Meta-Analysis: Mortality Trends and Risk Factors in Severe Alcohol-Associated Hepatitis.

BACKGROUND

Severe alcohol-associated hepatitis (sAH) is a life-threatening condition. Despite advances in clinical management, prognosis remains poor and long-term effectiveness of available therapies is uncertain. We conducted a systematic review and meta-analysis to evaluate short-term mortality (28, 60, and 90-day) trends in sAH over the past five decades.

METHODS

We searched PubMed, EMBASE, and Scopus from database inception to February 2024 for studies reporting 28, 60, and 90-day mortality in patients with sAH. Pooled mortality estimates were calculated using a random-effects meta-regression model. We assessed heterogeneity using the I2 statistic and explored sources of heterogeneity through subgroup and meta-regression analyses. Separate Bayesian mixed-effects binomial models were used to estimate the posterior distribution of mortality probability, updated sequentially across calendar time.

RESULTS

34 studies comprising 1586 patients with sAH were included. Pooled mortality rates were 26.8% (95% CI: 21.0%-33.5%) at 28 days, 35.1% (95% CI: 28.3%-42.5%) at 60 days, and 43.7% (95% CI: 34.6%-53.3%) at 90 days. Mortality increased steadily with follow-up time. Substantial heterogeneity was observed, as expected in pooled proportion meta-analysis (I2 > 80%). Although cumulative Bayesian analysis showed that average 28-day mortality declined from over 50% in the 1970s to ~25% after 2000, a formal decade-based analysis indicated no statistically credible improvement in short-term mortality was detected overall in the past four decades. In multivariable models adjusting for follow-up time, the Model for End-Stage Liver Disease (MELD) score was significantly associated with mortality.

CONCLUSIONS

Short-term mortality in sAH remains high and has not improved in recent decades. These findings highlight the urgent need for effective therapies, improved patient selection for early liver transplantation, and better prognostic tools to guide clinical decision-making.