Sage Journals: Discover world-class research

Abstract

Background: Interventional therapies to relieve chronic deep vein thrombosis (DVT) fail through inability to penetrate, cross, and remove the occlusion. Development of suitable tools requires fundamental understanding of chronic DVT mechanical properties and a reliable model for testing. Methods: Female farm swine underwent a novel, endovenous generation of long-segment unilateral iliac vein thrombosis. Thrombus was confirmed via venogram, intravascular ultrasound, and transabdominal duplex for 14 days. Thrombus components were quantified via histology. Thrombus mechanical properties were assessed via uniaxial compression. Results: Among seven swine, technical success was 100%. Compared to subacute thrombi (7-day), chronic thrombi (14-day) showed organizing thrombus with diffuse myointimal thickening and collagen matrix formation on histology. The thrombi collagen content was 41% versus 55% (p = 0.17) and the thrombus erythrocyte percentage was 4.3% versus 2.2%, p = 0.21 in 7- versus 14-day thrombi, respectively. The onset point (compression required to load the thrombus fiber network) was 66.6% versus 35.3% (p = 0.004), the secant modulus (resistance to deformation) measured at the onset point was 153.8 versus 275.99 kPa (p = 0.18), and the average shear constant (resistance to shearing), as defined by the Yeoh hyperelastic model, was 1.85 kPa versus 2.85 kPa in 7- versus 14-day thrombi. Conclusions: This study demonstrates the feasibility of an endovenous model generating chronic unilateral venous thrombi in 2 weeks with similar anatomy to humans and provides critical mechanical properties of thrombi for future research.

Keywords

animal model deep vein thrombosis (DVT)endovascular therapy venous thromboembolism (VTE)vascular biology intravascular ultrasound (IVUS)

Background

Postthrombotic syndrome (PTS), a consequence of untreated venous hypertension, is characterized by leg swelling, pain, heaviness, and skin ulceration. This syndrome ultimately affects 40–60% of deep vein thrombosis (DVT) patients: 7–16% of whom develop severe PTS, with the highest incidence in patients with iliofemoral obstruction.^1
–3 Relieving venous outflow obstruction has been shown to decrease the incidence of severe PTS in patients with acute iliofemoral DVT.^2,4,5 Over time, the acute thrombus matures into an adherent, collagen-rich structure, occasionally causing persistent outflow obstruction.⁶ No medical therapies exist for these chronic luminal changes. Recanalization and stenting of chronic DVTs is effective at relieving symptoms of PTS by relieving venous hypertension.⁷ The technical success of such procedures is significantly limited due to (a) an inability to cross chronic thrombi to allow for recanalization and stenting, and (b) poor stent patency rates when compared with nonthrombotic venous stenting. These suboptimal results are thought to be driven in part by a lack of specific, suitable endovascular tools.⁸

To date, all thrombectomy and thrombus crossing devices have been designed for acute/subacute thrombi.⁹ Owing to the complex mechanical interactions between endovascular devices and organizing thrombus, devices effective in the setting of acute thrombus do not translate for application to chronic thrombi. Depending on the mechanism of action of the device, the thrombus may experience tensile, compressive, and shear forces in a time-dependent manner. Optimizing critical design features that govern these interactions requires a fundamental understanding of the mechanical properties of chronic DVT. Owing to the complex biologic response involved in thrombus organization and maturation,¹⁰ in vitro models of thrombus are limited by the inability to adequately reproduce DVT composition and adherence to the vein wall.

The vital need for a reliable, endovascular, unilateral, subacute, and chronic DVT preclinical model was recently highlighted by the American Heart Association’s venous thromboembolism priorities¹¹ and the Society of Interventional Radiology Foundation’s Research Consensus Panel on Deep Venous Disease.^12,13 Herein we characterize the mechanical and histopathologic properties of subacute and chronic, proximal DVT in a novel pig model utilizing components of stasis, endothelial injury, and hypercoagulability to generate unilateral long-segment iliac thrombosis, generating histologic and imaging features consistent with chronic thrombus. We describe, for the first time in a large animal model, variations in mechanical properties of subacute compared to chronic thrombus.

Methods

Study design

Time points for acute (2 hours), subacute (7 days), and chronic (14 days) DVT were determined a priori based on previous literature in porcine DVT models.¹³ Previous studies demonstrate that thrombolytic efficacy drops from 86% at 3 days to 11% at 14 days post-DVT with advanced signs of thrombus organization and abundant collagen deposition 12–15 days post-DVT.^14,15 Intravascular ultrasound (IVUS; Philips Vision 0.035 catheter), duplex, and venography were performed at the time of thrombosis. In accordance with our standard quality protocols, a midpoint duplex ultrasound was performed 4 days postthrombosis to identify early recanalization or dissolution of thrombi.

Animals

This study was approved by the University of Michigan Institutional Animal Care and Use Committee. Female farm swine weighing 60–80 kg underwent induction and maintenance general anesthesia with 1.5–3.0% isoflurane and 100% oxygen. All operations were performed under sterile conditions. Swine were placed supine on an endovascular surgery table and the following anatomic landmarks identified with duplex ultrasound: bilateral common femoral veins, iliac veins, inferior vena cava (IVC) confluence, and right external jugular (EJ). In swine, the EJ is larger than the internal jugular vein (IJ). The right EJ and common femoral vein (CFV) were cannulated with a 6F and 8F vascular sheath, respectively (Figure 1A). The sheaths were positioned in bilateral iliac veins and a diagnostic venogram was performed to identify the IVC confluence. IVUS confirmed the origin of the hypogastric vein and identified any large iliac tributaries. A vascular plug (9-MVP; Medtronic, USA) was used to occlude the origin of the ipsilateral hypogastric and a high-riding profunda femoris vein if present (Figure 1B). A 12-mm Wallstent (Boston Scientific, Marlborough, MA, USA) was partially deployed on the sterile field, frayed using a 19-gauge needle to unravel the ends of the stent for 15 mm, and recaptured into the sheath on the back table. The stent was placed into the iliac vein and partially deployed (Figure 1C). It was passed central to caudal four times to excoriate the endothelium, then resheathed and withdrawn. The iliac segment was isolated with 11.5-mm diameter Berenstein occlusion balloons (Boston Scientific). Thrombin (100 units/mL, 400 units total; Sigma-Aldrich, St Louis, MO, USA) was delivered in divided doses into the isolated segment via the balloon catheters (Figure 1D). A duplex ultrasound was performed at 1 hour and 2 hours. The balloons were deflated and vascular sheaths withdrawn. The animal was awoken from anesthesia and allowed to ambulate freely. DVT duration was: 0 days (n = 1), 7 days (n = 3), and 14 days (n = 3). The single day-0 animal was performed for proof of concept of the model. The 7-day and 14-day animals represented proposed subacute and chronic thrombi, respectively. Duplex ultrasound scans were performed on day 0, day 4, and day of harvest. Additionally, on the day of harvest, a venogram was performed. Technical success was defined as persistent partially or totally occlusive iliac thrombus at the prespecified time points. At the conclusion of the study, animals underwent a laparotomy and sternotomy. The pulmonary arteries were examined for pulmonary embolism. The iliac veins and IVC were isolated and immediately processed for mechanical testing or histopathology.

Figure 1.

Totally endovascular unilateral iliac vein thrombosis model. (A) External jugular and femoral vein access obtained to isolate the iliac vein segment. Balloons (B) are placed proximally and distally, and a plug (P) is inserted in the hypogastric vein. (B) The hypogastric vein is selected (carat) and the plug placed into the vein for occlusion. Occlusion confirmed with venography (inset). (C) The IV is excoriated with a partially deployed Wallstent to induce endothelial injury. (D) Balloons are inflated for 2 hours to induce venous stasis in the prepared iliac vein segment and thrombin is injected to induce thrombosis.

Human thrombus samples

The use of human samples was approved by the Institutional Review Board (IRB) under IRB HUM00107111. Consent was waived by the IRB, as the samples were part of routine clinical care. Human iliofemoral DVT samples were obtained from samples biopsied during open or endovenous treatment. They were fixed and processed in a similar fashion to the porcine samples, as described below.

Histopathology

En bloc iliac vein and thrombus samples were formalin fixed for 48 hours. The distal half of each thrombus was used for histologic evaluation. The specimens were paraffin-embedded, sectioned, and stained with standard hematoxylin and eosin (H&E), Masson’s trichrome, and Martius Scarlet Blue (MSB). Slides were analyzed by a cardiovascular pathologist blinded to the animal treatment and duration of thrombosis. The following nomenclature was used for characterization: fresh thrombus – intact blood elements without signs of degeneration, consistent with 0–2 days and inclusive of periprocedural thrombus; organizing thrombus – fibrin material with cellular infiltrate inclusive of leukocytes and myointimal cells; myointimal cell ingrowth – fibroblast/smooth muscle cells and collagen replacing the thrombus material in the lumen; hemosiderin deposition – presence of iron pigmentation within macrophages or extracellular matrix; diffuse intimal thickening – numerous myointimal cells with surrounding collagen and extracellular matrix material (e.g., proteoglycan accumulation) expanding the intima; may include leukocytes and/or form continuum with organizing thrombus. The vein wall and thrombus characteristics were summarized using a semiquantitative scoring system as follows: (–) denotes feature is not present, (+) denotes minor histologic feature, (++) denotes histologic feature present in majority of samples, and (+++) denotes histologic feature present throughout all samples. Thrombi were characterized and aged according to standard pathologic criteria applied to human hematoma and thrombus analysis by a cardiovascular pathologist.^16
–21 Thrombus composition, including collagen and erythrocytes, was quantified using Image J software.

Mechanical testing

Thrombus stiffness, representing the resistance of the sample to deform under an applied load, was evaluated using a TA Instruments RSA-G2 solids analyzer (TA Instruments, New Castle, DE, USA) with a force resolution of 10 µN. Samples of harvested thrombi were stored in phosphate buffered saline (PBS) at 4°C prior to mechanical testing, which occurred immediately after necropsy. The thrombus was freed from the vein wall and sectioned to a length of approximately 5 mm. The sample was subjected to uniaxial compression loading for 20 seconds of 0.025 N preload force, followed by unconfined compression with a displacement rate of 0.025 mm/s. The test was terminated when the force exceeded 15 N.²² Deformation of the sample was portrayed in an engineering stress–strain curve. Engineering stress was defined as force divided by the initial cross-sectional area and the engineering strain was defined as the current sample length minus the initial sample length divided by the initial sample length. The onset point was defined as the intersection of the lines tangential to the initial and final point on the stress–strain curve to quantify the differences in the inflection point of the curves between the two timepoints. Tangent lines were calculated using equation: $m = y_{2} - y_{1} / x_{2} - x_{1}$ and the onset point (x) was calculated using equation: $m_{i n i t i a l} * x + b_{i n i t i a l} = m_{f i n a l} * x + b_{f i n a l}$ , where m is slope, y is stress, x is strain, and b is y-intercept. The secant modulus at the onset point was utilized to understand the differences in the thrombi’s deformation resistance. The secant modulus was calculated using equation: $y_{o n s e t p o i n t} - y_{i n i t i a l} / x_{o n s e t p o i n t} - x_{i n i t i a l} .$ (Figure 2). Regression of the stress–strain curves was performed to approximate strain-based zero points for the data prior to averaging and model analysis, allowing for more appropriate approximation of mechanical properties. The stress–strain data from 7-day and 14-day swine was then averaged to create mean curves to represent the compressive behavior of the two DVT models. The Yeoh hyperelastic model was found to accurately fit the data and thus used as the constitutive model.²³ The material was assumed incompressible, leading to a strain energy density function of:

W = \sum_{i = 1}^{3} C_{i 0} {(I_{1} - 3)}^{i};

where $C_{i 0}$ are material constants and $I_{1}$ is the first invariant of the Cauchy–Green deformation tensor.

Figure 2.

Graphic representation of the onset point and secant modulus. The onset point was defined as the intersection of the lines tangential to the initial and final point on the stress strain curve to elucidate differences in the inflection points of the curves between the two timepoints. The thrombi’s deformation resistance was quantified with secant modulus at the onset point.

Statistical analysis

Continuous variables were evaluated with a Student’s t-test, where appropriate. Outliers were determined using Grubbs’ method at α = 0.1. Statistical analysis was performed in GraphPad Prism 10.0.2 and mechanical testing data were analyzed in Hyperfit, Minitab, Matlab R2019b, Microsoft Excel, and TA Instruments TRIOS V4.4.1.41651.

Results

Model

Among severn female farm swine undergoing this novel, totally endovascular unilateral DVT model, technical success was 100% in DVT creation as confirmed via duplex ultrasound. No animals suffered main branch pulmonary embolism. The average animal weight was 72 ± 6.7 kg. Perimeter-derived external iliac vein areas measured by IVUS were 68.5 ± 10.7 mm². The average length of iliac thrombus generated at the index procedure was 7.1 ± 1.4 cm. A total of three animals needed adjunct procedures to achieve complete venous occlusion during the index procedure, including: two requiring endovascular plugs to occlude high-riding profunda femoris veins and one requiring additional thrombin and balloon occlusion time due to nonocclusive thrombi at 2 hours.

Venographic, duplex, and gross histology characteristics demonstrated changes consistent with human DVT evolution over 14 days. At day 0, there was no evidence of venous collaterals, whereas hypertrophy of multiple pathways was evident by day 7, and robust collateralization noted by day 14 (Figure 3). Acute thrombi were consistently identified by large, noncompressible veins on duplex ultrasound. By day 14, the thrombosed veins met human diagnostic criteria for chronic DVT, including contracted and hyperechoic appearance.²⁴ Only day-0 and day-7 thrombi could be easily separated from the adjacent vein wall. By day 14, the thrombi had formed thick, collagen-rich material quite adherent at the vessel interface (Figure 3). One animal was maintained 21 days postthrombosis. This animal demonstrated near-complete recanalization of the thrombosed iliac vein via duplex ultrasound, venography, and IVUS (Supplemental Figures 1–3).

Figure 3.

Representative venography, duplex, and gross histology images postthrombosis. Acute (day 0) DVT, subacute (day 7) DVT, and chronic (day 14) DVT: typical imaging and gross pathology findings associated with acute, subacute, and chronic thrombus (top to bottom): normal venogram; 7-day venogram with noticeable development of collateral venous pathways around obstructing thrombus; and 14-day venogram with extensive, robust collateralization. White arrows denote the occluded area. Duplex ultrasonography, demonstrating progression of acute thrombus in the dilated vein to contracted, chronic thrombus over time. Gross pathology of days 0, 7, and 14 thrombi.

Histopathology

Day 0: Compared to normal (contralateral) iliac vein (Figure 4A, B), the postthrombotic vein wall (Figure 4C, D) was found to have occasional prominent neutrophils (black arrowhead) and focal endothelial denudation. Experimental thrombus (Figure 4E) was fibrin (white star) and erythrocyte rich with scattered, trapped leukocytes (black arrowhead), mirroring features seen in acute human thrombus of one and a half weeks in age (Figure 4F).

Figure 4.

Representative human and porcine histology throughout the chronicity spectrum demonstrating vein wall and thrombus changes over time. (A, B) Normal porcine vein wall architecture in H&E and Masson’s trichrome, respectively, featuring hobnailed endothelium (black arrowheads) and the classical vascular layers. (C, D) Porcine vein wall 2 hours postthrombosis in H&E and Masson’s trichrome, respectively, featuring acute inflammatory infiltration (black arrowhead), adjacent to intraluminal thrombus. (E, F) Porcine (2 hours) and human (~1.5 weeks of symptoms) thrombi, respectively, demonstrating scattered trapped leukocytes (black arrowheads) and fibrin deposition (white stars). (G) H&E of porcine thrombus–vein wall interface after 7 days, demonstrating organizing thrombus with chronic inflammatory changes (black arrowhead) in the vein wall. (H) Trichrome stained, 7-day-old porcine thrombus and vein wall demonstrating DIT (between black arrowheads) adjacent to the organizing thrombus (white star). (I, J) Trichrome staining of porcine (7 days) and human (~2 months) thrombi, respectively, demonstrating organizing thrombus with mixed collagen and fibrin. (K) H&E staining of porcine vein wall 14 days postthrombosis with hemosiderin deposit noted by black arrowheads. (L) Trichrome staining of porcine vein wall 14 days postthrombosis demonstrating DIT adjacent to recanalized vein lumen. (M, N) Trichrome staining of porcine (14 days) and human (~3 years) thrombi, respectively, demonstrating organized thrombus with dense collagen and scattered recanalization channels.

Day 7: Inflammatory cell infiltrate was noted at the thrombus–vein wall interface, and multinucleate giant cells were noted in focal areas of chronic inflammation (Figure 4G). Diffuse fibrosis and intimal thickening were observed in the vein wall adjacent to the thrombus in all specimens (Figure 4H). In all cases, a central thrombus appears fresher, < 2 days in age. Marked mesenchymal ingrowth and collagen deposition were noted in some cases. The organizing thrombus (Figure 4I) bears resemblance to human thrombus of approximately 2 months in age (Figure 4J).

Day 14: Focal hemosiderin deposits were noted within the vein wall–thrombus interface (Figure 4K). Diffuse intimal thickening was a feature of the vein wall regardless of the location of the thrombus (Figure 4L). Recanalization channels (Figure 4M) and foci of chronic inflammation were common within the thrombi. The dense collagen seen in the 14-day experimental thrombus was akin to human comparator thrombus of 3 years in age (Figure 4N).

Analysis of histopathology demonstrated features consistent with chronic thrombus at day 14 with myointimal cell growth, organized thrombus, and chronic inflammatory cells (Figure 4). Similarly, vein wall fibrosis and diffuse intimal thickening, characteristics of late vein wall injury,²¹ were present in 14-day animals but not 7-day animals. These findings are summarized in Table 1. Based on histologic analysis of MSB-stained thrombi (Figure 5A), the collagen content was 41% in 7-day compared to 55% in 14-day thrombi (p = 0.17), and the thrombus erythrocyte percentage was 4.3% in 7-day compared to 2.2% in 14-day thrombi (p = 0.21) (Figure 5B).

Table 1.

Characteristics of thrombus and vein wall in representative histology slides from days 0, 7, and 14 samples.

	Day 0	Day 7	Day 14
Thrombus
Acute inflammatory cells	+	+	–
Chronic inflammatory cells	–	++	+++
Fresh thrombus	+++	++	+
Organizing thrombus	–	+++	++
Myointimal cell ingrowth	–	+	++
Collagenous matrix	–	+	++
Hemosiderin deposition	–	–	++
Vein wall
Acute inflammatory cells	++	+	–
Chronic inflammatory cells	–	+	++
Fibrosis	–	+	+++
Diffuse intimal thickening	–	+	+++
Hemosiderin deposition	–	–	+

–, Feature is not present; +, minor histologic feature; ++, histologic feature present in majority of samples; +++, histologic feature present throughout all samples.

Figure 5.

Representative histology of thrombi with quantification of thrombi erythrocytes and collagen. (A) Histology demonstrating increasing organization of thrombus represented by increasing collagen content and decreasing erythrocyte content (Martius Scarlet Blue: yellow = erythrocytes, blue = collagen). (B) Quantification of thrombus erythrocytes and collagen via Image J. The collagen content was 41% in 7-day compared to 55% in 14-day thrombi (p = 0.17). The thrombus erythrocyte percentage was 4.3% in 7-day compared to 2.2% in 14-day thrombi (p = 0.21).

Mechanical testing

During uniaxial compression loading, the onset point, which represents the amount of compression required to load the thrombus fiber network, was significantly higher in 7-day compared to 14-day thrombi, 66.6% versus 35.3%, p = 0.005 (Figure 6). The secant modulus was measured at the onset point and was 153.8 kPa in 7-day thrombi and 275.99 kPa in 14-day thrombi, p = 0.18. The Yeoh hyperelastic material model fits, defines, and predicts the mechanical response to the uniaxial compressive testing of the 7-day and 14-day thrombi. For best fit approximation, outliers from the data set were disregarded. Mean stress–strain curves were created through 80% strain for 7-day thrombi and 44% strain for 14-day thrombi based on the remaining data, due to force-based limits applied to the mechanical testing. The material constants for these fits can be found in Figure 7. The model and experimental data showed considerable agreement with R² values of 0.987 and 0.976 for 7-day and 14-day thrombi, respectively. Yeoh hyperelastic model analysis of the mean stress–strain curves demonstrate a substantial increase in 2C₁₀ in 14-day thrombi, which represents the shear modulus of the thrombi (1.85 kPa vs 2.85 kPa).

Figure 6.

Mechanical characterization of day 7 and day 14 thrombi. (A) Stress–strain curves demonstrating increased stiffness in day 14 thrombi when compared to day 7 thrombi. (B) Quantitative comparisons of secant modulus and onset point. The secant modulus was measured at the onset point and was higher in 14-day thrombi (153.8 kPa vs 275.99 kPa, p = 0.18). The onset point was significantly higher in 7-day compared to 14-day thrombi (66.6% vs 35.3%, p = 0.005), representative of the increased amount of compression required to load the thrombus fiber network in newer thrombi.

Figure 7.

Mean stress–strain and Yeoh hyperelastic model curve fits with calculated Yeoh material constants. The increased stiffness of 14-day thrombi compared to 7-day thrombi is demonstrated by a leftward shift of the mean stress–strain curves and substantially larger C₁₀ in the 14-day Yeoh fit.

Discussion

The results of this study demonstrate a novel porcine model of subacute and chronic, unilateral, DVT producing thrombi with ultrasound, venographic, and histopathological characteristics resembling those of human samples.²¹ Current treatment paradigms indicate catheter-directed therapies in acute iliofemoral DVTs to reduce PTS severity; however, no specific, FDA-approved, chronic DVT crossing device is available.^25,26 In this model, analysis of 7- and 14-day histopathology suggests the development of representative thrombi for preclinical testing of novel endovascular devices to address chronic DVT.

Current models of subacute and chronic DVT are limited by inadequate vessel diameter for testing of endovascular devices, residual foreign material, and prolonged time (⩾ 3 weeks) to develop representative chronic DVTs.¹³ Through endothelial injury and temporary unilateral iliac vein balloon occlusion with thrombin injection, the proposed model produced consistent, long-segment DVT in large vessels without these limitations. At 14 days, gross and microscopic histopathology of the thrombi were consistent with chronic DVT.²⁷ Key features contributing to the increased complexity of catheter-based treatment of chronic DVTs – increased collagen content and vessel wall adherence – were successfully produced in this model.²⁸ Unilateral, large animal DVT model development has been pursued in an attempt to produce a model most representative of clinical scenarios, particularly to employ a vessel diameter appropriate for novel endovascular device testing. As recently as 2022, Schwein et al. described a totally endovenous model of cavoiliac DVT through balloon occlusion and thrombin injection.²⁹ This produced consistent cavoiliac thrombus, but occlusion of bilateral iliac veins eliminates the contralateral iliac vein as an internal control. Additionally, Shi et al. describe the need for external iliac vein occlusion to reduce collateralization to the thrombosed segment.³⁰ These models required open suture ligation, rendering them unsuitable for future endovascular testing due to complete interruption of the iliac vein. Through abrasive endothelial injury and external iliac vein endovascular plug, the model presently described incorporates the advantages of each model, while avoiding open surgery or caval thrombi. Methods employed elsewhere in preclinical vascular surgery models (i.e., Fogarty catheter, Edwards Lifesciences, Irvine, CA endothelial denudation and coil embolization of the external iliac vein) were not trialed herein but would be expected to be suitable alternatives.

In developing endovascular devices for crossing chronic DVTs, understanding and quantification of the mechanical properties imparted by increasing chronicity of thrombi are critical. Until this point, mechanical characterization of chronic DVTs has been limited to ultrasound elastography. Though noninvasive, there are limitations to ultrasound elastography. Namely, the calculation of Young’s modulus requires measurement of applied force. As the force is manually applied through the ultrasound transducer and various intervening tissues attenuate the applied force, it is not possible to accurately measure the force applied to the tissue of interest. As such, it is not possible to compute Young’s modulus or other mechanical characteristics reliant on applied force. Given the variations in measurement, some assert that measurements produced by ultrasound elastography may only be compared as binary measurements (softer or harder) when completed in the same study with the same transducer.³¹

In this study, we observed that the strain required to significantly load the fiber network of the thrombus in compression was lower in 14-day thrombi when compared to 7-day thrombi. This suggests that the fiber network contained within the 14-day thrombus is denser and/or less permeable to fluid. The compressive stiffness (secant modulus) of the thrombus was 1.8 times higher in 14-day thrombi when compared to 7-day thrombi. Yeoh model fits for 7-day and 14-day thrombi are presented. These nonlinear, hyperelastic representations of the experimental data characterize the mechanical response of the material under compressive loads and can be utilized in modeling efforts to develop devices that can cross or deploy within a chronic DVT to overcome the shear and compressive resistance of the thrombus.

The thrombi produced by this model exhibit a significant increase in stiffness/mechanical properties compared with acute thrombi and thrombus analogues previously reported in the literature.^32,33 The increase in compressive mechanical properties seen in this study validates the root of existing difficulties in addressing chronic DVTs with devices designed for acute DVTs. To overcome these observed mechanical changes, catheter or wire properties must be adjusted to reinforce tip design (i.e., stiffness and/or geometry) for targeting chronic DVT penetration while allowing flexibility for maneuvering the venous system. Similarly, mechanical thrombectomy catheter retrieval elements (i.e., stentrievers) must have adequate radial force to overcome the stiffness of chronic DVT. These observations may allow accurate ex vivo model design of subacute and chronic thrombi for high throughput preclinical testing of devices with varying mechanical characteristics.

Limitations

Though this study demonstrates a critical step forward on the path to endovascular devices designed specifically for chronic DVTs, there are several limiting factors. Foremost, though the thrombi and data were reproducible, the number of swine in each group was low. One 7-day thrombus was excluded from histologic analysis by a blinded vascular histopathologist. The thrombi demonstrated peripheral areas of chronicity with central acute thrombus, suggestive of perimortem thrombosis of a nonoccluded segment of vein. There are inherent limitations in validation of the model, as no parallel study exists on the mechanical characterization of human chronic DVTs. The mechanical testing performed utilized a force-based preload of the samples. However, due to the biologic variability of the samples, varying thrombi diameters create variable starting points in the strains of the samples. Regression of the stress–strain curves was performed to approximate strain-based zero points for the data prior to model analysis, which does create room for error in the data, but allows for more appropriate approximation of the mechanical properties. Despite this, the study provides mechanical properties to correlate with clinically available testing, allowing for improved diagnosis and management of chronic DVTs.

Conclusions

This study describes a newly developed, totally percutaneous model for generation of chronic thrombus in 2 weeks. The model demonstrates low morbidity and cost, and represents similar venographic and histopathologic characteristics to human correlates. Therein, the success of the model provided the opportunity for the first published mechanical characterization of chronic thrombus, quantifying increases in compressive stiffness related to histologic increases in collagen content. This clinically representative model and properties allow for targeted endovascular device development for crossing and retrieving chronic thrombi, ultimately preventing postthrombotic syndrome. Future efforts will be directed at the correlation of findings to clinically available testing such as ultrasound elastography and creation of ex vivo models to capture the mechanical properties of in vivo DVT more accurately for utilization in device design.

Supplemental Material

sj-pdf-1-vmj-10.1177_1358863X241308481 – Supplemental material for Variation in compressive mechanical properties between subacute and chronic venous thrombosis in a novel unilateral iliac thrombosis model

Supplemental material, sj-pdf-1-vmj-10.1177_1358863X241308481 for Variation in compressive mechanical properties between subacute and chronic venous thrombosis in a novel unilateral iliac thrombosis model by Nathaniel J Parchment, Christian Argenti, Sriganesh B Sharma, Oscar Moreno, David Gordon, Laura Durham, Veronica Forsythe, David M Williams, Thomas W Wakefield, Nathan Weidenhamer, Daniel Durant Myers, Minhaj S Khaja, Peter K Henke, Albert J Shih and Andrea Tara Obi in Vascular Medicine

Footnotes

Acknowledgements

The authors wish to thank Andy Poli (University of Michigan) for assistance with the technical aspects of mechanical testing and data analysis, and Angela Hawley (University of Michigan) for assistance with histologic processing. Partial data were presented at the virtual 2020 American Venous Forum. A commentary by Li and Jaffer accompanies this article.³⁴

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

This work was partially funded by a scientific grant from Medtronic, Inc. This funding was used for supplies and swine. The analysis and interpretation were completed independently of Medtronic, Inc. The final manuscript was prepared by the authors and was sent to Medtronic, Inc. for review prior to submission for publication.

ORCID iDs

Nathaniel J Parchment

Christian Argenti

David Gordon

Supplemental material

Supplemental material for this article is available online.

References

Notten

Ten Cate-Hoek

Arnoldussen

, et al. Ultrasound-accelerated catheter-directed thrombolysis versus anticoagulation for the prevention of post-thrombotic syndrome (CAVA): A single-blind, multicentre, randomised trial. Lancet Haematol 2020; 7: e40–e49.

Vedantham

Goldhaber

Julian

, et al. Pharmacomechanical catheter-directed thrombolysis for deep-vein thrombosis. N Engl J Med 2017; 377: 2240–2252.

Heit

JA.

Epidemiology of venous thromboembolism. Nat Rev Cardiol 2015; 12: 464–474.

Setacci

Benevento

de Donato

, et al. Acute deep vein thrombosis and pulmonary embolism: Is the thromboaspiration device an appropriate choice? Transl Med UniSa 2020; 21: 38–46.

Kohi

Kohlbrenner

Kolli

, et al. Catheter directed interventions for acute deep vein thrombosis. Cardiovasc Diagn Ther 2016; 6: 599–611.

Obi

Barnes

GD.

Continuing to advance the venous agenda: Long-term insights from the CAVA Trial. J Am Heart Assoc 2021; 10: e021659.

Wadhwa

Srinivasa

Cooper

, et al. Endovascular therapy for lower extremity chronic deep venous occlusive disease: State of practice. Semin Intervent Radiol 2018; 35: 333–341.

Sudheendra

Vedantham

Catheter-directed therapy options for iliofemoral venous thrombosis. Surg Clin North Am 2018; 98: 255–265.

Shah

Roy

TL.

Catheter-directed interventions for the treatment of lower extremity deep vein thrombosis. Life (Basel) 2022; 12: 1984.

10.

Mukhopadhyay

Johnson

Duru

, et al. Fibrinolysis and inflammation in venous thrombus resolution. Front Immunol 2019; 10: 1348.

11.

Cushman

Barnes

Creager

, et al. Venous thromboembolism research priorities: A scientific statement from the American Heart Association and the International Society on Thrombosis and Haemostasis. Circulation 2020; 142: e85–e94.

12.

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.

13.

Schwein

Magnus

Chakfe

, et al. Critical review of large animal models for central deep venous thrombosis. Eur J Vasc Endovasc Surg 2020; 60: 243–252.

14.

Geier

Muth-Werthmann

Barbera

, et al. Laparoscopic ligation of the infrarenal vena cava in combination with transfemoral thrombin infusion: A new animal model of chronic deep venous thrombosis. Eur J Vasc Endovasc Surg 2005; 29: 542–548.

15.

Lin

Chen

Surowiec

, et al. Evaluation of thrombolysis in a porcine model of chronic deep venous thrombosis: An endovascular model. J Vasc Surg 2001; 33: 621–627.

16.

Butany

Cardiovascular pathology. 4th ed. Elsevier, 2016.

17.

Leu

AJ.

Phlebosclerosis, phlebothrombosis, and thrombophlebitis: A current perspective. Cardiovasc Pathol 1996; 5: 183–192.

18.

Carol

Bernet

Curos

, et al. Thrombus age, clinical presentation, and reperfusion grade in myocardial infarction. Cardiovasc Pathol 2014; 23: 126–130.

19.

Silver

Kawakami

Jolly

, et al. Histopathologic analysis of extracted thrombi from deep venous thrombosis and pulmonary embolism: Mechanisms and timing. Catheter Cardiovasc Interv 2021; 97: 1422–1429.

20.

Fineschi

Turillazzi

Neri

, et al. Histological age determination of venous thrombosis: A neglected forensic task in fatal pulmonary thrombo-embolism. Forensic Sci Int 2009; 186: 22–28.

21.

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.

22.

Chueh

Wakhloo

Hendricks

, et al. Mechanical characterization of thromboemboli in acute ischemic stroke and laboratory embolus analogs. AJNR Am J Neuroradiol 2011; 32: 1237–1244.

23.

Yeoh

OH.

Some forms of the strain-energy function for rubber. Rubber Chem Technol 1993; 66: 754–771.

24.

Needleman

Cronan

Lilly

, et al. Ultrasound for lower extremity deep venous thrombosis: Multidisciplinary recommendations from the Society of Radiologists in Ultrasound Consensus Conference. Circulation 2018; 137: 1505–1515.

25.

Vedantham

Weinberg

Desai

, et al. Society of Interventional Radiology position statement on the management of chronic iliofemoral venous obstruction with endovascular placement of metallic stents. J Vasc Interv Radiol 2023; 34: 1643–1657.e6.

26.

Comerota

Kearon

, et al. Endovascular thrombus removal for acute iliofemoral deep vein thrombosis. Circulation 2019; 139: 1162–1173.

27.

Saha

Humphries

Modarai

, et al. Leukocytes and the natural history of deep vein thrombosis: Current concepts and future directions. Arterioscler Thromb Vasc Biol 2011; 31: 506–512.

28.

Czaplicki

Albadawi

Partovi

, et al. Can thrombus age guide thrombolytic therapy? Cardiovasc Diagn Ther 2017; 7: S186–S196.

29.

Schwein

Magnus

Markovits

, et al. Endovascular porcine model of iliocaval venous thrombosis. Eur J Vasc Endovasc Surg 2022; 63: 623–630.

30.

Shi

, et al. Two swine models of iliac vein occlusion: Which form most contributes to venous thrombosis? Thromb Res 2015; 135: 1172–1178.

31.

Franchi-Abella

Elie

Correas

JM.

Ultrasound elastography: Advantages, limitations and artefacts of the different techniques from a study on a phantom. Diagn Interv Imaging 2013; 94: 497–501.

32.

Tashiro

Shobayashi

Ota

, et al. Finite element analysis of blood clots based on the nonlinear visco-hyperelastic model. Biophys J 2021; 120: 4547–4556.

33.

Johnson

McCarthy

Gilvarry

, et al. Investigating the mechanical behavior of clot analogues through experimental and computational analysis. Ann Biomed Eng 2021; 49: 420–431.

34.

Jaffer

. A novel swine model of chronic deep vein thrombosis (DVT) suitable for endovascular device assessment. Vasc Med 2025; 30: 136-137.

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

0.31 MB