Laparoscopic Repair of Perforated Marginal Ulcer Following Roux-en-Y Gastric Bypass: A Case Series

Introduction

Marginal ulcer formation has a widely reported incidence after Roux-en-Y gastric bypass (RYGB), occurring in 0.6% to 16% of patients. ¹ Multiple etiologies have been proposed, including the presence of Helicobacter pylori infection, presence of a foreign body, smoking, nonsteroidal anti-inflammatory agent use, acid production, gastrogastric fistula, ischemia, use of nonabsorbable suture, and use of increased number of staple lines.^2–6 Patients may present with nausea, abdominal pain of varying severity, bleeding, and, occasionally, perforation. Although difficult to treat, most marginal ulcers are amenable to medical treatment, but, occasionally, the ulcer proceeds to perforation, presenting as a surgical emergency. The optimal means of managing perforated marginal ulcers after RYGB has yet to be established. Although patients may benefit from laparoscopic management, and thus avoid the morbidity associated with an open operation, only a few cases have been reported describing laparoscopic repair of a perforated marginal ulcer after RYGB using omental patch application.^7–9 We present the largest series to date describing laparoscopic omental repair of perforated marginal ulcer after RYGB.

Methods

A retrospective chart review from August 2005 to April 2007 was performed identifying all patients who underwent laparoscopic repair of perforated marginal ulcer after RYGB at one hospital by three different surgeons. The perforation was repaired either by laparoscopic primary suture closure followed by application of an omental patch or by laparoscopic Graham patch without primary suture repair. Data collected included presenting symptoms, operative time, duration of hospitalization, postoperative follow-up, and postoperative complications. All values are reported as mean ± standard deviation of the mean.

Results

Six patients were identified who had undergone laparoscopic repair of a perforated marginal ulcer. Three patients were male and 3 patients were female. Three of the 6 patients were smokers. Two patients were on preoperative acid reducing agents consisting of proton pump inhibitors. Time from RYGB to perforation ranged from 0.9 to 48 months. When presenting symptoms data was available, all patients had documented evidence of acute onset abdominal pain. Imaging studies performed at the study institution and thus available for review demonstrated free intraperitoneal air. One patient did present with a preperforation history of a gastrointestinal bleed. Demographic data is summarized in Table 1. Operative data and postoperative outcomes are described in Table 2. Four patients were treated with laparoscopic primary suture repair with omental patch reinforcement, whereas the other 2 patients were treated with laparoscopic omental patch repair without primary suture repair. Mean operative time was 101.8 ± 50 minutes with a mean hospitalization of 5.3 ± 2.7 days. Five patients had drains placed intraoperatively for postoperative drainage. Two patients had intraoperative feeding tubes placed for postoperative enteral nutrition. Postrepair hospital stay ranged from 2 to 10 days. Mean follow-up was 19.4 ± 15.2 months, with a range of 0.3 to 35.4 months. Four of the 6 patients experienced a complication or gastric complaint postoperatively. Two patients presented postoperatively with nausea and vomiting related to an exposed suture at the gastrojejunostomy. Both patients were treated with endoscopic suture removal. Of note, both patients did have suture closure of the perforation site before omental patch application. One patient with a preoperative stricture who underwent intraoperative endoscopic dilation during repair of the perforation redeveloped a gastrojejunal stricture 2.5 months postoperatively requiring subsequent endoscopic dilations and eventually revisional bariatric surgery. One patient was found to have gastritis postoperatively.

Discussion

We have presented the largest case series to date describing laparoscopic omental patch repair for perforated marginal ulcer after RYGB. We were able to achieve good results following this minimally invasive technique, thereby preventing the need for a laparotomy in the patients evaluated. Perforation of a marginal ulcer is typically considered a rare complication, but a recent large prospective database study reported a 1% incidence of perforated marginal ulcer after RYGB. ¹⁰ Although marginal ulcers have a relatively low rate of perforation, for a bariatric surgeon treating a large number of patients, this complication will undoubtedly present. However, the optimal means of treating these perforations has yet to be described in the literature. Several methods of laparoscopic management have been described, including primary suture closure, ¹¹ suture closure followed by gastrosplenic ligament patch and fibrin sealant, ¹² omental patch repair, ¹³ and laparoscopic omental patch repair with or without primary closure.^7–9 Laparoscopic omental patch application has only been described in a few cases before our report. We utilized omental repair via laparoscopic approach in 6 patients. We obtained good results in our small cohort of patients and avoided the morbidity associated with a laparatomy, especially in morbidly obese patients. Operative time was quite variable, but reasonable given the varied amount of adhesiolysis required. Four patients did suffer postoperative complications, one patient who suffered from gastritis, not necessarily attributed to the omental patch repair. In addition, the one patient who suffered from a postoperative stricutre refractory to dilations and eventually requiring revisional bariatric surgery had a preperforation stricture in the setting of chronic marginal ulcer disease, thus the continued stricture was likely in the setting of progressive disease rather than a new complication related to the omental patch repair. Two patients suffered from retained suture at the perforation site, and in these patients suture closure of the perforation followed by omental patch was performed. The suture could also have been present from the original gastric bypass; however, it would have been unlikely that the patients presented with nausea and vomiting due to a retained suture many months after the initial gastric bypass. Since 2/4 of the patients with suture closure presented with a retained suture requiring endoscopic suture removal, and the patients with omental closure did well postoperatively, suture closure can likely be omitted with possibly fewer complications. Regardless, both patients with a retained suture were treated with endoscopic removal of the suture. Considering the analysis of these complications, it can be concluded that overall laparoscopic omental repair of postgastric bypass perforation can be performed successfully with low rates of complications and low morbidity.

Felix et al. ¹⁰ has recently reviewed their series of perforated marginal ulcers after RYGB and found that 69% of those patients treated had identifiable, preventable risk factors, including smoking, use of nonsteroidal anti-inflammatory drugs, or steroid use. About 31% had no identifiable risk factor, but roughly one-third of these patients had a history of marginal ulcer after gastric bypass. Only 20% of those patients who developed marginal ulcer perforations or 0.2% of their total perforation had no identifiable risk factors or a history of a preperforation marginal ulcer. ¹⁰ Due to the small number of patients we treated, it is difficult to determine consistent specific causes of the marginal ulcer perforation, but 3 of the patients we treated were smokers, a likely contributing factor to the marginal ulcer and the subsequent perforation. On the basis of our findings and the literature, the optimal management of perforated marginal ulcers lies first in prevention, through risk factor reduction. However, in the event that a patient does develop a marginal ulcer perforation, we have demonstrated that a laparoscopic approach is feasible, with low operative times and short time to discharge from the hospital. This patient population must be followed carefully postoperatively as they are at risk for gastrojejunal stricture given the history of marginal ulcer and its predisposition for stricture formation.

In conclusion, we have presented the largest series to date of laparoscopic repair of perforated marginal ulcer utilizing an omental patch and describe the technical aspects of the laparoscopic repair. We demonstrate that a laparoscopic repair can be completed in a reasonable operative time, with minimal postoperative hospitalization and low associated morbidity. We beleive that patients who develop a perforated marginal ulcer after laparoscopic Roux-en-Y gastric bypass can be safely and effectively treated with laparoscopic repair with an omental patch.