Development of a microcontroller-based automatic control system for the electrohydraulic total artificial heart.

An automatic physiological control system for the actively filled, alternately pumped ventricles of the volumetrically coupled, electrohydraulic total artificial heart (EHTAH) was developed for long-term use. The automatic control system must ensure that the device: 1) maintains a physiological response of cardiac output, 2) compensates for an nonphysiological condition, and 3) is stable, reliable, and operates at a high power efficiency. The developed automatic control system met these requirements both in vitro, in week-long continuous mock circulation tests, and in vivo, in acute open-chested animals (calves). Satisfactory results were also obtained in a series of chronic animal experiments, including 21 days of continuous operation of the fully automatic control mode, and 138 days of operation in a manual mode, in a 159-day calf implant.

Adaptive full-ejection point detection for automatic control of the electrohydraulic total artificial heart.

An adaptive algorithm to detect full ejection points was developed for the automatic control of the electrohydraulic total artificial heart (EHTAH). To automatically control the EHTAH, systole in each ventricle is completed when a hyper-pressurization spike is detected on the oil side of that ventricle. A problem associated with the method of determining ventricular full ejection under variable afterload pressure is that a fixed level comparison method could fail to provide a normal full ejection condition. To increase system stability, safety, and efficiency, there is a definite need to vary the full ejection trigger level with changes in afterload. In this study, the full ejection trigger level for the current beat was changed based on the estimated afterload derived from the previous beat's pressure waveform. To increase this estimate's accuracy, an adaptive averaging window was used to determine which part of the previous pressure waveform would be used for the afterload estimation. With this enhanced control scheme, mock circulation tests with the EHTAH device demonstrated that estimated afterload tracked actual afterload. This scheme was also used successfully to control the EHTAH device implanted in three chronic calves.

Anaerobic threshold in total artificial heart animals.

The anaerobic threshold represents an objective measure of functional capacity and is useful in assessment of pulmonary and cardiovascular dysfunction. This study determined the anaerobic threshold in total artificial heart animals and evaluated the performance of the total artificial heart system. Five animals with total artificial hearts were put under incremental exercise testing after exercise training. The intensity of exercise ranged from 2.0 to 4.5 km/hr, with an increment of 0.5 km/hr every 3 min. The anaerobic threshold was 6.72 +/- 0.84 ml/kg/min as detected by the lactate method, and 6.48 +/- 0.79 by the CO2 method. The value of the anaerobic threshold in total artificial heart animals implies that the performance capacity of a total artificial heart is not sufficient to meet the oxygen requirements of vigorously exercising skeletal muscle. The protocol does not allow for driving parameter changes during exercise, and this situation, combined with the manual mode of the control system used, was inadequate to allow the total artificial heart animals to exercise more vigorously. Using an automatic control mode might be helpful, as well as considering the relationship between indices of oxygen metabolism, such as oxygen delivery, oxygen consumption, and oxygen extraction rate, in the control algorithms in total artificial heart control systems.

Oxygen metabolism in animals with total artificial hearts.

The relationship between indices of oxygen metabolism has been widely used in clinical practice to evaluate the adequacy of tissue perfusion, to predict the outcome of the critically ill patient, and to evaluate the effectiveness of therapies. This study quantitated and correlated the relationship between oxygen delivery (DO2), oxygen consumption (VO2), and oxygen extraction rate (EO2) in 14 animals with total artificial hearts (TAH) to investigate the oxygen metabolism in animals with TAH during different physiologic and pathologic conditions. These 14 animals were subdivided into healthy, critical, and exercise groups. There was a physiologic dependence of DO2 to VO2 in animals in the healthy and exercise groups, whereas a pathologic dependence of VO2 to DO2 appeared to occur in animals in the critical group. Reduced or inadequate VO2 leads to organ dysfunction, shock syndrome, multiple organ failure, and finally, mortality. Providing a higher level of DO2 by restoring circulating blood volume, increasing cardiac output, raising hematocrit levels, and improving pulmonary function to achieve a higher level of oxygen extract efficiency and oxygen consumption in animals with TAH that are in a critical condition might be helpful for the treatment of complications and result in decreasing mortality. Using the relationship between indices of oxygen metabolism as a physiologic modifier for TAH control algorithms also might improve the physiologic performance and quality of life of TAH recipients.

Development of a totally implantable artificial heart.

The first generation of an integrated, totally implantable electrohydraulic total artificial heart was designed for long-term cardiac replacement. The system consists of an elliptical blood pump with an interatrial shunt, Medtronic-Hall 27 mm and 25 mm inflow and outflow valves, respectively, an energy converter consisting of an axial-flow, hydraulic pump driven by a brushless DC motor, and an electronics system with transcutaneous energy transmission and telemetry. Energy is supplied by internal nickel-cadmium rechargeable batteries that supply power for 20 min and external silver-zinc batteries that are designed to supply energy to run the system for 5 hr. The blood pump consists of a single layer diaphragm cast from Biolon, with joined right and left ventricles sharing a common base. The dynamic stroke volume is 84 ml, and maximum cardiac output is 9.2 L/min at a heart rate of 110 beats/min on the mock circulation. A 4.3 mm diameter interatrial shunt is used to balance the volumetrically coupled ventricles. The energy converter pumps hydraulic fluid alternately between ventricles, with controlled, active filling in one ventricle during the systolic phase of the other ventricle. Internal or external controllers adjust the heart rate and motor speed to maintain normal atrial filling pressures and full stroke. Electromagnetic induction is used to transfer energy through the skin and a bidirectional infrared data link incorporated within the transcutaneous energy transmission coils is used to transmit information. The entire system is being assembled and refined for long-term animal implant studies.

Implantable centrifugal pump with hybrid magnetic bearings.

Test methods and results of in vitro assessment of a centrifugal pump with a magnetically suspended impeller are provided. In vitro blood tests have been completed with a resulting normalized milligram index of hemolysis (NmIH) of 12.4 +/- 4.1, indicating that hemolysis is not a problem. Hydraulic characterization of the system with water has shown that a nominal pumping condition of 6 L/min at 100 mmHg was met at 2,200 rpm. Maximum clinically usable cardiac output is predicted be 10 L/min. The magnetic bearing supported impeller did not contact the housing and was shown to be stable under a variety of pumping conditions. The driving motor efficiency is 75% at the nominal condition. Finally, a description of the clinical version of the pump under development is provided.

Development of a prototype magnetically suspended rotor ventricular assist device.

A continuous flow centrifugal blood pump with magnetically suspended impeller has been designed, constructed, and tested. The system can be functionally divided into three subsystem designs: 1) centrifugal pump and flow paths, 2) magnetic bearings, and 3) brushless DC motor. The centrifugal pump is a Francis vane type design with a designed operating point of 6 L/min flow and 100 mmHg pressure rise at 2,300 RPM. Peak hydraulic efficiency is over 50%. The magnetic bearing system is an all active design with five axes of control. Rotor position sensors were developed as part of the system to provide feedback to a proportional-integral-derivative controller. The motor is a sensorless brushless DC motor. Back electromotive force voltage generated by the motor is used to provide commutation for the motor. No slots are employed in the motor design in order to reduce the radial force that the bearings must generate. Tests pumping blood in vitro were very encouraging; an index of hemolysis of 0.0086 +/- 0.0012 was measured. Further design refinement is needed to reduce power dissipation and size of the device. The concept of using magnetic bearings in a blood pump shows promise in a long-term implantable blood pump.

Pulsatile operation of a centrifugal ventricular assist device with magnetic bearings.

A prototype bench top model of a continuous flow ventricular assist device using an impeller suspended by magnetic bearings has been developed. Generation of a pulsatile pressure was studied using both a computer model and in vitro loop tests of the prototype. The motivation for developing a computer model for a blood pump in the natural circulation is two-fold. First, it allows simulation of the pump under a large variety of operating conditions. Second, it provides insight into what parameters of the system design are important for achieving a specific result. For example, in one case, an aortic pressure of 118/87 mmHg was generated by varying the speed from 2,000 to 2,600 rpm. The computer model was verified by coupling the centrifugal pump prototype to a mock circulatory system. The results of the model were verified by generating an aortic pressure of 113/78 mmHg while varying the speed from 2,000 to 2,600 rpm. These experiments have shown that it is possible to generate pulsatile pressure similar to that of native physiology using a centrifugal left ventricular assist device. Further tests will be required to quantify the effects on hemolysis.

Noninvasive diagnosis of mechanical failure of the implanted total artificial heart using neural network analysis of acoustic signals.

Acoustic signal measurement has been proposed as a noninvasive method of detecting mechanical failure of the implanted total artificial heart. However, differences in acoustic spectra obtained from undamaged and damaged devices may be difficult to distinguish using standard techniques, such as visual inspection or statistical analysis. A new technique, artificial neural network analysis, which has been used successfully on other problems of pattern recognition and classification, was applied to improve the detectability of the acoustic method. Acoustic signals were measured using two different devices in one damaged and one undamaged electrohydraulic total artificial heart, both in a mock circulation set-up and in animal experiments where they were implanted in eight post mortem sheep and the acoustic signal measured using a microphone placed at the skin surface. Spectra of the acoustic waveforms were calculated by discrete Fourier transformation and 400 values (representing the log magnitude in each 2.5 Hz band of the spectrum between 0 and 1 kHz) and used as input to the neural network. A three layer backpropagation neural network containing 400 input nodes, 20 intermediate nodes, and one output node was able to forms. The trained neural network then perfectly distinguished damaged waveforms from undamaged ones, with good separability. Because the neural network's output can take on a value between two extremes denoting damaged and undamaged states, it is possible to detect any progressive failure at relatively earlier stages. With multiple output node configuration, it could also classify the different types of damage using single acoustic signal waveforms.

In vitro characterization of a magnetically suspended continuous flow ventricular assist device.

A magnetically suspended continuous flow ventricular assist device using magnetic bearings was developed aiming at an implantable ventricular assist device. The main advantage of this device includes no mechanical wear and minimal chance of blood trauma such, as thrombosis and hemolysis, because there is no mechanical contact between the stationary and rotating parts. The total system consists of two subsystems: the centrifugal pump and the magnetic bearing. The centrifugal pump is comprised of a 4 vane logarithmic spiral radial flow impeller and a brushless DC motor with slotless stator, driven by the back emf commutation scheme. Two radial and one thrust magnetic bearing that dynamically controls the position of the rotor in a radial and axial direction, respectively, contains magnetic coils, the rotor's position sensors, and feedback electronic control system. The magnetic bearing system was able to successfully suspend a 365.5g rotating part in space and sustain it for up to 5000 rpm of rotation. Average force-current square factor of the magnetic bearing was measured as 0.48 and 0.44 (kg-f/Amp2) for radial and thrust bearing, respectively. The integrated system demonstrated adequate performance in mock circulation tests by providing a 6 L/min flow rate against 100 mmHg differential pressure at 2300 rpm. Based on these in vitro performance test results, long-term clinical application of the magnetically suspended continuous flow ventricular assist device is very promising after system optimization with a hybrid system using both active (electromagnet) and passive (permanent magnets) magnet bearings.

A magnetic bearing system for a continuous ventricular assist device.

This article describes a prototype continuous flow ventricular assist device (CFVAD3) supported in magnetic bearings. The VAD is a small centrifugal four bladed pump. The pump's geometry is explained. The CFVAD3 is the first compact VAD completely supported in magnetic bearings. The magnetic bearings are composed of an inlet side actuator divided into eight pole sets, and an outlet side actuator, also divided into eight pole sets. The pump operating performance was tested and found to be within the design flow rate of up to 9 L/min, and head up to 170 mm Hg for human circulatory support. Magnetic bearing operation out of center positions under various operating orientations were measured and found to be < 1/6 of the bearing clearance, well within specifications. The expected magnetic bearing power loss has been calculated at approximately 6.5 watts.

In vivo long-term evaluation of the Utah electrohydraulic total artificial heart.

An electrohydraulic total artificial heart (EHTAH) has been developed and evaluated by long-term in vivo studies. The EHTAH is composed of blood pumps with an interatrial shunt (IAS), an energy converter, and electronics. The EHTAH with external electronics was implanted in four calves weighing from 81-90 kg. Two animals died on the 1st and 5th post operative days, the third animal survived for 32 days, and the fourth for 159 days. The IAS was free of thrombus at autopsy in all animals. The longest surviving animal increased in size from a pre operative weight of 81 kg to 134 kg on day 144. Cardiac output ranged from 9.3 to 10.5 L/min, whereas right and left atrial pressures increased with the calf's growth from 4-10 to 16-20 mmHg and from 8-14 to 18-22 mmHg, respectively. The animal favorably tolerated up to 3.4 km/hr of treadmill exercise, both hemodynamically and metabolically. The elevation of atrial pressures during treadmill exercise was significantly alleviated by employing an automatic control mode. It is concluded that the device has the potential to be a totally implantable system for permanent use.

A blood pump with an interatrial shunt for use as an electrohydraulic total artificial heart.

A recently designed blood pump subsystem for the completely implantable electrohydraulic total artificial heart (EHTAH) has been developed and is under evaluation. The subsystem consists of joined left and right ventricles, atrial cuffs with an interatrial shunt (IAS), and two outflow grafts. The ventricles were developed to fit within the pericardial space based on the results of anatomic fit trials. An optimized configuration for animal use, which was adaptable for human use with minimal modification, was identified. The core dimensions of the ventricles with an energy converter are approximately 10 x 11 x 7 cm. Maximum output and stroke volume are 9.2 L/min and 81 ml, respectively. The IAS is used to balance the volumetrically coupled EHTAH, and is made by forming an orifice in the common septum of the left and right atrial cuffs. Performance and durability of the IAS were examined in animal experiments for up to 9 days. The diameter of the IAS was 3.4-5.5 mm, and the left-right atrial pressure difference ranged from 2 to 10 mmHg, with 0.57-1.48 L/min of theoretically calculated shunt flow. No evidence of thrombus formation was found in or around the IAS at autopsy. The entire EHTAH system with a new blood pump is being assembled for long-term animal studies.

Right-left ventricular output balance in the totally implantable artificial heart.

Pneumatically powered artificial hearts readily accommodated the higher net stroke volumes by the right ventricle than from the left ventricle. We published that this discrepancy was approximately 8% of the left ventricular cardiac output. A variety of methods have been used to achieve balance between the right and left atrial pressures. Relatively large volume-displacement chambers (VDC) present potential problems, but do provide balance. The VDC in volumetrically coupled right-left stroke volumes was eliminated by using a small-diameter interatrial shunt (IAS). Preliminary studies demonstrated excellent balance in contracted and expanded blood volume (preload) and by hypotension and hypertension created with vasoactive drugs (afterload). At a mean aortic pressure of 120 mmHg, heart rate of 120 BPM, cardiac output of 8 L/minute and right atrial pressure of 13 mmHg, the peak IAS flow was 3.2 ml/beat in a right to left direction and 8.0 ml/beat in a left to right direction. The net left to right flow was 4.8 ml/beat. Over a wide range of preload (2 to 20 mmHg) and afterload (45 to 180 mmHg), the left atrial pressure was routinely 5 mm Hg more than the right atrial pressure. Elimination of the VDC reduces the number of components, volume, and weight of the totally implantable artificial heart. The IAS offers a simple solution to a very complex problem and provides a device that is simpler to implant and is possible to explant.

Determination of atrial shunt size needed to balance an electrohydraulic total artificial heart.

An interatrial shunt (IAS) was developed and tested to balance a volumetrically coupled electrohydraulic total artificial heart (EHTAH). The IAS diameter was determined theoretically with the equation: [formula: see text] where D = orifice diameter, delta P = pressure drop across the orifice, rho = fluid density, Q = volume flow through the orifice, and C = empiric constant. The calculated orifice diameter was 2.9-5.3 mm to keep a 3-5 mmHg pressure difference between left and right atria, with 0.42-1.05 L/min of expected flow through the IAS. Three different sizes (3.4, 5.4, and 5.5 mm) of IAS were implanted and tested in four calves. The in vivo left-right pressure difference ranged from 3 to 8 mmHg, with 0.57-1.38 L/min of theoretically calculated flow through the IAS. At autopsy, none of the IAS showed macroscopic thrombus formation.

In vitro analysis of an atrial shunt in balancing an electrohydraulic total artificial heart.

In vitro tests were performed to evaluate the use of an interatrial shunt in balancing a dual energy converter, actively filled, volumetrically coupled, electrohydraulic total artificial heart. The in vitro atrial shunt was comprised of a 8 mm (PTFE) Teflon graft placed between the left and the right atrium. Other features under study were 1) cardiac output (CO) response to preload, 2) CO relationship to mean aortic pressure, and 3) balance of ventricular outputs. The tests were performed by varying the right filling pressure and monitoring ventricular output and inflow/outflow pressures. Effects of changes in afterload were simulated by varying the (AoP) pressure from 80 mmHg to 120 mmHg, and the (PAP) pressure from 15 mmHg to 40 mmHg. The test results indicated a rise in CO from 4 L/min to 9 L/min, with a change in mean right atrial pressure from 0 mmHg to 12 mmHg. No significant difference in CO was found as afterload pressures were varied. The interatrial shunt (IAS) was effective in establishing ventricular balance over a wide range of preload and afterload conditions, and a mean positive flow from left to right was maintained in the atrial shunt, even at conditions simulating an extreme left-right imbalance.

Development of an automatic control algorithm for the electrohydraulic total artificial heart without transducers.

To achieve a reliable and simple implantable total artificial heart, the number of implanted transducers providing the physiologic information required for automatic control should be minimized. To address this need, a new automatic control algorithm, based on a transducerless electrohydraulic total artificial heart (EHTAH) system, is proposed. The current EHTAH physiologic control algorithm relies on two implanted pressure transducers. Without these transducers, the information required for automatic control must be extracted from the running motor's parameters. These parameters correlated with the differential pressure across the axial flow pump used to actuate the EHTAH. Changes in this differential were chosen as a cue for cardiac output control. This algorithm can be viewed as depending upon systemic vascular resistance determined by subtracting mean right atrial pressure (RAP) from mean aortic pressure (AoP) and dividing the result by total cardiac output (CO). The difference between mean AoP and mean RAP was confirmed to correlate with the differential hydraulic pressure across the energy converter during the left systolic phase. As an interim configuration, a single differential pressure transducer measuring the differential hydraulic pressure across the energy converter was tested on a Donovan mock circulation system. The resultant CO response shows good sensitivity according to changes in both preload and afterload.

Characterization of a magnetic bearing system and fluid properties for a continuous flow ventricular assist device.

This article presents the performance test results of the CFVAD3 continuous flow blood pump in an artificial human circulation system. The CFVAD3 utilizes magnetic bearings that support a thin pancake impeller, the shape of which allows for a very compact pump whose total axial length is less than 5 cm with a radial length of about 10 cm. This gives a total volume of about 275 cc. The impeller itself has 4 vanes with a designed operating point of 6 L/min at 100 mm Hg of differential pressure and 2,000 rpm. The advantages of magnetic bearings, such as large clearance spaces and no mechanical wear, are elaborated upon. Furthermore, bearing model parameters such as load capacity and current gains are described. These parameters in conjunction with the operating conditions during testing are then used to estimate the fluid forces, stiffness, and damping properties while pumping. Knowledge of these parameters is desirable because of their effects on pump behavior. In addition, a better plant model will allow more robust control algorithms to be devised that can boost pump performance and reliability.