CineCT imaging platform for in-vivo and ex-vivo measurement of myocardial biomechanics post myocardial infarction and following intramyocardial delivery of theranostic hydrogel

Abstract Funding Acknowledgements Type of funding sources: Public grant(s) – National budget only. Main funding source(s): National Institute of Health (NIH) Purpose Myocardial infarction (MI) induces acute regional changes in myocardial strain and stiffness in the infarct and the remote areas of the left ventricle (LV), which lead to adverse changes in LV geometry and function. We hypothesize that cineCT imaging could evaluate these biomechanical changes along with the effects of intramyocardial delivery of theranostic hydrogels. Introduction We present an experimental platform to assess changes in the deformation of the LV myocardium using contrast cineCT (CCT) imaging of the beating porcine heart (active deformation) before and after acute MI and intramyocardial delivery of an imageable theranostic hydrogel. We then assess the acute effects of hydrogel delivery early post-MI on biomechanics (passive deformation) using an ex vivo perfused heart preparation. Methods Contrast cineCT imaging was performed using 64-slice CT on 5 Yorkshire pigs without MI (n = 3) or with MI (n = 2). MI pigs had serial imaging performed before and 1 hour after acute surgical coronary occlusion to induce anterolateral MI. One MI pig was also imaged 1 hour after intramyocardial injection of a novel imageable theranostic iodinated hydrogel within the MI region. Post euthanasia, excised hearts were flushed with chilled UW cardioplegic solution and mounted on a custom inflation apparatus for cineCT imaging during LV inflation by external pump. LV pressure was cycled between 10 and 60 mmHg at 35 bpm. Dilute iohexol was injected into aortic root and UW perfusate (15 ml, 1 ml/sec). CineCT image series were reconstructed, contrast enhanced, resampled to the LV long axis (Z), and exported as a series of 10 CT volumes covering 0-90% of the cardiac/inflation cycle. Volumes were registered incrementally using nonlinear image registration (BioImageSuite) and the calculated displacement at each time point was exported at a resolution of 1 mm. A custom Matlab program was used to fit the displacement field to local trilinear polynomials and then calculate the displacement gradients and 3D Lagrangian strains. To estimate the accuracy of this approach, cardiac volumes were also numerically deformed using a 10 pixel translation and 5% triaxial stretch. Results We successfully acquired serial in-vivo and ex-vivo 3D CineCT images for assessment of the active and passive LV myocardial deformation and tracked deformation through the full cardiac/inflation cycle (Figure 2). Numerical deformation tests showed average tracking errors of < 0.2 mm (1/4 pixel) in the X,Y,Z directions of the volume. These resulted in Lagrangian strain errors of < 0.47% for the in-plane strains EXX and EYY (radial and circumferential plane) and < 0.5% for EZZ (long axis). Conclusions We have developed a novel CineCT imaging platform that allows for high resolution in-vivo and ex-vivo measurement of myocardial biomechanics post-MI and following intramyocardial delivery of imageable theranostic hydrogels, which may improve early active and passive biomechanics.