Liver Transplantation for Alcoholic Liver Disease in Europe: A Study from the ELTR (European Liver Transplant Registry)

Authors


Corresponding author: P. Burra, burra@unipd.it

Abstract

Alcohol-related liver disease (ALD) is one of the most common indications for liver transplantation (LT). Long-term outcome after LT for ALD versus other etiologies is still under debate. The aim of this study was to compare outcome after LT of patients with ALD, viral (VIR), and cryptogenic cirrhosis. Donor, graft and recipient ELTR variables were analysed in transplants for alcoholic and nonalcoholic cirrhosis (1988–2005) and were correlated with patient survival. Causes of death and/or graft failure were compared between groups. Nine thousand eight hundred eighty ALD, 10 943 VIR, 1478 ALD + VIR and 2410 cryptogenic (CRYP) liver transplants were evaluated. One, 3, 5 and 10 years graft survival rates after LT in ALD patients were 84%, 78%, 73%, 58%, significantly higher than in VIR and CRYP (p = 0.04, p = 0.05). By multivariate analysis, ALD + VIR (RR 1.14) and viral alone (RR 1.06) were significant risk factors for mortality. De novo tumors, cardiovascular and social causes were causes of death/graft failure in higher percentage in ALD groups versus other etiologies. LT for ALD cirrhosis has a favorable outcome, however, hepatitis C virus co-infection seems to eliminate this advantage. Screening for de novo tumors and prevention of cardiovascular complications are essential to provide better long-term results.

Introduction

Alcohol-related liver disease (ALD) is one of the commonest indications for liver transplantation in Europe and in North America (1–3). Graft and patient survival rates after liver transplantation are similar to those seen after transplantation for other etiologies of liver disease (4,5). A return to some degree of alcohol consumption after transplantation is reported in up to 50% of cases (6,7), but for most patients, this is not associated with significant graft damage or graft loss. Whether due to a direct effect on the graft or indirectly because of noncompliance or damage to other organs, the effect of histological recurrence due to alcohol recidivism is much less than that seen with some other indications, notably with hepatitis C virus (HCV) infection (8,9). Both acute and chronic rejections are uncommon following liver transplantation for ALD, whereas infections, cardiovascular complications and de novo malignancy are more common, in both abstinent and relapsing patients (10).

Although the outcome after liver transplantation for ALD compares very favorably with those for other causes of cirrhosis, there is still some reluctance for health care professionals to refer these patients for formal assessment (11,12). Therefore, we believe that only a wider analysis from the whole European Registry, might overcome the limitations of single center studies, giving better information about graft and patient outcome and complications in patients transplanted for ALD.

The European Liver Transplant Registry (ELTR) currently allows for the analysis of 68,776 liver transplantations (LTs) performed in Europe on 61,718 patients over a 37-year period and represents a tool to give a comprehensive overview of the status of liver transplantation for ALD in Europe. The aim of this study was to compare the outcome after liver transplantation of patients transplanted for ALD with patients transplanted for viral or cryptogenic cirrhosis and to evaluate the specific role of HCV infection in those with ALD, using data from ELTR registry.

Methods

The ELTR database contains information about all liver transplants done in 23 European countries since 1968 (4). The methods used to obtain the data and details of the data collected have been described previously (13); the results of an audit published in 2003 confirmed the validity of the data (14). Data are submitted anonymously either by national transplant registries (UK, Spain) or by individual centers. Here, we present analyses based on data from 24 711 transplants done in adults (aged 15 years or over) between January 1988 and December 2005. A new system was implemented for the data collection and analysis since 1996. The data collection has been further enhanced since 2005, but the data for the present analysis were collected according to the previous method.

The ELTR does not obtain individual consent from patients for inclusion of their data: the French National Ethics Committee (Commission Nationale de L’Informatique et des Libertes) which was consulted at the creation of the registry since the ELTR is based in Paris, has not considered it mandatory for giving its agreement. All data are anonymised.

Procedures and statistical analysis

We compared patients transplanted for ALD alone to three other groups: (1) Hepatitis Viral Infection (VIR; hepatitis C (HCV), hepatitis B (HBV) and combined hepatitis C and B (HCV-HBV), (2) cryptogenic cirrhosis (CRYP), and (3) ALD and VIR (ALD-VIR). Patients with other viral liver disease than HBV and HCV, cholestatic (PBC; PSC), autoimmune liver diseases and HCC patients were excluded from analysis. We used all the ELTR parametric and discrete variables related to the donor, graft and the recipient. The variables for the donor were age and gender, for the graft, total ischemic time, and for the recipient: age, gender, group matching, UNOS status, emergency transplantation, type of graft (heterotopic, auxiliary, split liver, reduced liver, living donor, domino), liver transplantation associated with other organs, initial immunosuppression and length of follow up. The variable ‘emergency transplantation’ indicates prioritization of the patient according to the clinical condition. All patients with acute liver failure or relisted for primary nonfunction (PNF) or hepatic artery thrombosis were excluded from the analysis.

The causes of death or graft failure after first liver tranpslant were analysed in all groups: primary nonfunction (PNF) or delayed function, intra-operative failure, technical problems, infection, acute or chronic rejection (which are combined in the analysis), renal disease, cardiovascular disease, pulmonary disease, cerebrovascular disease, gastrointestinal disease, multi-organ failure, recurrence of primary liver disease, recurrence of tumors, de novo solid tumors, lymphoproliferative disorders, social problems (including noncompliance with immunosuppressive therapy, suicide or trauma).

Patient survival over time according to the etiology of liver disease was evaluated using the life-table method and comparison between different groups based on the etiology of liver disease was performed by the log-rank test. Discrete variables were shown as percentages and parametric variables as mean values ± SD. We used the Chi-square test for the comparison of discrete variables and Student's t-test for parametric variables and ANOVA analyses when more than two groups to compare. Differences were considered statistically significant when the p-value was less than or equal to 0.05.

The effect of patient and donor variables on patient survival was evaluated by univariate Cox analysis. The variables corresponding to the statistically significant risk-ratio were used for the subsequent multivariate analysis.

Results

Data were available from 9880 patients transplanted for ALD, 10 943 for VIR (6672 for HCV, 3579 for HBV and 584 for combined HCV-HBV), 1478 transplants for ALD-VIR, and 2410 for cryptogenic cirrhosis (CRYP); 108 patients infected with hepatitis viruses other than HBV and HCV were excluded. A significant increase (8.3%) in the proportion of patients transplanted for ALD with or without concomitant viral infection was observed between period 1988–1995 and 1996–2005. This was associated with a concomitant decline in the proportion of transplants undertaken for CRYPT (14.3% vs. 8%, p < 0.001). The variation of the percentage of transplanted patients for different etiologies of liver disease in the two periods is shown in Figure 1.

Figure 1.

ELTR Indication to liver transplantation in periods 1988–1995 and 1996–2005.

The data on donors (age and gender), grafts (total ischemic time), recipients (age, gender, group matching, UNOS status, emergency transplantation), type of graft (heterotopic, auxiliary, split liver, reduced liver, living donor, domino, multiorgan transplantation), initial immunosuppression, and the length of follow up, according to different etiology of liver disease groups are shown in Table 1.

Table 1.  Recipient and donor characteristics
VariablesCIRRHOSISALC+VIRVIR
ALC n = 9880ALC + VIR n = 1478VIR n = 10 943CRYPT n = 2410pALC + HBV n = 309ALC + HCV n = 1119pHBV n = 3579HCV n = 6672p
Recipient age (yrs)52 ± 849 ± 851 ± 1049 ± 12<0.000150 ± 849 ± 8ns47 ± 1053 ± 9<0.0001
Recipient gender (% male)79886958<0.00019088ns7963<0.0001
Donor age (yrs)43 ± 1744 ± 1842 ± 1740 ± 17<0.000145 ± 1844 ± 18ns40 ± 1742 ± 17<0.0001
Donor gender (% male)64646361ns6663ns65620.02
Group matching (% uncompatible)0.40.30.40.5ns0.30.4ns0.40.4ns
Total ischemic time528 ± 204500 ± 198529 ± 208555 ± 221<0.0001509 ± 198493 ± 195ns552 ± 214512 ± 204<0.0001
UNOS status (% 1 + 2)17172022<0.000125150.0012218<0.0001
Emergency (% yes)3235<0.00013.61.9ns42<0.0001
Immunosuppression (% Tac)46464038<0.00014546ns38410.02
Lenght of follow-up (patients) (yrs)3.5 ± 3.63.2 ± 3.33.7 ± 3.94.2 ± 4.4-3.6 ± 3.53.1 ± 3.2-4.2 ± 4.33.5 ± 3.6-
Length of follow-up (graft1) (yrs)3.4 ± 3.53.1 ± 3.23.5 ± 3.93.8 ± 4.3-3.4 ± 3.42.9 ± 3.2-4.0 ± 4.33.2 ± 3.6-
Type of transplant
Heterotopic (%)0.180.070.190.42ns00-0.20.2ns
Auxiliary (%)0.380.350.300.26ns0.30.4ns0.30.3ns
Split (%)2232<0.000121ns33<0.0001
Reduced (%)0.20.30.20.4 0.30.3 0.20.2 
Living donor (%)1132 21 52 
Domino (%)0.70.80.70.4 11 0.40.9 
Decreased donor full size (%)96969395 9596 9294 
Donor age > 50 (%)36413431<0.00014142ns3036<0.0001
Multiorgan transplantation (%)2.01.62.51.80.0071.61.6ns2.52.5ns

Donor variables

The age of donors for ALD patients (43 ± 17 years), ALD-HCV (44 ± 18), and ALD-HBV (45 ± 18) was significantly greater compared to all the other groups. The majority of donors were male (63%) (range 61–66%) but the proportion was not significantly different between different etiologies. The total ischemic time was significantly higher in the CRYP group (555 ± 221 min) compared to ALD, VIR and ALD-VIR (p < 0.0001). Among the viral patients, the total ischemic time was significantly higher between HCV-HBV (558 ± 201), HBV (552 ± 214) and HCV (512 ± 204) (p < 0.0001).

Recipient variables

The recipient age in the HCV group was significantly higher (53 ± 9 years) compared to HBV group (47 ± 10 years), moreover ALD recipients were significantly older (52 ± 8 years) than ALD + VIR (49 ± 8 years) and CRYPT (49 ± 12 years) groups, The majority of the recipients were male (90% ALD + HBV, 88% ALD-VIR and 88% ALD-HCV group (p < 0.0001 between groups ALD, ALD-VIR, VIR and CRYP)). Very few incompatible group matching transplantations were performed (0.4%). The UNOS status 1 or 2 at transplant was reported in 22% of CRYP 17% in ALD, 20% in VIR and 17% in ALD-VIR patients (p < 0.0001), but similar to HCV-HBV patients (23%). Emergency transplantation was performed in 1.9% to 5% of cases, in a significant higher percentage in CRYP recipients compared to the other etiologies, p < 0.0001.

Concerning the type of graft, the split liver technique was performed in 1.4% to 3% of the cases, whereas the living donor liver transplantation was performed in 1% to 4.5% of the cases. Between 1.5 and 5% of the cases the liver was transplanted in combination with other organs, mainly kidney. Initial immunosuppression was ciclosporin (54%) in 51% to 60% of the recipients and tacrolimus (43%) in 38% to 46% of the recipients. Patients were followed up from 2.9 ± 3.2 to 3.9 ± 4.5 years after liver transplantation.

Causes of graft failure or death are reported in Table 2. De novo tumors were a major cause of death/graft failure in the ALD, ALD-VIR (mainly ALD-HBV) groups (13.7%, 9.1% and 15.1%, respectively) (p < 0.0001 compared with nonalcoholic etiologies); similarly, lymphoproliferative disorders accounted for graft failure or death in a higher percentage in ALD, VIR and mainly HCV recipients compared to other etiologies (1.5%, 1.3% and 1.6%, respectively), but this difference was not statistically significant. Interestingly among solid organ tumors 2/97 (2%) patients with viral cirrhosis developed oral or upper GI tract tumors compared to 17/314 (5.4%) patients with ALD (p = 0.09).

Table 2.  Cause of death or graft failure
Cause of death or graft failureCIRRHOSISALC+VIRVIR
ALD n = 2492ALD-VIR n = 416VIR n = 3144CRYP n = 746pALD-HBV n = 66ALD-HCV n = 336pHBV n = 882HCV n = 2065p
Infection15.5%15.1%15.7%17.6%ns12.1%15.5%ns14.4%16.0%ns
Tumor de novo13.7%9.1% 5.3%5.6%<0.000115.1% 8.0%ns5.7%5.0%ns
PNF or DF8.5%9.1%9.9% 9.9%ns19.7% 7.1%0.00111.1%9.4%ns
Rejection7.6%6.3%6.7%10.1%0.01 6.1% 6.6%ns7.9%6.0%0.05
Cardiovascular8.0%6.7%5.3% 7.6%0.0007 3.0% 7.4%ns5.1%5.3%ns
Pulmonary4.7%5.3%4.2%  5%ns 4.6% 5.4%ns3.3%4.6%ns
Recurence4.3%18.3%20.5% 4.6%<0.000112.1%19.4%ns16.2%23.2%<0.0001
Cerebrovascular4.3%3.6%4.5% 5.8%ns  0% 4.5%ns4.8%4.2%ns
MOF3.7%1.7%3.5% 5.4%0.01 1.5% 1.8%ns2.6%4.0%ns
Gastrointestinal2.7%2.4%2.1% 1.7%ns 3.0% 2.4%ns2.0%2.0%ns
Intraoperative2.5%1.9%2.0% 3.1%ns 3.0% 1.8%ns1.6%2.3%ns
Tum rec1.1%2.9%1.2% 0.8%0.01 3.0% 3%ns1.4%1.1%ns
Technical13.1%10.1%12.1%14.9%ns 9.1% 9.8%ns17.4%9.9%<0.0001
Renal1.7%0.7%1.3% 0.9%ns 1.5% 0.6%ns0.7%1.4%ns
Lymphoproliferative1.5%0.5%1.3% 1.1%ns  0% 0.6%ns0.7%1.6%ns
Social1.3%1.4%0.7% 0.4%0.03 3.0% 1.2%ns0.6%0.6%ns
Others5.1%3.4%2.8% 4.6%0.0001 6.1% 3.0%ns2.8%2.7%ns

Patients transplanted for ALD with and without associated viral infection had a significantly higher incidence of death due to cardiovascular events compared to other etiologies (8% ALD ± VIR vs. 5.3% VIR) (Table 2). The incidence of deaths due to all social causes was double in patients with an alcoholic etiology of liver disease with or without viral infection, compared to other etiologies (1.3% ALD and ALD + VIR vs. 0.7% VIRAL and 0.4% CRYPT, respectively; p = 0.03). In detail, transplanted patients or previous ALD had double the incidence of deaths for suicide compared to viral etiology, p = 0.02 (Table 3).

Table 3.  Cause of death or graft failure
Cause of death or graft failureCIRRHOSISALC + VIRVIR
ALC N = 2492ALC + VIR N = 416VIRAL N = 3144CRYPT N = 746pALC + HBV N = 66ALC + HCV N = 336pHBV N = 882HCV N = 2065p
Social1.3%1.4%0.7%0.4%0.033.0%1.2%ns0.6%0.6%ns
Noncompliance immunosup therapy0.2%0.2%0.2%0.1%ns1.5%0%ns0.1%0.2%ns
Suicide0.7%0.5%0.3%0%0.021.5%0.3%ns0.3%0.2%ns
Trauma0.4%0.7%0.3%0.3%ns0%0.9%ns0.1%0.3%ns

Patient survival at 1, 3, 5 and 10 years from first transplantation was 84%, 78%, 73%, 58%, significantly higher in ALD patients compared to VIR (82%, 74%, 70%, 60%) and CRYPT patients (78%, 73%, 69%, 61%) (log rank p = 0.04 and p = 0.05, respectively) (Figure 2). When ALD patients were analysed according to HCV or HBV infection, patient survival at 1, 3, 5 and 10 years from first transplantation was 84%, 75%, 65%, 52%, respectively, in ALD-HCV group, significantly lower than 89%, 85%, 81%, 64%, respectively, observed in ALD-HBV patients (log rank p = 0.0002) (Figure 3). When VIR patients were grouped according to HCV or HBV or HCV-HBV infection, patient survival at 1, 3, 5 and 10 years from first transplantation was 81%, 72%, 67%, 54% in HCV patients, 84%, 78%, 76%, 70% in HBV patients, and 83%, 77%, 70%, 60% in HCV-HBV patients (Figure 4).

Figure 2.

Patient survival after liver transplantation according to liver disease etiologies (HBV and HCV grouped as VIR), ALC = alcoholic liver cirrhosis; CRYPT = cryptogenic cirrhosis.

Figure 3.

Patient survival after liver transplantation according to ALD + HBV and ALD + HCV etiologies (ALD = alcoholic liver cirrhosis; HBV = hepatitis B virus; HCV = hepatitis C virus).

Figure 4.

Patient survival after liver transplantation according to different VIR etiologies (HCV = hepatitis C virus; HBV = hepatitis B virus).

When the variation of patient survival over the years—before 1990 and 1991–1995, 1996–2000, 2001–2005, overall, a lower survival rate was reported in patients transplanted before 1990 compared with patients transplanted after 1990 (Figure 5). When this was evaluated comparing the different etiologies of liver disease, the difference was statistically significant between the period before 1990 and the subsequent periods in the ALD, HBV and CRYPT groups (log-rank p < 0.0001 for each group); after 1990, the survival continued to improve but this was not statistically significant. The trend was confirmed when ALD-HVC and ALD-HBV (curves not shown) were analysed. Incidentally, in the HCV group, no difference in survival was seen between the different decades (log rank p = 0.25) (Figure 6).

Figure 5.

Patient survival after liver transplantation for alcoholic liver cirrhosis according to decades.

Figure 6.

Patient survival after liver transplantation according to decades in patients transplanted for HCV cirrhosis.

Some results of univariate analysis using Cox model of risk factors for mortality are reported in Table 3. Multivariate analysis using Cox model using significant variables confirmed the significant role of emergency indication to transplantation (RR 1.27, 95% CI 1.11–1.45, p = 0.0006) and donor characteristics (donor age > 50 years RR 1.29, 95% CI 1.22–1.37, p < 0.0001; ischemia time ≥ 12 h RR 1.09, 95% CI 1.02–1.17, p < 0.007). VIR etiology and ALD-VIR are significant risk factors both at uni- and multivariate analysis (respectively, RR 1.06, 95% CI 1.00–1.12, p = 0.05 and RR 1.14, 95% CI 1.02–1.28, p = 0.02). The list of significant risk factors of mortality after a first liver transplantation in multivariate analysis is shown in Table 4.

Table 4.  Risk factors of mortality after a first liver transplantation for cirrhosis in Europe in uni- and multivariate analysis (Cox model): 20 718 patients
 UnivariateMultivariate
pRR95% ICp
Heterotopic0.00022.011.16–3.480.01  
Emergency<0.0001  1.271.11–1.450.0006
Reduced graft0.00012.431.58–3.73<0.0001  
Other altern technics0.001 1.371.20–1.56<0.0001  
Donor age ≥50 yrs<0.0001  1.291.22–1.37<0.0001  
Viral etiology0.037 1.061.00–1.120.05  
Alcohol + viral etiology0.045 1.141.02–1.280.02  
Ischemia time ≥12 h0.00071.091.02–1.170.007
Recipient age ≥60 years<0.0001  1.621.49–1.76<0.0001  
Recipient age [45–60]<0.0001  1.241.16–1.32<0.0001  
Cryptogenic etiology0.045 1.010.91–1.11ns
Donor gender: Female0.03  1.061.00–1.120.05  
OLT between 1988 and 1995<0.0001  1.271.19–1.36<0.0001  

Discussion

Alcoholic liver disease is one of the most common causes of cirrhosis and it has become a frequent indication for liver transplantation in Europe and North America (1–3), and many centers have reported encouraging results in terms of survival (4,5). Nonetheless, public opinion and even some medical professionals continue to question the appropriateness of using a scarce resource for those with alcoholic liver disease (11,12). This comes in part from the view (inappropriate in our opinion) that alcoholics bear full responsibility for their illness, concerns about relapse after liver transplantation, and poor compliance due to social co-morbidities (3,15). In this study, which evaluated the largest cohort yet published on patients transplanted for alcoholic liver disease, we demonstrate that patients transplanted for ALD have a slightly better survival compared to viral etiologies and significantly better overall unadjusted survival compared to patients transplanted for cryptogenic cirrhosis. While analyses of registries have the advantage of large numbers, it is acknowledged that there are limitations despite robust audit: reporting is voluntary and may not include important variables, there may be inconsistencies in definitions, selection and posttransplant management. In particular, the treatment of patients with hepatitis B and C viral infection has changed considerably over the course of the study with a major improvement on the impact of re-infection.

Clearly, there are many donor and recipient factors that will contribute to the overall outcome, for example, patients with cryptogenic cirrhosis included a significantly higher proportion of patients in UNOS status 1 and 2 and had longer ischemic time. Although the Cox regression model used to determine risk factors for mortality confirmed the known correlation of severity of liver disease and donor characteristics with patient survival, ALD-VIR and VIR remain significant risk factors for mortality at multivariate analysis with relative risk, respectively, equal to 1.14 and 1.06.

The reported rates of alcohol relapse vary but most centers report relapse rates up to 50% at 5 years of follow up; most but not all centers suggest there is a significant impact of alcohol relapse on survival after transplant. In a previous study, we found that 33% of patients had some degree of relapse: in 64% alcohol consumption was occasional but in 36% alcohol consumption was heavy (classified as more than 200 g of alcohol per week) but the graft showed only minor histological changes at a median of 3 years (16). Although data regarding the rate and extent of alcohol relapse are absent in the ELTR database, recurrence of disease in alcoholic patients leading to death or graft failure was only 4%, arguing for a low percentage of recipients with end stage liver disease as a consequence of alcohol relapse. Our data suggest that, at least in the short and medium term, relapse of alcohol consumption has a modest effect, if any, on reducing survival and that, overall, the protocols used to select patients are good.

This picture, however, may be different when long-term survival is analysed. Cuadrado et al. reported a significantly lower 10-year survival in a cohort of patients with harmful alcohol relapse after liver transplantation, compared to the abstinent group (45% vs. 85%, respectively) (17). However, cardiovascular events and malignant tumors were the most frequent causes of death in the group with alcohol relapse, rather than liver failure due to alcohol directly, confirming our previous findings (16).

Chronic HCV infection often co-exists with ALD and the combination may act synergistically to cause more aggressive liver disease, thus patients with excessive alcohol intake and HCV infection may come to transplantation sooner (18). Furthermore, in patients with HCV infection, alcohol abuse may not be fully explored so supportive measures to maintain abstinence and detect relapse may not be implemented. Although some authors have shown no difference in the outcome of patients transplanted for ALD alone compared with ALD and HCV related cirrhosis, the published series included only small numbers of patients (19). In the present investigation, we have shown that patients with ALD have similar survival when compared to alcoholic disease associated to hepatitis of viral etiology, but patients with ALD plus HCV had a significant lower survival compared to ALD plus HBV infection.

Recidivism of viral disease was the cause of death or graft failure in 18.3% of ALD + HCV patients, significantly different from ALD group (18.3% vs. 4.3%; similarly to the group with HCV alone (23.2%), suggesting that HCV recurrence is an additional factor worsening long-term survival in patients transplanted for ALD, a conclusion indirectly supported by the trend of survival amongst the decades: patients transplanted for ALD showed an improvement in survival over time: this is probably due to multiple factors including better surgical techniques and general care in transplanted patients; in contrast there was no improvement over time in patients transplanted for HCV. As can be seen in Figure 6, concomitant infection with HCV eliminates this progressive improvement of survival over time. An adverse role of HCV co-infection in patients transplanted for ALD liver cirrhosis is in keeping with the finding of more aggressive liver damage was in ALD plus HCV coinfected patients.

Even though alcohol relapse may be considered a failure in selection and/or management of the patient, transplantation for ALD is a very effective, utilitarian use of a scarce resource and indeed may be a better use than if used for patients with end-stage HCV. However, we do not have enough information to estimate whether those transplanted for ALD have a greater benefit from transplantation because we do not have a robust measure of survival without transplantation.

This is the concept of transplant survival benefit introduced by Merion and Schaubel (20). Recently, Lucey et al. (21) have addressed the survival benefit of HCV and ALD candidates for transplantation. Their findings complement those of Burra et al. (8), with the additional observation that pretransplantation mortality is greater for HCV than ALD patients. Pretransplant mortality should ameliorate the impact on survival benefit of posttransplant recurrence of HCV, because in Merion and Schauble's model, the more likely a candidate is to die without a transplant, the less impact posttransplant mortality will have on overall survival benefit. It was surprising therefore, that despite the greater pretransplant mortality, the posttransplant mortality of HCV was such that the overall survival benefit was still less for HCV than ALD patients.

As already reported, the one of the most common cause of death in ALD group was de novo malignancy that developed in 13.7%; which have been reported to rise from 6% to 55% at 15 years after liver transplantation (22). Dumotrier et al. (23) showed in 305 ALD transplanted patients that de novo cancer occurred in 35 patients after LT (11.5%). Univariate analysis disclosed that male gender, history of smoking, and de novo carcinoma were significant survival prognostic factors. We have confirmed these data, as ALD transplanted patients had greater incidence of orolaryngeal and upper GI tract tumors compared to viral patients (5.4% vs. 2%, respectively).

These data confirm the importance of screening and prevention of de novo tumors in ALD patients, and because of the high incidence of oropharyngeal and oesophageal tumors reported in up to 4.2% by other groups (24). If correct, this would suggest that smoking cigarettes, a common problem in patients with ALD is a major factor (25), and this emphasises the need to help the patient stop smoking.

Attention should also be paid to social causes of death, including suicide. Patients with alcohol and alcohol associated with viral disease had twice the incidence of deaths caused by social problems compared to viral etiology alone (1.3% vs. 0.4%; p = 0.03). Patients with previous ALD had double the incidence of suicide compared to viral etiology and cryptogenic cirrhosis (p = 0.02). This strongly supports the need for careful selection and of candidates before liver transplantation, because often social co-moborbidities are already present, although overall there was no difference in term of compliance and quality of life between alcoholic and nonalcoholic in a group on 132 patients in the waiting list for liver transplantation (26).

In conclusion, liver transplantation for ALD cirrhosis compares favorably with other etiologies of liver cirrhosis. Attention should be paid to HCV co-infection, which is shown to eliminate this advantage and increase long-term mortality. Screening and prevention for de novo tumors, cardiovascular diseases and adverse social events after transplant is essential and could provide even better results in long-term survivors.

Conflict of Interest Statement

The authors declare that they have no conflict of interest.

Acknowledgments

The sponsors of the study had no role in study design, data collection, data analysis, data interpretation, or writing the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication. The participants of the European Liver and Intestine Transplant Association (ELITA) acknowledges: the list of the 137 centers contributing to ELTR is available in the ELTR website: http://www.eltr.org

The ELTR is supported by grants from Novartis, and Astellasand receives logistical support from Hopital Paul Brousse (Assistance Publique–Hôpitaux de Paris), Villejuif, France.

Ancillary