P128 Diagnostic accuracy of faecal calprotectin in Crohn's disease – Does disease location matter?

Abstract

Background: Faecal calprotectin (FC) is a highly sensitive disease activity biomarker in Inflammatory Bowel Disease. However there are conflicting reports on whether the diagnostic accuracy in Crohn's disease (CD) is influenced by disease location. The aim of this study is to undertake a systematic review of published literature to compare the sensitivity and specificity of FC to accurately measure disease activity at small bowel (SB) vs large bowel (LB) location. Methods: Databases (Medline, Embase, Web of Science and Cochrane) were searched from inception to November 8th 2016 for cohort and case control studies which had data on FC in patients with isolated SB and LB CD. Similarly, relevant conference proceedings were searched from 2005–2016. There was neither age nor language restriction. The reference standard for activity was either endoscopy, magnetic resonance imaging, computed tomography, technetium scan or a combination of these. We excluded studies reporting on post-operative CD or on a specific disease location in isolation. Screening was done independently in duplicate (EGS, RW), with any disagreements resolved by 2 other authors (GWM, SS). EGS & GWM independently completed data extraction form. To assess the risk of bias; EGS & GWM used QUADAS-2, a research tool to check the quality of systematic reviews of diagnostic accuracy studies. Any disagreement was resolved by consensus with co-authors. Communication was undertaken with all lead authors in order to obtain missing data sets. Whenever possible, sensitivities and specificities were obtained from the raw data or as reported in the publication. Results: 5619 records were identified at initial search. 2098 duplicates were removed and 3521 records were screened. From the latter, 61 full text articles were then assessed for eligibility. 45 studies were later excluded and 16 studies were included in the final review, with sensitivities and specificities per disease location available from seven studies. The sensitivity of FC in the SB location ranged from 42.9% to 100% with a median of 75%, while that in the LB location ranged from 78.9% to 100% with a median of 94%. FC specificity in the SB location ranged from 50% to 100% while that in the LB location ranged from 28.6% to 100% with similar median specificities of 75% and 71% respectively. Conclusions: The sensitivity of FC to accurately measure disease activity for CD in the SB appears to be lower than that in the LB. Limitations of the study include heterogeneity associated with the cut offs of FC, disease spectrum, study design, gold standard used and quality of studies as well as insufficient raw data.