We assessed the usefulness and limitations of utilizing apparent diŠusion coe‹cient (ADC) values on diŠusion-weighted imaging (DWI) for the diŠerential diagnosis of benign and malignant non-mass-like breast lesions. We retrospectively reviewed 27 such lesions (16 malignant, 11 benign) detected on magnetic resonance (MR) imaging and analyzed the enhancing patterns of dynamic contrast-enhanced DCE-MRI (distribution and internal enhancement), kinetic curve patterns, and ADC values. All images were obtained with a 1.5-tesla MR unit, with patients supine. On DCE-MRI, malignant lesions tended to show either segmental or branching-ductal distribution, and when lesions with these patterns were considered malignant, sensitivity was 68.8z; speciˆcity, 63.6z; positive predictive value (PPV), 73.3z; negative predictive value (NPV), 58.3z; and accuracy, 66.7z. Kinetic curve analysis did not reliably diŠerentiate benign and malignant non-mass-like lesions. There was no signiˆcant diŠerence between the mean ADC value of the malignant lesions, 0.968×10 -3 mm 2 /s at b=1000 s/mm 2 , and that of benign lesions, 1.207×10 -3 mm 2 /s (P =0.109). Receiver operating characteristic (ROC) analysis revealed the most eŠective threshold of ADC value for diŠerentiating tumors as 1.1×10 -3 mm 2 /s; values lower than this were observed more often in malignant than benign lesions (P=0.054). Us of this threshold yielded sensitivity of 68.8z; speciˆcity, 72.7z; PPV, 78.6z; NPV, 61.5z; and accuracy, 70.4z. Combining the ADC value criteria with the analysis of DCE-MRI pattern increased sensitivity to 93.8z, negative predictive value (NPV) to 85.7z, and accuracy to 77.8z but decreased speciˆcity to 54.5z. Use of ADC values does not adequately improve DCE-MRI performance for diŠerential diagnosis of non-mass-like breast lesions, but adding the ADC value criteria to the DCE-MRI pattern analysis improves sensitivity, NPV, and accuracy.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.