Depression occurs in 14% to 23% of pregnant women and can be devastating for the mother-to-be if left untreated. Yet treatment with an antidepressant during pregnancy has been associated in case reports and small case series with a risk of complications including cardiac malformations, preterm birth, and newborn respiratory distress. To help determine the magnitude of risk, Bandoli et al (10.1542/peds.2019-2493) turned to a massive data warehouse that includes over 200 million lives and identified 15,000 pregnancies where newborns had been exposed to antidepressants prenatally. The authors established five use trajectories for these medications including (1) low use with tapering down in the first trimester, (2) low sustained use, (3) moderate use with tapering down in first trimester, (4) moderate sustained use, and (5) high sustained use and compared the risk of major cardiac malformations, prematurity or respiratory distress in these subgroups to control groups. Mothers-to-be who fell into the low use category (either tapered or sustained) did not show an increase in cardiac defects or preterm birth but the moderate and high use groups had an increased risk for newborn respiratory distress in a dose response fashion. Pregnant women in the moderate to high groups also had an increase in preterm birth and an increase in cardiac malformations compared to the lower group but not when compared to a control group made up of pregnant women with depression and/or anxiety not on an antidepressant medication.
While these findings suggest lower doses of antidepressants may be safest for the developing fetus, the study has important limitations. We invited Drs. Sascha Dublin, Paige Wartko and Rita Mangione-Smith from the University of Washington to share their thoughts in an accompanying commentary (10.1542/peds.2020-1540). The authors point out the Bandoli et al study is perhaps one of the best examples of how we might study medication safety in pregnancy, but also point out the limitations of this kind of research using massive data repositories. Drs. Dublin, Wartko, and Mangione-Sminth also highlight how new approaches, use of smartphone technology, and collaborations between large data sets and interviewing or surveying individual patients and their electronic health records may help us get better information on the ramifications of various medications used during pregnancy and their outcomes on parent and child. Link to this study and commentary and learn more.