Background/Purpose: Women with systemic lupus erythematosus (SLE) are at greater risk of preterm delivery compared to women without lupus. A significant proportion of SLE pregnancies are complicated by maternal factors and delivery is medically indicated (iatrogenic). For example, preeclampsia risk is increased in SLE pregnancies. We investigated the types of preterm deliveries in SLE (iatrogenic vs spontaneous), and estimated the proportion mediated by preeclampsia.

Methods: Women with and without SLE with singleton pregnancies were identified from SLINK, a Swedish register-based cohort, 2001-2013.Maternal SLE was defined as ≥2 SLE-coded discharge diagnoses from the National Patient Register and ≥1 SLE code from an appropriate specialist. Preterm delivery was defined as <37 weeks gestation, extremely and very preterm as <32 weeks. Spontaneous preterm birth was defined as due to preterm contractions or premature rupture of the membranes (PPROM). Iatrogenic was defined as planned C-section or induced labor. Maternal/fetal complications were stratified by maternal SLE, parity, and gestational age (<32 weeks vs 32-37 weeks). Using Drug Register data (2006-2013), modified Poisson models estimated risk ratios of preterm delivery by maternal hydroxychloroquine (HCQ) use during pregnancy. Mediation analysis estimated the proportion of iatrogenic deliveries mediated via preeclampsia, using Poisson models accounting for maternal age/smoking. Autocorrelation of multiple births per women were accounted for using robust errors.

Results: There were 781 SLE-exposed and 11271 non-SLE-exposed pregnancies. Preterm delivery was more common in the SLE than non-SLE group for first (22% vs 6%) and subsequent (15% vs 4%) pregnancies. 27% of SLE-exposed first births delivered extremely or very preterm, 90% were iatrogenic (vs 47% of non-SLE first births). Among first and subsequent births, preterm deliveries were more frequently complicated by preeclampsia in SLE women. Preterm delivery was more common among those without HCQ use during pregnancy (RR=0.75, 95% CI (0.49 to 1.14)), but power was limited. For SLE pregnancies, preeclampsia mediated 12% of iatrogenic preterm births – as compared to spontaneous preterm births – among first pregnancies, and 11% among subsequent pregnancies.

Conclusion: Preterm delivery complicates a greater proportion of SLE compared to general population pregnancies, and a considerable proportion of risk is mediated through preeclampsia. This study is the largest to date to partition preterm delivery into spontaneous and iatrogenic, and the first to estimate the proportion of preterm birth mediated via preeclampsia. Our findings reinforce the importance of close monitoring for preeclampsia in lupus pregnancy, and suggest larger well-powered studies investigating the potential protective effect of HCQ on preterm delivery.

« Back to 2017 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/preterm-delivery-phenotypes-in-sle-pregnancies/