Risk of infections and hospitalisations in inflammatory bowel disease and other immune-mediated diseases on thiopurine therapy: a population-based study

This abstract has open access
Abstract Summary
Abstract ID :
HAC403
Submission Type
HA Staff
Authors (including presenting author) :
Mak WY (1), Tang, Whitney (2), Lam, TO (1), Chan, Francis KL (2), Sung, Joseph JY (2), Ng, Siew C (2)
Affiliation :
(1) Department of Medicine and Therapeutics, Prince of Wales Hospital (2) Department of Medicine and Therapeutics, The Chinese University of Hong Kong
Introduction :
Thiopurines is widely used as an immunosuppressive agent but is associated with a risk of leukopenia. Recently, the presence of NUDT15 variant has been associated with thiopurine-induced leukopenia. Real world data on complications and outcomes of patients with thiopurine-induced leukopenia are scarce.
Objectives :
We aimed to study risks of infections, rates of hospitalisation and intensive care unit (ICU) admission in patients with thiopurine-induced leukopenia.
Methodology :
Patient demographics, disease diagnosis, types of infection, number of hospitalisations and ICU admissions were collected from a territory-wide computerised database of patient records managed by the Hong Kong Hospital Authority covering over 7 million population. Leukopenia was defined as white cell count < 4 x 109/L during thiopurine therapy. The odds ratios for hospitalisation and ICU admissions within 30 days of development of leukopenia were calculated.
Result & Outcome :
From 1 January 2014 to 31 December 2017, 6,185 ethnically Chinese patients received thiopurine (azathioprine or 6-mercaptopurine) for immune-related diseases [32.5% rheumatological diseases, 20.8% inflammatory bowel disease (IBD), 13.3% organ transplantation]. Mean age was 51 years (+/- 19.6 years), and median follow-up was 36 months (interquartile range: 15-47 months). 2,256 patients (36.6%) developed leukopenia during 13,299 person-years of follow-up. Significantly more IBD patients developed thiopurine-induced leukopenia compared with non-IBD patients (45.0% vs. 34.6%, p< 0.01). Patients with thiopurine-induced leukopenia had higher risk of Clostridium difficile infections ( OR 4.95), gastrointestinal (GI) tract infections (OR 2.60), tuberculosis: (OR 2.20), septicaemia (OR: 1.87), fungal infections (OR: 1.75), urinary tract infection (UTI) (OR: 1.63) and pneumonia (OR: 1.30) compared with those without leukopenia. They also had significantly higher number of hospitalizations (mean number, 7.66 vs. 4.17; p< 0.01) and ICU admissions (5.25% vs. 3.25%; p< 0.01). Amongst all patients with thiopurine-induced-leukopenia, IBD patients had more pneumonia, UTI and septicemia than non-IBD patients, but no difference in the number of hospitalizations (mean number, 4.82 vs. 5.60; p=0.10). Conclusion: In a population-based study, 45% of IBD patients developed thiopurine-induced leukopenia and these subjects had a higher risk of infections than non-IBD patients . Thiopurine-induced leukopenia is also associated with more hospitalisations and ICU admissions. Our findings highlight the importance of checking NUDT15 polymorphisms before starting thiopurine therapy.
Presentation Upload :