Endoscopic Ultrasound-Guided Fine-Needle Aspiration of Pancreatic Duct Fluid in the Evaluation of Pancreatic Cancer

Abstract

or acute pancreatitis of probable biliary origin) with a non-dilated biliary tract on US (Group II; n Z 82). Patients with personal history of claustrophobia or gastroenteroanastomosis and peace-maker carriers were excluded. Radial EUS (Olympus GF UM20 and GF UM160) and MRCP with SSFSE technique were performed within a 24-hour period. The sequence of the two techniques was randomly assigned and the operators were blinded with respect to the results of the other procedure. Gold standard was endoscopic retrograde cholangiopancreatography (ERCP) (n Z 86), surgery with intraoperative cholangiography when needed (n Z 23) or clinical follow-up (n Z 26). Results: Of 156 patients (69 men, 87 women; mean age of 68 G 15) included, 21 were excluded for: lack of definitive diagnosis (n Z 8), EUS (n Z 6) or MRCP (n Z 7). Therefore, a total of 63 and 72 patients were evaluated in Groups I and II respectively for the analysis. Mean time between EUS/MRCP and ERCP or surgery was 11 G 13 and 31 G 32 days respectively. Mean follow-up in patients in whom this was considered as gold standard was 5.8 G 3.9 months. Final diagnosis was choledocolitiasis (Group I/II: n Z 31/30), cholelithiasis without choledocolithiasis (Group I/II: n Z 2/5), pancreatic cancer (GroupI/II: n Z 17/2), cancer of the biliary tract (Group I/II: n Z 5/3), others (GroupI/II: n Z 2/0) and lack of pathologic findings (Group I/II: 6/ 32). The agreement between the two techniques was 84% and 72% in Groups I and II respectively, as assessed by kappa statistics. There were no differences in the performance characteristics of EUS and MRCP in the two study groups (table):