This study was financially supported by the Leonard Davis Institute of Health Economics, the National Institutes of Health (grant 1-F32-DK-089694-01 from the National Institute of Diabetes and Digestive and Kidney Diseases), the Agency for Healthcare Research and Quality (grant K08 HS018406 to Scott D. Halpern), and the Health Resources and Services Administration (contract 234-2005-370011C). The contents are the responsibility of the authors alone and do 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.
Primary sclerosing cholangitis (PSC) is a chronic cholestatic liver disease of unclear etiology with an estimated incidence of approximately 1/100,000 person-years.1, 2 Liver transplantation is the only known beneficial therapy for this progressive and potentially fatal disease, and the survival rates after transplantation are favorable for patients with PSC and exceed 80% and 70% at 5 and 8 years, respectively.3
However, there are reasons to suspect that the Model for End-Stage Liver Disease (MELD) score, which has been used to allocate deceased donor livers in the United States since February 27, 2002, may poorly predict waitlist mortality for patients with PSC. First, although patients with PSC suffer general complications of portal hypertension, they also are at increased risk for specific adverse outcomes, including cholangiocarcinoma (CCA) and ascending cholangitis due to biliary strictures; the approximately 70% to 80% of patients with PSC and concomitant inflammatory bowel disease are at risk for complications such as colon cancer as well.1 The risk for CCA is a particular concern because CCA develops in 6% to 36% of patients with PSC and commonly precludes liver transplantation.4-16 Second, few patients with PSC were included in the cohort used to derive the MELD score, and recent evidence suggests that providing greater weight to bilirubin (the MELD component most affected in PSC patients) would provide superior predictions of waitlist mortality.17
To circumvent these perceived disadvantages for patients with PSC, some have speculated that early or preemptive transplantation should be considered.18 Indeed, we have recently shown that PSC patients preferentially receive living donor transplants, perhaps because of preemptive referral patterns.19 Additionally, some patients with PSC (eg, those with 2 documented episodes of cholangitis requiring intravenous antibiotics and hospitalization within 1 year) are awarded exception points to increase their access to transplantation (T. Schiano, unpublished data, 2010).
Although these efforts to increase access for PSC patients appear meritorious, there is little available evidence for assessing whether they are sufficient or even necessary for promoting equal access to transplantation. To date, only 2 studies have addressed the waitlist mortality of patients with PSC. The first study examined the waitlist mortality of all patients in the United States 1 year after the implementation of the MELD allocation score and showed that patients with PSC had a lower risk of death or removal; however, this was a limited 1-year study that presented an unadjusted analysis.20 The only long-term study focusing on the outcomes of patients with PSC on a transplant waitlist was performed in Scandinavia, and the MELD score is not used for liver allocation there. The authors observed waitlist mortality rates of 3% for patients with PSC and 7% for patients with other liver diseases, but they also found that 9% of PSC patients (n = 24) and only 4% of other patients were withdrawn from the waiting list, mainly because of CCA. These findings suggest that patients with PSC who are listed for transplantation may have reduced access to organs. These disadvantages may manifest even more strongly in the United States: the waitlist times are typically longer, so patients with PSC are at risk for complications for a longer time.21
We designed this study to determine whether and why patients with PSC have different risks for death or removal from the waitlist in comparison with patients with other forms of end-stage liver disease. Our goal was to provide evidence for the appropriateness, necessity, or sufficiency of preemptive transplantation or exception points for patients with PSC.
PATIENTS AND METHODS
We used the United Network for Organ Sharing (UNOS)/Organ Procurement and Transplantation Network database to identify all patients listed for liver transplantation on or after February 27, 2002 (the first day on which all these patients were assigned a MELD score). We also included patients remaining on the waitlist who were added before February 27, 2002 because these patients were assigned a MELD score on that day. By including the person-time that patients accrued from February 27, 2002 onward, we were able to fully compare the times to death or waitlist removal among patients with different etiologies of liver disease in the MELD era. The follow-up ended on May 31, 2009, at which time we censored all patients remaining on the waitlist.
CCA, cholangiocarcinoma; CI, confidence interval; HR, hazard ratio; INR, international normalized ratio; MELD, Model for End-Stage Liver Disease; PSC, primary sclerosing cholangitis; UNOS, United Network for Organ Sharing.
We excluded patients who were less than 18 years of age for 2 reasons: (1) livers are allocated to patients less than 12 years of age by a distinct model (the Pediatric End-Stage Liver Disease score), and (2) patients between the ages of 12 and 18 years exhibit a spectrum of primary diagnoses substantially different from those of patients older than 18 years. We also excluded patients listed for retransplantation to ensure that all observations represented unique individuals. Lastly, we excluded patients who were listed as status 1 (defined as fulminant hepatic failure with a life expectancy of less than 7 days without liver transplantation) because these patients rarely have chronic liver diseases and are allocated organs without a MELD score. We excluded 362 of the 72,338 eligible patients (0.5%) because of missing primary diagnoses.
The primary outcome variable was death or removal from the waitlist because the patient became too sick for transplantation or was otherwise medically unsuitable (which we collectively call clinical deterioration). We considered patients who were removed from the transplant list on account of clinical deterioration to be equivalent to patients who died because these chronic liver diseases are almost uniformly fatal in the short term without transplantation. This grouping is consistent with previous research in the field.22, 23 All other outcomes were censored, with the most common censoring events being transplantation or removal due to an improved condition that no longer required transplantation.
To compare patients with PSC and patients without PSC, we used Fisher's exact test and chi-square tests for dichotomous variables and Student t tests or Wilcoxon rank-sum tests for continuous variables (according to their distributions).
We developed a Cox regression model to determine whether the relative hazard of death or withdrawal from the waitlist differed between patients with or without PSC after we accounted for the proportions of these hazards captured by the MELD score. The MELD score was treated as a time-varying covariate to enable adjustments for the full proportion of risk over time captured by the MELD score. We selected other independent variables for inclusion in the final model if they were independently associated with the outcome (P < 0.05) or if their removal from the model changed the coefficient for PSC by ≥10%. The variables tested with these criteria included age, sex, race, blood type, UNOS region, and insurance type (private versus public).
For any patient given MELD exception points (including but not limited to patients with complications such as hepatocellular carcinoma), we used the higher MELD value (incorporating exception points) because the higher value reflected the patient's true prioritization on the waitlist. However, to adjust for the fact that patients with PSC may have improved survival in comparison with other patients because of the nonstandard practice of granting exception points to these patients, we created another time-varying binary covariate to indicate whether exception points had been granted. That is, among patients with PSC, for each individual MELD score that was greater than what would have been calculated with the international normalized ratio (INR), the bilirubin level, and the creatinine level alone, the covariate was assigned a value of 1; for all other scores, it was assigned a value of 0. By incorporating this adjustment, we could ascertain the extent to which outcomes among patients with PSC were influenced by the nonsystematic granting of exception points.
We introduced a final time-varying covariate to account for the time that waitlist patients were temporarily inactive (status 7) because this time enabled patients to accrue risk for outcomes such as death but not transplantation. The insertion of a time-varying covariate to adjust for this time prevented spurious results that could have arisen if patients with PSC and patients without PSC had differed in the time that they were inactive.
We stratified the baseline hazard function by the transplant center to accommodate heterogeneity in the hazard of death or removal from the transplant list across centers.24 We used the sandwich variance estimator to properly account for the correlation due to the clustering of patients with centers. This accounted for patients who underwent transplantation at centers at which living donation was not an option.
We conducted a set of secondary analyses to determine whether (1) differential rates of portal hypertension complications accounted for outcome differences between the groups, (2) differential rates of living donor transplantation explained the results, (3) the results were explained in part by the granting of exception points, and (4) the exclusion of patients with hepatocellular carcinoma changed the results.
First, to determine whether the complications of portal hypertension accounted for the observed differences in the waitlist outcomes of patients with PSC and patients without PSC, we performed a secondary Cox regression, which was adjusted for the occurrence of spontaneous bacterial peritonitis, the development of ascites, hepatic encephalopathy, and recent variceal bleeding (within the 2-week period before listing) at the time of listing. We could not adjust for complications after listing because they are not captured in the UNOS database.
Second, because patients with PSC are preferentially referred for living donor transplantation, we wanted to determine whether this practice of preemptive living donor transplantation for patients with PSC affected the results of our primary analysis.19 We conducted a secondary analysis in which we assumed that living donor transplantation was not possible to determine whether the differential rates of living donor transplantation mitigated the potentially increased hazard of death in patients with PSC due to early living donor transplantation. In this analysis, we recoded various proportions of actual living donor recipients as having died or having received a deceased donor liver. Specifically, we tested the most plausible scenario: all patients who actually received a living donor transplant were instead coded probabilistically according to the actual distributions of outcomes for other patients in their diagnostic group. We also determined what would happen if all patients who actually received a living donor transplant were instead coded as having died because although this is implausible, this would portray the worst-case scenario for outcomes among PSC patients. Lastly, as another means of assessing the impact of differential rates of living donor transplantation on our results, we constructed a Cox regression model that excluded centers performing living donor liver transplantation.
Third, we reran our primary analyses first after we excluded all patients who underwent transplantation with an exception point–adjusted MELD score and second after we excluded all patients other than those with hepatocellular carcinoma who underwent transplantation with an exception point–adjusted MELD.
Finally, we reran our primary analyses after we excluded all patients with hepatocellular carcinoma because these patients systematically receive exception points, and this affects their chances of receiving a transplant.
There were 71,976 patients meeting the inclusion criteria, and 3,165 (4.4%) had PSC. Table 1 displays the demographic characteristics of the patients with PSC and the patients without PSC. As expected because of the epidemiology of the different disease processes, the patients with PSC were younger and were more commonly male and white. The patients with PSC were also more likely to have private insurance.
Table 1. Demographics for All Listed Patients
PSC Patients (n = 3165)
Non-PSC Patients (n = 68,811)
Data were available only for patients who were listed after February 26, 2002.
Race/ethnicity data were missing for 12 PSC patients and 424 non-PSC patients. Eight PSC patients and 751 non-PSC patients were American Indian/Alaskan Native, Native Hawaiian, or multiracial.
Insurance was considered private if the UNOS code for private insurance was used, and insurance was considered public if the UNOS code for public insurance was used (ie, Medicare, Medicaid, or other governmental insurance).
Table 2 displays the reasons for which patients were removed from the waitlist during the study. Among the patients listed for transplantation, a significantly greater proportion of patients without PSC died or were removed from the waitlist because of clinical deterioration (20.5% versus 13.6%, P < 0.001). Death alone was also more common among patients without PSC (15.1% versus 9.2% for PSC patients, P < 0.001).
Table 2. Reasons for Removal
Reason for Removal
PSC Patients (n = 3165)
Non-PSC Patients (n = 68,811)
NOTE: The data are presented as numbers and percentages. Percentages are based on the total number of patients in each group.
Deceased donor transplantation
Refusal to undergo transplantation
Transfer to another center
Too sick for transplantation
Transplantation at another center
Living donor transplantation
Death during transplantation
Diagnostic Group Differences Among Those Who Died
Tables 3 and 4 report the characteristics of patients with PSC and other diagnoses who died or were removed because of clinical deterioration. In comparison with patients without PSC, patients with PSC who died or were removed experienced significantly greater wait times before removal. Among all patients listed for transplantation and among those who died or were removed because of clinical deterioration, patients without PSC more commonly experienced complications of portal hypertension. Among patients who died or were removed from the waitlist, the MELD scores at the time of listing [difference in means = 0.89, 95% confidence interval (CI) = −0.21 to 2.00] and at death/removal (difference in means = 0.04, 95% CI = −1.18 to 1.26) were similar for patients with PSC and patients without PSC. Patients with PSC had significantly higher serum bilirubin values and lower INR and creatinine values.
Table 3. Demographic, Clinical, and Laboratory Data for All Patients Who Died or Were Removed From the Waitlist
PSC Patients (n = 432)
Non-PSC Patients (n = 14,073)
NOTE: Percentages are based on the total number of patients who died or were removed from the waitlist because of clinical deterioration (unless otherwise specified).
The data are presented as medians and 25th and 75th percentiles.
One hundred sixty-six non-PSC patients (1.1%) were American Indian/Alaskan Native, Native Hawaiian, or multiracial.
Data were available only for patients who were listed after February 26, 2002. The presence of complications was defined dichotomously (yes or no).
History of hepatic encephalopathy at listing [n/N (%)]
Waitlist Mortality of Patients With PSC and Patients Without PSC
Figure 1 displays the Kaplan-Meier survival estimates for the 2 groups of patients. The log-rank test comparing the 2 unadjusted failure functions had a P value < 0.001, and this indicated that patients with PSC had a higher unadjusted survival rate than patients without PSC.
When we developed the multivariate Cox model assessing the survival of patients with PSC, each of the independent variables that were tested were significantly associated with the hazard of death/removal due to clinical deterioration and were included in the final model (Table 5). The adjusted hazard ratio (HR) for PSC in this final model was 0.72 (95% CI = 0.66-0.79), and this revealed that waitlist survival was greater for patients with PSC versus patients with other forms of liver disease.
Table 5. Univariate and Multivariate Cox Models Assessing Factors Associated With Death or Removal From the Waitlist
The multivariate model included 3 time-varying covariates: MELD score, exception score (yes or no), and temporarily inactive status (status 7).
The reported P values come from the multivariate model.
Age data are reported as increased HRs for every 10-year age increase.
Insurance was considered private if the UNOS code for private insurance was used, and insurance was considered public if the UNOS code for public insurance was used (ie, Medicare, Medicaid, or other governmental insurance).
Complications of portal hypertension at the time of listing were more prevalent among patients without PSC. Full information on the presence or absence of complications of portal hypertension was available for 52,625 patients (73.1% of the total cohort). The individual HRs for complications of portal hypertension at listing were as follows: (1) 1.62 for esophageal variceal bleeding within 2 weeks (95% CI = 1.31-1.99), (2) 1.40 for hepatic encephalopathy (95% CI = 1.32-1.48), (3) 1.43 for spontaneous bacterial peritonitis (95% CI = 1.30-1.58), and (4) 2.18 for ascites (95% CI = 1.94-2.44). Running the primary model on this restricted cohort (ie, adjusting for the same covariates), we obtained an HR for PSC of 0.67 (95% CI = 0.59-0.76). Further adjustments for the 4 markers of portal hypertension produced an HR for PSC that was closer to the null but was still significant (HR = 0.84, 95% CI = 0.74-0.97).
When we recoded living donor organ recipients according to the outcomes observed for other patients in their diagnostic groups (ie, 21% of living donor recipients without PSC were reclassified as having died or having been removed, whereas 14% of living donor recipients with PSC were reclassified), the observed HR for PSC increased slightly to 0.75 (95% CI = 0.69-0.82). In the worst-case scenario in which all living donor recipients were reclassified as having died or having been removed from the waitlist because of clinical deterioration, the HR for PSC was 0.96 (95% CI = 0.87-1.07). The HR for PSC was unchanged (0.70, 95% CI = 0.55-0.90) when we excluded centers that performed living donor transplantation.
When we excluded all patients who underwent transplantation with an exception point–adjusted MELD score (28.8% of transplant patients without PSC versus 14.8% of transplant patients with PSC) and all patients other than those with hepatocellular carcinoma who underwent transplantation with an exception point–adjusted MELD (12.0% of transplant patients with PSC versus 6.5% of transplant patients without PSC), the results did not significantly change; the respective HRs were 0.70 (95% CI = 0.64-0.77) and 0.75 (95% CI = 0.68-0.82).
Finally, the exclusion of patients with hepatocellular carcinoma yielded an HR for PSC that was unchanged (0.68, 95% CI = 0.62-0.75).
This analysis of the UNOS database of patients on the liver transplant waitlist since the introduction of the MELD allocation system demonstrates that the risk of death or removal from the waitlist due to clinical deterioration is lower for patients with PSC versus patients with other end-stage liver diseases. This reduced risk was identified in time-dependent analyses and remained after adjustments for (1) the MELD score during a patient's entire time on the waitlist, (2) complications of portal hypertension at the time of listing, (3) differential rates of living donor transplantation, and (4) the granting of exception points for patients with PSC and other diseases. This central finding calls into question the need for the current practice of preemptively referring PSC patients for living donor transplantation or providing them with exception points in efforts to enhance their access (T. Schiano, unpublished data, 2010).
In the initial study validating the MELD score as a predictor of patient survival after the placement of a transjugular intrahepatic portosystemic shunt, patients with alcohol-related and cholestatic liver diseases had better 3-month survival after shunt placement.26 Consequently, the initial MELD formula gave additional points to patients with hepatocellular etiologies of liver disease. However, because the MELD score did not incorporate complications of cholestasis or PSC-specific complications, giving points only to patients with hepatocellular etiologies of liver disease was deemed inequitable, and the formula was modified.27 Over time, many have speculated that the weighting of the MELD score still disadvantages patients with PSC and other forms of cholestatic liver disease because it does not capture unique clinical features that better reflect the severity of their illness. Thus, some physicians may recommend living donor transplantation preferentially for PSC patients, and in certain regions (eg, region 9), patients with PSC are granted exception points if they suffer from repeated bouts of cholangitis. Initially, these patients are given a MELD score of 25, and certain regional review boards will grant additional points if a patient suffers from subsequent bouts of cholangitis (T. Schiano, unpublished data, 2010). In contrast to the beneficent intentions of such practices, if the overall survival of patients with PSC who are listed for liver transplantation is superior to the overall survival of patients with other diseases (as this study suggests), practices that are intended to level the playing field for patients with PSC may inadvertently disadvantage other patients.
Our results also provide insights into the mechanisms that do and do not contribute to the lower risks of death or removal for patients with PSC versus other patients with end-stage liver disease. Our results suggest that differential rates in living donor transplantation among patients with PSC may explain part of the observed effect but cannot be the sole explanation because the effect favoring patients with PSC remained even when we reclassified large proportions of patients receiving living donor transplants as having died instead and when we excluded centers that performed living donor liver transplantation. Moreover, by making adjustments in our primary analysis for whether patients with PSC underwent transplantation with exception points and by excluding these patients in secondary analyses, we have shown that the survival benefit among patients with PSC exists independently of these practices.
Finally, our results suggest that differential rates of complications of portal hypertension at the time of listing among groups explain a sizeable proportion of the survival benefit experienced by PSC patients. Adjusting for the occurrence of these intermediate outcomes moved the HR for PSC toward the null (from 0.67 to 0.84), and this suggests that the presence of complications of portal hypertension mediates in part the observed survival benefit. Because patients with PSC are less likely to develop complications of portal hypertension, they may be less likely to die or be removed from the waitlist because of clinical deterioration. However, because the UNOS data set does not provide the specific reasons for removal from the waitlist, we cannot quantify precisely the proportion of the effect explained by complications of portal hypertension.
The UNOS data set limits our ability to explore other potential mechanisms by which patients with PSC have better waitlist survival than patients with other forms of end-stage liver disease. We intend to explore this question in future studies and rely on patient-level data from individual centers.
The results of our study are different from those of the study by Brandsaeter et al.21 One possible explanation is the substantially shorter wait times in the Nordic study (median = 1 month) versus our study (median = 1 year); this meant that fewer patients developed complications of portal hypertension on the waitlist, so the waitlist mortality of patients without PSC was higher in our study. Additionally, the case mixes are substantially different in the 2 studies: PSC is the most common indication for liver transplantation in the Nordic countries, whereas in the United States, hepatitis C and alcoholic liver disease are the most common indications. As we have shown, in comparison with PSC patients, patients with these diseases more commonly present with complications of portal hypertension. However, our study was limited because we could not identify specific reasons for removal due to clinical deterioration. As reported by Brandsaeter et al., many of these patients in Scandinavian nations develop CCA. Future studies using other data sources are needed to better delineate the reasons for removal among patients with PSC in the United States.
Our study has several limitations. First, the use of the UNOS database limited the data elements available for covariate adjustments in the secondary analyses. Data concerning the presence of ascites, spontaneous bacterial peritonitis, and hepatic encephalopathy were available only for patients listed after the MELD score was introduced, so we could perform our secondary analysis with complications of portal hypertension only for 73.1% of our cohort. Also, the severity of these complications could not be validated. We dichotomized the outcomes of ascites and hepatic encephalopathy to limit the risk of exposure misclassification based on the grade or stage of disease. Data on the MELD scores and the laboratory elements making up the MELD score were limited for patients from the pre-MELD era; therefore, we chose to focus only on patients listed in the MELD era so that we could adjust for the severity of illness. Furthermore, the question that we hoped to answer was whether there are differences in the waitlist mortality rates in the MELD era, so a pre-MELD comparison was not needed. It is impossible to predict the outcomes of patients who received exception points on a case-by-case basis (ie, patients with PSC and repeated bouts of bacterial cholangitis) had they not been granted exception points because these patients may have sought high-risk organs or may have been listed in regions with a lower average MELD score at the time of transplantation and thus received a transplant anyway. However, the exclusion of these patients did not significantly affect our results.
A secondary goal of our study was to determine whether the waitlist survival rates are different for patients with PSC who develop CCA and/or bacterial cholangitis and patients without these complications. However, a careful exploration of the UNOS dataset revealed significant underreporting of these complications, and this precluded us from performing this analysis. Because the waitlist survival for the entire group of patients with PSC was significantly better than that for the patients without PSC, the group of patients with PSC and complications (CCA and/or bacterial cholangitis) would have to be rather large and would have to have substantially worse outcomes in comparison with patients without PSC for the waitlist survival of patients with PSC and complications to be significantly worse than that of patients without PSC. We are beginning to plan future studies to address this question with patient-level data acquired directly from individual centers.
Finally, these pooled data cannot be applied perfectly to any given patient with PSC. It may be that some patients with PSC are in fact sicker than their MELD scores indicate, and they may have reductions in their quality of life that are not captured by their scores. However, even a conservative interpretation of these data would yield the conclusion that the survival of the vast majority of patients with PSC is at least comparable to the survival of patients with other forms of end-stage liver disease after they are listed for liver transplantation.
For these reasons, we see no basis for changing the MELD score to explicitly promote access among patients with PSC. In contrast, our results suggest caution in providing exception points to these patients. If physicians are referring patients with PSC for living donor transplantation to prevent them from developing CCA, such choices may erode access for other patients or lead to more living donations than are truly necessary to promote acceptable outcomes for patients with PSC and thereby may cause unnecessary risks to donors.28