Ann Thorac Surg 2001;71:S413-S416
© 2001 The Society of Thoracic Surgeons
Basic research
Effect of ethanol and ether in the prevention of calcification of bioprostheses
Ming Shen, MD, PhDa,
Ali Kara-Mostefa, MDb,
Lin Chen, MDa,
Michel Daudon, PhDb,
Marc Thevenin, PhDb,
Bernard Lacour, PhDb,
Alain Carpentier, MD, PhDa
a Laboratoire d’Etude des Greffes et Prothèses Cardiaques, Hôpital Européen Georges Pompidou, Paris, France
b Laboratoire de Biochimie A, Hôpital Necker, Paris, France
Address reprint requests to Dr Shen, Laboratoire d’Etude des Greffes et Prothèses Cardiaques, Hôpital Broussais, 96 rue Didot, 75014 Paris, France
e-mail: ming.shen{at}brs.ap-hop-paris.fr
Presented at the VIII International Symposium on Cardiac Bioprostheses, Cancun, Mexico, Nov 3–5, 2000.
 |
Abstract
|
|---|
Background. Lipids play a significant role in the process of calcification of bioprostheses. We assessed whether lipid extraction by ethanol, ether, or a surfactant could mitigate calcification of glutaraldehyde-treated bioprostheses.
Methods. On 200 bovine pericardium samples pretreated with 0.6% glutaraldehyde, lipid extraction was carried out by ethanol, ether, or the tween 80 surfactant, and combinations thereof. The treated tissues were implanted subcutaneously in 50 juvenile rats for 4 and 6 months. Lipids were analyzed by Fourier transform infrared spectrophotometer and chromatography before implantation. Calcium content of implanted tissues was assessed by atomic absorption spectrometer.
Results. Ethanol, ether, or surfactant did mitigate calcification. The most efficient pretreatments were the combination of ethanol and surfactant (calcium content: 15.5 ± 6.8 µg/mg dry tissue after 6 months implantation) or the combination of ethanol, ether, and surfactant (13.1 ± 6.2 µg/mg dry tissue) when compared with surfactant alone (42.9 ± 12.7 µg/mg dry tissue).
Conclusions. Ethanol or the combination of ethanol and ether added to the currently used glutaraldehyde-surfactant treatment further mitigates calcification.
|
Introduction
|
|---|
Since the first clinical implantation of a glutaraldehyde-treated porcine valve [1], excellent results have been reported with a low incidence of thromboembolism in the absence of anticoagulation. In children, however, calcification was a predominant cause of early valve failure [2]. In previous studies we showed that the process of calcification was associated with the adsorption of proteins and phospholipids [3, 4] and Vyavahare and colleagues [5] showed that ethanol preincubation of glutaraldehyde-treated porcine aortic valve bioprostheses may mitigate calcification. In the currently used Carpentier–Edwards pericardial valve, ethanol is used as well as a surfactant to mitigate calcification [6]. But we wondered whether a higher concentration of ethanol or the combination of ethanol and ether could further mitigate calcification by a more pronounced extraction of the lipids.
|
Material and methods
|
|---|
Tissue preparation and implantation
Lipid extraction was carried out on 200 bovine pericardium samples (8 mm diameter) pretreated in 0.625% glutaraldehyde (Glut) for 18 months. The Glut solution was prepared using a standard 25% commercially available solution (Merck, Darmstadt, Germany) in 20 mmol/L phosphate buffer, pH 7.4, according to our original technique [1].
These 200 samples were separated into the following five groups (40 samples each): Glut/surfactant (ST); Glut/ethanol; Glut/ethanol/ST; Glut/ethanol/ether; and Glut/ethanol/ether/ST.
Ethanol extraction was achieved by exposing the samples to three increasing concentrations of 50%, 75%, and 100% for 10 minutes each. The 50% and 75% concentrations were obtained by dilution in 20 mmol/L phosphate buffer, pH 7.4.
Ether extraction was performed for 4 hours at room temperature with agitation after ethanol extraction, and then the samples were placed immediately again in ethanol of three decreasing concentrations of 100%, 75%, and 50% successively, for 10 minutes each. These tissue samples were then transferred to glutaraldehyde.
The surfactant treatment of bovine pericardium was carried out using the following method: samples were incubated in tween 80 solution for 10 hours at 37°C [6] and then stored in glutaraldehyde.
Tissue samples were implanted in 50 12-day-old Wistar rats (CERJ, Genest-Saint-Isle, France). The animals were anesthetized using 100% diethyl ether gas. Four pieces of bovine pericardium were then placed in subcutaneous pouches over the backs of the animals. Five groups of 5 rats each were selected randomly after 4 and 6 months postimplantation and sacrificed with lethal injection of intraperitoneal pentobarbital.
Lipid composition analysis
Lipid composition of nonimplanted bovine pericardium was determined after extraction according to the method of Folch by thin layer chromatography (TLC) and Fourier transform infrared spectrophotometer (IRS) (Vector 22, Bruker Spectrospin, Wissembourg, France). Quantitative assessment of the main constituents was performed using TLC. Fractioning was realized on silicate plates containing a fluorescent indicator (Kieselgel F254 precoated, 10 x 20 cm, 0.25 mm thickness from Merck, Darmstadt, Germany).
The deposit of about 200 µL on a width of 2 cm was done onto silicate plates using a Hamilton microsyringe. The identification of lipidic fractions occurred in few minutes after phosphomolybdic acid reagent (Sigma Aldrich, Saint Quentin-Fallavier, France) and revealed the different fractions according to their increasing Rf depending on the solvent used. Lipid masses were determined for the major classes of the neutral lipids and of the phospholipids.
Fractioning was performed as described above except for thickness of the plates, which was 0.5 mm. Migration and revelation were obtained likewise.
To complement the TLC study, the lipid extracts were analyzed by Fourier transform IRS covering the range of 2.5 to 25 µm (4,000 to 400 cm-1), using the pellet technique in potassium bromide [4].
Each spectrum was the mean of 32 successive scans. To improve the separation between close infrared peaks, the second derivative was calculated for each spectrum. In the second derivative mode, the peaks corresponded to the minimal intensities. Bioprostheses spectra were normalized to allow easier comparisons in relative peak intensities. Ordinate scale was in arbitrary units.
Infrared spectra of various compounds belonging to different classes of lipids were recorded as references and their second derivative was calculated. The references used were as follows: cholesterol, cholesterol esters, triglycerides (C10 to C22), diglycerides, monoglycerides, fatty acids, and phospholipids.
Calcium content analysis
The calcium content of bovine pericardium explants was determined by atomic absorption spectrometer (Spectra AA, Varian, Melbourne, Australia) after digestion in 70% nitric acid (Merck, Nogent Sur Marne, France). Concentrations were expressed as micrograms per milligram of dry tissue.
The results of calcium content were expressed as mean ± standard error of the mean. Analysis of calcium content was carried out by a two-factor analysis of variance. The significance level was set at p less than 0.05.
|
Results
|
|---|
Lipid extraction
Thin layer chromatography analysis of tissue before implantation is summarized in Figure 1. Triglycerides and cholesterol were present in the control group (Glut/ST). In groups treated by ethanol or ethanol and surfactant, cholesterol was absent and triglycerides were diminished. In the groups treated by ethanol/ether and ethanol/ether/surfactant, both cholesterol and triglycerides were absent.
View larger version (148K):
[in this window]
[in a new window]
|
Fig 1. Thin layer chromatography results: the first two columns represent the triglyceride and the cholesterol standard. The third column represents group 1 (the control group), which demonstrates the presence of triglyceride and cholesterol. The sample of group 3 (ethanol and surfactant treatment) shows only a small quantity of triglyceride and the sample of group 5 (ethanol, ether, and surfactant treatment) shows the complete disappearance of triglyceride and cholesterol.
|
|
The IRS confirmed the significant decrease in cholesterol content in groups 2 to 5. This removal of cholesterol was especially observed on the second derivative spectrum (Fig 2), which revealed a decrease in the cholesterol peak at 1,057 cm-1 in groups 2 and 3 and a complete disappearance in groups 4 and 5. Another point of interest was the presence of a small peak at 800 cm-1 (Fig 2), which corresponds to cholesterol structure and suggests some residual cholesterol esters in all samples. In parallel, we observed two significant changes in the derivative spectrum. First, a shift of the carbonyl band from 1,746 cm-1 in groups 1, 2, and 3 to 1,736 cm-1 for groups 4 and 5; second, an increased intensity of the peak at 2,958 cm-1 in comparison with the peak intensity at 2,925 cm-1 (Fig 3). The relative lengthening of the peak at 2,958 cm-1 suggests a less effective removal of glycerides with short (CH2)n chain such as lauryl or myristyl glycerides in comparison with more lipophilic glycerides such as tripalmitin or tristearin by treatments 1, 2, or 3. These lipids were better extracted by treatments 4 and 5. On the other hand, derivative spectra of groups 4 and 5 presented small peaks around 1,250 cm-1, which correspond to phospholipids’ vibrations, suggesting that these compounds were not completely removed during the ethanol delipidation procedures (groups 2 and 3) and were more effectively removed by the ether delipidation procedures (groups 4 and 5). They were more visible on the spectra recorded from groups 4 and 5 because other lipids groups, namely triglycerides with long chains and cholesterol, were effectively removed during the first stages of the treatment.
View larger version (16K):
[in this window]
[in a new window]
|
Fig 2. The second derivative spectra revealed a decrease in the cholesterol peak at 1,057 cm-1 for groups 2 and 3 and its complete disappearance for groups 4 and 5 (arrows). The presence of a small peak at 800 cm-1, which corresponds to cholesterol structure and is suggestive of some residual cholesterol esters, appears in all samples. (Group 1 = glutaraldehyde [Glut]/surfactant [ST]; Group 2 = Glut/ethanol; Group 3 = Glut/ethanol/ST; Group 4 = Glut/ethanol/ether; and Group 5 = Glut/ethanol/ether/ST.)
|
|
View larger version (21K):
[in this window]
[in a new window]
|
Fig 3. The relative lengthening of the peak at 2,958 cm-1 could suggest a less effective removal of glycerides with short (CH2)n chains as in lauryl or myristyl glycerides, in comparison with more lipophilic glycerides such as tripalmitin or tristearin by treatment 1, 2, or 3, these lipids being better extracted by procedures 4 and 5. (Group 1 = glutaraldehyde [Glut]/surfactant; Group 2 = Glut/ethanol; Group 4 = Glut/ethanol/ether.)
|
|
Calcium content
The results of calcium content are shown in Figure 4. After 4 months of implantation, the calcium content significantly decreased in the samples of all treatment groups compared with the control group. After 6 months of implantation, the calcium level of the groups treated with ethanol or ether with surfactant treatment was significantly less than the control group.
View larger version (19K):
[in this window]
[in a new window]
|
Fig 4. After 4 months (4m) of implantation, the calcium content is less in all treatment groups compared with the control group. After 6 months (6m), calcium mitigation is significant only in groups treated with surfactant (ST). In those groups, ethanol and ether enhance calcium mitigation. (Group 1 = glutaraldehyde [Glut]/surfactant [ST]; Group 2 = Glut/ethanol; Group 3 = Glut/ethanol/ST; Group 4 = Glut/ethanol/ether; and Group 5 = Glut/ethanol/ether/ST.)
|
|
|
Comment
|
|---|
Since the first clinical implantation of a glutaraldehyde-pretreated porcine valve [1], various techniques have been used to minimize calcification. The use of monovalent, divalent, or trivalent cations during the fixation in glutaraldehyde did mitigate calcification, but their action was not durable [2]. Hydroxybisphosphonates or aminobisphosphonates or metallic cations (aluminum) have been proposed to prevent calcification of bioprosthetic valve tissue. However, the required doses of these drugs led to stunted growth by interfering with physiologic calcification [2].
More recently, 
-amino oleic acid and ethanol have been investigated for preventing valvular bioprostheses calcification [5, 7]. The results of these anticalcification agents are encouraging in the short term (2 to 3 months of the implantation), but a duration of implantation of 4 and 6 months is necessary to conclude that a treatment is efficient [6]. Other investigators have studied currently used glutaraldehyde-surfactant treatment comparing it with so-called glutaraldehyde control treatment used experimentally. Quintero and coworkers [8] demonstrated that glutaraldehyde control treatment is not equivalent to the treatment actually used for commercial production, which comprises glutaraldehyde fixation plus surfactant-ethanol treatment [6]. This treatment has an anticalcification effect after 2 to 3 months of subcutaneous implantation in 12-day-old rats and even longer in 21-day-old rats. This finding is why in this study the glutaraldehyde-surfactant treatment and not glutaraldehyde alone was used as the control group.
After 4 months of implantation, the calcium content of all four groups with anticalcification treatment was significantly lower than in the control group. After 6 months of implantation, only in the groups of ethanol with surfactant and ether with surfactant was the calcium level significantly lower than in the control group. In previous studies [3, 4], we showed that phospholipids play an important role in the process of calcification. Thin lay chromatography analysis showed that extraction by ethanol does not completely eliminate triglycerides in bovine pericardium, whereas extraction by ether totally removed triglycerides. The calcium content of these two groups, however, was not significantly different after 6 months of implantation.
In conclusion, treatments by ethanol or ether alone or surfactant alone are less efficient than the combination of these treatments. The fact that the combination of treatments is more efficient than any of the single treatments tends to prove that the mechanism of extraction and the products extracted by each treatment are slightly different. As a practical conclusion of this work it could be said that in the currently used procedure of bioprosthetic valve preservation, it might be beneficial to increase the concentration of ethanol or its length of incubation or to add ether treatment to the surfactant treatment. However, structural changes and increased stiffness of the tissue resulting from these treatments should be taken into consideration.
|
Acknowledgments
|
|---|
We thank Bernard Martinet and Martine Rancic for their technical help.
|
References
|
|---|
-
Carpentier A., Lemaigre G., Robert L., Carpentier S., Dubost C. Biological factors affecting long-term results of valvular hetero-grafts. J Thorac Cardiovasc Surg 1969;58:467-483.[Medline]
-
Schoen F.J., Harasaki H., Kim K.M., Andreson H.C., Levy R.J. Biomaterial-associated calcification: pathology, mechanisms, and strategies for prevention. J Biomed Mater Res 1988;22:11-36.[Medline]
-
Shen M., Marie P., Farge D., et al. Osteopontin is associated with bioprosthetic heart valve calcification in humans. C R Acad Sci III 1997;320:49-57.[Medline]
-
Shen M., Lajos P.S., Farge D., et al. Infrared spectroscopy in the evaluation of the process of calcification of valvular bioprostheses. Ann Thorac Surg 1998;66:S236-S239.
-
Vyavahare N., Hirsch D., Lerner E., et al. Prevention of bioprosthetic heart valve calcification by ethanol preincubation: efficacy and mechanisms. Circulation 1997;95:479-488.[Abstract/Free Full Text]
-
Carpentier A., Nashef A., Carpentier S., Ahmed A., Goussef N. Techniques for prevention of calcification of valvular bioprostheses. Circulation 1984;70(Suppl I):165-168.[Abstract/Free Full Text]
-
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]
-
Quintero L.J., Lohre J.M., Hernandez N., et al. Evaluation of in vivo models for studying calcification behavior of commercially available bovine pericardium. J Heart Valve Dis 1998;7:262-267.[Medline]
Top
Abstract
Introduction
