Search in the Abstract Database

Abstracts Search 2014

* = Presenting author

P370. Prospective withdrawal trial of azathioprine (AZA) in ulcerative colitis (UC)

C. Pratico', N. Capozzi, F. Rizzello, A. Calafiore, R. Brugnera, G. Straforini, G. Spuri Fornarini, M. Campieri, C. Calabrese, P. Gionchetti, University of Bologna, Internal Medicine, Bologna, Italy


There are limited data regarding the optimal duration of AZA treatment in UC. Deep remission seems to be a positive predictor of long-term remission. We evaluated the clinical outcome after AZA withdrawal in steroid-dependent UC patients (pts), that achieved stable deep remission.


Up to now, out of 415 UC treated with AZA at our IBD clinic, 46 pts have interrupted the treatment for stable deep remission, defined as a minimum of 3 years (y) period of free-steroids clinical, laboratoristic and endoscopic remission (Mayo Score of 0), at the time of AZA withdrawal. Each pt received a minimum AZA dose of 2 mg/kg/day (no cases of dose reduction due to intolerance or toxicity were considered) for a minimum of 3 y. Relapse was defined as any recurrence of symptoms of UC.


46 steroid-dependent UC were followed for a mean time of 4.4 y (range 1–13) after AZA withdrawal [M/F: 25/21; mean age at diagnosis: 28 y (range 5–55); mean age of start AZA: 37 y (range 18–59); mean disease duration before AZA: 8.4 y (range 1–23); pancolitis/left-sided colitis: 19/27; endoscopic activity at baseline: mild 14 (30%), moderate 32 (70%); mean AZA duration 4.7 y (range 3–10); concomitant therapies: 5-ASA: 42 pts (91%); traditional corticosteroid: 27 pts (59%); beclomethasone dipropionate: 4 pts (9%); need of steroid during AZA: 3 pts (6%); post-AZA therapies: 5-ASA - mean dosage 3.2 g/day -: 44 pts (96%)].

The overall relapse rate after AZA interruption was 37% (17 pts) after a mean time of 2.2 y (range 1–6); the overall colectomy rate was 6.5% (3 pts) after a mean time of 4.7 y (range 2–6).

No difference was registered between pancolitis and left-sided colitis in term of relapse rate [37% (7pts) vs. 37% (10pts)]. Similarly, disease duration before AZA treatment did not influence the relapse rate after AZA withdrawal [relapse rate according disease duration: <5 y, 44% (8 of 18 pts); >5 y: 32% (9 of 28 pts)].


Despite the limit of the small sample size, our results show that a minimum time of 3 y stable deep remission with AZA can be considered a good prognostic factor for maintaining remission after AZA withdrawal.

In this subgroup of pts, 5-ASA seems to be effective to maintain long-term remission.

The stable deep remission seems to be the major prognostic factor of good outcome despite disease extension and disease duration before the treatment.