The poor prognosis of pancreatic carcinoma after resection is related to distant metastases and local recurrence that is characterized by a strong tendency to infiltrate the retroperitoneal tissue and spread along the neural plexuses and lymph nodes. Thorough clearance of these tissues around the celiac and mesenteric axes, aorta, and inferior vena cava from the diaphragm to the inferior mesenteric artery (extended pancreaticoduodenectomy may lower the rate of local recurrence, but the procedure has been criticized for its higher morbidity and mortality. Our aim was to compare extended pancreaticoduodenectomy (EPD) with standard pancreaticoduodenectomy (SPD) in terms of postoperative morbidity and mortality. Data from 47 patients who underwent either EPD (n=24) or SPD (n=23) between November 1992 and October 1995 were retrospectively analyzed. Preoperative laboratory findings, operative risk (according to the American Society of Anesthesiologists classification), type of operation (classic Whipple vs. pylorus-preserving Whipple), operative time, intraoperative blood and plasma transfusion, postoperative morbidity and mortality, and postoperative hospital stay were scrutinized. The results showed that all of the parameters considered were similar in the EPD and SPD groups (intraoperative blood transfusion 800+/-490 ml vs. 700+/-586 ml, postoperative mortality 0% vs. 4.3%, overall morbidity 45.8% vs. 47.8%, surgical morbidity 37.5% vs. 34.7%, and postoperative hospital stay 16+/-8.1 days vs. 17+/-13.1 days. These two groups differed only in the operative time, which was significantly longer for EPD than for SPD (360+/-68.9 minutes vs. 330=66.9 minutes, P=0.02). Although the operative time is increased with EPD, there does not appear to be an increase in intraoperative complications, postoperative morbidity and mortality, or postoperative hospital stay with this procedure. However, definitive confirmation of these results can only be provided by a prospective randomized study.