Comprehensive Assessment of Neurocognitive Function, Inflammation Markers, and Adiposity in Treated HIV and Control

Abstract

To compare the neurocognitive scores between persons living with human immunodeficiency virus (PLWH) and persons without human immunodeficiency virus (HIV) and assess the relationship between neurocognition, HIV status and variables, inflammation, and body composition measures. Cross-sectional study involving 225 participants (126 PLWH on antiretroviral therapy [ART] and 99 persons without HIV). For the first time in HIV, we used Cognivue®, an food and drug administration (FDA)-approved computer-based test to assess cognitive function. The test was calibrated to individuals’ unique cognitive ability and measured 6 cognitive domains and 2 performance parameters. Markers of inflammation, immune activation, insulin resistance, and body fat composition (using dual-energy X-ray absorptiometry scan) were collected. Classical t tests, chi-square tests, and spearman correlations were used to compare and explore relationships between variables. Inverse probability weighting adjusted average treatment effect models were performed to evaluate the differences between PLWH and persons without HIV, adjusting for age, race, sex, and heroin use. Overall, 64% were male, 46% were Black, with a mean age of 43 years. Among PLWH, 83% had an undetectable HIV-1 RNA level (≤20 copies/mL). Compared persons without HIV, PLWH performed poorer across 4 domains: visuospatial (P = .035), executive function (P = .029), naming/language (P = .027), and abstraction (P = .018). In addition, PLWH had a significantly longer processing speed time compared to controls (1686.0,ms vs 1606.0,ms [P = .007]). In PLWH, lower cognitive testing domain scores were associated with higher inflammatory markers (high sensitivity C-reactive protein [hsCRP]) and with higher total fat and visceral adipose tissue (P $<$ .05). Neurocognitive impairment (NCI) in HIV is associated with inflammation and total and central adiposity.