Bio-Medical Materials and Engineering 25 (2015) 189–202 DOI 10.3233/BME-151269 IOS Press
189
Effects of plaque lengths on stent surface roughness Achmad Syaifudin a,b , Ryo Takeda c and Katsuhiko Sasaki c,∗ a
Department of Mechanical Engineering, Institut Teknologi Sepuluh Nopember, Surabaya, Indonesia Division of Human Mechanical Systems and Design, Graduate School of Engineering, Hokkaido University, Sapporo, Japan c Division of Human Mechanical Systems and Design, Faculty of Engineering, Hokkaido University, Sapporo, Japan b
Received 6 June 2014 Accepted 20 October 2014 Abstract. The physical properties of the stent surface influence the effectiveness of vascular disease treatment after stent deployment. During the expanding process, the stent acquires high-level deformation that could alter either its microstructure or the magnitude of surface roughness. This paper constructed a finite element simulation to observe the changes in surface roughness during the stenting process. Structural transient dynamic analysis was performed using ANSYS, to identify the deformation after the stent is placed in a blood vessel. Two types of bare metal stents are studied: a Palmaz type and a Sinusoidal type. The relationship between plaque length and the changes in surface roughness was investigated by utilizing three different length of plaque; plaque length longer than the stent, shorter than the stent and the same length as the stent. In order to reduce computational time, 3D cyclical and translational symmetry was implemented into the FE model. The material models used was defined as a multilinear isotropic for stent and hyperelastic for the balloon, plaque and vessel wall. The correlation between the plastic deformation and the changes in surface roughness was obtained by intermittent pure tensile test using specimen whose chemical composition was similar to that of actual stent material. As the plastic strain is achieved from FE simulation, the surface roughness can be assessed thoroughly. The study found that the plaque size relative to stent length significantly influenced the critical changes in surface roughness. It was found that the length of stent which is equal to the plaque length was preferable due to the fact that it generated only moderate change in surface roughness. This effect was less influential to the Sinusoidal stent. Keywords: Plastic deformation, balloon expandable stent, surface roughness, plaque length
1. Introduction Stent implantation is a well-known non-surgical method to treat vascular diseases. As an object implanted in the human body, it should have a certain physical properties to avoid the surface passivation, encrustation and tissue damage. The physical properties of the stent surface, well known as surface roughness, influence the effectiveness of vascular disease treatment [1]. In the deployment process, stent acquires high-level deformation that could change the magnitude of surface roughness. A considerable * Address for correspondence: Katsuhiko Sasaki, PhD, Professor, Division of Human Mechanical Systems and Design, Faculty of Engineering, Hokkaido University, N13 W8, Kita-ku, Sapporo 060-8628, Japan. Tel./Fax: +81 11 706 6378; E-mail: [email protected].
0959-2989/15/$35.00 © 2015 – IOS Press and the authors. All rights reserved
190
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
amount of literature has been published on surface roughness of stent and its influence. The first serious analysis reported that surface roughness beyond the scale of a micrometer strongly hinders the process of surface passivation and leads to thrombus formation, which may adhere to the vessel wall and block blood flow [2]. Roughness of biomaterial implant turned out to have a significant effect to thrombogenicity. An observation on the outer surface of some intravascular catheters confirmed that surface roughness is one of the factors which can cause thrombogenicity [3]. Materials with a smooth surface and low coefficient of friction appear to be more biocompatible by reducing the mechanical skinning and the shear forces at the biomaterial-tissue interface. This phenomenon has been supported by animal studies where stents with those surface properties have been shown to produce less urethral reaction [4]. The other study showed that grooved surfaces promoted increased rate of migration of endothelial cells, up to 64.6% when compared to smooth, control surfaces [5]. Furthermore, in vitro investigation into the effect of surface roughness of stainless steel on human vein endothelial cell confirmed that stainless steel with roughness of 671.8 ± 27.8 nm causes membrane cell injury that leads to endothelial cell activation and inflammation [6]. These studies emphasize that the surface roughness of stainless steel is an indispensable surface property in the vascular stent development. Based on these facts, controlling surface properties of stent after deployment is extremely important. Several techniques suggested, as acid pickling, annealing, electro polishing, and electrochemical polishing, to improve the surface roughness of stent [7–9]. Those methods succeed to decrease the surface roughness of stent, but as is well known, its plastic deformation still changes the surface condition after stent fully expands. Many investigators confirmed that changes in the grain structure of metals lead to changes in its surface roughness [10–14]. What has been done by the stent manufacturers is to determine the allowable surface roughness of the stent before implantation. This surface roughness actually changes during stent expansion. Conducting experiments either in vitro or in vivo on this subject in detail will encounter many obstacles due to the lack of strain gauge in micro-scale observation of expanding stent. By finite element simulation, the changes in surface roughness during the deployment process can be observed easily. There are numerous finite element simulations relating to the interaction between the expanding stent, plaque, and arterial vessel wall. However, no finite element simulation was found associating plastic deformations of the stent to its surface roughness. In this study, therefore, we attempted to observe the relationship between plastic strain and surface roughness of balloon expandable stents using finite element method; including various length of plaque that could influence the distribution of the changes in stent surface roughness. The results of investigation are expected to provide a better guideline for the manufacturing process of stents. 2. Method 2.1. Finite element model ANSYS R12.1 (ANSYS Inc., PA, USA) was used to design the geometric models presented in this work. It is a nonlinear dynamic transient finite element modeling with the presence of plaque and vessel wall. The three-dimensional finite element model consisted of two kinds of bare metal stents: Palmaz, representative of stent type with rigid structural geometry, and Sinusoidal stent, representative of stent type with flexible structural geometry [15]. The geometry of both models is shown in Fig. 1. The material models are defined as multilinear isotropic for stent and hyperelastic for the balloon, plaque and vessel wall. For hyperelastic constitutive equation, 5-parameters Mooney–Rivlin was used
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
191
(a)
(b) Fig. 1. Front and side views of two stents type used in simulation: (a) geometry of the Palmaz, as rigid stent and (b) the Sinusoidal stent, as flexible stent. All units in mm. Table 1 Material properties used in the finite element simulation Component Material
Balloon Polyurethane
Stent SUS316L
Plaque Hypocellular
Vessel wall Carotid
Young’s modulus (GPa)
0.03447
186.67
0.00219
0.00175
Poisson’s ratio
0.495
0.33
0.495
0.495
Material behavior
Hyperelastic
Multilinear isotropic
Hyperelastic
Hyperelastic
Material constant
Mooney–Rivlin (MPa): C10 = 1.03176 × 10−7 C01 = 3.69266 × 10−7
Fitting stress-strain curve resulted from pure tensile test of SUS316NG
Mooney–Rivlin (kPa): C10 = −802.723 C01 = 831.636 C11 = 1157.68 C20 = 0.0 C30 = 0.0
Mooney–Rivlin (kPa): C10 = 18.90 C01 = 2.75 C11 = 85.72 C20 = 590.43 C30 = 0.0
Note: Compiled using data from [16] and [17].
for the plaque and vessel wall while 2-parameters Mooney–Rivlin was assigned for the material characteristic of the balloon. All of parameter constants, except material properties for stent, are taken from previous literatures as summarized in Table 1. As a pilot study, using the same material for plaque design during all simulation will be useful in eliminating the effect of fracture mechanism of plaque during stent expansion. More sophisticated plaque structures will be discussed in future works. Considering cyclical and reflective symmetry of the model, a 1/12 model was generated for Palmaz and Sinusoidal simulation, as shown in Fig. 2(a). The plaque is modeled in three different lengths and having equal thickness in the central radial as illustrated in Fig. 2(b), to observe its influence to the changes of surface roughness. The balloon was located inside the stent with the outside diameter of the
192
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
(a)
(b)
Fig. 2. (a) Sectional view of the full computational model, 1/12 model was generated for the Palmaz and Sinusoidal simulation; 1/6 due to cyclical symmetry of geometry and 1/2 due to symmetry of boundary conditions. (b) Three different length of plaque compared to the length of stent. (Colors are visible in the online version of the article; http://dx.doi.org/10.3233/BME-151269.)
balloon being equal to the inside diameter of the stent. The vessel wall and the plaque were modeled in such a way so that the plaque surface was not in contact with the stent before the inflation process. The plaque was assumed to be attached to the vessel wall firmly, so there is no generated contact behavior and frictional force. The element type for meshing is used a higher order 3D 20-node solid element that exhibits quadratic displacement behavior, which is superior in irregular meshing and simulating deformations of incompressible hyperelastic materials [25]. Defining contact properties is a decisive stage in the procedure of nonlinear finite element modeling. For stent simulation, surface-to-surface contact model is the most well suited one because it covers interference fit assembly contact. Because it undergoes extremely large deformation, flexible contact to flexible target is the most suitable choice. Setting for normal stiffness should be conducted carefully to obtain rapid converged calculation with accuracy. The normal stiffness of 0.1 for contact from balloon to stent and 10 for contact from balloon to plaque and vessel produced satisfactory results. In order to achieve better-converged result in finite element simulation, the coefficient of friction should be defined and chosen properly. The coefficient of friction for contact between rubber to a metal surface without dynamic pressure is below the value of 0.4 [18] and for stent is around the value of 0.125 but no more than that of 0.15 [19]. Another research has reported that friction coefficient from experimental measurements on living endothelial cells due to polished glass pin is around 0.06 [20]. In this investigation, the value of 0.3, 0.15 and 0.06 were assumed respectively as the coefficient of friction for the balloon, the stent and the plaque/vessel wall. 2.2. Loading and solution Boundary conditions of the model are represented in Fig. 3(a), where y direction of the sides has been restrained. In the end-side of the model, degree of freedom of balloon is restricted in all direction, while that of the vessel wall is restrained in the y and x direction due to its infinite length and hyperelasticity properties. To achieve geometric symmetry; symmetry boundary condition was applied to the cross section of the model and both sides of the model were coupled with offset nodes coupling along the circumferential direction for cyclical symmetry. Loading for both models is applied on the inner surface
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
193
Fig. 3. Boundary conditions and loading for both stent models in polar coordinates, X, Y and Z are representatives of radial, circumferential and longitudinal directions, respectively: (a) boundary conditions on four sides of the model. The blue triangles represent applied constraints and the S letter represents symmetry boundary conditions; (b) off-side coupling on both side of model (represented by red area in the side of the model) to generate cyclical symmetry and pressure loading in outward direction on the inner surface of balloon (represented by red arrows); (c) time history of pressure loading for both stents, the ordinate is pressure value, in MPa and the abscissa is time, in second. (The colors are visible in the online version of the article; http://dx.doi.org/10.3233/BME-151269.)
194
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
of the balloon in the outward direction as illustrated in Fig. 3(b), expanding the stent radially until passing its elastic deformation limit to a maximum diameter before failure stress was reached. Ramped pressure is chosen as loading, as suggested by previous study [21], to inflate the balloon steadily until reaching 30–40% deformation and then to deflate it. The balloon was subjected to a uniform internal pressure increasing from 0 to 0.95 MPa for the Palmaz model and from 0 to 0.80 MPa for the Sinusoidal model, as illustrated in Fig. 3(c). In result, the magnitude of pressure is not similar for both models due to surface area being different. Pressure is applied in 1.5 s; expanding pressure is 1 s and the remaining time is for balloon removal at a pressure rate arranged by ANSYS implicit solver because of its automatic time stepping. 2.3. Measurement of surface roughness The measurement of the changes in surface roughness for actual stent is approached with large-scale experiment of specimen whose chemical composition is similar to that of commercial stent material. The kind of experiment is intermittent pure tensile test, which is conducted to acquire the experimental data regarding the correlation between the plastic deformation and changes in surface roughness. The data from intermittent tensile test is supposed to describe how plastic deformation affects the changes in surface roughness; despite the residual stress within expanded stent is having more complexity rather than that of tensile test specimen. Geometry of the specimen and its dimension is shown in Fig. 4(a). Steel sheet geometry is chosen due to having response to tension is the most appropriate response of stent struts while it is being expanded. The material for the test is SUS 316 with the chemical composition displayed in Table 2. Loading was not applied continuously until specimen failure. This was because after each unloading, surface roughness was measured, as illustrated in Fig. 4(b). The measurement of surface roughness is performed on the same 2D area of the specimen (indicated by cyan color on the specimen surface) using color laser scanning microscope, thus, the obtained surface roughness is 2D surface roughness. Two different load strains were applied in the intermittent tensile test with loading speed 0.005 mm/s: 0.5% load strain was for axial strain less than 5% and 2.5% load strain was for the remaining axial strain. The results of microscopic observation and measurement are represented in Fig. 5(a). For the same magnification (400×), it is obvious that plastic strain strongly affects the surface roughness of the specimen. From the 2D surface roughness-plastic strain relationship as displayed in Fig. 5(b), the correlation between plastic deformation and 2D surface roughness of SUS 316 can be expressed as follows: S.R. = 0.16εp + 0.19,
(1)
where S.R. is surface roughness in µm and εp is plastic strain in mm/mm. Equation (1) shows the lowest limit for SUS316 from experiencing surface roughness changes due to plastic deformation. We can derive from Eq. (1) that a strain endurance limit of SUS316 that does not change the surface roughness is the strain of 0.012 mm/mm. By using Eq. (1), as the unloading plastic strain is achieved from FE simulation, the surface roughness can be assessed and predicted immediately. Thus, the similar procedure can be applied to all node of finite element model. Table 2 Chemical composition of SUS 316 Fe 66.332
C 0.009
Si 0.64
Mn 1.25
P 0.018
S 0.001
Ni 12.06
Cr 17.65
Mo 2.04
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
195
(a)
(b) Fig. 4. (a) Specimen of tensile test, (b) the illustration of intermittent tensile test, microscopic observation is conducted at the same area assigned by cyan color at the surface of specimen. (Colors are visible in the online version of the article; http://dx.doi.org/10.3233/BME-151269.)
3. Result and discussion Analysis of plastic strain distribution of stent will lead to consideration of surface roughness changes, the relationship between plastic strain and surface roughness was revealed in Section 2.3. The distribution of the surface roughness changes for Palmaz and Sinusoidal model are shown respectively in Figs 6(a) and 7(a), with the central zone or boundary conditions symmetry side is on the right side of figure, and the distal zone is on the other side. Dark blue area of stent shows the endurance area, which represents the part of stent without any changes in surface roughness. The generated strain belongs to endurance area is very small that less than the strain of 0.012 mm/mm, which is equal to 0.190 µm of surface roughness. The surface roughness changes are concentrated in the end of struts and bridges area. It is apparent on Figs 6(a) and 7(a) that the length of plaque does not extremely influence where the critical changes in surface roughness are located. The critical changes in surface roughness for all type of plaque length are located on distal area. The distal area of stent in the simulation represented the part
196
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
(a)
(b) Fig. 5. (a) The results of microscopic scanning photograph at magnification of 400× for 5%, 10%, 15% and 20% strains. (b) The linear correlation between plastic strains vs. 2D surface roughness of SUS 316 after each their measurement. The measurement is conducted automatically by the microscope while it is scanning the surface of the specimen. (Colors are visible in the online version of the article; http://dx.doi.org/10.3233/BME-151269.)
of stent, which has the largest changing in plastic strain due to stent could expand with fewer obstacles from plaque. An important point should be noticed from both models that critical surface roughness changes in model of type A is not covered by plaque, which is the most likely to cause severe damage to the vessel wall due to it having the largest surface roughness changes at the end of stent that is in direct contact with blood vessel wall. Nevertheless, in case of stent having longer length than plaque length, the area of the uncovered plaque remains as a stenotic blood vessel and there is a possibility of the uncovered plaque
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
197
(a)
(b) Fig. 6. (a) Distribution of the changes in surface roughness for the Palmaz stent with a unified legend provided below. The X, Y , Z coordinates represent the global Y -coordinate system, where the X axis is the radial direction, the Y axis the circumferential direction, and the Z axis the axial direction. (b) Corresponding values of the critical changes in surface roughness for each plaque length in the field of central, medial and distal area. The changes in surface roughness is indicated by ordinate and focused areas as indicated by abscissa. (Colors are visible in the online version of the article; http://dx.doi.org/ 10.3233/BME-151269.)
198
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
(a)
(b) Fig. 7. (a) Distribution of the changes in surface roughness for the Sinusoidal stent with a unified legend provided below. The X, Y , Z coordinates represent the global Y -coordinate system, where the X axis is the radial direction, the Y axis the circumferential direction, and the Z axis the axial direction. (b) Corresponding values of the critical changes in surface roughness for each plaque length in the field of central, medial and distal area, which the vertical axis is the value of the changes in surface roughness and the horizontal one is focused areas. (Colors are visible in the online version of the article; http://dx.doi.org/ 10.3233/BME-151269.)
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
199
peeling off. On the other hand, the critical surface roughness changes in the simulation model of types B and C have no direct connection to the vessel wall. When the critical changes in surface roughness takes place in the direct contact area between the stent’s distal zone and critical vessel wall, the vessel tissue would become damaged more quickly, leading to open-lesion of cell membrane structure. Moreover, stent length is preferable to be as short as possible, because a stent is basically a foreign material and therefore the smaller surface area of a foreign material is preferable based on the immunologic response standpoint. Considering those factors, stent length equal to the plaque length is recommended. For better observation, the overall stress distribution within the vessel wall is demonstrated in Fig. 8. The stress distribution is focused on the vessel wall by neglecting that within the plaque, because the stresses induced within the vessel wall can be an undermined stimulus for the growth of restenotic tissue [22,23]. Figure 8(a) shows the residual von Mises stress distribution within the vessel wall after balloon removal induced by the Palmaz stent and Fig. 8(b) shows that induced by the Sinusoidal stent, combined with the appearance of the edge line of plaque model. The plaque length of type A from both models causes direct contact between the stent and vessel wall surface. As a result, the vessel wall of type A received larger pressure than that of types B and C. Meanwhile, stress within the vessel wall, induced by plaque types B and C, is shielded by the plaque thoroughly reducing the possibility of arterial lesion. Especially for type C plaque of Palmaz model, the critical stress emerges outside of stent surface area. Therefore, even though it shows the highest stress among the other types; instead it is much safer for vessel wall. The corresponding value of the changes in surface roughness for each model in respect of longitudinal direction is displayed in Fig. 6(b) and 7(b), stated in the field of central, medial, and distal area of the stent. Both figures shows similar tendency of the changes in surface roughness that take place in central and distal area. For Palmaz model, the struts in medial area represent endurance from having changes in surface roughness. Meantime, central and medial areas of Sinusoidal model show the endurance due to flexibility properties of the Sinusoidal stent. The important finding from the figures was that the plaque length strongly influenced the critical changes in stent surface roughness. This is indicated by the value of SR in distal area, which type A has the largest changes. In better correlating the effect of different plaque length to the changes in stent surface roughness, the changes in stent geometry while expanding should be characterized. In order to characterize those effects, there are 5 parameters recommended to be measured: elastic recoil (central radial recoil, distal radial recoil, and longitudinal recoil), foreshortening, and dogboning of the stent [24], which the results were summarized in Table 3. It is seen that all parameters of expanded stent are very sensitive to variation of the plaque length, but the sensitiveness is less with the increasing length of plaque. The longer plaque length, the larger elastic recoil of the stents is. On the other hand, the plaque length succeeds decreasing foreshortening and dogboning effect of the stents. Especially for Sinusoidal model, there is no dogboning effect when the plaque has similar length to the stent or more. In summary, elastic recoil which is followed by the increasing foreshortening and dogboning of stent, tend to enlarge the possibility of the changes in stent surface roughness.
4. Study limitations This investigation presents a finite element analysis of predicting the changes in surface roughness, due to the different length of concentric plaque. In actual conditions, the shape of the plaque is eccentric circumferentially and axially which could drastically cause plastic strain deformation in any part of
200
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
(a)
(b) Fig. 8. (a) Stress concentration after balloon removal within the vessel wall due to the Palmaz stent, with a unified legend provided below. The X, Y , Z coordinates represent the global Y -coordinate system, where the X axis is the radial direction, the Y axis the circumferential direction, and the Z axis the axial direction. (b) Stress concentration after balloon removal within the vessel wall due to the Sinusoidal stent, with a unified legend provided below. The X, Y , Z coordinates represent the global Y -coordinate system, where the X axis is the radial direction, the Y axis the circumferential direction, and the Z axis the axial direction. (Colors are visible in the online version of the article; http://dx.doi.org/10.3233/BME-151269.)
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
201
Table 3 Calculation results Stent
Type
Ratio plaque to stent
Distal radial recoil (%)
Central radial recoil (%)
Longitudinal recoil (%)
Foreshortening (%)
Dogboning (%)
Palmaz
A B C
0.67 1.00 1.33
6.20 8.35 15.72
8.12 9.62 10.27
−0.25 −0.21 −0.20
2.34 1.62 1.26
47.10 42.62 31.71
Sinusoidal
A B C
0.67 1.00 1.33
51.10 42.85 66.71
36.04 63.95 65.48
−6.44 −3.04 −2.72
10.63 6.06 5.38
55.27 −6.65 −6.28
stent surface and eventually alter the surface roughness. In addition, the experimental data was not obtained from the actual Palmaz or Sinusoidal stent, but it came from a test piece of tensile test specimen. As is known, there is only one directional stress acting in the tensile test specimen, i.e. axial stress. However, in the expanding and deflating process of stent, radial pressure loading could generate three directional stresses acting in the stent, i.e. radial, axial and circumferential stress [16]. Even though having equal material properties; elevated stress level could cause different grain boundary elongation, which ultimately would affect the magnitude of plastic strain and the changes in surface roughness comprehensively. More sophisticated expansion models will be discussed in future works. 5. Conclusion Deformation analysis using finite element simulation is particularly useful in identifying critical surface roughness of balloon-expanded stent, based on its plastic strain distribution. The results of the present study demonstrate that the plaque size relative to stent length significantly influences the critical surface roughness. It can be inferred for both type of stent that using shorter stent length than plaque length leads to decreasing the changes in surface roughness after balloon removal. This research demonstrated that the plaque length of type B, which had equal length to the stent, is the most recommended model to be implemented in the actual stenting due to the fact that it generated moderate change in surface roughness and corresponding stress concentration within the vessel wall. This information may be valuable for practitioners and doctors in choosing stent length and locating stent in the implantation process. In addition, the simulation results can also be used as a basis in designing initial surface roughness for improving stent performance and in performing stent coating at its critical surfaces. Acknowledgement The authors wish to express thanks to Kazuhiko Takahashi of the Laboratory of Micro Energy System (Division of Human Mechanical Systems and Design, Hokkaido University), for support and cooperation in the experiments of this study. Conflict of interest There are no actual or potential conflicts of interest.
202
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
References [1] B. O’Brien and W. Carroll, The evolution of cardiovascular stent materials and surfaces in response to clinical drivers: a review, Acta Biomaterialia 5(4) (2009), 945–958. [2] V.A. DePalma, R.E. Baier, J.W. Ford, V.L. Gott and A. Furuse, Investigation of three-surface properties of several metals and their relation to blood compatibility, Journal of Biomedical Materials Research 6 (1972), 37–75. [3] J.F. Hecker and L.A. Scandrett, Roughness and thrombogenicity of the outer surface of intravascular catheters, Journal of Biomedical Materials Research 19(4) (1985), 381–395. [4] J.D. Denstedt, T.A. Wollin and G. Reid, Biomaterials used in urology: current issues of biocompatibility, infection, and encrustation, Journal of Endourology 12(6) (1998), 493–500. [5] J.C. Palmaz, A. Benson and E.A. Sprague, Influence of surface topography on endothelialization of intravascular metallic material, Journal of Vascular and Interventional Radiology 10(4) (1999), 439–444. [6] E. McLucas, M.T. Moran, Y. Rochev, W.M. Carroll and T.J. Smith, An investigation into the effect of surface roughness of stainless steel on human umbilical vein endothelial cell gene expression, Endothelium 13(1) (2006), 35–41. [7] G. Tepe, H.P. Wendel, S. Khorchidi, J. Schmehl, J. Wiskirchen, B. Pusich, C.D. Claussen and S.H. Duda, Thrombogenicity of various endovascular stent types: an in vitro evaluation, Journal of Vascular and Interventional Radiology 13(10) (2002), 1029–1035. [8] H. Zhao, J. Van Humbeeck, J. Sohier and I. de Scheerder, Electrochemical polishing of 316L stainless steel slotted tube coronary stents, Journal of Materials Science: Materials in Medicine 13(10) (2002), 911–916. [9] T. Hryniewicz, R. Rokicki and K. Rokosz, Surface characterization of AISI 316L biomaterials obtained by electropolishing in a magnetic field, Surface and Coatings Technology 202(9) (2008), 1668–1673. [10] T. Mizuno and H. Mulki, Changes in surface texture of zinc-coated steel sheets under plastic deformation, Wear 198(1,2) (1996), 176–184. [11] C.M. Wichern, B.C. De Cooman and C.J. Van Tyne, Surface roughness changes on a hot-dipped galvanized sheet steel during deformation at low strain levels, Acta Materialia 52(5) (2004), 1211–1222. [12] C.M. Wichern, B.C. De Cooman and C.J. Van Tyne, Surface roughness of a hot-dipped galvanized sheet steel as a function of deformation mode, Journal of Materials Processing Technology 160(3) (2005), 278–288. [13] H.A. Al-Qureshi, A.N. Klein and M.C. Fredel, Grain size and surface roughness effect on the instability strains in sheet metal stretching, Journal of Materials Processing Technology 170(1,2) (2005), 204–210. [14] O. Wouters, W.P. Vellinga, R. Van Tijum and J.Th.M. de Hosson, On the evolution of surface roughness during deformation of polycrystalline aluminum alloys, Acta Materialia 53(15) (2005), 4043–4050. [15] F. Ju, Z. Xia and K. Sasaki, On the finite element modelling of balloon-expandable stents, Journal of the Mechanical Behavior of Biomedical Materials 1(1) (2008), 86–95. [16] S.N. David Chua, B.J. Mac Donald and M.S.J. Hashmi, Finite element simulation of slotted tube (stent) with the presence of plaque and artery by balloon expansion, Journal of Materials Processing Technology 155–156 (2004), 1772–1779. [17] I. Pericevic et al., The influence of plaque composition on underlying arterial wall stress during stent expansion: the case for lesion-specific stents, Medical Engineering and Physics 31 (2009), 428–433. [18] M.A. Jiménez, J.M. Bielsa, R. Rodríguez and S. Dobón, The influence of contact pressure on the dynamic friction coefficient in cylindrical rubber–metal contact geometries, IUTAM Bookseries of Symposium on Computational Methods in Contact Mechanics 3 (2007), 257–275. [19] J. Mummert, A. Reynolds, E. Sirois and W. Sun, Determination of radial force and coefficient of friction with a self-expanding transcatheter aortic valve stent, in: Bioengineering Conference (NEBEC) 38th Annual Northeast, 2012, pp. 189–190. [20] A.C. Dunn, T.D. Zaveri, B.G. Keselowsky and W.G. Sawyer, Macroscopic friction coefficient measurements on living endothelial cells, Tribology Letters 27(2) (2007), 233–238. [21] S.N. David Chua, B.J. Mac Donald and M.S.J. Hashmi, Finite element simulation of stent and balloon interaction, Journal of Material Processing Technology 143–144 (2003), 591–597. [22] C. Lally, F. Dolan and P.J. Prendergast, Elastic behavior of porcine coronary artery tissue under uniaxial and equibiaxial tension, Annals of Biomedical Engineering 32(10) (2004), 1355–1364. [23] C. Lally, F. Dolan and P.J. Prendergast, Cardiovascular stent design and vessel stresses: a finite element analysis, Journal of Biomechanics 38(8) (2005), 1574–1581. [24] F. Migliavacca, L. Petrini, M. Colombo, F. Auricchio and R. Pietrabissa, Mechanical behavior of coronary stents investigated through the finite element method, Journal of Biomechanics 35(6) (2002), 803–811. [25] ANSYS Inc., ANSYS Mechanical APDL Element Reference, 2012.
Copyright of Bio-Medical Materials & Engineering is the property of IOS Press and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
189
Effects of plaque lengths on stent surface roughness Achmad Syaifudin a,b , Ryo Takeda c and Katsuhiko Sasaki c,∗ a
Department of Mechanical Engineering, Institut Teknologi Sepuluh Nopember, Surabaya, Indonesia Division of Human Mechanical Systems and Design, Graduate School of Engineering, Hokkaido University, Sapporo, Japan c Division of Human Mechanical Systems and Design, Faculty of Engineering, Hokkaido University, Sapporo, Japan b
Received 6 June 2014 Accepted 20 October 2014 Abstract. The physical properties of the stent surface influence the effectiveness of vascular disease treatment after stent deployment. During the expanding process, the stent acquires high-level deformation that could alter either its microstructure or the magnitude of surface roughness. This paper constructed a finite element simulation to observe the changes in surface roughness during the stenting process. Structural transient dynamic analysis was performed using ANSYS, to identify the deformation after the stent is placed in a blood vessel. Two types of bare metal stents are studied: a Palmaz type and a Sinusoidal type. The relationship between plaque length and the changes in surface roughness was investigated by utilizing three different length of plaque; plaque length longer than the stent, shorter than the stent and the same length as the stent. In order to reduce computational time, 3D cyclical and translational symmetry was implemented into the FE model. The material models used was defined as a multilinear isotropic for stent and hyperelastic for the balloon, plaque and vessel wall. The correlation between the plastic deformation and the changes in surface roughness was obtained by intermittent pure tensile test using specimen whose chemical composition was similar to that of actual stent material. As the plastic strain is achieved from FE simulation, the surface roughness can be assessed thoroughly. The study found that the plaque size relative to stent length significantly influenced the critical changes in surface roughness. It was found that the length of stent which is equal to the plaque length was preferable due to the fact that it generated only moderate change in surface roughness. This effect was less influential to the Sinusoidal stent. Keywords: Plastic deformation, balloon expandable stent, surface roughness, plaque length
1. Introduction Stent implantation is a well-known non-surgical method to treat vascular diseases. As an object implanted in the human body, it should have a certain physical properties to avoid the surface passivation, encrustation and tissue damage. The physical properties of the stent surface, well known as surface roughness, influence the effectiveness of vascular disease treatment [1]. In the deployment process, stent acquires high-level deformation that could change the magnitude of surface roughness. A considerable * Address for correspondence: Katsuhiko Sasaki, PhD, Professor, Division of Human Mechanical Systems and Design, Faculty of Engineering, Hokkaido University, N13 W8, Kita-ku, Sapporo 060-8628, Japan. Tel./Fax: +81 11 706 6378; E-mail: [email protected].
0959-2989/15/$35.00 © 2015 – IOS Press and the authors. All rights reserved
190
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
amount of literature has been published on surface roughness of stent and its influence. The first serious analysis reported that surface roughness beyond the scale of a micrometer strongly hinders the process of surface passivation and leads to thrombus formation, which may adhere to the vessel wall and block blood flow [2]. Roughness of biomaterial implant turned out to have a significant effect to thrombogenicity. An observation on the outer surface of some intravascular catheters confirmed that surface roughness is one of the factors which can cause thrombogenicity [3]. Materials with a smooth surface and low coefficient of friction appear to be more biocompatible by reducing the mechanical skinning and the shear forces at the biomaterial-tissue interface. This phenomenon has been supported by animal studies where stents with those surface properties have been shown to produce less urethral reaction [4]. The other study showed that grooved surfaces promoted increased rate of migration of endothelial cells, up to 64.6% when compared to smooth, control surfaces [5]. Furthermore, in vitro investigation into the effect of surface roughness of stainless steel on human vein endothelial cell confirmed that stainless steel with roughness of 671.8 ± 27.8 nm causes membrane cell injury that leads to endothelial cell activation and inflammation [6]. These studies emphasize that the surface roughness of stainless steel is an indispensable surface property in the vascular stent development. Based on these facts, controlling surface properties of stent after deployment is extremely important. Several techniques suggested, as acid pickling, annealing, electro polishing, and electrochemical polishing, to improve the surface roughness of stent [7–9]. Those methods succeed to decrease the surface roughness of stent, but as is well known, its plastic deformation still changes the surface condition after stent fully expands. Many investigators confirmed that changes in the grain structure of metals lead to changes in its surface roughness [10–14]. What has been done by the stent manufacturers is to determine the allowable surface roughness of the stent before implantation. This surface roughness actually changes during stent expansion. Conducting experiments either in vitro or in vivo on this subject in detail will encounter many obstacles due to the lack of strain gauge in micro-scale observation of expanding stent. By finite element simulation, the changes in surface roughness during the deployment process can be observed easily. There are numerous finite element simulations relating to the interaction between the expanding stent, plaque, and arterial vessel wall. However, no finite element simulation was found associating plastic deformations of the stent to its surface roughness. In this study, therefore, we attempted to observe the relationship between plastic strain and surface roughness of balloon expandable stents using finite element method; including various length of plaque that could influence the distribution of the changes in stent surface roughness. The results of investigation are expected to provide a better guideline for the manufacturing process of stents. 2. Method 2.1. Finite element model ANSYS R12.1 (ANSYS Inc., PA, USA) was used to design the geometric models presented in this work. It is a nonlinear dynamic transient finite element modeling with the presence of plaque and vessel wall. The three-dimensional finite element model consisted of two kinds of bare metal stents: Palmaz, representative of stent type with rigid structural geometry, and Sinusoidal stent, representative of stent type with flexible structural geometry [15]. The geometry of both models is shown in Fig. 1. The material models are defined as multilinear isotropic for stent and hyperelastic for the balloon, plaque and vessel wall. For hyperelastic constitutive equation, 5-parameters Mooney–Rivlin was used
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
191
(a)
(b) Fig. 1. Front and side views of two stents type used in simulation: (a) geometry of the Palmaz, as rigid stent and (b) the Sinusoidal stent, as flexible stent. All units in mm. Table 1 Material properties used in the finite element simulation Component Material
Balloon Polyurethane
Stent SUS316L
Plaque Hypocellular
Vessel wall Carotid
Young’s modulus (GPa)
0.03447
186.67
0.00219
0.00175
Poisson’s ratio
0.495
0.33
0.495
0.495
Material behavior
Hyperelastic
Multilinear isotropic
Hyperelastic
Hyperelastic
Material constant
Mooney–Rivlin (MPa): C10 = 1.03176 × 10−7 C01 = 3.69266 × 10−7
Fitting stress-strain curve resulted from pure tensile test of SUS316NG
Mooney–Rivlin (kPa): C10 = −802.723 C01 = 831.636 C11 = 1157.68 C20 = 0.0 C30 = 0.0
Mooney–Rivlin (kPa): C10 = 18.90 C01 = 2.75 C11 = 85.72 C20 = 590.43 C30 = 0.0
Note: Compiled using data from [16] and [17].
for the plaque and vessel wall while 2-parameters Mooney–Rivlin was assigned for the material characteristic of the balloon. All of parameter constants, except material properties for stent, are taken from previous literatures as summarized in Table 1. As a pilot study, using the same material for plaque design during all simulation will be useful in eliminating the effect of fracture mechanism of plaque during stent expansion. More sophisticated plaque structures will be discussed in future works. Considering cyclical and reflective symmetry of the model, a 1/12 model was generated for Palmaz and Sinusoidal simulation, as shown in Fig. 2(a). The plaque is modeled in three different lengths and having equal thickness in the central radial as illustrated in Fig. 2(b), to observe its influence to the changes of surface roughness. The balloon was located inside the stent with the outside diameter of the
192
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
(a)
(b)
Fig. 2. (a) Sectional view of the full computational model, 1/12 model was generated for the Palmaz and Sinusoidal simulation; 1/6 due to cyclical symmetry of geometry and 1/2 due to symmetry of boundary conditions. (b) Three different length of plaque compared to the length of stent. (Colors are visible in the online version of the article; http://dx.doi.org/10.3233/BME-151269.)
balloon being equal to the inside diameter of the stent. The vessel wall and the plaque were modeled in such a way so that the plaque surface was not in contact with the stent before the inflation process. The plaque was assumed to be attached to the vessel wall firmly, so there is no generated contact behavior and frictional force. The element type for meshing is used a higher order 3D 20-node solid element that exhibits quadratic displacement behavior, which is superior in irregular meshing and simulating deformations of incompressible hyperelastic materials [25]. Defining contact properties is a decisive stage in the procedure of nonlinear finite element modeling. For stent simulation, surface-to-surface contact model is the most well suited one because it covers interference fit assembly contact. Because it undergoes extremely large deformation, flexible contact to flexible target is the most suitable choice. Setting for normal stiffness should be conducted carefully to obtain rapid converged calculation with accuracy. The normal stiffness of 0.1 for contact from balloon to stent and 10 for contact from balloon to plaque and vessel produced satisfactory results. In order to achieve better-converged result in finite element simulation, the coefficient of friction should be defined and chosen properly. The coefficient of friction for contact between rubber to a metal surface without dynamic pressure is below the value of 0.4 [18] and for stent is around the value of 0.125 but no more than that of 0.15 [19]. Another research has reported that friction coefficient from experimental measurements on living endothelial cells due to polished glass pin is around 0.06 [20]. In this investigation, the value of 0.3, 0.15 and 0.06 were assumed respectively as the coefficient of friction for the balloon, the stent and the plaque/vessel wall. 2.2. Loading and solution Boundary conditions of the model are represented in Fig. 3(a), where y direction of the sides has been restrained. In the end-side of the model, degree of freedom of balloon is restricted in all direction, while that of the vessel wall is restrained in the y and x direction due to its infinite length and hyperelasticity properties. To achieve geometric symmetry; symmetry boundary condition was applied to the cross section of the model and both sides of the model were coupled with offset nodes coupling along the circumferential direction for cyclical symmetry. Loading for both models is applied on the inner surface
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
193
Fig. 3. Boundary conditions and loading for both stent models in polar coordinates, X, Y and Z are representatives of radial, circumferential and longitudinal directions, respectively: (a) boundary conditions on four sides of the model. The blue triangles represent applied constraints and the S letter represents symmetry boundary conditions; (b) off-side coupling on both side of model (represented by red area in the side of the model) to generate cyclical symmetry and pressure loading in outward direction on the inner surface of balloon (represented by red arrows); (c) time history of pressure loading for both stents, the ordinate is pressure value, in MPa and the abscissa is time, in second. (The colors are visible in the online version of the article; http://dx.doi.org/10.3233/BME-151269.)
194
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
of the balloon in the outward direction as illustrated in Fig. 3(b), expanding the stent radially until passing its elastic deformation limit to a maximum diameter before failure stress was reached. Ramped pressure is chosen as loading, as suggested by previous study [21], to inflate the balloon steadily until reaching 30–40% deformation and then to deflate it. The balloon was subjected to a uniform internal pressure increasing from 0 to 0.95 MPa for the Palmaz model and from 0 to 0.80 MPa for the Sinusoidal model, as illustrated in Fig. 3(c). In result, the magnitude of pressure is not similar for both models due to surface area being different. Pressure is applied in 1.5 s; expanding pressure is 1 s and the remaining time is for balloon removal at a pressure rate arranged by ANSYS implicit solver because of its automatic time stepping. 2.3. Measurement of surface roughness The measurement of the changes in surface roughness for actual stent is approached with large-scale experiment of specimen whose chemical composition is similar to that of commercial stent material. The kind of experiment is intermittent pure tensile test, which is conducted to acquire the experimental data regarding the correlation between the plastic deformation and changes in surface roughness. The data from intermittent tensile test is supposed to describe how plastic deformation affects the changes in surface roughness; despite the residual stress within expanded stent is having more complexity rather than that of tensile test specimen. Geometry of the specimen and its dimension is shown in Fig. 4(a). Steel sheet geometry is chosen due to having response to tension is the most appropriate response of stent struts while it is being expanded. The material for the test is SUS 316 with the chemical composition displayed in Table 2. Loading was not applied continuously until specimen failure. This was because after each unloading, surface roughness was measured, as illustrated in Fig. 4(b). The measurement of surface roughness is performed on the same 2D area of the specimen (indicated by cyan color on the specimen surface) using color laser scanning microscope, thus, the obtained surface roughness is 2D surface roughness. Two different load strains were applied in the intermittent tensile test with loading speed 0.005 mm/s: 0.5% load strain was for axial strain less than 5% and 2.5% load strain was for the remaining axial strain. The results of microscopic observation and measurement are represented in Fig. 5(a). For the same magnification (400×), it is obvious that plastic strain strongly affects the surface roughness of the specimen. From the 2D surface roughness-plastic strain relationship as displayed in Fig. 5(b), the correlation between plastic deformation and 2D surface roughness of SUS 316 can be expressed as follows: S.R. = 0.16εp + 0.19,
(1)
where S.R. is surface roughness in µm and εp is plastic strain in mm/mm. Equation (1) shows the lowest limit for SUS316 from experiencing surface roughness changes due to plastic deformation. We can derive from Eq. (1) that a strain endurance limit of SUS316 that does not change the surface roughness is the strain of 0.012 mm/mm. By using Eq. (1), as the unloading plastic strain is achieved from FE simulation, the surface roughness can be assessed and predicted immediately. Thus, the similar procedure can be applied to all node of finite element model. Table 2 Chemical composition of SUS 316 Fe 66.332
C 0.009
Si 0.64
Mn 1.25
P 0.018
S 0.001
Ni 12.06
Cr 17.65
Mo 2.04
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
195
(a)
(b) Fig. 4. (a) Specimen of tensile test, (b) the illustration of intermittent tensile test, microscopic observation is conducted at the same area assigned by cyan color at the surface of specimen. (Colors are visible in the online version of the article; http://dx.doi.org/10.3233/BME-151269.)
3. Result and discussion Analysis of plastic strain distribution of stent will lead to consideration of surface roughness changes, the relationship between plastic strain and surface roughness was revealed in Section 2.3. The distribution of the surface roughness changes for Palmaz and Sinusoidal model are shown respectively in Figs 6(a) and 7(a), with the central zone or boundary conditions symmetry side is on the right side of figure, and the distal zone is on the other side. Dark blue area of stent shows the endurance area, which represents the part of stent without any changes in surface roughness. The generated strain belongs to endurance area is very small that less than the strain of 0.012 mm/mm, which is equal to 0.190 µm of surface roughness. The surface roughness changes are concentrated in the end of struts and bridges area. It is apparent on Figs 6(a) and 7(a) that the length of plaque does not extremely influence where the critical changes in surface roughness are located. The critical changes in surface roughness for all type of plaque length are located on distal area. The distal area of stent in the simulation represented the part
196
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
(a)
(b) Fig. 5. (a) The results of microscopic scanning photograph at magnification of 400× for 5%, 10%, 15% and 20% strains. (b) The linear correlation between plastic strains vs. 2D surface roughness of SUS 316 after each their measurement. The measurement is conducted automatically by the microscope while it is scanning the surface of the specimen. (Colors are visible in the online version of the article; http://dx.doi.org/10.3233/BME-151269.)
of stent, which has the largest changing in plastic strain due to stent could expand with fewer obstacles from plaque. An important point should be noticed from both models that critical surface roughness changes in model of type A is not covered by plaque, which is the most likely to cause severe damage to the vessel wall due to it having the largest surface roughness changes at the end of stent that is in direct contact with blood vessel wall. Nevertheless, in case of stent having longer length than plaque length, the area of the uncovered plaque remains as a stenotic blood vessel and there is a possibility of the uncovered plaque
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
197
(a)
(b) Fig. 6. (a) Distribution of the changes in surface roughness for the Palmaz stent with a unified legend provided below. The X, Y , Z coordinates represent the global Y -coordinate system, where the X axis is the radial direction, the Y axis the circumferential direction, and the Z axis the axial direction. (b) Corresponding values of the critical changes in surface roughness for each plaque length in the field of central, medial and distal area. The changes in surface roughness is indicated by ordinate and focused areas as indicated by abscissa. (Colors are visible in the online version of the article; http://dx.doi.org/ 10.3233/BME-151269.)
198
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
(a)
(b) Fig. 7. (a) Distribution of the changes in surface roughness for the Sinusoidal stent with a unified legend provided below. The X, Y , Z coordinates represent the global Y -coordinate system, where the X axis is the radial direction, the Y axis the circumferential direction, and the Z axis the axial direction. (b) Corresponding values of the critical changes in surface roughness for each plaque length in the field of central, medial and distal area, which the vertical axis is the value of the changes in surface roughness and the horizontal one is focused areas. (Colors are visible in the online version of the article; http://dx.doi.org/ 10.3233/BME-151269.)
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
199
peeling off. On the other hand, the critical surface roughness changes in the simulation model of types B and C have no direct connection to the vessel wall. When the critical changes in surface roughness takes place in the direct contact area between the stent’s distal zone and critical vessel wall, the vessel tissue would become damaged more quickly, leading to open-lesion of cell membrane structure. Moreover, stent length is preferable to be as short as possible, because a stent is basically a foreign material and therefore the smaller surface area of a foreign material is preferable based on the immunologic response standpoint. Considering those factors, stent length equal to the plaque length is recommended. For better observation, the overall stress distribution within the vessel wall is demonstrated in Fig. 8. The stress distribution is focused on the vessel wall by neglecting that within the plaque, because the stresses induced within the vessel wall can be an undermined stimulus for the growth of restenotic tissue [22,23]. Figure 8(a) shows the residual von Mises stress distribution within the vessel wall after balloon removal induced by the Palmaz stent and Fig. 8(b) shows that induced by the Sinusoidal stent, combined with the appearance of the edge line of plaque model. The plaque length of type A from both models causes direct contact between the stent and vessel wall surface. As a result, the vessel wall of type A received larger pressure than that of types B and C. Meanwhile, stress within the vessel wall, induced by plaque types B and C, is shielded by the plaque thoroughly reducing the possibility of arterial lesion. Especially for type C plaque of Palmaz model, the critical stress emerges outside of stent surface area. Therefore, even though it shows the highest stress among the other types; instead it is much safer for vessel wall. The corresponding value of the changes in surface roughness for each model in respect of longitudinal direction is displayed in Fig. 6(b) and 7(b), stated in the field of central, medial, and distal area of the stent. Both figures shows similar tendency of the changes in surface roughness that take place in central and distal area. For Palmaz model, the struts in medial area represent endurance from having changes in surface roughness. Meantime, central and medial areas of Sinusoidal model show the endurance due to flexibility properties of the Sinusoidal stent. The important finding from the figures was that the plaque length strongly influenced the critical changes in stent surface roughness. This is indicated by the value of SR in distal area, which type A has the largest changes. In better correlating the effect of different plaque length to the changes in stent surface roughness, the changes in stent geometry while expanding should be characterized. In order to characterize those effects, there are 5 parameters recommended to be measured: elastic recoil (central radial recoil, distal radial recoil, and longitudinal recoil), foreshortening, and dogboning of the stent [24], which the results were summarized in Table 3. It is seen that all parameters of expanded stent are very sensitive to variation of the plaque length, but the sensitiveness is less with the increasing length of plaque. The longer plaque length, the larger elastic recoil of the stents is. On the other hand, the plaque length succeeds decreasing foreshortening and dogboning effect of the stents. Especially for Sinusoidal model, there is no dogboning effect when the plaque has similar length to the stent or more. In summary, elastic recoil which is followed by the increasing foreshortening and dogboning of stent, tend to enlarge the possibility of the changes in stent surface roughness.
4. Study limitations This investigation presents a finite element analysis of predicting the changes in surface roughness, due to the different length of concentric plaque. In actual conditions, the shape of the plaque is eccentric circumferentially and axially which could drastically cause plastic strain deformation in any part of
200
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
(a)
(b) Fig. 8. (a) Stress concentration after balloon removal within the vessel wall due to the Palmaz stent, with a unified legend provided below. The X, Y , Z coordinates represent the global Y -coordinate system, where the X axis is the radial direction, the Y axis the circumferential direction, and the Z axis the axial direction. (b) Stress concentration after balloon removal within the vessel wall due to the Sinusoidal stent, with a unified legend provided below. The X, Y , Z coordinates represent the global Y -coordinate system, where the X axis is the radial direction, the Y axis the circumferential direction, and the Z axis the axial direction. (Colors are visible in the online version of the article; http://dx.doi.org/10.3233/BME-151269.)
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
201
Table 3 Calculation results Stent
Type
Ratio plaque to stent
Distal radial recoil (%)
Central radial recoil (%)
Longitudinal recoil (%)
Foreshortening (%)
Dogboning (%)
Palmaz
A B C
0.67 1.00 1.33
6.20 8.35 15.72
8.12 9.62 10.27
−0.25 −0.21 −0.20
2.34 1.62 1.26
47.10 42.62 31.71
Sinusoidal
A B C
0.67 1.00 1.33
51.10 42.85 66.71
36.04 63.95 65.48
−6.44 −3.04 −2.72
10.63 6.06 5.38
55.27 −6.65 −6.28
stent surface and eventually alter the surface roughness. In addition, the experimental data was not obtained from the actual Palmaz or Sinusoidal stent, but it came from a test piece of tensile test specimen. As is known, there is only one directional stress acting in the tensile test specimen, i.e. axial stress. However, in the expanding and deflating process of stent, radial pressure loading could generate three directional stresses acting in the stent, i.e. radial, axial and circumferential stress [16]. Even though having equal material properties; elevated stress level could cause different grain boundary elongation, which ultimately would affect the magnitude of plastic strain and the changes in surface roughness comprehensively. More sophisticated expansion models will be discussed in future works. 5. Conclusion Deformation analysis using finite element simulation is particularly useful in identifying critical surface roughness of balloon-expanded stent, based on its plastic strain distribution. The results of the present study demonstrate that the plaque size relative to stent length significantly influences the critical surface roughness. It can be inferred for both type of stent that using shorter stent length than plaque length leads to decreasing the changes in surface roughness after balloon removal. This research demonstrated that the plaque length of type B, which had equal length to the stent, is the most recommended model to be implemented in the actual stenting due to the fact that it generated moderate change in surface roughness and corresponding stress concentration within the vessel wall. This information may be valuable for practitioners and doctors in choosing stent length and locating stent in the implantation process. In addition, the simulation results can also be used as a basis in designing initial surface roughness for improving stent performance and in performing stent coating at its critical surfaces. Acknowledgement The authors wish to express thanks to Kazuhiko Takahashi of the Laboratory of Micro Energy System (Division of Human Mechanical Systems and Design, Hokkaido University), for support and cooperation in the experiments of this study. Conflict of interest There are no actual or potential conflicts of interest.
202
A. Syaifudin et al. / Effects of plaque lengths on stent surface roughness
References [1] B. O’Brien and W. Carroll, The evolution of cardiovascular stent materials and surfaces in response to clinical drivers: a review, Acta Biomaterialia 5(4) (2009), 945–958. [2] V.A. DePalma, R.E. Baier, J.W. Ford, V.L. Gott and A. Furuse, Investigation of three-surface properties of several metals and their relation to blood compatibility, Journal of Biomedical Materials Research 6 (1972), 37–75. [3] J.F. Hecker and L.A. Scandrett, Roughness and thrombogenicity of the outer surface of intravascular catheters, Journal of Biomedical Materials Research 19(4) (1985), 381–395. [4] J.D. Denstedt, T.A. Wollin and G. Reid, Biomaterials used in urology: current issues of biocompatibility, infection, and encrustation, Journal of Endourology 12(6) (1998), 493–500. [5] J.C. Palmaz, A. Benson and E.A. Sprague, Influence of surface topography on endothelialization of intravascular metallic material, Journal of Vascular and Interventional Radiology 10(4) (1999), 439–444. [6] E. McLucas, M.T. Moran, Y. Rochev, W.M. Carroll and T.J. Smith, An investigation into the effect of surface roughness of stainless steel on human umbilical vein endothelial cell gene expression, Endothelium 13(1) (2006), 35–41. [7] G. Tepe, H.P. Wendel, S. Khorchidi, J. Schmehl, J. Wiskirchen, B. Pusich, C.D. Claussen and S.H. Duda, Thrombogenicity of various endovascular stent types: an in vitro evaluation, Journal of Vascular and Interventional Radiology 13(10) (2002), 1029–1035. [8] H. Zhao, J. Van Humbeeck, J. Sohier and I. de Scheerder, Electrochemical polishing of 316L stainless steel slotted tube coronary stents, Journal of Materials Science: Materials in Medicine 13(10) (2002), 911–916. [9] T. Hryniewicz, R. Rokicki and K. Rokosz, Surface characterization of AISI 316L biomaterials obtained by electropolishing in a magnetic field, Surface and Coatings Technology 202(9) (2008), 1668–1673. [10] T. Mizuno and H. Mulki, Changes in surface texture of zinc-coated steel sheets under plastic deformation, Wear 198(1,2) (1996), 176–184. [11] C.M. Wichern, B.C. De Cooman and C.J. Van Tyne, Surface roughness changes on a hot-dipped galvanized sheet steel during deformation at low strain levels, Acta Materialia 52(5) (2004), 1211–1222. [12] C.M. Wichern, B.C. De Cooman and C.J. Van Tyne, Surface roughness of a hot-dipped galvanized sheet steel as a function of deformation mode, Journal of Materials Processing Technology 160(3) (2005), 278–288. [13] H.A. Al-Qureshi, A.N. Klein and M.C. Fredel, Grain size and surface roughness effect on the instability strains in sheet metal stretching, Journal of Materials Processing Technology 170(1,2) (2005), 204–210. [14] O. Wouters, W.P. Vellinga, R. Van Tijum and J.Th.M. de Hosson, On the evolution of surface roughness during deformation of polycrystalline aluminum alloys, Acta Materialia 53(15) (2005), 4043–4050. [15] F. Ju, Z. Xia and K. Sasaki, On the finite element modelling of balloon-expandable stents, Journal of the Mechanical Behavior of Biomedical Materials 1(1) (2008), 86–95. [16] S.N. David Chua, B.J. Mac Donald and M.S.J. Hashmi, Finite element simulation of slotted tube (stent) with the presence of plaque and artery by balloon expansion, Journal of Materials Processing Technology 155–156 (2004), 1772–1779. [17] I. Pericevic et al., The influence of plaque composition on underlying arterial wall stress during stent expansion: the case for lesion-specific stents, Medical Engineering and Physics 31 (2009), 428–433. [18] M.A. Jiménez, J.M. Bielsa, R. Rodríguez and S. Dobón, The influence of contact pressure on the dynamic friction coefficient in cylindrical rubber–metal contact geometries, IUTAM Bookseries of Symposium on Computational Methods in Contact Mechanics 3 (2007), 257–275. [19] J. Mummert, A. Reynolds, E. Sirois and W. Sun, Determination of radial force and coefficient of friction with a self-expanding transcatheter aortic valve stent, in: Bioengineering Conference (NEBEC) 38th Annual Northeast, 2012, pp. 189–190. [20] A.C. Dunn, T.D. Zaveri, B.G. Keselowsky and W.G. Sawyer, Macroscopic friction coefficient measurements on living endothelial cells, Tribology Letters 27(2) (2007), 233–238. [21] S.N. David Chua, B.J. Mac Donald and M.S.J. Hashmi, Finite element simulation of stent and balloon interaction, Journal of Material Processing Technology 143–144 (2003), 591–597. [22] C. Lally, F. Dolan and P.J. Prendergast, Elastic behavior of porcine coronary artery tissue under uniaxial and equibiaxial tension, Annals of Biomedical Engineering 32(10) (2004), 1355–1364. [23] C. Lally, F. Dolan and P.J. Prendergast, Cardiovascular stent design and vessel stresses: a finite element analysis, Journal of Biomechanics 38(8) (2005), 1574–1581. [24] F. Migliavacca, L. Petrini, M. Colombo, F. Auricchio and R. Pietrabissa, Mechanical behavior of coronary stents investigated through the finite element method, Journal of Biomechanics 35(6) (2002), 803–811. [25] ANSYS Inc., ANSYS Mechanical APDL Element Reference, 2012.
Copyright of Bio-Medical Materials & Engineering is the property of IOS Press and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
