Prevalence and predictors of proteinuria in HIV-infected and uninfected pregnant women in Cameroon

Abstract

Proteinuria during pregnancy has been associated with increased pregnancy complications. Furthermore, even low-grade proteinuria has been associated with increased mortality in the general population and in non-pregnant HIV-infected women. Urine dipstick protein was measured prospectively on HIV-infected and trace protein or more and quantified by urine protein:creatinine measurement (P:C). Logistic regression modeling was used to identify factors associated with proteinuria. About 199 human immunodeficiency virus (HIV)-infected and 190 HIV-uninfected normotensive pregnant women were evaluated. The median age was 27 years in both groups and 37% presented in the third trimester. Among HIV-infected women, median CD4 cell count was 417 cells/mm(3); 27% were on combination antiretroviral therapy (cART). Proteinuria was present in 39.2% of HIV-infected and 20.9% of uninfected women (P < 0.001). HIV infection was independently associated with proteinuria [adjusted odds ratio (OR) = 2.45; confidence interval (CI) = 1.56-3.85]. Among HIV-infected pregnant women, cART was protective (adjusted OR = 0.39; CI = 0.19-0.82). Results were qualitatively similar when urine P:C was evaluated as a continuous outcome variable. The prevalence of low-grade proteinuria in both HIV-infected and -uninfected Cameroonian pregnant women is high. HIV-infected pregnant women are at increased risk for proteinuria, and cART appears to exert a protective effect. Further studies are needed to elucidate the causes of increased proteinuria in African pregnant women, both HIV-infected and -uninfected.