Effect of Physical Activity after Recent Deep Venous Thrombosis: A Cohort Study

Ian Shrier; Susan R. Kahn


Med Sci Sports Exerc. 2005;37(4):630-634. 

In This Article


The details of this study have been published.[9] From April 2001 to July 2002, we prospectively screened consecutive patients with a new diagnosis of acute DVT at seven hospitals in Quebec, Canada. Inclusion criteria for the study were 1) objectively diagnosed acute DVT of the lower limb, and 2) ability to read and understand French or English. To avoid restricting the generalizability of the study, all patients meeting the above inclusion criteria were potentially eligible, regardless of the location of DVT, the coprevalence of pulmonary embolism, the type of anticoagulation given, or whether the treatment setting was in the hospital or at home. We excluded patients whose geographic inaccessibility prevented return for the follow-up visits, patients with an expected lifespan of less than 1 yr, and patients who were unable or refused to provide informed consent. The ethics committees of the participating hospitals approved the study.

A DVT was considered to be objectively confirmed if the venous duplex study showed a lack of compressibility or intraluminal thrombus in the affected vein,[11] or if the contrast venogram showed a constant intraluminal filling defect in at least two different projections.[15] The sensitivity and specificity of duplex has been examined against the contrast venogram gold standard in one of the participating institutions and found to be 98.7% and 100%, respectively, for proximal vein DVT, and 94% and 99%, respectively, for calf vein DVT.[16] These figures are similar to other reports.[11]

Patients were seen at baseline (time of diagnosis) and at 1 and 4 months post-DVT. We recorded the following independent variables at baseline: age, sex, site of DVT, symptoms at presentation, disease severity using the Venous Insufficiency Epidemiological and Economic Study-Quality of Life (VEINES-QOL) Questionnaire,[13] and pre-DVT habitual physical activity using the Godin Questionnaire.[6] At the 1- and 4-month follow-up visits, we again recorded disease severity and habitual physical activity.

The VEINES-QOL is a validated 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 and psychological impact of chronic venous disease, change over the past year, and the time of day that the leg problem is most intense. All variables are measured using a Likert scale. Both English and French Canadian language versions of the VEINES-QOL have been comprehensively evaluated and shown to be acceptable, reliable, valid, and responsive.[10] For all measures, lower scores indicate poorer quality of life. The Godin Questionnaire is a validated measure of leisure-time physical activity over a typical 7-d period.[6] It is scored as a continuous outcome based on a weighted average of frequency of physical activity per week in strenuous (e.g., jogging, vigorous swimming), moderate (e.g., fast walking, volleyball), and light activities (e.g., yoga, golf). Higher scores indicate increased physical activity.

Our main study outcome was the change in score on the postthrombotic syndrome (PTS) scale between the 1- and the 4-month visits. The PTS scale is a validated clinical scale that grades the severity, from 0 to 3, of five symptoms (pain, cramps, heaviness, pruritus, and paresthesia) and six clinician-observed signs (edema, skin induration, hyperpigmentation, venous ectasia, redness, and pain during calf compression) usually associated with the sequelae of a DVT.[23] Points are summed into a total score. A summative score of 5 or higher is classified as the presence of the postthrombotic syndrome. A score of 15 or higher, or the presence of a venous ulcer, is classified as severe post thrombotic syndrome. The interobserver (physician/nurse) reliability for the individual clinical sign components of the PTS scale in a pilot project in our institution were moderate to high range (kappa ranging from 0.59 to 0.84), and the total scores were highly correlated (Pearson correlation coefficient 0.88, P < 0.0001).

In addition, at the end of the 4-month follow-up, we asked the patient to estimate whether their overall physical activity level was increased, decreased, or unchanged compared with their pre-DVT physical activity level. If physical activity was decreased, we asked the patient to list the main reason (no time, no interest, leg problem, or other).

We describe all continuous variables as mean ± SD if normally distributed, and as median with interquartile ranges (IQR) if there is a skewed distribution. If the patient had bilateral DVT, we considered the side with the more proximal DVT to be the affected side. If both DVT were distal or proximal, we arbitrarily considered the right leg to be the affected side.

In a univariate analysis, we used the chi-square test to determine whether habitual physical activity level at 1 month affected the change in PTS score between 1 and 4 months. We categorized the continuous main independent variable (habitual physical activity at 1 month by Godin Questionnaire) into tertiles based on the 1-month activity levels, because this was the main exposure of interest (0 = no physical activity, 1—20 = mildly to moderately active, > 20 = highly active). For reference, a person involved in 15 min of fast walking 3 × wk-1 would score 15 (mild to moderate physical activity), and someone involved in 15 min of jogging (strenuous exercise) 2 × wk-1 and 15 min of fast walking (moderate exercise) 1 × wk-1 would score 23 (high physical activity). We a priori categorized the main outcome variable PTS score (score at 4 months post-DVT minus score at 1 month post-DVT) as worse if PTS > 0, unchanged if PTS = 0, and improved if PTS < 0. An increase in PTS of 1 unit (i.e., PTS > 0) correlates with approximately a decrease of 4 units in the VEINES-QOL score,[10] and 4 units in a quality of life score is generally considered an important clinically relevant difference. To assess the robustness of the results, we also analyzed the data with unchanged category redefined as -1 ≤ PTS ≤ 1.

Next, we used multiple regression analyses to adjust for the potential confounding effects of age, sex, and pre-DVT physical activity level on PTS score. For this analysis, we dichotomized the outcome. Because we were interested in whether exercise worsened symptoms, and should therefore be avoided, we a priori considered a patient's condition to have worsened if PTS score was > 1, and improved-unchanged if the PTS score was ≤ 1. As a sensitivity analysis, we also categorized the main outcome variable as worse if the PTS was > 0, and improved/unchanged if the PTS was ≤ 0. To control for confounding by indication (i.e., those individuals who are doing well might be more likely to be active by 1 month), we also included in the model the 1-month VEINES-QOL score as a measure of disease severity.

Data analysis was performed using Statview (SAS institute Inc. Cary, NC) computerized software.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: