Background: Changes in body fat distribution are an adverse effect of therapy with HIV protease inhibitors (PI). It has been suggested that nucleoside analogue reverse transcriptase inhibitors (NRTI) may also contribute to this so-called lipodystrophy syndrome, but the relative contribution of the two drug classes is unclear as they are usually administered concomitantly. Method: The occurrence of lipodystrophy, as reported by physicians using no standardized criteria, was followed in patients randomly assigned to treatment with either a PI alone or a PI combined with an NRTI. The patients were part of a multicenter, open-label, randomized comparison of ritonavir (RTV)/saquinavir (SQV) with or without the addition of stavudine (d4T) in HIV-1-infected patients without prior PI and d4T experience (the Prometheus study). Results: Lipodystrophy was reported in 29 of 175 (17%) patients during 96 weeks of follow up. Overall, it was reported significantly more frequently in patients who were randomized to RTV/SQV/d4T (22/88; 25%), than in patients randomized to RTV/SQV alone (7/87; 8%) (P = 0.003). When the analysis was limited to patients without any prior antiretroviral experience, lipodystrophy likewise was significantly more frequent in patients randomized to RTV/SQV/d4T (12/50; 24%) than in those randomized to RTV/SQV (2/44; 5%) (P = 0.008). Conclusion: This randomized clinical trial, in spite of not having been blinded, supports a contributory role of NRTI in the development of antiretroviral therapy-associated lipodystrophy. The low incidence of lipodystrophy in patients with no or limited NRTI exposure supports further evaluation of NRTI-sparing regimens as alternatives to current antiretroviral regimens.