Alcohol Relapse After Liver Transplantation for Alcoholic Cirrhosis—Impact on Liver Graft and Patient Survival: A Meta-analysis

Alcohol Relapse After Liver Transplantation for Alcoholic Cirrhosis—Impact on Liver Graft and... Abstract Aim We performed meta-analysis to determine effect of alcohol relapse after liver transplantation (LT) for alcoholic cirrhosis on graft histology and survival. Methods Studies were selected using following criteria: (a) LT for alcoholic cirrhosis, (b) reporting data on liver histology and/or patient survival among relapsers and abstainers, (c) minimum follow-up of 3 years. Random effects model was used to pool data to compare relapsers and abstainers on liver histology and patient survival. Results On analysis of seven studies, pooled prevalence of self-reported alcohol relapse was 26.3% (18.0–36.7%) over median (range) follow-up of 6.0 (3.7–8.3) years, with annual alcohol relapse rate of 4.7% (3.0–6.4%) for any alcohol use and 2.9% (0.5–5.3%) for heavy alcohol use. Relapsers compared to abstainers had higher odds for graft steatosis [4.1 (2.4–6.9)], steatohepatitis [4.5 (1.4–14.2)], alcoholic hepatitis [9.3 (1.01–85)], advanced fibrosis or cirrhosis [8.4 (3.5–20)]. Relapsers were over 3-fold more likely to die at 10 years of follow-up: [3.67 (1.42–9.50)] without differences in overall or 5-year survival. Recurrent alcoholic cirrhosis occurring in 9% of biopsied patients and 2% of all transplants was responsible for about 20% of all deaths on follow-up after LT. Extra-hepatic malignancy, and cardiovascular events were common causes for patient mortality. Conclusion Alcohol relapse after LT for alcoholic cirrhosis negatively impacts the graft and long-term patient survival. Studies are needed to develop strategies to reduce alcohol relapse after LT for alcoholic cirrhosis. Short Summary Alcohol relapse in liver transplant recipients can negatively affect graft histology and patient survival. Strategies to reduce alcohol relapse are needed to preserve graft function? INTRODUCTION Alcoholic cirrhosis is a common indication for liver transplantation (LT), and accounts for about 15–40% of all the liver transplants from the US and Europe LT registries (Burra et al., 2010; Singal et al., 2013b). Five-year graft and patient survival after LT for alcoholic cirrhosis are reported to be similar compared to transplantation for other indications, and better compared to patients transplanted for HCV (Lucey et al., 2009; Burra et al., 2010; Singal et al., 2013b). Prevalence rates on alcohol relapse after LT for alcoholic cirrhosis are reported to vary from 10% to 50% (Cuadrado et al., 2005; Pfitzmann et al., 2007; Dew et al., 2008; DiMartini et al., 2010, 2011; Faure et al., 2012; Berlakovich, 2014). Among alcohol dependent subjects without liver disease, about 27–90% report relapse to alcohol use over 3–4-year period (1989; Moos and Moos, 2006; Dawson et al., 2007). Apart from the duration of follow-up after achieving remission on alcohol use and the severity of alcohol relapse (any use or slip vs. heavy alcohol use or alcohol use disorder), other risk factors determine the risk of relapse such as younger age, polysubstance use, lack of employment, low educational status, evidence of life time drinking problem and poor insight with lack of attempt/s on reducing alcohol use. Among LT recipients for alcoholic cirrhosis, data on the impact of alcohol relapse on liver graft histology remain scanty and conflicting (Pageaux et al., 1999, 2003; Bjornsson et al., 2005; Cuadrado et al., 2005; Schmeding et al., 2011; Rice et al., 2013; Egawa et al., 2014). Further, data on the impact of alcohol relapse on patient survival are variable with 10-year survival rates varying from about 22–70% (Pageaux et al., 1999, 2003; Bjornsson et al., 2005; Pfitzmann et al., 2007; Schmeding et al., 2011; Faure et al., 2012; Rice et al., 2013; Egawa et al., 2014). Data are also emerging on occurrence of severe relapse to alcohol use after LT for alcoholic cirrhosis, which is responsible for accelerated graft injury (Erard-Poinsot et al., 2016). We performed a meta-analysis of studies reporting data on alcohol relapse, liver histology and patient survival among patients transplanted for alcoholic cirrhosis. The specific aims of our study are to examine the impact of alcohol relapse on liver graft and on patient survival among transplant recipients for alcoholic cirrhosis. MATERIALS AND METHODS Identification and selection of studies Literature search was performed (until August 2016) using PubMed, Ovid, and Embase for citations that evaluated outcomes in patients transplanted for alcoholic liver disease. Search terms used were ‘alcoholic cirrhosis’ and ‘liver transplantation’. In addition, a manual search was made for cross-references from reviewed manuscripts. Search and study selection were independently performed by three investigators (SK, MK and RT). Criteria for study selection Studies were included for analysis if the following criteria were met: (a) liver transplantation for alcoholic cirrhosis, (b) data provided on alcohol relapse, (c) data provided on liver histology and or patient survival in recidivists and abstainers and (d) minimum follow-up of 3 years. Review articles, non-English citations, non-human studies and studies not meeting inclusion criteria were excluded. Any conflict was resolved with detailed review of the manuscript and discussion among all the authors. There were two studies from the same institution but these two studies were done in two different time periods and data were not duplicated (Pageaux et al., 2003; Faure et al., 2012). Of the six studies that had information on liver biopsies, five were done for clinical indications and one study had protocol-based biopsies (Cuadrado et al., 2005). None of the studies included in the final analysis reported on concomitant use of other substances, the role of polysubstance use on the outcomes could not be examined in this study. Outcome measure Primary outcome was impact of alcohol relapse on graft histology. Alcohol relapse was defined as alcohol use after LT, which was stratified to any use or heavy use of alcohol. The secondary outcome was impact of alcohol relapse on 5- and 10-year survival. Assessment of study quality Each study included in the analysis was assessed for its quality using the following parameters: (a) reported quantification of alcohol use, (b) comparison of baseline characteristics between relapsers and abstainers, (c) stratification of histology to relapsers and abstainers, (d) blinding of biopsies to relapse, (e) median follow-up greater than or equal to 5 years, (f) data provided on patient survival and (g) description of loss to follow-up and deaths. Each parameter was given a numerical score of 0 or 1 with a maximum score of 7. Studies with a score greater than or equal to 4 were considered good quality. Data collection Data were collected on sample size, patient demographics and characteristics, follow-up duration, and rate of relapse. Data were collected for both relapsers and abstainers, on liver biopsy findings, overall patient survival, and survival at 5 and 10 years, and causes of patient mortality. On liver biopsy, steatosis was defined as >5% fat, steatohepatitis as findings of fat along with inflammation and hepatocyte injury (ballooning), fibrosis defined as collagen and extracellular matrix and staging was based on standard Metavir scoring system. Data were collected independently by three different investigators (SK, MK and RT). Any conflict was resolved with detailed review of the manuscript and discussion among all the authors. Data analysis and statistical methods Data were entered using the comprehensive meta-analysis software (Biostat, Engelwood, NJ, USA). Random effects model was used to pool the data. Data on alcohol relapse was pooled for any and for heavy alcohol use. These were reported as proportion (95% confidence interval (CI)) and also as incidence (95% CI) per hundred person years of post-transplant follow-up. To determine the incidence of alcohol relapse, variable follow-up across studies was weighted for the sample size in each study, using the Poisson distribution. Median follow-up duration in the study was used, with the assumption that each subject was followed for that time period. Pooled data on liver graft histology and on patient survival were compared for relapsers and abstainers, and reported as odds ratio (OR) (95% CI). Heterogeneity across studies was reported using the I2 statistics. At least two studies were needed in analysis for determining heterogeneity. For heterogeneous data, sensitivity analyses were performed after excluding poor quality studies or after excluding two studies with the highest and lowest odds ratio from the initial analysis. Publication bias was assessed with visual inspection of funnel plots and by Egger’s regression analysis. P-value <0.05 was considered statistically significant. At least three studies were needed in the analysis to determine publication bias. We did not have manuscript available for one study to be able to access complete data. RESULTS Selection of studies Of the 1239 citations retrieved on initial search, seven studies were included in the final analysis (Fig. 1), six full manuscripts and an abstract publication (Bade and Singal, 2014). One of the studies analyzed outcomes of alcohol relapse among LT recipients for all the indications (Faure et al., 2012; Bade and Singal, 2014). The data for patients transplanted for alcoholic cirrhosis were extracted from this study and included in the analysis. Fig. 1. View largeDownload slide PRISMA flow diagram. Fig. 1. View largeDownload slide PRISMA flow diagram. Study characteristics A total of seven studies including 995 transplant recipients for alcoholic cirrhosis were analyzed (Table 1). Mean age reported in four studies (N = 505) was 51 years. About 77% (63–87%) of patients were males as reported in six studies (N = 789). The median (range) follow-up duration was 6.0 (3.7–8.3) years. Five of the seven studies were of good quality and remaining two were of poor quality (Faure et al., 2012; Bade and Singal, 2014) (Table 2). Table 1. Baseline characteristics of studies included in the analysis Author, year  No. of patients  Mean age, years  % Males  Median f/u, years  Number of liver biopsies  % biopsies in relapsers  % biopsies in abstainers  Lucey et al. (2009)  50  NA  80  5.3  33  39  61  Pageaux et al. (2003)  128  49  81  4.5  93  41  59  Cuadrado et al. (2005)  54  49  80  8.3  120  28  72  Faure et al. (2012)  206  51  NA  6.8  NA  NA  NA  Rice et al. (2013)  300  NA  73  6.5  72  33  67  Egawa et al. (2014)  140  NA  63  3.7  73  27  73  Bade and Singal (2014)  117  54  87  7.0  55  22  78  Summary  995  51  77  6  446  32  68  Author, year  No. of patients  Mean age, years  % Males  Median f/u, years  Number of liver biopsies  % biopsies in relapsers  % biopsies in abstainers  Lucey et al. (2009)  50  NA  80  5.3  33  39  61  Pageaux et al. (2003)  128  49  81  4.5  93  41  59  Cuadrado et al. (2005)  54  49  80  8.3  120  28  72  Faure et al. (2012)  206  51  NA  6.8  NA  NA  NA  Rice et al. (2013)  300  NA  73  6.5  72  33  67  Egawa et al. (2014)  140  NA  63  3.7  73  27  73  Bade and Singal (2014)  117  54  87  7.0  55  22  78  Summary  995  51  77  6  446  32  68  NA, not available. Table 2. Assessment of study quality Author, year  Quantifying alcohol use  Recidivists and abstainers comparable on baseline factors  Histology stratified to recidivists and abstainers  Biopsies blinded to exposure status  Median follow-up ≥5 years  Cohort survival over time delineated  Proportion of cohort followed over time described  Total score  Lucey et al. (2009)    ✓  ✓  ✓  ✓      4  Pageaux et al. (2003)  ✓  ✓  ✓      ✓  ✓  5  Cuadrado et al. (2005)  ✓  ✓  ✓    ✓  ✓    5  Faure et al. (2012)  ✓        ✓  ✓    3  Rice et al. (2013)  ✓    ✓    ✓  ✓    4  Egawa et al. (2014)  ✓    ✓      ✓  ✓  4  Bade and Singal (2014)      ✓    ✓  ✓    3  Author, year  Quantifying alcohol use  Recidivists and abstainers comparable on baseline factors  Histology stratified to recidivists and abstainers  Biopsies blinded to exposure status  Median follow-up ≥5 years  Cohort survival over time delineated  Proportion of cohort followed over time described  Total score  Lucey et al. (2009)    ✓  ✓  ✓  ✓      4  Pageaux et al. (2003)  ✓  ✓  ✓      ✓  ✓  5  Cuadrado et al. (2005)  ✓  ✓  ✓    ✓  ✓    5  Faure et al. (2012)  ✓        ✓  ✓    3  Rice et al. (2013)  ✓    ✓    ✓  ✓    4  Egawa et al. (2014)  ✓    ✓      ✓  ✓  4  Bade and Singal (2014)      ✓    ✓  ✓    3  Prevalence of alcohol relapse Any alcohol use A total of 264 (27%) patients in all the seven studies reported any use of alcohol on follow-up after liver transplantation with pooled prevalence of 26.3% (95% CI: 18.0–36.7%) over a median follow-up period of 6 years, with an annual incidence rate of about 4.7% (95% CI: 3.0–6.4%). Data on both these analyses were heterogeneous (I2 = 90 and 93, respectively, P < 0.0001), but without any publication bias (Egger’s P = 0.78 and 0.4, respectively). Sensitivity analysis performed after excluding two poor quality studies (Faure et al., 2012; Bade and Singal, 2014) showed pooled prevalence of 26.1% (21.3–31.6%) at 6 years and annual incidence of 4.9% (3.1–6.7%), with homogeneous data for both analyses (I2 = 39, P = 0.16 and 49, P = 0.09). Heavy alcohol use Alcohol relapse was defined as daily alcohol intake of more than 30 g/day (Cuadrado et al., 2005). Heavy drinking was defined as more than 14 units per week and/or periods of time with more than 4 units per day (Pageaux et al., 2003). In another study, continuous heavy alcohol use was defined as alcohol use that was continuous or nearly continuous after transplantation without clear intervals of sobreity (Rice et al., 2013). Amount of alcohol use was reported in only three studies. Of these, 178 (26.8%) reported any alcohol use and 94 (14.2%) used alcohol in heavy amount. Pooled prevalence of heavy alcohol use after liver transplantation was 14.5% over a median follow-up period of 6 years (95% CI: 6.1–30.9%), with an annual incidence rate of 2.9% (95% CI: 0.5–5.3%). Data on both these analyses were heterogeneous (I2 89.7 and 94.3, respectively, P < 0.0001) without any publication bias (Egger’s P = 0.06 and 0.17, respectively). Sensitivity analysis from one study after excluding two poor quality studies showed pooled prevalence of 21.1% (14.9–29.0%) and annual incidence of 4.9% (2.9–6.5%), respectively. Heterogeneity could not be assessed with only one study left in the analysis. Liver histology A total of 446 liver biopsies from six studies were analyzed to compare relapsers (N = 141) and abstainers (N = 305) on histological findings (Table 3). In one study, liver biopsies were performed when clinically indicated, however the biopsy results were not reported (Faure et al., 2012). Of the remaining five studies, biopsies in four were done for clinical indications, one study had protocol-based biopsies (Cuadrado et al., 2005). As not every study reported on all the histological lesions, the number of biopsies for different histological outcomes is variable. Table 3. Liver biopsy findings comparing recidivists and abstainers Author, year  Total biopsies  Minimal changes  Steatosis  Hepatitis  Cholestasis  ACR  Steatohepatitis  Alcoholic hepatitis  F1–F2 fibrosis  F3–F4 Indication Fibrosis  R  A  R  A  R  A  R  A  R  A  R  A  R  A  R  A  R  A  R  A  Lucey et al. (2009)  13  20  4  4  0  1  8  13  –  –  4  8  2  3  1  0  –  –  –  –clinical  Pageaux et al. (2003)  38  55  7  12  15  9  –  –  –  –  –  –  –  –  3  0  8  8  2  0 clinical  Cuadrado et al. (2005)  34  86  15  40  5  4  –  –  6  9  5  28  –  –  0  0  –  –  0  0 protocol  Rice et al. (2013)  24  48  –  –  14  12  –  –  –  12  23  –  10  4  –  –  –  –  15  9 clinical  Egawa et al. (2014)  20  53  2  10  9  7  1  10  0  4  –  –  –  –  –  –  0  0  NA  NA NA  Bade and Singal (2014)  12  43  –  –  –  –  –  –  –  –  –  –  3  2  –  –  –  –  2  7 NA  Total  141  305  28  66  43  33  9  23  6  13  9  36  15  9  4  0  8  8  19  18  Author, year  Total biopsies  Minimal changes  Steatosis  Hepatitis  Cholestasis  ACR  Steatohepatitis  Alcoholic hepatitis  F1–F2 fibrosis  F3–F4 Indication Fibrosis  R  A  R  A  R  A  R  A  R  A  R  A  R  A  R  A  R  A  R  A  Lucey et al. (2009)  13  20  4  4  0  1  8  13  –  –  4  8  2  3  1  0  –  –  –  –clinical  Pageaux et al. (2003)  38  55  7  12  15  9  –  –  –  –  –  –  –  –  3  0  8  8  2  0 clinical  Cuadrado et al. (2005)  34  86  15  40  5  4  –  –  6  9  5  28  –  –  0  0  –  –  0  0 protocol  Rice et al. (2013)  24  48  –  –  14  12  –  –  –  12  23  –  10  4  –  –  –  –  15  9 clinical  Egawa et al. (2014)  20  53  2  10  9  7  1  10  0  4  –  –  –  –  –  –  0  0  NA  NA NA  Bade and Singal (2014)  12  43  –  –  –  –  –  –  –  –  –  –  3  2  –  –  –  –  2  7 NA  Total  141  305  28  66  43  33  9  23  6  13  9  36  15  9  4  0  8  8  19  18  R = recidivists, A = abstainers hepatitis includes histological categories of mild hepatitis, progressive hepatitis and lobular hepatitis as reported by studies; F1–F2: Stage 1 or 2 Metavir fibrosis stage; F3–F4: bridging fibrosis or cirrhosis. Steatosis Of 391 liver biopsies (129 recidivists) from five studies, steatosis was reported on 76 (43 relapsers) biopsies. Relapsers as compared to 262 abstainers, had higher prevalence of steatosis (33% vs. 13%), with over 4-fold risk for presence of steatosis among recidivists [4.08 (2.41–6.91)]. Data were homogeneous (I2 = 0, P = 0.53) without publication bias (Egger’s P = 0.78). Steatohepatitis Of 160 biopsies (49 recidivists) from three studies, steatohepatitis was reported on 24 (15 relapsers) biopsies. Relapsers as compared to 111 abstainers had higher prevalence of steatohepatitis (31% vs. 8%) with over 4-fold risk for presence of steatohepatitis among relapsers, [4.5 (1.4–14.2)], P = 0.01 (Fig. 2A). Data were homogeneous (I2 = 28, P = 0.25) without any publication bias (Egger’s P = 0.54). Fig. 2. View largeDownload slide Forrest plots showing pooled data on liver biopsy findings comparing abstainers and recidivists on the risk of (A) steatohepatitis, (B) Alcoholic hepatitis and (C) Advanced fibrosis or cirrhosis. Fig. 2. View largeDownload slide Forrest plots showing pooled data on liver biopsy findings comparing abstainers and recidivists on the risk of (A) steatohepatitis, (B) Alcoholic hepatitis and (C) Advanced fibrosis or cirrhosis. Of 246 biopsies (85 recidivists) analyzing on alcoholic hepatitis in four studies, one study did not report any case of alcoholic hepatitis, leaving three studies for final analysis. Relapsers as compared to 161 abstainers had higher prevalence of alcoholic hepatitis changes on liver biopsy (5% vs. 0%) with over 9-fold risk for alcoholic hepatitis among relapsers, [9.3 (1.01–85)], P = 0.049 (Fig. 2B). Data were homogeneous (I2 = 0, P = 0.61). Publication bias could not be assessed with only two studies in the analysis. Fibrosis Of 340 biopsies (108 recidivists) from four studies, 37 (19 relapsers) reported presence of advanced fibrosis or cirrhosis. Relapsers as compared to 232 abstainers had higher prevalence of advanced fibrosis or cirrhosis (18% vs. 8%) with over 8-fold risk for advanced fibrosis or cirrhosis, [8.4 (3.5–20)], P < 0.0001 (Fig. 2C). Data were homogeneous (I2 = 0, P = 0.87) without any publication bias (Egger’s P = 0.76). Patient survival A total of 240 patients died on a median follow-up of about 6.0 years after LT. Five-year survival reported on 824 recipients was similar comparing 224 relapsers and 600 abstainers (86% vs. 88%), OR (95% CI) of 1.2 (0.7–2.1), (Fig. 3A). Data were homogeneous (I2 = 20, P = 0.29) without any publication bias (Egger’s P = 0.89). Ten-year survival reported on 692 recipients in four studies was worse among relapsers (N = 184) compared to 508 abstainers (45.2% vs. 65.4%), with over 3-fold risk of death among relapsers: 3.67 (1.42–9.50) (Fig. 3B). Data were heterogeneous (I2 = 82, P = 0.001) without any publication bias (Egger’s P = 0.28). With only one of four studies of poor quality in the initial analysis, sensitivity analysis performed after excluding two studies with highest and lowest OR showed homogeneous data (I2 = 36, P = 0.21), with similar effect size: 3.8 (1.6–8.9). Publication bias could not be assessed with only two studies left in the analysis. Fig. 3. View largeDownload slide Forrest plots showing pooled data on comparing abstainers and recidivists on (A) 5 years and (B) 10-year patient survival after liver transplantation for alcoholic cirrhosis. Fig. 3. View largeDownload slide Forrest plots showing pooled data on comparing abstainers and recidivists on (A) 5 years and (B) 10-year patient survival after liver transplantation for alcoholic cirrhosis. Causes of patient mortality Of 240 patients dying on post-transplant follow-up, causes of death were reported in 175 patients (Table 4). Of these, 19% died from liver failure related to continued alcohol use. More common causes were malignancy in 42% (non-HCC malignancy in 35%), cardiovascular-related mortality in 17%, and infections in 21%. Remaining patients died from liver failure due to graft rejection (Table 4). Table 4. Causes of patient mortality Author, year  Graft rejection  Infection or sepsis  Cardiovascular  Liver failure  HCC  Non-HCC cancers  T  R  A  T  R  A  T  R  A  T  R  A  T  R  A  T  R  A  Lucey et al. (2009)  1  1  0  0  0  0  0  0  0  0  0  0  1  0  1  0  0  0  Pageaux et al. (2003)  3  1  2  2  1  1  3  1  2  0  0  0  2  0  2  7  3  4  Cuadrado et al. (2005)  0  0  0  2  1  1  3  2  1  0  0  0  1  0  1  4  2  2  Faure et al. (2012)  2  –  –  6  –  –  8  –  –  13  –  –  –  –  –  18  –  –  Rice et al. (2013)  –  –  –  11  2  9  15  2  13  19  7  12  3  0  3  30  4  26  Egawa et al. (2014)  1  –  –  16  –  –  1  –  –  2  –  –  5  –  –  2  –  –  Total  7  2  2  37  4  11  30  5  16  34  7  12  12  0  7  61  9  32  Author, year  Graft rejection  Infection or sepsis  Cardiovascular  Liver failure  HCC  Non-HCC cancers  T  R  A  T  R  A  T  R  A  T  R  A  T  R  A  T  R  A  Lucey et al. (2009)  1  1  0  0  0  0  0  0  0  0  0  0  1  0  1  0  0  0  Pageaux et al. (2003)  3  1  2  2  1  1  3  1  2  0  0  0  2  0  2  7  3  4  Cuadrado et al. (2005)  0  0  0  2  1  1  3  2  1  0  0  0  1  0  1  4  2  2  Faure et al. (2012)  2  –  –  6  –  –  8  –  –  13  –  –  –  –  –  18  –  –  Rice et al. (2013)  –  –  –  11  2  9  15  2  13  19  7  12  3  0  3  30  4  26  Egawa et al. (2014)  1  –  –  16  –  –  1  –  –  2  –  –  5  –  –  2  –  –  Total  7  2  2  37  4  11  30  5  16  34  7  12  12  0  7  61  9  32  DISCUSSION Main findings of this meta-analysis are that alcohol relapse after liver transplantation for alcoholic cirrhosis (a) occurs at a rate of about 4.7% per year for any alcohol use and about 2.9% for heavy alcohol use, (b) negatively impacts the liver graft with risk for development of steatohepatitis, alcoholic hepatitis, and advanced fibrosis or cirrhosis; and (c) negatively impacts the long-term survival. Further, malignancy, graft failure and cardiovascular causes are common causes of patient mortality. Self-reported alcohol relapse rates after liver transplantation for alcoholic cirrhosis vary from 10% to 50% (Pageaux et al., 1999; Cuadrado et al., 2005; DiMartini et al., 2010). Pooled data of about 26% alcohol relapse rate in this meta-analysis is consistent with the published literature. Alcohol relapse commonly occurs at 2 years after transplantation and then increases with follow-up time (DiMartini et al., 2010; Russ et al., 2016). In a meta-analysis of 50 studies, annual incidence of alcohol relapse was 5.7% for any alcohol use and 2.5% for heavy alcohol use, similar to incidence rates in this meta-analysis (Dew et al., 2008). Among patients with alcohol use disorder without liver disease or receiving any solid organ transplantation, about 90% after achieving remission report at least one alcohol relapse over 4-year period, as reported by the National Institute on Alcohol Abuse and Alcoholism (1989). In one study, alcohol relapse with alcohol use disorder was reported in about 51% among at risk drinkers, 27% among low risk drinkers and about 7% among abstainers (Dawson et al., 2007). In another study, over a long-term period of 16 years, patients with alcohol use disorder entering into remission with the use of addiction therapy were less likely to report alcohol relapse compared to remitters without obtaining any help for alcohol use disorder (43% vs. 61%, P < 0.01) (Moos and Moos, 2006). Other variables influencing relapse rates are reported to be age, psychosocial functioning, educational status, employment status and life time drinking problem (Vaillant et al., 1991; McCallum and Masterton, 2006; Moos and Moos, 2006). Lack of information on these variables in our study limited our assessment of risk factors predictive of relapse to alcohol use among recipients of LT for alcoholic cirrhosis. Prevalence of steatosis and steatohepatitis was reported in about 33% and 31% of recidivists undergoing liver biopsy evaluation. There is likely selection bias on these findings as these liver biopsies were driven by clinical indication and were not protocol biopsies. Presence of steatosis and steatohepatitis in 13% and 8% of abstainers may likely be due to post-transplant metabolic syndrome or inaccurate self-report on alcohol use (Watt and Charlton, 2010; Singal et al., 2013a). Advanced fibrosis or cirrhosis was present in 18% of recidivists in this study. In a recent study, about 32% relapsers showed biopsy changes of advanced fibrosis or cirrhosis on liver biopsy. Protocol biopsies performed in this study, relatively longer duration of about 9 years, and examining only patients with heavy alcohol use likely explains difference between the studies. Overall recurrent alcoholic cirrhosis occurs in about 5–8% of the whole cohort of liver transplant recipients for alcoholic cirrhosis (Dumortier et al., 2015; Verbeek et al., 2016). In our current analysis, recurrent alcoholic cirrhosis occurred in about 2% of all transplant recipients, about 9% of all biopsied patients, and about 20% of relapsers with heavy alcohol use. Compared to abstainers, relapsers had similar (86% vs. 88%) survival rates at 5 years and lower survival (45% vs. 65%) at 10 years after liver transplantation. These results are similar to a recently reported prospective study on 712 (128 relapsers) patients transplanted for alcoholic cirrhosis, with survival rates comparing relapsers and abstainers at 5 and 10 years of follow-up of 88% vs. 89%, P > 0.05 and 49% vs. 70%, P < 0.005, respectively (Dumortier et al., 2015). Extra-hepatic malignancy and cardiovascular events were common causes of patient mortality, as reported earlier from single center and on analysis of the NIDDK LT database. Development of malignancy among transplanted patients for alcoholic cirrhosis is associated with smoking (Doycheva et al., 2016; Verbeek et al., 2016). However, this could not be analyzed in this analysis due to lack of information on smoking in most studies. Graft failure accounted for about 20% of all the deaths in this pooled data. The data pooled from seven studies with large sample size and stringent criteria for study selection with liver histology comparison of relapsers and abstainers are strengths of our analysis. However, our study suffers from some limitations. For example, all the studies included in this analysis were retrospective or observational with variations on follow-up period and definition of relapse. Also, alcohol use, diagnosis and variability of selection between teams is a limitation of our study. Further, alcohol consumption in all these studies was self-reported, which may often be inaccurate underestimating the true prevalence of alcohol use (Bajaj et al., 2008; Carbonneau et al., 2010). Moreover, biopsies in all studies were performed for clinical indications and were not protocol biopsies, and thus a potential for selection bias. Not all patients studied had biopsies and only 45% (446 out of 995) had liver biopsies. In summary, post-transplant alcohol relapse negatively impacts the liver graft and long-term patient survival among transplant recipients for alcoholic cirrhosis. Our study findings highlight the need and importance of maintaining alcohol abstinence after LT. Well-designed prospective studies are needed for developing strategies to maintain abstinence among LT recipients for alcoholic cirrhosis. FUNDING The study is supported by faculty development grant from the American College of Gastroenterology and R21 grant from the National Institute of Alcohol Abuse and Alcoholism (1R21AA023273-01A1), awarded to Ashwani K. Singal. CONFLICT OF INTEREST STATEMENT None declared. REFERENCES ALCOHOLISM. ( 1989) Relapse and craving. National Institute on Alcohol Abuse and Alcoholism. https://pubs.niaaa.nih.gov/publications/aa06.htm. Bade KS, Singal AK. ( 2014) Recidivism among liver transplant recipients for alcoholic cirrhosis: prevalence, predictors, and impact on outcomes. Am J Gastroenterol  109: S164. Bajaj JS, Saeian K, Hafeezullah M, et al.  . ( 2008) Failure to fully disclose during pretransplant psychological evaluation in alcoholic liver disease: a driving under the influence corroboration study. Liver Transpl  14: 1632– 6. Google Scholar CrossRef Search ADS PubMed  Berlakovich GA. ( 2014) Challenges in transplantation for alcoholic liver disease. World J Gastroenterol  20: 8033– 9. Google Scholar CrossRef Search ADS PubMed  Bjornsson E, Olsson J, Rydell A, et al.  . ( 2005) Long-term follow-up of patients with alcoholic liver disease after liver transplantation in Sweden: impact of structured management on recidivism. Scand J Gastroenterol  40: 206– 16. Google Scholar CrossRef Search ADS PubMed  Burra P, Senzolo M, Adam R, et al.   ELITA & CENTERS, E. L. T. ( 2010) Liver transplantation for alcoholic liver disease in Europe: a study from the ELTR (European Liver Transplant Registry). Am J Transplant  10: 138– 48. Google Scholar CrossRef Search ADS PubMed  Carbonneau M, Jensen LA, Bain VG, et al.  . ( 2010) Alcohol use while on the liver transplant waiting list: a single-center experience. Liver Transpl  16: 91– 7. Google Scholar CrossRef Search ADS PubMed  Cuadrado A, Fabrega E, Casafont F, et al.  . ( 2005) Alcohol recidivism impairs long-term patient survival after orthotopic liver transplantation for alcoholic liver disease. Liver Transpl  11: 420– 6. Google Scholar CrossRef Search ADS PubMed  Dawson DA, Goldstein RB, Grant BF. ( 2007) Rates and correlates of relapse among individuals in remission from DSM-IV alcohol dependence: a 3-year follow-up. Alcohol Clin Exp Res  31: 2036– 45. Google Scholar CrossRef Search ADS PubMed  Dew MA, DiMartini AF, Steel J, et al.  . ( 2008) Meta-analysis of risk for relapse to substance use after transplantation of the liver or other solid organs. Liver Transpl  14: 159– 72. Google Scholar CrossRef Search ADS PubMed  DiMartini A, DEW MA, DAY N, et al.  . ( 2010) Trajectories of alcohol consumption following liver transplantation. Am J Transplant  10: 2305– 12. Google Scholar CrossRef Search ADS PubMed  DiMartini A, Crone C, Dew MA. ( 2011) Alcohol and substance use in liver transplant patients. Clin Liver Dis  15: 727– 51. Google Scholar CrossRef Search ADS PubMed  Doycheva I, Amer S, Watt KD. ( 2016) De Novo malignancies after transplantation: risk and surveillance strategies. Med Clin North Am  100: 551– 67. Google Scholar CrossRef Search ADS PubMed  Dumortier J, Dharancy S, Cannesson A, et al.  . ( 2015) Recurrent alcoholic cirrhosis in severe alcoholic relapse after liver transplantation: a frequent and serious complication. Am J Gastroenterol  110: 1160– 6. quiz 1167. Google Scholar CrossRef Search ADS PubMed  Egawa H, Nishimura K, Teramukai S, et al.  . ( 2014) Risk factors for alcohol relapse after liver transplantation for alcoholic cirrhosis in Japan. Liver Transpl  20: 298– 310. Google Scholar CrossRef Search ADS PubMed  Erard-Poinsot D, Guillaud O, Hervieu V, et al.  . ( 2016) Severe alcoholic relapse after liver transplantation: what consequences on the graft? A study based on liver biopsies analysis. Liver Transpl  22: 773– 84. Google Scholar CrossRef Search ADS PubMed  Faure S, Herrero A, Jung B, et al.  . ( 2012) Excessive alcohol consumption after liver transplantation impacts on long-term survival, whatever the primary indication. J Hepatol  57: 306– 12. Google Scholar CrossRef Search ADS PubMed  Lucey MR, Schaubel DE, Guidinger MK, et al.  . ( 2009) Effect of alcoholic liver disease and hepatitis C infection on waiting list and posttransplant mortality and transplant survival benefit. Hepatology  50: 400– 6. Google Scholar CrossRef Search ADS PubMed  McCallum S, Masterton G. ( 2006) Liver transplantation for alcoholic liver disease: a systematic review of psychosocial selection criteria. Alcohol Alcohol  41: 358– 63. Google Scholar CrossRef Search ADS PubMed  Moos RH, Moos BS. ( 2006) Rates and predictors of relapse after natural and treated remission from alcohol use disorders. Addiction  101: 212– 22. Google Scholar CrossRef Search ADS PubMed  Pageaux GP, Michel J, Coste V, et al.  . ( 1999) Alcoholic cirrhosis is a good indication for liver transplantation, even for cases of recidivism. Gut  45: 421– 6. Google Scholar CrossRef Search ADS PubMed  Pageaux G-P, Bismuth M, Perney P, et al.  . ( 2003) Alcohol relapse after liver transplantation for alcoholic liver disease: does it matter? J Hepatol  38: 629– 34. Google Scholar CrossRef Search ADS PubMed  Pfitzmann R, Schwenzer J, Rayes N, et al.  . ( 2007) Long-term survival and predictors of relapse after orthotopic liver transplantation for alcoholic liver disease. Liver Transpl  13: 197– 205. Google Scholar CrossRef Search ADS PubMed  Rice JP, Eickhoff J, Agni R, et al.  . ( 2013) Abusive drinking after liver transplantation is associated with allograft loss and advanced allograft fibrosis. Liver Transpl  19: 1377– 86. Google Scholar CrossRef Search ADS PubMed  Russ KB, Chen NW, Kamath PS, et al.  . ( 2016) Alcohol use after liver transplantation is independent of liver disease etiology. Alcohol Alcohol  6: 6. Schmeding M, Heidenhain C, Neuhaus R, et al.  . ( 2011) Liver transplantation for alcohol-related cirrhosis: a single centre long-term clinical and histological follow-up. Dig Dis Sci  56: 236– 43. Google Scholar CrossRef Search ADS PubMed  Singal AK, Chaha KS, Rasheed K, et al.  . ( 2013a) Liver transplantation in alcoholic liver disease current status and controversies. World J Gastroenterol  19: 5953– 63. Google Scholar CrossRef Search ADS PubMed  Singal AK, Guturu P, Hmoud B, et al.  . ( 2013b) Evolving frequency and outcomes of liver transplantation based on etiology of liver disease. Transplantation  95: 755– 60. Google Scholar CrossRef Search ADS PubMed  Vaillant GE, Schnurr PP, Baron JA, et al.  . ( 1991) A prospective study of the effects of cigarette smoking and alcohol abuse on mortality. J Gen Intern Med  6: 299– 304. Google Scholar CrossRef Search ADS PubMed  Verbeek J, Cassiman D, Nevens F. ( 2016) De novo malignancy and recurrent alcoholic cirrhosis account for 70% of deaths in patients transplanted for end-stage alcoholic liver disease. Am J Gastroenterol  111: 436– 7. Google Scholar CrossRef Search ADS PubMed  Watt KD, Charlton MR. ( 2010) Metabolic syndrome and liver transplantation: a review and guide to management. J Hepatol  53: 199– 206. Google Scholar CrossRef Search ADS PubMed  © The Author 2017. Medical Council on Alcohol and Oxford University Press. All rights reserved. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Alcohol and Alcoholism Oxford University Press

Alcohol Relapse After Liver Transplantation for Alcoholic Cirrhosis—Impact on Liver Graft and Patient Survival: A Meta-analysis

Loading next page...
 
/lp/ou_press/alcohol-relapse-after-liver-transplantation-for-alcoholic-cirrhosis-OYHNtb4r0q
Publisher
Oxford University Press
Copyright
© The Author 2017. Medical Council on Alcohol and Oxford University Press. All rights reserved.
ISSN
0735-0414
eISSN
1464-3502
D.O.I.
10.1093/alcalc/agx098
Publisher site
See Article on Publisher Site

Abstract

Abstract Aim We performed meta-analysis to determine effect of alcohol relapse after liver transplantation (LT) for alcoholic cirrhosis on graft histology and survival. Methods Studies were selected using following criteria: (a) LT for alcoholic cirrhosis, (b) reporting data on liver histology and/or patient survival among relapsers and abstainers, (c) minimum follow-up of 3 years. Random effects model was used to pool data to compare relapsers and abstainers on liver histology and patient survival. Results On analysis of seven studies, pooled prevalence of self-reported alcohol relapse was 26.3% (18.0–36.7%) over median (range) follow-up of 6.0 (3.7–8.3) years, with annual alcohol relapse rate of 4.7% (3.0–6.4%) for any alcohol use and 2.9% (0.5–5.3%) for heavy alcohol use. Relapsers compared to abstainers had higher odds for graft steatosis [4.1 (2.4–6.9)], steatohepatitis [4.5 (1.4–14.2)], alcoholic hepatitis [9.3 (1.01–85)], advanced fibrosis or cirrhosis [8.4 (3.5–20)]. Relapsers were over 3-fold more likely to die at 10 years of follow-up: [3.67 (1.42–9.50)] without differences in overall or 5-year survival. Recurrent alcoholic cirrhosis occurring in 9% of biopsied patients and 2% of all transplants was responsible for about 20% of all deaths on follow-up after LT. Extra-hepatic malignancy, and cardiovascular events were common causes for patient mortality. Conclusion Alcohol relapse after LT for alcoholic cirrhosis negatively impacts the graft and long-term patient survival. Studies are needed to develop strategies to reduce alcohol relapse after LT for alcoholic cirrhosis. Short Summary Alcohol relapse in liver transplant recipients can negatively affect graft histology and patient survival. Strategies to reduce alcohol relapse are needed to preserve graft function? INTRODUCTION Alcoholic cirrhosis is a common indication for liver transplantation (LT), and accounts for about 15–40% of all the liver transplants from the US and Europe LT registries (Burra et al., 2010; Singal et al., 2013b). Five-year graft and patient survival after LT for alcoholic cirrhosis are reported to be similar compared to transplantation for other indications, and better compared to patients transplanted for HCV (Lucey et al., 2009; Burra et al., 2010; Singal et al., 2013b). Prevalence rates on alcohol relapse after LT for alcoholic cirrhosis are reported to vary from 10% to 50% (Cuadrado et al., 2005; Pfitzmann et al., 2007; Dew et al., 2008; DiMartini et al., 2010, 2011; Faure et al., 2012; Berlakovich, 2014). Among alcohol dependent subjects without liver disease, about 27–90% report relapse to alcohol use over 3–4-year period (1989; Moos and Moos, 2006; Dawson et al., 2007). Apart from the duration of follow-up after achieving remission on alcohol use and the severity of alcohol relapse (any use or slip vs. heavy alcohol use or alcohol use disorder), other risk factors determine the risk of relapse such as younger age, polysubstance use, lack of employment, low educational status, evidence of life time drinking problem and poor insight with lack of attempt/s on reducing alcohol use. Among LT recipients for alcoholic cirrhosis, data on the impact of alcohol relapse on liver graft histology remain scanty and conflicting (Pageaux et al., 1999, 2003; Bjornsson et al., 2005; Cuadrado et al., 2005; Schmeding et al., 2011; Rice et al., 2013; Egawa et al., 2014). Further, data on the impact of alcohol relapse on patient survival are variable with 10-year survival rates varying from about 22–70% (Pageaux et al., 1999, 2003; Bjornsson et al., 2005; Pfitzmann et al., 2007; Schmeding et al., 2011; Faure et al., 2012; Rice et al., 2013; Egawa et al., 2014). Data are also emerging on occurrence of severe relapse to alcohol use after LT for alcoholic cirrhosis, which is responsible for accelerated graft injury (Erard-Poinsot et al., 2016). We performed a meta-analysis of studies reporting data on alcohol relapse, liver histology and patient survival among patients transplanted for alcoholic cirrhosis. The specific aims of our study are to examine the impact of alcohol relapse on liver graft and on patient survival among transplant recipients for alcoholic cirrhosis. MATERIALS AND METHODS Identification and selection of studies Literature search was performed (until August 2016) using PubMed, Ovid, and Embase for citations that evaluated outcomes in patients transplanted for alcoholic liver disease. Search terms used were ‘alcoholic cirrhosis’ and ‘liver transplantation’. In addition, a manual search was made for cross-references from reviewed manuscripts. Search and study selection were independently performed by three investigators (SK, MK and RT). Criteria for study selection Studies were included for analysis if the following criteria were met: (a) liver transplantation for alcoholic cirrhosis, (b) data provided on alcohol relapse, (c) data provided on liver histology and or patient survival in recidivists and abstainers and (d) minimum follow-up of 3 years. Review articles, non-English citations, non-human studies and studies not meeting inclusion criteria were excluded. Any conflict was resolved with detailed review of the manuscript and discussion among all the authors. There were two studies from the same institution but these two studies were done in two different time periods and data were not duplicated (Pageaux et al., 2003; Faure et al., 2012). Of the six studies that had information on liver biopsies, five were done for clinical indications and one study had protocol-based biopsies (Cuadrado et al., 2005). None of the studies included in the final analysis reported on concomitant use of other substances, the role of polysubstance use on the outcomes could not be examined in this study. Outcome measure Primary outcome was impact of alcohol relapse on graft histology. Alcohol relapse was defined as alcohol use after LT, which was stratified to any use or heavy use of alcohol. The secondary outcome was impact of alcohol relapse on 5- and 10-year survival. Assessment of study quality Each study included in the analysis was assessed for its quality using the following parameters: (a) reported quantification of alcohol use, (b) comparison of baseline characteristics between relapsers and abstainers, (c) stratification of histology to relapsers and abstainers, (d) blinding of biopsies to relapse, (e) median follow-up greater than or equal to 5 years, (f) data provided on patient survival and (g) description of loss to follow-up and deaths. Each parameter was given a numerical score of 0 or 1 with a maximum score of 7. Studies with a score greater than or equal to 4 were considered good quality. Data collection Data were collected on sample size, patient demographics and characteristics, follow-up duration, and rate of relapse. Data were collected for both relapsers and abstainers, on liver biopsy findings, overall patient survival, and survival at 5 and 10 years, and causes of patient mortality. On liver biopsy, steatosis was defined as >5% fat, steatohepatitis as findings of fat along with inflammation and hepatocyte injury (ballooning), fibrosis defined as collagen and extracellular matrix and staging was based on standard Metavir scoring system. Data were collected independently by three different investigators (SK, MK and RT). Any conflict was resolved with detailed review of the manuscript and discussion among all the authors. Data analysis and statistical methods Data were entered using the comprehensive meta-analysis software (Biostat, Engelwood, NJ, USA). Random effects model was used to pool the data. Data on alcohol relapse was pooled for any and for heavy alcohol use. These were reported as proportion (95% confidence interval (CI)) and also as incidence (95% CI) per hundred person years of post-transplant follow-up. To determine the incidence of alcohol relapse, variable follow-up across studies was weighted for the sample size in each study, using the Poisson distribution. Median follow-up duration in the study was used, with the assumption that each subject was followed for that time period. Pooled data on liver graft histology and on patient survival were compared for relapsers and abstainers, and reported as odds ratio (OR) (95% CI). Heterogeneity across studies was reported using the I2 statistics. At least two studies were needed in analysis for determining heterogeneity. For heterogeneous data, sensitivity analyses were performed after excluding poor quality studies or after excluding two studies with the highest and lowest odds ratio from the initial analysis. Publication bias was assessed with visual inspection of funnel plots and by Egger’s regression analysis. P-value <0.05 was considered statistically significant. At least three studies were needed in the analysis to determine publication bias. We did not have manuscript available for one study to be able to access complete data. RESULTS Selection of studies Of the 1239 citations retrieved on initial search, seven studies were included in the final analysis (Fig. 1), six full manuscripts and an abstract publication (Bade and Singal, 2014). One of the studies analyzed outcomes of alcohol relapse among LT recipients for all the indications (Faure et al., 2012; Bade and Singal, 2014). The data for patients transplanted for alcoholic cirrhosis were extracted from this study and included in the analysis. Fig. 1. View largeDownload slide PRISMA flow diagram. Fig. 1. View largeDownload slide PRISMA flow diagram. Study characteristics A total of seven studies including 995 transplant recipients for alcoholic cirrhosis were analyzed (Table 1). Mean age reported in four studies (N = 505) was 51 years. About 77% (63–87%) of patients were males as reported in six studies (N = 789). The median (range) follow-up duration was 6.0 (3.7–8.3) years. Five of the seven studies were of good quality and remaining two were of poor quality (Faure et al., 2012; Bade and Singal, 2014) (Table 2). Table 1. Baseline characteristics of studies included in the analysis Author, year  No. of patients  Mean age, years  % Males  Median f/u, years  Number of liver biopsies  % biopsies in relapsers  % biopsies in abstainers  Lucey et al. (2009)  50  NA  80  5.3  33  39  61  Pageaux et al. (2003)  128  49  81  4.5  93  41  59  Cuadrado et al. (2005)  54  49  80  8.3  120  28  72  Faure et al. (2012)  206  51  NA  6.8  NA  NA  NA  Rice et al. (2013)  300  NA  73  6.5  72  33  67  Egawa et al. (2014)  140  NA  63  3.7  73  27  73  Bade and Singal (2014)  117  54  87  7.0  55  22  78  Summary  995  51  77  6  446  32  68  Author, year  No. of patients  Mean age, years  % Males  Median f/u, years  Number of liver biopsies  % biopsies in relapsers  % biopsies in abstainers  Lucey et al. (2009)  50  NA  80  5.3  33  39  61  Pageaux et al. (2003)  128  49  81  4.5  93  41  59  Cuadrado et al. (2005)  54  49  80  8.3  120  28  72  Faure et al. (2012)  206  51  NA  6.8  NA  NA  NA  Rice et al. (2013)  300  NA  73  6.5  72  33  67  Egawa et al. (2014)  140  NA  63  3.7  73  27  73  Bade and Singal (2014)  117  54  87  7.0  55  22  78  Summary  995  51  77  6  446  32  68  NA, not available. Table 2. Assessment of study quality Author, year  Quantifying alcohol use  Recidivists and abstainers comparable on baseline factors  Histology stratified to recidivists and abstainers  Biopsies blinded to exposure status  Median follow-up ≥5 years  Cohort survival over time delineated  Proportion of cohort followed over time described  Total score  Lucey et al. (2009)    ✓  ✓  ✓  ✓      4  Pageaux et al. (2003)  ✓  ✓  ✓      ✓  ✓  5  Cuadrado et al. (2005)  ✓  ✓  ✓    ✓  ✓    5  Faure et al. (2012)  ✓        ✓  ✓    3  Rice et al. (2013)  ✓    ✓    ✓  ✓    4  Egawa et al. (2014)  ✓    ✓      ✓  ✓  4  Bade and Singal (2014)      ✓    ✓  ✓    3  Author, year  Quantifying alcohol use  Recidivists and abstainers comparable on baseline factors  Histology stratified to recidivists and abstainers  Biopsies blinded to exposure status  Median follow-up ≥5 years  Cohort survival over time delineated  Proportion of cohort followed over time described  Total score  Lucey et al. (2009)    ✓  ✓  ✓  ✓      4  Pageaux et al. (2003)  ✓  ✓  ✓      ✓  ✓  5  Cuadrado et al. (2005)  ✓  ✓  ✓    ✓  ✓    5  Faure et al. (2012)  ✓        ✓  ✓    3  Rice et al. (2013)  ✓    ✓    ✓  ✓    4  Egawa et al. (2014)  ✓    ✓      ✓  ✓  4  Bade and Singal (2014)      ✓    ✓  ✓    3  Prevalence of alcohol relapse Any alcohol use A total of 264 (27%) patients in all the seven studies reported any use of alcohol on follow-up after liver transplantation with pooled prevalence of 26.3% (95% CI: 18.0–36.7%) over a median follow-up period of 6 years, with an annual incidence rate of about 4.7% (95% CI: 3.0–6.4%). Data on both these analyses were heterogeneous (I2 = 90 and 93, respectively, P < 0.0001), but without any publication bias (Egger’s P = 0.78 and 0.4, respectively). Sensitivity analysis performed after excluding two poor quality studies (Faure et al., 2012; Bade and Singal, 2014) showed pooled prevalence of 26.1% (21.3–31.6%) at 6 years and annual incidence of 4.9% (3.1–6.7%), with homogeneous data for both analyses (I2 = 39, P = 0.16 and 49, P = 0.09). Heavy alcohol use Alcohol relapse was defined as daily alcohol intake of more than 30 g/day (Cuadrado et al., 2005). Heavy drinking was defined as more than 14 units per week and/or periods of time with more than 4 units per day (Pageaux et al., 2003). In another study, continuous heavy alcohol use was defined as alcohol use that was continuous or nearly continuous after transplantation without clear intervals of sobreity (Rice et al., 2013). Amount of alcohol use was reported in only three studies. Of these, 178 (26.8%) reported any alcohol use and 94 (14.2%) used alcohol in heavy amount. Pooled prevalence of heavy alcohol use after liver transplantation was 14.5% over a median follow-up period of 6 years (95% CI: 6.1–30.9%), with an annual incidence rate of 2.9% (95% CI: 0.5–5.3%). Data on both these analyses were heterogeneous (I2 89.7 and 94.3, respectively, P < 0.0001) without any publication bias (Egger’s P = 0.06 and 0.17, respectively). Sensitivity analysis from one study after excluding two poor quality studies showed pooled prevalence of 21.1% (14.9–29.0%) and annual incidence of 4.9% (2.9–6.5%), respectively. Heterogeneity could not be assessed with only one study left in the analysis. Liver histology A total of 446 liver biopsies from six studies were analyzed to compare relapsers (N = 141) and abstainers (N = 305) on histological findings (Table 3). In one study, liver biopsies were performed when clinically indicated, however the biopsy results were not reported (Faure et al., 2012). Of the remaining five studies, biopsies in four were done for clinical indications, one study had protocol-based biopsies (Cuadrado et al., 2005). As not every study reported on all the histological lesions, the number of biopsies for different histological outcomes is variable. Table 3. Liver biopsy findings comparing recidivists and abstainers Author, year  Total biopsies  Minimal changes  Steatosis  Hepatitis  Cholestasis  ACR  Steatohepatitis  Alcoholic hepatitis  F1–F2 fibrosis  F3–F4 Indication Fibrosis  R  A  R  A  R  A  R  A  R  A  R  A  R  A  R  A  R  A  R  A  Lucey et al. (2009)  13  20  4  4  0  1  8  13  –  –  4  8  2  3  1  0  –  –  –  –clinical  Pageaux et al. (2003)  38  55  7  12  15  9  –  –  –  –  –  –  –  –  3  0  8  8  2  0 clinical  Cuadrado et al. (2005)  34  86  15  40  5  4  –  –  6  9  5  28  –  –  0  0  –  –  0  0 protocol  Rice et al. (2013)  24  48  –  –  14  12  –  –  –  12  23  –  10  4  –  –  –  –  15  9 clinical  Egawa et al. (2014)  20  53  2  10  9  7  1  10  0  4  –  –  –  –  –  –  0  0  NA  NA NA  Bade and Singal (2014)  12  43  –  –  –  –  –  –  –  –  –  –  3  2  –  –  –  –  2  7 NA  Total  141  305  28  66  43  33  9  23  6  13  9  36  15  9  4  0  8  8  19  18  Author, year  Total biopsies  Minimal changes  Steatosis  Hepatitis  Cholestasis  ACR  Steatohepatitis  Alcoholic hepatitis  F1–F2 fibrosis  F3–F4 Indication Fibrosis  R  A  R  A  R  A  R  A  R  A  R  A  R  A  R  A  R  A  R  A  Lucey et al. (2009)  13  20  4  4  0  1  8  13  –  –  4  8  2  3  1  0  –  –  –  –clinical  Pageaux et al. (2003)  38  55  7  12  15  9  –  –  –  –  –  –  –  –  3  0  8  8  2  0 clinical  Cuadrado et al. (2005)  34  86  15  40  5  4  –  –  6  9  5  28  –  –  0  0  –  –  0  0 protocol  Rice et al. (2013)  24  48  –  –  14  12  –  –  –  12  23  –  10  4  –  –  –  –  15  9 clinical  Egawa et al. (2014)  20  53  2  10  9  7  1  10  0  4  –  –  –  –  –  –  0  0  NA  NA NA  Bade and Singal (2014)  12  43  –  –  –  –  –  –  –  –  –  –  3  2  –  –  –  –  2  7 NA  Total  141  305  28  66  43  33  9  23  6  13  9  36  15  9  4  0  8  8  19  18  R = recidivists, A = abstainers hepatitis includes histological categories of mild hepatitis, progressive hepatitis and lobular hepatitis as reported by studies; F1–F2: Stage 1 or 2 Metavir fibrosis stage; F3–F4: bridging fibrosis or cirrhosis. Steatosis Of 391 liver biopsies (129 recidivists) from five studies, steatosis was reported on 76 (43 relapsers) biopsies. Relapsers as compared to 262 abstainers, had higher prevalence of steatosis (33% vs. 13%), with over 4-fold risk for presence of steatosis among recidivists [4.08 (2.41–6.91)]. Data were homogeneous (I2 = 0, P = 0.53) without publication bias (Egger’s P = 0.78). Steatohepatitis Of 160 biopsies (49 recidivists) from three studies, steatohepatitis was reported on 24 (15 relapsers) biopsies. Relapsers as compared to 111 abstainers had higher prevalence of steatohepatitis (31% vs. 8%) with over 4-fold risk for presence of steatohepatitis among relapsers, [4.5 (1.4–14.2)], P = 0.01 (Fig. 2A). Data were homogeneous (I2 = 28, P = 0.25) without any publication bias (Egger’s P = 0.54). Fig. 2. View largeDownload slide Forrest plots showing pooled data on liver biopsy findings comparing abstainers and recidivists on the risk of (A) steatohepatitis, (B) Alcoholic hepatitis and (C) Advanced fibrosis or cirrhosis. Fig. 2. View largeDownload slide Forrest plots showing pooled data on liver biopsy findings comparing abstainers and recidivists on the risk of (A) steatohepatitis, (B) Alcoholic hepatitis and (C) Advanced fibrosis or cirrhosis. Of 246 biopsies (85 recidivists) analyzing on alcoholic hepatitis in four studies, one study did not report any case of alcoholic hepatitis, leaving three studies for final analysis. Relapsers as compared to 161 abstainers had higher prevalence of alcoholic hepatitis changes on liver biopsy (5% vs. 0%) with over 9-fold risk for alcoholic hepatitis among relapsers, [9.3 (1.01–85)], P = 0.049 (Fig. 2B). Data were homogeneous (I2 = 0, P = 0.61). Publication bias could not be assessed with only two studies in the analysis. Fibrosis Of 340 biopsies (108 recidivists) from four studies, 37 (19 relapsers) reported presence of advanced fibrosis or cirrhosis. Relapsers as compared to 232 abstainers had higher prevalence of advanced fibrosis or cirrhosis (18% vs. 8%) with over 8-fold risk for advanced fibrosis or cirrhosis, [8.4 (3.5–20)], P < 0.0001 (Fig. 2C). Data were homogeneous (I2 = 0, P = 0.87) without any publication bias (Egger’s P = 0.76). Patient survival A total of 240 patients died on a median follow-up of about 6.0 years after LT. Five-year survival reported on 824 recipients was similar comparing 224 relapsers and 600 abstainers (86% vs. 88%), OR (95% CI) of 1.2 (0.7–2.1), (Fig. 3A). Data were homogeneous (I2 = 20, P = 0.29) without any publication bias (Egger’s P = 0.89). Ten-year survival reported on 692 recipients in four studies was worse among relapsers (N = 184) compared to 508 abstainers (45.2% vs. 65.4%), with over 3-fold risk of death among relapsers: 3.67 (1.42–9.50) (Fig. 3B). Data were heterogeneous (I2 = 82, P = 0.001) without any publication bias (Egger’s P = 0.28). With only one of four studies of poor quality in the initial analysis, sensitivity analysis performed after excluding two studies with highest and lowest OR showed homogeneous data (I2 = 36, P = 0.21), with similar effect size: 3.8 (1.6–8.9). Publication bias could not be assessed with only two studies left in the analysis. Fig. 3. View largeDownload slide Forrest plots showing pooled data on comparing abstainers and recidivists on (A) 5 years and (B) 10-year patient survival after liver transplantation for alcoholic cirrhosis. Fig. 3. View largeDownload slide Forrest plots showing pooled data on comparing abstainers and recidivists on (A) 5 years and (B) 10-year patient survival after liver transplantation for alcoholic cirrhosis. Causes of patient mortality Of 240 patients dying on post-transplant follow-up, causes of death were reported in 175 patients (Table 4). Of these, 19% died from liver failure related to continued alcohol use. More common causes were malignancy in 42% (non-HCC malignancy in 35%), cardiovascular-related mortality in 17%, and infections in 21%. Remaining patients died from liver failure due to graft rejection (Table 4). Table 4. Causes of patient mortality Author, year  Graft rejection  Infection or sepsis  Cardiovascular  Liver failure  HCC  Non-HCC cancers  T  R  A  T  R  A  T  R  A  T  R  A  T  R  A  T  R  A  Lucey et al. (2009)  1  1  0  0  0  0  0  0  0  0  0  0  1  0  1  0  0  0  Pageaux et al. (2003)  3  1  2  2  1  1  3  1  2  0  0  0  2  0  2  7  3  4  Cuadrado et al. (2005)  0  0  0  2  1  1  3  2  1  0  0  0  1  0  1  4  2  2  Faure et al. (2012)  2  –  –  6  –  –  8  –  –  13  –  –  –  –  –  18  –  –  Rice et al. (2013)  –  –  –  11  2  9  15  2  13  19  7  12  3  0  3  30  4  26  Egawa et al. (2014)  1  –  –  16  –  –  1  –  –  2  –  –  5  –  –  2  –  –  Total  7  2  2  37  4  11  30  5  16  34  7  12  12  0  7  61  9  32  Author, year  Graft rejection  Infection or sepsis  Cardiovascular  Liver failure  HCC  Non-HCC cancers  T  R  A  T  R  A  T  R  A  T  R  A  T  R  A  T  R  A  Lucey et al. (2009)  1  1  0  0  0  0  0  0  0  0  0  0  1  0  1  0  0  0  Pageaux et al. (2003)  3  1  2  2  1  1  3  1  2  0  0  0  2  0  2  7  3  4  Cuadrado et al. (2005)  0  0  0  2  1  1  3  2  1  0  0  0  1  0  1  4  2  2  Faure et al. (2012)  2  –  –  6  –  –  8  –  –  13  –  –  –  –  –  18  –  –  Rice et al. (2013)  –  –  –  11  2  9  15  2  13  19  7  12  3  0  3  30  4  26  Egawa et al. (2014)  1  –  –  16  –  –  1  –  –  2  –  –  5  –  –  2  –  –  Total  7  2  2  37  4  11  30  5  16  34  7  12  12  0  7  61  9  32  DISCUSSION Main findings of this meta-analysis are that alcohol relapse after liver transplantation for alcoholic cirrhosis (a) occurs at a rate of about 4.7% per year for any alcohol use and about 2.9% for heavy alcohol use, (b) negatively impacts the liver graft with risk for development of steatohepatitis, alcoholic hepatitis, and advanced fibrosis or cirrhosis; and (c) negatively impacts the long-term survival. Further, malignancy, graft failure and cardiovascular causes are common causes of patient mortality. Self-reported alcohol relapse rates after liver transplantation for alcoholic cirrhosis vary from 10% to 50% (Pageaux et al., 1999; Cuadrado et al., 2005; DiMartini et al., 2010). Pooled data of about 26% alcohol relapse rate in this meta-analysis is consistent with the published literature. Alcohol relapse commonly occurs at 2 years after transplantation and then increases with follow-up time (DiMartini et al., 2010; Russ et al., 2016). In a meta-analysis of 50 studies, annual incidence of alcohol relapse was 5.7% for any alcohol use and 2.5% for heavy alcohol use, similar to incidence rates in this meta-analysis (Dew et al., 2008). Among patients with alcohol use disorder without liver disease or receiving any solid organ transplantation, about 90% after achieving remission report at least one alcohol relapse over 4-year period, as reported by the National Institute on Alcohol Abuse and Alcoholism (1989). In one study, alcohol relapse with alcohol use disorder was reported in about 51% among at risk drinkers, 27% among low risk drinkers and about 7% among abstainers (Dawson et al., 2007). In another study, over a long-term period of 16 years, patients with alcohol use disorder entering into remission with the use of addiction therapy were less likely to report alcohol relapse compared to remitters without obtaining any help for alcohol use disorder (43% vs. 61%, P < 0.01) (Moos and Moos, 2006). Other variables influencing relapse rates are reported to be age, psychosocial functioning, educational status, employment status and life time drinking problem (Vaillant et al., 1991; McCallum and Masterton, 2006; Moos and Moos, 2006). Lack of information on these variables in our study limited our assessment of risk factors predictive of relapse to alcohol use among recipients of LT for alcoholic cirrhosis. Prevalence of steatosis and steatohepatitis was reported in about 33% and 31% of recidivists undergoing liver biopsy evaluation. There is likely selection bias on these findings as these liver biopsies were driven by clinical indication and were not protocol biopsies. Presence of steatosis and steatohepatitis in 13% and 8% of abstainers may likely be due to post-transplant metabolic syndrome or inaccurate self-report on alcohol use (Watt and Charlton, 2010; Singal et al., 2013a). Advanced fibrosis or cirrhosis was present in 18% of recidivists in this study. In a recent study, about 32% relapsers showed biopsy changes of advanced fibrosis or cirrhosis on liver biopsy. Protocol biopsies performed in this study, relatively longer duration of about 9 years, and examining only patients with heavy alcohol use likely explains difference between the studies. Overall recurrent alcoholic cirrhosis occurs in about 5–8% of the whole cohort of liver transplant recipients for alcoholic cirrhosis (Dumortier et al., 2015; Verbeek et al., 2016). In our current analysis, recurrent alcoholic cirrhosis occurred in about 2% of all transplant recipients, about 9% of all biopsied patients, and about 20% of relapsers with heavy alcohol use. Compared to abstainers, relapsers had similar (86% vs. 88%) survival rates at 5 years and lower survival (45% vs. 65%) at 10 years after liver transplantation. These results are similar to a recently reported prospective study on 712 (128 relapsers) patients transplanted for alcoholic cirrhosis, with survival rates comparing relapsers and abstainers at 5 and 10 years of follow-up of 88% vs. 89%, P > 0.05 and 49% vs. 70%, P < 0.005, respectively (Dumortier et al., 2015). Extra-hepatic malignancy and cardiovascular events were common causes of patient mortality, as reported earlier from single center and on analysis of the NIDDK LT database. Development of malignancy among transplanted patients for alcoholic cirrhosis is associated with smoking (Doycheva et al., 2016; Verbeek et al., 2016). However, this could not be analyzed in this analysis due to lack of information on smoking in most studies. Graft failure accounted for about 20% of all the deaths in this pooled data. The data pooled from seven studies with large sample size and stringent criteria for study selection with liver histology comparison of relapsers and abstainers are strengths of our analysis. However, our study suffers from some limitations. For example, all the studies included in this analysis were retrospective or observational with variations on follow-up period and definition of relapse. Also, alcohol use, diagnosis and variability of selection between teams is a limitation of our study. Further, alcohol consumption in all these studies was self-reported, which may often be inaccurate underestimating the true prevalence of alcohol use (Bajaj et al., 2008; Carbonneau et al., 2010). Moreover, biopsies in all studies were performed for clinical indications and were not protocol biopsies, and thus a potential for selection bias. Not all patients studied had biopsies and only 45% (446 out of 995) had liver biopsies. In summary, post-transplant alcohol relapse negatively impacts the liver graft and long-term patient survival among transplant recipients for alcoholic cirrhosis. Our study findings highlight the need and importance of maintaining alcohol abstinence after LT. Well-designed prospective studies are needed for developing strategies to maintain abstinence among LT recipients for alcoholic cirrhosis. FUNDING The study is supported by faculty development grant from the American College of Gastroenterology and R21 grant from the National Institute of Alcohol Abuse and Alcoholism (1R21AA023273-01A1), awarded to Ashwani K. Singal. CONFLICT OF INTEREST STATEMENT None declared. REFERENCES ALCOHOLISM. ( 1989) Relapse and craving. National Institute on Alcohol Abuse and Alcoholism. https://pubs.niaaa.nih.gov/publications/aa06.htm. Bade KS, Singal AK. ( 2014) Recidivism among liver transplant recipients for alcoholic cirrhosis: prevalence, predictors, and impact on outcomes. Am J Gastroenterol  109: S164. Bajaj JS, Saeian K, Hafeezullah M, et al.  . ( 2008) Failure to fully disclose during pretransplant psychological evaluation in alcoholic liver disease: a driving under the influence corroboration study. Liver Transpl  14: 1632– 6. Google Scholar CrossRef Search ADS PubMed  Berlakovich GA. ( 2014) Challenges in transplantation for alcoholic liver disease. World J Gastroenterol  20: 8033– 9. Google Scholar CrossRef Search ADS PubMed  Bjornsson E, Olsson J, Rydell A, et al.  . ( 2005) Long-term follow-up of patients with alcoholic liver disease after liver transplantation in Sweden: impact of structured management on recidivism. Scand J Gastroenterol  40: 206– 16. Google Scholar CrossRef Search ADS PubMed  Burra P, Senzolo M, Adam R, et al.   ELITA & CENTERS, E. L. T. ( 2010) Liver transplantation for alcoholic liver disease in Europe: a study from the ELTR (European Liver Transplant Registry). Am J Transplant  10: 138– 48. Google Scholar CrossRef Search ADS PubMed  Carbonneau M, Jensen LA, Bain VG, et al.  . ( 2010) Alcohol use while on the liver transplant waiting list: a single-center experience. Liver Transpl  16: 91– 7. Google Scholar CrossRef Search ADS PubMed  Cuadrado A, Fabrega E, Casafont F, et al.  . ( 2005) Alcohol recidivism impairs long-term patient survival after orthotopic liver transplantation for alcoholic liver disease. Liver Transpl  11: 420– 6. Google Scholar CrossRef Search ADS PubMed  Dawson DA, Goldstein RB, Grant BF. ( 2007) Rates and correlates of relapse among individuals in remission from DSM-IV alcohol dependence: a 3-year follow-up. Alcohol Clin Exp Res  31: 2036– 45. Google Scholar CrossRef Search ADS PubMed  Dew MA, DiMartini AF, Steel J, et al.  . ( 2008) Meta-analysis of risk for relapse to substance use after transplantation of the liver or other solid organs. Liver Transpl  14: 159– 72. Google Scholar CrossRef Search ADS PubMed  DiMartini A, DEW MA, DAY N, et al.  . ( 2010) Trajectories of alcohol consumption following liver transplantation. Am J Transplant  10: 2305– 12. Google Scholar CrossRef Search ADS PubMed  DiMartini A, Crone C, Dew MA. ( 2011) Alcohol and substance use in liver transplant patients. Clin Liver Dis  15: 727– 51. Google Scholar CrossRef Search ADS PubMed  Doycheva I, Amer S, Watt KD. ( 2016) De Novo malignancies after transplantation: risk and surveillance strategies. Med Clin North Am  100: 551– 67. Google Scholar CrossRef Search ADS PubMed  Dumortier J, Dharancy S, Cannesson A, et al.  . ( 2015) Recurrent alcoholic cirrhosis in severe alcoholic relapse after liver transplantation: a frequent and serious complication. Am J Gastroenterol  110: 1160– 6. quiz 1167. Google Scholar CrossRef Search ADS PubMed  Egawa H, Nishimura K, Teramukai S, et al.  . ( 2014) Risk factors for alcohol relapse after liver transplantation for alcoholic cirrhosis in Japan. Liver Transpl  20: 298– 310. Google Scholar CrossRef Search ADS PubMed  Erard-Poinsot D, Guillaud O, Hervieu V, et al.  . ( 2016) Severe alcoholic relapse after liver transplantation: what consequences on the graft? A study based on liver biopsies analysis. Liver Transpl  22: 773– 84. Google Scholar CrossRef Search ADS PubMed  Faure S, Herrero A, Jung B, et al.  . ( 2012) Excessive alcohol consumption after liver transplantation impacts on long-term survival, whatever the primary indication. J Hepatol  57: 306– 12. Google Scholar CrossRef Search ADS PubMed  Lucey MR, Schaubel DE, Guidinger MK, et al.  . ( 2009) Effect of alcoholic liver disease and hepatitis C infection on waiting list and posttransplant mortality and transplant survival benefit. Hepatology  50: 400– 6. Google Scholar CrossRef Search ADS PubMed  McCallum S, Masterton G. ( 2006) Liver transplantation for alcoholic liver disease: a systematic review of psychosocial selection criteria. Alcohol Alcohol  41: 358– 63. Google Scholar CrossRef Search ADS PubMed  Moos RH, Moos BS. ( 2006) Rates and predictors of relapse after natural and treated remission from alcohol use disorders. Addiction  101: 212– 22. Google Scholar CrossRef Search ADS PubMed  Pageaux GP, Michel J, Coste V, et al.  . ( 1999) Alcoholic cirrhosis is a good indication for liver transplantation, even for cases of recidivism. Gut  45: 421– 6. Google Scholar CrossRef Search ADS PubMed  Pageaux G-P, Bismuth M, Perney P, et al.  . ( 2003) Alcohol relapse after liver transplantation for alcoholic liver disease: does it matter? J Hepatol  38: 629– 34. Google Scholar CrossRef Search ADS PubMed  Pfitzmann R, Schwenzer J, Rayes N, et al.  . ( 2007) Long-term survival and predictors of relapse after orthotopic liver transplantation for alcoholic liver disease. Liver Transpl  13: 197– 205. Google Scholar CrossRef Search ADS PubMed  Rice JP, Eickhoff J, Agni R, et al.  . ( 2013) Abusive drinking after liver transplantation is associated with allograft loss and advanced allograft fibrosis. Liver Transpl  19: 1377– 86. Google Scholar CrossRef Search ADS PubMed  Russ KB, Chen NW, Kamath PS, et al.  . ( 2016) Alcohol use after liver transplantation is independent of liver disease etiology. Alcohol Alcohol  6: 6. Schmeding M, Heidenhain C, Neuhaus R, et al.  . ( 2011) Liver transplantation for alcohol-related cirrhosis: a single centre long-term clinical and histological follow-up. Dig Dis Sci  56: 236– 43. Google Scholar CrossRef Search ADS PubMed  Singal AK, Chaha KS, Rasheed K, et al.  . ( 2013a) Liver transplantation in alcoholic liver disease current status and controversies. World J Gastroenterol  19: 5953– 63. Google Scholar CrossRef Search ADS PubMed  Singal AK, Guturu P, Hmoud B, et al.  . ( 2013b) Evolving frequency and outcomes of liver transplantation based on etiology of liver disease. Transplantation  95: 755– 60. Google Scholar CrossRef Search ADS PubMed  Vaillant GE, Schnurr PP, Baron JA, et al.  . ( 1991) A prospective study of the effects of cigarette smoking and alcohol abuse on mortality. J Gen Intern Med  6: 299– 304. Google Scholar CrossRef Search ADS PubMed  Verbeek J, Cassiman D, Nevens F. ( 2016) De novo malignancy and recurrent alcoholic cirrhosis account for 70% of deaths in patients transplanted for end-stage alcoholic liver disease. Am J Gastroenterol  111: 436– 7. Google Scholar CrossRef Search ADS PubMed  Watt KD, Charlton MR. ( 2010) Metabolic syndrome and liver transplantation: a review and guide to management. J Hepatol  53: 199– 206. Google Scholar CrossRef Search ADS PubMed  © The Author 2017. Medical Council on Alcohol and Oxford University Press. All rights reserved.

Journal

Alcohol and AlcoholismOxford University Press

Published: Mar 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

Monthly Plan

  • Read unlimited articles
  • Personalized recommendations
  • No expiration
  • Print 20 pages per month
  • 20% off on PDF purchases
  • Organize your research
  • Get updates on your journals and topic searches

$49/month

Start Free Trial

14-day Free Trial

Best Deal — 39% off

Annual Plan

  • All the features of the Professional Plan, but for 39% off!
  • Billed annually
  • No expiration
  • For the normal price of 10 articles elsewhere, you get one full year of unlimited access to articles.

$588

$360/year

billed annually
Start Free Trial

14-day Free Trial