Jump to
Abstract
Abstract—Biocompatible stent coatings may alleviate problems of increased (sub)acute thrombosis after stent implantation. Hyaluronic acid (HA), a ubiquitous, nonsulfated glycosaminoglycan, inhibits platelet adhesion and aggregation and prolongs bleeding when administered systemically. However, the effects of immobilized HA for reducing stent platelet deposition in vivo are unknown. We therefore quantified the antithrombotic effects of coating stainless steel stents and tubes with HA using an established baboon thrombosis model under physiologically relevant blood flow conditions. HA-coated and uncoated (control) stents (3.5 mm in diameter, n=32) and stainless steel tubes (4.0 mm in diameter, n=18) were deployed into exteriorized arteriovenous shunts of conscious, nonanticoagulated baboons. Accumulation of 111In-radiolabeled platelets was quantified by continuous gamma-camera imaging during a 2-hour blood exposure period. HA coating resulted in a significant reduction in platelet deposition in long (4 cm) tubes (0.24±0.15×109 versus 6.12±0.49×109 platelets; P<0.03), short (2 cm) stainless steel tubes (0.18±0.06×109 versus 3.03±0.56×109 platelets; P<0.008), and stents (0.82±0.20×109 versus 1.83±0.23×109 platelets; P<0.02) compared with uncoated control devices. Thus, HA coating reduces platelet thrombus formation on stainless steel stents and tubes in primate thrombosis models. These results indicate that immobilized HA may represent an attractive strategy for improving the thromboresistance of endovascular devices.
-
Genzyme Corporation is a manufacturer of hyaluronic acid.
- Received November 29, 1999.
- Accepted January 27, 2000.
Stent thrombosis results from a series of complex interactions involving the presence of a thrombogenic surface, the damaged vascular wall, altered blood flow, and the activation of platelets and coagulation proteins.1 The effectiveness of antiplatelet agents in reducing thrombosis in atherosclerotic disease has been demonstrated with aspirin and clopidogrel.2 3 Ticlopidine has been shown to reduce periprocedural thrombotic events in coronary stenting to 1% to 2% in low- and intermediate-risk groups.4 5 However, the incidence of stent thrombosis is substantially increased in higher-risk patients.6 In addition, aspirin and ticlopidine may produce significant side effects.7 These findings suggest a need for further improvements in antithrombotic strategies associated with coronary stenting.
Coating stents with biocompatible and nonthrombogenic materials is an attractive alternative for further reducing (sub)acute stent thrombosis. A number of different stent coatings have been evaluated previously.8 9 10 11 12 13 Results with these coatings, including antithrombotic agents and components of cell membranes (biomimicry), appear promising, although larger studies and more cost-effective coatings are needed.14 15
Hyaluronic acid (HA) is a ubiquitous, nonsulfated glycosaminoglycan component of the extracellular matrix. Both HA and immobilized sulfated HA have been shown to inhibit platelet aggregation and platelet adhesion, as well as to prolong bleeding-time measurements when administered systemically.16 17 However, the effects of HA coating of stents on in vivo platelet reactions are presently unknown. The objective of this study was to assess in primates the effects of HA coating on platelet thrombus formation on metallic tubes and stents under controlled and physiologically relevant conditions of blood flow and exposure time.
Methods
All procedures were approved by the Institutional Animal Care and Use Committee in compliance with National Institutes of Health guidelines (Guide for the Care and Use of Laboratory Animals, 1985).
Coated Stents and Tubes
Stainless steel tubes and stents were coated with HA (molecular weight 1 to 3×106 Da) by Genzyme Corporation. Briefly, HA was covalently immobilized onto plasma-treated 316L stainless steel tubes (Small Parts, Inc) (2 and 4 cm long; 4 mm ID) and stents (Multilink, ACS Guidant) (15 mm long and 3.5 mm in diameter) with water-soluble carbodiimide, as described previously.18
Baboon Thrombosis Model
Devices were deployed into exteriorized arteriovenous (AV) shunts in baboons. The chronic AV shunts were composed of silicone rubber tubing (3-mm ID, Dow Corning) placed between the baboon femoral artery and vein. Sixteen juvenile male baboons (Papio anubis/cynocephalus) weighing 9 to 12 kg were studied. The animals were quarantined and observed to be disease free for ≥3 months. Platelet deposition was measured with 111In-labeled platelets as previously described.19 In brief, autologous baboon platelets were labeled with 1 mCi of 111In-oxine. The accumulation of 111In-labeled platelets was measured continuously with a gamma scintillation camera (General Electric 400T). Data were stored at 5-minute intervals and analyzed with a computer-assisted image-processing system interfaced with the camera. We calculated the total number of deposited platelets by dividing the deposited platelet radioactivity (counts per minute) by the whole-blood 111In-platelet activity (counts per min/mL) and multiplying by the circulating platelet count (platelets per milliliter).19
Coated and uncoated stainless steel stents were fully deployed with 3.5-mm noncompliant angioplasty balloons (ACS Guidant) at 10 atm into silicone rubber tubing of 3.2 mm ID. Stainless steel tubes (4.0-mm ID) were interposed between segments of 4.0-mm ID silicone rubber tubing. The tubing segments were prefilled with sterile saline, connected to the AV shunt, and maintained at arterial flow rates (100 mL/min) for 120 minutes without systemic anticoagulation.
Scanning Electron Microscopy
After blood exposure, the stents and steel tubes were disconnected from the shunt, flushed gently with sterile saline, filled with 2.5% buffered glutaraldehyde for 15 minutes, and then filled with 0.1 mol/L cacodylate and stored at 4°C until further processing. Metallic tubes were opened longitudinally with a milling bit and drill press. The stents were carefully removed from the rubber tubing and delicately opened longitudinally with surgical scissors. Scanning electron microscopy (SEM) was performed as described previously.20
Statistical Analysis of Data
Data are presented as mean±SEM. Statistical comparisons were performed with Sigma-Stat (Jandel Scientific). Two-way ANOVA was used to analyze the effects of HA coating and time on platelet deposition between the HA-coated stents or steel tubes versus the uncoated controls. The unpaired Student’s t test was used for comparisons between treated and control groups.
Results
AV Shunt Studies
Eighteen stainless steel tubes (9 coated and 9 uncoated) and 32 stents (8 coated and 24 uncoated) were studied. The measurements of platelet deposition on the stainless steel tubes are shown in Figure 1⇓. For the 2-cm-long tubes, platelet deposition on the controls increased during the first 80 minutes of the study, then either plateaued or occluded the tube (Figure 1A⇓). The HA coating significantly reduced platelet deposition (by 94% versus controls; 0.18±0.06×109 versus 3.03±0.56×109 platelets; P<0.008). In the 4-cm-long stainless steel tubes (Figure 1B⇓), platelet deposition was reduced by 96% versus the controls (0.24±0.15×109 versus 6.12±0.49×109 platelets; P<0.03). Platelet deposition was directly proportional to the area of exposed metal surface, because in the control studies, twice as many platelets deposited on the 4-cm tubes than on the 2-cm tubes.
Time course of platelet deposition onto 4.0-mm-ID stainless steel tubes deployed into AV shunts in baboons. The blood flow rate was 100 mL/min. Platelet deposition was monitored by measuring the accumulation of 111In-radiolabeled platelets. A, 2.0-cm-long tubes. B, 4.0-cm-long tubes. Note that twice as many platelets were deposited on the 4-cm-long tubes as the 2-cm-long tubes. Values are mean±SEM.
As shown in Figure 2⇓, HA coating of stents reduced platelet deposition by 55% after 2 hours of blood exposure (0.82±0.20×109 versus 1.83±0.23 × 109 platelets; P<0.02). This reduction was significantly greater at earlier time points (75% reduction at 1 hour; P<0.0001).
Time course of platelet deposition onto stainless steel stents (ACS Multilink; 3.5 mm nominal diameter) deployed into AV shunts (3.2-mm-ID silicone rubber) in baboons. The blood flow rate was 100 mL/min. Platelet deposition was monitored by measuring the accumulation of 111In-radiolabeled platelets. Values are mean±SEM.
Macroscopy and SEM
Thrombus was grossly evident on both uncoated steel tubes and uncoated stents but was virtually absent on the HA-coated tubes and less apparent on the HA-coated stents (Figure 3⇓, A through D). By SEM, uncoated stainless steel tubes (Figure 3E⇓) showed substantial adherent mixed thrombus (erythrocytes, leukocytes, platelets, and fibrin). In contrast, the HA-coated tubes displayed a smooth luminal surface with occasional adherent red cells and leukocytes (Figure 3F⇓). The uncoated stents also accumulated mixed thrombus, which covered the entire stent surface in a continuous layer (Figure 3⇓, C and G). In contrast, HA-coated stents had only small isolated thrombi, composed typically of dendritic platelets, which accumulated primarily at areas of stent strut convergence (Figure 3⇓, D and H).
Macrophotography of (A) uncoated 4-cm tubes, (B) HA-coated 4-cm tubes, (C) uncoated stents, and (D) HA-coated stents. Note the differences in thrombus deposition. SEM of (E) uncoated 4-cm tubes, (F) HA-coated 4-cm tubes, (G) uncoated stents, and (H) HA-coated stents. Platelets are shown by small white arrows, leukocytes by large white arrows, erythrocytes by arrowheads, and fibrin by a star.
Discussion
Stent thrombosis remains an important problem despite improvements in stent deployment techniques and the use of aspirin and ticlopidine or clopidogrel. The incidence of stent thrombosis is between 0.6% and 2.2% for low-risk patients undergoing elective stenting,4 21 but in higher-risk patients, it is 5.6%.6 One approach to this problem is to coat stents with an immobilized antithrombotic agent.
HA, a high-molecular-weight nonsulfated glycosaminoglycan, is a ubiquitous component of extracellular matrix22 and has been shown to inhibit platelet aggregation and adhesion16 and to prolong the bleeding time at high concentrations.17 Because of its antithrombotic effects and its known coating abilities,17 23 24 HA may provide a potential biocompatible and thromboresistant coating for endovascular devices. The baboon AV shunt model was used to assess coated tubes and stents because this test system allows for exposure of metal surfaces to nonanticoagulated blood under arterial blood flow conditions. Additionally, the chronic shunt does not activate platelets or coagulation without an inserted metal surface.25 In this model, HA coating significantly reduced platelet deposition versus the results with uncoated control devices.
For stainless steel tubes, HA coating reduced platelet thrombus by >94% after 2 hours of blood exposure compared with uncoated tubes (P<0.03). For stainless steel stents, HA coating resulted in a similar reduction of platelet deposition (75% at 1 hour; P<0.0001) compared with uncoated stents. The amount of platelet deposition seen in the uncoated stents and tubes was consistent with previous stent evaluations in this model.8 10 26 27 28 29 Platelet deposition in the stainless steel tubes was also directly proportional to lumenal surface area for the 2- and 4-cm-long tubes. This result is important because it indicates that (1) stainless steel is inherently thrombogenic, and (2) thrombus forms uniformly and reproducibly for tubular devices having relatively small dimensions. Control tubes (2-cm long×4.0-mm ID) showed somewhat higher platelet deposition than control stents (1.5-mm long; deployed into 3.2-mm-ID tubing), a finding that can be attributed in part to the greater metallic surface area of the tubes versus the stents.
Interestingly, although thrombus accumulation on the HA-treated tubes was effectively abolished (Figure 1⇑), the HA-treated stents still accumulated some thrombus (P=0.033 at 120 minutes, HA-treated stents versus HA-treated 2-cm tubes). The remarkable effectiveness of HA for abolishing platelet deposition on the smooth-walled tubes is probably explained by (1) HA coverage of the thrombogenic metallic surface, (2) the absence of surface irregularities that could serve as sites for thrombus attachment and growth, and (3) the rapid dilution under unidirectional laminar blood flow of procoagulant and platelet activating factors, such as thrombin. Similarly, the observations that HA coating reduced thrombus accumulation to a lesser degree on stents than on smooth-walled tubes is presumably due to stent geometric irregularities (stent struts), which can generate regions of flow recirculation and stasis, leading to localized blood coagulation, platelet activation, thrombus attachment, and thrombus growth. Thus, whereas uncoated stents and tubes accumulated considerable gross thrombus, HA-coated stents showed only small isolated thrombi at the areas of stent strut convergence (Figure 3⇑), which suggests that flow disturbances may indeed contribute to thrombus formation in this setting. This finding is also consistent with other observations with this model showing that smooth surfaces tend to be less thrombogenic than textured or irregular devices.19
Previous studies with other stent coatings (eg, phosphorylcholine and heparin) using the same experimental model have shown similar effects for reducing platelet deposition.8 10 However, there is sparse information on the stability and longevity of these alternative coating materials. Immobilized HA was found to be stable for ≥2 months in PBS at 37°C as demonstrated by x-ray photoelectron spectroscopy, Fourier-transformed infrared spectroscopy, and glucosamine assays.27 In addition, immobilized HA is significantly resistant to hyaluronidase digestion.27
The baboon model, as used in the present study, is advantageous because the effects of stent-related variables are emphasized. However, the model differs from human stent applications in several important respects. First, the stents were assessed in an ex vivo shunt system rather than in native arteries, thereby excluding the possible effects of both prothrombotic and antithrombotic components of the vessel wall. Second, no pharmacological inhibitors of platelets or coagulation were administered in the present study, although antithrombotic therapy is commonly administered to stent recipients clinically. In previous baboon studies, stent thrombus formation was reduced by oral aspirin and clopidogrel28 but not by systemic heparin.29 For clinical applications, it seems likely that thrombus formation on HA-coated stents could be reduced by antiplatelet therapy as well.
Thus, our results may not directly predict clinical efficacy. Nonetheless, primates are hemostatically similar to humans, and immobilized HA significantly reduced platelet deposition on both thrombogenic stainless steel tubes and clinical endovascular stents. This study, the first to document such effects in vivo, suggests that additional studies with immobilized HA are warranted to assess this thromboresistant material as a coating for stents and other cardiovascular devices.
Acknowledgments
This study was supported in part by research grants HL31469 and RR-00165 (Yerkes Regional Primate Research Center) from the National Institutes of Health; by the ERC Program of the National Science Foundation under award EEC-9731643; and by a grant from Genzyme Corporation, Cambridge, Mass. The technical assistance of Evan Dessasau, Deborah White, and Steven Marzec is greatly appreciated.
References
This Issue
Jump to
Article Tools
Share this Article
- Reduced Thrombus Formation by Hyaluronic Acid Coating of Endovascular Devices
