Abstract
BACKGROUND: Breast cancer is one of the commonest malignancy cancer worldwide and the Breast Imaging-Reporting and Data System (BI-RADS) classification has been extensively utilised as an adjunct to histopathological examination for malignant breast diseases. This study aims to analyse the concordance between radiological and histopathological findings, demonstrate the high predictive value in the BI-RADS category and evaluate the impact of these findings on surgical intervention and treatment outcome. METHODS: This is a single-centre retrospective study, analysing patients who underwent radiological examination with BI-RADS categories 3, 4 and 5 followed by histopathological examination confirming the diagnosis based on breast core biopsy or excision specimen over 3 years. RESULTS: A total of 316 specimens from 310 patients were included in this study; 75 cases were categorised BI-RADS 3, 166 as BI-RADS 4 and 75 as BI-RADS 5. Of these, 66 (20.8%) patients in BI-RADS category 3, 82 (25.9%) in BI-RADS category 4 and 5 (1.6%) in BI-RADS category 5 were reported as benign on histopathological examination. Malignant cases were reported in nine (2.8%) cases in BI-RADS category 3, 84 (26.6%) in BI-RADS category 4 and 70 (22.2%) in BI-RADS category 5. The positive predictive value (PPV), negative predictive value (NPV), sensitivity and specificity were 63.9%, 88%, 94.48%, and 43.14%, respectively. CONCLUSION: There is a significant correlation between BI-RADS score and histopathological results of breast cancer. A higher BI-RADS score is associated with a higher possibility of malignancy (P < 0.001). Our institution's performance is comparable to other previously published data.