Introduction Risk of complications in Crohn's disease (CD) is substantial but varies according to age at onset, smoking and possibly disease location. We examined a well characterised cohort of Scottish CD patients to explore how disease location modified time to first resection and time to complications.

Methods Consenting CD patients attending GI clinics in Edinburgh and Dundee participated. Patients' disease behaviour was documented according to the Montreal classification. First surgical resection was defined as the first operation removing a portion of intestine, excluding simple appendicectomies or defunctioning procedures. Time to complication was defined as the time from diagnosis to the first stricturing or internally penetrating complication.

Results 1150 patients were recruited, of which full phenotypic data were available in over 95%. The median age at onset was 28.6 years (IQR 21.2–43.1) with 10.2% A1, 60.6% A2 and 29.2% A3. The median length of follow-up was 10.3 years (IQR 4.1–19.8 years). The maximum Montreal location was L1 in 27.9%, L2 in 35.2%, L3 in 24.4%, pure L4 in 2% and L4 plus another location in 10.5%. On Kaplan Meier analysis the median time to first resection was 8.9 years (95% CI 7.5 to 10.1). At 20 years following diagnosis only 31% of patients were resection free. On Kaplan–Meier analysis the median time to first complication was 14.5 years (95% CI 10.5 to 19.5 years), and at 20 years 45% of patients were without complication. Cox proportional hazards models were used to examine disease location with respect to time to first resection and time to first complication, adjusting for age at diagnosis and smoking status at diagnosis as necessary.

When compared to the average risk of resection in the population studied, pure colonic disease had an HR for resection of 0.45 (95% CI 0.41 to 0.49) whereas pure ileal disease had an HR of 2.55 (95% CI 1.96 to 1.62). Ileocolonic disease conferred an intermediate risk of 1.23 (95% CI 0.93 to 1.62). When compared to the average risk of complication in the population, pure colonic disease conferred an HR of 0.34 (95% CI 0.34 to 0.34). Pure ileal disease conferred an HR of 1.87 (1.37 to 2.54). Ileocolonic disease had an intermediate HR of 1.17 (95% CI 0.84 to 1.63).

Conclusion This study, while confirming that colonic disease confers a lower risk of surgery and complications than other disease locations, also suggests that the risks are greater in pure ileal disease when compared with ileocolonic disease. Clinicians making treatment decisions should take CD location into account.