Impeller design for a miniaturized centrifugal blood pump.

The impeller design for a miniature centrifugal blood pump is an important consideration since the small diameter impeller requires higher rotational speed, which may cause more blood trauma compared to the larger diameter impeller. Three different impeller vanes (straight vanes with a height of 4 mm and 8 mm, and 8 mm curved vanes) of which the diameter was 35 mm were subjected to hydraulic performance and hemolysis tests in the same pump housing. Both straight vane impellers attained left ventricular assist condition (5 L/min against 100 mm Hg) at 2,900 rpm while the curved vane required 3,280 rpm. There was no significant hemolysis difference between the tall and short vanes. The curved impeller vanes did not exhibit sufficient hydraulic performance when compared to the straight vanes. The straight vane impellers, even with different heights, were incorporated into the same pump housings, and the vane heights did not drastically change the hydraulic performance or hemolysis.

Control system for an implantable rotary blood pump.

Rotary blood pumps can be used for long-term left ventricular assist devices. These pumps have several advantages over the conventional pulsatile pumps including smaller size, higher efficiency, and simple design and construction. However, one of the difficulties associated with the rotary blood pump is the proper control method to maintain an optimum flow rate in different physiological conditions. The rotary blood pump can be controlled by two methods. The first is to utilize the measured pump flow rate from its servo signal. The second is to detect and avoid abnormal pumping conditions such as; back flow and sudden increase in the pressure head. This abnormal situation typically occurs from excessive suction of blood when there is a functional or mechanical occlusion in the inflow cannula. The ultrasound flow meter is durable and reliable but it is difficult to continually monitor the blood flow rate of an implantable pump. Therefore, another method is needed instead of the continuous flow monitoring. One chronic calf having an LVAD was subjected for the development of this control system. This calf survived more than 6 months. Voltage, current, motor speed, heart rate and the pump flow rate were recorded and stored at 30-min intervals in a computer. Utilizing these parameters, attempts were made (1) to achieve indirect flow assessments and (2) to reveal abnormal operating parameters of the centrifugal pump (1). Indirect flow measurement, the predicted pump flow rate was calculated from these pump derived parameters (required power, motor speed and heart rate). The value of the coefficient of determination (R) between the measured and estimated pump flow rate was 0.796. (2) Abnormal operating indicator, there was an association between the required current and pump flow waves. The current was differentiated, and then calculated to the power of the differentiated current. The normal range of this value was 0.02+/-0.54. In abnormal conditions, this abnormal operating indicator increased 500 times. The predicted flow estimation method and abnormal operating indicator were available from intrinsic operating parameters of the pump and need no sensors. These two methods were simple, yet they are possibly effective and reliable servo control methods for a rotary blood pump.

Gyro pump wear and deformation analysis in vivo study: creep deformation.

The Gyro pump has a double pivot bearing system to support its impeller. In this study, the integrity of the bearing system was examined after ex vivo studies. The pumps were implanted into calves and evaluated for different periods as a paracorporeal left ventricular assist device (LVAD). One pump was subjected to a test of 30 days, 1 for 15 days, 4 for 14 days, 1 for 10 days, 1 for 7 days, 2 for 4 days, and 4 for 2 days. One additional pump was subjected to percutaneous cardiopulmonary support (PCPS) condition for 6 days (total pressure head 500 mm Hg with a pump flow rate of 3 L/min). The anticoagulation treatment consisted of a continuous administration of heparin to maintain an achieved clotting time (ACT) of 200-250 s during the LVAD study and 250-300 s during the PCPS study. After the experiment, the pumps were disassembled, and the wear and deformation of male and female bearings were analyzed. There were no dimensional changes on male bearings but there were on female bearings. Wear and deformation of the female bearings were calculated as follows: wear and deformation = (depth of female before pumping) - (depth after pumping). Thirteen assembled Gyro pumps were disassembled to measure the depth of the female bearings before pumping. There was no statistical relationship between the wear and deformation and the motor speed x driving period. From these results, the deformation was not due to wear but to the creep or elastic deformation. This study suggested that the double pivot bearing system of the Gyro pump is highly durable.

Antithrombogenicity evaluation of a centrifugal blood pump.

The Gyro C1E3 pump was developed not only for cardiopulmonary bypass but also as a short-term assist device. The main purpose of this study was to examine the correlation between the thrombus formation factor and the Gyro C1E3 pumps. Seven pumps were implanted into 3 calves and evaluated for different periods of duration as a paracorporeal left ventricular assist device (LVAD). One pump was subjected to percutaneous cardiopulmonary support condition (PCPS) (total pressure head 500 mm Hg with a pump flow rate of 3 L/min). The anticoagulation treatment consisted of a continuous administration of heparin to maintain an activated clotting time (ACT) of 200-250 during the LVAD study and 250-300 during the PCPS study. After the experiment, the pumps were disassembled and examined. In cases where there were any blood-derived deposits inside the pumps, the dry weight of these thrombi that adhered to the bearing area of the pump was measured. A multiple correlation was attempted to speculate possible thrombus formation. The estimated dry weight of thrombi was calculated from pump flow rate, pumping day, motor speed, and activated clotting time. This equation was estimated dry weight of thrombi = 1.140 x pump flow rate -0.001 motor speed + 1.652 pumping time -0.041 x ACT + 2.198 R2 = 0.944. This study suggested that there was a possibility to calculate the amount of adhered thrombus formation from pump flow rate, motor speed, pumping day, and ACT.

Particles released from the Gyro C1E3 during simulated extracorporeal circulation.

Evaluation of released particles from the blood pump during extracorporeal circulation is an important aspect because the particles may cause microembolism. The Gyro C1E3 is a centrifugal blood pump that has an impeller suspended by double pivot bearings inside the housing; therefore, it is important to evaluate the released particles. The C1E3 was driven for 14 days to simulate clinical left ventricular assist device (LVAD) and percutaneous cardiopulmonary support (PCPS). Also, a roller pump was driven for 2 days as a comparison. Released particles were weighed and examined by SEM. After 14 days of pumping, the particles from the C1E3 were 238.6 microg in an LVAD condition. The particles with the roller pump were 270.2 microg after only 2 days. Average particle sizes with the roller pump and C1E3 were 3.7 and 0.6 microm, respectively. These results suggest that the Gyro C1E3 substantially reduces the risk of microembolism from released particles.

Estimation of pump flow rate and abnormal condition of implantable rotary blood pumps during long-term in vivo study.

The control system for an implantable rotary blood pump is not clearly defined. A detection system is considered to be necessary for pump flow monitoring and abnormal conditions such as back flow or a sucking phenomenon where the septum or left ventricle wall is sucked into the cannula, etc. The ultrasound flowmeter is durable and reliable but the control system should not be totally dependent on the flowmeter. If the flowmeter breaks, the rotary blood pumps have no control mechanism. Therefore, the authors suggest controlling the pumps by an intrinsic parameter. One left ventricular assist device (LVAD) calf model was studied where the flow rate and waveform of the pump flow proved to identify the sucking phenomenon. Thus, the pump flow rate was calculated from the required power, motor speed, and heart rate. The value of the coefficient of determination (R2) between the measured and estimated pump flow rate was 0.796. To estimate this abnormal phenomenon, 2 methods were evaluated. One method was the total pressure head in which the pump flow rate and motor speed were estimated. During normal conditions the total pressure head is 79.5 +/- 7.0 mm Hg whereas in the abnormal condition, it is 180.0 +/- 2.8 mm Hg. There was a statistical difference (p < 0.01). Another method is using a current waveform. There is an association between the current and pump flow waves. The current was differentiated and squared to calculate the power of the differentiated current. The normal range of this value was 0.025 +/- 0.029; the abnormal condition was 11.25 +/- 15.13. There was a statistical difference (p < 0.01). The predicted flow estimation method and a sucking detection method were available from intrinsic parameters of the pump and need no sensors. These 2 methods are simple, yet effective and reliable control methods for a rotary blood pump.

Development of an antithrombogenic and antitraumatic blood pump: the Gyro C1E3.

The Gyro C1E3 is a centrifugal blood pump. Its antithrombogenic and antitraumatic blood features were demonstrated by prior studies. Based upon these studies, a mass production model of the C1E3 is becoming commercially available. Therefore, this feasibility study was conducted using the mass production models of the Gyro C1E3 for long-term cardiac assist in ex vivo animal experiments. Five healthy calves were used and 15 pump heads were applied for different time periods (Group 1, 30 days; Group 2, 14 days; Group 3, 10 and 7 days; Group 4, 4 days; and Group 5, 2 days). Activated clotting time (ACT) was kept at 200-250 sec. All five calves demonstrated neither abnormal signs nor abnormal blood examination data throughout the experiment. During necropsy, no thromboembolism was found in any downstream organs. Groups 1-4 showed thrombi inside the pump heads while two pumps in Group 5 had no thrombi formations. Bearing deformation or possible wear did not increase after 2 days of pumping. The C1E3 is capable of long-term assist circulation. However, after 2 days of pumping, careful observation is necessary since thrombi may occur inside the pump when ACT is controlled under 250 sec. During the weaning stage or low flow (under 2 L/min), over 250 sec of ACT is recommended to assure the safety of the patient.

Feasibility of a tiny Gyro centrifugal pump as an implantable ventricular assist device.

The Gyro pumps were developed for long-term circulatory support. The first generation Gyro pump (C1E3) achieved 1 month paracorporeal circulatory support in chronic animal experiments; the second generation (PI702) implantable ventricular assist device (VAD) was successful for over 6 months. The objective of the next generation Gyro pump is for use as a long-term totally implantable VAD and for pediatric circulatory support. This tiny Gyro pump (KP101) was fabricated with the same design concept as the other Gyro pumps. The possibility of an implantable VAD was determined after performance and hemolysis test results were compared to those of the other Gyro pumps. The pump housing and impeller were fabricated from polycarbonate with an impeller diameter of 35 mm. The diameter and height of the pump housings are 52.3 mm and 29.9 mm, respectively. At this time, a DC brushless motor drives the KP101, which is the same as that for the C1E3. The pump performance was measured in 37% glycerin water at 37 degrees C. Hemolysis tests were performed utilizing a compact mock loop filled with fresh bovine blood in a left ventricular assist device (LVAD) condition at 37 degrees C. The KP101 achieved the LVAD conditions of 5 L/min and 100 mm Hg at 2,900 rpm; generated 10 L/min against 100 mm Hg at 3,200 rpm; 3 L/min against 90 mm Hg at 2,600 rpm; and 2 L/min against 80 mm Hg at 2,400 rpm. In addition, the pump efficiency during this experiment was 12.5%. The other Gyro pumps. that is, the C1E3, PI601, and PI701, in an LVAD condition require 1,600, 2,000, and 2,000 rpm, respectively. The KP101 produced a normalized index of hemolysis (NIH) value of 0.005 g/100 L. With regard to the NIH, the other Gyro pumps, namely the C1E3, PI601, and PI701 demonstrated 0.0007, 0.0028, and 0.004 g/100 L, respectively. The KP101 produced an acceptable pressure flow curve for a VAD. The NIH value was higher than that of other Gyro pumps, but is in an acceptable range.

A new control method that estimates the backflow in a centrifugal pump.

The rotary blood pump will be widely used in the near future as an implantable left ventricular assist device (LVAD). However, one obstacle for the centrifugal pump is a control method that can maintain an optimum flow rate in a physiological condition. Thus, the object of this study is to develop this optimum control system for the centrifugal pump. If the heart function and pump efficiency are stable, the ratio of the systole current to the the diastole current (S/D) will be a fixed value. However, if the heart function and pump efficiency are unstable, S/D will not be a fixed value. This control system was investigated with a calf that was subjected to an ex vivo LVAD study. The LVAD was a Gyro C1E3 centrifugal pump. The pump flow rate was changed to 1.5, 3.5, 5.2, and 6.2 L/min. According to the changes of the pump flow rates, the S/D values were 1.01 +/- 0.01, 1.06 +/- 0.05, 1.03 +/- 0.01, and 1.03 +/- 0.01, respectively. There was no statistical difference among the 3 groups. In a separate experiment, the backflow condition S/D was 1.88 +/- 0.6, and the normal condition S/D was 1.35 +/- 0.5. There was a statistical difference between the 2 groups. The results of this study suggest that S/D is not influenced by the pump flow rate. However, the S/D was changed when the pump was in a backflow condition. This method will be useful in controlling a centrifugal pump requiring only electrical current information.

Evaluation of platelet adhesion and activation on materials for an implantable centrifugal blood pump.

A totally implantable centrifugal artificial heart has been developed in which a pivot bearing supported centrifugal pump is used as a blood pump. The following have been adopted as blood contacting materials in our pump: titanium alloy (Ti-6A1-4V) for the housing and impeller, alumina ceramic (Al2O3) for the male pivots, and ultrahigh molecular weight polyethylene (PE) for the female pivots. Greater antithrombogenicity is required for an implantable blood pump. To examine the thrombogenicity of these materials, we evaluated in vitro platelet adhesion and activation, which may play key roles in thrombogenesis on foreign surfaces. Ti-6A1-4V, Al2O3, and PE were compared with polycarbonate (PC), silicone carbide (SiC), and pure titanium (pTi). Platelet adhesion was assessed using monoclonal antibody (CD61) directed against glycoprotein IIIa. Platelet activation was evaluated by measuring P-selectin (GMP-140) released from irreversibly activated platelets. Each material with a surface area of 16.6 cm2 was incubated with 2.5 ml of plasma or 2.5 ml of heparinized fresh whole blood for 3 h at 37 degrees C. The optical density (OD) at a wavelength of 450 nm for CD61 was 0.93+/-0.35 in PC, 0.34+/-0.13 in PE, 0.27+/-0.13 in pTi, 0.26+/-0.01 in Al2O3, 0.21+/-0.04 in SiC, and 0.12+/-0.12 in Ti-6A1-4V. The GMP-140 levels of the tested materials were not significantly different from the control value (45.9+/-7.2 ng/ml). These results indicate that Al2O3, PE, and Ti-6A1-4V, which are incorporated into our implantable centrifugal pump, have satisfactory antithrombogenic properties in terms of platelet adhesion. However, platelet activation by any material was not observed under the static condition in this study.

Protein adsorption onto ceramic surfaces.

Ceramics seldom have been used as blood-contacting materials. However, alumina ceramic (Al2O3) and polyethylene are incorporated into the pivot bearings of the Gyro centrifugal blood pump. This material combination was chosen based on the high durability of the materials. Due to the stagnant flow that often occurs in a continuous flow condition inside a centrifugal pump, pivot bearing system is extremely critical. To evaluate the thombogenicity of pivot bearings in the Gyro pump, this study sought to investigate protein adsorption, particularly albumin, IgG, fibrinogen, and fibronectin onto ceramic surfaces. Al2O3 and silicon carbide ceramic (SiC) were compared with polyethylene (PE) and polyvinylchloride (PVC). Bicinchoninic acid (BCA) protein assay revealed that the amount of adsorbed proteins onto Al2O3 and SiC was significantly less than that on PVC. The sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) indicated that numerous proteins adsorbed onto PVC compared to PE, Al2O3, and SiC. Identification of adsorbed proteins by Western immunoblotting revealed that the adsorption of albumin was similar on all four materials tested. Western immunoblotting also indicated lesser amounts of IgG, fibrinogen, and fibronectin on Al2O3 and SiC than on PE and PVC. In conclusion, ceramics (Al2O3 and SiC) are expected to be thromboresistant from the viewpoint of protein adsorption.

Recent advances in the gyro centrifugal ventricular assist device.

The gyro pump was developed as an intermediate-term assist pump (C1E3) as well as a long-term centrifugal ventricular assist device (VAD). The antithrombogenic design concept of this pump was confirmed throughout three 1 month ex vivo studies. The normalized index of hemolysis (NIH) of this gyro C1E3 model was lower than that of the BP-80. In the next step, a miniaturized centrifugal blood pump (The Gyro permanently implantable model PI-601) has been developed for use as a permanently implantable device after design optimization. A special motor design of the magnet circuit was utilized in this system in collaboration with the University of Vienna. The priming volume of this pump is 20 ml. The overall size of the pump actuator package is 53 mm in height, 65 mm in diameter, 145 ml of displacement volume, and 305 g in weight. This pump can provide 5 L/min against 120 mm Hg total pressure head at 2,000 rpm. The NIH value of this pump was comparable to that of the BP-80. The gyro PI-601 model is suitable for a VAD. The expected life from the endurance study is approximately 8 years. The evolution from C1E3 to the PI-601 converts this pump to a totally implantable centrifugal pump. Recent technologic advances in continuous flow devices are likely to realize a miniaturized and economical totally implantable VAD.

Biocompatibility of alumina ceramic and polyethylene as materials for pivot bearings of a centrifugal blood pump.

The double pivot bearings in the Gyro C1E3 centrifugal blood pump incorporate a high-purity alumina (Al2O3) ceramic and an ultra-high-molecular-weight polyethylene (UHMWPE). This centrifugal pump has been developed as a completely sealless pump for long-term usage. The combination of Al2O3 and UHMWPE are the materials of choice for the acetabular bearing in artificial joints, which have proven to be clinically reliable for over 10 years. Previous studies have examined the biocompatibility of Al2O3 and UHMWPE as bulky implant materials. The present study investigated this material as a blood-contacting material using a standard assessment in vitro and in vivo analysis. The examined items were systemic toxicity, sensitization (guinea pig maximization test), cytotoxicity (elution test), mutagenicity (Ames test), direct contact hemolysis, and thrombogenicity. The studies were performed according to the United States Pharmacopoeia and published previous studies. The samples of both Al2O3 and UHMWPE demonstrated no differences from the negative controls in all tests. These findings indicate that both Al2O3 and UHMWPE are biocompatible materials for double-pivot bearings in the centrifugal blood pump.

Mapping of pump efficiency on the pressure-flow curve of a centrifugal blood pump.

Because pump efficiency is closely related to heat generation and blood trauma in a centrifugal blood pump, it is quite important to study pump efficiencies in a variety of conditions. In the present study, pump efficiencies were mapped on the pressure head-flow rate curves of 4 different pumps; BioMedicus BioPump (BP-80), Nikkiso (NK), Gyro C1E3, and Gyro PI601 (diameter of the impeller, NK: 50 mm, C1Ee3: 65 mm, and PI601: 50 mm). The mapping of pump efficiency revealed the following findings. First, the cone type (BP-80) has less pump efficiency than the impeller type (NK and C1E3); second, the miniaturization of the C1E3 to the PI601 has resulted in an increase in pump efficiency; and third, the diameter of the impeller may contribute to the pump efficiency of an im peller type pump. The mapping of the pump efficiency, as demonstrated in this study, is useful for the analysis of hydraulic pump performance in a wide range of clinically applied conditions.

Hemolytic effect of surface roughness of an impeller in a centrifugal blood pump.

The present study investigates how the surface roughness of an impeller affects hemolysis in the pivot bearing supported Gyro C1E3 pump. This study focuses on particular areas of the impeller surface in the impeller type centrifugal pump. Seven Gyro C1E3 pumps were prepared with smooth surface housings and different impeller parts with different surface roughnesses. The vanes, top side, and backside of the impeller were independently subjected to vapor polishing, fine sand blasting, or coarse sand blasting to produce three different grades of surface roughness. These surfaces were then examined by a surface profile instrument. Using these pumps with different impellers, in vitro hemolysis tests were performed simulating cardiopulmonary bypass (5 L/min, 350 mm Hg). The findings of this study conclusively proved that surface roughness of the back side of the impeller has the greatest effect on hemolysis, followed by the top side and then the vanes. The following are reasons for these findings. First, the shear rate may be greater on the back side than on the top side because of the smaller gap between the back and the housing and the greater relative speed against the impeller. Second, the fluid beneath the impeller may have a longer exposure time because there is little chance for the fluid to mix beneath the impeller. Third, the shear rate may be greater on the top side of the impeller than on the vanes because a vortex formation occurs behind the vanes.

Reconsideration of total erythrocyte destruction phenomenon.

During a particular long-term in vitro hemolysis test, the plasma free hemoglobin suddenly increased even though the hemolysis level had risen linearly for the previous several hours. This phenomenon was dubbed the total destruction of erythrocytes (TDE) phenomenon, and it was hypothesized that this was the result of the accumulation of sublethal damage to erythrocytes. It was suggested that the TDE might demonstrate the hemolytic characteristics of a pump more sensitively than a conventional hemolysis test. However, the previous report did not consider the effects of temperature or contamination. To study these effects, 3 long-term hemolysis tests were concluded under the following conditions. For Study 1 blood temperature was maintained at 27 degrees C (n = 2); for Study 2, at 37 degrees C (n = 4); and for Study 3, at 37 degrees C with gentamicin (n = 4). The BioMedicus and Nikkiso pumps were used as they were in our previous report. Gas sterilization of all circuits and pumps preceded experimentation. In Studies 1 and 3, hemolysis increased linearly for 29 h. However, in Study 2 a sudden increase of hemolysis occurred for both pumps. Possible causes of this were the dramatic changes in environmental factors such as severe acidosis, high O2 and glucose consumption, and CO2 accumulation. In contrast, neither Study 1 nor Study 3 showed a sudden increase in hemolysis. The plasma free hemoglobin increased linearly in both groups until 29 h of pumping. The environmental changes resulting from contamination were considered to be the cause of the sudden increase in hemolysis. In conclusion, the TDE did not reflect mechanical blood cell damage, but rather different environment situations. Hemolysis increased linearly up to 29 h in either 27 degrees C or germ-free conditions.

Comparison of the Gyro C1E3 and BioMedicus centrifugal pump performances during cardiopulmonary bypass.

The compact eccentric inlet port (C1E3) centrifugal blood pump was developed as a cardiopulmonary bypass (CPB) pump. The C1E3 pump incorporated a sealless design with a blood stagnation free structure. The pump impeller was magnetically coupled to the driver magnet in a sealless manner. To develop an atraumatic and antithrombogenic centrifugal pump without a shaft seal junction, a double pivot bearing system was introduced. Recently, a mass production model of the C1E3 was fabricated and evaluated. The ratio of the normalized index of hemolysis (NIH) of the C1E3 was 0.007 g/ 100 L, in comparison to the NIH of the BP-80, 0.018 g/ 100 L, each in a CPB condition of 5 L/min against 325 mm Hg. Both pumps were compared in identical in vitro circuits. To further evaluate the pumps during cardiopulmonary bypass for reliability and function, 6 h of CPB was performed on each of 8 bovines using either the C1E3 or BP-80 centrifugal pump. The BP-80 and C1E3 provided pump flows of 50-60 ml/kg/min without incident. The hemodynamics were stable, and the hematology and biochemistry data were within normal ranges. There were no statistically significant differences between the 2 groups. Concerning the plasma free hemoglobin values, a mass production model of the C1E3 pump had the same hemolysis levels as the BP-80. Our preliminary studies reveal that the C1E3 pump is reliable. Also, the C1E3 will satisfy clinical requirements as a cardiopulmonary bypass pump.

The hemolysis test of the Gyro C1E3 pump in pulsatile mode.

While a centrifugal pump is generally used for nonpulsatile blood flow, it can also produce a pulsatile flow by alternating the impeller rotational speed (rpm) periodically. However, there is a concern that this centrifugal pump pulsatile mode may induce added hemolysis as a result of the repeated acceleration and deceleration of rpm. Thus, a hemolysis study of the pulsatile modes of the Gyro C1E3 centrifugal pump (Gyro-P) was conducted. The results were then compared with the nonpulsatile mode of the same pump (Gyro-N) and the nonpulsatile BioMedicus BP-80 (Bio-N) pump. Three different conditions were simulated: left ventricular assist device (LVAD), cardiopulmonary bypass (CPB), and percutaneous cardiopulmonary support (PCPS). The beating rate of the Gyro-P was set at 40 bpm, with repetition of two different impeller speed (the lower being 70% of the higher speed). The 2 impeller speeds were set to obtain the same average flow as that of the nonpulsatile mode. The hemolysis results of the Gyro-P were comparable to or better than those of Bio-N, and no excessive hemolysis was observed, compared to the Gyro-N. In conclusion, The Gyro-P had an excellent hemolytic characteristic and generated no excessive hemolysis in most clinical usage conditions. With the concern of hemolysis eliminated, this pulsatile mode may have various possible mode advantages.

In vitro thrombogenic evaluation of centrifugal pumps.

One of the major considerations in the development of a circulatory assist device is its antithrombogenecity. Although the precise evaluation should be accomplished by in vivo tests, these tests are costly and require a relatively long period. In this study, we established a simple in vitro test and assessed feasibility using 2 clinically available centrifugal pumps, the BioMedicus and Nikkiso pumps. Two identical mock loops were fabricated, and fresh heparinized human blood (activated clotting time of 150-250 s) was circulated at 5 L/min against a total pressure head of 100 mm Hg. After 3 h of pumping, only the BioMedicus pumps had thrombi while the Nikkiso pumps were thrombus free. Following 6 h of pumping, thrombi were observed in both pumps. Clotting patterns and locations were reproducible in each pump and similar to the results of clinical or ex vivo studies. This simple in vitro test was considered to be feasible as a pilot study, particularly to predict thrombogenic sites.

Hemolytic effects of surface roughness of a pump housing in a centrifugal blood pump.

The surface roughness of artificial blood contacting devices is an important surface property that is closely related to blood cell trauma. The present study investigated the effect of the surface roughness of a pump housing on hemolysis in an impeller-type centrifugal blood pump, a pivot bearing supported Gyro C1E3 pump. The purpose of the study was to determine which part of a housing has the greatest surface roughness effect on hemolysis in a centrifugal pump. Seven Gyro C1E3 pumps were prepared, each with a smooth surface impeller and a housing with differing areas of altered surface roughness. Both top and bottom housings were divided into half subregions, each with the same area. Seven test pumps were produced by subjecting various subregions of the housings to vapor polishing and sandblasting. The treated surfaces were then examined by a surface profile instrument. Using these 7 pumps with different areas of altered housing roughness, in vitro hemolysis tests were performed simulating cardiopulmonary bypass (5 L/min, 350 mm Hg). The results of this study are as follows. First, the surface roughness of the top housing had a greater effect on hemolysis than that of the bottom housing. Second, on the surface of the top housing, the surface roughness of the outer half area had a greater effect on hemolysis than that of the inner half area. Third, on the surface of the bottom housing, the surface roughness of the inner half area had a greater effect on hemolysis than that of the outer half area. These findings concur with previous studies of flow patterns in pumps. Thus, it is expected that the method in this study, comparative in vitro hemolysis tests of the pumps with surfaces of the same roughness but different locations, can be used to detect the high shear area inside a pump.