Systematic Review and Meta-Analysis to Assess the Safety of Bupropion and Varenicline in Pregnancy.

Abstract

Smoking in pregnancy is a substantial public health issue, but, apart from nicotine replacement therapy (NRT), pharmacological therapies are not generally used to promote cessation. Bupropion and varenicline are effective cessation methods in nonpregnant smokers and this systematic review investigates their safety in pregnancy. We searched MEDLINE, EMBASE, CINAHL, and PsychINFO databases for studies of any design reporting pregnancy outcomes after bupropion or varenicline exposure. We included studies of bupropion used for smoking cessation, depression, or where the indication was unspecified. Depending on study design, quality was assessed using the Newcastle-Ottawa Scale or Cochrane Risk of Bias Tool. Most findings are reported narratively but meta-analyses were used to produce pooled estimates for the proportion of live births with congenital malformations and of the mean birthweight and gestational age at delivery following bupropion exposure. In total, 18 studies were included: 2 randomized controlled trials, 11 cohorts, 2 case- control studies, and 3 case reports. Study quality was variable. Gestational safety outcomes were reported in 14 bupropion and 4 varenicline studies. Meaningful meta-analysis was only possible for bupropion exposure, for which the pooled estimated proportion of congenital malformations amongst live-born infants was 1.0% (95% CI = 0.0%-3.0%, I2 = 80.9%, 4 studies) and the mean birthweight and mean gestational age at delivery was 3305.9 g (95% CI = 3173.2-3438.7 g, I2 = 77.6%, 5 studies) and 39.2 weeks (95% CI = 38.8-39.6 weeks, I2 = 69.9%, 5 studies), respectively. There was no strong evidence that either major positive or negative outcomes were associated with gestational use of bupropion or varenicline. PROSPERO registration number CRD42017067064. We believe this to be the first systematic review investigating the safety of bupropion and varenicline in pregnancy. Meta-analysis of outcomes following bupropion exposure in pregnancy suggests that there are no major positive or negative impacts on the rate of congenital abnormalities, birthweight, or premature birth. Overall, we found no evidence that either of these treatments might be harmful in pregnancy, and no strong evidence to suggest safety, but available evidence is of poor quality.