Comparison of Three Magnetization Transfer Ratio Parameters for Assessment of Intestinal Fibrosis in Patients with Crohn's Disease

ABSTRACT

OBJECTIVE: To establish a novel standardized magnetization transfer ratio (MTR) parameter which considers the element of the normal bowel wall and to compare the efficacy of the MTR, normalized MTR, and standardized MTR in evaluating intestinal fibrosis in Crohn's disease (CD).MATERIALS AND METHODS: Abdominal magnetization transfer imaging from 20 consecutive CD patients were analyzed before performing elective operations. MTR parameters were calculated by delineating regions of interest in specified segments on MTR maps. Specimens with pathologically confirmed bowel fibrosis were classified into one of four severity grades. The correlation between MTR parameters and fibrosis score was tested by Spearman's rank correlation. Differences in MTR, normalized MTR, and standardized MTR across diverse histologic fibrosis scores were analyzed using the independent sample t test or the Mann-Whitney U test. The area under the receiver operating characteristic curve (AUC) was computed to test the efficacies of the MTR parameters in differentiating severe intestinal fibrosis from mild-to-moderate fibrosis.RESULTS: Normalized (r = 0.700; p < 0.001) and standardized MTR (r = 0.695; p < 0.001) showed a strong correlation with bowel fibrosis scores, followed by MTR (r = 0.590; p < 0.001). Significant differences in MTR (t = −4.470; p < 0.001), normalized MTR (Z = −5.003; p < 0.001), and standardized MTR (Z = −5.133; p < 0.001) were found between mild-to-moderate and severe bowel fibrosis. Standardized MTR (AUC = 0.895; p < 0.001) had the highest accuracy in differentiating severe bowel fibrosis from mild-to-moderate bowel wall fibrosis, followed by normalized MTR (AUC = 0.885; p < 0.001) and MTR (AUC = 0.798; p < 0.001).CONCLUSION: Standardized MTR is slightly superior to MTR and normalized MTR and therefore may be an optimal parameter for evaluating the severity of intestinal fibrosis in CD.