AIM: To evaluate the significance of extended radical operation and its indications.METHODS: Between January 1995 and December 1998, 56inpatients with pancreatic head cancer received operation.Among them 35 patients (group 1) experienced the Whipple operation, and 21 patients (group 2) received the extended radical operation. The 1-, 2-, 3-year cumulative survival rates were used to evaluate the efficacy of the two operative procedures. Clinical stage (CS) was assessed retrospectively with the help of CT. The indications for extended radical operation were discussed.RESULTS: There was no difference in hospital mortality and morbidity rates. Whereas the 1-, 2-, 3-year cumulative survival rates were 84.8%, 62.8%, 39.9% in the extended radical operation group, and were 70.8%, 47.6%, 17.2%in the Whipple operation group, there was a significant difference between the two groups (P＜0.001, P＜0.001,P＜0.001, respectively). Most of the deaths within 3 years after operation were due to recurrence in the two groups.However, the 1-, 2-, 3-year cumulative rates of death due to local recurrence were decreased from 37.4% in patients that received the Whipple procedure to 23.8% in those who received by extended radical operation. Patients who survived for more than 3 years were only noted in those with CS1 in the Whipple procedure group and were founded in cases with CS1, CS2 and part of CS3 in the extended radical operation group.CONCLUSION: The extended radical operation appears to benefit patients with pancreatic head carcinoma which was indicated in CS1, CS2 and part of CS3 without severe invasion.