P247. Development of a new tool to assess Crohn's disease magnetic resonance inflammation severity
Y. Bouhnik1, M. Zappa2, V. Abitbol3, M. Lewin4, M. Boudiaf5, J. Cosnes6, J.Y. Mary7, 1Hopital Beaujon, Gastro Enterologie Nurtition, France, France, 2Hopital Beaujon, Radiologie, Clichy, France, 3Hôpital Cochin, Gastroenterology, Paris, France, 4Hopital Paul Brousse, Imagerie, Villejuif, France, 5Hopital Larioboisiere, Imagerie, Paris, France, 6Hopital Saint Antoine, Gastro Enterologie, Paris, France, 7Hopital Saint Louis, SBIM, Paris, France
The degree of inflammation of Crohn's disease (CD) lesions is a key component of the therapeutic choice as inflammatory lesions may reverse under medical treatment.
We aimed to develop a small bowel CD Magnetic Resonance index of Inflammation Severity (CDMRIS). Thirteen centres participated in this prospective transversal study and selected MR-DVDs from 6 patients with small bowel CD, stratified on considered inflammation severity, 2 nil-mild, 2 intermediate and 2 severe. The 78 MR-DVDs were blinded and allocated to centres using balanced incomplete block design per stratum of severity. A combined reading was conducted by a pair of gastroenterologist and radiologist in each centre. Each DVD was read by 4, 9 and 4 pairs and each pair evaluated 8, 18 and 8 DVDs per severity stratum. Globally, each reader-pair had to evaluate 34 MR-DVDs sent in 3 successive sets at 2 mo interval. Total length and number of 20 cm-diseased segments were noted. In each segment, the following data were recorded: maximal wall thickness, minimal luminal diameter, T2 wall hypersignal, deep ulceration without fistula, comb sign, abscess, sclerolipomatosis, lymph nodes, T1 degree of intensity enhancement (mild or moderate, severe) and pattern (homogeneous, layered), fatty proliferation (blurred wall or inflammatory mass) and fistula (blind, internal cutaneous). For each MR examination, global inflammation severity (GIS) was evaluated quantitatively between zero and 100 using a linear analog scale. To construct the CDMRIS, multivariate linear mixed model was used with forward selection through likelihood ratio test. The dependent variable was the GIS and the independent variables were the number and the length of diseased segments, the number of segments in which each sign was observed, the maximal wall thickness and the minimal luminal diameter. Reader pair and examinations were considered as random factors.
438 readings are available. Mean±SD of GIS was 20.7±20.0. The independent predictors of the GIS were the number of segments (n) with the following independent predictors: T1 mild–moderate intensity of enhancement (MMT1), T1 severe intensity of enhancement (ST1), deep ulceration without fistula (DU), comb sign (CS), any fistula (F) and abscess (A). This leads to build the indices' formula: CDMR-IS = 2·nMMT1 + 3·nST1 + nDU + nCS + 3·nF + 4·nA. These predictors explained 70% of the GIS variance.
The CDMR-IS under construction is a standardized index to evaluate the severity of inflammation in patients with small bowel CD. Further studies are mandatory for validation of CDMR-IS in new patients.