Differentiation of benign and malignant breast lesions: MR imaging versus Tc-99m sestamibi scintimammography.

Abstract

To compare the accuracies of magnetic resonance (MR) imaging and scintimammography in differentiating benign from malignant breast lesions. MR imaging was performed in 66 women with 75 lesions during intravenous administration of gadopentetate dimeglumine. Planar and single photon emission computed tomographic (SPECT) scintimammography were performed (with 740 MBq technetium-99m sestamibi administered intravenously) in all 66 patients with 75 lesions and in 64 patients with 73 lesions, respectively. MR imaging and scintimammographic studies were independently evaluated by using signal intensity measurements versus time or focal tracer uptake to differentiate benign from malignant lesions. Histopathologic proof was obtained in 63 lesions. Twelve lesions were monitored with follow-up. MR imaging was false-negative in one and false-positive in nine lesions. Planar scintimammography was false-negative in 10 and false-positive in six lesions. SPECT scintimammography was false-negative in four and false-positive in 10 lesions. Sensitivities and specificities for malignancy were, respectively, 96% and 82% for MR imaging, 62% and 88% for planar scintimammography, and 83% and 80% for SPECT scintimammography. Both MR imaging and scintimammography are useful in the evaluation of breast cancer. MR imaging is more sensitive and as specific as scintimammography.