Dynamic contrast-enhanced MRI in patients with luminal Crohn's disease.

OBJECTIVES: To prospectively assess dynamic contrast-enhanced (DCE-)MRI as compared to conventional sequences in patients with luminal Crohn's disease. METHODS: Patients with Crohn's disease undergoing MRI and ileocolonoscopy within 1 month had DCE-MRI (3T) during intravenous contrast injection of gadobutrol, single shot fast spin echo sequence and 3D T1-weighted spoiled gradient echo sequence, a dynamic coronal 3D T1-weighted fast spoiled gradient were performed before and after gadobutrol. Maximum enhancement (ME) and initial slope of increase (ISI) were calculated for four colon segments (ascending colon+coecum, transverse colon, descending colon+sigmoid, rectum) and (neo)terminal ileum. C-reactive protein (CRP), Crohn's disease activity index (CDAI), per patient and per segment Crohn's disease endoscopic index of severity (CDEIS) and disease duration were determined. Mean values of the (DCE-)MRI parameters in each segment from each patient were compared between four disease activity groups (normal mucosa, non-ulcerative lesions, mild ulcerative and severe ulcerative disease) with Mann-Whitney test with Bonferroni adjustment. Spearman correlation coefficients were calculated for continuous variables. RESULTS: Thirty-three patients were included (mean age 37 years; 23 females, median CDEIS 4.4). ME and ISI correlated weakly with segmental CDEIS (r=0.485 and r=0.206) and ME per patient correlated moderately with CDEIS (r=0.551). ME was significantly higher in segments with mild (0.378) or severe (0.388) ulcerative disease compared to normal mucosa (0.304) (p<0.001). No ulcerations were identified at conventional sequences. ME correlated with disease duration in diseased segments (r=0.492), not with CDAI and CRP. CONCLUSIONS: DCE-MRI can be used as a method for detecting Crohn's disease ulcerative lesions.

Grading luminal Crohn's disease: which MRI features are considered as important?

OBJECTIVES: Magnetic resonance imaging (MRI) is increasingly used for disease activity grading in small bowel Crohn's disease. It is not known which imaging features are essential for grading. For further insight, we solicited the opinion of expert radiologists. METHODS: A questionnaire about the grading of Crohn's disease was sent to 36 radiologists who had published on MRI grading of Crohn's disease between January 2006 and January 2010. Radiologists were asked which MRI protocol they used, how they graded luminal Crohn's disease, which features they used, how important they considered those features for grading, and which reference standards they used. RESULTS: Twenty-four radiologists responded (66%). They used different protocols and features; most frequently T2-weighted sequences (79%) and contrast enhanced fat saturated T1-weighted sequences (83%). MR-enterography was more often used than MR-enteroclysis (88% versus 33%). Features most frequently considered important for grading were bowel wall thickness (79% of radiologists), the presence of an abscess (75%), T1 enhancement (75%), and T1 stratification (46%). Reference standards differed; most commonly (ileo-) colonoscopy (88%) or surgery (75%) were used. CONCLUSIONS: Bowel wall thickness, abscess, T1 enhancement and T1 stratification are most often used for grading. Because of difference in grading, there is need for an international consensus on MRI grading of Crohn's disease.

Optimization of alternating TR-SSFP for fat-suppression in abdominal images at 3T.

Magnetic resonance imaging is widely used in the work-up and monitoring of patients with Crohn's disease. Balanced steady-state free precession sequences are an important part of the imaging protocol and until now primarily 1.5T scanners have been used in daily clinical practice. This is largely because running balanced steady-state free precession sequences in 3T magnets has technical problems related to increased B(0) inhomogeneity and specific absorption rate (SAR) deposition. A modified form of alternating repetition time steady-state free precession sequence is presented to acquire 3D-isotropic abdominal images with fat-suppression at 3T within a breath-hold. The modifications include an adjusted radiofrequency pulse shape, suitable phase-cycling scheme and TR(1)/TR(2) ratio. Results show that the proposed sequence is successful in obtaining high contrast 3D-isotropic abdominal images within a breath-hold. Furthermore, the proposed methodology is easy to implement in a clinical setting and does not require any postprocessing steps.