854    

Background:  It remains unclear whether rosiglitazone has a beneficial effect on HIV lipodystrophy. We aimed to evaluate whether treatment with rosiglitazone would slow peripheral fat loss in patients with established HIV lipodystrophy, treated with protease inhibitors (PI).

Methods:  HIV⁺ patients treated on stable, PI-containing HAART and affected by HIV lipodystrophy were randomized to receive rosiglitazone 4 mg by mouth once daily or matching placebo in a prospective, double blind trial. The primary endpoint was the median percentage decrease in arm-fat by DEXA over a 24-week period. Patients with diabetes were excluded. Clinical and anthropometric evaluations, fasting lipid parameters, oral glucose tolerance testing, and DEXA scanning were performed at baseline and week 24. HOMA, QUICKI, and McAuley indices of insulin resistance were calculated from baseline oral glucose tolerance testing results. Median changes in DEXA fat mass, lipid profile, and insulin resistance measures were compared using the Wilcoxon rank sum test.

Results:  Of the 96 enrolled patients, 73 were evaluable with 24 weeks of follow-up data available. Mean age was 47.2 years, 98.6% were male, median CD4 cell count was 430/mm³ (361 to 620), median viral load was 1.70 log₁₀ copies/mL (74.0% had a viral load of  ≤ 50 copies/mL). Median body mass index was 24.2 kg/m² (22.3 to 26.4); median total cholesterol 5.52 mmol/L (4.71 to 6.57). Median baseline arm-fat percentage was 12.4% (10.1 to 17.2%), leg fat 8.6% (7.6 to 11.6%), trunk fat 17.5% (13.7 to 22.0%), total body fat 14.5% (12.2 to 18.3%). The median percentage change in arm fat, leg fat, trunk fat, and total body fat at 24 weeks was not significantly different between the rosiglitazone and the placebo groups (7.7% vs 5.2%, p = 0.87), (2.0% vs  –2.4%, p = 0.78), (2.0% vs  –1.7%, p = 0.50), and (2.1% vs 0.2%, p = 0.53), respectively. Median percentage changes at 24 weeks in cholesterol, triglycerides, HDL, and oral glucose tolerance testing responses, as well as 24-week changes in HOMA, QUICKI, and McAuley indices were not significantly different between groups. Five subjects were started on lipid-lowering therapy during the study. Four patients had grade3 or 4 adverse events (2 with elevation in aspartate aminotransferase and 2 with elevations in triglycerides).

Conclusions:  This randomized placebo-controlled trial failed to show a benefit of rosiglitazone (4 mg once daily) on HIV lipodystrophy over 24 weeks follow-up. No significant differences in lipid profiles or measures of insulin resistance were found between treatment and control groups. Adverse effects were uncommon.

Keywords: HIV lipodystrophy; rosiglitazone; fat redistribution