Background & aims: We report the usefulness of cholangioscopy in patients with indeterminate pancreaticobiliary pathology.
Methods: A prospective collection of 62 consecutive patients during a period of 2.5 years who were referred to our tertiary referral center for cholangioscopy for indeterminate strictures suspicious for malignancy were included. Tissue sampling followed cholangioscopic visualization. Biopsies were obtained under direct visualization (cholangioscopy-directed) or through the duodenoscope (cholangioscopy-assisted).
Results: Sixty-two patients had 72 examinations. Forty patients had nondiagnostic sampling before cholangioscopy. Indications were stricture (n = 67: 16 primary sclerosing cholangitis, 51 non-primary sclerosing cholangitis), ductal dilation, or intraductal mass (n = 5). Biopsies were not performed in 19 because cholangioscopy did not identify suspicious lesions. Of the remaining 53 procedures, 29 underwent either cholangioscopy-directed or cholangioscopy-assisted biopsy, and 24 had both. Cholangioscopy findings consisted of primary sclerosing cholangitis only (n = 18), benign stricture or inflammatory changes (n = 18), bile duct cancer (n = 14), normal (n = 10), pancreatic cancer (n = 5), and other (n = 7). Fifty-eight patients (94%) had follow-up for a mean of 12.4 months (95% confidence interval, 10.1-14.7). Sixteen of 18 (89%) patients with a final diagnosis of malignancy were detected with cholangioscopy. The 2 missed cancers were intrahepatic cholangiocarcinomas. Overall, sensitivity to detect malignancy by cholangioscopy with and without biopsy was 89%, specificity 96%, positive predictive value 89%, and negative predictive value 96%.
Conclusions: Cholangioscopy with and without biopsy is highly accurate in diagnosing and excluding pancreaticobiliary malignancy in patients with indeterminate strictures.