A New Stent Graft

Summary

The use of stents improves the result after balloon coronary angioplasty. Restenosis due to neointimal hyperplasia and proliferation of smooth muscle cells are, however; a concern. In the present report, we studied the prevention of restenosis to allow endothelial cell migration and growth to proceed through micropores using our developed stent graft with micropored segmented polyurethane (SPU) thin film in a normal beagle model.

Our developed stent graft was made from Palmaz stent and micropored SPU thin film. The SPU film was arranged into four different micropore densities around the circumference: no micropores, arrangement 4; micropores of 30µm in diameter with an orderly distance of 250µm; (arrangement 1), 500µm; (arrangement 2) and 125µm (arrangement 3) between the neighboring two pores. Micropores were made using the Excimer laser ablation technique. The Palmaz stent was wrapped with micropored film, sutured, and glued with DMF (dimethyl formamide) under aid of a microscope. These stents were placed in the common carotid arteries of beagles (n = 5). They were sacrificed at 1 month, and a histological study and scanning electron microscopy study were performed for evaluation of endoluminal endothelialization.

In 10 arteries applied with stent grafts, there was no severe stenosis although it did occur to some extent. All stented arteries were patent. Endothelial cell migration and growth through micropores were observed histologically on micropored SPU thin film in this model, which did not affect the intraluminal diameter. In most non-porous regions, significant thrombi were found between the SPU film and the neointimal layer. On the other hand, in the porous region, little thrombosis was observed except in the lowest density region. In 125µm of distance between two neighboring pores, the neointimal layer was the thinnest, which was suitable for wide intraluminal space after placement of a stent graft.

Endothelial cell migration and growth through micropores were confirmed in the animal model using our developed micropored stent graft. The proceeding of their migration was controlled by micropore density under a constant micropore diameter. The stent graft with micropored SPU thin film is promising for the prevention of restenosis due to neointimal hyperplasia.

Introduction

In the field of balloon percutaneous transluminal coronary angioplasty (PTCA), acute or subacute occlusion of the dilated artery occurs in 3% to 8% of the cases¹. Another problem is the late restenosis of 30% to 50% of the cases within 3 to 6 months clinically^2,3. The use of coronary stents reduces late restenosis by 25% to 31% ^4,5. In the field of balloon percutaneous transluminal angioplasty (PTA) for peripheral arterial occlusive disease, patients with short lesions in larger vessels (> 6 mm diameter) do well, with restenosis rates of less than 30% at 1 year⁶.

We already presented endothelial cell (EC) migration and growth through micropores on microprocessed SPU thin films using the Excimer laser ablation technique as an in vitro model of transmural endothelialization in open cell-structured small-diameter vascular grafts⁷.

This time, a stent graft with micropored segmented polyurethane (SPU) thin film of 30µm in thickness was devised and introduced for prevention of late restenosis.

Material and Methods

Fabrication of the Stent graft with micropored SPU thin film

Our developed stent grafts were made from Palmaz stents (Palmaz stent: 2.1 mm in initial diameter, 2 cm in length, Cordis, Johnson & Johnson, Japan) and microprocessed SPU thin film. We fabricated a segmented polyurethane (PU)-based artificial graft with well controlled micropores in terms of their diameter and distribution, which was achieved using the computer-assisted Excimer laser (KrF) ablation technique⁸. The SPU thin film was arranged into four different micropore densities around the circumference: no micropores (arrangement 4); micropores 30 µm in diameter with an orderly distance of 250 µm (arrangement 1): 500 µm (arrangement 2) and 125 µm (arrangement 3) between two neighboring pores. They were made using the Excimer laser ablation system. The Palmaz stent was wrapped with the treated film as shown, sutured with a 10-nylon thread and glued with DMF (dimethyl formamide) under aid of a microscope (figure 1). They were sterilized with alcohol (0.5% (W/V) Maskin-ethanol) and irrigated with saline before use.

Figure 1.

Stent graft fabrication, A schema of fabrication is shown. A stent is wrapped with a treated SPU thin film, and the film is fixed on it with 10.0 nylon thread under aid of a microscope.

Stent graft placement protocol

For definition of an appropriate micropore size

Five beagles (10-17 Kg) were used in this study. Prior to stent graft placement, the beagles were daily fed with diets including aspirin (175 mg), dypiridamole (25 mg) for 1 month⁹. The beagles were anesthetized with ketamine (50 mg/Kg), xylazine (50 mg) and atropine sulfate (0.5 mg), and intubated endotracheally. Anesthesia was maintained with Ravonal (100 mg/H) with spontaneous respiration. Using a standard surgical technique, a 10cm in length 6F-sheath introducer was inserted into the femoral artery. A 5F catheter was inserted into each common carotid artery (CCA) and then control angiography was performed. These stent grafts were slipped retrogradely over the collapsed standard angioplasty balloon (Opta 5, 4 mm in diameter, 2 cm in length, Cordis, J&J, Japan), mounted and crimped down manually between the platinum markers. The balloon and stent graft assembly were passed through the sheath introducer and navigated into each common carotid artery. The stent graft was placed in the proximal CCA near the origin of the neck. Three dilations at 6 atmospheres for a few seconds were performed in the middle of the stent and at both ends of the stent. After delivery, patency of the artery was confirmed repeatedly on angiogram. Then, the dog was allowed to recover. They were fed with the above diets and drugs, and sacrificed at 1 month after stenting, and a histological study was performed in order to evaluate endoluminal endothelialization according to their micropore size.

Microscopic examination

After angiography at the 1 month follow-up, the cervical was opened and a lethal dose of sodium pentobarbital was injected intravenously, followed by the immediate clamping of the bilateral carotid arteries with clips. The clamped stented segment of the artery was removed and perfused with 5% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.2 and placed in the same buffer for more than 48 hours. Resin embedding was performed on them using glycol methacrylate (Historesin, AIKA, Japan).

They were cut into thick slices with a saw, and were subsequently cut into thin slices with a microtome. Light microscopic examination was performed on most tissues. Hematoxylin and Eosin stain and Masson-Trichrome stain were used as the routine stain. Scanning electron microscopy was performed on some tissues ¹⁰.

Results

Angiographic studies

In 5 beagles (10 arteries), the stent graft could be inserted easily into the sheath introducer via the femoral artery and placed accordingly. Bilateral carotid arteries expanded with the micropored stent graft were sampled out at 1 month after stenting. The measured balloon-to-artery ratio was 1.0, demonstrating a precise matching. All stented arteries were patent by angiography. Moderate stenosis of 20% to 50% in diameter occurred in 3 vessels, stenosis of less than 20% in 5 vessels, and no stenosis in 2 vessels.

Macroscopic findings

There was no major disruption of the vessel wall, incomplete expansion or marked overdilation of the stent. Macroscopic findings revealed no white thrombosis on the surface of the transplanted stent graft. At the border between the stent graft and normal endothelium, some additional white thrombus were found.

Microscopic findings

The cross-sectional specimen of the stent grafts showed neointimal growth, which induced confluent endothelial cells on the luminal surface of the SPU film. Beneath the confluent endothelial cell layer, a neointimal wall of tissue existed, which was filled with tissues migrated from the native vascular tissue through micropores. On each micro-patterned region of the micropored SPU film, the walls of the neointimal layer were almost the same regardless of the cutting location. The transverse sections of the grafts showed the following findings. In the most non-porous region, significant thrombi were found between the SPU film and the neointimal layer. On the other hand, in the porous region, little thrombus formation was observed except for the lowest density region. The neointimal layer was the highest in the non-porous region. The thickness decreased with the increase of the pore density in the porous region. In 125 µm of distance between two neighboring pores, the neointimal layer was the thinnest, which was suitable for wide intraluminal space in a stent graft. There was no endothelial or smooth muscle cell hypertrophy. The micropored thin film was intact, and there was a strong fence against the growth of it. Endothelial cell migration and growth through micropores of SPU thin films on the surface of the Palmaz stent graft were observed (figure 2).

Figure 2.

Endothelial cell migration and growth, 1 month after placement of a stent graft. This microscopic photo shows endothelial cell migration and growth through micropores of a SPU thin film.

SEM findings

In carotid arteries of the beagles, 1 month after the placement of non-treated micropored stent graft, there was confluent and constant growth of endothelium on the intraluminal surface of the stent graft, which passed through micropores from outside vascular tissues and migrated from the edge between the stent and normal vessel (figure 3).

Figure 3.

Endothelial cell migration and growth, lmonth after placement of a stent graft. This SEM photo shows endothelial cellular flow through placed micropores on the intraluminal surface of the stent graft.

Discussion

In this paper, we introduced our developed new-concept stent graft with micropored SPU thin film, and showed its promising and powerful role for long patency after stenting in a normal animal model.

There are two modalities to treat atherosclerotic stenosis. One is surgery and the other is intravascular surgery, that is, balloon PTA and stent placement. Surgery is an established method for atherosclerotic steno-occlusive lesions ^11,12,13. However, intravascular surgery has been recently become a less invasive surgery and gained popularity¹⁴.

Although the internal elastic membrane was broken by angioplasty, neither the plaque nor the media were disrupted with intact adventitia ^15,16. Endovascular manipulation by its very nature induces damage to the vessel wall, and the repair process can lead to recurring symptoms. The high strength and expansion ratio by the Palmaz stent prevent elastic recoil and allow highly accurate placement through a low-profile delivery system¹⁷. Stent integration is a multifactorally triggered process with proliferating smooth muscle cells (SMCs) generating regenerative tissue. Thrombotic material can be predominantly observed at the site of stenting, followed by the invasion of SMCs, T lymphocytes and macrophages¹⁸. The sequence of reactions after stent graft placement begins with the formation of a thin fibrin and platelet layer over the stent struts. Mural fibrin thrombus formation and incorporation contribute significantly to neointimal formation after deep vascular injury of pig coronary arteries or rabbit carotid arteries ¹⁹. This thrombotic layer is progressively replaced by fibromuscular tissue. Transformation of the fibrin and platelet layer into stable fibroblastic tissue is completed after 4 weeks^9,20 and is accompanied by neovascularization²¹. This tissue has been reported to reach maximal thickness 8 weeks after stent placement. In this manner, the normal regeneration process proceeds. The injured intimal layer is repaired and kept intact.

Recently, less invasive endovascular technique has dramatically developed. The restenosis, however, remained to be resolved even with the stent placement for a long follow-up study in coronary stent placement. No animal model can make a complete clinical restenosis, which means a process of smooth muscle proliferation, matrix deposition, vascular remodeling, elastic recoil, and a complex underlying atherosclerotic lesion²². The incidence of delayed reendothelialization and occurrence of deep dissections may be associated with excessive SMCs hyperplasia ¹⁸. Therefore, we must induce reendothelialization on the surface treated by a stent graft to avoid restenosis.

Long-term patency by stents depends upon the status of the inflow, runoff, length of disease, primary versus secondary stenting, and whether the lesion was initially a stenosis or occlusion. Restenosis rates for the coronary artery lesion have been 30% to 35% and for iliac artery lesions 10%. Within a stent, 0.45 to 1.45 mm of circumferential intimal hyperplasia develops that currently appears unavoidable ^23,24. Stents expanded to a diameter less than 6mm are prone to high rates of restenosis. If the diameter of a stent is too large for the recipient vessel (>l.3 times larger), it may induce excessive intimal proliferation around those stent struts in intimate contact with the vessel wall²⁵.

In patients with an occlusive disease, the purpose of the stent graft is to act as a microporous lining for the treated vascular segment. The stent portion of the device provides mechanical enlargement of the narrowed arterial lumen and prevents elastic recoil and constrictive remodeling²⁶, and the graft covering acts as a barrier to smooth muscle ingrowth intimal thickness. A true endoluminal bypass can be created²⁷. The Dacron cover is responsible for the acute inflammatory reaction with granulocyte invasion, which causes cell proliferation and finally occlusion of the stent and vessel lumen ²⁸. Inflammatory reaction with granulocyte infiltration contributes to the increased thrombogenic activity of the Dacron-covered stent graft and a high rate of occlusion²⁹. Partially covered stents by nonporous polytetrafluoroethylene inhibit stent-related restenosis in human iliac arteries³⁰. Stents grafts by an autologous porcine arterial graft result in accelerated endothelialization, less vascular injury, thinning of the arterial media and a trend to reduce the intimal hyperplasia^31,32.

Endothelialization is essential for the expression and maintenance of nonthrombogenecity when an artificial graft is implanted to replace a diseased vessel. EC ingrowth from surrounding native tissues, via unidirectional migration or transmural cell migration, to vascular prosthesis is one of the most important healing mechanisms of implanted artificial grafts. In the field of peripheral occlusive arteries with a stent graft, short-term thrombogenecity of the implanted graft and the development of restenosis at the ends of the graft were reported. They stressed that restenosis appeared to have been limited primarily to the native vessels at the ends of the stent-graft itself. The stent graft does appear to effectively limit the ingrowth of intimal hyperplasia on the lumen of the treated segment^27,33. For early endothelialization inside implanted devices, they were actively induced from native luminal tissues through placed micropores in the covering films of stent grafts. The highest endothelialization rates in an early incubation period were found on films with micropores between 18 and 50 µm in diameter, which were arranged by the Excimer laser ablation technique ⁷. In this study, our developed stent graft with micropored SPU thin film was placed into carotid arteries of beagles, and endothelialization was evaluated at 1 month after implantation. Proper and thin endothelialization of the endoluminal surface was performed via micropores 30 µm in size, and 125 µm in distance between the two pores in vivo.

Stent thrombogenecity is an inherent feature of stent composition, and that vigorous anticoagulation is needed to attenuate the propensity for stent thrombosis³⁴. Antiplatelets like aspirin and ticlopidine were transorally used in our canine model. The dogs were fed with food mixed with antiplatelets before and after procedures. Aspirin exerts its formation of thromboxane A2, a powerful mediator of platelet degranulation, through the permanent inactivation of the cyclooxygenase enzyme. In contrast, the primary action of ticlopidine is to irreversibly block the binding of fibrinogen to platelets, an effect that appears to be 85% effective against inhibiting platelet aggregation³⁴.

In our stent graft, the outside covering of the stent with thin walled controlled micropored polymer could offer better endoluminal flow dynamics without compromising its role in intimal hyperplasia.

The occurrence of restenosis by both vessel remodeling and neointimal hyperplasia might be reduced in the future by the combined mechanics and characteristics of our developed microporous graft³⁵. Our devices remain to be used in the restenosis animal model in the near future and to confirm their promising powerful anti-restenotic effects.

Conclusions

Our developed new-concept stent graft with micropored SPU thin film showed its promising and powerful role for long patency after stenting in a normal animal model.