Hazardous alcohol consumption is not associated with CD4+ T-cell count decline among PLHIV in Kampala Uganda: A prospective cohort study

INTRODUCTION There is limited data on the effects of alcohol on immunological response among persons living with HIV (PLHIV) in sub-Saharan Africa. We assessed the relationship between hazardous alcohol use and CD4+ T-cell count, among PLHIV in Uganda. METHODS PLHIV aged ≥ 18 years were enrolled in a cohort study at the Infectious diseases clinic Kampala, Uganda. Alcohol consumption was assessed at enrolment (baseline) and 6 monthly thereafter using the alcohol use disorders identification test (AUDIT). The CD4+ T-cell counts, assessed at baseline and over the next 12 months were compared between alcohol use strata, using linear mixed effects regression. Using longitudinal mediation analysis methods, we estimated the effect of alcohol induced ART non-adherence on CD4+ T-cell count. RESULTS Of the 1566 participants enrolled, 863(44.1%) were non-alcohol users (AUDIT score = 0), 433(27.7%) were non-hazardous (AUDIT score 1-7) alcohol users while 270 (17.2%) were hazardous (AUDIT score ≥ 8) alcohol users. The overall median (IQR) baseline CD4+ T-cell count was 356 (243-516) cells/μl. There were no differences in the median baseline CD4+ T-cell count between hazardous and non-hazardous alcohol users compared to non-alcohol users in both the non-ART (p = 0.43) and ART group (p = 0.77). The mean CD4+ T-cell count over 12 months was not different between hazardous alcohol users and non-alcohol users (non-ART group p = 0.88 and ART group p = 0.62), nor between non-hazardous alcohol users and non-alcohol users (and non-ART group p = 0.66 and ART group p = 0.20). Alcohol use was not associated with a significant natural direct effect on CD4+ T-cell count (1.37 95%CI [-1.78, 4.52] cells/μl, p = 0.39) but had a statistically significant natural indirect effect on reduction of CD4+ T-cell count (-0.91 cells/μl [-1.36, -0.45], p < 0.001) mediated through ART non-adherence. CONCLUSION Hazardous alcohol use among PLHIV was not directly associated with lower CD4+ T-cell count but had a significant natural indirect effect on CD4+ T-cell count mediated through ART non-adherence. Among PLHIV with lower than expected CD4+ T-cell count, alcohol consumption should be excluded as an underlying factor for non-adherence to ART and any interventions targeting alcohol use should tackle possible ART non-adherence.