Polo and colleagues examined the effect of switching nucleoside reverse transcriptase inhibitors (NRTIs) in lipodystrophy patients receiving 2 NRTIs + protease inhibitor (PI). One hundred-fifty patients received either AZT + 3TC + PI (n = 46), d4T + 3TC + PI (n = 87), or d4T + ddI + PI (n = 17). After approximately 16-18 weeks of follow up, the d4T + ddI + PI arm and the d4T + 3TC + PI arm had a much higher incidence of lipodystrophy compared to the AZT + 3TC + PI arm (23.5 %, 26.4%, and no cases, respectively). Ten patients receiving d4T + ddI + PI had their NRTIs switched to AZT + 3TC while maintaining IDV; after 6 months, the patients showed decreases in triglycerides, increases in fat mass, increases in CD4+ cells, and lower viral load. Based on the data, the researchers concluded that certain NRTIs, specifically d4T and ddI, may be associated with lipodystrophy.

The study has some serious limitations, though. The sampling size differences between the arms suggests that the patients were highly selected. Additionally, more than one quarter of the d4T + 3TC arm and two thirds of the d4T + ddI arm presented at baseline with lipodystrophy -- the study endpoint. Furthermore, the follow up period was short. Antiretroviral history was not reported and the authors did not control or adjust for disease progression.