PURPOSE: To prospectively analyze the value of fast, dynamic, subtraction magnetic resonance (MR) imaging in the characterization of musculoskeletal tumors. MATERIALS AND METHODS: In 175 consecutive patients with a musculoskeletal mass, dynamic contrast-enhanced subtraction MR imaging was performed after administration of 0.1 mmol per kilogram of body weight gadopentetate dimeglumine or gadoteridol. A turbo gradient-echo technique was used, with a temporal resolution of 1-3 seconds. The interval between arterial and early tumor enhancement, the pattern (peripheral or diffuse) of enhancement, and the progression of tumor enhancement, as visualized on time-signal intensity curves, were assessed. MR enhancement features were related to the histopathologic diagnoses. RESULTS: Differentiation of benign from malignant soft-tissue masses was possible with a sensitivity of 91% and specificity of 72% based on start of enhancement, a sensitivity of 73% and specificity of 97% based on peripheral or diffuse enhancement, and a sensitivity of 86% and specificity of 81% based on progression of enhancement. Benign bone tumors could not be accurately differentiated from malignant bone tumors on the basis of the three defined parameters (sensitivity, 63%-76%; specificity, 50%-76%). CONCLUSION: Dynamic, contract-enhanced, subtraction MR images may be useful to differentiate malignant from benign soft-tissue masses.
PURPOSE: To prospectively analyze the value of fast, dynamic, subtraction magnetic resonance (MR) imaging in the characterization of musculoskeletal tumors. MATERIALS AND METHODS: In 175 consecutive patients with a musculoskeletal mass, dynamic contrast-enhanced subtraction MR imaging was performed after administration of 0.1 mmol per kilogram of body weight gadopentetate dimeglumine or gadoteridol. A turbo gradient-echo technique was used, with a temporal resolution of 1-3 seconds. The interval between arterial and early tumor enhancement, the pattern (peripheral or diffuse) of enhancement, and the progression of tumor enhancement, as visualized on time-signal intensity curves, were assessed. MR enhancement features were related to the histopathologic diagnoses. RESULTS: Differentiation of benign from malignant soft-tissue masses was possible with a sensitivity of 91% and specificity of 72% based on start of enhancement, a sensitivity of 73% and specificity of 97% based on peripheral or diffuse enhancement, and a sensitivity of 86% and specificity of 81% based on progression of enhancement. Benign bone tumors could not be accurately differentiated from malignant bone tumors on the basis of the three defined parameters (sensitivity, 63%-76%; specificity, 50%-76%). CONCLUSION: Dynamic, contract-enhanced, subtraction MR images may be useful to differentiate malignant from benign soft-tissue masses.