P173 Contrast-enhancement at magnetic resonance enterography does not differentiate between fibrosis and inflammation in Crohn's disease: a prospective cohort study
G. Fiorino*1, A. Spinelli2, C. Bonifacio3, P. Spaggiari4, F. Grizzi5, A. Gandelli1, E. Morenghi6, M. Allocca1, M. Sacchi2, M. Roncalli4, L. Balzarini3, M. Montorsi2, A. Malesci7, 8, S. Vetrano1, S. Danese1
1Humanitas Research Hospital, IBD Center, Gastroenterology, Rozzano, Milan, Italy, 2Humanitas Research Hospital, IBD Surgery, Colo-rectal Surgery, Rozzano, Milan, Italy, 3Humanitas Research Hospital, Radiology, Rozzano, Milan, Italy, 4Humanitas Research Hospital, Anatomopathology, Rozzano, Milan, Italy, 5Humanitas Research Hospital, Immunology and Inflammation , Rozzano, Milan, Italy, 6Humanitas Research Hospital, Biostatistic, Rozzano, Milan, Italy, 7Humanitas Research Hospital, Gastroenterology, Rozzano, Milan, Italy, 8University of Milan, Translational Medicine, Milan, Italy
Contrast-enhancement (CE), measured at magnetic resonance enterography (MRE), is one of the established parameters of inflammation in Crohn's disease. No studies at the best of our knowledge has quantified fibrosis and correlated with MRI parameters. Therefore, we aimed to investigate the correlation the relative CE (RCE) seen at MRI prior to surgical ileo-colonic resection for complicated Crohn's disease (CD) and the quantity (measured as the actual percentage) of fibrosis and inflammation.
We prospectively enrolled CD subjects with planned ileo-colonic surgery for complicated CD. They underwent MRE not earlier than 4 weeks prior to surgery. RCE was calculated at the level of 1 cm above the ileo-ciecal valve or anastomosis to be resected. After surgery, surgical samples were cut and processed. All the histological sections were digitized using a computer-aided image analysis system at 10x objective magnification. Ad-hoc software automatically selected the total CD45+ immunoreactive surface (IRA, %) as a marker of inflammation, or Sirius red stained surface (SS, %) as a marker of fibrosis, on the basis of RGB (Red, Green, Blue) color segmentation. Spearman's correlation test was used to assess the correlation between RCE values and the percentage of inflammation and fibrosis, as well as the pattern of inflammation (dispersed/aggregated). Statistical significance was set as p<0.05.
Thirty-six subjects were enrolled, 27 subjects were included into the final analysis. Ten subjects (37%), underwent surgery for stricturing disease 5 subjects (18.5%) for stricturing and penetrating disease, 12 subjects (44.5%)for penetrating disease. In all subjects, fibrosis and inflammation were found coexisting. In the entire cohort, median RCE ranged from 0.18 to 2.33 (median 0.99), inflammation ranged from 2% to 20% (median 8.86%), fibrosis from 2% to 32% (median 10.23%) No correlation was found between RCE and the percentage of inflammation (R2= 0.0036, p=0.77), and with fibrosis (R2= 0.008, p=0.15). Also no correlation was found between RCE and patterns of inflammation, both dispersed and aggregated (R2= 0.0014, p=0.19).
In CD RCE was found not to correlate with the grade of inflammation and fibrosis. RCE may not be able to discriminate fibrosis and inflammation in CD in the preoperative settings.