A portion of this work was presented at the American Society of Hematology 47th Annual Meeting, Atlanta, GA, USA, December 2005.
Susan R. Kahn, Sir Mortimer B. Davis Jewish General Hospital, 3755 Cote Ste. Catherine Rm. A-127, Montreal, QC H3T 1E2, Canada. Tel.: +1 514 340 8222 ext. 7587; fax: +1 514 340 7564. E-mail: email@example.com
Summary. Background/objectives: We prospectively measured change in quality of life (QOL) during the 2 years after a diagnosis of deep vein thrombosis (DVT) and evaluated determinants of QOL, including development of the post-thrombotic syndrome (PTS). Patients/methods: Consecutive patients with acute DVT were recruited from 2001 to 2004 at eight hospitals in Canada. At study visits at baseline, and 1, 4, 8, 12 and 24 months, clinical data were collected, standardized PTS assessments were performed, and QOL questionnaires were self-completed. Generic QOL was measured using the Short-Form Health Survey-36 (SF-36) questionnaire. Venous disease-specific QOL was measured using the Venous Insufficiency Epidemiological and Economic Study (VEINES)-QOL/Sym questionnaire. The change in QOL scores over a 2-year follow-up was assessed. The influence of PTS and other characteristics on QOL at 2 years was evaluated using multivariable regression analyses. Results: Among the 387 patients recruited, the average age was 56 years, two-thirds were outpatients, and 60% had proximal DVT. The cumulative incidence of PTS was 47%. On average, QOL scores improved during follow-up. However, patients who developed PTS had lower scores at all visits and significantly less improvement in QOL over time (P-values for PTS*time interaction were 0.001, 0.012, 0.014 and 0.006 for PCS, MCS, VEINES-QOL and VEINES-Sym). Multivariable regression analyses showed that PTS (P < 0.0001), age (P = 0.0009), proximal DVT (P = 0.01) and inpatient status (P = 0.04) independently predicted 2-year SF-36 PCS scores. PTS alone independently predicted 2-year VEINES-QOL (P < 0.0001) and VEINES-Sym (P < 0.0001) scores. Conclusions: Development of PTS is the principal determinant of health-related QOL 2 years after DVT. Our study provides prognostic information on patient-reported outcomes after DVT and emphasizes the need for effective prevention and treatment of the PTS.
Deep vein thrombosis (DVT) of the leg, a common cardiovascular condition with substantial morbidity and mortality, is estimated to affect more than 460 000 people in the European Union and 300 000 people in North America each year [1,2]. Even with optimal anticoagulant treatment, acute symptoms of DVT such as leg pain and swelling can take weeks to subside, and 20–40% of patients develop the chronic post-thrombotic syndrome (PTS), which is characterized by leg pain, heaviness, swelling and, in severe cases, skin ulcers .
For chronic conditions such as PTS, assessment of health-related quality of life (QOL) can yield information on ‘patient-important outcomes’  that is not obtained by standard clinical measures. QOL defines health outcomes in broader terms than morbidity and mortality, and is measured from the patient’s perspective, which enables evaluation of the impact of the condition on patients’ health and daily functioning . Comprehensive assessment of QOL should include the use of both generic and disease-specific measures [5,6]. Generic measures allow comparison across populations of patients with different conditions, whereas disease-specific measures tend to be more responsive in detecting treatment effects and changes over time in patients who have the same disease [6,7].
Recently, our group developed and validated the Venous Insufficiency Epidemiological and Economic Study (VEINES)-QOL/Sym, a venous disease-specific measure of QOL for use in patients with chronic venous disorders as well as acute DVT [8,9]. In order to characterize the long-term effects of DVT on QOL, we performed a multicenter prospective cohort study in which we evaluated the influence of the PTS and other patient characteristics on QOL during the 2 years following DVT.
Materials and methods
Patients were recruited from among consecutive eligible patients with acute DVT seen in the emergency departments, outpatient clinics and inpatient wards of eight university-affiliated hospitals in Quebec and Ontario, Canada from April 2001 to September 2004. Patients were eligible to participate if they had symptomatic venous thrombosis of the deep calf or proximal veins of the lower limb that was objectively diagnosed [10,11] within the preceding 7 days and if they were able to read and understand English or French. Patients were excluded if: thrombosis was limited to the intramuscular (gastrocnemius or soleus) calf veins; estimated lifespan was less than 3 months; they were unable to complete a questionnaire; they were geographically inaccessible or unwilling to return for follow-up visits; or they were unable or refused to provide informed consent. Excluded patients were asked for permission to record basic demographic and clinical data in order to assess their comparability with participants.
Ethics approval was obtained from the relevant committees at each hospital center. Written informed consent was obtained from all patients prior to study entry.
At study entry, demographic and clinical characteristics were recorded. Patients were treated with low molecular weight heparin or unfractionated heparin for 5–7 days, followed by a minimum of 3 months of warfarin (target International Normalized Ratio 2–3) as per consensus guideline recommendations . Decisions regarding duration of warfarin treatment and use of compression stockings were made by the treating physicians.
Patients were evaluated at clinic visits that took place at baseline (within 7 days of DVT diagnosis), and at 1, 4, 8, 12 and 24 months. At each visit, a standardized assessment for PTS was performed, and patients self-completed generic and venous disease-specific QOL questionnaires, as described below.
Assessment of PTS
At each clinic visit, trained study personnel assessed subjects’ legs using the Villalta scale, a validated clinical PTS measure that grades the severity, from 0 (absent) to 3 (severe), of each of five patient-rated symptoms (pain, cramps, heaviness, pruritus, and paresthesia) and six clinician-rated clinical signs (edema, redness, skin induration, hyperpigmentation, venous ectasia, and pain during calf compression) [13,14]. Study personnel who rated the clinical signs were blind to patients’ responses to the symptom component of the scale and did not have access to scores obtained on previous visits.
Patients were classified as having developed PTS if the Villalta score for the same leg as the index DVT was ≥ 5 on at least two visits, starting at the 4-month visit or later, or was ≥ 5 at the final follow-up visit . A Villalta score of 5–9 at all assessments was considered to represent mild PTS, a score of 10–14 on at least one assessment was considered to represent moderate PTS, and a score of ≥ 15 or a venous ulcer was considered to represent severe PTS.
Assessment of QOL
Validated, self-administered instruments were used to measure generic and venous disease-specific QOL. These were the Short-Form Health Survey-36 (SF-36)  and the VEINES-QOL/Sym . The SF-36 produces Physical Component Summary (PCS) and Mental Component Summary (MCS) scores, which reflect physical and mental health status, respectively. The VEINES-QOL is a 26-item questionnaire that measures venous symptoms (heavy legs, aching legs, swelling, night cramps, heat or burning sensation, restless legs, throbbing, itching, tingling, intensity of leg pain), limitations in daily activities due to chronic venous disease, psychological impact of chronic venous disease, change over the past year, and time of day when the leg problem is most intense. The VEINES-Sym is a validated subscale of the VEINES-QOL that measures venous symptoms. Both English and French Canadian language versions of the VEINES-QOL have been shown to be acceptable, reliable, valid and responsive for use in patients with chronic venous disease as well as acute DVT [8,9]. For all measures, lower scores indicate poorer QOL.
Statistical analysis and sample size
Scores for the SF-36 PCS, SF-36 MCS, VEINES-QOL and VEINES-Sym measures at baseline and each follow-up visit were computed using standard scoring algorithms [8,17]. Mean within-patient change scores (baseline to 2 years) were assessed using paired t-tests (Ho: Δ = 0), where positive change scores indicate improvement in QOL. Change in QOL over time was compared between patients with and without PTS using repeated measures analysis of variance with an orthogonal polynomial transformation to account for uneven spacing between time point measures (Proc Glm procedure, SAS Statistical Software Release 9.1, 2002; SAS Institute Inc., Cary, NC, USA). Differences in mean QOL scores by category of severity of PTS were assessed using analysis of variance. For all analyses, a P-value of < 0.05 was considered to represent a statistically significant difference.
Potential determinants of QOL were evaluated using multivariable linear regression analyses. In separate models, the dependent variable was the 2-year score for each of the QOL measures (SF-36 PCS, SF-36 MCS, VEINES-QOL, VEINES-Sym); if 2-year QOL scores were not available, 12-month scores were substituted. The independent variable of interest for all models was PTS. Relevant covariates, including those found to be associated with baseline QOL on bivariate analysis, were also included in the initial models. A backward elimination procedure (P < 0.05 to remain in model) was used to select final models constructed of variables that were independently predictive of QOL. To adjust for the effects of baseline differences in QOL, baseline QOL scores were included in all models. Multivariable models were constructed in SAS (SAS Statistical Software Release 9.1).
On the basis of previous work that reported QOL scores in patients with chronic venous disease , we estimated that the expected mean difference in SF-36 PCS scores between patients with and without PTS in our sample would be at least 5 points, where a difference of 3–4 points is considered to be clinically meaningful to patients . Assuming a 20% incidence of PTS, we required 250 patients to detect a 3-point difference in mean QOL scores with 80% power and a two-sided alpha of 0.05.
In total, 1032 potentially eligible patients were screened for inclusion in the study cohort. Of these, 645 were excluded for the following reasons: unable to respond to a questionnaire in English or French (n = 138); estimated life-expectancy < 3 months (n = 116); geographically inaccessible for follow-up (n = 80); unable to provide informed consent (e.g. cognitive impairment) (n = 184); refused to provide informed consent (n = 124); and unspecified reason (n = 3). As compared with the 387 patients who were enrolled in the study, those excluded were older (mean age 65 vs. 56 years, P = 0.001), and more likely to be female (59% vs. 51%, P = 0.03) and to have proximal rather than distal DVT (70% vs. 60%, P = 0.02). Nineteen per cent of the 387 study patients had previous DVT; additional baseline characteristics are shown in Table 1.
Table 1. Baseline characteristics of study participants, n = 387*
*n (%) unless indicated otherwise.
†Cancer that was diagnosed within last 6 months or is being treated, metastatic or terminal.
‡Transient risk factors defined as surgery, trauma or immobilization for ≥ 3 days within the last 3 months.
§No cancer or transient risk factors.
Highest level of education completed
College or university
Type of employment
Body mass index (kg m−2): mean (SD)
Characteristics of deep vein thrombosis
Transient risk factors‡
Concurrent symptomatic pulmonary embolism
Previous deep vein thrombosis
During follow-up, 37 patients died and 31 patients developed recurrent venous thromboembolism. Three hundred and sixty-eight patients attended the 1-month visit, 347 attended the 4-month visit, 318 attended the 8-month visit, 302 attended the 12-month visit, and 260 attended the 24-month visit. The mean duration of heparin treatment was 7.4 [standard deviation (SD) 6.0] days. Warfarin was administered to study participants for a mean (SD) of 23.2 (10.9) weeks. Overall, 52% of patients reported the use of compression stockings during follow-up.
Incidence and severity of PTS
Three hundred and forty-four of the 387 study patients had sufficient follow-up data to be classified (as described above) as having or not having developed PTS during the 2-year study follow-up period; of these, 155 developed PTS, for a cumulative risk of 47%. The cumulative risk of PTS was lower in patients with distal DVT than proximal DVT (41% vs. 52%; P = 0.03). Among patients with PTS, 87 (57%) were classified as having mild, 43 (27%) as having moderate and 25 (16%) as having severe PTS.
Baseline quality of life Baseline QOL scores in the study population are shown in Table 2. For all QOL measures, women had significantly lower scores than men. SF-36 (generic QOL) scores were significantly lower among inpatients and patients with comorbidity (defined as active cancer, cardiac conditions such as angina or myocardial infarction, or respiratory conditions such as obstructive lung disease or pneumonia) than among outpatients and patients without comorbidity. Patients with higher body mass index (BMI) had significantly lower SF-36 PCS, VEINES-QOL and VEINES-Sym scores than patients with lower BMI. Baseline QOL scores were not influenced by whether DVT was the first episode or recurrent, or accompanied by symptomatic pulmonary embolism.
Table 2. Baseline quality of life (QOL) by patient characteristics*
Generic quality of life
Venous disease-specific quality of life
For all QOL measures, lower scores signify poorer quality of life.
P-value for two-sample t-test or anova.
SF-36, Short-Form Health Survey-36; PCS, Physical Component Summary score; MCS, Mental Component Summary score; VEINES, Venous Insufficiency Epidemiological and Economic Study; Sym, symptoms; DVT, deep vein thrombosis; PE, pulmonary embolism.
*The population consists of the 344 patients who were able to be classified as having, or not having, developed post-thrombotic syndrome during study follow-up. Generic QOL data available for 337 patients; VEINES-QOL/Sym data available for 342 patients.
†Defined as active cancer, cardiac disease, or respiratory disease.
Body mass index (kg m−2)
No concurrent PE
Change in QOL during follow-up Overall, QOL tended to improve over time. From the baseline to 24-month visit, mean SF-36 PCS scores improved by 6.5 points [95% confidence interval (CI): 4.9–8.0], SF-36 MCS scores improved by 6.2 points (95% CI: 4.8–7.6), VEINES-QOL scores improved by 4.9 points (95% CI: 4.2–5.7) and VEINES-Sym scores improved by 3.6 points (95% CI: 2.8–4.3) (P < 0.0001 for all four QOL measures). However, the magnitude of improvement in PCS, VEINES-QOL and VEINES-Sym scores during study follow-up was significantly less among patients who developed PTS than among those without PTS (Table 3). Furthermore, as seen in Fig. 1, for all QOL measures at all time points (with the exception of SF-36 MCS scores at baseline and at 24 months), patients with PTS had lower QOL scores than those without PTS, and there was evidence for interaction between PTS status and time such that for all measures, improvement in QOL scores during follow-up occurred more slowly in patients with PTS (P-value for PTS*time interaction 0.001, 0.012, 0.014 and 0.006 for PCS, MCS, VEINES-QOL and VEINES-Sym, respectively). Figure 1 also highlights that, overall, QOL may initially not improve or may even worsen in the first month after DVT, and that most of the subsequent improvement in QOL occurs between 1 and 4 months after DVT.
Table 3. Change in quality of life (QOL) from baseline to 2 years according to post-thrombotic syndrome (PTS) status
Subjects with PTS
Subjects without PTS
Mean (SD) change score
Mean (SD) change score
A positive change score indicates improvement in QOL. When 24-month data were missing, 12-month data were substituted (n = 39 observations).
SF-36, Short-Form Health Survey-36; PCS, Physical Component Summary score; MCS, Mental Component Summary score; VEINES, Venous Insufficiency Epidemiological and Economic Study; Sym, symptoms; NS, not significant.
Generic quality of life
Venous disease-specific QOL
QOL and PTS severity As seen in Table 4, SF-36 PCS, SF-36 MCS, VEINES-QOL and VEINES-Sym scores decreased in a graded fashion with increasing clinical severity of PTS (P < 0.001 for trend for all measures).
Table 4. Effect of post-thrombotic syndrome (PTS) severity on 2-year quality of life (QOL) scores
When 24-month data were missing, 12-month data were substituted (n = 39 observations).
PTS severity classification: a Villalta score of < 5 at all assessments was considered to represent no PTS, a score of 5–9 at all assessments was considered to represent mild PTS, a score of 10–14 on at least one assessment was considered to represent moderate PTS, and a score of ≥ 15 or a venous ulcer was considered to represent severe PTS.
Predictors of QOL at 2-year follow-up The results of the multivariable analyses are shown in Table 5. PTS, age, proximal DVT and being an inpatient at time of DVT diagnosis were independently predictive of 2-year SF-36 PCS scores; adjusted scores in patients with PTS were, on average, 7.1 points lower than in patients without PTS. Age, but not PTS, independently predicted 2-year MCS scores; adjusted scores increased, on average, by 1 point per 10-year increase in age. PTS was the only independent predictor of 2-year VEINES-QOL scores and VEINES-Sym scores; adjusted scores in patients with PTS were, on average, 4.4 points and 5.2 points lower, respectively, than in patients without PTS. Gender, BMI, concurrent pulmonary embolism at time of DVT diagnosis, history of previous DVT or recurrent DVT during follow-up did not independently influence 2-year scores for any QOL measure.
Table 5. Independent predictors of quality of life (QOL) at 2 years after deep vein thrombosis (DVT)
When 24-month data were missing, 12-month data were substituted (n = 39 observations).
*Variables examined in models: post-thrombotic syndrome (primary variable of interest; results for post-thrombotic syndrome shown in the table for all QOL measures, regardless of statistical significance), age, gender, body mass index, inpatient vs. outpatient, comorbid conditions at baseline (defined as active cancer, cardiac disease, or respiratory disease), extent of DVT (proximal vs. distal), concurrent pulmonary embolism, previous history of DVT, and recurrent DVT during follow-up. Baseline scores for respective QOL measures were included in all models.
Age (per year)
Proximal (vs. distal) DVT
Inpatient (vs. outpatient) at time of DVT diagnosis
Age (per year)
We evaluated health-related QOL during the 2 years following an episode of acute DVT in a cohort of patients recruited at eight hospital centers in Canada. We found that, on average, both generic and venous disease-specific QOL scores tended to improve during follow-up and that most of the improvement in QOL occurred between 1 and 4 months after DVT. However, patients who developed PTS, as compared with those who did not, had poorer QOL at all study visits, less overall improvement in QOL, and slower improvement in QOL over time. We identified that PTS was the predominant determinant of generic physical and venous disease-specific QOL at 2 years. Additional determinants of generic physical health-related QOL were age, proximal DVT, and being an inpatient at time of diagnosis, whereas the sole identified determinant of generic mental health-related QOL in our population was age.
Our findings are of direct importance to patients. Among study patients who did not have PTS, mean SF-36 PCS scores at 2 years were similar to population norms for the general Canadian adult population  standardized to the study cohort’s distribution of gender and age deciles . Hence, on average, patients with DVT who do not develop PTS can expect that at 2 years, their QOL will improve to a level similar to that of the general population. In contrast, among study patients who developed PTS, average 2-year SF-36 PCS scores were lower than that reported for US patients of similar age with arthritis, chronic lung disease, hearing impairment, or diabetes. Furthermore, patients with severe PTS had scores that were comparable to published norms for persons with angina, cancer, or congestive heart failure .
Population norms for US and Canadian populations have established that, on average, SF-36 PCS scores worsen with older age, whereas SF-36 MCS scores do not [16,19]. In our cohort of patients with DVT, a condition for which published norms are not available, we found that increasing age independently predicted modestly lower PCS scores and modestly higher MCS scores at 2 years. We previously reported that in our cohort, proximal DVT affecting the common femoral or iliac veins was an independent predictor of PTS . In the present analysis, it is of interest that, independently of PTS, proximal DVT was predictive of generic physical QOL at 2 years. This might be explained by a higher frequency of post-thrombotic symptoms in such patients, even when diagnostic criteria for PTS are not fully met, or may relate to a higher incidence of active cancer or other comorbidities that could affect QOL in patients with proximal as compared with distal DVT.
Our study extends previous work performed by our group to develop, validate and use the VEINES-QOL/Sym to measure disease-specific QOL in patients with venous disease, including DVT and PTS [8,9,18,21]. In the present study, the differences in VEINES-QOL and VEINES-Sym scores between patients with and without PTS, and among categories of severity of PTS, were of a magnitude previously reported to be clinically meaningful .
We prospectively evaluated the long-term effects of DVT on QOL. Although two previous studies have assessed QOL longitudinally after DVT, follow-up time was limited to the first few months. In the first study, a randomized trial that compared home-administered low molecular weight heparin with hospital-administered intravenous heparin, QOL tended to improve over time but characteristics of patients whose QOL remained poor during follow-up were not reported and a validated disease-specific QOL measure was not used . In the second study, an earlier analysis performed by our group on a subsample of the population reported in this article, we found that the main predictor of lack of improvement in QOL during the first 4 months following DVT was worsening of venous symptoms and signs over the same time period . Other studies that have reported on QOL in patients with DVT have had several limitations that impact on the interpretation of their results, such as small numbers of patients, retrospective data collection, no baseline evaluation, no adjustment for covariates that could independently influence QOL, lack of use of a disease-specific QOL measure, or QOL assessment limited to the acute phase only [21,23–28].
Our study has a number of strengths. We evaluated QOL at multiple time points from the time of DVT diagnosis, using measures that have been shown to be reliable, valid, and responsive to change [8,16]. In order to comprehensively assess QOL, we used both generic and disease-specific measures [5,7]. We enrolled consecutive eligible and consenting patients, study inclusion criteria were broad, and the incidence of PTS in our study population was similar to that of previous reports [14,29]; hence, our results are likely to be generalizable to other populations of DVT patients. Reasons for non-participation and clinical characteristics of patients who declined participation were documented, so the source population for the study is well defined. To avoid bias in the assessment of PTS, patients completed QOL questionnaires independently from the clinical evaluation for PTS. Finally, we collected data on a number of patient characteristics that could independently influence QOL so that these factors could be adequately controlled for in our analyses.
The main limitations of our study relate to the fact that, by design, decisions regarding patient management factors such as the duration of warfarin anticoagulation and the use of elastic compression stockings were left to be made by the treating physicians. Hence, we were not able to evaluate whether these factors independently influenced QOL. Also, we did not have complete follow-up data for all patients, primarily due to deaths or illnesses preventing attendance at some follow-up visits.
In conclusion, we found that in patients with DVT, the principal factor that influences long-term QOL is whether or not a patient develops PTS, and that older age and proximal DVT also influence QOL. Self-reported physical QOL in patients with PTS is poor, to a degree that is comparable to that of patients with serious chronic diseases such as diabetes, obstructive lung disease, and congestive heart failure. As PTS is a frequent complication of DVT, trials evaluating treatments for DVT should include assessment of QOL in conjunction with more traditional clinical endpoints. Our study provides valuable prognostic information on patient-important health outcomes after DVT and highlights the need for effective prevention and treatment of PTS.
Addendum: role of each author
S. R. Kahn: Conception and design, acquisition of data, analysis and interpretation of data, drafting of manuscript, critical revision of manuscript, obtaining funding, administrative support. H. Shbaklo: Statistical analysis, analysis and interpretation of data, administrative support. D. L. Lamping: Conception and design, critical revision of manuscript, administrative support. C. Holcroft: Statistical analysis, analysis and interpretation of data, critical revision of manuscript. I. Shrier: Conception and design, analysis and interpretation of data, critical revision of manuscript. M.-J. Miron: Acquisition of data, critical revision of manuscript, administrative support. A. Roussin: Acquisition of data, critical revision of manuscript, administrative support. S. Desmarais: Acquisition of data, critical revision of manuscript. F. Joyal: Acquisition of data, critical revision of manuscript. J. Kassis: Acquisition of data, critical revision of manuscript, administrative support. S. Solymoss: Acquisition of data, critical revision of manuscript. L. Desjardins (deceased 2006): Acquisition of data, administrative support. M. Johri: Conception and design, critical revision of manuscript. J. S. Ginsberg: Conception and design, acquisition of data, critical revision of manuscript.
We sincerely thank the VETO study personnel at the clinical study sites for their dedication to this project. S. R. Kahn and I. Shrier are recipients of Clinical Investigator Awards from the Fonds de la Recherche en Santé du Québec. M. Johri is a recipient of a New Investigator Award from the Canadian Institutes of Health Research. J. S. Ginsberg is a recipient of a Career Award from the Heart and Stroke Foundation of Ontario and holds the David Braley and Nancy Gordon Chair in Thromboembolic Disease at McMaster University. The VEnous Thrombosis Outcomes (VETO) Study was funded by the Fonds de la Recherche en Santé du Québec and by an unrestricted grant-in-aid from GlaxoSmithKline. The study sponsors had no role in the design and conduct of the study, the collection, management, analysis, interpretation and reporting of study data, the writing of the manuscript or the decision to submit the manuscript for publication.
Disclosure of Conflict of Interests
The authors state that they have no conflict of interest.