Quantifying Coronary Morphology in Mice: Towards a Computational Approach to Understanding Heart Transplant Rejection

The rejection rate of heart transplant (HTx) remains considerably high despite recent medical advancements. The major cause of chronic rejection is cardiac allograft vasculopathy (CAV), a pathology that affects 50% of HTx recipients with a 32% 5-year mortality. CAV is triggered by a strong immune response of the host to the graft and it is characterized by neointimal formation in the coronary arteries, ultimately culminating in critical vessel occlusion and graft failure. Currently, no effective treatments or diagnostic biomarkers are available for CAV, mainly due to a poor understanding of the disease initiation mechanisms. The morphology of coronary arteries is believed to play an important role within CAV etiology and progression, and, consequently, within HTx long-term outcome. However, a clear cause-effect relationship has not been outlined to date. In this context, the in vivo models used in pre-clinical studies have proven to be useful tools, but they are often associated with ethical concerns, high resource demand, and long protocol timeline. Additionally, they carry difficulties in measuring biomechanical quantities of interest and do not allow an exhaustive understanding of pathology complexity. In silico models can overcome these roadblocks since they can simulate different pathology scenarios and predict the effect of treatments while requiring fewer resources. We hypothesize that integrating in silico modeling with in vivo-based morphometric analysis of mouse coronaries will enhance the understanding of CAV development and optimize anti-chronic rejection research. Accordingly, the aim of this thesis is to build a portfolio of morphometric measurements from mice coronaries, to i) provide the scientific community with solid foundation to study CAV, ii) create the fundament to study the correlation between coronary anatomy and pathology development, and iii) supply pivotal data to calibrate/validate predictive computational models of CAV, thereby advancing research toward effective therapies. Mice hearts (n=6) were ex-vivo perfused with Microfil solution, to enable µCT-imaging of coronary arteries and cardiac tissue. The coronary tree, with main focus on left coronary artery (LCA), was 3D reconstructed through an automatic segmentation protocol and subjected to a centerline-based morphometric analysis to evaluate specific geometrical measurements of interest. The parameters retrieved were: i) curvature, torsion and tortuosity to characterize the complexity of the tree-like structure; ii) angle of bifurcation and branching ratio as markers for further investigation of the correlation between blood flow and pathology insurgency; iii) single coronaries’ cross-sectional areas and diameters to feed future computational fluid-dynamic models of CAV with blood flow repartition, aiming to elucidating pathology onset and development. The intra and inter-sample morphometric analysis reported: i) consistent curvature and torsion along LCA branches, ii) decreasing diameter and area from mother to daughter vessels, iii) diffuse asymmetry on average. The parameters retrieved were validated by comparing them with the corresponding ones in the ex vivo models, measured on microscope images. The proposed integrated approach provides the scientific community with a portfolio of measurements to create solid foundation to study CAV and to fill a fundamental gap of knowledge in CAV literature.