Ulcer disease after gastric bypass surgery

Ulcer disease after gastric bypass surgery

Surgery for Obesity and Related Diseases 2 (2006) 455– 459 Original article Ulcer disease after gastric bypass surgery Ramsey M. Dallal, M.D., F.A.C...

118KB Sizes 2 Downloads 124 Views

Surgery for Obesity and Related Diseases 2 (2006) 455– 459

Original article

Ulcer disease after gastric bypass surgery Ramsey M. Dallal, M.D., F.A.C.S.*, Linda A. Bailey, P.A.C. Albert Einstein Healthcare Network, Philadelphia, Pennsylvania Received February 3, 2006; revised March 4, 2006; accepted March 8, 2006


Background: The mechanism of marginal ulceration after laparoscopic gastric bypass surgery is poorly understood. We reviewed the incidence, presentation, and outcome of ulcer disease in consecutive patients undergoing laparoscopic gastric bypass surgery. Methods: The outcomes of 201 consecutive laparoscopic gastric bypass surgery procedures were prospectively analyzed for complications. All procedures were performed using a linear stapled anastomosis and absorbable suture. Results: The incidence of marginal ulcer disease was 3.5% (7 patients). One patient, the only smoker, presented with an acute perforation 4 months postoperatively. Three other patients presented with bleeding—all required transfusion. The remaining 3 patients presented with severe pain. At endoscopy, all patients had ulcerations associated with the Roux limb mucosa and were all successfully treated using proton pump inhibitors and sucralfate therapy. Symptoms of marginal ulceration occurred an average of 7.4 months (range 3–14) after surgery. The average follow-up was 19.8 months. No preoperative factors were predictors of ulcer disease, including body mass index, age, gender, or co-morbidities. Conclusion: Marginal ulcers using the linear-stapled technique occurred in 3.5% of patients. Three distinct clinical presentations occurred: bleeding, pain, or perforation. No preoperative risk factors were identified that predicted for this complication. Medical management is an effective treatment. © 2006 American Society for Bariatric Surgery. All rights reserved.


Stomal ulcer; Marginal ulcer; Ulcer; Gastric bypass; Complications

Marginal ulceration after gastric bypass surgery can be a surprisingly morbid complication. Ulceration can present with severe pain, nausea, unheralded bleeding, or, rarely, perforation. Patients with painful ulcers are particular difficult to treat because medical therapy is not often immediately effective at relieving severe discomfort. Ulcer disease has been reported in 1–16% of patients after gastric bypass surgery [1– 4]. The incidence seems to vary widely depending on the surgical technique, and theories about marginal ulcer etiology are abundant, including acid, fistula, ischemia, foreign body reactions, inflammation and Helicobacter pylori [5– 8]. Although the published data are scant, revision surgery has been performed for recalcitrant ulcer disease with mixed success [5,9]. *Reprint requests: Ramsey Dallal, M.D., Einstein at Elkins Park, 60 East Township Line Road, Elkins Park, PA 19027. E-mail: [email protected]

Our aim was to document the incidence, patient characteristics, and treatment of patients who developed marginal ulceration after laparoscopic gastric bypass surgery. Furthermore, we wanted to describe ulcer disease using a single surgical technique performed in a group of patients after the surgeon’s learning curve in the setting of high patient follow-up. Methods Our prospectively collected database was reviewed to identify patients who were diagnosed with a marginal ulceration after undergoing gastric bypass surgery. A single attending surgeon performed all procedures. Patients who had undergone surgery between July 2003 and July 2005 were included for evaluation and follow-up up to January 2006 was included. These patients had undergone surgery after the surgeon’s learning curve and after the surgeon had

1550-7289/06/$ – see front matter © 2006 American Society for Bariatric Surgery. All rights reserved. doi:10.1016/j.soard.2006.03.004


R.M. Dallal and L.A. Bailey / Surgery for Obesity and Related Diseases 2 (2006) 455– 459

completed a high-volume laparoscopic bariatric fellowship. All patients with symptoms suggestive of ulcer disease underwent endoscopy. The symptoms that commonly resulted in endoscopy included chronic nausea, dysphagia, and abdominal pain. Only patients with endoscopic evidence of ulcer disease and clinical symptoms were included in this study. Ulcer prophylaxis protocol

One Month of PPI BID Sucralfate suspn 1g QID

One Month of PPI QD Sucralfate suspn 1g QID

All patients were counseled specifically on the ulcerogenic properties of tobacco and were prescribed a proton pump inhibitor (typically Prevacid) to take once daily for 3 months after surgery. Patients were also specifically told not to take nonsteroidal anti-inflammatory drugs (NSAIDs) for longer than a few days without consulting the bariatric team. Those who required chronic NSAIDs received proton pump inhibitor prophylaxis. Routine preoperative endoscopy was not performed and the H. pylori was not studied preoperatively.

One Month of Sucralfate suspn 1g QID


One Month of Sucralfate suspn 1g TID

The procedure was an antecolic laparoscopic gastric bypass. The Roux limb was created by transecting the bowel 40 –50 cm from the ligament of Trietz. The mesentery was transected using two 45-mm staple fires. The Roux limb was 75–150 cm, depending on the body mass index. The omentum was vertically divided to also decrease the tension of the Roux limb as it traveled above the transverse colon. The gastric pouch was created to allow for the smallest safe size (10 –15 cm3). The pouch was transected approximately 1 cm below the fat pad of Belsey. The pouch was based on the lesser curve and vertically oriented such that the transection line came to the angle of His. The gastrojejunal anastomosis was performed using the linear-stapled approach after the pouch was mobilized from its posterior and diaphragmatic attachments. First, a seromuscular nonabsorbable 2-0 Surgidec (Tyco Healthcare, U.S. Surgical, Norwalk, CT) stitch was placed from the Roux limb to the posterior gastric pouch. Gastrostomy and enterotomy were then made using the ultrasonic scalpel (Ethicon Endosurgery, Cincinnati, OH). A linear stapler was then placed through these enterotomies, and a 1.2-cm anastomosis was created. A 32F orogastric tube was then advanced through the anastomosis. A 2-0 absorbable Polysorb suture (Tyco Healthcare) was used to close the common enterotomy tightly over the 32F tube. A seromuscular anterior layer of 2-0 Surgidec suture was used to complete the two-layered anastomosis. Medical management of ulcers The medical management of ulcer disease was implemented for all patients with ulcer as described in Fig. 1.

If symptoms return, go to previous month’s medication schedule.

One Month of Sucralfate suspn 1g BID

One Month Sucralfate suspn 1g qD

Off medication Fig. 1. Slow wean from sucralfate and proton pump inhibitor used to treat patients with marginal ulcers.

Statistical analysis Age, gender, co-morbidities, medications, and initial body mass index were examined for predictors of ulcer formation. Student’s t test was used to determine statistical significance. Postoperative complications were defined as those that occurred within the first 30 days after surgery.

R.M. Dallal and L.A. Bailey / Surgery for Obesity and Related Diseases 2 (2006) 455– 459


Table 1 Patient characteristics Pt. No.

Age at Surgery (y)

Preoperative Smoking

Postoperative NSAID Use

Initial BMI (kg/m2)

Postoperative Complications

Major Co-morbidities

1 2 3 4 5 6

48 39 35 59 37 32

No No No Yes No No

40.6 36.4 46.8 43.9 46.1 48.1

None None None None None None

Depression, GERD, USI Asthma, Hypertension, USI Diabetes, GERD, OSA, GERD, USI, arthritis, USI, OSA, GERD



No No No No No Yes and continued to smoke No





Pt. No. ⫽ patient number; NSAID ⫽ nonsteroidal anti-inflammatory drug; BMI ⫽ body mass index; GERD ⫽ gastroesphogeal reflux disease; USI ⫽ urinary stress incontinence; OSA ⫽ obstructive sleep apnea.

Results In the 24-month period, 201 gastric bypasses were performed. Seven patients (3.5%) developed marginal ulceration after laparoscopic gastric bypass surgery. The average length of follow-up was 19.6 months (range 6 –28). The proportion of patients in follow-up at 1 and 2 years was 95% and 84%, respectively. No mortalities occurred; one patient developed an anastomotic (gastrojejunal) leak but did not develop an ulcer postoperatively. The stricture rate for the entire population was 2.0%. In these patients, no ulcer was associated with the suture material at endoscopy. None of the patients had a concomitant stricture. The surgeon, if needed, performed all follow-up endoscopies. Three patients underwent an open procedure (two had had previous multi-abdominal trauma and one had undergone previous radical nephrectomy); none of these three developed marginal ulcers. The difference in the operative times between the first 50 cases (median 85 ⫾ 17.8 minutes, range 42–190) and the last 50 cases (median 75 ⫾ 14 minutes, range 54 –146) was not statistically significant (P ⫽ .22).

Table 1 documents the pertinent demographic features and potential risk factors for ulcer formation. All patients who used tobacco were strongly encouraged to abstain. The 1 patient who smoked who developed an ulcer had the most life-threatening disease. No significant differences in comorbidities, gender, or age between the patients who developed ulcer disease and those who did not. Table 2 documents the specific ulcer history of the patients. The 1 patient who had a perforation had continued epigastric pain after emergent laparoscopic repair. Although her ulcer was first diagnosed at emergent laparoscopy, follow-up endoscopy demonstrated a persistent large marginal ulcer. Only until she abstained from smoking did maintenance medical therapy heal her ulcer. She was unable to wean off the medication completely. All ulcers in these 7 patients were found on the mucosa of the Roux limb. Another patient became pregnant 6 months after gastric bypass surgery. Despite close follow-up and normal nutritional laboratory findings, she had a stillbirth at 35 weeks’ gestation. Two days afterward, she had gastrointestinal bleeding requiring blood transfusion from a marginal ulcer.

Table 2 Specific ulcer histories Pt. No.

Postoperative Interval to Ulcer Diagnosis (mo)


EGD Diagnosis


1 2 3 4 5

10 3 5 10 14

Pain Bleeding Pain Pain Bleeding

Yes Yes Yes Yes Yes









Resolved with medication Required transfusion; resolved with medication Resolved with Medication Resolved with medication Developed ulcer 2 days after D&E at 35 weeks’ gestation; required transfusion; resolved with medication Emergency repair; developed recurrent symptoms when weaning off Sucralfate; required twice-daily schedule Required transfusion; resolved with medication

Pt. No. ⫽ patient number; EGD ⫽ esophagogastroduodenoscopy; D&E ⫽ dilation and evacuation.


R.M. Dallal and L.A. Bailey / Surgery for Obesity and Related Diseases 2 (2006) 455– 459

Discussion One of the most vexing postoperative problems after gastric bypass surgery is the diagnosis and management of marginal ulceration. We have documented an ulcer incidence of 3.5% after an average follow-up of 19.6 months in patients undergoing laparoscopic gastric bypass using the linear-stapled technique. This was despite typical, although unproved, measures of ulcer prevention—smoking cessation and NSAID avoidance. Only 1 of the 7 patients with an ulcer had identifiable risks for developing ulcer disease. Marginal ulcers may present with severe pain. Unlike in classic gastric ulcers, food often makes the pain worse. The pain can be quite disabling and often does not seem to correlate with the size of the ulcer. Whether ulcers can give symptoms of nausea without pain is unclear. Anecdotally, in patients whose only symptom is nausea, frequently no pathologic finding is noted at endoscopy. However, in some, small ulcerations have be appreciated, usually associated with a foreign body (staple line or suture remnants). Whether these small ulcers are the cause of nausea or are an unrelated chance phenomenon is unclear. Often patients with marginal ulcers complain of such severe and constant pain that difficult revisions or even gastric bypass reversals have been performed—sometimes with recurrence of the ulcer [5,8]. Csendes et al. [10] have even advocated routine resection of the excluded gastric remnant to prevent the formation of ulcers stimulated by gastrin secretion. Significant bleeding can occur from an ulcer without any preceding symptoms of pain. Why some patients develop pain, sometimes from very small ulcers, and others do not is unclear. Although the motto “no acid, no ulcer” is valid in the nonbariatric population, this may not be true in bariatric patients. The cause of marginal ulceration may be more complex in patients who have undergone gastric bypass surgery. Ischemia, H. pylori, inflammation, foreign body reaction, and acid have all been implicated in the pathophysiology [5– 8]. Bariatric surgeons have a long-held belief that the gastric pouch produces only insignificant amounts of acid. Smith et al. [11] documented significantly decreased basal and pentagastrin-stimulated acid production in the gastric pouch. However, this may not be universal for all gastric bypass patients. Although gastric acid is mainly produced in the body and fundus of the stomach, Hedberg et al. [5] and Siilin et al. [12] have both documented an increase in acid production in patients with marginal ulcers despite the use of a small gastric pouch. This finding suggests that individual differences in parietal cell distribution may be the most important risk factor for the development of marginal ulceration. Most likely, because the jejunal mucosa does not have the intrinsic protection from an acid environment, only small amounts of acid production can cause ulceration. Sapala et al. [4,13] have advocated very

small pouches to decrease the number of parietal cells in the remaining pouch, as well as to prevent late weight regain. Experience likely contributes to the incidence of complications after bariatric surgery. Thus, longitudinal series that include their initial cases may skew the complication rates. As a proxy for the measurement of the “learning curve,” the operative times between the initial 50 patients and the last 50 patients were not different. The use of suture material also likely has a significant impact on the ulceration rate. Sacks et al. [14] has documented a significantly decreased incidence of marginal ulceration after switching from a nonabsorbable suture (Surgidec, 2.6%) to an absorbable one (Polysorb, 1.3%). Others have suggested that H. pylori may increase gastrointestinal symptoms after gastric bypass [7]. No group has definitively established a relationship between H. pylori and marginal ulcer formation. Schirmer et al. [8] documented that 30.6% of patients had H. pylori before surgery and an incidence of marginal ulceration of 2.4% compared with 6.8% in those who did not have preoperative treatment. Animal studies, however, have not found an increased incidence of ulceration in rats or mice with H. pylori using a gastric bypass model [15]. We did not check for the presence of H. pylori before gastric bypass or after a diagnosis of ulcer disease in these patients. No optimal therapy has yet been determined either for prophylaxis or treatment of ulcer disease in the bariatric patient. Our prolonged treatment regimen was created because of our experience with a high rate of recurrent ulceration after short treatment courses. The liquid suspension of sucralfate seems the most logical formulation, because a large pill would be unlikely to coat the gastric pouch. Anecdotally, patients seemed to respond better to sucralfate than to acid suppression. Revision surgery should be advocated for patients with documented gastrogastric fistulas; otherwise, the role of revision surgery is unclear. Although always difficult, we endoscopically removed all suture remnants during endoscopy when associated with an ulcer or in patients with complaints of dysphagia. Whether the suture was from the nonabsorbable Surgidec that had migrated into the lumen (or was sutured too deeply) or from yet undissolved absorbable suture is unknown. None of the patients in this cohort had an ulcer associated with suture or had an “incidental” ulcer, with the only symptom being nausea. The incidence of symptomatic ulcers likely varies depending on the type of gastric bypass (divided versus nondivided), the route (antecolic versus retrocolic), the type of anastomosis (hand sewn, linear stapled, circular stapled), the experience of the surgeon, and the suture material used (absorbable versus nonabsorbable) [16 –18]. Pouch configuration, tension on the Roux limb, ischemia to the tip of the Roux limb, and the fine details of the gastrojejunostomy construction may all affect marginal ulcer rates. All are difficult to study.

R.M. Dallal and L.A. Bailey / Surgery for Obesity and Related Diseases 2 (2006) 455– 459

Our study had limitations. The follow-up period in this study was not long enough to capture patients who might develop ulcer disease years after surgery. Although the follow-up was not 100%, it is probable that the patients who developed problems associated with their gastric bypass procedure would be more likely to return to our office. Furthermore, we did not perform repeat endoscopy to document cure of ulcer disease in patients who became asymptomatic. We were not able to define the etiology of the disease. We did not test for H. pylori either at baseline or after the diagnosis. We did not perform pH studies or measure for possible ischemia using tonometry. Furthermore, we believe that the results of this study cannot be extrapolated to other surgical techniques. Our ulcer rate was greater than that reported by Sacks et al. [14] (n ⫽ 2190), even though exactly the same techniques were used. The difference could have been a result of the smaller number of patients in this study, the longer length of follow-up (12 versus 19.6 months), or differences in patient follow-up in our practice setting.









Conclusion [11]

Patients who develop a marginal ulcer after gastric bypass surgery are often denied the quality-of-life improvements desired from massive weight loss. Long-term medical treatment with sucralfate suspension and acid inhibitors seems to heal ulcers in most patients. The current understanding of the pathophysiology of marginal ulcer disease suggests that the avoidance of tobacco and NSAIDs, as well as minimizing foreign materials, performing a divided gastric partitioning with Roux limb interposition, and the creation of a small pouch are all reasonable methods to decrease the risk of ulcer formation. Minimizing Roux limb tension and ensuring a well-perfused Roux limb may also contribute to decreased ulceration rates. Understanding the cause of ulcer disease through additional study may help to modify techniques to avoid this complication. References [1] MacLean LD, Rhode BM, Nohr C, Katz S, McLean AP. Marginal ulcer after gastric bypass. J Am Coll Surg 1997;185:1–7. [2] DeMaria EJ, Schweitzer MA, Kellum JM, Meador J, Wolfe L, Sugerman HJ. Hand-assisted laparoscopic gastric bypass does not im-









prove outcome and increases costs when compared to open gastric bypass for the surgical treatment of obesity. Surg Endosc 2002;16:1452–5. Leifsson BG, Gislason HG. Laparoscopic Roux-en-Y gastric bypass with 2-metre long biliopancreatic limb for morbid obesity: technique and experience with the first 150 patients. Obes Surg 2005;15:35– 42. Sapala JA, Wood MH, Sapala MA, Flake TM Jr. Marginal ulcer after gastric bypass: a prospective 3-year study of 173 patients. Obes Surg 1998;8:505–16. Hedberg J, Hedenstrom H, Nilsson S, Sundbom M, Gustavsson S. Role of gastric acid in marginal ulcer after gastric bypass. Obes Surg 2005;15:1375– 8. Pope GD, Goodney PP, Burchard KW, et al. Peptic ulcer/stricture after gastric bypass: a comparison of technique and acid suppression variables. Obes Surg 2002;12:30 –3. Ramaswamy A, Lin E, Ramshaw BJ, Smith CD. Early effects of Helicobacter pylori infection in patients undergoing bariatric surgery. Arch Surg 2004;139:1094 – 6. Schirmer B, Erenoglu C, Miller A. Flexible endoscopy in the management of patients undergoing Roux-en-Y gastric bypass. Obes Surg 2002;12:634 – 8. Martin MJ, Mullenix PS, Steele SR, See CS, Cuadrado DG, Carter PL. A case-match analysis of failed prior bariatric procedures converted to resectional gastric bypass. Am J Surg 2004;187:666 –71. Csendes A, Burdiles P, Papapietro K, et al. Results of gastric bypass plus resection of the distal excluded gastric segment in patients with morbid obesity. J Gastrointest Surg 2005;9:121–31. Smith CD, Herkes SB, Behrns KE, Fairbanks VF, Kelly KA, Sarr MG. Gastric acid secretion and vitamin B12 absorption after vertical banded Roux-en-Y gastric bypass for morbid obesity. Ann Surg 1993;218:91– 6. Siilin H, Wanders A, Gustavsson S, Sundbom M. The proximal gastric pouch invariably contains acid-producing parietal cells in Roux-en-Y gastric bypass. Obes Surg 2005;15:771–7. Sapala JA, Wood MH, Sapala MA, Schuhknecht MP, Flake TM Jr. The micropouch gastric bypass: technical considerations in primary and revisionary operations. Obes Surg 2001;11:3–17. Sacks BC, Mattar SG, Qureshi FG, et al. Incidence of marginal ulcers and the use of absorbable anastomotic sutures in laparoscopic Rouxen-Y gastric bypass. Surg Obes Relat Dis 2006;2:11– 6. Stenstrom B, Loseth K, Bevanger L, Sturegard E, Wadstrom T, Chen D. Gastric bypass surgery does not increase susceptibility to Helicobacter pylori infection in the stomach of rat or mouse. Inflammopharmacology 2005;13:229 –34. Lujan JA, Frutos MD, Hernandez Q, Cuenca JR, Valero G, Parrilla P. Experience with the circular stapler for the gastrojejunostomy in laparoscopic gastric bypass (350 cases). Obes Surg 2005;15:1096 – 102. Capella, JF, Capella RF. Gastro-gastric fistulas and marginal ulcers in gastric bypass procedures for weight reduction. Obes Surg 1999;9: 22– 8. Sanyal AJ, Sugerman HJ, Kellum JM, Engle KM, Wolfe L. Stomal complications of gastric bypass: incidence and outcome of therapy. Am J Gastroenterol 1992;87:1165–9.