Value of BI-RADS ultrasonic scores of direct and indirect ultrasonographic signs in diagnosis of solid breast lesions / 中华肿瘤杂志

ABSTRACT

<p><b>OBJECTIVE</b>To explore the value of BI-RADS ultrasonic scores of direct and indirect ultrasonographic signs in diagnosis of solid breast lesions.</p><p><b>METHODS</b>Reference to the standard BI-RADS score, ultrasonic scores of direct and indirect ultrasonographic signs of 132 solid breast lesions were assigned, and were compared with pathological results.</p><p><b>RESULTS</b>By the direct signs of breast lesions (aspect ratio, shape, border, internal echo, posterior echo, flow grade, sand-like calcification) and indirect signs (changes in local skin thickness, Cooper ligament changes, axillary lymph nodes, depth of reinforcement membrane changes, mass changes in the surrounding burr), the integral from the total scores in benign and malignant breast masses showed a statistically significant difference. The total score of malignant lesions (8.94 ± 2.85) was significantly higher than that of benign tumors (3.09 ± 1.97, P < 0.05). Except skin thickness, all the remaining scores of the signs of benign and malignant breast tumors showed a significant difference between the two groups (P < 0.05). By receiver operating curve (ROC) analysis, the best critical value of the total score of direct signs was ≥ 4, with a sensitivity and specificity of 0.84 and 0.93, respectively, in distinguishing breast carcinoma from benign lesions. The best critical value of the total score of indirect signs was ≥ 1, with a sensitivity and specificity of 0.82 and 0.74, respectively. The critical value of the combination of the direct and indirect signs was ≥ 5 in differential diagnosis of malignant and benign lesions, with a sensitivity and specificity value of 0.88 and 0.90, respectively.</p><p><b>CONCLUSION</b>The assignment score to the ultrasound characteristics of the direct and indirect signs of solid breast lesions can make a more objective diagnosis, yet it is a simple, effective, comprehensive and semi-quantitative analysis method.</p>