Comparison of cardiovascular magnetic resonance feature tracking and tagging for the assessment of left ventricular systolic strain in acute myocardial infarction

Abstract

AIMS: To assess the feasibility of feature tracking (FT)-measured systolic strain post acute ST-segment elevation myocardial infarction (STEMI) and compare strain values to those obtained with tagging. METHODS: Cardiovascular MRI at 1.5T was performed in 24 patients, 2.2 days post STEMI. Global and segmental circumferential (Ecc) and longitudinal (Ell) strain were assessed using FT and tagging, and correlated with total and segmental infarct size, area at risk and myocardial salvage. RESULTS: All segments tracked satisfactorily with FT (p<0.001 vs. tagging). Total analysis time per patient was shorter with FT (38.2±3.8 min vs. 63.7±10.3 min, p<0.001 vs. tagging). Global Ecc and Ell were higher with FT than with tagging, apart from FT Ecc using the average of endocardial and epicardial contours (-13.45±4.1 [FT] vs. -13.85±3.9 [tagging], p=0.66). Intraobserver and interobserver agreement for global strain were excellent for FT (ICC 0.906-0.990) but interobserver agreement for tagging was lower (ICC<0.765). Interobserver and intraobserver agreement for segmental strain was good for both techniques (ICC>0.7) apart from tagging Ell, which was poor (ICC=0.15). FT-derived Ecc significantly correlated with total infarct size (r=0.44, p=0.03) and segmental infarct extent (r=0.44, p<0.01), and best distinguished transmurally infarcted segments (AUC 0.77) and infarcted from adjacent and remote segments. FT-derived Ecc correlated strongest with segmental myocardial salvage (rs=-0.406). CONCLUSIONS: FT global Ecc and Ell measurement in acute STEMI is feasible and robust. FT-derived strain is quicker to analyse, tracks myocardium better, has better interobserver variability and correlated more strongly with infarct, area at risk (oedema), myocardial salvage and infarct transmurality.