Tobacco exposure and risk of spontaneous abortion, a dose-dependent association: A systematic review and meta-analysis

Abstract

INTRODUCTION This systematic review and meta-analysis aimed to quantify the dose-dependent association between tobacco exposure (active and passive smoking) and the risk of spontaneous abortion (SA), incorporating subgroup analyses to evaluate the influence of study design. METHODS Following PRISMA guidelines, we conducted a systematic search of PubMed, Embase, and Cochrane Library databases for English-language observational studies published between 1991 and 2023. Studies were included if they reported on the association between active or passive tobacco exposure during pregnancy and SA risk (defined as pregnancy loss before 20 weeks of gestation). Studies involving induced abortion, ectopic pregnancy, or molar pregnancy were excluded. Eligible participants included pregnant women with documented smoking status. Methodological quality was assessed using MMAT, NOS, and GARD. Data were analyzed using fixed-effects or random-effects models, with heterogeneity assessed using I2 statistics. Interaction p-values were reported to evaluate heterogeneity between study designs. RESULTS Fourteen studies (5 cohort, 7 case-control, 2 nested case-control) with a combined sample size of 741698 pregnancies met the inclusion criteria. Active smoking was significantly associated with increased SA risk (OR=1.35; 95% CI: 1.18–1.55; I2=46.8%), with the highest risk observed among individuals consuming ≥20 cigarettes/day (OR=1.45; 95% CI: 1.04–2.03). Secondhand smoke exposure also elevated SA risk (OR=1.32; 95% CI: 1.14–1.55; I2=37.6%). Significant heterogeneity was observed between cohort and case-control studies (interaction p=0.001). No significant interaction was found between active and passive smoking (interaction p=0.842), but a dose-dependent interaction was observed (interaction p=0.049). CONCLUSIONS Tobacco exposure is associated with increased SA risk, particularly at higher levels. Interventions targeting heavy smokers and those exposed to secondhand smoke are needed. Limitations include imprecise smoking exposure measurement and incomplete adjustment for confounders. Future research should focus on biomarker-guided cessation strategies and explore underlying mechanisms.