Accuracy of mammography in predicting histological extent of ductal carcinoma in situ (DOS)

Accuracy of mammography in predicting pathological extent of ductal carcinoma in situ (DCIS). Background and aims: Mammographic extent is the main determinant for offering wide local excision (WLE) for DCIS. It is recognized that this is not always accurate. Patients who prove to have larger lesions than predicted require further surgery. The aim of this study was to define the degree of variance between mammographic (MMG) and pathological (path) measurements of DCIS and to analyse the factors predicting a significant discrepancy. Methods: The pathological and mammographic data for 174 cases of DCIS were reviewed. Results: The mammographic size was bigger than the histological size in 97 (55.7%) and there was > 10 mm difference in 18 (10.3%) cases. The histological size was bigger than the mammographic size in 69 (39.7%) cases and > 10 mm difference was found in 30 (17.2%) cases. There was a significant relationship between larger MMG size, MMG size measured in two dimensions (MMG bi-dimensional product) and MMG-path size discrepancy (p < 0.01). In addition, the larger the size discrepancy, the greater the chance of requiring more than one therapeutic procedure (p < 0.01). There was no significant correlation between age, histological grade, mammographic density and shortest distance from nipple with degree of mammographic-pathological size discrepancy. (c) 2006 Elsevier Ltd. All rights reserved.