Comparison of the diagnostic value of mri and whole body 18F-FDG pet/ct in diagnosis of spondylodiscitis

Abstract

Spondylodiscitis is a spine infection for which a diagnosis by a magnetic resonance imaging (MRI) is considered the most appropriate imaging technique. The aim of this study was to compare the role of an18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) and an MRI in this field. For 56 patients with suspected spondylodiscitis for whom MRI and18F-FDG PET/CT were performed, we retrospectively analyzed the results. Cohen’s was applied to evaluate the agreement between the two techniques in all patients and in subgroups with a different number of spinal districts analyzed by the MRI. Sensitivity, specificity, and accuracy were also evaluated. The agreements of the18F-FDG PET/CT and MRI in the evaluation of the entire population, whole-spine MRI, and two-districts MRI were moderate (κ= 0.456, κ= 0.432, and κ= 0.429, respectively). In patients for whom one-district MRI was performed,18 F-FDG PET/CT and MRI were both positive and completely concordant (κ= 1). We also separately evaluated patients with suspected spondylodiscitis caused by Mycobacterium tuberculosis for whom the MRI and18 F-FDG PET/CT were always concordant excepting in 2 of the 18 (11%) patients. Sensitivity, specificity, and accuracy of the MRI and18 F-FDG PET/CT were 100%, 60%, 97%, and 92%, 100%, and 94%, respectively. Our results confirmed the18F-FDG PET/CT diagnostic value in the diagnosis of spondylodiscitis is comparable to that of MRI for the entire spine evaluation. This could be considered a complementary technique or a valid alternative to MRI.