Evaluation of treatment response in active Crohn's disease by low-field magnetic resonance imaging

Background: To evaluate low-field magnetic resonance imaging (MRI) in detecting therapeutic response in active Crohn's disease during treatment with systemic steroids. Methods: Eight patients with active Crohn's disease were examined before and during treatment with systemic steroids (1 mg/kg/day) using low-field MRI (0.1 T) in transverse and coronal planes before and after an intravenously administered bolus of gadodiamide. Five healthy persons were once examined in the same way. MRI images were evaluated without knowledge of diagnosis, treatment, or findings of endoscopy, conventional radiography, and surgery. Proximal and mid small bowel, terminal ileum, right-sided colon, transverse colon, and left-sided colon were evaluated separately. Results: Statistically significant differences were shown for both signal intensity on T2- (SIT2) and increment in signal intensity on T1-weighted images after contrast (%SIT1) when comparing diseased bowel segments with both nondiseased bowel segments (SIT2: p = 0.0001; %SIT2: p = 0.0009) and segments from the control group (SIT2: p < 0.00005; %SIT1: p < 0.00005). In 53 of 56 bowel segments evaluated (95%), agreement was found between findings by MRI, conventional radiography, endoscopy and/or surgery regarding disease extension. Extension was underestimated in two patients. All bowel segments in the control subjects were evaluated to be normal on MRI. Significant correlation was found between both SIT1 (p < 0.0025) and %SIT1 (p < 0.025) versus endoscopic activity gradings. During treatment, significant decrements of both SIT2 (p < 0.00005), %SIT1 (p = 0.002), and bowel wall thickness (I, = 0.03) were found. Conclusions: Low-field MRI seems to be a promising noninvasive method in the evaluation of response regarding both disease extension and activity in Crohn's disease during treatment with systemic steroids.