The value of ultrasonic examination to detect and diagnose breast carcinomas. Analysis of the results obtained in 125 tumors using radiographic and ultrasound mammography.

A series of 125 breast carcinomas studied by both radiological and ultrasound mammography, and which were observed in 102 patients, is presented. Of the lesions, 76.5% were less than 20 mm in diameter, and 16.8% had a diameter 10 mm or less. Sixteen lesions were in situ or microinvasive. Analysis of the results shows a 40.8% rate of false-negative results with radiological mammography, when only 'highly suspicious' and 'suspicious' images are considered. This figure is reduced to 28% when 'benign-appearing' masses are included in the computation. Using similar diagnostic categories, the false-negative rates obtained with ultrasound mammography were 14.4% and 12.8%, respectively. Combining radiological and ultrasound mammography resulted in a dramatic decrease in the false-negative rate to 4% The number of false-negative results observed with either radiological or ultrasound mammography was in the higher range of reported values when compared to selected studies. However, they may, in fact, closely reflect the real performance of these investigations, as they are available to patients in everyday clinical practice. Analysis of the data shows that radiological mammography is less efficient in younger patients. The mean age of the patients in the radiological mammography positive group was 54.3 +/- 1.2 years, as opposed to 47.5 +/- 2.1 (p = 0.004) in the radiological mammography negative group. In contrast, the efficiency of ultrasound mammography is independent of the patient's age, and consequently of breast density. In this series, no obvious size effect was apparent on tumor visibility using either imaging modality. The present study shows that the false-negative rate of radiological as well as ultrasound mammography is higher than usually perceived. Given the complementarity of radiological and ultrasound mammography, their combination proves highly sensitive and highly specific. However, this conclusion will only apply when both breasts are examined completely, but not when ultrasound examination of the breasts is restricted to the imaging of an anomaly detected with radiological mammography. The time needed to perform a complete breast examination reduces its potential use in a screening program. On the other hand, its use appears of major relevance in patients with breasts where the performance of radiological mammography is questionable, such as younger women, or women with dense breasts. Copyright 1992 International Society of Ultrasound in Obstetrics and Gynecology