- Email: [email protected]
Implantable rate-responsive counterpulsation assist system
Implantable Rate-Responsive Counterpulsation Assist Svstem J
Carlos M. Li, MD, Andrew Hill, MD, Michael Colson, MS, Carolyne Desrosiers, BS, and Ray C.-J. Chiu, MD, PhD Division of Cardiovascular and Thoracic Surgery, McGill University, Montreal, Canada, and Medtronic, Inc, Minneapolis, Minnesota
To apply the potential energy source available from skeletal muscle in cardiac assistance, we developed an implantable counterpulsation assist system. This study reports the results using this implantable counterpulsation assist system in an acute in vivo animal model. Twelve dogs had a dual-chambered, extraaortic counterpulsation pump anastomosed in parallel to the thoracic aorta. The left latissimus dorsi muscle was used to power the pump. A newly developed implantable stimulator was used to make the muscle contract in synchrony with the diastolic phase. The unique feature of this stimulator is its ability to adjust timing of muscle contraction according to changing heart rates. The stimulator is also able to detect arrhythmias, and as a safety measure, shuts down until a normal rhythm is resumed. During counterpulsation assist with the implantable counterpulsation assist system, diastolic pressure increased an aver-
age of 34 mm Hg from baseline, equivalent to a 69% augmentation. Systolic peak pressure decreased an average of 10 mm Hg, equivalent to an 11%unloading. With induced heart rate changes, the implantable counterpulsation assist system readjusted its timing, maintaining optimal counterpulsation without systolic interference. Induced ventricular tachycardia resulted in immediate shutdown of the stimulator until resumption of a normal rhythm. The feasibility of using an intraaortic balloon pump console as back-up was also demonstrated. Excellent counterpulsation was obtained with either muscle power or balloon pump console. We conclude that the implantable counterpulsation assist system can provide effective counterpulsation assist and has the potential for continuous cardiac support.
p i r c u l a t o r y counterpulsation with the intraaortic balloon is-a widely- accepted form of cardiac assist. Recently, reports of extended support with the balloon pump have confirmed its efficacy in the chronic situation [ 1 4 ] . In several cases, patients were supported from weeks to almost a year. Many more patients with endstage failure could benefit from chronic counterpulsation support. A major limitation is the patient’s dependency on an external power source with its risk of infection and restriction in mobility. A totally implantable counterpulsation assist system would offer an important therapeutic option for patients with end-stage heart failure. Recent advances in technology and skeletal muscle biology now allow for a viable alternative source in cardiac assistance [5, 61. It has been demonstrated that skeletal muscle can be made fatigue-resistant and powerful enough to continuously assist the heart [7]. We have attempted to apply this autologous power source to a totally enclosed circulatory support system. The main purpose of this investigation was to evaluate the efficacy of our implantable counterpulsation assist system (ICAS) in an acute in vivo model.
Material and Methods Extraaortic Dual Chambered Pump
Presented at the Twenty-fifth Anniversary Meeting of The Society of Thoracic Surgeons, Baltimore, MD, Sep 11-13, 1989. Address reprint requests to Dr Chiu, Montreal General Hospital, 1650 Cedar Ave, Room 947,Montreal, Que, Canada H3G 1A4.
0 1990 by The
Society of Thoracic Surgeons
(Ann Thoruc Surg 1990;49:356-62)
The muscle-driven counterpulsation pump used in this study is as previously described except for some slight modifications [8]. The inner blood sac of the chamber consists of a silicone bladder located within a rigid cylindrical housing. The outlets of the pump were 12 mm in diameter. The pump was powered pneumatically by the left latissimus dorsi muscle through compression of a bulb connected to the pump with Silastic (Dow Corning Corp, Midland, MI) tubing (Fig 1). A port in the Silastic tubing allows optional connection to an intraaortic balloon pump console.
Muscle Stimulator A new, microprocessor based, implantable pulse generator (Prometheus) was provided by Medtronic, Inc (Minneapolis, MN) [9]. This unit has a built-in microcomputer that is programmed through telemetry with software operated on an IBM-compatible personal computer. In our study, software developed by Medtronic in collaboration with us for synchronized counterpulsation was programmed into the stimulator. It allows for pulse burst stimulation of the muscle, synchronized to the cardiac R wave. Timing for onset delay of the pulse burst and its duration period is adjusted automatically by the stimulator according to the varying RR interval that occurs with heart rate changes. In addition, adjustment of the pulse 0003-4975/90/$3.50
LIETAL MUSCLE-POWERED COUNTERPULSATION
Ann Thorac Surg 1990;49:35&62
357
Amplitude: 2 V
Frequency: 33 Hz
Duration
~~l~~
IABP ConSol9
Fig 1 . Schematic representation of implantable counterpulsation assist system. A dual-chambered pump is anastomosed in parallel to the descending thoracic aorta. The latissimus dorsi contracts on a pneumatic bulb displacing a silicone membrane in the pump. The stimulator coordinates muscle contraction with the cardiac rhythm. An intruaortic balloon pump (IABP) console can be attached as an alternative source of power.
voltage, frequency, and assist ratio can be easily made on the personal computer and sent via telemetry. A safety feature available with this pulse generator is an automatic shutdown of stimulation should any cardiac arrhythmias occur. Twelve mongrel dogs of approximately 30 kg were induced with pentobarbital and maintained under 0.5%to 1% halothane endotracheal anesthesia. Through a left thoracotomy, the descending thoracic aorta was exposed, and a dual-chambered pump was anastomosed in parallel to the vessel with 12-mm Dacron grafts. Heparin (100 U/kg) was infused to partially anticoagulate the animal. A collapsible bulb was placed underneath the left latissimus dorsi, and the bulb was connected to the pump with Silastic tubing. The left thoracodorsal nerve was dissected free from the neurovascular bundle, and a bipolar neural cuff lead was placed around it. The lead was then connected to the stimulator. For sensing, epicardial leads were attached onto the myocardium and connected to the stimulator. All animals received humane care in compliance with the ”Guide for the Care and Use of Laboratory Animals” published by the National Institutes of Health (NIH publication No. 85-23, revised 1985) and the Canadian Council on Animal Care. The stimulator was programmed to deliver pulse bursts of 2 V, 33 Hz frequency, and a pulse width of 210 ps. The delay period between triggering R wave and onset of stimulus was set to equal 30% of the RR interval, whereas duration of the stimulus was set to equal 60% of the RR interval (Fig 2). These parameters offered optimal timing and contraction force of the muscle for counterpulsation assist. The assist ratio during the studies ranged from 1:l to 3:l. A Millar catheter (Millar Instruments, Houston, TX) pressure transducer was inserted through the left carotid artery into the aortic arch for hemodynamic monitoring.
60% R-R
Fig 2 . Parameter settings of the muscle stimulator. The burst pulses were set at 33 H z frequency and 2 V amplitude. Timing of the bursts was determined by the sensed RR interval of the electrocardiogram (EKG). Delay interval between triggering R wave and burst onset was equivalent to 30% of the RR interval. Duration of burst was 60% of the RR interval.
Studies were performed with the stimulator on and off to observe the effects of counterpulsation on arterial pressure. Each animal served as its own control for comparison. The change in diastolic and systolic pressures were compared using paired t tests for significance. To evaluate the system’s capability to track at different heart rates, esmolol or isoproterenol, or both, was infused as needed to vary the rhythm. The original stimulator settings were maintained throughout these rate changes, and arterial pressures were continuously monitored. The system’s response to acute cardiac dysrhythmias was also tested in 2 dogs. While undergoing counterpulsation assist, the surface of the heart was manually irritated, inducing ventricular ectopy. The response to this insult was monitored continuously throughout the episode. Finally, we tested the feasibility of using an intraaortic balloon pump console (Datascope 82, Paramus, NJ) to drive the dual-chambered pump in 1 dog. This was achieved by connecting the gas outlet of the console to a side port of the Silastic tubing, which connects the pneumatic bulb to the pump (Fig 1). By temporarily clamping and excluding the pneumatic bulb from the system, we were able to drive the extraaortic pump with 40- to 50-mL gas volumes from the console.
Results
Hemodynamic Counterpulsation Arterial counterpulsation was observed in all 12 dogs. A typical tracing obtained with the ICAS is shown in Figure 3. Diastolic pressure increase ranged from 20 to 50 mm Hg
Table 1. Pressure Results in 12 Dogs Pressure Diastolic (mm Hg) Systolic (mm Hg) a
Off
On
Augmentation
51.3 ? 14.3 85.8 ? 14.2
83.0 ? 15.5a 77.1 +- 12.3”
34 (69%) -10 (-11%)
Significance: p < 0.001 versus pump off.
358
LIETAL MUSCLE-POWEREDCOUNTERPULSATION
Ann Thorac Surg 1990;49:356-62
Fig 3. Representative arterial tracing with the implantable counterpulsation assist system (ICAS) compared with that with an intraaortic balloon pump (IABP). Assist ratio for both systems was set at 1:2. Note the burst artifact from the muscle stimulator at every second beat in the electrocardiographic (EKG) tracing.
(Fig 4). The mean increase was 34 mm Hg, which represented a 69% increase from baseline. Systolic pressure decrease ranged from 0 to 20 mm Hg (Fig 5 ) . The mean decrease was 10 mm Hg, which represented a 10% decrease from baseline. When analyzed by the paired t test, both diastolic and systolic changes were significant ( p < 0.001) (Table 1).
of the stimulator. At no instance was there any systolic interference seen from mistimed counterpulsation. Tracings from a dog in which heart rate was varied while undergoing counterpulsation are shown in Figure 6.
Arrhythmia In 2 dogs, ventricular tachycardia was induced by mechanical irritation of the heart during counterpulsation. In both instances, the stimulator immediately shut down and resumed function only after a normal rhythm was reestablished. No systolic interference was observed during this period of arrhythmia. A tracing during this period of arrhythmia is demonstrated in Figure 7.
Heart Rate Changes By using a P-blocker (esmolol) and a P-agonist (isoproterenol), heart rate changes from 70 to 194 bpm were obtained. Throughout this range, good counterpulsation was maintained without any need for timing adjustment Diastolic Pressure (mmW
Systolic Pressure I
. I "
IZ0
0'
Off
On
I
*p
Fig 4. Comparison of diastolic pressures with the implantable counterpulsation assist system off and on. Error bars represent mean values 2 standard deviation. The difference was statistically significant (p < 0.001 by paired t test). See Table 1 for further details.
3
Off
On
*p<0.00 1
Fig 5. Comparison of systolic pressures with the implantable counterpulsation assist system off and on. Error bars represent mean values 2 standard deviation. The difference was statistically significant (p < 0.001 by paired t test). See Table 1 for further details.
Ann Thorac Surg 1990;49:35&62
Fig 6 . Electrocardiographic and arterial tracings at three different heart rates (HRs). Note as the heart rate decreases, the RR interval increases. Delay between onset of pulse bursts and duration of bursts also increase proportionately. Optimal timing of counterpulsation is maintained with no systolic interference.
Back-up Intraaortic Balloon Pump Console In 1 animal, an intraaortic balloon pump (IABP) power console was hooked to the ICAS. Comparable counterpulsation assist was obtained with either the muscle or the IABP console. Tracings of these two modes of power source are shown in Figure 8.
Comment In the late 1960s, Kantrowitz [lo] began reporting the first successful clinical cases of intraaortic balloon pumping. Since then, the IABP has become the most extensively used form of circulatory assist. The physiological benefit of counterpulsation derives from the favorable balance between myocardial supply and demand ihrough diastolic
LIETAL MUSCLE-POWERED COUNTERPULSATION
359
augmentation and systolic unloading [ll,121. In our study, we attempted to emulate these effects of diastolic augmentation and systolic unloading with a muscle-powered ICAS. Muscle-powered cardiac assist is now feasible because of two recently introduced concepts. They include the fact that skeletal muscle can be made fatigue-resistant for continuous work and that with pulse-train stimulation, enough force can be generated for effective cardiac assistance [13, 141. Since the early 1980s, we have attempted to apply this biomechanical energy source in a counterpulsation assist system [15, 161. In 1988, Kochamba and coworkers [8] introduced the concept of a dual-chambered extraaortic pump for skeletal muscle-powered counterpulsation. In our ICAS, we used a slightly modified version of this pump that provides improved hemodynamic flow and allows attachment to an IABP console. A stimulator (Prometheus) recently developed for the ICAS by Medtronic, Inc, provides synchronized contraction of the muscle with the diastolic phase of the cardiac cycle [9]. Using this new stimulator and the dual-chambered pump, we tested the ICAS in an animal model. Our results demonstrate that the ICAS is capable of counterpulsation assist equivalent to the IABP. In every animal, significant diastolic pressure augmentation was achieved. Average diastolic augmentation was nearly 70%. In addition, peak systolic pressures were decreased by an average of 10 mm Hg. This is equivalent to the degree of systolic unloading reported in other studies on the intraaortic balloon [ l l , 171. By reproducing the effects of the intraaortic balloon on diastolic/systolic pressure contours, the benefit of the ICAS can be inferred. For a counterpulsation system to function independently, it must be able to adjust the timing according to changing heart rates. The newly developed Medtronic muscle stimulator serves this purpose [9]. With a built-in microprocessor, the stimulator adjusts the delay and duration intervals of its burst pulses according to the sensed RR intervals. Therefore, as the RR interval changes with heart rate, the stimulator senses these changes and adjusts the delaylduration intervals to maintain optimal timing. Through preliminary trials, we found that excellent counterpulsation could be achieved by setting the burst delay at 30% and burst duration at 60% of the RR interval. The determination of optimal timing of counterpulsation was made by visual inspection of the pressure tracings, where there was good diastolic augmentation starting near the dicrotic notch, associated with appropriate systolic unloading. All our tracings demonstrated these features. Throughout our study, we maintained this setting of 30%/60%in all animals. The consistently optimal counterpulsation obtained in every animal attests to the efficacy of this system. In some animals, heart rate changes were deliberately effected by infusion of either pblocker or P-agonist. Automatic adjustments of delay and duration occurred, and optimal timing was maintained. We did not adjust the timing of stimulation with the different heart rates because the main advantage of the ICAS is its ability to automatically self-adjust according to spontaneous rate changes. Pressure tracings at various heart rates demonstrated that excellent counterpulsation had been maintained.
360
LIETAL MUSCLE-POWERED COUNTERPULSATION
Ann Thorac 1990;49:3%42
Fig 7 . Response of implantable counterpulsation assist system to ventricular arrhythmia. Implantable counterpulsation assist system is providing assist at 1:2 ratio when ventricular tachycardia is induced (arrow). The stimulator shuts down immediately and resumes function only after sinus rhythm is reestablished. No systolic interference is observed during this period. (EKG = electrocardiogram.)
The stimulator possesses a safety feature against cardiac arrhythmias. With sudden large changes in RR intervals, as occurring in arrhythmias, the stimulator automatically shuts down until a more regular rhythm is reestablished. We studied this safety feature in 2 animals in which ventricular tachycardia was induced. In both instances, the stimulator promptly shut down on induction of the arrhythmia and resumed function only after a normal
Fig 8. Comparison of counterpulsation obtained with either the muscle or the intraaortic balloon pump (IABP) console. Equivalent assist is achieved with either system powering the pump. (EKG = electrocardiogram.)
rhythm was reestablished. There was never any evidence of systolic interference due to mistiming during this period of arrhythmia. The main goal of the ICAS is to exploit the potential source of energy from skeletal muscle. However, the possibility exists for the muscle or stimulator system to fail. Should such a situation arise, it is important to have an accessible external back-up power source for the extraaortic pump. We determined the feasibility of using a readily available IABP power console to drive our extraaortic pump to maintain counterpulsation. With minor modification of the pneumatic system, we could easily attach a balloon console to drive the system. In 1 animal, we alternated driving the pump with either the muscle or the power console. Comparably excellent counterpulsation was obtained with either system. Therefore, effective back-up can be easily achieved with an IABP console for the ICAS. In addition, patients requiring immediate circulatory support would benefit from temporary console assistance to power the pump while their muscle undergoes transformation into a fatigue-resistant form. With the recent advances in artificial assist devices, it is important to define a role for the ICAS. It is not our purpose to introduce the ICAS as a substitute for the ventricular assist devices presently being developed. We envision the ICAS as providing an alternative therapeutic modality in the wide spectrum represented by "endstage" heart failure. At one end, there is the patient manageable with medicinal modes; whereas on the other end, we have someone requiring immediate dramatic intervention with a mechanical ventricle. In between these two extremes, there is the patient refractory to medical therapy but with residual function not requiring total ventricular support. This group of patients, in whom intermittent reports are being described of successful chronic IABP support, would benefit from a system like
Ann Thorac Surg 1990;49:356-62
the ICAS. At this time, we are continuing with the chronic phase of the study in evaluating the ICAS and attempting to further refine its components. This study was supported by a research grant from the Medical Research Council of Canada.
References 1. Gaul G, Blazek G, Deutsch M, et al. Chronic use of an intraaortic balloon pump in congestive cardiomyopathy. In: Unger F, ed. Assisted circulation. Berlin: Springer-Verlag, 1984: 28-37. 2. Disler PB, Millar RNS, Obel IWP. Prolonged circulatory support with the intra-aortic balloon pump after myocardial infarction. Thorax 1978;33:50&7. 3. Ashar B, Turcotte LR. Analyses of longest IAB implant in human patient (327 days). Trans Am SOCArtif Intern Organs 1981;27:372-9. 4. Rubenfire M, Krakauer J, Ciborski M, et al. Prolonged circulatory support by intraaortic balloon pumping [Abstract]. Circulation 1972;4546(Suppl 2):214. 5. Chiu RC-J, Neilson IR, Khalafalla AS. The rationale for skeletal muscle-powered counterpulsation devices: an overview. J Cardiovasc Surg 1986;1:385-92. 6. Chiu RC-J. Biomechanical cardiac assist: cardiomyoplasty and muscle-powered devices. Mount Kisco, NY: Futura Publishing, 1986. 7. Acker MA, Hammond RL, Mannion JD, et al. An autologous biologic pump motor. J Thorac Cardiovasc Surg 1986;92:73346. 8. Kochamba G, Desrosiers C, Dewar M, et al. The musclepowered dual-chamber counterpulsator: rheologically supe-
LIETAL
MUSCLE-POWERED COUNTERPULSATION
361
rior implantable cardiac assist device. Ann Thorac Surg 1988; 45:620-5. 9. Li CM, Hill A, Desrosiers C, et al. A new implantable burst generator for skeletal muscle powered aortic counterpulsation. Trans Am SOCArtif Intern Organs 1989;35:4057. 10. Kantrowitz A, Tjenneland S, Freed PS, et al. Initial clinical experience with intraaortic balloon pumping in cardiogenic shock. JAMA 1968;203:113-8. 11. Weber KT, Janicki JS. Intraaortic balloon counterpulsation. Ann Thorac Surg 1974;17602-36. 12. Powell WJ, Daggett WM, Magro AE, et al. Effects of intraaortic balloon counterpulsation on cardiac performance, oxygen consumption, and coronary blood flow in dogs. Circ Res 1970;26:75344. 13. Macoviak JA, Stephenson LW, Armenti F, et al. Electrical conditioning of in situ skeletal muscle for replacement of myocardium. J Surg Res 1982;32:429-39. 14. Dewar ML, Drinkwater DC, Wittnich C, et al. Synchronously stimulated skeletal muscle graft for myocardial repair. J Thorac Cardiovasc Surg 1984;87:325-31. 15. Neilson IR, Brister SJ, Khalafalla AS, et al. Left ventricular assistance in dogs using a skeletal muscle powered device for diastolic augmentation. Heart Transplant 1985;6:34>7. 16. Neilson IR, Chiu RCJ. Skeletal muscle-powered cardiac assist using an extra-aortic balloon pump. In: Chiu RC-J, ed. Biomechanical cardiac assist: cardiomyoplasty and musclepowered devices. Mount Kisco, NY: Futura Publishing, 1986: 141-50. 17. Mueller H, Ayres SM, Conklin EF, et al. The effects of intraaortic counterpulsation on cardiac performance and metabolism in shock associated with acute myocardial infarction. J Clin Invest 1971;50:1885900.
DISCUSSION D R ADRIAN KANTROWITZ (Detroit, MI): First of all, if you are going to have a pumping chamber that is in parallel with the aorta and approximately the same diameter, presumably, then the flow through both the natural aorta and the parallel aorta will be reduced by half. That then raises the question of clotting on a long-term basis, and I think that is not a trivial problem. Let me put it this way. I have been concerned with a device very similar to this and have been struggling with the clotting problem for about 25 years. Now, let’s assume you are twice as smart as I am; this is a problem that you may have to think about for 10 years. Second, presumably if you are going to have the possibility of plugging into an intraaortic balloon console, you are going to drive this with a gas such as air or helium. Does this imply that the transfer fluid in your system is a gas? If that were true, then you certainly could plug in an intraaortic balloon pump, but it is far more efficient to do this with a fluid, a liquid, in which case it would be difficult to plug an intraaortic balloon pump in. I appreciate your efforts and I do hope that you will continue with this.
longest survivor we have had was for two days, and the problem is, as Dr Kantrowitz mentioned, that of thromboembolism. It is a problem that we are faced with and are trying to deal with. I do not know if I have the answers to this problem, but there is a lot of technology available out there, and hopefully we will be able to take advantage of some of that knowledge. The second issue is what is the best medium to drive the system. Dr Kantrowitz stated that a fluid system would be much more efficient than a pneumatic system. We used air in our system, so we could easily attach a balloon pump console to it. The argument that I have heard is that in a pneumatic system, air is compressible and therefore you lose a lot of energy as you try to ”compress” the air. However, our system already has a resting pressure of approximately 60% to 70% of systemic pressure, and in that case the air is already compressed. We compared using fluid or air in this situation and found that the air actually provided better counterpulsation in that it evacuated the pump more quickly and provided better unloading, while providing equally good diastolic augmentation. Therefore, we prefer to use a pneumatic system.
DR LI: I think Dr Kantrowitz brought up several important points that we are dealing with. The issue that he first brought up was that of problems with thromboembolism in the pump. Right now we are proceeding with the chronic phase of the studies. Initially the problem was that we did not have an adequate stimulator, until now, that could provide optimal timing for counterpulsation. We have conducted some preliminary trials attempting to have chronic survivors with this system. The
DR KANTROWITZ: If you are going to fill the system with air, the air isn’t going to stay there very long; it is going to leak out of whatever you seal it into. Therefore, you are going to have to have an opening on the outside to add more air. If you are going to do that, you might as well have the pump on the outside. And let met say that we have had some experience with intraaortic balloon pumping. In reviewing about 700 patients, we had about 20 patients who were treated long-term with balloon pumps, 20
362
Ann Thorac Surg 1990;49:356-62
LIETAL MUSCLE-POWERED COUNTERPULSATION
to 71 days, and it was very discouraging. We could keep these patients on the balloon pump very well but the moment we sent them home, almost all of them died very rapidly in chronic heart failure. So, when the patient reaches that stage, he or she has to have more or less permanent support. DR LI: I agree with the statement that a continuous leak of air is an issue in a chronic model. In the animals that we have attempted to keep for a long amount of time, we introduced a subcutaneous injection port in order to fill the system with air. We found that we needed to fill the system with air almost twice a day. That is a problem, and perhaps we can deal with better materials or eventually go to a fluid mechanism. Our plan for this muscle-powered counterpulsation system is that it will be a permanent form of cardiac assistance. We envision in the future that the patient will be going home with the system functioning within him. DR KANTROWITZ: Not so easy, not so easy. DR W. GERALD RAINER (Presiding Officer): Nothing is. DR JOHN A. JACOBEY (Mountain Lakes, NJ): My first question is, is excellent counterpulsation achieved when you do not increase the diastolic pressure to a point at which it is higher than the systolic pressure? D R LI: My understanding of the physiology is that the higher the difference between aortic diastolic pressure and the left ventricular filling pressure, the better the coronary perfusion drive would be. So, in my opinion, I would think that optimally we would benefit from a system that could produce as much diastolic augmentation as possible. DR JACOBEY You may or may not be aware of our work in the past in which we showed diastolic pressures that were significantly higher than the systolic pressures, which I believe would be considered to be excellent counterpulsation. But, then, ours was an external device designed to have both systolic and diastolic effects and you are troubled with the difficulties of producing effects with internal apparatus. And along that same line, my second question has to do with the placing of your device. Is this in the descending thoracic aorta or the abdominal aorta? D R LI: It is in the descending thoracic aorta. D R JACOBEY So the pressure you apply to increase the diastolic pressure is a function of the volume and the rapidity of your device as it sits against the descending thoracic aorta, and then that pressure wave goes both distally and proximally so that by the time the pressure change reaches the coronary ostia, which is where you want to make that pressure change, there has been a great deal of diastolic expansion of the aortic wall. This leads to the question that arose in the report by Chachques and associates [l].Isn’t the reason why you would like to have those pressure changes made as near the coronary ostia as you can so that they do not have to travel through that much aorta and have that much dispersion of effect? We have presented results of cannula counterpulsation, which was the type of counterpulsation done before the balloon, in which the cannula was inserted in the left subclavian artery and brought down to a point 3 cm from the aortic valve. When
measuring pressure at the coronary ostia, not the aorta near where the intraaortic balloon is, we could produce diastolic pressures of 160 mm Hg and systolic pressures of 80 mm Hg. DR LI: I agree with the positioning of the pump: the closer you get to the root, the more advantageous it would be. I think that if we put the pump closer to the root, we may be able to get better diastolic augmentation for coronary flow. Our results that you saw were made with tracings from a catheter located in the ascending root of the aorta. DR GORDON N . OLINGER (Milwaukee, WI): I would like to pursue the question asked by Greg Misbach but not really fully answered from the paper by Chachques and associates [l],and that is the question of unloading. Peak systolic pressure is really not a good indicator of unloading. We are more interested in isovolumetric systole during peak developed tension within the ventricle, and as I examined your tracings from animal to animal at various rates of augmentation and unloading, some of them seemed to indicate that you dropped end-diastolic pressure whereas others seemed to indicate that you did not. Do you have any consistent pattern with respect to unloading as it relates to end-diastolic pressure, which is more a reflection of isovolumetric systole? DR LI: The effect of systolic unloading from counterpulsation on myocardial oxygen consumption is difficult to measure. We used the peak systolic pressure because we believed it was the best index of myocardial oxygen consumption. By measuring peak systolic pressure, we did find significant decreases in pressure. To consistently produce a presystolic dip is difficult with our system. There might be a slight variation in the RR interval between one and the next RR interval, and the system is unable to instantaneously adjust its burst duration to produce maximal unloading. This drawback is unavoidable because the stimulator depends on the previous RR interval to determine its burst duration. DR CHIU: I am sure that when Dr Kantrowitz was doing his pioneering work in the 1960s, someone told him it could not be done, and if he had stopped doing it, he would not be honored here today as a pioneer. Certainly, there are still many questions remaining regarding this approach, but we will continue to address them the best we can. The most important message in this paper is that the development of a totally implantable muscle-powered counterpulsation device requires a rate-responsive and arrhythmia-responsive muscle stimulator. This stimulator will be useful not only for our particular pump configuration, but also for others such as that reported by Chachques and associates [l].Using the endogenous power source of transformed skeletal muscle for cardiac assist is still a young field, and at this point, we are still not sure what is the best way to use it. We are continuing to explore various possibilities and hope these efforts will lead to effective therapeutic modes applicable to patients in the years to come.
Reference 1. Chachques JC, Grandjean PA, Cabrera Fischer EI, et al. Dynamic aortomyoplasty to assist cardiac failure. Ann Thorac Surg 1990;49:22!5-30.
Carlos M. Li, MD, Andrew Hill, MD, Michael Colson, MS, Carolyne Desrosiers, BS, and Ray C.-J. Chiu, MD, PhD Division of Cardiovascular and Thoracic Surgery, McGill University, Montreal, Canada, and Medtronic, Inc, Minneapolis, Minnesota
To apply the potential energy source available from skeletal muscle in cardiac assistance, we developed an implantable counterpulsation assist system. This study reports the results using this implantable counterpulsation assist system in an acute in vivo animal model. Twelve dogs had a dual-chambered, extraaortic counterpulsation pump anastomosed in parallel to the thoracic aorta. The left latissimus dorsi muscle was used to power the pump. A newly developed implantable stimulator was used to make the muscle contract in synchrony with the diastolic phase. The unique feature of this stimulator is its ability to adjust timing of muscle contraction according to changing heart rates. The stimulator is also able to detect arrhythmias, and as a safety measure, shuts down until a normal rhythm is resumed. During counterpulsation assist with the implantable counterpulsation assist system, diastolic pressure increased an aver-
age of 34 mm Hg from baseline, equivalent to a 69% augmentation. Systolic peak pressure decreased an average of 10 mm Hg, equivalent to an 11%unloading. With induced heart rate changes, the implantable counterpulsation assist system readjusted its timing, maintaining optimal counterpulsation without systolic interference. Induced ventricular tachycardia resulted in immediate shutdown of the stimulator until resumption of a normal rhythm. The feasibility of using an intraaortic balloon pump console as back-up was also demonstrated. Excellent counterpulsation was obtained with either muscle power or balloon pump console. We conclude that the implantable counterpulsation assist system can provide effective counterpulsation assist and has the potential for continuous cardiac support.
p i r c u l a t o r y counterpulsation with the intraaortic balloon is-a widely- accepted form of cardiac assist. Recently, reports of extended support with the balloon pump have confirmed its efficacy in the chronic situation [ 1 4 ] . In several cases, patients were supported from weeks to almost a year. Many more patients with endstage failure could benefit from chronic counterpulsation support. A major limitation is the patient’s dependency on an external power source with its risk of infection and restriction in mobility. A totally implantable counterpulsation assist system would offer an important therapeutic option for patients with end-stage heart failure. Recent advances in technology and skeletal muscle biology now allow for a viable alternative source in cardiac assistance [5, 61. It has been demonstrated that skeletal muscle can be made fatigue-resistant and powerful enough to continuously assist the heart [7]. We have attempted to apply this autologous power source to a totally enclosed circulatory support system. The main purpose of this investigation was to evaluate the efficacy of our implantable counterpulsation assist system (ICAS) in an acute in vivo model.
Material and Methods Extraaortic Dual Chambered Pump
Presented at the Twenty-fifth Anniversary Meeting of The Society of Thoracic Surgeons, Baltimore, MD, Sep 11-13, 1989. Address reprint requests to Dr Chiu, Montreal General Hospital, 1650 Cedar Ave, Room 947,Montreal, Que, Canada H3G 1A4.
0 1990 by The
Society of Thoracic Surgeons
(Ann Thoruc Surg 1990;49:356-62)
The muscle-driven counterpulsation pump used in this study is as previously described except for some slight modifications [8]. The inner blood sac of the chamber consists of a silicone bladder located within a rigid cylindrical housing. The outlets of the pump were 12 mm in diameter. The pump was powered pneumatically by the left latissimus dorsi muscle through compression of a bulb connected to the pump with Silastic (Dow Corning Corp, Midland, MI) tubing (Fig 1). A port in the Silastic tubing allows optional connection to an intraaortic balloon pump console.
Muscle Stimulator A new, microprocessor based, implantable pulse generator (Prometheus) was provided by Medtronic, Inc (Minneapolis, MN) [9]. This unit has a built-in microcomputer that is programmed through telemetry with software operated on an IBM-compatible personal computer. In our study, software developed by Medtronic in collaboration with us for synchronized counterpulsation was programmed into the stimulator. It allows for pulse burst stimulation of the muscle, synchronized to the cardiac R wave. Timing for onset delay of the pulse burst and its duration period is adjusted automatically by the stimulator according to the varying RR interval that occurs with heart rate changes. In addition, adjustment of the pulse 0003-4975/90/$3.50
LIETAL MUSCLE-POWERED COUNTERPULSATION
Ann Thorac Surg 1990;49:35&62
357
Amplitude: 2 V
Frequency: 33 Hz
Duration
~~l~~
IABP ConSol9
Fig 1 . Schematic representation of implantable counterpulsation assist system. A dual-chambered pump is anastomosed in parallel to the descending thoracic aorta. The latissimus dorsi contracts on a pneumatic bulb displacing a silicone membrane in the pump. The stimulator coordinates muscle contraction with the cardiac rhythm. An intruaortic balloon pump (IABP) console can be attached as an alternative source of power.
voltage, frequency, and assist ratio can be easily made on the personal computer and sent via telemetry. A safety feature available with this pulse generator is an automatic shutdown of stimulation should any cardiac arrhythmias occur. Twelve mongrel dogs of approximately 30 kg were induced with pentobarbital and maintained under 0.5%to 1% halothane endotracheal anesthesia. Through a left thoracotomy, the descending thoracic aorta was exposed, and a dual-chambered pump was anastomosed in parallel to the vessel with 12-mm Dacron grafts. Heparin (100 U/kg) was infused to partially anticoagulate the animal. A collapsible bulb was placed underneath the left latissimus dorsi, and the bulb was connected to the pump with Silastic tubing. The left thoracodorsal nerve was dissected free from the neurovascular bundle, and a bipolar neural cuff lead was placed around it. The lead was then connected to the stimulator. For sensing, epicardial leads were attached onto the myocardium and connected to the stimulator. All animals received humane care in compliance with the ”Guide for the Care and Use of Laboratory Animals” published by the National Institutes of Health (NIH publication No. 85-23, revised 1985) and the Canadian Council on Animal Care. The stimulator was programmed to deliver pulse bursts of 2 V, 33 Hz frequency, and a pulse width of 210 ps. The delay period between triggering R wave and onset of stimulus was set to equal 30% of the RR interval, whereas duration of the stimulus was set to equal 60% of the RR interval (Fig 2). These parameters offered optimal timing and contraction force of the muscle for counterpulsation assist. The assist ratio during the studies ranged from 1:l to 3:l. A Millar catheter (Millar Instruments, Houston, TX) pressure transducer was inserted through the left carotid artery into the aortic arch for hemodynamic monitoring.
60% R-R
Fig 2 . Parameter settings of the muscle stimulator. The burst pulses were set at 33 H z frequency and 2 V amplitude. Timing of the bursts was determined by the sensed RR interval of the electrocardiogram (EKG). Delay interval between triggering R wave and burst onset was equivalent to 30% of the RR interval. Duration of burst was 60% of the RR interval.
Studies were performed with the stimulator on and off to observe the effects of counterpulsation on arterial pressure. Each animal served as its own control for comparison. The change in diastolic and systolic pressures were compared using paired t tests for significance. To evaluate the system’s capability to track at different heart rates, esmolol or isoproterenol, or both, was infused as needed to vary the rhythm. The original stimulator settings were maintained throughout these rate changes, and arterial pressures were continuously monitored. The system’s response to acute cardiac dysrhythmias was also tested in 2 dogs. While undergoing counterpulsation assist, the surface of the heart was manually irritated, inducing ventricular ectopy. The response to this insult was monitored continuously throughout the episode. Finally, we tested the feasibility of using an intraaortic balloon pump console (Datascope 82, Paramus, NJ) to drive the dual-chambered pump in 1 dog. This was achieved by connecting the gas outlet of the console to a side port of the Silastic tubing, which connects the pneumatic bulb to the pump (Fig 1). By temporarily clamping and excluding the pneumatic bulb from the system, we were able to drive the extraaortic pump with 40- to 50-mL gas volumes from the console.
Results
Hemodynamic Counterpulsation Arterial counterpulsation was observed in all 12 dogs. A typical tracing obtained with the ICAS is shown in Figure 3. Diastolic pressure increase ranged from 20 to 50 mm Hg
Table 1. Pressure Results in 12 Dogs Pressure Diastolic (mm Hg) Systolic (mm Hg) a
Off
On
Augmentation
51.3 ? 14.3 85.8 ? 14.2
83.0 ? 15.5a 77.1 +- 12.3”
34 (69%) -10 (-11%)
Significance: p < 0.001 versus pump off.
358
LIETAL MUSCLE-POWEREDCOUNTERPULSATION
Ann Thorac Surg 1990;49:356-62
Fig 3. Representative arterial tracing with the implantable counterpulsation assist system (ICAS) compared with that with an intraaortic balloon pump (IABP). Assist ratio for both systems was set at 1:2. Note the burst artifact from the muscle stimulator at every second beat in the electrocardiographic (EKG) tracing.
(Fig 4). The mean increase was 34 mm Hg, which represented a 69% increase from baseline. Systolic pressure decrease ranged from 0 to 20 mm Hg (Fig 5 ) . The mean decrease was 10 mm Hg, which represented a 10% decrease from baseline. When analyzed by the paired t test, both diastolic and systolic changes were significant ( p < 0.001) (Table 1).
of the stimulator. At no instance was there any systolic interference seen from mistimed counterpulsation. Tracings from a dog in which heart rate was varied while undergoing counterpulsation are shown in Figure 6.
Arrhythmia In 2 dogs, ventricular tachycardia was induced by mechanical irritation of the heart during counterpulsation. In both instances, the stimulator immediately shut down and resumed function only after a normal rhythm was reestablished. No systolic interference was observed during this period of arrhythmia. A tracing during this period of arrhythmia is demonstrated in Figure 7.
Heart Rate Changes By using a P-blocker (esmolol) and a P-agonist (isoproterenol), heart rate changes from 70 to 194 bpm were obtained. Throughout this range, good counterpulsation was maintained without any need for timing adjustment Diastolic Pressure (mmW
Systolic Pressure I
. I "
IZ0
0'
Off
On
I
*p
Fig 4. Comparison of diastolic pressures with the implantable counterpulsation assist system off and on. Error bars represent mean values 2 standard deviation. The difference was statistically significant (p < 0.001 by paired t test). See Table 1 for further details.
3
Off
On
*p<0.00 1
Fig 5. Comparison of systolic pressures with the implantable counterpulsation assist system off and on. Error bars represent mean values 2 standard deviation. The difference was statistically significant (p < 0.001 by paired t test). See Table 1 for further details.
Ann Thorac Surg 1990;49:35&62
Fig 6 . Electrocardiographic and arterial tracings at three different heart rates (HRs). Note as the heart rate decreases, the RR interval increases. Delay between onset of pulse bursts and duration of bursts also increase proportionately. Optimal timing of counterpulsation is maintained with no systolic interference.
Back-up Intraaortic Balloon Pump Console In 1 animal, an intraaortic balloon pump (IABP) power console was hooked to the ICAS. Comparable counterpulsation assist was obtained with either the muscle or the IABP console. Tracings of these two modes of power source are shown in Figure 8.
Comment In the late 1960s, Kantrowitz [lo] began reporting the first successful clinical cases of intraaortic balloon pumping. Since then, the IABP has become the most extensively used form of circulatory assist. The physiological benefit of counterpulsation derives from the favorable balance between myocardial supply and demand ihrough diastolic
LIETAL MUSCLE-POWERED COUNTERPULSATION
359
augmentation and systolic unloading [ll,121. In our study, we attempted to emulate these effects of diastolic augmentation and systolic unloading with a muscle-powered ICAS. Muscle-powered cardiac assist is now feasible because of two recently introduced concepts. They include the fact that skeletal muscle can be made fatigue-resistant for continuous work and that with pulse-train stimulation, enough force can be generated for effective cardiac assistance [13, 141. Since the early 1980s, we have attempted to apply this biomechanical energy source in a counterpulsation assist system [15, 161. In 1988, Kochamba and coworkers [8] introduced the concept of a dual-chambered extraaortic pump for skeletal muscle-powered counterpulsation. In our ICAS, we used a slightly modified version of this pump that provides improved hemodynamic flow and allows attachment to an IABP console. A stimulator (Prometheus) recently developed for the ICAS by Medtronic, Inc, provides synchronized contraction of the muscle with the diastolic phase of the cardiac cycle [9]. Using this new stimulator and the dual-chambered pump, we tested the ICAS in an animal model. Our results demonstrate that the ICAS is capable of counterpulsation assist equivalent to the IABP. In every animal, significant diastolic pressure augmentation was achieved. Average diastolic augmentation was nearly 70%. In addition, peak systolic pressures were decreased by an average of 10 mm Hg. This is equivalent to the degree of systolic unloading reported in other studies on the intraaortic balloon [ l l , 171. By reproducing the effects of the intraaortic balloon on diastolic/systolic pressure contours, the benefit of the ICAS can be inferred. For a counterpulsation system to function independently, it must be able to adjust the timing according to changing heart rates. The newly developed Medtronic muscle stimulator serves this purpose [9]. With a built-in microprocessor, the stimulator adjusts the delay and duration intervals of its burst pulses according to the sensed RR intervals. Therefore, as the RR interval changes with heart rate, the stimulator senses these changes and adjusts the delaylduration intervals to maintain optimal timing. Through preliminary trials, we found that excellent counterpulsation could be achieved by setting the burst delay at 30% and burst duration at 60% of the RR interval. The determination of optimal timing of counterpulsation was made by visual inspection of the pressure tracings, where there was good diastolic augmentation starting near the dicrotic notch, associated with appropriate systolic unloading. All our tracings demonstrated these features. Throughout our study, we maintained this setting of 30%/60%in all animals. The consistently optimal counterpulsation obtained in every animal attests to the efficacy of this system. In some animals, heart rate changes were deliberately effected by infusion of either pblocker or P-agonist. Automatic adjustments of delay and duration occurred, and optimal timing was maintained. We did not adjust the timing of stimulation with the different heart rates because the main advantage of the ICAS is its ability to automatically self-adjust according to spontaneous rate changes. Pressure tracings at various heart rates demonstrated that excellent counterpulsation had been maintained.
360
LIETAL MUSCLE-POWERED COUNTERPULSATION
Ann Thorac 1990;49:3%42
Fig 7 . Response of implantable counterpulsation assist system to ventricular arrhythmia. Implantable counterpulsation assist system is providing assist at 1:2 ratio when ventricular tachycardia is induced (arrow). The stimulator shuts down immediately and resumes function only after sinus rhythm is reestablished. No systolic interference is observed during this period. (EKG = electrocardiogram.)
The stimulator possesses a safety feature against cardiac arrhythmias. With sudden large changes in RR intervals, as occurring in arrhythmias, the stimulator automatically shuts down until a more regular rhythm is reestablished. We studied this safety feature in 2 animals in which ventricular tachycardia was induced. In both instances, the stimulator promptly shut down on induction of the arrhythmia and resumed function only after a normal
Fig 8. Comparison of counterpulsation obtained with either the muscle or the intraaortic balloon pump (IABP) console. Equivalent assist is achieved with either system powering the pump. (EKG = electrocardiogram.)
rhythm was reestablished. There was never any evidence of systolic interference due to mistiming during this period of arrhythmia. The main goal of the ICAS is to exploit the potential source of energy from skeletal muscle. However, the possibility exists for the muscle or stimulator system to fail. Should such a situation arise, it is important to have an accessible external back-up power source for the extraaortic pump. We determined the feasibility of using a readily available IABP power console to drive our extraaortic pump to maintain counterpulsation. With minor modification of the pneumatic system, we could easily attach a balloon console to drive the system. In 1 animal, we alternated driving the pump with either the muscle or the power console. Comparably excellent counterpulsation was obtained with either system. Therefore, effective back-up can be easily achieved with an IABP console for the ICAS. In addition, patients requiring immediate circulatory support would benefit from temporary console assistance to power the pump while their muscle undergoes transformation into a fatigue-resistant form. With the recent advances in artificial assist devices, it is important to define a role for the ICAS. It is not our purpose to introduce the ICAS as a substitute for the ventricular assist devices presently being developed. We envision the ICAS as providing an alternative therapeutic modality in the wide spectrum represented by "endstage" heart failure. At one end, there is the patient manageable with medicinal modes; whereas on the other end, we have someone requiring immediate dramatic intervention with a mechanical ventricle. In between these two extremes, there is the patient refractory to medical therapy but with residual function not requiring total ventricular support. This group of patients, in whom intermittent reports are being described of successful chronic IABP support, would benefit from a system like
Ann Thorac Surg 1990;49:356-62
the ICAS. At this time, we are continuing with the chronic phase of the study in evaluating the ICAS and attempting to further refine its components. This study was supported by a research grant from the Medical Research Council of Canada.
References 1. Gaul G, Blazek G, Deutsch M, et al. Chronic use of an intraaortic balloon pump in congestive cardiomyopathy. In: Unger F, ed. Assisted circulation. Berlin: Springer-Verlag, 1984: 28-37. 2. Disler PB, Millar RNS, Obel IWP. Prolonged circulatory support with the intra-aortic balloon pump after myocardial infarction. Thorax 1978;33:50&7. 3. Ashar B, Turcotte LR. Analyses of longest IAB implant in human patient (327 days). Trans Am SOCArtif Intern Organs 1981;27:372-9. 4. Rubenfire M, Krakauer J, Ciborski M, et al. Prolonged circulatory support by intraaortic balloon pumping [Abstract]. Circulation 1972;4546(Suppl 2):214. 5. Chiu RC-J, Neilson IR, Khalafalla AS. The rationale for skeletal muscle-powered counterpulsation devices: an overview. J Cardiovasc Surg 1986;1:385-92. 6. Chiu RC-J. Biomechanical cardiac assist: cardiomyoplasty and muscle-powered devices. Mount Kisco, NY: Futura Publishing, 1986. 7. Acker MA, Hammond RL, Mannion JD, et al. An autologous biologic pump motor. J Thorac Cardiovasc Surg 1986;92:73346. 8. Kochamba G, Desrosiers C, Dewar M, et al. The musclepowered dual-chamber counterpulsator: rheologically supe-
LIETAL
MUSCLE-POWERED COUNTERPULSATION
361
rior implantable cardiac assist device. Ann Thorac Surg 1988; 45:620-5. 9. Li CM, Hill A, Desrosiers C, et al. A new implantable burst generator for skeletal muscle powered aortic counterpulsation. Trans Am SOCArtif Intern Organs 1989;35:4057. 10. Kantrowitz A, Tjenneland S, Freed PS, et al. Initial clinical experience with intraaortic balloon pumping in cardiogenic shock. JAMA 1968;203:113-8. 11. Weber KT, Janicki JS. Intraaortic balloon counterpulsation. Ann Thorac Surg 1974;17602-36. 12. Powell WJ, Daggett WM, Magro AE, et al. Effects of intraaortic balloon counterpulsation on cardiac performance, oxygen consumption, and coronary blood flow in dogs. Circ Res 1970;26:75344. 13. Macoviak JA, Stephenson LW, Armenti F, et al. Electrical conditioning of in situ skeletal muscle for replacement of myocardium. J Surg Res 1982;32:429-39. 14. Dewar ML, Drinkwater DC, Wittnich C, et al. Synchronously stimulated skeletal muscle graft for myocardial repair. J Thorac Cardiovasc Surg 1984;87:325-31. 15. Neilson IR, Brister SJ, Khalafalla AS, et al. Left ventricular assistance in dogs using a skeletal muscle powered device for diastolic augmentation. Heart Transplant 1985;6:34>7. 16. Neilson IR, Chiu RCJ. Skeletal muscle-powered cardiac assist using an extra-aortic balloon pump. In: Chiu RC-J, ed. Biomechanical cardiac assist: cardiomyoplasty and musclepowered devices. Mount Kisco, NY: Futura Publishing, 1986: 141-50. 17. Mueller H, Ayres SM, Conklin EF, et al. The effects of intraaortic counterpulsation on cardiac performance and metabolism in shock associated with acute myocardial infarction. J Clin Invest 1971;50:1885900.
DISCUSSION D R ADRIAN KANTROWITZ (Detroit, MI): First of all, if you are going to have a pumping chamber that is in parallel with the aorta and approximately the same diameter, presumably, then the flow through both the natural aorta and the parallel aorta will be reduced by half. That then raises the question of clotting on a long-term basis, and I think that is not a trivial problem. Let me put it this way. I have been concerned with a device very similar to this and have been struggling with the clotting problem for about 25 years. Now, let’s assume you are twice as smart as I am; this is a problem that you may have to think about for 10 years. Second, presumably if you are going to have the possibility of plugging into an intraaortic balloon console, you are going to drive this with a gas such as air or helium. Does this imply that the transfer fluid in your system is a gas? If that were true, then you certainly could plug in an intraaortic balloon pump, but it is far more efficient to do this with a fluid, a liquid, in which case it would be difficult to plug an intraaortic balloon pump in. I appreciate your efforts and I do hope that you will continue with this.
longest survivor we have had was for two days, and the problem is, as Dr Kantrowitz mentioned, that of thromboembolism. It is a problem that we are faced with and are trying to deal with. I do not know if I have the answers to this problem, but there is a lot of technology available out there, and hopefully we will be able to take advantage of some of that knowledge. The second issue is what is the best medium to drive the system. Dr Kantrowitz stated that a fluid system would be much more efficient than a pneumatic system. We used air in our system, so we could easily attach a balloon pump console to it. The argument that I have heard is that in a pneumatic system, air is compressible and therefore you lose a lot of energy as you try to ”compress” the air. However, our system already has a resting pressure of approximately 60% to 70% of systemic pressure, and in that case the air is already compressed. We compared using fluid or air in this situation and found that the air actually provided better counterpulsation in that it evacuated the pump more quickly and provided better unloading, while providing equally good diastolic augmentation. Therefore, we prefer to use a pneumatic system.
DR LI: I think Dr Kantrowitz brought up several important points that we are dealing with. The issue that he first brought up was that of problems with thromboembolism in the pump. Right now we are proceeding with the chronic phase of the studies. Initially the problem was that we did not have an adequate stimulator, until now, that could provide optimal timing for counterpulsation. We have conducted some preliminary trials attempting to have chronic survivors with this system. The
DR KANTROWITZ: If you are going to fill the system with air, the air isn’t going to stay there very long; it is going to leak out of whatever you seal it into. Therefore, you are going to have to have an opening on the outside to add more air. If you are going to do that, you might as well have the pump on the outside. And let met say that we have had some experience with intraaortic balloon pumping. In reviewing about 700 patients, we had about 20 patients who were treated long-term with balloon pumps, 20
362
Ann Thorac Surg 1990;49:356-62
LIETAL MUSCLE-POWERED COUNTERPULSATION
to 71 days, and it was very discouraging. We could keep these patients on the balloon pump very well but the moment we sent them home, almost all of them died very rapidly in chronic heart failure. So, when the patient reaches that stage, he or she has to have more or less permanent support. DR LI: I agree with the statement that a continuous leak of air is an issue in a chronic model. In the animals that we have attempted to keep for a long amount of time, we introduced a subcutaneous injection port in order to fill the system with air. We found that we needed to fill the system with air almost twice a day. That is a problem, and perhaps we can deal with better materials or eventually go to a fluid mechanism. Our plan for this muscle-powered counterpulsation system is that it will be a permanent form of cardiac assistance. We envision in the future that the patient will be going home with the system functioning within him. DR KANTROWITZ: Not so easy, not so easy. DR W. GERALD RAINER (Presiding Officer): Nothing is. DR JOHN A. JACOBEY (Mountain Lakes, NJ): My first question is, is excellent counterpulsation achieved when you do not increase the diastolic pressure to a point at which it is higher than the systolic pressure? D R LI: My understanding of the physiology is that the higher the difference between aortic diastolic pressure and the left ventricular filling pressure, the better the coronary perfusion drive would be. So, in my opinion, I would think that optimally we would benefit from a system that could produce as much diastolic augmentation as possible. DR JACOBEY You may or may not be aware of our work in the past in which we showed diastolic pressures that were significantly higher than the systolic pressures, which I believe would be considered to be excellent counterpulsation. But, then, ours was an external device designed to have both systolic and diastolic effects and you are troubled with the difficulties of producing effects with internal apparatus. And along that same line, my second question has to do with the placing of your device. Is this in the descending thoracic aorta or the abdominal aorta? D R LI: It is in the descending thoracic aorta. D R JACOBEY So the pressure you apply to increase the diastolic pressure is a function of the volume and the rapidity of your device as it sits against the descending thoracic aorta, and then that pressure wave goes both distally and proximally so that by the time the pressure change reaches the coronary ostia, which is where you want to make that pressure change, there has been a great deal of diastolic expansion of the aortic wall. This leads to the question that arose in the report by Chachques and associates [l].Isn’t the reason why you would like to have those pressure changes made as near the coronary ostia as you can so that they do not have to travel through that much aorta and have that much dispersion of effect? We have presented results of cannula counterpulsation, which was the type of counterpulsation done before the balloon, in which the cannula was inserted in the left subclavian artery and brought down to a point 3 cm from the aortic valve. When
measuring pressure at the coronary ostia, not the aorta near where the intraaortic balloon is, we could produce diastolic pressures of 160 mm Hg and systolic pressures of 80 mm Hg. DR LI: I agree with the positioning of the pump: the closer you get to the root, the more advantageous it would be. I think that if we put the pump closer to the root, we may be able to get better diastolic augmentation for coronary flow. Our results that you saw were made with tracings from a catheter located in the ascending root of the aorta. DR GORDON N . OLINGER (Milwaukee, WI): I would like to pursue the question asked by Greg Misbach but not really fully answered from the paper by Chachques and associates [l],and that is the question of unloading. Peak systolic pressure is really not a good indicator of unloading. We are more interested in isovolumetric systole during peak developed tension within the ventricle, and as I examined your tracings from animal to animal at various rates of augmentation and unloading, some of them seemed to indicate that you dropped end-diastolic pressure whereas others seemed to indicate that you did not. Do you have any consistent pattern with respect to unloading as it relates to end-diastolic pressure, which is more a reflection of isovolumetric systole? DR LI: The effect of systolic unloading from counterpulsation on myocardial oxygen consumption is difficult to measure. We used the peak systolic pressure because we believed it was the best index of myocardial oxygen consumption. By measuring peak systolic pressure, we did find significant decreases in pressure. To consistently produce a presystolic dip is difficult with our system. There might be a slight variation in the RR interval between one and the next RR interval, and the system is unable to instantaneously adjust its burst duration to produce maximal unloading. This drawback is unavoidable because the stimulator depends on the previous RR interval to determine its burst duration. DR CHIU: I am sure that when Dr Kantrowitz was doing his pioneering work in the 1960s, someone told him it could not be done, and if he had stopped doing it, he would not be honored here today as a pioneer. Certainly, there are still many questions remaining regarding this approach, but we will continue to address them the best we can. The most important message in this paper is that the development of a totally implantable muscle-powered counterpulsation device requires a rate-responsive and arrhythmia-responsive muscle stimulator. This stimulator will be useful not only for our particular pump configuration, but also for others such as that reported by Chachques and associates [l].Using the endogenous power source of transformed skeletal muscle for cardiac assist is still a young field, and at this point, we are still not sure what is the best way to use it. We are continuing to explore various possibilities and hope these efforts will lead to effective therapeutic modes applicable to patients in the years to come.
Reference 1. Chachques JC, Grandjean PA, Cabrera Fischer EI, et al. Dynamic aortomyoplasty to assist cardiac failure. Ann Thorac Surg 1990;49:22!5-30.