Low-Dose Radioactive Endovascular Stents Prevent Smooth Muscle Cell Proliferation and Neointimal Hyperplasia in Rabbits
Background Restenosis induced by smooth muscle cell (SMC) migration and proliferation and neointimal thickening limits the clinical success of balloon angioplasty and stent implantation. In this study, the long-term effect of endovascular irradiation via low-dose radioactive stents on neointima formation was compared with conventional stent implantation in a rabbit model.
Methods and Results Palmaz-Schatz stents were made radioactive in a cyclotron. The stents had a very low activity (maximum, 35 μCi), and thus, manipulation did not require extensive radiation protection. One, 4, 12, and 52 weeks after the implantation of nonradioactive stents and radioactive stents in rabbit iliac arteries, neointimal thickening was analyzed by quantitative histomorphometry. Immunostaining for endothelial cell von Willebrand factor, macrophages, SMC α-actin, collagen type I, and proliferating cell nuclear antigen (PCNA) was performed to determine radiation-induced changes in the arterial wall. SMC proliferation was quantified by computer-assisted cell counting of PCNA-immunoreactive cells. Neointima formation was markedly suppressed by the implantation of radioactive stents in a dose-dependent fashion at all observed time points. At peak proliferative activity of SMCs 1 week after nonradioactive stent implantation, 30±2% of SMCs in the neointima were proliferating, compared with 0.5±0.1% of SMCs after implantation of stents with an initial activity of 35 μCi (P<.001). The neointima covering radioactive stents was characterized by decreased smooth muscle cellularity and increased extracellular matrix formation. Further, we observed a delayed endothelialization depending on the radiation dose. No difference in vascular thrombosis was found after nonradioactive and radioactive stent implantation.
Conclusions The results of this study clearly indicate that low-dose radioactive endovascular stents potently inhibit SMC proliferation and neointimal hyperplasia in rabbits.
Neointimal hyperplasia is the predominant feature of human restenosis after percutaneous transluminal coronary angioplasty (PTCA) or stent implantation. Migrating and proliferating smooth muscle cells (SMCs) responding to the initial vascular injury accompanied by the deposition of extracellular matrix are thought to be key events in this process.1 2 3 The antiproliferative effect of ionizing radiation has been used for many years to reduce cancer growth. Studies on potential antiproliferative effects of vascular irradiation have been hampered by the fact that the radiation therapy in cancer patients exposed to doses >35 Gy can cause occlusive vessel disease.4 5 6 Conflicting results have been reported from studies using ionizing radiation to prevent restenosis after angioplasty. The endovascular irradiation of porcine coronary arteries with a 192Ir source reduced neointimal hyperplasia after application of a dose of 20 Gy.7 A previous investigation applying 4 to 8 Gy external x-irradiation to porcine coronary arteries after stent implantation demonstrated an accentuation in the development of a neointima.8
Very little is known about the repair mechanisms of injured arteries after angioplasty and ionizing radiation. Intact arteries are very resistant to radiation. Only doses >10 Gy induce alterations in vascular morphology.9 It is well known that proliferating cells are more susceptible to radiation than quiescent cells. Therefore, the radiosensitivity of an injured vessel wall may be much higher than the radiosensitivity of an intact vessel. This study tested the hypothesis that the neointimal response to arterial injury is inhibited by vascular stents emitting low-dose ionizing radiation. Stents with very low activity were implanted in rabbit iliac arteries, and the resulting neointimal hyperplasia was compared with that from nonradioactive stents (NRS).
Activation of Metallic Stents
Stainless steel Palmaz-Schatz stents (7.5 mm in length) were activated by a special irradiation technique developed at the Nuclear Research Center in Karlsruhe (KfK), Germany.10 In brief, the stents were bombarded with suitable charged particles of adapted energy in the external beam of the KfK cyclotron to create a proper mixture of the radioisotopes 55Co (main activity), 56Co, 51Cr, 52Mn, 57Ni, and 55Fe within the metal emitting β-, γ-, and x-radiation with half-lives between 17.5 hours (55Co) and 2.7 years (55Fe). Stents were activated to an activity of 17.5 and 35 μCi at the date of implantation in rabbit arteries. The predominant amount of radiation is caused by β-particles. The low-level activity allows manual handling, not requiring a radioisotope license according to International Atomic Energy Agency (IAEA) regulations.
The main features of this stent surface activation are predominantly short-range radiation, homogeneously distributed over the length and circumference of the stent, and absolutely fixed to the metal. Dose measurements were performed using LiF (Mg/Ti-doped) thermoluminescence detectors (3×3×1 mm) in a phantom (300-mm diameter, 150-mm thickness) of polyamide material [HOOC(CH2)10COOH; density, ρ=1.01 g/cm3]. The radioactive stent (RS) was centered in the midplane of the phantom and was expanded to a diameter of 3 mm, and the single calibrated dose detectors were arranged at 15 different radial distances outward from the surface of the stent. Retroaction of shadowing was avoided by the specific arrangement of the detectors around the stent. The integral radiation doses were expressed in grays.
Animal Care and Surgical Procedure
All experiments were performed in accordance with the guidelines for animal research established by the American Heart Association and were approved by the state committee for animal research. Thirty-three New Zealand White rabbits of both sexes weighing between 2.5 and 3.0 kg were housed in individual cages and maintained on standard rabbit chow and water ad libitum. The animals were anesthetized with ketamine (35 mg/kg) and xylazine (5 mg/kg) IM. Both femoral arteries were exposed and ligated, and two 4F pediatric sheaths were introduced via arteriotomy. Heparin 500 IU and of aspirin 60 mg were given IV before the stent implantation. An RS was mounted manually on a 3-mm balloon angioplasty catheter (ACS Inc) and expanded in one common iliac artery at 10 atm for 2 minutes. An NRS was implanted likewise in the contralateral iliac artery. The arteries had diameters of 2.5 mm; therefore, the ratio of balloon-expanded stent to artery was 1.2:1. The femoral arteries were ligated, the wounds were closed, and the animals received 60 mg of aspirin IM every third day for 4 weeks. Rabbits receiving RS were divided into three groups on the basis of RS activity. Group 1 stents had an initial activity of 3.9 μCi (stents of group 3 stored for 30 days), group 2 had an activity of 17.5 μCi, and group 3 had an activity of 35 μCi (Table 1⇓).
After a lethal dose of sodium pentobarbital (120 mg/kg) was given to the rabbits, the abdominal aorta was cannulated and the animals were exsanguinated by flushing with lactated Ringer’s solution at 100 mm Hg pressure. The iliac arteries were harvested, and two thirds of the stented region was cut off of each one and immersed in 1.5% formaldehyde and 1.5% glutaraldehyde overnight. The specimens were stepwise dehydrated with graded alcohols and embedded in epoxy-araldite resin (Serva). Thereafter, stented arteries were sectioned into 70-μm slices with a rotating diamond-coated saw (Leica). The sections were stained with toluidine blue. The vessel perimeters delineated by the external elastic lamina and the neointimal areas were measured as described previously by computer-assisted morphometry using a light microscope (Olympus) connected to a video camera (Sony) and a computer-based high-resolution digitizing image analyzer (Pavlov Inc).11
After the wires had been removed from the remaining one third of the iliac segments, specimens were immersed in Carnoy’s fixative (60% methanol, 30% chloroform, and 10% glacial acetic acid) for 18 hours. The arteries were embedded in paraffin, cut into 4- to 6-μm serial sections, and dried on albumin-glycerol–coated slides overnight at 56°C. The next day, slides were deparaffinized and postfixed with acetone. Sections were then preincubated with 3% hydrogen peroxide to reduce the endogenous peroxidase activity and rinsed in 0.05 mol/L PBS. The sections were incubated with the primary antibody at 37°C for 1.5 hours. To minimize nonspecific antibody binding, monoclonal primary antibodies were diluted in 10% sheep serum (Sigma Chemical Co), whereas polyclonal antibodies were diluted in 10% donkey serum or goat serum (Dianova). After three washings with PBS, species-appropriate biotinylated secondary antibodies were applied, followed by a streptavidin horseradish peroxidase complex (Amersham). Antibody binding was visualized with 3,3-diaminobenzidine (Kem-En-Tec), yielding a brown color. Counterstaining was performed with Gill’s hematoxylin. Staining with type- and class-matched irrelevant antibodies served as negative control for each antibody.
Antibodies Used in the Study
To visualize SMCs, mouse IgG2a monoclonal antibody to rabbit SMC α-actin (Boehringer Mannheim) was applied at a 1:800 dilution. Endothelial cells were identified by von Willebrand factor (vWf) immunostaining using a polyclonal goat anti-human vWf antibody (Atlantic Antibodies) that cross-reacts with rabbit vWf at dilutions of 1:100.12 A monoclonal mouse antibody (clone PC10, Dako Corp) against proliferating cell nuclear antigen (PCNA) was used at 1:100 dilution to study SMC proliferation. To detect macrophages, mouse anti-rabbit macrophage IgG1 (RAM 11, Dako) was applied at dilutions of 1:100. The extracellular matrix component collagen I was identified by a polyclonal anti-rat collagen type I raised in guinea pigs (gift of Dr Luisa Iruela-Arispe, Seattle) that cross-reacts with rabbit collagen I (dilution, 1:100). It does not recognize collagen type II, III, or IV, fibronectin, serum proteins, or thrombospondin by Western blotting or ELISA. Uninjured iliac arteries, lung with alveolar macrophages, ileum, and skin were used as controls for positive immunostaining of the applied antibodies.
The SMC density of five randomly chosen 0.1-mm2 areas of the neointima was measured with computer assistance at ×200 light magnification. The number of PCNA-immunoreactive SMC nuclei in the neointima was expressed as percentage of PCNA-positive cells per total number of SMC nuclei.
Transmission Electron Microscopy
Arterial segments were immersed overnight in 1.5% formaldehyde and 1.5% glutaraldehyde in 0.1 mol/L PBS, postfixed in 1% osmium tetroxide and 0.1 mol/L cacodylic acid for 1 to 2 hours, dehydrated in graded ethanol baths, and embedded in epoxy-araldite resin. Tissues were sectioned into 200-μm slices, and stent struts were removed. All segments were reembedded in epoxy-araldite resin, and one segment per study group at each time point was cut into ultrathin (40- to 80-nm) sections. The sections were mounted on copper nets, and after 2% uranyl acetate/lead citrate was added for contrast, they were examined on a Zeiss microscope (EM 10C/CR) at 80-kV accelerating voltage.
All data are presented as mean±SEM. The two-tailed paired Student’s t test was used to compare group means. Simultaneous comparisons of more than two means were performed with ANOVA followed by Scheffé’s test. A probability value of P<.05 was considered significant.
The dosimetry of a radioactive stent with an activity of 35 μCi at date of implantation is given in Fig 1⇓. The greatest amount of radiation was delivered within a period of 5 days. After 5 days, a 16-Gy dose of radiation was delivered within a 0- to 1-mm distance outward from the stents. The mean radiation dose within 4 to 5 mm outward from the stent was decreased by 96% compared with the mean dose within 0 to 1 mm outside the stent after a 5-day exposure. The decrease in radiation dose was 99.3% when measured 10 to 11 mm outward from the stent.
The rabbits did not present any macroscopic evidence of radiation damage at any time up to 1 year after radioactive stent implantation. Analysis of blood samples did not reveal radiation damage as indicated by leukopenia or thrombocytopenia.
The histological examination of arteries treated with RS compared with NRS revealed a dose-dependent inhibitory effect of radiation on neointima formation. Although the lowest dose (group 1) did not reduce neointima formation, groups 2 and 3 RS showed markedly inhibited neointimal hyperplasia compared with NRS 4 weeks after implantation (Fig 2⇓). However, the highest dose (group 3) was more effective than the intermediate dose (group 2). Table 2⇓ gives a summary of the comparisons of neointimal area (mm2)/arterial cross section 1, 4, 12, and 52 weeks after implantation. Vascular thrombosis after stent implantation was minor and did not differ between NRS and RS at all time intervals.
Immunocytochemistry and Transmission Electron Microscopy
The evaluation of arteries after conventional NRS implantation revealed degenerated medial SMCs compared with the media of uninjured arterial wall. A loss of medial SMCs in the strut region due to mechanical compression was evident by immunostaining and electron microscopy. The more intense extracellular matrix formation in the adventitia 4 and 12 weeks after NRS implantation compared with uninjured controls was probably due to vessel distension. The arterial morphology after stent implantation has already been analyzed in depth by other investigators.13
In this study, the cellular characteristics of the neointima of arteries after NRS and RS implantation were evaluated.
Smooth Muscle Cells
The cellularity of the neointima of arteries with RS was markedly smaller than that of the neointima after NRS implantation. SMCs immunoreactive to anti-SMC α-actin were found in the neointima of all stented arteries. Four weeks after NRS implantation, 705±23 SMCs/0.1 mm2 neointima were counted, compared with 366±18 after group 1 RS implantation (P<.001). At 4, 12, and 52 weeks after group 2 RS implantation, 242±10, 321±7, and 214±30 SMCs/0.1 mm2 neointima, respectively, were measured (P<.001 versus NRS after 4 weeks). After 1, 4, and 12 weeks, 315±59, 229±9, and 237±13 SMCs/0.1 mm2 neointima were found in arteries after group 3 RS implantation, compared with 870±44, 705±23, and 850±25 SMCs after NRS implantation, respectively (P<.001, Fig 3A⇓ and 3B⇓).
After double staining of PCNA and SMC α-actin was performed to identify the proliferating cells as SMCs, the number of PCNA-immunoreactive SMCs in the neointima was assessed. SMC proliferation was markedly inhibited by radioactive stents in a dose-dependent fashion (group 1<group 2<group 3) up to 52 weeks after implantation. The ratio of proliferating to nonproliferating SMCs expressed as percentage is given in Table 3⇓. At peak proliferation after 1 week, 30±2% of neointimal SMCs were PCNA immunoreactive in the arteries with NRS, compared with 0.5±0.1% PCNA-immunoreactive SMCs in arteries after group 3 RS implantation (P<.001).
In arteries with NRS, single macrophages and giant cells, both staining positive for antibody RAM-11, were observed around the stent struts after 1 week. After 4 and 12 weeks, only a few macrophages were still located around the struts of NRS. No difference was found after implantation of group 1 RS and NRS. In contrast, groups 2 and 3 RS presented more macrophage giant cells around the struts 4 weeks after implantation. After 12 and 52 weeks, macrophages were still found around the base of the radioactive wires (groups 2 and 3 RS); however, giant cells had disappeared.
After 1 week, arteries with NRS began to endothelialize. The endothelialization of vessels with RS was delayed in a dose-dependent fashion. The endothelial cell coverage of NRS and group 1 RS was completed after 4 weeks. In arteries irradiated with higher doses, endothelialization was present between the stent struts and, to a lesser degree, on top of the struts. After 4 weeks, 60% to 80% of the inner surface of RS was covered with endothelium (group 2>group 3), and the stent struts were partly covered with macrophages (group 2<group 3). Arteries exposed to group 2 RS for 12 and 52 weeks showed a completely restored endothelial lining, and arteries with group 3 stents were endothelialized by ≈95% after 12 weeks.
After 1 week, collagen type I was not found in the neointima of arteries stented with NRS or RS. Collagen type I immunostaining after NRS implantation at later times showed only weak staining intensity diffusely distributed in the neointima. An increased collagen type I content in the neointima after 4 weeks was already observed with group 1 RS but was more pronounced with groups 2 and 3 RS. In addition to extracellular staining, we observed an intracellular immunoreactivity after implantation of groups 2 and 3 RS (Fig 3C⇑ and 3D⇑), indicative of ongoing collagen type I production. Collagen type I was predominant around the stent struts of RS and in the adventitia close to the struts, whereas with NRS, collagen type I was distributed diffusely in the arterial wall. After 52 weeks, collagen type I staining was still intense in arteries with RS. Electron microscopy revealed that the extracellular matrix in the neointima covering NRS exhibited abundant microfibrils, whereas a dense fibrous material without microfibrils characterized the extracellular matrix of the neointima after RS implantation.
Arterial restenosis after PTCA or stent implantation in patients with coronary artery disease is caused by SMC proliferation and has not yet been successfully inhibited by pharmacological means.1 2 3 The severity of the initial arterial injury is one of the major determinants of postprocedural neointimal thickening.2 Vascular stent implantation induces comparatively more neointimal thickening than balloon angioplasty.14 Nevertheless, stent implantation is widely applied by clinicians because it reduces vascular elastic recoil after PTCA.15 In a previous study, we demonstrated that elastic recoil after experimental balloon angioplasty in a rabbit model markedly augments lumen narrowing compared with stent implantation.11 The initial gain in vessel diameter after stent implantation more than compensates for its enhanced neointimal thickening.16 Because metallic stents induce extensive neointima formation and cause restenosis, stents coated with eluting antiproliferative drugs have been developed. The initial results with these stents, however, have not shown any beneficial effect over conventional stents.17
In the present study, we demonstrated that neointimal hyperplasia after implantation of low-dose radioactive stents is inhibited compared with that with conventional stents. Endovascular ionizing irradiation of the vessel wall via radioactive stents prevents SMC proliferation. These stents can be manipulated by hand without extensive radiation protection. A radioisotopic license is not required for usage according to IAEA guidelines. Because the stents are not coated with radioisotopes, radioactive material does not enter the vessel wall and is not carried off with the bloodstream after stent expansion.
Neointimal Thickening and Cell Distribution
We observed a narrow margin of effective to ineffective inhibition of neointima formation within our dose range. The lowest radiation doses used in this study reduced SMC proliferation but did not change neointimal thickening, probably because of a reactive extracellular matrix formation in the neointima. However, the production of extracellular matrix did not induce excessive neointimal thickening. The most likely explanation for this observation is that SMC migration and proliferation are potently inhibited by ionizing radiation. The cellularity of the neointima covering radioactive stents was decreased compared with nonradioactive stents. However, electron microscopy did not reveal any morphological defects of neointimal SMCs after radioactive stent implantation. Migration of medial SMCs into the neointima occurred and may have been less severely depressed by radiation than the proliferation of SMCs. Although the endothelialization of radioactive stents was delayed, we did not observe an increase in thrombus formation in arteries treated with radioactive stents compared with nonradioactive stents. We consistently noted that macrophages covered the wires of radioactive stents before they were endothelialized. Despite inflammatory reactions of the vessels to both conventional and radioactive stent implantation, macrophages were more often found in arteries treated with radioactive stents. The presence of macrophages in the neointima after radioactive stent implantation, however, may be determined by a delay in endothelialization rather than by the radiation effect. The injury to the media was comparable after conventional and radioactive stent implantation and caused medial cell degeneration. Previously, medial cell degeneration has been related in particular to the mechanical compression of the arterial wall by stents.18
Several clinical studies have documented that radiation therapy in cancer patients can cause accelerated atherosclerosis and vascular stenosis several years after treatment.4 5 6 In our study, radiation-induced vascular stenosis was not observed. On the contrary, SMC proliferation and neointimal hyperplasia were inhibited by vascular irradiation. The integral radiation doses delivered by radioactive stents in our study are in the dose range of short-term irradiation via 192Ir sources used by other investigators, showing an inhibition of restenosis after angioplasty.7 19 The radiation dose via radioactive stents is emitted continuously, and thus, very low doses are delivered during the actual angioplasty. The radioactive stents used in this study emit β-particle radiation, low-energy x-radiation, and high-energy γ-radiation. The portion of high-energy γ-radiation emitted by these stents is small, eg, 0.7% of the total dose within 5 days, and may not have contributed to the observed effects.
Limitations of the Study
Radioactive stents were implanted only in a single animal species. Further studies using other animal species are necessary to evaluate the efficacy and safety of radioactive stents. Iliac arteries are elastic arteries, and the more muscular coronary arteries may require different radiation doses to prevent neointimal hyperplasia.
We report, for the first time, that very low ionizing radiation doses emitted via endovascular stents are effective over the long term in preventing SMC proliferation and neointimal hyperplasia after experimental angioplasty.
Presented in part at the 66th Scientific Sessions of the American Heart Association, Atlanta, Ga, November 8-11, 1993.
- Received January 9, 1995.
- Revision received March 21, 1995.
- Accepted March 26, 1995.
- Copyright © 1995 by American Heart Association
Liu MW, Roubin GS, King SB III. Restenosis after coronary angioplasty: potential biologic determinants and role of intimal hyperplasia. Circulation. 1989;79:1374-1387.
El-Naggar AM, El-Baz LM, Carsten AL, Chanana AD, Cronkite EP. Radiation induced damage of blood vessels: a study of dose-effect relationship with time after X-irradiation. Int J Radiat Biol. 1978;34:359-366.
Fehsenfeld P, Kleinrahm A, Schweikert H. Radionuclide technique in mechanical engineering in Germany. J Radioanal Nucl Chem. 1991;160:141-151.
Hehrlein C, Zimmermann M, Pill J, Metz J, Kübler W, von Hodenberg E. The role of elastic recoil after balloon angioplasty of rabbit arteries and its prevention by stent implantation. Eur Heart J. 1994;15:277-280.
Tanaka H, Sukhova GK, Swanson SJ, Clinton SK, Ganz P, Cybulsky MI, Libby P. Sustained activation of vascular cells and leukocytes in the rabbit aorta after balloon injury. Circulation. 1993;88(pt 1):1788-1803.
Schwartz RS, Murphy JG, Edwards WD, Camrud AR, Vlietstra RE, Holmes DR. Restenosis after balloon angioplasty: a practical proliferative model in porcine coronary arteries. Circulation. 1990;82:2190-2200.
De Jaegere PP, Hermans WR, Rensing BJ, Strauss BH, De Feyter PJ, Serruys PW. Matching based on quantitative coronary angiography as a surrogate for randomized studies: comparison between stent implantation and balloon angioplasty of native coronary lesions. Am Heart J. 1993;125:310-319.
Cox DA, Anderson PG, Rubin S, Chou CY, Argrawal SK, Cavender JB. Effect of local delivery of heparin and methotrexate on neointimal proliferation in stented porcine coronary arteries. Coron Artery Dis. 1992;3:237-248.
Schatz R, Palmaz J, Tio F, Garcia F, Garcia O, Reuter S. Balloon-expandable intracoronary stents in the adult dog. Circulation. 1987;76:450-475.
Waksman R, Robinson KA, Crocker IR, Cipolla GD, King SB III. Endovascular irradiation inhibits a restenosis-like response after balloon angioplasty of pig coronaries: dose-response relationship. Circulation. 1994;90(suppl I):I-142. Abstract.