previous reports of this risk, citing an odds ratio of 1.81 (95 percent confidence interval, CI = 0.96–3.09) for ventricular and atrial defects (Kallen and Olausson, 2007). This finding was based on 13 cases among 959 exposures. A subsequent analysis of data pertaining to over 3,000 exposures was performed by the Motherisk Program in Canada, which found that paroxetine was associated with a decreased risk of malformations (0.7 percent versus an established population risk of 1.0 percent) (Einarson et al., 2008). On the basis of previous reports, however, the U.S. Food and Drug Administration (FDA) chose to demote paroxetine to Category D1 status in pregnancy, discouraging its use unless absolutely necessary.

Other adverse outcomes associated with SSRI in pregnancy are worth considering, including persistent pulmonary hypertension, preterm labor, and neonatal adaptation syndrome. Persistent pulmonary hypertension (PPHN) is a relatively rare complication occurring shortly after delivery that has been associated with a 20 percent mortality rate (Hageman, Adams, and Gardner, 1984). Results of a case-control study identified 14 cases of SSRI-induced PPHN in the case group (n = 377) and 6 cases of PPHN among controls (n = 836) (Chambers et al., 2006). The risk appeared to be highest with exposure after 20 weeks gestation. While the SSRIs were associated with a sixfold increase in the relative risk of this serious phenomenon, it should also be remembered that the absolute risk remains quite low (6–12 cases per 1,000 exposures) and is probably not as serious as the risks posed by untreated depression on the mother and the infant alike.

Several studies have noted that SSRIs are associated with a decrease in gestational age, birth weight, or both. Although depression itself has been associated with these two effects, comparisons between SSRIs and other antidepressants and between SSRIs and matched controls appear to confirm these findings. For example, Suri et al. followed the outcomes of three different cohorts (antidepressants, depressed without antidepressants, nondepressed controls; n = 90) and reported significant differences in gestational age (38.5, 39.4, and 39.7 weeks, respectively), rates of preterm birth (14, 0, and 5 percent, respectively) and special care nursery admissions (21, 9, and 0 percent, respectively) (Suri et al., 2007). Results of this investigation were confirmed in the large retrospective study by Oberlander cited above. In comparison to depressed mothers not receiving pharmacotherapy, mothers who were prescribed SSRIs were much more likely to give birth before 37 weeks gestation (6 percent versus 9 percent; 95 percent CI = –0.009 to –0.04), and their infants were more likely to suffer respiratory distress


For detailed description of U.S. Food and Drug Administration risk categories for drug use in pregnancy, see Categories depend on the type of studies available and the risk of fetal abnormalities and include: A, B, C, D, and X.

The National Academies of Sciences, Engineering, and Medicine
500 Fifth St. N.W. | Washington, D.C. 20001

Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement