Background The most ominous adverse event of primary sclerosing cholangitis (PSC) is development of cholangiocarcinoma (CCA). There is a wide variation in the reported diagnostic yield of bile duct brush cytology in PSC strictures. Objective To determine the diagnostic utility of biliary brush cytology for CCA detection in patients with PSC. Design Meta-analysis. Systematic search of PubMed, EMBASE, Web of Science, and the Cochrane Library for relevant studies published up to December 2012. Setting Meta-analysis of diagnostic parameters. Patients A total of 747 patients in studies (both retrospective and prospective) in which histopathologic correlation of CCA was available. Intervention Meta-analysis. Construction of 2 × 2 contingency data. Main Outcome Measurements Sensitivity, specificity, likelihood ratio, and pooled diagnostic odds ratio. Results The search yielded 54 studies of which 11, involving 747 patients, were included in our meta-analysis. The pooled sensitivity and specificity of bile duct brushings for a diagnosis of CCA in patients with PSC were 43% (95% confidence interval [CI], 35%-52%) and 97% (95% CI, 95%-98%), respectively. The pooled diagnostic odds ratio to detect CCA was 20.23 (95% CI, 8.75-46.79). The heterogeneity indices of χ2 statistics, I 2 measure of inconsistency, and the Cochran Q test were 0.156, 14.4, and 30.5%, respectively. Visual inspection of the funnel plot showed low potential for publication bias. Limitations Inclusion of low-quality studies, study heterogeneity. Conclusion Our study suggests that bile duct brushing is a simple and highly specific technique for detection of CCA in patients with PSC. However, the modest sensitivity from bile duct brushing precludes its utility as a diagnostic tool for early detection of CCA in patients with PSC.
Bibliographical noteFunding Information:
DISCLOSURE: U. Navaneethan received a research grant from the American College of Gastroenterology . J. Vargo is a speaking consultant for Olympus, Boston Scientific, Cook Endoscopy, and Ethicon. All other authors disclosed no financial relationships relevant to this publication.