majority of these models involve invasive tests and have Clinical decisions in primary sclerosing cholangitis never been demonstrated to be superior to more simple, dis-(PSC) depend upon understanding its variable natural ease-nonspecific approaches. We sought to determine if the history. Several prognostic models for survival haveChild-Pugh Classification (CPC), [30][31][32] with its ready applicabeen developed. We explored the Child-Pugh Classificability to clinical practice, provides a similar degree of accution (CPC) to determine if it predicts survival as well as racy in estimating survival as the most widely used diseasea disease-specific model (DSM). We identified 208 PSC specific model (DSM rhage, splenomegaly, and encephalopathy. Patients who reported nonspecific complaints such as fatigue, weight loss, or anorexia were classified as being asymptomatic. All blood tests were completed Primary sclerosing cholangitis (PSC) is a chronic chole-using standard laboratory techniques at the Cleveland Clinic. Cholstatic disease that is characterized by inflammation and fi-angiographic evidence of PSC included multifocal, diffuse strictures with alternating normal or dilated segments that produced a characbrosis of the intra-and extrahepatic bile ducts. 1-5 The etiology teristic ''beaded'' appearance and mural irregularities, involving both of PSC remains unknown, 6,7 although its relationship with the intra-and extrahepatic bile ducts. 33,34 Histological data from a inflammatory bowel disease is well recognized. [8][9][10][11][12] Recent liver biopsy performed within 6 months of entry into the study was studies have implicated genetic and immunological mechaused in calculating the DSM score. Ludwig's criteria 38,39 were used nisms as factors in its pathogenesis. [13][14][15] PSC is a progressive in the staging of liver biopsy specimens. For purposes of statistical disease in most individuals 16,17 advancing at a variable rate analyses, numerical values were allotted to various stages, as in towards bile duct obliteration, biliary cirrhosis, and hepatic earlier studies (the value: 1 for stages 1 and 2; 2 for stage 3; and 4 failure. No effective treatment for this disorder has been iden-for stage 4). 29 tified, 18-22 and orthotopic liver transplantation remains the Predictive Models: CPC. 30-32 We used the Pugh modification of the classification originally proposed by Child and Turcotte, both with only option for those with end-stage disease. [23][24][25] and without the variation developed for cholestatic disease. The DSM Accurate methods of estimating survival in PSC are of imto which we compared the CPC employs the following four variables: portance in the context of classifying patients in therapeutic log e bilirubin, histological stage, age, and presence of splenomegaly. trials and in the decision-making process regarding the timThe DSM is used to predict the probability of surviving 1 to 7 ing and candidacy of orthotopic liver transplantation in eligi-years. 27,29 ble individuals. Disease-specific multivari...