Association between preeclampsia and HIV: a case-control study in urban South Africa

BACKGROUND: Preeclampsia is a considerable cause of maternal and infant morbidity and mortality. Although its etiology is unknown, preeclampsia has been described as a state of exaggerated maternal inflammatory response. Therefore, it has been hypothesized that preeclampsia would occur less commonly in states of immune deficiency. OBJECTIVE: This study aimed to compare the prevalence of treated and untreated HIV infections among preeclamptic cases and controls, determine infant outcomes, and evaluate the association between HIV and preeclampsia after adjusting for known predictor variables, including maternal age, gravidity, body mass index, and smoking. STUDY DESIGN: This case-control study investigated the association between preeclampsia and HIV infection using secondary data from an unrelated study. We defined preeclamptic cases as pregnant women who were normotensive until 20 weeks of gestation and thereafter had at least 1 high blood pressure measurement either before or at delivery and proteinuria, defined as protein excretion of ≥300 mg within 24 hours or >2 protein on dipstick urinalysis. The prevalence of HIV infection was compared between cases and controls. Multivariate logistic regression analysis was used to assess the association between preeclampsia and potential confounding variables and reported using odds ratios and 95% confidence intervals. RESULTS: There were 571 cases with preeclampsia and 596 normotensive controls included in this study. The median age was 27 years for cases and 26 years for controls (P=.008). Most participants (69%) had ≥2 previous pregnancies with no difference between the cases and controls (P=.176). Overall, 43% of the participants were obese, with a mean body mass index of 29 (interquartile range, 24.5–34.2), with higher proportions of women who were overweight and obese in the group with preeclampsia (P=.031). The prevalence of HIV was significantly lower in cases than in controls (24% vs 30%, respectively; P=.014). Compared with 16% of infants born preterm to normotensive controls, 48% of infants were born preterm born to women with preeclampsia (P<.001). Compared with 14% of infants born with low birthweight to normotensive controls, 53% of infants were born with low birthweight to women with preeclampsia (P<.0001). Untreated HIV infection was negatively associated with preeclampsia (unadjusted odds ratio, 0.330; 95% confidence interval, 0.197–0.552; P<.0001), whereas factors associated with preeclampsia were advanced maternal age (odds ratio, 1.673; 95% confidence interval, 1.209–2.316; P=.002) and obesity (odds ratio, 1.611; 95% confidence interval, 1.023–2.537; P=.040). After adjusting for maternal age, gravidity, smoking, and body mass index in the multivariate regression, only obesity remained significantly associated with preeclampsia (adjusted odds ratio, 1.624; 95% confidence interval, 1.024–2.575; P=.039). CONCLUSION: Before the large-scale rollout of antiretroviral therapy in a setting with a high burden of HIV and preeclampsia, untreated HIV infection was found to have a protective effect against preeclampsia. The protective effect against preeclampsia was not apparent for HIV infection treated with antiretroviral therapy.