Posted: Thursday, February 3, 2022
Patients’ likelihood of being recalled appears to increase when their digital breast tomosynthesis images were interpreted later in the day by less-experienced radiologists, according to the findings of a retrospective study published in the journal Radiology. Ana P. Lourenco, MD, of Alpert Medical School of Brown University in Providence, Rhode Island, and colleagues propose the likelihood of radiologists reporting a false-positive result increases every hour when interpreting digital breast tomosynthesis images.
“The standard 2D mammogram is four views,” stated Dr. Lourenco in a press release from the Radiological Society of North America. “With digital breast tomosynthesis, we are looking at 1-mm reconstructions of the breast, which means hundreds of views for each exam. It’s useful to know that fatigue is a factor that impacts performance metrics.”
This study screened mammograms from January 2018 to December 2019 at 1 of 12 community radiology sites. The study analyzed the association between the time of day of the mammogram interpretation, the type of image interpreted (digital mammography or digital breast tomosynthesis), and the radiologist’s experience (less than and more than 5 post-training years). Additionally, the study analyzed the likelihood of a patient being recalled after mammography and whether the interpretation was a false-positive or true-positive result.
Of the 97,671 mammograms reviewed by 18 radiologists, 40,220 were from digital breast tomosynthesis and 57,451 were from digital mammography. Of the 18 radiologists, 9 had up to 5 post-training years of experience and 9 had more than 5 post-training years of experience.
The overall recall rate of digital mammography was significantly higher at 10.2% compared with digital breast tomosynthesis at 9% (P = .006). Odds of recall increased 11.5% with every hour when less-experienced radiologists used digital breast mammography. These results were not found for digital mammography. The false-positive rate was significantly different for digital mammography and digital breast tomosynthesis (9.8% and 8.6%, P = .004).
Disclosure: For full disclosures of the study authors, visit pubs.rsna.org.