Diagnostic potential of strain ratio measurement and a 5 point scoring method for detection of breast cancer: Chinese experience.

AIM: To evaluate the differential diagnostic potential of lesion stiffness assessed by the sonoelastographic strain index ratio (SR) and elastographic color scoring system (UE) for breast lesions. MATERIALS AND METHODS: Three hundred and forty two breast masses (158 benign and 184 malignant) from 325 consecutive patients (mean age 44.2 years; range 16-81)who had been scheduled for a sonographically guided core biopsy were examined with grey scale sonography,doppler sonography and sonoelastography. Each lesion was classified with ACR's BI-RADS assessment category (2, 3 and 4A=Benign and; 4B, 4C, 5=Malignant) and the 5-point scoring system proposed by Itoh et al, with scoring 1-3=benign and 4-5=malignant. Strain and area ratios of each lesion were calculated within the same machine. Histological diagnosis was used as the reference standard. The area under the curve (AUC) and cut-off point were obtained by receiver operating curve and the cross table Fischer Test was carried out for assessing diagnostic value. Sensitivity, specificity, PPV, NPV, accuracy and false-discovery rates were compared. RESULTS: The mean strain ratios for benign and malignant lesions were 1.87 and 7.9 respectively. (P<0.0001). When a cutoff point of 3.54 was used, SR had a sensitivity of 94.6%, a specificity 94.3%, a PPV of 95.1%, an NPV of 93.7% and an accuracy of 94.4%. The AUC values were 0.90 for the 5 point scoring system (UE) and 0.96 for the strain index ratio. The overall diagnostic performance was SR method was better (P<0.05). CONCLUSIONS: Strain ratio measurement could be another effective predictor in elastography imaging besides 5 the point scoring system for differential diagnosis of breast lesions.

Breast elastography: a hospital-based preliminary study in China.

The performance of ultrasound elastography in the characterization of 170 breast lesions with histologic correlation was evaluated in 150 consecutive patients who were diagnosed as having benign or malignant masses on B mode sonography. Each lesion was classified with ACR's BI-RADS lexicon (benign=2 and 3, malignant 4 and 5) using Ueno elastography classification scores (benign=1-3, malignant=4-5). Of the 170 lesions, 70 were histologically malignant, and 100 were benign. Ultrasound elastography was superior in detecting breast cancer, since the accuracy (95.8%), sensitivity (98.6%), specificity (96.0%), and positive predictive values (94.5%) were higher than those of B mode sonography (90.6%, 91.4%, 90.0% and 86.5% respectively). The sensitivity (98.57%) and false discovery rate (1.1%), when both modalities were jointly used (sonography and /UE) was better then those of sonography and UE singly. However, the specificity (90%) was found similar to sonography. In conclusion, ultrasound elastography is superior to B mode sonography in assessing the nature of breast lesions.