Histopathological and duplex characteristics of deep venous stents: A novel microsurgical small animal model

Abstract

Background:

Venous stent preclinical testing typically uses large animal models. We created two models of infrarenal inferior vena cava (IVC) stenting to study patency, flow rate, wall shear stress, and endothelial coverage over the stents in a microsurgery rodent model.

Methods:

In Group 1, Sprague-Dawley rats underwent IVC stenting. Patency was measured by ultrasonography immediately postoperatively (POD-0) and at postoperative day 1 (POD-1). Stents underwent histological analysis up to 21 days postimplantation. Group 2 stented rats received periprocedural anticoagulation (enoxaparin 6 mg/kg SC) and were harvested to evaluate stent coverage using scanning electron microscopy (SEM). En face immunofluorescence (IF) staining was performed in Group 3. Human samples from stent occlusions were collected in the endovascular suite and compared with the animal histological findings.

Results:

In Group 1, 65% of the stented animals (13/20) were patent POD-0, increasing to 85% (17/20) by POD-1. Histologically, in-stent restenosis (ISR) was 6.2 ± 3.8%, and residual thrombosis (RT) was 32.01 ± 23.12% of the IVC total luminal stent area (LSA). These animals’ pathological processes are similar to those in humans. The mean IVC vein wall thickness (VWT) was 80.18 ± 13.14 μm. In Group 2, all animals were patent (21/21) postimplantation and remained patent in POD-1. IF demonstrated stent coverage with CD31⁺ endothelial cells. Total stent endothelialization by SEM reached 69.53 ± 14.84% at 7 days and increased to 91.51 ± 11.23% at 21 days.

Conclusion:

Quantifying rat venous stent patency and endothelialization via histology and microscopy revealed complete patency with anticoagulation and pathological occlusions (RT, ISR) when withholding anticoagulation. This rodent model will facilitate the preclinical testing of future venous stents.

Keywords

chronic venous disease deep vein thrombosis (DVT)post-thrombotic syndrome restenosis stent vascular biology

Background

Venous stenting procedures in the United States have increased significantly, with over 69,500 stents implanted from 2005 to 2025.^1,2 The approval of the first dedicated venous stent by the US Food and Drug Administration (FDA) in 2019, along with ongoing technological progress and increased clinical awareness, has driven this rapid growth.^3,4 Projections show that this trend will continue, possibly reaching 21,300 procedures annually by 2035 as venous thromboembolism (VTE) rates increase, with yearly costs between US$7 and US$10 billion.^5–7 Despite advances in technology, stent complication rates remain significant. In a recent meta-analysis of 17 trials, venous stents achieved a primary patency of approximately 80% at 12 months.⁸

Venous-specific pathological processes lead to stent complications, including in-stent restenosis (ISR), arising from diffuse intimal thickening (DIT), and thrombus formation. Modifiable factors that can be manipulated to improve outcomes include the technical performance of the procedure and stent design characteristics that affect flexibility, radial strength, and biocompatibility.^9–15 Antithrombotic therapy is the mainstay of therapy for maintenance of arterial stent patency; however, there is no standard agent, dose, or duration that has been agreed upon as standard of care to prevent the progression of these pathological processes in veins.¹⁶ Arterial stents have benefited from innovations such as gene and protein cargo delivery, peptide modifications, drug-eluting coatings, and novel biomaterials, including resorbable scaffolds, which promote endothelialization and reduce neointimal hyperplasia to improve clinical outcomes.^17–23 However, these features are not available in venous stents, and the currently available preclinical models to test such features are expensive and require the use of large animals.

To address this gap, we developed a novel rat model of venous stenting that allows detailed assessment of venous stent performance using ultrasound, histology, scanning electron microscopy (SEM), and immunofluorescence (IF) with or without anticoagulation therapy, to better understand patency and stent endothelialization.

Methods

This research followed University of Michigan Institutional Animal Care and Use Committee (IACUC) policies, US Government Principles for the Use and Care of Vertebrate Animals in Testing, Research, and Training, and ARRIVE guidelines 2.0.²⁴ Consent was waived by the institutional review board as the samples were part of routine clinical management. Human iliofemoral deep vein thrombosis (DVT) samples were obtained via biopsy during open or endovenous treatment.

Study design

Sample size was defined a priori utilizing power calculations, for an alpha = 0.05 and power of 80%, allocating n ⩾ 3 animals per group per condition as specified in the supplemental data. Sprague-Dawley rats (Charles River Breeding Labs, USA) (n = 54) were assigned into Group 1 (n = 22), which included surgical shams (n = 1), a stasis thrombotic model (n = 1), and inferior vena cava (IVC) stenting (n = 20) with no anticoagulation therapy, with endpoints at 7, 14, and 21 days (n = 6–8), evaluating postoperative patency, flow rate, and wall shear stress by ultrasonography and histological characterization of the vein wall. Group 2 (n = 25) included surgical shams (n = 2) and stented rats with endpoints at 1, 3, 5, 7, 14, and 21 days (n = 3–4) to evaluate endothelial coverage using SEM. Group 3 (n = 7) included surgical shams (n = 2) and stented rats with endpoints at 1 (n = 1) and 21 days (n = 4), who received anticoagulation, to evaluate cell characterization using en face IF staining. Human histological samples from stent thrombosis were collected from the endovascular suite (n = 3) and compared with animal samples in Group 1.

Animal procedures

Adult male Sprague-Dawley rats, 8–10 weeks old, were anesthetized with 3% isoflurane. Eye lube (SKU-065441; Optixcare, USA) was used to prevent ocular damage. Preprocedure, the anteroposterior (AP) infrarenal IVC diameter was measured by ultrasound (ACUSON Antares; Siemens, USA) on B-mode to guarantee adequate sizing of the stent with an overstretch of the vein wall of approximately 20%, correlated with accepted human sizing.²⁵ Preoperatively, the abdomen was cleaned aseptically with gauze soaked in chlorhexidine scrub and solution (17030-003-16; Vetoquinol, USA). Buprenorphine 0.65 mg/kg (SKU-099114; Ethiqa XR, 1.3 mg/mL; Fidelis Animal Health, USA) was administered subcutaneously for analgesia. The abdomen of the rat was covered with ethylene oxide-exposed sterile Press’ n Seal (SKU PSS-140; Glad Products Co., USA).²⁶

Using a surgical microscope (M690; Leica, USA), through a midline laparotomy, the abdominal contents were reflected carefully, displacing organs on the right side, and covered with sterile saline-moistened gauze (Figure 1). The infrarenal IVC was carefully dissected from the aorta. The IVC posterior venous branches were cauterized with a low-temperature fine-tip Bovie cautery (AA00; Symmetry Surgical Inc., USA), and side branches were ligated using 7-0 DemeLENE (PM197011F13M; DemeTech, USA). Two 12 × 2.25-mm Heifetz microclips were temporarily placed (3795-28; SCANLAN, USA) on the infrarenal IVC and proximal to the IVC bifurcation approximately 1.5 cm apart. An 8-0 DemeLON (NL868007C7P; DemeTech) U-stitch was placed in the proximal IVC. A stent sheath was inserted retrograde through the U-stitch into the IVC and advanced to deploy a venous self-expanding 3 × 5-mm bare metal nitinol stent with a stent introducer. After the stent deployment and removal of the stent sheath, the U-stitch was tightened for hemostasis, and the microclips were removed (Figure 1). Muscle and skin layers were closed with continuous 3-0 DemeCRYL sutures (G183019F4P; DemeTech). After the procedure, the animals were monitored in a recovery cage with a heat source (a heating lamp) until they were fully recovered.

Figure 1.

(A) In vivo microsurgical model of IVC stent deployment in rodents. (B) Rat IVC with stent deployment, postimplantation, and gross specimen after harvesting. (C) Study design.

Duplex ultrasound

After closing the abdomen (POD-0) and on POD-1, color/pulsed-wave Doppler was performed to evaluate stent patency by assessing venous flow with a 10-MHz probe. The patency was determined as the IVC stent area with flow divided by the total IVC area. We characterized the mid-portion of the stent as patent (defined as > 75% of the intraluminal area visualized with spontaneous flow via color duplex), partially patent (< 75% and > 25%), or obstructed (< 25%). The area of the IVC with flow was measured using IVC longitudinal and transverse 2D images on ImageJ Fiji software (https://imagej.net/software/fiji/, National Institutes of Health, USA). To calculate the murine wall shear stress, we quantify the ultrasound spectral Doppler velocity contours from the IVC using the open‑source PlotDigitizer software (https://plotdigitizer.sourceforge.net, PORBITAL, USA). The digitized contours represent the peak velocities within the Doppler spectrum. Under the assumption of a parabolic velocity profile, mean velocities were calculated as one half of the maximum velocity. The time-averaged mean velocity across the contour was then used in Equation (1) to calculate the average flow rate, assuming a circular vessel cross-section to estimate the vessel area:

Q = v_{a v g} A = v_{a v g} π R^{2},

(1)

where $Q = a v e r a g e f l o w r a t e$ , $A = v e s s e l c r o s s s e c t i o n a l a r e a,$ , $v_{a v g} = t i m e a v e r a g e d v e l o c i t y, R = r a d i u s .$ Wall shear stress was then calculated analytically using Equation (2):

τ_{w} = 4 u Q / π R^{3},

(2)

where $τ_{w} = w a l l s h e a r s t r e s s$ , $u = b l o o d v i s c o s i t y$ , $Q = a v e r a g e f l o w r a t e$ , $R = r a d i u s$ . Equation (2) is derived from the Hagen-Poiseuille equation, under the assumptions of steady-state flow and Newtonian fluid behavior.^27,28 Blood viscosity was set to 6 centipoise, based on experimentally measured literature values of rat blood viscosity at comparable flow rates.²⁹

Histopathology

En bloc human biopsy samples were formalin-fixed for 8 hours and paraffin-embedded, sectioned, and slide-stained with hematoxylin & eosin (H&E) and Masson trichrome at the Orthopaedic Research Laboratories (ORL) Histology Core at the University of Michigan. The slides were classified by a cardiovascular pathologist according to previously defined pathological features: fresh thrombus, organizing thrombus, old thrombus, or DIT.¹⁵ Images were taken using a brightfield microscope (Eclipse E400; Nikon, USA) with a Digital Sight camera (DS-Ri1; Nikon) and controller (DS-U4; Nikon).

Stents were harvested after 1, 3, 5, 7, 14, and 21 days of stent implantation. The IVC was carefully dissected from its proximal and distal ends, separated from the aorta, removed, and washed in sterile saline (PCS1650; Medline, USA) for 1 × 5 minutes. Rat IVC stent specimens in Group 1 were fixed in 10% neutral buffered formalin (MCHEM110; Medline) for 24 hours and subsequently processed for plastic embedding at the ORL Histology Core.^30,31 Blocks were cut in half transversely and then sectioned at a thickness of 5 μm using a rotary microtome with a 55° angle tungsten carbide blade for plastic-embedded tissues.³² H&E and trichrome stained slides were used to quantify vein wall thickness (VWT), ISR, and residual thrombosis (RT).^33,34 Sham rats were housed and treated under the same conditions as the stented rats, but just underwent laparotomy with retroperitoneal dissection, without IVC stenting. The thrombotic stasis model was performed as described by Varma et al.³⁵ The mean VWT was quantified by taking 12 measurements from the intima to the adventitia per sample.^36,37 In-stent restenosis was calculated as ISR% = residual lumen area/area of stent wire border. RT was calculated as RT% = thrombus area/total IVC lumen. The stent area (SA) percentage was calculated as SA% = stent struts area/total imaged IVC area.

Scanning electron microscopy (SEM)

Harvested IVCs from Group 2 were fixed in a 10% formalin solution for 24 hours, followed by a 2-hour fixation in 5% glutaraldehyde and 2% paraformaldehyde solution (16536-06; Electron Microscopy Sciences, USA), ensuring minimal disturbance to the endothelial cells, followed by a phosphate-buffered saline (PBS) (10010049; Gibco, USA) wash to remove calcium deposits, 2 × 5 minutes. The samples were fixed in 1% osmium tetroxide (75632-10ML; Sigma-Aldrich, USA) for 60 minutes, followed by a series of ethanol (2701; Decon Laboratories, Inc., USA) washes for dehydration: 1 × 15 minutes each at 50%, 60%, 70%, 80%, 90%, 95% concentration, and 2 × 15 minutes at 100% concentration.³⁸ The IVCs containing the stents were longitudinally sectioned, creating two half-cylinders. For critical point drying, the conditions used were T_c: 23°C, P_c: 1.0 bar, and CO₂ IN: medium for 4 hours (EM CPD300; Leica) at the University of Michigan Microscopy Core. Samples were then mounted on SEM holders (RS-MN-12-002920-3; Rave Scientific, USA), coated with a thin 10-nm layer of gold, and imaged using the scanning electron microscope (EVO15; Zeiss, USA) on the same day to maintain imaging quality. After SEM processing, the image quality was assessed in the two IVC sections for analysis per animal. Images were taken at 60× to 1000× magnification, with a scan speed of 6, and an electron high tension (EHT) of 5 kV.³⁹ Stent endothelial coverage (SEC) was calculated as the percentage of endothelial-covered area relative to the total stent area in 60× images. SEM images were analyzed using QuPath 0.6.0 software (https://qupath.github.io/, Queen’s University Belfast, UK).

En face immunofluorescence (IF) staining

For en face IF microscopy, the stented vessels in Group 3 were fixed with 4% formaldehyde solution (28908; ThermoFisher, USA) in a HEPES buffer (25-060-CI; Corning, USA) and permeabilized with 0.2% Triton X-100 (T9284; Sigma-Aldrich) in PBS (10010049; Gibco, USA).^40,41 After washing with PBS for 5 minutes, the slides were blocked with 0.2% bovine serum albumin (BSA) (B4287-25G; Sigma-Aldrich) in PBS. The stented vessels were incubated for 2 hours in a PBS-BSA solution containing primary antibodies at 1/100 dilution: CD31-647 conjugate (34624S; Cell Signaling Technology [CST], USA) and αSMA-594 conjugate (36110S; CST). After double staining with 4′,6-diamidino-2-phenylindole (DAPI), the stented vessels were placed on Superfrost Plus microscope slides (1255015; Fisher Scientific, USA), mounted using Prolong Gold antifade reagent with DAPI (P36931; Fisher Scientific, USA), and coverslipped under micro-coverglass (72204-04;Electron Microscopy Sciences). Images were acquired and processed using an automated fluorescence microscope (BZ-X800E; Keyence Corp., USA) from the University of Michigan Frankel Cardiovascular Center. High-precision image stitching and Z-stacking for multilayer capture features were used to obtain full-focused images and advanced 3D analyses of the stent structures.

Statistical analysis

Statistical analysis was performed in Prism 10.6.0 software (https://www.graphpad.com, GraphPad). Statistical significance between groups was determined using Brown-Forsythe and Welch analysis of variance (ANOVA) tests with multiple comparison corrections or unpaired Student’s t-test with Welch correction. The presented quantifications were measured blindly by two separate individuals. Results were considered significant when p ⩽ 0.05. Data are presented as mean ± SD.

Results

Preprocedure, the infrarenal IVC diameters by ultrasound were 2.59 mm (± 0.14 mm). The mean rat weight was 521.80 ± 128 g. All the implanted stents were deployed in the infrarenal IVC. All the animals survived until their predefined time.

Ultrasound postimplantation and postoperative patency

In Group 1, the intraluminal stent area with flow was measured 15 minutes after skin closure at postimplantation (POD-0), with 65% (13/20) of the stented animals patent (patency > 75%: 9/20, patency < 75%: 4/20), and 35% (7/20) obstructed. After 24 hours, 85% of the stents (17/20) were patent (patency > 75%: 9/20, patency < 75%: 8/20) on postoperative day 1 (POD-1), with partial recanalization observed in four of seven obstructed stents (Figure 2A). The mean stent patency at POD-0 was 44.51 ± 40.54%, increasing at POD-1 to 56.82 ± 32.42%. Group 2, which received enoxaparin 6 mg/kg SC postprocedure, had complete patency in all animals (21/21) at both POD-0 (87.01 ± 4.83%) and POD-1 (86.67 ± 5.12%). There was a significant difference in the patency between Groups 1 and 2 at POD-0 (p = 0.0004) and at POD-1 (p = 0.0010) (Figure 2E). The mean volumetric flow rate at baseline (preoperative) was 3.48 × 10^–7 ± 3.19 × 10^–7 m³/s; for the patent stents, it was 3.33 × 10^–7 ± 2.52 × 10^–7 m³/s; and for the partially patent stents, it was 1.18 × 10^–7 ± 1.68 × 10^–7 m³/s. No significant differences were found between these groups. The mean wall shear stress at baseline was 12.62 ± 11.82 dyn/cm², 17.46 ± 10.70 dyn/cm² for the patent stents, and 34.64 ± 19.68 dyn/cm² for the partially patent stents. There was a significant increase in the wall shear stress in the partially patent stents compared with the patented stents (p = 0.0011), and when compared to baseline (p ⩽ 0.0001).

Figure 2.

(A) Ultrasound evaluation of IVC stents. (B) Outcomes by ultrasound. (C) Stent patency group distribution postoperatively, 15 minutes after skin closure (POD-0) and at follow-up postoperative day 1 (POD-1). (D) Ultrasound Doppler velocity contours quantification. (E) Stent patency mean % between Groups 1 and 2. (F) Volumetric flow rate (Q) preoperatively (baseline) compared to patent and partially patent stents. (G) Wall shear stress (τ_w) preoperatively (baseline) compared to patent and partially patent stents.

Stent histological findings (Group 1)

Histological changes in venous cross-sections stained from both Sprague-Dawley rats and humans were reviewed. Both samples showed similar features, including normal veins, DVT, and DIT (Figures 3A and 3B). The IVC generally exhibited a larger diameter compared to the aorta. In the rat model, venous thrombosis (VT) was identified on days 7, 14, and 21 after stent implantation without anticoagulation. Ultrasound assessment confirmed stent occlusion with reduced venous flow.²⁵ In Group 1, the mean VWT was 80.18 ± 13.14 µm, with no significant differences between days 7 (75.75 ± 13.20 µm), 14 (86.56 ± 12.40 µm) (p = 0.2524), or 21 (69.92 ± 4.08 µm) (p = 0.8389). Overall, the ISR% was 6.2 ± 3.8%. No significant differences were observed between the 7, 14, and 21-day time points. In 12/20 (60%) of the stented rats, RT was observed. Across all time points, the mean RT was 32.01 ± 23.12%. At 7 days, the RT was 4.59 ± 0.20%; at 14 days, 37.91 ± 20.69% (p = 0.1527); and at 21 days, 36.54 ± 27.42% (p = 0.2494) without significant differences (Figure 3C).

Figure 3.

(A) Histological analysis comparing normal and pathological features of the IVC in stented rats (left panel) and samples from human ISR cases (right panel). (B) IVC measurements. (C) Group 1 mean vein wall thickness, in-stent restenosis, and residual thrombosis at 7, 14, and 21 days postimplantation.

Endothelial coverage by SEM (Group 2)

Across all animals, the area of each stent in relationship with the total area of the IVC was 23.35 ± 4.92%, with no significant differences among the groups. High-power magnification SEM (1000×) stent strut images showed fine surface morphology. A nonstented normal IVC exhibited an aligned monolayer of venous endothelial cells (VECs) situated above the extracellular matrix (ECM) (Figures 4A and 4B). Postimplantation, the ECM was exposed and visible through the stent fenestrae at day 7. The deposited ECM was later covered by endothelium growing from the nonstented edges until day 21, covering the stent (Figure 4C). After 1 day, stent coverage by endothelial cells was 21.57 ± 11.11%; after 3 days, it was 29.67 ± 8.55%; after 5 days, 68.06 ± 6.90%; after 7 days, 69.53 ± 14.84%; after 14 days, 87.33 ± 15.39%; and after 21 days, 91.51 ± 11.23% (Figures 4D–4F). By day 21, endothelial cells appeared to align parallel to the direction of blood flow on SEM images. This rapid progression after day 5 demonstrates that the rat IVC achieves near-complete stent endothelialization within approximately 3 weeks postimplantation.

Figure 4.

(A) Scanning electron microscopy of a normal IVC (left) and a stent strut covered by endothelium (right). (B) Stent coverage quantification on QuPath software for stent coverage. (C) Total area of stent implanted in the IVC and percentage of stent coverage measurements at 1, 3, 5, 7, 14, and 21 days postimplantation. (D) Timeline of stent coverage after implantation. Day 0: Nitinol stent struts with RBCs. Day 7: extracellular matrix deposition. Day 14: venous endothelial sprouting. Day 21: complete endothelial coverage. (E) Time course comparison of our novel rat IVC models with large animal venous and arterial stent models.

En face IF staining (Group 3)

To better characterize the cell identity of the layer covering the venous stents, we first performed en face IF of DAPI, CD31, αSMA markers on IVCs of sham rats, which demonstrated the presence of VECs and alignment of blood flow direction on the IVC. After 1 day of stenting, we observed a few cells lining the stent struts. After 21 days, the stented vessels had a continuous surface coverage of positive CD31⁺ VEC cells (Figures 5A–5D). The overall morphology of these CD31⁺ VECs over the wire is similar to that observed on the IVC of the sham rat.

Figure 5.

(A) En face immunofluorescence staining of control and stented IVC demonstrating nuclei (DAPI, blue), CD31 (purple) venous endothelial cells, and αSMA vascular smooth muscle cells (red). (B) En face immunofluorescence of a 1-day postplantation stent with DAPI, CD31, and αSMA staining; 4× and 40× high-precision, full-focused images featuring CD31 and DAPI staining. (C) En face immunofluorescence of a 21-day postplantation stent with DAPI, CD31, and αSMA staining; 4× and 40× high-precision, full-focused images featuring CD31 and DAPI staining. (D) High-precision (4× and 40×), full-focused images featuring CD31 and DAPI staining of struts in 2D and 3D reconstructions of the stent structures.

High-precision (40×) full-focused 3D image reconstruction featuring CD31 and DAPI staining suggests that the scaffold deposited next to the stent struts forms an endothelial lining resembling a speed bump-like appearance at 45° (Figure 5). We also demonstrated how the CD31⁺ VEC cells can grow vertically over the wire until a full monolayer is achieved. These detailed endothelial cell features cannot be observed unless specific 3D reconstructions are performed at high magnification in advanced IF imaging or by altering the angle of SEM imaging, as shown in Figure 4D, which shows the lateral side of a stent strut covered at 21 days.

Discussion

The main findings of these preclinical deep vein stenting models were complete stent patency by ultrasound with the use of a single dose of perioperative anticoagulation, and the virtually complete stent coverage by day 21 with CD31⁺ VECs. In the first stenting model without anticoagulation, 6% ISR was identified by histopathological analysis. Without anticoagulation, the stent patency improved after 24 hours as assessed by ultrasound. Histopathological analysis demonstrated a mean RT of 37% of the lumen of the stented vessels at 21 days without anticoagulation. This model can enable the use of preclinical thrombosis models and imaging methods to help characterize thrombosis resolution in deep vein stents.^42–45

In our nonanticoagulated stenting model, we also observed a modest luminal ISR of 6%. We believe that this stenting model can evaluate the amount of ISR of the IVC, and facilitate the testing of future drug-eluting coatings. The structural and pathological characteristics of thrombosis, ISR, and DIT observed in human stented tissue have gained significant relevance due to their association with clinical outcomes.^15,46 Recently, medical interest in the pathology of DIT within stents has intensified due to the capacity of these features to predict restenosis/occlusion. Zhao et al. suggested that integrating histopathological assessment into postthrombectomy evaluation can help optimize postthrombotic (PTS) management strategies for iliofemoral in-stent occlusion. They reported that DIT was the predominant feature in occlusions beyond 3 months and that old thrombus and RT higher than 30% predicted restenosis/occlusion.⁴⁷

In the second stenting model that received anticoagulation, all stents remained patent during the perioperative period, and endothelial coverage increased from 69% at 7 days to 91% at 21 days. Our data show that simply by withholding periprocedural anticoagulation, it is possible to generate a model that has histologic characteristics that mimic those of partially or completely occluded human venous stents. This model could be used to study combinations of anticoagulant, antiinflammatory, and antiplatelet therapies poststenting, and their patency outcomes. Establishing optimal medical therapy poststenting was recently identified as the number-one research priority outlined by the Society of Interventional Radiology Foundation (SIRF) Research Consensus Panel.⁴⁸

The time to complete stent endothelialization is a critical factor in determining the need and duration of antiplatelet therapy in arterial stents and remains an unanswered question in the field of venous stents. Historically, in vivo models and computational models have attempted to recreate endothelial behavior after stenting and forecast cell growth and drug interactions.⁴⁹ Nevertheless, no model can fully capture all the biological interactions present in clinical trials. In our animal study, rapid endothelialization was observed with SEM, consistent with other venous stent models and the SEM IVC findings in rats.^50–52 In humans, the stent endothelialization of veins has only been documented in a case report using angioscopy after 30 days, with few studies examining cell features via SEM.^53–55 This direct observation method for endothelial assessment was employed by Che et al. to confirm the full endothelial coverage of venous stents 3 months after implantation in the iliac vein of a sheep.⁵⁶

When compared to established arterial stent models, the pace of endothelialization in the rat IVC is intermediate — faster than the pace in larger venous animal models but generally slower than arteries (Figure 5E).^{18,39,50,51,57} Previous animal (porcine and rabbit) arterial models and human coronary stents have demonstrated faster endothelial coverage, reflecting a faster regenerative capacity, possibly due to different flow dynamics and biological differences.^{18,39,57–61} Overall, our rat model offers a more efficient method for studying venous stent endothelialization as an alternative to large animal models due to reduced space utilization, a short procedural duration, and reduced personnel requirements for surgical and anesthesia assistance.

Technical considerations

For successful stent implantation, several considerations are crucial. First, maintain the rat’s temperature with a heating source for survival and to monitor anesthesia during procedures. Keep total anesthesia under an hour per procedure. Second, to prevent blood loss while separating the aorta from the IVC, start at the infrarenal IVC and continue distally. Third, when the IVC was sutured, the first two sutures should leave enough space for adequate canulation with the stent sheath and to avoid postimplantation IVC narrowing. Fourth, during stent deployment, it is recommended that the stent sheath is maintained with one hand while the stent introducer is pushed with the other hand, slowly maintaining alignment with the IVC.

In Group 2, which received anticoagulation, a superficial skin hematoma was observed, presenting as a palpable abdominal mass that was confirmed by ultrasound in three anticoagulated rats. The rats showed no gastrointestinal symptoms and remained hemodynamically stable. Two of the superficial hematomas resolved spontaneously. Only one rat with a persistent hematoma required treatment with ultrasound-guided aspiration under anesthesia after aseptic preparation, which involved removing 2 mL of sanguineous fluid without recurrence or complications in accordance with institutional animal care guidelines. Finally, immunohistochemistry is not recommended for plastic-embedded nitinol stent staining, as this process which requires advanced cutting techniques, drastically disrupts the morphology, and presents antigen retrieval difficulties.⁶² En face IF is a more robust method for cell characterization.

Limitations

The main limitation of our study arises from its focus on only one venous stent design, with predetermined animal baseline conditions including a fixed stent and IVC diameter. These considerations were established to increase the reproducibility of this model, but may not be fully reflective of the changes that the respiratory rate, hematocrit, and temperature can have on local hemodynamics and shear rates. These conditions, together with the assumed circular vessel cross-sectional area, the assumed steady-state Newtonian flow, and the predefined blood viscosity, make the observed hemodynamics and wall shear stress calculations rough approximations rather than exact measurements. Second, human stenting encompasses substantial differences in vessel diameter, flow, and immunobiology.⁶³ Our model does not completely reflect the complexity of the human postthrombotic condition; instead it resembles nonthrombotic iliac vein lesions (NIVL).

In this study, it was not possible to compare the rates of ISR formation observed between the groups at later time points by histological analyses without substantially increasing the number of animals. This is because the animal specimens can only be processed for plastic embedding for detailed histological morphology (Group 1), or for SEM/IF microscopy (Groups 2 and 3). This limitation can be addressed by novel imaging modalities that can overcome the resolution limitations of ultrasound, such as molecular magnetic resonance imaging (MRI) and fluorodeoxyglucose positron emission tomography/computed tomography (FDG-PET/CT).^64,65

Finally, en face IF and SEM have been the gold standard for assessing stent endothelialization; however, other emerging methods, such as intravascular ultrasound (IVUS), optical coherence tomography (OCT), micro-CT, and others, are being investigated and should be compared in future studies.^66–77 Further characterization of cell proliferation (Ki-67), apoptosis (Caspase-3), spontaneous thrombolysis (tissue plasminogen activator [tPA]), immune cell characterization (CD68), and other features of VECs (von Willebrand factor [VWF], vascular endothelial [VE]-cadherin) can be achieved with advanced IF imaging by using conjugated antibodies in the far-infrared wavelengths. With the incorporation of these technologies, it will be possible to characterize the role of anticoagulation and antiaggregation in venous in-stent restenosis prevention.

Future directions

Future uses of this model should focus on postthrombotic stenting to investigate the whole spectrum of the disease, allowing for comparative studies of different stent designs in combination with antithrombotic/antiplatelet and antiinflammatory therapies. Another area of recent development is the study of biomaterial interactions to improve endothelialization, similar to the approach taken by Wu et al. in sheep and Oliver et al. in rabbits.^78,79 Finally, future studies should incorporate the use of unbiased technologies (transcriptomics and proteomics) to establish the specific drivers of stent failure (DIT and ISR) more accurately.^80–82

Conclusions

The results of this study demonstrate that this novel preclinical model for venous stenting, characterized by ultrasound imaging, histopathology, and microscopy, enables the identification of pathological processes (thrombosis, ISR, DIT) present in human pathological stents and can correlate with clinical outcomes, helping to tailor and optimize management strategies. Administering a single periprocedural dose of anticoagulation allows for the analysis of stent coverage (SEM) and cell phenotype (IF) up to 21 days after stenting. With this robust model, we provide the means to further assess the performance of new stent designs, coatings, and pharmacological combinations in the future.

Supplemental Material

sj-pdf-1-vmj-10.1177_1358863X261436329 – Supplemental material for Histopathological and duplex characteristics of deep venous stents: A novel microsurgical small animal model

Supplemental material, sj-pdf-1-vmj-10.1177_1358863X261436329 for Histopathological and duplex characteristics of deep venous stents: A novel microsurgical small animal model by Oscar Moreno, Trevin Eggleston, Catherine E Luke, Amber Clay, Kate Micallef, Kiran Kumar, Sabrina Rocco, Kevin Hughes, David Gordon, Daniel D Myers, Thomas W Wakefield, Peter K Henke and Andrea T Obi in Vascular Medicine

Footnotes

Acknowledgements

We would like to thank the University of Michigan Frankel Cardiovascular Center for their support with microscope equipment assistance, Hongyu Liu for his help on IF imaging, Carol Whitinger and Emma Snyder-White from the Orthopaedic Research Laboratories (ORL) Histology Core for their assistance during the histology processing, Jing Liang and Megan Bennett from the Microscopy Core for their help with SEM sample processing and imaging, and Ismael Assi for his help with the wall shear stress calculations.

Declaration of conflicting interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article. Daniel Myers and Andrea Obi received funding for preclinical trials related to deep vein stenting from Medtronic. Kevin Hughes is currently a Principal R&D Engineer at Medtronic. The remaining authors have no conflicting interests.

Funding

This work was supported by the Elizabeth Anne Baiardi Research Fund, the Jobst Foundation, the National Institutes of Health (P30AR069620), and Medtronic. Medtronic provided custom stents at no cost and supported the development of this microsurgical technique in animals. Medtronic did not participate in data analysis or interpretation of the results. Medtronic reviewed the final version of the manuscript.

ORCID iDs

Oscar Moreno

Trevin Eggleston

Kevin Hughes

David Gordon

Peter K Henke

Andrea T Obi

Supplemental material

Supplemental material for this article is available online.

References

Tang

Lakhter

Zack

, et al. Contemporary nationwide trends and in-hospital outcomes of adjunctive stenting in patients undergoing catheter-directed thrombolysis for proximal deep venous thrombosis. J Vasc Surg Venous Lymphat Disord 2021; 9: 62–72.e1.

Divakaran

Meissner

Kohi

, et al. Utilization of and outcomes associated with intravascular ultrasound during deep venous stent placement among Medicare beneficiaries. J Vasc Interv Radiol 2022; 33: 1476–1484.e2.

Berger

Jaffe

York

TN.

Iliac compression syndrome treated with stent placement. J Vasc Surg 1995; 21: 510–514.

Breen

Role of venous stenting for venous thromboembolism. Hematology Am Soc Hematol Educ Program 2020; 2020: 606–611.

Deitelzweig

Johnson

Lin

, et al. Prevalence of clinical venous thromboembolism in the USA: Current trends and future projections. Am J Hematol 2011; 86: 217–220.

Grosse

Nelson

Nyarko

, et al. The economic burden of incident venous thromboembolism in the United States: A review of estimated attributable healthcare costs. Thromb Res 2016; 137: 3–10.

Transparency Market Research, Inc. Venous Stents Market, https://www.transparencymarketresearch.com/venous-stents-market.html (2025, accessed Aug 22, 2025).

Badesha

Black

Khan

, et al. A meta-analysis of the medium- to long-term outcomes in patients with chronic deep venous disease treated with dedicated venous stents. J Vasc Surg Venous Lymphat Disord 2024; 12: 101722.

Massi

Zamboni

What are the ideal characteristics of a venous stent?

Veins and Lymphatics. Published online Jun 18, 2021. DOI: 10.4081/vl.2021.9739.

10.

Smith

Butts

Owens

, et al. Outcomes following stenting for symptomatic chronic iliofemoral venous stenosis – a comparison of three stent types. J Vasc Surg Venous Lymphat Disord 2025; 13: 102208.

11.

Chick

JFB

Abramowitz

Osher

, et al. Radiographic findings of distressed venous stents and inferior vena cava filters: Clinical implications. AJR Am J Roentgenol 2017; 209: 1150–1157.

12.

Dabir

Feisst

Thomas

, et al. Physical properties of venous stents: An experimental comparison. Cardiovasc Intervent Radiol 2018; 41: 942–950.

13.

Jayaraj

Fuller

Raju

, et al. In-stent restenosis and stent compression following stenting for chronic iliofemoral venous obstruction. J Vasc Surg Venous Lymphat Disord 2022; 10: 42–51.

14.

Saleem

Raju

An overview of in-stent restenosis in iliofemoral venous stents. J Vasc Surg Venous Lymphat Disord 2022; 10: 492–503.e2.

15.

Williams

Nicklas

Obi

, et al. Pathologic characteristics of human venous in-stent stenosis and stent occlusion. J Vasc Surg Venous Lymphat Disord 2023; 11: 109–118.e2.

16.

Kishore

Khaja

Thornburg

, et al. Antithrombotic therapy after venous interventions: AJR Expert Panel Narrative Review. AJR Am J Roentgenol 2022; 219: 175–187.

17.

Brito

Chandrasekhar

Little

, et al. Non-viral eNOS gene delivery and transfection with stents for the treatment of restenosis. Biomed Eng Online 2010; 9: 56.

18.

Van Belle

Tio

Couffinhal

, et al. Stent endothelialization: Time course, impact of local catheter delivery, feasibility of recombinant protein administration, and response to cytokine expedition. Circulation 1997; 95: 438–448.

19.

Van Beusekom

HMM

Ertaş

Sorop

, et al. The Genous^TM endothelial progenitor cell capture stent accelerates stent re-endothelialization but does not affect intimal hyperplasia in porcine coronary arteries. Catheter Cardiovasc Interv 2012; 79: 231–242.

20.

Ranjan

Kumar

Hardikar

, et al. Human blood vessel-derived endothelial progenitors for endothelialization of small diameter vascular prosthesis. PLoS One 2009; 4: e7718.

21.

Wen

Qiu

, et al. Vascular stent with immobilized anti-inflammatory chemerin 15 peptides mitigates neointimal hyperplasia and accelerates vascular healing. Acta Biomater 2024; 179: 371–384.

22.

Lim

Varcoe

Current status of and future prospects for drug-eluting stents and scaffolds in infrapopliteal arteries. J Clin Med 2024; 13: 1757.

23.

Paul

Elias

Shum-Tim

, et al. Bioactive baculovirus nanohybrids for stent based rapid vascular re-endothelialization. Sci Rep 2013; 3: 2366.

24.

Percie Du Sert

Ahluwalia

Alam

, et al. Reporting animal research: Explanation and elaboration for the ARRIVE guidelines 2.0. PLoS Biol 2020; 18: e3000411.

25.

Moreno

Luke

Clay

, et al. Rodent inferior vena cava venoplasty balloon model. J Vis Exp 2024; e66820.

26.

Wrobleski

Farris

Diaz

, et al. Mouse complete stasis model of inferior vena cava thrombosis. J Vis Exp 2011; (52): 2738.

27.

Vlachopoulos

O’Rourke

Nichols

WW.

McDonald’s blood flow in arteries: Theoretical, experimental and clinical principles. 6th ed. CRC Press, 2011.

28.

Katritsis

Kaiktsis

Chaniotis

, et al. Wall shear stress: Theoretical considerations and methods of measurement. Prog Cardiovasc Dis 2007; 49: 307–329.

29.

Jun Kang

Yeom

Lee

S-J

. A microfluidic device for simultaneous measurement of viscosity and flow rate of blood in a complex fluidic network. Biomicrofluidics 2013; 7: 054111.

30.

Van Beusekom

Whelan

van de Plas

, et al. A practical and rapid method of histological processing for examination of coronary arteries containing metallic stents. Cardiovasc Pathol 1996; 5: 69–76.

31.

Quentin

Poppe

Bär

, et al. A novel method for processing resin-embedded specimens with metal implants for immunohistochemical labelling. Acta Histochem 2009; 111: 538–542.

32.

Rippstein

Black

Boivin

, et al. Comparison of processing and sectioning methodologies for arteries containing metallic stents. J Histochem Cytochem 2006; 54: 673–681.

33.

Langeveld

Van Gilst

Tio

, et al. Angiotensin-(1–7) attenuates neointimal formation after stent implantation in the rat. Hypertension 2005; 45: 138–141.

34.

Henke

Wakefield

Kadell

, et al. Interleukin-8 administration enhances venous thrombosis resolution in a rat model. J Surg Res 2001; 99: 84–91.

35.

Varma

Moaveni

Dewyer

, et al. Deep vein thrombosis resolution is not accelerated with increased neovascularization. J Vasc Surg 2004; 40: 536–542.

36.

Dowling

Luke

Cai

, et al. Modulation of interleukin-6 and its effect on late vein wall injury in a stasis mouse model of deep vein thrombosis. JVS Vasc Sci 2022; 3: 246–255.

37.

Kessinger

Kim

Henke

, et al. Statins improve the resolution of established murine venous thrombosis: Reductions in thrombus burden and vein wall scarring. PLoS One 2015; 10: e0116621.

38.

Mukhamadiyarov

Bogdanov

Glushkova

, et al. EMbedding and backscattered scanning electron microscopy: A detailed protocol for the whole-specimen, high-resolution analysis of cardiovascular tissues. Front Cardiovasc Med 2021; 8: 739549.

39.

De Prado

Pérez-Martínez

Cuellas-Ramón

, et al. Time course of reendothelialization of stents in a normal coronary swine model: Characterization and quantification. Vet Pathol 2011; 48: 1109–1117.

40.

Andries

Brutsaert

Sys

SU.

Nonuniformity of endothelial constitutive nitric oxide synthase distribution in cardiac endothelium. Circ Res 1998; 82: 195–203.

41.

Yan

Yin

, et al. M2 macrophage-derived exosomes promote the c-KIT phenotype of vascular smooth muscle cells during vascular tissue repair after intravascular stent implantation. Theranostics 2020; 10: 10712–10728.

42.

Diaz

Obi

Myers

Jr , et al. Critical review of mouse models of venous thrombosis. Arterioscler Thromb Vasc Biol 2012; 32: 556–562.

43.

Diaz

Saha

Cooley

, et al. Choosing a mouse model of venous thrombosis. Arterioscler Thromb Vasc Biol 2019; 39: 311–318.

44.

Saha

Andia

Modarai

, et al. Magnetic resonance T₁ relaxation time of venous thrombus is determined by iron processing and predicts susceptibility to lysis. Circulation 2013; 128: 729–736.

45.

Stein-Merlob

Kessinger

Erdem

, et al. Blood accessibility to fibrin in venous thrombosis is thrombus age-dependent and predicts fibrinolytic efficacy: An in vivo fibrin molecular imaging study. Theranostics 2015; 5: 1317–1327.

46.

Oguri

Nakamura

, et al. Expression of fibroblast activation protein-α in human deep vein thrombosis. Thromb Res 2024; 241: 109075.

47.

Zhao

Liles

Obi

, et al. Histopathology of iliofemoral vein in-stent occlusion in postthrombotic syndrome: Findings from endovascular thrombectomy. Vasc Med 2026; 31: 81–89.

48.

Khaja

Obi

Sharma

, et al. Optimal medical therapy following deep venous interventions: Proceedings from the Society of Interventional Radiology Foundation Research Consensus Panel. J Vasc Interv Radiol 2022; 33: 78–85.

49.

McQueen

Escuer

Aggarwal

, et al. Do we really understand how drug eluted from stents modulates arterial healing? Int J Pharm 2021; 601: 120575.

50.

Barbati

Bechter-Hugl

Thomis

, et al. Evaluation of safety and performance of a new prototype self-expandable nitinol venous stent in an ovine model. JVS Vasc Sci 2023; 4: 100113.

51.

Zhang

Wang

Guo

An experimental study of inferior vena caval stent in canine. Chin Med J (Engl) 1999; 112: 816–821.

52.

Zhou

May

Liao

, et al Inferior vena cava ligation rapidly induces tissue factor expression and venous thrombosis in rats. Arterioscler Thromb Vasc Biol 2009; 29: 863–869.

53.

Tanabe

Mitarai

Ishibashi

, et al. Endothelialization of a venous stent at 1 month post implantation: First-in-human angioscopic assessment. J Invasive Cardiol 2020; 32: E248.

54.

Liu

Chen

, et al. Adventitial collagen cross-linking by glutaraldehyde reinforcing human saphenous vein—Implication for coronary artery bypass grafting. Braz J Cardiovasc Surg 2022; 37: 439–446.

55.

Barboriak

Van Horn

Pintar

, et al. Scanning electron microscope study of human veins and aorta-coronary artery vein grafts. J Thorac Cardiovasc Surg 1976; 71: 673–679.

56.

Che

Jiang

Liu

, et al. Preliminary report of a new type of braided vein stent in animals. Phlebology 2019; 34: 246–256.

57.

Finn

Nakazawa

Joner

, et al. Vascular responses to drug eluting stents: Importance of delayed healing. Arterioscler Thromb Vasc Biol 2007; 27: 1500–1510.

58.

Joner

Finn

Farb

, et al. Pathology of drug-eluting stents in humans. J Am Coll Cardiol 2006; 48: 193–202.

59.

Bourantas

Torii

, et al. Local hemodynamic forces after stenting: Implications on restenosis and thrombosis. Arterioscler Thromb Vasc Biol 2017; 37: 2231–2242.

60.

Plass

Sabdyusheva-Litschauer

Bernhart

, et al. Time course of endothelium-dependent and -independent coronary vasomotor response to coronary balloons and stents. Comparison of plain and drug-eluting balloons and stents. JACC Cardiovasc Interv 2012; 5: 741–751.

61.

Kipshidze

Dangas

Tsapenko

, et al. Role of the endothelium in modulating neointimal formation. J Am Coll Cardiol 2004; 44: 733–739.

62.

Singhrao

Nicholson

Crean

Informed choices for challenging specimens when choosing methacrylate resin systems for histology. Microsc Res Tech 2012; 75: 576–585.

63.

Virmani

Drug eluting stents: Are human and animal studies comparable?

Heart 2003; 89: 133–138.

64.

Gao

Chaher

Serralha

, et al. Molecular MRI of collagen enables evaluation of fibrosis and therapeutic response in venous thrombosis. Circ Cardiovasc Imaging 2026; 19: e018784.

65.

Hara

Truelove

Tawakol

, et al. ¹⁸ F-fluorodeoxyglucose positron emission tomography/computed tomography enables the detection of recurrent same-site deep vein thrombosis by illuminating recently formed, neutrophil-rich thrombus. Circulation 2014; 130: 1044–1052.

66.

Marston

Chinubhai

Kao

, et al. In vivo evaluation of safety and performance of a nitinol venous stent in an ovine iliac venous model. J Vasc Surg Venous Lymphat Disord 2016; 4: 73–79.

67.

Liao

Abramowitz

Choi

, et al. Lower extremity woven and nonwoven venous stent morphology and luminal changes. J Vasc Surg Venous Lymphat Disord 2024; 12: 101893.

68.

Teng

Zhu

, et al. A novel glycyrrhizin acid-coated stent reduces neointimal formation in a rabbit iliac artery model. Front Pharmacol 2023; 14: 1159779.

69.

Murase

Suzuki

Yamaguchi

, et al. The relationship between re-endothelialization and endothelial function after DES implantation: Comparison between paclitaxcel eluting stent and zotarolims eluting stent. Catheter Cardiovasc Interv 2014; 83: 412–417.

70.

Gori

Polimeni

Indolfi

, et al. Predictors of stent thrombosis and their implications for clinical practice. Nat Rev Cardiol 2019; 16: 243–256.

71.

Diletti

Garcia-Garcia

Bourantas

, et al. Clinical and angiographic outcomes following first-in-man implantation of a novel thin-strut low-profile fixed-wire stent: The Svelte Coronary Stent Integrated Delivery System first-in-man trial. EuroIntervention 2013; 9: 125–134.

72.

Lin

Zhang

, et al. In vivo degradation and endothelialization of an iron bioresorbable scaffold. Bioact Mater 2021; 6: 1028–1039.

73.

Hahn

Lee

Hyun

, et al. Faster smooth muscle cell coverage in ultrathin-strut drug-eluting stent leads to earlier re-endothelialization. Front Bioeng Biotechnol 2023; 11: 1207858.

74.

Jinnouchi

Otsuka

Sato

, et al. Healthy strut coverage after coronary stent implantation: An ex vivo human autopsy study. Circ Cardiovasc Interv 2020; 13: e008869.

75.

Zun

Narracott

Chiastra

, et al. Location-specific comparison between a 3D in-stent restenosis model and micro-CT and histology data from porcine in vivo experiments. Cardiovasc Eng Tech 2019; 10: 568–582.

76.

Musick

Coffey

Irazoqui

PP.

Sensor to detect endothelialization on an active coronary stent. Biomed Eng Online 2010; 9: 67.

77.

Sheridan

Wetterling

Testani

, et al. Safety and performance of a novel implantable sensor in the inferior vena cava under acute and chronic intravascular volume modulation. Eur J Heart Fail 2023; 25: 754–763.

78.

Zhou

Bian

, et al. In vivo evaluation of safety and performance of a tapered nitinol venous stent with inclined proximal end in an ovine iliac venous model. Sci Rep 2024; 14: 7669.

79.

Oliver

Cortese

Ognard

, et al. Magnetic capture of autologous mesenchymal stem cells promotes the rapid endothelialization of peripheral venous stents in rabbits. Acta Biomater 2026; 211: 188–203.

80.

Cornelissen

Guo

Fernandez

, et al. Endothelial recovery in bare metal stents and drug-eluting stents on a single-cell level. Arterioscler Thromb Vasc Biol 2021; 41: 2277–2292.

81.

Jia

Cui

Song

Application of spatial omics in the cardiovascular system. Research (Wash D C) 2025; 8: 0628.

82.

Lorentzen

Hansen

Iversen

, et al. Proteomic characterization of atherosclerotic lesions in situ using percutaneous coronary intervention angioplasty balloons—Brief report. Arterioscler Thromb Vasc Biol 2022; 42: 857–864.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.08 MB