Endoscopic electrocautery incision therapy for benign lower gastrointestinal tract anastomotic strictures

Etiology and location

Studies included in this review identified patients with post-surgical anastomotic stricture [4-13]. Post-surgical anastomotic stricture can occur as a result of tissue ischemia, inflammation, hemorrhage at the site of the suture line, anastomotic site leakage or recurrence of disease [17,18]. Anastomotic leakage has been shown to contribute towards increased inflammation, thereby contributing to an increased risk of fibrosis and stenosis. According to a Cochrane Database systemic review, a side-to-side stapled anastomosis has been recommended for ileocolic anastomosis, especially following a right hemicolectomy for colonic adenocarcinomas, because it results in fewer anastomotic leaks [20]. A recent meta-analysis for Crohn’s disease showed a higher rate of anastomotic leakage with end-to-end anastomosis compared to side-to-side [21].

The spectrum of surgical resections included those for malignant primary lesions, such as low anterior resection for rectosigmoid colon cancer and rectal cancer, intersphincteric resection for lower rectal cancer, hemicolectomy or subtotal colectomy [5,7-13]. Patients undergoing proctocolectomy with ileal-pouch anal anastomosis for ulcerative colitis, sigmoid resection for diverticulitis and volvulus, prophylactic proctocolectomy for familial adenomatous polyposis were also included by some studies in this review [4,6,9,10]. Stricture formation across all studies included occurred at the site of the surgical anastomosis [4-13].

Criteria, characteristics and timing

Anastomotic strictures following colonic and/or rectal resections can present with symptoms of abdominal pain, cramping, constipation or symptoms of bowel obstruction. Studies included in this review identified strictures after patients presented with symptoms or as a result of follow-up assessments. Various radiologic and/or endoscopic techniques were utilized to identify the stricture. Strictures of <10 mm diameter were commonly seen to have presented with obstructive symptoms. Strictures in the descending colon or rectum may often present clinically once luminal diameter is 20 mm or less.

The majority of the studies defined the stricture as the inability to pass an endoscope across the anastomosis site [6-13]. The size of the endoscope used varied across the studies, hence creating a wide variability in the stricture definition. Few studies used radiologic techniques, such as barium enema, barium swallow or CT scan, to define the stricture [7,10,11,13]. The diagnosis in these studies was followed by endoscopic confirmation of the stricture. Wallstabe et al [4] used digital palpation and pouchoscopy to identify the area of fibrotic tissue, confirmed on biopsy. Truong et al, in a prospective study of 36 patients, defined three grades of stenosis according to the luminal diameter [12]. The severity of stenosis was closely related to the symptomatic presentation. Symptoms of bowel obstruction correlated with a luminal diameter of <5 mm, defined as grade-3 stenosis. The predominant use of stapled sutures in the index surgery was consistent across all three grades of stenosis. Anastomosis closure with a circular stapling device has been linked to an increase in the incidence of stricture to as high as 30% when compared with hand-sewn anastomosis [14-16]. This risk, however, is independent of the stapler diameter.

Benign stenosis of colonic, rectal or colorectal anastomosis can occur from as early as within the first postoperative month to as late as several years after the surgery [19]. Most subjects were diagnosed with a stricture in the first postoperative year [4,7-13]. One case reported by Dieuruf et al had a stricture diagnosed 12 years after the surgery [6].

Prior interventions

Only four studies reported the use of alternative techniques to treat the stricture before considering them for EECI therapy. The case reported by Wallstabe et al underwent six balloon dilations over 10 weeks after identification of the stricture [4]. Four months after the first dilation, a second series of treatment with balloon dilation, triamcinolone injections, as well as transanal advancement flap anoplasty, was carried out without success. Of the seven cases in the retrospective case series by Osera et al, five underwent bougie dilation and the rest underwent balloon dilation at least more than twice before they were termed as refractory [8]. In the case reported by Dieruf et al, balloon dilation was performed to a maximum diameter of 18 mm, but was followed by recurrence of abdominal distention after 48 h [6]. Hagiwara et al attempted two to three sessions of balloon dilations on all 41 patients, 12% (5/41) of whom failed to improve [11]. Balloon dilation of stenotic segments of the gastrointestinal tract has proved to be a convenient and less invasive treatment method. However, the majority of the patients require multiple sessions for the resolution of symptoms and recurrence is observed in about 18% of patients [3,22].

All details regarding stricture etiology, location, criteria, characteristics, time interval to develop and interventions performed prior to consideration for EECI therapy have been summarized in Table 2.

Table 2 Stricture characteristics for each study

Anesthesia and bowel preparation

Only four studies reported the type of anesthesia used for the procedure. Osera et al reported using conscious sedation via 2-3 mg of IV midazolam and analgesia via 35 mg of IV pethidine [8]. Bravi et al also used 5-10 mg of IV diazepam or 2.5-5 mg of IV midazolam for conscious sedation in 53 of the 60 patients included [9]. The remainder (7/60) opted out of receiving any conscious sedation for the procedure. No patient discomfort or pain was reported by either group of patients [9]. Two prospective studies by Truong et al and Schubert et al also utilized conscious sedation and/or analgesia [12,13]. The former used 5-10 mg of IV diazepam with 40 mg of IV butyl-scopolamine and the latter used 4-10 mg of IV midazolam [12,13]. No complications as a result of conscious sedation were reported in the studies detailed above.

With regard to bowel preparation, investigators used a regimen of liquid diet, oral laxatives and cleansing enemas to achieve a quality of bowel preparation that would allow adequate visualization for the therapy [9,10,12,13].

Technical details

All studies included in the review utilized the EECI technique, either alone or in combination with another modality. Three studies described the isolated use of EECI technique for stricture [4,7,9]. Wallstabe et al incised the stricture at 3 points using an ERCP sphincterotome and a precut knife [4]. Case reports by Kwon et al described incisions based on the stricture length and caliber determined either endoscopically or radiologically [7]. The length and depth of the incision were gauged by the endoscopist and never crossed the muscular layer. The electrocautery therapy lasted for about 5-10 min. An insulated tip (IT) knife and a bimodal electrocautery were used along with a 12 mm colonoscope [7]. A similar approach was used in the case series by Bravi et al, where two expert endoscopists performed the technique, making at least four incisions with a precut sphincterotome [9].

Three studies described using an endoscopic electrocautery RIC technique for stricture management. This is essentially same as EECI, but also involves resection of mucosal tissue in between the incisions [5,8,10]. Kawaguti et al and Osera et al both described making radial incisions with excision of the tissue in between the incisions along the lumen with an IT knife. The former took 12 min for the case and the latter reported a median duration of 18 min for the procedure [5,8]. In the retrospective case series by Osera et al, the majority of patients underwent a single inpatient procedure, with the median length of hospital stay being 3 days. The prospective study by Brandimarte et al utilized sphincterotome and bimodal electrocautery to make six radial incisions, the length and depth of which were operator dependent [10].

EECI with adjunctive corticosteroid injection was performed by Dieruf et al [6]. A 5-Fr needle knife was used to make incisions in four quadrants, with each cut being 5-8 mm in depth, resulting in a 20 mm luminal diameter. This was followed by injecting 7-10 mg of triamcinolone at each incision site using a sclerotherapy needle, up to a total of 35 mg.

EECI in combination with balloon dilation was described in two prospective studies [11,12]. Hagiwara et al used an endoscopic diathermic knife and a balloon dilator [11]. Two to three shallow radial incisions were made using repeated, short bursts in the stricture scar. A flexible tipped guide wire was then inserted, over which a balloon dilator was positioned at the stricture under fluoroscopy. Water-soluble contrast medium was used to inflate the balloon, which was then pulled out to have the stricture aligned in the middle. The balloon was left for 15-20 min and resulted in loosening of the scar. Of the five patients in this study, two who had a diverting colostomy underwent two sessions a week for two weeks and the remaining three patients without a colostomy had one treatment session only. In another prospective study by Troung et al, a similar technique of hydraulic balloon dilatation was performed after radial incisions were made [12]. The type of incision was dependent on whether the stenosis was central or eccentric. A higher number of consecutive dilatations were needed for a higher grade of stenosis.

Subjects enrolled in the prospective study by Schubert et al underwent incisions by a polypectomy snare, followed by reduction of the scar tissue in between the incisions by APC [13]. No more than one therapeutic session was needed for all subjects.

Details of the instruments used and techniques performed in each study have been summarized in Table 3.

Table 3 Instruments and technical details for intervention in each study

Success

All ten studies together had 186 subjects with lower gastrointestinal tract benign anastomotic stricture who underwent EECI, either alone or in combination with another treatment modality. Individually, 63 subjects underwent EECI alone [4,7,9]; 47 subjects underwent endoscopic electrocautery with RIC [5,8,10]; 41 subjects underwent combination therapy with EECI and balloon dilation [11,12]; 34 subjects had combination therapy with EECI and steroid injection [6]; and 1 subject had combination treatment with EECI and APC [13]. The overall success rate for early post procedure and on long-term follow up was 98.4% (183/186) and 93.0% (173/186), respectively. On long-term follow up, the success rate for EECI alone was 95.2% (60/63), for EECI with RIC it was 95.8% (45/47) and for combination treatment with EECI and balloon dilation it was 87.8% (36/41). Individual studies of combination treatment by EECI and steroid injection or APC reported success rates of 91.2% and 100%, respectively [6,13]. Six studies reported a 100% success rate on long-term study-specific follow up [4-7,10,11].

Of 186 total subjects, 14 had a refractory stricture and 172 were treatment-naïve. Among the subjects with a refractory stricture, the success rate both early post-procedure and over long-term follow up was 85.7% (12/14). Among treatment naïve subjects, the early post-procedure and long-term success rates were 99.4% (171/172) and 93.6% (161/172) respectively Figure 1 and 2 depicts EECI of a benign colon stricture.

Figure 1 Colon stricture (3 cm long, benign, non inflammatory)

Figure 2 Post-endoscopic electrocautery incision of colon stricture

Failure

EECI, alone or in combination with another treatment modality, failed to achieve stricture dilation in 1.6% (3/186) of the total population [4-13]. All studies except two reported a failure rate of 0% in achieving post-intervention stricture dilation [4-13]. Osera et al reported that 2 of 7 subjects (28.6%) failed to achieve stricture dilation, despite 4 and 6 sessions of endoscopic electrocautery with RIC, respectively [8]. In another study, the operator was unable to perform EECI and balloon dilation in 1 of 36 subjects (2.8%), because of a lack of epithelialization of the stricture that increased the risk of perforation [12].