Objective: To compare the postoperative course of patients in the Dutch nationwide randomised trial of R1 (conventional) compared with R2 resection (including extended lymph node dissection) in the treatment of gastric cancer.
Design: Prospective randomised controlled trial.
Setting: National multicentre trial with 72 participating hospitals in The Netherlands.
Subjects: 192 patients who were operated on between August 1989 and May 1990.
Interventions: 96 patients were randomised for a R1, and 96 for a R2 resection.
Main outcome measures: Morbidity and mortality among 131 patients (64 R1 and 67 R2) for whom the resection was performed with curative intent.
Results: The groups were comparable for age, sex, type of resection, site of tumour and depth of invasion. Complications developed in 23 R1 (36%) and in 29 R2 patients (43%). Seven patients died in the postoperative period. Median hospital stay was significantly longer after R2 (18 days, range 7-122) than after R1 resection (15 days, range 2-63) (p < 0.05). Morbidity and mortality among the patients whose R2 resection was done by the Japanese instructor (n = 34) did not differ significantly from those among patients operated on by the Dutch supervisors (n = 33), but those operated on by the Japanese instructor stayed in hospital significantly longer (20 compared with 16 days, p < 0.05).
Conclusions: If R2 resections are carried out by properly trained surgeons under supervision, they can be done safely. The reported high morbidity after R2 resection in Western countries seems to result from a lack of proper instruction and quality control.