Correlation Engine 2.0
Clear Search sequence regions


Sizes of these terms reflect their relevance to your search.

To elucidate whether breast imaging can predict final histologic diagnosis of lesions of uncertain malignant potential diagnosed at ultrasound core needle biopsy (CNB). The imaging characteristics (mammography, ultrasound, and magnetic resonance imaging [MRI]) of lesions of uncertain malignant potential in the breast that were obtained by ultrasound CNB were retrospectively analyzed in 87 women. Radiologic characteristics of lesions were compared to definitive histopathologic findings. Out of 87 breast lesions of uncertain malignant potential, 27 (31%) were diagnosed as papillary lesions, 24 (28%) atypical ductal hyperplasia, 19 (22%) lobular intraepithelial neoplasia, 9 (10%) phyllodes tumors, 3 (3%) radial sclerosing lesions, and 5 (6%) unspecified lesions of uncertain malignant potential. The underestimation rate of malignancy at CNB based on the total number of lesions on final follow-up was 22%. Using multivariate logistic regression, Breast Imaging-Reporting and Data System (BI-RADS) score (odds ratio [OR] = 12.29, p = 0.027) and Göttingen MRI scoring system (OR = 8.1, p = 0.008) were found to be independent predictors of malignancy. Receiver operating characteristic analysis showed that Göttingen MRI score >3 provides a plausibly good cutoff value with sensitivity of 100 (95% confidence interval [CI] 74%-100%) and specificity of 76% (95% CI 61%-88%). Lesions of uncertain malignant potential classified as BI-RADS 5 and Göttingen score 4 or higher are at significantly higher risk of harboring malignancy and therefore should be recommended for surgical excision.

Citation

Marko Petrovecki, Ana Marija Alduk, Jasminka Jakic-Razumovic, Ranka Stern-Padovan, Vinko Bubic, Maja Prutki. Role of breast imaging in predicting outcome of lesions of uncertain malignant potential (B3) diagnosed at core needle biopsy. Tumori. 2016 Mar-Apr;102(2):203-8

Expand section icon Mesh Tags


PMID: 26350202

View Full Text