Crohn's disease (CD) is a chronic inflammatory digestive disease with a long-term course that is usually progressive, disabling, and relapsing, and most patients with CD undergo intestinal surgery during their lifetime.^1,2 Effective therapy with immunomodulators and anti-tumor necrosis factor α antibody have the potential to change the natural course of CD. Actually, a large cohort study suggested that effective medical treatment has reduced the number of surgeries being performed for CD.³ On the other hand, intestinal strictures are still the main reason for surgeries such as intestinal resection and strictureplasty, which are performed to relieve stricture-related symptoms or ileus. Surgery is, however, not a basic treatment modality, and relapse at the anastomotic site is often a problem in CD.^4,5 Surgery sometimes lead to stricture recurrence, with the need for repeat surgery, and frequent surgery can lead to short bowel syndrome.⁶ Endoscopic balloon dilation (EBD) is an alternative to surgery for strictures secondary to CD. This endoscopic treatment is mainly indicated for Crohn's strictures of the gastric outlet, duodenum, colon and, ileo-colonic anastomosis.^{7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24} Many reports concluded that EBD is effective and safe, and its use can avoid surgery. Moreover, balloon-assisted enteroscopy (BAE), including double-balloon enteroscopy and single-balloon enteroscopy have been widely used for diagnosis and endoscopic treatment of small bowel diseases.^25,26 Since BAE enables us to perform EBD for small bowel strictures in patients with CD, the indications for EBD in CD have increased.^{27,28,29,30,31,32,33,34} Hence, it is important for gastroenterologists to know the current status and future direction of EBD for Crohn's strictures. This article reviews the literature on EBD performed using the traditional approach to strictures of the colon and ileo-colonic anastomosis, in addition to that on recent approaches to small bowel strictures using BAE.

EBD using BAE for small bowel strictures is almost the same as EBD for colorectal and ileo-colonic strictures in terms of procedure and technique. However, special insertion techniques for deeper portions of the small intestine in BAE are necessary. Moreover, the proper insertion route should be determined according to the results of small bowel imaging. In addition, there are some technical difficulties, compared to EBD performed with a conventional colonoscope. For instance, it is not easy to stabilize the tip of the scope and to maintain a good visual field because of the limited space available, severe angulation, and adhesion in the small intestine. Therefore, EBD for small bowel strictures in CD is still challenging, although reports regarding this endoscopic intervention have recently increased, mainly from Japan (Table 2).^{27,28,29,30,31,32,33,34} According to published studies, maximum dilation diameter for strictures of the small intestine varied from 15 mm to 20 mm which is relatively small compared to EBD for colorectal and ileo-colonic strictures. It is speculated that the reasons for the small dilation diameter are more severe strictures and the operator's concerns regarding perforation.

There are also no standardized definitions of short-term success of EBD for small bowel strictures. Despite the cited difficulties, the reported technical success rate is 72% to 100%, which is similar to the results of EBD for colorectal and ileo-colonic strictures. While clinical efficacy was evaluated based on the patient's obstructive symptoms in almost all studies, Despott et al.³⁰ adopted the VAS score for abdominal pain in their evaluation. The clinical success rate was >70% (72%-87%) overall, which is comparable to published data on EBD for colorectal strictures. Perforations occurred in 0% to 10% of subjects in most published studies, although one small cohort reported a perforation rate of 20% (2/10).³² Judging from these results, EBD using BAE for small bowel strictures is equal to EBD using a conventional colonoscope in terms of short-term efficacy and safety.

There is insufficient reliable data on the long-term outcomes of the procedure for small bowel strictures, because the observation periods were very short, as shown in Table 2. However, in two studies that investigated the long-term outcomes of EBD for small bowel strictures, the average observation period after initial EBD was more than 3 years.^33,34 Hirai et al.³³ reported that the cumulative surgery-free rate in 65 subjects was 79% and 73% at 2 and 3 years, respectively. However, the cumulative redilation-free rate in their analysis was 64% and 47% at 2 and 3 years. This rate was similar to the results reported in previous studies evaluating EBD for colonic and ileo-colonic anastomosis. Therefore, the need for frequent redilation seems to be an issue with EBD regardless of the target stricture site. According to a recent study published by Sunada et al.,³⁴ the cumulative surgery-free rate after initial EBD for small bowel strictures in CD was 87.3% at 1 year and 78.1% at 3 years. Even though the cumulative surgery-free rate was similar to that in the study by Hirai et al.,³³ they included patients with fistulas as subjects, and the presence of a fistula was significantly associated with the need for surgical intervention (hazard ratio, 5.50; 95% CI, 2.16-14.0; P=0.01). Hence, for small bowel strictures with fistula, EBD should be considered only in patients who are likely to benefit (e.g., for completion of important life events) from postponement of surgery.

The tendency for CD patients to have multiple small bowel strictures is one of the limitations of EBD using BAE. In fact, the proportion of patients with multiple small bowel strictures was high in published studies. Although the therapeutic strategy for this situation is not established, in my opinion, in the absence of exclusion criteria such as fistulas, EBD can be attempted before performing surgery (Fig. 1). Further strict analysis will be needed for confirmation of efficacy of EBD for multiple small bowel strictures.

The disease behavior of CD has changed over time, with most patients having a stricturing type or a penetrating type.^45,46 In particular, since patients with ileal type CD tend to have small bowel strictures, performing EBD using BAE would be more meaningful to improve their quality of life. However, as described above, there is only limited evidence regarding the utility of this endoscopic intervention in patients with small bowel strictures. Further studies confirming the efficacy and safety of this procedure are desirable. In fact, a nationwide, multicenter, open-label, prospective cohort study is now ongoing in Japan. According to the results of interim analysis in this study, the short-term success rate using VAS score as an indicator was 69%, and no perforations were reported.⁴⁷ The results also indicated equal efficacy and safety compared to published studies. The full data of this study will be published in the near future.

Many issues should be addressed regarding EBD in CD. First, it is difficult to state whether EBD or surgery is best for release of intestinal strictures. Direct comparison between these two interventions is not easy, due to the differences in invasiveness and the goal of each treatment. Surgical interventions are superior to EBD in terms of certainty. According to a small retrospective cohort study, the proportion of patients who required any reintervention during follow-up was significantly lower in the surgical group than in the EBD group (OR, 5.62; 95% CI, 1.66-19.01; P=0.005).⁴⁸ However, surgical interventions are inferior to EBD in terms of both the invasiveness and relapse at the anastomotic site. Strictureplasty has been performed for intestinal strictures in CD as a small bowel-saving strategy.^49,50,51,52 Although the postoperative recurrence rate after strictureplasty is comparable to that of surgical resection, clinicians should also pay attention to the risk of malignancy in the preserved bowel. Since the performance of a randomized controlled study to compare EBD and surgical intervention is extremely difficult, gastroenterologists need to select an intervention method for intestinal strictures in CD by considering various factors (e.g., patients background, status of stricture, disease activity, length of small intestine, and so on) on a case by case basis.

The basic treatments for CD are not endoscopic or surgical interventions but medical therapies to control disease activity and improve quality of life. There are many effective medical treatments for CD, such as anti-tumor necrosis factor α antibody and immunomodulators. Additional candidates for treatments of CD will emerge in the future. Ideally, a combination of EBD and these effective medical treatments should be used for CD patients with intestinal strictures.

New endoscopic treatment modalities for Crohn's strictures such as the use of self-expanding metallic stents^53,54 and needle-knife treatment⁵⁵ have been reported. Implantation of metallic stents has mainly been used for patients with malignant gastrointestinal strictures because these enable improved food intake without surgery. As EBD has the limitation of loss of long-term efficacy, stents have been used for Crohn's strictures in several institutions. However, there are some disadvantages, such as inflammatory reactions to a foreign body, recurrent stricture due to mucosal growth into the stent, and perforation or hemorrhage with self-expansion. Therefore, further large-scale studies to confirm the efficacy and safety of stenting are needed. Needle-knife treatment is a limited option for Crohn's stricture. With only a few case reports,^55,56 its efficacy is still uncertain and safety is questionable. Further experience is necessary in order to apply this intervention in CD patients with intestinal strictures.