Association between Alcohol Consumption and All-Cause Mortality in Patients Undergoing Hemodialysis: A Survival Analysis from the DIET-HD Cohort.

KEY POINTS: Low-to-moderate alcohol consumption was not associated with increased mortality among adults receiving maintenance hemodialysis. Findings were consistent when alcohol intake was modeled continuously and when accounting for beverage type. Heavier drinking patterns were uncommon in this cohort, limiting inference at higher levels of alcohol exposure.

BACKGROUND

Patients on hemodialysis face a mortality risk eight times greater than the general population. Although alcohol consumption is associated with increased mortality in the general population, its effect has not been studied in hemodialysis patients. This study aimed to assess the association between alcohol consumption and all-cause mortality in this population.

METHODS

The Diet-hemodialysis prospective cohort included adult patients undergoing hemodialysis in Europe and Argentina. Alcohol intake was assessed at baseline using a food frequency questionnaire and categorized as abstinent, occasional drinker (<14 grams per day [g/d]) or regular drinker (>14 g/d). Participants were followed until death or discharge from the dialysis unit. The primary outcome was all-cause mortality; secondary outcomes were cardiovascular (CV) and non-CV mortality. We examined associations using Cox proportional hazards models as the primary analysis and Fine and Gray competing risks models as secondary analysis. Confounding by demographic, clinical, and lifestyle factors was addressed using propensity score inverse probability weighting.

RESULTS

Among 7917 participants, 4097 (52%) were abstinent, 2981 (38%) were occasional drinkers, and 839 (11%) were regular drinkers. Wine accounted for most alcohol intake, and the mean consumption among regular drinkers was 26 g/d. Over a median follow-up of 4.0 (interquartile range, 1.6-4.8) years, 3753 deaths (36%) occurred. In the weighted and adjusted Cox proportional hazard model, the hazard ratios for all-cause mortality were 0.96 (95% confidence interval, 0.87 to 1.06; P = 0.46) and 1.05 (95% confidence interval, 0.88 to 1.26; P = 0.57) for occasional and regular drinkers, indicating no significant association between all-cause mortality compared with nondrinkers. Similar findings were observed for CV and non-CV mortality subanalyses.

CONCLUSIONS

In this large cohort of adults receiving maintenance hemodialysis, low-to-moderate alcohol consumption was not associated with all-cause or cause-specific mortality. These findings may not apply to heavier drinking patterns, which were uncommon in this population.