Liver transplantation (LT) is a definitive treatment option for patients with end-stage liver disease. Over the last two decades, progress in the surgical technique and availability of better immunosuppression regimens have led to improvement in posttransplantation graft and patient outcomes with 1-year survival rates of more than 90% (1).
Currently, approximately 6000 LT are performed annually in the United States, with approximately 2000 dying while waiting for the organ (1, 2). Living-donor LT (LDLT) and use of non-heart-beating donors (NHBD) have improved the donor pool (3). These factors and the introduction of the Model for End-stage Liver Disease (MELD) score for listing and transplantation allowing transplanting sickest first have considerably reduced wait-list mortality (4, 5).
Multiple recipient and donor factors affect the posttransplantation graft and patient survival. Donor risk index (DRI), based on donor-related factors, such as age, gender, race, height (cm), cause of donor death, cold ischemia time, partial/split LT or not, and organ location (local, regional, or national), is an important variable that affects the posttransplantation outcomes (6). Recipient factors affecting the outcomes after LT include age, baseline medical condition, prior dialysis, and MELD score.
With the evolution of LT including the introduction of MELD score, hepatocellular carcinoma (HCC) appeal criteria, and change in the definition of cryptogenic cirrhosis (CC) with the introduction of nonalcoholic steatohepatitis (NASH) as a specific disease entity, we performed this exploratory analysis of the United Network for Organ Sharing (UNOS) database with the aim of determining the impact of etiology of liver disease on outcomes of posttransplantation graft and patient survival.
Frequency of Transplants With Trends Over Time for Disease Etiology
A total of 54,687 adult liver transplants performed in the United States during 1994 to 2009 were analyzed. The number of transplants increased from 2906 in 1994 to 5357 in 2009. Since 2004, the number of liver transplants has remained stable between 5000 and 5500 per year. Hepatitis C virus (HCV) infection and alcohol were most common etiologies of liver diseases contributing to approximately half of all the liver transplants (Fig. 1). Over time, the proportion of transplants for HCV decreased from approximately 31% in 2000 to 26% in 2009, remained similar at approximately 15% for alcoholic cirrhosis (AC), and increased slightly from 7% in 1994 to approximately 10% in 2009 for HCV-positive drinkers. NASH was established as a distinct category in 2000 and is rapidly emerging as an indication. For example, NASH was listed as an indication for LT in only approximately 1% of all annual LT performed in 2001 and this increased to approximately 8% in 2009. Listing of NASH as disease etiology in 2001 also resulted in decrease in the proportion of cases transplanted for CC (∼15% in 1994 to 8% in 2009). A decrease in proportion of transplants was also noted for primary biliary cirrhosis (PBC; 11% in 1994 to ∼3% in 2009), primary sclerosing cholangitis (PSC; 10% in 1994 to ∼5% in 2009), and hepatitis B virus (HBV; 4% in 1994 to ∼2% in 2009). Approximately 13% of all transplants were performed for HCC. There is a more than fivefold increase in the proportion of transplants for HCC from approximately 5% in 1994 to more than 24% in 2009, with most of this change occurring in approximately 2002 (Fig. 1).
Type of Transplants
A total of 2970 (5.4%) transplants were performed as simultaneous liver-kidney (SLK). HCV, CC/NASH, and HCV+alcohol were common etiologies contributing to 30%, 14%, and 8%, respectively, of all SLK. The proportion of all transplants as SLK increased over time from 2.5% in 1994 to approximately 10.3% in 2009. A total of 2695 (3.8%) were performed using live donors (LDLT), with PBC and PSC being common etiologies requiring LDLT in 8.4% and 4% cases, respectively (see Figure S1, SDC,http://links.lww.com/TP/A759). The proportion of transplants as LDLT increased initially from less than 1% in 1994 to approximately 10% in 2001. However, there has been a consistent fall since then and only approximately 3% of all national LT in 2009 were performed using live donors. NHBD were used in 1885 (2.8%) cases, with CC and NASH being common etiologies contributing to 9% and 5.5% of LT, respectively (see Figure S1, SDC, http://links.lww.com/TP/A759). The use of NHBD increased from less than 1% in 1994 to more than 5% in 2009 (see Figure S1, SDC,http://links.lww.com/TP/A759).
Comparison of Recipient Characteristics
For this analysis, PBC and PSC were grouped as chronic cholestatic liver disease (CCLD) (Table 1). Patients transplanted for NASH, CC, and HCC were older compared with patients transplanted for other indications. Although the mean age at the time of transplantation for patients transplanted for HCV or alcohol was similar, liver transplant was needed approximately 3 years earlier in HCV-positive drinkers. Male gender dominates as LT recipient for most etiologies, except for transplants for PBC being dominated by females and transplants for NASH have equal gender distribution. Of all LT analyzed, 47% to 81% were among Caucasians, 15% to 23% among Hispanics, 1% to 8% among blacks, and 1% to 28% among other races. Black as recipient ethnicity was least (1%) among transplants performed for NASH, whereas Caucasians contributed least (47%) to LT when HBV was the disease etiology. In contrast, other races, especially Asians, contributed most to HBV-related LT (Table 1).
Approximately 50% to 60% of transplants were performed after 2001 (post-MELD era), except for NASH (99.6%) and HCC (80%). MELD score was highest for HBV, NASH, and alcohol liver disease (ALD) and lowest for HCC. Patients transplanted for NASH also had the highest body mass index (BMI), proportion with diabetes mellitus, age at transplantation, and DRI compared with other indications, including CC. In contrast, NASH patients were least likely to be on ventilator or on dialysis at the time of transplantation (Table 1).
Patients with HCV-associated HCC compared with HCC secondary to other indications were younger (mean age, 55.3 vs. 56.4 years; P<0.0001), more likely to be Hispanics (26% vs. 22%; P<0.0001), less likely to be diabetics (22% vs. 26%; P<0.0001), and had shorter waiting time to transplantation from the time of listing (162±267 vs. 186±293 days; P=0.0002).
Graft and patient survival after transplantation at 1, 3, 5, and 10 years were best for PBC, PSC, and NASH, whereas they were worst for HCC and HCV (Table 2). Cox proportional hazards regression analyses adjusted for recipient’s age, gender, race, MELD score, ventilator support, and DRI showed graft and patient survival at 10 years to be worse compared with CCLD with hazard ratios (HR) between 1.0 and 1.5 for CC and ALD and 1.5 or more for HCV, HCV-positive drinkers, and HCC (see Table S1, SDC,http://links.lww.com/TP/A759). These results are also observed on the adjusted survival curves where survival compared with “CCLD” was worse for all other indications, with CC and ALD falling between “CCLD” and HCV, HCV+alcohol, and HCC (Fig. 2A,B). Survival for HBV and NASH was similar to CCLD. Other variables predictive of worse posttransplantation outcomes were recipient age, black race, ventilator support at the time of surgery, MELD score, and DRI (see Table S1, SDC,http://links.lww.com/TP/A759). Five-year graft and patient survival were slightly less for HCV-associated HCC compared with HCC for other indications (63% vs. 69%; P<0.0001, log-rank test, and 70% vs. 74%; P<0.0001, log-rank test, respectively; Fig. 3A,B).
In summary, our analysis of the UNOS database for first liver transplantation performed in adults during 1994 to 2009 showed (a) increasing frequency of transplants especially for NASH and HCC, (b) HCV and alcohol as commonest indications for LT, (c) distinct recipient and donor characteristics of NASH recipients, and (d) graft and patient survival similar for PSC, HBV, and NASH, worse by 60% to 132% for HCV, ALD+HCV, and HCC, and worse by 28% to 77% for ALD and CC compared with patients transplanted for PBC.
HCV and alcohol remain most common etiologies accounting for almost half of all LT in the United States. Earlier data showed a bias among gastroenterologists and hepatologists not to refer patients with ALD for LT (7, 8). However, this was not noted in the present analysis of the UNOS database. Good outcomes after LT for ALD as any other indication have encouraged physicians and the transplant community to refer and transplant ALD patients (9, 10). With the emerging data on beneficial outcomes of LT for patients with severe alcoholic hepatitis who do not respond to medical management, transplants for ALD are likely to be more widely accepted (11, 12).
CC is heterogeneous group of liver disease and causes may include unknown autoimmune disease, drug-induced liver disease, or NASH (13). Awareness and more comprehensive workup of patients with liver disease resulted in decrease in proportion of transplants for CC. NASH recognized in the UNOS database since 2001 is rapidly emerging as the third most common indication for LT. It is projected that NASH would likely overtake HCV as an indication for LT by the year 2020 (14).
Availability of effective and potent treatment options such as oral antiviral agents for HBV and ursodiol for PBC also resulted a decrease in the proportion of annual LT performed for these indications (15, 16). Reasons for decrease in proportion of transplants for PSC are not clear. Excellent liver posttransplantation outcomes of patients with HCC for tumors within Milan criteria and introduction of MELD exception points have resulted in an increase in proportion of transplants for HCC (4, 17). Serum creatinine being a major component of MELD score also resulted in an increase in proportion of SLK also since the introduction of MELD score (18).
NASH patients despite being older, having higher DRI, and more likely to be diabetics, obese, or hypertensive had excellent 5-year outcomes of 80% to 85%, which were similar to patients transplanted for PBC, PSC, or HBV. Although metabolic syndrome or its components frequently occur among patients transplanted for NASH, recurrence rates of NASH and cirrhosis in the allograft remain low, explaining good liver posttransplantation outcomes (19). In our own analysis at the Mayo Clinic, recurrence of NASH occurred in 4.3% at 1 year and 6.3% at 3 years (20). Fibrosis stage ≥2 occurred in only 8% of patients transplanted for NASH compared with 46% of HCV, 15% of PBC/PSC, and 13% of AC (20). In an analysis of the UNOS database, death due to graft failure occurred in only 8.6% of cases transplanted for NASH compared with 16.6% of cases transplanted for diseases other than NASH (21).
Although recurrent PBC occurs in 0% to 35% cases, long median time to recurrence of approximately 10 years and lack of significant impact on graft or patient survival explain excellent and best outcomes of these patients (22). Recurrent PSC occurs in 20% to 25% cases and data on impact on graft survival are controversial. However, outcomes of PSC patients as good as PBC as seen in this study have been reported earlier (22). Regular protocol use of hepatitis B immunoglobulin with the availability of effective and potent oral anti-HBV agents have resulted in reducing the recurrent HBV to less than 5% at 5 years with excellent outcomes (23, 24).
Patients transplanted for AC or CC had 5-year survival rates of approximately 75% to 80%. Although recidivism to alcohol use after LT is common, return to harmful drinking is reported in only 10% to 15% of cases (25). Further, graft failure from alcohol relapse is rare (25–27). The commonest cause for death in patients transplanted for AC is the development of malignancy especially of oropharynx or lungs (9, 28). Heterogeneity among patients diagnosed with CC in terms of etiology of the liver disease limits comparison of their outcomes to patients transplanted for other etiologies.
Patients transplanted for HCV, HCC, and HCV+alcohol have worst outcomes after LT with 5-year patient survival rates of approximately 70% to 75%. Universal recurrence of HCV infection with faster progression of fibrosis resulting in cirrhosis in 40% at 5 years results in poor outcomes compared with other indications (29, 30). Results in the current analysis showed that the outcome of HCV-positive drinkers is similar to HCV-positive patients and worse than AC. Similar data using the UNOS database have been reported earlier (10). However, single-center European study did not show similar findings. Apart from geographic differences, the use of more frequent antiviral therapy in patients with HCV and alcohol compared with either etiology alone may have contributed to these differences (31). Recurrence of HCC in the liver posttransplantation follow-up period remains a risk resulting in approximately 70% patient survival at 5 years (17). HCV-positive HCC had worse outcomes compared with HCC related to other indications. The mechanisms of these findings remain to be determined but are likely related to HCV recurrence and aggressive nature of recurrent HCV in the allograft.
In conclusion, with an epidemic of obesity in the United States, liver transplants performed for NASH are increasing. Effective treatment options have resulted in a decrease in the number of transplants for HBV and PBC. Availability of better treatment modalities, especially combination of direct-acting intervals with interferon-free regimens, is expected to further reduce the number of transplants for HCV-infected cases and would also help better management of posttransplantation HCV recurrence. Excellent posttransplantation outcomes of patients transplanted for NASH and alcoholic liver disease are encouraging for wider acceptability of transplants for these etiologies.
MATERIALS AND METHODS
Establishing the Study Population
The Organ Procurement and Transplantation Network database (1994–2009) was queried for adults (age ≥18 years) undergoing first LT. The study population was stratified into nine groups using the UNOS etiology codes at the time of listing for transplantation: PBC, PSC infection, HCV infection, AC, AC+HCV, NASH, CC, HBV, and HCC. Patients transplanted for the first eight etiologies and with concomitant HCC were included and analyzed as HCC for homogeneity. Because the proportion of patients listed for HCV cirrhosis may be alcohol abusers and patients listed for AC may be positive for HCV, patients with the UNOS codes for both AC and HCV were analyzed as AC+HCV. Transplants from deceased donors as well as from liver donors (LDLT) were included. Patients receiving SLK transplants were included, but those with simultaneous heart, lungs, intestine, or pancreas were excluded.
Our main outcome measures graft and patient survival on follow-up. Patients with functioning graft or alive at their last follow-up visit or at the time of data collection are censored in the UNOS database.
Groups were compared for baseline (a) recipient characteristics, such as age, race, gender, BMI, whether on ventilator before transplant, diabetes, patient receiving dialysis or not, SLK transplant or not, and MELD score, and b) donor characteristics, such as DRI, gender, BMI, proportion of LDLT, or NHBD. MELD score for patients transplanted before 2002 was calculated using serum bilirubin, serum creatinine, and international normalized ratio at the time of transplantation. DRI was calculated using the donor-specific factors, such as age, race, height (cm), split transplant or not, cold ischemia time, cause of donor death, and share type (local, regional, or national) (6).
Cox proportional hazards models were built for the time to graft loss and time to survival to assess whether liver disease etiology predicted the outcomes. Variables that are significantly different between groups at baseline and could have confounded the outcomes were entered into the model. Effect size on Cox regression analyses were given as HR with 95% confidence interval. PBC was used as reference for comparing the graft and patient survival for the other etiologies. If the 95% confidence interval on the HR for any group crossed 1, survival was considered similar to PBC.
Categorical variables were compared using chi-square test. Continuous variables were compared using the analysis of variance test. Kaplan-Meier survival curves for graft and patient survival adjusted for recipient (age, donor, race, and MELD score) and donor (DRI and gender) factors were obtained. Log-rank test was used for comparing disease etiologies for the graft and patient survival. P<0.05 was considered significant. Analyses were performed using the Statistical Analysis Software (SAS Institute, Cary, NC).
The authors thank the UNOS for providing the database for the analysis.
1. Lim KB, Schiano TD. Long-term outcome after liver transplantation. Mount Sinai J Med 2012; 79: 169.
2. Wiesner RH, Rakela J, Ishitani MB, et al.. Recent advances in liver transplantation. Mayo Clinic Proc 2003; 78: 197.
3. Wertheim JA, Petrowsky H, Saab S, et al.. Major challenges limiting liver transplantation in the United States. Am J Transplant 2011; 11: 1773.
4. Kamath PS, Wiesner RH, Malinchoc M, et al.. A model to predict survival in patients with end-stage liver disease. Hepatology 2001; 33: 464.
5. Wiesner R, Edwards E, Freeman R, et al.. Model for end-stage liver disease (MELD) and allocation of donor livers. Gastroenterology 2003; 124: 91.
6. Feng S, Goodrich NP, Bragg-Gresham JL, et al.. Characteristics associated with liver graft failure: the concept of a donor risk index. Am J Transplant 2006; 6: 783.
7. Kotlyar DS, Burke A, Campbell MS, et al.. A critical review of candidacy for orthotopic liver transplantation in alcoholic liver disease. Am J Gastroenterol 2008; 103: 734.
8. Julapalli VR, Kramer JR, El-Serag HB. Evaluation for liver transplantation: adherence to AASLD referral guidelines in a large Veterans Affairs center. Liver Transpl 2005; 11: 1370.
9. Burra P, Senzolo M, Adam R, et al.. Liver transplantation for alcoholic liver disease in Europe: a study from the ELTR (European Liver Transplant Registry). Am J Transplant 2010; 10: 138.
10. Lucey MR, Schaubel DE, Guidinger MK, et al.. Effect of alcoholic liver disease and hepatitis C infection on waiting list and posttransplant mortality
and transplant survival benefit. Hepatology 2009; 50: 400.
11. Mathurin P, Moreno C, Samuel D, et al.. Early liver transplantation for severe alcoholic hepatitis. N Engl J Med 2011; 365: 1790.
12. Singal AK, Bashar H, Anand BS, et al.. Outcomes after liver transplantation for alcoholic hepatitis are similar to alcoholic cirrhosis: exploratory analysis from UNOS
database. Hepatology 2012; 55: 1398.
13. O’Leary JG, Lepe R, Davis GL. Indications for liver transplantation. Gastroenterology 2008; 134: 1764.
14. Charlton MR, Burns JM, Pedersen RA, et al.. Frequency and outcomes of liver transplantation for nonalcoholic steatohepatitis in the United States. Gastroenterology 2011; 141: 1249.
15. Kim WR, Benson JT, Hindman A, et al.. Decline in the need for liver transplantation for end stage liver disease secondary to hepatitis B in the US. Hepatology 2007; 46: 238A.
16. Lee J, Belanger A, Doucette JT, et al.. Transplantation trends in primary biliary cirrhosis. Clin Gastroenterol Hepatol 2007; 5: 1313.
17. Mazzaferro V, Regalia E, Doci R, et al.. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med 1996; 334: 693.
18. Davis CL, Feng S, Sung R, et al.. Simultaneous liver-kidney transplantation: evaluation to decision making. Am J Transplant 2007; 7: 1702.
19. Kim WR, Poterucha JJ, Porayko MK, et al.. Recurrence of nonalcoholic steatohepatitis following liver transplantation. Transplantation 1996; 62: 1802.
20. Singal AK, Watt KD, Heimbach JH, et al.. Recurrence of metabolic syndrome and non-alcoholic steatohepatitis after liver transplantation—a comparative analysis. Hepatology 2012; 56(suppl): 499A.
21. Afzali A, Berry K, Ioannou GN. Excellent posttransplant survival for patients with nonalcoholic steatohepatitis in the United States. Liver Transpl 2012; 18: 29.
22. Carbone M, Neuberger J. Liver transplantation in PBC and PSC: indications and disease recurrence. Clin Res Hepatol Gastroenterol 2011; 35: 446.
23. Markowitz JS, Martin P, Conrad AJ, et al.. Prophylaxis against hepatitis B recurrence following liver transplantation using combination lamivudine and hepatitis B immune globulin. Hepatology 1998; 28: 585.
24. Seehofer D, Berg T. Prevention of hepatitis B recurrence after liver transplantation. Transplantation 2005; 80: S120.
25. DiMartini A, Dew MA, Day N, et al.. Trajectories of alcohol consumption following liver transplantation. Am J Transplant 2010; 10: 2305.
26. Lucey MR. Liver transplantation in patients with alcoholic liver disease. Liver Transpl 2011; 17: 751.
27. Bellamy CO, DiMartini AM, Ruppert K, et al.. Liver transplantation for alcoholic cirrhosis: long term follow-up and impact of disease recurrence. Transplantation 2001; 72: 619.
28. Watt KD, Pedersen RA, Kremers WK, et al.. Evolution of causes and risk factors for mortality
post-liver transplant: results of the NIDDK long-term follow-up study. Am J Transplant 2010; 10: 1420.
29. Gane EJ, Portmann BC, Naoumov NV, et al.. Long-term outcome of hepatitis C infection after liver transplantation. N Engl J Med 1996; 334: 815.
30. Triantos C, Samonakis D, Stigliano R, et al.. Antiviral therapy for recurrent hepatitis C infection after liver transplantation. Transplantation 2005; 80: 540.
31. Aguilera V, Berenguer M, Rubin A, et al.. Cirrhosis of mixed etiology (hepatitis C virus and alcohol): posttransplantation outcome—comparison with hepatitis C virus-related cirrhosis and alcoholic-related cirrhosis. Liver Transpl 2009; 15: 79.