The effect of active cigarette smoking on reproductive outcomes in women undergoing assisted reproductive technology...

Active cigarette smoking is a prevalent, modifiable risk factor hypothesised to impair success in assisted reproductive technology (ART). Prior meta‑analyses have combined never‑smokers and former smokers in control groups, potentially diluting effect estimates.

To determine the effect of active cigarette smoking on live birth and other ART outcomes using a strict comparison between active smokers and never‑smokers.

We conducted a systematic review and meta‑analysis (PROSPERO: CRD420251128852) per PRISMA 2020. Databases were searched for observational studies comparing ART outcomes (using autologous oocytes) in active smokers versus never‑smokers. Two reviewers independently screened records, extracted data, and assessed risk of bias (Newcastle‑Ottawa Scale). Random‑effects models (REML‑HKSJ) pooled risk ratios (RRs) for dichotomous outcomes and mean differences (MD) for oocytes retrieved. Sensitivity analyses included restriction to strict never‑smoker controls. Certainty of evidence was assessed using GRADE.

From 2585 records, 26 studies (16,359 participants) were included. Active smoking was associated with a significant 15% reduction in live birth (13 studies; RR 0.85, 95% CI: 0.79-0.92, I² = 72.4%) and 18% reduction in clinical pregnancy (21 studies; RR 0.82, 95% CI: 0.77-0.88, I² = 66.7%), but both were attenuated to non‑significance when restricted to strict never‑smoker controls (RR 0.88 for both). Miscarriage was not significantly associated with smoking (7 studies; RR 1.15, 95% CI: 0.89-1.48, I² = 78.1%). A small reduction in oocytes retrieved was observed (18 studies; MD -1.06, 95% CI: -2.01 to -0.12, I² = 98.9%). Certainty of evidence was moderate for live birth and clinical pregnancy, and low for miscarriage and oocytes retrieved.

While pooled estimates suggest active smoking is associated with reduced live birth and clinical pregnancy rates in ART, these associations are not robust when a methodologically pure never‑smoker comparator is applied, and miscarriage was non‑significant. The moderate certainty for live birth and clinical pregnancy indicates a more robust evidence base than previously appreciated, though the precise risk remains uncertain. High‑quality prospective studies with biochemical verification and dose‑response data are urgently needed.