BACKGROUND & AIMS: Long-term outcomes of patients with branch-duct intraductal papillary mucinous neoplasms (IPMNs), particularly those after 5 years of surveillance, have not been fully evaluated in large studies. We analyzed incidences of IPMN-derived carcinoma and concomitant ductal adenocarcinoma (pancreatic ductal adenocarcinoma [PDAC]) over 20 years in a large population of patients. METHODS: We identified 1404 consecutive patients (52% women; mean age, 67.5 years) with a diagnosis of branch-duct IPMN, from 1994 through 2017, at the University of Tokyo in Japan. Using a competing risk analysis, we estimated cumulative incidence of pancreatic carcinoma, overall and by carcinoma type. We used competing risks proportional hazards models to estimate subdistribution hazard ratios (SHRs) for incidences of carcinomas. To differentiate IPMN-derived and concomitant carcinomas, we collected genomic DNA from available paired samples of IPMNs and carcinomas and detected mutations in GNAS and KRAS by polymerase chain reaction and pyrosequencing. RESULTS: During 9231 personyears of follow-up, we identified 68 patients with pancreatic carcinomas (38 patients with IPMN-derived carcinomas and 30 patients with concomitant PDACs); the overall incidence rates were 3.3%, 6.6%, and 15.0% at 5, 10, and 15 years, respectively. Among 804 patients followed more than 5 years, overall cumulative incidence rates of pancreatic carcinoma were 3.5% at 10 years and 12.0% at 15 years from the initial diagnosis. The size of the IPMN and the diameter of the main pancreatic duct associated with incidence of IPMNderived carcinoma (SHR 1.85; 95% confidence interval 1.38-2.48 for a 10-mm increase in the IPMN size and SHR 1.56; 95% confidence interval 1.33-1.83 for a 1-mm increase in the main pancreatic duct diameter) but not with incidence of concomitant PDAC. CONCLUSIONS: In a large long-term study of patients with branch-duct IPMNs, we found the 5year incidence rate of pancreatic malignancy to be 3.3%, reaching 15.0% at 15 years after IPMN diagnosis. We observed heterogeneous risk factor profiles between IPMNderived and concomitant carcinomas.

Endoscopic ultrasound (EUS)-guided biliary drainage (EUS-BD) is potentially complicated by bile leak and stent migration. The aim of this study was to evaluate the safety and effectiveness of a long (≥ 10 cm), partially covered metal stent (LP-CMS) for EUS-guided hepaticogastrostomy (EUS-HGS) for malignant biliary obstruction. Both the stent length and the uncovered portion at the proximal end of the LP-CMS are designed to prevent stent migration. A total of 33 patients undergoing EUS-HGS using an LP-CMS in four centers were retrospectively studied. Technical and clinical success, adverse events, and recurrent biliary obstruction were evaluated. Gastric outlet obstruction (76 %) and surgically altered anatomy (15 %) were two major reasons for EUS-HGS. The technical and clinical success rates were 100 %. The median intragastric stent length was 54 mm. The adverse event rate was 9 %. No stent migration was observed. Recurrent biliary obstruction developed in 24 %, with a median cumulative time to recurrence of 8.5 months. EUS-HGS using an LP-CMS for unresectable malignant biliary obstruction was safe and effective.

No abstract

No abstract