Background: The heightened inflammatory profile resulting from both HIV infection and obesity is of increasing importance in many HIV-related comorbidities. Little is known about the association between the change in body mass index (BMI) with antiretroviral (ART) initiation and the change in inflammatory markers, particularly in resource-limited settings.
Methods: AIDS Clinical Trials Group study A5175 was a randomized trial comparing 3 ART regimens in resource-diverse international settings; the following is a country stratified random sub-cohort of 270 subjects; 246 subjects had stored samples. BMI (weight [kg]/height [m]2) was categorized as underweight (UW, <18.5), normal weight (NW, 18.5-24.9), and overweight/obese (OW/OB, ≥25.0). Inflammatory markers were measured (TNF-α, IFN-γ, IL-6, IL-18, IP-10, CRP, sCD14) at weeks 0, 24, 48. Effect of baseline and change in BMI on changes in biomarkers was assessed using random effects models fitted for natural spline at BMI categories and adjusted for age, sex, country, log10 HIV-1 RNA, and treatment arm. A separate model assessed the effect of change to OB BMI (>30 versus ≤30).
Results: Of 246 participants, 50% were female, 53% black, with a median age 35 and CD4 count 179. 37% were assigned to ZDV/3TC+EFV, 33% to ATV+FTC+DDI, and 30% to TDF/FTC+EFV. At week 0, 8% were UW, 65% NW, 27% OW/OB including 7% OB; at week 48, 3% were UW, 60% NW, 37% OW/OB including 9% OB. In multivariate analyses, among baseline UW subjects, an incremental BMI increase was associated with decreased CRP (β -9.32; p=0.001) and trend towards decreased sCD14 (β -0.09; p=0.09). For baseline OW/OB subjects, an increase in BMI was associated with increased sCD14 (β 0.02; p=0.05). No significant associations were detected in the NW group or within other inflammatory markers (p>0.05). In multivariate analyses comparing OB vs not OB participants, OB was associated with an increase in sCD14 (β 0.19; p=0.02) and trend towards higher IL-18 (β 127.7; p=0.056); there were no associations with other markers.
Conclusions: Among HIV-infected persons initiating ART in resource-diverse settings, weight gain among underweight persons may reduce inflammation. In contrast, weight gain among obese persons appeared to heighten inflammation. As sCD14 is a marker of mortality during HIV treatment, the data highlight the potential impact of obesity on treatment outcomes. Further investigation into the impact of obesity on HIV treatment outcomes in resource-limited settings is needed.