Higher Baseline Viral Load and Black Race Predict Severe Weight Gain on HIV Treatment

Among people starting their first antiretroviral therapy (ART) in ACTG A5257, more advanced HIV disease and black race independently predicted severe weight gain through 96 weeks of follow-up. Severe weight gain risk was lower with atazanavir (Reyataz) or darunavir (Prezista) than with raltegravir (Isentress).

Weight gain may follow ART initiation and heighten risk of cardiovascular disease or diabetes. AIDS Clinical Trials Group (ACTG) investigators conducted this retrospective study to identify predictors of severe weight gain in people randomized to atazanavir/ritonavir (Norvir), darunavir/ritonavir or raltegravir in study A5257. They defined severe weight gain as a 10% or greater increase in weight through 96 weeks of treatment. Severe body mass index (BMI) gain meant an increase of one or more BMI categories through 96 weeks.

ACTG A5257 had 1,809 participants, 24% women, 42% black, 34% white and 22% Hispanic. Baseline age averaged 37 years, and pretreatment CD4 count averaged 310 cells/mm3. When the trial began, BMI indicated that 3% of participants were underweight, 47% normal weight, 31% overweight and 19% obese.

Related: Online Weight Loss Program Helps Overweight and Obese Individuals With HIV Lose Weight

Through 96 weeks of follow-up in participants with weight change data, weight rose by an average 3.8 kg and BMI by an average 1.3 kg/m2. While 373 participants (23.3%) had severe weight gain, 361 (22.7%) had severe BMI gain. Logistic regression analysis adjusted for potential confounders identified two independent predictors of severe weight gain, at the following adjusted odds ratios (aOR) and 95% confidence intervals (CI): every 10-fold higher baseline viral load (aOR 2.52, 95% CI 2.00 to 3.16, P < .0001) and black versus white race (aOR 1.55, 95% CI 1.10 to 2.20, P = .0129).

Three factors predicted lower odds of severe weight gain: every 100-cell higher baseline CD4 count (aOR 0.78, 95% CI 0.72 to 0.85, P < .0001), starting atazanavir/ritonavir versus raltegravir (aOR 0.72, 95% CI 0.53 to 0.99, P = .0427) and starting darunavir/ritonavir versus raltegravir (aOR 0.74, 95% CI 0.54 to 1.01, not significant at P = .0555). The protective effect of darunavir became significant in an analysis of severe BMI gain (aOR 0.73, 95% CI 0.53 to 0.99, P = .0414), but the protective effect of atazanavir/ritonavir lost significance. The other three variables significant in the severe weight gain analysis remained significant in the severe BMI gain analysis.

Receiver operating characteristic (ROC) curve analysis indicated that multivariable models for both severe weight gain and severe BMI gain had reasonable predictive accuracy: area under the curve (AUC) 0.7758 for severe weight gain, AUC 0.7236 for severe BMI gain.

The ACTG investigators suggest the link between worse pre-ART HIV disease severity and greater weight gain on ART may mean early treatment can prevent excess weight gain. The association between black race and weight gains underlines "a need for additional focus on investigating such complications" in minority groups.