Lésions infracliniques mammaires : corrélations radio-pathologiques sur les pièces d’exérèse chirurgicale. Intérêt pour le choix des techniques d’exérèse a minima

Abstract

Objective – To decrease the number of open excisional breast biopsies, percutaneous breast biopsies have been developed to perform the histologic diagnosis of non-palpable breast lesions. Some techniques allow a complete removal of small radiologic lesions. To evaluate the accuracy of a pathologic analysis performed only on the radiologic findings, we studied the radiologic and histologic correlation on open excisional biopsy (OEB) specimens. The aim of our study was to determine how many carcinomas were found distant from the radiologic findings. Material and Methods – Non-palpable breast lesions have been excised after preoperative localization. The radiologic findings were classified in microcalcifications, masses, architectural distorsion and in sonographic masses. The correlation between the radiologic and pathologic findings was studied and all the lesions incidentally discovered on histology with no correlation with the imaging findings or distant from the radiologic findings were described. One hundred and twenty-nine lesions have been evaluated in 99 patients. Radiologic findings included 79 clusters of microcalcifications, 30 masses, six architectural distortions and 14 sonographic lesions. Results – The histopathological results were benign in 41.9%, malignant in 49.6% and atypical hyperplasia lesions were discovered in 8.5%. All benign lesions were correlated to radiologic findings. No malignant lesion was found at a distance. In three cases atypical hyperplasia lesions were coexistent with the radiologic findings correlated with benign lesions and in three cases they were found at a distance. All but two malignant lesions were correlated with radiologic findings. In two cases a microinvasive carcinoma and foci of lobular carcinoma in situ were found coexistent with benign lesions. The subgroup of small lesions lesser than 10 mm included 31 benign lesions. A complete removal of these lesions should be obtained with percutaneous biopsies. Conclusion – A pathologic analysis targeted on radiologic findings is accurate. Numerous open excisional biopsies performed for benign pathology shoud be avoided.