The use of patients’ pre-existing full-field digital breast mammography (FFDM) and current digital breast tomosynthesis (DBT) images are largely independent contributing factors for reducing mammography recall rates in noncancer cases, according to results of a study published online March 21 in the journal Radiology.
In addition to prior FFMD images, which are effective at reducing recall rates when used in mammography interpretation, using DBT images has become an important tool for patients and doctors, with several recent studies showing their availability to be helpful in reducing recall rates while simultaneously bolstering cancer detection. “However, some studies did not include the availability of prior [FFDM] images during the interpretation of DBT images,” wrote lead author Christiane M. Hakim, MD, and her colleagues from the University of Pittsburgh Medical Center. “To our knowledge, none have specifically reported on the effect of the availability of prior FFDM images or the interaction between the availability of prior FFDM and DBT images on the resulting interpretation.”
Hakim and her team set out to examine this relationship between FFDM and DBT images and how each factors into recall decisions during mammogram interpretation. To do so, the researchers asked eight radiologists to independently interpret anonymous mammograms from 153 female patients with a mean age of 54 years old. Current and prior FFDM images and DBT images acquired in the researchers’ facility between June 2009 and January 2013 were used for the studies, which included 50 verified cancer cases, 60 negative and benign cases that were not recalled, and 43 benign cases meeting recall criteria. Two distinct sequential readings were used: one sequence in which current and prior FFDM images were used with and without DBT images, and one in which current mammography images and DBT were used with and without prior FFDM images.
Their results showed that despite the availability of both FFDM and DBT images for mammography studies, their ability to reduce recall rates are distinct and not cooperative. “Average recall rates in noncancer cases were significantly reduced with the addition of prior FFDM images by 34 percent (145 of 421) and 32 percent (106 of 333) without and with DBT images, respectively,” wrote Hakim et al. “In contrast, availability of DBT images resulted in a smaller reduction in recall rates (false-positive interpretations) of 19 percent (76 of 409) and 26 percent (71 of 276) without and with prior FFDM images, respectively.” While using FFDM images without and with DBT images resulted in lowered sensitivity of 7 percent (23 of 345) and 4 (14 of 353), using DBT without and with FFDM images increased sensitivity by 4 percent (15 of 338) and 8 percent (25 of 322).
“Our finding of no significant interaction between the two primary factors of prior FFDM and DBT images suggests that the effect of the two types of information is largely independent of each other,” the team wrote. “Our results suggest that in cases in which prior images are not readily available (eg, baseline studies, first time in a clinic without prior images for some reason, loss of prior images because of archiving errors) the use of DBT as a primary modality should yield accuracy that is comparable to, if not better than, that of current and prior FFDM images.”
Despite limitations including a small sample size and an enriched data set, Hakim and her colleagues believe their results will have positive practical benefits for clinical practices. “Understanding the primary variables that effect recall rates and the relative magnitude of their effect during interpretation of mammograms could improve our ability to optimize clinical practices,” the researchers concluded.