Aim: No useful indicator is available for preoperative risk evaluation when considering treatment with pancreatic surgery. The aim of this study was to clarify a usefulness of the POSSUM audit system (Physiological and Operative Severity Score for the Enumeration of Mortality and Morbidity) in patients with pancreatic resection. Methods: From May 1996 to April 2004, 201 patients underwent pancreatic resection. The operative procedure was pancreaticoduodenectomy (PD) in 148 cases and distal pancreatectomy (DP) in 53 cases. The underlying disease was pancreas cancer in 82 cases, intrapapillary mucinous tumor (IPMT) in 25 cases, bile duct cancer in 24 cases, ampullary cancer in 22 cases, gastric cancer in 32 cases, duodenal cancer in 5 cases and chronic pancreatitis in 11 cases. There were 8 hospital deaths (4.0%). Postoperative complications were seen in 67 cases (33.3%). The subjects were divided into two groups by the presence or absence of postoperative complications. Eighteen preoperative factors and 5 operative factors, and POSSUM parameters were compared between these two groups. Results: In terms of the 18 preoperative factors, there was no significant difference between the complicated group and the non-complicated group. In the 4 operative factors, the operative time (p=0.0004) and the blood loss (p=0.0026) differed significantly between the two groups. Of the POSSUM parameters analyzed, the physiological score (p=0.0001), the operative score (p=0.0002), the predicted mortality rate (p=0.0001) and the predicted morbidity rate (p=0.0001) differed significantly between the two groups. Conclusion: POSSUM is useful as a means of risk assessment in individuals scheduled to undergo pancreatic resection.