Background Late gadolinium enhancement (LGE) imaging is the most established technique in cardiovascular magnetic resonance (CMR) for myocardial tissue characterization. It is routinely performed for reliable assessment and quantification of myocardial fibrosis, irrespective of ischemic or non-ischemic origin. Beside the diagnostic value, the presence of LGE serves as a predictor of outcome. Beside the conventional bright-blood LGE technique, the technique of dark-blood LGE has shown to be a reliable CMR method for viability assessment in ischemic cardiomyopathy. 

Objectives To evaluate dark-blood LGE imaging compared to conventional bright-blood LGE for the detection of myocardial fibrosis in non-ischemic cardiomyopathies. 

Methods 306 consecutive patients underwent 1,5T CMR due to clinically suspected non-ischemic cardiomyopathy. N=102 patients were excluded due to a history of coronary artery disease or an ischemic LGE pattern. N=36 patients were excluded due to incomplete dataset acquisition. N=18 cases were excluded due to insufficient image quality. 150 patients (age 49 ± 17 years; 34% female) with different types of non-ischemic cardiomyopathy were consecutively included. All patients underwent 1.5 T CMR with acquisition of both dark-blood and conventional bright-blood LGE imaging. Corresponding short-axis stacks of both techniques wereanalysed for both presence and pattern of LGE in a head-to-head comparison. 

Results Absence of LGE was reported in 44% of the patients (n=66). For the detection of non-ischemic scars, dark-blood LGE showed a sensitivity of 96% (CI 85-100), a specificity of 99% (CI 95-100), and an accuracy of 98% (CI 64-100). All scars were of a non-ischemic or inflammatory type located in a midwall or subepicardial region. The positive predictive value was 98% (CI 86-100) and the negative predictive value was 98% (CI 93-100). No significant difference in total scar burden between the two LGE techniques was observed (dark-blood imaging with mean 4.97 ± 3.41 % enhanced volume of total myocardial volume, and bright-blood imaging with mean 5.03 ± 3.37 % enhanced volume, p=0.532). In the assessment of myocardial scar burden, dark-blood LGE showed no systematic bias (on average, dark-blood LGE measurements were 0.1% lower than bright-blood LGE, p=0.505).

Conclusion In this study, consisting of prospectively enrolled patients with clinically suspected non-ischemic or inflammatory cardiomyopathies, CMR dark-blood LGE has shown to be non-inferior to conventional bright-blood LGE imaging in the detection of non-ischemic scars analyzed by pattern and extent.