Neurocognitive performance and cardiovascular risk in a cohort of HIV-infected patients

Introduction: the use of highly active antiretroviral therapy (HAART) has dramatically reduced the morbidity and mortality related to AIDS. However, HIV infected patients, even if successfully treated, are at increased risk of cardiovascular disease and neurocognitive impairment. Recent evidence suggest a possible relation between cardiovascular disease and neurocognitive impairment in the general population. Purpose of the study: we evaluated the relationship between cardiovascular risk and neurocognitive performance in HIV infected patients. Methods: we enrolled consecutive HIV-infected outpatients. In each patient demographic, clinical and therapeutic characteristics were recorded from clinical records. Anthropometric measures and blood pressure were measured during the study visit. Each patient underwent a complete neurocognitive assessment with a battery of 8 tests: Trail Making Test-A (TMT-A), Trail Making TestB (TMT-B), Digit Span (DSp), immediate (Rey 15) and delayed (D-Rey 15) recall of Rey’s 15, Digit Symbol (DSy), Letter fluency test (Flu), Rey complex figure (R-Fig). We also considered 2 global zscore NPZ-4 (TMA-Z+TMB-Z+DSp-Z+DSy-Z) and NPZ-8 (TMA-Z+TMB-Z+DSp-Z+DSy-Z+15-Z +15dif-Z+Flu-Z+FigR-Z). Cardiovascular risk was estimated with the Framingham risk score (FRS). Results: we enrolled 80 HIV infected patients with a median age of 49 (IQR 41.5-53), 55 (68.7%) were males and 36 (45%) heterosexuals. 22 had a previous AIDS diagnosis and 28 (35%) were HCV-infected. Median CD4 cells count was 809 (IQR 607-1187) cells/mm3, median nadir CD4 was 271 (IQR 147,5-380,5) cells/mm3 and 69 (86.2%) had an undetectable HIV-RNA. According to FRS, 21 (26.2%) patients had a 10-year FRS ≥20% (high risk) whereas 59 (73.8%) had <20% (low/intermediate). Patients with FRS≥20% had a significantly lower level of education, longer duration of known HIV infection, HAART duration and proportion of HCV coinfection when compared to patients with FRS <20%. Patients with high FRS score had also lower values of both NPZ-4 (p=0.052) and NPZ-8 (p=0.018). Among single tests, DSy test z-scores were significantly lower in high FRS patients (p=0002). At univariate analysis, level of education, Rey 15, D-Rey 15, DSy z-scores, NPZ-4 and NPZ-8 were negatively correlated with FRS (p<0.05), whereas HIV infection and HAART duration were positively correlated (p<0.05). At multivariate analysis, after adjusting for HIV-related and demographic variables, NPZ-8 negatively predicted FRS (Beta -0.35, p=0.034). Among single tests included in the neurocognitive examination, DSy z-scores showed the strongest negative correlation (Beta -0.38, p=0.033) with the FRS. Conclusions: We found a high prevalence of patients with elevated cardiovascular risk. Neurocognitive performance was negatively correlated with FRS suggesting a tight correlation between cardiovascular disease and cognitive impairment in HIV-infected patients.