Azathioprine for prevention of postoperative recurrence in Crohn's disease: a retrospective study

Background The efficacy of azathioprine (AZA) in chronically active Crohn's disease (CD) is well established. Whether this drug is also useful to prev ent recurrences after surgery is unknown. We report here our experience of AZA in this therapeutic goal. Methods Between 1987 and 1996, 38 patients with CID were treated with AZA t o prevent postoperative recurrence. Twenty-three of these had undergone a c urative resection with removal of all previously involved parts of the gut. In the other 15 patients, resection was limited to the parts of the gut ma croscopically abnormal at the time of surgery; those parts that were previo usly involved but normal at this time were conserved. The operative procedu res were ileocolonic resection (n = 18), subtotal colectomy with ileorectal anastomosis (n = 12), coloproctectomy with ileoanal anastomosis (n = 4) or ileostomy (n = 2), ileal resection (n = 1) and segmental colectomy (n = 1) . Twelve patients had been treated previously with AZA before surgery; in 2 6 patients, AZA was started within the 2 months following surgery. Results The median duration of postoperative follow-up was 29 months. Proba bilities of clinical recurrence according to the Kaplan-Meier method were 9 , 16 and 28% at 1, 2 and 3 years, respectively. For the 25 patients who had a colonoscopy or a small bowel barium X-ray during the follow-up, probabil ities of anatomical recurrence were 16, 36 and 59% at 1, 2 and 3 years, res pectively. The probability of anatomical recurrence was significantly highe r in patients who had segments of the gut previously involved but not remov ed because they were macroscopically normal at the time of surgery. Conclusion In patients treated with AZA, the rate of postoperative endoscop ic recurrence was lower than that previously reported in untreated patients . Our results suggest that AZA should be evaluated prospectively for preven tion of postoperative CD recurrence, at least in highrisk patients. (C) 200 1 Lippincott Williams & Wilkins.