Systematic review and meta-analysis of the association between miscarriage and maternal exposure to tobacco smoke during pregnancy

Background: The association of maternal cigarette smoking with miscarriage has been studied extensively, but the evidence has not yet supported a causal conclusion. Between 5 and 35% of pregnant women in each state smoke (average of 14% nationally), and many more are exposed to secondhand smoke. Methods: We conducted a systematic review and meta-analysis of studies that examined the association of active and/or passive smoking with pregnancy loss. Eligible articles were identified through a PubMed search and review of bibliographies, and had relative risk (RR) and variance data available on the association between smoking and spontaneous abortion. Summary RRs were obtained using a random-effects model with inverse-variance weighting. Results: Of 13,400 article titles reviewed, 1,093 were selected for full-text review based on the title/abstract and 105 articles were included. The RR associated with any smoking during pregnancy is 1.26 (95% confidence interval (CI): 1.16-1.36, 38 studies). Using only studies that ascertained smoking prospectively or through biochemical measures, the RR was 1.51 (95% CI: 1.27-1.78, 17 studies). Among analyses adjusted for confounders, the RR was 1.24 (95% CI: 1.05-1.47, 13 studies). Of these, 11 adjusted for maternal age, 5 adjusted for socioeconomic status, and 5 adjusted for prior pregnancy loss. Women smoking 1-10, 11-20, and >20 cigarettes per day had RRs of 1.20, 1.29, and 1.56, respectively (all p's <0.05, 16 studies). Women exposed to secondhand smoke had a RR for miscarriage of 1.13 (95% CI: 0.96-1.32, 10 studies). Conclusion: Maternal smoking increases the risk of miscarriage in a dose-related manner.