Meme Lezyonlarının Kontrastlı Mamografi İle Kantitatif Analizi ve Histopatolojik Sonuçlar ile Karşılaştırılması
Özet
The aim of this study is to quantitatively evaluate the contrast-enhancing lesions in contrast-enhanced mammography (CEM) and to investigate the effectiveness of contrast-enhanced mammography in distinguishing benign and malignant lesions. Secondly, it was aimed to compare the enhancement values in contrast-enhanced mammography according to histopathological type, grade and molecular subtypes in malignant lesions. In this study, 170 lesions of 164 patients who underwent CEM between January 01, 2019, and January 11, 2022 were retrospectively analyzed. All lesions had a histopathological result or were stable for at least 2 years during follow-up to be considered benign. Lesions were grouped as benign, non-invasive, and invasive cancer. 44 lesions (25.8%) were benign, 9 lesions (5.3%) were noninvasive cancer, and 117 lesions (68.9%) were invasive cancer. For quantitative analysis, the measurements were made with the similar region of interest (ROI) sizes from each lesion, glandular (G) and fatty (F) tissue on both craniocaudal (CC) and mediolateral oblique (MLO) CEM images. The relative enhancement of the lesion for the fatty tissue and glandular tissue was calculated with the %R formula. The mean %R CC G values were 2.32, 2.88, and 4.76 for the benign, non-invasive, and invasive cancer groups, respectively. The mean %R CC F values were 3.70, 4.06, and 5.97 for benign, non-invasive, and invasive cancer groups, respectively. When the binary groups that created the significant difference were examined using the Bonferroni-corrected Mann-Whitney U test, the difference between invasive cancer and the benign group was found to be significant in both % R CC F and % R CC G (p<0.001). While the highest %R MLO G and %R MLO F means were found in the invasive cancer group, the lowest %R MLO G and %R MLO F means were found in the benign group (p<0.001). ROC analysis was performed for the differentiation of benign and malignant (noninvasive and invasive cancers) for %R CC G, %R CC F, %R MLO G, and %R MLO F, and the area under the curve was significant for all four variables (p<0.001). Cut-off values were 2.786 for %R CC G, 3.257 for %R CC F, 3.118 for %R MLO G, and 3.826 for %R MLO F. For the contrast dynamics of the lesions, the relative signal change between MLO and CC images was calculated with the %RSD formula. No significant difference was found between the types 1 (persistent), type 2 (plateau), and type 3 (washout) contrast enhancement patterns grouped according to %RSD G and %RSD F values between benign, non-invasive and invasive cancer groups (p>0.05). In the invasive cancer group, there was no significant difference for any of the quantitative CEM measurements according to the histopathological type, grade, and molecular subtypes (p>0.05). In conclusion, quantitative measurements in CEM can be used to distinguish between benign and malignant breast lesions, regardless of fat or glandular tissue. However, when the contrast-enhancement patterns of type 1, type 2, and type 3 are examined, no difference is observed between benign and malignant lesions. This can show that contrast enhancement patterns in CEM and breast MRI may differ from each other due to the different times of contrast-enhanced images.
