- Top of page
- Material and methods
Bowel strictures are a major cause of morbidity, hospitalisation and surgery in Crohn's disease (CD). Strictures are caused by transmural inflammation leading to tissue remodelling, mesenchymal cell proliferation and fibrosis, but the underlying cellular and molecular mediators have not yet been elucidated.[1-3] Any segment of the gastrointestinal tract can be affected, but most commonly the terminal ileum, ileocolonic anastomosis or colon is involved. Traditionally, fibrotic strictures have been treated by surgical resection, as medical treatment is ineffective. The cumulative risk for surgery 10 years after diagnosis of CD is approximately 40–55%. The recurrence risk after resection is high. New endoscopic lesions will be found in a large majority of patients as early as after 1 year,[5, 6] which may in due course cause recurrent clinical symptoms requiring a second surgical procedure in 20–44% of patients after 10 years.[4, 7] In the long-term perspective, multiple resections may be associated with a risk for short-bowel syndrome. Strictureplasty has therefore evolved as a surgical bowel-preserving alternative and has been shown to be as safe and efficient as bowel resection in treating fibrotic strictures in CD.[8-10]
Endoscopic balloon dilation is a nonsurgical alternative in treatment of fibrotic strictures in CD. It was first described in 1986 and has generally been accepted as an effective and safe treatment for short anastomotic strictures, although the evidence is limited.[12-15] Most previous studies are retrospective uncontrolled, observational studies with small numbers of patients and short follow-up time. In general, the immediate technical success is high, but the clinical efficacy is variable. A review of 23 publications including between 5 and 59 patients reported a technical success rate of 90%, a rate of major complications like bowel perforation or severe bleeding of 3%, and a surgical recurrence rate of 27.6% after a median follow-up of 21 months. However, the evidence for long-term efficacy is sparse.
The aim of this study is to report a single referral centre experience of short-term and long-term efficacy and safety of endoscopic balloon dilation in treatment of intestinal strictures in CD.
- Top of page
- Material and methods
This is one of the largest studies evaluating endoscopic balloon dilation of fibrostenotic strictures in Crohn's disease. The data, strengthened by the long follow-up period of a median 12 years and the large number of patients and procedures, emphasise that endoscopic dilation is an effective and safe method of treating anastomotic strictures, which constituted the majority of strictures. At index dilation, all patients suffered from obstructive bowel symptoms that likely would have required surgery if endoscopic dilation had not been accessible. During 5-year follow-up, 52% of the patients required no further intervention or just one additional dilation, and only 36% of patients underwent another surgical resection. Technical success was achieved in 89% of all dilations, which is similar or slightly inferior compared with other studies.[13, 19] The immediate clinical success rate, defined as relief of clinical symptoms during the following month after the procedure, was 77% in patients from the primary catchment area of the hospital.
Of greater importance is the long-term efficacy. This analysis was based on a subset of 75 patients from the primary catchment area to reduce selection bias of referral cases, and excluding cases having repeated dilations owing to asymptomatic recurrence of strictures, as the latter would not reflect the true long-term efficacy of the procedure. During the first year, 80% of the patients required no further intervention or one additional dilation only, and at 3 and 5 years corresponding figures were 57% and 52% (Figure 1). At 5-year follow-up, 4/75 (5%) patients required ≥3 dilations, which illustrates that the patients can be well managed in the long-term perspective, as well, with repeated endoscopic dilations. The length of symptom-free interval, the technical feasibility and the patient's preference will aid in the clinical choice between surgery and repeated dilation. The recurrence of stricture does not reflect a failure of the method but rather the natural history of stricturing CD.
In their large study of 237 dilations performed in 138 patients, Thienpont et al. reported the need of repeated dilation in 46% or surgery in 24% of patients during a follow-up of 5.8 years. Similarly, in a systematic review of 13 studies enrolling 347 cases, 58% of patients avoided surgery during a mean follow-up of 33 months (range: 2–168 months) and 22% required two dilation procedures, while 19% required more than two dilations, ranging from 3 to 18. Surgery was ultimately required in 42% of patients after a mean period of 15 (range 1–70) months. Our results are quite consistent with those data but have the advantage of presenting cumulative proportion of patients undergoing no further invention, repeated dilations or surgery for each year during 5 years following index dilation (Figure 1). This gives a more detailed view of long-term outcome and underscores the value of endoscopic dilation in the treatment of strictures in CD.
In our study, complication rates per procedure were similar to other studies,[16, 19] with a frequency of 5.3% for any complication requiring hospital stay and of 1.3% for a complication requiring surgery. Considering that a symptomatic stricture is an indication of surgery, this complication risk is acceptable. We found that the largest balloon (25 mm) was associated with significantly greater risk of any complication compared to smaller balloons. This has previously been observed but not confirmed.
In previous studies, the procedure was performed under general anaesthesia[19, 22] or conscious sedation.[23-27] In this study, only 11/776 (1.4%) required general anaesthesia and a large majority of procedures were done during conscious sedation using benzodiazepines and/or opioids. A minority of dilation procedures (37/776, 4.8%) was performed without any sedative at all. Theoretically, conscious sedation or no sedation has the advantage of allowing the patient's discomfort to be monitored as an indicator of impending perforation during the procedure. In this study, 1/11 (9%) patient under general anaesthesia suffered from a complication, but the figures are too small to assess whether general anaesthesia carries a higher risk than conscious sedation.
The limitation of this study, consistent with previous reports, is the retrospective, observational study design without a control group. The study period is very long with a median follow-up time of 12 years. This is strength of the study, but it also makes the data more heterogeneous. The first dilation in our hospital was done in 1987. In the early period, repeated dilations were performed in patients with recurrent strictures but without clinical symptoms of bowel obstructions out of a belief that this might reduce the risk of surgery. Today, only patients with obstructive symptoms are referred for dilation. During the years, the indication for referral to surgery may have changed, albeit treatment in general was discussed in a medico-surgical team. Comparing outcome of two time periods, 1987–1998 vs. 1999–2009, showed no significant differences with respect to repeated dilation or surgery within the first year (data not shown). This supports that changes in the medical or surgical treatment have not significantly influenced our data. At index dilation, only 20% of patients were treated with thiopurines, and biologics have not been used at all in this cohort. Another potential limitation is that several different physicians with variable endoscopic skill performed the procedures. This, however, reflects a day-to-day clinical situation, which makes the results even more relevant.
Factors influencing outcome after endoscopic balloon dilation in fibrostenotic Crohn's disease are largely unknown. Technically successful dilation,[28, 29] a stricture length <4 cm and absence of ulcer in the stricture were found to be good prognostic factors. Data on influence of smoking after dilation are sparse and inconsistent. An increased risk for surgery or recurrent dilation was reported in smokers, whereas others found no significant association.[24, 31] Thienpont et al. reported that neither disease activity, assessed by CRP levels or endoscopic disease activity score, nor medical treatment after index dilation influenced subsequent disease course. We are currently analysing these factors in our material.
Fibrotic complications are one of the less understood manifestations of CD, and there is a need for further research to improve therapy and reduce hospitalisation and surgery. Intralesional steroid injections have been attempted to prolong the results of endoscopic dilation with various results.[32-35] In a small series of three patients, dilation of colonic strictures was achieved by intralesional injection of infliximab. Metallic stent insertion has been tried in a few patients but was limited by stent migration. Whether a biodegradable stent can be an alternative remains to be studied. Whether drugs like azatioprine or anti-TNF therapy can change the natural history of CD in general and the risk of recurrent fibrostenotic stricture in particular is not yet known.
In contrast to other reports,[25, 26] this study may suggest that dilation of de novo strictures is as efficient and safe as dilation of anastomotic strictures (Figure 2). However, we do not put too much emphasis on this observation, as the number of such dilations is small compared with the number of anastomotic stricture dilations. Furthermore, we had restricted indication for dilation of de novo strictures, and there is obviously potential selection bias in this group of patients.
In conclusion, this study demonstrates that endoscopic balloon dilation of intestinal strictures in CD is an effective and safe alternative to surgical resection. At 5-year follow-up, 52% of the patients had undergone no further intervention or one additional dilation only, and 36% had undergone surgical resection. Complication rate per procedure was low.