Ultrasound Assessment of Invasive Breast Cancer: Correlation with Histologic Grade

ABSTRACT

PURPOSE: To correlate the final assessment of ultrasound and histologic grade in patients with invasive breast cancer. MATERIALS AND METHODS: The study consisted of one hundred eighty-six women with breast masses that were evaluated by ultrasound and subsequently found to have invasive ductal carcinoma not previously diagnosed. Two radiologists prospectively analyzed the results of the ultrasounds and issued an American College of Radiology Breast Imaging Reporting and Data System (BI-RADS) assessment category which indicates the probability of malignancy. The histologic grade of the 186 invasive ductal carcinomas were separated as follows, low grade in 22 lesions (9%), intermediate grade in 91 lesions (50%), and high grade in 73 lesions (41%). RESULTS: In the 186 invasive ductal carcinomas not previously diagnosed, a circumscribed margin was seen in 4% (three of 73) of high-grade tumors compared to none in intermediate-grade and low-grade tumors. A spiculated margin was seen in 59% (13 of 22) of low-grade tumors compared, to 20% (18 of 91) in intermediate-grade and 7% (five of 73) in high-grade tumors. Posterior acoustic enhancement was seen in 33% (24 of 73) of high-grade tumors compared, to 20% (18 of 91) in intermediate-grade and none in low-grade tumors. Posterior acoustic shadowing was seen in 59% (13 of 22) of low-grade tumors compared, to 34% (31 of 91) in intermediate-grade and 15% (11 of 73) in high-grade tumors. The final BI-RADS assessment of the 186 cases separated as follows, three lesions (2%) were category 3 probably benign, 115 lesions (62%) were category 4 suspicious, and 68 lesions (37%) were category 5 highly suggestive of malignancy. The three cases misclassified as probably benign were high-grade tumors. CONCLUSION: Breast cancer showing a spiculated margin and posterior acoustic shadowing on ultrasound were mainly low-grade tumors whereas breast cancer showing benign features were high-grade tumors.