Ultrasound detection of nonpalpable mammographically occult malignancy

William L. Simpson, George Hermann, Dana R. Rausch, Joshua Sherman, Stephen A. Feig, Ira J. Bleiweiss, Shabnam Jaffer, Abraham Port

Research output: Contribution to journalArticlepeer-review

2 Scopus citations

Abstract

Objective: To evaluate the prevalence of occult malignancy with screening breast ultrasound. Methods: All ultrasound-guided core needle breast biopsies performed between January 1, 1999, and June 30, 2001, were retrospectively reviewed. Lesions were identified during screening breast ultrasound in high-risk women with no mammographic or palpable abnormality in either breast, a unilateral mammographic or palpable abnormality in the contralateral breast, or a unilateral mammographic or palpable abnormality in a different quadrant of the same breast. All ultrasound-detected lesions were histologically verified. Results: Six hundred and fifty-two women with a mean age of 49 years underwent 698 biopsies during the study period. Three hundred and forty-nine of these lesions were detected at screening breast ultrasound. Out of 349, 11 (3.2%) had a mammographically and clinically occult malignancy. Nine cancers were found in women with no mammographic or palpable abnormality. Two cancers were found in the same breast as the mammographic or palpable abnormality. None were found in the breast contralateral to a palpable or mammographic abnormality. Conclusion: Screening breast ultrasound of high-risk women has a similar detection rate for occult carcinoma as screening mammography, but has a low positive predictive value in cases where biopsy is performed.

Original languageEnglish
Pages (from-to)70-76
Number of pages7
JournalCanadian Association of Radiologists Journal
Volume59
Issue number2
StatePublished - Apr 2008

Keywords

  • Breast cancer
  • Mammography
  • Screening
  • Ultrasound

Fingerprint

Dive into the research topics of 'Ultrasound detection of nonpalpable mammographically occult malignancy'. Together they form a unique fingerprint.

Cite this