A Method for Automatically Optimizing Medical Devices for Treating Heart Failure: Designing Polymeric Injection Patterns

Abstract

Heart failure continues to present a significant medical and economic burden throughout the developed world. Novel treatments involving the injection of polymeric materials into the myocardium of the failing left ventricle (LV) are currently being developed, which may reduce elevated myofiber stresses during the cardiac cycle and act to retard the progression of heart failure. A finite element (FE) simulation-based method was developed in this study that can automatically optimize the injection pattern of the polymeric "inclusions" according to a specific objective function, using commercially available software tools. The FE preprocessor TRUEGRID((R)) was used to create a parametric axisymmetric LV mesh matched to experimentally measured end-diastole and end-systole metrics from dogs with coronary microembolization-induced heart failure. Passive and active myocardial material properties were defined by a pseudo-elastic-strain energy function and a time-varying elastance model of active contraction, respectively, that were implemented in the FE software LS-DYNA. The companion optimization software LS-OPT was used to communicate directly with TRUEGRID((R)) to determine FE model parameters, such as defining the injection pattern and inclusion characteristics. The optimization resulted in an intuitive optimal injection pattern (i.e., the one with the greatest number of inclusions) when the objective function was weighted to minimize mean end-diastolic and end-systolic myofiber stress and ignore LV stroke volume. In contrast, the optimization resulted in a nonintuitive optimal pattern (i.e., 3 inclusions longitudinallyx6 inclusions circumferentially) when both myofiber stress and stroke volume were incorporated into the objective function with different weights.

Injectable Acellular Hydrogels for Cardiac Repair

“…Various volumes of 1, 2, and 4 mL were injected; 2-and 4-mL injections led to superior LVDA and LVSAwith 2-mL injections resulting in significant thickening (despite material degradation) and more pronounced trends for functional improvements. These findings illustrate the importance of injection volume in stabilizing the myocardial wall [67]. Other parameters, such as number and pattern of injections, although relevant through theoretical analysis [67], have yet to be investigated in a clinically relevant model.…”

“…Theoretical models have implied that material properties, specifically mechanics and volume, are important to consider when selecting the type of bulking agent to ameliorate dilation and increased stress in the myocardial wall [35,[65][66][67]. Using a finite element (FE) model to simulate the effects of injecting a non-contractile material into the myocardium, Wall et al showed that bulking the myocardium was sufficient to attenuate post-MI geometric changes and, thus, decrease stress in the myocardial wall [35].…”

“…FE models have also been employed to evaluate the effects of material volume and distribution in the myocardium and showed that they influence the extent of remodeling, depending on their pattern of injection (Fig. 2) [35,67]. After testing several injection patterns modeled with an objective function that was weighted to minimize mean end-diastolic and end -systolic myofiber stress and LV stroke volumes, it was determined that the maximum number of injections leads to the highest reduction in fiber stress [67].…”

“…2) [35,67]. After testing several injection patterns modeled with an objective function that was weighted to minimize mean end-diastolic and end -systolic myofiber stress and LV stroke volumes, it was determined that the maximum number of injections leads to the highest reduction in fiber stress [67]. These simulations provide insight to the relevance of bulking material properties, specifically mechanics and volume distribution, and present more evidence to pursue injectable material therapies to control LV remodeling.…”

“…Reference FE model for infarcted canine heart (a) and modified FE model with polymeric inclusion injection pattern into dilated LV (b) can be used to assess the effects of material volume and distribution on myofiber stress. Figure adapted from [67] while unable to promote cell adhesion, alginate, like fibrin, resulted in significantly higher arteriole density. Despite alginate's superiority in long-term functional outcomes, fibrin treatment groups had significantly smaller infarcts compared to controls.…”