Diagnostic Accuracy of Breast Imaging Reporting and Data System (BI-RADS) in Determination of Breast Malignancy Taking Histopathology as Gold Standard

Abstract

Objective: To determine the yield, reliability and diagnostic accuracy of Breast Imaging Reporting and Data System in evaluation of breast lesions taking histopathology as gold standard.

Design: This cross-sectional, analytical study was conducted at the Pathology and Radiology departments at Dow University of Health Sciences, Karachi from January 2013 to December 2015.

Patients and methods: The data of the core needle biopsies of breast lesions received at the Pathology department was reviewed. The cases which had been categorized on mammogram, according to the Breast Imaging-Reporting and Data System (BI-RADS), were selected for the study. According to BI-RADS, categories II and III were classified as benign and BI-RADs IV and V as malignant. The breast core biopsies were classified as benign and malignant, according to the diagnosis.

Results: A total of 359 patients were included in the study. Females were predominant [n=355 (98.9%)] as compared to males [n=4 (1.1%)] with age 45.80 ±13.39 years [mean±SD]. There were 191 (53.2%) patients with left sidedbreast involvement, while 168 (46.8%) patients had right sided involvement. BI-RADS system for reporting when compared with histopathology had sensitivity of 92.38%, specificity 88.89%, positive predictive value 99.37%, negative predictive value 38.10%and diagnostic accuracy92.20%.

Conclusion:

The findings of this study report a high diagnostic accuracy of BI-RADS in the diagnosis of breast carcinoma; however, based on low negative predictive value we recommend training of radiologists reporting the mammogram, regular reviewing of discordant cases by histopathologist and radiologist, and early follow up in all such patients