|
|
||||||||
J Thorac Cardiovasc Surg 2005;129:1119-1127
© 2005 The American Association for Thoracic Surgery
Surgery for Congenital Heart Disease |
a IMM Recherche, Centre dExpérimentation et de Recherche Appliquée, Paris
b Cardiac Surgery Department, Rui Jin Hospital, Shanghai, China
c Service dAnatomie Pathologique, Institut Mutualiste Montsouris, Paris
d Département de Pathologie Cardiaque, Institut Mutualiste Montsouris, Paris
e Laboratoire dAnatomie Pathologique, Hôpital Européen Georges Pompidou, Paris
f Unité de Génétique Moléculaire du Développement, Institut Pasteur, Paris, France
Received for publication May 21, 2004; revisions received September 3, 2004; accepted for publication September 20, 2004. * Address for reprints: Nicolas Borenstein, IMM Recherche, 42 boulevard Jourdan, 75014 Paris, France (E-mail: nicolas.borenstein{at}imm.fr).
| Abstract |
|---|
|
|
|---|
METHODS: Sixteen rams were assigned to one of the following groups: group 1, simple pulmonary artery banding (n = 5); group 2, pulmonary artery banding and cell implantation in the right ventricle (n = 7); and group 3, pulmonary artery banding and placebo injection in the right ventricle (n = 4). Hemodynamic assessment with pressure-volume loops was performed on days 0 and 60. The pulmonary artery banding and the injections were achieved through a left fourth intercostal thoracotomy. Autologous myogenic cell implantation was carried out with a noncultured cell preparation, as previously described by our group. Implanted sites were processed with monoclonal antibodies to a fast skeletal-specific isoform of myosin heavy chain (MY32).
RESULTS: Skeletal myosin heavy chain expression was detected at 2 months after noncultured cell implantation in all grafted animals. Right ventricular training resulted in statistically significant increased signs of contractility in all three groups. There was no observed difference, however, between the cell therapy group and the other two groups with respect to signs of cardiac function.
CONCLUSION: Successful engraftment of noncultured cells into right ventricular myocardium did not translate into a functional benefit that we could demonstrate in our ovine model. Cellular therapy thus is probably not useful to strengthen a left ventricle being retrained through pulmonary artery banding before arterial switch operation. However, cell transplantation may affect the outcome of right ventricular failure long term after atrial switch operation. Although preliminary, this investigation paves the way for further research into cellular cardiomyoplasty, right ventricular failure, and congenital heart disease.
|
Transposition of the great arteries is a heart defect that is found in approximately 5% of all neonates with congenital heart malformations.1 Currently, the operation of choice for transposition of the great arteries is the one-stage arterial switch operation (ASO). After birth, progressive regression of the left ventricular mass in simple transposition of the great arteries2,3 restricts the safe period for primary ASO to the first 14 or 25 months of life. If the left ventricle is not able to maintain a systemic pressure, left ventricular retraining is mandatory before switching the morphologically left ventricle under the aorta. Currently, this is achieved by creating a ventricular overload through pulmonary artery banding (PAB) and possibly associated aortopulmonary shunt. First performed by Yacoub and colleagues6 in 1977, left ventricular retraining has shown its limits7; the increase in mass of the left ventricle is accomplished beyond the neonatal period by simple hypertrophy and not by hyperplasia of cardiomyocytes.8
The concept of myogenic cell transplantation into the myocardium, cellular cardiomyoplasty (CCM), is based on the contribution of exogenous cells to replace lost or altered cardiomyocytes to restore functional performance of the heart.9 There is a large body of evidence showing that CCM can improve cardiac function in ischemic heart disease (see recent review10). Thus far, teams have mostly addressed acquired diseases of the adult.11 The aim of this study was therefore to explore whether myogenic cell transplantation could contribute to the hypertrophy obtained by increased right ventricular afterload through PAB in an ovine model. We hypothesized that CCM and PAB would yield better hemodynamic results than PAB alone.
| Methods |
|---|
|
|
|---|
Experimental design
The experimental design was as follows. Sixteen 40 ± 4-kg 4-month-old Ile de France rams were randomly assigned to one of the following groups: group 1, simple PAB (n = 5); group 2, PAB and cell implantation in the right ventricle (n = 7); and group 3, PAB and placebo injection in the right ventricle (n = 4). Hemodynamic assessment was performed on day 0 (just before PB) and at day 60 (just before the animals were put to death).
Anesthesia and analgesia
Animals were anesthetized with midazolam and etomidate, allowing endotracheal intubation and maintenance with isoflurane in oxygen, delivered with a ventilator (Hallowell Engineering and Manufacturing Corporation, Pittsfield, Mass). During pressure-volume loop analysis, fluid therapy, depth of anesthesia, and ventilation were adjusted so that pressure and end-tidal carbon dioxide were within normal ranges. The animals were left to recover with the required analgesic regimen after surgery and the following day as needed.
Right ventricular catheterization and pressure-volume loop analysis
Contractile performance of the right ventricle was assessed before and 2 months after PAB with preload recruitable stroke work (PRSW), the relationship between stroke work and right ventricular end-diastolic volume. Right ventricular volume measurement was performed with the conductance catheter method applied to the right ventricle, as previously described by Hon and coworkers.12 Parallel conductance was measured with the hypertonic saline method.13 Reference cardiac output was measured in triplicate in all animals by using the thermodilution technique.14 The mean of three consecutive measurements was recorded and saved for off-line factor
calibration.
After pressure calibration, a 6F combined conductance-pressure catheter (Millar Instruments, Inc, Houston, Tex) was advanced through the femoral vein introducer into the right ventricle. The conductance catheter was connected to a Leycom Sigma-5 signal conditioner processor (CardioDynamics, San Diego, Calif) to obtain an instantaneous right ventricular volume signal. Real-time signal processing was obtained on a dedicated software (IOX; Emka Technologies SA, Paris, France). At least 10 pressure-volume loops were obtained during apnea at end-expiratory pressure and analyzed off-line for
calibration and to calculate developed pressure (systolic pressure minus diastolic pressure), maximum and minimum rates of change of pressure inside the right ventricle (dP/dtmax and dP/dtmin, respectively). A 23-mm balloon catheter was advanced into the caudal vena cava for inflow occlusion. A series of three caval occlusions was obtained. Pressure-volume loop assessment at day 0 was followed by PAB in all three groups; additionally, either noncultured cell implantation (group 2) or placebo injection (group 3) was performed in the right ventricular wall.
Pulmonary artery banding
A left fourth intercostal thoracotomy was performed. The pulmonary trunk was dissected at its mid portion. The right ventricular pressure was monitored by direct catheterization. Baseline right ventricular pressure, pulmonary pressure, and maximal right ventricular pressure were recorded. The PAB was achieved with a 4-mm wide Dacron polyester fabric band. After dissection, the Dacron tape was passed around the pulmonary trunk. The degree of PAB was adjusted as previously described by Mee.15 The band was then sutured in its tightened position with 5-0 Prolene suture (Ethicon, Inc, Somerville, NJ). Animals were left to recover and were reoperated on 2 months later for final assessment.
Skeletal muscle biopsy
Animals of group 2 were anesthetized 3 hours before PAB to perform the muscle biopsy. Approximately 10 g muscle tissue was explanted from the left femoral biceps under sterile conditions. The biopsy tissue was kept at room temperature until mechanical and enzymatic digestions were started. The operative wound was closed in a routine fashion.
Muscle cell extraction
In animals of group 2, the noncultured cell preparation was performed as previously described by our group.16 This technique is based on the extraction of satellite cells from a skeletal muscle, mechanical and enzymatic digestion, and reimplantation of the cell preparation approximately 3 hours after muscle biopsy. Briefly, the muscle was minced with scissors to a slurry. To release satellite cells, the muscle fragments were then incubated at 37°C under agitation in 10 mL of Dulbeccos modified Eagles medium (DMEM) supplemented with 0.2 % (weight/volume) crude collagenase. After 20 minutes, the fragments were centrifuged at 300 rpm for 2 minutes. The supernatant containing isolated cells was stored in DMEM; the pellet was then subjected to four more rounds of digestion. The extracted cells were then filtered through a nylon cell strainer (VWR-Polylabo, Strasbourg, France). The cell preparation was then resuspended in 2 mL (first 2 animals of group 2) or 1.2 mL (remaining 5 animals of group 2) of serum-free DMEM and kept at 4°C until implantation. A previous study allowed to determine that 1.7 ± 0.3 x 107 mononucleated cells were injected per animal.16
Cell culture controls
To confirm the presence of muscle precursor cells in the cell preparation, a 100-µL aliquot of the cell suspension was plated in a flask containing fetal calf serum completed DMEM. The cells were then grown as a control in humid air with 5% carbon dioxide. One week after plating, cells were processed for immunodetection of MyoD as previously described by Montarras and associates.17
Cell implantation into right ventricular myocardium
After PAB, landmarks were made with 5-0 polypropylene suture material on the right ventricular free wall. Twenty (first 2 animals of group 2) and then 10 epicardial 100-µL injections of the cell preparation (5 remaining animals of group 2) or medium (4 control animals of group 3) were carried out with a 27-gauge infusion set (Vygon Corporation, East Rutherford, NJ). Five injections were performed in the infundibulum about 10 mm apart; five additional injections were performed in the rest of the right ventricle. A 0.1-mg/kg intramuscular injection of dexamethasone was performed before closing to reduce the inflammation associated with the numerous intramyocardial punctures (pilot study, data not shown).
Histologic examination
All sheep were killed 60 ± 2 days after PAB with or without cellular or placebo injection in the right ventricle. After hemodynamic assessment with pressure-volume loops, heparin (10,000 IU) and 60 mg/kg sodium pentobarbital were injected intravenously. The heart was exposed, and the pulmonary trunk and the atria were removed from the heart, which was then weighed. The right ventricle was cut and weighed before being fixed with 10% formalin for evaluation.
Formalin-fixed, paraffin-embedded blocks were processed. Serial 5-µm sections from the harvested area were prepared for hematoxylin-eosin staining and immunostaining. Immunohistochemical examination was performed with an automated immunoperoxidase technique (Ventana Medical Systems, Inc, Tucson, Ariz) with primary monoclonal antibodies directed against a fast skeletal-specific isoform of myosin heavy chain (MY32; Sigma, St Louis, Mo) and connexin 43 (Sigma), a component of gap junctions.
Statistical analysis
Indices of cardiac function (developed pressure, dP/dtmax, dP/dtmin, PRSW slope) and weight from baseline to last day of protocol and parameters evaluated at death only (heart, right ventricular, and left ventricular weights, right ventricular/heart weight ratio, and right ventricular diameter) were recorded and analyzed with the SAS software package (8.2; SAS Institute, Inc, Cary, NC) by either Wilcoxon signed rank test for paired and unpaired data, nonparametric analysis of covariance, or Kruskal-Wallis test. The data are presented as mean ± SD.
| Results |
|---|
|
|
|---|
Aliquots of each grafted cell preparation were analyzed to confirm cell viability and myogenic differentiation. Control culture flasks were inspected daily. Immunodetection of the regulatory factor MyoD showed that at least 50% of the cultured cells were skeletal muscle precursor cells (Figure 1). When the cells were left to fuse, all flasks were covered with numerous myotubes after 1 week of culture.
|
|
|
|
|
| Discussion |
|---|
|
|
|---|
The report presented here draws the attention to the interest in noncultured muscle cells for right ventricular CCM. Our group in a previous report16 showed that noncultured muscle cells can be a cell source for CCM. Advantages to using noncultured cells as opposed to cultured satellite cells are the lower cost, the reduced time lag between harvest and grafting, and the diminished risk of contamination. This technique requires a biopsy in the morning and a grafting procedure 3 hours later.
Numerous right ventricular sections had areas covered with skeletal myosin heavy chain (MY32)positive fibers at 2 months after implantation. Considering the small number of cells injected (a maximum of 20 x 106 cells, 50% identified as skeletal muscle precursor cells in control cultures) and the engraftment observed on many sections of all the hearts, active cellular divisions must have happened. Therefore the fact that we did not observe a functional benefit in the cell transplantation group is not related to a complete failure of the engraftment.
Reliable assessment of global right ventricular function was achieved with the analysis of pressure-volume loops, thanks to the conductance catheter technique, which is a well-validated tool in animals and human subjects for the assessment of left ventricular function. Right ventricular function has also been explored with conductance catheter-acquired pressure-volume loops in rabbits, dogs, pigs, and lambs.12,1825
Our results show that noncultured cells can massively engraft into ovine right ventricular myocardium. Successful engraftment, however, did not translate into a functional benefit that we could demonstrate. Several hypotheses could be proposed.
CCM has been proposed as an alternative strategy for augmenting the function of diseased myocardium. However, the myocardial environment we investigated was not diseased but subject to chronic pressure overload. It is likely that the cell implantation contribution could only be a moderate adjunct to the remarkable hypertrophy induced by PAB.
Targeted regional administration implies either transmyocardial or intracoronary delivery. The very small muscle fragments still present in our cell preparation precluded intracoronary administration. Indeed, a pilot study we carried out on a dilated cardiomyopathy model yielded acute myocardial ischemia and death in all animals exposed to intracoronary noncultured cells (unpublished data). Epicardial injections were therefore carried out.
Myocardial injury induced by the numerous punctures remains a serious concern. The first 2 animals of the cell therapy group died of right heart failure from myocardial edema of the right ventricle. Considering the thin wall of the right ventricle, it is likely that both the punctures and the volume occupied by the cell preparation were detrimental. Diminishing the number of injections resolved the problem but also limited the extent of treated myocardium. It is possible that this limitation in technique of administration accounts for lack of hemodynamic benefit; perhaps more injections would be required. Considering our experience, only sequential injections (at varying time points) could be performed. Minimally invasive and interventional approaches would therefore have to be used.
Perhaps another cell type should be investigated. For the left ventricle, mesenchymal or other stem cells are interesting options. If satellite cells have no demonstrable effects, however, it is doubtful that stem cells would have a much more visible effect.
One limitation of this study is the small sample size and thus lack of statistical power. Had we benefited from a larger series of animals in all groups, perhaps we would have observed a small difference between the groups. However, it is likely that regardless of the cell type and the route of delivery, cellular therapy is probably not useful to strengthen a retrained left ventricle through PAB before ASO. Whether cell transplantation can really affect the outcome of right ventricular failure long term after ASO is another unsettled question. A development of our research would be studying the long-term effect (6 months to 1 year) of cell therapy as an adjunct to PAB. Indeed, CCM might result in a more prolonged period of preserved right ventricular function. Another option would be studying the effect of cell therapy performed only after right ventricular failure has developed. This clinical context would be somewhat similar to that of patients with former ASOs who could not benefit from double-switch preparation and transformation and thus might receive functional benefit. Although no data are available regarding the precise number of potential patients, this indication would be of great interest. Still preliminary, this investigation paves the way for further research on CCM, right ventricular failure, and congenital heart disease.
In conclusion, our data indicate that noncultured muscle cell transplantation in the retrained right ventricle is feasible. However, successful engraftment did not translate into a hemodynamic benefit.
| Acknowledgments |
|---|
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. H. M. Hessel, P. Steendijk, B. den Adel, C. I. Schutte, and A. van der Laarse Characterization of right ventricular function after monocrotaline-induced pulmonary hypertension in the intact rat Am J Physiol Heart Circ Physiol, November 1, 2006; 291(5): H2424 - H2430. [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 |