Search in the Abstract Database

Abstracts Search 2014

* = Presenting author

DOP081. Long-term natural history of postoperative recurrence in patients on preventive treatment with azathioprine

M. Mañosa1, Y. Zabana2, L. Marin1, I. Bernal1, J. Boix1, M. Piñol1, E. Cabré1, E. Domènech1, 1Hospital Universitari Germans Trias i Pujol and CIBERehd, Gastroenterology Unit, Badalona, Spain, 2Hospital Universitario Mutua de Terrassa and CIBERehd, Gastroenterology Unit, Terrasa, Spain


The postoperative recurrence (POR) in Crohn's disease (CD) occurs in >75% within the first year after intestinal resection if no preventive treatment is started. Nowadays, azathioprine (AZA) is the most prescribed drug to prevent POR, but its long-term efficacy is unknown and no recommendations about POR monitoiring beyond the first year after surgery are available.

Aims: To evaluate the long-term clinical and endoscopic outcomes of CD after intestinal resection and early preventive therapy with AZA.


From an specific database in which all patients with CD who underwent resection with anastomosis at our institution since 1998 were prospectively included and followed, we identified those who initiated AZA (associated or not with metronidazole or 5-ASA) within the first month after surgery and with at least a follow-up of 3 years. Endoscopic recurrence (ER) was defined as a Rutgeerts score >1 and clinical recurrence (CR) as the development of symptoms that required changes in the treatment for CD. Surgical recurrence (SR) was considered as the need for surgery. We defined a Combined Outcome as any combination of the following events: rescue with biological agents, CR or SR.


189 patients were included of whom 57% male, 64% active smokers at the time of surgery, 54% penetrating behaviour. 58% of patients had ER after a median of 22 months (IQR 11.5–44.5). The cumulative probability of ER was 35%, 48% and 59%, the probability of CR was 18%, 27 and 34% and for SR was 3% 10% and 16%, at 3, 5 and 10 years, respectively. Only active smoking after surgery was associated with POR. The risk for the combined outcome was 21%, 23% and 46% at 3, 5 and 10 years. In patients without ER at the first endoscopic control, the probability at 3, 5 and 10 years of CR was 14%, 22% and 27%; for SR 6%, 9% and 9%; and for the combined outcome of 13%, 26% and 38%, respectively. In the log-rank analysis, the cumulative probability of CR or SR was significantly higher among those patients with early ER (at the first control after surgery - p = 0.044 and p = 0.05).


The use of AZA after surgical resection in Crohn's disease is associated with a low rate of CR and SR, probably because of early introduction of rescue therapy with biological in those patients with advanced endoscopic lesions. Patients without early ER, although at lower risk have a slow but steady increase in the development of ER and CR upon time, suggesting that periodical assessment of POR should be kept indefinitely.