IMPORTANCE: Nicotine replacement therapy (NRT), varenicline, and bupropion are effective smoking cessation pharmacotherapies, but evidence on fetal safety is limited.

OBJECTIVE: To assess whether prenatal use of smoking cessation pharmacotherapies was associated with increased risks of major congenital malformations (MCMs).

DESIGN, SETTING, AND PARTICIPANTS: This retrospective cohort study was conducted across 4 countries, and results were pooled via meta-analyses. Records of all births (2001-2020) in New South Wales (NSW; Australia), New Zealand (NZ), Norway, and Sweden were linked to prescribed medicine dispensings, hospital admission, outpatient, and death data. Follow-up ended December 31, 2021, and the data were analyzed between August and October 2023. The base cohort comprised 391 474 infants among 267 522 women who smoked during the first trimester or were dispensed a smoking cessation pharmacotherapy 90 days before conception or during the first trimester.

EXPOSURES: Supply of NRT, varenicline, and bupropion overlapping the first trimester. Unexposed infants were born to women who smoked but were not dispensed a pharmacotherapy 90 days preconception and the first trimester. Propensity score matching (1:10) was used.

MAIN OUTCOMES AND MEASURES: MCM overall and subgroups.

RESULTS: The mean (SD) maternal age at childbirth was 27.2 (6.0) years. Analyses included 9325 infants exposed to NRT (NSW, NZ), 3031 to varenicline (NSW, NZ, Norway, and Sweden), and 1042 to bupropion (NSW, NZ). Compared with unexposed infants, there were no differences in prevalence of MCMs overall following NRT exposure (37.6 vs 34.4 per 1000 live births; adjusted relative risk [aRR], 1.10; 95% CI, 0.98-1.22), varenicline (32.7 vs 36.6; aRR, 0.90; 95% CI, 0.73-1.10), or bupropion (35.5 vs 38.8; aRR, 0.93; 95% CI, 0.67-1.29). NRT analyses showed no difference in the risk of MCMs of the heart, limbs, genital organs, kidney/urinary tract, respiratory system, and orofacial clefts but a higher risk of digestive organ MCMs (3.8 vs 2.5 per 1000 live births; aRR, 1.53; 95% CI, 1.05-2.23; P = .41 after multiple comparison adjustment). Varenicline analyses revealed no difference in the risk of heart, limb, and genital MCMs but a higher risk of kidney/urinary tract MCMs (11.5 vs 4.2 per 1000 live births; aRR, 2.75; 95% CI, 1.42-5.34; P = .09 after multiple comparison adjustment), with findings for other MCMs being too imprecise. For bupropion, data were too sparse to estimate the risk of MCM subgroups.

CONCLUSIONS AND RELEVANCE: The results of this cohort study suggest that there is no clear increased risk of MCMs associated with prenatal use of NRT and varenicline compared with smoking during the first trimester.