Metabolic Syndrome (MS) is a common disorder combining obesity, dyslipidemia, hypertension, and insulin resistance. Its prevalence among HIV-infected people is still debated. Besides, how antiretroviral therapy and HIV infection per se are related to MS is still unclear. All treatment-naïve patients attending scheduled visits at CISAI group hospitals between January and December 2007 were eligible for the study. Patients without MS at enrolment were followed-up for 3 years or until they developed MS, diagnosed according to the National Cholesterol Education Program (NCEP) definition. The main objective was to assess the 3-years incidence of MS. MS was evaluated for 188 subjects. Out of them, 62 (33.0%) had started HAART at enrolment, whereas 67 (35.6%) more started during the observation. 59 (31.4%) were still treatment-naive at the study end. MS was newly diagnosed in 14 patients. The incidence was 2.60 cases/100 person-years (95% CI 1.47–4.51), 2.75 (1.11–5.72) among HAART-naïve patients and 2.65 (1.23–5.03) in subjects on HAART. Blood pressure did not change in the study period, whereas in naive patients the HDL level significantly lowered (median –6.0 vs. 4.0, P<0.0001) compared to HAART-treated patients. Triglicerides increased significantly in HAART subjects (median 12.0 vs. 1.0, P=0.02), as well as blood glucose (median 6.0 vs. 1.0, P=0.01). In our population, the overall MS incidence was low and largely similar in patients who started HAART or remained naive. However, the feature of MS was different in the two groups, suggesting that in untreated and treated patients MS developed through different metabolic pathways.