Posted on

Perinatal Psychology

by John M Rathbun, MD


A young woman and her fiancé came to see me because they wanted to have a baby and they desired my opinion about when she should stop taking her lithium. She was also on Wellbutrin and Cytomel. What would you tell them?

A pregnant woman with a history of intractable depression came to me because she knew from past experience that one electroconvulsive therapy treatment would relieve her depression for up to a month, but several health care professionals had told her that she couldn't have ECT because it would hurt the baby. What would you tell her?

Another patient of mine called because she was close to parturition and wanted to know if she should stop her antidepressant before delivery so she could breast feed. What would you tell her?

We used to think that pregnancy somehow protected women from severe psychiatric illness, but new data indicate that women are at least as likely to show psychiatric illness during pregnancy as at any other time. A recent literature review showed rates of depression in pregnancy anywhere from 5% to nearly 30%. When ten women with previous history of panic disorder were studied during pregnancy, seven continued symptomatic throughout the pregnancy.

A retrospective study of 29 women with OCD prior to pregnancy showed that nearly 70% experienced no change in symptoms during pregnancy; the other 30% were evenly divided between those who experienced improvement during pregnancy and those whose OCD got worse during pregnancy. Discontinuation of psychotropic medications in pregnant women has been associated with high rates of relapse in numerous reported cases of mood disorders, anxiety disorders, and schizophrenia.

Psychiatric illness during pregnancy is not only common, but causes considerable distress for the mother and risk for the fetus. Compliance with recommended prenatal care is poor in proportion to the severity of the illness. Common consequences of severe psychiatric illness include malnutrition, dehydration, heavy smoking, abuse of alcohol and other toxic substances, lack of rest, lack of exercise, and chronic high levels of stress hormones such as cortisol. These are all known risk factors for poor fetal outcomes, including intellectual and behavioral problems which can persist throughout life.

Diagnosis of psychiatric illness in pregnancy is complicated by the frequency with which sleep and appetite disturbance are found in uncomplicated pregnancies. Vivien Burt, who directs the Women's Life Center at the University of California in Los Angeles, has noted that during pregnancy, women with significant depression are much more likely than healthy women to complain vociferously of total insomnia.

Given the impersonal and hurried nature of so many clinical encounters these days, I suggest routine screening of pregnant women with a psychiatric symptom questionnaire, such as the Beck Depression Inventory. The Beck is particularly useful in medical settings because it ignores neurovegetative symptoms and concentrates on depressive thinking patterns. [seereferences]

Back to Top

It's important to distinguish illness which is dangerous from that which is merely uncomfortable. Pregnancy is an inherently uncomfortable condition. First pregnancies are among the most profoundly disturbing events in most people's lives. One's anatomy and physiology undergo cataclysmic changes, and one's personal identity becomes absorbed in the universal human drama of bringing a new life into the world.

Pregnancy can have many meanings, from the mysterious merger of two lives into a common biological destiny, to an experience of profound frustration, violation, and failure. The active, carefree woman may become bedfast and subject to annoying regimens which she doesn't dare resent because, of course, we all want to do what's best for the baby. Morning sickness, uncharacteristic emotional lability, and extreme dietary preferences are common stressors in the first trimester. Nicotine craving and abstinence from the solace of an occasional stiff drink may complicate the expectant mother's adjustment. As delivery approaches, women often begin to feel awkward and vulnerable, with additional discomfort from backache and from compression of abdominal organs such as the urinary bladder.

Given these stressors, it shouldn't be surprising that anxiety and depressive reactions are the most common emotional disturbances in pregnancy. They are usually manageable with counseling, support groups, environmental manipulation, and nonspecific diversions such as walking, massage, warm baths, and keeping up social contacts.

The provision of environmental support is particularly important in all phases of reproduction. Our fragmented society is seriously deficient in the supports offered to young families. Only in the past few decades has it become common for young families to relocate beyond easy reach of their families of origin. This practice leaves the expectant mother with a dearth of competent help when she most needs it.

Many young families need encouragement to ask for live-in assistance from relatives in distant cities. Pregnant women and new mothers need this kind of assistance. Provision of adequate assistance is often the most important step to resolution of emotional turmoil in a pregnant or lactating woman.

When depression and anxiety fail to respond to common-sense approaches, professional psychotherapy is indicated. Brief hospital stays and intensive outpatient programs will often be as effective as medication in moderately severe cases.

Very severe cases of depression and anxiety often will not respond to the low-risk management strategies previously discussed. Even continuous hospitalization is not a sufficient treatment for the most severely ill patients.

I hope you're now eager to hear which psychotropic medications have been approved by the FDA for use in pregnant women. It's a very short list. In fact, the list is still waiting for its first entry.

Nobody is eager to do research that involves giving psychotropic medications to pregnant women. Furthermore, the species that have been chosen for animal reproduction studies were selected because they are cheap to keep, and because they breed rapidly, not because their physiology is very similar to ours.

Cost considerations weigh against exposing a million rodents to therapeutic blood levels of psychotropic medications; the alternative is to expose a thousand rodents to very large doses. Not surprisingly, bunnies fed lovely fresh salads during pregnancy do better than those whose diet is largely pharmaceutical.

Given that corporate legal eagles have enormous influence on what gets published in a drug's prescribing information, you won't find much encouragement in the PDR for humane treatment of psychiatrically ill pregnant women.

Even those of us who practice in Indiana, where the malpractice situation is less obnoxious than elsewhere in the U.S., have reason to hesitate before medicating pregnant women. After all, congenital malformations are found at a base rate between two and seven per one hundred live births. The use of unapproved medications during a series of pregnancies seems likely to result in some cases of birth defects which might be attributed to the medication by parents whose guilt or anger can cause complications for themselves and for their physicians.

OTOH, it is not true that doing nothing is always preferable to an intervention that carries some risk. You can be sued as easily for acts of omission as for acts of commission, and our profession cannot pursue its stated goals without assuming some risk.

Back to Top

Keys to avoiding litigation are:

  • be aware of reasonable standards of practice
  • be familiar with research findings which may lead to changes in the standards
  • provide written justification of departures from standard practices
  • cultivate trust in your patients by taking time to show your interest and concern for their welfare
  • refer patients for consultation in cases where there is above average potential for liability.

Malpractice litigation more commonly reflects a failure of empathy on the doctor's part than a failure of technique. Doctors are selected and schooled to be tough in the face of tough situations, but this appearance of toughness is out of place in the presence of a patient who's struggling to adjust to a bad outcome. Doctors who are truly compassionate may be so affected by grief and anxiety when things go out of control that they walk away when the patient most needs their presence.

It's much better for all concerned if you can stay around and let the patient see your sorrow over what's happened. You want to emphasize whatever hope remains in a difficult case, but not to the point of pretending that something's truly OK when it's obviously not.

You can say, "I'm sorry about how this turned out!" without it sounding like, "I'm sorry I botched your case!" Either statement would be better than abrupt withdrawal or some other defensive reaction.

Dr. Vivien Burt counsels her pregnant patients, in the presence of a significant other, about the risks, benefits, and alternatives of medical treatment. Then she asks the patient to write on one page all the reasons for taking the treatment, and on a second page all the reasons against taking the treatment. On the third page go the woman's decision and her justification of the decision. A copy is kept by the patient and another goes on the patient's chart.

Not only is this excellent documentation of informed consent, it's a very useful tool in counseling women who are later overcome by anxiety over having placed their baby at risk. The physiologic stress of pregnancy and severe psychiatric illness can virtually eliminate the patient's ability to recall, after the pregnancy, how bad she felt and what she was told. Seeing the details in her own handwriting is very helpful in such cases.

Concerns over gestational exposure to psychotropic medications have focused on three areas: teratogenicity, neonatal toxicity, and postnatal behavioral sequelae. Teratogenicity is of primary concern when exposure occurs during the first trimester, although the fetal brain develops throughout pregnancy and damage could occur after the first trimester.

A relatively small number of cases of first trimester exposure to antidepressants have been reported. These reports have suggested no increased risk of birth defects. Given that major depression occurs in nearly 10% of women, and that we've been freely treating these women with tricyclic antidepressants for over 30 years, there are probably millions of cases of accidental first trimester exposure. In this context, the dearth of reports suggesting teratogenicity is encouraging.

The newer antidepressants are generally less toxic for adults than are the tricyclics. The serotonergic antidepressants are thought to be relatively pure in their pharmacologic activity, and their main effect is rather similar to that of amitriptyline, which is one of the oldest antidepressants. Therefore, we would not expect the newer antidepressants to be more fetotoxic than the tricyclics. Observation, however, is better than theory.

Back to Top

Prozac was the first of the new breed of antidepressant, having been introduced in this country by Eli Lilly ten years ago. It immediately became the most prescribed antidepressant in the U.S., and has probably been used by twenty million people, of whom two- thirds were likely to have been women. Lilly has maintained a worldwide pregnancy registry which includes pregnancies occurring during scientific studies as well as cases reported from naturalistic clinical settings. As of 1996, they had over 1700 cases on file, which showed no increase in birth defects, no suspicious clustering of abnormalities, and no increase in spontaneous abortions. Data for other serotonergic antidepressants are less voluminous, but no less reassuring.

With respect to perinatal complications, the picture is a bit less clear. Tricyclic antidepressants are known to cause an uncomfortable abstinence syndrome in some adults, lasting several days. The same sort of thing has been seen in infants exposed to these agents shortly before delivery. Interestingly, the long half- life of Prozac's active metabolite norfluoxetine may be advantageous when children are exposed to it immediately prior to delivery: no abstinence syndrome has been identified with Prozac in adults or in newborns.

The best long-term study of antidepressant-exposed fetuses followed 80 cases up to age seven. Compared to an unexposed control group, the exposed children showed no adverse effect on IQ, no adverse effect on language development, and no adverse effect on behavior.

Antidepressants such as Nardil and Parnate, which achieve their effects through the inhibition of monoamine oxidase, are contraindicated during pregnancy. They require dietary restriction which can compromise the expectant mother's nutritional status, they can raise or lower blood pressure, and there's a severe adverse interaction with terbutaline, which is used to suppress premature labor.

Given the known risks to mother and fetus from severe depression during pregnancy, with the absence of any data suggestive of increased risk to the fetus from exposure to antidepressants that have been in widespread use for over 30 years, I would favor use of antidepressants after appropriate counseling in selected cases. The most evidence supports use of Prozac and the tricyclic antidepressants, while MAOI antidepressants are contraindicated.

These data have application to the counseling of psychiatrically ill women who wish to become pregnant. The preponderance of the evidence is that depression is a progressive illness in many cases, with a pronounced tendency to recur under stress. Although it has long been felt appropriate to subject depressed persons to repeated trials of medication reduction, that approach does more harm than good. Reduction or discontinuation of antidepressant medication is contraindicated when there's a strong family history of depression, two or more distinct episodes of major depression, minimally- provoked episodes of major depression, onset of major depression before adulthood, or a stressful life situation in a person with a history of major depression.

While there is around 15% risk of transmission of the illness to each offspring, and the risk is doubled if both parents are affected, the prognosis of major depression with appropriate treatment is now so good that few physicians would discourage conception because of this illness.

Given the high risk of relapse and low risk of fetal damage, discontinuation of antidepressant medication in women who wish to become pregnant cannot be recommended as a routine practice. Many low-threat cases can reasonably be given a trial of dose reduction, but exposing those with severe or chronic history to such risk is inappropriate.

Electro-Convulsive Therapy (ECT) is the safest and most effective treatment for severe mood disorders, and especially so in pregnant women. The risk to the fetus is no more than the risk of brief general anesthesia.

There are some drawbacks to the use of ECT. It costs nearly $500 per treatment, and usually requires six to twelve treatments in the first few months, followed by at least one treatment monthly for maintenance. It causes acute memory problems, which are almost never persistent, but which require the patient who is getting several treatments weekly to be supervised closely by friends and relatives if the patient is not in the hospital.

The effects of ECT are rapid and dramatic: the response rate is as high as 85%, even in refractory illness. Contrary to what you may think, most people don't find ECT very unpleasant, and many prefer it to medication because of its effectiveness and freedom from daily side effects.


After delivery, things become a lot more difficult. Those of you with children have probably already noticed this!

About 80% of women have mildly depressed mood during some portion of the period between days three and twelve postpartum. This is not surprising, given the stress of labor and delivery, the enormous changes in hormone levels that occur at the end of pregnancy, and the demand for the new mother to make an instant transition from being taken care of by others to being the primary caretaker for an infant that requires almost constant attention. Sleep deprivation adds considerably to the stress of the first few months after parturition.

Ordinary postpartum blues are easily managed if enough support and assistance are available to the new mother, but the outcome is not always benign. Severe postpartum psychiatric illness is the most common complication of pregnancy, and can devastate both mother and infant.

A woman is twenty times as likely to require psychiatric hospitalization in the first month after pregnancy as in any other month of her life. Between ten and fifteen per cent of women suffer significant postpartum psychiatric illness, and the severity is often extreme. Infanticide and suicide are very real risks in this situation, and failure to thrive is the best that can be expected in an infant whose mother is psychotic. A study of 700 cases indicated that these infants do poorly for years afterward, with intellectual, behavioral, and emotional problems.

The increased risk for postpartum psychiatric illness persists for at least six months, and for as long as twenty-four months in some studies. For persons with no prior psychiatric history, the risk of severe postpartum psychiatric illness is 10%. If there's a history of prior major depression, the risk rises to 25%. For women with a history of postpartum psychiatric illness, the risk of postpartum recurrence is 50%. If the prior postpartum illness was major depression, the recurrence rate following a subsequent pregnancy is 62%. If she was psychotic during her prior postpartum illness, the risk of postpartum recurrence is 75%. If a woman has postpartum psychosis and becomes pregnant again within two years, the relapse rate is nearly 100%.

Fortunately, severe postpartum psychiatric illness is largely preventable. For example, if a woman with a prior history of postpartum major depression is started on antidepressant medication within 24 hours of delivery, the recurrence rate drops from sixty- two percent to seven percent. Prompt and vigorous treatment with medications or ECT likewise improves the outcome for other severe postpartum psychiatric illness.

Some authorities have suggested that psychotropic medications be reduced or discontinued for a week or two before the expected delivery date. They base this recommendation on scattered case reports of neonatal syndromes which have been attributed to persistence of medication effects or to withdrawal effects. These problems may occur, but they are infrequent and mild compared to the high frequency and serious consequences of perinatal relapse in the mother. I believe it's inappropriate to deprive a woman of pharmacologic treatment for a severe, persistent psychiatric illness at the very time when she's most likely to experience recurrence.

Back to Top>


The issues surrounding treatment of psychiatric illness in women following childbirth are considerably complicated by questions related to breast feeding. This activity is highly valued by most women and apparently quite beneficial to infants.

Breast feeding is less expensive than bottle feeding, and it supplies the infant with nucleotides and enzymes which facilitate digestion and absorption of nutrients. Breast-fed infants also receive immune factors which prevent infections, with a measurable improvement in survival. The longer-chain polyunsaturated fatty acids found in human milk are thought to promote continued development of the nervous system, which is relatively immature at birth in a human infant compared to other animals. When evaluated during the first year of school, both full-term and premature infants show enhanced intellectual development if they were breast fed. Mothers who breast feed benefit from hormonal factors which shrink the uterus, suppress ovulation, and reduce risk of breast cancer.

Of course there's no such thing as a free lunch. All psychotropic medications appear in breast milk to some degree. Furthermore, the infant's liver is relatively inadequate during the first two weeks after full-term delivery. These facts cause much tension between mothers and their doctors, and between one doctor and another.

Our increasingly fragmented and litigious system of health care delivery often requires the collaboration of doctors specializing in pediatrics, obstetrics, and psychiatry, each of whom may develop a narrow focus on some issues without due consideration of competing interests. Maximizing the welfare of both mother and infant requires complex thinking.

Since major depression is the most common postpartum illness, there are substantial data on babies exposed to antidepressants in breast milk. It appears that normal newborns exposed to tricyclic antidepressants through breast feeding rarely suffer ill effects, and rarely have detectable blood levels of parent compounds or active metabolites.

The main exception is doxepin. One eight-week-old infant became sedated while being breast fed by a mother taking doxepin, and was found to have accumulated an active metabolite of doxepin to a concentration comparable to that found in his mother's blood. In another report, a 10 week-old infant was found to have a quantifiable level of desmethyldoxepin, but there was no ill effect.

One case has been reported of an infant aged four weeks being nursed by a woman taking nortriptyline, in which the infant had a serum level about 10% of the mother's; there was no apparent effect on the baby. There are numerous case reports of infants being breast fed by women taking various tricyclic antidepressants other than doxepin, of whom none had levels high enough to be quantified, and none suffered an ill effect.

The serotonergic antidepressants have less data, and these data are mixed. Fluoxetine and sertraline have the most data in this group; both agents have active metabolites with long half lives, and both have been found in the blood of infants being nursed by mothers taking these medications.

In only one case was there a report of an adverse effect on the infant: that child developed colic which got better within a few days after breast feeding was discontinued. The problem returned when the infant was given mother's milk in a bottle. This infant's blood was found to contain levels of fluoxetine and norfluoxetine comparable to those found in the mother. The accuracy of this report has been questioned because the infant's blood was drawn only two days after resumption of nursing following three weeks of formula feeding, and the mother's milk contained an insufficient amount of medication to give rise to such high serum levels in the infant. Either there was laboratory error, or that infant maintained a very high drug level for three weeks on formula, in which case one wonders why the colic got better after only a few days back on formula.

Back to Top


To summarize, women with chronic or recurrent depression may require antidepressant medication during pregnancy, which may be continued up to initiation of labor and then restarted within 24 hours thereafter. There are no compelling data against breast feeding on a tricyclic antidepressant other than doxepin. Serotonergic antidepressants remain controversial because they more commonly generate substantial serum levels, even though there's only one report of a possible adverse effect.

It would be reasonable to consider reducing the doses of serotonergic antidepressants in pregnant and lactating women with augmentation by a tricyclic antidepressant other than doxepin, thereby reducing fetal and infant exposure to the newer compound while maintaining maternal health. This approach can be criticized because it involves exposure to two agents, but lower doses of both agents can be used, so the side effects may be less.

The situation I have just described calls for extensive discussion of the risks, benefits, and alternatives. While we can't promise safety, the risk of the mother going off medication so she can breast feed is unacceptably high, and this should be clearly communicated. Careful monitoring of the infant's development is mandatory, and it's reasonable to obtain a serum level if the baby's behavior changes substantially from baseline. Serum levels may also be indicated to alleviate anxiety in parents and physicians. After ten weeks of age, infants can metabolize drugs more than twice as fast as adults, so the risk of accumulation of toxic metabolites is very low.

Back to Top


When it comes to bipolar disorder, the situation becomes considerably more troublesome. Bipolar disorder is a more dangerous illness to the mother and fetus, with a very high rate of relapse when medication is discontinued. Tricyclic antidepressants are known to make bipolar disorder worse, sometimes very much worse. Serotonergic antidepressants appear to be better tolerated, but have some risk of precipitating mania. Almost all the treatment options for mania are unattractive during pregnancy.

There are four anticonvulsants with antimanic properties. Of these, carbamazepine and valproate are known teratogens. Gabapentin and lamotrigine are too new to have much data.

Lithium is the oldest known mood stabilizer. It is also a known teratogen, although the risks are manageable. Initial reports of lithium's teratogenicity came from a registry that attracted mostly problem reports and lead to substantial overestimation of the risks.

First trimester exposure to lithium causes Ebstein's anomaly, which is almost always lethal. The rate is now thought to be about one case for every 1000 exposures. This low rate of a serious outcome must be compared with the 50% relapse rate for bipolar disorder after 20 weeks off lithium. A bipolar in full relapse presents serious dangers to self and others, much more so during pregnancy. Only in the mildest cases should lithium withdrawal be considered in the first trimester.

When first-trimester exposure to lithium occurs, a level II ultrasound should be performed between 16 and 18 weeks gestation to rule out Ebstein's anomaly, which can be diagnosed reliably by this method.

Lithium doses must often be higher during pregnancy due to increased renal clearance. Maintenance of hydration during pregnancy and delivery are particularly important for pregnant women on lithium. Lithium-induced diabetes insipidus in the fetus can occasionally cause maternal respiratory embarassment due to expansion of amiotic fluid.

Bipolar disorder is the most strongly inherited illness in psychiatry. With one affected parent, the morbid risk is 33%, and with two affected parents, around 67% become ill. Any family history of bipolar disorder confers significant increased risk. Bipolar disorder is also a lot more difficult to control than unipolar depression, and more hazardous to all concerned. These facts must be made known to any woman of childbearing age who has any family or personal history of bipolar disorder, either in herself or in her partner.

Lithium is present in breast milk in rather high concentrations, so breast feeding is not recommended for mothers who must take lithium. Bipolar disorder has a high risk of catastrophic relapse, making it inappropriate to suggest stopping the lithium as an alternative to stopping the breast feeding.

Back to Top


Psychosis during pregnancy is a dire situation for mother and fetus. Psychotic persons will not follow any prescribed regimen, and they are seriously deficient in basic self-care, self- protection, and common-sense restraints on behavior. Such cases often require involuntary hospitalization for diagnostic studies and initiation of pharmacologic treatment.

In cases where a diagnosis of schizophrenia has been established, there's usually no alternative to maintenance treatment with antipsychotic medication. Intermittent medical treatment of schizophrenia is unsafe because full or partial relapse is inevitable when adequate levels of antipsychotic medication are not maintained throughout the patient's life. Relapse commonly triggers a major crisis which places the patient, the fetus, and other close contacts at risk for serious injury or death.

The oldest antipsychotic medications are known as "low-potency" neuroleptics because the effective dose is on the order of 1000 mg daily. These agents have prominent anticholinergic and antiadrenergic side effects. They used to be given to nonpsychotic pregnant women in low doses to treat hyperemesis gravidarum. Data from these cases suggest that such exposure in the first trimester increases the rate of birth defects by about four cases per thousand.

The second generation antipsychotic medications are known as "high- potency" neuroleptics, because they are effective in daily doses under 100 mg. These agents are much less likely to cause anticholinergic and antiadrenergic side effects, but they can cause imbalance in the extrapyramidal motor system leading to stiffness and rigidity of voluntary muscles. There is no evidence linking these agents with birth defects, and they've been around nearly thirty years. Continuous administration throughout pregnancy is preferred, because intermittent administration carries a high risk of relapse and requires much larger doses, leading to increased fetal exposure and uncomfortable muscular side effects.

Most of the medications we use to relieve muscular side effects in persons taking neuroleptics are suspected teratogens; this includes benztropine, trihexyphenidyl, and amantadine. The safest agent for managing extrapyramidal syndromes in pregnant women appears to be diphenhydramine.

The newest group of antipsychotic medications includes clozapine, risperidone, and olanzapine. We don't yet have enough data on these agents to be reassured about fetal risk.

Current data favor using the lowest effective dose of a high- potency neuroleptic on a continuous basis throughout pregnancy for schizophrenic women. The most data in this regard are for haloperidol, which comes in a long-acting injectable form that has a half life of nearly a month and keeps serum levels as low as possible. Since medication noncompliance is a major problem in the maintenance treatment of schizophrenia, the long-acting injectable medications are strongly preferred.

Children of schizophrenic mothers are at high risk for poor outcome. There's a high rate of perinatal death, fetal malformations, and postnatal difficulties with intellectual and social development. Schizophrenia is as strongly inherited as major depression, and the consequences are so devastating as to make genetic counseling obligatory.

Those offspring who don't turn out schizophrenic have a high rate of other significant psychiatric illness. Few schizophrenic women can meet a normal child's needs for affection, stability of environment, and consistent discipline appropriate to the child's developmental level. Compliance with most forms of birth control has been poor among chronic schizophrenics; those forms of birth control which don't require the mother's consistent compliance are to be preferred despite their drawbacks.

There's little data on the safety of antipsychotic agents in breast feeding. Prudence would suggest against the practice, given what we know about the effects of antipsychotic agents on the adult brain. Stopping treatment to allow breast feeding would not be advisable in any chronic psychosis, because the risk of relapse is too high and psychotic relapse can be an irremediable disaster.

Back to Top


The situation regarding use of benzodiazepine tranquilizers in pregnancy and lactation is very unclear. The best review I was able to find [McElhatten in references] indicates that diazepam accumulates in the fetus to double or triple the maternal level. Some studies have shown increased risk for oral cleft anomalies when diazepam is given in the first trimester, while others have not. Diazepam use during labor has been associated with low APGAR scores, apnea, hypotonia, poor feeding, and loss of thermoregulation. Diazepam and desmethyldiazepam were found in neonatal serum up to a week after maternal exposure. Exposure to diazepam in breast milk leads to ingestion by the infant of a dose about 5% of mother's dose on a weight adjusted basis; neonates can't clear diazepam very well in their first week, so infants exposed to diazepam in mother's milk should be observed for sedation.

Lorazepam crosses the placenta more slowly than diazepam and is metabolized more rapidly. Several studies have indicated a lack of accumulation in the fetus above maternal blood levels. There have been no reports of increased malformations in babies exposed to lorazepam. One study indicated rapid clearance of lorazepam by neonates, essentially complete within 24 hours. Apgar scores in this study were unimpaired, although another study found that when lorazepam was used during labor, the babies were often impaired in a manner similar to those whose mothers were exposed to diazepam during labor. Babies whose mothers take lorazepam and breast feed provide a clinically insignificant dose to the infant unless they ingest unusually high doses.

The situation for other benzodiazepines is intermediate between the diazepam case and the lorazepam case. In general, benzodiazepines with shorter half-lives and fewer active metabolites seem to be less toxic to the fetus and neonate. Fortunately, benzodiazepine use is rarely essential during pregnancy. For women who may become pregnant while taking a benzodiazepine, lorazepam would be preferred.

Back to Top


In summary, we know a great deal more now than we did a few years ago regarding indications and contraindications for psychotropic medications in pregnant and lactating women.

Cautious optimism and thoughtful therapeutic activism are appropriate when severe mood disorders make their appearance perinatally.

Postpartum psychiatric illness is particularly common, severe, and recurrent, but can often be managed without undue disruption of bonding.

Breast feeding while taking tricyclic antidepressants is not contraindicated, and the limited data regarding the newer antidepressants is not particularly adverse.

Bipolar disorder and schizophrenia continue to present dilemmas which warrant specialty consultation in every case, the more so in conjunction with pregnancy.


Copyright 1999 John M Rathbun MD

Reviewed by athealth on February 6, 2014.