Telephone: 54 2322 482884; FAX: 54 2322 482236
Liver transplantation in patients with cystic fibrosis: Analysis of united network for organ sharing data†
Article first published online: 7 MAR 2011
Copyright © 2011 American Association for the Study of Liver Diseases
Volume 17, Issue 3, pages 243–250, March 2011
How to Cite
Mendizabal, M., Reddy, K. R., Cassuto, J., Olthoff, K. M., Faust, T. W., Makar, G. A., Rand, E. B., Shaked, A. and Abt, P. L. (2011), Liver transplantation in patients with cystic fibrosis: Analysis of united network for organ sharing data. Liver Transpl, 17: 243–250. doi: 10.1002/lt.22240
This work was supported in part by the Health Resources and Services Administration (contract 231-00-0115). The content is the responsibility of the authors alone and does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the US Government.
- Issue published online: 7 MAR 2011
- Article first published online: 7 MAR 2011
- Accepted manuscript online: 13 DEC 2010 02:21PM EST
- Manuscript Accepted: 19 NOV 2010
- Manuscript Received: 7 MAY 2010
The improved life expectancy of patients with cystic fibrosis (CF) has led to a change in the impact of liver disease on the prognosis of this population. Liver transplantation has emerged as the procedure of choice for patients with CF and features of hepatic decompensation and for intractable variceal bleeding as a major manifestation. We retrospectively reviewed the United Network for Organ Sharing database to analyze the outcomes of 55 adults and 148 children with CF who underwent liver transplantation, and we compared them to patients who underwent transplantation for other etiologies. We additionally compared the benefits of liver transplantation among patients who underwent transplantation for cystic fibrosis–related liver disease (CFLD) and those who remained on the waiting list. The 5-year survival rates for children and adults undergoing liver transplantation were 85.8% and 72.7%, respectively (P = 0.016). A multivariate Cox regression analysis comparing pediatric and adult CF patients to patients who underwent transplantation for other etiologies noted lower 5-year survival rates (P < 0.0001). However, compared to those remaining on the waiting list, pediatric transplant recipients with CF (hazard ratio = 0.33, 95% confidence interval = 0.16-0.70, P = 0.004) and adult transplant recipients with CF (hazard ratio = 0.25, 95% confidence interval = 0.11-0.57, P = 0.001) gained a significant survival benefit. In conclusion, long-term outcomes in patients with CFLD are acceptable but are inferior in comparison with the outcomes of those undergoing transplantation for other etiologies. Despite such observations, a survival benefit was noted in transplant patients versus those who remained on the waiting list. Liver Transpl, 2011. © 2011 AASLD.
Cystic fibrosis (CF) is an autosomal, recessive, multisystem disease.1 Pulmonary involvement accounts for a majority of the morbidity and mortality. However, with improved management of respiratory complications, increased life expectancy can be achieved; thus, liver disease, including biliary tract disease, has emerged as a significant medical issue.2 Most commonly, liver disease complicating CF presents during the first decade of life; nevertheless, a small percentage of patients develop decompensated cirrhosis during adulthood.2, 3 The cumulative incidence of liver disease in CF patients ranges from 27% to 35%, but progression to cirrhosis is uncommon and occurs in only 3% to 7% of all patients.2-4 Focal biliary cirrhosis is the pathognomonic hepatic manifestation and results from biliary obstruction and progressive periportal fibrosis. Over time, focal biliary cirrhosis can progress to multilobular cirrhosis with clinically significant portal hypertension and related complications.5
ΔF508 is present in approximately 70% of defective cystic fibrosis transmembrane regulator (CFTR) alleles and in 90% of CF patients within the United States.1 It is unclear why only one-third of the patients with the CFTR mutation present with liver disease. Genetic polymorphisms have been described as modifiers of liver disease in CF patients. A higher incidence of severe mutations in patients developing liver disease has been reported,2, 6, 7 but no specific mutation was identified until recently; Z allele has been significantly associated with cystic fibrosis–related liver disease (CFLD) and portal hypertension.8 There is increasing evidence that an absence of CFTR can influence the expression of other genes that up-regulate fibrosis, inflammation, and oxidative stress and that genetic modifiers inherited independently from the CFTR gene can explain the considerable differences in clinical severity among patients with the same CFTR mutations.1, 8, 9
As in other cholestatic diseases, biochemical improvement has been observed with ursodeoxycholic acid, but no medical therapy has been proven to prevent progression to cirrhosis.10, 11 Isolated liver transplantation has emerged as the procedure of choice for those patients with CF and progressive liver failure or predominantly intractable variceal bleeding. Successful outcomes after liver transplantation have been reported with a 5-year survival rate of 75%,12-14 and beneficial effects on nutritional status and lung function are seen in most cases.15 A careful assessment of pulmonary function is critical during the process of patient selection because lung complications, which are primarily related to pulmonary infections, remain major causes of posttransplant morbidity and mortality.12 Both improvements and deterioration in pulmonary function have been reported after liver transplantation.14, 16 For those patients with end-stage liver disease and significant pulmonary complications, combined liver-lung transplantation may be considered. Combined liver-lung transplantation and combined heart-lung-liver transplantation carry a worse prognosis in comparison with liver transplantation alone; the 5-year actuarial survival rate ranges from 37.5% to 64.2%.14, 17 However, the published series are small, and the optimal timing and selection of candidates remain critical issues for a high probability of success.
A central problem in the literature examining liver transplantation for CFLD is the fact that most studies have been conducted in single centers and have relatively few patients. In order to gain a better perspective of the field of CF and liver transplantation, we analyzed the United Network for Organ Sharing (UNOS) database so that we could describe the outcomes of adult and pediatric patients with CFLD who underwent liver transplantation, and we compared these patients to those who underwent transplantation for other etiologies. To further understand the benefits of liver transplantation, we compared patients with CFLD who underwent transplantation to those who remained on the waiting list.
PATIENTS AND METHODS
Data Sources and Study Population
Data from the UNOS database were provided in the form of a Standard Transplant Analysis and Research file with follow-up files from the liver and thoracic organ transplant registry for 1987 to 2008. The data included all listed candidates and transplant recipients in the United States submitted by the members of the Organ Procurement and Transplantation Network during that time period. Adult and pediatric patients who received a liver transplant alone or a combined liver-lung transplant were included in the analysis. Institutional review board approval was obtained from the University of Pennsylvania.
The primary outcome measure was patient survival. The patients were divided into 2 groups, adults (18 years or older) and children (younger than 18 years), because of inherent differences among pediatric and adult populations with CF. The survival of these 2 groups was estimated with the Kaplan-Meier method and was compared with the log-rank test. 1-month and 5-year survival after liver transplantation for CF was compared to survival after transplantation for other indications with multivariate Cox proportional hazards models. Significant univariate predictors of mortality (P < 0.1) were incorporated into the multivariate models in a stepwise fashion to assess the effect on mortality. Only well-represented variables (with <20% of the data missing in the registry) were included in the analysis. In the pediatric group, the models were adjusted for the following: the recipient age, liver disease etiology, race, creatinine level, and location at the time of transplantation [home, hospital but not intensive care unit (ICU), or ICU] and the graft type (cadaveric or living donor). In the adult group, the analysis was adjusted for the following: the recipient age, liver disease etiology, race, creatinine level, gender, and location at the time of transplantation (home, hospitalized or ICU); the presence of diabetes; the donor age, gender, height, and race; the cold ischemia time; and the liver graft type (cadaveric or living donor). The donor risk index was calculated for adult recipients of cadaveric grafts.18
Survival Benefit From Transplantation
To quantitate the potential survival benefit of liver transplantation for patients with CF, we compared the hazard of death for patients after liver transplantation with the hazard of death for patients remaining on the list. With a method similar to that used by Merion et al.,19 a time-dependent survival model was constructed for pediatric and adult patients where wait-list patients contributed time at risk to the wait-list group and after transplant contributed time at risk to the recipient group. Patients in the wait-list group were censored at the time of entry into the transplant group. The 3-year survival of the transplant risk set was compared to the wait-list risk set by Cox proportional hazards analysis. Models were adjusted for the recipient age, gender, and race and for the presence of diabetes at the time of listing.
Comparisons of baseline characteristics were performed with 2-tailed t tests for continuous variables and with the chi-square test for categorical variables. All analyses were performed with SPSS, version 16.0 (SPSS, Inc., Chicago, IL).
In all, 93,918 liver transplants were performed during the study period. We identified 203 liver transplants (0.22%) for CFLD in 196 patients. Of these patients, 189 underwent liver transplantation alone, 13 underwent combined liver-lung transplantation, and 1 underwent triple transplantation (heart, lung, and liver). Seven patients underwent retransplantation for liver graft failure. After liver transplantation alone, 5 patients required a subsequent lung transplant, and 5 other patients remained on the waiting list for lung transplantation. No patients required or were awaiting liver transplantation after lung transplantation.
For the analysis of the characteristics of the study population, the patients were divided into 2 groups: 148 patients were included in the pediatric group (<18 years), and 55 patients were included in the adult group (≥18 years; Table 1). The mean ages at the time of transplantation were 11.1 ± 4.7 years and 26.6 ± 10.2 years for the pediatric and adult groups, respectively. Both groups presented male and white predominance: 62.2% and 85.1% for the pediatric group and 52.7% and 94.5% for the adult group, respectively. At the time of transplantation, fewer pediatric recipients were hospitalized (P < 0.0001). Ascites and encephalopathy were present in 22.9% and 10.8% of the pediatric patients and in 36.3% and 18.2% of the adult patients, respectively. Diabetes mellitus was diagnosed in 18.2% of the pediatric patients and in 54.5% of the adult patients (P < 0.0001). Fifteen pediatric patients (10.1%) received living donor grafts, whereas 14 adult patients (25.5%) received living donor grafts (P = 0.006). Since the initiation of the Pediatric End-Stage Liver Disease (PELD) and Model for End-Stage Liver Disease (MELD) allocation policy, 95 patients underwent transplantation. For the pediatric group, the mean calculated PELD and MELD scores at the time of transplantation were 9.7 ± 13.5 and 21.4 ± 7.1, respectively. The mean calculated MELD score for the adult group was 18.2 ± 8.8. Nineteen children (12.8%) and 4 adults (7.3%) received PELD or MELD score exceptions. The reasons for the MELD/PELD exception points were malnourishment in 7 patients, intractable variceal bleeding in 6 patients, refractory ascites in 3 patients, incapacitating encephalopathy in 3 patients, hepatopulmonary syndrome in 2 patients, and hepatocellular carcinoma in 1 patient. Information was missing for 1 patient (Table 1). The mean follow-up from the time of transplantation was 4.4 ± 3.8 years and 3.1 ± 3.1 years in the pediatric and adult CF groups, respectively, and 4.8 ± 4.0 years and 3.7 ± 2.8 years in the pediatric and adult non-CF groups, respectively.
|Characteristic||Pediatric Patients (n = 148)||Adult Patients (n = 55)||P Value|
|Age, years (mean ± SD)||11.1 ± 4.7||26.6 ± 10.2||N/A|
|Sex, male (%)||62.2||52.7||NS|
|Recipient ethnicity (%)|
|Creatinine, mg/dL (mean ± SD)||0.66 ± 0.72||0.96 ± 0.73||NS|
|International normalized ratio (mean ± SD)||1.6 ± 0.52||1.6 ± 0.66||NS|
|Diabetes mellitus (%)||18.2||54.5||<0.0001|
|Living donor liver transplantation (%)||10.1||25.5||0.006|
|Combined liver-lung transplantation (%)||3.4||16.4||0.003|
|Location at the time of transplantation (%)|
|Hospital, not ICU||12.1||36.4|
|PELD/MELD score (mean ± SD)||9.7 ± 13.5/21.4 ± 7.1||18.2 ± 8.8||N/A|
|PELD/MELD exception points (%)||12.8||7.3||NS|
|Reasons for MELD/PELD exception points for both groups (n)|
|Intractable variceal bleeding||6|
|Incapacitating hepatic encephalopathy||3|
|Donor age, years (mean ± SD)||18.9 ± 15.1||29.3 ± 14.3||<0.0001|
|Donor ethnicity (%)|
|Cold ischemia time, hours (mean ± SD)||7.59 ± 3.2||7.6 ± 5.7||NS|
|Donor risk index (mean ± SD)*||N/A||1.75 ± 0.34||N/A|
Patient Survival and Transplant Benefit
The 30-day survival rates for the 148 pediatric patients and 55 adult patients were 94.6% and 92.7%, respectively (P = 0.67). The 5-year unadjusted survival rates were 85.8% and 72.7% for the pediatric and adult patients, respectively (P = 0.016; Fig. 1). In an adjusted analysis, the pediatric CF group had inferior short- and long-term survival in comparison with the survival of patients with all other indications for liver transplantation (except those with noncholestatic liver disease) at 5 years (Table 2 and Fig. 2A)
|Variable||30 Days||5 Years|
|HR||CI||P Value||HR||CI||P Value|
|Recipient age, years||0.95||0.92-0.97||0.001||0.98||0.97-0.99||0.3|
|Recipient creatinine level, mg/dL||1.22||1.14-1.32||<0.0001||1.12||1.05-1.18||<0.0001|
|Location at the time of transplantation|
|Hospital, not ICU||1.52||1.01-2.30||0.04||1.32||1.07-1.63||0.008|
In the group of adults who underwent transplantation, the CF patients had a 30-day adjusted survival rate similar to that of the transplant recipients without CF (Table 3). However, adult recipients with CF had a lower 5-year adjusted survival rate than patients with other indications for transplantation (P < 0.0001; Fig. 2B).
|Variable||30 Days||5 Years|
|HR||CI||P Value||HR||CI||P Value|
|Recipient age, years||1.01||1.01-1.02||<0.0001||1.02||1.01-1.02||<0.0001|
|Recipient creatinine level, mg/dL||1.08||1.05-1.11||<0.0001||1.08||1.07-1.09||<0.0001|
|Location at the time of transplantation|
|Hospital, not ICU||1.53||1.37-1.68||<0.0001||1.33||1.27-1.38||<0.0001|
|Diabetes mellitus at transplant||0.89||0.82-0.97||0.01||0.86||0.83-0.89||<0.0001|
|Cold ischemia time, hours||1.03||1.02-1.04||<0.0001||1.02||1.01-1.02||<0.0001|
|Donor age, years||1.01||1.00-1.01||<0.0001||1.01||1.00-1.01||<0.0001|
The 3-year unadjusted survival rates for those pediatric patients who received a liver transplant and those who remained on the waiting list were 91.3% and 84.5%, respectively (P = 0.001). Among adult patients, the 3-year unadjusted survival rates were 81.2% for those who underwent liver transplantation and 61.7% for those who remained on the waiting list (P = 0.001). In an adjusted analysis, pediatric transplant recipients with CF [hazard ratio (HR) = 0.33, 95% confidence interval (CI) = 0.16-0.70, P = 0.004; Fig. 3A] and adult transplant recipients with CF (HR = 0.25, 95% CI = 0.11-0.57, P = 0.001; Fig. 3B) gained a significant survival benefit in comparison with those remaining on the wait list.
End-stage liver disease, although uncommon, has a major impact on the clinical course of CF. Liver transplantation is a viable option for CF patients who develop CFLD. Several studies have addressed this issue, but most have been limited in size because of the small number of patients who require liver transplantation; therefore, many questions remain unanswered.12-14, 16, 20 To the best of our knowledge, this is the largest study evaluating the role of liver transplantation in patients with CFLD. The present analysis demonstrates a survival benefit for patients with CFLD versus patients on the waiting list, and it provides essential validation of the procedure as lifesaving therapy in selected patients. However, the transplant team must assess each patient on the waiting list individually and consider whether that particular candidate's expected lifetime will be greater with a liver transplant.
In this analysis, survival among pediatric patients was superior to survival among adults when they underwent transplantation for CFLD. Pediatric and adult recipients with CF had similar perioperative mortality rates but inferior survival rates during long-term follow-up compared to recipients with other indications for liver transplantation. It is unclear why suboptimal outcomes are observed with liver transplantation for CFLD. CF is a chronic disease, and it is conceivable that the association of malabsorption and poor nutritional status, worsening pulmonary function, and infection may play a significant role and thus contribute to increasing long-term morbidity and mortality after liver transplantation.
Using the European Liver Transplant Registry, Melzi et al.13 analyzed the outcomes of 57 patients who underwent transplantation for CFLD and reported a survival rate of 81.4% at 5 years, although the data were not analyzed separately for pediatric and adult transplant recipients. Results comparable to those from the UNOS database have been described in the literature. Recently, Nash et al.14 reported a 5-year actuarial survival rate of 69% in 18 patients older than 15 years who underwent liver transplantation alone. On the other hand, Fridell et al.12 reviewed the outcomes of 12 children who received an isolated liver transplant at the University of Pittsburgh and reported a 5-year survival rate of 75%. Unlike us, when they compared liver transplant patients in this group with patients with other indications for liver transplantation, they found similar patient and graft survival rates. This may be related to patient selection or a small sample size.
The ideal timing of liver transplantation in CF patients is still a matter of debate. Although some centers encourage transplantation after the onset of portal hypertension and before signs of hepatocellular dysfunction, other transplant programs wait for signs of liver failure to emerge.14, 16 Currently, liver transplantation is reserved for those patients with intractable complications of portal hypertension and/or advanced liver failure. Preferably, patients should undergo isolated liver transplantation before their lung function declines to a critical stage because combined liver-lung transplantation carries a worse prognosis.14, 16 Clinical scoring systems have been proposed to evaluate the need and timing for liver transplantation. Noble-Jamieson et al.21 first described a simple scoring system that assessed the severity of portal hypertension, hepatic synthetic function, signs of hypersplenism, and nutritional status.21 Later, slight modifications were proposed with greater prognostic accuracy.16 Notably, both studies were retrospective; thus, further validation in a larger cohort of patients is needed.
In comparison with the general population, CF patients have a high prevalence of diabetes mellitus. This is not an unexpected finding because pancreatic insufficiency is a component of the multisystem involvement in patients with CF.15, 22 In concordance with the natural history of CF, diabetes was diagnosed in more than half of the adult patients of our series; this rate was significantly higher than the rate in the pediatric group (P < 0.0001). Presumably with the use of immunosuppressive drugs after transplantation (mainly corticosteroids and calcineurin inhibitors), the prevalence of diabetes has been observed to be even higher.13 Interestingly, combined en bloc liver and pancreas transplantation in patients with CF has been reported to lower posttransplant insulin requirements in comparison with those patients receiving a liver transplant alone.23
Additional priority for CF patients with liver disease on the transplant list is a subject of ongoing discussion. Whether such patients, on the basis of their mortality risk without transplantation, benefit from MELD exception points is a challenging question because robust data on the natural history of CFLD are lacking. Recently, it has been proposed that patients who are listed for liver transplantation alone and have markedly compromised lung function would benefit from additional priority because subsequent and progressive lung disease would raise the consideration of combined lung-liver transplantation.24 This proposal is partly supported by the fact that those listed for dual liver-lung transplantation are at a greater risk of dying of complications while they are on the waiting list.17, 21 Interestingly, patients with CFLD who benefited from exception points and underwent liver transplantation primarily had malnourishment, intractable variceal bleeding, encephalopathy, and refractory ascites. These cirrhosis-related complications are not considered in the MELD/PELD score.
We considered separate analyses of the UNOS database by adult and pediatric groups to be important for several reasons:
- 1UNOS organ allocation policies differ between pediatric and adult populations.
- 2The percentages of living donor transplants are significantly different.
- 3Lung disease may be more advanced in adults who have developed cirrhosis.21
- 4Adults present with a lower body mass index because of malabsorption and other causes that have been shown to be poor predictors of outcomes for isolated liver transplantation.16
- 5The rate of diabetes, which is an independent risk factor of mortality after liver transplantation, is increased in adults with CF.25-27
As with all registry-based observational studies, the results of this study are subject to certain limitations. Our study is retrospective and cannot account for inherent and undocumented patient characteristics. Additionally, nationwide clinical registries can contain errors related to data entry. Finally, using the UNOS database, we are dependent on the variables and definitions present in the data set, and important information such as detailed, specific indications for transplantation, posttransplant infectious complications, genetic mutations, and causes of death are missing.
In summary, this analysis of the UNOS database has found a significant difference in the survival of pediatric and adult patients who received a liver transplant for CFLD. In comparison with patients with other indications for liver transplantation, adult and pediatric CF patients fared less well; however, pediatric patients with CF achieved results not too dissimilar to those of other children with liver disease.