PURPOSE: To evaluate the MRI findings helpful in differentiating between osteoporosis and metastatic vertebral compression fracture. MATERIALS AND METHODS: This study involved 52 patients with nontraumaticvertebral compression fracture; 23 had osteoporosis and 29, metastatic diseases. We retrospectively analyzed signal abnormality and the extent to which bone marrow was seen within the involved vertebral body, sharpness of margin of abnormal signal intensity, morphological characteristics of the vertebral endplate, and the presence or absence of posterior element involvement and paraspinal mass, as seen on T1- and T2*-weighted MR images. For statistical analysis, the chi-square test was used. RESULTS: In 14 of 23 patients (61%) with osteoporotic benign compression, and 27 of 29 (93%) with metastatic compression fracture, the bone marrow of the compressed vertebral body showed both low signal intensity on T1-weighted image and high signal intensity on T2-weighted image. In 13 of 14 osteoporotic compression fractures, the extent of abnornal bone marrow signal was incomplete, but in 20 of 27 metastatic compression fractures, this was complete. In all 13 cases of osteoporosis, incomplete abnormal signal showed a sharp margin, whereas in five of seven metastatic compression fractures, this margin was ill-defined. Morphologically, the endplate was concave in 20 of 23 osteoporosis cases (87%), but was angled in 13 of 29 metastatic compression fractures (45%). Only the latter showed vertebral posterior element involvement (21/29) and paraspinal mass (16/29). CONCLUSION: The extent of signal abnormality, margin of compressed bone marrow, morphologic characteristics of the endplate, and the presence or absence of posterior element involvement and paraspinal mass on T1- and T2*-weighted MR images, as described above, may be helpful in differentiating between benign osteoporotic and malignant metastatic compression fractures.