Introduction Strictures are a common complication of Crohn's disease (CD), both de novo and following surgery (Sx). While endoscopic balloon dilatation (EBD) offers a valuable alternative to Sx in managing them, there is paucity of data on factors that may influence the safety and efficacy of this technique. Our aim was to perform a multi-centre audit to determine our experience and outcomes of EBD in symptomatic CD strictures.

Methods A retrospective audit across three major hospitals in Northwest England was performed on patients between 1998 and 2011. Demographics, smoking status, immunomodulation, CRP, endoscopic findings, EBD details including complications and subsequent surgery at follow-up were all recorded. Success of EBD was defined as symptomatic improvement without need for surgery at follow-up.

Results Patient & Disease Demographics: 71 patients (43 female; age range 17–85 years, median 47) were audited. Duration of CD was <1–64 years (median 17). 45% had smoking history. CD distribution and concomitant immunomodulation was heterogenous. 58 (82%) patients had previous surgery. 46 (65%) were ileo-colonic anastomotic strictures while rest had it de novo (nine colon only, six terminal ileum, six jejunum, three duodenum, one oesophagus and one small and colon). The disease activity at anastomosis was i0–14 (30%), i1, i2–25 (55%) and i3, i4–5 (11%) while two had no record. Disease in de novo stricture was mild—5 (20%), moderate—10 (40%) and severe—7 (28%) while three had no record. The stricture length were 0.5–7 cms (median 2). CRP at first EBD was between 2 and 188 (median 5). Procedure: Maximum diameters of first and subsequent EBD were similar, 10–20 mm (median 15). 60 were performed at colonoscopy and 11 were performed at enteroscopy. 177 (range 1–11, median 2) EBD over median 8.5 months (1–84) were carried out total 84 (range 1–5, median 1) strictures. Outcomes: There were no serious complications. Success at index EBD was 31% with another 37% achieving long-term symptomatic relief from further EBDs. 18 (25%) patients needed surgery and five were lost to follow-up. Time to surgery following first EDB was 1–59 months (median 16.5).

Conclusion EBD was safe and seemed fairly effective (68% cases) in achieving long-term symptom improvement and avoided the need for surgery in 75% of cases. Further large prospective trials with control groups (those going straight to surgery) are needed to evaluate effects of CD phenotypes, endoscopic techniques and patient factors to help identify those that would best achieve palliation of symptoms with EDB compared to surgery.

Competing interests None declared.