pret clearly his understanding of story). The limitations embedded in this
his wife's perceptions or his own case leave the clinician with the question:
perceptions. "What should be done?"
3. The patient's mother disqualified her-
self as a surrogate decision maker by What Should Be Done?
appealing to burdens she might face
personally. The AMA Board of Trustees Report:
4. The attending physician left questions
undiscussed that should have been anti-
Legal Interventions During
-cipated in the event the mother deter- Pregnancy
iorated more rapidly than expected. A few months after the Angela Carder deci-
5. The attending physician's senior col- sion was reversed, the Board of Trustees
league brought the case to the atten- for the American Medical Association pub
tion of an administrator without strong lished a report in which their legal and
reason. policy concerns about court-ordered medical
6. The alerted administrator failed to treatments were discussed. 55 The report
gather sufficient information before argued that "a woman who chooses to carry
bringing in legal counsel. her pregnancy to term has a moral respon-
7. The hospital's ethics committee, ironi- sibility to make reasonable efforts toward
cally meeting just down the hall at cru- preserving fetal health," but drew a clear
cial points on the day of decision, was distinction between moral and legal respon-
never consulted. sibilities. The report advised physicians that
8. The judiciary, minus the time needed "the duty to protect the health of both the
to function appropriately, failed to pregnant woman and the fetus precludes
evaluate the mother's competency balancing one against the other" and that
and failed to establish a decision- the principle of informed consent "indicates
making process. that a pregnant woman's refusal of treat-
9. All parties involved failed to enable ment should not be overridden for the bene-
Angela's husband to take a significant fit of the fetus." The report adopted a "rule"
position in the process as the legitimate utilitarian position (rather than an "act"
surrogate decision maker. utilitarian position, in which the conse-
10. Commentators on the case, both in quences taken into consideration are limited
defense of and in opposition to the court- to the case at hand)36 in reasoning that
ordered surgery, failed, with few excep- "while the health of a few infants may be
tions, to show respect for the profoundly preserved by overriding a pregnant woman's
human and tragic dimensions of this decision, the health of a great many more
case in their grasp for political gain. may be sacrificed."
Far from closing discussion about these The report concluded that physicians,
cases, the significance of the Court of Ap- whose duty is to ensure that the pregnant
peals' ruling has been widely disputed. Some woman makes an informed and thoughtful
saw in the decision a sweeping policy that decision, "should refrain from using the
"turned the law in the proper direction.,,53 courts to impose personal value judgments
Others felt that the decision did not "provide on a pregnant woman who refused medical
any practical assistance to doctors, hospital advice." However, the report did leave room
200 T.E. Elkins and D. Brown

for "an exceptional circumstance ... in social status of the viable fetus. By the late
which a medical treatment poses an insigni- stages of gestation, the burden of proof sits
ficant-or no-health risk to the woman, with decision makers who propose actions
entails a minimal invasion of her bodily that diminish or eliminate the well-being of
integrity, and would clearly prevent sub- the fetus. This follows because by that time
stantial and irreversible harm to her fetus." the fetus can be treated as a patient, has
In such a situation, "it might be appropriate certain legal rights, and is being incor-
for a physician to seek judicial intervention." porated into the moral community.
This report attempts to provide guidance A comply-or-refer approach to decision
for cases such as the Angela Carder case, making is clearly inappropriate if carried
thus agreeing with the Court of Appeals beyond pregnancy into routine gynecologic
judge who, in his dissent in part, argued that care. For instance, the ACOG committee on
these cases are medically and ethically ethics debated and finally rejected a comply-
complex. 45 By contrast, one commentator or-refer approach to requests for steriliza-
concluded that there are no medical and tion for women with mental retardation. 57
ethical complexities to resolve "if the only In a similar way, a comply-or-refer approach
relevant issue is what the woman wants collapses when applied to requests for tubal
done.,,39 He further exempted differences of ligation reversal from a patient with a his-
judgment from these cases by dismissing tory of child abuse and manic-depressive
footnotes in a Court's opinion as space re- disorder.
served "for losers.,,39 We find these remarks Clinicians can easily multiply these illus-
difficult to apply to clinical realities and in trations. The point is that patients should
conflict with the recent revision by the rarely be in a position to dictate medical
ACOG Committee on Ethics of its position decision making. Patient and physician
on informed consent56 (in which the com- have critical and mutually dependent roles.
mittee affirmed that patient will is not the Physicians, although deeply respectful of
only factor to consider in medical decisions). patient judgment, will on occasion find
However, the Board of Trustees Report themselves with no ethical alternative but
and the commentator do encourage one cri- to hold to their judgment. Decisions about
tical question: Is the obstetrician's role the utilization of cesarean section for de-
simply a matter of complying with maternal livery fall between abortion decisions and
preference or referring to someone who will? prenatal genetic counseling on the one
As noted at the beginning of the chapter, hand and sterilization decisions on the other
"obstetrics ethics" does not occur in isolation hand.
from the many other clinical situations
faced in the full range of obstetric and gyne-
cologic care. "Comply or refer" may be good Ethics Statements by the American
law, but it is not good medicine. College of Obstetricians and
Complying with an informed patient's Gynecologists and the American
preference or referring the patient to a phy- Academy of Pediatrics on Conflicts
sician who will comply has been accepted in
our society as both legally and ethically
in Maternal-Fetal Care
appropriate management in cases involving The ACOG committee on ethics and the
pregnancy termination or prenatal genetic American Academy of Pediatrics (AAP)
counseling. However, continuing that pro- issued opinions (in October 1987 and June
cess past the mother's decision to remain 1988, respectively) about the ethical impli-
pregnant and into the late stages of gesta- cations of decisions involving conflict over
tion is seriously debated among equally the use of invasive interventions such as
thoughtful clinicians. The difference lies in cesarean section. 13,58 Both statements
the judgment about the medical, legal, and recognized that new technology has pene-
14. Ethical Issues in the Utilization of Cesarean Section 201

trated the mystery that had surrounded the to maternal choice, which "must not be
fetus. AAP differentiated the various pro- violated."
cedures as "standard practices of proven In response to maternal refusal, the
efficacy," "routine practices," and practices ACOG advised obstetricians to make the
founded on "a more ambiguous data base." case for the recommended procedure, to
The ACOG, with a tenuous view of all new encourage "responsible behavior," and to
techniques, focused on the limitations and urge the pregnant woman "to seek consul-
fallibility of the technology. tation and counseling." When the recom-
Both statements expected these multi- mended procedure is "of proven efficacy and
faceted conflicts to be rare. However, they has concomitant low maternal risk," the
differed sharply on two points in their des- AAP encouraged even more pressure by
cription of these cases. First, the AAP called sanctioning efforts to make parents "feel
attention to the attending physician's obligated to participate." Accordingly, the
"dilemma" when forced to decide which attending physician should stress "the
patient-the mother or the unborn child- responsibility of the mother to accept some
will receive primary attention. Although personal risk for the potential benefit of her
concern for the well-being of the mother and fetus."
the fetus was expressed, the ACOG assumed Both statements discouraged "actions of
that the relationship between the pregnant coercion to obtain consent or force a course of
woman and the obstetrician is the primary action." However, they took contrasting
relationship. Second, the AAP included the positions about managing an impasse that
father and extended family in the decision- cannot be resolved in a "forum of discussion"
making process. The ACOG made no men- (ACOG). The AAP listed three criteria:
tion of either the father or the extended "substantial likelihood" offetal morbidity or
family. mortality, a clearly appropriate interven-
Neither statement commented on the tion procedure, and low maternal risk.
moral or legal status of the viable fetus. These criteria are intended to identify those
Only the AAP referred to the pregnant "unusual cases" in which the physician's
woman as "mother" or "parent." Both lack of sympathy with the treatment refusal
statements expected physicians to respect warrants "actively opposing the woman's
the rights of maternal autonomy and bodily choice." In regard to those cases, the AAP
integrity. Both statements expected the sounded confrontational (e.g., allowing the
attending physician to be medically in- attending physician to inform the mother
formed, to formulate treatment plans de- that he or she "finds her refusal extremely
signed to "foster the greatest benefit with unreasonable" and to warn her "that the
the least risk" (ACOG), to educate the preg- hospital ethics committee might be brought
nant woman, and to make recommendations. into the case"). The AAP cautiously ap-
The ACOG left a surprisingly pater- proved, as "a last resort," an appeal to the
nalistic picture of a conflict-free pregnancy. courts.
Accordingly, the obstetrician should assess The ACOG clearly opposed such a res-
risks and benefits, decide on a course of ponse. Although stopping short of categori-
action, and ask the pregnant woman to cally rejecting court-ordered interventions,
consent. If she fails to "cooperate" (a term the ACOG offered no criteria for identifying
ACOG used three times), the obstetrician the implied exceptional cases (as did the
then attempts to "convey the reasons for Court of Appeals in the Angela Carder case).
the current recommendations to the preg- The ACOG did make a four-point case
nant woman." However, ACOG also implied against "the use of judicial authority to
that the obstetrician's strong leadership implement treatment regimens in order to
in decision making, when confronted with protect the fetus": (1) benefit to the fetus
persistent maternal refusal, should yield cannot be "guaranteed absolutely"; (2) court
202 T.E. Elkins and D. Brown

orders for fetal intervention often undercut 5. When it is clear that the refusal would be
the reliability of the judicial system by re- contested without regard to the woman's
quiring a quick decision; (3) such man- social privilege or minority status
agement threatens or even destroys the 6. When plans are in place to diligently
physician's relationship with the preg- seek ways to repair any damage to the
nant woman; and (4) "undesirable societal pregnant woman's "life story"
consequences, such as the criminalization
We are persuaded that active opposition
of noncompliance with medical recommen-
should be exercised when, given the fore
dations," may follow.
going set of circumstances, maternal intent
is to harm or kill the fetus at term (i.e., truly
Suggested Criteria maternal-fetal conflict). In such truly excep-
Given the potential for disagreement over tional situations, the conflict is no longer
how to respond to maternal refusal of a between patient and physician over method
recommended cesarean section, it should be of delivery with both hoping for a "good
noted that both the ACOG and AAP en- outcome."
couraged consensus by affirming a pregnant
woman's right to privacy, defining an ideal
outcome in terms of maternal and fetal well- Conclusion
being, and expressing skepticism about
appeals to the judicial system. These cases, although very rare, lend cre-
However, we find that, when adapted to dence to the judgment that "obstetrics and
clinical situations, both statements fall gynecology are being devoured by ethical
short. The criteria in the AAP statement are issues.,,5 In" this chapter, we have disting-
too elastic to keep the cases "extremely uished obstetric ethics within the larger
rare." The ACOG statement, by failing to study of "ethics," looked through clinician
offer criteria by which to identify truly ex- eyes at the dynamics surrounding the deci-
ceptional cases, leaves the reader unpre- sion to recommend cesarean section de-
pared to detect or respond thoughtfully to livery, placed a literature context around
such cases. cases in which the pregnant woman refuses
The literature addresses the need for cri- such recommendation, reviewed profes-
teria that are clinically realistic, that take sional guidelines, and offered criteria for
seriously the complexities of these cases, conflict management.
and that keep the cases exceptiona1. 33 ,59-66 The minimal attention to perinatal ethics
We propose that actively opposing informed that has traditionally characterized resi-
maternal refusal of cesarean section de- dency training needs to be radically revised,
livery should only be considered: if physicians attending childbirth are to
1. When maternal opposition occurs at or participate responsibly in cases involving
near term67 complex ethical decisions. It is false to
2. When the procedure puts the pregnant assume that technical authority in matters
woman at essentially no greater risk of medical indications implies ethical auth-
ority in matters of values and beliefs. 69
than nonintervention does
3. When the morbidity/mortality risk to the
fetus and/or newborn is virtually certain
and very significant68 References
4. When the fetus' condition is clearly dis- 1. The APGO Task force on Medical Ethics.
sociated from neonatal reference points Exploring issues in obstetrics and gynecolo-
such as anencephaly, Potter's syndrome, gic medical ethics. Washington, DC: The
extreme osteoporosis, and other lethal or Association of Professors of Gynecology and
near-lethal anomalies Obstetrics, 1990.
14. Ethical Issues in the Utilization of Cesarean Section 203

2. Frankena W. Ethics, 2nd Ed. Englewood 20. Pellegrino E. The anatomy of clinical-ethical
Cliffs, New Jersey: Prentice-Hall, 1983. judgments in perinatology and neonatology:
3. Ryan K. Ethics in obstetrics and gynecology. a substantive and procedural framework.
Am J Obstet Gynecol 1985;151:840-843. Semin Perinatol 1987;11:202-209.
4. Hill E. Your morality or mine? an inquiry 21. Bosk C. Social, medical, and ethical dilem-
into the ethics of human reproduction. Am J mas in fetal medicine. In: Milunsky A, Annas
Obstet Gynecol 1986;154:1173-1180. GJ, eds. Genetics and the law, Vol. III. New
5. Park R. Old bedfellows: ethics and obstetrics York: Plenum, 1985.
and gynecology. Obstet Gynecol 1989;73: 22. Chervenak FA, McCullough LB. What
1-3. is obstetric ethics? Clin Obstet Gynecol1992;
6. Moffic S, Coverdale J, Bayer T. The 35:709-719.
Hippocratic oath and clinical ethics. J Clin 23. Nelson LJ, Milliken N. Compelled medical
Ethics 1990;1:287 -292. treatment of pregnant women: life, liberty,
7. Bulger R, ed. In search of the modern and law in conflict. JAMA 1988;259:1056-
Hippocrates. Iowa City: University of Iowa 1066.
Press, 1987. 24. Price MR, Broomberg J. The impact of the
8. Beauchamp T, Childress J. Principles of fee-for-service reimbursement system on the
biomedical ethics, 3rd Ed. New York: Oxford utilization of health serVIces. Part III: a
University Press, 1989. comparison of caesarean section rates in
9. Elkins T. Introductory course in biomedical white nulliparous women in the private and
ethics in the obstetrics-gynecology residency. public sectors. S Afr Med J 1990;78:136-138.
J Med Educ 1988;63:294-300. 25. Goyert GL, Bottoms SF, Treadwell MC,
10. Silver RK, Minogue J. When does a statis- Nehra PC. The physician factor in cesarean
tical fact become an ethical imperative? Am birth rates. N Engl J Med 1989;320:706-709.
J Obstet Gynecol 1987;157:229-233. 26. Stafford RS. The impact of nonclinical factors
11. MacIntyre A. After virtue. Notre Dame: on repeat cesarean section. JAMA 1991;
University of Notre Dame Press, 1983. 265:9-63.
12. Engelhardt T. The foundations of bioethics. 27. Kirk EP, Doyle KA, Leigh J, Garrard ML.
New York: Oxford University Press, 1986. Vaginal birth after cesarean or repeat cesa-
13. The American College of Obstetricians and rean section: medical risks or social realities?
Gynecologists Committee on Ethics. Patient Am J Obstet GynecoI1990;160:1398-1403.
choice: maternal-fetal conflict. Washington, 28. Evans MI, Richardson DA, Sholl JS, Johnson
DC: American College of Obstetricians and BA. Cesarean section: assessment of the
Gynecologists, 1987. convenience factor. J Reprod Med 1984;
14. Reich W. Caring for life in the first of it: 29:670-676.
moral paradigms for perinatal and neonatal 29. Johnson SR, Elkins TE, Strong C, Phelan JP.
ethics. Semin Perinatol 1987;11:279-287. Obstetric decision-making: responses to pa-
15. Pellegrino E, Thomasma D. For the patient's tients who request cesarean delivery. Obstet
good: the restoration of beneficence in health Gynecol 1986;69:847-850.
care. New York: Oxford University Press, 30. Morreim EH. Ethical considerations in obs-
1988. tetric management. In: Evans MI, et aI., eds.
16. Dunstan G, Shinebourne E, eds. Doctor's Fetal diagnosis and therapy: science, ethics,
decisions: ethical conflicts in medical and the law. Philadelphia: Lippincott, 1989:
practice. New York: Oxford University Press, 233-242.
1989. 31. Brown D, Elkins TE. Ethical issues in obste-
17. Drane J. Becoming a good doctor: the place of tric cases involving prematurity. Clin
virtue and character III medical ethics. Perinatol 1992;19:469-481.
Kansas City: Sheed and Ward, 1988. 32. Annas GJ, Glantz LH, Mariner WK. The
18. Brody H. My story is broken; can you help me right of privacy protects the doctor-patient
fix it? Medical Ethics and the joint construc- relationship. JAMA 1990;263:858-861.
tion of narrative. Lit and Med Spring 1994; 33. Jonsen AR. Involuntary treatment in medi-
13:79-92. cine. Annu Rev Med 1986;37:41-48.
19. Guillemin J. Babies by cesarean: who 34. Rhoden NK. Informed consent in obstetrics:
chooses, who controls? Hastings Center Rep some special problems. West Engl Law Rev
1981;11:15-18. 1987;9:67 -88.
204 T.E. Elkins and D. Brown

35. Obade CC. In re A.C. reversed: judicial re- 53. Curran WJ. Court-ordered cesarean sections
cognition of the rights of pregnant women. receive judicial defeat. N Engl J Med 1990;
J Clin Ethics 1990;1:251. 23:489-492.
36. Fleischeman AR, Macklin R. Fetal therapy: 54. Pacheco DA. Court Ordered Cesarean Sec-
ethical considerations, potential conflicts. tions. [Letter to the editor.] N Engl J Med
In: Weil WB, Benjamin M, eds. Ethical issues 1991;324:272-273.
at the outset of life. Boston: Blackwell, 55. Law and Medicine/Board of Trustees Report.
1987:121-148. Legal interventions during pregnancy: court-
37. Kaufmann CL, Williams PRo Fetal surgery: ordered medical treatments and legal
the social implications of medical and surgi- penalties for potentially harmful behavior
cal treatment of the unborn child. Women by pregnant women. JAMA 1990;264:2663-
Health 1985;10:25-37. 2670.
38. Shultz M. Informed consent: a symbol anal- 56. The American College of Obstetricians and
yzed. Hastings Center Rep 1987;17:24-25. Gynecologists Committee on Ethics. In-
39. Annas GJ. Foreclosing the use of force: A.C. formed Consent. Washington, DC: American
reversed. Hastings Center Rep 1990;20:27- College of Obstetricians and Gynecologists,
29. 1992.
40. Brody H. The healer's power. New Haven, 57. The American College of Obstetricians and
Connecticut: Yale University Press, 1992. Gynecologists Committee on Ethics. Sterili-
41. Kolder VE, Gallagher J, Parsons MT. Court- zation of women who are mentally handi-
ordered obstetrical interventions. N Engl J caped. Washington, DC: American College of
Med 1987;316:1192-1196. Obstetricians and Gynecologists, 1988.
42. Annas GJ. Protecting the liberty of preg- 58. American Academy of Pediatrics Committee
nant patients. N Engl J Med 1987;316: on Bioethics. Fetal therapy: ethical consider-
1213-1214. ations. Pediatrics 1988;81:898-899.
43. Elkins TE, Brown D, Barclay M, Andersen 59. Elkins TE, Andersen HF, Barclay M, et al.
HF. Maternal-fetal conflict: a study ofphysi- Court-ordered cesarean section: an analysis
cian concerns in court-ordered cesarean of ethical concerns in compelling cases. Am J
sections. J Clin Ethics 1990;1:316-319. Obstet Gynecol 1989;161:150-154.
44. In re A.C. No. 87-609, District of Columbia. 60. Elkins TE. The case for a middle ground.
Atlantic Reporter, 2nd series. November 10, Women's Health Issues 1990;1:34-36.
1987 (date of decision); 533:611-617. 61. Rosner F, Bennett AJ, Cassell EJ, et al. Fetal
45. In re A.C. No. 87-609, District of Columbia. therapy and surgery: fetal rights versus
Court of Appeals, en banco Atlantic Reporter, maternal obligations. N Y State J Med 1989;
2nd series. April 26, 1990 (date of decision); 89:80-84.
573:1235-1264. 62. Fletcher JC. Drawing moral lines in fetal
46. DS Davis. Reflections on A.C. BioLaw 1990; therapy. Clin Obstet Gynecol 1986;29:595-
(special section);448-451. 601.
47. Annas GJ. She's going to die: the case of 63. Chervenak FA, McCullough LB. Clinical
Angela C. Hastings Center Rep 1988;18:23- guides to preventing ethical conflicts be-
25. tween pregnant women and their physicians.
48. Mishkin B. [Letter to the editor.] Hastings Am J Obstet Gynecol1990;162:303-307.
Center Rep 1988;18:40-41. 64. Kluge EW. When cesarean section opera-
49. Annas GJ. [Reply]. Hastings Center Rep tions imposed by a court are justified. J Med
1988;18:41-42. Ethics 1988;14:206-211.
50. White ML. Reflections on In re A.C.: a trage- 65. Chervenak FA, McCullough LB. Perinatal
dy of decision making. BioLaw 1990;(spechll ethics: a practical method of analysis of
section):433-442. obligations to mother and fetus. Obstet
51. Allen AE. In re A.C.-an affirmation of Gynecol 1985;66:442-446.
ACOG committee opinion number 55: 66. Strong C. Ethical conflicts between mother
maternal-fetal conflict. Women's Health and fetus in obstetrics. Clin Perinatol 1987;
Issues 1990;1:37-40. 14:313-328.
52. Childress JF. Analogical reasoning: organ/ 67. Anderson G, Strong C. The premature
tissue donation and caesarean sections. breech: cesarean section or trial of labor? J
BioLaw 1990;(special section):443-447. Med Ethics 1988;14:18-24.
14. Ethical Issues in the Utilization of Cesarean Section 205

68. Chervenak FA, McCullough LB. Ethical medical authority: the nature of authority
analysis of the intrapartum management of in medicine for decisions by lay persons
pregnancy complicated by fetal hydroce- and professionals. In: Agich G, ed. Responsi-
phalus with macrocephaly. Obstet Gynecol bility in health care. Dordrecht: Reidel, 1982:
1986;68:720-724. 127-139.
69. Veatch R. Medical authority and professional
15
The Patient Who Demands
Cesarean Delivery
BRUCE L. FLAMM

The Cesarean Decision doctor- patient interactions as a fairly

symmetrical bell-shaped curve centered
Spectrum around the point of total agreement is hypo-
thetical but does serve to illustrate two very
Although very little research has been done important points.
on the subject of cesarean decisions, it is First, the shape of the curve would almost
probably safe to say that in most cases the certainly change depending on whether the
patient and her physician are in full agree- interaction was measured before or dur-
ment about the need (or lack of need) for ing labor. While the majority of patients
cesarean delivery. The doctor-patient in- and their physicians remain in agreement
teraction would thus fall near the middle of about the need (or lack of need) for cesarean
what I have termed the "cesarean decision throughout the entire antepartum and in-
spectrum" (Fig. 15.1). However, there are trapartum interval, a significant number of
many cases in which at least some degree of women request or even demand medically
conflict exists and the interaction would fall unindicated cesarean delivery in the later
slightly on either side of the middle of the stage of labor. This might be illustrated
spectrum. For example, position 4 on the as a shift of the curve to the right with the
spectrum would represent a woman who majority of docto~-patient interactions still
would prefer not to have a cesarean section centered near the line of total agreement
but certainly would not refuse one if it was (Fig. 15.2). The secondary peak at the far
recommended by her physician. Position 6 right of the curve represents those women
on the scale represents a woman who would who strongly request or even emphatically
prefer a cesarean delivery but would not demand a cesarean at some point during
demand one if her physician explained that their labor. Note that the numbers at the
it was not medically indicated. Position bottom of the spectrum signify the strength
number 1 at the far left end of the spectrum of agreement or disagreement between the
represents the unusual case in which the patient and her doctor, not cervical dilata-
patient absolutely refuses a cesarean in tion. If dilatation was plotted against pa-
spite of her doctor's insistence that the oper- tient demand for cesarean the resulting
ation is urgently needed. Finally, number 10 graph might look something like Figure 15.3.
at the far right of the spectrum represents The second point that is easy to illustrate
the case in which the patient absolutely on the cesarean decision spectrum is that
demands a cesarean operation although her while almost all discussions of cesarean
doctor has explained it is clearly not needed. decision conflicts in the medical, ethical, and
The representation of the distribution of legal literature have focused on the cases

207
 208 B.L.Flamm
100
FIGURE 15.1. The
cesarean decision
Pt and MD Agree spectrum. Pt, patient;
80
MD, physician; CS,
cesarean section.

60

40

20 Pt Demands CS

1 2 3 4 5 6 7 8 9 10

Interaction Scale (5 = total agreement)

100

80 Pt and MD Agree

60

40
Pt Demands CS

Pt Refuses CS
20

o ~.
1 2 3 4 5 6 7 8 9 10 FIGURE 15.2. The
cesarean decision
Interaction Scale (5 =total agreement) spectrum in active labor.

that fall at the extreme left side of the spec- end of the spectrum have been virtually
trum (coerced or court-ordered cesareans), ignored. This is unfortunate because con-
such cases are extremely rare (Fig. 15.4). flicts at the right of the spectrum occur far
Although they often make national head- more frequently in the day-to-day practice of
lines, only a few dozen such cases have ever obstetrics.
been documented. 1 - 5 This is not meant to be Although obstetricians must frequently
critical of the amount of attention that has deal with requests or demands for medically
been devoted to court-ordered cesareans but unindicated cesareans, methods of hand-
only to point out that issues at opposite ling such situation are not even mentioned
15. The Patient Who Demands Cesarean Delivery 209

FIGURE 15.3. Opinions about ces- 12

arean change during labor.

10

UJ 8
(.)
Cl
c
'5
c 6
III
E • Hypothetical Data
~
:.e
0
4

o
1 2 3 4 5 6 7 8 9 10

Cervical Dilatation (cm)

100

Pt and MD Agree
80

• Interactions
E3 Media Coverage
60

40 pt Demands CS

20

FIGURE 15.4. The media has o~.

1 2 3 4 5 6 7 8 9 10
focused on rare, court-ordered
cesareans. Interaction Scale (5 = total agreement)

III current obstetric textbooks. Williams When the search was broadened to include
Obstetrics does not discuss the issue but does "request cesarean" rather than "demand
point out that, "In modern obstetrical prac- cesarean" (moving from the right end of the
tice, there are virtually no contraindications decision spectrum toward the middle), only
to cesarean section."6 Perhaps even more two additional papers were found. s.9 The
surprisingly, a computerized search of 10 ethics chapter (Chapter 14) in this volume
years of medical literature revealed only one delves deeply into the issue of the court-
paper devoted specifically to the subject of ordered cesarean (left end of spectrum) but
women who demand cesarean operations. 7 only briefly mentions patient-demanded
210 B.L. Flamm

cesarean. Such a focus was not requested by

the editors; it simply reflects the current No prior Prior
emphasis on court-ordered cesarean in the cesarean cesarean
field of medical ethics. In contrast, this chap-
ter focuses on conflicts that arise on the Patient demands AI, A2 Bl, B2
right side ofthe spectrum: patient-requested cesarean section
and patient-demanded cesarean. before labor
Patient demands Cl, C2 Dl, D2
cesarean section
The Cesarean Decision Grid during labor

Before going on to discuss patient-requested Again, "I" is designated as definitely no

and patient-demanded cesareans, it is useful medical indication for a cesarean, and "2" is
to introduce another straighforward concept designated as a possible (but not definite)
that I call the cesarean decision grid. Con- medical indication for a cesarean.
flicts that arise when a patient demands a Almost every possible cesarean decision
medically unindicated cesarean can be di- conflict that could arise will fit into one of
vided into four separate types, as shown in the eight divisions of this grid. Examples of
the following grid. Because special circum- each of these eight scenarios are now dis-
stances arise in each of these cases that do cussed in detail. Note that although the
not apply to the others it is useful to discuss distinction between patients who request,
them separately. The only information strongly request, or actually demand a
needed for the classification is whether the cesarean is not always clear, the following
patient is in labor and whether she has had discussions would apply fairly well to all
a previous cesarean operation. these scenarios. It should be again em-
phasized that all the cases discussed here
reflect situations in which some degree of
No prior Prior
conflict exists. Decisions that fall in the
cesarean cesarean
middle of the cesarean decision spectrum
Patient demands imply that the patient and her physician
A B
cesarean section agree about the need (or lack of need) for
before labor cesarean delivery. This encompasses, by far,
Patient demands C D the majority of cases. The eight divisions of
cesarean section the grid could also be used to categorize and
during labor evaluate conflicts that arise at the left end of
the cesarean decision spectrum (court-
ordered cesarean, etc.); however, that is
To ensure that the grid fully addresses all beyond the scope of this chapter.
the conflicts that might occur in a typical
obstetric practice, it is useful to take the
process one step further and subdivide the Type "A" Cases
grid into situations in which there is de- Case A: The patient has no prior ce-
finitely no medical indication for a cesarean sarean and demands a cesarean
(designated "I") and cases in which there is before labor (i.e., an elective
a possible but not definite medical indication primary cesarean).
for the operation (designated "2"). The final
grid then looks like the following: These cases are by far the rarest of the
possible scenarios. Recall that we are not
discussing primary cesareans in which no
conflict exists, such as the case of a woman
15. The Patient Who Demands Cesarean Delivery 211

who has premature labor and breech presen- weeks) and are subsequently involved in
tation along with ruptured membranes and some type of catastrophe resulting in severe
requests a cesarean. neurologic damage or fetal death. They
plugged this information into an equation
Case AI: The patient has no prior cesa- along with data involving the risks of cesa-
rean and demands a cesarean rean and reached some very interesting
before labor but there is no conclusions. For one thing, they concluded
medical indication (i.e., an that both societal and legal factors and, "our
elective primary cesarean with own data all seem to support the notion that
no medical indication). patients ought to be given a definitive oppor-
tunity to consider electing prophylactic
The prime example of this type of case is cesarean section, especially since personal
the woman who has no prenatal problems and subjective value judgments are as im-
whatsoever but decides, at some point dur- portant to the decision as uncertain quanti-
ing her pregnancy, that she wants a cesa- tative estimates of morbidity, mortality, and
rean delivery before labor begins. Such a cost." They went on to say, "If an informed
woman may have a great fear of labor pain patient opts for prophylactic cesarean sec-
or perhaps a concern that labor is risky for tion at term, can it be denied?" Recall that
her baby and hence demands to be delivered they were not discussing women with prior
by cesarean for personal reasons. Although cesareans or high-risk patients with special
this scenario may seem ludicrous to many risk factors. Their conclusions were meant
physicians (and nonphysicians), a scholarly to apply to normal low-risk pregnant women.
argument supporting a woman's right to One of the letters to the editor prompted
make such a decision was made by Drs. by this article pointed out that, "requiring
Feldman and Freiman in an article pub- that healthy women be informed of a sur-
lished in the prestigious New England gical alternative to a normal physiologic
Journal of Medicine in 1985. 9 The authors process" would probably not represent rea-
became interested in the subject after re- sonable legal doctrine or public policy.10 The
viewing a lawsuit involving a placental letter went on to say, "As malpractice suits
abruption that occurred during labor at based on unnecessary cesarean sections
term. The doctor had apparently acted become more common, might not a court
quickly and did all that was medically pos- someday rule that a physician should have
sible under the circumstances, but the baby informed a pregnant woman of the availa-
had permanent neurologic damage. Al- bility of midwife-attended, out-of-hospital
though the malpractice charge was even- birth as a prophylactic measure against an
tually dismissed after a lengthy trial, the unwanted cesarean section?" Although the
authors were haunted by a novel approach Feldman and Freiman article was initially
taken by the plaintiff's attorney. The attor- considered by many readers (including
ney had asked the question, "Would this myself) to be some type of joke or parody, it
baby be alive, healthy, and undamaged was not. It did, in fact, raise several impor-
today if it had been delivered by cesarean tant issues that will no doubt be debated
section one week earlier?" The attorney further in coming years. In the meantime,
went on to ask, "Don't you think a mother where do we stand with the issue of elective
has the right to assume the extra risks of primary cesarean on demand?
cesarean section for the sake of her unborn With regard to the scientific (medical)
child if she wishes to?" aspects of this question, it would seem that
In an attempt to answer this question, most obstetricians currently believe that the
Drs. Feldman and Freiman set up a hypo- cesarean operation has not become so safe
thetical model. They assumed that between that we can offer it to anyone who desires it.
0.2% and 2% of fetuses reach maturity (37 Although there are apparently a few obste-
212 B.L. Flamm

tricians who do perform elective primary rious experiences. Some of the women were
cesareans (with no medical indication) on believed to have "substantial psychiatric
request, they are currently in the distinct problems," and several did have elective
minority. With regard to the legal aspects primary cesareans (some under general
of this question, the fact that the doctor anesthesia). In cases in which fear becomes
prevailed in the lawsuit in the case that pathologic and does not respond to counse-
spawned the Feldman and Freiman article ling or psychotherapy, the line between type
is reassuring. However, if the judicial sys- Al (no medical indications) and A2 (possi-
tem allows the same argument ("Would this ble medical indications) begins to blur.
baby be alive, healthy, and undamaged
today if it had been delivered by cesarean Case A2: The patient has no prior cesa-
section one week earlier?") to be used in rean and demands a cesarean
similar cases, other physicians may not fare before labor but there is no
so well. definite medical indication (an
Perhaps the key element that must be elective primary cesarean with
proven for a plaintiffs attorney to win a possible medical indications).
malpractice case is that the doctor breached
a standard of care. If it is ever decided that An example of the A2 type of case is a
one such "standard of care" is that all preg- woman with a difficult prior vaginal birth
nant women should be informed of the op- that resulted in a fourth-degree laceration
tion of elective primary cesarean before (tearing into the rectum). If the tear was
labor (for any reason they desire), then any repaired without lasting sequelae, most
physician who failed to do so would risk obstetricians would certainly not think that
losing a lawsuit any time a labor did not subsequent vaginal birth was contrain-
have a perfect outcome. This would be true dicated. But if the patient went on to develop
even if the quality of care during labor was a rectovaginal fistula that was successfully
not contested. In other words, the doctor repaired, the decision becomes more dif-
could lose the case even if he or she did ficult. It certainly possible that the previous
absolutely nothing wrong and had not repair could break down during childbirth.
breached any other standards of care. In a survey that addressed this very situa-
The A1 type of case, elective primary tion (although the demand for cesarean was
cesarean with no medical indication, was made during labor), fully 55% of obstetri-
also discussed in a Swedish study in which cians said that they would agree to perform
10 of 33 women who demanded cesarean a cesarean. 8
before labor did so solely because of fear of Because type A (and type B) cases occur,
the pain of childbirth. 7 Many of the women by definition, before labor, there are some
had feared childbirth from an early age, and options available. First, there is time to
several reported that they had been brought discuss the situation at length with the pa-
up by mothers who told them horrendous tient and her partner. With time and under-
childbirth stories. The publication pointed standing, a consensus can often be reached.
out that, "There are no private obstetric It may also be wise to obtain a second opin-
hospitals in Sweden. Thus, a Swedish wo- ion from another physician or perhaps even
man cannot buy a cesarean." However, the present the case at a departmental meeting
women's concerns were taken seriously and to obtain the opinions of several colleagues.
each patient· was seen from 1 to 13 times If the doctor and patient still cannot reach
during her pregnancy for counseling and an agreement, there may also be time for the
short-term psychotherapy. Of the 33 women patient to seek the care of another physi-
who demanded cesarean, 14 went on to have cian. Of course, these options do not exist
uncomplicated vaginal births and 11 of when conflicts arise during labor (see case
these described their births as good or glo- types C and D).
15. The Patient Who Demands Cesarean Delivery 213

Type "B" Cases involved a prior cesarean and a ruptured

uterus. The tragic outcome occurred in spite
Case B: The patient has had a prior
of the rapid performance of a cesarean sec-
cesarean and demands a cesa-
tion at the first signs of fetal distress. The
rean before labor (an elective
physician had documented the prior uterine
repeat cesarean).
scar to be low transverse and had managed
Type B cases represent some of the most the labor appropriately. This was reaffirmed
common and controversial dilemmas faced by the fact that the plaintiff's attorney was
by contemporary obstetricians. While obste- unable to show that the obstetrician had
tric textbooks have yet to discuss this situa- breached any standard of care.
tion, such cases are currently being debated Just as it appeared that the case was over
on a daily basis at obstetric staff meetings and that the defendant physician was not
all over the United States and in many other guilty of any negligence or malpractice, the
countries around the world. plaintiff's attorney raised the issue of in-
formed consent. He argued that his client
Case B1: The patient has had a prior had wanted an elective repeat cesarean and
cesarean and demands a cesa- that if this "simple" procedure had been
rean before labor but there is done, the baby would not have suffered any
no medical indication (an elec- injury. Although the patient claimed to have
tive repeat cesarean with no insisted on an elective repeat cesarean long
medical indications). before the onset of labor, the fact that she
presented at term in labor argued against
An example of a B1 type of case is the that contention. If the doctor had indeed
patient who had a primary low transverse refused to go along with her request she
cesarean several years ago for breech pre- would have had ample time to seek the care
sentation and refuses a subsequent trial of of, or at least the opinion of, another physi-
labor. A multitude of data show that such a cian. This had not been done. But the plan-
patient has a very high (;;:=:75% probability) tiff's attorney continued to argue that it is
of vaginal birth and that the risk of uterine the right of the patient, not a physician, to
rupture is very low «1%).11- 13 decide if an operation will be performed. He
During the past 20 years, dozens of studies went on to forcefully point out to the jury
have addressed technical issues related to that doctors are not "gods".
trial of labor. Several of these issues are Perhaps not coincidentally, just before the
discussed in detail in another chapter of this case went to trial, a movie entitled "Malice"
book. However, one perplexing issue has not had shown all over America. Alec Baldwin
been resolved. Who decides if a given patient starred as a brilliant but arrogant young
will have an elective repeat cesarean or a surgeon who, in a chilling courtroom scene
trial of labor? This question has already relating to the need for informed consent,
surfaced as a major issue in malpractice essentially stated that he was indeed a god.
litigation. Some doctors fear that if a uterine Undoubtedly some of the jurors had recently
rupture (or any significant problem) occurs seen the movie. In spite of the plaintiff at-
during a trial of labor, an attorney will torney's cleverly timed innuendos, the jury
always be able to argue that the doctor sided with the defendant doctor and agreed
should have ''just'' done an elective repeat that the trial of labor had been reason-
cesarean. This chain of events has now able and the uterine rupture had been
taken place. Similar to the case presented by unavoidable.
Drs. Feldman and Freiman in their New The outcome of this case should not be
England Journal of Medicine article dis- misconstrued as signifying that poor out-
cussed earlier, this case involved a neurolo- comes relating to uterine rupture carry little
gically impaired infant. However, this case medical-legal risk. For example, a Utah
214 B.L.Flamm

obsetrician recently lost a case in which a if a given patient will have an elective
trial of labor had ended with a uterine rup- repeat cesarean or a trial oflabor? The most
ture and a brain-damaged baby. The plain- recent American College of Obstetricians
tiffs attorney argued that trial of labor and Gynecologists (ACOG) Guidelines for
had been instituted as a cost-containment Vaginal Delivery after Previous Cesarean
measure. The judge apparently directed a Birth recommend that " ... a woman with
defense verdict on the cost-containment one previous cesarean delivery with a low
issue because this charge was not proven by transverse incision should be counseled and
the evidence, but the jury came back with encouraged to attempt labor in her cur-
a guilty verdict and awarded the plain- rent pregnancy.,,15 But how strongly should
tiff $8 million. Whether juries composed she be encouraged? What if, after coun-
of lay individuals have the capacity to seling, she is not interested in a trial of
render reasonable verdicts in cases involv- labor? These are very difficult questions
ing extremely complex medical issues is to answer.
debatable. Clearly, the way trial oflabor is presented
Sensational trials are even shedding to a patient will have a profound effect on
doubt on the ability of the jury system to whether or not she is willing to give it a try.
deal with cases that do not require an ad- A doctor could say, "If you opt for a trial of
vanced degree of medial sophistication. For labor you must fully understand that your
example, in Los Angeles, separate trials of uterus could rupture and your baby could
Lyle and Eric Menendez both concluded die." This is certainly true but does not put
with mistrials because neither jury could the risks into proper context. Few women
decide whether the verdict should be murder given such a morbid description would even
or manslaughter. This was true in spite of ponder an attempt at labor. Another doctor's
the fact that the brothers, both of whom version of informed consent might be, "If you
were adults, had already confessed to killing opt for a trial of labor, many large studies
their wealthy parents with repetitive shot- have shown that you will probably be able to
gun blasts. 14 After pumping several rounds avoid a repeat cesarean and that the risk of
into .each of their parents, the brothers your uterus rupturing is less than 1%. In the
admitted to going outside to reload and then small number of cases where the uterus does
returning to the house to finish the job. Al- rupture, the baby will often do well. But in
though both brothers claimed to have been some of these c;lses the baby may have pro-
previously abused by their father they were blems and in very rare cases the baby could
now adults and could have moved away from die."
home at any time. Retrial, if it occurs, may Some doctors state that few if any of their
take more than a year and cost taxpayers patients agree to trial of labor. This may
additional millions of dollars. If the current have a lot to do with exactly how the options
jury system cannot handle cases where de- are explained. A recent publication reported
fendants openly confess to homicide, how a trial oflabor rate of only 19% (81 % elective
can the same system be expected to deal repeat cesarean sections) and concluded that
with complex issues such as deciding exactly because of both medical and patient-choice
what is the proper dose of oxytocin to utilize factors in favor of cesarean, "it will be dif-
during labor after a prior low transverse ficult to substantially decrease the present
cesarean or exactly what electronic fetal rate of repeat cesarean births.,,16 However,
heart monitor patterns indicate that a cesa- 1 year later a prospective study of 7229
rean operation should have been performed? patients with prior cesareans reported that
Patients with prior cesareans represent a fully 70% opted for trial of laborP Rates of
significant and ever-increasing portion of trial of labor varying from 19% to 70% can-
the pregnant population. This brings .us not be explained by medical factors or demo-
back to the initial question. Who decides graphics and must therefore reflect the
15. The Patient Who Demands Cesarean Delivery 215

way that the options are being presented to panel should rule that no lawsuit is war-
patients. ranted. As is discussed later in this volume
This is not to say that the typical patient (Chapter 17, Cesarean Projects at the State
will jump for joy at the prospect of going and National Level), such a system is now
through labor if her physician has an opti- being tested in the state of Maine. The cur-
mistic attitude. Rather it implies that the rent hostile environment, in which any
majority of women will at least be willing to doctor can be dragged through a horrendous
give it a try. Nevertheless, some reports lawsuit simply because a patient "wanted a
contend that it is often reluctance of the cesarean and didn't get it," has led many
patient, not her obstetrician, that prevents American doctors to conclude that the only
greater utilization of vaginal birth after defense is to quickly comply with every pa-
cesarean section (VBAC). One study found tient's request for cesarean delivery.
that 50% of patients who were encouraged It should also be pointed out, however,
by their obstetrician toward VBAC opted for that in the current state of affairs doing a
elective repeat cesarean without trial of repeat cesarean does not free an obstetrician
labor. 18 Another study found that 40% of from potential liability . In a recent malprac-
patients demanded elective repeat cesarean tice case in Massachusetts a woman claimed
although they fulfilled the criteria for eligi- that her doctor had agreed to letting her
bility in a VBAC program. 19 A reason fre- have a trial of labor. However, 2 weeks be-
quently given by these patients was that fore her due date the doctor told her he was
they wished to know their dates of admis- going on vacation and was going to schedule
sion, delivery, and discharge months in her for a repeat cesarean. When she objected
advance in order to arrange their schedules he told her that labor would be risky to her
accordingly. Obviously, plaQ.ned elective and her baby and allegedly compared her
repeat cesarean provides similar scheduling uterus to a "hydrogen bomb." The women
advantages for the physician and hospital developed severe gastrointestinal complica-
staff. tions after the repeat cesarean and required
It is my opinion that the issue of who prolonged hospitalization. She sued, claim-
decides if a given patient will have an elec- ing that the cesarean was unwanted and
tive repeat cesarean or a trial of labor has unnecessary. The doctor was found negli-
not been resolved. To many attorneys and gent and the patient was awarded $1.53
perhaps to many jury members, it may seem million by a superior court jury.20,21
obvious that the decision to have (or not to
have) surgery is always an uninfringable Case B2: The patient has had a prior
right of a mentally competent patient. cesarean and demands a cesa-
However, things are not quite so simple. rean before labor but there
Does any women who is concerned about is no definite medical indica-
the risk of uterine cancer have the right to tion (an elective repeat cesa-
demand an elective hysterectomy? Can a rean with possible medical
physician be compelled to perform an opera- indications).
tion that he or she feels to be unnecessary?
If society agrees that the current cesarean A typical example of a B2 type of case
rate is too high, then tort reform must em- would be a woman who planned to have a
phasize the fact that a bad outcome is not trial of labor but changes her mind near
synonymous with malpractice. term because it appears that her baby is too
A logical system would mandate an initial large. She then requests or demands an elec-
review which would screen out frivolous tive repeat cesarean. This creates a very
lawsuits. In cases in which a poor outcome difficult dilemma. Studies have shown that
occurs in spite of good obstetric care with both clinical and sonographic estimates
adherence to basic guidelines, an arbitration of fetal macrosomia are inaccurate. 22 - 27
216 B.L.Flamm

Even if we could accurately predict fetal sign of fetal distress. This reasonable plan
macrosomia, it has been shown that the should not be confused with a "perfunctory"
majority of trials of labor will result in un- trial of labor in which the real plan is to
eventful vaginal births. We retrospectively perform a repeat cesarean an hour or two
reviewed 301 trials of labor in patients with after admission to the hospital regardless of
prior cesareans in which the actual docu- how the labor progresses.
mented birthweight for the current preg-
nancy was more than 4000 g (8 pounds, 13
ounces). In the 4000-4499 g weight range,
Type "C" Cases
58% of the patients had vaginal births after Case C: The patient has no prior cesa-
cesarean, and the perinatal morbidity and rean and demands a cesarean
mortality was no higher than for the infants during labor (a primary cesa-
with birthweights less than 4000 g.28 These rean during labor).
data are reassuring, but are no panacea for
Because of the enormous physical and
the obstetrician's dilemma. If a woman with
emotional stresses of labor, type C cases are
a prior cesarean requests a repeat operation
perhaps the most common of all the cesa-
but is strongly encouraged to have a trial
rean decision conflicts. However, it should
of labor and then goes on to experience a
emphasized that only cases that involve
traumatic birth of a 4700-g infant, litigation
little or no medical indication for cesarean
is likely to occur. On the other hand, if a
are discussed. For example, assume a
physician agrees to operate in any case in
woman in labor has a fetus with a suspicious
which the baby appears to be a little large,
fetal heartrate monitor pattern. Further
many unnecessary cesareans will be per-
assume that her cervical dilatation has not
formed for normal-sized infants.
changed from 6 cm despite several hours of
Another example of a B2 type of case is a
adequate oxytocin augmentation. If this
woman who had a prior cesarean performed
woman were to request or even demand
for failure to progress after a long and dif-
a cesarean, her doctor would almost cer-
ficult labor. Although studies have shown
tainly agree. There would be no conflict, the
that many of these patients can go on to
interaction would rank near the middle
have subsequent normal vaginal births, this
of the cesarean decision spectrum, and
outcome can certainly not be guaranteed. If
everyone would be happy. Unfortunately,
such a patient demands an elective repeat
things rarely are so simple in the practice of
cesarean, should her request be denied? On
obstetrics.
the other side of the coin, it must be remem-
bered that the single most common reason
Case C1: The patient has no prior cesa-
for primary cesarean in the United States is
rean and demands a cesarean
failure to progress, or dystocia. If all these
during labor although there is
women had elective repeat operations, there
no medical indication (a pri-
would be few VBAC candidates and overall
mary cesarean during labor
cesarean rates would continue to rise.
with no medical indication).
While the B I-type cases can strain the
doctor-patient relationship, the B2-type An obstetric intern quickly learns that
cases can destroy it. There are currently no many women demand cesarean delivery in
easy solutions. A management plan often the midst of labor. The obstetric intern also
used in cases in which there is a possible but learns that most of these women will go on
not definite medical indication for repeat to experience normal vaginal births. If a
cesarean is to recommend a trial of labor cesarean was performed every time an ex-
with the understanding and agreement that hausted woman screamed "Take my baby
cesarean will be performed "expeditiously" now!," the cesarean rate would probably be
if labor does not progress or if there is any more than 50%. Although some attorneys
15. The Patient Who Demands Cesarean Delivery 217

would argue that even during the most dif- they might agree to performing the cesa-
ficult moments of labor a woman has the rean. This was true even though only 1 % of
right to demand a cesarean operation, most respondents felt that stress incontinence
physicians would disagree. For example, a represented an actual medical indication for
survey of 112 obstetricians found that only cesarean. Similarly, 8% of respondents said
2% would agree to a laboring patient's re- that they would agree to a laboring patient's
quest for cesarean if she were at term with request for primary cesarean if she had a
no medical problems. 8 Interestingly, if it history of postpartum hemorrhage with a
was assumed that the same patient was previous birth. An additional 15% of respon-
married to a lawyer, the number of doctors dents said that they might agree to perform-
who said they would agree to perform a ing the cesarean in this situation. But only
cesarean rose to only 3%. To be sure, this 1 % of respondents thought that postpartum
was a survey and reflected only what indivi- hemorrhage experienced after a previous
dual physicians said they might do in hypo- vaginal birth represented an actual medical
thetical situations! In any case, the current indication for cesarean. Clearly, nonmedi-
state of affairs seems to be that the majority cal factors do influence the decision-making
of doctors would not agree to perform a pri- process.
mary cesarean section in the absence of
medical indications.
Type "D" Cases
Case C2: The patient has no prior cesa- Case D: The patient has had a prior
rean and demands a cesarean cesarean and demands a cesa-
during labor yet there is no rean during labor (a repeat
definite medical indication cesarean during labor).
(a primary cesarean during
Case Dl: The patient has had a prior
labor with possible medical
cesarean and demands a cesa-
indications).
rean during labor yet there is
no medical indication (a repeat
The survey of obstetricians mentioned
cesarean during labor with no
earlier (case Cl) also addressed the issue of
medical indications).
making the decision to perform a primary
cesarean during labor with possible medical This is a unique situation. At least in the
indications. 8 The method of the study can be United States, it is a common practice in
described briefly as follows. A survey re- many hosptials to perform a repeat cesarean
questing opinions about the decision to at any time during a trial of labor if the
perform a cesarean section in several hypo- patient requests it. This may explain why
thetical situations was sent to obstetricians some reports have been unable to reproduce
in three cities, and 112 of the doctors res- the rate of vaginal birth after cesarean of
ponded (97 university staff and 15 in private 75% or more found in large studies. Al-
practice). Four of the situations were clearly though most doctors would not even ponder
type Al cases (elective primary cesarean an affirmative response to such a request
with no medical indication). However, made by a woman with no prior cesarean
several scenarios involved the decision- (type Cl case), many will rush to the operat-
making process when there were possible ing room if she has had one. Most programs
(but not definite) indications for surgery. For with high rates of vaginal birth after cesa-
example, 7% of respondents said that they rean have adopted the policy of treating
would agree to a patient's request for pri- trial-of-Iabor patients essentially the same
mary cesarean if she had a history of stress as any other patients and using similar indi-
urinary incontinence since her last delivery. cations for primary and repeat cesarean.
An additional 12% of respondents said that However, in the survey of 112 obstetricians
218 B.L.Flamm

discussed here, if a woman was progressing quests that her trial of labor be terminated
well during a trial of labor after prior cesa- because a prior postpartum hemorrhage,
rean and changed her mind at 9 cm of dilata- how many doctors would refuse? In contrast
tion and requested a repeat operation, 30% to the Dl case where everything is going fine
of the physicians said they would definitely with the trial oflabor but the patient simply
comply and an additional 24% said they changes her mind and requests a cesarean,
might comply. This was true in spite of the this situation involves additional risk fac-
fact that only 4% believed that a medical tors. As another example, what if variable
indication for the repeat operation actually decelerations of the fetal heart rate are
existed. noted during a trial of labor and the con-
In another type of survey study addres- cerned patient (or her partner) demands an
sing the decision to choose trial of labor immediate repeat cesarean? In the majority
or repeat cesarean section, question- of cases, these decelerations are innocuous
naires were sent to women rather than but at times they could be an indication of
physicians. 29 Of those respondents who uterine rupture. What is the solution to
opted for repeat cesarean, about 25% listed these dilemmas? It may be a reasonable
concerns about the risks of labor as impor- policy to have a "lower threshold for inter-
tant in their decision-making process. Far vention" during trial of labor than for
more women listed such factors as conven- patients with no prior cesarean. This is
ience of timing the birth, knowing what to certainly not the same as saying that every
expect, and avoidance of the pain of labor. woman who changes her mind during a trial
Also, 73% of these patients reported that of labor should be rushed to the operating
their physicians either discouraged vaginal room.
birth after cesarean or had neutral at-
titudes. The authors concluded that the
physician has a central role in the decision- How the Cesarean Operation
making process. Differs from Other Surgical
Case D2: The patient has had a prior Procedures
cesarean and demands a cesa-
rean during labor yet there is Part of the reason why the cesarean deci-
no definite medical indication sion-making process can become so compli-
(a repeat cesarean during cated is that the cesarean operation differs
labor with possible medical somewhat from all other operations. In a
indications). strictly anatomic sense, cesarean involves
Everything said about the C2 type of cases cutting through and then repairing the
(primary cesarean during labor with pos- same or similar layers of tissue as those
sible medical indications) applies to the D2 encountered during many abdominal opera-
situations. However, when the additional tions. But when nonanatomic factors are
risk factor of a prior cesarean birth is added, considered, several differences become.
the scale tips toward repeat operation. For apparent. These differences may explain
example, if a laboring patient with no prior why doctor-patient conflicts may arise
cesarean requested a primary cesarean in obstetrics that would not occur in other
because she had a history of postpartum surgical fields.
hemorrhage with a previous birth, 8% of 1. There are very limited alternatives.
doctors surveyed said they would agree to When an individual presents with symp-
perform the operation even though only 1% toms or signs of illness, diagnostic testing
thought the prior event had any bearing on may lead to the conclusion that surgery
the current labor. 8 If a laboring patient with could be beneficial. In many cases there is
a history of prior cesarean operation re- often an alternative to surgery such as
15. The Patient Who Demands Cesarean Delivery 219

administration of appropriate medications band's opinion may be of little significance

and waiting to see if the symptoms improve to the surgeon. However, if a woman in labor
or resolve. With the cesarean operation has a fetal monitor pattern indicating pos-
there are generally no such alternatives. sible fetal distress and she is unclear as to
When a woman is in labor the baby must be whether or not she wants a cesarean, her
born and there are only two possible routes. husband's opinion can become quite impor-
The woman can not go home and come back tant. If he demands a cesarean delivery to
in a couple of weeks after she has had a protect the health of his unborn child, can
chance to ponder her options or try other his request be ignored?
treatment regimens.
2. The decision to operate is often made
on an emergent basis. With the exception of
procedures performed on trauma patients,
Conclusions
the vast majority of operations that take
place in most surgical disciplines are sche- 1. Approximately 1 million cesareans are
duled elective procedures. In contrast, the performed in the United States each year.
decision to perform a cesarean operation It has been claimed that 500,000 of these
(except for elective repeat procedures) is operations are unnecessary.30 While most
often made only minutes before the surgery experts would argue that this estimate is
actually begins. vastly inflated, many would agree that cesa-
3. The patient is often under great physi- reans are indeed beIng performed that are
cal and emotional stress. No surgeon would not medically indicated.
consider jogging along with a patient during 2. The decision to perform a cesarean may
a marathon to discuss the potential risks involve some degree of disagreement or con-
versus benefits of a proposed major opera- flict between a patient and her physician.
tion. Yet this is analogous to the situation Although media attention has focused on
that commonly occurs when one obtains the rare cases where a patient refuses a medi-
consent of a woman in the midst of active cally indicated cesarean, the opposite situa-
labor. tion, where a patient demands a medically
4. The health of two patients must be unnecessary cesarean, occurs much more
considered simultaneously. Yet another dif- frequently. Yet the "patient-demanded
ference between the cesarean operation and cesarean" has been virtually ignored in both
all other operations is the fact that two pa- .medical literature and media coverage of
tients are always involved. An individual rising cesarean rates.
who might be quite comfortable considering 3. The entire concept of a chapter devoted
risks versus benefits involving their own to "the patient who demands a cesarean"
health may be far less comfortable handling may infuriate members of the natural child-
decisions that also affect the health of their birth movement along with some childbirth
unborn child. In addition to the patient's educators and midwives. The idea that large
opinions about the welfare of her fetus, the numbers of women are requesting or de-
actual rights of the unborn child also must manding cesareans may seem absurd to
be considered. These fetal rights and the them. However, these individuals may not
issue of maternal versus fetal beneficence realize that they are dealing with a select
have been extensively debated by specialists portion of the population. Women who seek
in medical ethics. The opinion of the unborn out midwifery care may do so because they
child's father may also have an impact on strongly desire natural childbirth. Women
the decision as to whether or not a cesarean who desire cesareans do not contact mid-
is performed. For example, if a woman has wives. Women who are upset after cesarean
gallstones and is unclear as to whether or delivery may contact cesarean-related sup-
not she wants a cholecystectomy, her hus- port groups to ventilate negative feelings.
220 B.L. Flamm

Women who are pleased with their opera- in the legal field, that a woman has the right
tions do not contact support groups. to request or demand the operation at any
4. Although this chapter focused on pa- time, raises ominous implications. This was
tients who demand or strongly request discussed under case AI.
cesareans, all the eight cases discussed 8. The cesarean operation differs from
could also be applied to a discussion of other surgical procedures in several impor-
patients who prefer not to have a cesarean tant ways. These differences may explain
(left side of the cesarean decision-making why the cesarean decision-making process
spectrum), for example, the woman who is can become quite complicated.
very interested in natural birth but who is
told by her doctor early in her pregnancy
that the baby "looks too big." It must be References
acknowledged that, while cases of court- 1. Kolder VE, Gallagher J, Parsons MT. Court
ordered cesareans are rare, situations where ordered obstetrical interventions. N Engl J
women are subtly urged to accept a cesarean Med 1987;316:1192-1197.
occur frequently. 2. Elkins TE, Brown D, Barclay M, Anderson
5. This chapter only considers cases in HF. Maternal-fetal conflict: a study ofphysi-
which there is some degree of conflict in the cian concerns in court-ordered cesarean sec-
cesarean decision-making process. In most tions. J Clin Ethics 1990;1:316-319.
cases the patient and her doctor agree on 3. Curran WJ. Court-ordered cesarean sections
the decision to perform, or not to perform, receive judicial defeat. N Engl J Med 1991;
324:272-273.
a cesarean. The fact that a patient and
4. Elkins TE, Andersen HF, Barclay M. Court-
her doctor agree on the decision to per- ordered cesarean section: an analysis of ethi-
form a cesarean does not necessarily imply cal concerns in compelling cases. Am J
that the procedure is medically indicated. Obstet Gynecol 1989;161:150-154.
The agreement does imply that no conflict 5. Kluge EW. When cesarean section opera-
exists. tions imposed by a court are justified. J Med
6. Conflicts that occur before labor allow Ethics 1988;14:206-211.
time for many options to be considered (see 6. Cunningham GF, MacDonald PC, Gant NF.
case A2), whereas conflicts which occur dur- Williams obstetrics, 18th Ed. Norwalk, Con-
ing labor allow for few options. This raises necticut: Appleton & Lange, 1989:445.
the question of whether the "cesarean issue" 7. Ryding EL. Investigation of 33 women who
demanded a cesarean section for personal
should be discussed during routine prenatal
reasons. Acta Obstet Gynecol Scand 1993;
care. Contemporary prenatal care involves 72:280-285.
the discussion of a plethora of subjects, 8. Johnson SR, Elkins TE, Strong C, Phelan JP.
many of which will have little bearing on Obstetric decision-making: responses to pa-
the pregnancy outcome. Whether the pa- tients who request cesarean delivery. Obstet
tient should take one or two prenatal iron GynecoI1986;67:847-850.
tablets a day is probably of far less signi- 9. Feldman G, Freiman J. Prophylactic cesa-
ficance than whether the patient will end rean section at term. N Engl J Med 1985;
her pregnancy on an operating table. With 312:1264-1267.
one of every four women currently being 10. Brodsky A. Letter to the editor: prophylactic
delivered by cesarean in the United States, a cesarean section at term. N Engl J Med
1985;313:753.
prenatal discussion about why a cesarean
11. Flamm B, Lim 0, Jones C, Fallon D,
operation will or will not be recommended Newman L, Mantis J. Vaginal birth after
would be far from esoteric. cesarean section: results of a multicenter
7. The issue of who actually decides study. Am J Obstet Gynecol 1988;158:
whether a cesarean will be performed needs 1079-1084.
further discussion in the ethical, legal, and 12. Flamm B, Neuman L, Fallon D, Thomas S,
medical literature. The prevailing opinion Yoshida M. Vaginal birth after cesarean sec-
15. The Patient Who Demands Cesarean Delivery 221

tion: results of a five-year multicenter study. 22. Ong HC, Sen DK. Clinical estimation offetal
Obstet GynecoI1990;76:750-754. weight. Am J Obstet Gynecol 1972;112:877-
13. Phelan J, Clark S, Diaz F, Paul R. Vaginal 880.
birth after cesarean section. Am J Obstet 23. Simon NV, Levisky JS, Shearer DM, O'Lear
Gynecol 1987;157:1510-1515. MS, Flood JT. Influence of fetal growth pat-
14. Smolowe J. Waiting for the verdicts. Time terns on sonographic estimation of featl
Magazine. 1993;142:26:48-49 (December 20). weight. JCU 1987;15:376-383.
15. ACOG. Guidelines for vaginal delivery after 24. Miller JM, Kissling GA, Brown HL, Gabert
a previous cesarean birth. Committee opin- HA. Estimated fetal weight: applicability to
ion number 64. Washington, DC: American small and large-for-gestational-age fetus.
College of Obstetricians and Gynecologists, JCU 1988;16:95-97.
1988. 25. Miller JM, Brown HL, Khawli OF, Pastorek
16. Kline J, Arias F. Analysis of factors deter- JG, Gabert HA. Ultrasonographic identifi-
mining the selection of repeated cesarean cation of the macrosomic fetus. Am J Obstet
section or trial of labor in patients with his- Gynecol 1988;159:1110-1114.
tories of prior cesarean delivery. J Reprod 26. Benacerraf BR, Gelman R, Frigoletto FD.
Med 1993;38:289-292. Sonographically estimated fetal weights:
17. Flamm B, Goings J, Liu Y, Wolde-Tsadik G. accuracy and limitation. Am J Obstet
Elective repeat cesarean section versus trial Gynecol 1988;159:1118-1121.
of labor: a prospective multicenter study. 27. Deter RL, Hadlock FP. Use of ultrasound in
Obstet Gynecol 1994;83(6):1-6. the detection of macrosomia: a review. JCU
18. Joseph G, Stedman C, Robichaux A. Vaginal 1985;13:519-524.
birth after cesarean section: the impact of 28. Flamm B, Goings J. Vaginal birth after cesa-
patient resistance to a trial of labor. Am J rean section: is suspected fetal macrosomia
Obstet GynecoI1991;164(6):1441-1447. a contraindication? Obstet Gynecol 1989;
19. Abitbol M, Castillo I, Taylor U, Rochelson B, 74:694-697.
Shmoys S, Monheit A. Vaginal birth after 29. Kirk E, Doyle K, Leigh J, Garrard M.
cesarean section: the patient's point of view. Vaginal birth after cesarean or repeat cesa-
Am Fam Physician 1993;47:129-134. rean section: medical risks or social reali-
20. Lehman B. Woman wins $1.53 million suit ties? Am J Obstet Gynecol 1990;162(6):
on unwanted cesarean. The Boston Globe 1398-1405.
Newspaper, June 16, 1993 (Metro section), 30. Silver L, Wolfe SM. Unnecessary cesarean
pI. sections: How to cure a national epidemic.
21. Laska LL. Malpractice casebook. OBG Washington, DC: Health Research Group,
Management 1993 December: 49-50. 1989.
16
The Impact of Midwifery Care,
Childbirth Preparation, and Labor
Support on Cesarean Section Rates
JANICE R. GOINGS

The last few years have witnessed the develop- mortality did decrease dramatically, but
ment of radical surgical obstetrics and brilliant much of this decrease can be attributed
the results have been. Long lists of Cesarean to nonmedical factors, such as improved
sections have been published with little if any nutrition and sanitation. Further decreases
mortality, and the indications for the operation resulted from medical, not surgical, advances
have been extended .... The question before us is:
such as the availability oftype-specific blood
Are we in our enthusiasm over' radical obstetric
surgery neglecting the fundamentals of obstetrics;
transfusions, antibiotics, and the use of
the routine precautionary methods which may aseptic techniques. 8
make the resort to radical obstetric surgery If, in fact, we believe that cesarean section
unnecessary?1 is overutilized, how can we intervene to
Edwin B. Cragin lower this rate? Should we employ the active
management model? Will more liberal use of
In 1916, when Dr. Cragin published his external cephalic versions and trial of labor
paper, which was titled "Conservatism in after prior cesarean have sufficient impact?
Obstetrics," the cesarean section rate was Is the diagnosis of fetal distress made too
less than 1%.2 Three-quarters of a century frequently?
later, it has risen dramatically and has, The vast majority of literature devoted to
through the intervening years, been the this topic analyzes outcomes of the medical
subject of much debate. In 1965 the rate was model of care. In this model, the laboring
only 4.5%.3 By 1980 the rate had risen so woman is seen as a patient and the process
high that the National Institutes of Health oflabor as abnormal until proven otherwise. 9
convened a task force to study the topic. But other models of caring for pregnant
When this task force issued its report in women also have the potential for impacting
1981, the national rate was 17.6%.4 It has the cesarean rate. This chapter explores
since escalated to 23.5%,5 an increase of these models, which view pregnancy as a
33.5% in 10 years. normal physiologic state, one that is only
Almost 80 years later, Dr. Cragin's ques- pathologic in retrospect. Included in this
tion remains germane. Some have argued discussion are the midwifery model of care,
that the current cesarean rate has resulted childbirth education, and labor support.
in healthier mothers and babies. 6 However,
many industrialized countries with cesarean
rates much lower than ours also have much Midwifery
lower perinatal mortality rates than we
do. 7 During the first half of this century, The midwifery model of care is common in
maternal and neonatal morbidity and industrialized European countries. One of

223
224 J.R. Goings

the best known and well-studied examples (CNMs) were matched case by case with 29
is at the National Maternity Hospital in women attended by physicians. Patients
Dublin. There, midwives are responsible for included in each group were low risk and
the management of all laboring women, were delivered in a tertiary-care hospital.
including private patients who are delivered Results showed no differences in complication
by the midwives with their personal phys- rates, but very different care practices by
ician in attendance. Midwives are closely the two provider groups. Intravenous fluids
involved with the education and training of were administered to 38% of the CNM
resident physicians. The active manage- patients, compared to 72% of the physician
ment of labor, one of the components of (MD) patients. Electronic fetal monitoring
this system, is partially credited with the was used 100% of the time by physicians,
impressive outcomes of care there, but compared to 34% of the time by CNMs.
the inclusion of midwives in their mater- Oxytocin was administered to 56% of the
nity service is, undoubtedly, an important MD patients, but only 22% of the CNMs.
factor. 10 Amniotomy was performed more often by
In this country midwives consist of two physicians than midwives. Pain-relieving
groups of practitioners: nurse-midwives who medication was administered less frequently
receive certification through the American to patients cared for by CNMs. The use of
College of Nurse-Midwives and who may episiotomy was much higher in the physician
legally practice in all 50 states; and a second group. Two of the women in the midwife
group consisting of licensed midwives, lay group sustained third-degree lacerations,
midwives, and foreign-trained midwives, compared to 3 third- and 3 fourth-degree
who have varying degrees of education and lacerations in the physician group. More
training and, depending on the state in women in the midwife group planned to
which they practice, mayor may not have breast-feed their infants at discharge.
legal standing for the practice of midwifery. This study was limited by its small sample
The philosophy of care given by the mid- size and retrospective nature; women were
wife differs from that of the medical prac- not randomly assigned to providers, a factor
titioner, although both have a common goal, that might have accounted for some of the
that each birth end with a healthy mother findings. With patients self-selecting for
and baby. Midwives have historically been CNM or MD care, the differing patterns
viewed as noninterventionists in the birth of care might have been a reflection of
process, whereas physicians have been seen dissimilar needs in each groUp.15
as much more likely to intervene, even in a Another study addressing the philosophic
normal, low-risk birth. Midwifery philosophy difference in pregnancy and childbirth care
views pregnancy and childbirth as a normal given by physicians and CNMs was conducted
physiologic female function, to be considered at the School of Nursing at the University of
pathologic only in retrospect. The practice of Michigan at Ann Arbor. A sample of 15
midwifery has been described as a "com- obstetricians and seven CNMs in a private
bination of science, craft, art, talent, know- service within a large tertiary-care medical
ledge and experience". 11 center completed questionnaires on various
The safety, quality, cost-effectiveness, aspects of their philosophy of care. These
and patient acceptance of midwifery care questionnaires were constructed by clinical
in a variety of settings have been well experts familiar with the functioning of both
documented. 12- 14 But just how does this obstetricians and nurse-midwives. Self-
care differ from that provided by physicians? reports of amount of clinic time spent with
A case-comparison retrospective review patients were randomly verified by auditing
was conducted in 1985 in an attempt to provider office schedules.
answer this question. Twenty-nine laboring Results of this study showed consistent
women attended by certified nurse-midwives differences between the CNMs and MDs.
16. Midwifery Care, Childbirth Preparation, and Labor Support and Cesarean Section Rates 225

Midwives spent significantly more time on but also siblings and other family members in
initial and return prenatal visits and were the birth process. Physicians and nurses who
more likely to do teaching in the areas of work side by side with midwives in a variety of
nutrition, infant feeding, exercise, parenting, settings are familiar with such midwifery
and minor maternal illnesses than were techniques as ambulation or nipple massage
physicians. Midwives reported that they as an alternative to the administration
personally did almost all the teaching, while of oxytocin to stimulate labor; employing
physicians relied on ancillary staff members, massage and water therapy as alternatives to
such as nurses. the use of medication and regional anes-
While there were differences in the two thesia; encouraging the intake oforal fluids in
groups in their rating of the importance of labor instead of routinely administering
management of intrapartum care, such as IV solutions; the avoidance of routine episio-
the use of pitocin, intravenous (IV) fluids, tomies, and the use of alternative laboring
continuous electronic fetal monitoring and birthing positions.
(EFM), use of lithotomy position, etc., only An important question arises when con-
the importance of ambulation in labor sidering these midwifery interventions.
reached statistical significance. The authors Does this philosophic divergence and dif-
noted that differences which did not reach fering practice style between medicine and
statistical significance were, nonetheless, all midwifery result in different outcomes for
in the expected directions, with CNMs patients, especially when considering their
placing more importance on management impact on cesarean section? If midwifery
techniques such as not using IVs, not limiting care does lead to a lower cesarean section
oral fluids in labor, etc. rate, what components of midwifery care are
Study limitations included the small responsible?
sample size and the volunteer nature of the
study participants. In addition, the non-
participation of five of the physicians in the
Midwifery Care and Cesarean Rates
group who failed to return their question- One of the largest studies to address this
naires might have influenced the outcomes, issue, conducted in the Netherlands from
because there could have been differences 1969 to 1983, is known as the Wormerveer
in those choosing to participate and those study. It was a prospective study of 8480
choosing not to participate. 16 women who were booked for delivery either
These two studies found some significant at home, in a small maternity unit, or in
differences in the philosophy of and care given hospital, with deliveries under the supervi-
by midwives and physicians. Other studies sion of a midwife. After exclusion for risk
have affirmed these differences, including factors, spontaneous abortions, ectopic preg-
less instrumentation used in deliveries,17,ls nancies, and those lost to follow-up, the
lower rates of epidural anesthesia and episio- study group consisted of 4000 nulliparous
tomy,19 and significantly more third- and and 3980 multiparous women.
fourth-degree lacerations III physician The majority of births occurred at home,
deliveries. IS in the maternity unit, or in the hospital
Although small and limited in design, the under the supervision of the midwives
foregoing studies basically describe much of (74.3%); 17.3% births occurred in a hospital
the midwifery philosophy. Through such in- after referral of the woman during preg-
terventions as health education, midwives nancy, and the remainder (7.9%) occurred
encourage women to accept responsibility for after referral during delivery.
their own care. They have long embraced an The women were divided into two groups:
opposition to routine interventions in child- the "selected group" were low-risk women
birth and have strongly encouraged family- selected for labor and delivery under care
centered care, to include not only husbands of the midwife, and the other group con-
226 J.R. Goings

sisted of women referred to obstetricians episodic, were established. In the com-

because of complications during pregnancy prehensive model, patients were assigned
or childbirth. to the midwives in the prenatal period,
The overall cesarean section rate of 1.4% comanaged with physicians if complications
for the total group was very low, but was arose, or transferred to medical manage-
even lower (0.4%) for the selected, low-risk ment if complications arose that required
group cared for exclusively by the midwives. ongoing medical management. Episodic care
This was in comparison to a national rate of applied to patients who had not been cared
4.2% in 1969 and 9.7% in 1980. The perinatal for prenatally by the midwives but who
mortality rate was low, especially, as anti- received such care in the intrapartum period.
cipated, in the low-risk group. Of the 6671 patients who received episodic
The authors concluded that selection of care (low-risk, screened in labor), only 0.2%
pregnant women into high- and low-risk required cesarean section. For the remaining
groups is possible and felt that within the 4095 patients in the comprehensive group,
scope of the Dutch system of obstetric care it the cesarean rate was 8.2%.22
is possible to achieve very good results with In 1981 a study was conducted retro-
midwifery care for selected women. 20 spectively on the first 1000 births at the
In a prospective evaluation study of the midwifery service of the San Francisco
effectiveness of nurse-midwives conducted General Hospital (California). The popula-
at the University of Mississippi from October tion was screened to exclude very high risk
1972 to April 1973, low-risk maternity women but included some with risk factors
patients were randomly allocated to the such as adolescence and advanced maternal
nurse-midwifery or house staff service; the age. The overall cesarean section rate was
study was not blinded. There were 298 9.0% for the study period between 1975 and
patients in the midwife group and 140 1979. It was noted that the hospital rate
patients in the house staff group. Although was 12.6% in 1977 and 12.2% in 1978.
differences were observed between the two The perinatal mortality rate of 9 in 1000
groups for other parameters, the cesarean compared favorably to 13 to 1000 for the
section rate for patients cared for by the hospita1. 23
midwives was 3.7% as compared to 4.3% for Stewart and Clark24 described the out-
the resident staff, a number that did not comes of patients admitted to an in-hospital
reach statistical significanceP birthing center in Douglas County, Georgia.
In a report on the development of the Their experience spanned 5 years from 1976
midwifery service that was initiated in 1973 to 1981, and included 2050 births in a private
at the Hennepin County Medical Center, CNM-MD practice. Patients were accepted
Hewitt analyzed the outcomes of their into the practice regardless of risk factors;
experience with more than 2400 births the only exclusions were patients admitted
occurring at an in-hospital birth center. in active labor before 32 weeks of gestation.
High-risk patients were comanaged with Eighty percent of the deliveries were per-
obstetric residents, both antepartum and formed by the nurse-midwives. The authors
intrapartum, and the delivery was done by reported a primary rate of 7.9% and a repeat
the midwife when appropriate. From July rate of 1.5%, which they believed to be low in
1973 to January 1982, the nurse-midwives comparison to national averages. 24
cared for 2418 women. Of these, 117 (4.8%) The Tucson N urse-Midwifery Service
required cesarean section. 21 reported the outcomes of patients to whom
Sharp and Lewis 22 reviewed the experience the CNMs provided prenatal, intrapartum,
of nurse-midwives at a tertiary-level, and postpartum care from 1978 to 1982.
university-affiliated hospital from 1973 to Their population was low risk, with physician
1982. In caring for this indigent population, consultation being obtained in the an-
two models of care, comprehensive and tepartum period by a physician at the
16. Midwifery Care, Childbirth Preparation, and Labor Support and Cesarean Section Rates 227

Arizona Health Sciences Center. Ifphysician The total cesarean rate for all 6313 patients
consultation was needed once patients were was 2.1%, with group I having a rate of
admitted in labor, it was provided by resident 8.2%, compared to group II, which had a rate
staff from the University of Arizona. Con- of 1.3%. The overall cesarean rate for the
sultation was required on almost 50% of hospital was noted to be 17% at the time of
their patients. The midwives delivered 711 publication of the study.28
(86%) of 823 patients on their service; 39 At the University of Vermont between
women required cesarean section, for an 1980 and 1985, nurse-midwives cared for
overall rate of 5%.25 1966 women and retrospectively reported on
The Nurse-Midwifery Service at Women's their outcomes. Their data included all
Hospital of the Los Angeles County/Uni- women followed prenatally by the midwives,
versity of Southern California Medical regardless of intrapartum complications.
Center has been in existence since 1974. In a Physician consultation was required for
1985 report on the role of the nurse-midwife 46.5% of the women in labor. The authors
in a large teaching hospital, although not reported a 10.4% primary cesarean section
specifically addressing the topic of cesarean rate and noted that this was somewhat less
section, they noted that only 2% of the than the hospital rate of 15%.29
patients cared for primarily by the CNMs Cavero reported on the outcomes of the
during the last 6 months of their study first 1000 births at the Nurse-Midwifery
required a cesarean, compared with an Group of Fresno (California), where the
primary rate of 8.7% for the hospital during population served was at mixed obstetric
that period. 26 risk; more than 82% of the patients served
A retrospective chart review was conducted were Medicaid recipients. Data were re-
on 175 consecutive patients who received trieved through a retrospective review of the
care from the Yale N urse-Midwifery Practice, charts of all women admitted to nurse-
a nurse-midwifery service run by the Yale midwifery care. The primary cesarean rate
School of Nursing students and faculty. was 6.7% with an overall rate of 7.7%, which
Their population was predominantly middle was less than half that for the hospital in
class, with insurance or self-paying, married, which the midwives practice and which was
and with a planned pregnancy. Women with lower than the county and state statistics. 3o
prior cesareans were accepted into their An interesting comparison of midwife
practice. Patients were followed by the versus physician management was made
midwives from their initial visit through the using data obtained in a retrospective
postpartum period. If complications arose review of pregnant adolescents attending a
requiring MD consultation or management, tertiary-care center clinic. Nurse-midwives
the midwives remained with the patients were on call for their patients from Monday
throughout their labors and births. The through Friday; on the weekend, when
nurse-midwives delivered 76.4% of their patients from the midwifery case load were
patients; all vaginal birth after cesarean admitted in labor, they were managed by
(VBAC) candidates had a successful trial resident physicians. Fifty-three patients
of labor. Their cesarean section rate was were cared for by the CNMs and 32 by the
5.1%.27 physicians. The population was essentially
Scupholme28 analyzed the outcomes of low risk, except that they were adolescents.
6313 births conducted by nurse-midwives at The cesarean section rate in the CNM
Jackson Memorial Medical Center (Florida) group was 1.9% and in the MD group it was
from 1977 to 1981. Group I, which consisted 3.1 %, a difference that was not statistically
of 690 women, received both prenatal care significant. 19
and intrapartum management by CNMs. In a retrospective descriptive study of
The remaining 5623 patients were assigned midwife-attended home births in Toronto
to midwifery care when admitted in labor. from January 1983 through July 1988,
228 J.R. Goings

Tyson described the outcomes of 1001 low- A study on the effect of midwifery inter-
risk patients delivered by 26 midwives. The ventions in lowering the cesarean section
total number of cesareans was 35 for a rate rate was conducted at Los Angeles County/
of 3.5%.31 University of Southern California (LAC/
In one of the largest studies ever conducted USC) Medical Center in 1991. Chambliss et
on out-of-hospital births, Rooks et a1. 32 al. 18 compared cesarean birthrates between
analyzed the outcomes of care in U.S. birth physician- and nurse-midwife-attended
centers from 1985 to 1987, a study that was births. Low-risk patients were randomly
undertaken to evaluate the safety of care in assigned to either MD or CNM management
such centers. While physicians, primarily with the birth attendant being unable to
obstetrician, provided some of the labor determine the patient's group assignment.
(9.9%) and delivery (16%) care, the majority Patients assigned to the midwifery service
of care was given by certified nurse- were cared for during their labor and delivery
midwives, student nurse-midwives, and experience on the in-hospital birth center,
midwives who were not CNMs. while those assigned to physician care were
A total of 11,814 women were admitted for placed in the regular labor and delivery
labor and delivery to 84 free-standing birth area. A total of 492 patients were included
centers. There were no maternal deaths, and in the study.
the overall perinatal mortality rate was 1.3 Outcomes related to cesarean section
of 1000 births, excluding lethal anomalies. showed a 0.4% cesarean rate on the MD
There was no delineation of cesarean section service, compared to 2.0% on the CNM
rate by type of practitioner, but overall only service, a difference that did not reach
4.4% of patients required cesarean section. statistical significance. The authors believed
The authors then compared their outcomes that the selection criteria used for triage
to those of five other studies of low-risk was responsible for this outcome, not mid-
hospital births. This included almost 40,000 wifery management per se. The authors
births from hospitals such as Parkland concluded that in selected low-risk popu-
Memorial in Dallas and the University of lations, a low cesarean rate can be obtained
Illinois during the period from 1969 to 1985. with either physician- or midwife-managed
Cesarean section rates for low-risk patients services. 18
at these hospitals ranged from a low of 8.3% In a Finnish study on the rate of inter-
to a high of 17.6%.32 ventions by midwife, researchers obtained
Most of the literature on outcomes of data from a population-based trial on iron
midwifery practice involves low-risk pa- prophylaxis during 1986-1987. A research
tients. However, at the North Central Bronx assistant abstracted data from 2135 births
Hospital in New York City, nurse-midwives involving 25 midwives. Patients admitted in
care for low- and high-risk patients with labor were routinely assigned to midwives
approximately 70% of the patients being at without selection, but if possible women who
risk or high risk. Haire and Elsberry33 were anticipated to have a difficult birth
reported on the outcomes of 3287 patients were assigned to the more experienced
delivered in 1988. Midwives were the midwives. The hospital in which the mid-
primary providers of antepartum, intra- wives practiced had a low overall cesarean
partum, and postpartum care for all low-risk rate of9.8% in 1987, compared to 15% for all
women and comanaged high-risk women of Finland. The combined cesarean rate for
with attending obstetricians. The CNMs patients managed by the midwives was
delivered 86.1% of these patients, including 8.4%, with division of cesareans into elective
38 vaginal breech deliveries performed with and emergency not possible with the data
the physician assisting or in attendance. retrieval system used. 34
Their cesarean rate was 11.8%, which was The Menlo Park Birth Place (California)
one half the national average. 33 registered 898 women during a 7-year
16. Midwifery Care, Childbirth Preparation, and Labor Support and Cesarean Section Rates 229

period ending in 1985. Of this number, 208 They further excluded pregnancies with 1 or
were not admitted because of medical and more of 52 medical or obstetric conditions
other factors, leaving a total of690 who were that they felt predisposed to cesarean de-
followed through delivery. The population livery. The final study group consisted
was predominantly Caucasian, married, and of 4607 low-risk women; 3551 were in the
well educated; 82% gave birth at the center, physician group and 1056 were in the nurse-
with 18% requiring intrapartum transport. midwife group.
Of the 127 women who required transfer, 18 The overall incidence of cesarean section
(3%) were delivered by cesarean section. 35 during their study period was 19.8%. After
In 1990 Schimmel et a1. 36 analyzed the the exclusions described, the cesarean sec-
outcomes of patients registered at a mid- tion rate for the low-risk cohort included in
wifery service established to serve low- their study was 11.8%. When analyzed by
income women who previously had no access type of provider, the nurse-midwives had a
to prenatal care. Their patients were pro- 9.75% cesarean rate and the physicians
spectively enrolled, and data were collected 12.3%. This difference was statistically
using a standardized form obtained from the significant.
American College of Nurse Midwives. Their Because the two groups differed signi-
population included high-risk women who ficantly by age, race, parity, and year of
were cared for collaboratively with attending delivery, multivariate analysis was used to
physicians. Women who were planning trials estimate the risk of cesarean while control-
of labor after prior cesarean were accepted ling for these potential confounding varia-
into their practice. Comparison data for the bles, as well as birthweight. Even after this
private obstetricians was obtained from the analysis, the differences remained statis-
hospital's delivery logbook. There were 496 tically significant.
patients in the midwifery group and 669 Because this was a low-risk cohort, all
patients in the physician group. The primary cesarean sections were performed for either
and repeat cesarean rates for the midwives labor abnormalities or fetal distress. In
were 3.7% and 3.0%, respectively, for a total analyzing the frequency of these two diag-
rate of 6.7%. For physicians, the primary noses by provider, the authors found that
rate was 11.4% with a total rate of 20.6%. physician patients were 50% more likely
The nulliparous cesarean rates for the to be diagnosed with fetal distress and
midwives and the physicians were 7.6% and 25% more likely to be diagnosed with labor
18.5%, respectively.36 abnormalities. 37
A study that found a statistically signi-
ficant difference in the cesarean rates in Discussion
low-risk patients cared for by midwives
compared to physicians was conducted Of the 21 studies analyzed, the majority
by the University of California at San were retrospective reviews of the outcomes
Francisco. U sing computerized records, of midwifery management in a variety of
researchers selected patients cared for by settings, both in and out of hospital (Table
resident physicians or nurse-midwives at 16.1). Of the 6 prospective studies, only 2
Moffitt Hospital, which is affiliated with the were randomized, and only 1 was blinded.
University. They then excluded all patients The study at the University of Mississippi,
who were not delivered at term (37-42 which was randomized but not blinded, con-
weeks), did not have at least five prenatal sisted of a group of 438 women. The differ-
visits, had a multiple gestation, were ad- ences in the cesarean section rates of CNMs
mitted in labor with any fetal presentation and resident physicians were not statis-
other than an cephalic or position other than tically significant. 17 The randomized,
occiput, and who had infants who were not blinded study at LAC/USC Medical Center
liveborn or who had congenital anomalies. of 492 women showed a similar result. 18 The
230 J.R. Goings

TABLE 16.I. Studies on the impact of midwifery care on cesarean section rate
Cesarean
Study Limitations Type section rate Comparison Population
Wormeveer Not randomized Prospective 1.4%, total 4.2%, 1969a Low risk
(1969-1983), No control group 0.4%, select 9.7%, 1980a
n = 7,980
University of Missippi Not blinded Prospective 3.7% 4.3%b Low risk
(1972-1973), Small sample Randomized
n = 438
Minnesota No control group Retrospective 4.8% Low risk
(1973-1981),
n = 2,418
Grady Memorial No control group Retrospective 0.2%C Mixed
(1973-1982), 8.2%d
n = 10,776
San Francisco General No control group Retrospective 9.0% 12.6%,1977 Mixed risk
Hospital 12.2%,1978
(1975-1979),
n = 1,005
Georgia (1977 -1982), No control group Retrospective 7.9%, primary Mixed risk
n = 2,050 1.5%, repeat
Tucson (1978-1982), No control group Retrospective 5.0% Low risk
n = 838
LACIUSC Cesarean section Retrospective 2.0% 8.7% Low risk
(1979-1980), rate incidental
n = 863 finding
Yale (1985), No control group Retrospective 5.1% Low risk
n = 175 Small sample Consecutive
Jackson Memorial No control group Retrospective 2.1% 17% Mixed risk
(1977 -1981),
n = 6,313

University of Vermont No control group Retrospective 10.4% 15% Low risk

(1980-1985),
n = 1,852
Fresno (1980-1985), No control group Retrospective 7.7% 15%a Mixed risk
n = 1,000
Yale (1984) Small sample Retrospective 1.9%, CNM 3.1%, MDb Low risk
n = 85 No random
assignment
Toronto (1983-1988), No control group Retrospective 3.5% Low risk
n = 1,001
National Birth Center No control group Prospective 4.4% 8.3%-17.6% Low risk
(1985-1987),
n = 11,814
Bronx Hospital (1988), No control group Retrospective 8.1%, primary 1/2 the High risk
n = 3,287 3.4%, repeat national rate
LACIUSC (1992), Small sample; Prospective 2.0% 0.4%b Low risk
n = 492 important Randomized
variables not Blinded
analyzed
16. Midwifery Care, Childbirth Preparation, and Labor Support and Cesarean Section Rates 231

TABLE 16.1. Continued

Cesarean
Study Limitations Type section rate Comparison Population
Finland (1988-1987), No random Prospective, 8.4% 9.8%, hospital Low risk
n = 2,135 assignment population 15%a
based
Menlo Park No control group Retrospective 3.0% Low risk
(1979-1985),
n = 690
Yolo County (1990), Data on control Prospective 6.7% 20.6% Mixed risk
n = 496 group
obtained from
delivery
logbook
Moffitt Hospital Self-selection of Retrospective 9.75% 12.3%" Low risk
(1981-1988), patients cohort
n = 4,607
Booth Medical Self-selection of Retrospective 4.3% 12.9%" Low risk
Centers, Thomas patients cohort
Jefferson University
Hospital
(1977 -1978),
n = 1,600

aNational rate.
b Not statistically significant.
cEpisodic care.
d Comprehensive care.
"Statistically significant.

study by Beal19 in which the same caseload pare with midwives in their incidence of
of women were cared for intrapartum by cesarean section? Goyert et al. 39 analyzed
nurse-midwives during the week and re- the method of delivery of 1533 women who
sidents on the weekend also showed no sta- were cared for by 11 board-certified obste-
tistical difference in cesarean rates between trician-gynecologists at a community hos-
the two types of practitioners. 19 pital in an affluent suburb of Detroit
Does this mean that midwifery care has (Michigan). Their study was conducted
no effect on the cesarean section rates? Are during a 12-month period from 1986 to 1987.
the low rates achieved by midwives simply Only very low risk private patients gave
a reflection of selection bias? Do either birth at this 320-bed hospital; patients with
physicians or midwives achieve low cesa- risk factors were transferred to tertiary-care
rean rates when caring for low-risk women? centers. The mean primary cesarean rate for
Risk status of patients is an important this low-risk population was 17.2%, with a
variable to consider, because it is known range of 9.6% to 31.8%.39
from a Canadian study that women with an In the randomized, blinded LAC/USC
absence of risk factors at the beginning study, Chambliss et al. 18 concluded that the
of their pregnancy have a cesarean rate that low cesarean rates achieved by the nurse-
is half the rate of the general hospital midwives resulted from selection bias.
population. 38 However, nurse-midwives are an integral
How do physicians caring for patients in part of the health care team at this hospital
low-risk settings that are not large tertiary- and have been so for approximately 20
care centers with residency programs com- years. When physicians are exposed to the
232 J.R. Goings

midwifery style of care, they will often adopt After complete analysis of their data
many midwifery management techniques. Baruffi et al. believed that the most likely
Could the results at this hospital be a reflec- explanation for the differences in the cesa-
tion of the philosophy of care there, rather rean rates related to "differences in ma-
than selection bias? The authors stated that ternity care practices that may arise from
physician management at this hospital different philosophies about the manage-
rarely includes ambulation, the presence of ment oflabor and delivery." They noted that
a support person for the laboring woman, or the model of care at Booth Maternity Center
the use of alternative laboring and birthing was Similar to that in many European coun-
positions, but does include a high rate of tries with low intervention rates and perin-
use of epidurals. However, none of these atar mortality rates lower than the U.S.
variables, which have the potential to im- rate.41
pact outcomes, were analyzed. And, as was As was discussed, it is often the contention
pointed out by Butler et al.,37 the small that midwives have low cesarean rates only
sample size of the USC study did not have because they care for low-risk patients. But
sufficient power to detect the 30% lower midwives also care for high-risk patients,
incidence found in their own study. usually in tertiary-care centers. The review
In a comparison of the differences in by Haire and Elsberry33 at the North Cen-
intrapartum care oflow-risk women, Reid et tral Bronx Hospital in 1988 was conducted
a1. 40 found lower intervention rates by in a midwifery service in which 70% of
family physicians when compared to obste- the women were at risk or high risk. Risk
tricians. They thought this reflected a more factors present in patients in their study
expectant style of practice by family physi- population included medical induction of
cians. Could this principle of expectant labor, diabetes (A-R) , pregnancy-induced
practice style be extrapolated to midwifery hypertension, abruptio placenta, placenta
care? Do midwives in general have more of a previa, maternal addiction, lack of prenatal
commitment to achieve a normal birth when care, etc. Even in this high-risk population,
possible than do physicians, especially when the midwives managed to achieve a cesa-
compared to obstetricians? rean rate of only 11.8% with a neonatal
This topic was addressed by Baruffi et mortality rate of 9.2 per 1000 live births for
a1. 41 in a 1990 study investigating the all birthweights and 3.7 per 1000 for infants
differences in cesarean section rates at over 1000 g.33 At Jackson Memorial, where
two institutions in Philadelphia. Booth nurse-midwives care for high-risk patients
Maternity Center (BMC) was staffed by intrapartum, Schupholme28 reported an
CNMs with a consulting physician present overall cesarean rate of 2.1% in the study
on the premises. Thomas Jefferson Univer- conducted there. Five-minute Apgars of 7 -
sity Hospital (TJUH) was staffed by resident 10 were present in 99.65% of their infants.
and attending physicians. A sample of Many of the studies reviewed were con-
796 women who received care at BMC was ducted in large teaching hospitals, where
randomly selected and then matched to 804 the patients, although ostensibly low risk,
women delivering at TJUH. After adjusting were from the lower socioeconomic strata, a
for demographic factors and medical condi- group that generally has all the medical
tions that might explain any differences, the problems associated with being disadvan-
authors found a significantly higher pri- taged such as poor nutrition, lack of con-
mary cesarean rate at TJUH. The rate at traception (resulting in closely spaced
TJUH was 2.5 fold higher than at BMC, conceptions), and poor compliance with
being 12.9 for the former and 4.3 for the prenatal care.
latter. Importantly, the higher cesarean rate Mayes et al. 15 pointed out that when
at the tertiary hospital was not associated patients have a choice between midwifery
with improved newborn outcomes. and physician care, such women may choose
16. Midwifery Care, Childbirth Preparation, and Labor Support and Cesarean Section Rates 233

providers according to their own internal tients with high-risk pregnancies and com-
needs, resulting in a self-selection bias. In plicated deliveries.
many of these busy, tertiary-care centers, Nurse-midwives also collaboratively care
however, patients frequently are assigned to for selected at-risk patients, such as those
midwifery or physician care solely on the with mild PIH, gestational diabetes, post-
basis of the demands of the labor and de- dates inductions, mild abruptio placenta,
livery unit, not patient preference. previous cesarean section, etc. At the River-
In all the studies reviewed, covering the side Medical Center, nurse-midwives an-
years from 1969 to 1982, cesarean sections nually deliver almost 70% of the patients.
were infrequently performed for patients For the 3-year period from 1991 to 1993, the
cared for by midwives. Rates ranged from CNMs have managed and comanaged 5812
0.2% to 11.8%, with the highest rate occur- women of mixed risk status. Their overall
ing in a high-risk population. Although cesarean section rate is 5.5%.44
many of the studies were methodically Many theories have been postulated as
flawed, one cannot ignore the universally to how or why midwives are successful in
low cesarean rates reported, which often lowering the incidence of operative inter-
have been achieved despite the fact these vention. Butler37 proposed two possible
midwives work on labor and delivery units mechanisms. The first is that care provided
in busy, tertiary, high-risk medical centers. by the midwife empowers women, enabling
In a study on the effects of hospital en- them to cope more successfully with child-
vironment on the practice of low-risk obs- birth. Second, she explored the possibility
tetrics, Carroll et al. 42 found that caring for that the midwife provides a mechanism for
low-risk patients in a high-risk setting is delaying the decision to interfere (physician
associated with higher intervention rates for consultation), thus lengthening the deci-
family physicians. And yet midwives have sion-making time, during which some
achieved low intervention rates in such women may make progress in their labor or
high-risk settings. be delivered. 37 Hemminki 34 believed that
In examining some of the most successful midwives may influence cesarean rates in
programs discussed, it becomes apparent three ways: that the manner in which they
that these programs have one thing in com- provide care may influence the mother's at-
mon, that of a strong working relationship titude and her progress in labor; that the
between physicians and midwives. This frequency and style of consultation may
team approach to obstetric practice is a influence the physician's decision regarding
model that has been utilized in a variety the need for cesarean section; and that how
of settings to achieve good maternal and actively the midwife intervenes in each
neonatal outcomes. One such setting in labor (such as labor augmentation, analge-
Kaiser-Permanente, the largest and oldest sia) may influence its outcome.
health maintenance organization (HMO) in The possibility exists that other personnel
this country. In this HMO setting the mid- who elect to work in maternity care settings
wife-physician team approach has been used where midwives function may influence the
successfully since its inception in the South- outcomes of laboring women. Nurses who
ern California region in 1980. 14,43 Nurse- support the midwifery philosophy of care
midwives and nurse-practitioners are in- frequently will choose to work in hospitals
volved in the antepartum care of low- and and birth centers with midwives. The type of
high-risk women. At several of the medical supportive care that they provide in labor
centers in the region, all laboring women may additionally impact cesarean section
with low-risk pregnancies are managed rates.
independently by nurse-midwives with in- The available evidence seems to suggest
house physician consultation available. that the midwifery model of care can lower
Physicians are, thus, freed to care for pa- cesarean section rates. Utilization of cesa-
234 J.R. Goings

rean section is decreased in maternity Childbirth Education

services in which midwives provide intra-
partum care in industrialized countries Historical Perspective
around the world. Large randomized trials
are needed to assess the specific ways in There is evidence to show that attempts to
which midwives achieve their outcomes. An manage the pain of childbirth date back to
interesting finding of the Finnish study was ancient times. Recently discovered evidence
the wide variation in cesarean rates among in Jerusalem from the fourth century A.D.
their midwives. While the overall cesarean indicates that marijuana was used to streng-
rate was low for patients cared for by mid- then labor contractions and reduce the pain
wives, individual rates ranged from 0.0% to of labor. Skeletal remains in an undisturbed
17.6%.34 tomb were identified as those of a 14-year-
More investigation is needed to deter- old with a term fetus in the pelvic area. The
mine if the low cesarean rates of midwives young girl, who was found lying on her back
are achieved by means of their philosophy in an extended position, apparently in the
of care, types of care given, professional last stages of pregnancy or of giving birth,
standards, practice environments, or other had very contracted pelvic measurements.
factors that perhaps have yet to been Residue of burnt cannabis was present near
considered. the abdominal area. Researchers concluded
One final area that should be investigated that this drug was inhaled by the pregnant
in future research concerns physicians girl to possibly increase uterine contractions
who ally themselves with midwives. It is and/or relieve the pain of labor. They also
important to remember that a midwife may cite the use of cannabis during childbirth in
care for a laboring woman, but it is always a several nineteenth-century publications. 45
physician who decides to perform a cesarean In the nineteenth century, methods intro-
section. While the practice style of the mid- duced to reduce the pain of childbirth were
wife who is obtaining the consult may have surrounded by controversy. When choloro-
an effect on the decision, the ultimate judg- form was used in 1847 by Simpson in
ment to resort to surgical intervention lies Scotland to alleviate the pain of childbirth,
with the consultant. Midwives obtain physi- it was condemned by the clergy as being
cian consultation frequently,25,29 but rarely contrary to biblical teaching that women
is surgery needed for their patients. More should suffer in childbirth to atone for origi-
information is needed about this decision- nal sin. 46
making process. Additionally, more data on In 1848, in Boston, Dr. Walter Channing
the characteristics and motivations of these introduced the use of ether for childbirth,
physicians would be helpful in determining and encouraged other physicians to try
the reasons for their support of midwives. it. Most were reluctant to do so and, thus,
Possibilities might include such factors as the use of anesthesia was infrequent in the
personal belief systems that value the birth latter half of the century. Various reasons
process or prior exposure to midwives and for this reluctance included uncertainty
their style of practice. The latter could have about the necessity and safety of anesthesia
profound implications for the manner in for most births, and moral objections, with
which students in medical school and resi- some doctors fearing that the drugs might
dency programs are educated about the excite women into erotic behavior. When
normal birth process. used, it was most often employed for first
In the final analysis, it is physicians and births or when internal versions or instru-
midwives together who can have a positive ments were needed. 46
impact on lowering the cesarean section rate With the movement to hospital birth in
in this country. the early part of the twentieth century, the
16. Midwifery Care, Childbirth Preparation, and Labor Support and Cesarean Section Rates 235

use of analgesia and anesthesia became Obstetrics (ASPOlLamaze), a national

commonplace, with women feminists and organization that promotes the principles of
suffragettes spearheading the efforts to Lamaze childbirth and family-centered
introduce "twilight sleep' into this country maternity care. Also founded in 1960,
from Germany. A hospital devoted to the the International Childbirth Education
utilization and promotion of this method of Association (ICEA) is a nonprofit organiza-
pain relief was established in Boston by a tion devoted to the support offamily-centered
female physician in 1914. It was called the maternity care and the encouragement
Twilight Sleep Maternity Hospital and of freedom of choice based on the know-
eventually was the site of 3000 births at- ledge of alternatives available to child-
tended by Dr. Eliza Ransom. Mrs. John bearing families.
Jacob Astor's picture even appeared in One of the first organizations in this
newspaper articles extolling the virtues of country to focus on the issue of cesarean
this method. 46 section was the Cesarean Support Education
However, concerns for such extensive use and Concern or C/SEC, which was founded
of sedation during hospital childbirth were in 1973. Its purpose was to increase aware-
being voiced in the 1920s and 1930s. Dr. ness of the rising cesarean section rates,
Grantly Dick-Read, a British physician, prevent the unnecessary utilization of the
introduced the concept of natural childbirth cesarean operation, and to humanize the
into this country with the publication of his cesarean childbirth experience. In 1982
classic text in 1933. He advocated teaching another cesarean awareness group, the
relaxation techniques to handle what he Cesarean Prevention Movement (CPM), was
termed the fear-tension - pain cycle that founded, and later became the International
occurred with laboring women. 47 Other Cesarean Awareness Network (ICAN).
methods of managing the pain of child- Their goals have been to provide a forum for
birth by relying on preparation during the exchange of ideas about birth, to provide
the prenatal period instead of the use of a childbirth support network for women, and
medications in labor were gradually in- to lower the rising cesarean section rate
troduced. These included the methods of through education.
Lamaze and Vellay in France,48 which were
introduced by MaIjorie Karmel in 195949
and Elizabeth Bing in 1967,50 and Bradley's
Limitations of Research Studies
husband-coached childbirth. 51 Many of There are many problems inherent In
these techniques of childbirth preparation attempting to assess the effectiveness of
remain popular today and are an accepted childbirth education programs. The most
part of prenatal care. Various other methods apparent is that of selection bias. The type of
of handling the pain of childbirth have been patient who chooses to attend childbirth
employed in prenatal classes in this country, classes can have a significant effect on
including such techniques as hypnosis, self- outcomes. The attitudinal factors of these
hypnosis, acupressure, water therapy, women may make it difficult to ascertain if
and biofeedback. outcomes result from the classes themselves
In the United States, there are various or occur because of psychologic and sociologic
organizations, many of which date back to differences. Women who choose to attend
the early years of the natural childbirth childbirth classes seem to differ from those
movement, devoted to the education and who choose not to attend such classes. Lumley
certification of childbirth educators. One and Brown52 found that nonattenders were
of the first of these, founded in 1960 by younger, less well educated, poorer, and more
Elizabeth Bing and MaIjorie Karmel, is the likely to be unmarried. They were also more
American Society for Psychoprophylaxis in likely to miss clinic appointments, had
236 J.R. Goings

higher rates of alcohol and tobacco use, andhave been concerned with the safety of
were less likely to initiate and maintain mothers and neonates and addressed such
breast-feeding. Davis and Morrone53 found issues as the amount of time required of the
that women who attended classes were of a physicians and nurses who worked with
prepared women; the effects on forcep
higher socioeconomic group, older, and better
educated than those who did not attend. rates, which were quite high in that era;
Leonard 54 found that women attending and the "success or failure" of the laboring
childbirth education classes were older, with
woman.
higher educational levels of both the women Goodrich and Thoms57 introduced the
and their spouses, were of a higher socioe- concept of childbirth preparation to Yale
conomic level, and had a higher incidence ofUniversity Hospital in the late 1940s. In
planned pregnancies. 54 their initial publication from 1948, they
Prenatal classes vary greatly in their looked at the outcomes of 156 women trained
content, with some focused on early preg- by their method, which was based on the
nancy and covering such topics as nutrition,work of Read. The patients they selected to
fetal growth and development, exercise, and receive childbirth education were from two
maternal body changes during pregnancy. groups: volunteers, who came to the clinic
Later classes can focus on any subject from requesting natural childbirth; and every
hospital policies to sibling preparation, third woman registering for prenatal care at
breast-feeding, and parenting. Even classes their clinic, providing they had completed
that are supposedly devoted to labor pre- at least 28 weeks of gestation and were
paration can differ greatly in the their "deliverable from below." These women then
approach. Standardization of class content, attended classes on anatomy/physiology,
relaxation and breathing techniques, the
as well as the focus of such educational efforts,
is an important factor. An additional variable
course of labor, and prenatal exercises.
that must be considered is the expertise andOutcomes were only evaluated for the
effectiveness of the childbirth educator. amount of analgesia/anesthesia required by
Also, problems arise when trying to quantifythe patient in labor, being characterized
instruction that is given; simply attending from excellent to poor. This study, while
class does not ensure internalization of thevery flawed by today's standards, was
material presented. important however because it showed that
childbirth education could successfully be
Goals of prospective parents vary greatly,
and certainly may differ from that of their taught to patients in a clinic service. 57
childbirth educators and caregivers. Hether- These authors expanded their study in
ington55 conducted a study in an inner-city 194958 by combining the earlier group with
hospital and found that motivating factors another study, reporting on a total of 546
for the majority of women were simply to be women with childbirth preparation. The
better prepared for delivery and to share additional women were recruited con-
secutively from their clinic population;
their childbirth experience with their pamer.
Charles et al. 56 related that most women apparently after their initial report, all
who take childbirth classes do so not for patients were given the opportunity to
obstetric reasons but to increase enjoyment attend classes. In analyzing their data, only
of their birth experience and to reduce and vaginal deliveries were considered, but they
control pain. did report an operative vaginal delivery rate
of 8.4%. They again concluded that natural
childbirth could be taught successfully in a
Early Studies teaching ward obstetric service, and that the
Reports examining the effectiveness of method had advantages for mother and child
childbirth preparation seem to have begun and was psychologically desirable for most
in the 1940s. Early investigators seemed to women. 58
16. Midwifery Care, Childbirth Preparation, and Labor Support and Cesarean Section Rates 237

Interestingly, they never even considered patients who were delivered by cesarean
the impact that this method might have on section from their data. They found that
cesarean sections. In their total series, it is women who received childbirth classes had
difficult to determine their cesarean section shorter labors, less instrumented deliveries,
rate, because they only reported on patients less perineal trauma, and a decreased use of
who delivered vaginally. The cesarean drugs and anesthetics in labor. 60
rate in this country in 1948-1949 was Davis and Morrone in 1962 53 studied 463
approximately 3%.2 One can only surmise patients receiving childbirth preparation and
that to the authors this was not an important labor support. The subjects were divided into
variable to analyze when looking at four groups: 320 received both preparation
the effects of childbirth preparation on labor and support, 35 received preparation but no
and delivery. support, 85 had support but no preparation,
In 1951, Thoms, this time in conjunction and 23 had no preparation or support. They
with Wyatt, analyzed the outcomes of 1000 concluded that the type of person selecting
deliveries under a childbirth training pro- childbirth preparation is more important
gram and in 1954 expanded this study with than the preparation itself. In considering
Karlovsky 59 to include 2000 consecutively this influence on delivery factors, they only
selected deliveries. Although interesting, noted an increased use of elective low forceps
these studies were quite flawed, because not in women who were not supported in labor,
all the women in the study group participated but failed to comment on cesarean sections.
in classes; 4% of the women included in the Beck and Ha1l61 in their 1978 review
analysis of their data were unregistered and article on natural childbirth examined
seen for the first time in labor. In addition, approximately a dozen studies from the
there was no control group for comparison. Of 1940s to 1977, including some of those dis-
the patients whom they evaluated in their cussed here. They believed that these studies
first series ofl 000, they noted a 2.3% cesarean tended to show such effects as decreased
rate for primigravidas and 1.8% for multi- pain perception, increased cooperativeness
paras. In the second series oflOOO, the rate for on the part of the mother, reduced use of
primigravidas was 0.8% and for multiparas analgesia/anesthesia, decreased rate of
2.4. The primary rate was 1.8%. There were 14 episiotomies and operative deliveries, etc.
repeat cesareans in the second group, but it is But they concluded that a variety of errors
not clear if these were failed trial-of-Iabor in study design, such as absence of control
(TOL) patients, unlikely given the obstetric groups, failure to provide a group undergoing
philosophy at the time. And if they were not attention-placebo treatment, failure to report
failed TOL patients, their inclusion in this statistical methods used to analyze data,
study of the effects of childbirth preparation and nonrandom assignment when there
on labor and delivery outcomes is puzzling. 59 were control groups confounded all the
One ofthe first studies to use a control group measured treatment outcomes. 61
when assessing the effects of childbirth
training was reported by Van Auken and
Tomlinson in 1953. 60 They made a compara-
Recent Research
tive study of 400 primigravidas in their It was not until the 1970s that researchers
practice in Troy, New York. One-half re- began to consider the impact of childbirth
ceived prenatal training consisting of a preparation on cesarean rates. Scott and
hospital tour and four classes conducted at Rose 62 from the University of Iowa retro-
their office; the other half of the group acted spectively analyzed the data from 129
as controls. It does not appear that their primiparous patients who completed Lamaze
subjects were randomly assigned, however. classes and delivered at the University
This might have shown some interesting Hospital in 1967 -1974. These patients were
comparisons, but they specifically excluded matched with a control group of 129 patients
238 J.R. Goings

of the same parity who had not received Sturrock and Johnson 65 reviewed the
preparation and were of the same age and records of 207 primigravid patients who
economic status as the study group. The attended childbirth education classes at
cesarean rate was 6.2% in the study group a military hospital from December 1986
and 4.7% in the controls, a finding that did through July 1987. Patients were divided
not reach statistical significance. into two groups: the treatment group, which
Charles et aI., in 1977,56 analyzed the attended two to four classes, and the com-
obstetric and psychologic outcomes of parison group, which attended one or no
patients who had received childbirth pre- prenatal classes. Family practice residents
paration in a large urban teaching hospital. delivered the majority of the patients. The
249 women who delivered at this hospital cesarean section rate in the group who
were interviewed 1-3 days postpartum, and attended classes was 38% compared with
information about obstetric complications 29% in the group who did not attend classes;
was obtained by reviewing their medical this did not reach statistical significance.
records. Of this group, 95 had received A population-based study comparing the
training in the Psychoprophylatic method of differences in women attending childbirth
childbirth education. It appears that there classes and those who did not attend such
were more cesareans in the prepared group, classes was conducted in Australia in 1989.
but this was probably related to their selec- Data were obtained through analysis of a
tion method, thus making it difficult to questionnaire mailed to a representative
interpret their data on cesarean sections. sample of women 8 months after they had
In a 1978 retrospective study of the given birth for the first time. The question-
maternal and fetal outcome of Lamaze- naire contained 66 pre coded questions about
prepared patients, Hughey and McElin63 such aspects as reproductive history, duration
analyzed the outcome of 500 consecutive of breast-feeding, sociodemographic data,
Lamaze deliveries at the Evanston Hospital and emotional well-being. The emergency
in Illinois and compared this group with 500 cesarean section rate for those women
controls, who were consecutively matched attending classes was not significantly dif-
for age, race, parity, and educational level. ferent from those who did not attend class
The authors reported a statistically sig- (9.8% versus 10.6%). Interestingly, the dif-
nificant decrease in the cesarean rate ference in women who chose elective cesarean
of the Lamaze group, 3% versus 12% in section was slightly higher in the attenders
the controls. While acknowledging that (11.0% versus 8.5%), but this also was not a
analyzing the effects of consistency of class significant difference. 52
attendance might have altered their out-
comes, the authors did not collect data on
this aspect. 63 Discussion
An Australian study conducted from 1981
to 1982 consisted of a random sample of 398 When the literature related to childbirth
low-risk women who gave birth in five preparation is reviewed, results are con-
Sydney teaching hospitals. Data were flicting. Some have thought that childbirth
obtained by questionnaire-based interviews preparation results in substantial benefits
3 weeks after birth and by abstracting and has positive outcomes on birth ex-
information from hospital charts. The women periences. 55 ,66 Others believe that at-
were divided into groups based on the tendance at such classes is not associated
number of hours of class attendance: none, with differences in emotional well-being,
low (1-12h), medium (13-19h), and high satisfaction with care received, or differences
(20 or more h). The overall cesarean rate in birth procedures. 52 ,53 All authors do,
was 5% and was not significantly related to however, seem to agree that the process is
the extent of antenatal preparation. 64 not harmful to mother or infant.
16. Midwifery Care, Childbirth Preparation, and Labor Support and Cesarean Section Rates 239

As Beck and Hall61 pointed out, to evaluate education on cesarean section, none were
the outcomes of childbirth classes, large randomized, prospective studies. The two
randomized trials are needed that include studies reviewed with matched controls,
three groups: a control group, which receives both of which were retrospective, revealed
no treatment; the experimental group; and a differing results; one showed a statistical
third group consisting of those who receive difference in the incidence of cesarean
placebo-attention. Shearer67 believes that section,63 and the other did not. 62 The
even this might not be adequate, because so retrospective review by Charles56 was
much ofintrapartum management, including designed in such a manner as to make
use ofEFM, labor induction, use of analgesia interpretation of data uncertain. In neither
and anesthesia, and cesarean section is based the retrospective reviews by Bennett64 and
on practice habits within hospitals and Sturrock65 nor the population-based study
communities. Intrapartum care, thus, con- by Lumley52 were there significant difference
forms mostly to physician peer practices and in cesarean rates between attenders and
not to patients' wishes, diluting whatever nonattenders of classes (Table 16.2).
effect is present from the classes. If the desired goal is to reduce the incidence
Perhaps we are attempting to measure of cesarean section through educating
the wrong outcomes when evaluating women and their partners, the first require-
the "success" or "failure" of prenatal classes. ment would be specific instruction in cesarean
As Enkin68 stated, "It is important to prevention techniques. Classes such as these
remember ... that these classes are an are already provided through the Inter-
educational rather than an obstetrical national Cesarean Awareness Network
intervention, and the outcomes used to (ICAN) in their Birthworks program.
evaluate them must reflect their objectives." Research is needed on the effectiveness
Certainly, prenatal and childbirth pre- of classes that are specifically focused
paration classes seem to fill a perceived need on reducing cesarean section rates by
on the part of expectant parents. While empowering women to cope with their labors.
concrete data are not available, estimates It is ironic that an educational movement
indicate that between 30% and 70% of all which began in response to the overuse of
expectant women in English-speaking sedating drugs during childbirth seems to
countries attend such classes. 69 Perhaps it have failed, somewhat, in its purpose. All
would be more relevant to evaluate the effect too frequently childbirth preparation classes
of classes on the parent's actual goals, rather in this country simply prepare women to
than on outcomes that health care providers accept hospital policies while reviewing the
perceive as important. types of analgesia and anesthesia available.
One of the initial aims of the childbirth While not negating the efforts of dedicated
education in this country was to help women childbirth educators or denying that child-
cope with their labors by using nonpharma- birth preparation is beneficial, it is, perhaps,
logic means, so it is not unexpected that too simplistic to believe that attending
researchers have paid relatively little sessions which review the process of labor,
attention to the impact of childbirth classes describe support and comfort measures for
on cesarean section rates. They have focused, the laboring woman, and discuss hospital
instead, on analyzing such factors as use procedures can significantly impact a
of pain medication, use of instrumented woman's ability to avoid a major surgical
deliveries, incidence of complications, length procedure.
of time in labor, incidence of episiotomy, Educational efforts that are directed more
use of forceps and vacuums, and patient specifically at cesarean prevention can
satisfaction with their birth experience. undoubtedly play a large role in reducing
Of the available research that addresses the number of cesarean sections performed
the question of potential impact of childbirth in this country. More effort is required on
240 J.R.Goings

TABLE 16.2. Comparison of cesarean section rates between women

who did or did not attend birth preparation classes
Cesarean
Study Type section rate Comparison
Scott Retrospective, 6.2% 4.7%
(1967-1974),62 matched controls
n = 359
Charles (1978),56 Retrospective, ? (increased)
n = 249 control group
Hughey (1978),63 Retrospective, 3% 12%a
n = 1000 control group
Bennett Random sample,
(1981-1982),64 questionnaire-
n = 398 based interview
Sturrock Retrospective 38% 29%
(1986-1987),65
n = 207
Lumley (1989),52 Population based 9.8% 10.6%
n = 292

a Statistically significant.

the part of childbirth educators to promote doulas, or companions whose sole respon-
this type of childbirth education. sibility is to support the laboring woman,
are increasingly being utilized in hospital
births. These doulas are not necessarily
Labor Support professionals, but function as an ombudsman
for women in labor and after birth. 70
Before this century, women were always That labor support might be an important
supported in labor by other women- variable in outcome studies was touched
midwives, female friends, and relatives. upon by Goodrich and Thoms57 as early
When hospital births became the norm, this as 1948, when they advocated constant
form of labor support all but vanished. Jf attention by an understanding nurse who
support was given, it was usually by the would help the laboring woman adapt her
bedside nurse who was providing care to the relaxation exercises. Van Auken and
laboring woman. Tomlinson60 in their 1953 study commented
An outgrowth of the childbirth education that "labor patients should be out of bed and
and to some extent the feminist movement with their husbands as much as possible
was an increased awareness that support during labor."
was an important element of care for women In 1980 a study was conducted in
in labor. Starting with the natural childbirth Guatemala by Sosa et al. 71 on the effects of a
movement in the 1930s and 1940s, first supportive labor companion. Hospital policy
husbands and then other family members did not allow the presence of husbands, other
and friends joined laboring women in family members, or friends in the labor or
hospital (and later birth) rooms across delivery rooms. Primigravid low-risk women
America. Today, it is unusual for a woman were randomly assigned to either an experi-
who is laboring in a hospital to do so alone. mental (labor support) or control (no labor
Midwives and nurses have traditionally support) group when admitted in active
supported laboring women. In addition, labor. Women in the experimental group
16. Midwifery Care, Childbirth Preparation, and Labor Support and Cesarean Section Rates 241

received constant support from an untrained it would seem important to consider the ef-
lay woman from admission to delivery. fects of their care on cesarean rates. In 1991
There was a 27% rate of cesareans in the Radin et al. 74 studied labor and delivery
control group, compared to 15% in nurses in a nonprofit, tertiary referral hospi-
the experimental, a rate that was stati- tal with approximately 6700 deliveries per
stically significant. 71 year. A staff of 45 registered nurses provided
Klaus and Kennell 72 expanded that study care for patients throughout their labor and
to include low-risk primigravid women at birth. These nurses were ranked according
the same hospital, using the same methods to the cesarean rates of their nulliparous,
described by Sosa et al., and reported the low-risk women who went into spontaneous
outcome in 1986. There were 249 women labor. Cesarean rates differed markedly be-
who labored alone and 168 who labored in tween nurses in the lowest quintile (4.9%)
the presence of a supportive female com- and the highest quintile (19%) and were
panion. Once again, there was a statistically not explained by various factors, such as
significant difference in the cesarean section maternal age, gravidity, childbirth prepara-
rates of the two groups, with the supported tion, insurance status, reliance on public
group having fewer cesareans than the assistance, physician who attended the birth,
unsupported group (7% versus 17%).72 labor augmentation, epidural anesthesia,
The two foregoing studies were criticized infant weight/gestational age, or dilatation
on the grounds that the care environment in when the nurse assumed care. The authors
which they were conducted in Guatemala concluded that the care given by nurses
was vastly different from that in this country. during labor has an important influence on
In an attempt to see if this type of care cesarean birthrates. 74
would have similar results in a hospital Hodnett and Osborn75 conducted a ran-
with modern obstetric practices, these domized trial on the effects on labor support
authors conducted a study at Jefferson on childbirth outcomes. Low-risk women
Davis Hospital in Houston, which is affiliated who were attending childbirth preparation
with Baylor College of Medicine. Hospital classes were enrolled in the study, and ran-
policy prohibited the presence of support domly assigned to either the experimental
persons during labor and delivery. Nulli- group, who received prenatal and intrapar-
parous low-risk women admitted in active tum support from a monitrice, or the control
labor who consented to participate in the group, who received prenatal and post-
study were assigned to either a doula sup- partum support from the same monitrice.
port group (n = 212) or an observed group (n The study's purpose was to evaluate the
= 204) in which an inconspicuous observer effect of labor support on women having
remained in the laboring woman's room. vaginal births, so those requiring cesareans
A control group (n = 204) was added to were excluded from the analysis of the data.
examine the potential supportive effects of The final sample consisted of 103 subjects.
the nonparticipant observer. Their findings The authors did comment that the cesarean
in relation to cesarean section rates were section rate in the experimental group
similar to those observed in Guatemala. was 17%, compared to 18% in the controls.
Of the patients in the supported group, 8% Results failed to show that continuous
required a cesarean, compared to 13% in the supportive care had any evident direct effect
observed group and 18% in the controls. on cesarean delivery rates. Obstetric prac-
It is interesting to note that the presence tices at the hospital in which the study
of someone who simply remains with the was conducted may have been an important
mother at all times (observed group) can factor in outcomes; there was a high inci-
lower the cesarean rate. 73 dence of a forcep deliveries (31 of 103),
Because nurses provide much of the care and epidurals, both continuous (89%) and
given to laboring women in this country, terminal. 75
242 J.R. Goings

Discussion tivities that are more highly valued and

which consume more of their time than
The foregoing studies show the significant labor support. Using work-sampling,
impact that labor support can have in lower- McNiven et al. 77 found that the proportion
ing cesarean section rates (Table 16.3). of time that nurses spent on supportive
However, support for many women laboring care on the day shift at a Toronto teaching
in hospitals is this country is often minimal. hospital was only 9.9%. And yet patients
Most commonly, it is the baby's father who view nurses who remain with them for longer
assumes the supporting role, a role for which periods of time as more helpful. 78
he may be ill equipped. Many fathers can Midwives have traditionally been seen as
function quite well in assuming an active a category of caregiver who remains con-
role in caring for their partners in labor; in stantly with the laboring woman. However,
fact, this is the basis of the Bredley method midwives in this country, who are working
of childbirth preparation. Other fathers find in increasing numbers in busy, tertiary
the experience distressing and overwhelm- medical centers, may fail to provide adequate
ing, and may, perhaps, contribute to the labor support. Not infrequently, a midwife
decision to perform a cesarean by their in such a setting may care for more than one
inability to cope with their partner's pain. patient at a time.
Other family members or friends of the Doulas are assuming an increasing role in
laboring woman may be involved in sup- supporting women during labor and after
portive care. The usual practice during birth. One organization offering certification
home or birth center births is to have several for these caregivers is Doulas of North
people present. As birth rooms have become America. It is estimated that there are 800-
more common in hospitals, there has been 1000 birth doulas and 500-1000 postpartum
an increase in the number of women who doulas in the United States. 79
choose to have more than one person present Thus, although it appears that labor
during their births. But here are some voices support decreases obstetric interventions,
of dissension over this practice. Odent76 lowers cost, and has a positive impact on
believes that this is disruptive to the laboring maternal and neonatal health,73 providing
woman and may, in fact, result in a long, such support does not seem to be a priority
difficult labor. in this country. This is not the case in other
Nurses have increasingly become involved industrialized countries with low cesarean
with the technologic aspects of birth, ac- section rates. At the National Maternity

TABLE 16.3. Effect of support during labor on cesarean rate

Cesarean section rate
Study Type Experimental Controls
Sosa (1980)71 Prospective- 15%; n = 20 27%a; n = 20
randomized
Klaus (1986)72 Prospective- 7%; n = 168 17%a; n = 249
randomized
Kennell (1991j13 Prospective- 8%; n = 212 13%, observed
randomized group; n = 204
18%, control
groupa; n = 204

Hodnett (1988)75 Retrospective 17%; n = 72 18%; n = 73

a Statistically significant.
16. Midwifery Care, Childbirth Preparation, and Labor Support and Cesarean Section Rates 243

Hospital in Dublin, whe.re the cesarean 10. Boylan P. Active management of labor:
section and perinatal mortality rates are results in Dublin, Houston, London, New
very low, laboring women are never left Brunswick, Singapore, and Valparaiso. Birth
alone. Student midwives or medical students 1989;16:122.
sit continuously with patients to provide 11. Swinnerton T. Traditional midwifery skills.
Nurs Times 1990;86:74-75.
emotional support. Such activity is con-
12. Levy B, Wilkinson F, Marine W. Reducing
sidered to be an important part of the educa- neonatal mortality rate with nurse-midwives.
tion and training of both midwives and Am J Obstet Gynecol 1971;109:50-58.
physicians. 1o 13. Reid M, Morris J. Perinatal care and cost
Ifwe are to lower the cesarean section rate effectiveness: changes in health expenditures
in this country, we need to reassess the following the establishment of a nurse-
practice, that is so common in many hospitals, midwife program. Med Care 1979;17:
of leaving the laboring woman alone with 491-500.
her significant other, who may be poorly 14. Bell KE, Mills, JI. Certified nurse-midwife
prepared to be her sole labor support. Nurses, effectiveness in the health maintenance
midwives, doulas, and physicians together organization obstetric team. Obstet Gynecol
1989;74:112-116.
must investigate ways in which continuous
15. Mayes F, Oakley D, Wranesh B, et al. A
support can be given to each and every retrospective comparison of Certified Nurse-
laboring woman in this country. Midwife and physician management of
low risk births. J Nurse-Midwifery 1987;
32:216-221.
16. Yankou D, Petersen B, Oajkley D, et al.
References Philosophy of care: a pilot study comparing
1. Cragin E. Conservatism in obstetrics. N Y Certified Nurse-Midwives and physicians. J
Med J 1916;CIV:1-3. Nurse-Midwifery 1993;38:159-164.
2. Levine W. Thirty-year survey of cesarean 17. Slome C, Wetherbee H, Daly M, et al. Ef-
section. Obstet Gynecol 1962;19:42-49. fectiveness of certified nurse-midwives: a
3. Marieskind H. Cesarean section in the United prospective evaluation study. Am J Obstet
States: has it changed since 1979? Birth Gynecol 1976;124:177-182.
1989;16:196-201. 18. Chambliss L, Daly C, Medearis A, et al. The
4. National Institute of Health: Consensus role of selection bias in comparing cesarean
Development Conference on Cesarean Child- birth rates between physician and midwifery
birth, September 1980, sponsored by the management. Obstet Gynecol 1992;80:
National Institute of Child Health and 161-165.
Human Development. Bethesda, maylands: 19. Beal M. Nurse-midwifery intrapartum
NIH publication no. 82-2067,1981. management. J Nurse-Midwifery 1984;29:
5. Taffel S, Placek P, Rosary C. U.s. Cesarean 13-17.
section rates 1990: an update. Birth 1992; 20. Van Alten D, Eskes M, Treffers PE. Midwifery
19:21-22. in the Netherlands. The Wormerveer study;
6. Williams RL, Hawes WE. Cesarean section, selection, mode of delivery, perinatal mor-
fetal monitoring, and perinatal mortality in tality and infant morbidity. Br J Obstet
Callifornia. Am J Public Health 1979;69: Gynecol 1989;96:656-662.
864-870. 21. Hewitt M, Hangsleben KL. Nurse-midwives
7. O'Driscoll K, Foley M. Correlation of a in a hospital birth center. J Nurse-Midwifery
decrease in perinatal mortality and increase 1981;26:21-29.
in cesarean section rates, Obstet Gynecol 22. Sharp ES, Lewis LE. A decade of nurse-
1984;63:205-208. midwifery practice in a tertiary university-
8. Stewart D, Stewart L. Compulsory hospi- affiliated hospital. J Nurse-Midwifery 1984;
talization: freedom of choice in childbirth?, 29:353-365.
Vol. II. Missouri: Napsac Reproductions, 23. Mann R. San Francisco General Hospital
1979. nurse-midwifery practice: the first thousand
9. Klein M. Active management oflabor: whose births. Am J Obstet Gynecol 1981;40:
agenda? Birth 1993;20:97-99. 676-682.
244 J.R. Goings

24. Stewart RB, Clark L. Nurse-midwifery 39. Goyert G, Bottoms S, Treadwell M, et al. The
practice in an in-hospital birthing center: physician factor in cesarean birth rates. N
2050 births. J Nurse-Midwifery 1982;27: Engl J Med 1989;320:706-709.
21-26. 40. Reid A, Carroll J, Ruderman J, et al. Dif-
25. SchreIer A. The Tucson Nurse-Midwifery ferences in intrapartum obstetric care
Service: the first four years. J Nurse-Mid- provided to women at low risk by family
wifery 1983;28:24-30. physicians and obstetricians. Can Med Assoc
26. Platt LD, Angelini DJ, Paul RH, et al. Nurse- J 1989;140:625-633.
midwifery in a large teaching hospital. 41. Baruffi G, Strobino D, Paine L. Investigation
Obstet Gynecol 1983;66:816-820. of institutional differences in primary ce-
27. Nichols C. The Yale Nurse-Midwifery sarean birth rates. J Nurse-Midwifery 1990;
practice: addressing the outcomes. J Nurse- 35:274-28l.
Midwifery 1983;30:159-165. 42. Carroll J, Reid A, Ruderman J, et al. The
28. Scupholme A. Nurse-midwives and phy- influence of the high-risk care environment
sicians: a team approach to obstetrical care on the practice of low-risk Obstetrics. Fam
in a perinatal center. J Nurse-Midwifery Med 1991;23:184-188l.
1982;27:21-27. 43. Goings JR. Success in a health maintenance
29. Keleher K, Mann L. Nurse-midwifery care in organization. In: Rooks J, Haas JE,eds.
an academic health center. J Obstet Gynecol Nurse-Midwifery in America. Washington,
Neonatal Nurs 1986;September/October: DC: The American College of Nurse-Midwives
369-372. Foundation, 1986:36-37.
30. Cavero C, Fullerton J, Bartlome J. Assess- 44. Unpublished data. nurse-midwifery service,
ment of the process and outcomes of the first Kaiser-Permanente, Riverside, California,
1000 births of a nurse-midwifery service. J 1993.
Nurse-Midwifery 1991;36:104-110. 45. Zias J, Stark H, Seligman J, et al. Early
31. Tyson H. Outcomes of 1001 midwife-attended medical use of cannabis. Nature 1993;
home births in Toronto, 1983-1988. Birth 363:215.
1991;18:14-19. 46. Wertz R, Wertz D. Lying-in: a history of
32. Rooks J, Weatherby N, Ernst E, et al. Out- childbirth in America. New York: Free Press,
comes of care in birth centers: the national 1977.
birth center study. N Engl J Med 1989;321: 47. Dick-Read G. Natural childbirth. London:
1804-1811. Heinemann, 1933.
33. Haire D, Elsberry CC. Maternity care and 48. Lamaze F. Painless childbirth. New York:
outcomes in a high-risk service: the North Pocket Books, 1972.
Central Bronx Hospital experience. Birth 49. Karmel M. Thank you, Dr. Lamaze. Philadel-
1991;18:33-37. phia: Lippincott, 1959.
34. Hemminki E, Kojo-Austin H, Malin M, et al. 50. Bing E. Six practical lessons for an easier
Variations in obstetric interventions by childbirth. New York: Grossop & Dunlap,
midwife. Scand J Caring Sci 1992;6:81-86. 1967.
35. Eakins P, O'Reilly W, May L, et al. Obstetric 51. Bradl(;ly R. Husband-coached childbirth. New
outcomes at the Birth Place in Menlo Park: York: Harper & Row, 1965.
the first seven years. Birth 1989;16:123-129. 52. Lumley J, Brown S. Attenders and non-
36. Schimmel L, Hogan P, Boehler B, et al. The attenders at childbirth education classes in
Yolo County Midwifery Service: a descriptive Australia: how do they and their births differ?
study of 496 singleton birth outcomes, 1990. Birth 1993;20:123-129.
J Nurse-Midwifery 1992;37:398-403. 53. Davis CD, Morrone FA. An objective evalua-
37. Butler J, Abrams B, Parker J, et al. Sup- tion of a prepared childbirth program. Am J
portive nurse-midwifery care is associated Obstet GynecoI1962;84:1196-120l.
with a reduced incidence of cesarean section. 54. Leonard RF. Evaluation of selection ten-
Am J Obstet Gynecol1993;168:1407-1413. dencies of patients preferring prepared child-
38. Moutquin J, Gagnon R, Rainville C, et al. birth. Obstet GynecoI1973;42:371-377.
Maternal and neonatal outcome in preg- 55. Hetherington S. A controlled study of the
nancies with no risk factor. Can Med Assoc J effect of prepared childbirth classes on
1987;137:728-732. obstetric outcomes. Birth 1990;17:90.
16. Midwifery Care, Childbirth Preparation, and Labor Support and Cesarean Section Rates 245

56. Charles AG, Norr KL, Block CR, et al. 68. Enkin MW. Commentary: are the correct
Obstetric and psychological effects of psycho- outcomes of prenatal education being
prophylactic preparation for childbirth. Am J measured? Birth 1990;17:90-91.
Obstet Gynecol 1978;131:44-52. 69. Simkin P, Enkin M. Antenatal classes. In:
57. Goodrich FW, Thoms H. A clinical study of Chalmers I, Enkin M, Keirse MJN, eds
natural childbirth: a preliminary report from Effective care in pregnancy and childbirth.
a teaching ward service. Am J Obstet Gynecol Oxford: Oxford University Press, 1989:318-
1948;56:875-883. 333.
58. Thoms H, Goodrich FW. Training for child- 70. Raphael D. New patterns in doula-client
birth. JAMA 1949;140:156-158. relations. Midwife, Health Visitor and Com-
59. Thoms H, Karlovsky ED. Two thousand munity Nurse 1988;24:376-379.
deliveries under a training for childbirth 71. Sosa R, Kennell J, Klaus M, et al. The effect
program: a statistical survey and com- of a supportive companion on perinatal pro-
mentary. Am J Obstet Gynecol 1954;68: blems, length of labor, and mother-infant
279-285. interaction. N Engl J Med 1980;303:597 -600.
60. Van Auken WB, Tomlinson DR. An appraisal 72. Klaus M, Kennell J, Robertson S, et al. Effects
of patient training for childbirth. Am J Obstet of social support during parturition on
Gynecol 1953;66:101-105. maternal and infant morbidity. Br Med J
61. Beck NS, Hall D. Natural childbirth: a 1986;293:585-587.
review and analysis. Obstet GynecoI1978;52: 73. Kennell J, Klaus M, McGrath S, et al. Con-
371-379. tinuous emotional support during labor in a
62. Scott JR, Rose NB. Effect of psychopro- US hospital: a randomized controlled trial.
phylaxis (Lamaze preparation) on labor and JAMA 1991;265:2197-2201.
delivery in primiparas. N Engl J Med 1976; 74. Radin TG, Harman JS, Hanson DA. Nurses'
294:1205-1207. care during labor: its effect on the cesarean
63. Hughey MJ, McElin TW. Maternal and fetal birth rate of healthy, nulliparous women.
outcome of Lamaze-prepared patients. Obstet Birth 1993;20:14-21.
GynecoI1978;51:643-647. 75. Hodnett ED, Osborn RW. A randomized trial
64. Bennett A, Hewson D, Booker E, et al. of the effects of monotrice support during
Antenatal preparation and labor support in labor: mothers' views two to four weeks post-
relation to birth outcomes. Birth 1985; partum. Birth 1989;16:177 -183.
12:9-16. 76. Odent M. Birth reborn. London: Souvenir
65. Sturrock WA, Johnson JA. The relationship Press, 1984.
between childbirth education classes and 77. McNiven P, Hodnett E, O'Brien-Pallas LL.
obstetric outcome. Birth 1990;17:82-85. Supporting women in labor: a work sampling
66. Cogan R. Effects of childbirth preparation. study of the activities of labor and delivery
In: Clinical obstetrics and gynecology. New nurses. Birth 1992;19:3-7.
York: Harper & Row, 1980. 78. Klein RPD, Gist NF, Nicholson J, et al. A
67. Shearer MH. Commentary: effects of pre- study of father and nurse support during
natal classes cannot be measured by ob- labor. Birth 1981;8:161-164.
stetric management. Birth 1993;20:130- 79. Queenan JT, ed. New mothers add to their
131. support staff. Contemp Ob/Gyn 1994;39:145.
17
Cesarean Projects at the State and
National Level
BRUCE L. FLAMM

The 450% increase in the U.S. cesarean The third biennial report to the governor
rate over 20 years did not go unnoticed by and the state legislature focused on cesarean
governmental agencies. In response to per- section. 1 Cesarean was chosen for investi-
sistently increasing rates, several states gation because of rising cesarean rates
embarked upon cesarean-related projects. and a very wide variation of rates observed
Related committees have been formed to in different parts of the state. 2 Data were
address rising cesarean rates in other indus- analyzed from the Department of Health
trialized nations. Summaries of some of Services Birth Cohort Data File for 1987,
these programs are outlined here. This which included information on more than
information will be of value to physicians, 500,000 births. The report noted that the
hospitals, and others who are interested in cesarean delivery rate for 1987 in California
developing quality assurance programs to was 24.4% but that there were wide varia-
evaluate the utilization of cesarean opera- tions in different areas. For example, in one
tions. Other nations and other states within area of Los Angeles (including 10 hospitals)
the United States have no doubt formed the cesarean rate was 32.9%. By compari-
similar committees, and hence the following son, an area (including 13 hospitals) with
examples are not meant to be an exhaustive similar demographics and economics in
listing of such projects. another part of the state had a cesarean rate
of 22.9%. In spite of the very different cesa-
rean rates, perinatal outcomes for the two
United States areas were found to be similar. There was no
obvious explanation why cesareans were
California used 44% more frequently in one area than
The California Collaborative Cesarean in the other.
Appropriateness Review and Education The report recommended that various
organizations be brought together to address
(CARE) Project
"what appears to be unnecessarily high C-
California State Assembly Concurrent Reso- section rates in certain parts of California."
ution 199 (Resolution Chapter 124, 1984) The CHPDAC subsequently initiated the
requested studies and reports that focused formation of a Cesarean Work Group, which
on the rising costs of health care. In 1985, convened for the first time in January 1992.
responsibility for these projects was as- This group included representatives from
signed by Governor George Deukmejian to The American College of Obstetrics an
the California Health Policy and Data Advi- Gynecology (ACOG), the California Associa-
sory Commission CHPDAC. tion of Hospitals and Healthcare Systems

247
248 B.L.Flamm

(CAHHS), the California Medical Associa- that detailed information about cesarean
tion (CMA), the California Office of State- rates for specific physicians and specific
wide Health Planning and Development hospitals would be confidential. These data
(OSHPD), Kaiser Permanente (America's would be compiled and fed back to the volun-
largest health maintenance organization, tarily participating hospitals and physicians
or HMO), Blue Cross of California, the to be used in confidential educational and
California Health Information Association quality assurance programs. As a member of
(CHIA), the Bay Area Business Group on the California Cesarean Work Group and
Health, SRI International, and the Univer- the California Collaborative CARE project,
sity of California San Francisco. I can verify that public dissemination of
The California Cesarean Work Group abstracted data was not a consideration.
reviewed state and national data along with In contrast, California State Assembly
the work that had already been done in this Bill No. 686 would make cesarean rates
area in other states (see following). It was public. First introduced by Assembly Mem-
noted that one of the U.S. Public Health ber Roybal-Allard in 1989, AB 686 would
Service's goals was to reduce the national add Chapter 6.6 to Division 2 of the Health
cesarean rate from 24.4% in 1987 to no more and Safety Code, relating to cesarean sec-
than 15% by the year 2000. It was also noted tions. The legislative counsel's digest
that the U.S. Public Health Service's goals pointed out that existing law provides speci-
for the year 2000 were based on the work of fic informed consent requirements which
a consortium of nearly 300 national organi- must be satisfied before a hysterectomy may
zations and all state health departments. To be performed but that no similar require-
reach that national target, California would ments exist for the cesarean operation. AB
have to perform about 44,000 fewer cesarean 686 would mandate that a specific informa-
operations each year. tion about cesarean delivery be given to all
During 1992-1993, the California Cesa- women during the first 6 months of preg-
rean Work Group developed the The nancy. This would be accomplished by
California Collaborative CARE (Cesa- means of a standardized written summary
rean Appropriateness Review and Education) in layman's language that would discuss
Project. This was designed to be a voluntary risks, benefits, and alternatives to cesarean
program in which cesarean-related data along with several other topics. Other topics
would be collected at voluntarily parti- would include a discussion of vaginal birth
cipating hospitals. Medical records staff at after cesarean section.
participating hospitals would abstract rele- The Bill would also require documentation
vant data from the charts of all patients that standardized written information was
who underwent cesarean operations onto a given to the patient. Perhaps the most con-
specifically designed data-collection form. troversial aspect of AB 686 was that it would
The data would then be analyzed and sent also allow for public disclosure of cesarean
back to the participating hospital's obstetric rates. "This bill would require every hospital
staff in a series of confidential reports. Indi- that provides obstetrical services to, upon
vidual physicians would then be able to request, provide information to any person
compare their practice patterns to those of regarding the rate of cesarean sections per-
their peers (with names deleted) both locally formed at that hospital." Perhaps because of
and across the state. At the time of this this controversial aspect, the Bill faced
writing, the California Collaborative CARE opposition from the California Medical
project is in its pilot phase. Association (CMA) and the American College
of Obstetricians and Gynecologists (ACOG).
AB 686 has been amended several times, but
California State Assembly Bill No. 686
thus far it has not been ratified.
A fundamental principle of the California
Collaborative CARE project just described is
17. Cesarean Projects at the State and National Level 249

California State Assembly Bill No. 524 hospital's cesarean rate over the others may
thus be somewhat misleading. Even with
AB 524 (California Hospital Outcomes these limitations, RAMO reports on cesa-
Assessment) was approved by the state rean section would be of great value if fed
legislature and signed by Governor Pete back to hospitals and physicians for use
Wilson in late 1991. AB 524 requires that in their educational and quality assurance
the California Office of Statewide Health programs.
Planning and Development (OSHPD) begin
Risk-Adjusted Monitoring of Outcomes
(RAMO) of hospital inpatient care. A sum- Florida
mary of the evolution of AB 524 and a des-
In 1991 the Florida state legislature passed
cription of the RAMO approach to outcome
a bill which requires that hospitals receiv-
analysis has been recently published. 3 AB
ing public funds for maternity care must
524 established a 12-member technical
comply with practice parameters that in-
advisory committee to make recommen-
clude the assessment of the feasibility of
dations about production of RAMO reports.
attempting vaginal birth after a previous
The committee chose cesarean section, the
cesarean delivery.
most common major operation performed in
California, as one of the RAMO projects.
One of the benefits of the RAMO process is
Maine
that it is apparently better suited to the
analysis of statewide hospital discharge Although often not discussed in publications
abstract databases than "severity adjust- on rising cesarean rates, fear of malpractice
ment" methodologies which have tradi- litigation is clearly one of the most impor-
tionally been applied to situations in which tant factors. The general rule is that the
the medical record of each patient is availa- only cesarean a doctor is sued for is the one
ble. However, it must be noted that the data he or she does not do. In other words, if a
contained in the statewide hospital dis- cesarean is done early in labor, it would be
charge database related to indications for difficult to blame any neonatal or childhood
cesarean delivery is limited. In contrast, the problems on the obstetrician. This type of
data abstraction form developed by the thinking and the resultant practice of "de-
California Collaborative CARE project was fensive medicine" has no doubt resulted in
designed specifically to address the indica- countless cesarean operations.
tions for cesarean. Several members of the In April of 1990, Maine Governor John R.
Clinical Advisory Panel on Obstetric Out- McKernon, Jr. signed into law Public Law
comes (of the AB 524 California Hospital 1990, Chapter 931, which included sub-
Outcomes Assessment Project), including chapter IX, the 5-year Medical Liability
myself, cautioned that any report on state- Demonstration Project (MLDP). With this
wide cesarean rates must therefore be inter- law, Maine will become the first state to
preted with caution. This is particularly formally use medical practice guidelines for
true because the public may be misled by malpractice defense. If an obstetrician elects
confusing results. For example, analysis to participate in this program, and if he or
may reveal a slightly higher cesarean rate she practices in accordance with the practice
at one hospital than at another hospital in guidelines, the guidelines may be used as
the same area. This difference may persist an affirmative defense in a medical malprac-
even after "risk adjustment" by the RAMO tice suit. By proving compliance with the
process. However, a small difference in the guidelines, a physician will be able to win
two neighboring hospital's cesarean rates dismissal of a suit. In other words, the
might be explainable by factors that are not MLDP protects physicians from liability in
contained in the state's discharge abstract exchange for following specific practice
database. Newspaper reports touting one guidelines. 4
250 B.L.Flamm

Five obstetric protocols developed for the State Assembly Bill No. 686 that was des-
project potentially affect cesarean rates: cribed earlier.
cesarean for failure to progress, breech, fetal
distress, repeat cesarean, and management
of prolonged pregnancy. There is often con-
Michigan
fusion about what actually constitutes a The Greater Detroit Area Health Council in
legally defensible standard of care. If guide- conjunction with the ACOG began a vaginal
lines are accepted as the standard of care, birth after cesarean section (VBAC) educa-
and if a physician adheres to these guide- tional project in 1990. The program involves
lines, it follows that a bad outcome does 13 hospitals and includes presentations to
not imply negligence. The Public Health the medical staff along with patient educa-
Research Institute of Portland, Maine plans tion in the form of specific pamphlet.
to evaluate this project. If successfully im-
plemented, the Maine MLDP Law could
serve as a model for the rest of the naiton.
Minnesota
The Minnesota Clinical Comparison and
Assessment program developed cesarean-
Maryland
related guidelines and distributed them to
Cesarean rates varied from 17.8% to 42.6% all providers of obstetric care in 1991. More
at hospitals in Maryland in 1987. An inner- than 40 Minnesota hospitals began abstract-
city hospital, presumably handling many ing medical records in 1992 to evaluate the
high-risk obstetric cases, had a cesarean impact of the guidelines.
rate of 14.5% while a rural hospital had a
rate of 37.6%. In 1987 the Maryland Hospi-
tal Association's Council for Quality Health
New York
Care began to investigate cesarean rates in In 1986, cesarean rates for hospitals in New
the state and to search for reasons for inter- York State varied from 11% to 40%. This
hospital variations. The Council established huge variation could not be explained by
an obstetrics advisory committee that in- demographic or geographic factors. In 1987
cluded representatives from the Maryland a joint Task Force was formed by ACOG
Medical Society and the Obstetrical and District II and the New York State Depart-
Gynecological Society of Maryland. The ment of Health. The Task Force developed a
Council developed guidelines and data col- two-tiered review program in which were
lection forms that hospital quality assur- included both internal and external (site
ance (QA) staff and obstetrics chiefs could visit) reviews of cesarean charts. 7
use. The Maryland Hospital Association is Similar to the California CARE project,
currently helping hospitals perform internal participation in the program was voluntary.
review of records to investigate reasons for In 1989-1990, 24 of 176 hospitals in the
large interhospital variations in cesarean state with obstetric services had site visits.
rates. 5 ,6 Perhaps partially as a result of this program
the New York State cesarean rate fell from
more than 25% in 1987 to 23.4% in 1991.
Massachusetts National rates also leveled off at about 24%
In 1985 legislation was passed in Massachu- during the same time interval, but in New
setts which mandated that every pregnant York State the downward trend was greater
woman has the right to know a hospital's in those hospitals that participated than in
cesarean section rate. This includes the total those which did not.
rate, the primary rate, the repeat rate, and
the vaginal birth after cesarean rate. This is
very similar to the controversial California
17. Cesarean Projects at the State and National Level 251

Vermont The Washington Business Group

In 1990 the Hospital Data Council accepted on Health
the Vermont Program for Quality in Health The Washington Business Group on Health
Care's proposal to develop a cesarean section (WBGH) is a national nonprofit health
quality improvement project. 8 The project policy and research organization that repre-
focused specifically on VBAC. Data were sents more than 150 corporations represent-
collected from all 12 hospitals with obstetric ing all segments of industry. In 1991 the
services. Results indicated that 72% of pa- group researched the topic of VBAC and
tients with prior cesarean were delivered by published a 23-page document aimed at the
repeat cesarean. Overall cesarean rates purchasers of health care. 13
varied widely from less than 12% to more
than 30% at some hospitals in Vermont. A The Public Citizen Health
program was developed that included both Research Group
patient and professional education.
The Public Citizen Health Research Group
is a division ofthe Washington-based Public
U.S. National Cesarean Programs Citizen, a group founded in 1971 by con-
and Reports sumer advocate Ralph Nader. The Health
u.s. National Consensus Conference Research Group obtained and analyzed
on Cesarean Birth statewide hospital discharge databases that
included information on cesarean opera-
Between 1970 and 1980, the U.S. cesarean tions. In 1989, the group released a contro-
rate increased from 5% to 15%. Because the versial report that listed cesarean rates at
rate had tripled in a single decade, the Na- more than 2400 hospitals across the United
tional Institutes of Health (NIH) formed States. 14 The report concluded that 475,000
a 19-member Task Force on Cesarean of the 934,000 cesareans performed in 1987
Childbirth. 9 In 1980 a major conference was were unnecessary resulting in more than 1
held at the NIH headquarters to review the million extra hospital days and extra costs of
Task Force findings and bring together more than $1 billion. The report generated
experts from many fields to evaluate the much media attention and was sharply
rising cesarean rate. The Task Force find- criticized by the American College of Obste-
ings were published in 1981 in a 537-page tricians and Gynecologists. 15 Although
report. 10 A summary of the Task Force many experts felt that the report greatly
recommendations was also published in overestimated the number of unnecessary
several major medical journals. A subse- operations, it was difficult to argue with the
quent study demonstrated that the cesarean conclusion that at least some unnecessary
project received more media coverage than cesareans were being performed.
any of the seven other NIH consensus de-
velopment projects even though other topics
included controversial issues such as breast Canada
cancer and coronary bypass surgery. 11 In
spite of the widespread dissemination of The Canadian National Consensus
recommendations aimed at lowering the Conference on Cesarean Birth
utilization of this operation, the national
cesarean rate continued to rise. 12 In 1985 a project similar to the cesarean
consensus development program of the U.S.
NIH was developed in Canada. The com-
mittee reviewed the relevant literature and
held a meeting to develop recommendations.
They then circulated an interim statement
252 B.L.Flamm

to more than 30,000 individuals and organi- Conclusion

zations. A final statement (Canadian Con-
sensus Statement on Cesarean Birth), which Rising cesarean rates have clearly cap-
included cesarean reduction guidelines, was tured the attention of agencies at state and
then developed and published in the Journal national levels. Because cesarean section is
of the Canadian Medical Association. 16 The the most frequently performed major opera-
Society of Obstetricians and Gynecologists tive procedure in the United States and be-
of Canada and the Association of Professors cause it has held this position each year for
of Obstetrics and Gynecology fully endorsed the past decade, governmental interest is
the final statement. not surprising. In fact, in the era of health
care reform it would be surprising if com-
monly performed operations were not sub-
jected to governmental scrutiny. Physicians
The Ontario Cesarean Birth Quality and hospitals that voluntarily develop qua-
Assurance Committee lity assurance programs for cesarean de-
As had been the case in the United States livery will undoubtedly fair well in any
there was evidence that the Canadian Con~ future health care environment.
sensus Statement on Cesarean Birth had Finally, some physicians are fearful that
little effect on cesarean rates. Ontario governmental or third-party agencies, per-
Canada's largest province, contains 35% haps motivated by cost-containment issues,
of the country's population and 44% of its will attempt to dissuade obstetricians from
obstetricians; 154 hospitals offer maternity performing cesarean sections that truly are
care in Ontario. A study conducted in On- medically indicated. The American College
tario in 1987 -1988 showed that although of Obstetricians and Gynecologists is work-
about 90% of obstetricians were aware of the ing to avoid this potentially dangerous
cesarean guidelines, only one-third reported scenario by ensuring that ACOG represen-
changing their practice because of them. tatives participate in cesarean projects at
Although self-reported data from physician the state and national levels.
surveys indicated that cesarean rates de-
clined after publication of the guidelines, References
actual data showed that rates of cesarean
were 15%-49% higher than rates reported 1. Kerr C, et al. Variations in cesarean section
by the obstetricians. 17 rates in California, Biennial report on cost
In 1989 the Minister of Health announced containment. Sacramento, California Health
Policy and Data Advisory Commission,
the formation ofthe Ontario Cesarean Birth
1991.
Quality Assurance Committee. The Minis- 2. Dumbauld S. Trends in cesarean births in
ter of Health suggested a 15% cesarean rate California, 1970-1986. Sacramento: State of
as a target goal to be achieved within 2 California Department of Health Services
years. The committee consisted of members Health Data and Statistics Branch, No. 88~
from the Ontario Medical Association, the 05171, July 1988.
Ontario Section of the Society of Obstetri- 3. Johns L. State Report: measuring quality in
cians and Gynecologists of Canada, the California. Health Affairs 1992;Spring:266-
Ontario chapter of the Canadian College of 270.
Family Practice, the Ontario College ofPhy- 4. Smith GH. Maine's liability demonstration
project: Relating liability to practice para-
sicians and Surgeons, the Ontario Nurses'
meters. Washington, DC: American Medical
Association, the Ontario Hospital Associa- Association, AMA State Health Legislation
tion, childbirth consumer groups, and the Report, Fall 1990.
Ministry of Health. In 1991 the committee 5. Kazandjian VA, Summer SJ. Evaluating the
published a 76-page report that analyzed appropriateness of care: a study of cesarean
national and international cesarean rates section rates. Qual Rev J Qual Assur 1989;
and made many specific recommendations. IS 15:206-214.
17. Cesarean Projects at the State and National Level 253

6. Kazandjian VA, Summer SJ. Cesarean sec- the National Consensus Development Con-
tion guideline: the Maryland experience. ference in 1980. JAMA 1984;252:3273-3276.
JAMA 1990;263:1491. 13. Vaginal birth after cesarean section: infor-
7. Dillon WP, Choate JW, Nusbaum ML, mation for health care providers. Quality
McCarthy MA, McCALL M, Rosen MG. Obs- Resource Center Report. Washington, DC:
tetric care and cesarean birth rates: a pro- Washington Business Group on Health, 1991.
gram to monitor quality of care. Obstet 14. Silver L, Wolfe SM. Unnecessary cesarean
Gynecol 1992;80:731-737. sections: how to cure a national epidemic.
8. Vermont Program for Quality in Health Washington, DC: Health Research Group,
Care, Inc. (VPQHC), 136 Main Street, Box 1989.
1356, Montpelier, Vermont 05601. 15. Half the cesarean operations in U.S. called
9. Jacoby I. The consensus development pro- unnecessary. Los Angeles Times, January
gram of the National Institutes of Health. 27,1989.
Current practices and historical perspec- 16. Lomas J, et al. Consensus Conference Report.
tives. Int J Tech Assess Health Care 1985; Indications for cesarean section: final state-
2:420-432. ment of the panel of the National Consensus
10. U.S. Department of Health and Human Conference on Aspects of Cesarean Birth.
Services. Cesarean childbirth. Washington, Can Med Assoc J; 134:1348-1352.
DC: Government Printing Office, 1981. 17. Lomas J, Anderson G, Domnick-Pierre K,
(National Institutes of Health publication Vayda E, Enkin M, Hannah W. Do practice
No. 82-2076.) guidelines guide practice? N Engl J Med
11. Winkler JD, Kanouse DE, Brodsley L, Brook 1989;321:1306-1311.
RH. Popular press coverage of eight National 18. Cesarean Birth Quality Assurance Com-
Institutes of Health Consensus Development mittee. Appropriate use of cesarean section:
Topics. JAMA 1986;255:1323-1327. recommendations for a quality assurance
12. Gleicher H. Cesarean section rates in the program. Ministry of Health Ontario,
United States: the short-term failure of Canada: 1991;ISBN 0-7729-8614-2.
18
Guidelines for Appropriate
Utilization of Cesarean Operations
BRUCE L. FLAMM

The U.S. Department of Health objectives "cephalopelvic disproportion," and "failed

for the 1990s call for decreasing the national induction."
cesarean rate to no more than 15% of all
deliveries by the year 2000. 1 This would A. Avoid "latent-phase" cesareans.
mean performing about 400,000 fewer ce- Cesareans for "failure to progress" performed
sareans each year in the United States. To before the cervix has reached 4cm of dilata-
some this may seen like an impossible tion are generally not indicated. Make sure
goal. However, by using guidelines, many that the patient is truly in labor before
hospitals have already lowered their overall admitting her to the hospital. This is one of
cesarean rates to less than 15%. Importantly, the key components of Dublin's "active
this has been achieved with no increase in management" program, and may contribute
perinatal or maternal morbidity or mortality. more to the lowering of cesarean rates than
In today's medical-legal and social climate, the aggressive use of oxytocin. Patients with
it may be unreasonable to assume that a set reassuring fetal monitoring results who are
of guidelines aimed solely at physicians will not in labor should be sent home to await the
lead to a major reduction of the U.S. cesarean onset of labor. For those patients who are
rate. If the U.S. Department of Health's experiencing a prolonged latent phase,
objectives are to be achieved and sustained, therapeutic rest with morphine sulfate or a
obstetricians will need the support of nurses, similar medication is often very beneficial.
childbirth educators, attorneys, govern- B. Avoid failed inductions. If the cervix
mental agencies, and the lay public. Similar is not favorable and does not respond to
cooperation will be necessary in any country oxytocin, consider delaying the induction for
that plans to address the cesarean issue on a a few days if fetal monitoring is reassuring.
national level. Alternatively, consider using prostaglandins
for cervical ripening. Above all, confirm that
the indication for induction is medically
Guidelines for Physicians valid. Many patients strongly request elec-
tive induction near term. Such patients
1. Decrease the Rate of Primary should be reminded that awaiting a more
Cesarean Section for "Failure to favorable cervix and the onset of natural
Progress" to 5% (or less) of All labor may result in an easier and less com-
Deliveries plicated birth.
This category should also include all ce- C. A lack of change in cervical dilatation
sareans for "dystocia," "failure to descend," after 1 h (multiparous patient) or 2 h (pri-

255
256 B.L.Flamm

miparous patient) in the active phase of that fetal monitoring is an inexact science
labor is an indication that some type of in- and that true cases of "fetal distress" ne-
tervention may be necessary. This finding is cessitating immediate operative delivery
not an indication for cesarean delivery. are infrequent. One potential benefit of
Appropriate interventions may include electronic fetal monitoring is that, if the
ambulation, placement of an internal uterine pattern is reassuring, there is no need for
pressure catheter to assess the labor pattern, intervention. This is particularly important
oxytocin augmentation oflabor, or continued when a patient is making slow progress
observation. in labor.
D. Avoid placing arbitrary time limits B. Genital herpes: Assuming a 5% pre-
on the second stage oflabor. This is especially valence of genital herpes in the reproductive-
true if epidural anesthesia is being used. aged population, approximately 200,000
Consider alternative positions for pushing, women with a history of herpes will deliver
such as squatting. each year in the United States. Studies from
Washington State revealed that increasing
E. The performance ofa primary cesarean
the cesarean rate to 75% for women with a
will place the patient at increased risk during
history of recurrent genital herpes failed to
future pregnancies. Repeat cesarean section
reduce the incidence of neonatal herpes.
and its alternative, vaginal birth after
Recent reports indicate that cesarean de-
cesarean, are each associated with specific
livery is indicated only for those women
risks. This fact should be part of the informed-
with lesions or symptoms of a recurrence
consent process at the time of the decision to
during labor. Specific details are presented
perform an initial cesarean, particularly if
in Chapter 10.
there is no emergent need to proceed with
the operation. The prime example is a ce- C. Breech presentation: The prevalence
sarean performed solely because of "slow of breech presentation at term is approxi-
progress in labor." mately 3%-4%. Currently about 85% of
breech presentations are delivered by cesar-
ean section in the United States. Many of
2. Decrease the Rate of Primary these 100,000 operations per year could
Cesarean Section for All Indications be prevented by external cephalic version
Other Than "Failure to Progress" to while others could be safely avoided by
5% (or less) of All Deliveries allowing vaginal breech delivery in selected
cases. Specific details are presented III
A. Fetal distress: Use electronic fetal Chapter 6.
heartrate monitoring appropriately and
D. Twin Gestation. Twins comprise
avoid overinterpretation of subtle monitor
approximately 1% of all pregnancies. Cesar-
findings. In recent years there has been
ean section is generally not indicated when
a proliferation of for-profit educational
both twins are in vertex presentation,
courses on the interpretation of electronic
regardless of gestational age. Although ex-
fetal monitormg. Some of these meetings
ternal cephalic version of the second non-
have had unfortunate effects on both phy-
vertex twin has recently become popular, 2
sicians and nurses. Presentations linking
breech extraction remains a reasonable
minimally suspicious monitor strips with
alternative. 3 Specific guidelines for the intra-
subsequent poor outcomes leave attendees
partum management of twin gestation are
with the mistaken impression that almost
presented in Chapter 9.
any change in the fetal heartrate is an
indication for immediate cesarean delivery.
Although such cases do occur they are the
exception, not the rule. Most experts agree
18. Guidelines for Appropriate Utilization of Cesarean Operations 257

3. Decrease the Rate of Repeat Current data do not support the use of x-ray
Cesarean Section to 5% (or less) pelvimetry to screen such patients before
of All Deliveries trial of labor.

A. Follow the American College of Obs- F. Many women request repeat cesarean
tetricians and Gynecologists (ACOG) guide- operations. However, many women request
lines for vaginal delivery after previous primary cesarean operations. It is a double
cesarean birth.4 Approximately 75% of standard to treat these requests differently.
women who are candidates for trial of labor The decision to perform a major operation
will deliver vaginally if labor is allowed. must be made on the basis of medical indi-
cations.
B. Most women with a prior cesarean are
appropriate candidates for trial of labor.
Women who have contraindications for Guidelines for Midwives
labor with the current pregnancy such as
complete placenta previa or transverse lie 1. Reaffirm That Pregnancy Is a
are obviously not candidates for trial oflabor. Normal Condition, Not a Disease
C. Several studies indicate that trial of A. Nurse-midwifery educational prog-
labor with two prior cesareans is relatively rams usually require that registered nurse
safe; however, the risk of uterine rupture (RN) candidates for admission have a mini-
appears to be higher than for patients with mum of 2 years of experience working in a
one prior cesarean (approximately 1% com- labor and delivery unit, which is generally
pared to 0.5%). Minimal data are available in a medical center environment. Thus,
on trial of labor with more than two prior nurses accepted to nurse-midwifery schools
cesareans; therefore, it may be prudent to may have only experienced birth in a high-
perform a repeat cesarean unless the patient tech, high-risk environment. Furthermore,
has a very strong desire for vaginal birth some nurses apply to midwifery school be-
and is willing to assume a potentially in- cause they want to "do" more. Admissions
creased risk. panels may want to focus on those candi-
D. Unknown uterine scar type is defin- dates who have a great interest in normal
itely not a contraindication to trial of labor. childbirth and a lesser interest in advanced
The majority of patients in several large ultrasound and electronic fetal monitoring
studies on trial of labor had undocumented techniques.
uterine scar types. 5 However, if the records B. Nurse-midwifery educational prog-
cannot be obtained and the primary cesarean rams are commonly located in university
was performed because of transverse lie, medical centers. This may expose midwifery
placenta previa, or premature breech, elect- students to a continuous stream of com-
ive repeat cesarean may be indicated. Al- plicated pregnancies. Nurse-midwifery
though some centers have allowed trial of educators must continue to emphasize that
labor with a known prior classical or vertical midwives are experts in normal childbirth.
uterine incision, experts generally agree They should strive to make educational
that these patients are at significantly in- experiences available, perhaps at out-of-
creased risk for uterine rupture. hospital birth centers, which promote the
E. Prior cesarean for "failure to progress" commitment that midwifery has to normaliz-
or "cephalopelvic dysproportion" is not a ing the birth process.
contraindication to trial of labor. The
majority of patients in several large studies
actually had "failure to progress" listed as
the indication for their primary cesarean.
258 B.L.Flamm

Guidelines for Nurses and delivery nurse's primary role. Every

effort should be made to streamline unneces-
1. Recognize the Impact of sary paperwork so that the nurse can spend
Intrapartum Nursing Care on more time with the patient and less time
with the chart.
Cesarean Rates
A. Often a nurse is the only health care D. In settings where one-on-one nursing
professional at the bedside during most of is not possible, consideration should be given
labor. Labor can be physically and emo- to trained "doula" support. The patient's
tionally exhausting not only for the patient husband or partner, who may be attending a
and her partner but for the labor and delivery birth for the very first time, is not equipped
nurse as well. In particular, assisting the to handle alone the physical and emotional
couple with the pushing stage of labor needs of a laboring woman. (See point num-
makes great demands on the nurse. Anyone ber four under guidelines for the lay public,
who has been with a woman during the following.)
second stage of labor knows that this can be
a difficult time. If the patient becomes angry
or "gives up," it may be tempting for the Guidelines for Childbirth
nurse to call the doctor and say, "she's just Instructors
not going to make it." Such comments are
often taken very seriously by physicians, 1. Recognize the Impact of
and at times the response is to order that the Childbirth Education on Cesarean
patient be prepared for surgery. Clearly,
this may end a difficult situation for the
Rates
nurse but may not be in the patient's best A. Studies on the impact of childbirth
interest. The bedside nurse must be aware of education on cesarean rates have had mixed
this potential scenario and request to be results (see Chapter 16, this volume). How-
relieved by a colleague if she feels that she is ever, this may simply reflect the difficulty in
becoming emotionally or physically ex- conducting such studies. Experienced doc-
hausted while caring for the patient. tors and midwives agree that women who
have no preparation for labor and delivery
B. Nursing administrators must be
are far more likely to decompensate and
aware of the importance of a nurse at the
demand unnecessary cesareans than women
bedside during active labor. This one-on-one
who have a good understanding of what to
labor support is a key component of Dublin's
expect.
"active management" program and un-
doubtedly contributes to their low cesarean B. Present an optimistic but realistic
rates. If a nurse is assigned to care simul- picture of the normal birth process. Promin-
taneously for several women in active labor, ent childbirth educators have pointed out
it will not be physically possible for her to that some of their colleagues strictly avoid
offer the support needed by each individual the use of the word "pain" in their classes.
patient. Although one-on-one labor support This may be a disservice to those women
may appear to be an inefficient use oflimited who find labor to be far worse than anything
nursing resources, it must be remembered they expected.
that every patient who "gives up" during
C. Avoid the use of comprehensive birth
labor and demands an unnecessary cesarean
plans that specifically list each of the inter-
will require additional nursing care during
ventions that a woman will (or will not)
her postoperative hospitalization.
allow during labor. These point-by-point
C. Charting duties that require exten- birth plans often alienate both the physician
sive paperwork also detract from the labor and nursing staffs. It is my opinion that it is
18. Guidelines for Appropriate Utilization of Cesarean Operations 259

far more fruitful for a couple to stress their residents understand when the operation is
desire for natural childbirth during the pre- really necessary rather than simply teaching
natal visits and to reiterate these feelings them how to best perform the procedure or
during labor. what antibiotics to use.

Guidelines for Medical School

and Residency Faculty Guidelines for Attorneys
1. Emphasize Normal Pregnancy 1. Ponder the Impact of Your
and Birth Before Presenting Actions and Those of Your
Detailed Discussions of Pathologic Colleagues on Society
Conditions Malpractice attorneys now commonly adver-
Professors involved in the development of tise in newspapers and telephone books,
medical school and residency curriculum seeking out cases that involve alleged "birth
should place more emphasis on the normal injuries." These ads generally offer incen-
aspects of pregnancy and childbirth. Cur- tives such as a "no recovery, no fee" guaran-
rent teaching often leads medical students tee and occasionally include testimonials by
to believe that pregnancy is an extremely clients who have successfully won millions
dangerous condition that often culminates of dollars. To many individuals such ads are
in a series of life-threatening catastrophes. an open invitation to obtain wealth far
In reality, many if not most pregnancies beyond their wildest dreams. These ads will
would have good outcomes with little or no certainly attract some individuals who have
medical intervention. Students should be actually been harmed by medical malprac-
taught that the goal of obstetric care is tice. Unfortunately, they will also attract
to screen for risk factors and to avoid inter- fortune hunters. Initial review of most cases
ventions when no problems exist. Ove- solicited in this manner will probably lead
remphasis on pregnant patients with the attorney to doubt ifthere was any wrong-
insulin-dependent diabetes and other serious doing. If consultation with medical experts
medical diseases may lead students to the confirms this opinion, it would be unethical
mistaken conclusion that every pregnancy is to proceed with litigation. Still, the potential
fraught with danger. In addition to lectures for a huge verdict or out-of-court settlement
by perinatologists on high-risk obstetrics, must be very tempting. Attorneys faced with
medical students and obstetric residents such temptations must understand that
should be exposed to the nurse-midwifery thousands upon thousands of unnecessary
care of uncomplicated pregnant women. cesarean operations have been performed by
doctors fearful of frivolous litigation.
Many obstetricians fear that any birth
2. It Is Far Easier to Teach a Young with a poor outcome will result in a lawsuit
Physician How to Do a Cesarean even if superior medical care is provided.
This fear is not totally unjustified. Although
Than When to Do One a physician who has provided excellent care
In the past, cesarean-related education has will typically prevail in such a lawsuit, the
focused almost entirely on discussions of agony of being dragged through the litiga-
subtle differences in surgical technique. tion process cannot be ignored. Formulating
Entire grand rounds programs are often a page of statements repetitively charging
dedicated to topics such as antibiotic pro- that a physician is guilty of gross negligence
phylaxis for cesarean delivery. Clearly, and incompetence may be all in a day's work
more attention must be devoted to helping for a malpractice attorney but such a docu-
260 B.L.Flamm

ment will be devastating for the physician culminated in the vaginal birth of vigorous
who receives it. Likewise, taking a deposi- infants with normal Apgar scores. Such
tion from an accused physician may be a cases are not difficult to find. In fact it is
mundane matter for an experienced mal- unusual to find a monitor strip that does not
practice attorney, but it may be the most include at least some decelerations. Apgar
difficult day in the life of the doctor. scores should not be revealed until after
Attorneys may argue that induring such everyone has been allowed to voice their
pain is an unavoidable consequence of com- opinions.
mitting "negligent" actions, but it must be
remembered that standard-of-care issues
are often confusing. Unintentionally miss-
2. Initiate Cesarean Review
ing a few subtle late decelerations on a fetal Meetings
monitor recording is definitely not equi- In contrast to fetal monitor review meetings,
valent to robbing a bank, yet some plaintiff which tend to focus on cases in which cesar-
attorneys treat doctors like armed robbers. eans should have been done (or should have
Worse yet, attorneys who knowingly pursue been done sooner), cesarean review meetings
frivolous malpractice cases should be aware . focus on cases where cesareans might have
that they are committing legal malpractice been avoided. As for fetal monitor review
and that their actions will have profound meetings, these meetings should not be
consequences not only for innocent physi- punitive lest obstetricians be placed in the
cians but also for society. double-bind situation of being criticized at
one meeting if a cesarean is performed and
at the other meeting if it is not. At some
Guidelines for Medical hospitals, every cesarean is briefly presented.
Directors and Chiefs of This is unnecessary because many cases
Obstetric Services involve clear-cut or at least very reasonable
indications for surgery. However, "failure to
1. Avoid Punitive Obstetric progress" cesareans performed at 2 em and
Meetings elective repeat cesareans (or repeat opera-
tions performed after an hour or two oflabor)
It is common practice at many hospitals to make for interesting discussions.
review electronic fetal monitor strips at
weekly or monthly obstetric conferences.
Generally, cases with low Apgar scores are 3. Perform a Confidential Audit of
chosen for review. The doctor who attended Individual Cesarean Rates
the birth squirms in his or her seat as other At some hospitals, the cesarean rates of one
physicians review every inch of the monitor or two physicians are so high that the entire
strip. With the clarity of hindsight (and pre- hospital cesarean rate is elevated above the
ordained knowledge that the Apgar scores national average. If an audit reveals that
were low because the case is being presented), nine physicians have annual cesarean rates
participants call out their opinions and point in the range of12%-19% while one physician
out exactly at which point they would have has a rate of more than 45%, it would be
done a cesarean if they had been managing logical to focus educational efforts on that
the case. Eventually some obstetricians physician.
learn that they can avoid being placed on
the "hot seat" by doing a cesarean any time
there is any suspicious area on a fetal moni-
tor tracing. This is the wrong lesson.
A more balanced meeting would include
the review of suspicious monitor strips that
18. Guidelines for Appropriate Utilization of Cesarean Operations 261

Guidelines for Third-Party tional disability compensation offered after

cesarean delivery is a tempting incentive
Payers for some women to lean toward operative
delivery.
1. Pay Physicians More for Normal
Vaginal Delivery
Historically, physicians have been paid Guidelines for Hospital
more for performing a cesarean than for
attending a vaginal birth. In other words, a Administrators
physician could stay in the hospital all night
helping a woman through a difficult labor Be Wary of Potential Conflicts
and make less money than for performing a of Interest
I-h operation and going home. This does not Hospital obstetric services are generally
make sense. financially rewarding. In 1987, U.S. hospitals
billed more than $10 billion for obstetric
2. Do Not Reimburse Patients care (this did not include physician fees).6
for Operations That Are Not This makes obstetric services the greatest
revenue generator for most hospitals. It is
Medically Necessary common knowledge that the hospital length
Some patients demand cesarean delivery of stay for cesarean delivery is approximately
solely as a matter of convenience. Whether twice as long as for vaginal birth. For hospi-
it is reasonable to perform such operations is tal administrators faced with difficult finan-
an ethical issue that is discussed elsewhere cial situations and empty beds, it may be
(see Chapter 15, this volume). Whether difficult to support initiatives aimed at
society or third-party payers should cover lowering cesarean rates.
the expenses of such operations is a separate
issue. A woman might not find the conveni-
ence of knowing her baby's exact birth date Guidelines for Legislatures
to be quite so appealing if this option was
offered along with the condition that she Pass Legislation That Eliminates
agree to pay the $3000 in extra hospital bills
Frivolous Lawsuits
generated by the operation and postoperative
care. The Maine plan (see Chapter 17, this vol-
ume) could serve as a model for other states.
The impact of legal issues on cesarean rates
3. Do Not Offer Additional Weeks of is difficult to measure, but many experts
Paid Disability for Operations That believe that it is of monumental importance.
Ate Not Medically Necessary Fear of frivolous malpractice suits may be
In some states it is common practice to pay the single most important factor driving up
for 6 weeks of postpartum disability after a cesarean rates, and this factor alone has
vaginal birth and 8 weeks after a cesarean probably resulted in hundreds of thousands
delivery. The same woman who might insist of unnecessary cesarean operations.
on cesarean delivery because it is "easier on
her body than going through labor" might
also insist on the 2 additional weeks of paid Guidelines for the Lay Public
disability. In reality, most women are suf-
ficiently recovered by 6 weeks postpartum to 1. Inquire about cesarean rates for hospi-
resume full activities regardless of the mode tals and physicians in your area. Small dif-
of birth. The "traditional" 2 weeks of addi- ferences may simply reflect the fact that
262 B.L.Flamm

some doctors care for more patients with References

complications than others. However, very
1. Healthy People 2000. Washington, DC: U.S.
high cesarean rates (greater than 30%) may
Department of Health Services, Public Health
be a warning if you hope to have a natural Services (publication no. 91-50212:378-379).
birth. 2. Chervenak FA, Johnson RE, Berkowitz RL, et
2. During prenatal visits and again on al. Is routine cesarean section necessary for
admission to the hospital, stress the fact vertex-nonvertex and vertex-transverse twin
that you would like to avoid a cesarean gestations? Am J Obstet Gynecol 1984;148:
operation if possible. 1-5.
3. Women should recognize that child- 3. Gocke SE, Nageotte MP, Garite T, Towers CV,
birth involves hard but valuable work and Dorcester W. Management of the nonvertext
that a request for the attending physician to second twin: primary cesarean, external ver-
''just do a cesarean" is inappropriate when sion, or primary breech extraction. Am J Obstet
Gynecol 1989;151:111-114.
there is no medical indication for surgery.
4. Guidelines for vaginal delivery after a prev-
4. Consider the help of a doula (trained ious cesarean birth. Committee Opinion Num-
labor support person). Although husbands ber 64. Washington, DC: American College of
can often be very helpful during labor, many Obstetricians and Gynecologists, 1988.
have never seen a baby born except perhaps 5. Leung A, Farmer R, Leung E, Medearis A,
in a film shown during a childbirth educa- Paul R. Risk factors associated with uterine
tion class. A trained support person who rupture during trail of labor after cesarean
has attended dozens of births will be better delivery: a case control study. Am J Obstet
equipped to reassure a woman who is having Gynecol 1993;168:1358-1363.
trouble with her labor. Also, the husband 6. The stork's share of America's hospital tab.
OBG Management, May 1994.
may be too emotionally involved to make
objective decisions. The husband's response
to seeing his wife in pain may be to beg the
doctor to do a cesarean and just "get it over
with."
Index

A Ambulation recommendations of
Abdominal circumference, and alternative to oxytocin ad- Mueller-Hillis maneuver,
macrosomia, 116 ministration, 55 34
Abortion, spontaneous, in twin use of, nurse-midwives ver- on oxytocin in dilatation ar-
gestations, 125 sus physicians, 225 rest, 32
Acidosis, 111 American Academy of Pediat- on trial of labor, 179
in asphyxia, 96 rics (AAP), on conflicts American Medical Association
effect of, on fetal heart rate, in maternal-fetal care, (AMA)
100, 108 200-202 on In re: Angela Carder, 197
scalp pH American College of Nurse- on legal interventions in preg-
for confirmation, 109 Midwives, certification nancy, 199-200
and fetal heart rate, 100 by,224 American Society for Psycho-
umbilical artery, breech American College of Obstetri- prophylaxis in Obstet-
presentation, 74-75, cians and Gynecologists rics (ASPO/Lamaze), 235
85 (ACOG),17 Amniocentesis, 117
Active management of labor Cesarean Work Group partic- Amnionitis, 108
(AML), 39, 43-48 ipation, California, 247 Amniotic fluid
in cephalopelvic dispropor- committee on ethics, 192 amnioinfusion to reestablish
tion, 152 informed consent consider- volume of, 168
National Maternity Hospital, ations, 200 volume of, and breech presen-
Dublin,224,255,258 on conflicts in maternal-fetal tation,65
Northwestern Memorial Hos- care,200-202 See also Oligohydramnios
pital, Chicago, 149 on fetal size for vaginal deliv- Amniotomy
Acupressure, to manage pain ery in breech presenta- in labor induction, 117
of childbirth, 235 tion, 120 use of, nurse-midwives ver-
Acyclovir, for suppressing geni- guidelines of sus physicians, 224
tal herpes, 136 for cesarean delivery capa- in vertex/nonvertex twin de-
Age bility, 167 livery, 126
gestational, and fetal macro- for vaginal delivery after Analgesia! anesthesia
somia,116 cesarean birth, 57, 214, choice of, with VLBW,
maternal 257 breech presentation, 86
indication for cesarean de- hypoxia defined by, 97 conduction
livery, 14-15, 153 on In re: Angela Carder, 197 for abdominal deliveries,
and risk of fetal macro- on Public Citizen Health Re- 80
somia, 115, 116 search Group conclu- contraindication with
Alabama, University of, study sions, 251 VLBW fetus, 86
of single-layer uterine on rate of cesarean delivery, for vaginal delivery with
closure, 60 reduced,37,142-143 breech presentation, 85

263
264 Index

Analgesia/anesthesia (cant.) with shoulder dystocia, 118 of electronic fetal monitor-

effect of, on fetal heart rate Ataxic syndromes, with cere- ing,106
variability, 108 bral palsy, 97 Benzodiazepines, effect of, on
epidural Atropine, effect of, on fetal fetal heart rate, 108
in active management of heart rate variability, 108 Bergstreser v. Mitchell, 168
labor, 47 Attorneys, guidelines for, 259- Beta-blockers, fetal bradycar-
fetal heart rate with, 110 260 dia associated with, 107
with oxytocin administra- Audit Bing, Elizabeth, 235
tion, association with of active management of la- Biofeedback, to manage pain of
uterine rupture, 56 bor program, 44 childbirth, 235
and rate of cesarean deliv- of cesarean rates, for physi- Biparietal diameter (BPD), and
ery,16 cians, 260 fetal macrosomia, 116
and rate of descent of the external, for reducing cesar- Birth certificates, cesarean
head,46 ean delivery rate, 154 data source, 2-3
and socioeconomic status, See also External review; Birthweight
39 Peer review low
use of, nurse-midwife versus Auscultation, for fetal heart and cerebral palsy, 98
physician deliveries, 225 rate monitoring, 95-96, indication for cesarean de-
in vertex/nonvertex twin 100, 101-102 livery, 14
delivery, 126 Australia maternal or sibling, and risk
Wisconsin study, 150 childbirth education study, of fetal macrosomia, 115
general, risks of, in obstet- 238 and outcomes of vaginal de-
rics, 80 prevalence of cerebral palsy livery
historic place of in childbirth, in,99 after prior cesarean sec-
235 randomized controlled trial, tion,57
narcotic, sinusoidal fetal EMF versus ausculta- for twins, 126-127
heart rate baseline asso- tion, 101-102 with very low birthweight,
ciated with, 108 rate of cesarean delivery, 9 69
Anemia, fetal, 108 Authorization, for specialist and perinatal outcome, 159-
Anencephaly, 202 consultation, 164-165 160
Anoxia, and fetal heart rate, Autonomy, patient's and rate of cesarean delivery
110. See also Asphyxia; and medical ethics, 191-192 for dystocia, 24
Hypoxia and requests for cesarean de- See also Very low birth-
Antibodies, for HSV, by viral livery, 193 weight (VLBW); Weight
type, 132-133 and risks of fetal interven- Bleeding, with external ce-
Apgar scores tion, 195, 201 phalic version, 83
and birthweight, 159 Blood sampling, fetal, and ce-
after breech delivery, 85 sarean delivery rate, 13
in low birthweight twins, 126 B Blue Cross of California, 248
Arizona, Nurse-Midwifery Ser- Barbiturates, in prolonged la- Bodily integrity
vice, Tucson, 226-227 tent phase labor, 31 maternal,201
Arm Basal ganglia, hypoxic- right to, 198
nuchal, in breech delivery, 78 ischemic injury to, and Booth Maternity Center, 232
posterior delivery of, 119- cerebral palsy, 97 Bowers v. Olch, 169
120 Bay Area Business Group on Brachial plexus injury
Arrest disorders, 31-34 Health,248 in breech delivery, 79
with protraction disorder, 35 Baylor College of Medicine in macrosomia and shoulder
Arrhythmias, fetal, tachycar- study dystocia, 118
dia associated with, 108 of scar type and vaginal de- Bradycardia, fetal
Asphyxia livery,58 defined,107-108
animal studies of anoxia, 98- of single-layer uterine clo- with external cephalic ver-
99 sure, 60 sion, 126
fetal, 96-97 Benefits with oxytocin use, 55
and fetal heart rate (FHR), of cesarean delivery, extrapo- See also Fetal heart rate
100 lation of, 141 (FHR)
Index 265

Brazil, rate of cesarean deliv- Los Angeles study, 227, cesarean delivery in
ery,9,15-16 228,231-232 guidelines for reducing rate
and insurance, 17 Menlo Park study, 228-229 of, 255, 257
Breech extraction, 126, 256 rate of cesarean delivery, 17 indication for, 13-14, 29,
Breech presentation, 52-53, University of, at San Fran- 84-85,117,145
65-88 cisco, 248 evaluating, 30, 31-34
decision analysis model, ex- uterine rupture rate, in trials and protraction disorders,
ample, 180-184 oflabor,53 34-35
with fetal macrosomia, 120 California Association of Hos- Cerebral palsy
incidence of, 65 pitals and Healthcare and asphyxia, 97-99
management of, 12-13 Systems (CAHHS), 247- association with breech pre-
guidelines for cesarean de- 248 sentation, 66
livery,256 California Health Information choreoathetoid, 97
physician training for, 143 Association (CHIA), 248 kernicterus associated with,
outcome, vaginal versus ce- California Health Policy and 97,98
sarean delivery, 145, 152 Data Advisory Commis- prevention of, 4-5
and vaginal birth after cesar- sion (CHPDAC), 247 spastic, 97
ean delivery, 59 California Medical Association, Certification
Wisconsin study, 150 248 of doulas, 242
See also External cephalic California Office of Statewide of nurse-midwives, 224
version Health Planning and De- Cervical dilatation
Britain/England/United velopment (OSHPD), 248 in latent phase labor, 29-30
Kingdom Risk-Adjusted Monitoring of and the Mueller-Hillis maneu-
active management of labor Outcomes, hospital inpa- ver validity, 34
in,47 tient care, 249 rate of, 25, 27
cesarean delivery in Canada and intervention, 255
criteria for, 144 active management of labor and oxytocin administra-
rate of, 9 in,47 tion,45
prevalence of cerebral palsy Cesarean Birth Quality As- secondary arrest of, 28
in,99 surance Committee, On- Cervical ripening
study of prostaglandin for tario, 252 before induction of labor,
cervical ripening, 60 mode of breech delivery in, 117,255
trial 68 prostaglandin E2 for, 60
EMF versus auscultation, National Consensus Confer- Cervicographic analysis, 119
Oxford,103 ence on Cesarean Birth, Cervix, effacement of, 26-27,
EMF versus auscultation, 251-252 45
Sheffield,102 nurse-midwife study, To- Cesarean Birth Quality AssJlr-
Bronx, North Central Bronx ronto, 227-228 ance Committee, On-
Hospital, nurse-midwife rate of cesarean delivery, pro- tario, 252
study, 228, 232 gram to reduce, 38, 149 Cesarean delivery
trial, EMF versus ausculta- in breech presentation, 85-
tion, 104 87
C Carder, Angela, case of, 197- fetal risks,· 80
California 198 maternal risks, 80-81
Collaborate Cesarean Appro- Cardiac conduction defects, 107 distribution in patient de-
priateness Review and Catheter, intrauterine pres- mand for, 207-208
Education Project, 247- sure, for measuring con- elective
249 tractions, 32, 54 in macrosomia, 120, 215-
hospital discharge data Centers for Disease Control 216
on breech presentation, 181 (CDC), report on na- primary, 210-212
risks of cesarean delivery, tional cesarean delivery primary, demanded by the
174 rate, 23, 37 patient in labor, 216-217
nurse-midwifery in Cephalopelvic disproportion repeat, 213-216
San Francisco study, 229 (CPD) repeat, versus trial of la-
Fresno study, 227 active management in, 152 bor, 60-61
266 Index

Cesarean delivery (cont.) Closure, single-layer, of uterine hip dislocation, association

third-party payment for, incision, 59-60 with breech presenta-
261 Cobbs v. Grant, 165 tion,66
error in estimating benefits Coercion indications for cesarean deliv-
of, 141 to attain patient compliance, ery in, 153-154
in fetal macrosomia, after the 201 Connecticut
Zavanellimaneuver,120 force, place of, in medicine, childbirth preparation, Yale
and incidence of neonatal 195 University Hospital,
HSV infection, 135 Collaborate Cesarean Appropri- 236-237
incision for breech delivery, ateness Review and Edu- Yale Nurse-Midwifery Prac-
80-81 cation Project, 248 tice,227
information needed for in- Collaborative Perinatal Proj- Contractions, 26-27
formed consent to, 166, ect, cerebral palsy data, and diagnosis of labor, 45
256 98,99 decelerations of fetal heart
repeat, 87, 151,213-216 Colorado rate during, 109, 110
decision analysis model, trial, EMF versus ausculta- measuring, 32
178 tion, Denver, 101, 102- Contraindications
elective, versus trial of la- 103 to conduction anesthesia,
bor,60-61 uterine rupture data, trial of VLBW fetus, 86
guidelines for, 257 labor, 53-54 to external cephalic version,
increase in, 6 Competence, maternal, to au- 121
international comparisons, thorize cesarean deliv- to morphine, for prolonged la-
11-12 ery,198 tent phase, 31
with macrosomia, indica- Compliance, with patients' in- to succinylcholine, for cesar-
tion for, 117 formed decisions, 200 ean delivery, 86
patient demand for, during Complications to thiopental, for cesarean de-
labor, 217-218 with external cephalic ver- livery, 86
rate of, 52 sion,83 to vaginal delivery in breech
and total rate of cesarean of fetal asphyxia, 97 presentation, 84
delivery, 52-53, 145 maternal See also Indications
timeliness of, 163, 167-168 breech presentation, Controversy, over cesarean de-
See also Rate of cesarean de- VLBW infants, 87 livery, nature of, 3-4
livery trial of labor study, 61 Convenience
Cesarean Prevention Move- postpartum, cost of, 174 physician's, and cesarean de-
ment (CPM), 4, 235 with transverse incisions, 80 livery, 176-177
Cesarean Support Education Computed tomography, for and rate of cesarean delivery,
and Concern (C/SEC), 4, evaluating breech pre- 39,176-177
235 sentations, 78, 84 and repeat cesarean delivery,
Cesarean Work Group, Califor- Computer, for data collection, 51
nia,247 with peer review, 146-147 of scheduling, 215
Chile, active management of la- Conflict Cord. See Umbilical cord
borin, 48 in maternal-fetal care, 200- Cost
Chloroform, historic use in 202 of cesarean delivery, 11, 12, 23
childbirth, 234 physician-patient, 220 1987 survey, 173
Chorioamnionitis, tachycardia Congenital anomalies postpartum complications,
associated with, 108 atrioventricular block, associ- 174
Clavicle ation with fetal brady- with recurrent genital her-
fracture of, in macrosomia, cardia, 107 pes infection, 134
118 and breech presentation, 65- versus vaginal delivery, 51,
injury to, in breech delivery, 66,79 146,155
79 central nervous system, 99 as a criterion in decision anal-
Clinical practice guidelines cerebral palsy associated ysis,184
(CPGs), for cesarean de- with, 98, 99 of health care
livery, 147, 162 and fetal heart rate variabil- California State Assembly-
Clinicians, perspective of, 192 ity, 108, 153-154 requested study, 247
Index 267

United States, 23-24 Angela Carder case, 198-199 Documentation, of prior inci-
of midwifery care, 224 in nurse-midwife practices, sion, 168
of vaginal delivery, 51, 146, 234 Dosages, oxytocin, 62, 117-
155,173 options available in delivery, 118
See also Economics 218-219 Doulas, 240, 242, 258, 262
Cost-containment, and mal- role of the physician, repeat presence of, and cesarean de-
practice, 214 cesarean deliveries, 218 livery rate, 154
Counseling, for patients request- Decision tree Texas study, 241
ing elective primary cesar- examples of, breech presenta- Dublin,5
ean delivery, 212 tion, 182 cesarean deliveries in, and
Court order costs and probability esti- perinatal mortality, 18
for cesarean delivery mates, 183 labor support in, 242-243
frequency of, 208 cost-effectiveness, 185 National Maternity Hospital,
outcome of, 195-196 folding back of, 180 5,43-48
for intervention in a preg- Delahunt v. Fenton, 170 midwifery in, 224
nancy, 195 DeLee-Hillis fetoscope, 96 trial, EMF versus ausculta-
In re: Angela Carder, 196- Delivery tion, 103
199 database on, Mount Sinai trials of labor after cesarean
Courts study, 160-162 deliveries, 53
on cesarean delivery, 211-212 intrapartum management Dyskinetic syndromes, with ce-
intervention by of fetal macrosomia, rebral palsy, 97
AAP position on, 201 117 Dystocia, 23-40, 216
ACOG position on, 201- normal, cost of, 173 active management in, 152
202 See also Labor in breech presentation, 75
AMA position on, 199-200 Denmark cesarean delivery in, 6
reform of tort system, sug- trial, EMF versus ausculta- guidelines for, 255
gestions, 215 tion, 104 indication for, 145
Cragin's dictum, 12 rate of cesarean delivery, 14 international comparison,
Culture Descent, of the head 14
and childbirth, 48,194 arrest of, 28 clinically correctable, 39-40
and perspective on medical failure of, guidelines for ce- defined,24-25
ethics, 191 sarean delivery, 255 shoulder, 57
and rate of cesarean delivery, and progress in labor, 46 with macrosomia, 118-119,
15-16,151,154 Developing countries, issues in, 120
Cytomegalovirus, heart block 192,193-194
caused by, 107 Diabetes, risks in
Czechoslovakia, rate of cesar- offetal macrosomia, 115 E
ean delivery, 9 of shoulder dystocia, 118, Economics
120 of cesarean delivery, 173-188
Diagnosis hospital incentives for cesar-
D offetal macrosomia, 57,116- ean delivery, 177-178
Decelerations 117 and rate of cesarean delivery,
evaluating of labor, 44-45 15,17
in breech presentation, 84- specific, and cesarean deliv- See also Cost
85 ery rate, 36-38 Education, childbirth, 234-240,
historic pattern, 100 of uterine rupture, 54 258
late, during contractions, 110 differential by midwives, versus physi-
patterns of, 109-111 between normal latent cians, 225
Decision analysis model phase labor and secon- for physicians, 259
decision grid, 210-218 dary arrest of dilatation, availabilityof,143
decision tree, 178-187 30 guidelines for training
Decision making between true and false la- in delivery methods,
ACOG and AAP positions, bor,30-31 144
conflicts in maternal- Disability payment, after cesar- for primigravida and part-
fetal care, 200-202 ean delivery, 261 ners, 43-44
268 Index

Education, childbirth External review, and rate of ce- in breech presentation, 72-
(cont.) sarean delivery, 38. See 75,85
in vaginal birth after cesar- also Audit and cesarean delivery rate,
ean section, Michigan, 13,256
250 in external cephalic ver-
Electrocardiograph (ECG), fe- F sion,82
tal,96 Failure to progress in labor randomized controlled tri-
Endometritis guidelines for cesarean deliv- als,101-106
and cesarean section rate, ery,257 periodic changes in baseline,
133 reasons for, 25-26 109-111
in vaginal delivery, 79 Family, involvement of sinusoidal baseline pattern,
Endomyometritis, in cesarean in decisions about maternal- 108
delivery, 80 fetal care conflict, variability in, 108-109
England. See Britain/England! 201 See also Bradycardia; Decel-
United Kingdom at labor and birth, 225 erations; Tachycardia
Entrapment, in breech presen- Family physicians Fetal-pelvic index (FPI), 118
tation,79 expectant style of practice predictive value of, 121
with cesarean delivery, 87 by,232 Fetal station
Episiotomy interventions by, in high-risk plotting, as a function of
in breech presentation, 71, 79 settings, 233 time,25-26,27
utilization of, nurse-midwife Fat, body, in macrosomic fe- rate of change in, 26
versus physician deliver- tuses,117 abnormal, 28
ies, 224, 225 Father Fetus
Error, diagnostic involvement of, in decisions interest of, 198
interobserver error, 34-35 about cesarean delivery, rights of, 165, 219
in failure to progress in labor, 219 mother's and physician's
28-36 involving maternal-fetal perceptions, 194
Ether, historic use in child- care conflict, 201 cesarean delivery of, in
birth,234 support from, during deliv- breech presentation
Ethics ery,225,242 risks, 80-81
defined, 191 Fear, and request for cesarean at term, 85-86
issues of, 191-202 delivery,211-212 Finland
See also Fetus, rights of Fetal distress, 95-111 nurse-midwife study, 228
Ethics committees, 192 defined,99 prevalence of cerebral palsy
Expectant management diagnosis of, nurse-midwives in,99
alternative to oxytocin ad- versus physicians, 229 Florida
ministration, 55 fetal heart rate indicators of, legal requirement for assess-
style of practice by family verification, 152 ing feasibility of vaginal
physicians, 232 hospital facilities for detect- birth,249
Expected value, in decision ing,16 nurse-midwife study, 227
analysis, 180 increase in cesarean rates Force, place of, in medicine,
External cephalic version, 52- dueto,6 195
53,82-83 indication for cesarean deliv- Forceps delivery, 35
attempted, and cesarean de- ery in, 62, 146, 150 Foreign bodies left in patient
livery rate, 184 international comparison, after cesarean delivery, 169-
evaluating, decision analysis 13 170
model,180-181 intrapartum, management after surgery, 163
and macrosomia, 121 of,107 Fractures during delivery
after a previous cesarean de- Fetal heart rate (FHR) cesarean, breech presenta-
livery, 59 acceleration patterns, 109 tion, 80-81, 87
rate of success, 52,145-146, and asphyxia, 100 of macrosomic fetuses, 118
150,256 baseline, 107-108 Freedom of choice, in a just so-
with vertexlnonvertex twins, monitoring of ciety, 194
126 in active labor, 36, 77-78 Fundal height, and fetal weight
See also Breech presentation by auscultation, 95-96 estimation, 116
Index 269

G H Hospitals
Genital herpes, 131-137 Head monitoring of, California leg-
cesarean delivery with, 153 entrapment of, 79, 87 islation, 249
guidelines, 256 hyperextension of rate of cesarean delivery
Georgia, nurse-midwife study, with breech presentation, facilities of, 16, 149
226-227 78-79 incentives affecting, 155,
Glucose tolerance test (GTT), and choice of delivery 177-178
abnormal, correlation route, 84 Humerus, injury to, in breech
with fetal macrosomia, Health maintenance organiza- delivery, 79
115 tion (HMO), rate of ce- Hungary, rate of cesarean de-
Goals sarean delivery, 175-176 livery, 9
of childbirth education, Health Research Group, esti- Husband-coached childbirth,
239 mate of unnecessary ce- 235
for reduced rate of cesarean sareans,4 Hyperbilirubinemia, in choreo-
delivery, 37, 248, 255 Heart block athetoid cerebral palsy, 97
Graphing, to follow labor, 25- caused by cytomegalovirus, Hyperglycemia, maternal, and
26,45-46 107 fetal macrosomia, 115
breech presentations, 71 fetal heart rate variable decel- Hyperthyroidism, maternal
cervicographic analysis, erations with, 110 cerebral palsy associated
119 Hemolytic disease, cerebral with,99
National Institutes of Health palsy associated with, 98 tachycardia associated with,
on,37-38 Hemorrhage 108
Greece, electronic monitoring intracranial, in breech deliv- Hypnosis, to manage pain of
study, Athens, 105 ery,79 childbirth,235
Growth retardation, intrauter- intraventricular Hypoglycemia, fetal bradycar-
ine,65 in low birthweight twins, dia associated with, 107
in HSV-2 infection, 135 126 Hypotension, maternal, fetal
in twin gestations, 125 in preterm delivery, 152 heart rate variable decel-
See also Birthweight Herpes simplex virus (HSV-1 erations with, 110
Guatemala, study of labor sup- and HSV-2) Hypothermia, fetal bradycar-
port, 240-241 neonatal infection with, 132, dia associated with, 107
Guidelines 133-135 Hypothyroidism, fetal bradycar-
for cesarean delivery, 147, primary and nonprimary diaassociated with, 107
162,250 first episodes of infec- Hypoxia
in latent phase labor, 255 tion, 131 cerebral palsy associated
local,142 transmission risk, 134-135 with,97-98
timely performance of, High-risk patients defined,96
167-168 breech presentation, 88 perinatal death associated
utilization of, 255-262 in midwifery studies, 232-233 with,105
clinical practice, 147, 162 study of EFM in, 105 risk of, with oxytocin infu-
for determining dystocia, 24- randomized controlled sion,47
25 trial, 101-102, 104 tachycardia associated with,
for diagnosis of fetal dis- History 108
tress, 150 of cesarean delivery, 1-6 See also Asphyxia
suggested, in opposing ma- of childbirth education, 234- Hysterectomy
ternal refusal of cesarean 235 after complications of cesar-
delivery, 202 of fetal heart rate monitor- ean delivery, 80
for vaginal delivery ing,100 and cost containment, 24
in breech presentation, 12- of maternal mortality in ce- with placenta accreta, 87
13 sarean delivery, 95
after previous cesarean de- Home delivery, Netherlands, 9
livery,57,142-143,151, Homeostasis, fetal blood, in I
214 breech presentation, 74 Illinois
for vertexlnonvertex twin de- Hospital administrators, guide- study of childbirth educa-
livery, 126 lines for, 261 tion, Evanston, 238
270 Index

Illinois (cont.) postpartum, in cesarean de- Intervention

University of, rate of cesar- livery,174 court-ordered, 195-196
ean deliveries, 228 recurrent, HSV-1 and HSV-2, midwifery philosophy, 225
Incision for breech delivery, 131 minimizing, active manage-
80-81 tachycardia associated with, ment of labor, 43-48
infected, 80 108 and monitoring a pregnant
type of, with VLBW fetus, 86 wound, in cesarean delivery, woman's life-style, 196
See also Scar 80 Intrauterine growth retarda-
Indications Infectious Disease Society for tion, 65, 125, 135
for cesarean delivery Obstetrics-Gynecology, Intrauterine pressure catheter
cephalopelvic dispropor- on cesarean delivery in (IUPC),32
tion (CPD), 13-14, 29, genital herpes, 153 and diagnosing uterine rup-
84-85,117,145 Information ture, 54
congenital anomalies, 153- to pregnant women, about ce- Intravenous fluids, use of,
154 sarean delivery, 248 nurse-midwives versus
dystocia, 14, 145 public, on cesarean rates, 248 physicians, 224, 225
fetal weight, 117, 120 Informed consent, 163 In vitro fertilization, and rate
macrosonria,56-58,l17 medical-legal perspective, of cesarean delivery, 153
maternal age, 14-15, 153 165-167 Iowa, study of effects of child-
medical, international com- in prenatal intervention birth education, 237-
parison, 11-14 American Medical Associa- 238
meningomyelocele, 153- tion on,199 Ireland, breech delivery after
154 American College of Obste- previous cesarean sec-
nonmedical, international tricians and Gynecolo- tion, 59. See also Dublin
comparison, 14-18 gists on, 200 Ischemic injury, and choreoath-
placenta previa with macro- suggested guidelines, 202 etoid cerebral palsy, 97
somia,117 Injury Israel, rate of vaginal breech
prolapsed umbilical cord, brachial plexus, 79, 118 delivery after previous
84 fetal, time of occurrence of, cesarean section, 59
repeat, 56-57 167 Italy
vasa previa with macro- ischemic, and choreoathetoid malpractice claims in, 17
somia,117 cerebral palsy, 97 rate of cesarean delivery
for forceps delivery, 95-96 See also Anoxia; Hypoxia; and insurance, 17
for oxytocin use, 55 Neurologic impairment; and socioeconomic status,
See also Contraindications; Trauma 15
Fetal distress In re: Angela Carder, 196-199
Induced labor Insulin, exogenous, and fetal
breech presentation, monitor macrosomia, 116 J
tracings, 73, 74 Insurance Jackson Memorial Hospital,
decision analysis model, 178 payment for cesarean deliv- nurse-midwifery at, 232
and dystocia, 14 ery,51,261 Jamaica, rate of cesarean deliv-
failure of, and cesarean deliv- and rate of cesarean delivery, ery,9
ery,255 17 Japan, prevalence of cerebral
in fetal macrosomia, 117-118 International Cesarean Aware- palsy in, 99
Infection ness Network (lCAN), 4, Judicial system
cerebral palsy associated 235 functioning of, in prenatal in-
with,99 classes provided by, 239 tervention, 198
cytomegalovirus, 107 International Childbirth Edu- informed decisions by juries,
fetal heart block associated cation Association 214
with,107 (ICEA),235 See also Court order; Courts
genital herpes, stages of, International Classification of
131-132 Disease (ICD) codes, 2
maternal International comparisons, K
in electronic monitoring, rates of cesarean deliv- Kaiser Permanente hospitals
101,106 ery,9-19 cesarean delivery rate, 176
Index 271

reduction of, 151 randomized controlled Location, urbanlrural, and rate

at Riverside, 52-53 trial, EMF versus aus- of cesarean delivery, 17
Cesarean Work Group partic- cultation, 104-105 Low-risk patients
ipation, California, 248 in twin gestations, 125 randomized controlled trial,
nurse-midwives and physi- prolonged, 34-35 EMF versus ausculta-
cians at, working rela- indication for oxytocin use, tion, 102, 103
tionship, 233 55 rate of cesarean delivery, De-
study, elective repeat cesarean program for reducing inci- troit study, 231
versus trial of labor, 61 dence of, 43-48 Lung maturity, in fetal macro-
uterine rupture rate, after spontaneous, in breech pre- somia
prior cesarean delivery, sentation, 77 at delivery, 121
53 support during, 240-243, 258 and induction oflabor, 117
Kareml, Marjorie, 235 See also Active management Lutheran Hospital, La Crosse,
Kernicterus, and cerebral of labor (AML); Induced Wisconsin, 150
palsy, 97, 98 labor
Key v. Caldwell, 169 Lacerations, nurse-midwife ver-
Kleihauer-Betke test, following sus physician deliveries, M
external cephalic ver- 224,225 McKernon, John R., Jr., 249
sion,83 Lactate, maternal and fetal lev- McRobert's maneuver, 119
els, 36 Macrosomia, fetal, 115-121
Lamaze method, 235 evaluating, before elective re-
L effects of childbirth educa- peat cesarean delivery,
Labor tion, study, 237-238 215-216
active phase of, 27 Large-for-gestational-age as an indication for repeat ce-
effects of epidural anesthe- (LGA) fetus, 116 sarean, 56-58
siaon, 47 Lecithin/sphingomyelin (LIS) Macrosomic index, 116
oxytocin administration in, ratio, to document lung Magnesium sulfate, 108, 110
45 maturity, 117 Magnetic resonance imaging
prolonged, 35-36 Legislation (MRI), for evaluating
time limits on, 256 California State Assembly breech presentations, 84
arrest disorders, 31-34 Bill No. 524, 249 Maine, legislation on medicalli-
decision-making about cesar- California State Assembly ability, 249-250
ean delivery during, Bill No. 686, 248 Malpractice
193-194 Florida, law requiring assess- guidelines for attorneys,
diagnosis of, 44 ment of feasibility of va- 259-260
disorders of, 31-35 ginal birth, 249 and medical practice guide-
diagnosing, 229 Maine Public Law 1990, lines,249-250
objectively identifiable, 28 Chapter 931,249-250 rate of claims and size of
evaluating, 25-26 to reduce incidence of mal- awards, 163
objective definitions, 27-28 practice suits, frivolous, and trials oflabor, 61-62,
probability of diagnostic er- 261 213-214,215
ror, 29 Legislatures, guidelines for, See also Medical-legal issues
false, 30-31 261 Malpresentation
latent phase of, 26, 27 Leukomalacia, periventricular cerebral palsy associated
guidelines for cesarean de- and prenatal injuries, 99 with,98
livery in, 255 and spastic cerebral palsy, with fetal macrosomia,
normal, 29-30 97 117
prolonged,30-31 Linkage, in the hospital- Marijuana, historic use during
obstructed, contractility pat- physician relationship, childbirth, 234
terns in response to, 32 177 Maryland, rates of cesarean de-
preterm, 152 Lithotomy position, use of, livery in, 250
in maternal HSV-2 infec- nurse-midwives versus Massachusetts, information
tion, 135 physicians, 225 about hospital's cesarean'
in multiple gestations, Litigation. See Malpractice; delivery rate in, 250
127-128 Medical-legal issues Mauriceau maneuver, 72
272 Index

Medical components Meningomyelocele, indication role of, medical-legal perspec-

active management of la- for cesarean delivery, tive, 167-168
bor program, 44-48 153-154 scalp electrode, risk for HSV
indication versus patient Mercury toxicity, association transmission accompa-
demand, for cesarean de- with cerebral palsy, 99 nying, 135
livery, 211-212 Metropolitan Life Insurance use of, nurse-midwives ver-
Medical ethics, defined, 191 Company, cost compari- sus physicians, 224, 225
Medical-legal issues, 6, 163- son, cesarean versus va- Monoamniotic twinning, 127
170 ginal delivery, 146 Montevideo units
in cesarean delivery, 142, 146 Mexico City, rate of cesarean for evaluating protraction
example of, oxytocin use, 54- delivery, 9-10 disorders,35
55 Michigan for measuring uterine con-
litigation, and rate of cesar- certified nurse-midwife traction, 32
ean delivery, 17 study, 224-225 Moral responsibility, of the
malpractice reform recom- rate of cesarean deliveries, mother, 199-200
mendation, 155 Detroit, 231 Morbidity
medical practice guidelines, vaginal birth after cesarean following blood transfusion,
Maine, 249-250 delivery project, 250 80
and oxytocin infusion, 48, 62 Midwifery, 223-234 cesarean delivery associated
example, 54-55 in active management of la- with,11
in vaginal birth after cesar- bor,44 and decision analysis, 187
ean section, 51, 61-62 and cesarean delivery rate, 154 effect on, of cesarean reduc-
See also Court order; Courts; studies of, summary, 230- tion initiative, 148
Legislation; Malprac- 231 in external cephalic version,
tice; Tort reform choice of, and preference for 126
Medical Liability Demonstra- vaginal delivery, 219- maternal
tion Project (MLDP), 220 in breech presentation with
Maine,249 guidelines for, 257 abdominal delivery, 70,
Medical model of care, 223 and labor support, 242 71,83,125
Medical schools, emphasis on La Crosse, Wisconsin, 150 in cesarean delivery, 174
normal childbirth in, licensing of nurse-midwives, meta-analysis, and trial of la-
259 224 bor, 151
Medications in the Netherlands, 9 neonatal
acyclovir, for genital herpes, Minnesota analysis of, cesarean reduc-
136 guidelines, cesarean-related, tion initiative, 147
atropine, 108 250 and birthweight, 159
barbiturates, 31 study of nurse-midwife deliv- and breech presentation, ef-
benzodiazepines, 108 eries, Hennepin County fect of route of delivery,
insulin, 116 Medical Center, 226 70
phenothiazines, 108 Mississippi, University of, and herpes infection, 133
pitocin, 225 nurse-midwife prospec- perinatal
prostaglandin E2 (PGE 2), 60, tive evaluation, 226 in breech presentation, 65
255 Monitoring in shoulder dystocia, 119-
succinylcholine, 86 in breech presentation, 71-75 120
sympathomimetic, tachycar- electronic, 96 in twin births, 126
dia associated with, versus auscultation, ran- pulmonary complications in
108 domized trials, 101-106 cesarean delivery, 174
terbutaline, 108, 110, 168 in fetal macrosomia, 117 second twin, low birthweight,
thiopental, 86 of quality of obstetric care, 126
tocolytics, 31, 82 and rate of cesarean de- Morphine, contraindication to
utilization of, nurse- livery, 38 use of, 31
midwives versus physi- and rate of cesarean delivery, Mortality
cians, 224 256 in cesarean delivery, 11
Membranes, artificial rupture review of, to evaluate physi- effect on, of cesarean reduc-
of,45 cian performance, 260 tion initiative, 148, 150
Index 273

fetal data on electronic fetal moni- Neonatal service, participation

following external cephalic toring,96 in cesarean reduction ini-
version, 83 data on rate of repeat cesar- tiative, 147
with interlocking twins, 127 ean delivery, 145 Netherlands
maternal National Hospital Discharge cesarean deliveries in, in
with cesarean delivery, 71, Survey, 2 breech presentation, 12
174,181 National Consensus Conference home delivery in, 9
with breech presentation of Aspects of Cesarean mode of breech delivery in,
and cesarean delivery, 71 Birth (NCCACB), 149, 69
following external cephalic 251-252 rate of cesarean delivery in, 9
version, 83 National Consensus Develop- with cephalopelvic dispro-
historic pattern, 95 ment Conference, portion, 13-14
meta-analysis of, and trial of 142 midwifery study, 225-226
labor, 151 guidelines on cesarean deliv- Neurologic impairment
neonatal ery,147 accompanying breech presen-
in herpes infection, 133 recommendation on rate of tation, 70-71
for low birthweight twins, cesarean delivery, 141 and disorders of develop-
126 outcome of, 146 ment,98
See also Neonatal deaths National data on cesarean de- electronic fetal monitoring,
(NND); Perinatal mortal- livery rates, 9-19 and incidence of, 106
ity(PNM) National Hospital Discharge and parasagittal neuronal
Mother Survey (NHDS), 1 damage,99
effects on, vaginal versus ce- data for 1989, 3 prenatal causes of, 99
sarean delivery, 46-47 data on patients with prior after uterine rupture in trial
moral responsibility of, 199- cesarean delivery, of labor, 54
200 52 New York, city of
risk to, in vaginal delivery factors related to rate of ce- mode of breech delivery in,
with breech presenta- sarean delivery, 5-6 68
tion, 80-81 National Center for Health North Central Bronx Hospi-
See also Morbidity, mater- Statistics (NCHS), 2 tal, nurse-midwife study,
nal; Mortality, maternal National Institutes of Health 228,232
Mount Sinai, Chicago, cesarean (NIH) New York, state of
section reduction initia- cerebral palsy literature re- effects of childbirth educa-
tive, 146-149 view, 5 tion, 237
database, 159-162 Consensus Development labor disorders, rules for di-
Mueller-Hillis maneuver statement on cesarean agnosis,38
(MHM), 33-34 childbirth,37-38, 163 rates of cesarean delivery, re-
Multiparity, and risk of fetal Task Force on Cesarean view program, 250
macrosomia, 115 Childbirth, 1,223,251 review process, 151
Multiple gestations, 125-128 Natural childbirth, Dick-Read New Zealand, rate of cesarean
association with cerebral philosophy of, 235 delivery, 9
palsy, 98 Negligence, in cesarean deliv- Nonstress test (NST), after re-
and repeat cesarean delivery, ery, 168-169 jecting diagnosis of la-
52 Neonatal deaths (NND) bor, 45
and vaginal birth after cesar- and breech presentation, ef- Northwestern Memorial Hospi-
ean delivery, 59 fect of route of delivery, tal, cesarean reduction
Multivariate analysis, trial of 70,181,184 program, 149
labor outcomes, 61 and prematurity, 67 Nurse-Midwifery Service, Tuc-
Myomectomy, 151-152 See also Mortality, neonatal son, study of low risk pa-
Neonatal depression, with tients, 226-227
breech presentation, 86 Nurse-midwives. See Mid-
N Neonatal Research Network, wifery
Nader, Ralph, 251 National Institute of Nurses
National Center for Health Sta- Child Health and Hu- effect of, on rate of cesarean
tistics (NCHS) man Development, 152 delivery, 241
274 Index

Nurses (cont.) in breech presentation, 75- Northwestern Memorial Hos-

guidelines for, 258 77,84 pital,149
personal, for primigravida, for dilatation arrest, 32 quality of, 260
44 discontinuing, in variable de- See also Audit; External re-
responsibility of, for in- celerations, 110 view
formed consent, 167 dosages of, 62 Peer support, and decisions un-
for cervical ripening and la- der clinical uncertainty,
bor induction, 117-118 177
o for inefficient uterine contrac- Pelvimetry
Obesity, and risk of fetal mac- tion,45 clinical, for evaluating breech
rosomia, 115, 116 labor arrest during adminis- presentations, 84
Obstetric ethics, defined, 192 tration of, 31-32 for evaluating labor aberra-
Obstetrician, senior, active after premature rupture of tions,40
management of labor membranes, 55 prenatal, 33
program, 44 in prolonged labor, 36, 39-40 x-ray, for evaluating breech
O'Driscoll, Kieran, 43 for protraction disorders, 35 presentations, 78, 84-85
Oligohydramnios, 45 risks in use of, 46-47, 62 Pennsylvania, nurse midwifery,
and external cephalic ver- in trial of labor, 54-56 and rate of cesarean de-
sion,82 use of, nurse-midwives ver- liveries, 232
and variable decelerations, sus physicians, 224 Perinatal centers, rate of cesar-
110 in vertexlnonvertex twin de- ean deliveries at, 145
Opportunity cost, and physi- livery, 126 Perinatal (PN) deaths, rate of,
cian preference for cesar- 67. See also Perinatal
ean delivery, 175 mortality (PNM); Neona-
Outcome P tal deaths (NND)
analysis of, quality assurance Pain Perinatal morbidity. See Mor-
tool,155-156 control of, 85 bidity, perinatal
assigning values to, decision water therapy for, 235 Perinatal mortality (PNM)
analysis, 180 and evaluation of cesarean Booth Maternity Center
of breech presentation, by delivery,3-4 study, 232
size and prematurity, 68 Paperwork, and patient care, in breech presentation, 65,
California Hospital Out- 258 66-67
comes Assessment, AB Parametritis, in cesarean deliv- Dallas-Dublin study, 5
524,249 ery,80 following external cephalic
of cesarean delivery Parasagittal neuronal damage, version, 83
breech presentation, 180 99 international comparison,
court-ordered,195-196 Parkland Hospital, Dallas, 4 223
very low birthweight in- Partogram, 45-46. See also nurse-midwifestudy,226-227
fant, 174 Graphing and prolonged active stage of
by fetal characteristics and Patients labor, 36
delivery route, Mount Si- cesarean delivery demanded prospective study, 5
nai database, 159-162 by, 207-220 rate of
neonatal, cesarean reduction perspective of, in medical eth- and cesarean delivery rate,
programs, 148 ics, 191 18-19,223
perinatal, and rate of cesar- primary, fetus versus and prematurity, 67
ean deliveries, 247 mother, 195 in shoulder dystocia, 119-120
in preterm delivery, cesarean See also Autonomy, pa- studies of electronic fetal
versus vaginal route, tient's; Fetus; High-risk monitoring, 106
152 patients; Low-risk pa- VVormerveerstudy, 226
and rate of cesarean delivery, tients; Mother Perinatology, 164
18-19, 144-145, 175~ Peer review effect of, on rate of cesarean
176,247 of clinical practice guidelines, delivery, 143
trial of labor study, 61 147 Personal practice profile, 147
Oximetry, 31 for control of cesarean deliv- Phenothiazines, tachycardia as-
Oxytocin, 255 ery rates, 146, 154 sociated with, 108
Index 275

Phosphatidylglycerol (PG) postterm quality improvement

level, to document lung indication for oxytocin use, project, 251
maturity, 117 55 Quality of care, referral to spe-
Physicians and risk of fetal macro- cialists, 164-165
disclosure of personal inter- somia,115
est of, 167 Premature rupture of mem-
education of, in delivery tech- branes (PRM) R
niques, 143 in breech presentation, moni- Radiograph, abdominal, to ver-
factors affecting preference of tor tracings, 76 ify hyperextension of the
for avoiding VBAC deliv- fetal heart rate and contrac- fetal head, 84
ery, 51 tion pattern in, 72 Rate of cesarean delivery, 219,
for cesarean delivery, 17- as an indication for oxytocin 223
18,154,176-177 use, 55 with breech presentation,
economic, 175-177 Prematurity 146
fees of, vaginal and cesarean association with cerebral cultural factors in, 15-16,
delivery, 175 palsy, 98 151
guidelines for cesarean deliv- and breech presentation, 65- determination of, 1-3
eries,255-257 67 studies of midwifery, 225-
interactions with midwives, iatrogenic, 174 234
232-233,234 Prenatal care, time spent on, and dystocia, 6, 24
interactions with patients, 224-225 effect on
200 Private practice, and cesarean of childbirth education,
personal practice profile, 147 reduction programs, 149, 237-238
support of, for cesarean re- 150-151 of electronic fetal monitor-
duction, 150 Probability, assigning for un- ing,106
Piper's forceps, 78 certain events, 179 of repeat cesarean delivery,
Pitkin, Roy, 51 Professionals, perspective of, 51-52
Pitocin, use of, nurse-midwives in medical ethics, 191 of support during labor,
versus physicians, 225 Prospective study, 5 240-241
Placenta Prostaglandin E2 (PGE 2) in health maintenance organi-
abnormalities of for cervical ripening, 255 zations, 175-176
in breech presentation, 65 for induction after a prior ce- and HSV-2 infection, 133
in twin gestations, 125 sarean section, 60 international data, 10, 151
diagnosis of insufficiency, 45 Proteinuria, maternal, associa- in low-risk patients, Detroit
location of, and external ce- tion with cerebral palsy, study, 231
phalic version, 82 99 mechanism of nurse-midwife
Placenta accreta, and prior ce- Public, guidelines for, 261-262 influence on, 233-234
sarean delivery, 87 Public Citizen Health Research in multiple births, 125
Placenta previa Group, 251 in Ontario, Canada, 252
with fetal macrosomia, 117 Public Health Research Insti- at perinatal centers, 145
and prior cesarean deliveries, tute,250 primary, guidelines for reduc-
87 Public Health Service, U.S., ing rate of, 256
Polymerase chain reaction goal for reduced rate of reasons for change in, 5-6
(PCR), for identifying cesarean deliveries, 248 reduction of
HSV-2,132 Puerto Rico, rate of cesarean safe methods for, 141-156
Potter's syndrome, 202 delivery, 10 strategies for, 154-155
Precedent, for prenatal inter- U.S., 1968-1988,2
vention, 198-199 W ormerveer study, 226
Preeclampsia Q See also Cesarean delivery
cerebral palsy associated Quality assurance Referral in complicated births,
with,98 in cesarean reduction initia- 163, 164-165
in twin gestations, 125 tive, 147 midwifery study, 225-226
Pregnancy and outcomes, analysis of, Reimbursement, for cesarean
normality of, and cesarean 155-156 versus vaginal delivery,
delivery, 257 Vermont cesarean section 155
276 Index

Relaxation, to reduce pain in la- Rupture of membranes Sleep, effect of, on fetal heart
bor, 235 premature rate variability, 108
Reproductive ethics, defined, in breech presentation, Small for gestational age
192 monitor tracings, 76 (SGA), and breech pre-
Resident staff fetal heart rate and con- sentation, 70-71
active management of labor traction pattern, 72 Smedra v. Stanek, 169
program,44 indication for oxytocin use Social factors in cesarean deliv-
guidelines for, 259 in,55 eryrates
Respiratory distress syndrome and prolapse of the umbilical increased, 6
(RDS) cord,67 socioeconomic status, 15
effect of, on perinatal mortal- See also Culture
ity rate, 18 Sonography, for estimating fe-
in low birthweight twins, 126 S tal weight, 57
risk of, in elective cesarean Safety Spastic syndromes, with cere-
delivery,174 of acyclovir, in late preg- bral palsy, 97
Responsibility, by women, for nancy,136 Specialization, effects of, 143,
their own care, 225 of cesarean delivery, 4 164-165
Rest, therapeutic, during labor, of midwifery care, 224 SRI International, Cesarean
30 of vaginal birth after cesar- Work Group participa-
Rights ean delivery, 61 tion, California, 248
of bodily integrity, 198 of vaginal delivery, 46-47 Standard for informed consent
of the fetus, 165, 194 San Francisco, University of patient-oriented,165-166
and decisions about cesar- California at, 248 professional,165-166
ean delivery, 219 Saudi Arabia, rate of cesarean Standard of care
See also Ethics; Medical- delivery, 16 cesarean section-driven ap-
legal issues Scar proach,142
Risks rupture of, with prostaglan- medical-legal perspective,
of electronic fetal monitor- din use, 60 164-165
ing,106-107 type of, and vaginal birth Sterilization
factors in after prior cesarean de- and cesarean delivery, 15-16
with diabetes, 118, 120 livery,58,61-62,257 for retarded women, 200
in with fetal macrosomia, See also Incision Stories
115-116 Scheduling, and repeat cesar- patient's, 192, 194
with neonatal herpes, 135 ean deliveries, 215 and opposition to informed
long-term, of cesarean sec- Second opinion, 147-148 refusal, 202
tion, 80-81 Seizures physician's, 192
and outcomes, midwife care, in low birthweight twins, 126 Stress, and cesarean delivery
232-233 neonatal, Dallas-Dublin rate, 154
of oxytocin infusion, 46-47 study, 5 Stroke, in utero, association
Ritodrine, tachycardia associ- Selection of patients, for active with cerebral palsy, 99
ated with, 108 management of labor, 43 Succinylcholine, contraindica-
Roark v. Peters, 169 Selection bias tion to, 86
Rubin's maneuver, 119 in midwife care, and rate of Surgery
Rupture, uterine cesarean delivery, 224, cesarean delivery's unique at-
risk of 231 tributes, 218
with oxytocin infusion, 46- in studies of childbirth educa- emergent, decisions about, 219
47,62 tion, 235-236, 237 Sweden
with prostaglandin admin- Sensitivity analysis, 180 cesarean deliveries
istration, 60 Sepsis, fetal heart rate baseline in breech presentation, 12-
in trials of labor in,108 13
incidence of, 61-62 Serotyping, to classify genital elective primary, study of,
Los Angeles' County study, herpes, 132 212
53 Sherman Hospital, cesarean re- and perinatal mortality, 18
after previous cesarean sec- duction program, 150- uterine rupture rate, in trials
tion, 53-54, 179 151 oflabor,53
Index 277

Systemic lupus erythematosis, from external cephalic ver- Umbilical cord

maternal,107 sion,83 compression of
See also Injury in breech presentation, 72-
Trends in cesarean delivery, 74
T worldwide, 11 variable decelerations with,
Tachycardia Trial of labor (TOL) 109-110
associated with atropine, 108 in breech presentation, 71- prolapse of
in breech presentation, 72 77,78 in breech presentation, 65,
Technology effect on 75,77-78
and conflicts in maternal- of birthweight, 57 in external version, twin
fetal care, 200-201 of scar type, 58-59 births, 126
and patients' perspectives, ending on patient request, indication for cesarean de-
191 217-218 livery, 84
Terbutaline in fetal macrosomia, 117 and rupture of membranes,
to arrest uterine activity, 168 informed consent about op- 67
effect of, on fetal heart rate tion, 166 variable decelerations with,
variability, 110 with more than one prior ce- 110
tachycardia associated with, sarean delivery, 56 Uncertainty, and decision for
108 oxytocin use in, 55 cesarean delivery, 176
Texas patient United Kingdom. See Britain/
active management of labor refusal of, 213 EnglandlUnited
in,47-48 acceptance of, 214-215 Kingdom
rate of cesarean deliveries in, recommendation of ACOG, United States
Dallas, 228 179 breech vaginal delivery, 59
study of prostaglandin for and repeat cesarean rate, 52- cesarean deliveries
cervical ripening, 60 53,145,150,257 in breech presentation, 12-
support during labor, doula with twins, 59 13
study, 241 uterine rupture during, 53- issues surrounding, 193
Thiopental, contraindication 54 and perinatal mortality, 18
to,86 Triplets, 127-128 rate of cesarean delivery, 9,
Third-party payers, guidelines Truman v. Thomas, 166 251
for, 261 Tubal ligation, and the comply- and nurse midwifery, 226-
Thomas v. Ellis, 168 or-refer approach, 200 227,229
Thomas Jefferson University Tucson Nurse-Midwifery Ser- public health goal for, 248,
Hospital, 232 vice, 226-227 255
Threshold analysis, 184, 186- Twin gestation, 125-128 and socioeconomic status,
187 guidelines for cesarean deliv- 15
Time, economic value of, and ery, 256 study of out-of-hospital
route of delivery, 175 vaginal birth after cesarean births, 228
Tocolytics delivery, 59 United States Cesarean Birth
with external cephalic ver- Task Force, 13
sion,82 United States Standard Certifi-
sedative, with prolonged la- U cate of Live Birth, cesar-
tent phase labor, 31 Ultrasound ean data from, 1989,2-3
Tokodynamometry, 32 for detecting potential scar University of Southern Califor-
Toronto, nurse-midwife study, defects, 58 nia/Los Angeles County
227-228 during external cephalic ver- Medical Center, 58-59
Tort reform, 6, 58, 215. See also sion,82 Utah, uterine rupture data,
Legislation; Malprac- in fetal macrosomia, 120-121 trial of labor, 53-54
tice; Medical-legal issues for fetal size determination, Uterine malformation, and
Transmission, herpes simplex 33 breech presentation, 65,
virus, 134-135 for fetal weight estimation, 67
Trauma 116 Uterotonic stimulation, 30
in delivery of macrosomic fe- during vertex/nonvertex twin Utility, of maternal and neona-
tuses, 118-120 delivery, 126 tal survival, 184
278 Index

v Validity, agreement on criteria Washington, state of

Vaginal birth after cesarean for cesarean delivery, study of genital herpes man-
(VBAC) delivery, 51-62, 144 agement, 133-134
151 Vanderbilt University Hospi- trial, EMF versus ausculta-
disclosure of option, and in- tal, Labor and Delivery tion, 104
formed consent, 166 guidelines, 126 Washington Business Group
education project, Michigan, Vasa previa, with fetal macro- on Health, 251
250 somia,117 Water intoxication, risk of,
guidelines for, 142-143 Vermont with oxytocin infusion,
Northwestern Memorial Hos- cesarean section quality im- 47
pital program, 149 provement project, 251 Water therapy, to manage pain
nurse-midwife delivery, Yale nurse-midwife study, 227 of childbirth, 235
School of Nursing study, Vertex/nonvertex twins, man- Weight, excessive fetal, 117,
227 agement of delivery, 120. See also Birth-
rate of, 1, 52, 145, 214-215 125-127 weight
Centers for Disease Con- Very low birthweight (VLBW) Western blot assay, for HSV-2,
trol report on, 23 and breech presentation, 69, 132-133
Vermont study and follow- 79 Williams Obstetrics, 54, 209
up, 251 cesarean delivery in, 86-87 Wisconsin
Vaginal delivery outcomes of cesarean deliv- cesarean delivery reduction
in breech presentation ery, 152, 174 program, 150
evaluating the option, 83- See also Birthweight study of obstetricians' prefer-
85 Vibroacoustic stimulation, ence, 17-18
risks to the fetus, 77-79 109 Woods' maneuver, 119
risks to the mother, 79-80 Wormerveer study, 225-226
cost of, average, 173
Vaginal examination, for evalu- w
ating cephalopelvic dis- Wales, rate of cesarean deliv- Z
proportion, 33 ery,9 Zavanelli maneuver, 120