default inclusion

The incidence of clinically significant anastomotic leaks (AL) after upper gastrointestinal (GI) surgery is approximately 4 % - 20 %, and the associated mortality can be as high as 80 % . Nutritional support is a key component of therapy in such cases, related to high prevalence of malnutrition and nil per month required for leak treatment.

In the prophylactic setting, before the occurrence of any AL, a literature review based on seven randomised trials showed that enteral nutrition (EN) is associated with shorter hospital stay, lower incidence of severe of infectious complications, lower severity of complications and decreased cost compared to parenteral nutrition (TPN) following major upper GI surgery .

In the curative setting, after the AL occurrence, very few evidence is available. Only one randomized clinical trial suggested the superiority of EN versus TPN after pancreatic surgery with a increase of the 30-day fistula closure rate from 37% in the TPN group to 60% in the EN group .

This sole randomised study available did not include all postoperative upper GI AL (PUGIAL) that can occur after esophageal, gastric, duodenal, pancreatic surgery (including obesity surgery), whereas the concept of enteral nutritional support is highly relevant for all these situations.

However surgeons are usually reluctant to provide EN in case of AL. A randomized study suggested the feasibility of EN in 47 patients with upper GI AL but no randomized study to date has been designed to test the superiority of EN versus TPN in PUGIAL.

The study aim is to demonstrate the superiority of EN versus TPN to accelerate AL healing after upper GI surgery.

Hypothesis:

EN increases the 30-day fistula closure rate in PUGIAL, allowing better HRQOL without increasing morbi-mortality.

Inclusion Criteria:

• Patients ≥ 18 years
• Patients having received upper GI surgery for benign or malignant disease including, oesophageal, gastric, duodenal or pancreatic surgery or bariatric surgery
• Diagnosis of an active postoperative fistula or persisting after a failure of surgical or endoscopic procedure dedicated to close the fistula
• AL diagnosed from less than 72h AND confirmed on at least two criteria among the followings:
• clinical symptoms of AL
• ct scan / ultrason imaging and /or endoscopic diagnosis of AL
• biologic/bacteriology diagnosis on fluid output
• intraoperative diagnosis of AL at time of reoperation
• Indication of nil per mouth
• American society of anesthesiologist score 1, 2 or 3
• In case of neoplasm, absence of peritoneal carcinomatosis or distant metastasis
• No severe concomitant uncontrolled disease
• Life expectancy more than 6 months
• No history of allergy or study product intolerance
• Ongoing healthcare insurance

Exclusion Criteria:

• - Scheduled surgical or endoscopic treatment with the aim to close the fistula (suture, prosthesis, clip or glue). In case of such treatment failure, patients are eligible to participate to the study. Endoscopic or surgical drainage are not exclusion criteria (meaning that drainage only is authorized before randomization)
• History of or current severe uncontrolled cardiovascular, pulmonary, renal or liver failure
• Uncontrolled sepsis related to the AL
• Malnutrition requiring combined nutritional treatment with the enteral AND parenteral routes together
• Untreated or persistent Peritoneal carcinomatosis or distant metastasis
• Pregnant and/or lactating women
• Freedom privacy