Objective: Because large-scale reports of PSC in the Mediterranean area we are still lacking, in this study we evaluated by Kaplan-Meyer analysis the natural history of primary sclerosing cholangitis (PSC) in Italy and by means of other statistical methods we identified the variables most useful in predicting survival of such patients.
Design: Retrospective multicentre study of unselected patients with PSC. Several variables involving sex, age, associated diseases, clinical features, laboratory, cholangiographic and histological findings at presentation and clinical outcome at data recording were collected by means of a detailed questionnaire.
Setting: 16 Italian university and regional hospitals all over the country, thus giving a geographically representative population.
Patients: A total of 117 PSC patients (73 men and 44 women); median age 35 years.
Methods: Survival analysis was performed by the Kaplan-Meyer method; the prognostic influence on survival of collected data was evaluated by univariate chi(2) analysis with Wilcoxon and log-rank tests. The same prognostic variables were also evaluated by multivariate analysis (Cox model), using a stepwise regression procedure. All statistical analyses were performed using the SAS statistical software.
Results: At presentation 70% of patients were symptomatic; symptoms did not relate to liver histology. Both intra- and extrahepatic bile duct lesions were detected in 46% of patients at cholangiography. Inflammatory bowel disease was found in 54% of symptomatic patients, ulcerative colitis was 36% of total. Clinical outcome (91/117): 15 underwent liver transplantation or died from liver disease (cholangiocarcinoma). Survival at 10 years was 74%. Features of poor prognosis were cholesterol, aspartate aminotransferase (AST), haemoglobin and albumin.
Conclusion: PSC in Italy mainly follows a benign course and among clinical features recorded at presentation, serum cholesterol, AST, haemoglobin and albumin may provide some objective criteria to assess disease severity.