Dilation of Strictures in Patients with Inflammatory Bowel Disease

Stricture formation occurs in up to 40% of patients with inflammatory bowel disease (IBD). Patients are often symptomatic, resulting in significant morbidity, hospitalizations, and loss of productivity. Strictures can be managed endoscopically in addition to traditional surgical management (sphincteroplasty or resection of the affected bowel segments). About 3% to 5% patients with IBD develop primary sclerosing cholangitis (PSC), which results in stricture formation in the biliary tree, managed for the most part by endoscopic therapies. In this article, we discuss endoscopic management of strictures both in the alimentary tract and biliary tree in patients with IBD and/or PSC.

Key points

•

Strictures—de novo versus anastomotic, locations, inflammatory versus fibrotic—in inflammatory bowel disease (IBD) are discussed.
•

The pros and cons of dilation, symptoms, confirmation of inflammation with cross-sectional imaging, and j pouch and anal-pouch strictures are discussed.
•

The technical details, size of balloons, use of fluoroscopy of dilation are presented.
•

Other adjuncts procedures include steroids, needle knife, and stent.
•

About 10% to 15% patients with IBD develop primary sclerosing cholangitis, which results in stricture formation in the biliary tree.

Introduction

Patients with Crohn’s disease may develop strictures either before surgery or after surgery. We will discuss when and how to dilate strictures in IBD patients. We will discuss endoscopic retrograde cholangiopancreatography (ERCP) in patients with primary sclerosing cholangitis (PSC) and the proper evaluation and management of strictures in this condition.

Transmural inflammation is a hallmark of Crohn’s disease, which leads to stricturing of the bowel wall. These strictures may be inflammatory, fibrotic, or, most often, a combination of both. In patients with Crohn’s disease, around one-third will develop strictures within 10 years of diagnosis. These strictures occur most commonly at the colon, ileocolonic region, and ileum and are termed de novo strictures. Also, the cumulative risk of surgery for bowel resection within 10 years of diagnosis is around 50%. In these patients, recurrence of disease occurs in 44% to 55%, which is often associated with restenosis at the anastomotic site. At times, anastomotic strictures can occur owing to altered blood supply at the level of the anastomosis, and these are usually not associated with active inflammation.

In ulcerative colitis, rates of colectomy in more recent cohorts in the era of biologics is around 10% to 16.6%. Among these patients, anastomotic strictures can form at the pouch–anal anastomosis and rarely at the pouch–inlet anastomosis. In about 5% to 10% of colitis, indeterminate colitis is diagnosed. Both classic ulcerative colitis patients and indeterminate colitis patients can progress to Crohn’s disease of the ileal–pouch–anal anastomosis, often manifesting with strictures in the pouch and prepouch ileum.

Strictures in IBD occur as a result of scarring and fibrosis in the setting of chronic inflammation in the bowel wall. As a result, most strictures demonstrate a combination of fibrosis and active inflammation. Endoscopically, active ongoing inflammation is manifest as ulcerations and edema at the narrowed site. Radiographically, too, active inflammation can be demonstrated in a strictured segment both on computed tomography enterography and MR enterography. In a lesser number of situations, the stricture is almost completely fibrotic without any evidence of active inflammation. The presence of active inflammation endoscopically is an important factor in dilation of these strictures because it can limit the extent of dilation that is performed. Actively inflamed bowel wall is more friable and more likely to result in a complication of perforation.

Strictures in IBD patients have diverse manifestations. When they affect the gastric outlet or duodenum, patients can present with symptoms of gastric outlet obstruction including nausea, vomiting, and an inability to eat. When they affect the small bowel, patients usually present with intermittent partial small bowel obstruction symptoms including abdominal pain, abdominal distention, obstipation, and vomiting. At times, the symptoms may be more subtle, with just postprandial abdominal cramping, especially a few hours after intake of high residue foods. Colonic strictures are the most well-tolerated, and hence often present very late. Patients complain of change in the caliber or stool, and at times increasing constipation. When the stricture is at the pouch–anal anastomosis, tenesmus, and a feeling of incomplete evacuation are common symptoms. Also, excessive straining and urgency with watery stool can occur in this scenario.

Imaging of the strictures is very important before attempting endoscopic therapy. This is both to delineate the location, extent, and possibly diameter of the stricture, and also to document the presence and extent of concurrent inflammation. Ideal imaging is cross-sectional in the form of a computed tomography enterography or MR enterography. Both of these techniques are able to provide a fairly precise location of the stricture, and an assessment of degree of inflammation. Small bowel series have been used in the past, but cannot provide any information on inflammation within the bowel wall. Barium enemas can be useful in colonic and pouch strictures because the sensitivity of computed tomography and MR is lower. They can outline the length and location of strictures well, and should be considered before therapy if this information is unknown. Imaging, both cross-sectional and contrast radiography, can also help to identify other complications of IBD including fistulae, abscesses, or sinus tracts, which can occur in conjunction with stricturing disease.

When to Dilate

Strictures in IBD should be treated if they are symptomatic, as described. If active inflammation is present, medical therapy should be optimized first. In the case of anastomotic strictures, even if the patients are not overtly symptomatic, endoscopic dilation should be performed if the endoscope is unable to pass through the stricture, to allow for examination of the more proximal bowel and to prevent symptoms from developing down the road.

How to Dilate

The first principle in endoscopic dilation is choosing the correct endoscope to reach the site of the stricture. Upper endoscopes are used for gastric and proximal duodenal strictures. For distal jejunal and proximal jejunal strictures, a push enteroscopy using the pediatric colonoscope is usually necessary. Often times, however, in these patients, the bowel can be fairly angulated, and we recommend using the more flexible enteroscope (either with single- or double-balloon technology) to reach these sites. In the colon and for ileocolonic anastomotic strictures, the pediatric colonoscope is usually the best choice given its flexibility. Here too, if the ileocolonic anastomosis is a side-to-side anastomosis and fairly angulated, use of an enteroscope may facilitate access to and passage through this site. Ileal strictures almost always require an enteroscope to reach the site. Strictures in the left colon can often be sharply angulated especially in the sigmoid colon or at colorectal anastomoses. Here, the more flexible upper endoscope is favorable.

Dilation is performed using graduated, through-the-scope balloon dilators, except at the anal verge, which is discussed elsewhere in this paper. If the exact diameter of the stricture is unknown beforehand, it is advisable to use a suite with fluoroscopic capability that can be used if necessary. Once the stricture is identified, if the endoscope is able to traverse it, the balloon can be advanced beyond the stricture, and then pulled back to straddle the stricture before inflation. If the stricture cannot be traversed with the scope, then the balloon should be advanced across the stricture. This is safely done if there is minimal surrounding inflammation and no resistance to balloon passage is encountered. If the stricture seems to be long or if any resistance is encountered, then the dilation should be undertaken over a guidewire. The guidewire can be passed through the stricture under fluoroscopic guidance to ensure the presence of the wire in the bowel lumen past the stricture. Over the guidewire, the balloon can then be advanced across the stricture.

The choice of the diameter of the balloon depends on the initial diameter of the stricture, the presence or absence of ulcers and inflammation, and the previous maximal diameter achieved if this is a repeat dilation. The diameter of the stricture can be estimated based on the diameter of the endoscope being used as these diameters are readily available. Another trick is to use the diameter of a known device such as an open biopsy cable as a reference to guess the approximate stricture diameter. The initial balloon diameter chosen should be a little larger than the estimated baseline diameter of the stricture. The balloon can be inflated with water, saline, or water-soluble contrast. The duration the inflated balloon should be held in position is not well-defined. It can vary from 30 seconds to 2 minutes. If fluoroscopy is being used, the balloon can be held in place until effacement of the waist is seen. Once the balloon is deflated, it should be withdrawn and the stricture site examined for mucosal tearing, and any complication including persistent bleeding or perforation. Adequate mucosal tearing ( Fig. 1 ) indicates disruption of the mucosal layer of the gut lining, but not beyond the submucosal layer. If there is inadequate mucosal tearing, further dilation with the next larger balloon diameter should be carried out. It is not recommended to dilate more than 3 to 4 mm beyond the initial diameter of the stricture, especially if there is active inflammation at the site, unless there has been prior dilation of the same site to a higher diameter and it has now repeatedly scarred down. Even so, great caution should be used. Wire-guided hydrostatic dilation balloons (eg, CRE, Boston Scientific Corporation, Natick, MA) are also safer to use than non–wire-guided balloons, especially around angulated strictures, because they can guide the more distal tip of the balloon to remain within the bowel lumen and avoid perforation of the viscus wall beyond the stricture by the distal balloon tip. Another suggestion is to examine repeatedly the patient’s abdomen to watch for distention, which may be a sign of a perforation. Maximal balloon diameter is usually 18 to 20 mm, although balloons as large as 25 mm have been described in studies. These larger diameter balloons have been associated with a definite higher risk of perforation without any evidence to suggest increased efficacy. Hence, we do not recommend dilating strictures to more than 18 to 20 mm.

Sixteen percent of patients undergoing ileal pouch–anal anastomoses develop stricturing at the pouch–anal site. These strictures can be digitally dilated if they are soft and pliable. However, once they become more fibrotic, they can be dilated during endoscopy using either a graduated balloon as described above or Hegar’s dilators, up to 18 mm in size.

Success of Endoscopic Balloon Dilation

Multiple studies over the years have documented the efficacy and safety of endoscopic balloon dilation in Crohn’s disease strictures. One of the largest of these was published by Gustavsson and colleagues in 2012. They reported the outcomes in 178 patients who underwent a total of 776 dilations. A stricture was defined as inability to pass the endoscope, and 81% of these patients were reported symptomatic. Eighty percent of the strictures were anastomotic (77% at an ileocolonic anastomosis). The maximal balloon dilation was 18 to 20 mm in the majority, but in 31% patients, a 25-mm balloon was used. Technical success was seen in 89%, with a complication rate of 5.3% (perforation was commonest). Of note, a significantly higher rate of complications (9.3%) was seen with use of the 25-mm balloon compared with 3.5% with balloons under 20 mm. Median follow-up time for a subset of 125 patients was 12 years. No further intervention or one additional dilation was needed in 80%, 57%, and 52% at 1, 3, and 5 years after the first dilation. Surgery-free survival did not differ between anastomotic versus de novo strictures.

The second largest study was published by Van Assche and colleagues in 2010 where they reported the long-term efficacy of dilations in 138 patients followed for a median of 5.8 years; clinical results were robust and outweighed the risks of complications. Anastomotic strictures made up 84% of the cohort and ileal location was seen in 85%. All strictures were reported to be less than 5 cm in length, and a standard 16- to 18-mm balloon was used in all cases. The starting diameter of the strictures in the study was not described. This is a unique technique because usually we recommend starting dilation with a balloon that is slightly larger than the approximate stricture diameter, rather than using a standard diameter balloon for all strictures regardless of baseline lumen diameter. Immediate success rate of the dilation defined as ability to pass the colonoscope through the stricture was very high at 97.1%. No further dilation was needed in 54.3% of patients and 76.1% had no need for surgery after the first dilation. The complication rate was 5.1%, with 6 out of the 12 complications being perforation. It can be hypothesized that this complication rate could have been decreased if the starting balloon diameter was chosen based on the stricture diameter, rather than using a standard 18-mm balloon for all cases. Nevertheless, the success rate and surgery-free survival is impressive, and the authors also found that neither active disease nor medical therapy could predict the need for recurrent stricture dilation.

Atreja and colleagues in 2014 reported outcomes in 128 patients with 169 Crohn’s related strictures with a median follow-up of only 1.8 years. This cohort was unique because 52.1% of the strictures were de novo and only 47.9% were anastomotic. The median stricture length was 2 cm, and mean balloon size used was 16.6 mm. Active inflammation was seen in 60.1% of the strictures. Also unique was that 54.4% of these strictures could be traversed by the endoscope even before dilation. Redilation was performed in 58.6% patients, but the median number of redilations was only one. However, because the median follow-up was only 1.8 years, this study does not document long-term results. Complication rate was low at 3.1% per patient and only 0.93% per procedure, and 3 out of the 4 complications seen were perforations requiring surgical resection.

There have been 2 systematic reviews with metaanalyses published within the last year describing the efficacy of endoscopic balloon dilation for Crohn’s disease strictures. Morar and colleagues included 25 studies with 1089 patients and 2664 dilations. Mean age at first dilation was 41 years. The follow-up times varied from a median minimum of 4 months to a median maximum of 83.5 months. Across 13 studies, 79.1% were anastomotic strictures and only 3.8% were upper GI strictures (but this was estimation because of missing data in 9 studies). A symptomatic response rate of 70.2% was shown but technical response was reported in 92.6% patients. No difference was seen on subgroup analyses between anastomotic versus de novo strictures or active versus quiescent inflammation at the stricture site, but there was heterogeneity in the reported data. The pooled complication rate was calculated at 6.4%, but perforation rate was 3%.

Navaneethan and colleagues reported on 24 studies including 1163 patients with 1571 strictures. Mean age of the patients was similar; mean stricture length was 1 to 4 cm, and median follow-up time was 15 to 70 months. Ileocolonic anastomotic strictures were the majority (69%), and 31% were de novo strictures. Overall technical success was 89%. A single dilation was effective in 44% patients, and in 27%, surgical intervention was required either for failure of dilation or complication. Complication rate was found to be 4%, with perforation in 3%. Subanalyses showed that though success of endoscopic balloon dilation seems to be higher in anastomotic strictures compared with de novo strictures, this was not statistically significant. And, although only 5 studies had comparative data, they found a stricture length less than 4 cm had a significantly decreased risk of surgical intervention when compared with greater than 4 cm.

Krauss and colleagues published an interesting and unique report in 2014 where they included 88 Crohn’s disease patients with strictures and compared those with only endoscopic therapy, only surgical therapy, endoscopy followed by surgery and surgery followed by endoscopy. They found that endoscopic therapy was applied in 20 patients, of whom 9 required only a single dilation but the rest required up to 5 dilations. Stenoses in the surgical group had an average length of 6.5 cm compared with the endoscopic group which was 3.0 cm. Endoscopic balloon dilation also worked more effectively in fibrotic strictures, which were shorter (mean, 4.0 cm) than inflammatory strictures (mean length, 7.5 cm). In the surgical group, 35% developed new stenoses, of which 75% were treated with endoscopic balloon dilation. Also, smoking was found to be a risk factor for recurrent stenosis.

A recent study by Sunada and colleagues reports on the long-term outcomes of 321 double balloon enteroscopy-assisted dilations of Crohn’s strictures in 85 patients. Seventy-one percent were antegrade enteroscopies, the mean balloon diameter at first dilation was 12.4 mm, mean follow-up period was 41.9 months, and surgery-free rate after the initial double balloon enteroscopy-assisted endoscopic balloon dilation was 87.3% at 1 year, 78.1% at 3 years, and 74.2% at 5 years. Complication rate was 1% overall and perforation rate was 0.8% per procedure and 4.7% per patient.

The success of endoscopic dilation of ileal pouch–anal anastomosis strictures was published in a report by Shen and colleagues where 646 strictures in 150 patients were dilated during 406 pouchoscopies. Median stricture length was 1 cm and median balloon size was 20 mm. Two perforations occurred. In a median follow-up period of 9.6 years, 87.3% patients were able to retain their pouches.

Other Adjunctive Techniques

Intralesional injection of a long-acting corticosteroid has been used as an adjunct to endoscopic balloon dilation to help maintain the efficacy of the dilation owing to a proposed local antiinflammatory effect. The most commonly used agent is triamcinolone, which is injected in a dose from 40 to 100 mg usually in 4 to 5 aliquots at different sites within the stricture. This injection is performed after maximal dilation. The data regarding the efficacy of intralesional steroid injection is conflicting. Some studies did not find any beneficial effects but other small studies suggested a favorable effect of prolonging the need for further dilation or surgery. Similarly, 2 small case series have described the use of intralesional infliximab to treat strictures, but these data are not robust enough to support widespread use of the agent in this application.

Another adjunctive treatment modality for strictures is needle-knife strictureplasty, wherein a needle-knife from the pancreaticobiliary endoscopy armamentarium is used as an adjunctive technique to perform multisite monopolar electroincision of refractory fibrotic strictures. However, this technique has only been described in a few case reports and should not be performed routinely owing to high risk of perforation Self-expanding stents have been reported in a few recent studies as an alternative to repeated Endoscopic balloon dilation of strictures in Crohn’s disease. These have included both fully covered metallic stents and biodegradable stents. They have had reasonable success rates in delaying the need for surgery, but have high migration rates. Hence, these are not yet recommended for routine use.

In conclusion, endoscopic balloon dilation is a safe and efficacious therapy for luminal strictures in inflammatory bowel disease and should be used to treat symptoms and prevent future morbidity.