Highlights
This multinational prospective study evaluated 936 hospitalized patients with alcohol-associated hepatitis across 32 centers in 14 countries between 2015 and 2024.
Nearly half of the cohort had at least one cardiometabolic risk factor, yet diabetes, hypertension, dyslipidemia, and overall cardiometabolic risk factor burden were not independently associated with 180-day mortality.
Mortality was driven primarily by age, alcohol intake, MELD score, and acute-on-chronic liver failure grade, reinforcing that short-term prognosis in severe alcohol-associated hepatitis remains dominated by liver failure severity.
A prespecified nonlinear analysis suggested slightly lower mortality at body mass index values between 25 and 40 kg/m², with higher risk above 40 kg/m², a finding more consistent with nutritional reserve than a true protective effect of adiposity.
Background
Alcohol-associated hepatitis (AH) is one of the most severe clinical syndromes in hepatology, characterized by acute hepatic inflammation and decompensation in the setting of heavy alcohol use. Hospitalized patients frequently present with jaundice, coagulopathy, encephalopathy, infection, kidney dysfunction, and multiorgan failure. Short-term mortality remains substantial despite advances in supportive care, infection management, and more selective use of corticosteroids. In parallel, the global burden of metabolic disease has risen sharply, and metabolic dysfunction now coexists with alcohol-related liver injury in a large proportion of patients.
The interaction between alcohol use and metabolic dysfunction is biologically plausible and clinically important. Obesity, insulin resistance, diabetes, hypertension, and dyslipidemia contribute to hepatic steatosis, oxidative stress, mitochondrial dysfunction, and inflammatory signaling. In steatotic liver disease, these factors influence fibrosis progression, cardiovascular risk, and long-term outcomes. However, whether cardiometabolic risk factors meaningfully worsen short-term survival once a patient develops severe AH has remained uncertain.
This question matters because AH is increasingly encountered in patients who do not fit the historical stereotype of isolated alcohol-related liver disease. Clinicians now frequently manage overlap phenotypes: heavy alcohol exposure combined with overweight or obesity, diabetes, hypertension, or preexisting metabolic steatotic liver disease. If metabolic dysfunction independently worsens AH prognosis, it would have implications for risk stratification, trial design, and multidisciplinary inpatient management. The current global study by Cari and colleagues directly addresses this knowledge gap.
Study Design
Design and setting
This was a multinational prospective cohort study conducted from 2015 to 2024 across 32 centers in 14 countries. The study enrolled hospitalized patients with AH, using National Institute on Alcohol Abuse and Alcoholism diagnostic criteria.
Population
A total of 936 participants were included. The mean age was 48 ± 11.2 years, and 88.9% were male. Disease severity was high: the median MELD score was 24.4 (IQR 19.3-31.4), 86.7% had severe AH, and overall 180-day survival was 72.9%.
Exposure of interest
The primary exposure was the presence of cardiometabolic risk factors (CMRFs). At least one CMRF was present in 46.6% of patients. Median body mass index (BMI) was 24.2 kg/m² (IQR 22.8-28.2). The prevalence of diabetes was 17.6%, hypertension 16.5%, and dyslipidemia 5.8%.
Statistical approach
The investigators used adjusted competing-risk models, accounting for liver transplantation as a competing event. Covariates included age, sex, ethnicity, history of cirrhosis, cardiometabolic risk factors, corticosteroid use, MELD score, and acute-on-chronic liver failure (ACLF) grade. This is a clinically appropriate approach, because standard survival analysis can overestimate mortality risk when transplantation removes patients from the risk set in a nonrandom way.
Endpoints
The principal endpoint was mortality, analyzed with emphasis on 180-day outcomes. The study also assessed the relationship between individual cardiometabolic risk factors and mortality, including a prespecified nonlinear analysis of BMI.
Key Results
Cardiometabolic risk factors were common but not prognostically dominant
Almost half the cohort had at least one cardiometabolic abnormality, underscoring that metabolic dysfunction is now common in contemporary AH. This alone is clinically relevant. It suggests that inpatient hepatology practice must increasingly accommodate comorbid diabetes management, blood pressure control, and broader metabolic assessment. Nevertheless, simple prevalence does not equal prognostic significance.
In unadjusted survival analysis, 180-day survival did not differ significantly by cardiometabolic risk factor status (log-rank p = 0.453). More importantly, in adjusted competing-risk models, no individual cardiometabolic risk factor independently increased mortality. This includes diabetes, hypertension, and dyslipidemia, as well as the broader category of metabolic dysfunction assessed in the study.
These findings indicate that in hospitalized patients with severe AH, short-term survival is primarily determined by the intensity of acute liver failure and systemic decompensation rather than by baseline metabolic comorbidity.
Traditional severity markers remained the main drivers of death
Several variables retained independent prognostic value. Higher age was associated with increased mortality, with a subdistribution hazard ratio (sHR) of 1.03 per year (95% CI 1.01-1.04). Greater alcohol intake also predicted mortality, although the effect size per gram/day was modest (sHR 1.001; 95% CI 1.000-1.002), reflecting cumulative risk across large differences in consumption.
As expected, MELD score remained a major predictor of outcome, with an sHR of 1.04 per point (95% CI 1.01-1.06). This is clinically coherent, because MELD integrates bilirubin, creatinine, and INR, capturing core dimensions of hepatic and renal dysfunction.
ACLF grade was especially powerful. Compared with lower grades, ACLF grade 2 and grade 3 were associated with substantially increased mortality, with sHRs of 2.34 and 4.34, respectively. This reinforces a critical bedside principle: in severe AH, the burden of extrahepatic organ failure often outweighs chronic comorbidity in determining near-term survival.
The BMI signal was nonlinear and should be interpreted cautiously
One of the more intriguing observations was the prespecified nonlinear BMI analysis. Mortality appeared modestly lower between BMI 25 and 40 kg/m², but rose above 40 kg/m². At first glance, this could be misread as a protective effect of overweight or class I-II obesity. The authors appropriately caution against that interpretation.
In advanced liver disease, BMI is an imperfect surrogate for adiposity and can be distorted by ascites, edema, and sarcopenic obesity. A moderately higher BMI in a hospitalized AH population may partly reflect preserved nutritional reserves rather than metabolically harmful adiposity. Conversely, a lower BMI may identify frailty, protein-calorie malnutrition, muscle wasting, or catabolic illness. Therefore, the observed U- or J-shaped relationship is biologically plausible but not evidence that excess body weight is beneficial.
The increase in risk above BMI 40 kg/m² is also clinically credible. Extreme obesity may worsen respiratory mechanics, infection risk, hemodynamic instability, and overall critical illness complexity, while also compounding chronic inflammatory stress.
Clinical interpretation of the null cardiometabolic findings
The study’s central message is not that metabolic dysfunction is irrelevant in alcohol-related liver disease. Rather, it suggests that once a patient is hospitalized with severe AH, the acute syndrome is so dominant that common metabolic comorbidities do not independently add measurable short-term mortality risk after adjustment for disease severity.
This distinction is important. Metabolic dysfunction may still contribute upstream by accelerating steatosis, fibrosis, portal hypertension, or susceptibility to decompensation before hospitalization. It may also influence longer-term outcomes after the acute episode, including cardiovascular events, recurrent liver-related admissions, or progression to chronic advanced liver disease. The present analysis addresses short-term mortality in hospitalized AH, not the full natural history of dual metabolic and alcohol-related liver injury.
Clinical Implications
Risk stratification should remain centered on liver failure severity
For bedside prognostication, these data support continued reliance on established severity markers such as MELD score and ACLF grade rather than routine up-classification based on metabolic comorbidities alone. A patient with severe AH and multiorgan failure remains high risk regardless of whether diabetes or hypertension is present.
Metabolic assessment still matters, but for different reasons
Even though cardiometabolic risk factors did not independently predict 180-day mortality, they remain clinically relevant. Diabetes can complicate inpatient glucose management, infection risk, and nutritional planning. Hypertension and obesity can influence kidney perfusion, cardiovascular reserve, venous thromboembolism risk, and candidacy for rehabilitation. These comorbidities may also shape post-discharge care, including relapse prevention, cardiovascular prevention, and long-term hepatology follow-up.
Nutritional status deserves closer attention than BMI alone
The BMI findings highlight the limitations of weight-based metrics in AH. Nutritional reserve, sarcopenia, frailty, and body composition likely carry more prognostic information than BMI in isolation. Future inpatient protocols may benefit from incorporating dietary assessment, muscle mass evaluation, and standardized frailty tools into routine AH care.
Expert Commentary
This study has several major strengths. First, its multinational prospective design improves external relevance across different healthcare systems and referral patterns. Second, the sample size is substantial for AH research, a field often constrained by single-center cohorts. Third, the use of competing-risk modeling is methodologically sound given the potential impact of liver transplantation on observed mortality. Fourth, the study reflects modern AH care, where overlap between alcohol-related and metabolic liver disease is increasingly common.
There are also important limitations. The abstract-level data do not provide granular definitions for each cardiometabolic risk factor, the duration or control of these conditions, or the extent of treatment exposure. For example, diabetes is a heterogeneous disorder; insulin-treated, longstanding diabetes with end-organ damage may carry different risk than recently diagnosed mild hyperglycemia. Similarly, dyslipidemia prevalence was low, which may reflect underrecognition, interruption of outpatient treatment, or the fact that lipid levels are often altered in advanced liver disease and severe systemic illness.
BMI is a particularly imperfect measure in decompensated liver disease. Ascites and edema can inflate weight, while sarcopenia may coexist with obesity. Without imaging-based or functional measures of body composition, causal interpretation is limited. Residual confounding is also possible. A patient with higher BMI but better muscle mass, better socioeconomic support, or earlier presentation may fare differently for reasons not directly related to adiposity.
Another point is time horizon. The study focuses on short-term mortality, which is appropriate for severe AH, but cardiometabolic dysfunction may exert more influence over medium- and long-term outcomes. As alcohol abstinence, relapse patterns, transplant access, and chronic metabolic risk all evolve over time, longer follow-up may reveal prognostic interactions not visible at 180 days.
These findings align with the broader understanding of AH pathobiology. Severe AH is marked by profound systemic inflammation, immune dysregulation, hepatocellular injury, portal hypertension, gut barrier dysfunction, and frequent organ failure. In this context, established severity scores often outperform baseline comorbidity measures in predicting near-term outcomes. At the same time, the epidemiologic convergence of alcohol-associated and metabolic liver disease remains highly relevant, particularly for prevention, outpatient disease modification, and long-term survivorship.
How This Fits With Existing Literature
Current guidance from major liver societies emphasizes prognostic scoring, infection screening, nutritional support, alcohol cessation treatment, and selective corticosteroid use in severe AH. The present study reinforces this framework by showing that acute disease severity and ACLF remain the main determinants of early mortality.
Prior literature has established that obesity and metabolic syndrome can worsen chronic liver injury in alcohol-related disease and can interact synergistically with alcohol to promote steatosis and fibrosis. However, evidence has been less consistent regarding their role in acute AH outcomes. This global cohort helps clarify that short-term mortality during hospitalization and the early post-discharge period may not be independently worsened by common cardiometabolic risk factors after accounting for severity of hepatic failure.
The study therefore helps reconcile two truths: metabolic dysfunction is highly relevant to the development and background burden of liver disease, yet in severe AH the acute inflammatory and organ failure phenotype appears to be the dominant determinant of survival.
Conclusion
In this large international prospective cohort, cardiometabolic risk factors were frequent among hospitalized patients with alcohol-associated hepatitis but did not independently increase 180-day mortality after adjustment for disease severity and transplantation as a competing event. Age, alcohol intake, MELD score, and especially ACLF grade were the principal predictors of death. The nonlinear BMI signal suggests that nutritional reserve and body composition may be more informative than BMI alone.
For clinicians, the practical message is clear: in severe AH, short-term management and prognostication should remain focused on liver failure severity, organ support, infection surveillance, nutrition, and alcohol-focused care. Metabolic comorbidity should still be recognized and treated, but not assumed to be a major independent driver of early mortality in the acute AH setting. Future studies should examine longer-term outcomes, more precise body composition metrics, and whether overlap metabolic-alcohol phenotypes influence recovery trajectories after the initial hospitalization.
Funding and ClinicalTrials.gov
Funding information was not provided in the abstract supplied here. A ClinicalTrials.gov registration number was not reported in the provided citation or abstract text.
References
1. Cari V, Perez MI, Tellez I, Idalsoaga F, Valsan A, Nair G, Soni K, Rhoads SF, Arnold J, Ayala-Valverde M, Ramirez-Cadiz C, Jalal PK, Ibrahim MA, Elfeki MA, Roblero JP, Simian D, Velarde-Ruiz Velasco JA, Cordova J, Higuera-de-la-Tijera F, Silva R, Rocha CM, Araujo RC, Fontani GM, Pereira GH, Bessone F, Tanno M, Kisch A, Mendizabal M, Marciano S, Gomez-Perdiguero G, Montes P, Guerra Salazar P, Ramos G, Chacko KR, Skladaný L, Maiwall R, Simonetto D, Sanyal AJ, Shah VH, Singal AK, Dunn W, Kamath PS, Loomba R, Arrese M, Bataller R, Díaz LA, Arab JP. Exploring the interaction between metabolic dysfunction and alcohol-associated hepatitis: A global study. Hepatology (Baltimore, Md.). 2026-05-06. PMID: 42090455.
2. Crabb DW, Im GY, Szabo G, Mellinger JL, Lucey MR. Diagnosis and treatment of alcohol-associated liver diseases: 2019 practice guidance from the American Association for the Study of Liver Diseases. Hepatology. 2020;71(1):306-333.
3. Forrest EH, Atkinson SR, Richardson P, Masson S, Ryder S, Thursz MR, Allison M. Application of prognostic scores in the STOPAH trial: discriminant function is no longer the optimal scoring system in alcoholic hepatitis. Journal of Hepatology. 2018;68(3):511-518.
4. EASL Clinical Practice Guidelines on acute-on-chronic liver failure. Journal of Hepatology. 2023;79(2):461-491.
5. Louvet A, Labreuche J, Artru F, Bouthors A, Rolland B, Saffers P, Naassila M, Dharancy S, Lassailly G, Mathurin P. Main drivers of outcome differ between short term and long term in severe alcoholic hepatitis: a prospective study. Hepatology. 2017;66(5):1464-1473.
