Kohei Ishii

Results of Animal Experiments With the Fourth Model of the Undulation Pump Total Artificial Heart

Animal experiments using a total artificial heart in a goat are not easy to perform. The fourth model of the undulation pump total artificial heart (UPTAH4), which was designed to perform a long-term physiological experiment including pulsatile and nonpulsatile TAH operations with a conductance- and arterial pressure-based control method named 1/R control, was implanted in 31 goats weighing 38.5 to 60.4 kg (average of 46.8 kg). The 1/R control is a physiological flow control method of TAH developed with a conductance (1/R: reciprocal of a resistance) parallel circuit model. The survival periods were from 0.1 to 153 days (average of 14.5 days). The causes of termination were postoperative bleeding in eight goats, respiratory failure in five goats, device failure in 14 goats, dissected aneurysm in two goats, and thrombus in one goat. The thrombus case was the longest surviving goat. The respiratory failure tended to occur when the extracorporeal circulation time was prolonged. Autotransfusion was effective for the prolongation of survival time. The left-right balance control and the suction control were performed successfully in all goats. The 1/R control was performed for a long time in five goats that survived for more than 1 month. With three goats that survived for 48, 52, and 53 days mainly with the pulsatile mode, the 1/R control was stable. With a goat that survived for 73 days, the nonpulsatile mode with the 1/R control could be tested for 3 weeks. With the longest surviving goat that was maintained mainly with the pulsatile mode, the 1/R control was unstable, possibly due to the mismatching of the response time of the control system between the computer and the body. However, liver and kidney functions were almost normal, and the total protein level recovered. Further study to stabilize the 1/R control in the UPTAH is necessary.

Animal Experiments of the Helical Flow Total Artificial Heart

Severe cardiac failure patients require a total artificial heart (TAH) to save life. To realize a TAH that can fit a body of small stature and has high performance, high durability, good anatomical fitting, good blood compatibility, and physiological control, we have been developing the helical flow TAH (HFTAH) with two helical flow pumps with hydrodynamic levitation impeller. Animal experiments of the HFTAH were conducted to perform in vivo studies. The HFTAH was implanted in 13 adult female goats weighing 45.0-64.0 kg. After surgery, neither anti-coagulant nor anti-platelet medication was given systemically. The HFTAH was usually driven with a quasi-pulsatile mode. The 1/R control or ΔP control was applied to control the circulation. The ΔP control is a new method using simplified equation of the 1/R control. The HFTAH could be implanted in all goats with good anatomical fitting. Two goats survived for a long time (100 and 68 days). Major causes of termination were device failure and surgical complications. In the device failure, trouble with hydrodynamic bearing was conspicuous. In the two long-term survived goats, experiments were terminated with bearing instability that was probably caused by the suction effect. In these goats, hemolysis occurred on postoperative day 88 and 44, which was considered to be relevant to the bearing trouble. Thrombus was found at the broken right bearing of the 100-day survived goat. However, antithrombogenicity of the pump is expected to be good unless bearing trouble occurs. In two long-term survived goats, the 1/R control or ΔP control worked appropriately to prevent the elevation of right atrial pressure. In both goats, hemodynamic parameters changed with the condition of the animals, liver and kidney functions remained almost normal except when recovering from surgery and during hemolysis, and total protein recovered 2 weeks after surgery. Although instability of the hydrodynamic bearing should be improved, performance of the HFTAH with physiological control could be demonstrated.

The helical flow total artificial heart: Implantation in goats

To realize a total artificial heart (TAH) with high performance, high durability, good anatomical fitting, and good blood compatibility, the helical flow TAH (HFTAH) has been developed with two helical flow pumps having hydrodynamic levitation impeller. The HFTAH was implanted in goats to investigate its anatomical fitting, blood compatibility, mechanical stability, control stability, and so on. The size of the HFTAH was designed to be 80 mm in diameter and 84 mm wide. The maximum output was 19 L/min against 100 mmHg of pressure head. Eight adult female goats weighting from 45 to 56.3 kg (average 49.7 kg) were used. Under the extracorporeal circulation, natural heart was removed at the atrioventricular groove and the HFTAH was implanted. The HFTAH was driven with a pulsatile mode. The 1/R control was applied when the right atrial pressure recovered. The HFTAH could be implanted with good anatomical fitting in all goats. Two goats survived for more than a week. One goat is ongoing. Other goats did not survive for more than two days with various reasons. In the goats that survived for more than a week, the hydrodynamic bearing was worn and broken, which indicated that the bearing touched to the shaft. The cause was supposed to be the influence of the sucking effect. The potential of the HFTAH could be demonstrated with this study. The stability of the hydrodynamic bearing in a living body, especially the influence of the sucking effect, was considered to be very important and a further study should be necessary.

Preliminary study of physiological control for the undulation pump ventricular assist device

The undulation pump ventricular assist device (UPVAD) is a small implantable ventricular assist device using an undulation pump. The UPVAD can produce pulsatile flow by changing the motor rotation speed of the UPVAD. Because the undulation pump is a volume displacement type pump, the inflow sucking occurs easily. The purpose of this study is to develop a suitable control method for the UPVAD. The UPVAD inflow cannula equipped with an implantable pressure sensor was inserted into the ventricle. Therefore, pressure variation that synchronized with the natural heartbeat and negative pressure spike caused the inflow sucking were observed. By changing the motor rotation speed that responded to the inflow pressure, the UPVAD could synchronize with the natural heartbeat and the UPVAD could generate a co-pulse assist flow. The inflow sucking could be released by reducing the motor rotation speed, if the inflow sucking was detected. The newly developed control method exhibited superior characteristics than existing ones due to high immunity against pressure sensor drift. The assist flow could be increased more than 15% and the inflow sucking occurrence could be decreased with this control method. The UPVAD could generate the suitable assist flow with the developed control method.

Hydrodynamic characteristics of the helical flow pump

The helical flow pump (HFP) was invented to be an ideal pump for developing the TAH and the helical flow TAH (HFTAH) using two HFPs has been developed. However, since the HFP is quite a new pump, hydrodynamic characteristics inside the pump are not clarified. To analyze hydrodynamic characteristics of the HFP, flow visualization study using the particle image velocimetry and computational fluid dynamics analysis were performed. The experimental and computational models were developed to simulate the left HFP of the HFTAH and distributions of flow velocity vectors, shear stress and pressure inside the pump were examined. In distribution of flow velocity vectors, the vortexes in the vane were observed, which indicated that the HFP has a novel and quite unique working principle in which centrifugal force rotates the fluid in the helical volutes and the fluid is transferred from the inlet to the outlet helical volutes according to the helical structure. In distribution of shear stress, the highest shear stress that was considered to be occurred by the shunt flow across the impeller was found around the entrance of the inlet helical volute. However, it was not so high to cause hemolysis. This shunt flow is thought to be improved by redesigning the inlet and outlet helical volutes. In distribution of pressure, negative pressure was found near the entrance of the inlet helical volute. However, it was not high. Negative pressure is thought to be reduced with an improvement in the design of the impeller or the vane shape.

Computational fluid dynamics analysis of the pump parameters in the helical flow pump

The helical flow pump (HFP) was invented to develop a total artificial heart at the University of Tokyo in 2005. The HFP consists of the multi-vane impeller involving rotor magnets, a motor stator and pump housing having double-helical volutes. To investigate the characteristics of the HFP, computational fluid dynamics analysis was performed. Validation of the computational model was performed with the data of the actual pump. A control computational model in which the vane area corresponded approximately to that of the actual pump was designed for the parametric study. The parametric study was performed varying the vane height, vane width and helical volute pitch. When the vane height was varied from 0.5 to 1.5 times that of the control computational model, the H–Q (pressure head vs. flow) and efficiency curves were translated in parallel with the vane height. When the vane height was two and three times that of the control computational model, the profile of these curves changed. From the results, the best proportion for the vane was considered to be a vane height between 1.5 and 2 times the vane width. The effect of vane width was not very strong compared to that of the vane height. A similar tendency in vane height was observed by varying the helical volute pitch. The best helical volute-pitch size is considered to be between 1.5 and 2 times the vane width. Although further study is necessary to determine the best values for these parameters, the characteristics of the pump parameters in the HFP could be approximately clarified.

Development of normal-suction boundary control method based on inflow cannula pressure waveform for the undulation pump ventricular assist device.

It is desirable to obtain the maximum assist without suction in ventricular assist devices (VADs). However, high driving power of a VAD may cause severe ventricle suction that can induce arrhythmia, hemolysis, and pump damage. In this report, an appropriate VAD driving level that maximizes the assist effect without severe systolic suction was explored. The target driving level was set at the boundary between low driving power without suction and high driving power with frequent suction. In the boundary range, intermittent mild suction may occur. Driving power was regulated by the suction occurrence. The normal-suction boundary control method was evaluated in a female goat implanted with an undulation pump ventricular assist device (UPVAD). The UPVAD was driven in a semipulsatile mode with heartbeat synchronization control. Systolic driving power was adjusted using a normal-suction boundary control method developed for this study. We confirmed that driving power could be maintained in the boundary range. Occurrences of suction were evaluated using the suction ratio. We defined this ratio as the number of suction occurrences divided by the number of heartbeats. The suction ratio decreased by 70% when the normal-suction boundary control method was used.

Pulsatile driving of the helical flow pump

The helical flow pump (HFP) is newly developed blood pump for total artificial heart (TAH). HFP can work with lower rotational speed than axial and centrifugal blood pump. It can be seen reasonable feature to generate pulsatile flow because high response performance can be realized. In this article, pulsatility of HFP was evaluated using mock circulation loop. Pulsatile flow was generated by modulating the rotational speed in various amplitude and heart rate. In the experiment, relationship between Pump flow, pump head, rotational speed amplitude, heart rate and power consumption is evaluated. As the result, complete pulsatile flow with mean flow rate of 5 L/min and mean pressure head of 100 mmHg can be obtained at ±500rpm with mean rotational speed of 1378 to 1398rpm in hart rate from 60 to 120. Flow profiles which are non-pulsatile, quasi-pulsatile or complete flow can be adjusted arbitrarily. Therefore, HFP has excellent pulsatility and control flexibility of flow profile.

Emergency life support system aiming preprimed oxygenator

Development have been achieved of a new blood pump for next generation Percutaneous Cardio-Pulmonary Support (PCPS) system and a novel surface coating method for silicone membrane hollow fiber by physical adsorption using a copolymer composed of a 2-Methacryloyloxyethyl phosphorylcholine (MPC) unit and a hydrophobic unit. The new blood pump, named the Troidal Convolution Pump (TCP), is based on the principle of a cascade pump and perfused 5 L/min and 350 mmHg at 2450 rpm. The novel copolymer composed of 30% MPC unit and 3-(methacryloyloxy) propyltris (trimethylsiloxy) silane (MPTSSi) unit (PMMSi30) was the most suitable molecular design on a silicone surface. The PMMSi30 coated surface adsorbed 7.2 % as much protein a non-coated surface adsorbed.

Development of an Implantable Small Observation Camera with Scaffold Chamber for Angiogenesis

A miniaturized video camera system integrated with a scaffold for blood vessels and tissue induction that is implantable into an animal body was developed. A small autofocus camera with USB interface was used as an imaging device. A close-up lens was added to reduce the focusing point. Also 45-degree reflecting prism was added for the compact system design. A fabric poly-glycolic acid sheet of 0.3 mm thickness was used as a scaffold. The observation area was about 8 x 6 mm and the tissue was induced from sideways. The camera was implanted into a muscular layer of a goat for more than 9 weeks to observe the skeletal muscle induction. Tissue induction to the scaffold started from third week and it took about 9 weeks to cover the entire area by the tissue, Vigorous angiogenesis was observed at the front region of tissue induction resulting dense distribution of capillary vessels and red blood cells. The density of capillary vessels reduced considerably behind the front region and arterioles and venules started to appear.