Peptic ulcer/stricture after gastric bypass: a comparison of technique and acid suppression variables.

Background: Mason's original animal experiments on the gastric bypass (GBP) showed little acid production in the gastric pouch, a finding confirmed in humans. Despite this, GBP in humans is associated with an incidence of ulcer/stricture (U/S) at the gastrojejunostomy of 3 to 20%, with both acid secretion and staple-line dehiscence considered important risk factors or etiologies. Our series of GBP patients was reviewed to determine what technical or management factors, if any, were associated with U/S.

Methods: All patients undergoing first time GBP at Dartmouth-Hitchcock Medical Center by one surgeon from June 1991 until June 2000 were reviewed. The incidence of U/S as confirmed on upper endoscopy was determined by retrospective chart review. The technique of surgery, frequency of acid suppressive therapy at discharge, postoperative day of U/S diagnosis by endoscopy, length of follow-up with a member of the multidisciplinary bariatric team, and incidence of staple-line dehiscence were tabulated.

Results: 158 patients (72% female, mean BMI 53, mean age 42) underwent GBP. Two gastric stapling methods were used to create the gastric pouch: 4-rows (136 patients) and 8-rows (22 patients). No other technical feature was adjusted in the series. The two patient groups were similar in gender, age, and BMI. Acid suppressive therapy at the time of discharge was similar in each group with U/S (4-rows 64% and 8-rows 50%, p = 0.5). U/S developed in 12 (55%) of the 8-row group and in 14 (10%) of the 4-row group (p < 0.001). U/S typically occurred within the first 2 months postoperatively (mean 48 days, SD 40). No patients in our series developed a staple-line dehiscence.

Conclusion: U/S occur in the first few months following GBP. Twice the number of gastric staple-lines is associated with over five times the incidence of U/S, whereas post-discharge acid suppressive therapy is not predictive of U/S. Thus, a technique performed to decrease the risk of staple-line breakdown was associated with a much higher incidence of U/S. Staple-line dehiscence is not the etiology of this condition. Therefore, U/S after GBP does not appear to be explained by acid injury. We speculate that local, tissue injury related factors may be more responsible, a speculation that invokes a novel pathophysiologic mechanism for U/S formation following gastrojejunostomy.