Pitfalls in the development of a rotary blood pump controller

The controller presents a major obstacle in the development of the rotary blood pump as a left ventricular assist device (LVAD). Clinically, LVAD flow is a good indicator in the regulation of circulatory conditions and pump flow changes, depending on pump preload and afterload. Many investigators have tried estimating pump flow by referencing the motor current. There have been pitfalls in in vitro experimental settings, however. Using a test loop with a pneumatically driven LV chamber and a centrifugal pump as an LVAD, we monitored pump flow and pressure head to evaluate the pump performance curve (H-Q curve). Under pulsatile LV conditions, the H-Q curve was a loop that changed, depending on LV contractility. The pneumatically driven LV chamber cannot mimic the Starling phenomenon, so the developed LV pressure does not change according to the LV preload. Rotary pump flow estimation is the most effective control method. In pulsatile conditions, however, the H-Q curve is a loop that changes under various LV contractility conditions, complicating determination of linear equation for calculating flow. In addition, the LV chamber in the test loop cannot mimic native heart contractility as described by Starling's law. This finding can lead to a misanalysis of the H-Q curve under pulsatile conditions.     

Recovery directed left ventricular assist device: a new concept

To promote cardiac recovery, we developed a recovery directed left ventricular assist device (RDLVAD) that consists of a valved apical conduit, an afterload controlling chamber (ACC), and a pump. We evaluated its efficacy by comparison with an ordinary LVAD. In each of six pigs with ischemia-induced heart failure, flow and pressure measurements were made while maintaining the total blood flow and arterial pressure equal in the two groups. RDLVAD was able to direct all the blood ejected from the LV into the ACC (0-15 mm Hg) but not into the aorta (73 mm Hg). In the ordinary LVAD, however, some ejection occurred into the aorta despite vigorous suction of the LV. Thus, RDLVAD increased DPTI/SPTI 2.3 times (p < 0.005) and decreased left ventricular end-diastolic pressure by 40% and maximum dP/dt by 20% (p < 0.05). Even the apical valve, at approximately half the diameter of the aortic valve, was able to allow all the blood ejected from the LV to enter the ACC. In one control group pig that achieved almost no ejection into the aorta, left ventricular relaxation and dilatation was extremely limited. RDLVAD may promote cardiac recovery by ensuring less LV work, a greater blood supply/demand ratio in the coronary circulation, and full ventricular relaxation.     

Preload sensitivity of the Jarvik 2000 and HeartMate II left ventricular assist devices

The in vitro sensitivity of continuous flow pumps to preload and afterload pressure has been well characterized. We compared flow in the Jarvik 2000 and HeartMate II continuous flow left ventricular assist devices (LVADs) at different inflow and outflow pressures and different pump speeds. This allowed us to measure the impact of a changing inflow pressure on the pump flow rate at different speeds but against a constant afterload. The resulting preload sensitivity curves showed that, overall, both LVADs have a mean preload sensitivity of 0.07 L/min/mm Hg in the physiologic ranges of pressures and flows encountered during normal operation. The HeartMate II pump had an increased preload sensitivity (up to approximately 0.1 L/min/mm Hg) as the preload was increased. The preload sensitivity of the Jarvik 2000 LVAD was more variable, having several peaks and troughs as the preload was increased. In future LVADs, improved preload sensitivity may allow passive regulation of pump output, optimize ventricular unloading, and decrease the risk of ventricular suction by the pump.     

In vitro controllability of the MagScrew total artificial heart system

The purpose of this study was to evaluate the in vitro responses to preload and afterload of our total artificial heart (TAH), the MagScrew TAH. The TAH consists of two blood pumps and a control logic, developed at the Cleveland Clinic, OH, and the MagScrew actuator and its electronic control system, developed by Foster-Miller Technologies, Inc., Albany, NY. Tests were performed on a mock circulatory loop, using water as a test fluid. Preload sensitivity of the Mag-Screw TAH demonstrated a Frank-Starling response to preload in automatic mode. A peak flow of 10 L/min was obtained, with a left atrial pressure of 13 mm Hg. The relationship between right atrial pressure and left atrial pressure was well balanced when tested with a left bronchial shunt flow of 5% and a range of pulmonary artery and aortic pressures. With respect to afterload response, the left pump showed a relatively low sensitivity, which allowed the pump to maintain perfusion over a wide range of aortic pressures. The right pump, on the other hand, was much more sensitive to pulmonary artery pressure, which provided a measure of protection against pulmonary congestion. The very effective physiologic response of the MagScrew TAH is believed to result from employment of a left master, alternating ejection control logic, high inherent sensitivity of the blood pumps to atrial pressure, a lower effective stroke volume for the right pump, and a scaling of right side motor ejection voltage to 80% of that used for the left side ejection.     

Effect of pump flow mode of novel left ventricular assist device upon end organ perfusion in dogs with doxorubicin induced heart failure

End organ effects of nonpulsatile (NP) and pulsatile (P) left ventricular assist device (LVAD) flow were compared in a canine model of doxorubicin-induced heart failure. After heart failure induction, a prototype bimodal LVAD was implanted. Hemodynamics, cardiac dimensions, and myocardial metabolism were monitored with the LVAD off (baseline) and on (in NP and P modes at 70% or 100% power). End organ perfusion was assessed by colored microsphere analysis. Seven dogs were used: two died before pump implantation and were excluded from analysis, and the remaining five survived to study termination. At 70% NP, ascending aortic flow and myocardial oxygen consumption (MVO2) decreased significantly. At 100% NP, LV dimensions decreased, aortic systolic, pulse, and LV pressures decreased but not significantly, and ascending aorta flow reversed. At 100% NP, coronary blood flow, MVO2, and LV free wall subepicardial and subendocardial blood flows decreased significantly. However, as NP support increased, the subepicardial/subendocardial blood flow ratio remained near baseline. At 100% NP, right ventricular perfusion decreased but not significantly, cerebral perfusion decreased significantly, and renal perfusion stayed constant. P mode results were similar, except that ascending aorta flow decreased significantly at 100% P instead of reversing as at 100% NP. These results suggest that end organ perfusion is not differentially affected by LVAD flow mode during chronic heart failure.     

In vitro hemodynamic evaluation of ventricular suction conditions of the EVAHEART ventricular assist pump

Purpose: Mismatches between pump output and venous return in a continuous-flow ventricular assist device may elicit episodes of ventricular suction. This research describes a series of in vitro experiments to characterize the operating conditions under which the EVAHEART centrifugal blood pump (Sun Medical Technology Research Corp., Nagano, Japan) can be operated with minimal concern regarding left ventricular (LV) suction. Methods: The pump was interposed into a pneumatically driven pulsatile mock circulatory system (MCS) in the ventricular apex to aorta configuration. Under varying conditions of preload, afterload, and systolic pressure, the speed of the pump was increased step-wise until suction was observed. Identification of suction was based on pump inlet pressure. Results: In the case of reduced LV systolic pressure, reduced preload (=10 mmHg), and afterload (=60 mmHg), suction was observed for speeds =2,200 rpm. However, suction did not occur at any speed (up to a maximum speed of 2,400 rpm) when preload was kept within 10-14 mmHg and afterload =80 mmHg. Although in vitro experiments cannot replace in vivo models, the results indicated that ventricular suction can be avoided if sufficient preload and afterload are maintained. Conclusion: Conditions of hypovolemia and/or hypotension may increase the risk of suction at the highest speeds, irrespective of the native ventricular systolic pressure. However, in vitro guidelines are not directly transferrable to the clinical situation; therefore, patient-specific evaluation is recommended, which can be aided by ultrasonography at various points in the course of support.     

Computational analysis of the effect of the type of LVAD flow on coronary perfusion and ventricular afterload

We developed a computational model to investigate the hemodynamic effects of a pulsatile left ventricular assist device (LVAD) on the cardiovascular system. The model consisted of 16 compartments for the cardiovascular system, including coronary circulation and LVAD, and autonomic nervous system control. A failed heart was modeled by decreasing the end-systolic elastance of the ventricle and blocking the mechanism controlling heart contractility. We assessed the physiological effect of the LVAD on the cardiovascular system for three types of LVAD flow: co-pulsation, counter-pulsation, and continuous flow modes. The results indicated that the pulsatile LVAD with counter-pulsation mode gave the most physiological coronary blood perfusion. In addition, the counter-pulsation mode resulted in a lower peak pressure of the left ventricle than the other modes, aiding cardiac recovery by reducing the ventricular afterload. In conclusion, these results indicate that, from the perspective of cardiovascular physiology, a pulsatile LVAD with counter-pulsation operation is a plausible alternative to the existing LVAD with continuous flow mode. An erratum to this article can be found at     

Analysis of the relationship between left ventricular pressure and motor current for evaluation of native cardiac function during left ventricular support with a centrifugal blood pump

Control of the ventricular assist device (VAD) for native heart preservation should be attempted, and the VAD could be one strategy for dealing with the shortage of donors in the future. In the application of nonpulsatile blood pumps for ventricular assistance from the ventricular apex to the aorta, bypass flow and hence the motor current of the pumps change in response to the ventricular pressure change. Utilizing these intrinsic characteristics of the continuous-flow pumps, in this study we investigated whether motor current could be used as an index for continuous monitoring of native cardiac function. In study 1, a centrifugal blood pump (CFP) VAD was installed between the apex and descending aorta of a mock circulatory loop. In this model, a baseline with a preload of 10 mmHg, afterload of 40 mmHg, and LV systolic pressure of 40 mmHg was used. The pump speed was fixed at 1300, 1500, and 1700 rpm, and LV systolic pressure was increased up to 140 mmHg by steps of 20 mmHg while the changes in LV pressure, motor current, pump flow, and aortic pressure were observed. In study 2, an in vivo experiment was performed using three sheep. A left heart bypass model was created using a centrifugal pump from the ventricular apex to the descending aorta. The LVP was varied through administration of dopamine while the changes in LV pressure, pump flow, and motor current at 1500 and 1700 rpm were observed. An excellent correlation was observed in both in vitro and in vivo studies in the relationship between motor current and LV pressure. In study 1, the correlation coefficients were 0.77, 0.92, and 0.99 for 1300, 1500, and 1700 rpm, respectively. In study 2, they were 0.88 (animal no. 1), 0.83 (animal no. 2), and 0.88 (animal no. 3) for 1500 rpm, and 0.95 (animal no. 2) and 0.93 (animal no. 3) for 1700 rpm. These results suggest that motor current amplitude monitoring could be useful as an index for the control of VAD for native heart preservation.     

Hydrodynamic characterisation of ventricular assist devices

A new mock circulatory system (MCS) was designed to evaluate and characterise the hydraulic performance of ventricular assist devices (VADs). The MCS consists of a preload section and a multipurpose afterload section, with an adjustable compliance chamber (C) and peripheral resistor (Rp) as principal components. The MCS was connected to a pulse duplicator system for validation, simulating a wide range of afterload conditions. Both pressure and flow were measured, and the values of the different components calculated. The data perfectly fits a 4-element electrical analogon (EA). The MCS was further used to assess the hydrodynamic characteristics of the Medos VAD as an example of a displacement pump. Data was measured for various MCS settings and at different pump rates, yielding device specific pump function graphs for water and pig blood. Our data demonstrate (i) flow sensitivity to preload and afterload and (ii) the effect of test fluid on hemodynamic performance.     

The influence of pump rotation speed on hemodynamics and myocardial oxygen metabolism in left ventricular assist device support with aortic valve regurgitation

Aortic valve regurgitation (AR) is a serious complication under left ventricular assist device (LVAD) support. AR causes LVAD-left ventricular (LV) recirculation, which makes it difficult to continue LVAD support. However, the hemodynamics and myocardial oxygen metabolism of LVAD support with AR have not been clarified, especially, how pump rotation speed influences them. An animal model of LVAD with AR was newly developed, and how pump rotation speed influences hemodynamics and myocardial oxygen metabolism was examined in acute animal experiments. Five goats (55 ± 9.3 kg) underwent centrifugal type LVAD, EVAHEART implantation. The AR model was established by placing a vena cava filter in the aortic valve. Hemodynamic values and the myocardial oxygen consumption, delivery, and oxygen extraction ratio (O₂ER) were evaluated with changing pump rotation speeds with or without AR (AR+, AR?). AR+ was defined as Sellers classification 3 or greater. AR was successfully induced in five goats. Diastolic aortic pressure was significantly lower in AR+ than AR? (p = 0.026). Central venous pressure, mean left atrial pressure, and diastolic left ventricular pressure were significantly higher in AR+ than AR? (p = 0.010, 0.047, and 0.0083, respectively). Although systemic flow did not improve with increasing pump rotation speed, LVAD pump flow increased over systemic flow in AR+, which meant increasing pump rotation speed increased LVAD-LV recirculation and did not contribute to effective systemic circulation. O₂ER in AR? decreased with increasing pump rotation speed, but O₂ER in AR+ was hard to decrease. The O₂ER in AR+ correlated positively with the flow rate of LVAD-LV recirculation (p = 0.012). AR caused LVAD-LV recirculation that interfered with the cardiac assistance of LVAD support and made it ineffective to manage with high pump rotation speed.     

Development of a novel drive mode to prevent aortic insufficiency during continuous-flow LVAD support by synchronizing rotational speed with heartbeat

Aortic insufficiency (AI) is a serious complication for patients on long-term support with left ventricular assist devices (LVAD). Postoperative aortic valve opening is an important predictor of AI. A system is presently available that can promote native aortic flow by reducing rotational speed (RS) for defined intervals. However, this system can cause a reduction in pump flow and lead to insufficient support. We therefore developed a novel “delayed copulse mode” to prevent AI by providing both minimal support for early systole and maximal support shortly after aortic valve opening by changing the RS in synchronization with heartbeat. To evaluate whether our drive mode could open the aortic valve while maintaining a high total flow (sum of pump flow and native aortic flow), we installed a centrifugal LVAD (EVAHEART^®; Sun Medical) in seven goats each with normal hearts and acute LV dysfunction created by micro-embolization of the coronary artery. We intermittently switched the drive mode from continuous (constant RS) with 100 % bypass to delayed copulse mode with 90 % bypass. Total flow did not significantly change between the two modes. The aortic valve opened when the delayed copulse mode was activated. The delayed copulse mode allowed the aortic valve to open while maintaining a high total flow. This novel drive mode may considerably benefit patients with severe heart failure on long-term LVAD support by preventing AI.     

In vitro evaluation of the PUCA II intra-arterial LVAD

The "pulsatile catheter" (PUCA) pump is a minimally invasive intra-arterial left ventricular assist device intended for acute support of critically ill heart failure patients. To assess the hydrodynamic performance of the PUCA II, driven by an Arrow AutoCat IABP driver, we used a (static) mock circulatory system in which the PUCA II was tested at different loading conditions. The PUCA II was subsequently introduced in a (dynamic) cardiovascular simulator (CVS) to mimic actual in vivo operating conditions, with different heart rates and 2 levels of left ventricular (LV) contractility. Mock circulation data shows that PUCA II pump performance is sensitive to afterload, pump rate and preload. CVS data demonstrate that PUCA II provides effective LV unloading and augments diastolic aortic pressure. The contribution of PUCA II to total flow is inversely related to LV contractility and is higher at high heart rates. We conclude that, with the current IABP driver, the PUCA II is most effective in 1:1 mode in left ventricles with low contractility.     

The isolated working mouse heart: methodological considerations

 Our aim was to develop a working isolated murine heart model, as the extensive use of genetically engineered mice in cardiovascular research requires development of new miniaturized technology. Left ventricular (LV) function was assessed in the isolated working mouse heart perfused with recirculated oxygenated Krebs-Henseleit bicarbonate buffer (37 °C pH 7.4) containing 11.1 mM glucose and 0.4 mM palmitate bound to 3% albumin. The hearts worked against an afterload reservoir at a height equivalent to 50 mmHg, and heart rate was controlled by electrical pacing of the right atrium. LV pressure was measured with a micromanometer connected to a small steel cannula inserted through the apex of the heart. The experimental protocol consisted of two interventions. First, following instrumentation and stabilization, the preload reservoir was raised from a pressure equivalent of 7 to 22.5 mmHg, while pacing at 390 beats·min^–1. Thereafter the height of the preload reservoir was set to 10 mmHg, and the pacing rate was varied from 260 to 600 beats·min^–1. Aortic and coronary flows were measured by timed collections of effluent from the afterload line and that dripping from the heart, respectively [aortic+coronary flow=cardiac output (CO)]. Elevation of LV end-diastolic pressure (LVEDP) from approximately 5 to 10 mmHg resulted in a twofold increase in average cardiac power [product of LV developed pressure (LVDevP) and CO], whereas myocardial contractility (first derivative of LV pressure, dP/dt) and LVDevP (LV systolic pressure–LVEDP) increased only minimally (5–10%). Measured LVEDP was lower than the equivalent height of the preload reservoir by an amount that was related to the heart rate. Cardiac power, LVDevP and dP/dt were stable at heart rates up to 400 beats·min^–1, but declined markedly with higher rates, consistent with the decrease in LVEDP. Thus, cardiac power was reduced to 50% of its maximum value when stimulated at approximately 500 beats·min^–1, and at even higher rates there was little ejection. By systematic manipulation of the height of the preload reservoir and heart rate, we conclude that LV afterload and preload can be assessed only by high-fidelity measurement of intraventricular pressures. The heights of the afterload column and the preload reservoir are unreliable and potentially misleading indicators of LV afterload and preload. Received: 28 September 1998 / Accepted: 25 January 1999     

Development of counterpulsation algorithm for a moving-actuator type pulsatile LVAD

A pulsatile left ventricular assist device (LVAD) was used to support the aortic blood pumping function of an injured left ventricle, and as a result helped its recovery. It is important to observe a left ventricle's pumping status and to adjust the operating status of a LVAD to reduce the left ventricle's pumping load and thus to enhance its recovery. To observe the left ventricle's pumping status, an electrocardiogram (ECG) signal is generally used because it is a result of the natural heart's blood pumping function. In this paper, we describe the development of an ECG based counterpulsation control algorithm that prevents simultaneous aortic blood co-pumping by a left ventricle and a moving-actuator type pulsatile LVAD and as a result, reduces the natural heart's pumping load. In addition, to verify the algorithm's applicability for LVAD control we designed three ECG based automatic pump control algorithms that use a developed counterpulsation control algorithm. These algorithms control the operating status of a LVAD automatically and, at the same time, maintain a counterpulsing status. The results of in vitro experiments show that the counterpulsing effect between a left ventricle and a LVAD was successfully produced and that the newly designed automatic pump control algorithms met their own control purposes with a counterpulsing effect.     

Daily transient discontinuation of extracorporeal LVAD to prevent thromboembolism of mechanical aortic valve prosthesis

Patients with mechanical aortic valves are generally contraindicated for left ventricular assist device (LVAD) insertion because the prosthetic valve often becomes fixed in closed position. A 41-year-old woman with mechanical aortic valve prosthesis experienced sudden chest pain and developed cardiogenic shock. A paracorporeal pulsatile LVAD and a monopivot centrifugal pump as a right VAD (RVAD) were implanted. The mechanical aortic valve was intentionally left in place. Soon after the operation, LVAD support was discontinued daily for few seconds to allow the mechanical aortic valve to open and to avoid thrombus formation. The patient was successfully weaned off RVAD and received anticoagulation therapy with warfarin. On postoperative day 141, she was transferred to a university hospital where a HeartMate II LVAD was implanted, and the aortic valve was successfully replaced with a bioprosthetic valve. The patient is currently awaiting heart transplantation.     

Use of a soft reservoir bag in a fully heparin-coated closed-loop cardiopulmonary bypass system for distal aortic perfusion during aortic surgery

A fully heparin-coated closed-loop cardiopulmonary bypass system has recently been introduced into clinical practice. Without a venous reservoir, however, it does not allow control of the preload to the heart. We connected a soft reservoir bag in parallel with a centrifugal pump to enable preload control and clinically evaluated this modified system for distal aortic perfusion during aortic surgery. We have used the modified system in 17 patients since November 2002. For venous drainage, we use long narrow cannulae (21 ± 2 French). We administered 1 mg/kg heparin without cardiotomy suction and 2 mg/kg heparin with suction. We compared the clinical results with those in 13 patients who underwent distal aortic perfusion with an open cardiopulmonary bypass circuit between January 2002 and February 2004. We also analyzed factors affecting the coagulation system in these 30 patients using multiple regression analysis. With the modified system, venous drainage was adequate despite the use of smaller cannulae, and heparin reduction was not associated with thrombotic complication or elevated D-D dimer levels. Abrupt rises in proximal aortic pressure on aortic cross-clamping could be avoided by allowing blood to drain into the soft reservoir bag. Clinical results were not different from those with an open system. In the multiple regression analysis, the peak activated clotting time tended to correlate with postoperative platelet counts. This system is effective in controlling the preload to the heart and allows the safe reduction of heparin dosage. It therefore seems useful for distal aortic perfusion during aortic surgery.     

In vitro evaluation of an axial flow pump: mock-pump interaction and an approach to control

Continuous flow pump support has emerged as an alternative therapy in patients with congestive heart failure. For long-term applications, it is important to have a control system that changes the pump function according to the physiological conditions of the patient, thereby preventing risk situations. In the early stages of development, the evaluation of control algorithms for artificial blood pumps can be done in vitro using cardiovascular mock systems. A systemic cardiovascular mock loop was constructed and an axial flow pump was connected to it. The level of pump assistance was estimated using a pulsatility index (IPAo) obtained from the aortic pressure wave. An IPAo proportional-integral control system was implemented and its responses to peripheral resistance and systemic compliance changes were evaluated. IPAo is an indicator of the assistance level of a continuous flow pump operated at constant speed. The IPAo control algorithm responds by increasing the pump speed when peripheral resistance or systemic compliance is reduced. Control system operation around an IPAo fixed value provides a safety point for pump operation by maintaining aortic pressure pulsatility and avoiding ventricular suction. In vitro experimental results show that the IPAo can be taken into consideration in multiobjective control algorithm designs.     

Mini hemoreliable axial flow LVAD with magnetic bearings: part 2: design description

This paper gives the preliminary configuration of the flow geometry used to eliminate bearing thrombus by forced pressure wash-out of the bearing gaps. This left ventricular assist device (LVAD) is physiologically controllable without extraneous sensors based on the measurement of pump differential pressure using the magnetic bearings. Knowing the LVAD differential pressure allows safe cyclic variation of impeller rpm with feedback around differential pressure, which obtains desired pressure pulsatility. Flow pulsatility is known to be of major benefit for minimizing thrombus in both the pump and arteries. It also results in improved perfusion of many organs. The ability of a conventional virtual zero power feedback loop to axially control the bearing in a long-term drift free manor is also explained.     

World-smallest LVAD with 27 g weight, 21 mm OD and 5 l min−1 flow with 50 mmHg pressure increase

To investigate the feasibility of a long-term left ventricular assist device (LVAD) placed in the aortic valve annulus, an implantable aortic valve pump (21 mm outer diameter, weighing 27 g) was developed. The device consists of a central rotor and a stator. The rotor assembly incorporates driven magnets and an impeller. The stator assembly has a motor coil with an iron core and outflow guide vanes. The device is to be implanted identically to an aortic valve replacement, occupying no additional anatomic space. The pump delivers the blood directly from left ventricle to the aortic root, like a natural ventricle, therefore causing less physiologic disturbance to the natural circulation. Neither connecting conduits nor ‘bypass’ circuits are necessary. The pump is designed to cycle between a peak flow and zero net flow to approximate systole and diastole. Bench testing indicates that the pump can produce a blood flow of 5 l min^?1 with 50 mmHg pressure increase at 17 500 rpm. At zero net flow rate, the pump can maintain a diastole aortic pressure against 80 mmHg at the same rotating speed.     

World-smallest LVAD with 27 g weight, 21 mm OD and 5 l min-1 flow with 50 mmHg pressure increase

To investigate the feasibility of a long-term left ventricular assist device (LVAD) placed in the aortic valve annulus, an implantable aortic valve pump (21 mm outer diameter, weighing 27 g) was developed. The device consists of a central rotor and a stator. The rotor assembly incorporates driven magnets and an impeller. The stator assembly has a motor coil with an iron core and outflow guide vanes. The device is to be implanted identically to an aortic valve replacement, occupying no additional anatomic space. The pump delivers the blood directly from left ventricle to the aortic root, like a natural ventricle, therefore causing less physiologic disturbance to the natural circulation. Neither connecting conduits nor 'bypass' circuits are necessary. The pump is designed to cycle between a peak flow and zero net flow to approximate systole and diastole. Bench testing indicates that the pump can produce a blood flow of 5 l min(-1) with 50 mmHg pressure increase at 17,500 rpm. At zero net flow rate, the pump can maintain a diastole aortic pressure against 80 mmHg at the same rotating speed.