Antiretroviral treatment patterns and incident HIV-associated morphologic and lipid abnormalities in a population-based chort.

This study provides population-based estimates of the incidence of constituent symptoms associated with HIV-related lipodystrophy syndrome. Possible predictors of symptomatology based on analysis of accrued cases are provided after adjustment for a broad range of personal, clinical, and treatment characteristics. Patients enrolled in a province-wide HIV/AIDS treatment program reported annually on the occurrence of lipoatrophy, lipohypertrophy, and elevated triglyceride and cholesterol levels. Of 1261 individuals who provided baseline data, 745 were available at follow-up, among whom incidence was 27% for lipoatrophy, 21% for lipohypertrophy, and 10% and 16% for increased triglyceride and cholesterol levels, respectively. In logistic multivariate modeling, incident lipoatrophy was associated with duration of stavudine (per quarter) (adjusted odds ratio [AOR] 1.18; 95% confidence interval [CI] 1.09-1.27) and having been diagnosed with AIDS (AOR 2.07; 95% CI 1.20-3.56). Lipohypertrophy risk increased with use of protease inhibitor (AOR 3.53; 95% CI 1.81-6.86) and stavudine (AOR 3.67; 95% CI 1.61-8.38). Incident cholesterol or triglyceride abnormalities were associated with protease inhibitor use (AOR 7.17; 95% CI 2.46-20.96) and duration of ritonavir (per quarter) (AOR 1.12; 95% CI 1.04-1.21). Our findings suggest high annual rates of incidence and a role of first line antiretroviral therapies in symptom development. These outcomes, in conjunction with the findings of others have important implications for evolving treatment patterns.