SAN ANTONIO, TX—Radiologist’s time and x-ray exposure to the patient could be saved by using synthetic 2D mammograms rather than using 2D/3D combinations. Furthermore, review of the digital breast tomosynthesis (DBT) is necessary only if the synthetic 2D is reported M3, 4, or 5. This study was presented at the 2015 San Antonio Breast Cancer Symposium.1
A synthetic 2D mammogram (C-view) can be created by combining the individual optimally enhanced 1-mm slices of a DBT. Screening with 2D/3D combination has been shown to increase the cancer detection rate by approximately 40%, reduce the recall rate by 27%, and reduce false-positive rates by 15%; however, the method doubles the x-ray exposure to the patient and reading time for the radiologist compared with standard 2D screening. A synthetic 2D instead of a standard 2D film may theoretically overcome these problems by avoiding a second exposure and presenting the detail normally available in DBT in 1 picture. The researchers sought to determine if any advantage could be had in reviewing the DBTs if the synthetic 2D is normal/benign.
The researchers prospectively collected data on 2500 unselected cases presenting symptomatically or at follow up, all of whom underwent DBT on a Hologic Dimensions machine. The average age of the women was 58.4 years (range 28 to 95 years). Of these, some were under follow-up after mastectomies so a total of 4589 individual mammogram sets were reported.
Continue Reading
From the 3D data sets synthetic 2D mammograms were constructed (Hologic C-view). One breast radiologist with 13 years’ experience interpreting mammograms and 5 years’ interpreting DBTs was asked to review the 2D synthetic mammograms (CC and MLO) and report them before then reviewing the DBTs and issuing a final report. The mammograms were reported M1 to M5 using standard BIRADs criteria. BIRADs scores for each breast were recorded prospectively. Breast density, assessed by eye, was recorded (fatty/average density/dense).
Although the correlation is very close, the synthetic 2D was reported as normal or benign (M1 or M2) but the DBT was reported as M3 for 11 patients. Of these, 10 were benign and 1 malignant on assessment. For 1 patient, synthetic 2D was reported as M2 but M4 on DBT. Assessment confirmed malignancy. Sixteen cases reported as suspicious of malignancy (M3/4) by synthetic 2D were subsequently downgraded to benign after review of the DBT. We estimate that 2D mammography alone would have detected only 68 of the 94 detected by synthetic 2D.
Of the 4589 examinations, 1131 (25%) were assessed as fatty, 1851 (40%) as average density, and 1607 (35%) as dense. One cancer was missed in an averagely dense and 1 in a dense breast.
Synthetic 2D was found to reduce exposures by half from the 2D/3D combination, compression force is reduced, easier to compare with prior 2D images and to compare left and right breasts, and retains much of the information from the key 3D slices. The researchers acknowledge that a larger trial in the screening setting is required to confirm these findings; however, these results suggest that screening with 3D and reading the derived synthetic 2D films with selected review of the 3D images before recall could be cost effective way of significantly increasing the sensitivity and specificity of screening.
REFERENCE
1. Holt SD, Sharaiha YM, Moalla A, Williams HR, Thomas D, Huws AM. A comparison of the diagnostic performance of 2D synthetic mammography versus digital breast tomosynthesis in 2500 patients. Oral presentation at: San Antonio Breast Cancer Symposium; December 9-12, 2015; San Antonio, TX.
