Neurocognitive Impairment in Well-Controlled HIV-Infected Patients: A Cross-Sectional Study.

Abstract

The reported prevalence of HIV-associated neurocognitive disorders in HIV people depends on the population studied and the methodology used. We analyze the prevalence of neurocognitive impairment (NCI) and associated factors in patients on successful antiretroviral therapy (ART), without comorbidities. Cross-sectional observational study in HIV subjects, ≥18 years old, on stable ART, and HIV viral load of <50 copies/mL. Patients with medical or psychiatric comorbidities and substance abuse were excluded. NCI was diagnosed using Frascati criteria, examining seven neurocognitive domains (NDs). We analyzed the association between NCI and HIV-related clinical variables, carotid intima-media thickness, bacterial translocation, and plasma inflammatory biomarkers [soluble CD14, interleukin-6 (IL-6), and tumor necrosis factor-α]. The prevalence of NCI was calculated with a 95% confidence interval (CI). We fitted a logistic regression model to assess the strength of the associations. Eighty-four patients were included with an observed NCI prevalence of 29.8% (95% CI: 21.0-40.2): 19% had asymptomatic NCI, 8.3% had mild neurocognitive disorder, and 2.4% had HIV-associated dementia. Delayed recall was the most commonly affected ND (27.4%). People diagnosed at least 10 years ago (odds ratio [OR]: 6.5, 95% CI: 1.6-21.7) and those with IL-6 levels above 1.8 pg/mL (OR: 6.0, 95% CI: 1.1-31.3) showed higher odds of NCI in adjusted analyses. Participants with carotid plaques had lower scores for delayed recall: -0.9 ± 1.1 versus -0.2 ± 1.1 (p = .04). Prevalence of NCI is high in otherwise healthy adults with HIV-infection. In this population, more than 10 years since HIV diagnosis and high IL-6 levels are associated with NCI. Delayed recall ND is worse in patients with subclinical atherosclerosis.