Purpose: To evaluate whether breast ultrasound should be routine practice following contrast enhanced spectral mammography.
Materials and Methods: Consecutive screening and diagnostic contrast mammography examinations with concurrent breast ultrasound were retrospectively collected (05/2012 - 02/2016). A radiologist assigned a separate Breast Imaging Reporting and Data System (BIRADS) score for contrast mammography and for ultrasound. BIRADS scores were grouped into three categories: benign appearing (BIRADS 1, 2); probably benign, short term follow-up (BIRADS 3); or suspicious appearing (BIRADS 0, 4, 5). Patients with a suspicious appearing lesion on either ultrasound or contrast mammography underwent biopsy. The associations between malignant pathology to either suspicious appearing contrast mammography or suspicious appearing ultrasound were calculated. The sensitivities and specificities of contrast mammography and ultrasound were calculated.
Results: Eighty-seven breasts was biopsied, 37 (43%) biopsies were malignant and 50 (57%) were benign. Among malignant biopsies, contrast mammography had a sensitivity of 97%, compared to 92% with ultrasound. None of the malignant biopsies were benign appearing with contrast mammography. One case of follow-up contrast mammography had a malignant appearing ultrasound and proved to be malignant in biopsy.
Specificity of contrast mammography was 40%, significantly higher than ultrasound 8%.
While suspicious appearing contrast mammography was associated with malignant biopsies (p<0.0001), suspicious appearing ultrasound was not (p=0.985).
Conclusion: When contrast mammography is suspicious appearing, subsequent ultrasound and biopsy is appropriate. When contrast mammography score is BIRADS 3 we suggest correlation with ultrasound. If the contrast mammography is benign appearing, the routine use of ultrasound is questionable as it may lead to unnecessary biopsies.
