Interwoven versus Knitted Self-Expandable Metallic Stents: A Comparison Study of Stent-Induced Tissue Hyperplasia in the Rat Esophagus

Abstract

This study investigated whether interwoven self-expandable metallic stent (I-SEMS) placement suppresses stent-induced tissue hyperplasia compared with conventional knitted self-expandable metallic stent (K-SEMS) placement in a rat esophageal model. Twelve rats were randomly assigned to the I-SEMS (n = 6) and K-SEMS (n = 6) groups. All rats were sacrificed 4 weeks after the stent placement. The degree of stent-induced tissue hyperplasia on esophagography and histologic examination was compared between the groups. Stent placement was technically successful in all rats. Esophagography revealed significantly greater mean luminal diameter of the stented esophagus in the I-SEMS group than in the K-SEMS group (p < 0.001). Histologic examination revealed a significantly lower percentage of tissue hyperplasia area and a significantly thinner submucosal fibrosis in the I-SEMS group than in the K-SEMS group (all p < 0.001). There were no significant differences in the thickness of the epithelial layers (p = 0.290) and degree of inflammatory cell infiltration (p = 0.506). Formation of stent-induced tissue hyperplasia was evident with both I-SEMS and K-SEMS. Placement of I-SEMSs with a small cell size and high flexibility seems to be effective in suppressing stent-induced tissue hyperplasia compared with placement of K-SEMSs in rat esophageal models.

1. Introduction

Placement of hand-crafted and knitted self-expandable metallic stents (SEMSs; K-SEMSs) is the most common therapeutic strategy for malignant or benign esophageal strictures [1,2,3]. However, excessive proliferative response to mechanical injury is a consequential obstacle for successful treatment [3,4,5]. Fully or partially covered SEMSs have been developed for esophageal strictures to overcome this problem [6,7,8]. Although the covering membrane could prevent tissue hyperplasia through the wire meshes, granulation tissue formation could still occur at the uncovered ends of the stent [8,9]. Furthermore, covered SEMS placement significantly increases the stent migration rate, and tissue ingrowth through the wire mesh can occur through the disrupted covering membrane [8,10]. Stent-induced tissue hyperplasia can lead to recurrent symptoms and technical difficulties in stent removal [2,3].

An interwoven SEMS (I-SEMS) is a weave nitinol alloy stent designed to resist the unique stressor in vascular stenosis [11,12]. Its helical interwoven structure and small-sized cells help to minimize the stress on the entire strut when it is bent, making the stent highly flexible [12,13]. Additionally, this design makes the I-SEMS resistant to torsion caused by friction between the nitinol wires [14]. Compared with the design of conventional K-SEMSs, that of I-SEMSs increases the resistance for kinking owing to the structure with a smaller size of cells, thereby preventing stent-related complications, such as stent fracture and stent-in-restenosis. The K-SEMS is a closed-cell design, and it is a knitted design by joining the intersection of the stent wires. This closed-cell design gives high radial force and uniformity but limited flexibility [13]. On the other hand, open-cell stents have the feature of increasing flexibility and low radial force by sacrificing strut uniformity by eliminating the intersection of the stent wire. Although the I-SEMS structure has a lower radial force than the K-SEMS designed with the same number of wires, it combines the advantages of a closed-cell stent and an open-cell stent to provide high flexibility while maintaining a uniform cell size [15]. Armstrong et al. reported that I-SEMS placement improved stent patency and decreased the stent fracture rates compared with conventional K-SEMS placement in the femoropopliteal segment [16]. Despite their better performance in the vascular lumen, no study has identified the value of I-SEMSs with small-sized cells in suppressing stent-induced tissue hyperplasia in non-vascular luminal organs. We hypothesized that I-SEMS placement would reduce stent-induced tissue hyperplasia owing to the small size of the stent cell. Therefore, the purpose of this study was to investigate whether I-SEMS placement suppresses stent-induced tissue hyperplasia formation compared with conventional K-SEMS placement in a rat esophageal model.

2. Materials and Methods

2.1. Stent Preparation

The I-SEMSs used were designed and manufactured by S&G Biotech Co., Ltd. (Yongin, Korea). A total of 32 nitinol wires with a thickness of 0.09 mm were interwoven using a braiding machine. When fully expanded, the I-SEMSs used for the rat esophagus were 5 mm in diameter and 10 mm in length. One gold wire was wrapped around the stent in a spiral shape to facilitate precise placement under fluoroscopic guidance. Having the same size at full expansion, the K-SEMSs were knitted from a single thread of 0.127 mm-thick nitinol wire filament into a tubular configuration with six bent points. The stent delivery system consisted of a 6 Fr Teflon sheath (Cook, Bloomington, IN, USA) and a pusher catheter. To match the radial force of the two types of SEMSs, different wire thickness was used.

2.2. Stent Cell Size Measurement and Bending Test

The I-SEMSs and knitted SEMSs used for measurement of the stent cell size and the bending test were 5 mm in diameter and 50 mm in length with 10 samples each. Images of the two types of SEMSs were obtained and analyzed using the ImageJ 1.53c software (National Institutes of Health, Bethesda, MD, USA) to measure the cell size of the stents. The areas of the cell size were calculated, and the average value at 10 points from the proximal to the distal ends of the stents was compared between the two groups.

The modeling of the bending test phantom was designed and manufactured by a local manufacturer (ANYMEDI, Seoul, Korea). The phantom was made of a photocurable resin material (Xiamen Zhisen Electro, Xiamen, China) consisting of a body part and two fixation parts. It was 150 mm in length, 60 mm in width, and 50 mm in height. The fixation parts had a central cylindrical bar to fix the stent. The bending test was performed to evaluate and compare the degree of stent deformation, such as bending and kinking, of the I-SEMSs and K-SEMSs. The proximal and distal ends of the stent samples were placed in each cylindrical bar of the phantom 20 mm apart.

2.3. Animal Study Design

This study was approved by the Institutional Animal Care and Use Committee of our center and conformed to US National Institutes of Health guidelines for humane handling of laboratory animals. A total of 12 Sprague–Dawley rats weighing 330–362 g (median weight = 345.5 g) were used in this study. The rats were randomly assigned to two groups: an I-SEMS group (n = 6) and a K-SEMS group (n = 6). All rats were provided with food and water ad libitum and maintained at 24 ± 2 °C with a 12 h day–night cycle. They were sacrificed 4 weeks after stent placement via administration of inhalable pure carbon dioxide.

2.4. Stent Placement

Anesthesia was induced via intramuscular injection of 50 mg/kg zolazepam and tiletamine (Zoletil 50l, Virbac, Carros, France) and 10 mg/kg xylazine (Rompun, Bayer Healthcare, Leverkusen, Germany). Under fluoroscopic guidance, a 30 cm micro guidewire (Boston Scientific/Medi-Tech, Watertown, MA, USA) was inserted through a custom-made mouthpiece and advanced into the stomach (Figure 1a). The stent delivery system loaded with the I-SEMS or K-SEMS was advanced over the guidewire into the middle portion of the esophagus (Figure 1b). The stent was released in the cervical esophagus below the clavicle level by withdrawing the sheath while the pusher catheter stayed in place. The sheath and the pusher catheter were removed after the stent was completely released (Figure 1c). Esophagography was performed immediately after the procedure to confirm stent patency (Figure 1d).

2.5. Esophagography

Follow-up esophagography was performed using a contrast medium (Telebrix 30 meglumine, Guerbet, Aulnay-sous-Bois, France) immediately before all rats were sacrificed. The stented esophagus was divided into proximal and distal segments. The luminal diameter of each segment was measured on the esophagographic images using RadiAnt DICOM viewer (version 1.1.20, Medixant Company, Poznan, Poland).

2.6. Histologic Examination

The esophagus and stomach were extracted surgically. The esophageal tissue samples were fixed in 10% neutral buffered formalin for 24 h. The samples were dehydrated sequentially with alcohols of different concentrations and embedded in a resin block via infiltration with glycol methacrylate (Technovit 7200^® VLC; Heraus Kulzer GMBH, Wertheim, Germany). The embedded samples were axially sectioned at the stented esophagus. Thereafter, the blocks obtained were mounted on an acrylic slide. Microgrinding and polishing with silicon carbide papers using a grinding system (Apparatebau GMBH, Hamburg, Germany) were performed for all slides until a thickness of 30 µm was reached. The slides were stained with hematoxylin and eosin (H&E). Histologic evaluation via H&E staining was based on the percentage of the tissue hyperplasia area, degree of inflammatory cell infiltration, thickness of the submucosal fibrosis, and thickness of the epithelial layers. The tissue hyperplasia formation-related percentage of the esophageal cross-sectional area stenosis was calculated using the following equation:

$$\mathrm{The}\mathrm{percentage}\mathrm{of}\mathrm{tissue}\mathrm{hyperplasia}\mathrm{area}=100\times (\frac{\mathrm{Stenotic}\mathrm{stented}\mathrm{area} \left({\mathrm{mm}}^2\right)}{\mathrm{original}\mathrm{stented}\mathrm{area}\left({\mathrm{mm}}^2\right)})$$

The degree of inflammatory cell infiltration was subjectively determined according to the distribution and density of the inflammatory cells (graded as 1, mild; 2, mild to moderate; 3, moderate; 4, moderate to severe; and 5, severe) [16,17]. The average values of the submucosal fibrosis and epithelial layer thickness and the degree of inflammatory cell infiltration were calculated from eight points around the circumference. Histologic analysis of the esophagus was performed using a digital slide scanner (Pannoramic 250 FLASH III, 3D HISTECH Ltd., Budapest, Hungary). Measurements were obtained using a digital microscope viewer (CaseViewer, 3D HISTECH Ltd.).

2.7. Statistical Analysis

The Mann–Whitney U test was used to analyze the differences between the groups as appropriate. A p-value of <0.05 was considered statistically significant. A Bonferroni-corrected Mann–Whitney U test was further performed for p-values of <0.05 to detect group differences (p < 0.008 indicated statistical significance). Statistical analyses were performed using the SPSS software (version 27.0; SPSS, IBM, Chicago, IL, USA).

3. Results

3.1. Cell Size and Bending Test Findings

The mean ± standard deviation value of the cell size of the I-SEMS was 0.251 ± 0.02 mm², whereas that of the K-SEMS was 3.712 ± 0.37 mm² (Figure 2a,c). The cell size of the I-SEMS was 14.7 times smaller than that of the K-SEMS. In the bending test, the I-SEMS showed highly flexible bending without kinking and maintained an inner lumen of the stent (Figure 2b). However, the K-SEMS was not able to withstand the bending force and showed severe kinking in the middle portion, which means an inner lumen of the stent could not be maintained (Figure 2d).

3.2. Procedural Outcomes

Stent placement was technically successful in all rats, and all of them survived until the end of the study. A small amount of hematemesis occurred immediately after the procedure in three rats (one in the I-SEMS group and two in the K-SEMS group), but subsided spontaneously. Stent-related complications, such as stent migration and in-stent stenosis, were not observed during follow-up.

3.3. Esophagographic Findings

The esophagographic findings are shown in Figure 3. The mean overall luminal diameter of the stented esophagus was significantly higher in the I-SEMS group (3.81 ± 0.63 mm) than in the K-SEMS group (2.08 ± 0.86 mm, p < 0.001). This significant difference was also observed in the proximal portion (K-SEMS group: 2.40 ± 0.92 mm vs. I-SEMS group: 4.29 ± 0.51 mm, p < 0.001) and distal portion (K-SEMS group: 1.76 ± 0.77 mm vs. I-SEMS group: 3.35 ± 0.35 mm, p < 0.001). Moreover, severe narrowing was seen around the stent in the K-SEMS group, whereas a straight-line filling defect was observed in the stented esophagus without narrowing in the I-SEMS group (Figure 3a,b).

3.4. Histological Findings

The histologic findings are shown in Figure 4. The mean percentage of the tissue hyperplasia area was significantly higher in the K-SEMS group (48.07% ± 7.30%) than in the I-SEMS group (26.68% ± 8.24%, p < 0.001). The submucosal fibrosis was significantly thicker in the K-SEMS group (0.65 ± 0.13 mm) than in the I-SEMS group (0.25 ± 0.06 mm, p < 0.001). However, the mean thickness of the epithelial layer (0.18 ± 0.07 mm in the K-SEMS group vs. 0.22 ± 0.08 mm in the I-SEMS group, p = 0.290) and the degree of inflammatory cell infiltration (3.00 ± 0.76 in the K-SEMS group vs. 2.75 ± 0.71 in the I-SEMS group, p = 0.506) were not significantly different between the two groups.

4. Discussion

I-SEMSs were developed and designed to increase flexibility and fracture resistance [12,13,14]. They have a uniform and small cell size resulting from fabrication by braiding machines using multiple nitinol wires [13]. I-SEMSs have a significantly smaller cell size than conventional K-SEMSs. Stent-induced tissue hyperplasia commonly and definitely grows through the wire mesh within 4 weeks after stent placement. According to the results of previous studies, stent-induced tissue hyperplasia occurred after stent placement in the rat within 4 weeks, so this study also processed a follow-up for 4 weeks [17,18,19,20]. Our results demonstrated that I-SEMS placement successfully inhibited stent-induced tissue hyperplasia caused by mechanical injury after stent placement in a rat esophageal model. Tissue penetration through the wire meshes may have been relatively preventable due to the small cell size. The high flexibility also relatively reduced the mechanical damages to the esophageal wall. After stent placement, the luminal diameter of the stented esophagus was significantly larger in the I-SEMS group than in the K-SEMS group. Consistently, the histologic results also demonstrated that the tissue hyperplasia-related variables in the K-SEMS group were significantly higher than those in the I-SEMS group. These findings support that the smaller cell size and high flexibility of I-SEMSs are key factors for inhibiting stent-related tissue hyperplasia formation.

The physical properties of I-SEMSs yielded high flexibility to maintain a round shape when they are bent. The flexibility of these stents is associated with stent-related complications, such as stent collapse, fracture, abutment, and kinking [20,21,22,23,24]. In our study, the small cell size of the I-SEMSs increased the stent flexibility. In the bending test, the K-SEMSs showed severe kinking at the middle portion owing to their low flexibility. Conversely, the I-SEMSs successfully maintained an inner lumen without any stent deposition. With their superior physical advantages, I-SEMSs may prevent various stent-related complications in non-vascular luminal organs, including the esophagus, biliopancreatic duct, colon, and airway. Furthermore, they seem to be particularly useful in tortuous anatomical structures of non-vascular luminal organs.

I-SEMSs have been extensively used in managing vascular stenosis [11,12,25]. Clinical investigators reported excellent outcomes of I-SEMS placement for primary superficial femoral artery lesions to prolong stent patency and reduce stent fracture [26,27,28,29,30]. However, I-SEMSs have high variations of stent length caused by shortening and elongation during stent placement [12]. In our study, wherein I-SEMSs were applied for the first time in the rat esophagus, stent shortening and elongation during stent placement did not occur because the esophageal stent delivery system has a better profile than the vascular stent delivery system.

Our study has several limitations. First, tissue responses were observed in the normal rat esophagus after stent placement. Stent-induced hyperplasia formation after stent placement may differ with esophageal strictures. Second, only a few tests of the stent samples were conducted in this study. Although the bending test was successfully performed to compare the degree of stent deformation, the bending stiffness could not be investigated because of the lack of a measuring device. Additional mechanical properties of the newly developed I-SEMS for non-vascular luminal organs, including radial force, axial tension, and torsion test findings, should be evaluated. Although the tortuous anatomical structures of non-vascular luminal organs have not been investigated to prove the benefits of the high flexibility of I-SEMSs, our results demonstrated the basic concept of the effectiveness and safety of these stents in the rat esophagus.

5. Conclusions

Formation of stent-induced tissue hyperplasia was evident in both the I-SEMS and K-SEMS groups. Placement of I-SEMSs with a small cell size and high flexibility relatively suppressed stent-induced tissue hyperplasia compared with placement of K-SEMSs in the rat esophageal model. Placement of I-SEMSs with excellent flexibility seems to be applicable to highly tortuous non-vascular luminal organs for palliative treatment of malignant obstructions. Although further preclinical studies are required to investigate its efficacy and safety, I-SEMS placement can be a palliative therapeutic option for malignant stricture diseases of non-vascular organs.