Ann Thorac Surg 2002;73:1528-1533
© 2002 The Society of Thoracic Surgeons
Original article: cardiovascular
Use of autologous auricular chondrocytes for lining artificial surfaces: a feasibility study
Timothy Scott-Burden, PhDb,*,
Jennifer P. Bosley, BSb,
Doreen Rosenstrauch, RN, MDa,b,
Kimberly D. Henderson, BSb,
Fred J. Clubb, Jr, DVM, PhDc,
Harald C. Eichstaedt, MDa,
Kazuhiro Eya, MDa,
Igor Gregoric, MDa,
Timothy J. Myers, BSa,
Branislav Radovancevic, MDa,
O.H. Frazier, MD*a
a Cardiovascular Surgical Research Laboratories, Texas Heart Institute at St. Luke’s Episcopal Hospital, and The University of Texas Health Science Center at Houston, Houston, Texas, USA
b Vascular Cell Biology Laboratory, Texas Heart Institute at St. Luke’s Episcopal Hospital, and The University of Texas Health Science Center at Houston, Houston, Texas, USA
c Department of Cardiovascular Pathology, Texas Heart Institute at St. Luke’s Episcopal Hospital, and The University of Texas Health Science Center at Houston, Houston, Texas, USA
Accepted for publication December 3, 2001.
* Address reprint requests to Dr Frazier, 1101 Bates Ave, Suite P357, Houston, TX 77030 USA
e-mail: Doreen.Rosenstrauch{at}uth.tmc.edu
 |
Abstract
|
|---|
Background. Auricular elastic cartilage is a potential source of autologous cells for lining the luminal surfaces of cardiovascular prostheses. We tested this potential in vitro and in vivo using a left ventricular assist device (LVAD) and a calf model.
Methods. In vitro, auricular cartilage was harvested from the anesthetized ear of a calf, isolated, and cultured on tissue culture dishes. Primary chondrocytes were typed by immunocytochemistry, transferred into culture media, passaged twice, and seeded onto the blood-contacting luminal surfaces of four LVADs (HeartMate; Thoratec Corporation, Woburn, MA). Seeded cell linings were preconditioned under simulated flow conditions to promote cell adhesion to luminal surfaces. Seeding efficiency and cumulative cell loss under flow conditions were quantitated. In vivo, one of the four autologous chondrocyte-lined and preconditioned LVADs was implanted into the tissue-donor calf; run for 7 days; explanted; and evaluated grossly, by scanning electron microscopy, and by transmission electron microscopy.
Results. The efficiency of seeding chondrocytes onto the luminal surfaces of the four LVADs was 95.11% ± 4.23% (n = 4). Cumulative cell loss during preconditioning under flow conditions in vitro did not exceed 12% (n = 4). After 7 days of in vivo implantation, the luminal surfaces of the implanted LVAD demonstrated an intact, strongly adherent cellular lining.
Conclusions. Auricular elastic cartilage is a ready and easily accessible source of chondrocytes whose ability to produce collagen II and other important extracellular matrix constituents allows them to adhere strongly to the luminal surfaces of LVADs. The simple method of isolating and expanding auricular chondrocytes presented here could be used to provide strongly adherent autologous cell linings for LVADs and other cardiovascular devices. If and when chondrocytes can be genetically engineered to produce antithrombogenic factors and then used to line the luminal surfaces of LVADs or other cardiovascular prostheses, they may be able to improve the hemocompatibility of the blood–biomaterial interface in such devices. Our successful feasibility study in a calf model warrants further studies of this concept in vivo.
|
Introduction
|
|---|
The use of endothelial cells as autologous cell lining has been shown to improve the biocompatibility of cardiovascular prostheses [1]. Nevertheless, the ability of endothelial cells to adhere to the artificial surfaces in such devices is poor. When exposed to physiologic flow conditions, they slough off easily [2]. One successful alternative has been to use autologous smooth muscle cells that adhere better to biomaterials and that have been genetically engineered to produce nitric oxide so as to improve the hemocompatibility of the blood–biomaterial interface [3].
However, the process of harvesting, isolating, and cultivating both endothelial cells and smooth muscle cells from autologous vessels is invasive and time consuming. In contrast, auricular chondrocytes are abundant, readily accessible, and easily and efficiently harvested. One potential source of chondrocytes is auricular cartilage, which can be harvested by a minimally invasive technique that preserves cell viability, decreases surgical time, and minimizes postoperative complications [4]. In vivo, chondrocytes are naturally nourished by diffusion and produce substantial amounts of extracellular matrix components [5]. Therefore, cultured chondrocytes may offer a more efficient and less invasive means of covering artificial surfaces with a viable and adherent cell layer. Furthermore, if the chondrocytes are genetically engineered to act like endothelial cells, then they might also improve the hemocompatibility of blood–biomaterial interfaces.
As the first step in proving the feasibility of this concept, we harvested, isolated, and cultured auricular cartilage from ears of calves and studied its adherence to the luminal surfaces of LVADs in vitro and in vivo in a calf model.
|
Material and methods
|
|---|
In vitro experiments
Tissue harvesting
Six weeks before the in vivo experiments began, a 2-mm-diameter tissue sample was harvested from the ear of a 4-month-old longhorn-crossbreed calf weighing 70 kg (4B Livestock, Midway, TX) by punch biopsy with a trephine (Nasco, Fort Atkinson, WI). The biopsy was done under sterile conditions and local anesthesia. The tissue sample was immediately placed in sterile cell culture medium containing antibiotics (Table 1)
and transferred to a sterile hood. The sample was kept at 4°C for a minimal amount of time to keep the tissue viable.
Tissue isolation
Under the sterile hood, auricular elastic cartilage was removed from the surrounding dermal tissue of the biopsy sample by microdissection. The isolated cartilage was placed in 10- x 10-mm tissue culture dishes with 2% antimycotic-antibiotic phosphate-buffered saline (Table 2)
and incubated at 37°C for 10 minutes. Multiple small pieces of the cartilage were placed in six-well tissue culture plates in 500 µL of modified RPMI cell culture medium and incubated at 37°C.
Cell culture
Chondrocytes growing from the auricular cartilage were cultured under sterile conditions and maintained in a humidified 5% CO2 incubator at 37°C. Every 12 hours, drops of medium were added to each culture well. After 5 to 7 days, when chondrocytes became adherent to the culture dishes, cartilage pieces were removed. The chondrocytes were then passaged as follows. First, they were trypsinized in a solution of 0.5 g of trypsin and 0.2 g of EDTA (Sigma, St. Louis, MO). Then, when the cells became detached, fetal bovine serum was added to inactivate the trypsin. The resulting single cell suspension was then transferred to tissue culture plates to establish a monolayer of chondrocytes. Thereafter, the cell culture medium was changed every 48 hours. Chondrocytes were passaged twice, and cells of the second passage were used for subsequent experiments.
Histology
Fine (1-µm-thick), paraffin-embedded sections of pure elastic cartilage tissue were stained with Verhoff-van Gieson stain to visualize elastic fibers, Masson’s trichrome stain to visualize collagen fibers, and hematoxylin and eosin to assess tissue and cell morphology.
Immunocytochemistry
A portion of the second-passage chondrocytes from each culture plate was plated on slides and fixed in 1% formalin for immunocytochemistry using a double-antibody labeling technique. Paraffin-embedded sections of pure elastic cartilage tissue immunostained in the same way served as a control. The primary antibody used was collagen type II (NCL-COLL-IIp; Novocastra, Newcastle, United Kingdom); the secondary antibody used was a biotinylated immunoglobulin specific for the primary antibody (Vectastain Elite IgG; Vector Laboratories, Burlingame, CA). In brief, deparaffinized and hydrated cartilage tissue sections and cells were incubated in 3% hydrogen peroxide for 10 minutes to quench endogenous peroxidase activity. All specimens were then incubated with normal mouse serum for 20 minutes to block nonspecific binding sites. Next, all specimens were exposed to a 1:50 diluted solution of the primary antibody for 1 hour. After repeated washes, all specimens were incubated with the secondary antibody for 30 minutes. Next, all specimens were exposed to avidin DH and biotinylated enzyme (Vectastain Elite ABC reagent) for 30 minutes and then to diaminobenzidine as a peroxidase substrate for 30 seconds. Finally, all specimens were counterstained with hematoxylin for 1 minute. Control incubations were performed in the absence of the primary antibody.
Cell seeding
Four implantable pneumatic LVADs (HeartMate; Thoratec Corporation, Woburn, MA) were used for the in vitro cell seeding experiments. The HeartMate has two luminal artificial surfaces: a flexible diaphragm made of a "biomer" (ie, polyurethane flocked with polyester microfibrils) and a metal housing made of sintered titanium microspheres [6]. One of the four LVADs used in the in vitro experiments was later used in the in vivo experiments, described later.
Seven days before the in vivo experiments began, each LVAD was seeded with a total of 3 x 107 autologous cells under sterile conditions. The inlet and outlet conduits of each LVAD were closed, and the LVAD was placed in a humidified 5% CO2 incubator at 37°C. Each LVAD was tilted 15 degrees every 2 hours for 24 hours during cell seeding to ensure complete cell coverage of surfaces. After seeding was completed, three samples of cell culture medium were collected from each LVAD and assessed for the number of nonadherent cells using a Coulter counter (Coulter, Hialeah, FL). Each LVAD was incubated for 4 days in the same incubator. Cell culture medium supplemented with 50 µg/mL of sodium ascorbate (Sigma) was changed every 48 hours to promote extracellular matrix synthesis to maximize the adherence of cells to both artificial surfaces. After each medium change, samples were collected to assess the number of nonadherent cells as before and to calculate seeding efficiency based on the number of initially seeded cells.
Cell preconditioning
Because preconditioning promotes good cell adherence once an LVAD is implanted and perfusion initiated, each seeded LVAD was subjected to an in vitro preconditioning regimen that exposed it to flow conditions [7]. In brief, 12 hours before the in vivo experiments began, each LVAD was incorporated under sterile conditions into an in vitro flow loop (Fig 1).
The in vitro flow loop was connected to a pneumatic drive console operated at 70 beats/min and an ejection fraction of 30%. During the 12-hour preconditioning period, 18 samples of cell culture medium were collected (three each at 0, 0.3, 3, 5, 7, and 9 hours) to assess the amount of cell loss and to calculate preconditioning efficiency based on the number of initially seeded cells.
View larger version (120K):
[in this window]
[in a new window]
|
Fig 1. In vitro flow loop used for preconditioning of seeded cells. The loop consisted of a reservoir filled with cell culture medium (1) and the cell-seeded left ventricular assist device (2). Samples of cell culture medium emptied into the reservoir through a stop cock (3) were collected from the reservoir at several time points during the 12-hour preconditioning period as described in Material and Methods. The in vitro flow loop was connected to a pneumatic drive console (not shown).
|
|
In vivo experiments
Animal care
The tissue-donor calf used in the in vivo experiments received humane care in compliance with the Principles of Laboratory Animal Care prepared by the National Society for Medical Research and the Guide for the Care and Use of Laboratory Animals prepared by the Institute of Laboratory Animal Resources and published by the National Academy Press (1996).
LVAD preparation
Immediately before implantation, one of the four seeded LVADs used in the in vitro experiments was disconnected from its preconditioning in vitro flow loop and discharged of cell culture medium. To eliminate any remaining cell culture medium, the LVAD was rinsed twice with pure RPMI medium and washed three times with 37°C phosphate-buffered saline. Once the LVAD was filled with phosphate-buffered saline, its inflow and outflow conduits were capped and its external surface sterilized by rinsing twice with 70% ethanol. The LVAD was then transported under sterile conditions to the operating room for immediate implantation.
LVAD implantation
Once prepared, the seeded LVAD was implanted into the tissue-donor calf under cardiopulmonary bypass using a standard procedure described previously [8]. In brief, an abdominal incision was made through which the LVAD was placed. The LVAD’s percutaneous driveline was tunneled subcutaneously and exteriorized high on the left flank. The inflow and outflow conduits were passed through separate 1- to 2-cm-long incisions in the anterior left hemidiaphragm. A 20-mm-diameter low-porosity Dacron outlet graft was preclotted using autologous serum and blood and then anastomosed end-to-side at the descending thoracic aorta. The sewing ring of the LVAD was sutured to the left ventricular apex, using interrupted 2-0 braided polyester sutures with Teflon felt pledgets. A small crux incision was made in the apex, and a coring knife was inserted into the left ventricular cavity. A full-thickness circular segment of the apical myocardium was then excised. The pump inlet tube was inserted into the left ventricle. Air was removed from the LVAD through a needle inserted into the Dacron graft. Protamine sulfate was administered intravenously to antagonize heparin. The LVAD was kept in automatic mode at all times while implanted.
In the postoperative period, dextrose 5% in Ringer’s lactate solution was infused intravenously as necessary to maintain central venous pressure. Potassium chloride was added to the intravenous infusion as indicated by serial measurements of serum potassium. Sodium cefonicid (1 g) was administered daily for 4 days to prevent infection. Butorphanol (10 mg) was given intramuscularly every 4 to 8 hours for pain. No anticoagulative therapy was administered.
Postmortem examination and gross observations
Seven days after LVAD implantation, the calf was euthanized (by intravenous administration of 3 mg/kg heparin and then 1 mL/kg beuthanasia-D) and necropsied. The LVAD and pertinent organs (eg, heart, lungs, liver, rumen, spleen, kidney, adrenal glands, and brain) were examined grossly and photographed. The organs were evaluated for the presence of emboli, ischemia, and infarction.
Scanning electron microscopy and transmission electron microscopy
The LVAD was disassembled, inspected, photographed, and examined by scanning electron microscopy and transmission electron microscopy. In brief, both artificial surfaces of the LVAD, along with the attached cellular lining, were immediately fixed in Millonig’s phosphate buffer supplemented with 3% glutaraldehyde and incubated for several days at 4°C. Scanning electron microscopy samples were further fixed in 1% osmium tetroxide for 1 hour at room temperature. The samples were then rinsed with Millonig’s phosphate buffer and gradually dehydrated with ethanol [7].
For scanning electron microscopy, samples of the diaphragm and the titanium housing of the LVAD were coated with gold using a Denton Vacuum 502A Cold Sputter Module. The samples were then placed in a digital scanning electron microscope (model 960; Zeiss, Jena, Germany) to visualize the cell-lined sintered titanium and textured polyurethane surfaces.
For transmission electron microscopy, samples of the LVAD’s diaphragm and housing were infiltrated with resin and ethanol, embedded in the resin overnight, cut with a diamond knife to a thickness of 60 to 80 nm, pulse-stained in uranyl acetate and lead citrate, and finally viewed under a transmission electron microscope (model 1200EX; JEOL, Peabody, MA).
|
Results
|
|---|
In vitro experiments
Histology
Isolated auricular cartilage tissue stained positive for elastic and collagen fibers, indicating the presence of pure elastic cartilage. Tissue culture cells derived from isolated auricular cartilage tissue at zero passage stained positive for elastic and collagen fibers, indicating the presence of chondrocytes (Fig 2).
View larger version (142K):
[in this window]
[in a new window]
|
Fig 2. Chondrocytes (Cc) growing from a piece of elastic cartilage (Ec) at zero passage. Tissue culture cells derived from isolated auricular cartilage tissue at zero passage stained positive for elastic and collagen fibers, indicating the presence of chondrocytes. See Material and Methods for stains used.
|
|
Immunocytochemistry
Elastic cartilage tissue and cells at second passage stained positive for collagen II, indicating the presence of chondrocytes (Fig 3).
Cells treated with cell culture medium supplemented with sodium ascorbate showed much stronger staining of collagen II, indicating improved collagen synthesis.
View larger version (97K):
[in this window]
[in a new window]
|
Fig 3. Immunocytochemistry of elastic cartilage (left) and passaged chondrocytes (right). Positive staining for collagen II indicated the presence of chondrocytes. See Material and Methods for stains used.
|
|
Seeding efficiency
The luminal surfaces of the LVADs, consisting of sintered titanium and polyurethane, were seeded with autologous auricular chondrocytes at a first seeding efficiency of 92.66% ± 10.08% and a second seeding efficiency of 98.14% ± 0.93% (n = 4). The result was a total seeding efficiency of 95.11% ± 4.23%.
Cumulative cell loss during preconditioning
During the 12 hours of preconditioning under flow conditions in vitro, the average cumulative cell loss was 11.45% ± 0.21% (n = 4). The average cumulative cell loss was 2.22% ± 0.32% after 30 minutes, 5.13% ± 0.15% after 3 hours, 7.47% ± 0.20% after 5 hours, and 9.78% ± 0.35% after 7 hours.
In vivo experiments
Postmortem examination and gross observations
Gross examination of the luminal surfaces of the LVAD revealed an intact cell layer and complete coverage of both artificial surfaces after 7 days of implantation (Fig 4).
There was no evidence of infection or thrombus on either luminal surface, within the graft, or within the aortic anastomosis. Lungs, liver, rumen, spleen, kidney, rete mirabile, adrenal glands, and brain showed no gross and no histologic evidence of emboli, ischemia, or infarction.
View larger version (66K):
[in this window]
[in a new window]
|
Fig 4. Gross appearance of the implanted left ventricular assist device’s biomaterial surfaces after 7 days of implantation in vivo. (Left) Textured polyurethane surface; (right) sintered titanium surface. The luminal surfaces of the left ventricular assist device are completely covered with an intact cell layer.
|
|
Scanning electron microscopy and transmission electron microscopy
Scanning electron microscopy revealed an extensive amount of extracellular matrix components and an intact, well-incorporated cellular lining on the sintered titanium and polyurethane surfaces of the implanted LVAD (Fig 5).
Transmission electron microscopy revealed a well-established monolayer of chondrocytes (Fig 6).
No endothelial cells were seen.
View larger version (75K):
[in this window]
[in a new window]
|
Fig 5. Scanning electron microscopy of the implanted left ventricular assist device’s biomaterial surfaces after 7 days of implantation in vivo. An extensive amount of extracellular matrix and an intact, well-incorporated cellular coating were noted on the textured polyurethane (left) and sintered titanium (right) surfaces of the implanted device. Bar (left) = 10 µm; bar (right) = 20 µm.
|
|
View larger version (108K):
[in this window]
[in a new window]
|
Fig 6. Transmission electron microscopy of the implanted left ventricular assist device’s biomaterial surfaces after 7 days of implantation in vivo. (Left) Textured polyurethane surface; (right) sintered titanium surface. A well-established monolayer of chondrocytes was revealed. No endothelial cells were seen. Bar = 500 nm.
|
|
|
Comment
|
|---|
Using LVADs and a calf model, we have demonstrated that it is feasible to line the luminal artificial surfaces of a cardiovascular prosthesis with autologous chondrocytes derived from auricular elastic cartilage and to expose that lining to the bloodstream for relatively short periods of time without damaging or loosening it. Our experience shows that auricular elastic cartilage is an accessible source of autologous cells, easily harvested from the ear under local anesthesia. It also shows that chondrocytes can be isolated more efficiently than vascular smooth muscle cells, that they can be efficiently preconditioned and seeded, and that they can adhere strongly to artificial surfaces. Overall, our results support our initial hypothesis that chondrocytes adhere strongly to the artificial surfaces of the LVAD because of their ability to produce collagen II and possibly other important constituents of extracellular matrix.
In performing our feasibility study, we chose to use the HeartMate LVAD. The point should be made that this choice was not based on any need or desire on our part to improve the biocompatibility of the luminal surfaces of the HeartMate, which are already relatively nonthrombogenic, but rather on our extensive experience with this particular type of cardiovascular prosthesis. In early calf [9] and human [10] studies, endothelial cells were found on samples taken from the luminal surfaces of this LVAD after implantation. Presumably, blood-borne endothelial cells or endothelial cell precursors had been deposited on the blood-contacting surfaces through a process known as fallout healing [11], encouraging the immediate deposition of a stable, uniform, antithrombogenic, nonhemolytic cell lining. This may, in part, explain the low reported incidence of thrombogenicity and clinical thromboembolic problems associated with the use of the HeartMate.
It has been reported that endothelial cells exhibit cell surface receptors involved in adhesion to collagen [12]. Therefore, one could imagine that the chondrocyte lining would function as an initial cell layer to which subsequent endothelial cells could adhere with relative ease. However, a lining of chondrocytes, with its ability to produce negatively charged collagen and present it to the circulating blood of the host, is more likely to attract platelets and increase the thrombogenicity of the interface. This may explain why in the present study endothelial cells were not identified in the one LVAD that was implanted into the tissue donor calf for 7 days. The short time of exposure of the cell lining to the bloodstream in vivo may be another contributing factor, as fallout endothelialization reportedly occurs after 4 weeks [11, 13].
It is important to remember that in the present study we used a chondrocyte lining that was not genetically engineered and that was, therefore, not expected to reduce thrombogenicity or improve biocompatibility. It is our plan, in subsequent studies, to show that these easily accessible chondrocytes can also be genetically engineered to produce endothelial-like factors while still remaining functional after seeding onto artificial surfaces. This may result in the desired combination of availability, strong surface adhesion, and effective production of antithrombogenic factors at a blood–biomaterial interface while simultaneously increasing the possibility of fallout healing (ie, the deposition of circulating endothelial cells or endothelial cell precursors on the seeded surface).
Our group has already shown that smooth muscle cells can be genetically engineered to produce nitric oxide, a platelet aggregation inhibitor [3]. Our next goal, therefore, is to show that the same can be done with auricular chondrocytes, thereby improving the accessibility of the cell line while retaining its adhesive properties. Our successful feasibility study in a calf model warrants further studies in vivo for longer exposure periods.
Our present results should find application not only in LVADs, but also in other devices and vascular prostheses (eg, stents and grafts) whose geometry and design might lend them to easier seeding at lower cost. Our findings also have some implications beyond the immediate scope of this study for the use of auricular cartilage in tissue engineering. For example, the ideal cell source for tissue engineering a heart valve still remains a mystery [14]. Other investigators have successfully used a mixed-cell population of vascular cells from ovine carotid arteries to create a heart valve on a scaffold of biodegradable porous polyhydroxyalkanoate [14]. Perhaps our methodology might be useful in this regard.
In conclusion, auricular elastic cartilage is a ready and easily accessible source of chondrocytes whose ability to produce extracellular matrix constituents, such as collagen, allows them to adhere strongly to the luminal surfaces of cardiovascular prostheses. The simple method of harvesting, isolating, and expanding auricular chondrocytes presented here could be used to provide strongly adherent autologous cell linings for LVADs and other cardiovascular prostheses.
|
Acknowledgments
|
|---|
This work was partially funded by a grant (HL 53233) from the National Institutes of Health (TS-B) and a Roderick MacDonald Research Fellowship (DR). We sincerely thank Jacki Abrams, MD, Kamuran A. Kadipasaoglu, PhD, David Engler, PhD, and Christine L. Tock, MD, PhD, for their helpful discussions; JudeRichard for his editorial assistance; and Patricia Dillard, Ralph Nichols, Melissa Mayo, and Tommy Riess for their technical assistance.
|
Footnotes
|
|---|
* Doctor Scott-Burden passed away on April 19, 2000. 
|
References
|
|---|
-
Bordenave L., Remy-Zolghadri M., Fernandez P., Bareille R., Midy D. Clinical performance of vascular grafts lined with endothelial cells. Endothelium 1999;6:267-275.[Medline]
-
Luscher T.F., Vanhoutte P.M. The endothelium: modulator of cardiovascular function. Boca Raton: CRC Press, 1990.
-
Scott-Burden T., Tock C.L., Bosely J.P., et al. Nonthrombogenic, adhesive cellular lining for left ventricular assist devices. Circulation 1998;98(suppl 19):II339-II45.
-
Megerian C.A., Weitzner B.D., Dore B., Bonassar L.J. Minimally invasive technique of auricular cartilage harvest for tissue engineering. Tissue Eng 2000;6:69-74.[Medline]
-
Fedewa M.M., Oegema T.R., Jr, Schwartz M.H., MacLeod A., Lewis J.L. Chondrocytes in culture produce a mechanically functional tissue. J Orthop Res 1998;16:227-236.[Medline]
-
Bernhard W.F., Poirier V., LaFarge C.C., Carr J.G. A new method for temporary left ventricular bypass: preclinical appraisal. J Thorac Cardiovasc Surg 1975;70:880-895.[Abstract]
-
Tock C.L., Bosley J.P., Parnis S.M., et al. A genetically engineered, nonthrombogenic cellular lining for LVADs: in vitro preconditioning before in vivo implantation. ASAIO J 1999;45:172-177.[Medline]
-
Robinson W.J., Benedict B.S., Daly D.T., et al. An abdominal left ventricular assist device: preclinical studies. Ann Thorac Surg 1975;19:540-551.[Abstract]
-
Liss R.H., Edmonds C.H., Fuqua J.M., Norman J.C. Pseudo or neointimal viability: a controversial criterion of artificial heart pump linings. Med Instrum 1975;9:81.
-
Frazier O.H., Baldwin R.T., Eskin S.G., Duncan J.M. Immunochemical identification of human endothelial cells on the lining of a ventricular assist device. Tex Heart Inst J 1993;20:78-82.[Medline]
-
Shi Q., Wu M.H., Hayashida N., et al. Proof of fallout endothelialization of impervious Dacron grafts in the aorta and inferior vena cava of the dog. J Vasc Surg 1994;20:546-557.[Medline]
-
Fernandez P., Bareille R., Conrad V., Midy D., Bordenave L. Evaluation of an in vitro endothelialized vascular graft under pulsatile shear stress with a novel radiolabeling procedure. Biomaterials 2001;22:649-658.[Medline]
-
Shi Q., Wu M.H., Onuki Y., et al. The effect of flow shear stress on endothelialization of impervious Dacron grafts from circulating cells in the arterial and venous systems of the same dog. Ann Vasc Surg 1998;12:341-348.[Medline]
-
Sodian R., Hoerstrup S.P., Sperling J.S., et al. Tissue engineering of heart valves: in vitro experiences. Ann Thorac Surg 2000;70:140-144.[Abstract/Free Full Text]
Top
Abstract
Introduction
