|
Bipolar, or manic-depressive, disorder (BD) is a serious, recurrent
psychiatric illness with a lifetime prevalence well above 1% (1,2).
The disorder typically begins in adolescence or early adulthood
and tends to be a lifelong condition characterized by high relapse
rates, comborbid anxiety and substance use disorders, persistent
subsyndromal morbidity and dysfunction, and premature mortality,
mainly due to extraordinarily high suicide rates (3–8). Owing
to its relatively high prevalence, BD is perhaps the most common
idopathic psychotic disorder, but only recently has it been recognized
as a major public health problem (2,3,9).
BD represents a significant source of distress, disability, and
family burden. Its prevention and treatment are particularly important
and complicated for women of reproductive age. Yet, despite its
undoubtedly great clinical importance, remarkably little is known
about the impact of the female reproductive life cycle—the
menstrual cycle, pregnancy, postpartum, nursing, and menopause—on
the course and treatment of BD (10). Typically, women diagnosed
with the disorder encounter significant obstacles from the professional
community with respect to pregnancy. They are often counselled to
avoid or terminate pregnancy to prevent fetal exposure to potentially
teratogenic medications and to avoid the risk for recurrence of
manic or depressive illness (11). Women with BD seeking prepregnancy
consultation at a major medical centre were recently surveyed (12).
These women wanted advice regarding management of their mood disorder
during pregnancy. Of those surveyed, 45% reported that a psychiatrist
or other mental health professional had advised them to avoid pregnancy
altogether. Following their prepregnancy consultation, 37% decided
to avoid pregnancy. The most frequently stated reasons for this
decision were fear of adverse effects on fetal development from
the medicines and fear that, if treatment were discontinued, the
illness would recur.
Physicians caring for pregnant women with BD face a complex clinical
challenge: they must minimize risk to the fetus while limiting the
impact of maternal morbidity on the mother, her unborn offspring,
and her family that might result from potentially severe untreated
psychiatric illness. Patients and their clinicians also face the
difficult reality that decisions either to use or not to use psychotropic
medications can be associated with complications. Deciding what
constitutes reasonable risk during pregnancy requires shared responsibility
but ultimately rests with the informed patient. Such informed choices,
coupled with close psychiatric follow-up and coordinated care with
the obstetrician, are components of an emerging model of care aimed
at optimizing the clinical management of women with BD during pregnancy.
This overview considers information about the course of BD in pregnancy
and about the reproductive safety of established and proposed mood
stabilizers. A major aim of this report is to stimulate further
discussion and research on this important, complex, and insufficiently
considered public health problem for women of childbearing age with
BD. We also present tentative guidelines for improved clinical care
of such women.
There is wide agreement that the early postpartum period presents
unusually high risk for recurrence of BD and other psychiatric illnesses
(13,14). In the mid-19th century, Marcé provided compelling
early reports of severe psychotic and affective disorders arising
in women during the postpartum period (15). Later, in his classic
descriptions of manic-depressive syndrome, Kraepelin observed that
attacks of mania and melancholia were common in pregnancy, but even
more so following childbirth (16). Modern studies have sustained
this general impression that the postpartum period brings high risks
of acute psychiatric illness in women with manic-depressive illness.
However, the risks associated with pregnancy itself remain less
well characterized, and evidence is conflicting as to whether pregnancy
alters the risk that major affective illness will recur. Some clinical
observations suggest that pregnancy may reduce the risks of acute
psychiatric illness and specifically protect against recurrences
of BD, major depression, psychotic disorders, and suicide (17–20).
Other studies have found rates of psychiatric hospitalization to
be either somewhat lower or unchanged during pregnancy, but these
studies did not evaluate morbidity in nonhospitalized pregnant women
with BD (13,14,21–23).
Grof and colleagues recently suggested that pregnancy has an apparent
protective effect on the course of lithium-responsive BD I (with
mania) (24). They describe a benign course and even improvement
during pregnancy, based on comparisons before and after pregnancy
in women whose illness could be managed without mood-stabilizing
medication for prolonged periods. They propose that these findings
support the view that pregnancy may protect against and prevent
recurrences of BD, but the study sample may not represent broader
groups of women with BD (25). Moreover, other recent research and
growing clinical experience suggest that pregnancy probably does
not consistently protect against recurrences of mania or major depression
in women with BD. Instead, pregnancy has been found to be a time
of substantial risk for relapse, particularly following discontinuation
of ongoing mood-stabilizing treatment (25–28). In a large,
well-characterized clinical sample, Blehar found that about 45%
of women with BD experienced an exacerbation of their illness during
pregnancy (29). More recently, Freeman and colleagues also found
that at least 50% of a sample of women with BD became symptomatic
during pregnancy (30).
We recently studied the course of BD I and BD II (recurrent major
depression with hypomania) in a sample of 101 age-matched, pregnant
(n = 42) and nonpregnant (n = 59) women who discontinued lithium
maintenance treatment. Survival analysis demonstrated that BD recurred
in 52% of the pregnant women and 58% of the nonpregnant women, with
indistinguishable relapse time-courses (25,26). In contrast, within
the preceding year only 21% of the entire sample had experienced
a recurrence during ongoing treatment with lithium. Risks were similar
with both BD I and II subtypes, but significantly higher for women
with a history of 4 or more prior episodes and for women who abruptly
or rapidly (< 2 weeks) discontinued lithium treatment proximate
to conception.
|
|
These findings are consistent with the view that either pregnancy
may have little effect on recurrence risk in BD or discontinuation
of maintenance treatment itself represents a major, and perhaps
dominant, stressor (25–36).
These studies seem to suggest that any protective effects of pregnancy
on risk for recurrences of mania or depression in women with BD
are limited. Moreover, these effects are probably insufficient to
protect most patients from recurrence if ongoing maintenance mood-stabilizing
treatment is discontinued. To some extent, the risks for recurrence
may be predicted by the history of illness or severity, as well
as by a history of prolonged wellness or proven ability to tolerate
long periods without mood-stabilizing treatment. Clearly, more studies
that control specifically for past illness, DSM-IV diagnostic subtypes
(I, II, and rapid-cycling), and treatment status are required to
clarify the course of BD during pregnancy.
Whereas data on the course of BD during pregnancy are sparse, the
postpartum period has received more systematic study. This period
of the female reproductive cycle has been recognized consistently
for more than a century as a time of heightened vulnerability to
relapse of mood disorders and acute psychosis, although quantitative
specification of that risk has been inconsistent. Recurrence rates
within the first 3 to 6 postpartum months among women with BD have
ranged from 20% to 80% (15,16,26,28,30,37–42). These rates
have been well above 60% (range, 67% to 82%) in more recent studies,
perhaps reflecting more reliable diagnosis and greater interest
in the problem (26,29,30,42).
BD is also closely associated with postpartum psychosis (13,15,37–48).
Several studies have demonstrated that many women presenting with
postpartum psychosis later develop BD (13,15, 37–48). Postpartum
psychosis is a rare condition in the general population, with an
estimated prevalence in postpartum women of 0.1% to 0.2%. However,
for women with BD, the risk is perhaps 100 times higher, at 10%
to 20% (13,39,42). Postpartum psychosis is characterized by rapid
onset of symptoms—often within the first 48 to 72 hours after
delivery. The disorder may present with delirium, but it is often
indistinguishable from a manic or mixed manic-depressive episode
with psychotic features. Postpartum psychosis is a psychiatric emergency
associated with high risk of infanticide and potential suicide;
it requires immediate treatment with a mood stabilizer and antipsychotic
agent or electroconvulsive treatment (ECT), usually in an inpatient
setting (39,40,43–48). Following an episode of such illness,
risk for a recurrent episode of postpartum psychosis with a subsequent
pregnancy is estimated to be as high as 90% (43–45).
Several investigators have evaluated the ability of treatment to
attenuate the high postpartum risk for recurrence of BD or acute
psychosis (43–49). Most of this research has been limited to
use of lithium prophylaxis. When lithium was given either several
weeks prior to delivery or immediately postpartum, the risk for
postpartum recurrences of BD was reduced on average by two- to fivefold,
compared with untreated women (17,43–49). These reports leave
important questions unanswered, including those of optimal dose
and treatment timing and the comparative efficacy of lithium vs
other mood stabilizers. A preliminary study of 11 women with BD
found that introducing the anticonvulsant divalproex shortly after
delivery resulted in fewer recurrences than observed in otherwise
similar, but untreated, women (49). In view of the very limited
research on this important problem, further systematic study of
perinatal and postpartum prophylaxis with anticonvulsants, atypical
antipsychotics (ATPs), and nonpharmacologic interventions is urgently
needed.
Clinicians face particularly urgent challenges when a woman with
BD plans to conceive or becomes pregnant. Information accumulated
in recent decades suggests that some psychotropic drugs may be safe
for use during pregnancy (50–55). Nevertheless, knowledge regarding
the risks of prenatal exposure to psychotropic medications remains
far from complete.
All psychotropics diffuse readily across the placenta, and no psychotropic
drug has been approved by the US Food and Drug Administration (FDA)
for use during pregnancy. For obvious ethical reasons, it is not
possible to conduct randomized placebo-controlled studies on medication
safety in pregnant women. Accordingly, most information about the
reproductive safety of drugs derives from case reports, case series,
and retrospective studies. Very few reports involve prospective
designs (56,57). To guide physicians seeking information about the
reproductive safety of various prescription drugs, the FDA has established
a system that classifies medications into 5 risk categories (A,
B, C, D, and X), based on data derived from human and animal studies.
Category A medications are designated as safe for use during pregnancy
(no psychotropics have this rating), while Category X drugs are
contraindicated by having demonstrated fetal risks that outweigh
any benefit to the patient. Drugs in Categories B to D are considered
to have intermediate risks, which are greatest in category D. Most
psychotropic drugs are classified as Category C agents for which
adequate human studies are lacking and risk cannot be ruled out.
This classification system is often ambiguous, inaccurate, and misleading.
In fact, the Teratology Society recommends that the FDA pregnancy
categories be deleted from drug labelling and replaced with narrative
statements summarizing and interpreting available data on teratogenic
risk (58). At present, physicians must rely on other sources of
information when recommending the use of psychotropic medications
during pregnancy (50–55).
|
