Вспомогательное кровообращение с использованием скелетных мышц


От редакции

Автор - один из наиболее известных в мире кардиохирургов, активно разрабатывающих теоретические и экспериментальные основы вспомогательного кровообращения с помощью биологического скелетно-мышечного желудочка. В настоящей статье представлены результаты изучения в хроническом эксперименте на животных динамической кардиомиопластики, аортомиопластики (контрпульсации) и скелетно-мышечного желудочка в целях органосохраняющего хирургического лечения кардиомиопатии. Также дан обзор современного клинического опыта и отдаленных результатов использования кардиомиопластики. 

Ключевые слова:кардиомиопатии, сердечная недостаточность, динамическая кардиомиопластика, скелетно-мышечный желудочек

Клин. и эксперимент. хир. Журн. им. акад. Б.В. Петровского. 2017. № 3. С. 71-80.

Статья поступила в редакцию: 20.06.2017. Принята в печать: 07.07.2017. 

Energy derived from the contraction of skeletal muscle has been used experimentally and clinically to augment native heart function. Skeletal muscle is capable of efficient transformation of chemical energy into mechanical work. Investigators have shown that it is possible to transform a fatigue-sensitive skeletal muscle into a fatigue-resistant muscle capable of repeated contractions over a sustained period of time. Investigators have used this transformed muscle in an attempt to assist the heart and circulation in a number of novel ways.

Two methods of skeletal muscle cardiac assist - cardiomyoplasty and aortomyoplasty have been used clinically and will be discussed. Cardiomyoplasty involves wrapping the latissimus dorsi muscle around the cardiac ventricles and stimulating the muscle during cardiac systole in an attempt to assist the failing heart. Aortomyoplasty requires that skeletal muscle be wrapped around the aorta and stimulated during diastole to unload the failing ventricle.

A third approach to skeletal muscle cardiac assistance uses skeletal muscle ventricles (SMVs), which so far have only been used in laboratory animals. In contrast to the procedures already mentioned, separate pumping chambers are constructed from the muscle and then connected to the circulation in various ways and used as auxiliary blood pumps.

Dynamic Cardiomyoplasty

Since Alain Carpentier and Juan Carlos Chachques in Paris, France introduced clinical dynamic cardiomyoplasty in 1985, an estimated 1000 patients have had these procedures worldwide (Fig. 1) [1]. Latissimus dorsi muscle is wrapped around the heart and stimulated to contract in synchrony with the cardiac ventricles. In the phase II clinical trials conducted in the United States under the auspices of FDA (Food and Drug Administration), about 80-85% of hospital survivors showed improvement in their signs and symptoms for heart failure from several months up to a year and sometimes much longer [2].

The phase III randomized clinical trial, again under the auspices of the FDA, commenced in June 1995 and ended in 1998. Slightly more than 100 patients entered this study which was designed to determine similar improvement in the signs and symptoms of heart failure as in the phase II trial and also when compared with control (medically treated) patients. There was no improvement, however, in survival of the cardiomyoplasty patients when compared to the medically randomized controls followed for about two years. This fact and difficulty in recruiting patients into the study resulted in Medtronic, who was sponsoring the study, terminating the phase III trial. This study was thought to need outcomes from about 200 patients for valid statistical comparisons. With Medtronic’s withdrawal from the study and their decision to no longer manufacture cardiomyostimulators, cardiomyoplasty, from a clinical standpoint, was essentially abandoned in the United States [2].

Some groups in other countries continue to perform this procedure, using other available cardiomyostimulators. As new clinical data and laboratory research becomes available, it is possible that there could be a resurgence of cardiomyoplasty.

In 2003, Benicio and colleagues reported on long-term results with cardiomyoplasty. Their group from Sao Paulo University in Brazil has been among the leaders and has been one of the most reputable groups in this field for many years. They presented their results with cardiomyoplasty over a 10-year period, and compared their results by patients who were initially NYHA (New York Heart Association) Functional Class III for heart failure versus those who were Functional class IV. When cardiomyoplasty was first performed, there was controversy as to whether the latissimus dorsi muscle should be stimulated with every heartbeat or every other heartbeat. Their data clearly showed that over time the 1:2 mode of stimulation was better. They concluded, based on the marked improvement of the long term performance of skeletal muscle obtained with a 1:2 stimulation mode, that the use of cardiomyoplasty continues to be justified as an alternative treatment for patients with dilated cardiomyopathies [2].

In 2009 the French reported on a multicenter experience (6 centers) with long term results of cardiomyoplasty since the very first clinical case in 1985 to the last patient entered in the series in 2006. This totaled 212 cardiomyoplasty procedures. Patient all had preoperative symptoms of chronic heart failure despite maximal medical management. The etiology was ischemic (48%), idiopathic (45%), or other (7%). During the follow up 88% of the hospital survivor’s patients improved clinically. Hospital deaths occurred in 14% of the patients and were related to the severity of the preoperative heart failure symptoms. Late mortality occurred in 99 patients due to heart failure (44%), sudden death (37%), or non-cardiac causes (18%). Cardiomyoplasty was combined with implantation of a cardiac rhythm management system in 22 patients and 26 eventually underwent heart transplantation after recurrent heart failure occurred. Long term functional improvement was observed in most patients undergoing cardiomyoplasty and the best outcome was achieved in those with isolated right ventricular failure. The authors concluded that dynamic cardiomyoplasty can be considered a destination therapy or a mid to long term bridge to heart transplantation [3].


Aortomyoplasty is a surgical procedure, where the skeletal muscle is wrapped directly around the aorta instead of the heart. This operation has been performed in more than 29 humans worldwide and shows some promise [4].

The muscle has been wrapped around the ascending aorta just as it comes out of the heart or wrapped around the descending aorta (Fig. 2). In the animal laboratory sometimes the aorta is enlarged with a patch to increase stroke volume as the muscle contracts around it.

Because aortomyoplasty and cardiomyoplasty work on different parts of the anatomy, the muscle is trained to contract at different times. In cardiomyoplasty, the muscle is stimulated to contract at the same time as the heart ventricles. In aortomyoplasty, the latissimus muscle is stimulated to contract during diastole while the heart ventricles are relaxing which is similar in principle to that of an intraortic balloon pump.

One advantage this procedure has over cardiomyoplasty is that the sick heart itself does not need to be manipulated. In some cases, the heart is so large that the latissimus muscle cannot even be wrapped around it. In this case, cardiomyoplasty would not be performed, but an aortomyoplasty could be an option.

Skeletal Muscle Ventricle Construction

Our laboratory has used the latissimus dorsi muscle of dogs for construction of the SMV. Both the right and left latissimi have been used, depending on the type of assist to be performed. The operation is begun by making an incision from the axilla to the tip of the eleventh rib. The overlying platysma and soft tissue are elevated from the muscle. The attachments to the eleventh and twelfth rib are then incised. The attachment along the posterior spinous processes is then incised, taking care to include a wide margin of the thoracodorsal fascia. The underlying collateral blood vessels from the chest wall are then divided. The muscle is fully mobilized up to its humeral attachment, taking care to avoid injury to the neurovascular pedicle.

The thoracodorsal nerve is encircled with a bipolar nerve lead, which is then connected to a nerve stimulator and placed in a subcutaneous pocket over the rectus abdominis muscle. Next, a thoracotomy at the fifth intercostal space is performed to harvest the pericardium, which will line the inner surface of the SMV. The pericardium is then sutured to a Dacron cuff that encircles a plastic mandrel, and the chest is closed. The muscle is wrapped around the mandrel, anchored by the Dacron cuff. Generally, one to two wraps of the muscle are obtained. The thoracodorsal fascia is oriented so that this firm tissue is sewn to the Dacron cuff. Absorbable sutures are then used to secure the layers of the muscle together and to close the end of the muscle to form the apex of the pumping chamber. The chamber is then sewn to the surrounding tissue. The SMV has thus been either anchored in the subcutaneous tissue on the chest wall or placed in an intrathoracic position after excision of several ribs.

 After mobilization of the muscle, there is relative ischemia in the distal portion. This portion is not able to increase its blood flow in response to the increased demands of stimulation. Mannion and colleagues showed that over the next 3 to 4 weeks, the muscle gradually recovers its ability to increase its blood flow in response to stimulation. We refer to this 3- to 4-week period as the period of vascular delay [5, 6].

 The muscle is allowed to recover during this vascular delay period and is then stimulated at 2 Hz continuously over the next 5 to 7 weeks. Following this period of electrical conditioning, the muscle of the SMV has become fatigue resistant and can be used to assist the native heart in a variety of configurations. A second operative procedure is then performed whereby the mandrel is extracted and the SMV is connected to the circulation. A ventricular sensing lead is placed on the myocardium, and the SMV is stimulated to contract synchronously with the heart, using an implantable cardiomyostimulator.

 Skeletal Muscle Ventricles as Cardiac Assist Devices

 Aortic Diastolic Counterpulsators

The following configuration represents the model most studied in our laboratory. The SMV is connected to the circulation using the bifurcated graft that is anastomosed to the base of the SMV and then to two locations on the descending thoracic aorta. The aorta is ligated between the two limbs of the graft to obligate blood flow through the circuit. A myocardial lead is then used to sense the electrical activity of the native ventricle, and an implantable cardiomyostimulator is used to synchronize contraction to occur in cardiac diastole (Fig. 3).


The contraction of the SMV during cardiac diastole has several useful purposes. First, blood is pumped proximally and distally from the descending aorta to the periphery. Second, because the coronary arteries are perfused during diastole, there is an increase in coronary artery flow. Finally, relaxation of the SMV chamber at the end of diastole provides a low-pressure system into which the native heart is able to eject, decreasing the energy required for the heart to pump blood, thereby decreasing the heart’s oxygen consumption. These hemodynamic improvements are similar to those produced with an intra-aortic balloon pump. Representative figure tracings are shown in Fig. 4.

Fig. 4. Pressure and electrocardiographic tracings recorded from the longest surviving animal at the time of connection to the aorta (A), and after 1 year (B), 2 years (C), and 4 years (D) in the circulation. Stimulation burst frequency is 33 Hz at a 1:2 assist ratio. Carotid P - pressure measured at the carotid artery. Fem P - blood pressure measured at the femoral artery 

 Early acute experiments showed that electrically pre-conditioned SMVs were able to generate a power output of 0.68×106 erg, which was approximately half the power output of the native left ventricle and roughly three times the power output of the right ventricle [7]. In 1987, Acker reported on experiments involving five dogs that had SMVs constructed with a modification of the SMV design [8]. These chambers had a cylindrical geometry with inflow and outflow at opposite ends of the chamber. The SMVs were then monitored over time while pumping blood continuously in the circulation. These chambers functioned as diastolic counterpulsators for up to 11 weeks. During periodic measurements of SMV function, the burst frequency was increased from chronic 25 Hz setting to 43 Hz and then to 85 Hz. These pumps improved aortic flow by 29, 40, and 63% at 25, 43, and 85 Hz of thoracodorsal nerve stimulation, respectively. Two-dimensional short-axis echocardiograms of the SMB chambers were obtained; they showed a 70, 90, and 100% decrease in the cross- sectional area at the midpoint of the SMV as the burst stimulation frequency increased. The decrease in cross-sectional area was somewhat similar to the ejection function of the SMV. These chambers, however, were prone to thrombus formation, and although all animals had a functional SMV at the time of death, the two longest-surviving animals, 5 and 11 weeks, demonstrated multiple splenic and renal infarctions at the time of autopsy. Neither animal, however, showed evidence of cerebral or coronary embolization.

 Over the past decade, modification in the diastolic counterpulsator model have allowed for improvements in survival. In 1992, Mocek reported on a series of four dogs that survived for more than 6 months with an SMV pumping continuously in circulation [9]. One animal from this series survived for 836 days but showed evidence of some thrombus formation within the chamber at the time of death.

The use of autologous pericardium as a blood- SMV surface lining was then investigated as a possible method of decreasing the incidence of thrombosis. The animal’s pericardium was removed at the time of the initial construction of the SMV and wrapped around the mandrel before the muscle wrap was applied. The tissue was oriented so that the inner surface of the pericardium was in contact with the plastic mandrel. When the mandrel was removed several weeks later and the SMV connected to the circulation at the second operation, the blood came in contact with the inner surface of the pericardium. Interestingly, there was no thrombosis noted in either the group with autologous lining or the control group (constructed with the inner layer of the muscle serving as the contact surface for the circulatory blood flow). However, the group with the autologous lining demonstrated a significantly reduced rate of rupture. Sixty-three percent of the SMVs in the control group ruptured over time, as compared to 0% in the group with autologous lining. These investigators concluded that the autologous pericardium improved the structural integrity of the pumping chamber [10]. One animal in the group was electively sacrificed after continuously pumping blood for more than 4 years. To our knowledge, this represents the longest reported survival - clinical or experimental - with a functioning, indwelling cardiac assist device of any type [11].

 Thomas also demonstrated that it is possible to line these chambers with autologous endothelial cells and to retain the endothelial surface while the SMV pumps blood in the arterial circulation. The animal’s own jugular vein was used to obtain endothelial cells. After the cells were harvested and grown in culture, the suspended cells were then delivered into the space between the muscle itself and the plastic mandrel at a separate surgical procedure [12]. After allowing several weeks for the endothelial cells to grow and attach, a confluent monolayer of endothelial cells was histologically shown to be present on the inner surface of the pumping chambers. This group also demonstrated that this same result could be obtained by percutaneously injecting the suspended endothelial cells into the space around the mandrel [13]. Finally, they showed that this endothelial cell layer was retained after the SMV had pumped blood in the arterial circulation for 3 h [14]. We believe that this was the first report of the endothelium remaining intact on the surface of a heart assist device while the device pumped blood in the circulation.

The aortic diastolic counterpulsator model has been investigated in the setting of heart failure [15]. Because blood supply to the SMV muscle itself may be impaired in the setting of low cardiac output, the possibility of impaired SMV hemodynamic function exists. Propranolol was used to induce heart failure, and it was shown that the percentage improvement in several hemodynamic parameters in this setting was actually better than without propranolol with the SMV functioning. Mean diastolic pressure increased 27.6%, as compared to a 12.9% increase in the same group of SMVs before the induction of heart failure. The endocardial viability ratio, a ratio of myocardial oxygen delivery to myocardial oxygen demand, also increased 28.7% in the setting of heart failure, versus an 11.2% increase before heart failure induction. However, the studies were performed in an acute setting over 1 h, and the animal’s cardiac output promptly returned to normal upon discontinuation of the propranolol.

More recently, our laboratory used a stable chronic heart failure model that allows evaluation of the function of SMVs in the setting of chronic low cardiac output. We used the rapid ventricular pacing (RVP) technique in conjunction with the aortic diastolic counterpulsator model. Patel examined six dogs with pericardium-lined SMV’s created from latissimus dorsi muscles. Each SMV was anastomosed to the descending thoracic aorta with a two-limbed bifurcated polytetrafluoroethylene (PTEE) graft after the usual electrical conditioning period, and the aorta was ligated between the limbs. The SMV was stimulated to contract during diastole at a 1:2 to 1:3 ratio. Chronic heart failure was then induced over the next 7 weeks with the initiation of rapid ventricular pacing at 220 to 230 bpm. SMV contraction resulted in augmentation of the diastolic pressure-time index (DPTI) by 12.1% prior to initiation of RVP and by 33.6% after 7 weeks of RVP [16].

The rapid ventricular pacemaker was turned off temporarily during measurement of the left ventricle function, while the SMV was appropriately stimulated with the cardiomyostimulator to again contract synchronously with the heart in a 1:2 or 1:3 ratio. In addition, significant afterload reduction was demonstrated, with increases in peak left ventricular ejection velocity of 22.7% and stroke volume of 6.2%. In three of the six animals coronary blood flow was shown to be augmented by an average of 47.6%.

Guldner’s group in Germany has recently reported success in chronic studies using an aortic diastolic counterpulsation configuration somewhat similar to ours but with a unique blood contacting chamber device. They used the latissimus dorsi muscle of the boer goat and reported pumping blood in the circulation for up to 414 days [17].

Left-Heart Bypass

Hooper and colleagues constructed SMVs and connected the left atrium to the SMV and the SMV to the aorta by using two valved conduits [18]. In this parallel circuit model, the left atrial pressure served as the preload for the SMV. A portion of the systemic cardiac output that would normally have been pumped by the left ventricle (LV) was routed through the parallel SMV circuit and pumped by the contraction of the SMV. Thus, the work required by the native heart was decreased even though the net blood flow produced the LV and SMV were similar to control. The SMVs in this configuration, as in the aortic diastolic counterpulsator model, are stimulated to contract during diastole because contraction during systole would result in the SMV attempting to eject against an afterload equal to the systolic pressure generated by the LV. Acute experiments over 3 h by Hooper showed that the SMV was able to pump between 21 and 27% of the cardiac output. Although these results showed promise, we have not pursued this model in a chronic setting because of our observation that higher SMV preloads seem to be necessary for optimal SMV performance.

Left Ventricular Apex-to-Aorta Model

In terms of hemodynamic augmentation, the left ventricular apex-to-aorta model is a highly effective experimental model for ventricular assistance, both in vivo and via computer model [19]. This model involves construction of a SMV that is connected in circulation by two valved conduits (Fig. 5) One conduit is placed from the apex of the left ventricle to the SMV, and the other joins the SMV to the descending thoracic aorta. This model makes use of the higher pressure generated by the left ventricle to serve as preload for the SMV. In addition, when the SMV relaxes and the left ventricle ejects into this low-pressure system, the left ventricle is effectively “unloaded”. Figure 6 shows representative hemodynamic tracings of this model after 1 year of functioning in circulation [20].

The blood flow to the muscle layers of the SMV is also likely to be improved in the left ventricular apex-to-aorta configuration when compared to that of the aortic diastolic counterpulsator model. With the left ventricular apex-to-aorta model, there is a substantial period of time during which the pressure inside the SMV itself is much lower than the systemic diastolic pressure. In contrast, with the aortic diastolic counterpulsator model, the walls of the pumping chamber are always exposed to a pressure at least equal to the systemic arterial pressure, which could potentially cause problems with impaired blood flow to the SMV muscle layers.

Fig. 6. Hemodynamic recording obtained after 1 year in circulation from a canine with an SMV positioned between the LV apex and the aorta. The SMV is contracting at a 1:2 ratio with the native heart and stimulated at a 33 Hz burst frequency. Arrows indicate effects of SMV contraction in the pressure and flow traces ECG - electrocardiogram; LV - left ventricle; SMV - skeletal muscle ventricle 

Initially, acute 3 h experiments were performed; they showed significant improvement in the systolic tension-time index, a measure of myocardial oxygen demand, when the SMV was pumping [21]. The endocardial viability ratio was also significantly improved. At the time of implant and at 1, 2, and 3 h in circulation, the ratio was increased by 68, 66, 62, and 63%, respectively. The SMV circuit in these acute studies pumped 47% of the cardiac output. Stevens and colleagues demonstrated a 31% increase in cardiac output in chronic heart failure in dogs with a left ventricular apex-to-aorta SMV of their own design and constructed from the rectus abdominis muscle [22].

Fig. 7. SMV in LV apexto- aorta configuration. Pressure recordings were made at the time of connection of the SMV to the systemic arterial circulation. Transition from the control state with the SMV off to stimulation at 33 Hz, 1:2 ratio, can be seen, with corresponding changes in the flow and pressure tracings 

Subsequently, Thomas documented an SMV in a left ventricular apex-to-aorta configuration that was electively sacrificed after functioning well for 1 year [20]. In a chronic heart failure study, skeletal muscle ventricles were constructed from the latissimus dorsi muscle in 10 dogs. After conditioning, the SMVs were connected to the left ventricle and aorta with two valved conduits, and the SMV was programmed to contract during diastole [23]. At the time of implantation, SMVs stimulated at 33 Hz and in a 1:2 ratio, the power output of the SMVs was 59% of left ventricular power output at 33 Hz and 93% at 50 Hz (Fig. 7). Animals survived 7, 11, 16, 17, 72, 99, 115, 214, and 248 days. Three deaths were directly related to the SMV. In the animal that survived 248 days, SMV power output at 8 months with a 33 Hz stimulation frequency and a 1:2 contraction ratio was 57% of left ventricular power output and 82% at 50 Hz. At a 1:1 contraction ratio, SMV power output was 97% and 173% of the left ventricle at 33 and 50 Hz, respectively (Fig. 8).

Fig. 8. A. The SMV in the LV apex- to-aorta configuration is contracting at 33 Hz at a 1:2 ratio (chronic stimulation setting) with the heart. The shaded areas in the SMV flow trace indicate SMV stroke volume B. Pressure recoding made after 248 days of pumping blood continuously in the circulation. The SMV is shown contracting at 50 Hz, 1:1 ratio 

This study demonstrated that SMVs in a LV apex-to-aorta configuration are able to function effectively in the circulation. Maintenance of significant power output was confirmed in one animal at the one year follow-up when the animal was then electively sacrificed according to the studies protocol. Skeletal muscle ventricles significantly unloaded the left ventricle, resulting in decreases in the LV peak pressure, LV end-diastolic pressure, the LV tension- time index, and LV stroke and minute work. During SMV contraction, flow was redirected from the LV outflow tract and through the aortic valve instead of through the SMV. Total systemic flow did not change significantly. 

We have found that with the chronic model it is desirable to narrow the aorta just above the anastomoses of the efferent SMV conduit to the aorta. By doing this, a slight pressure gradient in the aorta occurs at this point, which allows some passive flow of blood through the SMV system into the aorta during every other cardiac cycle (ie, the unassisted beats) and thereby reduces the chances of blood clot formation in the system. To determine whether the redistribution of flow dynamics caused by this 50% constriction of the aorta would compromise cerebral circulation during SMV contraction, a flow probe was placed around the carotid artery in one animal during a measure session. During SMV stimulation at 33 and 50 Hz, mean carotid flow was 97.1 and 97.4% of control flow with the SMV off.

The left ventricular apex-to-aorta configuration is the most hemodynamically efficient model for left ventricular assistance that we have tested.

Right-Heart Bypass

Bridges showed that skeletal muscle ventricles were effective in replacing the native right ventricle [24]. As with other series, the SMVs received 3 weeks of vascular delay and 4 to 6 weeks of continuous low-frequency electrical preconditioning at 2 Hz. Larger mandrels (49 to 69 mL) were used in the construction of these SMVs in an effort to improve chamber compliance (ie, fewer muscle wraps around the mandrel and, consequently, decreased SMV wall thickness) so as to make it more suitable to accommodate the low-pressure right-heart circulation. At the second operation, valved conduit grafts were used to route all systemic blood flow from the superior and interior venae cavae to these high-compliance chambers. Outflow from the SMVs was returned to the pulmonary artery. Upon stimulation of the SMV, the systolic blood pressure increased to the 100 mm Hg range from a systolic pressure of 50 to 60 mm Hg associated with passive flow without SMV stimulation. The SMV stroke work was 169 and 174% of the canine right ventricular stroke work at 2 and 4 h of continuous pumping, respectively.

Despite the intriguing nature of these experiments, our laboratory observed that the SMVs seemed to function better at higher preload levels than that achieved with the venous system alone. Consequently, we developed a model where the pressure generated by the native right ventricle was used for the SMV pre-load [25]. One valved conduit connected the right ventricle to the SMV, and a second valved conduit connected the SMV to the pulmonary artery. The pulmonary artery was then ligated proximal to the conduit to obligate right ventricular flow through the SMV circuit. At 1 h, the cardiac output of these animals increased by 27% with the SMV stimulated, as compared to an increase of 30% at 4 h with the stimulator off. Similarly, systemic arterial pressure increased by 12 and 13% at 1 and 4 h, respectively. Further studies have demonstrated that this model can continue to pump effectively and augment the native circulation for up to 16 weeks [26].


Dynamic Cardiomyoplasty (DCM)

Currently survival at 1 and 2 years stand around 72 and 60% respectively. Late postoperative deaths were usually caused by progressive heart failure and ventricular arrhythmia. Incorporation of a cardioverter defibrillator into the device may help to improve the outcome. Approximately 80% of survivors report subjective benefits including improvement in functional status (NYHA) and quality of life. A survival advantage at 18 months over medically treated controls had also been demonstrated [2, 27]. However, such parameters have not been matched by improvements in objective measures such as haemodynamic and exercise testing.


Although there are reports that patients with failing hearts have benefited from aortomyoplasty, it should be considered experimental, particularly until more clinical cases are reported with long-term follow-up.

Skeletal Muscle Ventricles

Even though we have had our best long-term success withaorticdiastoliccounterpulsatorconfiguration,the left ventricular apex-to-aorta configuration remains the most hemodynamically efficient model. Although several animals have survived for periods between 6 months and 1 year, the morbidity and mortality of this model must be reduced significantly, and the model must be tested during chronic heart failure prior to attempting clinical studies.

The use of an autologous biologic pump would obviate the need for patients to wear an external power unit, and transformed skeletal muscle is an efficient means of assisting the failing heart. With continued study and refinement of technique, it is hoped that SMVs will be available someday for clinical application. 


1. Carpentier A., Chachques J.C. Myocardial substitution with a stimulated skeletal muscle: First successful clinical case. Lancet. 1985; 1: 1267.

2. Stephenson L.W. Invited commentary on: dynamic cardiomyoplasty: long-term outcomes. Ann Thorac Surg. 2003; 76: 821-7.

3. Chachques J.C., Jegaden.O., Mesana.T., Glock Y., Grandjean.P., Carpentier.A. Cardiac Bioassist: results of the French multicenter cardiomyoplasty study. Asian Cardiovasc Thorac Ann. 2009; 17: 573-80.

4. Chachques J.C., Radermecker M., Grandjean P., et al. Dynamic aortomyoplasty for long-term circulatory support: experimental studies and clinical experience. In: Carpentier A.F., Chachques J.C., Grandjean P.A., eds. Cardiac Bioassist. Armonk, NY: Futura Publishing Co., 1997: 488.

5. Mannion J.D., Bitto T. Hammond R.L., et al. Histochemical fatigue characteristics of conditioned canine latissimus dorsi mus- cle. Circ Res. 1986; 58: 298-304.

6. Mannion J.D., Hammond R.L., Stephenson L.W. Canine latissimus dorsi hydraulic pouches. Potential for left ventricular assistance. J Thorac Cardiovasc Surg. 1986; 91: 534-44.

7. Mannion J.D., Acker M.A., Hammond R.L., Stephenson L.W. Four-hour circulatory assistance with canine skeletal muscle ventricles. Surg Forum. 1986; 27: 211-3.

8. Acker M.S., Hammond R.L., Mannion J.D., et al. An autologous biologic pump motor. J Thorac Cardiovasc Surg. 1986; 92: 733-46.

9. Mocek F.W., Anderson D.R., Pocchetino A., et al. Skeletal muscle ventricles in circulation long-term: 191 to 836 days. J Heart Lung Transplant. 1992; 11: S334-40.

10. Thomas G.A., Lu H., Isoda S., et al. Skeletal muscle ventricles in circulation: decreased incidence of rupture. Ann Thorac Surg. 1996; 61: 430-6.

11. Thomas G.A., Hammond R.L., Greer K.A., et al. Functional assessment of skeletal muscle ventricles after pumping for up to four years in circulation. Ann Thorac Surg. 2000; 70: 1281-90.

12. Thomas G.A., Lelkes P.I., Chick D.M.C., et al. Skeletal muscle ventricles seeded with autologous endothelium. ASAIO J. 1995; 41: 204-11.

13. Thomas G.A., Lelkes P.I., Chick D.M., et al. Endothelial- lined skeletal muscle ventricles: open and percutaneous techniques. J Card Surg. 1995; 10: 245-56.

14. Thomas G.A., Lelkes P.I., Chick D.M., et al. Endothelial-lined skeletal muscle ventricles in circulation. J Thorac Cardiovasc Surg. 1995; 109: 66-73.

15. Thomas G.A., Lu H., Isoda S., et al. Pericardial-lined skeletal muscle ventricles in circulation up to 589 days. Ann Thorac Surg. 1994; 58: 978-88.

16. Patel B.G., Shah S.H., Astra L.I., et al. Skeletal muscle ventricle aortic counterpulsation: function during chronic heart failure. Ann Thorac Surg. 2002; 73 (2): 588-93.

17. Guldner N.W., Klapproth P., Zimmerman H., Sievers, H. Sketetal muscle ventricles (SMVs) and biomechanical hearts (BMHs) with a self-endothelializing titanized blood contacting surface. In: Tissue engineering and regenerative medicine by Andrades JA, intechopen.com, open access ccby 3.0, license pub May 22, 2013 pp 4-35. (www.intechopen.com/books/regenerative medicine-and-tissue-engineering/skeletal-....accessed June 22, 2017.)

18. Hooper T.L., Niinami H., Lu H., et al. Skeletal muscle ven- tricles as left atrial-aortic pumps: short-term studies. Ann Thorac Surg. 1992; 54: 316-22.

19. Platt K.L., Moore T.W., Barnea O. Performance optimization of left ventricular assistance: a computer model study. ASAIO J. 1993; 39: 29-38.

20. Thomas G.A., Stephenson L.W.: Update on skeletal muscle ventricles: Left ventricular apex to aorta configuration. Ann Thorac Surg. 71: 1736-7, 2001.

21. Lu H., Fietsam R., Hammond R.L., et al. Skeletal muscle ventricles: left ventricular apex-to-aorta configuration. Ann Thorac Surg. 1993; 55: 78-85.

22. Stevens L., Badylak S.F., Janas W., et al. A skeletal muscle ventricle made from rectus abdominis muscle in the dog. J Surg Res. 1989; 46: 84-9.

23. Thomas G.A., Baciewicz F.A., Hammond R.L., Greer K.A., Lu H., Bastian S., Jindal P., Stephenson L.W. Power output of pericardium-lined skeletal muscle ventricles, left ventricular apex to aorta configuration: up to 8 months in circulation. J Thorac Cardiovasc Surg. 1998; 116: 1029-42.

24. Bridges C.R., Anderson W.A., Hammond R.L., et al. Functional right heart replacement with skeletal muscle. Circulation.1989; 80 (5pt.2): III-183-91.

25. Niinami H., Hooper T.L., Hammond R.L., et al. A new configuration for right ventricular assist with a skeletal muscle ventricle: short-term studies. Circulation. 1991; 84: 2470-5.

26. Niinami H., Hooper T.L., Hammond R.L., et al. Skeletal muscle ventricles in the pulmonary circulation: up to sixteen weeks’ experience. Ann Thorac Surg. 1992; 53: 750-7.

27. Magovern G.J., Sr., Simpson K.A.. Clinical cardiomyoplasty: review of the ten-year United States experience. Ann Thorac Surg. 1996; 61 (1): 413-9.