S L Savaridas1, D B Taylor2, D Gunawardana1, M Phillips3. 1. Royal Perth Hospital, Radiology Department, GPO Box X2213, Perth, WA, 6001, Australia. 2. Royal Perth Hospital, Radiology Department, GPO Box X2213, Perth, WA, 6001, Australia; School of Medicine and Pharmacology, University of Western Australia, Crawley, WA, 6009, Australia. Electronic address: donna.taylor@health.wa.gov.au. 3. Harry Perkins Institute for Medical Research, Medical Research Foundation of RPH Building, Wellington St, Perth, WA, 6001, Australia.
Abstract
AIM: To compare background parenchymal enhancement (BPE) on contrast-enhanced (CE) spectral mammography (CESM) with CE magnetic resonance imaging (MRI), and evaluate how these relate to hormonal status, mammographic breast density (MBD) and MRI fibroglandular tissue volume (FGTV). METHODS AND MATERIALS: Between June 2012 to October 2015, participants in a cancer staging study underwent full-field digital mammography (FFDM), CEMRI, and CESM. Two readers independently rated FGTV, MBD, and BPE using the Breast Imaging-Reporting and Data System (BI-RADS) criteria. Inter-reader reliability was estimated using weighted kappa (k) and correlations between BPE, MBD, and FGTV calculated using Spearman's correlation coefficient. Associations with hormonal status were evaluated using multilevel ordinal regression analysis. RESULTS: Of the 96 eligible participants, 66 women (35-77 years) underwent CESM and CEMRI. Reasons for exclusion were declined or withdrawn consent (n=18), inadequate renal function (n=2), claustrophobia (n=2), previous reaction to contrast medium (n=2), mild reaction to contrast medium following CESM (n=2), lack of vascular access (n=1), neoadjuvant chemotherapy (n=1), CESM equipment failure (n=1), and unclear in one case. Inter-reader agreement was substantial (k=0.67) for CESM BPE, slight (k=0.19) for CEMRI BPE, moderate (k=0.57) for MRI FGTV and fair (k=0.35) for MBD. CESM BPE showed significant correlation with MBD (rho=0.36, p<0.0001), FGTV (rho=0.52, p<0.0001), and MRI BPE (rho=0.49, p<0.0001). BPE was significantly reduced in the post-menopausal group for CEMRI and CESM (p<0.05). CESM BPE did not significantly fluctuate during the menstrual cycle. CONCLUSION: CESM BPE is correlated with MBD, FGTV, and CEMRI BPE, has better inter-reader reliability than CEMRI, and is not influenced by the menstrual cycle. Grading the degree of BPE on CESM could be a useful addition to breast cancer risk assessment tools.
AIM: To compare background parenchymal enhancement (BPE) on contrast-enhanced (CE) spectral mammography (CESM) with CE magnetic resonance imaging (MRI), and evaluate how these relate to hormonal status, mammographic breast density (MBD) and MRI fibroglandular tissue volume (FGTV). METHODS AND MATERIALS: Between June 2012 to October 2015, participants in a cancer staging study underwent full-field digital mammography (FFDM), CEMRI, and CESM. Two readers independently rated FGTV, MBD, and BPE using the Breast Imaging-Reporting and Data System (BI-RADS) criteria. Inter-reader reliability was estimated using weighted kappa (k) and correlations between BPE, MBD, and FGTV calculated using Spearman's correlation coefficient. Associations with hormonal status were evaluated using multilevel ordinal regression analysis. RESULTS: Of the 96 eligible participants, 66 women (35-77 years) underwent CESM and CEMRI. Reasons for exclusion were declined or withdrawn consent (n=18), inadequate renal function (n=2), claustrophobia (n=2), previous reaction to contrast medium (n=2), mild reaction to contrast medium following CESM (n=2), lack of vascular access (n=1), neoadjuvant chemotherapy (n=1), CESM equipment failure (n=1), and unclear in one case. Inter-reader agreement was substantial (k=0.67) for CESM BPE, slight (k=0.19) for CEMRI BPE, moderate (k=0.57) for MRI FGTV and fair (k=0.35) for MBD. CESM BPE showed significant correlation with MBD (rho=0.36, p<0.0001), FGTV (rho=0.52, p<0.0001), and MRI BPE (rho=0.49, p<0.0001). BPE was significantly reduced in the post-menopausal group for CEMRI and CESM (p<0.05). CESM BPE did not significantly fluctuate during the menstrual cycle. CONCLUSION: CESM BPE is correlated with MBD, FGTV, and CEMRI BPE, has better inter-reader reliability than CEMRI, and is not influenced by the menstrual cycle. Grading the degree of BPE on CESM could be a useful addition to breast cancer risk assessment tools.