Biomarkers in Relation to Patency, Popliteal Reflux, and Post-thrombotic Syndrome: A Subanalysis of the Ultrasound-accelerated Catheter-directed Thrombolysis versus Anticoagulation for the Prevention of Post-thrombotic Syndrome Trial

Abstract

Background

Despite restored patency, catheter-directed thrombolysis (CDT) has variable efficacy in preventing post-thrombotic syndrome (PTS); biomarkers may clarify PTS pathophysiology and guide patient selection for CDT.

Objectives

To investigate relationships between biomarkers, patency, popliteal reflux, and PTS.

Methods

This prespecified CAVA trial subanalysis included patients with first acute iliofemoral deep vein thrombosis (DVT), randomized to standard treatment (ST) or ultrasound accelerated CDT (UACDT). Baseline blood samples were analyzed for fibrinogen, CRP, IL-6, IL-10, VEGF-A, P-selectin, E-selectin, ICAM-1, VCAM-1, MMP-2, MMP-9, and adiponectin. Patency and reflux (duplex ultrasound), and PTS (Villalta score) were assessed at 1-year and long-term follow-up (LT).

Results

Among 108 patients (51 UACDT, 57 ST), absence of patency at 1-year was associated with higher baseline CRP and fibrinogen in both groups, and elevated IL-6 and VEGF-A in the ST group. Reflux at LT was associated with lower IL-6 and adiponectin in the UACDT group. (Moderate-to-severe) PTS at LT was associated with higher baseline MMP-2 and lower IL-10 in the UACDT group, and lower baseline VCAM-1 and adiponectin in the ST group.

Conclusion

Pro-inflammatory processes are linked to reduced patency, with UACDT improving patency in patients with enhanced inflammatory responses. LT reflux is associated with impaired vasoprotective properties. PTS involves impaired anti-inflammatory responses and tissue remodelling both not modifiable by UACDT. Therefore, biomarker-guided treatment selection may potentially improve treatment outcome.

Keywords

peripheral catherization - post-thrombotic syndrome - biomarkers/blood - vascular patency - venous thrombosis

Introduction

Post-thrombotic syndrome (PTS), a chronic complication of deep vein thrombosis (DVT), significantly impacts quality of life and imposes substantial healthcare costs.[1] [2] The current recommended strategy for PTS prevention involves anticoagulation combined with elastic stockings.[3] However, despite advancements in anticoagulation therapy, 30 to 50% of DVT patients still develop PTS, with twice the risk in cases with iliofemoral DVT.[4] [5]

The exact pathogenesis of PTS remains incompletely understood. It is hypothesized that PTS arises from venous hypertension caused by impaired vascular patency and valvular reflux due to impaired thrombus resolution.[6] Several studies have investigated the value of additional catheter-directed thrombolysis (CDT) in preventing PTS.[7] [8] [9] [10] In these CDT trials, it was anticipated that rapid thrombus removal through CDT, in addition to standard treatment, would reduce the risk of PTS. However, despite the effectiveness of CDT in restoring patency and preventing reflux,[11] [12] [13] results on PTS prevention have been less consistent.[14] Both a better understanding of the pathogenesis of PTS and improved patient selection are probably essential to further improve treatment to reduce PTS burden.[4] Besides thrombus age, the effectiveness of CDT is partly determined by fibrin clot properties, such as density and permeability, which are affected by various processes, including coagulation and inflammation, promoting the formation of denser thrombi that are more resistant to lysis.[15]

A previous sub-study of the Ultrasound-Accelerated Catheter-Directed Thrombolysis Versus Anticoagulation for the Prevention of Post-Thrombotic Syndrome (CAVA) trial demonstrated that levels of fibrinogen are highly correlated with clot properties and were found to be a strong predictor of treatment success following ultrasound accelerated CDT (UACDT) in those with higher fibrinogen levels at baseline.[16] Additionally, older thrombi have been shown to dissolve less effectively and at a slower rate compared with fresher thrombi.[17] Biomarkers may therefore play a critical role in the selection of patients most likely to benefit from additional CDT.

To better understand the etiology of PTS, its underlying risk factors, and the role of CDT herein, this sub-study of the CAVA trial investigates a panel of literature-supported biomarkers and outcomes, in relation to patency, popliteal reflux, and PTS.

Methods

Study Design and Participants

For this prespecified sub-study, patients were included from the multicentre, randomized, single-blinded CAVA trial who had available blood samples at baseline. The CAVA trial was designed to evaluate whether adding UACDT to standard treatment prevents PTS in patients with acute iliofemoral DVT more effectively than standard treatment (ST) alone.[9] [10] Patients were eligible for the CAVA trial if they were 18 to 85 years old, had a objectively documented first-time iliofemoral DVT (i.e., complete or partial thrombosis of the common femoral vein or more cranial vein segments), had acute symptoms lasting no longer than 14 days, a life expectancy of more than 6 months, and no previous thrombosis in the affected limb. Patients were excluded if they were known to have varicosities or pre-existing chronic venous insufficiency (Clinical, Etiological, Anatomical, and Pathophysiological [CEAP] classification C3 or higher); history of gastrointestinal bleeding, cerebrovascular accident, or central nerve system disease within 1 year; severe hypertension (systolic blood pressure >180 mm Hg or diastolic blood pressure > 100 mm Hg); active malignancy (metastatic, progressive, or treated within the previous 6 months); increased alanine transaminase levels (more than three times the upper limit of normal [34 international units (IU)/L for women and 45 IU/L for men]); renal failure (estimated glomerular filtration rate <30 mL/min); had major surgery within 6 weeks; pregnancy; or impaired mobility. For patients with bilateral thrombosis the leg with the most cranial localization was considered as the index leg. Enrolment took place at 15 hospitals in the Netherlands. The study was approved by the institutional review boards of all 15 centres, and all patients provided written informed consent before randomization.

Blood Collection

Venous blood was drawn from the antecubital vein at two time points: at baseline (after study inclusion and initiation of anticoagulation, but before UACDT in patients randomized to the UACDT group) and at 1-year follow-up. Blood was drawn using citrate (3.2% w/v) polypropylene tubes (Becton Dickinson Vacutainer) and free-flowing blood. Within 30 minutes of collection, samples were processed into platelet-poor plasma. Centrifugation was performed in two steps: first, for 5 minutes at 2,500 g (room temperature), then for 10 minutes at 10,000 g (18°C). Plasma samples were stored at –80°C and analyzed in a single batch after study completion.

Selected Biomarkers and Laboratory Assays

This sub-study assesses biomarkers ([Table 1]) related to coagulation and fibrinolysis (fibrinogen[16]), inflammation (pro-inflammatory interleukin [IL]-6, and CRP and anti-inflammatory IL-10[18] [19] [20] [21] [22]), tissue remodelling (vascular endothelial growth factor A [VEGF-A], matrix metalloproteinase [MMP]-2, and MMP-9[23] [24]), adhesion/endothelial function (P-selectin, E-selectin, vascular cell adhesion protein 1 [VCAM-1], and intercellular adhesion molecule 1 [ICAM-1][19] [22] [25] [26]). Furthermore, adiponectin is included in the panel, given its previously demonstrated association with PTS and its known anti-inflammatory and vasoprotective properties.[27] [28] The MMP-9/MMP-2 ratio was calculated as a potential marker of thrombus age: MMP-9 remains elevated during the entire tissue remodelling process, whereas MMP-2 increases predominantly in later stages, making a decreasing ratio indicative of older thrombi.[29]

Table 1 Overview of selected biomarkers

Biological process/function	Related biomarker	Literature based on:
Coagulation and fibrinolysis	Fibrinogen	Iding et al, 2024[16]
Inflammation	CRP	Bouman et al, 2012[18]; Siudut et al, 2020[20]; Polak et al, 2019[21]; Rabinovich et al, 2015[22]
	IL-6	Shbaklo et al, 2009[19]; Siudut et al, 2020[20]; Rabinovich et al, 2015[22]
	IL-10
Tissue remodelling	MMP-2	de Franciscis et al, 2016[23]
	MMP-9
	VEGF-A	Ranero et al, 2023[24]
Adhesion and endothelial function	P-selectin	Iding et al, 2023[25]
	E-selectin	Bittar et al, 2020[26]
	ICAM-1	Shbaklo et al, 2009[19]; Rabinovich et al, 2015[22]
	VCAM-1
Others	Adiponectin	Mrozinska et al, 2018[27]; Ekmekci et al, 2006[28]

Fibrinogen levels were measured by the Clauss method, and CRP levels were determined by nephelometry (Siemens) with correction for citrate dilution. The remaining markers (IL-6, IL-10, MMP-2, MMP-9, VEGF-α, P-selectin, E-selectin, ICAM-1, VCAM-1, and adiponectin) were measured in citrate-treated samples using a multiplex assay (ProcartaPlex, Thermo Fisher), following the manufacturer's instructions. All assays were performed by laboratory personnel blinded to participant data associated with the samples.

Vein Assessment

Extended duplex ultrasound (DUS) was performed at two time points: at 1 year after enrolment and at long-term follow-up. Trained vascular technicians, blinded to treatment allocation, performed the DUS examination using a standardized protocol. Different duplex machines were used depending on availability at each interventional centre. During the DUS examination, patients were placed in supine position to assess the inferior vena cava to the tibial venous confluence in the affected leg using a convex array transducer. In standing position, venous morphology and valve competence were assessed from the groin to below the knee using either a high-frequency linear array transducer or a pulsed-wave Doppler. Vein patency was assessed by compressing each vein segment in the transverse plane to check for obstruction, focusing on the common iliac, external iliac, common femoral, femoral, and popliteal veins. For iliac vessels, diameters were compared with the contralateral side, with the healthy vein diameter set to 100% to calculate the affected to healthy ratio. Patency was defined as ≥90% venous compressibility and the presence of flow in all venous segments. Popliteal reflux was assessed using colour and pulsed wave Doppler to determine flow and flow direction in both transverse and longitudinal planes. Flow augmentation and venous dilatation were stimulated through crural compression, controlled breathing (guided inhalation and exhalation commands), the Valsalva manoeuvre, weight shift, and dorsiflexion of the foot. Reflux was defined as reverse flow lasting >0.5 seconds.

Post-thrombotic Syndrome

PTS was diagnosed using the Villalta score, according to the consensus definition of the International Society on Thrombosis and Haemostasis (ISTH).[30] The Villalta score assesses five symptoms (pain, cramps, heaviness, paraesthesia, and pruritus) and six clinical signs (pretibial oedema, skin induration, hyperpigmentation, redness, venous ectasia, and pain upon calf compression). Each symptom and sign are rated on a severity scale: 0 = absent, 1 = mild, 2 = moderate, and 3 = severe. PTS was defined as a Villalta score of ≥5 or the presence of venous ulceration at the 6-month, 1-year, or long-term follow-up. Moderate-to-severe PTS was defined as a Villalta score of ≥10 or the presence of venous ulceration.

Statistical Analysis

Continuous variables, including all biomarker levels, are presented as medians with interquartile ranges (median [IQR]), while categorical variables are reported as counts and percentages (n (%)). Pearson correlation coefficients were used to assess correlations between baseline biomarker levels.

Analyses of biomarker levels in relation to patency, reflux, and PTS were stratified by treatment group, as CDT has been shown to significantly improve both patency and popliteal reflux as well as long-term PTS compared with standard treatment alone.[11] [12] [13] Group comparisons of baseline biomarker levels in relation to patency, popliteal reflux, and (moderate-to-severe) PTS, at 1-year and long-term follow-up, were performed using the non-parametric Mann-Whitney U test. Statistical significance was defined as a two-sided p-value of ≤0.05, and no adjustments were made for multiple comparisons. Correlations were calculated in R (version 4.2.0) using the ggcorrplot package, all other analyses were performed in IBM SPSS Statistics (version 28).

Results

Study Population

A total of 108 patients were enrolled in this study: 51 in the UACDT group and 57 in the standard treatment group. The median duration of symptoms at the time of blood draw was 10 days (7–14) in the UACDT group and 13 days (7–18) in the standard treatment group. Median long-term follow-up period was 38 (12–63) months in the UACDT group and 39 (14–65) months in the standard treatment group. In both groups, the minimum long-term follow-up period was 12 months. Demographic and clinical characteristics at baseline were similar between groups and are presented in [Table 2]. Baseline blood samples were analyzed for all included patients ([Fig. 1]).

Fig. 1 Flowchart of study population.

Table 2 Baseline characteristics

	Additional UACDT (n = 51)	Standard treatment (n = 57)
Age
Median (years)	51.0 [38.0–67.0]	52.0 [37.5–63.5]
Sex ( n (%))
Female	22 (43.1)	27 (47.4)
Body mass index
Median (kg/m2)	27.8 [24.0–30.8]	27.0 [24.2–29.4]
Provoked DVT ( n (%))
Provoked DVT	26 (51.0)	24 (42.1)
Risk factors ( n (%))
One	20 (39.2)	15 (26.3)
More than one	6 (11.8)	9 (15.8)
Surgery in previous 2 months	7 (13.7)	7 (12.3)
Trauma in previous 2 months	1 (2.0)	3 (5.3)
Immobility	13 (25.5)	15 (26.3)
Pregnancy or childbirth in previous 3 months	5 (9.8)	3 (5.3)
Hormonal substitution	1 (2.0)	0 (0.0)
Oral anticonception	6 (11.8)	90 (15.8)
Previous contralateral DVT	8 (15.7)	5 (8.8)
Previous pulmonary embolism	2 (3.9)	5 (8.8)
Active malignancy	2 (3.9)	1 (1.8)
Thrombus location and characteristics ( n (%))
Left extremity	32 (62.7)	39 (68.4)
Right extremity	17 (33.3)	16 (28.1)
Bilateral	2 (3.9)	2 (3.5)
Popliteal vein not involved	17 (36.2)	22 (38.6)
Popliteal vein involved	30 (63.8)	35 (61.4)
Procedural characteristics ( n (%))
May-Turner syndrome	19 (37.3)	24 (42.1)
Stent placement UACDT	20 (39.2)	−
Post-procedural stent placement <year	6 (12.2)	0 (0.0)
Post-procedural stent placement >year and <LT	3 (6.1)	7 (13.5)

Abbreviations: LT, long-term follow-up; UACDT, ultrasound-accelerated catheter-directed thrombolysis.

Note: Categorical data are presented as number (%). Age, body mass index, and thrombus load are expressed as median [IQR]; baseline thrombus characteristics were determined by DUS, in the absence of DUS by MRV, and in the absence of both DUS and MRV by venography.

Correlation of Baseline Biomarker Levels

Several correlations were identified between baseline biomarker levels. Strong positive correlations were observed between fibrinogen and CRP, between IL-10 and IL-6, and between VCAM-1 and adiponectin. Additionally, a moderately negative correlation was found between fibrinogen and MMP-2 ([Fig. 2]).

Fig. 2 Correlations between baseline biomarker levels. Pearson correlation coefficients were used to assess correlations between baseline biomarker levels.

Baseline Biomarker Levels in Relation to Patency

At 1-year follow-up, associations between baseline biomarker levels and patency were similar for both UACDT and standard treatment. Baseline levels of fibrinogen, CRP, and VEGF-A were lower in patients who achieved patency at 1 year. MMP-9/MMP-2 ratio was significantly lower in patients who regained patency in the UACDT group only ([Table 3]).

Table 3 Association of biomarkers at baseline and patency at 1 year stratified for treatment

Biomarker	UACDT			Standard treatment
	Patency (n = 26) Median [IQR]	No patency (n = 17) Median [IQR]	p-value	Patency (n = 12) Median [IQR]	No patency (n = 36) Median [IQR]	p-value
Coagulation and fibrinolysis
Fibrinogen (g/L)	4.25 [3.33–5.13]	5.10 [4.10–6.65]	0.045	4.45 [3.38–5.18]	5.15 [4.23–6.10]	0.048
Inflammation
CRP (mg/L)	12.21 [4.02–26.45]	27.06 [12.71–76.47]	0.020	6.69 [2.18–17.56]	15.14 [5.33–59.95]	0.054
IL-6 (pg/mL)	0.51 [0.21–1.17]	1.35 [0.28–2.95]	0.214	0.49 [0.22–0.92]	1.73 [0.60–2.73]	0.004
IL-10 (pg/mL)	0.64 [0.28–1.59]	0.84 [0.38–1.80]	0.555	0.57 [0.30–1.81]	1.05 [0.49–1.34]	0.468
Tissue remodelling
MMP-2 (103 pg/mL)	5.72 [3.57–12.34]	3.64 [1.62–6.94]	0.065	3.50 [2.28–9.51]	4.00 [1.98–7.39]	0.924
MMP-9 (103 pg/mL)	2.10 [1.45–3.49]	2.24 [1.68–7.13]	0.249	2.55 [1.71–3.94]	3.93 [2.13–6.70]	0.199
MMP-9/MMP-2	0.35 [0.17–1.02]	0.76 [0.38–4.49]	0.039	0.70 [0.30–1.39]	1.26 [0.29–2.46]	0.263
VEGF-A (pg/mL)	17.42 [9.58–33.70]	44.21 [16.41–83.09]	0.059	22.53 [15.87–27.45]	29.45 [19.61–41.73]	0.038
Adhesion and endothelial function
P-selectin (103 pg/mL)	2.70 [1.90–4.56]	2.92 [1.55–4.68]	0.710	2.67 [1.85–3.92]	2.58 [1.58–6.07]	0.686
E-selectin (103 pg/mL)	4.41 [2.74–6.01]	2.94 [2.34–5.76]	0.377	3.88 [2.16–6.31]	4.33 [2.82–5.30]	0.617
ICAM-1 (104 pg/mL)	8.14 [5.92–12.71]	8.91 [5.83–15.26]	0.838	9.09 [7.82–15.07]	9.38 [6.47–16.50]	0.915
VCAM-1 (105 pg/mL)	1.44 [0.47–2.05]	1.29 [0.71–1.78]	0.828	1.42 [0.70–1.98]	1.29 [0.89–2.37]	0.757
Others
Adiponectin (106 pg/mL)	4.15 [1.13–29.48]	2.51 [1.22–15.67]	0.691	3.97 [1.75–30.02]	4.55 [2.05–15.51]	0.884

Abbreviation: UACDT, ultrasound-accelerated catheter-directed thrombolysis.

Notes: Differences in baseline biomarker levels were compared using a non-parametric Mann-Whitney U test. Patients were divided into groups based on achievement of patency in all vein segments; patency defined as >90% compressibility in presence of flow.

At long-term follow-up, CRP levels were significantly lower in patients with patency in the standard treatment group (3.26 [0.94–6.78] versus 20.26 [6.72–56.29]; p = 0.002) ([Fig. 3]). There were no associations with any of the other baseline biomarkers and patency in either treatment group ([Supplementary Table S1], available in the online version only).

Fig. 3 Association of biomarkers at baseline and patency at 1 year and in the long term.

Baseline Biomarker Levels in Relation to Popliteal Reflux

At 1-year follow-up, popliteal reflux was not associated with any of the baseline biomarkers in either of the treatment groups ([Supplementary Table S2], available in the online version only). At long-term follow-up, however, in patients with popliteal reflux after UACDT both baseline IL-6 (0.60 [0.12–0.92] versus 1.17 [0.49–2.33]; p = 0.023) and adiponectin (0.92 [0.73–15.01] versus 6.31 [1.55–24.79]; p = 0.05) were significantly lower ([Fig. 4]). In the standard treatment group, none of the baseline biomarkers were associated with long-term popliteal reflux ([Supplementary Table S3], available in the online version only).

Fig. 4 Association of biomarkers at baseline and popliteal reflux at 1 year and in the long term.

Baseline Biomarker Levels in Relation to PTS

At 1-year follow-up baseline CRP levels were significantly lower in patients in the UACDT group who developed PTS compared with patients without PTS. Furthermore, MMP-2 levels were significantly higher in patients with PTS after UACDT compared with patients without PTS. There were no associations of baseline biomarker levels and (moderate-to-severe) PTS at 1 year in the standard treatment group ([Table 4]).

Table 4 Association of baseline biomarker levels and (moderate–severe) PTS at 1 year stratified for treatment

Biomarker	UACDT					Standard treatment
	No PTS (n = 28) Median [IQR]	All PTS (n = 23) Median [IQR]	p-value All PTS versus No PTS	M.-s. PTS (n = 8) Median [IQR]	p-value M.-s. PTS versus No PTS	No PTS (n = 29) Median [IQR]	All PTS (n = 28) Median [IQR]	p-value All PTS versus No PTS	M.-s. PTS (n = 16) Median [IQR]	p-value M.-s. PTS versus No PTS
Coagulation and fibrinolysis
Fibrinogen (g/L)	4.50 [3.93–6.38]	4.30 [3.70–6.00]	0.358	4.15 [3.55–6.13]	0.537	4.70 [4.25–5.95]	4.90 [3.73–5.70]	0.854	5.40 [3.90–5.70]	0.722
Inflammation
CRP (mg/L)	23.72 [9.83–78.14]	11.11 [4.56–25.49]	0.045	13.87 [5.43–37.98]	0.284	12.01 [5.38–44.14]	15.34 [5.34–44.70]	0.762	15.34 [4.69–60.15]	0.831
IL-6 (pg/mL)	0.92 [0.28–2.20]	0.57 [0.28–1.20]	0.333	0.97 [0.24–2.16]	0.955	1.09 [0.43–2.49]	1.49 [0.47–2.81]	0.528	1.22 [0.66–2.81]	0.374
IL-10 (pg/mL)	1.05 [0.33–1.73]	0.40 [0.31–0.83]	0.104	0.38 [0.16–0.40]	0.091	1.05 [0.46–1.77]	1.12 [0.45–1.44]	0.892	1.06 [0.44–1.52]	0.991
Tissue remodelling
MMP-2 (103 pg/mL)	3.69 [1.69–5.56]	6.14 [3.97–11.77]	0.022	6.81 [4.36–13.07]	0.070	3.69 [1.94–5.74]	4.18 [2.17–13.96]	0.235	5.54 [2.21–13.96]	0.223
MMP-9 (103 pg/mL)	2.38 [1.49–5.40]	2.03 [1.57–3.42]	0.674	1.92 [1.80–2.59]	0.896	2.94 [1.45–4.72]	4.32 [2.03–7.64]	0.177	2.97 [1.62–6.37]	0.696
MMP-9/MMP-2 (pg/mL)	0.66 [0.31–2.93]	0.49 [0.20–1.23]	0.226	0.34 [0.21–1.03]	0.180	0.81 [0.27–1.69]	1.01 [0.25–1.59]	0.948	0.81 [0.23–1.59]	0.643
VEGF-A (pg/mL)	26.43 [13.32–57.90]	17.84 [10.51–33.30]	0.132	18.50 [13.21–57.24]	0.576	25.79 [16.94–29.71]	31.94 [17.52–57.07]	0.193	31.94 [16.33–47.93]	0.275
Adhesion and endothelial function
P-selectin (103 pg/mL)	2.92 [1.27–5.67]	2.25 [2.00–3.87]	0.845	3.31 [2.07–3.59]	0.723	2.82 [1.34–5.30]	2.58 [1.66–5.79]	0.981	2.46 [1.71–5.02]	0.887
E-selectin (103 pg/mL)	3.56 [2.19–6.29]	4.89 [3.11–8.52]	0.218	4.62 [2.84–8.24]	0.399	3.85 [2.26–5.46]	4.30 [2.51–5.82]	0.690	4.66 [2.62–5.82]	0.507
ICAM-1 (104 pg/mL)	8.17 [5.57–14.52]	8.37 [6.73–12.30]	0.860	7.95 [6.51–14.80]	0.837	8.20 [6.06–10.57]	11.23 [6.82–16.50]	0.082	12.53 [6.49–21.54]	0.200
VCAM-1 (105 pg/mL)	1.24 [0.56–1.87]	1.67 [0.42–2.41]	0.446	2.10 [1.53–2.69]	0.064	1.28 [0.91–2.21]	1.18 [0.69–1.92]	0.576	1.04 [0.69–1.84]	0.393
Others
Adiponectin (106 pg/mL)	3.41 [0.99–26.50]	7.17 [1.32–16.17]	0.573	17.34 [2.72–60.22]	0.143	3.33 [1.84–23.41]	3.28 [1.04–15.19]	0.724	2.29 [1.04–15.19]	0.511

Abbreviations: PTS, post-thrombotic syndrome; UACDT, ultrasound-accelerated catheter-directed thrombolysis.

Notes: Differences in baseline biomarker levels were compared using a non-parametric Mann-Whitney U test. Patients were divided into groups based on PTS diagnosis: all PTS if Villalta score ≥5 or venous ulcers present, moderate–severe (m.-s.) PTS if Villalta score ≥10 or venous ulcer present.

At long-term follow-up MMP-2 levels were significantly higher in patients with PTS (6.02 [3.73–12.75] versus 3.64 [1.72–4.62]; p = 0.013) and moderate-to-severe PTS (6.61 [3.94–13.33]; p = 0.013) in the UACDT group. Additionally, baseline IL-10 levels were significantly lower in patients with long-term PTS (0.38 [0.29–0.74] versus 1.15 [0.67–1.93]; p = 0.003) and moderate-to-severe PTS (0.36 [0.19–0.70]; p = 0.003) after UACDT. In the standard treatment group, baseline VCAM-1 levels were significantly lower in patients with long-term PTS (1.08 [0.64–1.75] versus 1.53 [1.31–2.77]; p = 0.012) and moderate-to-severe PTS (1.04 [0.69–1.84]; p = 0.038) compared with patients without PTS. In addition, in the standard treatment group, baseline adiponectin levels were significantly lower in patients with long-term PTS (2.21 [1.28–10.07] versus 8.01 [2.79–58.43]; p = 0.031) compared with those without PTS ([Fig. 5]). The other baseline biomarkers investigated did not show significant associations with PTS ([Supplementary Table S4], available in the online version only).

Fig. 5 Association biomarkers at baseline and (moderate-to-severe) post-thrombotic syndrome (PTS) at 1 year and in the long term. UACDT, ultrasound-accelerated catheter-directed thrombolysis.

Discussion

This prespecified sub-study of the CAVA trial not only highlights the associations of biomarkers with patency, popliteal reflux, and PTS, but also the changing associations with outcomes depending on the intervention. In both treatment groups, patients who achieved patency had lower baseline levels of fibrinogen and CRP compared with those without patency. Interestingly, among patients with patency in the UACDT group, median CRP levels were substantially higher compared with those with patency in the standard treatment group. Despite these higher CRP levels, the UACDT group had double the patency rate. A similar trend was observed for fibrinogen: while baseline levels were comparable between groups, patency was twice as frequent after UACDT. These findings suggest that UACDT improves patency rates even in the presence of elevated inflammation.

In addition to CRP, a significant difference in pro-inflammatory IL-6 levels was observed in the standard treatment group only. Both CRP and IL-6 have been associated with denser thrombi that are more resistant to lysis.[15] The association of IL-6 may not be observed in patients undergoing UACDT because those with denser thrombi, influenced by IL-6, could still achieve thrombus resolution with the additional assistance of UACDT. Altogether, it suggests that patients with more active inflammation upon presentation with an iliofemoral DVT would benefit most from UACDT.

Additionally, in the UACDT group, the MMP9/MMP2 ratio was significantly lower in the patients who achieved patency. This ratio has been proposed as a potential indicator of thrombus age.[29] MMP-2 and MMP-9 contribute to extracellular matrix degradation, particularly collagen. In mouse models, MMP-2 levels were more elevated in later stages of thrombus resolution, MMP-9 levels increase early in the process and remain elevated throughout.[31] The observed difference in MMP9/MMP2 ratio may suggest that patients with older thrombi are less likely to achieve patency. Furthermore, lower VEGF-A levels were observed in patients with patency in the standard treatment group. VEGF-A facilitates thrombus resolution by stimulating the formation of vascular channels within the resolving thrombus and enhancing macrophage recruitment.[32] Therefore, the elevated VEGF-A levels observed in patients who did not achieve patency may indicate impaired recanalization. Together, thrombus age and clot properties influenced by pro-inflammatory processes appear to be crucial factors in effective thrombus resolution to achieve patency.

No significant associations were found between baseline biomarkers and 1-year reflux in either treatment group, suggesting that early reflux is not primarily driven by the baseline biological processes examined. However, in the UACDT group, long-term reflux was associated with lower levels of both adiponectin and IL-6. Adiponectin is a well-known adipokine and modulator for endothelial adhesion molecules, with anti-inflammatory and vasoprotective properties.[33] [34] These findings suggest that reduced adiponectin may impair the vasoprotective response, thereby increasing the risk of reflux development.

The association between lower IL-6 levels and increased long-term reflux may seem counterintuitive, especially since higher IL-6 levels were previously linked to reduced patency. However, IL-6 also plays a key role in thrombus resolution by enhancing the expression of proteolytic enzymes via macrophages, which are critical for thrombus remodelling and recanalization.[35] [36] These seemingly contradictory associations may indicate the need for a balanced IL-6 response, given its dual role in promoting thrombus resolution and contributing to prothrombotic and fibrotic processes.

Despite the evidence that CDT is more effective than standard treatment in achieving patency and preventing popliteal reflux,[11] [12] [13] this did not yet translate into better clinical outcomes for PTS. There is no one-to-one relationship between patency achieved and the development of PTS. This is also reflected in the difference in biomarkers associated with patency (including fibrinogen, CRP, IL-6) and PTS (including MMP2 and IL-10), suggesting that other processes are also relevant in PTS development.

In relation to the clinical outcome PTS, the association with MMP-2 was observed only in the UACDT group. Higher baseline MMP-2 levels were associated with PTS both at 1-year and long-term follow-up. The higher MMP-2 levels observed in PTS patients suggest that their thrombi were older at treatment initiation. A post hoc analysis of the CAVA trial demonstrated that chronic thrombi are resistant to thrombolytic therapy, reducing UACDT efficacy. However, in acute and subacute thrombi UACDT was found to be equally effective, although thrombolysis required more time in patients with subacute thrombosis.[17]

Lower baseline levels of the anti-inflammatory cytokine IL-10 were associated with long-term (moderate-to-severe) PTS in patients following UACDT. IL-10 was found to reduce inflammation and thrombus formation in a rodent model of stasis-induced venous thrombosis.[37] However, studies investigating the association between IL-10 and PTS development show mixed results. The BioSOX study identified a significant association of IL-10 (6 months after DVT) with PTS development, whereas other studies did not observe a significant association between IL-10 (3–6 months after DVT) and PTS.[19] [20] The differences in results could be explained by the different timing of blood collection and PTS diagnosis. Notably, in the present study IL-10 was only associated with PTS in the long term, suggesting that patients with a prolonged impaired anti-inflammatory response may be at heightened risk of PTS.

In the standard treatment group, baseline VCAM-1 and adiponectin levels were positively correlated, and both were significantly lower in patients with moderate-to-severe long-term PTS. One previous study examined the association between adiponectin levels 3 months after DVT and the development of PTS 2 years later and found that lower adiponectin levels were predictive of PTS, independent of BMI.[27] This is consistent with our observation of reduced levels of the vasoprotective adiponectin in patients with long-term PTS. We also observed that lower VCAM-1 levels in the acute phase were associated with moderate-to-severe long-term PTS. Two previous studies have examined the association between VCAM-1 and PTS. One reported significantly higher VCAM-1 levels in the chronic phase in PTS patients.[38] The other study found no difference in VCAM-1 levels, measured at 4 months after DVT, between patients with and without PTS.[19] Also, these discrepancies may be explained by timing of blood sampling. The associations with PTS of the anti-inflammatory adiponectin and the linked VCAM-1 levels were not observed in the UACDT group. This may be explained by the faster removal of thrombi, which could attenuate thrombus-related inflammation and diminish the significance of these processes.

There are several limitations to our study that need to be noted. First, this is a subanalysis where patients were selected based on available blood samples, which may have introduced selection bias and, although baseline characteristics appeared to be similar between treatment groups, unobserved differences cannot be excluded. Second, this is an exploratory study, with a limited sample size and the results were not adjusted for other known predictors of PTS such as age, gender, and obesity, nor did we correct for multiple testing. Third, pre-existing reflux at baseline was not objectively assessed by DUS in the contralateral leg. However, the likelihood of pre-existing reflux is expected to be low in this cohort given the relatively young age of participants and the exclusion of patients with CEAP C3 or higher. Fourth, this study focused on biomarkers with specific roles in biological pathways hypothesized to contribute to PTS, patency, and reflux. Biomarkers that have shown potential as clinical predictors of these outcomes but lack specificity for an underlying (patho)physiological process (e.g., D-dimer) were not included.

The main strength of this study is that it clearly demonstrates treatment-dependent biomarker associations with patency as well as with the development of PTS in patients undergoing adjunctive UACDT. Therefore, this study not only provides novel information on biomarkers and their association with different processes in the development of PTS but may also be a step toward a more tailored treatment approach.

In conclusion, pro-inflammatory processes are linked to reduced patency, with UACDT improving patency in patients with enhanced inflammatory responses. Long-term popliteal reflux is associated with impaired vasoprotective properties. PTS involves multiple mechanisms, notably impaired anti-inflammatory responses and tissue remodelling both not modifiable by UACDT. Therefore, biomarker-guided patient selection may potentially improve treatment outcome.

Conflict of Interest

A.J.t.C.-H. reports personal fees from Alveron, Galapagos, Astra Zeneca, Novostia and he is shareholder with Coagulation Profile; all revenues deposited at CARIM for research. The other authors declares that they have no conflict of interest.

Contributors' Statement

R.H., A.I., and A.T.C. performed the data analysis and interpretation; R.H. and A.T.C. wrote the manuscript; A.T.C., H.T.C., and C.H.A.W. contributed to study concept and study design of the CAVA trial. All other authors contributed equally to data collection and review of the manuscript. Authors involved in analyzing the data (R.H., A.I., and A.T.C.) had full access to all of the study data. A.T.C. had final responsibility for the decision to submit for publication. All authors approved the final version for submission.

• References

• 1 Lubberts B, Paulino Pereira NR, Kabrhel C, Kuter DJ, DiGiovanni CW. What is the effect of venous thromboembolism and related complications on patient reported health-related quality of life? A meta-analysis. Thromb Haemost 2016; 116 (03) 417-431
• 2 Grosse SD, Nelson RE, Nyarko KA, Richardson LC, Raskob GE. The economic burden of incident venous thromboembolism in the United States: a review of estimated attributable healthcare costs. Thromb Res 2016; 137: 3-10
• 3 Kakkos SK, Gohel M, Baekgaard N. et al; Esvs Guidelines Committee. Editor's Choice—European Society for Vascular Surgery (ESVS) 2021 clinical practice guidelines on the management of venous thrombosis. Eur J Vasc Endovasc Surg 2021; 61 (01) 9-82
• 4 Hupperetz R, Ten Cate-Hoek A. The postthrombotic syndrome, where are we now?. Vasa 2025; 54 (01) 7-19
• 5 Tick LW, Doggen CJ, Rosendaal FR. et al. Predictors of the post-thrombotic syndrome with non-invasive venous examinations in patients 6 weeks after a first episode of deep vein thrombosis. J Thromb Haemost 2010; 8 (12) 2685-2692
• 6 Phillips II LJ, Sarkar R. Molecular characterization of post-thrombotic syndrome. J Vasc Surg 2007; 45 (Suppl A): A116-A122
• 7 Vedantham S, Goldhaber SZ, Julian JA. et al; ATTRACT Trial Investigators. Pharmacomechanical catheter-directed thrombolysis for deep-vein thrombosis. N Engl J Med 2017; 377 (23) 2240-2252
• 8 Enden T, Haig Y, Kløw NE. et al; CaVenT Study Group. Long-term outcome after additional catheter-directed thrombolysis versus standard treatment for acute iliofemoral deep vein thrombosis (the CaVenT study): a randomised controlled trial. Lancet 2012; 379 (9810) 31-38
• 9 Notten P, Ten Cate-Hoek AJ, Arnoldussen CWKP. 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 (01) e40-e49
• 10 Notten P, de Smet AAEA, Tick LW. et al. CAVA (Ultrasound-Accelerated Catheter-Directed Thrombolysis on Preventing Post-Thrombotic Syndrome) trial: long-term follow-up results. J Am Heart Assoc 2021; 10 (11) e018973
• 11 Hupperetz RD, Iding AFJ, van Laanen J. et al. Patency and reflux in relation to postthrombotic syndrome: a subanalysis of the Ultrasound-Accelerated Catheter-Directed Thrombolysis Versus Anticoagulation for the Prevention of Post-Thrombotic Syndrome trial. J Thromb Haemost 2024; 22 (12) 3552-3561
• 12 Weinberg I, Vedantham S, Salter A. et al; ATTRACT Trial Investigators. Relationships between the use of pharmacomechanical catheter-directed thrombolysis, sonographic findings, and clinical outcomes in patients with acute proximal DVT: results from the ATTRACT multicenter randomized trial. Vasc Med 2019; 24 (05) 442-451
• 13 Enden T, Kløw NE, Sandvik L. et al; CaVenT study group. Catheter-directed thrombolysis vs. anticoagulant therapy alone in deep vein thrombosis: results of an open randomized, controlled trial reporting on short-term patency. J Thromb Haemost 2009; 7 (08) 1268-1275
• 14 Broderick C, Watson L, Armon MP. Thrombolytic strategies versus standard anticoagulation for acute deep vein thrombosis of the lower limb. Cochrane Database Syst Rev 2021; 1 (01) CD002783
• 15 Ząbczyk M, Ariëns RAS, Undas A. Fibrin clot properties in cardiovascular disease: from basic mechanisms to clinical practice. Cardiovasc Res 2023; 119 (01) 94-111
• 16 Iding AFJ, Alkarithi G, Cate HT, Ariëns RAS, Ten Cate-Hoek AJ. Fibrinogen levels and clot properties identify patients who benefit from catheter-directed thrombolysis after DVT. Blood Adv 2024; 8 (11) 2924-2932
• 17 Arnoldussen CWKP, Notten P, Brans R. et al. Clinical impact of assessing thrombus age using magnetic resonance venography prior to catheter-directed thrombolysis. Eur Radiol 2022; 32 (07) 4555-4564
• 18 Bouman AC, Smits JJ, Ten Cate H, Ten Cate-Hoek AJ. Markers of coagulation, fibrinolysis and inflammation in relation to post-thrombotic syndrome. J Thromb Haemost 2012; 10 (08) 1532-1538
• 19 Shbaklo H, Holcroft CA, Kahn SR. Levels of inflammatory markers and the development of the post-thrombotic syndrome. Thromb Haemost 2009; 101 (03) 505-512
• 20 Siudut J, Natorska J, Son M, Plens K, Undas A. Increased levels of histidine-rich glycoprotein are associated with the development of post-thrombotic syndrome. Sci Rep 2020; 10 (01) 14419
• 21 Polak MW, Siudut J, Plens K, Undas A. Prothrombotic clot properties can predict venous ulcers in patients following deep vein thrombosis: a cohort study. J Thromb Thrombolysis 2019; 48 (04) 603-609
• 22 Rabinovich A, Cohen JM, Cushman M. et al. Inflammation markers and their trajectories after deep vein thrombosis in relation to risk of post-thrombotic syndrome. J Thromb Haemost 2015; 13 (03) 398-408
• 23 de Franciscis S, Gallelli L, Amato B. et al. Plasma MMP and TIMP evaluation in patients with deep venous thrombosis: could they have a predictive role in the development of post-thrombotic syndrome?. Int Wound J 2016; 13 (06) 1237-1245
• 24 Ranero S, Silveira G, Trias N. et al. Endothelium-biomarkers for postthrombotic syndrome: a case-control study. Blood Coagul Fibrinolysis 2023; 34 (03) 144-151
• 25 Iding AFJ, Kremers BMM, Nagy M. et al. Translational insights into mechanisms underlying residual venous obstruction and the role of factor XI, P-selectin and GPVI in recurrent venous thromboembolism. Thromb Res 2023; 221: 58-64
• 26 Bittar LF, Silva LQD, Orsi FLA. et al. Increased inflammation and endothelial markers in patients with late severe post-thrombotic syndrome. PLoS One 2020; 15 (01) e0227150
• 27 Mrozinska S, Cieslik J, Broniatowska E, Undas A. Elevated leptin and decreased adiponectin independently predict the post-thrombotic syndrome in obese and non-obese patients. Sci Rep 2018; 8 (01) 6938
• 28 Ekmekci H, Ekmekci OB. The role of adiponectin in atherosclerosis and thrombosis. Clin Appl Thromb Hemost 2006; 12 (02) 163-168
• 29 Henke PK, Nicklas JM, Obi A. Immune cell-mediated venous thrombus resolution. Res Pract Thromb Haemost 2023; 7 (08) 102268
• 30 Kahn SR, Partsch H, Vedantham S, Prandoni P, Kearon C. Subcommittee on Control of Anticoagulation of the Scientific and Standardization Committee of the International Society on Thrombosis and Haemostasis. Definition of post-thrombotic syndrome of the leg for use in clinical investigations: a recommendation for standardization. J Thromb Haemost 2009; 7 (05) 879-883
• 31 Deatrick KB, Eliason JL, Lynch EM. et al. Vein wall remodeling after deep vein thrombosis involves matrix metalloproteinases and late fibrosis in a mouse model. J Vasc Surg 2005; 42 (01) 140-148
• 32 Modarai B, Burnand KG, Humphries J, Waltham M, Smith A. The role of neovascularisation in the resolution of venous thrombus. Thromb Haemost 2005; 93 (05) 801-809
• 33 Ouchi N, Kihara S, Arita Y. et al. Novel modulator for endothelial adhesion molecules: adipocyte-derived plasma protein adiponectin. Circulation 1999; 100 (25) 2473-2476
• 34 Ouedraogo R, Gong Y, Berzins B. et al. Adiponectin deficiency increases leukocyte-endothelium interactions via upregulation of endothelial cell adhesion molecules in vivo. J Clin Invest 2007; 117 (06) 1718-1726
• 35 Nosaka M, Ishida Y, Kimura A. et al. Crucial involvement of IL-6 in thrombus resolution in mice via macrophage recruitment and the induction of proteolytic enzymes. Front Immunol 2020; 10: 3150
• 36 Achyar AC, Hara T, Adinata A. et al. Interleukin-6 enhances localized immune cell infiltration and deep vein thrombosis resolution at the distal edge. Am J Physiol Heart Circ Physiol 2025; 329 (02) H603-H621
• 37 Downing LJ, Strieter RM, Kadell AM. et al. IL-10 regulates thrombus-induced vein wall inflammation and thrombosis. J Immunol 1998; 161 (03) 1471-1476
• 38 Bouman AC, Cheung YW, Spronk HM. et al. Biomarkers for post thrombotic syndrome: a case-control study. Thromb Res 2014; 134 (02) 369-375

Publication History

Received: 29 August 2025

Accepted: 21 January 2026

Accepted Manuscript online:
27 January 2026

Article published online:
10 February 2026