ABBREVIATIONS
AGA
American Gastroenterological Association
AL
anastomotic leak
APP
advance practice provider
ASA
acetylsalicylic acids
ASCRS
American Society of Colon and Rectal Surgery
ASGE
American Society for Gastrointestinal Endoscopy
BCIR
Barnett continent intestinal reservoir
CAD
coronary artery disease
CD
Crohn’s disease
CRA
colorectal anastomosis
CRC
colorectal cancer
CRCS
colorectal cancer surgery
CRP
C-reactive protein
CRS
colorectal surgery
CSP
cold snare polypectomy
CTE
computed tomography enterography
CT
computed tomography
DOACS
direct oral anticoagulant agents
DVT
deep vein thrombosis
EBD
endoscopic balloon dilation
ECF
enterocutaneous fistula
EEA
end-to-end anastomosis
EMR
endoscopic mucosal resection
ESD
endoscopic submucosal dissection
ESRD
end-stage renal disease
ESt
endoscopic stricturotomy
ESTx
endoscopic strictureplasty
GGE
gastrografin enema
GI
gastrointestinal
HSP
hot-snare polypectomy
IBD
inflammatory bowel disease
ICA
ileocolonic anastomosis
ICR
ileocolonic resection
IPAA
ileal pouch–anal anastomosis
IQR
interquartile range
KSA
Kono-S anastomosis
LAR
lower anterior resection
LMWH
low-molecular-weight heparin
NKSt
needle-knife stricturotomy
NSAID
nonsteroidal antiinflammatory drugs
PCP
primary care provider
PSC
primary sclerosing cholangitis
SBO
small bowel obstruction
SEMS
self-expandable metal stent
SSI
surgical site infection
STx
strictureplasty
TME
total mesorectal excision
TIPS
transjugular intrahepatic portosystemic shunting
TNF
tumor necrosis factor
UC
ulcerative colitis
INTRODUCTION
Diagnostic and therapeutic endoscopy along with surgery are the main modalities for the management of benign and malignant colorectal disease (CRD). Commonly performed endoscopy for lower gastrointestinal (GI) disorders diagnostic colonoscopy, colonoscopy with polypectomy, colonoscopy with endoscopic balloon dilation (EBD) or electroincision of strictures, colonoscopy with resection or dissection of neoplastic lesions. Commonly performed colorectal surgical procedures for benign or malignant CRD include partial or complete colectomy, ileocolectomy with anastomosis, diverting ostomies, strictureplasty, and ileal pouch–anal anastomosis (IPAA).
Procedure- or surgery-associated complications are common in the patients at risk and occur in more invasive procedures or surgeries. Common complications during or after endoscopic or surgical interventions are GI bleeding, perforation, acute or chronic anastomotic leaks (AL), and their associated abscess, sepsis, or sinus, and ileus. In most cases, direct causes of these postprocedural or operative complications can be identified and managed. However, sometimes the etiology of the complications is not readily found beforehand. The risk factors for these complications may be related to patients’ comorbidities, the nature of the disease, characteristics of targeted lesions, settings, endoscopic and surgical techniques, and the expertise of endoscopists and surgeons. This chapter aims to highlight risk factors for complications that may occur during postendoscopic or postsurgical procedures. This will help healthcare providers, including primary care providers (PCPs), gastroenterologists, endoscopists, advanced practice providers (APPs), gastroenterologists, and colorectal surgeons (CORS), to prevent and manage these complications.
OVERVIEW OF INFLAMMATORY BOWEL AND COLORECTAL DISORDERS
Lower GI endoscopy consisting of colonoscopy, flexible sigmoidoscopy, ileoscpy, and pouchoscopy plays a key role in the diagnosis and management of IBD (i.e., Crohn’s disease [CD], ulcerative colitis (UC), and ileal pouch disorders) and other benign or malignant CRD.
Inflammatory Bowel Diseases
Inflammatory bowel disease is a chronic and debilitating spectrum of disorders that require life-long medical therapy, and endoscopic, and/or surgical interventions in some. Most patients with inflammatory phenotypes of CD would eventually develop structural complications such as stricturing or penetrating phenotypes, despite advances in medical therapy. The complications in CD encompass GI bleeding, severe anemia, malnutrition, failure to thrive abdominal abscess or sepsis from fistula or perforation, and bowel obstruction secondary to primary or anastomotic strictures. Severe GI bleeding is rarely seen in CD and incidence was reported to be 0.6% to 4%. The risk of GI bleeding in CD was documented to be associated with the use of corticosteroids. The use of nonsteroidal antiinflammatory medications can also aggravate the damage to GI mucosa leading to strictures and bleeding. More common than CD, colonic and/or rectal bleeding is more insidious in UC. The structural complications usually require endoscopic and/or surgical interventions, with modalities such as EBD, endoscopic stricturotomy (ESt), endoscopic strictureplasty (ESTx), endoscopic fistulotomy, ileocolonic resection (ICR) and ileocolonic anastomosis (ICA), and strictureplasty (STx).
The recurrent mucosal inflammation and ulceration in patients with UC, as well as CD, can trigger hyperplastic changes to GI mucosa and lead to the development of pseudopolyps. These polyps can easily enflame GI bleeding. These inflammatory polyps are more commonly seen in UC patients as compared to CD patients. Long-standing UC increases the risk for the development of colitis-associated neoplasia (CAN), especially in those with concurrent primary sclerosing cholangitis (PSC). Some polypoid, slightly raised, or clear-demarcated flat CAN lesions may be managed with endoscopic mucosal resection (EMR) or endoscopic submucosal resection (ESD). Approximately 10% to 20% of patients with UC would require colectomy for medically refractory disease or CAN. Restorative proctocolectomy with IPAA, commonly with “J” or “S” pouch, is the preferred surgical procedure after colectomy for UC. Structural or inflammatory complications of IPAA are common, including anastomotic strictures, acute and chronic leaks, fistula or abscess formation, sepsis pouchitis, cuffitis, and floppy pouch complex. ,
Colorectal Neoplasia
Colorectal cancer (CRC) is the third most common cancer in the United States. Therefore, it is essential for CRC screening in the general population as well as those at risk. The early detection of colonic neoplasia results in onsite or timely endoscopic management. Endoscopic removal of premalignant polyps has been shown to prevent CRC. Colonoscopic polypectomy requires the skills of endoscopists. It is safe and feasible to perform EMR or ESD for large or superficial neoplastic lesions in experienced hands. The removal of diminutive (1–5 mm) polyps can effectively be performed using cold-snare polypectomy (CSP). The complete resection rate for adenomatous polyps (1–7 mm) has been considerably higher in the CSP group than in excisional cold biopsy. Submucosal lifting is less frequently needed during CSP thus decreasing the potential risk of bleeding or perforation. Hot-snare polypectomy (HSP) helps resect bulkier polyps, such as 4 to 9 mm polyps, 10 to 19 mm nonpedunculated polyps, and pedunculated polyps with head size of >20 mm and stalk size of >10 mm. Interestingly, HSP has been modified by adding submucosal injections, the use of adjuncts, or an underwater technique to reduce diathermy-related complications. The pretreatment by using dilute adrenaline and/or mechanical hemostasis to a pedunculated colorectal polyp with a stalk of ≥10 mm and head ≥20 mm before HSP may help prevent bleeding. The addition of methylene blue or indigo carmine to the solution of submucosal injection during the resection process allows demarcation of the lateral margin of polyps including the submucosal plane.
Most patients with CRC require colectomy. A partial colectomy is most commonly performed. For rectal cancer, lower anterior resection (LAR) with colorectal anastomosis (CRA) is often performed. However, AL is common, especially in those with chemoradiation therapy. Temporary diverting ileostomy or colostomy is performed in some patients undergoing colectomy with CRA and pelvic radiation therapy to reduce the risk of anastomotic stricture.
Other Benign Colorectal Disorders
Diverticulosis is mostly seen during colonoscopy or abdominal imaging as outpouching in a weak colonic mucosal wall layer. The diverticular diseases encompass a range of mild to severe clinical manifestations from asymptomatic self-limiting infection to symptomatic diverticular bleeding, and diverticulitis with complications of perforation, abscesses, sepsis, and fistula to the skin, bladder, or vagina. While the role of endoscopic therapy is limited in the treatment of diverticulitis, colonoscopy or flexible sigmoidoscopy is the main treatment modality for diverticular bleeding. Surgery with partial colectomy with or without a diverting ostomy is warranted after frequent or recurrent attacks of diverticulitis or diverticular bleeding to prevent further complications. This author (Dr. Bo Shen) has performed an endoscopic extraction of trapped fecal bezoars in the diverticula.
In addition to IBD-associated colonic neoplasia and diverticular disease, sporadic CRC, colectomy is also performed in complicated diverticular colitis, complicated ischemic colitis, severe slow-transit constipation, and volvulus of the colon. Rectal pexy or resection surgery is performed for the treatment of rectocele, rectal prolapse, rectal/sigmoid intussusception, and solitary rectal ulcer syndrome.
Postcolectomy complications in these benign disorders are common. These complications include anastomotic bleeding, anastomotic strictures, acute and chronic AL, abscesses, sepsis, and fistulas. The temporary ostomy is constructed for those at risk for anastomotic complications. In addition, complications can occur in the stoma (e.g., prolapse, retraction, bleeding, and stenosis), and diverted large bowel (e.g., diversion colitis or proctitis, diversion-associated strictures) stoma complications should be monitored for unusual bleeding, prolapse, perforation, or retraction issues over time. Stoma formation can be evident in conditions like IBD, CRC, or complicated diverticulitis. At Columbia University Irving Medical Center, we have been performing endoscopic treatment (such as band ligation and plication) for postoperative recurrent prolapse and intussusception. Some patients have elected to try endoscopic therapy before surgery.
COLONOSCOPIC POLYPECTOMY
Endoscopic removal of premalignant polyps or polypectomy is essential to prevent CRC. Safely and effectively performing polypectomy is considered a fundamental skill for GI endoscopists and colorectal surgeons. However, therapeutic colonoscopy (such as polypectomy) carries a higher risk for bleeding and perforation than diagnostic colonoscopy, particularly endoscopic polypectomy in the right colon.
Techniques
The preparation for polypectomy (e.g., excellent bowel cleansing and management of anticoagulants, antiplatelet agents, and systemic corticosteroids) is an important first step. Correct positioning of the patient during the procedure along with straightening of the endoscope shaft, that is, the short-scope position is warranted. The correct size (10–30 mm) and type of snare should also be selected vigilantly. The removal of diminutive (1–5 mm) polyps is effectively performed using cold snare polypectomy (CSP) with or without submucosal lifting or cold biopsy forceps (CBF). The complete resection rate for adenomatous polyps (1–7 mm) is shown to be higher in the CSP group than in the CBF group. During CSP persuaded polypectomy a cuff of normal mucosa is resected surrounding the polyp. Hot snare polypectomy (HSP) is routinely used to resect bulkier polyps, such as 4 to 9 mm in size, 10 to 19 mm in size nonpedunculated polyps, and pedunculated polyps with head size of >20 mm and stalk size of >10 mm. Interestingly, HSP has been modified by adding submucosal injections, the use of adjuncts, or an underwater technique to reduce diathermy-related complications. Submucosal injection of methylene blue or indigo carmine allows demarcation of the margin of polyps including the submucosal plane. There are other endoscopic procedures for the removal of larger, flatter polyps, such as EMR or ESD.
Common Complications
Common complications of colonoscopic polypectomy include bleeding, bowel perforation, and postpolypectomy syndrome. The severity of bleeding and perforation depends upon factors such as age, sex, comorbidities, coagulopathies, the quality of bowel preparation, the size and shape of polyps, and techniques. Bowel perforation can be extraperitoneal or intraperitoneal, where extraperitoneal perforation can cause retroperitoneal passage of air leak that may lead to pneumoretroperitoneum, pneumomediastinum, pneumothorax, subcutaneous emphysema, or even pneumopericardium. These patients may present with symptoms of shortness of breath, chest pain, subcutaneous crepitus, or neck swelling after colonoscopy. On the other hand, intraperitoneal perforation triggers a leak of air or colonic contents into the peritoneum, which can be recognized by acute failure to maintain appropriate intramural distention, acute change in the patient’s clinical status, or sudden onset of acute abdominal pain. If these symptoms are delayed or gradually worsen along with tachycardia, abdominal distention, no flatus, and abdominal wall tenderness then urgent clinical evaluation is warranted. The chest and abdominal radiographs may be evidence of free air under the diaphragm. The high suspicion of perforation necessitating urgent rescue endoscopic or surgical interventions should be immediately evaluated with abdominal series and/or computed tomography.
The reported frequency of bleeding during and postpolypectomy ranged from 0.3% to 10%. The bleeding can happen during the procedure or delays up to 30 days after the polypectomy. , Polypectomy thermocoagulation with transmural burn may cause postpolypectomy syndrome. Patients with polypectomy syndrome present with abdominal pain, fever, leukocytosis, and even peritoneal inflammation without a frank perforation.
Risk Factors for Complications
Various measures are taken to reduce the risk of bleeding, perforation, or postpolypectomy syndrome by reviewing the patient’s age, past medical history, and concurrent current medications as well as endowing standard practice of endoscopy. Most risk factors are preventable and manageable. Risk factors for postpolypectomy complications have been extensively studied. A large population-based study demonstrated that the risk of perforation was increased in patients >65 years old and even more in patients >80 years old. Females are at higher risk for colonic perforation due to endoscopy even with or without polypectomy as females have a longer colon with a rounded and deeper pelvis to cause sigmoid looping. , Perforation as well as the bleeding risk during polypectomy is greater in the right colon. The perforation rate was relatively elevated in >60-year-old patients. It seemed to be associated with general anesthesia. Coronary artery disease (CAD), atrial fibrillation, pulmonary conditions, chronic kidney disease, acute or chronic liver problems, or coagulopathies can increase the risk of adverse outcomes, such as perforation, bleeding, myocardial infarction, respiratory failure, and stroke, and during any endoscopic interventions. The perforation risk during polypectomy has recently been reported at 1.78% in non–end-stage renal disease (ESRD) versus 5.83% in patients with ESRD. From a technical perspective, instant or delayed perforation is caused while removing large polyps or barotrauma of excessive bowel looping. The weakened colonic wall especially the cecum with a thin mucosal wall is at the highest risk of perforation from over-insufflation, thermocoagulation, or barotrauma. Transmural injury can occur if the muscularis propria is trapped by snares during the polypectomy. The reduced colonic contractions due to age-related degenerative changes in the myenteric plexus may lead to decreased neuronal density and an increased fibrous component of the neuron. These changes may alter elasticity in colonic tissue, causing excessive looping of the large bowel and making the colon (especially the sigmoid colon or rectosigmoid junction) vulnerable to barotrauma. , Anatomically, the intraperitoneal perforation above the rectum during polypectomy should be addressed immediately to avoid peritonitis and subsequent sepsis. The assessment of the polyp per its size, morphology, location, and easy accessibility to the colon is thoroughly screened before planning a polypectomy. The preparation for polypectomy is the second most important step followed by the assessment of the polyp. The correct position of the patient during the procedure along with an appropriate rotation/alignment of the endoscopic shaft is warranted to seek a better view of a targeted polyp. The correct size (10–30 mm) and type of snare should also be selected vigilantly. Even poor bowel preparation, or technical errors during polypectomy by the endoscopy team can lead to an increased risk of polypectomy-associated complications. Therefore, the experienced gastroenterologist may have a lessened rate of complications performing routine or emergency polypectomy.
Multiple factors contribute to postpolypectomy bleeding, for which the risk is approximately 4.5 times higher while removing ≥10 mm polyps than smaller polyps. Large and thick stalked polyps, sessile polyps with or without probable risk of tubular adenoma or malignancy, laterally spreading polyps, and the presence of multiple polyps can also increase the risk for bleeding , , ( Table 3.1 ). Interestingly, it has been documented that CAD can increase the risk of the development of colonic polyps and CRC. This high-risk patient population would require anticoagulation therapy to increase overall life expectancy. It is therefore warranted to demand standard peri-endoscopic management of anticoagulant therapy before initiating any endoscopic procedures. If patients are on anticoagulants, such as antiplatelet agents, for example, acetylsalicylic acid (ASA), clopidogrel, prasugrel or warfarin, low-molecular-weight heparin (LMWH), or direct oral anticoagulant agents (DOACS), for example, apixaban, betrixaban, dabigatran, rivaroxaban, edoxaban, these medications should be appropriately optimized before endoscopy to reduce the risk of immediate or postpolypectomy bleeding. A multicenter study showed that there is a lower risk of postpolypectomy bleeding if patients continue to take aspirin. It is always safe to review the most recent platelets and coagulation laboratory values before endoscopic interventions, especially for patients who are on anticoagulants or antiplatelet agents. In general, DOACS or warfarin agents are stopped within 5 to 7 days before colonoscopy with possible polypectomy procedures. Patients with cirrhosis or a Child-Pugh B score have shown an increased risk of bleeding after colonoscopic polypectomy. The electrocoagulation injury during polypectomy can not be underestimated, since it can lead to adverse events. Intraprocedural hemostasis can be managed by applying pressure on the pedicle to stop excessive bleeding. Additionally, epinephrine injection, cauterization, deployment of endoclips, applying loops, and band ligation may also help control bleeding. Endoclips can be placed prophylactically during the removal of the larger-size polyp to prevent delayed postpolypectomy bleeding and/or perforation. It should be noted that immediate or delayed bowel perforation is also at stake during polypectomy or while endeavoring hemostasis efforts.
Authors | Sample Size and Study Design | Risk Factors Reported |
---|---|---|
Park et al. | Prospective (n = 2716) | Female sex. |
Forsberg et al. | Retrospective (n = 593,315) | Colonoscopy under general anesthesia; age >60 years. |
Rutter et al. | Retrospective (n = 130,831) | Increasing polyp size. Polyps at the cecum. |
Zhang et al. | Retrospective (n = 5600) | >10 mm size polyp and colon polyp pathology. |
Blotiere et al. | Retrospective (n = 947,061) | Age >80 years; larger polyps (>1 cm); resection of >4 polyps; emergency colonoscopy; experience of gastroenterologist (performing <244 colonoscopies per year). |
Buddingh et al. | Retrospective case-control (n = 156) | Polyps located in the right colon. |
Consolo et al. | Retrospective (n = 1038) | Malignant polyp; polyp located at the sigmoid colon. |
Kim et al. | Prospective (n = 5152) | >65 years old; cardiovascular or renal comorbidities; use of anticoagulants; thick polyp stalk; laterally spreading polyp; technical procedural errors; poor bowel preparation. |
Nivatvongs et al. | Prospective (n = 1172) | Use of anticoagulants (aspirin, coumadin); electrocoagulation injury; polyp in the right side of the colon; sessile or pedunculated polyp. |
Hot-snare or hot biopsy for polypectomy has a higher risk for CSP or cold excision biopsy, respectively. A multicenter retrospective case-control study found that hypertension, large lesion size, and nonpolypoid configuration were independent risk factors associated with postpolypectomy syndrome. EMR or ESD for the removal of polyps with thermocoagulation to deeper layers carries a higher risk for postpolypectomy syndrome than conventional polypectomy. Approximately 1% of postpolypectomy syndrome cases either with EMR, or 7% to 8% of cases with ESD have been reported.
ENDOSCOPIC MUCOSAL RESECTION AND ENDOSCOPIC SUBMUCOSAL DISSECTION
Advances in therapeutic endoscopy have made EMR or ESD amenable to the removal of colorectal neoplasia. EMR and endoscopic ESD have evolved into valid treatment modalities to treat early-stage colonic neoplasia. EMR was first described by Deyhle et al. in 1973. However, EMR and ESD techniques were more popularized in the 1990s. The difference between EMR and ESD is related to the level of a submucosal plane that is “transected.” ESD which requires more technical expertise than EMR gives a more histopathological margin of removal of neoplastic lesions than EMR. The lesions can manifest flat to raised morphology, which needs to be discerned for an adenoma-like lesion versus sporadic adenomas. EMR or ESD has been performed in selected patients with CAN in patients with underlying IBD. EMR and ESD even evolved to endoscopic full-thickness rection for the treatment of colonic neoplasia. , In addition, hybrid ESD-EMR procedures allow skilled endoscopists to incise lateral mucosal margins around the suspected neoplastic lesions. A skilled endoscopist can choose the appropriate techniques for the resection of the neoplastic lesion based on certain characteristics of the targeted lesions while emphasizing the best patient-centered approach. A repeat diagnostic and surveillance colonoscopy may be performed to document the healing and eradication of the neoplastic lesions.
Techniques
Both EMR and ESD require augmenting the submucosal bowel wall layer by creating a safe mucosal platform and thus avoiding injury to the muscularis propria leading to bleeding or bowel perforation. EMR is less time-consuming and cost-effective with a steeper learning curve than ESD. EMR can resect 20 mm or mucosal lesions en bloc and resect large >2 cm, complex, flat, or sessile lesions in a piecemeal fashion. Incomplete resection, however, is associated with the recurrence of malignant lesions. ESD which is more technically demanding has been used to resect larger and flat lesions by dissecting the submucosal area underneath these lesions. The submucosal layer can be expanded by injecting fluids (e.g., saline, hypertonic glucose, indigo or methylene blue or sodium hyaluronate, or low-dose adrenaline) to prevent injury to abutting deep muscle layers.
Common Complications
The most common complications secondary to EMR or ESD are onsite or delayed bleeding and perforation, with some patients requiring hospitalization, or even surgical intervention. Naturally, the risk for bleeding or perforation is higher in ESD than that in EMR. It is estimated that the risk from EMR is 2.3% to 3.5% as compared to 4.8% to 5.7% from ESD, whereas the risk for perforation risk from EMR is reported to be 0.9% to 1.4% as compared with 4.8% to 5.7% from ESD. An international survey showed that perforation or delayed bleeding due to ESD has been commonly seen in up to 12% of resecting colorectal lesions as compared to the esophagus (6%) or stomach (7%). A multicenter study reported a rate of delayed bleeding of 6.7% in 2424 EMR procedures. Terasaki et al. conducted a retrospective cohort study of the adverse outcomes in the resection of laterally spreading tumors with either EMR, ESD, or hybrid ESD. This study reported that the perforation rate was 1.4% in EMR, 0% in ESD, and 7.1% in hybrid ESD and the rate of bleeding was 7.1% in EMR, 11.5% in ESD, and 0% in hybrid ESD. ESD was reported to have an increased rate (6.9–9.9%) of subsequent surgery as compared to EMR (4.1–5.8%), presumably because ESD allowed the resection of larger submucosal lesions with invasive cancers.
Patients with long-standing IBD in the colon are at risk of developing CAN. Some investigators have proposed and practiced EMR and ESD for CAN. However, the long-term oncological outcomes of EMR or ESD in CAN remain to be defined.
Risk Factors for Complications
A systemic review and meta-analysis from the American Society for Gastrointestinal Endoscopy (ASGE) reported bleeding and perforation as major adverse events secondary to EMR and ESD. The risk of bleeding, perforation, or postpolypectomy syndrome after EMR shall not be overlooked. Several risk factors shall be meticulously assessed before performing EMR, such as polyps with adenomatous pit pattern, high-grade dysplasia, adenocarcinomatous lesions, diameter of polyps, >40 mm size lesions, laterally spreading colorectal tumors, large sessile colorectal polyps, and lesions in the right colon that can otherwise lead to potential adverse events ( Table 3.2 ). Expert endoscopists could lower complication rates (<3%) secondary to ESD than endoscopists under training with relatively less experience in ESD performance (5–10%). High-risk patients are those >65 years old with comorbid conditions, concurrently on anticoagulants, or having colonic lesions located at the cecum. The right colon wall is thinner (1.5 mm) as compared to the thicker (3 mm) left colon or rectum wall and therefore bleeding and perforation risk is greater in the right-sided colon wall. A multicenter retrospective study of cirrhotic patients undergoing EMR showed risk factors for delayed bleeding, including advanced age ≥65 years, >15 mm polyp size, right colon polyps, and use of antiplatelets within 5 days of colonoscopy. Delayed bleeding is infrequent (<2.2%) after ESD, since major perforating vessels could be assessed, isolated, coagulated, or clipped if needed during a submucosal lift before transecting the targeted neoplastic lesion. Up to 7% of patients have experienced severe or clinically significant delayed bleeding after resecting large colonic lesions, which required hospitalization, blood transfusions, or even repeat endoscopy with clipping or cauterization of the bleeding lesions. Pohl et al. reported that 3.5% of patients with postprocedural bleeding were treated with EMR with the use of endoscopic clip closure of the defect and 7.1% with post-EMR bleeding without the application of endoscopic clips. This multicenter randomized control trial also reported that EMR for a large nonpedunculated colon polyp (≥20 mm) was a risk factor for bleeding. The removal of larger polyps has been seen to cause delayed bleeding after EMR in several other studies. , The presence of a “cherry red spot” or visible muscle fibers in the resection defect was found to be associated with delayed bleeding after resecting large nonpedunculated colorectal polyps by EMR. A modified technique with the endoscopist transecting through mid or superficial submucosal layers during EMR may reduce the risk of traumatizing large feeding vessels in deep mucosal layers and subsequently the risk of bleeding or perforation.
Endoscopic Technique for Polypectomy | Authors | Sample Size and Study Design | Risk Factors Reported |
---|---|---|---|
Endoscopic mucosal resection | Son et al. | Retrospective (n = 275) | Adenomatous pit pattern; high-grade dysplasia; adenocarcinoma. |
Hong et al. | Retrospective (n = 594) | Comorbidities; adenocarcimatous lesions; EMR-P (piecemeal resection), perforation in laterally spreading colorectal tumors. | |
Moss et al. | Prospective (n = 1134) | Intraprocedural bleeding; lesion size >40 mm; use of argon plasma coagulation. | |
Burgess et al. | Prospective (n = 1172) | Large sessile colorectal polyp. | |
Metz et al. | Prospective (n = 288) | Right colon/proximal lesion location; anticoagulation. | |
Sawhney et al. | Case-control (n = 4592) | Anticoagulation; diameter of polyps. | |
Endoscopic submucosal dissection | Tachikawa et al. | Retrospective (n = 535) | Obesity. |
Ogiyama et al. | Retrospective (n = 87) | Anticoagulation. | |
Pérez-Cuadrado-Robles et al. | Retrospective (n = 225) | Nonlifting lesion; nongranular-type laterally spreading tumor; difficult retroflexion. | |
Arimoto et al. | Prospective (n = 223) | Female sex; lesion located in the cecum; ESD time > 90 minutes. | |
Okamoto et al. | Retrospective (n = 451) | Rectal tumor location; antithrombotic therapy. | |
Bae et al. | Retrospective (n = 220) | Bleeding after resecting protruding tumors with size > 60 mm. | |
Suzuki et al. | Retrospective (n = 317) | Lesion at the cecum and significant bleeding during ESD. |
Perforation leading to surgical interventions after EMR is infrequent with an estimated frequency of <1% of overall cases. The deployment of endoclips at the site of mucosal bleeding or a breach in the muscular layer reduces the risk of the progression of bleeding or perforation. One must note that delayed bleeding can happen within three weeks after performing polypectomy, EMR, or ESD. However, it should be safe to educate high-risk patients on alarming symptoms after the procedure so that urgent medical or surgical management will not be delayed in case it happens.
Electrocoagulation syndrome also named transmural burn syndrome is defined as transmural thermoelectric injury from nonintended-layer electroincision of endoscopy. Electrocoagulation syndrome can present after hot-snared or hot-biopsy polypectomy, EMR, or ESD, but it is more often in ESD than polypectomy. Similar to the postpolypectomy syndrome, patients with electrocoagulation syndrome typically have symptoms of localized pain and fever. Reported risk factors for electrocoagulation syndrome after ESD are female sex, the presence of submucosal fibrosis, larger tumor size, prolonged procedure time, and right colon lesions (mainly the cecum). , Moreover, risk factors related to potentially unfavorable outcomes due to ESD include obesity, female sex, anticoagulation therapy, lesions located at the cecum or rectum, larger lesions >60 mm in size, nonlifting, nongranular laterally spreading lesion, and even procedure time >90 minutes ( Table 3.2 ).
ENDOSCOPIC BALLOON DILATION OF STRICTURES
Stricture is defined as a luminal narrowing of the GI tract. Patients with strictures can have symptoms of nausea, vomiting, bloating, abdominal pain, constipation, and weight loss. With CD being a classic example, the luminal narrowing can result from intraluminal materials or lesions (e.g., large pedunculated inflammatory polyps and retained capsule endoscope), intramural inflammation, fibrosis, or muscular/neuronal hyperplasia, or extraluminal compression (e.g., from mass, enlarged lymph nodes, mesh, and adhesions). Strictures can be further classified into (1) inflammatory versus fibrotic versus mixed; (2) short versus long with a cut-off of 4 to 5 cm; (3) single versus multiple; (4) with versus without prestenotic dilation; (5) primary versus anastomotic; (6) benign versus malignant; and (7) simple versus complex (e.g., these with associated fistulas or abscesses).
Endoscopic treatment with EBD has become a valid treatment option for strictures in IBD and other CRD. EBD is performed for the treatment of primary strictures and anastomotic strictures in the surgically altered bowel, such as ileocolonic anastomosis (ICA), ileostomy, colostomy, stricturoplasty, or ileal pouches. To improve the outcome of EBD, preprocedural bowel preparation, cross-sectional imaging, such as computed tomography (CT), CT enterography (CTE), magnetic resonance imaging (MRI) or MR enterography (MRE), and contrasted enema via the anus or stoma can provide a roadmap to delineate anatomy and characteristics of the strictures. Video capsule endoscopy should be avoided in the patients suspective of having small or large bowel strictures.
The goal of endoscopic therapy for strictures is to attain relief from ongoing or recurrent obstructive symptoms, reduce mucosal inflammation from the consequence of fecal stasis, prevent the formation of abscess, fistula, or sinus secondary to chronic downstream bowel obstruction, and bridge medical and surgical therapies by spacing out or even avoiding potential surgery. The consensus guideline for CD strictures recommended that both symptomatic and asymptomatic strictures be treated endoscopically. The same rule may be applied to non-CD IBD strictures or strictures from other benign large bowel or ileal pouch disorders.
Techniques
Endoscopic balloon dilation is mostly performed under conscious sedation or monitored anesthesiology care in inpatient or outpatient settings. We recommend the use of a gastroscope or pediatric colonoscopy to perform EBD. A carbon dioxide insufflation is recommended to reduce the risk of ileus and perforation. There is no significant evidence available in the literature on the most appropriate duration for balloon insufflation. Both wire-guided short balloons (5.5 cm in length) and non–wire-guided long balloons (8 cm in length) can be used for EBD, depending on the nature of the strictures. A graded dilation with balloon sizes of 8 to 20 mm can be deployed in a step-wise fashion to the targeted stricture. High-grade, angulated, and nontransversable strictures can be safely dilated using wire-guided balloons. If possible, EBD with a retrograde fashion is preferred to antegrade dilation without seeing the bowel lumen proximal to the stricture. EBD may be avoided in the stricture with deep ulcers. Pre- and postprocedure inspection of targeted strictures should be performed carefully for deep ulcers, fistula, abscess, or presence of fistula and procedure-associated bleeding or perforation. Usually, patients are observed in the endoscopy recovery room for 30 to 45 minutes after the procedure. Prolonged observation and even overnight may be required when excessive bleeding perforation and ileus are suspected.
Endoscopic Balloon Dilation of Strictures in Crohn’s Disease
According to the Montreal classification, the phenotype of CD is divided into inflammatory (B1), stricturing (B2), and penetrating (B3). The most common locations of strictures in CD are the terminal ileum, ileocecal valve, colon, and anorectal ring. The myofibroblasts produce extracellular matrix proteins. Excessive deposition of matrix proteins at various sites of injured bowel mucosa during several cycles of remission and relapses in CD can lead to stricture formation. It has been recently reported in a metaanalysis study that fibrotic stenosis of the intestinal mucosa in CD can occur in 10% to 20% of patients within 10 to 20 years from the initial diagnosis. Intramural inflammation, fibrosis, muscular hyperplasia, and neuronal hyperplasia lead to intrinsic strictures. The role of medical management is limited in the treatment of fibrotic strictures, giving two other therapeutic options, that is, endoscopic and surgical approaches. Surgical resection and anastomosis are more effective, yet more invasive than EBD. Surgical resection and anastomosis also have inherent issues of postoperative complications and postoperative disease recurrence. In most cases with CD strictures, surgical interventions eventually become necessary. However, multiple surgeries can be debilitating to CD patients and lead to short-gut syndrome.
Multiple case series reported favorable short-term (passage of endoscope after EBD) and long-term (i.e., reintervention and surgery-free survivals) efficacy and procedure-associated complications. Previous studies showed an immediate success rate of 50% to 95% on the first stricture dilation by EBD and 89% following redilation in managing CD-related strictures. , The response of long stricture (>5 cm). Adverse events can occur after EBD. , The authors’ group conducted a series of case-control studies comparing endoscopic vs surgical therapy for IBD and non-IBD strictures. We compared EBD and surgical ICR and ICA in the treatment of anastomotic strictures at the ICA in CD. The results showed that EBD delayed surgical intervention by nearly 6.45 years in these patients. The two surgery-free survival curves from EBD versus ICR and ICA merged at 10 years. However, the surgery-free survival curves of EBD versus surgical ICR and ICA for primary ileocolonic CD strictures were still widely separated at 10 years. The results suggest that EBD is more effective in treating anastomotic strictures at ICA than primary ileocolonic strictures. Therefore, EBD being a less invasive approach is more applicable to anastomotic strictures in CD with the main goal set to reduce the number of surgical resections.
Significant complications of EBD are bleeding, ileus, and bowel perforation. The latter can lead to peritonitis and sepsis. It is estimated that EBD-associated perforation or major bleeding was 3% to 5% in CD per patient. , A prospective study claimed a 9.3% complication rate after balloon dilations >25 mm (20/216 EBD procedures). These complications can result in further complications, such as surgical intervention, admission to the intensive care unit, or even mortality. ,
Risk factors for complications and complications have been extensively studied with prospective and retrospective chart reviews ( Table 3.3 ). The short interval between two consecutive endoscopic interventions may increase the risk of bleeding or perforation. Reported or purported risk factors of EBD-associated perforation include severe inflammation at the site of stricture, the use of systemic corticosteroids, strictures with small diameter, multiple and/or angulated small bowel strictures with or without deep ulcers, length of strictures, for example, >5 cm, asymmetric strictures, and strictures with adjacent fistula or abscess. , Spindle-shaped strictures are at greater risk of perforation on EBD as compared to weblike strictures. Of all risk factors, the most important factor contributing to procedure-associated perforation and subsequent complications is the diagnosis of severe IBD, , and the systemic use of corticosteroids. Smoking and previous abdominal surgeries are also reported to be risk factors. The use of a large-size balloon can elevate the risk of adverse events. It is recommended that patients undergoing endoscopic therapy for CD strictures should have a clean bowel preparation and minimum or no systemic corticosteroids. The concurrent use of immunomodulator or antitumor necrosis factor (TNF) does not appear to increase the risk for IBD-associated complications. , , ,
Authors | Sample Size and Study Design (n > 75) | Risk Factors |
---|---|---|
Hirai et al. | Prospective (n = 112) | Potentially small diameter of stricture. |
Takenaka et al. | Prospective (n = 209) | Strictures: deep ulcerated, sharp angulated, long length, intestinal fistula. |
Lian et al. | Retrospective (n = 176) | Smoking; previous use of corticosteroids; previous abdominal surgeries. |
Gustavsson et al. | Retrospective (n = 178) | Technical: using the largest balloon for EBD (diameter of 25 mm). |
Shen et al. | Historical cohort (n = 150) | Crohn’s disease-associated stricture; stricture of redo pouch. |
Couckuyt et al. | Prospective (n = 55) | Ileosigmoidal or ileorectal anastomoses; the presence of fistula near the targeted stricture. |
Suchan et al. | Retrospective (n = 94) | Technical error and previous surgery for malignancy. |
Endoscopic Balloon Dilation in Ulcerative Colitis and Ileal Pouches
Benign and malignant strictures can occur in patients with long-standing UC. All colonic strictures should be extensively biopsied to rule out malignancy. Gumaste et al. conducted a retrospective study of 1156 patients with UC and reported 5% of intestinal strictures, of which 70% (out of 5%) of strictures were malignant and located proximal to the sigmoid colon. All malignant strictures should be treated with colectomy. However, the development of benign strictures in long-standing UC has been noticed, largely from submucosal fibrosis and hyperplasia of the muscularis mucosae. On endoscopy, the benign strictures usually are nonulcerated and present in the distal rectum or rectosigmoid junction in the background of the tubular rectum and colon. The pathophysiology of stricture formation in UC appears to be similar to that in CD, but benign strictures in UC are not transmural. The authors have experience in EBD of short UC-related benign strictures in a few patients.
Endoscopic therapy is largely applied for the treatment of strictures after colectomy and construction of the ileal pouches. Restorative proctocolectomy and IPAA are usually performed in stages for patients with underlying UC or some patients with FAP. During the staged IPAA and subsequent follow-up, strictures can occur at various locations, including the distal rectum at the Hartmann pouch, ileostomy, stoma closure site, pouch inlet, and pouch-anal anastomosis. Commonly constructed pouches include the J- or S- or K-pouches. The prevalence of pouch strictures with complications ranges from 5% to 38%. ,
The causes of pouch strictures include CD of the pouch, surgery-associated ischemia, the use of NSAIDs, anastomotic dehiscence, anastomotic tension, dysfunctional ileostomy, pelvic abscess, and pelvic sepsis. CD of the pouch can also lead to ileal pouch strictures with an incidence of 2.7% to 13%. One of the hypotheses suggests that the fibrosis ensued by partial dehiscence at IPAA or ischemic changes at the marginal anastomosis may lead to the development of a fibrous web at the anal anastomosis. Other contributing factors for strictures refractory to EBD include the presence of prolapse mucosa above the stricture.
Pouchoscopy is a main diagnostic and therapeutic modality for pouch strictures. It can accurately measure the number, length, location, severity, and concurrent inflammation of strictures. EBD is routinely performed in patients with strictures with favorable outcomes, although endoscopic treatment of the anastomotic strictures and pouch inlet strictures has been challenging in terms of the requirement of repeat endoscopic intervention. Investigators at Mayo Clinic reported that EBD was successful in 95% of the nonfibrotic anastomotic strictures when compared to 45% of the fibrotic strictures. The same investigators also claimed that a hand-sewn anastomosis (12%) seemed to be at an increased risk of developing strictures than stapled anastomosis (4%). Our group has published a series of articles on the endoscopic treatment of pouch strictures, including EBD. , EBD is considered to be an effective therapy in treating a pouch inlet and afferent limb strictures with a pouch failure rate of 5.6%. EBD can safely be performed in patients with pouch strictures, with a low risk for bleeding or perforation. Peritonitis and sepsis are rare.
Risk factors for post-EBD bleeding or perforations have not been systemically studied but possible complications are secondary to EBD in ileal pouches. Purported risk factors for EBD-associated bleeding could be inflammatory or ulcerated strictures, the use of corticosteroids or anticoagulants or antiplatelet agents, and the presence of PSC with portal hypertension and thrombocytopenia.
Bowel perforation can occur after EBD for pouch strictures with a reported frequency of 0.8% in one of the studies. A separate study of 150 patients with pouch strictures undergoing EBD showed the perforation rate to be as low as 0.46% (2 episodes) and bleeding to be 0.98% (4 episodes). It has been reported that CD-associated strictures or strictures in redo-ileal pouches are at risk of causing potential complications during EBD. The presence of a fistula adjacent to targeted stricture can also pose a risk to adverse events. The main disadvantage of EBD (as compared with ESt or ESTx) is the endoscopist’s ability to control the location and depth of the tear. Purported risk factors for perforation include multiple angulated strictures, inflammatory or ulcerated strictures, and the use of systemic corticosteroids. The risk for complications is lower if EBD is performed by a well-trained endoscopy interventionist in treating ileal pouch strictures.
Endoscopic Balloon Dilation in Non-IBD Colorectal Diseases
The non–IBD-related colonic strictures, such as NSAIDs-related, ischemic strictures, and diverticulitis-associated can be treated with EBD. The strictures can be short or long, ulcerated, and single or multiple. Postischemic strictures can be caused by mesenteric vein thrombosis, ischemic enteritis/colitis, abdominal trauma, or iatrogenic injury secondary to inadvertent manipulations of abdominal vessels during bowel surgery. In addition, comorbid cardiac conditions with low ejection fraction, morbid obesity, uncontrolled diabetes, chronic kidney disease (CKD), or coagulopathies may contribute to compromised blood flow to vulnerable segments of the colon triggering the formation of ischemic strictures. These strictures are normally present as circumferential segmental ulcers. In non-IBD strictures, the poor technical skills of endoscopists or the presence of previous surgery for cancer may be associated with a high complication rate from EBD. Strictures of the upper rectum may develop after abdominal aortic aneurysm repair with successful treatment of EBD without bleeding or perforation. Benign anastomotic strictures are evident in 22% of patients with colonic or rectal resection. These anastomotic strictures can also be treated with EBD. EBD in this setting also carries the risk of bleeding and perforation.
ENDOSCOPIC BALLOON DILATION IN COLORECTAL CANCER
Endoscopic balloon dilation of malignant stricture in the large bowel may have a role as a temporary measure. Since anastomotic strictures are common after colectomy and ICA, colocolonic anastomosis, colorectal anastomosis, and anorectal stricture. EBD can be safely performed. It was reported that EBD is safe with a success rate of 87% to 92% with short- and long-term benefits. , Two randomized controlled trials mentioned intralesional injection of a long-acting corticosteroid at the stricture site to improve the efficacy of dilation or space out the need for redilatation showed conflicting results. , Endoscopic submucosal dissection of neoplastic large bowel lesions may result in the formation of strictures, especially in those having >90% circumferential resection by ESD. These strictures can be dilated using EBD. Placement of a self-expandable metal stent (SEMS) is an alternative to EBD for malignant strictures of the large bowel as a palliative measure or a bridge definitive surgery. Stent migration and perforation can occur after the deployment of SEMS. The reported frequency of perforation from SEMS ranged from 1% to 17%.
Bleeding and perforation can occur in patients undergoing EBD for malignant or benign anastomotic stricture. The risk for perforation and subsequent surgical intervention is higher in those who require multiple EBD. For strictures refractory to EBD, alternative endoscopic approaches, such as ESt and placement of SEMS, may be attempted.
ENDOSCOPIC STRICTUROTOMY
Endoscopic stricturotomy (ESt) with needle knife stricturotomy (NKSt) or an insulated-tip (IT) knife has become a widely utilized endoscopic intervention treating IBD- and other colorectal disease-associated strictures. The need for repetitive EBD treatments and subsequent surgical interventions in managing refractory strictures prompted the development of endoscopic electroincision therapy. The advantage of ESt over EBD is that skilled endoscopists can pick up precise locations and control depth to cut at the targeted stricture during electroincision. Clear identification of anatomical landmarks is crucial while performing ESt, especially at the distal rectal, anorectal, or IPAA strictures to avoid iatrogenic trauma to the anal sphincter or a vaginal wall. Similar to EBD, it is important to have abdominal imaging before ESt to delineate the characteristics of the targeted strictures.
Both EBD and ESt have been used for the treatment of anastomotic strictures in CD. We performed a historical cohort study of the efficacy of EBD versus ESt in treating anastomotic strictures in CD and found reported symptomatic and endoscopic improvement in both groups. The strictures were tight and nontransversable to the endoscope with a length of 1.5 cm to 2.0 cm. Most of these strictures were present at the neo-terminal ileum. The subsequent bowel resection after stricture therapy with the ESt group was seen in 9.5% of CD patients as compared to 33.5% in the EBD group. Subsequent surgery was needed in 2/21 (9.5%) patients in the ESt group and 55/164 (33.5%) in the EBD group ( P = .03), during a median of 0.8 (interquartile range [IQR]: 0.1–1.6) year and 4.0 (IQR: 0.8–6.9) years, respectively. We also reported a case series of ESt in successfully treating ileal pouch strictures refractory EBD in 85 patients. ESt can be formed in IBD- as well as non–IBD-associated anastomotic strictures. Our previous study showed comparable outcomes of ESt in IBD and non–IBD-associated (colorectal anastomotic strictures). The non-IBD patients seemed to have an increased need for subsequent surgery as compared to the IBD group after ESt. Importantly, bleeding was seen in 5/106 procedures (4.7%) in the IBD group after ESt versus 0/27 procedures (0.0%) risk in non-IBD patients. No perforation was reported in both groups. In a separate study, ESt was performed in 57 patients with benign anastomotic strictures from colectomy for CRC, and immediate technical success was achieved in 84% of the patients. In the mean follow-up of 31.3 ± 15.8 months, postprocedure benign anastomotic stricture recurred in 11 patients after initial successful endoscopic stricturotomy; 10 of them required salvage surgery.
Techniques
Techniques of ESt to treat primary or anastomotic strictures in non-IBD and IBD conditions involve the use of NK or IT knife. Unlike the replacement of tissues in EBD or stenting, ESt can remove strictured tissue. ESt is set at ERCP endo-cut mode in the ERBE machine (ERBE USA Inc, Marietta, Georgia). Endoscopic electroincision consists of two modes, endoscopic stricturotomy (ESt) and endoscopic strictureplasty (ESTx). The latter needs placement of endoclips as spacers in between the eletroincised strictures. Electroincision is performed in a circumferential, horizontal, or radial fashion. , Radial cut is reserved for patients with anorectal or distal rectal strictures or strictures >2 cm. Longer strictures (>2 cm) may require a combination of EBD and electroincision. The horizontal or radial cut is performed in patients with shorter (≤2 cm) strictures with or without placement of endoclips. Usually circumferential cut creates larger areas of ulcers than a horizontal or radial cut, thus carrying a higher risk for bleeding.
Common Complications of Endoscopic Procedures
Endoscopic electroincision leads to a higher risk of bleeding and a lower risk of perforation than EBD. Notably, adverse outcomes due to ESt, such as perforation requiring hospitalization, were noted in 0.4% (one patient) while excessive bleeding was seen in 3.3% (N = 9) patients. However, few patients with bleeding or perforation required hospitalization, blood transfusion, or surgical intervention. The subsequent surgery rate for disease-associated complications was reported to be 15.3% at 25 years. Many patients with IBD- or non–IBD-associated stricture still require surgical for more definitive therapy.
Risk Factors for Complications
The high risk of bleeding during and after ESt has hindered the wide application of stricture therapy outside expert endoscopists. The common pattern of post-ESt bleeding is delayed onset, which is likely attributed to the ulcer area created by electroincision. We feel that the setting of electroincision and cauterization may play a limited role in the prevention of bleeding. We believe that the use of a sharp knife (such as NK) instead of a dull knife (such as IT knife, Olympus Medical Co., Tokyo, Japan) with the same setting on the ERBE machine may induce less thermal injury and shallower and smaller ulcers, then reducing the risk of bleeding. Radial cut with the placement of clips as in ESTx has reduced the risk for postprocedural bleeding as compared with circumferential cut in ESt without placement of endosclips.
It appears that tissue ischemia plays an important role in the development of postprocedure bleeding. Tissue ischemia also exerts a key role in the primary and anastomotic strictures in IBD or non-IBD patients. Thermal injury during electroincision in stricture therapy further exacerbates tissue ischemia. Additionally, obese patients or patients with a short mesentery may be predisposed to the formation of ischemic strictures. Strictures in ischemic colitis often occur in the “watershed” zone, such as splenic flexure and rectosigmoid junction. Primary or anastomotic strictures in IBD or other benign colorectal strictures can occur anywhere. The strictures can be short or long, ulcerated or nonulcerated, and with or without prestenotic dilation. The increased length of strictures, location of strictures, such as ileocolonic versus neo-terminal ileum strictures in addition to longer duration of CD to the first abdominal surgery, and/or concurrent evidence of significant inflammation, for example, cuffitis or pouchitis in ileal pouch patients can predispose these patients to ESt-related complications , ( Table 3.4 ). In general, EBD may be avoided in patients with ulcerated stricture to reduce the risk of bleeding or perforation. Ulcerated strictures may be carefully treated with ESt in a radial fashion with the location and depth of intended treated strictures being carefully selected. The risk for perforation is higher in EBD higher than in ESt. It appears that the performance and outcome of salvage endoscopic procedure for perforation may be easier in ESt-associated perforation than in EBD-associated perforation. Extreme caution should be exerted when performing ESt or ESD in the strictures with underlying inflammation and ischemia. Radiation should be cautiously used in radiation enteritis in CRC patients, since radiation leads to fibrotic and ulcerated strictures that may perforate easily. CRC can cause stricture formation as well with subsequent obstruction in the large colon.
Authors | Sample Size and Study Design | Risk Factors Reported |
---|---|---|
Lan et al. | Retrospective (n = 200) | Increased length of strictures; concurrent pouchitis/cuffitis. |
Zhang et al. | Cohort (n = 74) | Strictures with IBD than non-IBD conditions. |
Lan et al. | Cohort (n = 85) | Advanced age, corticosteroid, and antitumor-necrosis use, location of strictures (ilecocolic anastomosis, neo-terminal ileum, or pouch inlet). |
Lan et al. | Comparative (n = 185) | Radiographic prestenotic bowel dilation and longer duration from CD diagnosis to the first surgery. |
Chronic bowel inflammation, healing, and fibrosis, such as CD and UC increase the risk for the development of malignant strictures. For example, the frequency of bowel cancer in CD patients with strictures was higher (6.8%) than in those without strictures (0.7%). Patients with partial colectomy and anastomosis for CRC also carry the risk of the formation of malignant strictures. Therefore, it is recommended that all patients underlying the index therapeutic endoscopy for bowel strictures should have a biopsy of the stricture area and subsequently yearly biopsy.
Fecal diversion with ileostomy or colostomy is formed for the treatment of refractory inflammation or diseases in the distal bowel or perianal diseases or the protection of anastomosis of the downstream bowel. Fecal diversion can exacerbate anastomotic strictures in the distal bowel and the creation of a stoma may result in diversion-associated primary strictures. Diversion-associated primary or anastomotic strictures can be friable and endoscopic biopsy, EBD, or ESt is often associated with onsite bleeding or bacterial translocation. However, ESt may carry a lower risk for bleeding than mechanical EBD or bougie dilation. We reported a case report in which sealed distal rectal stricture in a 27-year-old female with a severe CD and a diverted large bowel was effectively treated with CT-assisted ESt with symptomatic relief and no complications. Long-standing fecal diversion in patients with IPAA can also result in diversion-associated primary or anastomotic strictures. We described a case in which stricture in a sealed diverted pouch was treated with wire-guided ESt.
The authors noticed a growing number of patients with anastomotic strictures refractory ESt as well as EBD who had prolapse bowel proximal to the anastomosis. We believe that prolapsed proximal bowel is a contributing factor to the development of strictures. The authors have performed a combined ESt and septectomy in a 68-year-old female with prolapsed stricture at an ileal pouch. The combined endoscopic therapy may carry a higher risk for bleeding or perforation than mono-endoscopic therapy.
SURGERY FOR CROHN’S DISEASE
Crohn’s disease is a chronic inflammatory process involving any parts of the GI tract from the mouth to the anus. Common CD-associated complications include strictures, fistulas, abscesses, malnutrition, and anemia. The availability of various medical, endoscopic, and surgical therapies has drastically changed the disease course of CD. Approximately 80% of patients with CD would eventually need surgery for medically refractory inflammatory diseases or disease-associated complications, such as severe bleeding, refractory strictures, fistulizing disease, bowel perforation, intra-abdominal or perianal abscess, peritonitis, and malignancy. CD predominantly affects the terminal ileum but can involve any part of the GI tract. Approximately 75% of patients with CD develop eventually develop strictures and one-third of CD patients develop bowel fistulas. Common locations of fistulas in CD are the ileocecal valve, distal ileum, anorectal ring, colonic, and duodenum cap. Single, short, and endoscopically accessible strictures may be treated with EBD, ESt, and ESTx. Multiple, long, or complex strictures (e.g., strictures with fistulas and abscesses) usually require surgical intervention. Fistulas in CD can be ileoileal, ileocecal, ileosigmoid, colorectal, enterocutaneous (ECF), and rectovaginal (RVF). Most complex fistulae with or without abscess, peritonitis, or bowel perforation poorly respond to medical or endoscopic therapy and eventually require surgical intervention.
Patients with Crohn’s colitis, likely UC, may present with fulminant colitis or toxic megacolon, free perforation, or intra-abdominal abscess that warrant emergent surgery. In addition, Crohn’s colitis moreover elicits a 4 to 20 times increased risk of CRC as compared to the general population. The presence of colonic strictures may make surveillance colonoscopy difficult. All colonic strictures in CD should be biopsied at least once a year. EBD or ESt of colonic stricture creates access for surveillance colonoscopy in the proximal bowel. Symptomatic, multiple, long colonic strictures or neoplastic strictures require colectomy.
Perianal disease is common in CD. Patients often presented with complex perianal fistulas or RVF. We also notice that a significant number of patients with perianal fistulas or RVF have concurrent strictures at the anorectal ring, making surgical treatment challenging.
Preoperative diagnosis and characterization of disease phenotype, extension, and complications with cross-sectioning imaging, such as CT, CTE, MRI, and MRE, is required. Some patients may require preoperative or intraoperative colonoscopy for the diagnosis and assessment of disease activity and disease extent. Some patients may benefit from contrasted enemas via the anus or stoma.
Postoperative recurrent CD is common and adequate prophylactic medical therapy is critical for the prevention of disease recurrence and further surgery. Routine ileocolonoscopy is recommended for disease monitoring and treatment of anastomotic strictures, if present.
Techniques
Surgical options for strictures or fistulas in the bowel include bowel resection (e.g., ileal resection, ileocolonic resection, and partial colectomy) and anastomosis, surgical strictureplasty, bypass, and fecal diversion with an ileostomy or colostomy. For patients undergoing subtotal colectomy and Hartmann procedure at risk for stump leaks, mucus fistulas may be constructed.
The surgical management of Crohn’s colitis normally involves partial colectomy with colo-colonic anastomosis or colorectal anastomosis. In highly selected patients with CD only limited to the colon, sparing the small bowel and perianal area, total proctocolectomy (TPC) with IPAA has been performed similarly to that for UC.
Surgical management of perianal CD includes modalities of incision and drainage of abscesses, fistulotomy, seton placement, and proctectomy with fecal diversion. Topical stem cell therapy is safe and effective for the treatment of perianal fistulas from CD.
Common Complications of the Surgery
Common surgical complications in CD surgery are postoperative ileus, anastomotic bleeding, acute or chronic anastomotic leaks (AL), abscess, sepsis, fistula, sinus, and anastomotic strictures. AL or strictures can also occur in the staple or suture lines, diverted bowel. Patients with multiple sessions of small bowel resections carry the risk of short gut syndrome or malabsorption. Acute AL can lead to potential abscess and sepsis. On the other hand, chronic AL can trigger the formation of a fistula or sinus at the anastomosis. One of the largest historical cohort studies from the Cleveland Clinic (150/2959 CD patients) reported 8.7% of frequency in early-onset pelvic sepsis, whereas late-onset pelvic sepsis secondary to AL was 3.3%. Interestingly, the late onset of AL in CD patients was 2.0%. It seemed that patients with CD manifested a frequent risk of developing early- and late-onset postoperative complications at the anastomotic sites as compared to UC, indeterminate colitis, or familial adenomatous polyposis (FAP).
Anastomotic strictures are common after bowel resection and anastomosis or surgical strictureplasty. Common locations of strictures are ileoileal, ICA, IRA, colocolonic, or colorectal anastomoses. In patients with surgical strictureplasty, strictures often occur at the inlet or outlet. Fecal diversion can also cause strictures in the downstream bowel.
Risk Factors for Complications
Multiple risk factors have been identified for postoperative complications of CD ( Table 3.5 ). Some of the risk factors are modifiable, such as smoking, malnutrition, anemia, and obesity. Malnutrition and anemia reflect poorly controlled inflammation or disease. Elevation of inflammatory markers, such as CRP (C-reactive protein >40 g/L) may predict the risk of postoperative complications. Zuo et al. reported that high CRP was associated with the risk of postoperative intra-abdominal sepsis. The study showed reported 11.34% of intra-abdominal septic cases in which AL was seen in 4.65% of patients with preoperative CRP at >10 mg/L.
Authors | Sample Size and Study Design | Risk Factors Reported |
---|---|---|
Zager et al. | Retrospective (n = 121) | Lower psoas muscle area; higher count of white blood cells. |
Ge et al. | Retrospective (n = 315) | Preoperative hypoalbuminemia. |
Galata et al. | Retrospective (n = 230) | Low preoperative skeletal mass index. |
Zhu et al. | Retrospective cohort (n = 949) | Upper GI involvement. CRP ≥40 mg/L. Albumin level <30 g/L. Preoperative use of infliximab or glucocorticoids. |
Galata et al. | Retrospective (n = 126) | Preoperative elevation in CRP and hypoalbuminemia; emergency surgery. |
Fumery et al. | Prospective (n = 209) | Preoperative use of corticosteroids. |
Waterland et al. | Meta-analysis (n = 5425) | Preoperative use of anti-TNFα therapy. |
Li et al. | Prospective (n = 498) | Malnutrition. |
Heimann et al. | Prospective (n = 130) | Preoperative low serum albumin; extensive bowel resection can lead to sepsis; permanent or temporary ileostomy. |
Malnutrition and hypoalbuminemia are known factors for postoperative perioperative morbidity and mortality in CD. It is estimated that 65% to 80% of CD patients had malnutrition or failure to thrive. Preoperative optimization of nutrition status reduces postoperative complications. A case-matched study showed intra-abdominal septic complications were 4% in CD patients who received preoperative optimization with enteral nutrition versus 25% of patients with a regular preoperative diet. Sarcopenia is considered a marker as well as a byproduct of chronic inflammation that has been shown to have a prognostic value in postoperative complications in IBD. Low psoas muscle area (PMA) as a marker for sarcopenia measured radiographically was found to be associated with postoperative complications in CD. Similarly, preoperative lower skeletal muscle mass index (SMI) also has been shown as an independent risk factor for major postoperative complications in CD. On the other hand, obesity can also increase the risk of postsurgical complications, such as ischemic bowels, delayed recovery, prolonged hospital stay, hospital-acquired infections, or complications due to comorbid diabetic pathology. Excessive visceral fat is associated with a high risk for postoperative recurrence of CD. , There is an elevated risk of deep vein thrombosis (DVT) and pulmonary embolism (PE) in IBD patients perioperatively. Obesity can logically compromise earlier mobility in these postoperative CD patients, which may further increase the risk of venus thromboembolism.
Anemia is another risk factor for postoperative complications in CD. A multicenter study showed that low hemoglobin and iron deficiency were risk factors associated with postoperative complications in CD. , Multiple factors contribute to anemia in CD, including bleeding, poor oral intake of iron or vitamin B12 and other nutrients, and chronic inflammatory status. It was estimated that 0.9% to 6% of CD patients had bleeding due to severe recurrent mucosal inflammation. On the other hand, red blood cells are considered an immunosuppressant, and perioperative blood transfusion was associated with a risk for postoperative complications in CD, similar to other disease conditions. ,
Smoking can lead to adverse surgical outcomes. It has been claimed that 50% of smoking reduction in general within 6 to 8 weeks before surgery can decrease postoperative incisional issues and heart-related adverse events. , Smoking along with longer strictures (>4 to 5 cm) may predispose patients with CD-associated stricture to have surgical reintervention.
The immunosuppressive therapy before surgery can also predispose patients to postoperative surgical complications including intra-abdominal abscess, AL, or sepsis. , The association between the utilization of corticosteroids before surgery and postoperative complications in CD, as well as UC (such as AL and septic complications), is well established. , The frequency of postoperative complications, such as AL was reported as 0% to 12.7%, abdominal abscess at 3.1% to 18.7%, and wound sepsis at the rate of 0% to 18%. For example, stuck et al. reported a 12.7% postoperative infection rate in CD patients with corticosteroid therapy before surgery as compared to 8.0% in the control group. Similar postoperative complications were reported in 209 patients with CD with ileocecal resection who were on preoperative corticosteroid therapy within 4 weeks. Unfortunately, severe complications were reported in 33% of patients, reoperation in 7% of patients, and secondary stoma in 4.5% of patients. There has been a long debate on the adverse impact of the preoperative use of biological agents on postoperative infectious complications. For example, a retrospective study by McKenna et al. showed a frequency of 8% of intra-abdominal sepsis in 30 days post-ICR resection in CD, and a combination of immunosuppression (corticosteroids, immunomodulators, and biological agents) was associated with the risk for intra-abdominal sepsis. A recent prospective cohort study showed that preoperative use of biologics may not increase the risk for postoperative infectious complications.
Emergent surgery has been documented to predispose CD patients to postoperative complications. Laparoscopic surgery has benefits over open surgery with faster recovery, shorter hospital stay, and lower risk for postoperative complications than the open approach, as reported in randomized controlled trials on laparoscopic versus open ICR. , It appears that patients with IRA may have a high risk for AL, disease recurrence, and surgical reintervention. The reported rate of surgical reintervention was 30% in IRA patients with Crohn’s colitis sparing rectum. Surgical redo resection of bowels was still associated with a higher risk of AL (17%) than no prior intestinal resection (5%) in CD. O’Riordan et al. published their study suggesting that colectomy with IRA with a spared rectum was a relatively safe and appropriate option for Crohn’s colitis. This study reported 7.4% of AL, which was not associated with anastomotic techniques (stapled versus handsewn) or one-stage versus two-stage surgeries. Interestingly, no AL was found in a defunctioning ileostomy. Extensive bowel resection has been shown to lead to sepsis.
The impact of the type of anastomotic configuration on postoperative CD recurrence and complications has been explored. Shimada found a lower rate of AL (5.1%) in the Kono-S anastomosis (KSA) group than that (17.3%) in the end-to-end anastomosis (EEA) group. Another nonrandomized study in CD reported that hand-sewn EEA after ileocecal or ileocolonic resection could lead to a higher rate of AL (14.1%) than those (2.0%) with side-to-side stapled ICA. Some investigators felt that over-sewing the staple line in EEA might prevent AL. However, Kanazawa et al reported that penetrating CD, hand-sewn anastomosis, and excessive operative time (>180 minutes) significantly increase the risk of intra-abdominal septic complications.
Surgical stricturoplasty is routinely performed in patients with small bowel strictures, and occasionally in ICA strictures. The rectal stricture in CD can be treated with transanal strictureplasty. The bowel resection could have a lesser recurrence rate than stricturoplasty. The morbidity rate after stricturoplasty was reported to be 18% with septic complications in only 5% of patients, whereas the recurrence of CD requiring surgical bowel resection was seen in 34% of CD patients. There are limited data available on stricturoplasty-related AL in Crohn’s colitis.
Anastomotic strictures are common after bowel resection and anastomosis. The disease status, surgical techniques, and the site of anastomosis could affect the development of anastomotic strictures. , The recurrence rate of CD within 1 year was reported to be 72% in one of the endoscopic and histologic studies. This study published that CD recurrence was noticed at the neo-terminal ileum and anastomotic sites in 88% of patients. The anastomosis strictures were found stenosed, and rigid with large ulcers approximately within 3 to 10 years of the postsurgical period in CD patients. The coordination between patients, gastroenterologists, interventional endoscopists, and CORS is imperative in delivering the best patient-centered quality care to CD patients. The multidisciplinary approach to managing CD would more likely reduce the postsurgical complications and recurrence rate of CD.
SURGERY FOR ULCERATIVE COLITIS
Surgical options for UC have been evolving past decades. Approximately 15% to 25% of UC patients may experience flare at any time. It is estimated that 15% to 20% of patients with UC would eventually need colectomy for medically refractory disease or CAN. Urgent colectomy is warranted in UC patients with medically refractory disease (70%), toxic colonic dilatation (20%), colon perforation (<10%), and severe bleeding (<5%). Approximately 5% of patients with UC would develop CAN. A meta-analysis study estimated that UC patients had an 18.4% risk of developing CRC after 30 years of their diagnosis. Another meta-analysis showed that the risk of CRC development in Asian UC patients increases only after 10 to 20 years.
Total proctocolectomy with Brooke end ileostomy was originally established as a standard of surgery until the late 1960s when Kock or K-pouch was introduced. Parks and Nicholls then designed a pelvic S-pouch in 1978. Restorative proctocolectomy (RPC) with IPAA became the surgical treatment of choice for patients with UC or FAP who require colectomy. The commonly fashioned configurations are J- and S-pouch.
Techniques
The standard surgical procedure to manage UC is open or laparoscopic RPC with IPAA or continent ileostomy. RPC with IPAA or continent ileostomy can be performed in one, two, or three stages. The anastomosis can be handsewn or stapled, with or without mucosectomy. Mucosectomy is routinely performed in those with CAN or FAP. RPC and IPAA are also performed in highly selected patients with Crohn’s colitis in the absence of perianal or small bowel diseases. Occasionally colectomy with IRA is constructed in patients with UC with relative rectal sparing. Extreme caution should be taken to identify and preserve the ureter, pelvic nerves, pelvic structures, and retroperitoneal organs during the operation.
Common Complications of the Surgery
Postoperative complications are common in patients with RPC and pouch surgery. The International Ileal Pouch Consortium published a series of guidelines for the diagnosis, classification, and management of pouch disorders. , , The complications were divided into five categories: (1) structural (such as acute and chronic AL and anastomotic strictures); (2) inflammatory (such as pouchitis, cuffitis, and CD of the pouch); (3) functional; (4) neoplastic; and (5) systemic and metabolic. Here we mainly discussed structural complications of RPC and ileal pouches. The structural complications mainly have two subcategories, intrinsic (such as anastomotic strictures) and intrinsic (such as floppy pouch complex [FPC]) obstruction, acute (e.g., abscess and sepsis) and chronic (e.g., fistula and sinus) leaks.
Acute AL can happen within 6 months of stoma closure and lead to potential abscess formation and sepsis. A large cohort study from Cleveland Clinic reported early-onset AL to be at 4.8% and late-onset AL at 1.5%, whereas frequency of early-onset pelvic sepsis at 6.3% and late-onset pelvic sepsis at 2.7% in UC patients with IPAA. The first occurrence of AL could potentially happen at the suture line of the Hartmann pouch after subtotal colectomy. On the other hand, chronic LA develops >6 to 12 months of poststoma closure and can lead to a fistula or sinus in the ileal pouch. The vulnerable sites for AL in the pouch are located at suture lines, the tip of the “J” to the pelvis, and/or tract from the anastomotic site to the presacral space or vagina. Therefore, a mucus fistula is constructed with a long Hartmann pouch and suture line to be placed at the fascia level. UC patients with left-sided severe colitis or even proctitis may benefit from a mucous fistula. Thus, the leak from an intentionally constructed mucous fistula results in ECF. Anastomotic leaks can lead to pouch failure in 31% of IPAA cases due to pelvic sepsis, pouch fistulas, pouch sinus, or abscess formation with symptoms of fever, pelvic pain, or anal discharge.
Strictures are another commonly encountered complication after IPAA. The overall incidence rate of pouch strictures in IPAA ranges from 5% to 11%. The most common locations of strictures are the stoma closure site, pouch inlet, and anastomosis, while the most common areas of the leaks are anastomosis, the tip of the “J,” stoma closure site, and vertical staple line. Strictures in other ileal pouches, such as the Kock pouch or BCIR have been found at the skin level, nipple valve, or pouch inlet.
Currently, the main phenotypes of FPC are pouch prolapse, pouchocele, afferent limb syndrome, efferent limb syndrome, and twisted pouch or pouch volvulus. The hand-sewn anastomosis can potentially predispose patients to stricture formation in the pouch. This could happen due to increased mesenteric tension at the anastomotic site. Strictures are frequently seen at the loop ileostomy site, afferent limb, and pouch inlet sites. Other locations for strictures in the J-pouch have also been reported to be at the mid-pouch or by the dome of the J-pouch. Remarkably, strictures in other ileal pouches, such as the Kock pouch or BCIR have been found at the skin level, nipple valve, or pouch inlet.
The structural complications can be detected by MRI, MRE, CT, CTE, or contrasted pouchogram. Suspected AL should be diagnosed earlier and managed promptly to salvage the pouch from potential pouch failure. AL is confirmed and characterized by radiographic, endoscopic, or surgical approaches with a multidisciplinary team.
Risk Factors for Complications
The reported frequency of AL after restorative proctocolectomy and IPAA ranged from 5% to 18%. , Purported risk factors for AL entail male sex, age >40, morbid obesity, severe or fulminant colitis, mesenteric tension at the anastomotic site, bowel ischemia, sarcopenia, the use of corticosteroids, and preoperative hypoalbuminemia ( Table 3.6 ). The presence of >2 cardiovascular diseases, concurrent use of anti-TNF or corticosteroids, higher body mass index (BMI), and a high score from the American Society of Anesthesiologists have recognized potential risk factors for AL. It appears, however, that the preoperative use of anti-TNF agents, vedolizumab, or ustekinumab is not associated with an increased risk for postoperative AL. , Preoperative use of vedolizumab, however, has been shown to increase risk of postoperative ileus. The formation of defunctioning ileostomy may reduce the incidence of AL by 50%. It has been seen that a lack of fecal diversion or an absence of a stoma can lead to postsurgical complications. Emergent surgery or delayed surgical intervention for severe UC can lead to adverse postsurgical outcomes. Total abdominal colectomy has been shown to have greater complication rates including overall morbidity as compared to staged restorative proctocolectomy. Interestingly, hand-sewn pouch-anal anastomosis seemed to have greater adverse outcomes and lower quality of life as compared to patients with stapled IPAA ( Table 3.6 ). Patients with Hartmann pouch are predisposed to suture or staple line leak that could subsequently lead to a pelvic abscess, sepsis, or even nonintentional ECF. The endo-sponge-assisted early surgical closure had been found safe and effective in controlling pelvic sepsis after pouch leakage.
