Malignant versus benign vertebral compression fractures

Abstract

To determine whether it is possible to use bone single-photon emission tomography (SPECT) as a substitute for magnetic resonance (MR) imaging for differentiating malignant from benign vertebral compression fractures (VCFs). Ninety-seven fractures in patients who underwent both bone SPECT and MR imaging were included in our study. These 97 fractures were divided into two groups: malignant and benign VCFs. Two radiologists separately classified VCFs into malignant and benign according to the criteria for classifying VCFs on the planar, SPECT, and MR imaging as malignant or benign, as reported by many investigators earlier. In addition, 97 fractures were divided into two groups based on the MR images: VCFs with complete and partial replacement of the normal fatty marrow. The sensitivity, specificity, and accuracy of both the observers were calculated. The accuracy of the MR images was significantly greater than that of the SPECT images for both the observers (observer 1, P<0.001; observer 2, P<0.05). However, there were no significant differences in sensitivity, specificity, and accuracy for differentiating malignant from benign VCFs with complete replacement of the normal fatty marrow for both the observers [(observer 1: sensitivity, P=0.85; specificity, P=0.58; accuracy, P=0.63), and (observer 2: sensitivity, P=0.85; specificity, P=0.58; accuracy, P=0.87)]. Bone SPECT may be comparable with MR imaging for differentiating malignant from benign VCFs, especially, in the case of VCFs with a complete replacement of the normal fatty marrow.