Clinical usefulness of repeated pancreatic juice cytology via endoscopic naso-pancreatic drainage tube in patients with pancreatic cancer

Background Cytological examination of pancreatic juice obtained during endoscopic retrograde cholangiopancreatography (ERCP) is well established, but its sensitivity for pancreatic cancer has not been satisfactory. The aim of this study was to evaluate the usefulness of repeated pancreatic juice cytology (PJC) via the endoscopic naso-pancreatic drainage (ENPD) tube in patients with pancreatic cancer compared with conventional PJC. Methods We retrospectively investigated 139 patients with pancreatic disease. Between April 2004 and November 2007, conventional PJC was performed in 56 patients with pancreatic cancer and 23 with benign pancreatic stricture. Between January 2008 and November 2010, ENPD was used in 40 patients with pancreatic cancer and 20 with benign pancreatic stricture. The ENPD tube was placed into the main pancreatic duct for up to 3 days, and cytological samples of pancreatic juice were collected up to 6 times in total. Results Sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy of the ENPD method for pancreatic cancer were 80, 100, 100, 71, and 87 %, respectively, revealing significantly higher sensitivity than the conventional method ( p = 0.0001). Sensitivities according to tumor location and size were 90 % (19/21), 69 % (9/13), and 67 % (4/6) in the head, body, and tail of the pancreas, 88 % (7/8), 79 % (19/24), and 75 % (6/8) in tumors with a diameter less than 20 mm including carcinoma in situ, 21–40, and greater than 41 mm, respectively. Conclusions The ENPD method was found to have high diagnostic yield, especially for tumors less than 20 mm or located in the pancreatic head, and might be useful for the diagnosis of early-stage pancreatic cancer.