PURPOSE: to evaluate the agreement between histopathologic diagnoses of breast lesions made by general pathologists and by a specialist in breast pathology. METHODS: a cohort retrospective study comparing histopathologic diagnoses of 329 cases of breast lesions received in consultation for a second opinion was carried out. The material received for consultation included slides (152 cases), paraffin blocks (59 cases) or slides and blocks (118 cases). Cases were reviewed and the original diagnoses and diagnoses from a specialist in breast pathology were compared. The main diagnoses, nuclear grade of ductal carcinoma in situ, and the histopathologic grade of invasive mammary carcinomas were evaluated. The kappa index and percentual concordance were used in the statistical analyses. RESULTS: a moderate agreement was observed between the original histopathologic diagnoses and the second opinion (kappa index=0.48; percentual concordance=59.9%). The diagnosis of malignancy was confirmed in 185/225 cases (82.2%) and diagnosis of benign lesions was confirmed in 89/104 cases (85.6%). The highest agreement was observed in the diagnosis of invasive mammary carcinomas (81%) and the highest disagreement was observed among diagnoses of ductal carcinoma in situ with microinvasion (74%), lobular carcinoma in situ (70%), and atypical epithelial hyperplasias (61%). There was a moderate agreement in the nuclear grade of ductal carcinoma in situ (kappa index=0.52; percentual concordance=68.8%), and good concordance in the histologic grade of invasive carcinomas (kappa index=0.61; percentual concordance=74.3). CONCLUSIONS: the results show higher concordance rate in the diagnosis of invasive carcinomas and lower concordance in the diagnosis of ductal carcinoma in situ with microinvasion and premalignant breast lesions, especially lobular neoplasia in situ, and atypical epithelial hyperplasias.
Breast; Breast neoplasms; Carcinoma, ductal, breast; Referral and consultation
