Catheter and stent-based brachytherapy have shown promise

Transcription

1 Late Arterial Responses (6 and 12 Months) After 32 -Emitting Stent lacement Sustained Intimal Suppression With Incomplete Healing Andrew Farb, MD; Sweta Shroff, MS; Michael John, BA; William Sweet, MD; Renu Virmani, MD Background Three-month studies of stent-delivered brachytherapy in the rabbit model show reduced neointimal growth. However, intimal healing is delayed, raising the possibility that intimal inhibition is merely delayed rather than prevented. The purpose of this study was to explore the long-term histological changes after placement of -emitting radioactive stents in normal rabbit iliac arteries. Methods and Results Three-millimeter -emitting 32 stents (6, 24, and 48 Ci) were placed in normal rabbit iliac arteries with nonradioactive stents as controls. Animals were euthanatized at 6 and 12 months, and histological assessment, morphometry, and analysis of endothelialization were performed. Morphometric measurements demonstrated a 50% reduction in intimal growth and percent lumen stenosis within 24- and 48- Ci stents versus control nonradioactive stents at both 6 and 12 months. However, the 24- and 48- Ci stents also showed delayed healing of the intimal surface, characterized by persistent fibrin thrombus with nonconfluent areas of matrix, incomplete endothelialization, and increased intimal cellular proliferation. Stent edge stenosis was present at 12 months in the 24- and 48- Ci stent groups, characterized by both intimal thickening and negative arterial remodeling. Conclusions Inhibition of intimal growth is maintained 6 and 12 months after 32 -emitting stent placement. However, delayed arterial healing, incomplete endothelialization, and edge effects are present. (Circulation. 2001;103: ) Key Words: angioplasty atherosclerosis pathology stents Catheter and stent-based brachytherapy have shown promise for preventing coronary restenosis. 1 3 However, with radioactive stents, arterial healing is incomplete at 1 month in porcine coronary arteries treated with 3.0- Ci radioactive stents 4 and at 3 months in rabbit iliac arteries treated with 6- Ci stents. 5 The issue of incomplete healing by stentdelivered brachytherapy raises the question of whether early suppression of neointimal growth is sustained chronically. Further concerns with this device include the potential for increased neointimal growth 6 and the development of adverse edge effects, resulting in an increased frequency of non target-lesion revascularization. 7 The objectives of the present study were to analyze histologically neointimal suppression, healing responses, endothelialization, and edge effects at 6 and 12 months after 32 -emitting stent implantation in normal rabbit iliac arteries. Methods The protocol was approved by the Institutional Animal Care and Use Committee and conformed to the position of the American Heart Association on research animal use. Stent Treatment Groups BX stents ( mm; Isostent, Inc) were rendered radioactive by established ion implantation techniques. 8 A 6-, 24-, or 48- Ci or a nonradioactive control stent was randomly placed, under fluoroscopic guidance, in each iliac artery of male New Zealand White rabbits. At euthanasia, the iliac arteries were perfusion-fixed with 10% formalin. Animals received an injection of bromodeoxyuridine (BrdU) before euthanasia. 5 The cumulative 6- and 12-month radiation dose at a tissue depth of 0.5 mm from the mid portion of the stent was 95 Gy for the 6- Ci stent, 381 Gy for the 24- Ci stent, and 763 Gy for the 48- Ci stent. Angiographic Analysis restenting, immediately poststenting, and preeuthanasia iliac artery lumen diameters were measured with digital calipers. The arterial diameter just proximal and distal to the stent was measured at the time of euthanasia to determine angiographic edge effects. Tissue rocessing In brief, stented segments were embedded whole in methylmethacrylate as described previously. 5 Four-micrometer sections from the proximal, middle, and distal portion of the stents were stained with hematoxylin-eosin and Movat pentachrome stains. A 3.0-mm arterial Received June 22, 2000; revision received October 17, 2000; accepted October 18, From the Department of Cardiovascular athology, Armed Forces Institute of athology (A.F., S.S., M.J., R.V.), Washington, DC, and the Isostent Corporation (W.S.), Belmont, Calif. The opinions and assertions contained herein are the private views of the authors and are not to be construed as official or reflecting the views of the Department of the Army or the Department of Defense. Correspondence to Renu Virmani, MD, Department of Cardiovascular athology, Armed Forces Institute of athology, Washington, DC American Heart Association, Inc. Circulation is available at

2 Farb et al Long-Term Arterial Responses After -Emitting Stents 1913 TABLE 1. Morphometry at 6 and 12 Months Stent Activity n Neointimal Thickness, mm Neointimal Area, mm 2 Lumen Area, mm 2 % Stenosis Adventitial Thickness, mm 6 Months Control Ci * 24 Ci * * * * * 48 Ci * * * * * vs control; 0.02 vs 6 Ci * vs control; vs 6 Ci * 0.02 vs control and 6 Ci * vs control; vs 6 Ci * vs control 12 Months Control Ci * * 24 Ci * * * * * 48 Ci * * * * * vs control; vs 6 Ci; 0.04 vs 6 Ci n indicates number of stents in each treatment group. * vs control; vs 6 Ci * 0.03 vs control; 0.07 vs control * vs control; 0.02 vs 6 Ci * vs control segment just proximal and distal to the stents was processed and stained to evaluate edge effects. Morphometry All histological sections were magnified and digitized with the observer blinded to the treatment group. Computerized morphometry (ILab Spectrum software) was performed on stented segments to determine lumen area, area within the internal elastic lamina (IEL), neointimal area, percent luminal stenosis, neointimal thickness at and between each stent wire site, and adventitial thickness between each stent strut. To evaluate stent edge stenosis, the IEL area, external elastic lamina (EEL) area, and adventitial area were measured in the proximal and distal arterial edge segments. A negative remodeling index was defined as the ratio of the area within the outer boundary of the adventitia to the area within the EEL. This ratio normalizes the adventitial outer boundary to the vessel size (determined by the EEL), thus correcting for the expected increase in adventitia present in larger relative to smaller vessels. The maximal intimal thickness in the stent edges was measured. To assess intimal cellular proliferation, mid sections were incubated with a mouse monoclonal anti-brdu antibody, and BrdUpositive intimal cells were counted as a percent of total cells. 5 Actin-positive intimal cells and inflammatory cells (neutrophils and macrophages) in midstent sections were counted from 8 randomly selected 400 fields, and intimal cell densities were calculated. Stains for fibrin were performed, and the percentage of the intima occupied by fibrin was measured. Evaluation of Stent Endothelialization Stented iliac arteries were processed for scanning electron microscopy (SEM) as described previously. 5 The percent of the luminal surface that was endothelialized was determined from digitized photographs. Statistical Analysis Numerical data are presented as mean SD. Continuous variables were compared with an ANOVA. Results Thirty-six BX stents (6 Ci, n 10; 24 Ci, n 9; 48 Ci, n 9; nonradioactive controls, n 8) were deployed in 18 rabbits and used for light microscopy studies at 6 months. A total of 32 BX stents (6 Ci, n 7; 24 Ci, n 9; 48 Ci, n 8; nonradioactive controls, n 8) were deployed in 16 rabbits and used for light microscopy studies at 12 months. Twenty iliac artery stents (n 5 for each radioactive stent treatment group plus 5 control stents) were placed in 10 rabbits for endothelialization studies at 6 months and a similar number of stents and rabbits at 12 months. The vast majority of struts had injury scores of 0 or 1, with no differences among treatment groups. Angiographic Assessment Iliac artery lumen diameters before and immediately after stent placement were similar in all stent treatment groups (data not shown). At euthanasia (6 and 12 months), stent lumen diameters were similar in all groups. All stents were angiographically patent at the time of euthanasia without aneurysm formation. Histologic Morphometry Morphometric data are shown in Table 1 (6- and 12-month analyses). At 6 months, compared with control stents, there was a 50% reduction in mean intimal thickness, mean neointimal area, and mean percent arterial stenosis by the 24- and 48- Ci stents. Mean intimal thickness and area and percent arterial stenosis were similar between 6- Ci and control stents. Compared with control stents, lumen area was increased 17% ( 0.02) by the 24- Ci stents. At 12 months, there was maintenance of the 50% reduction in intimal thickness, neointimal area, and percent arterial stenosis by the 24- and 48- Ci stents. Compared with control stents, lumen area was increased 17% ( 0.01) by the 24- Ci stents and 15% ( 0.075) by the 48- Ci stents. There was a significant dose-dependent increase in mean adventitial thickness for all radioactive stent groups compared with controls at both 6 and 12 months (Table 1). Intimal suppression at 6 and 12 months was maximal in the mid portion of the stents.

3 1914 Circulation April 10, 2001 Figure 1. Iliac artery morphology at low-power and high-power (inset) 12 months after stent placement. Control stent is shown in panel A; collagen-rich neointima is present. No inhibition of neointimal growth is present with 6- Ci stent (B), and neointima contains greater concentration of proteoglycan (blue-green extracellular proteoglycan matrix). Significant reduction in neointimal thickness is present within 24- Ci (C) and 48- Ci (D) stents. Note increased adventitial thickness (arrows) associated with 6-, 24-, and 48- Ci stents. Movat pentachrome; bar 0.18 mm. Morphological Evaluation None of the stents was thrombotically occluded. The intima (neointima) in control stents was well developed at 6 and 12 months, consisting of smooth muscle cells (SMCs) in a proteoglycan/collagen matrix (Figure 1A, 12 months). The 6- Ci stents at 6 and 12 months (Figure 1B, 12 months) demonstrated a well-defined neointima, but compared with control stents, there was a relatively greater amount of proteoglycan-rich matrix with fewer SMCs. Clear evidence of delayed healing of the intima was present in the 24- and 48- Ci stent groups (Figure 1C and 1D, 12 months), characterized by a generally hypocellular intima containing fibrin and trapped erythrocytes around struts, with frequent inflammatory cells and incomplete endothelialization. SMCs were occasionally seen within the proteoglycan-rich hypocellular extracellular matrix, and there was medial thinning with medial SMC loss, especially beneath stent struts. The changes of delayed intimal healing were similar in the 24- and 48- Ci stents. Intimal Fibrin, Cellularity, roliferation, Atherosclerosis, and Endothelialization Intimal fibrin deposits were significantly more common in arteries containing radioactive stents. At 6 months, fibrin deposits accounted for % of the total intimal area of the mid segment of control stents compared with % for 6- Ci stents ( 0.05), % for 24- Ci stents ( 0.03 versus control and 0.03 versus 6

4 Farb et al Long-Term Arterial Responses After -Emitting Stents 1915 Figure 2. Dose-dependent increase in intimal fibrin deposition (brownish-red reaction product) and cell proliferation (brownish-red stained nuclei) 12 months after deployment of control (A and E), 6- Ci (B and F), 24- Ci (C and G), and 48- Ci (D and H) stents. A through D, Anti-fibrin antibody staining with hematoxylin counterstain, bar 0.15 mm; E-H, anti-brdu antibody staining with hematoxylin counterstain, bar 0.02 mm. Ci), and % for 48- Ci stents ( 0.05 versus control and 6 Ci). At 12 months, intimal fibrin deposits (Figure 2, A through D) were not seen in the control stents and accounted for % of the intimal area within the 6- Ci stents compared with % for the 24- Ci stents ( versus control and versus 6 Ci) and % for the 48- Ci stents ( versus control and 0.02 versus 6 Ci). Levels of cell proliferation were elevated in all radioactive stents at 6 and 12 months versus controls (Table 2 and Figure 2, E through H). Total intimal cell density and SMC density were substantially reduced in the high-activity stents at 6 and 12 months (Table 2), but of the cells present, there were greater numbers of inflammatory cells (Figure 3, A through D). Focal atherosclerotic change, consisting of intimal collections of foam cells with or without cholesterol clefts and focal calcification, was observed in the intima of one 6- Ci stent, one 24- Ci stent, and three 48- Ci stents at 6 months and one 24- Ci stent and four 48- Ci stents at 12 months (Figure 3E). The absence of complete healing was confirmed by SEM. At 6 months, only % and % of the intimal TABLE 2. Intimal Cell Density and Cell roliferation (BrdU-Labeling Index) at 6 and 12 Months surface was endothelialized in the 24- and 48- Ci stents, respectively, compared with % for 6- Ci stents and % for control stents ( 0.005). Endothelialization remained incomplete even at 12 months with these higherradioactivity stents: % and % of the 24- and 48- Ci stent surface was endothelialized, respectively. In the 24- and 48- Ci stents, endothelialization was present at the ends of the stents, and the central portion of the stents was only partially endothelialized with focally adherent inflammatory cells and platelets (Figure 4). Edge Effects At 12 months (Table 3), all radioactive stent groups had significantly greater maximal intimal thickness versus controls in the distal nonstented edge segment, with a smaller lumen diameter, increased percent stenosis, and increased negative remodeling index in the 24- and 48- Ci groups (Figure 5). The IEL and EEL areas were smallest in the 48- Ci stents. Adverse edge effects were present but slightly less marked in the proximal nonstented arterial segment (Table 3). At 6 months (data not shown), maximal intimal 6 months 12 months Inflammatory Stent Activity TICD, cells/mm 2 SMCs, cells/mm 2 Inflammatory Cells, cells/mm 2 BrdU Labeling Index, % TICD, cells/mm 2 SMCs, cells/mm 2 Cells, cells/mm 2 BrdU Labeling Index, % Control Ci * * * * * * * * 24 Ci * * * * * * * * 48 Ci * 27 20* * * * * * * 0.03 vs control * vs control; 0.04 vs 6 Ci * vs control; 0.02 vs 6 Ci * 0.02 vs control; 0.05 vs 6 Ci * vs control; 0.05 vs control; 0.03 vs 48 Ci BrdU-labeling index (total BrdU-positive cells/total cells) 100%; TICD, total intimal cell density. * vs control * vs control; 0.04 vs 6 and 24 Ci * 0.04 vs control; 0.07 vs 6 Ci

5 1916 Circulation April 10, 2001 Figure 3. Intimal cellularity at 12 months. Control stent intima (A) consists of numerous spindle-shaped SMCs. Six-microcurie stent intima (B) is relatively hypocellular and contains occasional mononuclear cells. Intima within 24- and 48- Ci stents (C and D, respectively) contains rare spindle-shaped cells and frequent inflammatory cells with focal collections of red blood cells and fibrin. E, Focal atherosclerosis with large needle-shaped cholesterol clefts (arrows) with focal foam cells within 48- Ci stent. Hematoxylin-eosin; bar 0.05 mm in A through D, 0.11 mm in E. thickness in the distal nonstented arterial segment was significantly greater ( 0.03) in the 48- Ci group than in controls, and angiographic lumen diameter was smallest in the 48- Ci group ( 0.04 versus other groups). The negative remodeling index was significantly greater in the 48- Ci stents compared with controls. In the proximal nonstented edge segment, there were trends ( 0.10) toward a smaller lumen diameter, greater percent stenosis, and a greater negative remodeling index in the 48- Ci stents compared with controls. Discussion This is the longest and largest experimental study that confirms long-term suppression of in-stent neointimal growth with 32 -emitting stents. Concerns raised about whether the intimal suppression seen at 3 months would be maintained appear to be allayed by the current 6- and 12-month data. Furthermore, this modality appeared to be safe and well tolerated in this animal model, and there were no cases of late occlusive stent thrombosis. However, numerous observations indicate persistent delayed healing of the intimal surface: incomplete stent endothelialization, intimal fibrin deposition, ongoing cellular proliferation, and inflammation. SEM clearly showed impaired endothelialization out to 1 year, in contrast to previous studies in which complete endothelialization was suggested after only limited analysis. 9 These data provide morphological insights into the initial results seen with the use of -emitting radioactive stents in humans. Mechanisms of Delayed Healing and Endothelialization Reestablishment of an intact endothelium is believed to be a critical factor in limiting the neointimal expansion that produces restenosis. 10,11 After endothelial denudation, endothelialization proceeds, via cell migration and proliferation, from the edges of the denuded segment toward the center. 12 Migration of endothelial cells from branch vessels and vasa vasorum toward the center of the stent is another source for intimal reendothelialization. One month after nonradioactive stent deployment in rabbits, the neointima is fully healed, stent endothelialization is complete, and intimal SMC proliferation is at low levels. 13,14 However, in the present study, healing and endothelialization remain far from complete out to 1 year after placement of 24- and 48- Ci stents despite extremely low levels of radioactivity levels below which biologic activity might not be expected to be present. For example, the radioactivity present on the 24- Ci stent fell below 0.1 Ci on day 129. The radiation delivered may have had its desired effect of producing lethal DNA damage to the local SMCs and endothelial cells during cell division. It is uncertain whether the remaining SMCs and endothelial cells within the stent and at the stent margins have sufficient proliferative potential to ever completely heal the intimal surface. Adventitial fibrosis and radiation-induced damage to branch vessels and vasa vasorum may deplete the pool of cells necessary to restore a complete endothelial lining. Furthermore, injury to the arterial wall may interfere with endothelial cell and SMC migration and adhesion even if cell proliferation continues. resently, no long-term animal study has demonstrated a brachytherapy regimen (beta or gamma, stent or wire) that both effectively inhibits neointimal growth and results in complete healing. In humans, the timing of complete intimal healing remains to be established. An intact endothelial lining also plays an antithrombotic role, and one can postulate that impaired endothelialization may be responsible for the persistent intimal fibrin deposition observed out to 12 months in the present study. Incomplete endothelialization and lumen thrombus are likely to be a feature of other methods of arterial brachytherapy; Salame et al 15 showed impaired endothelialization, persistent intramural

6 Farb et al Long-Term Arterial Responses After -Emitting Stents 1917 Figure 4. Low-power SEM of control stent (A) and 48- Ci stent (B) at 12 months. Entire control stent is endothelialized, whereas 48- Ci stent is endothelialized at ends (* in B). arts of central portion of 48- Ci stent are nonendothelialized, and bare or partially covered struts are present. Transition zone between endothelialization and nonendothelialization is outlined in B and shown at medium power in C. Two areas are further magnified (arrow and box in C corresponding to D and E, respectively) to show adherent platelets and inflammatory cells on partially endothelialized (D) and nonendothelialized (E) parts of stent. Bars 0.5 mm in A and B, 0.1 mm in C, 0.01 mm in D and E. hemorrhage, and increased platelet recruitment 1 month after overstretch balloon injury and 90 Y/Sr brachytherapy utilizing a catheter-based system. Concerns regarding a persistently prothrombotic lumen surface have led to the practice of long-term antiplatelet therapy in patients treated with coronary brachytherapy. This recommendation followed the reported 6.6% incidence of sudden thrombotic events 2 to 15 months after TCA in patients treated with intracoronary -radiation. 16 The incompletely endothelialized intima with persistent fibrin deposition, observed in the present study, may be an important substrate for late thrombosis. Animal Models of -Emitting Stents To date, rabbit, porcine, and canine arteries have been used to study these brachytherapy devices. Rabbit studies have consistently yielded intimal suppression at 3 months. 5,9 Before the present study, few long-term data were available, and 2 studies showed no intimal suppression with radioactive stents placed in dogs 17 and pigs. 6 The results of the present study suggest the rabbit iliac artery may be the superior animal model to study responses likely to be seen in humans. In the recently reported Milan Dose-Response Study 7 of 32 -emitting stents, there was a dose-dependent reduction in pure intrastent restenosis rates at 6-month follow-up: 16%, 3%, and 0% in arteries treated with to 3.0- Ci, 3.0- to 6.0- Ci, or 6.0- to Ci stents, respectively. The authors noted near-complete inhibition of neointimal growth, at stent activities 3 Ci, in the mid portion of the stent, with increased tissue ingrowth toward the stent margins. This impressive intimal suppression was similar to that seen in the present study out to 1 year with 24- and 48- Ci stents. Furthermore, the mechanism of edge restenosis in the present study involved the combined effects of intimal growth and negative remodeling, similar to data

7 1918 Circulation April 10, 2001 Figure 5. Edge effects at 12 months. Arterial segment distal to control stent (A and B) and 48- Ci stent (C and D) are shown. Adventitia is thin (adv), and there is minimal intimal thickening (arrowhead) distal to control stent (B). Distal to 48- Ci stent, there is eccentric, proteoglycan-rich intimal thickening (arrows) with negative remodeling characterized by increased adventitial thickness (adv) and smaller area enclosed within IEL (A and C). Movat pentachrome; bars 0.13 mm in A and C, 0.06 mm in B and D. TABLE 3. Distal and roximal Stent Edge Effects at 12 Months Stent Activity Lumen Diameter, mm Neointimal Thickness, mm % Stenosis IEL Area, mm 2 EEL Area, mm 2 Negative Remodeling Index Distal stent edge effects Control Ci * Ci * * Ci * * * * * * * 0.04 vs control and 6 Ci; 0.06 vs control and 6 Ci roximal stent edge effects * 0.02 vs control * 0.03 vs control * 0.04 vs control and 6 Ci * 0.05 vs control and 6 Ci * 0.02 vs control and 6 Ci; 0.09 vs control and 6 Ci Control Ci Ci * 48 Ci * * * 0.01 and 0.07 vs control NS * 0.02 vs control, 6 Ci, and 24 Ci NS NS * 0.04 vs control and 6 Ci; 0.04 vs 6 Ci Negative remodeling index adventitial area/eel area (see text). Lumen diameter measured from angiograms; other variables from histology.

8 Farb et al Long-Term Arterial Responses After -Emitting Stents 1919 derived from intravascular ultrasound studies in the restenosis cases of human radioactive stent implants. 7 Atherosclerosis In-stent atherosclerotic lesions observed in a minority of the stents at 6 and 12 months were an unexpected finding in normal rabbit arteries, a model that does not develop atherosclerosis in the absence of hypercholesterolemia. Radiation therapy induced accelerated atherosclerosis is a recognized complication in long-term surviving patients treated with external-beam radiation (XRT). 18 The latency period for this complication is typically 10 years. 19 One cannot extrapolate the data in the present study to suggest that accelerated atherosclerosis is likely to occur in patients treated with -emitting stents. On the other hand, patients who develop significant radiation-induced accelerated atherosclerosis presumably had no or minimal coronary disease before the initiation of XRT, in contrast to coronary brachytherapy patients who already have advanced atherosclerosis. Therefore, any latent tendency toward atherosclerosis progression by coronary brachytherapy could have negative clinical consequences. Study Limitations The major finding of the present study is the demonstration of incomplete intimal healing concurrent with intimal growth suppression by radioactive stents. However, only selected stent activities were examined (6, 24, and 48 Ci). Additionally, results from radioactive stent deployment in normal peripheral arteries may differ from stent placement in atherosclerotic epicardial coronary arteries. Conclusions At 12 months, 32 -emitting stents show a treatment effect consisting of reduced intimal thickness and area, increased lumen area, and reduced luminal percent stenosis. However, impaired intimal healing persists, characterized by late luminal fibrin deposition, inflammation, incomplete endothelialization of the stent surface, and ongoing cellular proliferation. Neointimal growth and negative arterial remodeling contribute to stent edge effects. The finding of incomplete healing strongly supports the use of prolonged antithrombotic therapy. Longer-term follow-up studies are needed to assess the significance of atherosclerotic change. Acknowledgment This study was supported by a grant from Isostent, Inc, Belmont, Calif. References 1. Wardeh AJ, Kay I, Sabate M, et al. -article-emitting radioactive stent implantation: a safety and feasibility study. Circulation. 1999;100: Teirstein S, Massullo V, Jani S, et al. Three-year clinical and angiographic follow-up after intracoronary radiation: results of a randomized clinical trial. Circulation. 2000;101: Waksman R, White RL, Chan RC, et al. Intracoronary gamma-radiation therapy after angioplasty inhibits recurrence in patients with in-stent restenosis. Circulation. 2000;101: Carter AJ, Laird JR, Bailey LR, et al. Effects of endovascular radiation from a -particle-emitting stent in a porcine coronary restenosis model: a dose-response study. Circulation. 1996;94: Farb A, Tang AL, Shroff S, et al. Neointimal responses 3 months after 32 -emitting stent placement. Int J Radiat Oncol Biol hys. 2000;48: Carter AJ, Scott D, Bailey LR, et al. Dose-response effects of 32 radioactive stents in an atherosclerotic porcine coronary model. Circulation. 1999;100: Albiero R, Adamian M, Kobayashi N, et al. Short- and intermediate-term results of 32 radioactive -emitting stent implantation in patients with coronary artery disease: the Milan Dose-Response Study. Circulation. 2000;101: Carter AJ, Laird JR. Experimental results with endovascular irradiation via a radioactive stent. Int J Radiat Oncol Biol hys. 1996;36: Hehrlein C, Stintz M, Kinscherf R, et al. ure -particle-emitting stents inhibit neointima formation in rabbits. Circulation. 1996;93: Clowes AW, Reidy MA, Clowes MM. Kinetics of cellular proliferation after arterial injury, I: smooth muscle growth in the absence of endothelium. Lab Invest. 1983;49: Asahara T, Bauters C, astore C, et al. Local delivery of vascular endothelial growth factor accelerates reendothelialization and attenuates intimal hyperplasia in balloon-injured rat carotid artery. Circulation. 1995;91: Reidy MA, Clowes AW, Schwartz SM. Endothelial regeneration, V: inhibition of endothelial regrowth in arteries of rat and rabbit. Lab Invest. 1983;49: Van Belle E, Maillard L, Tio FO, et al. Accelerated endothelialization by local delivery of recombinant human vascular endothelial growth factor reduces in-stent intimal formation. Biochem Biophys Res Commun. 1997; 235: Bai H, Masuda J, Sawa Y, et al. Neointima formation after vascular stent implantation: spatial and chronological distribution of smooth muscle cell proliferation and phenotypic modulation. Arterioscler Thromb. 1994;14: Salame MY, Verheye S, Mulkey S, et al. The effect of endovascular irradiation on platelet recruitment at sites of balloon angioplasty in pig coronary arteries. Circulation. 2000;101: Costa MA, Sabate M, van der Giessen WJ, et al. Late coronary occlusion after intracoronary brachytherapy. Circulation. 1999;100: Taylor AJ, Gorman D, Farb A, et al. Long-term coronary vascular responses to 32 -particle-emitting stents in a canine model. Circulation. 1999;100: Stewart JR, Fajardo LF, Gillette SM, et al. Radiation injury to the heart. Int J Radiat Oncol Biol hys. 1995;31: Hancock SL, Tucker MA, Hoppe RT. Factors affecting late mortality from heart disease after treatment of Hodgkin s disease. JAMA. 1993; 270: