CMR-Analysis (including machine learning)

1345549 - Changes in Remote and Border Zone Circumferential Strain after a Myocardial Infarction using DENSE MRI

Wednesday, January 25, 2023

Location: E-POSTER

Presenting Author(s)

TH

Tawfik Hussein, MSc

Graduate Student
Georgia Institute of Technology, Georgia, United States

Primary Author(s)

TH

Tawfik Hussein, MSc

Graduate Student
Georgia Institute of Technology, Georgia, United States

Co-Author(s)

MP

Muralidhar Padala, PhD

Associate Professor of Cardiothoracic Surgery
Emory University, Georgia, United States
JO

John N. Oshinski, PhD

Professor of Radiology and Biomedical Engineering
Emory University
Atlanta, Georgia, United States

Background: Global indices of ventricular function fail to capture the local adaptive remodeling that occurs in the heart post-myocardial infarction (MI). Local indices, however, such as strain, could provide insights into early changes in myocardial function. Hence, the goal of this work was to quantify the local changes in circumferential and longitudinal Green strain (E_CC and E_LL, respectively) in porcine myocardium using displacement encoding with simulated echoes (DENSE) [1]. The hypothesis is that MI will cause differential changes in strain in the regions adjacent and remote to the myocardial infarct.

Methods: A postero-lateral MI was induced in 4 healthy Yorkshire pigs by injecting ethanol into the left circumflex artery (LCX). ECG-gated MRI was acquired in the mid short axis (SA) as well as the two-chamber (2CH) long axis (LA) to measure strain (i.e., E_CC and E_LL) at baseline (pre-MI) and 3 months post-MI. Both the SA and LA images were divided into 3 zones (i.e., infarcted, border, and remote), based on late Gadolinium enhancement images and the AHA 17-segment model (Figure 1). Measured strains were averaged for each zone, and their respective mean and standard deviation were calculated. Finally, the peak systolic strain was calculated for both timepoints, and a paired t-test was used to assess if the difference between the mean baseline and MI strains per zone was significant.

Results: In the mid SA images, E_CC changed over both timepoints across all three zones. Peak systolic E_CC in the remote zone increased by – 0.03 ± 0.01 (from baseline to MI) with a p-value = 0.007 while it did not change significantly in the border zone (Figure 2). However, in the 2CH LA images, peak systolic E_LL did not change from baseline to MI (p = 0.50).

Conclusion: Myocardial strains in areas remote to the infarct increase in radial and circumferential direction post-MI, but not longitudinal direction. Measurements of myocardial strain can provide more mechanistic insights into the LV remodeling process and may be predictive markers of adverse remodeling.