Surg Endosc (2015) 29:2885–2890 DOI 10.1007/s00464-014-4022-8
and Other Interventional Techniques
The presence of Helicobacter pylori is not associated with longterm anastomotic complications in gastric bypass patients John J. Kelly • Richard A. Perugini • Qi L. Wang • Donald R. Czerniach • Julie Flahive • Philip A. Cohen
Received: 1 April 2014 / Accepted: 28 November 2014 / Published online: 1 January 2015 Ó Springer Science+Business Media New York 2014
Abstract Background Eradication of Helicobacter pylori prior to Roux-en-Y gastric bypass (RYGB) has been advocated as a measure to reduce the complications of anastomotic ulceration. However, evidence to support a causal relationship between preoperative H. pylori status and postoperative anastomotic ulceration is weak. Methods Intraoperative gastric biopsies were obtained on consecutive patients who underwent laparoscopic RYGB at our institution from December 2007 to June 2010. These samples were analyzed by Warthin–Starry stain for H. Pylori organisms. Retrospective chart review was conducted to determine the preoperative presence of acid dyspepsia and acid suppression therapy and to determine postoperative ulcer symptoms, smoking, NSAID or steroid use, and compliance with ulcer prophylaxis. The incidence of ulcer visualization, perforation, and stricture were obtained from a prospectively collected database. Fisher’s exact test was used for analyzing associations between
Presented at the 22nd EAES Congress, June 25–28, 2014, Paris, France. J. J. Kelly (&) R. A. Perugini D. R. Czerniach P. A. Cohen Division of General, Minimally Invasive, and Bariatric Surgery, Department of Surgery, University of Massachusetts Medical Center, 55 Lake Ave N., H1-760, Worcester, MA 01655, USA e-mail:
[email protected] Q. L. Wang Department of Surgery, University of Massachusetts Medical Center, Worcester, MA, USA e-mail:
[email protected] J. Flahive Center for Outcomes Research, University of Massachusetts Medical School, Worcester, MA, USA
discrete groups. Multiple logistic regression was used to assess associations between anastomotic ulcer complications and potential predictors. Results Histologic evaluation for H. pylori was available in 708 of the 728 patients who underwent RYGB. Fourteen patients were lost to follow up leaving 694 patients available for review. H. pylori was positive in 66 (9.5 %) patients who did not go on to receive definitive treatment for eradication. Marginal ulcers or related late complications were seen in a total of 113 (16.3 %) patients. In the H. pylori positive group, five patients (7.6 %) developed ulcer complications compared to 108 (17.1 %) in the H. pylori negative group (p = 0.05). Groups were not different in terms of preoperative demographics, postoperative ulcer prophylaxis compliance, steroid, NSAIDs, and cigarette use. Conclusion The presence of H. pylori infection at the time of RYGB was found to be associated with a significantly lower incidence of anastomotic ulcer complications postoperatively. This study brings into question efforts and expense allocated to identify and eradicate H. pylori prior to RYGB. Keywords
Bariatric Obesity Complications Ulcer
Morbid obesity remains a serious health problem and is responsible for a significant portion of health care expenditures in the United States and elsewhere. Bariatric surgery, including the Roux-en-Y gastric bypass (RYGB), has proven to be a safe and effective strategy for the treatment of morbid obesity. Over the last 20 years, the field of bariatric surgery has achieved a perioperative safety profile similar to that of more minor surgical procedures, particularly concerning key outcome indices such as mortality, bleeding, anastomotic leak, and venous thromboembolic events. However, despite improved safety, diligent
123
2886
bariatric programs, and compliant patients, certain late complications remain difficult to prevent and will predictably occur following RYGB. One of the most common late complications is the development of marginal ulceration and associated complications including pain, stricture, bleeding, and perforation. The incidence reported varies widely between 1 and 16 % [1, 2], and is believed to be underreported. The exact pathogenesis of marginal ulcers is not completely clear, but is thought to involve the direct toxic effect of gastric acid on jejunal mucosa, which is prone to chemical injury due to limited barrier protection. In addition, various risk factors for anastomotic ulcers have been identified, such as smoking, non-steroidal anti-inflammatory drug (NSAID) use, steroid use, foreign body presence, surgical technique, gastric pouch size, time from surgery, and co-morbid illness [3, 4]. However, the perceived significance and relative contribution of each risk factor varies by study. Some authors have implicated Helicobacter pylori as a risk factor for anastomotic ulceration, although studies looking directly at the association between H. pylori and subsequent anastomotic ulcer yield conflicting results [5–10]. Of concern, many insurance carriers have formulated best practice management plans that demand preoperative diagnosis and eradication of H. pylori prior to bariatric surgery approval without strong supportive evidence.
Materials and methods Between December 2007 and June 2010, intraoperative gastric full thickness biopsies were obtained on consecutive patients who underwent laparoscopic RYGB at the University of Massachusetts/Memorial Medical Center. These samples were sent for histological evaluation for the presence of H. pylori organisms using Warthin–Starry stain. Institutional Review Board approval was obtained prior to initiation of the study. Four experienced bariatric surgeons all using the same technique performed 728 laparoscopic RYGB during the study period. The RYGB procedure included a 120-cm Roux limb passed in an antecolic and antegastric fashion. The gastro-jejunostomy was formed with 25-mm circular stapler with the anvil passed by a transabdominal, transgastric approach. Histologic specimens were obtained from residual tissue created during closure of the gastrotomy with a linear stapler (created at the distal body of the stomach for the transgastric anvil passage). The gastric pouch volume was 30 mL by visual estimation. Mesenteric and retro-roux defects were closed with permanent suture. Patients were treated postoperatively with prophylactic H2blocker therapy once a day for at least 1 year. All patients
123
Surg Endosc (2015) 29:2885–2890
received routine postoperative follow-up by the same weight center team at the following intervals: 2, 6 weeks, 6, 12, 18, 24 months then yearly. For each patient, retrospective chart review was conducted to determine demographics including age, gender, preoperative body mass index (BMI), preoperative symptoms of acid dyspepsia, use of proton pump inhibitor (PPI) or H2-blocker therapy, diagnosis of gastroesophageal reflux disease (GERD), and history of peptic ulcer disease (PUD). Postoperative data collected included patient reported ulcer symptoms or ulcer visualized on EGD, perforation, stricture, or episodes of gastrointestinal bleeding. Other factors examined included smoking, NSAID or steroid use, and compliance with ulcer prophylaxis (PPI and H2-blocker use as documented in the electronic medical record). The presence of marginal ulcer was defined by the following: (1) Simple presumed ulcer identified as having symptoms (epigastric pain with a relationship to meals) that improved with only conservative management (prescription acid suppression and/or sucralfate use); (2) Simple visualized ulcer identified as having endoscopic documentation of ulcer; (3) Stricture with ulcer identified if stenosis was seen on upper endoscopy and dilated more than 3 months postoperatively; and (4) Perforation identified if operative intervention was required for free air or contrast leak seen on imaging study. If patients had more than one complication, the most severe complication was documented (in order of increasing severity: simple presumed ulcer, simple visualized ulcer, stricture, perforation), so that each patient was only counted once. Fisher’s exact test was used for analyzing associations between discrete groups. Multiple logistic regression was used to assess associations between anastomotic ulcer complications and potential predictors.
Results A total of 728 patients underwent laparoscopic RYGB during the study period from December 2007 to June 2010. Histologic evaluation for H. Pylori organisms was possible in 708 specimens, and 20 specimens were inconclusive using Warthin–Starry stain due to inadequate tissue sampling. Of these 708 patients, 14 patients did not have reviewable postoperative data and were excluded from the analysis. A total sample of 694 patients with adequate follow-up and definitive histological evaluation were included in the analysis. Histological analysis of tissue specimens was positive for H. pylori in 66 (9.5 %) patients at the time of RYGB. No patients in this group received definitive eradication therapy following surgery.
Surg Endosc (2015) 29:2885–2890
2887
Table 1 Demographics for total study population, H. pylori negative and H. pylori positive groups
Table 2 Postoperative findings for study population, H. pylori negative and H. pylori positive groups
Total population (n = 694)
(-) H. pylori (n = 628)
(?) H. pylori (n = 66)
p value NSAID use (%)
52 (7.5)
50 (8)
2 (3)
0.22
Age (mean)
46
46
45
0.70
Smoking (%)
30 (4.5)
27 (4.3)
3 (4.6)
0.76
BMI
47
47
47
0.84
Sex
76 % female
76 % female
73 % female
0.58
24 (3.5) 511 (74)
21 (3.3) 463 (74)
3 (4.6) 48 (73)
0.49 0.86
Pre-op GERD
195 (28 %)
183 (29 %)
12 (18 %)
0.06
Steroid use (%) Ulcer prophylaxis (%)
Pre-op PUD
6 (0.9 %)
5 (0.8 %)
1 (1.5 %)
0.45
Pre-op PPI
150 (22 %)
143 (23 %)
7 (11 %)
0.02
Pre-op H2B
29 (4.2 %)
28 (4.5 %)
1 (1.5 %)
0.51
Follow-up time (median)
729 days
729 days
708 days
0.75
Patients [ 1-year follow-up
583 (78 %)
487 (78 %)
51 (78 %)
0.96
Patients had a mean age of 46 years with an average preoperative BMI of 47. The patients were predominantly female (n = 524, 76 %). Twenty eight percent of patients carried a preoperative diagnosis of GERD, 22 % were on PPI prior to surgery, and 4.2 % were on H2-blockers [Table 1]. The median postoperative follow-up time was 729 days with an interquartile range of 386–1,149 days. The H. pylori positive group was not different in terms of preoperative age, gender, BMI, diagnosis of GERD or PUD, but was less likely to be utilizing PPIs prior to surgery [Table 1]. There was no difference in the percentage of patients with follow-up time greater than 1 year between the two groups. Following surgery, there was no identifiable difference in postoperative ulcer prophylaxis compliance, use of steroids, use of NSAIDS, or cigarette smoking between the H. pylori positive and the H. pylori negative groups [Table 2]. A total of 123 (17.7 %) patients underwent endoscopic evaluation postoperatively for a variety of symptoms, primarily epigastric pain and/or vomiting. Counting each patient once based on the most severe complication, the overall incidence of ulcer or ulcer complication was seen in a total of 113 (16.3 %) patients. The diagnosis of simple ulcer was made in 56 (8.1 %) patients either on the basis of symptoms (n = 20, 2.9 %) or visualized on endoscopy (n = 36, 5.2 %). Stricture with ulcer was seen on upper endoscopy in 31 patients (4.5 %). Perforations presented emergently in 26 patients (3.7 %), all of whom were treated surgically. There was no statistical difference between the H. pylori biopsy positive group (n = 66) versus the group with negative tissue biopsy (n = 628) for the development of simple ulcers (3.0 vs. 8.6 %, p = 0.15), stricture (3.0 vs. 4.6 %, p = 0.76), or perforation (1.5 vs. 4.0 %, p = 0.5)
Total (n = 694)
Table 3 Postoperative H. pylori status
(-) H. pylori (n = 628)
anastomotic
(?) H. pylori (n = 66)
complications
(-) H. pylori (n = 628)
p value
grouped
(?) H. pylori (n = 66)
p value
Presumed simple ulcer (%)
19 (3)
1 (1.5)
0.71
Visualized simple ulcer (%)
35 (5.6)
1 (1.5)
0.24
Stricture (%) Ulcer perforation (%)
29 (4.6) 25 (4.0)
2 (3.0) 1 (1.5)
0.76 0.50
108 (17.1)
5 (7.6)
0.05
All anastomotic ulcerrelated complications combined
by
[Table 3]. However, when the incidence of all ulcer-related complications were combined, the H. pylori positive group had a significantly lower rate of complications compared to the H. pylori negative group (7.6 vs. 17 %, p = 0.05). Postoperatively, there was no significant difference between H. pylori positive and negative groups for symptoms of abdominal pain, epigastric pain, nausea and/or vomiting, pain with eating, or symptoms of GERD [Table 4]. Regarding behaviors generally considered risk factors for ulcer, we found no significant difference between groups for the incidence of simple ulcer or stricture and the subsequent use of NSAIDs, steroids, smoking, or compliance with ulcer prophylaxis [Table 2]. Postoperative BMI at 1 year was lower in the H. pylori negative group (31 vs. 33 kg/m2, p = 0.04). The mean and median times from surgery to ulcerrelated complications were 596 and 362 days, respectively, which did not vary significantly depending on type of complication. The mean and median follow-up periods in this study were 802 and 729 days, respectively, and there was no statistically significant difference in follow-up between H. pylori negative or positive groups.
Discussion The prevalence rate of H. pylori varies depending on demographic location, socioeconomic status, ethnicity, and
123
2888
Surg Endosc (2015) 29:2885–2890
Table 4 Comparison of postoperative symptoms between H. pylori negative and H. pylori positive groups (-) H. pylori (n = 628)
(?) H. pylori (n = 66)
p value
Abdominal pain (%)
496 (79)
83 (55)
0.42
Epigastric pain (%)
66 (11)
6 (9.1)
0.72
Nausea/vomiting (%) Pain with eating (%)
62 (10) 25 (4)
7 (10.6) 3 (4.6)
0.85 0.72
GERD symptoms (%)
39 (6.2)
2 (3.0)
0.42
method of detection. In developing countries, most children are infected before 10 years of age. However, in the United States, prevalence increases with age and is more common in certain ethnicities, but seems to be unrelated to BMI [11, 12]. It is estimated that up to 30–40 % of the United States population may be infected with H. pylori [11, 13]. Symptomatic patients infected with H. pylori can present with acid dyspepsia, chronic gastritis, and peptic ulcers. Treatment of these conditions is advocated to prevent the more serious sequela of ulcer complications including bleeding and perforation, or the insidious development of cancer (adenocarcinoma, lymphoma) [14, 15]. Obligatory testing and eradication of H. pylori infection in asymptomatic patients, however, remains controversial. While the literature supports a well-established causal relationship between H. pylori infection and PUD, almost all western literature and established practice guidelines advise against general screening and subsequent treatment in the asymptomatic patient [14]. Indeed, as evidence of unintended consequences of treatment for asymptomatic conditions, there is an evolving line of research suggesting a direct association between the eradication of H. pylori and an increase in the rate of reflux esophagitis [16]. The most rational explanation for this finding is that H. pylori can induce atrophic gastritis, limiting acid production from the parietal cells and subsequent esophageal acid exposure. It is estimated that 15–30 % of patients evaluated for weight loss surgery will test positive for H. pylori [17]. Previous studies have demonstrated that this approaches the rates seen in the general population after accounting for patient demographics [7, 11]. Several authors have demonstrated an association between higher complication rates after bariatric surgery and H. pylori positivity noted preoperatively [4–6]. Schirmer reported on a case series of 560 patients undergoing RYGB during a 16-year period [5]. Marginal ulcer was identified less often in those preoperatively screened and treated for H. pylori versus those from an earlier cohort who were not screened. Importantly, this study was retrospective, impacted by a mid-study change in the standard
123
care for all patients; this raises the possibility that surgeon experience and transitions in management had some effect on the results. Given these limitations, it is difficult to draw conclusions regarding the true effect of H. pylori status on outcome. Rasmussen [4] concluded that the postoperative marginal ulcer rate was greater in the group of patients found to be positive on preoperative H. pylori testing versus those testing negative (32 vs. 12 %, p = 0.02). All patients who tested positive received standard eradication therapy prior to surgery. Limitations of this study included a change in operative technique mid-study, a relatively small patient population (n = 260) and short follow-up (10 months). Another finding, which may implicate additional contributing patient factors, was the high rate of H. pylori negativity noted at the time of marginal ulcer diagnosis (68 %). These studies and others [6, 23] appear to suggest that despite fairly aggressive preoperative eradication efforts, the risk of marginal ulcer persists. Furthermore, it is not clear what identification and eradication protocols can accomplish, given the natural history of H. pylori infections in individuals undergoing RYGB. Csendes [20] followed 130 patients with routine endoscopies and assessment for H. pylori by pouch biopsy for up to 9 years following RYGB. In this population, 26.7 % of individuals tested positive for H. pylori in the preoperative period but not in the postoperative period, though no treatment for H. pylori had been administered. Another group of patients (9.5 %) who tested negative for H. pylori preoperatively, showed positive histology in the postoperative period. Thus, even if H. pylori is associated with attendant pathology, preoperative eradication does not guarantee that the organism will not be present postoperatively. Given the limitations of these studies along with the lack of evidence of active H. pylori infection at the time of surgery, the results must be considered carefully and do not provide definitive evidence of a link between H. pylori status and postoperative anastomotic complications after RYGB. Similar to our results, many papers dispute any association between H. pylori and marginal ulcer complications after gastric bypass [7–10]. In the largest series of 636 patients, Yang discovered (similar to our data) that rates of preoperative H. pylori positivity were lower in the group with endoscopically proven ulcers, but this failed to reach significance [6 (27 %) of 22 vs. 26 (43 %) of 60, p = 0.21]. Rawlins [10] evaluated 228 consecutive patients for the preoperative presence of H. pylori through serum antigen testing. Patients testing positive were eradicated with standard triple therapy (n = 68). A group of patients retested positive and received a second round of therapy (n = 24). Postoperatively, no significant difference was found in the rate of complications (p = 0.11) between H. pylori negative patients (11 of 160, 6.9 %) and those who
Surg Endosc (2015) 29:2885–2890
tested positive preoperatively (1 of 68, 1.5 %). In addition, the group of patients requiring repeat therapy for eradication developed no postoperative complications. Papasavas [8] retrospectively reviewed 259 patients who were tested preoperatively for H. pylori prior to RYGB. He found there was no difference in the rate of positive findings on postoperative endoscopy between patients tested preoperatively for H. pylori (5 %) and patients not tested (3.7 %). The study concluded that H. pylori testing did not lower the risk of anastomotic ulcer or pouch gastritis. The present study represents the largest series in the literature examining tissue biopsy proven intraoperative H. pylori colonization of the stomach and its relationship to postoperative anastomotic ulcer complications. The data collected have confirmed our suspicion that intraoperative H. pylori status does not correlate with the development of ulcer-related complications after RYGB. In fact, when all ulcer complications are taken together, tissue biopsy proven H. pylori colonization of the gastric mucosa at the time of RYGB seems to protect against the development of postoperative ulcer complications. This seemingly paradoxical finding should not be overlooked or discounted. There is scant evidence to support H. pylori infection as the causative agent in the development of anastomotic ulcers in patients undergoing laparoscopic RYGB. The vast majority of marginal ulcer complications, whether discovered endoscopically or laparoscopically, are located on the jejunal side of the anastomosis. H. pylori organisms are known to preferentially congregate along mucosa of the stomach (primarily antrum and body). The presence of H. pylori in jejunal cells is extremely rare and can only be found in one case report [18]. Furthermore, the precipitating cause of most anastomotic ulcers is as yet undetermined. More than likely the etiology is multifactorial with chronic inflammation (foreign body), microvascular changes (ischemia), decreased cytoprotective barriers (NSAID, steroids), or direct injury (ingested pathogen or caustic agent, trauma) as potential factors. If one of the primary causes is considered to be peptic-induced jejunal mucosal injury, it is possible that H. pylori promoted atrophic gastritis within the pouch may actually protect against post-anastomotic ulceration. Observational studies have suggested that H. pylori may protect against GERD-related esophageal injury through a similar mechanism. Curiously, despite several papers suggesting no impact of H. pylori status on outcomes after bariatric surgery, the eradication of H. pylori prior to RYGB has been advocated by some as a measure to reduce anastomotic ulceration and minimize complications. H. pylori screening and treatment is currently recommended by InterQual as a preoperative quality measure. Many insurance carriers subscribe to
2889
InterQual standards and demand screening and eradication prior to authorization of bariatric surgery, including sleeve gastrectomy. The literature used in the InterQual review, however, is outdated and not specific to the development of anastomotic ulceration or its attendant complications. The review is based on the consensus that H. pylori is the predominant cause of generalized PUD and is associated with gastric cancer in the general population [17]. Given our findings and the conclusions of others cited previously, it may be time to re-evaluate these recommendations and to update the standards to reflect current evidence. Unnecessary H. pylori workup and treatment are of course lengthy, complex, and costly. Eradication efforts can lead to resistance if compliance is low or if an inadequate course is prescribed [19]. As a result of anti-microbial pressure and non-compliance, the success rate of first time treatment with PPI, Clarithromycin, and Amoxicillin has decreased to 70–85 %, mostly due to resistance [14]. Bismuth-containing quadruple regimens for persistent infections are not well tolerated and have not been validated in the United States [14]. For patients that have been taking PPI, antibiotics, or bismuth, endoscopic biopsies from gastric body or antrum with or without rapid urease testing may be inaccurate [14]. Inconsistencies seen in the accuracy of diagnosing H. pylori colonization, the efficacy of H. pylori eradication, and our finding that H. pylori positivity does not seem to predispose patients to ulcer complications suggest that valuable resources expended to identify and eradicate H. pylori prior to RYGB may be misguided. Histology has previously been considered an imperfect gold standard for diagnosing H. pylori infection given its reliance on multiple variables including site, number and size of gastric biopsies, method of staining, and the level of experience of the pathologist [21]. Sensitivity of histologic evaluation may also be impacted by medications such as bismuth, antibiotics, and PPIs [22]. This raises a theoretical concern in our body of data, as the preoperative rate of PPI usage was higher in the H. pylori negative group than in the H. pylori positive group. This discrepancy could potentially have led to incorrectly identifying some H. pylori positive patients as H. pylori negative. The degree to which this error, if present, impacted our results is unknown but is likely present in other studies where histology is used as the primary diagnostic method. Additional potential weaknesses of this study include reliance on a retrospective chart review for a significant portion of the data. Also, due to a reliance on documentation review and identification/reporting of symptoms, the possibility exists for underestimating the incidence of anastomotic complications in both groups. However, the length of follow-up in our study is quite reasonable and should have provided ample opportunity for documenting
123
2890
and addressing patient complaints. The percentage of patients followed for more than 1 year was 78.5 %, and there was no significant difference in follow-up between the H. pylori negative and H. pylori positive groups. Lastly, arriving at a diagnosis of ulcer in this population, either by symptoms and observed treatment response or direct endoscopic visualization, seems to be an efficient and cost-effective model. Selective endoscopy, initiated and performed by the treating bariatric surgeon, allows for equipoise between a hair trigger interventional approach and a strict symptom-based clinical approach to postoperative concerns. This approach is more reflective of real world practice and may address the under reporting concerns of the endoscopy only model and the over reporting of the symptom-based approach [10].
Conclusions The preoperative presence of H. pylori was not found to correlate with the incidence of postoperative anastomotic ulcer or its complications. Preoperative colonization by H. pylori may be protective against anastomotic ulcer complications after RYGB. This study brings into question efforts and expense allocated to identify and eradicate H. pylori prior to RYGB. Acknowledgments Dr. Fred Anderson and the Center for Outcomes Research for statistical analysis. Disclosures Richard A. Perugini, Qi L. Wang, Donald R. Czerniach, Julie Flahive, Philip A. Cohen, and John J. Kelly have no conflict of interest or financial ties to disclose.
References 1. Garrido AB Jr., Rossi M, Lim SE Jr. et al (2010) Early marginal ulcer following Roux-en-Y gastric bypass under proton pump inhibitor treatment: prospective multicentric study. Arg Gastroenterol. 47(2):130–134 2. Buschwald H, Cowan G, Pories W (2007) Surgical management of obesity. WB Saunders, Philadelphia, pp 297–303 3. Bhayani NH, Oyetunji TA, Chang DC et al (2012) Predictors of marginal ulcers after laparoscopic Roux-en-Y gastric bypass. J Surg Res 177(2):224–227 4. Rasmussen JJ, Fuller W, Ali MR (2007) Marginal ulceration after laparoscopic gastric bypass: an analysis of predisposing factors in 260 patients. Surg Endosc 21(7):1090–1094 5. Schirmer B, Erenoglu C, Miller A (2002) Flexible endoscopy in the management of patients undergoing Roux-en-Y gastric bypass. Obes Surg 12:634–638
123
Surg Endosc (2015) 29:2885–2890 6. Cleator IG, Rae A, Birmingham CL (1996) Ulcerogenesis following gastric procedure for obesity. Obes Surg 6(3):260–261 7. Papasavas PK, Gagne DJ, Donnelly PE, Salgado J et al (2008) Prevalence of Helicobacter pylori infection and value of preoperative testing and treatment in patients undergoing laparoscopic Roux-en-Y gastric bypass. Surg Obes Relat Dis. 4(3):383–388 8. Yang CS, Lee WJ, Wang HH, Huang SP (2006) The Influence of Helicobacter pylori infection on the development of gastric ulcer in symptomatic patients after bariatric surgery. Obes Surg 16(6):735–739 9. Hartin CW Jr., ReMine DS, Lucktong TA (2009) Preoperative bariatric screening and treatment of Helicobacter pylori. Surg Endosc 23(11):2531–2534 10. Rawlins L, Rawlins MP, Brown CC, Schumacher DL (2013) Effect of Helicobacter pylori on marginal ulcer and stomal stenosis after Roux-en-Y gastric bypass. Surg Obes Relat Dis. 9(5):760–764 11. Renshaw AA, Rabaza JR, Gonzalez AM, Verdeja JC (2001) Helicobacter pylori infection in patients undergoing gastric bypass for morbid obesity. Obes Surg 11(3):281–283 12. Verma S, Sharma D, Kanwar P et al (2013) Prevalence of Helicobacter pylori infection in bariatric patients: a histologic assessment. Surg Obes Relat Dis. 9(5):679–685 13. Peterson WL, Fendrick AM, Cave DR et al (2000) Helicobacter pylori-related disease: guidelines for testing and treatment. Arch Intern Med 160:1285–1291 14. Chey WD, Wong BCY (2007) Practice parameters committee of the American College of Gastroenterology. American College of Gastroenterology guideline on the management of Helicobacter pylori infection. Am J Gastroenterol 102:1808–1825 15. Malfertheiner P, Megraud F, O’Morain C et al (2007) Current concepts in the management of Helicobacter pylori infection: the Maastricht III concensus report. Gut 56(6):772–781 16. Labenz J, Blum AL, Bayerdorffer E et al (1997) Curing Helicobacter pylori infection in patients with duodenal ulcer may provoke reflux esophagitis. Gastroenterology 112:1442–1447 17. McKesson Corporation. Interqual Smart Sheets. 2006–Procedure Criteria Bariatric Surgery. San Francisco, CA, McKesson Corporation 2006 18. Kim EunYoung et al (2012) A case of small bowel ulcer associated with Heliocobacter pylori. Pediatric Gastroenterol Hepatol Nutr. 159(4):266–271 19. Cerqueira RM, Correia MR, Fernandes DC, Vilar H et al (2013) Cumulative Helicobacter pylori eradication therapy in obese patients undergoing gastric bypass surgery. Obes Surg 23(2): 145–149 20. Csendes A, Smok G, Burgos AM, Canobra M (2012) Prospective sequential endoscopic and histologic studies of the gastric pouch in 130 morbidly obese patients submitted to Roux-en-Y gastric bypass. Arq Bras Cir Dig. 25(4):245–249 21. el-Zimaity HM (2000) Accurate diagnosis of Helicobacter pylori with biopsy. Gastroenterol Clin N Am 29:863–869 22. Woo JS, el-Zimaity HM, Genta RM et al (1996) The best gastric site for obtaining a positive rapid urease test. Helicobacter 1:256–259 23. Vanek VW, Catania M, Triveri K, Woodruff RW Jr. (2006) Retrospective review of the preoperative biliary and gastrointestinal evaluation for gastric bypass surgery. Surg Obes Relat Dis 2:17–23
