Purpose To evaluate the diagnostic performance, costs, and impact on therapy of one-step magnetic resonance imaging (MRI) of the axial skeleton (MRIas) for detecting bone metastases in patients with high-risk prostate cancer (PCa). Patients and Methods Sixty-six consecutive patients with high-risk PCa prospectively underwent MRIas in addition to the standard sequential work-up (SW) of bone metastases (technetium-99m bone scintigraphy [BS] completed with targeted x-rays [TXR] in patients with equivocal BS findings and with MRI obtained on request [MRIor] in patients with inconclusive BS/TXR findings). Panel review of initial and 6-month follow-up MRI findings, BS/TXR, and all available baseline and follow-up clinical and biologic data were used as the best valuable comparator to define metastatic status. Diagnostic effectiveness of MRIas alone was compared with each step of the SW. Impact of MRIas screening on patient management and costs was evaluated. Results On the basis of the best valuable comparator, 41 patients (62%) had bone metastases. Sensitivities were 46% for BS alone, 63% for BS/TXR, 83% for BS/TXR/MRIor, and 100% for MRIas; the corresponding specificities were 32%, 64%, 100%, and 88%, respectively. MRIas was significantly more sensitive than any other approach (P < .05, McNemar). MRIas identified metastases in seven (30%) of 23 patients considered negative and eight (47%) of 17 patients considered equivocal by other strategies, which altered the initially planned therapy. Economic impact was variable among countries, depending on reimbursement rates. Conclusion MRIas is more sensitive than the current SW of radiographically identified bone metastases in high-risk PCa patients, which impacts the clinical management of a significant proportion of patients.
1 July 2003 in Journal of Computer Assisted Tomography
A patient with a history of pain, paresthesias, and weakness in both legs is reported. Magnetic resonance imaging (MRI) of the lumbar spine demonstrated subacute subdural hematoma. Brain MRI obtained 1 day later because of progressive headache showed hemorrhagic cortical metastasis and extensive subdural hematoma. It is hypothesized that the lumbar hematoma originated from the intracranial bleeding, which was substantiated by the observation of a thin hemorrhagic collection connecting cranial and lumbar hematomas on MRI. Cranial origin should be included in the differential diagnosis of spontaneous spinal subdural hematomas.
It is the radiologist's cornerstone to decide if imaging findings are normal or abnormal and to differentiate between clinically significant and insignificant findings. This challenge is extremely common in routine clinical practice when performing magnetic resonance (MR) imaging of the spine because it is frequently performed to assess patients with cancer or with spine-related symptoms. MR appearance of the normal vertebral marrow shows important variations not only with age but also among individuals of the same age range. On the contrary, marrow distribution and signal intensity patterns show little variation among each vertebral body of the same subject. Focal alterations in signal intensity can be observed that reflect local variation in the amount of normal expected vertebral components, including fat and hematopoietic cells, bone, and vessels. A more recently recognized condition related to the presence of notochordal cells deserves further study because it could account for some frequent tiny marrow changes. Diffuse alteration in vertebral signal intensity can also be observed and can be difficult—or even impossible—to differentiate from diffuse marrow infiltration by an abnormal process, given the lack of specificity of MR imaging. This article highlights the normal variants and frequent alterations of the vertebral bone marrow as encountered on MR studies of the spine and that can simulate significant lesions.