Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

First Clinical Experience With the Pressure Sensor–Based Autoregulation of Blood Flow in an Artificial Heart

First Clinical Experience With the Pressure Sensor–Based Autoregulation of Blood Flow in an... ASAIO Journal 2021 Adult Circulatory Support First Clinical Experience With the Pressure Sensor–Based Autoregulation of Blood Flow in an Artificial Heart IVAN NETUKA,* YURIY PYA ,† BASTIEN POITIER,‡ PETER IVAK ,* MIROSLAV KONARIK ,* JEAN-CHRISTOPHE PERLÈS,§ ZUZANA BLAŽEJOVÁ,¶ HYNEK RIHA ,¶ MAKHABBAT BEKBOSSYNOVA,† ASSEL MEDRESSOVA ,† FABIEN BOUSQUET,§ CHRISTIAN LATRÉMOUILLE,‡ AND PIET JANSEN§ The CARMAT-Total Artificial Heart (C-TAH) is designed to pro- showed a range of average inflow pressures of between 5 and vide heart replacement therapy for patients with end-stage 20 mm Hg during their daily activities, resulting in cardiac out- biventricular failure. This report details the reliability and effi- put responses of between 4.3 and 7.3 L/min. Operator adjust- cacy of the autoregulation device control mechanism (auto- ments were cumulatively only required on 20 occasions. This mode), designed to mimic normal physiologic responses to report demonstrates that the C-TAH auto-mode effectively pro- changing patient needs. Hemodynamic data from a continu- duces appropriate physiologic responses reflective of changing ous cohort of 10 patients implanted with the device, recorded patients’ daily needs and represents one of the unique char- over 1,842 support days in auto-mode, were analyzed with acteristics of this device in providing almost physiologic heart respect to daily changing physiologic needs. The C-TAH uses replacement therapy. ASAIO Journal 2021; 67;1100–1108 embedded pressure sensors to regulate the pump output. Right and left ventricular outputs are automatically balanced. The Key Words: total artificial heart, autoregulation, physiologic, operator sets target values and the inbuilt algorithm adjusts the hemodynamics, pulsatile, bioprosthetic stroke volume and beat rate, and hence cardiac output, auto- matically. Auto-mode is set perioperatively after initial postcar- Biventricular heart replacement pumps are used to support or diopulmonary bypass hemodynamic stabilization. All patients substitute the cardiac output of patients suffering from severe biventricular failure. One of the key objectives of this therapy is to discharge a patient back to a home environment and asso- From the *Department of Cardiovascular Surgery, Institute for Clinical and Experimental Medicine, Prague, Czech Republic; †National ciated usual activities that make for a normal quality of life. A Research Cardiac Surgery Center (NRCSC), Nur-Sultan (Astana), central requirement for achieving this is to have a device that Kazakhstan; ‡Georges Pompidou European Hospital, Paris, France; can be autonomous while requiring minimal attention. From a §Carmat SA, Vélizy, France; and ¶Department of Anesthesiology physiologic standpoint, the device should emulate the function and Intensive Care Medicine, Institute for Clinical and Experimental Medicine, Prague, Czech Republic. of a native heart. This should preferably be accomplished by Submitted for consideration September 2020; accepted for publica- providing fully pulsatile flow with normal flow and pressure tion in revised form April 2021. profiles, a Starling-like response to changes of input pressure Disclosure: Dr. Netuka reports grants, personal fees, and nonfinan- by modifying the cardiac output, along with maintaining a low cial support from Carmat SA, during the conduct of the study; grants, right atrial pressure so as to avoid edema without creating neg- personal fees and nonfinancial support from Carmat SA, grants, per - sonal fees and nonfinancial support from Abbott Int., nonfinancial ative atrial pressures. Therefore, left/right cardiac outputs need support and other from LeviticusCardio Ltd., personal fees and non- to be balanced to avoid pulmonary edema, while providing financial support from Evaheart Inc., outside the submitted work. Dr. adequate end-organ perfusion. From a patient management Poitier reports personal fees from Carmat, during the conduct of the perspective, the device needs to require minimal interventions study; personal fees from Carmat, outside the submitted work. Dr. Ivak reports personal fees and nonfinancial support from CARMAT, during and the patient needs to require minimal medications. the conduct of the study; grants, personal fees and nonfinancial sup- The CARMAT-Total Artificial Heart (C-TAH) has been port from Abbott, grants from Ministry of Health of Czech Republic, designed to achieve above objectives by the incorporation outside the submitted work. Dr. Konarik reports personal fees and non- of proven biocompatible materials, a pumping mechanism financial support from CARMAT during the conduct of the study; grants which mimics ventricular dynamics and a control algorithm and nonfinancial support from Abbott, grants from Ministry of Health of Czech Republic outside the submitted work. Perlès and Bousquet report which has a Starling-like response. being employed by Carmat SA, during the conduct of the study. Dr. This report details the clinical experience with the entire first Riha reports grants from Carmat SAS, during the conduct of the study; cohort of 10 patients enrolled in the ongoing CE Mark trial (pro- personal fees from Abbott, outside the submitted work. Dr. Latrémouille tocol details at ClinicalTrials.gov Identifier: NCT02962973). reports to be a Carmat consultant during the conduct of the study. Dr. Jansen reports being employed by Carmat SA, during the conduct of the LWW study. The other authors have no conflicts of interest to report. Materials and Methods This study was sponsored by Carmat SA, Vélizy-Villacoublay Cedex, France. Correspondence: Ivan Netuka, Department of Cardiovascular Device Description Surgery, Institute for Clinical and Experimental Medicine, Vídeňská The C-TAH is a single-unit device with bioprosthetic blood-con- 1958/9, Prague 4, Czech Republic. Email: ivne@ikem.cz. Copyright © ASAIO 2021 This is an open-access article distributed tacting surfaces, designed for orthotopic placement. Each ventricle under the terms of the Creative Commons Attribution-Non consists of two compartments, separated by a hybrid membrane Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is (Figure 1). Two electrohydraulic rotary pumps create systolic and permissible to download and share the work provided it is properly diastolic phases by rapidly reversing the direction of hydrau- cited. The work cannot be changed in any way or used commercially without permission from the journal. lic fluid-flow that, alternately, pushes and pulls the membranes. Pressure sensors in each ventricle provide information on preload DOI: 10.1097/MAT.0000000000001485 1100 EXPERIENCE WITH AN AUTOREGULATED ARTIFICIAL HEART Figure 1. The CARMAT-TAH. LV, left ventricle; RV, right ventricle; TAH, total artificial heart. and afterload, while ultrasound transducers measure the position ventricular inflow pressure (RVIP), which is the right ventricu- of the membranes. The pressure sensors have a low 1 year drift of lar target pressure during diastole; the objective is to achieve a 0.375 mm Hg that might influence the cardiac output. However, near-empty right atrium without creating suction. The second this can be managed by adjusting the device settings. The cardiac is the average left-right inflow pressure gap (ALRIPG), which output is calculated as the product of the ejected ventricular vol- is the target for the difference between the average left and ume and the beat rate. An algorithm responds to changes in pre- right inflow ventricular pressures; the objective is to achieve an load and afterload by adjusting the beat rate (35–150 beats per unloaded left atrium, and to manage left/right balance, to take minute) while stroke volume is maximized to 55-60 ml (autoregu- into account the bronchial circulation. Finally, the left minimal lation). The resulting pulsatile blood flow ranges from 2 to 9 L/ outflow pressure threshold (LMOPT) is the target for a mini- min with automated adjustments on the left side to correct for the mum average of the left outflow pressure, designed to accom- contribution of the bronchial circulation. Electronics and micro- modate instances of low arterial pressure. processors are contained within the device. Bioprosthetic valves The CARMAT-TAH function is divided into a diastolic and a (Edwards Lifesciences, Irvine, CA), located at the inflow and out- systolic phase, as described later and in Figure 3. flow of each blood compartment, maintain unidirectional flow. The prosthesis is partially surrounded by a flexible polyurethane Diastolic Phase compliance bag, that contains the hydraulic fluid. A percutane- ous driveline (8 mm diameter) delivers power to the C-TAH and The diastolic phase targets a maximum stroke volume of retrieves information on the device performance. The driveline 60 ml according to the three periods. connects to a portable controller and battery pack while provid- Period 1: Each diastolic phase starts with a filling of around ing an uninterrupted power supply and to display device data 4 ml of blood into the ventricle to open the inflow valve. and alarms. A pressure sensor is integrated inside the controller This is achieved by an initial rapid movement of the to automatically calibrate the pressure sensors located inside the membrane. device. The physician connects a medical console to the controller Period 2: The ventricular pressure sensor monitors the pres- to access medical information and change settings, as required. sure inside the ventricle, providing an evaluation of the venous return. The speed of the membrane is dynamically Autoregulation adjusted every millisecond to initially fill up to half of Auto-mode is an operating mode in which the C-TAH behaves the ventricular volume (30 ml), modulated by the differ- similarly to Starling’s law: it automatically adapts ventricular ence between the ventricular pressure and the set RVIP. outputs in response to changes in preload detected by the pres- When this difference is low, the speed of the membrane is sure sensors located inside the device. Changes in filling dura- slowed down to avoid dropping below the RVIP. tion, ejection duration, and end-diastolic volume result in beat Period 3: The speed of the membrane is decreased to target rate variation, while complying with the set target values. the end-diastolic position for a maximum stroke volume Three main adjustments allow physicians to modify the of 60 ml while taking into account the difference between C-TAH performance envelope (Figure  2). The first is the right the RVIP and the ventricular pressure. NETUKA ET AL. Figure 2. Real-time ventricular pressure curves with a RVIP set at 0 mm Hg. RVIP, right ventricular inflow pressure; TAH, total artificial heart. In case of a negative pressure lower than 50 mm Hg develop- Study Population ing during period 2 of the diastolic phase (due to inflow obstruc- The first 10 consecutive patients enrolled in the European tion or severe hypovolemia), the diastolic phase is terminated premarket clinical trial were included in this study (NCT early, as a safety measure to avoid atrial suction. The next systolic 04475393). All patients were males with end-stage biven- phase is then immediately initiated to minimize blood stasis. tricular heart failure, predominantly in Interagency Registry for Mechanically Assisted Circulatory Support profile 3 at the Systolic Phase time of implant (Table  1). The date of inclusion was the date of implantation for each patient. The follow-up end date was The systolic phase is achieved in a time equivalent to a third the May 31, 2019, or the time of device explantation before of the full cardiac cycle and is designed to completely empty transplantation or death. the ventricular blood volume. The speed of the membrane is increased until half of the blood volume (30 ml) is ejected and Data Collection then decreased until reaching the end-systolic position at rest. That position of the membrane is controlled by an ultrasonic The autoregulation function was evaluated by examining sensor located in the technical compartment of the ventricle. how the device responds to changes in preload. Functional This enables the membrane to be adjusted at each beat to data are stored in the portable controller and the medical reach the optimal end-systolic position. console and were collected throughout the study. Beat by With respect to outflow pressures, only the left afterload beat cardiac output, stroke volume, beat rate, and pressures influences the operation of the autoregulation algorithm. A low were evaluated after the switch from manual mode to auto- left afterload results in speeding up the systolic ejection time to mode, while a 10 minute rolling average was used to ana- eject the full volume in a shorter time and thus restore systemic lyze the behavior of the device, according to the venous blood pressure. Both ventricles are subject to a safety mecha- return. nism which halts systole early, and initiates the next diastolic We also wanted to determine how often the physicians phase, if the output pressures exceed 220 mm Hg. made autoregulation setting adjustments. Therefore, we ana- lyzed all setting changes that were maintained for more than 24 hours. Temporary changes were excluded for two reasons. Left/Right Balance Management First, every single setting change is automatically stored by the The stroke volume of each ventricle is continuously adjusted device; thus, there can be several recorded lines for the same to avoid an imbalance between the average left and right parameter change. Second, since less experienced physicians inflow pressure of more than the LRIPG setting (2.5 mm Hg may change a parameter experimentally, to observe the effect, by default). If the actual difference is above the LRIPG, the they will also cancel it on the same day. right stroke volume is decreased to reduce the average left Other clinical data, such as pre- and postoperative characteris- inflow pressure. When the average inflow pressure difference tics, were collected from the Case Report Forms. Preimplant data is below the LRIPG, the right stroke volume is maximized. were recorded from the baseline or the screening data, if needed. The above general functional description provides an Analyses were performed with SAS software 9.4. Continuous overview of the algorithmic logic employed by the autoreg- data were expressed as mean ± standard deviation. Categorical ulation of the device. The principle is to generate a cardiac variables were expressed by an absolute value and as a output adjusted according to variations in the venous return. percentage. Therefore, the membrane speed is increased during the cur- rent cardiac cycle if the detected ventricular pressure (after Results opening the inflow valve) is higher than it has been in the previous cycle, thus creating a shorter diastolic phase. As a The device was successfully switched from manual mode to consequence, the beat rate increases while the stroke volume auto mode in the operating room, following weaning from car- remains maximized, resulting in a higher flow. diopulmonary bypass (CPB), in all patients. This resulted in an EXPERIENCE WITH AN AUTOREGULATED ARTIFICIAL HEART Figure 3. Flowchart describing the autoregulation mechanism of the CARMAT-TAH. TAH, total artificial heart. immediate appropriate cardiac output response to the targeted The cumulative auto-mode duration for all 10 patients RVIP and variations of venous return (Figure  4). In manual throughout the study was 1,842 days. The default settings mode, the cardiac output is fixed. When switched to auto- (RVIP = 0 mm Hg, average LRIPG = 2.5 mm Hg, LMOPT = mode, the cardiac output adapts to the venous return resulting 90 mm Hg) were most commonly used in these patients, for in stable inflow pressures, according to the targeted left and a cumulative duration of 1,580 days (86% of the time in auto- right inflow pressure settings. regulation). Analysis of the recorded device hemodynamic NETUKA ET AL. Table 1. Patient Characteristics Management of Autoregulation Device Settings Characteristics Mean ± SD Twenty auto-mode setting changes were performed by the medical teams during a cumulative implant duration of 5 years Male 10 (100%) (Table  2). Nearly all the changes (n = 18) were related to the Age at implant, years 60.1 ± 10.1 RVIP. One was related to the LRIPG and one was an adjust- Weight, kg 82.9 ± 7.3 BSA, m 2.0 ± 0.1 ment to the LMOPT. Thirteen device setting changes were done INTERMACS patient profile during the first 30 days while 11 were done in the intensive 2 1 (10%) care unit (ICU). After discharge, only four setting changes were 3 8 (80%) 4 1 (10%) made, on the seven outpatients, during more than 44 cumula- Indication tive months. This represents approximately one setting change Bridge-to-transplant 6 (60%) per 11 patient months. Destination Therapy 4 (40%) Cardiomyopathy Dilated 6 (60%) Ischemic 4 (40%) Hemodynamic Recovery Continuous data are expressed as mean ± standard deviation and Hemodynamic normalization was evident almost immediately categoric data as a number with a percentage. after implantation (Table  3). The C-TAH provided a satisfactory BSA, body surface area; INTERMACS, Interagency Registry for cardiac index of 2.80 ± 0.33 L/min/m as early as postoperative Mechanically Assisted Circulatory Support; SD, standard deviation. day 1 (vs. 1.57 ± 0.52 L/min/m at the baseline) and with much trends show the expected variations in left and right ventricle lower preloads (central venous pressure 10.4 mm Hg on day 1 outputs, corresponding to changes in the inflow pressures, as a vs. 12.8 ± 6.8 mm Hg at the baseline and left atrial pressure 10.2 consequence of beat rate variations, while stroke volumes were ± 2.9 mm Hg on day 1 vs. 23.3 ± 10.4 mm Hg at the baseline). maximized (Figure 5). Left ventricular outputs ranges from 4.3 The cardiac index was subsequently well maintained during the to 7.3 L/min for average left inflow pressures ranges of 6–19 mm support duration with an average of 3.03 ± 0.27 L/min/m at 6 Hg. On the right side, the ventricular output ranges from 4.2 to months. Preloads continued to remain low, as indicated by the 7.2 L/min for average right inflow pressures ranging of 4–17 mm lower jugular vein diameter (12.8 ± 2.1 mm at 6 months vs. 15.0 Hg. The average beat rate ranges from 78 to 128 bpm. ± 4.6 mm at baseline) and by device data recordings of mean Figure 4. Responsiveness of the C-TAH when switched from manual to auto-mode with a targeted right ventricular inflow pressure of 0 mm Hg is demonstrated. Cardiac output in the first day, in the first week, and in the first month after the switch into auto-mode is shown for one patient in the three figures at the bottom. C-TAH, CARMAT-Total Artificial Heart; LV, left ventricle; RV, right ventricle. EXPERIENCE WITH AN AUTOREGULATED ARTIFICIAL HEART Figure 5. Pump output variation in response to inflow pressures is accomplished by beat rate changes while stroke volume (50–60 cc) is maximized. The C-TAH data of the device of the 10 patients when set at the default auto-mode setting were accumulated. C-TAH, CARMAT- Total Artificial Heart; LV, left ventricle; RV, right ventricle. values throughout the study; C-TAH Right Inflow Pressure (9.9 ± interaction with the patient is more complicated because both 3.3 mm Hg) and C-TAH Left Inflow Pressure (12.0 ± 3.4 mm Hg) devices are placed in parallel to the existing cardiovascular sys- at 6 months versus 6.8 ± 1.5 mmHg and 9.56 ± 1.4 mm Hg at tem, where the patient’s heart makes a variable contribution within day 1. Patient’s systolic blood pressures were satisfactory from the combined system. As LVADs are not designed to support the postoperative day 1, with average pressures over 100 mm Hg, right ventricle, surgical techniques are modified to address this trending up by 20 mm Hg at 6 months. need, such as shortening the length of the inflow cannula, insert- ing the right inflow cannula in the right atrium, instead of the right ventricle free wall, to avoid thrombosis or downsizing the right- Discussion sided outflow graft diameter to avoid an elevated right outflow. One of the challenges of mechanical circulatory support The C-TAH autoregulation aims to come closer to a natural (MCS) therapy has been the lack of devices with interactive physiologic interaction with the patient. The manual mode of control systems that automatically and physiologically adjust the C-TAH used a nominal operating mode during the feasibil- to the patients’ hemodynamic changes. Normal cardiac ity study which is now only used during deairing and weaning physiologic control involves neural, hormonal, and intrinsic from CPB. In all patients, observed in this study, the device myocardial mechanisms. These mechanisms are obviously was successfully switched to auto-mode perioperatively. The unavailable to an inanimate device. C-TAH produces a significant increase in cardiac output in The pneumatically powered SynCardia temporary Total response to increased venous return, by beat rate adjustments Artificial Heart (TAH) is employed as a bridge-to-transplant (more than 3 L/min as observed in Figure  5) without device (BTT) device in those patients with irreversible biventricular setting changes, and diminishes the risk of blood stasis by tar- heart failure, at imminent risk of death. The beat rate, positive geting a full ejection. The only event which could lead to a and negative pneumatic pressures, and the systolic duration decreased stroke volume is a reduced venous return, due to are set manually. Thus, to respond to changes in the preload, either severe hypovolemia or a cardiac tamponade. the device is only partially filled at rest. Increases in output The management of the left/right balance is a challenge therefore rely on an increased preload causing an elevated when using biventricular MCS. After implantation of two stroke volume, up to a maximum of 70 ml. This results in a CF-LVADs, speed optimization of the 2 independent pumps very truncated physiologic response due to a cardiac output requires echocardiography guidance to achieve a neutral inter- increase of only 9% during exercise. ventricular septal position. Subsequently, any modification of In the case of the off-label use of two implantable continu- speed implemented on one pump requires a matching speed ous flow left ventricular assist devices (CF-LVADs), physiologic modification of the other pump. Patients supported with two such pumps suffer a limited quality of life due to the necessity Table 2. Auto-Mode Device Settings Changes During of two sets of external equipment, as well as two drivelines. Patient Follow-Up This may not only increase the risk of infection but also lead to additional patient discomfort. Number of Number of Setting There is also growing evidence that a pulsatile flow brings Cumulative Setting Changes/Patient 9,10 Location Months Changes Month some significant advantages in MCS. Pulsatility is difficult to achieve with small rotary continuous flow pumps and is also ICU (n = 10) 5.8 11 1.91 negatively influenced by interactions with the native heart. General ward (n = 8) 10.2 5 0.49 Several methods to compensate such challenges have been Out of hospital (n = 7) 44.4 4 0.09 Total 60.4 20 0.33 used, including the variation of pump speed and the sensing of native ventricular pressures. Additionally, the aforementioned ICU, intensive care unit. NETUKA ET AL. Table 3. Pre- and Postoperative Hemodynamic Characteristics Preimplant D1 D7 M1 M3 M6 Variable (n = 10) (n = 10) (n = 10) (n = 9) (n = 7) (n = 5) Patient hemodynamic characteristics Systolic blood pressure, mm Hg 99.0 ± 10.0 105.1 ± 16.2 (p = 0.389) 109.6 ± 12.1 115.2 ± 10.1 114.0 ± 6.6 124.6 ± 23.0 Diastolic blood pressure, mm Hg 65.9 ± 5.0 57.2 ± 6.7 (p = 0.016) 60.1 ± 10.7 67.3 ± 10.2 75.3 ± 7.7 79.4 ± 5.6 Mean blood pressure, mm Hg 76.7 ± 6.6 72.8 ± 9.5 (p = 0.719) 75.9 ± 10.9 82.9 ± 8.8 87.9 ± 4.9 94.0 ± 10.0 Central venous pressure, mm Hg 12.8 ± 6.8 10.4 ± 2.6 (p = 0.207) 11.7 ± 5.9 (n = 9) N.A. N.A. N.A. PCWP (preimplant)/left atrial 21.6 ± 7.8 10.2 ± 2.9 (n = 9) 11.2 ± 6.1 N.A. N.A. N.A. pressure (postimplant), mm Hg (n = 9) (p = 0.221) (n = 5) Right cardiac output, L/min 3.14 ± 0.97 N.A. N.A. N.A. N.A. N.A. C-TAH hemodynamic characteristics C-TAH left cardiac output, L/min N.A. 5.73 ± 0.63 5.87 ± 0.79 6.11 ± 0.66 5.85 ± 0.57 6.15 ± 0.58 C-TAH right cardiac output, L/min N.A. 5.64 ± 0.59 5.78 ± 0.74 6.03 ± 0.66 5.78 ± 0.62 6.09 ± 0.57 C-TAH R/L cardiac output ratio N.A. 0.99 ± 0.01 0.99 ± 0.02 0.99 ± 0.02 0.99 ± 0.02 0.99 ± 0.01 C-TAH left stroke volume N.A. 56.9 ± 1.7 55.8 ± 1.9 55.9 ± 2.8 57.2 ± 2.7 56.2 ± 2.1 C-TAH right stroke volume N.A. 56.2 ± 2.2 55.0 ± 2.5 55.3 ± 3.6 56.5 ± 2.9 55.7 ± 2.4 C-TAH beat rate, bpm N.A. 99.7 ± 13.3 104.4 ± 14.4 108.6 ± 13.9 101.6 ± 13.3 108.4 ± 9.5 C-TAH left inflow pressure, mm Hg N.A. 9.6 ± 1.4 10.2 ± 2.1 10.2 ± 2.0 9.9 ± 2.3 12.0 ± 3.3 C-TAH right inflow pressure, mm Hg N.A. 6.8 ± 1.5 8.2 ± 2.4 8.4 ± 1.8 7.4 ± 2.7 9.9 ± 3.4 C-TAH left outflow pressure, mm Hg N.A. 97.8 ± 8.8 94.0 ± 6.3 95.0 ± 4.2 99.5 ± 4.2 105.3 ± 12.3 C-TAH right outflow pressure, mm Hg N.A. 46.1 ± 7.4 48.5 ± 10.4 48.6 ± 9.4 43.9 ± 7.1 48.2 ± 10.3 p value denotes difference at D1 from preimplant value. C-TAH, CARMAT-total artificial heart; D1, day 1; D7, day 7; M1, month 1; M3, month 3; M6, month 6; N.A., not available; PCWP, pulmonary capillary wedge pressure. approaches may result in additive blood trauma due to rapid Adjustments to the RVIP were also made for hypovolemic epi- speed modulation and associated shear rates. sodes while clinical interventions (diuretics dose adjustments, For the SynCardia TAH, the left/right balance is managed volume infusion, etc.) were optimized. After the restoration semiautomatically by the device, using the mechanism of of optimal intravascular volume, the device was generally set partial-fill, along with independent negative pressure settings back to the default setting. Seven patients were discharged for each ventricle. However, with this device a higher risk of home with the default device settings, demonstrating that, stroke and bleeding is incurred due to the combination of par- despite temporary setting modifications while the patient is in tial ventricular fill and mechanical valves. Furthermore, the hospital, the clinicians considered that default settings provide driving system is, despite its refinements, still relatively noisy an optimum function. In addition, two patients did not require thus negatively impacting the patients’ quality of life. setting adjustments at any time throughout their entire support. The C-TAH, in comparison, automatically manages the left/ For illustrative purposes, the 1 year hemodynamic trends of right balance by maintaining a preset difference between right one of these patients are shown in Figure  7. This depicts the and left inflow pressures. Thus, the left and right outputs are autonomous variation of the cardiac output according to the automatically adjusted to always maintain an optimal inflow average inflow pressure. pressure difference, compensating for the bronchial flow. The current study has several limitations. It is a nonrandom- This preset difference (ALRIPG) was only modified for the first ized observational study recording the first clinical experience autoregulated patient, 3 days after implantation. It was not with the new autoregulation system. The small number of modified for any other patient. The reduced need for device patients (n = 10) and the low number of only 3 study centers management changes may contribute to greater autonomy for may limit the significance of the data. All patients were male, patients outside of the hospital environment and thus improve however this is common in biventricular support. Since this their quality of life. is an ongoing study, total cohort outcomes were not available The modification of settings of biventricular MCS systems at the time of submission of this article. Additionally, the mini- can be required in several situations. Mitigation of this need mal long-term drift might have an effect on the accuracy of the is appreciated by both the medical team and the patient. autoregulation response. However, there was no clinical situ- Conversely, LVAD management requires optimization of pump ation on any patient in the current study requiring adjustment speed according to the patient’s hemodynamic status to unload of auto-mode device settings that might have been caused by the left ventricle, without inducing suction or right ventricular this drift. dysfunction. Nevertheless, it represents a significant experience of more On the C-TAH, most setting changes were executed dur- than 4 years of device performance with an overall positive ing the first 30 days postoperative, predominantly in ICU, and promising outcome for the patients yet requiring only min- while only four were performed after discharge. Among the imal intervention from the clinicians. 20 changes, 18 concerned the RVIP. This corresponds to the In fact, this experience will allow further improvements of targeted right venous return pressure. It has been suggested the system. Although an observation on exercise response was that an abrupt increase in blood flow in some patients adapted not part of this study, we intend to do so in the future. to chronic low cardiac outputs may result in a renal reperfu- In summary, the C-TAH autoregulation system ensures a sion injury, during the early postoperative recovery phase,. It fully pulsatile cardiac output, that is automatically adjusted, is therefore possible that lifting the RVIP parameters up dur- according to the venous return. It allows an immediate and ing this early phase might bring some benefits (Figure  6). durable hemodynamic recovery, with low preloads, and EXPERIENCE WITH AN AUTOREGULATED ARTIFICIAL HEART Figure 6. The left and right ventricular cardiac output variations in response to inflow pressures, at three RVIP settings, in a representative patient. A lower set RVIP resulted in a higher cardiac output, at the same averaged inflow pressure. C-TAH, CARMAT-Total Artificial Heart; LV, left ventricle; RV, right ventricle; RVIP, right ventricular inflow pressure. Figure 7. Daily average cardiac output and inflow pressure in autoregulation without setting changes, measured over 1 year of support. C-TAH, CARMAT-Total Artificial Heart. NETUKA ET AL. right atrial and right ventricular implantation outcomes. J Heart normal systolic blood pressures. Whether the low number of Lung Transplant 35: 466–473, 2016. device setting interventions will result in reduced readmis- 7. Latrémouille C, Carpentier A, Leprince P, et al: A bioprosthetic sion rates needs to be elucidated in broader scale future stud- total artificial heart for end-stage heart failure: Results from a ies with the C-TAH. pilot study. J Heart Lung Transplant 37: 33–37, 2018. 8. Tran HA, Pollema TL, Silva Enciso J, et al: Durable biventricular support using right atrial placement of the HeartWare HVAD. Acknowledgment ASAIO J 64: 323–327, 2017. 9. Soucy KG, Koenig SC, Giridharan GA, Sobieski MA, Slaughter MS: The authors thank Dr. Dereck R. Wheeldon for editorial guidance Defining pulsatility during continuous-flow ventricular assist on this manuscript. device support. J Heart Lung Transplant 32: 581–587, 2013. 10. Cheng A, Trivedi JR, Van Berkel VH, Massey HT, Slaughter MS: Comparison of total artificial heart and biventricular assist References device support as bridge-to-transplantation. J Card Surg 31: 648–653, 2016. 1. Carpentier A, Latrémouille C, Cholley B, et al: First clinical use of 11. Ng BC, Kleinheyer M, Smith PA, Timms D, Cohn WE, Lim E: a bioprosthetic total artificial heart: Report of two cases. Lancet Pulsatile operation of a continuous-flow right ventricular assist 386: 1556–1563, 2015. device (RVAD) to improve vascular pulsatility. PLoS One 13: 2. Smadja DM, Susen S, Rauch A, et al: The carmat bioprosthetic e0195975, 2018. total artificial heart is associated with early hemostatic recov- 12. Sorensen EN, Dees LM, Kaczorowski DJ, Feller ED: The heartware ery and no acquired von Willebrand syndrome in calves. J lavare cycle: A cautionary tale. ASAIO J 66: e114–e116, 2020. Cardiothorac Vasc Anesth 31: 1595–1602, 2017. 13. Arabía FA, Cantor RS, Koehl DA, et al: Interagency registry for 3. Bizouarn P, Roussel JC, Trochu JN, Perlès JC, Latrémouille C: mechanically assisted circulatory support report on the Effects of pre-load variations on hemodynamic parameters with total artificial heart. J Heart Lung Transplant 37: 1304–1312, a pulsatile autoregulated artificial heart during the early post- operative period. J Heart Lung Transplant 37: 161–163, 2018. 14. Demondion P, Fournel L, Niculescu M, Pavie A, Leprince P: The 4. Canada JM, Evans RK, Abbate A, et al: Exercise capacity in patients challenge of home discharge with a total artificial heart: The La with the total artificial heart. ASAIO J 65: 36–42, 2018. Pitie Salpetriere experience. Eur J Cardiothorac Surg 44: 843– 5. Lavee J, Mulzer J, Krabatsch T, et al: An international multicenter 848, 2013. experience of biventricular support with HeartMate 3 ventricular 15. Smith EM, Franzwa J: Chronic outpatient management of patients assist systems. J Heart Lung Transplant 37: 1399–1402, 2018. with a left ventricular assist device. J Thorac Dis 7: 2112–2124, 6. Shehab S, Macdonald PS, Keogh AM, et al: Long-term biventricu- lar HeartWare ventricular assist device support–Case series of http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png ASAIO Journal Wolters Kluwer Health

First Clinical Experience With the Pressure Sensor–Based Autoregulation of Blood Flow in an Artificial Heart

Download PDF
9 pages

Loading next page...
 
/lp/wolters-kluwer-health/first-clinical-experience-with-the-pressure-sensor-based-ywA1GCIcZT

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Wolters Kluwer Health
Copyright
Copyright © ASAIO 2021
ISSN
1058-2916
eISSN
1538-943X
DOI
10.1097/mat.0000000000001485
Publisher site
See Article on Publisher Site

Abstract

ASAIO Journal 2021 Adult Circulatory Support First Clinical Experience With the Pressure Sensor–Based Autoregulation of Blood Flow in an Artificial Heart IVAN NETUKA,* YURIY PYA ,† BASTIEN POITIER,‡ PETER IVAK ,* MIROSLAV KONARIK ,* JEAN-CHRISTOPHE PERLÈS,§ ZUZANA BLAŽEJOVÁ,¶ HYNEK RIHA ,¶ MAKHABBAT BEKBOSSYNOVA,† ASSEL MEDRESSOVA ,† FABIEN BOUSQUET,§ CHRISTIAN LATRÉMOUILLE,‡ AND PIET JANSEN§ The CARMAT-Total Artificial Heart (C-TAH) is designed to pro- showed a range of average inflow pressures of between 5 and vide heart replacement therapy for patients with end-stage 20 mm Hg during their daily activities, resulting in cardiac out- biventricular failure. This report details the reliability and effi- put responses of between 4.3 and 7.3 L/min. Operator adjust- cacy of the autoregulation device control mechanism (auto- ments were cumulatively only required on 20 occasions. This mode), designed to mimic normal physiologic responses to report demonstrates that the C-TAH auto-mode effectively pro- changing patient needs. Hemodynamic data from a continu- duces appropriate physiologic responses reflective of changing ous cohort of 10 patients implanted with the device, recorded patients’ daily needs and represents one of the unique char- over 1,842 support days in auto-mode, were analyzed with acteristics of this device in providing almost physiologic heart respect to daily changing physiologic needs. The C-TAH uses replacement therapy. ASAIO Journal 2021; 67;1100–1108 embedded pressure sensors to regulate the pump output. Right and left ventricular outputs are automatically balanced. The Key Words: total artificial heart, autoregulation, physiologic, operator sets target values and the inbuilt algorithm adjusts the hemodynamics, pulsatile, bioprosthetic stroke volume and beat rate, and hence cardiac output, auto- matically. Auto-mode is set perioperatively after initial postcar- Biventricular heart replacement pumps are used to support or diopulmonary bypass hemodynamic stabilization. All patients substitute the cardiac output of patients suffering from severe biventricular failure. One of the key objectives of this therapy is to discharge a patient back to a home environment and asso- From the *Department of Cardiovascular Surgery, Institute for Clinical and Experimental Medicine, Prague, Czech Republic; †National ciated usual activities that make for a normal quality of life. A Research Cardiac Surgery Center (NRCSC), Nur-Sultan (Astana), central requirement for achieving this is to have a device that Kazakhstan; ‡Georges Pompidou European Hospital, Paris, France; can be autonomous while requiring minimal attention. From a §Carmat SA, Vélizy, France; and ¶Department of Anesthesiology physiologic standpoint, the device should emulate the function and Intensive Care Medicine, Institute for Clinical and Experimental Medicine, Prague, Czech Republic. of a native heart. This should preferably be accomplished by Submitted for consideration September 2020; accepted for publica- providing fully pulsatile flow with normal flow and pressure tion in revised form April 2021. profiles, a Starling-like response to changes of input pressure Disclosure: Dr. Netuka reports grants, personal fees, and nonfinan- by modifying the cardiac output, along with maintaining a low cial support from Carmat SA, during the conduct of the study; grants, right atrial pressure so as to avoid edema without creating neg- personal fees and nonfinancial support from Carmat SA, grants, per - sonal fees and nonfinancial support from Abbott Int., nonfinancial ative atrial pressures. Therefore, left/right cardiac outputs need support and other from LeviticusCardio Ltd., personal fees and non- to be balanced to avoid pulmonary edema, while providing financial support from Evaheart Inc., outside the submitted work. Dr. adequate end-organ perfusion. From a patient management Poitier reports personal fees from Carmat, during the conduct of the perspective, the device needs to require minimal interventions study; personal fees from Carmat, outside the submitted work. Dr. Ivak reports personal fees and nonfinancial support from CARMAT, during and the patient needs to require minimal medications. the conduct of the study; grants, personal fees and nonfinancial sup- The CARMAT-Total Artificial Heart (C-TAH) has been port from Abbott, grants from Ministry of Health of Czech Republic, designed to achieve above objectives by the incorporation outside the submitted work. Dr. Konarik reports personal fees and non- of proven biocompatible materials, a pumping mechanism financial support from CARMAT during the conduct of the study; grants which mimics ventricular dynamics and a control algorithm and nonfinancial support from Abbott, grants from Ministry of Health of Czech Republic outside the submitted work. Perlès and Bousquet report which has a Starling-like response. being employed by Carmat SA, during the conduct of the study. Dr. This report details the clinical experience with the entire first Riha reports grants from Carmat SAS, during the conduct of the study; cohort of 10 patients enrolled in the ongoing CE Mark trial (pro- personal fees from Abbott, outside the submitted work. Dr. Latrémouille tocol details at ClinicalTrials.gov Identifier: NCT02962973). reports to be a Carmat consultant during the conduct of the study. Dr. Jansen reports being employed by Carmat SA, during the conduct of the LWW study. The other authors have no conflicts of interest to report. Materials and Methods This study was sponsored by Carmat SA, Vélizy-Villacoublay Cedex, France. Correspondence: Ivan Netuka, Department of Cardiovascular Device Description Surgery, Institute for Clinical and Experimental Medicine, Vídeňská The C-TAH is a single-unit device with bioprosthetic blood-con- 1958/9, Prague 4, Czech Republic. Email: ivne@ikem.cz. Copyright © ASAIO 2021 This is an open-access article distributed tacting surfaces, designed for orthotopic placement. Each ventricle under the terms of the Creative Commons Attribution-Non consists of two compartments, separated by a hybrid membrane Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is (Figure 1). Two electrohydraulic rotary pumps create systolic and permissible to download and share the work provided it is properly diastolic phases by rapidly reversing the direction of hydrau- cited. The work cannot be changed in any way or used commercially without permission from the journal. lic fluid-flow that, alternately, pushes and pulls the membranes. Pressure sensors in each ventricle provide information on preload DOI: 10.1097/MAT.0000000000001485 1100 EXPERIENCE WITH AN AUTOREGULATED ARTIFICIAL HEART Figure 1. The CARMAT-TAH. LV, left ventricle; RV, right ventricle; TAH, total artificial heart. and afterload, while ultrasound transducers measure the position ventricular inflow pressure (RVIP), which is the right ventricu- of the membranes. The pressure sensors have a low 1 year drift of lar target pressure during diastole; the objective is to achieve a 0.375 mm Hg that might influence the cardiac output. However, near-empty right atrium without creating suction. The second this can be managed by adjusting the device settings. The cardiac is the average left-right inflow pressure gap (ALRIPG), which output is calculated as the product of the ejected ventricular vol- is the target for the difference between the average left and ume and the beat rate. An algorithm responds to changes in pre- right inflow ventricular pressures; the objective is to achieve an load and afterload by adjusting the beat rate (35–150 beats per unloaded left atrium, and to manage left/right balance, to take minute) while stroke volume is maximized to 55-60 ml (autoregu- into account the bronchial circulation. Finally, the left minimal lation). The resulting pulsatile blood flow ranges from 2 to 9 L/ outflow pressure threshold (LMOPT) is the target for a mini- min with automated adjustments on the left side to correct for the mum average of the left outflow pressure, designed to accom- contribution of the bronchial circulation. Electronics and micro- modate instances of low arterial pressure. processors are contained within the device. Bioprosthetic valves The CARMAT-TAH function is divided into a diastolic and a (Edwards Lifesciences, Irvine, CA), located at the inflow and out- systolic phase, as described later and in Figure 3. flow of each blood compartment, maintain unidirectional flow. The prosthesis is partially surrounded by a flexible polyurethane Diastolic Phase compliance bag, that contains the hydraulic fluid. A percutane- ous driveline (8 mm diameter) delivers power to the C-TAH and The diastolic phase targets a maximum stroke volume of retrieves information on the device performance. The driveline 60 ml according to the three periods. connects to a portable controller and battery pack while provid- Period 1: Each diastolic phase starts with a filling of around ing an uninterrupted power supply and to display device data 4 ml of blood into the ventricle to open the inflow valve. and alarms. A pressure sensor is integrated inside the controller This is achieved by an initial rapid movement of the to automatically calibrate the pressure sensors located inside the membrane. device. The physician connects a medical console to the controller Period 2: The ventricular pressure sensor monitors the pres- to access medical information and change settings, as required. sure inside the ventricle, providing an evaluation of the venous return. The speed of the membrane is dynamically Autoregulation adjusted every millisecond to initially fill up to half of Auto-mode is an operating mode in which the C-TAH behaves the ventricular volume (30 ml), modulated by the differ- similarly to Starling’s law: it automatically adapts ventricular ence between the ventricular pressure and the set RVIP. outputs in response to changes in preload detected by the pres- When this difference is low, the speed of the membrane is sure sensors located inside the device. Changes in filling dura- slowed down to avoid dropping below the RVIP. tion, ejection duration, and end-diastolic volume result in beat Period 3: The speed of the membrane is decreased to target rate variation, while complying with the set target values. the end-diastolic position for a maximum stroke volume Three main adjustments allow physicians to modify the of 60 ml while taking into account the difference between C-TAH performance envelope (Figure  2). The first is the right the RVIP and the ventricular pressure. NETUKA ET AL. Figure 2. Real-time ventricular pressure curves with a RVIP set at 0 mm Hg. RVIP, right ventricular inflow pressure; TAH, total artificial heart. In case of a negative pressure lower than 50 mm Hg develop- Study Population ing during period 2 of the diastolic phase (due to inflow obstruc- The first 10 consecutive patients enrolled in the European tion or severe hypovolemia), the diastolic phase is terminated premarket clinical trial were included in this study (NCT early, as a safety measure to avoid atrial suction. The next systolic 04475393). All patients were males with end-stage biven- phase is then immediately initiated to minimize blood stasis. tricular heart failure, predominantly in Interagency Registry for Mechanically Assisted Circulatory Support profile 3 at the Systolic Phase time of implant (Table  1). The date of inclusion was the date of implantation for each patient. The follow-up end date was The systolic phase is achieved in a time equivalent to a third the May 31, 2019, or the time of device explantation before of the full cardiac cycle and is designed to completely empty transplantation or death. the ventricular blood volume. The speed of the membrane is increased until half of the blood volume (30 ml) is ejected and Data Collection then decreased until reaching the end-systolic position at rest. That position of the membrane is controlled by an ultrasonic The autoregulation function was evaluated by examining sensor located in the technical compartment of the ventricle. how the device responds to changes in preload. Functional This enables the membrane to be adjusted at each beat to data are stored in the portable controller and the medical reach the optimal end-systolic position. console and were collected throughout the study. Beat by With respect to outflow pressures, only the left afterload beat cardiac output, stroke volume, beat rate, and pressures influences the operation of the autoregulation algorithm. A low were evaluated after the switch from manual mode to auto- left afterload results in speeding up the systolic ejection time to mode, while a 10 minute rolling average was used to ana- eject the full volume in a shorter time and thus restore systemic lyze the behavior of the device, according to the venous blood pressure. Both ventricles are subject to a safety mecha- return. nism which halts systole early, and initiates the next diastolic We also wanted to determine how often the physicians phase, if the output pressures exceed 220 mm Hg. made autoregulation setting adjustments. Therefore, we ana- lyzed all setting changes that were maintained for more than 24 hours. Temporary changes were excluded for two reasons. Left/Right Balance Management First, every single setting change is automatically stored by the The stroke volume of each ventricle is continuously adjusted device; thus, there can be several recorded lines for the same to avoid an imbalance between the average left and right parameter change. Second, since less experienced physicians inflow pressure of more than the LRIPG setting (2.5 mm Hg may change a parameter experimentally, to observe the effect, by default). If the actual difference is above the LRIPG, the they will also cancel it on the same day. right stroke volume is decreased to reduce the average left Other clinical data, such as pre- and postoperative characteris- inflow pressure. When the average inflow pressure difference tics, were collected from the Case Report Forms. Preimplant data is below the LRIPG, the right stroke volume is maximized. were recorded from the baseline or the screening data, if needed. The above general functional description provides an Analyses were performed with SAS software 9.4. Continuous overview of the algorithmic logic employed by the autoreg- data were expressed as mean ± standard deviation. Categorical ulation of the device. The principle is to generate a cardiac variables were expressed by an absolute value and as a output adjusted according to variations in the venous return. percentage. Therefore, the membrane speed is increased during the cur- rent cardiac cycle if the detected ventricular pressure (after Results opening the inflow valve) is higher than it has been in the previous cycle, thus creating a shorter diastolic phase. As a The device was successfully switched from manual mode to consequence, the beat rate increases while the stroke volume auto mode in the operating room, following weaning from car- remains maximized, resulting in a higher flow. diopulmonary bypass (CPB), in all patients. This resulted in an EXPERIENCE WITH AN AUTOREGULATED ARTIFICIAL HEART Figure 3. Flowchart describing the autoregulation mechanism of the CARMAT-TAH. TAH, total artificial heart. immediate appropriate cardiac output response to the targeted The cumulative auto-mode duration for all 10 patients RVIP and variations of venous return (Figure  4). In manual throughout the study was 1,842 days. The default settings mode, the cardiac output is fixed. When switched to auto- (RVIP = 0 mm Hg, average LRIPG = 2.5 mm Hg, LMOPT = mode, the cardiac output adapts to the venous return resulting 90 mm Hg) were most commonly used in these patients, for in stable inflow pressures, according to the targeted left and a cumulative duration of 1,580 days (86% of the time in auto- right inflow pressure settings. regulation). Analysis of the recorded device hemodynamic NETUKA ET AL. Table 1. Patient Characteristics Management of Autoregulation Device Settings Characteristics Mean ± SD Twenty auto-mode setting changes were performed by the medical teams during a cumulative implant duration of 5 years Male 10 (100%) (Table  2). Nearly all the changes (n = 18) were related to the Age at implant, years 60.1 ± 10.1 RVIP. One was related to the LRIPG and one was an adjust- Weight, kg 82.9 ± 7.3 BSA, m 2.0 ± 0.1 ment to the LMOPT. Thirteen device setting changes were done INTERMACS patient profile during the first 30 days while 11 were done in the intensive 2 1 (10%) care unit (ICU). After discharge, only four setting changes were 3 8 (80%) 4 1 (10%) made, on the seven outpatients, during more than 44 cumula- Indication tive months. This represents approximately one setting change Bridge-to-transplant 6 (60%) per 11 patient months. Destination Therapy 4 (40%) Cardiomyopathy Dilated 6 (60%) Ischemic 4 (40%) Hemodynamic Recovery Continuous data are expressed as mean ± standard deviation and Hemodynamic normalization was evident almost immediately categoric data as a number with a percentage. after implantation (Table  3). The C-TAH provided a satisfactory BSA, body surface area; INTERMACS, Interagency Registry for cardiac index of 2.80 ± 0.33 L/min/m as early as postoperative Mechanically Assisted Circulatory Support; SD, standard deviation. day 1 (vs. 1.57 ± 0.52 L/min/m at the baseline) and with much trends show the expected variations in left and right ventricle lower preloads (central venous pressure 10.4 mm Hg on day 1 outputs, corresponding to changes in the inflow pressures, as a vs. 12.8 ± 6.8 mm Hg at the baseline and left atrial pressure 10.2 consequence of beat rate variations, while stroke volumes were ± 2.9 mm Hg on day 1 vs. 23.3 ± 10.4 mm Hg at the baseline). maximized (Figure 5). Left ventricular outputs ranges from 4.3 The cardiac index was subsequently well maintained during the to 7.3 L/min for average left inflow pressures ranges of 6–19 mm support duration with an average of 3.03 ± 0.27 L/min/m at 6 Hg. On the right side, the ventricular output ranges from 4.2 to months. Preloads continued to remain low, as indicated by the 7.2 L/min for average right inflow pressures ranging of 4–17 mm lower jugular vein diameter (12.8 ± 2.1 mm at 6 months vs. 15.0 Hg. The average beat rate ranges from 78 to 128 bpm. ± 4.6 mm at baseline) and by device data recordings of mean Figure 4. Responsiveness of the C-TAH when switched from manual to auto-mode with a targeted right ventricular inflow pressure of 0 mm Hg is demonstrated. Cardiac output in the first day, in the first week, and in the first month after the switch into auto-mode is shown for one patient in the three figures at the bottom. C-TAH, CARMAT-Total Artificial Heart; LV, left ventricle; RV, right ventricle. EXPERIENCE WITH AN AUTOREGULATED ARTIFICIAL HEART Figure 5. Pump output variation in response to inflow pressures is accomplished by beat rate changes while stroke volume (50–60 cc) is maximized. The C-TAH data of the device of the 10 patients when set at the default auto-mode setting were accumulated. C-TAH, CARMAT- Total Artificial Heart; LV, left ventricle; RV, right ventricle. values throughout the study; C-TAH Right Inflow Pressure (9.9 ± interaction with the patient is more complicated because both 3.3 mm Hg) and C-TAH Left Inflow Pressure (12.0 ± 3.4 mm Hg) devices are placed in parallel to the existing cardiovascular sys- at 6 months versus 6.8 ± 1.5 mmHg and 9.56 ± 1.4 mm Hg at tem, where the patient’s heart makes a variable contribution within day 1. Patient’s systolic blood pressures were satisfactory from the combined system. As LVADs are not designed to support the postoperative day 1, with average pressures over 100 mm Hg, right ventricle, surgical techniques are modified to address this trending up by 20 mm Hg at 6 months. need, such as shortening the length of the inflow cannula, insert- ing the right inflow cannula in the right atrium, instead of the right ventricle free wall, to avoid thrombosis or downsizing the right- Discussion sided outflow graft diameter to avoid an elevated right outflow. One of the challenges of mechanical circulatory support The C-TAH autoregulation aims to come closer to a natural (MCS) therapy has been the lack of devices with interactive physiologic interaction with the patient. The manual mode of control systems that automatically and physiologically adjust the C-TAH used a nominal operating mode during the feasibil- to the patients’ hemodynamic changes. Normal cardiac ity study which is now only used during deairing and weaning physiologic control involves neural, hormonal, and intrinsic from CPB. In all patients, observed in this study, the device myocardial mechanisms. These mechanisms are obviously was successfully switched to auto-mode perioperatively. The unavailable to an inanimate device. C-TAH produces a significant increase in cardiac output in The pneumatically powered SynCardia temporary Total response to increased venous return, by beat rate adjustments Artificial Heart (TAH) is employed as a bridge-to-transplant (more than 3 L/min as observed in Figure  5) without device (BTT) device in those patients with irreversible biventricular setting changes, and diminishes the risk of blood stasis by tar- heart failure, at imminent risk of death. The beat rate, positive geting a full ejection. The only event which could lead to a and negative pneumatic pressures, and the systolic duration decreased stroke volume is a reduced venous return, due to are set manually. Thus, to respond to changes in the preload, either severe hypovolemia or a cardiac tamponade. the device is only partially filled at rest. Increases in output The management of the left/right balance is a challenge therefore rely on an increased preload causing an elevated when using biventricular MCS. After implantation of two stroke volume, up to a maximum of 70 ml. This results in a CF-LVADs, speed optimization of the 2 independent pumps very truncated physiologic response due to a cardiac output requires echocardiography guidance to achieve a neutral inter- increase of only 9% during exercise. ventricular septal position. Subsequently, any modification of In the case of the off-label use of two implantable continu- speed implemented on one pump requires a matching speed ous flow left ventricular assist devices (CF-LVADs), physiologic modification of the other pump. Patients supported with two such pumps suffer a limited quality of life due to the necessity Table 2. Auto-Mode Device Settings Changes During of two sets of external equipment, as well as two drivelines. Patient Follow-Up This may not only increase the risk of infection but also lead to additional patient discomfort. Number of Number of Setting There is also growing evidence that a pulsatile flow brings Cumulative Setting Changes/Patient 9,10 Location Months Changes Month some significant advantages in MCS. Pulsatility is difficult to achieve with small rotary continuous flow pumps and is also ICU (n = 10) 5.8 11 1.91 negatively influenced by interactions with the native heart. General ward (n = 8) 10.2 5 0.49 Several methods to compensate such challenges have been Out of hospital (n = 7) 44.4 4 0.09 Total 60.4 20 0.33 used, including the variation of pump speed and the sensing of native ventricular pressures. Additionally, the aforementioned ICU, intensive care unit. NETUKA ET AL. Table 3. Pre- and Postoperative Hemodynamic Characteristics Preimplant D1 D7 M1 M3 M6 Variable (n = 10) (n = 10) (n = 10) (n = 9) (n = 7) (n = 5) Patient hemodynamic characteristics Systolic blood pressure, mm Hg 99.0 ± 10.0 105.1 ± 16.2 (p = 0.389) 109.6 ± 12.1 115.2 ± 10.1 114.0 ± 6.6 124.6 ± 23.0 Diastolic blood pressure, mm Hg 65.9 ± 5.0 57.2 ± 6.7 (p = 0.016) 60.1 ± 10.7 67.3 ± 10.2 75.3 ± 7.7 79.4 ± 5.6 Mean blood pressure, mm Hg 76.7 ± 6.6 72.8 ± 9.5 (p = 0.719) 75.9 ± 10.9 82.9 ± 8.8 87.9 ± 4.9 94.0 ± 10.0 Central venous pressure, mm Hg 12.8 ± 6.8 10.4 ± 2.6 (p = 0.207) 11.7 ± 5.9 (n = 9) N.A. N.A. N.A. PCWP (preimplant)/left atrial 21.6 ± 7.8 10.2 ± 2.9 (n = 9) 11.2 ± 6.1 N.A. N.A. N.A. pressure (postimplant), mm Hg (n = 9) (p = 0.221) (n = 5) Right cardiac output, L/min 3.14 ± 0.97 N.A. N.A. N.A. N.A. N.A. C-TAH hemodynamic characteristics C-TAH left cardiac output, L/min N.A. 5.73 ± 0.63 5.87 ± 0.79 6.11 ± 0.66 5.85 ± 0.57 6.15 ± 0.58 C-TAH right cardiac output, L/min N.A. 5.64 ± 0.59 5.78 ± 0.74 6.03 ± 0.66 5.78 ± 0.62 6.09 ± 0.57 C-TAH R/L cardiac output ratio N.A. 0.99 ± 0.01 0.99 ± 0.02 0.99 ± 0.02 0.99 ± 0.02 0.99 ± 0.01 C-TAH left stroke volume N.A. 56.9 ± 1.7 55.8 ± 1.9 55.9 ± 2.8 57.2 ± 2.7 56.2 ± 2.1 C-TAH right stroke volume N.A. 56.2 ± 2.2 55.0 ± 2.5 55.3 ± 3.6 56.5 ± 2.9 55.7 ± 2.4 C-TAH beat rate, bpm N.A. 99.7 ± 13.3 104.4 ± 14.4 108.6 ± 13.9 101.6 ± 13.3 108.4 ± 9.5 C-TAH left inflow pressure, mm Hg N.A. 9.6 ± 1.4 10.2 ± 2.1 10.2 ± 2.0 9.9 ± 2.3 12.0 ± 3.3 C-TAH right inflow pressure, mm Hg N.A. 6.8 ± 1.5 8.2 ± 2.4 8.4 ± 1.8 7.4 ± 2.7 9.9 ± 3.4 C-TAH left outflow pressure, mm Hg N.A. 97.8 ± 8.8 94.0 ± 6.3 95.0 ± 4.2 99.5 ± 4.2 105.3 ± 12.3 C-TAH right outflow pressure, mm Hg N.A. 46.1 ± 7.4 48.5 ± 10.4 48.6 ± 9.4 43.9 ± 7.1 48.2 ± 10.3 p value denotes difference at D1 from preimplant value. C-TAH, CARMAT-total artificial heart; D1, day 1; D7, day 7; M1, month 1; M3, month 3; M6, month 6; N.A., not available; PCWP, pulmonary capillary wedge pressure. approaches may result in additive blood trauma due to rapid Adjustments to the RVIP were also made for hypovolemic epi- speed modulation and associated shear rates. sodes while clinical interventions (diuretics dose adjustments, For the SynCardia TAH, the left/right balance is managed volume infusion, etc.) were optimized. After the restoration semiautomatically by the device, using the mechanism of of optimal intravascular volume, the device was generally set partial-fill, along with independent negative pressure settings back to the default setting. Seven patients were discharged for each ventricle. However, with this device a higher risk of home with the default device settings, demonstrating that, stroke and bleeding is incurred due to the combination of par- despite temporary setting modifications while the patient is in tial ventricular fill and mechanical valves. Furthermore, the hospital, the clinicians considered that default settings provide driving system is, despite its refinements, still relatively noisy an optimum function. In addition, two patients did not require thus negatively impacting the patients’ quality of life. setting adjustments at any time throughout their entire support. The C-TAH, in comparison, automatically manages the left/ For illustrative purposes, the 1 year hemodynamic trends of right balance by maintaining a preset difference between right one of these patients are shown in Figure  7. This depicts the and left inflow pressures. Thus, the left and right outputs are autonomous variation of the cardiac output according to the automatically adjusted to always maintain an optimal inflow average inflow pressure. pressure difference, compensating for the bronchial flow. The current study has several limitations. It is a nonrandom- This preset difference (ALRIPG) was only modified for the first ized observational study recording the first clinical experience autoregulated patient, 3 days after implantation. It was not with the new autoregulation system. The small number of modified for any other patient. The reduced need for device patients (n = 10) and the low number of only 3 study centers management changes may contribute to greater autonomy for may limit the significance of the data. All patients were male, patients outside of the hospital environment and thus improve however this is common in biventricular support. Since this their quality of life. is an ongoing study, total cohort outcomes were not available The modification of settings of biventricular MCS systems at the time of submission of this article. Additionally, the mini- can be required in several situations. Mitigation of this need mal long-term drift might have an effect on the accuracy of the is appreciated by both the medical team and the patient. autoregulation response. However, there was no clinical situ- Conversely, LVAD management requires optimization of pump ation on any patient in the current study requiring adjustment speed according to the patient’s hemodynamic status to unload of auto-mode device settings that might have been caused by the left ventricle, without inducing suction or right ventricular this drift. dysfunction. Nevertheless, it represents a significant experience of more On the C-TAH, most setting changes were executed dur- than 4 years of device performance with an overall positive ing the first 30 days postoperative, predominantly in ICU, and promising outcome for the patients yet requiring only min- while only four were performed after discharge. Among the imal intervention from the clinicians. 20 changes, 18 concerned the RVIP. This corresponds to the In fact, this experience will allow further improvements of targeted right venous return pressure. It has been suggested the system. Although an observation on exercise response was that an abrupt increase in blood flow in some patients adapted not part of this study, we intend to do so in the future. to chronic low cardiac outputs may result in a renal reperfu- In summary, the C-TAH autoregulation system ensures a sion injury, during the early postoperative recovery phase,. It fully pulsatile cardiac output, that is automatically adjusted, is therefore possible that lifting the RVIP parameters up dur- according to the venous return. It allows an immediate and ing this early phase might bring some benefits (Figure  6). durable hemodynamic recovery, with low preloads, and EXPERIENCE WITH AN AUTOREGULATED ARTIFICIAL HEART Figure 6. The left and right ventricular cardiac output variations in response to inflow pressures, at three RVIP settings, in a representative patient. A lower set RVIP resulted in a higher cardiac output, at the same averaged inflow pressure. C-TAH, CARMAT-Total Artificial Heart; LV, left ventricle; RV, right ventricle; RVIP, right ventricular inflow pressure. Figure 7. Daily average cardiac output and inflow pressure in autoregulation without setting changes, measured over 1 year of support. C-TAH, CARMAT-Total Artificial Heart. NETUKA ET AL. right atrial and right ventricular implantation outcomes. J Heart normal systolic blood pressures. Whether the low number of Lung Transplant 35: 466–473, 2016. device setting interventions will result in reduced readmis- 7. Latrémouille C, Carpentier A, Leprince P, et al: A bioprosthetic sion rates needs to be elucidated in broader scale future stud- total artificial heart for end-stage heart failure: Results from a ies with the C-TAH. pilot study. J Heart Lung Transplant 37: 33–37, 2018. 8. Tran HA, Pollema TL, Silva Enciso J, et al: Durable biventricular support using right atrial placement of the HeartWare HVAD. Acknowledgment ASAIO J 64: 323–327, 2017. 9. Soucy KG, Koenig SC, Giridharan GA, Sobieski MA, Slaughter MS: The authors thank Dr. Dereck R. Wheeldon for editorial guidance Defining pulsatility during continuous-flow ventricular assist on this manuscript. device support. J Heart Lung Transplant 32: 581–587, 2013. 10. Cheng A, Trivedi JR, Van Berkel VH, Massey HT, Slaughter MS: Comparison of total artificial heart and biventricular assist References device support as bridge-to-transplantation. J Card Surg 31: 648–653, 2016. 1. Carpentier A, Latrémouille C, Cholley B, et al: First clinical use of 11. Ng BC, Kleinheyer M, Smith PA, Timms D, Cohn WE, Lim E: a bioprosthetic total artificial heart: Report of two cases. Lancet Pulsatile operation of a continuous-flow right ventricular assist 386: 1556–1563, 2015. device (RVAD) to improve vascular pulsatility. PLoS One 13: 2. Smadja DM, Susen S, Rauch A, et al: The carmat bioprosthetic e0195975, 2018. total artificial heart is associated with early hemostatic recov- 12. Sorensen EN, Dees LM, Kaczorowski DJ, Feller ED: The heartware ery and no acquired von Willebrand syndrome in calves. J lavare cycle: A cautionary tale. ASAIO J 66: e114–e116, 2020. Cardiothorac Vasc Anesth 31: 1595–1602, 2017. 13. Arabía FA, Cantor RS, Koehl DA, et al: Interagency registry for 3. Bizouarn P, Roussel JC, Trochu JN, Perlès JC, Latrémouille C: mechanically assisted circulatory support report on the Effects of pre-load variations on hemodynamic parameters with total artificial heart. J Heart Lung Transplant 37: 1304–1312, a pulsatile autoregulated artificial heart during the early post- operative period. J Heart Lung Transplant 37: 161–163, 2018. 14. Demondion P, Fournel L, Niculescu M, Pavie A, Leprince P: The 4. Canada JM, Evans RK, Abbate A, et al: Exercise capacity in patients challenge of home discharge with a total artificial heart: The La with the total artificial heart. ASAIO J 65: 36–42, 2018. Pitie Salpetriere experience. Eur J Cardiothorac Surg 44: 843– 5. Lavee J, Mulzer J, Krabatsch T, et al: An international multicenter 848, 2013. experience of biventricular support with HeartMate 3 ventricular 15. Smith EM, Franzwa J: Chronic outpatient management of patients assist systems. J Heart Lung Transplant 37: 1399–1402, 2018. with a left ventricular assist device. J Thorac Dis 7: 2112–2124, 6. Shehab S, Macdonald PS, Keogh AM, et al: Long-term biventricu- lar HeartWare ventricular assist device support–Case series of

Journal

ASAIO Journal – Wolters Kluwer Health

Published: Oct 16, 2021

References

  • First clinical use of a bioprosthetic total artificial heart: Report of two cases.
    Carpentier, A; Latrémouille, C; Cholley, B
  • The carmat bioprosthetic total artificial heart is associated with early hemostatic recovery and no acquired von Willebrand syndrome in calves.
    Smadja, DM; Susen, S; Rauch, A
  • Effects of pre-load variations on hemodynamic parameters with a pulsatile autoregulated artificial heart during the early post-operative period.
    Bizouarn, P; Roussel, JC; Trochu, JN; Perlès, JC; Latrémouille, C
  • Exercise capacity in patients with the total artificial heart.
    Canada, JM; Evans, RK; Abbate, A
  • An international multicenter experience of biventricular support with HeartMate 3 ventricular assist systems.
    Lavee, J; Mulzer, J; Krabatsch, T
  • Long-term biventricular HeartWare ventricular assist device support–Case series of right atrial and right ventricular implantation outcomes.
    Shehab, S; Macdonald, PS; Keogh, AM
  • A bioprosthetic total artificial heart for end-stage heart failure: Results from a pilot study.
    Latrémouille, C; Carpentier, A; Leprince, P
  • Durable biventricular support using right atrial placement of the HeartWare HVAD.
    Tran, HA; Pollema, TL; Silva Enciso, J
  • Defining pulsatility during continuous-flow ventricular assist device support.
    Soucy, KG; Koenig, SC; Giridharan, GA; Sobieski, MA; Slaughter, MS
  • Comparison of total artificial heart and biventricular assist device support as bridge-to-transplantation.
    Cheng, A; Trivedi, JR; Van Berkel, VH; Massey, HT; Slaughter, MS
  • Pulsatile operation of a continuous-flow right ventricular assist device (RVAD) to improve vascular pulsatility.
    Ng, BC; Kleinheyer, M; Smith, PA; Timms, D; Cohn, WE; Lim, E
  • The heartware lavare cycle: A cautionary tale.
    Sorensen, EN; Dees, LM; Kaczorowski, DJ; Feller, ED
  • Interagency registry for mechanically assisted circulatory support report on the total artificial heart.
    Arabía, FA; Cantor, RS; Koehl, DA
  • The challenge of home discharge with a total artificial heart: The La Pitie Salpetriere experience.
    Demondion, P; Fournel, L; Niculescu, M; Pavie, A; Leprince, P
  • Chronic outpatient management of patients with a left ventricular assist device.
    Smith, EM; Franzwa, J