AIMS: To determine the efficacy and safety of nicotine replacement therapy (NRT) with or without behavioural support when used to support smoking cessation in pregnancy. DESIGN, SETTING AND PARTICIPANTS: A systematic review of randomized controlled trials (RCTs) in which NRT was used with or without behavioural support to promote smoking cessation; trials providing unequal behavioural support to different trial groups were excluded. MEASUREMENTS: EFFICACY: self-reported smoking cessation in later pregnancy, validated where possible by biochemical measures with appropriate cut-points; infants' safety: mean and low birth weights (LBW), preterm birth, fetal demise and neonatal intensive care unit (NICU) admissions. FINDINGS: Five trials, enrolling 695 pregnant, regular smokers were included in the review. The pooled risk ratio (RR) and 95% confidence Interval (CI) for smoking cessation in later pregnancy after using NRT was 1.63 (0.85, 3.14). Subgroup analysis comparing studies at lower risk of bias (placebo-RCTs) with those at higher risk of bias (non-placebo-RCTs) found that efficacy estimates varied with trial design [RR (95% CI) for cessation in placebo-RCTs 1.17 (0.83, 1.65) versus 7.81 (1.51, 40.35) for non-placebo-RCTs]. Five of the seven safety outcomes were more positive among infants born to women who had used NRT, but none of the observed differences between trial groups reached statistical significance. CONCLUSIONS: There is currently insufficient evidence to determine whether or not nicotine replacement therapy is effective or safe when used in pregnancy for smoking cessation; further research and, in particular, placebo-randomized controlled trials are required.