Management of Alcohol Use Disorder in Patients Requiring Liver Transplant

Approximately 17 million adults in the United States (7.2% of Americans age 18 and over) had an alcohol use disorder in 2012 (1). About 1.4 million adults received treatment for alcohol use disorder in 2012 (8.4% of adults in need), which included 416,000 women (7.3% of those in need) and 1.0 million men (8.9% of those in need) (1). Alcohol misuse represents the third largest risk factor for disease burden worldwide, including end-organ damage (2).

The CAGE questionnaire and the Alcohol Use Disorders Identification Test may be used to screen for consequences of drinking and hazardous alcohol consumption. The DSM-5 diagnosis of alcohol use disorder uses a list of 11 symptoms, with mild (two to three symptoms), moderate (four to five symptoms), and severe (six or more symptoms) subclassifications. The DSM-IV diagnoses of alcohol abuse and dependence have been eliminated.

The DSM-5 criteria do not quantify alcohol use. The timeline followback method (3), using a semistructured interview, has become the gold standard for obtaining quantitative drinking information; it allows the interviewer to guide the patient to obtain more accurate information compared with self-report measures. Objective biomarkers have been sought, although none, such as mean corpuscular volume, has the sensitivity to reliably detect heavy drinking. Candidate biomarkers are ethyl glucuronide, ethyl sulfate, and carbohydrate-deficient transferrin, but none of them has reached standard point-of-care in clinical practice (4). Notably, a combination of carbohydrate-deficient transferrin and GGT using a formulated equation improves the sensitivity of detecting heavy drinking without a loss in assay specificity (5).

Liver transplant is a standard treatment for end-stage liver disease, and successful liver transplant extends life expectancy and enhances quality of life. The American Association for the Study of Liver Diseases (AASLD) guideline on evaluation for liver transplantation states that patients with cirrhosis should be referred for transplantation when they develop evidence of hepatic dysfunction, quantified as a score >10 on the Model for End-Stage Liver Disease (MELD) or with onset of ascites, variceal bleeding, and/or hepatic encephalopathy (12). The MELD is a mathematical model that incorporates serum creatinine and bilirubin levels with the international normalized ratio; it is the main measure used to place patients on the transplant waiting list and to prioritize their rank on the list.

Alcoholic liver disease, either alone or in combination with hepatitis C infection, accounted for 20% of all the primary liver transplants in the United States between 1988 and 2009, with >19,000 recipients, and it is the second most common indication for liver transplantation (13).

Five-year survival in end-stage liver disease increases from 23% to 88% with liver transplant (14). Despite this, it is estimated that 95% of patients with end-stage alcoholic liver disease are not referred for evaluation, even when they meet AASLD guidelines for referral (15). There are also disparities in rates of liver transplant waiting list assignment based on geography, gender, acuity of illness, and ethnicity (16).

One important reason for the low referral rate is the bias against patients with alcohol use disorder as candidates for liver transplantation, particularly by community practitioners (17), who may be reluctant to consider giving a scarce resource to those who have failed to treat their alcohol dependence but who develop end-stage liver disease for reasons beyond their control (18). The probability of referral in end-stage liver disease with alcohol-related causes remains significantly lower than that for liver disease with non-alcohol-related causes regardless of current drinking status (19). Nevertheless, patients with alcoholic liver disease who receive transplants have graft survival rates comparable to or even higher than those of patients with non-alcohol-related liver diseases (20). Furthermore, judgments of moral responsibility are not made for other addictions, such as smoking (20).

Current drinking is another major reason for not referring a patient for liver transplant evaluation. A 6-month period of abstinence (the “6-month rule”) before being considered eligible for liver transplant is commonly enforced and is required by most insurance companies (21). Evaluating sobriety is challenging, as the accuracy of self-report measures is questionable and most objective measures have limited specificity and only reflect recent use of alcohol. The 6-month abstinence rule is controversial, as there is evidence that a threshold of 6 months of abstinence is a poor indicator of future abstinence, even though there is an inverse correlation between duration of sobriety and risk of relapse (22). In cases where patients undergo transplant before being required to be abstinent for 6 months, as in patients with severe alcoholic hepatitis unresponsive to medical therapy, there is a long-term survival advantage (23). Importantly, while return to excessive drinking has an impact on posttransplant survival, return to occasional or moderate drinking does not affect posttransplant outcomes (13). Therefore, instead of evaluating drinking in terms of presence or absence of abstinence, it is important to evaluate severity of drinking in order for this parameter to be clinically relevant. Furthermore, there are multiple factors that predict posttransplant relapse, including absence of social support, family history of alcoholism, and presence of psychiatric comorbidity, among others (24).

In the majority of U.S. transplant centers, resumption of drinking after being added to the liver transplant waiting list results in removal from the list either permanently (15% of centers) or temporarily (almost half remove the patient for an additional 6 months), potentially compromising survival in severely ill patients (21). In addition, only a small number of centers (<10%) require treatment for patients who relapse (21).

Nevertheless, according to the AASLD guidelines for liver transplantation, it is optimal to refer patients with alcoholic liver disease early for evaluation for liver transplant to provide time for psychosocial assessment and to set addiction treatment goals in addition to 6 months of abstinence and compliance with an addiction specialist’s recommendations (12).

There are fewer women than men on the liver transplant waiting list (38% compared with 62%). Women have a higher risk of hepatic damage at lower doses of alcohol exposure (25). Once diagnosed with alcoholic liver disease, women have a more rapid acceleration to liver fibrosis than men (26). Accordingly, while women may have greater access to the transplant list (16), they have worse waiting list outcomes and longer waiting list times: women are 30% less likely than men to receive a transplant within 3 years of listing (27). They are also at higher risk of death or becoming too sick for liver transplant (28). To some extent, this has been linked to lower MELD scores due to lower serum creatinine levels in women, thereby underestimating reduction in true renal function in female compared with male patients (29). Longer waiting list times necessitate more careful treatment to minimize the risk of removal from the list for relapse to drinking. After transplant, there is evidence of increased relapse rates and poorer psychosocial outcomes in women (30), highlighting the particular need for ongoing relapse prevention treatment in this group.

According to the AASLD guidelines, evaluation for liver transplantation in adults requires a psychosocial evaluation (12). The major issues to be addressed are ability to comply with medical directives, existence of social supports, particularly in the peritransplant period, and presence of psychiatric disorders, including drug and alcohol use disorders, that can compromise well-being and adherence to medical directives (12).

Factors predicting relapse have recently been reviewed comprehensively (31). Important factors to assess that are associated with a higher risk of alcohol relapse are inability to remain abstinent before transplant, presence of alcohol dependence (as opposed to alcohol abuse), longer duration of heavy drinking, greater number of drinks per drinking day, family history of alcoholism, previous unsuccessful attempts to treat alcoholism, lack of social supports, and psychiatric comorbidity. There is no psychiatric disorder that is an absolute contraindication to liver transplant, including psychosis or mental retardation (12). Although a diagnosis of depression has been associated with an increased risk of relapse to drinking after transplant, it does not appear to be associated with poorer posttransplant morbidity or mortality; rather, depression in the immediate posttransplant period is associated with a higher risk of morbidity and mortality (32). Again, this supports the important role of close, careful posttransplant psychiatric monitoring and care.

While up to 50% of patients with alcohol use disorder manifest withdrawal symptoms on alcohol discontinuation, only a small percentage of them develop clinically significant symptoms requiring medical treatment. End-stage liver disease patients are more likely to present with hypoglycemia and electrolyte disturbances secondary to the neurohormonal changes that occur in liver failure, including hepatorenal syndrome. Concomitant medications for end-stage liver disease, such as furosemide and spironolactone, also contribute to fluid management problems.

General supportive care should correct fluid balance, electrolyte disturbances, and hypoglycemia and should include hydration and vitamin supplementation, including folate and thiamine. Since prolonged glucose supplementation without the addition of thiamine can be a risk factor for the development of Wernicke’s encephalopathy, thiamine supplementation should be started promptly (35).

The severity of alcohol withdrawal is typically measured by using ranked scales, most commonly the Clinical Institute Withdrawal Assessment for Alcohol, Revised (CIWA-Ar) scale. Scores ≥8 indicate a potential need for pharmacological treatment, and scores ≥15 must be treated pharmacologically. Benzodiazepines are considered the gold standard treatment for alcohol withdrawal syndrome since they also reduce the risk of withdrawal seizures and delirium tremens. Benzodiazepines may be delivered in a fixed dosage or triggered by CIWA-Ar score. The latter approach is usually preferred, as it minimizes unnecessary dosing of benzodiazepines, the clearance of which is compromised in end-stage liver disease. There is no evidence that any benzodiazepine is significantly superior to the others in the treatment of alcohol withdrawal syndrome. However, while long-acting benzodiazepines such as diazepam and chlordiazepoxide provide more protection against seizures and delirium tremens because of their long half-lives, shorter-acting benzodiazepines, such as lorazepam, temazepam, and oxazepam, are preferable because they do not require hepatic oxidation.

Several other medications have been investigated for use in this population to treat alcohol withdrawal syndrome, including clonidine, atenolol, carbamazepine, valproic acid, topiramate, baclofen, gabapentin, and pregabalin (35). Among them, gabapentin, topiramate, and baclofen have the advantage of having little or no hepatic metabolism and can be continued through the postwithdrawal period to prevent relapse (for a review, see reference 9). However, unlike benzodiazepines, none of these medications has proven efficacy in preventing complications such as seizures and delirium tremens.

Psychosocial interventions prior to transplant appear to reduce drinking in the pretransplant period. In a recent randomized study, motivational enhancement therapy combined with contingency management was superior to treatment as usual with respect to daily drinking during the waiting list period (36). Recently, a retrospective study (37) compared the drinking outcomes of patients receiving addiction counseling from providers who or were not integrated with the transplant team. The patients whose counselors were part of the transplant team had less alcohol recidivism and lower mortality rates after liver transplant. This emphasizes the utility of an integrated team in caring for this population. The timing of the intervention seems to be important, as treatment before and after transplant reduced posttransplant relapse rates compared with treatment only before transplant or no treatment (38). Abstinence contracts do not seem to have an impact on relapse rates (39), although more work needs to be done on this approach as a potential effective intervention.

Consistent with our increasing knowledge of the neurobiology of addictions, treatment options for alcohol use disorder have been expanded from psychosocial and behavioral approaches alone to include adjunctive pharmacotherapy interventions (Table 1). Of the three medications approved by the U.S. Food and Drug Administration (FDA) to treat alcohol use disorder (disulfiram, naltrexone, and acamprosate), acamprosate, which undergoes minimal hepatic metabolism, has few, mild side effects and no liver-related side effects. Thus, its use in patients with end-stage liver disease should be safe.

Promising medications that are FDA-approved for other indications have been tested in phase II and phase III trials in patients with liver disease. These medications are sometimes used off-label for alcohol use disorder, including topiramate, ondansetron, baclofen, gabapentin, and varenicline (Table 1; for a review, see reference 9). Most of these medications have no evidence of liver toxicity, although only baclofen has been formally tested for safety and efficacy in patients with alcohol use disorder and end-stage liver disease (9). In a 12-week randomized clinical trial (N=84), treatment with baclofen, titrated up to 30 mg/day in three divided doses, compared with placebo, resulted in significantly larger numbers of abstinent patients and total days of alcohol abstinence and a significant decrease in alcohol craving. Additionally, treatment with baclofen was associated with improved liver function tests compared with placebo and was safe, including in a subset of patients with comorbid hepatitis C virus infection (9). In further open-label studies, baclofen was found to promote alcohol abstinence and improve liver function in patients with alcohol use disorder with liver damage as well as in a small group of posttransplant patients with alcohol use disorder (37).

Combined pharmacotherapies for alcohol use disorder have been studied. This approach may offer advantages by using potentially hepatotoxic drugs at lower dosages in combination with drugs that have shown some efficacy that do not undergo hepatic metabolism (for a review, see reference 40).

Up to 60% of liver transplant candidates have a lifetime history of smoking, and the prevalence in alcoholic liver disease is as high as 70% (41). Smoking contributes to the increased incidence of cardiovascular disease and cancers in this group after transplant (20, 42). Therefore, ongoing treatment for comorbid nicotine addiction when managing liver transplant candidates is important. Varenicline, an FDA-approved drug for smoking cessation with minimal hepatic metabolism, was shown to reduce drinking and smoking in a randomized controlled trial in patients with alcohol use disorder (N=200) (43). Varenicline may thus represent a promising addition to potential pharmacologic treatment for this comorbidity, including in the context of liver transplant.

DiMartini et al. (44) found that the trajectory of depressive symptoms in the first year after transplant was predictive of survival during the years that followed: patients with persistently high levels of depression and patients with increasing levels of depression were more than twice as likely to have died on follow-up compared with the group with consistently low depression scores. Of note, 50% of each group abstained from alcohol and there were no significant differences between groups in alcohol consumption or death due to recurrent alcoholic liver disease. In another study (45), early return to problematic drinking after transplant was associated with poorer health and more stress, pain, and demoralization that the transplant did not restore feelings of well-being.

There is considerable comorbidity between alcoholic liver disease and hepatitis C infection, and alcohol use worsens the course of hepatitis C infection (46). In addition, waiting list mortality for patients who have both diseases is higher than for those who have either disease alone (47). Liver transplant graft and patient survival are lowest for hepatocellular carcinoma, hepatitis C, and comorbid hepatitis C and alcohol etiologies for liver transplant (48). Outcomes of patients with alcoholic and hepatitis C are worse compared to those with alcohol and hepatitis B etiologies of liver failure (20). Patients with combined hepatitis C and alcohol etiologies do not have an increased risk of recurrence of hepatitis C after transplant compared to either etiology alone (49). There is no difference in posttransplant survival between patients transplanted for alcoholic cirrhosis and hepatitis C compared to alcohol cirrhosis only (50). By contrast, patients who undergo transplant for both hepatitis C and alcoholic liver disease realize fewer improvements in quality of life compared with patients who undergo transplant for other etiologies (51). At 1-year follow-up, those who had transplants for comorbid alcohol-related and hepatitis C liver disease had worse quality of life measured in all domains. Over 12 years of follow-up, ratings for physical functioning and symptoms continued to decline. These differences remained after controlling for pretransplant medical factors, and there were no group differences in alcohol or tobacco use. Special efforts should be made in posttransplant rehabilitation programs for this at-risk group to restore quality of life and to maintain maximal quality of life levels attained.

Hepatic encephalopathy is a brain dysfunction caused by liver insufficiency and/or portosystemic shunting. It is recurrent and episodic, and it manifests as a wide spectrum of neurological or psychiatric abnormalities, ranging from subclinical findings (unimpaired) to coma (52). It generally resolves after liver transplant. Hepatic encephalopathy will occur in 30%–40% of those with cirrhosis at some time in their clinical course and is one of the factors that elevates the MELD score for those on the transplant waiting list, as it carries a poor prognosis. On one end of the clinical spectrum, minimal hepatic encephalopathy is found in the setting of normal clinical examinations but manifests as abnormalities in neuropsychological tests probing attention, working memory, visual spatial processing, and electroencephalography (53). Overt hepatic encephalopathy presents with a panoply of neuropsychiatric symptoms in the domains of neurological, cognitive, affective, behavioral, and bioregulatory symptoms. When evaluating cognitive fluctuations in patients awaiting transplant, it is important to consider factors that precipitate mental status changes, such as gastrointestinal bleeding, electrolyte disturbances, medications, and infections (for a review, see reference 52). It is important to distinguish alcohol-related dementia from hepatic encephalopathy, as the former is irreversible, and therefore a consideration in waiting list decisions. Of particular relevance, given the high incidence of neuropsychiatric disturbances in this population, psychotropic medications, especially those that undergo hepatic metabolism, must be administered with caution. Whenever possible, psychotropic medications that are excreted renally should be used, such as lithium, topiramate, gabapentin, and pregabalin. Overall, drugs that undergo glucuronidation rather than oxidation as their primary hepatic metabolism should be favored, as this is relatively preserved late in hepatic failure (54).