Effects of discontinuing stavudine or protease inhibitor therapy on human immunodeficiency virus-related fat redistribution evaluated by dual-energy x-ray absorptiometry.

STUDY OBJECTIVE

To determine whether discontinuation of stavudine or protease inhibitor therapy improves human immunodeficiency virus (HIV)-related fat distribution in men.

DESIGN

Observational, retrospective study consisting of a cross-sectional (part 1) and a longitudinal (part 2) study.

DATA SOURCE

Medical records from Purpan University Hospital and La Grave University Hospital, Toulouse, France. Subjects. Eighty men with HIV infection treated with antiretrovirals and 151 healthy male controls matched for age.

MEASUREMENTS AND MAIN RESULTS

In part 1, body composition and fat distribution of the HIV-infected men were compared by dual energy x-ray absorptiometry (DEXA) with those of the controls to determine whether body fat distribution is altered in HIV-infected men. In part 2, we analyzed modifications of body composition and fat distribution in 45 of the 80 patients. These 45 had been exposed to antiretroviral drugs, including stavudine and a protease inhibitor, for at least 5 months before the first of two DEXA assessments. They received three different treatment strategies for several months. In group 1, stavudine was withdrawn; in group 2, protease inhibitor was discontinued, and in group 3, stavudine plus protease inhibitor were continued. Group 1 showed a significant fat gain in the lower extremities 31.7 +/- 5.9 months after stavudine discontinuation (p<0.0001). Group 2 did not show any significant modification of total body, lower limb, or trunk fat despite protease inhibitor discontinuation for 35.2 +/- 6.6 months. Findings were similar for group 3, who continued receiving stavudine-protease inhibitor therapy for 21.2 +/- 12.8 months.

CONCLUSION

These data suggest that long-term withdrawal of stavudine from the antiretroviral therapy regimen may be associated with significant improvement in lipoatrophy in the lower extremities, whereas long-term protease inhibitor withdrawal did not modify fat distribution.