Search

Continuous-flow left ventricular assist devices (CF LVADs) are rotary blood pumps that improve mean blood flow, but with potential limitations of non-physiological ventricular volume unloading and diminished vascular pulsatility. In this study, we tested the hypothesis that left ventricular unloading with increasing CF LVAD flow increases myocardial flow normalized to left ventricular work. Healthy (n = 8) and chronic ischemic heart failure (IHF, n = 7) calves were implanted with CF LVADs. Acute hemodynamics and regional myocardial blood flow were measured during baseline (LVAD off, clamped), partial (2-4 L/min) and full (>4 L/min) LVAD support. IHF calves demonstrated greater reduction of cardiac energy demand with increasing LVAD support compared to healthy calves, as calculated by rate-pressure product. Coronary artery flows (p < 0.05) and myocardial blood flow (left ventricle (LV) epicardium and myocardium, p < 0.05) decreased with increasing LVAD support in normal calves. In the IHF model, blood flow to the septum, LV, LV epicardium, and LV myocardium increased significantly with increasing LVAD support when normalized to cardiac energy demand (p < 0.05). In conclusion, myocardial blood flow relative to cardiac demand significantly increased in IHF calves, thereby demonstrating that CF LVAD unloading effectively improves cardiac supply and demand ratio in the setting of ischemic heart failure.

Experiments were performed to test the hypothesis that viscoelastic properties of the swine myocardium are independent of heart rate (HR), preload (PL), and afterload (AL). Left ventricular pressure and aortic flow (AoF) waveforms were recorded in 13 swine. At different paced heart rates, an inferior vena caval occlusion (IVC) was used to reduce PL, then the IVC was released and simultaneously the aorta was clamped to increase AL. Equivalent left ventricular pressure waveform pairs consisting of an ejecting waveform (denoted as LVP) and isovolumic waveform (denoted as hydromotive pressure, HMP) were selected according to specified criteria resulting in 371 equivalent waveform pairs. From the selected waveform pairs and corresponding aortic flow waveforms, the viscoelastic properties (k and epsilon1) were estimated by HMP = LVP + epsilon1 V(EJ) + k x LVP x AoF. Here epsilon1 is the parallel elastance, k is the myocardial friction, and V(EJ) is the integral of AoF over ejection. Next, using k, epsilon1, LVP, and AoF waveforms, HMP was estimated using the equation above. To validate the model, the measured HMP and model-calculated HMP were compared for 371 matched waveform pairs (R2 = 0.97, SEE = 3.7 mmHg). The viscoelastic parameters (k and epsilon1) did not exhibit any clear or predictable dependence on HR, PL, and AL.

Numerous parameter estimation techniques exist for characterizing the arterial system using electrical circuit analogs. These techniques are often limited by requiring steady-state beat conditions and can be computationally expensive. Therefore, a new method was developed to estimate arterial parameters during steady and transient beat conditions. A four-element electrical analog circuit was used to model the arterial system. The input impedance equations for this model were derived and reduced to their real and imaginary components. Next, the physiological input impedance was calculated by computing fast Fourier transforms of physiological aortic pressure (AoP) and aortic flow. The approach was to reduce the error between the calculated model impedance and the physiological arterial impedance using a Jacobian matrix technique which iteratively adjusted arterial parameter values. This technique also included algorithms for estimating physiological arterial parameters for nonsteady physiological AoP beats. The method was insensitive to initial parameter estimates and to small errors in the physiological impedance coefficients. When the estimation technique was applied to in vivo data containing steady and transient beats it reliably estimated Windkessel arterial parameters under a wide range of physiological conditions. Further, this method appears to be more computationally efficient compared to time-domain approaches.