| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 2006;132:89-98
© 2006 The American Association for Thoracic Surgery
Cardiopulmonary Support and Physiology |
a Laboratory of Experimental Cardiac Surgery, Department of Cardiovascular Diseases, Katholieke Universiteit Leuven, Leuven, Belgium
b Department of Pathology, Katholieke Universiteit Leuven, Leuven, Belgium.
Received for publication July 25, 2005; revisions received December 11, 2005; accepted for publication February 6, 2006. * Address for reprints: Willem Flameng, MD, PhD, CEHA, Provisorium I, Minderbroedersstraat 17, B-3000 Leuven, Belgium. (Email: willem.flameng{at}med.kuleuven.be).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
METHODS: Calcium content was measured by means of atomic absorption spectrometry in bioprostheses implanted in 120 sheep (age <1 year) for a period of 3 or 6 months.
RESULTS: Bioprostheses calcified significantly in adolescent sheep, but the extent of calcification was multifactorial. Multivariate analysis of the calcium content reveals that age, mitral or pulmonary implant position, prosthesis design (stented or stentless), structure (porcine or pericardial, wall portion or cusp), and antimineralization treatment are independent factors influencing calcification; implant duration beyond 3 months was not. In juvenile sheep (age 5 months) the wall portion, as well as the cusps of the prosthesis, calcified significantly more than in adolescent sheep (age 11 months). Irrespective of age, the cusps of valves implanted in the mitral position calcified more than those in the pulmonary position. The wall portion of stentless valves calcified more than that of stented valves, and pericardial valves calcified less than porcine valves. The surfactant (Tween 80, No-React, and 
-amino-oleic acid) and alcohol (ethanol and octanediol) treatment significantly reduced cusp calcification; sodium dodecylsulfate did not. None of the anticalcification treatments was able to prevent wall calcification in stentless porcine valves.
CONCLUSION: These findings suggest that tissue valve calcification is determined by many independent factors, which can be identified by using adolescent sheep as a preclinical in vivo model.
-amino-oleic acid; KUL = Katholieke Universiteit Leuven; SDS = sodium dodecylsulfate
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Calcium content of the valves was determined by means of atomic spectrometry after implantation for 20 weeks in the mitral or tricuspid position. The objective was to study various antimineralization treatments, and he found that there was a very wide distribution range of calcium content for each valve treatment as a function of individual animals. Therefore a large number of complete experiments are needed to determine the efficacy of a given treatment. In spite of the variability of the calcifications between animals and the inhomogeneity of the calcium deposits within the same valve, calcium content measurement with atomic absorption spectroscopy is still accepted as the gold standard. Alternative techniques are soft radiographic assessment of valve calcifications or histologic determination of calcium deposits in the valve tissue. How much the results of these techniques correlate with each other is unclear. A problem with the use of juvenile animals is their growth potential: they still grow considerably, and a severe patient-prosthesis mismatch can occur. This could be avoided by using adolescent or adult animals for valve testing. Unfortunately, little is known about accelerated calcification in these grown-up animals.
The aim of the present study was to evaluate accelerated calcification of bioprosthetic heart valves in adolescent sheep and to study the potential to differentiate between antimineralization treatments in this animal model. Therefore we studied the effect of age (juvenile vs adolescent), implant position (left or right side), implant duration (3 vs 6 months), valve tissue material (porcine aortic valve or bovine pericardium, aortic wall portion or cusps), valve design (stented or stentless), and antimineralization treatment on valve calcification.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Valves Studied
Fifteen different types of tissue valves were used in this study. The types and their major characteristics are presented in Table 1. All valves are aortic prostheses and are made by the following manufacturers: Medtronic, Inc, Minneapolis, Minn (Mosaic, Hancock MO, Hancock II, Hancock I pulmonary valved conduit, Freestyle aortic root); St Jude Medical, Inc, St Paul, Minn (Epic, Toronto SPV; Sulzer Carbomedics—Photofix); Labcor Laboratorios, Bela Horizonte, Brazil (Labcor, Labcor Duranol, Labcor pulmonary conduit); BioVascular, Inc, St Paul, Minn (Biocor genetic); Sorin Biomedica, Saluggia, Italy (Pericarbon stented, Pericarbon stentless); Edwards Lifesciences, Irvine, Calif (Prima plus, Perimount); Biocor Industria e Pesquisas Ltda, Belo Horizonte, Brazil (Biocor No-react I); AorTech Europe Ltd, Strathclyde, Scotland (Aspire AorTech).
|
Pulmonary valve replacement
After achievement of general anesthesia and artificial ventilation, a left thoracotomy was carried out. After administration of heparin (3 mg/kg), a pneumatic right ventricular assist device (Medos HIA-VAD; Medos-Helmholtz Institute, Aachen, Germany) was installed, with the inflow cannula in the right atrium and the outflow cannula 1 cm proximal from the pulmonary bifurcation. The prosthetic valve was implanted as an interposition in the pulmonary artery. In this way stentless valves were placed into position as a whole root. The native pulmonary valve was destroyed by tearing 2 cusps, and then the right ventricular assist device was stopped and removed.
3
After mitral or pulmonary valve replacement, removal of the cannulas, and careful hemostasis, the chest was closed in layers, with a chest drain in the left pleural space.
No inotropic drugs are used during the procedure. The animal was ventilated at all times. After extubation, the animals returned to the controlled animal facility, where the general health of the sheep was checked daily.
Experimental Models
The combination of animal age, implant position, type of valve design, and type of tissue resulted in the following experimental models: (1) adolescent, mitral, stented, porcine or pericardial; (2) adolescent, pulmonary, stented or stentless, porcine or pericardial; and (3) juvenile, pulmonary, stented or stentless, porcine.
Echocardiography
Transthoracic echocardiography was performed by an experienced echocardiographer. This included a description of heart valve motion from the prosthetic implant in the pulmonic or mitral position.
Animal Death
The animal was prepared and anesthetized as described above, administered heparin (3 mg/kg) intravenously, and killed with an overdose of a KCl solution administered intravenously, and the heart was removed. The implanted bioprosthesis was excised, and macroscopic and soft radiographic images were taken (Figure 1). The valves were analyzed as described below.
|
|
Quantitative calcium content analysis was performed on about 50% of the valve tissue material, as described above.
Nonimplanted control valves
In 23 valves that were never implanted, calcium content was determined, as described above. This series included 345 samples originating from 2 Toronto, 1 Labcor Duranol, 2 Freestyle, 3 Epic, 2 Labcor, 2 Mosaic, 5 Pericarbon stented, 4 Perimount, and 2 Aspire Aortech valves.
Short-term implants
Six valves were recovered from animals that died early (1-5 days) after the operation: 1 Hancock II, 1 Aspire Aortech, 1 Toronto, and 3 Pericarbon stented valves. Calcium content was determined in a total of 90 samples.
Chronically implanted valves
A total of 120 chronically implanted valves (Table 1) were used in this study. Fifteen samples per valve were analyzed on calcium content, which resulted in 1721 values.
Data Management and Statistical Analysis
Normal probability plots (with the Shapiro-Wilks test for normality) were constructed to judge data distribution. Given the important skewed, nonnormal distribution of the calcium content data, logistic transformation was performed (Figure 3). Further statistical analysis consisted of multiple linear regression modeling with a backward stepwise selection technique after insertion of all variables with a univariate P value of less than .1 (univariate analysis by t tests).
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Echocardiography
All prosthetic valves functioned well at 1 week. There were no signs of valve thrombosis recorded. Also, at autopsy, there was no additional thrombus found in and around the native pulmonary valve, which was destroyed during the operation.
Spacial Distribution of Valve Calcification
When all the data obtained from explanted bioprostheses after chronic implantation are considered, calcification in the cusps is significantly lower than in the aortic wall portions of the valve (P < .001). Within the cusps, the highest calcium levels are found in the commissures, and the lowest calcium levels are found in the edges of the cusp. There is no prevalence for calcification between cusps (Figure 4, A). In the wall portion of the valve, the highest values are found in the outflow part of the valve (Figure 4, B). The nature of the calcification was studied in the histologic sections. Extrinsic calcification was found in 9 of 120 valves and was located in the pannus formation at the base of the leaflets. This was found in 4 stentless valves (Biocor I, Prima plus, Toronto, and Pericarbon) and 5 stented valves (Labcor, Labcor Duranol, Mosaic, and 2 Pericarbon valves). All theses valves were implanted in the pulmonary position. All other calcifications were intrinsic calcifications. The differences in calcification between the porcine valve types are analyzed by using multiple comparisons, as presented in Table 2. It is shown that many valve types differ in their calcium content, which might be related, in addition to the age of the sheep, to factors like fixation process, antimineralization treatment, and design. Indeed, these valves represent multiple generations of valves and valve design. To unravel the significance of the different factors influencing valve calcification, multiple types of valves were grouped together, and a univariate and multivariate analysis was performed by using the logistically transformed calcium content data.
|
|
|
These results were compared with those originating from valves that were implanted for only 1 to 5 days. Calcium content was significantly higher than in corresponding areas of similar types of nonimplanted valves: 1.47 ± 0.85 µg/mg in the leaflets, 1.59 ± 0.96 µg/mg in the wall portion, and 1.52 ± 0.89 µg/mg in the entire valve (P > .05).
In adolescent sheep the overall calcium content of valves increased about 10 times during the first 3 months of implantation (to 14.3 µg/mg) but remained at this level during the next 3 months.
Juvenile versus adolescent sheep
As described above, multivariate analysis of the calcium content data revealed that age was a significant and independent factor of increased calcification for the cusp as for the aortic wall portion (Table 3). In an effort to quantify these calcification characteristics, calcium content was determined in all juvenile (5 months of age) and adolescent (11 months of age) sheep. The results are as follows: in juvenile sheep overall calcium content was several times higher than in adolescent sheep, in the leaflets (Figure 5, A) as in the aortic wall portion (Figure 6, A).
|
|
When the calcium contents at 3 to 6 months of all valves implanted in the pulmonary position are compared with those of all valves at the same time interval but implanted in the mitral position, calcium content of the leaflets was about twice as high in the mitral than in the pulmonary implants (Figure 5, B).
Valve design and tissue material
The multivariate analysis indicated that valve design (stented or stentless) and tissue material (pericardium or porcine aorta) are independent factors influencing calcification (Table 3). Stenting is only an independent predictor of wall calcification and not for cusp calcification, but tissue material is important for both wall and cusp calcification. When the results of all stented valves are compared with those of all stentless valves, stentless valves calcify more than stented valves in the wall portion (Figure 6, B). Valves constructed from pericardial tissue calcify less than valves made of porcine aortic roots. When all porcine valves are compared with all pericardial valves, calcium content is higher in the porcine valves compared with that in the pericardial valves, in the leaflets (Figure 5, C) as in the wall portion (Figure 6, C).
Antimineralization treatment
When the calcium content data of valves that received any kind of antimineralization treatment were compared with those of nontreated valves, a significantly lower calcification was found (P < .001). Multivariate analysis indicated that treatment was an independent factor influencing valve calcification (Table 3).
The following antimineralization treatments were included:
-amino-oleic acid (AOA; Mosaic, Freestyle valves) and For the leaflets, any form of antimineralization treatment tested resulted in less calcification than the nonuse of treatment, except for SDS (Figure 7, A). Concerning the wall portion in stentless valves, no treatment had a significant effect in reducing calcification (Figure 7, C). In stented valves the limited wall portion that is present calcifies less when an antimineralization treatment is applied (Figure 7, B).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although the calcification potential in adolescent sheep is less than that in juvenile sheep, it is powerful enough to reveal the factors influencing calcification processes and to distinguish between the effects of different antimineralization treatments.
The finding that young age is an important determinant of tissue prosthesis calcification is not surprising. It has been long known from clinical implants that prosthetic valve calcification and its consequences are strongly related to patient age at implantation.
5
It was, and perhaps still is, believed that pericardial and porcine tissue equally calcifies with time when used as a prosthetic construct. This originates from the finding that both tissues have the same potential for calcification when implanted subcutaneously in rats. This, however, does not imply that they, as valve constructs, will calcify to the same extent. Indeed, results from rats cannot be freely extrapolated to other species
1
and certainly not when blood contact conditions are involved.
6
Nevertheless, the extensive work in sheep from Jones and colleagues
1
and Gallo and associates
7
could not differentiate between the calcification rates of porcine or pericardial valves. The problem is that a mixture of different porcine valves was compared with a mixture of different pericardial valves from different manufacturers. Recently, a sophisticated in vitro technique using holographic interferometry was developed to test tissue calcification.
8
When comparing pericardial valves from one manufacturer with porcine bioprostheses from different manufacturers, no differences in calcification rates can be found.
8
It was concluded that the biologic material of the prosthesis (pericardial or porcine) is not the crucial factor but that mechanical stresses have a much more pronounced effect on calcification. Nevertheless, using the same technology, Glasmacher and coworkers
9
demonstrated that pericardial valves are less prone to calcification than porcine valves when valves from the same manufacturer are compared. Our findings in the present study suggest also that tissue source is an independent predictor of valve calcification in the sense that pericardial valves calcify less than porcine valves. This was confirmed by clinical data when the long-term performance of Carpentier-Edwards porcine bioprostheses and Carpentier-Edwards pericardial bioprostheses were compared.
10
It was found that 10-year freedom from explantation was lower for the porcine (90% ± 2%) than for the pericardial (97% ± 2%, P = .04) valves. The incidence of structural valve degeneration was much higher in porcine than in pericardial valves.
This, however, does not mean that pericardial bioprostheses do not calcify: it was extensively shown in sheep
1,7
and in human subjects
11-13
that pericardial prostheses of different designs and from different manufacturers calcify at different rates. It has long been known that mechanical stress stimulates calcification,
14
and it was clearly shown by Deiwick and colleagues
8
that valve calcification strongly correlates with stress distribution within the valve. This explains the significant influence of valve design on valve calcification.
In the present study several aspects of this interrelationship among valve design, valve structure, and calcification characteristics come about in the adolescent sheep model. First, there is the distribution of valve calcification. We could clearly show that at sites of increased mechanical stress and strain, calcification was pronounced: predominantly at the cuspal commissures, to a lesser extent at the base of the leaflet, and in the least at the free edge. This is in agreement with the findings of Schoen and associates
15
and demonstrates again the sensitivity of the adolescent sheep model.
Remarkable in our studies comparing stented with stentless valves
16
was that the wall calcification occurs more progressively and more rapidly in stentless valves than in their stented counterparts. In previous studies we already focused attention on porcine wall mineralization of tissue valves and revealed an important cell-mediated calcification process.
17
Besides the importance of dead cells and cell remnants as foci of early mineralization, the contribution of elastin, elastolysis, and subsequent calcification of elastic fibers in the aortic wall portion was shown. Thus in general the aortic wall calcifies more than the cusps, most probably because of its typical highly cellularized structure, which contains more elastin. Obviously, stenting in itself reduces aortic wall calcification, so that in stented valves the difference between cusp and wall calcification fades away or even reverses, especially after left-sided (mitral) valve implantation when increased stress accentuates cusp calcification. This might explain why Biedrzycki and associates
18
found that calcification in the aortic wall segments of failed stented human porcine bioprostheses was approximately half that of cuspal calcification. All valves were failing in the aortic or mitral position and were heavily calcified (66 ± 6.3 µg/mg). It was suggested by Trantina-Yates and coworkers
19
that the predominant calcification of the aortic wall is species related and typical for sheep. They found a distinctly higher value of aortic wall calcification in the sheep model compared with that seen in nonhuman primates. The problem with this study is that the difference in aortic wall calcification can be age related rather than species related because the sheep were juveniles and the age of the baboons was unknown. Also, the finding that the aortic wall portion, mainly in stentless valves, calcifies more in the pulmonary position is confirmed by human data: allograft aortic valves or roots implanted in the right-sided circulation frequently fail because of calcific stenosis, often predominantly in the aortic wall portion.
20
Midterm echocardiographic follow-up studies after Ross operations also have shown that homografts in the pulmonary position are prone to a calcification and stenotic process. Independent predictors of homograft failure in the pulmonary position were young age at the time of the operation and the use of aortic instead of pulmonary homografts.
21
Schoen and colleagues
22
suggested that the reduced aortic wall calcification in stented valves might be related to the fact that the stent with cloth might block diffusion of calcium-containing fluid from the adventitial side of the aortic wall. Another possibility is that the valve stent reduces mechanical stress on the small aortic wall segment, which was trimmed contiguous with the cusps to a level slightly distal to the commissures and incorporated into the prosthesis. Until now, there have been no available systematic data on calcification of explanted stentless valves in human subjects, which makes further comparison with the sheep model impossible.
Another independent factor that influences cusp and wall calcification is the implant position (left or right side). Prosthetic valve implantation in the left side of the heart (mitral position) accentuates calcification of the cusps, whereas right-sided implantation accentuates calcification of the wall portion. This was confirmed in human implants,
18
with the highest calcium concentrations shown in the cusps in mitral valve implants.
In general, biologic tissue used in the construction of heart valve bioprostheses is glutaraldehyde treated to prevent degeneration of the material after implantation. However, glutaraldehyde treatment is related to in vivo calcification of the fixed tissue. Various preglutaraldehyde and postglutaraldehyde tissue treatments have been attempted to overcome these detrimental effects. These include detergents or surfactants, such as SDS and polysorbate 80 (Tween 80). Their mode of action is mainly related to removal of cellular remnants and phospholipids from the tissue, and their effect was first described in juvenile sheep by Jones and colleagues.
1
One has to distinguish between the anticalcification effects of a given treatment on the leaflets or on the valve wall portion. On the leaflets, both the detergents and surfactants Tween 80, No-React, and AOA and the alcohols ethanol and octanediol have a similar effect as anticalcification agents. SDS had no effect on cusp protection (Figure 7, A). AOA, Tween 80, and No-React had no effect against aortic wall calcification in porcine stentless valves. In a previous study we could show that ethanol and AOA are superior to SDS as an antimineralization treatment.
23
Whether the No-React process reduces calcification to the level of the detergents or of the more effective anticalcification treatments could not be determined from our adolescent sheep study. Because the mode of action of this procedure was never clarified,
24
it is difficult to explore the real effects of these kinds of procedures.
In conclusion, we demonstrates that the adolescent sheep model is useful and sensitive in the study of bioprosthetic heart valve prostheses, provided sufficient animals are used to overcome variability in calcification data. Also, it became clear that optimal valve testing should take into account the different independent factors influencing bioprosthesis calcification. For example, valved conduits for right-sided implantation in congenital heart disease should be tested in juvenile sheep and in the pulmonary position, whereas stented bioprostheses can be tested preferentially in adolescent sheep in the mitral position.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  W. Flameng, R. Jashari, G. De Visscher, L. Mesure, and B. Meuris Calcification of allograft and stentless xenograft valves for right ventricular outflow tract reconstruction: An experimental study in adolescent sheep J. Thorac. Cardiovasc. Surg., June 1, 2011; 141(6): 1513 - 1521. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Flameng, M. C. Herregods, M. Vercalsteren, P. Herijgers, K. Bogaerts, and B. Meuris Prosthesis-Patient Mismatch Predicts Structural Valve Degeneration in Bioprosthetic Heart Valves Circulation, May 18, 2010; 121(19): 2123 - 2129. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Flameng, B. Meuris, G. De Visscher, C. Cunanan, E. Lane, E. Verbeken, P. Herijgers, and M.-C. Herregods Trilogy Pericardial Valve: Hemodynamic Performance and Calcification in Adolescent Sheep Ann. Thorac. Surg., February 1, 2008; 85(2): 587 - 592. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. De Visscher, H. Blockx, B. Meuris, H. Van Oosterwyck, E. Verbeken, M.-C. Herregods, and W. Flameng Functional and biomechanical evaluation of a completely recellularized stentless pulmonary bioprosthesis in sheep. J. Thorac. Cardiovasc. Surg., February 1, 2008; 135(2): 395 - 404. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Englberger, J. Noti, F. F. Immer, M. Stalder, F. S. Eckstein, and T. P. Carrel The Shelhigh No-React(R) bovine internal mammary artery: a questionable alternative conduit in coronary bypass surgery? Eur J Cardiothorac Surg, February 1, 2008; 33(2): 222 - 224. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. C Santos, L. R Gerola, I. Casagrande, E. Buffolo, and D. T Cheung Stentless Valves Treated by the L-Hydro Process in the Aortic Position in Sheep Asian Cardiovasc Thorac Ann, October 1, 2007; 15(5): 413 - 417. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Vincentelli, F. Wautot, F. Juthier, O. Fouquet, D. Corseaux, S. Marechaux, T. Le Tourneau, O. Fabre, S. Susen, E. Van Belle, et al. In vivo autologous recellularization of a tissue-engineered heart valve: Are bone marrow mesenchymal stem cells the best candidates? J. Thorac. Cardiovasc. Surg., August 1, 2007; 134(2): 424 - 432. [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 |
