Over the last decade, endoscopists have been increasingly involved in the management of the post-bariatric surgical patient. Additionally, over the last few years, there's been the exciting introduction of new technologies which may have a role in the primary therapy of obese patients. In this DVD, we will be discussing the current and potential future roles for an endoscopist in the management of a bariatric patient. This video was commissioned by the ASGE for the Endoscopic Learning Library. In this video project, altered anatomy following bariatric surgery and the endoscopic management of the complications encountered following these surgical procedures will be reviewed. Additionally, the primary endoscopic treatments for obesity, none of which are presently approved by the FDA for clinical use, will also be discussed. Obesity is a worldwide health problem of epidemic proportions. According to the World Health Organization, in 2005, 1.6 billion adults were overweight and at least 400 million adults were obese. The World Health Organization projects that by 2015, 2.3 billion adults will be overweight and more than 700 million will be obese. Additionally, in 2005, over 20 million children under the age of five years were overweight. In the United States, according to the Centers for Disease Control, 72.5 million adults, or 26.7% of the American population, were obese in 2007 to 2008. An additional 34% of the adults in the U.S. are overweight. When studied by state, the prevalence of obesity ranges from 18.6% in Colorado to 34.4% in Mississippi. This rather dramatic slide showing the topographic distribution of obesity across the United States was obtained from the Centers for Disease Control website. Lifestyle changes and pharmacologic approaches have largely been unsuccessful in stemming the tide of obesity, and bariatric surgery has become the mainstay of treatment over the last decade. The rapidly growing number of obese individuals, coupled with the recognition and increased awareness of obesity-related morbidity and comorbidity, and the advent of laparoscopic techniques with reduced complications and shorter hospital stays, have led to an exponential increase in the number of bariatric surgical procedures over the last decade. Currently, in the United States, over 110,000 bariatric surgical procedures are performed annually. Given the large number of bariatric surgical procedures, it is not surprising that endoscopists are increasingly being called on to assist in the management of post-operative complications. For optimal care, it is crucial that the endoscopist is familiar with the post-surgical anatomy, complications that can occur, the endoscopic evaluation, as well as the management. Significant technology advances over the last decade have led to a growing interest in minimally invasive techniques which shorter hospital stay and reduce risk of complications. These technology advances have been instrumental in the search for a primary endoscopic treatment for obesity. In this project, the nascent area of primary therapy of obesity and associated pilot data will also be reviewed. The outline for this presentation is as follows. We will first discuss post-bariatric surgical anatomy. We will then look at the endoscopic management of complications, including anastomotic strictures, acute anastomotic leaks, stomal ulcers, the management of gastrogastric fistulas, and complications that can occur following lap band placement. The next section will look at ERCP following gastric bypass. We will then also discuss the primary endoscopic therapy for obesity, and finally discuss the current status of the endoscopic treatment of weight regain following gastric bypass. In this chapter, we will discuss the altered anatomy that is present following bariatric surgery. A thorough understanding of the different types of bariatric surgical procedures and the potential complications one may encounter with each of these procedures is extremely important for any endoscopist who is interested in participating in the multidisciplinary management of these patients. Knowledge of the altered anatomy also allows the endoscopist to choose the right tools for the planned procedure. Bariatric surgical procedures can be divided into three broad groups. Malabsorptive procedures, restrictive procedures, and combined procedures. Malabsorptive procedures, such as the jejuno-ileal bypass, allow the mixture of biliary and pancreatic secretions with the food a short distance from the ileocecal valve. As such, the surface area that's available for absorption of the digested food is limited. Restrictive procedures, such as the vertical banded gastroplasty, gastric banding, which may be either fixed or adjustable, and sleeve gastrectomy, reduce the gastric volume. In addition to reducing the gastric volume, these procedures also work, probably, by altering the neurohumeral mechanisms associated with satiety. Finally, the combined restrictive and malabsorptive procedures, such as the biliopancreatic diversion, duodenal switch, and gastric bypass, combine benefits that are seen with both the malabsorptive and restrictive procedures. We will now discuss the Roux-en-Y gastric bypass, which is the most commonly performed bariatric surgery in the United States. The Roux-en-Y gastric bypass is considered to be the most effective bariatric surgery currently available and is the gold standard against which other bariatric procedures are compared. The Roux-en-Y gastric bypass accounts for 85% of bariatric surgical procedures in the U.S. and 65% worldwide. This surgical procedure involves the creation of an isolated gastric pouch, measuring 15 to 30 cc's in volume, with a 10 to 12 millimeter gastrojejunal anastomotic outlet. When the BMI is less than 50, a roux limb is created measuring 75 to 100 centimeters in length. When the BMI is greater than 50, the roux limb can measure 150 centimeters in length. At the end of the roux limb are two orifices. One leads to the afferent or the biliopancreatic limb. This limb usually measures 35 to 75 centimeters and is the route to access the duodenum and the excluded stomach. The second orifice leads to the efferent limb, which runs on down to the colon. During the Roux-en-Y gastric bypass surgery, the gastrojejunostomy that is created is an end-to-side anastomosis. This results in a 1 to 2 centimeter blind jejunal limb just distal to the anastomosis. The blind passage of wires and catheters across the gastrojejunal anastomosis, for instance in patients who have a gastrojejunal anastomotic stricture, can result in perforation of this blind jejunal limb. Typically, an upper endoscope is sufficient to examine the gastrojejunostomy and the proximal roux limb. However, in order to reach the end of the roux limb, a pediatric colonoscope is usually required. If the roux limb is short, the excluded stomach can also be reached with a pediatric colonoscope. On the other hand, with a very long roux limb, deep small bowel technology, such as single balloon endoscopy, double balloon endoscopy, or spiral endoscopy, is required to access the end of the roux limb and the excluded stomach. This is a patient who has previously undergone a Roux-en-Y gastric bypass. Once a normal gastroesophageal junction is traversed, the gastric pouch is entered. The gastric pouch measures 15 to 30 cc's in volume, and at the distal end of the gastric pouch is the gastrojejunal anastomosis, which should measure between 10 and 12 millimeters in diameter. The endoscope then traverses the gastrojejunal anastomosis and encounters the jejunal aspect of the gastrojejunostomy. As mentioned previously, this is an end-to-side anastomosis, and there is a small blind limb which is a risk for perforation with gastrojejunal anastomotic stricture dilations. The endoscope is then advanced down the roux limb. Depending on the length of the endoscope and the length of the roux limb, the end of the roux limb may or may not be reached. With a pediatric colonoscope, the end of the roux limb is often reached. As one gets closer to the end of the roux limb, bile-stained fluid is visible within the jejunal lumen. Further advancement of the endoscope results in visualization of the jejunal-jejunostomy at the end of the roux limb. One of these two orifices leads to the efferent limb which runs down to the colon. The other orifice is the opening to the afferent limb or the biliopancreatic limb which leads to the duodenum, the major papilla, the minor papilla, and the excluded stomach. In this case, the pediatric colonoscope was switched to a deep small bowel technology for further advancement. The endoscope was then advanced up the afferent limb. Fluoroscopy is useful at this point to confirm that the endoscope is indeed advancing up towards a right upper quadrant into the region where the duodenum is expected. Once the duodenum is entered, the major papilla and the minor papilla are visualized. In this case, the major papilla is towards 3 o'clock and the minor papilla is towards a 12 o'clock position. The tangential approach to the major papilla renders cannulation of the bile duct and pancreatic duct difficult during ERCP. As the endoscope is advanced up the duodenum, the duodenal bulb is encountered. A retrograde view of the pylorus is seen. Once the pylorus is traversed with the endoscope, the excluded stomach can then be examined. This access into the excluded stomach will also permit the placement of a gastrostomy tube if required for endoscopic access for ERCP. In this chapter, we will discuss laparoscopic banding procedures which are being done in increasing numbers in the United States and worldwide. The lap band procedure involves the laparoscopic placement of a band around the proximal gastric wall, leading to a reduction in appetite and food intake. This procedure results in the creation of a gastric pouch which measures 15 to 20 ml in size. The lap band is connected to a subcutaneous reservoir by a thin tube. The subcutaneous port provides for access for injecting saline to adjust the degree of restriction in patients. The drawing on the left shows the lap band present around the upper portion of the stomach creating a small gastric pouch superior to it. The lap band is connected via a thin tube to the subcutaneous reservoir. The picture on the right side shows a close-up view of the lap band itself. There is a small buckle-like structure at the point that the thin tube connects to the lap band. During endoscopic extraction of eroded lap bands, this area can present significant resistance to extraction. These are endoscopic pictures in a patient who has previously undergone lap band placement. The picture on the left shows a normally appearing gastroesophageal junction. There is minimal resistance to advance of the endoscope through this area into the stomach. The picture on the right shows a normal appearing post-lap band anatomy. The gastric cardia is fairly snug around the 9.8 mm endoscope. This is endoscopic footage in a patient who has had a lap band placed. The distal esophagus, as you can see, looks relatively normal. The gastroesophageal junction also appears to be normal. Once the gastroesophageal junction is crossed, you can see the small gastric pouch, which should measure between 15 to 30 cc in volume. A perplexed view of the stomach shows a well-positioned lap band. The cardia is snug around the 9.8 mm endoscope. Occasionally, gastric bands may slip. It is important for the endoscopist and the treating surgeon to be comfortable with the normal appearance of the gastric band on radiologic studies. The normal orientation of the band is close to 40 degrees to the horizontal. The band should run from approximately 1 to 2 o'clock position on the dial to the 7 to 8 o'clock position. There should be unimpeded flow of contrast into the distal stomach across the location of the band. We will now briefly discuss the vertical banded gastroplasty procedure, also known as a VBG. During this procedure, a transgastric window is made 6 to 8 cm below the angle of hiss. A pouch of 30 ml is created with a circular and linear stapler. The narrow outlet, which measures 10 to 11 mm in diameter, is surrounded by a ring of polypropylene, PTFE or silicone to avoid dilation. Due to complications that can be encountered following vertical banded gastroplasty, as well as the availability of other surgical procedures with better results, the vertical banded gastroplasty has fallen out of favor over the last decade. Sleeve gastrectomies are also occasionally performed as bariatric surgical procedures to achieve weight reduction. During this procedure, a narrow tubular stomach is created along the lesser curvature with resection of the remnant greater curvature. This procedure may be performed either alone or in combination with a biliopancreatic diversion. This is a patient who has previously undergone a sleeve gastrectomy. The staple line from the surgery can be seen on the left upper side of the screen. As is evident in this endoscopy footage, the gastric volume has been significantly reduced. Once the endoscope is advanced beyond the sleeve gastrectomy, the endoscope enters into a normal appearing antrum. The pylorus and the duodenal anatomy beyond this point is also entirely normal. Another procedure that is occasionally performed, particularly in those who are super obese, is the biliopancreatic diversion with sleeve gastrectomy. This procedure includes a sleeve gastrectomy, as was previously discussed, together with a transection of the post-pyloric duodenum. A 250 cm rule limb is anastomosed to the proximal duodenum. The biliopancreatic limb and the rule limb are anastomosed about 50 to 150 cm proximal to the ileocecal valve. During this procedure, undigested food material traverses a large extent of the small bowel. The admixture between the biliary and pancreatic secretions with the food takes place only a short distance away from the ileocecal valve. This results in significant malabsorption. The presence of a sleeve gastrectomy also provides a restrictive component to this procedure. I will now list some of the more common complications associated with each of these bariatric surgical procedures. Following a Roux-en-Y gastric bypass, it is not uncommon to encounter anastomotic strictures, marginal ulcers, staple line disruption, and symptomatic cholelithiasis. On the other hand, following a lap band placement, the complications tend to be different. They include pouch dilation, displacement of the band, band erosion, and infection of the band or the reservoir. Following vertical banded gastroplasty, patients can experience staple line disruption, stenosis, occluded stoma, gastro-gastric fistula, enlarged stoma, band erosion, and anastomotic leaks. As is evident from these lists, vertical banded gastroplasty is associated with a higher number of complications as compared to lap bands and Roux-en-Y gastric bypass. Patients who have previously undergone bariatric surgical procedures can present with clinical problems that can vary from relatively simple to treat to extremely complex. It is extremely important for the endoscopist to recognize the altered anatomy that can be encountered following bariatric surgery and the complications which may occur. It is very useful for the endoscopist to review operative reports and imaging studies prior to embarking on an endoscopic evaluation. A thorough knowledge of the altered anatomy will allow the endoscopist to choose the appropriate endoscope system. In addition, it will also help plan for the procedure by having accessories of appropriate length and caliber available depending on the interventions that are planned. In this chapter, we will discuss the anastomotic strictures that typically occur following Roux-en-Y gastric bypass. In Roux-en-Y gastric bypass surgery, the gastrojejunal stoma is crafted to approximate 10-12 mm in diameter. Anastomotic strictures occur in about 2-27% of patients undergoing gastric bypass. The strictures are believed to be related to tension on the anastomosis, foreign body reaction, technical error, marginal ulcers, ischemia, or leaks associated with scarring. The surgical technique, as well as the choice of stapler, for instance, a circular stapler, an EEA stapler, or linear stapler, influence the incidence of stomal stenosis. Patients who have a gastrojejunal anastomotic stricture typically present with intolerance to solid foods, which may progress to liquids depending on the degree of stenosis. The typical history is that ingested food gets stuck or is promptly vomited. Delayed vomiting, or vomiting of material other than ingested food, often suggests an etiology other than stomal stenosis. This scenario is particularly common in individuals shortly after having a gastric bypass surgery who are still getting used to the new dietary restrictions. Strictures typically present 3-8 weeks following surgery. Occasionally, presentation may be delayed by months. The typical method of diagnosis of an anastomotic stenosis is by endoscopy. Barium swallow is also diagnostic, but however does not have any therapeutic potential. As such, in a patient who has a history consistent with stomal stenosis, an upper endoscopy is typically performed for both diagnosis and treatment. In the radiograph shown on this slide, a contrast study is being performed following gastric bypass. The contrast is noted to fill the gastric pouch. Immediately distant to the gastric pouch, there is a stenosis in the region of the gastrojejunal anastomosis. Some contrast is also seen in the jejunum. The typical treatment for anastomotic strictures following Rouen-Wei gastric bypass is endoscopic balloon dilation. In the first picture here, you can see the gastrojejunal anastomotic stricture. This stricture probably measures about 2-3 mm in diameter. In the second picture, you can see balloon dilation being performed. And finally, in the third picture, you can see the appearance of the gastrojejunal anastomosis following full dilation. The degree to which an anastomotic stricture should be dilated is still a subject of some degree of controversy, and we will be talking about this further in the next few slides. It is important to employ a safe technique during dilation of these gastrojejunal anastomotic strictures, particularly the very tight strictures in which one cannot see the jejunal aspect across the tight anastomosis. It is best to advance only about 1-2 cm of the balloon across the stricture for the initial dilation, then to advance the endoscope across the anastomosis, confirm the correct positioning of the balloon in the roux limb, and then complete the dilation. A careful technique will reduce the chances of perforation of the blind jejunal limb distal to the gastrojejunal anastomosis. On occasion, one may encounter extremely tight strictures that do not allow easy passage of the dilating balloon. Such procedures may be performed more safely with fluoroscopic guidance. We shall now review two endoscopic dilations for gastrojejunal anastomotic strictures. In the video that you are seeing right now, this patient presented with significant dysphagia to swallowing solids. As you can see from this video, the luminal diameter is marginally less than that of the endoscope, which measures about 9.8 mm on its outer diameter. However, the patient is having significant problems with dysphagia and it was decided to proceed with dilation. You can see the presence of the blind limb here on the right side, which would be a very high perforation risk for inadvertent blind passage of wires and dilating catheters. A safe technique for all dilations, as is seen in this case, is to advance the endoscope down the root limb, push the balloon out of the tip of the endoscope, and under endoscopic visualization, withdraw the scope, leaving the tip of the balloon safely in the root limb to allow dilation of the gastrojejunal anastomosis. In this case, the gastrojejunal anastomosis was dilated to about 15 mm and the duration of inflation of the balloon can vary anywhere from 30 seconds to a minute. In the next patient, the gastrojejunal anastomotic stricture measures approximately about 5 mm in diameter. The balloon is easily advanced across the stenosis without meeting any resistance on the other side and sequentially inflated to 15 mm. The balloon is kept inflated for about 30 seconds prior to deflation. Following deflation of the balloon, it is easy to understand why inadvertent blind passage of the wire or the balloon catheter can cause perforation of the blind limb distal to the gastrojejunal anastomosis. The degree to which a gastrojejunal anastomotic stricture should be dilated has been the subject of some controversy. Our group has previously reported on 43 patients with strictures of the gastrojejunal anastomosis out of 801 patients who underwent Roux-en-Y gastric bypass. The strictures were dilated to a mean of about 15.5 mm. There were no perforations or significant bleeding following these dilations. 79% of the patients were managed with one dilation and 93% were managed with one or two dilations. A lower rate of repeat dilation was found if the first dilation was greater than or equal to 15 mm compared to those patients in whom the first dilation was less than 12 mm or equal to 12 mm in diameter. The weight loss in our cohort of patients at 6 months and 12 months was not adversely affected by aggressive dilation to 15 mm or higher. Concerns still continue to remain that excessive dilation of the gastrojejunal anastomosis may result in loss of efficacy of the Roux-en-Y gastric bypass. An uncontrolled retrospective analysis of 165 patients revealed a positive association between gastrojejunal stoma diameter and the percentage weight regain following Roux-en-Y gastric bypass. At 5 years after Roux-en-Y gastric bypass, each 10 mm increase in gastrojejunal stoma diameter was associated with an 8% increase in the percentage of maximum weight loss that was regained. The precise significance of such data, however, with regard to dilation of the gastrojejunal anastomotic stricture to 15 mm remains unclear and in our institution, based on our experience, we have continued to dilate an anastomotic stricture to 15 mm during the initial presentation. Another location where a small bowel stricture may be encountered following Roux-en-Y gastric bypass is in the transverse mesocolon. During the surgical creation of the Roux-en-Y gastric bypass, the Roux limb is tunneled through a window in the transverse mesocolon and brought up to the gastric remnant for the gastrojejunal anastomosis. Small bowel obstruction can occur at this site due to internal herniation. Endoscopy typically reveals a narrowed Roux limb several centimeters distal to the gastrojejunal anastomosis. Treatment is usually surgical and endoscopic attempts at dilation may result in perforation. In this chapter, we will discuss acute anastomotic leaks that can occur following Roux-en-Y gastric bypass. These leaks occur in the immediate post-operative setting and can sometimes be difficult to diagnose. Acute anastomotic leaks result in intra-abdominal sepsis and can be a life-threatening complication. The incidence can vary from 1 to 6% following bariatric surgery. The leakage rates, however, may be higher following revisional bariatric surgery. As mentioned earlier, acute anastomotic leaks may be difficult to diagnose. These are patients who have recently undergone a bariatric surgery and abdominal examination can be difficult. Such a diagnosis can be suspected in the presence of persistent drain output, concerns for intra- abdominal sepsis, the presence of fever, tachycardia, tachypnea, and finally the presence of fluid collections on cross-sectional imaging. Often the key to early diagnosis of an acute anastomotic leak rests with a high index of suspicion in a post-operative patient with signs suggestive of sepsis. CAT scans with oral contrast may assist in establishing the diagnosis as well as the location of the leak. Since these patients are often critically ill, laparotomy is the mainstay of treatment in the acute setting. It is important to remember that the leak site might not be very large. Leaks can arise from defects as small as one to two millimeters in size or a few centimeters in diameter. The diagnosis of an acute anastomotic leak or fistula is often a clinical diagnosis. An endoscopy is not required to establish the diagnosis of an anastomotic leak or fistula in a critically ill patient. Such patients based on their clinical presentation oftentimes will undergo exploratory laparotomy without the need for an endoscopy. If an endoscopy were to be performed, the size of the fistula may be very very small. It may be very difficult to appreciate a one to two millimeter fistula in a setting of recent surgery. On the other hand, occasionally large defects as demonstrated in this particular slide can be seen. In a series of 3,018 patients, 2.1% of the patients had an anastomotic leak. In about 49% of the patients, the leak was at the gastrojejunostomy site. In 25% of the patients, the leak was at the excluded stomach. In 13% of the patients, the leak was at the jejunojejunostomy. And in about a quarter of the patients, the leak was at the divided staple line of the gastric pouch. In patients who are clinically stable for further evaluation, endoscopy and radiologic contrast studies may assist in identifying the precise location of the leak. While the precise reason for the occurrence of an anastomotic leak in any given patient will remain unclear, there are several potential etiologies that may be playing a role. This could include increased tension of the anastomosis, failure of the stapling device, poor surgical technique, obstruction distal to the site of the leak, or ischemia. Given that these patients are critically ill, treatment usually consists of early re-operation. Occasionally, leaks may result spontaneously with percutaneous drainage of fluid collections. Endoscopic therapy is limited to medically stable patients with a subacute presentation. In clinically stable patients, the use of self-expanding covered stents to treat acute anastomotic leaks has been reported. In this series, 17 patients with a gastrojejunal leak following gastric bypass underwent self-expanding metal stent placement one to three weeks after surgery. All the stents were removed endoscopically approximately about three months later. 16 of the 17 patients had a completely healed leak. One patient, however, had a persistent gastro-gastric fistula. It is important to note that none of the covered stents available on the market today are approved by the FDA for this particular indication, so the use of a covered stent for therapy of an anastomotic leak is an off-label indication. The choices that one has available would include the self-expanding covered plastic stents as well as a self-expanding covered metal stent, all of which should have the ability for removal at a later date. We will now review the care provided to a patient who had an anastomotic leak following a Roux-en-Y gastric bypass. In the immediate post-surgical setting, a percutaneous drain was placed to provide drainage of the fluid collection. Since the patient was clinically stable, the surgeon elected to observe the patient for a couple of weeks. At that point, the drainage via the percutaneous drain persisted and an upper GI series was performed. The radiographs on the slide show leakage of contrasts from the gastric pouch into the region where the drain has been placed. The radiographs also suggest that the size of the fistula itself is very small. A decision was then made to pursue endoscopic treatment for this leak. The goal of endoscopic therapy was to place a self-expanding fully covered metal stent across the region of the leak. During the upper endoscopy, not surprisingly, the precise site of the leak itself could not be identified. Using standard techniques for self-expanding metal stent placement, a wire was left running from the root limb across the gastrogynostomy and gastric pouch. Under fluoroscopic and endoscopic visualization, a self-expanding metal stent, which is fully covered, is then deployed across the region. Subsequent endoscopic visualization confirms that the stent now spans the lower esophageal sphincter region, the gastric pouch, the gastrogynal anastomosis, and extends down into the root limb. A post-procedure radiograph demonstrates good placement of the self-expanding metal stent. Following placement of the stent, the drainage via the percutaneous tube stopped completely. The patient was subsequently brought back for stent removal about three months after initial placement of the stent. At the time of stent removal, the wire at the proximal end of the stent is grasped and pulled into the endoscope. This effectively cinches or narrows down the proximal end of the stent. Gradual traction is applied and the stent is removed under endoscopic and fluoroscopic visualization, depending on the type of stent that is being removed and the degree of difficulty. Occasionally, the use of a double channel endoscope with the forceps going down each channel grabbing opposite sides of the stent may be useful for easy stent extraction. Endoscopic visualization following stent removal demonstrates stent related changes in the lining. These changes extend from the distal esophagus, where the proximal end of the stent was located, to the gastric pouch, where you can now see ulceration and staples in the region of the gastrointestinal anastomosis. These changes also extended down into the rulum, where the distal end of the stent was located. The leak did not recur in this patient following removal of the stent. A follow-up contrast study following stent removal confirms that the leak has indeed healed completely. The following panels show the radiographic and endoscopic correlates in a patient who was undergoing management of an anastomotic leak with the placement of a self-expanding plastic stent. The radiographs show a demonstration of the leak by barium study and the subsequent placement of a self-expanding plastic stent. The endoscopic picture A shows the region of the anastomotic leak. Endoscopic picture B shows the self-expanding plastic stent following deployment. Endoscopic picture C shows the self-expanding plastic stent that is barely visible at the time of stent removal. And finally, endoscopic picture D shows the site of the anastomotic leak once the stent has been removed and the mucosal surface has healed. In summary, patients with acute anastomotic leaks following Roux-en-Y gastric bypass are typically critically ill. These patients usually require surgical exploration and closure of the leak site. If the leak is subacute and the patient is clinically stable, off-label use of a covered, removable stent may assist in closure of the leak and avoid reoperation. While there is no definite data regarding the timing of stent removal, anecdotal experience and published literature suggests that a duration of approximately three months following initial stent placement should be adequate to ensure healing of the leak site. In this chapter, we will discuss the clinical presentation, evaluation, and management of stomal ulcers. This is a patient who underwent a Roux-en-Y gastric bypass approximately four years ago. This patient presented to our hospital with abdominal pain. On the endoscopic picture on the left side, the yellow arrow shows the gastrogastric fistula. The red arrowhead shows the gastrojejunal anastomosis, and the blue arrowhead shows the anastomotic ulcer. In the right side is a different view of the gastrogastric fistula on the lower aspect of the endoscopic photograph, and the blue arrow shows the anastomotic ulcer. Anastomotic ulcers occur following Roux-en-Y gastric bypass at a rate of 0.6 to 16 percent. Patients may present with retrosternal or epigastric pain, dyspepsia, or bleeding. It is believed that surgical technique, ischemia, use of non-steroidal agents, and H. pylori may play a role in the pathogenesis of these stomal ulcers. A larger gastric pouch with greater amounts of acid production, reaction to sutures or staples, and the presence of H. pylori may be contributory to the formation of these ulcers. The clinical presentation of a stomal ulcer varies from patient to patient. If there is an ulcer and a stenosis of the gastrojejunal anastomosis, patients may present with pain and dysphagia. On the other hand, patients with ulcers alone may have persistent stabbing or burning pain in the epigastrium of varying severity. Additionally, sometimes patients with stomal ulcers can have hematemesis or Melna with or without epigastric pain. When a stomal ulcer is suspected, an upper endoscopy is usually performed to make the diagnosis. When a stomal ulcer is identified, evaluation for H. pylori is important. This can be accomplished either by gastric pouch biopsies or subsequent fecal antigen tests. Occasionally, deep stomal ulcers may lead to gastrogastric fistulas. As such, when a deep ulcer is noted, it should be carefully examined for the presence of such a fistula. An upper GI series may be required if a suspicion for fistula is present but is not visualized in endoscopic examination of the ulcer. Gastrogastric fistula may range in size from a pinpoint opening to a few centimeters in diameter. The smaller size openings may not be readily appreciated at endoscopy. This patient has previously undergone a Rouen Y gastric bypass. Upon entry into the gastric pouch, the gastrointestinal anastomosis is visualized. An ulcer is seen at the anastomosis here in the 12 o'clock position. The ulcer appears to be cratered. The endoscope is advanced into the ulcer in an attempt to see if there is a communicating gastrogastric fistula. However, no such fistula is identified in this case. Treatment of a stomal ulcer includes the eradication of H pylori. It is not yet clear whether this eradication should be performed empirically without necessarily testing for H pylori or should be performed only upon histologic confirmation of the presence of H pylori. Cessation of the use of non-steroidal agents and steroids is also extremely important to assist in the healing of these ulcers. Additionally, therapy with proton pump inhibitors, sucralphate, and smoking cessation are equally important in assisting in the healing of these ulcers. If sutures or staples are present in the ulcer bed, endoscopic removal of these sutures or staples may also assist in healing. With these recommendations, the stomal ulcers usually heal in over 90% of the patients. Occasionally, an ulcer that is resistant to such therapy may require surgical revision. As a preventive measure, some surgeons also recommend pre-operative testing for H pylori and treatment if present. This endoscopic footage is from a patient who has undergone Roux-en-Y gastric bypass several years previously. The gastric pouch is entered and the gastrojejunal anastomosis is visualized. As the scope is advanced, on the jejunal aspect of the anastomosis, a deep, cratered ulcer is visualized. This patient presented with severe, persistent abdominal pain. This same patient happened to undergo a repeat endoscopy about three weeks after the recommendations, which included PPI therapy, stopping smoking, stopping NSAIDs, and sucralfate. The abdominal pain had significantly improved, and as you can see on this endoscopic footage, over this three-week period, the appearance of the ulcer is also dramatically better. Occasionally, non-absorbable suture may persist in the region of the anastomosis and cause either luminal obstruction related to the impaction of food or marginal ulceration. The use of endoscopic scissors to assist in the removal of suture material has been described. This video clip, provided by Drs. Mulledy and Thompson, demonstrates the use of an endoscopic scissor to cut residual suture material in the region of the gastrojejunal anastomosis. The suture material can then be easily extracted with the forceps. In summary, stomal ulcers can present with abdominal pain, bleeding, or gastrogastric fistula. An upper endoscopy is usually diagnostic. PPI therapy, cessation of NSAID use, cessation of smoking, treatment of H. pylori, and the use of sucralfate can successfully treat over 90% of ulcers. Surgical revision may be required for non-healing ulcers. In this chapter, we will be discussing the management of gastrogastric fistulas. Gastrogastric fistulas occur as a result of dehescence of the staple line. In our anecdotal experience, these fistulae are progressive in that over time the size gradually increases. The incidence can vary from 4 to 29%, and these patients may present with nausea, epigastric pain, weight gain, or occasionally patients may remain totally asymptomatic. An upper GI contrast study is typically diagnostic for a gastrogastric fistula. It should be noted that sometimes gastrogastric fistula may be extremely small, and when they are very small, occasionally the upper GI contrast study is the only way to establish a diagnosis of a gastrogastric fistula. Larger fistula can also be well visualized at endoscopy. In the presence of a fistula following ruon by gastric bypass, the endoscope can be advanced through the fistula to examine the excluded stomach, the pylors, and the duodenum, which normally would not be accessible via the gastric pouch. In this radiograph, the gastric pouch is shown by the black arrow. The excluded gastric stomach is outlined by the white arrows. Contrast is seen to flow down the esophagus into the gastric pouch. Then, via a small gastrogastric fistula, contrast then begins filling the excluded stomach. Occasionally, small fistulas may be identified easily at radiographic studies, but however, may prove to be difficult to identify endoscopically. This is a patient who underwent an upper endoscopy about three years following a ruon by gastric bypass for abdominal pain. During the endoscopy, a small gastrogastric fistula is visible in the 10 o'clock position on the screen. The fistula is not large enough to allow passage of the endoscope. The same patient underwent a repeat upper endoscopy about four years later. At the time of this endoscopy, you can see that the fistula has increased in size significantly. Additionally, the endoscope is able to be advanced through the fistula into the excluded stomach, as you can see on the right side. The presence of such a large fistula, as you can imagine, can easily lead to loss of effectiveness of the ruon by gastric bypass. Occasionally, patients may also begin to have problems with acid reflux disease, because a larger amount of gastric acid can now reflux through the gastrogastric fistula into the gastric pouch and up the esophagus. In order to minimize the weight regain following the development of a gastrogastric fistula, as well as other symptoms which may occur, in the past, surgical revision has been the definitive therapy. Given the dramatic advances in endoscopic tools and accessories over the last decade, increasing attention has been paid to the possibility of an endoscopic treatment for gastrogastric fistula. However, long-term successful closure of gastrogastric fistula with endoscopic therapy has remained elusive. Several traditional and newly developed endoscopic tools have been employed in attempts to close gastrogastric fistulas. During endoscopic therapy, the initial step is oftentimes a mucosectomy. The purpose of the mucosectomy is to remove the mucosa from the lining of the fistula to promote closure and healing of the fistula once tissue apposition is achieved. This mucosectomy can be performed either with argon plasma coagulation, the use of electrosurgical energy and snares, needle-knife cautery, hot or cold biopsy forceps, followed by one or more of the following, including fibrin glue, endoscopic clips, and endoscopic suturing devices. Over the last decade, there have been some rather dramatic advances in our ability to perform tissue apposition via the endoscope. These advances have opened up the possibility and the hope that an endoscopic therapy can be devised to avoid re-operation for gastrogastric fistulas following Roux-en-Y gastric bypass. We will now review some of the therapies that have been employed for endoscopic closure of fistulas following Roux-en-Y gastric bypass. The radiograph on this slide demonstrates the presence of a jejunal cutaneous fistula following a Roux-en-Y gastric bypass. The upper endoscopic photograph shows that a mucosectomy has been performed with the use of the argon plasma coagulator to ablate the edges of the fistula, and in the lower picture you can see clip closure of the fistula. Despite seemingly excellent endoscopic results, unfortunately the jejunal cutaneous fistula in this patient was not successfully treated, and a few months later the patient underwent surgical revision for treatment of this problem. A recent publication also looked at the use of clips to close gastrogastric fistula in eight patients. There was short-term failure in four patients, and the fistula remained closed in the remaining four patients for a time duration of 8 to 48 months. Small series, such as this series, as well as anecdotal experience, suggest that for very small gastrogastric fistula, clip closure may be a viable option. In another series of 95 patients with gastrogastric fistula, amongst those with prior Roux-en-Y gastric bypass, fistula size was noted to be about 2 to 40 millimeters, with the average being about 12.7 plus or minus 8.6 millimeters. 49% of the fistula were located in the proximal pouch, 23% in the mid pouch, 18% in the distal pouch, and 10% at the gastrojejunal anastomosis. Therapies used to try to close the gastrogastric fistula varied across these patients. They included closure with either the endosynch system or hemoclips, along with glue and argon plasma coagulation. The following video demonstrating the use of the endosynch suturing device to close a gastrogastric fistula was submitted by Dr. Christopher DiMaio and Dr. Christopher Thompson. This is a patient with a relatively large gastrogastric fistula following a Roux-en-Y gastric bypass. The endoscope is easily advanced through the gastrogastric fistula into the excluded stomach. The rim of the fistula is denuded of its mucosa with the use of the argon plasma coagulator. The suturing device with its suture loaded is advanced down an overtube up to the fistula. Suction is then applied to bring the tissue into the chamber and the needle is advanced, resulting in the suture material passing through the acquired tissue. The scope and the suturing device are removed, the suturing material is reloaded, and the suturing device is reintroduced. This time, the opposing wall of the rim of the gastrogastric fistula is acquired for suturing. This process is then repeated with a new suture to result in two interrupted stitches at the fistula site. Using a new delivery device, the sutures are cinched together and cut. Finally, to conclude the procedure, the area is injected with the fibrin glue for complete closure of the fistula. Of the 95 patients, 73 were available for follow-up at a median of 217 days. 14 of these 73 patients demonstrated durable closure. None of the fistula which were initially greater than 20 millimeters in diameter remained closed at the end of the study period. Additionally, only 32% or 10 out of 31 patients with fistula less than 10 millimeters in diameter remained closed at the end of the study period. Dr. Lee Swanstrom and his group recently reported their experience at closing gastrogastric fistulas using a tissue acquisition system. The system essentially consists of a T-tag applier, which is a hollow needle attached to a plastic sheet and with a stylet to eject the T-tag. The T-tag applier is loaded with one tissue anchor at a time. The tissue anchor consists of a monofilament non-resorbable polypropylene thread fixed to a stainless steel anchor element. The T-tag is deployed by using a pusher knob at the top of the handle. When one T-anchor is in place, the next anchor is loaded the same way and deployed at the opposite side of the fistula by using the same working channel. The knotting element consisting of an implantable polymer is delivered with its own knotting element applier. During the procedure, the first step was the mucosectomy. This was achieved by using the needle knife cautery, snares and hot cold biopsy forceps. The tissue anchors with the attached permanent sutures were then deployed through the full thickness of the gastric wall by using delivery needles to the working channel of the endoscope. After two anchors were deployed on both sides of the fistula's orifice, a knotting element was cinched down to approximate the two sides of the fistula's orifice. The procedure was repeated until the fistula's orifice was securely closed. This video demonstrating the use of the tissue apposition system to close a gastrogastric fistula was provided by Dr. Lee Swanstrom. Mucosectomy was initially performed with a needle knife sphincter tone and with a grasper. The T-tag applier, which is basically a hollow needle with a plastic sheath and a stylet to eject the T-tag, has to be loaded with one tissue anchor at a time. The tissue anchor consists of a monofilament non-resorvable 3-0 polypropylene thread fixed to a stainless steel anchor element. The system can be introduced through a 2.8 millimeter working channel, but we preferred the 3.7 millimeter channel to have less friction for the introduction of the second anchor. Friction can lead to malfunction like ripping of the thread. Once in place, the needle can be advanced and locks in half centimeter steps on the handle. The T-tag can be deployed separately using a push knob at the top of the handle. When one T-anchor is in place, the next anchor is loaded the same way and deployed at the opposite side of the fistula using the same working channel. The knotting element, which consists of implantable polymer, is delivered with its own knotting element applier. After pushing down the loaded knotting element applier through the same working channel that the tissue anchor threads are in, the suture can be tightened by pulling on both ends of the thread while the endoscope is fixed and suture cut in the same process and can be pulled out of the working channel. This tissue apposition system was used in four patients with five gastrogastric fistula by Dr. Swanstrom's group. The mean fistula diameter was 18.6 millimeters with a range of 10 to 30 millimeters. The treatment, as was shown in the video, involved an initial mucosectomy followed by tissue apposition. Unfortunately, at six months, all of the fistula had rickered. The last fistula to ricker was the smallest fistula, which measured approximately 10 millimeters in diameter at the initial endoscopy. In summary, patients who have a gastrogastric fistula may present with weight gain, reflux symptoms, or pain. Occasionally, these patients may be asymptomatic. An upper GI series or an upper endoscopy is diagnostic. Surgical treatment is the definitive treatment for a gastrogastric fistula. Several investigators have studied modalities such as endoscopic clips, endoscopic suturing, and tissue apposition techniques following mucosectomy. Unfortunately, all of these endoscopic techniques have been unsuccessful at achieving a durable closure of the gastrogastric fistula in the majority of patients. In this chapter, we will discuss the complications that can arise following adjustable lap band surgery. Common complications that may require endoscopic evaluation and management include pouch dilation. This might be related to overinflation of the band or excessive food intake, band erosion, band slippage, and port or tube migration. Of these complications, band erosion is probably the most common. The incidence ranges from 0.5% to 3.8%. Band erosion is best diagnosed at endoscopy. While occasionally band erosion may be asymptomatic, typically patients present with abdominal pain, nausea, or access port site infection. Once the band erodes into the gastric lumen, bacteria can track along the band and the tube to cause an infection at the port site. An eroded band is also not functional, hence occasionally these patients may present with increased food intake or weight gain. Finally, as the band erodes into the stomach, gastrointestinal bleeding is also a possible clinical presentation. While in the past an eroded lap band was often treated surgically, there has been increased attention over the last few years to endoscopic removal of these eroded bands. The techniques that have been employed to disrupt the band include lasers, electrosurgical tools, endoscopic scissors or band cutters, and mechanical lithotripsy devices. In our experience, and that reported by others, the mechanical lithotripsy devices to disrupt the eroded lap band and then subsequent endoscopic removal appear to be the most effective. We will next review the endoscopic removal of an eroded lap band in a patient. These procedures are typically carried out under general anesthesia in the operating room. In the first part of the procedure, the surgeon performs a skin incision in order to extract the axis port from its subcutaneous site. The endoscope is introduced trans-orally and advanced down the esophagus. In the vicinity of the GE junction, the eroded lap band is clearly visible. This is much better apparent when the endoscope is retroflexed and you can see the eroded lap band in the region of the cardiac. A 0.025 inch guide wire is advanced around the lap band. The use of a balloon or a sphincter tube may help if the lap band has only minimally eroded into the stomach. On the other hand, if a significant portion of the band is in the stomach, the endoscope can be advanced on either side of the band. Once the wire is advanced on one side of the band, the endoscope is removed, leaving the wire in place. The endoscope is then reintroduced. The free end of the wire is grasped and brought out trans-orally. This leaves the wire passing from the mouth down the esophagus around the band and the free end also traverses back up the esophagus and exits the mouth. The two free ends of the guide wire, which is now looped around the lap band, are advanced into a mechanical lithotriptor. While the mechanical lithotriptor is being slowly cranked, the metal sheath is advanced over the guide wires to reach the lap band. The advancement of the metal sheath over the guide wires is best monitored endoscopically. Under endoscopic visualization, once it is confirmed that the metal sheath is indeed up against the lap band, the mechanical lithotriptor is progressively cranked. In our experience, this is best done slowly, otherwise the wire itself might break. As gradual pressure is applied on the lap band, the lap band eventually breaks. Once the lap band has been disrupted, it is grasped with a snare or a forceps, freed from the gastric wall, and then extracted trans-orally. Prior to the extraction of the lap band, the surgeon cuts the connection between the tubing and the axis port at the skin incision site. Dr. DeVere and his group recently reported on 13 patients, of whom 3 had eroded lap bands, 4 had eroded sylastic rings, and 6 had refractory outlet stoma stenosis following vertical-banded gastroplasty. A unique feature of their approach was that in patients who had minimally migrated bands, a self-examination of the lap band was performed. A unique feature of their approach was that in patients who had minimally migrated bands, a self-expanding polyethylene stent was placed across the band to induce complete migration. This was done in 10 of their 13 patients. Six weeks later, the self-expanding polyethylene stent was first removed, and subsequently the band of the ring was disrupted with a wire and Atkinson extractor. The band was easily extracted via a trans-oral route. In the first picture, you see the endoscopic view of a partially eroded lap band. The second picture is a radiograph showing the deployment of a self-expanding plastic stent through the partially migrated lap band. And the third picture is an endoscopic view of the self-expanding plastic stent traversing the lap band. Lap band slippage is a problem which has to be addressed surgically. On the radiograph on the left, you can see that the lap band has migrated from its typical position. The radiograph on the right side is representative of a contrast study that was performed following lap band slippage. In this study, you can see that contrast fills the gastric pouch immediately above the lap band, however, passage across the lap band itself is very slow. In contrast to the open ring appearance of the slipped lap band, a normally placed lap band should be seen as an opaque block extending from the 1 o'clock or 2 o'clock position to 8 o'clock or 7 o'clock. Following surgical revision, the lap band now is in its expected position, and the contrast studies demonstrate that contrast flows very easily across the restriction from the lap band. Erosion of the lap band tubing into the stomach is an unusual complication following lap band surgery. This is a patient who presented to his surgeon with infection around the axis port site. An upper endoscopy was performed with the expectation that an eroded band would be seen. However, on the left side, you can see that the band itself is in good position without the appearance of any erosion. On the other hand, lower down in the gastric body, the lap band tubing has eroded into the stomach. This patient was treated with surgical explantation of the entire lap band system. We have previously mentioned in the chapter on post-bariatric surgery anatomy that occasionally bands are placed to reinforce and prevent dilation of the gastric pouch outlet following Roux-en-Y gastric bypass or vertical banded gastroplasty. Occasionally, these sylastic bands can erode into the stomach. Removal can be performed endoscopically. In this patient, endoluminal erosion of the sylastic band is clearly visible. The sylastic band was disrupted with a needle knife in this instance, and following disruption, it was grasped with the forceps and easily extracted trans-orally. In summary, an increasing number of lap bands are being placed to treat obesity. As the cumulative numbers of these patients rises, there is an increasing need for endoscopic evaluation. Successful endoscopic management of these complications requires that the endoscopist is familiar with the various tools and approaches that are required for a successful outcome. In this chapter, we will discuss endoscopic retrograde cholangiopancreatography or ERCP following Roux-en-Y gastric bypass surgery. The indications for an ERCP in patients following bariatric surgery are similar to that encountered in patients with normal anatomy. These procedures, however, are far more technically challenging in view of the altered anatomy. There are three main modes of access to the biliary tree following Roux-en-Y gastric bypass. The trans-oral route, which may involve the use of a pediatric colonoscope or deep small bowel intubation technologies. Access via a gastrostomy tract, which could be placed surgically or endoscopically. And finally, access via a percutaneous route by an interventional radiologist. There are several technical challenges associated with this procedure. Access via a percutaneous route by an interventional radiologist. There are several technical challenges one encounters during ERCP via a per-oral route. Often the length of the roux limb requires the use of deep small bowel intubation technologies such as a single balloon, double balloon, or spiral endoscopy. Once the end of the roux limb is reached, it is often difficult for the endoscopist to identify which orifice to intubate in order to traverse the biliopantreatic limb. Fluoroscopy can confirm that the endoscope is headed up towards the right upper quadrant and is in the biliopantreatic limb. Once the duodenum is entered, the tangential approach to the papilla is the next hurdle. This approach prevents easy cannulation of the bile duct or the pancreatic duct. The limited number of accessories that can be used with deep small bowel intubation technology or a pediatric colonoscope, as well as the absence of the elevator, further compound the difficulties that one encounters. In patients with a short limb Roux-en-Y gastric bypass, a pediatric colonoscope will suffice to reach the papilla often. Patients with a long limb Roux-en-Y gastric bypass, on the other hand, typically will require deep small bowel intubation technologies. Other modified techniques in order to reach the major papilla have also been described. For instance, a single balloon overtube or a spiral overtube is introduced by traditional deep small bowel intubation technique. Once the biliopantreatic limb is accessed, the endoscope is removed, leaving the overtube in place, and a shorter scope is advanced via the overtube, which may need to be trimmed. Occasionally, a pediatric colonoscope is long enough to reach the major papilla. However, the tangential approach prevents cannulation. In such cases, a wire can be left in the stomach. The pediatric colonoscope can be withdrawn. The wire is then backloaded onto a balloon passed through a side-viewing duodenoscope, and an inflated balloon is used to draw up the duodenoscope into the duodenum to enable better approach to cannulate the major papilla. Clearly, per-oral techniques for ERCP following Roux-en-Y gastric bypass are complicated. What are the advantages of such an approach? The main advantages are that compared to a surgical approach or a percutaneous approach, these procedures are less invasive. Additionally, if access to the major papilla is easy, this might allow for repeated procedures in the future. We will now review some endoscopic footage of ERCP procedures following Roux-en-Y gastric bypass completed via a per-oral route. This is a patient who is undergoing an ERCP following Roux-en-Y gastric bypass for a residual stone in the bile duct. This patient has previously undergone a cholecystectomy. Deep small bowel intubation technology such as Spiros or single balloon is being used for this procedure. The endoscope is advanced beyond the gastrojejunostomy. The roux limb is traversed and the end of the roux limb is reached. The biliopancreatic limb is intubated and the scope is advanced up to the major papilla. The initial attempts at cannulation result in access to the biliary tree. A stent is placed into the bile duct and a needle-knife sphincterotomy is performed over the biliary stent. Note the direction of the sphincterotomy towards the bile duct. Note the direction of the sphincterotomy towards the excluded stomach. Cut is being made in the intra-duodenal portion of the bile duct similar to a traditional sphincterotomy. Once an adequate biliary sphincterotomy has been performed, a wire is advanced into the biliary tree and balloon dilation of the papillary orifice and distal bile duct is performed. Following balloon dilation, the large stone is easily extracted from the biliary tree. In this next patient who has also previously undergone a Roux-en-Y gastric bypass surgery, single balloon technology is being used to access the biliopancreatic limb and the major papilla. The use of fluoroscopy enables the endoscopist to confirm that the endoscope is in the biliopancreatic limb and moving up towards the right upper quadrant. A straight catheter with a wire is then used to cannulate the biliary tree. In this particular patient, a stricture in the distal bile duct related to chronic pancreatitis is seen. The wire is left in place and the catheter is withdrawn. Next, a stent is advanced over the wire into the biliary tree. Once the stent is in place, a needle-nose sphincterotomy is performed over the biliary stent. In this patient, the stent was left in place in order to enable a discussion with the surgeons about further management options. In this next patient, access to the duodenum and the major papilla was achieved with an entroscope. However, the tangential approach to the major papilla resulted only in impacted cannulations and no deep cannulation. A wire was therefore left in the excluded stomach and the entroscope was gradually withdrawn. This was done in order to attempt to advance a side-viewer into the duodenum to enable an on-fuss approach to the major papilla. The wire was then back-loaded onto a balloon catheter in a side-viewing duodenoscope. Once the endoscope was in the rule limb, the balloon catheter was advanced over the wire all the way into the stomach. The balloon was then inflated to 20 mm under fluoroscopic visualization, in the stomach. Using the inflated balloon as an anchor, the duodenoscope is advanced into the duodenum and the major papilla is deeply cannulated. This modified technique allowed us to gain on-fuss access to the major papilla. Additionally, the use of the elevator and the sphincterotomy also enabled easy access into the papilla. Because there was a concern of the possibility of residual stones in the bile duct in this patient, a biliary sphincterotomy was performed with the sphincterotome, followed by balloon dilation and sweep of the bile duct. As mentioned previously, success rates with these procedures are lower than traditional ERCP. A recent multicenter study looked at 156 ERCPs in 129 patients, which included 64 patients in the ER. The papilla was not reached in 23 patients, and cannulation failure was noted in 11 patients. The interventions performed included biliary sphincterotomy in 41 patients, of which 21 were performed with a needle-nosed sphincterotome. The mean procedure time was 110 minutes. Of the patients who had a Roux-en-Y gastric bypass, spiral endoscopy was successful in reaching the papilla in 73.1%. Single balloon endoscopy was successful in 72.7%, and double balloon endoscopy in 86.7% of the patients. ERCP success was 61.5%, 59.1%, and 66.7% respectively. It was interesting to find that the choice of technology did not significantly impact the success of ERCP following Roux-en-Y gastric bypass between these three modalities. Accessing the major papilla with a forward-viewing endoscope is technically very challenging. As such, it is not surprising that the rates of success are lower in Roux-en-Y gastric bypass patients as compared to patients who have a Roux-en-Y hepaticogenostomy. A recent study looking at the single balloon system for ERCP following Roux-en-Y gastric bypass found that the procedure was successful in 47% of the patients following a Roux-en-Y gastric bypass with a native papilla as compared to 78% of patients with a Roux-en-Y hepaticogenostomy. Pre-procedure planning is extremely important. Pre-procedure planning is extremely important for the success of ERCP procedures following Roux-en-Y gastric bypass. It is very useful to review the indications for the procedure as well as operative notes and prior imaging studies. The choice of the endoscope system also needs to be made very carefully. Patients who have a short limb Roux-en-Y gastric bypass may do well with a pediatric colonoscope. On the other hand, a long limb Roux-en-Y gastric bypass may require spiral endoscopy, single balloon endoscopy or double balloon endoscopy. Depending on the choice of endoscope, it is equally important to make sure that endotherapy tools of appropriate caliber and length are available. The pediatric colonoscope and balloon endoscopes permit seven French devices only. Finally, these procedures do take a fair amount of time and it is important to schedule adequate time for these procedures. The typical side-viewing duodenoscope has a working length of 124 cm and a channel diameter of 4.2 mm. In comparison, the pediatric colonoscope is 168 cm long and has a channel diameter of 3.2 mm. The single balloon endoscope and the double balloon endoscope both measure 200 cm long and the channel diameter can vary between 2.2 or 2.8 mm. It is important to keep these limitations in mind in comparison to that of the standard duodenoscope prior to embarking on an ERCP with these systems. In order to make it easier to perform future ERCP procedures in these patients, some endoscopists tattoo the biliopancreatic limb at its entrance. It is also useful in the future if the endoscopy report details distances traversed, the scope system used, and the type of anatomy encountered. Additionally, if the patient has to be in a particular position for the endoscope system to more easily advance to the duodenum, this should also be noted. If local expertise for deep small bowel intubation is not available or if deep small bowel intubation is unable to reach the major papilla, either the surgical options or a percutaneous approach with the assistance of an interventional radiologist should be considered. If access is available to the excluded stomach via a gastrostomy tract, then a side-viewing duodenoscope and standard accessories can be used to accomplish an ERCP. An ERCP with the assistance of a surgeon can be performed either intra-operatively or by placement of a G-tube into the excluded stomach, allowing the tract to mature for six weeks and then subsequently performing an ERCP via the tract. Additionally, if a pediatric colonoscope or a deep small bowel intubation technology is able to reach the excluded stomach, an endoscopic gastrostomy can be performed with subsequent dilation of the tract and ERCP via the tract. As seen in this video, the gastrostomy tract in an obese individual is very, very thick and simple balloon dilation is usually not sufficient to dilate the tract. In our institution, once the tract has matured for about six weeks, the G-tube is removed. An endoscope is advanced through the gastrostomy tract into the stomach and into the small bowel. A sabry wire or a super-stiff wire is placed through the endoscope into the small bowel. The endoscope is then removed and sabry dilation to 42 French is performed prior to introduction of the side-viewing duodenoscope for the ERCP. The ERCP can then be performed by standard techniques. Getting the tip of the endoscope close to the major papilla with this kind of an approach is sometimes a little difficult. In this instance, the side-viewing duodenoscope is advanced into the small bowel via the dilated gastrostomy tract. The bile duct is cannulated with a sphingotome. Note that the scope positioning is very similar to that encountered during normal ERCP. An alternate approach is to perform the ERCP at the time that the initial gastrostomy is performed. This essentially is a laparoscopic assist ERCP. The advantage of this approach is that you do not have to wait six weeks for the gastrostomy tract to mature prior to proceeding with an ERCP. During a laparoscopic assisted ERCP, the duodenoscope is inserted via a 15 millimeter trocar into the stomach. All standard ERCP techniques can be performed by this approach. Leaving a gastrostomy tube at the end of the procedure allows short-term access for stent removal or the management of any complications. Additionally, the tract can be used for future procedures if repeated access is desired. An added advantage of a laparoscopic assisted ERCP is the occasional occurrence of internal hernias which can be identified and repaired at the time of the ERCP. The third and final approach that is available for ERCP following rheumatoid gastric bypass is a percutaneous approach with the assistance of the interventional radiologist. In contrast to surgery, this is far less invasive and typically the interventional radiologist can manage most biliary issues via the percutaneous access route. This approach might require multiple sessions and is hampered by the inability to perform a biliary sphincterotomy or undertake any pancreatic endotherapy. Ultimately, several factors play a role in the choice of method for performing an ERCP following rheumatoid gastric bypass. The length of the roux limb, the need for repeated procedures and maintaining continued access into the excluded stomach, as well as local expertise, play a role in this decision tree. We will now briefly discuss endoscopic ultrasound following rheumatoid gastric bypass. There is a very limited amount of literature and published experience regarding EUS in this post-surgical setting. In a recent series of a very limited number of patients, the pancreatic body and tail could be visualized from the gastric pouch following rheum-Y gastric bypass. The authors felt that fine needle aspiration of the pancreatic body and the tail was probably feasible from the gastric pouch if it was indicated. The head of the pancreas and the bile that could not be imaged from the gastric pouch in any of these patients. In summary, ERCP following roux-en-Y gastric bypass can be technically very, very challenging. The difficulties relate both to reaching the papilla and then the subsequent cannulation with a forward-viewing endoscope. There are significant failure rates for endoscopic approaches. It is hoped that technologic improvements in the near future will improve endoscopic success. Access via a surgically created gastrostomy or percutaneous radiologic approaches are fallback options. Occasionally, following roux-en-Y gastric bypass, some people stop losing weight and in fact begin regaining weight. The etiology of this weight regain following gastric bypass is believed to be multifactorial. Amongst the many factors, dilation of the gastric pouch and dilation of the gastrointestinal anastomosis or stoma are believed to be responsible for weight regain in at least a small number of individuals. Surgical revision can be challenging and carries a significant risk of complications. As such, it is natural for endoscopists to explore the possibility that the stoma and the pouch can be modified to reverse the weight regain that occurs sometimes several years following roux-en-Y gastric bypass. These efforts have revolved around the placement of tissue anchors to create tissue plications and reduce pouch size as well as stoma size, or the injection of sodium muruate in the region of gastrojejunostomy to reduce stoma size, or endoscopic suturing for stoma reduction. While the short-term results following such endoscopic manipulation have been encouraging, it remains to be seen if the results following such endoscopic revisions are durable in the long term. In this chapter, we will discuss endoscopic treatment of obesity. There are several devices in various stages of research and development. None of these devices are as yet approved by the FDA for the endoscopic treatment of obesity. Over the last decade, there have been tremendous technologic advances within the field of endoscopy. With the growing epidemic of obesity in the country, it is but natural to see if any of these endoscopic advances can be applied towards a primary therapy of obesity. Surgical treatment of obesity has a long history and there are several lessons that an endoscopist can pick up from the surgical experience. To begin with, it is important for us to understand the basic mechanisms by which bariatric surgical procedures bring about weight loss. Bariatric surgical procedures cause weight loss by either intestinal malabsorption or restriction or a combination of the two. Intestinal malabsorption can be brought about by diverting enteral nutrients around most of the small bowel. This can be achieved by anastomosing the proximal jejunum close to the ileocecal valve. Another way to promote intestinal malabsorption is to keep bile and pancreatic juices separate from the food for extended lengths of the small bowel and allowing admixture close to the ileocecal valve. The restrictive procedures work by a reduction in stomach volume and possibly the suppression of rel-insecretion. There are probably other neurohumoral factors that play a role in weight loss following bariatric surgery that are not yet well understood. This slide gives us a good idea of the kind of weight loss one can expect following different bariatric surgical procedures. To date, the gastric bypass procedure remains the gold standard. The vertical banded gastroplasty and banding do not achieve as much weight loss as gastric bypass, but banding is nevertheless a well-accepted surgical alternative at this point in time. Endoscopic therapies hoping to compete in this market need to have an efficacy similar to that of laparoscopic banding. Following laparoscopic banding, the weight loss at two years is about 21 to 22 percent. Beyond the two-year time point in all of these surgical procedures, there appears to be a mild loss of effectiveness, but in the long run, the weight loss trends do seem to plateau. For an endoscopic therapy to be deemed acceptable and successful, several factors will play a role. This will include ease of performing the procedure, cost of the procedure, safety of the procedure, and finally long-term durability of the procedure. There are a few potential roles for endoscopy in the management of obesity. First is as a primary therapy for obesity. Second is a reduction of the impact of comorbidities such as diabetes, even if the weight loss is marginal. Third, as a bridge to definitive surgical therapy. And finally, for revision of failed bariatric surgical procedures. There are several devices currently in various stages of clinical research and development. In this chapter, we will discuss intragastric balloons, the transoral gastroplasty procedure, the duodenal jejunal bypass sleeve, endoluminal vertical gastroplasty, and the transoral endoscopic restrictive implant system. The amount of published data and literature that is available regarding each of these procedures is variable. None of these procedures, as mentioned previously, are approved by the FDA for use in the United States. We will first discuss intragastric balloons as a means of bringing about weight loss in the bariatric patient. There are three types of intragastric balloons which are important from a historical perspective. The first is the Garen Edwards gastric bubble. This was a polyurethane cylindrical device inflated with 220 cc's of air. Over 25,000 of these balloons have been deployed worldwide. They were removed from the American market in 1987. The second is the Taylor intragastric balloon. This was a pear-shaped smooth silicone balloon inflated with 550 cc's of saline. And finally, there's a below this intragastric balloon which was inflated with 475 cc's of air. These balloons have either not been effective or there have been concerns for complications in the past and hence they are currently not used in clinical practice. The intragastric balloons which are currently in use outside the United States include the Bioentrix intragastric balloon, the Heliosphere bag, and the Endogast system. The Bioentrix intragastric balloon system has been available for over nine years. It consists of a transparent silicone elastomer filled with 400 to 700 cc's of sterile saline. The maximum period of placement for these balloons is six months. The Heliosphere bag is a similar balloon inflated with 900 cc's of air. The Endogast balloon, on the other hand, has a subcutaneous port that is connected to an intragastric balloon. As such, it requires a combined laparoscopic endoscopic insertion. The subcutaneous port in the Endogast system allows for inflation or deflation of the intragastric balloon without need for additional invasive procedures. This video demonstrating the placement of the Bioentrix intragastric balloon was provided by Dr. Nagesh Reddy of the Asian Institute of Gastroenterology in Hyderabad, India. Once the intragastric balloon is introduced into the stomach, under endoscopic visualization, the restraining string is removed and about 400 to 700 cc's of saline are injected into the balloon. Once the balloon is fully inflated, the catheter is easily detached from the balloon, leaving the intragastric balloon in place. Here is an endoscopic view of the intragastric balloon upon full inflation. This is a lady who presented to our institution for removal of an intragastric balloon that was placed abroad. The balloon was punctured with an endoscopic insertion. The endoscopic ultrasound needle and it was aspirated completely. Following aspiration, the balloon was grasped firmly with a rat tooth forceps and extracted with minor difficulty across the lower esophageal junction via a transoral route. We will now review some of the data regarding the effectiveness and safety of intragastric balloons. This is a trial that was published in 2006 which was sham controlled. Patients with a BMI of 40 to 45 were enrolled and they were randomized to group A or group B. Patients in group A received intragastric balloon placement for a period of three months, followed by balloon removal and crossover into the sham group. The sham group on the other hand did not receive the intragastric balloon for the first three months but then underwent the balloon placement and subsequent removal three months later. The results from the study conclusively demonstrate the short-term effectiveness of the intragastric balloon in bringing about weight loss in the obese patient. In both arms, when the intragastric balloon was in place, striking weight loss was achieved and when the intragastric balloon was not in place, the weight did not change significantly. Based on the results from their study, the authors concluded that bioentrics intragastric balloon placement is safe and feasible. There was no mortality or significant complications related to balloon placement. There were no erosions or ulcers since these patients were largely on proton pump inhibitors. The authors felt that the intragastric balloon was effective as an adjunct device for weight reduction and the effect was not a placebo effect. They felt that the temporary weight reduction might be useful in the pre-operative treatment in preparation for bariatric or other surgical procedures. This study, which was published earlier in 2005 in gastrointestinal endoscopy, has a slightly different study design. The first three months of the study involved two groups of patients. Patients were either randomized to a balloon arm or a sham arm. As you can see in the study, there was no significant difference between the two arms, whether you look at BMI or at weight loss at the three-month point. After three months, the sham controls were crossed over into the balloon group and in all patients, the balloons were removed at 12 months. As is evident in this graph, while patients did lose weight while the intragastric balloon was in place, a significant portion of this weight was regained during the additional 12-month follow-up period following balloon removal. The heterogeneity of the sham-controlled balloon trials probably relates to differences in study design and controls. Another study looked at the ability of the intragastric balloon to reduce liver volume in the super obese patient prior to Roux-en-Y gastric bypass. In this study, 31 patients were enrolled with a BMI of 55.2 plus or minus 6.9. These patients underwent placement of the intragastric balloon and underwent a Roux-en-Y laparoscopic gastric bypass six months later. Liver volume was measured before intragastric balloon placement and prior to Roux-en-Y gastric bypass. In previous studies, decreased liver volume has been shown with a low energy diet in the preoperative setting. During the six months following intragastric balloon placement, the BMI decreased from 55.2 to 47.4 and the mean weight decreased from 149.3 kilograms to 128 kilograms. The mean liver volume was also significantly reduced over the six-month period. A reduction in liver volume can render laparoscopic Roux-en-Y gastric bypass surgery easier. A review of the complications of intragastric balloon placement is important because these balloons have been pulled out of the market previously in the United States for fear of complications. A meta-analysis was published in 2008 which looked at 4,240 patients across 20 studies. There were three deaths out of this population due to gastric perforation. Two of them were in patients who had nissen fundoplication previously and one was a death following aspiration. There were nine perforations in the study group and five of these patients had previously had gastric surgery. Bowel obstruction requiring surgical endoscopic or combined removal of the intragastric balloon was seen in seven patients. Intolerance led to early removal in about 2.43% or 103 patients. Some patients also experienced abdominal pain, spontaneous deflation and gastroduonal ulcers. This data suggests that in patients without prior gastric or gastroesophageal junction surgery, the placement of an intragastric balloon for six months is safe. In contrast to the sham-controlled randomized trials that we spoke about earlier, there is an abundance of single-arm studies using the intragastric balloon which have shown positive results with this technology. This is one study that was published in gastrointestinal endoscopy in 2010 which looked at weight, insulin resistance and steatosis following the placement of an intragastric balloon. This prospective trial enrolled 130 patients who met NIH criteria for treatment for obesity. In these patients, following a six-month period of intragastric balloon placement, patients were followed up for a mean of 21 months. Patients were put on a 1,000 to 1,200 kilocalorie diet a day during the study and they were maintained on regular follow-up program after removal of the intragastric balloons. This chart shows the BMI distribution of the patients at baseline in the dark bars and the BMI distribution at the time of intragastric balloon removal in the gray colored bars. At baseline, 23% of the patients had a BMI of 50 or higher and 43% of the patients had a BMI that ranged between 40 and 49.9. In contrast, at the time of BIB removal, 7% of the patients had a BMI that was 50 or higher. 28% of the patients had a BMI that ranged from 40 to 49.9. What is more striking is that 23% of the patients now had a BMI that ranged between 30 and 34.9 and 19% of the patients had a BMI that ranged between 25 and 29.9. Besides the significant changes in BMI in these patients, significant improvements were also noted in glycemia, insulinemia, triglyceride levels, ALT levels, gamma-GT levels, as well as steatosis. Finally, likely as a consequence of the regular follow-up program that these patients were enrolled in following the removal of the intragastric balloon, after a median 22-month follow-up period, 50% of the responders maintained or continued to lose weight. Given the fact that single-arm studies have consistently shown that the intragastric balloon is efficacious in patients with obesity, it is not surprising that there are several proponents of this technology who have continued to develop further iterations of this device. One such device that was recently introduced is the adjustable intragastric balloon, which has been approved outside the United States for a one-year implantation period. Pilot study data with the use of the adjustable intragastric balloon was presented at DDW in 2011. The device was successfully implanted in 18 patients. If additional weight loss was desired, the balloon was further inflated, and on the other hand, if patients tolerated the balloon poorly, the balloon was marginally deflated. At one year, the percentage excess weight loss was 67.3%, and the weight loss was 35.5 kilograms. Another modification of the intragastric balloon is seen in this slide. This device consists of two balloons joined together by a flexible shaft with a capacity of 900 CCs. Clinical data with this device are currently awaited. In summary, intragastric balloons appear to be effective in promoting short-term weight loss. They are safe if contraindications are observed. There is also a short-term improvement in comorbidities. There has been a recent resurgence of interest along with other newer endoscopic devices, and currently we are awaiting results from newly modified intragastric balloons. In this chapter, we will discuss the transoral gastroplasty, or the TOGA procedure. Of the procedures that are being evaluated for primary therapy of obesity, the transoral gastroplasty, or the TOGA procedure, is the farthest along with regard to clinical studies. The TOGA device consists essentially of two components. The first is the sleeve stapler, which you see on this slide here. The sleeve stapler is a 54 French diameter device, and it allows the passage of an 8.6 millimeter endoscope through its channel. This endoscope allows direct visualization during key procedure steps. The sleeve stapler creates a stapled sleeve along the lesser curvature of the stomach. The second device is the TOGA restrictor, which is 45 French in diameter, and is delivered alongside an endoscope. The restrictor creates stapled pleats at the distal end of the sleeve, restricting outflow. The objective of the TOGA procedure is to perform a gastric stapling by the endoluminal route, with the use of vacuum-based tissue acquisition. This procedure uses titanium staples, similar to that used in surgery, and the goal of the TOGA procedure is to eliminate the key risks of surgery, such as leaks, pulmonary and cardiovascular adverse events, wound infections, hernia, and banned migration. Following creation of the sleeves, the staple line runs from the cardio along the lesser curvature down towards the antrum. Depending on the anatomy of the individual patient, the distal end of the sleeve sometimes may be very close to the angularis. This is a photograph of an explanted stomach from a cadaver in which the TOGA procedure has been performed. You can clearly see the staple line running along the lesser curvature, and the staple line appears to be robust in that it is trans-serosal. We will now review a TOGA procedure being performed. Following esophageal dilation, the sleeve stapler is advanced over a savoury wire into the stomach. The endoscope is advanced through the channel in the sleeve stapler to provide visualization. Under endoscopic visualization, the sleeve stapler pods are opened up. The sale septum and the retractor wire are then deployed. After ensuring that the staple line is going to start at or slightly above the cardio without any gaps, suction is applied. The sale is pushed out of the way and the staples are fired. Subsequently, the device is opened up and removed. A new cartridge is placed and the device is reintroduced over a savoury wire for the placement of the second sleeve immediately distal to the first sleeve. The restrictor is then advanced over a savoury wire and an endoscope is advanced alongside to provide visualization. Restrictions are placed in the distal end of the sleeve at two separate locations. Inspection of the sleeve at the end of the procedure shows that the sleeve outlet is fairly snug around the end of the 5.9 millimeter endoscope. This is an anti-grade view of the sleeve. This radiograph shows a barium study at three months following a toga procedure. You can see the contrast coming down the esophagus passing through the gastric sleeve prior to entering into the rest of the stomach. This is the endoscopic appearance of the intact toga sleeve three months following a toga procedure. This is a retroflexed view showing the distal end of the sleeve. In the distal end of the sleeve are also visible the two restrictions that were placed towards the end of the procedure. The anti-grade view shows the two restrictions in the 3 o'clock and 7 o'clock position and the sleeve staple line in the 10 to 11 o'clock position. This patient has a small gap in the mid portion of the sleeve as is visible on the endoscopic footage. Often the gaps that are seen following toga procedures can be fairly large adhesions of the staple lines. While the pilot procedures in Europe did not reveal any significant loss of effectiveness of the toga procedure in the presence of a gap, the size and the number of gaps that are seen in fallout patients does raise concern about the long-term durability of the effectiveness of the toga procedure in bringing about sustained weight loss. This is data from the pilot study of the toga procedure presented by Dr. Jacques Devere at the IFSO conference in 2009. In this pilot report, 41 patients received their procedures either in Brussels, Mexico City or Rome. The average age was 45 years and the average BMI was 42.7. The average procedure time was about an hour and 33 minutes. This is not surprising for procedures that are still under development. The average anesthesia time was 1 hour 54 minutes. In these 41 patients, 28 double sleeves, 12 triple sleeves and one slingled sleeve were placed. The staple line gaps were present in 11 of 24 patients at six months and the average size of these gaps was 1.8 centimeters. Six patients received additional restrictions on account of insufficient weight loss. Over the course of 12 months, in the 18 patients who had completed one year of follow-up, the BMI dropped from an average of 43 to 35. At the 12-month time point, the percentage of excess BMI loss was 52% and the percentage of excess weight loss was 46%. As shown on this graph, the presence or absence of gaps following the toga procedure did not impact excess weight loss at 12 months. In follow-up, on quality of life questionnaires, these patients also reported significant improvements in physical function, self-esteem, sexual life, public distress and work. In these pilot studies, there were no serious adverse events. 40 of the 41 patients were discharged the day following the procedure. The most common adverse events noted was abdominal pain in 28 patients, nausea in 10 patients, gastric ulcers in 8 patients, dysphagia in 7 patients, sore throat in 6 patients and vomiting in 5 patients. A pivotal study evaluating the effectiveness and safety of the toga procedure has been completed. At the time of this recording, the results of the study are not available for public dissemination. The study design was very robust. It was a prospective, randomized, multicenter, sham-controlled study. The study enrolled patients who were surgical candidates for bariatric surgery treatments. The BMI had to be between 40 and 55 or between 35 and 40 with comorbidities. 275 patients at 9 U.S. sites and 1 international site were enrolled in the study. The randomization scheme was 2 is to 1 with 2 procedures for every 1 sham patient. Crossover was offered to the sham patients at 12 months. The primary endpoint was percentage excess weight loss at 1 year. The other endpoints that were determined in the study include comorbidity improvement, BMI change and quality of life scores. In summary, the pilot studies have shown that the transoral gastroplasty procedure is feasible, safe, effective and improves quality of life in the short term. These patients have a short hospital stay and there are fewer complications compared to traditional surgeries. We are currently awaiting the results from the U.S. multicenter study. Long-term durability of the results will be key to the success of this device. Prior to the publication of the results of the TOGA pivotal trial, all assets and intellectual property of the company initially developing the TOGA device, Satiety, were acquired by another company. At this time, it is not known if there are plans to develop the TOGA device further for the primary therapy of obesity. In this segment, we will discuss the duodenal jejunal bypass sleeve. The device consists of a nitinol anchor and a liner. The anchor is made of nitinol and has barbs on it. This allows the anchor to lodge in the duodenal bulb and reduce the risks of migration. Attached to this nitinol anchor is a retrieval drawstring which enables removal of the device. Attached to the anchor is a two-foot long impermeable fluoropolymer liner. This fluoropolymer liner prevents admixture of pancreatic and biliary juices with food following ingestion. The sleeve is implanted endoscopically under fluoroscopic guidance. This animation showing the deployment of the duodenal jejunal bypass sleeve was provided by GI Dynamics, the company that is developing the device for clinical use. A guide wire is advanced through the upper endoscope into the small bowel and the endoscope is then removed. A delivery capsule is then threaded over the guide wire and placed into the duodenal bulb, straddling the pylorus. The guide wire is removed and the upper endoscope is reintroduced to provide endoscopic visualization. Under fluoroscopic guidance, a coaxial catheter is used to deploy the sleeve two feet into the small bowel. The atraumatic ball tip is released into the small bowel. Finally, the nitinol anchor is deployed within the duodenal bulb and contrast is flushed to the device to assure patency of the entire length of the sleeve. The endoscope and the delivery system can then be removed. When it is time for removal of the duodenal jejunal bypass sleeve, a retrieval hood is attached to the tip of the endoscope and the endoscope is advanced into the duodenal bulb. Once in the bulb, the duodenal jejunal bypass sleeve and its anchor are visualized. A grasper is used to grasp one of the drawstrings in the nitinol anchor. The drawstring is pulled and this results in collapse of the nitinol anchor basket. This is then drawn into the retrieval hood and under fluoroscopic and endoscopic visualization with gentle traction, the sleeve and the anchor are removed in an atraumatic manner. In this procedure being performed by Dr. Sakai and Galvao in Brazil, the duodenal jejunal bypass liner device is advanced over a guide wire that has been placed into the small bowel. Under fluoroscopic and endoscopic guidance, the sleeve is gradually deployed. Once the sleeve has been fully deployed, the next step is to release the nitinol anchor cage in the duodenal bulb. This is an endoscopic view of the duodenal bulb following complete placement of the liner. The initial human trials with the duodenal jejunal bypass sleeve were conducted in Chile. The trial was an open-label, prospective, randomized control trial with duodenal jejunal bypass sleeve versus a low-calorie diet. The sleeve was left in place for 12 weeks. Twenty-five patients received the duodenal jejunal bypass sleeve and there were 14 controls. The last 15 consecutive patients received the duodenal jejunal bypass sleeve. The average BMI was 42 in the device arm and 40 in the control group. The primary endpoint sought was a difference in the percentage excess weight loss. Primary endpoints included the reduction in HbA1c of 0.5% or a reduction or elimination of diabetic medications. An additional secondary endpoint that was evaluated was the percentage of patients who lost greater than 10% of excess weight. Eighty percent of the patients maintained the duodenal jejunal bypass sleeve for 12 weeks without any adverse events. Upper GI bleeding was seen in three patients. The bleeding was seen at a mean of 13.8 days and did not require any blood transfusions. The anchor migrated in one patient on day 47. Sleeve obstruction was noted in one patient who presented with nausea and vomiting on day 30. The mean weight loss at 12 weeks was 25% for the device and 5% for the control group. Subsequently, a 12-week open-label, randomized trial was conducted to investigate the use of the duodenal jejunal bypass sleeve versus a sham operandoscopy for weight loss before bariatric surgery. Patients who received the sleeve were explanted at 12 weeks. Sham-treated subjects were unblinded at a 12-week visit and exited the study. Sham-treated subjects underwent general anesthesia for implantation and removal of the device. On the other hand, sham-treated subjects underwent conscious sedation during which an EGD and mock procedure was performed. Altogether 69 patients were screened and 58 patients were enrolled. Fifty-six of these patients were randomized. Twenty-seven patients were randomized to duodenal jejunal bypass sleeve group. Two of the patients withdrew prior to treatment and in four of the 27 patients, implantation was unsuccessful. In the sham group, there were 29 patients, of whom three withdrew prior to treatment. In the completer population at week 12, which included 13 of the duodenal jejunal bypass sleeve and 24 sham patients, excess weight loss was 11.9% in the sleeve group as compared to 2.7% in the sham arm. In the sleeve group, 62% achieved an at least 10% excess weight loss at 12 weeks compared to 17% in the sham arm. Total body weight change at week 12 in the sleeve group was minus 8.2 kg compared with minus 2 kg in the sham arm. All subjects in the sleeve arm lost weight. In contrast, six subjects in the sham arm gained weight during the course of the study. The most common adverse events noted in this trial in patients who received the sleeve included upper abdominal pain, procedural nausea, nausea and vomiting. The majority of these adverse events were deemed to be mild or moderate. There were no signs or symptoms of biliary obstruction, pancreatic duct obstruction or migration of the device in any subject. There were no clinically significant abnormal blood values observed during the study with the exception of three subjects who presented with a decrease in hemoglobin and hematocrit associated with GI bleeding. The three subjects with hematomasis presented with their complication at 11, 25 and 43 days post implantation. The devices were removed endoscopically with no subsequent sequelae in two patients. In one patient, the source of bleeding was identified and successfully treated with sclerotherapy and endoscopic clips. Two of these subjects required transfusions. Four patients who received the duodenal jejunal bypass sleeve discontinued early because of abdominal pain, nausea and vomiting on days 3, 9, 30 and 36. All the removed devices appeared normal and there was no evidence of gastritis, esophagitis or ulcerations in these patients. These symptoms resolved without further treatment or clinical sequelae. Given the initial successes with the device, the device has been modified and currently we are beginning to see longer term results with the duodenal jejunal bypass liner. In these results presented in 2010, preoperative weight loss was determined in a non-randomized single arm study in Chile. The planned duration of the study was 52 weeks. There were a total of 39 subjects with a BMI ranging from 30 to 65. Following placement of the liner, patients were put on a liquid diet for a week followed by a bariatric standard of care diet. During this study, it took a mean of 23.7 minutes to implant the liner and 16.4 minutes to explant the liner. Thanks to modification of the device, in this trial, the mean implantation time was slightly less than 40 weeks. Over the 52 week study period, the BMI decreased from a mean of 45.3 to 36 and 46.3% excess weight loss was achieved in 24 patients. In the subset of 6 patients with diabetes, there were improvements in the HbA1c noted over the 52 week period. Besides a beneficial impact on weight and BMI, significant improvements were also seen in blood pressure, total cholesterol, LDL, and triglyceride levels. In summary, in the short term, the duodenal jejunal bypass sleeve or liner is successful in reducing BMI and has a beneficial impact on diabetes. More longer term studies are needed if this device is to gain acceptance as an endoscopic treatment for obesity and diabetes. Additionally, there may be a role for the duodenal jejunal bypass liner in the pre-operative period preceding bariatric surgery. In this chapter, we will discuss endoluminal vertical gastroplasty. During this procedure, an endoscopic sewing machine is used to suture the anterior and posterior walls of the stomach together. This effectively reduces the compliance and distensibility of the stomach in response to ingested food. On the left side, you can see the native stomach prior to endoscopic intervention and on the right side, you can see the appearance of the stomach following endoluminal vertical gastroplasty. The white pledge that you see there on the screen is the cinching mechanism for the suture material used with the endoluminal vertical gastroplasty. The first clinical results from the endoluminal vertical gastroplasty relate to a study that was performed in Venezuela. Patients with BMIs of 28 to 44 were divided into three groups. Group 1 had a BMI greater than 40, the number of patients in this group was 33. Group 2 patients had a BMI of 35 to 40 and the number of patients in this group was 19 and group 3, the BMI was less than 35 and the number of patients was 12. A total of 64 patients were enrolled in the study. The procedures were performed on an outpatient basis and each procedure took approximately 45 minutes. Recovery time was less than two hours. The data published in 2008 looked at the 12-month percentage excess weight loss in each of these three groups. In the group with the BMI less than 35 to begin with, the percentage excess weight loss was over 80%. In the second group, where the BMI at the point of entry into the study ranged from 35 to 40, the percentage excess weight loss was 56.5%. In the third group, where the BMI was 40 at the beginning of the study, the percentage excess weight loss was 48.9%. Other trends in the BMI are also seen at 12 months in each of the three groups. The same group subsequently reported longer-term 24-month follow-up on a group of 233 patients who underwent endoluminal vertical gastroplasty. Of this group, 45 patients had 24-month follow-up and the percentage excess weight loss in these 45 patients was 49 plus or minus 28% at 24 months. The published experience with the endoluminal vertical gastroplasty procedure is very, very exciting. The data that has been published to date about this procedure shows that this procedure can be safely performed on an outpatient basis. Additionally, prolonged hospitalization is not required and recovery periods are short. Finally, the published techniques suggest that the endoluminal vertical gastroplasty procedure can be repeated again in the future in the same patient if there is dehiscence of the suture lines. We eagerly await further clinical trials with sham controls. For endoluminal vertical gastroplasty, besides initial efficacy, the ability to repeat this procedure again at a later date as well as the longer-term durability of the results are extremely important. We will now discuss one of the newest endoluminal technologies that have been used in an attempt to treat patients with obesity. This is the transoral endoscopic restrictive implant system. The TERO system consists of a 66-fringe overtube which can accommodate a 5-millimeter endoscope. Also advanced with this overtube is an articulating endoscopic circular stapler. With this circular stapler, five gastric placations can be created three centimeters below the gastroesophageal junction. Each of these placations is made with the use of two concentric rings of 3.5-millimeter staples reinforced with a plastic ring. In the placation, a placation hole is created. Subsequently, silicone anchors are pulled through the placation hole. Next, a multilumen guide is used. This multilumen guide has a central lumen for a 5-millimeter endoscope. Additionally, there are five lumens around it for the locking anchor graspers. There is one locking anchor grasper for each silicone anchor. The gastric restrictor is then delivered and attached to these anchors via the multilumen guide. This animation of the TEROS procedure was provided by Dr. Paul Falkens of the Academic Medical Center in Amsterdam. Through the 66 French overtube, the 5-millimeter endoscope and the articulating endoscopic stapler have been introduced into the stomach. The endoscopic circular stapler is used to create placations three centimeters below the GE junction. The tissue is drawn into the stapler, and two concentric rings of 3.5-millimeter staples reinforced with a plastic ring are used to create the placation. Next, a placation hole is created, and the silicone anchor is pulled through the placation hole. The silicone anchor is then released. The circular stapler and endoscope are withdrawn, leaving the overtube in place. The cartridge is reloaded into the stapler, and the stapler and endoscope are reintroduced. Five gastric placations, each with a silicone anchor, is created circumferentially three centimeters below the GE junction. This is a cross-sectional view at the point that all five silicone anchors have been deployed. Next, the multilumen guide, which will accommodate both the endoscope as well as the locking anchor graspers, is introduced. A locking anchor is advanced and is used to grasp a silicone anchor. This is repeated for each of the other four silicone anchors. Finally, the gastric restrictor is attached and delivered to the anchors along the locking anchor graspers. The locking anchor graspers are then detached, and the device can be removed. This is the final animated appearance of the gastric restrictor in place, anchored to each of the silicone anchors that were applied to placations in the gastric wall about three centimeters below the GE junction. The pilot study in Netherlands for the TERAS trial enrolled 13 patients. Of the first seven patients, two patients were found to have a pneumoperitoneum. One patient was noted to have perforation after stapler mounting. One patient was noted to have perforation after stapler malfunction. The device was then modified, and then subsequently, in the next six patients, there were no adverse events. The mean procedure time was 142 minutes. Ten patients were discharged in one day, two patients in two days, and one patient in three days. The weight loss was 11.2% at three months, and the excess weight loss was 22.2% at three months. For many endoscopists, the ability to deliver such endoluminal therapies is indeed very exciting. Unfortunately, the TERAS system is currently not being developed further due to unacceptable durability of the lesser curvature anchor points, and alternative implant design and plication patterns are being studied. In summary, endoscopic treatment of obesity is clearly not yet ready for prime time. We are awaiting results from randomized, blinded, sham-controlled trials for each of these devices. The precise mechanisms for weight loss with endotherapy remain to be elucidated. The long-term durability and safety are important in addition to short-term effectiveness. The results, as well as the costs of endotherapy for bariatric goals, should be compared with existing surgical options. Similar to other bariatric interventions, endoscopic therapies, if validated, have to be delivered via a multidisciplinary weight management team rather than as a stand-alone endoscopic therapy for obesity.

endoscopists

post-bariatric surgery

obesity management

bariatric surgical procedures

altered anatomy

endoscopic evaluation

complications

Roux-en-Y gastric bypass

strictures

leaks

ulcers

fistulas

lap band system

ERCP

intragastric balloons