Preliminary report of a new type of braided vein stent in animals

Structural features of the stent and delivery system

The novel stent specifically for use in the iliofemoral venous system was designed by Suzhou Yinluo, Inc. The novel stent is a woven-mesh self-expanding stent composed of 12 nickel nitinol alloy wires through positive and negative cross-winding. The rotations are switched at the convoluted round portions at both ends of the stent such that the stent is in looped-ends segments. The diameter of the stent is designed to be 10–16 mm. Larger (14 mm, 16 mm) stents are woven from 0.01-inch Ni-Ti alloy wire, and smaller (10 mm, 12 mm) stents are woven from 0.009-inch wire to provide the same radial support force. Each end of the stent is designed to have a 6-mm long “wide mouth,” which is 2 mm larger than the diameter of the stent body. The purpose of the wide mouth is to increase the anchorage at both ends of the stent and reduce displacement (Figure 1).

OPEN IN VIEWER

Figure 1.

The stent composes of 12 nickel nitinol alloy wires through positive and negative cross-winding. Both ends of the stent are looped and “wide mouth” shape.

Delivery system

The delivery system consists of a three-layer catheter structure. The driving gear drives a double-screw structure to move the middle tube and the outer tube in opposite directions during stent release, thereby compensating for the axial shortening of the stent. In other words, during the release process, the stent was pushed forward through the forward movement of the middle tube to compensate the retraction of the stent. By adjusting the number of screw gears, different speed ratios can be realized to adapt to different stent shrinkage ratios, avoiding axial displacement in the release process and achieving accurate positioning (Figure 2).

OPEN IN VIEWER

Figure 2.

The gear 2 drives 1 and 3 so that the inner and outer sheath can move coaxially in opposite directions. The stent is pushed forward during the release process to compensate for the “retraction” of the stent. By adjusting the number of screw gears, different speed ratios can be realized to adapt to different stent shrinkage ratios.

At the same time, we sent the stent to the State Food and Drug Administration Medical Device Quality Supervision and Inspection Center for physical property testing.

Animal models and experimental design

This experiment was reviewed and approved by the Animal Experimental Ethics Committee of Peking University People’s Hospital. The agency complies with the National Institutes of Health (NIH) guidelines for experimental animal use and feeding released in 2011.

Healthy 4- to 6-month-old male sheep weighing 50–60 kg were selected as experimental animals. According to the weight of the animals, they were randomly divided into two groups: the novel stent group (NS) and the Wallstent stent group (WS). Twelve animals were in each group. The same type of stent was placed on each side of the iliac vein in each animal and observed at 30 days and 90 days after placement. Six animals in each group were randomly selected at each observation point for angiography, sacrifice and sampling.

Stent placement

Experimental animals were fasted for 12 hours before surgery, and 1.6 million units of penicillin were intramuscularly injected 30 min before the operation. After the animals were pretreated with an intramuscular injection of atropine sulfate (0.04 mg/kg), general anesthesia was induced by the intravenous administration of propofol (6 mg/kg). After endotracheal intubation, anesthesia was maintained with isoflurane (0.8–1.25%) inhalation. Cardiac and respiratory parameters were monitored throughout the operation. The animal was placed in a supine position on the operating table. Each animal received intravenous heparin (100 U/kg). Both femoral veins were punctured under guidance by ultrasound, a 5-F sheath was placed (10-F sheaths could cause vein occlusion and inaccurate venous diameter measurements), and a catheter was inserted for venography. Angiography was performed in the positive, left anterior oblique 45° and right anterior oblique 45° positions. The diameters of the bilateral iliac veins and the bifurcation of the common iliac vein were determined by angiography. The 5-F sheath was replaced with a 10-F sheath under guidance of the hard guidewire, and a stent was placed according to the measurement. The proximal end of one stent exceeded 0.5–1 cm of extension into the IVC, and the contralateral stent did not exceed the common iliac vein. After bilateral stent deployment, bilateral iliac venography was performed for measurements. All sheep were given 100 mg of aspirin qd and 75 mg of clopidogrel qd orally after surgery until the end of follow-up.

Follow-up

The follow-up examinations at the observation points were performed by an independent researcher who was blinded to the animal groups.

All animals were subjected to venography in the same position as in placement to evaluate the stent for patency, displacement, and fracture. The animals were sacrificed, and the stents were collected. At the ends of each specimen, more than 5 mm of vein past the stent coverage was retained. A Fuji EB-270S electronic bronchoscope (Fuji Co., Ltd., Japan) was used to observe and assess the stent for thrombosis and neointimal formation. Samples were fixed with formaldehyde for further pathological examination.

Pathomorphological examination

The specimens were fixed with 10% formaldehyde solution, dehydrated by alcohol, and solidified by liquid polymer. Sections 150 µm thick were prepared using a precision cutting machine (Buehler IsoMet 5000, USA) and thinned to 10–20 µm with a polishing machine (Buehler EcoMet 250, USA). All histological sections were stained with hematoxylin and eosin, and histological slides were prepared for pathomorphological examination. Inflammation, injury, and vessel/wall dimensions were evaluated according to methods described in the literature.⁹ Pathological sections containing more than four stent points were selected for statistical analysis. The average values are calculated from three measurements.

Inflammation. 0: 0–3 inflammatory cells (lymphocytes, eosinophils, macrophages, etc.) around strut. 1: 4–10 inflammatory cells around strut. 2: >10 inflammatory cells around strut area, can extend into neointima, but does not efface surrounding tissue 3: >10 cells, effaces surrounding tissue.

Injury. 0: Inner Elastic Lamina (IEL) intact. 1: IEL lacerated. 2: Media lacerated. 3: External elastic lamina lacerated.

The mean integral of injury score is the sum of the damage integral of each stent and the number of stent wires.

The average injury integral = the sum of stent wires injury scores/the number of stent wires

Vessel/wall dimensions

Neointima Area (NA)＝Internal Elastic Lamina Area (IELA) − Lumen area (LA)

Stenosis Rate＝NA／IELA × 100％

Neointima thickness. The average thickness of the intima from the stent to the lumen

In-stent restenosis. Stenosis of luminal area >75% (stenosis of lumen diameter >50%)

Statistics

Imaging, histology, and integration data were analyzed by t-test for comparisons between the groups. SPSS 23.0 statistical software was used for the analysis. P < 0.05 was considered statistically significant.

Physical performance

The maximal radial resistive force (N/cm) of the NS vs. the WS¹⁰ was 10.8–12.9 vs. 3.89. The value for the novel stent was approximately four times that for the Wallstent.

The chronic outward force at the 90% diameter (N/cm) was 0.9 vs. 0.45 for the NS vs. the WS.¹⁰

Stent placement

The “release compensation” function of the new delivery system stabilized the tip of the stent during release. The novel stent does not exhibit obvious shortening during the release process. However, it is still necessary for the operator to make fine adjustments based on the position of the vein. Stents were successfully placed in all animals. According to quantitative vessel angiography (QVA), all proximal ends of the stents remained at a stent:vessel diameter oversize ratio of 1–1.3:1. As there was much difference concerning the proximal and distal diameters of the iliac veins in the sheep, after stenting, the stent:vessel diameter ratio in the distal part of some stents exceeded 1.3:1. There were no significant differences in the diameter of the iliac vein or the implanted stents between the two groups (Table 1).

OPEN IN VIEWER

Table 1

The diameter of iliac vein in quantitative vessel angiography.

QVA	30 days			90 days
	NS	WS	P	NS	WS	P
Before Implantation (mm)	8.89±1.21	8.81±0.94	0.901	9.24±1.16	8.75±0.97	0.445
After Implantation (mm)	10.16±0.91	10.10±1.02	0.912	10.55±1.06	10.08±1.16	0.480
Follow-up (mm)	9.91±0.86	9.78±0.99	0.813	8.32±0.80	7.92±0.97	0.450

QVA: quantitative vessel angiography; NS: the novel stent; WS: the Wallstent stent.

Angiography

Follow-up at 30 days and 90 days after the operation showed good image development for all iliac veins and stents in both groups. All stents were normal in shape, without fracture, distortion, or displacement (Figures 3 and 4). At 30 days and 90 days, angiography showed slight lumen stenosis in both groups. The diameter of the vessels was measured by QVA. There were no significant differences in the vein diameter between the two groups after 30 or 90 days (Table 1).

OPEN IN VIEWER

Figure 3.

The DSA images after implantation and 30-day follow-up. (a and b) Instant image after implantation of group WS; (c and d) 30-day follow-up image of group WS; (e and f) instant image after implantation of group NS; (g and h) 30-day follow-up image of group NS. The red arrow shows a slight stenosis in the stent.

OPEN IN VIEWER

Figure 4.

The DSA images after implantation and 90-day follow-up. (a and b) Instant image after implantation of group WS; (c and d) 90-day follow-up image of group WS; (e and f) instant image after implantation of group NS; (g and h) 90-day follow-up image of group NS. The red arrow shows a slight stenosis in the stent.

Endothelialization and thrombosis

After the stents were collected, the specimens were observed by endoscopy. After 30 days, the stents were essentially covered by a layer of neointima with an even thickness. Through the neointima, the structure of the scaffold was visible, and no thrombosis or stenosis was observed in the stent. There was no intima coverage in part of the stent (internal iliac vein opening or venular branch site). After 90 days, stents in the two groups were completely endothelialized without thrombosis. The endothelium was still not covered at the site of the internal iliac vein opening (Figure 5).

OPEN IN VIEWER

Figure 5.

Endoluminal images of endoscopy: at 30 days, the stents were essentially covered by a layer of neointima. Through the neointima, the structure of the scaffold was visible, and no thrombosis was observed. At 90 days, stents were completely endothelialized without thrombosis. The red arrows showed that the endothelium was not covered at the site of the internal iliac vein opening.

Histomorphology

At the 30-day and 90-day follow-up, a continuous intima layer was observed on the stent surface in both groups. The stent stem was completely covered by neointima with mild hyperplasia. No significant differences were found in IELA, LA, NA and Stenosis Rate. Numerically, the stenosis rate was higher in the WS than in the NS, but there was no significant difference. There were no stents with a stenosis rate ≥75% in either group (Table 2).

OPEN IN VIEWER

Table 2

The area of different parts of the vascular cross section.

Histopatho-logical Measurement	30 days			90 days
	NS	WS	P	NS	WS	P
IELA (mm2)	97.54±16.64	91.21±13.08	0.479	101.01±16.17	86.22±12.24	0.105
LA (mm2)	73.75±8.69	70.64±11.52	0.609	67.97±14.51	58.26±14.63	0.275
NA (mm2)	23.82±11.92	20.46±9.05	0.594	33.04±16.69	27.96±6.30	0.510
Stenosis rate (%)	23.55±9.80	22.60±7.99	0.857	31.97±11.90	33.06±8.34	0.858

IELA: internal elastic lamina area; LA: lumen area; NA: neointima area.

Histopathology

At the 30-day follow-up, no intimal thickening at the proximal or distal end of the stents, vascular injury, degeneration or necrosis was observed. Mild intimal hyperplasia was observed in the stent segment. Smooth muscle cells and collagen fibers were found in the proliferative intima, accompanied by many new small blood vessels. A small number of stent wires were not completely endothelialized. Several stent stems exhibited foreign body reactions, with a small amount scattered in inflammatory cell infiltration. The vascular wall was slightly compressed without medial membrane injury. No fibrin deposition was found around the stent, nor was intraluminal thrombosis observed. There were no significant differences in the inflammatory response or vascular injury (Figure 6).

OPEN IN VIEWER

Figure 6.

Thirty-day histopathological image of the stent. (a) The vein at the proximal end of the stent; (b) proximal end of the stent; (c) the middle of the stent; (d) distal end of the stent; (e) the vein at the distal end of the stent; and (f–j) enlarged images of the corresponding black frame parts of the above figure.

After 90 days, the stents in both groups were completely endothelialized. There was moderate neointimal hyperplasia in the stent segment. Many smooth muscle cells and a large amount of collagen fibers were observed in the proliferative intima, and many small vessels were formed (Figure 7). There were no significant differences in the inflammatory response or vascular injury.

OPEN IN VIEWER

Figure 7.

Ninety-day histopathological image of the stent. (a) The vein at the proximal end of the stent; (b) proximal end of the stent; (c) the middle of the stent; (d) distal end of the stent; (e) the vein at the distal end of the stent; and (f–j) enlarged images of the corresponding black frame parts of the above figure.

Iliac venous stents

With the development of interventional technology and profound understanding of chronic iliac vein obstruction, endovenous stenting is increasingly used to treat IVCS. Iliac venous stent placement has shown good clinical results. Raju et al.¹¹ reported that the pain relief rate was 86–94%, the swelling remission rate was 66–89%, and the venous ulcer healing rate was 58–89%. CIRSE and SVS guidelines recommend stenting for severe venous obstructive disease.^7–8 The reality is that the development of iliac venous stents is behind in both clinical awareness and practice. For a long time, there were no specially designed stents for the peripheral venous system, so various types of self-expanding arterial stents were applied to the iliac vein, including the Wallstent (Boston Scientific, USA), Protégé (ev3, USA), and SMART (Cordis, USA). These stents were originally designed for peripheral arteries or other purposes, and thus do not fully meet the needs of IVCS treatment.^12–14

Characteristics of venous stents

Veins differ from arteries in anatomy, function and hemodynamics. Venous stents require a larger caliber than arterial stents. Post-thrombotic venous lesions usually have a tough fibrous bundle, and the thin wall of veins is prone to narrowing or occlusion after compression by a peripheral artery or tumor, which requires the stent to provide a high radial support force to maintain the dilated lumen. Histological examination of IVCS-thickened vessel walls demonstrates a predominance of fibroblasts with a thick intermediary layer of collagen. As a result, flexible stents with high radial forces are needed to treat chronic obstruction.^15,16 The vein across the pelvic cavity has natural physiological bending, which requires good stent compliance. Therefore, it is necessary to design a special stent for venous characteristics. A venous stent must be characterized by larger diameters, longer lengths, higher radial forces, and increased flexibility.^17,18

The status of venous stents and woven stents

In recent years, attention has been directed toward research on venous stents in professional fields. Several types of venous stents for iliofemoral vein stenosis are available in the US and Europe, such as the Vici (Boston Scientific, USA), sinus-Venous (Optimed, Germany), Zilver Vena (Cook Medical, USA), Venovo (Bard Medical, USA) and Abre (Medtronic, USA). These stents are laser-cut scaffolds that have greater diameters and higher radial forces than stents designed for arteries. Laser-cutting stents from tubes does not result in a high level of both strength and flexibility; improvements in radial force will inevitably result in decreased flexibility. Although some stents were designed to overcome this contradiction through the open-loop structure, this structure introduces instability in the scaffold, causing it to easily break under force.¹⁹

Most researchers have concentrated their efforts on studying laser-cut stents, with few focusing on braided stents. A braided stent can be designed with high radial strength to provide maximal blood flow to affected areas from the time of stent deployment while maintaining good flexibility and high kink and fracture resistance. Braided stents have a certain inherent ductility, which can allow changes in the diameter and length with pressure changes inside and outside the cavity; additionally, the lumen is smooth. These features are more suitable for the venous system. In terms of structure, braided stents are more in line with the concept of “vascular mimetic stents.” The Wallstent is a well-known woven stent, and although it was not designed for venous lesions, it is most widely used in IVCS. As such, we designed a new braided venous stent to meet these needs.

Materials

The NS wire is a Ni-Ti alloy, and the WS is composed of Co-Cr wire. Ni-Ti alloy allows faster formation of an oxide layer in vivo, so it is more resistant to corrosion and is more biocompatible than Co-Cr alloy.^20,21 Compared with Co-Cr alloys, Ni-Ti alloys have better elasticity and radial force.²² Therefore, nitinol is more suitable for stent materials.

Structure

Wide mouth: After an iliac venous stent is placed, due to pulsation of the iliac artery in front of the iliac vein, the stent can be easily displaced and extrude into the distal end of the vein, leading to restenosis or occlusion. Each end of the novel stent is designed to have a wide mouth to avoid stent displacement at the distal end of the vein. Meanwhile, the distal wide mouth strengthens the landing between the stent and the vein wall to prevent displacement. Although the wide mouth slightly increased the local oversize ratio of the stent, our results indicate that this did not increase the risk of thrombosis or stenosis in the stent.

Special delivery system

A braided stent allows for strength and flexibility with high radial force. However, foreshortening and/or precise placement can be challenging and requires experience.²⁵ Our delivery system can compensate for axial retraction during the release process for precise positioning. We also found that the delivery system could not completely counteract stent retraction because the different diameters of the implanted vessels cause different degrees of stent retraction after release. Thus, fine-tuning by the operator is still required during release.

Additional findings

The endoscopic observations of the two groups after 30 days and 90 days showed that most scaffolds remained unobstructed in the branches of the veins, especially the opening of the internal iliac veins. Zhang et al.²⁶ have demonstrated that an iliac venous stent extending over the contralateral iliac vein does not impede the contralateral backflow of blood for at least 12 weeks in a canine model. One possible reason for this phenomenon is that the blood flow at the branch vessels affects the attachment and proliferation of inflammatory cells, endothelial cells and smooth muscle cells. Whether the opening of the branch vein can remain unobstructed over time needs further investigation.

Limitations

Because of the limitations of objective experimental conditions, intravascular ultrasound was not used in this study for accurate measurement of the iliac vein. During the placement process, we compensated for this deficiency by performing multiangle angiography of the target vessel. At 30 days and 90 days after stent placement, the diameter of the vessel was accurately measured by hard tissue pathology.