Persons living with HIV (PLWH) have more aging-related diseases, use more prescribed drugs, and are more likely to have neurocognitive (NC) impairment than the general population. Medical and psychiatric comorbidities increase risk of NC impairment in PLWH but the neurotoxicity of drugs used to treat these diseases remains underinvestigated.
956 PLWH taking antiretroviral therapy (ART) and enrolled in neuroHIV cohorts underwent NC assessment. Demographically-adjusted T-scores were computed for seven cognitive domains and global performance. The most common classes of concomitant drugs were antidepressants (31.1%), antimicrobials (26.2%), non-steroidal anti-inflammatory drugs (21.0%), opioids (16.0%), gastric acid drugs (15.8%), antipsychotics (12.6%), antihypertensives (11.2%), and anxiolytics (10.9%). Polypharmacy was defined as taking ≥5 concomitant drugs. Psychiatric and substance use diagnoses were available for 719 participants. Stepwise multivariable linear regressions using the Akaike Information Criterion modeled NC performance as a function of concomitant medications, adjusting for age, sex, and HIV disease and treatment characteristics.
Participants were generally middle-aged (mean 44.1) white (53.1%) men (86.2%) who had AIDS (56%), viral suppression (71.7%), and immune recovery (median CD4+ count 488/µL). The mean number of concomitant drugs was 3.3 (range 0-24). Overall, PLWH who took more concomitant drugs had worse global performance (r=0.15, p<0.001), as did those who took ≥5 concomitant drugs (d=0.28, p<0.001). Worse global performance was associated with use of anxiolytics (p<0.001), protease inhibitors (p=0.006), opioids (p=0.008), antimicrobials (p=0.009), and antipsychotics (p=0.03) (Model p<0.001). Concomitant drug classes were most frequently associated with worse executive functioning and learning (see Figure). Accounting for psychiatric, medical, and substance use diagnoses and did not significantly weaken the associations between concomitant drugs and global performance.
Concomitant drug use is associated with worse NC performance. Distinguishing the effects of concomitant drugs from those of underlying diseases is complex but our analyses support that certain drug classes may cause reversible or irreversible neurotoxicity. Different drug classes were associated with different cognitive patterns, suggesting that they differently affect the neurobiological pathways underlying these abilities.