Subcategorization of Ultrasonographic BI-RADS Category 4: Assessment of Diagnostic Accuracy in Diagnosing Breast Lesions and Influence of Clinical Factors on Positive Predictive Value

Abstract

To evaluate the diagnostic accuracy of subcategories 4a–4c of the second edition of the Breast Imaging Reporting and Data System (BI-RADS) ultrasonography (US) lexicon, and to investigate whether clinical factors influence the positive predictive values (PPVs). Overall, 1240 breast lesions in 1227 women diagnosed on ultrasound as category 4 and with pathology were included. The PPV with 95% confidence interval (CI) was 13.6% (95% CI: 11%, 16%) in BI-RADS 4a, 50.0% (95% CI: 44%, 56%) in BI-RADS 4b and 86.0% (95% CI: 82%, 90%) in BI-RADS 4c. Patients’ age significantly affected PPVs for subcategories 4a–4c, whereas radiologists’ experience and application time had little influence on PPVs for subcategories 4a–4c. In conclusion, the diagnostic accuracies of subcategories 4b and 4c were superior to subcategory 4a. Patients’ age significantly affected PPVs for subcategories 4a–4c. Utilizing the subcategories of category 4 was a feasible method regardless of radiologists’ experience and application time.