®

PSYCHIATRY BOARD REVIEW MANUAL PUBLISHING STAFF PRESIDENT, PUBLISHER

Bruce M.White

Peripartum Mood Disorders

EXECUTIVE EDITOR

Debra Dreger SENIOR EDITOR

Becky Krumm, ELS

Series Editor and Contributing Author: Jerald Kay, MD Professor and Chair, Department of Psychiatry,Wright State University School of Medicine, Dayton, OH

ASSISTANT EDITOR

Deidre Yoder EDITORIAL ASSISTANT

Matthew T. Patton

Contributing Author: Deborah Y. Liggan, MD Resident, Department of Family Practice, Bethesda Hospital, Inc., Cincinnati, OH

SPECIAL PROGRAMS DIRECTOR

Barbara T.White, MBA PRODUCTION MANAGER

Suzanne S. Banish

Table of Contents

PRODUCTION ASSISTANTS

Tish Berchtold Klus Christie Grams Mary Beth Cunney

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Case Presentation . . . . . . . . . . . . . . . . . . . . . . . . 2

ADVERTISING/PROJECT MANAGER

Patricia Payne Castle

Spectrum of Disorders . . . . . . . . . . . . . . . . . . . . 2 Diagnostic Approach . . . . . . . . . . . . . . . . . . . . . . 4

NOTE FROM THE PUBLISHER: This publication has been developed without involvement of or review by the American Board of Psychiatry and Neurology.

Endorsed by the Association for Hospital Medical Education The Association for Hospital Medical Education endorses HOSPITAL PHYSICIAN for the purpose of presenting the latest developments in medical education as they affect residency programs and clinical hospital practice.

Pathophysiology . . . . . . . . . . . . . . . . . . . . . . . . . 5 Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Summary Points . . . . . . . . . . . . . . . . . . . . . . . . . 11 Board Review Questions. . . . . . . . . . . . . . . . . . . 11 Answers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Cover Illustration by Vanessa Ray

Copyright 2000, Turner White Communications, Inc., 125 Strafford Avenue, Suite 220, Wayne, PA 19087-3391, www.turner-white.com. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, mechanical, electronic, photocopying, recording, or otherwise, without the prior written permission of Turner White Communications, Inc. The editors are solely responsible for selecting content. Although the editors take great care to ensure accuracy, Turner White Communications, Inc., will not be liable for any errors of omission or inaccuracies in this publication. Opinions expressed are those of the authors and do not necessarily reflect those of Turner White Communications, Inc.

Psychiatry Volume 4, Part 2 1

PSYCHIATRY BOARD REVIEW MANUAL

Peripartum Mood Disorders I. INTRODUCTION Pregnancy has been described as a time of emotional well-being during which many psychiatric disorders become quiescent. In truth, the reproductive years represent a time of enhanced vulnerability for the onset or reemergence of mood disorders that range from the minor, commonly occurring postpartum blues to manic episodes with severe psychotic symptoms. The erroneous belief that one should be happy about giving birth adds to the shame and stigma of mood disorders. Consequently, a pregnant woman or adolescent may be reticent to reveal her emotional pain for fear that disclosure will result in social isolation, a judgment that she is unfit to care for her child, or hospitalization. There is overwhelming evidence that the biologic, psychologic, and psychosocial aspects of pregnancy exert a varied effect on mood disorders. During pregnancy, the placenta excretes estrogens and progesterone into the maternal system. With the removal of the placenta at delivery, estrogen and progesterone levels drop sharply, reaching pregravid levels by the fifth postpartum day. The impact of these hormonal changes on the course of mood disorders is unclear. However, growing clinical experience suggests that mood may be labile at the onset of pregnancy. For many patients who have either unipolar or bipolar disorder, the mood disorder stabilizes as the pregnancy advances but is followed by an increased postpartum risk for recurrence of the specific disorder. However, it is not uncommon for women to become symptomatic during the pregnancy. Emotional experiences vary during the 3 trimesters of pregnancy, thus contributing to mood changes. A woman’s feelings during the first trimester appear to be determined mainly by the physiologic changes that she is undergoing. During the last trimester, feelings are influenced by the approach-avoidance conflict regarding delivery. For the psychologically vulnerable patient, the stress of approaching delivery may be characterized by anxiety and increased emotional lability, tension, irritability, nightmares, depression, and insomnia.1 More than 50% of females who have depression or mania never receive psychiatric attention, even though their contact with health services is greater around the

2 Hospital Physician Board Review Manual

time of pregnancy and childbirth than at any other time. When mood disorders are not recognized or treated, parent health, parent-infant interaction, child development, and family stability may be seriously compromised. In extreme cases, untreated mood disorder may result in maternal suicide or infanticide.

II. CASE PRESENTATION A 27-year-old pregnant mother of two is brought in to an appointment with a psychiatrist by her husband, who states that she has been increasingly irritable and impatient with the children. She becomes easily distracted when they ask her a question, and she no longer reads to the children at bedtime, an activity that had been a nightly ritual. The patient has been eating poorly, and she sleeps approximately 3 to 4 hours each night; sometimes she completes household chores and pays bills during the early hours of the morning. The patient’s husband also notes that conversations have become increasingly one-sided. Often he cannot follow the discussion because his wife jumps from one topic to another. These symptoms have progressed over the past 5 to 6 days. The patient is starting her second trimester of pregnancy. Her current medications are limited to prenatal vitamins. During her last pregnancy she was both mildly hypertensive and hyperglycemic, although that child was born without neonatal distress and weighed 7 lb, 11 oz. The husband is concerned because his wife had one serious depressive episode at the age of 19 that required antidepressant therapy. Her symptoms at that time were sleeplessness, fatigue and lethargy, frequent crying spells, and weight loss (approximately 10 pounds over a 2-month period).

III. SPECTRUM OF DISORDERS • What diagnostic criteria differentiate the patient’s symptoms from other affective disorders? • How does pregnancy affect the course of mood disorders? • What factors predispose a female to mood disorders during pregnancy or the postpartum period?

Peripartum Mood Disorders DEPRESSION Depression is the most common psychiatric illness seen in pregnancy, with a reported prevalence of 9% to 16% in pregnant women in the United States. The importance of this mental health issue lies in its potential impact on mother-child attachment, parenting attitudes and behaviors, and newborn irritability.2 The diagnosis of depression during or after pregnancy is established using the same criteria as those for nongravid patients. Unipolar major depression is defined by the presence of 5 or more symptoms during the same 2-week period that represent a change from previous functioning. At least 1 of the symptoms must be depressed mood or loss of interest or pleasure in daily activities. Other symptoms include significant weight loss, insomnia or hypersomnia, psychomotor retardation or agitation, feelings of worthlessness or inappropriate guilt, diminished concentration, or recurrent thoughts of death. Symptoms of perinatal depression vary in intensity, frequency, and duration, and they intensify when sleep deprivation occurs. These symptoms must cause clinically significant distress or impairment in social or occupational functioning. Severe depression may be complicated by delusions or hallucinations. The course of depression during pregnancy has not been systematically evaluated. Some studies report a tendency toward improvement of depression as pregnancy progresses; others report that the highest level of depressive symptomatology occurs in the third trimester. Because depression is a recurrent disorder in the majority of patients, a history of major mood disorder increases risk for subsequent depression, particularly when this has occurred during a previous pregnancy. Patients should be evaluated for maternal risk factors for perinatal depression (Table 1). Postpartum Depression Overwhelming evidence shows that the onset of depression peaks in the postpartum period. However, controversy exists regarding 2 diagnostic facets of postpartum depression. First, many investigators believe that postpartum depression represents a separate entity from unipolar major depression. The task force associated with the DSM-IV concurred with this view by using a modifier for coding postpartum depression. Second, the DSM-IV has somewhat rigorously defined postpartum depression as depression occurring within 4 weeks of delivery. However, this strict criterion is not used in most epidemiologic studies. Throughout the literature, postpartum depression is regarded as a major depressive episode that occurs within the first 6 months after delivery.

Table 1. Maternal Risk Factors for Perinatal Depression Depression during previous pregnancy or postpartum course History of depression before pregnancy History of premenstrual syndrome Maternal family history of Axis I disorders, especially affective disorders Maternal or paternal alcoholism An unsupportive partner or marital distress Adverse life events (sexual abuse, fetal/infant death) Poor social support Ambivalence about the pregnancy

The mildest of the postpartum mood disturbances, termed postpartum blues, is experienced by up to 85% of new mothers within the first 2 weeks after delivery. Peak symptoms occur between the third and seventh day after delivery, with patients commonly complaining of insomnia, anxiety, tearfulness, headaches, irritability, appetite changes, and feeling overwhelmed and oversensitive. Postpartum blues often has a minor functional impact and resolves over 2 weeks to 3 months. Symptoms resolve spontaneously without sequelae; therefore, this condition does not require treatment.3 Adolescent Depression Pregnancy during adolescence can consolidate the resolution of the normal separation/individuation developmental crisis of puberty. It can also revive primitive anxieties and conflicts that lead to maladaptive behaviors as well as negative emotional states.4 Of particular concern is the occurrence of depression in the especially vulnerable population of pregnant adolescents. The role transition into early parenthood causes many adolescent mothers to experience stress beyond that normally associated with childbearing. These new mothers already have diminished coping abilities before delivery. During the puerperium, adaptation to motherhood and the infant’s demands requires coping skills not yet developed. This may lead the mother into depression, which further diminishes her ability to adapt to the new role. Adolescents in particular tend to minimize their stressors and interpret their psychological state in moral terms. This makes them less likely to seek psychiatric help and consequently prevents them from receiving appropriate treatment. Accordingly, study results have revealed a tendency for depressed young mothers to be hostile, indifferent, and rejecting

Psychiatry Volume 4, Part 2 3

Peripartum Mood Disorders of their children.5 These reports also identified selfesteem and social support as the primary predictors of depressive symptoms in pregnant adolescents. BIPOLAR DISORDER Bipolar disorder is marked by distinct periods of depression and mania or hypomania. Mania manifests itself as a state of persistently elevated, expansive, or irritable mood that lasts for at least 1 week. The manic period is associated with pressured speech, flight of ideas, distractibility, increased goal-directed activities, and excessive involvement in high-risk activities. Either the depressive or manic phases can have signs of psychosis. Hypomania, which is characterized by milder symptoms of excitement of less than 1 week’s duration with no psychotic features, is not severe enough to cause impairment in social or occupational functioning. A variety of data exist regarding the clinical course of bipolar disorder during pregnancy and the postpartum period. First, despite case reports that describe bipolar patients who experience clinical mood stability during pregnancy, there is no empirical evidence that pregnancy offers protection from the onset or relapse of bipolar disorder. In fact, studies associating bipolar disorder with puerperal worsening of mood have reported rates of relapse as high as 30% to 50%. Second, multiple studies have shown that bipolar females are at high risk of having an affective episode in the postpartum period. Third, rapid cycling (four or more affective episodes per year) has been associated with pregnancy. Rapid cycling occurs in 13% to 20% of all patients suffering from bipolar mood disorders. In many females, rapid cycling begins in the postpartum period. One theory posited to explain why parturition has been associated with the induction of rapid cycling is the “kindling model.” When postpartum hormonal changes trigger a first episode in genetically predisposed females, changes produced by that first episode render these females more vulnerable to subsequent episodes.6 SUICIDE AND PREGNANCY The most serious possible consequences of mood dysfunction are suicide attempts and completed suicide. Although pregnancy may provide little protection against attempted suicide, it has been argued that completed suicide during pregnancy is uncommon and less than expected for females of a similar age. The hypothesis is that pregnancy may exert a behavioral inhibiting effect on suicidal urges.7 Clinically, this theory is difficult to substantiate. However, it is known that most suicide attempts appear to occur early in pregnancy. Motivating factors cited by females who have attempted suicide

4 Hospital Physician Board Review Manual

during pregnancy include prior loss of children, potential loss of a lover, and the desire for an abortion. Immediate hospitalization should be provided in the presence of psychotic symptoms that threaten maternal or fetal safety or suicidal thoughts in which a method, time, and place is chosen.

IV. DIAGNOSTIC APPROACH • Who should be assessed for mood disorders in pregnancy? • What tools can assist in screening for perinatal mood disorders? SCREENING FOR MOOD DISORDERS IN PREGNANCY Beginning with the first trimester of pregnancy, all pregnant patients should be assessed by either their obstetrician/gynecologist or their primary care physician for depression or risk factors for mood disorder. Universal screening can be accomplished with one of the readily available depression screening instruments (eg, the Hamilton Anxiety and Depressive Scales, the Beck Depression Inventory, and the Zung Anxiety and Depressive Self-Assessment Scale) or by inquiring about the cardinal signs of major depression (eg, depressed mood, incapacity to experience pleasure, or diminished interest). In the case presented here, the patient’s current depressive episode could have been detected earlier in her pregnancy through screening questions regarding decreased need for sleep, irritability, and distractibility. Several specific postpartum depression scales are also available, including the Postpartum Support Questionnaire and the Edinburgh Postnatal Depression Scale (EPDS). For each question on the EPDS, the mother is asked to choose which of 4 possible responses best describes how she has been feeling during the past week. Mothers who score above a threshold of 12 or 13 (of a maximum 30) are likely to be suffering from a depressive illness. Severity is indicated by score. After parturition, it is best to continue to screen during wellchild examinations and postnatal maternal visits. Mothers who develop depression during pregnancy or who have risk factors should receive close follow-up postpartum screening. Pregnant women and adolescents who exhibit symptoms or have increased risk for affective disorder should also receive education about symptoms and the complicated interface between mood disorder and pregnancy. In this way, they will be able to actively participate in defining the general goals of therapy, should it be

Peripartum Mood Disorders needed. Active education and involvement of the patients’ partners will also contribute to family stability and patient recovery. DIFFERENTIAL DIAGNOSIS In most cases, the diagnostic work-up for mood disorder should begin with a careful history, thorough physical examination, and laboratory tests that include determination of serum hemoglobin and thyroid-stimulating hormone (TSH). Advanced central nervous system (CNS) imaging is reserved for the presence of neurologic signs and symptoms. Certain medical conditions should be ruled out before considering a diagnosis of depression. For example, fatigue may be an indication of anemia caused by excessive blood loss during delivery or sleep loss associated with nocturnal child care. Loss of initiative, mood changes, and weight gain may be manifestations of thyroid dysfunction. In a subgroup of females, depressive symptoms are associated with positive thyroid colloidal antibody status during the postpartum period. If mania is considered as a diagnosis, medical conditions such as hyperthyroidism, Addison’s disease, Cushing’s disease, and vitamin B12 deficiency must be ruled out. Mania can also be induced by substance abuse and by antidepressive therapy.

V. PATHOPHYSIOLOGY • What are the neurologic underpinnings of mood disorders? • What hormones and neurotransmitter systems are implicated in perinatal affective disturbance? Mood disorders associated with childbearing have different times of onset in different patients and are heterogeneous in their presentation. Consequently, the exact mechanism of mood changes due to pregnancy may involve a number of hormonal systems that precipitate the complex physiologic changes of gestation. The study of these alterations in the context of pregnancy and the postpartum period offers a heuristic model for the study of biologic factors in mood states. A controversy exists regarding whether the hormones of pregnancy directly cause mood dysfunction or whether the hormonal changes in pregnant patients with mood disorders are within normal limits but nevertheless trigger pathologic mechanisms at the neurotransmitter or receptor levels that result in the release of estrogen, progesterone, prolactin, cortisol, thyroxin, and thyrotropin hormones.

The placenta produces high levels of estrogens (estradiol, estriol, and estrone) and progesterone during pregnancy. Although a reciprocal relationship between the serotoninergic system and gonadal hormones is speculative, preliminary studies have indicated that serotonin receptor changes in the limbic area are negatively correlated with progesterone levels. With removal of the placenta at delivery, estrogen and progesterone levels— which are at their highest levels at term—drop sharply, reaching pregravid levels by the fifth postpartum day. Recognition of this massive drop prompted the hypothesis that in some vulnerable or genetically predisposed women, postpartum withdrawal of gonadal hormones may cause changes along the serotoninergic cascade that may lead to a mood disorder.8 High estrogen levels during pregnancy stimulate production of thyroid hormone–binding globulin, which mimics hyperthyroidism, with thyroid enlargement and increased total triiodothyronine (T3) and thyroxine (T4) levels that reflect the increase of binding globulins. However, free T3 and T4 remain within normal limits. Serum cortisol levels gradually increase over gestation to 2 to 3 times the normal level, with rapid decline to baseline level within the first 3 days postpartum. Levels of β-endorphin, human chorionic gonadotropin, and cortisol also rise during pregnancy, reaching a maximum level near term and declining at delivery. Prolactin levels rise during pregnancy, peak at delivery, and, in nonlactating females, return to pregravid levels within 3 weeks postpartum. By inducing the release of oxytocin, a hormone that stimulates pituitary lactotrophic cells, breastfeeding maintains high prolactin levels. Even in breastfeeding females, however, prolactin levels eventually return to pregravid levels.9

VI. TREATMENT • Should psychotropic medications be discontinued before conception or as soon as pregnancy is diagnosed? • What are the consequences of untreated mood disorders during childbirth? • Why is postpartum prophylaxis recommended for bipolar patients? PSYCHOTHERAPY Pregnancy may influence a woman’s perceptions, interpretations, and expectations. It has been suggested that how a woman adapts to psychosocial changes during pregnancy may be a more important determinant of

Psychiatry Volume 4, Part 2 5

Peripartum Mood Disorders later outcomes than the nature of the changes themselves. The willingness to relate to a psychotherapist and the built-in “termination date” make pregnant patients exceptionally receptive to psychotherapeutic intervention, for which the focus will often be reexamining prior and current relationships, sorting out identifications, and integrating somatic changes.10 Typically, this readiness to interact allows a rapid and stable therapeutic alliance to be built between the patient and therapist. During psychotherapy, the somatic concerns associated with pregnancy may be approached from 2 primary perspectives. The first entails an exploration of various psychologic conflicts and external stressors that may be related to somatization. The second involves encouraging the patient to cope with any medical complications that may impact pregnancy (eg, diabetes, high blood pressure, bleeding) to prevent negative developmental consequences for the pregnancy and the anticipated mother-infant relationship.11 Although psychotherapeutic themes tend to undergo some transformations during various stages of pregnancy, they remain organized along a limited number of dimensions.11 Themes may include conflicts regarding dependency needs, narcissistic disturbances, reconciliation, and working through losses while giving life. Career conflicts are also frequent concerns for some women. During pregnancy, not only the baby but also the female as mother is in gestation; thus she uses the therapeutic setting as a holding environment where she can be nurtured. Therefore, an important transference consideration is to notice how being mothered by the therapist is received—as a pleasure, a due, a source of shame, or a seduction.11 PHARMACOTHERAPY Risks and Benefits Psychotropic drugs have not been tested or approved by the Food and Drug Administration for use during pregnancy. Therefore, the safest pregnancy and postpartum course would involve no fetal or infant exposure to medications. However, mood dysfunction during childbirth is a complex reality. The mother may be at risk for impulsivity and lack of attention to prenatal care because of untreated psychiatric illness. Relapse may also predict a more chronic clinical course, because with repeated episodes, the length of time between subsequent episodes diminishes. Of equal concern, certain psychotropic agents used in pregnancy may be associated with adverse outcomes for the infant. First, congenital anomalies may occur with exposure during the first trimester. Second, perinatal complications may include poor obstetric outcomes (eg, prematurity or low birth

6 Hospital Physician Board Review Manual

weight) or syndromes related to drug use late in the third trimester (eg, neonatal withdrawal or toxicity in the first few days after birth). Third, neurobehavioral sequelae may be characterized by developmental delays, learning difficulties, and neurologic deficits that result from drug exposure at any time during pregnancy.12 Pharmacokinetics During Pregnancy It is important for the clinician to appreciate the physiologic changes that occur during pregnancy and to understand how the pharmacokinetics of psychotropic medications are affected by pregnancy (Table 2). Glomerular filtration rate and creatinine clearance both increase during pregnancy; these changes cause many medications to be excreted more rapidly. If the medication dose is not increased, serum levels may fall and the mother may be at increased risk for symptom recurrence. Special attention to serum drug levels is needed around the period of childbirth, when dramatic changes in drug clearance can produce a sudden increase to toxic blood levels, which could be injurious to mother and fetus. Clinician familiarity with each medication will ensure an optimal outcome for both mother and infant. Antidepressants Selective serotonin reuptake inhibitors. No antidepressant has been approved as a category A (ie, lowest risk category) agent for use during pregnancy or lactation (Table 3). Once the clinician and patient have decided to use medication to treat depression during pregnancy, fluoxetine (category C) should be the agent of first choice because it is the only selective serotonin reuptake inhibitor (SSRI) for which human data exist regarding safety in pregnancy. Documentation is lacking for sertraline, paroxetine, and fluvoxamine.13 Although no prospective, double-blind, crossover trials have addressed the teratogenicity or short- and long-term developmental effects of SSRIs, retrospective studies indicate no evidence of an increased rate of spontaneous abortion in the first trimester or of an increased rate of major malformations or neurodevelopmental delays in children whose mothers took SSRIs in the first trimester. One study found increased rates of premature delivery, admission to special care nurseries, and poor neonatal feeding when fluoxetine was administered in the third trimester. However, major methodological problems plagued this study.14,15 Some evidence exists of increased rates of premature delivery, admission to special-care nurseries, and poor neonatal adaptation (ie, cyanosis on feeding, jitteriness) when fluoxetine was taken during the third trimester.14

Peripartum Mood Disorders Table 2. Perinatal Pharmacokinetic Changes

Table 3. Classification for Medication Safety in Pregnancy

Gastric acid secretion alters normal drug absorption Gastrointestinal motility renders drug absorption unpredictable

Category A

Controlled studies show no risk: Adequate, well-controlled studies in pregnant women have failed to demonstrate risk to the fetus in any trimester.

B

No evidence of risk in humans: Human studies show no increased risk but animal studies do; or, if no adequate human studies have been done, animal studies show no risk.

C

Risk cannot be ruled out: Human studies are lacking, and animal studies are either positive for fetal risk or lacking as well; however, potential benefits may justify potential risk.

D

Positive evidence of risk: Human studies, or investigational or postmarketing data, show risk to the fetus; nevertheless, potential benefits may outweigh risks.

X

Contraindicated in pregnancy: Studies in animals or humans, or investigational or postmarketing reports, have shown fetal risks that clearly outweigh any possible benefit to the patient.

Serum volume and total body water increase Hepatic microsomal enzyme activity increases Serum protein binding capacity increases Glomerular filtration rate increases Creatinine clearance doubles Urinary excretion of many medications increases

These trends may be related to observations that antidepressant levels fluctuate across pregnancy as a result of increases in serum volume, hepatic microsomal enzyme activity, and renal clearance rates.16 Physiologic changes that occur during pregnancy (eg, increased total body water) can cause unpredictable shifts in the serum level of antidepressants. Doses may need to be increased 1.5 times over baseline as the pregnancy progresses. It is recommended that dosages of SSRIs be halved during the week before delivery to minimize any potential withdrawal in the infant. The infant should then be observed for withdrawal symptoms. Antidepressant treatment continued for 12 months postpartum following an infant’s first withdrawal episode usually results in complete recovery. Tricyclic antidepressants. These agents (eg, imipramine, amitriptyline) are believed to exert antidepressant activity by reducing CNS uptake of norepinephrine and serotonin and blocking the response of dopamine receptors. The majority of tricyclic antidepressants (TCAs) are either category C or D agents for use during pregnancy (Table 3). If a TCA is used, nortriptyline is the preferred choice because it is the most extensively studied agent of this class in pregnancy, it exerts less anticholinergic activity, and it maintains a stable relationship between serum concentration and therapeutic effect. TCAs elevate heart rate and cause orthostatic hypotension, which can be compounded by the physiologic effects of pregnancy. Tricyclic agents were once thought to increase the rate of congenital anomalies, particularly limb reduction defects. However, there is no statistical evidence to support this assumption. When treating depressed pregnant adolescents, TCAs have not been as effective as in adults, which may be due to the influence of pubertal hormones or an immature noradrenergic system. Cardiotoxic effects of TCAs in adolescents necessitate monitoring by baseline electrocardiogram after steady-state serum levels are attained.

Interpretation

Adapted from Physicians’ Desk Reference, 54th ed. Montvale, NJ: Medical Economics Company, 2000:345.

At birth, TCAs can cause neonatal withdrawal symptoms such as tachypnea, tachycardia, cyanosis, irritability, and diaphoresis. Fetal withdrawal symptoms can be avoided by tapering the dose 3 weeks before delivery. Treatment with TCAs may also be complicated in new mothers who are concurrently prescribed oral contraceptive agents. Recent literature suggests that oral contraceptives inhibit the hepatic metabolism of TCAs; therefore, dose reductions of TCAs may be necessary in this situation.3 Monoamine oxidase inhibitors. These agents (eg, phenelzine and tranylcypromine) are category C agents (Table 3); that is, they are known teratogens in animals, but data have been too small to conclude whether they are human teratogens. Use of monoamine oxidase inhibitors (MAOIs) results in increased concentrations of epinephrine, norepinephrine, and serotonin in the CNS. A limiting drawback preventing use of MAOIs is that ingestion of foods or drugs containing tyramine can cause an MAOI-induced hypertensive crisis that may terminate the pregnancy. Side effects such as sedation, palpitations, dizziness, insomnia, and constipation also occur more frequently in pregnant females who are using this class of antidepressant. MAOIs can also interact adversely with

Psychiatry Volume 4, Part 2 7

Peripartum Mood Disorders tocolytic agents that may be necessary to forestall premature labor. Therefore, most clinicians avoid using MAOIs in pregnant patients unless these agents are absolutely necessary. Other antidepressants. Scant information is available regarding reproductive safety when other antidepressant agents are used during pregnancy. Little is known about such newer agents as moclobemide, venlafaxine, and nefazodone. The increased risk of seizures associated with maprotiline and bupropion deters their use during pregnancy, especially in females who are subject to eclampsia. Psychostimulants (amphetamines, methylphenidate, and pemoline) are sometimes prescribed to treat depressive disorders. However, amphetamines cause neuronal release of dopamine and norepinephrine as well as the blockade of catecholamine reuptake. Clinically significant maternal hypertension may result as well as cardiac, biliary, and palatal defects in the fetus. Mood Stabilizers Clinical management of pregnant patients who have bipolar disorder poses 3 major dilemmas: (1) teratogenic risks of mood stabilizers, (2) risks of untreated psychiatric illness, and (3) risks of discontinuing maintenance psychotropic medications. Whenever possible, plan a drug-free first trimester by slowly tapering both lithium and anticonvulsants (to avoid acute relapse and withdrawal seizures, respectively) before conception. Unfortunately, there is no ideal pregnancy for the bipolar patient; the decision about whether to continue or discontinue medication depends on the patient’s insight, social supports, and therapeutic alliance. Better candidates for attempting discontinuation of medical therapy are patients who can recognize their own early symptoms or have family members who recognize symptoms and can bring them to a treatment facility. Patients who have had a single episode of mania and a long period of affective well-being may be able to taper their medications and gradually discontinue therapy before attempting to conceive. However, a safer option may be to continue lithium until the first missed menstrual period and then gradually discontinue the drug after a positive pregnancy test result.17 Women who have severe bipolar disorder require medication before and during pregnancy. All bipolar patients should be closely monitored for a relapse in symptoms, which demands aggressive therapy regardless of the gestational period. Lithium. The use of lithium carbonate or sustainedrelease lithium compounds in female patients of reproductive potential remains controversial and is empirically discouraged. It was previously thought that firsttrimester exposure caused Ebstein’s anomaly, a rare con-

8 Hospital Physician Board Review Manual

dition characterized by right ventricular hypoplasia and congenital downward displacement of the tricuspid valve into the right ventricle. However, recent data suggest a more modest estimate of risk associated with fetal exposure to lithium. Moreover, the alternative pharmacologic treatments to lithium have significant teratogenic potential. The current standard of practice is that if firsttrimester exposure occurs, a perinatal consultation combined with a level II ultrasound is recommended at 16 to 18 weeks’ gestation to assess fetal heart and vertebral body development. When inadvertent exposure to anticonvulsants has occurred, folate (which may reduce the risk of neural tube defects) should be given until the end of the first trimester. With the exception of electroconvulsive therapy (ECT), lithium may be the safest first-line alternative for the treatment of bipolar disorder during pregnancy; lithium may also have distinct advantages in the postpartum period. A unique feature of lithium is that it is not metabolized at all but eliminated unchanged, mostly by renal clearance. The greatest and most predictable factor affecting lithium clearance is reduced renal functioning, although monitoring of serum levels provides a way of dealing with this problem. Some authors have suggested that serum lithium levels should be monitored weekly during pregnancy; however, a more reasonable approach is monthly monitoring of serum medication levels, serum electrolytes, and thyroid function. Alterations in thyroid panel results (eg, increases in TSH) during pregnancy make interpretation difficult; any test result abnormalities should prompt the physician to seek an endocrinology consultation.18 The main risks of lithium use later in pregnancy include polyhydramnios, premature labor, neonatal toxicity, and neonatal hypothyroidism. Polyhydramnios results from lithium-induced nephrogenic diabetes insipidus in the fetus; this condition can compromise maternal respiration because the uterus becomes so enlarged that the lungs do not have enough room to expand. Neonatal lithium toxicity can occur even when the mother’s serum drug level is therapeutic. Toxicity may result in abnormal respiratory patterns, cyanosis, hypotonia, decreased sucking and Moro reflexes, poor myocardial contractility, and hypoglycemia. Lithium requirements increase during the third trimester. However, lithium doses are usually reduced by 25% to 30% just before delivery to avoid possible neonatal toxicity; neonates are monitored for lithium toxicity for up to 10 days after birth. This practice is being reviewed in light of data from nonpregnant patients indicating that even partial reductions in lithium serum concentrations may precipitate relapse.

Peripartum Mood Disorders Administration of relatively low doses of lithium in the early postpartum period may be particularly dangerous to the mother because bipolar disorder appears to be especially prevalent at this time. Clinically, a more prudent option may be to follow serial lithium levels during labor and delivery, as well as during the first few days postpartum, and adjust the dose accordingly.19 Monitoring of maternal serum levels after delivery, continued hydration, and avoidance of nonsteroidal antiinflammatory drugs for pain management will minimize the risk of maternal toxicity. Other primary mood-stabilizing agents. Alternative treatment options for patients who have bipolar disorder include valproate, carbamazepine, and newer anticonvulsant agents such as gabapentin and lamotrigine. Teratogenic risks from carbamazepine are deemed greater than any benefit from treatment. The reproductive safety data on the newer mood stabilizers are limited and based mainly on case reports. An advantage of using valproic acid for treatment of acute mania is that loading doses can be administered to reach therapeutic serum concentrations within 24 hours, whereas with lithium several days are needed to achieve the same effect. Additionally, patients who have rapid cycling disorder and mixed or dysphoric mania are more likely to respond favorably to valproate therapy. Valproate use during pregnancy carries a 1% to 2% risk of neural tube defects. However, both valproate and carbamazepine can affect neural tube development, as well as increase the risk of neonatal hemorrhage due to depletion of vitamin K–dependent clotting factors. A distinctive pattern of minor anomalies of the face and digits, termed valproic acid syndrome, has been described. Common characteristics are midface hypoplasia, epicanthal folds, short nose, broad nasal bridge, thin upper lip, and micrognathia. Also, a rare but potentially fatal neonatal complication of valproate use is acute liver failure. As with the use of all medications during pregnancy, the physician and patient must seriously consider the risk-benefit ratio. Prophylactic mood stabilization. The high rate of bipolar disorder recurrence during the postpartum period underscores the need to consider prophylactic mood stabilization for those patients who have completed a pregnancy (regardless of whether medication was used during the pregnancy). This practice is often neglected, thus placing the new mother at significant risk for psychiatric morbidity. The expectant mother should be clearly informed of the high risk for relapse as she enters the postpartum period. Lithium prophylaxis has been recommended for patients at risk for puerperal affective instability. Post-

partum relapse can be reduced fivefold when lithium is given shortly before birth (week 36) or within 48 hours of delivery and continued into the postpartum period. Carbamazepine and valproic acid have not been proven effective for postpartum prophylaxis. Antipsychotic Agents In an attempt to avoid the use of mood stabilizers during pregnancy, neuroleptic agents are often administered. Unfortunately, these agents are rarely effective treatments for mania and are not necessarily safer during pregnancy. Although prospective studies have failed to demonstrate significant teratogenic risks of highpotency neuroleptics such as haloperidol, adequate data are lacking concerning the safety profiles of these agents. Even less is known about the reproductive safety of atypical antipsychotics such as clozapine, risperidone, and olanzapine. In general, avoidance of these agents in the first trimester is highly recommended. When an antipsychotic is the best choice, first-trimester exposure should be kept at the lowest dose for the briefest time possible. High-potency neuroleptics (eg, haloperidol) may be prescribed during this interval. Neonatal case reports have demonstrated motor restlessness, tremor, difficulty with oral feedings, hypertonicity, dystonic movements, and parkinsonian-like effects. Although these symptoms are typically of short duration, authorities recommend tapering neuroleptics 5 to 10 days before delivery to minimize the chances of neonatal extrapyramidal symptoms. Abrupt neuroleptic discontinuation just before labor and delivery heightens the risk of maternal psychiatric decompensation within weeks of delivery. Electroconvulsive Therapy ECT is a modality that is not contraindicated in pregnancy and is generally successful, often providing rapid improvement for patients who have vegetative symptoms, suicidal ideation, or inability for self-care. Because uterine muscle does not automatically contract as part of a generalized tonic-clonic seizure, it is unlikely that ECT hastens the onset of labor. Modifications of ECT can be implemented that decrease the potential risks of this modality (Table 4).20 Informed consent for ECT should include the patient’s capacity to understand and rationally evaluate the risks and benefits to herself and the fetus. PHARMACOTHERAPY DURING LACTATION • What are the pharmacokinetics of medication excretion into breast milk? • What approach should be used to determine appropriate treatment for patients who wish to breast-feed?

Psychiatry Volume 4, Part 2 9

Peripartum Mood Disorders Table 4. Modifications of Electroconvulsive Therapy (ECT) During Pregnancy Perform a pelvic examination to look for vaginal bleeding or cervical dilatation Discontinue nonessential anticholinergic medication the day before ECT to avoid excessive delay in gastric emptying Assess for contractions using uterine tocodynamometry within the hour before ECT Ensure adequate hydration with lactated Ringer’s solution or normal saline solution 12 hours before ECT Give 30 mL of 0.3 mol/L sodium citrate 15 to 20 minutes before anesthesia to increase gastric pH Place a pillow or rolled up sheet under patient’s right hip to displace uterus to the left Place external fetal cardiac monitor just before ECT Intubate patient just before ECT if pregnancy is beyond the first trimester Avoid inducing excessive hyperventilation during ECT Recheck for uterine contractions after ECT Recheck for vaginal bleeding after ECT

The increasing number of pregnant patients who plan to breast-feed and the high rate of psychiatric illness during the postpartum period underscore the complex considerations involved in a risk-benefit assessment of medications as well as the impact of untreated maternal mental illness (Table 5). Most psychotropic drugs are known to be secreted into milk with great variability in concentration. The amount of medication available to enter the breast milk depends on several maternal factors, including (1) maternal rate of drug metabolism, (2) maternal volume of distribution, (3) medication dosing schedule, (4) breast-feeding schedule, and (5) the bioavailability of the medication in the maternal circulation.21 Based on these data, the American Academy of Pediatrics has provided guidelines for breast-feeding in mothers taking antidepressants and mood stabilizers.22 When any psychotropic medication is used during lactation, the infant should, at a minimum, be routinely monitored as often as an adult would be, because the infant is also exposed to the medication. Antidepressants During Lactation Because all SSRIs are excreted in breast milk, several options are available to minimize infant drug exposure. One strategy is to adjust the feeding schedule based on the time that the dose is administered. This may be difficult with some medications; for example, sertraline’s serum and breast milk concentrations in a mother peak

10 Hospital Physician Board Review Manual

1 to 9 hours after the dose, so the mother would need to refrain from nursing for this long interval. A more reasonable option is to prescribe an SSRI with a shorter halflife and no active metabolites (eg, paroxetine).23 Because SSRI concentration is greatest in the milk first excreted in a feeding session, another technique is for the mother to express the initial breast milk before feeding the infant. Breast milk concentrations of TCAs are equal to those measured in maternal serum, but breast-fed infants have nearly undetectable TCA serum levels. The best advice is that the mother should minimize her infant’s drug exposure by breast-feeding immediately before the next dose and refraining from nursing at least 4 hours from the last dose.22 This technique does not allow time for the medication to be absorbed and distributed in the mother so that less drug is available to be transferred to the infant. MAOIs are not often used as first-line treatment in lactating mothers because of dietary constraints and the potential for hypertensive crisis. It is not known whether the novel antidepressants nefazodone, mirtazapine, and venlafaxine or their metabolites are excreted in human milk; therefore, caution should be exercised when these agents are administered to nursing mothers.24 Mood Stabilizers During Lactation The use of lithium during lactation has been categorically discouraged in previous reports and is contraindicated by the American Academy of Pediatrics.22 There have been reports of lithium toxicity in breastfed infants, including cyanosis, electrocardiogram abnormalities, and hypotonia. This toxicity is believed to be due to the relatively immature kidney function in neonates and is especially likely to occur during times of dehydration. Although anticonvulsant mood stabilizers do pass into maternal milk, the documented levels are not known to be harmful. For example, valproic acid’s concentration in breast milk is approximately 1% to 10% of its concentration in maternal serum. In fact, valproic acid and carbamazepine are considered compatible with breast-feeding by the American Academy of Pediatrics.22 However, presently available data indicate that the use of valproic acid during lactation is preferable to using either carbamazepine or lithium. Nevertheless, some authors have recommended against nursing while taking valproic acid because of the potential threat of fatal hepatotoxicity in children under the age of 2 years who are exposed to valproate. These sources also discourage concurrent breast-feeding and mood stabilizer use because of the need for invasive infant monitoring (ie, serum level monitoring).

Peripartum Mood Disorders Table 5. Factors to Consider in a Risk-Benefit Assessment of Using Psychotropic Medications During Lactation • Breast-feeding, which is supported by most professional organizations as the ideal form of newborn nutrition, provides immunologic benefits to the infant. • The postnatal period is a high-risk time for onset/relapse of maternal psychiatric illness. • Untreated maternal mental illness adversely affects mother-infant attachment and infant development. • Because all psychotropic medications are excreted in breast milk, the infant is exposed. • The pharmacokinetics of psychotropic medication excretion into breast milk are complicated. • The clinical significance of detectable infant serum concentrations of psychotropic medications remains unknown. • Little is known about the long-term neurobehavioral effects of breast-feeding exposure to psychotropic medication. • All environmental exposures to medication, alcohol, and drugs during pregnancy should be documented, and factors that may affect infant development should be minimized. • The medication of prior response should be used unless it is contraindicated in lactation (eg, monoamine oxidase inhibitors, lithium). • A medication with known efficacy and side effects should be chosen to ensure identification of adverse effects and provide reassurance for the nursing mother.

Another concern that has been raised is that no data exist on the safety of multiple drug exposures to mother and infant, such as exposure to lithium during pregnancy and valproic acid during lactation. Changing the mother to a novel medication during the high-risk period may enhance the possibility of recurrent symptoms, thus increasing risk by exposing the infant to both medication and maternal mental illness.19

VII. SUMMARY POINTS • All patients presenting with perinatal mood symptoms should be screened for physical illness, psychiatric history, and family history of Axis I disorders. • More than 50% of pregnant patients who have bipolar disorder may suffer relapse during the postpartum period. The risk of postpartum relapse in patients with depression and psychosis is also high. • With patients for whom medication is indicated, the physician should choose antidepressants that have fewer anticholinergic effects, no active metabolites, and shorter half-lives to minimize drug exposure and side effects to the fetus. • Electroconvulsive therapy is a relatively safe and effective treatment during pregnancy provided that steps are taken to decrease potential risks.

1.

2. A pregnant woman who has bipolar disorder experiences increased impulsivity and delusions related to body changes and fetal movements at 28 weeks’ gestation. All of the following treatment options would be appropriate in this setting EXCEPT: A) Lithium B) Valproic acid C) Haloperidol D) Continued observation without medications until depressive symptoms develop 3. Steps to decrease potential maternal and fetal risks during electroconvulsive therapy (ECT) include: A) Elevate the patient’s right hip B) Hyperventilate the patient C) Administer an anticholinergic agent in conjunction with ECT D) Avoid pelvic examinations, which could induce contractions and bleeding 4.

BOARD REVIEW QUESTIONS Choose the single best answer to each of the following questions.

All of the following abnormalities are probably biologic common denominators for perinatal mood disorders EXCEPT: A) Thyroid axis abnormalities B) Serum elevation and abrupt decrease in estrogen and progesterone C) Serotonin dysfunction D) Blunted growth hormone response

When educating a pregnant woman who has bipolar disorder, she should be told that: A) Recent research has proved that lithium carbonate is safe during pregnancy B) Studies have demonstrated the safety of

Psychiatry Volume 4, Part 2 11

Peripartum Mood Disorders mood stabilizers such as valproate and carbamazepine C) Relapse may predict a more chronic clinical course D) Postpartum relapse can be reduced fivefold when valproic acid is given shortly before birth 5.

A few days after giving birth to her first child, a 23-year-old woman experiences mood lability, tearfulness, and disturbances in appetite and sleep. All of the following statements about her condition are correct EXCEPT: A) Any mood lability after birth reflects psychopathology in the mother B) Symptoms may reflect the benign course of postpartum blues C) Women with postpartum blues who also have histories of recurrent mood disorder require close monitoring D) If these symptoms persist beyond 2 weeks postpartum, further evaluation is indicated to rule out a more significant affective disorder

cide during pregnancy. Am J Psychiatry 1997;154:122–123. 8. Abou-Saleh MT, Ghubash R, Karim L, et al: Hormonal aspects of postpartum depression. Psychoneuroendocrinology 1998;23:465–475. 9. Hendrick V, Altshuler LL, Suri R: Hormonal changes in the postpartum and implications for postpartum depression. Psychosomatics 1998;39:93–101. 10. Ballou J: The significance of reconciliative themes in the psychology of pregnancy. Bull Menninger Clin 1978;42: 383–413. 11. St-Andre M: Psychotherapy during pregnancy: opportunities and challenges. Am J Psychother 1993;47:572–590. 12. Austin MP, Mitchell PB: Psychotropic medications in pregnant women: treatment dilemmas. Med J Aust 1998; 169:428–431. 13. Gupta S, Masand PS, Rangwani S: Selective serotonin reuptake inhibitors in pregnancy and lactation. Obstet Gynecol Surv 1998;53:733–736. 14. Chambers CD, Johnson KA, Dick LM, et al: Birth outcomes in pregnant women taking fluoxetine. N Engl J Med 1996;335:1010–1015. 15. Wisner KL, Gelenberg AJ, Leonard H, et al: Pharmacologic treatment of depression during pregnancy. JAMA 1999;282:1264–1269.

ANSWERS

16. Altshuler LL, Hendrick VC: Pregnancy and psychotropic medication: changes in blood levels. J Clin Psychopharmacol 1996;16:78–80.

1. D

2. D

3. A

17. Cohen LS, Friedman JM, Jefferson JW, et al: A reevaluation of risk of in utero exposure to lithium. JAMA 1994; 27:146–150.

4. C

5. A

REFERENCES 1. Rofe Y, Blittner M, Lewin I: Emotional experiences during the three trimesters of pregnancy. J Clin Psychol 1993; 49:3–12. 2. Mayberry LJ, Affonso DD: Infant temperament and postpartum depression: a review. Health Care Women Int 1993; 14:206–211. 3. Susman JL: Postpartum depressive disorders. J Fam Pract 1996;43(suppl 6):S17–S24.

18. Llewellyn A, Stowe ZN, Strader JR Jr: The use of lithium and management of women with bipolar disorder during pregnancy and lactation. J Clin Psychiatry 1998; 59(suppl 6):57–65. 19. Viguera AC, Cohen LS: The course and management of bipolar disorder during pregnancy. Psychopharmacol Bull 1998;34:339–346. 20. Miller LJ: Use of electroconvulsive therapy during pregnancy. Hosp Community Psychiatry 1994;45:444–450.

4. Szigethy EM, Ruiz P: Treatment of depression during adolescent pregnancy. Journal of Practical Psychiatry and Behavioral Health 1999;5:256–264.

21. Llewellyn A, Stowe ZN: Psychotropic medications in lactation. J Clin Psychiatry 1998;59(suppl 2):41–52.

5. Koniak-Griffin D, Walker DS, de Traversay J: Predictors of depression symptoms in pregnant adolescents. J Perinatol 1996;16:69–76.

22. American Academy of Pediatrics Committee on Drugs: The transfer of drugs and other chemicals into human milk. Pediatrics 1994;93:137–150.

6. Parry BL: Reproductive-related depressions in women: phenomena of hormonal kindling? In Postpartum Psychiatric Illness: A Picture Puzzle. Hamilton JA, Harberger PN, eds. Philadelphia: University of Pennsylvania Press, 1992.

23. Altshuler LL, Burt VK, McMullen M, Hendrick V: Breastfeeding and sertraline: a 24-hour analysis. J Clin Psychiatry 1995;56:243–245.

7. Marzuk PM, Tardiff K, Leon AC, et al: Lower risk of sui-

24. Iqbal MM: Effects of antidepressants during pregnancy and lactation. Ann Clin Psychiatry 1999;11:237–256.

Copyright 2000 by Turner White Communications Inc., Wayne, PA. All rights reserved.

12 Hospital Physician Board Review Manual