Differentiation of Colonic IBD: Re-examination of PIBD Classes Algorithm With Resected Colon As Gold Standard

OBJECTIVES:Differentiation of Crohnʼs disease (CD) from Ulcerative Colitis (UC) is challenging when inflammation is predominantly colonic. The pediatric inflammatory bowel disease (PIBD) classes algorithm was developed to bring consistency to labelling, but used physician-assigned diagnosis as gold...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of pediatric gastroenterology and nutrition 2019-10
Hauptverfasser: Dhaliwal, Jasbir, Siddiqui, Iram, Muir, Jennifer, Rinawi, Firas, Church, Peter C, Walters, Thomas D, Griffiths, Anne M
Format: Artikel
Sprache:eng
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:OBJECTIVES:Differentiation of Crohnʼs disease (CD) from Ulcerative Colitis (UC) is challenging when inflammation is predominantly colonic. The pediatric inflammatory bowel disease (PIBD) classes algorithm was developed to bring consistency to labelling, but used physician-assigned diagnosis as gold standard. We aimed to reassess the PIBD classes using pathology of subsequently resected colon as gold standard. METHOD:Single-centre study of patients diagnosed with colonic IBD between 2002–2017 and subsequently treated with colectomy. Baseline pre-treatment data were reviewed and the PIBD classes algorithm was independently applied by two reviewers to assign a label of UC/IBD-unclassified (IBD-U)/colonic-CD. Concordance between the algorithm-based, pre-colectomy clinical, and pathologic examination of resected colon diagnosis were assessed. Changes in diagnosis during post-colectomy follow-up were recorded. RESULTS:Sixty-two children underwent colectomy for medically refractory colonic IBD. Diagnosis based on pathologic review of resected colon CD:4; UC:56; IBDU:2. The clinical, PIBD classes algorithm, and colectomy diagnoses were concordant in 51/62patients (81%, Fleiss’Kappa0.48). Pre-colectomy clinical diagnosis was concordant with colectomy diagnosis in 58/62patients (94%, weighted-kappa0.65). The PIBD classes label was concordant with colectomy diagnosis in 51/62patients (82%, weighted-kappa0.38); resected colon pathology was typical of UC in 6 patients with PIBD classes label of IBD-U based on single class 2 feature and in 3 with PIBD classes label of CD based on single class 1 feature. CONCLUSION:Concordance of PIBD classes algorithm diagnosis applied prior to colectomy with a diagnostic label based on pathologic examination of a subsequently resected colon is only fair. Caution is needed in stringent application of colonic CD and IBD-U labels based on presence of single feature.
ISSN:0277-2116
1536-4801
DOI:10.1097/MPG.0000000000002544