B-mode ultrasound diagnostic flowchart for solid breast masses -JABTS BC-01 study

Abstract

Introduction The recall criteria for breast ultrasound screening developed by the Japan Association of Breast and Thyroid Sonology (JABTS) have been widely used in Japan since 2004. Furthermore, these recall criteria have also served as diagnostic criteria. However, there are problems. For example, although it is important to judge whether biopsy or follow-up is necessary at a diagnosis of breast masses, these criteria cannot be used to make this determination. This study aimed to develop breast ultrasound diagnostic flowchart for solid breast masses. Methods During the period from September 2009 to January 2010, we enrolled women with pathologically-confirmed breast masses and women with breast masses showing no significant change for more than 2 years. Since the number of patients who completed 2-year follow-up was small, we conducted an additional survey in January 2014. Static B-mode digital images, clinical and pathological data were collected. The B-mode images were evaluated by members of the centralized image interpretation committee without clinical information or pathological data. We carried out univariate and multivariate analyses. Based on the results, we developed new diagnostic flowchart. Results In total, 1045 solid breast masses (799 pathologically-confirmed breast masses and 246 2-year follow-up masses) were registered from 22 hospitals. Multivariate analysis showed that shape, margin, echogenic halo, interruption of the interface between fat and glandular tissues, and the Depth Width (DW) ratio were significant findings for distinguishing between benign and malignant lesions. The finding most strongly suggesting malignancy was echogenic halo (malignant: 98%), followed by interruption of the interface between fat and glandular tissues (84%).The finding most strongly suggesting benignity was the combination of oval shape, circumscribed margin, and DW ratio Conclusion Despite specificity being slightly inferior, the new diagnostic flowchart was superior to an expert in sensitivity.