RESUMEN
BACKGROUND AND AIMS: Little is known about the risk factors of colectomy in patients with ulcerative colitis (UC) under thiopurine treatment. The aim of the study was to determine the prevalence and the predictive risk factors of colectomy in an extensive cohort of patients with UC treated with thiopurines in Spain. METHODS: Among 5753 UC patients, we identified those diagnosed between 1980 and 2009 and treated with azathioprine or mercaptopurine (AZA/MP). We analyzed the age at diagnosis, familial history of IBD, extraintestinal manifestations (EIMs), disease extent, smoking status and treatment requirements (AZA/MP, cyclosporine (CsA) or anti-TNFα). Colectomies for dysplasia or cancer were excluded. Survival analysis and Cox proportional hazard regression were performed. Results were reported as hazard ratios (HR) with 95% CI. RESULTS: Among the 1334 cases included, 119 patients (8.9%) required colectomy after a median time of 26 months (IQR 12-42) after AZA/MP initiation. Independent predictors of colectomy were: Extensive UC (HR 1.7, 95% CI: 1.1-2.6), EIMs (HR 1.5, 95% CI: 1.0-2.4), need for antiTNFα (HR 2.3, 95% CI: 1.5-3.4) and need for CsA (HR 2.4, 95% CI: 1.6-3.7). Patients requiring early introduction of AZA/MP had an increased risk of colectomy with a HR of 4.9 (95% CI: 3.2-7.8) when AZA/MP started in the first 33 months after UC diagnosis. CONCLUSIONS: Nearly one-tenth of patients with UC under thiopurines require colectomy. Extensive UC, EIMs, need for CsA or anti-TNFα ever and an early need for AZA/MP treatment were associated with a higher risk of colectomy. These risk factors of colectomy could help to stratify risk in further controlled studies in UC.