In this cross-sectional study, we evaluate potential predictors of Metabolic Syndrome (MS) in a group of 54 Caucasian chronically HIV-infected patients with lipodystrophy. According to ATP-III criteria, 22 patients were affected by MS and 32 were not. The mean age of the sample was 41.2+/-8.6 years, and most patients were males (74.1%); the two groups were homogeneous for gender, age, viro-immunologic status and the duration of antiviral therapy. The independent association between MS and several factors including demographic characteristics, type of highly-active antiviral therapy (HAART), viro-immunologic response, common cardiovascular risk factors (including Framingham scores), and selected cytokines (IFN-gamma, TNF-alpha, IL-2, IL-4, IL-6, IL-10 and IL-18), was investigated using stepwise forward logistic regression. At multivariate analysis, the only independent predictors of the metabolic syndrome were triglycerides and IL-18. A 10 mg/dL increase in triglycerides corresponds to an adjusted risk ratio for MS of 1.11 (95 percent IC: 1.04-1.19); and patients in the top tertile of IL-18 (those with IL-18 >/= 530 pg/L) had more than three times the likelihood of MS, as compared to the bottom and medium tertiles of IL-18 (patients with IL-18< 530 pg/L). This relationship was not attenuated by the inclusion of any other variable in the multivariate model. However, the association between metabolic syndrome and IL-18 is no longer significant when IL-18 is treated as a continuous variable (trend p = 0.087). Our results on HIV patients with lipodystrophy confirm previous findings on a strong independent association between IL-18 and MS in the general population. Further research is needed to clarify the mechanism of this association and its role in the development of cardiovascular disease in HIV patients.