Longitudinal strain reflects the interaction of myocardial contractility to afterload in rat models of hemodynamic overload-induced heart failure

Abstract

Abstract Background Two-dimensional (2D) speckle tracking echocardiography (STE)-derived myocardial strain parameters are sensitive markers of left ventricular (LV) systolic function. Novel findings suggest that the contractile state of the myocardium, afterload and preload are major determinants of STE measurements. However, the hypothesis that longitudinal strain expresses the interaction between contractility and loading conditions rather than contractility alone in hemodynamic overload-induced heart failure (HF) has not been tested. Purpose This study aimed to explore the connection between longitudinal strain and contractility, afterload and preload in rat models of pressure overload (PO)- and volume overload (VO)-induced heart failure (HF). Methods Pressure overload (PO)-induced HF was evoked by transverse aortic constriction ([TAC], n=14). Volume overload (VO)-induced HF was established by an aortocaval fistula ([ACF], n=12). Age-matched sham operated animals served as controls. Pressure-volume analysis was carried out to compute cardiac contractility (slope of end-systolic pressure-volume relationship [ESPVR]), afterload (arterial elastance [Ea]) and ventriculo-arterial coupling ([VAC] = Ea/ESPVR). Preload was evaluated by meridional end-diastolic wall stress (σend-diastolic). STE was performed to assess global longitudinal strain (GLS). Results GLS was impaired in both PO-induced HF (−5.9±0.6 vs. −12.9±0.5%, TAC vs Sham, P<0.001) and VO-evoked HF (−11.7±0.7 vs. −13.5±0.4%, ACF vs Sham, P=0.048). Hemodynamic measurements indicated that the TAC group presented with maintained ESPVR, increased Ea and enhanced σend-diastolic. In contrast, the ACF group was characterized by reduced ESPVR, decreased Ea and enhanced σend-diastolic. Ordinary least squares non-linear regression revealed that GLS was predominantly determined by afterload (Ea) in the TAC model and by contractility (ESPVR) in the ACF model. In accordance, GLS showed a strong correlation with Ea in case of PO-induced HF (R= 0.848, P<0.001) and with ESPVR in case of VO-evoked HF (R=−0.526; P=0.008), respectively. Furthermore, GLS also demonstrated strong correlation with VAC in both the TAC and the ACF models. Of particular interest, a robust correlation between VAC and GLS could also be detected in the entire study population (R= 0.654, P<0.001). Conclusion Both afterload and contractility define GLS. Hence, under conditions when both factors become altered, GLS reflects VAC. Funding Acknowledgement Type of funding source: Public grant(s) – National budget only. Main funding source(s): NVKP_16-1-2016-0017