{"title":"Breast MR Imaging Helps Differentiate Malignant and Benign Mammographic Microcalcifications: A Study Based on the 5th Edition of BI-RADS","authors":"L. R., W. L., W. D., Z. Y., Chen Y","doi":"10.47829/coo.2021.5301","DOIUrl":null,"url":null,"abstract":"Objective: To investigate whether breast MR imaging could help in differentiating malignant from benign mammographic microcalcifications. Methods: The study consecutively included 106 patients with 112 mammographic microcalcifications between January 2014 and April 2017 in our institute. Pre-operative mammograms and breast MR images were analyzed in a blind manner by two trained breast imaging subspecialists. Each lesion was described and categorized according to the 5th BI-RADS atlas. AUC, sensitivity, specificity, positive Likelihood Ratio (LR) were used to evaluate the value of MR imaging in differentiating malignancy from benignity. Results: Of the 112 lesions, pathologic results revealed 81 benign, 12 pre-cancerous and 19 malignant (10 invasive cancer and 9 ductal carcinoma in situ) findings. The number of lesions assigned to BIRADS 3, 4B, 4C, and 5 was 2, 92, 16, and 2, respectively, resulting in a PPV of 14.7% (17/108) for MG-BI-RADS 4 microcalcifications. The number of MRI classification 1 to 5 to the corresponding BI-RADS 4B microcalcifications were 37, 2, 33, 27, and 9 respectively. MR-BI-RADS criteria ruled out 72 benign MG-BI-RADS 4 lesions with none malignancy missed, while MRI enhancement criteria ruled out 46 benign lesions. MR-BI-RADS criteria were significantly better in AUC (0.896, P<0.0001), specificity (79.12%, P<0.0001), and positive LR (4.79) than mammography and MRI enhancement criteria. Conclusion: Breast MR imaging is useful in the evaluation of BI-RADS 4 mammographic microcalcifications by avoiding 79.12% unnecessary biopsies with none false-negative diagnosis.","PeriodicalId":92766,"journal":{"name":"Clinics of oncology","volume":"1 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Clinics of oncology","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.47829/coo.2021.5301","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
Objective: To investigate whether breast MR imaging could help in differentiating malignant from benign mammographic microcalcifications. Methods: The study consecutively included 106 patients with 112 mammographic microcalcifications between January 2014 and April 2017 in our institute. Pre-operative mammograms and breast MR images were analyzed in a blind manner by two trained breast imaging subspecialists. Each lesion was described and categorized according to the 5th BI-RADS atlas. AUC, sensitivity, specificity, positive Likelihood Ratio (LR) were used to evaluate the value of MR imaging in differentiating malignancy from benignity. Results: Of the 112 lesions, pathologic results revealed 81 benign, 12 pre-cancerous and 19 malignant (10 invasive cancer and 9 ductal carcinoma in situ) findings. The number of lesions assigned to BIRADS 3, 4B, 4C, and 5 was 2, 92, 16, and 2, respectively, resulting in a PPV of 14.7% (17/108) for MG-BI-RADS 4 microcalcifications. The number of MRI classification 1 to 5 to the corresponding BI-RADS 4B microcalcifications were 37, 2, 33, 27, and 9 respectively. MR-BI-RADS criteria ruled out 72 benign MG-BI-RADS 4 lesions with none malignancy missed, while MRI enhancement criteria ruled out 46 benign lesions. MR-BI-RADS criteria were significantly better in AUC (0.896, P<0.0001), specificity (79.12%, P<0.0001), and positive LR (4.79) than mammography and MRI enhancement criteria. Conclusion: Breast MR imaging is useful in the evaluation of BI-RADS 4 mammographic microcalcifications by avoiding 79.12% unnecessary biopsies with none false-negative diagnosis.