Layer-specific analysis of myocardial deformation for assessment of infarct transmurality: comparison of strain-encoded cardiovascular magnetic resonance with 2D speckle tracking echocardiography.

AIMS Separate analysis of endocardial and epicardial myocardial layer deformation has become possible using strain-encoded cardiovascular magnetic resonance (SENC) and 2D-dimensional speckle tracking echocardiography (Echo). This study evaluated and compared both modalities for the assessment of infarct transmurality as defined by late gadolinium enhancement (LGE) cardiovascular magnetic resonance (CMR). METHODS AND RESULTS In 29 patients (age 62.4 ± 11.7 years, 23 male) with ischaemic cardiomyopathy, SENC using 1.5 T CMR and Echo were performed. Peak circumferential systolic strain of the endocardial and the epicardial layer of 304 myocardial segments was assessed by SENC and by Echo. The segmental transmurality of myocardial infarction was determined as relative amount of LGE (0%: no infarction; 1-50%: non-transmural infarction; 51-100%: transmural infarction). Endocardial and epicardial strain defined by SENC and by Echo differed significantly between segments of different infarct transmurality determined by CMR. Endocardial layer circumferential strain analysis by Echo and by SENC allowed distinction of segments with non-transmural infarction from non-infarcted segments with similar accuracy [area under the curve (AUC) 0.699 vs. 0.649, respectively, P = 0.239]. Epicardial layer circumferential strain analysis by Echo and by SENC allowed distinction of transmural from non-transmural myocardial infarction defined by LGE CMR with similar accuracy (AUC 0.721 vs. 0.664, respectively, P = 0.401). Endocardial strain by SENC correlated moderately with endocardial strain by Echo (r = 0.50; standard error of estimate = 5.2%). CONCLUSION Layer-specific analysis of myocardial deformation by Echo and by SENC allows discrimination between different transmurality categories of myocardial infarction with similar accuracy. However, accuracy of both methods is non-optimal, indicating that further tools for improvement should be evaluated in the future.