Intrapericardial Paclitaxel Delivery Inhibits Neointimal Proliferation and Promotes Arterial Enlargement After Porcine Coronary Overstretch
Background—Catheter-based intrapericardial (IPC) delivery of therapeutic agents has recently been demonstrated. Paclitaxel is known to inhibit vascular smooth muscle cell proliferation. This study examined the effect of IPC instillation of paclitaxel on neointimal proliferation after balloon overstretch of porcine coronary arteries.
Methods and Results—Overstretch injury of coronary arteries was followed by IPC administration of micellar paclitaxel at low dose (LD, 10 mg; n=6) or high dose (HD, 50 mg; n=7) or of control micelles (50 mg, n=5). Animals were euthanized 28 days after balloon dilation. Arterial injury indices were no different among the groups. The neointimal area, maximal intimal thickness, and adventitial thickness were significantly reduced in both LD (0.47±0.04 mm2, 0.43±0.03 mm, and 0.35±0.02 mm, respectively) and HD (0.51±0.06 mm2, 0.42±0.03 mm, and 0.38±0.03 mm, respectively) paclitaxel groups compared with the control group (0.79±0.07 mm2, 0.56±0.02 mm, and 0.47±0.02 mm, respectively; P<0.001). Meanwhile, the vessel circumference measured at the external elastic lamina of paclitaxel-treated vessels was significantly larger than the control circumference. Apoptotic cells were found in the neointima. The apoptotic cell percentage was not different between the control (1.72%) and LD (2.31%) groups but was higher in the HD group (7.07%, P<0.0001 versus control and LD groups). Immunostaining for matrix metalloproteinase-2 revealed concurrent reduction in the HD group compared with the control and LD groups.
Conclusions—IPC space delivery of a single dose of paclitaxel significantly reduces vessel narrowing in this balloon-overstretch model. This effect is mediated by reduction of neointimal mass as well as positive vascular remodeling.
Although tremendous advances in coronary recanalization have been developed during the past 20 years, a solution to restenosis after coronary angioplasty is still lacking.1 It had been hoped that intracoronary delivery would allow achievement of high concentrations of therapeutic bioactive agents within the coronary artery wall. However, such approaches are limited by inconsistent delivery, low localization efficiency, and rapid washout of agent from the target vascular wall after delivery.2 3
Recently, catheter-based methods for access to the normal pericardial space have been described.4 5 6 7 The benefits of delivering agents to the pericardial sac include enhanced consistency of local agent levels, reduced acute systemic delivery of agent, and prolonged exposure of coronary arteries to the therapeutic material.8 Intrapericardial (IPC) delivery of NO donors reduces neointimal proliferation in a porcine coronary balloon-overstretch model after 2 weeks9 and in a porcine stent-restenosis model after 4 weeks.10 Favorable local pharmacokinetics and consistency of tissue loading after pericardial delivery compared with endoluminal delivery have been demonstrated by our group.3
Smooth muscle cell (SMC) hyperplasia and vessel remodeling are critical events in restenosis.11 Paclitaxel inhibits microtubular disassembly and function, resulting in apoptosis. Because SMCs are actively dividing after arterial injury, paclitaxel may be a good candidate for halting neointimal formation. It has been shown to inhibit SMC proliferation and migration in vitro in rat and human SMCs as well as to reduce the intimal area and restenosis after vascular injury in rat and rabbit carotid artery models.12 13
Such results suggested a study of paclitaxel in the context of coronary balloon injury with use of IPC placement to access the epicardial vessels. The present study examines the effects of catheter-based IPC instillation of paclitaxel on the arterial occlusion induced by balloon overstretch of pig coronary arteries.
Paclitaxel compounded with a micellar polymer and control micelles containing the polymer alone (PN 7018, lot No. AR-BC020-97, and PN 7053, lot No. 98026A, Angiotech Pharmaceuticals, Inc) were dissolved in 0.9% sodium chloride at 50±5°C. The solutions were sterile-filtered and used within 4 hours for pericardial instillation.
Eighteen juvenile female domestic pigs weighing 23 to 25 kg were used. The animals were divided into 3 groups: low-dose (LD, 10 mg paclitaxel; n=6) and high-dose (HD, 50 mg paclitaxel; n=7) groups and a control group (50 mg copolymer, n=5). All animals received a normal diet. The study was approved by the Indiana University Animal Care and Use Committee and was based on National Institutes of Health laboratory standards.
Animals were fasted overnight and premedicated with aspirin (325 mg) 24 hours before sedation with intramuscular ketamine (20 mg/kg), xylazine (2 mg/kg), and atropine (0.05 mg/kg). Anesthesia was initiated with thiopental sodium (25 mg/kg IV). After intubation, the animals were ventilated by using air mixed with oxygen (2 L/min) and isoflurane (2.5%). The ECG and blood pressure were monitored.
Animals underwent coronary balloon dilation, as previously described.14 15 After systemic heparinization (200 U/kg) and lidocaine (30 mg), an 8F guiding catheter was used to engage the left coronary artery. After intracoronary nitroglycerin (200 μg), coronary angiography was performed. The left anterior descending and left circumflex coronary artery diameters were determined by use of NIH Image, and a 20-mm balloon with a 1.3 balloon/artery diameter ratio was used to dilate the target 3 times for 30 seconds each time.
Percutaneous IPC Space Delivery
After the balloon procedure, a pericardial access device (PerDUCER, Comedicus Inc) was used for transthoracic insertion of a guidewire into the normal pericardial space as previously described (Figure 1⇓). The sheathed needle was inserted into the mediastinum through an introducer and positioned on the anterior surface of the pericardial sac, which was drawn into the hemispherical tip by suction, and pierced. Finally, a 0.018-in guidewire was placed through the needle and advanced several centimeters to confirm confinement within the pericardial space. A 4F hydrophilic-coated catheter was inserted, IPC placement was tested by contrast injection, and 25 mL of paclitaxel with copolymer or the copolymer alone was delivered over 5 minutes into the pericardial sac.
At 28 days after the procedure, animals were anesthetized, and final coronary angiography was performed after heparin (200 U/kg) administration. The animals were euthanized by a lethal dose of pentobarbital (65 mg/kg), heart and pericardial tissues were harvested, and coronary arteries were perfusion-fixed with 10% buffered zinc formalin for 15 to 20 minutes at 80 mm Hg pressure.
Tissue Preparation and Immunocytochemical Staining
Gross pericardial adhesions were quantified according to the scoring system of Hurewitz et al.16 The grades were as follows: 0, normal; 1, focal thin adhesions; 2, diffuse widespread adhesions; and 3, complete obliteration of the pericardial space. After paraffin embedding and sectioning were performed, pericardial tissue was stained with hematoxylin-eosin and Masson’s trichrome. Mesothelial cells on the parietal pericardium were noted as absent or present. The thickness of visceral pericardium was measured at 4 sites overlying the 4 chambers.
Left anterior descending and left circumflex coronary vessels were sectioned at 3-mm intervals from the proximal to distal end and embedded in paraffin. Sections were cut at 6 μm, affixed to glass microscope slides, and stained with hematoxylin-eosin and Verhoeff–van Gieson’s reagents. Immunostaining was performed on selected segments with the use of primary antibodies, including anti–smooth muscle α-actin (1:1000, Dako), von Willebrand factor (1:600, Dako), and anti–matrix metalloproteinase antibody (MMP-2, 1:100, Oncogene). Secondary antibody binding was revealed by avidin complex, with a staining reaction performed with the use of 3,3′-diaminobenzidine solution (Sigma). Nuclei were counterstained with hematoxylin or methyl green. Endogenous peroxidase activity was blocked with 3% H2O2 solution for 5 minutes. Negative controls were generated by using nonimmune serum. To permit comparative qualitative analysis of the staining intensity of the study groups, staining of multiple segments from distinct study groups was conducted at the same time with the use of consistent development protocols for each antigen.
Apoptotic cells were detected by use of a Klenow fragment end labeling kit (Oncogene). After deparaffinization, the tissue sections were treated with 20 μg/mL proteinase K/10 mmol/L Tris-HCl, pH 8.0, for 10 minutes. After rinsing in 1× Tris-buffered saline, the Klenow labeling reaction mixture was added. In each experiment, positive and negative controls were included. The positive control was treated with DNase l (1 mg/mL, 20 minutes, room temperature) to induce DNA strand breaks; the negative control was exposed only to Klenow labeling reaction mix (without Klenow enzyme).
Morphometric measurements were performed as described previously17 by use of a light microscope (Olympus) at low power (×2.5) linked to a video camera (Sony) and computer-interfaced with NIH Image. The endoluminal length and the circumference bounded by the internal and external elastic laminae (IEL and EEL, respectively) were traced manually, and luminal and intimal areas were determined. Fracture length (FL) was defined as the arc length between the 2 fracture points of the IEL. Intimal area was measured directly. Maximal intimal thickness was defined as the maximal distance between the lumen and EEL, and maximal adventitial thickness was defined as the analogous length between EEL and adventitia, normal to the arterial circumference. The percent stenosis was described as the histological luminal diameter at the site of maximal stenosis divided by the preangioplasty luminal diameter determined at the midpoint of the target segment.
Hematoxylin-stained cells were counted at ×40 microscopic magnification, randomly evaluating areas encompassing 20% to 40% of the total neointimal cross-sectional area.18 19 The cells within the media were counted in 5 regions: regions 1 and 2, composed of the 2 medial ends adjacent to the medial tear; region 3, the site 180° opposite the neointimal mass; and regions 4 and 5, at 90° radials with respect to the neointima. Apoptotic cells in the neointima or media were counted positive when showing morphological features characteristic of apoptosis as well as positive nuclear Klenow labeling.
Results are presented as mean±SEM. An unpaired t test was used to compare the 3-group histomorphometric measurement data. Differences are considered significant at P<0.05. All statistical calculations used the SigmaStat software package.
Pericardial instillation was well tolerated by all animals. No complications developed, and no clinical evidence of paclitaxel-related toxicity was noted.
Baseline Angiographic Characteristics
The artery diameters before dilation (control 2.65±0.12 mm, LD 2.51±0.13 mm, and HD 2.47±0.12 mm; P=NS) and the balloon/artery ratios (control 1.34±0.02, LD 1.32±0.03, and HD 1.32±0.02; P=NS) were no different among the groups.
Pericardial Tissue and Contents
The gross and histological changes of the pericardium after IPC paclitaxel delivery are summarized in Table 1⇓. IPC adhesions were absent in the control and LD groups, except for 3 pigs with a few thin adhesions limited to the puncture site (1 pig in the control group and 2 pigs in the LD group). In the HD group, widespread pericardial adhesions routinely extended from the visceral to the parietal layer, and the pericardial space was obliterated in 4 pigs. Macroscopic scoring of the pericardial space adhesions, as described above, confirmed that the HD, but not the LD, group was significantly different from the control group (adhesion scores were 0.20±0.20 for control, 0.33±0.21 for LD, and 2.57±0.20 for HD groups; P<0.001 for HD versus control or LD group). Microscopically, the control and LD groups had intact mesothelial layers, with multilayering noted in some regions; no alteration was associated with exposure to control micelles. In addition to intracavitary adhesions, the pericardium of the HD group was demonstrably thicker compared with the control and LD pericardium. The interlaminar adhesion tissue displayed fibrin and collagen deposition as well as infiltration with mononuclear cells. Cells staining positively for SMC α-actin expression were found throughout the connective tissue of the visceral pericardium in all groups, and the most intense staining was in the HD group.
Morphometric Analysis of Arteries
IPC paclitaxel delivery significantly inhibited neointimal proliferation, as demonstrated in Figure 2⇓. Table 2⇓ displays morphometric data for the vessels constituting each group. The extent of vessel injury, expressed as an injury index (FL/FL+IEL) was equivalent among the 3 groups (control 0.21±0.02, LD 0.22±0.03, and HD 0.21±0.01; P=NS). The neointimal response correlated with the degree of vessel injury in the control as well as the experimental groups (Figure 3⇓; control R2=0.69, HD R2=0.66, and LD R2=0.44), with markedly diminished slope for vessels receiving paclitaxel at either dose. The absolute neointimal area (Figure 4A⇓) was smaller in both experimental groups (LD 0.47±0.04 and HD 0.51±0.06 mm2, P=NS) compared with the control group (0.79±0.07 mm2, P<0.001). The neointimal area normalized to FL was also significantly smaller for both treatment groups (LD 0.32±0.02 and HD 0.39±0.04) than in the control group (0.68±0.03, P<0.001). Similarly, the maximal intimal and adventitial thicknesses were lower in both paclitaxel groups than in the control group (P<0.001). However, the medial area did not differ among the groups. Compared with the control group, both paclitaxel groups evidenced outward vascular remodeling, with the EEL circumference and enclosed area significantly larger in the LD and HD groups (Table 2⇓).
The degree of luminal occlusion, expressed as percent stenosis, was significantly reduced in both treated groups (LD 10±2% and HD 22±3% versus control 39±3%, P<0.001; Figure⇑ 4B) for relative stenosis reductions of 74% in the LD and 42% in the HD groups. This is dominantly achieved by the effect on vessel remodeling. Comparative evaluation of the area contributions of the decreased neointima and the increased vessel circumference (Figure 5⇓) shows that the latter accounts for 70% to 80% of the luminal expansion relative to control vessels.
Cell Quantification and Immunohistochemistry
Medial cell density was not different among the control (3983±128 cells/mm2), LD (3875±244 cells/mm2), and HD (4089±422 cells/mm2) groups. However, the neointimal cell density in the HD group (3571±128 cells/mm2) was significantly lower than that in the control (4574±201 cells/mm2) and LD (4196±120 cells/mm2) groups (P<0.001). The Klenow-positive cells were predominately detected in the neointima, with few found in the media or adventitia. Most staining cells also demonstrated hyperchromatic fragmented nuclei. Some had histologically normal nuclei, possibly representing early apoptosis. There were no significant differences between the control and LD groups, but the HD group had a greater percentage of apoptotic cells than either of these groups (control 1.72% and LD 2.31% versus HD 7.07%, P<0.0001).
Immunohistochemical staining demonstrated that neointimal cells were predominantly immunoreactive for α-actin in all groups. The neointima was composed of spindle-shaped cells and a large amount of loose extracellular matrix (Figure 6⇓). At 28 days after balloon injury, complete vessel reendothelialization had been achieved in most vessel segments of all 3 groups, as measured by von Willebrand factor staining. Because paclitaxel has shown altered MMP-2 expression in other systems, we investigated for such modulation after IPC paclitaxel delivery. In control as well as LD vessels, MMP-2 immunoreactivity was found in endothelial cells, neointima, and media. Conversely, there was generally diminished MMP-2 staining in all vessel layers in the HD sections. MMP-2 staining was not characteristically present in the adventitia of any groups.
The present study demonstrates that a single dose of paclitaxel into the pericardial space is associated with significant restenosis reduction after porcine coronary balloon injury, mediated by reduced neointimal formation and enhanced arterial enlargement. The 10-mg dosage produced more optimal results, although favorable effects were present at both doses. These findings have implications for clinical utility given the growing experience with systemic paclitaxel as an antiproliferative agent.
The clear effect identified in vivo 28 days after a single dose is remarkable and likely relates to the high affinity of paclitaxel for its specific intracellular target sites on microtubules as well as its hydrophobicity, which will favor slow redistribution after local delivery. The reservoir formed by the pericardial sac would also be expected to contribute to the persistence of effective concentrations after IPC placement. A single paclitaxel exposure for either 20 minutes or 24 hours in vitro caused a complete and prolonged inhibition of human arterial SMC growth up to day 14.13 In vivo experiments using endovascular porous-balloon delivery of paclitaxel have resulted in the reduction of rabbit carotid artery restenosis by 24% at 28 days after balloon injury.13 Several groups have recently reported antistenotic effects of paclitaxel when it was applied to stents as a direct coating or with a polymeric matrix,20 21 22 whereas other studies using local delivery of paclitaxel before stent implantation have not shown benefit.23
The EEL circumference becomes remarkably larger after paclitaxel therapy (7.04±0.22 mm for control versus 7.71±0.23 mm for HD and 8.12±0.18 mm for LD groups, P=0.017). Such an effect, found in the context of conserved medial area, reflects modulated disposition of tissue mass consistent with altered vascular remodeling compared with the control condition. Such changes of vessel circumference have also been noted after surgical application of 20% paclitaxel-loaded paste to the perivascular surface (L.S. Machan, personal communication, October 1999). The reduced adventitial thickness found in the present study provokes the hypothesis that decreased adventitial fibrosis may contribute to this positive remodeling. The result of the increment in vessel circumference and the reduction in neointimal mass due to IPC paclitaxel is an increase in luminal size from 5.12±0.23 mm in the control group to 6.15±0.25 mm in the HD group and 7.02±0.18 mm in the LD group (P=0.002 and P= 0.006, respectively). This dual effect of paclitaxel on remodeling and proliferation is encouraging because multiple studies have suggested that increased total SMC bulk and vascular remodeling both contribute to restenosis after angioplasty, whereas numerous therapeutic agents affecting predominantly SMC proliferation have been found insufficient to prevent vessel renarrowing. However, the basis for such inward remodeling and the mechanisms by which paclitaxel promote outward remodeling remain largely speculative.
Recent studies have suggested that MMPs and their inhibitors, which regulate matrix homeostasis, might play a significant role in normal and pathological vessel remodeling. Degradation of the elastic laminae by MMP-2 is accentuated in inward remodeling due to low flow and outward remodeling due to high flow and appears to be an important component of structural modification of the vessel wall.24 25
MMP-2 expression was detectable in the control and LD groups after porcine coronary angioplasty but was generally lost or reduced in the HD vessel segments. The mechanism of MMP-2 downregulation after exposure to paclitaxel at the 50 mg dose is unknown. The presence of MMP-2 immunoreactivity in the control and LD groups must be interpreted cautiously because of the absence of data confirming zymogen activation and molar excess with respect to tissue metalloprotease inhibitor levels, a limitation of the present study. Nevertheless, the absence of MMP-2 staining in the HD group suggests a lack of activity in these specimens. This, in turn, generates the hypothesis that the loss of MMP-2 is linked to the diminished outward remodeling found in the HD group. To more fully assess the importance of MMP-2 expression for vascular remodeling after PTCA and after IPC delivery of paclitaxel, future polyacrylamide gel electrophoresis and in situ zymography studies will be required.
IPC paclitaxel at both doses does not cause overt damage to either the endothelial or medial layers. Endothelial regeneration was nearly complete in all groups, consistent with reports showing reendothelialization at 4 to 8 weeks after injury.26 Likewise, medial cell densities and areas were no different among the 3 groups.
The diminished outward remodeling in the 50-mg dosage group is largely responsible for decrease of the antistenotic effect with the lower dose. This biphasic dose-response relation may define the transition into a supratherapeutic level for this approach. Indeed, diminished neointimal cell densities accompanied by a higher percentage of apoptotic cells in HD but not LD vessels (P<0.001) may be a further reflection of vascular toxicity and specifically does not correlate with enhanced outward remodeling. Although apoptosis has been noted early after balloon dilation, persistent apoptosis at 28 days has been associated with chronic vascular insult.27 28 Hui et al29 reported that paclitaxel causes synovial toxicity by inducing apoptosis in vitro. High concentrations of paclitaxel (≥50.0 μmol/L) also cause SMC apoptosis in vitro.13
The present study also identifies that administration of 50 mg micellar paclitaxel as a single IPC dose is above the tolerable threshold for pericardial tissue. At this dose, mesothelial layer destruction is accompanied by pericardial adhesions characterized by mononuclear infiltration and an expansion of α-actin–positive cells within the subepicardial connective tissue layer. Local secretion of proinflammatory molecules in this group may also contribute to the diminished antistenotic effect found in the HD group. An implicit limitation of these data resides in the absence of knowledge concerning the pericardial reaction at later time points. Accordingly, it will be necessary to conduct studies extending the length of observation after such intrapericardial deliveries to support the safety of these approaches for clinical application.
Our findings demonstrate that a single-dose perivascular delivery of paclitaxel into the pericardial space preserves luminal patency in the porcine coronary balloon–overstretch model. The mechanism of vascular luminal maintenance involves promotion of positive vascular remodeling as well as inhibition of SMC hyperplasia. The present study further establishes a maximum dose for IPC paclitaxel by using the polymeric formulation described and suggests that a carefully chosen dose of paclitaxel may inhibit postangioplasty restenosis via IPC delivery. Further experimentation will be required to support long-term safety of IPC paclitaxel and to determine the dose for optimizing its therapeutic efficacy.
The authors express gratitude to David Mendel and Larry Solomon for excellent technical assistance. We acknowledge Teresa Knight, Kelly Palmer, Frederick M. Rauscher, and Tonya J. Dickson for their editorial assistance.
- Received January 26, 2000.
- Revision received April 26, 2000.
- Accepted May 2, 2000.
- Copyright © 2000 by American Heart Association
Topol EJ, Serruys PW. Frontiers in interventional cardiology. Circulation. 1998;98:1802–1820.
Stoll HP, Carlson K, Keefer LK, et al. Pharmacokinetics and consistency of pericardial delivery directed to coronary arteries: direct comparison with endoluminal delivery. Clin Cardiol. 1998;21(suppl III):III-10–III-16.
March KL, Woody M, Mehdi K, et al. Efficient in vivo catheter-based pericardial gene transfer mediated by adenoviral vectors. Clin Cardiol. 1999;22(suppl I):I-23–I-29.
Seferovic PM, Ristic AD, Marsimovic R, et al. Initial clinical experience with PerDUCER device: promising new tool in the diagnosis and treatment of pericardial disease. Clin Cardiol. 1999;22(suppl. I):I-30–I-35.
March MP, Igo SR. Minimally invasive access of the normal pericardium: initial clinical experience with a novel device. Clin Cardiol. 1999;22(suppl I):I-36–I-39.
Hou D, March KL. Intrapericardial approach for therapeutic angiogenesis. In: Kornowski R, Leon MB, eds. Handbook of Myocardial Revascularization and Angiogenesis. London, UK: Martin-Dunitz; 1999:189–200.
Baek SH, Keefer L, Mehdi K, et al. Intrapericardial nitric oxide donor reduces neointimal and adventitial thickening following porcine coronary overstretch. J Am Coll Cardiol. 1997;29:51A. Abstract.
Makkar RR, Shah PK, Terhakopian A, et al. Intrapericardial delivery of a nitric oxide donor inhibits in-stent stenosis in porcine coronary arteries. Am J Cardiol. 1998;82(suppl 7A):104S. Abstract.
Deuel TF, Bianchi C, Cantley L. Restenosis injury: a problem in regulation of growth. In: Chien KR, ed. Molecular Basis of Cardiovascular Disease. Philadelphia, PA: WB Saunders; 1999:367–391.
Sollott SJ, Cheng L, Pauly RR, et al. Taxol inhibits neointimal smooth muscle cell accumulation after angioplasty in the rat. J Clin Invest. 1995;95:1869–1876.
Axel DI, Kunert W, Goggelmann C, et al. Paclitaxel inhibits arterial smooth muscle cell proliferation and migration in vitro and in vivo using local drug delivery. Circulation. 1997;96:636–645.
Robinson KA, Chronos NA, Schieffer E, et al. Pharmacokinetics and tissue localization of antisense oligonucleotides in balloon-injured pig coronary arteries after local delivery with an iontophoretic balloon catheter. Cathet Cardiovasc Diagn. 1977;41:354–359.
Liu MW, Anderson PG, Luo JF, et al. Local delivery of ethanol inhibits intimal hyperplasia in pig coronary arteries after balloon injury. Circulation. 1997;96:2295–2301.
Wilensky RL, March KL, Gradus-Pizlo I, et al. Vascular injury, repair, and restenosis after percutaneous transluminal angioplasty in the atherosclerotic rabbit. Circulation. 1995;92:2995–3005.
Strauss BH, Robinson R, Batchelor WB, et al. In vivo collagen turnover following experimental balloon angioplasty injury and the role of matrix metalloproteinases. Circ Res. 1996;79:541–550.
Waksman R, Rodriguez JC, Robinson KA, et al. Effect of intravascular irradiation on cell proliferation, apoptosis, and vascular remodeling after balloon overstretch injury of porcine coronary arteries. Circulation. 1997;96:1944–1952.
Heldman AW, Cheng L, Heller P, et al. Paclitaxel applied directly to stents inhibits neointimal growth without thrombotic complications in a porcine coronary artery model of restenosis. Circulation. 1997;96(suppl I):I-288. Abstract.
Farb A, Heller PF, Carter AJ, et al. Paclitaxel polymer-coated stents reduce neointima. Circulation. 1997;96(suppl I):I-608. Abstract.
Drachman DE, Edelman ER, Kamath KR, et al. Sustained stent-based delivery of paclitaxel arrests neointimal thickening and cell proliferation. Circulation. 1998;98(suppl I):I-740. Abstract.
Oberhoff M, Cetin S, Alghobainy R, et al. Local delivery of paclitaxel before stent implantation using the perfusion double balloon in a porcine restenosis model. Circulation. 1999;100(suppl I):I-306. Abstract.
Kleijn D, Velema E, Schoneveld A, et al. Increase of metalloproteinase-2 and -9 activity during flow induced remodeling is regulated at the post-transcriptional level. J Am Coll Cardiol. 1999;33:268A. Abstract.
Van Belle E, Bauters C, Asahara T, et al. Endothelial regrowth after arterial injury: from vascular repair to therapeutics. Cardiovasc Res. 1998;38:54–68.
Isner JM, Kearney M, Bortman S, et al. Apoptosis in human atherosclerosis and restenosis. Circulation. 1995;91:2703–2711.