Model-based estimation of internal heart power in aortic valve disease patients

Aortic valve disease (AVD) causes pressure overload of the left ventricle (LV) which may trigger adverse remodeling and, eventually, precipitate progression towards heart failure (HF). A frequent therapy is transvenous aortic valve implant (TAVI) which aims at reducing the transvalvular pressure gradient. However, TAVI does not always reverse HF symptoms. We aim to develop personalized computer models of LV electromechanics (EM), suitable for predicting acute TAVI-induced changes in LV function and derive clinical markers such as internal heart power (IHP) which are believed to offer prognostic value of longer term outcomes. EM simulations of a heartbeat fitted to pre-treatment conditions were carried out and validated against complementary clinical data that were not used for model fitting. Based on the assumption of a reduced transvalvular pressure gradient <; 20 mmHg, a change in characteristic impedance of the aorta was estimated to simulate post-treatment conditions. IHP computed for pre- and post-treatment conditions was compared against the image-based clinical estimation of these quantities. All in silico models replicated clinical markers within prescribed margins of accuracy. Model predictions of IHP were comparable to the clinically derived analog.