Hydroxychloroquine in Lupus Pregnancy and Risk of Preeclampsia

Abstract

Objective: Systemic lupus erythematosus (SLE) disproportionately affects women during childbearing years, and hydroxychloroquine (HCQ) is the standard first-line treatment. Preeclampsia complicates up to one-third of pregnancies in lupus patients, although reports vary by parity and multifetal gestation. We investigated whether taking HCQ early in pregnancy may reduce the risk of preeclampsia. Methods: We studied 1,068 live birth singleton pregnancies among 1,020 privately insured patients with SLE (2007–2016). HCQ treatment was defined as three months preconception through the first trimester, and prescription fills were a proxy for taking HCQ. Modified Poisson regression estimated risk ratios (RRs) and 95% confidence intervals (CIs), stratified by parity. Propensity scores accounted for confounders, and stratified analyses examined effect modification. Results: Approximately 15% of pregnant patients were diagnosed with preeclampsia. In 52% of pregnancies, patients had one or more HCQ fills. Pregnant patients exposed to HCQ had more comorbidities, SLE activity, and azathioprine treatment. We found no evidence of a statistical association between HCQ and preeclampsia among nulliparous (RR 1.26 [95% CI 0.82–1.93]) and multiparous pregnancies (RR 1.20 [95% CI 0.80–1.70]). Additional controls for confounding decreased the RRs toward the null (nulliparous pregnancy, propensity score–adjusted [PS-adj] RR 1.09 [95% CI 0.68–1.76]; multiparous pregnancy, PS-adj RR 1.01 [95% CI 0.66–1.53]). Conclusion: Using a large insurance-based database, we did not observe a decreased risk of preeclampsia associated with HCQ treatment in pregnancy, although we cannot rule out residual and unmeasured confounding and misclassification. Further studies leveraging large population-based data and prospective collection could characterize how HCQ influences preeclampsia risk in pregnant patients with SLE and among persons at greater risk of hypertensive disorders of pregnancy.