Abstract Archives of the RSNA, 2004
SSK01-01
The Value of Adding Quantitative 1H MRS for Improving Diagnostic Accuracy of Breast MRI: Results from an Observer Performance Study at High Field
Scientific Papers
Presented on December 1, 2004
Presented as part of SSK01: Breast (MR Imaging)
Sina Meisamy MD, Presenter: Nothing to Disclose
Patrick J. Bolan MS, Abstract Co-Author: Nothing to Disclose
Michael Thomas Nelson MD, Abstract Co-Author: Nothing to Disclose
Lenore I. Everson MD, Abstract Co-Author: Nothing to Disclose
Frederick Kelcz MD, PhD, Abstract Co-Author: Nothing to Disclose
Michael G. Garwood PhD, Abstract Co-Author: Nothing to Disclose
Tim Harrison Emory MD, Abstract Co-Author: Nothing to Disclose
Kathy R. Brandt MD, Abstract Co-Author: Nothing to Disclose
Kimberly Katz Amrami MD, Abstract Co-Author: Nothing to Disclose
Mary Carol L. Lechner MD, Abstract Co-Author: Nothing to Disclose
et al, Abstract Co-Author: Nothing to Disclose
To determine whether the addition of in vivo quantitative proton MR spectroscopy (1H MRS) can improve the radiologist’s accuracy in interpreting breast MRI.
Fifty-five breast MR cases with pathologic findings confirmed by biopsy were retrospectively evaluated by 4 radiologists. Patients were examined with contrast-enhanced, fat-suppressed, T1-weighted MRI using a 4T MRI/MRS scanner. The concentration of choline-containing compounds [tCho] was quantified using single-voxel 1H MRS. For each case, the radiologists were asked to state the probability of malignancy and to provide a recommendation on patient management. Two interpretations were performed for each case. The initial interpretation was based on lesion morphology and time intensity curve; the second interpretation was based on lesion morphology, time intensity curve, and [tCho].
All lesions evaluated in this study enhanced with Gd-DTPA. Of the 55 cases evaluated in this study, 35 were invasive carcinoma and 20 were benign. The addition of 1H MRS resulted in higher sensitivity, specificity, and accuracy (AZ) for all 4 radiologists. More specifically, two of the 4 radiologists showed a significant improvement in sensitivity (p=0.03, p=0.03), and all 4 radiologists showed a significant improvement in AZ (p=0.01, p=0.05, p=0.009, p=0.0008).
These results suggest that the addition of quantitative 1H MRS to breast MRI may improve the radiologist’s ability to distinguish benign from malignant breast lesions. The additional information provided by 1H MRS is particularly valuable in cases where the lesion morphology or time-intensity curve is indeterminate.
Meisamy, S,
Bolan, P,
Nelson, M,
Everson, L,
Kelcz, F,
Garwood, M,
Emory, T,
Brandt, K,
Amrami, K,
Lechner, M,
et al, ,
The Value of Adding Quantitative
1H MRS for Improving Diagnostic Accuracy of Breast MRI: Results from an Observer Performance Study at High Field. Radiological Society of North America 2004 Scientific Assembly and Annual Meeting, November 28 - December 3, 2004 ,Chicago IL.
http://archive.rsna.org/2004/4404908.html