RESUMO
BACKGROUND: Differential therapy requires repeated diagnostic assessment for mapping and monitoring of disease activity in Crohn's disease (CD). PURPOSE: The purpose of this prospective study was to evaluate the accuracy of (18)F-fluorodexyglucose positron-emission tomography (FDG-PET) for non-invasive assessment of disease activity in CD. METHODS: Forty-three patients with CD underwent ileocolonoscopy and hydromagnetic resonance imaging (hydro-MRI) as reference standards. In addition, FDG-PET was performed and correlated with clinical data, hydro-MRI, and endoscopy findings. Diagnostic accuracy was determined for all methods. RESULTS: Two-hundred and forty-one bowel segments could be analyzed by all methods. Of 80 endoscopically inflamed segments in CD, FDG-PET detected 72 and hydro-MRI 53 segments. Overall sensitivity was 90 % (FDG-PET) versus 66 % (hydro-MRI), and specificity was 92.6 % versus 99 %. In the proximal ileum, hydro-MRI revealed inflammation in eight out of 49 patients and FDG-PET, also, detected all of these inflamed segments. Seventeen stenoses could be identified in 43 CD patients. With regard to assessment as inflammatory or fibrotic stenosis, there was good concordance between colonoscopy, hydro-MRI, and FDG-PET. In one case only, the nature of the stenosis was assessed differently. In contrast with leukocyte numbers and CDAI, there was significant correlation of FDG-PET activity with C-reactive protein and CDEIS levels (P = 0.019 and P = 0.007, respectively). CONCLUSION: FDG-PET is able to detect mucosal inflammation in CD with high sensitivity and specificity and to enable proper assessment of inflammatory activity in stenoses. FDG-PET is, thus, a promising non-invasive technique for clinical management of CD.
