Addictive behavior after solid organ transplantation: What do we know already and what do we need to know?^†

Transplantation of kidneys, livers, hearts, and lungs has changed from a last-hope effort to become part of the standard armamentarium of tertiary care throughout the first world. As part of this gradual evolution, liver transplantation in the United States has attained many of the markers of an established area of clinical practice, such as concentration on liver transplantation at the annual Liver Meeting of the American Association for the Study of Liver Diseases, a thriving journal, and, most recently, specific subspecialty training programs in Transplant Hepatology accredited by the Accreditation Council for Graduate Medical Education and a certifying examination in Transplant Hepatology under the auspices of the American Board of Internal Medicine. At the same time, there has been a gradual evolution in how outcome after solid organ transplantation is assessed. Although organ survival and patient survival remain crucial endpoints, as shown by the outstanding annual data reports published by the Scientific Registry of Transplant Recipients, as long-term survival has become commonplace, an increasing emphasis is being placed on morbidity after solid organ transplantation.1, 2 The recurrence of the original disease, the development of new diseases, the morbid consequences of immunosuppression, and the psychological health of the transplant recipient all contribute to a more complete understanding of the lives of recipients of solid organ transplants. As such, alcohol abuse and drug abuse after transplantation are important areas of interest in assessing the quality of life after solid organ transplantation.

Unfortunately, there are many inherent barriers to getting accurate data about alcohol and/or illicit drug use by transplant recipients. Several of the most common methods used to obtain information, such as self-reports, assessment by clinical findings, and collateral reports, are subject to many biases. In particular, the lack of an objective and reliable instrument to measure alcohol intake and the perceived risk to the recipient that candor about drug or alcohol use could harm his or her chances of receiving further care from the transplant program tend to encourage underreporting.3 On top of this, the medical literature dealing with alcohol and drug use after transplant is replete with retrospective data, whereas only a few studies have been performed in a prospective fashion.4 Finally, as a result of our interaction with addiction specialists, an important evolution in the understanding of addictive behavior has occurred within the transplant community as shown by the gradual change in the definition of alcohol relapse used in liver transplant studies. Early publications tended to define relapse as any alcohol use by the alcoholic patient.4 In contrast, recent retrospective and prospective studies have adopted the nomenclature of addictive medicine by distinguishing a transient relapse (a “slip”) from harmful drinking.5-7

Given these caveats, it is understandable that, despite the increased interest in posttransplant morbidity, it has been difficult to apply the tools of meta-analysis to the assessment of the frequency of addictive behavior after transplantation. Consequently, Dew et al.8 are to be congratulated for their outstanding effort to analyze alcohol and other drug use after transplantation, which is published in this issue of Liver Transplantation. Among the 54 papers that they identified as dealing with this topic, 50 were concerned with liver transplant recipients, 3 dealt with kidney recipients, and there was 1 on heart transplant recipients. No paper dealing with addiction in lung transplant patients was found that met their entry criteria. The first question posed by this review is why have so few studies been directed to addictions in recipients of transplants other than liver transplants. We do not know whether the disparity in the number of papers is a reflection of a lack of concern that alcoholism or drug use occurs in these populations. One explanation for a lesser degree of interest could be that alcoholism and intravenous drug use are primary drivers of many of the liver diseases that lead to liver transplantation, whereas they would, when present, constitute a secondary problem in the kidney, heart, and lung transplant populations. Another possible factor is the different pattern of transition of care in kidney transplant recipients, who are returned to their community heath care providers, where the impact of addiction in this population may not be as easily recognized. In any case, Dew et al.'s study is a call to clinical investigators in the kidney, heart, and lung transplant worlds that an opportunity awaits whoever wishes to document and analyze the problem of addiction in these populations.

Thus, Dew et al.'s study8 concentrates almost exclusively on liver transplant recipients and seeks to discover how frequently these patients' addictions relapse after transplantation. According to their meta-analysis, when alcoholic liver disease is the pretransplant diagnosis, approximately 6 out of 100 recipients per year will use alcohol after transplantation, and less than 3 will resume heavy alcohol use. The rate for illicit drug use is even lower. For the addiction specialist, the persistence of sobriety after transplantation is surprising and unexplained.9 Among the possible explanations for these findings are that they are a consequence of liver transplant recipients being a highly selected population with less craving for alcohol than typical alcoholics, that they are a result of potential therapeutic properties of transplantation on addictions, and that they represent an underestimate because alcoholic transplant recipients hide their drinking.3, 10

Another curious observation in Dew et al.'s study8 is that the rate of alcoholic relapse, defined as either any or heavy use, seems higher in Europe than in North America, whereas the rate of nonadherence to immunosuppressives is higher in North America than Europe. As the authors point out, the geographic divergence of alcohol relapse and nonadherence, although unexplained, shows that interpretation of these phenomena amounts to medical anthropology and needs to take into account such societal forces as attitudes to alcohol use and universal access to medical care.

Three pretransplant factors were found to assist in predicting alcoholic relapse after transplantation: lack of social support, a family history of alcoholism, and less than 6 months of abstinence from alcohol. Both social support and family history are in agreement with the studies by Vaillant11 in nontransplant alcoholic populations. The 6-month period of abstinence is a more questionable prognostic tool, and here, like the authors, we must read between lines. As they point out, use of the 6-month rule to select alcoholic patients for liver transplantation is almost ubiquitous, and this exerts a profound bias on their meta-analysis because it means that the highest risk patients, those with less than 6 months of abstinence, are likely to have been excluded before the data could be collected. Data on the natural history of alcoholism do not provide much support for the 6-month abstinence concept. Valliant's longitudinal observational studies of a cohort of alcoholics found that 6 months of abstinence by itself was of little prognostic significance.11 Indeed, at least 59% of relapsing patients had already achieved 6 months of abstinence, whereas 41% of patients who achieved 2 years of abstinence relapsed. More recently, the same author has extended the interval of observation in his prospective cohorts to 60 years. The vast majority of relapses in his study occurred before the seventh year of abstinence, and by analogy to the concept of a cancer-free period as a definition of cure, he concluded that a follow-up of 5 years rather than one of 1 or 2 years would be necessary to determine stable recovery from alcoholism.12 Similarly, DiMartini et al.'s prospective study of a cohort of transplanted alcoholics,6 albeit confounded by the use of the 6-month rule, nevertheless indicates that it is a weak prognostic indicator at best. The absence of a link between pretransplant rehabilitation treatment and posttransplant alcoholic relapse is an unexpected finding in the present study. Rehabilitation support has been considered a good predictor of abstinence outside transplant.11 In order to understand whether rehabilitation has a positive or negative influence, we need to better understand how a history of rehabilitation treatment influenced selection for transplantation, something that is difficult to know from an analysis of published, largely retrospective studies.

Therefore, on the basis of Dew et al.'s study,8 what we do we know already about addictions after solid organ transplantation, and what do we need to know? First, we can conclude that addiction has been largely ignored in transplant recipients other than liver recipients. There is a need for studies of addiction in kidney, heart, and lung recipients. These studies, like future studies in liver patients, would be best if they included a prospective study design beginning before transplantation. Next, we know that relapse rates for alcoholism after liver transplantation are quite low. However, the studies included in the meta-analysis that led to this conclusion have many biases and confounders. We need studies designed to limit these biases by using study designs that recognize the difficulty of identifying alcohol or drug relapse, especially because the patients perceive that it is in their interest to conceal their behavior. One such method is to separate the study personnel from the transplant personnel and to keep revelations of alcohol or drug use confidential, that is, within the study. We know that predicting future alcohol or drug relapse remains imperfect, as is well demonstrated by the still contentious status of the 6-month abstinence rule. We need future studies that confront the issue of the suitability for transplantation of patients with duration of abstinence shorter than 6 months.