A patient with alcoholic liver failure referred for liver transplantation


  • Philippe Mathurin,

    Corresponding author
    1. Service d'Hépato-Gastroentérologie, Höpital Claude Huriez, Lille, France
    • Service d'Hépato-Gastroentérologie, Höpital Claude Huriez 4ème étage Ouest, Avenue Michel Polonovski, CHRU Lille, 59037, France
    Search for more papers by this author
    • Telephone: +33 3 20-44-55-97; FAX: +33 3 20-44-55-64

  • Michael R. Lucey

    1. Department of Medicine, University of Wisconsin School of Medicine and Public Health, Madison, WI
    Search for more papers by this author

A 47-year-old woman with liver failure due to alcoholism was referred for evaluation for liver transplantation. Approximately 1 year ago, she developed hepatic encephalopathic coma requiring admission into an intensive care unit and assisted ventilation. When she recovered, she agreed that she needed treatment for her alcoholism.

She has been drinking heavily for at least 20 years. Her primary alcoholic drug is vodka. She reports that she drank at home alone. She hid her drinking from her husband and her children. Her family doctor told her about 7 years ago that she had alcoholic liver disease, but she did not stop drinking. She decided to stop drinking 1 year ago on account of her hospital admission for coma. She entered a 7-day inpatient treatment program. She remained abstinent for 6 months. She then relapsed and began drinking covertly. After a few weeks, she developed ascites and swollen legs. After an intervention by the family, she acknowledged her problem drinking and agreed to more treatment. She then underwent a 28-day inpatient treatment for alcoholism. She claims to have been abstinent both during this experience and for 2 months since her discharge from inpatient rehabilitation. Her family confirms that she is abstinent. She attends Alcoholics Anonymous when she feels strong enough, but has been unable to attend for the past few days. She has smoked cigarettes for 30 years, previously anywhere between 20 and 30 cigarettes a day, and now 5 to 6 a day. Although she acknowledges occasional marijuana use, she denies any intravenous use of illicit drugs.

She is married. Her father and brother are recovered alcoholics. She lives with her husband and 2 children. Her husband drinks rarely and is supportive of her sobriety. She is not working outside the home on account of her illness.

Her current symptoms include decreased energy, intermittent pruritus, and intermittent muscle cramps in her hands and legs. As long as she has 2 bowel movements a day, her comprehension is adequate, but she admits to difficulty concentrating. She naps for 3 hours a day. She has difficulty falling asleep and awakens early, which she attributes to diuretics. Her ascites and lower extremity edema are controlled with diuretics and a salt-restricted diet.

Her medical and surgical history includes a cholecystectomy, appendectomy, and hysterectomy.

She was accompanied to the transplant evaluation clinic by her husband, father, brother, and older daughter. Clinical examination revealed an ill-appearing woman. She had marked loss of muscle mass in her upper torso. Her sclerae were icteric. She had numerous spider angiomas and some bruises on her arms and upper body. There was palmar erythema. Cardiovascular and respiratory systems were normal. Her abdomen was protuberant with ascites that was not tense. There was a caput medusae. The liver edge was palpable and tender. The spleen tip was easily detected. There was edema of both legs. She was alert and lucid, and answered questions appropriately. There was no asterixis.

Test results were as follows: hemoglobin, 11.2 g/dL, white cell count 11,600/mm3, platelet count 68,000/mm3, international normalized ratio of prothrombin time 2.7 seconds, serum creatinine 1.8 mg/dL, total bilirubin 7.8 mg/dL, aspartate aminotransferase 56 IU/l, alanine aminotransferase 35 IU/l, alkaline phosphatase 188 IU/l, serum albumin 2.7 mg/dL, and total protein 5.2 mg/dL. Her calculated Model for End-Stage Liver Disease (MELD) score is 31.

Serum markers for hepatitis A, B, and C, autoantibodies, and iron studies were negative or normal. A liver sonogram with Doppler images showed ascites, a shrunken liver, no evidence of a mass lesion in the liver, and patency of the hepatic artery and portal vein.


Questions are as follows: (1) Give a brief account of the clinical therapies available and her prognosis with medical management. (2) What are the risks and benefits of requiring a 6-month abstinence in order to become acceptable for orthotopic liver transplantation? (3) What is her prognosis as she undergoes transplantation?


MELD, Model for End-Stage Liver Disease; DF, Maddrey function


Question 1

Give a brief account of the clinical therapies available and her prognosis with medical management.

In this patient with advanced alcoholic cirrhosis, liver insufficiency started >3 months ago after she resumed to heavy drinking. A recent episode of severe alcoholic hepatitis is excluded because the patient has been sober for 3 months. Therefore, in the present case, specific therapies such as corticosteroids or pentoxifylline are not indicated.1–5 These molecules have been validated only in nonabstinent patients with jaundice developing at <3 months. Therefore, in this patient with >3 months of abstinence, I would not recommend any specific therapy.

However, it is important to underline that at the onset of the decompensation, 3 months ago, it would have been important to calculate the Maddrey function (DF), an accurate tool for identifying severe forms of alcoholic hepatitis. In the absence of treatment, patients with DF ≥32 survived in 65% and 50% of cases at 1 and 2 months. Therefore, if 3 months ago the DF had been ≥32, I would have initiated corticosteroid therapy. Individual data analysis restricted to patients with a DF ≥32 observed that that corticosteroids induce a marked and early improvement in hepatic function, and corticosteroid patients had far higher survival at 28 days (84.6% ± SE3.4% vs. 65.1% ± SE4.8%).4 The biological improvement was sustained to the end of treatment and associated with improved survival.4 This supports an improved healing process related to corticosteroid therapy and provides some additional insights to support their use in severe alcoholic hepatitis.6 In addition, the evolution of bilirubin during the first week of treatment is a useful and simple criterion to identify the patients who benefit from corticosteroids.7

Recently, we developed a prognostic model we call the Lille model that is a powerful tool to identify the patient whose disease does not respond to corticosteroid therapy. All patients with a Lille score >0.45 we now classify as having nonresponsive disease. When this Lille model cutoff of 0.45 is used, close to 40% of patients have early nonresponsive disease. In contrast, in patients with disease that responds to corticosteroid therapy (Lille score <0.45), the therapy induces a rapid improvement of hepatic function, and 3-6 months after the onset of alcoholic hepatitis, most of the patients do not have severe liver insufficiency. Therefore, I speculate that if 3 months ago the DF had been >32, corticosteroid therapy would have decreased the probability that this patient would be a good candidate for receiving a liver transplant.

This patient developed acute renal failure, which is common in patients with cirrhosis, although its exact incidence is unknown. In most cases, acute renal failure results from renal hypoperfusion without cellular injury.8 Patients with cirrhosis have marked circulatory alterations such as low arterial pressure, renal vasoconstriction, and decreased renal blood flow. Prerenal failure may be rapidly reversible if the underlying cause is corrected. In the present case, I do not notice most of the causes of prerenal failure such as vomiting, diarrhea, sepsis, gastrointestinal bleeding, or administration of nonsteroidal antiinflammatory drugs, angiotensin-converting enzyme inhibitors, or angiotensin II receptor antagonists.8 Thus, the more likely mechanism of acute renal failure is a true hypovolemia induced by diuretic treatment. As recommended by the experts, it is necessary to perform a renal ultrasound scan and evaluate the urine (urine osmolality, urinary sodium concentration, fractional excretion of sodium, and urinary protein concentration). To improve the renal function, I recommend suspension of diuretics because renal failure related to diuretics is always reversible with their cessation; and I recommend performing a plasma volume expansion with 1.5 L of isotonic saline or with 60 g of human albumin infusion. After correction of the renal function, I would recalculate the MELD score because its formula includes serum creatinine. In the case of normalization of renal parameters, the MELD score will drop from 31 to 25.

Question 2

What are the risks and benefits of requiring a 6-month abstinence in order to become acceptable for orthotopic liver transplantation?

I do think that this patient is a candidate for liver transplantation when considering the severity of the liver disease assessed as example by Child-Pugh and MELD scores. In patients with alcoholic cirrhosis, the efficacy of liver transplantation has been estimated by comparison of transplanted subjects with matched and simulated controls.9, 10 Transplanted patients of Child-Pugh class C had higher 1- and 5-year survival than their matched controls, whereas among those with Child-Pugh classes A or B, there was no statistically significant survival difference between transplanted patients and their matched and simulated controls.9, 10 The continuing imbalance between the few available livers and the increasing numbers of patients on waiting lists has led clinicians to develop prognostic factors to determine disease severity in order to list and allocate donor organs to the sickest patients. Among available scores, MELD is now considered the gold standard when selecting candidates for liver transplantation.11–13 An American conference held in 2003 stated that MELD meets the goal of providing a system that emphasizes the urgency of the candidate's needs.14 A study compared waiting list and posttransplantation mortality in a cohort of 12,996 adult patients placed on a waiting list.12 Overall, transplanted patients had a 79% lower mortality risk than candidates. With MELD scores of 18-20, the mortality risk was lower among recipients compared with candidates. Survival benefit increased with increasing MELD score. In contrast, at lower MELD scores, recipient mortality risk was much higher in recipients than in candidates (hazard ratio 3.64 at MELD 6-11; hazard ratio 2.35 at MELD 12-14).12 Taking into account that result, this American national conference stated that listing only patients with a MELD score ≥15 is the best option.14

Nevertheless, alcoholic liver disease continues to be the most controversial in terms of public reaction. The main source of concern about liver transplantation in alcoholic patients is the perception that alcoholics are likely to relapse into alcohol abuse after transplantation. In order to ration organs, most programs require a 6-month period of abstinence before evaluation of alcoholic patients. The 6-month period of abstinence is presumed to permit some patients to recover from their liver disease and lessen the need for liver transplantation, but also to identify subsets of patients who are likely to remain abstinent after liver transplantation.

However, the usefulness of the 6-month rule is controversial15, 16 because there is little evidence to document the validity of this criterion alone in predicting alcoholic relapse. Numerous studies lend support to the validity of the 6-month abstinence criterion, but such studies also observe that its use alone forced many candidates who were unlikely to relapse to wait for transplant listing.16–22 Barely 41% of patients with <6 months' abstinence relapsed after liver transplantation, and 87% of patients with ≥6 months' abstinence remained abstinent after liver transplantation.20 Despite an increased incidence of alcoholism relapse observed in recipients with <6 months' pretransplantation abstinence, receiver operating characteristic curve analysis of the value of pretransplantation abstinence revealed this variable to be a relatively poor indication.20 For patients with mild alcoholism, 6 months does not greatly affect the risk of relapse, and such a prerequisite excludes many patients at low relapse risk from prompt listing and eventual transplantation.16 Despite the frequent use of the 6-month rule, the United Network for Organ Sharing23 and the French Conference Consensus on Liver Transplantation24, 25 do not consider this measure as a formal recommendation.

In summary, decisions on transplant eligibility should be made on an individual basis, with careful prediction of short-term survival. In this patient with severe liver insufficiency, a strict application of the 6-month rule as a policy for transplant eligibility may be unfair because she is at risk of dying before the end of the 6-month sober period.

In the present case, I notice the following: the sober period has not improved her liver function, as observed by the MELD score; and her family is supportive of her sobriety. Therefore, I do not recommend waiting 3 additional months to put this patient on a waiting list.

Question 3

What is her prognosis as she undergoes transplantation?

To predict this patient's prognosis after liver transplantation, it is important to differentiate between short- and long-term outcome after liver transplantation.26, 27

The mortality risk on the waiting list while a patient has any given MELD score can be reliably estimated. Conversely, the prediction of short-term posttransplant mortality is still unsolved. In this setting, MELD score cannot be used by itself as a predictive model of posttransplantation mortality. Although unadjusted waiting list mortality covers a 300-fold range from low to high MELD scores, the range of the corresponding posttransplant mortality rates is about 2-fold.28 More completely specified models of posttransplantation mortality are being constructed by means of variables that are good predictors of posttransplantation mortality such as race, age, diagnosis, serum creatinine, diabetes mellitus, need for pretransplantation inotropic support, live support, and care in the intensive care unit. In the present case report, the patient did not disclose any of those features. Therefore, in terms of short-term outcome, I predict a 1-year posttransplant survival of close to 90%.

Patterns of alcohol consumption result in clinically important differences in long-term outcome.26, 27 Long-term studies showed that occasional or moderately heavy drinking did not affect graft function or patient survival.19, 29–31 Conversely, the effect of excessive drinking seems to affect long-term survival. Initial studies focusing on short-term survival did not observe any survival difference between recipients who resumed heavy drinking and others.15, 30–32 Conversely, recent studies evaluating long-term effect of the different patterns of alcohol relapse33, 34 showed that recipients who resumed abusive drinking had far lower long-term survival than abstinent recipients or those with minor relapses.

In conclusion, long-term outcome after liver transplantation will be linked at least in part to the control of her alcohol dependency.