Elastography Ultrasound for Breast Tumors: Gland-to-Lesion Strain Ratio or Fat-to-Lesion Strain Ratio?

To evaluate the diagnostic efficiency of the gland-to-lesion strain ratio versus the fat-to-lesion strain ratio for breast tumor diagnosis.

The prospective study included 375 breast masses in 330 patients. B-mode ultrasound and elastography were performed for each mass. The gland-to-lesion strain ratio and fat-to-lesion strain ratio of the masses were calculated. The elasticity score of the lesions was also evaluated. The area under the curve, sensitivity, and specificity were calculated to compare the gland-to-lesion strain ratio with the fat-to-lesion strain ratio. Histopathological examination was considered the gold standard for final diagnosis.

Three hundred and seventy-five breast masses were included in our study (298 benign and 77 malignant). The gland-to-lesion strain ratio (GLR) and fat-to-lesion strain ratio (FLR) of the malignant lesions were significantly higher than those of benign lesions (P<0.0001 for both). The sensitivity and specificity of fat-to-lesion strain ratio were significantly better than gland-to-lesion strain ratio, (sensitivity, 96.1% versus 72% and specificity, 93.3% versus 81.1%). The area under the curve values for the fat-to-lesion strain ratio (0.990) and the elasticity score (0.972) were significantly higher than those for the gland-to-lesion strain ratio (0.820) (P < 0.0001). However, there was no significant difference between the area under the curve of the fat-to-lesion strain ratio and the area under the curve of the elasticity score (P=0.64).

The fat-to-lesion strain ratio provided better diagnostic performance than the gland-to-lesion strain ratio in breast mass characterization.