Objective: To reevaluate the risk associated with in utero exposure to lithium.
Data sources and study selection: Data were obtained from all published studies, in multiple languages, referenced in MEDLINE, Toxline, and the Lithium Information Center databases. Unpublished studies were not included. The search terms were lithium, pregnancy, teratogen, abnormalities (drug induced), Ebstein's anomaly, and adverse effects.
Data extraction and synthesis: In the 1970s a very strong association was suggested between maternal lithium treatment during pregnancy and Ebstein's anomaly of the heart in the offspring. The relative risk for Ebstein's anomaly among such children was estimated to be 400 on the basis of data collected from a registry of voluntarily submitted cases. More recent controlled epidemiologic studies have consistently shown a lower risk. No women who took lithium during pregnancy were found among four case-control studies of Ebstein's anomaly involving 25, 34, 59, and 89 affected children, respectively. In two cohort studies, risk ratios of 3.0 (95% confidence interval [CI], 1.2 to 7.7) and 1.5 (95% CI, 0.4 to 6.8) for all congenital anomalies have been observed. The risk ratios for cardiac malformations in these studies were 7.7 (95% CI, 1.5 to 41.2) and 1.2 (95% CI, 0.1 to 18.3), respectively.
Conclusion: While initial information regarding the teratogenic risk of lithium treatment was derived from biased retrospective reports, more recent epidemiologic data indicate that the teratogenic risk of first-trimester lithium exposure is lower than previously suggested. The clinical management of women with bipolar disorder who have childbearing potential should be modified with this revised risk estimate.