Comparison of cancer yields and diagnostic performance of screening mammography vs. Supplemental screening ultrasound in 4394 women with average risk for breast cancer

Purpose: The effectiveness of supplemental screening ultrasound (US) was investigated in women ≥40 years at average risk for breast cancer regardless of breast parenchymal density. A total of 4394 women at average risk and having previously undergone screening mammography were classified as the mammography group. Materials and Methods: Of 4394 women, 2005 underwent screening US after a final assessment of category 1 or 2 on screening mammography, and were categorized as the US group.Category 0, 4, and 5 on mammography and 3, 4, and 5 on US were defined as positive. The cancer yields per 1000 women and diagnostic performance of two groups were compared. Results: The total cancer and invasive cancer yields for the mammography group were 3.0 (95% confidence interval 1.6, 5.1) and 2.0 (95% CI, 0.9, 3.9) per 1000 women, higher than the US values of 2.0 (0.5, 5.1) and 1.0 (0.1, 3.6), not statistically significant. The specificity, accuracy, and positive predictive value (PPV) for mammography were 88.90% (87.93, 89.81), 88.85% (87.88, 89.76), and 2.61% (1.39, 4.41), significantly higher than the US values of 69.07% (66.99, 71.09), 69.13% (67.05, 71.15), and 0.64% (0.18, 1.64). The short-term follow-up rate of mammography was 5.51% (4.85, 6.22), significantly lower than the rate of 26.58 (24.66, 28.58) for US. Conclusion: Supplemental screening US in mammographically negative breasts can find additional carcinomas in women at average risk but is not as effective as screening mammography because of the lower cancer yield, invasive cancer yield, specificity, accuracy, PPV and a high short-term follow-up rate.