Digital breast tomosynthesis provided mass characterization comparable to mammographic spot views in terms of visibility ratings, reader performance and BI-RADS assessment, according to a study published online Oct. 13 in Radiology. The findings could set the stage for a new paradigm in clinical breast imaging practice that substitutes tomosynthesis for spot views and may spare women from diagnostic mammography recalls.
To assess whether tomosynthesis performs comparably to mammographic spot views, researchers evaluated tomosynthesis and mammographic spot views of 67 masses (30 malignant, 37 benign) in 67 women.
Four academic radiologists individually reviewed the images in random order and rated them for visibility, likelihood of malignancy and BI-RADS classification.
Mean mass visibility ratings were slightly higher with tomosynthesis than with mammographic spot views. “All readers reported that, on average, masses appeared more obvious on digital breast tomosynthesis images, although only one reader achieved a statistically significant difference,” wrote Mitra Noroozian, MD, of the department of radiology at the University of Michigan Health System in Ann Arbor.
When the researchers compared histopathologic results and biopsy recommendations, they demonstrated that the four readers would have recommended biopsy for an additional seven cancers and five benign masses based on tomosynthesis results. This produced a mean increase of 1.8 true positives for every 1.3 false positives per reader.
“While this also did not achieve statistical significance, it supports the hypothesis that digital breast tomosynthesis performs comparably to mammographic spot views for characterizing a breast mass as benign or malignant,” wrote Noroozian.
The study also supported earlier subjective reports suggesting that tomosynthesis could provide an alternative to additional mammographic views for mass characterization. The results indicate that as tomosynthesis is adopted into clinical practice, spot views may be unnecessary.
“If future studies confirm that tomosynthesis can be integrated into screening protocols and that tomosynthesis can substitute for mammographic spot views to characterize masses, then women might be spared from being recalled for diagnostic views and from the associated incremental radiation exposure,” wrote Noroozian.
The researchers acknowledged several limitations, particularly the small size of the study, small numbers of masses and matched sets and limited number of readers. They noted that larger studies with more diverse mammographic findings and full sets of diagnostic mammographic images could help better define the role of tomosynthesis in practice.