Background: Pancreaticoduodenectomy, with either gastrectomy (Whipple procedure) or pylorus-preserving pancreaticoduodenectomy (PPPD), is a complex procedure. Technical diversity, variation and sampling bias exist among surgeons. Previous reports comparing these two procedures are retrospective and not randomized. These factors should be considered seriously and eliminated in comparisons between the two procedures.
Methods: From August 1994 to August 1997, a prospective randomized comparison was conducted between the Whipple procedure and PPPD performed by the same surgeon with the same approach and same anastomotic fashion for periampullary cancer. After exclusion of seven patients, 31 patients were eligible for the study, 16 receiving PPPD and 15 a Whipple procedure. No significant difference in the age, sex distribution, tumour localization or staging was noted between the two groups.
Results: One operative death after PPPD and no operative death after the Whipple procedure resulted in a 3 per cent mortality rate in the 31 patients. Median duration of the Whipple operation was 235 (range 195-305) min, with a median blood loss of 500 (range 230-3100) ml and a median blood transfusion of 0 (range 0-10) units. In the patients who had PPPD, median operating time was 230 (range 170-275) min, median blood loss was 350 (range 100-1200) ml and median blood transfusion was 0 (range 0-4) units. There were two minor leaks from the pancreaticojejunostomy after the Whipple procedure and no leakage after PPPD, resulting in 6 per cent minor leakage in 31 patients. These outcomes were not significantly different. Delayed gastric emptying was observed more frequently after PPPD (six of 16 patients) than after the Whipple procedure (one of 15 patients), with marginal significance (P = 0.08, two-sided Fisher's exact test).
Conclusion: In this prospective randomized study, both PPPD and the Whipple procedure were associated with low mortality and operative morbidity rates. There was no significant difference between PPPD and Whipple resection in terms of operative mortality and morbidity, operating time, blood loss and blood transfusion. PPPD was associated with more frequent delayed gastric emptying, although study of more patients is needed to confirm this.