Hydroxychloroquine and Preeclampsia in a Diverse Cohort of Women with SLE.

OBJECTIVE: Patients with systemic lupus erythematosus (SLE) are at risk for pregnancy complications such as preeclampsia and eclampsia. These clinically important complications are associated with maternal morbidity, mortality and postpartum cardiovascular disease. Some studies suggest that hydroxychloroquine (HCQ) may reduce preeclampsia risk in lupus pregnancy. Using a cohort of pregnancies in prevalent SLE patients at Kaiser Permanente Northern California (KPNC), we investigated whether HCQ use in early pregnancy reduced the risk of preeclampsia/eclampsia. METHODS: Among SLE pregnancies from 2011-2020, we assessed HCQ use from three months before pregnancy through the first trimester. HCQ exposure was defined multiple ways to account for adherence and duration of use. Propensity scores accounted for multiple confounders and modified Poisson models estimated risk ratios (RR) and 95% confidence intervals of the association between HCQ and preeclampsia/eclampsia. Effect modification by pregestational hypertension, history of nephritis, and antiphospholipid antibody status was investigated through stratified analysis. RESULTS: There were 399 pregnancies among 324 patients with SLE at KPNC between 2011 and 2020. Considering multiple exposure definitions, we consistently found a null association between HCQ and preeclampsia/eclampsia. The RRs were consistently lower among the nullipara pregnancies, and RRs were consistently protective but not statistically significant among the high-risk subgroup of those with history of nephritis, aPL positivity, or pregestational hypertension (both nullipara and multipara). DISCUSSION: Although this study found no reduced risk of HCQ on preeclampsia/eclampsia, residual confounding may be attenuating the effect despite an integrated health care delivery system setting with detailed clinical data.

Pregnancy Outcomes in a Diverse US Lupus Cohort.

OBJECTIVE: Although the population of patients with systemic lupus erythematosus (SLE) is racially and ethnically diverse, many study populations are homogeneous. Further, data are often lacking on critical factors, such as antiphospholipid antibodies (aPLs). We investigated live birth rates in patients with SLE at Kaiser Permanente Northern California, including race and ethnicity and aPL data. METHODS: Electronic health records of pregnancies with outcomes observed from 2011 to 2020 were identified among patients with SLE. Prevalent SLE was defined as two or more International Classification of Diseases-coded visits seven or more days apart before the last menstrual period. We summarized patient characteristics, medication orders, health care use, and medication use. Pregnancy outcomes (live birth, stillbirth, spontaneous abortion, ectopic pregnancy, and molar pregnancy) were presented overall and stratified by race and ethnicity, aPL status, and nephritis history. RESULTS: We identified 657 pregnancies among 453 patients with SLE. The cohort was diverse, reflecting the Northern California population (27% Asian, 26% Hispanic, 26% Non-Hispanic White, 13% Non-Hispanic Black, 5% multiracial, and approximately 2% Pacific Islander and Native American). Approximately 74% of observed pregnancies ended in live birth, 23% resulted in spontaneous abortion, 2% were ectopic or molar pregnancies, and <1% were stillbirths. There was limited variability in live births by race and ethnic group (72%-79%), aPL status (69.5%-77%), and nephritis history (71%-75%). CONCLUSION: Our findings are consistent with previous studies; however, some methodologic differences may yield a range of live birth rates. We found that approximately 74% of pregnancies in patients with SLE ended in live birth, with modest variability in spontaneous abortion by race and ethnicity, nephritis history, and aPL status.

Reconciling Between Medication Orders and Medication Fills for Lupus in Pregnancy.

OBJECTIVE: Most studies consider either medications ordered or filled, but not both. Medication underuse based on filling data cannot necessarily be ascribed to patient nonadherence. Using both data sources, we quantified primary medication adherence in a cohort of prevalent systemic lupus erythematosus (SLE) pregnancies. METHODS: We identified 419 pregnancies in Kaiser Permanente Northern California in patients with prevalent SLE from 2011 to 2020. We calculated the number of physician-initiated orders or pharmacy-initiated reorders during pregnancy and a comparable 9-month window the year before (prepregnancy) and the proportion of orders ever filled and filled within 30 days for hydroxychloroquine (HCQ), azathioprine, and corticosteroids. For pregnancies without an order or reorder, we identified the proportion with previous prescription fills overlapping into the respective study period. RESULTS: New orders for lupus medications were usually filled. HCQ was prescribed most often (45.8% pregnancies) and usually filled (89.7% in prepregnancy, 93.2% during pregnancy). The majority filled within 30 days (80.5% prepregnancy, 83.3% pregnancy). Some pregnancies without new HCQ orders had continuous refills from prior orders; 53% of 2011-2015 pregnancies either had a new order or fill coverage from a previous period, compared to 63.2% of pregnancies delivering in 2016-2019. Corticosteroid fill frequencies were 90.6% in prepregnancy and 83.6% during pregnancy. Fewer patients used azathioprine; however, most new orders were filled (94.3% prepregnancy, 91.7% pregnancy). For azathioprine and corticosteroids, fill rates were modestly higher in prepregnancy compared to pregnancy. CONCLUSION: We observed that patients have high adherence to filling new orders for lupus medications, such as HCQ and azathioprine, in pregnancy.