Optimising triage of urgent referrals for suspected IBD: results from the Birmingham IBD inception study

Abstract

Objective Diagnostic delays in inflammatory bowel disease (IBD) result in adverse outcomes. We report a bespoke diagnostic pathway to assess how best to combine clinical history and faecal calprotectin (FCP) for early diagnosis and efficient resource utilisation. Methods A rapid-access pathway was implemented for suspected IBD patients referred outside urgent ‘two-week wait’ criteria. Patients were triaged using symptoms and FCP. A 13-point symptom history was taken prediagnosis and clinical indices, including repeat FCP, collected prospectively. Results Of 767 patients (January 2021–August 2023), 423 were diagnosed with IBD (208 Crohn’s disease (CD), 215 ulcerative colitis (UC)). Most common symptoms in CD were abdominal pain (84%), looser stools (84%) and fatigue (79%) and in UC per-rectal bleeding (94%), urgency (82%) and looser stools (81%). Strongest IBD predictors were blood mixed with stools (CD OR 4.38; 95% CI 2.40–7.98, UC OR 33.68; 15.47–73.33) and weight loss (CD OR 3.39; 2.14–5.38, UC OR 2.33; 1.37–4.00). Repeat FCP testing showed reduction from baseline in non-IBD. Both measurements >100 µg/g (area under the curve (AUC) 0.800) and >200 µg/g (AUC 0.834) collectively predicted IBD. However, a second value ≥220 µg/g considered alone, regardless of the first result, was more accurate (Youden’s index 0.735, AUC 0.923). Modelling symptoms with FCP increased AUC to 0.947. Conclusion Serial FCP measurement prevents unnecessary colonoscopy. Two FCPs >200 µg/g could stream patients direct to colonoscopy, with two >100 µg/g prompting clinic review. A second result ≥220 µg/g was more accurate than dual-result thresholds. Coupling home FCP testing with key symptoms may form the basis of effective self-referral pathways. Data are available upon reasonable request.