Patients who suffer deep venous thrombosis (DVT) have at least a 30% recurrence rate during 10 years. Current murine models of inferior vena cava (IVC) DVT all model primary thrombosis and not recurrent thrombosis. The aim of this project was to create and characterize the first murine model to study recurrent DVT in the IVC.

Introduction: Deep vein thrombosis (DVT) resolution occurs via an inflammatory response. As most DVT are sterile, endogenous danger signals, such as nucleic or uric acids, must incite the inflammatory response. Neutrophils and monocyte/macrophages are important in DVT resolution; the role of dendritic cells is not well defined. Toll-like receptor 9 (TLR9) is expressed on these cells and activation by foreign or endogenous nucleic acids triggers an innate immune response. Previously, we have shown that loss of TLR9 signaling is associated with impaired venous thrombosis (VT) resolution and increased necrotic cell products in mice. in this study, we investigated whether extracellular nucleic acids are present in experimental VT in mice, and further defined the role of TLR9 and sterile inflammation in early VT resolution. Methods: VT were induced in Balb/c mice by inferior vena cava ligation. at 2 days, thrombi were harvested and histological sections were stained for DNA or RNA. to determine the effects of necrotic cell products on VT resolution, bone marrow-derived macrophages (BMMO) or dendritic cells (BMDC) were exposed to TLR9 agonist (CpG DNA), monosodium urate (MSU), necrotic cells, or RNA in vitro with expression analysis of genes important for VT resolution. to further elucidate the role of TLR9 signaling in thrombus-associated macrophages and dendritic cells, Tlr9+/+ BMMO or BMDC were adoptively transferred to Tlr9-/- mice prior to VT induction, with examination of the effect on early VT resolution. Results: We found that extracellular DNA, but not extracellular RNA, is present in experimental VT in mice. the extracellular DNA is located in proximity to neutrophils in the thrombus, suggesting the presence of neutrophil extracellular traps (NETs). in vitro, BMMO and BMDC responded to TLR9 agonist, MSU, and necrotic cells with changes in expression of genes important for VT resolution (e.g. BMMO responded to MSU with an 11-fold increase in IL-1α, 1.7-fold increase in MMP-9, and 60% decrease in uPA compared with control; P<0.05), but did not respond to RNA. Adoptive transfer of Tlr9+/+ macrophages or DC to Tlr9-/- mice did not improve VT resolution (P=0.31 and 0.69, respectively, for thrombus size when compared with appropriate controls). Conclusions: These findings show that extracellular DNA, but not RNA, is present in experimental VT in mice. While monocytes and DCs are activated by sterile inflammatory products in vitro, these cells did not restore normal VT resolution, suggesting that TLR9 signaling in these cells is not important in early VT resolution.

Post-thrombotic syndrome (PTS) is characterized by a fibrotic vein injury after deep vein thrombosis (DVT), resulting in a less compliant vein wall. We sought to quantify the change in vein wall thickness and to determine if vein wall damage, defined as wall thickening, is worsened in patients when DVT fails to resolve by 6 months, and also whether there were differences in blood or plasma levels of proteins associated with tissue remodeling. Patients presenting with suspected lower extremity DVT were evaluated. Ultrasound imaging of the lower extremity venous system was performed, and blood was collected. Patients with DVT received repeat evaluation with blood draw and ultrasound imaging at 6 months. DVT resolution was assessed using ultrasound examination. The thickness of the vein wall was quantified by ultrasound imaging in each segment affected by thrombus, and a contralateral, unaffected vein wall served as a control. Messenger RNA was extracted from whole blood using the PAXgene system, and serum proteins were analyzed using enzyme-linked immunosorbent assay (ELISA). Analysis of variance or t tests were used. P < .05 was significant. Thirty patients (10 with DVT resolution at 6 months, 10 with persistent thrombus, and 10 healthy controls) were compared. Resolving and nonresolving DVT were both associated with 1.5- to 1.8-fold increased vein wall thickness at 6 months (n = 10-12; P = .008) compared with nonaffected vein wall segments. However, the thickness of the affected segments was 1.4-fold greater in patients who had total resolution of the DVT by 6 months than in patients who had persistent chronic thrombus 6 months after presentation (n = 10-12; P = .01,). There was a four- to fivefold increased level of MMP-9 in all thrombosed groups compared with controls (n = 5; P < .05). Toll-like receptor-9 (TLR-9) expression was threefold less than in controls (n = 5; P < .05). There were no statistically significant differences in the levels of associated factors such as D-dimer, P-selectin, or inflammatory and remodeling markers such as SLC or matrix metalloproteinase (MMP)-2 by ELISA. There were no significant differences in the gene expression of C-reactive protein, MMP-2, MMP-9, or TLR-4. This preliminary study suggests ongoing vein wall remodeling after DVT. At 6 months, the vein wall is markedly thickened, but this change is independent of thrombus resolution and is associated with elevated MMP-9 but not other inflammatory markers. This suggests that the vein wall damage is initiated early after thrombus formation and persists even in the presence of total resolution.

Background: Post-thrombotic syndrome (PTS) is characterized by a fibrotic vein injury following deep vein thrombosis (DVT), resulting in a less compliant vein wall. We sought to quantify the change in vein wall thickness, and to determine if vein wall damage, defined as wall thickening, is worsened in patients who fail to resolve DVT by six months, and whether there were differences in blood or plasma levels of proteins associated with tissue remodelling. Methods: Patients presenting with suspected lower extremity DVT were evaluated. Ultrasound imaging of the lower extremity venous system was performed and blood was collected. Patients with DVT received repeat evaluation with blood draw and ultrasound imaging at six months. DVT resolution was assessed using ultrasound examination. The thickness of the vein wall was quantified by ultrasound imaging in each segment affected by thrombus, and a contralateral, unaffected vein wall served as a control. mRNA was extracted from whole blood using the PAXgene system, and serum proteins were analysed using enzyme-linked immunosorbent assay (ELISA). Analysis of variance or Student’s t-tests were used and a P , 0.05 was significant. Results: Thirty patients (10 patients with DVTresolution at six months, 10 patients with persistent thrombus and 10 healthy controls) were compared. Both resolving and non-resolving DVT were associated with 1.5–1.8-fold increased vein wall thickness at six months (n 1⁄4 10–12; P 1⁄4 0.008) as compared with non-affected vein wall segments. However, the thickness of the affected segments was 1.4-fold greater in patients who had a total resolution of the DVT by six months than in patients who had persistent chronic thrombus six months after presentation (N 1⁄4 10–12; P 1⁄4 0.01). There was a 4–5-fold increased level of matrix metalloproteinase (MMP)-9 in all thrombosed groups compared with controls (n 1⁄4 ; P , 0.05), while Toll-like receptor-9 (TLR-9) expression was threefold less than controls (n 1⁄4 ; P , 0.05). There were no statistically significant differences in the levels of associated factors such as D-dimer, P-selectin or inflammatory and remodelling markers such as SLC or MMP-2 by ELISA. There were no significant differences in the gene expression of C-reactive protein, MMP-2, MMP-9 or TLR-4. Conclusion: This preliminary study suggests ongoing vein wall remodelling after DVT. At six months, the vein wall is markedly thickened, but this change is independent of thrombus resolution, and associated with elevated MMP-9 but not other inflammatory markers. This suggests that the vein wall damage is initiated early following thrombus formation and persists even in the presence of total resolution.

Objective: Post-thrombotic syndrome (PTS) is clinically characterized by a fibrotic vein wall injury following deep vein thrombosis (DVT). Previous studies have demonstrated exacerbated injury with loss of both matrix metalloproteinase (MMP)-9 and tissue inhibitor of matrix metalloproteinase (TIMP)-1, each associated with an increase in MMP-2 activity. We hypothesized that genetic deletion of MMP-2 would attenuate fibrotic vein wall injury. Methods: DVT was produced in the mouse by ligation of the infrarenal inferior vena cava (IVC) or sham operation in MMP-2 -/- (KO) and c57 (WT) animals, and tissue was harvested at 2, 8, 21 and 42 days. The vein wall tissue was processed for real-time reverse transcriptase-polymerase chain reaction, gelatin zymography (5 per time point), reverse zymography, ELISA and immunohistochemistry. Measurement of collagen deposition and scarring was performed using picrosirius red staining in polarized light. ANOVA was used for multiple comparisons, Student's test was used for direct pair comparisons, and a P < .05 was significant. Results: Thrombus resolution was documented by a decrease in the thrombosed IVC weight from 2 days to 42 days in both groups. At baseline and 2d following DVT formation, vein wall thickness and collagen scores were similar. At 8 days following DVT formation, collagen content and thickness of the vein wall was 32% less in the MMP-2 KO vs WT mice (P= .017). At 21d, the collagen content and thickness of the vein wall was less in the MMP-2 KO, but this difference did not attain significance (P=0.12). Collagen III gene expression was 5.8-fold lower at 8d (P=.009), and collagen I gene expression was 25.7-fold lower at 21d (P=0.034) in MMP-2 KO than in WT. Tropoelastin gene expression was 6.7-fold lower in MMP-2 KO than in WT at 21d (P=.008). There were no significant alterations in expression of MMP-9, although by gelatin zymography, 3 fold more MMP-9 activity was present in MMP-2 KO than WT at 8d (P=.006). TGF-β expression was similar at baseline and at 2d and 8d after DVT creation, but was 4.4 fold higher in MMP -2 KO at 21 days (P=.009). There were no differences in TNFα or MCP-1 levels by ELISA, nor were differences found in vein wall monocyte influx. There were no significant differences in the levels of either TIMP-1 or TIMP-2 at 8, 21, or 42 days in MMP-2 KO as compared with WT. Lastly, no significant differences in cellular proliferation (Ki-67 positive staining) or in apoptosis (TUNEL assay) at 8d were found between groups. Conclusions: In an experimental DVT injury model, loss of MMP-2 activity attenuates the post thrombotic vein wall injury with diminished expression and accumulation of collagen following DVT. These changes are not dependent on alterations in cellular proliferation or apoptosis. Further study will be required to determine if this is a suitable target for reducing PTS.