ABBREVIATIONS
CD
Crohn’s disease
CT
computed tomography
GI
gastrointestinal
EBD
endoscopic balloon dilation
ES
endoscopic stricturotomy
EVAC
endoluminal vacuum-assisted closure
IBD
inflammatory bowel disease
IPAA
ileal pouch-anal anastomosis
MRI
magnetic resonance imaging
RYGB
Roux-en-Y gastric bypass
UC
ulcerative colitis
QoL
quality of life
INTRODUCTION
Surgical techniques to divert the bowel may occur for a myriad of indications including emergent management of trauma, gangrenous bowel, distal gastrointestinal (GI) malignancy, bowel obstruction, inflammatory bowel disease (IBD), transplant surgery, and weight loss. These surgeries can provide significant benefits and potentially lifesaving outcomes, though postoperative complications can occur. The management of these complications was previously the exclusive domain of the surgical service, but newer endoscopic technologies and techniques have allowed the gastroenterologist to offer nonsurgical options. These may reduce the resulting morbidity and, in some cases, avoid corrective surgery.
Common post-diversion complications that can be either diagnosed or managed endoscopically include anastomotic leak, fistula, strictures, malignancy, IBD, and diversion colitis. The initial indication for diversion surgery may predispose the patient to complications and can be used to create a differential diagnosis. An obese or malnourished patient may be more likely to have an anastomotic leak. A history of Crohn’s disease (CD) increases the risk of strictures or fistula as part of the natural course of the disease. Rectal bleeding may be related to ulcerative colitis or diversion colitis, in the appropriate clinical setting.
Every gastroenterologist should feel empowered to help diagnose these conditions. The key elements are a combination of a thorough history, imaging tests such as computerized tomography (CT) or magnetic resonance imaging (MRI), and endoscopy. The latter allows direct visual examination of the tissue as well as histologic evaluation. Not every endoscopist may be qualified to manage some conditions, but timely referral can lead to improved morbidity and superior outcomes. A team of gastroenterologists, surgeons, radiologists, and pathologists can share information to aid in determining the most effective and safest therapy.
TYPES OF DIVERSION
The gastroenterologist should be familiar with the postsurgical anatomy before endoscopy to ensure a thorough, safe examination, while also being able to recognize pathology. Bowel diversion can be classified anatomically based on where the surgery is located. Gastric bypass, small bowel bypass, small bowel diversion, and colon diversion may all result in different postsurgical anatomy.
Common indications for gastric bypass include weight loss, distal obstruction, malignancy, pancreatic transplant, or trauma. Surgery may result in a decreased capacitance of the gastric reservoir and can be classified as restrictive, malabsorptive, or both. The most common techniques are the Roux-en-Y gastric bypass (RYGB), biliopancreatic diversion, and sleeve gastrectomy ( Fig. 23.1 ).

The ileum and colon may be bypassed by the creation of an ostomy. The ileostomy and colostomy allow stool to drain into a bag on the anterior abdominal wall that is either emptied or exchanged when full. An end ostomy is created when the bowel is transected and brought to the surface ( Fig. 23.2A ). A loop or diversion ostomy is created when the bowel is brought to the abdominal wall, an enterotomy is performed, and the bowel is fixed into afferent and efferent limbs ( Fig. 23.2B ). A mucous fistula is created when the defunctionalized bowel is brought to the abdominal wall ( Fig. 23.2C ). Finally, the Kock pouch, a continent ileal reservoir also known as a K-pouch, is created when the bowel is reformed into a reservoir that drains without an ostomy bag to the abdominal wall ( Fig. 23.3 ).





The gastroenterologist will rarely be asked to evaluate the middle of the surgically altered small bowel, but in cases such as an RYGB that necessitate an endoscopic retrograde cholangiopancreatography, knowledge of the anatomy may change the instruments needed to perform the procedure. A distally placed bypass anastomosis may result in a standard duodenoscope not being long enough to reach the ampulla. In that case, a pediatric colonoscope may be used in its place.
INDICATIONS FOR DIVERSION
Malignancy
The diverted bowel should still receive appropriate screening for common GI malignancies. In patients with IBD who underwent a colectomy with a Hartman’s pouch (rectal stump), 1.7% subsequently developed dysplasia and 1.5% of patients ultimately developed carcinoma. The rate of rectal cancer in IBD patients with an ileorectal anastomosis was 1.8%, and in those with an ileal pouch-anal anastomosis (IPAA) it was 0.06%. A previous history of colorectal cancer and ongoing rectal inflammation were considered risk factors for subsequent dysplasia. The diagnosis of metachronous cancer was made a median of 15 years after colectomy. Despite being uncommon, anorectal cancer following fecal diversion was often advanced when discovered and associated with a 5-year survival rate of only 20%.
In an asymptomatic patient who does not have a history of malignancy, dysplasia, or IBD, endoscopic screening should be performed as per standard societal guidelines but no longer than every 10 years between examinations. If there is a history of dysplasia as an indication for the diversion and bowel resection, then screening should begin 6 months after surgery. In patients with IBD, screening should be performed every 1 to 3 years. Surveillance endoscopy is associated with a lower rate of future colorectal cancer.
The goal of endoscopic assessment is to perform the best visual assessment possible given the circumstances. An oral bowel cleanse will not be possible due to the diversion. Rectal enemas may help purge the distal GI tract, but some patients may have colonic atrophy or diversion colitis that will make enemas uncomfortable and of limited benefit. In many cases, the stool burden cannot be removed, and the best attempts must be made to visualize the mucosa despite the circumstances.
High-definition white light endoscopy is the standard of care for most evaluations. If there is a history of dysplasia or IBD, then consider random biopsies for histologic examination in addition to direct visualization. In IBD there is a role for dye-based chromoendoscopy with targeted biopsies, or virtual chromoendoscopy with targeted biopsies instead of a white light examination. There is no data on the efficacy of these techniques in an IPAA or diverted bowel, and they are not recommended in this population at this time.
Due to the presence of diversion colitis, it is advised to examine the mucosa on insertion of the scope as well as withdrawal. In diversion colitis, there can be frank oozing of blood with insufflation, and visualization can be compromised such that examination on withdrawal is not feasible ( Fig. 23.4 ).
Malignancy in a diverted bowel is an uncommon complication that typically occurs years after a diversion surgery. It is important to be cognizant of its risk and to screen appropriately despite the limitations such as the inability to remove the stool burden, poor visibility, and a lack of data on the optimum techniques.
Luminal Strictures
The indication for bowel bypass surgery may affect the risk of subsequent luminal strictures. Strictures associated with complicated diverticulitis or medication sequalae may progress if there is an ongoing insult but can often stabilize over time. Inflammation and atrophy from disuse such as diversion colitis or relative ischemia of an anastomosis after surgery can lead to stricture formation. Disease progression of CD and malignancy may cause the strictures to become more severe over time. The cellular structure of the stricture can vary based on the etiology of the stricture. In CD, there is often an expansion of the muscularis mucosa with fibrosis, submucosal fibrosis, submucosal vascular changes, expansion and hypertrophy of the muscularis propria, and infiltration of several inflammatory cell lines. The varying degrees of fibrosis or muscular hypertrophy may affect if the stricture is rigid or elastic, respectively. It can also dictate which therapy is most effective in treating a stricture. Endoscopic balloon dilation (EBD) is more effective in a rigid fibrotic stricture than an elastic muscular hypertrophy stricture.
Strictures in the diverted bowel can be diagnosed by a combination of history, imaging, and endoscopy. Typical obstructive symptoms include nausea, abdominal distension, pain, and a general sensation that something is “stuck” in a particular area. Using imaging tests such as CT and MRI of the abdomen and pelvis can identify and localize the stricture. They can measure its length, determine if there is a curvature, identify associated pathology such as fistulas and abscesses, and identify if there is an inflammatory or fibrotic component of the stricture. This information can help determine if endoscopy or surgery would be the best initial therapy of the stricture. The techniques and role of imaging in classifying and grading the strictures are evolving, and endoscopic assessment may determine that the stricture is not as severe or long as imaging would suggest. Endoscopy is the gold standard for grading the severity of the stricture. Mild strictures allow the scope to pass easily. Moderate strictures allow the scope to pass with resistance. Severe strictures prevent the scope from passing. A limitation of endoscopy is that it may not be capable of visualizing across the stricture to determine the length, course, or extraluminal complications.
The presence of luminal strictures in most patients who are not diverted can be managed by a combination of dietary intervention, stool softeners, EBD, endoscopic stricturotomy (ES), or surgery. There are several unique challenges in patients with strictures distal to a GI tract diversion. In the diverted patient, there is no role for dietary avoidance of vegetable matter, since the passage of chyme will not be impeded, by the stricture. Likewise, stool softeners are not needed, as there is no passage of stool. The patient is unlikely to experience obstructive symptoms. The question arises whether there is a role for endoscopic evaluation and treatment of these strictures. Coordination with a colorectal surgeon to determine the timing of endoscopic versus surgical management can help guide the course. If the patient would ultimately require surgery regardless of endoscopic interventions, delays in the definitive therapy may lead to worsening patient health. The need for endoscopic treatment of the stricture in a diverted or bypassed area of the bowel is typically related to a need to access the area of the bowel on the other side of the stricture, for example in someone who needs colorectal cancer screening.
Previous studies have assessed the rate of colon cancer detected in colonic strictures as roughly 6.8% to 12.5%. , More recently, the GETAID study assessed CD or ulcerative colitis (UC) patients with strictures who underwent surgery. Despite negative biopsies prior to surgery, 3.5% of patients with presumed benign colon strictures were found to have dysplasia or cancer at the time of resection. It is important to visualize and histologically sample the stricture periodically. We do not have an evidence-based approach to surveillance in the diverted bowel, but current recommendations are to follow established guidelines for either average-risk colorectal cancer screening or IBD surveillance based on the underlying comorbidities of the patient. , The bowel on the far side of the stricture may also need to be evaluated, and since CT colonography is not established practice for IBD surveillance due to the potential for flat lesions being missed, it may become necessary to endoscopically treat the stricture to perform surveillance of the bowel.
The two most common endoscopic techniques to treat a stricture are EBD and ES. EBD has been reported successful in 97% of CD strictures, with 46% requiring repeat dilation and 5% experiencing severe complications such as bleeding or perforation. EBD should be performed in strictures that are less than 5 cm in length, by dilating the balloon to a maximum of 18 mm diameter, for at least 1 minute. Success is defined by the ability to pass the scope through the stricture. There is no longer a need to follow the “Rule of three” for esophageal strictures, as bougie dilation beyond three stages does not increase the risk of adverse events. Dilation of strictures in the diverted bowel should be performed until the stricture has been disrupted. Dilation to less than 15 mm or the presence of multiple strictures increases the risk of repeat EBD. The stricture may also be treated with EBD multiple times over a period of months to years to provide a more durable response or to increase the lumen diameter sequentially. Fluoroscopy was not associated with a higher rate of technical success or a lower rate of complications in the treatment of Crohn’s strictures.
Endoscopic stricturotomy is a technique of using endoscopic cautery through an insulated-tip needle-knife to cut back the stricture ( Fig. 23.5 ). This has shown efficacy in terms of resolution of the stricture and improvement of the stenosis in a variety of locations (esophagus, stomach, common bile duct, colon, and IPAA) and settings (IBD, surgical anastomosis, and sleeve gastrectomy). Compared to EBD, ES is more likely to be successful in treating the stricture (100% vs. 90%), less likely to result in perforation (0 vs. 5 cases), less likely to go to surgery for the stricture (9.5% vs. 33.5%), and more likely to require transfusion for bleeding (8.8% vs. 0%). Fortunately, endoscopic management of iatrogenic bleeding is effective in many cases. Key principles include endoscopically identifying the bleeding source, using electrocautery with endoscopic coagulation forceps or mechanical clips to control the bleeding vessel ( Fig. 23.5 ), and, in select cases with impaired visibility, either submucosal injection of 1:10,000 epinephrine in four quadrants or tamponade using an endoscopic balloon. ES is an effective technique that can improve the quality of life (QoL) and in some cases delay surgery.
A risk of ES not shared with EBD is bleeding or perforation caused endoscopic electrocautery. The energy used for cautery creates heat that dissipates the further it travels from the instrument. During the procedure, this heat is used to resect the stricture, but it may lead to denaturation of proteins and cell death. The patient may present with bleeding or perforation after 2 to 3 days of the intervention. As this is a known risk, counseling the patient prior to endoscopy on signs of delayed complications, and when to return for medical care, can lead to prompt recognition and treatment. Repeat endoscopy is often effective for attaining hemostasis at the site of intervention ( Fig. 23.6 ). In select cases, prompt recognition with appropriate antibiotic use and bowel rest can treat the perforation medically without the need for surgical resection of the affected bowel.
If a stricture is not amenable to endoscopic therapy as it is too long, angulated, not in a position that would allow treatment of complications, or refractory to endoscopic therapy, then it is appropriate to consider surgery. The criteria for surgical treatment of a small intestine or colonic stricture include evidence of a stricture, obstructive symptoms, and upstream dilation. Additional criteria include penetrating complications and concern for malignancy. A note of caution for these traditional criteria is that upstream dilation rarely significantly improves with EBD or ES. The utility of upstream dilation as a criteria is of dubious significance after its presence has been established.
Many fibrostenotic strictures of the GI tract can be safely treated endoscopically. This can improve the patient’s QoL, allow for endoscopic screening for malignancy, and potentially reduce their risk of repeat surgery. The ability to recognize and endoscopically treat strictures is changing peoples’ lives in meaningful ways.
Postsurgical Anastomotic Leaks
During the creation of a J-pouch, some patients may experience an anastomotic leak, particularly from the blind limb “tip of the J.” The most common presenting signs are abdominal pain and fever. The diagnosis can be established by imaging such as CT or MRI of the abdomen and pelvis with contrast, contrast enema, or pouchoscopy. Surgery has been the standard therapy and included pouch repair by suturing or stapling the tip of the J-pouch excision with the creation of a new pouch, and repeat anastomosis. This has yielded an 86% 5-year pouch survival.
The endoluminal vacuum-assisted closure (EVAC) using endoscopic sponges provides a nonsurgical option. Initially used for rectal anastomotic leaks, they have increased in utilization throughout the GI tract. The principles are that a polyurethane sponge is attached to a negative pressure system, usually a nasogastric tube, and placed intraluminally or intracavitary. The sponge prevents debris from entering the defect. The negative pressure collapses the cavity and lumen while removing liquids that pool at the site. Changing of the sponge provides debridement of the tissue. With time granulation tissue can form and the wound may close. In the case of an IPAA leak, a series of 8 patients showed 100% resolution after a median of 12 (3–42) days and 3 (1–10) sponge changes.
This is a complex procedure to be performed in patients who are likely quite sick at baseline. The decision to place an EVAC should be made between the surgeon and the gastroenterologist based on the probability of endoscopic success. The endoscopist should be experienced with this procedure, and if no experienced physician is available, consideration for referral to a specialty center may be warranted.
ENDOSCOPIC MANAGEMENT OF FISTULAS
Diversion of the bowel may be performed to treat distal GI fistula, though in some cases a fistula may form in a bowel that has already been diverted. Fistulas can arise from any area of the GI tract and communicate with other locations along the tract. Common extraintestinal sites for the fistula to reach include the skin, bladder, vagina, urethra, and cervix. Fistulas between sections of the bowel may be symptomatic with pain but are often less severe and easier to manage than those that communicate with other organs. The current classification system uses a taxonomy to describe if the fistula is internal or external, the locations of the origin and end of the fistula, and if the fistula is associated with other pathology such as an abscess, stricture, or previous surgical site. Examples would include enterocutaneous fistula, perianal fistula associated with abscess, enterovesicular fistula, or postsurgical rectovaginal fistula.
Principles for managing fistulas regardless of site include controlling the infection through drainage and antibiotic use, optimizing fluid and electrolytes if the fistula results in a high output, nutritional support if the enterocutaneous fistula is located proximally in the GI tract resulting in malnutrition, bowel rest, psychological support, medical control of causative comorbid conditions such as CD, and surgical repair when indicated. Each case is individualized, and treatment is determined by factors such as location, severity, and comorbid conditions.
Advancements in endoscopic techniques and devices allow for endoscopic closure of the fistula in select cases. The case should be discussed between a colorectal surgeon and gastroenterologist both of whom are experienced in managing fistulas. An enteroenteric fistula that is short, not associated with pain or weight loss, and does not have an associated infectious complication such as abscess may not require endoscopic or surgical intervention. This type of fistula may be treated only with antibiotic therapy. It is appropriate to review imaging such as CT or MRI of the affected site prior to endoscopic intervention. The location should be assessed as a lesion deep in the small bowel may not be amenable to endoscopic therapy. Additionally, the fistula os is often quite small and can be difficult to identify. A detailed understanding of the anatomy of the fistula can reduce the chances that treatment will not lead to infectious complications. The presence of inflammation in the area, from an IBD-like CD, should be treated medically prior to endoscopic or surgical fistula closure to offer the best chance of success.
Endoscopic techniques depend on the nature of the fistula. For a small fistula, an endoscopic clip may be used to close the internal os. If the opening is larger than a few millimeters, then over-the-scope clips or endoscopic sutures may be needed to close the fistula. In recalcitrant strictures, the author has used an endoscopic brush to irritate the fistula lumen to the point of bleeding, thus disrupting the surface epithelium. Then doxycycline is injected into the lumen causing sclerosis of the fistula tract. Lastly, the internal os of the fistula along with a 5-mm circumference of tissue around it may be treated with argon plasma coagulation to promote scarring, making the endoscopic clip more effective at definitively closing the fistula.
The risk with endoscopic treatment is that by closing one opening, ongoing pus may accumulate, and the infection may not have an avenue to drain. Abscesses can form along the treated fistula tract, and then treatment may shift from endoscopic therapy to either surgical or interventional radiology options. Current research is evaluating the role of hyperbaric oxygen therapy or mesenchymal stem cell therapy as an adjuvant to current strategies of fistula closure, but these are still being investigated and not the standard of care.
INFLAMMATORY BOWEL DISEASE
Patients with IBD may be diverted for multiple reasons. In perianal CD the diverted bowel tends to become less active and the perianal complications may improve. Patients report fewer CD-related symptoms with a diversion stoma than without. In some cases, with good medical control of the disease the diversion may be reversed. After reversal of a stoma, 23% of patients had an increase in perianal disease activity and a second stoma was needed. Factors associated with a shorter time to stoma reversal included a primary laparotomy and the absence of perianal/rectal CD. In patients with UC with refractory disease or dysplasia, an IPAA is typically formed over two to three surgeries, and at various points, an end ileostomy or loop ileostomy may be used to divert stool away from the distal bowel.
The gastroenterologist may perform endoscopy to assess disease activity, screen for dysplasia, or determine the etiology of symptoms. The same principles apply in the diverted bowel as in other IBD patients for scope selection and interventions performed. The key difference is that due to the diversion, oral preparation of the bowel is not possible. Rectal enemas may be attempted but would not be able to clear stool that is proximal in the colon.
In patients who undergo colectomy for UC, 2% to 8% develop CD. This can be a difficult diagnosis for the patient, as often the formation of the IPAA was performed with the understanding that their UC should improve, and there was a possibility to avoid immunosuppressing medications. The diagnostic criteria are still under debate, but in a patient with inflammation in the pouch resistant to antibiotics, evidence of inflammation or strictures in afferent ileal limb more than 20 cm from the pouch body and fistulizing disease, it is reasonable to make the diagnosis of CD of the pouch.
DIVERSION COLITIS
Diversion colitis occurs in the majority of patients after bowel diversion surgery, and typically within 3 to 36 months after the surgery. In a functioning bowel part of the nutrition to the mucosa is derived from the stool passing through the colon. The bacterial fermentation of short-chain fatty acids including butyric acid is compromised due to diversion, leading to a relative malnutrition. Early after surgery diversion colitis is largely asymptomatic, but can progress to tenesmus and bleeding. During endoscopy the mucosa may appear pale at first, then after insufflation develop stria of oozing blood due to the friable tissue ( Fig. 23.1 ). Key features include erythema, edema, erosions, aphthous ulcers, erosions, and friability. Histologic features may include lymphoid follicular hyperplasia, aphthous ulcerations, and inflammation limited primarily to the submucosa. Care should be taken during endoscopy to avoid traumatizing the tissue, and if visibility is compromised due to bleeding, then the procedure should be aborted.
There is no endoscopic management of this condition. Rather, the need for ongoing diversion should be reevaluated, as restoration of the normal luminal function can effectively, treat diversion colitis. If the diversion cannot be reversed, then removal of the diverted bowel can be considered. The benefit of preserving defunctioned bowel includes improved urinary and sexual performance. If the GI tract is not restored and the diverted bowel is not to be removed, then rectal short-chain fatty acid enemas may treat the diversion colitis. Several formulations have been evaluated and can result in complete endoscopic and histologic remission.
SUMMARY AND RECOMMENDATIONS
Surgical advancements have improved patient outcomes in a myriad of conditions by diverting the fecal stream away from the GI tract. Complications from this process can occur, but fortunately, endoscopic techniques have also been advancing. The gastroenterologist should feel comfortable working within a multidisciplinary team to make the diagnosis. The ability to treat these complications varies based on experience. While many practitioners may be able to provide endoscopic therapies, some may refer when appropriate to centers with expertise. Physicians working together across specialties deliver the best outcomes for our patients.
REFERENCES

Stay updated, free articles. Join our Telegram channel

Full access? Get Clinical Tree


