Ann Thorac Surg 2002;73:588-593
© 2002 The Society of Thoracic Surgeons
Original article: cardiovascular
Skeletal muscle ventricle aortic counterpulsation: function during chronic heart failure
Bhavik G. Patel, MDa,
Sachin H. Shah, MDa,
Lou I. Astra, MDa,
Robert L. Hammond, PhDa,
Zulfikar A. Sharif, MDa,
Phillip J. McDonalda,
Larry W. Stephenson, MD*a
a Department of Surgery, Cardiothoracic Division, Wayne State University, Detroit, Michigan, USA
Accepted for publication November 1, 2001.
* Address reprint requests to Dr Stephenson, Cardiothoracic Surgery, Harper Professional Bldg, Suite 2102, 3390 John R St, Detroit, MI 48201, USA
e-mail: lstephenson{at}intmed.wayne.edu
 |
Abstract
|
|---|
Background. Skeletal muscle ventricles (SMVs) are pumping chambers formed from latissimus dorsi muscle. The SMV aortic counterpulsator model has been proven to be stable in the long term and provide effective diastolic pressure augmentation in normal dogs. This study seeks to prove that the aortic counterpulsator model can function effectively in chronic heart failure.
Methods. In 6 dogs, pericardium-lined SMVs were created from latissimus dorsi muscle and electrically conditioned for fatigue resistance. Each SMV was attached to the descending thoracic aorta with a two-limb bifurcated graft and the aorta ligated between the limbs. The SMV was stimulated to contract during cardiac diastole at 1:2 to 1:3 ratio. Rapid ventricular pacing was initiated at 220 to 230 beats/min for 7 weeks to induce chronic heart failure.
Results. SMV contraction resulted in augmentation of the diastolic pressure time-index by 12.1% (32.8 ± 15.4 versus 36.1 ± 14.7 mm Hg-s, p < 0.05) at baseline, then by 33.6% (12.9 ± 4.4 versus 16.8 ± 4.3 mm Hg-s, p < 0.05) after 7 weeks of rapid ventricular pacing. After 7 weeks of rapid ventricular pacing, SMV counterpulsation provided significant afterload reduction with increases in peak left ventricular ejection velocity and stroke volume of 22.7% (142 ± 55 versus 168 ± 45 mL/s, p < 0.05) and 6.2% (13.0 ± 5.1 versus 13.7 ± 5.2 mL, p < 0.05), respectively. Coronary blood flow was measured in 3 animals at the 7-week measurement; augmentation averaged 47.6% (0.357 ± 0.29 versus 0.432 ± 0.26 mL/beat, p < 0.05).
Conclusions. The SMV aortic counterpulsator provides improved cardiac assistance relative to the failing heart.
|
Introduction
|
|---|
Skeletal muscle ventricles (SMVs) of various types have been studied in several laboratories for over 15 years. SMVs differ from the other forms of skeletal muscle cardiac assist, cardiomyoplasty and aortomyoplasty, in that they consist of a separate pumping chamber connected to the circulation. The most stable model is the aortic counterpulsator, which has been shown to pump blood effectively in the arterial circulation for over 4 years [1].
SMVs have the potential to benefit patients with chronic heart failure. Although the aortic counterpulsator model has been shown to be effective in the long term and in propanolol-induced acute heart failure, no previous study has shown that a SMV of any type can perform effectively during chronic heart failure [1, 2]. Using rapid ventricular pacing as a technique to induce chronic heart failure, this study evaluates the aortic counterpulsator in dogs with moderate-to-severe heart failure.
|
Material and methods
|
|---|
Construction of skeletal muscle ventricles
SMVs were constructed from left latissimus dorsi muscle (LDM) in 6 mongrel dogs weighing from 16 to 26 kg. The animals were operated on in accordance with the "Guide for the Care and Use of Laboratory Animals" published by the National Institutes of Health (NIH publication 85-23, revised 1985).
Each animal underwent a two-stage procedure. The first to create the SMV and the second to connect the SMV to the systemic circulation. Anesthesia was induced using intravenous thiopental 20 to 25 mg/kg, and maintained after endotracheal intubation with isoflurane (1.5% to 3.0%). At the first operation, an incision from the left axilla to tip of the 11th rib was made, and the LDM was mobilized from the subcutaneous tissue and the chest wall using electrocautery, leaving the humeral attachment and the thoracodorsal nerve and blood supply intact. A cuffed electrode (Medtronic, Inc, Minneapolis, MN) was placed around the thoracodorsal nerve and sutured in place. The left chest was opened at the fifth interspace and the pericardium was harvested to serve as an inner lining for the SMV. Subsequently, a 20-mm ultrasonic flow probe (Transonic Systems, Inc, Ithaca, NY) was placed around the root of the aorta to measure left ventricular ejection velocity, stroke volume, and cardiac output. After the chest was closed, the pericardium was sewn to a polypropylene mandrel, which is constructed from a 50-mL vial with a 3-cm base. The top of the mandrel is removed to adjust the volume to 25 to 30 ml and a 1-cm Dacron (C. R. Bard, Haverhill, PA) felt ring is placed around the opening. The LDM was then wrapped around the mandrel approximately 2 times and sewn to the mandrel with 2-0 polypropylene suture, thereby creating the SMV. This technique has been previously diagrammed [3]. The SMV was left in a subcutaneous position near the left axilla. A pacemaker (Itrel models 7420-1; Medtronic, Inc) was placed in the left flank and connected to the cuffed electrode.
Muscle conditioning
After a 3-week vascular delay period, the SMV was stimulated with an electrical pulse at a frequency of 2 Hz, duration of 210 µs, and amplitude of 1.0 to 4.0 V. This regimen has been shown to result in nearly 100% conversion of fast-twitch (type II) muscle fibers to fatigue-resistant slow-twitch (type I) muscle fibers [4]. This period of muscle conditioning was carried out for 4 to 6 weeks.
Connection to circulation
After conditioning of the SMV, a second operation was performed to connect the SMV to the systemic circulation. An incision was made caudal to the SMV, and the mandrel was removed from the SMV. A left lateral thoracotomy at the fifth interspace was performed, and the fifth rib resected. The two 13-mm PTFE (Impra, Inc, Tempe AZ) limbs of a bifurcated graft were anastamosed (end to side) to the descending thoracic aorta during partial aortic occlusion. The mouth of the graft was then attached to the base of the SMV (see Fig 1).
A cotton umbilical tape was used to ligate the aorta between the limbs to obligate flow though the SMV. Then a cardiac electrode was placed on the epicardium and attached to a cardiomyostimulator (4710 Medtronic, Inc) along with the cuffed nerve lead. The chest was closed, and a chest tube was left in place for 1 to 2 days. Aspirin 81 mg p.o. per day was started 5 to 7 days postoperatively as the only means of anticoagulation.
View larger version (70K):
[in this window]
[in a new window]
|
Fig 1. The skeletal muscle ventricle (SMV) aortic counterpulsator model. (ECG = electrocardiogram; PTFE = polytetrafluoroethylene.)
|
|
At this procedure, after the SMV was connected to the arterial circulation, it was set to contract during cardiac diastole at a ratio of 1:2 to 1:3. The SMVs were pumping blood for at least 3 weeks before initiation of the rapid ventricular pacing.
Rapid ventricular pacing
After the 3-week recovery period, the animals were returned to the operating room for institution of rapid ventricular pacing (RVP) at 220 to 230 beats per minute, which has been shown to produce a reliable model of dilated cardiomyopathy with biventricular dysfunction [5, 6]. This was accomplished by moving the epicardial lead from the cardiomyostimulator to a modified rapid pacemaker (Medtronic, Inc). The nerve electrode was left in the cardiomyostimulator, which was then set to stimulate the SMV asynchronously at 60 contractions per minute (33 Hz, 4.0 to 6.0 V). Before initiating RVP, baseline hemodynamic measurements were taken. Repeat measurements were taken after 4 and 7 weeks of RVP. After several weeks of RVP, some dogs experienced anorexia or respiratory distress as a result of the development of severe heart failure. For respiratory distress, lasix was initiated at 1.5 to 3 mg/kg/day in 3 dogs after the 4-week measurement and withheld for at least 24 hours before the 7-week measurement. If symptoms were severe, the RVP was discontinued for 24 hours and then reinitiated with no measurements taken within 7 days of resuming RVP. In 1 dog, the RVP was reduced to 200 beats per minute at the 6-week mark.
Hemodynamic data collection
At each measurement session, the dogs were placed under general anesthesia. Thiopental was used to induce anesthesia at the baseline measurement, while a combination of intravenous ketamine (0.5 to 1.0 mg/kg) and valium (0.25 to 0.5 mg/kg) was used instead of thiopental at the 4- and 7-week measurements, because of the risk of cardiac arrest with the use of thiopental in the heart failure dogs. Again, anesthesia was maintained using isoflurane (1.5% to 2.0%). A left neck incision was made, and the left carotid artery was located and cannulated with a 6F micro-transducer tipped catheter (Millar Instruments, Inc, Houston, TX). Arterial pressure readings were taken at the proximal aorta and then the catheter was advanced into the left ventricle to obtain a left ventricular pressure tracing. For the purpose of data collection, the rapid pacemaker was temporarily turned off during the 4- and 7-week measurements sessions and the dog allowed to return to sinus rhythm. The cardiomyostimulator was programmed to send a burst of stimuli (33 to 50 Hz, 4.0 to 6.0 V) during cardiac diastole at a rate of 1:2 or 1:3. The burst duration was roughly 40% of the R-R interval and the burst was delayed from the R-wave so that the last pulse was at, or near, the next R-wave. Cardiac output and stroke volume were recorded using the ultrasonic flow probe. At the 7-week measurement, 3 of the dogs also had a 2-mm ultrasonic flow probe placed around either the left anterior descending artery (n = 1) or the circumflex artery (n = 2) to assess coronary blood flow. All data were recorded with the Gould 4600 Series Signal conditioner and CODAS data acquisition software (Dataq, Akron, OH).
Data reduction and analysis
For standardization purposes, 10 consecutive beats were averaged for each parameter. The left ventricular end-diastolic pressure (LVEDP) readings were taken as the left ventricular pressure at the peak of the R-wave. The diastolic pressure-time index (DPTI) was defined as the area under the aortic pressure tracing during diastole minus the area under the left ventricular pressure tracing during diastole. The tension-time index (TTI) was defined as the area under the left ventricular pressure curve from the first upward deflection of the left ventricular (LV) pressure signal during systole to the maximum negative rate of change of LV pressure. The endocardial viability ratio was calculated as DPTI/TTI. For those parameters measuring diastolic function (mean diastolic pressure, DPTI) assisted beats were defined as beats in which there was contraction of the SMV. The parameters of systolic function (peak left ventricular ejection velocity, stroke volume, TTI) were defined as beats following relaxation of the SMV. The data presented throughout the article includes all 6 dogs entered into the study, 6 dogs for the baseline and 4-week measurements, and 4 dogs for the 7-week measurements. However, only those animals that completed the full 7-weeks of RVP were included in the statistical analysis (n = 4). Assisted versus nonassisted beats were compared using two-factor repeated-measures analysis of variance (time of RVP and assist), and where significant interactions were found, comparisons of interest were performed by a C-matrix test, with statistical significance defined as p value less than 0.05 (Systat). At the terminal experiment, coronary blood flow during assisted versus nonassisted beats was compared by using a paired t test.
|
Results
|
|---|
All 6 of the animals entered into the study had uneventful postoperative courses and underwent baseline measurements along with measurements after 4 weeks of RVP. One of the animals did not recover from the 4-week hemodynamic measurement and expired. Another animal developed thrombosis of its SMV circuit at approximately 6 weeks and was terminated shortly thereafter. The other 4 dogs underwent a terminal hemodynamic study after 7 weeks of RVP.
After 4 weeks of RVP, the animals were in moderate-to-severe heart failure. The LVEDP rose from 8.5 ± 1.6 mm Hg at baseline to 29.0 ± 7.3 mmHg at the 4-week measurement. LVEDP continued to increase with further RVP to 34.9 ± 5.9 mm Hg after 7 weeks of RVP. Other measures of left ventricular function such as stroke volume and mean diastolic pressure were significantly reduced from baseline at 4 weeks, but did not change after an additional 3 weeks of RVP. Table 1
shows the decline in the unassisted hemodynamic parameters with RVP.
Diastolic pressure augmentation during assisted beats was moderate at baseline with an increase in the DPTI of 12.1% (p < 0.05). This was markedly increased at the 4-week and 7-week measurement with increases of 31.3% (p < 0.05) and 33.6% (p < 0.05), respectively. At the 4-week and 7-week measurements, there was also significant afterload reduction with SMV functioning as seen by an increase in peak LV ejection velocity and stroke volume. The LV ejection velocity was increased by 24.3% at 4 weeks and 22.7% (p < 0.05) after 7 weeks of RVP, while stroke volume increased 5.0% (p < 0.05) at 4 weeks and 6.2% (p < 0.05) at 7 weeks. If the two early deaths are included, the increases in peak LV ejection velocity and stroke volume averaged 23.2% and 7.2%, respectively, at each animal’s last measurement.
Tension-time index, which tends to correlate with myocardial oxygen consumption, was only slightly reduced and did not reach significance at the 4-week or 7-week measurements with a reduction of 1.9% and 1.6%, respectively. This decrease in TTI resulted in increases in the calculated endocardial viability ratio of 34.1% (p < 0.05) at the 4-week measurement and 35.7% (p < 0.05) at the 7-week measurement. Coronary blood flow was also increased during assisted beats by an average of 47.6% (0.375 ± 0.29 versus 0.432 ± 0.26 mL/beat, p = 0.044), although the numbers are small (n = 3) and there was considerable variability (ie, 8%, 18%, and 117% increases). Complete assessment of hemodynamic assistance from the SMV is presented in Table 2.
|
Comment
|
|---|
Complications
Two of the 6 dogs in the study did not make it to the 7-week measurement. One dog developed significant aortic regurgitation shortly after introduction of the pressure catheter into the left ventricle at the 4-week measurement. This dog was in severe heart failure and was unable to recover from the measurement session. Fortunately, this occurred at virtually the end of the measurement session with the required data already being recorded. In the 2nd dog, the SMV circuit thrombosed between the 4- and 7-week measurement. Therefore no 7-week measurement was made for this dog; however, the late thrombosis did not affect the previously completed 4-week measurement. A recent large review of SMVs showed that thrombosis of the entire circuit is relatively rare [7], and this problem has declined with the use of a pericardial lining for the SMV [1]. This may be related to the lack of tissue factor production by pericardial mesothelial cells, leading to less activation of the extrinsic coagulation pathway [8]. In this study, during the period of RVP, the SMV was set to contract asynchronously at a rate of 60/min. This asynchronous contraction may have led to an increased incidence of thrombosis in this group (5 of 6 animals demonstrated some thrombus formation at the apex of the SMV) and possibly contributed to the circuit thrombosis in the 1 dog. A second explanation may be that, on autopsy examination of this animal, it was found that the aorta was not completely ligated allowing for some flow around the SMV circuit. Incomplete ligation of the aorta has been shown in a previous study to increase the incidence of thrombosis [9].
Diastolic pressure augmentation
A goal of diastolic pressure augmentation is increased coronary blood flow, and can be estimated by the DPTI, which subtracts the LV diastolic pressure from the aortic diastolic pressure. In this group of animals, SMV contraction resulted in a moderate, but highly significant, increase in the DPTI of 12.1% (p = 0.005). Pressure augmentation was sustained after development of congestive heart failure, and because of the increase in the LV diastolic pressure, the percentage of increase in the DPTI rose to greater than 30%.
Afterload reduction and timing of SMV contraction
Relaxation of the SMV should result in afterload reduction similar to what is seen with deflation of the balloon in an intraaortic balloon pump (IABP). During each measurement session (ie, baseline, 4-week, and 7-week) with the rapid pacer temporarily off, the inotropic state of the heart was not altered and should not have changed from beat to beat. Therefore, significant afterload reduction should lead to an increase in peak LV ejection velocity or stroke volume. A previous study with the aortic counterpulsators from this laboratory failed to show an increase in these parameters [2]. Perhaps, this was because those animals were not studied in chronic heart failure. Another explanation may be that the timing of SMV contraction selected in previous studies was not appropriate for maximum afterload reduction.
In prior experiments, timing of SMV contraction was based upon a study showing maximum diastolic pressure increase with a 40% delay from the R-wave [10], as well as early work done with the IABP where the goal was to reduce presystolic aortic pressure. This was known as "systolic unloading" and was thought to reduce LV stroke work by reduction of afterload. The benefit of "systolic unloading" has been questioned. One study where coronary blood flow was controlled, found no benefit in myocardial oxygen consumption or cardiac index from "systolic unloading" at lower coronary perfusion pressures [11]. Recent work with the IABP demonstrates that balloon deflation with conventional timing may be too early, and that balloon deflation coupled with isovolumic contraction leads to a larger increase in LV mechanical efficiency [12]. In our study, we found that delaying contraction of the SMV (thereby delaying SMV relaxation), with the last stimulus occurring at, or near, the next R-wave did not lead to the characteristic reduction in presystolic aortic pressure, but resulted in increased peak LV ejection velocity and stroke volume. Figure 2
compares conventional timing of SMV versus delayed contraction in the same dog after 7 weeks of RVP. Note the minimal change in aortic root flow with conventional timing.
View larger version (71K):
[in this window]
[in a new window]
|
Fig. 2. Electrocardiographic (ECG), aortic pressure, and aortic root flow tracings from a typical dog after 7 weeks of rapid ventricular pacing). Conventional timing (A), and delayed contraction (B). Note the decrease in pre-systolic pressure, however with no effect on aortic root flow with the conventional timing.
|
|
Coronary blood flow
At the 7-week terminal measurement coronary blood flow was measured for 3 animals in either the left anterior descending or circumflex artery. Both the IABP and aortomyoplasty have been shown to increase coronary blood flow in animals with heart failure [13, 14]. Coronary blood flow increased on average 47.6% (p = 0.044) in this study. Because of the few numbers and large variability, it is difficult to say from this study what amount of coronary blood flow augmentation can be expected in this model. It can be said that there is augmentation of coronary blood flow that correlates with augmentation of the DPTI. Another interesting result from this study was that traditional timing of SMV contraction may actually lead to reversal of coronary flow during the presystolic drop in aortic pressure, and that delayed contraction may improve augmentation of coronary blood flow. Figure 3
illustrates the difference in coronary blood flow with conventional and delayed SMV contraction.
View larger version (43K):
[in this window]
[in a new window]
|
Fig 3. Electrocardiographic (ECG) and coronary blood flow (from the left anterior descending artery) tracings in a dog after 7 weeks of rapid ventricular pacing. Conventional timing (A), and delayed contraction (B). Note the reversal of coronary blood flow with the conventional timing (arrow).
|
|
Conclusion
The results of this study indicate three important facts. First, SMVs can perform effectively during chronic heart failure. Secondly, because of the decreased performance of the LV in this heart failure model, the relative benefit of the aortic counterpulsator is increased. This can be seen as increases in the percentage of augmentation of the DPTI as well as increases in the peak LV ejection velocity and stroke volume, which was not seen in the normal dogs. Lastly, the conventional timing of SMV contraction, used by our laboratory in the past and others [2, 14], may not be appropriate for maximum afterload reduction and augmentation of coronary blood flow.
|
Acknowledgments
|
|---|
This work was supported by NIH grant NHLBI-RO1-34778-15.
|
References
|
|---|
-
Thomas G.A., Hammond R.L., Greer K., et al. Functional assessment of skeletal muscle ventricles after pumping for up to four years in circulation. Ann Thorac Surg 2000;70:1281-1290.[Abstract/Free Full Text]
-
Thomas G.A., Lu H., Isoda S., et al. Pericardium-lined skeletal muscle ventricles in circulation up to 589 days. Ann Thorac Surg 1994;58:978-988.[Abstract/Free Full Text]
-
Thomas G.A., Lelkes P.I., Chick D.M., et al. Skeletal muscle ventricles seeded with autogenous endothelium. ASAIO J 1995;41:204-211.[Medline]
-
Mannion J.D., Bitto T., Hammond R.L., Rubinstein N.A., Stephenson L.W. Histochemical and fatigue characteristics of conditioned canine latissinmus dorsi muscle. Circ Res 1986;58:298-304.[Abstract/Free Full Text]
-
Wilson J.R., Douglas P., Hickey W.F., et al. Experimental congestive heart failure produced by rapid ventricular pacing in the dog: cardiac effects. Circulation 1987;4:857-867.
-
Patel H.J., Pilla J.J., Polidori D.J., et al. Ten weeks of rapid ventricular pacing creates a long-term model of left ventricular dysfunction. J Thorac Cardiovasc Surg 2000;119:834-841.[Abstract/Free Full Text]
-
Nakajima H., Nakajima H.O., Thomas G.A., et al. Chronic morphologic changes of skeletal muscle ventricles in circulation. Ann Thorac Surg 1994;57:912-920.[Abstract/Free Full Text]
-
Chung J.H., Gikakis N., Rao A.K., Drake T.A., Colman R.W., Edmunds L.H. Pericardial blood activates the extrinsic coagulation pathway during clinical cardiopulmonary bypass. Circulation 1996;93:2014-2018.[Abstract/Free Full Text]
-
Greer K., Lu H., Hammond R.L., Bastian S., Vanderheide R., Stephenson L.W. Skeletal muscle ventricles full versus half ligation. J Card Surg 1998;13:242-251.[Medline]
-
Mocek F.W., Hammond R.L., Thomas G.A., et al. Skeletal muscle ventricles: improved left ventricular performance associated with a prolongation of ventricular relaxation. J Surg Res 1995;59:569-577.[Medline]
-
Akyurekli Y., Taichman G.C., Keon W.J. Effectiveness of intra-aortic balloon counterpulsation on systolic unloading. Can J Surg 1980;23:122-126.[Medline]
-
Sakamoto T., Suzuki A., Kazama S., Komatsu S., Sasaki S., Shoji Y. Effects of timing on ventriculoarterial coupling and mechanical efficiency during intraaortic balloon pumping. ASAIO J 1995;41:M580-M583.[Medline]
-
Feola M., Haiderer O., Kennedy J.H. Intra-aortic balloon pumping (IABP) at different levels of experimental acute left ventricular failure. Chest 1971;59:68-76.[Abstract/Free Full Text]
-
Ukoha O.O., Braxton J.H., Letsou G.V., Hogan J.F., Ariyan S., Elefteriades J.A. Skeletal muscle ventricles in continuity with the bloodstream. J Card Surg 1996;11:271-279.[Medline]
This article has been cited by other articles:
|
|
|
|
 
I. R. Ramnarine, M. Capoccia, Z. Ashley, H. Sutherland, M. Russold, N. Summerfield, S. Salmons, and J. C. Jarvis
Counterpulsation From the Skeletal Muscle Ventricle and the Intraaortic Balloon Pump in the Normal and Failing Circulations
Circulation,
July 4, 2006;
114(1_suppl):
I-10 - I-15.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
