Endoscopic Management of Bowel Obstruction

Abstract

The traditional morbidity and mortality associated with traditional management has stimulated exploration of endoscopic approaches. Success depends on patient selection, the location and etiology of obstruction, patient status, and the capability of the endoscopist. This article discusses techniques of intralumenal dilation and stent placement and results from systematic reviews.

Bowel obstruction remains a common source of morbidity and mortality. An endoscopic approach is becoming an attractive approach in selected cases. Factors related to patient selection include the location and etiology of obstruction, patient status, and the capability of the endoscopist. With advanced endoscopic techniques, including overtube and balloon-assisted endoscopy, the entire gastrointestinal tract is potentially accessible.

Colonic strictures result from inflammatory bowel disease (Crohn's and ulcerative colitis), anastomotic problems, ischemia, malignancy (primary and secondary), radiation injury, nonsteroidal anti-inflammatory drugs, and the consequences of diverticulitis. ^{1 2} The location and degree of the stricture often determine the significance of symptoms. High-grade strictures may lead to a complete obstruction and a surgical emergency. Colonic obstruction is a common clinical problem in the United States and the majority are due to malignant processes. Approximately 8 to 29% of patients with colorectal cancer initially present with symptoms of partial or complete bowel obstruction, most of which are left sided and either Stage III or IV. ^{2 3}

There are several options to manage patients with large bowel obstruction. Traditionally, patients were resuscitated, evaluated with a contrast study such as a water-soluble enema ( Fig. 1 ), CT scan, or colonoscopy, and then taken to surgery. Surgical options for right-sided lesions usually involved a segmental resection and an ileocolic anastomosis or ileostomy depending on the status of the bowel and patient as well as the experience of the surgeon. Management of left-sided lesions traditionally varied from a segmental resection and colostomy, a subtotal colectomy, and ileorectal anastomosis or ileostomy. These emergency procedures had an associated mortality of 10 to 30% and a morbidity of 10 to 36%. ^{2 3 4} Due to their age and comorbidities, many of the patients receiving ostomies did not have their stomas closed and the time spent recovering from therapy or associated with their morbidity or complications from surgery often delayed or prevented chemotherapy and radiotherapy. Quality of life and the cost of ostomy supplies have also been an issue. These limitations encouraged development of other options such as endoscopic balloon dilation and self-expanding intraluminal stenting. This article will review the indications, techniques, and published results of endoscopic colonic balloon dilation and intralumenal stents for colonic obstructions.

Fig. 1.

Contrast enema demonstrating an obstruction colonic lesion. (Reproduced with permission of Ochsner Clinic Foundation).

Patient Selection and Indications

Proper patient selection is important to obtain a successful outcome. The history, examination, and diagnostic studies will often suggest an etiology and provide information of characteristics of the stricture which are predictive. Predictors of a successful outcome include narrow stenosis (<10 mm) and short strictures (<4 cm) that are accusable. ¹ Anastomotic strictures do better with dilation while malignancies respond better with intraluminal stents. Additional poor predictors are numerous strictures, complete obstruction, associated fistulas within the stricture, and active inflammation at or around the stricture. Finally, the patient's lesion must be accessible to an endoscope and should not have an absolute indication for a laparotomy (e.g., perforation).

Balloon dilatation may resolve an anastomotic or inflammatory stricture, but repeat treatments are often necessary. Intralumenal stenting can be used for palliation or as a bridge to surgery (converting an emergency procedure to an elective one). The use of stents for nonmalignant obstruction has been limited. Since their introduction in 1991, colonic stents have become an important method of palliation for obstruction in colorectal cancer patients, especially those with unresectable metastatic disease. ⁵ These self-expanding metallic stents can potentially dilate the lumen to a near-normal diameter, providing quick relief of symptoms and, in some cases, allowing endoscopic or radiologic assessment of the proximal colon. An experienced endoscopist can place a stent in patients using minimal sedation, without need of prior endoscopic dilation and a low risk of complications such as perforation or tumor fracture. Moreover, these stents can be placed across relatively long lesions by overlapping stents in a “stent-within-stent” fashion. Most patients with benign strictures are better managed with dilatation or traditional operative techniques as the morbidity associated with stents in this patient group has been prohibitive. However, experience with stents used as a bridge to surgery in selected patients with benign strictures is being accumulated.

Technique

Endoscopic Balloon Dilation

The endoscope is passed via the anus until the stricture is encountered ( Fig. 2 ). Fluoroscopic assistance is rarely needed. A guidewire or balloon catheter is passed via the accessory port of the endoscope through the stricture under direct visualization ( Fig. 3 ). The balloon catheter is advanced over the guidewire or to the midpoint of the stricture and inflated with saline using a pressure- or volume-controlled handle. It is important to keep the balloon centered on the stricture during inflation to prevent it slipping out of the stricture. Inflating the balloon slowly and using longer balloons help to minimize the risk of migration. ¹ After removal of the balloon, the dilated stricture is usually examined endoscopically. Several endoscopically guided balloons are available. For colonic use, the balloons range in size from 6 to 20 mm diameter.

Fig. 2.

Colonoscopic view of obstructing lesion. (Reproduced with permission of Ochsner Clinic Foundation).

Fig. 3.

Insertion of a through-the-scope (TTS) guidewire through the obstruction. (Reproduced with permission of Ochsner Clinic Foundation).

Intraluminal Stents

Several intralumenal stents are currently available ( Table 1 ). They are composed of stainless steel or alloys. Nitinol is an alloy of nickel and titanium which has increased flexibility and is helpful for stenting sharply angulated regions at the cost of lesser radial force relative to stents made from other metals. Elgiloy is an alloy of cobalt, nickel, and chromium and is corrosion resistant and capable of generating high radial forces. Stents are produced in coated and uncoated versions. Coated stents may have lower rates of tissue ingrowth which may produce a longer patency but has a higher rate of migration. There is some evidence that they might aid in fistula closure. ^{1 6} Currently, only uncoated stents are approved for use in the colon. Colonic stents range in predeployment diameter from 10 to 30 F and post-deployment diameter of 20 to 35 mm, and 40 to 120 mm in length. Most have a proximal and/or distal flare to prevent migration. The stents are generally packaged in compressed form and constrained within a delivery devise.

Table 1. Available stents.

Stent	Manufacturer	Composition	Deployment	Size (diameter/length)
Wallflex	Boston Scientific	Elgiloy	TTS	27–30 mm 60–120 mm
Wallstent	Boston Scientific	Stainless steel	TTS	20 and 22 mm 60 and 90 mm
Evolution colonic controlled-release.	Cook Medical		TTS	25 mm 60, 80, 100 mm
Evolution duodenal uncovered controlled-release.	Cook Medical		TTS	22 mm 60, 90, 120 mm
Z-stent	Cook Medical	Stainless Steel	OTW	25 mm 40–120 mm
Ultraflex precision	Boston Scientific	Nitinol	OTW	30/25 mm 57–117 mm

Abbreviations: OCW, over the wire; TTS, through the scope.

Stents can be deployed through the scope (TTS) with or without guidewires or passed under fluoroscopic control over an endoscopically or fluroscopically placed guidewire. Enteral stents may be uncovered or covered (referring to whether the meshwork of the stent is barewire or has a covering to decrease tissue ingrowth into the stent). ⁶ The author's preference is an Enteral Wallflex 120 mm in length (Boston Scientific). TTS stents are easier to place through proximal or angulated lesions.

The following is the technique used by the author. Patients are taken to the endoscopy unit and placed on a fluoroscopically capable table. After placement of monitors and nasal oxygen, the patient is placed into a left lateral position and sedated. A colonoscope is inserted and advanced to the obstructing lesion. Fluoroscopy is used to observe the lesion location and the patient's position is altered to obtain the best fluoroscopic view. Intralumenal contrast may assist in this localization and positioning. If the lumen appears tight a flexible guidewire in inserted through the working port of the colonoscope and under direct vision it is gently passed across the obstruction ( Fig. 3 ). If the lumen is not so tight the stent itself may be passed across the lesion. Fluoroscopy confirms that the wire or catheter has passed through the lesion. If a guidewire was used, a TTS stent is then passed over the guidewire and passed across the lesion ( Fig. 4 ). Radio-opaque markers of the stent help to position it across the obstructing lesion. Under fluoroscopic visualization the stent is deployed slowly. Some repositioning is possible prior to complete stent deployment. Optimally the configuration of the deployed stent will have a flare proximal and distal to the stricture with a narrowed neck at the stricture. Successful deployment is often associated with a rush of gas or colonic contents through the deployed stent. Following successful deployment, a contract study or plain film will confirm appropriate deployment ( Fig. 5 ). If the stricture is long or the stent does not have adequate extension across the stenosis a second stent can be placed overlapping a portion of the initial stent. Attempting to pass the colonoscope through the stent after deployment should be avoided as the metal edge of the stent can damage the colonoscope ( Fig. 6 ).

Fig. 4.

( A ) Guidewire passed across stricture. ( B ) Undeployed stent is passed through stricture. Black markers indicate radio-opaque markers. (Reproduced with permission of Ochsner Clinic Foundation).

Fig. 5.

X-ray of stent in place. (Reproduced with permission of Ochsner Clinic Foundation).

Fig. 6.

Endoscopic view of deployed stent. (Reproduced with permission of Ochsner Clinic Foundation).

Results

Stenting has been more successful in shorter, distal lesions and lesions with colonic primaries. ⁷ Technical failure is usually due to inability to pass through a guidewire or stent catheter across the obstruction, usually due to angulation of the tumor. Clinical failure is usually defined as failure to relieve the obstruction or the occurrence of a major complication. The published experience on management of colonic strictures and obstruction is derived from case series, small comparative trials, and retrospective studies. To overcome these limitations several systematic reviews have been performed.

Malignant Strictures

Since the technique of using self-expandable metallic stents as a bridge to surgery was described over 20 years ago, there have been conflicting results. ^{8 9} Watt and colleagues in 2007 reviewed 88 articles (15 of which were comparative) and found a technical success of 96% (range 66–100%) and a clinical success of 92% (range 46–100%). ¹⁰ The hospital stay was shorter with stents in most studies. The median rate of reintervention was 20% (range 0–100%). The median stent migration rate was 11% (range 0–50%) and the medial perforation rate was 4.5% (range 0–83%).

In 2014, the European Society for Gastrointestinal Endoscopy stated that self-expanding metallic stents (SEMS) placement as a bridge to surgery was not recommended as a standard treatment of symptomatic left-sided malignant colonic obstruction. ¹¹ Concern was expressed about the short-term adverse events associated with stenting and the long-term survival of potentially curable patients. ^{12 13}

Arezzo and colleagues performed a systemic review and meta-analysis of eight randomized controlled trials and 497 patients with left-sided malignant colonic obstruction. ¹⁴ The 60-day overall mortality was 9.6% in colonic stent with a bridge to surgery (SBTS) and 9.9% in emergency surgery (ES) treated patients ( p  = 0.97). Morbidity was 33.9% in SBTS and 51.2% in ES ( p  = 0.23). The temporary stoma rate was 33.9% in SBTS and 51.4% after ES ( p  < 0.001) while the permanent stoma rate was 22.2% after SBTS and 35.2% after ES ( p  = 0.003). Primary anastomosis was successful in 70.0% of SBTS patients and 54.1% of ES patients. ( p  = 0.043). The authors concluded that SBTS was associated with lower short-term overall morbidity and lower rates of temporary and permanent stoma. Depending on multiple factors such as local expertise, clinical status including level of obstruction, and certainty of diagnosis, SBTS offers advantages over ES in the short term.

A Cochrane meta-analysis in 2011 of five randomized trials of 207 patients compared stents with surgery. ¹⁵ Stent placement was successful in 86 of 207 patients, with a perforation rate of 6%, a migration rate of 2%, and an obstruction rate of 2%. The average time of clinical relief of obstruction was significantly lower with stent placement compared with ES. There was no difference with regard to 30-day mortality or complication rates.

A 2015 meta-analysis of 11 studies and 1,136 patients compared the long-term outcome of stents as a bridge to surgery (38%) to ES (62%). ¹⁶ The overall and disease-free survival was similar between the groups suggesting that stenting is a promising alternative for malignant large bowel obstruction.

Yang and colleagues compared 71 patients who underwent direct surgery and 182 patients who underwent SEMS placement as a bridge to surgery. ¹⁷ There were no differences in long-term oncologic outcomes between the groups. The primary anastomosis rate was higher in the stent group than in the direct surgery group. In a subgroup analysis of the bridge to surgery group, postoperative complication rates were lower in the minimally invasive group than in the open surgery group. Time to flatus and time to soft diet resumption were shorter in the minimally invasive group, as was length of hospital stay. The authors concluded that elective surgery after stent insertion does not adversely affect long-term oncologic outcomes. Furthermore, minimally invasive surgery combined with stent insertion for malignant colonic obstruction is safe and feasible.

Nonmalignant Strictures

Intestinal strictures result from several conditions. Benign conditions such as inflammation (inflammatory bowel disease) or postoperative anastomosis often respond to dilation. In 1989, Aston et al reported their experience with nine patients who underwent balloon dilatation of colon and rectal anastomotic strictures. Six of the nine strictures resolved after a single dilatation. ¹⁸ Park and colleagues retrospectively reviewed 43 consecutive patients with benign structures. ¹⁹ Twenty-nine had balloon dilation, seven had stents, and seven had both procedures. The clinical success was similar (balloon 89.1% and stent 87.5%). Reobstruction rates were similar (balloon 54.4% and stent 57.1% but the duration of patency was longer for the balloon, i.e., 65.5 months). The authors felt that endoscopic balloon dilatation was safe and effective as an initial treatment.

Small et al described 23 patients who underwent intralumenal stent placement for benign colonic strictures. ²⁰ Technical success was obtained in 95% of the patients but complications occurred in 38%. Eighty-seven percent of these complications occurred after 7 days of placement. The authors felt that intralumenal stents were useful as a bridge to surgery which should be performed within 7 days if possible.

Summary

The published data suggests that in experienced centers, an initial attempt at stent placement by an experienced multidisciplinary team is the preferred option for malignant obstruction. Endoscopic procedures can serve as a bridge to surgery in potentially curable, fit patients and provide effective and durable palliation for selected patients with malignant obstruction. Dilatation may be used in selective benign strictures while stenting may provide a bridge to surgery.