¹ From the Experimental Cardiology Laboratory, University Medical Center, Heidelberglaan 100, Room G02.523, Utrecht, the Netherlands. Received March 6, 2001; revision requested April 9; revision received June 1; accepted July 5. Supported by grant NHS 99.209 from the Dutch Heart Foundation. M.J.S. supported by the Sorbo Foundation.

	ABSTRACT

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
PURPOSE: To determine the minimal duration of oral matrix metalloproteinase (MMP) inhibition to prevent constrictive remodeling after balloon dilation.

MATERIALS AND METHODS: In 37 nonatherosclerotic pigs, balloon dilation was performed in 145 peripheral arteries. Pigs were treated with an MMP inhibitor for 2, 7, 14, 28, or 42 days, or they served as controls and were killed 42 days after intervention. Arteries were visualized with angiography and intravascular ultrasonography.

RESULTS: A 69% reduction in late vessel area (VA) loss was achieved after 14 days of treatment: 1.27 mm² ± 0.55 (standard error of the mean [SEM]) versus 4.04 mm² ± 0.93 (SEM) in the control group (P = .1). A consistent inhibition of late VA loss was observed in the 28-day (0.89 mm² ± 0.83 [SEM], P = .03) and 42-day (0.74 mm² ± 0.66 [SEM], P = .02) groups treated with the MMP inhibitor. After 14 and 28 days of treatment, late lumen area loss was 65% and 55% of control values, and it decreased to 41% (P = .04) after 42 days of treatment.

CONCLUSION: MMP inhibition for 14–28 days is sufficient to inhibit constrictive remodeling after balloon dilation. This implies that an essential MMP-dependent initiator of constrictive remodeling was mainly active in the first 2 weeks after intervention.

Index terms: Animals • Arteries, peripheral, 92.454, 98.454 • Arteries, transluminal angioplasty, 92.454, 98.454 • Arteries, US, 92.1298, 98.1298 • Experimental study • Ultrasound (US), intravascular, 92.1298, 98.1298

	INTRODUCTION

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In addition to elastic recoil and intimal hyperplasia, constrictive arterial remodeling is an important determinant of restenosis after balloon angioplasty (1). Arterial remodeling is a time-dependent phenomenon. In peripheral arteries of pigs, constrictive remodeling starts within days after balloon angioplasty and progresses for as long as 6 weeks of follow-up (2). It is unknown what initiates or maintains this remodeling response.

Matrix metalloproteinases (MMPs) belong to a group of zinc- and calcium-dependent proteases and cause breakdown of the extracellular matrix, which is an important feature in arterial remodeling. After balloon angioplasty, a transient increase in MMP activity has been observed (3). In the pig, inhibition of MMP activity with oral administration of MMP inhibitor (BB-2516, Marimastat; British Biotech Pharmaceuticals, Oxford, England) (4) substantially reduces late lumen area (LA) loss after balloon dilation by means of inhibition of constrictive remodeling. BB-2516 has been clinically applied to prove its role as a tumorostatic agent (5). It is well absorbed and has excellent bioavailability. BB-2516 is composed of a synthetic molecule with a molecular weight of 331.42 that is designed to mimic the substrate of MMPs. One substituent of the molecule of BB-2516 is a hydroxamate group that will combine with the zinc atom at the active site of the MMPs. This feature, together with stereochemical aspects in the molecule of BB-2516 that mimic the natural substrate, leads to potent reversible inhibition of the majority of MMPs.

A short duration of treatment necessary to inhibit constrictive remodeling may limit potential side effects of MMP inhibitors and is cost-effective (5). However, it is still unknown how long MMP inhibition is needed to block the remodeling process permanently. The purpose of the present study was to determine the minimal duration of oral MMP inhibition to prevent constrictive arterial remodeling after balloon dilation.

	MATERIALS AND METHODS

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Animals
Thirty-seven nonatherosclerotic Landrace pigs with an average weight of 22.6 kg ± 2.8 (SD) were included in this study. This population is different from that in our previous study with BB-2516 (4). Balloon dilation was performed bilaterally in femoral and internal iliac arteries; however, it was not performed in two femoral arteries and one internal iliac artery because the balloon catheter could not be advanced owing to extreme vessel tortuosity. All animals were killed 42 days after intervention, except for three of nine pigs that were treated with BB-2516 for 42 days and killed 84 days after intervention to investigate whether a delayed remodeling response would occur. The investigation was approved by the Ethics Committee on Animal Experimentation of the Faculty of Medicine, Utrecht University, the Netherlands.

Anesthesia
During intervention and killing, the animals were anesthetized with an intravenous dose of 0.3 mg of midazolam hydrochloride per kilogram of body weight per hour and 2.5 µg/kg/h of sufentanil. They were ventilated with a mixture of O₂, air in a volume-to-volume ratio of 1:1, and 1% halothane after premedication with 4 mg/kg of azaperone, 10 mg/kg of ketamine hydrochloride, and 4 mg/kg of thiopental sodium.

Intervention
All experiments were attended by two authors (M.J.S., E.V.), who performed the catheterization procedures in a random order. Acetylsalicylic acid (80 mg daily) was administered to all pigs, starting with 320 mg 1 day prior to the intervention. Heparin sodium (100 IU/kg) was administered to the animals, and a continuous infusion of nitroglycerin (20 µg/min) was administered to prevent arterial spasm. The arterial tree was accessed via the right carotid artery. An arterial 8-F sheath was introduced into the descending aorta.

For balloon dilation, a 2–4-cm long 4–6-mm-diameter standard peripheral balloon catheter (Opta 5; Cordis Endovascular, Warren, NJ) was used. The adequate balloon size was chosen to achieve a balloon-to-artery ratio of approximately 1.2. We performed balloon dilation with the following balloon sizes in 145 peripheral arteries: in 13 arteries, 4-mm-diameter balloon; in 12 arteries, 4.5-mm-diameter balloon; in 53 arteries, 5-mm-diameter balloon; and in 67 arteries, 6-mm-diameter balloon. In the femoral arteries, balloon dilation was performed between the two main side branches near the hip joint. In the internal iliac arteries, the midsegments were dilated with the balloon. The balloon was inflated three times for 1 minute at a pressure of 6–10 atm. After intervention, the right carotid artery was ligated. For adequate pain relief, 10 µg/kg of buprenorphine hydrochloride was intramuscularly injected immediately after intervention and at 2 days afterward. The pigs were prophylactically treated with amoxicillin, starting with 250 mg administered intravenously at intervention (day 0) and thereafter with 375 mg of long-acting medication administered intramuscularly at days 1, 3, and 5.

Angiography and Intravascular Ultrasonography
Angiography and intravascular ultrasonography (US) were performed before and after intervention and at killing. At killing, access was achieved through a left carotid artery approach. Radiopaque rulers were used to localize the treated segments. First, 0.5 mg of nitroglycerine was selectively injected to prevent spasm. Then, contrast material (Telebrix; Guerbet, Roissy, France) was injected selectively into each artery. Intravascular US was performed with a 4.1-F 30-MHz US catheter (Princeps; Du-MED/Endosonics Europe, Rijswijk, the Netherlands) with a transverse resolution of 0.1 mm. Fluoroscopy was performed during intravascular US to document the images relative to an anatomic landmark for adequate matching. During manual pullback and at regular (0.5-cm) intervals, while the intravascular US catheter was held, the images were recorded on VHS videotape for analysis with a digital video analyzer. After follow-up angiography and intravascular US, the animals were killed with an overdose of pentobarbital sodium.

MMP Inhibition
Starting 1 day prior to the intervention, 10 mg of BB-2516 per kilogram of body weight was mixed with food and administered twice a day for 2, 7, 14, 28, or 42 days in four, four, six, six, and nine pigs, respectively. Eight pigs served as controls. In the 42-day treatment group, three of nine pigs were killed at 84 days. Findings in an unpublished dosimetry study in Landrace pigs in our laboratory established that a dose of 10 mg/kg twice a day would provide an exposure of 100–200 ng/mL of plasma, a concentration that had previously been effective in animal models of cancer. In a previous study (4), 1–10 ng/mL of functional BB-2516 was detected in vessel extracts of treated animals.

Data Analysis
The intravascular US scans were analyzed at regular intervals (every 0.5 cm) by one experienced observer (M.J.S.). An additional observer (G.P.) analyzed scans in case delineation of the different layers within the intravascular US scan was difficult to assess. Consensus was reached with the first observer in all cases.

In each intravascular US scan, LA and vessel area (VA) were measured. In this nonatherosclerotic model, intimal hyperplasia was absent at baseline. VA before and after intervention was, therefore, defined as the outer border of the hypoechoic lumen within the intravascular US scan, and it was equal to LA before and after intervention. At killing, VA was defined as the outer border of the hypoechoic layer behind the hyperechoic intima and therefore represents luminal, intimal, and medial areas.

Anatomic landmarks, visualized with angiography and intravascular US, were used to match the image locations at different times. The segment dilated with the balloon was identified by means of comparison with the angiogram, and this finding was confirmed by the existence of acute LA gain, which was defined as the difference between postintervention and preintervention LA. Untreated segments (proximal parts of external iliac arteries, parts proximal and distal to the dilated segments of the femoral arteries, and parts distal to the dilated segments of the internal iliac arteries) were used for correction of growth of all areas by means of the following calculation: 1 - ([mean r²_k - mean r²_p]/[mean r²_p]), where k signifies killing and p signifies preintervention status. The radius, or r, was determined angiographically. Within the segment dilated with the balloon, the location with the smallest LA at follow-up was selected for further calculations.

Definitions
Late LA loss was calculated as postintervention LA minus follow-up LA. Loss in VA, a measure of remodeling, was calculated as postintervention VA minus follow-up VA. Intimal hyperplasia was calculated as follow-up VA minus follow-up LA.

Table 1 displays the number of pigs represented in each treatment group, the number of arteries in each group that were dilated with the balloon with an intention to treat, and the number of arteries on which the results were based. Arteries were excluded because of lack of gain, which indicated local spasm or immediate procedural failure (gain  0 mm, seven arteries), procedure-induced aneurysm formation (six arteries), extravasation (12 arteries), excessive thrombus formation (one artery), intravascular US recording of related problems (four arteries), or untimely death (ventricular fibrillation, four arteries; retroperitoneal bleeding due to the intervention, four arteries).

	RESULTS

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
An example of intravascular US scans is shown in Figure 1 for both the group treated with BB-2516 and the control group. Table 2 lists the intravascular US measurements at different times. The growth correction factor was 0.74 ± 0.03 (standard error of the mean [SEM]) for the control group and 0.75 ± 0.02 (SEM) and 0.79 ± 0.09 (SEM) for the groups treated with BB-2516 with a follow-up of 42 days and with a prolonged follow-up of 84 days, respectively. The increase in weight during follow-up was 11.3 kg ± 0.6 (SEM) in the control group and 10.4 kg ± 0.5 (SEM) and 12.8 kg ± 0.6 (SEM) in the groups treated with BB-2516 with a follow-up of 42 days and with a prolonged follow-up of 84 days, respectively. Figure 2 shows the effect of BB-2516 for different durations of administration on late VA loss, late LA loss, and intimal hyperplasia at 42 days of follow-up. Acute LA gain did not differ significantly (P = .98) among groups.

	DISCUSSION

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The principal findings of the present study were that oral MMP inhibition with BB-2516 for 14–28 days seemed sufficient to inhibit constrictive remodeling after balloon dilation. Furthermore, findings in the group of animals killed 42 days after therapy provided evidence that a delayed constrictive remodeling response did not occur.

MMP Inhibitor and Constrictive Remodeling
In the pig, inhibition of MMP activity after balloon angioplasty by means of either intraperitoneal injection of another agent (BB-93, Batimastat; British Biotech Pharmaceuticals) (6) or oral administration of BB-2516 (4) markedly reduced late LA loss through inhibition of constrictive remodeling.

Although the cellular and molecular basis of constrictive arterial remodeling after balloon angioplasty is still unclear, parallels can be drawn to wound healing. In the atherosclerotic monkey, Geary et al (7) showed that the pattern of matrix and integrin expression within the injured wall is in many ways analogous to that of healing wounds. In an in vitro wound contraction model (8), BB-2516 inhibited lattice contraction, mediated by fibroblasts, which is a modeling that may occur when granulation tissue contracts in a healing wound.

Arterial remodeling after balloon angioplasty is a time-dependent phenomenon. In the atherosclerotic pig (2), constrictive remodeling in peripheral arteries starts within days after balloon angioplasty and progresses for as long as 42 days. In the nonatherosclerotic pig, constrictive remodeling started between 7 and 14 days after intervention (Sierevogel MJ, unpublished data, 2001). Surprisingly, the present study showed that 14-day MMP inhibition seemed sufficient to inhibit constrictive remodeling. Furthermore, an additional follow-up of 42 days after 42 days of treatment did not result in a catch-up of the remodeling response, as previously reported (9) for intimal hyperplasia. These results indicate that the effect of MMP inhibition precedes the geometric remodeling response. This implies that an essential initiator of the constrictive remodeling response, related to MMP activity, is mainly active in the first 2 weeks after intervention and is blocked by the MMP inhibitor.

Strauss et al (3) demonstrated that peak collagen synthesis and degradation occur at 7 days after balloon angioplasty. MMPs play a role not only in collagen turnover (3) and elastin metabolism (10) but also in processes such as solubilization of plasma membrane receptors through proteolytic cleavage of the ligand-binding domain at the cell surface, which is called shedding (11), activation of integrins (12), regulation of vascular reactivity (13), and inflammation (14). Consequently, the effects of MMP inhibition by means of a nonspecific MMP inhibitor are likely the complex result of general inhibition of various processes mediated by MMPs.

MMP Inhibitor and Intimal Hyperplasia
In our previous studies with MMP inhibitors (4,6), intimal hyperplasia at the 42-day follow-up did not differ markedly among control groups and groups treated with MMP inhibitors, which was probably due to a catch-up phenomenon (9). In this study, augmented intimal hyperplasia at 42 days was found after 7 days of treatment with BB-2516; however, this difference was not significant (P = .12). This difference might be partly explained by the slightly larger (but not noticeable) LA gain as a result of the procedure in this group, although we previously demonstrated that in this model the effect of acute LA gain on intimal hyperplasia after balloon dilation is relatively small (4).

Other Methods to Reduce Constrictive Remodeling
In addition to MMP inhibition, other methods have been reported that reduce constrictive remodeling after balloon angioplasty. Placement of a percutaneous endovascular stent eliminates constrictive remodeling mechanically (15), but it also results in excessive intimal growth. The latter leads to augmented risk of in-stent restenosis in small arteries (16).

Constrictive remodeling has been inhibited nonmechanically as well. After coronary angioplasty in the pig, both intimal hyperplasia and constrictive remodeling are attenuated by means of inhibition of tyrosine kinases, transducers of a variety of extracellular signals that regulate smooth muscle cell proliferation, and differentiation (17).

In humans, the antioxidant probucol exerts its antirestenotic effects by enhancing expansive remodeling after angioplasty (18). In addition, Daida et al (19) recently demonstrated the potential benefits of probucol administration in prevention of restenosis.

In several animal studies, another target for pharmacologic intervention in restenosis has been the dysfunctional endothelium. The nitric oxide precursor L-arginine reduces neointima formation, but the effect on arterial remodeling varies from none (20) to reduced constrictive remodeling (21). With adenovirus-mediated transfer of human endothelial nitric oxide synthase in the pig, Varenne et al (22) showed a reduction in late LA loss after balloon angioplasty by means of both reduction of neointima and enlargement of VA.

In humans, ß-radiation therapy after balloon angioplasty increases VA at the 6-month follow-up (23). In the study of Sabaté et al (23), however, LA remained unaltered since plaque area increased as well.

Compared with other methods to prevent restenosis, probucol and BB-2516 have the advantage of being orally administered.

Limitations
The time during which constrictive remodeling develops differs between humans and animals. In humans (24), the coronary artery shows expansive remodeling, together with neointima formation, in the 1st month, whereas constrictive remodeling is observed 1–6 months after the intervention. This study shows that the effect of MMP inhibition preceded the geometric constrictive remodeling response. Thus, it is unlikely that treatment will need to exceed 1 month.

This study has been performed in a nonatherosclerotic model, which may hamper extrapolation of the results to the human atherosclerotic artery. However, our previous studies with MMP inhibitors were performed in atherosclerotic (6) and nonatherosclerotic (4) models. In both models, MMP inhibition markedly reduced constrictive remodeling after balloon dilation. One must keep in mind, however, that in both models, balloon dilation resulted in an overdilation of the artery, which could result in reduced local shear force. This is known to induce MMP activity (25) and might also be influenced by MMP inhibition. Late VA loss, the primary endpoint of the present study, was influenced by treatment with BB-2516. We also determined the effect on late LA loss and intimal hyperplasia. However, since we did not correct for multiple testing, these secondary outcome measurements cannot be considered conclusive.

In the pig, oral MMP inhibition for 14–28 days was sufficient to inhibit constrictive arterial remodeling after balloon dilation. This implies that an essential MMP-dependent initiator of the constrictive remodeling response was mainly active in the first 2 weeks after intervention. The short treatment necessary to inhibit constrictive remodeling limits potential side effects and may make BB-2516, or comparable compounds, suitable for prevention of restenosis after balloon angioplasty in humans.

Practical application: To our knowledge, BB-2516 is the first orally administered MMP inhibitor to be included in clinical trials in the field of oncology (5). It has excellent bioavailability, and phase I, II, and some phase III trials with it have been completed. In phase I studies, short courses of BB-2516 were well tolerated by healthy volunteers. With long-term treatment, however, patients with various malignancies began to experience joint and muscle pain. This was seen in more than 60% of patients receiving a dose of BB-2516 greater than 50 mg twice a day. The symptoms were reversible after discontinuation of the drug, and occurrence of symptoms was reduced by using a 10-mg dose of BB-2516 twice a day. Improved survival was not observed in the completed phase III studies of pancreatic or gastric carcinoma and glioma. It is likely that MMP inhibitors will be most effective in the setting of minimal tumor volume. Phase III studies of coadministration with conventional cytotoxic agents and radiation therapy are currently ongoing.

If the present findings in the pig hold true in humans, the short duration of treatment necessary to inhibit constrictive remodeling will limit potential side effects and may make BB-2516, or comparable compounds, suitable for prevention of restenosis in humans.

	ACKNOWLEDGMENTS

 
BB-2516 (Marimastat) was supplied by British Biotech Pharmaceuticals, Oxford, England.

	REFERENCES

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Post MJ, Borst C, Kuntz RE. The relative importance of arterial remodeling compared to intimal hyperplasia in lumen renarrowing following balloon angioplasty: a study in the normal rabbit and the hypercholesterolemic Yucatan micropig. Circulation 1994; 89:2816-2821.

2. De Smet BJ, van der Zande J, van der Helm YJ, Kuntz RE, Borst C, Post MJ. The atherosclerotic Yucatan animal model to study the arterial response after balloon angioplasty: the natural history of remodeling. Cardiovascular Res 1998; 39:224-232.

3. 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.

4. Sierevogel MJ, Pasterkamp G, Velema E, et al. Oral matrix metalloproteinase inhibition and arterial remodeling following balloon dilation: an intravascular ultrasound study in the pig. Circulation 2001; 103:302-307.

5. Steward WP, Thomas AL. Marimastat: the clinical development of a matrix metalloproteinase inhibitor. Expert Opin Investig Drugs 2000; 9:2913-2922.

6. De Smet BJGL, De Kleijn DPV, Hanemaaijer H, et al. Metalloproteinase inhibition reduces constrictive arterial remodeling following balloon angioplasty: a study in the atherosclerotic Yucatan micropig. Circulation 2000; 101:2962-2967.

7. Geary R, Nikkari S, Wagner W, Williams J, Adams M, Dean R. Wound healing: a paradigm for lumen narrowing after arterial reconstruction. J Vasc Surg 1998; 27:96-108.

8. Scott K, Wood E, Karran E. A matrix metalloproteinase inhibitor which prevents fibroblast-mediated collagen lattice contraction. FEBS Lett 1998; 441:137-140.

9. Bendeck MP, Irvin C, Reidy MA. Inhibition of matrix metalloproteinase activity inhibits smooth muscle cell migration but not neointimal thickening after arterial injury. Circ Res 1996; 78:38-43.

10. Treharne GD, Boyle JR, Goodall S, Loftus IM, Bell PR, Thompson MM. Marimastat inhibits elastin degradation and matrix metalloproteinase 2 activity in a model of aneurysm disease. Br J Surg 1999; 86:1053-1058.

11. Lombard MA, Wallace TL, Kubicek MF, et al. Synthetic matrix metalloproteinase inhibitors and tissue inhibitor of metalloproteinase (TIMP)-2, but not TIMP-1, inhibit shedding of tumor necrosis factor-alpha receptors in a human colon adenocarcinoma (Colo 205) cell line. Cancer Res 1998; 58:4001-4007.

12. Deryugina EI, Bourdon MA, Jungwirth K, Smith JW, Strongin AY. Functional activation of integrin alpha V beta 3 in tumor cells expressing membrane-type 1 matrix metalloproteinase. Int J Cancer 2000; 86:15-23.

13. Fernandez-Patron C, Radomski MW, Davidge ST. Vascular matrix metalloproteinase-2 cleaves big endothelin-1 yielding a novel vasoconstrictor. Circ Res 1999; 85:906-911.

14. Kalela A, Ponnio M, Koivu TA, et al. Association of serum sialic acid and MMP-9 with lipids and inflammatory markers. Eur J Clin Invest 2000; 30:99-104.

15. Post MJ, de Smet B, van der Helm Y, Borst C, Kuntz RC. Arterial remodeling after balloon angioplasty or stenting in an atherosclerotic experimental model. Circulation 1997; 96:996-1003.

16. Elezi S, Kastrati A, Neumann F, Hadamitzky M, Dirschinger J, Schömig A. Vessel size and long-term outcome after coronary stent placement. Circulation 1998; 98:1875-1880.

17. Fukumoto Y, Shimokawa H, Kozai T, et al. Tyrosine kinase inhibitor suppresses the (re)stenotic changes of the coronary artery after balloon injury in pigs. Cardiovasc Res 1996; 32:1131-1140.

18. Côté G, Tardif JC, Lespérance J, et al. Effects of probucol on vascular remodeling after coronary angioplasty: Multivitamins and Protocol Study Group. Circulation 1999; 99:30-35.

19. Daida H, Kuwabara Y, Yokoi H, et al. Effect of probucol on repeat revascularization rate after percutaneous transluminal coronary angioplasty (from the Probucol Angioplasty Restenosis Trial [PART]). Am J Cardiol 2000; 86:550-552.

20. Le Tourneau T, van Belle E, Corseaux D, et al. Role of nitric oxide in restenosis after experimental balloon angioplasty in the hypercholesterolemic rabbit: effects on neointimal hyperplasia and vascular remodeling. J Am Coll Cardiol 1999; 33:876-882.

21. Bosmans JM, Vrints CJ, Kockx MM, Bult H, Cromheeke KMC, Herman HG. Continuous perivascular L-arginine delivery increases total vessel area and reduces neointimal thickening after experimental balloon dilatation. Arterioscl Thromb Vasc Biol 1999; 19:767-776.

22. Varenne O, Pislaru S, Gillijns H, et al. Local adenovirus-mediated transfer of human endothelial nitric oxide synthase reduces luminal narrowing after coronary angioplasty in pigs. Circulation 1998; 98:919-926.

23. Sabaté M, Serruys PW, van der Giessen WJ, et al. Geometric vascular remodeling after balloon angioplasty and beta-radiation therapy: a three-dimensional intravascular ultrasound study. Circulation 1999; 100:1182-1188.

24. Kimura T, Kaburagi S, Tamura T, et al. Remodeling of human coronary arteries undergoing coronary angioplasty or atherectomy. Circulation 1997; 96:475-483.

25. Bassiouny HS, Song RH, Hong XF, Singh A, Kocharyan H, Glagov S. Flow regulation of 72-kD collagenase IV (MMP-2) after experimental arterial injury. Circulation 1998; 98:157-163.

查询更多Minimal相关信息在本站>>

《放射学杂志》2003年2月第226卷第2期