Thrombosis Research (2008) 122, 763–773

www.elsevier.com/locate/thromres

REVIEW ARTICLE

Physical activity in patients with deep venous

thrombosis: A systematic review ☆
Susan R. Kahn a,⁎, Ian Shrier a , Clive Kearon b

a
Center for Clinical Epidemiology & Community Studies, Sir Mortimer B. Davis Jewish General Hospital,
McGill University, Montreal, Quebec, Canada
b
McMaster University and the Henderson Research Centre, Hamilton, Ontario, Canada

Received 20 September 2007; received in revised form 19 October 2007; accepted 23 October 2007
Available online 21 December 2007

KEYWORDS Abstract
Deep venous
thrombosis; Objectives: We performed a systematic review to assess the benefits or risks of
physical activity in patients with an acute or previous DVT of the leg.
Postthrombotic
Data sources: PubMed, EMBASE and Science Citation Index were searched without
syndrome;
language restrictions up to July 2007. Bibliographies of retrieved articles and per-
Exercise;
sonal files were also searched.
Walking;
Review methods: Randomized trials and prospective cohort studies that included
Training
patients with acute or previous DVT, described an exercise intervention or exercise
exposure, and described any related clinical outcome were selected. Data were
independently extracted by 2 investigators.
Results: Seven randomized trials and two prospective observational studies were
included. Early exercise, compared with bed rest, was associated with a similar short-
term risk of pulmonary embolism in patients with acute DVT and led to more rapid
resolution of limb pain. In patients with acute DVT, a 6 month daily walking program
led to similar degrees of vein recanalization and improvement in quality of life as
controls. In patients with previous DVT, 30 min of vigorous treadmill exercise did not
worsen venous symptoms and improved calf muscle flexibility; a 6 month exercise
training program improved calf muscle strength and pump function; and high levels of
physical activity at one month tended to be associated with reduced severity of
postthrombotic symptoms during the subsequent 3 months.

☆
Grant support and acknowledgements: Dr. Kahn and Dr. Shrier are recipients of Clinical Research Scientist Awards from Fonds de la
Recherche en Santé du Québec. Dr. Kearon is an Investigator of the Canadian Institutes for Health Research.
⁎ Corresponding author. Department of Medicine, McGill University, and Center for Clinical Epidemiology & Community Studies, Sir
Mortimer B. Davis Jewish General Hospital, 3755 Cote Ste. Catherine, Room A-127, Montreal, Quebec, Canada H3T 1E2. Tel.: +1 514 340
8222x7587; fax: +1 514 340 7564.
E-mail address: susan.kahn@mcgill.ca (S.R. Kahn).

0049-3848/$ - see front matter © 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.thromres.2007.10.011
764 S.R. Kahn et al.

Conclusions: Early walking exercise is safe in patients with acute DVT and may help
to reduce acute symptoms. Exercise training does not increase leg symptoms acutely
in patients with a previous DVT and may help to prevent or improve the postthrom-
botic syndrome.
© 2007 Elsevier Ltd. All rights reserved.

Contents

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 764
Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 764
Data sources and searches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 764
Study selection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 764
Data extraction and quality assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 765
Data synthesis and analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 765
Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 770
Study identification and selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 770
Study quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 770
Study characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 770
Exercise intervention or exposure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 770
Early walking versus bed rest for acute deep venous thrombosis . . . . . . . . . . . . . . . . . . 770
Effects of short session exercise in patients with previous deep venous thrombosis . . . . . . 771
Effects of longer term exercise in patients with recent or previous deep venous thrombosis . . 771
Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 771
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 772

Introduction systematic review of studies that have reported on

Deep venous thrombosis of the leg is a common
clinical problem that affects more than 250,000
people in the U.S. and 25,000 people in Canada each
year [1]. Despite receiving optimal anticoagulant
treatment, symptoms of acute deep venous throm-
bosis such as leg swelling or pain can take weeks to
subside [2–5]. Subsequently, up to 40% of patients
with deep venous thrombosis develop the post-
thrombotic syndrome [6]. Symptoms of postthrom-
botic syndrome include chronic pain, heaviness,
swelling and cramping in the leg which are often
aggravated by standing and are lessened by leg
elevation or lying down [6].
There is clear evidence that regular physical
activity contributes to the primary and secondary
prevention of several chronic diseases such as car-
diovascular disease, diabetes and osteoporosis, is
associated with a reduced risk of premature death
and improves quality of life [7]. In addition, many
patients with deep venous thrombosis or postthrom-
botic syndrome are eager to resume physical activ-
ity and seek information about when an exercise
program can be restarted and at what intensity.
In order to evaluate the clinical effects of exercise
in patients with venous thrombosis, we performed a
the benefits or risks of physical activity in patients
with lower extremity deep venous thrombosis.
Methods
We prospectively developed a protocol that defined objectives of
this review, a search strategy for study identification, criteria for
study selection, how data would be extracted, study outcomes,
and statistical methodology.
Data sources and searches
We performed a computerized search of English-language
publications listed in the electronic databases PubMed, EMBASE
and Science Citation Index up to July 2007 using the following
text and key words in combination both as MeSH terms and text
words: [“thromboembolism”, “venous thrombosis”, “pulmonary
embolism”, “deep vein thrombosis”, “deep-vein thrombosis”,
“postthrombotic syndrome”, “post-thrombotic syndrome”,
“postphlebitic syndrome”, “post-phlebitic syndrome”, “venous
insufficiency”] and [ “physical activity”, “walk”, “jog”, “run”,
“exercise”] and [“randomized controlled trial”, “cohort stu-
dies”, “prospective studies”, “clinical trial”]. We also hand
searched bibliographies of retrieved articles and our own files to
identify additional potentially relevant articles.
Study selection
Two investigators (SK, IS) assessed articles for eligibility. If the
title and the abstract were judged to be potentially eligible,
Exercise and venous thrombosis 765

published in duplicate, in which case the article reporting the

results in greatest detail was included.

Data extraction and quality assessment

Two investigators (SK, IS) independently extracted data using

Figure 1 Identification of eligible studies. DVT= deep-vein
thrombosis; PTS = postthrombotic syndrome.
both reviewers assessed the complete study manuscript. Studies
were included for review if: 1) patients had objectively
diagnosed deep-vein thrombosis; 2) an exercise intervention or
exposure was described; and 3) a related clinical outcome was
described. Studies that satisfied all inclusion criteria were
excluded if the study was not prospective or if results had been
a standardized form. Data were extracted for study
design, study quality, number, sex, and age of enrolled patients,
type and duration of exercise intervention or exposure, dura-
tion of follow-up, and the effect of exercise on outcome
measures.
We assessed the methodological quality of included studies
using criteria adapted from guidelines for the evaluation of
articles on therapy and on prognosis [8]. The criteria for
randomized controlled trials were a clear description of valid
randomization; concealment of randomization; intent-to-treat
analysis; patient similarity in treatment and control groups;
outcome assessors blinded to group allocation; and follow-up of
at least 90%. The criteria for observational studies of exercise
exposure were enrolment of unselected patients with venous
thromboembolism; a clearly defined inception cohort (i.e.,
patients enrolled at the time of the initial diagnosis or at a
uniform time-point after diagnosis); and follow-up of at least
90%.
Data synthesis and analysis
When more than one similar randomized trial addressed a ques-
tion, we calculated the pooled relative risk (95% confidence

Table 1 Methodological quality of eligible studies

Study Randomization Concealment of Patient similarity in Outcome assessors Intent- Completeness
Randomization treatment and blinded to group to-treat of follow-up
control groups allocation analysis
A. Randomized controlled trials
Schellong, Yes Not stated Yes Yes Not 97%
1999 [12] stated
Aschwanden Yes Not stated Yes Probably not Probably 100%
2001 [11] yes
Jünger, 2006 Yes Yes Yes “where possible” Yes 99%
[14]
Blättler, Yes Not stated Yes Yes, for some No 100%/70%b
2003 [13] endpointsa
Partsch, Yes Not stated Yes Yes, for one endpointc No 70%
2004 [15]
Kahn, 2003 Yes Yes Yes Partial Yes 100%
[17]
Padberg, Yes Yes Yes Probably not Yes 93%
2004 [20]
Isma, 2007 Yes Not stated Yes Yes, for primary No 93%
[19] endpointd

B. Observational studies
Consecutive patients Clearly defined inception cohort % Follow-up
Kahn, 2003 [16] Yes Yes 100%
Shrier, 2005 [18] Yes Yes 100%
a
Pulmonary embolism and thrombus extension.
b
For endpoint thrombus extension.
c
Ultrasound examination.
d
Phlebographic score.

Table 2 Studies of exercise after deep venous thrombosis
Author, year Study Patients
design
Studies of early walking vs. bed rest in acute deep venous thrombosis
Schellong, 1999 [12] Single center RCT 126 patients with acute
symptomatic proximal deep
venous thrombosis
Females: 35%
Mean age: 60 years
Patients with symptomatic PE
excluded
Asymptomatic PE at
baseline a: 66%
Aschwanden 2001 [11] Single center RCT 129 patients with acute
proximal deep venous
thrombosis
Females: 44%
Mean age: 65 years
Symptomatic PE at baseline:
24%
Asymptomatic PE at
baseline a: 27%
Jünger, 2006 [14] Multicenter RCT 103 patients with acute
proximal deep venous
thrombosis (1 lost to
follow-up
Females: 44%
Mean age: 60 years
Patients with symptomatic PE
were excluded.
Asymptomatic PE at
baseline a: 49%
766
Intervention/exposure Outcomes assessed Duration of follow-up Results Comments
2 days of leg Perfusion defect in 10 days New PE on V/Q at 10 days
elevation followed previously normal Ambulation: 14/63 (22%)
by free ambulation region on ventilation­ Bed rest: 10/59 (17%)
for 6 days (n = 64) perfusion lung scan, RR 1.31; 95%CI 0.63, 2.72
vs. compared between
Strict bed rest for baseline and day 10 Deaths: 0
8 days (n = 62)
Death by day 10
All patients were
treated in hospital
and received
compression
bandages or
graduated
compression
stockings
Supervised New high probability 4 days At 4 days All episodes of PE
ambulation for ≥ 4 perfusion defects on PE on V/Q scan: at 4 days were
hours per day for lung scan, baseline to Ambulation: 10/69 (14%) asymptomatic
first 4 days plus day 4 Bed rest: 6/60 (10%)
compression therapy RR 1.45; 95%CI 0.56, 3.75 12 out of 16 PE
(n = 69) Change in leg occurred in patients
vs. circumference, leg pain Leg circumference, leg pain: who had PE at
strict bed rest for by VAS, baseline to Both groups improved to same baseline.
first 4 days without day 4 degree
compression therapy
(n = 60) Death at day 4 Death: 0
3 months
All patients were Recurrent VTE, death At 3 months
treated in hospital at 3 months Recurrent VTE:
Ambulation: 2 (3%)
Bed rest: 1 (2%)
Deaths: All deaths occurred
Ambulation: 3 (4%) in patients with
Bed rest: 2 (3%) active cancer.
Free movement Primary combined 12 days Primary endpoint:
around the ward endpoint (any one of Ambulation: 7/52 (13.5%)
(n = 52) of the following): Bed rest: 14/50 (28.0%)
vs. objectively proven new RR 0.48; 95%CI 0.21, 1.09
Strict bed rest for at symptomatic PE,
least 5 days (n = 50) asymptomatic PE, leg Symptomatic PE:
thrombus progression, Ambulation: 1/52 (1.9%)
Patients were nosocomial infection, Bed rest: 5/50 (10.0%)
hospitalized for at other adverse events or RR 0.19; 95%CI 0.02, 1.59
least 6 days death
S.R. Kahn et al.
Thrombus progression or
All patients received Individual components new thrombus on
thigh length of primary endpoint ultrasound:
compression Ambulation: 4/52 (7.7%)
bandages Leg pain (VAS) Bed rest: 10/50 (20%)
RR 0.38; 95%CI 0.13, 1.15
Deaths: 0

Blättler, 2003 [13] Single center RCT 53 patients with acute
proximal deep venous
thrombosis
Patients characteristics not
provided
Partsch, 2004 [15] Extended follow-up of above RCT Same patients as above but
n = 37
Studies of short session exercise in patients with deep venous thrombosis
Kahn, 2003 [16] Single center prospective cohort 41 patients with deep venous
thrombosis (71% proximal,
29% distal) at least 1 year
prior; 19 had postthrombotic
syndrome
Females: 41%
Mean age: 51 years
Mean time since deep venous
thrombosis: 2.2 years
Exercise and venous thrombosis
Leg pain: improved over 12
days to a similar degree in
both groups.
Walking for 9 days New perfusion defects 9 days PE (all asymptomatic) Thrombus
and inelastic on lung scan, baseline Ambulation + bandages: 1/18 progression only
compression to day 9 Ambulation + compression assessable in 37
bandages (n = 18) socks: 1/18 patients whose
vs. Thrombus progression Bed rest: 1/17 initial DVT was
Walking for 9 days on ultrasound RR 0.94; 95%CI 0.09, 9.71 below the inguinal
and elastic ligament
compression Other outcomes: Thrombus progression
stockings (n = 18) QOL Combined ambulation/
vs. Leg pain (VAS) compression groups: 6/27
Bed rest for 9 days Calf pain during (22%)
(n = 17) pressure cuff inflation Bed rest: 4/10 (40%)
Leg circumference RR 0.56; 95%CI 0.20, 1.57
Walking started the Clinical status assessed
day of diagnosis and by physician QOL improved more (p b 0.05)
was progressively and limb pain decreased more
increased to an (p b 0.01) with ambulation +
average of 4 km/day compression than with bed
rest
Data on other outcomes
presented in Figures in paper.
Same interventions Postthrombotic 2 years % with post thrombotic 16/53 patients from
as above syndrome, diagnosed syndrome (Villalta score N4) original trial lost to
with Villalta scale Ambulation: 14/26 (54%) follow-up, with
Bed rest: 9/11 (82%) equal losses among
RR 0.66; 95%CI 0.42, 1.03 groups
Median Villalta score:
Ambulation: 5.1 [IQR 3­6.5]
Bed rest: 8 [IQR 6.5­11]
p b 0.01
Treadmill walking Severity of venous 30 minutes After vs. before exercise
or running at a speed symptoms (heaviness, comparisons, affected leg.
that caused aching, swelling, etc.)
breathlessness and Δ symptom ratings on 10-cm
sweating, up to a Leg volume visual analog scale ranged
maximum of 30 from ­ 0.01 to +0.52; p N 0.05
minutes. Calf muscle for all symptoms (all p values
(gastrocnemius and N 0.05)
Mean treadmill soleus) flexibility
speed: 3.2` ± 1.9 Leg volume increased by
miles per hour, 5% 76±110 ml (p b 0.0001)
incline
In patients with post
thrombotic syndrome,
gastrocnemius flexibility
increased by 4.5° (p = 0.003),
soleus flexibility increased by
5.7° (p = 0.001)
767
(continued on next page)
 768
Table
Table 22 (continued)

Author, year Study Patients Intervention/exposure Outcomes assessed Duration of follow-up Results Comments
design

Kahn, 2003 [17] Single center Same patients as above 2 periods of Severity of venous 30 minutes Stocking vs. no stocking
randomized treadmill walking symptoms (headiness, comparisons, affected leg:
cross-over for 30 min (as aching, swelling, etc.)
trial described above Leg volume No differences in severity of
(16)) with, or Calf circumference post-exercise leg symptoms,
without, Calf muscle flexibility leg volume, calf
compression (30 mm Hg) circumference, calf flexibility
stocking (all p values N 0.05)
(order of exercise
randomized)

Studies of longer term exercise in patients with deep venous thrombosis

Padberg, 2004 [20] Single center RCT 30 patients with moderate to 3 months of Measures of calf pump 6 months Mean change in ejection Non-blinded
severe venous insufficiency, supervised exercise function: fraction outcome
half due to prior deep venous followed by 3 Ejection fraction (%) Exercise (n = 15): 3.5% assessment
thrombosis (i.e. post months Residual volume Control (n = 13): −1.4%
thrombotic syndrome) unsupervised fraction (%) p = 0.03 No women
exercise, designed to included
Females: 0% strengthen calf Calf muscle strength: Mean change in residual
Mean age: 70 years muscles and isokinetic peak torque/ volume fraction Only half had prior
Mean time since deep venous increase joint body weight at slow Exercise (n = 15): − 8.8% DVT; results not
thrombosis not stated mobility and fast speed Control (n = 13): 3.4% presented
vs. p = 0.03 sepatately for this group
No exercise control Reflux (venous filling group
group index) Mean change isokinetic peak
Venous severity score torque/body weight
All patients Quality of life Slow speed
prescribed Exercised (n = 16): 3.1
compression Control (n = 13): −1.0
stockings p b 0.05
Fast speed
Exercise (n = 16): 2.8
Control (n = 13): −0.3
p b 0.03

No differences in reflux
(p = 0.64), venous severity
score (p = 0.51) or quality of
life (data not provided)
Isma, 2007 [19] Single center RCT 72 patients with a first episode Daily walking Recanalization of 6 months Median (SD) phleborgraphic Study likely
of acute DVT (68% proximal) exercise (at least 15 occluded veins score at enrollment underpowered

S.R. Kahn et al.

min) plus supervised (phlebographic score; Exercise (n = 32): 9.8 (8.3)
Females: 48% exercise (45 min; range 0–42) Control (n = 35): 10.2 (8.3) Postthrombotic

Mean age: 54 years mostly, lower limb
but also upper limb
and torso) once
weekly for 1 month
then once monthly
beginning 5–7 days
after DVT diagnosis
vs.
No exercise control
group
All patients treated
with routine
anticoagulation and
prescribed
compression
stockings
Shrier 2005 [18] Multicenter 301 patients with acute deep Frequency and
prospective venous thrombosis (50% intensity of leisure
cohort proximal) time physical
activity during a
Females: 50% typical 7 day period
Mean age: 55 years (categorized by
Godin leisure time
questionnaire [32] as
inactive, mild-
moderately or highly
active), assessed 1
month after deep
venous thrombosis
Notes: RCT, randomized controlled trial; VAS, visual analogue scale; PE, pulmonary embolism; VTE, venous thromboembolism; V/Q, ventilation perfusion; IQR, interquartile range; RR, relative risk; OR, odds ratio; CI, confidence interval; SD, standard
deviation.
a
As detected by baseline V/Q scan.
Exercise and venous thrombosis
p N 0.05 syndrome not
Quadriceps muscle Median (SD) phlebographic specifically
strength (method of score at 6 months assessed
testing not provided) Exercise (n = 32): 3.0 (4.9)
Control (n = 35): 1.1 (2.8)
Thigh and calf p N 0.05
circumference (lower phlebographic score
indicates greater
Quality of life (VAS) recanalization)
No differences in quadriceps
muscle strength, change in
thigh and calf circumference,
change in quality of life
between groups (all p N 0.05)
Worsening of Villalta 4 months OR for worsening of post
postthrombotic thrombotic syndrome,
syndrome score by N 1 compared to the inactive
point between 1 and 4 group:
months Mild-moderately active: OR
0.93 (95%CI 0.47, 1.87)
Highly active: 0.52 (95%CI
0.24, 1.15)
(analyses adjusted for age,
sex, habitual physical activity
before deep venous
thrombosis and venous
disease-specific QOL at 1
month)
769
770
intervals) using a random effects model with Comprehensive
Meta-Analysis software, version 2.2.034 [9]. Heterogeneity of
study findings was assessed using the I2 statistic, where a value
greater than 50% is considered to indicate substantial hetero-
geneity [10]. A two-side probability of less than 0.05 was deemed
statistically significant for all analyses. When only one rando-
mized controlled trial addressed a question, or if studies were
dissimilar in terms of patients, methodologies, interventions and
outcomes, study results were not combined but were reported
descriptively.
Results
Study identification and selection
Using the predefined search strategy, 653 titles were retrieved
and scanned, with subsequent review of 41 abstracts and 19 full
manuscripts (Fig. 1). Of these, 10 studies were eligible for
inclusion in the review [11–20] (list of ineligible studies and
reasons are available from the authors).
Study quality
The methodological quality of the included studies was moderate
to good, with most randomized trials satisfying at least 4 of 6
quality criteria, and all observational studies satisfying all 3
criteria (Table 1).
Study characteristics
Detailed characteristics of the study design, patient population,
exercise intervention or exposure, outcomes assessed and results
of individual studies are outlined in Table 2.
Four were randomized trials of the short-term effects of
walking compared with bed rest early after diagnosis and
initiation of anticoagulation for acute deep venous thrombosis
[11–14]; one was a follow-up of one of the previously noted
studies [13] that evaluated if early ambulation influenced devel-
opment of postthrombotic syndrome after 2 years [15]; one study
evaluated if treadmill exercise had an acute influence on limb
symptoms and function in patients with previous deep venous
thrombosis [16]; one study was a randomized cross-over trial that
evaluated if wearing elastic compression stockings during
Figure 2 Difference in risk of pulmonary embolism in patients with deep venous thrombosis assigned to early walking versus
bed rest. Tests of heterogeneity: q-value = 3.129, df (Q) = 3, p = 0.372; I2 = 4.13%.
S.R. Kahn et al.
exercise in patients with previous deep venous thrombosis
influenced acute symptoms, leg swelling or joint flexibility
[17]; one was a cohort study that evaluated the relationship
between self-reported habitual physical activity levels one
month after a diagnosis of deep venous thrombosis and severity
of postthrombotic symptoms and signs in the subsequent 3 months
[18]; one was a randomized trial that assessed the effects of a 6
month daily walking program, initiated early after DVT, on deep-
vein recanalization, leg circumference and quality of life [19];
and one was a randomized trial that evaluated the effect of a
structured exercise program in patients with chronic venous
insufficiency (half had prior deep venous thrombosis) on
measures of venous function, calf muscle function, leg symptoms
and quality of life [20].
Because there were four randomized controlled trials that
assessed if early walking influenced the incidence of pulmonary
embolism, observational studies that addressed this issue were not
considered further.
Exercise intervention or exposure
Early walking versus bed rest for acute deep
venous thrombosis
Four RCTs have addressed the effects of early walking compared
with a period of bed rest in patients with acute deep venous
thrombosis. In two of the trials, all patients received early
compression therapy (bandages or stockings) [12,14], while in
two others only patients randomized to early walking received
early compression therapy [11,13]. Outcomes assessed in these
trials included development of new pulmonary embolism (4
studies [11–14]), thrombus extension (2 studies [13,14]) and
change in leg symptoms during short-term follow-up (3 studies
[11,13,14]). In the four trials combined, early walking was
associated with a relative risk of symptomatic or asymptomatic
pulmonary embolism after about 10 days of 1.16 (95% CI 0.66 to
2.05; p = 0.61; I2 = 4.13%) (Fig. 2). In two studies, early walking
did not influence the rate of improvement of acute leg pain
[11,14] while in one study early walking was associated with
greater improvement in acute pain (p b 0.01) and quality of life
(p b 0.05) [13]. The relative risk of thrombus progression asso-
ciated with early walking was 0.38 (95%CI 0.13, 1.15) [14] in one
trial and 0.56 (95%CI 0.20, 1.57) in a second, smaller trial [13].
Exercise and venous thrombosis
During two year follow-up of participants in the latter trial, early
walking was associated with a reduced risk of developing the
postthrombotic syndrome (RR 0.66; 95%CI 0.42, 1.03) [15], which
the authors postulated may have been due to lower rates of early
thrombus progression.
Effects of short session exercise in patients with
previous deep venous thrombosis
A cohort study performed by our group evaluated whether
previous deep venous thrombosis (mean of 2 years earlier)
limited the ability to perform treadmill exercise and whether
exercise increased the severity of venous symptoms and signs
[16]. Nineteen of the 41 subjects studied had the postthrombotic
syndrome. A 30 minute treadmill exercise session did not worsen
venous symptoms such as heaviness, swelling and aching (Δ
symptom ratings on 10-cm visual analog scale ranged from − 0.01
to +0.52; p N 0.05 for all symptoms) and in subjects with the
postthrombotic syndrome, led to significant increases in calf
muscle flexibility (gastrocnemius 4.5°, p = 0.003; soleus 5.7°,
p = 0.001). In a randomized cross-over trial of the same 41 sub-
jects, wearing elastic compression stockings during exercise did
not influence acute symptoms, leg swelling or joint flexibility (all
p N 0.05) [17].
Effects of longer term exercise in patients with
recent or previous deep venous thrombosis
Two randomized trials have evaluated the effects of structured,
long-term exercise in patients with deep venous thrombosis. In the
first trial, 72 patients with acute deep venous thrombosis were
randomized within one week of diagnosis to daily walking exercise
plus periodic physiotherapist-supervised exercise sessions, or no
exercise. At 6 months, degree of recanalization of occluded
venous segments (measured by a venographic scoring system), leg
circumference and quality of life all improved to a similar extent in
both groups (p N 0.05 for all between-group comparisons) [19]. In
the second trial, 30 patients with severe chronic venous
insufficiency, half with previous deep venous thrombosis, were
randomized to 6 months of exercise training designed to strength-
en calf musculature and enhance joint mobility or to no exercise.
Exercise training improved calf muscle strength (p = 0.03) and
pump function (p b 0.03) but did not improve valvular reflux
(p = 0.64), venous clinical severity scores (p = 0.78) or quality of life
(p value not provided) [20].
Finally, in a multicenter prospectively followed cohort of 301
patients with recently diagnosed deep venous thrombosis, higher
levels of self-reported habitual physical activity at one month
tended to be associated with less severe postthrombotic
symptoms during the subsequent three months in a dose–
response fashion, although the findings were not statistically
significant and require confirmation in larger studies. Odds ratios
were 0.93 (95%CI 0.47, 1.87) for patients who reported mild-
moderate activity at one month and 0.52 (95%CI 0.24, 1.15) for
patients who reported high levels of activity at one month [18].
Overall, by four months after the diagnosis of deep venous
thrombosis, more than half of patients were able to resume their
usual levels of physical activity.
Discussion
The overall findings of this review support a positive
role for exercise in patients with acute or previous
771
deep venous thrombosis and can be summarized as
follows.
First, based on the results of four randomized
controlled trials [11–14], there is strong prospective
evidence that early walking does not increase the
risk of pulmonary embolism in the days after diag-
nosis and initiation of anticoagulant therapy for
deep venous thrombosis. A previous systematic re-
view that included three of the four trials found
similar results [21]. Also, high quality evidence from
randomized trials showing that home treatment of
DVT with anticoagulants is effective and no more
liable to complications than hospital treatment
[22,23] further supports the safety of resuming
walking activity early after diagnosis and initiation
of treatment for deep venous thrombosis. While a
retrospective study previously reported that reha-
bilitation patients with deep venous thrombosis who
subsequently developed pulmonary embolism had
returned to active physiotherapy earlier (average
48 h) than those who did not develop pulmonary
embolism (average 123 h), this study had a number
of limitations, including that only 6 patients with
pulmonary embolism were studied and there was no
screening for asymptomatic pulmonary embolism at
the time of study entry [24].
Results of two of the trials reviewed [13,14] also
suggest that early walking may reduce the risk of
extension of proximal deep venous thrombosis
shortly after diagnosis. The trials differed with re-
gard to whether compression stockings were pre-
scribed to all patients [12,14] or only to patients
randomized to early walking [11,13] and in one trial,
early walking was preceded by a two day period of
leg elevation [12]. Hence, it is not certain whether
compression or leg elevation provide additional
benefit to early walking.
Second, based on the findings of three trials, it
appears that early walking either has no effect, or
improves, resolution of acute symptoms [11,13,14].
Two studies [11,14] found no effect of early walking
on symptom resolution whereas one study [13] found
that walking improved general and disease-specific
quality of life and leg pain. However, as the three
studies included a total of only 285 patients, and as
patients, caregivers, and outcome assessors were not
blinded in all three studies, the influence of early
walking on resolution of acute symptoms is uncertain.
Third, based on the findings of one randomized
trial [15], there is a suggestion that early walking
may reduce long-term symptoms of the postthrom-
botic syndrome. However, as this study was small,
was unable to evaluate long-term outcomes in 30%
of the patients, and as assessors of the postthrom-
botic syndrome were not blinded to whether or not
patients were mobilized early, it is uncertain if early
772
mobilization influences the risk of developing the
postthrombotic syndrome.
Fourth, based on comparison of symptoms before
and after 30 min of treadmill walking or running in
one study [16], there is evidence that vigorous ex-
ercise generally does not aggravate leg symptoms in
patients with previous deep venous thrombosis.
However, we cannot exclude that exercise-induced
symptoms differ among patients with previous deep
venous thrombosis and, therefore, exercise may
aggravate symptoms in some patients.
Fifth, based on a small-randomized cross-over
study [17], it appears that graduated compression
stockings have little influence on symptoms during
vigorous exercise.
Sixth, a small RCT suggests that a six-month
exercise program improves leg muscle strength and
venous return but may not influence the clinical
manifestations of postthrombotic syndrome [20],
while another found no effect of a six-month exer-
cise program on vein recanalization or quality of life,
but did not specifically assess the postthrombotic
syndrome [19]. However, a tendency for higher
levels of physical activity one month after deep-vein
thrombosis to be associated with less severe post-
thrombotic symptoms three months later suggests
that increasing physical activity may reduce symp-
toms of the postthrombotic syndrome [18]. As
regular exercise has been shown to be beneficial in
patients with arterial insufficiency [25,26], and as
there may be overlap in the mechanisms of leg
discomfort that occur in patients with chronic
venous and chronic arterial insufficiency, including
hypoxic damage to the calf muscle with resultant
muscle fiber necrosis [27–31], there is the potential
for exercise training to help prevent, or treat, the
postthrombotic syndrome.
We developed and utilized a structured study
protocol to guide our search strategy, study selec-
tion, extraction of data and statistical analysis, and
the quality of included studies was generally good.
However, there are several limitations inherent to
our review. Publication bias is possible, as is language
bias, since we restricted our search to English-
language publications. As there were few randomized
trials, we included some observational studies, the
results of which may be affected by confounding bias
due to the absence of random assignment to exercise.
Also, many of the studies were small and may have
lacked statistical power to show differences, if such
differences were present. Finally, due to the hetero-
geneity of outcomes reported in the studies, our
search strategy focused on an intervention, i.e. exer-
cise, rather than on prespecified outcomes.
In conclusion, current evidence indicates that
early walking may be encouraged in patients with
S.R. Kahn et al.
acute deep-vein thrombosis and that a previous
deep-vein thrombosis is not a contraindication to
regular exercise. Further large, adequately pow-
ered studies are required to determine if more
vigorous exercise training initiated early after deep-
vein thrombosis is safe and perhaps of benefit for
the recovery of the leg, and whether regular exer-
cise can prevent, or be an effective treatment for,
the postthrombotic syndrome. These results suggest
that the potential role of exercise as a treatment
modality for deep venous thrombosis patients
should be explored.
References
[1] Anderson FA, Wheeler B, Goldberg RJ, Hosmer DW, Patwardhan
NA, Jovanovic B, et al. A population-based perspective of the
hospital incidence and case-fatality rates of deep vein
thrombosis and pulmonary embolism. Ann Intern Med
1991;151:933–8.
[2] Kahn SR, Ducruet T, Lamping DL, Arsenault L, Miron MJ,
Roussin A, et al. Prospective evaluation of health-related
quality of life in patients with deep venous thrombosis. Arch
Intern Med 2005;165:1173–8.
[3] Kearon C, Salzman EW, Hirsh J. Epidemiology, pathogenesis
and natural history of venous thrombosis. In: Colman RW,
Hirsh J, Marder VJ, Clowes AW, George JN, editors.
Hemostasis and Thrombosis: Basic principles and Clinical
Practice. 4th edition. Philadelphia, PA: Lippincott, Williams
& Wilkins; 2001. p. 1153–77.
[4] Koopman MM, Prandoni P, Piovella F, Ockelford PA, Brandjes
DP, van Der MJ, et al. Treatment of venous thrombosis with
intravenous unfractionated heparin administered in the
hospital as compared with subcutaneous low-molecular-
weight heparin administered at home. The Tasman Study
Group. N Engl J Med 1996;334:682–7.
[5] Van Korlaar IM, Vossen CY, Rosendaal FR, Bovill EG, Cushman
M, Naud S, et al. The impact of venous thrombosis on quality
of life. Thromb Res 2004;114:11–8.
[6] Kahn SR. The post-thrombotic syndrome: progress and
pitfalls. Br J Haematol 2006;134:357–65.
[7] Warburton DER, Nicol CW, Bredin SSD. Health benefits of
physical activity: the evidence. Can Med Assoc J 2006;174:
801–9.
[8] The Evidence-Based Medicine Working Group. Users' guides
to the medical literature: a manual for evidence-based
clinical practice. Chicago: AMA press; 2002.
[9] Borenstein M, Higgins J, Hedges L, Rothstein H. Comprehen-
sive Meta-analysis Version 2. Englewood NJ: Biostat; 2005.
[10] Cochrane Reviewers' Handbook. Cochrane Collaboration
(ed 4 2 2) 2004.
[11] Aschwanden M, Labs KH, Engel H, Schwob A, Jeanneret C,
Mueller-Brand J, et al. Acute deep vein thrombosis: early
mobilization does not increase the frequency of pulmonary
embolism. Thromb Haemost 2001;85:42–6.
[12] Schellong SM, Schwarz T, Kropp J, Prescher Y, Beuthien-
Baumann B, Daniel WG. Bed rest in deep vein thrombosis
and the incidence of scintigraphic pulmonary embolism.
Thromb Haemost 1999;82(Suppl 1):127–9.
[13] Blattler W, Partsch H. Leg compression and ambulation is
better than bed rest for the treatment of acute deep venous
thrombosis. Int Angiol 2003;22:393–400.
[14] Junger M, Diehm C, Storiko H, Hach-Wunderle V, Heidrich H,
Karasch T, et al. Mobilization versus immobilization in the
Exercise and venous thrombosis
treatment of acute proximal deep venous thrombosis: a
prospective, randomized, open, multicentre trial. Curr Med
Res Opin 2006;22:593–602.
[15] Partsch H, Kaulich M, Mayer W. Immediate mobilisation in
acute vein thrombosis reduces post-thrombotic syndrome.
Int Angiol 2004;23:206–12.
[16] Kahn SR, Azoulay L, Hirsch A, Haber M, Strulovitch C, Shrier I.
Acute effects of exercise after deep venous thrombosis: Impact
of the post-thrombotic syndrome. Chest 2003;37:410–9.
[17] Kahn SR, Azoulay L, Hirsch A, Haber M, Strulovitch C, Shrier
I. Effect of graduated elastic compression stockings on leg
symptoms and signs during exercise in patients with deep
venous thrombosis: a randomized cross-over trial. J Thromb
Haemost 2003;1:494–9.
[18] Shrier I, Kahn SR. Effect of physical activity after recent deep
venous thrombosis: a cohort study. Med Sci Sports Exerc
2005;37:630–4.
[19] Isma N, Johanssson E, Bjork A, Bjorgell O, Robertson F,
Mattiasson I, et al. Does supervised exercise after deep venous
thrombosis improve recanalization of occluded vein segments?
A randomized study. J Thromb Thrombolysis 2007;23:25–30.
[20] Padberg J, Johnston MV, Sisto SA. Structured exercise
improves calf muscle pump function in chronic venous
insufficiency: a randomized trial. J Vasc Surg 2004;39:79–87.
[21] Trujillo-Santos AJ, Martos-Perez F, Perea-Milla E. Bed rest or
early mobilization as treatment of deep vein thrombosis: a
systematic review and meta-analysis. Med Clin (Barc)
2004;122:641–7.
[22] Buller HR, Agnelli G, Hull RD, Hyers TM, Prins MH, Raskob
GE. Antithrombotic therapy for venous thromboembolic
disease: the Seventh ACCP Conference on Antithrombotic
and Thrombolytic Therapy. Chest 2004;126:401S–28S.
773
[23] Schraibman IG, Milne AA, Royle EM. Home versus in-patient
treatment for deep vein thrombosis. Cochrane Database
Syst Rev 2001:CD003076.
[24] Kiser TS, Stefans VA. Pulmonary embolism in rehabilitation
patients: relation to time before return to physical therapy
after diagnosis of deep vein thrombosis. Arch Phys Med and
Rehabil 1997;78:942–5.
[25] Gardner AW, Poehlman ET. Exercise rehabilitation programs
for the treatment of claudication pain. A meta-analysis.
JAMA 1995;274:975–80.
[26] Leng GC, Fowler B, Ernst E. Exercise for intermittent
claudication. Cochrane Database Syst Rev 2000:CD000990.
[27] Remijnse-Tamerius HC, Duprez D, De Buyzere M, Oeseburg
B, Clement DL. Why is training effective in the treatment of
patients with intermittent claudication? Int Angiol 1999;18:
103–12.
[28] Stewart KJ, Hiatt WR, Regensteiner JG, Hirsch AT. Exercise
training for claudication. N Engl J Med 2002;347:1941–51.
[29] Szmolenszky T, Mark B, Kiss T, Trombitas K. Ultrastructural
damage of crural muscle in post-thrombotic syndrome. Acta
Chir Hung 1984;25:107–18.
[30] Makitie J. Muscle changes in patients with varicose veins.
Acta Pathol Microbiol Scand [A] 1977;85:864–8.
[31] Qvarfordt P, Eklof B, Ohlin P, Plate G, Saltin B. Intramuscular
pressure, blood flow, and skeletal muscle metabolism in
patients with venous claudication. Surgery 1984;95:191–5.
[32] Godin G, Shephard RJ. A simple method to assess exercise
behavior in the community. Can J Appl Spt Sci 1985;10:
141–6.