Impact of background parenchymal enhancement on the diagnosis of enhancing lesions in breast MRI: a systematic approach
摘要
Dynamic contrast-enhanced MRI (DCE-MRI) enhances both suspicious lesions and normal tissue, known as background parenchymal enhancement (BPE). There is controversial evidence regarding whether BPE compromises the accuracy of breast MRI. Our aim was to investigate the influence of BPE in the detection of breast lesions by DCE-MRI using an ROI-based approach.
Materials and methodsMRI Images of consecutive patients with indeterminate or suspicious findings on mammography or ultrasound (BI-RADS 0, 3, 4, 5). tested between May 2020 and September 2023 were retrospectively analyzed. Both breasts were divided into 10 regions of interest (ROIs), and BPE, BI-RADS and lesion type were assessed according to the MRI BI-RADS lexicon for each ROI by two blinded readers. BPE was dichotomized into minimum/mild (BPE1) and moderate/marked (BPE2). Patients without a standard reference were excluded. The standard reference was defined as histology or 1 year follow-up. The MRI performance was assessed using ROC analysis in an ROI-based analysis. Inter-reader agreement was assessed using k statistics.
ResultsA total of 316 patients (mean of 54 years, SD 10.6; range 31–84) were included. Among the 282 lesions histopathologically analyzed, 129 were malignant and 153 were benign. BPE1 was observed in 78% of patients and BPE2 in 22%. AUC did not differ between BPE1 (0.989) and BPE2 (0.988), p > 0.05. Sensitivity was 98.1% for BPE1 and 100% for BPE2, with comparable specificity: 95.5% (BPE1) vs. 92.6% (BPE2), p > 0.05.
ConclusionOur study demonstrated that BPE does not have a negative impact on the detection and identification of suspicious lesions on breast MRI.
Key Points