Using CT and cholangiography to diagnose biliary tract carcinoma complicating primary sclerosing cholangitis

OBJECTIVE. The objective of our study was to determine the value of CT and cholangiography for diagnosing biliary tract carcinoma complicating primary sclerosing cholangitis. MATERIALS AND METHODS. One hundred thirteen abdominal CT examinations and cholangiograms in 45 patients with primary sclerosing cholangitis, including 18 patients with established biliary tract carcinoma, were analyzed for tumor. Four radiologists who were unaware of the presence or absence of carcinoma rated each study as to the probability of malignancy. Receiver operating characteristic curve analysis was used to assess the diagnostic performance of CT and cholangiography, the value of imaging signs, and the degree of interobserver variation in interpretation. Sensitivity and specificity values were calculated. RESULTS. CT outperformed cholangiography in the detection of carcinoma. The average area under the receiver operating characteristic curve was 0.82 for CT and 0.57 for cholangiography (p = 0.003). Sensitivity and specificity for detecting carcinoma using CT were good, with average values of 82% and 807o, respectively. Average sensitivity and specificity for cholangiography were 54% and 53%, respectively. The most reliable sign of tumor on CT was a discrete mass. Progressive biliary dilatation on sequential studies was the most useful sign on cholangiography. Interobserver agreement assessed using the Cronbach alpha was fair for cholangiography and good for CT. CONCLUSION. CT provides good sensitivity and specificity and significantly outperforms cholangiography in detecting biliary tract carcinoma complicating primary sclerosing cholangitis. Despite limitations, CT and cholangiography provide useful information not otherwise available in the treatment of patients with primary sclerosing cholangitis.