Mechanoelectric feedback in ventricular myocardium

The influence of three-dimensional diastolic strain on action potential morphology in intact myocardium was investigated in a series of integrative modeling studies. Action potential morphology is known to be altered by stretch of the ventricles---a process known as mechanoelectric feedback. The mechanisms governing this process are not well understood, and experimental investigations are complicated by the three-dimensional anatomy and anisotropy of the heart. Our hypothesis was that nonuniform three-dimensional diastolic strains in the intact myocardium contribute significantly to nonuniform alterations in action potential morphology.; The first objective of this dissertation was to investigate three-dimensional distributions of stress and strain in the passively inflated rabbit left ventricle. A high-order finite element model of the rabbit ventricles was developed that represented the geometry of the heart to within +/-0.55 millimeters and the fiber orientation to within +/-18.6 degrees. We estimated material properties of the passive rabbit ventricular myocardium using the ventricular model and the finite element method on a scalable parallel processing computer. Simulations of passive left ventricular inflation to 25 mm Hg pressure required 5.3 hours. Predicted epicardial fiber and cross fiber strains were within the accuracy of experimental measurements. The results suggest that transmural fiber strain is relatively uniform and cross fiber strain increases from epicardium to endocardium. Predicted Cauchy stress was generally ally higher in the fiber direction than the cross fiber direction. Rabbit myocardium may be more stiff than that of the canine or rat.; The second objective of this dissertation was to investigate possible mechanisms governing mechanoelectric feedback in the intact myocardium. Previous theories based on experimental observations and models of mechanoelectric coupling have focused on changes in length of the myofiber or sarcomere as the primary governing mechanical stimulus. Our results suggest that fiber strain alone cannot account for the transmural variation in action potential amplitude in response to stretch. A model of mechanosensitive conductance based on fiber and cross fiber strain produced both the reduction in epicardial action potential amplitude and corresponding transmural gradient. Mechanoelectric feedback may thus be governed by deformation other than changes in fiber length.