Background Early preeclampsia is a serious pregnancy complication characterised by abnormal placentation, diffuse maternal endothelial cell dysfunction, and requiring emergent delivery which may be very premature. SLE has been associated with preeclampsia, but little is known about the risks of early onset preeclampsia – a pregnancy morbidity associated with stroke, placental abruptions, and perinatal death.

Materials and methods SLE was defined as ≥2 visits in the Swedish National Patient Register (NPR, inpatient and outpatient specialist) with ≥1 SLE diagnosis from a specialist who typically treats, manages, or diagnoses SLE in Sweden (2001–2012). General population comparators (non-SLE) were sampled from the Total Population Register. We restricted to singleton births in the Medical Birth Register (MBR). Preeclampsia was defined by first ICD-coded visit during pregnancy in NPR and early-onset defined as <34 weeks. Obesity (BMI > 30), age, smoking, and pregestational hypertension and diabetes were defined using NPR and MBR. NPR ICD-coded visit and/or heparin dispensing during pregnancy from the Prescribed Drug Register (2006–2012) was a proxy for antiphospholipid syndrome (APS). The association between early preeclampsia and SLE was estimated by multivariable-adjusted modified Poisson models for first, subsequent, and all births. Robust standard errors and preeclampsia history were accounted for in non-first-births analyses. We investigated effect modification by pregestational hypertension, examined residual confounding by APS and misclassification of lupus nephritis as preeclampsia.

Results There were 742 births to women with SLE (343 first births) and 10484 births to women from the general population (4443 first births). Among the 32 pregnancies with early preeclampsia and SLE, 75% were first births and 34% were positive for the defined APS proxy. SLE was associated with a significantly increased risk of early preeclampsia for all, first, and subsequent births compared to non-SLE [RR = 7.3, (95% CI: = 4.5, 11.9), all births]. Although adjustment for APS proxy attenuated the association SLE remained statistically significantly associated with early preeclampsia (RR = 3.7, 95% CI: = 1.7, 7.9). Findings were similar among women with no pregestational hypertension, as well as in the absence of recent nephrology care. Risk ratios for early preeclampsia were smaller, but significant, for subsequent births compared to first and all births [RR = 2.8 (95% CI: = 1.2, 6.4) subsequent].

Conclusions Women with SLE are at increased risk of preeclampsia before 34 weeks gestation, and importantly, this increased risk may be independent of pregestational hypertension, APS, obesity, or smoking. Traditional risk factors alone may not explain the increased risk of early preeclampsia among women with SLE for first, subsequent, or any birth. Women with SLE during pregnancy should continue to be monitored carefully for early preeclampsia and future research is needed to identify what non-traditional preeclampsia factors might be causing this serious outcome.