Pharmacologic treatment of depression during pregnancy

Context Despite the frequency of depression in women of childbearing age, i nformation to guide patients and physicians through a consideration of trea tment during pregnancy is limited. Objective To identify risk factors associated with treatment of major depre ssion during pregnancy to help physicians develop treatment plans that opti mize clinical care. Data Sources Reports of prospective controlled trials in English were ident ified from MEDLINE and Health STAR using the search terms antidepressant du ring pregnancy and depression during pregnancy, by manually searching bibli ographies of review articles, and through discussions with investigators fo r 1989-1999. Study Selection We selected studies in which maternal and infant health out comes associated with antidepressant exposure were compared with those of n on-teratogen-exposed controls. Four studies published since 1993 were ident ified and included in the analysis. Data Extraction We abstracted information about identification of subjects, comparison groups, pregnancy, and birth outcomes. We organized the data al ong 5 domains of reproductive toxicity: intrauterine fetal death, morpholog ic teratogenicity, growth impairment, behavioral teratogenicity, and neonat al toxicity. Data Synthesis Data were available for tricyclic antidepressants (collectiv ely), fluoxetine, and newer selective serotonin reuptake inhibitors (collec tively), Exposure to these agents did not increase risk for intrauterine de ath or major birth defects. Decreased birth weights of infants exposed to f luoxetine in the third trimester were identified in 1 study. The developmen t of children whose mothers took tricyclics or fluoxetine during gestation did not differ from that of controls. Direct drug effects and withdrawal sy ndromes occurred in some neonates whose mothers were treated with antidepre ssants near term, Conclusions Although few in number, new information from prospective studie s provides a welcome change from decision making based on nonprospective da ta. Monitoring and interventions for patients with identified risks leg, po or weight gain) are recommended.