Introduction Thiopurines (TPs) including azathioprine and mercaptopurine have an established role in maintaining disease remission in ulcerative colitis (UC) but their impact on the risk of colectomy remains unknown. Our aim was to establish predictors of colectomy and determine the effect of TPs on the risk for colectomy using the Clinical Practice Research Datalink (CPRD).
Methods We conducted a population based study using electronic primary care records in the UK. We identified incident cases of UC between 1989 and 2009 in the CPRD which contains prescribing and clinical data for 8% of the UK population and is a validated research database. Patients were included if registered with a practice for at least 12 months prior to diagnosis. We compared rates of colectomy between TP users and non-users and examined the impact of treatment duration. We performed survival analysis using the Kaplan-Meier method. Independent risk factors for surgery were determined using a multivariable Cox proportional hazards model.
Results Overall, 8673 patient met our inclusion criteria of which 479 (5.5%) went on to have a total colectomy during 21 years of follow up. 51.4% were male and the median age was 46 years (IQR: 32–62 years). Female patients were less likely to undergo colectomy (HR 0.70, 95% CI: 0.58–0.84, p < 0.001). There was a threefold increase in the risk of colectomy amongst TP users compared with non-users (HR 3.48, 95% CI: 2.84–4.37, p < 0.001). Of the TP users, those receiving greater than 12 months of treatment, were least likely to undergo colectomy compared with users for less than 3 months (HR 0.29, 95% CI: 0.21–0.40, p < 0.001). Early steroid users were almost twice as likely to undergo colectomy (HR 1.94, 95% CI: 1.59–2.37, p < 0.001). 5-ASA use was protective, with users 65% less likely to undergo colectomy than non-users (HR 0.35, 95% CI: 0.28–0.44, p < 0.001).
Conclusion Male sex, TP use and early steroid use within 3 months of diagnosis are predictors of colectomy in UC. Amongst TP users, sustained TP use for greater than 12 months duration, was associated with a reduction in colectomy rates. 5-ASA use was associated with a two thirds reduction in risk of colectomy. Our findings support the role of 5-ASA use and prolonged TP treatment for UC patients with a severe disease phenotype.
Disclosure of Interest None Declared.