Preservation of pulsatility with universal ventricular assist device: In vitro assessment for biventricular support.

BACKGROUND: The objective of this study was to assess the pulsatility preservation capability of the universal ventricular assist device (UVAD) when used as a biventricular assist device (BVAD). This evaluation was conducted through an in vitro experiment, utilizing a pulsatile biventricular circulatory mock loop. METHODS: Two UVAD pumps were tested in a dual setup (BVAD) in the circulatory model with the simulated conditions of left heart failure (HF), right HF, and moderate/severe biventricular HF (BHF). The total flow, aortic pulse pressure, the pulse augmentation factor (PAF), the energy-equivalent pressure (EEP), and the surplus hemodynamic energy (SHE) were observed at various pump speeds to evaluate the pulsatility. RESULTS: The aortic pulse pressure increased from the baseline (without pump) in all simulated hemodynamic conditions. The PAF ranged from 17%-35% in healthy, left HF, right HF, and mild BHF conditions, with the highest PAF of 90% being observed in the severe BHF condition. The EEP correlated with LVAD flow in all groups (R2 = 0.87-0.97) and increased from the baseline in all cases. The SHE peaked at approximately 5-6 L/min of LVAD support and was likely to decrease at higher LVAD pump flow. The largest decrease in SHE from the baseline, 53%, was observed in the mild BHF conditions with the highest LVAD and RVAD support. CONCLUSIONS: The UVAD successfully demonstrated the ability to preserve pulsatility in vitro, and to optimize the cardiac output, as an isolated circulatory support device option (RVAD or LVAD) and when used for BVAD support.

Improving hydraulic performance of the left atrial assist device using computational fluid dynamics.

BACKGROUND: The left atrial assist device (LAAD) is a novel continuous-flow pump designed to treat patients with heart failure with preserved ejection fraction, a growing type of heart failure, but with limited device-treatment options. The LAAD is implanted in the mitral plane and pumps blood from the left atrium into the left ventricle. The purpose of this study was to refine the initial design of the LAAD, using results from computational fluid dynamics (CFD) analyses to inform changes that could improve hydraulic performance and flow patterns within the LAAD. METHODS: The initial design and three variations were simulated, exploring changes to the primary impeller blades, the housing shape, and the number, size, and curvature of the diffuser vanes. Several pump rotational speeds and flow rates spanning the intended range of use were modeled. RESULTS: Guided by the insight gained from each design iteration, the final design incorporated impeller blades with improved alignment relative to the incoming flow and wider, more curved diffuser vanes that better aligned with the approaching flow from the volute. These design adjustments reduced flow separation within the impeller and diffuser regions. In vitro testing confirmed the CFD predicted improvement in the hydraulic performance of the revised LAAD flow path design. CONCLUSIONS: The CFD results from this study provided insight into the key pump design-related parameters that can be adjusted to improve the LAAD's hydraulic performance and internal flow patterns. This work also provided a foundation for future studies assessing the LAAD's biocompatibility under clinical conditions.

Left atrial assist device for heart failure with preserved ejection fraction: initial results with torque control mode in diastolic heart failure model.

A novel pump, the left atrial assist device (LAAD), is a device specifically for the treatment of heart failure with preserved ejection fraction (HFpEF). The LAAD is a mixed-flow pump that is implanted in the mitral position and delivers blood from the left atrium to the left ventricle. During the development process, we aimed to explore whether device activation in torque control (TC) mode would improve the function of the LAAD. The TC mode causes adjustment of the pump speed automatically during each cardiac cycle in order to maintain a specified torque. In this study, we tested four different TC settings (TC modes 0.9, 1.0, 1.25, and 1.5) using an in vitro mock circulatory loop. Mild, moderate, and severe diastolic heart failure (DHF) conditions, as well as normal heart condition, were simulated with the four TC modes. Also, we evaluated the LAAD in vivo with three calves. The LAAD was implanted at the mitral position with four TC settings (TC modes 0.9, 1.0, 1.1, 1.2). With LAAD support, the in vitro cardiac output and aortic pressure recovered to normal heart levels at TC 1.25 and 1.5 even under severe DHF conditions with little pump regurgitation. The TC mode tested in vivo with three calves, and it also showed favorable result without elevating the left ventricular end-diastolic pressure. These initial in vitro and in vivo results suggest that the TC mode could be potentially effective, and the LAAD could be a treatment option for HFpEF patients.

Characterization of left atrial assist device implantation: Early results of ex vivo anatomical assessment.

BACKGROUND: The left atrial assist device (LAAD) is a novel pump that was developed specifically for the treatment of heart failure with preserved ejection fraction. The device is surgically implanted in the mitral position. This study aimed to characterize the various device-fitting configurations in the mitral annular position. METHODS: Rapidly prototyped LAAD models (n = 5) were fabricated with five different driveline configurations: (A) annulus level/intra-cuff running; (B) supra-cuff/below coronary sinus (CS); (C) infra-cuff; (D) supra-annulus/supra-CS; (E) left ventricular free wall level. The 3D-printed models were implanted in extracted fresh porcine hearts (80-100 kg, adult, healthy porcine) and the proximity of anatomical structures between the driveline and CS and coronary artery (CA) were measured. RESULTS: All five device configurations were evaluated for fitting. For the purpose of preventing blood clot formation around the driveline, the mitral annulus (MA) as a driveline pass-way (configuration A) has been considered advantageous with the current device, in that the driveline exposure to blood has been avoided. The CS does not exist at exactly the same level as the MA, and there is less risk of injuring it than using the left atrial free wall. However, there is an inevitable risk of damaging the CA, so careful visual inspection before inserting the driveline is needed. CONCLUSIONS: Several options of driveline exteriorization were demonstrated, and the safety of each configuration was evaluated. Using the MA as a pathway for the driveline exit is considered to be a reasonable and safe method.

Left Atrial Circulatory Assistance in Simulated Diastolic Heart Failure Model: First in Vitro and in Vivo.

BACKGROUND: We are developing a left atrial assist device (LAAD) that is implanted at the mitral position to treat diastolic heart failure (DHF) represented by heart failure with preserved ejection fraction. METHODS: The LAAD was tested at 3 pump speeds on a pulsatile mock loop with a pneumatic pump that simulated DHF conditions by adjusting the diastolic drive. The LAAD was implanted in 6 calves, and the hemodynamics were assessed. In 3 cases, DHF conditions were induced by using a balloon inserted into the left ventricle, and in 2 cases, mitral valve replacement was also performed after the second aortic cross-clamp. RESULTS: DHF conditions were successfully induced in the in vitro study. With LAAD support, cardiac output, aortic pressure and left atrial pressure recovered to normal values, whereas pulsatility was maintained for both in vivo and in vitro studies. Echocardiography showed no left ventricular outflow tract obstruction, and the LAAD was successfully replaced by a mechanical prosthetic valve. CONCLUSIONS: These initial in vitro and in vivo results support our hypothesis that use of the LAAD increases cardiac output and aortic pressure and decreases left atrial pressure, while maintaining arterial pulsatility.

Pulsatility hemodynamics during speed modulation of continuous-flow total artificial heart in a chronic in vivo model.

BACKGROUND: The evaluation of pulsatile flow created by the new Cleveland Clinic continuous-flow total artificial heart (CFTAH100), which has a re-designed right impeller and motor, had not been tested in vivo. The purpose of this study was to evaluate the feasibility of pulsatility with the CFTAH100 during the application of pump speed modulation protocols in a chronic animal model. METHODS: A 30-day chronic animal experiment was conducted with a calf. Five pulsatile studies were performed on the alert animal. The mean pump speed was set at 2800 rpm, and modulated sinusoidally within a range of 0 to ± 35% of mean speed, in increments of 5% at 80 beats per minute (bpm). The pressures and pump flow were collected and a pulsatility index (PI) was calculated. RESULTS: The calf was supported with the CFTAH100 without any major complications. The maximum and minimum pump flows changed significantly from baseline in all conditions, while the mean pump flow did not change. All flow pulsatility (FP) readings in all conditions significantly increased from baseline, and the percent modulation (%S) and FP had a strong positive correlation (r = 0.99, p < 0.01). The PI also increased significantly in all conditions (maximum at %S of 35%, 2.2 ± 0.05, p < 0.01), and a positive correlation between %S and PI (r = 0.99, p < 0.01) was observed. CONCLUSION: The CFTAH100 showed the feasibility of creating pulsatile circulation with sinusoidal pump speed modulation.

Analysis of Cleveland Clinic continuous-flow total artificial heart performance using the Virtual Mock Loop: Comparison with an in vivo study.

The Virtual Mock Loop (VML) is a mathematical model designed to simulate mechanism of the human cardiovascular system interacting with mechanical circulatory support devices. Here, we aimed to mimic the hemodynamic performance of Cleveland Clinic's self-regulating continuous-flow total artificial heart (CFTAH) via VML and evaluate the accuracy of the VML compared with an in vivo acute animal study. The VML reproduced 124 hemodynamic conditions from three acute in vivo experiments in calves. Systemic/pulmonary vascular resistances, pump rotational speed, pulsatility, and pulse rate were set for the VML from in vivo data. We compared outputs (pump flow, left and right pump pressure rises, and atrial pressure difference) between the two systems. The pump performance curves all fell in the designed range. There was a strong correlation between the VML and the in vivo study in the left pump flow (r2 = 0.84) and pressure rise (r2 = 0.80), and a moderate correlation in right pressure rise (r2 = 0.52) and atrial pressure difference (r2 = 0.59). Although there is room for improvement in simulating right-sided pump performance of self-regulating CFTAH, the VML acceptably simulated the hemodynamics observed in an in vivo study. These results indicate that pump flow and pressure rise can be estimated from vascular resistances and pump settings.

Effects of blood pump orientation on performance: In vitro assessment of universal advanced ventricular assist device.

An advanced ventricular assist device (VAD), which is under development in our institution, has specific features that allow changes in the axial rotor position and pump performance by intrapump pressure difference. However, performance could be influenced by the pump orientation because of the effect of gravity on the rotor position. The purpose of this study was to evaluate the effects of pump orientation on the pump performance, including pulse pressure and regurgitant flow through the pump when the pump was stopped. Bench testing of the VAD was performed on a static or pulsatile mock loop with a pneumatic device to simulate the native ventricle. The pump performance, including pressure-flow curve, pulsatility, and regurgitant flow, was evaluated at several angles, ranging from -90° (inlet pointed upward) to +90° (inlet pointed downward) at pump speeds of 2000, 2500, 3000, and 3500 rpm. The pump performance was slightly lower at +90° at all rotational speeds, compared with -90°. The pulse pressure on the pulsatile mock loop (80 bpm) was 50 mm Hg without pump support, remained at 50 mm Hg during pump support, and was not changed by orientation (-90°, 0°, and +90°). When the pump was stopped, the regurgitant flow was near 0 L/min at all angles. Pump orientation had a minor effect on pump performance, with no effect on pulse pressure or regurgitant flow when the pump was stopped. This indicates that the effect of gravity on the rotor assembly is insignificant.

A simulation tool for mechanical circulatory support device interaction with diseased states.

We have created a simulation model to investigate the interactions between a variety of mechanical circulatory support (MCS) devices and the circulatory system with various simulated patient conditions and disease states. The present simulation accommodates a family of continuous-flow MCS devices under various stages of consideration or development at our institution. This article describes the mathematical core of the in silico simulation system and shows examples of simulation output imitating various disease states and of selected in vitro and clinical data from the literature.

The design modification of advanced ventricular assist device to enhance pulse augmentation and regurgitant flow shut-off.

The new Advanced ventricular assist device (Advanced VAD) has many features such as improving pulsatility and preventing regurgitant flow during pump stoppage. The purpose of this study was to evaluate the effects of design modifications of the Advanced VAD on these features in vitro. Bench testing of four versions of the Advanced VAD was performed on a static or pulsatile mock loop with a pneumatic device. After pump performance was evaluated, each pump was run at 3000 rpm to evaluate pulse augmentation, then was stopped to assess regurgitant flow through the pump. There was no significant difference in pump performance between the pump models. The average pulse pressure in the pulsatile mock loop was 23.0, 34.0, 39.3, 33.8, and 37.3 mm Hg without pump, with AV010, AV020 3S, AV020 6S, and AV020 RC, respectively. The pulse augmentation factor was 48%, 71%, 47%, and 62% with AV010, AV020 3S, AV020 6S, and AV020 RC, respectively. In the pump stop test, regurgitant flow was -0.60 ± 0.70, -0.13 ± 0.57, -0.14 ± 0.09, and -0.18 ± 0.06 L/min in AV010, AV020 3S, AV020 6S, and AV020 RC, respectively. In conclusion, by modifying the design of the Advanced VAD, we successfully showed the improved pulsatility augmentation and regurgitant flow shut-off features.

Continuous-flow total artificial heart: hemodynamic and pump-related changes associated with posture in a chronic calf model.

This study aimed to evaluate the effects of posture (sitting [lying down]/standing) on hemodynamic and pump-related parameters in calves implanted with our institution's continuous-flow total artificial heart (CFTAH). These parameters were analyzed with posture information in four calves that had achieved the intended 14-, 30-, or 90-day durations of implantation. In each animal, postoperative hourly data gathered throughout the study were used to compare average values with the animal sitting vs. standing. Pump flow became significantly higher in the standing than sitting position at the same pump speed (standing 7.9 ± 0.8, sitting 7.4 ± 1.0 L/min, p = 0.028). Systemic vascular resistance (SVR) and aortic pressure (AoP) were significantly lower in the standing than sitting position (SVR standing 779 ± 145, sitting 929 ± 206 dyne s/cm5, p = 0.027; AoP standing 93 ± 7, sitting 103 ± 7 mm Hg, p < 0.001). No substantial change occurred in pulmonary vascular resistance (PVR) or pulmonary arterial pressure (PAP) with posture (PVR standing 161 ± 39, sitting 164 ± 48 dyne s/cm5, p = 0.639; PAP standing 32 ± 3, sitting 33 ± 4 mm Hg, p = 0.340). Posture affected some hemodynamic and pump-related parameters in calves with CFTAH, with implications for patients with implanted pumps.

Use of a Mechanical Circulatory Support Simulation to Study Pump Interactions With the Variable Hemodynamic Environment.

The Virtual Mock Loop, a versatile virtual mock circulation loop, was developed using a lumped-parameter model of the mechanically assisted human circulatory system. Inputs allow specification of a variety of continuous-flow pumps (left, right, or biventricular assist devices) and a total artificial heart that can self-regulate between left and right pump outputs. Hemodynamic inputs were simplified using a disease-based input panel, allowing selection of a combination of cardiovascular disease states, including systolic and diastolic heart failure, stenosis, and/or regurgitation in each of the four valves, and high to low systemic and pulmonary vascular resistance values. The menu-driven output includes a summary of hemodynamic parameters and graphical output of selected flows, pressures, and volumes in the heart's four chambers as well as in the pulmonary artery and aorta. New tools to augment experimental research on implantable heart-assist devices and to increase our understanding of patient-specific pump interactions are in high demand. The purpose of this ongoing study is to demonstrate the use of a system analysis computer simulation to explore and better comprehend the interactions of mechanical circulatory support pumps with a more extensive combination of patient-specific or simulation conditions than can be established by practical experimentation. Usability is an important factor in constructing computer models for research purposes, and among our primary objectives in creating this simulation model were to make it as portable and useful as possible outside the lab environment, by people not involved in the creation of its operational software.

Generating pulsatility by pump speed modulation with continuous-flow total artificial heart in awake calves.

The purpose of this study was to evaluate the effects of sinusoidal pump speed modulation of the Cleveland Clinic continuous-flow total artificial heart (CFTAH) on hemodynamics and pump flow in an awake chronic calf model. The sinusoidal pump speed modulations, performed on the day of elective sacrifice, were set at ±15 and ± 25% of mean pump speed at 80 bpm in four awake calves with a CFTAH. The systemic and pulmonary arterial pulse pressures increased to 12.0 and 12.3 mmHg (±15% modulation) and to 15.9 and 15.7 mmHg (±25% modulation), respectively. The pulsatility index and surplus hemodynamic energy significantly increased, respectively, to 1.05 and 1346 ergs/cm at ±15% speed modulation and to 1.51 and 3381 ergs/cm at ±25% speed modulation. This study showed that it is feasible to generate pressure pulsatility with pump speed modulation; the platform is suitable for evaluating the physiologic impact of pulsatility and allows determination of the best speed modulations in terms of magnitude, frequency, and profiles.

Cleveland Clinic Continuous-Flow Total Artificial Heart: Progress Report and Technology Update.

Cleveland Clinic's continuous-flow total artificial heart (CFTAH) is being developed at our institution and has demonstrated system reliability and optimal performance. Based on the results from recent chronic in vivo experiments, CFTAH has been revised, especially to improve biocompatibility. The purpose of this article is to report our progress in developing CFTAH. To improve biocompatibility, the right impeller, the pump housing, and the motor were reviewed for design revision. Updated design features were based on computational fluid dynamics analysis and observations from in vitro and in vivo studies. A new version of CFTAH was created, manufactured, and tested. All hemodynamic and pump-related parameters were observed and found to be within the intended ranges, and the new CFTAH yielded acceptable biocompatibility. Cleveland Clinic's continuous-flow total artificial heart has demonstrated reliable performance, and has shown satisfactory progress in its development.

Biventricular circulatory support using single-device and dual-device configurations: Initial pump characterization in simulated heart failure model.

Objective: Severe biventricular heart failure (BHF) can be remedied using a biventricular assist device (BVAD). Two devices are currently in development: a universal ventricular assist device (UVAD), which will be able to assist either the left, right, or both ventricles, and a continuous-flow total artificial heart (CFTAH), which replaces the entire heart. In this study, the in vitro hemodynamic performances of two UVADs are compared to a CFTAH acting as a BVAD. Methods: For this experiment, a biventricular mock circulatory loop utilizes two pneumatic pumps (Abiomed AB5000™, Danvers, MA, USA), in conjunction with a dual-output driver, to create heart failure (HF) conditions (left, LHF; right, RHF; biventricular, BHF). Systolic BHF for four different situations were replicated. In each situation, CFTAH and UVAD devices were installed and operated at two distinct speeds, and cannulations for ventricular and atrial connections were evaluated. Results: Both CFTAH and UVAD setups achieved our recommended hemodynamic criteria. The dual-UVAD arrangement yielded a better atrial balance to alleviate LHF and RHF. For moderate and severe BHF scenarios, CFTAH and dual UVADs both created excellent atrial pressure balance. Conversely, when CFTAH was atrial cannulated for LHF and RHF, the needed atrial pressure balance was not met. Conclusion: Comprehensive in vitro testing of two different BVAD setups exhibited self-regulation and exceptional pump performance for both (single- and dual-device) BHF support scenarios. For treating moderate and severe BHF, UVAD and CFTAH both functioned well with respect to atrial pressure regulation and cardiac output. Though, the dual-UVAD setup yielded a better atrial pressure balance in all BHF testing scenarios.

Artificial Deep Neural Network for Sensorless Pump Flow and Hemodynamics Estimation During Continuous-Flow Mechanical Circulatory Support.

The objective of this study was to compare the estimates of pump flow and systemic vascular resistance (SVR) derived from a mathematical regression model to those from an artificial deep neural network (ADNN). Hemodynamic and pump-related data were generated using both the Cleveland Clinic continuous-flow total artificial heart (CFTAH) and pediatric CFTAH on a mock circulatory loop. An ADNN was trained with generated data, and a mathematical regression model was also generated using the same data. Finally, the absolute error for the actual measured data and each set of estimated data were compared. A strong correlation was observed between the measured flow and the estimated flow using either method (mathematical, R = 0.97, p < 0.01; ADNN, R = 0.99, p < 0.01). The absolute error was smaller in the ADNN estimation (mathematical, 0.3 L/min; ADNN 0.12 L/min; p < 0.01). Furthermore, strong correlation was observed between measured and estimated SVR (mathematical, R = 0.97, p < 0.01; ADNN, R = 0.99, p < 0.01). The absolute error for ADNN estimation was also smaller than that of the mathematical estimation (mathematical, 463 dynes·sec·cm -5 ; ADNN, 123 dynes·sec·cm -5 , p < 0.01). Therefore, in this study, ADNN estimation was more accurate than mathematical regression estimation. http://links.lww.com/ASAIO/A991.

Computational Fluid Dynamics Model of Continuous-Flow Total Artificial Heart: Right Pump Impeller Design Changes to Improve Biocompatibility.

Cleveland Clinic is developing a continuous-flow total artificial heart (CFTAH). This novel design operates without valves and is suspended both axially and radially through the balancing of the magnetic and hydrodynamic forces. A series of long-term animal studies with no anticoagulation demonstrated good biocompatibility, without any thromboemboli or infarctions in the organs. However, we observed varying degrees of thrombus attached to the right impeller blades following device explant. No thrombus was found attached to the left impeller blades. The goals for this study were: (1) to use computational fluid dynamics (CFD) to gain insight into the differences in the flow fields surrounding both impellers, and (2) to leverage that knowledge in identifying an improved next-generation right impeller design that could reduce the potential for thrombus formation. Transient CFD simulations of the CFTAH at a blood flow rate and impeller rotational speed mimicking in vivo conditions revealed significant blade tip-induced flow separation and clustered regions of low wall shear stress near the right impeller that were not present for the left impeller. Numerous right impeller design variations were modeled, including changes to the impeller cone angle, number of blades, blade pattern, blade shape, and inlet housing design. The preferred, next-generation right impeller design incorporated a steeper cone angle, a primary/splitter blade design similar to the left impeller, and an increased blade curvature to better align the incoming flow with the impeller blade tips. The next-generation impeller design reduced both the extent of low shear regions near the right impeller surface and flow separation from the blade leading edges, while maintaining the desired hydraulic performance of the original CFTAH design.

Characterization and Development of Universal Ventricular Assist Device: Computational Fluid Dynamics Analysis of Advanced Design.

We are developing a universal, advanced ventricular assist device (AVAD) with automatic pressure regulation suitable for both left and right ventricular support. The primary goal of this computational fluid dynamics (CFD) study was to analyze the biventricular performance of the AVAD across its wide range of operating conditions. An AVAD CFD model was created and validated using in vitro hydraulic performance measurements taken over conditions spanning both left ventricular assist device (LVAD) and right ventricular assist device (RVAD) operation. Static pressure taps, placed throughout the pump, were used to validate the CFD results. The CFD model was then used to assess the change in hydraulic performance with varying rotor axial positions and identify potential design improvements. The hydraulic performance was simulated and measured at rotor speeds from 2,300 to 3,600 revolutions/min and flow rates from 2.0 to 8.0 L/min. The CFD-predicted hydraulic pressure rise agreed well with the in vitro measured data, within 6.5% at 2300 rpm and within 3.5% for the higher rotor speeds. The CFD successfully predicted wall static pressures, matching experimental values within 7%. High degree of similarity and circumferential uniformity in the pump's flow fields were observed over the pump operation as an LVAD and an RVAD. A secondary impeller axial clearance reduction resulted in a 10% decrease in peak flow residence time and lower static pressures on the secondary impeller. These lower static pressures suggest a reduction in the upwards rotor forces from the secondary impeller and a desired increase in the pressure sensitivity of the pump. The CFD analyses supported the feasibility of the proposed AVAD's use as an LVAD or an RVAD, over a wide range of operating conditions. The CFD results demonstrated the operability of the pump in providing the desired circumferential flow similarity over the intended range of flow/speed conditions and the intended functionality of the AVAD's automated pressure regulation.

Total Artificial Heart Computational Fluid Dynamics: Modeling of Stator Bore Design Effects on Journal-Bearing Performance.

Cleveland Clinic's continuous-flow total artificial heart (CFTAH) is a double-ended centrifugal blood pump that has a single rotating assembly with an embedded magnet, which is axially and radially suspended by a balance of magnetic and hydrodynamic forces. The key to the radial suspension is a radial offset between the stator bearing bore and the magnet's steel laminations. This offset applies a radial magnetic force, which is balanced by a hydrodynamic force as the rotating assembly moves to a "force-balanced" radial position. The journal-bearing blood passage is a narrow flow path between the left and right impellers. The intent of this study was to determine the impact of the stator-bearing bore radius on the journal-bearing hydraulic performance while satisfying the geometric design constraints imposed by the pump and motor configuration. Electromagnetic forces on the journal bearing were calculated using the ANSYS EMAG program, Version 18 (ANSYS, Canonsburg, PA). ANSYS CFX Version 19.2 was then used to model the journal-bearing flow paths of the most recent design of the CFTAH. A transient, moving mesh approach was used to locate the steady state, force-balanced position of the rotating assembly. The blood was modeled as a non-Newtonian fluid. The computational fluid dynamics simulations showed that by increasing stator bore radius, rotor power, stator wall average shear stress, and blood residence time in journal-bearing decrease, while blood net flow rate through the bearing increases. The results were used to select a new bearing design that provides an improved performance compared with the baseline design. The performance of the new CFTAH-bearing design will be confirmed through upcoming in vitro and in vivo testing.

Left atrial assist device to treat patients with heart failure with preserved ejection fraction: Initial in vitro study.

OBJECTIVES: Many patients with heart failure have preserved ejection fraction but also diastolic dysfunction, with no effective therapy. We are developing a new pump (left atrial assist device, LAAD) for implantation at the mitral position to pump blood from the left atrium to sufficiently fill the left ventricle. The purpose of the initial in vitro study was to demonstrate that the LAAD can reduce left atrial pressure (LAP) and increase cardiac output (CO) while maintaining arterial pulsatility and normal aortic valve function using a proof-of-concept device. METHODS: The LAAD concept was tested at 3 pump speeds on a pulsatile mock loop with a pneumatic pump that simulated the normal function of the native ventricle as well as 3 levels of diastolic heart failure (DHF 1, 2, and 3) by adjusting the diastolic drive pressure to limit diastolic filling of the ventricle. RESULTS: Without the LAAD, CO and aortic pressure (AoP) decreased dramatically from 3.8 L/min and 100 mm Hg at normal heart condition to 1.2 L/min and 35 mm Hg at DHF 3, respectively. With LAAD support, both CO and AoP recovered to normal heart values at 3200 rpm and surpassed normal heart values at 3800 rpm. Furthermore, with LAAD support, LAP recovered to almost that of the normal heart condition at 3800 rpm. CONCLUSIONS: These initial in vitro results support our hypothesis that use of the LAAD increases CO and AoP and decreases LAP under DHF conditions while maintaining arterial pulsatility and full function of the aortic valve.