Factors influencing the agreement on histopathological assessments of breast biopsies among pathologists

Abstract

It has been recommended that the histopathology results of core biopsies of the breast are categorized according to the B-categorization scheme. We measured the interobserver variability of the B-categorization of core biopsies of the breast. Core biopsies were taken among 765 women at the University of Halle between 2006 and 2008. All histological slides were reviewed in a blinded fashion by two experienced breast pathologists. We calculated observed and chance-corrected agreements (kappa) and 95% confidence intervals (CI). The prevalence of B3-B5 biopsies was 41.6%. The observed and weighted kappa agreement of the five-level B-categorization scheme was 0.87 (95% CI: 0.84 -0.89) and 0.89 (95% CI: 0.89-0.91), respectively. The most frequent disagreement was between B2 and B3 (47 of 103 disagreements, 45.6%). Overall, 49.5% of all disagreements were clinically relevant disagreements that would imply different therapeutic strategies. Agreement was modified by referral group, Breast Imaging Reporting and Data System (BIRADS) level, radiological breast density, imaging guidance and application of immunohistological staining. Interobserver agreement of the B-categorization scheme was high and was modified by referral status, level of radiological suspicion of breast cancer, breast density, imaging guidance of core biopsies and requirement of additional immunohistological staining.