Rick Schreurs

110 Chapter 6 and hemodynamics during pacing delay variations as applied in the animal experiments. Previous experimental and clinical studies have shown that the CircAdapt model realistically relates local ventricular myofiber mechanics to global cardiovascular hemodynamics in dyssynchronous and paced hearts [16,21,24–26]. The source code of the CircAdapt model used for all simulations as well as the Matlab® (The MathWorks, Natick, MA) scripts to perform all simulations and analysis are provided as an online data supplement. In summary, the CircAdapt model is a reduced order model of the human four-chamber heart connected to a closed-loop cardiovascular system, with lumped pulmonary and systemic circulations. It uses a simplified ventricular geometry, where cardiac walls are represented by thick spherical shells consisting of contractile myocardium. The MultiPatch module enables cardiac walls to be subdivided into an arbitrary number of wall segments (patches). Tissue properties and activation time can differ between patches, but all patches within a wall share a common wall tension and curvature. Since wall tension is the same in all patches within a wall, spatial location within a wall is not required to calculate deformation in a patch. In CircAdapt wall tension and curvature determine cavity pressure through Laplace’s law [21,23] Fiber stress in a patch is the sum of an active component, representing myofiber contraction, and a passive component. The active stress component incorporates length-dependence of the force generated and the duration of contraction. The passive component provides a non-linear relationship between myofiber stress and strain. More details on the phenomenological model of myocardial contraction and the validation of the MultiPatch module are previously published by Walmsley et al [21]. Simulating a healthy human reference heart and circulation Cardiac adaptation implemented in the CircAdapt model was used to obtain a reference parameterization that represents a healthy human cardiovascular system [22,27,28]. The tissue volumes and areas in the cardiac walls and large blood vessels were adapted as described previously (Arts et al., 2005, 2012). A resting cardiac output of 5.1L/min and heart rate of 70bpm were assumed. Cardiac output was tripled and the heart rate was doubled during the stress-state of the adaptation process. Mean arterial pressure was maintained at 92mmHg during the adaptation process. The resulting reference simulation was used as the basis for subsequent pacing simulations. Using electrical activation to simulate the pacing delay optimization protocol We divided the ventricular wall in the same amount of segments as the number of available electrodes (52 LV free wall, 50 RV free wall and 10 septal segments). Time of onset of activation was assigned based on the electrical activation times measured in the animal experiments. As previously stated, in the current MultiPatch module the segments were considered to be mechanically coupled in series, meaning that the order in which patches were placed was not significant [21]. This allowed sorting of the activation times per wall