PTH-125 The clinical utility and diagnostic accuracy of faecal calprotectin for IBD in paediatric patients. (June 2019)

• Record Type: Journal Article
• Authors: Soubieres, Anet; Shandro, Benjamin; Mathur, Jai; Boa, Frances; Paul, Thankam; Poullis, Andrew
• Is Part Of: Gut. Volume 68(2019)Supplement 2
• Page Start: A97
• Page End: A97
• Publication Date: 2019-06
• Subjects: Gastroenterology -- Periodicals; 616.33
• DOI: 10.1136/gutjnl-2019-BSGAbstracts.184
• Languages: English
• ISSNs: 0017-5749

Abstract

Abstract : Introduction: Faecal calprotectin (FCP) has an established place in the adult diagnostic pathway. Its role in the paediatric population, where triage for colonoscopy is vital, is less well studied. There is increasing awareness that the normal range for FCP and the prevalence of IBD vary with age. We aimed to determine the most effective use of FCP in paediatric patients presenting with GI symptoms. Methods: We conducted a retrospective analysis of FCP results for patients aged ≤18 years presenting to paediatric gastroenterology at a London teaching hospital from 2013 to 2014. Demographic and clinical information, including final diagnosis of IBD, was extracted from the Electronic Patient Record. In patients with multiple FCP results, the earliest was used. Abnormal FCP was defined as ≥50µg/g. Contingency tables for FCP and IBD were generated for the total cohort and grouped by age <10 years and ≥10 years. Sensitivity, specificity, positive predictive values (PPVs), negative predictive values (NPVs) and pre- and post-test probabilities were calculated. The analysis was repeated in patients aged <10 years using an FCP threshold ≥160µg/g, which has been adopted locally as the cut-off in this group. Complete case analyses were used where data were missing. Stata version 13.1 was used for all statistical analyses. Results: 356 FCP samples were sent from 328 patients. 49.9% were male, and median age was 10.9 years (range 0.1–18.7). 134 patients (41%) had an abnormal FCP. 90 patients (27.4%) were diagnosed with IBD. The median FCP for patients with IBD was 408.5µg/g vs. 18µg/g for those without IBD. Using an FCP threshold of ≥50µg/g for IBD vs. non-IBD the overall sensitivity was 76.7%, specificity 72.4%, PPV 52.6% and NPV 88.6%. In patients <10 years old, the sensitivity was 100%, specificity 70.2%, PPV 21.7% and NPV 100%. The pre-test probability was 7.6% (low) and post-test probability 21.6%. In patients ≥10 years old, the sensitivity was 73.8%, specificity 75%, PPV 69.4% and NPV 78.8%. The pre-test probability was 43.5% (intermediate) and post-test probability 69.3%. Increasing the FCP threshold to ≥160µg/g in patients <10 years old improved the specificity to 85.1%, but at the expense of sensitivity, which decreased to 80%. The PPV was 30.8%, NPV 98.1% and post-test probability 30.6%. Colonoscopy was carried out in 133 patients (42%), median age 12.5 years (range 0.2–18.5), of which 49 (36.8%) had a normal FCP. Conclusions: FCP is highly accurate at excluding IBD in paediatric patients with GI symptoms and should guide the need for colonoscopy. In low prevalence populations, such as those aged<10 years, a positive result should be interpreted with caution.