[The diagnostic role of breast echography]

The authors reviewed a consecutive series of 4,160 breast lesions (511 carcinomas and 3,649 benign lesions) studied with 10-MHz US in 1990-1991. US sensitivity was 72.6%, specificity was 97.3% and positive predictive value was 78.9%. Sensitivity was independent of age whereas it strongly depended on tumor size (pT1a = 54.2%, pT1b = 70.6%, pT1c = 80.6%, pT2-4 = 90.2%) (81.1% palpable masses and 41.3% nonpalpable masses) and histologic type (11.4% intraductal carcinomas and 69.4% special types). The association was confirmed of the classic US signs with carcinoma (irregular margins and posterior shadowing) and benign lesions (lateral shadowing, no echogenicity and posterior acoustic enhancement) patterns, even though the former are not accurate enough for diagnostic purposes. US demonstrated 376 of 402 palpable carcinomas (50 diagnosed as benign and 326 as suspicious) and 58 of 109 nonpalpable carcinomas (13 diagnosed as benign and 45 as suspicious). US was the only imaging method to yield findings suggestive of cancer in 13 of 511 total cases and contributed to final diagnosis in 6 more clinically and radiologically negative cases in which US findings suggested a benign lesion but indicated the need for aspiration cytology. If on the one hand US was the only method responsible for unnecessary biopsy in 21 of 141 surgical benign lesions, on the other hand negative or benign US findings helped avoid unnecessary biopsy in 277 cases diagnosed as suspicious at palpation or mammography. US should be used as the routine examination in the assessment of clinical and mammographic abnormalities. Its contribution to cancer detection is limited but it allows the number of unnecessary biopsies to be markedly reduced as well as a more accurate guide for fine needle aspiration in most palpable cancers and in approximately half of nonpalpable ones.