Brush cytology for pancreatic carcinoma: An analysis of factors influencing results

Background: Despite recent advances in cytology brush design, yield of endoscopic brush cytology in suspected pancreatic carcinoma remains low. Methods: We prospectively evaluated 32 such patients by ERCP to analyze differences in yield based on anatomic location of the pancreatic stricture, and the role of concurrent biliary stricture brush cytology, in improving the overall yield. Endoscopic brush cytology was performed on all strictures following ERCP. A final diagnosis of pancreatic carcinoma was confirmed in all patients. Results: Twenty-three of the 32 patients had positive cytology for pancreatic malignancy (71.9%). Eight patients had positive brushings from biliary strictures alone (25%) and 15 had positive brushings obtained from pancreatic strictures (46.9%). The yield varied widely depending on the anatomic location of the stricture; ampullary, genu, and tail regions had low rates of positive cytology, in part due to technical factors and brush design (1 of 8, 2 of 6, and 1 of 4, respectively). Strictures of the head and body yielded high rates of positive cytology (7 of 8 and 4 of 6, respectively). Conclusions: The yield of endopancreatic brush cytology is related to the location of malignancy, with overall yield enhanced by concurrent brushing of bile duct strictures.