Bariatric surgery is an effective treatment for obesity and its metabolic related disorders. With the numbers of bariatric procedures being performed rising annually endoscopists are now more likely than ever to encounter this patient cohort on routine diagnostic lists. Endoscopy plays a vital role in both the pre and postoperative periods to diagnose and treat certain complications associated with bariatric surgery. Preoperative endoscopy may be useful in identifying pathologies that then influence the type of bariatric surgery being proposed. Postoperative endoscopy has an established role in identifying and managing complications such as anastomotic leaks and gastrointestinal bleeding. As endoscopic techniques develop, these complications may be managed without the need for surgical intervention or revision. Increased knowledge and awareness of the indications for endoscopy, together with the altered anatomy and common complications of bariatric surgery, is paramount in managing these patients effectively.
- diagnostic and therapeutic endoscopy
- obesity surgery
- surgical complications
Statistics from Altmetric.com
Obesity is defined as a body mass index >30 kg/m2 and is a rising epidemic with WHO estimating that over 700 million people worldwide are obese.1 UK obesity rates have doubled from 1993 to 2011 and the UK now ranks eighth for overweight prevalence for men and women combined.2 Obesity is linked to a multitude of illnesses including type 2 diabetes mellitus (T2DM), fatty liver, heart disease, gallstones and gastro-oesphageal reflux disease so presents a major economic burden. The Department of Health estimates that obesity could cost society and the economy £50 billion by 2050 if rates continue to increase at the same rate.3 Unfortunately little progress has been made with non-surgical interventions such as behavioural modifications and pharmacotherapy in the treatment of obesity. Contrastingly, the field of bariatric surgery has emerged rapidly as an effective treatment strategy not only in the management of obesity but also its related metabolic disorders such as T2DM.4 There is now long-term data supporting the efficacy of weight loss surgery in reducing mortality and morbidity as well as leading to improvements and remission of T2DM in up to 80% of patients.5
The steep rise in morbid obesity coupled with advances in bariatric surgery has led to a dramatic increase in the number of bariatric operations being performed. The UK National Bariatric Surgical Registry 2014 reported 16 956 procedures from 2011 to 2013 compared with 7045 operations between 2009 and 2010.6
Endoscopists are now more likely than ever to encounter patients who are undergoing, or have undergone, bariatric surgery on routine diagnostic lists so it is vital to have an understanding of the anatomical alterations that occur following surgery in order to conduct a thorough endoscopic assessment. This review article addresses the gastroenterology curriculum 2010 (table 1) with particular focus on bariatric surgery and provides a succinct summary of the indications for preoperative endoscopic assessment but also how to recognise the complications associated with bariatric surgery endoscopically in order to diagnose and manage these patients effectively.
The rationale for preoperative endoscopy is to identify pathologies that may alter management, such as hiatus hernia, oesophagitis, peptic ulcer disease, premalignant and malignant conditions. Obesity is associated with an increased prevalence of a number of these conditions which if identified on oesophagogastroduodenoscopy (OGD) may require medical treatment, delay to surgery or alteration of the surgical approach.5 Preoperative OGD prior to sleeve gastrectomy (SG) may detect reflux with oesophagitis and Barrett’s. These findings are a relative contraindication to SG as it may exacerbate reflux symptoms, and may prompt a switch to gastric bypass surgery.7 Hiatus hernia is another common finding in preoperative OGD with patients often being asymptomatic. Smaller hernias with no evidence of reflux may not need to be repaired but larger ones may influence the type of surgery that is performed.8
Current practice and evidence base
There is a wide range in practice in relation to the use of OGD for preoperative assessment of patients undergoing bariatric surgery due to differing opinion over its diagnostic value and variation in clinical guidelines. A systematic review by Parikh et al including 28 studies encompassing 6616 patients found that 92.4% had a normal OGD and only 7.6% had findings that delayed or altered surgery. Only 0.08% were found to have malignancy.9 A survey of National Health Service (NHS) bariatric units in 2016 reported that 10% of units deemed preoperative OGD ‘completely unnecessary’ and 25% stated they would not be able to accommodate routine OGD in all patients.10 There are significant financial implications to screening all patients undergoing bariatric surgery preoperatively which need to be considered in publicly funded health systems. One systematic review estimated the cost of routine screening as US$22 141 per OGD that altered surgical management. The cost to identify malignancy was US$1.8 million per endoscopy.11
The Society of American Gastrointestinal and Endoscopic Surgeons (SAGES) guidance suggests that preoperative OGD is indicated if a suspicion of pathology exists whereas the European Association for Endoscopic Surgery (EAES) make a conditional recommendation that OGD is considered as a routine test prior to bariatric surgery.12 13 The American Society for Gastrointestinal Endoscopy advises that the decision for preoperative endoscopy should be individualised following discussion with the surgeon taking into account the type of procedure planned.14
Who should be scoped?
If a selective approach is adopted then most would agree it is important that all patients with upper gastrointestinal (GI) symptoms should have an OGD, as this may help predict those likely to develop complications.15 Consideration of risk factors is also likely to be important in identifying those that warrant preoperative OGD. A retrospective review identified that proton-pump inhibitor (PPI) use, age and male gender were associated with abnormal OGD that resulted in a change in surgical approach.16
Helicobacter pylori is a one of the most common findings on preoperative OGDs and eradication is common practice.8 The relevance of a positive result is debated but it seems to correlate with increased risk of developing postoperative marginal ulceration. A recent meta-analysis including 255 453 patients found H. Pylori to be the largest independent predictor for development of marginal ulceration in Roux-en-Y gastric bypass (RYGB), but there was little impact on other surgical outcomes.17 However, the studies forming the evidence base are of relative low quality and EAES has been unable to draw a conclusion. Older SAGES guidance does however advocate eradication therapy.12 While H. pylori determination preoperatively may be important, the option of non-invasive alternatives exist to test H. pylori status, with good sensitivities, such as breath testing or stool antigen, so there is little justification from this perspective alone for preoperative OGD.18
Endoscopy in the postoperative period
Postoperative symptoms in patients who have undergone bariatric surgery include nausea, vomiting and abdominal pain, which may in part be due to non-compliance with the recommended diet, but a persistence of symptoms should warrant further investigations.
Endoscopy has a role in investigating symptoms, diagnosing and managing postoperative complications, as well as investigating causes of weight loss failure. Before attempting to proceed with an endoscopic procedure familiarisation with the postoperative anatomy (figure 1, table 2) and review of any imaging available is critical. Carbon dioxide insufflation may be preferable as it is quickly absorbed. This prevents excessive distension particularly if there are healing anastomoses or a suspected leak.14
Complications following bariatric surgery
Knowledge of the type of operation performed is vital as each technique has different complications (table 3).
Leaks and fistulas
Leaks are defined as the exit of luminal content at an anastomosis. Fistulas are abnormal passages between two epithelialised surfaces, which are the consequence of chronic, healing leaks in the surrounding area. Following RYGB surgery, fistulas usually occur at the gastrojejunal and jejunojejunal anastomoses.5 For SGs, the most common site of leaks is proximally along the oesophagogastric junction staple line.19 The management of leaks involves maintaining the patient nil by mouth, initiating parenteral nutrition, treating sepsis, and draining any collections or repairing the leak surgically where appropriate. If a leak or fistula is suspected, CT with oral contrast should precede endoscopy. Conservative measures including drain insertion are effective 90% of the time.20 Drainage of abdominal collections to promote closure of a leak or fistula can be achieved endoscopically with an internal drain, radiologically with a percutaneous drain, or surgically. Endoscopic techniques for leak closure include the use of fully covered and partially covered self-expanding metal stents, or silicone-coated polyester stents, which aim to protect the area of leakage from gastric secretions, in order to promote healing of the site. Stents are often used in combination with fibrin-glue injection, clip placement or suturing.19 A double pigtail stent can be used in addition to stent placement to allow internal drainage of a collection while promoting re-epithelialisation over the stent until the leak or fistula heals. A residual cavity, a pseudodiverticulum, may remain which has no clinical repercussions.21 A number of studies have shown the efficacy of endoscopic vacuum therapy for the treatment of leaks and in the future this may be adopted as routine practice for their management. It is performed by the endoscopic insertion of an open-pore polyurethane sponge into the abscess cavity. The sponge is sutured in place and continuous suction is applied to the sponge via drainage tubes with an electronic vacuum pump. The sponge is replaced every few days, as the abscess reduces in size, until the leak has fully healed.22
Most GI bleeds occur within 48 hours and are more common after RYGB (1%–4%) than SG (2%).4 The bleeding source is often from the staple lines at the gastrojejunostomy, or more rarely from the jejunojejunostomy, gastric pouch or bypassed stomach.5 Late GI bleeds (>30 days) mostly originate from marginal or peptic ulcers. Minor bleeds can be managed conservatively with fluid resuscitation, PPIs and blood transfusions where indicated. Intraluminal sources of severe haemorrhages will require endoscopic or surgical management, and may require revisional surgery. Endoscopic management of bleeds includes epinephrine injections, thermal therapy and clips. Following the endoscopic intervention, contrast imaging can be performed to rule out a perforation. If an extraluminal source is identified, radiological or surgical intervention will be required.23
Strictures are a late complication of RYGB in 3%–28% of patients and are most commonly found at the gastrojejunal anastomosis, and less often at the jejunojejunal anastomosis.4 Patients will typically present with dysphagia, vomiting and nutritional deficiencies such as protein, vitamin B12, vitamin D, iron and calcium. Persistence of these symptoms require endoscopic evaluation where a stenotic lumen, dilated gastric pouch and non-digested food may be visible.24 Endoscopic dilation of stomal stenosis using balloon dilation is effective and should be considered as first line management for an RYGB stricture.23 The majority of balloon dilations are successful after one or two attempts, and few require surgical revision.5 Where balloon dilation fails, endoscopic scissors have been shown to be successful in removing sutures which are obstructing the lumen.23 The most serious complication is perforation (1.9%–3%).4 Although there has been concern that over dilation might cause weight gain, it has been shown that dilation to 15 mm does not result in increased weight.5 Strictures can also occur following SG (1%–4%) and can arise at the angularis incisura or the gastro-oesophageal junction. There is a lower rate of efficacy of endoscopic therapy for strictures originating from SGs, and surgical revision is usually required.23
Marginal ulcers occur at or near the gastrojejunostomy following RYGB (figure 2). They usually occur within the first 2 years postoperatively and present insidiously with epigastric pain, nausea, and vomiting, or more acutely with bleeding or perforation. The aetiology of ulceration is unclear but local ischaemia, non-steroidal anti-inflammatory use, suture or staple erosion, smoking gastric acidity, and infection with H. pylori are thought to play a role.23
Endoscopy can diagnose ulceration, remove sutures, assess response to treatment, and ensure there is no fistula at the site. Medical management of marginal ulcers with PPIs or sucralfate is usually effective, however small numbers require surgical revision.5 Furthermore, patients should be advised on modifying risk factors, in particular smoking cessation.
Gastric band complications
Complications of laparoscopic adjustable silicone gastric banding (LAGB) include stenosis with oesophageal dilation, reflux, band slippage and band erosion. Eroded bands, which usually occur 1–2 years following surgery, can present with weight regain, abdominal pain and port infection.19 They can be diagnosed at endoscopy, and can be removed endoscopically or surgically.5 Endoscopic removal using a gastric band cutter is possible when the band has eroded sufficiently (more than 50% of the band circumference). The subcutaneous port is removed first, then the cutting wire is inserted via the gastroscope working channel and passed around the band. A mechanical lithotripter is used to cut through the band. The wire and band are removed with the gastroscope.23 A postoperative gastrograffin swallow is usually performed to rule out a leak shown by extraluminal contrast.
Obesity is a risk factor for gallstone formation, as is rapid weight loss, thus gallstones are a common issue for patients undergoing bariatric surgery. Endoscopic retrograde cholangiopancreatography (ERCP) can be performed following an SG or LAGB but following RYGB this procedure introduces technical challenges including the small bowel length and the sharp angle at the jejunojejunal anastomosis. There are a number of ways to perform ERCP following RYGB including laparoscopic assisted access, endoscopic access through a gastrostomy with interventional radiology, double or single-balloon enteroscopy and rotational enteroscopy.5 Some centres carry out prophylactic cholecystectomy at the time of the bariatric procedure in order to reduce the risk of future gallstone disease. Ursodeoxycholic acid is commonly used in identified gallstone disease following bariatric surgery, and is also being used as a prophylactic measure to reduce rates of cholelithiasis.25
Weight loss failure and weight recidivism
Suboptimal weight loss and weight recidivism are recognised problems post-surgery. Weight recidivism occurs in 10%–20% of patients due to factors including non-dietary compliance, physical inactivity, hormonal or metabolic causes and psychological issues around binge eating. RYGB leads to weight loss by restricting food intake and promoting malabsorption. Enlargement of the gastric pouch and gastrojejunostomy reduces the level of restriction, resulting in weight regain. Endoscopy can be used to evaluate the cause of the weight gain as it demonstrates the size of the gastric pouch and gastrojejunostomy. Surgical revision is typically used to manage weight regain, in conjunction with dietetic, psychological and physical activity support. Endoscopic techniques including sodium morrhuate injections surrounding the gastrojejunostomy to reduce the stoma size, and suturing to tighten the gastrojejunal anastomosis have been shown to be effective in reinstating weight loss.4 Argon plasma coagulation is another option which can be used to induce scarring at the site of the anastomosis leading to a smaller diameter of the stoma.26 More recently, radiofrequency ablation has been used to reduce the gastric pouch size which has led to weight loss in a small study of 25 patients.27
Weight recidivism is also seen following SG and gastric banding however these are largely managed by surgical revision. In LAGB, the stomach pouch size can be adjusted by the amount of saline in the balloon around the gastric band, accessed by a subcutaneous port, however band slippage may cause this to be ineffective, requiring further surgery. A retrospective study of five patients has shown promising results using an endoscopic suturing device for sleeve revision by plications, however, this technique is not used routinely, and surgical revision is the mainstay of treatment.28
As the number of bariatric operations being performed continues to rise endoscopy is likely to play an increasing role in the diagnostic work up and management of these patients. The evidence base for preoperative OGD is of low-quality and therefore the utility of endoscopy in this setting remains unclear. The risks of endoscopy along with the financial implications of performing large numbers of procedures must be balanced against the potential benefits of identifying pathology that may alter management. High-quality randomised studies are needed to address this further.
Endoscopists must familiarise themselves both with the anatomical alterations seen following bariatric surgery and also the common complications of these operations. Management of these patients requires a coordinated effort with the surgical team.
Single best answer questions
A patient with obesity is due to undergo a laparoscopic SG. What is the current European guidance regarding indications for preoperative OGD prior to bariatric surgery?
OGD is not required routinely prior to bariatric surgery.
OGD is only justified if the gastric remnant will become inaccessible as a result of surgery.
OGD can be considered as a routine diagnostic test prior to bariatric surgery.
OGD is only required when risk factors for oesophageal cancer are present.
OGD is only required if upper GI symptoms are present.
Answer: option C
Explanation: The the European Association for Endoscopic Surgery (EAES) make a conditional recommendation that OGD is considered a routine diagnostic test prior to bariatric surgery.
A patient undergoes a RYGB.
Two days following her procedure while recuperating on the ward she develops generalised abdominal pain.
Observations: Heart rate 120 beats per minute, blood pressure 98/71 mm Hg, respiratory rate 20 breaths per minute, oxygen saturations 98% on room air, temperature 38.2°C.
Blood test results: C reactive protein 130 mg/L; white cell count 16×10∧9/L.
It is suspected that she may have developed an anastomotic leak. Which of the following is the most appropriate next investigation?
Barium follow through.
CT abdomen with oral contrast.
Answer: option D
Explanation: CT abdomen with water soluble oral contrast such as gastrograffin is the most sensitive imaging modality for detecting anastomotic leaks. This will usually show free fluid, extravasation of contrast material or a perianastomotic fluid collection.
A patient on the ward who is day 1 following a laparoscopic RYGB develops melaena.
What is the most likely cause of her upper GI bleeding?
Bleeding from the staple line at the gastrojejunostomy.
Bleeding from the staple line at the jejunojejunostomy.
Answer: option C
Explanation: This is a case of early upper GI bleeding postoperatively. Most early bleeds occur within 48 hours and the source is most frequently from the staple lines at the gastrojejunostomy. More rarely it can originate from the jejunojejunostomy, gastric pouch or bypassed stomach. Late GI bleeds (>30 days) usually result from peptic or marginal ulcers.
A patient presents several months after a RYGB surgery with 2 weeks of worsening dysphagia to solids and vomiting shortly after meals.
It is suspected that he has developed a stricture and so preparations are made for an OGD with a view to balloon dilatation.
Which site is the stricture most likely to have occurred?
Answer: option D
Explanation: Strictures are usually a late complication of RYGB and are most commonly found at the gastrojejunal anastomosis. Less frequently they can occur at the jejunojejunal anastomosis. Strictures can present with dysphagia, vomiting and nutritional deficiencies such as protein, vitamin B12, vitamin D, iron and calcium. Persistent symptoms require endoscopic evaluation and balloon dilation is considered as first line management of RYGB strictures.
A patient presents to the emergency department with abdominal pain and vomiting. She has a history of previous laparoscopic adjustable gastric banding surgery. Despite initial weight loss she has been gaining weight over the last few months and has struggled to limit her food intake compared with previously.
In light of the probable diagnosis what investigation will confirm the diagnosis?
CT with intravenous and oral contrast.
Answer: option A
Explanation: Band erosion most frequently occurs 1–2 years following surgery. It can present with weight regain, difficulty controlling food intake, abdominal pain and port site infections. OGD will confirm the presence of band erosion. Removal can be performed endoscopically but surgical removal may be required.
Patient consent for publication
Correction notice This article has been corrected since it published Online First. A typographical error in the second sentence has been corrected.
Contributors LT performed the literature review, wrote and edited the manuscript and is responsible for the overall content of the manuscript. AS performed the literature review, wrote and edited the manuscript and and constructed the best of five questions. AR came up with the concept for the review article and was the senior author who reviewed the manuscript before submission and is responsible for the overall content of the manuscript.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.