Background Some studies have reported an increased risk of major congenital malformations (MCM) associated with hydroxychloroquine (HCQ), but others find no increased risk. We aim to assess the risk of MCM associated with 1st-trimester HCQ exposure in the offspring of women with lupus.

Methods We conducted a population-based cohort study of pregnancies (2006–2020) with a singleton birth among women with prevalent lupus in Sweden. Prevalent lupus was defined as having ≥ two ICD-coded visits in the National Patient Register before pregnancy, with ≥ one with a lupus specialist. HCQ exposure was defined as filling ≥ one HCQ prescription during the 1st trimester (Prescribed Drug Register). MCM was assessed at birth by any ICD code in the Swedish Medical Birth Register. Inverse probability of treatment weighting (IPTW) was applied to adjust for confounding. Risk ratios and 95% confidence intervals (RR 95%CI) were estimated using modified Poisson regression models with robust variance estimation.

Results We included 407 exposed births and 520 unexposed births. The risks of MCM in the full cohort, the exposed, and the unexposed were 2.3%, 2.7%, and 1.9%, respectively (unadjusted RR 1.42, 95%CI 0.60–3.28). The IPTW-adjusted population achieved a good balance across patient characteristics. The IPTW-adjusted RR was 1.59 (0.67–3.75). The adjusted risk difference was 0.01 (-0.01–0.03). When the exposure was defined as having ≥ one HCQ dispensation from three months preconception to the end of the 1st trimester, the IPTW-adjusted RR was 1.56 (0.69–3.54).

Conclusions Our findings show an increased, but not statistically significant, risk of MCM at birth among births born to women with lupus exposed to HCQ during the 1st trimester compared to those without HCQ exposure. Future studies are warranted and should utilize a longer follow-up for MCM ascertainment (i.e., within one year of birth). For managing lupus during pregnancy, the benefits of HCQ may still outweigh the risks.