Stereoscopic mammo boosts breast cancer screening accuracy

Stereoscopic digital mammography significantly improved specificity for breast cancer detection and reduced the recall rate, according to a study published online Nov. 13 in Radiology.

Stereoscopic digital mammography (SDM) employs a medical stereo display comprised of two 5-megapixel, gray-scale monitors mounted one above the other with an angular separation of 110 degrees. The interpreting radiologist wears passive cross-polarized glasses. The configuration results in a single in-depth image of the breast and offers the potential to reduce false positive findings and screening recalls compared with conventional screening mammography.

Carl J. D’Orsi, MD, from the department of radiology and imaging sciences at Emory University School of Medicine in Atlanta, and colleagues designed a prospective clinical trial to compare SDM with standard digital mammography, focusing on patient recall and cancer detection in a screening population at elevated risk for breast cancer.

The study population consisted of 779 patients who underwent 1,298 exams from September 2004 to December 2007. Women underwent digital mammography, followed by SDM. Exams were read independently by two radiologists on the same day, without knowledge of each other’s results. Radiologists reported findings using BI-RADS categories and judged breast density according to the BI-RADS scale. Cancers reported within one year of the exam provided truth for the presence of malignancy. If no cancer was discovered, it was considered truth for the absence of malignancy.

D’Orsi and colleagues reported a recall rate of 12.9 percent with digital mammography, which was significantly higher than the SDM rate of 9.6 percent. SDM delivered significantly higher specificity at 91.2 percent, compared with 87.8 percent for digital mammography. Accuracy was 90.9 percent with SDM and 87.4 percent with digital mammography.

“This finding is consistent with two of the main predicted effects of stereoscopic DM: reduced susceptibility to tissue superposition and improved analysis of the volumetric structure of suspicious lesions and the surrounding tissue,” wrote D’Orsi and colleagues.

Although the results were promising for SDM, D’Orsi et al reported several limitations to the study. Because the study population was comprised of women at elevated risk for breast cancer, results may not be generalizable to the broad population. The fused stereo image contains lower noise than standard digital images, which could have caused some of the reduction in recall rates. In addition, each image in a stereo pair required a full x-ray dose. To be accepted as a routine screening exam, SDM would require that x-ray dose is halved.

D’Orsi and colleagues noted that SDM and tomographic breast imaging modalities—tomosynthesis and dedicated breast CT—could serve in complementary roles. Tomosynthesis hardware could provide a platform for acquisition of stereoscopic images and it could enable stereo projections through all or part of a stack of reconstructed tomographic data without any additional radiation dose.

The researchers recommended additional research to determine the efficacy of SDM at a dose equal to digital mammography in a non-high-risk population and have launched a second study addressing these factors.