In Reply: As Dr Fujita points out, our study showed that supplemental ultrasound screening increased the cancer detection rate. To achieve this, recall rates increased from 10.2% to 20.3% on average (16.8% with incidence screenings). While a benchmark mammography recall rate of 10% has been suggested, acceptable benchmarks for combined screening with other modalities have not been established.
We agree with Fujita that a better management strategy for sonographically detected BI-RADS 3 lesions, as well as BI-RADS 4a, is needed. In year 1, of 2637 mammogram results, 177 (6.7%) were BI-RADS 3 (including 1 cancer), as were 401 (15.2%) mammography plus ultrasound results (4 malignancies).1 In additional analyses from this study, we found that, in part due to the elevated risk of our participants, nearly 17% of sonographically BI-RADS 3 lesions underwent biopsy and 0.8% proved to be malignant. For sonographically BI-RADS 4a lesions, the biopsy rate was 82.5% with 2.8% proving to be malignant. BI-RADS 3 and BI-RADS 4a ultrasound lesions together accounted for 71.1% of biopsies in our study. Multiple similar benign-appearing masses that should be classified as BI-RADS 2, benign, were frequent among BI-RADS 3 lesions. Adding elastography, a measure of stiffness, can reduce unnecessary biopsy of sonographically BI-RADS 4a masses.2 With any screening strategy, continued monitoring of outcomes and physician education can help reduce unnecessary recall and biopsy.