Ann Thorac Surg 1999;67:105-111
© 1999 The Society of Thoracic Surgeons
Original Articles
Evaluation of a compressive-type skeletal muscle pump for cardiac assistance
Hisao Mizuhara, MDa,
Takaaki Koshiji, MDa,
Kazunobu Nishimura, MDa,
Shin-ichi Nomoto, MDa,
Katsuhiko Matsuda, MD, PhDa,
Toshihiko Ban, MDa
a Department of Cardiovascular Surgery, Kyoto University, Kyoto, Japan
Accepted for publication June 22, 1998.
Address reprint requests to Dr Mizuhara, Department of Cardiovascular Surgery, Rakuwakai Otowa Hospital, 2, Otowachinjicho, Yamashina-ku, Kyoto 607-8062, Japan
 |
Abstract
|
|---|
Background. Recent investigations have focused on using the latissimus dorsi muscle for cardiac assistance. Although cardiomyoplasty has been applied clinically, other procedures remain experimental, but promising, modes of cardiac assistance. We assessed the latissimus dorsi muscle as an in situ energy source for circulatory assist devices.
Methods. We developed a pneumatic chamber as a compressive-type muscle actuator. The chamber was implanted under the latissimus dorsi muscle and converted contractile power into pneumatic pressure. The effect of chamber position and size and the influence on muscle blood flow were examined. After muscle conditioning, the pump performance of a circulatory assist device driven by the chamber was evaluated.
Results. The chamber functioned better when placed in the proximal position of the latissimus dorsi muscle. The size affected the generated pneumatic pressure, and the higher resting pressure of the chamber reduced the muscle blood flow. The maximum stroke work of the circulatory assist device was greater than that of the right ventricle but less than that of the left ventricle. The chamber could drive the circulatory assist device against the systemic range of afterload in which a high preload was available. Long-term adhesion surrounding the chamber reduced the pressure generation capability.
Conclusions. The compressive-type muscle actuator using the latissimus dorsi muscle generated acceptable hemodynamic work for right ventricular bypass or aortic counterpulsation.
|
Introduction
|
|---|
Configurations that couple skeletal muscle contraction directly to the circulation are potentially efficient and relatively inexpensive. The three principal areas of usefulness are cardiomyoplasty [1–3], aortomyoplasty [4], and skeletal muscle ventricles [5, 6]. However, morphologic changes in the wrapped latissimus dorsi muscle (LDM) consistent with fatty degeneration have been detected by magnetic resonance imaging [7], and doubt has been cast on whether the distal part of the LDM flap can function as a power source. The proximal part of the LDM is thicker muscle with better contractility than the distal part, but the proximal part cannot be used in these principal procedures. An alternative is to use skeletal muscle as an energy source for a circulatory assist device (CAD). Several reports have proposed a linear-type muscle actuator system with the use of linear contraction geometry [8–10]. A pneumatic (or hydraulic) chamber that is squeezed by the contraction of a muscle is another actuator system. This type of actuator, which we call the compressive type, is a simpler system. There are previous reports of similar types of actuators, but they did not mention the basic information of the muscle in situ and the performance of the muscle actuator. To investigate the application of the LDM as an in situ energy source for CAD, we performed a series of experiments.
|
Material and methods
|
|---|
Animal preparation
Twenty-two adult mongrel dogs were used in this study. All of the 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 85-23, revised 1985). Anesthesia for all animals was induced and maintained by an intramuscular injection of ketamine (10 mg/kg) and an intravenous injection of pentobarbital sodium (20 mg/kg). Ventilation was maintained by volume-control ventilation (Model-B3; Igarashi Inc, Tokyo, Japan). The animals were divided into three groups. In group 1 (n = 6, body weight = 11.2 ± 1.4 kg), a preliminary experiment was performed to find the most efficient method of using the LDM in situ. In group 2 (n = 8, body weight = 14.0 ± 4.5 kg), a control study was performed for comparison with group 3. In group 3 (n = 6, body weight = 14.6 ± 3.8 kg), a long-term study with the conditioned muscle was performed.
Compressive-type muscle actuator
A pneumatic chamber (PC) was designed as a compressive-type muscle actuator that converted muscle contractile power into pneumatic pressure. The PC has the configuration of a jellyfish and consists of a polyurethane dome and vinyl chloride baseplate equipped with a driving port (Fig 1). The polyurethane dome membrane, 700 µm thick, tolerated a pressure of 0.5 kg/cm2. We made four PCs of different sizes to evaluate the effect of size on generating pneumatic pressure. Table 1 shows the diameters, height of the dome, and volume of the PCs.
View larger version (88K):
[in this window]
[in a new window]
|
Fig 1. Pneumatic chamber (left) connected to circulatory assist device (right). The pneumatic chamber has the configuration of a jellyfish and consists of a polyurethane dome and vinyl chloride baseplate equipped with a driving port.
|
|
The PC was inserted between the LDM and the chest wall. The PC was compressed by the burst-stimulated LDM, and the muscle contractile power was converted into pneumatic pressure.
Generating pneumatic pressure
In group 1, the left LDM was exposed by a longitudinal incision made close to the proximal portion of the LDM. Two electrodes (TAK-1; Matsuda Medical Inc, Tokyo, Japan) were implanted in the muscle for electrical stimulation; one electrode was placed near the main branch of the thoracodorsal nerve and the other approximately 10 cm distally. The PC was inserted and fixed between the LDM and the chest wall. The driving port of the PC was connected to a pressure transducer (TP-101T; Nihon Kohden, Tokyo, Japan) to measure pneumatic pressure in the PC. A multiprogrammable electronic stimulator was used (SEN-3301; Nihon Kohden, Tokyo, Japan). The stimulation parameters were as follows: a burst duration of 280 msec, a burst frequency of 25 Hz, a pulse width of 300 µsec, and a pulse amplitude of 5 V. The contraction rate was 30 per minute.
The effects of the position and size of the PC on the developed pneumatic pressure were investigated. First, the PC-III, 45 mL in volume, was placed in the proximal, middle, and distal positions of the LDM corresponding to the third, fifth, and seventh intercostal spaces, respectively. The air pressure in the PC during the muscle resting phase was controlled to the desired resting chamber pressure (RCP). Then, the muscle was activated and the peak pressure in the PC was observed by changing the RCP. Second, the peak pressure was compared in four different-sized PCs placed in the proximal position.
Evaluation of muscle tissue blood flow
The influence of the expanded PC under the LDM on regional muscle blood flow was evaluated. After inserting the PC (PC-II) in the proximal position, the PC was expanded to reach the desired RCP. Immediately after a 5-minute period of continuous expansion, the muscle blood flow was measured on top of the distended LDM by the expanded chamber. The muscle blood flow was measured using a laser blood flow meter (ALF 2100; Advance Co, Tokyo, Japan) and a needle type probe. The value of the blood flow is expressed as the mean blood flow for 1 minute. The regional muscle blood flow could be measured only during the muscle resting phase. During the muscle activation phase, stable fixing of the probe could not be obtained.
Electrical conditioning and pneumatic chamber implantation
In group 3, under aseptic conditions, the left LDM was exposed and two pacing electrodes (TAK-1; Matsuda Medical, Tokyo, Japan) were implanted in the manner described for group 1. The placing position and the size of the chamber were determined according to the results in group 1. The PC (PC-II) was implanted in the proximal position with minimal dissection. The PC was filled with sterile saline solution and sealed with a chemical bonding agent. The pacing electrodes were connected to a stimulator (Vista DDD; CPI, Minneapolis, MN) implanted subcutaneously in the chest. After these procedures the skin incision was closed. Antibiotic prophylaxis with cefazolin (500 mg) was intravenously administered postoperatively. Two weeks after the operation, the stimulator was activated and continued for 10 weeks. The stimulation parameters were as follows: pulse width, 0.2 ms, and frequency of stimulation, 2 Hz, resulting in 120 twitches per minute. Voltages were adjusted within the range of 2.5 to 5.0 volts to give visible and forcible contractions of the LDM.
Evaluation of the pneumatic chamber as an energy source
After a 12-week period of muscle conditioning, the second operation was performed. The left axillary incision was performed along the scar from the previous operation. The driving port of the PC implanted under the muscle was exposed, and the saline solution in the PC was released.
First, the driving port of the PC was connected to a pressure transducer. The air pressure in the PC during the muscle resting phase was maintained at the desired RCP. The pressure generation capability of the PC enveloped in fibrous adhesion was evaluated by measuring developed peak pressure at several levels of RCP. The fibrous adhesion was then dissected and the PC was exposed. The pressure generation capability of the PC without fibrous adhesion around the PC was evaluated in the same manner.
Second, the driving port of the PC was connected to the air chamber of an air-driven CAD (full stroke volume, 37 mL; Tomasugiken, Himeji, Japan) (Fig 1). The CAD was driven by the developed pneumatic pressure in the PC. The CAD was mounted on an overflow type mock circuit designed to allow the independent control of preload and afterload. The circuit was filled with isotonic saline solution. The flow and pressure of the circuit were measured using pressure transducers and an electromagnetic flow meter (MFV-2100; Nihon Kohden, Tokyo, Japan). The flow velocity wave was integrated and converted into stroke volume (SV), was looped to the pump pressure (pressure-volume loop), the area of which represents the stroke work of the CAD, and was calculated and recorded by a computer (PC9801; NEC, Tokyo, Japan). The SV and stroke work were measured at CAD preloads of 10, 20, and 60 mm Hg. The applied preload pressure simulated a ventricular bypass (low preload, 10 or 20 mm Hg) or aortic counterpulsation (high preload, 60 mm Hg).
The PC design reexpanded passively during the muscle resting phase by volume replacement from the air chamber of the CAD. Therefore, the RCP should be maintained lower than the CAD preload level. When the preload of the CAD was 10, 20, and 60 mm Hg, the applied RCP was 5, 10, and 30 mm Hg, respectively. In addition, we recorded the changes in the SV of the CAD produced by increasing the stimulation rate from 30 beats per minute (bpm) to 80 bpm. After the evaluation, the animals were sacrificed, and a histologic study using ATPase staining was done on the conditioned LDM and the unconditioned contralateral LDM.
In group 2, the same study as in group 3 was performed as a short-term experiment. At the initial operation, two electrodes and the PC were implanted in a protocol similar to that used for group 3. The evaluation of the PC as an energy source was then immediately performed.
Data collection and statistical analysis
All data were collected from ten consecutive muscle activations after continuous activation for 1 minute. Data were recorded on a thermal array recorder (RAT-1300, Nihon Koden) and calculated by personal computer. All data are expressed as mean ± standard deviation of the mean. Statistical analysis was performed by Wilcoxon’s signed-rank test or Student’s paired t test. P values less than 0.05 were considered statistically significant.
|
Results
|
|---|
Pressure generation capabilities of the pneumatic chamber
Figure 2 illustrates the developed peak pressure associated with increasing RCPs in the three different positions of the PC. The developed peak pressure was significantly higher when the PC was in the proximal position than when it was in the middle or distal positions (p < 0.05). There was no significant difference between the values obtained in the middle and distal positions. The pressure increase, which was determined by subtracting the RCP from the developed peak pressure, is shown in Figure 2. There were slight changes as the RCP increased in each position. The highest pressure peak was 140 mm Hg, which occurred when the PC was in the proximal position at an RCP of 40 mm Hg.
View larger version (23K):
[in this window]
[in a new window]
|
Fig 2. Pressure generation function in three different positions of the latissimus dorsi muscle. The developed peak pressure (A) and the pressure increase (B) of the pneumatic chamber versus resting chamber pressure are shown. (Proximal = third intercostal space; middle = fifth intercostal space; distal = seventh intercostal space.)
|
|
The developed peak pressures as the RCP was increased in the PCs of different sizes are shown in Figure 3. The developed peak pressure with PC-II was the highest at all RCPs. There was a significant difference between the PC-II and PC-IV at each RCP (p < 0.005). At RCPs higher than 15 mm Hg, the PC-II could generate a peak pressure of greater than 150 mm Hg.
View larger version (21K):
[in this window]
[in a new window]
|
Fig 3. The developed peak pressure in four different-sized pneumatic chambers versus resting chamber pressure.
|
|
Regional muscle blood flow
The average regional muscle blood flow at each RCP, normalized by the average muscle blood flow at an RCP of 0 mm Hg, is shown in Figure 4. The regional muscle blood flow decreased significantly at RCPs higher than 40 mm Hg (p < 0.05). The blood flow decreased by 13.5% at an RCP of 40 mm Hg (p < 0.05), 23.3% at an RCP of 60 mm Hg (p < 0.01), and 41.4% at an RCP of 80 mm Hg (p < 0.001).
View larger version (14K):
[in this window]
[in a new window]
|
Fig 4. Normalized regional muscle blood flow decreases with increasing resting chamber pressure. The flow was normalized by the resting flow at a resting chamber pressure of 0 mm Hg. Decrease in flow is statistically significant when the resting chamber pressure exceeded 40 mm Hg. (*p < 0.05; **p < 0.005; ***p < 0.001.)
|
|
Influence of chronic adhesions on the pneumatic chamber
In group 3, the PC remained well expanded for 12 weeks. A fibrous adhesion surrounding the PC was observed at the second operation. Moreover, the PC was enveloped by a thin fibrous capsule that was usually seen in a permanent pacemaker pocket. There were no macroscopic findings that suggested ischemic damage or degenerative changes on the inner surface of the LDM in contact with the PC (Fig 5). Figure 6 shows the developed peak pressure in the PC of group 3, with and without the fibrous adhesion. The developed peak pressure in the PC with fibrous adhesion was significantly lower than that after the fibrous adhesion was removed (p < 0.01).
View larger version (95K):
[in this window]
[in a new window]
|
Fig 5. Pneumatic chamber implanted for 12 weeks in the proximal position of the latissimus dorsi muscle. There were no findings that suggested ischemic damage or degenerative changes in the muscle.
|
|
View larger version (17K):
[in this window]
[in a new window]
|
Fig 6. The developed peak pressure in the pneumatic chamber of group 3 is shown. The fibrous adhesion significantly reduced the pressure generating function of the pneumatic chamber. (*p < 0.001; **p < 0.005; ***p < 0.05 between the adhesion group and the nonadhesion group.)
|
|
Energy source capabilities of the pneumatic chamber
Figure 7 shows the SV of the CAD driven by the developed pneumatic pressure of the PC that was compressed by conditioned muscle. At a CAD preload of 10 mm Hg, an SV greater than 15 mL was obtained with an afterload of 30 mm Hg or lower, and the SV decreased as the afterload increased. When the CAD preload was increased to 20 mm Hg, the SV reached 15 mL, even with an afterload of 50 mm Hg. When the CAD preload was increased to 60 mm Hg, ie, when it simulated aortic counterpulsation, the SV reached 15 mL, even with an afterload of 80 mm Hg.
View larger version (31K):
[in this window]
[in a new window]
|
Fig 7. Stroke volume of the circulatory assist device driven by the developed pressure of the pneumatic chamber compressed by conditioned muscle.
|
|
Figure 8 shows the comparison of the SV of the CAD between groups 2 and 3. The SV of the CAD powered by conditioned muscle (group 3) was maintained at 83% to 73% of that of unconditioned muscle (group 2) at an afterload of less than 70 mm Hg. However, at an afterload of 70 mm Hg or greater, there was a marked decrease in the SV of conditioned muscle compared with that of nonconditioned muscle. The calculated maximum stroke work at the CAD preload of 20 mm Hg was 0.87 x 106 erg, which was greater than that of the canine right ventricle (0.2 x 106 erg), but less than that of the canine left ventricle (2.0 x 106 erg).
View larger version (22K):
[in this window]
[in a new window]
|
Fig 8. Comparison of the stroke volume of the circulatory assist device between being driven by nonconditioned muscle (Group II) and being driven by conditioned muscle (Group III) without fibrous adhesion. Circulatory assist device preload = 20 mm Hg. Resting chamber pressure = 10 mm Hg. (*p < 0.05; **p < 0.01 between Groups II and III.)
|
|
The changes in SV as the stimulation rate was increased are shown in Figure 9. The stimulation rate was increased from 30 to 80 bpm. The SV at the CAD preload of 20 mm Hg decreased significantly at stimulation rates above 60 bpm. The change of the SV at the CAD preload of 20 mm Hg decreased by 36% at the stimulation rate of 70 bpm (p < 0.01), and by 52% at the stimulation rate of 80 bpm (p < 0.005). In contrast, the SV at the CAD preload of 60 mm Hg did not decline at all.
View larger version (16K):
[in this window]
[in a new window]
|
Fig 9. The changes in stroke volume of the circulatory assist device (CAD) at two different preloads rates versus CAD rate are shown. (*p < 0.01; **p < 0.005 between CAD rate of 30 per minute.)
|
|
|
Comment
|
|---|
Heart transplantation does not have the possibility of widespread application because of donor limitations [11]. Artificial hearts present problems that remain unresolved [12, 13]; one of the most important problems is the development of a permanent energy source system. The application of autologous skeletal muscle has the potential to overcome this problem.
During the past several years, there has been intense worldwide interest in the use of skeletal muscle as a form of cardiac assistance. The procedure was originally described by Kantrowitz and McKinnon in 1959 [14]. There are three major approaches: dynamic cardiomyoplasty [1–3], dynamic aortomyoplasty [4], and a skeletal muscle ventricle [5, 6, 15]. Another possibility is the use of skeletal muscle as an energy source for a CAD. Several methods have been proposed. The linear-type muscle actuator is thought to be the most efficient method of extracting maximum contractile power [8–10]. The disadvantages of the linear actuator, however, are the complicated mechanism of converting muscle contractile force into energy and anatomic compatibility. A compressive-type muscle actuator such as our model is simpler and smaller than a linear-type actuator. In addition, the muscle dissection area involved is limited, and thus we expected that it would be less affected by adhesions.
Several reports describe the application of the compressive-type of muscle actuators [16–20]. However, these reports do not describe the site of insertion or the size and design of the chamber. We found that the pneumatic chamber functioned best when it was placed in the proximal position (third intercostal space) rather than in the middle or distal position. Because the proximal portion of the LDM is thicker than the middle and distal portions, the proximal portion might provide optimal function. In addition, Egoh and associates [21], who studied the contraction pattern of the LDM by a computerized graphic analysis, found that the proximal third of the total muscle length showed better contraction compared with the distal portion.
In the resting phase, because the LDM is partially stretched by the dome of the chamber, the regional muscle length changes in accordance with the size and configuration of the chamber. Reichebach [22] investigated the relationship between muscle length and contraction force with an isometric contraction test. However, the relationship between regional muscle stretch and contraction force has not been studied yet. Kolff and Stephenson [18] proposed "a large low-pressure pouch" as an inserting chamber. Our study shows, however, that a large chamber (PC-IV) was significantly inferior to a smaller one (PC-II) in generating pneumatic pressure. We did not examine the chamber shape in detail. An optimal chamber size and configuration to convert the contractile force into pneumatic pressure must still be determined. A more precise understanding of muscle contraction biomechanics will help to identify the best configuration and size of the actuator.
Muscle blood flow is one of the most important factors in the function of skeletal muscle pumps. Ischemia of the LDM might be responsible for the deterioration of the muscle flap structure and nerve branches [23]. Kalil-Filho and associates [7] used nuclear magnetic resonance imaging to detect morphologic changes in wrapped LDM consistent with fatty degeneration. It was suggested that deterioration of the muscle might result from the interruption of the distal collateral vessels or kinking of the thoracodorsal artery. Therefore, the distal part of the LDM flap is unlikely to be an efficient power source. These problems could be avoided by using the proximal portion of the muscle in situ, thus keeping the distal collateral vessels intact. Gealow and associates [23] investigated the blood flow to the latissimus dorsi pouch during chronic counterpulsation. They reported that the resting blood flow decreased significantly at a preload greater than 60 mm Hg. Therefore, there is a potential hazard of ischemic degenerative change when a skeletal muscle ventricle is connected directly to the circulation. In our system, the RCP corresponds to the resting preload pressure of the muscle in situ. For that reason, the resting preload of the muscle might be reduced by adjusting the RCP, even when the system is applied to counterpulsation. However, our results showed that the regional muscle blood flow was impaired at a resting chamber pressure of more than 40 mm Hg in the short-term experiment. In group 3, degenerative changes were not observed at gross inspection of the inner surface of the muscle that had been in contact with the PC. However, long-term direct muscle injury might be caused by compression of the device.
Our results showed that the fibrous adhesions between the muscle and the PC impaired the efficacy of pressure generation. The developed pneumatic pressure in the PC with fibrous adhesions was approximately 65% of that of the PC without fibrous adhesions. The negative influence of chronic adhesions on the linear-type muscle actuator system has not been reported. However, it is likely that chronic adhesions yield a great negative influence on the efficacy of the linear-type actuator. The interface between tissue and the muscle actuator is important for the development of muscle-powered CADs and requires further investigation of other device-muscle interface materials.
The SV of the CAD, provided at CAD preloads of 10 and 20 mm Hg, exceeded the predicted SV of dogs used in our experiment when the afterload was 40 mm Hg or less [24]. The calculated stroke work obtained in this experiment was greater than that of the canine right ventricle (0.2 x 106 erg). This result suggests that the PC driven by conditioned muscle might be an energy source for right ventricular bypass. These results are similar to those of a previous study [25]. The SV provided at a CAD preload of 60 mm Hg was maintained at greater than 10 mL, even with an afterload of 100 mm Hg (Fig 6). A sufficient SV against a systemic range of afterload combined with a high CAD preload might be obtained when the PC is applied to simulate an energy source for aortic counterpulsation. Moreover, a high CAD preload could help prompt the reexpansion of the PC, so that the PC can match higher rates of pumping. An active expansion mechanism of the PC will be necessary to allow for higher pumping rates under the low preload conditions of ventricular bypass. The results of this study indicate that a better application of the PC would be as an energy source for aortic counterpulsation.
The present findings suggest that the compressive-type of skeletal muscle pump is a promising design for extracting efficient power from skeletal muscle. The LDM in situ would be an acceptable energy source for right ventricular bypass and aortic counterpulsation in which high preload is available. The establishment of the effective powering of a CAD by skeletal muscle compression requires a greater understanding of muscle contraction dynamics and the development of a high-efficiency power transmission system.
|
References
|
|---|
-
Carpentier A., Chachques J.C., Acar C., et al. Dynamic cardiomyoplasty at seven years. J Thorac Cardiovasc Surg 1993;106:42-52.[Abstract]
-
Magovern G., Sr, Simpson K.A. Clinical cardiomyoplasty: review of the ten-year United States experience. Ann Thorac Surg 1996;61:413-419.[Abstract/Free Full Text]
-
Moreira L.F., Bocchi E.A., Bacal F., et al. Present trends in clinical experience with dynamic cardiomyoplasty. Artif Organs 1995;19:211-216.[Medline]
-
Chachques J.C., Grandjean P.A., Cabrera-Fischer E.I., et al. Dynamic aortomyoplasty to assist left ventricular failure. Ann Thorac Surg 1990;49:225-230.[Abstract]
-
Mannion J.D., Hammond R., Stephenson L.W. Hydraulic pouches of canine latissimus dorsi. Potential for left ventricular assistance. J Thorac Cardiovasc Surg 1986;91:534-544.[Abstract]
-
Oda T., Ban T., Okamoto Y., Miyamoto A.T. Skeletal muscle powered ventricle: comparison of double-layered small ventricle and single-layered large ventricle. J Card Surg 1991;6:154-163.[Medline]
-
Kalil-Filho R., Bocchi E., Weiss R.G., et al. Magnetic resonance imaging evaluation of chronic changes in latissimus dorsi cardiomyoplasty. Circulation 1994;90:II102-II6.
-
Farrar D.J., Hill J.D. A new skeletal muscle linear-pull energy convertor as a power source for prosthetic circulatory support devices. J Heart Lung Transplant 1992;11:S341-S350.[Medline]
-
Sasaki E., Hirose H., Murakawa S., et al. A skeletal muscle actuator for an artificial heart. ASAIO J 1992;38:M507-M511.[Medline]
-
Takahashi M., Misaki T., Watanabe G., et al. Efficacy of a skeletal muscle-powered dynamic patch: part 1. Left ventricular assistance. Ann Thorac Surg 1995;59:305-312.[Abstract/Free Full Text]
-
Stevenson L.W., Warner S.L., Steimle A.E., et al. The impending crisis awaiting cardiac transplantation. Modeling a solution based on selection. Circulation 1994;89:450-457.[Abstract/Free Full Text]
-
Nose Y. Is a totally implantable artificial heart realistic?. Artif Organs 1992;16:19-42.[Medline]
-
Pierce W.S., Sapirstein J.S., Pae W., Jr Total artificial heart: from bridge to transplantation to permanent use. Ann Thorac Surg 1996;61:342-346.[Abstract/Free Full Text]
-
Kantrowitz A., Mckinnon W.M.P. The experimental use of the diaphragm as an auxillary myocardium. Surg Forum 1959;9:266-268.
-
Acker M.A., Hammond R.L., Mannion J.D., Salmons S., Stephenson L.W. An autologous biologic pump motor. J Thorac Cardiovasc Surg 1986;92:733-746.[Abstract]
-
Chiu R.C., Walsh G.L., Dewar M.L., De-Simon J.H., Khalafalla A.S., Ianuzzo D. Implantable extra-aortic balloon assist powered by transformed fatigue-resistant skeletal muscle. J Thorac Cardiovasc Surg 1987;94:694-701.[Abstract]
-
Kochamba G., Desrosiers C., Dewar M., Chiu R.C. The muscle-powered dual-chamber counterpulsator: rheologically superior implantable cardiac assist device. Ann Thorac Surg 1988;45:620-625.[Abstract]
-
Kolff W.J., Stephenson L.W. Total artificial hearts, LVADs or nothing? And muscle and air-powered LVADs. In: Akutu T., Koyanagi H., eds. . Artifical heart. Tokyo: Springer-Verlag, 1993:3-11.
-
Neilson I.R., Brister S.J., Khalafalla A.S., Chiu R.C. Left ventricular assistance in dogs using a skeletal muscle powered device for diastolic augmentation. J Heart Transplant 1985;4:343-347.[Medline]
-
Novoa R., Jacobs G., Sakakibara N., et al. Muscle powered circulatory assist device for diastolic counterpulsator. Asaio Trans 1989;35:408-411.[Medline]
-
Egoh Y., Nakajima H., Kaneko S., Andoh H. Computer graphical analysis for contraction and relaxation pattern of the latissimus dorsi muscle in cardiomyoplasty. Jpn J Artif Organs 1994;23:7-9.
-
Reichebach S.H. Characterization and work optimization of skeletal muscle as a VAD power source. ASAIO J 1994;40:M356-M369.
-
Gealow K.K., Solien E.E., Lang G.R., et al. Blood flow to the latissimus dorsi muscle pouch during chronic counterpulsation stimulation. J Heart Lung Transplant 1992;11:S306-S314.[Medline]
-
Mannion J.D., Velchik M.A., Acker M., et al. Transmural blood flow of multi-layered latissimus dorsi skeletal muscle ventricles during circulatory assistance. Asaio Trans 1986;32:454-460.[Medline]
-
Sakakibara N., Takemura H., Tedoriya T., Kawasuji M., Misaki T., Iwa T. Ex vivo performance of muscle powered cardiac assist device: potential for right ventricular support. J Card Surg 1991;6:171-174.[Medline]
Top
Abstract
Introduction
