Read 019.pdf text version

BloodCompatibilityofNitinolComparedtoStainlessSteel Thierry,Merhi,Bilodeau,Yahia,Tabrizian ProceedingsoftheInt'lConferenceonShapeMemoryandSuperelasticTechnologies SMST2000 (eds.)S.Russell,A.Pelton pp.285290 2000

We are Nitinol.TM 47533 Westinghouse Drive Fremont, California 94539 t 510.683.2000 f 510.683.2001


B. Thierry,· Y. Merhi. 2 C. Trepanier,J L. Bi lodcau,2 L. H. Yahia,l and M. Tabrizianl

Engineering /nsritwe-BioIlUllerials/8iomedwmics RI!Xt!MCh Group Mechanical Ellgi/leeri/lg Deparrmellt. Ecole Polyreclll1iqllt of Montreal Quebec. H3C 3A7. Cwwda, l Molllrea/ Heart Instill/re, 5000 Bellinger. MOil/real, QI/ebee, H IT I CB, Callo(l" JCordis Corpomtioll-Nirillo! Dellice.~ & Compol/ems. 47533 Westinghollse Drive,

FremOIll. CA 94539

I Biomedical


Becliuse of its superelasticity, shape memory. corrosion resistance. and biocompatibility. Nitinol is becomi ng increasingly popular for minimally invasi\'e devices such as endoluminal stenlS. Despite several studies o n ill vitro or ;/1 vivo biocompatibiliy of NiTi. few studies have been conducted on the interactions of the material with blood. In this study. blood compatibility telits were conducted on NitinoJ and stainless steel stems using an ex \·;IIQ. AV-shunt porcine model. We have demonsmned thilt Nitinol is significantly less thrombogenic than stainless steel as indicated by 1251_ human fi brinogen (p = 0.03) and I III_platelets (p = 0 .01) quant ification. These differences may be related to the Nitinoltitanium-oxide rich surface layer that may prevem denaturation of fi brinogen and minimize platelet-rich thrombus fonnati on with in the stem after impl:mtation.


Si nce the implantation of the first human coronary stent in 1986. many dev ices manufactured rrom different materials (stainless steel. tantalum. Nitinol. etc.). exhibiting various designs have entered the market. Along with the use of a bener anti platelet and amicoagulant therapy. optimal stent deployment has promoted stent implantation as a procedure of choice by red uci ng the risk of acute thrombosis and chronic restenosis. To achieve such an optimal deployment. conventional balloon-cxpandable stainless steel stents require high-pressure dilatation to plastically defoml their structure in the atherosclerotic vesscl, which may increase the level of vessel injury and endothelial denudation. and in tum enhance neointimal proliferation within the slents [II. Moreover. longitudinal Shortening during expansion and lack of elasticity remain limitations associmed to the balloon-expandable stems. In contrast. Nitinol's supcrelastic properties can be favorably used 10 design self-ex panding devices thai minimize the need for balloon postdilatation and allow a good self-expansion ratio and a more uniform radial expansion with less longitudinal shortening [2-4). In addition. Nitinol is characterized by biocompalibility and corrosion resislance comparable or superior to stainless steel 15-7). These propenies promOlc its uses



for stenting in peripheral artery disease (renal. carotid. iliac. and subclavian artery diseases. aon ic intervenlion. and femoral PTA) through the development of such self-expanding devices. Despite several studies on ill "jim or ill 1';1'0 biocompat ibility of Nitinol. few studies have been conducted on the interactions of the material wi th blood. Even though cl inical d~lIa are missing o n this point. differences in haemocompatibility can be expeCioo to have consequences on the clinical

thrombogenic occlusion and/or reslenosis rates. In the present study, we have investigated the

thrombogcnici ty of NiTi stcnts in comparison to Stai nless Steel (SS) Sl en L'i. To achieve this goal. an

ex villo. AV-shunt porci ne model was used 10 measure fibrinogen adsorpt ion as well as platelet

adhesion on both devices.



The NiTi stems were prototype devices manufactured by Cordis Corpomtion-N itinol Devices & ComlXlnents (Calif.. U.S.). They consisted of 3 mm di ameter by 30 mOl lo ng stents that were lasercut fro m NiTi tubing to re plicate the Palmaz 3l6L stainless steel ste m geometry (P294M. Cordis Corp.). Palmaz 3 16L 5S stents wcre used as reference.


All procedures fo llowed the American Heart Association Guidelines for Animal Research and were approved by the An imal Ethics Committee of the Montreal Heart Institute. Experiments were performed using six pigs weighing 25 ± 3 kg. Animal platelets were isolated and radio labeled with Indium-I I J (11 11n oxi ne. Merck FroSl Canada Inc .. Canada) as described previously 181. The quanl ifi calion of fibri nogen deposition was done by injection of approximately 10 J.lCi 1251-human fibrinoge n (Amersham International. U.K.) one hour before the experiment. Prior to each perfusion, IwO stents were inserted manually in two Silastic tu bes using a sterile fil ament to position each device. The tu bes were 118 inch Internal Diameter (10) and 9 em long. A 3 by 20 mm conventional semicompliant balloon catheter was then insened in each tube and inflaled at 12 atm to deploy the stem in the tubing. Even though the Nitinol stents were self-ex panding devices and did not need to be deployed usi ng a balloon catheter, they were inserted using the same method as the S5 stenls to avoid variation in manipulation of each device. The extracorporeal AV shu nt consisted of a silicon tubi ng circ uit connecting the left fe moral artery to the right femora l vein through the perfusion channe ls (see Figure 1). The main extracorporeal AV shunt was d ivided in two paralle l tubi ng systems inside Ihe perfusion chamber and was connected to the stented tubes. During the experiment. the blood flow in the main circuit was maintained at a stable role of 160 mUm;n using a roller pump. The perfusion chamber was immersed in a 37± 1°C water bath. One hour after re-i njection of the labeled platelets and fibrinoge n. the stems (one SS and one NiTi) were set in each channe l of the cxtmcorporeal c ircuit and rinsed for 20 seconds usi ng saline solution. Blood was then allowed to c ircu late for 15 minutes at a wall-shear rate of 456 S- I (oorresponding to 80 mUmin). This wall-shear rate represents the nonnal human shear rale in large- to medium-siz.ed aneries 19J. At the end of the perfusion. circulalion of saline solution was done 10 remove unattached cells and blood from the stents and the perfusion circuit. The stented tubi ng segments were CUt at both e nds and removed from lhe circui t. A 1.5 cm long segment althe distal end of each lube was CUI and used as control. Before testing the next series o f stents. the



,------, Roller

Ste t


maintained at 37 ± 1°C

Figure J Sci/eli/otic represelllarioll of the ex vivo AV-slwilf model.

extrncorporeal circui t was washed again with saline. All tubing segments were fixed in 1.5% glutaraldehyde solut ion and processed for quantifical ion of platelet adhesion and fi brinogen adsorption . Using a gamma counter. the amount of II lin-platelet and 1251 fibrinogen was quantified by measuring the radioactivity of each tubing segment (8J.


Results are expressed as mean value ± SO. The data in tenns of fibrinogen adsorption and platelet adhesion were analyzed using paired SlUdent-T tests. A p-value of less than 0.05 was considered significant.


After Ihe perfusion, macroscopic analysis of the Siems revealed obvious differences between NiTi and SS. NiTi stents had only small amounts ofwhitc and/or red thrombus. princi pally localed 011 the strut intersections. SS stcnts. however, clearly exh ibited more thrombus. Typical thrombus observed on NiTi and SS stcnts is presented in Figure 2. FurthemlOre. one SS stem was occluded after 15 minutes of perfusion. These macroscopic observations were confinncd by !:iubsequent quantification of the radio-labeled platelets and fibrinogcn. Results indicated that the nature of thc stent material played a major role on the quantity of fi brinogen and platelets deposited on the devices. t251_fibrinogen adsorption was significantly lowcron Nitinol than on SS (p = 0.03) (see Figure 3). The fibri nogen count averolgcd 3653 ± 913 cpm/stcn! on Nilinol stents and 5707 ± 1556 cpm/slent on SS stents. Nitinol device~ hlld also significantly less platelet adhesion on their surface than had SS Slcnts (p 0.01) (see Figure 4). The mean quantity of platelets was 925 ± 248 x 106 platelets/slent for thc Nitinol group while the quantity of platelet averaged 2526 ± 770 x 106 platelets/stent o n SS. Control tubing segments did not promote significant fibrinogen adsorption (23.8 ± 7.3 cpm/tube) or plalelels adhesion (2.96 ± 0.73 x 106 platelets/tube).


Based on our results. Nilinol exhibits a lower acute thrombogenicity than SS. Indeed. the amount of labeled fibrinogen adsorption and platelel dcposition on Nilinol stents aftcr 15 minutes of perfusion was significantly lower: 36% and 63%. respectively on Nitinol compared 1 SS stents. In o ur study. 0


Figure 2 Ni7i and 55 .~leIllS a/ler !)erfusion.

8000 T

Fibrinogen 7000 adsorption (cpmlstents) 6000 -

5000 4000 · 3000 2000 10000Control




Figure 3 J2S (-fibrinogell adso'1)fion on Nil;l1ol SIems, 5S sIems, alld Silaslic lubes (ColI/ rol) afler 15 mil/IlleS o/peifll.fioll ( "p < O .OOll's. COII/rol),

the effect of the design and the surface fini sh were eliminated as device-related variables since both slenlS exhibited the exact same geometry and similar smooth and unifonn e leclropolished surface. Stent surface composition and topography will determine the n,lIure of the protein-adsorbed layer. which will affect thrombus formation following Slenl implantation IIO}. Previous studies have reponed Ihlll the surface of clcctropolished SS is mainly covered by a chromium- and iron-rich oxide 171. On Ihe olher hand. eleclropolished Nilinol has been shown 10 be covered by a lilaniumrich oxide layer (mainly 1102) simi lar to the oxide on tilanium alloys. which is recogn ized for ils good haemocompalibililY 17.1 11 . Nygren. el al. have shown thai depending on the charnclerislics of the Ti0 2 surfaces. the Icvels of surface-adsorbed plasma proteins such as fibrinogen and of plalclcls were significantly different. Funhennore. thrombus formation on biomaterial surfaces is also con-



Plateletdeposition (10.fstent)


3000 -


2SOO T 2000 r 1500 1000 . SOO 0

Contral Nitinal Stainless steel

Figllre 4 II JIII -platelet deposition on Nilino! stems, mi liUl e.f ofperfl.l.~ioll (*P < 0.001 VS. C 01l1rol).

SS stems, Gnd Sila.I·lic tubes (C Olltrol) afler J5

trolled by the denaturation of fibrinogen into fibrin monomers and fibrinopeptides (10). This event may rely on an e lectron exchange from occupied valence band states a f the fibrinogen to the surface of the stentl1 2J . Since thrombus growth depends on the polymerization of fibrin monomers incorporating activated phl1elets. prevention of such denaturation can improve the haemocompatibility. II has been recently suggested that a Ti0 2_ -ox ide film may prevent the denaturation of fibrin ogen in a similar way (13 ). x Based on o ur results_ the tilani um-rich oxide surface of Ni tinol stems may have minimized the formation of fibrin-platelet rich thrombus. Previous ill vitro studies have reported similar quantities of plateletlldhesion between NiTi and SS in contact with platelet-rich plasma in absence of fibrinogen f I4. 15) . These results may be explained by the differe nt test models that were used in these studies. They also emphasize the importanl role of fibrin ogen and its denaturation to fibrin in the thrombusformation processe.<; in dynamic fl ow condi tion. Our results are in agreement with those obtained by Sheth, et al. using a r'ol.bbit carotid artery model 121. They reported lower thrombogenicity of Nil ino l stents compared 10 SS stents. St ill. because stent surface fini sh. design, and deployment mechanism were d ifferent for the Nitino l and SS stents. their results could nOI ctearly show the individual effcct of the material composition on thrombogenicity. Indeed, all these parameters have been shown to be very important fa ctors in the modulation of blood-biomaterial response f 16 ). It is important to e mphasi7-c that in our study use of similar ste nt geometry and surface fin ish allowed us to isolate the effect of material composition on blood interaction.


This study was conducted to assess the relative thrombogenici ty of Niti no l and SS stcnts in an ex vivo. AV-shum porcine model. NiT! siems manufactured 10 replicate the geometry and the surface fin ish of the Palmaz stainless steel stem were tested. Our resu lts show that Nitinol is significantly less thrombogenic than SS-based on 1 1SI_human fibrinogen (p = 0.03) and I Ill_platelets (p 0.0 1) quantification. These differences may be. related to a Nitinoltitanium-oxide rich surface layer thaI may prevent denaturation of fibrinogen and minimize platelet-rich thrombus formation within the stent after implantation.




The authors wish to thank J. F. Theoret from the Montreal Heart Institute for his technical assistance.



S. L. Goldberg. el al" Am. 1. Cordiol. 81 (1998),708. S. Sheth, e l al.· eircl/llI/ioll 94 {I996). 1733.


3. 4.


E. Rechavia. et a l.. Carll. Cardiomsc. Diagn. 41 (1997), 85. A. J. Carter. et al.. Calli. Can/iova.f e. Diagn. 44 (1 998). 193.

J. Ryhanen. Bicx:ompatibilil)' evaillotion of "ickel·tita"ium shape memory mctal alloy (Ou lu. Finland: Oulu University Press. 1999).

R. VcnugopaJan and



C. Trepanier, M in.

III\'. Tiler. alld Allied Tech. (forthcoming).

B. Thierry, el aI. , 1. Biomed. Mater. Re.f. (forthcoming).



Y. Mcrhi. M. King. and R. Guidoin, 1. Biamed. Mater. Res. 34 (1997), 477 .

S. R. Hanson and K. S. Sakariassen. Am. Heart J. 135, no. 5, pI. 2, supp!. ( 1998), S 132.

10 . H. Nygren . C. Erikson. and J. Lausma, J. Lab. c/ill. Med. 129 ( 1997), 35.

I I. C. Trepanier. et al.. J. Biollled. Mat. Res. 43 ( 1998), 423.

12. P. Baursc hmidt and M. Schaldach. J. Bioellgilleerillg 1 (1977). 261. 13. H. Nan. el al .. Biollluterials 19 ( 1998), 77 1. 14. C. S. Sunon. P. M. Consigny, and M. Thakur. Circulario" 90. suppl. I (1994), 1-9. 15. D. A. Annitage. e! al.. in SMST·97: Proceedings of the Second I/ltenla/ional Conference Otl Shape Memory {/lid SlIperelasric Technologies. ed. AR. PelIon. et al. (Pacific Grove. Califor· nia: International O rganization on SMST. 1997).411. 16. C. R. Rogers and E. R. Edelman. Circl/laliotl 91 (1995). 2995.


7 pages

Find more like this

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate