298
Views
0
CrossRef citations to date
0
Altmetric
Breast

Ultrasonography-Guided Core Needle Biopsy in Differential Diagnosis of Papillary Breast Tumors

, &
Pages 722-729 | Published online: 01 Sep 2009
 

Abstract

Background: Papillary lesions of the breast are considered diagnostically challenging for various reasons. A relatively high malignancy rate in final pathological analysis has in many cases necessitated excision of these lesions, regardless of core needle biopsy (CNB).

Purpose: To assess mammographic, sonographic, and CNB findings of papillary lesions, and to correlate them with final histology obtained by surgical excision.

Material and Methods: From 2000–2006, 29 benign and 19 malignant papillary tumors examined with CNB were surgically removed. Mammographic, sonographic, and CNB results were analyzed and correlated with final histology.

Results: On ultrasonography (US), 69% (20/29) of the benign lesions were solid and 31% (9/29) were cystic, and 47% of the malignant lesions (9/19) were solid and 53% (10/19) were cystic. The mammographic findings were nonspecific, although most of the malignant tumors (67%, 12/18) were categorized as BI-RADS 4 lesions. The sensitivity, specificity, and positive predictive value of the core needle biopsy histology was 32% (6/19), 100% (29/29), and 100% (6/6), respectively, for papillary carcinoma. A negative predictive value of 91% (21/23) for malignancy and 48% (11/23) for either atypia or malignancy was shown. Of the lesions with CNB diagnosis of benign papillary lesion with atypia, 58% (11/19) turned out to be malignant and 11% (2/19) were benign on surgery.

Conclusion: The probability of malignancy is low when the CNB result shows a benign papillary lesion with no atypia. However, the only way to reliably diagnose atypical papillary lesions is to surgically remove all papillary tumors, irrespective of the CNB result. Differentiation between benign and malignant lesions or malignant noninvasive and invasive tumors is not possible based on sonographic or mammographic appearance.

Log in via your institution

Log in to Taylor & Francis Online

There are no offers available at the current time.

Related Research

People also read lists articles that other readers of this article have read.

Recommended articles lists articles that we recommend and is powered by our AI driven recommendation engine.

Cited by lists all citing articles based on Crossref citations.
Articles with the Crossref icon will open in a new tab.