Optimizing the detection of biliary dysplasia in primary sclerosing cholangitis before liver transplantation.

BACKGROUND: Patients with primary sclerosing cholangitis (PSC) have increased risk of cholangiocarcinoma (CCA). We evaluated pre-transplant work-up in PSC patients, to search for the most effective strategy for the detection of biliary dysplasia or early CCA. METHODS: Two hundred and twenty five consecutive PSC patients undergoing liver transplantation (LTx) in Sweden between 1999 and 2013 were studied. Patients with CCA or dysplasia in the explanted liver were compared with those with benign histopathology. Measures of test performance were calculated for patients having brush cytology on one endoscopic retrograde cholangiopancreaticography (ERCP) occasion, for those having repeated examinations with or without cholangioscopy, and for fluorescence in situ hybridization (FISH). Survival after LTx was analyzed. RESULTS: Brush cytology on a single ERCP occasion had moderate sensitivity (57%) and high specificity (94%) for the detection of CCA/high grade dysplasia (HGD) in the explanted liver. The corresponding sensitivity and specificity for FISH were 84% and 90%, respectively. Utilizing repeated ERCP and brush cytology to confirm the initial finding improved sensitivity to 82%. Using single operator cholangioscopy (SOC) for targeted examination at the second ERCP improved sensitivity (100%) and specificity (97%) significantly. Mortality rate in patients with incidentally discovered CCA (n = 16) in the explanted liver was significantly higher than in patients with HGD or benign histopathology (HR 16.0; 95% CI, 5.6-45.4; p < .001). CONCLUSIONS: Repeated brush cytology especially when combined with targeted examination under SOC guidance is superior to single brush examinations. This strategy improves the detection of malignancy in PSC and is of importance for selection of patients for LTx.