Higher Comorbidity Burden Predicts Worsening Neurocognitive Trajectories in People with HIV

Background: Comorbidities linked to aging such as diabetes mellitus, pulmonary disease and renal insufficiency accumulate at a faster rate in people with HIV (PWH) than in the general population. We evaluated whether the Charlson Index, a multimorbidity scale comprising 17 variables that has been validated in previous studies, predicted neurocognitive trajectories in PWH. Methods: Neurocognition was measured by averaging scaled scores from all assessments in a comprehensive neuropsychological battery. Multilevel modeling was used to examine between- and within-person predictors of global neurocognition. At the between-person level, average Charlson Index (averaged within each person across all their visits) was examined as a predictor of neurocognitive change over time, covarying for the effect of HIV disease characteristics (proportion of visits virally suppressed, average CD4). At the within-person level, Charlson Index was used to predict fluctuations in global neurocognition at the same and next visit, covarying for visit-specific effects of viral load detectability and current CD4 count. Results: Participants were 1195 PWH (mean age at baseline = 43·0; SD 9·7) followed for an average of over 7·1 years (SD = 5·0; range = 0·5 to 20·5 years). Between persons, higher average Charlson index scores were associated with faster rates of global neurocognitive decline (standardized β = -0·050 [0·015], p = 0·001)· This global effect was driven by significant declines in the domains of executive functioning (p = 0·001) and working memory (p = 0·007). HIV disease characteristics did not predict trajectories of neurocognitive change (ps > 0·05). At the within-person level of the model, lower current CD4+ lymphocytes (β = 0·043 [0·009]; p < 0·001), detectable plasma HIV RNA (β = 0·018 [0·006]; p = 0·001), and higher Charlson Index score (β =-0·046 [0·015]; p = 0·003) independently predicted worse concurrent global neurocognitive performance. Time-lag analyses demonstrated that increasing comorbidities occurred concurrent with, not before, neurocognitive decline. Conclusion: The impact of comorbidities on trajectories of neurocognitive decline was greater than that of HIV disease factors and independent of chronological age. Although correlative, the temporal relationship between accumulating comorbidities and neurocognitive decline suggests that interventions to prevent or ameliorate a variety of comorbidities may improve neurocognitive prognosis for PWH. Funding Statement: This publication was made possible through funding by the NIMH and NINDS by the following grants: HIV Neurobehavioral Research Center (HNRC): P30 MH62512 Manhattan HIV Brain Bank (MHBB): U24MH100931 Texas NeuroAIDS Research Center (TNRC): U24MH100930 National Neurological AIDS Bank (NNAB): U24MH100929 California NeuroAIDS Tissue Network (CNTN): U24MH100928 Data Coordinating Center (DCC): U24MH100925 Translational Methamphetamine AIDS Research Center (TMARC): P50 DA026306 Declaration of Interests: None to declare. Ethics Approval Statement: UCSD's Human Research Protections Program (irb.ucsd.edu) approved all study procedures, and all participants provided written informed consent.