Background: Bursectomy is regarded as a standard surgical procedure during gastrectomy for serosa-positive gastric cancer in Japan. There is little evidence, however, that bursectomy has clinical benefit. We conducted a randomized controlled trial to demonstrate non-inferiority of treatment with the omission of bursectomy.
Methods: Between July 2002 and January 2007, 210 patients with cT2-T3 gastric adenocarcinoma were intraoperatively randomized to radical gastrectomy and D2 lymphadenectomy with or without bursectomy. The primary endpoint was overall survival (OS). Secondary endpoints were recurrence-free survival, operative morbidity, and levels of amylase in drainage fluid on postoperative day 1. Two interim analyses were performed, in September 2008 and August 2010.
Results: Overall morbidity (14.3%) and mortality (0.95%) rates were the same in the two groups. The median levels of amylase in drainage fluid on postoperative day 1 were similar in the two groups (P = 0.543). In the second interim analysis, the 3-year OS rates were 85.6% in the bursectomy group and 79.6% in the non-bursectomy group. The hazard ratio for death without bursectomy was 1.44 (95% confidence interval [CI] 0.79-2.61; P = 0.443 for non-inferiority). Among 48 serosa-positive (pT3-T4) patients, the 3-year OS was 69.8% for the bursectomy group and 50.2% for the non-bursectomy group, conferring a hazard ratio for death of 2.16 (95% CI 0.89-5.22; P = 0.791 for non-inferiority). More patients in the non-bursectomy group had peritoneal recurrences than in the bursectomy group (13.2 vs. 8.7%).
Conclusions: The interim analyses suggest that bursectomy may improve survival and should not be abandoned as a futile procedure until more definitive data can be obtained.