There is a severe discrepancy between the number of liver transplant candidates and number of donor livers available for transplantation in the United States at the present time. Nothing illustrates this better than the rapid rise in the number of patients currently on the United Network for Organ Sharing (UNOS) waiting list for liver transplantation (1). This relative shortage of donor organs has fueled the recent controversy over the distribution and allocation of donor organs. One of the principal issues in this controversy is the marked difference in the number of days on the transplant waiting list before receiving a liver transplant in different parts of the country. There are many possible explanations for these regional differences. The explanations include the possibility that the present system of donor organ identification, retrieval, distribution and allocation results in unequal distribution of livers throughout the country. This is a contentious area, about which many interested parties, both inside and outside the transplant community, hold conflicting views. In addition, there is the possibility that patients in different parts of the country are placed on the waiting list at different points in the natural history of their diseases. This practice could result in a patient getting priority for liver transplantation based on nothing more than an accident of geography. Furthermore, there is the concern that, because of long waiting times in certain regions, there is a pressure on transplant programs to list patients early, before they actually require transplantation, a practice referred to as 'waiting list inflation.'
In order to address these concerns, United Network for Organ Sharing (UNOS) recently asked the various organ specific committees to develop organ specific criteria for entry into the national waiting list. On February 21, 1997, the liver transplant community met at the Natcher Center on the campus of the National Institutes of Health (NIH) to formulate minimal listing criteria. The meeting was cosponsored by the American Society of Transplant Physicians (ASTP) and the American Association for the Study of Liver Diseases (AASLD). It also was endorsed by the Liver and Intestinal Committee of UNOS. A subcommittee of the UNOS Liver and Intestinal Committee had already developed liver-specific minimal listing criteria. These were used as a starting point for discussion and changes were recommended where the UNOS criteria were felt to be too restrictive, not restrictive enough or in need of modification. While we, as organizers of this meeting, recognize that the access of particular patients to an allograft is influenced by rules governing the allocation and distribution of donor organs, the objectives of this conference were confined to the development of minimal listing criteria for potential liver transplant recipients. The present paper summarizes the views of the participants regarding minimal listing criteria for orthotopic liver transplantation for adult patients. A companion paper discusses minimal criteria for pediatric candidates (2).
WHAT IS MEANT BY 'MINIMAL CRITERIA'
Before considering the significance of minimal criteria, it is necessary to define the purpose of the liver transplant waiting list. It was agreed by the participants of the conference that when a program places a patient on the waiting list, it should signify that the program would be prepared to transplant that patient immediately, should a donor liver become available. Patients should not be placed on the waiting list simply because it is perceived that he or she will likely need liver transplantation a some point in the future. It was agreed that the minimal listing criteria should be simple, practicable, based on existing published or, in some cases, unpublished clinical research, and have received broad agreement among the liver transplant community. The simplicity of the criteria would be enhanced by avoiding subjective qualifiers such as 'refractory', 'intractable', 'severe', etc. The criteria should be readily verifiable, and the verification process should become a mechanism for UNOS and the regional review boards (see below) to collect ongoing data.
It was also agreed that the minimal listing criteria should be regularly reviewed by the transplant community, and modified where appropriate. This provision is necessary to accommodate the rapid developments in liver transplantation medicine and surgery, whereby previous contraindications cease to be impediments to successful transplantation. A frequently expressed fear of the participants of the conference was that by writing a set of specific minimal criteria which reflect current practice, we might be constructing a set of immutable rules which, in later years, would restrict access to transplantation of patients with transplantable disorders. Finally, it must be emphasized that it is not mandatory for all patients meeting the minimal listing criteria to be listed for transplantation or be transplanted. The final decision whether to list a patient ultimately depends on the clinical judgment of the patient's physicians and the local transplant program's selection committee.
MONITORING THE LISTING PROCESS
A key component of the development of listing criteria is the establishment of regional review boards. The composition of the review boards has yet to be determined by UNOS, but would comprise physicians drawn from liver transplant programs within that area. The purpose of the regional boards will be 3-fold:
- To monitor adherence to the minimal listing criteria.
- To arbitrate in difficult or exceptional cases, that fall outside the standard minimal listing criteria.
- To facilitate UNOS to gather data on the conduct of, and consequences of, the minimal listing criteria in practice.
It is presumed that there will be regional variation in the extent of review conducted by the regional review boards. However, it is anticipated that a standard report form will allow collection of standardized data on all patients entering the listing process, including the cases, presumably the majority, who clearly meet the minimal criteria outlined above, and who do not need to be reviewed in detail.
It was agreed by the participants that, in general, a minimum criterion for a patient with chronic liver disease to qualify for the waiting list for liver transplantation should be a 90% or less chance of surviving one year. It is also imperative that a mechanism to enter the waiting list exist for exceptional patients, whose predicted one year survival exceeds 90%, but who have other compelling indications for liver transplantation. An example would be a patient with well-compensated Primary Biliary Cirrhosis, who has severely impaired quality of life as a result of pruritus. It is believed that presentation of such patients to the regional review boards is the best way to provide access to the waiting list for exceptional cases, and that consideration of these cases will be one of the important functions of the review boards. However, for several of the busiest UNOS regions, reviewing all listings would be impracticable. Thus, it was believed that criteria developed needed to be straightforward and simple to allow easy verification by the transplant program and by UNOS.
The survival of patients with cirrhosis is dependent on Child-Pugh score (Table 1) (3). This observation is supported by a 15-year study of 620 patients with chronic liver disease from Innsbruck University (4). The estimated 1-year and 5-year survivals were 95% and 75% for patients with Child-Pugh class C, respectively. Based on this institutional review and other studies of natural history, it is apparent that the survival of patients with cirrhosis is significantly decreased after the development of decompensation, usually defined as the onset of any of the following: ascites, jaundice, variceal hemorrhage, or encephalopathy. For example, in a 7-year study of the natural history of chronic hepatitis C with cirrhosis, the probability of decompensation after the diagnosis of cirrhosis was relatively low: 12% at 3 years, 18% at 5 years and 29% at 10 years (5). However, survival was significantly different after decompensation: 91% of cirrhotic patients who remained without decompensation were alive 5 years later, whereas only 50% were alive 5 years following an episode of decompensation. Once decompensation occurred, approximately 20% of patients died within one year. Ascites and variceal bleeding were the most common manifestations of decompensation. Similar natural history data, in a cohort of patients with cirrhosis due to many causes, were reported by the Barcelona group (6). Decompensation occurred in 58% of patients in the first 10 years of observation. Ascites was the most common decompensating event, followed by jaundice, hepatic encephalopathy, and variceal bleeding. The median survival in the cirrhotic who remained stable without decompensation was 8.9 years, whereas the median survival from the first presentation of decompensation was 1.6 years.
Ascites confers a poor prognosis, in part, because of the associated development of spontaneous bacterial peritonitis, and hepatorenal syndrome, both of which further shorten survival. In another report from the University of Barcelona, 25% of cirrhotic patients who developed ascites, experienced an episode of spontaneous bacterial peritonitis within the following year (7). The 1-year cumulative probability of survival was 66% in the cirrhotics with ascites but without an episode of spontaneous bacterial peritonitis, compared to 38% in those with an episode of spontaneous bacterial peritonitis. Hepatorenal syndrome is an even more ominous complication of ascites. The probability of hepatorenal syndrome after the onset of ascites was 18% at 1 year, and 39% at 3 years in the Barcelona series. The median survival after the onset of hepatorenal syndrome was only 1.7 weeks (8).
Regarding the influence of variceal bleeding on the natural history of cirrhosis, the Barcelona studies indicated that just under 40% of cirrhotics without previous decompensation experienced bleeding due to portal hypertension over a 10 year period (7). The survival after an episode of variceal hemorrhage may be confounded by therapeutic interventions. Combining data from disparate series, it appears that mortality after variceal hemorrhage is high despite treatment with either sclerotherapy, shunt surgery, or transjugular portosystemic shunts (TIPS). For example, in a study comparing sclerotherapy and distal splenoral shunt surgery in the prevention of recurrent variceal bleeding, the 1-year survivals were 92% versus 75%, and 4-year survivals were 75% and 43% respectively (9). In a different study of TIPS for management of recurrent bleeding, the overall 1-year survival was 87%, with rates 100%, 86%, and 73% among patients with Child-Pugh classes A, B, and C respectively (10).
In summary, after reviewing published data, it is clear that patient with cirrhosis and a Child-Pugh score of 7 or greater, which places the patient in Class B or C, would meet the requirement for a predicted 1-year survival of 90% or less. In addition, patients who experience gastrointestinal bleeding caused by portal hypertension, or a single episode of spontaneous bacterial peritonitis (SBP), irrespective of their Child-Pugh score, would meet this minimal criterion. By contrast, a patient with well compensated-cirrhosis, i.e., Child-Pugh class A, without a history of portal hypertensive bleeding or SBP, would not meet these minimal criteria, and would require approval of the regional review board in order to gain access to the waiting list on exceptional grounds.
Much of the day's discussion related to whether unique criteria should be developed for the various causes of liver disease. In order to accomplish this, the parenchymal liver diseases were divided into specific categories including chronic hepatitis B, chronic hepatitis C, autoimmune hepatitis, alcoholic liver disease (ALD), cholestatic liver disease: primary biliary cirrhosis (PBC), primary sclerosing cholangitis (PSC) and secondary biliary cirrhosis, and finally metabolic liver diseases (hematochromatosis, Wilson's disease and alpha1 anti-trypsin deficiency) and each disease was addressed individually. It was a general consensus that for all causes of cirrhosis, a Child-Pugh score of 7 or more, i.e. Class B or C, or the presence of portal hypertensive gastrointestinal bleeding should allow entry onto the waiting list. For cirrhosis caused by chronic hepatitis B or C, or autoimmune liver disease, these general minimal listing criteria were adequate without the addition of any diagnosis-specific criteria. Many speakers emphasized that the first requirement was for appropriate medical or surgical management, even in patients who met the minimal standards for entry onto the waiting list. An example is autoimmune liver disease, which is very steroid responsive, even when showing features of decompensated cirrhosis (11). The minimal listing criteria merely allow access to the waiting list, without requiring prior review by the regional review board. It remains in the domain of the individual transplant center to determine whether to place an individual patient with decompensated autoimmune disease on the list, or to initiate medical management, or to do both.
Fulminant Hepatic Failure
The sole minimal listing criterion for patients with fulminant hepatic failure (FHF), regardless of etiology, is the onset of Stage 2 hepatic encephalopathy in a patient who meets the standard definition of FHF, shown in Table 2. It important to emphasize that many of the patients who meet this criterion will recover without transplantation (11). Therefore, continuous monitoring for recovery of liver function is required in these patients.
Patients with subacute hepatic failure (see Table 2 for definition), also have a high mortality, and should be eligable for placement on the list, without prior approval of the regional review committee.
Alcoholic Liver Disease
It was agreed that the general minimal listing criteria described above for patients with chronic parenchymal liver disease should apply to patients with alcoholic liver disease also. It is the possibility of returning to alcohol abuse that separates patients with alcoholic liver disease from those with other forms of chronic liver disease. The discussion at the February meeting drew on the results of a recent NIH-sponsored conference entitled 'Liver Transplantation for Alcoholic Liver Disease,' at which data were presented showing a broad consensus among transplant professionals that a history of abstinence from alcohol is an important criterion in selection of alcoholic patients for liver transplantation. Eighty five per cent of the liver transplant programs in the United States and Europe stated that they either require a fixed period of abstinence, most commonly lasting 6 months, or at least place a value on established abstinence, when assessing which alcoholic patients should be approved for liver transplantation (13). The impact of the requirement for 6 months abstinence on rates of posttransplant alcohol use is controversial. However, one of the basic tenets of UNOS is that policy be made based on consensus opinion. Clearly, there is a strong consensus for requiring that most alcoholic patients should be abstinent from alcohol for at least 6 months before they can be listed for liver transplantation.
The 6 month's abstinence rule serves a number of purposes. First, it provides time for recovery from the acute inflammatory effects of recent alcohol exposure. Decompensation in patients with chronic alcoholic liver disease is often the result of acute alcoholic hepatitis supervening on established liver injury. When these patients maintain complete abstinence from alcohol, many will improve to the point that liver transplantation may not be needed (14,15,16). The 6 month abstinence requirement, in effect, excludes patients with alcoholic hepatitis from consideration for liver transplantation, a group of patients who were specifically excluded from transplantation in the original criteria developed by the UNOS Liver and Intestinal Committee. Second, it is recognized that the data regarding the utility of the 6 month abstinence period as a predictor of long-term sobriety are conflicting (17). However, it should be emphasized that 6 months abstinence from alcohol is not an adequate minimal standard by itself for placement on the liver transplant waiting list. There is widespread agreement that alcoholic patients must be evaluated by a substance abuse professional and must comply with the recommendations of the substance abuse professional (13). We recommend that the minimal criteria for listing an alcoholic patient with serious liver disease should be all of the following:
- Approval by the center evaluation committee,
- Child-Pugh point score of 7 or greater, or portal hypertensive bleeding, or an episode of spontaneous bacterial peritonitis,
- Favorable assessment by a substance abuse professional and reported abstinence of at least 6 months at time of listing.
- Exceptional patients with alcoholic liver disease, who have not been abstinent for 6 months and yet whom the transplant program believes are good candidates for liver transplantation, can be referred to the regional review boards for consideration.
- It also bears mentioning that the criteria regarding abstinence from alcohol, and assessment by a substance abuse professional, also apply to those patients with duel diagnoses which include alcoholic liver disease, e.g., alcoholic liver disease and chronic hepatitis C infection.
- There was much discussion regarding the utility of a contract or agreement to abstain, signed at the time of evaluation by the transplant candidate. Although many programs include such documents in their evaluation protocols, it was agreed that the policy to adopt such contracts should remain center-driven, and not become a UNOS requirement.
- Finally, the conference participants discussed the appropriate response to the alcoholic patient who resumes alcohol use while in the evaluation process, or while on the transplant waiting list.
The consensus was that such patients should be removed from the list. These patients could reenter the evaluation process after therapy, at the discretion of the transplant center evaluation committee, and as directed by the center's substance abuse professional. The rules regarding abstinence from alcohol and approval by the transplant center evaluation committee and substance abuse professional before placement on the waiting list, still apply. Exceptional cases may be brought to the regional review committee.
CHRONIC CHOLESTATIC LIVER DISEASE
The cholestatic liver diseases, primary biliary cirrhosis (PBC) and primary sclerosing cholangitis (PSC), represent the 2 diseases for which the natural history of the disease is perhaps best defined. The Liver and Intestinal Committee of UNOS recommended using the Mayo risk assessment scores which were developed as prognostic tools for PBC and PSC (18,19). These 2 prognostic instruments are shown in the addendum. They also recommended the minimum listing criterion would be a risk score which predicts a < 95% chance of surviving the next year. However, the participants at the conference believed that the 95% standard was too lax, given that the standard minimum criterion for patients with parenchymal liver disease was a greater than 10% risk of death in one year.
Primary Biliary Cirrhosis
The Mayo mathematical model of PBC identified five independent factors from which a prognotic equation was derived (18). The factors were the patient's age, total serum bilirubin, and serum albumin concentrations, prothrombin time and severity of edema. However, using the Mayo risk score as a sole criterion of suitability for the waiting list has some limitations. For example, although age is one of the 5 factors used to calculate the Mayo Risk Score, it may lead to an overestimation of the patient's risk for liver disease-related mortality. For example, a 63 year old asymptomatic patient with normal liver function would meet the criteria proposed by the UNOS Liver and Intestinal committee as a criterion for listing. Second, ursodeoxycholic acid (Ursodiol) therapy, which is currently the treatment of choice for patients with PBC, has a marginal effect on survival (20), yet has a more striking effect on the Mayo Risk Score (21), because the serum bilirubin concentration is so heavily weighted in the model. Thus, using the Mayo Risk Score in ursodiol-treated patients may underestimate their mortality. Third, the model tends to underestimate mortality in patients with life-threatening complications such as variceal bleeding. Fourth, the risk score is not easy to calculate without the use of electronics (see *addendum*). Finally, the variability of the disease makes it more difficult to generate predictions for individual patients given that the model represents only the mean of a disparate population. We are working to have these scores available on the UNOS website to facilitate their incorporation into routine pretransplant assessment.
Primary Sclerosing Cholangitis
The Mayo Risk Score for PSC uses 4 factors (age, bilirubin, splenomegaly and presence of cirrhosis on liver biopsy) to predict risk of death (19). This model shares several of the limitations listed above for the PBC model (e.g. age, disease variability) but presents some new concerns as well. The first is the lack of validation of this model in a second PSC population. Second, use of the model requires performing a liver biopsy in a patient where the biopsy would not be otherwise indicated. Third, the variability in serum bilirubin concentrations over even short periods can be striking, often related to episodes of cholangitis or transient biliary obstruction. Finally, the system is difficult to use again requiring electronics.
In the interest of keeping the minimal listing criteria simple and practical, we recommend the following standard listing criteria for patients with PBC or PSC:
- Child-Pugh score of 7 or greater (using the adjustments for cholestatic liver diseases as shown in Table 1) or the development of portal hypertensive gastrointestinal bleeding.
- In addition, PBC or PSC patients with a Mayo Risk Score that predicts greater than 10% 1-year mortality meet minimal criteria for listing for liver transplantation. Since many patients with PSC do not have a recent liver biopsy, we suggest that the presence of thrombocytopenia and splenomegaly could be used as surrogate markers for the histologic finding of cirrhosis and be assigned a score of 2 for histology when calculating the Mayo Risk Score.
Patients with 'quality of life' type indications, e.g. refractory pruritus, fatigue or recurrent cholangitis, but who do not meet the criteria listed above, should be referred to the regional review board.
OTHER ETIOLOGIES OF CHRONIC LIVER DISEASE
For virtually all other forms of chronic liver disease leading to cirrhosis (including hemochromatosis, Wilson's disease, alpha 1 anti-trypsin deficiency, cryptogenic cirrhosis, secondary biliary cirrhosis and Budd-Chiari syndrome) the standard non-disease specific minimal listing criteria would apply. There were some unique listing criteria identified. For Budd-Chiari syndrome, we recommend that a portal decompressive procedure is often the treatment of choice in the patient who does not yet have cirrhosis (22). Liver transplantation is appropriate in patients with Budd-Chiari syndrome:
- When portal decompression has failed or is not technically feasible,
- In patients with established cirrhosis,
- In patients who have a fulminant presentation.
Patients with acute Wilson's disease can be listed as status 1, if they meet appropriate criteria. All of these patients would almost certainly meet the standard listing criteria, thus it was not believed that unique criteria need be developed. Nazer et al. have reported a useful prognostic score, based on serum bilirubin, serum transaminases, and prothrombin time, which stratifies patients according to risk of death without transplantation (23).
Although earlier reports on the outcome of liver transplantation in patients were not encouraging on account of tumor recurrence, more recent studies show an encouraging survival in patients with tumors confined to the liver (24). There was widespread agreement at the February 1997 conference that the minimal listing criteria for patients with primary liver tumors should admit any patient with tumor confined to the liver, irrespective of size or number of tumors, after careful investigation had failed to reveal spread to lymph nodes, the portal vein, or distant organs. The tumors under consideration for liver transplantation include: hepatocellular cancer (AKA: HCC, or hepatoma), hepatoblastoma, rare selected cholangiocarcinomas, and finally selected neuroendocrine tumors. Before considering whether a particular patient with neoplastic disease meets minimal listing criteria, primary resection of the tumor should be considered. Patients with tumor outside of the hepatobiliary tree, i.e. portal vein invasion, lymph node involvement, distant metastases or local extension, should not be listed for liver transplantation. Patients with unusual tumor indications, e.g. extensive cavernous hemangiomas should be reviewed by the regional review board.
It was recognized that severe restrictions on the entry of hepatic malignancy patients to the waiting list could stifle future development of potentially life-saving therapy, involving hepatic transplantation, for previously untransplantable hepatic tumors. It was agreed that innovative protocols, that include liver transplantation, to treat hepatic malignancy, should be reviewed and approved by the regional review boards.
Liver transplantation may be appropriate in circumstances in which the risk of hepatocellular carcinoma is high, even in the absence of an overt tumor. For example, it was agreed that listing for transplantation was appropriate for patients with cirrhosis associated with chronic hepatitis B or C, accompanied by an alpha feto-protein (AFP) level of > 250 ng/ml or an AFP level of 100 ng/ml with the AFP having risen on 3 consecutive measurements, in the absence of a mass lesion.
UNUSUAL INDICATIONS FOR LIVER TRANSPLANTATION
It is obviously impossible to cover in a one day conference minimal listing criteria, which adhere to the requirement for simplicity and ease of use, for every conceivable indication for liver transplantation. There are a number of conditions for which liver transplantation is indicated wherein affected patients will never meet the minimal listing criteria stated here. Some examples include polycystic liver disease, hereditary oxalosis, cystic fibrosis, porphyrias and familial amyloidosis. These diseases or conditions arise so infrequently that it would not be burdensome for the regional review boards to deal with them on a case by case basis, rather than trying to draw up criteria for each of the diseases. The same considerations apply to multiple organ transplants that include liver transplantation.
CONTRAINDICATIONS, AND REMOVAL FROM THE LIST
It was the consensus of the attendees that there should be no absolute contraindications to placement of patients on the liver transplant waiting list. It is reasonable to identify relative contraindications, which require referral to the regional review committee in order to gain access to the waiting list. These relative contraindications are in some cases the converse of the indications mentioned above:
- Well-compensated cirrhosis, without a history of complications such as variceal hemorrhage or SBP.
- Alcohol use in the previous 6 months by a patient with a diagnosis of alcohol abuse or dependence
- Illicit drug use in the previous 6 months by a patient with a diagnosis of substance abuse
- Extra hepatic malignancy other than skin cancer
- Systemic sepsis
- Other relative contraindications are reflective of current medical practice, and may become redundant in the future:
- HIV positive status
- Such has been the advance in treatment of HIV infection, that this limitation may go the way of previous restrictions on transplantation of patients with HBV infection, which are now considered inappropriate.
It was the consensus of the attendees that UNOS should not formulate criteria to remove already-listed patients from the liver transplant waiting list. This was a clinical decision best left to the transplant centers.
The Mayo Model for Primary Biliary Cirrhosis R=0.871 loge (bilirubin in mg/dl) + -2.38 loge (albumin in g/dl) + 0.039 age in years + 2.38 loge (prothrombin time in sec) + 0.859 edema.
Albumin should be measured by serum electrophoresis.
Edema is estimated on a simple clinical scale: none, and no diuretic=0; edema without diuretics, or no edema after diuretics=0.5; and edema despite diuretics=1.
Underlying survival function for the Mayo model: Table
So(t) gives the survival probabilities for a patient with a risk score 5.07, the mean of the combined Mayo data set. To calculate the survival of a given patient, use the following equation: S(t)=So(t)exp(r-3.326)
The Mayo Model for Primary Sclerosing Cholangitis
R=(0.535 × log e total bilirubin in mg/dl) + (0.486 × histologic stage) + (0.041 × age in years) + (0.705 × presence of splenomegaly)
For histologic stages 1 and 2, score 1; for stage 3, score 2; and for stage 4, score 4. For splenomegaly, score 1.
Underlying survival function for the Mayo model: Table
So(t) gives the survival probabilities for a patient with a risk score 3.326, the mean of the combined data set. To calculate the survival of a given patient, use the following equation: S(t)=So(t)exp(r-3.326)
1. Data supplied by UNOS (United Network for Organ Sharing), Richmond, VA.
2. McDiarmid S, Millis M, Oltoff K. Minimal listing criteria for pediatric orthotopic liver transplantation. Liver Transplant Surg 1997 (Submitted)
3. Pugh RNH, Murray-Lyon IM, Dawson JL, Pietrony MC, Williams R. Transection of the esophagus for bleeding esophageal varices. Br J Surg 1973; 60: 649-654
4. Propst A, Propst T, Zangerl G, Ofner D, Judmaier G, Vogel W. Prognosis and life expectancy in chronic liver disease. Dig Dis Sci 1995; 40: 1805-15
5. Fattovich G, Giustina G, Degos F, Tremolada F, Diodati G, Almasio P, et al. Morbidity and mortality in compensated cirrhosis type C: a retrospective follow-up study of 384 patients. Gastroenterology 1997; 112: 463-472
6. Gines P, Quintero E, Arroyo V, Teres J, Bruguera M, Rimola A, Caballeria J, Rodes R, Rozman C. Compensated cirrhosis: Natural history and prognostic factors. Hepatology 1987; 7: 122-28
7. Andreu M, Sola R, Sitges-Serra A, Alia V, Gallen M, Vila MC et al. Risk factors for spontaneous bacterial peritonitis in cirrhotic patients with ascites. Gastroenterology 1993; 104: 1133-1138
8. Gines A, Escorsell A, Gines P, Salo J, Jimenez W, Inglada W, et al. Incidence, predictive factors and prognosis of the hepatorenal syndrome in cirrhosis in cirrhosis with ascites. Gastroenterology 1993; 105: 229-236
9. Henderson JM, Kutner MH, Millikan WJ, Galambos JT, Riepe SP, Brooks WS. Endoscopic variceal sclerosis compared with distal splenorenal shunt to prevent recurrent variceal bleeding in cirrhosis: a prospective, randomized trial. Ann Intern Med 1990; 112: 262-269
10. Rossle M, Haag K, Ochs A, Sellinger M, Noldge G, Perarnau JM, et al. The transjugular intrahepatic portosystemic stent-shunt procedure for variceal bleeding. N Engl J Med 1994; 330: 165-71
11. Czaja AJ. Diagnosis and therapy of autoimmune liver disease. Med Clin of N Am 1996; 80: 973-994
12. O'Grady J, Alexander GJ, Hayllar KM, Williams R. Early indicators of prognosis in fulminant hepatic failure. Gastroenterology 1989; 97: 439-445
13. Everhart J A survey of liver transplant programs in the United States. Liver Transplant Surg 1997 (in press)
14. Chedid A, Mendenhall CL, Gartside P, French SW, Chen T, Rabin L, and the VA Cooperative Study Group. Prognostic factors in alcoholic liver disease. Am J Gastroenterol 1991; 86: 210-216.
15. Alexander JF, Lischner MW, Galambos JT. Natural history of alcoholic hepatitis. II. The long-term prognosis. Am J Gastro 1971; 56: 515-525.
16. Borowsky A, Stomre S, Lott E. Continued heavy drinking and survival in alcoholic cirrhotics. Gastroenterology 1981; 80: 1405-1409.
17. Lucey MR, Carr K, Beresford TP, Fisher L, Shieck V, Brown K A, Campbell DA, Appelman HD. Alcohol use after liver transplantation in alcoholics - a clinical cohort study. Hepatology 1997 (in Press)
18. Dickson ER, Grambsch PM, Fleming TR, Fisher LD, Langworthy A. Prognosis on primary biliary cirrhosis: model for decision making. Hepatology 1989; 40: 1-7
19. Dickson ER, Murtaugh PA, Wiesner RH, Grambsch PM, Fleming TR, Ludwig J, LaRusso NF, Malinchoc M, Chapman RW, Kaplan MM et al. Primary sclerosing cholangitis: refinement and validation of survival models. Gastroenterology 1992; 103: 1893-901
20. Lindor KD, Therneau TM, Jorgensen RA, Malinchoc M, Dickson ER. Effects of ursodeoxycholic acid on survival in patients with primary biliary cirrhosis. Gastroenterology 1996; 110: 1515-8
21. Kilmurry MR, Heathcote EJ, Cauch-Dudek K, O Rourke K, Bailey RJ, Blendis LM, Ghent CN, Minuk GY, Pappas SC, Scully LJ, Steinbrecher UP, Sutherland LR, Williams CN, Woroberz LJ. Hepatology 1995; 23; 1148-53
22. Shaked A, Goldstein RM, Klintmalm GB, Drazan K, Husberg B, Busuttil RW. Portosystemic shunt versus orthotopic liver transplantation for Budd-Chiari syndrome. Surg Gynecol Obstet 1992; 174: 453-459
23. Nazer H, Ede RJ, Mowat AP, Williams R. Wilson's disease: Clinical presentation and use of a prognostic index. Gut 1986; 27: 1377-81
24. Mazzaferro V, Regalia E, Doci R, Andreola S, Pulvirento A, Bozzetti F, Montalto F, Ammatuna M, Morabito A, Gennari L. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med 1996; 334: 693-9