An Image-driven Approach to Micromechanical Characterization of Infarcted Myocardium

Introduction:: Myocardial infarction (MI) is accompanied by the formation of a fibrotic scar in the left ventricle (LV) and initiates significant alterations in the architecture and constituents of the LV free wall (LVFW). Despite valuable studies on the evolution of mechanical and structural properties of MI scar tissues, structure-function analysis of LVFW tissues as a whole, including infarct and remote regions and their region-specific kinematic information, remains largely unexplored. Previous studies have indicated that LV adaptation is highly individual; thus, the identification of remodeling mechanisms post-MI demands a fully subject-specific approach that can integrate a myriad of structural alterations at the fiber-level up to changes in bulk biomechanical adaptation at the tissue-level. We present an image-driven micromechanical approach that assimilates subject-specific biaxial mechanical data, histological studies, and digital image correlation (DIC), within an in-silico framework delineating the fiber-level remodeling mechanisms driving tissue-level adaptation for each subject. The in-silico model presented in this work can serve as a key platform for the personalized design of post-MI myocardial tissues at any specific timepoint post-MI and will aid in delineating subject-specific LV remodeling, enhancing the traditional measures of LV performance.

Materials and Methods:: MI was induced in twenty-four (n=24) rats by ligation of the left anterior descending artery. Six rats (n=6) underwent sham surgery and were used as control subjects. Animals were sacrificed at five timepoints: control, 1-wk, 2-wk, 3-wk, and 4-wk post-MI. Echocardiographic and hemodynamic measurements were obtained prior to sacrifice. Full-thickness rectangular LVFW samples were isolated from excised hearts and underwent biaxial mechanical testing. Samples then underwent strain mapping via DIC. The LVFW samples were then sectioned in to 5 µm thick slices and stained with picrosirius red. The transmural fibrous architecture, including both myo- and collagen fibers, and fiber content area fractions were characterized via an automated pixel-counting algorithm. Additionally, collagen fiber undulation was calculated from 2-D images. Subject-specific finite-element (FE) models were developed (Fig. 1), incorporating the 3-D distribution of myo- and collagen volume fractions, fiber architecture, and collagen undulation. A fiber-specific hyperelastic incompressible constitutive model was used to model the passive behavior of the myocardium. An inverse problem approach was used to estimate model parameters such that the LVFW models matched tissue-level biaxial mechanical data.

Results, Conclusions, and Discussions:: The LVFW exhibited a transition from anisotropic behavior at control to isotropic behavior at late MI (Fig. 2-I). Both myo- and collage fibers exhibited reduced alignment and greater angular dispersion at later post-MI timepoints (Fig 2-II). We found that a combination of a progressively diffused collagen fiber structure in the scar region and a similarly disorganized myofiber architecture in the healthy region leads to the loss of overall LVFW anisotropy post-MI, offering an important tissue-level hallmark for LV maladaptation post-MI. In contrast, our results suggest that reductions in collagen undulation are an adaptive mechanism competing against LV wall thinning. DIC indicated the maximum strains in the LVFW are experienced by the scar region. This observation, combined with our modeling results, lead us to suggest LV wall thinning is a primary determinant of LVFW regional strain. Our approach offers a platform to design personalized hydrogel injection strategies to reinforce and suppress native adaptive and maladaptive mechanisms, respectively, to prevent the transition of MI to heart failure. We have taken an essential step towards modeling the multiscale, non-uniform, and subject-specific
fiber-level remodeling of the LVFW following MI. We used our models to clarify the relationship between fiber-level remodeling and development of passive cardiac kinematics post-MI. Ultimately, the delineation of subject-specific LV remodeling, with a focus on
regional biomechanical changes throughout the LV, could enhance traditional measures of LV performance and inform novel subject-specific interventions.