Karen S. Johnson MD, Presenter: Nothing to Disclose

Mary Scott Soo MD, Abstract Co-Author: Nothing to Disclose

Sheila S. Lee MD, Abstract Co-Author: Nothing to Disclose

Jay Alan Baker MD, Abstract Co-Author: Consultant, Siemens AG

To review the imaging features and histological and clinical outcomes in a large series of suspicious, mammographically occult breast lesions detected at 3.0 T (tesla) MRI (magnetic resonance imaging).

Approval was obtained from the Institutional Review Board. A HIPPA-compliant retrospective review was performed of 121 consecutive suspicious, mammographically occult, lesions detected on contrast-enhanced breast MRI. MRI examinations were performed on 3.0 T scanners.  All 121 suspicious lesions underwent ultrasound guided core needle biopsy, MRI guided vacuum assisted core needle biopsy or MRI-guided needle localization.  Radiology and clinic reports were reviewed for patient age, menopausal status, indication for breast MRI, imaging findings including MRI lesion size and type, MRI-directed ultrasound findings, biopsy and localization details, histological results, and follow-up MRI and mammogram findings.  Positive predictive value (PPV) of biopsy recommendations were calculated and compared for screening versus diagnostic cases. Likelihood of malignancy was also compared with lesion size. Statistical analyses were performed using chi-square, Fisher’s exact, and two-tail z-tests.

Overall 43 malignancies were diagnosed from 121 suspicious, mammographically occult 3.0 T MRI-detected lesions. Seventy-eight (64%) of the 121 were benign. The overall positive predictive value of 3.0 T MRI-detected lesions was 36% (43/121). The positive predictive value for biopsy in the screening setting [22% (10/45)] was statistically significantly less than the PPV of a biopsy recommendation in the diagnostic setting [43% (33/76)]. There was no correlation between lesion size and the likelihood of malignancy; no significant difference in the frequency of malignancy as a function of lesion size was found (p=0.9176).

Our PPV of suspicious, mammographically occult, breast lesions detected at 3.0 T breast MRI is similar to the PPV reported previously for suspicious breast lesions detected at 1.5 T.  PPV of biopsy recommendation was greater for lesions detected at diagnostic breast MRI compared with screening-detected lesions.  This study supports the use of 3.0 T breast MRI for both screening and diagnostic breast imaging.

Our study supports the use of 3.0 T MRI for both screening and diagnostic breast imaging.

Johnson, K, Soo, M, Lee, S, Baker, J, Suspicious Breast Lesions Detected at 3.0 T Magnetic Resonance Imaging: Clinical and Histological Outcomes.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11000992.html