Abstract Body

Non-pregnant women on nevirapine(NVP)-based antiretroviral therapy (ART) have increased risk of incident hypertension. We evaluated the risk of hypertension in pregnancy by ART regimen in Botswana.

Data were collected from all live and stillbirths <=24 weeks gestational age (GA) at 8 hospitals throughout Botswana (~45% of nationwide births). We recorded maternal demographics, medical history, HIV and ART history, and blood pressures and weights during antenatal care. Women were included if they started ART prior to pregnancy, with either zidovudine/lamivudine (ZDV/3TC) or tenofovir/emtricitabine (TDF/FTC) combined with NVP, lopinavir-ritonavir (LPV/r) or efavirenz (EFV).  Using log binomial regression, we compared risk of any hypertension (SBP >=140 or DBP >=90), severe hypertension (SBP >=160 or DBP >=110), gestational hypertension (onset of hypertension >=20 weeks GA) and early gestational hypertension (onset of gestational hypertension <34 weeks GA) by maternal ART regimen at conception, adjusting for potential confounders (maternal age, occupation, parity, weight and alcohol/tobacco use).

From Aug 2014-Aug 2016, 5087 women on NVP, LPV-r or EFV-based ART from conception delivered singletons at a surveillance maternity.  Of these, 4915 (97%) had at least 1 blood pressure recorded in pregnancy including 2128 (43%) on NVP-based ART and 2797 (57%) on non-NVP based ART.  Overall 1090 (22%) had hypertension, 106 (2.2%) had severe hypertension, 691 (14%) had gestational hypertension and 438 (9%) had early gestational hypertension. In adjusted analyses, women on NVP-based regimens were more likely to have hypertension (30% vs. 16%), severe hypertension (3.3% vs. 1.2%), gestational hypertension (18% vs.10%) and early gestational hypertension (12% vs. 7%) compared with women on non-NVP based ART (Table 1).  There was no difference in outcomes when NVP was combined with ZDV/3TC vs. TDF/FTC except for severe hypertension which was more common in ZDV/3TC/NVP (6% vs. 2%). There were 48 stillbirths (7.8%) among hypertensive women on NVP, 25 stillbirths (5.6%) among hypertensive women on non-NVP containing ART, and 88 (2.3%) stillbirths among non-hypertensive women. Although hypertensive women on NVP accounted for only 13% of the population, they had 30% of the stillbirths.

HIV-infected women on NVP-based ART have increased risk of developing gestational, severe and early hypertension in pregnancy, which may explain their increased risk for stillbirth.