Background: Our aim was to evaluate low-field magnetic resonance imaging (MRI) in the assessment of disease extension and activity in inflammatory bowel disease.
Methods: Nineteen patients with Crohn's disease (CD), 8 with ulcerative colitis (UC), and 5 healthy controls (HC) were examined using MRI (0.1 T) before and after intravenously administered gadodiamide and glucagon. MRI images were evaluated in a blinded fashion and compared with findings at endoscopy, double-contrast barium enema, small-bowel follow-through, and surgery.
Results: Comparisons of diseased with both non-diseased bowel segments and segments from HC showed significant differences for both CD and UC with regard to signal intensity on T2-weighted (SI(T2)) images and post-contrast increment of signal intensity on T1-weighted images (%SI(T1). Agreements with regard to disease extension in CD between MRI and other examinations were 97%, underestimating the extension in two patients. For SI(T2) in CD a cut-off value of 1.0 showed a predictive value of a positive finding (PVpos) = 1.0 and a predictive value of a negative finding (PVneg) = 0.96. For %SI(T1) in CD a cut-off value of 15.0% showed values of PVpos = 0.95 and PVneg = 0.92. Agreements between MRI and conventional methods (disease extension) in UC was 87.5%. Extension was underestimated in two patients and overestimated in two patients as compared with barium enemas. Values of PVpos were 1.0 (SI(T2) >1.0) and 1.0 (%SI(T1) >15.0%), respectively, with corresponding values of PVneg being 0.94 and 0.94.
Conclusion: Low-field MRI seems a promising non-invasive, non-radiating method in the evaluation of inflammatory bowel disease.