P709 Is azathioprine safe as long-term treatment in paediatric patients with inflammatory bowel disease?
Abstract Background The toxicity of azathioprine (AZA) includes myelosuppression, infections, pancreatitis, photosensitivity, and hepatotoxicity. The aim of this study was to describe the adverse effects profile of azathioprine as long-term treatment in paediatric inflammatory bowel disease (IBD). Methods An observational, descriptive and retrospective study was performed in the paediatric IBD Unit of a tertiary care hospital from September 2008 to December 2018. It was included patients under 18 diagnosed with IBD who were treated with AZA during their follow-up. We recorded epidemiological data, thiopurine methyltransferase (TPMT) enzyme activity, AZA side effects, and the dosage the patients were receiving when these effects took place. Bone marrow suppression (BMS) was defined as leukopenia, thrombocytopenia and/or anaemia. Acute pancreatitis (AP) induced by azathioprine was considered when two of these criteria (Atlanta 2012) were met: lipase increase (> 3 times normal value), congruent signs and symptoms and/or echographic findings, without other possible aetiology and with complete recovery after AZA withdrawal. Results We included 52 patients, being 31 men (59.6%). They were diagnosed with Crohn′s disease (CD) (73%), ulcerative colitis (UC) (21%) and IBD-unclassified (6%). The median TPMT activity was 17 U/ml (14.2–19.2). Up to 63.5% developed adverse effects by AZA with a median time from the beginning of treatment of 11.4 months (2.6–26.4) and a median dosage of 2 mg/kg/day (1.7–2.3). The most frequent side effect was BMS (52%). These patients had a median TPMT activity of 16.9 U/ml (14.2–18.9), the median duration of treatment was 14 months (3.9–27.7), and the median dosage was 2 mg/kg/d (1.8–2.5). BMS was more frequent in patients with UC (p 0.04) and longer treatment (p 0.08). No differences were found according to age, sex or TPMT activity. Up to 11.5% developed AP, the median duration of treatment until its appearance was 1.5 months (0.7–43.3) and the median dosage was 2 mg/kg/d (1.5–2.5). No differences were found related to age, sex, diagnosis or dosage. Other side effects were: 3 flu-like symptoms, 3 opportunistic infections, 2 hypertransaminasemia, and 1 patient with elevated pancreatic enzymes and hyperbilirubinemia. AZA was discontinued in 14 patients (43.8%): in 6 due to AP, in 4 due to severe lymphopenia, in 2 because of Epstein-Barr virus infection, in 1 due to flu-like symptoms and in 1 with several adverse effects. Conclusion More than half of the patients treated with AZA presented side effects, mainly BMS, although most of them were mild and temporary, and the withdrawal of the drug was not necessary. It seems that TPMT activity is not useful to predict BMS, but this adverse effect could be related to a longer treatment.
