Gray matter volumes discriminate cognitively impaired and unimpaired people with HIV

BACKGROUND: Current diagnostic criteria of HIV-associated neurocognitive disorders (HAND) rely on neuropsychological assessments. The aim of this study was to evaluate if gray matter volumes (GMV) can distinguish people with HAND, neurocognitively unimpaired people with HIV (unimpaired PWH), and uninfected controls using linear discriminant analyses. METHODS: A total of 231 participants, including 110 PWH and 121 uninfected controls, completed a neuropsychological assessment and an MRI protocol. Among PWH, HAND (n = 48) and unimpaired PWH (n = 62) designations were determined using the widely accepted Frascati criteria. We then assessed the extent to which GMV, corrected for intracranial volume, could accurately distinguish the three groups using linear discriminant analysis. Sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, area under the curve (AUC), and accuracy were computed for each model using the classification results based on GMV compared to the neuropsychological assessment. RESULTS: The best performing model was comprised of bilaterally combined GMV and was stratified by sex. Among males, sensitivity was 85.2% (95% CI: 66.3%–95.8%), specificity was 97.0% (95% CI: 91.6%-99.4%), and the AUC was 0.91 (95% CI: 0.83–0.99). Among females, sensitivity was 100.0% (95% CI: 83.9%–100.0%), specificity was 98.8% (95% CI: 93.4%-100.0%), and the AUC was 0.99 (95% CI: 0.98–1.00). CONCLUSIONS: GMV accurately discriminated HAND from unimpaired PWH and controls. Measures of GMV may be highly sensitive to HAND, and revisions to the Frascati criteria should consider including GMV in conjunction with a neuropsychological assessment to diagnose HAND.