Natural history of Crohn's disease and changes in the use of immunomodulators over time: the Sydney inflammatory bowel disease cohort 1942-2012

Abstract

Background: Immunomodulators (IM) are proven to be effective in the induction and maintenance of remission in Crohn’s disease (CD). However, few studies have evaluated changes in its use over longitudinal follow-up and impact this has made on the natural history of CD.Aims The study aims to determine the changes in use of immunomodulators during longitudinal follow-up in a large metropolitan cohort of ambulatory inflammatory bowel disease (IBD) patients. Methods: The Sydney IBD Cohort is a cohort study based around the Sydney Local Health District with recruitment by specialists in private rooms and hospitals. The medical records were reviewed for patient demographics, disease characteristics, drug treatments and surgical management. Three time periods were assessed: pre-1990, 1990–2000, and post-2000. Kaplan-Meier analysis was used to estimate the cumulative probability of initiation of IM and the log rank test to determine significance. Predictive factors affecting the likelihood of IM initiation over time were investigated using the Cox proportional hazard regression model. Results: A total of 406 CD patients (53% females) with complete follow-up data were included in the analysis. The median length of follow-up was 9 years (IQR 4–17) with an overall 4,800 patient-years of follow-up. Age of diagnoses was over 16 years old in 92.6%. Location phenotypes were ileo-colonic 39.8%, ileal 24.6%, and colonic 35.6%. The stricturing phenotype was found in 20.4%, the penetrating phenotype in 5.3% of patients and perianal disease in 17.9%. Ever smokers comprised of 36% of the cohort. There were total of 241 (60.2%) patients who were exposed to IM with 146 (36.1%) being current users (31.8% thiopurines and 4.7% methotrexate at the time of follow-up). Percentage of patients on IM in the 3 time periods considered did not differ (P = 0.6). However, the time of introduction of an IM significantly reduced over the three time periods [P < 0001], Figure 1.On multivariate Cox regression analysis, earlier time periods of diagnosis (hazard ratio (HR): 7.81, 95% CI: 3.7–16.48, P < 0001 and 21.23, 95% CI: 9.51–47.40, P < 0.001), age at onset (HR: 0.977, 95% CI: 0.96–0.99, P < 0.0001), perianal involvement at diagnosis (HR: 1.69, 95% CI: 1.14–2.52, P = 0.009), need for systemic steroids (HR: 2.83, 95% CI: 1.69–3.35, P < 0.001) and requirement for surgical resection (HR: 1.493, 95% CI: 1.00–2.22, P = 0.05) were all significantly associated with the time to initiation of IM therapy. In total, 138 first surgical resections were noted in the cohort, and 41.3% of these had surgical recurrence. IM significantly decreased the need for subsequent surgical recurrence (P = 0.039).Conclusion: Earlier introduction of IM over time is associated with reduced surgical recurrence rate.