| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL ARTICLE |
Centro de Pesquisas Labcor Laboratórios Belo Horizonte, Minas Gerais, Brazil
Gláucio Furlanetto, MD Tel: +55 11 3284 7686 Fax: +55 11 3284 7686 Email: gfurlanetto{at}terra.com.br, Rua Maestro Cardim 560, Sala 73, Liberdade, CEP 01323-000 São Paulo, SP, Brazil.
ABSTRACT
As there is currently no suitable valved pulmonary conduit for small children, porcine conduits treated by the L-Hydro process were implanted into 9 newborn lambs to investigate growth potential. Of the 8 survivors, 7 were kept alive for 12 months after implantation. The diameter of the conduit and gradient across the valve were evaluated at surgery and at 3 and 9 months postoperatively using bidirectional echocardiographic and angiographic methods. After sacrifice, histological and radiological analyses were performed. The mean weight of the animals was 4.2 ± 1.1 kg at implantation and 43.1 ± 6.2 kg at sacrifice. There was a significant increase in mean valve area from 139.9 ± 18.0 mm2 at implantation to 443.5 ± 89.2 mm2 at 12 months. Pre-sacrifice angiography showed no transvalvular gradient, and radiographic analysis did not reveal significant conduit wall or leaflet calcification in any of the animals. Histological examination of the grafts demonstrated total integration, with native-like intact valve leaflets. Thus functional evaluation, echocardiography, and histology demonstrated growth of the grafts with completely endothelialized and apparently normal pulmonary valve leaflets without calcification.
Key Words: Bioprosthesis • Heart Defects • Congenital • Pulmonary Valve • Tissue Engineering
INTRODUCTION
There are several congenital cardiopathies with impaired connection between the right ventricle and the pulmonary circulation, in which surgical correction using a valved conduit is needed. A valved conduit can be constructed using various materials fixed by glutaraldehyde, such as a porcine valve, Dacron or bovine pericardial tube with bovine pericardial valves, and most recently, bovine jugular vein with native valves.1,2 Pulmonary and aortic homograft roots are used for this application.3 Although surgical techniques have been developed for correction of these congenital defects, the latest results are inadequate, especially in small children, due to stenosis, thrombosis, proliferation of the intima, valvular calcification and, most importantly, inability to grow as the child matures. The objective of this experimental work was to evaluate the potential for biointegration and growth of L-Hydro-treated porcine pulmonary root xenografts implanted in the orthotopic position in newborn lambs.
MATERIALS AND METHODS
The pulmonary trunks with the pulmonary valves were removed from newborn pigs and treated by the L-Hydro process that consists of 3 steps: extraction and masking of porcine antigens using chemical oxidation and poly-ethylene glycol; incorporation of nonsteroidal antiinflammatory and antithrombotic agents; and sterilization of the graft in aqueous hydrogen peroxide. The sterilized prostheses were stored in 50% ethanol. The treated porcine roots contained no living cells.4
Nine male newborn Santa Ines lambs aged <2 weeks, weighing 2.8–5.5 kg, were selected for the study, following guidelines set forth by the American Association for Accreditation of Laboratory Animal Care. Anesthesia was induced with 10 mg · kg–1 sodium pentobarbital, followed by administration of cephalothin and methylprednisolone; halothane (kept at 1.5%) was used as a continuous anesthetic. Vital signs and rectal temperature were monitored. A longitudinal incision parallel to the sternocleidomastoid muscle was made with the animal resting on its right side. After administration of 350 U · kg–1 heparin, a cannula was introduced into the left carotid artery. A left lateral thoracotomy was performed in the 4th intercostal space, and another cannula was introduced into the right atrium. Normothermic extracorporeal circulation was initiated with a flow rate of 100 mL · kg–1 · min–1 using a pediatric membrane oxygenator. The trunk of the pulmonary artery was divided in the transverse plane, and the native pulmonary valve was removed. The L-Hydro-treated porcine xenograft was placed in the orthotopic position with proximal and distal continuous suturing with 7/0 polypropylene. The diameter of both grafts in 2 pilot-study animals (nos. 1 and 2) was 23 mm. In the other 7 animals, the L-Hydro-treated grafts were harvested from newborn piglets; 6 had a diameter of 13 mm and 1 had a diameter of 15 mm. After implantation of the graft, extracorporeal circulation was discontinued, and the cannulas were removed. Protamine sulfate was administered intravenously. The mean duration of extracorporeal circulation was 32.6 ± 7.0 min (in 8 animals). A chest tube was placed in the 6th intercostal space, and the chest was closed. After the animals were extubated, the chest tube was removed. Cefalotin and gentamicin were administered as prophylactic antibiotics for 7 days.
Of the 9 animals operated on, 8 survived: 6 were kept alive for 379 days, 1 for 354 days, and 1 for 684 days. The weight of each animal was monitored monthly until sacrifice. They did not receive any antiplatelet or anticoagulant therapy postoperatively. Transthoracic Doppler echocardiography was carried out to evaluate leaflet mobility and transvalvular gradients after 3, 9, and 12 months, using an ATL Ultramark 6 (Philips, Netherlands). Prior to sacrifice, sodium pentobarbital 12.5 mg · kg–1 was administered, followed by intubation and mechanical ventilation. Continuous anesthesia was maintained with halothane. The left jugular vein was dissected, and a 7F Swan-Ganz catheter (Baxter Healthcare, Irvine, CA, USA) was introduced into the pulmonary artery. A multiparametric monitor (Bese, Belo Horizonte, Brazil) was used to measure pressure in the right ventricle and pulmonary trunk. To evaluate valve competence, contrast medium with 75% diatrizoate of meglumine (Hypaque M 75%; Sanofi-Syntelabo, Brazil) was injected into the pulmonary trunk, and images were recorded using a Philips XG 4002/00 fluoroscope (Philips, The Netherlands). At the end of the study, the animals received 350 U · kg–1 of heparin prior to being sacrificed by intravenous injection of 1 g of sodium pentobarbital and 26 mEq potassium chloride. A left lateral thoracotomy was performed in the 4th intercostal space. The heart and lungs were removed and examined macroscopically. After the implants were dissected and recovered, they were opened longitudinally for evaluation of function, integrity of the leaflets and graft wall tissues, leaflet mobility, and the presence of thrombus, vegetation, calcification, or fibrosis. The implants were photographed and fixed in Histochoice (Amresco, Inc., Solon, OH, USA) for histological and radiological examination. Radiographic analysis of the xenograft valves was performed using a Senographe DMR mammograph (GE, France) at 28 kV and 300 mA to determine the distribution and intensity of calcium deposits.
Histological studies by light microscopy of hematoxylin-eosin stained specimens from animals nos. 2–8 were carried out to verify re-endothelization of the valve leaflets and re-cellularization of the collagenous matrix by host cells. For immunochemical histological studies, tissue sections (5 µm) were de-paraffinized and stained for endothelial nitric oxide synthase (eNOS), inducible nitric oxide synthase (iNOS), von Willebrand factor (vWF), and 
-smooth muscle actin (ASMA). Antibodies to vWF (Rbt polyclonal) and ASMA (Mo monoclonal) were purchased from Sigma (St. Louis, MO, USA), anti-iNOS (Rbt polyclonal) from Upstate (Lake Placid, NY, USA), and anti-eNOS (Mo monoclonal) from BD Biosciences (San Jose, CA, USA). Bound antibody to iNOS or eNOS was labeled with a biotin-streptavidin amplification kit (LSAB2, Dako, Carpinteria, CA, USA), while Protein A-HRP (Sigma, St. Louis, MO, USA) was used to label bound antibody to vWF and ASMA actin. After incubation with the appropriate label, the sections were rinsed with phosphate-buffered saline and developed with 3,3'-diaminobenzidine (DAB; Sigma, St. Louis, MO, USA). Slides were rinsed with water to stop development, counterstained with hematoxylin-eosin, dehydrated/cleared, and cover-slips were applied. Sections of native sheep pulmonary roots were stained together with study implant samples, as controls. Control samples showed endothelial cells that stained positive for eNOS and vWF but negative for iNOS. Cells inside the native root wall also stained positive for ASMA.
Comparisons of the values were made using the nonparametric test of Friedman. The level of significance used in the tests was 5%.
RESULTS
The mean weights of the animals throughout the study are given Table 1
. The increases in weight between time 0 and 9 months and between 3 and 12 months were significant (p < 0.05). Macroscopic evaluation of the xenografts revealed that the integrity of the tissue matrices of the valve leaflets was preserved: there was no visible thrombosis, vegetation, calcification, or fibrosis of the leaflets (Figure 1). The surfaces of the wall and leaflets appeared smooth in all grafts. The coaptation of the 3 leaflets in each implant was properly aligned at the central axes of the valve (Figure 2). Macroscopic evaluation of the graft in one animal showed underdevelopment of 1 of the 3 pulmonary valve leaflets, and there were small nodules of calcification in the wall of the graft in another animal, which did not compromise valvular function. Graft diameter was determined at the level of the valve in sheep nos. 3–8 by serial Doppler echocardiography (Tables 1 and 2); there was a significant increase during the experimental period. The increase from 3 to 12 months was also significant. Calculation of valve area by Doppler echocardiography showed significant increases from time 0 to 9 and 12 months, and between 3 and 12 months (Table 1). The maximum transvalvular gradients measured postoperatively by Doppler echocardiography in animals nos. 2–8 were not significantly different (Table 1). The maximum gradient in one animal sacrificed at 24 months was 0.9 mm Hg. Direct measurement of pressure readings in these animals (2–8) before sacrifice showed a mean systolic gradient of 9.1 ± 3.6 mm Hg between the right ventricle and the pulmonary trunk. The differences in pressure between the right ventricle and the pulmonary artery were not significantly different. Angiographic analysis of graft valve insufficiency, obtained by injection of contrast medium into the pulmonary trunk, revealed moderate to severe insufficiency in animal no. 7, and mild insufficiency in animal no. 6 (angiographic data not shown). No evidence of valvular insufficiency was observed in the other animals (Figure 3). Radiographic analysis of the xenograft valves did not reveal calcification in any of the leaflets. In one animal, there were 3 points of calcification in the wall of the graft, and there were microcalcification loci in the graft-host suture lines in the others (Figure 4).
|
|
|
|
|
|
). In animal no. 7, one valve cusp appeared to be underdeveloped with a large number of plasmocytes consistent with a degenerative process due to a chronic inflammatory reaction (data not shown). There was no evidence of significant numbers of inflammatory cells in the other explanted leaflets or the valve root walls. Immunochemical histological analysis was performed using antibodies to eNOS, iNOS, and vWF to identify endothelial cells. Cells positive for eNOS but negative for iNOS were observed on the tips of valve leaflets from implants recovered at 12 months (Figure 6). These cells stained positive for vWF. Cells that stained positive for ASMA were also observed aligned in a native-like manner within the interstitial and medial layers of the root wall (Figure 6).
|
|
Commercial tissue valves involve glutaraldehyde preservation. Due to the cytotoxic nature of the preservation chemicals in these bioprostheses, unsatisfactory long-term results with inflammation, thrombosis, and calcification have been reported.5 Correction of congenital cardiopathies, such as truncus arteriosus and Fallot’s tetralogy, using monocuspid bovine pericardial devices fixed with glutaraldehyde demonstrated satisfactory early performance but long-term failure of monocuspid valve function.6 One alternative is a Dacron or bovine pericardial tube with bovine pericardial valves fixed by glutaraldehyde, but late stenotic dysfunction has been reported.7 Conduits with glutaraldehyde-treated porcine valves have been widely used, and the reoperation rate due to valve dysfunction is variable.8,9 Cryopreserved pulmonary and aortic homografts have also been used for right ventricular outflow tract reconstruction. These grafts offer good results in the young adult or older patient, but are not satisfactory in children. Possible reasons for poor results include an immunologic response.10 Recently, glutaraldehyde-preserved bovine jugular veins with native trileaflet venous valves have shown good hemodynamics and freedom from calcification in midterm follow-up;11 however, the long-term outcome is expected to be similar to that of other glutaraldehyde-treated devices. The long-term results of valved conduits fabricated intraoperatively using fresh autologous pericardial tissue were excellent;12 but difficulty in fabrication has limited their use. Calcification is the most common cause of degeneration of tissues preserved by glutaraldehyde.13 Several anti-mineralization strategies have been used to mitigate this problem, including detergent substances and other non-glutaraldehyde processes such as epoxy, photo-oxidation, and carbodiamide.
Alternative non-aldehyde methods of tissue preservation have been attempted to encourage implants to re-cellularize with host endothelial cells and transform into a living structure with potential regeneration and growth. These methods include decellularization by enzymatic digestion, washing with detergents, and photo-oxidation. Tissue-engineering techniques for seeding implants made from different materials with cells harvested from donors or autologous cells have also been developed.14 Hoerstrup and colleagues15 reported 30% growth of tissue-engineered vascular grafts seeded with autologous cells and cultured in vitro for 21 days prior to implantation into a growing sheep model, with 100 weeks of follow-up. Much work has been undertaken to develop decellularization technology to minimize degeneration in the initial post-implant period;16 however, long-term durability of these implants has yet to be demonstrated. Bioprostheses obtained by tissue engineering using a biodegradable scaffold have provided encouraging short-term results.17,18 Preservation of the structure and natural function of the tissue is thought to be important for in-vivo endothelization. Polyethylene glycol, a well-known agent with low toxicity and beneficial immunosuppressive characteristics due to its interaction with lipids and antigens on cell membranes, has been used by others.19 The lack of cytotoxicity in L-Hydro-treated heart valves provides a cytocompatible tissue matrix that allows spontaneous re-endothelization with host cells.20
All animals in this study grew asymptomatically without cardiac murmurs or clinical signs of valve insufficiency. Doppler echocardiography revealed negligible transvalvular gradients 1 year post-implantation in 7 sheep, and after 2 years in one. There were normal diameters at the level of the pulmonary valve annulus, and macroscopic evidence of tissue growth with development of the wall and valve leaflets without fibrosis, thrombosis, or calcification. Although there was under-development of one leaflet in animal no. 7, the valves in the other lambs showed normal development of the pulmonary trunk of the xenograft, with proper leaflet coaptation. These observations are evidence of growth rather than dilatation of the xenograft. Radiography did not reveal calcification, and histology confirmed total integration of the xenograft by the host, with endothelial cells on the tissue surface and cellularity in the collagen matrix. Immunohistochemical studies showed that these endothelial cells expressed eNOS but not iNOS; it has been suggested that the expression of iNOS is a response of endothelial cells to oxidative stress or endothelial injury. The absence of a random orientation of ASMA expression cells in the tissue matrix indicates that the cells in the implanted tissue matrix were not stimulated to grow or synthesize extracellular matrix in an uncontrolled manner. Thus the non-vital L-Hydro-treated porcine xenografts were not only covered by the host endothelial cells, but they served as templates for significant growth and maintenance of valvular function without stenosis.
We theorize several reasons why the L-Hydro-treated implants integrated host cells into living tissue in a controlled manner: tissues in the treated implants contain intact extracellular components and thus are not subjected to degradation by nonspecific enzymes in a random fashion; donor antigens were either extracted or masked by polyethylene glycol polymers; there was no living donor cell continuously expressing antigens; the noninflammatory tissue matrix had antiinflammatory agents added to prevent degradative processes that can destroy the implants before the rebuilding process; and the compliance of the implants was similar to that of the replaced tissues, therefore, a chronic reaction from the invading cells to aggressively modify or synthesize new matrices was not stimulated. The ability of the L-Hydro-treated implants to be revitalized and remodeled, as observed in this study, prompted us to further speculate that the implants may eventually become indistinguishable from the host tissue, and thus should have the same properties as the original tissue, including long-term durability. We further theorize that these treated tissues facilitate a slow rebuilding process because of inhibition of the host inflammatory reaction by the L-Hydro process, preventing breakdown of the implanted material. Inhibiting host inflammatory responses can also reduce the stimulation of uncontrolled growth that leads to hypertrophy. Further studies are warranted to elucidate the mechanism by which the biointegrated tissues are programmed to grow. To confirm behavior of this type of graft in humans, a clinical study with late follow-up in children with congenital cardiopathies associated with malformation of the ventricular outflow tract should be considered.
REFERENCES
Asian Cardiovasc Thorac Ann 2009;
17:350-356
© 2009 by SAGE Publications
DOI: 10.1177/0218492309338096
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
