Recent observational studies suggest a role for lipopolysaccharide (LPS) as a marker of immune activation in HIV-infected patients, with potential repercussions on the effectiveness of antiretroviral regimens.

Objective: To assess changes over time in neuropsychological test results (NPr) and risk factors among a regularly followed HIV-infected patient population. Methods: Prospective cohort of HIV-infected patients randomly selected to undergo neuropsychological follow-up. Test score was adjusted for age, sex and education. Patients were divided into five groups: normal tests, neuropsychological deficit (one impaired cognitive domain), asymptomatic neurocognitive disorders (ANIs), mild neurocognitive disorders (MNDs) and HIV-associated dementia (HAD). Demographic and background parameters including CSF drug concentration penetration effectiveness (CPE) score 2010 were recorded. Changes in NPr and associated risk factors were analyzed. Results: Two hundred and fifty-six patients underwent neuropsychological tests and 96 accepted follow-up approximately 2 years later. The groups were comparable. Upon neuropsychological retesting, six patients improved, 31 worsened and 59 were stable. The proportion of patients with HIV-associated neurocognitive disorders (HANDs) rose from 26 to 45%, with ANIs and MNDs still mostly represented. Most patients initially diagnosed with HANDs remained stable, five of 25 showed clinical improvement and three of 25 deteriorated. Of 33 patients with normal tests, four deteriorated, whereas 24 of 38 with initial neuropsychological deficit had poorer NPr, and contributed most of the new HAND cases. Patients with clinical deterioration had a lower CPE score both at inclusion (6.9 vs. 8.1; P = 0.005) and at the end of follow-up (7.2 vs. 7.8; P = 0.08) than those with improved or stable performance. This was confirmed by multivariate analysis. Conclusion: Patients with higher CPE scores upon inclusion and at the end of follow-up were at lower risk of clinical worsening, suggesting that combination antiretroviral therapy with better CSF penetration could protect against cognitive deterioration.

Objectives: Inversion of the CD4:CD8 ratio is a marker of immune activation and age‐associated disease. We measured the CD4:CD8 ratio as a marker of cognitive impairment in HIV‐infected patients and explored differences according to clinical severity.Methods: Post hoc analysis of data from two prospective cohorts of HIV‐infected patients randomly selected to undergo neuropsychological tests was performed. Test scores were adjusted for age, gender and education. Inclusion criteria were undetectable viral load and stable treatment for at least 6 months. Subjects with HIV‐associated dementia were excluded. Patients were divided into an unimpaired group, a group with asymptomatic neurocognitive disorder (ANI) and a group with symptomatic HIV‐associated neurocognitive disorder (sHAND), represented by mild neurocognitive disorder (MND). Demographic and background parameters, immune activation markers and the CD4:CD8 ratio were recorded.Results: Two hundred patients were included in the study. The mean age was 52 years, 78% were male, the mean CD4 count was 624 cells/μL, the mean nadir CD4 count was 240 cells/μL, 27% were hepatitis C virus (HCV)‐coinfected, the mean duration of HIV infection was 16 years, and the mean time on current combination antiretroviral therapy (cART) was 2.9 years. Twenty‐nine per cent of subjects had HAND (21% had ANI and 8% had MND). In multivariate analysis, a CD4:CD8 ratio < 1 was associated with a nadir CD4 count < 200 cells/μL [odds ratio (OR) 3.68] and with the presence of CD4⁺CD38⁺HLA⁺ cells (OR 1.23). Multinominal logistic regression showed that, in comparison with the unimpaired group, diagnosis of sHAND was associated with a CD4:CD8 ratio < 1 (OR 10.62), longer HIV infection (OR 1.15) and longer current cART (OR 1.34), while the ANI group differed from the unimpaired group only for education level.Conclusions: Aviraemic patients with sHAND did not display the same pattern of immune activation as subjects with ANI, suggesting that the underlying pathophysiological mechanisms could be different.