Strain-based characterization of electromechanical tissue properties using patient-specific simulation of dyssynchronous hearts: a pilot study

Abstract

Abstract Funding Acknowledgements Type of funding sources: Foundation. Main funding source(s): Dutch Heart Foundation (ERA-CVD JTC2018 grant 2018T094; Dr. Dekker Program grant 2015T082 to J.L.) and the Netherlands Organisation for Scientific Research (NWO- ZonMw, VIDI grant 016.176.340 to J.L.) Background/Introduction Interpretation of regional strain patterns in patients with failing hearts is often challenging due to presence of different underlying tissue abnormalities, including electrical conduction delay and myocardial infarction (MI). Purpose To test whether an automatic strain-based computer simulation algorithm can differentiate between electrical and ischemic contractile dysfunction in patients with mechanical dyssynchrony. Methods Three heart failure patients with reduced left ventricular ejection fraction (33 ± 3%) and wide QRS (160 ± 7ms, LBBB morphology) were retrospectively included, two with and one without MI. Longitudinal strain patterns obtained in 18 myocardial segments using speckle tracking echocardiography (STE) and left ventricular (LV) volume measurements were used as an input to the patient-specific simulation algorithm, which automatically optimizes a set of global and regional myocardial tissue parameters such that the regional strain patterns and global volumes simulated by the CircAdapt model of the human heart and circulation were similar to the strains and volumes measured in the patients. As output, the patient-specific simulations revealed onset time of mechanical activation (relative to first activated segment) and relative severity of contractile dysfunction in all 18 segments. We compared predicted mechanical activation delays and segmental contractile dysfunction values with independent measurements of QRS width and transmurality of late gadolinium enhancement (LGE), respectively. Results The model was able to reproduce the dyssynchronous regional strain patterns as measured in the patients (Figure 1: representative example). In all patients, significant septal—to-lateral wall mechanical activation delays (33-55ms) were predicted (Figure 2: blue bullseye plot of same patient used for figure 1), matching the presence of electrical activation delays on the patient’s ECGs (i.e. wide QRS and LBBB morphology). Furthermore, all simulations predicted a global loss of contractile dysfunction, matching the low ejection fractions reported in those patients. The simulations of the two patients with MI predicted more severe contractile dysfunction (>60%) in the regions with transmural LGE (Figure 2: red bullseye plot versus gold standard LGE-MRI image of same patient used for figure 1). Conclusion This pilot study demonstrated the ability of an automatic patient-specific simulation algorithm to differentiate between electrical and ischemic myocardial disease substrates in heart failure patients with electrical conduction abnormalities with and without MI. Future studies will investigate whether this computer algorithm can improve diagnosis of patients with mechanical dyssynchrony. Abstract Figure 1 Abstract Figure 2