Control of a Pediatric Pulsatile Ventricular Assist Device: A Hybrid Cardiovascular Model Study.

The aim of this work is to study pediatric pneumatic ventricle (PVAD) performance, versus VAD rate (VADR) and native heart rate (HR) ratio Rr (VADR/HR). The study uses a hybrid model of the cardiovascular system (HCS). HCS consists of a computational part (a lumped parameter model including left and right ventricles, systemic and pulmonary arterial and venous circulation) interfaced to a physical part. This permits the connection of a VAD (15 mL PVAD). Echocardiographic and hemodynamic data of a pediatric patient (average weight 14.3 kg, HR 100 bpm, systemic pressure 75/44 mm Hg, CO 1.5 L/min) assisted apically with asynchronous PVAD were used to set up a basal condition in the model. After model tuning, the assistance was started, setting VAD parameters (ejection and filling pressures, systole duration) to completely fill and empty the PVAD. The study was conducted with constant HR and variable VADR (50-120, step 10, bpm). Experiments were repeated for two additional patients' HRs, 90 and 110 bpm and for two values of systemic arterial resistance (Ras ) and Emax . Experimental data were collected and stored on disk. Analyzed data include average left and right ventricular volumes (LVV, RVV), left ventricular flow (LVF), VAD flow (VADF), and total cardiac output (COt). Data were analyzed versus Rr. LVV and RVV are sensitive to Rr and a left ventricular unloading corresponds in general to a right ventricular loading. In the case of asynchronous assistance, frequency beats are always present and the beat rate is equal to the difference between HR and VADR. In the case of pulsatile asynchronous LVAD assistance, VADR should be chosen to minimize frequency beat effects and right ventricular loading and to maximize left ventricular unloading.

Reproduction of continuous flow left ventricular assist device experimental data by means of a hybrid cardiovascular model with baroreflex control.

Long-term mechanical circulatory assistance opened new problems in ventricular assist device-patient interaction, especially in relation to autonomic controls. Modeling studies, based on adequate models, could be a feasible approach of investigation. The aim of this work is the exploitation of a hybrid (hydronumerical) cardiovascular simulator to reproduce and analyze in vivo experimental data acquired during a continuous flow left ventricular assistance. The hybrid cardiovascular simulator embeds three submodels: a computational cardiovascular submodel, a computational baroreflex submodel, and a hydronumerical interface submodel. The last one comprises two impedance transformers playing the role of physical interfaces able to provide a hydraulic connection with specific cardiovascular sites (in this article, the left atrium and the ascending/descending aorta). The impedance transformers are used to connect a continuous flow pump for partial left ventricular support (Synergy Micropump, CircuLite, Inc., Saddlebrooke, NJ, USA) to the hybrid cardiovascular simulator. Data collected from five animals in physiological, pathological, and assisted conditions were reproduced using the hybrid cardiovascular simulator. All parameters useful to characterize and tune the hybrid cardiovascular simulator to a specific hemodynamic condition were extracted from experimental data. Results show that the simulator is able to reproduce animal-specific hemodynamic status both in physiological and pathological conditions, to reproduce cardiovascular left ventricular assist device (LVAD) interaction and the progressive unloading of the left ventricle for different pump speeds, and to investigate the effects of the LVAD on baroreflex activity. Results in chronic heart failure conditions show that an increment of LVAD speed from 20 000 to 22 000 rpm provokes a decrement of left ventricular flow of 35% (from 2 to 1.3 L/min). Thanks to its flexibility and modular structure, the simulator is a platform potentially useful to test different assist devices, thus providing clinicians additional information about LVAD therapy strategy.

Continuous-flow pump model study: the effect on pump performance of pump characteristics and cardiovascular conditions.

This model study evaluates the effect of pump characteristics and cardiovascular data on hemodynamics in atrio-aortic VAD assistance. The model includes a computational circulatory sub-model and an electrical sub-model representing two rotary blood pumps through their pressure-flow characteristics. The first is close to a pressure generator-PG (average flow sensitivity to pressure variations, -0.047 l mmHg(-1)); the second is closer to a flow generator-FG (average flow sensitivity to pressure variations, -0.0097 l mmHg(-1)). Interaction with VAD was achieved by means of two interfaces, behaving as impedance transformers. The model was verified by use of literature data and VAD onset conditions were used as a control for the experiments. Tests compared the two pumps, at constant pump speed, in different ventricular and circulatory conditions: maximum ventricular elastance (0.44-0.9 mmHg cm(-3)), systemic peripheral resistance (781-1200 g cm(-4) s(-1)), ventricular diastolic compliance C p (5-10-50 cm(3) mmHg(-1)), systemic arterial compliance (0.9-1.8 cm(3) mmHg(-1)). Analyzed variables were: arterial and venous pressures, flows, ventricular volume, external work, and surplus hemodynamic energy (SHE). The PG pump generated the highest SHE under almost all conditions, in particular for higher C p (+50 %). PG pump flow is also the most sensitive to E max and C p changes (-26 and -33 %, respectively). The FG pump generally guarantees higher external work reduction (54 %) and flow less dependent on circulatory and ventricular conditions. The results are evidence of the importance of pump speed regulation with changing ventricular conditions. The computational sub-model will be part of a hydro-numerical model, including autonomic controls, designed to test different VADs.

A modular computational circulatory model applicable to VAD testing and training.

Aim of this work was to develop a modular computational model able to interact with ventricular assist devices (VAD) for research and educational applications. The lumped parameter model consists of five functional modules (left and right ventricles, systemic, pulmonary, and coronary circulation) that are easily replaceable if necessary. The possibility of interacting with VADs is achieved via interfaces acting as impedance transformers. This last feature was tested using an electrical VAD model. Tests were aimed at demonstrating the possibilities and verifying the behavior of interfaces when testing VADs connected in different ways to the circulatory system. For these reasons, experiments were performed in a purely numerical mode, simulating a caval occlusion, and with the model interfaced to an external left-VAD (LVAD) in two different ways: with atrioaortic and ventriculoaortic connection. The caval occlusion caused the leftward shift of the LV p-v loop, along with the drop in arterial and ventricular pressures. A narrower LV p-v loop and cardiac output and aortic pressure rise were the main effects of atrioaortic assistance. A wider LV p-v loop and a ventricular average volume drop were the main effects of ventricular-aortic assistance. Results coincided with clinical and experimental data attainable in the literature. The model will be a component of a hydronumerical model designed to be connected to different types of VADs. It will be completed with autonomic features, including the baroreflex and a more detailed coronary circulation model.

A new hybrid electro-numerical model of the left ventricle.

The paper presents a new project of a hybrid numerical-physical model of the left ventricle. A physical part of the model can be based on electrical or hydraulic structures. Four variants of the model with numerical and physical heart valves have been designed to investigate an effect of a heart assistance connected in series and in parallel to the natural heart. The LabVIEW real time environment has been used in the model to increase its accuracy and reliability. A prototype of the hybrid electro-numerical model of the left ventricle has been tested in an open loop and closed loop configuration.

A hybrid mock circulatory system: testing a prototype under physiologic and pathological conditions.

Hydraulic models of circulation are used to test mechanical heart assist devices and for research and training purposes. However, when compared with numerical models, they are rather expensive and often not sufficiently flexible or accurate. Flexibility and accuracy can be improved by merging numerical models with physical models, thus obtaining a hybrid model where numerical and physical sections are connected by an electrohydraulic interface. This concept is applied here to represent left ventricular function. The resulting hybrid model is inserted into the existing closed loop model of circulation. The hybrid model reproduces ventricular function by a variable elastance numerical model. Its interaction with the hydraulic sections is governed by measuring left atrial and systemic arterial pressures and computing the left ventricular output flow by the resolution of the corresponding equations. This signal is used to control a flow generator reproduced by a gear pump driven by a DC motor. Results obtained under different circulatory conditions demonstrate the behavior of the ventricular model on the pressure-volume plane and report the trend of the main hemodynamic variables.