Redistribution of body fat in HIV-infected women undergoing combined antiretroviral therapy

Objectives: To investigate the prevalence, metabolic features and risk fact ors of a particular pattern of fat redistribution (FR), characterized by a progressive enlargement of breast and abdominal girth associated with a was ting of the lower limbs, observed in HIV-infected women treated with combin ed antiretroviral (ARV) therapy. Design: Cross-sectional study. Setting: Outpatients attending the Institute of Infectious Diseases, Univer sity of Milan, Milan, Italy. Patients and methods: HIV-infected women treated with two or more ARV drugs , observed between December 1997 and February 1998. FR was confirmed by mea ns of a physical examination and dual-energy X-ray absorptiometry (DEXA). T he metabolic and endocrinological measurements in patients with FR were com pared with those in FR-free women. Results: FR was observed in 32 out of 306 women (10.5%). DEXA revealed more trunk fat (P < 0.01) and less leg fat (P < 0.001) in the patients with FR than in the matched controls. There were no significant differences in labo ratory test results between the two groups. All of the FR patients were tak ing lamivudine-containing regimens; 20 of them were also taking a protease inhibitor (PI). The association of FR with lamivudine-including regimens wa s statistically significant (P = 0.017). Among the patients taking lamivudi ne, the risk associated with treatments including PI was 1.8 (95% CI 0.8-3. 8, P = 0.12). A total duration of ARV therapy of more than 1000 days was as sociated with a greater risk of developing FR (OR 10.8; 95% CI 1.4-80.5; P = 0.0207). Stepwise logistic regression analyses indicated that prolonged A RV therapy and a viral load of more than 10 000 copies per ml at the beginn ing of the last ARV regimen were the only variables that significantly and independently correlated with the risk of FR. Conclusions: The observed body modifications are caused by a redistribution of body fat without fat loss that is apparently not associated with hyperl ipidemia, altered glucose metabolism or other endocrinological disorders. T he development of FR in patients receiving only reverse transcriptase (RT) inhibitors suggests the presence of a PI-independent mechanism that deserve s further investigation. (C) 1999 Lippincott Williams & Wilkins.