Anticonvulsant Mood Stabilizer and Lithium Use and Risk of Adverse Pregnancy Outcomes

Objective: To determine the comparative safety of mood stabilizers with respect to risk of preeclampsia, placental abruption, growth restriction, and preterm birth.

Methods: A cohort study was carried out using Medicaid Analytic eXtract data for pregnant women linked to live born infants enrolled from 2000 to 2010. Exposure to lamotrigine, valproate, topiramate, carbamazepine, oxcarbazepine, and lithium during the first 20 weeks of pregnancy was assessed. The reference group did not fill a prescription for an anticonvulsant or lithium during the 3 months prior to conception or the first half of pregnancy. Women who continued mood stabilizer monotherapy after 20 weeks were also compared to those who discontinued. Risk ratios (RRs) and 95% CIs were estimated with propensity score stratification to control for confounding.

Results: Among 1,472,672 pregnancies, 10,575 (0.7%) were exposed to anticonvulsant mood stabilizer or lithium monotherapy and 917 (0.06%) were exposed to polytherapy. In unadjusted analyses, exposure to each specific mood stabilizer and polytherapy was associated with increased risks of all adverse outcomes considered compared to no exposure (RR ranged from 1.15 to 1.56). However, these RR estimates were not meaningfully elevated with adjustment for confounding (0.89 to 1.16). Continuation of mood stabilizers was not associated with an increased risk for any outcomes compared to discontinuation and was associated with a reduced risk of placental abruption and growth restriction.

Conclusions: Anticonvulsant mood stabilizers and lithium are not associated with an increased risk of placenta-mediated complications or preterm birth after accounting for confounding by indication.