Search in the Abstract Database

Abstracts Search 2019

P101 Simple five-point classification of MR enterocolonography for Crohn’s disease based on enteroscopic findings

T. Fujii*1, Y. Kitazume2, K. Takenaka1, K. Suzuki1, M. Motobayashi1, E. Saito1, M. Nagahori1, K. Ohtsuka1, M. Watanabe1

1Tokyo Medical and Dental University, Gastroenterology and Hepatology, Tokyo, Japan, 2Tokyo Medical and Dental University, Radiology, Tokyo, Japan

Background

Crohn’s disease (CD) is a chronic progressive inflammatory bowel disease. Assessing the severity and extent of the disease is critical to determine appropriate therapeutic strategies in patients with CD. Magnetic resonance (MR) enterography can assess both intestinal walls and extraintestinal structures without radiation exposure and anaesthesia, which makes it appropriate for repeated evaluation in CD patients. We developed novel MR enterocolonography (MREC) for simultaneously evaluating large and small intesntinal lesions of CD. The aim of this study was to establish the efficacy of the simplified 5-point MREC classification for assessing CD activity, comparing to the validated MR score of magnetic resonance index of activity (MaRIA) and endoscopic findings.

Methods

A total of 120 patients (70 for derivation cohort and 50 for validation cohort) with CD were enrolled and undergone MREC and ileocolonoscopy or balloon-assisted enteroscopy (BAE). MREC results were evaluated for each bowel segment; rectum, sigmoid, descending, transverse, ascending colon, terminal, proximal ileum, and jejunum, by one observer in the derivation phase, and independently by three observers in the validation phase, using the simplified 5-point MREC (sMREC) classification lexicon and MaRIA. Areas under the receiver-operating characteristic curves (AUCs) were obtained to assess the accuracy of discriminating deep ulcers. Inter-observer reproducibility was assessed using weighted Kappa coefficients.

Results

The AUCs of sMREC classification were 89.0% in the derivation phase and 88.5, 81.0, and 77.3% for three observers in the validation phase. The AUCs of MREC classification were statistically non-inferior to those of MaRIA (p < 0.001). The cross-validation accuracy was 81.9% in the derivation and 81.5% in the validation phase. sMREC classification showed enough reproducibility.

Conclusion

In clinical practice, scoring systems should be simple and provide appropriate levels of accuracy and reproducibility. sMREC classification met these requirements, and was demonstrated to be useful for evaluating CD activity in the large and small intestine.