ARTICLE IN PRESS doi:10.1510/icvts.2008.196220

Interactive CardioVascular and Thoracic Surgery 8 (2009) 553–557 www.icvts.org

ESCVS article - Valves

St Jude Epic heart valve bioprostheses versus native human and porcine aortic valves – comparison of mechanical properties夞,夞夞 Martins Kalejsa,*, Peteris Stradinsa,b, Romans Lacisa,b, Iveta Ozolantab, Janis Pavarsa,b, Vladimir Kasyanovb Center of Cardiac Surgery, Pauls Stradins Clinical University Hospital, 13 Pilsonu Street, LV-1002 Riga, Latvia b Riga Stradins University, 16 Dzirciema Street, LV-1007 Riga, Latvia

a

Received 11 October 2008; received in revised form 9 January 2009; accepted 14 January 2009

Abstract Objectives: The major problem with heart valve bioprostheses made from chemically treated porcine aortic valves is their limited longevity caused by gradual deterioration, which has a causal link with valve tissue mechanical properties. To our best knowledge, there are no published studies on the mechanical properties of modern, commercially available bioprostheses comparing them to native human valves. The objective of this study is to determine the mechanical properties of St Jude Epic bioprostheses and to compare them with native human and porcine aortic valves. Methods: Leaflets from eight porcine aortic valves and six Epic bioprostheses were analyzed using uni-axial tensile tests in radial and circumferential directions. Mechanical properties of human valves have been previously published by our group. Results are represented as mean values"S.D. Results: Circumferential direction. Modulus of elasticity of Epic bioprostheses in circumferential direction at the level of stress 1.0 MPa is 101.99"58.24 MPa, 42.3"4.96 MPa for native porcine and 15.34"3.84 MPa for human aortic valves. Ultimate stress is highest for Epic bioprostheses 5.77"1.94 MPa, human valves have ultimate stress of 1.74"0.29 MPa and porcine 1.58"0.26 MPa. Ultimate strain in circumferential direction is highest for human valves 18.35"7.61% followed by 7.26"0.69% for porcine valves and 5.95"1.54% for Epic bioprostheses. Radial direction. Modulus of elasticity in radial direction is 9.18"1.81 MPa for Epic bioprostheses, 5.33"0.61 MPa for native porcine, and 1.98"0.15 MPa for human aortic valve leaflets. In the radial direction ultimate stress is highest for Epic bioprostheses 0.7"0.21 MPa followed by native porcine valves 0.55"0.11 MPa and 0.32"0.04 MPa for human valves. For human valves ultimate strain is 23.92"4.87%, for native porcine valves 8.57"0.8% and 7.92"1.74% for Epic bioprostheses. Conclusions: Epic bioprostheses have non-linear stress–strain behavior similar to native valve tissue, but they are significantly stiffer and hence less elastic compared to native porcine and human aortic valves. These differences in mechanical properties may cause variations in stress distribution within leaflets of the bioprosthetic valves and accelerate their deterioration. 䊚 2009 Published by European Association for Cardio-Thoracic Surgery. All rights reserved. Keywords: Aortic valve; Mechanical properties; Bioprosthesis

1. Introduction All bioprostheses have one important and serious drawback – limited longevity w1, 2x – within 10 years of implantation 50% to even 60% of all bioprosthetic valve recipients will undergo a repeated surgical intervention caused by valve failure w3x. The major cause of limited longevity is gradual tissue deterioration by calcific and non-calcifying degeneration w1–3x. The biological tissues used in bioprostheses undergo physical and chemical treatment before use. One of the steps in this treatment is fixation with glutaraldehyde which forms cross-links between collagen fibers. Application of this cross-linking agent also modifies the mechanical properties of the biological tissues making them mechanically stronger, stiffer and more rigid w4x. When designing new bioprostheses, mechanical properties of the native human 夞 The work was supported by the Latvian National Research Programme in Medicine 2006–2009. 夞夞 Presented at the 57th International Congress of the European Society for Cardiovascular Surgery, Barcelona, Spain, April 24–27, 2008. *Corresponding author. Tel.: q371 67069221; fax: q371 67069421. E-mail address: [email protected] (M. Kalejs). 䊚 2009 Published by European Association for Cardio-Thoracic Surgery

aortic valve should be treated as an example to follow, this stands true also for prostheses created by tissue engineering or made from polymer materials. It is supported by the fact that changes in mechanical properties of valve leaflets appear to accelerate tissue deterioration (this issue will be discussed more in the discussion section of this article). The goal of this study is to determine the mechanical properties of St Jude Epic bioprostheses and to compare them with native human and porcine aortic valves. 2. Materials and methods For comparison with the native aortic human and porcine valves we used St Jude Medical xenoartic bioprosthesis Epic (St Jude Medical, Minnesota, USA) which is created from porcine aortic valve leaflets mounted on a polymer material stent. We chose this bioprosthetic valve because it has a long application history of close to 20 years w5, 6x (the design of Epic valves is borrowed from St Jude Biocor), and because it is one of the most widely used xenoaortic bioprostheses in the world and in our hospital. For the mechanical studies we used valve leaflets harvested from eight porcine hearts within 24 h after the

ARTICLE IN PRESS 554

M. Kalejs et al. / Interactive CardioVascular and Thoracic Surgery 8 (2009) 553–557

human aortic valves and the methods used have been published previously w7x. Experimental data were analyzed using SPSS for Windows 16.0 (SPSS Inc, Chicago, USA). Statistical significance among the mechanical properties of the three tested materials was evaluated using single-parameter ANOVA and the appropriate post-hoc tests. Statistical significance of differences between values was defined as having P-0.05. All measurement values are shown as mean values"S.D. 3. Results 3.1. Circumferential direction Fig. 1. A scheme depicting the cutting directions of specimens from aortic valve leaflets.

animals’ death and from six St Jude Epic bioprostheses which all were within 1–2 weeks due to their expiry date. We determined their mechanical properties using uni-axial tensile tests with a universal testing machine ZwickyRoell BDO-FB0.5TS (Zwick GmbH & Co, Ulm, Germany) equipped with test Xpert software. The tested valve leaflet material was cut into 3.5 mm wide and 20 mm long specimens in circumferential and radial directions (Fig. 1). The thickness of all leaflets was measured using a cathetometer MK-6 (LOMO, Saint Petersburg, Russia) with a precision of "0.01 mm. Uni-axial tensile tests were performed to examine the deformability and strength of the tissues. The mechanical properties of pathologically unchanged native

The modulus of elasticity values in the circumferential direction at the level of stress 1.0 MPa are significantly different amongst all three tested specimens, the biggest being for the Epic bioprostheses with 101.99"58.24 MPa, followed by native porcine 42.3"4.96 MPa and human valve leaflets 15.34"5.3 MPa (Fig. 2a). Epic bioprostheses have also the highest ultimate stress value 5.77"1.94 MPa, which is significantly higher than that of native human 1.74"0.37 MPa and porcine aortic valves 1.58"0.26 MPa (Fig. 2b). Ultimate strain in the circumferential direction is highest for the native human valves 18.35"7.61%, followed by 7.26"0.69% for native porcine valve leaflets and 5.95"1.54% for Epic bioprostheses, respectively (Fig. 2c). Also in radial direction at the level of stress 0.1 MPa modulus of elasticity is significantly different among all tested samples (Fig. 2d), the biggest for St Jude Epic bioprostheses 9.18"1.81 MPa, with 5.33"0.61 MPa for por-

Fig. 2. Modulus of elasticity: (a) in circumferential, (d) radial direction; ultimate stress: (b) in circumferential, (e) radial direction; ultimate strain: (c) in circumferential, (f) radial direction. In all graphs: (i) native porcine valve, (ii) Epic bioprosthesis, (iii) human valve. Closed bars represent the 95% confidence interval of the mean values which are marked with a bold line, open bars represent the whole range of values.

ARTICLE IN PRESS M. Kalejs et al. / Interactive CardioVascular and Thoracic Surgery 8 (2009) 553–557

555

cine valves and 1.98"0.24 MPa for human aortic valves. Human native valve tissue has the smallest ultimate stress values 0.32"0.04 MPa, compared with 0.55"0.11 MPa for porcine tissue and 0.7"0.21 MPa for Epic bioprostheses (Fig. 2e). Similar like in circumferential direction also in radial direction ultimate strain values are the highest for human valve tissue 23.92"4.87%, followed by native porcine valves 8.57"0.8% and Epic bioprostheses 7.92"1.74% (Fig. 2f). There is no statistically significant difference between sample thickness of native human valve leaflet samples 0.57"0.16 mm and the thickness of chemically treated porcine aortic valve leaflets used in Epic bioprostheses 0.59"0.15 mm. The native porcine valve tissue is of 0.92"0.17 mm thickness, and significantly thicker than the other two samples. 4. Discussion Experimental results show that the mechanical properties of aortic valve leaflets are different in circumferential and radial directions which is in accordance with previously published data about the anisotropy of this material w8, 9x. This feature is determined by the structural composition of heart valve leaflets, more precisely – the orientation of connective tissue fibers in the leaflets. Several authors have demonstrated by histological and ultrastuctural studies that elastin fibers are oriented more or less equally both in radial and circumferential directions but collagen fibers are mostly aligned in the circumferential direction w7, 10x. Our study has shown that the ultimate stress is 2.9–8.3 times higher in the circumferential direction compared to radial direction, as it is dependent on the concentration of collagen fibers. Modulus of elasticity in circumferential direction is even 7.6–11.1 times higher than in radial direction, a similar tendency has been observed also by Sauren and Missirlis with colleagues w9, 11x for porcine valve tissue – they reported the modulus of elasticity being respectively 21 and 3 times higher in circumferential direction. The apparent discrepancy between the results could be explained by different stress levels at which the elasticity modulus was calculated. Although all three tested tissue types have a similar nonlinear stress–strain response (Figs. 3 and 4), Epic bioprostheses show a prominent shift to the stress axis and significantly higher ultimate stress values. As shown previously w4, 12x, mechanical properties of the tissue are determined by fixing the tissue in a loaded or unloaded state. Fixation in a loaded state causes a shift to the left of the stress–strain curve closer to the stress axis but fixation in an unloaded state to the right. Tissue thickness also has been shown to be dependent on loading during fixation; Rousseau in 1983 w4x showed that after fixation in a loaded state the thickness is reduced by ;40%, in our study the difference between thickness of Epic bioprostheses leaflets and native porcine leaflets is ;36%. Thubrikar and colleagues w13x have come to a conclusion that valve leaflets to function properly require mechanical strength during diastole to prevent excess bulging and prolapse of valve leaflets as well as elasticity during the

Fig. 3. Representable stressystrain curves of the tested materials in circumferential direction. s, stress; ´, strain; S, porcine aortic valve; H, human aortic valve.

Fig. 4. Representable stressystrain curves of the tested materials in radial direction. s, stress; ´, strain; S, porcine aortic valve; H, human aortic valve.

first part of systole – it is required so that the valve can open as fast and efficient as possible, and is dependent on the flexion rigidity of the valve leaflets. Flexion rigidity on the other hand is directly proportional to modulus of elasticity and tissue thickness w14x. In a recent study, Mirnajafi and colleagues have shown that the region most exposed to flexion deformity is around comissures – the leaflets during opening undergo a rotation of ;658 at this region w15x. It is supported by observations that bioprostheses are very often damaged exactly in this region w16x. Flexion fatigue is thought to be one of the main causes of leaflet rupture w17, 18x. Arcidiacono and colleagues w19x with means of computer modeling have shown that even the slightest differences in rigidity of valve leaflets have an impact on the dynamics of valve opening and closure, stressing the importance of leaflet mechanical properties homogeneity within one valve. All these previous studies point out that increased leaflet rigidity characterized by

ARTICLE IN PRESS 556

M. Kalejs et al. / Interactive CardioVascular and Thoracic Surgery 8 (2009) 553–557

higher modulus of elasticity values may have significant impact on bioprosthesis longevity and hemodynamic properties. There is also a significant difference in ultimate strain amongst the tested materials, especially between native human valves and Epic bioprostheses in both directions. Several authors have pointed out that deformability in the radial direction is very important for proper leaflet coaptation and prolapse prevention w20x. In our study, the ultimate strain of Epic bioprostheses and native porcine valves does not differ very much and both are significantly smaller than for native human valves, which causes concerns about the suitability of this material for creation of an ‘ideal prosthesis’. Data on the failure of bioprostheses to copy the mechanical properties of the native human heart valves are in accordance with the large number of publications about non-calcifying valve leaflet degeneration w17, 21x. Although it is most likely that these processes – gradual structural deterioration and calcification work hand in hand. It is supported by many studies which demonstrate that calcification often begins in the regions of increased stress and deformation w16, 22, 23x. Further on calcification causes a loss of elasticity and an increase in flexion stress which accelerates further tissue degeneration. Although, when talking about the current generation of bioprostheses calcification is not the most actual problem – there is an increasing tendency to highlight damage caused solely by mechanical factors w17x. This study has certain limitations – we, limited by the expensiveness of bioprostheses, analyzed the mechanical properties of only one xenoaortic bioprosthetic heart valve model. At the moment we have no data on other xenoaortic or pericardial bioprostheses to compare. Still, taking into account that the main steps in chemical and physical treatment of biological tissues prior to use in bioprostheses are similar, we believe that the observed mechanical properties and drawn conclusions to some extent can be extrapolated to most of the traditional bioprostheses made from porcine aortic valves on the market. It should be kept in mind that this study shows neither superiority nor weakness of the analyzed bioprosthesis type compared to other bioprostheses. Only a study comparing mechanical properties of different bioprostheses before implantation and after in vivo or in vitro aging could give an answer on the superiority of a certain biological material or its treatment for use in bioprostheses. 5. Conclusions Epic bioprostheses have a non-linear and anisotropic response to stress in uniaxial tensile tests similar to native human and porcine aortic valve leaflets. They have the highest ultimate stress values but together with the gained mechanical strength they have lost tissue elasticity and are significantly more rigid compared to native valve tissue. The before-mentioned differences in mechanical properties between bioprostheses and native valves may cause

variations in stress distribution within leaflets of the prosthetic valve and accelerate its deterioration. References w1x Vesely I. The evolution of bioprosthetic heart valve design and its impact on durability. Cardiovascular Pathology 2003;12:277–286. w2x Morsi YS, Birchall IE, Rosenfeldt FL. Artificial aortic valves: an overview. Int J Artif Organs 2004;27:445–451. w3x Schoen FJ, Levy RJ. Calcification of tissue heart valve substitutes: progress toward understanding and prevention. Ann Thorac Surg 2005; 79:1072–1080. w4x Rousseau EPM, Sauren AAHJ, van Hout MC, van Steenhoven AA. Elastic and viscoelastic material behaviour of fresh and glutaraldehyde-treated porcine aortic valve tissue. J Biomech 1983;16:339–348. w5x Lehmann S, Walther T, Leontjev S, Kempfert J, Rastan A, Garbade J, Borger MA, Falk V, Mohr FW. Mid-term results after Epic xenograft implantation for aortic, mitral, and double valve replacement. J Heart Valve Dis 2007;16:641–648. w6x Myken PS. Seventeen-year experience with the St. Jude Medical Biocor porcine bioprosthesis. J Heart Valve Dis 2005;14:486–492. w7x Stradins P, Lacis R, Ozolanta I, Purina B, Ose V, Feldmane L, Kasyanov V. Comparison of biomechanical and structural properties between human aortic and pulmonary valve. Eur J Cardiothorac Surg 2004;26: 634–639. w8x Purinya B, Kasyanov V. Biomechanical and structural properties of the explanted bioprosthetic valve leaflets. J Biomech 1994;27:1–11. w9x Sauren AAHJ, van Hout MC, van Steenhoven AA, Veldpaus FE, Janssen JD. The mechanical properties of porcine aortic valve tissues. J Biomech 1983;16:327–337. w10x Sauren AAHJ, Kuijpers W, van Steenhoven AA, Veldpaus FE. Aortic valve histology and its relation with mechanics – preliminary report. J Biomech 1980;13:97–104. w11x Missirlis YF, Chong M. Aortic valve mechanics – Part I: material properties of natural porcine aortic valves. J Bioeng 1978;2:287–300. w12x Broom ND, Thomson FJ. Influence of fixation conditions on the performance of glutaraldehyde-treated porcine aortic valves: towards a more scientific basis. Thorax 1979;34:166–176. w13x Deck JD, Thubrikar MJ, Schneider PJ, Nolan SP. Structure, stress, and tissue repair in aortic valve leaflets. Cardiovascular Research 1988; 22:7–16. w14x Thubrikar MJ, Piepgrass WC, Bosher LP, Nolan SP. The elastic modulus of canine aortic valve leaflets in vivo and in vitro. Circ Res 1980;47:792– 800. w15x Mirnajafia A, Raymera JM, McClurea LR, Sacks MA. The flexural rigidity of the aortic valve leaflet in the commissural region. J Biomech 2006;39:2966–2973. w16x 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. w17x Vesely I, Barber JE, Ratliff NB. Tissue damage and calcification may be independent mechanisms of bioprosthetic heart valve failure. J Heart Valve Dis 2001;10:471–477. w18x Vesely I, Boughner D, Song T. Tissue buckling as a mechanism of bioprosthetic valve failure. Ann Thorac Surg 1988;46:302–308. w19x Arcidiacono G, Corvi A, Severi T. Functional analysis of bioprosthetic heart valves. J Biomech 2005;38:1483–1490. w20x Driessen NJB, Mol A, Bouten CVC, Baaijens FPT. Modeling the mechanics of tissue-engineered human heart valve leaflets. J Biomech 2007;40: 325–334. w21x Sacks MS, Schoen FJ. Collagen fiber disruption occurs independent of calcification in clinically explanted bioprosthetic heart valves. J Biomed Mater Res 2002;62:359–371. w22x Levy RJ, Schoen RJ, Levy JT, Nelson AC, Howard SL, Oshry LJ. Biologic determinants of dystrophic calcification and osteocalcin deposition in glutaraldehyde-preserved porcine aortic valve leaflets implanted subcutaneously in rats. Am J Pathol 1983;113:143–155. w23x Schoen FJ, Levy RJ, Nelson AC. Onset and progression of experimental bioprosthetic heart valve calcification. Lab Invest 1985;52:523–532.

ARTICLE IN PRESS M. Kalejs et al. / Interactive CardioVascular and Thoracic Surgery 8 (2009) 553–557 eComment: CorBeat trileaflet mechanical full-flow heart valve prosthesis versus native human aortic valves – evaluation of functional performance Authors: Leo A. Bockeria, Bakoulev Center for Cardiovascular Surgery, 121552 Moscow, Russia; Aleksandr Fadeev, Olga Bockeria, Osman Makhachev doi:10.1510/icvts.2008.196220A We read with great interest the findings of Martins Kalejs and colleagues w1x evaluating new bioprostheses by following the mechanics of the native human aortic valve. There are two good things about bioprostheses. First, they follow the natural structure of native human and reproduce the mode of its functioning. Second, they have a favorable effect on both the physiological blood flow constancy and the shortening of the patient’s hemodynamics recovery. Attempts to develop a new bioprosthesis with regard to mechanics of human aortic valve leaflets are worthwhile. But there is still room for improvement in mechanical valves possessing one common disadvantage: they are stenotic due to unnatural design w2x. The imperfection of the design, particularly noticeable when using small size prostheses, has a number of impacts on the patient’s hemodynamics. The valve occluder (disc, leaflets) located directly in the blood flow may cause its obstruction and separation, high pressure gradients, etc. Finally, the conventional mechanical valves do not keep the physiological blood flow constancy. The need for a new valve led the Bakoulev Center to the development of the CorBeat trileaflet mechanical valve prosthesis which is close by its

557

design and mode of functioning to the native human. The all-carbon CorBeat has been specially designed for aortic valve replacement in children and adults with small aortic valves. CorBeat has a three hinge mechanism located on the top of housing. The major design achievement is a free of occluder full-flow orifice area of the prosthesis when leaflets open. In vitro values of effective orifice area of CorBeat were as much as 0.95–0.97 of its geometric values. The first implantations of the CorBeat trileaflet prosthesis were performed in the Bakoulev Center of Moscow on November 2007 w3x, and June–July 2008. References w1x Kalejs M, Stradins P, Lacis R, Ozolanta I, Pavars J, Kasyanov V. St. Jude Epic heart valve bioprostheses versus native human and porcine aortic valves – comparison of mechanical properties. Interact CardioVasc Thorac Surg 2009;8:553–557. w2x Gott VL, Alejo DE, Cameron DE. Mechanical heart valves: 50 years of evolution. Ann Thorac Surg 2003;76:2230–2239. w3x Bockeria LA, Bockeria OL, Fadeev AA, Soboleva NN, Agafonov AV, Melnikov AP, Kuznetsov VO, Nikolaev DA, Makhachev OA, Melnikov DA. The first experience of ‘CorBeat’ prosthetic tricuspid valve in patient with mitral lesion and persistent atrial fibrillation. Ann Surgery (in Russian) 2008;2:25–31.