BACKGROUND: The objective of this study was to determine whether the use of ultrasound and percutaneous breast biopsies in patients with screen-detected nonpalpable abnormalities can reduce benign open surgical biopsies of the breast without increasing cost or sacrificing detection of potentially curable breast carcinomas. METHOD: Using a computerized mammography database and consecutive logs of needle localization procedures and fine- and large core needle biopsies of a single university-based breast imaging practice, the authors determined the breast carcinoma yield and cost of diagnosis over a 14-year period and the changes that occurred over time with the sequential introduction of ultrasound, ultrasound-guided biopsies, and stereotactic biopsies. RESULTS: The overall breast carcinoma yield for needle localization biopsies of nonpalpable lesions increased from 21% in 1984 to 68% in 1998 (P < 0.0001). The yield for nonpalpable masses increased from 21% to 87% (P < 0.0001) over the same period. The selective use of ultrasound alone and percutaneous fine- and large core needle biopsy resulted in a substantial reduction in benign open surgical biopsies. A cost analysis showed a 50% reduction in the average expense of discovering breast carcinoma. The breast carcinomas detected after introduction of these methods were prognostically favorable with 88% measuring 1.5 cm or less in size and 66% measuring less than 1 cm. CONCLUSIONS: Selective use of ultrasound and imaging-guided percutaneous biopsies can significantly reduce the number of benign open surgical biopsies generated by mammographic screening. This can result in substantial cost savings without decreasing the sensitivity for detecting small potentially curable lesions.