Background: The pathological Q-wave (QW) is an important indicator of infarcted myocardial volume indicating a worse prognosis compared to non-Q-wave (NQW) infarctions. Traditional classification divides infarcts into transmural and non-transmural based on QW and NQW. This view has been challenged by the advent of late gadolinium enhancement (LGE) MR imaging. Conventional LGE (Conv-LGE) detection of subendocardial MI is limited by bright blood pool. Dark Blood LGE imaging (DB-LGE) nulls the blood pool improving the conspicuity and accuracy of detection of subendocardial infarcts. We hypothesize that improved detection of subendocardial enhancement with DB-LGE will result in improved correlation of electrocardiogram (ECG) and extent of infarction.
Methods: Sixty-four clinically confirmed infarction patients were enrolled in this prospective study. All the participants underwent cardiac MR imaging including conv-LGE and DB-LGE. Twelve-lead ECG were performed on the same day. The patients were divided into QW and NQW groups by one experienced cardiologist. MI quantitation was by MI% (the ratio of MI volume to whole myocardial volume) and transmural grading, compared using paired t-test and Wilcoxon-test, respectively. The image quality obtained by Conv-LGE and DB-LGE were evaluated according to the signal intensity ratio (SIR) and contrast-to-noise ratio (CNR).
Results: Fifty-six subjects were enrolled in the final analysis [23 (41%) QW and 33 (59%) NQW infarcts]. For the QW cohort, both sequences classified infarcts as transmural in 21/23 (91%) subjects and subendocardial in 2/23 (9%). For the NQW cohort, both sequences classified infarcts as transmural in 16/33 (48%) subjects and subendocardial in 17/33 (52%). Using BB-LGE there were significant differences in detecting subendocardial infarcts in QW and NQW cohorts (Z=-5.85, P<0.001). The MI% of QW group was greater than in NQW group (24.2±10.3 vs.15.9±9.8, P=0.003). Compared to Conv-LGE, BB-LGE provided higher CNR and SIR between infarcted myocardium and blood pool (6.3±2.6 vs. 2.1±1.3, P<0.001; 5.4±1.9 vs. 1.3±0.2, P<0.001). BB-LGE detected more subendocardial infarcted segments in the QW group and NQW group (Z=-4.24, P<0.001; Z=-5.57, P<0.001). The larger MI% was displayed in BB-LGE than in Conv-LGE in both QW group and NQW group (24.2±10.3 vs. 22.6±10.3, P<0.001; 15.9±9.8 vs.14.6±9.6, P=0.001).
Conclusions: Compared to conventional LGE, DB-LGE can provide more accurate detection and characterization of infarction in terms of transmurality and subendocardial extent. This is important for evaluating QW and NQW MIs. Due to nulling the high signal of blood pool, DB-LGE can effectively improve the identification of subendocardial MI which may be missed on conventional LGE. Therefore, in both QW and NQW MIs, DB-LGE detects more subendocardial MIs and larger MI% is found. This may facilitate more accurate quantitative MR assessment of both QW and NQW MIs and further empower LGE volume as a predictive biomarker.