Objective: To compare body composition, serum lipid profile, parameters of insulin secretion and endocrine measurements in HIV-1-infected patients whose first combination antiretroviral regimen differed only in a nucleoside reverse transcriptase inhibitor (NRTI).
Design and setting: Cross-sectional study in an AIDS clinic of a university hospital.
Patients: One-hundred-and-fifty HIV-infected patients on long-term first highly active antiretroviral therapy including stavudine (n=75) or zidovudine (n=75).
Main outcome measure: Fat wasting was assessed by physical examination. Regional fat distribution was estimated using calliper measurements of skinfold thickness at four sites. Central adiposity was assessed by measurement of waist-hip ratio. Fasting glucose, insulin, triglyceride, cholesterol and its fractions, testosterone, follicle stimulating hormone, luteinizing hormone levels, CD4 cell count and HIV viral load were determined. Daily caloric intake and physical activity level were also calculated.
Results: Total body fat was significantly lower in patients taking stavudine, whereas the lean body mass was not statistically different amongst both groups. Ninety-four patients (62.7%; 95% CI: 54.9-70.4%) had fat redistribution, being isolated lipoatrophy in 20 (13.3%; 95% CI: 7.9-18.8%), isolated lipohypertrophy in 33 (22.0%; 95% CI: 15.4-28.6%) and mixed syndrome in 41 (27.3%; 95% CI: 20.2-34.5%). There were not statistically significant differences between stavudine- and zidovudine-treated patients with respect to the overall prevalence of fat redistribution syndromes (P=0.34). The prevalence of lipoatrophy (OR=1.86; 95% CI: 0.58-6.33, P=0.37), lipohypertrophy (OR=0.65; 95% CI: 0.25-1.69, P=0.45) and mixed syndromes (OR=1.05; 95% CI: 0.43-2.54, P=0.93) was not statistically different in both groups of patients. The only independent predictor for the appearance of mixed syndrome and lipoatrophy was sedentarism (OR=4.418; 95% CI: 1.565-12.472, P=0.005) and (OR=4.515; 95% CI: 1.148-17.761, P=0.03), respectively. Independent predictors of lipohypertrophy were age (OR=1.138; 95% CI: 1.061-1.220, P<0.0001) and prior AIDS (OR=0.305; 95% CI: 0.100-0.931, P=0.04). There were no statistically significant differences between stavudine and zidovudine-based groups with respect to metabolic and hormonal parameters.
Conclusion: The use of stavudine or zidovudine in the context of the first combination antiretroviral therapy is not associated either with an increased likelihood of lipid or gonadal hormones abnormalities, and although there was a trend to a lesser body fat content in the stavudine group, there was no increase in the overall likelihood of fat redistribution syndromes with respect to zidovudine group. Physical activity is a protective factor for the development of fat redistribution syndromes.