Alcohol use disorder and liver transplant: new perspectives and critical issues

Article information

Korean J Intern Med. 2020;35(4):797-810

Publication date (electronic) : 2020 April 3

doi : https://doi.org/10.3904/kjim.2019.409

Stefano Gitto^1,*, Silvia Aspite^1,*, Lucia Golfieri², Fabio Caputo³, Francesco Vizzutti¹, Silvana Grandi², Valentino Patussi⁴, Fabio Marra^1,5

¹Department of Experimental and Clinical Medicine, University of Florence, Florence, Italy

²Department of Psychology, University of Bologna, Bologna, Italy

³Department of Internal Medicine, SS Annunziata Hospital, University of Ferrara, Cento, Italy

⁴Alcohology Unit, University Hospital Careggi, Florence, Italy

⁵Research Center Denothe, University of Florence, Italy

Correspondence to Stefano Gitto, M.D. Department of Experimental and Clinical Medicine, Univerity of Florence, Largo Brambilla 3, Florence 50134, Italy Tel: +390557947845 Fax: +390557949500 E-mail: stefano.gitto@unifi.it

^*

These authors contributed equally to this work.

Received 2019 December 1; Accepted 2020 March 3.

Abstract

INTRODUCTION

SELECTION OF CANDIDATES FOR LIVER TRANSPLANTATION DUE TO ALCOHOL-RELATED CIRRHOSIS

LT is indicated in patients affected by end-stage liver disease in whom it would increase life expectancy, regardless of the aetiology of the liver disease. LT should be considered when a patient with cirrhosis develops major complications after applying the Model End-Stage Liver Disease (MELD) score to assess the severity of the liver disease, the timing of transplantation and prioritisation of organ allocation [20,21]. In general, both indications and contraindications for LT in patients with ALD cirrhosis are the same as those for patients with cirrhosis of any aetiology [22]. ALD is one of the most common indications for LT in western countries [18,19] and is also one of the most controversial in terms of prioritisation. Moral judgment on what is considered a self-inflicted disease might influence public opinion [23-25]. However, ethical principles recommend whole and effective treatment of these patients without discrimination [26-30].

Today, many ALD patients have no access to an LT program. Kotlyar et al. [31] demonstrated that 100,000 patients per year with ALD and decompensated ALD-related cirrhosis could be candidates for LT. Nevertheless, only 10% were referred, and only 1,200 underwent LT in 2018. Thus, every year, 4% of patients with decompensated ALD-related cirrhosis are on the waiting list and 1.2% undergo LT. This pattern of referral may lead to as many as 12,000 deaths per year [31].

Prediction of alcohol relapse and the 6-month rule

Pre-LT alcohol abstinence represents one of the most controversial questions on this topic. The main predictors of alcohol relapse are summarized in Table 1. Many transplant centers consider the criterion of 6-month abstinence (6-month rule) to establish whether ALD patients can receive livers. Notably, this rule allows the patient’s liver a chance to recover and reduces the risk of alcohol relapse [52-55]. However, the 6-month rule presents some limitations: (1) improved liver function usually occurs during the first 3 months of abstinence; (2) pre-LT abstinence is only one of the factors that might influence the risk of alcohol relapse; (3) a relevant proportion of candidates could die before achieving the 6-month abstinence period [23].

Table 1.

Main predictors of post-transplant alcohol relapse

Chuncharunee et al. [56] analysed the probability of alcohol relapse and its predictors through a systemic review and meta-analysis. The authors reported that the alcohol relapse rate and heavy alcohol relapse rate, defined as harmful alcohol consumption [27], were 22% and 14% respectively. Other authors have reported that psychiatric comorbidities [19,27,57-59] and pre-LT abstinence less than 6 months [27,28,59-65] are the most significant variables predicting alcohol relapse. These results were comparable to the ones obtained in a previous meta-analysis by Dew et al. [66]. Dew et al. [66] considered 12 potential psychosocial risk factors for any kind of alcohol relapse after LT and showed that only three were significantly correlated with alcohol relapse, including low social support, a family history of alcohol abuse and pre-transplant abstinence ≤ 6 months. Notably, the association between these psychosocial variables and alcohol relapse was modest in size and not strong enough to predict the risk of recurrence accurately [66]. Both of these meta-analyses concluded no strong statistical evidence for indiscriminate use of the 6-month rule.

The American Association for the Study of Liver Diseases, the European Association for the Study of the Liver, the United Network for Organ Sharing (UNOS) and the French Consensus Conference clearly stated that 6 months of zero alcohol consumption before LT should no longer be an absolute rule or a defining factor to determine whether a patient can be accepted onto the waitlist [24,52,67-69]. It must be emphasized that a minimum period of abstinence (1 to 3 months) is useful to recover liver function and to assess the possibility of avoiding LT [38].

Given the complexity of the setting as well as the organic, social and the psychological nature of AUD, it is not surprising that other patterns or conditions, other than the “rigid” concept of length of sobriety, might predict the risk of alcohol relapse and, more generally, the adherence of patients to a transplant program. In particular, conditions, such as psychiatric co-morbidity [27], absence of a companion in life and/or the presence of young children [28] represent independent predictors of alcohol relapse. In contrast, patient recognition, family support, employment, permanent residence and sharing a rehabilitation program correlate with a low risk of recidivism and optimal adherence [22,26].

CLINICAL IMPACT OF ALCOHOL RELAPSE

No standardized definition of alcohol relapse has been widely approved. Therefore, the reported percentages vary widely among studies due to different definitions and follow-up periods. However, it is particularly important to discriminate between patients who sporadically drink small amounts of alcohol from those who habitually drink moderate amounts and from those who continuously drink large amounts of alcohol [45]. The risk of alcohol consumption relapse after LT is estimated to be 15% to 40% over 5 years [70,71].

Kodali et al. [72] reported that prevalence was as high as 26.3% over a median follow-up of 6 years; in particular, the authors demonstrated an annual rate of 4.7% for any alcohol use and 2.9% for heavy alcohol intake. The most important finding of this study was the three-fold increased risk of death within 10 years in relapsers compared to abstainers [72]. Many other studies have shown that alcohol relapse could lead to a significant worsening of the survival rate compared to abstinence [73]. Therefore, transplant recipients, considering the risk of loss of control or craving, should be supported to remain completely abstinent throughout their life [1,24,70,73,74]. Few data are available about the potential impact of alcohol relapse on the quality of life of transplant recipients and their family members [75].

Prevention and early diagnosis should be mandatory given the clinical influence of alcohol relapse and the potential social effects of this condition. Although self-reported alcohol use is usually of little value, biomarkers can be a supportive replacement. In particular, metabolites of alcohol, such as ethyl glucuronide, can disclose alcohol use 3 to 4 days after the last drink was consumed [76].

SPECIFIC POST-TRANSPLANT CLINICAL ISSUES

CVD and extra-hepatic malignancies represent the leading causes of long-term mortality in LT recipients [77]. CVD represents one of the most common reasons for death in ALD patients who remained abstinent or with a non-harmful drinking pattern behavior. However, recurrence of HCC and other alcohol-related malignancies is usually observed in patients who relapse with alcohol [28]. The risk of a CV event ranges from 10% at 5 years post-LT to 25% at 10 years [78,79]. This high incidence of CV events can be at least partly explained by the presence of pre-transplant CV risk factors. Moreover, post-LT conditions, such as immunosuppressive agents and de novo metabolic syndrome features, play a significant role [23,80-82]. The prevalence of metabolic syndrome in LT recipients is 60% [83]; that of diabetes mellitus is 10% to 64%, that of obesity (body mass index > 30 kg/m2) 24% to 64%, that of dyslipidaemia is 40% to 66% and that of arterial hypertension is 40% to 85% [84]. The risk of a CV event doubles in patients who develop hypertension or diabetes during the post-LT period [85]. Immunosuppressive therapy can worsen prior CVD and determine the onset of de novo arterial hypertension, dyslipidaemia, diabetes and obesity [86]. As CVDs are responsible for nearly one-quarter of long-term deaths [79,86], these disorders must be identified and promptly treated even by modifying an immunosuppressive regimen [23].

The incidence of de novo malignancies ranges between 3.1 and 14.4% [87,88], representing the first cause of death in patients surviving at least 1 year after LT. In particular, de novo cancer accounts for approximately 40% of deaths [86,89]. The status of the transplant recipient correlates with twice the risk of developing a de novo malignancy compared to the general population [90].

The higher incidence is not the only feature of malignancies in transplant recipients. Malignancies are generally more aggressive, show an earlier onset and have a more relevant impact on survival (with mortality rates of 55%) compared with analogous disorders occurring in the general population [91,92].

Post-transplant lymphoproliferative disorders are the most common malignancies described in LT recipients (20% of cases) [92]. Kaposi sarcoma represents another frequent type of cancer with an incidence similar to head and neck neoplasms (17%). Then, we see oesophageal tumors (12%), lung cancer (10%), gastric cancer (7%) and, less frequently, melanoma, colorectal cancer, cervical cancer, and breast cancer [92].

Intensity and length of immunosuppression [93,94], age and pre-LT liver disease [77,86] are the main predictors of de novo malignancies with some relevant differences according to the tumor site. Notably, ALD correlates with an increased occurrence of skin malignancies when compared to other liver diseases [95,96]. Alcohol is a well-known risk factor for developing oesophageal and gastric tumors [97] that occur more often in LT recipients with ALD [32,96,98]. Furthermore, higher rates of lung cancer [99] and head and neck cancers [100,101] are detected in patients transplanted for ALD. It can be difficult to establish the real burden of alcohol consumption compared to tobacco use as a risk factor for cancer development. Alcohol increases the carcinogenic consequences of smoking [102] and the epidemiological association between smoking and heavy drinking is very strong [103].

OVERALL POST-TRANSPLANT OUTCOME OF ALCOHOLIC CIRRHOSIS

Patients transplanted because of ALD usually show a satisfying overall outcome, comparable to the other main aetiologies [19,104,105]. Patient survival rates after LT for ALD are 81% to 92%, 78% to 86%, and 73% to 86% at 1, 3, and 5 years, respectively [106]. The European Liver Transplant Registry [107] analysed data from 121,546 patients who underwent LT during 1968 to 2015 in Europe. Survival rates for ALD recipients (n = 22,648) at 1, 5, and 10 years after LT were 86%, 73%, and 59%, respectively. These percentages were greater than those for hepatitis C virus (HCV) (80%, 65%, and 53%), HCC (83%, 62%, and 49%), acute liver failure (70%, 64%, and 58%), and hemochromatosis (76%, 66%, and 53%), but comparable to those registered for patients with ALD plus HCV or hepatitis B virus (HBV) (85%, 69%, and 54%) and slightly lower than those described for patients with cholestasis (87%, 78%, and 70%), autoimmune hepatitis (85%, 76%, and 67%), and HBV (83%, 74%, and 68%).

A recently published large meta-analysis [108] indicated that LT significantly reduces mortality of ALD patients (relative risk [RR], 0.25; 95% confidence interval [CI], 0.13 to 0.46; p < 0.02). Moreover, LT decreases the relapse risk (RR, 0.25; 95% CI, 0.14 to 0.45; p < 0.01) if the analysis is restricted to prospectively collected data [109-112]. In fact, limiting the analysis to prospectively collected data exceeds the substantial heterogeneity of a random effects meta-analysis but it can slip on smallstudy effects [108]. Because of the complexity of AUD, further studies are mandatory to characterize its severity, treatment (medication and behavioral therapies) and spectrum more fully, to overcome low reproducibility and methodological heterogeneity [108].

ACUTE ALCOHOLIC HEPATITIS AND LIVER TRANSPLANT

AAH is defined by rapid onset of jaundice because of alcohol intake [113]. Ascites, gastrointestinal haemorrhage or encephalopathy represent the first clinical signs in patients with severe liver disease. Underlying cirrhosis (often undiagnosed) is present in the vast majority of cases [114]. The National Institute on Alcohol Abuse and Alcoholism (NIAAA) Alcoholic Hepatitis Consortia recommend the following standard criteria for defining alcoholic hepatitis: jaundice onset within 8 weeks, ongoing heavy alcohol intake for at least 6 months with less than 60 days of abstinence before the jaundice, aspartate transaminase (AST) > 50 IU/L, AST/alanine transferase ratio > 1.5 and serum total bilirubin > 3 mg/dL [115]. Moreover, a liver biopsy remains indicated in the presence of a doubtful diagnosis [115].

Alcoholic hepatitis represents a subgroup of ALD, accounting for 10% of ALD-related deaths [116]. Few effective medical therapies exist for alcoholic hepatitis [117], and LT as rescue therapy for SAH remains controversial. Patients with SAH are not considered acceptable candidates for LT as they frequently cannot satisfy the 6-month rule with a hypothetical unacceptable risk of alcohol relapse. In fact, to wait for a relevant period of abstinence before LT is often unfeasible for the very high short-term mortality risk (75% to 90% of deaths within 2 months) [118]. The very high mortality and lack of effective therapies indicate that early LT can be a concrete therapy option.

Early identification of patients with a high probability of short-term mortality is mandatory. Maddrey’s discriminant function, the MELD, age-bilirubin-international normalised ratio-creatinine, Glasgow Alcoholic Hepatitis Score, and the Lille model all show predictive ability [119]. Corticosteroids represent the only evidence-based drug option for treating SAH other than alcohol abstinence and treatment of decompensated liver disease [120].

Notably, many scores have been developed to predict the survival of patients with AAH. Among them, the consolidated MELD score has a solid role. Moreover, Maddrey’s score is useful to determine the indication for the use of corticosteroids (in patients with the worst prognosis, score > 32) [119]. The Lille model was proposed in 2007, which differs in utilizing both baseline values and the day 7 serum bilirubin level [118]. The Lille score can be used to assess the clinical response to 7-day of corticosteroids. LT can be avoided in responders while transplant is the only therapeutic option in non-responders.

In 2011, Mathurin et al. [109] conducted a pilot study of early LT for SAH with restrictive inclusion criteria: first liver decompensating event, nonresponse to medical therapy, absence of severe comorbid disorders and motivation for abstinence. Using these criteria, only 7.7% of patients with SAH were eligible for LT. A survival benefit was demonstrated, with 6-month survival of 77% in the LT group (relapse percentage of 12%) and 23% in the non-LT group, respectively. Considering data from the UNOS (2004 to 2010) [121], 5-year graft and patient survival were similar in patients transplanted for AAH and alcoholic cirrhosis (75% vs. 73% and 80% vs. 78%, respectively).

A recently published meta-analysis [122] investigated survival and alcohol relapse rate in patients transplanted for SAH (by definition without 6 months of alcohol abstinence). The authors conducted two different studies. The first study compared early LT with medical therapy alone in patients with SAH not responding to medical therapy. The second study analysed the post-LT outcomes of patients transplanted for SAH and ALD cirrhosis. The first meta-analysis showed a clear 1-year survival benefit of the LT group compared to the medical therapy group. The second study showed no differences in 1, 3, or 5-year survival between patients transplanted for SAH of alcoholic cirrhosis. Finally, there were no differences in the alcohol relapse rate between the two subgroups.

BEHAVIORAL INTERVENTIONS

Adopting a comprehensive AUD treatment for the pre and post-transplant periods is mandatory [123] because abstinence from alcohol represents a prerequisite to enter the transplant list [23] and recurrence of alcohol consumption after LT negatively alters long-term survival [73]. The main goal of patients with AUD, particularly during the transplant process, is to reach and maintain abstinence. Psychosocial interventions are a crucial treatment component to achieve this purpose [124].

This approach can be challenging in ALD patients due to medical and psychiatric comorbidities [125]. Possible interventions vary widely in their content, length, and direction [123]. The intensity and duration of behavioral interventions for alcohol use are controversial, but some show a positive relationship between the length of therapy and maintenance of sobriety [126,127]. Emerging evidence suggests that integrating the transplant setting with addiction therapy should be an optimal goal [126,128]. One study demonstrated that the degree of this integration could bring encouraging outcomes [128]. Behavioral interventions for alcohol are most useful for increasing abstinence rates if put into effect during both the pre and post-transplant periods [64]. Acting after LT can be more difficult due to concomitant medical issues [25] but some groups have carried it out with benefit [64,126,128]. Cognitive behavioral therapy, included in a multi-component intervention, can be effective [126,128]. Defining a skill-based plan [129,130] might be suitable for highly motivated subjects (such as people waiting for LT) [123]. Another possible intervention is socalled “motivational enhancement therapy” [131], which would be appropriate for LT candidates with low insight or who have lapsed [125,132]. Abstinence contracting is relatively spread out over transplant centers [133] but its role in facilitating abstinence has been questioned [134]. Behavioral and psychosocial interventions are feasible and useful for treating AUD in the setting of LT recipients and candidates and usually takes place in a multidisciplinary context [123].

In general, a multidisciplinary transplant team should include an expert psychologist and psychiatrist for early diagnosis and effective treatment of pre-transplant mental illness [135]. Particularly in the specific setting of AUD, the multidisciplinary team should perform a quick and effective evaluation of social support, adequate training about the LT process and an early psychological/psychiatric diagnosis. The multidisciplinary team should favor the development of mechanisms, such as active coping, that improve the overall clinical outcome [135].

No study has proposed a specific psychological approach for the LT setting. However, as we have recently reported [135], the following suggestions are suitable for the design of further studies: (1) any psychological approach beneficial for achieving an active coping strategy; (2) family therapy might be beneficial; (3) group or selfhelp group therapy seem to be beneficial for patients with a history of illicit drug use or AUD; indeed, also in the transplant context, these kinds of therapies might determine positive outcomes.

CONCLUSIONS

ALD and non-alcoholic steatohepatitis represent the true future challenge for the transplant community. Patients with ALD are candidates for LT according to consolidated indications represented by worsening of liver function (MELD > 15), one or more complications of portal hypertension and/or HCC. In this way, patients are often approached equally concerning other indications, such as viral infection. However, there are many clinical conditions present in patients with ALD that might represent or be contraindications for LT. Patients with ALD deserve a specific pre-LT screening program. At the same time, psychological and addiction support should start early during the transplant process despite the length of abstinence. Familial and social support should have absolute priority in this specific setting. Patients transplanted for ALD should be followed-up in three ways. The transplant team should verify the constant presence of a caregiver and social support and should guarantee an intensive psychological support program. These are essential issues to reduce the risk of relapse that seems to negatively alter long-term survival. Secondly, the CV risk profile should be specifically considered. In this regard, limitations on consolidated risk factors associated with an alcohol habit, such as tobacco use, are mandatory. Cancer surveillance should be personalized and reinforced, as patients transplanted for ALD show specific risk profiles. Sufficient evidence exists to consider well-selected patients with SAH as good candidates for LT. The length of sobriety should not be considered the only predictor of a good outcome in patients with ALD, including those with SAH. Other stronger factors are valid predictors of post-LT abstinence and optimal adherence. Active family members, a valid caregiver, social support, motivation, multidisciplinary care and the absence of major psychiatric comorbidities are the best predictors. Notably, different from the length of sobriety, the vast majority of them are changeable with wide and aggressive multi- and interdisciplinary work. An inter- and multidisciplinary approach is the only effective work approach in the transplant context, especially when caring for patients with ALD. Social, psychological and clinical issues influence each other and only a wide overview can obtain noteworthy results.

Article information Continued