Objectives Leflunomide is known to be embryotoxic and teratogenic in rodents. However, there is less evidence in humans. We quantified the risk of major congenital malformation (MCM), prematurity, low birth weight (LBW) and spontaneous abortion associated with leflunomide exposure during pregnancy in humans.

Methods From a cohort of 289 688 pregnancies in Montreal, Quebec, Canada, from 1998 to 2015, first-trimester leflunomide exposure and other antirheumatic drug exposures were studied for their association with MCM and spontaneous abortions. Also second or third-trimester leflunomide exposures were examined for associations with prematurity and LBW. Logistic regression model-based generalised estimating equations were used.

Results 51 pregnancies were exposed to leflunomide during the first trimester, and 21 during the second/third trimesters. Adjusting for potential confounders, use of leflunomide during the first trimester of pregnancy was not associated with the risk of MCM (adjusted OR (aOR) 0.97, 95% CI 0.81 to 1.16; 5 exposed cases). No association was found between second/third-trimester exposure to leflunomide and the risk of prematurity (aOR 4.03, 95% CI 0.91 to 17.85; 7 exposed cases) nor LBW (aOR 1.06, 95%CI 0.90 to 1.25; 8 exposed cases). Pregnancy exposure to leflunomide was also not associated with the risk of spontaneous abortion (aOR 1.09, 95% CI 0.90 to 1.32; 11 exposed cases).

Conclusions Maternal exposure to leflunomide during pregnancy was not associated with statistically significant increased risk of MCMs, prematurity, LBW or spontaneous abortions. However, given that relatively few women were exposed to leflunomide during pregnancy in this cohort, caution remains warranted.