P745 Perianal disease in Crohn’s disease is associated with higher rate of hospitalisation and mortality despite a higher utilisation of biologics: Report from the epiIIRN cohort

O. ATIA1, N. Asayag1, G. Focht1, A. Cahan2, I. Dotan3, R. Balicer4, M. Gavish5, B. Feldman4, I. Brufman4, Z. Haklai6, D. Turner1, Israeli IBD Research Nucleus (IIRN)

1Shaare Zedek Medical Center- The programmesity of Jerusalem- Israel., Juliet Keidan Institute of Pediatric Gastroenterology Hepatology and Nutrition, Jerusalem, Israel, 2Maccabi Health Services- Tel-Aviv- Israel, Maccabi Health Services, Tel-Aviv, Israel, 3IBD Center- Department of Gastroenterology and Liver diseases and the Research Center for Digestive Diseases- Tel-Aviv Sourasky Medical Center- and the Sackler School of Medicine-, IBD Center- Department of Gastroenterology and Liver diseases, Tel-Aviv, Israel, 4Clalit Research Institute- Chief Physician’s Office- Clalit Health Services, Clalit Research Institute, Tel-Aviv, Israel, 5The Hebrew University, School of Computer Science and Engineering, Jerusalem, Israel, 6Ministry of health Israel, Ministry of health Israel, Jerusalem, Israel

Background

Perianal disease is a disabling complication of Crohn’s disease (CD). In this study, we aimed to explore the incidence of fistulising perianal CD (PCD) and its association with treatment pattern and disease outcomes, within the validated epiIIRN cohort which includes all IBD patients in Israel (n = 45,074). We also aimed to differentiate simple from complicated PCD.

Methods

The current analysis was performed on data from the two largest Health Maintenance Organisations (HMOs), covering 75% of the Israeli population. First, we developed and validated algorithms to identify fistulising PCD patients and differentiate simple from complex disease, by establishing two reference cohorts of true positive PCD patients (TP, patients with perianal abscess/fistula) classified into simple or complicated PCD, and true negatives (TN) of CD patients without perianal disease (validated by chart reviews). We then applied the algorithms on the complete epiIIRN IBD cohort. Random charts were reviewed to validate again PCD diagnosis.

Results

The chosen algorithm included patients with at least one perianal-related diagnosis, procedure or surgery (sensitivity/specificity 75%/96% for identifying PCD, and 70%/79% for differentiating complicated vs. simple PCD). Of the total 26,440 CD patients, 4237 (16%) had fistulising PCD, of whom 1004 (4% of the total) had complicated PCD. Those with complicated PCD were diagnosed with IBD at a younger age than simple PCD and non-PCD (29.2 ± 13.6 vs. 32.2 ± 16 and 32.8 ± 16.7 years, respectively) (p < 0.001). PCD patients were treated more often with biologics (32% in non-PCD compared with 47% in simple PCD and 70% in complicated PCD, p = 0.01), steroids (74%, 78% and 83%, respectively; p < 0.001) and immunomodulators (56%, 64% and 67%, respectively; p < 0.001). Despite this, PCD patients had more hospitalisations (median 0 (IQR 0–2), 5 (2–11) and 13 (7–22) in non PCD, simple PCD and complicated PCD, respectively, p < 0.001), more colon surgery (5%, 18% and 52% in non PCD, respectively, p < 0.001) and higher mortality compared with CD patients without perianal disease (p < 0.001).

Conclusion

Conclusion: Despite the higher utilisation of immunomodulators and biologics, patients with PCD are at higher risk of hospitalisation and surgical resection, higher need for steroids and higher mortality. PCD is a strong predictor of disease course and thus should prompt early intensified treatment and close monitoring.

This study was supported by a grant from the Leona M. and Harry B. Helmsley Charitable Trust.