A. Riecher-Rossler, Meir Steiner Perinatal Stress, Mood and Anxiety Disorders From Bench To Bedside Bibliotheca Psychiatrica 2005

Perinatal Stress, Mood and Anxiety Disorders
Bibliotheca Psychiatrica
No. 173
Series Editors
A. Riecher-Rssler Basel
M. Steiner Hamilton
Perinatal Stress, Mood
and Anxiety Disorders
From Bench to Bedside
Basel Freiburg Paris London New York
Bangalore Bangkok Singapore Tokyo Sydney
Volume Editors
A. Riecher-Rssler Basel
M. Steiner Hamilton
11 figures, 1 in color, and 5 tables, 2005
Prof. Dr. med. A. Riecher-Rssler M. Steiner, MD, PhD
Psychiatrische Poliklinik Department of Psychiatry
Universittsspital Basel McMaster University
Petersgraben 4 Womens Health Concerns Clinic
CH4031 Basel (Switzerland) St. Josephs Healthcare
Hamilton, Ont. (Canada)
Bibliographic Indices. This publication is listed in bibliographic services, including Current Contents
and
Index Medicus.
Drug Dosage. The authors and the publisher have exerted every effort to ensure that drug selection and
dosage set forth in this text are in accord with current recommendations and practice at the time of publication.
However, in view of ongoing research, changes in government regulations, and the constant flow of information
relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for
any change in indications and dosage and for added warnings and precautions. This is particularly important when
the recommended agent is a new and/or infrequently employed drug.
All rights reserved. No part of this publication may be translated into other languages, reproduced or
utilized in any form or by any means electronic or mechanical, including photocopying, recording, microcopying,
or by any information storage and retrieval system, without permission in writing from the publisher.
Copyright 2005 by S. Karger AG, P.O. Box, CH4009 Basel (Switzerland)
www.karger.com
Printed in Switzerland on acid-free paper by Reinhardt Druck, Basel
ISSN 00678147
ISBN 3805578652
Library of Congress Cataloging-in-Publication Data
Perinatal stress, mood, and anxiety disorders : from bench to bedside /
volume editors, A. Riecher-Rssler, M. Steiner.
p. ; cm. (Bibliotheca psychiatrica ; no. 173)
Includes bibliographical references and indexes.
ISBN 3-8055-7865-2 (hard cover : alk. paper)
1. Postpartum psychiatric disorders. 2. ChildbirthPsychological
aspects. 3. MotherhoodPsychological aspects. 4. Postnatal
carePsychological aspects. 5. MothersMental health. 6. Stress
(Psychology) 7. Anxiety in women. 8. Postpartum depression. 9. Mental
illness in pregnancy. 10. Pregnant womenMental health.
[DNLM: 1. Postpartum Periodpsychology. 2. Anxiety Disorders. 3. Mood
Disorders. 4. Pregnancy Complicationspsychology. WQ 500 P4456 2005] I.
Riecher-Rssler, Anita. II. Steiner, Meir. III. Series.
RG850.P475 2005
618.76dc22
2005006482
Bibliotheca Psychiatrica
Formerly published as Abhandlung aus der Neurologie, Psychiatrie, Psychologie und
ihren Grenzgebieten (Founded 1917)
Edited by K. Bonhoeffer, Berlin (19171939), J. Klaesi, Bern (19481952), J. Klaesi
and E. Grnthal, Bern (19551967), E. Grnthal, Bern (1968), E. Grnthal and
Th. Spoerri, Bern (19691971), Th. Spoerri, Bern (1973), P. Berner and E. Gabriel,
Wien (19751986), B. Saletu, Wien (19862003)
V
Contents
VII Foreword
Riecher-Rssler, A. (Basel); Steiner, M. (Hamilton)
1 A Historical Perspective on the Psychiatry of Motherhood
Brockington, I. (Bredenbury)
6 Diagnostic Classification of Perinatal Mood Disorders
Riecher-Rssler, A.; Rohde, A. (Basel)
28 Maternal Adversity,Vulnerability and Disease
Matthews, S.G. (Toronto); Meaney, M.J. (Montral)
50 Behavioral Perinatology
Wadhwa, P.D. (Irvine, Calif.)
70 Maternal Depression: An Adverse Early Environment
Beach, A.J.; Henry, A.L.; Stowe, Z.N.; Newport, D.J. (Atlanta, Ga.)
85 Perinatal Infanticide and Suicide
Spinelli, M.G. (New York, N.Y.)
100 Relevance of Gonadal Hormones to Perinatal Mood and
Anxiety Disorders
Ahokas, A.; Kaukoranta, J.; Wahlbeck, K.; Aito, M. (Helsinki)
112 Pharmacotherapy for Psychiatric Disorders in Pregnancy
Ross, L.E.; Gunasekera, S.; Rowland, M.; Steiner, M. (Hamilton)
137 Maternal Depression in the Postpartum Period:
Impact of Breast-Feeding on Treatment Planning
Stowe, Z.N.; Ragan, K.; Newport, D.J. (Atlanta, Ga.)
150 The Use of Interpersonal Psychotherapy for Perinatal
Mood and Anxiety Disorders
Stuart, S.; OHara, M.W. (Iowa City, Iowa)
167 Group Psychotherapy for Depression in Early Stages
of Motherhood
Hofecker-Fallahpour, M.; Riecher-Rssler, A. (Basel)
182 Alternative Treatment Strategies for Perinatal
Depression and Anxiety
OHiggins, M.; Glover, V. (London); Corral, M. (Vancouver)
194 Author Index
195 Subject Index
Contents VI
Foreword
Mood and anxiety disorders are very common during the perinatal period
and confront us with specific needs. Thus, help seeking is often delayed due to
shame and stigma, and diagnosis is often missed due to misinterpretation of
symptoms. Services often do not adequately meet the needs of the women
concerned as they do not take into account their specific situation, problems
and fears. Untreated, peripartum disorders can have especially severe long-term
consequences, not only for the mother but also for the whole family and espe-
cially for the child. Recent research has shown how maternal stress, anxiety and
depression can adversely influence a childs early and later development.
Despite this, perinatal mental health has until recently been very much
neglected, overlooked, underdiagnosed and undertreated.
A recent surge of clinical and research interest, prompted by some very
active pioneers in this field and also by media exposures of several very tragic
outcomes of mothers who committed infanticide and/or suicide during an
episode of postpartum depression, is a most welcome change in this field.
It is now widely accepted that mental disorders in the perinatal period need
special attention and special treatments with modifications of the classical
pharmacological, non-pharmacological and psychotherapeutic approaches.
Universities around the world have started research programmes and imple-
mented subspecialty clinics for perinatal mental health. Residency programmes
in psychiatry and in obstetrics and gynaecology are including the topic in their
curriculum and granting agencies have declared it a priority for research.
We have assembled a group of internationally renown experts in the field to
contribute to this volume. The twelve articles included here offer a comprehensive
VII
up-to-date overview of the most relevant research and treatment considerations in
this neglected field.
A brief, though very interesting introduction to the history of psychiatry of
motherhood is followed by the discussion of the old question if perinatal disor-
ders are specific entities and if they should have a specific place in our classifi-
cation systems.
Further articles focus on the effects of chronic maternal stress on fetal
developmental outcomes and on the relationship between adversity in early
pre- and postnatal life and the risk for later cognitive and behavioural impairments
and chronic illness. Maternal depression during pregnancy and early childhood
is shown to be one of the childs earliest adverse life events which can con-
tribute to a childs vulnerability to later psychiatric illness. Both clinical as well
as preclinical evidence is discussed.
In an article of outstanding importance for each clinician, the danger of
perinatal infanticide and suicide is reviewed.
Further articles deal with specific forms and modifications of therapies for
perinatal mental disorders. The relevance of gonadal hormones in the pathogen-
esis of perinatal mental disorders is discussed, an innovative field of research
which could offer new therapeutic possibilities. The most important questions
of pharmacotherapy during pregnancy and breast-feeding are discussed in
depth. New psychotherapeutic approaches are presented such as interpersonal
psychotherapy as well as a new form of group therapy for mothers with depres-
sion and anxiety disorders. Last but not least, alternative treatment strategies
for pregnant and breast-feeding women are reviewed.
This book is aimed primarily at clinicians, teachers and researchers from
the fields of psychiatry (adult as well as child and adolescent psychiatry),
obstetrics and gynaecology, paediatrics, psychology, psychotherapy, neurobiol-
ogy and psychoneuroendocrinology as well as their students and learners.
The editors wish to thank all contributors as well as our administrative
assistants, Ms. Cindy Tasch, Hamilton, Mrs. Elisabeth von Castelmur and
Ms. Brigitte Howald, Basel, and the staff of S. Karger AG, Basel, for their
professional help with the publication of this volume.
Anita Riecher-Rssler
Meir Steiner
Foreword VIII
Riecher-Rssler A, Steiner M (eds): Perinatal Stress, Mood and Anxiety Disorders.
From Bench to Bedside. Bibl Psychiatr. Basel, Karger, 2005, No 173, pp 15
A Historical Perspective on the
Psychiatry of Motherhood
Ian Brockington
Bredenbury, Herefordshire, UK
The early literature on the psychiatric disorders of reproduction was
focused on psychosis. Hippocrates [1] may have encountered postpartum delir-
ium complicating infections, as in the women from Thasos and Cyzicus (3rd
book of epidemics, cases 2 and 14), and possibly the enigmatic 40th aphorism
from the 5th book of aphorisms. The next case was published in Basel about
2,000 years later in 1593 by Plater [2]. In the 17th and 18th centuries, there
were 3040 other brief reports, before the Gttingen obstetrician Osiander [3]
(1797) wrote the first full description of puerperal mania. A 3rd form of
psychosis posteclamptic psychosis was described by Mercatus [4] in 1614.
At the same time, Society became aware of another group of disorders,
apparently a disturbance of the maternal instinct. Murder of the newborn
became a major public health problem between the 17th and mid 19th centuries
and led to the development of a whole branch of forensic medicine. The fury of
the mother was shown in some cases by extreme violence, such as beheading or
butchering the infant. Almost all cases followed clandestine illegitimate births,
but occasionally neonaticide was reported within marriage, e.g. Biermann [5]
(1839). Simultaneously, parental child abuse came to light, initially in older
children. Brutal assaults betrayed parental rage, but children dying from starva-
tion first described by Rothamel [6] in 1845 indicated persistent states of
hatred and rejection. Tardieus [7] classic paper (1860) showed that this cruelty
could extend to very young children. Of his 32 cases, 5 concerned breast-fed
infants and 1 was only 15 days old. He wrote:
When we consider the tender age of these poor defenceless beings, subjected daily and
almost hourly to savage atrocities, unimaginable tortures and harsh privation and when we
face the fact that their tormentors are the very mothers who gave them life, we are confronted
with one of the most appalling problems that can disturb the soul of a moralist, or the con-
science of justice.
Brockington 2
A year later, Boileau de Castlnau [8] coined the term misopdie (hatred of
children) to denote this maternal disorder. Just as the phenomenon of child
abuse met obstinate resistance, until forced on the profession by advances in
paediatric radiology nearly 100 years later, child hatred is still resisted by many
practitioners, who prefer to regard it as a manifestation of postnatal depres-
sion. A paper by Luft [9] in 1964 indicated that these disorders were common:
he collected 40 cases of Wochenbettdepression all following legitimate, not
unwanted, pregnancies in which the most prominent symptom was hostility to
the newborn. In the USA, Robson and Moss [10] (1970) made a survey and
found that a delay in the maternal emotional response was found in about
10% of mothers.
From the standpoint of psychological medicine, childbearing is the most
complex event in human experience. This is reflected in the large number of
disorders described. Plater [2] may have been the first to observe obsessions of
fetal and child harm, and Woodward [11] described another case in 1757.
Kirkland [12] (1774) described the delirium that sometimes complicated the
most painful stage of parturition, before analgesics were introduced. Moll [13]
(1920) drew attention to the prevalence of persistent anxiety in the care of
young children, which he called Maternittsneurose. De Armond [14] (1954)
and Weightman et al. [15] (1998) described specific forms of anxiety, and
Sved-Williams [16] (1992) wrote about phobia of the newborn. Bydlowski and
Raoul-Duval [17] (1978) drew attention to post-traumatic stress disorder fol-
lowing prolonged and painful parturition, and a parallel state of Querulantenwahn
or bitterness disorder is also seen.
As for postpartum depression, some claim that the oracular remark of
Trotula of Salerno [18], an 11th century professor of medicine or a 13th century
midwife, is the first recognition of this entity.
If the womb is too moist, the brain is filled with water, and the moisture running over
the eyes, compels them to involuntarily shed tears.
A clearer, though brief, description was given by Joao Rodrigues de
Castello Branco (Amatus Lusitanus) [19] a Portuguese physician practising in
Rome in 1547.
The beautiful wife of Carcinator (a merchant) who always enjoyed the best of health,
was attacked after childbirth by melancholy, and remained insane for a month, but recovered
with treatment.
About the same epoch, Plater [2] described a melancholic country lass,
who, when breast-feeding, hanged herself, but was cut down by one that chanced
to see her. Other cases were described by Ledelius [20] (1696) and Schulzius
[21] (1705). As reports of puerperal melancholia began to multiply, recurrent
A Historical Perspective on the Psychiatry of Motherhood 3
prepartum and postpartum depression became a focus. Esquirol [22] (1818)
observed a woman who developed depression during each of 5 pregnancies, and
about a dozen other authors reported multiple episodes, culminating in Menzies
[23] (1893) whose patient had 12 episodes. As for postpartum depression,
Castro [24] (1617) another Portuguese physician (practising in Hamburg)
described recurrent episodes in another merchants wife. Hoppe [25] (1893),
Meyer [26] (1901) and Cruickshank [27] (1940) all encountered mothers with
8 episodes of depression after childbirth.
Interest in puerperal melancholia was sharpened by reports of melancholic
filicide. Calmeil [28] described a case which occurred in 1557, and Hume [29]
(1797) another that occurred in Aylesbury in 1668. Dapping [30] (1823) and
Teilleux [31] (1865) described mothers who killed all 3 infants. There are two
relevant French theses: Dumas [32] (1892) wrote on Libricide or murder of
small children by their parents: 47 of his 92 cases (collected from the litera-
ture) involved the mother alone. There were many cases of child abuse, but
some mothers were melancholic. Equally common is combined suicide and fil-
icide. Dumas described a woman reduced to penury, who sold everything to
provide her children with new clothes and a farewell feast, then, before attempt-
ing suicide, strangled all 5 of them one by one while they were sleeping.
Perrussel [33] (1923) wrote his thesis on altruistic homicide committed by
melancholic patients. Depressive filicide is probably the commonest setting for
the killing of a child more than 24 h old. In Denmark, Harder [34] (1967) stud-
ied 92 parents (84 of them mothers) who poisoned their children: 82 completed
suicide.
After the Second World War, psychiatry turned its attention to milder,
more frequent disorders. The Gordons of New Jersey [35] deserve the credit for
drawing attention to the frequency of milder forms of emotional disturbance
after delivery. They studied the aetiology, pioneered follow-through studies and
conducted a prophylactic trial. Main [36], in 1948, was the first to introduce a
child into the psychiatric ward where its mother was receiving treatment. This
courageous initiative led to the development of psychiatric mother-and-baby
units which, particularly in England and Australia, have accelerated the
growth of knowledge, as illustrated by the brilliant contributions of the late
Prof. Channi Kumar, who founded the largest of these units at the Bethlem
Royal Hospital in London.
The multitude of psychiatric disorders seen in pregnancy, parturition and
the puerperium is due to the complexity of the life event itself, with somatic,
social and psychological components. The pioneers were general physicians on
the continent of Europe, but the full picture has emerged through the labours of
many disciplines. The study of these disorders has now spread to every corner
of the world.
Brockington 4
References
1 Hippocrates. English translation by WHS Jones. Cambridge, Heineman, 1962.
2 Plater F: Praxeos medicae. Basel, Conradi Waldkirchii, 1602, 1656.
3 Osiander FB: Neue Denkwrdigkeiten fr Aerzte und Geburtshelfer. Gttingen, Rosenbusch, 1797.
4 Mercati L: Consultationes, observation 30, 1614.
5 Biermann LCA: Gerichtsrztliches Gutachten ber ein von einer Ehefrau heimlich gebornes und
bald nach der Geburt durch Erstickung gettetes Kind. Henkes Z Staatsarzneik 1839;38:303322.
6 Rothamel, of Fulda: Eine Mutter fhrt durch allmlige Entziehung der Nahrungsmittel den Tod
ihres ehelichen Kindes herbei. Henkes Z Staatsarzneik 1845;50:139156.
7 Tardieu A: tude mdico-lgale sur les vices et mauvais traitements exercs sur des enfants. Ann
Hyg 1860;15:361398.
8 Boileau de Castlnau P: Misopdie ou lsion de lamour de la progniture. Ann Md Psychol (Paris)
1861;7:553568.
9 Luft H: Die Wochenbettdepression. Klinik und pathogenetische Faktoren. Nervenarzt 1964;35:
185194.
10 Robson KS, Moss HA: Patterns and determinants of maternal attachment. J Pediatr 1970;77:
976985.
11 Woodward J: Autochthonous delusion (1757); in Hunter R, McAlpine I (eds): Three Hundred
Years of Psychiatry. London, Oxford University Press, 1963.
12 Kirkland T: A Treatise on Childbed Fevers and on the Methods of Preventing Them. London,
Baldwin & Dawson, 1774.
13 Moll L: Die Maternittsneurose. Wien Klin Wochenschr 1920;33:160162.
14 De Armond M: A type of postpartum anxiety reaction. Dis Nerv Syst 1954;15:2629.
15 Weightman H, Dalal BM, Brockington IF: Pathological fear of cot death. Psychopathology
1998;31:246249.
16 Sved-Williams AE: Phobic reactions of mothers to their own babies. Aust NZ J Psychiatry
1992;26:631638.
17 Bydlowski M, Raoul-Duval A: Un avatar psychique mconnu de la puerpralit: La nvrose
traumatique post-obsttricale. Perspect Psychiatr 1978;4:321328.
18 Trotula of Salerno (11th Century): The diseases of women. A translation of Passionibus mulierum
curandorum by Elizabeth Mason-Hohl. The Ward Ritchie Press, 1940.
19 de Castello Branco RJ: Second Century. 1551, 1620.
20 Ledelius S: Melancholia in praegnantibus. Miscellanea curiosa sive ephemeridum medico-physi-
carum germanicorum academiae naturae curiosorum, observation 28, 1696.
21 Schultzius G: De melancholia ex utero; in puerpera observata et curata; thesis, Frankfurt, 1705.
22 Esquirol JED: Observations sur lalination mentale la suite de couches. J Gn Md Chir Pharm
Fr trang 1818;1:148164.
23 Menzies WF: Puerperal insanity. An analysis of 140 consecutive cases. Am J Insanity 1893;50:
148185.
24 Castro R: De universa mulierum medicanovo et antehac a nemine tentato ordine. Hamburg, 1617,
p 314.
25 Hoppe H: Symptomatologie und Prognose der im Wochenbett entstehenden Geistesstrungen
(zugleich ein Beitrag zur Lehre von der acuten hallucinatorischen Verwirrtheit). Arch Psychiatr
Nervenkr 1893;25:137210.
26 Meyer E: Zur Klinik der Puerperalpsychosen. Allg Z Psychiatr 1901;58:700706.
27 Cruickshank WH: Psychoses associated with pregnancy and the puerperium. Can Med Assoc J
1940;74:571576.
28 Calmeil LF: De la folie, considre sous le point de vue pathologique, philosophique, historique et
judiciaire. Paris, 1845.
29 Hume D: Commentaries on the law of Scotland, respecting the description and punishment of
crimes. Edinburgh, Bell; in Hunter R, McAlpine I (eds): Three Hundred Years of Psychiatry.
London, Oxford University Press, 1963.
A Historical Perspective on the Psychiatry of Motherhood 5
30 Dapping S (Frankenthal): Gutachten ber den Seelenzustand der Charlotte Sorg, welche am 10ten
Mrz 1822 drei ihrer Kinder ttete. Henkes Z Staatsarzneik 1823;5:340375.
31 Teilleux: Triple infanticide. Rapport mdico-lgal sur ltat mental dAdle-Hlne Brevard-
Lacroix, femme Donnier-Blanc. Ann Md Psychol (Paris) 1865;5:419454.
32 Dumas E: Du libricide ou meurtre des enfants mineurs par leurs parents; thse, Lyon, 1892.
33 Perrussel G: LHomicide altruiste des mlancoliques et des perscuts; thse, Paris, 1923.
34 Harder T: The psychopathology of infanticide. Acta Psychiatr Scand 1967;43:196245.
35 Gordon RE, Gordon K: Some social-psychiatric aspects of pregnancy and childbearing. J Med
Soc NJ 1957;54:569572.
36 Main TF: Mothers with children in a psychiatric hospital. Lancet 1958;ii:845847.
Prof. emer. Ian Brockington
Lower Brockington Farm
Bredenbury
Herefordshire HR7 4TE (UK)
E-Mail i.f.brockington@bham.ac.uk
Diagnostic Classif ication of Perinatal
Mood Disorders
Anita Riecher-Rssler, Anke Rohde
Psychiatrische Poliklinik, Universittsspital Basel, Basel, Schweiz
About 2540% of mothers suffer from mood lability and mild depression
during the first week after parturition called postpartum blues or postpartum
dysphoria, about 1015% suffer from a depressive disorder and one or two of
every thousand women present with psychosis during the infants first year
[15; for reviews, see ref. 69]. These disorders were traditionally called post-
partum disorders.
But what is so special about these disorders that we even use a specific
diagnostic label for them? Do they have a specific aetiology? And, if not, is it
nevertheless justified to use this term, as these disorders confront us with
specific problems and needs?
Recently, attention has also been drawn to the fact that depression during
pregnancy is not less frequent than at any other given time in a womans life [10]
and that other disorders, especially anxiety and posttraumatic stress disorders,
can emerge in the postpartum period, at times with very unique presentations and
causing very specific problems.
Thus, the question to be put forward is whether these disorders should have
a separate entry in our classification systems.
Postpartum Depression and Postpartum
Psychosis Should We Keep These Diagnostic Labels?
Methodological Problems
The traditional textbook presentation of the diagnostic classification of
perinatal mood disorders usually includes a table labelled as the continuum of
Classification of Perinatal Disorders 7
postpartum emotional reactions listing maternity blues at one end, postpartum
depression in the middle and psychosis at the other end of the spectrum (table 1).
However, reviewing the literature, neither the diagnostic categories post-
partum depression nor postpartum psychosis or postpartum blues have so
far been exactly defined [for reviews, see ref. 6, 7, 11]. Thus, for example, psy-
chotic depression is in some studies categorised as psychosis, but in other stud-
ies as depression. Moreover, the time period postpartum is not unequivocally
defined, differing between 6 weeks [e.g. 12, 13] and 12 months [e.g. 14, 15].
While some studies refer to an incidence, i.e. new onset of a disorder during
this period, others refer to new-onset episodes and others to prevalence,
i.e. including pre-existing ongoing diseases as well. Postpartum blues is even
less clearly defined [for reviews, see e.g. ref. 6, 16, 17], with the consequence
of very strongly differing prevalence rates in the different studies.
This unclear definition and classification of postpartum disorders has lead
to severe problems in research. In fact, although using the same diagnostic
labels, different studies have examined quite different patient groups with the
consequence of quite inconsistent results. Thus, e.g. in a study examining women
during the first 6 weeks postpartum only, the kind and frequency of disorders
found will quite distinctly differ from the results of a study looking at the entire
postpartum year.
Inconsistent and contradictory results are of course not only due to the
described problems of definition and classification, but also to other method-
ological shortcomings of studies. Thus, only few studies have been based on
representative populations or examine control groups. Standardised diagnostic
systems have not always been used and some studies rely on self-assessments
only [e.g. 1820]. However, problems of definition and classification have in
our opinion so far been one of the main obstacles for research.
Table 1. Mental disorders in the postpartum period
Characteristics Frequency, %
Postpartum blues mild depression with mood 2540 (80)
lability and tearfulness, during the
first postpartum week, self-limiting
Postpartum all types of depressive disorders, during 1015
depression the first postpartum months/year
Postpartum depressive, manic, schizo-affective, 0.10.2
psychosis schizophrenic or atypical symptoms, during
the first postpartum months
Riecher-Rssler/Rohde 8
Current Classification Systems
The current diagnostic systems have different approaches to the classification
of postpartum disorders.
ICD-10 [21] suggests to code mental disorders associated with the puer-
perium according to the presenting psychiatric disorder, i.e. with the usual
classification number, while a second code (O 99.3) indicates the association
with the puerperium.
However, in exceptional circumstances, ICD-10 also allows a special code,
F53, to be used if there is insufficient information to classify according to given
classification criteria or if special additional features are noted (table 2). This
latter statement has opened the door for too many options with the consequence
that classification has become somewhat arbitrary. In fact, F53 is sometimes used
just because symptomatology such as the content of depressive thoughts is
influenced by the specific situation of motherhood. Theoretically, F53 only refers
to cases with onset within 6 weeks after delivery. However, it has also often been
used if the onset was earlier (pre-existing disorder) or later and in fact, ICD does
not clearly state if this time criterion refers to the first onset of a disorder only or
also to the onset of an episode of a recurrent disorder. This makes classification
even more problematic. Therefore, F53 should only be a rest category for cases
which do not fulfil the other criteria (i.e. depressive episode).
DSM-IV [22] only offers special classification possibilities for mood disor-
ders, i.e. a postpartum onset specifier can be applied to all mood disorders in
Table 2. ICD-10 classification of postpartum depression
Code
(1) Presenting psychiatric disorder
(2) O 99.3 for mental disorder complicating puerperium
Only use exceptionally
F53: Mental and behavioural disorders associated with the
puerperium, not elsewhere classified
commencing within 6 weeks of delivery
not meeting the criteria for disorders classified elsewhere
because of insufficient information or
because of special additional features
F53.0 Postpartum depression
F53.1 Postpartum psychosis
F53.8 Other disorders
F53.9 Unspecified
addition to standard classification. It refers to the current or most recent
episode if the onset of the episode lies within the first 4 weeks postpartum
(table 3). However, this specifier cannot be coded in DSM-IV.
There is no specific classification of maternity or postpartum blues as this
is a very brief, self-limiting episode with mild depression and mood lability
with minimal dysfunction. It occurs in the first days after delivery and usually
resolves within hours or days without any treatment. In the following sections,
we will therefore focus on postpartum depression and postpartum psychosis.
Specific Entities?
What is so specific about postpartum depression and postpartum psy-
chosis that these terms were coined? Are they really different diagnostic enti-
ties with descriptive, construct and predictive validity according to the criteria
of Spitzer and Williams [23]? If this were the case, they should e.g. show a
specific symptomatology, specific risk factors and a specific course [11].
Specific Symptomatology?
Postpartum depression does not show a very specific symptomatology, i.e.
no high descriptive validity. Patients with postpartum depression may rather
present with all different symptoms of the classical forms of depression [3, 24].
These symptoms are principally not different from the symptoms known to
occur with mood disorders not related to childbearing. However, addition-
ally, emotional lability similar to that seen in the blues has often been described.
Furthermore, having a newborn also influences the cognitive content of depres-
sion. Postpartum depressed mothers often feel guilty for not being good enough
mothers, are preoccupied with, at times, irrational worries about the well-
being of the infant and many of them report obsessional thoughts, e.g. of
harming their child [2527]. Many mothers complain about their inability to
have any warm feelings for their newborns. This is very often the reason to
seek help.
Table 3. DSM-IV classification of postpartum depression
Postpartum onset specifier
Can be applied to mood disorders only
Refers to current or most recent episode
Onset of episode within 4 weeks postpartum
Cannot be coded
In severe or psychotic depression, again, symptoms may be non-specific,
but in addition to the risk of suicide [28], there is also the added risk of infanti-
cide as part of an extended suicide [25] (also compare Spinelli, this volume).
All in all, symptomatology is not specific, but influenced by the specific
situation of being a mother of a newborn. In other words, the special situation of
the mother only plays a pathoplastic role.
The same seems to be true for postpartum psychoses [8, 2933]. Patients
with postpartum mania (usually as part of a bipolar disorder) may for example
present with delusions of grandiosity which are projected onto the child (e.g.
the baby is believed to be Jesus). Besides these postpartum-specific contents of
delusions, symptomatology of psychoses in the postpartum period usually does
not differ from that of psychoses occurring independently from parturition.
Also the frequency distribution of the different types of psychoses (affective vs.
schizophrenic psychosis and their subtypes) seems to correspond to the usual
distribution [3, 34]. Although a strong link to bipolar affective and schizoaffec-
tive psychoses has been noted [8, 35], it has not clearly been shown that these
disorders are overrepresented in women with newborns as compared to women
of the respective age group without delivery.
The majority of the so-called atypical psychoses with disorientation, con-
fusion and amentia, as often described some decades ago [36], might have been
of organic or exogenous origin due to obstetric complications, e.g. sepsis or
eclampsia [9]. With the progress in medical care, these presentations have
become very rare. If atypical psychoses still occur in the postpartum time, diag-
nosis should be made according to the diagnostic system used (e.g. delirium).
Specifically High Incidence or Prevalence?
The prevalence of postpartum depression depends on the diagnostic crite-
ria used and the time period observed. Looking at studies with standardised
diagnoses only, the reported prevalence rates vary between 6.5 [14] and 22%
[13, for review, see ref. 6]. Based on a meta-analysis of 59 studies, the mean
1-year prevalence rate is as high as 13% [37]. In studies which used self-rating
scales only, such as the Edinburgh Postnatal Depression Scale (EPDS) [38], the
estimated rates were even higher [e.g. 18, 19, 22]. However, probably not all of
those depressive syndromes would meet diagnostic criteria if a standardized
instrument for diagnosis was applied.
There has been some controversy about the question whether the risk for
depression is increased in the postpartum period. Most studies, however, find
prevalence rates comparable to that in the non-childbearing population [for
reviews, see ref. 6, 25]. OHara et al. [1], in a case-control study, compared
182 childbearing with 179 non-childbearing women matched for age, marital
status, job and number of children. Using research diagnostic criteria for major
depression, in the first 9 postpartum weeks no significant differences in preva-
lence were found. However, postpartum women suffered from milder sub-
threshold depression associated with physical complaints, sexual dysfunction
and partner problems. Cox et al. [39], who also conducted a case-control study
using research diagnostic criteria, could not find an enhanced prevalence
during the first year either, but he observed an enhanced incidence of depres-
sion in the first 5 postpartum weeks.
Thus, there might be an enhanced risk of mild depression and possibly also
a slightly enhanced incidence of major depression in the first postpartum
weeks; however, thereafter, risk seems to be similar to that of other women of
the same age group who according to the large epidemiological studies [e.g.
4042] have quite a high risk for depression anyway. The 1-year prevalence
rate of depression reaching the diagnostic threshold does not seem to be
enhanced.
This is obviously quite different in psychosis. The incidence of psychosis
in the postpartum period seems to be significantly enhanced. Kendell et al. [5]
reported a 22-fold increased risk in the first postpartum month and an only
slightly less increased risk during the next 2 postpartum months as compared to
the 2 years before childbirth. Terp and Mortensen [2] compared admission rates
of puerperal women (within 91 days after delivery) with admission rates among
non-puerperal women in the general Danish female population and found espe-
cially the risk of first admissions increased.
Specific Aetiology or Pathogenetic Mechanisms?
If postpartum disorders were specific entities, they would also show a
specific aetiology, i.e. a construct validity [23].
When the diagnostic labels postpartum disorders and postpartum psy-
chosis were introduced, some researchers hypothesized that these were specific
entities with specific aetiologies. Nowadays, most researchers have aban-
doned this theory. They rather agree that in the postpartum period, we see in
principle the same disorders that can also be seen independently from parturi-
tion [e.g. 3, 8, 9, 43]. There does not seem to be a specific aetiology in the strict
sense. However, there might be a specific psychoneuroendocrine vulnerability
in some of the mothers concerned. Moreover, delivery and the changes and bur-
dens of the perinatal situation might act as specific triggers.
The main risk factor which has been identified is an individual predisposi-
tion for the respective disorder quite independent of parturition. About one third
of all women with postpartum depression and about 15% of all women with
postpartum psychosis had already been suffering from similar episodes earlier
on even without a delivery, and there is often a corresponding family history [12,
13, 31, 34, 36, 44, 45; for a review, see ref. 6]. A further sign of the underlying
general vulnerability of many of these mothers is the risk of relapse, which also
exists independently of further deliveries (especially in psychosis), although it is
especially high just after further deliveries [4648]. This vulnerability mainly
seems to be genetically transmitted. Thus, genetic factors have been shown to
explain 2538% of the variance of postnatal depression in a twin study [49].
A vast amount of other risk factors and predictors have been found in differ-
ent studies, but have not consistently been replicated [for reviews, see ref. 68, 11,
50, 51]. Thus, women with postpartum depression as compared to women who
stay healthy after delivery show neither differences regarding actual hormonal or
obstetrical factors nor regarding psychosocial factors like age, family status,
education or socio-economic status. Apart from the individual, partly genetic
predisposition, the only additional predictors consistently found were anxiety and
depression already during pregnancy, baby blues, stress in child care or general
stress, little social support and marital problems [for meta-analyses, see ref.
37, 52, 53]. However, it is not clear to what extent these predictors are of patho-
genetic relevance and to what extent they are early consequences of the beginning
depression [7, 11].
Thus, the pathogenetic model for developing postpartum depression or
psychosis seems to correspond to the general vulnerability-stress model of
depression. If a woman is predisposed to depression or psychosis, normal par-
turition and normal postpartum changes can trigger the outbreak of the disorder
or enhance a pre-existing disease and also influence its symptomatology and
course. Giving birth to a child with all its consequences seems to act as a major
stressor not only in the sense of a psychosocially stressful life event, but also
in the sense of a biological stressor.
Concerning biological stressors, we know for example that during preg-
nancy, the oestrogen serum level is increased about 200-fold and drops to normal
values within a few days after delivery. If a woman is breast-feeding, values even
drop to subnormal. On the other hand, we know that oestrogens can modulate
many neurotransmitter systems like the serotonergic, the noradrenergic, the
dopaminergic, the GABA system and also influence many other brain activities
in a way that they have even been called natures psychoprotectants [5456].
The sudden drop of oestrogens might especially contribute to the mood
lability of the postpartum blues [6, 16] and to the outbreak of psychosis [55,
57]. However, it might also be relevant in postpartum depression, especially if
oestrogen deficiency is sustained during breast-feeding.
Interestingly, Bloch et al. [58] found that an artificial induction of a drop in
oestrogen levels provoked depression in 63% of women with a history of post-
partum depression, but not in women without such a history. This underlines the
importance of a specific vulnerability, but also of the oestrogen withdrawal as a
specific trigger.
On the other hand, there is probably hardly any life event which is
emotionally more stirring than giving birth to a child. In the period thereafter,
the mother has to cope with massive psychosocial changes in almost all areas of
her life, with multiple burdens, role changes and role conflicts, losses and
ambivalences. Thus, new motherhood can also be a massive psychosocial stres-
sor which can trigger the outbreak of a disease in the postpartum period. This
does not seem to be true for depression only, but also for psychosis. Thus, the
observation that in primiparae the risk of psychosis is even more severely
enhanced than in multiparae [5, 43, 47] could be a consequence of these
psychosocial stressors, which are probably higher for first-time mothers than
for experienced mothers.
All in all, while hormonal influences seem to be very important for
provoking blues and psychosis in (genetically) vulnerable women, psychosocial
factors might be of more relative importance in depression.
Specific Course?
The third prerequisite required for a specific entity would be a predictive
validity, e.g. a specific course. It has sometimes been reported that postpartum
psychoses show a slightly better course than disorders occurring at other times
of womens lives [31, 59]. However, long-term studies could not confirm this. It
could rather be demonstrated that course and outcome are similar to the respec-
tive diagnostic category without postpartum onset [60]. If there are any differ-
ences, they are not very marked and could be explained in the framework of the
vulnerability-stress model. Thus, giving birth and adapting to the new mother
role are obviously major stressors. If a disorder is provoked by such a massive
stressor only, the underlying vulnerability to the respective disorder might not
be very high. After waning of this stressor, the further course of these disorders
might thus be quite mild due to the relatively low underlying vulnerability.
In conclusion, it has to be stated that postpartum depression or postpartum
psychosis do not seem to be valid entities with a specific aetiology, symptoma-
tology or course. Nevertheless, depression as well as psychotic disorders can be
strongly influenced by parturition and early motherhood.
Specific Needs
Specific Consequences of Postpartum Disorders
Depression as well as psychosis in the postpartum period need our
special attention and specific treatment, as they can have particularly severe
consequences not only for the mother, but also for the infant and the whole
family.
We very often see vicious circles in these cases, which we have to interrupt
at a very early stage. Thus, the early mother-infant bonding is often severely
impaired, as the mothers concerned often have difficulties in making emotional
contact with their infants [61]. Depressed mothers realize this, react with feel-
ings of guilt, which enhances depression, and the vicious circle starts. In fact,
the children of these mothers may not only develop emotional and behavioural
difficulties, but also a delay of their cognitive development, especially if the
mother-infant interaction is disturbed [for reviews, see ref. 8, 6264]. Pawlby
et al. [65] presented very alarming results from a prospective study showing
cognitive deficits, especially in boys, even until the end of primary school.
Mothers with psychotic disorders might have even severer problems of
interaction. During acute psychosis, they might not only neglect the child, but
even show dangerous behaviour. In severe cases, they might harm or even kill
their own child, e.g. due to delusions or imperative voices, although this is very
rare [66] (see also Spinelli, this volume).
Specific Treatment Delay
Unfortunately, diagnosis and treatment of depressive disorders in the post-
partum period are often seriously delayed [67, 68]. Feelings of shame and many
fears prevent mothers from seeking help. Furthermore, they often do not realize
that they suffer from a disease. Depressed mothers might rather feel that they
fail as a mother. In a working class population in Scotland, McIntosh [69] could
show that only one quarter of the women concerned sought professional help.
Most women stated that medicine could not help anyway. They also feared to be
stigmatized as mentally ill or to be separated from their babies. Furthermore, in
some cases, doctors and midwives might not listen sensitively enough or misin-
terpret symptoms as stemming from normal exhaustion due to delivery and
child care.
Mothers with psychosis in the acute stages of their disease often show an
illness-inherent lack of illness insight. During phases of remission, they might
nevertheless be reluctant to seek help because of the realistic fear that the infant
could be taken away from her care.
A lack of specific, low-threshold treatment offers and facilities further
contributes to the treatment delay [68].
Specific Therapeutic Needs
Having the vulnerability-stress model as pathogenetic model in mind,
therapy must be directed not only to the underlying vulnerability, but also to the
actual biological and psychosocial stressors/triggers. This means that therapy
although in principle the same as the classical therapy for these disorders has
to put a special emphasis on the needs of the postpartum period.
In depressed mothers, education about the fact that this is a disease and not
due to failure as a mother is of utmost importance to relieve the patients from
feelings of guilt. Counselling and practical advice for the mother and her family
is often needed as well as the help of a midwife, a family nurse, a social worker
and/or other health care professionals.
A good mother-infant bonding and relationship should be a main concern
from the start [8]. This means that the mother should not be separated from her
infant for longer periods of time. However, she needs a lot of help in order to
allow for relaxed contacts with the infant. Baby massage and mother-infant play
therapy can be helpful, especially if there are bonding problems [25, 70, 71]
(see also OHiggins et al., this volume).
Psychotherapy should be initiated as soon as possible and should be very
supportive in the beginning. It has only recently been realized that this specific
situation might need specific psychotherapeutic approaches taking into account
specific problems such as the role changes of the women or the relationships
with the infant and the partner. Thus, e.g. Stuart and OHara [72] developed a
specific manualized interpersonal psychotherapy for women with postpartum
depression, whereas Riecher-Rssler and Hofecker-Fallahpour [11] and
Hofecker-Fallahpour et al. [73, 74] have developed a specific group therapy for
depressed mothers with young children (see also Hofecker-Fallahpour and
Riecher-Rssler, this volume).
Fathers should be involved as much as possible [7, 8], as father-infant
bonding and good care by the father seem very important, especially if the
mother suffers from recurrent illness. In some cases, also couple therapy might
be indicated, as we know that the relation of couples can be severely disturbed
in postpartum depression [1].
If medication is needed, breast-feeding has to be discussed with the
mother, as all psychotropics are secreted into breast milk (although to different
degrees) and effects of even trace amounts of medication on the developing
brain cannot be safely excluded, as so far there have been not enough method-
ologically sound long-term studies with sufficient numbers of cases [for
reviews, see e.g. ref. 10, 75, 76] (see also Stowe et al., this volume). Mothers are
often reluctant to stop breast-feeding, and it might be helpful to offer baby mas-
sage as a substitute regarding the physical contact part of breast-feeding and
potentially also regarding its positive psychological, physiological and hor-
monal effects (compare OHiggins et al., this volume).
However, the risks of ongoing psychosis or severe depression or a relapse are
certainly higher than the potential risks of medication [77]. Thus, if after balanced
information about the potential risks, a mother still decides to breast-feed, she has
to be educated about side-effects and the baby has to be carefully monitored
regarding these effects [75, 77] (for review, see also Stowe et al., this volume).
In the meantime, some studies have also shown positive effects of oestrogen
treatment in postpartum depression [7880] and psychosis [81] (see also Ahokas
et al., this volume) especially if there is a sustained postpartum oestrogen defi-
ciency, although large controlled studies are still lacking. Non-pharmacological
treatments such as bright light therapy could also be promising new approaches
[82] (see also OHiggins et al., this volume).
Hospitalization is often needed, especially if the safety of the mother or
infant cannot be guaranteed because of acute psychosis or severe depression.
However, mothers are often very reluctant, as they cannot bear the separation
from their child, and develop even more feelings of guilt. Moreover, the mother-
child relationship can be severely hampered by the separation. In some coun-
tries, starting in Great Britain [83], mother-baby units have therefore been
implemented, which allow the admission of mothers together with their infants.
These units can also offer evaluation of the quality of the relationship and care
for the infant as well as a respective training and therapy if needed. However, it
has to be stated that in most countries, the number of available mother-baby
units does not meet the needs. Domiciliary assessment, home treatment, day
hospital and day care centre, motherhood classes or playgroups for parents with
their children are also urgently needed.
Very important to note is also that women in the peripartum period have a
special need to be cared for by multidisciplinary teams including not only
psychiatrists and psychologists, but also gynaecologists/obstetricians, paedia-
tricians, midwives, social workers and sometimes also lawyers for medicolegal
advice or child protection agencies.
Specific Classification?
In conclusion, it has to be stated that neither postpartum depression nor
psychosis are specific entities which would need separate entries in classifica-
tion systems. There is neither evidence for a specific aetiology nor for a clearly
distinct psychopathology or long-term course and outcome [6, 11].
The prevalence of severe depression is not significantly enhanced in the
first year postpartum as compared to women of the same age group without
delivery who have quite a high prevalence of depression anyway. Only in the
first postpartum weeks might the prevalence of mild, subthreshold depression
be slightly enhanced and potentially also the incidence of depressive disorders.
It is mainly in the first postpartum months that first onset and probably also the
remanifestation of pre-existing psychosis seem to present excessively in vulner-
able women.
Giving birth and adapting to the new mother role seem to be stressors
which can trigger the outbreak of depression and especially of psychosis in
vulnerable, predisposed women or enhance a pre-existing disorder. Biological,
especially hormonal, as well as psychosocial factors can be relevant stressors in
this situation.
Although these are no specific entities, the diagnostic terms or better spec-
ifiers postpartum depression and postpartum psychosis might still be justi-
fied, as these disorders in early motherhood show many important specificities.
Apart from the fact that giving birth seems to be a special trigger, there are also
specific needs in diagnosis, treatment and care.
Diagnosis is particularly difficult, as women, due to shame, stigma and
many fears, do not seek help, and doctors, due to misinterpretation of symp-
toms, sometimes do not recognize the severity of the situation. Untreated, these
disorders can have severe consequences, not only for the mother, but also for
the child and the whole family. Therefore, these disorders need our special
attention.
They also need special treatment with modifications of our pharmacologi-
cal, non-pharmacological and psychotherapeutic methods. Furthermore, new
services are necessary, such as special low-threshold outpatient clinics with
attached child care, mother-baby units, day clinics and day centres for mothers.
Support for the mother and the father for establishing a good bonding and
relationship with the infant is of utmost importance, not only to improve the
course of the disorder in the mother, but also to avoid traumatization of the
infant with all its potential long-term consequences.
This means that as suggested in the DSM and ICD depressive and psy-
chotic disorders associated with the postpartum period should primarily be
coded according to the presenting psychiatric disorder. However, the diagnostic
specifier or additional coding of during postpartum period should be allowed
for and encouraged. This additional coding should in the future be used accord-
ing to clearer criteria as concerns incidence or prevalence or the exact time
period covered. Thus, as the above-named specific problems do not only
concern new-onset cases, the specifier should also be used for pre-existing dis-
orders, and as the specific needs involve at least the first postpartum year, this
time period should be covered. Used in this way, such an additional coding
would be of great practical value.
Coding of postpartum blues has as such not been possible so far. In line
with our argumentation, it would be feasible to classify it as an adjustment dis-
order of short duration with a postpartum onset specifier if the psychiatric
classification is necessary at all. One would, however, have to bear in mind that
it is not a psychological adjustment only, but also a physiological, probably hor-
monal one [6, 16, 17].
Other Postpartum Disorders
Post-Traumatic Stress Disorders after Childbirth
One important differential diagnosis if a patient is presenting severe
depressive symptoms after childbirth is post-traumatic stress disorder (PTSD).
This disorder has not been very well known in obstetrics as well as in psychia-
try until recently, although epidemiological data show that it is not a rare event.
In a Swedish cross-sectional study with 1,640 women, Wijma et al. [84] found
a PTSD prevalence of 1.7% after childbirth. A prospective study in England
[85] showed a comparable percentage of 1.8% women (out of 289) developing
PTSD symptoms within 6 months after delivery. A current study from the USA
found a percentage in the same range of 1.9% out of 103 women. Thirty-four
per cent of these women reported traumatic birth experiences, 30.1% of them
were only partially symptomatic [86].
All the typical symptoms of PTSD are present in severe cases. The trauma,
in this case the delivery, is persistently re-experienced, nightmares and uncon-
trolled recollections (flashbacks) occur. The patient avoids contact with other
mothers or talking about the delivery. Other typical PTSD symptoms are
enhanced physiological reactivity when exposed to internal or external cues,
loss of interest, sleep difficulties and in some cases the wish for compensation
or restitution. In clinical practice, the patients show affective symptoms like
depressive mood, restlessness, loss of energy, feelings of insufficiency and
numbing. Often anxiety symptoms can be found like worrying, feelings of con-
cern or generalized anxiety. Also formal thought disorders such as ruminating,
poor concentration or constricted thinking are reported.
There seem to be certain predisposing factors for the development of
PTSD. One important aspect is the patients personality. A high need for
control, a distinct rationality or an intense sense of decency can rise the risk for
PTSD after childbirth [87]. Other predisposing factors are an individual history
of sexual abuse, former traumata or psychiatric disorders. Although the deliv-
ery method itself is not of importance, PTSD seems to occur more often when
the delivery is finished by secondary caesarean section, in most cases after a
long and painful period of trying to give birth in a normal way. Subjective
perceptions, like the feeling of menace or threat in this situation, the feeling of
helplessness or being at someones mercy seem to be more important than the
objective situation. Furthermore, a long duration, unbearable pain, loss of
control and offending the sense of decency is often reported as well as a lack of
information during and after delivery and the perception that the people
involved are insensitive, inconsiderate or thoughtless. Soet et al. [86], in their
study of 103 patients mentioned above, identified the pain experienced during
birth, levels of social support, self-efficacy, internal locus of control, trait anxi-
ety and coping as significant predictors of the development of PTSD after
childbirth.
If the patients perceptions of a traumatic birth lead to a subsyndromal
form or even full-blown PTSD and the symptoms are not treated, most women
develop problems during future pregnancies, if they do not avoid to get pregnant
again anyway. Further pregnancies are often accompanied by severe anxiety,
panic and re-actualisation of PTSD symptoms. These patients have a higher
need for medical care and not seldom the wish for a controlled birth, for
example an elective caesarean section.
The treatment of PTSD after childbirth is the same as for any other post-
traumatic disorder, i.e. mainly a psychotherapeutic one, if possible using
trauma-specific techniques. Psychopharmacological treatment with an SSRI
can be useful, if depression and hyperarousal are prominent symptoms.
Obsessive-Compulsive Disorder
Obsessional symptoms can occur in two ways in connection with a deliv-
ery. Rather frequent is the occurrence of obsessions of hurting the child (or
even infanticide) as symptoms of postpartum depression (41% of postpartum
depressed mothers compared to 6.5% of non-depressed mothers [27]). On the
other hand, also an independent obsessive-compulsive disorder (OCD) can
manifest for the first time after delivery [88, 89]. The exact incidence or pre-
valence rates of postpartum-onset OCD are unknown, but pregnancy and
childbirth are by some authors believed to be among the main precipitants
of obsessional disorders [for reviews, see ref. 9, 90]. If a patient already suf-
fered from OCD before pregnancy, a worsening of the symptomatology is also
possible.
While obsessive symptoms as part of postpartum depression respond very
well to the usual treatment strategies such as cognitive-behavioural psychother-
apy and antidepressants (preferably a serotonergic agent), OCD with postpar-
tum onset shows a tendency to chronicity.
Other Anxiety Disorders
All anxiety disorders, especially panic disorder, can be influenced by
pregnancy and delivery or manifest for the first time in the postpartum period
[8, 91, 92]. All in all, postpartum anxiety disorders may be more common than
depression [93, 94]. Regarding panic disorder, the postpartum period may be a
time of particular vulnerability. Panic attacks can also be a prominent symptom
of postpartum depression [95], which in clinical practice could justify the
differentiation of a panic type of postpartum depression [96]. Especially for
pregnant and breast-feeding women, psychotherapy should be the first-line
treatment, although in severe cases also the use of antidepressants can be
indicated and without relevant problems. A serious risk-benefit discussion is,
however, necessary in these cases.
Disorders during Pregnancy
The incidence of new mental disorders and their prevalence during preg-
nancy seem to be similar to the 12-month incidence of mental disorders in
women of the same age group without pregnancy, according to a review
recently published by Halbreich [97]. However, similarly to a disorder in the
postpartum period, a mental disorder during pregnancy can have particularly
severe short-term and long-term consequences. A higher rate of perinatal com-
plications, preterm deliveries, low birth weight, postpartum depression and a
longer-term impact on the childs development have been reported [for a review,
see ref. 97]. This means that questions of therapy, especially pharmacotherapy
(for which we refer to the article by Ross et al., in this volume), are very spe-
cific ones, not only in the postpartum period, but also during pregnancy. It
would therefore be of high clinical utility to label mental disorders during this
period with the specifier during pregnancy.
Mood Disorders
Changes of mood positive as well as negative changes and rapid mood
swings are not unusual during pregnancy. They can be triggered for example by
the individual perception of the pregnancy, the actual social situation and part-
nership or somatic changes and symptoms. Furthermore, an influence of
the extreme hormonal changes during pregnancy can be postulated. The most
frequent affective symptoms during pregnancy are depressive symptoms and
anxiety, but also euphoria or hypomania and irritability.
While euphoria or hypomania only rarely lead to a psychiatric consultation
because the women in most cases do not suffer, depressive symptoms and
anxiety are more frequent reasons for seeking help. Bennett et al. [10] have
recently reviewed 29 studies carried out between 1985 and 2003, which inves-
tigated the number of depressive women during pregnancy using different
instruments (i.e. BDI, SADS, SCID, EPDS). They calculated the average
percentage of depressive patients with 18.3 (with a broad range from 2 to 50%).
Rates determined by a structured interview were much lower and ranged from
2 to 21%.
Not all instruments used have been validated for pregnancy and it is possible
that pregnancy-related somatic symptoms have been measured and weighted as
depressive symptoms, thus leading to false-positive diagnoses and too high esti-
mations of incidence and prevalence rates. It might thus well be that incidence
and prevalence of depression during pregnancy do not substantially differ from
the rates at other times in womens lives.
Nevertheless, it can be argued quite similarly as in postnatal depression
that depression during pregnancy needs our specific attention. Thus, a major
percentage of the women concerned will not get adequate pharmacological
treatment because of the fear of negative effects on the unborn child and yet,
at the same time, the possible consequences of untreated depression are
especially severe during this period including substance abuse, functional
impairment, increased risk of postnatal depression and poor pregnancy out-
comes [10, 98].
Anxiety Disorders
Regarding panic disorder, pregnancy has for a long time been believed to
be protective [for a review, see ref. 90], which has, however, not been supported
by a prospective study [99]. The onset or worsening of OCD during pregnancy
has also been frequently described [for a review, see ref. 90]. Symptoms of all
anxiety disorders can be so severe that in some cases patients even consider an
interruption of the pregnancy because they feel they cannot bear the symptoms
until delivery and/or the fear of the adverse effects of medication on the unborn.
Furthermore, anxiety during pregnancy has been associated with an increased
incidence of perinatal complications [for a review, see ref. 90] and postpartum
depression [100].
Psychoses
If psychoses or bipolar disorder occur during pregnancy, it is rarely the
first manifestation. More frequently, it will be the relapse of a pre-existing ill-
ness, i.e. when because of the pregnancy or because of the wish for a child a
successful pharmacological prophylaxis was stopped in some cases, very
soon after diagnosing pregnancy. The danger of relapse after stopping the pro-
phylaxis is especially high for bipolar disorders [101].
Summary and Conclusions
The diagnostic terms postpartum depression and postpartum psychosis
are still widely used. This paper discussed if this is still justified in the light of
recent research and if also other postpartum disorders or disorders during preg-
nancy should have a special entry or term in our classification systems.
Based on a comprehensive review of the literature, it has to be stated that
postpartum depression or postpartum psychosis are no specific entities in
terms of having a specific aetiology, symptomatology or course. Rather, giving
birth to a child with all its biological and psychosocial consequences seems to
act as a major stressor, which within a general vulnerability-stress model
can trigger the outbreak of all classical disorders in predisposed women.
Symptomatology, although sometimes influenced by the specific situation of
(new) motherhood in a pathoplastic way, principally corresponds to the typical
symptomatology of the classical disorders. The same is true for the course.
From a scientific point of view, the concept and term postpartum depres-
sion should therefore be abandoned as it is not a specific valid entity. A further
use of this concept might even hinder research as postpartum illnesses include
as we have shown a variety of classical disorders which have to be clearly dif-
ferentiated from each other [see also ref. 8]. So far, these illnesses might some-
times have been too uncritically categorized under postpartum depression.
However, there are good reasons to continue to use postpartum as a diag-
nostic specifier, i.e. as an addendum to the main diagnostic category (as in
DSM-IV). Thus, in many cases, becoming mother in fact seems to trigger the
onset or relapse of a disorder. Giving birth not only seems to be a unique
psychosocial stressor in the sense of a life event, which is associated with a
massive emotional upstirring, it also is a biological stressor inducing massive
hormonal and other changes. Furthermore, some women seem to have a spe-
cific psychoneuroendocrine vulnerability, which might lower the threshold for
the outbreak of a disease and potentially also influence symptomatology, e.g.
contribute to emotional lability.
Most importantly, mental disorders in early motherhood confront us with
specific needs and call for specific treatments. Thus, help seeking is often delayed
due to shame and stigma, and diagnosis is often missed due to misinterpretation
of symptoms. Services often do not adequately meet these womens needs, as they
do not take into account their specific situation, problems and fears. Untreated,
post- and peripartum disorders can have especially severe long-term conse-
quences, not only for the mother, but also for the child and the whole family.
Therefore, special attention and special treatment is necessary; this includes mod-
ifications of our pharmacological, non-pharmacological and psychotherapeutic
treatments as well as provision of new low-threshold mother-infant services.
Finally, postnatal depression has proven to be helpful as a lay term and
concept [8, 9]. It reduces stigma and encourages mothers to seek help. In fact, a
lay movement including self-help groups and lobby groups to improve services
has developed based on this obviously very appealing concept. A careful use of
the addendum with all the above-named restrictions could therefore be helpful
in practice.
In conclusion, although postpartum depression or postpartum psychoses are
no specific entities from an aetiological point of view, the diagnostic specifier
postpartum should not be abandoned. Clinical utility would in this case be the
criterion for the specifier rather than diagnostic validity, which has also been sug-
gested as worthwhile by First et al. [102]. Other peripartum disorders, such as
anxiety, OCD or PTSD, could also be marked with a corresponding specifier.
In line with our argumentation, the postpartum specifier should be used
not only for new-onset cases but for all cases prevalent during the first postpar-
tum year independent of onset to indicate the specific needs for care and
treatment during this period.
A pregnancy specifier should be introduced correspondingly.
Both specifiers would, in our opinion, help to improve clinical practice and
research if clearly defined and used as an addendum to the main diagnostic
category.
References
1 OHara MW, Zekoski EM, Philipps LH, Wright EJ: Controlled prospective study of postpartum
mood disorders: Comparison of childbearing and nonchildbearing women. J Abnorm Psychol
1990;99:315.
2 Terp IM, Mortensen PB: Post-partum psychoses. Clinical diagnoses and relative risk of admission
after parturition. Br J Psychiatry 1998;172:521526.
3 OHara MW: Post-partum blues, depression, and psychosis: A review. J Psychosom Obstet
Gynaecol 1987;7:205227.
4 Harding JJ: Postpartum psychiatric disorders: A review. Compr Psychiatry 1989;30:109112.
5 Kendell RE, Chalmers JC, Platz C: Epidemiology of puerperal psychoses. Br J Psychiatry
1987;150:662673.
6 Riecher-Rssler A: Psychische Strungen und Erkrankungen nach der Entbindung. Fortschr
Neurol Psychiatr 1997;65:97107.
7 Riecher-Rssler A, Hofecker Fallahpour M: Die Depression in der Postpartalzeit eine diagnos-
tische und therapeutische Herausforderung. Schweiz Arch Neurol Psychiatr 2003;154:106154.
8 Brockington I: Postpartum psychiatric disorders. Lancet 2004;363:303310.
9 Brockington I: Diagnosis and management of post-partum disorders: A review. World Psychiatry
2004;3:8995.
10 Bennett HA, Einarson A, Taddio A, Koren G, Einarson TR: Prevalence of depression during
pregnancy: Systematic review. Obstet Gynecol 2004;103:698709.
11 Riecher-Rssler A, Hofecker Fallahpour M: Postpartum depression: Do we still need this diagnostic
term? Acta Psychiatr Scand 2003;418(Suppl):5156.
12 Paykel ES, Emms EM, Fletcher J, Rassaby ES: Life events and social support in puerperal depres-
sion. Br J Psychiatry 1980;136:339346.
13 Watson JP, Elliott SA, Rugg AJ, Brough DI: Psychiatric disorder in pregnancy and the first
postnatal year. Br J Psychiatry 1984;144:453462.
14 Kumar R, Robson KM: A prospective study of emotional disorders in childbearing women. Br J
Psychiatry 1984;144:3547.
15 Cooper PJ, Campbell EA, Day A, Kennerley H, Bond A: Non-psychotic psychiatric disorder after
childbirth. A prospective study of prevalence, incidence, course and nature. Br J Psychiatry
1988;152:799806.
16 Miller LJ, Rukstalis M: Beyond the blues, hypotheses about postpartum reactivity; in Miller
(ed): Postpartum Mood Disorders. Washington, American Psychiatric Press, 1999, pp 319.
17 Henshaw C: Mood disturbances in the early puerperium: A review. Arch Women Ment Health
2003;6:S33S42.
18 Herz E, Thoma M, Umek W, Gruber K, Linzmayer L, Walcher W, Philipp T, Putz M: Nicht-
psychotische postpartale Depression. Pilotstudie zu Epidemiologie und Risikofaktoren.
Geburtshilfe Frauenheilkd 1997;57:282288.
19 Augusto A, Kumar R, Calheiros JM, Matos E, Figueiredo E: Post-natal depression in an urban area of
Portugal: Comparison of childbearing women and matched controls. Psychol Med 1996:26:135141.
20 Webster ML, Thompson JM, Mitchell EA, Werry JS: Postnatal depression in a community cohort.
Aust NZ J Psychiatry 1994;28:4249.
21 World Health Organisation: The ICD-10 Classification of Mental and Behavioral Disorders.
Geneva, World Health Organisation, 1992.
22 American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, ed 4.
Washington, American Psychiatric Press, 1994.
23 Spitzer RL, Williams JBV: Classification in psychiatry; in Kaplan HI, Freedman AM, Sadock BJ
(eds): Comprehensive Textbook of Psychiatry. Baltimore, Williams and Wilkins, 1985, pp 591613.
24 Pitt B: Atypical depression following childbirth. Br J Psychiatry 1968;114:13251335.
25 Brockington I: Motherhood and Mental Health. Oxford, Oxford University Press, 1996.
26 Wisner KL, Peindl KS, Gigliotti T, Hanusa BH: Obsessions and compulsions in women with
postpartum depression. J Clin Psychiatry 1999;60:176180.
27 Jennings KD, Ross S, Popper S, Elmore M: Thoughts of harming infants in depressed and nonde-
pressed mothers. J Affect Disord 1999;54:2128.
28 Appleby L, Mortensen PB, Faragher EB: Suicide and other causes of mortality after post-partum
psychiatric admission. Br J Psychiatry 1998;173:209211.
29 McNeil TF: A prospective study of postpartum psychoses in a high-risk group. 1. Clinical charac-
teristics of the current postpartum episodes. Acta Psychiatr Scand 1986;74:205216.
30 Brockington I, Cernik KF, Schofield EM, Downing AR, Francis AF, Keelan C: Puerperal Psychosis.
Phenomena and diagnosis. Arch Gen Psychiatry 1981;38:829833.
31 Schpf J: Postpartum-Psychosen. Berlin, Springer, 1994.
32 Gold LH: Postpartum disorders in primary care: Diagnosis and treatment. Prim Care 2002;29:
2741, vi.
33 Altshuler LL, Hendrick V, Cohen LS: Course of mood and anxiety disorders during pregnancy and
the postpartum period. J Clin Psychiatry 1998;59(suppl 2):S29S33.
34 Meltzer ES, Kumar R: Puerperal mental illness, clinical features and classification: A study of 142
mother-and-baby admissions. Br J Psychiatry 1985;147:647654.
35 Chaudron LH, Pies RW: The relationship between postpartum psychosis and bipolar disorder:
A review. J Clin Psychiatry 2003;64:12821294.
36 Dean C, Kendell RE: The symptomatology of puerperal illnesses. Br J Psychiatry 1981;139:
128133.
37 OHara MW, Swain AM: Rates and risk of postpartum depression A meta-analysis. Int Rev
38 Cox JL, Holden JM, Sagovsky R: Detection of postnatal depression. Development of the 10-item
Edinburgh Postnatal Depression Scale. Br J Psychiatry 1987;150:782786.
39 Cox JL, Murray D, Chapman G: A controlled study of the onset, duration and prevalence of post-
natal depression. Br J Psychiatry 1993;163:2731.
40 Robins LN, Regier DA: Psychiatric Disorders in America. The Epidemiological Catchment Area
Study. New York, The Free Press, 1991.
41 Kessler RC, McGonagle KA, Zhao S, Nelson CB, Hughes M, Eshleman S, Wittchen HU, Kendler KS:
Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States.
Results from the National Comorbidity Survey. Arch Gen Psychiatry 1994,51:819.
42 Meyer C, Rumpf HJ, Hapke U, Dilling H, John U: Lifetime prevalence of mental disorders in a
general adult population. Results of TACOS study. Nervenarzt 2000;71:535542.
43 Brockington I, Winokur G, Dean C: Puerperal psychosis; in Brockington I, Kumar C (eds):
Motherhood and Mental Illness. New York, Grune and Stratton, 1982, pp 3770.
44 OHara MW: Social support, life events, and depression during pregnancy and the puerperium.
Arch Gen Psychiatry 1986;43:569573.
45 Steiner M, Fairman M, Doyle W: Mood disorders with postpartum onset: A family study. Abstract
presented at the XIXth CINP Congress, Washington DC. Neuropsychopharmacology 1994;10:176.
46 Davidson J, Robertson E: A follow-up study of post partum illness, 19461978. Acta Psychiatr
Scand 1985;71:451457.
47 Rohde A, Marneros A: Postpartum psychoses: Onset and long-term course. Psychopathology
1993;26:203209.
48 Garfield P, Kent A, Paykel ES, Creighton FJ, Jacobson RR: Outcome of postpartum disorders:
A 10-year follow-up of hospital admissions. Acta Psychiatr Scand 2004;109:434439.
49 Treloar SA, Martin NG, Bucholz KK, Madden PA, Heath AC: Genetic influences on post-natal
depressive symptoms: Findings from an Australian twin sample. Psychol Med 1999;29:645654.
50 Kumar R: Postpartum mood disorders: The psychoses; in Demers L, McGuire M, Phillips A,
Rubinow D (eds): Premenstrual, Postpartum and Menopausal Mood Disorders. Baltimore, Urban
& Schwarzenberg, 1989, pp 163172.
51 Boyce PM: Risk factors for postnatal depression: A review and risk factors in Australian populations.
Arch Women Ment Health 2003;6(suppl 2):S43S50.
52 Wilson LM, Reid AJ, Midmer DK, Biringer A, Carroll JC, Stewart DE: Antenatal psychosocial
risk factors associated with adverse postpartum family outcomes. CMAJ 1996;154:785799.
53 Beck CT: Meta-analysis of predictors of postpartum depression. Nurs Res 1996;45:297303.
54 Fink G, Sumner BE, Rosie R, Grace O, Quinn JP: Estrogen control of central neurotransmission:
Effect on mood, mental state, and memory. Cell Mol Neurobiol 1996;16:325344.
55 Riecher-Rssler A: Oestrogens and schizophrenia. Curr Opin Psychiatry 2003;16:187192.
56 Bergemann N, Riecher-Rssler A: Estrogen Effects in Psychiatric Disorders. Wien, Springer,
2004.
57 Mahe V, Dumaine A: Oestrogen withdrawal associated psychoses. Acta Psychiatr Scand 2001;104:
323331.
58 Bloch M, Schmidt PJ, Danaceau M, Murphy J, Nieman L, Rubinow DR: Effects of gonadal
steroids in women with a history of postpartum depression. Am J Psychiatry 2000;157:924930.
59 Whiffen VE, Gotlib IH: Comparison of postpartum and nonpostpartum depression: Clinical pre-
sentation, psychiatric history, and psychosocial functioning. J Consult Clin Psychol 1993;61:
485494.
60 Rohde A, Marneros A: Schizoaffective disorders with and without onset in the puerperium. Eur
Arch Psychiatry Clin Neurosci 1992;242:2733.
61 Kumar RC: Anybodys child: Severe disorders of mother-to-infant bonding. Br J Psychiatry
1997;171:175181.
62 Murray L, Cooper J: Postpartum depression and child development. Psychol Med 1997;27:
253260.
63 Brennan PA, Hammen C, Andersen MJ, Bor W, Najman JM, Williams GM: Chronicity, severity,
and timing of maternal depressive symptoms: Relationships with child outcomes at age 5. Dev
Psychol 2000;36:759766.
64 Grace SL, Evindar A, Stewart D: The effect of postpartum depression on child cognitive develop-
ment and behavior: A review and critical analysis of the literature. Arch Women Ment Health
2000;6:263274.
65 Pawlby SJ, Hay DF, Sharp DJ: The effects of postnatal depression on the development of boys.
Arch Women Ment Health 2001;3(suppl 2):5.
66 Rohde A, Raic D, Varchmin-Schultheiss K, Marneros A: Infanticide: Sociobiographical back-
ground and motivational aspects. Arch Women Ment Health 1998;1:125130.
67 Parry BL: Postpartum psychiatric syndromes; in Kaplan HI, Sadock B (eds): Comprehensive
Textbook of Psychiatry. Philadelphia, Williams & Wilkins 1995, vol 1, pp 10591066.
68 MacArthur C: What does postnatal care do for womens health? Lancet 1999;353:343344.
69 McIntosh J: Postpartum depression: Womens help-seeking behaviour and perceptions of cause.
J Adv Nurs 1993;18:178184.
70 Field T, Grizzle N, Scafidi F, Schanberg S: Massage therapy for infants of depressed mothers.
Infant Behav Dev 1996;19:107112.
71 Onozawa K, Glover V, Adams D, Modi N, Kumar RC: Infant massage improves mother-infant
interaction for mothers with postnatal depression. J Affect Disord 2001;63:201207.
72 Stuart S, OHara MW: Treatment of postpartum depression with interpersonal psychotherapy.
Arch Gen Psychiatry 1995;52:7576.
73 Hofecker Fallahpour M, Riecher-Rssler A, Wsten G, Zinkernagel C, Stieglitz RD:
Gruppentherapie fr depressive Mtter; in Riecher-Rssler A, Rohde A (eds): Psychische
Erkrankungen bei Frauen fr eine geschlechtersensible Psychiatrie und Psychotherapie. Basel,
Karger, 2001, pp 307320.
74 Hofecker Fallahpour M, Zinkernagel Burri C, Stckli B, Wsten G, Stieglitz RD, Riecher-Rssler A:
Gruppentherapie bei Depression in der frhen Mutterschaft: erste Ergebnisse einer Pilotstudie.
Nervenarzt 2003;74:767774.
75 Burt VK, Suri R, Altshuler L, Stowe Z, Hendrick VC, Muntean E: The use of psychotropic med-
ications during breast-feeding. Am J Psychiatry 2001;158:10011009.
76 Yonkers A, Wisner KL, Stowe Z, Leibenluft E, Cohen L, Miller L, Manber R, Viguera A, Suppes T,
Altshuler L: Management of bipolar disorder during pregnancy and the postpartum period. Am J
Psychiatry 2004;161:608620.
77 Newport DJ, Hostetter A, Arnold A, Stowe ZN: The treatment of postpartum depression:
Minimizing infant exposures. J Clin Psychiatry 2002;63(suppl 7):S31S44.
78 Gregoire AJ, Kumar R, Everitt B, Henderson AF, Studd JW: Transdermal oestrogen for treatment
of severe postnatal depression. Lancet 1996;347:930933.
79 Ahokas A, Kaukoranta J, Wahlbeck K, Aito M: Estrogen deficiency in severe postpartum depression:
Successful treatment with sublingual physiologic 17beta-estradiol: A preliminary study. J Clin
Psychiatry 2001;62:332336.
80 Lawrie TA, Herxheimer A, Dalton K: Oestrogens and progestogens for preventing and treating
postnatal depression (Cochrane Review); in: The Cochrane Library. Chichester, Wiley, 2004.
81 Ahokas A, Aito M, Turiainen S: Association between oestradiol and puerperal psychosis. Acta
Psychiatr Scand 2000;101:167169.
82 Corral M, Kuan A, Kostaras D: Bright light therapys effect on postpartum depression. Am J
Psychiatry 2000;157:303304.
83 MainTF: Mothers with children in a psychiatric hospital. Lancet 1958;ii:845847.
84 Wijma K, Soderquist J, Wijma B: Posttraumatic stress disorder after childbirth: A cross sectional
study. J Anxiety Disord 1997;11:587597.
85 Ayers S, Pickering AD: Do women get posttraumatic stress disorder as a result of childbirth?
A prospective study of incidence. Birth 2001;28:111118.
86 Soet JE, Brack GA, DiIorio C: Prevalence and predictors of womens experience of psychological
trauma during childbirth. Birth 2003;30:3646.
87 Pantlen A, Rohde A: Psychische Auswirkungen traumatisch erlebter Entbindungen. Zentralbl
Gynkol 2001;123:4247.
88 Abramowitz JS, Schwartz SA, Moore KM, Luenzmann KR: Obsessive-compulsive symptoms in
pregnancy and the puerperium: A review of the literature. Anxiety Disord 2003;17:461478.
89 Brandes M, Soares CN, Cohen LS: Postpartum onset obsessive-compulsive disorder: Diagnosis
and management. Arch Women Ment Health 2004;7:99110.
90 Levine RE, Oandasan AP, Primeau LA, Berenson AB: Anxiety disorders during pregnancy and
postpartum. Am J Perinatol 2003;5:239248.
91 Shear MK, Mammen O: Anxiety disorders in pregnant and postpartum women. Psychopharmacol
Bull 1995;31:693703.
92 Villeponteaux VA, Lydiard RB, Laraia MT, Stuart GW, Ballenger JC: The effects of pregnancy on
preexisting panic disorder. J Clin Psychiatry 1992;53:201203.
93 Matthey S, Barnett B, Howie P, Kavanagh DJ: Diagnosing postpartum depression in mothers and
fathers: Whatever happened to anxiety? J Affect Disord 2003;74:139147.
94 Hertzberg T, Wahlbeck K: The impact of pregnancy and puerperium on panic disorder: A review.
J Psychosom Obst Gynaecol 1999;20:5964.
95 Metz A, Sichel DA, Goff DC: Postpartum panic disorder. J Clin Psychiatry 1988;49:278279.
96 Rohde A: Psychiatrische Erkrankungen in der Schwangerschaft und im Wochenbett. Gynkologe
2001;34:315323.
97 Halbreich U: Prevalence of mood symptoms and depression during pregnancy: Implications for
clinical practice and research. CNS Spectr 2004;9:177184.
98 Freeman MP, Smith KW, Freeman SA, McElroy SL, Kmetz GE, Wright R, Keck PE Jr: The impact
of reproductive events on the course of bipolar disorder in women. J Clin Psychiatry 2002;63:
284287.
99 Cohen LS, Sicher DA, Faraone SV, Robertson LM, Dimmock JA, Rosenbaum JF: Course of panic
disorder during pregnancy and the puerperium: A preliminary study. Biol Psychiatry 1996;39:
950954.
100 Heron J, OConnor TG, Evans J, Golding J, Glover V: The ALSPAC Study Team. The course of
anxiety and depression through pregnancy and the postpartum in a community sample. J Affect
Disord 2004;80:6573.
101 Viguera AC, Nonacs R, Cohen LS, Tondo L, Murray A, Baldessarini RJ: Risk of recurrence of
bipolar disorder in pregnant and nonpregnant women after discontinuing lithium maintenance. Am
J Psychiatry 2000;157:179184.
102 First MB, Pincus HA, Levine JB, Williams JBW, Ustun B, Peele R: Clinical utility as a criterion
for revising psychiatric diagnoses. Am J Psychiatry 2004;161:946954.
Prof. Dr. med. Anita Riecher-Rssler
Psychiatrische Poliklinik
Universittsspital Basel
Petersgraben 4
CH4031 Basel
Tel. 41 61 265 51 14, Fax 41 61 265 45 99, E-Mail ariecher@uhbs.ch
Maternal Adversity,Vulnerability
and Disease
Stephen G. Matthews
a
, Michael J. Meaney
b
a
Departments of Physiology, Obstetrics and Gynaecology and Medicine, Faculty of
Medicine, University of Toronto, Toronto, and
b
McGill Program for the Study of
Behaviour, Genes and Environment, Douglas Hospital Research Centre, Montral,
Canada
The probability of chronic illness is strongly influenced by environmental
conditions prevailing during development. The frequency of heart disease, dia-
betes and mental illness is substantially greater amongst those born and raised
in poverty, regardless of social class in adulthood. Adversity associated with
poverty produces incomplete, dysfunctional families often rife with domestic
violence, drug use, child abuse and neglect. Reproduction within this context
results in maternal stress, increased risks for infection and thus preterm labour,
perinatal deaths, birth insults, poor nutrition for mother and offspring and
serious compromises in the quality of parent-child interactions. These condi-
tions define high-risk families. For the child, these conditions bear the potential
to influence the development of physiological systems that regulate metabo-
lism and cardiovascular function, as well as neural processes underlying
emotion and cognition. These effects form the basis for vulnerability to illness
in later life.
Perhaps the most dramatic illustration of the relation between adversity in
early life and the risk for later illness emerges from studies on birth weight,
early growth and health in adulthood. Individuals that were small for gestational
age, suggesting some measure of intra-uterine growth restriction (IUGR), are at
significantly greater risk for type II diabetes, visceral obesity and hypertension
(collectively referred to as the metabolic syndrome). The major predictors
of fetal growth restriction and preterm labour are maternal stress, infections,
malnutrition and tobacco/alcohol consumption. These risk factors are far more
prevalent in low socio-economic status (SES) families and, predictably, low
SES emerges as a major risk factor for both IUGR and preterm labour.
Maternal Adversity, Vulnerability and Disease 29
It is increasingly clear that growth restriction also predicts an increased
risk for mood disorders [1]. Additionally, the risk factors (i.e., maternal stress,
tobacco/alcohol use, nutritional deprivation) for IUGR are also the major
predictors for attentional deficit disorder (ADD) and, predictably, ADD is far
more prevalent in low SES environments. Follow-up studies of growth-
restricted infants reveal evidence for problems associated with attention and
impulse control, functions that are heavily dependent on the hippocampus and
prefrontal cortex. These same brain regions are also among the major sites
implicated in both depression and anxiety disorders. Interestingly, growth
restriction is also associated with increased behavioural inhibition, shyness and
timidity in childhood, all of which appear to be predictors of mood disorders in
later life. Studies with primate and rodent models show that the development of
the hippocampus is impaired by fetal glucocorticoid exposure or by prenatal
stress, and glucocorticoids are a mechanism for fetal growth restriction. Thus,
it emerges that impairments in the development of the hippocampus and
prefrontal cortex serve as the mechanism for the relationship between fetal
growth restriction and impaired cognitive and emotional development.
These studies suggest that endocrine conditions that promote fetal growth
restriction can also compromise neural development. However, follow-up
studies with even very-low-birth-weight babies reveal an impressive level of
variability in cognitive outcomes. Evidence from clinical intervention studies
with high-risk infants and developmental neurobiology suggests that neurocog-
nitive development is influenced by the quality of the postnatal environment.
Thus, we can predict that the developmental outcomes associated with fetal
adversity, reflected in growth retardation, are determined by the quality of the
postnatal environment.
Intra-Uterine Growth Restriction and the
Hypothalamo-Pituitary-Adrenal Axis
A major paradox of pregnancy involves the increased levels of the highly
catabolic, adrenal glucocorticoids in the maternal circulation. Increased cortisol
levels make perfect sense for the mother. Pregnancy represents a period of
metabolic demand, and glucocorticoids increase the availability of energy sub-
strates through both gluconeogenesis and lipolysis. However, since steroids
readily cross the placental barrier, the same catabolic hormones could compro-
mise the growth of the fetus. Glucocorticoids antagonize the anabolic effects of
growth hormone through, among other effects, a disruption of the growth
hormone-insulin-like growth factor pathway. Thus, increased fetal glucocorticoid
exposure in animals can produce growth retardation.
Matthews/Meaney 30
Natures resolution to the maternal-fetal glucocorticoid paradox is the
presence of a placental enzyme, 11-hydroxysteroid dehydrogenase type 2
(11-HSD2). 11-HSD2 converts about 95% of cortisol to the biological inert,
cortisone, thus maintaining about a 10:1 mother-fetus ratio in circulating corti-
sol. However, placental expression of 11-HSD2 is sensitive to environmental
regulation and decreased 11-HSD2 function is associated with increased fetal
exposure to cortisol and thus a risk for IUGR. Placental tissue obtained from
low-birth-weight pregnancies show significant reductions in 11-HSD2 mRNA
and a decreased capacity for the conversion of cortisol to cortisone [24].
Moreover, chord blood samples obtained from low-birth-weight babies
reveal significant elevations in both corticotrophin-releasing factor (CRF) and
cortisol [5]. Elevated fetal exposure to cortisol thus emerges as a mechanism for
IUGR. Interestingly, maternal stress, protein deprivation, tobacco and alcohol all
increase maternal adrenal glucocorticoid release, which increases placental CRF
expression; hence the relationship between maternal and fetal glucocorticoids-
placental CRF in IUGR babies. Additionally, maternal stress and anxiety are
commonly associated with an increased catecholamine output that could serve to
impair uteroplacental circulation, further decreasing oxygen and nutrient
passage to the fetus. Nevertheless small-birth-weight babies commonly show a
postnatal catch-up in growth, such that the evidence of growth retardation will
largely have disappeared by 1 year of age. Why, then, would these children be at
greater risk for illness in later life?
Fetal Programming of Gene Expression
Elevated exposure to biologically active glucocorticoids in fetal life can
permanently alter the expression of hepatic genes (i.e., PEPCK, glucose
kinases) that are key regulators of glucose and fat metabolism. Collectively,
these effects increase the capacity for lipid metabolism, decrease glucose clear-
ance and render the organism vulnerable to hyperlipidaemia, glucose intoler-
ance and hyperglycaemia. Thus, administration of dexamethasone (which is not
subject to conversion by placental 11-HSD2) to pregnant rat mothers results in
offspring which, as adults, show increased blood pressure, hyperlipidaemia and
decreased glucose tolerance, thus replicating the condition associated with
IUGR. These findings represent examples of what is known as early life pro-
gramming of gene expression: situations where events in early life exert an
apparently permanent effect on gene expression, and thus phenotype.
One recent finding from the Southampton studies dramatically expands the
scope of the fetal programming hypothesis beyond that of the liver [6]. Among
elderly male human subjects, birth weight was a highly significant predictor of
basal cortisol levels. Individuals that were small for gestational age showed
increased circulating levels of cortisol. This finding is actually predictable based
on animal studies of the effects of prenatal stress or prenatal glucocorticoid expo-
sure, both of which increase basal and stress-induced hypothalamic-pituitary-
adrenal (HPA) activity in adulthood [4, 712]. These effects are associated with
alterations in mineralocorticoid and glucocorticoid receptor expression in brain
regions that mediate corticosteroid negative feedback inhibition (fig. 1). Decreased
negative feedback sensitivity to circulating glucocorticoids is associated with
enhanced hypothalamic CRF expression and increased glucocorticoid release.
Moreover, prenatal stress/glucocorticoid exposure also programs CRF gene
expression in the amygdala, further increasing HPA activity through the regulation
of central noradrenergic projections to the hypothalamic CRF neurons (fig. 2).
Fetal Adversity and Neural Development
The effects of prenatal stress/glucocorticoid exposure on HPA activity in
adulthood are also mediated by alterations in the structural development of
ACTH
Cortisol
CRH mRNA
AVP mRNA
Hippocampus
Hypothalamic
PVN
Anterior
pituitary
Adrenal
cortex
POMC mRNA
POMC
CRH
AVP
Glucocorticoid
receptor
Mineralocorticoid
receptor
Dentate
gyrus
CA1
CA2
CA3
CA4
Fig. 1. Schematic illustrating the HPA axis and the major routes of glucocorticoid
feedback within the axis. CRH Corticotropin-releasing hormone; AVP vasopressin;
POMC proopiomelanocortin; ACTH adrenocorticotropin.
Matthews/Meaney 32
brain regions that normally serve to regulate hypothalamic CRF expression and
HPA activity. The principal among such regions is the hippocampus. Prenatal
exposure to glucocorticoids dramatically inhibits neuronal proliferation in the
hippocampus in rodents as well as primates, resulting in a severely impaired
pattern of hippocampal development. In adult rats, hippocampal lesions
produce increased hypothalamic CRF expression and elevated HPA activity.
Thus, prenatal stress can increase HPA responses to stress through effects on
neural structures that normally inhibit CRF synthesis and release.
The hippocampus and prefrontal cortex are also critical structures for
impulse control and emotional states such as fear. As adults, animals exposed to
stress during fetal development exhibit increased fearfulness and anxiety under
conditions of stress. Interestingly, fetal dexamethasone administration in humans
for congenital adrenal hyperplasia was associated with increased shyness
and timidity in children. Both anxiety disorders and depression are accom-
panied by significant decreases in hippocampal volume and, in the case of
Hippocampus
BNST
(CRF)
Autonomic
output
HPA output
(NA)
Baso-
lateral
complex
Central
n.
mPFC
P
V
N
Sensory
input
(Stressor)
BNST
(CRF)
Locus coer.
PB n
NTS
Fig. 2. A simplified version of the neural circuits mediating behavioural, autonomic
and HPA responses to stress. Note the critical role for the CRF from the central nucleus
(central n.) of the amygdala in activating the release of noradrenaline from the noradrenergic
cell body regions in the locus coeruleus, parabrachial nucleus (PB n.) and the nucleus of the
solitary tract (NTS). Noradrenaline is released into virtually all areas of the corticolimbic
regions, including the hippocampus and medial prefrontal cortex (mPFC). At the level of the
PVN, noradrenaline stimulates the release of CRF from the PVN, activating the HPA
responses to stress. BNST Bed nucleus of the stria terminalis.
depression, dramatically reduced volume and activity in the prefrontal cortex. The
hippocampus and the prefrontal cortex are primary targets for both antidepres-
sant and anti-anxiety medications. These correlational findings do not clarify
whether decreased hippocampal volume is a pre-existing cause of depression.
Indeed, studies with both rodent and primate models show that chronic stress can
result in the shrinkage of dendritic branching and even neuron loss in the
hippocampus. It is possible that the angst of severe mental illness might provoke
hippocampal atrophy. Alternatively, we argue that the decreases in hippocampal
and prefrontal cortical volumes occur as a function of early adversity and thus
define a state of increased vulnerability for cognitive and emotional impair-
ments. Studies in rhesus monkeys indicate that prenatal exposure to glucocorti-
coids has dramatic and lasting effects on hippocampal volume [13].
Hippocampal atrophy is not unique to depression. Patients with post-
traumatic stress disorder (PTSD) also exhibit a decrease in hippocampal volume.
Interestingly, recent studies of monozygotic twins where one of the pair suffers
from PTSD, reveal that the non-traumatized, otherwise healthy sibling also
shows evidence of reduced hippocampal volume. Moreover, individuals with a
history of parental abuse in childhood show decreased hippocampal volume.
These findings suggest that the origins of the decreased hippocampal volume
might lie in early development and be considered a predisposing condition. This
idea is consistent with clinical data, which shows that even among seriously
traumatized individuals (e.g. rape victims) only about 25% develop chronic
PTSD; the single best predictor of vulnerability to PTSD is the quality of early
family life and parent-child relations. Epidemiological studies of depression
and anxiety disorders reveal that early life events, including both fetal growth
retardation and the quality of postnatal parental care, are major predictors of
mood disorders. While these findings are entirely consistent with the results of
animal studies (see below), the definitive developmental studies will require a
longitudinal, within-subject design in which the effects of environmental adversity
and family function are determined early in life.
The hippocampus is, of course, intimately associated with cognitive
function, notably episodic learning/memory, which refers to the ability to acquire
factual information, such as time and location (i.e., spatiotemporal relation-
ships). The hippocampus and the prefrontal cortex are also critical in attentional
processes. Predictably, in animal studies, including those with primate models,
prenatal stress compromises hippocampal development and results in impaired
attention [14]. These findings suggest that fetal adversity would be associated
with both cognitive and emotional disturbances in childhood, and this is indeed
the case. Both prematurity and IUGR are consistently associated with cognitive
impairments; the existing data suggest that cognitive outcomes are more seriously
compromised by growth restriction than by prematurity [15]. Earlier studies,
Matthews/Meaney 34
using more global measures of cognitive development, such as IQ tests, reported
significant impairments among IUGR children [e.g. 1620]. As adults, IUGR
babies are less likely to hold managerial/professional employment, and earn less
income [21], suggesting that growth-related impairments in cognitive develop-
ment are functionally relevant. Recent studies focus on more specific aspects of
cognitive function. The Avon Longitudinal Study found that among low-birth-
weight children there were significant reductions in many cognitive functions,
including attention and spatial learning/memory. Glover and OConnor [22] and
Van der Reijden-Lakeman et al. [23] found that at 2 years of age, IUGR children
were impaired on tests of divided, focused and sustained attention. On such tests,
the IUGR population was less accurate and more impulsive than controls. These
findings suggest that low-birth-weight children might be at greater risk for the
development of ADD. Indeed, the risk factors for IUGR, including maternal
stress, maternal psychopathology, protein malnutrition, tobacco/alcohol con-
sumption are identical to those for ADD. To date, no study has systematically
examined whether low-birth-weight children are indeed at greater risk for ADD.
These events occur against a genomic background and a number of genetic
polymorphisms have been related to ADD. The first is a 40-bp repeat polymor-
phism in the 3-region of the gene for the dopamine transporter (DAT), bearing 9
or 10 copies of the repeat. Such repeat polymorphisms influence gene expres-
sion and are therefore functionally relevant. Individuals heterozygous for the
9/10-repeat variant exhibit reduced DAT protein in the caudate putamen [24].
A second potentially relevant polymorphism resides in a 48-bp repeat in exon 3
of the dopamine D
4
receptor (DRD4) gene and appears to be functionally rele-
vant [25]. Linkage and association studies associated both the DRD4 7-repeat
allele and the DAT polymorphism with ADD [2628], although there are also
negative reports [24]. Genes encode for proteins, not syndromes and ADD is a
complex disorder that emerges from a persistent interaction between gene and
environment. Linkage analysis studies rarely examine the influence of genomic
variants within a developmental context. It is possible that the functional impor-
tance of the DRD4 or DAT variants for cognitive function might be best revealed
within a high-risk population. Research in schizophrenia underscores this point.
Birth insults (e.g. hypoxia) increase the risk for schizophrenia marginally in the
normal population; amongst those with a family history of schizophrenia, there
is an approximately 100-fold increase in the disorder.
Maternal stress during gestation is a predictor of both fetal growth restric-
tion and negative temperament (irritability, negative emotionality) and anxiety in
children [22, 29]. Behavioural disorders, characterized by poor impulsive con-
trol, shyness, timidity and negative temperament (greater frequency of negative
emotional states) are positively related to the degree of growth restriction
[e.g. 3032]. In one study, the risk for this constellation of behavioural problems
increased by 30% for each kilogram below the average birth weight [33].
Interestingly, growth restriction, examined on a continuum, was associated with
increased stress susceptibility [34]. This behavioural profile defines a condition
described as behavioural inhibition and predicts a greater risk for adult depres-
sion and anxiety; IUGR is linked to an increased risk for depression [1].
It is important to note the potential for gene-environment interactions.
Ebstein et al. [35] found a highly significant effect of the DRD4 repeat variant
on infant temperament, with 2-month-olds bearing short DRD4 variants and
showing more negative emotions than those bearing a longer number of repeats.
Interestingly, this association was only apparent among children bearing the
short variant of the 5-HT transporter (5-HTTP) gene [36]. Among children with
the longer version of the 5-HTTP gene, the variants of the DRD4 gene were
inconsequential with respect to temperament. Interestingly, variants of the
DRD4 gene are also associated with attachment behaviour in the Stranger
Situation Test at 1 year of age [37, 38]. Such effects are apparent even in a
non-clinical, low-risk population. Certainly, infants with more negative emo-
tions or those who differ in attachment elicit differential parenting. The early
appearance of the functional correlates of these genetic variants underscores the
potential gene environment interactions. An example of such interactions
emerges from non-human primate studies [39]. Rhesus monkeys show the same
5-HTTP polymorphism, with strong linkage to temperament and stress reactiv-
ity, as well as evidence for effects of central 5-HT activity. However, animals
bearing the short variant of the 5-HTTP gene, but reared by highly nurturing
mothers, do not differ from animals with the long 5-HTTP variant on measures
of temperament, stress reactivity or 5-HT function.
Fetal growth retardation is associated with problems of impulse control and
attention. The preliminary results of recent imaging studies of low-birth-weight
children reveal significantly decreased hippocampal volume. These clinical find-
ings are entirely consistent with the results of animal studies described above.
Indeed, alterations in hippocampal and prefrontal development might explain the
unusually high level of comorbidity between attentional deficits, impulse control
disorders and affective disorders. The findings to date suggest that adversity
during fetal development can result in increased exposure to glucocorticoids
which promote growth restriction and compromise neural development, resulting
in vulnerability to both cognitive and emotional disorders (fig. 3).
Postnatal Programming of Neural Development
The quality of family life influences the development of individual differ-
ences in vulnerability to illness throughout life. As adults, victims of childhood
Matthews/Meaney 36
abuse are at greater risk for mental illness, as well as for obesity, type II
diabetes and heart disease [4043]. Children need not be beaten to be compro-
mised. Persistent emotional neglect, family conflict and conditions of harsh,
inconsistent discipline compromise growth [e.g. 44], intellectual development
[45, 46] and increase the risk for obesity [47], depression and anxiety disorders
[4850] to a level comparable to that for abuse. More subtle relationships exist.
Low scores on measures of parental bonding (parental-infant attachment),
reflecting cold, distant parent-child relationships increase the risk of depression
and anxiety in later life [5153]. Of course, the sword cuts both ways. Family
life can also serve as a source of resilience in the face of chronic stress [54].
Thus, warm, nurturing families tend to promote resistance to stress and to
diminish vulnerability to stress-induced illness [55].
The relationship between early life events and health appears to be, in part,
mediated by parental influences on the development of neural systems which
Maternal
endocrine state
Fetal growth
Outcomes
Cognition function
Emotional well being
Endocrine function
Family life
Neural
development
Maternal depression/anxiety
Nutrition
Maternal adversity
Tobacco/alcohol
Environmental factors
Low SES High SES
Partner relationship
Violence Social support
Relationship to mother
Community
Workplace
Individual resources
Self-esteem
Optimism
Vulnerability
Adaptations
Prenatal
Maternal depression/anxiety
Nutrition
Maternal adversity
Tobacco/alcohol
Postnatal
Fig. 3. A schema outlining the proposed major pathways by which environmental
adversity influences development. Note that in the case of both pre- and postnatal life, this
influence is mediated by maternal emotional well-being and behaviour.
underlie the expression of behavioural and endocrine responses to stress
[5658]. Chronic exposure to elevated levels of the stress mediators, such as the
glucocorticoids and catecholamines as well as CRF, are associated with
increased visceral obesity, insulin resistance as well as mood and cognitive
disorders [57, 59]. Interestingly, each of these disorders is associated with
IUGR. This raises the possibility that, in part, increased HPA and cate-
cholaminergic responses to stress might mediate the relationship between
IUGR and multiple forms of chronic illness. Moreover, under most conditions,
the forms of adversity that promote IUGR also prevail in postnatal life and the
results from animal studies reveal that the nature of the postnatal environment
also influences the development of stress responsess (fig. 3).
Experimental manipulations during postnatal life that compromise the
quality and stability of mother-infant interactions (e.g. prolonged periods of
separation) in the rat or monkey increase HPA responses to stress in adulthood
[6062]. Moreover, maternal care directly influences the development of
individual differences in HPA and behavioural responses to stress [39, 63]. In
the rat, naturally occurring variations in maternal care are associated with the
development of individual differences in HPA responses to stress. As adults, the
offspring of mothers that exhibit increased levels of pup licking/grooming (LG)
and arched-back nursing (ABN) are less fearful and show more modest HPA
responses to stress [64, 65]. Cross-fostering completely reverses the phenotype.
The biological offspring of low LG-ABN mothers fostered at birth by high
LG-ABN dams are comparable on measures of stress reactivity to the normal
offspring of high LG-ABN mothers; the reverse is also true for the offspring of
high LG-ABN dams fostered by low LG-ABN mothers [66, 67].
The increased stress reactivity of the adult offspring of low LG-ABN
mothers is associated with elevated CRF gene expression in both the paraven-
tricular nucleus of the hypothalamus (PVN) and the amygdala [64] (fig. 1, 2).
CRF activates both behavioural and HPA responses to stress, and this effect is,
in part, mediated by CRF action at the noradrenergic cell bodies in the locus
coeruleus and the nucleus of the solitary tract [68, 69]. Predictably, the off-
spring of the low LG-ABN mothers show increased stress-induced release of
noradrenaline in the hypothalamus that then stimulates the release of CRF.
Increased noradrenaline release is also associated with fear, and the offspring of
low LG-ABN mothers show increased fearfulness in response to stress.
Environmental effects on the development of stress responses involve
systems that normally serve to constrain CRF synthesis and release, such as the
hippocampal glucocorticoid receptor system. Glucocorticoids inhibit CRF
mRNA expression in the PVN (i.e., glucocorticoid negative feedback; fig. 1).
The hippocampus is a critical site for feedback regulation [70]. The adult
offspring of high LG-ABN mothers show increased glucocorticoid receptor
Matthews/Meaney 38
mRNA in the hippocampus, increased feedback sensitivity to glucocorticoids
and thus reduced CRF mRNA levels in the PVN [64]. The GABA
A
receptor
system also inhibits CRF activity, particularly at the level of the amygdala and
locus coeruleus [71, 72] (fig. 2). Predictably, behavioural responses to stress
are inhibited by benzodiazepines (BZs), which exert their potent anxiolytic
effect by enhancing GABA-mediated Cl
currents through GABA

A
receptors
[73, 74] especially at the level of the amygdala. The adult offspring of
high and low LG-ABN mothers differ in GABA
A
/BZ receptor binding in the
amygdala [65].
The results of human studies are consistent with the idea that the quality of
postnatal parent-infant interactions influences the development of behavioural
and HPA responses to stress. Abuse in early life increases endocrine and auto-
nomic responses to stress in adulthood. DeBellis et al. [75], Heim et al. [76] and
Pruessner et al. [77] found that measures of maternal care predicted cortisol
responses to stress in healthy human subjects: low maternal care was associated
with significantly greater responses to stress. These findings suggest that
family distress can lead to impairments in child care and increased HPA
responses to adversity in the offspring. Such effects could compromise cogni-
tive function. In human subjects, acute elevations of glucocorticoids disrupt
hippocampal/prefrontal cortex function and thus learning/memory, while
chronic elevations in cortisol are associated with decreased hippocampal vol-
ume and more severe impairments of attention and learning [78]. There is a
remarkable potential for a feed-forward series of effects, since both the hip-
pocampus and the prefrontal cortex provide inhibitory regulation over HPA
activity [70, 79].
Parent-child interactions might influence cognitive and emotional develop-
ment through even more direct pathways. Attentional processes are regulated by
catecholaminergic innervation of the hippocampus and prefrontal cortex [80].
Stress increases dopamine and noradrenaline release and impairs attention
through actions on D
1
and
1
receptors in the prefrontal cortex. Increased
dopamine release at the level of the ventral striatum (or nucleus accumbens in
the rat) is associated with poor impulse control and hyperactivity. In the rat,
postnatal maternal separation increases dopamine release to stress and produces
behavioural hyperactivity [81]. Likewise, the offspring of low LG-ABN mothers
show increased stress-induced release of dopamine in the prefrontal cortex and
noradrenaline release in the hippocampus and prefrontal cortex [82]. Thus, in
rodents, mother-pup interactions in early life regulate the development of the
mesolimbic catecholamine systems. In humans, Pruessner et al. [77] found that
in adults the magnitude of stress-induced increases in dopamine release in the
ventral striatum is strongly correlated with the quality of maternal care. Poor
maternal care was associated with increased dopamine responses to stress in a
PET imaging study. Of course, the converse is also true: more nurturing forms
of maternal care are associated with more modest stress-induced increases in
cortisol and dopamine. Thus, parent-child interactions appear to influence the
development of both HPA and central catecholamine responses to stress and
might thus alter cognitive performance.
These findings are consistent with the clinical literature that has identified
ADD as a familial disorder [26]. Of course, disorders may run in families
because of environmental as well as genetic factors pathogenic families are
certainly defined by multiple gene environment interactions. Indeed, family
function and SES are important predictors of ADD [83, 84]. Family conflict,
parental psychopathology and low SES are significant predictors of ADD, each
accounting for a 7-fold increase in the risk for ADD. For children burdened with
all three risk factors, as so commonly occurs, the odds ratio for ADD was
increased by greater than 40-fold. These factors also predict impulse control
disorders [33, 85]. Moreover, amongst non-ADD controls, these same condi-
tions predicted learning disorders, reflecting a more general relationship
between familial environment and cognitive development.
Animal studies suggest that the quality of the maternal care exerts a
substantial influence on neural development, in part, through the regulation of
endocrine function in the pup. Tactile stimulation associated with maternal LG
dampens adrenal sensitivity to ACTH, thus limiting adrenal glucocorticoid
release even in response to acute stress. Prolonged deprivation of maternal care
increases glucocorticoid levels, an effect that is reversed with experimental
stroking, which mimics maternal LG. The next question is whether under
normal circumstances tactile stimulation derived from maternal LG might alter
neural development. A cDNA array study [86] showed that on day 6 offspring
of high LG-ABN mothers exhibited significantly (i.e., 3-fold levels)
increased expression of (1) genes related to cellular metabolic activity (glucose
transporter, fructose transporter, c-Fos, cytochrome oxygenase, LDL receptor,
growth hormone receptor, insulin receptor), (2) genes related to glutamate
receptor function, including NMDA receptor subunits and (3) genes encoding
for growth factors, including BDNF, bFGF and -NGF [87]. Both NMDA
receptor activation and neurotrophic factor expression are primary regulators of
neuronal development and thus the offspring of high LG-ABN mothers showed
increased neuronal survival, enhanced synaptic development, as well as
increased spatial learning/memory [88]. In contrast, maternal separation results
in elevated levels of glucocorticoids which downregulate the expression of
neurotrophic factors, such as BDNF and -NGF, and impair hippocampal
development [8991]. Postnatal adversity might therefore contribute to the risk
for cognitive impairments, such as ADD, through direct effects on the develop-
ment of neural structures such as the hippocampus.
Matthews/Meaney 40
Environmental Adversity and Parental Care
These findings suggest that positive forms of parental care can enhance
neural development, dampen catecholamine and HPA responses to stress, and
thus influence both emotional responses to stress and cognitive performance.
The next question concerns the origins of these variations in parental care. If
nurturing forms of parenting decrease the vulnerability to chronic illness and
enhance cognitive development, then why should all parents not behave in this
manner towards their children?
Environmental adversity can compromise the emotional well-being of the
parent and thus influence the quality of parent-child relationships. While many
factors contribute to the quality of the mothers attitude towards her newborn,
none are correlated more highly than the womens level of anxiety [92].
Mothers who feel depressed and anxious are, not surprisingly, less positive
towards their baby. High levels of maternal stress are associated with less sensi-
tive child care [93, 94]. The children of highly stressed primary caregivers tend
to develop more insecure parental attachment [94, 95], which predicts behav-
ioural inhibition in childhood and an increased risk for depression. Vaughn
et al. [95] found that unstable/stressful environments were associated with
greater variability in the quality of infant-mother attachments. When parents
suffer from poverty or other environmental stressors, they experience more
negative emotions, irritable, depressed and anxious moods, which lead to more
punitive parenting [9698]. The greater the number of environmental stressors
(e.g. lesser education of parents, low income, many children, being a single
parent), the less supportive the mothers were of their children; they were more
likely to threaten, push or grab them, and they displayed more controlling
attitudes toward child rearing. Generally, high levels of stress are associated
with negative outcomes for both mothers and children [94].
Family dysfunction and low income are associated with significantly
increased risks of psychopathology and cognitive development is inversely
related to familial adversity [99]. Interestingly, the effects of poverty on intel-
lectual and emotional development are mediated by variations in parental care.
When the parental care factor is statistically factored out of the equation, there
is no relationship between SES and child development [100102]. Low SES
serves as a potent source of adversity and Lupien et al. [103] found that
children from low SES homes exhibited significantly higher levels of basal
salivary cortisol by comparison to middle-class peers. Two points are impor-
tant: first, elevated basal cortisol levels are associated with hippocampal dys-
function and cognitive impairments [104] and second, the effect of SES
on basal cortisol levels was entirely accounted for by the level of maternal
depression.
Environmental adversity appears to compromise the quality of child
care and thus alters neural development. For humans, these are not isolated
conditions: 1 in 5 teenage females and 1 in 6 adult women experience abuse
during pregnancy [105, 106]. Such severe forms of stress promote emotional
states of depression and anxiety, and these states predict cold, punitive forms of
parenting. Not surprisingly, highly anxious mothers are more likely to have
children who are shy and timid, and the behaviour of the mother predicts the
level of such behavioural inhibition in the child [107]. Moreover, there is
evidence for the behavioural transmission of anxiety. Cross-fostering studies in
the rat and monkey show that parental care can serve as a mechanism for the
transmission of individual differences in stress reactivity from parent to
offspring [39, 66]. In rats, stress during the latter half of gestation produces sig-
nificant reductions in pup LG among high LG-ABN mothers, such that group
differences between high and low LG-ABN mothers are eliminated [108]; ges-
tationally stressed, high LG-ABN mothers lick/groom pups no more frequently
than do stressed or control low LG-ABN mothers. Predictably, the offspring of
stressed/high LG-ABN mothers do not differ from those of low LG-ABN moth-
ers on measures of hippocampal development or stress reactivity.
Of course, the influence of parental care can also function to enhance
cognitive development and dampen stress reactivity. Protective environmental
influences, such as social support, enhance the quality of parental care and thus
improve developmental outcomes for the child. More nurturing forms of mater-
nal care can enhance neurocognitive development and dampen emotional and
HPA responses to stress. These considerations thus raise the question of
whether a more nurturing home environment during postnatal life might offset
the risk-associated fetal adversity.
Prenatal/Postnatal Interactions and the Question
of Reversibility
The critical question here concerns the potential for the reversibility of
the developmental effects associated with adversity in fetal life. The results of
animal studies suggest ample potential for functional reversal. Postnatal han-
dling, which increases maternal LG, reverses the effects of prenatal stress on
HPA responses to stress, fearfulness and spatial learning/memory in the rat.
Postnatal handling of the excessively fearful and intellectually challenged
BALBc mouse produces a decrease in timidity and a marked improvement in
cognitive performance [109]. Interestingly, evidence from embryo transfer
experiments using BALBc and C57 mice [110] suggests that the increased fear-
fulness of the BALBc mouse develops during fetal life and is, in part at least,
Matthews/Meaney 42
independent of the genome. Thus, the handling studies with these mice repre-
sent an example of postnatal reversibility of impairments derived from fetal
life. Cross-fostering BALBc mice by C57 mothers also reverses the excessive
fearfulness of the BALBc offspring as well as differences in GABA
A
/BZ recep-
tor levels [111]. Importantly, C57 mothers lick/groom their pups about 34 times
more frequently than do BALBc dams.
Interestingly, animals that are more vulnerable by virtue of adversity in
fetal life may actually be more susceptible to the effects of postnatal conditions.
For example, the effects of postnatal handling are consistently greater in prena-
tally stressed than control animals [112]. In rhesus monkeys [39], anxious
infants cross-fostered by highly nurturing mothers showed dramatic decreases
in timidity and behavioral inhibition. Less anxious infants were unaffected by
maternal care. This same point emerges from studies of environmental enrich-
ment. Postweaning enrichment of the offspring of low LG-ABN mothers pro-
duces an increase in hippocampal synaptogenesis and cognitive performance,
with little or no effect on the offspring of high LG-ABN dams [88].
The same also emerges from studies of human populations. Belsky [98]
examined parenting styles over the 2nd and 3rd year of life in children, measur-
ing positive (the degree to which parents were affectionate and sensitive to the
child) and negative (hostile, irritable and intrusive) parenting and evaluated the
timidity/behavioural inhibition in the children at 36 months of age. Parental
style accounted for only 4% of the variance on the outcome measure among
children who were evaluated as low on negative emotionality at the beginning
of the study. Among those high in negative emotionality, parental style
accounted for almost 30% of the variance in behavioural inhibition. Likewise,
among children with a negative temperament in infancy, there are significant
effects of the quality of parental care or day care on emotional problems in
childhood; no such effects emerge amongst children exhibiting a positive tem-
perament in infancy [113]. The NICHD [114] revealed a highly significant rela-
tionship between parental sensitivity and emotional/behavioural disorders in
childhood, but only amongst those children with a negative temperament in
infancy. Morris et al. [115] found that hostile/harsh maternal care predicted
behavioural problems in children, but again, only in children scoring high on
irritability distress in infancy. The more phlegmatic infants were not signifi-
cantly affected by negative parental style.
These findings suggest that high-risk infants are more susceptible to the
influence of postnatal family life than are children low in vulnerability. Hence,
the proverbial 0.3 correlation that so routinely emerges between parental rear-
ing styles and developmental outcomes in children may likely be the result of a
zero-order association in the non-susceptible child, with associations in the
order of 0.5, 0.6 or greater in more susceptible (i.e., highly vulnerable) children.
If true, the implications for early childhood treatment programs are enormous.
The results of clinical intervention studies would seem to be consistent with this
line of reasoning. Indeed, the rather moderate correlations between maternal
care and infant attachment in the general population [116] contrast with clinical
data. For example Van den Boom [117] randomly assigned low SES mothers
with highly irritable infants to an experimental group that received an interven-
tion structured to promote maternal sensitivity or to control conditions. As
intended, the intervention significantly increased maternal sensitivity and
secure forms of infant attachment. Among controls, 22% of the infants showed
secure attachment, in contrast to 66% of those in the treatment group; a 300%
increase. These findings suggest that very impressive treatment effects are
observed when interventions focus on more vulnerable populations. Finally,
Blair [118] found that among low-birth-weight babies from economically dis-
advantaged homes an enriched form of education day care, which included
home visiting and parental support, significantly reduced the risk for emo-
tional/behavioural disorders, but only amongst those children that exhibited
highly negative emotionality in infancy; no treatment effect was detected
amongst children with normal temperament in infancy. Interestingly, the same
pattern of effects emerged in the analysis of cognitive outcomes. Among the
infants of poor temperament, those enrolled in the enrichment program were
5 times less likely to exhibit cognitive impairments (75 IQ) than those in the
control group; there were no treatment effects on cognitive development among
children with a normal or positive temperament in infancy. In each case, chil-
dren with a history of negative mood and irritability in infancy were most
affected by parental care. Evaluative research conducted with the Abecedarian
Project shows that early (years 14) enrichment interventions have profound
effects, in the order of 11.5 SD on IQ tests, in children from seriously disad-
vantaged homes. In children from more functional families and better educated
parents, the program had no effect on cognitive development [119].
Taken together, these findings also suggest that the outcomes associated with
fetal adversity are determined by the quality of the postnatal environment. Two
studies on developmental outcomes in growth-restricted babies bear directly on
this point. First, Barker et al. [120] found that the risk for cardiovascular disease
among men that were small for gestational age was apparent only in low SES
individuals; higher SES, low-birth-weight men were at no greater risk for heart
disease than individuals born at normal weight. Second, Kelly et al. [33] found
that the effects of fetal growth restriction on behavioural development were medi-
ated by SES. Small babies reared in lower SES conditions showed increased
evidence of behavioural disorders characterized by irritability and impulsivity.
There was no effect of birth weight on behavioural/emotional disorders among
children from higher SES settings.
Matthews/Meaney 44
Conclusions
Not surprisingly, there is strong evidence for cognitive/behavioural impair-
ments associated with fetal growth retardation. What is surprising is that not all
IUGR babies are seriously impaired, even under conditions that produce severe
growth restriction and prematurity. In populations of preterm, very-low-birth-
weight babies estimates are that 40% exhibit normal cognitive function in
childhood and school performance within the normal range. We propose that
the variability in the outcomes associated with fetal adversity is due to the influ-
ence of postnatal life.
References
1 Thompson C, Syddall H, Rodin I, Osmond C, Barker DJ: Birth weight and the risk of depressive
disorder in late life. Br J Psychiatry 2001;179:450455.
2 Seckl JR, Nyirenda MJ, Walker BR, Chapman KE: Glucocorticoids and fetal programming.
Biochem Soc Trans 1999;27:7478.
3 McTernan CL, Draper N, Nicholson H, Chalder SM, Driver P, Hewison M, Kilby MD, Stewart
PM: Reduced placental 11-hydroxysteroid dehydrogenase type 2 mRNA levels in human preg-
nancies complicated by intrauterine growth restriction: An analysis of possible mechanisms. J Clin
Endocrinol Metab 2001;86:49794983.
4 Seckl JR: Glucocorticoid programming of the fetus adult phenotypes and molecular mechanisms.
Mol Cell Endocrinol 2001;185:6171.
5 Goland RS, Jozak S, Warren WB, Conwell IM, Stark RI, Tropper PJ: Elevated levels of umbilical
cord plasma corticotropin-releasing hormone in growth-retarded fetuses. J Clin Endocrinol Metab
1993;77:11741179.
6 Phillips DJW: Birth weight and the future development of diabetes. Diabetes Care 1998;21:
B150B155.
7 Matthews SG: Early programming of the hypothalamo-pituitary-adrenal axis. Trends Endocrinol
Metab 2002;13:373380.
8 Lingas R, Matthews SG: A short period of maternal nutrient restriction in late gestation modifies
pituitary-adrenal function in adult guinea pig offspring. Neuroendocrinology 2001;73:302311.
9 Liu L, Li A, Matthews SG: Maternal glucocorticoid treatment programs HPA regulation in adult
offspring: Sex-specific effects. Am J Physiol Endocrinol Metab 2001;280:E729E739.
10 McCabe L, Marash D, Li A, Matthews SG: Repeated antenatal glucocorticoid treatment decreases
hypothalamic corticotropin releasing hormone mRNA but not corticosteroid receptor mRNA
expression in the fetal guinea-pig brain. J Neuroendocrinol 2001;13:425431.
11 Weinstock M: Does prenatal stress impair coping and regulation of hypothalamic-pituitary-adrenal
axis. Neurosci Biobehav Rev 1997;21:110.
12 Welberg LA, Seckl JR, Holmes MC: Prenatal glucocorticoid programming of brain corticosteroid
receptors and corticotrophin-releasing hormone: Possible implications for behaviour. Neuroscience
200;104:7179.
13 Uno H, Eisele S, Sakai A, Shelton S, Baker E, De Jesus O, Holden J: Neurotoxicity of glucocorti-
coids in the primate brain. Horm Behav 1994;28:336348.
14 Schneider ML, Moore CF, Kramer GW, Roberts AD, DeJesus OT: The impact of prenatal stress,
fetal alcohol exposure, or both on development: Perspectives from a primate model. Psycho-
neuroendocinology 2002;27:285298.
15 Lagercrantz H: Better born too soon than too small. Lancet 1997;350:10441045.
16 Seidman DS, Laor A, Gale R, Stevenson DK, Mashiach S, Danon YL: Birth weight and intellectual
performance in late adolescence. Obstet Gynecol 1992;79:543546.
17 Breslau N, Klein N, Allen L: Very low birthweight: Behavioural sequelae at nine years of age.
J Am Acad Child Adolesc Psychiatry 1988;27:605612.
18 Martyn CN, Gale CR, Sayer AA, Fail C: Growth in utero and cognitive function in adult life:
Follow-up study of people born between 1920 and 1943. Br Med J 1996;312:13931396.
19 Hutton JL, Pharoah POD, Cook RWI, Stevenson RC: Differential effects of preterm birth and
small for gestational age on cognitive and motor development. Arch Dis Child 1997;76:F75F81.
20 Landry SH, Denson SE, Swank PR: Effects of medical risk and socioeconomic status on the rate of
change in cognitive and social development for low birth weight children. J Clin Exp Neuropsychol
1997;19:261274.
21 Strauss RS: Adult functional outcome of those born small for gestational age: Twenty-six-year
follow-up of the 1970 British Birth Cohort. JAMA 2000;283:625632.
22 Glover V, OConnor TG: Effects of antenatal stress and anxiety: Implications for development and
psychiatry. Br J Psychiatry 2002;180:389391.
23 Van der Reijden-Lakeman IEA, de Sonneville LMJ, Swaab-Barneveld HJT, Slipper F, Verhulst
FC: Evaluation of attention before and after 2 years of growth hormone treatment in interuterine
growth retarded children. J Clin Exp Neuropsychol 1997;19:101118.
24 Holmes J, Payton A, Barrett JH, Hever T, Fitzpatrick H, Trumper AL, Harrington R, McGuffin P,
Owen M, Ollier W, Worthington J, Thapar A: A family-based and case-control association study of
the dopamine D
4
receptor gene and dopamine transporter gene in attention deficit hyperactivity
disorder. Mol Psychiatry 2000;5:523530.
25 Asghari V, Sanyal S, Buchwaldt S, Paterson A, Jovanovic V, Van Tol HH: Modulation of intracellular
cyclic AMP levels by different human dopamine D
4
receptor variants. J Neurochem 1995;65:157165.
26 Faraone SV, Biederman J: Genetics of attention-deficit hyperactivity disorder. Child Adolesc
Psychiatr Clin N Am 1994;3:285302.
27 Faraone SV, Doyle AE, Mick E, Biederman J: Meta-analysis of the association between the
7-repeat allele of the dopamine D
4
receptor gene and attention deficit hyperactivity disorder.
Am J Psychiatry 2001;158:10521057.
28 Faraone SV, Doyle AE: Genetic influences on attention deficit hyperactivity disorder. Curr
Psychiatry Rep 2000;2:143146.
29 Huizink AC, de Medina PG, Mulder EJ, Visser GH, Buitelaar JK: Psychological measures of pre-
natal stress as predictors of infant temperament. J Am Acad Child Adolesc Psychiatry
2002;41:10781085
30 Szatmari P, Saigal S, Rosenbaum P, Campbell D, King S: Psychiatric disorders at five years among
children with birthweights 1,000g: A regional perspective. Dev Med Child Neurol 1990;32:
954962.
31 Pharoah POD, Stevenson CJ, Cooke RWI, Stevenson RC: Prevalence of behaviour disorders in low
birthweight infants. Arch Dis Child 1994;70:271274.
32 Sykes DH, Hoy EA, Bill JM, McClure BG, Halliday HL, Reid MM: Behavioural adjustment in
school of very low birth weight children. J Child Psychol Psychiatry 1997;38:315325.
33 Kelly YJ, Nazroo JY, McMunn A, Boreham R, Marmot M: Birthweight and behavioural problems
in children: A modifiable effect? Int J Epidemiol 2001;30:8894.
34 Nilsson PM, Nyberg P, Ostergren PO: Increased susceptibility to stress at a psychological assess-
ment of stress tolerance is associated with impaired fetal growth. Int J Epidemiol 2001;30:7580.
35 Ebstein RP, Levine J, Geller V, Auerbach J, Gritsenko I, Belmaker RH: Dopamine D
4
receptor and
serotonin transporter promoter in the determination of neonatal temperament. Mol Psychiatry
1998;3:238246.
36 Arbelle S, Benjamin J, Golin M, Kremer I, Belmaker RH, Ebstein RP: Relation of shyness in
grade school children to the genotype for the long form of the serotonin transporter promoter
region polymorphism. Am J Psychiatry 2003,160:671676.
37 Lakatos K, Nemoda Z, Toth I, Ronai Z, Ney K, Sasvari-Szekely M, Gervai J: Further evidence for
the role of the dopamine D
4
receptor (DRD4) gene in attachment disorganization: Interaction of the
exon III 48-bp repeat and the -521 C/T promoter polymorphisms. Mol Psychiatry 2002;7:2731.
38 Lakatos K, Toth I, Nemoda Z, Ney K, Sasvari-Szekely M, Gervai J: Dopamine D
4
receptor
(DRD4) gene polymorphism is associated with attachment disorganization in infants. Mol
Matthews/Meaney 46
39 Suomi SJ: Early determinants of behaviour: Evidence from primate studies. Br Med Bull 1997;53:
170184.
40 Bifulco A, Brown GW, Adler Z: Early sexual abuse and clinical depression in adult life.
Br J Psychiatry 1991;159:115122.
41 Brown GR, Anderson B: Psychiatric morbidity in adult inpatients with childhood histories of
sexual and physical abuse. Am J Psychiatry 1993;148:5561.
42 McCauley J, Kern DE, Kolodner K, Dill L, Schroeder AF, DeChant HK, Ryden J, Derogatis LR,
Bass EB: Clinical characteristics of women with a history of childhood abuse: Unhealed wounds.
JAMA 1997;277:13621368.
43 Felitti VJ, Anda RF, Nordenberg D, Williamson DF, Spitz AM, Edwards V, Koss MP, Marks JS:
Relationship of childhood abuse and household dysfunction to many of the leading causes of
death in adults. Am J Prev Med 1998;14:245258.
44 Montgomery SM, Bartley MJ, Wilkinson RG: Family conflict and slow growth. Arch Dis Child
1997;77:326330.
45 Ammerman RT, Cassisi JE, Hersen M, van Hasselt VB: Consequences of physical abuse and
neglect in children. Clin Psychol Rev 1986;6:291310.
46 Trickett PK, McBride-Chang C: The developmental impact of different forms of child abuse and
neglect. Dev Rev 1995;15:311337.
47 Lissau I, Sorensen TIA: Parental neglect during childhood and increased risk of obesity in young
adulthood. Lancet 1994;343:324327.
48 Holmes SJ, Robins LN: The influence of childhood disciplinary experience on the development of
alcoholism and depression. J Child Psychol Psychiatry 1987;28:399415.
49 Holmes SJ, Robins LN: The role of parental disciplinary practices in the development of depression
and alcoholism. Psychiatry 1988;51:2436.
50 Gottman JM: Psychology and the study of marital processes. Annu Rev Psychol 1998;49:
169197.
51 Canetti L, Bachar E, Galili-Weisstub E, De-Nour AK, Shalev AY: Parental bonding and mental
health in adolescence. Adolescence 1997;32:381394.
52 Parker G: Parental representations of patients with anxiety neurosis. Acta Psychiatr Scand
1981;63:3336.
53 Russak LG, Schwartz GE: Feelings of parental care predict health status in midlife: A 35-year
follow-up of the Harvard Mastery of Stress Study. J Behav Med 1997;20:111.
54 Rutter M: Protective factors in childrens responses to stress and disadvantage; in Kent MW, Rolf
JE (eds): Primary Prevention of Psychopathology: Social Competence in Children. Hanover,
University Press of New England, 1979, vol 3, pp 4974.
55 Smith J, Prior M: Temperament and stress resilience in school-age children: A within-families
study. J Am Acad Child Adolesc Psychiatry 1995;34:168179.
56 Seckl JR, Meaney MJ: Early life events and later development of ischaemic heart disease. Lancet
1994;342:1236.
57 Nemeroff CB: The corticotropin-releasing factor (CRF) hypothesis of depression: New findings
and new directions. Mol Psychiatry 1996;1:336342.
58 Francis D, Meaney MJ: Maternal care and the development of stress responses. Curr Opin
Neurobiol 1999;9:128134.
59 McEwen BS: Protective and damaging effects of stress mediators. N Engl J Med 1998;338:171179.
60 Plotsky PM, Meaney MJ: Early, postnatal experience alters hypothalamic corticotropin-releasing
factor (CRF) mRNA, median eminence CRF content and stress-induced release in adult rats.
Brain Res Mol Brain Res 1993;18:195200.
61 Ladd CO, Huot RL, Thrivikraman T, Nemeroff CB, Meaney MJ, Plotsky PM: Long-term behav-
ioral and neuroendocrine adaptations to adverse early experience. Prog Brain Res 2000;22:
79101.
62 Higley JD, Haser MF, Suomi SJ, Linnoila M: Nonhuman primate model of alcohol abuse: Effects
of early experience, personality and stress on alcohol consumption. Proc Natl Acad Sci USA
1991;88:72617265.
63 Meaney MJ: The development of individual differences in behavioral and endocrine responses to
stress. Annu Rev Neurosci 2001;24:11611192.
64 Liu D, Diorio J, Tannenbaum B, Caldji C, Francis D, Freedman A, Sharma S, Pearson D, Plotsky
PM, Meaney MJ: Maternal care, hippocampal glucocorticoid receptors and hypothalamic-
pituitary-adrenal responses to stress. Science 1997;277:16591662.
65 Caldji C, Tannenbaum B, Sharma S, Francis D, Plotsky PM, Meaney MJ: Maternal care during
infancy regulates the development of neural systems mediating the expression of behavioral
fearfulness in adulthood in the rat. Proc Natl Acad Sci USA 1998;95:53355340.
66 Francis DD, Diorio J, Liu D, Meaney MJ: Nongenomic transmission across generations in maternal
behavior and stress responses in the rat. Science 1999;286:11551158.
67 Caldji C, Diorio J, Meaney MJ: Variations in maternal care alter GABA
A
receptor subunit expression
in brain regions associated with fear. Neuropsychopharmacology 2003;23:150159.
68 Valentino RJ, Curtis AL, Page ME, Pavcovich LA, Florin-Lechner SM: Activation of the locus
ceruleus brain noradrenergic system during stress: Circuitry, consequences, and regulation. Adv
Pharmacol 1998;42:781784.
69 Bakshi VP, Shelton SE, Kalin NH: Neurobiological correlates of defensive behaviors. Prog Brain
Res 2000;122:105115.
70 De Kloet ER, Vreugdenhil E, Oitzl MS, Joels M: Brain corticosteroid receptor balance in health
and disease. Endocr Rev 1998;19:269301.
71 Owens MJ, Vargas MA, Knight DL, Nemeroff CB: The effects of alprazolam on corticotropin-
releasing factor neurons in the rat brain: Acute time course, chronic treatment and abrupt with-
drawal. J Pharmacol Exp Ther 1991;258:349356.
72 Skelton KH, Nemeroff CB, Knight DL, Owens MJ: Chronic administration of the triazolobenzo-
diazepine alprazolam produces opposite effects on corticotropin-releasing factor and urocortin
neuronal systems. J Neurosci 2000;20:12401248.
73 McKernan RM, Whiting PJ: Which GABA
A
-receptor subtypes really occur in the brain? Trends
Neurosci 1997;19:139143.
74 Mehta AK, Ticku MK: An update on GABA
A
receptors. Brain Res Brain Res Rev 1999;29:
196217.
75 DeBellis MD, Chrousos GP, Dom LD, Burke L, Helmers K, Kling MA, Trickett PK, Putnam FW:
Hypothalamic pituitary adrenal dysregulation in sexually abused girls. J Clin Endocrinol Metab
1994;78:249255.
76 Heim C, Newport DJ, Heit S, Graham YP, Wilcox M, Bonsall R, Miller AH, Nemeroff CB:
Pituitary-adrenal and autonomic responses to stress in women after sexual and physical abuse in
childhood. JAMA 2000;284:592597.
77 Pruessner JL, Champagne FA, Meaney MJ, Dagher A: Parental care and neuroendocrine
and dopamine responses to stress in humans: A PET imaging study. J Neurosci 2004;24:
28252831.
78 Lupien S, Meaney MJ: Stress, glucorticoids, and hippocampus aging in rat and human; in Wang E,
Snyder S (eds): Handbook of the Aging Brain. San Diego, Academic Press, 1998, pp 1950.
79 Diorio D, Viau V, Meaney MJ: The role of the medial prefrontal cortex (cingulate gyrus)
in the regulation of hypothalamic-pituitary-adrenal responses to stress. J Neurosci 1993;13:
38393849.
80 Arnsten AFT: Stress impairs prefrontal cortical function in rats and monkeys: Role of dopamine
D
1
and norepinephrine -1 receptor mechanisms. Prog Brain Res 2000;126:183192.
81 Meaney MJ, Brake W, Gratton A: Environmental regulation of the development of mesolimbic
dopamine systems: A neurobiological mechanism for vulnerability to drug abuse? Psychoneuro-
endocrinology 2002;27:127138.
82 Lovic V, Fleming AS: Artificially reared female rats show reduced prepulse inhibition and deficits
in the attentional set shifting task Reversal of effects with maternal-like licking stimulation.
Behav Brain Res 2004;148:209219.
83 Barkley RA, Anastopoulos AD, Guevremont DC, Fletcher KE: Adolescents with attention deficit
hyperactivity disorder: Mother-adolescent interactions, family beliefs and conflicts, and maternal
psychopathology. J Abnorm Child Psychol 1992;20:263288.
84 Biederman J, Milberger S, Faraone SV, Kiely K, Guite J, Mick E, Ablon S, Warburton R, Reed E:
Family-environment risk factors for attention-deficit hyperactivity disorder. A test of Rutters indi-
cators of adversity. Arch Gen Psychiatry 1995;52:464470.
Matthews/Meaney 48
85 Barkley RA: Attention Deficit Hyperactivity Disorder: A Handbook for Diagnosis and Treatment.
New York, Guilford Press, 1990.
86 Diorio J, Weaver ICG, Meaney MJ: A DNA array study of hippocampal gene expression regulated
by maternal behavior in infancy. Abstr Soc Neurosci 2000;26:1366.
87 Liu D, Diorio J, Day JC, Francis DD, Mar A, Meaney MJ: Maternal care, hippocampal synaptoge-
nesis and cognitive development in the rat. Nat Neurosci 2000;3:799806.
88 Bredy TW, Humpartzoomian RA, Cain DP, Meaney MJ: The influence of maternal care and
environmental enrichment on hippocampal development and function in the rat. Neuroscience
2003;118:571576.
89 Rocceri M, Hendriks W, Racagni G, Ellenbroek BA, Riva MA: Early maternal deprivation reduces
the expression of BDNF and NMDA receptor subunits in rat hippocampus. Mol Psychiatry
2002;7:609616.
90 Huot RL, Plotsky PM, Lenox RH, McNamara RK: Neonatal maternal separation reduces
hippocampal mossy fiber density in adult Long Evans rats. Brain Res 2002;950:5263.
91 Zhang LX, Levine S, Dent G, Zhan Y, Xing G, Okimoto D, Kathleen Gordon M, Post RM, Smith
MA: Maternal deprivation increases cell death in the infant rat brain. Dev Brain Res 2002;133:111.
92 Fleming AS: Factors influencing maternal responsiveness in humans: Usefulness of an animal
model. Psychoneuroendocrinology 1988;13:189212.
93 Dix T: The affective organization of parenting adaptive and maladaptive processes. Psychol Bull
1991;110:325.
94 Goldstein LH, Diener ML, Mangelsdorf SC: Maternal characteristics and social support across
the transition to motherhood: Associations with maternal behavior. J Fam Psychol 1996;10:
6071.
95 Vaughn B, Egeland B, Sroufe LA, Waters E: Individual differences in infant-mother attachment at
twelve and eighteen months: Stability and change in families under stress. Child Dev 1979;50:
971975.
96 Conger RD, McCarty JA, Yang RK, Lahey BB, Kropp JP: Perception of child, childrearing values,
and emotional distress as mediating links between environmental stressors and observed maternal
behavior. Child Dev 1984;55:22342247.
97 Grolnick WS, Gurland ST, DeCourcey W, Jacob K: Antecedents and consequences of mothers
autonomy support: An experimental investigation. Dev Psychol 2002;38:143155.
98 Belsky J: Theory testing, effect-size evaluation, and differential susceptibility to rearing influence:
The case of mothering and attachment. Child Dev 1997;64:598600.
99 Offord DR, Boyle MH, Racine YA, Fleming JE, Cadman DT, Blum HM, Byrne C, Links PS,
Lipman EL, MacMillan HL, et al: Outcome, prognosis, and risk in a longitudinal follow-up study.
J Am Acad Child Adolesc Psychiatry 1992;31:916923.
100 Eisenberg L, Earls FJ: Poverty, social depreciation and child development; in Hamburg DA (ed):
American Handbook of Psychiatry. New York, Basic Books, 1975, vol 6, pp 275291.
101 Conger R, Ge X, Elder G, Lorenz F, Simons R: Economic stress, coercive family process and
developmental problems of adolescents. Child Dev 1994;65:541561.
102 McLloyd VC: Socioeconomic disadvantage and child development. Am Psychol 1998;53:185204.
103 Lupien SJ, King S, Meaney MJ, McEwen BS: Childs stress hormone levels correlate with
mothers socioeconomic status and depressive state. Biol Psychiatry 2000;48:976980.
104 Lupien SJ, Nair NPV, Briere S, Maheu F, Tu MT, Lemay M, McEwen BS, Meaney MJ: Increased
cortisol levels and impaired cognition in human aging: Implications for depression and dementia
in later life. Rev Neurosci 1999;10:117139.
105 Newberger EH, Barkan SE, Lieberman ES, et al: Abuse of pregnant women and adverse birth out-
comes. Current knowledge and implications for practice. JAMA 1992;267:23702372.
106 Parker B, McFarlane, Soeken K: Abuse during pregnancy: Effects on maternal complications and
birth weight in adult and teenage women. Obstet Gynecol 1994;84:323328.
107 Hirschfield DR, Biederman J, Brody L, Faraone SV, Rosenbaum JF: Expressed emotion towards
children with behavioral inhibition: Association with maternal anxiety disorder. J Am Acad Child
Adolesc Psychiatry 1997;36:910917.
108 Champagne FA, Meaney MJ: Stress during gestation alters postpartum maternal care and the
development of the offspring in a rodent model. Submitted.
109 Zaharia MD, Kulczycki J, Shanks N, Meaney MJ, Anisman H: The effects of early postnatal stim-
ulation on Morris water-maze acquisition in adult mice: Genetic and maternal factors.
Psychopharmacology (Berl) 1996;128:227239.
110 Francis DD, Szegda K, Campbell G, Martin WD, Insel TR: Epigenetic sources of behavioral
differences in mice. Nat Neurosci 2003;6:445446.
111 Caldji C, Diorio J, Anisman H, Meaney MJ: Maternal behavior regulates benzodiazepine/GABA
A
receptor subunit expression in brain regions associated with fear in BALB/c and C57BL/6 mice.
Neuropsychopharmacology 2004;29:13441352.
112 Smythe JW, McCormick CM, Meaney MJ: Median eminence corticotrophin-releasing hormone
content following prenatal stress and neonatal handling. Brain Res Bull 1996;40:195199.
113 Deater-Deckard K, Dodge K: Spare the rod, spoil the authors: Emerging themes in research on
parenting. Psychol Inquiry 1997;8:230235.
114 NICHD Early Child Care Research Network: Early child care and self-control, compliance, and
problem behavior at 24 and 36 months. Child Dev 1998;69:11451170.
115 Morris A, Silk J, Steinberg L, Sessa F, Avenevoli S, Essex M: Temperamental vulnerability and
negative parenting as interactings predictors of child adjustment. J Marriage Fam 2002;64:
461471.
116 De Wolff M, van Ijzendoorn M: Sensitivity and attachment: A meta-analysis on parental
antecedents of infant attachment. Child Dev 1997;68:571591.
117 Van den Boom D: The influence of temperament and mothering on attachment and exploration:
An experimental manipulation of sensitive responsiveness among lower-class mothers and irrita-
ble infants. Child Dev 1994;65:14571477.
118 Blair C: Early intervention for low birth weight preterm infants: The role of negative emotionality
in the specification of effects. Dev Psychopathol 2002;14:311332.
119 Ramey CT, Ramey SL: Early intervention and early experience. Am Psychol 1998;53:109120.
120 Barker DJ, Forsen T, Eriksson JG, Osmond C: Growth and living conditions in childhood and
hypertension in adult life: A longitudinal study. J Hypertens 2002;20:19511956.
Dr. Stephen G. Matthews
Department of Physiology, Faculty of Medicine, University of Toronto
Medical Sciences Building, 1 Kings College Circle
Toronto, ON M5S 1A8 (Canada)
Tel. 1 416 978 1974, Fax 1 416 978 4940, E-Mail Stephen.matthews@utoronto.ca
Behavioral Perinatology
Pathik D. Wadhwa
Departments of Psychiatry & Human Behavior and Obstetrics & Gynecology,
University of California, Irvine, Calif., USA
Developmental processes involved in transforming a single-cell human
embryo into a fully functioning organism within a mere span of 40 weeks are
exceedingly complex and fascinating; indeed, one would be hard pressed to
come up with any other example in the physical or biological world that even
begins to approximate the sheer elegance of intrauterine development.
Biologists over the ages have asked the question: does the genetic material of
the fertilized egg already contain a full set of building specifications for the
organism? Over the last decade or so, there has been a major paradigm shift in
developmental biology regarding fundamental concepts of how the central
nervous system and the rest of the organism develops and functions. The answer
to the above question is now believed to be an unequivocal no. Genes and
environment are no longer considered to exert separate influences, and develop-
ment is viewed not as a gradual elaboration of an architectural plan preconfig-
ured in the genes, but rather as a dynamic interdependency of genes and
environment characterized by a continuous process of interactions in a place-
and time-specific dependent manner, and involving short- and long-term infor-
mation storage, whereby genetic and epigenetic processes
1
, at every step of
development, become represented in the evolving structural and functional
design of the organism [24]. According to this epigenetic view of develop-
ment, events at one point in time have consequences that are manifested later in
Major sections of this chapter are reprinted from an article that appeared in Wadhwa
et al. [1], and are reproduced here with permission from Elsevier.
1
For the purpose of this discussion, we use the term genetic to refer to the effects of
variations in DNA sequences on protein physiology, and the term epigenetic to refer to
alterations in gene expression and protein physiology without changes in DNA sequences
(e.g. genetic imprinting via DNA methylation).
Behavioral Perinatology 51
the developmental process, and afferent activity has a profound influence on
the developmental trajectory [5]. In other words, it appears that within the
constraints imposed by the heritable germ line at conception, each developing
organism plays an active role in its own construction. This dynamic process is
effected by evolving various systems during embryonic and fetal life to acquire
information about the nature of the environment, and to use this information to
guide development. In the context of this formulation, not only does environ-
ment play a necessary role for development to occur, but the nature of the
environment may play either an advantageous role for normal or optimal develop-
ment, or may play a pernicious role to harm development [6]. The degree of
congruence or discongruence between fetal, childhood and adult environments
determines trajectories leading to either optimal or suboptimal developmental
and health outcomes [79].
Behavioral perinatology is broadly defined as an interdisciplinary area of
research that involves conceptualization of theoretical models and conduct of
empirical studies of the dynamic time-, place-, and context-dependent interplay
between biological and behavioral processes in fetal, neonatal and infant life
using an epigenetic framework of development. The biobehavioral processes of
particular interest to our research group relate to the effects of maternal
pre- and perinatal stress and maternal-placental-fetal stress physiology. Our
choice of stress and stress physiology is guided by the following two major
considerations: first, empirical studies in humans and animals support a signif-
icant role for pre- and perinatal stress as an independent risk factor for adverse
developmental outcomes [10]; second, stress and stress physiology offer an
excellent model system for the study of early developmental processes because
it appears that the developing fetus acquires and incorporates information
about the nature of its environment via the same systems that in a developed
individual are known to mediate adaptation and central and peripheral
responses to challenge/stress (i.e., the neuroendocrine, immune and vascular
systems) [11].
We propose that behavioral perinatology research may have important
implications for a better understanding of the processes that underlie or con-
tribute to the risk of at least three sets of outcomes: prematurity, adverse neu-
rodevelopment, and chronic degenerative diseases in adulthood. Each of these
classes of adverse health outcomes represent major public health issues in the
United States and other developed nations, their prevalence is characterized by
substantial disparities along factors associated with sociodemographic disad-
vantage and racial/ethnic minority status (which we and others have argued
may, in part, reflect the effects of variations in stress and stress physiology in
affected populations), and growing evidence supports a crucial role for an early
developmental process in their origins [5, 8, 9, 1214].
Wadhwa 52
Biobehavioral Model of Prenatal Stress and Stress
Physiology in Human Fetal Development
From a biological perspective, the term stress is used to describe any phys-
ical or psychological challenge that threatens or is perceived to have the potential
to threaten the stability of the internal milieu of the organism (homeostasis). The
neuroendocrine, immune and vascular systems play a major role in adaptation to
stress. The principal effectors of these adaptive responses are the corticotropin-
releasing hormone (CRH) and locus ceruleus-noradrenaline/autonomic (sympa-
thetic) neurons in the hypothalamus and brain stem, which regulate the peripheral
activities of the hypothalamic-pituitary-adrenal (HPA) axis and the systemic/
adrenomedullary sympathetic nervous system, respectively. Activation of the HPA
axis and locus ceruleus-noradrenaline/autonomic system results in the systemic
elevation of glucocorticoids and catecholamines, respectively, which act in concert
on target tissues to mobilize and redistribute available resources, and also to
maintain or effect a return to the state of homeostasis [15].
The adoption of an epigenetic framework for early development, wherein the
organism plays an active role in its own construction by evolving systems to
acquire and use information about the nature of the environment to guide devel-
opment, gives rise to two important questions. First, how do the fetal and maternal
compartments communicate with one another? And second, in light of the fact
that the fetal nervous system is itself in a state of evolution and has yet to acquire
its full repertoire of structural and functional capabilities, what are the modalities
available to the developing fetus to receive, process and act on information
acquired from the environment? There are no direct neural, vascular or other con-
nections between the mother and her developing fetus. One of the remarkable
adaptations of pregnancy is the evolution in early gestation of a transient organ of
fetal origin the placenta. All communication between the maternal and fetal
compartments is mediated via the placenta through one or both of two mecha-
nisms: the actions of maternal and fetal factors on placental activity, or transpla-
cental passage of blood-borne substances [13]. In addition to the long-recognized
multiple roles played by the placenta, it now appears that the placenta may also
take on some functions that are usually ascribed to the central nervous system, i.e.,
the capability of receiving, processing and acting upon certain classes of stimuli.
Indeed, we propose one of the important roles of the placenta is to act on behalf of
the fetus as both a sensory and effector organ to facilitate the transduction and
incorporation of environmental information into the developmental process.
Based on our understanding of the ontogeny of human fetal development
and physiology of pregnancy and fetal development, we have articulated a
neurobiological model of pre- and perinatal stress. Our model proposes that
chronic maternal stress may exert a significant influence on fetal developmental
outcomes [810]. The effects of maternal stress may be mediated through
biological and/or behavioral mechanisms. Maternal stress may act via one or
more of three major physiological pathways: neuroendocrine, immune/inflam-
matory, and vascular. We further suggest that placental CRH plays a central role
in coordinating the effects of endocrine, immune/inflammatory and vascular
processes on fetal developmental outcomes. Finally, we hypothesize that the effects
of maternal stress are modulated by the nature, duration and timing of occur-
rence of stress during gestation (fig. 1). We specifically postulate that individual
differences in maternal stress appraisals exert a larger impact than exposure per se
to stressful events; that prenatal stress in early gestation exerts a larger impact
on outcomes related to the length of gestation and fetal growth than stress in the
latter part of gestation; that among spontaneous births (i.e., those following
spontaneous labor or rupture of fetal membranes) prenatal stress directly influ-
ences the length of gestation, whereas among elective births, prenatal stress
indirectly influences the length of gestation by contributing to increased risk
of obstetric complications (e.g. preeclampsia) that are indicators for elective
delivery, and that the maternal-placental-fetal neuroendocrine system is the
primary physiological mediator of the effects of prenatal stress on adverse fetal
outcomes because it constitutes the fundamental substrate for fetal growth,
development and parturition, and because pathways through which alterations in
other systems (e.g. immune, vascular) produce pathophysiological consequences
are mediated, in part, by maternal-placental-fetal neuroendocrine processes.
Maternal prenatal stress
nature, timing, duration
Immune Neuro-endocrine Vascular
Utero-placental function
Intrauterine
environment
Fetal
environment
Fetal, newborn and infant development and health
Fig. 1. Biobehavioral model of prenatal stress and human fetal developmental and
health outcomes.
Wadhwa 54
Starting very early in gestation, the placenta produces hormones, neuro-
peptides, growth factors and cytokines, and appears to function in a manner
resembling that of compressed hypothalamic-pituitary target systems [16]. The
physiology of placental CRH serves as an excellent illustration of our concept
that the placenta acts in some ways as a sensory and effector organ on behalf of
the fetus. CRH, a 41-amino acid neuropeptide of predominantly hypothalamic
origin, is one of the primary mediators through which the brain regulates the
activity of the HPA axis and the physiological responses to stress and inflam-
mation [17, 18]. During human pregnancy, the CRH gene and receptors are also
richly expressed in the placenta. Placental CRH is identical to hypothalamic
CRH in structure, immunoreactivity, and bioactivity [19]. The expression of the
CRH gene increases exponentially in the placenta over the course of gestation,
resulting in the production of placental CRH and its release into the maternal
and fetal compartments (fig. 2). With respect to the role of the placenta as an
effector of fetal development, we and others have proposed various crucial roles
for placental CRH in regulating human reproductive biology, including implan-
tation [20], modulation of maternal and fetal pituitary-adrenal function, partic-
ipation in fetal cellular differentiation, growth and maturation, and involvement
in the physiology of parturition [19, 2123]. With respect to the role of the
placenta as a sensory organ, several lines of evidence have now converged to
suggest that the activity of placental CRH is, in turn, regulated by characteris-
tics of the maternal and intrauterine environment. For example, in vitro and
in vivo studies have demonstrated placental CRH output is modulated in a
Non-pregnant state
ACTH
Cortisol
CRH
Placental CRH in human pregnancy
Cortisol
CRH
ACTH
Pregnant state
CRH
Placenta
Fig. 2. Placental CRH in human pregnancy. In pregnancy, the placenta is a major

extrahypothalamic site for CRH production and action. In contrast to the negative control
exerted on the brain and pituitary, cortisol stimulates the production of CRH in the placenta,
establishing a positive feedback loop that terminates upon delivery.
positive, dose-response manner by all the major biological effectors of stress,
including cortisol, catecholamines, oxytocin, angiotensin II, both forms of
interleukin (IL)-1, and hypoxia [19, 24, 25].
Prenatal Stress and Fetal Developmental Outcomes:
Overview of Epidemiological Findings
Disruption of reproductive function in mammals is a well-known conse-
quence of stress. Results from experimental approaches in animal models strongly
support a causal role for prenatal stress as a developmental teratogen, with
large effects of even relatively mild behavioral perturbation in pregnancy on
outcomes including, but not limited to, maternal-fetal physiology, length of
gestation and fetal growth [10, 2629]. Psychosocial/behavioral stress in
human pregnancy has also been associated with outcomes at various points
along the developmental continuum, including fertilization and conception,
early pregnancy loss (spontaneous abortion), fetal structural and functional
developmental outcomes (malformations, physiological activity, neurobehav-
ioral maturation, growth), the length of gestation, infant birth weight, neonatal
neurological optimality, neonatal complications, infant neurodevelopmental
indices related to cognition, affect and behavior, and childhood and adult psy-
chopathology [for a review, see ref. 10]. In humans, the length of gestation and
fetal growth/infant birth weight are the outcomes that have been most
commonly studied and found to be associated with maternal stress during
pregnancy. We have recently conducted a comprehensive review of human empi-
rical research published in English language journals over the past 14 years
(19902004) and identified empirical reports that examined the association of
maternal psychosocial stress and/or social support with pregnancy outcomes
related to the length of gestation and birth weight. Findings from this review are
consistent with our own previously published studies in this area [3033] and
support the notion that pregnant women reporting high levels of psychosocial
stress and/or low levels of social support during pregnancy are significantly
more likely to deliver earlier/preterm and to deliver a smaller/low-birth-weight
infant. Moreover, the effects of maternal stress appear to be independent from
those of other established obstetric and sociodemographic risk factors. The
effects of maternal stress are observed across the entire range of the out-
come distribution, as opposed to only at one end of the distribution. Subjective
measures of stress perceptions and appraisals are more strongly associated with
adverse outcomes than measures of exposure to potentially stressful events or
conditions. In many instances, the effects of stress are moderated by other
person or situation characteristics, such as maternal age, body mass index,
Wadhwa 56
occupation, personality and coping styles. In terms of the magnitude of the
effect, pregnant women reporting high levels of stress are at approximately
doubled risk for preterm birth or fetal growth restriction compared to women
reporting low levels of stress (the adjusted relative risk ratios vary between
1.5 and 2.5).
Based on these findings, we suggest the following two implications: first,
maternal psychosocial processes in pregnancy are at least as important and
warrant the same degree of further consideration and study as other established
obstetric risk factors because the overall magnitude of their independent effect
size on prematurity-related outcomes is comparable to that of most other obstetric
risk factors; second, the translation of this knowledge to clinical application for
risk assessment and intervention has, however, been limited by the fact that
commonly used measures of maternal stress and related constructs have low
sensitivity and specificity in predicting adverse outcomes at the individual (as
opposed to population) level. Thus, there is a need to better capture and quantify
determinants of inter- and intraindividual variability in the link between maternal
stress and adverse birth outcomes, to identify which subgroup(s) of pregnant
women, under what circumstances, and at what stage in pregnancy, are especially
susceptible to the deleterious effects of prenatal stress on birth outcomes. For any
individual, the probability of a stress-related adverse health outcome is a joint
function of not only cumulative stress exposure but also individuals biological
responsivity to stress. Biological stress responsivity refers to an individuals
propensity for biological perturbation upon stress exposure. Two major limita-
tions of the maternal stress and birth outcome literature are that (a) previous stud-
ies have considered only the stress exposure side of the above equation, but not
the issue of individual differences in biological responsivity to stress, and (b) the
approach to the measurement of maternal stress has relied exclusively on self-
report, retrospective recall measures of psychological state or affect over time,
which may be prone to numerous biases that undermine validity. Self-report,
summary measures of an individuals states and experiences over time rely on
autobiographical memory (as opposed to semantic memory), which is as much a
matter of reconstruction as of accurate recall, and is known to be highly suscepti-
ble to numerous, systematic biases that adversely impact accuracy [34, 35].
Prenatal Stress and Physiological Processes in
Human Fetal Development: Role of Placental
Corticotropin-Releasing Hormone
Fetal growth and development involves a complex interplay of factors and
signaling molecules within the maternal, placental and fetal tissues. Pregnancy
is associated with major alterations in physiological function, including changes
in hormone levels and control mechanisms (feedback loops) that are crucial
in providing a favorable environment within the uterus and fetus for cellu-
lar growth and maturation and conveying signals when the fetus is ready for
extrauterine existence [16]. Fetal maturation and parturition are tightly syn-
chronized processes. Recent advances have implicated placental CRH as one of
the primary endocrine mediators of parturition and fetal development (fig. 3)
[for reviews, see ref. 19, 21, 3638].
Placental Corticotropin-Releasing Hormone and Parturition
It is well recognized that a shift in the balance from a progesterone-
dominant to an estrogen-dominant milieu over the course of gestation results
in a sequence of events in the gestational tissues to promote labor, including
gap junction formation, expression of oxytocin receptors, and synthesis of
prostaglandins [for reviews, see ref. 19, 21]. In most mammals, this shift is
effected by the conversion of progesterone to estrogen in the placenta. However,
unlike most other mammals, the human (primate) placenta cannot convert
progesterone to estrogen because it lacks the enzyme 17-hydroxylase requi-
red for this conversion. Instead, the placenta utilizes another precursor hormone
dehydroepiandrosterone sulfate (DHEA-S) which is produced by the
fetal adrenal zone, to synthesize estrogen [estriol (E
3
)] [for reviews, see
ref. 3841].
Fetus Placenta Mother
ACTH
DHEA-S
Liver
Lungs
Gut
Progesterone
Estrogens
Cholesterol
Cortisol
CRH
Myometrial contractility
cervical softening
CRH
Cortisol
CRH
ACTH
11-HSD 1 and 2
Fig. 3. Maternal-placental-fetal neuroendocrine axis in human pregnancy and fetal
development.
Wadhwa 58
Placental CRH is believed to coordinate and control the physiology of
parturition via its actions on the fetal endocrine system (fetal HPA axis) and
also within the gestational tissues. Placental CRH has recently been shown to
directly and preferentially stimulate DHEA-S secretion by human fetal adrenal
cortical cells [42]. Placental CRH also exerts direct actions on the uterus and
cervix to augment changes produced by estrogen on these tissues by interacting
with both prostaglandins and oxytocin, the two major uterotonins that stimulate
and maintain myometrial contractility at term and during labor [43, 44].
The overwhelming evidence from clinical studies of CRH and parturition
conducted by us and others suggests that women in preterm labor have signifi-
cantly elevated levels of CRH compared to gestational age-matched controls,
and that these elevations of CRH, assessed in some studies as early as 15 weeks
of gestation, precede the onset of preterm labor [2224, 4553]. Studies that
conducted serial assessments of CRH over the course of gestation found that
compared to term deliveries, women delivering preterm had not only signifi-
cantly elevated CRH levels but also a significantly accelerated rate of CRH
increase over the course of their gestation [47, 50, 54]. Moreover, we have
shown that the effects of placental CRH on spontaneous preterm birth are inde-
pendent from those of other biomedical risk factors [22, 23].
Placental Corticotropin-Releasing Hormone and
Fetal Growth
Placental CRH is believed to also regulate fetal growth via its effects on
placental perfusion and fetal cortisol production. Placental CRH elevations are
associated with decreased uteroplacental flow and hypoxemia known risk
factors for fetal growth restriction [55]. Fetal cortisol plays a critical role in
organ growth and maturation [56], and placental CRH may also participate in
this process via its positive feedback loop with fetal cortisol [37, 38, 41].
Several clinical studies have found that CRH levels in maternal and/or cord
blood during gestation or at the time of delivery are significantly higher in low
birth weight/small-for-gestational-age births [23, 5759].
Immune-Inflammatory Processes in Pregnancy
Microbial infection and inflammation in the gestational tissues has
emerged as one of the major risk factor for adverse birth outcomes such as
early preterm birth (30 weeks of gestation) and adverse neurodevelopmental
outcomes such as white matter brain damage and cerebral palsy [60]. These
adverse outcomes in the setting of infection are believed to result from the
actions of proinflammatory cytokines secreted as part of the maternal and/or
fetal host response to microbial invasion [61, 62]. Proinflammatory cytokines
have been shown to promote spontaneous labor and rupture of membranes via
their actions in the gestational tissues to stimulate the synthesis and release of
prostaglandins and metalloproteases, in the fetus to stimulate the production of
inflammatory cytokines, cortisol and DHEA-S, and in the placenta to stimulate
CRH synthesis and release (fig. 4) [6264]. Endocrine and immune processes
extensively cross-regulate one another in pregnancy. For example, the proin-
flammatory cytokine IL-1 stimulates the production of placental CRH, and
CRH in turn regulates cytokine production by immune cells. Because maternal
stress is associated with preterm birth, abnormalities in the regulation of CRH
and the production of proinflammatory cytokines may be a mechanism that
could form the pathophysiological basis for this association [61].
Although maternal stress and infection have each been implicated as
risk factors in preterm birth and the effects of stress on immune function are
well established, very little research to date has examined the nature of the
stress-infection-immune relationship in human pregnancy. Our review of the
relevant literature found only two studies linking maternal stress with immune
processes in human pregnancy [65, 66], and one in vivo study reporting that
women in preterm labor with microbial invasion of the amniotic cavity had sig-
nificantly higher CRH levels than those in preterm labor without infection [51].
Maternal
local infection
Fetal
inflammatory response
Maternal
systemic infection
Cytokines
(IL-1, IL-6)
Placenta and
fetal membranes
Uterine contractility
CRH Prostaglandins
Fig. 4. Endocrine-immune interactions in human pregnancy and fetal development.
Wadhwa 60
In collaboration with colleagues in Philadelphia, we have recently published a
report that supports a role for maternal psychosocial stress in the development
of reproductive tract infection in human pregnancy [67]. Moreover, our prelim-
inary findings to date also support the notion that stress-related endocrine
processes modulate immune function in the context of reproductive tract infec-
tion and inflammation [68, 69].
Fetal Neurodevelopment
The developing human central nervous system may be more vulnerable to
environmental perturbations than any other system because it develops over a
much longer period of time (1112 years); it has limited repair capabilities; its
units have highly specific functional roles; the blood-brain barrier is not fully
developed in utero, and the sensitivity of neurotransmitter systems, which is set
during critical developmental periods, affects the organisms response to all
subsequent experience [70]. However, the influence of the maternal and
intrauterine environment on the developing human fetal brain is poorly under-
stood, in part, because the assessment and quantification of human fetal brain
development presents many theoretical and methodological challenges [71]. To
date, we have performed four studies in an effort to quantify and examine the
influence of biobehavioral processes on fetal brain development.
The first study was performed on a sample of 84 fetuses at 3132 weeks of
gestation to examine the ability of the human fetus to learn and recall informa-
tion. Three series of vibroacoustic stimuli were presented at pseudorandom
intervals over the fetal head, and fetal heart rate (FHR) responses to the first
series of 15 stimuli (S1) were compared to responses to an identical second
series of 15 stimuli (S1) separated from the first set by the administration of a
single novel stimulus of different intensity and frequency (S2). A significant
habituation pattern of responses was observed across trials for both series of
stimuli, but this habituation pattern was attenuated for the series following the
novel stimulus. These findings suggest the 32-week-old human fetus may be
capable of detecting, habituating, and dishabituating to an external stimulus,
and support the premise that areas of the human fetal central nervous system
critical for some aspects of learning and memory have developed by the early
third trimester [72].
In a subsample of 33 mother-fetus pairs from the above study, the relation-
ship was examined between maternal (placental) levels of CRH and the above-
described fetal pattern of habituation and dishabituation in response to external
stimulation. Results indicated the fetuses of mothers with highly elevated CRH
levels did not respond significantly to the presence of the novel stimulus,
thereby providing preliminary support for the notion that abnormally elevated
levels of placental CRH may play a role in impaired neurodevelopment, as
assessed by the degree of dishabituation [73].
In the third study, the association of circulating maternal ACTH and BE
levels with measures of fetal responses to challenge was determined in a sample
of 132 women at 3132 weeks of gestation. Fetal responses were measured by
measuring heart rate habituation to a series of repeated vibroacoustic stimuli.
Individual differences in habituation were determined by computing the number
of consecutive responses above the standard deviation during a control (nonstim-
ulated) period. There was no significant relation between levels of ACTH, BE
and fetal responses to challenge. However, an index of POMC dysregulation
the degree of uncoupling between ACTH and BE was significantly related to
fetal responses, such that fetal exposure to relatively high levels of the maternal
opiate, BE, relative to ACTH, was associated with a significantly lower rate of
habituation [74].
In the fourth study, we performed nonlinear statistical analyses on our
complete sample of 156 mother-fetus pairs studied at 3133 weeks of gestation.
These analyses of FHR arousal and reactivity data, using a nonlinear repeated-
measures model with autocorrelated errors within subjects and independence
across subjects, suggest a host of maternal processes, including factors related
to prenatal stress, elevated levels of placental hormones, and the presence of
obstetric risk conditions, exert significant influences on the fetus and predict
individual differences in patterns of fetal responses to external challenges. Our
results specifically indicate the following: fetuses exhibited a significant,
nonlinear FHR increase in response to the vibroacoustic stimulation protocol;
baseline FHR, the presence of uterine contractions during trials, and character-
istics of the challenge protocol such as intertrial interval significantly
influenced the magnitude of FHR responses; after accounting for the effects
of baseline FHR, uterine contractions and characteristics of the challenge
protocol, maternal conditions related to psychological and physiological stress
(i.e., psychosocial stress levels, placental CRH concentrations, umbilical blood
flow, and the presence of maternal medical risk conditions) were significantly
associated with the pattern of FHR responses; after an initial response period,
fetuses exhibited an FHR response decrement to subsequent stimuli, indicating
habituation; a two-parameter growth curve (power) model to assess habituation
rate accounted for approximately 70% of the variance in FHR response, and
fetal sex and conditions related to maternal stress (i.e., maternal ACTH concen-
trations, the presence of medical risk conditions) were significantly associated
with the rate of habituation [75]. Thus, this set of findings provides further
support for the role played by the prenatal environment in modulating aspects
Wadhwa 62
of human fetal brain development that underlie processes related to recognition,
appraisal, response, memory and habituation.
Stress and Placental Corticotropin-Releasing
Hormone Function
Placental CRH is stress-sensitive. As mentioned earlier, a series of in vitro
studies by Petraglia et al. [25, 76, 77] have shown that CRH is released from
cultured human placental cells in a dose-response manner in response to all the
major biological effectors of stress, including cortisol, and catecholamines.
In vivo studies by our group [78] and other investigators [7981] have found
significant correlations among maternal pituitary-adrenal stress hormones
(ACTH, cortisol) and placental CRH levels. Moreover, we and others have
reported that maternal psychosocial stress is significantly correlated with maternal
pituitary-adrenal hormone levels (ACTH, cortisol) [82] both of which are
known to stimulate placental CRH secretion. Some [46, 47], but not all studies
[83], have also reported direct associations between maternal psychosocial
processes and placental CRH function. Thus, depending on the chronicity of
the stressor, the resultant increase in CRH production may be a critical factor
that contributes to the early initiation of spontaneous labor and impaired fetal
growth [84].
The fetus may also be directly sensitive to maternal stress [85, 86]. Romero
et al. [62] have recently described a condition, fetal inflammatory response
syndrome, characterized by a multisystem fetal stress response in human preg-
nancy, with activation of endocrine and immune systems, including elevated
fetal cortisol/DHEA-S ratio [87] and elevated levels of inflammatory cytokines
in fetal circulation [87], all of which are important biochemical mediators of
fetal development and spontaneous preterm birth.
Conclusions, Issues and Future Directions
Adverse fetal developmental outcomes and their sequela are recognized as
significant health problems in the United States. Women reporting high levels of
pre- and perinatal stress are, on average, twice as likely to experience an adverse
outcome as women reporting low levels of stress. Although the magnitude of this
effect of prenatal stress is comparable to that of other established obstetric risk
factors, the specificity and sensitivity of these measures as predictors of adverse
outcome(s) in any individual pregnancy is, at best, only modest. These self-report
measures of psychosocial stress rely exclusively on retrospective recall, and may
be subject to numerous, systematic biases that undermine measurement validity.
Moreover, these measures do not capture several important dimensions that are
known to moderate the stress and adverse health outcome relation, such as
individual differences in psychological or biological responsivity to potentially
stressful circumstances in subjects everyday lives, and individual differences in
the context specificity of stressful responses. Recent advances in momentary
experience sampling methodology now afford the opportunity of not only mini-
mizing biases associated with retrospective recall measures but also of assessing
the dynamic interplay of psychological, behavioral and biological processes in
natural, everyday settings. We suggest that these new methods hold great promise
in addressing many of the shortcomings in the stress and fetal development
literature, and recommend importing and adapting these methods to conduct
ambulatory studies of psychological, biological and behavioral processes in
human pregnancy.
Stress-related physiological parameters such as placental CRH and proin-
flammatory cytokines have been shown to significantly predict the risk of
adverse fetal developmental outcomes. However, studies have examined the
role of these parameters separately and have uniformly reported low specificity
and sensitivity. For example, low levels of placental CRH in pregnancy are a
good negative predictor of preterm birth but high levels are a poor positive pre-
dictor. Similarly, absence of intrauterine infection is a good negative predictor
of early preterm birth, but presence of intrauterine infection/inflammation is a
poor positive predictor. This may suggest that parameters such as placental
CRH and infection/inflammation are, in and of themselves, necessary but not
sufficient causes of adverse outcomes. Rather than proposing other novel
physiological parameters, we suggest that these major parameters need to be
examined simultaneously to determine the manner in which they interact to pre-
dict the risk of adverse developmental outcomes. We are not aware of any
study that, for instance, has looked systematically at both endocrine and
immune/inflammatory processes in human pregnancy. We suggest this is a crit-
ical future direction for this work because it is well known that endocrine and
immune processes extensively regulate and counterregulate one another, and
that the effect of either of these processes on a biological outcome of interest is
modulated by the state/context of the other.
We and others have uncovered evidence of stress-related dysregulation in
adverse fetal developmental outcomes in early gestation [47, 50]. Moreover,
measures of stress and stress-related physiological dysregulation in early gesta-
tion are better predictors of adverse outcomes than the same measures assessed
later in gestation [33, 54]. This brings up the important question of the possibil-
ity of an underlying susceptibility to stress and stress-related physiological
dysregulation that may even precede the index pregnancy. We are not aware of
Wadhwa 64
any studies that have examined stress and stress biology processes in women
before they became pregnant to track the physiological and psychosocial transi-
tions from the nonpregnant to the pregnant state, and we suggest this is an
important direction in order to better understand individual vulnerabilities to
the adverse effects of prenatal and perinatal stress.
Finally, to return to the concept of an epigenetic framework of development,
it appears that fetal developmental processes ultimately represent the dynamic
interplay between two sets of information systems (i.e., fetal and maternal
DNA) and two sets of cellular machinery (i.e., the fetal and maternal environ-
ments). It is crucial to identify the genetic and environmental determinants of
fetal development, and to model the gene, environment, and maternal-fetal
interactions that may underlie the risk of pathophysiological outcomes. In this
context, the genetic architecture of fetal development and parturition is defined
as the number of loci, their genomic positions, the number of functional alleles
per locus, and the patterns of dominance, epistasis, pleiotropy, and gene-
environment interactions that characterize the transition from genotype to
phenotype.
Genomic imprinting is an epigenetic mechanism by which certain genes
become repressed on one of the two parental alleles. Imprinting is now known
to play important roles in many other aspects of mammalian development, and
its deregulation may result in disease. Recent evidence supports a role for stress
in dysregulating the imprinting process during development. Studies in the
mouse, for example, demonstrate that environmental stress, such as in vitro cul-
ture, affect the somatic maintenance of epigenetic marks at imprinted loci [88].
Other studies have shown that bovine in vitro produced and nuclear transfer-
derived embryos differ from their in vivo produced counterparts in a number of
characteristics, including a complete lack of expression, an induced expression,
or a significant up- or downregulation of a specific gene [89]. These alterations
are considered a kind of stress response of the embryos to deficient environ-
mental conditions, are believed to be caused primarily by changes in the methy-
lation patterns, and are associated with aberrant growth and morphology at fetal
and perinatal stages of development. Nonimprinted genes can also undergo epi-
genetic change in response to the environment. For example, a recent study in
rodents reported that the choice of exon usage in the glucocorticoid receptor
gene is altered in both prenatal glucocorticoid exposure and neonatal behavioral
manipulation via histone acetylation and DNA methylation in a transcriptional
factor binding site, and that these changes persist throughout life as manifested
in altered HPA activity [90]. We are not aware of any studies to date that have
systematically examined the physiological genomics of maternal and fetal
stress-related neuroendocrine systems and pathways in human pregnancy, and
suggest this is yet another important future avenue for this line of research.
Some 60 years ago, the Fels study of early development, probably the first
systematic investigation of factors that affect development before birth, sug-
gested that such factors as.... [maternal] emotional life and activity level during
gestation may contribute to the shaping of physical status, behavioral patterns,
and postnatal progress of the children they bear [91]. Clearly, although we have
come a long way since then, the study of the interface between biology and
behavior in prenatal life continues now more than ever before to hold great
promise in realizing the full implications of this statement to shed light on the
nature of our origins and their consequences for our health and well-being.
Acknowledgement
This study was supported, in part, by US PHS (NIH) grants HD-33506, HD-41696 and
HD-28413 to P.D.W.
References
1 Wadhwa PD, Glynn L, Hobel CJ, Garite TJ, Porto M, Chicz-DeMet A, Wiglesworth A, Sandman
CA: Behavioral perinatology: Biobehavioral processes in human fetal development. Regul Pept
2002;108:149157.
2 Arenander AT, de Vellis J: Development of the nervous system; in Siegel GJ, Agranoff BW, Albers
RW, Molinoff PB (eds): Basic Neurochemistry: Molecular, Cellular and Medical Aspects. New
York, Raven Press, 1989.
3 Ackerman S: Discovering the Brain. Washington, National Academy Press, 1992.
4 Smotherman WP, Robinson SR: Tracing developmental trajectories into the prenatal period; in
Lecanuet JP, Fifer WP, Krasnegor NA, Smotherman WP (eds): Fetal Development: A Psycho-
biological Perspective. Hillsdale, Erlbaum, 1995.
5 Kolb B: Brain Plasticity and Behavior. Mahwah, Erlbaum, 1995.
6 Bornstein MH: Sensitive periods in development: Structural characteristics and causal interpreta-
tions. Psychol Bull 1989;105:179197.
7 Gluckman PD, Hanson MA: Living with the past: Evolution, development and patterns of disease.
Science 2004;305:17331736.
8 Wadhwa PD, Culhane JF, Rauh VA, Barve SS: Stress and preterm birth: Neuroendrocrine
immune-inflammatory and vascular mechanisms. Matern Child Health J 2001;5:119125.
9 Wadhwa PD, Culhane JF, Rauh VA, Barve SS, Hogan VK, Sandman CA, Hobel CJ, Dunkel-
Schetter C, Garite TJ, Glynn L: Stress, infection and preterm birth: A biobehavioral perspective.
Paediatr Perinat Epidemiol 2001;15:1729.
10 Wadhwa PD: Prenatal stress and life-span development; in Friedman HS (ed): Encyclopedia of
Mental Health. San Diego, Academic Press, 1998, vol 3, pp 265280.
11 Chrousos GP: Stressors, stress, and neuroendocrine integration of the adaptive response. The 1997
Hans Selye Memorial Lecture. Ann NY Acad Sci 1998;851:311335.
12 Barker DJP: Mothers, Babies and Health in Later Life, ed 2. Edinburgh, Churchill Livingston,
1998.
13 Creasy RK: Preterm labor and delivery; in Creasy RK, Rosnik R (eds): Maternal Fetal Medicine:
Principles and Practice. Philadelphia, Saunders, 1994.
14 Nathanielsz PW: Life in the Womb: The Origin of Health and Disease. Ithica, Promethean Press,
1999.
Wadhwa 66
15 Chrousos GP, Gold PW: The concepts and stress and stress systems disorders. JAMA 1992;267:
12441252.
16 Yen SC: Endocrinology of pregnancy; in Creasy RK, Resnick R (eds): Maternal-Fetal Medicine:
Principles and Practice. Philadelphia, Saunders, 1994.
17 Chrousos GP: Regulation and dysregulation of the hypothalamic-pituitary-adrenal axis. The
corticotropin-releasing hormone perspective. Endocrinol Metab Clin North Am 1992;21:833858.
18 Vale W, Spiess J, Rivier C, Rivier J: Characterization of a 41-residue ovine hypothalamic peptide
that stimulates secretion of corticotropin and beta-endorphin. Science 1981;213:13941397.
19 Petraglia F, Florio P, Nappi C, Genazzani AR: Peptide signaling in human placenta and membranes:
Autocrine, paracrine, and endocrine mechanisms. Endocr Rev 1996;17:156186.
20 Kalantaridou SN, Makrigiannakis A, Zoumakis E, Chrousos GP: Stress and the female reproduc-
tive system. J Reprod Immunol 2004;61:6168.
21 Challis JRG, Matthew SG, Gibb W, Lye SJ: Endocrine and paracrine regulation of birth at term,
and preterm. Endocr Rev 2000;21:514550.
22 Wadhwa PD, Porto M, Chicz-DeMet A, Sandman CA: Maternal CRH levels in early third trimester
predict length of gestation in human pregnancy. Am J Obstet Gynecol 1998;179:10791085.
23 Wadhwa PD, Garite TJ, Porto M, Glynn L, Chicz-DeMet A, Dunkel-Schetter C, Sandman CA:
Placental corticotropin-releasing hormone (CRH), spontaneous preterm birth and fetal growth
restriction: A prospective investigation. Am J Obstet Gynecol 2004;191:10631069.
24 Korebrits C, Ramirez MM, Watson L, Brinkman E, Bocking AD, Challis JR: Maternal CRH is
increased with impending preterm birth. J Clin Endocrinol Metab 1998;83:15851591.
25 Petraglia F, Sutton S, Vale W: Neurotransmitters and peptides modulate the release of immunore-
active corticotropin-releasing factor from cultured human placental cells. Am J Obstet Gynecol
1989;160:247251.
26 Coe CL, Lubach GR, Karaszewski JW, Ershler WB: Prenatal endocrine activation alters postnatal
cellular immunity in infant monkeys. Brain Behav Immun 1996;10:221234.
27 Coe CL, Lubach GR: Prenatal influences on neuroimmune set points in infancy. Ann NY Acad Sci
2000;917:468477.
28 Schneider ML, Moore CF, Roberts AD, Dejesus O: Prenatal stress alters early neurobehavior,
stress reactivity and learning in non-human primates: A brief review. Stress 2001;4:183193.
29 Weinstock M: Alterations induced by gestational stress in brain morphology and behaviour of the
offspring. Prog Neurobiol 2001;63:427452.
30 Wadhwa PD, Sandman CA, Porto M, Dunkel-Schetter C, Garite TJ: The association between
prenatal stress and infant birth weight and gestational age at birth: A prospective study. Am J Obstet
Gynecol 1993;169:858865.
31 Rini C, Dunkel-Schetter C, Wadhwa PD, Sandman CA: Psychological adaptation and birth
outcomes: The role of personal resources, stress and sociocultural context during pregancy. Health
Psychol 1999;18:333345.
32 Feldman PJ, DunkelSchetter C, Sandman CA, Wadhwa PD: Maternal social support predicts birth
weight and fetal growth in human pregnancy. Psychosom Med 2000;62:715725.
33 Glynn L, Wadhwa PD, Dunkel-Schetter C, Sandman CA: When stress happens matters: Effects of
earthquake timing on stress responsivity in pregnancy. Am J Obstet Gynecol 2001;184:637642.
34 Gorin AA, Stone AA: Recall biases and cognitive errors in retrospective self-reports: A call for
momentary assessments; in Baum A, Revenson T, Singer J (eds): Handbook of Health Psychology.
Mahwah, Erlbaum, 2001, pp 405413.
35 Shiffman S: Real-time self-report of momentary states in the natural environment: Computerized
ecological momentary assessment; in Stone AA, Turkkan JJ, Bachrach CA, et al (eds): The
Science of Self-Report: Implications for Research and Practice. Mahwah, Erlbaum, 2000.
36 Grammatopoulos DK, Hillhouse EW: Role of corticotropin-releasing hormone in onset of labour.
Lancet 1999;354:15461549.
37 Majzoub JA, Karalis KP: Placental corticotropin-releasing hormone: Function and regulation. Am
J Obstet Gynecol 1999;180:S242S246.
38 Smith R(ed): The Endocrinology of Parturition. Front Horm Res. Basel, Karger, 2001, vol 27.
39 Jaffe RB: Role of human fetal adrenal gland in the initiation of parturition; in Smith R (ed): The
Endocrinology of Parturition. Front Horm Res. Basel, Karger, 2001, pp 7585.
40 Mesiano S: Roles of estrogen and progesterone in human parturiton; in Smith R (ed): The
Endocrinology of Parturition. Front Horm Res. Basel, Karger, 2001, vol 27, 86104.
41 Smith R: The timing of birth. Sci Am 1999;3:6875.
42 Smith R, Mesiano S, Chan EC, Brown S, Jaffe RB: Corticotropin-releasing hormone directly and
preferentially stimulates dehydroepiandrosterone sulfate secretion in fetal adrenal cortical cells.
J Clin Endocrinol Metab 1998;83:29162920.
43 Jones SA, Brooks AN, Challis JR: Steroids modulate corticotropin-releasing hormone production
in human fetal membranes and placenta. J Clin Endocrinol Metab 1989;68:825830.
44 Jones SA, Challis JR: Steroid, corticotrophin-releasing hormone, ACTH and prostaglandin inter-
actions in the amnion and placenta of early pregnancy in man. J Endocrinol 1990;125:153159.
45 Campbell EA, Linton EA, Wolfe CD, Scraggs PR, Jones MT, Lowry PJ: Plasma corticotropin-
releasing hormone concentrations during pregnancy and parturition. J Clin Endocrinol Metabol
1987;64:10541059.
46 Erickson K, Thorsen P, Chrousos G, Grigoriadis DE, Khongsaly O, McGregor J, Schulkin J:
Preterm birth: Associated neuroendocrine, medical, and behavioral risk factors. J Clin Endocrinol
Metab 2001;86:25442552.
47 Hobel CJ, Dunkel-Schetter C, Roesch SC, Castro LC, Arora CP: Maternal plasma corticotropin-
releasing hormone associated with stress at 20 weeks gestation in pregnancies ending in preterm
delivery. Am J Obstet Gynecol 1999;180:S257S263.
48 Holzman C, Jetton J, Siler-Khodr T, Fisher R, Rip T: Second trimester corticotropin-releasing
hormone levels in relation to preterm delivery and ethnicity. Obstet Gynecol 2001;97:657663.
49 Kurki T, Laatikainen T, Salminen-Lappalainen K, Ylikorkala O: Maternal plasma corticotrophin-
releasing hormone Elevated in preterm labour but unaffected by indomethacin or nylidrin. Br J
Obstet Gynaecol 1991;98:685691.
50 McLean M, Bisits A, Davies J, Woods R, Lowry P, Smith R: A placental clock controlling the
length of human pregnancy. Nat Med 1995;1:460463.
51 Petraglia F, Aguzzoli L, Florio P, Baumann P, Genazzani AD, DiCarlo C, Romero R: Maternal
plasma and placental immunoreactive corticotrophin-releasing factor concentrations in infection-
associated term and pre-term delivery. Placenta 1995;16:157164.
52 Warren WB, Patrick SL, Goland RS: Elevated maternal and plasma corticotropin-releasing
hormone levels in pregnancies complicated by preterm labor. Am J Obstet Gynecol 1992;166:
11981207.
53 Wolfe CDA, Patel SP, Linton EA, Campbell EA, Anderson J, Dornhorst A, Lowry PJ, Jones MT:
Plasma corticotrophin-releasing factor (CRF) in abnormal pregnancy. Br J Obstet Gynaecol
1988;95:10031006.
54 McGarth S, McLean M, Smith D, Bisito A, Giles W, Smith R: Maternal plasma corticotropin-
releasing hormone trajectories vary depending on the etiology of preterm birth. Am J Obstet
Gynecol 2002;186:257260.
55 Giles WB, McLean M, Davies JJ, Smith R: Abnormal umbilical artery Doppler waveforms and
cord blood corticotropin-releasing hormone. Obstet Gynecol 1996;87:107111.
56 Fencl MD, Stillman RJ, Cohen J, Tulchinsky D: Direct evidence of sudden rise in fetal corticoids
late in human gestation. Nature 1980;287:225226.
57 Goland RS, Jozak S, Warren WB, Conwell IM, Stark RI, Tropper PJ: Elevated levels of umbilical
cord plasma corticotropin-releasing hormone in growth-retarded fetuses. J Clin Endocrinol Metab
1993;77:11741179.
58 Ruth V, Hallman M, Laatikainen T: Corticotropin-releasing hormone and cortisol in cord plasma
in relation to gestational age, labor, and fetal distress. Am J Perinatol 1993;10:115118.
59 Tropper PJ, Warren WB, Jozak SM, Conwell IM, Stark RI, Goland RS: Corticotropin releasing
hormone concentrations in umbilical cord blood of preterm fetuses. J Dev Physiol 1992;18:8185.
60 Goldenberg RL, Hauth JC, Andrews WW: Mechanisms of disease: Intrauterine infection and
preterm delivery. N Engl J Med 2000;342:15001507.
61 Dudley DJ: Immunoendocrinology of preterm labor: The link between corticotropin-releasing
hormone and inflammation. Am J Obstet Gynecol 1999;180:s251s256.
62 Romero R, Gomez R, Chaiworapongsa T, Conoscenti G, Kim JC, Kim YM: The role of infection
in preterm labour and delivery. Paediatr Perinat Epidemiol 2001;15:4156.
Wadhwa 68
63 Falkenberg ER, Davis RO, DuBard M, Parker CR: Effects of maternal infections on fetal adrenal
steroid production. Endocr Res 1999;25:239249.
64 Gomez R, Romero R, Edwin SS, David C: Pathogenesis of preterm labor and preterm premature
rupture of membranes associated with intraamniotic infection. Infect Dis Clin North Am 1997;11:
135176.
65 Arck PC, Rose M, Hertwig K, Hagen E, Hildebrandt M, Klapp BF: Stress and immune mediators
in miscarriage. Hum Reprod 2001;16:15051511.
66 Herrera JA, Alvarado JP, Matrinez JE: The psychosocial environment and cellular immunity in the
pregnant patient. Stress Med 1998;4:4956.
67 Culhane JF, Rauh V, Farley-McCollum K, Hogan V, Agnew K, Wadhwa PD: Maternal stress is asso-
ciated with bacterial vaginosis in human pregnancy. Matern Child Health J 2001;5:127134.
68 Culhane J, Rauh V, Hogan V, McCollum K, Wadhwa PD: Racial/ethnic differences in bacterial
vaginosis in pregnancy: The role of individual and contextual-level maternal stress (abstract). Am
J Obstet Gynecol 2001;185:S145.
69 Wadhwa PD, Culhane J, Chicz-DeMet A, Barve SS, Rauh V, Farley-McCollum K, Hogan V:
Maternal stress, endocrine/immune-inflammatory processes and vaginal infection in human preg-
nancy: Preliminary findings. Am J Obstet Gynecol 2001;185:S231.
70 Rodier PM, Cohen IR, Buelke-Sam J: Developmental neurotoxicology: Neuroendocrine manifes-
tations of CNS insult; in Kimmel CA, Buelke-Sam J (eds): Developmental Toxicology. New York,
Raven Press, 1994, pp 6592.
71 DiPietro JA, Hodgson DM, Costigan KA, Hilton SC, Johnson TR: Fetal neurobehavioral develop-
ment. Child Dev 1996;67:25532567.
72 Sandman CA, Wadhwa PD, Hetrick W, Porto M, Peeke HS: Human fetal heart rate dishabituation
at 32 weeks gestation. Child Dev 1997;68:10311040.
73 Sandman CA, Wadhwa PD, Chicz-DeMet A, Garite TJ, Porto M: Maternal corticotropin-releasing
hormone (CRH) and fetal habituation in human pregnancy. Dev Psychobiol 1999;34:163173.
74 Sandman CA, Glynn L, Wadhwa PD, Chicz-DeMet A, Porto M, Garite TJ: Maternal HPA dysreg-
ulation influences fetal behavior in human pregnancy. Dev Neurosci 2003;25:4149.
75 Wadhwa PD, Truszczynska H, Garite TJ, Porto M, Sandman CA: Maternal environment influences
evoked fetal heart rate responses in human pregnancy. Ann Behav Med 1999;21:S104.
76 Petraglia F, Sawchenko PE, Rivier J, Vale W: Evidence for local stimulation of ACTH secretion by
corticotropin-releasing factor in human placenta. Nature 1987;328:717719.
77 Petraglia F, Volpe A, Genazzani A, Riveier J, Sawchenko P, Vale W: Neuroendrocinology of the
human placenta. Front Neuroendrocinol 1990;11:637.
78 Wadhwa PD, Sandman CA, Chicz-DeMet A, Porto M: Placental CRH modulates maternal
pituitary adrenal function in human pregnancy. Ann NY Acad Sci 1997;814:276281.
79 Chan EC, Smith R, Lewin T, Brinsmead MW, Zhang HP, Cubis J, Thornton K, Hurt D: Plasma
corticotropin-releasing hormone, -endorphin and cortisol inter-relationships during human
pregnancy. Acta Endocrinol 1993;128:339344.
80 Goland RS, Conwell IM, Warren WB, Wardlaw SL: Placental corticotropin-releasing hormone
and pituitary-adrenal function during pregnancy. Neuroendocrinology 1992;56:742749.
81 Sasaki A, Shinkawa O, Yoshinaga K: Placental corticotropin-releasing hormone may be a stimula-
tor of maternal pituitary adrenocorticotropic hormone secretion in humans. J Clin Invest 1989;84:
19972001.
82 Wadhwa PD, Dunkel-Schetter C, Chicz-DeMet A, Porto M, Sandman CA: Prenatal psychosocial
factors and the neuroendocrine axis in human pregnancy. Psychosom Med 1996;58:432446.
83 Petraglia F, Hatch MC, Lapinski R, Stomati M, Reis FM, Cobellis L, Berkowitz GS: Lack of effect
of psychosocial stress on maternal corticotropin-releasing factor and catecholamine levels at 28
weeks gestation. J Soc Gynecol Investig 2001;8:8388.
84 Lockwood CJ, Kuczynski E: Markers of risk for preterm delivery. J Perinat Med 1999;27:520.
85 Gitau R, Cameron A, Fisk NM, Glover V: Fetal exposure to maternal cortisol. Lancet 1998;352:
707708.
86 Gitau R, Fisk NM, Teixeira JM, Cameron A, Glover V: Fetal hypothalamic-pituitary-adrenal stress
responses to invasive procedures are independent of maternal responses. J Clin Endocrinol Metab
2001;86:104109.
87 Yoon BH, Romero R, Park JS, Chang JW, Kim AH, Kim JC, Kim JS: Microbial invasion of the
amniotic cavity with Ureaplasma urealyticum is associated with a robust host response in fetal,
amniotic, and maternal compartments. Am J Obstet Gynecol 1998;179:12541260.
88 Thompson SL, Konfortova G, Gregory RI, Reik W, Dean W, Feil R: Environmental effects on
genomic imprinting in mammals. Toxicol Lett 2001;120:143150.
89 Niemann H, Wrenzycki C, Lucas-Hahn A, Brambrink T, Kues WA, Carnwath JW: Gene expression
patterns in bovine in vitro-produced and nuclear transfer-derived embryos and their implications
for early development. Cloning Stem Cells 2002;4:2938.
90 Weaver ICG, Cervoni N, Champagne FA, DAlessio AC, Sharma S, Seckl JR, Dymov S, Szyf M,
Meaney MJ: Epigenetic programming by maternal behavior. Nat Neurosci 2004;7:847854.
91 Sontag LW: The significance of fetal environmental differences. Am J Obstet Gynecol 1941;124:
9961003.
Pathik D. Wadhwa, MD, PhD
Behavioral Perinatology Research Program, University of California, Irvine
3117 Gillespie Neuroscience Building
Irvine, CA 926974260 (USA)
Tel. 1 949 824 8238, Fax 1 949 824 8218, E-Mail pwadhwa@uci.edu
Maternal Depression: An Adverse
Early Environment
Aquila J. Beach
a
, Autumn L. Henry
a
, Zachary N. Stowe
a,b
,
D. Jeffrey Newport
a
Departments of
a
Psychiatry and Behavioral Sciences, and
b
Gynecology and
Obstetrics, Emory University School of Medicine, Atlanta, Ga., USA
Innate differences in the vulnerability to major depressive disorder (MDD)
and other stress-related psychiatric illnesses have long been recognized. Some
individuals tolerate stress of great magnitude without succumbing, whereas
others exhibit a constitutional vulnerability to the deleterious effects of stress,
i.e., a lower threshold of tolerance predisposing them to stress-induced illness.
This inherent vulnerability, or diathesis, provides the basis for the diathesis/stress
model of disease that has been applied to an extensive array of psychiatric and
psychosomatic disorders. The origin of the diathesis, its components, and the
contribution to individual vulnerability remain central foci of investigation. The
relative contribution of genetic inheritance and environmental exposure has
been deliberated in often contentious debates of nature vs. nurture. These argu-
ments are frequently couched in overly absolute terms, but diathesis/stress
models provide a more balanced consideration, recognizing that both genetic
and acquired factors likely interact to contribute to the vulnerability to stress-
related illness (fig. 1).
A burgeoning line of research has moved beyond the study of psychiatric
syndromes to investigate individual vulnerability to the development of these
disorders. This perspective acknowledges that both psychosocial and neurobio-
logical factors mediate the diathesis. Furthermore, it works from the assump-
tion that these factors can be characterized and potentially quantified with
respect to the biobehavioral determinants of vulnerability engendered by
genetic heritability and exposure to early adverse life events. Much of the extant
early life stress research has focused on the persistent sequelae of traumatic
childhood experiences such as child abuse and neglect. However, increasing
Maternal Depression: An Adverse Early Environment 71
evidence indicates that parental depression, particularly maternal depression,
may have untoward effects on child development.
Freuds Oedipal conflict [1], Kleins notion of infant splitting of the mother
[2, 3], and Bowlbys concepts regarding attachment [4] underscore the long-
standing consensus within modern psychiatry that mother/child interaction is
critical to normal human psychological development. It should not be surpris-
ing, therefore, that when conditions such as maternal depression impact the
mother/infant dyad, child development suffers deleterious effects. Some of this
risk is undoubtedly transmitted in part by genetic means, though other potential
pathways of transmission include: (1) effects of exposure to the neurobiological
aberrations associated with antenatal maternal depression, (2) effects of expo-
sure to the affective, behavioral, and cognitive alterations of the depressed
parent, and (3) effects of exposure to familial dysfunction that can occur when
the childs mother is depressed [5]. Because its effects can even be communi-
cated prenatally, maternal depression, within the context of the diathesis/stress
model, must be recognized as a childs earliest adverse life event.
This review assesses both the clinical and preclinical evidence that exposure
to maternal depression and stress constitutes an aberrant change in the environ-
ment which comprises an adverse life event for the developing child. The discus-
sion focuses first on the prevalence of maternal depression during the peripartum
period, and then on the clinical data regarding the effects of maternal depression
on child well-being. After reviewing the evidence from preclinical research of
the harmful effects of maternal stress, the discussion concludes by addressing
the clinical implications for depressed women who have children.
Illness
Vulnerable
phenotype
Heredity
Early life
stress
Later stress
Fig. 1. Diathesis/stress model of illness.
Beach/Henry/Stowe/Newport 72
Prevalence of Maternal Depression
It has been well established across a disparate array of ethnographic groups
that the lifetime rate of MDD among women doubles that of men. However, the
heightened vulnerability to MDD is not evenly distributed across the female
reproductive life cycle but clusters within the childbearing years [6], raising the
specter, of course, that depression is most likely to strike when a woman is rais-
ing children. Despite the concerns that should be raised by such findings, there
persists the popular notion that pregnancy is a time of emotional well-being.
Such clinical lore is not supported, however, by available research data, which
indicates that the rate of antenatal depression is comparable to nonpuerperal
depression [712]. Up to 70% of pregnant women endorse depressive symp-
toms, with 1016% fulfilling diagnostic criteria for MDD [1316]. In fact, over
10% of women presenting for evaluation of postpartum depression (PPD)
report symptom onset actually occurred during pregnancy [17]. Whereas sys-
tematic research regarding antenatal depression is a relatively new area of
investigation, PPD has been documented for millennia. PPD affects between 10
and 22% of adult women and up to 26% of adolescent mothers [18, 19].
Furthermore, there is a dramatic rise in psychiatric hospitalizations during the
first postpartum month [20], and up to 12.5% of all psychiatric admissions for
women occur during the first postpartum year [21].
Effects of Fetal Exposure to Antenatal Maternal
Depression
The effects of depression during pregnancy on fetal development can be
transmitted in any of 3 ways. First, the fetal sequelae of antenatal depression
can be directly mediated by certain neurobiological substrates of depression,
e.g. glucocorticoids and/or proinflammatory cytokines, that can cross the
placenta into the fetal circulation. Although the direct mediation of fetal effects
has not been incontrovertibly demonstrated, it is clear that many such substrates
readily cross the placenta. Second, the fetus can be indirectly affected via
neuroendocrine mechanisms whereby prenatal depression modulates the physi-
ological maintenance of pregnancy. For example, there is convincing evidence
that depression during pregnancy is associated with higher rates of preterm
delivery [2225]. Although the mechanism whereby antenatal depression might
lead to preterm delivery is indeterminate and arguably multifactorial, it has
been postulated that depression-associated hyperactivity of the pituitary-adrenal
axis might induce placental hypersecretion of corticotropin-releasing factor
[2629], which in turn potentiates myometrial contractility and ultimately
parturition [26, 30]. Finally, the effects of antenatal depression on fetal devel-
opment can be mediated indirectly by its impact on maternal behavior. Indeed,
depressed pregnant women are more likely to experience heightened stress,
decreased social support, poor weight gain, and are more likely to use tobacco,
drugs and alcohol than euthymic gravidas [31].
The pathways of fetal exposure in most cases remain obscure. However,
several studies have documented the association of depression during preg-
nancy with deleterious obstetrical outcomes. As previously mentioned, antena-
tal depression is associated with a greater risk for preterm delivery [2225].
Other data demonstrate that stress or depression during pregnancy may impede
fetal growth, resulting in a higher rate of small-for-gestational-age babies
[3236] and smaller head circumferences at delivery [37]. Maternal stress
during the embryonic phase of gestation has even been reported to increase the
risk for congenital anomalies such as cranial neural crest deformities [38].
Children exposed to maternal stress, anxiety, or depression during preg-
nancy may also suffer adverse neurodevelopmental sequelae that affect various
aspects of cognitive and psychosocial function. For example, newborns of
depressed mothers perform poorly on neurological examinations, exhibiting
poorer motor skills, activity, coordination, and resilience [39]. Additional evi-
dence indicates that the neurodevelopmental effects of antenatal depression
persist beyond infancy. Antenatal maternal stress is associated with higher rates
of developmental delay and a variety of emotional and behavioral problems
[40, 41] in preschool children. Furthermore, children of elementary school age
have also demonstrated cognitive delay, behavioral and emotional difficulties
[42, 44], in addition to maladaptive social interactions [42, 45].
Recent data have begun to delineate the neurobiological alterations in
children exposed to antenatal depression that conceivably underlie the obser-
ved neurodevelopmental effects. For example, stress-associated elevations in
maternal catecholamine and glucocorticoid secretion during late pregnancy
are predictive of similar elevations in neonates 24 h after delivery [46]. The
neuroendocrine effects of prenatal exposure to maternal depression may be per-
sistent, as 6-month-old children of euthymic mothers who had been depressed
during pregnancy exhibit heightened salivary cortisol responses to a standard-
ized psychosocial stressor [47]. Fetal exposure to prenatal depression is also
associated with increases in right frontal electroencephalograph (EEG) asym-
metry evident at 1 month postpartum [48]. This EEG pattern is consistent with
that of left hypofrontality commonly observed in depressed adults. Toddlers
with this pattern of EEG asymmetry more often display negative facial expres-
sions, hostility, aggression, reduced motor activity during observed play and
fewer displays of affectionate behavior towards their mother regardless of infant
temperament [49]. Preschool-aged children with right-sided EEG asymmetry
also exhibit less emphatic responses both to novel emotional triggers and their
mothers simulated distress [50].
Effects of Infant Exposure to Postnatal Maternal
Depression
The effect of maternal depression after delivery on child development has
been more extensively studied. The majority of these investigations have
focused on postdelivery events with sparse documentation of antenatal events.
Although the effects of postnatal depression can be theoretically mediated
directly via depression-related alterations in the constituents of the breast milk
of depressed lactating women [51], the children of depressed women are more
likely to be effected indirectly by changes in maternal behavior. A meta-analysis
of 46 observational studies of depressed mothers demonstrated a moderate
association of maternal depression with negative (i.e. hostile, coercive) parent-
ing behaviors and disengaged parenting behaviors [52]. Depressed mothers
have recently been reported to be less likely to put their children in car seats,
give vitamins to their children, talk with their children, or play with their
children [53, 54]. These mothers may also have greater difficulty completing
routine parenting tasks such as feeding [55] or bathing [56] their infants.
Numerous studies indicate that alterations in the parenting behaviors of
depressed women may in turn affect child development. For example, a meta-
analysis of the relationship between maternal mental health and infant attach-
ment, encompassing 35 studies and over 200 mother-infant pairs, indicated that
maternal stress and depression are associated with a greater prevalence of inse-
cure infant attachment [57]. Disordered attachment in the infants of depressed
women is typically manifested by less facial expression, less frequent head ori-
entation, less crying, and more fussiness [58]. As they age, the children of
depressed mothers more frequently display ineffective emotional regulation
when interacting with their mothers [59], and less prolonged and less efficient
attachment behaviors even when interacting with other nondepressed women
[58, 60]. The clinical consequences of impaired attachment in the infant of
depressed women include difficulties in eating and sleeping observed as early
as 3 months of age [56, 61], neurodevelopmental delays evident as early as
6 months of age [61, 62], and diminished verbal interaction and poorer perfor-
mance on object concept tasks at 18 months of age [54]. Aberrant attachment
behaviors have been proposed to result from a deranged interaction between
mother and child in which the prototypical withdrawn or intrusive maternal
styles of depressed mothers produce insufficient arousal and stimulation in
their infants [63].
Consistent with the implications of the diathesis/stress model, the neurobe-
havioral sequelae of exposure to postnatal maternal depression appear to persist
well beyond infancy. Children of depressed mothers are ultimately more likely
to exhibit emotional instability, suicidal behavior, conduct problems, and
require psychiatric treatment [39, 64]. They are also more often troubled during
the school years by low self-esteem [59], diminished social competence [45],
more fear and anxiety [63], delayed language and cognitive development [43, 65],
and a greater propensity to overt violence and aggression [60, 66]. Ultimately,
the children of depressed mothers are more likely to experience social, educa-
tional, and vocational difficulties [59, 67, 68] as well as psychiatric illness
[64, 67, 6971].
Less studied are the neurobiological underpinnings of childhood exposure
to postnatal maternal depression, but a few recent studies have begun to delin-
eate the presence of neurobiological alterations in these children. For example,
elevated serum cortisol concentrations in the elementary-school-aged children
of depressed mothers were positively correlated in one study with the severity
of maternal depression [72]. In addition, preschool-aged children of depressed
women exhibit EEG alterations in frontal lobe activity that correlate with
diminished empathy and other behavioral problems [49, 50].
Preclinical Evidence Regarding Child Exposure to
Maternal Depression
Due to the ethical and logistical complexities of conducting psychiatric
research during pregnancy or with children, clinical data regarding the effects of
exposure to maternal depression have been understandably slow to accrue.
Similarly, the myriad of variables encountered such as obstetrical complications,
method of delivery, medication and environmental toxin exposure will require a
considerable sample size to provide definitive information. This is particularly
true when attempting to measure in children the neurobiological impact of expo-
sure. Fortunately, these data are complemented by an extensive line of laboratory
animal research bearing homology to maternal depression [73].
Prenatal stress studies in laboratory animals are largely consistent with the
clinical data, demonstrating an adverse impact on offspring growth [74, 75],
learning [7678], and attainment of developmental milestones [79]. Furthermore,
preclinical research indicates that certain biobehavioral aberrations induced by
prenatal stress persist into adulthood. For example, prenatally stressed adult rats
exhibit depression-like behaviors [80], anxiety-like behaviors in novel situa-
tions [8187], and behaviorally exaggerated responses to stress [8890]. The
behavioral alterations associated with prenatal stress are accompanied by
lasting neurobiological alterations. In particular, prenatally stressed animals
demonstrate multiple alterations in hypothalamic-pituitary-adrenal axis func-
tion including: (1) increased basal concentrations of plasma corticosterone
and adrenocorticotropin hormone [9193]; (2) heightened production of
corticotropin-releasing factor in the fetal hypothalamus [94], and (3) exaggerated
corticosterone responses to subsequent mild stressors [79, 9597]. The activity
of catecholamine [95, 98102] and serotonin [95, 103106] systems are also
altered in prenatally stressed laboratory animals.
Because considerable evidence indicates that depression interferes with
maternal parenting behavior, laboratory animal protocols that interfere with
maternal care (e.g. maternal separation, variable foraging) serve as potential
models of postnatal maternal depression. Like the prenatal stress models, these
models reliably precipitate similar adverse biobehavioral outcomes in the
offspring. Offspring subjected to neonatal maternal stress paradigms exhibit
insecure attachment behaviors [107110], depression-like and anxiety-like
behaviors [111], and alterations in stress-respondent neurobiological systems
[112116].
Conclusions
An impressive array of evidence now demonstrates that early adverse life
events contribute significantly to the vulnerability to adulthood psychopathol-
ogy, medical illnesses, and maladaptive psychosocial functioning. Whereas
parental nurture fosters adaptive responses to early life stressors that should
protect the child from such vulnerability, studies indicate that parental depres-
sion, specifically maternal depression since it has been subjected to greater
scrutiny than paternal depression, not only undermines this protection but can
itself represent an early adverse life event. Considered in tandem, the clinical
and preclinical data raise a persuasive argument that maternal depression dur-
ing pregnancy or early childhood represents a childs earliest adverse life event.
From a developmental standpoint, fetal and infant exposure to maternal depres-
sion can thereby contribute to a childs vulnerability to psychiatric illness in the
same manner as other adverse childhood experiences such as child abuse or the
loss of a parent. It is a unique feature of maternal depression, when compared to
other childhood psychosocial stressors, that it can even effect the antenatal
environment. Consequently, the effects of maternal depression add another
level of definition to diathesis/stress models in that the vulnerability to psychi-
atric illness can be conveyed from mother to child even in the prenatal context
via both genetic inheritance and environmental exposure (fig. 2).
Current findings regarding maternal depression carry profound clinical
implications for the children of depressed women. It is imperative that we better
understand the mechanisms whereby parental depression contributes to the
development of the childs subsequent psychiatric vulnerability. In particular, the
animal data, because it more readily affords assessment of the neurobiological
outcomes of exposure to maternal stress, may ultimately direct the development
of pharmacological interventions for a child once a developmental insult from
exposure to maternal depression has occurred. Such treatments could theoreti-
cally be useful either as a prophylactic intervention for children in the acute
aftermath of exposure or a symptom-alleviating treatment for adults with a
persistent vulnerability to illness that is, at least in part, a consequence of the
long-term effects of exposure to maternal depression. Quantification of the
severity or level of maternal depression/stress that poses these risks will serve to
directly inform treatment guidelines for pregnancy and the postpartum period.
The clinical implications of these data also extend to the psychiatric care of
depressed women during the reproductive years. Due caution should be exercised
when the decision to treat maternal depression would expose a fetus or nursing
child to an antidepressant medication; however, the risk/benefit assessment is
incomplete unless the adverse consequences of fetal or infant exposure to
untreated maternal depression have been considered. Furthermore, if successful
treatment of maternal depression protects the developing child from the unto-
ward effects of exposure to this earliest of adverse life events, then the restora-
tion of maternal mental health may in fact represent primary preventive care for
the otherwise exposed child. Consequently, a thorough risk/benefit analysis for
women at risk for experiencing depressive illness during pregnancy or lactation,
Illness
Vulnerable
phenotype
Antenatal
depression
Heredity
Postnatal
depression
Later stress
Maternal
depression
Fig. 2. Maternal depression and the diathesis/stress model of illness.
given our present state of knowledge, should take into account clinical factors
such as the potential deleterious effects of untreated illness on both mother
and child, the possible adverse effects of offspring exposure to alternative
psychotropic medications within particular developmental windows, and the
maternal psychiatric history that is useful in predicting the likelihood and
severity of maternal relapse in the event that treatment is withheld.
These data also raise important research questions regarding the future of
clinical investigation during pregnancy and lactation and the development of
animal models appropriate to the study of maternal depression. Although it has
been argued and is widely accepted that conducting clinical trials in pregnant
and nursing women is on its face unethical [117], failing to conduct such stud-
ies arguably deprives women of available treatment and may over time increase
the overall risk to women and their infants. Alternative ethical positions regard-
ing perinatal clinical research have been raised. In considering alternative inter-
ventions following evidence of in utero failure to thrive, obstetrical researchers
have advocated applying the uncertainty principle [118] to the ethical question
of antenatal clinical research [119]. The uncertainty principle dictates that a
patient, even a pregnant patient, should be ethically eligible to participate in a
randomized clinical trial if the preferred clinical course of action is uncertain.
Because the relative risks to fetus and infant of exposure to maternal depression
versus exposure to maternal psychotropic medication is often unclear, the
uncertainty principle may be readily applicable to the study of maternal depres-
sion and other psychiatric disorders during pregnancy and lactation. Such
issues must be thoughtfully deliberated so that perinatal psychiatric research
can be advanced in a productive, yet ethically responsible, manner.
Finally, continued refinement of animal models of human parental behav-
ior will undoubtedly hasten our understanding of the impact of depression and
other disorders that interfere with parental nurture on offspring vulnerability to
stress-related illness. Animal models offer several distinct advantages over
human research that can serve to expedite the pace of research. Despite these
advantages, an animal model with poor homology to the human experience
is not particularly useful. Among the existing animal paradigms, the variable
foraging model demonstrates the greatest face validity for human maternal
depression. This model avoids direct interaction between researchers and
animals, and it does not abolish maternal care but instead introduces an envi-
ronmental stressor that challenges the maternal capacity for care giving [73].
Although this more closely parallels the difficulties confronted by depressed
mothers than other animal models implemented to date, it must be acknowl-
edged that the stress of a systematically controlled environmental perturbation
is not fully homologous to the presence of a stress-related psychiatric disorder.
To advance our understanding of the developmental effects of exposure to
maternal depression, future preclinical studies should endeavor both: (1) to
continue refining the homology of the models to the human experience (per-
haps by using biparental species in which both mother and father routinely par-
ticipate in care of the offspring), and (2) to expand the domain of neurobiological
inquiry in the offspring beyond that of a single, isolated substrate such as corti-
sol or serotonin to the simultaneous exploration of the effects of maternal stress
on multiple biological systems while tracing the association of any observed
systemic changes with behavioral outcomes.
Acknowledgements
This project was supported in part by a National Institutes of Health (NIH) Patient-
Oriented Research Career Development Award (K23 MH-63507) and an NIH Specialized
Center of Research Grant (P50 MH-68036).
References
1 Freud S: The Complete Introductory Lectures on Psychoanalysis (translated and edited by
J. Strachey). New York, Norton, 1966.
2 Klein M: The Psychoanalysis of Children (translated by A. Strachey). New York, Grove Press, Inc,
1960.
3 Klein M: Notes on some schizoid mechanisms. J Psychother Pract Res 1996;5:164179.
4 Bowlby J: Attachment and Loss. New York, Basic Books, 1969.
5 Goodman S, Gotlib I: Risk for psychopathology in the children of depressed mothers: A develop-
mental model for understanding mechanisms of transmission. Psychol Rev 1999;106:458490.
6 Weissman MM, Bland RC, Canino GJ, Faravelli C, Greenwald S, Hwu HG, Joyce PR, Karam EG,
Lee CK, Lellouch J, Lepine JP, Newman SC, Rubio-Stipec M, Wells JE, Wickramaratne PJ,
Wittchen H, Yeh EK: Cross-national epidemiology of major depression and bipolar disorder.
JAMA 1996;276:293299.
7 Manly PC, McMahon RJ, Bradley CF, Davidson PO: Depressive attributional style and depression
following childbirth. J Abnorm Psychol 1982;91:245254.
8 OHara MW, Rehm LP, Campbell SB: Predicting depressive symptomatology: Cognitive-behavioral
models and postpartum depression. J Abnorm Psychol 1982;91:457461.
9 Watson JP, Elliott SA, Rugg AJ, Brough DI: Psychiatric disorder in pregnancy and the first postnatal
year. Br J Psychiatry 1984;144:453462.
10 Kumar R, Robson KM: A prospective study of emotional disorders in childbearing women.
Br J Psychiatry 1984;144:3547.
11 Cutrona CE: Causal attributions and perinatal depression. J Abnorm Psychol 1986;92:161172.
12 Buesching DP, Glasser ML, Frate DA: Progression of depression in the prenatal and postpartum
periods. Women Health 1988;11:6178.
13 OHara MW, Neunaber DJ, Zekoski EM: Prospective study of postpartum depression: Prevalence,
course, and predictive factors. J Abnorm Psychol 1984;93:158171.
14 Beck CT: Screening methods for postpartum depression. J Obstet Gynecol Neonatal Nurs
1995;24:308312.
15 Affonso DD, Lovett S, Paul SM, Sheptak S: A standardized interview that differentiates pregnancy
and postpartum symptoms from perinatal clinical depression. Birth 1990;17:121130.
16 Weissman MM, Olfson M: Depression in women: Implications for health care research. Science
1995;269:799801.
17 Stowe ZN, Hostetter AL, Newport DJ: The onset of postpartum depression: Implications for clinical
screening in obstetrical and primary care. Am J Obstet Gynecol, submitted.
18 Troutman B, Cutrona C: Nonpsychotic postpartum depression among adolescent mothers.
J Abnorm Psychol 1990;99:69.
19 Stowe ZN, Nemeroff CB: Women at risk for postpartum-onset major depression. Am J Obstet
Gynecol 1995;173:639645.
20 Kendler KS, Kessler RC, Neale MC, Heath AC, Eaves LJ: The prediction of major depression in
women: Toward an integrated etiologic model. Am J Psychiatry 1993;150:11391148.
21 Duffy CL: Postpartum depression: Identifying women at risk. Genesis 1983;11:21.
22 Perkin MR, Bland JM, Peacock JL, Anderson HR: The effect of anxiety and depression during
pregnancy on obstetric complications. Br J Obstet Gynaecol 1993;100:629634.
23 Steer RA, Scholl TO, Hediger ML, Fischer RL: Self-reported depression and negative pregnancy
outcomes. J Clin Epidemiol 1992;45:10931099.
24 Orr ST, Miller CA: Maternal depressive symptoms and the risk of poor pregnancy outcome.
Epidemiol Rev 1995;17:165171.
25 Korebrits C, Ramirez MM, Watson L, Brinkman E, Bocking AD, Challis JR: Maternal
corticotropin-releasing hormone is increased with impending preterm birth. J Clin Endocrinol
Metab 1998;83:15851591.
26 Majzoub J, McGregor JA, Lockwood CJ, Smith R, Taggart MS, Schulkin J: A central theory of
preterm and term labor: Putative role for corticotropin-releasing hormone. Am J Obstet Gynecol
1999;180:232241.
27 Lockwood C, Kuczynski E: Markers of risk for preterm delivery. J Perinat Med 1999;27:520.
28 Lockwood C: Stress-associated preterm delivery: The role of corticotropin-releasing hormone.
Am J Obstet Gynecol 1999;180:264266.
29 Wadhwa PD, Porto M, Garite TJ, Chicz-DeMet A, Sandman CA: Maternal corticotropin-releasing
hormone levels in the early third trimester predict length of gestation in human pregnancy.
30 Grammatopoulos D, Hillhouse E: Role of corticotropin-releasing hormone in onset of labor.
Lancet 1999;354:15461549.
31 Zuckerman B, Amaro H, Bauchner H, Cabral H: Depressive symptoms during pregnancy:
Relationship to poor health behaviors. Am J Obstet Gynecol 1989;160:11071111.
32 Shell L: Environmental noise and human prenatal growth. Am J Phys Anthropol 1981;56:6370.
33 Hedegaard M, Henriksen TB, Sabroe S, Secher NJ: The relationship between psychological
distress during pregnancy and birth weight for gestational age. Acta Obstet Gynecol Scand
1996;75:3239.
34 Hoffman S, Hatch M: Depressive symptomatology during pregnancy: Evidence for an association
with decreased fetal growth in pregnancies of lower social class women. Health Psychol
2000;19:535543.
35 Lobel M, DeVincent CJ, Kaminer A: The impact of prenatal maternal stress and optimistic dispo-
sition on birth outcomes in medically high-risk women. Health Psychol 2000;19:544553.
36 Gennaro S, York R, Brooten D: Anxiety and depression in mothers of low birthweight and very
low birthweight infants: Birth through 5 months. Issues Compr Pediatr Nurs 1990;13:97109.
37 Lou HC, Hansen D, Nordentoft M, Pryds O, Jensen F, Nim J, Hemmingsen R: Prenatal stressors of
human life affect fetal brain development. Dev Med Child Neurol 1994;36:826832.
38 Hansen D, Lou HC, Olsen J: Serious life events and congenital malformations: A national study
with complete follow-up. Lancet 2000;356:875880.
39 Lyons-Ruth K, Wolfe R, Lyubchik A: Depression and the parenting of young children: Making the
case for early preventive mental health services. Harv Rev Psychiatry 2000;8:148153.
40 Stott D: Follow-up study from birth of the effects of prenatal stresses. Dev Med Child Neurol
1973;15:770787.
41 OConnor TG, Heron J, Glover V: Antenatal anxiety predicts child behavioral/emotional
problems independently of postnatal depression. J Am Acad Child Adolesc Psychiatry 2002;41:
14701477.
42 Meijer A: Child psychiatric sequelae of maternal war stress. Acta Psychiatr Scand 1985;72:
505511.
43 Nulman I, Rovet J, Stewart DE, Wolpin J, Pace-Asciak P, Shuhaiber S, Koren G: Child develop-
ment following exposure to tricyclic antidepressants or fluoxetine throughout fetal life: A prospective,
controlled study. Am J Psychiatry 2002;159:18891895.
44 OConnor TG, Heron J, Golding J, Glover V: Maternal antenatal anxiety and behavioral/emotional
problems in children: A test of a programming hypothesis. J Child Psychol Psychiatry 2003;44:
10251036.
45 Luoma I, Tamminen T, Kaukonen P, Laippala P, Puura K, Salmelin R, Almqvist F: Longitudinal
study of maternal depressive symptoms and child well-being. J Am Acad Child Adolesc
Psychiatry 2001;40:13671374.
46 Lundy BL, Jones NA, Field T, Nearing G, Davalos M, Pietro PA, Schanberg S, Kuhn C: Prenatal
depression effects on neonates. Infant Behav Dev 1999;22:119129.
47 Calhoun K, Brennan P, Walker E, Stowe Z: The impact of maternal depression on infant salivary
cortisol (abstract). Am Psychiatr Assoc Annu Meet, New Orleans, 2001.
48 Jones NA, Field T, Fox NA, Lundy B, Davalos M: EEG activation in 1-month-old infants of
depressed mothers. Dev Psychopathol 1997;9:491505.
49 Dawson G, Frey K, Self J, Panagiotides J, Hessl D, Yamada E, Rinaldi J: Frontal brain electrical
activity in infants of depressed and nondepressed mothers: Relation to variations in infant behavior.
Dev Psychopathol 1999;11:589605.
50 Jones NA, Field T, Davalos M: Right frontal EEG asymmetry and lack of empathy in preschool
children of depressed mothers. Child Psychiatry Hum Dev 2000;30:189204.
51 Cox MT, Stowe ZN, Hostetter A, Ramsey C, Sternberg K, Ritchie JC: Cortisol concentrations in
the breast milk of women with major depression (abstract). Am Psychiatr Assoc Annu Meet,
Toronto, 2000.
52 Lovejoy C, Graczyk P, OHare E, Neuman G: Maternal depression and parenting behavior:
A meta-analytic review. Clin Psychol Rev 2000;20:561592.
53 Leiferman J: The effect of maternal depressive symptomatology on maternal behaviors associated
with child health. Health Educ Behav 2002;29:596607.
54 Righetti-Velterna M, Bousquet A, Manzano J: Impact of postpartum depressive symptoms on
mother and her 18-month-old infant. Eur Child Adolesc Psychiatry 2003;12:7583.
55 Blank DM: Relating mothers anxiety and perception to infant satiety, anxiety, and feeding behavior.
Nurs Res 1986;35:347351.
56 Righetti-Velterna M, Conne-Perreard E, Bousquet A, Manzano J: Postpartum depression and
mother-infant relationship at 3 months old. J Affect Disord 2002;70:291306.
57 Atkinson L, Paglia A, Coolbear J, Niccols A, Parker K, Guger S: Attachment security: A meta-
analysis of maternal mental health correlates. Clin Psychol Rev 2000;20:10191040.
58 Martinez A, Malphurs J, Field T, Pickens J, Yando R, Bendell D, Valle C, Messinger D: Depressed
mothers and their infants interactions with nondepressed partners. Infant Ment Health J 1996;17:
7480.
59 Downey G, Coyne JC: Children of depressed parents: An integrative review. Psychol Bull
1990;108:5076.
60 Jameson PB, Gelfand DM, Kulcsar E, Teti DM: Mother-toddler interaction patterns associated
with maternal depression. Dev Psychopathol 1997;9:537550.
61 Field T, Diego MA, Dieter J, Hernandez-Reif M, Schanberg SM, Kuhn C, Yando R, Bendell D:
Depressed withdrawn and intrusive mothers effects on their fetuses and neonates. Infant Behav
Dev 2001;24:2739.
62 Galler JR, Harrison RH, Ramsey F, Forde V, Butler SC: Maternal depressive symptoms affect
infant cognitive development in Barbados. J Child Psychol Psychiatry 2000;41:747757.
63 Field T: Maternal depression effects on infants and early interventions. Prev Med 1998;27:200203.
64 Weissman MM, Prusoff BA, Gammon GD, Merikangas KR, Leckman JF, Kidd KK: Psycho-
pathology in the children (ages 618) of depressed and normal parents. J Am Acad Child Adolesc
65 Kurstjens S, Wolke D: Effects of maternal depression on cognitive development of children over
the first 7 years of life. J Child Psychol Psychiatry 2001;42:623636.
66 Hay DF, Pawlby S, Angold A, Harold GT, Sharp D: Pathways to violence in the children of mothers
who were depressed postpartum. Dev Psychol 2003;39:10831094.
67 Billings A, Moos R: Comparisons of children of depressed and nondepressed parents: A social-
environmental perspective. J Abnorm Child Psychol 1983;11:483486.
68 Beardslee W, Keller M, Klerman G: Children of parents with affective disorder. Intl J Fam
69 Anderson C, Hammen C: Psychosocial outcomes of unipolar depressed, bipolar, medically ill, and
normal women: A longitudinal study. J Consult Clin Psychol 1993;61:448454.
70 Goodman S, Brogan D, Lynch M, Fielding B: Social and emotional competence in children of
depressed mothers. Child Dev 1993;64:516531.
71 Hammen C, Gordon D, Burge D, Adrian C, Jaenicke C, Hiroto D: Maternal affective disorders,
illness, and stress: Risk for childrens psychopathology. Am J Psychiatry 1987;144:736741.
72 Lupien SL, King S, Meaney MJ, McEwen BS: Childs stress hormone levels correlate with
mothers socioeconomic status and depressive state. Biol Psychiatry 2000;48:976980.
73 Newport DJ, Stowe ZN, Nemeroff CB: Parental depression: Animal models of an adverse life
event. Am J Psychiatry 2002;159:12651283.
74 Herrenkohl LR, Gala RR: Serum prolactin levels and maintenance of progeny by prenatally
stressed female offspring. Experientia 1979;35:702704.
75 Schneider M, Roughton E, Koehler A, Luback G: Growth and development following prenatal
stress exposure in primates: An examination of ontogenetic vulnerability. Child Dev 1999;70:
263274.
76 Smith B, Wills G, Naylor D: The effects of prenatal stress on rat offsprings learning ability.
J Psychol 1981;107:4551.
77 Weller A, Glaubman H, Yehuda S, Caspy T, Ben-Uria Y: Acute and repeated gestational stress
affect offspring learning and activity in rats. Physiol Behav 1988;43:139143.
78 Jaiswal A, Bhattacharya S: Effects of gestational undernutrition, stress and diazepam treatment on
spatial discrimination learning and retention in young rats. Indian J Exp Biol 1993;31:353359.
79 Fride E, Weinstock M: The effects of prenatal exposure to predictable or unpredictable stress on
early development in the rat. Dev Psychobiol 1984;17:651660.
80 Secoli S, Teixeira N: Chronic prenatal stress affects development and behavioral depression in
rats. Stress 1998;2:273280.
81 Fride E, Weinstock M: Prenatal stress increases anxiety related behavior and alters cerebral later-
alization of dopamine activity. Life Sci 1988;36:10591065.
82 Weinstock M, Fride E, Hertzberg R: Prenatal stress effects on functional development of the
offspring. Prog Brain Res 1988;73:319331.
83 Szuran T, Zimmerman E, Pliska V, Pfister H, Welzl H: Prenatal stress effects on exploratory activity
and stress-induced analgesia in rats. Dev Psychobiol 1991;24:361372.
84 Schneider M: Prenatal stress exposure alters postnatal behavioral expression under conditions of
novelty challenge in rhesus monkey infants. Dev Psychobiol 1992;25:529540.
85 Clarke AS, Schneider ML: Prenatal stress has long-term effects on behavioral responses to stress
in juvenile rhesus monkeys. Dev Psychobiol 1993;26:293304.
86 Poltyrev T, Keshet G, Kay G, Weinstock M: Role of experimental conditions in determining
differences in exploratory behavior of prenatally stressed rats. Dev Psychobiol 1996;29:453462.
87 Vallee M, Mayo W, Dellu F, Le Moal M, Simon H, Maccari S: Prenatal stress induces high anxiety
and postnatal handling induces low anxiety in adult offspring: Correlation with stress-induced
corticosterone secretion. J Neurosci 1997;17:26262636.
88 Fride E, Dan Y, Feldon J, Halevy G, Weinstock M: Effects of prenatal stress on vulnerability to
stress in prepubertal and adult rats. Physiol Behav 1986;37:681687.
89 Wakshlak A, Weinstock M: Neonatal handling reverses behavioral abnormalities induced in rats
by prenatal stress. Physiol Behav 1990;48:289292.
90 Pfister H, Muir J: Prenatal exposure to predictable and unpredictable novelty stress and oxy-
tocin treatment affects offspring development and behavior in rats. Int J Neurosci 1992;62:
227241.
91 Takahashi L, Kalin N: Early developmental and temporal characteristics of stress-induced secre-
tion of pituitary-adrenal hormones in prenatally stress rat pups. Brain Res 1991;558:7578.
92 McCormick CM, Smythe JW, Sharma S, Meaney MJ: Sex-specific effects of prenatal stress on
hypothalamic-pituitary-adrenal responses to stress and brain glucocorticoid receptor density in
adult rats. Brain Res 1995;84:5561.
93 Takahashi LK: Prenatal stress: Consequences of glucocorticoids on hippocampal development
and function. Int J Dev Neurosci 1998;16:199207.
94 Fujioka T, Sakata Y, Yamaguchi K, Shibasaki T, Kato H, Nakamura S: The effects of prenatal stress
on the development of hypothalamic paraventricular neurons in fetal rats. Neuroscience 1999;92:
10791088.
95 Peters D: Prenatal stress: Effects on brain biogenic amine and plasma corticosterone levels.
Pharmacol Biochem Behav 1982;17:721725.
96 Weinstock M, Matlina E, Maor G, Rosen H, McEwen B: Prenatal stress selectively alters the
reactivity of the hypothalamic-pituitary-adrenal system in female rats. Brain Res 1992;595:
195198.
97 Henry C, Kabbaj M, Simon H, Le Moal M, Maccari S: Prenatal stress increases the hypothalamo-
pituitary-adrenal axis response in young and adult rats. J Neuroendocrinol 1994;6:341345.
98 Peters D: Prenatal stress: Effect on development of rat brain adrenergic receptors. Pharmacol
Biochem Behav 1984;21:417422.
99 Fride E, Dan Y, Gavish M, Weinstock M: Prenatal stress impairs maternal behavior in a conflict
situation and reduces hippocampal benzodiazepine receptors. Life Sci 1985;36:2103.
100 Alonso SJ, Navarro E, Rodriguez M: Permanent dopaminergic alterations in the nucleus accum-
bens after prenatal stress. Pharmacol Biochem Behav 1994;49:353358.
101 Henry C, Guegant G, Cador M, Arnauld E, Arsaut J, Le Moal M, Demotes-Mainard J: Prenatal
stress in rats facilitates amphetamine-induced sensitization and induces long-lasting changes in
dopamine receptors in the nucleus accumbens. Brain Res 1995;685:179186.
102 Alonso SJ, Navarro E, Santana C, Rodriguez M: Motor lateralization, behavioral despair
and dopaminergic brain asymmetry after prenatal stress. Pharmacol Biochem Behav 1997;58:
443448.
103 Peters D: Prenatal stress increases the behavioral response to serotonin agonists and alters open
field behavior in the rat. Pharmacol Biochem Behav 1986;25:873877.
104 Peters D: Effects of maternal stress during different gestational periods on the serotonergic system
in adult rat offspring. Pharmacol Biochem Behav 1988;31:839843.
105 Peters D: Maternal stress increases fetal brain and neonatal cerebral cortex 5-hydroxytryptamine
synthesis in rats: A possible mechanism by which stress influences brain development. Pharmacol
Biochem Behav 1990;35:943947.
106 Hayashi A, Nagaoka M, Yamada K, Ichitani Y, Miake Y, Okado N: Maternal stress induces synaptic
loss and developmental disabilities of offspring. Int J Dev Neurosci 1998;16:209216.
107 Rosenblum L, Paully G: The effects of varying environmental demands on maternal and infant
behavior. Child Dev 1984;55:305314.
108 Andrews M, Rosenblum L: Relationship between foraging and affiliative social referencing in
primates; in Fa J, Southwick C (eds): Ecology and Behavior of Food-Enhanced Primate Groups.
New York, Alan R Liss, 1988, pp 247268.
109 Andrews M, Rosenblum L: Attachment in monkey infants raised in variable and low-demand
environments. Child Dev 1991;62:686693.
110 Andrews M, Rosenblum L: Developmental consequences of altered dyadic coping patterns in
bonnet macaques; in Roeder JJ, Thierry B, Anderson JR, Herrenschmidt N (eds): Current
Primatology: Social Development, Learning and Behavior. Strasbourg, Presses de lUniversit
Louis Pasteur, 1994, pp 265271.
111 Levine S: Maternal and environmental influences on the adrenocortical response to stress in weanling
rats. Science 1967;156:258260.
112 Pauk J, Kuhn C, Field T, Schanberg S: Positive effects of tactile vs kinesthetic or vestibular stimu-
lation on neuroendocrine and ODC activity in maternally deprived rat pups. Life Sci 1986;39:
20812087.
113 Walker C, Scribner K, Cascio C, Dallman M: The pituitary-adrenocortical system of neonatal rats
is responsive to stress throughout development in a time-dependent and stressor-specific fashion.
Endocrinology 1991;128:13851395.
114 Pihoker C, Owens MJ, Kuhn CM, Schanberg SM, Nemeroff CB: Maternal separation in neonatal
rats elicits activation of the hypothalamic-pituitary-adrenocortical axis: A putative role for
corticotropin-releasing factor. Psychoneuroendocrinology 1993;7:485493.
115 Coplan J, Andrews M, Rosenblum LA, Owens MJ, Friedman S, Gorman J, Nemeroff CB:
Persistent elevations of cerebrospinal fluid concentrations of corticotropin-releasing factor in
adult nonhuman primates exposed to early-life stressors: Implications for the pathophysiology of
mood and anxiety disorders. Proc Natl Acad Sci USA 1996;93:16191623.
116 Coplan J, Trost R, Owens MJ, Cooper T, Gorman J, Nemeroff CB, Rosenblum L: Cerebrospinal
fluid concentrations of somatostatin and biogenic amines in grown primates reared by mothers
exposed to manipulated foraging conditions. Arch Gen Psychiatry 1998;55:473477.
117 Kerns LL: Treatment of mental disorders in pregnancy: A review of psychotropic drug risks and
benefits. J Nerv Ment Dis 1986;174:652659.
118 Peto R, Baigent C: Trials: The next 50 years. BMJ 1998;317:11701171.
119 Vail A, Hornbuckle J, Spiegelhalter DJ, Thornton JG: Prospective application of Bayesian moni-
toring and analysis in an open randomized clinical trial. Stat Med 2001;20:37773787.
D. Jeffrey Newport, MD, MS, MDiv
Emory University School of Medicine, Womens Mental Health Program
1365 Clifton Road NE, Suite B6100
Atlanta, GA 30322 (USA)
Tel. 1 404 778 2524, Fax 1 404 778 2535, E-Mail jeff.newport@emory.edu
Perinatal Infanticide and Suicide
Margaret G. Spinelli
College of Physicians and Surgeons of Columbia University,
and The New York State Psychiatric Institute, New York, N.Y., USA
The subject of maternal infanticide is met with complex reactions. Worldwide
estimates are grossly underreported, and infant death statistics are likewise under-
estimated. Similarly, there is a small literature on postpartum maternal suicide.
The purpose of this work is to provide an overview of maternal infanticide
and suicide. The goal is to identify deficiencies in postpartum research and lit-
erature on the causes, prevention and phenomenology of childbirth-associated
disorders in order to prevent these tragedies.
History of Infanticide
In the year 2000, the worldwide prevalence of violent deaths of children
(between birth and 4 years old) accounted for 55,000 of 702,000 deaths [1].
Infanticide or infant murder in the first year of life accounts for 1/3 of all US
deaths due to injury. According to available data, one infant is killed every day
in the USA [2]. Estimates double this number [3, 4]. Yet, data and research on
infanticide are scarce. Infant death statistics are likewise glaringly underesti-
mated [5]. In particular, there is a scarcity of data on infant fatalities from abuse
or neglect particularly as related to the perpetrators. This underreporting is
accounted for by poor documentation, infanticides reported as deaths from
sudden infant death syndrome, lack of death certificates and undocumented
births due to pregnancy denial and unfounded corpses [4, 6]. Undoubtedly, the
very nature of such a tragedy makes it an unappealing subject of research.
Parts of this paper have been adapted from Infanticide: A worldwide view in review at
the Archives of Womens Mental Illness.
Spinelli 86
Historically, the subject of infanticide has been treated with ambivalence.
The very same tension between the demand for condemnation and the impulse
towards mercy describes the evolution of infanticide laws [7] from ancient
times in both Western society and non-Western, primitive cultures [8].
In the Greek and Roman era, birth control and eugenic reasons were
primary causes for infant murder by fathers who had absolute rule in the family.
Infants were sacrificed to pagan gods, and unwanted newborns were exposed to
the elements as a method of population control [9, 10].
As the practice of infanticide became more common, the Catholic Church
was the first to institute penalties. Overlaying was a practice in which mothers
lay on the infant smothering it to death [7]. In the manuals for parish priests,
overlaying was identified as a venial sin comparable to failing to teach a child
proper manners.
When the church elevated infant murder to a mortal sin, societies adopted
laws in hopes for prevention [11], and secular penalties became increasingly
severe. By the 17th century, infanticide was so common that concealment of a
murdered newborn became a capital offense [7, 9]. Such punishments as sack-
ing were initiated in which a woman was placed in a sack with a dog, a cock,
and a snake and thrown into the water.
In the 18th century, laws in the USA, Canada and Europe became increas-
ingly strict particularly for unmarried women. In France, England and Russia,
the growing public awareness of the problem of dead and live abandoned
newborns led to the creation of the foundling homes, but they were unable to
overcome the profound effects of infanticide and abandonment.
In 1647, Russia became the first country to adopt a more humane attitude
and by 1888, all European states except England established a legal distinction
between infanticide and murder by assigning more lenient penalties to infanti-
cide [7]. In 1922 and 1938, England passed the Infanticide Act in recognition of
the time surrounding childbirth as biologically vulnerable and made infanticide
a less severe crime proscribing sentences of probation and mandatory psychi-
atric treatment for women found guilty. Today, almost all Western societies have
adjusted the penalty for infanticide [11] by recognizing the unique biological
changes that occur at childbirth.
These early legal statutes have evolved into contemporary and contrasting
legal views across the USA, UK and other countries. Oberman [7] estimated that
29 countries describe statutes explicitly governing the crime of infanticide. All
nations that have statutes make infanticide a less severe crime than ordinary
homicide with sentences considerably less severe than for manslaughter or mur-
der. This is the case for example in Austria, Finland, Greece, India, Italy, Korea,
New Zealand, the Philippines, New South Wales, Western Australia, Tasmania
and Canada. Moreover, New South Wales ascribes diminished responsibility to
Perinatal Infanticide and Suicide 87
infant murder. However, there is one exception: Luxembourg provides a more
severe penalty for killing a child than for other homicides.
According to the Italian Penal Code, killing a parent is punishable by
2430 years while infanticide is punishable by 310 years. Furthermore, Italys
law specifically provides for those who commit infanticide in order to save
their honor [7]. Killing an illegitimate compared to a legitimate child is a less
serious crime in the Philippines and Austria.
Some statutes also differ with regard to the infants age. For example New
Zealands law includes infant murder from immediately after birth to age 10
with a maximum of 3 years in jail. Most statutes follow the British rule, which
pertains to the first 12 months of life. Women are often given probation and
mandatory psychiatric treatment in countries with such statutes [12].
Generally speaking, infanticide has been treated as a far different crime
than other homicides. Yet contemporary cries for greater punishment can be
heard from abolitionists in countries where law reformers propose a change in
legislation by overturning the existing infanticide law.
The United States has no particular laws governing infanticide. A woman
who kills an infant is charged with the crime of homicide. If convicted in the
American judicial system, she may face a long prison sentence or even the
death penalty. Due to the scarcity of psychiatric treatment in the overcrowded
US prison system, these women exit the system in their childbearing years with
the same psychopathology that brought them into prison. Yet, there is no differ-
ence in the prevalence of infanticide between countries that mandate treatment
and those that mandate punishment [13].
The Case of Andrea Yates: A Failure of Society and
US Legislation
In 2001, when a psychotic Andrea Yates drowned her 5 children in the
bathtub of her Houston, Texas home [14], the nation was riveted and the
Western world responded. Andrea Pia Yates was a nurse, a devoted mother and
wife who home-schooled her 5 children. She had a long history of psychiatric
illness including previous postpartum psychotic episodes. Several hospitalizations
were due to suicide attempts driven by efforts to resist satanic voices command-
ing her to kill her infant and children [15].
Six months after her 5th child was born, Andrea Yates family and friends
described her as a caged animal as she continued to deteriorate. For unclear
reasons, her psychiatrist discontinued her antipsychotic medication, precipitat-
ing a florid psychotic delusional state. She believed that Satan was directing her
to kill her children; otherwise they would be condemned to the fires of hell.
Spinelli 88
Yates was found guilty of capital murder after only 3h of jury delibera-
tion. Spared the death penalty by the same jury, she was remanded to prison for
life [14]. Her trial attracted international attention. Organizations dedicated to
postpartum disorders requested clarification of postpartum DSM-IV diagnostic
criteria, improved medical education, guidelines for treatment and consideration
of infanticide legislation. Advocates for the mentally ill blamed the inadequacy of
the courts, the use of an archaic insanity [15] defense and the troubling nature
of expert psychiatric witnesses whose opinions differed so remarkably.
US Insanity Laws
The outcome of Yates vs. Texas is representative of many insanity pleas in
the US courts. A diagnosis of psychosis does not imply insanity under the US
law. Despite overwhelming agreement by the defense and the prosecution that
Andrea Yates was psychotic at the time of her actions, she was found not
legally insane [16].
Diagnostic standards do not exist for postpartum psychiatric illness,
and a woman who commits infanticide may receive sentences that vary remark-
ably [15, 17]. Depending on the state, the defendant must pass the test of that
jurisdiction in order to be found not guilty by reason of mental illness.
A woman who receives a prison sentence in one state may receive the death
penalty in another despite the identical circumstances of the crime. Outcomes
vary depending on the state, county or even the presiding judge [18].
While some states provide no defense of insanity, the insanity defense in
most American jurisdictions is based on two main formulations for insanity: the
MNaghten Test (1843) and the Model Penal Code/American Law Institute Test
(MPC).
The MPC is the second test of insanity, which provides that a defendant is
not responsible for a crime if she lacked capacity to appreciate the criminality
(right or wrong) of her conduct or was unable to conform her conduct to the
law. The expert psychiatric witness makes these respective determinations.
The MNaghten Test, or the right and wrong test, was derived from the
landmark English case decided in 1843 and represents the cognitive test for
insanity in the state of Texas where Andrea Yates was prosecuted. According to
MNaghton, the finding of insanity under the law depends on the cognitive
ability of the individual or the ability to know right from wrong at the time of
the crime.
There are inherent problems with this test. First, the likelihood that a
160-year-old legal case can be applied to 21st century neuroscience to accu-
rately determine a state of insanity is improbable. Second, this test relies on the
defendants recall at the time of the crime. Since psychosis and particularly
organic states are frequently associated with amnesia, the reliability of these
retrospective reports is debatable. Finally, the psychiatric literature is replete
with clinical case findings of postpartum psychosis and associated mood labil-
ity confounded by delirium, amnesia and impaired cognition.
The MPC approach has been adopted by about half of the states and the
majority of the federal circuit courts of appeal [19]. It recognizes that mental
disease may impair functioning in several ways. The satisfaction of either the
cognitive or volitional prong is grounds for an insanity verdict in a MPC juris-
diction. It is likely that Andrea Yates would have been found not guilty by
reason of mental illness in a state using the MPC test.
Ambiguity in the law is further complicated by psychiatrys failure to
provide diagnostic guidelines for postpartum psychiatric illness. Consequently, we
have little to offer the legal justice system [19]. This presents a virtual catch-22.
In a court of law, expert witness testimony must be founded on scientific stan-
dards that are recognized in the psychiatric community, yet few standards exist.
Therefore, the defenses for these women are limited to early and outdated liter-
ature and laws. Our reluctance to place postpartum disorders within a diagnos-
tic framework often leads to tragic outcomes for women, family and society.
Moreover, it continues to result in disparate treatment for women in the legal
system overall [20].
For example, when evidence for postpartum illness could assist womens
interests in criminal cases, it is often barred from admission [20, 21]. On the
other hand, postpartum syndromes are readily admitted into evidence during
civil proceedings [20] where they are almost always in opposition to a womans
interests such as child custody or adoption decisions.
Contemporary and Contrasting Views
Contrary to US legislation, those of England and Wales provide that a
woman who has killed her infant aged under a year can be indicted for infanti-
cide. This law contained in the Infanticide Act of 1938 [22] represents a model
of infanticide legislation often described in the literature.
Where a woman by any wilful act or omission causes the death of her child aged less
than a year but at the time the balance of her mind was disturbed by reason of her not hav-
ing fully recovered from the effect of giving birth to the child or by reason of the effect of lac-
tation the offence which would have amounted to murder is deemed to be infanticide and
is dealt with and punished as if it were manslaughter.
The Infanticide Act assumes that should a woman kill an infant under
1 year of age, it is likely a consequence of mental instability. For example,
Spinelli 90
a 19-year-old British mother who killed her 6-month-old baby was due to stand
trial for murder, but the Crown Prosecution Service accepted a guilty plea
to infanticide because she had suffered from postnatal depression [23]. She
was given a 4-year sentence in a hospital to be treated for her psychiatric
problems.
Marks [13] describes the ongoing debate over the Infanticide Act in the
UK. Proponents of abolition in the UK argue that medicalisation of the
offence encourages tolerance of infant murder. Marks argues that rather than
abolish the Infanticide Act, what is needed is research into the reasons that lead
the prosecution to bring charges of murder, manslaughter or infanticide. Further
evidence to support this Act comes from a comparison of England and Wales
with Scotland.
Scottish legislation makes no special provision for maternal infanticide.
A mother who kills her infant in Scotland will be charged with either murder or
homicide. If harsher legislation is a deterrent, then rates should be lower.
However, rates of the offence, the characteristics of victims and perpetrators
and the patterning of both convictions and sentences are similar in the two
regions [24]. Abolition of the Act is unlikely to result in a reduction in the
number of infants killed, nor to facilitate research into the precursors of these
crimes.
The legislative debate and dilemma are not particular to the UK. The para-
doxical responses to infanticide are aroused regardless of geography, legislation
or social structure.
China illustrates the paradoxical manner in which these cases are viewed
from a social perspective. Despite a policy of sex selection and population con-
trol [25], a 15-year-old Hong Kong girl who threw a newborn to death went on
trial for murder.
Economic and cultural realities can contribute to the prevalence and even
acceptance of infanticide. Although most countries have specific laws that rule
out harsh sentences, there are no simple predictions to the United States experi-
ence of seemingly similar crimes, but radically different sentences. In this
nation, states set their own policies for handling such cases.
In Texas, Andrea Yates began serving a life prison term last year for
drowning her 5 children in a bathtub. Despite overwhelming agreement by
the defense and the prosecution that she was psychotic, she was found not
legally insane. In the USA, the strong political and social basis for the failure
of the insanity defense makes no provision for the scientific fact of mental
illness.
In a report by the Calgary Herald [26], psychologist Tana Dineen of
Victoria opined in the case of Andrea Yates that psychiatric testimony should
not be allowed in court. All you get is a lot of speculation.
Law, Gender and Science
The basis of infanticide legislation in most countries reflects concern for
the biologically vulnerable state of women after childbirth. The Victorian Law
Reform Commission challenges this accepted wisdom as it lobbies for abolition
of the Infanticide Act.
Journalist Janet Albrechtsen [27] demands that punishment for infanticide
should be based on justice, not gender, with mothers and fathers treated equally.
Patricia Pearson exclaims if a hand rocks the cradle it must bear the blame it
comes with the territory of equality [27].
Ironically, the proponents of abolition claim that medicalization is outdated.
Yet contemporary neuroscience overwhelmingly supports the neurohormonal
basis for postpartum illness [28].
Despite evidence for physiological underpinnings of postpartum disorder,
our limited diagnostic guidelines and the scarcity of research limit the role of
the psychiatric community in the American court system.
Contemporary perinatal psychiatrists have failed to demonstrate a unique
phenomenology to postpartum disorders, precluding a distinct diagnostic cate-
gory in the DSM-IV. Cognitive disturbance has been described and demon-
strated by a formal systematic study. Wisner et al. [29] described a cognitive
disorganization psychosis in women with childbearing-related psychoses com-
pared to women with non-childbearing-related psychoses. The clinical picture
in the postpartum group demonstrated thought disorganization, bizarre behav-
ior, confusion, delusions of persecution and impaired sensorium/orientation, all
consistent with a clinical picture of delirium. This biologically driven state pres-
ents as any other toxic organic psychosis [28] such as thyrotoxicosis, severe
B
12
deficiency, and hypopituitary states. Unusual psychotic symptoms, such as
tactile, olfactory and visual hallucinations [29] present in a woman who looks
well at one moment, but floridly psychotic in the next.
The lack of formal diagnostic criteria for postpartum disorders is particu-
larly deleterious in the US courtroom where the judicial community relies on
the DSM for testimony. In the legal arena, the systematic presentation of the
perinatal literature is usually absent from the repertoire of psychiatric experts
who testify in courtrooms. The consequence is that jurors responsible for the
fate of these women are not informed about the illness associated with child-
birth. Consequently, jurors are more likely to attribute secondary gain to pleas
for insanity in order to avoid prosecution. Improved documentation in the
psychiatric literature could indeed prompt major changes if communicated to
the legal justice system and the juries who decide culpability.
The challenge for psychiatry is to educate the legal community. The task
for the expert witness is to communicate our scientific and biologically based
Spinelli 92
knowledge to the jury, to use the courtroom as a classroom and encourage
verdicts based on informed understanding of the facts.
The trial of Andrea Yates represents further inequities in the American
system. The very fact that two experts can find drastically different outcomes
from one defendant speaks to the dearth of scientific evidence underlying these
decisions. Psychiatric expert witnesses defer to the legal community and judi-
cial system by fitting our criteria into archaic laws that have no basis in fact.
The medical ethics surrounding these cases demands that the patient defendant
be given the best treatment based on our contemporary biopsychosocial model
of psychiatry.
After sentencing Andrea Yates, the American Psychiatric Association made
a public announcement on the insanity defense and mental illness [30].
The American Psychiatric Association hopes that the Yates case will lead to broad
public discussion of how our society and its legal system deals with defendants who are
severely mentally ill. Advances in neuroscience have dramatically increased our
understanding of how brain function is altered by mental illness, and how psychotic illness
can distort reality. Unfortunately, public understanding has not kept pace with these
advances.
A failure to appreciate the impact of mental illness on thought and behavior often lies
behind decisions to convict and punish persons with mental disorders. Defendants whose
crimes derive from their mental illness should be sent to a hospital and treated not cast into
a prison, much less onto death row.
The legal and medical consensus was that Andrea Yates was psychotic
[31]. The prosecuting psychiatrist as expert witness testified that Yates knew
the act was wrong at the time of the killings. The psychiatric expert for the
defense testified that the patient knew that it was legally wrong to kill. He
also testified that Yates knew it was morally right to kill her children because
Satan commanded that her children would suffer the fires of hell if she did not
drown them as if her delusional thinking and decision making could be
applied to reality-based thinking of the law.
As MNaghton does not specify that cognition should apply to legal or
moral knowledge suggests further distortion of the facts. The empathic human
being might ask whether cognitive factors should be considered when a
mothers distorted reality demands that she save her children from an eternity of
pain. We, in psychiatry, make decisions that conform to legal standards that
have no application to our scientific determination and therefore cannot possi-
bly determine culpability. The fact that we attribute decision-making capacity to
a psychotic mother speaks to the inaccuracies in our system. Impaired cognition
or inability to know belongs to the pathophysiology that describes postpartum
psychosis. How then can we use this phenomenon to indight an individual with
postpartum psychosis?
The single most important piece of judicial evidence for the existence of a
clinical entity lies in the description of the phenomenology in peer-reviewed lit-
erature. The dearth of descriptive symptomatology on infanticide and postpar-
tum psychosis leaves the expert witness with few scientific tools. The courts
rely on scientific knowledge that is accepted in the medical community [32].
Absent systematic clinical descriptions or research-based case reports, each act
is judged in isolation with little or no regard for similar cases [18]. And so, the
case of Texas vs. Yates was tried in the media and courts with little intervention
from the psychiatric community.
Absent research-based information on the temporal relationship between
childbirth and infanticide, and a clinical framework for understanding the diagno-
sis and phenomenology that underlie infanticide, we are in all likelihood missing
the signs of potential tragedy as evidenced by the case of Texas vs. Yates [16].
The Tragedy of the Yates Family: What Can We Learn?
Before she killed her children, Andrea Yates spent days in her bed pulling
out her hair to demonstrate 666 (the mark of the beast) on her scalp. Was she
begging to be safeguarded, to be recognized as Satan?
Yates had a family and personal history of psychiatric illness. While preg-
nant or lactating over the past 7 years, she experienced three postpartum psy-
chotic episodes complicated by infanticidal ideation, suicide attempts and
hospitalizations. Mr. Yates seemed indifferent to his wifes illness and the
squalor of the household. During a 1999 hospitalization, Mrs. Yates reported to
the staff that she was overwhelmed [16] living in a converted Greyhound bus
with her growing family of 4 children [14]. The social worker filed the report
with Childrens Protective Services but the state agency did not pursue the case.
Mr. Yates testified that he gave his wife 2 weekly hours of personal time. He
stated that their religious beliefs encouraged additional children to which
Andrea agreed when she was feeling better after deliveries.
The fact that both the inpatient and outpatient hospital team released a psy-
chotic mother to her home and children then withdrew her antipsychotic med-
ication speaks to poor medical education and management, and overall failure
of the system. The final word of the legal system to determine her culpability
using archaic laws and values was the ultimate failure of humanity.
We, as a society, failed Andrea Yates. We share equal responsibility for the
tragedy. Friends, neighbors and family failed to see or report as Mrs. Yates con-
tinued to decompensate. The medical community failed to provide appropriate
protection, social work assistance and child services to a severely psychotic
Spinelli 94
mother of 5 children. When the legal community and her state failed to appreciate
the severity of her illness, they eliminated her last opportunity for appropriate
treatment.
After 80 years of using probation and treatment in lieu of incarceration,
the British legal system has demonstrated that this method is as effective at
preventing or deterring infanticide as is incarceration, while being considerably
more efficient and cost-effective [7, 13].
The question then to ask ourselves is what we seek to gain by this punish-
ment and how can we prevent these needless tragedies in the future?
The fact that the insanity defense is nonexistent in some states and
extremely limited in others speaks to our disregard for mental illness and the
rights of those who suffer. The task for the psychiatric community is certain.
Until we treat mental illness with the same dignity afforded to other illnesses,
the course will remain unchanged.
Perinatal Suicide
One theme which has received even less attention in the epidemiological
literature is perinatal suicide. Seen against the backdrop of the general population,
the mortality rate of childbearing women from suicide is reportedly low [33].
However, measured against the backdrop of the perinatal population, the preva-
lence is considerably higher. In general, patients with affective disorders have
significantly higher rates of mortality than age- and gender-matched controls
[34]. In addition, the risk of suicide is highest in the first year after admission,
particularly within the few weeks of discharge. Similarly, when measured
against the mortality of new mothers with peripartum mental illness, the risk of
suicide is surprisingly high [35].
Eighty-eight percent of homicide-suicides are perpetrated by family mem-
bers. The victims of females tend to be their children. Premenopausal mothers
will tend to kill their young children before they kill themselves [36].
There appear to be 2 primary motives for parents who murder their chil-
dren and then commit suicide [37]. Some may kill on purpose or accidentally
then commit suicide as a result of despair. The second motive is usually a result
of suicide in a parent who feels that they cannot leave their children in a world
that is so harsh and merciless. These altruistic [38] filicides or mercy
killings are committed to relieve the real or imagined suffering of the children.
In a qualitative study of filicide by mentally ill mothers, Stanton et al. [39]
performed semi-structured interviews of 6 postpartum women with severe psy-
chiatric illness (major depression, schizoaffective disorder or schizophrenia).
Each described their motivation as altruistic or an extension of suicide. The
childrens ages ranged from a few weeks to 7 years. They described an intense
relationship with their children and little warning about their filicidal actions.
Motives were altruistic and driven by delusions. I thought that they were
going to use my daughter, that the devil would take her in a cot death and she
would be lost. This mother believed that if she herself killed her daughter, the
child could return her to the angels. These women rescued their children
from some awful fate that was indicated by their delusional system.
For those who described their killing in the context of their own suicide, a
recurrent theme related a sense of identification with the children. I did not
want to live and thought that if we could turn on the gas we could all go to
sleep and wake up in heaven. Another mother killed her infant because I cre-
ated her, because she was my responsibility.
In general, the filicides were ego-syntonic within the delusional system,
but clearly ego-dystonic once the woman had recovered.
A Case of Infanticide-Suicide
In August 2003, the headline in the Philadelphia Inquirer [40] said
Professor charged with murder in babys death. E., a tenured professor, distin-
guished academic and recognized expert on poverty was charged with murder
for killing her 6-month-old baby. She faced 40 years in an American prison.
E. had refused an amniocentesis at 20 weeks of gestation. Her infant
daughter was born with Downs syndrome, could not breast-feed and used a
feeding tube. E. felt exceedingly guilty. She believed that she was to blame for
her babys future life of suffering. She had been contemplating suicide for 2 or
3 weeks.
Within days of her incarceration, she attempted to hang herself in the
county jail. She was forced to wear Kevlar clothing because it cannot be tied
into a noose. Her mattress was placed in the dayroom, in direct sight of jail
deputies and other inmates at the detention center.
Her last hours were chronicled by jail deputies who had orders to check E.
every 30min. She was seen on her mattress in the dayroom for the 2 p.m. check.
At 2:45 p.m., a deputy noticed her feet moving under the Kevlar blanket. By
3:15, a deputy concluded that she was sleeping. But at 3:40, the deputy pulled
the blanket away to find her dead. She suffocated herself with a plastic garbage
bag in the presence of 12 inmates and jail deputies.
E.s attorney blamed law enforcement, which did not transfer E. to a
psychiatric hospital after her first suicide attempt. This case describes one
more tragedy triggered by the American judicial systems treatment of our
mentally ill.
Spinelli 96
The literature is scarce on this subject. Although the presence of children is
protective against parental suicide, it is greater with young and hospitalized
psychiatrically ill children [41]. Suicide risk decreases with younger age and
greater number of children, while parents of children who have died are at high
risk for suicide.
The risk of suicide in postpartum women is associated with severe mental
illness. Appleby et al. [35] investigated the risk of suicide in women with post-
partum illness by calculating standardized mortality ratio (SMRs) for patients
discharged from the psychiatric hospital within 1 year. They found that suicides
and deaths from unnatural causes were more likely to occur within the first year
after childbirth.
During the study (19731993), 1,567 women were admitted to a psychi-
atric hospital within 1 year of childbirth of whom 107 died (6.8%). There were
52 suicides (49% deaths, 3.3% of sample), of whom 14 (27% of suicides, 0.9%
of sample) died within 1 year of childbirth. There were 71 unnatural deaths
(66% of deaths, 4.5% of sample) of whom 19 (27% of unnatural deaths, 1.2%
of sample) died within 1 year of childbirth.
Although postpartum women seem to have a low rate of suicide, the rates
of natural and unnatural deaths increase within the first year in women with
severe postpartum illness. In summary, the rate of suicide increases 70-fold in
the first postpartum year and the overall long-term risk increases approximately
17-fold.
Forty-seven (3%) of the postpartum psychiatric admissions were associ-
ated with the death of an infant in the first year after birth, most occurring
before admission and likely to have precipitated the maternal illness.
Recent findings by Oates [42] identified suicide as the leading cause of
maternal death. Of those maternal deaths reported to the Confidential Enquiries
into Maternal Deaths (CEMD 2001), 12% were due to psychiatric causes and
10% to suicide. Although women are less likely to commit suicide in a violent
manner, most postpartum suicides died violently, mainly by hanging or jump-
ing. Only 3 died from overdose of medication.
Oates estimates the suicide rate for puerperal psychosis as 2/1,000 suffer-
ers and 0.3/1,000 for women referred to psychiatric service following child-
birth. The fact that 46% of suicides were in contact with psychiatric services
emphasizes the need for close supervision in this population of women.
Oates described the profile of a woman who is likely to commit suicide as
one who has had a previous experience of psychiatric hospitalization without
her baby, who suffers from a severe mental illness with an early onset following
childbirth, who is older and whose act of deliberate self-harm is violent.
Furthermore, no death occurred in a woman admitted at any time to a mother
and baby unit.
In an interview follow-up study [43], 54 postpartum mentally ill women
and 89 controls were followed for 6 years; the author describes the outcomes as
dismal. One third of these mothers did not live with their child compared to
only 3% of controls. Index mothers had more problematic relationships with
their children predisposed by the mothers own poor parenting. Readmission
rate was 46% for these mothers and the suicide rate was 4.2%. In 2003, 18 post-
partum suicides in American women were reported to Postpartum Support
International [pers. commun.].
Similar to infanticide, the prevalence of peripartum suicide is likely greater
than reported. Suicides are less likely reported in the media. Suicide and/or the
mothers postpartum status may not be reported on death certificates.
Conclusion
As a major public health problem, postpartum psychiatric illness is predic-
table, identifiable, treatable and therefore, most importantly, preventable [44].
The request for greater understanding is hardly new to the 21st century. It is a
long-standing plea for research and knowledge.
The potential benefit of a formal diagnosis for postpartum illness is greater
awareness, education and greater likelihood of early identification and treat-
ment. The risk of challenging the standard method of choosing criteria for diag-
nostic consideration must be weighed against the potential benefit to maternal/
infant health.
The great promise of prevention through education could play out in incal-
culable saved lives. We, as a world society, could do a far better job of prevent-
ing these tragedies.
References
1 Murray CJK, Lopez AD: The Global Burden of Disease. Boston, Harvard University Press, 1996,
p 651.
2 Overpeck MD, Brenner RA, Trumble AC, Trifilletti LB, Berendes HW: Risk factors for infant
homicide in the United States. N Engl J Med 1998;339:12111216.
3 McClain PW, Sacks JJ, Froehlke RG, Ewigman BG: Estimates of fatal child abuse and neglect,
United States, 1979 through 1988. Pediatrics 1993;91:338343.
4 Herman-Giddens ME, Brown G, Verbiest S, Carlson PJ, Hooten EG, Howell E, Butts JD:
Underascertainment of child abuse mortality in the United States. JAMA 1999;282:463467.
5 Overpeck MD: Epidemiology of Infanticide; in Spinelli M (ed): Infanticide: Psychosocial and
Legal Perspectives on Mothers Who Kill. Washington, American Psychiatric Press, 2002, pp 318.
6 Ewigman B, Kivlahan C, Land G: The Missouri child fatality study: Underreporting of maltreatment
fatalities among children younger than five years of age, 1983 through 1986. Pediatrics 1986;91:
330337.
Spinelli 98
7 Oberman M: Mothers who kill: Coming to terms with modern American infanticide. Am Crim
Law Rev 1996;34:1110.
8 Moseley KL: The history of infanticide in Western society. Issues Law Med 1986;1:346357.
9 Lagaipa SJ: Suffer the little children: The ancient practice of infanticide as a modern moral
dilemma. Issues Compr Pediatr Nurs 1990;13:241251.
10 Langer WL: Infanticide: A historical survey. Hist Child Q 1974;1:353365.
11 Brockington I: Motherhood and Mental Health. Oxford, Oxford University Press, 1996, pp 400449.
12 Linzer L: When the blues turn deadly, where is the help? New Zealand Press Association, August
13, 2001.
13 Marks MN: Infanticide in Britain; in Spinelli M (ed): Infanticide: Psychosocial and Legal
Perspectives on Mothers Who Kill. Washington, American Psychiatric Press, 2002, pp 185200.
14 Yardley J: Despair plagued mother held in childrens deaths. New York Times at A-07, September
8, 2001.
15 MNaghtens Case 8 Eng Rep 718 HL, 1843.
16 Spinelli M: Infanticide: Psychosocial and Legal Perspectives on Mothers Who Kill. Washington,
American Psychiatric Press, 2002, pp 6180.
17 Model Penal Code: 4.01(1) 1995.
18 McFarlane J: Criminal Defenses in the Cases of Infanticide and Neonaticide; in Spinelli M (ed):
Infanticide: Psychosocial and Legal Perspectives on Mothers Who Kill. Washington, American
Psychiatric Press, 2002, pp 133166.
19 Robinson PH: Criminal Law Defenses. Minneapolis, West Publishing Co, 1984, vol 12.
20 Meyer C, Spinelli MG: Medical and Legal Dilemmas of Postpartum Psychiatric Disorders; in
Spinelli M (ed): Infanticide: Psychosocial and Legal Perspectives on Mothers Who Kill.
Washington, American Psychiatric Press, 2002, pp 167184.
21 Meyer CL, Proano T, Franz J: Postpartum syndromes: Disparate treatment in the legal system; in
Muraskin R (ed): Its a Crime: Women and Justice. Englewood Cliffs, Prentice Hall, 1999,
pp 91104.
22 Infanticide Act of 1938 1 & 2 Geo 6, Ch 36 sec, 1938.
23 Watson P: Four years for disturbed mother who killed her baby. Press Association Limited, June
28, 2001.
24 Marks MN, Kumar R: Infanticide in England and Wales, 19821988. Med Sci Law 1993;33:
329339.
25 Deutsche Presse Agentur: Teenager who threw newborn to death in Hong Kong on probation.
January 4, 2003.
26 Woodard J: Experts grapple with reasons for infanticide. Calgary Herald, August 13, 2001.
27 Albrechtsen J: If a hand rocks the cradle, it must bear the blame. Sydney Morning Herald, August
28, 2001.
28 Sichel DA: Neurohormonal aspects of postpartum depression and psychosis; in Spinelli M (ed):
Infanticide: Psychosocial and Legal Perspectives on Mothers Who Kill. Washington, American
Psychiatric Press, 2002, pp 235256.
29 Wisner KL, Gracious BL, Piontek CM, Peindl K, Perel JM: Postpartum disorders: Phenomenology,
treatment approaches, and relationship to infanticide; in Spinelli M (ed): Infanticide: Psychosocial
and Legal Perspectives on Mothers Who Kill. Washington, American Psychiatric Press, 2002,
pp 3660.
30 APA Statement on the Insanity Defense and Mental Illness in response to the Andrea Yates case by
Richard K. Harding MD, President, American Psychiatric Association Release No 0208, March
2002.
31 CourtTV: Texas mom drowns kids. Available at http://www.courttv.com/trials/yates, March 2002.
32 Frye vs United States, 293 F 1013 (DC Cir 1923).
33 Marzuk PM, Tardiff K, Leon AC, Hirsch CS: Lower risk of suicide during pregnancy. Am J
Psychiatry 1997;154:122123.
34 Hoyer EH, Mortensen PB, Olesen AV: Mortality and causes of death in a total national sample of
patients with affective disorders admitted for the first time between 19731993. Br J Psychiatry
2000;176:7682.
35 Appleby L, Mortensen PB, Faragher EB: Suicide and other causes of mortality after postpartum
36 Barraclough B, Harris EC: Suicide preceded by murder: The epidemiology of homicide-suicide in
England and Wales 19881992. Psychol Med 2002;32:577584.
37 Collins Pl, Shaughnessy MF, Bradley L, Brown K: Filicide: Suicide in search of meaning. North
Am J Psychol 2001;3:277292.
38 Resnick PJ: Child murder by parents: A psychiatric review of filicide. Am J Psychiatry 1969;126:
325334.
39 Stanton J, Simpson A, Wouldes T: A qualitative study of filicide by mentally ill mothers. Child
Abuse Negl 2000;24:14511460.
40 Gray C: Professors last hours chronicled. Philadelphia Inquirer, August 2003.
41 Qin P, Mortensen PB: The impact of parental status on the risk of completed suicide. Arch Gen
Psychiatry 2003;60:797802.
42 Oates M: Suicide: The leading cause of maternal death. Br J Psychiatry 2003;183:279281.
43 Bagedahl-Strindlund M: Parapartum mental illness: An interview follow-up study. Acta Psychiatr
Scand 1992;95:389395.
44 Wisner KL, Peindl KS, Hanusa BH: Symptomatology of affective and psychotic illnesses related
to childbearing. J Affect Disord 1994;30:7787.
Margaret G. Spinelli, MD
Assistant Professor of Clinical Psychiatry
College of Physicians and Surgeons of Columbia University,
Director of Maternal Mental Health, The New York State Psychiatric Institute
PO Box 123, 1051 Riverside Drive, New York, NY 10032 (USA)
Tel. 1 212 543 5860, Fax 1 212 543 6700, E-Mail mgs8@columbia.edu
Relevance of Gonadal Hormones to
Perinatal Mood and Anxiety Disorders
Antti Ahokas
a
, Jutta Kaukoranta
a
, Kristian Wahlbeck
b
, Marjatta Aito
c
a
Mehilainen Clinic and University of Helsinki,
b
National Research and
Development Center for Welfare and Health, and
c
Adnex Research Clinic of
Gynecology, Helsinki, Finland
Epidemiological studies have repeatedly and consistently demonstrated
that there is a gender difference in depression, with a 2-fold greater prevalence
in women than in men [13]. The findings that the gender differences are man-
ifested during reproductive age suggest an important involvement of hormone-
related processes, especially the hypothalamo-pituitary-ovarian (HPO) axis and
hormonal changes in the pathophysiology of depression in women. Periods of
hormonal fluctuations or low estrogen levels, namely prior to the first ovulation
after menarche, premenstrually, postpartum and perimenopausally, have been
associated with increased vulnerability to depression and psychotic disorders
among susceptible women [410]. After our transition to the new millennium
with increasing life expectancy, including hormonal fluctuation during the
reproductive age and the use of low-dose oral contraceptive pills, women
are likely to live a substantial part of their life in a hypoestrogenic state.
Concomitantly with these demographic and physiologic changes, a wealth of
information derived from basic, clinical and epidemiological research has
increased our knowledge also of other areas such as cognitive disorders,
Alzheimers dementia, cardiovascular diseases, and osteoporosis, as major
factors affecting womens health and the role that estrogen deficiency plays in
contributing to those disorders.
Given the fact that postpartum depression is common, it still often remains
unrecognized, and thus, undiagnosed and undertreated. The incidence of
depression has been estimated from 10 to 16% during the 612 weeks after
delivery, and up to 22% during the first 6 months postpartum according to con-
trolled epidemiologic studies [1113]. Depression may also manifest with a
longer delay and have a long duration; a quarter of the untreated women is still
Gonadal Hormones 101
depressed at the childs first birthday [11]. Furthermore, postpartum depression
can be resistant to conventional psychiatric treatment interventions [6, 14]. The
postpartum period is typically described as a time of happiness and excitement.
However, the expectations of happiness in a young family can shipwreck into
deep depression and no less than 5% of women with severe postpartum depres-
sion may commit suicide [15, 16]. Instead of happiness, the postpartum period
is often experienced as a time of significant stress.
From a physiological point of view, hardly any other life event brings about
changes that rival the neuroendocrine and psychosocial changes associated with
pregnancy and childbirth. The drastic physiologic events occurring after deliv-
ery may offer a window for evaluating the effects of gonadal steroids on mood
and behavior. Hormonal changes, increasing hormonal levels during pregnancy
and abrupt postpartum withdrawal, and their multiple actions on the central ner-
vous system (CNS) may constitute a precipitating factor for depression in vul-
nerable women. Besides the obvious significance of gonadal steroids in the
pathophysiology of postpartum depression, the main circulating physiologic
estrogen in the female, 17-estradiol, may be an option in the treatment of this
condition.
Hormonal Physiology in Pregnancy and Postpartum
Levels of gonadal hormones, estrogens (estradiol, estriol, and estrone) and
progesterone, increase steadily during pregnancy. The biologically active forms
of estrogen, 17-estradiol and estriol, rise by 100-fold and 1,000-fold, respec-
tively [17]. At the end of pregnancy, serum estradiol levels are very high (about
100,000pmol/l), and this estradiol is mainly of placental origin [17]. With
removal of the placenta at delivery, serum estradiol concentration declines
sharply within a few days to 100 pmol/l or lower. The recovery of ovarian estra-
diol production can be slow, and thus, postpartum estradiol deficiency can be
prolonged and profound, mimicking menopause. The sharp drop of estradiol
concentration, however, differs from events in peri- and postmenopausal women,
in whom the concentrations change in a slowly declining manner allowing more
time for the brain to adapt to the changes [18].
Besides the elevations of the concentrations of gonadal steroids, levels of
other bioactive substances also progressively rise during pregnancy. Levels of
placental corticotrophin-releasing hormone, cortisol, human chorionic gonado-
tropin, prolactin, -endorphin and thyroid hormone-binding globulin rise
across pregnancy, reaching a maximum near term and declining at delivery
[17, 1922]. The levels of free thyroid hormones prior to delivery and postpartum
remain relatively constant [19, 20].
Ahokas/Kaukoranta/Wahlbeck/Aito 102
There is a complex interaction and feedback system between HPO and
hypothalamic-pituitary-adrenal (HPA) axis. Stress is known to affect the repro-
ductive axis, leading in the severest cases to amenorrhea or it may delay the
recovery of estradiol production postpartum. The HPA axis abnormalities seen
in depression suggest that activation of the HPA axis may be linked to inhibition
of the HPO axis [23]. The increased activity of the HPA axis during the third
trimester of human pregnancy may be a consequence of progressively increas-
ing circulating levels of placental corticotrophin-releasing hormone [21, 22].
The resetting of the HPA axis during the postpartum period has not yet been
studied in detail. A relatively high cortisol response has been reported, suggest-
ing that postpartum women have hypertrophic adrenal cortices [22, 24]. Stress-
like psychiatric symptoms (e.g. anxiety) may further trigger the activation of
the HPA axis. This may result in a postpartum vicious circle, in which stress-
like symptoms increase HPA activity inhibiting HPO function, which may fur-
ther trigger psychiatric symptoms in vulnerable women [22]. Thus, the HPA
axis exerts profound, mostly inhibitory effects on the reproductive axis inhibit-
ing hypothalamic gonadotropin hormone-releasing hormone, pituitary luteiniz-
ing hormone and estrogen production [22]. Furthermore, elevation of prolactin
and breast-feeding are known to inhibit HPO function [21, 22].
The hormonal levels in women with postpartum depression have not been
investigated systematically and extensively. Those few studies that exist have
not found significant differences in depressed and nondepressed women [20].
Thus, although quantitative hormonal levels do not appear to distinguish
women with postpartum depression from normals, their susceptibility to changes
or low levels of gonadal steroid concentration may differentiate them.
Effects of Gonadal Hormones on the Central Nervous System
Ovarian steroid hormones are neuroactive steroids and have a wide range
of actions on the regulation of mood, behavior and cognition. These actions
include modulation of the blood-brain barrier, glucose metabolism, neuronal
growth, neurotransmitter activity, intraneuronal signal transduction, gene
expression and multiple neuroprotection mechanisms [2527]. Owing to their
lipophilic nature, reproductive hormones easily cross the blood-brain barrier,
and the central levels of estradiol are reported to correlate with plasma levels
[28, 29]. Estrogens have their own receptors (ER) and (ER), or and
estradiol receptors, that act in the CNS in the cell nucleus and at the
neuronal membrane [2527]. Estradiol nuclear receptors belong to nuclear
ligand-activated transcription factors that act to turn genes on and off, and can
interact with neuronal membrane receptors, neurotransmitter transporters and
G-protein-linked receptors [25, 26]. ER appears to dominate in the hypothal-
amus and amygdala, indicating that the -subtype might modulate neuronal
cells involved in autonomic and reproductive neuroendocrine functions as well
as emotional interpretation and processing. The dominant areas of ER appear
to be hippocampal formation, entorhinal cortex and thalamus, suggesting a
putative role for ER in cognition, nonemotional memory and motor functions
[25, 30, 31].
The distribution of estradiol receptors in different brain regions, and their
distinct up- or downregulation by estradiol can contribute to the complexity of
the effects of estrogen on the CNS. The experimental paradigms point to two
general mechanisms of gonadal steroid hormones: a rapid, short-term non-
genomic membrane effect and a slower, long-term, possibly genomic effect on
neurotransmitter function [27, 31].
Considerable evidence from basic and clinical research demonstrates that
gonadal hormones, especially estradiol, can modulate neurotransmission at
multiple points in the serotonin (5-HT), norepinephrine [noradrenalin (NA)]
and dopamine pathways as well as influence monoamine oxidase (MAO)
activity. Estrogen modulation of the monoamine neurotransmitter system is a
possible mechanism in the link between gonadal hormones and depression,
considering that an imbalance in the neurotransmitter system is strongly
hypothesized in the etiology of depression. Estradiol exerts an agonistic effect
on 5-HT activity by increasing the number of serotonergic receptors, transport
and uptake of the neurotransmitter as well as synthesis of 5-HT [27, 3234].
Estradiol upregulates 5-HT
1
receptors and downregulates 5-HT
2
receptors, and
decreases significant MAO changes in the density of 5-HT
2
receptors in the
anterior frontal, and cingulate activity [27, 33]. In ovariectomized animals, an
acute drop in estrogen induces significant changes in the density of 5-HT
2
receptors in the anterior frontal, cingulate and primary cortex and in the nucleus
accumbens [35, 36] higher centers of the forebrain associated with the control
of mood, behavior and cognition. The changes could be restored towards
normal levels by estradiol treatment. The findings were later confirmed in
humans by positron emission tomography during the estradiol treatment of
postmenopausal women [37, 38]. Thus, recent neuroimaging studies confirm
that estrogen can modulate neurotransmitter function in the human brain.
Estradiol selectively increases the activity of NA in the brain. Increased
NA activity may be due to decreased reuptake of NA, and decreased metabo-
lism due to inhibition of MAO [27, 31, 33]. Studies further indicate that estra-
diol modulates the dopamine-2 receptors and dopamine transport in various
brain regions [39, 40]. Estradiol decreases dopamine-2 receptor sensitivity and
probably that of other dopaminergic receptors as well, and can inhibit dopamin-
ergic activity in various steps of dopamine transmission [27, 33, 41, 42].
In contrast to estrogen, progesterone appears to have an inhibitory effect
on most neuronal activities in the brain. Progesterone can increase MAO activ-
ity and thus counteract estrogens action of decreasing MAO activity [24, 25,
41]. With regard to neurotransmitter function and activity, progesterone has
mainly an opposite effect to estrogen [27, 43].
Conclusively, estrogen/estradiol has antidepressant and antipsychotic
properties, whereas progesterone exerts an opposite action to estrogen.
Theoretically, this could possibly also explain the manifestation of mood
changes called premenstrual dysphoric disorder in susceptible women at the
luteal phase, in which the level of estradiol is decreasing and progesterone
increasing, and perhaps progesterone is dominating.
Intervention Studies
Only limited clinical investigations have been conducted in depression and
the influence of disrupted hormonal homeostasis postpartum. Direct evidence
in support of the involvement of the reproductive hormones in the development
of postpartum depression in vulnerable women is provided by a double-blind
study, in which hormonal withdrawal was simulated [44]. In this study, the sup-
raphysiologic hormonal levels of pregnancy and withdrawal from these high
levels to a hypogonadal state were simulated by blocking estradiol production
with leuprolide in euthymic women with and without a history of postpartum
depression. Five of 8 (62.5%) women with a history of postpartum depression
and none in the comparison group developed significant mood symptoms.
Positive results have been reported from the treatment studies using differ-
ent estrogens and regimens, including physiologic 17-estradiol in transdermal
[45] and sublingual [6, 46, 47] form for postpartum depression, as well as high-
dose oral conjugated estrogen for the prevention of postpartum affective dis-
order [48]. The results of the double-blind, placebo-controlled study of 61 women
with severe postpartum depression treated with transdermal 17-estradiol
patches (200 g/day) confirmed the presence of an antidepressant effect com-
pared with placebo [45]. The onset of the effect was rapid and occurred within
the first month of treatment. The adverse effect profile during the 6-month
treatment period did not differ between the estradiol-treated and placebo
groups. A limitation in evaluating the estradiol effect is that in the actively
treated group nearly half the patients (47%) concomitantly used antidepressants
and the serum concentration of estradiol was measured only once, i.e. after
3 months of treatment.
The first published studies based on the measurement of estradiol concen-
tration at baseline and monitoring the concentration during the treatment with
sublingual physiologic 17-estradiol were presented in case series [46, 47] and
in an open-label study lasting 8 weeks of 23 patients with severe postpartum
depression [6]. Fourteen of the 23 patients had received psychiatric treatment
without an adequate effect before estradiol treatment. The pretreatment values
of serum estradiol were very low (mean 47.1 pmol/l, lower than the threshold
value for gonadal failure, i.e. 110 pmol/l), suggesting estradiol deficiency. The
estradiol dose was determined according to the serum concentration after a
weekly measurement. The mean daily dose varied from 3.9 to 4.8mg. During
the treatment, the serum level rose similarly in all patients and approached the
values of the follicular phase: mean 342 and 478 pmol/l at week 1 and 8, respec-
tively. The concurrent rise in serum concentration of estradiol coincided with
the decline in depressive symptoms (fig. 1). The estradiol level remained low in
2 patients, and in both cases the relief of the symptoms was minimal. The find-
ing suggests that there may be an individual threshold effect of the concentra-
tion of bioavailable estradiol and psychiatric symptoms. The observation that
depressive symptoms diminished rapidly may have important theoretical and
clinical implications, because there is a typical delay of about 2 weeks before
antidepressants begin to take effect.
The results from studies on women with puerperal psychosis show promis-
ing evidence of estrogens antipsychotic efficacy. In a prophylactic study of
7 women with histories of puerperal psychosis and 4 with histories of postpar-
tum major depression, only 1 experienced dysphoria again when high-dose oral
conjugated estradiol was administered immediately following childbirth [48].
None of the women had histories of nonpuerperal affective disorder, and they
0
10
20
30
40
50
0 1 2 3 4 8
Time (weeks)
M
A
D
R
S

t
o
t
a
l

s
c
o
r
e
s
0
100
200
300
400
500
600
700
800
C
o
n
c
e
n
t
r
a
t
i
o
n

o
f

e
s
t
r
a
d
i
o
l

(
p
m
o
l
/
l
)

MADRS S-E
2
Fig. 1. Serum estradiol (S-E
2
) concentration (mean SD) and Montgomery-sberg
Depression Rating Scale (MADRS) total scores during treatment with 17-estradiol [6].
were euthymic throughout the current pregnancy and delivery. In contrast, in a
series of 29 patients with preexisting bipolar disorder or schizoaffective disor-
der, treatment with transdermal estradiol did not reduce the rate of recurrence
[49]. These patients, however, represent rather the primary diagnosis than post-
partum disorder. In case series [50, 51] and an open-label study [52] of patients
with puerperal psychosis treated with sublingual physiologic 17-estradiol for
6 weeks, the treatment effect was positive and quite quickly achieved. Six of
10 patients had undergone psychiatric treatment without adequate effect prior
to estradiol treatment. The baseline serum estradiol values were very low (mean
49.5, range 1390 pmol/l), suggesting estradiol deficiency. By correcting the
documented estradiol deficiency with physiologic estradiol, the psychotic
symptoms declined.
The role of progesterone in the treatment or prevention of postpartum
depression has remained obscure, as can be expected theoretically. The results
from the study of 180 women given long-acting synthetic progesterone
(norethisterone) within 48 h of delivery show that the use of progesterone was
associated with an increased risk of developing postnatal depression [53].
The promising effect of estradiol in the treatment of postpartum depression
is in line with the reports from the treatment studies of estradiol for premen-
strual syndrome [54], premenstrual psychosis [55], peri/postmenopausal
depression [56, 57] and as an adjunct therapy for schizophrenia [58, 59].
Different Estrogens and Regimens
Various estrogen types are available in several forms and preparations.
Different types include natural, physiological 17-estradiol normally produced
by the ovary in women, artificial estrogens (e.g. ethinyl estradiol in oral contra-
ceptives) and various conjugated equine estrogens extracted from the urine of
pregnant mares [33]. They have a different nature, effects and adverse effect
profiles, and thus, in evaluating the effect and clinical practice, the generalizing
word estrogen should be used with caution, and not without specifying the
type of estrogen. Most data in the USA regarding the safety of estrogen therapy
have been gathered from women taking conjugated estrogens alone or com-
bined with progesterone [33, 60]. The results from studies of postmenopausal
women using 17-estradiol or oral conjugated estrogen have yielded different
results [33].
In Europe, the most commonly used estrogen is 17-estradiol, which is
available in several forms, including oral pills, micronized tablets, transdermal
patches and gels as well as intranasal spray [6163]. Significant differences in
the pharmacokinetic aspects of oral versus nonoral (e.g. transdermal, sublingual)
estrogens may contribute to the lack of consistency across studies regarding the
effect of estrogen treatment on mood [33, 60]. Oral estrogens are metabolized
through the hepatic portal system, with extensive conversion to the less active
estrone, whereas nonoral preparations avoid the first-pass metabolism and
result in higher physiologic circulation ratios of estradiol to estrone [61, 62].
The transdermal delivery (patches) provides sustained near-constant release
and nearly constant serum levels, but there is a noncompliance risk of poor
absorption throughout the skin in some patients [64]. Sublingual administration
of micronized resorbable estradiol tablets is easy and convenient, and avoids
noncompliance sometimes associated with the transdermal system. The rapid
but short duration of action via the sublingual route mimics the natural pulsatile
ovarian function, and thus may contribute by feedback to a faster recovery of
ones own ovarian function to produce estradiol [22, 65]. Several daily doses are
needed to maintain estradiol levels because of the rapid absorption, short half-
life of 17-estradiol, and quick steroid metabolism in young women [61].
By treating young women with physiologic estradiol during the postpartum
period, one of the most important issues is to measure the concentration of serum
estradiol at the beginning and during the treatment [6, 52]. The measurement of
the concentration is needed to ensure the absorption and to maintain the concen-
tration within physiologic limits compared with the values across the normal
menstrual cycle. The main problem in young women may be that the concentra-
tion does not reach the individual threshold value, but remains below the effec-
tive level [6], because young women have faster steroid metabolism than
postmenopausal women [66, 67].
Conclusions
Research on the etiology of postpartum psychiatric disorders, in particular
postpartum depression, has been characterized by many claims but few con-
firmed findings. A better understanding of the physiologic bases of psychiatric
symptoms postpartum may lead to the correction of the underlying pathology of
postpartum depression and puerperal psychosis rather than treatment of symp-
toms. Postpartum psychiatric disorders are serious illnesses with a potentially
long-lasting negative impact on the mother, the marital relationship and the
childs emotional, social and cognitive development. These disorders can also
be resistant or respond with a long delay to conventional psychiatric treatment
methods. Consequently, safe and rapidly effective therapies are needed that
have a specific etiologic relation to the disorder.
Accumulating data suggest intrinsic and complex interactions between
reproductive hormones and brain functioning. Existing evidence supports the
view that estrogen, or its absence, plays a significant role in the regulation of
mood, behavior and cognition, as well as in the pathophysiology of mood dis-
orders. Virtually no other life event brings changes for a woman that can compete
with the hormonal, neuroendocrine and psychological changes associated with
pregnancy and childbirth. The promising results of benefits of estradiol in treat-
ing postpartum depression and puerperal psychosis highlight the potential for
new, rapid and clinically effective hormone treatments for women vulnerable to
these disorders. When estradiol deficiency is documented together with psychi-
atric symptoms, hormonal intervention could be encouraged, without forgetting
psychotherapeutic and social support. However, further research and clinical
trials on the use of estradiol need to be performed before conclusive clinical
guidance can be given.
References
1 Rieger DA, Mayers JK, Kramer M, Robins LN, Blazer DG, Hough RL, Eaton WW, Locke BZ:
The NIMH epidemiological catchment area program: Historical context, major objectives, and
study population characteristics. Arch Gen Psychiatry 1984;41:934941.
2 Kessler RC, McGonagle KA, Zhao S, Nelson CB, Hughes M, Eshleman S, Witchen HU, Kendler
KS: Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in United States. Arch
Gen Psychiatry 1994;51:819.
3 Leibenluft E: Gender Differences in Mood and Anxiety. Reviews of Psychiatry. Washington,
American Psychiatric Association Press, 1999, vol 18.
4 Riecher-Rssler A, Hfner H, Dutsch-Strobel A, Stumbaum M: Gonadal function and its influence
on psychopathology: A comparison of schizophrenic and non-schizophrenic female inpatients. Arch
Women Ment Health 1998;1:1526.
5 Young EA, Midgley AR, Carlson NE, Brown MB: Alteration in the hypothalamic-pituitary-ovarian
axis in depressed women. Arch Gen Psychiatry 2000;57:11571162.
6 Ahokas A, Kaukoranta J, Wahlbeck K, Aito M: Estrogen deficiency in severe postpartum depres-
sion: Successful treatment with sublingual physiologic 17--estradiol: A preliminary study. J Clin
Psychiatry 2001;62:332336.
7 Wisner KL, Parry BL, Piontek CM: Postpartum depression. N Engl J Med 2002;347:194199.
8 Hfner H: Gender differences in schizophrenia. Psychoneuroimmunology 2003;28:1754.
9 Halbreich U: The etiology, biology, and evolving pathology of premenstrual syndromes. Psycho-
neuroimmunology 2003;28:5599.
10 Riecher-Rssler A: Oestrogens and schizophrenia. Curr Opin Psychiatry 2003;16:187192.
11 Cox JL, Murray D, Chapman G: A controlled study of the onset, duration and prevalence of postnatal
depression. Br J Psychiatry 1993;163:2731.
12 OHara MW, Swain AM: Rates and risk of postpartum depression: A meta-analysis. Int Rev
13 Llewellyn AM, Stowe ZN, Nemeroff CB: Depression during pregnancy and postpartum. J Clin
Psychiatry 1997;58(Suppl 15):2662.
14 Stuart S, OHara MW, Blehar HC: Mental disorders associated with childbearing: Report of the
biennial meeting of the Marc Society. Psychopharmacol Bull 1998;34:333338.
15 Appleby L, Mortensen PB, Faraghner EB: Suicide and other causes of mortality after post-partum
16 Georgiopoulos AM, Bryan TL, Yawn BP, Houston MS, Rummans TA, Therneau TM: Population-
based screening for postpartum depression. Obstet Gynecol 1999;93:653657.
17 Willcox DL, Yovich JL, McColm SC, Phillips JM: Progesterone, cortisol and oestradiol-17 beta
in the initiation of human parturition: Partitioning between free and bound hormone in plasma.
Br J Obstet Gynaecol 1985;92:6571.
18 Teede H, Burger H: The menopausal transition; in Studd J (ed): The Management of Menopause
Annual Book Review. London, Parthenon, 1998, pp 112.
19 Learoyd DL, Fung HYM, McGregor AM: Postpartum thyroid dysfunction. Thyroid 1992;2:7380.
20 Hendrick V, Altshuler L, Suri R: Hormonal changes in the postpartum and implications for post-
partum depression. Psychosomatics 1998;39:93101.
21 Taylor RN, Martin MC: The endocrinology of pregnancy; in Greenspan FC, Strewler GJ (eds):
Basic and Clinical Endocrinology, ed 5. Stamford, Appleton & Lange, 1997, pp 548574.
22 Magiakou MA, Mastorakos G, Webster E, Chrousos GP: The hypothalamic-pituitary-adrenal axis
and female reproductive system. Ann NY Acad Sci 1997;816:4256.
23 Yong EA, Midgley AR, Carlson NE, Brown M: Alteration in the hypothalamic-pituitary-ovarian
axis in depressed women. Arch Gen Psychiatry 2000;57:11571162.
24 Magiakou MA, Mastorakos G, Rabin D, Bubbert B, Gold PW, Chrousos GP: Hypothalamic CRH
suppression during the postpartum period: Implications for the increase of psychiatric manifesta-
tions in this period. J Clin Endocrinol Metab 1996;81:19121917.
25 McEwen BS, Alves SE: Estrogen actions in the central nervous system. Endocr Rev 1999;20:
279307.
26 Rupprect E, Holsboer F: Neuroactive steroids: Mechanisms of action and neuropsychophramaco-
logical perspectives. Trends Neurosci 1999;22:410416.
27 McEwen BS: The molecular and neuroanatomical basis for estrogen effects in the central nervous
system. J Clin Endocrinol Metab 1999;84:17901797.
28 Backstrom T, Carstensen H, Sodergard R: Concentration of estradiol, testosterone and progesterone
in the cerebrospinal fluid compared to plasma unbound and total concentrations. J Steroid Biochem
1976;7:469472.
29 Schwarz S, Pohl P: Steroid hormones and steroid hormone binding globulins in cerebrospinal
fluid studied in individuals with intact and with disturbed blood-cerebrospinal fluid barrier.
Neuroendocrinology 1992;55:174182.
30 Osterlund MK, Hurds YL: Estrogen receptors in the human forebrain and the relation to
neuropsychiatric disorders. Prog Neurobiol 2001;64:251267.
31 Rupprecht R: Neuroactive steroids: Mechanisms of action and neuropsychopharmacological
properties. Psychoneuroimmunology 2003;28:139168.
32 Rubinow DR, Schmidt PJ, Roca CA: Estrogen-serotonin interactions: Implications for affective
regulation. Biol Psychiatry 1998;44:839850.
33 Halbreich U, Kahn LS: Role of estrogen in the aetiology and treatment of mood disorders. CNS
Drugs 2001;15:797817.
34 Stahl S: Effects of estrogen on the central nervous system. J Clin Psychiatry 2001;62:317318.
35 Fink G, Sumner BE, McQueen JK, Wilson H, Rosie R: Sex steroids control mood, mental state
and memory. Clin Exp Pharmacol Physiol 1998;25:764775.
36 Cyr M, Bosse R, Paolo T: Gonadal hormones modulate 5-hydroxytryptamine2A receptors:
Emphasis on the rat frontal cortex. Neuroscience 1998;83:829836.
37 Moses EL, Drevets WC, Smith G, Mathis CA, Kalro BN, Butters MA, Leondires MP, Greer PJ,
Lopresti B, Loucks TL, Berga SL: Effects of estradiol and progesterone administration on human
serotonin 2A receptor binding: A PET study. Biol Psychiatry 2000;48:854860.
38 Kugaya A, Epperson CN, Zoghbi S, van Dyck CH, Huu Y, Fujita M, Staley JK, Garg PK, Seibyl JP,
Innis RB: Increase in prefrontal cortex serotonin 2A receptors following estrogen treatment in
postmenopausal women. Am J Psychiatry 2003;160:15221524.
39 Di Paolo T: Modulation of brain dopamine transmission by sex steroids. Rev Neurosci 1994;5:2741.
40 Bosse R, Rivest R, Di Paolo T: Ovariectomy and estradiol treatment affect dopamine transporter
and its gene expression in the rat brain. Brain Res Mol Brain Res 1997;46:343346.
41 Wieck A, Kumar R, Hirst AD, Marks MN, Campbell IC, Checkley SA: Increased sensitivity
of dopamine receptors and recurrence of affective psychosis after childbirth. BMJ 1991;303:
613616.
42 Cyr M, Calon F, Morisette M, Grandbois M, Fi Paolo T, Callier S: Drugs with estrogen-like
potency and brain activity: Potential therapeutic application for the CNS. Curr Pharm Des 2000;6:
12871312.
43 Arpels JC: The female brain hypoestrogenic continuum from premenstrual syndrome to
menopause. A hypothesis and review of supporting data. J Reprod Med 1996;41:633639.
44 Bloch M, Schmidt PJ, Danacceau M, Murphy J, Nieman L, Rubinow DR: Effects of
gonadal steroids in women with a history of postpartum depression. Biol Psychiatry 2000;157:
924930.
45 Gregoire AJP, Kumar R, Everitt B, Henderson AF, Studd JWW: Transdermal oestrogen for treat-
ment of severe postnatal depression. Lancet 1996;347:939933.
46 Ahokas A, Turtiainen S, Aito M: Sublingual oestrogen treatment of postnatal depression. Lancet
1998;351:109.
47 Ahokas A, Kaukoranta J, Aito M: Effect of oestradiol on postpartum depression. Psycho-
pharmacology 1999;146:108110.
48 Sichel DA, Cohen LS, Robertson LM, Ruttenberg A, Rosenbaum JF: Prophylactic estrogen in
recurrent postpartum affective disorder. Biol Psychiatry 1995;38:814818.
49 Kumar C, Melvor RJ, Davies T, Brown N, Papadopoulos A, Wieck A, Checkley SA, Campbell IC,
Marks MN: Estrogen administration does not reduce the rate of recurrence of affective psychosis
after childbirth. J Clin Psychiatry 2003;64:112118.
50 Ahokas A, Aito M: Role of estradiol in puerperal psychosis. Psychopharmacology 1999;147:
108110.
51 Ahokas A, Aito M, Turtiainen S: Association between oestradiol and puerperal psychosis. Acta
Psychiatr Scand 2000;101:167170.
52 Ahokas A, Aito M, Rimn R: Positive treatment effect of estradiol in postpartum psychosis:
A pilot study. J Clin Psychiatry 2000;61:166169.
53 Lawrie TA, Hofmyer GJ, De Jager M, Berk M, Paiker J, Viljonen E: A double-blind randomized
placebo-controlled trial of postnatal norethisterone enanthate: The effect on postnatal depression
and serum hormones. Br J Obstet Gynaecol 1998;105:10821090.
54 Smith RN, Studd JW, Zamblera D, Holland EF: A randomised comparison over 8 months of
100 micrograms and 200 micrograms twice weekly doses of transdermal oestradiol in the treatment
of severe premenstrual syndrome. Br J Obstet Gynaecol 1995;102:475484.
55 Korhonen S, Saarijrvi S, Aito M: Successful oestradiol treatment of psychotic syndrome in the
premenstrual phase: A case report. Acta Psychiatr Scand 1995:92;237238.
56 Schmidt PJ, Nieman L, Danaceau MA, Tobin MB, Roca CA, Murphy JH, Rubinow DR: Estrogen
replacement in perimenopause-related depression: A preliminary report. Am J Obstet Gynecol
2000;183:414420.
57 Soares CN, Almeida OP, Joffe H, Cohen LS: Efficacy of estradiol for the treatment of depressive
disorders in perimenopausal women: A double-blind, randomized, placebo-controlled trial. Arch
58 Kulkarni J, Riedel A, de Castella AR, Fitzgerald PB, Rolfe TJ, Taffe J, Burger H: Estrogen:
A potential treatment for schizophrenia. Schizophr Res 2001;48:137144.
59 Riecher-Rssler A: Oestrogen effects in schizophrenia and their potential therapeutic implications.
Arch Women Ment Health 2002;5:111118.
60 Li CI, Malone KE, Porter PL, Weiss NS, Tang MC, Cushing-Haugen KL, Daling JR: Relationship
between long duration and different regimens of hormone therapy and risk of breast cancer. JAMA
2003;289:32543263.
61 Price TM, Blauer KL, Hansen M, Stanxzyk F, Lobo R, Bates GW: Single-dose pharmacokinetics
of sublingual versus oral administration of micronized 17- estradiol. Obstet Gynecol 1997;89:
340345.
62 Frase I, Wang Y: New delivery systems for hormone replacement therapy; in Studd J (ed): The
Management of Menopause Annual Review. London, Parthenon, 1998, pp 101110.
63 Studd J, Pornel B, Marton I, Bringer J, Varin C, Tsouderos T, Christiansen C: Efficacy and accept-
ability of intranasal 17 -oestradiol for menopausal symptoms: Randomised dose-response study.
Lancet 1999;353:15741578.
64 Hall G, Spector T: The use of estrogen replacement as an adjunct therapy in rheumatoid arthritis;
in Berg G, Hammer M (eds): Modern Management of Menopause. London, Parthenon, 1994,
pp 369375.
65 Rossmanith WG, Laughlin DA, Mortola JF, Yen SS: Secretory dynamics of oestradiol (E2) and
progesterone (P4) during periods of relative pituitary LH quiescence in the midluteal phase of
menstrual cycle. Clin Endocrinol (Oxf ) 1990;32:1323.
66 Zumoff B, Fishman J, Cassouto JU, Gallagher TF, Hellman L: Influence of age and sex on normal
estradiol metabolism. J Clin Endocrinol Metab 1968;28:937941.
67 Hartman BW, Kirchengast S, Albrecht A, Soregy C, Knogler W, Huber JC: Absorption of orally
supplied natural estrogens correlated with age and somatometric parameters. Gynecol Endocrinol
1994;8:101107.
Antti Ahokas, MD, PhD, Ass. Prof.
Merikannontie 3 A 14, FIN00260 Helsinki (Finland)
Tel. 358 50 3699850, Fax 358 9 43191550, E-Mail antti.ahokas@kolumbus.fi
Pharmacotherapy for Psychiatric
Disorders in Pregnancy
Lori E. Ross, Shiny Gunasekera, Mary Rowland, Meir Steiner
Womens Health Concerns Clinic, St. Josephs Healthcare, Hamilton, Canada
Historically, pregnancy was thought to be a time of protection from
psychiatric disorders [1], but the available research has since demonstrated that
rates of mood disorders are approximately equivalent in pregnant and non-
childbearing populations [2, 3]. While there is insufficient research evidence to
allow for prevalence estimates of anxiety and psychotic disorders during preg-
nancy, these disorders are also commonly observed in clinical settings.
Many have hypothesized a role for pregnancy-related changes in sex steroid
hormone concentrations in prenatal psychiatric disorders, though to date, no
consistent relationships between pregnancy-specific variables and psychiatric
symptoms or syndromes have been established. As such, there is little evidence
to suggest that the etiology of, or response to treatment for, psychiatric disorders
will be substantially different during pregnancy relative to the non-pregnant
state. However, very few studies have systematically examined the efficacy of
treatment for perinatal psychiatric disorders: virtually all studies of prenatal use
of psychotropic drugs have focused on infant, rather than maternal, outcomes.
Treatment of any medical condition, including a psychiatric disorder,
during pregnancy, is complicated by concerns about the safety of the fetus.
Nearly all drugs, including psychotropic medications, pass through the placenta
into the fetal circulation [4]. Therefore, there is concern that in utero exposure
could result in complications to the fetal development and neonatal adaptation,
or in long-term neurobehavioural sequelae. What is known about the risks to
the fetus due to administration of psychotropic agents, therefore, is an impor-
tant component of treatment decision making with pregnant patients. To aid in
these decisions, this chapter will first review the medical literature on potential
teratogenic effects and adverse neonatal outcomes associated with prenatal
exposure to antidepressant, anxiolytic, mood-stabilizing and antipsychotic
Pharmacotherapy in Pregnancy 113
medications. We close with a review of the literature on neurobehavioural
development of children exposed to psychotropic medications in utero.
Antidepressant Medications
Selective Serotonin Reuptake Inhibitors
Presently, selective serotonin reuptake inhibitors (SSRIs) are considered
the first line of treatment for perinatal depression/anxiety of sufficient severity
for pharmacotherapy [5]. The data that have been collected with respect to
safety of the use of SSRIs during pregnancy are largely reassuring.
Teratogenic Effects
Of 5 controlled studies and a number of uncontrolled trials, there are no
reports of an increased risk for major malformations in SSRI-exposed infants
[610]. This finding is supported by the results of 2 meta-analyses [11, 12] and a
recent chart review of 138 non-smoking women [13]. One study found an
increased risk for minor malformations [6]; however, this study raises several
methodological questions: data for minor malformations are reported for less than
half of the exposed infants and no details are provided as to how infants were cho-
sen for examination, leaving open the possibility of selection bias. In addition,
minor malformation rates were not adjusted for potentially confounding variables,
including the use of other psychotropic medications. In fact, when infants also
exposed to benzodiazepines (BZs) were excluded from the analysis, the difference
between the groups was no longer statistically significant. No differences in the
frequency of minor malformations have been reported in other studies [10, 14].
Birth Outcomes
Exposure to SSRIs has been associated with decreases in gestational age
and/or increased risk for premature birth in some [6, 7, 10, 15] but not all [9, 13]
studies. Shorter gestational duration was observed after the use of either SSRI
or non-SSRI antidepressants [15]. However, in another study of infants with
prenatal exposure to either SSRIs or tricyclic antidepressants (TCAs), only
SSRI, but not TCA exposure, was associated with a 0.9-week decrease in gesta-
tional age, a 175-gram decrease in mean birth weight, and a decrease of 0.29 in
the 5-min Apgar score. The differences in birth weight and Apgar scores were
no longer significant when gestational age was controlled for [10]. These
authors note that while SSRI exposure was associated with a twofold increase
in premature delivery, the absolute risk for this adverse outcome was still only
10% and may therefore be outweighed by the potential benefits of SSRI use.
Ross/Gunasekera/Rowland/Steiner 114
With respect to the relationship between birth weight and prenatal SSRI
exposure, a recent chart review of 138 non-smoking SSRI users revealed 4 cases
of low birth weight (2.9%), all involving infants who had been exposed to fluox-
etine throughout pregnancy [13]. However, the authors note that this rate was
lower than the national low-birth-weight rate of 7.6% (although the national
sample probably included a substantial proportion of cigarette smokers, which
would be expected to result in a higher proportion of low-birth-weight babies).
In another study of 969 babies exposed to various antidepressants, an increased
rate of low-birth-weight infants was observed in the group exposed to antide-
pressants [15]. However, this difference was no longer statistically significant
after stratification for maternal age, parity and smoking. It is notable that in the
study by Chambers et al. [6], which did report increased rates of low birth weight
in exposed infants, the results were not controlled for these potential confound-
ing variables, nor for the use of other psychotropic medications.
Risk for Spontaneous Abortion
There have recently been some questions whether fetal exposure to SSRIs
may result in an increased risk for spontaneous miscarriage [7, 9]. However,
differences between exposed and non-exposed groups in the rates of sponta-
neous abortion do not attain statistical significance and do not seem to be
specific to any particular drug or even drug class: similar findings have been
reported in women using other antidepressant agents, including venlafaxine
[16], trazodone and nefazodone [17]. This lack of specificity has led to the
suggestion that it may be depression itself, rather than the antidepressant, that is
responsible for any increased risk for early pregnancy loss [17]. Additional
research is required to clarify this issue; however, clinicians can be reassured
that the rate of spontaneous abortion in women with prenatal SSRI exposure
remains within the expected range for the general population [7, 9].
Neonatal Complications
Recently, a primary concern regarding the use of SSRIs during pregnancy
has been potential effects of third-trimester SSRI exposure on neonatal adapta-
tion. Symptoms of neonatal irritability, crying, shivering, increased tonus,
respiratory distress and in rare cases, convulsions, have been associated with
exposure to SSRIs at term. There has been a debate, however, as to whether
these symptoms can be attributed to neonatal withdrawal from SSRIs or to sero-
tonergic overstimulation [18].
In most reports, adverse neonatal adaptation has been described as a neona-
tal withdrawal syndrome. Earliest descriptions appear in case reports, including
a case report of a neonate who developed symptoms of agitation and restlessness
on the second postpartum day after third-trimester exposure to sertraline, described
by the authors as potential sertraline withdrawal [19]. A further 2 case reports of
jitteriness, hypertonia and/or respiratory distress after third-trimester exposure to
fluoxetine were also published [20, 21]. The largest report of adverse neonatal
adaptation found that infants with third-trimester fluoxetine exposure had higher
rates of poor neonatal adaptation (including respiratory distress and jitteriness)
than did infants with first- and/or second-trimester exposure after adjustment for
a number of potentially confounding variables [6]. However, as described above,
neonatal adaptation data were available for only a subset of the infants in this
sample.
Case series and chart reviews have also reported on neonatal symptoms in
infants with late-pregnancy SSRI exposure. In one chart review of 64 mother-
infant pairs with prenatal exposure to fluoxetine, there was a threefold increase
in the frequency of newborn complications (3 of 17 infants with neonatal
complications were reported to be jittery, the other infants had non-specific
symptoms) and a twofold increase in the frequency of special care nursery
admissions for infants with third-trimester exposure, relative to infants with
exposure in the first and/or second trimester of pregnancy. However, these
effects did not reach statistical significance, perhaps due to the limited sample
size. Similarly, a trend towards a longer duration of fluoxetine exposure among
infants admitted to the special care nursery was observed, although again, this
finding did not attain statistical significance [22].
A large case series review reported data from the worldwide fluoxetine
pregnancy registry. One hundred and twelve prospectively identified pregnan-
cies with third-trimester fluoxetine exposure were reported to Eli Lilly and it
was found that 15 (13%) were associated with postnatal complications unre-
lated to malformations. In all but 1 case, the symptoms were mild and transi-
tory. However, since there was no consistent pattern of symptoms (e.g.
jitteriness reported in 2 infants, irritability in 4 infants, hyperbilirubinaemia in
4 infants and other non-specific symptoms in the remaining infants), and since
the rate was similar to the frequency of complications reported in surveys of
community samples, the author concluded that the complications were unlikely
to be attributable to fluoxetine. It is notable, however, that these data are based
on the reporters assessment of outcome and whether the assessment of the
infant in each case is thorough and unbiased is unknown [23].
Another chart review of 138 infants of non-smoking mothers exposed to
SSRIs found a number of transient, non-specific neonatal complications, but
concluded that these occurred at expected rates [13]. A recent report of 5 cases
of neonates with third-trimester exposure to paroxetine, citalopram or fluoxe-
tine observed symptoms of irritability, crying, shivering, increased tonus, and
in 1 case, convulsions. Four of these infants were treated with chlorpromazine
(CPZ) to resolve the symptoms, and in 1, symptoms persisted for up to 4 weeks
after birth [24]. Finally, a study of 17 SSRI-exposed full-birth-weight newborns
found significantly more tremulousness in exposed infants aged between
14 and 39 h when studied for 1 h than in a non-exposed control group; however,
the degree to which these groups are comparable on the basis of maternal use of
other drugs, including marijuana and BZs, is debatable [25].
Concern has been expressed about third-trimester exposure to paroxetine
in particular, since there are reports of serious discontinuation symptoms asso-
ciated with adult use of this SSRI [26]. A search of the medical literature
revealed a case report of paroxetine withdrawal syndrome in a neonate with
third-trimester exposure: at 12 h of age, this infant developed an increased res-
piratory rate and jitteriness, which was later followed by increased muscle tone
and tremor. Symptoms gradually decreased without treatment over the third and
fourth days postpartum [27]. A further 4 cases of infants with irritability, jitter-
iness and other symptoms which were considered to be associated with mater-
nal use of paroxetine during the third trimester of pregnancy were reported [28].
However, 2 of these infants had increased serum paroxetine concentrations,
raising the possibility of toxicity as opposed to withdrawal (see discussion
below). A further study of this same group of 55 mother-infant pairs with third-
trimester exposure to paroxetine found that complications necessitating treat-
ment and prolonged hospitalization were reported for 12 infants (22%) [29].
The primary symptom in these infants was respiratory distress, and symptoms
were alleviated within 12 weeks. It should be noted that unlike other studies of
this group [9], data were based solely on maternal reports and were not corrob-
orated by medical records, and therefore, these findings await replication.
A large-scale prospective investigation of neonatal outcomes in 997 infants
exposed in utero to a variety of antidepressants found that an increased risk for
preterm birth, low birth weight, low Apgar score, respiratory distress, convul-
sions and hypoglycaemia was noted especially after exposure to tricyclic drugs.
Most effects did not seem to be SSRI specific and outcomes after exposure to
paroxetine were not worse than after exposure to other SSRIs [30].
The existence of a neonatal serotonin withdrawal syndrome has recently
been disputed, and instead it has been suggested that symptoms of jitteriness,
irritability and/or respiratory distress represent SSRI toxicity [31]. This study
reports on 20 mother-infant pairs with third-trimester exposure to either citalo-
pram or fluoxetine. In comparison with 20 matched controls without psychotropic
exposure, the exposed neonates had a fourfold increase in the serotonergic
symptom score during the first 4 days of life, with the most prominent symptoms
being tremor, restlessness and rigidity. The difference in serotonergic symptoms
between the exposed and control groups was only apparent at days 14 postpar-
tum; there was no difference between groups at either 2 weeks or 2 months post-
partum. None of the exposed infants required treatment for these symptoms.
When the exposed group was subdivided by the SSRI used, the difference
between exposed and control groups in serotonergic symptoms was found to be
largely attributable to the infants exposed to fluoxetine. What makes these data
particularly notable is that the difference in serotonergic symptoms between
SSRI-exposed and control groups was associated with significant decreases in
umbilical vein whole blood serotonin and cord blood 5-hydroxyindoleacetic acid
concentrations. Further, an inverse correlation was found between the serotoner-
gic symptom score and venous umbilical 5-hydroxyindoleacetic acid concentra-
tions, providing a physiological rationale for these symptoms [31].
While these findings require replication, the association of serotonergic
symptoms with physiological markers of increased central nervous system sero-
tonin activity suggests that adverse effects in neonates with third-trimester SSRI
exposure may be the result of serotonergic overstimulation, rather than serotonin
withdrawal. It should be noted, however, that the serotonergic symptoms that
have been reported in infants with late-pregnancy SSRI exposure generally
resolve quickly and do not usually require intervention. As the bulk of these data
is drawn from case reports and case series, the proportion of exposed infants
affected by this neonatal withdrawal/discontinuation syndrome cannot be
reliably estimated. As described below, similar symptoms have been described in
infants with late-pregnancy exposure to TCAs. Prospective studies are required
to determine whether neonates who develop these symptoms are at increased
risk for additional exposure-related complications later in life. Although gradual
discontinuation of maternal antidepressant treatment in the last weeks of preg-
nancy has been proposed to reduce the likelihood of neonatal complications
[31], this regimen could predispose mothers to postpartum relapse, which would
also be associated with potential adverse effects for the infant.
The balance of evidence clearly indicates that discontinuation of clinically
needed antidepressants in women near term is unwarranted and may put the
mother at an unjustified perinatal risk. The neonatal discontinuation syndrome
occurs in a minority of cases and is self-limited. However, it is extremely
important to observe the infants of mothers taking antidepressants at term for
more than the typical 12 days to be able to recognize and treat, if necessary, the
discontinuation syndrome, when it occurs [18].
Tricyclic Antidepressants
Teratogenic Effects
A meta-analysis of 14 studies performed in 1996 investigating potential
effects of TCA exposure during pregnancy detected no significant effects on rates
of congenital malformations. This analysis included over 300,000 births [32].
Furthermore, a report using data from the European Network of Teratology
Information Services did not find higher than expected rates of congenital
malformations in 109 infants exposed to TCA monotherapy and 174 infants
exposed to TCAs plus other, non-antidepressant drugs [14]. More recently, a
chart review of 209 infants exposed to TCAs (as determined by maternal
prescriptions filled during the 270 days prior to delivery) found no effects on
congenital malformations or neonatal outcomes (including gestational age,
birth weight, and Apgar scores) relative to a group of unexposed matched
control infants [10].
There are reports of a transient TCA withdrawal syndrome in neonates
exposed during pregnancy: symptoms are reported to include jitteriness, irri-
tability and convulsions [3335]. These reports appear to be based on data
solely from case reports, so the prevalence of the TCA withdrawal syndrome
among all infants with late-pregnancy TCA exposure is unknown, nor is it clear
whether these neonatal symptoms have any lasting clinical significance. However,
as already mentioned above, a recent prospective study which included 395
mothers who used tricyclic drugs (mostly clomipramine) and 558 who used
SSRIs has identified an increased risk of poorer birth outcomes and of with-
drawal symptoms in the neonates notably after exposure to TCAs, suggesting
that SSRIs may be the drugs of choice during pregnancy [30].
Monoamine Oxidase Inhibitors
As a result of the development of safer and more convenient antidepressant
agents, monoamine oxidase inhibitors (MAOIs) are now seldom used for the
treatment of depression. Extensive interactions with food and other medica-
tions, together with their tendency to exacerbate hypertension, make these drugs
problematic for many patients, but in particular for pregnant populations [36].
For patients who do not respond to, or cannot tolerate, other classes of anti-
depressants, clinicians may wish to consider the MAOIs; however, there are few
data to suggest whether these agents can be safely used. An early study of 21
mother-infant pairs exposed to various MAOIs reported a relative risk of 3.4 for
congenital malformations [37]; however, this report provided little data with
respect to both the exposure and the malformations in question and so remains
difficult to interpret. Two cases of phenelzine exposure throughout pregnancy
have been reported with no congenital malformations in either case [38, 39].
Management of analgesia and anaesthesia during labour and delivery must be
carefully considered in women using MAOIs due to potential drug interactions
with opioid narcotics; successful use of epidural anaesthesia with bupivacaine
has been reported [38].
In the context of these limited data, together with concerns about dietary
restrictions and the potential risk for hypertension associated with MAOI use,
these agents are recommended for pregnant patients only after all other treat-
ment options have been exhausted.
Other Antidepressant Medications
There are limited data with respect to the safety profiles of other classes of
antidepressant medications for use during pregnancy. One study has reported
on first-trimester exposure to venlafaxine. One hundred and fifty women
exposed to venlafaxine were compared to women with first-trimester exposure
to other SSRIs and to women with exposure to non-teratogenic drugs. There
were no statistically significant differences between groups in the frequency of
major malformations, spontaneous and therapeutic abortions, gestational age or
birth weight as assessed by maternal report [16].
A report by the same group has examined the safety of trazodone and nefa-
zodone during pregnancy. One hundred and forty-seven women with first-
trimester exposure (58 to trazaodone, 89 to nefazodone) were assessed by
telephone at approximately 6 months postpartum and outcomes were compared
to previously published data on women with exposure to another SSRI or to a
non-teratogenic drug. There were no statistically significant differences in
major malformations, spontaneous or therapeutic abortions, gestational age or
birth weight between the three groups [17]. Further prospective research is
required to confirm these promising results regarding safety of the newer anti-
depressants.
Clinical Management
Any risk associated with treatment must be weighed against the known
risks associated with untreated disease, particularly perinatal major depression.
Risk-benefit decision making is best performed by the informed patient in con-
sultation with a responsible health care provider on a case-by-case basis.
When weighing the relative risk, one has to consider the diagnosis, the indi-
cation, the potential consequences of not treating and the safety of available
treatments, i.e. teratogenic risk and neonatal discontinuation/withdrawal risk. Up
to 70% of pregnant women endorse depressive symptoms, with up to 16% ful-
filling criteria for major depressive disorder [5]. The diagnosis is hindered due to
the overlap of symptoms between depression and normal effects of pregnancy,
the limited assessment of comorbid medical conditions and the denial/reluctance
to seek help. The indication to treat must always be 100%, but accepting the
treatment should be a womans choice if at all possible. The pregnant woman and
her partner should also be made aware of the risks of not treating depression.
Controlled clinical studies of the effects of untreated prenatal stress and depres-
sion report a long list including a higher rate of both acute and developmental
consequences of miscarriages, preterm deliveries, obstetrical complications,
pre-eclampsia, admissions to the Neonatal Intensive Care Unit, small for gesta-
tional age, smaller head circumferences, lower Apgar scores, congenital anom-
alies, neurodevelopmental delays, cognitive, psychosocial, behavioural and
emotional problems which persist beyond infancy and postpartum depression.
Anxiolytic Medications
Benzodiazepines
The SSRIs are now widely prescribed for anxiety disorders, including
obsessive-compulsive disorder and generalized anxiety disorder. However, BZs
are still considered for administration to pregnant women, in particular, as
sleeping aids.
Teratogenic Effects
Early reports suggesting a tenfold increase in cleft palate following first-
trimester exposure to BZs led to caution in prescribing this class of drug to preg-
nant women [40]. However, these findings have not been supported when data
from multiple studies are pooled: in a meta-analysis of 23 controlled studies with
BZ exposure in at least the first trimester, there was no significant association
between fetal exposure and major malformations or oral cleft specifically. When
only case-control studies were examined, however, the frequency of major mal-
formations and of oral cleft alone were both increased in the exposed group
(odds ratios of 3.01 and 1.79, respectively) [41]. The authors recommend that in
the absence of additional data, level 2 ultrasonography should be used to rule out
oral cleft in the case of first-trimester exposure to BZs. Others have recom-
mended that since BZs are seldom urgently required, other options should be
considered until patients are safely through the first trimester [42].
There is concern about neonatal withdrawal syndrome in infants exposed
to BZs in the third trimester of pregnancy: there have been a number of case
reports of symptoms of hypertonia and hyperreflexia after exposure to
diazepam [43] and chlordiazepoxide [44], and apnoea, lethargy and hypotonia
after exposure to clonazepam [45] and diazepam [46]. As for neonatal symp-
toms associated with third-trimester antidepressant use, symptoms associated
with late-pregnancy BZ exposure generally remit gradually without treatment.
Mood Stabilizers
The following mood stabilizers are commonly used in the treatment of
psychiatric illnesses, especially bipolar affective disorder: lithium, valproic
acid, carbamazepine, lamotrigine, topiramate and gabapentin. Issues of fetal
safety with in utero exposure to this class of medications will be discussed in
this section.
Lithium
Teratogenic Effects
The overall risk of congenital abnormalities following the use of lithium in
the first trimester is estimated to be 412%, an increase in the risk of 24%
reported in the general population [47]. Lithium administration in early preg-
nancy has been associated with cardiac abnormalities in particular, including
coarctation of the aorta, dextrocardia, patent ductus arteriosus, atresia of the
tricuspid valve, mitral atresia and intraventricular septal defect [48]. However,
the main cardiac defect of concern is Ebsteins anomaly, a malformation in the
tricuspid valve leading to downward displacement and adherence of portion of
the valve to the right ventricular wall [47]. Depending on the severity of the
anomaly, the presentation of the infant can range from asymptomatic to neonatal
cyanosis. Older children with a minor form of this anomaly may become symp-
tomatic with right ventricular failure only with exercise. Surgical correction of
this condition is difficult since a major complication of the procedure is damage
to the atrioventricular conduction system with significant mortality rates.
However, the risk of Ebsteins anomaly with the use of lithium is signifi-
cantly lower than once estimated [47]. Moreover, lithium is not consistently
found to be teratogenic: one prospective, multicentre study of 148 mother-
infant pairs found no significant increase in congenital malformations in
the exposed group relative to a matched control group [49]. If there has been
exposure to lithium in the first trimester, a high-resolution ultrasound and
echocardiogram at 1618 weeks of gestation should be ordered to detect the
presence of cardiac anomalies.
In the newborns of women treated with lithium near term or during deliv-
ery, lithium toxicity can occur. The symptoms include lethargy, hypotonia, poor
suck and grasp, muscle flaccidity, shallow respirations, cardiac arrhythmias
and cyanosis, and they can take up to 10 days to resolve. Studies have reported
an association between lithium use and premature delivery, macrosomia (large
size for gestational age) [48], nephrogenic diabetes insipidus [50] and non-toxic
goitres in the newborn [51]. If at all possible, lithium should be tapered and dis-
continued close to delivery to prevent lithium toxicity in the neonate; if this is
not possible, the infant should be closely monitored for up to 10 days of life.
Because renal excretion of lithium is increased in the third trimester and then
falls to pre-pregnancy levels, a decrease in the lithium dose close to delivery
may also be needed to prevent lithium toxicity in the mother.
Valproic Acid and Carbamazepine
Teratogenic Effects
The prevalence of congenital malformations associated with valproic acid
and carbamazepine exposure has been reported to be 11.1% and 5.7%, respec-
tively [52]. The main concern with the prenatal use of these medications is the
risk of neural tube defects. In the general population, neural tube defects occur in
6/10,000 pregnancies. The prevalence of spina bifida with valporic acid exposure
has been estimated to be 12% and 0.5% with carbamazapine exposure [53, 54].
This risk may be even higher with polytherapy and a positive family history of
neural tube defects. Some studies suggest that with valproic acid, the teratogenic
risk may be reduced when doses are kept below 1,000mg/day and serum levels
are less than 70g/ml [52]. A constellation of features, known as fetal anticon-
vulsant syndrome, has also been observed in children exposed to valproic acid
and carbamazepine in utero. These features include: urogenital abnormalities,
hypertelorism, epicanthal folds, microcephaly, minor skeletal abnormalities,
transverse palmar creases, abnormal midface, congenital heart defects, respira-
tory tract abnormality and spina bifida [55]. If valproic acid and carbamezapine
are used during the first trimester, an ultrasound at 1618 weeks (with or without
amniocentesis) should be performed to detect the presence of neural tube defects.
The use of valproic acid during pregnancy has been associated with a num-
ber of adverse neonatal outcomes, including hypoglycaemia, hypocalcaemia,
cholestasis, hyperbilirubinaemia and acute liver failure with hepatic necrosis,
though these data are derived primarily from case reports [48]. Fetal distress
during labour, low Apgar scores [56] and withdrawal symptoms (including
jitteriness, irritability, abnormal tone, feeding difficulties and seizures) [57]
have also been reported. As with lithium, an attempt should be made to decrease
the dose of anticonvulsants close to delivery if possible. There are also a few
case reports in the literature of hepatic dysfunction occurring in neonates of
mothers using carbamazepine [58].
To prevent haemorrhagic diseases in the newborn, some authorities have
suggested that pregnant women on anticonvulsants should receive vitamin K
starting at 36 weeks and then the infant should receive a dose at birth [59].
Lamotrigine, Gabapentin and Topiramate
Teratogenic Effects
Because minimal data have been collected on pregnancies after exposure
to lamotrigine, gabapentin and topiramate, it is not yet known whether these
drugs have teratogenic effects. In postmarketing surveillance studies (with
small sample sizes), the prevalence of birth defects was reported to be 1.8% and
4.5% with lamotrigine and gabapentin, respectively [60, 61].
Clinical Management
Women of reproductive age who are prescribed mood stabilizers should be
informed of the risks of these medications on a developing fetus. In turn, they
should also be advised regarding effective methods of contraception to maxi-
mize the possibility of a planned pregnancy. For instance, carbamazepine (and to
a lesser extent topiramate) are inducers of the cytochrome P450 3A4 enzyme,
which also metabolizes oral contraceptives. Therefore, because these drugs can
decrease the effectiveness of oral contraceptives, the latter should contain at least
50g of oestrogen; otherwise an additional or alternate method of contraception
should be recommended [62]. Furthermore, because up to 50% of all pregnan-
cies are unplanned, and because by 28 days after conception (i.e. before most
women know they are pregnant) neural tube closure is completed, women of
reproductive age who are being treated with valproic acid, carbamazepine and
lamotrigine should consider prophylactic folate supplementation. Although folic
acid has been shown to reduce the risk of neural tube defects in the general
population, its ability to achieve similar results in offspring of women using
anticonvulsants has been more equivocal. Although in 1997 the ACOG [63]
recommended 45mg/day of folic acid supplementation for women treated with
anticonvulsants, the appropriate dose still remains unclear.
If possible, mood stabilizers should be tapered and discontinued prior to
conception (or at least during embryogenesis in the first trimester). However,
this option must be carefully weighed against the risks to the mother and fetus
resulting from untreated bipolar affective disorder (e.g. impaired judgment and
psychosis leading to dangerous behaviour, poor prenatal care, fetal abuse and
potential need for re-introduction of multiple medications at higher doses to
treat a relapse). If medications are continued throughout pregnancy, one should
aim for monotherapy at the lowest effective dose, especially during the first
trimester. Serum concentrations of mood stabilizers should be monitored at
least once each trimester since drug levels may drop as pregnancy progresses
(especially in the third trimester). However, dosage adjustments should be clin-
ically based and not solely decided upon by serum drug levels.
Antipsychotic Medications
Antipsychotic medications are used in the treatment of many psychiatric
illnesses including schizophrenia, schizoaffective disorder, psychotic mood
disorders and organic psychoses. The decision to start or continue antipsychotic
medication during pregnancy must be made by weighing the risk of illness to
the mother and fetus against the risk to the fetus from exposure.
It is unlikely that psychotic illness will improve during pregnancy. In fact,
a prospective study found that 59% of women reported deterioration in their
mental health while pregnant and 29% cited improvement [64]. It has also been
reported that when maintenance antipsychotics are discontinued during preg-
nancy, 65% of women with schizophrenia will relapse [65]. Younger women
and those for whom the pregnancy was unwanted may be at increased risk for
decompensation. Psychosis in pregnancy may lead to poor prenatal care, poor
nutrition, elevated stress levels, substance abuse, impulsive behaviour, suicidal-
ity, failure to recognize signs of labour, delusional interpretation of pregnancy,
attempts at premature self-delivery and violence.
Conventional Antipsychotics
Teratogenic Effects
While all antipsychotics diffuse across the placenta, it has generally been
agreed that the class of butyrophenones (e.g. Haldol) is safer to use than phe-
nothiazines (e.g. CPZ) during pregnancy. However, available studies looking
at the teratogenic potential of traditional antipsychotics are plagued with certain
limitations. In these studies, confounding variables (e.g. time of gestation, illness
effect, substance abuse, intensity and duration of exposure, nutritional status,
other medications) are not always well controlled for. Furthermore, most data
have been collected from non-psychiatrically ill women receiving the antipsy-
chotic to control hyperemesis gravidarum or for sedative purposes during
labour; therefore, doses of antipsychotics and the length of exposure of the fetus
to them are often much smaller than would be needed to control psychoses.
Nevertheless, retrospective studies have found no increase in congenital
abnormalities following in utero exposure to Haldol. In the largest study, 98
women were treated with Haldol with a mean dose of 1.2 mg/day for hypereme-
sis gravidarum over the course of a few days to a few weeks. No malformations
were noted in these infants and Haldol was not found to effect duration of ges-
tation, birth weight or fetal viability [6668].
In a large series of case reports, no increase in congenital malformation
has been found in offspring exposed to trifluoperazine in pregnancy. Again
however, many of these women were on low doses for the treatment of nausea
[69, 70].
Phenothiazines are the most extensively studied class of antipsychotics.
A prospective, longitudinal controlled study of 1,952 infants exposed to phenoth-
iazines during pregnancy showed a non-significant trend towards an increase in
major congenital anomalies when exposed between week 4 and 10 [71]. A meta-
analysis of large retrospective and small prospective controlled trials (conducted
between 1963 and 1995) showed a small but statistically significant increase in
non-specific congenital abnormalities by 0.4% [32]. However, there may be a
differential effect to the individual phenothiazines. A prospective survey of
12,764 pregnancies exposed to CPZ and piperazine showed a significant
increase in congenital malformations in infants exposed in utero to CPZ but not
piperazine. The congenital anomalies noted in the CPZ group included cleft lip,
cleft palate, limb deformities and cardiac abnormalities [72]. It should also be
noted, however, that other studies report no association between in utero phe-
nothiazines exposure and newborn malformations [73]. Finally, some authors
have suggested that schizophrenia itself may be associated with an increase in
congenital abnormalities as well as fetal and neonatal deaths [74]. Low socio-
economic class and non-compliance with prenatal care in this group of women
may be contributory [74].
Treatment of extrapyramidal side-effects common to traditional antipsy-
chotics may be problematic in pregnancy since antihistamines, anticholinergics
and amantadine have all been associated with an increase in congenital malfor-
mations [37, 75]. If possible, these medications should therefore be avoided in the
first trimester. Since low calcium levels can predispose to extrapyramidal side-
effects and since pregnancy is a time of high calcium demand, particular attention
should be paid to diet and vitamin supplementation in these women [76].
There have been case reports of perinatal effects in the neonate caused by
late-pregnancy exposure to antipsychotics. These symptoms (including tremor,
hypertonicity, motor restlessness, abnormal movements, poor suck, functional
bowl obstruction and jaundice in preterm infants) are often transient and
resolve within days [77]. As a result, some authors suggest discontinuing, or at
least lowering the dose of the antipsychotic 12 weeks before delivery and
then increasing the dose again following delivery. This recommendation should,
however, be weighed against the risk of decompensation in an individual
woman.
Atypical Antipsychotic Medications
Only 1 study has prospectively investigated the risk for major malforma-
tions, low birth weight and prematurity after prenatal exposure to atypical
antipsychotic medications [78]. Pregnancy outcomes of 151 women who took
olanzapine (n 60), risperidone (n 49), quetiapine (n 36) or clozapine
(n 6) were compared to those of a control group without psychiatric diag-
noses exposed to a non-teratogen. No statistically significant increase was
found in the rate of major malformations, low birth weight or prematu-
rity among the infants with prenatal exposure to atypical antipsychotic
medications.
Despite the small sample size (and therefore modest power) of this study,
the results are reassuring, particularly since women in the exposed group were
more likely to endorse a number of risk factors previously associated with poor
outcomes (e.g. smoking, alcohol consumption) than women in the control group.
Similarly, many women in the exposure group were using other psychotropic
medications besides atypical antipsychotics (including conventional antipsy-
chotics, antidepressants and anticonvulsants). As a result of these potentially
confounding factors, women in the exposed group would be expected to have a
higher than usual risk for malformations, even in the absence of the exposure to
atypical antipsychotic medications.
In the sections that follow, we review the available data for specific atypi-
cal antipsychotic medications.
Olanzapine
The manufacturer of olanzapine has maintained a safety database of infor-
mation about the use of this drug in pregnancy and breast-feeding. From 1983
to 2001, there were 144 prospectively reported cases of olanzapine use during
pregnancy for which the outcomes are known [79, 80]. In this data set, there
were 12 spontaneous abortions, 3 stillbirths, 6 premature births and 6 major
malformations. In addition, perinatal complications were reported in 12 cases.
When compared to rates of adverse outcomes in the general population, these
rates are within normal ranges. The manufacturers database also includes 98
retrospectively identified cases of olanzapine use in pregnancy within which
higher rates of adverse outcome were observed; however, the retrospective
nature of the reports makes it difficult to draw conclusions.
The case report literature for olanzapine primarily describes healthy out-
comes after prenatal exposure to doses ranging from 7.5 to 25 mg/day
[8185]. There is 1 report of a pregnancy complicated by a 79-lb weight gain
as well as gestational hypertension, gestational diabetes and pre-eclampsia
resulting in premature birth; however, the role of olanzapine in contributing to
these complications is unclear [86]. It should be noted that there does not
appear to be any specific pattern of abnormality or teratogenicity associated
with prenatal exposure to olanzapine. Based on these findings, olanzapine
may be used in pregnant women where benefits are felt to outweigh the risk to
the fetus.
Risperidone
The risks to human fetuses exposed to risperidone are largely unknown, as
there have not been any adequate human studies to date. It has been shown to be
fetotoxic, but not teratogenic, in animal studies [87]. There are 2 published case
reports of human exposure to risperidone during all three trimesters of preg-
nancy, both of which reported healthy outcomes [88]. A postmarketing study of
the tolerability of risperidone in 7,684 male and female patients included data
from 9 patients who took the drug during 10 pregnancies. In this group, there
were 7 live births and 3 therapeutic abortions; no abnormalities were reported
among the live births [89].
The manufacturer of risperidone received 25 spontaneous reports of expo-
sure to risperidone during pregnancy [Janssen-Ortho, written commun., 1997].
Of these 25 exposures, 10 resulted in normal outcomes. In addition, there were
9 spontaneous abortions, 3 therapeutic abortions, 1 episode of neonatal jitteri-
ness which developed on the 7th day of life in a neonate whose mother was also
taking sertraline, and 2 abnormal outcomes. These abnormalities consisted of a
case of agenesis of the corpus callosum in a fetus whose mother had been on
numerous medications including 3 days of risperidone at 13 weeks of gestation,
and a baby who was delivered at 30 weeks and suffered an intracerebral bleed as
a result of prematurity. This infants mother had also been on other neuroleptics
in addition to anticonvulsants and antidepressants.
Quetiapine
Little is known about the reproductive safety of quetiapine. To date, there
have been no reported cases of teratogenicity in humans; however, preliminary
animal studies have shown some delay in skeletal ossification, decreased fetal
weight and an increase in fetal and pup death [90]. There have been 2 published
case reports of patients who conceived while taking quetiapine [91, 92]. In both
cases, the drug was continued throughout pregnancy and healthy outcomes
were reported. Until more data are available, quetiapine should only be used in
pregnancy if the potential risks of discontinuing therapy are felt to outweigh the
unknown potential risks to the fetus.
Clozapine
The primary concerns with the use of clozapine during pregnancy and
breast-feeding are the potential for agranulocytosis as well as its association
with diabetes mellitus. Although agranulocytosis has not been reported in
fetuses/infants exposed to clozapine, the theoretical possibility exists and the
incapacity to monitor fetal hematology makes this a concern. Clozapine has
been associated with both new onset and exacerbation of diabetes mellitus,
besides its effects on weight gain [93]. This raises concerns about the potential
sequelae of gestational diabetes including macrosomia, shoulder dystocia and
fetal hypoglycaemia.
Clozapine has not been associated with any congenital anomalies and is
the only atypical antipsychotic to be considered a class B drug during preg-
nancy by the FDA (i.e. no evidence of risk in humans). The manufacturer
received over 100 spontaneous reports of clozapine exposure during pregnancy
between 1996 and 2000. Of these, 20% resulted in unknown outcomes, 6% in
elective terminations and over 40 healthy babies were born. There were 5 spon-
taneous abortions and 2 intra-uterine deaths including a fetus with Reyes syn-
drome and a pregnancy complicated by placenta previa marginalis [Novartis,
Montreal, Quebec, Canada, pers. commun., 2001].
A review published in 1995 reported on 61 children born to 59 mothers
who used clozapine during pregnancy. Of these, 51 babies were healthy, 5 had
congenital malformations and 5 experienced neonatal complications [94].
There are 8 case reports of clozapine use in pregnancy in the literature: 2 were
uncomplicated [95, 96], there are 2 cases of diabetes complicated by shoulder
dystocia at delivery [97, 98], 1 report of f loppy baby syndrome (possibly con-
founded by concomitant use of lorazepam) [99], 1 infant with a cephalhe-
matoma, hyperpigmentation folds and a coccygeal dimple who went on to
have a seizure at 8 days of life [96] and 1 case of decreased fetal heart rate vari-
ability [100]. Finally, there is a report of a stillbirth at 32 weeks of an infant
with intra-uterine growth retardation to a woman using clozapine who demon-
strated very poor self-care [101].
Long-Term Neurodevelopment after in utero Exposure to
Psychotropic Medications
Limited data are available to provide reassurance that fetal exposure to the
psychotropic medications described here has no lasting consequences for
neurodevelopment. Normal behavioural, cognitive and motor development has
been reported in 3 samples of children exposed to TCA and/or SSRI antide-
pressants in utero: 1 set of 18 children exposed to TCAs followed up between
4 months to 3 years [102], an additional 80 children exposed to TCAs and
55 children exposed to fluoxetine followed up between 16 and 86 months [8]
and most recently, a sample of 46 children exposed to TCAs and 40 children
exposed to fluoxetine followed up between 15 and 71 months [103]. It is
notable that in the latter study, both IQ and language development were nega-
tively associated with the duration/number of episodes of depression since
childbirth: this is strongly suggestive that untreated maternal depression can
have significant effects on child development.
Another recent study provided data on 31 children of depressed mothers
treated with SSRIs, in comparison with 13 children of depressed mothers who
elected not to take medication, assessed between ages 640 months. The
exposed children had lower scores on the Bayley Psychomotor Development
Indexes and the motor quality factor of the Bayley Behavioural Rating Scale
and specifically exhibited more frequent tremulousness and inappropriate fine
motor movements [104]. There are limitations to this study, particularly the lack
of data on the concomitant use of other psychotropic medications, including
BZs, and outcome testing across a wide age range. Thus, the findings await
replication.
A small study of 11 mother-infant pairs with prenatal citalopram exposure
found normal neurodevelopment up to 1 year of age [105], as did a similar
study of 11 infants exposed to fluoxetine [106]. However, the small sample size
and non-specific assessment of infant neurodevelopment make the clinical
significance of these studies questionable.
There are some data with respect to child development after fetal BZ expo-
sure. In 1 study of 17 infants with BZ exposure throughout gestation and
29 infants with no psychotropic exposure, the BZ-exposed infants had signifi-
cantly lower scores on all subscales of the Griffiths Developmental Scale at
10 months of age and on the personal-social behaviour and hearing and speech
subscales at 18 months of age [107]. However, the two groups were not matched
for cigarette smoking or psychiatric diagnosis, variables which could account
for effects on child development.
Most recently, a Danish study has reported on a comparison between 435
women using a psychotropic medication during pregnancy and 1,304 women
without psychotropic medication use who had given birth in the same region
and time period. Public health records were extracted and scores on the Boel
test (a test of psychomotor development which assesses hearing, sight and
motor attention) at 710 months of age were compared for the infants in each
group. Abnormal results on the Boel test were more frequent in the children
who were presumed to be drug exposed (16%) than in those presumed not to be
exposed (4%): the odds ratios were 4.5 for antidepressant medications, 4.6 for
anti-epileptic medications, 3.9 for neuroleptic medications and 4.0 for BZs
[108]. These odds ratios increased after adjustment for gestational age, birth
weight, breast-feeding, type of residence and childs age. However, it is notable
that data were not available for other potentially important confounding factors
such as smoking or use of illicit drugs. In addition, the authors note that assess-
ments were not blinded and children who were suspected to have problems may
have been more likely to undergo assessment with the Boel test. Thus, these
findings await replication.
Only limited data have specifically examined neurodevelopment after
prenatal exposure to mood stabilizers. A follow-up study of 60 children born to
mothers using lithium during pregnancy showed no difference in physical or
mental difficulties compared to their unexposed siblings [109]. A few follow-up
studies of offspring born to mothers having used valproic acid during preg-
nancy suggest that these children may experience some developmental delay
and require more help in school [110, 111]; another study however found
normal psychomotor development [112]. Similarly, some studies suggest an
association between maternal carbamazepine use and lowered cognitive scores
and developmental delays in the children [113], while other studies have not
found an association with cognitive dysfunction [114]. No follow-up studies
assessing neurobehavioural sequelae following lamotrigine, gabapentin and
topiramate use are yet available.
Data examining the neurodevelopmental effects on offspring exposed to
traditional antipsychotics in utero remain limited. However, no adverse out-
comes have been documented [115118]. One study noted that children
exposed to antipsychotics during pregnancy were taller and/or heavier than the
matched control group [119].
In summary, methodological issues make the collection of long-term
neurodevelopmental follow-up data difficult and the available studies have
substantial weaknesses. While the results of Nulman et al. [8, 103] are reassur-
ing with respect to the antidepressants, it should be noted that in each of these
studies, children were assessed only once and over a very wide age range,
potentially making the groups very heterogeneous and decreasing the likeli-
hood of finding statistically significant differences in the exposed children. The
recent studies that have identified subtle differences in drug-exposed children
are concerning, and further research employing specific definitions of expo-
sure, longitudinal methods and detailed, unbiased assessments of developmen-
tal functioning are urgently required.
Conclusions
Psychiatric disorders are common during pregnancy and pharmacotherapy
will often be necessary. Unfortunately, there are few data to guide clinical deci-
sion making about the choice of medication or dosage regimen, neither with
respect to efficacy nor with respect to fetal safety [120, 121]. Moreover, current
guidelines and clinical practice for the use of antipsychotic drugs in women
with psychotic disorders during pregnancy are not based on evidence from ran-
domized controlled trials [122]. Furthermore, the current classification systems
for teratogenic risk provide little help to clinicians, since specific drugs are not
consistently rated across international systems [123] and in the US system,
many of the psychotropic drugs are classified into category C (adverse effects
demonstrated in animals but no human data available) [124]. Additional
research on the effects of psychotropic medications, and in particular, the
potential effects on long-term neurodevelopment, is sorely needed. Until such
data become available, however, risk-benefit analyses must be made on a case-
by-case basis, working towards the best possible outcomes for mother and baby
as a unit.
References
1 Brockington I: Motherhood and Mental Health. Oxford, Oxford University Press, 1996.
2 Kumar R, Robson KM: A prospective study of emotional disorders in childbearing women. Br J
3 Bennett HA, Einarson A, Taddio A, Koren G, Einarson TR: Depression during pregnancy:
Overview of clinical factors. Clin Drug Invest 2004;24:157179.
4 Morgan DJ: Drug disposition in mother and foetus. Clin Exp Pharmacol Physiol 1997;24:
869873.
5 Steiner M, Yonkers KA: Evidence-based treatment of mood disorders in women. Ment Fit
2003;2:3467.
6 Chambers CD, Johnson KA, Dick LM, Felix RJ, Jones KL: Birth outcomes in pregnant women
taking fluoxetine. N Engl J Med 1996;335:10101015.
7 Pastuszak A, Schick-Boschetto B, Zuber C, Feldkamp M, Pinelli M, Sihn S, Donnenfeld A,
McCormack M, Leen-Mitchell M, Woodland C, Gardner A, Hom M, Koren G: Pregnancy out-
come following first-trimester exposure to fluoxetine (Prozac). JAMA 1993;269:22462248.
8 Nulman I, Rovet J, Stewart DE, Wolpin J, Gardner HA, Theis JG, Kulin N, Koren G: Neuro-
development of children exposed in utero to antidepressant drugs. N Engl J Med 1997;336:
258262.
9 Kulin NA, Pastuszak A, Sage SR, Schick-Boschetto B, Spivey G, Feldkamp M, Ormond K,
Matsui D, Stein-Schechman AK, Cook L, Brochu J, Rieder M, Koren G: Pregnancy outcome
following maternal use of the new selective serotonin reuptake inhibitors: a prospective controlled
multicenter study. JAMA 1998;279:609610.
10 Simon GE, Cunningham ML, Davis RL: Outcomes of prenatal antidepressant exposure. Am J
Psychiatry 2002;159:20552061.
11 Wisner KL, Gelenberg AJ, Leonard H, Zarin D, Frank E: Pharmacologic treatment of depression
during pregnancy. JAMA 1999;282:12641269.
12 Addis A, Koren G: Safety of fluoxetine during the first trimester of pregnancy: A meta-analytical
review of epidemiological studies. Psychol Med 2000;30:8994.
13 Hendrick V, Smith LM, Suri R, Hwang S, Haynes D, Altshuler L: Birth outcomes after prenatal
exposure to antidepressant medication. Am J Obstet Gynecol 2003;188:812815.
14 McElhatton PR, Garbis HM, Elefant E, Vial T, Bellemin B, Mastroiacovo P, Arnon J, Rodriguez-
Pinilla E, Schaefer C, Pexieder T, Merlob P, Dal Verme S: The outcome of pregnancy in 689
women exposed to therapeutic doses of antidepressants. A collaborative study of the European
Network of Teratology Information Services (ENTIS). Reprod Toxicol 1996;10:285294.
15 Ericson A, Kallen B, Wiholm B: Delivery outcome after the use of antidepressants in early preg-
nancy. Eur J Clin Pharmacol 1999;55:503508.
16 Einarson A, Fatoye B, Sarkar M, Lavigne SV, Brochu J, Chambers C, Mastroiacovo P, Addis A,
Matsui D, Schuler L, Einarson TR, Koren G: Pregnancy outcome following gestational exposure
to venlafaxine: a multicenter prospective controlled study. Am J Psychiatry 2001;158:17281730.
17 Einarson A, Bonari L, Voyer-Lavigne S, Addis A, Matsui D, Johnson Y, Koren G: A multicentre
prospective controlled study to determine the safety of trazodone and nefazodone use during preg-
nancy. Can J Psychiatry 2003;48:106110.
18 Koren G, Matsui DM, Einarson A, Steiner M: Is maternal use of selective serotonin reuptake
inhibitors in the third trimester of pregnancy harmful to neonates? Can Med Assoc J, in press.
19 Kent LSW, Laidlaw JDD: Suspected congenital sertraline dependence. Br J Psychiatry 1995;167:
412413.
20 Spencer MJ: Fluoxetine hydrochloride (Prozac) toxicity in a neonate. Pediatrics 1993;92:721722.
21 Mhanna MJ, Bennet JB 2nd, Izatt SD: Potential fluoxetine chloride (Prozac) toxicity in a newborn.
Pediatrics 1997;100:158159.
22 Cohen LS, Heller VL, Bailey JW, Grush L, Ablon JS, Bouffard SM: Birth outcomes following pre-
natal exposure to fluoxetine. Biol Psychiatry 2000;48:9961000.
23 Goldstein DJ: Effects of third trimester fluoxetine exposure on the newborn. J Clin Psychopharmacol
1995;15:417420.
24 Nordeng H, Lindemann R, Perminov KV, Reikvam A: Neonatal withdrawal syndrome after in
utero exposure to selective serotonin reuptake inhibitors. Acta Paediatr 2001;90:288291.
25 Zeskind PS, Stephens LE: Maternal selective serotonin reuptake inhibitor use during pregnancy
and newborn neurobehavior. Pediatrics 2004;113:368375.
26 Judge R, Parry MG, Quail D, Jacobson JG: Discontinuation symptoms: Comparison of brief interrup-
tion in fluoxetine and paroxetine treatment. Int Clin Psychopharmacol 2002;17:217225.
27 Dahl ML, Olhager E, Ahlner J: Paroxetine withdrawal syndrome in a neonate. Br J Psychiatry
1997;171:391392.
28 Stiskal JA, Kulin N, Koren G, Ho T, Ito S: Neonatal paroxetine withdrawal syndrome. Arch Dis
Child Fetal Neonatal Ed 2001;84:F134F135.
29 Costei AM, Kozer E, Ho T, Ito S, Koren G: Perinatal outcome following third trimester exposure
to paroxetine. Arch Pediatr Adolesc Med 2002;156:11291132.
30 Kallen B: Neonate characteristics after maternal use of antidepressants in late pregnancy. Arch
Pediatr Adolesc Med 2004;158:312316.
31 Laine K, Heikkinen T, Ekblad U, Kero P: Effects of exposure to selective serotonin reuptake
inhibitors during pregnancy on serotonergic symptoms in newborns and cord blood monoamine
and prolactin concentrations. Arch Gen Psychiatry 2003;60:720726.
32 Altshuler LL, Cohen L, Szuba MP, Burt VK, Gitlin M, Mintz J: Pharmacologic management of psy-
chiatric illness during pregnancy: Dilemmas and guidelines. Am J Psychiatry 1996;153:592606.
33 Webster PA: Withdrawal symptoms in neonates associated with maternal antidepressant therapy.
Lancet 1973;2:318319.
34 Eggermont E: Neonatal effects of maternal therapy with tricyclic antidepressant drugs. Arch Dis
Child 1980;55:81.
35 Cowe L, Lloyd DJ, Dawling S: Neonatal convulsions caused by withdrawal from maternal
clomipramine. Br Med J 1982;284:18371838.
36 Nonacs R, Cohen LS: Assessment and treatment of depression during pregnancy: An update.
Psychiatr Clin North Am 2003;26:547562.
37 Heinonen OP, Slone D, Shapiro S, Gaetano LF, Hartz SC, Mitchell AA, Monson RR, Rosenberg L,
Siskind V: Sedatives, tranquilizers and antidepressant drugs; in Kaufman DW (ed): Birth Defects
and Drugs in Pregnancy. Littleton, Publishing Sciences Group, 1977.
38 Pavy TJ, Kliffer AP, Douglas MJ: Anaesthetic management of labour and delivery in a woman
taking long-term MAOI. Can J Anaesth 1995;42:618620.
39 Gracious BL, Wisner KL: Phenelzine use throughout pregnancy and the puerperium: Case report,
review of the literature, and management recommendations. Depress Anxiety 1997;6:124128.
40 Barry WS, St Clair SM: Exposure to benzodiazepines in utero. Lancet 1987;1:14361437.
41 Dolovich LR, Addis A, Vaillancourt JM, Power JD, Koren G, Einarson TR: Benzodiazepine use in
pregnancy and major malformations or oral cleft: Meta-analysis of cohort and case-control stud-
ies. BMJ 1998;317:839843.
42 St Clair SM, Schirmer RG: First-trimester exposure to alprazolam. Obstet Gynecol 1992;80:
843846.
43 Mazzi E: Possible neonatal diazepam withdrawal: A case report. Am J Obstet Gynecol 1977;129:
586587.
44 Athinarayanan P, Pierog SH, Nigam SK, Glass L: Chlordiazepoxide withdrawal in the neonate.
45 Fisher JB, Edgren BE, Mammel MC, Coleman JM: Neonatal apnea associated with maternal
clonazepam therapy: A case report. Obstet Gynecol 1985;66:34S35S.
46 Rementeria JL, Bhatt K: Withdrawal symptoms in neonates from intrauterine exposure to
diazepam. J Pediatr 1977;90:123126.
47 Cohen LS, Friedman JM, Jefferson JW, Johnson EM, Weiner ML: A reevaluation of risk of in
utero exposure to lithium. JAMA 1994;271:146150.
48 Iqbal M, Sohhan T, Mahmud SZ: The effects of lithium, valproic acid, and carbamazepine during
pregnancy and lactation. Clin Toxicol 2001;39:381392.
49 Jacobson SJ, Jones K, Johnson K, Ceolin L, Kaur P, Sahn D, et al: Prospective multicentre study of
pregnancy outcome after lithium exposure during first trimester. Lancet 1992;339:530533.
50 Mizrahi EM, Hobbs JF, Goldsmith DI: Nephrogenic diabetes insipidus in transplacental lithium
intoxication. J Pediatr 1979;94:493495.
51 Karlsson K, Lindstedt G, Lundberg PA, Selstam U: Transplacental lithium poisoning: Reversible
inhibition of fetal thyroid. Lancet 1975;1:1295.
52 Kaneko S, Battino D, Andermann E, Wada K, Kan R, Takeda A, Nakane Y, Ogawa Y, Avanzini G,
Fumarola C, Granata T, Molteni F, Pardi G, Minotti L, Canger R, Dansky L, Oguni M, Lopes-
Cendas I, Sherwin A, Andermann F, Seni MH, Okada M, Teranishi T: Congenital malformations
due to antiepileptic drugs. Epilepsy Res 1999;33:145158.
53 Lindhout D, Schmidt D: In-utero exposure to valproate and neural tube defects. Lancet
1986;1:13921393.
54 Rosa FW: Spina bifida in infants of women treated with carbamazepine during pregnancy. N Engl
J Med 1991;324:674677.
55 Crawford P: Epilepsy and pregnancy. Seizure 2001;10:212219.
56 Jager-Roman E, Deichl A, Jakob S, Hartmann AM, Koch S, Rating D, Steldinger R, Nau H, Helge
H: Fetal growth, major malformations and minor anomalies in infants born to women receiving
valproic acid. J Pediatr 1986;108:9971004.
57 Thisted E, Ebbesen F: Malformations, withdrawal manifestations, and hypoglycaemia after
exposure to valproate in utero. Arch Dis Child 1993;69:288291.
58 Merlob P, Mor N, Litwin A: Transient hepatic dysfunction in an infant of an epileptic mother
treated with carbamazepine during pregnancy and breastfeeding. Ann Pharmacother 1992;26:
15631565.
59 Thorp JA, Gaston L, Caspers DR, Pal ML: Current concepts and controversies in the use of
vitamin K. Drugs 1995;49:376387.
60 Tennis P, Eldridge RR: Preliminary results on pregnancy outcomes in women using lamotrigine.
Epilepsia 2002;43:11611167.
61 Wilton LV, Shakir S: A postmarketing surveillance study of gabapentin as add-on therapy for
3,100 patients in England. Epilepsia 2002;43:983992.
62 McAuley JW, Anderson GD: Treatment of epilepsy in women of reproductive age: pharmaco-
kinetic considerations. Clin Pharmacokinet 2002;41:559579.
63 ACOG educational bulletin: Seizure disorders in pregnancy. Int J Gynecol Obstet
1997;56:279286.
64 McNeil TF, Kaij L, Malmquist-Larsson A: Women with nonorganic psychosis: Pregnancys effect
on mental health during pregnancy. Acta Psychiatr Scand 1984;70:140148.
65 Casiano ME, Hawkins DR: Major mental illness and childbearing. A role for the consultation-
liaison psychiatrist in obstetrics. Psychiatr Clin North Am 1987;10:3551.
66 van Waes A, van de Velde E: Safety evaluation of haloperidol in the treatment of hyperemesis
gravidarum. J Clin Pharmacol 1969;9:224227.
67 Hanson JW, Oakley GP Jr: Haloperidol and limb deformity. JAMA 1975;231:26.
68 Altshuler LL, Szuba MP: Course of psychiatric disorders in pregnancy. Neurol Clin 1994;12:
613635.
69 Moriarty AJ, Nance MR: Trifluoperazine and pregnancy. Can Med Assoc J 1963;88: 375376.
70 Schrire I: Trifluoperazine and foetal abnormalities. Lancet 1963;1:174.
71 Miklovich L, van den Berg BJ: An evaluation of the teratogenicity of certain antinauseant drugs.
72 Rumeau-Rouquette C, Goujard J, Huel G: Possible teratogenic effect of phenothiazines in human
beings. Teratology 1977;15:5764.
73 Sobel DE: Fetal damage due to ECT, insulin coma, chlorpromazine, or reserpine. Arch Gen
Psychiatry 1960;2: 606611.
74 Pinkofsky HB: Effects of antipsychotics on the unborn child: What is known and how should this
influence prescribing? Paediatr Drugs 2000;2:8390.
75 Gjere NA: Psychopharmacology in pregnancy. J Perinat Neonatal Nurs 2001;14:1225.
76 Kuny S, Binswanger U: Neuroleptic-induced extrapyramidal symptoms and serum calcium levels.
Results of a pilot study. Neuropsychobiology 1989;21:6770.
77 Tamer A, McKey R, Arias D, Worley L, Fogel BJ: Phenothiazine-induced extrapyramidal dysfunction
in the neonate. J Pediatr 1969;75:479480.
78 McKenna K, Koren G, Tetelbaum M, Wilton L, Shakir S, Diav-Citrin O, Levinson A, Zipursky RB,
Einarson A: Pregnancy outcome of women using atypical antipsychotic drugs: A prospective
comparative study. J Clin Psychiatry, in press.
79 Eli Lilly Canada Inc: Use of Zyprexa in pregnant or nursing women. Manufacturers Information.
2002.
80 Goldstein DJ, Corbin LA, Fung MC: Olanzapine-exposed pregnancies and lactation: Early experi-
ence. J Clin Psychopharmacol 2000;20:399403.
81 Kirchheiner J, Berghofer A, Bolk-Weischedel D: Healthy outcome under olanzapine treatment in
a pregnant woman. Pharmacopsychiatry 2000;33:7880.
82 Malek-Ahmadi P: Olanzapine in pregnancy. Ann Pharmacother 2001;35:12941295.
83 Lim LM: Olanzapine and pregnancy. Aust N Z J Psychiatry 2001;35:856857.
84 Nagy A, Tenyi T, Lenard K, Herold R, Wilhelm F, Trixler M: Olanzapine and pregnancy. Orv Hetil
2001;142:137138.
85 Neumann NU, Frasch K: Olanzapine and pregnancy. 2 case reports. Nervenarzt 2001;72:876878.
86 Littrell KH, Johnson CG, Peabody CD, Hilligoss N: Antipsychotics during pregnancy. Am J
87 Iqbal MM, Gundlapalli SP, Ryan WG, Ryals T, Passman TE: Effects of antimanic mood-stabilizing
drugs on fetuses, neonates, and nursing infants. South Med J 2001;94:304322.
88 Ratnayake T, Libretto SE: No complications with risperidone treatment before and throughout
pregnancy and during the nursing period. J Clin Psychiatry 2002;63:7677.
89 Mackay FJ, Wilton LV, Pearce GL, Freemantle SN, Mann RD: The safety of risperidone: A post-
marketing study on 7,684 patients. Hum Psychopharmacol Clin Exp 1998;13:413418.
90 Ernst CL, Goldberg JF: The reproductive safety profile of mood stabilizers, atypical antipsy-
chotics, and broad-spectrum psychotropics. J Clin Psychiatry 2002;63(suppl 4):4255.
91 Tenyi T, Trixler M, Keresztes Z: Quetiapine and pregnancy. Am J Psychiatry 2002;159:674.
92 Taylor TM, OToole MS, Ohlsen RI, Walters J, Pilowsky LS: Safety of quetiapine during preg-
nancy. Am J Psychiatry 2003;160:588589.
93 Popli AP, Konicki PE, Jurjus GJ, Fuller MA, Jaskiw GE: Clozapine and associated diabetes mellitus.
J Clin Psychiatry 1997;58:108111.
94 Dev V, Krupp P: The side-effects and safety of clozapine. Rev Contemp Pharmacother 1995;6:
197208.
95 Barnas C, Bergant A, Hummer M, Saria A, Fleischhacker WW: Clozapine concentrations in
maternal and fetal plasma, amniotic fluid, and breast milk. Am J Psychiatry 1994;151:945.
96 Stoner SC, Sommi RW Jr, Marken PA, Anya I, Vaughn J: Clozapine use in two full term pregnan-
cies. J Clin Psychiatry 1997;58:364365.
97 Waldman MD, Safferman AZ: Pregnancy and clozapine. Am J Psychiatry 1993;150:168169.
98 Dickson RA, Hogg L: Pregnancy of a patient treated with clozapine. Psychiatr Serv 1998;49:
10811083.
99 DiMichele V, Ramenghi LA, Sabatino G: Clozapine and lorazepam administration in pregnancy.
Eur Psychiatry 1996;11:214.
100 Yogev Y, Ben Haroush A, Kaplan B: Maternal clozapine treatment and decreased fetal heart rate
variability. Int J Gynaecol Obstet 2002;79:259260.
101 Mendhekar DN, Sharma JB, Srivastava PK, War L: Clozapine and pregnancy. J Clin Psychiatry
2003;64:850.
102 Misri S, Sivertz K: Tricyclic drugs in pregnancy and lactation: A preliminary report. Int J
Psychiatry Med 1991;21:157171.
103 Nulman I, Rovet J, Stewart DE, Wolpin J, Pace-Asciak P, Shuhaiber S, Koren G: Child development
following exposure to tricyclic antidepressants or fluoxetine throughout fetal life: A prospective,
104 Casper RC, Fleisher BE, Lee-Ancajas JC, Gilles A, Gaylor E, DeBattista A, Hoyme HE: Follow-
up of children of depressed mothers exposed or not exposed to antidepressant drugs during preg-
nancy. J Pediatr 2003;142:402408.
105 Heikkinen T, Ekblad U, Kero P, Ekblad S, Laine K: Citalopram in pregnancy and lactation. Clin
Pharmacol Ther 2002;72:184191.
106 Heikkinen T, Ekblad U, Palo P, Laine K: Pharmacokinetics of fluoxetine and norfluoxetine in
pregnancy and lactation. Clin Pharmacol Ther 2003;73:330337.
107 Viggedal G, Hagberg BS, Laegreid L, Aronsson M: Mental development in late infancy after prena-
tal exposure to benzodiazepines A prospective study. J Child Psychol Psychiatry 1993;34:295305.
108 Mortensen JT, Olsen J, Larsen H, Bendsen J, Obel C, Sorensen HT: Psychomotor development in
children exposed in utero to benzodiazepines, antidepressants, neuroleptics, and anti-epileptics.
Eur J Epidemiol 2003;18:769771.
109 Schou M: What happened later to the lithium babies? A follow-up study of children born without
malformations. Acta Psychiatr Scand 1976;54:193197.
110 Ardinger HH, Atkin JF, Blackstone RD, Elsas LJ, Clarren SK, Livingstone S, Flannery DB, Pellock
JM, Harrod MJ, Lammer EJ, Majewski F, Schinzel A, Toriello HV, Hanson JW: Verification of the
fetal valproate syndrome phenotype. Am J Med Genet 1988;29:171185.
111 Adab N, Jacoby A, Smith D, Chadwick D: Additional educational needs in children born to moth-
ers with epilepsy. J Neurol Neurosurg Psychiatry 2001;70:1521.
112 Granstrom ML: Development of the children of epileptic mothers Preliminary results from the
prospective Helsinki study; in Janz D, Dam M, Richens A (eds): Epilepsy, Pregnancy, and the
Child. New York, Raven Press, 1982.
113 Onroy A, Cohen I: Follow-up studies of children born to epileptic women treated with carba-
mazepine. Teratology 1995;51:169.
114 Scolnik D, Nulman I, Rovet J, Gladstone D, Czuchta D, Gardner HA, Gladstone R, Ashby P,
Weksberg R, Einarson T, Koren G: Neurodevelopment of children exposed in utero to phenytoin
and carbamazepine monotherapy. JAMA 1994;271:767770.
115 Edlund MJ, Craig TJ: Antipsychotic drug use and birth defects: An epidemiologic reassessment.
Compr Psychiatry 1984;25:3237.
116 Slone D, Siskind V, Heinonen OP, Monson RR, Kaufman DW, Shapiro S: Antenatal exposure to
the phenothiazines in relation to congenital malformations, perinatal mortality rate, birth weight,
and intelligence quotient score. Am J Obstet Gynecol 1977;128:486488.
117 Ayd FJ Jr: Children born of mothers treated with chlorpromazine during pregnancy. Clin Med
1964;71:17581763.
118 Kris EB: Children of mothers maintained on pharmacotherapy during pregnancy and postpartum.
Curr Ther Res Clin Exp 1965;7:785789.
119 Platt JE, Friedhoff AJ, Broman SH, Bond RN, Laska E, Lin SP: Effects of prenatal exposure to
neuroleptic drugs on childrens growth. Neuropsychopharmacology 1988;1:205212.
120 Cohen LS, Rosenbaum JF: Psychotropic drug use during pregnancy: Weighing the risks. J Clin
Psychiatry 1998;59(suppl 2):1828.
121 Patton SW, Misri S, Corral MR, Perry KF, Kuan AJ: Antipsychotic medication during pregnancy
and lactation in women with schizophrenia: Evaluating the risk. Can J Psychiatry 2002;47:
959965.
122 Webb RT, Howard L, Abel KM: Antipsychotic drugs for non-affective psychosis during pregnancy
and postpartum (Cochrane Review). Cochrane Database Syst Rev 2004(2);CD004411.
123 Addis A, Sharabi S, Bonati M: Risk classification systems for drug use during pregnancy: Are
they a reliable source of information? Drug Saf 2000;23:245253.
124 Merlob P, Stahl B: Classification of drugs for teratogenic risk: An anachronistic way of counsel-
ing. Teratology 2002;66:6162.
Meir Steiner, MD, PhD, FRCPC
Professor of Psychiatry and Behavioural Neurosciences and
Professor of Obstetrics and Gynecology, McMaster University
Director, Womens Health Concerns Clinic, St. Josephs Healthcare
Fontbonne Building 6th Floor
50 Charlton Avenue E., Hamilton, ON L8N 4A6 (Canada)
Tel. 1 905 522 1155, ext. 3605, Fax 1 905 521 6098, E-Mail mst@mcmaster.ca
Maternal Depression in the
Postpartum Period: Impact of
Breast-Feeding on Treatment
Planning
Zachary N. Stowe
a,b
, Kimberly Ragan
a
, D. Jeffrey Newport
a
Departments of
a
Psychiatry and Behavioral Sciences, and
b
Gynecology and
Obstetrics, Emory University School of Medicine, Atlanta, Ga., USA
There are numerous prospective, cross-sectional, retrospective case series
and review articles addressing the topic of maternal depression during the
postpartum period. The majority of these articles often utilize the term post-
partum depression. However, the accuracy of this classification is suspect.
Several investigations have demonstrated that depressive symptoms often arise
de novo during pregnancy [13] or represent exacerbations of preexisting
mood disorders. The current diagnostic systems (DSM, ICD) do not require an
individual to be asymptomatic during pregnancy to be classified postpartum.
As such, this introduces considerable confounds to delineate the potentially
unique etiological treatment and the impact of postpartum-onset maternal
depression. This is further complicated by the imprecise media utilization
of the term postpartum when reporting a postnatal tragedy such as infanti-
cide or when a celebrity/public figure publishes an autobiographical account
of her own experience with maternal depression. Whether or not these cases
represent postpartum onset versus exacerbation of a preexisting illness is
not typically discussed. Regardless of the timing of symptom onset, there
are unique issues in the treatment of maternal depression in nursing
women, including: (1) unique hormonal environment; (2) potential exposure to
nonpsychotropic medications that may influence mood (e.g. progestin-only
birth control pills, galactagogues to increase milk production), and
(3) pharmacological treatment options in breast-feeding women. These impor-
tant treatment issues are further complicated by the changing psychosocial
Stowe/Ragan/Newport 138
environment (e.g. career decisions, financial strain, increased interactions with
extended family).
It is beyond the scope of this review to cover the purported benefits of
breast milk for the neonate/infant and extended breast-feeding for the mother [4].
It suffices to note that virtually all professional organizations support breast
milk as the ideal form of nutrition for infants. Over the past decade, the number
of women choosing to breast-feed has been rising [5], and the postpartum
duration of breast-feeding has increased [6]. As such, this review will focus on
the unique treatment issues encountered during breast-feeding with empirical
thoughts on the use of antidepressants in lactation.
Unique Hormonal Environment
There is an overwhelming tendency to view postpartum-onset mental
illness as etiologically derived from the abrupt decline in sex steroids after
parturition. The empirical scientific evidence to support this hypothesis is lim-
ited [7]. However, these previous studies are restricted by the dependence on
peripheral measures of estrogens and progestins with limited access to central
markers of gonadal steroid metabolism and activity.
Furthermore, nursing itself engenders a distinct hormonal milieu that
women do not experience outside of lactation. Elevations in serum prolactin,
fluctuations in oxytocin, and the hypoestrogenic state may have a direct impact
on mental functioning during breast-feeding. There is evidence that symp-
toms of depression may occur after weaning [8]. Empirically, our program has
consistently recommended that women who choose to wean do so by taper-
ing over 710 days to minimize further abrupt changes in the hormonal
environment.
There is limited data on the impact of medications taken by breast-feeding
women on the incidence and/or severity of depression in the postpartum period.
Some women may receive galactagogues to increase breast milk production. It
is noteworthy that common side effects of the most commonly prescribed agent,
metoclopramide (Reglan) [9], include agitation and depression [10]. Similarly,
some women may elect to take oral contraceptives early in the postpartum
period. In breast-feeding women, the progestin-only compounds are the most
commonly prescribed. While there has been considerable debate, conflicting
data, and differing clinical opinions regarding their psychotropic effects, the
Physicians Desk Reference [10] lists mood changes as a side effect of these
agents. Empirically, nonpharmacological interventions for assistance with milk
production and contraception should be first-line options for women at high
risk for depression.
Treating Postpartum Depression while Breast-Feeding 139
Pharmacological Treatment Options in
Breast-Feeding Women
The literature is replete with review articles on the use of antidepressants
during pregnancy and breast-feeding [1117]. To aid in processing the rela-
tively large database on the use of antidepressants in lactating women, division
into several sections is warranted. These sections include: (1) the breadth of the
extant data; (2) quantitative comparisons of infant exposure; (3) documentation
of adverse effects, and (4) the clinical decision.
Antidepressants in Breast-Feeding: The Data
As a class, antidepressants have more published data in breast-feeding than
any other class of medications. A comprehensive literature search identified
86 separate reports including a total of 688 nursing infants exposed to antide-
pressants. Given the interest in this area, there will be additional reports prior to
the release of this review. The current data are listed in table 1.
Clearly, the majority of the reports (65%) and infant exposure cases (85%)
involve the selective serotonin reuptake inhibitors. There have been numerous
attempts to compare the extent of infant exposure and relative safety of individ-
ual antidepressants.
Quantifying Nursing Infant Exposure
The issues of what methodology most accurately defines a nursing infants
exposure (i.e., dose) to a particular antidepressant and how to report such expo-
sure for comparative analyses have generated some debate. Given that no
human study has determined the central nervous system concentration or the
receptor occupancy of a given neurotransmitter system associated with antide-
pressant exposure in nursing infants, only indirect measures are available for
comparing individual medications.
The historical standard for breast-feeding safety, i.e., medicines with an
estimated infant daily dose during lactation that is less than 10% of the mat-
ernal daily dose, was adopted in a recent meta-analysis of antidepressants
in breast-feeding [80]. Yet, this standard is essentially devoid of scientific
justification and to a large extent empiric. Such conclusions are laden with
confounds.
(1) Infant dose estimates are typically based on unreliable extrapolations
from the breast milk to maternal serum ratio (M/S). Our group [45, 51], in
collaboration with others [29], has clearly demonstrated that antidepressant
excretion into breast milk varies from the first portion of breast milk (foremilk)
to the latter portion of breast milk (hindmilk) for sertraline, paroxetine, and
fluoxetine. Similarly, the excretion into human breast milk varies with time
after dose. As such, estimation of infant dose from the M/S ratio based on spot
breast milk sampling is unlikely to be accurate.
(2) The decision to view 10% as clinically meaningful represents an
extension from previous investigations of nonpsychotropic medications. Our
literature search failed to identify an investigation that supported the 10%
versus 10% cutoff as clinically relevant, leading us to conclude that it is
more a psychologically based limit rather than a scientifically established
guideline.
(3) The evidence that a higher M/S ratio reliably predicts higher infant
serum concentrations is sparse, and there is no evidence that higher M/S ratios
have been associated with a greater risk for adverse effects.
These confounds limit the clinical utility of infant dose estimations derived
from breast milk concentrations in random spot samples as the concentrations
Table 1. Data on antidepressant use in breast-feeding
Medication Separate reports Exposed infants References
Selective serotonin reuptake inhibitors
Citalopram 10 120 [1826]
Fluoxetine 15 198 [2539, 49]
Fluvoxamine 7 14 [24, 26, 4044]
Paroxetine 9 90 [24, 26, 37, 4449]
Sertraline 15 162 [25, 26, 37, 44, 4959]
Subtotal 56 584
Tricyclic antidepressants
Amitriptyline 2 3 [60, 61]
Clomipramine 3 8 [49, 61, 62]
Desipramine 1 5
Dothiepin 3 25 [61, 63, 64]
Doxepin 3 3 [6567]
Imipramine 2 6 [49, 61]
Nortriptyline 5 23 [49, 59, 6870]
Subtotal 19 73
Other antidepressants
Bupropion 3 4 [7173]
Nefazodone 2 3 [74, 75]
Trazodone 1 6 [76]
Venlafaxine 5 18 [25, 26, 7779]
Subtotal 11 31
Total 86 688
vary with the portion of milk and time after dose. A recent study with fluoxetine
[29] compared the predictive value of dose estimates derived from M/S ratio,
total breast milk concentration over 24h, and mathematical modeling on nursing
infant serum concentrations. The mathematical modeling (using a gradient
from foremilk to hindmilk and excretion over time) was the best predictor of
infant serum concentrations.
Comparing serum concentrations in nursing infants has emerged over the
past decade as the preferred method for comparing antidepressant exposure via
lactation. This may indeed be a more reliable method of comparison, as con-
founds such as gastrointestinal absorption and infant metabolism are eliminated.
However, other potential confounds limit definitive comparisons.
(1) Nursing infant serum concentrations are low, usually 20ng/ml or
below the limits of detection for high performance liquid chromatography
(2ng/ml). The reliability of clinical laboratory assays in the range typically
encountered is suspect. Research assays with confirmed detection limits are
required. A single study of paroxetine using gas chromatography with mass
spectroscopy [47] demonstrated that infant serum concentrations are often
0.1ng/ml.
(2) Infant serum concentrations are reported in nanograms per milliliter
and do not account for the significant differences in binding affinity [8183]
between particular antidepressants. For example, the binding affinity of parox-
etine for the serotonin transporter is over 20 times greater than that of fluoxe-
tine, 10 times greater than that of citalopram, and 4 times greater than that of
sertraline [83]. Furthermore, estimates of functional exposure should be
extended to other sites such as norepinephrine and dopamine transporters to
provide a comprehensive picture of infant exposure. Functional exposure via
conversion to nanomoles and corrected for the individual medications binding
affinities would provide a rational means, though complex, for comparing the
relative exposures to two or more medications across studies. Such conversions
and analysis have yet to be completed.
(3) The correlation between serum concentration for antidepressants and
central nervous system concentrations for the newer antidepressants is poor. In
summary, infant serum concentrations have advantages over reliance on breast
milk concentrations and the M/S ratio, but the data in their present form (ng/ml)
are not a valid means for comparing antidepressants.
One group has taken another approach to assessing the effects of antide-
pressants in nursing infants [32, 52]. They measured plasma serotonin as an
indicator of central serotonin function in mothers and nursing infants before
and during antidepressant treatment. While the sample sizes are limited, fluoxe-
tine (n 11) and sertraline (n 14), no significant change in plasma serotonin
was demonstrated.
In summary, there have been myriad efforts to define nursing infant expo-
sure to antidepressants and to compare individual medications. A definitive and
widely accepted method remains to be established. Absent such standardized
comparisons, it is important to consider any potential adverse effects reported.
Adverse Effects in Nursing Infants Exposed to Antidepressants
There remains continued concern about antidepressant use during lacta-
tion. The American Academy of Pediatrics Committee on Drugs categorizes
most antidepressants as unknown and may be of concern [84]. However, the
latest report from this committee failed to cite the extant data set. A literature
review identified 11 publications comprised of 40 infants (5.2% of total pub-
lished breast-feeding cases) that noted purportedly adverse effects from antide-
pressant exposure during breast-feeding. These reports are listed in table 2.
An empirically conservative appraisal of these reports is warranted given
alternative feeding methods for infants are readily available. In the absence of
controlled studies, the greatest evidence for causality that can be derived from
the available case literature is that in some instances the infant symptoms were
observed to resolve when breast-feeding was discontinued. In any event, the
preponderance of adverse events consists of colic and other gastrointestinal
disturbances and sleep disruptions. Simply put, the clinician, patient, and fam-
ily must weigh the potential for gastrointestinal distress and difficulty sleeping
versus the benefits of breast-feeding. The principal exception is doxepin, which
in 2 cases appeared to cause clinically significant adverse effects in a nursing
infant.
The Clinical Decision
There is an understandable reluctance on the part of many new mothers
and their clinicians to use antidepressant medication during lactation. This deci-
sion is complicated by the lack of consensus with respect to infant monitoring,
and an international disparity with respect to defining the infants exposure as
discussed above. To date, all antidepressants investigated are found in human
breast milk [85, 86]. The medication concentrations in breast milk (ng/ml)
represent relatively low oral doses to the nursing infant; however, the signifi-
cance of such exposure remains obscure. The breast-feeding baby is exposed to
these medications regardless of whether it can be detected in their sera or not.
Unfortunately, the terminology often utilized in breast-feeding studies can be
misleading and even frightening. For example, some authors have described a
fear of accumulation of fluoxetine in breast-feeding infants [87] based on
2 case reports. Additional reports use the term toxicity to describe effects in
breast-feeding infants of sedation, sleep disturbance, and gastrointestinal
distress, all of which are listed as side effects for the individual medications.
Conversely, some investigators have used the essentially meaningless term
negligible exposure. Until these children have reached adulthood without
having demonstrated any sequelae of exposure, we cannot with certitude char-
acterize any exposure as negligible.
The clinician must take into account the large database on antidepressant
exposure in breast-feeding, the risk of adverse effects related to medication
exposure, the benefits of breast-feeding, and the adverse effects of untreated
maternal depression [88]. It is noteworthy that nursing women often receive
pharmacological treatment for a variety of conditions with limited concerns for
nursing infant exposure to the medication. One of the best examples is the use
of opiate analgesics following delivery. In addition, there has been a recent shift
to support breast-feeding in women treated for epilepsy with medications that
have far less data than antidepressants, and a higher proportion of adverse
effects noted [89]. Where on the clinical care continuum does maternal depression/
anxiety lie?
Our group has posited a detailed model aimed at minimizing infant exposure,
as a guide to clinical decision making during the postpartum period [15, 17]. For
example, if a breast-feeding mother is more likely to consume alcohol, smoke
cigarettes, or take other medications (both prescription and over-the-counter)
when shes depressed or anxious, these exposures should be considered in the
treatment planning. When treatment is indicated, the options for breast-feeding
women are reasonably straightforward: (1) use nonpharmacological treatments;
(2) wean and initiate pharmacological treatment, and (3) continue to breast-feed
and initiate pharmacological treatment. Additional steps to minimize risk for
option (3) include:
(a) Use a medication appropriate for the diagnosis and any comorbid
conditions.
(b) Use a medication of prior response. The postpartum breast-feeding
patient should not experiment with trials of new medications.
(c) Use a medication of prior infant exposure. If a patient has taken a
particular medication during pregnancy (even if discontinued at knowledge of
conception), and it was clinically effective, then the choice for breast-feeding
has already been made. Switching medications is not recommended (e.g. Med A
in pregnancy Med B in lactation no/limited data).
(d) Use a medication with data (i.e., new and improved no/limited data).
(e) Monotherapy at any dose is preferable to introducing a second medica-
tion in women who are breast-feeding.
(f) Infant serum monitoring is not recommended in the clinical setting, and
if adverse effects are suspected suspend breast-feeding.
Table 2. Adverse effects of antidepressant exposure
Medication Exposed Type of study Adverse effects Comments References
infants
Citalopram 1 case report decreased sleep sleep improved with [21]
dose reduction
Fluoxetine 1 case report, agitation, difficulty
Sertraline 2 retrospectively feeding, somnolence,
Paroxetine 1 obtained low muscle tone, hearing
problems, suspected
developmental delays
Fluoxetine 2/14 case series colic possible withdrawal
symptoms from
pregnancy
exposure (n 2)
Fluoxetine 1 case report seizure-like episode neurological testing
at 3 weeks of age, normal, mother also
limp and unresponsive treated with
episode at 4 months, carbamazepine,
peripheral cyanosis buspirone
at 5 1/2 months
Fluoxetine 26 case-controlled growth curve 7.7% of fluoxetine
retrospective significantly below group tobacco
identification, that of infants compared to 2.2%
compared to 38 without medication of comparison
women without exposure group
fluoxetine
exposure
Fluoxetine 1 case report increased irritability [38]
Fluoxetine 1 case report colic, sleep symptoms [39]
disturbance, resolved with
constant crying, cessation of
difficult to breast-feeding,
console, watery and returned
stools, vomiting with resumption
of breast-feeding
Paroxetine 1 case report agitation [37]
Subtotal 37 8 reports
Tricyclic antidepressants
Doxepin 1 case report pale, limp, shallow breast-feeding
breathing, drowsy stopped child
returned to normal
within 24 h
(g) If unsure, get a consultation. There are several web sites devoted to
womens mental health and the reproductive safety of psychotropic medications.
The majority of these links are available through www.emorywomensprogram.org.
Conclusion
The treatment of breast-feeding women will continue to raise important
avenues for scientific investigation. While breast-feeding may complicate treat-
ment, it by no means precludes it. If as a medical community we continue to
advocate for breast-feeding, then we are responsible for understanding the
impact of breast-feeding on medical conditions and their treatments. This man-
dates extending the data on the impact of maternal depression and anxiety on
the constituents of breast milk and understanding the impact on successful
breast-feeding. Similarly, we must extend our understanding of the impact of
breast-feeding on the pharmacokinetics of medications and the impact of med-
ication exposure on nursing infants. It is laudable that psychiatry is well ahead
of most other subspecialties in investigating the reproductive safety of pharma-
cotherapeutic agents during lactation.
Doxepin 1 case report poor sucking and symptoms [65]
swallowing, subsided
muscle hypotonia, within 48 h
vomiting, drowsiness of stopping
breast-feeding
Subtotal 2 2 reports
Other antidepressants
Nefazodone 1 case report drowsy, lethargic, symptoms
unable to maintain resolved
body temperature, within 72 h
difficulty feeding of stopping
in a 9-week-old breast-feeding
born prematurely
at 27 weeks
Subtotal 1 1 report
Total 40
Table 2. (continued)
Medication Exposed Type of study Adverse effects Comments References
infants
References
1 OHara MW, Schlechte JA, Lewis DA, Varner MW: Controlled prospective study of postpartum
mood disorders: Psychological, environmental, and hormonal variables. J Abnorm Psychol
1991;100:6373.
2 Cox JL, Murray D, Chapman D: A controlled study of the onset, duration, and prevalence or post-
natal depression. Br J Psychiatry 1993;163:2731.
3 Stowe ZN, Hostetter A, Newport DJ: The onset of postpartum depression: Implications for identi-
fication in obstetrical and primary care. Am J Obstet Gynecol, in press.
4 Newton ER: Breastmilk: The gold standard (management of breastfeeding). Clin Obstet Gynecol
2004;47:632642.
5 Ryan AS, Wenjun Z, Acosta A: Breastfeeding continues to increase into the new millenium.
Pediatrics 2002;110:11031109.
6 Nichols-Johnson V: Promoting breast feeding as an obstetrician/gynecologist. Clin Obstet
Gynecol 2004;47:624631.
7 Wisner KL, Stowe ZN: Psychobiology of postpartum mood disorders. Semin Reprod Endocrinol
1997;15:7789.
8 Susman VL, Katz JL: Weaning and depression: Another postpartum complication. Am J Psychiatry
1988;145:498501.
9 Gabay MP: Galactagogues: Medications that induce lactation. J Hum Lact 2002;18:274279.
10 Physicians Desk Reference (PDR), ed 58. Thomson PDR, 2004.
11 Cohen LS, Nonacs R, Viguera AC, Reminick A: Diagnosis and treatment of depression during
pregnancy. CNS Spectr 2004;9:209216.
12 Newport DJ, Stowe ZN: Clinical management of perinatal depression: Focus on paroxetine.
Psychopharmacol Bull 2003;37(suppl 1):148166.
13 Nonacs R, Cohen LS: Assessment and treatment of depression during pregnancy: An update.
Psychiatr Clin North Am 2003;26:547562.
14 Hendrick V, Altshuler L: Management of major depression during pregnancy. Am J Psychiatry
2002;159:16671673.
15 Newport DJ, Wilcox MM, Stowe ZN: Antidepressants during pregnancy and lactation: Defining
exposure and treatment issues. Semin Perinatol 2001;25:177190.
16 Spencer JP, Gonzalez LS 3rd, Barnhart DJ: Medications in the breast-feeding mother. Am Fam
Physician 2001;64:119129.
17 Stowe ZN, Calhoun K, Ramsey C, Sadek N, Newport DJ: Mood disorders in pregnancy and lactation:
Defining issues of exposure and treatment. CNS Spectr 2001;6:150166.
18 Lee A, Woo J, Ito S: Frequency of infant adverse events that are associated with citalopram use
during pregnancy. Am J Obstet Gynecol 2004;190:218221.
19 Rampono J, Kristensen JH, Hackett LP, Paech M, Kohan R, Ilet KF: Citalopram and demethylcitalo-
pram in human milk: Distribution, excretion and effects in breast fed infants. Br J Clin Pharmacol
2000;50:263268.
20 Jensen PN, Olesen OV, Bertelsen A, Linnet K: Citalopram and desmethylcitalopram concentra-
tions in breast milk and in serum of mother and infant. Ther Drug Monit 1997;19:236239.
21 Schmidt K, Olesen OV, Jensen PN: Citalopram and breast-feeding: Serum concentration and side
effects in the infant. Biol Psychiatry 2000;47:164165.
22 Heikkinen T, Ekblad U, Kero P, Ekblad S, Laine K: Citalopram in pregnancy and lactation. Clin
Pharmacol Ther 2002;72:184191.
23 Spigset O, Carleborg L, Ohman R, Norstrom A: Excretion of citalopram in breast milk. Br J Clin
Pharmacol 1997;44:295298.
24 Nordeng H, Bergsholm YK, Bohler E, Spigset O: The transfer of selective serotonin uptake
inhibitors to human milk. Tidsskr Nor Laegeforen 2001;121:199203.
25 Berle J, Breilid H, Spigset O, Aamo TO: Breastfeeding and treatment with the SSRIs. Am
Psychiatr Assoc Annu Meet, San Francisco, May 19, 2003.
26 Hendrick V, Smith LM, Hwang S, Altshuler LL, Haynes D: Weight gain in breastfed infants of
mothers taking antidepressant medications. J Clin Psychiatry 2003;64:410412.
27 Moretti ME, Sharma A, Bar-Oz B, Koren G, Ito S: Fluoxetine and its effects on the nursing infant:
A prospective cohort study. Clin Pharmacol Ther 1999;65:141.
28 Kristensen JH, Ilett KF, Hackett LP, Yapp P, Paech M, Begg EJ: Distribution and excretion of
fluoxetine and norfluoxetine in human milk. Br J Clin Pharmacol 1999;48:521527.
29 Suri R, Stowe ZN, Hendrick V, Hostetter A, Widawski M, Altshuler LL: Estimates of nursing
infant daily dose of fluoxetine through breast milk. Biol Psychiatry 2002;52:446451.
30 Hendrick V, Stowe ZN, Altshuler LL, Mintz J, Hwang S, Hostetter A, et al: Fluoxetine and norflu-
oxetine concentrations in nursing infants and breast milk. Biol Psychiatry 2001;50:775782.
31 Taddio A, Ito S, Koren G: Excretion of fluoxetine and its metabolite, norfluoxetine, in human
breast milk. J Clin Pharmacol 1996;36:4247.
32 Epperson CN, Jatlow PI, Czarkowski K, Anderson GM: Maternal fluoxetine treatment in the post-
partum period: Effects on platelet serotonin and plasma drug levels in breastfeeding mother-infant
pairs. Pediatrics 2003;112:e425e429.
33 Yoshida K, Smith B, Craggs M, Kumar RC: Fluoxetine in breast-milk and developmental outcome
of breast-fed infants. Br J Psychiatry 1998;172:175178.
34 Burch KJ, Wells BG: Fluoxetine/norfluoxetine concentrations in human milk. Pediatrics 1992;89:
676677.
35 Chambers CD, Anderson PO, Thomas RG, Dick LM, Felix RJ, Johnson KA, et al: Weight gain in
infants breastfed by mothers who take fluoxetine. Pediatrics 1999;104:e61.
36 Brent NB, Wisner KL: Fluoxetine and carbemazepine concentrations in a nursing mother/infant
pair. Clin Pediatrics 1998;37:4144.
37 Australian Adverse Drug Reactions Advisory Committee: SSRIs and neonatal disorders. Aust
Adverse Drug React Bull 1997, vol 16, No 4.
38 Isenberg KE: Excretion of fluoxetine in human breast milk. J Clin Psychiatry 1990;51:169.
39 Lester BM, Cucca J, Andreozzi L, Flanagan P, Oh W: Possible association between fluoxetine
hydrochloride and colic in an infant. J Am Acad Child Adolesc Psychiatry 1993;32:12531255.
40 Arnold L, Suckow RF, Lichtenstein PK: Fluvoxamine concentrations in breast milk and maternal
and infant sera. J Clin Psychopharmacol 2000;20:491493.
41 Piontek CM, Wisner KL, Perel JM, Peindl KS: Serum fluvoxamine levels in breastfed infants.
42 Kristensen JH, Hackett LP, Kohan R, Paech M, Ilett KF: The amount of fluvoxamine in milk is
unlikely to be a cause of adverse effects in breastfed infants. J Hum Lact 2002;18:139143.
43 Hagg S, Granberg K, Carleborg L: Excretion of fluvoxamine into breast milk. Br J Clin Pharmacol
2000;49:286288.
44 Hendrick V, Fukuchi A, Althsuler LL, Widawski M, Wertheimer A, Brunhuber MV: Use of sertra-
line, paroxetine, and fluvoxamine in nursing women. Br J Psychiatry 2001;179:163166.
45 Stowe ZN, Cohen LS, Hostetter A, Ritchie JC, Owens MJ, Nemeroff CB: Paroxetine in human
breast milk and nursing infants. Am J Psychiatry 2000;157:185189.
46 Ohman R, Hagg S, Carleborg L, Spigset O: Excretion of paroxetine into breast milk. J Clin
Psychiatry 1999;60:519523.
47 Misri S, Kim J, Riggs KW, Kostaras X: Paroxetine levels in postpartum depressed women, breast
milk, and infant serum. J Clin Psychiatry 2000;61:828832.
48 Begg EJ, Dufful SB, Saunders DA, Buttimore RC, Ilett KF, Hackett LP, et al: Paroxetine in human
milk. Br J Clin Pharmacol 1999;48:142147.
49 Birnbaum CS, Cohen LS, Bailey JW, Grush LR, Robertson LM, Stowe ZN: Serum concentrations
of antidepressants and benzodiazepines in nursing infants: A case series. Pediatrics 1999;104:e11.
50 Kristensen JH, Ilett, KF, Dusci LJ, Hackett LP, Yapp P, Wojnar-Horton RE, et al: Distribution and
excretion of sertraline and N-desmethylsertraline in human milk. Br J Clin Pharmacol 1998;45:
453457.
51 Stowe ZN, Owens MJ, Landry JC, Kilts CD, Ely T, Llewellyn A, et al: Sertraline and desmethylser-
traline in human breast milk and nursing infants. Am J Psychiatry 1997;154:12551260.
52 Epperson N, Czarkowski KA, Ward-OBrien D, Weiss E, Gueorguieva R, Jatlow P, et al: Maternal
sertraline treatment and serotonin transport in breast-feeding mother-infant pairs. Am J Psychiatry
2001;158:16311637.
53 Mammen OK, Perel JM, Rudolph G, Foglia JP, Wheeler SB: Sertraline and norsertraline levels in
three breast-fed infants. J Clin Psychiatry 1997;58:100103.
54 Epperson CN, Anderson GM, McDougle CJ: Sertraline and breast-feeding. N Engl J Med
1997;336:11891190.
55 Wisner KL, Perel JM, Blumer J: Serum sertraline and N-desmethylsertraline levels in breast-feeding
mother-infant pairs. Am J Psychiatry 1998;155:690692.
56 Stowe ZN, Hostetter A, Owens MJ, Ritchie JC, Sternberg K, Cohen LS, et al: The pharmacokinet-
ics of sertraline excretion into human breast milk: Determinants of infant serum concentrations.
57 Dodd S, Stocky A, Buist A, Burrows GD, Norman TR: Sertraline analysis in the plasma of breast-fed
infants. Aust NZ J Psychiatry 2001;35:545546.
58 Dodd S, Stocky A, Buist A, Burrows GD, Maguire K, Norman TR: Sertraline in paired blood
plasma and breast-milk samples from nursing mothers. Hum Psychopharmacol 2000;15:261264.
59 Altshuler LL, Burt VK, McMullen M, Hendrick V: Breast-feeding and sertraline: A 24-hour
analysis. J Clin Psychiatry 1995;56:243245.
60 Breyer-Pfaff U, Mill K, Entenmann A, Gaertner HJ: Secretion of amitriptyline and metabolites
into breast milk. Am J Psychiatry 1995;152:812813.
61 Yoshida K, Smith B, Craggs M, Kumar RC: Investigation of pharmacokinetics and of possible
adverse effects in infants exposed to tricyclic antidepressants in breast-milk. J Affect Disord
1997;43:225237.
62 Wisner KL, Perel JM, Foglia JP: Serum clomipramine and metabolite levels in four nursing
mother-infant pairs. J Clin Psychiatry 1995;56:1720.
63 Buist A, Janson H: Effect of exposure to dothiepin and northiaden in breast milk on child devel-
opment. Br J Psychiatry 1995;167:370373.
64 Ilett KF, Lebedevs TH, Wojnar-Horton RE, Yapp P, Roberts MJ, Dusci LJ, et al: The excretion of
dothiepin and its primary metabolites in breast milk. Br J Clin Pharmacol 1992;33:635639.
65 Frey OR, Scheidt P, von Brenndorff AI: Adverse effects in a newborn infant breast-fed by a mother
treated with doxepin. Ann Pharmacother 1999;33:690693.
66 Kemp J, Ilett KF, Booth J, Hackett LP: Excretion of doxepin and N-desmethyldoxepin in human
milk. Br J Clin Pharmacol 1985;20:497499.
67 Matheson I, Pande H, Alertsen AR: Respiratory depression caused by N-desmethyldoxepin in
breast milk. Lancet 1985;2:1124.
68 Wisner KL, Perel JM: Nortriptyline treatment of breast-feeding women. Am J Psychiatry
1996;153:295.
69 Wisner KL, Perel JM, Findling RL, Hinnes RL: Nortriptyline and its hydroxymetabolites in
breast-feeding mothers and newborns. Psychopharmacol Bull 1997;33:249251.
70 Wisner KL, Perel JM: Serum nortriptyline levels in nursing mothers and their infants. Am J
Psychiatry 1991;148:12341236.
71 Haas JS, Kaplan CP, Barenboim D, Jacob P 3rd, Benowitz NL: Bupropion in breast milk: An expo-
sure assessment for potential treatment to prevent post-partum tobacco use. Tob Control
2004;13:5256.
72 Baab SW, Peindl KS, Piontek CM, Wisner KL: Serum bupropion levels in 2 breastfeeding mother-
infant pairs. J Clin Psychiatry 2002;63:910911.
73 Briggs GG, Samson JH, Ambrose PJ, Schroeder DH: Excretion of bupropion in breast milk. Ann
Pharmacother 1993;27:431433.
74 Dodd S, Maguire KP, Burrows GD, Norman TR: Nefazadone in the breast milk of nursing moth-
ers: A report of two patients. J Clin Psychopharmacol 2000;20:717718.
75 Yapp P, Ilett KF, Kristensen JH, Hackett LP, Paech MJ, Rampono J: Drowsiness and poor feeding
in a breast-fed infant: Association with nefazodone and its metabolites. Ann Pharmacother
2000;34:12691272.
76 Verbeeck RK, Ross SG, McKenna EA: Excretion of trazodone in breast milk. Br J Clin Pharmacol
1986;22:367370.
77 Ilett KF, Kristensen JH, Hackett LP, Paech M, Kohan R, Rampono J: Distribution of venlafaxine
and its O-desmethyl metabolite in human milk and their effects in breastfed infants. Br J Clin
Pharmacol 2002;53:1722.
78 Ilett KF, Hackett LP, Dusci LJ, Roberts MJ, Kristensen JH, Paech MJ, et al: Distribution and excre-
tion of venlafaxine and O-desmethylvenlafaxine in human milk. Br J Clin Pharmacol 1998;45:
459462.
79 Hendrick V, Altshuler L, Wertheimer A, Dunn WA: Venlafaxine and breast-feeding. Am J
Psychiatry 2001;158:20892090.
80 Weissman AM, Levy BT, Hartz AJ, Bentler S, Donohue M, Ellingrod VL, et al: Pooled analysis of
antidepressant levels in lactating mothers, breast milk, and nursing infants. Am J Psychiatry
2004;161:10661078.
81 DeVane CL: Translational pharmacokinetics: Current issues with newer antidepressants. Depress
Anxiety 1998;8(suppl 1):6470.
82 Taylor JE, Richelson E: High affinity binding of tricyclic antidepressants to histamine H
1
-receptors:
Fact and artifact. Eur J Pharmacol 1980;67:4146.
83 Owens MJ, Morgan WN, Plott SJ, Nemeroff CB: Neurotransmitter receptor and transporter bind-
ing profile of antidepressants and their metabolites. J Pharmacol Exp Ther 1997;283:13051322.
84 American Academy of Pediatrics Committee on Drugs: The transfer of drugs and other chemicals
into human milk. Pediatrics 1994;93:137150.
85 Burt VK, Suri R, Altshuler L, Stowe Z, Hendrick VC, Muntean E: The use of psychotropic med-
ications during breast feeding. Am J Psychiatry 2001;158:10011009.
86 Llewellyn A, Stowe ZN: Psychotropic medications in lactation. J Clin Psychiatry 1998;59(suppl
2):4152.
87 Pons G, Rey W, Matheson I: Excretion of psychoactive drugs into breast milk. Clin Pharmacokinet
1994;27:270289.
88 Beach A, Henry A, Stowe ZN, Newport DJ: Maternal depression: A childs earliest life event. Bibl
Psychiatr, in press.
89 Ito S: Drug therapy for breast-feeding women. N Engl J Med 2000;343:118126.
Zachary N. Stowe, MD
Department of Psychiatry and Behavioral Sciences, Emory University School of Medicine
1365 Clifton Road NE, Suite B6100
Atlanta, GA 30322 (USA)
Tel. 1 404 778 2524, Fax 1 404 778 2535, E-Mail zstowe@emory.edu
The Use of Interpersonal
Psychotherapy for Perinatal
Mood and Anxiety Disorders
Scott Stuart
ac
, Michael W. OHara
b,c
a
Department of Psychiatry,
b
Department of Psychology, and
c
Iowa Depression and Clinical Research Center, University of Iowa,
Iowa City, Iowa, USA
Pregnancy and the puerperium are life transitions writ large. Women and
their families undergo major changes in relationships and role expectations.
Many women are faced with decisions about returning to work or staying at
home with their newborn. Conflicts often arise regarding child care, parenting,
or sexual intimacy during pregnancy and the postpartum period. In short, the
perinatal period is an interpersonal event that is characterized by relationship
changes.
Perinatal depression and anxiety are associated with difficulties in negoti-
ating these interpersonal changes. Interpersonal Psychotherapy (IPT) [1, 2] is a
well-defined treatment directed specifically at these interpersonal problems as
well as the symptoms that women experience while depressed or anxious. It has
proven to be an efficacious treatment for depression during pregnancy and the
postpartum period, and because of its intuitive appeal to perinatal women, it has
proven to be a highly acceptable and relevant treatment as well.
Perinatal Social Support: Impact on Depression
and Anxiety
One of the most consistently replicated findings regarding perinatal
depression is its association with poor social support [3]. Bernazzani et al. [4]
demonstrated that levels of antenatal depression are correlated with satisfac-
tion with interpersonal relationships. They also found that the impact of poor
IPT for Perinatal Depression 151
interpersonal relationships on the risk for the development of depression was
nearly as significant as having had previous psychiatric problems. Support from
a womans spouse or significant other [5] and the degree of intimacy in their
relationship [6] appear to be particularly important.
The role of social support is also critical during the postpartum period.
Support from spouse or partner appears to be particularly important [7].
Conflicts regarding child care responsibilities are common, as are disagree-
ments regarding the degree of intimacy in the couples relationship. This often
involves aspects of the couples sexual relationship, but even more frequently
reflects dissatisfaction on the part of both partners with the degree of commu-
nication and connection. Men may describe feeling displaced by the newborn,
while women may describe feeling emotionally distanced, with little energy to
give to the marital relationship. In all of these circumstances, problems in the
spousal relationship can contribute to or exacerbate depression, anxiety, and
poor psychological adjustment.
During the puerperium, women often look to mentors for both emotional
and practical support. Mothers of postpartum women often fill this role, though
other experienced female relatives or friends may substitute. The absence of
such a mentor can contribute to poor psychosocial adjustment; this absence
need not be physical but can also reflect the inability of important mentoring
figures to meet the expectations of the perinatal woman. For instance, a woman
may expect that her mother will provide practical physical support and emo-
tional support postpartum, only to find that her mother is unwilling to do so, or
is critical rather than supportive.
Poor social support in general is also associated with postpartum psychi-
atric problems [7]. Women often rely on others for practical help with child
care, and may use their social support network as a source of advice and feed-
back about caring for a newborn. Social isolation can contribute greatly to
psychological distress, making social connections and relationships extremely
important in the perinatal period. The ability of a womans social support
network to respond to her new life circumstances, and the sustainability of her
social connections are particularly crucial. For instance, being able to maintain
important social support from work colleagues while on maternity leave may be
important.
Other social factors, such as having an unplanned pregnancy, have also
been linked to an increased risk for postpartum depression [8]. Women from
lower social classes and with lower incomes are at higher risk [7].
In sum, social support and psychosocial factors are clearly associated
with perinatal distress and psychiatric problems. The biopsychosocial model
supports the role of these factors in the context of the physical changes that
occur in the perinatal period. A combination of biological vulnerabilities and
Stuart/OHara 152
inadequate social support met with the physical, emotional, and social changes
occurring in pregnancy and the postpartum period form the diatheses and stressors
which lead to depression or anxiety. All of these data suggest that psychosocial
interventions directed towards social support and specific interpersonal problems
should be effective in alleviating perinatal distress.
Interpersonal Psychotherapy
IPT [1, 2] is a time-limited, dynamically informed psychotherapy which
aims to alleviate patients suffering and improve their interpersonal functioning.
IPT focuses specifically on interpersonal relationships as a means of bringing
about change, with the goal of helping patients to either improve their interper-
sonal relationships or change their expectations about them. In addition, the
treatment also aims to assist patients in improving their social support networks
so that they can better manage their current interpersonal distress. Both of these
goals are highly relevant to women with perinatal mood and anxiety disorders,
particularly because such problems are associated with perinatal depression.
IPT was originally developed in a research context as a treatment for major
depression and was codified in a manual developed by Klerman et al. in 1984 [2].
Since that time, a great deal of empirical evidence supporting its use has
accumulated (for a comprehensive review, see Stuart and Robertson [1]). In
addition, as clinical experience with the treatment has increased, the use of IPT
has broadened to include not only the treatment of patients with a variety of
well-specified diagnoses as described in DSM-IV [9], but also the treatment of
patients presenting with a variety of interpersonal problems. Its application to
perinatal women, therefore, should be considered not only when there is a
clearly defined affective or anxiety disorder, but also when there is interper-
sonal distress to a degree that a woman wishes treatment.
IPT is based on a biopsychosocial model of psychological functioning [10],
which is consistent with the prevailing view of the perinatal period as one
involving physical and hormonal changes, interpersonal changes, and social
changes. Rather than narrowly viewing psychological distress or psychiatric
problems as medical problems, the IPT approach is to view the patients
functioning in broad terms as a product of her temperament, personality, and
attachment style, based upon a foundation of biological factors such as genetics
and physiological functioning, placed in the context of social relationships and
broad social support. This model is particularly relevant for perinatal women
because it incorporates the obvious biological changes of pregnancy and the
postpartum period into the context of the interpersonal event of childbirth. In
doing so, there is acknowledgment that even though IPT is directed towards
relationship issues and communication, the use of medication and other
somatic treatments may be of benefit and may be provided in conjunction
with IPT.
IPT is based on the premise that interpersonal distress is intimately
connected with psychological symptoms. In the case of perinatal depression,
this is a direct extension of the data regarding the impact of social factors
during the puerperium. The foci of IPT treatment are twofold. One focus is the
conflicts and transitions in relationships in which the perinatal woman is
engaged: the aim is to help her to either improve communication within those
relationships, or to change her expectations about those relationships. The
second focus is helping the perinatal woman to build or better utilize her
extended social support network so that she is better able to muster the inter-
personal support needed to deal with the crises which precipitated the distress,
e.g. to obtain more effective support for the perinatal transition.
This approach is extremely well suited to the treatment of perinatal depres-
sion [11, 12]. Distress which is linked to conflicts with partners or to difficul-
ties in making the transition from working woman to mother can be directly
addressed. A therapist using IPT would help the patient to resolve the conflicts
with her partner over issues such as division of child care labor, and would also
assist the woman to garner more support from her social network. This might
include connecting with and asking for support from other friends who have
had children, from extended family members, or colleagues at work. It could
also include encouraging the patient to get involved in a new mothers support
group. Resolution of the particular interpersonal conflict, along with improved
interpersonal support while the role transition is being negotiated then leads to
symptomatic improvement.
IPT is grounded in the interpersonal theories of Sullivan [13] and the
attachment theories described by Bowlby [14] among others. These theories
assert that humans have an innate biological drive to develop and establish
interpersonal relationships, and that the principal feature of mental health is the
capacity to form flexible attachments. In other words, mentally healthy individ-
uals are able to develop relatively secure relationships in which they can depend
on others to provide support and in which they are emotionally available to oth-
ers. In contrast, individuals with less flexible styles of attachment, such as those
with dependent or avoidant interpersonal styles, typically have relationships
which are less satisfying and are less adaptable to stress.
Attachment theorists further hypothesize that attachment styles develop
primarily during childhood, and are a result of both genetically based tempera-
ment and environmental influences (usually the accumulated experiences with
primary caregivers) [15]. Though these attachment styles tend to persist, they
are not fixed, and can be modified either positively or negatively as a result of
Stuart/OHara 154
additional interpersonal interactions. In general, however, an individuals
pattern of attachment tends to be consistent both within relationships and across
relationships, i.e., people tend to persist in their styles of relating to individuals
whom they have known for some time, and tend to replicate their characte-
ristic attachment styles with others with whom they have recently developed
relationships.
Psychiatric disorders are hypothesized to result from a combination of
interpersonal and genetic factors following a biopsychosocial model [1].
Genetic vulnerability can be modified either positively by healthy attachment
experiences or negatively by poor interpersonal experiences. Those patients
with less secure attachments, who are unable to constructively request and
receive psychological support in times of crises, have an increased vulnerability
to the development of psychopathology. In addition, circumstances in which
an individuals primary attachments are disrupted, such as death, divorce, or
illness, also tend to increase the chance that a psychiatric disorder will occur in
a vulnerable individual.
This model has direct application to the treatment of perinatal women.
Such women who may have less secure attachment styles are at greater risk for
depression or anxiety during the puerperium, both because they are more
vulnerable to the interpersonal changes which accompany childbirth and
because they have less extensive social support networks. Women are at addi-
tional risk if there are other adverse stressors, such as early delivery, previous
perinatal loss, or medical problems with their infant. In such circumstances,
even securely attached women may develop depression or anxiety disorders,
given the magnitude of the stressor.
IPT is therefore designed to treat psychiatric symptoms and to improve
interpersonal functioning by increasing effective social support to meet the
womans attachment needs more effectively and by focusing specifically on
patients primary interpersonal relationships, particularly in the problem areas
of grief and loss, interpersonal disputes, role transitions, and interpersonal
sensitivities [1]. Though fundamental change in either personality or attach-
ment style is unlikely during short-term treatment, symptom resolution is made
possible when patients are assisted in repairing their disrupted interpersonal
relationships and when they learn new ways to communicate their need for
emotional support.
The efficacy of IPT in the acute treatment of depression has been investi-
gated in a number of studies [1], the most notable of which is the National
Institute of Mental Health Treatment of Depression Collaborative Research
Program [16]. In this landmark study, IPT was found to be equal to both
imipramine and cognitive behavior therapy (CBT) in the treatment of mild to
moderate depression [17]. IPT has also been found to be effective in preventing
recurrence of depression for patients who have had at least three previous
episodes [18]. Our own laboratory has been active in developing and testing
modified forms of IPT for use with several populations in addition to perinatal
women [11, 12, 19]. Included among these are patients with depression follow-
ing myocardial infarction [20], social phobia [21], and somatization disorder
[21, 22].
Psychosocial Treatment of Perinatal Mood and
Anxiety Disorders: A Review of the Literature
Psychosocial treatments for perinatal depression can be placed in two
categories. The first include preventive interventions which involve treatment
of either large numbers of women from the general population, or involve only
those who are at high risk for postpartum psychopathology. These prevention
trials often begin during pregnancy. The second type of psychosocial intervention
is designed to treat perinatal women who have already developed depression or
anxiety.
The interventions can also be characterized on a spectrum from nonspecific
to targeted. Nonspecific interventions often consist of supportive listening,
empathy, and positive reinforcement [23], elements which are fundamental for
all psychotherapies, but which, though essential, may not be sufficient to treat
more severe psychiatric problems. In addition to these basic elements, targeted
interventions rely on specific techniques such as communication analysis
or cognitive restructuring which are ultimately directed towards symptom
resolution.
The literature regarding treatment of perinatal depression and anxiety is
strongly skewed towards depression and postpartum depression specifically. At
present, there are no controlled treatment trials of either medication or
psychotherapy for anxiety disorders either during pregnancy or the postpartum
period. This is obviously a huge gap, particularly as perinatal anxiety disorders
are known to be prevalent and disabling [2426]. Moreover, there are no con-
trolled psychopharmacologic treatment trials for depression during pregnancy,
and only three psychopharmacologic trials of any kind exist for the treatment of
postpartum depression [2729]. The majority of the empirical literature
involves the use of psychosocial interventions for the prevention or treatment of
postpartum depression; these studies are reviewed below.
Treatment of Postpartum Depression Prevention Treatment Trials
Preventive interventions for postpartum depression have been applied to
high-risk women as well as pregnant women in the general population who may
Stuart/OHara 156
not be at risk. Preventive interventions are divergent across a number of dimen-
sions including the type of professional conducting the intervention (e.g. men-
tal health vs. nonmental health professionals), individual versus group
treatment, the type of interventions provided (e.g. psychoeducation vs. psy-
chotherapy), number of sessions provided, and the timing of the intervention
(e.g. prenatal only vs. pre- and postnatal sessions).
The nondirective approaches to prevention include treatments such as
psychoeducational classes [30, 31] and relaxation [32]. Midwife debriefing,
which consists of psychoeducation about the birthing process delivered by a
midwife, has also been described [3335]. These interventions have great appeal
because they require no specialized training for treatment providers; in addition,
it might well be argued that such psychoeducation and debriefing should be
included routinely in perinatal care as good medical practice. They do not, how-
ever, appear to be effective in preventing episodes of postpartum depression.
In contrast to the provision of psychoeducational information, Wolman
et al. [36] conducted a treatment trial in which 189 nulliparous women were ran-
domly assigned to be accompanied by a doula during their delivery, or were
assigned to a group which received only standard medical care. Women with
doula support had a mean score of 10.4 on the Pitt Depression Questionnaire
[37] at 6 weeks postpartum, which was significantly better than the control
group (mean 23.3). The results of the study by Wolman et al. [36], however,
are limited by several factors. First, depressive symptoms were not assessed
during pregnancy, so that the effects of the intervention on reducing depression
are not clear. Additionally, women continued to report moderate levels of depres-
sion even in the experimental group. This study does, however, support the
hypothesis that social support is helpful in preventing postpartum depression.
More traditional psychotherapeutic interventions, designed to include
Rogerian factors such as a healing relationship, positive regard, and positive
reinforcement [23], have been conducted by several investigators [3840].
Brugha et al. [38] conducted a randomized prevention trial with 190 antenatal
women at risk. The intervention consisted of six 2-hour group sessions during
pregnancy with an additional session at 8 weeks postpartum. The sessions were
educational and focused on social support and problem-solving approaches to
avoid postpartum depression.
The intervention had no effect with respect to depressive symptomatology or
diagnostic status at 3 months postpartum. Roughly 55% of women completing
the outcome assessment at 3 months postpartum were not compliant in attending
sessions, a problem noted frequently in the prevention literature. It may be that
women without acute symptoms are not highly motivated to engage in treatment.
Oakley et al. [39] studied 509 women who had a history of at least 1 low-
birth-weight delivery. The women were randomly assigned to receive supportive
counseling from midwives or routine antenatal care. The midwife intervention
package consisted of a minimum of 3 home visits plus 2 telephone contacts
during pregnancy. The first contact usually occurred at about 34 months of
pregnancy. Midwives did not provide any clinical care, and were instructed to
give advice or information to women only if specifically asked to do so. They
were also instructed to assist women with referrals to other social agencies and
other health professionals as needed 80% of the women in the intervention
group received such referrals. At 1 year, 95% of the women in the midwife
group, and 90% of the women in the control group reported no postpartum
depression, a nonsignificant difference.
Elliott et al. [40] reported on the preventive effects of group treatment
provided from early in pregnancy to 6 months postpartum. Women were
selected for treatment if they had: (1) a previous psychiatric history; (2) high
levels of anxiety; (3) a poor marital relationship, and (4) no confidant. The
authors described the psychoeducational component of the treatment as antici-
patory guidance aimed at helping women to anticipate changes that would
occur after childbirth, as well as providing practical advice on how to avoid
potential problems. Therapists provided specific information about postpartum
depression and the need to establish adequate social supports.
Both first- and second-time mothers were treated, though in separate
groups. First-time mothers attended an average of 7 of 11 sessions, while
second-time mothers attended an average of only 4 of 11 meetings. There were
significant differences in depressive symptomatology only with primiparous
women at 3 months postpartum. Median Edinburgh Postnatal Depression Scale
(EPDS) scores for treated and control subjects were 3.0 and 8.0, respectively.
There have been 2 prevention trials which have targeted social support and
interpersonal relationships specifically. Both were based on the interpersonal
model utilized in IPT, though neither included all of the elements of a full course
of IPT. Gorman [41] randomly assigned high-risk women to receive a modified
and abbreviated course of IPT (n 24) or to a no treatment control group
(n 21). The intervention consisted of 2 individual sessions during pregnancy
and 3 weekly sessions between 2 and 4 weeks postpartum. The pregnancy
sessions consisted of psychoeducation about postpartum mood disorders and
discussion related to current or anticipated interpersonal difficulties. The postpar-
tum sessions focused on the womans mood and how it was associated with the
interpersonal issues discussed during pregnancy. At 1 month postpartum, women
receiving the intervention were significantly less likely to have experienced a
major depression compared to women in the no treatment control group (0 vs. 25%).
At 6 months postpartum, this difference was not significant (15 vs. 23.5%).
Zlotnick et al. [42] randomized 37 women at risk for postpartum depres-
sion to either 4 group sessions of interpersonally oriented psychoeducation
Stuart/OHara 158
(n 18) or to a control group (n 19). The group intervention included
46 women and involved four 60 -min group sessions over a 4-week period. The
first session provided psychoeducation about the baby blues and postpartum
depression; the remaining 3 focused on the IPT problem areas of role transi-
tions and interpersonal conflict along with goal setting. At 3 months postpar-
tum, none of the women in the intervention group had developed postpartum
depression compared to 33% in the control group, a significant difference. The
mean Beck Depression Inventory (BDI) score for the intervention group at
3 months postpartum was 8.4 and it was 11.3 for the control group.
The treatment trials designed to prevent postpartum depression have gen-
erally not demonstrated marked effects. However, there is some indication that
the more directed treatments, particularly those which target interpersonal
issues and social support, may be of benefit. The effect of these directed treat-
ments may be even greater if they are provided to women at risk for developing
postpartum depression, rather than towards perinatal women in general.
Prevention research is needed in a variety of areas. First, the determination
of risk for postpartum depression is unclear. It seems clear from the data that
intensive preventive interventions should not be used for women who are not at
risk for postpartum depression. However, the factors that should be used to
determine risk, and the extent of risk which makes preventive treatment viable
from a cost-benefit standpoint are yet to be determined. In addition, the risk
factors which are associated with likely response to a particular psychosocial
intervention are not clear. Reasonable hypotheses, for example, might be that
women with psychosocial risk factors such as poverty might benefit from
directive interventions designed to assist them to obtain practical financial sup-
port; those that face relationship conflicts may benefit more from an interper-
sonally based treatment, and those with biological risk factors such as a family
history of depression from a structured psychotherapy or medication.
In sum, there is some evidence that the targeted preventive interventions
studied to date are effective in preventing postpartum depression. The empirical
data suggest that preventive interventions should be reconceptualized using the
following guidelines: (1) the interventions should be directed towards high-risk
women; (2) the risk factors for postpartum depression which are also associated
with response to treatment need to be better characterized, and (3) the treat-
ments should be based on clear rationales which direct the interventions, such
as a focus on interpersonal issues and social support.
Treatment of Postpartum Depression Acute Treatment Trials
Several acute interventions have been designed to treat the symptoms
of postpartum depression. In contrast to preventive interventions, all of the
women who receive acute treatment are specifically selected because they are
experiencing postpartum depression. These interventions can also be character-
ized along a spectrum from those which are nondirective to those which target
specific symptoms.
The nondirective treatment trials include those conducted by Holden
et al. [43], and Wickberg and Hwang [45]. Holden et al. [43] conducted a study
involving 50 women who met Research Diagnostic Criteria (RDC) [44] for
either major or minor depression at 12 weeks postpartum. The women were
divided into two groups those who received 8 weekly sessions of counseling
provided by health visitors most of whom had midwife experience and several
of whom had additional psychiatric training and those in a no treatment
control group. The counseling was designed to be nondirective in orientation,
and patients received a mean of 8.8 sessions of therapy over about 13 weeks.
Twelve of the women in the study were receiving antidepressant medication in
addition to the counseling.
A significantly higher percentage of women in the treatment group (69%)
no longer met RDC for either major or minor depression at the conclusion of
treatment compared with the women in the control condition (38%).
Additionally, the median score on the EPDS decreased from 15.5 to 12.0 in the
control group compared to 16.0 at intake and 10.5 following treatment in the
group receiving counseling, a significant change only for the treatment group.
The outcome scores, however, were compared only within each group, rather
than between the two groups.
Wickberg and Hwang [45] studied the effects of nondirective counseling
provided by child health nurses. Women were recruited during routine visits to
a Child Health Care Clinic and were eligible for randomization if they had
scores of 12 or more on the EPDS at both 2 and 3 months postpartum, had
Montgomery-Asberg Depression Rating Scale (MADRS) [46] scores of 10 or
more, and met DSM-III-R [47] criteria for major depression. The treated group
received 6 weekly nondirective counseling sessions, and the control group
received routine care.
Twelve of 15 women in the treatment group compared to 4 of 16 women in
the control group no longer met criteria for DSM-III-R major depression at the
completion of treatment. The decrease in MADRS scores in the group receiving
counseling (19.610.9) was significantly greater than in the control group
(17.114.7). The authors reported that 4 severely ill women had to be dropped
from the study, which limits the findings to postpartum women with mild to
moderate depression.
Cooper and Murray [48] and Cooper et al. [49] also used nondirective
counseling in their study of postpartum depression, but compared it to 8 weeks
of cognitively oriented therapy, psychodynamically oriented psychotherapy,
and a no treatment control group. Routine care included care from general
Stuart/OHara 160
practitioners and health visitors, though the health visitors received no addi-
tional psychological training. The nondirective counseling was conducted by
health visitors, and followed the model described by Holden et al. [43] above, in
which the health visitors received specific training.
The cognitively oriented treatment was a modified form of the interaction
guidance treatment described by McDonough [50]. Bearing little resemblance
to the CBT described by Beck et al. [51], this therapy was described by the
authors as directed towards problems identified by the mother in the manage-
ment of her infant, as well as observed problems in the quality of the mother-
infant interaction. The therapist often provided advice about infant care. The
authors note that the treatment was not directed primarily at the maternal
depression.
The psychodynamically oriented psychotherapy followed techniques
described by Cramer et al. [52], in which an understanding of the mothers
representation of her infant and her relationship with her infant was promoted
by exploring aspects of the mothers own attachment history [49]. There was no
directed focus upon the postpartum depression.
A total of 193 primiparous depressed women screened from the commu-
nity using the EPDS were randomized to 1 of the 4 treatments. All of the treat-
ments produced improvement in EPDS scores relative to the control, but only
the psychodynamically oriented intervention reduced the rates of depression
relative to the control. Remission occurred in 54% of the women receiving
nondirective counseling, 57% of the women receiving cognitively oriented
counseling, 71% of the women treated with psychodynamically oriented
psychotherapy, and 40% of the women in the control group. There were no
differences between any of the groups at 9 months postpartum, and none of the
treatments reduced the rate of subsequent postpartum episodes of depression.
EPDS scores at 4.5 months ranged from 8.9 to 9.9 in the treatment groups
compared to 11.3 in the control group. No information was provided regarding
the EPDS scores at the beginning of the study, though study criteria required a
score of only 12 for entry. Based on the limited degree of reported change on
the EPDS, it appears that most women were only mildly depressed.
Appleby et al. [29] reported on the treatment of postpartum depression
using counseling based loosely on the CBT model. The counseling was based
on the nondirective model used by health visitors in the British system, with the
addition of some techniques from cognitive therapy. Each counseling session
was structured to provide reassurance to mothers and to offer practical advice in
four areas: feelings of difficulty in coping, lack of pleasurable activities, lack of
practical support, and caring for any older children. The first session lasted
about an hour, and subsequent visits lasted about 30min and were provided
once every 2 weeks for women who received 6 sessions of counseling.
Eighty-seven women were randomized to treatment if they scored above
10 on the EPDS and met RDC for major or minor depressive disorder at
68 weeks postpartum. The interventions included: (1) treatment with fluoxe-
tine and 1 session of counseling; (2) treatment with fluoxetine and 6 sessions
of counseling; (3) treatment with placebo and 1 session of counseling, and
(4) treatment with placebo and 6 sessions of counseling. The medication
arm of the study was double-blinded, and the total duration of the study was
12 weeks.
Significant improvements were seen in all 4 of the treatment groups, and
fluoxetine was superior to placebo on all measures. Six sessions of counseling
were superior to 1 session only on the Hamilton Rating Scale for Depression
(HRSD). There were no interaction effects between medication and psychother-
apeutic treatment. Women completing the fluoxetine plus 1 session of counsel-
ing treatment had a decline in their HRSD scores from 13.3 to 2.9, and ratings
of the women in the fluoxetine plus 6 sessions group declined from 13.2 to 2.8.
The HRSD scores of women receiving placebo and 1 session declined from
14.7 to 7.5 and from 13.3 to 3.7 if they received placebo and 6 counseling
sessions.
Both the studies by Cooper et al. [49] and Appleby et al. [29] are notable in
several respects. First, neither utilized treatments targeted specifically at post-
partum depression. The active counseling in Appleby et al. somewhat resem-
bled cognitive therapy, but was oriented largely towards the provision of
practical advice. In addition, it was delivered less intensively than standard
CBT both session duration and length of treatment were substantially
reduced. Second, most of the women in both studies were mildly depressed. In
addition, the magnitude of absolute change in depressive symptoms was rela-
tively small. In sum, both studies support the hypothesis that women with mild
depressive symptoms are likely to benefit from nonspecific treatments, but that
such treatments are likely not to be effective with women with moderate to
severe depression.
OHara et al. [11] evaluated the use of IPT in a sample of 120 postpartum
women from the community who met criteria for DSM-IV [9] major depres-
sion. Depressed women were assigned to either 12 weekly sessions of IPT or to
wait for 12 weeks before receiving weekly IPT. Therapists were PhD clinical
and counseling psychologists in clinical practice who were specifically trained
as IPT therapists, and the treatment was specifically directed at depressive
symptoms [1, 2, 12].
Treatment with IPT was found to be significantly superior as measured by
both the HRSD and BDI. Average HRSD scores for the IPT condition dropped
from 18.4 to 8.3 and in the waiting condition scores dropped from 19.8 to 16.8.
Similar changes were noted for the BDI scores, which decreased from 23.6 to
Stuart/OHara 162
10.6 in the IPT group and from 23.0 to 19.2 in the control group. For both the
HRSD and the BDI significant differences emerged by 4 weeks into therapy. At
the conclusion of treatment, significantly fewer women in the IPT group met
criteria for a major depressive episode (12.5%) compared to women in the con-
trol group (68.6%). Significant effects were also observed for several of the
measures of social adjustment, including the Social Adjustment Scale [53] and
the Postpartum Adjustment Questionnaire [54]. Women with moderate and
severe levels of depression responded as well to treatment as women with mild
depression.
The empirical data regarding the treatment of acute postpartum depression
are similar to the prevention data. Those treatments which are more specifically
directed towards interpersonal issues and social support appear to have the
greatest effects on depression. Nonspecific treatments do appear to be of help
for mild depression, but their effect on moderate to severe depression is less
impressive.
Conversely, the nonspecific treatments tend to be less expensive, and are
likely to be more readily available than the directed treatments. Though gener-
ally much easier to learn than CBT or psychodynamic psychotherapy, IPT does
require advanced training. In contrast, the nonspecific counseling used in the
British model can be delivered by health visitors. The ability of these providers
to deliver nonspecific treatments also makes them more feasible.
Taken together, these conclusions suggest that strong consideration ought
to be given to a sequential program of treatment for perinatal depression.
Preventive treatments are likely to be of benefit to those women at high risk for
postpartum depression, but are not likely to be cost-effective when delivered to
unselected groups of women. Acute treatment might begin with nonspecific
counseling delivered by health visitors or other similar personnel; women who
do not respond to such counseling could then be referred to more intensive
directed psychotherapeutic treatments. This would provide quality care while
minimizing cost and maximizing feasibility of treatment.
Though there are very little data regarding the treatment of perinatal
anxiety, the same is likely to be true for women with anxiety disorders. Sup-
portive treatment and reassurance may be of help for mild symptoms, while
more specifically directed treatments are likely to be of greater benefit for
those women with moderate to severe disorders. In the case of anxiety, it is
possible that CBT [55] may be a more plausible treatment than IPT, as the
direct link between anxiety and social support is not as well established as it is
in depression. In contrast, CBT is directed specifically towards anxiety symp-
toms, and is at present an established psychosocial treatment for anxiety disor-
ders. This clinically intuitive treatment selection remains to be empirically
tested.
Treatment of Depression during Pregnancy: Acute Treatment Trials
There are only 2 studies which have investigated the use of psychotherapy
of any sort for depression during pregnancy, both of which used IPT. An addi-
tional large-scale study of IPT is currently under way at the University of Iowa.
There are no medication treatment trials for depression during pregnancy, nor
are there any treatment trials with pregnant women with anxiety disorders.
Spinelli and Endicott [56] conducted a treatment trial comparing the out-
comes of 17 depressed pregnant women who received 16 weeks of parenting
education with 21 who received 16 weekly IPT sessions. All participants met
DSM-IV criteria for major depressive episode and had HRSD scores 12.
Assessments and interventions were conducted in either English or Spanish.
The IPT intervention relative to parenting education resulted in significant
improvement indexed by the EPDS, the BDI, and the HRSD. Moreover, a sig-
nificantly greater proportion of women receiving IPT (60%) relative to the
women receiving parenting education (15%) evidenced recovery on the Clinical
Global Improvement Scale. Recovery based on the HRSD was not different
across groups. It is notable that IPT did show a significant advantage over a
presumably credible psychosocial intervention for depressed pregnant women.
Moreover, these findings were obtained with a relatively impoverished popula-
tion of women who spoke Spanish as well as English. Limitations include the
fact that 24% of women initially randomized to treatment did not attend even
1 session and the study had a relatively small sample size. It does need to be
replicated, but the results are quite encouraging.
Stuart and OHara [12] conducted an open pilot study which included
7 depressed pregnant women who were treated with 12 weeks of IPT. At intake,
all met DSM-IV [9] criteria for a major depressive disorder. None of the women
took medication during the entire study, and all completed 12 weeks of IPT.
Outcome data revealed highly significant improvements in both BDI scores and
in HRSD scores [57]. BDI scores decreased from an average of 28.1 to 3.2
(p 0.001), and mean HRSD scores decreased from 19.8 to 8.4 (p 0.001).
Only 1 of the 7 women met criteria for major depression at the completion of
treatment. While the results are impressive, the scale of this open trial is too
limited to draw meaningful conclusions.
The results these 2 studies suggest that IPT has utility for the treatment of
depressed pregnant women. Much more research is needed, both regarding the
use of IPT and other psychosocial interventions. However, given the absolute
lack of data regarding other treatments, IPT stands at present as the only empir-
ically tested treatment for depression during pregnancy. It is likely that, as is the
case with postpartum depression, specifically targeted psychotherapeutic treat-
ments such as IPT are more effective for moderate to severe depression and
anxiety than are nonspecific approaches.
Stuart/OHara 164
Conclusion
At present, empirical data clearly support the use of acute interventions for
postpartum depression. The literature regarding both primary prevention and
acute treatment suggests that treatment which specifically targets depressive
symptoms is likely to be the most effective. Epidemiologic research also
suggests that psychosocial interventions should target factors associated with
perinatal depression, namely social support and interpersonal difficulties.
A number of additional factors warrant further investigation. First, the effi-
cacy studies of the treatment of acute postpartum depression suggest that there
may be benefit to a stepwise approach to treatment. Nondirective counseling
provided by public health nurses appears to be of benefit for women with mild
depression, while more intensive interventions such IPT are needed for moder-
ate to severe depression. A reasonable approach would be to utilize less costly
interventions as a first-line treatment for depressed perinatal women, with those
that do not respond to a nondirective approach moving on to more intensive
interventions such as IPT.
There is clearly a great need for research regarding perinatal anxiety disor-
ders. Intuitively, targeted treatments such as IPT would seem to be of benefit for
these problems, though CBT may be of equal or greater benefit for perinatal
anxiety. The efficacy of these targeted interventions remains to be empirically
tested, as does the concept of stepwise treatment for anxiety.
In sum, the efficacy of psychotherapeutic interventions for the treatment
and prevention of perinatal depression is strongly supported by the data. The
strength of study designs, the large number of subjects in the trials, and the
impressive results all suggest that counseling is of benefit as a stand-alone
treatment for postpartum depression. For more severe depressions, there is need
for treatments such as IPT, which are specifically directed towards maternal
depression and which target social and interpersonal problems.
References
1 Stuart S, Robertson M: Interpersonal Psychotherapy: A Clinicians Guide. London, Edward
Arnold, 2003.
2 Klerman GL, Weissman MM, Rounsaville BJ, Chevron ES: Interpersonal Psychotherapy of
Depression. New York, Basic Books, 1984.
3 Paarlberg KM, Vingerhoets AJ, Passchier J, Heinen AG, Dekker GA, van Geijn HP: Psychosocial
factors as predictors of maternal well-being and pregnancy-related complaints. J Psychosom
Obstet Gynaecol 1996;17:93102.
4 Bernazzani O, Saucier JF, David H, Borgeat F: Psychosocial factors related to emotional distur-
bances during pregnancy. J Psychosom Res 1997;42:391402.
5 Gjerdingen DK, Froberg DG, Fontaine P: The effects of social support on womens health during
pregnancy, labor and delivery, and the postpartum period. Fam Med 1991;23:370375.
6 Kitamura T, Shima S, Sugawara M, Toda MA: Clinical and psychosocial correlates of antenatal
depression: A review. Psychother Psychosom 1996;65:117123.
8 Cox JL, Connor YM, Henderson I, McGuire RJ, Kendell RE: Prospective study of psychiatric
disorders of childbirth by self-report questionnaire. J Affect Disord 1983;5:17.
9 American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, ed 4.
Washington, American Psychiatric Association, 1994.
10 Engel GL: The clinical application of biopsychosocial models. Am J Psychiatry 1980;137:
535544.
11 OHara MW, Stuart S, Gorman L, Wenzel A: Efficacy of interpersonal psychotherapy for postpartum
depression. Arch Gen Psychiatry 2000;57:10391045.
12 Stuart S, OHara MW: Interpersonal psychotherapy for postpartum depression: A treatment
program. J Psychother Pract Res 1995;4:1829.
13 Sullivan HS: The Interpersonal Theory of Psychiatry. New York, Norton, 1953.
14 Bowlby J: Attachment. New York, Basic Books, 1969.
15 Bowlby J: The making and breaking of affectional bonds: Etiology and psychopathology in the
light of attachment theory. Br J Psychiatry 1977;130:201210.
16 Elkin I, Parloff MB, Hadley SW, Autry JH: NIMH Treatment of Depression Collaborative
Treatment Program: Background and research plan. Arch Gen Psychiatry 1985;42:305316.
17 Elkin I, Shea MT, Watkins JT, et al: National Institute of Mental Health Treatment of Depression
Collaborative Research Program: General effectiveness of treatments. Arch Gen Psychiatry
1989;46:971982.
18 Frank E, Kupfer DJ, Perel JM, et al: Three-year outcomes for maintenance therapies in recurrent
19 Stuart S: Interpersonal psychotherapy for postpartum depression; in Miller L (ed): Postpartum
Psychiatric Disorders. Washington, American Psychiatric Press, 1999, pp 143162.
20 Stuart S, Cole V: Treatment of depression following myocardial infarction with interpersonal
psychotherapy. Ann Clin Psychiatry 1996;8:203206.
21 Stuart S: Use of Interpersonal Psychotherapy for Other Disorders. Directions in Mental Health
Counseling. New York, Hatherleigh, 1997, pp 416.
22 Stuart S, Noyes R: Attachment and interpersonal communication in somatization disorder.
Psychosomatics 1999;40:3443.
23 Rogers CR: The necessary and sufficient conditions of therapeutic personality change. J Consult
Psychol 1957;21:95103.
24 Cohen LS, Sichel DA, Dimmock JA, Rosenbaum JF: Postpartum course in women with preexisting
panic disorder. J Clin Psychiatry 1994;55:289292.
25 Sichel DA, Cohen LS, Dimmock JA, Rosenbaum JF: Postpartum obsessive compulsive disorder:
A case series. J Clin Psychiatry 1993;54:156159.
26 Stuart S, Couser G, Schilder K, OHara MW, Gorman LL: Postpartum anxiety and depression:
Onset and comorbidity in a community sample. J Nerv Ment Dis 1998;186:420424.
27 Cohen LS, Viguera AC, Bouffard SM: Venlafaxine in the treatment of postpartum depression.
28 Stowe ZN, Casarella J, Landry J, Nemeroff CB: Sertraline in the treatment of women with post-
partum major depression. Depression 1995;3:4955.
29 Appleby L, Warner R, Whitton A, Faragher B: A controlled study of fluoxetine and cognitive-
behavioral counseling in the treatment of postpartum depression. BMJ 1997;314:932936.
30 Gordon RE, Gordon KK: Social factors in the prevention of postpartum emotional problems.
Obstet Gynecol 1960;15:433438.
31 Hayes BA, Muller R, Bradley BS: Perinatal depression: A randomized controlled trial of antenatal
education intervention for primiparas. Birth 2001;28:2835.
32 Halonen JS, Passman RH: Relaxation training and expectation in the treatment of postpartum
distress. J Consult Clin Psychol 1985;53:839845.
33 Lavender T, Walkinshaw SA: Can midwives reduce postpartum psychological morbidity?
A randomized trial. Birth 1998;25:215221.
Stuart/OHara 166
34 Small R, Lumley J, Donohue L, Potter A, Waldenstrom U: Randomised controlled trial of midwife
led debriefing to reduce maternal depression after operative childbirth. BMJ 2000;321:10431047.
35 Stamp GE, Williams AS, Crowther CA: Evaluation of antenatal and postnatal support to overcome
postnatal depression: A randomized, controlled trial. Birth 1995;22:138143.
36 Wolman WL, Chalmers B, Hofmeyer GJ, Nikodem VC: Postpartum depression and companion-
ship in the clinical birth environment: A randomized, controlled study. Am J Obstet Gynecol
1993;168:13881393.
37 Pitt B: Atypical depression following childbirth. Br J Psychiatry 1968;114:13251335.
38 Brugha TS, Wheatley S, Taub NA, et al: Pragmatic randomized trial of antenatal intervention to pre-
vent post-natal depression by reducing psychosocial risk factors. Psychol Med 2000;30:12731281.
39 Oakley A, Rajan L, Grant A: Social support and pregnancy outcome: Report of a randomized con-
trolled trial. Br J Obstet Gynaecol 1990;97:155162.
40 Elliott SA, Leverton TJ, Sanjack M, et al: Promoting mental health after childbirth: A controlled
trial of primary prevention of postnatal depression. Br J Clin Psychol 2000;39:223241.
41 Gorman LL: Prevention of postpartum adjustment difficulties (doctoral dissertation, University of
Iowa). Diss Abstr Int B Sci Eng 1997;58:2674.
42 Zlotnick C, Johnson SL, Miller IW, Pearlstein T, Howard M: Postpartum depression in women
receiving public assistance: Pilot study of an interpersonal-therapy-oriented group intervention.
Am J Psychiatry 2001;158:638640.
43 Holden JM, Sagovsky R, Cox JL: Counseling in a general practice setting: Controlled study of
health visitor intervention in treatment of postnatal depression. BMJ 1989;298:223226.
44 Spitzer RL, Endicott J, Robins E: Research Diagnostic Criteria: Rationale and reliability. Arch
45 Wickberg B, Hwang CP: Counseling of postnatal depression: A controlled study on a population
based Swedish sample. J Affect Disord 1996;39:209216.
46 Montgomery SA, Asberg M: A new depression rating scale designed to be sensitive to change. Br
J Psychiatry 1979;134:382389.
47 American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, ed 3
revised. Washington, American Psychiatric Association, 1987.
48 Cooper PJ, Murray L: Three psychological treatments for postnatal depression: A controlled
comparison. Eighth Int Conf Marce Soc, Cambridge, 1994.
49 Cooper PJ, Murray L, Wilson A, Romaniuk H: Controlled trial of the short- and long-term effect
of psychological treatment of post-partum depression. Br J Psychiatry 2003;182:412419.
50 McDonough S: Interaction guidance: Understanding and treating early infant-caregiver relationship
disorders; in Zeanah C (ed): Handbook of Infant Mental Health. New York, Guilford Press, 1993,
pp 414426.
51 Beck AT, Rush AJ, Shaw BF, Emery G: Cognitive Therapy of Depression. New York, Guilford
Press, 1979.
52 Cramer B, Robert-Tissot C, Stern D: Outcome evaluation in brief mother-infant psychotherapy:
A preliminary report. Infant Ment Health J 1990;11:278300.
53 Weissman MM, Bothwell S: Assessment of social adjustment by patient self-report. Arch Gen
Psychiatry 1976;33:11111115.
54 OHara MW, Hoffman JG, Phillips LHC, Wright EJ: Adjustment in childbearing women: The
Postpartum Adjustment Questionnaire. Psycho Assess 1992;4:160169.
55 Beck AT, Emery G: Anxiety Disorders and Phobias. New York, Basic Books, 1985.
56 Spinelli M, Endicott J: Controlled clinical trial of interpersonal psychotherapy versus parenting
education program for depressed pregnant women. Am J Psychiatry 2003;160:555562.
57 Hamilton MA: Development of a rating scale for primary depressive illness. Br J Soc Cli Psychol
1967;6:278296.
Scott Stuart, MD
University of Iowa, Department of Psychiatry
1293 Medical Education Building, Iowa City, IA 52242 (USA)
Tel. 1 319 353 4230, Fax 1 319 353 3003, E-Mail scott-stuart@uiowa.edu
Group Psychotherapy for Depression
in Early Stages of Motherhood
Maria Hofecker-Fallahpour, Anita Riecher-Rssler
Psychiatrische Universittspoliklinik, Basel, Schweiz
Individual psychotherapy in the treatment of depression is well accepted and
has a firm stand in terms of evidence replicated by numerous studies showing its
efficacy particularly in the use of cognitive behavioral therapy (CBT) and inter-
personal therapy (IPT) [24]. Furthermore, there is common agreement in clini-
cal practice that psychotherapy and antidepressant medication show comparable
effects in the treatment of mild depression, whereas in severer depression the
combination of psychotherapy and medication is recommended. Initial treatment
response is quicker with antidepressants compared to psychotherapy, but relapse
is more effectively prevented by psychotherapy unless prolonged antidepressant
maintenance treatment up to 1 year is administered [5, 6].
Group psychotherapy in comparison to individual psychotherapy has
numerous additional benefits such as facilitating social contacts and reducing
isolation and stigma by exchanging and working on similar symptoms and
difficulties in a group of peers. Moreover, group psychotherapy is commonly
regarded as being more cost-effective than individual psychotherapy. However,
many patients hesitate to participate in group activities due to a number of fears
such as feeling uncomfortable to talk about ones own problems in a group or
being overwhelmed by the severity of symptoms of others. In addition, group
treatment restricts the amount of time and consideration allotted to the difficul-
ties of a single patient as compared to individual therapy.
When psychotherapy is offered to depressed mothers of infants, a number of
specific demands arising from the particular needs of early motherhood have to be
taken into consideration. In addition, many mothers feel ashamed to experience
Parts of this paper have been published in Hofecker-Fallahpour et al. [1].
Hofecker-Fallahpour/Riecher-Rssler 168
difficulties in their new role and hesitate to seek professional help. Therefore,
adaptations to conventional psychotherapeutic procedures and contents are neces-
sary to meet the specific needs of this particular group of patients.
However, specific psychotherapy for mentally ill mothers of infants had
not been available until recently. Only in the past 510 years did a few therapists
start to adapt standard treatment strategies to the needs of mentally ill mothers.
For this reason, we decided to develop a manualized group therapy for mothers
with depression (GMD) and evaluated it in the first 31 patients.
In the following sections, existing specific psychotherapies for mothers of
infants or small children with depression are described. Then we present the ratio-
nale for developing our specific group therapy and report on the results regarding
effects and acceptance. The GMD is adjusted not only to a certain diagnostic
group but also to the demands of a most significant period in the lives of women.
The Need for Specific Psychotherapy for Depression
and Anxiety in Early Motherhood
Early Motherhood A Period of Specific Demands
Depression and anxiety disorders are known to be very frequent during
pregnancy and the postpartum period in the otherwise mostly healthy women of
childbearing age. Postpartum depression is found in 1015% of new mothers
and recent studies show similar rates of depression during pregnancy [79],
which is at least as high as in women of the same age group without children.
The investigation of anxiety disorders in perinatal women had been neglected to
some extent until recently, when studies have shown that the prevalence of
anxiety disorders is even higher than the prevalence of postpartum depression
[1012]. Matthey et al. [11] found rates of anxiety disorders such as phobia,
panic disorder or generalized anxiety disorder (except time criterion) of up to
16% in a sample of 216 first-time mothers, which exceeded the rate of pure
depression without anxiety by more than three times. Wenzel et al. [12] also
reported a higher prevalence of generalized anxiety than postpartum depression
in their general population sample of 68 women. It is important to note that the
symptom range of pure anxiety disorders is largely independent of the depres-
sive symptom range so that anxiety disorders will be missed by commonly used
screening instruments for depression [13, 14].
Even though many mothers have a clear postpartum onset of depression or
anxiety, other mothers already suffer during pregnancy, continue to have symp-
toms long after the postpartum period or even start to develop mental problems
in the following years of early motherhood. Those mothers, suffering from a
Group Psychotherapy for Mothers 169
mental disorder outside the explicit perinatal period, are even more likely to be
neglected and they themselves have even more difficulties seeking help since
they have no group to identify with. All these mothers show specific needs for
care due to their particular life situation.
Therefore, we suggest to use a term like depression in early motherhood
in accordance with terms like depression of old age or depression in adoles-
cence. We suggest to include the time span from pregnancy up to prekinder-
garten age to define this period [15]. Due to an apparently high prevalence of
anxiety disorders in early motherhood, both terms, depression and anxiety,
should be commonly used.
During that age, a considerable number of infants and small children in
many western societies are cared for entirely or mainly by their mothers at
home. Such mothers, especially if they are mentally ill, very often have given
up their former jobs after childbirth and only have a restricted social network at
their disposal. They also hesitate or refuse to seek help and have difficulties
providing their children with an adequate environment and stimulation. Many
of these children of mentally ill mothers only draw attention to themselves
when entering kindergarten showing a variety of already long-lasting behav-
ioral, emotional and cognitive difficulties. Therefore, caregivers to mothers of
infants must be aware of the fact that many mental illnesses including depres-
sion and anxiety show a tendency to relapse or to worsen during this period of
early motherhood and need our attention.
The life event scale is ranking pregnancy as a potentially harmful life event
on place 12, and the gain of a new family member ranks on place 14 out of a list
of 43 items, underscoring the impact of childbirth on new parents even in the
absence of any complication on the part of the mother or the infant [16]. In
addition, the strain of caring for a newborn can be quite a challenge for a couple
and even more so for a single mother. Moreover, the couple has to come to
terms with far-reaching changes in social roles such as changes in working
environment, social networking and close relationships.
First-time mothers and fathers have to come along with a range of new
tasks which cannot be simply delegated to others if the couple finds itself as not
being fit to tackle the tasks. It is well accepted that new parents need time to
adapt and to grow into the responsibility of parents. Not only does the mother
experience considerable strain, the father does so as well. Several investigations
have pointed out that perinatal depressive disorders exert a considerable amount
of strain on the healthy partner and can lead to mental illness in himself. The
couple concordance in developing mental illness, if the mother suffers from
perinatal depression, is high [17, 18]. If the father falls mentally ill, the likeli-
hood of the mother to develop symptoms herself is even higher. Matthey et al.
[11] showed the same effect in postpartum anxiety disorders.
The Pathoplastic Influence of Early Motherhood
Pitt [19], one of the first authors to investigate and describe postpartum
depression, used the term atypical depression to describe typical mood alter-
ations following childbirth. In his sample, he found, alongside with 1011%
depressed mothers, other conditions which he classified as anxiety without
depression, reactive anxiety and depression due to infants ill-health, prolonged
fatigue, psychosomatic disorders, and diminution of libido. He described
depressed mothers as weary, irritable, despondent and anxious about the baby
[20]. The more recent studies clearly indicate that there is no evidence for a
specific set of symptoms pertaining only to postpartum depression [21]. Yet,
from a clinical point of view, mothers often complain about a marked increase
in irritability and fears alongside with classical symptoms of depression, both
resulting in particular difficulties when interacting with the baby or with their
partners.
Irritability is described as a distinct mood condition in premenstrual,
perinatal, and perimenopausal mood disturbances [22].
Obsessive thoughts and compulsions were also found to be more common
in postpartum depressed mothers (57%) as compared to depressed women
without children (39%) [23]. The content of obsessive thoughts was primarily
aggressive such as accidentally stabbing or drowning the baby or throwing it
down the stairs or imagining most horrifying events happening to the baby.
Specific Psychotherapies in the Treatment of
Perinatal Depression
Individual Psychotherapies
Cooper et al. [24] investigated 193 women with postpartum depression by
assigning them randomly to three different specific interventions including
CBT, nondirective counseling and psychodynamic therapy and compared these
interventions to routine primary care. The three specific therapies were carried
out in the homes of women by weekly sessions from the 3rd to the 5th month
after parturition, whereas routine care was performed in the usual way by
general practitioners and health visitors. A significant short-term effect on
improvement of mood in favor of the three specific interventions was shown
immediately after treatment, whereas in the long term, no superior effect in
comparison to routine primary care was evident.
OHara et al. [25] performed a 12-week randomized controlled trial
including 120 subjects by comparing patients attending weekly sessions of IPT
with waiting list controls showing a significant improvement of depression in
the treatment group.
Appleby et al. [26] compared fluoxetine treatment of 87 subjects to
placebo with additional counseling either once or six times after initiating med-
ical treatment by randomly assigning women to the four different conditions.
Fluoxetine treatment proved to be superior to placebo without any measurable
additional effect of counseling regardless of the amount of counseling. The
study had a remarkable dropout rate which might reflect reluctance to take
medication, being common in mothers during the perinatal period.
Two other studies compared counseling versus routine care without coun-
seling showing that counseling was helpful in mild depression; however, addi-
tional therapy might be necessary in severer depression [27, 28].
At present, there are no reports on specific treatments of anxiety disorders
in the perinatal period.
Group Psychotherapies
Over the last 30 years, a number of group treatments for mothers with
depressive disorders have been reported. Many of these reports merely
described the experience and used simple evaluation strategies without control
groups [2935]. These descriptive studies give some insight into advantages
and difficulties and possible procedures of group treatment for mothers of
infants. Klier et al. [35] adapted IPT for the group setting and found it helpful
as it focuses on interpersonal relationships and role transitions both of which
represent important topics in new mothers.
So far, only three studies on group therapy for depressed mothers have
used a randomized control design [3638].
Fleming et al. [36] used social support groups, in a total sample of 127
women controlling for interventions by mail and a nonintervention group.
There were two major disadvantages of the study; on the one hand, the recruit-
ment of mothers within 2 weeks after delivery, thereby including mothers who
suffered from baby blues but fully recovered within a short period of time, and,
on the other hand, healthy and depressed mothers attending the same group,
leading to increased feelings of failure in the depressed mothers.
Meager and Milgrom [37] reported on a pilot study comparing group ther-
apy using CBT, educational and social support components to a waiting list
control group. The 10-week group treatment resulted in a significant reduction
of depressive symptoms, whereas women on the waiting list remained severely
depressed. The same group of investigators then developed a manualized group
program and evaluated a larger sample of group participants controlling for
routine community care [38]. There was a significant improvement of depres-
sion in the CBT group compared to the group receiving routine community
care. Furthermore, levels of perceived social support improved in the CBT
group but not in the routine community care group, whereas parenting stress
was not changed by either treatment.
Mothers with postpartum depression were repeatedly reported as being
reluctant to enter group-based treatment [24, 39], which seems to be even more
the case in rural areas [40]. Despite these difficulties, depressed mothers of
infants represent a very homogenous sample predisposing them to mutually
support each other and benefit from each other within a group setting.
There are also an increasing number of reports on antenatal group
interventions primarily targeted at the prevention of postpartum depression by
identifying and treating pregnant women at risk. Several group approaches
reported on using psychoeducation classes, skill training and discussion of the
challenges of new motherhood [4149]. At present, there is no clear evidence
regarding the prevention of postpartum depression by the use of such ante- and
perinatal group interventions [50, 51].
The Basle Group Therapy for Mothers with Depression
Development of a Manualized Group Psychotherapy
We developed the GMD following the descriptions given by Meager and
Milgrom [37]. Well-established CBT techniques were primarily used adapting
them to the needs and concerns of mothers in early motherhood and supple-
mented by a systemic and psychoeducative approach. A manual was developed
to facilitate and standardize treatment performance.
The GMD was designed for a closed group of 68 mothers with 12 weekly
sessions of 1.5-hour duration. The group room was equipped with an overhead
projector and a flip chart as well as colored post-it papers (to optimize visual
anchoring of the 3- and 5-column technique) and pens.
Therapeutic Techniques and Contents of GMD
Group treatment was started by introducing and training cognitive techniques
like the tracing of negative automatic thoughts and errors in logical thinking as
well as the detection of dysfunctional core beliefs [52]. The cognitive con-
cepts were illustrated by examples out of the daily routine of depressed, irritable
and anxious mothers of infants. In further sessions, behavioral techniques
like encouraging mothers to take up positive activities and to improve their
help-seeking behavior were introduced. Psychoeducative elements of depres-
sion, mother-child interaction and couple relationship in the transition to par-
enthood were enclosed consecutively in the training program to provide
sufficient background information in order to facilitate a change of attitudes
and behavior. Two sessions were dedicated to rehearse cognitive and behavioral
skills and another session was targeted at early recognition of relapse and strate-
gies of crisis intervention. Group work was enhanced by homework to test dif-
ferent strategies in the reality of daily routine. Difficulties of the depressed
mothers in parenting and interaction with the child and the spouse were
approached with a systemic point of view by encouraging the mothers to reacti-
vate their family bonds and other social contacts.
The partner was invited to an additional couple session to air difficulties and
to improve the understanding of his wifes depression, irritability, and anxiety.
Nursery was offered during group sessions paying respect to the fact that mothers
of infants needed extra support to be able to attend psychotherapy.
Topics of the sessions
Session 1: Introduction into depression and 3-column technique
Session 2: Automatic thoughts, dysfunctional core beliefs, logical errors and 5-column
technique
Session 3: Dysfunctional core beliefs, vicious circles, stressors in daily routine with children,
positive activities
Session 4: Modifying dysfunctional core beliefs, models of stress coping, help-seeking
strategies
Session 5: Other treatment options for depression beside CBT
Session 6: Repetition of cognitive and behavioral strategies
Session 7: Mother-child interaction
Session 8: Role of parents and parenting
Session 9: Couple relationship and sexuality
Session 10: Relapse into depression, suicidality and crisis intervention
Session 11: Conclusion of cognitive and behavioral strategies
Session 12: Farewell to the group and planning of the near future
Evaluation Study
In order to evaluate this newly developed group therapy, we conducted a
first evaluation study in 31 mothers with depression. Mothers were referred to
the Psychiatric Outpatient Department in Basle by family doctors, pediatri-
cians, maternity centers or requested treatment by means of self-referral.
The following hypothesis was formulated: group therapy will result in a
significant reduction of depressive symptoms, which will remain stable in a
1-year follow-up.
Inclusion and Exclusion Criteria
Inclusion criteria were a youngest infant within an age range from third-
trimester pregnancy up to prekindergarten age, and mothers with a clinical
diagnosis of first or recurrent episodes of unipolar or bipolar depression (ICD-
10 F32, F33, F31.3, F31.4), dysthymia (ICD-10 F34.1) or adjustment disorder
with depressive or mixed anxiety and depressive reaction (ICD-10 F43.2) [53].
Comorbidity with anxiety disorders (ICD-10 F40, F41), obsessive-compulsive
disorder (ICD-10 F42) or eating disorders (ICD-10 F50) were accepted as long
as the depressive disorder was the main complaint. In addition, fluent knowl-
edge of the German language and willingness as well as ability to participate in
a group therapy were afforded.
Excluded were patients suffering from acute suicidality, severe degrees of
hostility against their infant, psychotic symptoms, addiction disorders or person-
ality disorders.
Setting
We formed five consecutive groups, which were conducted by one fully
trained female psychiatrist and one female psychiatrist in training. The majority
of therapists were mothers themselves.
A nursery for the infants and children was available, since many depressed
mothers have difficulties organizing a baby-sitter on their own due to a lack of
energy or out of guilt and shame to give the baby away for their own treatment.
The child minder responded sensitively to the needs of the depressed mothers
who suffered from a lack of self-esteem and easily felt frustrated by the slight-
est comments on their children or on their own behavior. Therefore, we chose a
child minder who was experienced in small talk and conversation with
depressed mothers. For the nursery, age-specific toys were available.
Sample
From 1998 to 2001, 52 women were referred to us for group therapy. Five
did not fulfill diagnostic criteria, and 1 decided not to participate before
commencement of the therapy. Out of the remaining 46 women, 31 attended
group therapy more than 6 times out of 12 sessions, whereas 15 women were
regarded as drop-outs attending less than 6 times. Eight of them had attended
only once.
The majority of women who dropped out had a mother tongue different
from German and followed different cultural concepts of motherhood. In addi-
tion, most of them had not achieved vocational qualification.
The 31 fully attending mothers were on average 33.5 years old (SD 4.1,
range 24.340.1). Twenty-six were married, 2 of them living separated from
their spouse. Five were unmarried mothers, 2 of them living with their partner.
The 31 mothers had 1.6 children on average (SD 0.6, range 13) who were
2.4 years old on average (SD 2.1), ranging in age from the 3rd trimester of
pregnancy (1 person) up to 9 years of age (older sibling of an infant). Twenty-
eight women were fully qualified in their former jobs, 7 mothers were working
part-time, 1 woman was a university student and working part-time and 1
woman was in job training.
As far as diagnosis was concerned, 11 women suffered from recurrent
major depressive episodes, 1 being mild, 9 being medium and 1 being severe in
degree. A first episode of major depression was the diagnosis in 9 women, 2
being mild, 3 medium and 4 severe in degree. Adjustment disorder with depres-
sive reaction was diagnosed in 6 women, 3 being of short duration, 3 of pro-
longed duration. One woman suffered from panic disorder without depression
and 1 woman from bipolar disorder with current depression. Three women were
admitted to group therapy for prophylactic reasons. They had no current depres-
sion but postpartum onset of severe depression after previous childbirth.
Comorbidity was found in 3 patients (panic disorder, dysthymia, and bulimia
nervosa).
At the beginning of group therapy, 14 women were on antidepressants, 8 of
them having taken medication for more than 3 months, 1 woman for less than
2 months and 5 for less than 1 month. On average, patients attended the group
program 10 times within a range of 612 times. Twenty-nine partners were
attending an additional couple session, 3 of them living separated from their
partner. Two single mothers refused to have a parent session due to severe dis-
cord with the father of the infant.
Instruments of Evaluation and Statistics
A clinical interview was performed to assess symptoms, diagnosis and
inclusion criteria for group therapy. The diagnostic criteria of ICD-10, chapter F
were used (WHO, 1992). For assessing the severity of psychopathology, we
used the following self-rating scales: the Beck Depression Inventory (BDI)
[54], Symptom Check List (SCL-90-R) [55] and a self-rating scale to evaluate
couple relationship [Familienbgen-Zweierbeziehung (FB-Z)] [56]. The same
instruments and an additional questionnaire to assess acceptance of group ther-
apy and subjective changes in relationship attributed to group therapy were
handed out at the end of the group therapy. The latter comprised 12 questions
with a Likert scale ranging from 3 (not at all true) to 3 (very true). In addi-
tion, in a small number of patients, these instruments were handed out again 1
year after termination of group therapy.
Means and standard deviations were calculated. One-tailed t tests were
performed to calculate significance. To compare group participants who used
antidepressant medication with those who did not, a one-factorial analysis of
variance was performed. Effect sizes were calculated as suggested by Rosenthal
[57]. Effect sizes between 0.40 and 0.79 indicate medium efficacy, below 0.40
poor efficacy and above 0.80 substantial effects of psychotherapy.
Results
BDI scores showed a highly significant reduction in depressive symptoms
from pretesting before group therapy to posttesting immediately after the last
session of group therapy. Mean BDI scores dropped from 21.5 to 12.9 points
(one-tailed t test, d.f. 29, p 0.001). Twenty patients showed a significant
drop in BDI scores, 3 patients remained unchanged, 4 deteriorated in BDI
scores and 4 showed scores below 10 before and after treatment.
Follow-up evaluation after 1 year was only done in 10 patients. They
showed a drop of initial BDI scores from 22.7 before treatment to 10.7 immedi-
ately after treatment and 13.6 1 year after treatment. The difference between
pretesting and posttesting was significant (one-tailed t test, d.f. 5, p 0.001)
just as between pretesting and 1-year follow-up (one-tailed t test, d.f. 5,
p 0.01). From posttesting to 1-year follow-up, there was no significant dete-
rioration, indicating that the achieved improvement was stable (one-tailed t test,
d.f. 5, p 0.18).
SCL-90-R scores were available for 24 patients. Similar reductions were
achieved on the depression subscale of this scale comparing the scores before
and after group therapy. A significant reduction in symptoms was also seen on
other subscales of the SCL-90-R, e.g. aggression/hostility, interpersonal sen-
sitivity and obsessive-compulsive.
Effect sizes were 1.01 in the BDI and 0.99 in the SCL-90-R, indicating
substantial effects of psychotherapy.
The instrument measuring couple relationship (FB-Z), which was available
in 23 patients, showed scores higher than average at both measure points, indi-
cating preexisting difficulties in couple relationship without improvement in
the course of treatment.
At the end of treatment, the patients expressed high levels of satisfaction,
on a Likert scale ranging from 3 (not at all true) to 3 (very true), with the
treatment (mean 3, SD 1.12) and the therapists (mean 3, SD 0.44),
and agreed strongly that their stress-coping behavior in daily routine had
improved with therapy (mean 2, SD 0.83). They felt, subjectively, that
their relationship to the infant had strongly improved over the course of treat-
ment (mean 3, SD 1.86), whereas couple relationship was felt to have
slightly deteriorated (mean 0, SD 2.02).
To control for the influence of medication, analysis of variance was per-
formed to distinguish group treatment effects in patients with and without anti-
depressants. It was shown that both subsamples of patients improved
significantly and to the same extent over the course of group treatment with a
parallel decline of depressive symptoms in both groups. Patients who were on
antidepressants had higher initial levels of depression compared to those with-
out antidepressants. The 3 asymptomatic patients who had been admitted for
primary prevention of depression and 1 patient with pure panic disorder were
not included into this analysis of variance.
Discussion
The GMD was developed to improve and facilitate the treatment of
depressed mothers of infants and small children by adapting well-established
antidepressant psychotherapeutic strategies to the specific needs and concerns
of early motherhood.
The significant drop in the scores of the BDI and in the depression sub-
scale scores of the SCL-90-R down to almost symptom-free levels could be
replicated in every single group of the five consecutive group treatments, indi-
cating the efficacy of the therapy program. The significant symptom reduction
was not only apparent in patients without antidepressants but also in patients on
antidepressants who had not reached full remission before entering group ther-
apy in spite of adequate medication.
The symptom reductions on several other dimensions of the SCL-90-R such
as aggression/hostility, interpersonal sensitivity, and obsessive-compulsive
also contributed to a mothers ability to care for her baby and adapt to a differ-
ent lifestyle. Participating mothers definitely benefited from feeling less hostile
and tense in handling their babies and their social skills improved by attending
the group therapy, which might have served as an incentive to resume social
activities in general.
The majority of women who had been offered group treatment showed
spontaneous and clear interest in participating. Some women expressed some
fears before entering group therapy, especially that their condition might be
worsened being confronted with more severely depressed mothers. However,
during the sessions, mothers repeatedly expressed their satisfaction with meet-
ing other mothers who experienced similar difficulties. Some of them started to
contact other group members between sessions, which was generally well toler-
ated by the women.
The high level of acceptance was also reflected by a high average rate of
attendance in patients who were prone to miss sessions due to illness or day
sleeping hours of the child and other typical hazards of early motherhood.
In addition, mothers expressed an overall high level of satisfaction with the
group meetings and the therapy program. This stands in contrast to the experi-
ence of some other authors who found mothers to be reluctant to enter group
therapy [24].
The nursery was very much appreciated by those mothers who had no
other possibility to get baby-sitting for their child. On average, up to 50% of
mothers made regular use of the nursery during group sessions.
The limitations of the study were as follows: the sample was recruited in a
clinical setting by consecutively admitting every referred patient who fulfilled
inclusion criteria. Therefore, the sample could have been biased as mainly
patients with adequate help-seeking behavior were included. Drop-outs
concerned mainly patients with a different cultural and unskilled vocational
background. Evaluation included only self-administered scales. Diagnosis
was made by a clinical interview only and not by a standardized instrument.
Randomization was not possible at the very beginning of the clinical imple-
mentation of the program. A control group using individual therapy as usual
was, however, started after completion of five consecutive groups. The results
of this control group will be reported in the near future.
Conclusions
Early motherhood is a period of increased vulnerability to develop mood
disorders due to challenging positive and possibly negative life events. Until
recently, only few studies have reported on specific psychotherapeutic interven-
tions for mothers of infants and small children suffering from depression and
anxiety. In general, psychotherapy seems to improve symptoms in this group of
patients. Yet, little is known which interventions in particular are most helpful
for these patients. Therefore, further investigations are needed to find out in
more detail which strategies are most efficacious and effective.
The Basle GMD is specific with regard to disorder, gender and life situa-
tion. The program proved to be efficacious and well accepted not only by reduc-
ing symptoms of depression in mothers but also by improving the relationship
with the child as rated by the mothers themselves. The manual facilitates clini-
cal implementation. The results of a post hoc control group with individual
therapy as usual will be reported in the near future. In addition, a randomized
study is planned with improved design and instruments for evaluation to com-
pare individual therapy with group therapy in terms of symptom reduction as
well as in terms of couple relationship and mother-child interaction.
References
1 Hofecker-Fallahpour M, Zinkernagel-Burri C, Stoeckli B, Wsten G, Stieglitz R-D, Riecher-
Rssler A: Gruppentherapie bei Depression in der frhen Mutterschaft. Nervenarzt 2003;74:
767774.
2 Beck AT, Rush AJ, Shaw BF, Emery G: Cognitive Therapy of Depression. New York, Guilford
Press, 1979.
3 Lewinsohn M, Antonuccio DO, Steinmetz JL, Teri L: The Coping with Depression Course.
A Psychoeducational Intervention for Unipolar Depression. Eugene, Castalia Publishing, 1984.
4 Klerman GL, Weissman MM, Rounsaville BJ, Chevron ES: Interpersonal Psychotherapy of
Depression. New York, Basic Books, 1984.
5 Clark DM: Cognitive therapy for depression and anxiety: Is it better than drug treatment in the
long term?; in Hawton K, Cowen P (eds): Dilemmas and Difficulties in the Management of
Psychiatric Patients. New York, Oxford University Press, 1990, pp 5564.
6 Glick ID: Treating Depression. San Francisco, Jossey-Bass, 1995.
8 Fergusson DM, Horwood LJ, Thorpe K: Change in depression during and following pregnancy.
ALSPAC Study Team. Study of pregnancy and children. Paediatr Perinat Epidemiol 1996;10:
279293.
9 Evans J, Heron J, Francomb H, Oke S, Golding J: Cohort study of depressed mood during preg-
nancy and after childbirth. BMJ 2001;323:257260.
10 Stuart S, Couser G, Schilder K, OHara MW, Gorman L: Postpartum anxiety and depression:
Onset and comorbidity in a community sample. J Nerv Ment Dis 1998;186:420424.
12 Wenzel A, Haugen EN, Jackson LC, Robinson K: Prevalence of generalized anxiety at eight weeks
postpartum. Arch Women Ment Health 2003;6:4349.
13 Ballard CG, Davis R, Handy S, Mohan RN: Postpartum anxiety in mothers and fathers. Eur J
14 Muzik M, Klier CM, Rosenblum KL: Are commonly used self-report inventories suitable for
screening postpartum depression and anxiety disorders? Acta Psychiatr Scand 2000;102:7173.
15 Hofecker-Fallahpour M, Riecher-Rssler A: Depression in der frhen Mutterschaft: Erschpft,
gereizt, berngstlich. Neurotransmitter 2003;2:3539.
16 Dimsdale JE, Keefe FJ, Stein MR: Stress and psychiatry; in Sadock BJ, Sadock VA (eds):
Comprehensive Textbook of Psychiatry. Philadelphia, Lippincott Williams & Wilkins, 2000,
p 1837.
17 Raskin VD, Richman JA, Gaines C: Patterns of depressive symptoms in expectant and new
parents. Am J Psychiatry 1990;147:658660.
18 Matthey S, Barnett B, Ungerer J, Waters B: Paternal and maternal depressed mood during the
transition to parenthood. J Affect Disord 2000;60:7585.
19 Pitt B: Atypical depression following child birth. Br J Psychiatry 1968;114:13251335.
20 Pitt B: Postpartum disorders; in Priest RG (ed): Psychological Disorders in Obstetrics and
Gynaecology. London, Butterworths, 1985.
21 Riecher-Rssler A, Hofecker-Fallahpour M: Postpartum depression: Do we still need this diagnostic
term? Acta Psychiatr Scand 2003;108:5156.
22 Born L, Steiner M: Irritability: The forgotten dimension of female-specific mood disorders. Arch
Women Ment Health 1999;2:153167.
23 Wisner KL, Peindl KS, Gigliotti T, Hanusa BH: Obsessions and compulsions in women with post-
partum depression. J Clin Psychiatry 1999;60:176180.
24 Cooper P, Murray L, Wilson A, Romaniuk H: Controlled trial of short- and long-term effect of
psychosocial treatment of postpartum depression. Br J Psychiatry 2003;182:412419.
25 OHara MW, Stuart S, Gorman L, Wenzel A: Efficacy of interpersonal therapy for postpartum
26 Appleby L, Warner R, Whitton A, Faragher B: A controlled study of fluoxetine and cognitive-
behavioral counseling in the treatment of postparum depression. BMJ 1997;314:932936.
27 Holden JM, Sagovsky R, Cox JL: Counseling in a general practice setting: Controlled study of
health visitor interventions in treatment of postnatal depression. BMJ 1989;298:223226.
28 Wickberg B, Hwang CP: Counseling of postnatal depression: A controlled study on a population
based Swedish sample. J Affect Disord 1996;39:209216.
29 Morris JB: Group psychotherapy for prolonged postnatal depression. Br J Med Psychol 1987;60:
279281.
30 Gordon J, Swan M, Robertson R: Babies dont come with a set of instructions. Running support
groups for mothers. Health Visit 1995;68:155156.
31 Jones A, Watts T, Romain S: Facilitating peer group support. Health Visit 1995;68:153.
32 May A: Using exercise to tackle postnatal depression. Health Visit 1995;68:146147.
33 Pitts F: Comrades in adversity: The group approach. Health Visit 1995;68:144145.
34 Morgan M, Matthey S, Barnett B, Richardson C: A group programme for postnatally distressed
women and their partners. J Adv Nurs 1997;26:913920.
35 Klier C, Muzik M, Rosenblum KL, Lenz G: Interpersonal psychotherapy adapted for the group
setting in the treatment of postpartum depression. J Psychother Pract Res 2001;10:124131.
36 Fleming AS, Klein E, Corter C: The effect of a social support group on depression, maternal
attitudes and behaviour in new mothers. J Child Psychol Psychiatry 1992;33:685698.
37 Meager I, Milgrom J: Group treatment for postpartum depression: A pilot study. Aust N Z J
Psychiatry 1996;30:852860.
38 Milgrom J, Martin PR, Negri LM: Treating Postnatal Depression. A Psychological Approach for
Health Care Practitioners. Chichester, Wiley, 1999.
39 Seeley S, Murray L, Cooper PJ: The outcome for mothers and babies of health visitor intervention.
Health Visit 1996;69:135138.
40 Lane B, Roufeil LM, Williams S, Tweedie R: Its just different in the country: Postnatal depression
and group therapy in a rural setting. Soc Work Health Care 2001;34:333348.
41 Gordon RE, Gordon KK, Englewood N: Social factors in prediction and treatment of emotional
disorders in pregnancy. Am J Obstet Gynecol 1959;77:10741083.
42 Gordon RE, Gordon KK: Social factors in prevention of postpartum emotional problems. Obstet
Gynecol 1960;15:433437.
43 Gordon RE, Kapostins EE, Gordon KK: Factors in postpartum emotional adjustment. Obstet
Gynecol 1965;25:158166.
44 Elliott SA, Sanjack M, Leverton TJ: Parent groups in pregnancy: A preventive intervention for
postnatal depression; in Gottlieb BH (ed): Marshaling Social Support: Formats, Process and
Effects. London, Sage, 1988.
45 Stamp GE, Williams AS, Crowther CA: Evaluation of antenatal and postnatal support to overcome
postnatal depression: A randomized controlled trial. Birth 1995;22:138143.
46 Buist A, Westley D, Hill C: Antenatal prevention of postnatal depression. Arch Women Ment
Health 1998;1:17.
47 Elliott SA, Leverton TJ, Sanjack M, Turner H, Cowmeadow P, Hopkins J, Bushnell D: Promoting
mental health after childbirth: A controlled trial of primary prevention of postnatal depression. Br
J Clin Psychol 2000;39:223241.
48 Brugha TS, Wheatley S, Taub NA, Culverwell A, Friedman T, Kirwan P, Jones DR, Shapiro DA:
A pragmatic randomized trial of antenatal intervention to prevent postnatal depression by reducing
psychosocial risk factors. Psychol Med 2000;30:12731281.
49 Zlotnick C, Johnson SL, Miller IW, Pearlstein T, Howard M: Postpartum depression in women
receiving public assistance: Pilot study of an interpersonal-therapy-oriented group intervention.
Am J Psychiatry 2001;158:638640.
50 Stuart S, OHara MW: The efficacy of psychotherapy for perinatal depression. Econ Neurosci
2001;3:4651.
51 Austin MP: Targeted group antenatal prevention of postnatal depression: A review. Acta Psychiatr
Scand 2003;107:244250.
52 Free ML: Cognitive Therapy in Groups. Guidelines and Resources for Practice. Chichester, Wiley,
1999.
53 World Health Organization: ICD-10 Classification of Mental and Behavioral Disorders with
Glossary and Diagnostic Criteria for Research ICD-10; DCR-10. Compilation and editorial arrange-
ments by J.E. Cooper. World Health Organization Geneva. Edinburgh, Churchill Livingstone, 1991.
54 Hautzinger M, Bailer M, Worall H, Keller F: Beck-Depressions-Inventar (BDI). Testhandbuch.
Bern, Huber, 1995.
55 Franke G: Die Symptom-Checkliste von Derogatis Deutsche Version Manual. Gttingen,
Beltz, 1995.
56 Cierpka M, Frevert G: Die Familienbgen. Ein Inventar zur Einschtzung von Familienfunktionen.
Gttingen, Hogrefe, 1995.
57 Rosenthal R: Metaanalytic Procedures for Social Research. Newbury Park, Sage, 1997.
Dr. Maria Hofecker-Fallahpour
Psychiatrische Universittspoliklinik
Petersgraben 4
CH4031 Basel (Switzerland)
Tel. 41 61 265 50 40, Fax 41 61 265 45 88, E-Mail hofeckerm@uhbs.ch
Alternative Treatment Strategies for
Perinatal Depression and Anxiety
Madeleine OHiggins
a
, Vivette Glover
a
, Maria Corral
b
a
Institute of Reproductive and Developmental Biology, Imperial College
London, London, UK;
b
Reproductive Mental Health Programme,
St. Pauls and BC Womens Hospitals, Vancouver, Canada
Complementary and alternative medicine (CAM) is the first choice for
many people suffering from anxiety and depression. These are the only medical
conditions for which the majority seek complementary or alternative treat-
ments, rather than orthodox ones. Kessler et al. [1] reported that, in the USA,
53% of those with symptoms of anxiety had used alternative treatments compared
with 41% who had sought conventional treatment. In the case of depression, the
figures were 57 and 36%, respectively.
It is possible that these percentages are even higher in the perinatal period
when women are likely to be even more reluctant to take medication. Mood dis-
turbances and anxiety problems are common at this time. Postnatal depression
is present in 1015% of mothers [2], and depression also affects at least this
percentage antenatally [3, 4]. There is a strong comorbidity between anxiety
and depression. Matthey et al. [5] note that anxiety may often be the presenting
symptom of a mother with postnatal depression.
While the symptoms of depression and anxiety appear to be the same in
the perinatal period as at other times, there is a growing fetus and developing
baby who must also be considered when deciding on treatment. Studies demon-
strating the efficacy and safety of pharmacological treatments for perinatal
depression, particularly those involving the use of tricyclic antidepressants and
serotonin reuptake inhibitors, are only just starting [6, 7]. Women remain
concerned about the possible effects these medications may have on the fetus
and breast-fed baby. Recent research shows that most of these medications
do not increase the risk of congenital malformations when used during preg-
nancy [8]. However, there have been few studies published that address the
Alternative Therapies 183
issue of long-term safety of these medications [9, 10]. Following childbirth,
most women choose to breast-feed their infants for at least several months.
The research to date indicates that relatively low levels of antidepressant
medication are excreted into breast milk, and similarly, infant serum levels of
these medications have been shown to be low [11]. Despite the encouraging
nature of these early data, many women continue to express a preference for
non-pharmacological treatments of anxiety or depression occurring during
pregnancy or in the postpartum period. In this paper, the current evidence for
some non-pharmacological, alternative treatment strategies is reviewed with
reference to their potential use in the perinatal period.
What Is Meant by Alternative Treatment Strategies?
For the purpose of this review, the term alternative treatment strategies is
used to refer to treatments for anxiety or depression other than antidepressants
or forms of counselling and psychotherapy. This encompasses a wide range,
including homoeopathy, acupuncture, herbal remedies, exercise or relaxation
and bright light therapy. Those discussed here will be yoga/relaxation/medita-
tion, exercise, infant massage and bright light therapy, as these are the ones for
which there is some evidence for efficacy in depression or anxiety [1, 1214].
We will also discuss the little research there is that is specific to the perinatal
period [15, 16].
There is a lack of good-quality research in the area of complementary
therapies. There is still a debate about the most appropriate type of research to
undertake and also a lack of consensus on terms and definitions. Much has
been written about the feasibility and applicability of randomised controlled
trials for investigating CAM. While this is considered the gold standard for
research in orthodox medicine, some argue that relevance is sacrificed in favour
of rigour when randomised controlled trials are used in this area [17].
Richardson [17] notes the importance of the therapeutic relationship, which
may not be measured quantitatively. Vuckovic [18] mentions the incidental
outcomes that may result from choosing a complementary treatment option.
She cites lifestyle changes, such as stopping smoking, as possible incidental
outcomes that may be important to the patient, but could go unrecorded in a
randomised controlled trial. One other concern about research into CAM has
been that randomised controlled trials do not fit well with the tailoring of treat-
ments to individual patients that is important in many such treatments [19]. This
is all true. However, the problem with these approaches is that they can be used
to explain away negative results for such reasons: evidence that a particular
treatment does not work is never accepted.
OHiggins/Glover/Corral 184
Our approach is that with research into CAM, the first requirement should
still be to establish with a trial, preferably a randomised controlled trial,
whether a particular treatment has better results than no treatment. One can then
try to unpick which components of the CAM are actually producing the benefit,
and also tailor treatments for individual patients. This can be combined with
qualitative or other hypothesis-generating research.
Exercise
Craft and Landers [20] have conducted a meta-analysis of studies that have
looked at exercise as an intervention for clinically depressed patients or those
that were seriously mentally ill and depressed as a result of that mental illness.
It was found that all types of exercise investigated were effective in lowering
depression scores. Larzelere and Wiseman [12] further note that changes in
physical fitness are only weakly related to depression reduction. They empha-
sise the importance of long-term maintenance of exercise behaviour to ensure
the reduction in depression persists. Results from the meta-analyses [21] show
that the largest effect size has been found with programmes of exercise lasting
912 weeks. Craft and Landers [20] suggest that research should now focus on
even longer programmes of exercise, as such protocols have not been examined.
While both aerobic and non-aerobic activities are reported to be equally effec-
tive in the treatment of depression in the article by Larzelere and Wiseman [12],
Craft and Landers [20] mention that running appeared to have the largest effect,
although this was not statistically significant. One suggested mechanism for
such effects is via the serotonin system in the brain. Acute physical exercise has
been shown to increase levels in rats [22].
Exercise is generally agreed to be useful in producing a small to moderate
reduction in both trait and state anxiety [23]. Research suggests that exercise
duration should be of 40 min or more with a training period of 16 weeks or
longer [24] in order to produce the largest decrease in trait anxiety. Larzelere
and Wiseman [12] suggest the same guidelines for exercise to reduce depres-
sion should be followed here. This implies that exercise should be undertaken at
a level and frequency that is comfortable for the patient. The effects of exercise
on anxiety reduction are said to be equal to effects of treatments including
meditation or relaxation [23].
One small randomised controlled trial has been conducted into exercise as
an intervention for mothers with depressive symptoms [25]. In this study, moth-
ers who participated in an exercise intervention showed a significant improve-
ment in depressive symptomatology compared to a control group. The mothers
in the exercise group also had improved physical fitness when compared to the
control group. The exercise intervention consisted of 2 group pram-walking
sessions a week with an additional pram-walking session to be done indepen-
dently each week. The exercise was of moderate intensity for 3040 min, which
is consistent with the guidelines for exercise protocols for anxiety described
above. The control group attended a support session once a week. The study
period was 12 weeks, which falls within the intervention period that appeared
to have maximum effects in the meta-analyses of exercise interventions for
depression previously described. While this study cannot distinguish whether
the reduction in depression was due to the exercise itself or other aspects of the
intervention, the finding that the groups did not differ in their reported levels of
social support before and after the study period of 12 weeks suggest that it is
unlikely that the social aspect of the exercise intervention is responsible for the
effects seen here. However, this was a small study (n 19) and more research
is needed to further investigate the benefits of exercise for postnatal depression.
Its benefits for anxiety in the postnatal period should also be examined.
Studies have not focused on the antenatal period and there are no specific
recommendations as to what type of exercise might be most suitable for pregnant
women. However, Da Costa et al. [26] looked at mothers reported physical activ-
ity during pregnancy and its relationship to depressed mood and state anxiety in
addition to other measures of psychological well-being. They started collecting
data on women in the third month of pregnancy. They found that in the first and
second trimesters, mothers who exercised reported less depressed mood and state
anxiety. Less state anxiety was reported by exercising mothers in the third
trimester. The overall findings appear to indicate that exercise may be a useful
means of increasing psychological well-being in mothers in the perinatal period,
but more research is needed in this area. It will also be important to examine what
type of exercise would be most suitable, particularly for mothers during pregnancy.
Cortisol is elevated during stress, anxiety and depression, but it is also
elevated in normal pregnancy. Bessinger et al. [27] report that moderate-intensity
exercise increased circulating cortisol levels more in women who were pregnant
than in women who had already delivered. This may indicate that moderate-
intensity exercise has a similar effect on pregnant women to intense exercise
at other times, since data from Jacks et al. [28] found that in healthy males,
cortisol only increased after exercise at high intensity for a long duration.
Previously, McMurray et al. [29] reported that pregnancy significantly alters
the metabolic responses to exercise in the water. They observed the usual
increase in cortisol levels with pregnancy but levels of cortisol remained lower
in the pregnant women during immersion and exercise.
These findings emphasise the need for research into exercise as a treatment
for anxiety and depression specifically in the perinatal period, and the need to
look for long-term effects on both mood and biology.
Touch and Massage in Anxiety and Depression
This has been the aspect of CAM which has been most studied in the
perinatal period. Tiffany Field [30] has been a pioneer in generating interest in
the area of touch and its therapeutic effects. She has reported that massage or
touch therapy has benefits for depression and anxiety in addition to facilitating
growth (for example, in premature infants) and enhancing immune function.
Massage in pregnancy has been found to reduce levels of anxiety-, stress- and
depression-related hormones, and has been reported to be superior in its effects
to relaxation therapy [31]. Field et al. [32] studied teenage mothers and found
massage to be an effective therapy for reducing their anxiety and depression.
Studies of brain activity have shown that receiving massage therapy reduces
frontal EEG symmetry in both mothers [33] and in infants of depressed moth-
ers who were massaged by a researcher trained in the procedure [34]. Right
frontal EEG asymmetry is a pattern associated with depression.
More recent research has found that, not only is there a reduction in dep-
ression in mothers who are taught infant massage, but they also have improved
interactions with their infants [13]. In this small study, mothers with postnatal
depression were all assigned to a support group, with one group of mothers also
attending baby massage classes. After the intervention, only those who attended
the massage class had statistically significantly improved interactions with their
babies; they improved into the normal range. Mothers who had attended the
massage class also had a greater reduction in their depression scores than the
controls, and their infants also scored more positively on the global interaction
scale [35]. This could partly be a result of mothers learning to interpret
their infants non-verbal cues in the massage classes. This is an important
finding as children of mothers who have suffered from postnatal depression
are reported to have impaired development [36, 37]; this is likely to be medi-
ated, at least in part, by the poor interactions between some depressed mothers
and their babies. An intervention, such as baby massage, that can improve the
interaction, may help combat the long-term effects of postnatal depression on
the child.
The results in this area are supported by animal research. Meaney et al.
[38] found lower levels of stress hormones in frequently handled rat pups, while
the touch/growth relationship has also been shown in rats [39]. It is further sug-
gested that massage may raise levels of oxytocin in both the mother and baby.
Massage therapy is known to increase plasma oxytocin in healthy adult women
[40], while tactile stimulation is reported to induce the release of oxytocin in
young dogs and in calves [41]. Oxytocin has been shown to increase maternal
behaviour in animal models [42] and it has been implicated in social bonding
[43]. Thus, baby massage could be helping to release oxytocin in both mothers
and babies and contribute to the development of a close mother-infant relation-
ship in this way.
Massage therapies, whether by the mother on her infant, or on the mother,
do not appear to have any contra-indications in the perinatal period. Of course,
the type of massage needs to be tailored to the recipient. In baby massage
classes run by the International Association for Infant Massage [44], this is an
important part of the teaching. Mothers are taught to be sensitive to their
infants cues and not to force massage on them unless they are receptive.
The oils used in massage also need to be considered. Aromatherapy has
been supposed as having an evidence base by the House of Lords Select
Committee on Science and Technology [16]. Amongst the suggested oils for
massage in postnatal depression are bergamot, lavender and rose [45], but
research is needed into their efficacy for helping women with postnatal depres-
sion. Research into what oils are appropriate in the antenatal period would also
be useful. In baby massage, oils used include sesame oil and organic vegetable
oils. At infant massage classes, a choice may be provided as some mothers, for
example, prefer to avoid nut oils in case of nut allergies.
Relaxation
Meditation/Yoga and Depression
A study conducted into the antidepressant effects of Sudarshan Kriya yoga
[46] has suggested that its effectiveness may be comparable to drug treatment
(imipramine) in severe hospitalised depressives. This form of yoga involves
rhythmic hyperventilation interspersed with normal breathing and concluding
with 1015 min of a tranquil state in a supine position. This type of breathing
has demonstrable effects on brain function, normalising an event-related
potential (P300) that is low in patients with severe endogenous depression [47].
This type of pattern has also been reported for antidepressant medications and
for electroconvulsive therapy [48, 49].
Relaxation has been shown to lower salivary cortisol levels in addition to
self-report measures of anxiety and perceived stress [50]. Practising yoga has
been shown to decrease serum cortisol levels in yoga instructors [51]. These
results, showing a biological effect of relaxation therapies, should encourage
more research into their effects as stand-alone interventions for anxiety.
Baby yoga is becoming increasingly popular, as shown by the number of
books and classes that are available. Baby yoga involves mothers handling their
babies and helping them stretch out. They may be put into certain poses and it
often incorporates a massage as a warm-up [52]. Babies who do baby yoga are
said anecdotally to be less likely to suffer from digestive problems such as
constipation and tend to be more relaxed. A happier, more relaxed baby is likely
to be easier for a mother to look after and this could help the relationship with a
depressed mother. However, there is no research in this area and this is needed
to establish and clarify any benefits it may have.
Relaxation and Anxiety
Research comparing relaxation with other therapies for anxiety is scarce,
possibly because relaxation is often seen as a technique to be used in conjunc-
tion rather than in its own right [12]. Relaxation training has been shown to be
effective as part of a treatment package for posttraumatic stress disorder or
generalised anxiety disorder [53, 54].
Research looking at the perinatal period is necessary. If this is positive,
these strategies, perhaps baby yoga in particular, could be most useful for
mothers and mothers-to-be suffering at this time. Including the baby in the
intervention seems practical and may make it easier for mothers to take part.
The research relating to the effects of touch are likely to be relevant in research
into baby yoga and it would also be interesting to look at any effects on mother-
infant interaction with this intervention.
Bright Light Therapy
Rosenthal et al. [55] at the National Institutes of Mental Health first
described seasonal affective disorder (SAD) and its treatment with bright light
therapy in 1984. Since then, bright light therapy has not only become estab-
lished as an evidence-based first-line treatment for SAD, but it has also been
used to treat various forms of non-seasonal depression with success [56]. Light
is usually administered using light boxes, which emit specific intensities of
light. Most currently used standard light boxes emit light intensities of
5,00010,000 lx, with minimal (less than 1%) ultraviolet emission. Patients
place the device on a table or desk and sit in front of the light box, with the
device approximately 2535 cm from eye level. Patients are instructed to sit in
front of the light for a prescribed period of time, usually ranging from 20min to
1 h, depending on the intensity of light exposure. The treatment is effective and
well tolerated, with few side-effects reported [57]. Ophthalmologic examina-
tions are recommended in individuals with a history of pre-existing ocular
problems. Some experts recommend that every new patient beginning treatment
with bright light therapy should have an eye exam [58].
The exact mechanism of action for bright light therapy is still unknown.
Bright light therapy induces phase shifts through its suppression of melatonin,
and this may account for some of its efficacy in treatment. Another mechanism
likely involves changes in the levels of various neurotransmitters in the hypo-
thalamic region of the brain. Serotonin has been most widely studied in this
regard, although dopamine has also been implicated [59].
Bright Light Therapy in the Treatment of Depression
during Pregnancy
One open study has reported on the use of bright light therapy in depressed
pregnant women [60]. In that study, Oren et al. [60] treated 16 pregnant women
with 60min of daily morning bright light therapy for 3 weeks. The treatment
was well tolerated, except in 2 patients, who experienced nausea during the
treatment. The nausea improved after light exposure was reduced from 60min
to 45min daily. Seven of the 16 patients were offered extended treatment with
bright light for 5 weeks. Patients were followed weekly during the course of
treatment and the main outcome measure was the SIGH-SAD score, a 29-item
modified Hamilton Depression Scale frequently used in treatment studies of
seasonal depression. In this cohort of pregnant women, improvement from
baseline was 49% after 3 weeks of treatment and 59% after 5 weeks of treat-
ment. The authors concluded that bright light therapy has an antidepressant
effect in pregnant women with depression. The improvement was more gradual
than that seen in the treatment of winter depression with bright light, but was
comparable with the time course for improvement generally seen when psy-
chopharmacological agents are used.
Bright Light Therapy in the Treatment of Postpartum Depression
There has only been one case report in the literature describing the use of
bright light therapy in postpartum depression [14]. Two women were treated
with 10,000 lx of morning bright light for 4 weeks. Both women responded to
treatment without side-effects and showed a 60% reduction in symptoms during
the course of treatment. One of the 2 women was nursing her infant.
A second small study investigating the use of bright light therapy in post-
partum women has recently been completed. This study was a double-blind,
randomised control study, comparing morning bright light (10,000lx) with
morning dim red light (600 lx) as the control condition [Corral et al., unpubl.
data]. Eighteen women were randomised into the study after a diagnosis of
postpartum depression was established and after meeting inclusion criteria and
consenting to participate. Ten women received bright light and 8 the red light
for a total of 5 weeks of treatment. Women used the light boxes at home
for 30min in the morning, between 07.00 and 09.00 h. The results are shown in
figure 1; two patients in the bright light group and 1 patient in the red light
group were excluded from the analysis because of missing data points.
The depression scores for the women treated with bright light dropped by
44%, whereas the women treated with red light had an improvement of 33%.
The improvement in the bright light group was much more consistent on a
week-by-week basis compared with the changes seen in the red light group,
which were more variable (fig. 1). There were no significant treatment differ-
ences between the two groups, but this was not unexpected given the small
sample size. Overall, the women tolerated the treatment well and were pleased
with the results. This small pilot study demonstrates that morning bright light
therapy shows promise in the treatment of postpartum depression and it
warrants further investigation in this patient population.
It may well be that bright light therapy is suitable for a specific subgroup
of women with postnatal depression. SAD has many features of atypical
depression such as oversleeping and carbohydrate craving, and it may be women
with these features who benefit.
Conclusion
There is some evidence for the efficacy of exercise, touch/massage,
yoga/meditation and bright light to treat anxiety and depression in general. In
the perinatal period, there is evidence for the benefit of touch therapies, espe-
cially massage, both for the mother and the baby as well as for their interaction.
40
30
20
10
0
N 8 8 8 8 8 7 7 7 7
Red light Bright light
Light condition
7
HAM-D 29 after 1 week
of light therapy
HAM-D 29 after 2 weeks
of light therapy
of light therapy
of light therapy
of light therapy
Fig. 1. Hamilton Depression Scale-29 items (HAM-D 29) scores in women with postna-
tal depression exposed either to bright light (n 8) or red light (n 7).
Evidence is also starting to show some benefit of the use of bright light therapy
for depression at this time.
Pregnancy and the time postpartum are periods during which depression
and anxiety are often present, and these disorders may also have an effect on the
fetus [61] and the baby [35]. CAM therapies are likely to be a popular choice
for women at this time. Clinicians and researchers working with these women
need to continue to explore novel and alternative treatments in a scientific and
controlled way.
References
1 Kessler RC, Soukup J, Davis R, Foster DF, Wickey SA, Van Rompay MI: The use of complementary
and alternative therapies to treat anxiety and depression in the United States. Am J Psychiatry
2001;158:189294.
2 Murray L: Childbirth as a life event: The Cambridge study of postnatal depression and infant
development; in Paykel ES (ed): Life Events and Postpartum Psychiatric Disorders. Southampton,
Southampton University Press, 1989.
3 Evans J, Heron J, Francomb H, Oke S, Golding J: Cohort study of depressed mood during and
after childbirth. BMJ 2001;323:257260.
4 Boath L, Bradley E, Henshaw C: The treatment of antenatal depression: A systematic review. The
Marc Society Newsletter, 2003, vol 17, p 6.
6 Hendrick V, Smith LM, Suri R, Hwang S, Haynes D, Altshuler L: Birth outcomes after prenatal
exposure to antidepressant medication. Am J Obstet Gynecol 2003;188:812815.
7 Burt VK, Suri R, Altshuler L, Stowe Z, Hendrick VC, Muntean E: The use of psychotropic
medications during breast-feeding. Am J Psychiatry 2001;158:10011009.
8 Kulin NA, Pastuszak A, Sage SR, Schick-Boschetto B, Spivey G, Feldkamp M, Ormond K,
Matsui D, Stein-Schechman AK, Cook L, Brochu J, Rieder M, Koren G: Pregnancy outcome
following maternal use of the new selective serotonin reuptake inhibitors: A prospective controlled
multicentre study. JAMA 1998;279:609610.
9 Nulman I, Rovet J, Stewart DE, Wolpin J, Gardner HA, Theis JG, Kulin NA, Koren G:
Neurodevelopment of children exposed in utero to antidepressant drugs. N Engl J Med 1997;336:
258262.
10 Nulman I, Rovet J, Stewart DE, Wolpin J, Pace-Asciak P, Shuhaiber S, Koren G: Child development
following exposure to tricyclic antidepressants or fluoxetine throughout fetal life: A prospective,
11 Hendrick VC, Fukuchi A, Altshuler L, Widawski M, Wertheimer A, Brunhuber MV: Use of sertraline,
paroxetine and fluvoxamine by nursing women. Br J Psychiatry 2001;179:163166.
12 Larzelere M, Wiseman P: Anxiety, depression and insomnia. Prim Care 2002;29:339360.
13 Onozawa K, Glover V, Adams D, Modi N, Kumar R: Infant massage improves mother-infant inter-
action for mothers with postnatal depression. J Affect Disord 2001;63:201207.
14 Corral MR, Kuan A, Kostaras D: Bright light therapys effect on postpartum depression. Am J
Psychiatry 2000;157:303304.
15 Gerrard J: Postnatal Depression. London, NT Books, Emap Healthcare, 2000.
16 Mantle F: The role of alternative medicine in treating postnatal depression. Complement Ther
Nurs Midwifery 2002;8:197203.
17 Richardson J: Evidence-based complementary medicine: Rigor, relevance, and the swampy lowlands.
J Altern Complement Med 2002;8:221223.
18 Vuckovic N: Integrating qualitative methods in randomized controlled trials: The experience of
the Oregon center for complementary and alternative medicine. J Altern Complement Med
2002;8:225227.
19 Wilson K, Mills E: Closing comment: Evidence-based complementary and alternative medicine:
Is it a viable concept? J Altern Complement Med 2002;8:875876.
20 Craft L, Landers DM: The effect of exercise on clinical depression and depression resulting from
mental illness: A meta-analysis. J Sport Exerc Psychol 1998;20:339357.
21 North TC, McCullagh P, Tran ZV: Effects of exercise on depression. Exerc Sports Sci Rev
1990;18:379415.
22 Chaouloff F: Effects of acute physical exercise on central serotonergic systems. Med Sci Sports
Exerc 1997;29:5862.
23 Palushka SA, Schwenk TL: Physical activity and mental health: current concepts. Sports Med
2000;29:167180.
24 Petruzello SJ, Landers DM, Hatfield BD, et al: A meta-analysis on the anxiety-reducing effects of
acute and chronic exercise: Outcomes and mechanisms. Sports Med 1991;11:143182.
25 Armstrong K, Edwards H: The effectiveness of a pram-walking exercise programme in reducing
depressive symptomatology for postnatal women. Int J Nurs Pract 2004;10:177194.
26 Da Costa D, Rippen N, Dritsa M, Ring A: Self-reported leisure-time physical activity during
pregnancy and relationship to psychological well-being. J Psychosom Obstet Gynecol 2003;24:
111119.
27 Bessinger RC, McMurray RG, Hackney AC: Substrate utilization and hormonal responses to
moderate intensity exercise during pregnancy and after delivery. Am J Obstet Gynecol 2002;186:
757764.
28 Jacks DE, Sowash J, Anning J, McGloughlin T, Andres F: Effect of exercise at three exercise inten-
sities on salivary cortisol. J Strength Cond Res 2002;16:286289.
29 McMurray RG, Katz VL, Berry MJ, Cefalo RC: The effect of pregnancy on metabolic responses
during rest, immersion, and aerobic exercise in the water. Am J Obstet Gynecol 1988;158:481486.
30 Field T: Massage therapy effects. Am Psychol 1998;53:12701281.
31 Hernandes-Reif M, Dieter J, Field T, Swedlow B, Diego M: Migraine headaches are reduced by
massage therapy. Int J Neurosci 1998;96:111.
32 Field T, Morrow C, Valdern C, Larson S, Kuhn C, Schanberg S: Massage reduces anxiety in child
and adolescent psychiatric patients. J Am Acad Child Adolesc Psychiatry 1992;31:125131.
33 Jones NA, Field T: Massage and music therapies attenuate frontal EEG asymmetry in depressed
adolescents. Adolescence 1999;34:529534.
34 Jones NA, Field T, Davalos M: Massage therapy attenuates right frontal EEG asymmetry in
one-month-old infants of depressed mothers. Infant Behav Dev 1998;21:527530.
35 Murray L, Stanley C, Hooper R, King F, Fiori-Cowley A: The role of infant factors in postnatal
depression and mother-infant interactions. Dev Med Child Neurol 1996;38:109119.
36 Beck CT: The effects of postpartum depression on child development: A meta-analysis. Arch
Psychiatr Nurs 1998;12:1220.
37 Murray L, Cooper P: Intergenerational transmission of affective and cognitive processes associated
with depression: Infancy and the preschool years; in Goodyer IM (ed): Unipolar Depression:
A Lifespan Perspective. Oxford, Oxford University Press, 2003.
38 Meaney MJ, Aitken DH, Bhatnagar M, Bodnoff SR, Mitchell JB, Sarrieau A: Neonatal handling
and the development of the adrenocortical reponse to stress; in Field T (ed): Advances in Touch.
Skillman, Johnson and Johnson, 1990.
39 Schanberg S: Genetic basis for touch effects; in Field T (ed): Touch in Early Development.
Hillsdale, Erlbaum, 1994, pp 6780.
40 Turner RA, Altemus M, Enos T, Cooper B, McGuinness T: Preliminary research on plasma
oxytocin in normal cycling women: Investigating emotion and interpersonal distress. Psychiatry
1999;62:97113.
41 Uvnas-Moberg K: Oxytocin may mediate the benefits of positive social interaction and emotions.
Psychoneuroendocrinology 1998;23:819835.
42 Levy F, Kendrick KM, Goode JA, Guevara-Guzman R, Keverne EB: Oxytocin and vasopressin
release in the olfactory bulb of parturient ewes: Changes with maternal experience and effects on
acetylcholine, gamma-aminobutyric acid, glutamate and noradrenaline release. Brain Res
1995;669:197206.
43 Carter CS: Developmental consequences of oxytocin. Physiol Behav 2003;79:383397.
44 International Association for Infant Massage UK homepage: http://www.iaim.org.uk/. In
vol 2003:IAIM.
45 Tiran D: Massage and aromatherapy; in Mack S (ed): Complementary Therapies for Pregnancy
and Childbirth. Edinburgh, Baillere Tindall, 2000, pp 129167.
46 Janakiramaiah N, Gangadhar PJ, Murthy NV, Harish MG, Subbakrishna DK, Vedamurthachar A:
Antidepressant efficacy of Sudarshan Kriya Yoga (SKY) in melancholia: A randomised comparison
with electroconvulsive therapy (ECT) and imipramine. J Affect Disord 2000;57:255259.
47 Naga Venkatesha Murthy PJ, Janakiramaiah N, Gangadhar BN, Subbakrishna DK: P300 ampli-
tude and antidepressant response to Sudarshan Kriya Yoga (SKY). J Affect Disord 1998;50:4548.
48 Blackwood DH, Whalley LJ, Christie JE, Blackburn IM, St Clair DM, McInnes A: Changes in
auditory P3 event-related potential in schizophrenia and depression. Br J Psychiatry 1987;150:
154160.
49 Gangadhar BN, Ancy J, Janakiramaiah N, Umapathy C: P300 amplitude in non-bipolar, melan-
cholic depressives. J Affect Disord 1993;28:5760.
50 Pawlow LA, Jones GE: The impact of abbreviated progressive muscle relaxation on salivary
cortisol. Biol Psychol 2002;60:116.
51 Kamei T, Toriumi Y, Kimura H, Ohno S, Kumano H, Kimura K: Decrease in serum cortisol during
yoga exercise is correlated with alpha wave activation. Percept Mot Skills 2000;90:10271032.
52 Freedman FB: Baby Yoga: Gentle Exercise for Babies, Mums and Dads. London, Gaia Books,
2000.
53 Barlow DH, Esler JL, Vitali AE: Psychosocial treatments for panic disorders, phobias and gener-
alized anxiety disorder; in Gorman JM (ed): A Guide to Treatments That Work. New York, Oxford
University Press, 1998, pp 288318.
54 Keane TM: Psychological and behavioral treatments for post-traumatic stress disorder; in Gorman
JM (ed): A Guide to Treatments That Work. New York, Oxford University Press, 1998, pp 398407.
55 Rosenthal NE, Sack DA, Gillin JC, Lewy AJ, Goodwin FK, Davenport Y, Mueller PS, Newsome
DA, Wehr T: Seasonal affective disorder. A description of the syndrome and preliminary findings
with light therapy. Arch Gen Psychiatry 1984;41:7280.
56 Kripke DF: Light treatment for nonseasonal depression: Speed, efficacy and combined treatment.
J Affect Disord 1998;49:109117.
57 Partonen T: Light therapy; in Magnusson A (ed): Seasonal Affective Disorder: Practice and
Research. Oxford, Oxford University Press, 2001.
58 Terman M, Rem CE, Rafferty B, Gallin PF, Terman JS: Bright light therapy for winter depression:
Potential ocular effects and theoretical implications. Photochem Photobiol 1990;51:781792.
59 Neumeister A, Konstantinidis A, Praschak-Reider N, Willeit M, Hilger E, Statsny J, Kasper S:
Monoamines; in Magnusson A (ed): Seasonal Affective Disorder: Practice and Research. Oxford,
Oxford University Press, 2001.
60 Oren DA, Wisner KL, Spinelli M, Epperson CN, Peindl KS, Terman JS, Terman M: An open trial of
morning light therapy for treatment of antepartum depression. Am J Psychiatry 2002;159:666669.
61 OConnor TG, Heron J, Golding J, Glover V: Maternal antenatal anxiety and behavioural/emotional
problems in children: A test of a programming hypothesis. J Child Psychol Psychiatry 2003;44:
10251036.
Madeleine OHiggins
Institute of Reproductive and Developmental Biology, Imperial College London
Du Cane Road
London W12 0NN (UK)
Tel. 44 207 594 2136, Fax 44 207 594 2138, E-Mail v.glover@imperial.ac.uk
194
Ahokas, A. 100
Aito, M. 100
Beach, A.J. 70
Brockington, I. 1
Corral, M. 182
Glover, V. 182
Gunasekera, S. 112
Henry, A.L. 70
Hofecker-Fallahpour, M.
167
Kaukoranta, J. 100
Matthews, S.G. 28
Meaney, M.J. 28
Newport, D.J. 70, 137
OHara, M.W. 150
OHiggins, M. 182
Ragan, K. 137
Riecher-Rssler, A. VII, 6,
167
Rohde, A. 6
Ross, L.E. 112
Rowland, M. 112
Spinelli, M.G. 85
Steiner, M. VII, 112
Stowe, Z.N. 70, 137
Stuart, S. 150
Wadhwa, P.D. 50
Wahlbeck, K. 100
Author Index
195
Subject Index
Anxiety disorders
benzodiazepine management in
pregnancy 119, 120
cognitive behavioral therapy 162, 164
complementary and alternative medicine,
see Complementary and alternative
medicine
demands of early motherhood 168,
169
exercise effects 184, 185
interpersonal psychotherapy, see
Interpersonal psychotherapy
postpartum 19, 20
pregnancy 21
relaxation therapy 188
Aromatherapy, anxiety and depression
studies 187
Attention deficit disorder (ADD)
dopamine system gene polymorphisms
34, 35
familial risks 39
intrauterine growth restriction risks 29, 34
Baby blues
definition 7
frequency 7
Basel group therapy for mothers with
depression, see Group therapy for
mothers with depression
Behavioral perinatology
maternal-placental-fetal neuroendocrine
axis 57
prenatal stress and fetal outcomes
biobehavioral model 5255
epidemiological findings 55, 56
placental corticotropin-releasing
hormone role
fetal growth 58
fetal neurodevelopment 6062
immune-inflammatory processes in
pregnancy 5860
preterm labor effects 57
stress sensitivity 62
prospects for study 6265
scope 50, 51
Benzodiazepines
long-term neurodevelopment effects of
prenatal exposure 129, 130
neonatal complications 120, 121
teratogenic effects 120
Bipolar disorder, management in pregnancy
123, 124
Bonding, mother and infants 15
Breast-feeding
advantages 138
hormonal changes 138
postpartum depression management
antidepressants
adverse effects 142, 144, 145
clinical decision-making 142, 143,
145
infant exposure quantification
139142
studies 139, 140
concerns 137
Bright light therapy, see Light therapy
Carbamazepine
Chlorpromazine
neonatal complications 126
teratogenic effects 124, 125
Clozapine, safety in pregnancy 126, 128,
129
Cognitive behavioral therapy (CBT)
anxiety management 162, 164
group therapy in perinatal depression
171, 172
160, 161, 167
Complementary and alternative medicine
(CAM)
aromatherapy 187
definition 183
exercise 184, 185
light therapy 188190
massage 186, 187
pharmacotherapy attitudes in postpartum
period 182, 183
popularity in treatment of maternal
depression and anxiety 182
prospects 190, 191
relaxation therapy 187, 188
research concerns 183, 184
Corticotropin-releasing factor (CRF),
intrauterine growth retardation infants
cord blood levels 30
fetal adversity and neural development
31, 32
postnatal programming of neural
development 37, 38
stress response mediation 31
Corticotropin-releasing hormone (CRH),
prenatal stress and fetal outcome role of
placental hormone
course of expression 54
fetal growth 58
fetal neurodevelopment 6062
immune-inflammatory processes in
pregnancy 5860
overview 52, 53
parturition role 57, 58
preterm labor effects 58
stress sensitivity 62
Cortisol
exercise response in pregnant women 185
relaxation effects 187
Depression, postpartum
animal offspring studies of postpartum
stress exposure 75, 76, 78, 79
demands of early motherhood 168, 169
diagnostic classification
depression versus psychosis 16, 17,
22, 23
DSM-IV 8, 9
ICD-10 8
limitations 137
methodological problems 6, 7
specificity requirements
consequences 13, 14
course 13
etiology 1113
therapeutic needs 1416
symptomatology 9, 10
frequency 7, 11, 72, 100, 168
historical perspective 2, 3
infant effects 74, 75
management
cognitive behavioral therapy 160, 161,
167, 171, 172
complementary and alternative
medicine, see Complementary and
alternative medicine
delay 14, 23
estrogen
formulations and regimens 104107
overview 16, 104
group psychotherapy, see Group
therapy for mothers with depression
interpersonal psychotherapy, see
Interpersonal psychotherapy
nursing women, see Breast-feeding
pharmacotherapy 119, 120, 171
pathoplastic influence of early
motherhood 170
suicide, see Suicide, perinatal
Depression, pregnancy
fetal effects 7274
interpersonal psychotherapy 163
light therapy 189
pharmacotherapy safety, see specific drugs
Subject Index 196
Diathesis/stress model of illness
maternal depression incorporation 76, 77
overview 70, 71
Dopamine, estradiol effects 103
Electroencephalography (EEG)
child studies following prenatal
depression 73, 74
massage studies 186
Estrogen
central nervous system effects 102, 103
changes in female reproduction 12, 101
mental effects during breast-feeding
138
formulations and regimens 104107
overview 16, 104
psychoprotectant effect 12
receptors 102, 103
Exercise
anxiety effects 184, 185
depression effects 184, 185
Family environment
environmental adversity and parental
care 40, 41
maternal depression and stress, see
Depression, postpartum
development 3539
Filicide, features 94, 95
Gabapentin, teratogenic effects 123
Genetic susceptibility, peripartum disorders
11, 12
Group therapy for mothers with depression
(GMD)
contents of sessions 173
development 172
evaluation study
instruments 175
limitations 178
outcomes 176178
patient inclusion and exclusion criteria
174
sample 173175
setting 174
statistics 175, 176
rationale 167, 168
therapeutic techniques 172, 173
Haldol (haloperidol)
Hippocampus, stress effects on volume 32,
33
Hospitalization, indications 16
Hypothalamic-pituitary-adrenal (HPA) axis
animal studies of stress and
developmental effects 3739
behavioral perinatology, see Behavioral
perinatology
environmental adversity and parental
care 40, 41
intrauterine growth restriction
fetal programming of gene expression
30, 31
glucocorticoid role 29, 30
neural development and fetal adversity
3134
ovarian axis interactions 102
development 3539
pregnancy changes 102
reversibility of developmental effects of
fetal adversity 4143
Infanticide, perinatal
filicide 94, 95
frequency 85
historical perspective 13, 8587
infanticide-suicide 9497
legal issues
gender influences 9193
insanity law in United States 88, 89
overview 86, 87
research-based information needs
9193
United Kingdom 89, 90
Yates case 87, 88, 9294
Insanity, law in United States 88, 89
Interpersonal psychotherapy (IPT)
depression treatment efficacy 154,
155
group therapy 171
overview and principles 152154
Subject Index 197
perinatal psychiatric disorder trials
postpartum depression
acute treatment 158162, 167, 170,
171
prevention 155158
prepartum depression 163
prospects 164
perinatal social support impact on
Intrauterine growth restriction (IUGR)
attention deficit disorder risks 29, 34
behavioral disorder risks 34, 35
fetal programming of gene expression
30, 31
glucocorticoid role 29, 30
hippocampal effects 3234
metabolic syndrome risks 28
neural development and fetal adversity
3134
outcomes and variability 44
risk factors 28
Lactation, see Breast-feeding
Lamotrigine, teratogenic effects 123
Light therapy
depression management
postpartum depression 189, 190
prepartum depression 189
seasonal affective disorder 188
principles 188
Lithium
Manualized group therapy for mothers with
depression, see Group therapy for
mothers with depression
Massage
anxiety and depression studies 186, 187
baby massage 186, 187
Metabolic syndrome, intrauterine growth
restriction risks 28
Model Penal Code (MPC), insanity law 88,
89
Monoamine oxidase
estradiol effects 103
inhibitor safety in pregnancy 118, 119
progesterone effects 104
Mood disorders
management in pregnancy 124129
pregnancy 20, 21
Nefazodone, safety in pregnancy 119
Noradrenaline, estradiol effects 103
Obsessive-compulsive disorder (OCD)
clinical features 19
pathoplastic influence of early
motherhood 170
Olanzapine, safety in pregnancy 126, 127
Oxytocin
baby massage effects 186, 187
mental effects during breast-feeding 138
Panic disorder
postpartum 19, 20
pregnancy 21
Posttraumatic stress disorder (PTSD),
postpartum
diagnostic classification 18, 19
frequency 18
hippocampal volume effects 33
historical perspective 2
predisposing factors 18, 19, 33
symptoms 18
treatment 19
Progesterone
central nervous system effects 104
changes in pregnancy 101
Prolactin, mental effects during
breast-feeding 138
Psychosis, postpartum
diagnostic classification
depression versus psychosis 16, 17,
22, 23
ICD-10 8
methodological problems 6, 7
specificity requirements
consequences 13, 14
course 13
etiology 1113
therapeutic needs 1416
Subject Index 198
Interpersonal psychotherapy (IPT)
(continued)
symptomatology 10
treatment delay 14
frequency 7, 10, 11
historical perspective 1
Psychosis, pregnancy 21
Psychosocial stressors, postpartum period
13, 168, 169
Psychotherapy, approaches 15
Quetiapine, safety in pregnancy 126, 128
Risperidone, safety in pregnancy 126, 127
Schizophrenia, management in pregnancy
124129
Seasonal affective disorder (SAD), light
therapy 188
(SSRIs)
birth outcomes 113, 114
prenatal exposure 129
postpartum depression management in
nursing women
adverse effects 142, 144, 145
clinical decision-making 142, 143, 145
infant exposure quantification 139142
studies 139, 140
spontaneous abortion risk 114
Serotonin
estradiol effects 103
exercise effects 184
transporter gene variants and infant
behavior 35
Social support, perinatal support impact on
Stress, see Behavioral perinatology;
Hypothalamic-pituitary-adrenal axis
Suicide, perinatal
filicide 94, 95
frequency 94, 96, 97
infanticide-suicide 9497
motivation 94, 95
risk factors 96
Topiramate, teratogenic effects 123
Touch therapy, anxiety and depression
studies 186, 187
Trazodone, safety in pregnancy 119
Tricyclic antidepressants (TCAs)
prenatal exposure 129
Uncertainty principle, ethical question of
antenatal clinical research 78
Valproic acid
Venlafaxine, safety in pregnancy 119
Vulnerability-stress model, postpartum
disorders 12, 14
Yates, Andrea 87, 88, 9294
Yoga
baby yoga 187, 188
depression studies 187
Subject Index 199

A. Riecher-Rossler, Meir Steiner Perinatal Stress, Mood and Anxiety Disorders From Bench To Bedside Bibliotheca Psychiatrica 2005

Enviado por

Dados do documento

Título original

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

A. Riecher-Rossler, Meir Steiner Perinatal Stress, Mood and Anxiety Disorders From Bench To Bedside Bibliotheca Psychiatrica 2005

Enviado por

Direitos autorais:

Formatos disponíveis

Perinatal Stress, Mood and Anxiety Disorders

currents through GABA

Fig. 2. Placental CRH in human pregnancy. In pregnancy, the placenta is a major

Você também pode gostar