[A multicenter study for the evaluation of the diagnostic efficiency of mammography and echography in nonpalpable breast neoplasms].

The Breast Section of the Italian Society of Radiology set up a cooperative study which included 17 Departments of Radiology and Breast Diagnosis in order to evaluate the diagnostic accuracy of US versus mammography in nonpalpable breast lesions. From January 1, 1989 to december 31, 1990, 400 nonpalpable breast lesions (142 benign lesions, 59 in situ and 199 infiltrating carcinomas) were detected by mammography and/or US; they had questionable/suspicious findings which called for further investigation by means of cytology and/or histology. US proved much less sensitive in non-palpable carcinomas than mammography (49.2% vs 93.8%), also in younger women, and failed to detect 50% of the benign/malignant nonpalpable lesions identified by mammography. US sensitivity was directly related to lesion diameter and probe frequency: 38.7% in infiltrating carcinomas < or = 5 mm vs 56.8% in those > 10 mm; 12% in the patients examined with a 5-MHz probe vs 57.7% in those examined with a > or = 7.5-MHz probe. Furthermore, the most significant US patterns of nonpalpable lesion were irregular contours, posterior attenuation and irregular internal echoes, while an irregular skin line and Cooper ligaments had no significant relation with carcinoma. Thus, breast US cannot be used as a screening test on asymptomatic patients not even on young women with radiologically dense breasts. On the contrary, US performed with high-frequency probes is useful in the assessment of nonpalpable lesions identified by mammography, and allows, in many cases, US-guided cytology and preoperative localization.