The fifth edition of the American College of Radiology (ACR) Breast Imaging and Reporting Data System (BI-RADS) showed better positive predictive values (PPVs) for categorizing suspicious calcifications compared to BI-RADS 4, according to a new study published in the American Journal of Roentgenology.
“Calcification seen on mammography has variable probabilities of malignancy, and its interpretation can differ among readers,” wrote Ga Eun Park, of the department of radiology, Seoul St. Mary’s Hospital, The Catholic University of Korea in Seoul, and colleagues. “To our knowledge, no study has examined how changes in categorization of calcification from the 4th to the 5th edition of BI-RADS have affected positive predictive values (PPVs).”
Those changes took place in 2013 when the fifth edition of BI-RADS was published, classifying calcifications into two categories: “typically benign or having suspicious morphology.” By comparison, BI-RADS 4 divided calcifications into three categories: “typically benign, of intermediate concern and showing potential for malignancy.”
Additionally, BI-RADS 5 ruled that three calcification types (amorphous, coarse heterogeneous and fine pleomorphic) should be assessed as moderate suspicion of malignancy (4B), and fine liner or branching patterns should be assessed as high suspicion of malignancy (4C).
In the retrospective study, Park and colleagues enrolled 444 women ( mean age of 50.1 years) with pathologically confirmed suspicious calcifications for a total of 469 cases. Two radiologists categorized the calcifications using both BI-RADS 4 and BI-RADS 5, and analyzed PPVs for morphology, distribution and categorization of calcifications.
Overall, “the PPVs of categorization using the 5th edition matched better with BI-RADS category assessment than did categorization using the 4th edition,” the authors noted.
A central reason for this, according to Park et al., was tumor morphology.
“The 5th edition defines amorphous, coarse heterogeneous, and fine pleomorphic calcifications as category 4B; amorphous and fine pleomorphic calcifications were seen in the majority of cases included in this study,” they added. “Distribution, especially the combination of grouped or regional and segmental or linear distribution, also influenced PPV changes, which made the differences in PPVs of categories 4B, 4C, and 5 more distinct and consistent with category assessment by BI-RADS.”
The study included limitations, such as its exclusion of suspicious calcifications which were either not observed on biopsy or misread as probably benign.
Park and colleagues suggested more research is needed to formulate improved guidelines.
“Further research with a large number of cases, including specific cases related to shift in PPV, is required to establish proper guidance for categorization of suspicious calcification,” the researchers concluded.