Role of Long-Term Nucleoside-Analogue Therapy in Lipodystrophy and Metabolic Disorders in Human Immunodeficiency Virus--Infected Patients

The role of nucleoside analogues (NAs) in lipodystrophy (LD) syndrome in human immunodeficiency virus (HIV)-infected patients remains controversial. We studied the prevalence of LD in previously untreated patients randomized to receive different NA combinations (in the ALBI-ANRS 070 trial) for 6 months. At month 30 of follow-up, 37 (31%) of 120 patients had >/=1 morphologic change, and 21 (57%) of 37 had isolated peripheral lipoatrophy; corresponding values for the patients who received only NAs throughout follow-up were 20 (30%) of 66 and 14 (67%) of 21, respectively. In multivariate analysis, factors associated with presence of LD at month 30 were initial assignment to the group receiving stavudine and didanosine (odds ratio [OR], 6.7; P=.02), age (OR for being 10 years older, 3.6; P=.002), and HIV RNA level at month 30 (OR, 0.4; P=.007). No difference was observed in cholesterol and glucose levels as a function of any pattern of antiretroviral exposure. Exposure to stavudine and didanosine was associated with LD syndrome (predominantly lipoatrophy).