| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1999;67:1304-1311
© 1999 The Society of Thoracic Surgeons
a Jewish Hospital Cardiovascular Research Center, Divisions of Thoracic and Cardiovascular Surgery, University of Louisville School of Medicine, Louisville, Kentucky, USA
b Division of Plastics and Reconstructive Surgery, Department of Surgery, University of Louisville School of Medicine, Louisville, Kentucky, USA
Accepted for publication October 6, 1998.
Address reprint requests to Dr. Slater, Division of Thoracic and Cardiovascular Surgery, Rudd Heart Lung Center, 201 Abraham Flexner Way, Suite #1200, Louisville, KY 40292
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsDiscussionStudy limitationsAcknowledgmentsReferences |
|---|
Methods. Mongrel dogs (n = 24) were divided in four groups: 1) controls (n = 6), single-stage CMP; 2) Group ES (n = 6), single-stage CMP with early LDM stimulation beginning 48 h, postoperatively; 3) Group VD (n = 6), vascular delay of the LDM followed by CMP without early LDM stimulation, and 4) Group VDES (n = 6), vascular delay of LDM (14–18 days), followed by CMP with early stimulation (48 h postoperatively). Two weeks after CMP, global cardiac dysfunction was induced by injecting microspheres into the left coronary artery. LDM-assisted (S) beats were compared with nonstimulated beats (NS) by measuring aortic pressure (AoP), LV pressure, aortic flow, and by calculating first derivative of LV contraction (±dP/dt), stroke volume (SV), and stroke work (SW).
Results. In ES, LDM stimulation had no effect on the hemodynamic parameters. In the other groups, LDM stimulation significantly (p < 0.05) increased AoP, LVP, dP/dt, SV, and SW. However, these increases were much larger in VD and VDES. In VD, LDM stimulation increased peak AoP by 21.5 ± 3.8 mm Hg, LVP by 22.1 ± 4.1 mm Hg, dP/dt by 512 ± 163 mm Hg/sec, SV by 10.4 ± 2.3 mL, and SW by 22.1 ± 5.4 g/m-1. Similarly, in VDES, LDM stimulation increased peak AoP by 24.1 ± 4.7 mm Hg, LVP by 26.2 ± 4.3 mm Hg, dP/dt by 619 ± 47 mm Hg/sec, SV by 6.5 ± 0.7 mL, and SW by 16.7 ± 4.1 g/m-1.
Conclusions. In dogs with global LV dysfunction, CMP after vascular delay resulted in a significant improvement in hemodynamic function measured 2 weeks after surgery. This improvement was not provided by single-stage CMP with or without early stimulation. Vascular delay of the LDM before surgery may play an important role for early benefit after CMP, shorten the overall muscle training period, as well as increase hemodynamic response to LDM stimulation.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionStudy limitationsAcknowledgmentsReferences |
|---|
The training of the LDM starts 2 weeks after CMP surgery and lasts for 12 weeks. During this 12-week period, the LDM is transformed into a fatigue-resistant muscle. The rationale for the delayed stimulation has been to allow the thoracodorsal artery to gradually adapt itself as the main blood supply to the LDM and to have adhesions develop between the LDM and the epicardium [3, 6]. Thus, substantial systolic assistance by the LDM is not available initially after surgery, when it is required the most.
Ischemic changes and muscle degeneration have been noted in the distal third of the LDM with standard single-staged CMP [4, 5, 8–11]. These degenerative changes are due to loss of muscle structure in the distal flap, and could be responsible for the limited hemodynamic benefits observed after surgery [9, 10].
Vascular delay of the LDM has been shown to result in improved blood supply and contractile strength by preserving the muscle architecture [12–16]. We hypothesized that cardiomyoplasty after vascular delay would significantly improve LDM performance, resulting in augmentation of the hemodynamic indices by active systolic support of the left ventricle. We also hypothesized that vascular delay of the LDM before cardiomyoplasty would allow for an earlier LDM stimulation after CMP surgery.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionStudy limitationsAcknowledgmentsReferences |
|---|
In this study, two interventions were examined: vascular delay of the LDM before CMP surgery, and early LDM stimulation after CMP surgery (A 2 x 2 block design). Accordingly, the dogs (27–31 kg, n = 24) were divided into 4 groups (Table 1). Control animals underwent standard CMP without LDM preconditioning and without early LDM stimulation. The dogs in the early stimulation group (Group ES) underwent standard CMP without LDM preconditioning but with early (48-h) LDM stimulation after CMP surgery. The dogs in the vascular delay group (Group VD) had a vascular delay of the LDM followed by CMP surgery, but without early LDM stimulation. Last, the dogs in the vascular delay group plus early stimulation group (Group VDES) had a vascular delay of the LDM followed by CMP surgery and early LDM stimulation after CMP surgery (Fig 1).
|
|
Vascular delay
The dogs in groups VDES or VD underwent a vascular delay procedure of the latissimus dorsi before CMP surgery. After fasting overnight, dogs were anesthetized with sodium thiopental (2.5% pentothal sodium) 15–25 mg/kg IV and atropine 0.01 mg/kg IM. The animals were intubated and placed on a ventilator (Quantiflex; VMC Anesthesia Machine, Orchard Park, NY). Anesthesia was maintained by 2% isoflurane (Isoflurane Vaporizer; Oharda, Aushell, GA), 0.5–1.0% nitrous oxide, and oxygen. Using a sterile technique, a 15–20-cm left oblique incision was made parallel to the free (anterior) border of the LDM. The margin of the LDM was elevated, and perforating intercostal arteries supplying the muscle were ligated. The subcutaneous surface of the muscle was left undisturbed. Incision was closed in layers using absorbable sutures. The animals were then returned to their cages and allowed to recover. Fourteen to 18 days later, CMP was performed as described below.
Cardiomyoplasty
The CMP procedure was performed by the technique used previously in our laboratory and as described below [2, 17, 18]. Anesthesia was induced as described above and animals were positioned with their left side up. An oblique 15–20-cm incision was made parallel to the anterior border of the left LDM (previous incision was used in dogs with vascular delay). The muscle was elevated and carefully taken down from all its attachments, except proximally on the humerus. The neurovascular pedicle with the thoracodorsal nerve and artery was identified and preserved. The tendon of the LDM was then carefully isolated and dissected out. Epimysial leads (model YY38403403; Medtronic Inc, Minneapolis, MN) were implanted on the pedicle with nylon sutures. A-V pacing system analyzer (5311; Medtronic Inc) was used to determine the stimulation threshold. An approximately 5-cm section of the left second rib was removed and the LDM rotational flap moved inside the chest. A muscle stimulator (ITREL 7420; Medtronic Inc.) was implanted and connected to the epimysial leads in all groups. The wound was closed in layers.
The animal was repositioned on its back and the chest opened by median sternotomy. After opening the pericardium, an aortic flow probe (A-series 16–20-mm flow probes; Transonic Systems Inc, Ithaca, NY) was placed around the ascending aorta. The lead for the aortic flow probe was placed subcutaneously under the muscles lateral to the spine for later access during the evaluation studies. The LDM was wrapped in a clockwise direction as a posterior wrap around the heart, approximating the costal surface of the LDM to epicardium. The margins of the LDM were anchored to the pericardium in the atrioventricular (AV) groove. The free margin of the LDM was approximated such as to completely wrap the heart. We were able to perform a 360-degree wrap of the LDM around the myocardium in all the animals. Bilateral chest tubes were inserted, and the chest was closed using steel sutures. Animals were extubated and returned to their cages. Intravenous bupreneorphine hydrochloride 0.3–0.6 mg (Bupernex; Reckitt & Coleman Pharmaceuticals Inc, Richmond, VA) was used for analgesia every 3–4 h as needed during the first 48 h. Acepromazine maleate 0.25–0.5 mg IM (PromAce; Fort Dodge Lab Inc, Fort Dodge, IA) was used every 10–12 h for sedation in the initial 24 h. Animals were positioned to lie on their right side overnight. Chest tubes were removed on the first operative day. Antibiotics (cefazolin 500 mg and gentamicin 75 mg every 12 h) were used for 48 h postoperatively.
Postoperatively, the ITREL stimulator was not switched on for controls and group VD animals. In the ES and VDES animals, asynchronous LDM stimulation was started from the second postoperative day using twice the threshold voltage, 30-Hz pulse frequency, and a pulse train duration of 185 msec. The LDM was stimulated at three pulse trains/min in the first week and six pulse trains/min during the second week.
Hemodynamic evaluation
Experimental prepararation
All animals were evaluated 15 ± 2 days after the CMP procedure. Anesthesia was induced as described previously and maintained on intravenous phenobarbital 50–100 mg/h. Atropine was not used. Four arterial cut-downs were made to place introducer catheters through both carotids and femoral arteries. Micro-manometer-tipped catheters (MILLAR; Instruments Inc, Houston, TX) were used to measure pressure. Using fluoroscopic guidance, the MILLAR catheters were placed via the femoral and carotid arteries into the left ventricle and the descending aorta, respectively. Analog signals from the pressure transducers were amplified (PM-1000; CWE Inc, Admore, PA). The aortic flow probe cable was dissected out and connected to a flow meter (206T; Transonic Systems Inc). Three animals in group ES and three controls did not have a flow probe. In these animals, stroke volume was calculated using a conductance catheter (Lycom; CardioDynamics BV, Rijnsburg, the Netherlands).
The epimysial leads were disconnected from the ITREL pulse generator and connected to an external muscle stimulator (8800; Grass Systems Inc, Quincy, MA). A cardiotachometer (1000; CWE Inc, Admore, PA) detected the QRS waveform from the analog electrocardiogram (ECG) signal and, in turn, triggered the muscle stimulator. This resulted in synchronized LDM stimulation with the R wave.
Data acquisition
Data were recorded simultaneously on a chart recorder (TA-11; Gould Instrument Systems Inc, Cleveland, OH) as well as on a computer (Winbook XP5 120-MHz Pentium; Winbook Computer Corp, Columbus, OH). The pressure, flow, and ECG signals were digitized using an A/D circuit board (DAS-1601; Keithley Metrabyte, Taunton, MA). The data were acquired using software (Labtec Notebook, Version 9.0; Laboratory Technologies Corp, Wilmington, MA). The muscle stimulator was switched on every fourth to sixth beat using at least four times the threshold voltage. Pulse train duration was adjusted between 150 and 190 msec, pulse duration at 0.5 msec, interpulse interval of 15–20 msec, and pulse train delay after the R wave was 20–80 msec. Each data run was 30 sec long, with an interval of at least 3–5 min between files. The ventilator was switched off during data acquisition to avoid respiratory variations.
Cardiac dysfunction
We induced cardiac dysfunction by sequential intracoronary microsphere injection [19]. Under fluoroscopy, a left 5 or 6 F coronary catheter was positioned in the left anterior descending artery. Latex microspheres (1.0–2.0 x 105, 90 ± 2 µm diameter; Polyscience Inc, Warrington, PA) were injected in the left main coronary artery to create acute cardiac dysfunction. The intracoronary injections were verified by using iodine base contrast material (Renografin, 76 Bristol; Myers-Squibb Co, Princeton, NJ). Latex microspheres were injected sequentially until the peak positive dP/dt decreased by at least 20%.
After inducing left ventricular dysfunction, four to five additional data runs were obtained. After the experiment, the anesthetized animals were killed by intravenous injection of pentobarbital (10 mg/kg), followed by intravenous saturated potassium chloride solution (20–30 mL).
Morbidity and mortality
One animal (#5) in the vascular delay group, while under going final evaluation, had a fatal arrhythmia. Thus, in this animal, we only included the hemodynamic evaluation of the normal myocardium. About 50% of the animals had large seroma collection after vascular delay and before cardiomyoplasty. The fluid collection had to be drained in 3 animals preoperatively due to infection. However, it did not prevent the animals from completing the study. It was not possible to leave self-contained drains in these animals. Almost all the animals had swelling of the left forelimb for 2–3 days after cardiomyoplasty, which subsided in all but one animal.
Data analysis
Using software developed in Visual Basic for Excel (Microsoft Excel 7.0; Microsoft Inc, Redwood, WA), hemodynamic variables were extracted from a digitally stored data file. Ectopic beats and postectopy beats were excluded from the analysis. For each beat, the end-diastolic pressure, the peak ventricular systolic pressure, the peak positive and negative first derivative of the left ventricle pressure (+dP/dt, -dP/dt), and peak and end-diastolic aortic pressures were determined, and stroke volume and stroke work were calculated. Absolute values of the changes (stimulated beat value - nonstimulated beat value) were calculated. Data were reported as mean ± standard error of the mean. In the results, only the data obtained after inducing left ventricular dysfunction were reported. The only exception was one animal in group VD, where the dog had a massive myocardial infarction while positioning the intracoronary catheter. In this animal, hemodynamic data obtained before the microsphere injections were used.
Statistical analysis
Software for statistical analysis (StatView 4.5; Abacus Concepts, Berkley, CA) was used for statistical analysis. Within each group, the stimulated beats were compared with the immediately preceding nonstimulated beats using a paired Student’s t test (p < 0.05). Between groups, a two-way analysis of variance (ANOVA) (vascular delay and early stimulation) was used to compare the absolute changes (
). A comparison between each group was performed using a post hoc ANOVA for multiple comparisons. Statistical significance was determined if p < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionStudy limitationsAcknowledgmentsReferences |
|---|
|
|
|
|
|
|
|
|
Figures 4–8 show absolute increases in peak systolic aortic and left ventricular pressures, stroke volume, stroke work, and peak positive left ventricular dP/dt for each group. The hemodynamic response to LDM stimulation was similar in the vascular delay and the vascular delay plus early stimulation group.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionStudy limitationsAcknowledgmentsReferences |
|---|
Vascular delay of the LDM
CMP surgery involves severing the perforating intercostal arteries to the LDM. The muscle is then transferred inside the chest as a rotational flap. The muscle’s resting tension is lost, as well as most of the blood supply to the distal half of the muscle. These changes have been shown to be detrimental to the muscle structure [8, 16]. Tobin and associates examined intramuscular vascular territories in canine LDM and from fresh human cadavers using radiographic planimetry [16]. They determined that intercostal arteries contributed 67% and 69% of the vascular supply to the LDM in humans and dogs, respectively. Moreover, LDM mobilization has been shown to decrease the perfusion in the distal LDM by more than 90%. In a canine model, Cruz and associates measured LDM force development, shortening, and blood flow [25]. They demonstrated that loss of LDM function was most apparent after mobilizing and reattaching the muscle, and that the resting blood flow was significantly decreased in the mid and distal LDM.
Associated with a decrease in distal blood flow, fibrosis of the LDM damage has been reported [8, 10, 19, 20, 25]. Cheng and associates examined full-thickness biopsies of the left ventricle and the LDM. The proximal LDM was normal. However, in the distal LDM, a large area of fibrosis was observed with extensive degeneration of the LDM [18]. Moreover, Lucas and associates have shown in goats that damage to the LDM correlated with poor hemodynamic outcome [10]. Clinically, Kalil-Filho and associates [22] used magnetic resonance imaging to evaluate chronic LDM. The thickness of the LDM decreased from 19.6 mm at 15 days after surgery to 7.6 mm at 24 to 52 months after surgery [22]. Additionally, the signal intensity of the LDM was comparable with thoracic skeletal muscle shortly after surgery, but by 24 months the signal intensity was comparable with subcutaneous fat.
Several studies have examined vascular delay as a means to preserve the LDM. In this procedure some of the arteries supplying a muscle or tissue are ligated. The muscle is left in its original position for several days (delayed) before being moved. In dogs, Isoda and associates demonstrated that a 1-month vascular delay period significantly enhanced muscle flap perfusion at rest and during exercise [15]. After a 10-day vascular delay, Carroll and associates [14] wrapped the LDM around a silicone tube simulating CMP. Two weeks after the procedure, they demonstrated that vascular delay resulted in improved perfusion to the middle and distal LDM during exercise and improved fatigue resistance [14].
You and associates [26] performed vascular delay of the LDM followed by cardiomyoplasty, and showed that vascular delay of the LDM preserved normal muscle architecture. In their study, LDM contraction resulted in significant improvement in hemodynamic indices (left ventricular elastance increased from 0.77 ± 0.14 to 1.00 ± 0.17 mm Hg/mL and stroke volume from 6.3 ± 1.2 to 8.3 ± 1.1 mL). However, the effects of LDM stimulation were only observed acutely after surgery [26]. In the present study, two groups of animals (VDES and VD) had vascular delay of the LDM before CMP. One of these groups had early LDM stimulation after surgery at 48 h. In both the groups, LDM simulation 2 weeks after CMP surgery caused very substantial increases in peak left ventricular aortic pressures, ± dP/dt, and stroke work. These changes were greater than the effect of LDM stimulation in the groups without vascular delay. Furthermore, augmentation of LV hemodynamic parameters was observed consistently in all 12 animals. The statistical differences between the hemodynamic changes with LDM stimulation was compared through the magnitude of the LDM assistance to LV systolic function. In another study, this was performed in normal and with progressive degree of myocardial damage. It was observed that the magnitude of LDM assistance was not affected by the degree of LV dysfunction [17]. Thus, in the present study, differences in the baseline pressures and stroke volume did not affect the absolute magnitude of changes in the left ventricular pressure and output with LDM stimulation.
Clinically, LDM training is initiated with one pulse every second to third heart beat starting at the third postoperative week. With only one, two, or three pulses, the LDM does not contract sufficiently to assist the heart. Thus, current clinical protocols for LDM stimulation do not benefit the patient in the first few weeks after surgery, and has been termed the "preassist period" [1, 2]. There is a significant morbidity and mortality associated with the CMP, and the early series have reported mortality up to 30%–35% during the first 90 days after CMP or during the "preassist" period [1–3].
Recent experimental studies suggest that early stimulation may be possible. In sheep, Chekanov and associates [21] performed a partial mobilization of the LDM. Each day, the LDM was stimulated, for 20 min and increased to 50 min over 2 weeks. On the final examination, during 20 min of continuous LDM contraction, no fatigue was evident at the end of testing.
In this study, complete muscle stimulation was initiated on the second postoperative day in the early stimulation animals (ES and VDES groups). Stimulation was initiated with pulse trains consisting of six pulses at 30 Hz. This is the current clinical pulse train that is achieved only after 12 weeks of training. By using the six-pulse train whenever the LDM contracts, the muscle can provide hemodynamic assistance. The LDM was stimulated with three pulse trains/min for the first week, and six pulse trains/min during the second week. In the early stimulation group without vascular delay, the effects of early LDM stimulation did not have any hemodynamic benefit. On the other hand, the group with early post-CMP stimulation that had vascular delay showed large hemodynamic responses for all hemodynamic parameters. Furthermore, the effects of LDM stimulation in the VDES group was similar to the VD group, where the LDM was not trained during the first 2 weeks. Thus, no loss of hemodynamic function was observed despite early stimulation for group VDES.
|
Study limitations |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionStudy limitationsAcknowledgmentsReferences |
|---|
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionStudy limitationsAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionStudy limitationsAcknowledgmentsReferences |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  S. Salmons Cardiac assistance from skeletal muscle: a reappraisal Eur. J. Cardiothorac. Surg., February 1, 2009; 35(2): 204 - 213. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Feindt, U. Boeken, J.D. Schipke, J. Litmathe, N. Zimmermann, and E. Gams Ventricular constraint in dilated cardiomyopathy: A new, compliant textile mesh exerts prophylactic and therapeutic properties J. Thorac. Cardiovasc. Surg., October 1, 2005; 130(4): 1107 - 1107. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Rigatelli, U. Carraro, M. Barbiero, R. Riccardi, F. Cobelli, M. Gemelli, and G. Rigatelli A Review of the Concept of Circulatory Bioassist Focused on the "New" Demand Dynamic Cardiomyoplasty: The Renewal of Dynamic Cardiomyoplasty? Angiology, May 1, 2003; 54(3): 301 - 306. [Abstract] [PDF]
|
|
|
|
 |
|
 G. Zakine, E. Martinod, P. Fornes, M. Sapoval, D. Barritault, A. F. Carpentier, and J. C. Chachques Growth factors improve latissimus dorsi muscle vascularization and trophicity after cardiomyoplasty Ann. Thorac. Surg., February 1, 2003; 75(2): 549 - 554. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Rigatelli, U. Carraro, M. Barbiero, M. Zanchetta, K. Dimopoulos, F. Cobelli, R. Riccardi, and G. Rigatelli Activity-rest stimulation protocol improves cardiac assistance in dynamic cardiomyoplasty Eur. J. Cardiothorac. Surg., March 1, 2002; 21(3): 478 - 482. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Kashem, W. P. Santamore, B. Chiang, L. Unger, A. T. Ali, and A. D. Slater Vascular delay and intermittent stimulation: keys to successful latissimus dorsi muscle stimulation Ann. Thorac. Surg., June 1, 2001; 71(6): 1866 - 1873. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. Magovern Update 2001: Right latissimus dorsi cardiomyoplasty augments left ventricular systolic performance Ann. Thorac. Surg., June 1, 2001; 71(6): 2077 - 2078. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
