HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Zilla, P. Articles by von Oppell, U. O. Search for Related Content PubMed PubMed Citation Articles by Zilla, P. Articles by von Oppell, U. O.

Ann Thorac Surg 2000;70:2091-2095
© 2000 The Society of Thoracic Surgeons

---

Original article: cardiovascular

High glutaraldehyde concentrations mitigate bioprosthetic root calcification in the sheep model

^a Department of Cardiothoracic Surgery, University of Cape Town Medical School, Cape Town, South Africa

Accepted for publication May 9, 2000.

Address reprint requests to Dr Zilla, Department of Cardiothoracic Surgery, Cape Heart Centre, University of Cape Town Medical School, Anzio Road, Observatory 7925, Cape Town, South Africa
e-mail: ctszilla{at}Samiot.uct.ac.za

	Abstract

Top Abstract Introduction Material and methods Results Comment Acknowledgments References

 
Background. Fixation at high glutaraldehyde (GA) concentrations mitigated bioprosthetic calcification in the rat model. The present study intended to verify this observation in the circulatory sheep model.

Methods. Porcine aortic roots were either fixed in 0.2%, 1.0%, or 3.0% GA. Eight roots per group were implanted in the distal aortic arch of sheep. After six weeks and six months calcification and inflammation were quantitatively and qualitatively assessed.

Results. By increasing the GA concentration from 0.2% to 3.0%, aortic wall calcification could be reduced by 38% after 6 weeks and 34% after 6 months of implantation (p < 0.01). Mineralization coincided with the presence of elastin although calcium was predominantly found in cell nuclei and membranes. Leaflet calcification was absent in all groups after 6 weeks but in a few leaflets presented as heterogeneous, nodular spongiosa deposits after 6 months. Overall, differences between 0.2%-, 1.0%-, and 3.0%-fixed tissue were quantitative but not qualitative regarding distribution patterns. There was no significant difference in inflammatory host reaction between all groups.

Conclusions. We have shown in the circulatory sheep model that the anticalcific effect of better cross-linking seems to outweigh the intrinsic pro-calcific effect of GA accumulation in bioprosthetic aortic wall tissue.

	Introduction

Top Abstract Introduction Material and methods Results Comment Acknowledgments References

 
Xenograft heart valves are cross-linked to reduce antigenicity [1] and increase degradation resistance [2]. The most widely used fixative is glutaraldehyde (GA). Unfortunately, the excellent cross-linking ability of this dialdehyde is opposed by its alleged potential to intrinsically facilitate tissue calcification [3]. For decades, the purported pro-calcific effect of GA was the reason why this dialdehyde was used at very low concentrations for commercial valve fixation [4]. However, such low concentrations may result in suboptimal tissue fixation, potentially leading to accelerated degeneration of prosthetic heart valves. Thus, the vilification of GA led to a compromise in tissue fixation that could only be justified with the perception that at least mineralization was mitigated by this measure. Recent studies indicate that this underlying paradigm of contemporary valve fixation may need to be revised. Fixation at increasing concentrations of GA was shown to result in decreasing tissue mineralization in the rat model [5, 6]. This phenomenon was seen in all three basic bioprosthetic tissue types including aortic wall that previously proved to be rather resistant to anticalcification treatment [7]. The present study aimed at confirming this observation in the circulatory sheep model.

	Material and methods

Top Abstract Introduction Material and methods Results Comment Acknowledgments References

 
Experimental design
Immediately fixed porcine aortic roots were implanted into the distal aortic arch of sheep [8]. Three different concentrations of GA were chosen for tissue fixation: 0.2%, 1.0%, and 3.0%. Eight roots were implanted per group: four were explanted after 6 weeks and four after 6 months.

Tissue preparation and fixation
The proximal ascending aorta, including the aortic valve of medium- to large-sized pigs (90 kg to 120 kg), was roughly dissected at the abattoir within 5 minutes of slaughter. In accordance with the previously reported effect of immediate fixation on ultrastructural integrity [9] and calcific degeneration [5], aortic tissue was immediately placed in GA (Saarchem, Holpro Analytic, Krugersdorp, Republic of South Africa; 15 g tissue/100 mL GA in phosphate buffered saline; 4°C). Three different GA concentrations were used: 0.2%, 1.0%, and 3.0%. Based on the fact that cold fixation at 4°C was least tissue damaging [9], the entire fixation procedure was performed at this low temperature (15 g tissue/100 mL GA in phosphate buffered saline; 0.1 mol/L; pH 7.4; 4°C; 7 days). After rinsing in phosphate buffered saline (12 mL/g tissue; 37°C; 24 hours), samples were stored in low-volume phosphate buffered saline (PBS, 0.1 mol/L; 4°C; 800 µL/g tissue) until implantation (3.6 ± 2.1 days). Low volume was defined as the minimum volume necessary to keep the tissue immersed inside a 50 mL tube. The goal was to prevent both dehydration and undefined GA washout.

Implantation
All anesthetic and surgical procedures were approved by the Animal Research and Ethics Committee of the University of Cape Town and complied with the Principles of Laboratory Care and the Guide for the Care and Use of Laboratory Animals (National Institutes of Health, Publication No. 85-23; revised 1985). Juvenile Merino sheep (41.7 ± 5.0 kg) were brought on site 5 days before implantation to allow adaptation. Surgical access was through a left lateral thoracotomy under Gott-shunt protection. Porcine aortic roots were interpositioned in an end-to-end fashion using running 4-0 Prolene sutures (Ethicon; Johnson & Johnson, Somerville, NJ). On day 10, skin sutures were removed and the sheep were transported back to the farm where they were kept until tissue retrieval.

Tissue retrieval and sample processing
After 6 weeks or 6 months, sheep were euthanized under full anesthesia with an overdose of potassium (40 mmol IV). Following excision of the implant together with a generous segment of native aorta, roots were longitudinally opened between the left and right coronary cusp. Subsequently, roots were dissected into three equal thirds, each consisting of an entire leaflet plus adjacent aortic wall. One half of the noncoronary third was placed in 10% formalin for light microscopy, whereas the other half was fixed in 2% GA for electron microscopy. The left coronary third was kept at –20°C for calcium analysis. The right coronary third was placed in formalin (10%) and kept at 4°C as a reserve sample.

Calcium analysis
The leaflet and the infravalvular tissue of the left coronary third were cut off distal to the leaflet insertion. The length of the tissue strips used for calcium analysis, comprising the sinus and supracommissural aorta, was an average 32.4 ± 1.7 mm in the 0.2% group, 29.9 ± 1.1 mm in the 1% group, and 28.2 ± 1.5 mm in the 3% group. Due to the sometimes nodular calcium deposits in the base of the leaflets reaching into the annulus, a demarcation of actual leaflet tissue was not accurately possible and therefore, quantitative but not qualitative calcium analysis was restricted to the supraannular sinus and aortic wall.

After drying at 104°C for 24 hours, samples were weighed and then ashed at 560°C for 12 hours in a Muffle furnace. Ashed tissue was dissolved in 20% hydrochloric acid (Fisher ISOTEMP Model 184, Springvale, Australia) at a ratio of 10 mg dried tissue:1 ml hydrogen chloride (HCl). A final dilution of this solution was made in 0.5% lanthanum chloride (BDH, Spectrosol, Cat. 140413U, BDH, Poole, England). Absorption was measured at 422.7 nanometers on an atomic absorption spectrophotometer (Varian Techtron, model AA1275; Springvale, Australia). A standard curve was obtained using a standard calcium solution (Titrisol, Merck, catalog No. 9943; Merck, Darmstadt, Germany) in 0.5% lanthanum chloride. The level of calcium in the solution was expressed as µg calcium per mg dry weight of tissue.

Histology
Paraffin sections were stained by either hematoxylin and eosin or von Kossa. Longitudinal sections of the entire root, including leaflet, sinus and supracommissural aortic wall, were investigated regarding calcium distribution, tissue erosion, and inflammatory host reaction. Inflammation was assessed using a scoring system from zero to three. A score of zero was the equivalent to no inflammatory cells at the interface of prosthesis and host tissue. A score of one stood for a mild demarcating rim of inflammatory cells; a score of two for a rim of less than 100 µm with or without signs of mild tissue infiltration/erosion; and a score of three for a broad rim of inflammatory cells (> 100 µm) or severe tissue infiltration/erosion.

Transmission electron microscopy
Ultrastructural analysis was restricted to the main site of calcification, namely the distal aortic wall. Two full-thickness blocks of the aortic wall were processed for transmission electron microscopy per explant. Sample processing and microscopy followed standard procedures previously described [9]. Ultrathin sectioning was done with a 5.5 mm diamond knife, allowing full-thickness sectioning of wall tissue.

Statistical analysis
Atomic absorption spectroscopy data were expressed as means ± standard error. Assessment of the effect of increased GA concentration on tissue calcium was performed by single-factor analysis of variance. Between group comparisons with 0.2%GA-fixed tissue was performed using Student’s t test for unpaired data. Inflammatory scores were assigned using a continuous scale. Assessment of the effect of increasing GA concentration on adventitial surface inflammation was performed by single-factor analysis of variance. A significance level of 0.05 or less was accepted as being statistically relevant.

	Results

Top Abstract Introduction Material and methods Results Comment Acknowledgments References

 
Calcium analysis
Increasing concentrations of GA fixation led to decreasing levels of aortic wall calcification (Fig 1). This tendency was consistent in both 6-week and 6-month implants. This decrease in calcification became statistically significant at 6 months although the trend correlated between the two groups. At 6 weeks an increase from 0.2% to 1.0% GA resulted in a 26% drop in aortic wall mineralization (76.5 ± 13.9 µg/mg vs 56.4 ± 14.1 µg/mg; p = 0.184) while fixation at 3% reduced the calcification level by 38% (47.6 ± 8.6 µg/mg; p = 0.076). At 6 months the difference between 0.2% and 1.0% was less distinct (14%; 174.1 ± 11.9 µg/mg vs 149.9 ± 15.7 µg/mg; p = 0.133), but after 6 weeks the difference between 0.2% and 3% fixation was equally as high (34%; 114.8 ± 10.0 µg/mg; p = 0.0076) and this difference was statistically significant.

View larger version (31K): [in this window] [in a new window]   Fig 1. Mean aortic wall calcium values for 6-week and 6-month root implants in the circulatory sheep model. (GA = glutaraldehyde.)

Morphologic analysis
Calcification patterns
Calcification was predominantly confined to the suprasinusoidal aortic wall and the distal sinus of valsalva (Fig 2).

View larger version (67K): [in this window] [in a new window]   Fig 2. Correlation of tissue calcification with the presence of elastin in aortic roots that were implanted in the distal aortic arch for 6 weeks (left pairs) and 6 months (right pairs). (GA = glutaraldehyde; VB = Victoria Blue stain for elastin; VK = von Kossa stain for calcium.)

After 6 months of implantation, calcification was more distinct but followed similar patterns. The suprasinusoidal aortic wall showed a similar sandwich mineralization after 6 weeks but with denser calcification between these surface layers. Increasing concentrations of GA could not fully eradicate but rather mitigated mineralization, whereas distribution patterns of calcium remained unaffected. Similar to 6-week implants, calcification was confined to aortic wall structures containing elastin, resulting in noncalcified sinus walls in the proximity of the annulus. In contrast to 6-week implants, however, a minority of 6-month implants showed huge calcium deposits in the spongiosa of the leaflet. The pattern of leaflet calcification was heterogeneous with mineral nodules either being close to the free edge or reaching deep into the annular tissue. By increasing the concentration of GA, leaflet calcification could also be reduced but not entirely eradicated.

Overall, there were no distinct differences regarding calcification sites between tissue fixed at lower and higher GA concentration but the extent of calcification was obviously more pronounced in the low GA group.

Inflammatory reaction
Histologically, there were no signs of higher cytotoxicity after fixation at high concentrations of GA. Prosthetic tissue was mildly demarcated from host tissue by either a rim of small lymphatic follicles or a patchy foreign body reaction along its adventitial surface. This was reflected in a score of 0.8 vs 0.5, 0.2 vs 0.4, and 0.7 vs 0.7 (p = 0.200/p = 0.642) for tissue fixed in 0.2%, 1.0%, and 3.0% GA concentrations after 6 weeks vs 6 months.

Ultrastructural sites of aortic wall-calcification
Ultrastructural assessment was often difficult because of coalescing mineral deposits. In control 0.2% GA-fixed aortic wall tissue calcification was primarily confined to cell nuclei (Fig 3a) and organelle debris after 6 weeks of implantation. Cell nuclei were often homogeneously filled with calcium masses. To a lesser extent, small globular nodules were found at the inside of cell membranes and occasionally at the interface of elastin and collagen (Figs 3b and c). By increasing the GA concentration, the overall ultrastructural tissue preservation improved. Outside the mineral agglomerates that clearly originated from cells, it was primarily free organelles in areas of poorer ultrastructural preservation that calcified.

View larger version (160K): [in this window] [in a new window]   Fig 3. (a) Transmission electron micrograph of nuclear calcification. The mineralization is confined to the cell nucleus whose entire chromatin structure is obliterated (broad arrow). Small calcium deposits are also scattered through cytoplasm (slim arrows) (aortic wall; 1.0% glutaraldehyde fixation; 6 weeks; mag. 8,000x). (b) High power magnification of calcified area at the interface of elastin and collagen. The infiltrating crystal is more intimately associated with the elastin than the surrounding collagen (aortic wall; 1.0% glutaraldehyde fixation; 6 weeks; mag. 40,000x). (C = Collagen; E = Elastin.) (c) Calcifying membrane vesicles (arrows) partly associated with elastin (aortic wall 3.0% glutaraldehyde fixation; 6 weeks; mag. 10,000x). (d) Massive calcium deposits reaching into elastin. The collagen fibers (arrow) surrounding the centrally located calcium structure indicate a tangential section of a surface-associated formation (0.2% glutaraldehyde-fixed aortic wall; 6 months; mag. 10,000x).

	Comment

Top Abstract Introduction Material and methods Results Comment Acknowledgments References

 
Contrary to a widely held belief, higher concentrations of GA lead to a lower degree of tissue mineralization when used for bioprosthetic cross-linking. After our previous demonstration of this phenomenon in leaflets, pericardium, and aortic wall tissue in the subcutaneous rat model [5, 6], the present study confirmed this finding for aortic wall tissue in the circulatory sheep model.

Because the aortic wall, which holds a key to bioprosthetic root replacement [10], proved to be relatively inert toward otherwise highly effective anticalcification treatments [7], it is remarkable that a reduction of more than one third of tissue calcification can be achieved solely through higher GA concentrations. This is particularly surprising in view of the fact that GA cross-links were regarded as the main culprits for the mineralization of bioprosthetic heart valves [3]. As a result, particularly low concentrations of GA were used, and are still used, in commercial tissue valve fixation. The low degree of cross-linking, however, resulting from this compromise, defies to a large extent the main purpose of tissue fixation, namely the suppression of antigenicity [1, 11] and enzyme activity [12]. In the past, residual antigenicity was not considered to be a major contributing factor to bioprosthetic tissue degeneration. Recent reports, however, showing a correlation between antigenicity and tissue calcification [13] illustrate that one cannot completely dismiss immune responses against tissue valves [11]. Although this lack of biological quiescence of bioprosthetic tissue has been known for a long time [12], it continues to be seen as the minor evil in view of the side effects of GA, such as the facilitation of calcification. This described tendency of GA to intrinsical facilitation of calcification [3] was later supported by the observation that tissue extraction of excess GA [14, 15] significantly reduces tissue mineralization [6, 14]. However, as much as this intrinsic effect of GA on tissue calcification is undisputed, the indiscriminate vilification of GA in the literature of the 1980s partly obscured deeper insight into possible mechanisms of bioprosthetic calcification for almost a decade. It took until the second half of the 1990s until various groups began to directly [5, 13] and indirectly [10] challenge the prevailing paradigm of the predominantly detrimental effect of GA. The apparent paradox of an intrinsically procalcific reagent leading to an overall mitigation of calcification can be explained with the anticalcific net effect of better cross-linking outweighing the inherent promineralizing effect of GA or its reaction products. Retrospectively, one may have surmised the role cross-linking plays in the mitigation of tissue mineralization from previous reports demonstrating significantly lower tissue calcification after long-term storage in GA [16] as opposed to short-term fixation. Because both GA condensation products [17] and cross-links [18] increase over time, the dramatically lower calcification potential after long-term storage further supports the conclusion that the better cross-linking effect outweighed that of accumulating reaction products. The striking coincidence between calcification sites and areas containing elastin suggests that better cross-linking primarily affects structures inherent to elastic arteries. Because our ultrastructural investigations showed elastin to be less affected by calcification than other structures, the reason for the colocalization of elastin and calcium on a light microscopy level may well lie in the colocalization of elastin with another structure. The fact that the calcification sites remained the same after fixation at higher concentrations of GA, whereas overall calcification diminished, further suggest that improved cross-linking only mitigates rather than alters the calcification mechanism.

In summary, it appears that bioprosthetic tissue fixation was based on a false compromise in the past that prevented us from discovering the benefits of thorough cross-linking. Even with an intrinsically procalcific reagent such as GA, we were able to show that better cross-linking not only reduces calcific tissue degeneration in both circulatory and subcutaneous models [5, 6] but also improves ultrastructural tissue preservation [9], reduces macrophage activation [19], and suppresses the immune response [11].

If one extrapolates the principle of high efficiency cross-linking to alternative cross-linking procedures that may not be burdened by the negative side effects of GA, we may hold a key to significantly improved bioprosthetic heart valves.

	Acknowledgments

Top Abstract Introduction Material and methods Results Comment Acknowledgments References

 
We thank Dr Yinxing Zhang, Mona Bracher, Jenny Molde, Nazlia Samodien, and Phillip Christopher for their dedicated technical assistance.

	References

Top Abstract Introduction Material and methods Results Comment Acknowledgments References

 

1. Nimni M.E., Cheung D., Strates B., Kodama M., Sheikh K. Chemically modified collagen: a natural biomaterial for tissue replacement. J Biomed Mater Res 1987;21:741-771.[Medline]
2. Vasudev S.C., Chandy T. Effect of alternative cross-linking techniques on the enzymatic degradation of bovine pericardia, and their calcification. J Biomed Mater Res 1997;35:357-369.[Medline]
3. Golomb G., Schoen F.J., Smith M.S., Linden J., Dixon M., Levy R.J. The role of glutaraldehyde-induced cross-links in calcification of bovine pericardium used in cardiac valve bioprostheses. Am J Pathol 1987;127:122-130.[Abstract]
4. Nimni M.E., Myers D., Ertl D., Han B. Factors which affect the calcification of tissue-derived bioprostheses. J Biomed Mater Res 1997;35:531-537.[Medline]
5. Zilla P., Weissenstein Ch, Bracher M., et al. High glutaraldehyde concentrations reduce rather than increase the calcification of aortic wall tissue. J Heart Valve Dis 1997;6:502-509.[Medline]
6. Weissenstein Ch, Human P., Bezuidenhout D., Zilla P. Glutaraldehyde detoxification on top of enhanced amine cross-linking dramatically reduces bioprosthetic tissue calcification in the rat model. J Heart Valve Dis 2000;9:230-240.[Medline]
7. Chen W., Schoen F.J., Levy R.J. Mechanism of efficacy of 2-amino oleic acid for inhibition of calcification of glutaraldehyde pretreated porcine bioprosthetic heart valves. Circulation 1994;90:323-329.[Abstract/Free Full Text]
8. Chanda J., Kuribayashi R., Abe T. Valved conduit in the descending aorta in juvenile sheep: a useful, cost effective model for accelerated calcification study in systemic circulation. Biomaterials 1997;18:1317-1321.[Medline]
9. Zilla P., Zhang Y., Human P., Koen W., von Oppell U. Improved ultrastructural preservation of bioprosthetic tissue. J Heart Valve Dis 1997;6:492-501.[Medline]
10. Melin G., Rubens M.B., Birks E.J., Bizzarri F., Yacoub M. A quantitative study of calcium deposition in the aortic wall following Medtronic Freestyle compared with Homograft aortic root replacement: a prospective randomized trial. J Heart Valve Dis 2000;9:97-103.[Medline]
11. Human P., Bracher M., Zilla P. The immune response to bioprosthetic tissue—influence of cross-link density. Cardiovasc Pathol 1998;7:334.
12. Maranto A.R., Schoen F.J. Alkaline phosphatase activity of glutaraldehyde-treated bovine pericardium used in bioprosthetic cardiac valves. Circ Res 1988;63:844-848.[Abstract/Free Full Text]
13. Vincentelli A., Latremouille C., Zegdi R., et al. Does glutaraldehyde induce calcification of bioprosthetic tissue?. Ann Thorac Surg 1998;66:S255-S288.
14. Grabenwoeger M., Grimm M., Eybl E., Leukauf C.H., Mueller M.M. Decreased tissue reaction to bioprosthetic heart valve material after L-glutamic acid treatment: a morphological study. J Biomed Mater Res 1992;26:1231-1240.[Medline]
15. Zilla P., Fullard L., Trescony P., et al. Glutaraldehyde detoxification of aortic wall tissue: a promising perspective for emerging bioprosthetic valve concepts. J Heart Valve Dis 1997;6:510-520.[Medline]
16. Schryer P.J., Tomasek E.R., Wright J.T.M. Anticalcification effect of glutaraldehyde-preserved valve tissue stored for increasing time in glutaraldehyde. In: Bodnar E., Yacoub M., eds. Biologic and bioprosthetic valves. New York: Yorke Medical Books, 1986:471.
17. Korn A.H., Feairheller S.H., Filachione E.M. Glutaraldehyde: nature of the reagent. J Mol Biol 1972;65:525-529.[Medline]
18. Huang-Lee L.L., Cheung D.T., Nimni M.E. Biochemical changes, and cytotoxicity associated with the degradation of polymeric glutaraldehyde-derived crosslinks. J Biomed Mater Res 1990;24:1185-1201.[Medline]
19. Dower T., Adler U., Davids L., Zilla P. Increasing cross-linking efficiency mitigates macrophage activation on bioprosthetic tissue. Cardiovasc Pathol 1998;7:295.

This article has been cited by other articles:

				R. A. Manji, L. F. Zhu, N. K. Nijjar, D. C. Rayner, G. S. Korbutt, T. A. Churchill, R. V. Rajotte, A. Koshal, and D. B. Ross Glutaraldehyde-Fixed Bioprosthetic Heart Valve Conduits Calcify and Fail From Xenograft Rejection Circulation, July 25, 2006; 114(4): 318 - 327. [Abstract] [Full Text] [PDF]

				P. Zilla, D. Bezuidenhout, and P. Human Carbodiimide Treatment Dramatically Potentiates the Anticalcific Effect of Alpha-Amino Oleic Acid on Glutaraldehyde-Fixed Aortic Wall Tissue Ann. Thorac. Surg., March 1, 2005; 79(3): 905 - 910. [Abstract] [Full Text] [PDF]

				F. J. Schoen and R. J. Levy Calcification of Tissue Heart Valve Substitutes: Progress Toward Understanding and Prevention Ann. Thorac. Surg., March 1, 2005; 79(3): 1072 - 1080. [Abstract] [Full Text] [PDF]

				P. Human and P. Zilla Characterization of the immune response to valve bioprostheses and its role in primary tissue failure Ann. Thorac. Surg., May 1, 2001; 71 (2007): S385 - S388. [Abstract] [Full Text] [PDF]

				A. Trantina-Yates, C. Weissenstein, P. Human, and P. Zilla Stentless bioprosthetic heart valve research: sheep versus primate model Ann. Thorac. Surg., May 1, 2001; 71 (2007): S422 - S427. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Zilla, P. Articles by von Oppell, U. O. Search for Related Content PubMed PubMed Citation Articles by Zilla, P. Articles by von Oppell, U. O.