Aliment Pharmacol Ther 31, 634–639
Background Endoscopic balloon dilation has been shown to be an alternative to surgery in the treatment of Crohn’s symptomatic strictures.
Aim To analyse the impact of the type of the strictures –de novo or anastomotic – their location and their length on the outcome of endoscopic balloon dilation.
Methods Between December 1999 and June 2008, 55 patients underwent 93 balloon dilations for 74 symptomatic strictures. One stricture was located in the duodenum, 39 strictures were in the terminal ileum, 17 at the ileocoecal anastomosis after a preceding resection and 17 in the colon.
Results Endoscopic treatment was successful in 76% of the patients during an observation period of 44 (1–103) months. Of the patients, 24% required surgery. All patients who underwent surgery had de novo strictures in the terminal ileum. These strictures were significantly longer compared with the ileal strictures that responded to endoscopic treatment [7.5 (1–25) cm vs. 2.5 (1–25) cm; P = 0.006].
Conclusions The long-term success of endoscopic balloon dilation depends on the type of the strictures, their location and their length. Failure of endoscopic treatment was observed only in long-segment strictures in the terminal ileum.
Gastrointestinal strictures are a feared complication of Crohn’s disease. Up to one-third of patients develop a stricture from time of diagnosis to 10 years later.1–3 Symptomatic strictures resistant to medical therapy usually require surgery. Reappearance of symptomatic strictures in up to 40% of the patients within 4 years after surgery may necessitate repeated bowel resections4 leading to the risk of short bowel syndrome. 5–8 In recent years, endoscopic through the scope (TTS) balloon dilation has offered a therapeutic alternative.9–11 This procedure has the advantage of being minimally invasive and of preserving intestinal length. Recently published data demonstrate that hydrostatic balloon dilation of Crohn’s strictures is safe and effective with immediate success rates ranging from 71% to 100%.12–20 However, long-term success rates are lower and vary in the different series.12–20 This variability might be as a result of different rates of de novo and post-operative strictures (59–100%) and their different location in the gastrointestinal tract, which has, to our knowledge, not yet been taken into account.
The aim of the present study was to analyse whether the outcome of dilation of Crohn’s strictures depends on the type of the strictures, their location and their length.
Materials and methods
Between December 1999 and June 2008, 93 endoscopic balloon dilations were prospectively performed to treat 74 symptomatic strictures in 55 patients [21 males, 34 females, 41 (15–71) years]. All patients had given their written informed consent to the procedure. The history of Crohn’s disease was 11 (0–35) years. All patients presented with obstructive symptoms such as abdominal pain, cramps, bloating and vomiting. Prior to endoscopy, the patients received a sonographic and radiological assessment (small bowel enteroclysis or MRI) to confirm the clinically suspected stricture, to exclude intestinal fistulas and angulated strictures and to measure the length of the strictures.
The patients’ strictures were located as follows: one in the duodenum, 39 in the terminal ileum or at the ileocoecal valve, 17 at the ileocolonic anastomosis after a preceding resection and 17 in the colon. Six patients had more than one affected location.
The median length of the strictures was 3 (1–25) cm.
In patients with an anastomotic stricture, the median time between the last resection and the first balloon dilation was 70 (17–412) months.
At the time of dilation, 27% of the patients had no treatment of their Crohn’s disease, 34% received a systemic or topic steroid therapy, 13% took an immunosuppressive agent. No patient was under treatment with infliximab. Characteristics of the patients are summarized in Table 1.
|Age(median, range)||41 (15–71)|
|Duration of Crohn’s disease (median, range)||11 (0–35) years|
|Location of the strictures||74||100|
|Terminal ileum or ileocoecal value||39||53|
|More than one location||6||11|
|Prior forms of therapy||55||100|
|Surgical (ileocoecal resection)||17||31|
Endoscopy was performed under medication with midazolam/pethidin or midazolam/propofol. The act of dilation was carried out under conscious sedation to perceive a possibly imminent perforation by the patient’s utterance of pain. At the first dilation, specimens were taken out of the strictures to rule out malignancy. A hydrostatic through the scope balloon with a diameter of 15–18 mm on inflation was used for dilation in all patients (Microvasive Rigiflex, Boston Scientific, Boston, MA, USA). The introduction of the balloon into the stricture was done under visual control. If this was not possible, the balloon was inserted using a guide wire under fluoroscopic control. After reaching the correct position, the balloon was inflated up to a diameter of 15 mm for 1 min. After deflation, another inflation of up to 18 mm was performed and maintained for 1 min. This technique was used to minimize the risk of perforation. In long-segment strictures, segmental dilations were carried out. In unclear situations, a contrast agent (Solutrast 300, Bracco ALTANA GmbH, 78467 Konstanz, Germany) was administered intraluminally.
The follow-up data were collected prospectively and analysed retrospectively.
Technical success was defined as the passage of the scope through the stricture after balloon dilation.
Complications were defined as perforation, active bleeding with the need of an endoscopic re-intervention or the transfusion of blood and abdominal pain after balloon dilation.
Data are presented as medians and range. Univariate analysis (Mann–Whitney-Test) was used to compare the characteristics of the strictures. A two-sided P value of <0.05 was considered statistically significant. Kaplan–Meier plots were used to analyse intervals free of surgery.
Technical success and safety
Technical success was achieved in 95% of all investigated patients. In three cases (5%), treatment failed. One patient had a procedure-related perforation of the terminal ileum, which required immediate surgical intervention. The patient recovered uneventfully. In the two other patients, the dilation procedure was technically impossible because the guide wire could not be passed over a long-segment stricture in the terminal ileum.
The procedure-related perforation was the only observed complication leading to a complication rate of 1.8% per patient and 1.1% per procedure.
In all, 13 patients (24%) needed surgery within 1.5 (0–20) months. In 42 patients (76%), the endoscopic treatment was successful during an observation period of 44 (1–103) months (Figure 1).
Indication for surgery
The indications for surgery in the 13 operated patients were perforation of the terminal ileum (n = 1), technical failure of the procedure (n = 2), persistent or recurrent abdominal symptoms (n = 6), newly developed intestinal fistulas just proximal to the stricture (n = 2) and the patient’s wish despite preceding successful balloon dilation (n = 2). In all surgically treated patients, the strictures were located at the ileocoecal valve or in the terminal ileum. In strictures in all other locations, surgery was not required. Thus, in patients with de novo strictures of the ileocoecal valve or the terminal ileum, endoscopic balloon dilation showed long-term response only in 57% of the patients (17 out of 30) (Figure 2).
Length of the strictures
The length of strictures in different locations is demonstrated in Table 2.
|Location||Length (total range)||P|
|Terminal ileum and ileocoecal valve||4 (1–25 cm)||0.69|
|Colon||3 (1–10) cm|
|Anastomosis||3 (1–15) cm|
|Operated (term. ileum, ileoecoecal valve)||7.5 (1–25) cm||0.006|
|Non-operated (term. ileum, ileocoecal valve)||2.5 (1–25) cm|
The patients who required surgery had a stricture length of 7.5 (1–25) cm, whereas the stricture length in patients with an exclusive endoscopic treatment of an ileal or valvular stricture was 2.5 (1–25) cm. This difference was statistically significant (P = 0.006).
Number of dilations
During the follow-up period, 17 patients had a second, seven patients a third, one patient a fourth and one patient a fifth dilation. Of these 26 patients, seven were finally operated on. The 13 patients who required surgery had 2 (1–3) balloon dilations prior to their operation. In the 42 patients who received endoscopic treatment as only therapy, 1 (1–3) dilation during follow-up was performed. There was no significant difference in the number of dilations between operated and non-operated patients (P = 0.65). The time intervals between repeated dilations were 8 (0–65) months.
Our study describes the response to endoscopic balloon dilation in 55 patients with Crohn’s strictures. The success rate during the observation period was 76%. The outcome was worse in the terminal ileum, especially in long de novo strictures. Best results were achieved in anastomotic and colon strictures where no surgery was required.
The management of symptomatic Crohn’s strictures is challenging, because of the poor response to medical therapy and the high rate of recurrence after surgery4 with a reoperation rate of 3–6% per year.21, 22
Through the scope balloon dilation has become an alternative to surgery for treatment of strictures accessible to endoscopy. Although recurrence of symptoms was reported in 13–100% of patients,12–20 not all of these patients needed surgery, because repeated balloon dilation is an option.
In contrast to other studies12–20 where most of the patients were dilated for anastomotic strictures (59–100%), majority of our patients presented with de novo strictures (69%). Interestingly, our technical success rate (95%) was similar to that reported by others.12–15, 19 In two of our patients, the dilation procedure was technically impossible because the guide-wire could not be passed over a long-segment stricture in the terminal ileum. Another patient had to be treated surgically and he recovered uneventfully after a procedure-related perforation of the terminal ileum. This perforation corresponds to a complication rate of 1.8% per patient, which is in the range of 0–10% published in the literature. 12–20 Higher complication rates were reported15 when larger dilation balloons (20–25 mm) were used and patients were anaesthetized for the procedure. In our study, initial balloon dilation was performed with a 15-mm balloon before increasing the diameter up to 18 mm under minimal conscious sedation. In agreement with Thomas-Gibson et al.,18 we believe that patient feedback during the procedure is important in guiding the endoscopist as to the insufflation and its duration that can be performed without complication. Singh et al.23 considered the passage of the scope through the stricture after dilation as a possible source of a higher complication rate. In contrast, this technique is used in most studies to define technical success of balloon dilation. As we did not observe any complications in our patients, we think that this predictor of a successful dilation can be performed safely.
Of our patients, 24% required surgery within a median period of 1.5 months after the first balloon dilation. Twenty months after the first dilation, none of the patients had to be operated on. Interestingly, all patients who underwent surgery had de novo strictures at the ileocoecal valve or in the terminal ileum. Our data demonstrate that nearly half of the patients with strictures of this location required surgery within the follow-up period.
On the other hand, all our patients with colonic and anastomotic strictures were successfully treated by dilation. However, 45% of them needed repeated dilations.
A major factor affecting long-term response to balloon dilation seems to be the length of the stricture. Within the group of patients with an ileal or a valvular stricture, patients who required surgery had significantly longer strictures than those who responded to endoscopic treatment. Thus, the length of the stricture as well as its location might be useful as prognostic factors, helping to decide whether endoscopy or surgery is the treatment of choice in each individual patient.
We suggest that the endoscopic approach to Crohn’s strictures is justified in every case if technically feasible. Anastomotic and primary colonic strictures seem to be the most suitable indication. The treatment of de novo strictures in the terminal ileum remains a matter of discussion because nearly half of them required surgery. In our opinion, surgery should be favoured in long-segment strictures of the terminal ileum, when repeated balloon dilation becomes necessary early after the first procedure. This strategy, however, must be verified by other studies in the future.
Primarily in the group of patients with de novo strictures in the terminal ileum, the role of concomitant treatment such as steroid injection into the stricture, placement of self-expanding stents, immunosuppressive therapy and the newly described treatment with transilast24 appears to be very promising.
The effectiveness of steroid injection into the stricture after balloon dilation is not sufficiently evaluated. Encouraging data14, 25, 26 contrast with reports of no benefit20 or even an increased complication rate.27 Data on the effect of endoscopically placed self-expanding stents are still very limited. 20, 28
While azathioprine has been shown to reverse inflammatory changes at the anastomotic site,29 no data are available whether azathioprine is effective prophylaxis against recurrent strictures after endoscopic balloon dilation. However, Readler et al.16 recently demonstrated in abstract form that patients receiving azathioprine together with budesonide after dilation had a better 1-year recurrence rate (20%) than the placebo group (53%).
Infliximab has proven to provide mucosal healing. A consequent therapy with this agent, following a balloon-dilation, might reduce the reappearance of strictures in Crohn’s disease, although no data are available on this topic.
In conclusion, endoscopic balloon dilation is a safe and effective procedure in the management of Crohn’s strictures. It can at least postpone and probably avoid surgery in many patients. The success rate of balloon dilation and the need of surgery seem to depend on the type of stricture, primary or anastomotic, its location and its length.
We assume that double balloon enteroscopy might extend endoscopic dilation to intestinal strictures not yet accessible to conventional endoscopy.
Further data collection is needed to identify those patients with concomitant treatment who benefit most in long-term from this minimally invasive procedure.
Parts of this paper were presented at DDW 2006 as oral presentation (Gastrointestinal Endoscopy 2006; 63: AB100) and DDW 2008 as poster presentation (Gastroenterology 2008; 134, Suppl. 1: A472–A473).
Declaration of personal and funding interests: None.