Prospective evidence that HIV lipoatrophy and visceral adiposity are partially independent processes

Correspondence: Handan Wand national Centre in HiV Epidemiology and Clinical research, University of new South Wales, Sydney, nSW 2052, Australia Tel +61 2 9385 0900 Fax +61 2 9385 0910 Email hwand@nchecr.unsw.edu.au Purpose: To investigate the patterns of change in objectively assessed body composition parameters and to determine to what extent the observed patterns correlate with modifiable variables and potential risk factors for lipodystrophy in human immunodeficiency virus (HIV)-infected lipoatrophic adults. Method: Changes from baseline in limb fat and visceral adipose tissue (VAT), and their associations with antiretroviral therapy, body composition, and metabolic variables were investigated using linear and logistic regression models. Results: Increases in limb fat were significantly associated with higher baseline limb fat (P , 0.0001), VAT (P = 0.023), and change from baseline to week 72 in VAT (P , 0.0001). On-study use of zidovudine or stavudine was negatively associated with a limb fat increase (P = 0.017). High baseline limb fat mass and VAT had negative effects on subsequent VAT increases at week 72 (P = 0.016 and P = 0.001, respectively). Conclusions: This large, prospective study in HIV-infected adults with moderate or severe lipoatrophy at baseline showed positive associations between changes in limb fat and VAT over 72 weeks. Risk factors for these two lipodystrophic features were different. Our findings suggest that lipoatrophy and fat accumulation are at least partially independent processes.