Incidence and management of enteric leaks after gastric bypass for morbid obesity during a 10-year period

Background: To describe the incidence, etiology, outcomes, and management of enteric leaks in patients who had undergone open or laparoscopic gastric bypass. Methods: From November 1996 to November 2006, 1133 patients underwent primary gastric bypass at Columbia University, New York-Presbyterian Hospital. A retrospective review of our prospective bariatric surgery registry identified 17 patients (1.5%) who developed a clinically apparent enteric leak after surgery. The demographic and outcome data were studied. Results: The mean body mass index was 52 kg/m² (range 35-65), and 15 (88%) of the 17 patients were women. The mean number of co-morbidities per patient was 1.3 and included hypertension in 11, diabetes in 9, and sleep apnea in 6. Ten patients had previously undergone abdominal surgery. The enteric leak was diagnosed by radiographic studies in 12 patients (9 of 12 by upper gastrointestinal series and 3 of 6 by computed tomography); the remaining 5 patients were diagnosed at re-exploration. Of the 17 patients, 12 (70%) were treated by laparoscopy and 5 (29%) by laparotomy. The mean time from completion of the index procedure to the diagnosis of the leak was 2 days (range 1-5) for patients treated by laparoscopy versus 4 days (range 1-6) for patients treated by laparotomy (P <.05). The patients treated by laparoscopy experienced a shorter hospital stay, but the difference was not statistically significant (mean 11.4 days, range 6-36, versus 18 days, range 7-33; P >.05). One of the laparoscopic patients (5.9%) died. Conclusion: Enteric leak is a significant complication after gastric bypass. Prompt treatment should be based on clinical suspicion, because contrast and cross-sectional imaging studies might not be reliable diagnostic tests. A laparoscopic index procedure might be associated with an earlier diagnosis.