Background: Lipid abnormalities are among the most frequent treatment-limiting adverse events during HAART in HIV-infected individuals. Lipid disturbances have also been associated with hepatitis C virus (HCV) chronic infection in HIV-uninfected participants. HAART-induced lipid abnormalities may then have peculiar features in HIV-HCV co-infected individuals. Purpose: To estimate the prevalence and incidence rates of hypertriglyceridemia and hypercholesterolemia and to identify associated factors in a large clinic population of HIV patients after HAART has been initiated. Design: We performed a retrospective longitudinal follow-up study in a large cohort of HIV patients on their first HAART. Patients and Method: The clinical databases of two major clinical centers in Italy participating in the MASTER study were merged. Treatment-emerging metabolic disorders in patients on their first HAART regimen (PI-based or NNRTI-based) who were stable for at least 4 months were prospectively analyzed by baseline parameters, drug regimens, and viroimmunological outcome of therapy. Follow-up was continued for 24 months or until drug discontinuation, whichever came first. Results: Two hundred and eighty two (282) HIV-infected patients undergoing HAART (203 PI + 79 NNRTI; 65 including stavudine [d4T]) met inclusion criteria and were enrolled in the study from 1997 to 2002. Mean follow-up was 18.5 ± 6.7 months. After HAART had been initiated, a statistically significant mean increase in total cholesterol over time was observed in comparison to baseline (p < .0001), without difference between treatment groups (PI vs. NNRTI, with or without d4T). In the univariate analysis, predictive factors for HAART-induced hypercholesterolemia were baseline total plasma cholesterol and triglycerides values and CD4+ cell count differential increase over time, while a negative correlation was found with zalcitabine-including regimens and baseline HCV seropositivity. At multivariate analysis, only high baseline total plasma cholesterol and triglycerides values retained their predictive value and baseline HCV seropositivity was significantly associated with lower increase in total cholesterol values under HAART, regardless of treatment groups (p < .001). Conclusion: HCV co-infection is an independent factor preventing the emergence of treatment-limiting total cholesterol increase under any HAART regimen, possibly reflecting impaired total cholesterol synthesis in the liver or total cholesterol hypercatabolism. On the contrary, no HCV influence on triglycerides plasma levels was noted. Our data do not suggest any favoring role of specific treatment or drugs (PI and/or d4T) on total cholesterol and triglycerides increase under HAART.