Do People Living with HIV and Low-to-Moderate Alcohol Use Have Improved Survival?
- Baseline alcohol use was 51% abstinent/very low, 20% low, 22% “moderate,” and 7% high.
- Over 46,719 person-years of follow-up, there were 491 deaths and 343 non-fatal cardiovascular events.
- Compared with abstinent/very low alcohol use, low and “moderate” use were associated with lower risk for the primary composite outcome (hazard ratio [HR], 0.79 and 0.78, respectively) and for death alone (HR, 0.57 and 0.60). Incidence of non-fatal cardiovascular events did not differ significantly across alcohol use categories.
* Categorized by the authors as: abstention/very low (<1 g/day), low (1–9 g/day), “moderate” (women, 10–29 g/day; men, 10–39 g/day), and high (exceeding “moderate” limits).
Comments:
This study suggests a J-curve relationship between alcohol use and overall mortality in PLWH, with decreased risk observed among those with low to “moderate” alcohol use. However, it is perplexing that cardiovascular events, the purported mechanism for decreased mortality among low-to-moderate drinkers, did not differ by alcohol use categories. Although the researchers did sensitivity analyses omitting some “sick quitters” (i.e., those with illnesses that result in decreased alcohol use and increased mortality) and found similar results, concerns remain given the high prevalence of abstinence/very low alcohol use in this cohort and the potential of the last time-updated alcohol use category to change towards abstinence/very low in advance of death. Similar analyses should be done in other prospective HIV cohorts with the addition of competing risks models and, as the researchers suggest, attention to types of alcoholic beverage and patterns of use.
Kevin L. Kraemer, MD, MSc
Reference:
Wandeler G, Kraus D, Fehr J, et al. The J-curve in HIV: low and moderate alcohol intake predicts mortality but not the occurrence of major cardiovascular events. J Acquir Immune Defic Syndr. 2016;71(3):302–309.