| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
© 2004 European Association of Cardio-Thoracic Surgery
No drive line, no seal, no bearing and no wear: magnetics for impeller suspension and flow assessment in a new VADService de Chirurgie Cardiovasculaire, Centre Hospitalier Universitaire Vaudois CHUV, 1011 Lausanne, Switzerland
* Corresponding author. Tel.: +41-21-314-22-80; fax: +41-21-314-22-78 Received October 30, 2003; received in revised form January 7, 2004; accepted January 9, 2004
The new magnetically suspended axial pump is free of seals, bearings, mechanical friction and wear. In the absence of a drive shaft or flow meter, pump flow assessment is made with an algorithm based on currents required for impeller rotation and stabilization. The aim of this study is to validate pump performance, algorithm-based flow and effective flow. A series of bovine experiments was realized after equipment with pressure transducers, continuous-cardiac-output-catheter, intracardiac ultrasound (AcuNav) over 6 h. Pump implantation was through a median sternotomy (LV  VADcalibrated transonic-flow-probeaorta). A transonic-HT311-flow-probe was fixed onto the outflow cannula for flow comparison. Animals were electively sacrificed and at necropsy systematic pump inspection and renal embolus score was realized. Observation period was 340±62.4 min. The axial pump generated a mean arterial pressure of 58.8±14.3 mmHg (max 117 mmHg) running at a speed of 6591.3±1395.4rev./min (min 5000/max 8500rev./min) and generating 2.5±1.0 l/min (min 1.4/max 6.0 l/min) of flow. Correlation between the results of the pump flow algorithm and measured pump flow was linear (  ). VAD explants were free of macroscopic thrombi. Renal embolus score was 0±0. The magnetically suspended axial flow pump provides excellent left ventricular support. The pump flow algorithm used is accurate and reliable. Therefore, there is no need for direct flow measurement.
Key Words: Ventricular assist device; Axial blood pump; Heart failure
Terminal congestive cardiomyopathy has become the leading cause of death in developed countries [1]. The shortage in donor hearts has led to the development of long-term mechanical support systems. Derived from the need for slow weaning or even prolonged support after cardiopulmonary bypass [2], is the first medium to long-term clinical left ventricular bypass pump [3]. Since then, several pulsatile ventricular assist devices (VADs) have been designed and did become the most used devices for bridge-to-transplantation or as substitute for cardiac transplantation, as in destination therapy [4]. Those pumps have shown to provide a reasonable quality of life outside the hospital environment [5]. While over a decade of pulsatile cardiac support, several limitations and shortcomings have been identified. First, the increased size of the pumps results in difficult surgical implantation techniques and severely restricting the potential recipient number; second, high pre-disposition of infections due to large drivelines crossing the natural skin barrier [6]. Furthermore, the large amount of moving mechanical elements increases the danger for malfunctioning. To overcome these shortcomings and after encouraging pre-clinical results of the physiological tolerance to non-pulsatile flow, the first-generation of miniaturized and non-pulsatile pumps for cardiac support was introduced into clinics and currently four axial flow pumps are undergoing clinical investigations including the Jarvic 2000 [7], HeartMate II [8], the MicroMed DeBekey VAD [9] and most recently the Incor Berlin Heart. The Incor pump is a new generation implantable VAD working as an axial flow pump on the principle of the Archimedes screw with a fully magnetically suspended impeller. Free of seals, bearings, mechanical friction and wear and in the absence of a drive shaft or flow meter, pump flow assessment is calculated with an algorithm based on currents required for impeller rotation and stabilization. The study was designed to describe the pump with its control system and the implantation techniques as well as to validate the integrated flow rate calculation without flow sensor.
2.1. Device components The Incor® is an implantable axial flow pump with full magnetic suspension of the containing impeller (Fig. 1). The system further permanently consists of an inflow cannula connected to the left ventricle, an angled connection to the outflow cannula anastomosed onto the ascending aorta, a percutaneous pump cable connected to the control unit and a main as well as a backup battery (Fig. 2). All tubings are made from medical grade silicone and are interconnected by snap-in connectors. The impeller is suspended by magnetic bearings (Fig. 1) inside the axial flow pump rotating at speeds between 5000 and 10,000rev./min generating up to 5 l/min of flow against a pressure of 100 mmHg at a power consumption of 2–4W. The pump made of titanium weighs 200 g, is 114 mm long, outer diameter measures 30 mm, internal diameter 16 mm and has a casting volume of 60 cm3. Blood flow is directed through a flow straightener at the pump inflow before being driven by the spinning impeller and exhaust through a flow diffuser located near the pump outflow. A special characteristic of this VAD is the integrated flow rate calculation by means of impeller position variability inside the generated magnetic field. Given the fact that for a known rotational speed a defined pressure head across the pump is generated leading to an impeller shift towards the outflow pole of the pump housing. In order to neutralize this flow force, a counteracting actuator force is generated to bring back the impeller into the centralized position. The amount of electrical energy needed to keep the impeller in its neutral or central position is the principal changing variable for the flow calculation algorithm.
2.2. Implantation The VAD is implanted using standard cardiothoracic surgery techniques. Access is via medial sternotomy or lateral thoracotomy. Implantation can be carried out under ventricular fibrillation, cardioplegic arrest or beating heart. After cannulation for standard ECC, either the outflow or the inflow cannula are placed first. The outflow cannula is anastomosed onto the ascending aorta using a tangential clamp and the inflow cannula is inserted into the apex of the left ventricle using an apex-coring knife supplied with the system. The ring holder for the apical inflow cannula is designed to be connected to a ventricular vent. After the air has been expelled from the pump and tubing, the components of the system are connected with the snap-in connectors and placed in the pre-pericardial space. Finally the pump cable is passed through a percutaneous tunnel and connected to the control unit and power supply.
Three bovine experiments (body weight 79.3±10.7 kg) were performed in compliance with the, ‘Principles of Laboratory Animal Care’, formulated by the National Society for Medical Research and the, ‘Guide for Care and Use of Laboratory Animals’, published by the National Institute of Health (INH publication 85-23, revised 1985). After pre-medication, general anesthesia was started with thiopental sodium. Anesthesia was maintained with Fluothane and a median sternal splitting incision was performed. The pericardium was opened, the diaphragm was mobilized in its ventral portion and a tunnel for the driveline was prepared. Instrumentation included electrocardiogram, a central venous pressure line, an arterial pressure line in the right carotid artery, and intracardiac ultrasound (AcuNav) in the right femoral vein. The hearts were cannulated after injection of 200 mg of heparin for ECC with a heparin-coated circuit. The ascending aorta was tangentially clamped and the outflow cannula anastomosed to the anterior surface. After initiation of ECC, ventricular fibrillation was started. The apical ventriculotomy was made once by standard incision and twice using the provided apex-coring knife. Care was taken to avoid the coronary arteries and to obtain a cylindrical ventriculotomy with its longitudinal axe parallel to the septum wall to prevent further inflow cannula obstruction. Free trabeculae that potentially obstruct the inflow cannula were resected and the holder mounted apex suture ring inserted. A venting tube was placed inside the cylindrical holder grip. Twelve to 16 felt-backed, double reinforced ethibond sutures were sawn in a circular fashion around the apex, passed through the suture ring and tied down. Finally the inflow cannula was inserted into the prepared port using a protection balloon, rotated into the optimum position and the suture collar of the inflow cannula fixed onto the apical suture ring. After the air has been expelled from the pump and tubing, the components of the system were connected with the snap-in connectors and placed in the pre-pericardial space of the lower anterior mediastinum. Remaining air was removed using a 24G needle. Finally the driveline was passed trough a percutaneous tunnel and connected to the control unit and power supply. A transonic HT 311 flow probe (flow accuracy 1%, resolution 1 ml/min, flow sampling rate 333 Hz) was fixed onto the silicone outflow cannula for flow comparison. The flow probe was previously calibrated by a roller-pump, which underwent itself a volumetric tank and timer calibration during the 3-day study period. The pump was started and the heart weaned from the ECC. Throughout the assist period, the ACT was kept >180 s with heparin. Systematic pump inspection and renal embolus score was realized at necropsy.
During the mean observation period of 6±1 h, the implantable magnetically suspended axial blood pump provided significant left ventricular support with complete unloading of the left heart in all animals, as shown by the hemodynamic parameters in Table 1. Generated mean arterial pressure was 58.8±14.3 mmHg with the Impeller spinning a speed of 6591.3±1395.4 (min 5000/max 8500rev./min) and generating 2.5±1.0 l/min (min 1.4/max 6.0 l/min) of flow.
Verification of the algorithm-based integrated flow calculation with the calibrated transonic flow probe documented a perfectly linear correlation between both flow measure and flow calculation (Fig. 3).
All animals were sacrificed electively. Explanted axial flow pumps were found to be free of thrombi. Necropsy provided evidence that renal parenchyma was free from embolic events leading to a renal embolus score of 0±0.
The comparison of the integrated flow calculation with an external Doppler-based flow probe showed a perfectly linear correlation. The pump flow algorithm used was shown to be accurate and reliable. Therefore, there is no need for direct flow measurement. The magnetically suspended axial flow pump provides excellent left ventricular support during the hall of the observation period. A particularity of axial flow pumps is that flow over the pump depends mainly on pre-load and only secondarily on the impeller spin rate, it is therefore critical to optimize the rotational speed to the highest generated flow. Flow measure becomes a crucial element for optimal pump performance. As the Incor Berlin Heart does not measure flow directly but calculates flow based on an algorithm integrating the current consumption needed to keep the impeller centralized in the electromagnetic field by creating a counteracting actuator force the precision of flow, assessment becomes of uppermost importance. At present there are no data available on the validation of integrated flow calculation in a clinical setting. However, the preliminary results of the Incor Initial Trial (results provided by Berlin Heart AG, Germany, Status of May 6th 2003) show good clinical results with a low mortality of 21% (5 pts). Mortality was mainly due to hemorrhagic complications (3 pts) as well as thrombo-embolic event in one patient (week 3) and multiple organ failure in an other patient (week 2). Out of the 24 patients, two could successfully be weaned from the pump and live at home and three underwent heart transplantation. Ten patients remained on the VAD. In two patients thrombus formation inside the pump system was suspected after gradual reduction of pump flow as well as increase in current requirements. Both patients underwent successful lysis and pump function recovered. Pump thrombus is a known problem in all of the axial pump systems but the exact mechanism remains unclear [10]. In order to minimize the risk of thrombus formation, all the surfaces within the pump coming into blood contact are coated with covalently bonded heparin previously used in cardiopulmonary bypass circuits [11], vascular grafts [12] or assist devices [13] (Carmeda® BioActive Surface). In our experimental setting intracardiac ultrasound (AcuNav) was used to visualize complete unloading of the left ventricle shown by the standstill of the aortic valve as well as to monitor the interaction between the left ventricle and ventricular septum with the pump's inflow cannula or to look for possible thrombus formation. Interestingly, enough optimal pump performance was highly dependent on the exact location and axis of the inflow cannula. Intracardiac ultrasound made evident that in those, ‘healthy calves heart’, without prior ventricular dilatation and thinning of the myocardium there was a potential of inflow occlusion by either the septum itself or the myocardial trabeculae. Care had to been taken to excise an exactly parallel to the septum directed myocardial cylinder and to free the pump inflow tract from all remaining trabeculae. An inflow cannula directed towards the septum leads to important pump underperformance and had to be redirected in order to generate sufficient flow for successful unloading. Intracardiac ultrasound showed to be a very valuable tool to monitor optimal heart pump interface. Evolving technologies and optimized materials have brought continuous progress to the assistance of the failing heart. Whatever cardiac assist system is used—pulsatile or continuous flow, left, right or bi-ventricular [14]—the pumps get smaller, more powerful, are working with increased efficiency and have lower energy consumption. Continuous axial pump systems are known to provide submaximal unloading of the left ventricle and are highly pre-load sensitive, furthermore some of the complications that can occur are, thrombosis within the left ventricle interfering with the pump inflow cannula or worse, obstructing the device inflow conduit, flow reversal through the pump if malfunction or pump standstill occurs in the presence of a residual cardiac activity. Nevertheless, their light weight, small dimensions, reduction of blood exposure to foreign surface and easy system operation make axial flow pump the first choice of cardiac assist device in selected patients and indications.
Until 28th of October 2003, the Incor axial-pump was implanted into 76 patients as bride-to-transplantation. Cumulative support time is actually 20 years and the mean support time is 100 days.
We thank Monique Augstburger, Isabelle Seigneul and Judith Horisberger for their technical assistance received for and during the animal experiments, which contributed to the success of this procedure.
Presented at the joint 17th Annual Meeting of the European Association for Cardio-thoracic Surgery and the 11th Annual Meeting of the European Society of Thoracic Surgeons, Vienna, Austria, October 12–15, 2003. doi:10.1016/j.icvts.2004.01.014
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
Top
Abstract
1. Introduction
).