Calcification Resistance With Aluminum-Ethanol Treated Porcine Aortic Valve Bioprostheses in Juvenile Sheep

Transcription

1 Calcification Resistance With Aluminum-Ethanol Treated Porcine Aortic Valve Bioprostheses in Juvenile Sheep Matthew F. Ogle, MS, Sheila J. Kelly, BS, Richard W. Bianco, BS, and Robert J. Levy, MD Heart Valve Division, St. Jude Medical Inc, St. Paul, Minnesota, Department of Experimental Surgery, University of Minnesota, Minneapolis, Minnesota, and Division of Cardiology, Children s Hospital of Philadelphia, Philadelphia, Pennsylvania Background. Calcification of glutaraldehyde fixed bioprosthetic heart valve replacements frequently leads to the clinical failure of these devices. Previous research by our group has demonstrated that ethanol pretreatment prevents bioprosthetic cusp calcification, but not aortic wall calcification. We have also shown that aluminum chloride pretreatment prevents bioprosthetic aortic wall calcification. This study evaluated the combined use of aluminum and ethanol to prevent both bioprosthetic porcine aortic valve cusp and aortic wall calcification in rat subcutaneous implants, and the juvenile sheep mitral valve replacement model. Methods. Glutaraldehyde fixed cusps and aortic wall samples were pretreated sequentially first with aluminum chloride (AlCl 3 ) followed by ethanol pretreatment. These samples were then implanted subdermally in rats with explants at 21 and 63 days. Stent mounted bioprostheses were prepared either sequentially as previously described or differentially with AlCl 3 exposure restricted to the aortic wall followed by ethanol pretreatment. Mitral valve replacements were carried out in juvenile sheep with elective retrievals at 90 days. Results. Rat subdermal explants demonstrated that sequential exposure to AlCl 3 and ethanol completely inhibited bioprosthetic cusp and aortic wall calcification compared with controls. However the sheep results were markedly different. The differential sheep explant group exhibited very low levels of cusp and wall calcium. The glutaraldehyde group exhibited little cusp calcification, but prominent aortic wall calcification. All sheep in the two groups previously described lived to term without evidence of valvular dysfunction. In contrast, animals in the sequential group exhibited increased levels of cusp calcification. None of the animals in this group survived to term. Pathologic analysis of the valves in the sequential group determined that valve failure was caused by calcification and stenosis of the aortic cusps. Conclusions. The results clearly demonstrate that a combination of aluminum and ethanol reduced aortic wall calcification and prevented cuspal calcification. Furthermore, this study demonstrates that exclusion of aluminum from the cusp eliminated the cuspal calcification seen when aluminum and ethanol treatments were administered in a sequential manner. (Ann Thorac Surg 2003;75: ) 2003 by The Society of Thoracic Surgeons Stentless bioprosthetic tissue heart valves have become recognized as an important therapeutic option in the surgical treatment of heart valve disease [1 3]. However, clinically available xenografts [1, 2] are at risk to undergo dystrophic calcification of the aortic wall. This calcification may affect the long-term function of these devices and the outcome of reoperation. Previous studies [3, 4] have demonstrated that ethanol pretreatment inhibits calcification of pretreated glutaraldehyde porcine aortic valve cusps. The mechanism of ethanol inhibition of dystrophic calcification is believed to be the action of ethanol extracting lipids and altering collagen structure of the bioprosthetic material [5, 6]. However, ethanol pretreatment of bioprosthetic valve constructs has little to no inhibitory effect on aortic wall Accepted for publication Sept 16, Address reprint requests to Dr Levy, Children s Hospital of Philadelphia, Abramson Pediatric Research Center, Suite 702, 3516 Civic Center Blvd, Philadelphia, PA 19104; levyr@ .chop.edu. calcification [7]. Aluminum chloride (AlCl 3 ) pretreatment has been demonstrated to inhibit aortic wall calcification [8, 9]. The mechanisms of this inhibition are related to prevention of elastin calcification and reduction of regional alkaline phosphatase activity [10, 11]. In the present study we investigated the combined pretreatment of ethanol and aluminum chloride for the prevention of both cuspal and aortic wall mineralization in porcine aortic valve bioprosthetic implants. Furthermore, prior research by our group [12] and the research of others [13] suggested that AlCl 3 cusp exposure could actually exacerbate leaflet calcification in the blood stream, but not in subdermal implants. Thus we hypothesized that the combined use of ethanol and AlCl 3 could Doctors Ogle and Kelly disclose that they have a financial relationship with St. Jude Medical, Inc, St. Paul, MN by The Society of Thoracic Surgeons /03/$30.00 Published by Elsevier Science Inc PII S (02)

2 1268 OGLE ET AL Ann Thorac Surg ALUMINUM TREATED VALVES IN SHEEP 2003;75: inhibit both bioprosthetic cusp and aortic wall calcification. However, rat subdermal implants may not reveal evidence of an adverse cuspal outcome because of the AlCl 3. Therefore we sought to compare rat subdermal and sheep circulatory model system results. To this end we systematically developed novel pretreatment conditions for the combined use of ethanol and AlCl 3, including the sequential treatment of the entire bioprosthetic valve(alcl 3 first, then ethanol) compared with a differential exposure (AlCl 3 exposed only to the aortic wall). Assessment of inhibition of bioprosthetic calcification was the key endpoint of interest in these studies. Material and Methods In Vitro Aluminum Deposition Porcine aortic valves were dissected and the cusp tissue fixed with buffered (0.5%) glutaraldehyde. Three sample groups were created. Group I contained six control cusps with no aluminum treatment. Group II contained six cusps incubated for 2 hours in 0.1 mol/l aluminum chloride hexahydrate, ph 7.4 (Sigma Chemical Co, St. Louis, MO). Group III contained six cusps incubated for 2 hours in 0.1 mol/l aluminum chloride hexahydrate, ph 3 (Sigma Chemical Co, St. Louis, MO). All sets of samples were rinsed once in 10 ml of HEPES (0.05 mol/l, ph 7.4) buffered saline. Samples were then bisected, and one half of each specimen was processed for histology and stained for aluminum using aluminum stain (Histoserv, Inc, Gatherburg MD), and the other half of each specimen was analyzed using inductively coupled plasmaatomic emission spectroscopy (ICP-AES) (Thermal Jarrell Ash, Franklin, MA). Implant Tissue Preparation Three groups of glutaraldehyde fixed (low pressure fixation, 2 to 4mm Hg) bioprosthetic valves were prepared. Group I consisted of glutaraldehyde fixed porcine aortic valves as previously described [14]. Group II consisted of glutaraldehyde fixed porcine aortic valves subjected to sequential aluminum and ethanol treatment; specifically these valves were first incubated in a ph 3 aluminum chloride solution (0.1 mol/l), then in a buffered (0.05 mol/l HEPES, ph 7.4) ethanol solution (80%). Group III consisted of glutaraldehyde fixed porcine aortic valves subjected to differential aluminum and ethanol treatment; specifically these valve walls were treated exclusively with aluminum chloride (0.1 mol/l) followed by a total valve incubation in a buffered ethanol solution (80%). A process was developed by using the fact that aluminum in solution is a weak acid, and when aluminum is adjusted to ph 7 it precipitates as aluminum hydroxide. While the root tissue is incubated in aluminum solution, the cusps are bathed with a ph 7.0 buffered solution which prevents association of aluminum with the valve cusps. All valves in all groups were fabricated on a prototype stent for sheep implants or were dissected for subcutaneous rat implants. Rat Study Rat studies were performed in compliance with the animal care committee guidelines for small animal operations. Three-week-old, male Sprague Dawley rats were used. They received a preoperative injection of ketamine hydrochloride 85 mg/kg by intraperitoneal injection. Samples in each treatment group were color-coded with surgical suture before implant. An incision was made in the back of each rat and six samples, randomly selected from each of three treatment groups (previously described) were implanted into individual subcutaneous pockets. At the termination of the study (21 and 63 days) animals were euthanized with intraperitoneal injection of Beuthanasia solution (Schering Plough, Union, NJ), 86 mg/kg mixed 1:1 with normal saline. Explanted samples were placed in 0.9% saline for storage before analysis. Each tissue sample was removed from the saline and sectioned in half. One half of the tissue sample was cleaned of host tissue and used for elemental analysis. The second half of the tissue was placed in 10% neutral buffered formalin and stored for histologic examination. Sheep Study Animal acquisition and preoperative and postoperative evaluations were conducted using protocols previously described [3]. Briefly, 18 sheep (Ovis aries), of either gender and between ages 3 and 5 months were used. An operation was performed under general anesthesia with cardiopulmonary bypass as previously published [3]. All animals were cared for at Experimental Surgery Services, University of Minnesota (Minneapolis, MN), an American Association of Laboratory Animal Care (AALAC) accredited facility. Serum studies were performed preoperatively and immediately before sacrifice at 90 days. The gross pathology was performed at Experimental Surgery Services. Under general anesthesia, the chest was entered through a fourth intercostal incision. The heart was entered through the left atrium. The leaflets of the native mitral valve (including the entire chordae apparatus) were removed and the bioprosthetic valve was placed in the mitral annulus. The valve was sutured with interrupted 3-0 braided polyester sutures in an inverted mattress pattern. All animals received broad-spectrum antibiotic prophylaxis, analgesia, and diuretics. Animals surviving less than 48 hours postoperatively were considered technical failures and were not included in the study. Only interoperative deaths were excluded from the result analysis. Early deaths that appeared to be related to calcific stenosis were included. Animals surviving more than 48 hours, but dying before the scheduled 90 day study period, were autopsied using the study protocol, and were included in all analyses. The animals were euthanized at 90 days, and total autopsy including evaluation of the brain, lungs, liver, spleen, and kidneys was performed. Macroscopic evaluation and photography of the heart and bioprosthetic valve was performed before fixation in 10% neutral buffered formalin. Portions of all organs were submitted

3 Ann Thorac Surg OGLE ET AL 2003;75: ALUMINUM TREATED VALVES IN SHEEP 1269 for paraffin embedding. Animal care complied with the Principles of Laboratory Animal Care and the Guide for the Care and Use of Laboratory Animals (National Institutes of Health Publication No , 1985). Photographs were made of each excised valve and surrounding tissue. A radiograph of each excised valve was made at 65 kv for 30 seconds. The valve was then removed from the stenting material. One section of each valve leaflet and supporting aortic wall tissue was submitted for histologic preparation at American HistoLabs, Inc (Gaithersburg, MD). Another section of each leaflet with wall tissue was submitted for elemental analysis at St. Jude Medical, Inc. Histology The degree and character of the general architectural features and the extent of inflammation were evaluated using the hematoxylin-eosin stain. The aluminum associated with the tissue was evaluated with aluminum stain. The degree of calcification was evaluated using the von Kossa stain. The character and extent of the fibrous healing response and the extent of fibrin accumulation was evaluated using the Movat s Pentachrome stain. A calcification grading scale was developed and applied in a blinded manner to all specimens with 0 no calcium staining identifiable, 1 rare granules of calcification, 2 nodular collections of calcium less than 25% surface area, 3 extensive collections of calcium 25% to 50%, and 4 near total calcification more than 50%. Elemental Analysis Explanted valve (cusp and aortic wall) tissue sections were cleaned of host tissue and washed with saline followed by distilled water, and then lyophilized and weighed. Each tissue sample was hydrolyzed in nitric acid and diluted to a standard volume. Elemental analysis was performed using inductively coupled plasmaatomic emission spectroscopy. Elemental concentrations of calcium and phosphorus were reported in mg per gram of dry tissue. Statistical Analysis One way analysis of variance was used to determine statistical significance between the results of each test group, cusp, and aortic wall. Subsequent differences were evaluated using either the Bonferroni or Tukey posthoc analysis. Linear regression was used to determine a correlation between rat and sheep study ICP-AES results and to determine a correlation between the sheep ICP-AES results and the histologic scale. Results In Vitro Aluminum Deposition This brief study sought to investigate the maximum loading potential of aluminum into bioprosthetic tissue for both acidic ( ph 3) and physiologic (ph 7.4) conditions. These experiments were necessary in order to establish the optimal incubation conditions for AlCl 3 Fig 1. Elemental analysis of cusp tissue after incubation with aluminum chloride solution. Note the high aluminum incorporation levels at acidic ph compared with those at physiologic ph that were barely detectable. Untreated control cusp and aortic wall were below the limits of detection (0.12 ppm). pretreatment of bioprostheses, and to assess aluminum loading levels with respect to potential systemic exposure. The results indicate that very little aluminum (0.13 mg/g in the cusps) was associated with the tissue under physiologic conditions (Fig 1). The limited aluminum association at ph 7.4 is likely due to the relative insolubility of aluminum salts at physiologic ph [15]. Under the acidic incubation conditions there was significantly more aluminum associated with the bioprosthetic tissue (7.56 mg/g aluminum). Histology results using aluminum staining to detect aluminum deposition (Fig 2) demonstrates a diffuse deposition of aluminum in the tissue primarily in or around cell membranes and structural proteins. In Vivo Rat Evaluation Both cusp and aortic wall samples treated sequentially with the combination of aluminum and ethanol demonstrated significant reduction in calcification versus the glutaraldehyde control, which demonstrated severe calcification of both wall and cusp tissue (Table 1). Furthermore, the host response to these aluminum and ethanol treated samples showed no additional inflammation or abnormal healing when compared with the glutaraldehyde-fixed control implants. In Vivo Sheep Explant Evaluation The calcification of all valves was evaluated by both a histopathologic grading scale and elemental analysis (Figs 3, 4). Groups I (glutaraldehyde control) and III (differential ethanol and AlCl 3 ) experienced no mortality or adverse valvular events for the 90-day duration of the study (Fig 5). However, group II (sequential ethanol and AlCl 3 ) experienced 100% mortality by 45 days, caused by dystrophic calcification of the bioprosthetic valve cusps. The control group I exhibited low cuspal calcification, but high levels of aortic wall calcification. The differential group III exhibited significantly lower levels of both cusp and wall calcification (p 0.001). In contrast, the sequen-

4 1270 OGLE ET AL Ann Thorac Surg ALUMINUM TREATED VALVES IN SHEEP 2003;75: Fig 3. Ninety-day sheep study calcium histopathologic grading. See Methods for conditions used for group I (control), group II (sequential AlCl 3 and ethanol), and group III (differential AlCl 3 and ethanol). Note grading for groups I and III were zero for cusp tissue, and thus no bar graphs are shown. Fig 2. Representative histomicrographs of cusp tissue (stained with aluminum stain 100) incubated in either 0.1 mol/l aluminum chloride at ph 3 (A) or at ph 7.4 (B). tial group II exhibited a statistically significant increase in cuspal calcification (p 0.001). Histopathologic examination of the valves from the 90-day sheep study found that the greatest degree of calcification was in the control group I, followed by the sequential ethanol and aluminum treated group 2. Calcification of the control valves (group I) was localized primarily in the aortic wall (Fig 4). Calcification of sequentially treated valves (group II) was found in both cusp and wall tissue. For all study groups, inflammation was limited to the interface between the valve and host tissue. Radiographs and histomicrographs provided additional information regarding the distribution of calcium throughout the bioprosthetic tissue (Fig 6). Severe aortic wall calcification was seen in group I. The low level of wall calcification seen in group II was limited to the cusp and wall interface at the line of leaflet attachment. No calcification was seen in group III, which received the differential ethanol and AlCl 3 pretreatment. There was a correlation noted between the actual calcium content of both cusps and aortic wall determined by ICP-AES and the histologic scale used to assess the extent of calcification, r (cusp) and r 0.941(aortic wall). In all groups no significant differences were observed between hematologic and blood chemistry profiles preoperatively and at termination. For all groups, no abnormal pathologic findings were observed in any of the organ samples. Comment This study showed that a combination of ethanol and AlCl 3 can be used to effectively prevent both aortic cusp and wall calcification. The novel approach of excluding aluminum from the cusp tissue using a buffer system eliminated the severe cuspal calcification seen in our sheep studies when ethanol and AlCl 3 were used sequentially. These results differed significantly from our Table 1. Rat Study Calcium Content (mg per Gram of Dry Tissue) TEST Group Cusp Aortic Wall 21 days 63 Days 21 Days 63 Days Pre-implant NA NA Glutaraldehyde controls (group I) Aluminum/ethanol sequential (group II) For the study sample size, n 10. NA not available.

5 Ann Thorac Surg OGLE ET AL 2003;75: ALUMINUM TREATED VALVES IN SHEEP 1271 Fig 4. Ninety-day sheep study calcium content (mg/g of dry tissue). See Methods for description of group I (control), group II (sequential AlCl 3 and ethanol), and group III (differential AlCl 3 and ethanol). Note preimplant levels and all cusp values except Group II were barely detectable. sequentially prepared rat subdermal implants that did not demonstrate cusp calcification associated with AlCl 3 pretreatment. Other research by our group [12] has demonstrated that this effect is most likely caused by the very low affinity of aluminum binding to cuspal structural proteins compared with high affinity binding to the elastin enriched aortic wall extracellular matrix; the net result of this is dissociation of cuspal aluminum with circulatory washout of loosely bound aluminum leading to aluminum precipitates with carbonate or phosphate, or both, and subsequent mineral deposition. It is widely accepted that stentless aortic valves have better hydrodynamic characteristics compared with their stented counterparts and more closely approximate the native hemodynamic characteristics of the aortic valves [15 17]. The advantage appears in part to relate to the maintained compliance of the prosthetic tissue [18]. Although low pressure fixation is often used to prepare bioprosthesis, Fig 5. Ninety-day sheep study percent survival. See Methods for pretreatments used in group I (control), group II (sequential AlCl 3 and ethanol), and group III (differential AlCl 3 and ethanol). Note group I and group III lines are coincident. there is no evidence this helps to prevent bioprosthetic calcification. Thus currently available stentless valves still undergo mineralization [19 21]. Mineralization may affect the long-term function of these devices, to the point where surgical replacement of the valve is necessary. To address this problem, several groups have proposed anticalcification pretreatments [22 24]. These methodologies appear to inhibit bioprosthetic cusp calcification. To date, none of these treatments have been shown to be effective in preventing aortic wall calcification, thus underlining the importance of the findings described here. It should be noted that although our model systems evaluated bioprosthetic cusp and aortic wall calcification in a stent mounted configuration, AlCl 3 inhibition of aortic wall calcification has been demonstrated in nonstented glutaraldehyde-fixed allografts in circulatory implants [25], thus providing further validation of the observed efficacy for preventing aortic wall calcification in the present studies. Furthermore the technical difficulty of placing a stentless bioprostheses in the aortic position in the juvenile sheep precludes a calcification study with a directly comparable clinical configuration. The anticalcific effects of aluminum and other metal cations on bioprosthetic tissue has been the subject of a number of studies over the past 10 years [8 11]. The mechanism of ethanol inhibition of cuspal calcification has been investigated in a number of studies that have demonstrated prevention of mineralization due to alterations in collagen structure and extraction of cuspal lipids [3 7]. Pretreatment with ethanol is now used clinically on St. Jude Medical (St. Paul, MN) bioprosthetic heart valves as an anticalcification treatment. Until now, the ability to combine the beneficial effects of ethanol and AlCl 3 pre treatments in a functioning preclinical model has not been reported. However, our group observed that incubation of cuspal tissue with aluminum chloride induced accelerated calcification. Aortic wall tissue did not calcify significantly compared with controls under similar conditions. There is no obvious explanation for the accelerated cuspal calcification observed in the sequential group in the sheep model. It is highly likely that this is due to some aspect of AlCl 3 exposure, because previous studies have shown that ethanol pretreatment significantly inhibits cuspal calcification, both in subdermal implants and in circulatory experiments [3] Ongoing experiments indicate that although AlCl 3 pretreatment of elastin, a major component of aortic wall, results in irreversible binding of aluminum, comparable AlCl 3 incubations of type I collagen, the most abundant collagen in porcine cusps, results in unstable binding of aluminum [12]. Thus, the leaching of Al 3 ions from collagen binding sites into the cuspal interstitium could lead to microprecipitates. Aluminum phosphate or carbonate microprecipitates could hypothetically damage the bioprosthetic cusps directly, or the formation of aluminum carbonates or phosphates could provide a nidus for dystrophic calcification. Furthermore, since rat subdermal implants demonstrated inhibition of cuspal calcification with sequential ethanol and AlCl 3 exposure, it can be concluded

6 1272 OGLE ET AL Ann Thorac Surg ALUMINUM TREATED VALVES IN SHEEP 2003;75: Fig 6. Roentgenogram analysis and histomicrographs showing calcium distribution in bioprosthetic sheep explants. (A) Group I (control) radiograph. (B) Group I aortic wall (Von Kossa stain, original magnification 100). (C) Group I aortic cusp (Von Kossa stain, original magnification 200). (D) Group II (sequential AlCl 3 and ethanol) radiograph. (E) Group II aortic wall (Von Kossa stain, original magnification 100). (F) Group II aortic cusp (Von Kossa stain, original magnification 200). (G) Group III (differential AlCl 3 and ethanol) roentgenogram. (H) Group III aortic wall (Von Kossa stain, original magnification 100) (I) Group III aortic cusp (Von Kossa stain, original magnification 200). that some aspect of blood material interactions may be of mechanistic importance in the cuspal calcification in the sequential group. Both ethanol and AlCl 3 pretreatments have been studied individually with demonstrated efficacy for inhibiting calcification of bioprosthetic aortic cusp [4] and the aortic wall [25], respectively, in circulatory implants. Thus the present study investigated the combined use of these pretreatments. We have demonstrated a decrease in calcification of the differentially treated tissues compared with sequential treatment in the sheep model. Of note, calcification of the differentially treated valve cusps was not significantly different than the control valve cusps, neither calcified, possibly because of the short duration of the study (90 days). A prior time course study by our group has shown that 90-day mitral valve bioprosthetic replacements in juvenile sheep have minimal cuspal calcification [26]. Nevertheless, we chose this time point to evaluate both the potential for exacerbated calcification, and the combined beneficial effects of ethanol and AlCl 3 on aortic wall calcification. The next step in the evaluation of the differential aluminum and ethanol treatment will be to

7 Ann Thorac Surg OGLE ET AL 2003;75: ALUMINUM TREATED VALVES IN SHEEP 1273 extend the duration of the study to 150 days to confirm the benefits of this treatment on cuspal tissue. In conclusion, ethanol and AlCl 3 can be used in combination to effectively inhibit both bioprosthetic cuspal and aortic wall calcification in this sheep model. We have shown the importance of a differential pretreatment approach, restricting AlCl 3 from the aortic valve cusps in order to prevent AlCl 3 -related cuspal mineralization. Sequential pretreatment with ethanol and AlCl 3 had a deleterious effect leading to cuspal calcification, but nevertheless resulted in inhibition of aortic wall mineralization. The authors thank Jennifer LeBold for her assistance with manuscript preparation. Research support was provided by St. Jude Medical Inc. Doctor Levy s efforts were supported in part by the National Heart, Lung, and Blood Institute grant HL38118 and by The William J. Rashkind Endowment of the Children s Hospital of Philadelphia. References 1. Schoen FJ, Kujovich JL, Levy RJ, St. John Sutton M. Bioprosthetic valve failure. Cardiovas Clin 1988;18: Chen W, Schoen FJ, Levy RJ. Mechanism of efficacy of 2-amino oleic acid for inhibition of calcification of glutaraldehyde-pretreated porcine bioprosthetic heart valves. Circulation 1994;90: Vyavahare, N, Hirsch, D, Lerner, et al. Prevention of bioprosthetic heart valve calcification by ethanol preincubation. Circulation1997;95: Langanki D, Ogle MF, Cameron JD, Lirtzman RA, Schroeder RF, Mirsch MW. Evaluation of a novel bioprosthetic heart valve incorporating anticalcification and antimicrobial technology in a sheep model. J Heart Valve Dis 1998;7: Vyavahare NR, Jones PL, Hirsch D, Schoen FJ, Levy RJ. Prevention of glutaraldehyde-fixed bioprosthetic heart valve calcification by alcohol pretreatment: further mechanistic studies. J Heart Valve Dis 2000;9: Vyavahare NR, Hirsch D, Lerner E, et al. Prevention of calcification of glutaraldehyde-crosslinked porcine aortic cusps by ethanol preincubation: mechanistic studies of protein structure and water-biomaterial relationships. J Biomed Mater Res 1998;40: Lee CH, Vyavahare N, Zand R, et al. Inhibition of aortic wall calcification in bioprosthetic heart valves by ethanol pretreatment: biochemical and biophysical mechanisms. J Biomed Mater Res 1998;42: Webb CL, Flowers WE, Boyd J, Rosenthal EL, Schoen FJ, Levy RJ. Al binding studies, and metallic cation effects on bioprosthetic heart valve. Calcification in the rat subdermal model. ASAIO 1990;36: Webb CL, Nguyen NM, Schoen FJ, Levy RJ. Calcification of allograft aortic wall in a rat subdermal model. Am J Pathol 1992;141: Levy RJ, Schoen FJ, Flowers WB, Staelin ST. Initiation of mineralization in bioprosthetic heart valves: studies of alkaline phosphatase activity and its inhibition by AlCl3 or FeCl3 preincubations. J Biomed Mater Res 1991;25: Vyavahare N, Ogle M, Schoen FJ, Levy RJ. Elastin calcification and its prevention with aluminum chloride pretreatment. Am J Pathol 1999;155: Levy, RJ, Vyavahare, N, Ogle, M, Ashworth, P, Bianco, R, Schoen, FJ. Inhibition of cups and aortic wall calcification in ethanol and aluminum treated bioprosthetic heart valves in sheep: background, mechanisms, and synergism. J Heart Valve Dis. In press, Carpentier SM, Carpentier AF, Chen L, Shen M, Quintero LJ, Witzel TH. Calcium mitigation in bioprosthetic tissues by iron pretreatment: the challenge of iron leaching. Ann Thorac Surg 1995;60:S Goldman BS, David TE, Del Rizzo DF, Sever J, Bos J. Stentless porcine bioprosthesis for aortic valve replacement. J Cardiovasc Surg 1994;35: McQuarrie, DA, Rock PA. Solubility and precipitation reactions. General Chemistry. New York: W H Freeman and Company, 1987: Sung HW, Le TN, Kingsbury CJ, Quintero LJ, Myeres KE, Quijano RC. In vitro pulsatile flow evlauation of a stentless porcine aortic bioprosthesis. ASAIO J 1995;41: Eberhart CE, Yoganathan AP, Capps M, Toomes C. Hydrodynamic performance of the Medtronic freestyle aortic root bioprosthetic. In: Piwnica A, Westaby S, eds. Stentless Bioprostheses, Oxford, UK, ISIS Media, 1995: Konertz W, Herrmann M, Knauth M, Stabenow I, David T. Preliminary experience with the Toronto SPV stentless porcine bioprosthesis for aortic valve replacement. Thorac Cardiovasc Surg 1994;42: Herijgers P, Ozaki S, Verbeken E, et al. Calcification characteristics of porcine stentless valves in juvenile sheep. Eur J Cardiothorac Surg 1999;15: Dittrich S, Alexi-Meskishvili VV, Yankah AC. Comparison of porcine xenografts and homografts for pulmonary valve replacement in children. Ann Thorac Surg 2000;70: Sundt TM 3rd, Rasmi N, Wong K, Radley-Smith R, Khaghani A, Yacoub MH. Reoperative aortic valve operation after homograft root replacement: surgical options and results. Ann Thorac Surg 1995;60:S Herijgers P, Ozaki S, Verbeken E, et al. The No-React anticalcification treatment: a comparison of Biocor No-React II and Toronto SPV stentless bioprostheses implanted in sheep. Semin Thorac Cardiovasc Surg 1999;11: Chen W, Kim JD, Schoen FJ, Levy RJ. Effect of 2-amino oleic acid exposure conditions on the inhibition of calcification of glutaraldehyde cross-linked porcine aortic valves. J Biomed Mater Res 1994;28: Hirsch D, Drader J, Thomas TJ, Schoen FJ, Levy JT, Levy RJ. Inhibition of calcification of glutaraldehyde pretreated porcine aortic valve cusps with sodium dodecyl sulfate: preincubation and controlled release studies. J Biomed Mater Res 1993;27: Levy RJ, Qu X, Underwood T, Trachy J, Schoen FJ. Calcification of valved aortic allografts in rats: effects of age, crosslinking, and inhibitors. J Biomed Mater Res 1995;29: Schoen FJ, Hirsch D, Bianco RW, Levy RJ. Onset and progression of calcification in porcine aortic bioprosthetic mitral valve replacements in juvenile sheep. J Thorac Cardiovasc Surg 1994;108:880 7.