P595 Finding predictors of azathioprine-induced pancreatitis in patients with inflammatory bowel disease

Abstract

Abstract Background Azathioprine (AZA)-induced pancreatitis (AIP) is a common, idiosyncratic side effect, whose incidence, clinical course and risk factors data in inflammatory bowel disease (IBD) patients are scarce. We aimed to establish the incidence, describe the clinical course and identify risk factors for AIP. Methods Retrospective study including all IBD patients on AZA between January 2013 and July 2020. Patients with AIP were considered. Demographic, clinical, biochemical and imaging data were collected. Results AIP occurred in 33 patients (7.5%; 442 patients on AZA): 81.8% had Crohn’s disease, 54.5% were male, and the mean age was 35±13 years. The mean time under AZA till AIP was 25±11 days, with a mean dosage of 88±44mg. Eighteen patients (54.4%) were hospitalized, with a mean hospital stay of 4±2days. All patients had a mild course of disease which resolved with suspension of AZA, and with no complications or need of invasive interventions or complications. Smoking (p=0.02), single daily dose of AZA (p<0.001) and concomitant treatment with budesonide (p=0.001) were risk factors for AIP. In multivariate analysis, concomitant treatment with budesonide (OR: 5.3; p=0.002) and single daily dose of AZA (OR: 4.8; p=0.002) were the only predictors of AIP. Conclusion Although AIP was a relatively common side effect, it presented a mild course in all patients. Smoking, concomitant treatment with budesonide and single daily dose of AZA were risk factors for AIP. This study suggests that smoking, concomitant use of budesonide and single dose regimen of AZA should be avoided in IBD patients treated with AZA.