Allowing normal food at will after major upper gastrointestinal surgery does not increase morbidity: a randomized multicenter trial.

ABSTRACT

OBJECTIVE: The aim of this trial was to investigate whether a routine of allowing normal food at will increases morbidity after major upper gastrointestinal (GI) surgery. SUMMARY BACKGROUND DATA Nil-by-mouth with enteral tube feeding is widely practiced for several days after major upper GI surgery. After other abdominal operations, normal food at will has been shown to be safe and to improve gut function. METHODS: Patients were randomly assigned to a routine of nil-by-mouth and enteral tube feeding by needle-catheter jejunostomy (ETF group) or normal food at will from the first day after major upper GI surgery. Primary end point was rate of major complications and death. Secondary outcomes were minor complications and adverse events, bowel function, and length of stay. All patients were invited to a follow-up at 8 weeks after discharge from the hospital. RESULTS: Four hundred fifty-three patients who underwent major open upper GI surgery in 5 centers were enrolled between 2001 and 2006. Four hundred forty-seven patients were correctly randomized. Of 227 patients 76 (33.5%) had major complications in the ETF group compared with 62 (28.2%) of 220 patients allowed normal food at will (P = 0.26, 95% CI for the difference in rate from -3.3 to 13.9). In the ETF group, 36 (15.9%) patients were reoperated compared with 29 (13.2%) in the group allowed normal food at will (P = 0.50) and 30-day mortality was 10 (4.4%) of 227 and 11 (5.0%) of 220 patients, respectively (P = 0.83). Time to resumed bowel function was significantly in favor of allowing normal food at will (P = 0.01), as were the total number of major complications, length of stay, and rate of postdischarge complications. CONCLUSIONS: Allowing patients to eat normal food at will from the first day after major upper GI surgery does not increase morbidity compared with traditional care with nil-by-mouth and enteral feeding.