Endoscopic management of ileal pouch–anal anastomosis strictures: meta-analysis and systematic literature review

Introduction

Restorative proctocolectomy with ileal pouch–anal anastomosis (IPAA) is the procedure of choice for the management of ulcerative colitis (UC) that has proven refractory to medical therapy, UC with dysplastic/neoplastic transformation of intestinal mucosa and familial adenomatous polyposis [1]. IPAA improves the quality of life in UC patients by decreasing the symptom burden and avoiding the need for an end ileostomy [2].

Adverse events following IPAA include pouch leaks, pouch sinus formation, pouch strictures, pouchitis, cuffitis, pouch neoplasia and irritable pouch syndrome [3]. The reported incidence of strictures following IPAA ranges from 5-38%, with the most common stricture locations being the pouch outlet, followed by the pouch inlet, the afferent limb and the mid-pouch [4,5].

The surgical management of IPAA strictures includes several options, including pouch resection, strictureplasty, stapler resection, pouch revision and re-anastomosis, depending on the stricture’s location and severity, as well as the surgeon’s experience [5]. While surgical treatments are effective, they are associated with significant morbidity [4,5]. Minimally invasive endoscopic interventions that have been used for other gastrointestinal tract strictures, such as balloon dilation, stricturotomy and stent placement, have been shown to be effective for IPAA strictures [4].

We conducted a meta-analysis of studies that analyzed the safety and efficacy of endoscopic interventions for the management of IPAA strictures.

Materials and methods

This study adhered to the preferred reporting items for systematic reviews and meta-analysis (PRISMA) checklist to identify the safety and efficacy of endoscopic interventions for the management of IPAA strictures (Appendix A).

Results

Search results and population characteristics

The initial search yielded 664 references. After the removal of duplicates, a total of 346 studies, including full articles and abstracts, underwent formal title and abstract screening. A total of 9 studies, including 607 patients, were included based on our inclusion and exclusion criteria (Fig. 1).

Figure 1 Study selection flow chart

A total of 607 patients (54% male, mean age 44.5±6.5 years) with IPAA strictures were included in the final analysis. The most common indication for IPAA was UC (97%). One or more strictures were present in each patient and the most common locations were pouch outlet (46%), followed by pouch inlet (33%), afferent limb (11%), and other (10%). Mean stricture length was 1.4±0.4 cm. Endoscopic interventions were performed with techniques including endoscopic balloon dilation (EBD), needle knife stricturotomy (NKSt), stent placement and digital dilation under endoscopic view. The mean number of sessions required was 2.5±0.9 per patient. The mean follow-up period following the first intervention was 3.4±3.0 years.

Characteristics and quality of included studies

Table 1 shows the characteristics of the included studies. The quality assessment of the included studies was conducted independently and blindly by 2 authors (PP and MP). Discrepancies that arose were resolved by a third author (MAE) in an independent and blinded manner. Our systematic review employed 2 types of quality assessment: the interventional NIH (National Institute of Health) scale for pre–post studies without control groups, and controlled intervention study assessment [15]. According to the NIH scale for pre–post studies, 1 study was rated as high quality, receiving a score of 9, 4 studies were considered fair quality, scoring between 5 and 8, and 1 study was considered of poor quality, receiving a score of 3, as detailed in Supplementary Table 1. Studies of poor to moderate quality lacked sufficient data, often sourced from abstracts rather than full-text articles, limiting our ability to evaluate them thoroughly. In the interventional controlled trials, 3 studies achieved fair quality, with a score of 6-8 out of 14, based on various aspects outlined in the NIH quality assessment for controlled intervention trials, as noted in Supplementary Table 2. Given the specialization in handling intricate cases with proficient endoscopists and support staff, these studies did not involve randomization or blinding, as physicians had the autonomy to choose the procedure they felt most comfortable performing.

Table 1 Characteristics of studies

Pooled outcomes

Success

Technical success, defined as the ability to pass the endoscope through the stricture following the intervention, was achieved in 97.4% patients (I²=44%) (Supplementary Fig. 1). Immediate clinical success, defined as improvement in symptoms following the first intervention, was achieved in 44.5% patients (I²=86%) (Supplementary Fig. 2). Clinical success at the end of the follow-up period was reported in 8/9 studies and was seen in 81.7% patients (I²=81%). However, 38 (6.8%) patients experienced pouch failure, defined as the need for surgical management such as pouch excision, diversion ileostomy or stricturoplasty during the follow-up period, due to refractory IPAA strictures (I²=67%), while 50 (14.5%) patients ultimately required surgical interventions for management of refractory stricture or related complications (I²=55%) (Figs. 2,3).

Figure 2 Forest plot: success at the end of follow up

CI, confidence interval

Figure 3 Forest plot: pooled event rate of pouch failure

CI, confidence interval

Adverse events

Nineteen (3.9%) patients experienced serious adverse events related to the endoscopic intervention (I²=0%)(Fig. 4): 7 (2%) patients who underwent EBD developed perforation requiring surgical intervention (I²=0%), while 12 (2.6%) patients, 9 and 3 in the EBD and NKSt groups, respectively, had major postprocedural bleeding requiring blood transfusion following the intervention (I²=0%) (Supplementary Fig. 3,4).

Figure 4 Forest plot: pooled severe adverse events

CI, confidence interval

Validation of meta-analysis

Sensitivity analysis

To assess whether any single study had a dominant effect on the meta-analysis, we excluded each study individually and analyzed the effect on the main summary estimate. No single study significantly affected the outcome or heterogeneity.

Heterogeneity

We assessed the dispersion of the calculated rates using the I² percentage values. Based on I² analysis for heterogeneity, considerable heterogeneity was noted for the pooled rate of immediate clinical success and success at the end of the follow-up period following endoscopic interventions. Low heterogeneity was noted for the pooled rate of intervention related to severe adverse events. The I² values for the pooled rates are summarized in Table 2.

Table 2 Pooled outcomes

Prediction interval

This meta-analysis was conducted using the random-effects model. Therefore, we calculated the prediction interval, which deals with the dispersion of the effects. The calculated prediction interval for clinical success at the end of follow up was 0.817 (95%CI 0.423-0.964) and for pouch failure it was 0.068 (95%CI 0.005-0.533) (Supplementary Fig. 5).

Discussion

Our study evaluated the efficacy and safety of endoscopic interventions for the management of IPAA strictures at various pouch locations. Our analysis found a high technical success rate for endoscopic interventions (97%) in the management of IPAA strictures. However, the immediate clinical success rate was substantially lower (44.5%). This discrepancy arose because, in our analysis, most patients required more than 1 treatment session (2.5±0.9 sessions per patient) to achieve symptomatic improvement [16,17]. In addition, serial dilations with a gradual increase in diameter, as opposed to single-session dilation, decreased the risk of perforation [18].

Strictures at different pouch locations may respond differently to various endoscopic and surgical interventions [5,16]. For proximal outlet (IPAA) strictures, non-endoscopic surgical dilation techniques have a success rate of 90-100% [5]. However, distal strictures (inlet, mid-pouch, afferent limb) often require more invasive surgical interventions, such as strictureplasty, bypass, or even reconstruction of the anastomosis, with success rates of 80-100% [5]. Distal strictures are accessible via minimally invasive endoscopic techniques and can potentially spare patients from having to undergo repeat surgery. Our analysis, including a comparable number of both proximal and distal strictures, showed an 81.7% success rate at follow up, with an acceptable complication rate.

A recent meta-analysis reported a 6% overall pouch failure rate, irrespective of the method used to treat the pouch-related complications [19]. This rate is similar to that seen in our analysis (6.8%), where pouch failure following failed endoscopic management of pouch strictures was analyzed. Strictures alone cannot be blamed for pouch failure, as most studies reported multiple risk factors, such as underlying disease activity prior to IPAA, surgical techniques, location of strictures, and post-IPAA non-mechanical complications [20,21].

In our analysis, the pooled event rate of iatrogenic perforation was 2%, comparable to rates following endoscopic dilation for other lower gastrointestinal strictures [18,22,23]. None of the patients treated with NKSt experienced perforation events. None of the patients in recent studies experienced iatrogenic perforation following endoscopic stricturotomy for strictures related to inflammatory bowel disease (IBD) [24,25]. Major bleeding events with EBD (1.8%) were more frequent, whereas with NKSt (6%) they were comparable to prior studies on EBD and NKSt for the management of IBD-related strictures [25,26].

To our knowledge, this is the first meta-analysis to specifically examine the safety and efficacy of endoscopic interventions for managing ileal pouch strictures following restorative proctocolectomy. Our study had several limitations. First, the predominance of retrospective studies within our analysis introduced a bias towards historical data. Second, 3 of the 9 studies included in the analysis originated from conference abstracts, which by their nature have not undergone a comprehensive inspection and peer-review process. Finally, the heterogeneity encountered in the analysis of clinical success, immediate and at end of follow up, was high. However, all the studies showed a high rate of clinical success, indicating that the heterogeneity may have been due to variations in the population sizes (3-200), the baseline severity of the disease and the length of the follow-up periods (1-9 years) across different studies, and not to other effects.

In conclusion, endoscopic treatment of ileal pouch strictures is effective and reasonably safe. Future studies comparing different endoscopic interventions for managing ileal pouch strictures at different locations are needed to develop effective treatments for each type of stricture.