Benign versus pathologic compression fractures of vertebral bodies: assessment with conventional spin-echo, chemical-shift, and STIR MR imaging.

Differentiation of benign from pathologic compression fractures of vertebral bodies was evaluated with magnetic resonance imaging in a prospective study of 53 patients. Twenty-six patients had 34 benign posttraumatic compression fractures. Twenty-seven patients had metastatic disease to the vertebral column and seven pathologic fractures. T1- and T2-weighted spin-echo (SE) sequences (1.5 T) were performed in all patients. A presaturation technique was used to obtain "fat" and "water" images to better assess the degree of normal fatty marrow replacement in fractured vertebrae. Short inversion-time inversion-recovery (STIR) images were also obtained. Discrimination between benign and pathologic compression fractures was generally possible with the SE sequences. Chronic benign fractures demonstrated isointense marrow signal intensity (SI), compared with that of normal vertebrae with all sequences. Pathologic fractures showed low SI on T1-weighted images and high SI on T2-weighted images. Fat images revealed complete replacement of normal fatty marrow, shown as absent SI in the involved vertebral body. Water and STIR images showed diffuse high SI in pathologic fractures, with STIR images having the highest contrast between abnormal and normal marrow. Acute benign compression fractures also demonstrated high SI on T2-weighted, water, and STIR images, but the SI was less pronounced and the pattern was generally more inhomogeneous than that of pathologic compressions. In general, fat images showed only partial replacement of normal fatty marrow by low SI, in contrast to the complete absence of marrow SI typical of pathologic fractures.