Thinking through acetaminophen hepatotoxicity and liver transplantation: Choosing worthy recipients


  • William M. Lee,

    Corresponding author
    1. Division of Digestive and Liver Diseases, Department of Internal Medicine, University of Texas Southwestern Medical Center at Dallas, Dallas TX
    • Division of Digestive and Liver Diseases, Department of Internal Medicine, University of Texas Southwestern Medical Center at Dallas, 5323 Harry Hines Boulevard, HP 4.420E, Dallas TX 75390
    Search for more papers by this author
    • Telephone: 214-645-6111; FAX: 214-645-6114

  • Anne M. Larson,

    1. Division of Digestive and Liver Diseases, Department of Internal Medicine, University of Texas Southwestern Medical Center at Dallas, Dallas TX
    Search for more papers by this author
  • Juan Arenas

    1. Division of Transplant Surgery, Department of Surgery, University of Texas Southwestern Medical Center at Dallas, Dallas TX
    Search for more papers by this author

  • See Article on Page 600

Transplant selection committees make truly vital decisions on candidacy for liver transplantation. At times, they deal with insufficient data or are forced to use nonobjective criteria in their decision-making. This is particularly true for patients with acute liver failure (ALF) and perhaps even more evident for patients with acetaminophen (APAP)-related ALF. With ALF, time is of the essence, patients generally cannot be interviewed, and formal committee meetings are often eliminated. Patients listed as United Network for Organ Sharing status 1, the highest priority in the United States, are limited to ALF and primary nonfunction within 7 days of a previous liver transplant; this category resembles the “super-urgent” listing in the United Kingdom and elsewhere in Europe. Once the decision to list is made, liver grafts are located quite rapidly, so that patients on both sides of the Atlantic typically receive a liver in about 48 hours.1–3 However, what are the criteria for consideration of organ grafting, and are they not somewhat idiosyncratic? Are they not unique to each institution and therefore not standardized across all centers?


ALF, acute liver failure; APAP, acetaminophen.

The criteria for listing for transplantation for ALF fit into 2 categories. First, there is consideration of prognosis: “Will the patient die within 7 days without liver grafting?” Second, there are the psychosocial issues: “What in the clinical history (or laboratory testing) precludes (or supports) providing an organ to this patient?” In this way, physicians are once more called to render a judgment of worthiness for receipt of this precious gift, a donor organ.

The study of outcomes of ALF and transplantation in the current issue of Liver Transplantation, from a single center in Edinburgh over a 13-year period, provides some insights into how this selection process unfolds.3 There are no close-up details given, but rather a view from 10,000 feet regarding assessment for transplantation for APAP-induced and non–APAP-induced ALF. In comparing transplant utilization in APAP and non-APAP cases, the authors reviewed their experience with 469 cases, including 104 (22%) who were transplanted. There were 326 APAP patients (70% of the overall number) and 143 non-APAP patients considered. APAP was defined with history only. We wonder whether there might have been some APAP cases hidden in the non-APAP group because history may be absent on account of the presence of coma or a failure to report the ingestion. We have recently confirmed our initial report that 18% of US cases of indeterminate cases (called seronegative in the article under discussion) were due to unrecognized APAP toxicity.4, 5 In the current study, at least 2 values for alanine aminotransferase of >9000 IU/L were reported in the non-APAP group, and we would consider these findings as very suggestive of an APAP etiology. The authors strictly applied the King's College criteria for both groups and found remarkable sensitivity and specificity, better than what others have been able to demonstrate in recent years: 84% and 96%, respectively, for APAP cases and quite similar numbers for non-APAP cases. Once severity was determined (meeting King's College criteria), the authors refined candidacy on the basis of psychosocial issues. Two-thirds of the APAP patients were refused on the basis of “alcohol dependence,” although an exact definition of this was not provided.

APAP cases are essentially always hyperacute. Patients become ill more rapidly than non-APAP cases and recover more rapidly as well. This was associated in Edinburgh with considerably more cerebral edema in the APAP group (62% versus 26% for non-APAP) and more need for dialysis and inotrope support. It is widely understood that outcomes are better for APAP than for most other causes; however, that was not observed in Edinburgh, where a poorer overall 30-day survival for APAP cases was observed by Kaplan-Meier analysis, although this was not statistically significant. This poorer outcome for APAP was driven by the large number who did not undergo transplant consideration either because they were too ill or because they declined for psychosocial reasons. Only 14% of APAP patients received a graft versus 41% for non-APAP patients. In studies from the US Acute Liver Failure Study Group during this same time period, transplant-free survival was 65%, with only 8% receiving a graft; this meant an overall 3-week survival of 71%. Although the differences between the 2 groups (APAP and non-APAP) in Edinburgh were not significant, there certainly was a trend toward poorer outcomes for the APAP patients, with a difference in overall survival of 20% (71% versus 51%) between the United States and Scotland, which remains unexplained. In addition, posttransplant survival was remarkably different: 69% 30-day survival in Edinburgh versus 89% 3-week survival in the United States. Scottish patients appear to have been sicker than their US counterparts: 43% of the APAP group were considered too sick to list, and 32% of the listed Scottish patients died on the list, whereas 16% of the non-APAP cases died before transplantation.

Why would disease severity be greater in the Edinburgh group, which presumably represents predominantly a suicidal patient group? The authors do not specify whether any patients represented unintentional overdoses in the current article. It should be recalled that US and UK patients appear to differ in the frequency of suicidal cases: US estimates show relatively similar numbers of intentional and unintentional cases, whereas UK estimates suggest that more than 90% of cases involve intentional self-harm.6, 7 Outcomes did not differ between the suicidal and unintentional groups in the United States. Other features such as alcohol and other substance abuse did; 35% overall were considered to have alcohol abuse. The unintentional group was more likely to be using or abusing alcohol or to be using multiple preparations of APAP than the suicidal group.6 In other studies, markers for poorer outcomes included larger ingested doses, alcohol abuse, and longer delay prior to presentation; these data were not provided in the present study.8 The striking feature of the Scottish APAP group was the high numbers that were turned down because of alcohol (56% of potentially listed patients—a much higher number than those considered to have alcohol abuse in the US study). It would appear then that the infrequent use of transplantation and the high death rate of the APAP cases resulted from more frequent alcohol abuse and possibly greater severity of illness in this referral population in Scotland. The authors also provide evidence to support an increase in the prevalence of alcohol abuse over the course of the study.

Interestingly, the number of transplanted APAP cases declined over the course of the observation period, from 16% in the first 4 years to 5% in the last 4 years, and this was consistent with more refusals due to psychiatric issues (largely increased numbers of alcohol abusers). As always, it is hard to understand what might be the differences between studies from 2 sides of the Atlantic. One wonders what the actual criteria for determining alcohol dependence were and how rigorously and consistently these were considered. Remarkably few (7%) patients were considered to have had repeated attempts at self-harm as a reason for not listing.

The APAP overdose problem, confined to the developed world, requires further study. After that, efforts should be made to limit these needless injuries and the estimated 500 deaths per year in the United States alone.2 Although blister packages and package size limits have been somewhat effective in the United Kingdom, there is no will being demonstrated from the US government even to take on package labeling!9, 10 Right now, it would seem wise for us to more carefully study APAP patients so that we can better understand this “disease.” What are the reasons for these overdoses? How many are impulsive 1-time events? What decision-making takes place when we list or do not list patients for psychiatric criteria? In these emergent settings, should skilled interviewers, such as psychiatrists or social workers, be involved in more rigorously determining alcohol dependence or suicide risk? How many overdoses, even in the United Kingdom, are unintentional cases? Does that mitigate the situation in any way, even if there was some alcohol involved? First, we must better understand the disease in order to optimally care for these complex patients. This will require a multidisciplinary approach and careful study to unravel the problem; the Edinburgh article is a good first step.