Outcomes of Liver Transplantation Without Alcohol Abstinence in Alcohol-Associated Hepatitis With Original MELD Score ≥30

Introduction

Alcohol-associated liver disease (ALD) is a major contributor to alcohol-related fatalities globally [1,2]. Over 40% of all liver disease deaths are linked to ALD [3]. In the past decade, the liver transplantation (LT) wait-listing scenario in the United States has seen substantial changes, especially with a marked shift in the number of patients listed for alcohol-associated hepatitis (AH) [4,5]. This trend not only reflects a heightened awareness of alcohol-related diseases and changes in treatment strategies for these patients, but is also the result of advances in antiviral therapies, particularly for chronic hepatitis B and C [6]. Alcohol-associated hepatitis (AH), an acute form of ALD, particularly in severe cases and non-responders to medical therapy, has a 6-month mortality rate of up to 70% for patients who do not receive liver transplantation (LT) [7–10]. While broader populations may experience lower mortality rates around 20% to 50%, those with severe disease face the highest risk, underscoring the critical need for timely intervention [10,11].

A required period of abstinence, generally set at a minimum of 6 months, was viewed as a fundamental requirement for transplantation, based on the assumption that it would greatly reduce the likelihood of future detrimental alcohol consumption [12]. However, these requirements excluded numerous patients who could have benefitted from liver transplantation but had limited chances of survival. Following the pivotal prospective pilot study conducted in France and Belgium in year 2011 [13], perspectives shifted significantly. The present study found a notable survival advantage for early liver transplant (ELT) in carefully selected individuals with severe AH who underwent transplantation without a mandatory 6-month period of sobriety, when compared to a historical control group. The empirical criterion for the first decompensating event, which is an episode of severe AH in a patient without previous knowledge of liver disease, is the core of the inclusion criteria. The benefits of early liver transplantation (ELT) for selected patients with alcohol-associated hepatitis (AH) are becoming increasingly evident [14]. As a result, major associations like the American Association for the Study of Liver Diseases (AASLD) and the European Association for the Study of Liver Diseases (EASL) have included these findings in their practice guidelines [15,16].

A recent study found that the Model for End-Stage Liver Disease (MELD) score is the most reliable predictor of 90-day mortality in patients with AH [17]. Among patients with severe AH undergoing liver transplantation, 40% have a MELD score exceeding 30, indicating very severe illness [18]. For severely ill AH patients with an extremely high MELD score (≥30), the prognosis is dismal without LT; instead of recovering to a compensated state of liver disease, most of them die within 90 days [19]. However, these patients are often declined from LT listing for psychosocial reasons despite their high MELD scores [20]. There are debates regarding whether to perform ELT for patients with severe AH without mandatory sobriety or for recidivists who continue to drink even after prior decompensating experience. While severe AH patients who sufficiently abstain from alcohol consumption have a significant ideal influence on long-term mortality [21], outcomes among severe AH patients (MELD score ≥30) with ELT versus standard LT (SLT) remains unclear.

Material and Methods

STUDY POPULATION:

During the period between May 2002 and October 2023, 68 consecutive LT recipients fulfilled the inclusion criteria of adult (age ≥18-year-old), high MELD score (≥30), and a diagnosis of clinically severe AH in Linkou Chang-gung Memorial Hospital. After excluding those who had viral hepatitis (n=25), liver cancer (n=7), and Chinese herb-related hepatitis (n=1), a total of 34 patients were included in the final analysis, providing detailed retrospective data. These patients were divided into 2 main groups based on their alcohol consumption status prior to transplantation: 1. The non-abstinence prior to transplant group (Group NA, n=22) consisted of patients with severe AH who reported recent alcohol consumption immediately preceding the acute hepatitis event; specifically, engaging in alcohol use within the month leading to their critical condition. 2. The abstinence prior to transplant group (Group A, n=12) comprised patients who had documented alcohol abstinence prior to transplant for either more than 6 months, or a shorter period (not exceeding 6 months) with abstinence status corroborated by family endorsement. For these patients, the abstinence status was primarily ascertained through patient reports corroborated by family history. In this retrospective study, informed consent was not required as the data were collected from existing medical records. The study was approved by the institutional review board (IRB No. 202401417B0).

ASSESSMENT AND CLINICAL OUTCOMES:

Collected data include baseline characteristics, LT types and details, personal history (tobacco and betel nut use history), and quantification/duration of pre-LT alcohol use. The outcome of interest was overall survival (OS) after LT, calculated since the date of receiving LT to the latest follow-up date or the date of death.

STATISTICAL ANALYSIS:

Categorical and continuous variables are reported as counts (percentages) and mean±standard deviation (median), respectively. To compare clinical parameters between Group NA and Group A, appropriate statistical tests were used based on data distribution. Specifically, the Pearson’s chi-square test or Fisher’s exact test (when expected counts were <5) were used for comparing categorical variables. For continuous variables, the independent samples t test was used for normally distributed data, and the Mann-Whitney U test was used for non-normally distributed data. Survival analyses were conducted using Kaplan-Meier methods, with differences between the 2 main groups and subgroups stratified by identified survival risks assessed using the log-rank test. All potential survival risks identified in the univariate Cox regression model (P<0.100) were included in the multivariate analysis to determine independent risk factors. To confirm the validity of the Cox regression results, the assumption of proportional hazards (PH) for the Cox model was graphically assessed using the Log Minus Log Survival Plot. Parallelism of the curves in the ln(-\ln(S(t))) plot across the comparison groups was considered evidence supporting the PH assumption. Furthermore, the final results of the Cox models are presented with 95% confidence intervals (CI) for all calculated hazard ratios (HR) to reflect the precision of the effect estimates. All P values were calculated as two-tailed, with a significance level set at less than 0.05. Statistical analyses were performed by SPSS Statistics software (version 24.0; SPSS, Inc., Chicago, IL, USA).

Results

COMPARISONS OF COVARIATES ACCORDING TO ALCOHOL ABSTINENCE OR NOT:

This study enrolled 34 patients with severe AH who underwent subsequent LT. Group NA consisted of 22 patients and Group A consisted of 12 patients. Patient demographic data are shown in Table 1. The mean ages were 48.8±10.2 and 51.4±8.0 and the mean MELD scores were 36.1±3.9 and 34.6±3.9 in Group NA and Group A, respectively (not significant P values). The rates of hospitalization and diagnoses from the emergency department were similar between the 2 groups. Regarding prior decompensating history, a higher proportion was observed in Group A (83.3%) versus Group NA (31.8%) (P=0.004).

The daily amount of pure alcohol consumed was 215.6±117.1 and 269.7±111.5 g for 19.1±9.1 and 24.6±9.9 patient-reported alcohol use years in Group NA and Group A, respectively (P=0.200), and half of the patients in Group A were heavy users with daily amount of pure alcohol consumed over 300 g. Six (50.0%) patients in Group A achieved successful alcohol abstinence for at least 6 months before LT, indicating standard LT (SLT). In other words, we had another 6 recipients who received ELT in Group A. In Group NA, the use of a living donor was significantly more common (n=16, 72.7% vs n=4, 33.3%; P=0.050), and over 80% of the living donors were the recipients’ spouse or first-degree relative.

OUTCOMES IN SEVERE AH PATIENTS WITH HIGH MELD SCORES AFTER LT:

After an average follow-up of 1407.7 days, we evaluated the short- and long-term overall survival (OS) outcomes according to pre-transplant abstinence or not (Table 2). The survival rates were similar in both groups, with 1-, 3-, 5-year OS rates of 75.3%, 69.5%, 52.1% in Group NA and 80.2%, 61.5%, 50.8% in Group A (P=0.908). There was a downward trend of survival in the fifth year in both groups. For all patients (n=34), the average survival days were 2137.8±297.0, with a median of 1705.0 days and a 95% confidence interval (CI) of 1555.7 to 2719.9 days. In Group NA (n=22), the average survival days were 2164.2±370.2, with a median of 1486.0 days and a 95% CI of 1438.5 to 2890.6 days. Conversely, in Group A (n=12), the average survival days were 2097.6±454.3, but the median survival was not reached, with a 95% CI ranging from 1207.2 to 2988.1 days.

PREDICTORS AND ASSOCIATED IMPACTS ON POST-LT SURVIVALS:

Univariate Cox regression analysis revealed that recipients with smoking history, diabetes, and clinical presentation of HE had a lower survival benefit. The multivariate Cox regression analysis confirmed that recipients with clinical presentation of HE (P=0.014, HR 4.52, 95%CI 1.35–15.10) had a higher risk of mortality (Table 3). Post-transplant recipient survival times were similar for Group NA and Group A (P=0.908; Figure 1A), while compared to patients without HE at presentation, patients with this clinical feature tended to have inferior outcomes (P=0.007; Figure 1B). These results remained consistent regardless of transplant types. Regarding short- and long-term survival, no difference was observed between the 28 recipients who received ELT and the 6 recipients who received SLT (P=0.986; Figure 2A). Further stratification analysis of OS in different LT types and pre-transplant abstinence status demonstrated comparable outcomes among patients in Group NA who received ELT, in Group A who received ELT, and in Group A who received SLT (P=0.992; Figure 2B).

Discussion

LIMITATIONS:

This study has several limitations that should be acknowledged. First, the relatively small sample size in Group A (n=12) may limit the generalizability of our findings, emphasizing the need for larger, multicenter studies to validate these results. Additionally, the retrospective nature of the analysis introduces potential biases, as data were collected from existing medical records without the benefit of standardized assessments for alcohol abstinence, such as phosphatidylethanol (PEth) testing. While we have provided confidence intervals for survival rates to enhance the reliability of our findings, the absence of objective measures may affect the accuracy of the reported alcohol consumption history. Furthermore, although we observed no significant differences in major complications or survival outcomes between LDLT and deceased donor liver transplantation DDLT, the potential impact of transplant type on long-term outcomes remains an area for further investigation. Lastly, while we have attempted to control for confounding variables, the inherent limitations of observational studies necessitate caution in drawing definitive conclusions regarding the effects of ELT in patients with severe alcohol-associated hepatitis. Future prospective studies are warranted to further elucidate the relationship between alcohol abstinence, transplant type, and post-transplant survival outcomes.

Conclusions

ELT and SLT demonstrate comparable long-term survival rates for patients with severe AH. These findings reinforce the continued application of ELT for medically refractory AH cases with extremely high MELD scores. Furthermore, this research suggests that individuals with severe AH should not be excluded from liver transplantation solely based on their alcohol abstinence status. Prospective, large-scale studies are needed to provide definitive conclusions. Additionally, hepatic encephalopathy at presentation may provide valuable insights for prognostic predictions.