Abstract
Background
Elastic compression stockings (ECS) are commonly used to prevent postthrombotic syndrome (PTS) after deep vein thrombosis (DVT). However, their efficacy is controversial. This could be due to suboptimal adherence in the negative study. Additional data on the impact of adherence on ECS efficacy are needed.
Objectives
To assess the impact of adherence on ECS efficacy to prevent PTS.
Methods
We conducted a prespecified analysis of the double-blind “Compression Elastique Evaluation du Syndrome post Thrombotique” trial comparing 25 mm Hg ECS with 35 mm Hg ECS to prevent PTS after DVT. The objectives were to 1) assess the proportion of patients with PTS at 2 years according to adherence to ECS; 2) assess predictors of PTS; and 3) build a decision tree algorithm to measure the impact of each independent predictor of PTS.
Results
The proportion of patients with PTS at 2 years who whore the study ECS <50%, 50% to 79%, and 80% of the time were 38.4% (95% CI, 28.8%-48.7%), 24.5% (95% CI, 13.3%-38.9%), and 27.5% (95% CI, 20.0%-36.0%), respectively. Obesity (adjusted odds ratio [aOR], 4.73 [95% CI, 2.13-10.51]), patient height (aOR, 1.07 [95% CI, 1.02-1.11]), and DVT involving 2 to 3 and ≥4 venous segments (aOR, 5.23 [95% CI, 1.11- 24.74] and aOR, 15.59 [95% CI, 3.22-75.43], respectively) increased PTS risk; male sex (aOR, 0.37 [95% CI, 0.16-0.83]), hypertension (aOR, 0.46 [95% CI, 0.21-0.99]), and wearing ECS 50% of the study time (aOR, 0.51 [95% CI, 0.28-0.92]) decreased this risk. Wearing ECS 50% of the study time was most beneficial among nonobese patients with extensive DVT; the proportion with PTS at 2 years was 29.8% vs 54.3% among nonadherent patients.
Conclusion
Adherence to ECS is an independent predictor of PTS that becomes apparent when ECS are worn 50% of the time. Adherence to ECS appears particularly advantageous in nonobese patients with extensive DVT.
Keywords: clinical trial, patient compliance, postthrombotic syndrome, stockings, compression, venous thrombosis
Essentials
-
•
The efficacy of ECS in preventing PTS is uncertain.
-
•
We conducted a prespecified CELEST trial analysis assessing the impact of ECS adherence on PTS risk.
-
•
Adherence to ECS independently predicted PTS, becoming apparent when worn ≥50% of the time.
-
•
Wearing ECS ≥50% for 2 years is most beneficial among nonobese patients with extensive DVT.
1. Introduction
Postthrombotic syndrome (PTS) is a clinical manifestation of chronic venous insufficiency that arises as a long-term complication of deep vein thrombosis (DVT) [1]. Although not life-threatening, PTS is a common condition that can significantly impair patients’ quality of life and impose substantial healthcare costs on both individuals and society [2].
Given the lack of simple, noninvasive, and effective treatments for established PTS, current management primarily focuses on prevention following DVT [3]. Elastic compression stockings (ECSs) are commonly used for this purpose [2,3]. However, their efficacy is controversial. While 2 historical open-label randomized controlled trials (RCTs) reported a 50% reduction in PTS risk with ECS use compared with no ECS, a large double-blind trial found no significant benefit of ECS over placebo [[4], [5], [6]]. These conflicting results may be attributed to a placebo effect in the open-label RCTs or to poor adherence to ECS use in the double-blind RCT [7]. Notably, in the latter trial, only 55.6% of participants were wearing ECS at least 3 days per week at 2 years of follow-up.
To better understand the role of patient adherence in the effectiveness of ECS, further data from a double-blind study are needed. In this context, we analyzed data from the “Compression Elastique Evaluation du Syndrome post Thrombotique” (CELEST) RCT to investigate how adherence to ECS influenced the risk of developing PTS during the trial.
2. Methods
This study is a prespecified analysis of the CELEST RCT. CELEST was a multicenter, double-blind RCT designed to assess whether 25 mm Hg ECS worn for 2 years is noninferior to 35 mm Hg ECS in preventing PTS following a first episode of acute proximal lower extremity DVT. The trial protocol was approved by the South East II Ethics Committee (Lyon, France) in November 2011 (number 2011-032). Written informed consent was obtained from all participating patients. The full study protocol has been previously published [7], and the trial is registered at ClinicalTrials.gov (NCT01578122).
The primary findings of the CELEST trial demonstrated that 25 mm Hg ECS were noninferior to 35 mm Hg ECS in preventing PTS at 2 years (relative risk [RR], 0.93 [95% CI, 0.65-1.33]). Among patients who were optimally adherent to ECS (defined as daily use of the study ECS 80% of the time and a modified GIRERD score of 0-2), the cumulative proportion of PTS was lower in the 35 mm Hg group than in the 25 mm Hg group, although this difference did not reach statistical significance (21.6% vs 29.4%, P = .3) [8].
The objective of this analysis was to evaluate how ECS adherence influences the risk of developing PTS. Specifically, we assessed: i) the proportion of patients with PTS at 2 years, stratified by adherence to ECS use, regardless of assigned treatment group, during the first 3 months and over the full 2-year follow-up period. Adherence was categorized as reasonable, at least reasonable, or optimal if the patient wore the study ECS for 50% to 79%, ≥50%, or ≥80% of the time, respectively, and had a modified GIRERD score of 0 to 2 [7]. Adherence was considered poor if these criteria were not met; ii) independent predictors of PTS development; and iii) the construction of a decision tree algorithm to evaluate the relative impact of each independent predictor of PTS on the proportion of patients who develop PTS (the study’s primary objective). Unlike traditional multivariable models, decision tree algorithms provide a clinically intuitive stratification of patients by translating complex data into simple, rule-based pathways that mirror clinical decision-making. This approach facilitates the identification of specific patient subgroups in which adherence to ECS may have a particularly meaningful impact on the risk of PTS. A key advantage of decision trees is their ability to capture nonlinear relationships and higher-order interactions among risk factors without requiring prior assumptions, thereby offering complementary insights beyond those obtained from conventional regression analyses.
All patients enrolled in the CELEST trial who were evaluated for the primary outcome (ie, PTS at 2 years) were included in the analysis. The primary outcome was PTS at 2 years. According to the International Society on Thrombosis and Haemostasis standards, PTS was considered present if the Villalta score was 5 or if a venous ulcer was present in the leg ipsilateral to the DVT at the 1- or 2-year follow-up visit [1].
Baseline characteristics are presented as counts and percentages for categorical variables and as medians and IQRs for continuous variables.
The proportions of patients with PTS at 2 years, according to ECS adherence during the first 3 months and throughout the 2-year follow-up, are presented as percentages with 95% CIs calculated using the “ci proportion” command in Stata version 17.0 (StataCorp) with the “exact” option.
Independent predictors of PTS were assessed using a backward stepwise logistic regression model. Variables entered into the model included the ECS allocation group and all baseline variables with P values .2 in univariate analysis. These variables are listed in Table 1 and include patient and DVT characteristics, ECS allocation group (25 mm Hg vs 35 mm Hg), anticoagulant type (direct oral anticoagulant [DOAC] vs vitamin K antagonist), and anticoagulation duration (6 months or less vs >6 months). Since the risk of PTS was found to be similar among patients who were reasonably (50%-79%) and optimally (≥80%) adherent to ECS during follow-up, we combined these groups for analysis to increase statistical power. Thus, we used the binary variable “at least reasonable adherence” (≥50%) in our risk factor analysis. To assess the predictive performance of multivariate logistic regression, we calculated the area under the receiver operating characteristic curve.
Table 1.
Baseline characteristics of the 302 patients enrolled in the prespecified study, according to adherence to elastic compression stockings during the 2-year follow-up.
| Overall population |
Poor adherence |
At least reasonable adherencea |
Data missing on adherence |
|
|---|---|---|---|---|
| N = 302 | n = 99 | n = 180 | n = 23 | |
| 25mm Hg ECS group (vs 35 mm Hg group) | 154 (51.0) | 46 (46.5) | 100 (55.6) | 8 (34.8) |
| Baseline characteristics | ||||
| Age ≥ 65 y | 103 (34.1) | 30 (30.3) | 65 (36.1) | 8 (34.8) |
| Male sex | 215 (71.2) | 69 (69.7) | 132 (73.3) | 14 (60.9) |
| Height | 172 (167-178) | 172 (167-180) | 172 (168-178) | 170 (160-178) |
| BMI, kg/m2 | ||||
| <25 | 105 (34.8) | 32 (32.3) | 67 (37.2) | 6 (26.1) |
| 25-30 | 133 (44.0) | 46 (46.5) | 74 (41.1) | 13 (56.5) |
| 30 | 64 (21.2) | 21 (21.2) | 39 (21.7) | 4 (17.4) |
| Previous VTE | 61 (20.3) | 21 (21.4) | 34 (18.9) | 6 (27.3) |
| CEAP C3, leg ipsilateral to the DVT | 103 (34.6) | 31 (31.6) | 60 (33.7) | 12 (54.5) |
| Baseline contralateral VS 5 | 27 (9.0) | 13 (13.1) | 8 (4.5)b | 6 (27.3)b |
| ABI < 0.9 | 5 (1.8) | 2 (2.2) | 3 (1.8) | 0 (0) |
| Hypertension | 70 (23.2) | 21 (21.2) | 43 (23.9) | 6 (26.1) |
| Initial in-hospital management | 225 (74.5) | 65 (65.7) | 141 (78.3) | 19 (82.6) |
| Cancer | 31 (10.3) | 7 (7.1) | 20 (11.1) | 4 (17.4) |
| At least reasonably adherent to ECS | 180 (64.5) | - | - | - |
| DVT characteristics | ||||
| Unprovoked DVT | 184 (62.0) | 67 (67.7) | 105 (58.7) | 12 (63.2) |
| Left leg DVT | 179 (59.3) | 62 (62.6) | 104 (57.8) | 13 (56.5) |
| Time from DVT diagnosis to randomization 3 d | 230 (76.2) | 76 (76.8) | 140 (77.8) | 14 (60.9) |
| Most proximal DVT extent | ||||
| Iliac or common femoral vein | 91 (31.2) | 24 (25.3) | 59 (33.5) | 8 (38.1) |
| Femoral or popliteal vein | 201 (68.8) | 71 (74.7) | 117 (66.5) | 13 (61.9) |
| No. of venous segments involved | ||||
| 1 | 35 (12.0) | 10 (10.5) | 22 (12.5) | 3 (14.3) |
| 2-3 | 162 (55.5) | 55 (57.9) | 96 (54.5) | 11 (52.4) |
| ≥4 | 95 (32.5) | 30 (31.6) | 58 (33.0) | 7 (33.3) |
| Anticoagulant treatment | ||||
| >6 mo | 201 (78.5) | 70 (80.5) | 126 (78.3) | 5 (62.5) |
| Initial treatment with LMWH | 202 (66.9) | 56 (56.6) | 127 (70.6) | 19 (82.6) |
| Long-term treatment with DOAC | 193 (68.2) | 74 (78.7) | 108 (62.4)b | 11 (68.8) |
ABI, ankle brachial index, measured in the leg ipsilateral to the DVT (patients with ABI < 0.6 were excluded from CELEST); BMI, body mass index; CEAP, Clinical, Etiological, Anatomical, and Pathophysiological; DOAC, direct oral anticoagulant; DVT, deep vein thrombosis; ECS, elastic compression stockings; LMWH, low-molecular-weight heparin; VS, Villalta score; VTE, venous thromboembolism.
Defined as the use of ECS 50% of the total study time and a modified GIRERD score of 0 to 2, poorly adherent otherwise.
P < .01.
Data are n (%) or median (IQR). Comparisons are between patients who are poorly adherent vs at least reasonably adherent, and between patients with adherence data (poorly adherent + at least reasonably adherent) vs patients with missing adherence data.
A decision tree model was then developed using the RPART package in R version 3.5.0 (R Foundation for Statistical Computing), employing the C4.5 algorithm for tree decision [9]. In this model, the dependent variable was PTS at 2 years, and the independent variables were all predictors that were significantly associated with PTS in the multivariable logistic regression.
All data were processed and analyzed using Stata version 17.0 (StataCorp). A 2-sided P value <.05 was considered statistically significant. However, comparisons of baseline characteristics and DVT treatment between poorly adherent and at least reasonably adherent patients, and between patients with and without available data on adherence to the study ECS, were considered exploratory; for these comparisons, we report significance only at P < .01.
3. Results and Discussion
Of the 350 patients originally enrolled in the CELEST study, 302 were included in the present analysis. Exclusions were made for patients who were ineligible for the RCT (n = 6), who withdrew consent, or who were withdrawn by the investigator (n = 42). Of the included patients, 154 were randomized to the 25 mm Hg ECS group and 148 to the 35 mm Hg ECS group. Baseline characteristics and details of DVT treatment for the overall study population and for subgroups stratified by ECS adherence are presented in Table 1. Compared with patients with missing data on ECS adherence (n = 23), those with complete data sets were less likely to have a contralateral Villalta score 5 (Table 1). Other baseline characteristics and DVT treatments did not differ significantly between the 2 groups.
In the overall study population, most patients were men (71.2%) and aged <than 65 years (65.9%). In about a third of cases, the most proximal extent of the DVT involved the iliac or the common femoral vein, and the thrombus extended across 4 or more venous segments, as classified by the lower extremity thrombosis US system [10]. Most patients (78.5%) received anticoagulant therapy for at least 6 months, predominantly DOACs. Baseline characteristics and DVT treatment did not differ significantly between ECS adherence groups. However, patients who were poorly adherent to ECS were more likely to have been treated with DOACs compared with those who were at least reasonably adherent (78.7% vs 62.4%).
Overall, 32.1% (95% CI, 26.9%-37.7%) of patients developed PTS at 2 years. The proportions of PTS by ECS adherence during the first 3 months and over the entire 2-year study period are presented in Table 2. Patients who were at least reasonably adherent to ECS were significantly less likely to develop PTS compared with those who were poorly adherent both at 3 months and 2 years. This association remained consistent in multivariable analyses (adjusted odds ratio, 0.51 [95% CI, 0.28-0.92]), as shown alongside the univariate findings in Table 3. The strength and direction of this association are consistent with findings from 2 historical open-label RCTs, which demonstrated that wearing ECS for 2 years reduced the risk of PTS by 50% [4,5]. More recent studies have also reported a lower incidence of PTS among ECS users [[11], [12], [13]]. This provides further, though not definitive, support for a causal role of ECS in modifying the natural course of PTS following DVT. Interestingly, there were no obvious differences in the proportion of patients with PTS at 2 years, whether patients had been reasonably or optimally adherent to ECS during the first 3 months or throughout the study period (29.6% [95% CI, 18.0%-43.6%] vs 29.5% [95% CI, 23.1%-36.5%] and 24.5% [95% CI, 13.3%-38.9%] vs 27.5% [95% CI, 20.0%-36.0%], respectively).
Table 2.
Proportion of postthrombotic syndrome at 2 years according to adherence to elastic compression stockings during the first 3 months and the whole 2-year study follow-up.
| N patients | PTS, n (% [95% CI]) | |
|---|---|---|
| Overall population | 302 | 97 (32.1 [26.9-37.7]) |
| Adherence during the first 3 mo | ||
| Poor | 53 | 24 (45.3 [31.6-59.6]) |
| At least reasonable | 244 | 72 (29.5 [23.9-35.7])a |
| Reasonable | 54 | 16 (29.6 [18.0-43.6]) |
| Optimal | 190 | 56 (29.5 [23.1-36.5]) |
| Adherence during the 2-y follow-up | ||
| Poor | 99 | 38 (38.4 [28.8-48.7]) |
| At least reasonable | 180 | 48 (26.7 [20.4-33.8])a |
| Reasonable | 49 | 12 (24.5 [13.3-38.9]) |
| Optimal | 131 | 36 (27.5 [20.0-36.0]) |
Poor adherence: use of the study elastic compression stockings (ECS) <50% of the time and/or a modified GIRERD score > 2; Reasonable adherence: use of the study ECS 50% to 79% of the time and a modified GIRERD score of 0 to 2; Optimal adherence: use of the study ECS 80% of the time and a modified GIRERD score of 0 to 2; At least reasonable adherence: use of the study ECS 50% of the time and a modified GIRERD score of 0 to 2.
PTS, postthrombotic syndrome.
Comparisons are between patients who are poorly adherent and those who are at least reasonably adherent.
P < .05.
Table 3.
Risk factors of postthrombotic syndrome at 2 years.
| OR (95% CI) | aOR (95% CI)b | |
|---|---|---|
| 25 mm Hg ECS group | 0.76 (0.47-1.24) | 0.82 (0.46-1.48) |
| Baseline characteristics | ||
| Age ≥ 65 y | 0.65 (0.39-1.11) | |
| Male sex | 0.69 (0.41-1.17)c | 0.37 (0.16-0.83) |
| Height | 1.02 (0.99-1.05)c | 1.07 (1.02-1.11) |
| BMI, kg/m2 | ||
| 25-30 | 1.53 (0.85-2.77)c | 1.85 (0.91-3.78) |
| 30 | 4.04 (2.06-7.93)c | 4.73 (2.13-10.51) |
| Previous VTE | 1.50 (0.84-2.70)c | |
| CEAP C3, leg ipsilateral to the DVT | 1.92 (1.16-3.17)c | |
| Baseline contralateral VS ≥ 5 | 1.79 (0.80-3.99)c | |
| Initial in-hospital management | 1.25 (0.71-2.20) | |
| Cancer | 1.01 (0.45-2.23) | |
| ABI < 0.9 | 0.53 (0.06-4.76) | |
| Hypertension | 0.67 (0.37-1.22)c | 0.46 (0.21-0.99) |
| At least reasonably adherent to ECSa | 0.58 (0.35-0.98)c | 0.51 (0.28-0.92) |
| DVT characteristics | ||
| Unprovoked DVT | 1.03 (0.63-1.71) | |
| Left leg DVT | 0.86 (0.52-1.40) | |
| Time from DVT diagnosis to randomization 3 d | 1.37 (0.79-2.39) | |
| Iliac or common femoral DVT | 1.77 (1.06-2.98)c | |
| No. of venous segments involved | ||
| 2-3 | 6.15 (1.42-26.72)c | 5.23 (1.11-24.74) |
| ≥4 | 16.16 (3.67-71.18)c | 15.59 (3.22-75.43) |
| Anticoagulant treatment | ||
| >180 d | 1.02 (0.54-1.94) | |
| Initial treatment with LMWH | 1.01 (0.60-1.68) | |
| Long-term treatment with DOAC | 0.71 (0.42-1.20)c |
ABI, ankle brachial index, measured in the leg ipsilateral to the DVT; aOR, adjusted odds ratio; BMI, body mass index; CEAP, Clinical, Etiological, Anatomical, and Pathophysiological; DOAC, direct oral anticoagulant; DVT, deep vein thrombosis; ECS, elastic compression stockings; LMWH, low-molecular-weight heparin; OR, odds ratio; VS, Villalta score; VTE, venous thromboembolism.
Defined as the use of ECS 50% of the total study time and a modified GIRERD score of 0 to 2.
All variables achieving a P value ≤ .2 and the ECS strength allocation group were entered into the multivariable model. The following variables were successively excluded: DVT extent, age, previous VTE, other chronic comorbidities, contralateral VS, type of anticoagulant treatment, and CEAP C3 in the leg ipsilateral to the DVT.
P.2 in univariate analysis.
Beyond identifying risk factors for PTS, our primary objective was to develop a decision tree to determine which patients would benefit most from ECS use (Figure). Thus, among nonobese patients with extensive DVT involving at least 4 venous segments, wearing ECS on at least a reasonably consistent basis for 2 years was associated with a nearly 2-fold reduction in the risk of PTS (29.8% vs 54.3%; RR, 0.55 [95% CI, 0.32-0.94]). In contrast, the benefit of ECS use was seemingly lower in patients with nonextensive DVT, whether they were nonobese (17.0% vs 24.0%; RR, 0.71 [95% CI, 0.36-1.38]) or obese (29.2% vs 46.7%; RR, 0.63 [95% CI, 0.27-1.43]). Among patients with obesity with extensive DVT, the benefit of wearing ECS appeared minimal, if any (73.3% vs 83.3%; RR, 0.88 [95% CI, 0.55-1.41]).
Figure.
Decision tree algorithm depicting the proportion of patients with postthrombotic syndrome (PTS) 2 years after a first proximal deep vein thrombosis, according to the presence/absence of independent risk factors for PTS. BMI, body mass index; ECS, elastic compression stockings.
Our study has several strengths and limitations that should be acknowledged. First, although we observed that greater adherence to the study ECS was associated with a lower risk of PTS, this does not necessarily imply a causal relationship. It is possible that patients who eventually developed PTS discontinued ECS use early, perceiving them as ineffective in alleviating their symptoms. However, the significantly lower proportion of patients with PTS at 2 years who were at least reasonably adherent to ECS during the first 3 months of treatment argues against reverse causality. Similarly, it is conceivable that ECS may have treated rather than prevented PTS symptoms, leading adherent patients to report fewer symptoms. Supporting this interpretation, a recent study demonstrated that quality-of-life scores for patients with PTS who wore ECS were comparable to those of patients without PTS [14]. Since adherence was self-reported by patients and not objectively assessed, an adherence reporting bias cannot be excluded. The magnitude of this bias, particularly in relation to PTS symptoms, is difficult to evaluate. However, the inclusion of the GIRERD score in the definition of adherence may have mitigated both the risk and the potential impact of such bias. Second, although our study included a relatively large cohort (302 patients, 97 with PTS), we likely lacked sufficient statistical power to detect a significant association between preexisting chronic venous insufficiency and PTS risk [15,16]. Specifically, neither the Clinical, Etiological, Anatomical, and Pathophysiological C3 or higher classification in the leg ipsilateral to the DVT nor a Villalta score ≥ 5 in the contralateral leg showed a statistically significant association with PTS risk in our multivariable model. Consequently, we could not evaluate the impact of ECS within these subgroups using our decision model. Similarly, consistent with the existing literature, we observed a trend toward approximately one-third fewer PTS cases in patients treated with DOACs than in those treated with vitamin K antagonists. Again, this difference did not reach statistical significance in multivariable analysis [17,18]. Third, overall adherence to ECS in our study was lower than that reported in historical open-label RCTs (64.5% vs >75%) [4,5]. However, the definition of adherence used in CELEST was stringent, incorporating both patient self-report and the GIRERD adherence assessment score. Originally developed for patients with hypertension, the GIRERD score captures patients’ attitudes and behaviors toward their treatment, providing insight into both intentional and unintentional nonadherence. We believe that adherence to CELEST was rigorously assessed. Considerable efforts were made to support and enhance adherence, extending beyond regular phone follow-ups by research coordinators. Patients were offered different ECS options (below- or above-knee length, open- or closed-toe models) that were replaced every 3 months as needed. Donning devices were also provided to facilitate daily use. Our seemingly suboptimal adherence rates also reflect the real-world difficulty of consistently wearing ECS daily over a prolonged period [19,20]. Nevertheless, our findings suggest that even moderate adherence to ECS use can confer benefits in preventing PTS. Improved adherence could be achieved by using customized graduated ECS rather than standard models, sending regular SMS reminders to patients, and reducing the expected duration of ECS therapy [[21], [22], [23]]. The recommendation of a 2-year duration, which prevailed when CELEST was conducted, may have appeared discouraging to some patients [24]. Finally, we did not collect data on adherence to anticoagulant therapy. Consequently, we cannot fully exclude that the imbalance in the proportion of poorly vs reasonably adherent patients treated with DOACs (78.7% vs 62.4%) may have slightly influenced our results. One of the key strengths of our study is its multicenter, double-blind design. Most participants were treated with DOACs, which aligns with current standard practice for DVT management. Our ultrasound protocol enabled precise assessment of thrombus burden, which appears to be a stronger predictor of PTS than the usual, well-described most proximal anatomical extent of the DVT [25]. This highlights the importance of clot distribution over anatomical location alone. To our knowledge, this is the first study to evaluate specific clinical scenarios in which ECS use may be most beneficial. Furthermore, our multivariable model demonstrated good predictive performance, with an area under the receiver operating characteristic curve of 0.77 (0.72-0.83).
Adherence to ECS is a significant independent predictor of PTS. The protective effect of ECS becomes apparent when they are worn at least 50% of the time. Adherence to ECS appears particularly advantageous in nonobese patients with extensive DVT. While ECS use is associated with a reduced risk of developing PTS, we believe that definitive evidence of their efficacy can only be established through a new double-blind RCT.
Declaration of AI and AI-Assisted Technologies in the Writing Process
During the preparation of this work the authors used OpenAI in order to improve the English of the manuscript. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.
Acknowledgments
We express our gratitude to the members of the study expert committee and to all investigators, residents, and physicians from the emergency departments and vascular medicine units at all participating centers. We would also like to thank all secretaries and clinical research associates involved in this study, as well as the Association des Médecins Vasculaires Hospitaliers, the French Society for Vascular Medicine, and the College Français de Pathologie Vasculaire for promoting the study among French Vascular Medicine Physicians. Last but not least, we would like to express our gratitude to the patients who made the study possible by agreeing to participate in the trial.
Funding
This research was funded by a grant from Laboratoires Innothéra. Laboratoires Innothéra approved the study protocol, manufactured the study elastic compression stockings, and funded both an independent audit of all study data and English-language editing of the primary manuscript. The company also approved the final version of the present manuscript. However, Laboratoires Innothéra had no role in data collection, analysis, or interpretation.
Author contributions
Conception and design of the study: J.-P.G. and J.-L.B. Funding: J.-L.B. Development of the “Compression Elastique Evaluation du Syndrome post Thrombotique” protocol: J.-P.G., B.T., F.V., and J.-L.B. Data acquisition or analysis: J.-P.G., C.G.-V., B.T., A.S., D.L., C.G., M.M., F.C., P.M., A.C., and J.-L.B. Drafting of the article: J.-P.G. Critical revision for important intellectual content: J.-P.G., C.G.-V., B.T., A.S., D.L., C.G., M.M., F.C., P.M., A.C., and J.-L.B. Final approval of the manuscript: J.-P.G., C.G.-V., B.T., A.S., D.L., C.G., M.M., F.C., P.M., A.C., F.V., and J.-L.B.
Relationship Disclosure
J.-P.G. is an investigator of the CanVECTOR Network, which receives grant funding from the Canadian Institutes of Health Research (funding reference: CDT-142654). F.V. is an employee of Laboratoires Innothéra. The other authors declare no conflicts of interest relevant to the present work.
Footnotes
Handling Editor: Professor Michael Makris
A portion of this study was presented at the 63rd American Society of Hematology Annual Meeting, Atlanta, USA, December 11-14, 2021.
References
- 1.Kahn S.R., 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;7:879–883. doi: 10.1111/j.1538-7836.2009.03294.x. [DOI] [PubMed] [Google Scholar]
- 2.Hupperetz R., Ten Cate-Hoek A. The postthrombotic syndrome, where are we now? Vasa. 2025;54:7–19. doi: 10.1024/0301-1526/a001156. [DOI] [PubMed] [Google Scholar]
- 3.Palareti G., Santagata D., De Ponti C., Ageno W., Prandoni P. Anticoagulation and compression therapy for proximal acute deep vein thrombosis. Vasa. 2024;53:289–297. doi: 10.1024/0301-1526/a001138. [DOI] [PubMed] [Google Scholar]
- 4.Brandjes D.P., Buller H.R., Heijboer H., Huisman M.V., de Rijk M., Jagt H., et al. Randomised trial of effect of compression stockings in patients with symptomatic proximal-vein thrombosis. Lancet. 1997;349:759–762. doi: 10.1016/S0140-6736(96)12215-7. [DOI] [PubMed] [Google Scholar]
- 5.Prandoni P., Lensing A.W., Prins M.H., Frulla M., Marchiori A., Bernardi E., et al. Below-knee elastic compression stockings to prevent the post-thrombotic syndrome: a randomized, controlled trial. Ann Intern Med. 2004;141:249–256. doi: 10.7326/0003-4819-141-4-200408170-00004. [DOI] [PubMed] [Google Scholar]
- 6.Kahn S.R., Shapiro S., Wells P.S., Rodger M.A., Kovacs M.J., Anderson D.R., et al. Compression stockings to prevent post-thrombotic syndrome: a randomised placebo-controlled trial. Lancet. 2014;383:880–888. doi: 10.1016/S0140-6736(13)61902-9. [DOI] [PubMed] [Google Scholar]
- 7.Galanaud J.P., Genty-Vermorel C., Rolland C., Comte A., Ouvry P., Bertaina I., et al. Compression stockings to prevent postthrombotic syndrome: literature overview and presentation of the CELEST trial. Res Pract Thromb Haemost. 2020;4:1239–1250. doi: 10.1002/rth2.12445. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Galanaud J.P., Genty-Vermorel C., Barrellier M.T., Becker F., Jabbour V., Blaise S., et al. 25 mm Hg versus 35 mm Hg elastic compression stockings to prevent post-thrombotic syndrome after deep vein thrombosis (CELEST): a randomised, double-blind, non-inferiority trial. Lancet Haematol. 2022;9:e886–e896. doi: 10.1016/S2352-3026(22)00247-2. [DOI] [PubMed] [Google Scholar]
- 9.Song Y.Y., Lu Y. Decision tree methods: applications for classification and prediction. Shanghai Arch Psychiatry. 2015;27:130–135. doi: 10.11919/j.issn.1002-0829.215044. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Strijkers R.H., Arnoldussen C.W., Wittens C.H. Validation of the LET classification. Phlebology. 2015;30:14–19. doi: 10.1177/0268355515569133. [DOI] [PubMed] [Google Scholar]
- 11.Nielsen J.D., Hermann T.S., Fredskilde P.C.A. Graduated elastic compression stockings in the prevention of post-thrombotic syndrome: a systematic review and meta-analysis. Phlebology. 2024;39:229–237. doi: 10.1177/02683555231225268. [DOI] [PubMed] [Google Scholar]
- 12.Yang X., Zhang X., Yin M., Wang R., Lu X., Ye K. Elastic compression stockings to prevent post-thrombotic syndrome in proximal deep venous thrombosis patients without thrombus removal. J Vasc Surg Venous Lymphat Disord. 2022;10:293–299. doi: 10.1016/j.jvsv.2021.06.023. [DOI] [PubMed] [Google Scholar]
- 13.Thapar A., Lawton R., Bradbury A., Cullum N., Gohel M., Horne R., et al. Compression hosiery to avoid post-thrombotic syndrome: CHAPS randomized clinical trial. Br J Surg. 2025;112:znaf018. doi: 10.1093/bjs/znaf018. [DOI] [PubMed] [Google Scholar]
- 14.Silva J.C., Constâncio V., Lima P., Nunes C., Silva E., Anacleto G., et al. Determinants of quality of life in patients with post-thrombotic syndrome. Ann Vasc Surg. 2022;85:253–261. doi: 10.1016/j.avsg.2022.03.002. [DOI] [PubMed] [Google Scholar]
- 15.Galanaud J.P., Ducruet T., Kahn S.R., SOX trial investigators group Accuracy of contralateral Villalta score to assess for pre-existing chronic venous insufficiency in patients with unilateral deep vein thrombosis. J Thromb Haemost. 2020;18:3309–3315. doi: 10.1111/jth.15091. [DOI] [PubMed] [Google Scholar]
- 16.Galanaud J.P., Holcroft C.A., Rodger M.A., Kovacs M.J., Betancourt M.T., Wells P.S., et al. Predictors of post-thrombotic syndrome in a population with a first deep vein thrombosis and no primary venous insufficiency. J Thromb Haemost. 2013;11:474–480. doi: 10.1111/jth.12106. [DOI] [PubMed] [Google Scholar]
- 17.Karathanos C., Nana P., Spanos K., Kouvelos G., Brotis A., Matsagas M., et al. Efficacy of rivaroxaban in prevention of post-thrombotic syndrome: a systematic review and meta-analysis. J Vasc Surg Venous Lymphat Disord. 2021;9:1568–15676.e1. doi: 10.1016/j.jvsv.2021.04.016. [DOI] [PubMed] [Google Scholar]
- 18.Makedonov I., Kahn S.R., Abdulrehman J., Schulman S., Delluc A., Gross P., et al. Prevention of the postthrombotic syndrome with anticoagulation: a narrative review. Thromb Haemost. 2022;122:1255–1264. doi: 10.1055/a-1711-1263. [DOI] [PubMed] [Google Scholar]
- 19.Kankam H.K.N., Lim C.S., Fiorentino F., Davies A.H., Gohel M.S. A summation analysis of compliance and complications of compression hosiery for patients with chronic venous disease or post-thrombotic syndrome. Eur J Vasc Endovasc Surg. 2018;55:406–416. doi: 10.1016/j.ejvs.2017.11.025. [DOI] [PubMed] [Google Scholar]
- 20.Raju S., Hollis K., Neglen P. Use of compression stockings in chronic venous disease: patient compliance and efficacy. Ann Vasc Surg. 2007;21:790–795. doi: 10.1016/j.avsg.2007.07.014. [DOI] [PubMed] [Google Scholar]
- 21.Yang W.T., Xiong Y., Wang S.X., Ren H.L., Gong C., Jin Z.Y., et al. A randomized controlled trial of standard vs customized graduated elastic compression stockings in patients with chronic venous disease. J Vasc Surg Venous Lymphat Disord. 2024;12 doi: 10.1016/j.jvsv.2023.08.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Uhl J.F., Benigni J.P., Chahim M., Fréderic D. Prospective randomized controlled study of patient compliance in using a compression stocking: importance of recommendations of the practitioner as a factor for better compliance. Phlebology. 2018;33:36–43. doi: 10.1177/0268355516682886. [DOI] [PubMed] [Google Scholar]
- 23.Ten Cate-Hoek A.J., Amin E.E., Bouman A.C., Meijer K., Tick L.W., Middeldorp S., et al. Individualised versus standard duration of elastic compression therapy for prevention of post-thrombotic syndrome (IDEAL DVT): a multicentre, randomised, single-blind, allocation-concealed, non-inferiority trial. Lancet Haematol. 2018;5:e25–e33. doi: 10.1016/S2352-3026(17)30227-2. [DOI] [PubMed] [Google Scholar]
- 24.Kearon C., Akl E.A., Comerota A.J., Prandoni P., Bounameaux H., Goldhaber S.Z., et al. Antithrombotic therapy for VTE disease: antithrombotic therapy and prevention of thrombosis, 9th ed: American College of Chest Physicians evidence-based clinical practice guidelines. Chest. 2012;141:e419S–e496S. doi: 10.1378/chest.11-2301. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Kahn S.R., Comerota A.J., Cushman M., Evans N.S., Ginsberg J.S., Goldenberg N.A., et al. The postthrombotic syndrome: evidence-based prevention, diagnosis, and treatment strategies: a scientific statement from the American Heart Association. Circulation. 2014;130:1636–1661. doi: 10.1161/CIR.0000000000000130. [DOI] [PubMed] [Google Scholar]