Els van de Vijver

63 INTRODUCTION Persistent rectal bleeding or perianal disease in children and teenagers justifies endoscopy to evaluate the presence of inflammatory bowel disease (IBD).(1-2) When the indication for endoscopy is less obvious, as in the case of patients with chronic abdominal pain and non-bloody diarrhoea, a triage test may help to distinguish who are in need of immediate referral to endoscopy. Several meta-analyses (3-6) have shown that measuring a single faecal calprotectin level can help to distinguish IBD from functional abdominal disorders. Calprotectin concentrations above 50 µg/g predict the presence of IBD with high sensitivity [99% (range 92 to 100%)], but the mediocre specificity [65% (range 54 to 74%)](6) is the reason that a substantial number of children are wrongly exposed to endoscopy. A refinement of the cut-point to 250 µg/g was insufficient to reduce the rate of unnecessary endoscopies.(7-9) Complications of endoscopy, related to the invasiveness of the procedure itself (colonic perforation or tear) or to anaesthesia, may be rare but could cause severe morbidity.(10- 12) A diagnostic strategy that includes a combination of tests would potentially further reduce the number of children exposed to this invasive and costly procedure. We evaluated four diagnostic strategies to predict the presence of IBD: [1] symptoms alone, [2] symptoms plus blood markers, [3] symptoms plus faecal calprotectin, and [4] symptoms plus blood markers plus faecal calprotectin. PATIENTS AND METHODS Study design This international multi-centre study was a planned ancillary study of the prospective CACATU cohort (clinicaltrials.gov NCT02197780). The cohort and the calprotectin results have previously been described (11) and are replicated here for the subgroup of previously undiagnosed children and teenagers presenting with persistent or recurrent non-bloody diarrhoea and abdominal pain. Patients were assessed by a local clinician and data collected during history taking and physical examination were entered on a secured study website (www.cacatustudie.eu) . Blood tests were performed at the local hospital and the results were uploaded to the study website. Stool samples were sent to the Test strategies to predict IBD 63