P599Native T1 mapping is useful for detection of myocardial fibrosis in cases with ischemic and non-ischemic myocardial diseases

Abstract

Abstract Background Evaluation of myocardial fibrosis (MF) as late gadolinium enhancement (LGE) on MRI is useful for differential diagnosis of various myocardial diseases and prediction of future adverse cardiac events in some specific myocardial diseases. Gadolinium contrast is contraindicated for cases with severe renal dysfunction, therefore non contrast MRI is necessary for detection of MF in cases with both myocardial disease and severe renal dysfunction. Purpose We aimed to evaluate diagnostic accuracy of native T1 mapping for detection of MF compared with LGE in cases with various myocardial diseases, including ischemic and non-ischemic myocardial diseases. Methods We selected consecutive 40 patients who were suspected of having various myocardial diseases and underwent cardiac MRI, using 1.5T MRI (Ingenia, Philips) in 10 cases (25%) or 3T MRI (Ingenia, Philips) in 30 cases (75%), including native T1 mapping (without contrast) and LGE using contrast media from Jan 2018 to Feb 2019 in our institution. We evaluated diagnostic accuracy for detection of MF in left ventricular myocardium (LVM) of native T1 mapping image compared with LGE as the gold standard, in a patient-based and segment-based analysis. In T1 mapping images, segmental high T1 lesions were defined as MF. In a segment-based analysis, MF was evaluated using 17 LVM segments model in American Heart Association. Results MF was detected in 139 LVM segments in 25 (63%) cases. Sensitivity, specificity, positive predictive value, negative predictive value and diagnostic accuracy of native T1 mapping for detection of MF were 90%, 89%, 95%, 80% and 90% in a patient-based analysis, and 63%, 96%, 84%, 89% and 88% in a segment-based analysis (left figure). Native T1-values of LVM with MF were significantly higher than LVM without LGE (1351±79 vs 1093±124 in 1.5T and 1562±131 vs 1291±43 in 3T) (p<0.05 and p<0.01). Interobserver agreement of native T1 mapping and LGE were not significantly different (0.88 and 0.89, P=0.70). Overall diagnostic accuracy of native T1 mapping for detection of MF in a patient-based analysis, was not significantly different in between the cases with ischemic (n=18) and non-ischemic (n=22) myocardial disease (90% and 83.3%, P=0.10). Conclusion Native T1 mapping (without contrast) is useful for detection of MF in various myocardial diseases and high diagnostic accuracy is expected especially in a patient-based analysis.