Diagnostic Yield and Clinical Implications of Preoperative Upper Gastrointestinal Endoscopy in Morbidly Obese Patients Undergoing Bariatric Surgery

Introduction

Routine upper gastrointestinal (UGI) endoscopy in morbidly obese patients prior to bariatric surgery is controversial. As UGI endoscopy is an invasive procedure, some bariatric centers suggest that its performance in the preoperative evaluation of patients undergoing bariatric surgery may be based on the presence or absence of symptoms as in the general population.^1,2 However, many observational studies performing routine UGI endoscopy in preoperative bariatric surgery have shown that there is no correlation between symptoms and pathological findings on UGI endoscopy.^3–6 A variety of pathologies, including hiatal hernias, esophagitis, gastritis, duodenitis, peptic ulcers, etc., can be identified on UGI endoscopy, the majority of which occur in asymptomatic patients. Some of these endoscopic findings like severe mucosal inflammation or ulcers and polyposis may necessitate delay of surgery for need of treatment and histopathological confirmation with or without imaging. The chosen procedure might be changed if specific pathological UGI findings, such as a large hiatal hernia or Barrett's esophagus, are detected preoperatively.^3–6 Rarely, findings may be a contraindication to surgery in case of malignancy or varices. Also, as procedures like Roux-en-Y gastric bypass, duodenal switch, biliopancreatic diversion, etc., render the distal stomach and/or duodenum inaccessible postoperatively, ideally, the stomach and duodenum should be screened before surgery, as there have been reports of malignancies occurring in the bypassed stomach.^7,8

However, some authors have pointed out that most pathological findings discovered on routine UGI endoscopy will have a low impact on surgical management, especially in asymptomatic patients, and may lead to unnecessary work-up prompted by endoscopic findings adding on to cost. ²

With this study we aimed to evaluate the diagnostic yield and clinical implications of pathological findings on routine UGI endoscopy before bariatric surgery in a cohort of morbidly obese Indian patients.

Materials and Methods

Results

The patients included in this study included 164 females and 119 males, with a mean age of 42.30±12.49 years (range, 14–75 years) and mean body mass index of 43.80±8.5 kg/m². There were 209 sleeve gastrectomies and 74 Roux-en-Y bypasses performed.

Discussion

The rationale for performing an UGI endoscopy before bariatric surgery is to detect and treat UGI lesions that might cause symptoms or complications in the postoperative period or to detect lesions that may need a change in bariatric procedure performed. Our institutional policy is to perform routine preoperative UGI endoscopy in all patients prior to bariatric surgery. Some suggest symptom-directed UGI endoscopy as in the general population.^1,2 However, no correlation has been shown in the occurrence of symptoms and significant pathological findings on endoscopy in a number of observational studies on routine endoscopy before bariatric surgery. The majority of lesions occur in asymptomatic patients, as were the finding in our study also.^3–6 This high incidence of UGI lesions (81% in our series) that lacks correlation with symptoms has been the basis of most guidelines suggesting routine endoscopy and screening preoperatively in bariatric surgery patients.^9,10 However, some authors have pointed out that pathological findings discovered will have a low impact on surgical management, especially in asymptomatic patients, and may lead to secondary unnecessary work-up adding on to cost. ²

In this study patients with normal findings or milder mucosal inflammatory lesions (mild to moderate esophagitis, gastritis, or duodenitis) or mild anatomical abnormalities (lax lower esophageal sphincter, hiatal hernias of <5 cm) proceeded with surgery as previously planned. Patients with severe erosive gastritis or duodenitis were treated with proton pump inhibitors for 2 weeks, needing delay of surgery. There were no gastroduodenal ulcers detected in this series. If present, these require treatment with proton pump inhibitors for 4 weeks with reassessment by UGI endoscopy to confirm mucosal healing. We diagnosed 1 case of Barrett's esophagus in this series in whom surgical approach was changed to Roux-en-Y bypass with enrollment into a screening program as reported in other studies also. ² Laparoscopic Roux-en-Y gastric bypass is the most effective and advantageous treatment option for gastroesophageal reflux disease and hence Barrett's esophagus in the morbidly obese patients because it treats gastroesophageal reflux disease effectively and provides the additional benefit of weight loss. ¹¹

All patients with polyposis in this series were biopsied for histopathological confirmation of diagnosis. Eight were found to have benign fundic polyposis. These polyps were excised along with the specimen if a sleeve gastrectomy was performed and left in situ in bypass patients. Gastric fundic gland polyps that are nonfamilial are associated with a very low or no risk of malignancy and hence are not usually resected. ¹² Azagury et al. ² performed subtotal gastrectomy if a Roux-en-Y bypass was planned as the fundic polyposis lesion would be without surveillance in a bypassed stomach, but there are no data to support this radical strategy. It may be advisable in polyps with greater risk of malignant degeneration (e.g., adenomatous, hamartomatous polyps, etc.). One patient having two duodenal polyps was found to have a carcinoid tumor on biopsy. The tumor was excised completely endoscopically, and the patient has been kept on a surveillance program after sleeve gastrectomy. Although rare, carcinoid tumors appear to be more frequent among obese patients. Keshishian et al. ¹³ and Mottin et al. ¹⁴ noticed a high incidence of carcinoid in obese patients (1.5%) compared with the general population, a finding underling the importance of screening endoscopy. Obesity has been implicated in the pathogenesis of these tumors. Some series have reported detection of gastrointestinal stromal tumors, which may be excised at the time of the bariatric procedure with adequate margins. ¹⁵ Malignancy may be uncommon but if suspected needs additional evaluation. ¹⁶

Performance of a sleeve gastrectomy in the presence of a hiatal hernia and/or gastroesophageal reflux is controversial. Some studies have reported exacerbation of gastroesophageal reflux disease in some or de novo reflux in others, whereas some have reported not much influence on gastroesophageal reflux disease. Some authors have suggested that Roux-en-Y gastric bypass is the best bariatric surgical procedure for obese patients with hiatal hernia. Others have suggested that patients can also be safely offered sleeve gastrectomy if combined with a simultaneous hiatus hernia repair. ¹⁷ Large hiatal hernias found at endoscopy may thus need the addition of hiatal hernia repair at sleeve gastrectomy or change of the surgical approach to a Roux-en-Y bypass,^17,18 with the latter done in 2 patients in this series.

All patients positive for H. pylori were treated with eradication and therapy with proton pump inhibitors. The prevalence of H. pylori in the Indian subcontinent can be as high as 80%, especially in rural areas. ¹⁹ The pathogen most commonly manifests as peptic ulcer disease. It may contribute to marginal ulcers after surgery, but its role is still unclear. ²⁰ Studies not routinely eradicating H. pylori infections prior to surgery have reported an increased incidence and association of this infection with early postoperative foregut symptoms. ²¹ Also, several epidemiological studies in India have shown a high incidence of gastric cancer in South India compared with North India with a definite association between H. pylori and gastric cancer in approximately 50% of patients.^22,23 In the rest it may still play a role by acting as a cocarcinogen. ²⁴ Atrophic pangastritis in these infected patients is the main precursor lesion associated with gastric cancer. ²⁵ This increased risk of gastric cancer may be of further concern in the stomach remnant of the gastric bypass as this would not be accessible for screening. We routinely eradicate H. pylori infection in all patients prior to bariatric surgery because of concerns of peptic ulceration and gastric cancer as seen in the general Indian population after chronic infection, although supporting data are lacking in the literature.

In conclusion, preoperative routine UGI endoscopy yielded a high proportion of endoscopic abnormalities. In a few of these cases, findings led to a change in surgical approach. In the majority, however, these findings did not affect the actual operative management, although medical management was required, without which these lesions could have potentially led to postoperative symptoms and complications. In a few cases, postoperative surveillance for malignancy was recommended. Preoperative UGI endoscopy in all patients before bariatric surgery should be performed regardless of the presence or absence of symptoms, and pathological findings have an impact on the course of management.