Myocardial extracellular volume estimation by CMR predicts functional recovery following acute myocardial infarction

Abstract

Background The transmural extent of myocardial infarction (MI) assessed by late gadolinium enhancement (LGE) cardiovascular magnetic resonance (CMR) is a reference standard for prediction of functional recovery. In acute MI (AMI), myocardial oedema and effects of reperfusion therapy reduce the predictive accuracy of LGE. LGE assesses tissue dichotomously as “viable” or “non-viable”, but does not consider severity of tissue damage within the hyperenhanced infarct zone. Extracellular volume (ECV) estimation, using native and post-contrast T1 mapping CMR, allows for quantitative assessment of severity of myocardial damage. We aimed to assess if CMR measurement of infarct ECV offers additional predictive value over LGE extent for contractile recovery in reperfused AMI. Methods 35 patients with first ST-segment elevation AMI treated by primary percutaneous coronary intervention underwent acute (day 2) and convalescent (3 months) CMR at 3 Tesla. Cine imaging, tissue tagging, modified Look-Locker inversion T1 mapping (3-3-5 acquisition with 3x R-R interval recovery epochs) natively and 15 min post gadolinium-contrast administration and LGE imaging at 20 min were performed. The ability of acute infarct ECV and acute transmural extent of LGE to predict convalescent wall motion, ejection fraction (EF) and strain were compared. A per-patient analysis was performed using a region of interest corresponding to the core of the infarct, excluding any microvascular obstruction (MO). Segmental analysis was also performed and evaluated using a multilevel linear mixed-effects model to account for non-independence of segmental data. Results Per-patient, ECV and transmural extent of LGE correlated with convalescent wall motion score (r = 0.43, p < 0.01; r = 0.41, p = 0.02 respectively) and convalescent EF (r = - 0.56, p < 0.01; r = -0.36, p = 0.04). Per-segment, acute ECV and LGE transmural extent were associated with convalescent wall motion score (β = 0.55, r = 0.54, p < 0.01; β = 0.50, r = 0.50, p < 0.01, Figure 1). ECV had higher accuracy than LGE to predict improvement in wall motion (area under receiver-operator-characteristics (ROC) curve 0.77 vs. 0.68, p = 0.02, Figure 2). ECV of ≤0.5 had sensitivity 81% and specificity 65% for prediction of improvement in segmental function. Adding ECV analysis to a 50% LGE transmural extent cut-off for prediction of improved wall motion in dysfunctional segments increased sensitivity from 87% to 89% and specificity from 34% to 81%. In multivariable analysis, acute infarct ECV was independently associated with both convalescent infarct strain and EF (β = 0.58, p < 0.001; β = -0.39, p = 0.02) whereas LGE was not (β = 0.13, p = 0.35; β = -0.20, p = 0.2). ECV in patients with and without MO was similar (0.54+/-0.18 vs. 0.57 +/-0.10, p = 0.6). Conclusions Acute infarct ECV in reperfused AMI predicts LV functional recovery, and is both independent of and more accurate than transmural extent by LGE. There is clinical utility for ECV mapping post-AMI. (Figure presented).