Alcoholic relapse damages liver allografts

The narrative concerning the place of liver transplantation (LT) in the treatment of life-threatening alcoholic liver disease (ALD) has evolved in the 30 years since the National Institutes of Health consensus conference declared that LT was no longer an experimental intervention.1 Over the years, attitudes to offering LT to patients with ALD have changed from the assertion that very few patients with ALD would be chosen for LT, to an acceptance that ALD had emerged as a common indication for LT and that the main issue had become how best to identify prospective LT candidates about whom the transplant program could have confidence of lifelong abstinence before and after LT.2-4

A weakness in these discussions emerged quickly with the realization that even with the adoption of stringent selection practices, often based on a requirement for a 6-month abstinence prior to transplantation or even prior to placement on the transplant list, invariably some ALD patients resumed drinking after LT. Early on, there was much conjecture on accurately estimating the actual proportion of ALD LT recipients who returned to drinking, regardless of relapse pattern. The focus shift to relapse pattern was driven by the addiction treatment community who drew a distinction between occasional lapses or “slips” and relapse to addictive or harmful drinking.5 The best data on rates and patterns of relapse come from a prospectively followed cohort of ALD transplant recipients assembled by DiMartini et al.6 at the University of Pittsburgh, among whom, 20% returned to significant or “harmful” drinking. The realization that alcohol relapse was inevitable has allowed an additional question to emerge: what is the impact of a return to drinking after LT?

In this issue of Liver Transplantation, Erard-Poinsot et al.7 add to the growing literature demonstrating that relapse to addictive drinking is harmful to the welfare of LT recipients. It might seem strange that this could come as a surprise. However, initial inquiries into graft and patient survival of LT recipients who underwent transplantation for ALD often lacked the clarification between harmful drinking and slips. However, once the character of the return to alcohol use was taken into account and was linked to longer durations of follow-up, a body of single-center data came forward to suggest that patient survival was impaired in the “relapsing” group.8, 9 Erard-Poinsot et al.7 have advanced our knowledge of this phenomenon by focusing on liver biopsy data rather than patient or graft survival. Their observation that a return to harmful drinking is associated with increased fibrosis in the allograft confirms earlier reports that documented similar findings, and it allows us to assert with conviction that addictive drinking leads to fibrosis in the liver allograft.10, 11 Furthermore, their data suggest that the progression of alcohol-related liver damage in the allograft might be accelerated compared to alcohol-related damage in a native liver. This finding also reinforces previous data demonstrating accelerated allograft fibrosis progression in patients who relapse.10

Why has it taken so long for the damaging effect of alcohol relapse to become evident? First, the authors who emphasized the lack of drinking by ALD patients after LT have a point, or as the aphorism attributed to Thomas Starzl puts it that “liver transplantation is the ultimate sobering experience.” That 80% of ALD patients should abstain from harmful drinking after LT is remarkable, considering that alcohol-use disorder is a lifelong condition characterized by remission and relapse. In Vaillant's12 longitudinal studies of men with ALD in the non-LT setting, remission was only secure after 5 years. Most LT recipients have not demonstrated longterm sobriety prior to transplantation, and the relatively low prevalence of relapse to heavy alcohol use posttransplant compared to the nontransplant population is surprising and encouraging.12 The second reason for failure to recognize alcohol's role in damaging LT allografts is well shown by the Pittsburgh cohort referred to above, and it is confirmed in the present and other retrospective reports: alcoholic relapse has variable patterns, with some patients resuming drinking soon after LT, whereas others do so after an extended period of post-LT abstinence. Some patients drink without stopping, whereas others stop and restart, and some stop for good. In addition, quantification of alcohol use posttransplant is exceedingly difficult because of the subjective nature of reporting by a patient who may feel inclined to try and hide relapses or minimize the quantity of alcohol consumed due to embarrassment or guilt. Consequently, unless an ALD LT cohort is large and is followed for several years, there will not be a sufficient number of recipients with significant drinking to see an effect.

If our understanding that alcoholic relapse may lead to fibrotic damage in the allograft has only recently become clearer, we remain in the dark about how we, as medical professionals, can improve this situation for our patients and prevent the progression to cirrhosis. We are hampered by our difficulty in identifying drinking among ALD LT recipients and in stratifying the risk of relapse in an individual LT recipient. The ALD patients who receive LTs are not uniform in regards to risk of relapse to addiction. Many ALD LT recipients have a strong sense of having recovered from “alcoholism.” These patients tend to deny craving and consequently express low motivation for undergoing treatment.13 This is in contrast to patients in addiction treatment units, where endorsing the treatment model is considered a favorable prognostic indicator. It is likely that some of these responses are genuine and indicate an absence of internal prompts to consume alcohol. Lack of interest in receiving treatment for alcoholism was 1 reason for failure to recruit to a trial of naltrexone in ALD LT recipients.14 Additionally, this resistance to treatment on the part of ALD LT recipients may reflect the fear that admitting to a craving would be interpreted by the transplant team as a sign of poor candidacy or a lack of commitment to sobriety. Indeed, fear that their candor about alcohol use could have negative consequences for their ongoing care in the transplant program may be an impediment to getting timely psychological intervention in those persons who have already started on the path to relapse.15 In view of the relatively low relapse rate and the heterogeneity of drinking after LT, it would be best to target the ALD LT recipients with the highest-risk profiles, including those of who have relapsed or who have experienced slips.

Thus, 2 important factors are necessary in order to design a successful intervention to interdict the progress to alcohol-related liver fibrosis in this selected population. First, as discussed above, we need to offer assistance to the population most “at risk” of returning to harmful drinking. Part of this process would be to recognize those patients who continue to have cravings for alcohol. A complementary element would be to identify the beginning stages of a resumption of alcohol use, such as a recent slip or early relapse. Developments in better biomarkers of alcohol exposure may be very helpful here.16 Second, we need better treatments for the “at risk” population. There are many treatments that offer hope in the nontransplant population that are unlikely to transfer into the post-LT cohort. For example, potentially hepatotoxic medications such as naltrexone are likely to be rejected by LT recipients.14 One way forward has been described by Björnsson et al.,17 who introduced into their transplant program a plan for structured management of addiction to alcohol, comprising assessment by a psychiatrist skilled in care of alcoholics, initiation of treatment in patients who had not been treated in the past, encouragement to participate in motivational enhancement, and use of an abstinence contract. The protocol was started before and continued after transplantation with interviews at 3 months and at 1, 3, and 5 years. In consecutive patients, they observed a reduction in the prevalence of alcohol use when compared to a matched historical control group (48% versus 22%), although they do not report their data in terms of harmful drinking. Furthermore, well-powered studies are needed to test this and other approaches to preventing alcohol-related damage to allografts.