| Abstract:Objective:
Human Immunodeficiency Virus (HIV) infection can lead to neuroinflammation, neural injury, and, subsequently, risk for neurocognitive (NC) impairment (NCI). Confounding risk is substance use, prompting inquiry into substances commonly used by people with HIV (PWH), particularly cannabis. Research findings have, however, been inconsistent, possibly due to variability in characteristics of cannabis use and level at which NCI is examined. On one hand, chronic cannabis exposure has been found to be associated with better NC functioning in specific domains through possible anti-inflammatory mechanisms. Other research suggests no or negative associations with NC functioning globally due to lingering acute effects of cannabis. The present study aimed to examine relationships between degree of cannabis exposure and various levels of NC functioning in PWH who report chronic cannabis use.
Participants and Methods:
Participants included 358 PWH (18-74 years old, 87% male, 51% white non-Hispanic, 74% on ART, 53% virally suppressed) endorsing past 30-Day cannabis use. The sample excluded individuals with current non-cannabis substance use disorders or positive urine toxicology for other substances. Cannabis exposure parameters obtained via Timeline Follow-back Interview included: Days Used, Quantity, Average Quantity per Day Used (AQPDU), and Average Quantity Across Time Period (Density) over the lifetime, last 12-months, and last 30-days. Short-term effects of recent cannabis exposure, represented by self-reported use within 24 hours or positive THC urine toxicology, were also explored. Independent multiple regression models examined associations between cannabis use parameters with NCI globally and by domain (Verbal, Executive, Processing Speed, Learning, Memory, Attention/Working Memory, Motor). Global and domain impairment was defined using demographically corrected global deficit scores (GDS≥0.5) and domain deficit scores (DDS>0.5) calculated across a comprehensive test battery. Deficits in specific domains were further examined with models of individual test performance T-scores as outcomes. Reading level (WRAT) and nadir (lowest-ever) CD4 cell count were included as covariates. All analyses were evaluated against a Type-I error rate of 0.05 without multiple comparison adjustment.
Results:
No associations between any cannabis exposure parameters and global NCI were found. Higher 30-day Quantity (Median [IQR]=3.0 [0.5,15.0]) was associated lower odds of impairment in Verbal Fluency (OR=0.62, 95% CI [0.40,0.97]) and higher 12-month AQPDU (0.5 [0.2,1.0]) was associated with lower odds of Processing Speed impairment (OR=0.68, 95% CI[0.46,0.99]). In contrast, higher 12-month Density (0.1 [0.01,0.5]; OR=1.21, 95% CI [1.01,1.47]) was associated with higher odds of Executive Dysfunction. Similarly, higher 30-day Density (0.1 [0.02,0.5]; OR=1.41, 95% CI [1.10,1.80]), 12-month Density (OR=1.21, 95% CI [1.00,1.45]), and lifetime Density (0.2 [0.04,0.6]; OR=1.32, 95% CI [1.02,1.70]) were associated with higher odds of Learning impairment. No relationships between domain NCI with positive urine toxicology or 24-hour use we found.
Conclusions:
Although no associations were found between cannabis exposure and global NCI in PWH, several exposure parameters were associated with better performance on speeded tasks, but worse performance on executive functions and learning. Future research should examine thresholds of cannabis exposure or HIV disease characteristics that may moderate these domain-specific effects and explore whether indices of cannabis use map onto neuroinflammatory or other mechanisms that confer vulnerability or protection. |