Does Timing of Imaging Affect Interpretation of Contrast-Enhanced Mammography?

A couple of minutes may make a significant difference in radiologist interpretation of contrast-enhanced mammography (CEM).

In a new study presented recently at the at the American Roentgen Ray Society (ARRS) 2023 Annual Meeting, researchers compared earlier-obtained imaging after contrast use (mean acquisition time of 2.5 minutes) and later image acquisition (mean acquisition time of 4.8 minutes) in 78 patients who had CEM. The study authors noted that five breast radiology attendings reviewed recombined images with the known cancer indicated and used a 5-point Likert scale to assess lesion visibility, margins, and conspicuity of the findings in comparison to background parenchymal enhancement (BPE).

The researchers noted intra-reader discrepancies in nine percent of the cases (35 out of 390 CEM interpretations) with 80 percent of those cases (28 of 35 cases) involving downgrading of lesion types to less conspicuous findings on later image acquisition. Nearly half of the downgrading instances (13 of 28 cases) reflected changes of mass identification on earlier-obtained imaging to non-mass enhancement (NME) on later image acquisition, according to the researchers.

The study authors also found that all of the reviewing breast radiologists indicated that earlier acquisition of imaging after contrast improved visibility of lesions, margins and conspicuity compared to BPE.

“Our study demonstrates that earlier-obtained recombined imaging is significantly preferred in cancer lesion characterization with a few instances demonstrating that biopsy-proven lesions may appear more conspicuously on earlier-obtained imaging,” noted study co-author Dennis Dwan, M.D., a clinical fellow of radiology affiliated with the Beth Israel Deaconess Medical Center in Boston.

(Editor’s note: For related content, see “Contrast-Enhanced Mammography Study Examines Impact of Clinical Factors on Background Parenchymal Enhancement,” “Study Highlights Use of Contrast-Enhanced Mammography in Women with Breast Implants” and “Key Considerations with Implementing Contrast-Enhanced Mammography in Your Practice.”)

The study authors also noted there was statistically significant difference in the assessment of lesion type when they compared craniocaudal views versus mediolateral views.

“This suggests that better conspicuity of cancer on early imaging is independent of projection,” added Dwan and colleagues.

In regard to study limitations, the researchers noted the modest sample size of 78 study participants and acknowledged that all cases were performed with units from one company (Senographe Essential Machine, GE Healthcare). Dwan and colleagues also noted that the reviewing breast radiology attendings in the study assessed annotated recombined images that would not reflect real-life presentations.