Abstract
BACKGROUND: Previous studies of HIV-infected patients have shown significant associations between highly active antiretroviral therapy (HAART) and increased blood pressure; however, the mechanisms involved are less clear. Therefore, we sought to investigate the potential impact of body fat changes in mediating the effects of HAART on blood pressure changes among people living with HIV. METHODS: Four hundred six consenting patients (≥18 years of age) attending a tertiary HIV clinic in semi-urban Nigeria were recruited between August and November 2014 as part of a cross-sectional study. We performed bias-corrected bootstrap tests of mediation using 95 % confidence intervals (CI) to determine the mediating effects of body mass index and waist circumference (mediators) on the total effects of HAART exposure (primary predictor) on blood pressure (outcome), while controlling for age, sex and other potential confounders. RESULTS: Waist circumference remained a significant partial mediator of the total effects of HAART exposure on increasing systolic blood pressure (coefficient: 1.01, 95 % CI: 0.33 to 2.52, 11 % mediated) and diastolic blood pressure (coefficient: 0.68, 95 % CI: 0.26 to 1.89, 9 % mediated) after adjusting for age, sex, smoking status, CD4 count and duration of HIV infection. No significant mediating effect was observed with body mass index alone or in combination with waist circumference after adjusting for all potential confounders. CONCLUSION: Waist circumference significantly mediates the effects of HAART on blood pressure in persons living with HIV, independent of the role of traditional risk factors. The use of waist circumference as a complementary body fat measure to body mass index may improve the clinical prediction of hypertension in HIV-infected patients on antiretroviral therapy.