Effect of the diastolic and systolic duration on valve cavitation in a pediatric pulsatile ventricular assist device

Minimization of cavitation is of high importance in the design of pulsatile ventricular assist devices because cavitation can cause blood and valve surface damage. Cavitation is associated with valve closure and has been previously correlated to high dP/dt, high valve closing velocity, and decreased pump filling. In this study, the effects of diastolic and systolic duration on the inlet and outlet valve cavitation were investigated. A low volume (280 ml) mock circulatory loop filled with room-temperature saline was used. A high-fidelity hydrophone was mounted into the inlet valve connector approximately 0.5 cm upstream from the inlet valve to quantify inlet valve cavitation. The inlet valve connector and hydrophone were placed symmetrically on the outlet side when measuring outlet valve cavitation. The RMS intensity of a 6-millisecond window pressure trace, bandpass filtered from 50 to 500 kHz, was used to quantify cavitation intensity. Approximately 80 beats were recorded at every test condition. High-speed video and an accelerometer were used to determine the position of the valves during closure. The cavitation intensity of the inlet valve was minimal when the onset of systole occurred at the moment when the pump just completed filling (RMS was approximately zero). The cavitation intensity increased when the onset of systole occurred before the pump was completely filled (valve partially opened), reaching a plateau of approximately 16 mm Hg when the valve was fully open. The cavitation intensity increased again when diastolic duration exceeded pump filling time by more than 30 milliseconds. The outlet valve cavitation intensity was very low (<4 mm Hg) regardless of the systolic duration, which can be attributed to the position of the hydrophone being on the opposite side of cavitation events. Although very small, the outlet cavitation intensities with respect to systolic duration show a trend similar to the inlet valve cavitation with respect to diastolic duration. Both inlet and outlet valve cavitation increased with increased peak regurgitant flow. An understanding of the relationship of the inlet and outlet valve cavitation to the diastolic and systolic duration can be used to determine the optimal operating conditions of the pulsatile pediatric pump.     

Effects of mechanical valve orifice direction on the flow pattern in a ventricular assist device

We have been developing a pneumatic ventricular assist device (PVAD) system consisting of a diaphragm-type blood pump. The objective of the present study was to evaluate the flow pattern inside the PVAD, which may greatly affect thrombus formation, with respect to the inflow valve-mount orientation. To analyze the change of flow behavior caused by the orifice direction (OD) of the valve, the flow pattern in this pump was visualized. Particle image velocimetry was used as a measurement technique to visualize the flow dynamics. A monoleaflet mechanical valve was mounted in the inlet and outlet ports of the PVAD, which was connected to a mock circulatory loop tester. The OD of the inlet valve was set at six different angles (OD = 0°, 45°, 90°, 135°, 180°, and 270°, where the OD opening toward the diaphragm was defined as 0°) and the pump rate was fixed at 80 bpm to create a 5.0 l/min flow rate. The main circular flow in the blood pump was affected by the OD of the inlet valve. The observed regional flow velocity was relatively low in the area between the inlet and outlet port roots, and was lowest at an OD of 90°. In contrast, the regional flow velocity in this area was highest at an OD of 135°. The OD is an important factor in optimizing the flow condition in our PVAD in terms of preventing flow stagnation, and the best flow behavior was realized at an OD of 135°.     

On the Characterization of a Non-Newtonian Blood Analog and Its Response to Pulsatile Flow Downstream of a Simplified Stenosis

Particle image velocimetry (PIV) was used to investigate the influence of a non-Newtonian blood analog of aqueous xanthan gum on flow separation in laminar and transitional environments and in both steady and pulsatile flow. Initial steady pressure drop measurements in laminar and transitional flow for a Newtonian analog showed an extension of laminar behavior to Reynolds number (Re) ~ 2900 for the non-Newtonian case. On a macroscale level, this showed good agreement with porcine blood. Subsequently, PIV was used to measure flow patterns and turbulent statistics downstream of an axisymmetric stenosis in the aqueous xanthan gum solution and for a Newtonian analog at Re ~ 520 and Re ~ 1250. The recirculation length for the non-Newtonian case was reduced at Re ~ 520 resultant from increased viscosity at low shear strain rates. At Re ~ 1250, peak turbulent intensities and turbulent shear stresses were dampened by the non-Newtonian fluid in close proximity to the blockage outlet. Although the non-Newtonian case’s recirculation length was increased at peak pulsatile flow, turbulent shear stress was found to be elevated for the Newtonian case downstream from the blockage, suggesting shear layer fragmentation and radial transport. Our findings conclude that the xanthan gum elastic polymer prolongs flow stabilization, which in turn emphasizes the importance of non-Newtonian blood characteristics on the resulting flow patterns in such cardiovascular environments.     

T型分叉血管中血液流动对动脉血栓形成的影响

目的 从血流动力学角度研究T型血管分叉处血液流动的改变对血栓形成的影响。主要从血流速度、分支直径、T型分叉角度及血液粘度方面研究血液流动对血栓形成的影响。同时结合相关医学病例,从多学科角度分析并验证医学研究中的有关血栓形成机理的猜测。方法 建立T-型动脉血管的几何模型,采用计算流体动力学方法对血管内流场进行数值分析研究,分析不同条件的流动对血栓形成影响。结果 在分叉血管附近的支血管和主血管中分别形成了两处较大区域的涡流区域,另外,在分叉交接处的下游位置也出现了一处较小的区域,这些区域速度较低,剪应力出现突然增大趋势,符合血栓形成,因此成为“最适成栓位置”。结论 血栓在“最适成栓位置”的形成还与分支血管直径、血管中血流速度、分叉角度以及血液粘度等有密切关系。     

Flow vectorial analysis in an artificial implantable lung

An artificial implantable lung would be a useful device to support patients awaiting lung transplantation. A suitable device must offer low resistance and adequate gas exchange, be impermeable to plasma, and nonthrombogenic. Although plasma permeability is an intrinsic quality of the materials, the other requirements are largely a function of device geometry, particularly as it relates to fluid dynamics. Using a CAD system and the requirements of a membrane surface area of 1.5 m2 and an inlet outlet port distance of 12 cm, we designed 10 models that varied in their other dimensions. Computational fluid dynamic (CFD) software was applied to the models to determine which minimized regions of low flow velocity. A prototype built to these specifications was used in an in vivo ovine experiment to verify the CFD predictions. The prototype was placed in parallel to the native pulmonary circulation (pulmonary artery to left atrium) for 120 minutes while the activated coagulation times were kept between 110 and 120 seconds and device flow was maintained between 1.5 and 2.5 L/min. Examination of the prototype confirmed a correlation between predicted areas of low flow and thrombus formation. Although nearly identical low flow velocity conditions exist at both the inlet and outlet ports, thrombus formation occurs only near the outlet port. This finding agrees with detailed vectorial analysis, which predicts a more complex flow pattern near the outlet port. Although near the inlet port flow vectors are nearly parallel, near the outlet port flow vectors collide. This area of flow collision corresponds to the area of thrombus formation in vivo. The addition of microflow vectorial analysis to flow velocity predictions allows for improved accuracy in predicting regions at risk of thrombosis in an artificial implantable lung.     

Flow field study comparing design iterations of a 50 cc left ventricular assist device

The Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart Failure (REMATCH) study shows that implanted ventricular assist devices improve survival time and quality of life when used as a permanent therapy in patients who do not qualify for heart transplant. The success of the pulsatile 70 cc stroke volume left ventricular assist device (LVAD) developed by Penn State has led to the development of a 50 cc stroke volume pump for use in patients with smaller chest cavities to benefit a larger patient population. The initial 50 cc pump shows regions of in vivo thrombus formation, which correlate to low wall shear rates within the device. In an in vitro evaluation of three new designs (V-2, V-3, and V-4) of the 50 cc LVAD, identical except for the location and orientation of their outlet ports, particle image velocimetry (PIV) is used to capture planar flow field data within the pumps. V-2 has an outlet port that is located parallel to the inlet. In V-3, the outlet port is rotated away from the inlet port, with the intention of minimizing the amount of fluid turning needed to exit the device. With V-4 the outlet port is moved to the center of the pump to prolong the desirable rotational flow. PIV data were taken at six planar locations within the pump. Although the modifications to the outlet port locations serve their intended purpose, they also introduce unwanted changes in the flow. Poorer wall washing and weaker rotational flow are observed with V-3 and V-4. Although the differences between the devices are subtle, the device that has the most desirable flow characteristics is V-2.     

Flow Residence Time and Regions of Intraluminal Thrombus Deposition in Intracranial Aneurysms

Thrombus formation in intracranial aneurysms, while sometimes stabilizing lesion growth, can present additional risk of thrombo-embolism. The role of hemodynamics in the progression of aneurysmal disease can be elucidated by patient-specific computational modeling. In our previous work, patient-specific computational fluid dynamics (CFD) models were constructed from MRI data for three patients who had fusiform basilar aneurysms that were thrombus-free and then proceeded to develop intraluminal thrombus. In this study, we investigated the effect of increased flow residence time (RT) by modeling passive scalar advection in the same aneurysmal geometries. Non-Newtonian pulsatile flow simulations were carried out in base-line geometries and a new postprocessing technique, referred to as “virtual ink” and based on the passive scalar distribution maps, was used to visualize the flow and estimate the flow RT. The virtual ink technique clearly depicted regions of flow separation. The flow RT at different locations adjacent to aneurysmal walls was calculated as the time the virtual ink scalar remained above a threshold value. The RT values obtained in different areas were then correlated with the location of intra-aneurysmal thrombus observed at a follow-up MR study. For each patient, the wall shear stress (WSS) distribution was also obtained from CFD simulations and correlated with thrombus location. The correlation analysis determined a significant relationship between regions where CFD predicted either an increased RT or low WSS and the regions where thrombus deposition was observed to occur in vivo. A model including both low WSS and increased RT predicted thrombus-prone regions significantly better than the models with RT or WSS alone.     

A numerical investigation on the steady and pulsatile flow characteristics in axi-symmetric abdominal aortic aneurysm models with some experimental evaluation

Steady and pulsatile flow characteristics in rigid abdominal aortic aneurysm (AAA) models were investigated computationally (using Fluent v. 4.3) over a range of Reynolds number (from 200 to 1600) and Womersley number (from 17 to 22). Some comparisons with measurements obtained by particle image velocimetry under the pulsatile flow conditions are also included. A sinusoidal inlet flow waveform 1 + sin omega t with thin inlet boundary layers was used to produce the required pulsatile flow conditions. The bulk features of the mean flow as well as some detailed features, such as wall shear stress distributions, are the foci of the present investigation. Recirculating vortices appeared at different phases of a flow cycle causing significant spatial and temporal variations in wall shear stresses and static pressure distributions. A high level of shear stresses usually appeared at the upstream and downstream ends of the bulge. Effects of pressure rise caused by the increase in cross-sectional area were transmitted into the downstream tube. Further simulation studies were conducted using simulated physiological waveforms under resting and exercise conditions so as to determine the possible implication of vortex dynamics inside the AAA model.     

Biventricular support using a centrifugal pump in a 6 year old with fulminant myocarditis

We experienced a case of ventricular assist with both a pulsatile-flow and a continuous-flow pump in a pediatric patient, and herein report the clinical course and characteristics of the pumps. A 6-year-old female was diagnosed with fulminant myocarditis and transferred to our hospital for mechanical support. After 12 days of extracorporeal membrane oxygenation, we implanted a left ventricular assist device (LVAD) and a right ventricular assist device (RVAD) using centrifugal Gyro pumps with a membrane oxygenator in a paracorporeal fashion. The membrane oxygenator was removed on postoperative day (POD) 4, and the patient was weaned from the respirator on POD 6. The LVAD was exchanged on POD 13 and 17, and the RVAD was exchanged on POD 14 because of thrombus formation inside the pumps. The RVAD was removed on POD 25. On POD 32, the patient experienced cerebral infarction and the centrifugal Gyro pump was switched to an extracorporeal pulsatile pump. No thromboembolic event occurred after pump conversion, although continuous administration of vasodilators was required to avoid hypertension. She underwent successfully heart transplantation in the USA after 8 months of ventricular support. A centrifugal pump is considered useful for pediatric patients, as pump flow and blood pressure can be relatively easily controlled in the postoperative acute phase compared with the pulsatile pump. However, special care should be taken to monitor for thrombus formation when support length becomes longer than 13 days, and a switch to a pulsatile pump should be considered once the hemodynamic status stabilizes.     

Development and evaluation of endurance test system for ventricular assist devices

We developed a novel endurance test system that can arbitrarily set various circulatory conditions and has durability and stability for long-term continuous evaluation of ventricular assist devices (VADs), and we evaluated its fundamental performance and prolonged durability and stability. The circulation circuit of the present endurance test system consisted of a pulsatile pump with a small closed chamber (SCC), a closed chamber, a reservoir and an electromagnetic proportional valve. Two duckbill valves were mounted in the inlet and outlet of the pulsatile pump. The features of the circulation circuit are as follows: (1) the components of the circulation circuit consist of optimized industrial devices, giving durability; (2) the pulsatile pump can change the heart rate and stroke length (SL), as well as its compliance using the SCC. Therefore, the endurance test system can quantitatively reproduce various circulatory conditions. The range of reproducible circulatory conditions in the endurance test circuit was examined in terms of fundamental performance. Additionally, continuous operation for 6 months was performed in order to evaluate the durability and stability. The circulation circuit was able to set up a wide range of pressure and total flow conditions using the SCC and adjusting the pulsatile pump SL. The long-term continuous operation test demonstrated that stable, continuous operation for 6 months was possible without leakage or industrial device failure. The newly developed endurance test system demonstrated a wide range of reproducible circulatory conditions, durability and stability, and is a promising approach for evaluating the basic characteristics of VADs.     

Modelling of pulsatile blood flow in arterial trees of retinal vasculature

The paper presents a numerical investigation of the pulsatile blood flow in the detailed arterial vasculatures of a mouse retina using the mathematical model based on frequency domain incorporating an appropriate outlet boundary impedance at the end of the terminal vessels of the arterial trees. The viscosity in the vessels was evaluated considering the Fahraeus-Lindqvist effect, the plasma skimming effect and in vivo viscosity effect in the microcirculation. Comparative studies of the pulsatile circulation were carried out for cases of rigid vessels, constant viscosity, zero and non-zero outlet boundary impedances. In addition, the dependence of the oscillating input impedance at the inlet of the arterial trees on angular frequencies of the oscillation and vessel elasticises was also studied. The study shows that the pressure wave continues in the pre-capillary vessels throughout the retina. In elastic vessels, the amplitude of oscillatory velocity and wall shear stress in larger vessels and in vessels at the periphery region of the retina is amplified. The pulsatile blood flow is significantly influenced by the outlet boundary (or load) impedance which simulates the effect of the capillary and venous vasculatures. The oscillating input impedance at the inlet of the arterial trees is also found to be dependent on the angular frequency and the Young modulus of the vessel segment. Insights into the potential variations of the dynamic responses of the system under retinal pathological condition of arteriosclerosis may be inferred from the findings of the present study.     

In vitro LDA and flow visualization analysis of the spiral vortex pump: effect of inlet valve orientation

The spiral vortex pump (SV), an innovative, penumatically driven ventricular assist device, was tested using the flow visualization technique and laser Doppler anemometry to study the effect of inlet valve orientation under steady and pulsatile flow conditions in a purposely constructed flow circuit aimed at obtaining flow field data. Qualitative information was obtained using the flow visualization technique. The slit-lighting technique and fluorescent bees provided a clear flow field view at the desired location, and a 200 frames/s high-speed video camera was used, capturing the vortex nature of the flow field. Mean velocity and fluctuating velocity profile were obtained using a Kanomax single-channel FLV system. Three diametrically transverse locations and three vertical locations were selected for measurements. The particle-tracking method was also incorporated to obtain velocity vectors. Based on the experimental data, the following general conclusions can be drawn: (1) The SV pump created a vortex flow field under steady and pulsatile flow conditions. (2) The inlet valve orientation sharply influenced the flow inside the SV pump. (3) A relatively strong circulatory flow field was observed when the major orifice was oriented toward the HD junction under steady flow. (4) A relatively weak circulatory flow field was observed when the major orifice was oriented toward the center under steady flow. (5) The directional flow field was more accentuated under pulsatile flow conditions. (6) A relatively stable flow field was observed when the major orifice was oriented upward (pump outlet direction). (7) Directional flow toward the diaphragm was observed when the major orifice was oriented downward. (8) A strong circulatory flow with possible colliding flow toward the peripheral area was observed when the major orifice was oriented outward. (9) A relatively weak circulatory flow was observed when the major orifice was oriented inward. (10) The strength of the circulatory flow during the peak flow phase under pulsatile conditions was not affected by the orientation of the inlet valve.     

LES of non-Newtonian physiological blood flow in a model of arterial stenosis

Large Eddy Simulation (LES) is performed to study the physiological pulsatile transition-to-turbulent non-Newtonian blood flow through a 3D model of arterial stenosis by using five different blood viscosity models: (i) Power-law, (ii) Carreau, (iii) Quemada, (iv) Cross and (v) modified-Casson. The computational domain has been chosen is a simple channel with a biological type stenosis formed eccentrically on the top wall. The physiological pulsation is generated at the inlet of the model using the first four harmonic series of the physiological pressure pulse (Loudon and Tordesillas [1]). The effects of the various viscosity models are investigated in terms of the global maximum shear rate, post-stenotic re-circulation zone, mean shear stress, mean pressure, and turbulent kinetic energy. We find that the non-Newtonian viscosity models enlarge the length of the post-stenotic re-circulation region by moving the reattachment point of the shear layer separating from the upper wall further downstream. But the turbulent kinetic energy at the immediate post-lip of the stenosis drops due to the effects of the non-Newtonian viscosity. The importance of using LES in modelling the non-Newtonian physiological pulsatile blood flow is also assessed for the different viscosity models in terms of the results of the dynamic subgrid-scale (SGS) stress Smagorinsky model constant, C(s), and the corresponding SGS normalised viscosity.     

A safe procedure for connecting a continuous renal replacement therapy device into an extracorporeal membrane oxygenation circuit

Patients receiving extracorporeal membrane oxygenation (ECMO) often require continuous renal replacement therapy (CRRT). The intra-circuit pressure of adult ECMO usually deviates from the physiological range. We investigated the use of CRRT connected to an ECMO circuit with physiological intra-circuit pressures (0–150 mmHg, defined as the “safety range”) using an in vitro experiment involving a water-filled ECMO circuit. The intra-circuit pressure pre-pump, post-pump, and post-oxygenator were measured while varying the height of the pump or ECMO flow. The bypass conduit pressure and distance from the post-oxygenator port were measured to find the “safety point”, where the bypass pressure remained within the safety range. Both drainage and return limbs of the CRRT machine were connected to the safety point and the inlet and outlet pressures of the hemofilter were recorded while varying the ECMO and CRRT flow. The pre-pump pressure only remained within the safety range for heights?>75 cm (ECMO flow?=?4 L/min) or ECMO flow?<3.5 L min (height?=?50 cm). The post-pump and post-oxygenator pressure was generally outside of the safety range. The bypass pressure decreased according to the distance from the post-oxygenator port and the safety point was found at 60 or 75 cm (in a 90-cm length conduit) regardless of ECMO flow. The hemofilter inlet and outlet pressures remained within the safety range for all conditions of ECMO and CRRT flow, findings validated in clinical cases. The bypass conduit within an ECMO circuit can be connected to a CRRT machine safely under physiological pressures in adult patients receiving ECMO.     

In vitro laser Doppler anemometry of pulsatile flow velocity and shear stress measurements downstream from a jellyfish valve in the mitral position of a ventricular assist device

Thrombus formation and hemolysis have both been linked to the dynamic flow characteristics of heart valve prostheses. To enhance our understanding of the flow characteristics past the mitral position of a jellyfish (JF) valve in the left ventricle under physiological pulsatile flow conditions, in vitro laser Doppler anemometry (LDA) measurements were carried out. The hydrodynamic performance of the JF valve was compared with that of a Bjork-Shiley tilting-disk valve (BS mono). The results indicated that both valves created disturbed flow fields and turbulence shear stress levels in the immediate vicinity and up to 1D (diameter of the valvering) downstream from the valve that were capable of causing lethal damage to blood elements. At a location further downstream, the JF valve showed better hydrodynamic performance than the BS in terms of back flow properties and velocity and turbulence stress characteristics. However, any imperfection in the manufacturing of the valve structure, particularly membrane thickness, adversely affected the performance of the JF valve.     

A numerical investigation on the steady and pulsatile flow characteristics in axi-symmetric abdominal aortic aneurysm models with some experimental evaluation

Steady and pulsatile flow characteristics in rigid abdominal aortic aneurysm (AAA)models were investigated computationally (using Fluent v.4.3) over a range of Reynolds number (from 200 to 1600)and Womersley number (from 17 to 22). Some comparisons with measurements obtained by particle image velocimetry under the pulsatile flow conditions are also included. A sinusoidal inlet flow waveform 1+sin omega t with thin inlet boundary layers was used to produce the required pulsatile flow conditions. The bulk features of the mean flow as well as some detailed features, such as wall shear stress distributions, are the foci of the present investigation. Recirculating vortices appeared at different phases of a flow cycle causing significant spatial and temporal variations in wall shear stresses and static pressure distributions. A high level of shear stresses usually appeared at the upstream and downstream ends of the bulge. Effects of pressure rise caused by the increase in crosssectional area were transmitted into the downstream tube. Further simulation studies were conducted using simulated physiological waveforms under resting and exercise conditions so as to determine the possible implication of vortex dynamics inside the AAA model.     

Volumetric quantification of in vitro sonothrombolysis with microbubbles using high-resolution optical coherence tomography

Several in vitro and in vivo studies have established accelerated thrombolysis using ultrasound (US) induced microbubble (MB) cavitation. However, the mechanisms underlying MB mediated sonothrombolysis are still not completely elucidated. We performed three-dimensional (3-D) volumetric optical coherence tomography (OCT) imaging before and after the application of contrast US to thrombus. The most dramatic reduction in clot volume was observed with US + MB + recombinant tissue plasminogen activator (rt-PA). Thrombus surface erosion in this group on the side of the thrombus exposed to MB and ultrasound was evident on the OCT images. This technique may assist in clarifying the mechanisms underlying sonothrombolysis, especially regarding the importance of US transducer orientation on lytic efficacy and the effects of MB cavitation on thrombus structure.     

体外血栓形成机理的实验研究

血栓形成与流体力学因素关系密切，用健康成人血样进行的实验表明，Chandler血栓环内血液呈现出三维流动和二次流动状态，在管壁处的剪切应力最大。体外血栓形成应考虑血栓环弯曲和内部血液回流因素。血栓环内下弯月面处血栓的形成可能是流速相对较高的流体元冲击下弯月面导致血小板和红细胞聚集所至。当血栓环旋转时间低于10min时，血栓形成不良。15min是血栓形成的最佳时间。转速为3．75rpm时，不能形成血栓。     

Evaluation of the hemodynamics in straight 6-mm and tapered 6- to 8-mm grafts as upper arm hemodialysis vascular access

The present study is intended to investigate and compare the hemodynamics in two different sizes of hemodialysis arteriovenous grafts for upper arm hemodialysis vascular access: 8-mm tapered to 6-mm at the arterial side and straight 6 mm. A computational simulation approach is presented for this study, which is validated against the available experimental and numerical pressure measurements in the literature. The imposed boundary conditions at the arterial inlet and venous outlet boundaries of the models are physiological velocity and pressure waveforms, respectively. Blood flow fields and distribution patterns of the hemodynamic indices including wall shear stress (WSS) as one of the major hemodynamic parameters of the cardiovascular system and spatial wall shear stress gradient (SWSSG) as an indicator of disturbed flow patterns and hence susceptible sites of lesion developments are analyzed and compared between the two grafts. The tapered 6- to 8-mm graft seemingly is associated with less disturbed flow patterns within the venous anastomosis (VA) and the vein downstream while benefiting from higher blood flow rates within. Also, it shows a definitive advantage in terms of WSS and SWSSG distribution patterns around the VA and throughout the vein downstream with significantly lower values, which reduce the risk of thrombosis formation and stenotic lesion developments. The only disadvantage encountered in using 6- to 8-mm tapered graft is higher values of hemodynamic parameters at the arterial junction attributable to its significantly higher mean blood flow rate within. The results clearly indicate that the tapered 6- to 8-mm graft entirely outperforms straight 6-mm graft hemodynamically as an upper arm hemodialysis vascular access graft and confirms clinical data in the literature, which suggests advantageous use of tapered 6- to 8-mm grafts in the creation of upper arm brachioaxillary hemodialysis vascular access grafts in selected groups of patients with expectably higher patency rates and lower complications.     

Results of animal experiments using an undulation pump total artificial heart: analysis of 10 day and 19 day survival

An undulation pump is a special rotary blood pump in which rotation of a brushless DC motor is transformed to an undulating motion by a disc in the pump housing attached by means of a special link mechanism. In the blood pump, a closed line between the disc and housing moves from the inlet to the outlet by this undulating disc motion, which sucks and pushes the blood from the inlet to the outlet. Because the same phenomena occurs at both sides of the disc, a continuous flow is obtained when the motor rotational speed is constant. The pump flow pattern can be easily changed from continuous flow to pulsatile flow by controlling the motor drive current pattern. A seal membrane made of segmented polyurethane protects the blood from invading the link mechanism as well as the motor. UPTAH is fabricated with two undulation pumps and two brushless DC motors. Its size is 75 mm in diameter and 80 mm long, and it has one of the great advantage of no compliance chamber required in the system. UPTAHs were implanted under cardiopulmonary bypass (CPB) into the chest cavities of 16 goats, each weighing between 41 and 72 kg. No anticoagulant and antiplatelet agent was used after the surgery. The left atrial pressure was automatically controlled to prevent its elevation and sucking of the atrial wall into the atrial cuff. The following results were obtained: (1) UPTAHs fit well into all the goats; (2) the longest survival was 19.8 days, the cause of death was bleeding from the aortic anastomosis; (3) No thrombus was observed in the blood pump despite no anticoagulant use. Hemolysis depended upon the length of CPB during surgery. When CPB time was within 2 hours, hemolysis level returned to baseline within a few days of the surgery. UPTAH is a promising implantable TAH, because of its small size and easy controllability.