Subacute Thrombosis and Vascular Injury Resulting From Slotted-Tube Nitinol and Stainless Steel Stents in a Rabbit Carotid Artery Model
Background Our objectives were to quantify the thrombogenicity and extent of vascular injury created by slotted-tube geometry stainless steel and nitinol coronary stents in a rabbit carotid artery model.
Methods and Results Stents were implanted in rabbit right carotid arteries without antiplatelet therapy. Stainless steel stents were implanted for 4 days while nitinol stents were placed for 4 and 14 days (n=8, 8, and 6, respectively). Stent thrombosis was assessed by thrombus weight, grading thrombus encroachment of the lumen, and by blood flow in the stented and contralateral arteries. Stainless steel stents at 4 days contained more thrombus than 4- and 14-day nitinol stents (20.0±5.9 versus 2.5±0.6 and 2.7±0.3 mg, respectively; P<.000001). Stainless steel stents were more often occluded by thrombus (6 of 8) or contained more subocclusive thrombus (2 of 8) than nitinol stents (0 of 14, P<.002). Resting blood flow was reduced in arteries with stainless steel stents compared with 4- and 14-day nitinol stents (1.5±2.8 versus 24.0±2.0 and 25.5±1.9 mL/min, respectively, P<.000001). Stainless steel stents were less uniformly expanded, had deeper strut penetration into the vascular wall, and were associated with more extensive medial smooth muscle cell necrosis. There were strong correlations (r=.77 to .95) between variables of thrombosis extent (thrombus weight and grade) and histologically determined vascular injury (strut penetration and medial necrosis).
Conclusions Slotted-tube stainless steel stents were more thrombogenic and created more extensive vascular injury than nitinol stents in a rabbit carotid artery model. The mechanisms underlying these differences probably are related to metallurgic and design geometry properties of the two stent types.
Coronary stenting is an important advance in interventional cardiology, with clinical trials demonstrating safety and efficacy for the treatment of acute or threatened coronary occlusion, reduction in angiographic restenosis rates, and improved patency of anatomically suitable saphenous vein graft lesions.1 2 3 4 5 The major limitation of stenting is a new syndrome: subacute stent thrombosis. The syndrome is manifest by acute coronary ischemia typically occurring 1 to 12 days (mean, 5±3) after implantation and has been observed with all stent designs thus far tested.6 7 8 9 10 11 12 Subacute thrombosis rates are 3% to 5% after elective stenting of ideal native coronary lesions.2 3 5 Predisposing factors include acute or threatened closure or discontinuation of anticoagulation therapy. In such cases, stent thrombosis rates have been reported to range from 7% to 16%.1 6 7 8 9 Moreover, data from a porcine stent model of restenosis suggest that early thrombus formation may stimulate neointimal proliferation.13 14 Approaches to limit stent thrombosis have included optimizing stent deployment with high-pressure inflations and intravascular ultrasound guidance,15 16 addition of more powerful antiplatelet drugs,17 18 and thromboresistant stent coatings.19 Animal model testing of different stent designs suggests that geometry influences the frequency thrombotic occlusion.20
We have developed a slotted-tube, balloon-expandable coronary stent fabricated from nitinol, a nickel titanium alloy. This alloy was selected for its unique metallurgic characteristics, which include the option of percutaneous removal by thermoelastic shape-memory recovery to its predeployment geometry; favorable tissue and blood biocompatibility characteristics; and a stress-strain relationship that allows uniform stent expansion at low pressure.21 22 23 24 This latter feature forms the basis for our experimental hypothesis: A stent that expands uniformly at low pressure may cause less vessel injury during deployment and therefore may be associated with less thrombus formation.
The goals of this study were threefold: first, to quantify the relative thrombogenicity of a slotted-tube nitinol stent and compare this with a commercially available stainless steel stent of similar design in a rabbit carotid artery model; second, to assess the extent to which the differing devices effect vascular injury; and finally, to correlate the extent of vascular injury with the development of stent thrombosis.
Palmaz-Schatz coronary stents (Johnson & Johnson Interventional Systems) composed of 316L stainless steel were divided in two at the articulated bridge and the cut ends were ground smooth. This resulted in monomeric stents 7 mm in length, with a wall thickness of 0.0025 in, and a mass of 8.0±0.2 mg.
Prototype slotted-tube nitinol stents (Advanced Coronary Technologies) also composed of two segments of 7 mm joined by a 1-mm bridge were similarly prepared. The nitinol stents were heavier (18.5±0.4 mg) and had thicker walls (0.007 in). The two stent types differed in several other respects. The slot length of the nitinol stent was approximately twice as long (see Fig 1⇓) and its surface area (inner, outer, and edge surfaces) 80% greater than the stainless steel stents. At 3.0-mm expansion, the nitinol stent covers 36% of the arterial wall compared with 20% for the stainless steel device. The nitinol stent has 18 loci where struts interconnect, whereas the stainless steel stents have 30. The surface of the stainless steel stents was brightly polished, whereas the nitinol stents had a duller matte finish. All stents were mechanically crimped on identical 3.0-mm-diameter, 20-mm-long stent delivery balloon catheters (Progressive Angioplasty Systems, Inc).
Animal experiments conformed to the guiding principles of the American Physiological Society and were approved by the Cedars-Sinai Medical Center Institutional Animal Care and Use Committee. Three groups of rabbits were studied: group A (n=8) had stainless steel stents implanted with euthanasia at 4 days; group B (n=8) had nitinol stents implanted with euthanasia at 4 days, and group C (n=6) had nitinol stents implanted with euthanasia at 14 days.
Normolipemic fasting male New Zealand White rabbits, weighing 3 to 4 kg, were anesthetized with intravenous xylazine and ketamine. A 6F sheath was inserted into the left femoral artery by cutdown, and 500 U of heparin was injected. No other anticoagulant or antiplatelet therapy was given either before or after stent implantation. A single stent/balloon delivery system was introduced over a 0.014-in guide wire and advanced to the right carotid artery. The balloon was inflated twice to 6 atm for 1 minute to deploy the stent. The femoral incision was closed, and the animals were maintained on a standard diet.
At the scheduled time of euthanasia, rabbits were reanesthetized, both carotid arteries were surgically exposed, and phasic flow was measured by transit-time probes (Transonic Systems, Inc). Stents were radiographically imaged on 35-mm cinefilm (Advantx-DXC GE Medical Systems) in orthogonal projections, with highly collimated exposures and a 4.5-in image intensifier field size. A 9F contrast-filled catheter was placed adjacent to each stent for image calibration. Stent outer diameters were measured at the mid portion and at each end. Cross-sectional area calculations assumed that orthogonal diameters were the major and minor axes of an ellipse.
Rabbits were euthanatized under anesthesia. The stented vessels were explanted and photographed. The vessels were incised and the stent plus all visible thrombus removed and placed overnight in a vacuum oven at 40°C. Dried specimens were weighed and photographed. Stents were examined under a dissecting microscope and a semiquantitative ordinal thrombus grade (T) assigned as follows: T1, minimal thrombus without lumen encroachment; T2, small thrombus with encroachment; T3, subocclusive thrombus ≥75% of lumen area obstructed; and T4, totally occlusive thrombus.
Thus, four independently measured indices of thrombus severity and vascular patency were obtained: (1) blood flow in the stented artery, (2) blood flow in the contralateral carotid artery, (3) thrombus dry weight, and (4) semiquantitative grading of the luminal encroachment by thrombus.
Scanning electron microscopy was performed on two stents from each of the three groups. The stented segments of carotid arteries were placed in formalin, cut, embedded in paraffin, sectioned, and stained with hematoxylin and eosin. Two sections from each stented region were assessed by an experienced cardiovascular pathologist (M.C.F.) who was blinded to the stent type and the data quantifying extent of stent thrombosis. Vessels were graded on a three-point scale to assess the degree of stent strut penetration (P), where P1 was assigned if the internal elastic lamina remained intact; P2 if at least one strut penetrated the internal elastic lamina and was embedded in the media; and P3 if at least one strut transected the arterial media. The extent of medial necrosis (N) was recorded, where N1 was assigned if vascular smooth muscle cell necrosis (loss of nuclear staining) was absent or confined to the region contacted by an individual stent strut; N2 if patches of necrosis were seen in the inter strut segments; and N3 if ≥50% of the vessel media circumference was necrotic.
Visualization of Stent Expansion
In one additional anesthetized rabbit, a stainless steel stent and a nitinol stent were positioned in the right and left carotid arteries as previously described. A cutdown was performed on the neck to expose both carotid arteries. Expansion of the two stents was visualized and photographed external to the vessel at stepwise increments of balloon pressurization.
Data are presented as mean±SD. For interval scaled data, comparisons were made between the three study groups with the use of ANOVA with an implementation of the general linear model. Intergroup post hoc testing was by the Tukey honest significant difference test for unequal sample sizes. For ordinal scaled data, comparisons between the three groups were made by Kruskal-Wallis ANOVA and pairwise comparisons were by the Mann-Whitney U test adjusted for multiple comparisons. The relationship between thrombosis and vascular injury variables was assessed by Spearman's rank correlation coefficient. Analysis was performed with the use of Statistica V4.2A (Statsoft).
Table 1⇓ summarizes the four independently measured variables of stent thrombosis. Carotid blood flow in the treated artery was significantly reduced in the group with stainless steel stents implanted for 4 days compared with nitinol stents for 4 and 14 days (1.5±2.8 versus 24.0±2.0 and 25.5±1.9 mL/min, respectively). In 6 of 8 cases, there was no measurable flow through the stainless steel stents, and in the remaining two cases resting blood flow was sharply reduced at 6.0 mL/min.
Blood flow in the contralateral artery was moderately but significantly elevated in rabbits with stainless steel stents compared with both nitinol stent groups. When nitinol stents were implanted, there was no difference between the stented and the contralateral artery blood flow (P=.88)
Thrombus weight was sevenfold to eightfold greater in the group with stainless steel stents compared with nitinol stents implanted for 4 and 14 days (20.0±5.9 versus 2.5±0.6 and 2.7±0.3 mg, respectively). There was no statistical difference in thrombus weight for nitinol stents examined after 4 or 14 days (P=.79). The sample sizes were sufficient to detect a 1-mg difference at 90% power.
All nitinol stents were graded T1 (minimal thrombus). In contrast, stainless steel stents were rated T3 thrombus (subocclusive) in 2 and T4 thrombus (occlusive) in the remaining 6 rabbits. Fig 1⇑ shows representative photographs of the explanted stents from the three experimental groups. Stainless steel stents were nonuniformly expanded and filled with adherent red thrombus. Nitinol stents were more uniformly expanded and had small patches of white or mixed red and white thrombus located principally at the strut intersections.
Fig 2⇓ shows scanning electron micrographs of the luminal surface of both stent types. Stainless steel stents had confluent masses of red cells and fibrin. The nitinol stents had a rougher surface topography (consistent with the known bare surface of the device) with scattered clumps of platelets, red cells, and fibrin. At 14 days, nitinol stents showed a thin covering of neointima.
Table 2⇓ summarizes the extent of stent expansion and vessel injury for the two types of stents. The ratio of large to smaller diameters and stent cross-sectional areas measured in the proximal, mid, and distal regions of the stents were nearly identical for both stent types. Histological grading of the stented vascular segments showed marked differences between groups. Stainless steel stents were associated with higher grades of stent strut penetration injury and medial necrosis than nitinol stents. Fig 3⇓ shows representative histological sections from the three stent groups. All eight vessels with stainless steel stents were graded as P3 with at least one strut transecting the media and N3 with >50% of the vessel circumference exhibiting medial necrosis. In all cases, the medial necrosis was transmural with thinning of the wall, scattered acute inflammatory infiltrates, and mural thrombi. In comparison, nitinol stents, whether implanted for 4 or 14 days, were associated with significantly less strut penetration and medial necrosis. Vessel histology in the 14-day nitinol stent group demonstrated neointimal proliferation adjacent to the sites of stent strut contact.
Irrespective of the type of stent or duration, the severity of medial necrosis correlated closely with the depth of stent strut injury (r=.91, P<.000001). There were also strong relationships (r=.77 to .95, P<.00002) between independently measured variables of thrombosis extent (thrombus weight and grade) and histologically determined vascular injury (strut penetration and medial necrosis).
Fig 4⇓ shows the two stent types at stepwise increments of balloon pressurization visualized from the external surface of the carotid arteries. The stainless steel stent first expanded at 2 atm by outward flaring of the distal ends, with hemorrhage immediately developing at these sites. The mid portion of the stent fully expanded at 5 atm, but several slots remained unopened. The nitinol stent did not flare at the ends or initiate hemorrhage and was fully and uniformly expanded at 2 atm.
Our study showed that under conditions used in this model, slotted-tube stainless steel stents implanted in normal rabbit carotid arteries for 4 days contained significantly more thrombus by weight, extent of lumen encroachment, and functional effect on blood flow than similarly configured and implanted nitinol stents. The thrombus burden on nitinol stents was similar at 4 and 14 days after implantation, suggesting that delayed thrombosis was unlikely. Although stents were expanded to similar diameters, stainless steel stents were less uniformly expanded and caused significantly deeper arterial wall injury and more extensive vascular medial necrosis. There were strong correlations between the extent of thrombosis measured by weight or degree of luminal encroachment and histological evidence of stent-mediated vascular injury assessed by the depth of strut penetration or medial smooth muscle cell necrosis. These correlations suggest the hypothesis that penetrating and necrotizing vascular injury may predispose to the development of subacute stent thrombosis.
Extent of Vascular Injury
Studies of the mechanisms underlying stent thrombosis have focused largely on the interactions of blood constituents with bioprosthetic surfaces.24 25 The present study shows that stent thrombosis is associated with the extent that a specific stent type injures the arterial media. Although both stainless steel and nitinol slotted-tube stents were expanded under identical circumstances and measured to have similar postdeployment dimensions, the nitinol stents were less deeply embedded into the arterial wall and were associated with significantly less medial necrosis. This begs the question, why should two stents with similar geometry create such different patterns of vascular injury? We believe the answer lies in the different expansion mechanics of the two devices. Mechanical differences are suggested by the comparative appearance of the explanted stents and the visualization of stent expansion in exposed carotid arteries. Stainless steel stents did not expand uniformly. In each instance, some of the struts were twisted, the stent ends were flared outward, and several slots were not expanded. In comparison, nitinol stents were more symmetrically expanded with respect to both radial and axial orientations.
Rogers and Edelman20 compared stainless steel stents with two geometries: slotted-tube and corrugated ring–type stents at 2 weeks after implantation in denuded rabbit iliac arteries. Both stent configurations were of identical mass, surface area, and surface finish. Both groups were aspirin-treated. Like the nitinol stents in our study, the corrugated ring stents had 29% fewer strut-strut intersections and did not flare at the ends. Similarly, the ring stents were associated with significantly less vascular injury, having a 42% lower vascular injury score; significantly less thrombosis (15% versus 42%); and significantly less neointimal hyperplasia. Whether the reduced vascular injury observed with nitinol slotted-tube stents will result in limited late restenosis was not determined and will require separate investigation.
Other possibilities for differences in thrombotic potential between the two stents include the interaction of the metal surface with blood and tissue procoagulant components. Although theoretically possible, the literature thus far shows that nitinol and stainless steel have a similar propensity for thrombosis.22 24
The differing behaviors of the two stent types during expansion can be explained in terms of the physical properties of the respective metal alloys and subtle but important differences in design geometry. Mechanical deformation of a physical solid material such as a metallic stent can be analyzed by generating a stress-strain diagram.26 Fig 5⇓ shows the stress-strain relationship for elongation of stainless steel and the type of nitinol used in our study. An expanding force such as a balloon creates tension in a metal, known as the apparent stress (force per original unit cross-sectional area). In response to the stress, deformation of the metal occurs. Strain is a measure of the fractional deformation. Depending on the direction of the applied stress, strain can be measured as fractional expansion or compression of length, volume, or twisting (shear). In a rough sense, stress and strain can be thought of as analogous to balloon expansion pressure and stent diameter, respectively.27 In actuality, for a given loading stress, strain varies throughout the meshwork, being greatest at the locations undergoing the most deformation. Strain can be mathematically modeled for complex structures such as stents by finite element analysis. For slotted tubes, the location of the greatest strain and therefore the greatest stress is at the slot ends on the luminal surface.
Stainless steel behaves as typical ductile alloy. Initial stress results in elastic deformation (vertical or linear portion of the curve). Once the elastic limit is exceeded, the alloy yields and considerable increases in strain are achieved. This increase in strain where the metal appears to flow like a viscous liquid is called plastic deformation and allows the material to acquire a “permanent set.” Nearly all of the expansion of slotted-tube stainless steel stents is by plastic deformation. The Palmaz-Schatz stent ends expand first because they require less stress to deform than the middle of each stent where two slots are adjoined, forming four corners (location of the highest strain). The stent ends flare outward to minimize the internal forces created by the stress. The stent ends penetrate into the arterial wall, securely anchoring the device but simultaneously creating deep wall injury. Small variations in stent strut thickness or inhomogeneous application of expanding stress results in nonuniform expansion of the slots. High pressure is required to achieve the desired strain needed for complete stent expansion.
Nitinol is unique among metallic alloys in that it exists in two crystalline phases called martensite and austinite.28 The nitinol stent we used remains in the martensite at body temperature. The martensitic stress-strain relationship is very different from other ductile alloys. Early deformation is linearly proportional to applied stress. Thereafter, nitinol becomes extremely ductile and can be further deformed up to ≈6%, with very little increase in stress. This deformation is said to be thermoelastic because upon heating to a specified temperature the material undergoes a rapid transition to the austinite crystalline phase, resulting in collapse of the stent to its preexpanded geometry. Strain ≈>8% results in plastic (unrecoverable) deformation of the martensite crystalline structure. The longer slot length was chosen to minimize the achieved strain at the corners. Finite element analysis of the nitinol stent shows that the majority of the strut wall thickness does not reach the maximum recovery strain. This has been demonstrated experimentally by complete thermoelastic recovery of the stent even after expansion to 6 mm in diameter.
Thus, for the martensitic nitinol stent, only very small changes in balloon pressure are required to initially expand all parts of the stent. Small differences in strut thickness or applied stress will have less of an effect on early expansion than with stainless steel. This results in a stent with uniform radial and axial expansion at low balloon pressures. Moreover, because the opening stress is so low, a much thicker and therefore radiopaque stent can be made. High balloon pressures may be required to fully expand all parts of the stent up to the limit of the maximum recovery strain. These metallurgical and design geometry differences create a device that more uniformly expands at lower pressures, does not flare at the ends, and thus may result in less vascular injury.
Our study has several limitations that should be considered before making inferences about the mechanisms of subacute stent thrombosis. The resting flow velocity in rabbit carotid arteries is typically lower than native coronary vessels of similar size. The present study thus excludes exposure to high blood shear, which is an important activator of platelets.29 Stents were implanted in normal vessels rather than a lipemic atherosclerotic model or a restenosis model. The influence of fatty plaques or preexisting vascular injury on subacute thrombosis cannot be predicted from the current data.
Expansion of the nitinol stent may not be as uniform in atherosclerotic vessels. The spatially heterogeneous viscoelastic properties of a diseased vessel result in less uniform balloon expansion, which in turn determines the uniformity of stent expansion. Nonuniform balloon expansion would, however, have an even more pronounced affect on stainless steel slotted tubes for the reasons described in the previous section on mechanical engineering aspects. The crush resistance or hoop strength of the stent may also affect expansion uniformity in rigid, fibrotic segments. Stress versus strain analysis has demonstrated that the two stent types have similar crush resistance whether in response to one-sided compression with a flat plate or circumferential compression with an iris.
It was not possible to study stents with identical geometry to discern the effect of each alloy. A stainless steel stent with a wall thickness the same as the nitinol stent would require much higher pressures to expand. Conversely, a nitinol stent with a wall thickness and slot length identical to the Palmaz-Schatz stent would be plastically deformed during deployment. We therefore chose to evaluate the thrombotic potential of one stent design relative to another stent design.
The time course of stent thrombosis was not specifically evaluated in this study. The stainless steel stents may have thrombosed earlier than 4 days after implantation. Although we cannot exclude this possibility, preliminary experiments in the same animal model with a different stent design showed less thrombosis at 1 and 3 days compared with 4 days.30 Alternatively, the patency of stainless steel stents at 14 days was not evaluated. Although it is conceivable that some stents may recanalize after complete thrombosis, Rogers and Edelman20 showed that 42% of stainless steel slotted-tube stents remain thrombotically occluded at 2 weeks in the rabbit, even with daily administration of aspirin.
The lack of aspirin in our model is a potential limitation for extrapolating the current data to clinical application in which aspirin is used in conjunction with other antiplatelet or antithrombin agents. From the animal model just cited, it is apparent that aspirin as a sole agent is not very effective for preventing thrombosis of stainless steel slotted-tube stents. The dose of aspirin that is effective for preventing stent thrombosis has not been systematically defined. In our experience, oral administration of aspirin in rabbits, whether dissolved in drinking water or by pill gun instillation, is unreliable because of frequent regurgitation. We have repeatedly found that some “aspirin-treated” rabbits have undetectable blood salicylate levels. Thus, we chose not to use aspirin because it was unlikely to substantially alter the experimental results and could create an additional source of variability.
Stent thrombosis appears to be modulated by several mechanisms including incomplete stent expansion, vascular injury, hemocompatibility of the device, local flow characteristics (shear rate, dissections, distal runoff), and systemic factors such as the effectiveness of platelet or thrombin inhibition. The present study shows that a nitinol stent, which is more completely expanded at low pressures and causes less vascular injury, has less thrombosis. The relative contribution of each of these mechanisms to thrombosis has not been studied. The recent trend to reduce stent thrombosis by optimizing Palmaz-Schatz stent deployment with high pressures (14 to 18 atm) and intravascular ultrasound guidance was not attempted.15 16 Although this practice may enhance vascular injury, our observations (Fig 4⇑) suggest that much of the deep necrotizing arterial wall injury seen with this stent is created during the early phase of expansion <5 atm. Higher deployment pressures may more completely and uniformly expand the stent without substantially worsening the extent of injury. This may in turn create better flow characteristics and less surface interaction with blood components, resulting in less thrombosis.
Acknowledging these limitations, our results remain consistent with the hypothesis that stent thrombosis is in part dependent on the extent of vascular injury, which may be augmented by incomplete stent expansion. The data also suggest that stent thrombosis can be modulated by design engineering considerations including metallurgical properties and subtle differences in stent geometry. A relationship between medial necrosis and stent thrombosis also may help explain why clinical stent thrombosis is typically delayed for several days after implantation. When the media necrotizes and thins, a space is created between the stent and the remaining arterial wall. This space contains relatively static blood, a bioprosthetic surface, and exposure to prothrombogenic deep vessel wall components. The clinical relevance of these findings will require validation in additional animal models and ultimately in patients.
This work was supported by a grant from the National Heart, Lung, and Blood Institute (RO1-HL-53226) and the generosity of Miriam and Al Winner, Rose and Richard Miller, and the Cedars-Sinai Grand Foundation. We are indebted to Susan Schauer, MS; Hao Zeng, MD; Adrian Glenn; and Tina Nguyen for technical assistance.
- Received November 2, 1995.
- Revision received March 28, 1996.
- Accepted April 11, 1996.
- Copyright © 1996 by American Heart Association
Scho¨mig A, Kastrati A, Mudra H, Blasini R, Schu¨hlen H, Klauss V, Richardt G, Neumann FJ. Four-year experience with Palmaz-Schatz stenting in coronary angioplasty complicated by dissection with threatened or present vessel closure. Circulation. 1994;90:2716-2724.
Fischman DL, Leon MB, Baim DS, Schatz RA, Savage MP, Penn I, Detre K, Veltri S, Ricci D, Nobuyoshi M, Cleman M, Heuser R, Almond D, Teirstein PS, Fish RD, Colombo A, Brinker J, Mosis J, Shaknovich A, Hishfeld J, Bailey S, Ellis S, Rake R, Goldberg S. A randomized comparison of coronary stent placement and balloon angioplasty in the treatment of coronary artery disease. N Engl J Med. 1994;331:496-501.
Serruys PW, DeJaegere P, Kiemeneij F, Macaya C, Rutsch W, Heyndrickx G, Emanuelsson H, Marco J, Legrand V, Materne P, Belardi J, Sigwart U, Colombo A, Goy JJ, Heuvel P, Delcan J, Morei MA. A comparison of balloon expandable stent implantation with balloon angioplasty in patients with coronary artery disease. N Engl J Med. 1994;331:489-495.
Savage MP, Fischman DL, Schatz RA, Teirstein PS, Leon MB, Baim D, Ellis SG, Topol EJ, Hirshfeld JW, Cleman MW, Buchbinder M, Bailey S, Heuser R, Walker CM, Curry RC Jr, Gebhardt S, Rake R, Goldberg S, for the Palmaz-Schatz Study Group. Long-term angiographic and clinical outcome after implantation of balloon-expandable stent in the native coronary circulation. J Am Coll Cardiol. 1994;24:1207-1212.
Herrmann HC, Buchbinder M, Clemen MW, Fischman D, Goldberg S, Leon MB, Schatz RA, Tierstein P, Walker CM, Hirshfeld JW. Emergent use of balloon-expandable coronary artery stenting for failed percutaneous transluminal coronary angioplasty. Circulation. 1992;86:812-819.
Vrolix MC, van der Krieken T, Piessens JH, for the Wiktor Study Group. Wiktor stent for acute and threatened closure after coronary angioplasty: an update of the European Registry. Circulation. 1992;86(suppl I):I-248. Abstract.
George BS, Voorhees WD, Roubin GS, Fearnot NE, Pinkerton CA, Raizner AE, King SB, Holmes DR, Topol EJ, Kereiakes DJ, Hartzler GO. Multicenter investigation of coronary stenting to treat acute or threatened closure after percutaneous transluminal coronary angioplasty: clinical and angiographic outcomes. J Am Coll Cardiol. 1993;22:135-143.
Hearn JA, King SB III, Douglas JS Jr, Carlin SF, Lembo NJ, Ghazzal ZMB. Clinical and angiographic outcomes after coronary artery stenting for acute or threatened closure after percutaneous transluminal coronary angioplasty: initial results with a balloon-expandable stainless steel design. Circulation. 1993;88:2086-2096.
Schwartz RS, Murphy JG, Edwards WD, Camrud AR, Vliestra RE, Holmes DR. Restenosis after balloon angioplasty: a practical proliferative model in porcine coronary arteries. Circulation. 1990;82:2190-2200.
Colombo A, Hall P, Nakamura S, Almagor Y, Maiello L, Martini G, Gaglione A, Goldberg SL, Tobis JM. Intracoronary stenting without anticoagulation accomplished with intravascular ultrasound guidance. Circulation. 1995;91:1676-1688.
Lablanch JM, Grollier G, Danchin N, Bonnet JL, Belle EV, McFadden E, Bertrand ME. Full antiplatelet therapy without anticoagulation. J Am Coll Cardiol. 1995;25(suppl I):181. Abstract.
Ha˚kan E, Serruys PW, Belardi J, Bonnier H, Colombo A, Fajadet J, Goy JJ, Heyndrickx G, de Jaegere P, Legrand V, Macaya C, Materne AP, Rutsch W, Sigwart U, Suryapranata H, the Benestent Study Group. Clinical experience with heparin-coated stents: the Benestent II pilot phase I. J Am Coll Cardiol. 1995;25(suppl I):181. Abstract.
Rogers C, Edelman ER. Endovascular stent design dictates experimental restenosis and thrombosis. Circulation. 1995;91:2995-3001.
Sutton CS, Consigny PM, Thakur M. Thrombogenicity of intravascular stent wires. Circulation. 1994;90(suppl I):I-9. Abstract.
Van Lack, Lawrence H. Elements of Materials Science and Engineering. 4th ed. Reading, Mass: Addison Wesley; 1980.
Schotky LM. Shape-memory alloys. Sci Am. 1979;241:74-83.
Ikeda Y, Handa M, Kawano K, Kamata T, Murata M, Araki Y, Anbo H, Kawai Y, Watanabe K, Itagaki I, Sakai K, Ruggeri ZM. The role of von Willebrand factor and fibrinogen in platelet aggregation under varying shear stress. J Clin Invest. 1991;87:1234-1240.
Seth S, Park KD, Dev V, Jacobs H, Kin SW, Lambert T, Forrester JS, Litvack F, Eigler N. Prevention of stent subacute thrombosis by segmented polyurethanurea-polyethylene oxide-heparin coating in the rabbit carotid. J Am Coll Cardiol. 1994;23:187. Abstract.