Myocardial scar burden is an important prognostic factor after myocardial infarction. This cohort study compared assessment of left ventricle scar burden between pathological Q waves on electrocardiography (ECG), Selvester multiparametric ECG scoring system for scar burden, and global longitudinal strain (GLS) by speckle-tracking echocardiography 6 months after myocardial infarction. The scar burden was defined by late gadolinium enhancement cardiac magnetic resonance as fraction of total left ventricle tissue. ECG measures were presence of pathologic Q waves and Selvester scores. GLS was the average of peak strain from 16 left ventricle segments. In 34 patients aged 58 ± 10 years (mean ± SD), the scar burden was 19% (9, 26) (median [quartiles]) and 79% had scar burden >5%. Patients with scar burden >5% more frequently had pathologic Q waves (63% vs 14%) and had worse Selvester scores (5 [3, 7] vs 0 [0, 1]) and worse GLS (-16.6 ± 2.4% vs -19.9 ± 1.1%). Pathologic Q waves, Selvester scores, ejection fraction, and GLS related to scar burden in univariable analyses. Sensitivity and specificity for detecting scar burden >5% was 63% and 86% (pathologic Q waves), 89% and 86% (Selvester score), 81% and 86% (ejection fraction), 89% and 86% (GLS), and 96% and 71% (combination of Q waves, Selvester score, and GLS). In conclusion, Selvester score and GLS related to scars 6 months after myocardial infarction, and pathologic Q waves were only weakly associated with scar and GLS was associated with scar independently of ECG markers.
Copyright © 2017 Elsevier Inc. All rights reserved.