Mechanisms of bioprosthetic heart valve failure: Fatigue ...

[Pages:7]Mechanisms of bioprosthetic heart valve failure: Fatigue causes collagen denaturation and glycosaminoglycan loss

Narendra Vyavahare,1 Matthew Ogle,2 Frederick J. Schoen,3 Robert Zand,4 D. Claire Gloeckner,5 Michael Sacks,5 Robert J. Levy1 1Division of Cardiology, Children's Hospital of Philadelphia and the Department of Pediatrics, University of

Pennsylvania School of Medicine, 34th Street and Civic Center Boulevard, Philadelphia, Pennsylvania 19104 2St. Jude Medical, Inc., St. Paul, Minnesota 3Department of Pathology, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts 4Department of Biophysics, University of Michigan, Ann Arbor, Michigan 5Department of Biomedical Engineering, University of Miami, Coral Gables, Florida

Received 14 October 1998; accepted 10 November 1998

Abstract: Bioprosthetic heart valve (BPHV) degeneration, characterized by extracellular matrix deterioration, remodeling, and calcification, is an important clinical problem accounting for thousands of surgeries annually. Here we report for the first time, in a series of in vitro accelerated fatigue studies (5?500 million cycles) with glutaraldehyde fixed porcine aortic valve bioprostheses, that the mechanical function of cardiac valve cusps caused progressive damage to the molecular structure of type I collagen as assessed by Fourier transform IR spectroscopy (FTIR). The cyclic fatigue caused a progressive loss of helicity of the bioprosthetic cuspal collagen, which was evident from FTIR spectral changes in the amide I carbonyl stretching region. Furthermore, cardiac valve fatigue in these studies also led to loss of glycosaminoglycans (GAGs) from the cuspal extracellular matrix. The GAG levels in glutaraldehyde crosslinked porcine

aortic valve cusps were 65.2 ? 8.66 g uronic acid/10 mg of dry weight for control and 7.91 ? 1.1 g uronic acid/10 mg of dry weight for 10?300 million cycled cusps. Together, these molecular changes contribute to a significant gradual decrease in cuspal bending strength as documented in a biomechanical bending assay measuring three point deformation. We conclude that fatigue-induced damage to type I collagen and loss of GAGs are major contributing factors to material degeneration in bioprosthetic cardiac valve deterioration. ? 1999 John Wiley & Sons, Inc. J Biomed Mater Res, 46, 44?50, 1999.

Key words: type I collagen; FTIR spectroscopy; in vitro accelerated fatigue; bioprosthetic heart valve degeneration; biomechanical strength

INTRODUCTION

Treatment of cardiac valve disease requires thousands of cardiac surgery procedures each year.1 Cardiac valve failure in both valvular disease and prosthetic valve deterioration is believed to be due in part

Correspondence to: Dr. Robert J. Levy; e-mail: levyr@ email.chop.edu

Contract grant sponsor: National Heart, Lung, and Blood Institute; Contract grant number: HL-38118

Contract grant sponsor: St. Jude Medical, Inc. Contract grant sponsor: American Heart Association, Florida Affiliate (to M. Sacks) Contract grant sponsor: Joseph Stokes, Jr., Research Institute of the Children's Hospital of Philadelphia (to R. J. Levy) Contract grant sponsor: University of Pennsylvania School of Medicine (to R. J. Levy)

? 1999 John Wiley & Sons, Inc. CCC 0021-9304/99/010044-07

to the unique hemodynamic forces intrinsic to this tis-

sue. This hypothesis appears to be particularly rel-

evant to cardiac valve replacements derived from

chemically fixed heterograft tissues (bioprostheses),

which have been widely used over the past three decades.2 Bioprosthetic heart valve (BPHV) replacements have significant advantages over mechanical cardiac valve prostheses in terms of a reduced risk for thromboembolic complications; however, they frequently fail clinically because of material deterioration and associated calcium phosphate deposits3?5 at stressed regions, such as valvular cusp commissures and points of maximal cuspal flexion.6?9 In addition to calcification, ultrastructural disruption of the extracellular matrix (ECM), including type I collagen fibrils has also been noted in clinical retrievals.10 Nevertheless, little is known about the molecular mechanisms whereby mechanical stress leads to tissue deterioration.

MECHANISMS OF BPHV FAILURE: FATIGUE CAUSES COLLAGEN DENATURATION AND GAG LOSS

45

We hypothesize that mechanical stress-induced matrix degeneration in bioprosthetic aortic valve cusps is caused by both molecular damage to type I collagen and loss of glycosaminoglycans (GAGs). These changes in the connective tissue ECM may explain the mechanism of the bioprosthetic cuspal structural degeneration observed with long-term clinical use, as well as that observed in aging valves. We report herein the results of a study that provides evidence of both molecular damage to the tertiary structure of type I collagen and loss of GAGs, secondary to the cyclic fatigue conditions associated with cardiac valvular biomechanics.

MATERIALS AND METHODS

Preparation of type I collagen films

Type I collagen solution (32 mL, bovine dermal collagen, Vitrogen, Collagen Corporation, Fremont, CA, concentration of 3 mg/mL in 0.012N HCl) was mixed with 4 mL of 10? phosphate buffer saline solution (0.2M Na2HPO4, 1.3M NaCl, pH 7.4) and 4 mL 0.1N NaOH at below 5?7?C. The final pH was adjusted to 7.4 by the addition of few drops of 0.1M HCl or 0.1M NaOH. The neutralized isotonic collagen solution was gelled at 37?C for 2 h followed by drying in a hood and exhaustive washing with sterile water to remove salts. The collagen films were glutaraldehyde crosslinked using conditions identical to those for preparing BPHVs (see below).

Preparation of porcine aortic valves

Stentless porcine aortic valves (25-mm diameter, Toronto SPV, St. Jude Medical Inc., St. Paul, MN) were used for the study. These aortic valves were glutaraldehyde crosslinked with 0.6% glutaraldehyde in 0.05M HEPES buffer (pH 7.4) for 24 h followed by additional crosslinking for 6 days in 0.2% glutaraldehyde in HEPES buffer. The valves were stored in glutaraldehyde solution until use.

In vitro cyclic fatigue experiments

bacterial growth) at 1000 cycles/min at 25?C for 5 million cycles. One million cycles equates to 11.5 days of valve function at a heart rate of 60 beats/min. Glutaraldehyde fixed porcine aortic valve bioprostheses (25-mm diameter, St. Jude Medical) were custom mounted and cycled under the same accelerated fatigue conditions for 5?500 million cycles.

IR spectroscopy of aortic valve cusps and collagen

The IR spectra (50 scans at 4 cm-1 resolution) of the surfaces of both collagen films and valve cusps in the hydrated state were obtained using Fourier transform IR spectroscopy (FTIR, Perkin Elmer model 1740, Stamford, CT) with a fixed 45? angle attenuated total reflectance (ATR, ZnSe) cell attachment. Thus, hydrated collagen film samples and porcine aortic valve cusps were placed on an ATR cell and the FTIR scans were obtained. Water was used as a background reference and was subtracted by the computer program from the sample spectra. Spectra were deconvoluted to resolve the amide peaks by the software program provided by Perkin Elmer (IR Data Manager).

GAG analyses and histology

Quantitative GAG content of the cusps was performed with a uronic acid assay using glucoronolactone standard curves.12 The glutaraldehyde fixed porcine aortic valve cusps were extracted by 1.0M sodium hydroxide treatment for 18 h followed by trichloroacetic acid (10%) precipitation to remove proteins. The supernatant was dialyzed against water and the GAGs were precipitated by cetylpyridinium chloride (CPC) addition. The CPC-GAG precipitate was dissolved in 10% sodium acetate solution and reprecipitated with the addition of ethanol according to the procedure reported for the human heart valves.13 Cuspal morphology was investigated with formalin fixed specimens of cycled valves embedded in paraffin, sectioned at 6 m, and stained with Movat's pentachrome reagent. The alcian blue component of this stain, in particular, stains the GAGs blue.

Sections of a glutaraldehyde fixed porcine bioprosthesis (25-mm diameter, St. Jude Medical), which was implanted in a sheep as a mitral valve replacement (150 days),14 were obtained for GAG staining. Sheep surgeries were performed according to the NIH guidelines (NIH Publication 85-23, Rev. 1985) for the care and use of laboratory animals.

Porcine aortic valves were subjected to accelerated flex fatigue conditions using a standard protocol utilized for valve prosthesis testing for United States Food and Drug Administration approval.11 Glutaraldehyde crosslinked type I collagen films (5 ? 1 cm-1 in dimensions, 100 m thick) were mounted as a partially occluding membrane positioned in the prosthesis mounting of a cardiac valve prosthesis accelerated fatigue tester (Shelhigh Inc., Springfield, NJ). Collagen films were then subjected to the pulsatile flow of sterile water (with 0.01% sodium azide added to prevent

Aortic valve cusp biomechanical studies

Because the aortic valve cusp is composed of three distinct structural layers (the ventricularis, spongiosa, and fibrosa), those with the curvature and against the curvature direction were tested. Similarly, circumferential and radially cut specimens were both tested for flexural rigidity. We quantitated the flexural rigidity in a series of specimens (10 per each fatigue condition) subjected to three point bending us-

46

VYAVAHARE ET AL.

ing an apparatus adapted from the design of Xie et al.15 Each sample (3 ? 12 ? 0.5 mm) was tested in three point bending. A load was applied to the center of each specimen by a thin stainless steel bar calibrated to a known displacement?load relationship. Specimens were preconditioned to a known displacement in both directions. Each test was recorded on an S-VHS video system. From the recorded test, an image grabbing board was used to take one image at a reference displacement and at the maximum displacement. The displacement, calculated force, and the specimen dimensions were used to calculate a flexural rigidity index (FRI). Maximum differences in the FRI between fatigue samples were obtained with the circumferential direction against curvature.

Statistical methodology

Statistical analyses of data were performed and probability values (p) for significance were calculated using Student's t test. The p values of less than 0.05 were considered to be statistically significant. Correlation coefficients and p values were calculated for the flexural rigidity data.

Figure 1. An overlay of deconvoluted ATR-FTIR spectra for type I collagen films: (A) control uncycled glutaraldehyde crosslinked collagen film; (B) glutaraldehyde crosslinked collagen film subjected to accelerated cyclic fatigue (1000 cycles/min) on a heart valve tester for 5 million cycles (equivalent to 2 months at a rate of 60 cycles/min). (B) In the amide I region, a new peak appeared at 1645 cm-1 after cyclic fatigue, indicating a structural rearrangement of the amide carbonyls.

RESULTS

Cyclic fatigue causes reorganization of type I collagen conformation: ATR-FTIR spectroscopy

Purified collagen films and glutaraldehyde fixed porcine aortic valve bioprostheses (25-mm diameter, St. Jude Medical) were subjected to accelerated flex fatigue conditions using a cardiac valve prosthesis accelerated fatigue apparatus. Representative samples were assessed for type I collagen conformational changes by ATR-FTIR spectroscopy. Type I collagen films subjected to accelerated cyclic fatigue (5 million cycles) demonstrated a significant change in absorbance occurring in the amide I stretching region compared to nonfatigued type I collagen films prepared from the same formulation (Fig. 1). The nonfatigued collagen films showed two prominent peaks (1658 and 1628 cm-1) in the amide I region. After 5 million cycles, the collagen films exhibited a new peak at 1645 cm-1 in the amide I region. Other regions of the FTIR spectra (amide II, NH bending, and CN stretching in the vicinity of 1555 cm-1; and amide III, CN stretching, and NH bending at 1240 cm-1) were comparable for control and cyclic fatigued collagen films (Fig. 1).

Glutaraldehyde fixed porcine aortic cusps subjected to increasing levels of cyclic fatigue demonstrated the same alterations in the amide I region noted in purified type I collagen samples (Fig. 2). In aortic cusps, the band at 1660 cm-1 was noted to broaden after 1 million cycles and resolved into two bands (1660 and

1645 cm-1) at 50 million cycles. These changes intensified with increasing levels of cyclic fatigue (Fig. 2). In the cusp studies, at 500 million cycles, the amide I band at 1631 cm-1 also separated into two peaks at 1631 and 1617 cm-1, suggesting further damage to the collagen structure (Fig. 2).

Furthermore, fatigued aortic valve cusps also demonstrated progressive changes in FTIR spectral peaks not noted in the collagen spectra: the appearance of new broad peaks at 1260, 1084, and 1015 cm-1. The intensity of these new peaks at 1260, 1084, and 1015 cm-1 also increased with the duration of cyclic fatigue (Fig. 2).

GAGs loss from aortic valve cusps

In order to quantitate the loss of GAGs from the aortic valve cusps after fatigue, we performed a uronic acid assay12 for total GAG content for control (uncycled) and cycled valve cusps. Control cusps contained 65.2 ? 8.66 g uronic acid/10 mg of dry weight, while valves subjected to cyclic fatigue (10?300 million) had GAG content ranging from 5.97 to 9.75 g/ 10 mg dry weight (mean of 7.91 ? 1.1; p < 0.01, cycled cusps vs. control).

We also performed morphology studies with cross sections of aortic valve cusps before and after cyclic fatigue using Movat's pentachrome staining. The alcian blue component of this method stains GAGs blue (Fig. 3). Following 200 million cycles, the cuspal surfaces were irregular and locally damaged compared to

MECHANISMS OF BPHV FAILURE: FATIGUE CAUSES COLLAGEN DENATURATION AND GAG LOSS

47

the uniform and smooth surface of uncycled cusps. Moreover, the alcian blue staining of cuspal GAGs was almost completely absent in the fatigued cusps except in the subsurface layers.

To demonstrate that our flex fatigue studies were also relevant to an in vivo situation, we obtained a porcine BPHV (25-mm diameter, St. Jude Medical) retrieved after 150 days from a sheep mitral valve replacement. Movat stained cuspal sections revealed that there was an absence of alcian blue staining within the cupsal spongiosa, which would normally be present due to GAGs [Fig. 3(C)].

Cyclic fatigue of cardiac valve cusps causes reduced bending strength

We sought biomechanical correlates to the molecular and morphologic events associated with progressive cuspal fatigue. We chose flexural rigidity as a mechanical test relevant to in vivo and in vitro cuspal deformations. The bending properties of aortic valve cusps were assessed in a series of specimens previously subjected to flex fatigue conditions in the same range as those noted in our studies as causing collagen damage and GAG loss. The results showed a progressive decrease in flexural rigidity (Fig. 4) after cyclic fatigue, indicating that the cusps had lost their native stiffness.

DISCUSSION

Figure 2. An overlay of deconvoluted ATR-FTIR spectra for glutaraldehyde crosslinked porcine aortic valve cusps subjected to cyclic fatigue on an accelerated heart valve tester (1000 cycles/min) for various numbers of cycles (M = million). Progressive collagen structural rearrangements are evident in the amide I carbonyl stretching region (between 1700 and 1600 cm-1) from 50 million cycles and up. The band at 1660 cm-1 broadens at 1 million cycles and resolves into two bands (1660 and 1645 cm-1) at 50 million cycles (equivalent to 20 months at 60 cycles/min) and above. These changes in the amide carbonyl stretching regions in the FTIR spectra of cuspal collagen indicate progressive abnormal rearrangements of the collagen amide carbonyls. The lower wavelength regions of the spectra show broad changes after the cycling of cusps that were not seen in the type I collagen studies. For example, a new peak at 1260 cm-1 was noted after 1 million cycles and it was progressively more evident with cyclic fatigue. Similarly two broad peaks at 1084 and 1015 cm-1 appeared at 5 million cycles and become increasingly stronger with cyclic fatigue.

In the present studies we showed that the mechanical function of glutaraldehyde crosslinked BPHVs leads to molecular level collagen damage and loss of GAGs. This in turn leads to reduced bending strength of the cusps.

FTIR spectroscopy results

IR spectroscopy was used in the past to study the secondary and tertiary structure of type I collagen.16?22 In particular, the amide I carbonyl stretching region (1700?1600 cm-1) was shown to be very sensitive to regional changes in the triple helical structure of collagen under various conditions including fibrillogenesis and denaturation.19,20,22 The amide carbonyls that are within the triple helix, which are associated intramolecularly with weaker hydrogen bonds, absorb at a higher wavelength in the amide I region (1660 cm-1) while the amide carbonyls, which form relatively stronger intermolecular hydrogen bonds with associated water, absorb at a lower wavelength (1630 cm-1).22 Other studies of the FTIR spectroscopy of

48

VYAVAHARE ET AL.

Figure 3. Histology of glutaraldehyde crosslinked porcine aortic valve cusps: (A) control uncycled cusp, (B) after 200 million cycles on a heart valve accelerated fatigue tester, (C) after sheep mitral valve replacement; implant time 150 days (approx. 13 million cycles). Slides are stained with Movat's pentachrome stain, which specifically stains extracellular glycoproteins and glycosaminoglycans (blue), collagen (yellow), and nuclei and elastic fibers (black). The blue staining in the cusp spongiosa (central layer) caused by glycosaminoglycans is very diminished (B) after in vitro cyclic fatigue, as well as (C) after in vivo sheep heart valve replacement. Original magnification ?25.

type I collagen in aqueous solutions observed three

distinct carbonyl stretching bands at 1660, 1645, and 1630 cm-1.20,22 In particular, the band at 1645 cm-1 was

assigned to the solvent hydrogen bonded glycine residues of the triple helix.20

In the present studies, control glutaraldehyde

crosslinked type I collagen films showed two peaks (1658 and 1628 cm-1) in the amide I region that are

characteristic for native type I fibrillar collagen in the solid state as observed by others16,17,23 and by our group.14 After in vitro cyclic fatigue for 5 million cycles a new peak at 1645 cm-1 in the amide I region was

noticed, indicating structural alterations in these car-

bonyl loci. Thus, type I collagen films in our experi-

ments after 5 million cycles exhibited IR spectral peaks

in the amide I region similar to those reported for type I collagen in aqueous solutions.20,22 The appearance of a peak at 1645 cm-1 suggests that in fatigued type I

collagen an increased number of carbonyls form hy-

drogen bonds with water, perhaps due to a change in helicity caused by mechanically induced molecular fatigue.

Similar amide I spectral changes (a progressive appearance of a new peak at 1645 cm-1) were noted for glutaraldehyde crosslinked porcine heart valves cusps after cycling. Thus, the appearance of a new peak at 1645 cm-1 in cusps at 50 million cycles and above was consistent with our observations in studies of type I collagen films, suggesting a common mechanistic event with comparable water associations with the amide carbonyls after cyclic fatigue. It is of interest that more extensive cyclic fatigue was necessary to cause comparable changes in collagen structure in valvular cusps compared to collagen films. This may be due to the additional ECM supportive structures present in the cusps such as elastin and GAGs that could stabilize the collagen structure to some extent. The progressive appearance of additional peaks in the lower end of the spectrum (at 1260, 1084, and 1015 cm-1) after in vitro cycling of BPHV cusps indicates the possibility of cyclic fatigue-induced cuspal surface alterations and molecular damage to other, as yet unidentified, structural components within the aortic valve cusp.

GAG loss from BPHV cusps

Figure 4. Diminished valve cusp flexural rigidity after accelerated fatigue of increasing durations. The valve cusps (specimen size 2 ? 10?15 mm) were bent against the natural curvature of the cusp. Flexural rigidity is expressed as Newton meters (Nm2 ? 106). n = 10, linear regression r = 0.997, p = 0.001.

The major finding of this study is that the GAGs from the cusps leach out under conditions of accelerated fatigue (at 10 million cycles, equal to 6 months of heart valve function). This is irrespective of the fact that BPHVs are chemically fixed with glutaraldehyde crosslinking. Similar cuspal GAG loss was recently reported in clinical BPHV retrievals.24 The present study suggests that flexural fatigue induces loss of GAGs (at 10 million cycles) prior to structural alterations in type I collagen, which occurs at 50 million cycles and above (Fig. 2). The GAG loss was also prevalent in the in vivo

MECHANISMS OF BPHV FAILURE: FATIGUE CAUSES COLLAGEN DENATURATION AND GAG LOSS

49

studies with sheep valve replacements [Fig. 3(C)]. These observations confirm our chemical analyses and morphological results from in vitro fatigue studies, indicating that circulatory functioning of bioprosthetic valve cusps leads to a depletion of GAGs.

Such a loss of GAGs from BPHV cusps could be detrimental for two reasons. First, these highly hydrated GAGs present in the spongiosa of the cusp are thought to be involved in modulating the physiologic cuspal biomechanical relationships between the major cuspal tissue planes (the fibrosa and the ventricularis) during valve flexion.25,26 Thus, the loss of GAGs could potentiate valvular deterioration. Second, the highly negative charge of GAGs has been hypothesized to attract calcium, thereby preventing calcium phosphate nucleation.27 Thus, fatigue related GAG loss from the cuspal matrix could also hypothetically contribute to cuspal calcification based on this mechanism.

Our bending strength results showed a progressive decrease in flexural rigidity (Fig. 4) after cyclic fatigue, indicating that the cusps had lost their native stiffness and thus could also be vulnerable to material failure. This in turn could cause each cusp to be more susceptible to progressive mechanical trauma due to continued flexing.28 We hypothesize that this change in cuspal stiffness is due to the major changes in the molecular components of the connective tissue matrix and cuspal surface damage as noted above. Thus, cuspal failure could supervene through an interaction of mechanisms involving a loss of GAGs, collagen structural modifications, and an overall diminution in cuspal strength necessary for bending and flexion. These mechanisms interact to create the substrate for cuspal failure.

CONCLUSIONS

In conclusion, we demonstrated the following:

1. Cyclic fatigue of glutaraldehyde crosslinked type I collagen under accelerated conditions simulating aortic valve biomechanics causes a characteristic structural change in collagen, which was evident within the amide I carbonyl stretching region noted with IR spectroscopy, that may indicate the change in helicity caused by mechanically induced molecular fatigue.

2. Cyclic fatigue of glutaraldehyde fixed porcine aortic valve cusps also results in similar changes in the amide I region of the IR spectra as seen in purified type I collagen, indicating comparable collagen structural alterations.

3. GAGs are removed from these aortic valve cusps during flex fatigue at earlier stages (by 10 million cycles approximating less than 6 months of heart

valve function) than collagen damage takes place. The GAG loss in turn could lead to increased interfibrillar calcification, as well as inadequate material stability to withstand in vivo mechanical stresses. 4. These molecular events associated with valvular fatigue may be the basis for the reduced flexural rigidity noted in our studies. Because specific molecular changes in collagen and other structural components may contribute to the mechanism of fatigue-induced cardiac valve failure and associated calcification, future designs of BPHVs should consider stabilizing these macromolecular components in order to avoid deterioration due to functional mechanical stresses.

References

1. Edmunds LH, Clark RE, Cohn LH, Grunkemeier GL, Miller DC, Weisel RD. Guidelines for reporting morbidity and mortality after cardiac valvular operations. Eur J Cardio Thorac Surg 1996;10:812?816.

2. Fann JI, Miller DC, Moore KA, Mitchell RS, Oyer PE, Stinson EB, Robbins RC, Reitz BA, Shumway NE. Twenty-year clinical experience with porcine bioprostheses. Ann Thorac Surg 1996; 62:1311?1312.

3. Pelletier L, Carrier M, Leclerc Y, Lepage G, DeGuise P, Dyrda I. Porcine versus pericardial bioprostheses: A comparison of late results in 1,593 patients. Ann Thorac Surg 1989;47:352?361.

4. Potkin BN, McIntosh CL, Cannon RB, Roberts WC. Bioprostheses in tricuspid and mitral valve positions for 95 months with heavier calcific deposits on the right-sided valve. Am J Cardiol 1988;61:947?949.

5. Schoen FJ, Collins JJ, Cohn LH. Long-term failure rate and morphologic correlations in porcine bioprosthetic heart valves. Am J Cardiol 1983;51:957?964.

6. Thubrikar MJ, Deck JD, Aouad J, Nolan SP. Role of mechanical stress in calcification of aortic bioprosthetic valves. J Thorac Cardiovasc Surg 1983;86:115?125.

7. Sabbah HN, Hamid MS, Stein PD. Mechanical stresses on closed cusps of porcine bioprosthetic valves: correlation with sites of calcification. Ann Thorac Surg 1986;42:93?96.

8. Grabenwoger M, Grimm M, Eybl E, Kadletz M, Havel M, Kostler P, Plenk H, Bock P, Wolner E. New aspects of the degeneration of bioprosthetic heart valves after long-term implantation. J Thorac Cardiovasc Surg 1992;104:14?21.

9. Jorge HE, Fernandez P, de la Torre Escudero C, Zurita N, Garcia M, Paez JM, Castillo OJ. Influence of stress on calcification of delipidated bovine pericardial tissue employed in construction of cardiac valves. J Biomed Mater Res 1996;30: 411?415.

10. Ferrans VJ, Spray TL, Billingham ME, Roberts WC. Structural changes in glutaraldehyde-treated porcine heterografts used as substitute cardiac valves: Transmission and scanning electron microscopic observations in 12 patients. Am J Cardiol 1978;41: 1159?1184.

11. Johnson DM, Chwirut DJ, Regnault WF. FDA's requirements for in-vitro performance data for prosthetic heart valves. J Heart Valve Dis 1994;3:228?234.

12. Bitter T, Muir HM. A modified uronic acid carbazole reaction. Anal Biochem 1962;4:330?334.

13. Murata K. Acidic glycosaminoglycans in human heart valves. J Mol Cell Cardiol 1981;13:281?292.

50

VYAVAHARE ET AL.

14. Vyavahare NR, Hirsch D, Learner E, Baskin JZ, Schoen FJ, Bianco R, Kruth HS, Zand R, Levy RJ. Prevention of bioprosthetic heart valve calcification by ethanol preincubation: efficacy and mechanism. Circulation 1997;95:479?488.

15. Xie J, Zhou J, Fung YC. Bending of blood vessel wall: stress? strain laws of the intima?media and adventitial layers. ASME J Biomech Eng 1995;117:136?145.

16. Susi H, Ard JS, Carroll RJ. The infrared spectrum and water binding of collagen as a function of relative humidity. Biopolymers 1971;10:1597?1604.

17. Doyle BB, Bendit EG, Blout ER. Infrared spectroscopy of collagen and collagen like peptides. Biopolymers 1975.14:937?957.

18. Jakobsen RJ, Brown LL, Hutson TB, Fink DJ, Veis A. Intermolecular interactions in collagen self-assembly as revealed by Fourier transform infrared spectroscopy. Science 1983;220: 1288?1290.

19. Lazarev YA, Grishkovsky A, Khromova TB. Amide I band of IR spectrum and structure of collagen and related polypeptides. Biopolymers 1985;24:1449?1478.

20. Payne KJ, Veis A. Fourier transform IR spectroscopy of collagen and gelatin solutions: deconvolution of the amide I band for conformational studies. Biopolymers 1988;27:1749?1760.

21. Renugopalkrishnan V, Chandrakasan G, Moore S, Hutson TB, Berney CV, Bhatnagar R. Bound water in collagen. Evidence

from Fourier transform infrared and Fourier transform infrared photoacoustic spectroscopic study. Macromolecules 1989; 22:4121?4124.

22. George A, Veis A. FTIRs in H2O demonstrates that collagen monomers undergo a conformational transition prior to thermal self-assembly in vitro. Biochemistry 1991;30:2372?2377.

23. Jackson M, Choo LP, Watson PH, Halliday WC, Mantsch HH. Beware of connective tissue proteins: assignment and implications of collagen absorptions in infrared spectra of human tissue. Biochim Biophys Acta 1995;1270:1?6.

24. Mako WJ, Calabro A, Ratliff NB, Vesely I. Loss of glycosaminoglycans (GAGs) from implanted bioprosthetic heart valves. In: Proceedings of the American Heart Association, 70th Scientific Sessions; Orlando, FL. 1997.

25. Schoen FJ. Aortic valve structure?function correlations: role of elastic fibers no longer a stretch of the imagination [Editorial]. J Heart Valve Dis 1997;6:1?6.

26. Thubrikar M, editor. The aortic valve. Boca Raton, FL: CRC Press; 1990.

27. Blumenthal NC. Mechanisms of inhibition of calcification. Clin Orthop Rel Res 1989;247:279?89.

28. Sacks MS, Smith DB. Effects of accelerated testing on porcine heart valve fiber architecture. Biomaterials. In press.

 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download