Rescue liver transplantation for severe alcoholic hepatitis: Arriving where we started?


  • Ewan H. Forrest M.D.,

    1. Department of Gastroenterology, Glasgow Royal Infirmary Glasgow, UK
    Search for more papers by this author
  • Michael R. Lucey M.D.

    Corresponding author
    1. Division of Gastroenterology and Hepatology, Department of Medicine, University of Wisconsin School of Medicine and Public Health, Madison, WI
    • Division of Gastroenterology and Hepatology, Department of Medicine, University of Wisconsin School of Medicine and Public Health, 4245 MFMB, 1685 Highland Ave., Madison, WI 53792-5124
    Search for more papers by this author

  • See Article on Page XXX.

  • Potential conflict of interest: Dr. Lucey consults for Alkerme. He received grants from Salix, Vertex, Abbott, and Hyperion.

We shall not cease from exploration

And the end of all our exploring

Will be to arrive where we started

And know the place for the first time

—T.S. Eliot, Four Quartets

Alcoholic liver disease (ALD), either alone or in conjunction with other diseases such as chronic hepatitis C infection, has become a major indication for liver transplantation in North America and Europe.1 How different were the predictions of the watershed National Institutes of Health (NIH) consensus meeting on liver transplantation in 1984 that declared: “Patients who are judged likely to abstain from alcohol and who have established clinical indicators of fatal outcomes may be candidates for transplantation. Only a small proportion of alcoholic patients with liver disease would be expected to meet these rigorous criteria.” The initial reticence centered not only on concern that ALD transplant recipients would relapse to alcohol use, but also that alcoholic patients were less deserving of scarce donor organs because of their complicity in causing liver damage, and that the low esteem in which the public holds alcoholics would translate into donor families declining to donate if the recipient were likely to be an alcoholic.3, 4


ALD, alcoholic liver disease; HCV, hepatitis C virus; MELD, Model for Endstage Liver Disease; UNOS, United Network for Organ Sharing.

This orthodoxy was challenged by Dr. Thomas Starzl who, in 1995, reported that alcoholic patients had excellent short-term outcomes after liver transplantation.5 In the subsequent years, we have learned that survival benefit of liver transplant for ALD exceeds that in hepatitis C virus (HCV) but also that, even with careful selection, ∼20% of ALD liver transplant recipients return to abusive drinking at some point and that relapse to abusive drinking is associated with liver damage.6, 7

In the United States in 2012, most transplant centers select ALD patients after evaluation by an addiction specialist but also require observation of an abstinence period, most commonly 6 months. In the United Kingdom, there is no mandatory time requirement for abstinence, but again, the 6-month abstinent period has been adopted by many as advisory. In both countries patients with acute alcoholic hepatitis have been specifically excluded, on the grounds that it is necessary to wait in order to give an opportunity to recover.8, 9 In practical terms, the “6-month rule” has been an insurmountable barrier for most.10

Patients with severe alcoholic hepatitis, who have failed medical therapy, have high 6-month mortality, exceeding 70% in some studies. Recent data from the U.S. and Europe challenge our easy acceptance of excluding patients with alcoholic hepatitis failing medical therapy.11 In a recent edition of HEPATOLOGY, Singal et al.12 reviewed the United Network for Organ Sharing (UNOS) database from 2004 to 2010 and found 130 patients with alcoholic hepatitis who had been “listed” for transplantation, of whom 59 received a transplant. Comorbid HCV was present in 14 (25%), whereas 11 had histologic appearances of alcoholic hepatitis on explant pathology, 33 had cirrhosis, and the remainder had other diagnoses. Notwithstanding the small numbers, and the heterogeneity surrounding the diagnosis of alcoholic hepatitis, it is encouraging that graft and patient survival was similar in the alcoholic hepatitis cohort compared to a control cohort of nonalcoholic recipients selected by sequential matching according to sex, race, year of transplant, age (±5 years), donor risk index, and Model for Endstage Liver Disease (MELD) score.

Wells et al.13 retrospectively reviewed the explanted livers of 148 patients transplanted for ALD alone who were drawn from a single center cohort of 1,097 patients transplanted over 18 years. The histological features of alcoholic hepatitis were found in 32 (22%) ALD recipients. In this series, recorded duration of pretransplant abstinence did not correlate with explant histology. Furthermore, patient and graft survival was the same in patients with bland alcoholic cirrhosis or cirrhosis plus alcoholic hepatitis, and among 125 matched non-ALD recipients.

These studies are limited by retrospection, and the infrequency of transplantation for alcoholic hepatitis either defined clinically in the UNOS database or on transplant histology. The unexpected finding of explant histology which is compatible with alcoholic hepatitis is clearly very different from the patient presenting acutely with the florid clinical syndrome of alcoholic hepatitis with its attendant jaundice, coagulopathy, and high short-term mortality.

To address this difficult and controversial patient group, Mathurin et al.14 published a prospective pilot study of liver transplantation as rescue for severe alcoholic hepatitis that failed to respond prednisolone 40 mgs/day for 7 days. All subjects had a favorable psychosocial profile, without comorbid psychiatric disease, and acknowledged their need for life-long abstinence. Selection required concurrence among five constituencies comprising nurses, doctors in training, addictions specialists, hepatologists, anesthetists and surgeons, and the patient and his or her family. The primary endpoint was 6-month survival from starting corticosteroids. Secondary endpoints included impact on transplant program and recipient abstinence from alcohol. The severity of illness of the 26 subjects undergoing transplantation is shown by a median MELD score of 34.

The median interval from stopping corticosteroids to placement on the waiting list was 13 days, and from listing to transplantation was 9 days. The 6-month survival was 77.8% compared with 23.8% in historical controls, matched to be “best fit.” Infection, particularly fungal infection, was the cause of most posttransplant deaths. The transplanted patients represented only 2.8% of transplants done and the study was ongoing in the seven centers. Only three subjects returned to drinking, all late after transplantation.

The crux of the question of transplantation for alcoholic hepatitis, and its acceptance by the hepatology community, is patient selection. Patients with other indications for liver transplant such as acute liver failure from deliberate acetaminophen overdose are readily accepted for transplantation. This is not because they have a superior posttransplant outcome to patients with ALD, but rather because we have highly specific tests of prognosis which can identify those who will almost certainly die without such intervention. In the study by Mathurin et al. the Lille Score was used to identify corticosteroid nonresponders.

Although a statistically robust score, using a Lille threshold of ≥0.45 will lead to nearly one in four patients being transplanted who otherwise would have survived with standard medical therapy.15 This survival rate may even rise to 50% in selected patient groups using the same Lille threshold.16 Justification for transplanting such patients when other patients whose prognosis is uniformly grim are dying on the waiting list is problematic. In the U.K. more than 10% of patients die while listed for transplantation, whereas in the U.S. in 2011, ∼4,000 patients died on the waiting list or were removed for reasons other than clinical improvement.17, 18

Where have we arrived with regard to liver transplantation as rescue therapy for alcoholic hepatitis failing medical therapy? Notwithstanding the tangential evidence from the UNOS database presented in June in HEPATOLOGY, and single-center studies such as that of Wells et al., and the extraordinary prospective data presented by Mathurin et al., alcoholic hepatitis remains excluded from the indications for liver transplantation, except in isolated cases. This will probably remain the norm until a new consensus emerges involving insurance payers, transplant hepatologists and surgeons and other members of the transplant teams, and the public. Until we have more specific objective criteria for selecting patients, we believe that it will prove difficult to introduce into standard practice transplantation for patients with severe alcoholic hepatitis who have failed medical therapy.

Whether a higher threshold of Lille Score will achieve this remains to be tested.11 However, it is imperative that transplant programs on both sides of the Atlantic remain flexible enough to allow further controlled assessment of liver transplantation for alcoholic hepatitis. We need prospective studies from both Europe and the U.S. to corroborate the findings of Mathurin et al. and to explore the ethical and fiscal impact of widening the net of transplantation to include alcoholic hepatitis.