Premium
Accuracy of point‐of‐care intestinal ultrasound for Crohn's disease
Author(s) -
Wright Emily K.,
Wang Ian,
Wong Darren,
Bell Sally J.,
Connell William R.,
Thompson Alexander J.,
Novak Kerri L.,
Kamm Michael A.
Publication year - 2020
Publication title -
australasian journal of ultrasound in medicine
Language(s) - English
Resource type - Journals
eISSN - 2205-0140
pISSN - 1836-6864
DOI - 10.1002/ajum.12218
Subject(s) - medicine , kappa , colonoscopy , ultrasound , diagnostic accuracy , receiver operating characteristic , cohen's kappa , gastroenterology , radiology , colorectal cancer , philosophy , linguistics , cancer , machine learning , computer science
Background Point‐of‐care ultrasound (POCUS), performed by a gastroenterologist, provides safe and convenient imaging allowing for immediate clinical decision in Crohn's disease. The minimum training required to gain competency, its accuracy and clinical utility requires evaluation. Methods In this pilot study, Crohn's disease activity and extent were assessed using POCUS (performed by a single gastroenterologist following the completion of 200 supervised scans), magnetic resonance enterography (MRE) and ileo‐colonoscopy. The presence of complications was assessed by POCUS and MRE. Accuracy of POCUS was analysed with respect to MRE and ileo‐colonoscopy. Agreement between modalities was assessed using kappa coefficient. Results Forty‐two patients had a POCUS paired with MRE. Thirty‐eight patients had a POCUS paired with ileo‐colonoscopy. When compared to MRE, POCUS was accurate in the assessment of disease activity (sensitivity 87.5%, specificity 61.1%, ROC 0.74), extent (sensitivity 77.8%, specificity 83.3%, ROC 0.81) and complications (sensitivity 85.7%, specificity 94.3%, ROC 0.90). Agreement between POCUS and MRE was moderate (kappa estimates 0.50, P < 0.001, 0.61, P < 0.001 and 0.76, P < 0.001) for disease activity, extent and complications, respectively. When compared to ileo‐colonoscopy, POCUS was accurate in the assessment of disease activity (sensitivity 72%, specificity 86%, ROC 0.79) and extent (sensitivity 85.7%, specificity 86%, ROC 0.86). For POCUS and ileo‐colonoscopy, kappa estimates were 0.55, P < 0.001 for disease activity and 0.62, P < 0.001 for disease extent. Conclusion POCUS performed by a gastroenterologist after completion of limited training is accurate for assessing Crohn's disease activity, extent and the presence of complications.