Abstracts
2000 Digestive Disease Week

# 2308 Preoperative Endoscopic Ultrasound in Pancreatic Malignancy.
Frank Makowiec, Alexander Pietsch, Albrecht Holle, Hartwig Riediger, Ulrich Adam, Stefan Liebe, Ulrich T. Hopt, Rostock, Germany

Despite of the availability of modern imaging methods staging and assessment of the resectability of malignant pancreatic tumors is sometimes difficult. During the recent years endoscopic ultrasound (EUS) has been used increasingly in the locoregional staging of those tumors. The aim of our study was to assess the value of EUS as compared with operative (OP) and pathological (Path) findings. METHODS: Fifty-four patients (39 with pancreatic, 10 with periampullary and five with distal bile duct cancer) with completed preoperative EUS were evaluated. All EUS were performed by one experienced gastroenterologist. A resection was performed in 49 patients, five patients underwent palliative procedures with intraoperative staging and pathological workup. The combination of OP and Path served as gold standard to assess accuracy of EUS. RESULTS: The mean tumor diameter was 3.6 cm (EUS) and 3.1 cm (Path/ OP), respectively. A false estimation of tumor size by EUS more than one cm difference to OP/Path occured in 24 %. The T stage (T1: n=3; T2: n=6; T3: n=28; T4: n=17) was assessed correctly by EUS in 59 %. An overestimation of the T stage occured in 4 %, an underestimation in 37 % of patients. The results of EUS to assess different parameters are shown in detail in the table below. The frequent underestimation of the T stage by EUS was mainly due to missed demonstration of infiltration (in Path frequently only microscopic) of the duodenum or peripancreatic fat. CONCLUSIONS: The accuracy of EUS in preoperative staging and assessment of resectability is high. The underestimation of the T stage is often due to microscopic peripancreatic disease which is not relevant in the preoperative planning of pancreatic resection.