Chronic Marginal Ulcers After Gastric Bypass

The epidemic of overweight and obesity in the United States of America along with its comorbidities continues to expand. Bariatric surgery has demonstrated to be the most effective and sustained method to control severe obesity and its comorbidities. For instance, type 2 diabetes mellitus was completely resolved in 76.8%, systemic arterial hypertension was resolved in 61.7%, dyslipidemia improved in 70% and obstructive sleep apnea-hypopnea syndrome was resolved in 85.7%. Furthermore, bariatric surgery significantly increases life expectancy (89%) and decreases overall mortality (30-40%), particularly deaths from diabetes, heart disease, and cancer. Lastly, preliminary evidence about downstream savings associated with bariatric surgery offset the initial costs in 2 to 4 years.

Since 1998, there has been a substantially progressive increase in bariatric surgery. In 2005, the ASMBS reported that 81% of bariatric procedures were approached laparoscopically and in 2007, 205,000 people had bariatric surgery in the United States from which approximately 80% of these were Gastric Bypass. Moreover, there is a mismatch between eligibility and receipt of bariatric surgery with just less than 1% of the eligible population being treated for morbid obesity through bariatric surgery. Along with the increasing number of elective primary weight loss procedures, up to 20% of post RYGB patients cannot sustain their weight loss beyond 2 to 3 years after the primary bariatric procedure. Thus, revisional surgery for poor weight loss and re-operations for technical or mechanical complications will rise in a parallel manner.

A common late complication after gastric bypass surgery is marginal ulceration, an ulcer at the margins of the gastrojejunostomy on the jejunal side. Its incidence after RYGB ranges from as low as 0.6 to as high as 16%. After 1,040 laparoscopic RYGB surgeries, the incidence rate, in our hands, is 1.4% and mainly related to NSAID´s use. In observational cohort studies, the presence of specific technical factors - staple-line dehiscence or gastro-gastric fistula, enlarged pouch, foreign material and local ischemia - and environmental factors - tobacco, NSAID´s, alcohol consumption, and H pylori infection among others - have been associated with marginal ulceration however the exact etiopathogenesis has not been completely elucidated.

Similar to peptic ulcer disease (PUD), most marginal ulcers respond to medical therapy, specifically sucralfate and acid-lowering medication. In contrast, when perforation, obstruction, penetration, bleeding and/or intractability presents, complex or complicated ulcer disease, warrants surgical intervention.

The intestinal mucosa is not typically exposed to gastric acid, which is neutralized by the alkaline biliopancreatic secretions. The jejunal mucosa has no natural barriers; when exposed to gastric acid, it ulcerates easily. Capella & Capella demonstrated that transecting the gastric segments significantly reduce staple-line dehiscence; this is the so-called divided gastric bypass. In the retrospective analysis of their consecutive series, the incidence for gastro-gastric fistula (GGF) formation after undivided gastric bypass (GBP) was 23%, after a partially divided GBP was 19%, after a completely divided GBP was 2% and after complete transection with interposition of the jejunal limb was 0% (p <0.001). MacLean et al confirmed that divided primary gastric bypass decreases GGF formation (29% vs. 3%). Also, patients who developed marginal ulcers had a lower pH as well as a greater time with a pH less than 2 correlating 100% with the presence of GGF; closure of the GGF increased the pH in the pouch with subsequent healing of the marginal ulcer.

An unusually large gastric pouch (such as horizontal pouches, retained fundus, long lesser curvature based pouches or enlarged after initially being sized adequately) contain more acid-producing parietal cells. Increased acid production in the pouch carries the risk of developing marginal ulcers. Acid secretion in the small pouch after RYGB is virtually absent. Smith et al measured basal and pentagastrin-stimulated gastric acid secretion from the pouch were significantly lower compared to age and sex-matched controls. Likewise, MacLean et al reported a significantly lower pH & greater time with pH <2 in the gastric pouches of marginal ulcers and/or GGF patients after RYGB compared to non-complicated RYGB controls. Thus, creating a esophagojejunostomy would solve the gastric acid factor for developing marginal ulcers however the high incidence of anastomotic failure and unknown weight loss results are prohibitive for this approach. Sapala et al created a micro-pouch or cardiojejunostomy to decrease at maximum the parietal cell mass with a low incidence of marginal ulcers (0.01% at 1 years of follow-up) as well as to limit the pouch dilation. By Histopathology with a semi-quantitative approach, Gustavsson et al reported less acid-producing parietal cells within smaller pouches. With his next study (n=12), Gustavsson et al, demonstrated a significantly higher time exposure to a pH<4 in patients with marginal ulcer after RYGB (4x3cm pouch) compared to controls (p< 0.01). Furthermore, after downsizing the pouch, repeated pH-metry showed the % of time with pH <4 declined from 100% prior to 6% after revisionary surgery.

The anastomotic techniques influence the incidence of marginal ulcers. Capella & Capella reported a consecutive series with significant decrement from 5.1% to 1.5% (p< 0.001) after switching from a stapled to a hand-sewn anastomosis. Likewise, after changing from an inner layer of absorbable suture and an outer layer of nonabsorbable material to a double-layer of absorbable suture the incidence rate improved from 1.6% to 0%. Dr Schauer´s group confirmed a significant improvement in the incidence rate of MU from a 2.6% with the use of nonabsorbable suture for the outer layer to 1.3% after the change to absorbable suture for both layers (p < 0.001).

Local ischemia, in the immediate postoperative period, is probably secondary to technical reasons. Fundamental aspects for decreasing tension and local ischemia at the gastrojejunostomy are dissection of the tissues around the pouch without devascularizing the lesser curvature and complete mobilization of a well-perfused Roux limb.

In epidemiological, clinical and experimental studies, NSAID´s have been identified as one of the three major risk factors for PUD. Wilson et al found NSAID´s consumption to significantly increase the risk for marginal ulcer following RYGB (adjusted OR 11.5, 95%CI 4.8-28).

In epidemiological, clinical and experimental studies, Tobacco is another major risk factor for PUD. Smoking carries an overall relative risk of 2.2 (95%CI, 2.0-2.3).

Helicobacter pylori (H pylori) infection carries an overall relative risk of 3.3 (95%CI, 2.6-4.4) for developing PUD. A synergistic relationship exists between H pylori infection and NSAID´s consumption for developing PUD with an overall risk of 3.5 (95%CI, 1.26-9.96) compared to either H pylori or NSAID´s negative individuals. In Papasavas et al study, preoperative H. pylori testing with prophylactic eradication did not decrease the incidence of MU or erosive pouch gastritis.

The pathophysiological mechanisms of damage to the gastric mucosa of ethanol and alcoholic beverages are poorly understood. There are no studies available about the effect of alcohol on marginal ulcer development after RYGB.

Cocaine use is responsible for approximately 143,000 Emergency Department visits annually; 19% of American, between 18 to 25 years old, have used cocaine: more than 1% of the Americans use cocaine at least once a week; and approximately 50% of all drug-related deaths were secondary to Cocaine. The temporal association between smoking cocaine (crack) and GI tract manifestations include ulceration, perforation, visceral infarction, and retroperitoneal fibrosis.

Re-operative strategies for addressing chronic marginal ulcers after gastric bypass have been scarcely described and mostly are reports of a case or small series of cases. The revisional strategies described are I) ulcer excision with revision of the gastrojejunostomy and gastric transection if needed, II) ulcer excision with pouch downsizing and redo of gastrojejunostomy, III) ulcer excision with resection of the ischemic Roux limb segment, and IV) ulcer excision and reversal. The possible adjuvant procedures includes I) proximal remnant gastrectomy (partial gastrectomy), and II) vagotomy.

In summary, there is scant information about late complications after gastric bypass especially after the widespread adoption of the laparoscopic approach and the modern anatomical construct of Roux-en-Y Gastric Bypass surgery.