These authors contributed equally to the article.
The post-thrombotic syndrome: risk factors and impact on the course of thrombotic disease
Version of Record online: 24 NOV 2005
Journal of Thrombosis and Haemostasis
Volume 3, Issue 12, pages 2671–2676, December 2005
How to Cite
STAIN, M., SCHÖNAUER, V., MINAR, E., BIALONCZYK, C., HIRSCHL, M., WELTERMANN, A., KYRLE, P. A. and EICHINGER, S. (2005), The post-thrombotic syndrome: risk factors and impact on the course of thrombotic disease. Journal of Thrombosis and Haemostasis, 3: 2671–2676. doi: 10.1111/j.1538-7836.2005.01648.x
- Issue online: 24 NOV 2005
- Version of Record online: 24 NOV 2005
- Received 8 June 2005, accepted 2 September 2005
- deep vein thrombosis;
- post-thrombotic syndrome;
- risk of occurrence
Summary. Background: The post-thrombotic syndrome (PTS) is a frequent complication of deep vein thrombosis (DVT). Patients with recurrent ipsilateral DVT have an increased risk of PTS; other risk factors are unknown. Objectives: To establish risk factors of PTS and its impact on venous thrombotic disease. Patients: We prospectively followed 406 patients after a first symptomatic DVT for a median of 60 months. Patients with recurrent DVT, a natural inhibitor deficiency, the lupus anticoagulant, cancer, long-term anticoagulation, an observation time < 18 months and DVT-recurrence prior PTS-assessment were excluded. Study outcomes were occurrence of PTS and recurrent symptomatic DVT. Results: PTS was assessed after 44 ± 23 months (mean ± SD) using a clinical classification score. PTS developed in 176 of 406 patients (43.3%). Severe PTS was rare (1.4%). Proximal DVT was the strongest risk factor of PTS [odds ratio (OR) 2.1, 95% confidence interval (CI) 1.3–3.7]. Male gender (OR 1.6, 95% CI 1.0–2.8) and elevated D-dimer levels (OR 1.9, 95% CI 1.0–3.9) were weaker risk factors. Factor V Leiden, factor II G20210A or high factor VIII did not confer an increased risk of PTS. At 4 years, the cumulative probability of recurrence was 7.4% (95% CI 3.2–11.7) among patients with PTS when compared with 1.6% (95% CI 0–3.5; P < 0.02) among patients without PTS. The risk of recurrence was 2.6-fold (95% CI 1.2–5.9) increased when PTS was present. Conclusions: Proximal DVT, male gender, and high D-dimer levels are independently associated with the development of PTS in patients with a first DVT. Patients with PTS have an increased risk of recurrent venous thromboembolism.
What threatens patients after a deep vein thrombosis (DVT) besides the risk of recurrence is the development of the post-thrombotic syndrome (PTS). The clinical characteristics of PTS range from light pain, occasional swelling, and venous ectasia to an extremely troublesome condition with chronic pain, intractable edema, skin alterations and, ultimately, leg ulcer. Approximately 20%–50% of patients with symptomatic DVT will develop PTS within 2 years despite adequate treatment of DVT [1–3].
Recurrent, ipsilateral DVT is the best established risk factor of PTS, increasing the risk as much as sixfold when compared with patients without recurrence [4–8]. There is no agreement, however, regarding the impact of other factors on the risk of PTS, such as location of DVT [4,9–14], or of patient characteristics, including age, sex, body mass index (BMI), or thrombophilia [15–20].
Therapeutic options for PTS are very limited. The frequency of PTS was halved with the use of graduated elastic compression stockings (ECS) (data coming from two randomized trials) [19,21]. Data that will allow stratifying patients according to whether or not they will benefit from ECS are not available. It may, hence, be important to define risk factors of PTS, in order to encourage the consequent use of ECS and to target prophylaxis in high-risk patients.
Within the frame of a large cohort study, we prospectively followed patients with a first DVT of the leg to investigate the development of PTS and to define risk factors of PTS and to study the association of PTS and risk of recurrent thrombosis.
Patients and study design
All patients in the present analysis are participants of the Austrian Study on Recurrent Venous Thromboembolism, which is an ongoing cohort study with the aim to define risk factors of recurrence in patients with venous thromboembolism (VTE) [22,23]. Between July 1992 and October 2004, 3291 patients older than 18 years and treated with vitamin K antagonists for at least 3 months after symptomatic VTE were screened. Inclusion criteria for the present study were a first, symptomatic, objectively confirmed DVT of the lower limbs. Patients with a history of VTE, natural anticoagulant deficiency, presence of the lupus anticoagulant, cancer, or need for long-term anticoagulant therapy were excluded. Further exclusion criteria were a pre-existing leg ulcer, an observation period of < 18 months, recurrent VTE before assessment of PTS or unavailability for PTS evaluation. A flowchart for the inclusion and exclusion of patients is shown in Fig. 1. Written informed consent was obtained and patients entered the study the day of withdrawal of vitamin K antagonists. Patients were seen at 3-month intervals during the first year and every 6 months thereafter. They were instructed to wear ECS (30 mmHg pressure at the ankle) and to report to the study center (Department of Internal Medicine I, Medical University of Vienna, Austria) in case of symptoms suggestive of recurrent VTE, such as leg pain, swelling of leg, dyspnea, and thoracic pain.
Diagnosis of VTE
The diagnosis of DVT was established by venography or color-coded duplex sonography (in case of proximal DVT only). If venography was used, one of the following direct or indirect criteria had to be fulfilled: a constant-filling defect present on two views, an abrupt discontinuation of the contrast-filled vessel at a constant point in the vein, failure of the entire deep vein system to fill without an external compressing process, with or without venous flow through collateral veins.
Diagnostic criteria for color-coded duplex sonography were the following: visualization of an intraluminal thrombus in a deep vein, lack of complete compressibility, and lack of flow spontaneously and after distal manipulation.
Diagnostic criteria for pulmonary embolism (PE) were a high probability ventilation-perfusion lung scan according to the criteria of the Prospective Investigation of Pulmonary Embolism Diagnosis  or the finding of one or several low attenuation areas that partly or completely filled the lumen of an opacified vessel in spiral computed tomography.
The primary endpoint of the study was the development of PTS. A clinical examination for the presence of PTS was performed no sooner than 18 months after diagnosis of DVT using the clinical component of a standard clinical scale [25,26] (Table 1). The presence of edema, teleangiectasias, venous ectasia, and varicose veins, hyperpigmentation, eczema, lipodermatosclerosis, and open or healed ulcer in the affected leg was determined. The patients were asked about symptoms of PTS, such as pain, heaviness, swelling, cramps, paresthesia, and itching. PTS was assessed by the same experienced physician in all patients. In addition, the use of ECS was registered (no or occasional use only vs. use for most of the daytime on most days).
|Class 0||Symptoms * only, no visible or palpable signs|
|Class 1||Teleangiectasias, reticular veins|
|Class 2||Varicose veins|
|Class 3||Edema, no skin changes|
|Class 4||Skin changes, pigmentation, lipodermatosclerosis|
|Class 5||Skin changes with healed ulcer|
|Class 6||Skin changes with active ulcer|
Another endpoint was recurrent symptomatic VTE confirmed by venography, color-coded duplex sonography, ventilation-perfusion lung scanning, or spiral computed tomography according to the aforementioned criteria. In the vast majority of patients, the diagnosis of recurrent DVT was established by venography and confirmed by an adjudication committee consisting of independent clinicians and radiologists unaware of the presence or absence of thrombotic risk factors. Recurrent DVT was diagnosed if the patient had a thrombus in another deep vein in the leg involved in the first event, a thrombus in the other leg, or a thrombus in the same venous system involved in the previous event with a proximal extension of the thrombus if the upper limit of the original thrombus had been visible or the presence of a constant-filling defect surrounded by contrast medium if it had not. In a few patients with recurrent proximal thrombosis in the contralateral leg, the diagnosis was established by duplex sonography.
Venous blood was obtained after overnight fasting approximately 3 weeks after withdrawal of vitamin K antagonists, placed in 1/10 volume of 0.11 mmol L−1 of trisodium citrate and centrifuged for 20 min at 2000 g. The plasma was stored at −80 °C. Routine laboratory methods were used to identify plasma levels of antithrombin, protein C, and protein S. Screening for factor (F)V Leiden and factor (F)II G20210A was carried out on genomic DNA as described previously [27,28]. Factor (F)VIII was measured by a one-step clotting assay with the use of FVIII-deficient plasma. D-dimer levels were measured by an enzyme-linked immunoassay (Asserchrom D-dimer, Boehringer Mannheim, Germany). The presence of the lupus anticoagulant was established on the basis of the criteria of the Subcommittee on Lupus Anticoagulant/Antiphospholipid Antibody of the Scientific and Standardisation Committee of the International Society on Thrombosis and Haemostasis .
For numerical operations SPSS software (SPSS Inc. Headquarters, Chicago, IL) was used. Values are given as mean ± standard deviation (SD). To test for homogeneity between strata, we applied the log–rank and the generalized Wilcoxon test. Categorical data were checked for homogeneity with the use of contingency-table analyses (by the chi-square test). Simple descriptive statistics were computed to provide a clear presentation of data. Logistic regression was used to test the following potential predictors of PTS for independence: age, sex, location of DVT, BMI (determined at study entry) as a dichotomized variable with a cutoff at 25 kg m−2), D-dimer (as a dichotomized variable with a cutoff at 500 ng mL−1), FV Leiden mutation, FII G20210A mutation, and FVIII (as a dichotomized variable with a cutoff at 234 IU dL−1). We applied linear regression with PTS as a dependent variable and the above-mentioned variables as explanatory covariates. PTS was entered as a categorical variable (0, ‘no PTS’; 1, ‘PTS’) to evaluate the odds ratio (OR) with ‘no PTS’ as a reference category. Times to recurrence (uncensored observations) or follow-up times in patients without recurrence (censored observations) were analyzed according to survival methods . The probability of recurrence was estimated according to the method of Kaplan and Meier .
A total of 406 patients (160 men and 246 women) were prospectively followed for a median of 60 months after the diagnosis of the first, symptomatic DVT of the lower limbs, the total follow-up was 2192 patient years. The mean age at diagnosis of DVT was 46 ± 16 years. In 75 patients (18.5%), DVT was provoked by surgery, trauma, or pregnancy; 331 patients (81.5%) had unprovoked DVT; 152 patients (37.4%) had distal; 254 (62.6%) had proximal thrombosis; and 118 patients (29.1%) had objectively confirmed PE in addition.
All patients except 13 (3.2%) were using ECS; 137 patients (33.7%) wore them for < 6 months, 118 (29.1%) for up to 2 years, and 138 (34%) for > 2 years during most of the daytime.
The post-thrombotic syndrome
The PTS evaluation was performed no sooner than 18 months (mean 44 ± 23 months; median 35 months; range 18–220 months) after diagnosis of DVT. A total of 176 of 406 patients (43.3%) developed signs and symptoms of PTS. PTS was considered mild or moderate in 170 patients (41.9%) [clinical-etiologic-anatomic-pathophysiologic (CEAP) classification, clinical class 1–4] and severe in six patients (1.4%) (CEAP, clinical class 5 and 6). The characteristics of patients with and without PTS are shown in Table 2.
|No PTS (n = 230)||PTS (n = 176)||P-value|
|Age (years)||44 ± 15||49 ± 15||0.002|
|Sex (male)||80 (35%)||80 (46%)||0.03|
|Duration of OAC (months)||7.2 ± 5.6||9.4 ± 17.3||ns|
|Location of DVT (proximal)||128 (55.7%)||126 (71.6%)||0.001|
|BMI > 25 kg m−2||133 (58.1%)||120 (68.6%)||0.03|
|D-dimer > 500 ng mL−1||20 (14.3%)||36 (28.3%)||0.006|
|Factor VIII > 234 IU dL−1||8 (3.5%)||12 (6.8%)||ns|
|Factor V Leiden||84 (36.5%)||58 (33%)||ns|
|Factor II G20210A||20 (8.7%)||13 (7.4%)||ns|
In a univariate analysis, advancing age [OR 1.2 for every 10-year increase, 95% confidence interval (CI) 1.1–1.3], male gender (OR 1.6, 95% CI 1.0–2.3), proximal DVT (when compared with distal DVT, OR 2.0, 95% CI 1.3–3.1), BMI > 25 kg m−2 (OR 1.6, 95% CI 1.0–2.4), and D-dimer > 500 ng mL−1 (OR 2.4, 95% CI 1.3–4.4) were associated with an increased risk of PTS. Factor V Leiden (OR 0.9, 95% CI 0.6–1.3), FII G20210A (OR 0.8, 95% CI 0.4–1.7) or high FVIII (OR 2.0, 95% CI 0.8–5.1) did not confer an increased risk of PTS. If the aforementioned variables were included in a multivariate model, proximal DVT was the strongest risk factor of PTS (OR 2.1, 95% CI 1.3–3.7). Male gender (OR 1.6, 95% CI 1.0–2.8) and elevated D-dimer (OR 1.9, 95% CI 1.0–3.9) were weaker risk factors. Duration of anticoagulant therapy (< 6 months vs. 6–12 months and > 12 months respectively) had no impact on the risk of PTS. There was no difference in the frequency of PTS in patients with unprovoked DVT compared with patients with DVT secondary to surgery or trauma (OR 1.0, 95% CI 0.6–1.7).
Recurrent venous thromboembolism
Venous thromboembolism recurred in 26 of 406 patients (6.4%) (18 men and eight women). Thirteen patients had recurrent contralateral DVT. Five patients had recurrent isolated non-fatal PE. In 18 of these 26 patients, the diagnosis of PTS was established in average 14 months before recurrence.
At 4 years, the cumulative probability of recurrence was 7.4% (95% CI 3.2–11.7) among patients with PTS when compared with 1.6% (95% CI 0–3.5) among patients without PTS (P < 0.02) (Fig. 2). Patients with PTS had an almost threefold higher risk of recurrence when compared with patients without PTS [relative risk (RR) 2.6, 95% CI 1.2–5.9]. After adjustment for age and sex, the RR was 2.3 (95% CI 1.0–5.3). After inclusion of D-dimer in addition to age and sex into the multivariate analysis, the RR was 3.0 (95% CI 1.1 –8.3).
The Austrian Study on Recurrent Venous Thromboembolism is a prospective cohort study, which offers the opportunity to study PTS in a large number of well-defined patients with a first DVT of the leg. Within an average of 3 years, 43% of these patients developed PTS. Severe, disabling PTS with ulcers was, however, a rare event seen in only < 2%. A similar frequency of PTS was reported in other prospective studies [4,13,19–21] although comparison of data is hampered by lack of standardized diagnostic criteria and different observation times. As PTS will become apparent within the first 2 years after the acute DVT in a majority of patients [1–4], we assessed the presence of PTS no sooner than 18 months after the acute event.
In our patient cohort, proximal DVT was the strongest independent risk factor conferring a more than twofold increased likelihood of PTS. This finding is in accordance with other studies in which the association between proximal DVT and PTS was found comparably strong [9–13]. In contrast, no relation between the location of DVT and PTS was found by other investigators [4,5]. This discrepancy can be explained by the inclusion of patients with recurrent DVT. As recurrent DVT is an established strong risk factor of PTS, the impact of proximal DVT might have been concealed.
Noteworthy, men are at higher risk of PTS than women. This important risk factor may not have emerged in previous studies because the study populations were small [13,15,20] or only women or patients with recurrent DVT were included [5,7,16].
An elevated BMI has been linked with an increased risk of PTS [15,16,20]. In our study, a BMI > 25 kg m−2 was a risk factor of PTS in the univariate analysis, but the risk disappeared after adjustment for confounding variables, including age, sex, and location of DVT. This contradiction may be explained by small patient numbers [15,16] and different populations [16,20]. One study  included only females and an elevated risk of PTS was described only in women with a BMI of > 22 kg m−2, a distinction of little relevance in clinical practice.
The role of laboratory thrombophilia with regard to the risk of PTS is not well established. We evaluated the association between PTS and various markers of thrombophilia, such as FV Leiden or FII G20210A, elevated clotting factors, and D-dimer. In our cohort, carriership of FV Leiden or the prothrombin mutation did not confer an increased risk of PTS. Accordingly, no influence of hereditary thrombophilia on the risk of PTS was found in an Italian study . In a recent trial, patients with FV Leiden unexpectedly had a significantly lower risk of PTS compared with non-carriers of the mutation . These discrepant findings could be explained by patient selection (two-thirds of the patients had recurrent VTE, 49% of patients without PTS had FV Leiden).
Elevated D-dimer is a risk factor of a first as well as of recurrent DVT [22,32,33]. We found that an initially high D-dimer (assessed approximately 3 weeks after withdrawal of vitamin K antagonists) predicted the development of PTS. Recently, elevated levels of D-dimer have been found to predict poor outcome (defined as development of PTS, residual thrombus, and recurrent VTE) in children . Elevated clotting factor levels, in particular high FVIII, were not associated with a higher risk of PTS in our patient cohort. This finding, however, must be interpreted cautiously. High FVIII is a strong risk factor of recurrence  and was present in a high percentage of patients with early recurrence. As only patients with a follow-up of at least 18 months before recurrence were eligible for PTS assessment, a large proportion of patients with high FVIII had been excluded from the present analysis.
An important finding of our study was that the presence of PTS conferred an almost threefold increased risk of recurrent VTE. One explanation for this association might be the fact that PTS and recurrent VTE share similar risk factors. Another reason, however, could be a reduced thrombolytic capacity in these patients. This assumption might be corroborated by reports of an increased risk of recurrence in patients with residual thrombosis . Patients with early edema are more likely to have residual thrombosis , but the association between residual thrombosis and development of PTS has not yet been established. In a Canadian study, ultrasound evidence of residual thrombosis 3–4 months after acute DVT was not associated with an increased risk of PTS . We did not evaluate the presence of residual thrombosis in our patient cohort. Thus, the relation between residual vein thrombosis and the development of PTS needs to be further investigated.
Clinical signs and symptoms of recurrent DVT and PTS can be very similar. Thus, diagnosis of recurrent DVT, particularly in the ipsilateral leg, can be challenging in patients with PTS. We, therefore, applied very strict predefined criteria for the diagnosis of recurrent ipsilateral DVT in all patients. Recurrence had to be diagnosed by venography and had to be confirmed by an independent adjudication committee.
There are several limitations to our study. First, the patients in our study cohort are younger and show a higher percentage of unprovoked DVT events compared with most other study cohorts, which could be related to the referral pattern to our center. Secondly, we excluded patients with recurrent DVT, an acknowledged, high-risk factor for PTS because we felt that the impact of other risk factors would be masked. Thirdly, because of the study concept of visits scheduled from the withdrawal of oral anticoagulants on, the time period from acute DVT to PTS assessment varied according to the duration of treatment with vitamin K antagonists. Fourthly, we excluded 234 eligible patients who were unavailable for PTS assessment, which may have influenced our results.
The PTS is a frequent complication in patients with DVT resulting in impaired quality of life and affecting the course of the thrombotic disease. In our cohort, PTS was strongly predicted by proximal DVT, and men with proximal DVT and high D-dimer levels constituted the highest risk group. Development of PTS is also associated with an increased risk of recurrent VTE.
- 30The Statistical Analysis of Failure Time Data. New York: John Wiley, 1980..