Developing Asymmetry at Mammography: Correlation with US and MR Imaging and Histopathologic Findings

To evaluate ultrasonographic (US) and magnetic resonance (MR) imaging findings, histopathologic etiologies, and outcomes for developing asymmetry at mammography. In this institutional review board-approved, informed consent-waived, HIPAA-compliant, retrospective review of a mammography database for...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Radiology 2016-05, Vol.279 (2), p.385-394
Hauptverfasser: Chesebro, Allyson L, Winkler, Nicole S, Birdwell, Robyn L, Giess, Catherine S
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:To evaluate ultrasonographic (US) and magnetic resonance (MR) imaging findings, histopathologic etiologies, and outcomes for developing asymmetry at mammography. In this institutional review board-approved, informed consent-waived, HIPAA-compliant, retrospective review of a mammography database for records from January 1, 2009 to December 31, 2012, 2354 consecutive diagnostic mammograms classified as showing focal asymmetry were identified. After patients with benign results, those considered stable, and those without prior mammograms were excluded, images from 521 studies were reviewed and 202 developing lesions were identified in 201 women. Patient demographics, US and MR imaging findings, and clinical and histopathologic outcomes were obtained from the electronic medical records. Equivocal US correlates of findings with developing asymmetry detected at mammography were excluded from statistical analysis. The Fisher exact test and Student t test analysis were performed and relative risk and 95% confidence intervals (CIs) were determined. Biopsy was performed in 73 (36%) of 201 patients with developing asymmetries, with 42 (58%) benign and 31 (42%) malignant results. Of 128 patients with nonbiopsied lesions, 110 (86%) were stable at 24 months (considered benign), 12 (9.4%) were stable at less than 24 months, and six (4.7%) were lost to follow-up. Diagnostic US was performed in 186 (93%) of 201 patients, 74 (40%) with correlates. US was performed in 30 (97%) of 31 patients with malignant developing asymmetries, 17 (57%) with correlates, and in 140 (92%) of 152 patients with benign lesions, 51 (36%) with correlates (risk ratio, 1.92; 95% CI: 1.001, 3.695; two-tailed P = .064, one-tailed P = .038). MR imaging was performed in 66 (33%) of 201 patients, 26 (39%) with correlates. MR imaging was performed in 10 (32%) of 31 patients with malignant developing asymmetries, all with correlates, and 53 (35%) of 152 patients with benign lesions, 15 (28%) with correlates (P < .0001). Developing asymmetries were malignant in 15% (95% CI: 11%, 21.1%) of patients. Presence of a US or MR imaging correlate was predictive of malignancy.
ISSN:0033-8419
1527-1315
DOI:10.1148/radiol.2015151131