The glymphatic system has been proposed as a clearance pathway relevant to neurodegenerative and neuroinflammatory conditions. This study aimed to assess glymphatic function in people with human immunodeficiency virus type 1 (HIV-1) using diffusion tensor image analysis along the perivascular space (DTI-ALPS). The investigators also evaluated associations between DTI-ALPS indices and cognition and tested the diagnostic utility of DTI-ALPS for identifying asymptomatic neurocognitive impairment (ANI) in people with HIV (PWH).
Participants comprised three groups: 40 PWH with ANI (PWH-ANI), 44 PWH with normal cognition (PWH-NC), and 29 age-, sex-, and education-matched persons without HIV. All subjects underwent MRI and standardized neuropsychological assessments. The study calculated DTI-ALPS indices from diffusion tensor imaging to index glymphatic-related diffusion along perivascular spaces.
DTI-ALPS indices were derived from DTI data to quantify diffusion characteristics along perivascular spaces; the bilateral hemisphere index (DTI-ALPS_B) was among the metrics analyzed. Group differences in DTI-ALPS were evaluated using general linear models. Partial correlation and multivariate regression analyses examined relationships between DTI-ALPS indices and cognitive or clinical variables. Mediation analyses tested whether DTI-ALPS mediated the effect of HIV status on cognition. Receiver operating characteristic (ROC) curve analysis assessed diagnostic performance for distinguishing ANI.
DTI-ALPS indices were significantly lower in the PWH-ANI group compared with both persons without HIV and PWH-NC (all P < .01). There were no significant DTI-ALPS differences between persons without HIV and PWH-NC. These results indicate reduced DTI-ALPS is specifically associated with the PWH subgroup who meet criteria for ANI in this sample.
Among PWH, lower DTI-ALPS values were associated with higher cognitive T scores (reflecting poorer performance) in specific domains: attention/working memory, speed of information processing, and abstract/executive function. These associations remained statistically significant after false discovery rate adjustment (all FDR-adjusted P < .05). DTI-ALPS therefore correlated with domain-specific cognitive impairment within the PWH cohort.
Multivariate analyses identified DTI-ALPS_B as the only independent predictor among the DTI-ALPS measures for the cognitive domains noted. Mediation analyses indicated DTI-ALPS_B partially mediated the relationship between HIV status and deficits in attention/working memory, processing speed, and abstract/executive function, suggesting glymphatic-related diffusion changes may be one pathway linking HIV infection to cognitive decline in ANI.
ROC analysis evaluated diagnostic discrimination for ANI. A combined model incorporating DTI-ALPS_B plus the Montreal Cognitive Assessment (MoCA) had an area under the curve (AUC) of 0.782, which significantly outperformed MoCA alone (AUC 0.619; P = .001). This result indicates that adding DTI-ALPS_B to MoCA improved the ability to detect ANI in this study population.
The study found reduced DTI-ALPS indices in PWH with ANI relative to PWH with normal cognition and persons without HIV. Lower DTI-ALPS correlated with and partially mediated cognitive impairment in attention/working memory, processing speed, and executive function. The authors propose that DTI-ALPS may serve as a noninvasive biomarker to facilitate early detection of ANI in people with HIV when combined with cognitive screening tools such as the MoCA.
The abstract reports key design elements, group sizes, statistical outcomes, and the principal conclusions. Detailed information not reported in the provided source text includes full demographic breakdowns beyond age, sex, and education matching, specific MRI acquisition parameters, precise DTI-ALPS computation formulas, covariates included in each multivariate model, and full numeric values for DTI-ALPS indices and cognitive T scores. If required, those methodological and numeric details should be obtained from the full-text article.