Birth Outcomes in Pregnant Women Taking Fluoxetine


Synopsis: Women taking fluoxetine during pregnancy do not have an increased risk of spontaneous pregnancy loss or major fetal anomalies, but women who take fluoxetine in the third trimester are at increased risk for perinatal anomalies.

Source: Chambers CD, et al. N Engl J Med 1996; 335:1010-1015.

The selective inhibitor of serotonin reuptake fluoxetine (Prozac) has become the most commonly used antidepressant drug in the United States. Its safety in pregnancy has not been formally studied. This study by Chambers and colleagues was possible because of the existence of the California Teratogen Information Service and Clinical Research Program. From 1989 through 1995, 1500 phone calls were received requesting information on the potential teratogenic effects of fluoxetine. Two hundred twenty-eight women who were pregnant and taking fluoxetine were willing to participate prospectively in this observational study, and they were matched with 254 pregnant women who called with questions about drugs and procedures not considered teratogenic. Most were enrolled in the first trimester, and all were enrolled before any outcomes of pregnancy were known. The primary indications for treatment with fluoxetine were depression (77%), anxiety (8%), panic disorder (6%), bipolar disorder (6%), and obsessive-compulsive disorder (4%). The proportion of women who continued to smoke was higher in those who were exposed only during the first and second trimesters (exposed-early group, 10%) and in those who were exposed all three trimesters (exposed-late group, 18%) vs. the control group (4%). This and other confounding variables were controlled for in the analysis.

The rate of major structural anomalies in the offspring of women in the two fluoxetine groups was not higher than the rate in control women. However, the proportion of infants with three or more minor anomalies was higher in the exposed groups (15%) than in the control group (6%). Miscarriage rates were similar. The rate of prematurity was higher in the exposed-late group than in the exposed-early or control groups. The rate of admission to special-care nurseries was 23% in the exposed-late group as compared to 9.5% for the exposed-early and 6.5% for the control group. Poor neonatal adaptation was much higher in the exposed-late vs. exposed-early group. For full-term infants in the exposed-late group, birth weight was lower, birth length was shorter, and there were two cases of pulmonary hypertension. The authors interpret the increased rate of minor anomalies in offspring exposed to fluoxetine as evidence of an effect on embryogenesis. However, their primary concern is for babies whose mothers take fluoxetine late in pregnancy. Whether the effects of fluoxetine use in late gestation are due to the underlying psychiatric disorder and its biological concomitants or due primarily to fluoxetine exposure cannot be directly addressed by this study. Nonetheless, the authors suggest that fluoxetine be discontinued if possible before the last trimester.


This is an important study for several reasons. First, it addresses a critical question about the fetal and neonatal effects of fluoxetine exposure. Second, the methodology is exemplary. Although observational studies are not without inherent limitations, the present investigation permitted the selection of a reasonable control group, and it was prospectively conducted. It seems unlikely that a randomized, controlled trial to address the teratogenicity and neonatal outcome of fluoxetine or any antidepressant would be feasible, so this is the next best approach in humans. Unfortunately, the findings are not reassuring. When possible, given the present results, nonpharmacological intervention during pregnancy is advised. Previous studies using less desirable methodology had suggested that fluoxetine was "safe" for use in pregnancy. The same limitation may apply to studies suggesting that other antidepressants are reasonably safe. On the other hand, psychiatric disorders carry their own risks. Therefore, an individualized risk-to-benefit analysis must be done for any woman taking psychotropics for a known psychiatric condition, either prior to attempting conception or as soon as the pregnancy is diagnosed. Personally, I would not want to attempt a risk-to-benefit analysis without the input of a psychiatrist with a special interest in psychiatric adjustment to pregnancy and the postpartum period. As the authors of this article point out, the stress of a psychiatric disorder also poses a threat to maternal adaptation to pregnancy and fetal and neonatal health. In countries where maternal-infant health is more of a priority, such specialists and special care units exist. Even before the legacy of managed care was upon us, maternal-infant mental health received less attention than it was due in the United States. Depression is a common disorder and very likely to be encountered by those practicing obstetrics. Thus, we need to continue to advocate for research and specialized clinical services in this domain.