Journal of Diabetes 8 (2016), 363–377

O R I G I N A L A RT I C L E

Intensive walking exercise for lower extremity peripheral

arterial disease: A systematic review and meta-analysis
Highlights
• Similar to PAD patients with intermittent claudication (IC), walking exercise may also be recommended to PAD
patients without IC.
• Early involvement in intensive walking exercise can bring more therapeutic benefit to PAD patients.
• The efficacy of no-to-mild pain exercise is comparable to that of moderate-to-maximal pain exercise in patiens
with PAD.
• Presence of diabetes may attenuate the improvement of walking performance in patients with PAD following
exercise.

Xiafei LYU,1,2* Sheyu LI,2* Shifeng PENG,3 Huimin CAI,3 Guanjian LIU4 and Xingwu RAN1,2
1
Diabetic Foot Center, 2Department of Endocrinology and Metabolism, 4Chinese Cochrane Center, West China Hospital, and 3Clinical
Medicine of Eight-year Program, West China School of Medicine, Sichuan University, Chengdu, China

Correspondence
Xingwu Ran, Department of
Endocrinology and Metabolism, Diabetic
Foot Center, West China Hospital,
Chengdu, Sichuan 610041, China.
Tel: +86 28 85422982
Fax: +86 28 85422982
Email: ranxingwu@163.com
*These authors contributed equally to this
work.
Received 29 November 2014; revised 25
March 2015; accepted 26 April 2015.
doi: 10.1111/1753-0407.12304
Abstract
Background: Supervised treadmill exercise is the recommended therapy for
peripheral arterial disease (PAD) patients with intermittent claudication (IC).
However, most PAD patients do not exhibit typical symptoms of IC. The aim
of the present study was to explore the efficacy and safety of intensive walking
exercise in PAD patients with and without IC.
Methods: The PubMed, Embase and Cochrane Library databases were sys-
tematically searched. Randomized controlled trials comparing the effects of
intensive walking exercise with usual care in patients with PAD were included
for systematic review and meta-analysis.
Results: Eighteen trials with 1200 patients were eligible for the present
analysis. Compared with usual care, intensive walking exercise significantly
improved the maximal walking distance (MWD), pain-free walking distance,
and the 6-min walking distance in patients with PAD (P < 0.00001 for all).
Subgroup analyses indicated that a lesser improvement in MWD was
observed in the subgroup with more diabetes patients, and that the subgroup
with better baseline walking ability exhibited greater improvement in walking
performance. In addition, similar improvements in walking performance were
observed for exercise programs of different durations and modalities. No
significant difference was found in adverse events between the intensive
walking and usual care groups (relative risk 0.84; 95% confidence interval
0.51, 1.39; P = 0.50).
Conclusions: Regardless of exercise length and modality, regularly intensive
walking exercise improves walking ability in PAD patients more than usual
care. The presence of diabetes may attenuate the improvements in walking
performance in patients with PAD following exercise.
Keywords: exercise, intermittent claudication, meta-analysis, peripheral arte-
rial disease, walk.

© 2015 Ruijin Hospital, Shanghai Jiaotong University School of Medicine and John Wiley Sons & Australia, Ltd 363
Walking for peripheral arterial disease
Significant findings of the study: Intensive walking exericse improved the walking ability in PAD patients with and
without IC. A lesser improvement in MWD was observed in trials with more diabetes patients, and patients with
better baseline walking ability exhibited greater improvement in walking performance.
What this study adds: Early adoption of walking exercise may result in a better efficacy in patients with PAD. The
presence of diabetes may attenuate the improvements in walking performance in patients with PAD following
exercise.
Introduction
Lower extremity peripheral arterial disease (PAD) is a
chronic atherosclerosis disease in which stenoses or
occlusions of the peripheral arteries result in insufficient
blood flow to the lower limbs. The prevalence of PAD is
approximately 5% in the general population aged >40
years, and increases to approximately 15% among
elderly people >70 years of age.1 Intermittent claudica-
tion (IC) is considered the most typical symptom of PAD
among a panoply of clinical manifestations. However,
patients with exertional leg pain other than IC and
asymptomatic patients without previous exertional leg
symptoms are common among PAD patients, particu-
larly among patients who are older, male, diabetic, or
identified on the basis of ankle–brachial index (ABI)
screening in the primary care setting.2–4 There is a signifi-
cant association between PAD and limited daily activity,
decreased quality of life, and increased mortality.5–7
Typical therapeutics for PAD include exercise, smoking
cessation, modification of risk factors (blood pressure,
glucose, and lipids), percutaneous angioplasty, and
bypass surgery.8,9 There have been intensive investiga-
tions focused on the effects of exercise in PAD patients
with IC and the results indicate that exercise is a safe and
effective treatment option.10 The guidelines of the Ameri-
can College of Cardiology Foundation/American Heart
Association (ACCF/AHA) also recommend supervised
treadmill exercise as the standard therapy for patients
with IC.8,9 Previous systematic reviews showed that com-
pared with usual care or placebo, exercise could achieve
a 50%–200% improvement in maximal walking time
(MWT) and increase the physical quality of life.10,11
However, most of the reviews have included studies with
selected PAD patients with IC, as well as studies that
have used exercise modes other than walking exercise,
such as lower limb resistance training, dynamic leg exer-
cise, and even upper limb exercise. Few data are available
addressing the efficacy of walking exercise in PAD
patients without IC. In the present study, we sought to
evaluate the efficacy and safety of intensive walking exer-
cise in PAD patients with and without IC. In addition,
using subgroup analyses, we explored whether different
clinical characteristics, such as the baseline walking per-
364 © 2015 Ruijin Hospital, Shanghai Jiaotong University School of Medicine and John Wiley Sons & Australia, Ltd
X. LYU et al.
formance of patients and exercise intervention param-
eters, were associated with differential efficacy of
walking exercise in patients with PAD.
Methods
Selection criteria
Studies were included in the present review if: (i) they
were randomized clinical controlled trials comparing
intensive walking exercise with usual care, such as advice
to maintain daily activity or “go home and walk”, in
PAD patients with and without IC; (ii) the diagnosis of
PAD was identified by typical IC symptoms or an ABI
<0.9 at rest or <0.73 after exercise;9 (iii) the length of the
exercise program was no less than 12 weeks;9 and (iv) the
studies reported walking assessments of maximal
walking distance (MWD) and/or MWT and pain-free
walking distance and/or time (PFWD and PFWT,
respectively) measured using a graded treadmill test.
Trials in which the exercise mode was something other
than walking were excluded from the analysis. We also
excluded duplicate reports, trials without a clear defini-
tion of the diagnosis of PAD, and trials comparing dif-
ferent exercise modes.
Outcome assessment
The primary outcomes included MWD and PFWD mea-
sured using a graded treadmill test. Secondary outcomes
included 6-min walking distance (6-MWD), resting ABI,
post-exercise ABI, adverse events, and lower limb func-
tional status measured using the Walking Impairment
Questionnaire (WIQ),12 which includes distance, speed
and stair-climbing scores. The WIQ is a qualitative and
validated tool to detect objective improvements in func-
tional ability in response to exercise therapy in patients
with PAD.12
Data sources and search strategy
The electronic databases of PubMed, Ovid Embase, and
the Cochrane Library up until October 2014 were
searched without language restriction using both MeSH
terms and key words. The electronic search strategy was
X. LYU et al. Walking for peripheral arterial disease

developed by XL and SL. The detailed search strategy Publication bias was assessed by the visual inspection
for PubMed publications is available as Supplementary of funnel plots, as well as Begg’s rank correlation test
Material to this paper. The reference lists of the publica- and Egger’s linear regression test.17,18
tions identified were also checked to find relevant cita-
tions. When several studies analyzed the same set of
patients, the articles that had the longest follow-up Results
period were selected for analysis in the present study.
The systematic review and meta-analysis was conducted Search results and study characteristics
following the Preferred Reporting Items for Systematic The flow chart for publication selection is shown in
Reviews and Meta-analyses (PRISMA) statement.13 Fig. 1. Eighteen studies with sample sizes ranging from
18 to 252 were eligible for the present systematic
Citation screening and data attraction review.19–36 The baseline characteristics of the studies
included are summarized in Table 1. The meta-analysis
Citations were scanned independently by XL and SP,
included 1200 PAD patients in all, with a mean age of
whereas XL and HC reviewed the full text articles and
67.1 ± 9.4 years, and 67% of them were male. Most
extracted the data independently using a self-designed
studies exclusively included PAD patients with IC, and
standardized data extraction form.
the mean PFWD ranged from 110 to 266 m. Three
studies also included asymptomatic PAD patients and
Quality assessment PAD patients with exertional leg pain other than IC.29–31
The studies included in the analysis were assessed by XL The mean ABI was 0.67 ± 0.18. There were 16 studies
and HC using a checklist described in the Cochrane that reported the proportion of patients with diabetes
Handbook for Systematic Reviews of Interventions to and five of these excluded diabetic patients from the
determine methodological quality.14 The adequacy of study design. The mean diabetes proportion among these
each category was assessed as low, unclear, or high risk. studies was 24%.
Blinding of personnel and patients during exercise was The walking exercise parameters of the studies
not applicable and therefore considered as low risk for included are summarized in Table 2. Most studies
all studies. adopted supervised treadmill training, whereas one study
assessed the efficacy of home-based walking exercise in
Statistical analysis
Review Manager (RevMan) Version 5.3 (The Cochrane
Collaboration, The Nordic Cochrane Centre, Copenha-
gen, Denmark; 2014) and STATA Version 12.0 (Stata
Corporation, College Station, TX, USA) were used for
statistical analyses. Unless indicated otherwise, baseline
data are given as the mean ± SD. Two-sided P < 0.05 was
considered significant in all tests. If the walking perfor-
mance was reported as walking time, it was transformed
to walking distance using the reported walking speed.
For all continuous outcome measures, the weighted
mean difference (WMD) with 95% confidential intervals
(CI) was calculated with mean difference (MD) and stan-
dard deviation (SD) between the baseline and follow-up
assessment of each group. For adverse events, the risk
ratio (RR) with 95% CI was calculated. Statistical het-
erogeneity was tested using Q test and I2 statistics.15 All
data were pooled using a random effect model for the
high clinical heterogeneity. Sensitivity analyses, Gal-
braith plot, and subgroup analyses were also used.16 Sub-
group analyses were stratified by age, sample size,
symptoms, ABI, body mass index (BMI), male propor-
tion, diabetes proportion, smoking proportion, exercise Figure 1 Systematic process for publication selection. RCT, ran-
program length, and exercise modality. domized controlled trial.

© 2015 Ruijin Hospital, Shanghai Jiaotong University School of Medicine and John Wiley Sons & Australia, Ltd 365
Walking for peripheral arterial disease X. LYU et al.

Table 1 Characteristics of randomized controlled trials included in the present meta-analysis

No. subjects BMI Mean Mean

Study Country (IntWalk/Ctrl) Age (years) Male (kg/m2) ABI Smokers Diabetes MWD (m) PFWD (m)

Allen et al.19 USA 15/18 67.0 ± 11.8 NR 27.8 ± 5.9 0.66 ± 0.24 31% 0.0% 505.7 171.0
Crowther et al.20 Australia 10/11 69.1 ± 7.8 48% 28.8 ± 4.8 0.69 ± 0.23 19% 19% 274.2 110.0
Cucato et al.21 Brazil 13/12 62.6 ± 7.0 100% 25.6 ± 3.1 0.61 ± 0.03 24% 20% 805.1 266.1
Gardner et al.22 USA 17/14 71.5 ± 3.9 92% 29.9 ± 4.3 0.68 ± 0.20 0.0% 42% 388.3 167.9
Gardner et al.23 USA 80/39 65.3 ± 10.9 48% 29.8 ± 6.0 0.73 ± 0.23 10% 43% 362.7 184.7
Gardner et al.24 USA 106/36 68.0 ± 8.0 85% 28.3 ± 4.7 0.66 ± 0.20 60% 32% 373.4 158.5
Hiatt et al.25 USA 10/9 60.1 ± 12.2 100% NR 0.67 ± 0.11 58% 0.0% 331.0 NR
Hiatt et al.26 USA 10/8 67.0 ± 6.0 100% NR 0.58 ± 0.15 70% 0.0% 459.5 187.7
Hodges et al.27 UK 14/14 68.0 ± 8.0 NR 26.7 ± 3.4 0.60 ± 0.10 NR NR 354.7 NR
Kruidenier et al.28 Netherlands 35/35 62.4 ± 9.8 61% 27.1 ± 4.2 0.70 ± 0.19 56% 20% 594.6 NR
McDermott et al.29 USA 17/8 69.6 ± 8.8 52% 28.7 ± 5.3 0.61 ± 0.14 19% 35% 115.9 NR
McDermott et al.30 USA 51/53 70.1 ± 10.5 47% 30.2 ± 6.7 0.60 ± 0.18 14% 43% 382.7 130.1
McDermott et al.31 USA 97/97 70.2 ± 9.6 50% 29.1 ± 6.7 0.67 ± 0.17 24% 33% 412.2 155.0
Mika et al.32 Poland 27/28 59.0 ± 8.1 87% NR 0.66 ± 0.18 76% 0.0% 386.2 175.9
Mika et al.33 Poland 30/31 62.8 ± 7.0 87% 27.7 ± 3.2 0.78 ± 0.11 80% 0.0% 542.8 240.9
Nicolaï et al.34 Netherlands 169/83 66.2 ± 9.3 63% 27.9 ± 4.6 0.66 ± 0.18 42% 23% 260.0 NR
Treat-Jacobson et al.35 USA 11/8 66.5 ± 10.4 74% 27.5 ± 4.3 0.68 ± 0.12 NR 21% 431.7 166.2
Tsai et al.36 China 27/26 76.2 ± 3.7 83% 23.3 ± 2.5 0.70 ± 0.10 NR NR 389.2 165.5

Unless indicated otherwise, data are given as the mean ± SD.

ABI, ankle–brachial index; BMI, body mass index; Ctrl, usual care (control) group; IntWalk, intensive walking exercise group; MWD, maximal
walking distance; NR, not reported; PFWD, pain-free walking distance.

patients with PAD.31 Exercise program length ranged
from 12 to 78 weeks with exercise frequency of no less
than twice per week. The duration of exercise sessions
was no less than 30 min. The walking speed was no less
than 3.2 km/h with metabolic equivalents (MET)/min
that of at least 2.5. In terms of exercise modality, the
patients were encouraged to walk to moderate-to-
maximal pain in most studies, whereas four studies were
preset to no-to-mild pain.21,32,33,36
Methodological quality
Results of quality assessment are given in Table 3. The
studies included in the meta-analysis were of mediocre
quality. Only approximately 50% reported detailed
random sequence generation, and the most common bias
occurred during blinding of assessors and adequate allo-
cation concealment.
Maximal walking distance
The meta-analysis demonstrated that compared with
usual care, intensive walking exercise could significantly
improve MWD in patients with PAD (WMD 218.01 m;
95% CI 161.54, 274.48; P < 0.00001; Fig. 2). Because
high heterogeneity was detected among studies
(Q = 103.92, P < 0.00001; I2 = 84%), a random-effect
model was used to calculate the combined effect esti-
mates. Sensitivity analysis by removing one study at a
time suggested that no single study markedly influenced
the pooled WMDs, which yielded a range from 203.38 m
(95% CI 153.90, 252.86) to 231.81 m (95% CI 182.65,
280.97; see Fig. S1). A Galbraith plot was used and four
outliers were detected (Fig. S2).29,31–33 After removing
these four outliers, the pooled result was just slightly
increased to 233.83 m (95% CI 197.72, 269.94). The
between-study heterogeneity was completely eliminated
after removal of the outliers (Q = 9.92, P = 0.70;
I2 = 0%).
To further confirm the results and identify potential
sources of heterogeneity, we performed subgroup
analyses using the covariates mentioned above and the
results are given in Table S1. In subgroup analysis
stratified by different diabetes proportion, greater
improvement in MWD was achieved in subgroups with
a lower diabetes proportion. The improvements in
MWD for subgroup with a diabetes proportion of
<25% and 25%–50% were 266.19 m (95% CI 201.37,
331.01) and 138.83 m (95% CI 55.38, 222.28), respec-
tively (P = 0.02; Fig. 3).
Intensive walking exercise resulted in significantly
greater improvements in MWD in PAD patients with IC
and a mean PFWD ≥ 200 m compared with PAD
patients with IC and a mean PFWD <200 m (WMD
321.39 m [95% CI 290.22, 352.56] and 200.90 m [95% CI
173.68, 228.11], respectively; P < 0.00001). However,
intensive walking exercise was not significantly better

366 © 2015 Ruijin Hospital, Shanghai Jiaotong University School of Medicine and John Wiley Sons & Australia, Ltd
 X. LYU et al.

Table 2 Exercise parameters of randomized controlled trials included in the meta-analysis

Program Session Session Walking

length duration frequency Walking advice for
Study Exercise (weeks) (min) (/week) Exercise modality MET/min speed (km/h) Supervision control group

Allen et al.19 Supervised treadmill exercise 14 40 3 Moderate pain 2.5 Initial at 3.2 Yes Yes
Crowther Supervised treadmill exercise 52 40 3 Maximal pain 2.5 3.2 Yes Yes
et al.20
Cucato et al.21 Supervised treadmill exercise 12 30 2 No to mild pain 2.5 3.2 Yes Yes
Gardner Supervised treadmill exercise 76 40 3 Sub-maximal pain 2.5 3.2 Yes No
et al.22
Gardner Supervised treadmill exercise 12 45 3 Sub-maximal pain 2.5 3.2 Yes Yes
et al.23 or home-based walking exercise
Gardner Supervised treadmill exercise 26 40 3 Sub-maximal pain 3.5 3.2 Yes Yes
et al.24
Hiatt et al.25 Supervised treadmill exercise 12 50 3 Moderate pain 2.5 3.2 Yes No
Hiatt et al.26 Supervised treadmill exercise 12 50 3 Moderate pain 2.5 3.2 Yes No
Hodges Supervised treadmill exercise 12 30 2 Maximal pain 2.5 3.2 Yes Yes
et al.27
Kruidenier Supervised community-based 26 30 2–3 Sub-maximal pain 2.5 3.2 Yes No
et al.28 walking exercise
McDermott Supervised treadmill exercise 14 50 3 Exercise to 11–12 on the Borg 2.5 3.2 Yes No
et al.29 RPES
McDermott Supervised treadmill exercise 26 40 3 Sub-maximal or 12–14 (moderately 2.5 3.2 Yes No
et al.30 hard) on the Borg RPES

© 2015 Ruijin Hospital, Shanghai Jiaotong University School of Medicine and John Wiley Sons & Australia, Ltd
McDermott Home-based walking exercise 26 50 5 Sub-maximal or 12–14 (moderately NR NR No No
et al.31 hard) on the Borg RPES
Mika et al.32 Supervised treadmill exercise 14 50 3 No pain 2.5 3.2 Yes No
Mika et al.33 Supervised treadmill exercise 12 55 3 No pain 2.5 3.2 Yes No
Nicolaï et al.34 Supervised treadmill exercise 52 30 3 Sub-maximal pain 2.5 3.2 Yes Yes
Treat-Jacobson Supervised treadmill exercise 12 60 3 Sub-maximal pain 2.5 3.2 Yes Yes
et al.35
Tsai et al.36 Supervised treadmill exercise 12 30 3 Mild pain 2.5 3.2 Yes No

MET, metabolic equivalents; NR, not reported; RPES, rating of perceived exertion scale.

367
Walking for peripheral arterial disease
Walking for peripheral arterial disease X. LYU et al.

Table 3 Risk of bias of randomized controlled trials included in the meta-analysis

Random Blinding of Blinding of Free of

sequence Allocation participants outcome Incomplete selective Free of
Study generation concealment and personnel assessment outcome data reporting other bias

Allen et al.19
Crowther et al.20
Cucato et al.21
Gardner et al.22
Gardner et al.23
Gardner et al.24
Hiatt et al.25
Hiatt et al.26
Hodges et al.27
Kruidenier et al.28
McDermott et al.29
McDermott et al.30
McDermott et al.31
Mika et al.32
Mika et al.33
Nicolaï et al.34
Treat-Jacobson et al.35
Tsai et al.36
? − + − − ? ?
? ? + − + + +
? − + − ? + ?
+ − + − + + ?
+ + + − + + +
+ + + − + + +
? − + − + + −
? − + − + + −
+ − + − ? + ?
+ + + − + + −
+ − + − + + −
+ + + + + + −
+ + + + + + +
? − + + + + ?
? − + − ? + ?
+ + + + + + +
+ − + − + + +
? − + − + + −

−, low risk; +, high risk; ?, unclear risk.

Figure 2 Effect of intensive walking exercise compared with usual care (control group) on maximal walking distance. CI, confidence interval.

than usual care in PAD patients without IC (WMD

75.83 m; 95% CI –17.32, 168.98; P = 0.11; Fig. S3).
Subgroup analyses indicated that differential improve-
ments in MWD were not associated with varied exercise
program lengths and modalities (P = 0.66 and P = 0.29,
respectively).
Pain-free walking distance
Intensive walking exercise was associated with a signifi-
cantly greater improvement in PFWD compared with
usual care (WMD 174.19 m; 95% CI 130.25, 218.13;
P < 0.00001; Fig. 4). High heterogeneity was also

368 © 2015 Ruijin Hospital, Shanghai Jiaotong University School of Medicine and John Wiley Sons & Australia, Ltd
X. LYU et al. Walking for peripheral arterial disease

Figure 3 Effect of intensive walking exercise on maximal walking distance compared with usual care (control group) stratified by diabetes
proportion. CI, confidence interval.

Figure 4 Effect of intensive walking exercise compared with usual care (control group) on pain-free walking distance. CI, confidence interval.

detected among the studies (Q = 46.00, P < 0.00001;
I2 = 78%). A Galbraith plot revealed that two studies
were outliers (Fig. S4).32,33 After removal of the outliers,
the pooled result was decreased slightly, albeit not sig-
nificantly, to 151.78 m (95% CI 115.15, 188.42), and a
significant reduction in between-study heterogeneity was
achieved (I2 from 78.5% to 19%).
Subgroup analyses were also conducted using the
covariates mentioned above and the results are presented
in Table S2. Greater improvement in PFWD was
observed in PAD patients with IC and a mean PFWD
≥200 m than in patients with a mean PFWD <200 m
(WMD 274.25 m [95% CI 242.77, 305.74] and 159.89 m
[95% CI 140.70, 179.08], respectively; P < 0.00001;

© 2015 Ruijin Hospital, Shanghai Jiaotong University School of Medicine and John Wiley Sons & Australia, Ltd 369
Walking for peripheral arterial disease X. LYU et al.

Figure 5 Effect of intensive walking exercise compared with usual care (control group) on pain-free walking distance stratified by diabetes
proportion. CI, confidence interval.

Figure 6 Effect of intensive walking exercise compared with usual care (control group) on 6-min walking distance. CI, confidence interval.

Fig. S5). In subgroup analysis stratified by different dia-

betes proportion, comparable improvements in PFWD
were achieved in different subgroups (Fig. 5). No signifi-
cant differences in PFWD improvements were observed
for other stratifications.
Six-minute walking distance
Compared with usual care, intensive walking exercise
significantly improved 6-MWD (WMD 42.91 m; 95%
CI 26.41, 59.41; P < 0.001; Fig. 6). Further subgroup
analyses suggested no significant difference in all
stratifications, including differential PAD symptoms
(Fig. S6).
Ankle–brachial index
Comparsion of intensive walking exercise with usual care
showed no significant differences in resting ABI and
post-exercise ABI (WMD –0.03 [95% CI –0.09, 0.02;
P = 0.22] and –0.03 [95% CI –0.12, 0.05; P = 0.49],
respectively; Figs S7,S8).
Walking Impairment Questionnaire score
The WIQ distance, speed, and stair-climbing scores were
all significantly improved with intensive walking exercise
compared with usual care (WMD 9.20 [95% CI 5.74,
12.70; P < 0.00001], 8.71 [95% CI 5.64, 11.77; P <
0.00001] and 8.02 [95% CI 4.84, 11.21; P < 0.00001],

370 © 2015 Ruijin Hospital, Shanghai Jiaotong University School of Medicine and John Wiley Sons & Australia, Ltd
X. LYU et al. Walking for peripheral arterial disease

Figure 7 All adverse events of intensive walking exercise compared with usual care (control group). CI, confidence interval.

respectively; Figs S9–S11). Further subgroup analyses

suggested the improvements were consistent among PAD
patients regardless of the existence of IC (Figs S12–S14).

Adverse events
The primary adverse events reported for walking exercise
included all-cause mortality, cardiovascular events,
amputation, progression of PAD, fracture, respiratory
infection, chest pain, and arrhythmia. Meta-analysis
showed that there were no significant differences between
the walking exercise and usual care groups in terms of the
incidence of all adverse events. The pooled result for the
incidence of all adverse events was RR 0.84 (95% CI
Figure 8 Funnel plots for visual assessment of the presence of
0.51, 1.39; P = 0.50; Fig. 7).
publication bias for maximal walking distance in the meta-analysis.

Publication bias
No significant evidence of publication bias was observed
for MWD and PFWD, as indicated by visual observa-
tion of the funnel plots (Figs 8,9), Begg’s test (P = 0.150
and P = 0.161, respectively) and Egger’s test (P = 0.730
and P = 0.625, respectively). Publication bias was also
not significant among the studies on 6-MWD, ABI, all
the three WIQ scores and all adverse events.
Discussion
The findings of the present meta-analysis indicated that
compared with usual care, intensive walking exercise
improved MWD, PFWD, 6-MWD, and WIQ scores in
PAD patients with IC. In PAD patients without IC,
intensive walking exercise could improve 6-MWD and
WIQ scores, but not MWD. Greater improvement in
walking performance was observed in PAD patients with
IC and a mean baseline PFWD ≥200 m, and no differ-
ence in improvement of walking performance was
observed among different exercise program lengths and
modalities. Intensive walking exercise had no effect on
ABI compared with usual care, and it did not increase
the incidence of adverse events.
For PAD patients without IC, improvements in
6-MWD but not MWD were observed after intensive
walking exercise. However, the differential efficacy of
intensive walking exercise in PAD with and without IC
may be associated with the quality of the studies included
and the different tools used to assess walking ability. The
three studies on PAD patients without IC included in this
systematic review exhibited significant heterogeneity,
which restricted the reliability of the study.29–31 The
dropout rate of one study was high, with only 14 of 24
participants completing the exercise training program;
despite randomization, the exercise training group had

© 2015 Ruijin Hospital, Shanghai Jiaotong University School of Medicine and John Wiley Sons & Australia, Ltd 371
Walking for peripheral arterial disease
Figure 9 Funnel plots for visual assessment of the presence of
publication bias for pain-free walking distance in the meta-analysis.
more patients with cardiovascular and lung diseases.29 In
2013, McDermott et al. reported on the adoption of
home-based walking exercise without supervision and
only with group-mediated cognitive behavioral interven-
tion.31 Previous systematic reviews have demonstrated
that patients with IC gain little benefit from home-based
walking exercise.37,38 A well-controlled study by McDer-
mott et al. assessed the efficacy of supervised treadmill
exercise on PAD without IC, and observed significant
improvements in the supervised treadmill exercise group
compared with the usual care group.30 With regard to the
different assessment tools, both the graded treadmill test
and 6-min walk test are used to assess walking ability.
Studies showed that walking on a treadmill could not
mimic walking in daily life well and there was also a
significant learning effect.39,40 Nevertheless, the 6-min
walk test is closely correlated to real-life outdoor walking
capacity, and is increasingly recognized as a meaningful
outcome measure in patients with PAD.40,41 Our results
demonstrate that walking exercise can improve the
6-MWD in PAD patients without IC. Thus, it seems
reasonable to hypothesize that intensive walking exercise
could also improve the walking ability in PAD patients
without IC. Further studies are required to address this
issue.
Our subgroup analysis found that improvement in
walking performance was significantly associated with
differences in baseline PFWD. Intensive walking exercise
led to greater improvements in PAD patients with IC
and better baseline walking performance, indicating that
early adoption of intensive walking exercise may result in
a better efficacy in patients with PAD.
Subsequent subgroup analyses indicated greater
improvements in MWD were achieved in subgroups
with a lower proportion of diabetic patients. In con-
trast, comparable improvements in PFWD were
372 © 2015 Ruijin Hospital, Shanghai Jiaotong University School of Medicine and John Wiley Sons & Australia, Ltd
X. LYU et al.
observed in groups with different proportions of dia-
betic patients. Studies focused on the effects of diabetes
on exercise tolerance and the physiological responses to
exercise in PAD showed inconsistent results. Three
studies reported that exercise tolerance was lower in dia-
betic patients, but exercise was equally effective in
improving walking distance for patients with and
without diabetes, suggesting that the beneficial effect of
exercise was not negatively influenced by the presence of
diabetes.42–44 However, another two studies found that
diabetes in PAD attenuated improvements in endothe-
lial function, net plasma nitrite, and PFWD following
exercise.45,46 In any event, these apparent inconsistencies
call for further investigations to confirm whether the
efficacy of walking exercise is compromised in PAD
patients with diabetes.
The present meta-analysis also showed that improve-
ments of walking ability were comparable following dif-
ferent exercise program lengths. The guidelines of the
ACCF/AHA have suggested a 3-month supervised
treadmill training program for PAD patients with IC.8,9
However, the parameters of the exercise regimen remain
contentious. Gardner et al. reported that increased
walking performance was achieved during the first 2
months exercise and maintained with a further 4 months
training.22 However, Pilz et al. evaluated 6- and
12-month supervised exercise training programs on the
walking performance of PAD patients and found that
even though improvements in treadmill performance
were similar, greater improvements in walking speed
were achieved in the 12-month training group (12.1 vs
5.3%; P < 0.001),47 indicating that patients could benefit
more from a longer exercise program. However, pro-
longed walking exercise is more demanding to patients
and requires more medical resources, which may greatly
limit patient adherence.
In contrast with existing data,48,49 subgroup analysis in
the present study found that walking exercise of no to
mild pain could achieve a similar improvement in
walking ability compared with moderate-to-maximal
pain exercise. Previous studies suggested that patients
should walk to maximal or near-maximal pain.49
However, recent studies reported that walking to sub-
maximal or maximal pain could induce inflammatory
responses that may exacerbate atherosclerosis.50 Consis-
tent with this notion, it has been suggested that pain-free
walking training still improves walking performance in
patients with PAD, and has no effect on inflammatory
status.32,33,51 Gardner et al. trained two groups of patients
with IC to either 40% or 80% of the maximal workload
and found that the improvements in walking ability
between the two groups were similar.52 Another random-
ized trial that compared moderate claudication pain
X. LYU et al.
Figure 10 Potential mechanisms by which walking exercise generate improvement in walking performance in patients with peripheral arterial
disease. HIF-1, hypoxia-inducible factor-1; VEGF, vascular endothelial growth factor; NOS, nitric oxide synthase
treadmill exercise with pain-free exercise also showed
similar results.53 Therefore, we feel comfortable to
recommend pain-free or mild pain exercise to our
patients given the similar efficacy and potentially better
compliance.
The mechanism by which walking exercise exerts ben-
eficial effects on walking performance remains incom-
pletely understood.54 There is accumulating evidence
that improved collateral circulation may be involved.23,25
Significant increases in calf blood flow under resting,
hyperemic, and maximal conditions have been demon-
strated after 6 months supervised treadmill training.25
Long-term aerobic exercise results in relative oxygen
insufficiency and enhances expression of hypoxia-
inducible factor-1 and vascular endothelial growth
factor, which, in turn, induces angiogenesis.55,56 Never-
theless, other studies showed that the improvement in
walking ability was correlated with altered metabolic
activity of the skeletal muscles, as assessed by muscle
biopsy.57,58 Another possible mechanism may rely on
increased nitric oxide synthase activity and increased
endothelium-dependent dilation of the vessels,59–61
improvements in cardiopulmonary capacity,27 improved
pain tolerance, and psychological adaptations to exer-
cise training.62 These potential mechanisms are described
and summarized in Fig. 10.
© 2015 Ruijin Hospital, Shanghai Jiaotong University School of Medicine and John Wiley Sons & Australia, Ltd
Walking for peripheral arterial disease
The present systematic review has some limitations.
Given the lack of allocation concealment and blindness
of outcome assessors, the overall quality of the literature
included in this review was suboptimal, limiting the level
of evidence of this systematic review. The sample size of
several studies was small (<20 patients per group). Most
studies did not describe statistical methods used to cal-
culate the required sample size, which could result
in imprecise effect estimation. Although the random-
effect model was used to calculate the pooled results,
considerable heterogeneity existed in the baseline char-
acteristics of the participants and walking exercise
parameters. We have conducted further analyses to
explore the heterogeneity. Sensitivity analyses indicated
that no single study exerted a marked effect on the
pooled results. Subgroup analyses found that different
symptoms of PAD and mean baseline PFWD were the
main sources of heterogeneity. Furthermore, drug and
risk factor modification were made in several studies
without a wash-out period, which may bring about
mixed effects. In addition, no attempts were made to
search ongoing or unpublished studies. Publication bias
may also mask the pooled results because the funnel
plots of the studies were not perfectly symmetrical, even
though Begg’s and Egger’s tests did not indicate substan-
tial publication bias.
373
Walking for peripheral arterial disease
The present systematic review may also be informa-
tive for further research. First, the study found that the
efficacy of walking exercise was associated with the
baseline walking ability of the patients. Therefore, it
sounds reasonable to incorporate baseline status into
future study designs to confirm this. Second, the results
indicated that the short-term effects of shorter (3
months) and longer (>6 month) exercise programs were
comparable. In future, studies could be directed to the
long-term effects of exercise programs of different
lengths. In addition, it would be interesting to determine
whether the efficacy of exercise is compromised in
PAD patients with diabetes, as well as the underlying
mechanism by which walking exercise improves walking
performance.
In conclusion, this review suggests that intensive
walking exercise therapy should play an important role
in the standard care of PAD patients. Similar to PAD
patients with IC, walking exercise may also be recom-
mended to PAD patients without IC, but more studies
are warranted to further elucidate this issue. Moreover,
early involvement in intensive walking exercise can
bring about a greater therapeutic benefit to PAD
patients. Intensive walking exercise can also improve
the walking ability in patients with PAD and diabetes,
but whether the efficacy of exercise is attenuated by
the presence of diabetes needs further investigation.
As for the parameters of the exercise program, our
results indicate that no-to-mild pain exercise is compa-
rable to moderate-to-maximal pain exercise and that
a 3-month exercise program may be preferred over
a longer program to treat patients with PAD, but
obviously more evidence is needed to substantiate this
argument.
Acknowledgements
The authors extend their thanks to the doctors and
nurses at the Diabetic Foot Center, Department of
Endocrinology and Metabolism, West China Hospital,
Sichuan University, for their hard work. This meta-
analysis was funded by the National Natural Science
Foundation of China (Grant no. 81170776, 81471043 to
XR; Grant no. 81400811 to SL), Science and Technology
Bureau of Sichuan Province, China (Grant no.
2011SZ0220 to XR), and Leaders of Disciplines Foun-
dation of Sichuan Personnel Bureau (Grant No. [2012]
319-1 to XR).
Disclosure
None declared.
374 © 2015 Ruijin Hospital, Shanghai Jiaotong University School of Medicine and John Wiley Sons & Australia, Ltd
X. LYU et al.
References
1. Selvin E, Erlinger TP. Prevalence of and risk factors for
peripheral arterial disease in the United States: Results
from the National Health and Nutrition Examination
Survey, 1999–2000. Circulation. 2004; 110: 738–43.
2. McDermott MM, Mehta S, Greenland P. Exertional leg
symptoms other than intermittent claudication are
common in peripheral arterial disease. Arch Intern Med.
1999; 159: 387–92.
3. McDermott MM, Greenland P, Liu K et al. Leg symp-
toms in peripheral arterial disease: Associated clinical
characteristics and functional impairment. JAMA. 2001;
286: 1599–606.
4. McDermott MM, Kerwin DR, Liu K et al. Prevalence
and significance of unrecognized lower extremity periph-
eral arterial disease in general medicine practice. J Gen
Intern Med. 2001; 16: 384–90.
5. Dolan NC, Liu K, Criqui MH et al. Peripheral artery
disease, diabetes, and reduced lower extremity function-
ing. Diabetes Care. 2002; 25: 113–20.
6. Smolderen KG, Hoeks SE, Pedersen SS, van Domburg
RT, de Liefde II, Poldermans D. Lower-leg symptoms in
peripheral arterial disease are associated with anxiety,
depression, and anhedonia. Vasc Med. 2009; 14: 297–
304.
7. Jain A, Liu K, Ferrucci L et al. Declining walking impair-
ment questionnaire scores are associated with subsequent
increased mortality in peripheral artery disease. J Am
Coll Cardiol. 2013; 6: 1820–29.
8. 2011 Writing Group Members, 2005 Writing Committee
Members, ACCF/AHA Task Force Members. 2011
ACCF/AHA Focused update of the guideline for the
management of patients with peripheral artery disease
(updating the 2005 guideline): A report of the American
College of Cardiology Foundation/American Heart
Association Task Force on practice guidelines. Circula-
tion. 2011; 124: 2020–45.
9. Hirsch AT, Haskal ZJ, Hertzer NR et al. ACC/AHA
2005 practice guidelines for the management of patients
with peripheral arterial disease (lower extremity, renal,
mesenteric, and abdominal aortic): A collaborative
report from the American Association for Vascular
Surgery/Society for Vascular Surgery, Society for Cardio-
vascular Angiography and Interventions, Society for Vas-
cular Medicine and Biology, Society of Interventional
Radiology, and the ACC/AHA Task Force on Practice
Guidelines (Writing Committee to Develop Guidelines
for the Management of Patients With Peripheral Arterial
Disease): Endorsed by the American Association of Car-
diovascular and Pulmonary Rehabilitation; National
Heart, Lung, and Blood Institute; Society for Vascular
Nursing; Trans Atlantic Inter-Society Consensus; and
Vascular Disease Foundation. Circulation. 2006; 113:
E463–654.
10. Watson L, Ellis B, Leng GC. Exercise for intermittent
claudication. Cochrane Database Syst Rev. 2008;
(4)CD000990.
11. Kruidenier LM, Viechtbauer W, Nicolaï SP et al. Treat-
ment for intermittent claudication and the effects on
X. LYU et al.
walking distance and quality of life. Vascular. 2012; 20:
20–35.
12. Nicolaï SP, Kruidenier LM, Rouwet EV, Graffius K,
Prins MH, Teijink JA. The walking impairment question-
naire: An effective tool to assess the effect of treatment
inpatients with intermittent claudication. J Vasc Surg.
2009; 50: 89–94.
13. Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA
Group. Preferred reporting items for systematic reviews
and meta-analyses: The PRISMA statement. BMJ. 2009;
339: B2535.
14. Higgins JP, Altman DG, Gøtzsche PC et al. The
Cochrane Collaboration’s tool for assessing risk of bias in
randomised trials. BMJ. 2011; 343: D5928.
15. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Mea-
suring inconsistency in meta-analyses. BMJ. 2003; 327:
557–60.
16. Galbraith RF. A note on graphical presentation of esti-
mated odds ratios from several clinical trials. Stat Med.
1988; 7: 889–94.
17. Begg CB, Mazumdar M. Operating characteristics of a
rank correlation test for publication bias. Biometrics.
1994; 50: 1088–101.
18. Egger M, Davey Smith G, Schneider M, Minder C. Bias
in meta-analysis detected by a simple, graphical test.
BMJ. 1997; 315: 629–34.
19. Allen JD, Stabler T, Kenjale A et al. Plasma nitrite flux
predicts exercise performance in peripheral arterial
disease after 3 months of exercise training. Free Radic
Biol Med. 2010; 49: 1138–44.
20. Crowther RG, Spinks WL, Leicht AS et al. Effects
of a long-term exercise program on lower limb mob-
ility, physiological responses, walking perform-
ance, and physical activity levels in patients with
peripheral arterial disease. J Vasc Surg. 2008; 47:
303–9.
21. Cucato GG, Chehuen Mda R, Costa LA et al. Exercise
prescription using the heart of claudication pain onset in
patients with intermittent claudication. Clinics (Sao
Paulo). 2013; 68: 974–8.
22. Gardner AW, Katzel LI, Sorkin JD, Goldberg AP.
Effects of long-term exercise rehabilitation on claudica-
tion distances in patients with peripheral arterial disease:
A randomized controlled trial. J Cardiopulm Rehabil.
2002; 22: 192–8.
23. Gardner AW, Parker DE, Montgomery PS, Scott KJ,
Blevins SM. Efficacy of quantified home-based exercise
and supervised exercise in patients with intermittent clau-
dication: A randomized controlled trial. Circulation.
2011; 123: 491–8.
24. Gardner AW, Montgomery PS, Parker DE. Optimal
exercise program length for patients with claudication. J
Vasc Surg. 2012; 55: 1346–54.
25. Hiatt WR, Regensteiner JG, Hargarten ME, Wolfel EE,
Brass EP. Benefit of exercise conditioning for patients
with peripheral arterial disease. Circulation. 1990; 81:
602–9.
26. Hiatt WR, Wolfel EE, Meier RH, Regensteiner JG. Supe-
riority of treadmill walking exercise versus strength train-
ing for patients with peripheral arterial disease.
© 2015 Ruijin Hospital, Shanghai Jiaotong University School of Medicine and John Wiley Sons & Australia, Ltd
Walking for peripheral arterial disease
Implications for the mechanism of the training response.
Circulation. 1994; 90: 1866–74.
27. Hodges LD, Sandercock GR, Das SK, Brodie DA. Ran-
domized controlled trial of supervised exercise to evaluate
changes in cardiac function inpatients with peripheral
atherosclerotic disease. Clin Physiol Funct Imaging. 2008;
28: 32–7.
28. Kruidenier LM, Nicolaï SP, Rouwet EV, Peters RJ, Prins
MH, Teijink JA. Additional supervised exercise therapy
after a percutaneous vascular intervention for peripheral
arterial disease: A randomized clinical trial. J Vasc Interv
Radiol. 2011; 22: 961–8.
29. McDermott MM, Tiukinhoy S, Greenland P et al. A pilot
exercise intervention to improve lower extremity func-
tioning in peripheral arterial disease unaccompanied by
intermittent claudication. J Cardiopulm Rehabil. 2004; 24:
187–96.
30. McDermott MM, Ades P, Guralnik JM et al. Treadmill
exercise and resistance training in patients with peripheral
arterial disease with and without intermittent claudica-
tion: A randomized controlled trial. JAMA. 2009; 301:
165–74.
31. McDermott MM, Liu K, Guralnik JM et al. Home-based
walking exercise intervention in peripheral artery disease:
A randomized clinical trial. JAMA. 2013; 310: 57–65.
32. Mika P, Spodaryk K, Cencora A, Mika A. Red blood
cell deformability in patients with claudication after
pain-free treadmill training. Clin J Sport Med. 2006; 16:
335–40.
33. Mika P, Wilk B, Mika A, Marchewka A, Nizankowski R.
The effect of pain-free treadmill training on fibrinogen,
haematocrit, and lipid profile inpatients with claudica-
tion. Eur J Cardiovasc Prev Rehabil. 2011; 18: 754–60.
34. Nicolaï SP, Teijink JA, Prins MH, Exercise Therapy in
Peripheral Arterial Disease Study Group. Multicenter
randomized clinical trial of supervised exercise therapy
with or without feedback versus walking advice for inter-
mittent claudication. J Vasc Surg. 2010; 52: 348–55.
35. Treat-Jacobson D, Bronas UG, Leon AS. Efficacy of
arm-ergometry versus treadmill exercise training to
improve walking distance in patients with claudication.
Vasc Med. 2009; 14: 203–13.
36. Tsai JC, Chan P, Wang CH et al. The effects of exercise
training on walking function and perception of health
status in elderly patients with peripheral arterial occlusive
disease. J Intern Med. 2002; 252: 448–55.
37. Wind J, Koelemay MJ. Exercise therapy and the addi-
tional effect of supervision on exercise therapy in patients
with intermittent claudication. Systematic review of ran-
domised controlled trials. Eur J Vasc Endovasc Surg.
2007; 34: 1–9.
38. Gommans LN, Saarloos R, Scheltinga MR et al. Editor’s
choice: The effect of supervision on walking distance in
patients with intermittent claudication: A meta-analysis.
Eur J Vasc Endovasc Surg. 2014; 48: 169–84.
39. Nordanstig J, Broeren M, Hensäter M, Perlander A,
Osterberg K, Jivegard L. Six-minute walk test closely
correlates to “real-life” outdoor walking capacity and
quality of life in patients with intermittent claudication. J
Vasc Surg. 2014; 60: 404–9.
375
Walking for peripheral arterial disease
40. McDermott MM, Guralnik JM, Criqui MH, Liu K,
Kibbe MR, Ferrucci L. Six-minute walk is a better
outcome measure than treadmill walking tests in thera-
peutic trials of patients with peripheral artery disease.
Circulation. 2014; 130: 61–8.
41. Hiatt WR, Rogers RK, Brass EP. The treadmill is a better
functional test than the 6-minute walk test in therapeutic
trials of patients with peripheral artery disease. Circula-
tion. 2014; 130: 69–78.
42. Van Pul KM, Kruidenier LM, Nicolaï SP et al. Effect of
supervised exercise therapy for intermittent claudication
in patients with diabetes mellitus. Ann Vasc Surg. 2012;
26: 957–63.
43. Mahé G, Ouedraogo N, Leftheriotis G, Vielle B, Picquet
J, Abraham P. Exercise treadmill testing in patients with
claudication, with and without diabetes. Diabet Med.
2011; 28: 356–62.
44. Green S, Askew CD, Walker PJ. Effect of type 2 diabetes
mellitus on exercise intolerance and the physiological
responses to exercise in peripheral arterial disease. Diabe-
tologia. 2007; 50: 859–66.
45. Gardner AW, Parker DE, Montgomery PS, Blevins SM.
Diabetic women are poor responders to exercise rehabili-
tation in the treatment of claudication. J Vasc Surg. 2014;
59: 1036–43.
46. Allen JD, Stabler T, Kenjale AA et al. Diabetes status
differentiates endothelial function and plasma nitrite
response to exercise stress in peripheral arterial disease
following supervised training. J Diabetes Complications.
2014; 28: 219–25.
47. Pilz M, Kandioler-Honetz E, Wenkstetten-Holub A,
Doerrscheidt W, Mueller R, Kurz RW. Evaluation of 6-
and 12-month supervised exercise training on strength and
endurance parameters in patients with peripheral arterial
disease. Wien Klin Wochenschr. 2014; 126: 383–9.
48. Parmenter BJ, Dieberg G, Smart NA. Exercise training for
management of peripheral arterial disease: A systematic
review and meta-analysis. Sports Med. 2015; 45: 231–44.
49. Gardner AW, Poehlman ET. Exercise rehabilitation pro-
grams for the treatment of claudication pain. A meta-
analysis. JAMA. 1995; 274: 975–80.
50. Tisi PV, Shearman CP. The evidence for the exercise
induced inflammation in intermittent claudication:
Should we encourage patients to stop walking? Eur J
Vasc Endovasc Surg. 1998; 15: 7–17.
51. Mika P, Spodaryk K, Cencora A, Unnithan VB, Mika A.
Experimental model of pain-free treadmill training in
patients with claudication. Am J Phys Med Rehabil. 2005;
84: 756–62.
52. Gardner AW, Montgomery PS, Flinn WR, Katzel LI.
The effect of exercise intensity on the response to exercise
rehabilitation in patients with intermittent claudication. J
Vasc Surg. 2005; 42: 702–9.
53. Mika P, Konik A, Januszek R et al. Comparison of two
treadmill training programs on walking ability and
endothelial function in intermittent claudication. Int J
Cardiol. 2013; 168: 838–42.
54. Stewart KJ, Hiatt WR, Regensteiner JG, Hirsch AT.
Exercise training for claudication. N Engl J Med. 2002;
347: 1941–51.
376 © 2015 Ruijin Hospital, Shanghai Jiaotong University School of Medicine and John Wiley Sons & Australia, Ltd
X. LYU et al.
55. Haas TL, Lloyd PG, Yang HT, Terjung RL. Exercise
training and peripheral arterial disease. Compr Physiol.
2012; 2: 2933–3017.
56. Laufs U, Werner N, Link A et al. Physical training
increases endothelial progenitor cells, inhibits neointima
formation, and enhances angiogenesis. Circulation. 2004;
109: 220–6.
57. Holm J, Dahllöf AG, Scherstén T. Metabolic activity of
skeletal muscle in patients with peripheral arterial insuf-
ficiency. Effect of arterial reconstructive surgery. Scand J
Clin Lab Invest. 1975; 35: 81–6.
58. Hiatt WR, Regensteiner JG, Wolfel EE, Carry MR,
Brass EP. Effect of exercise training on skeletal muscle
histology and metabolism in peripheral arterial disease. J
Appl Physiol. 1996; 81: 780–8.
59. Kojda G, Cheng YC, Burchfield J, Harrison DG. Dys-
functional regulation of endothelial nitric oxide synthase
(eNOS) expression in response to exercise in mice lacking
one eNOS gene. Circulation. 2001; 103: 2839–44.
60. Lima A, Ritti-Dias R, Forjaz CL et al. A session of resis-
tance exercise increases vasodilation in intermittent clau-
dication patients. Appl Physiol Nutr Metab. 2015; 40:
59–64.
61. Januszek R, Mika P, Konik A, Petriczek T, Nowobilski
R, Niżankowski R. Effect of treadmill training on
endothelial function and walking abilities in patients with
peripheral arterial disease. J Cardiol. 2014; 64: 145–51.
62. Collins EG, Bammert C, Edwards LC et al. Pole striding
exercise and vitamin E for management of peripheral
vascular disease. Med Sci Sports Exerc. 2003; 35: 384–93.
Supporting information
Additional Supporting Information may be found in the
online version of this article at the publisher’s web-site:
Methods S1 PubMed search strategy.
Figure S1 Sensitivity analyses by omitting one study at a
time for maximal walking distance.
Figure S2 Galbraith plot of maximal walking distance.
Figure S3 Subgroup analyses of maximal walking dis-
tance by different peripheral arterial disease symptoms
and pain-free walking distance.
Figure S4 Galbraith plot for pain-free walking distance.
Figure S5 Subgroup analyses of pain-free walking dis-
tance (PFWD) by different peripheral arterial disease
symptoms and baseline PFWD.
Figure S6 Subgroup analyses of 6-min walking distance
by different peripheral arterial disease symptoms and
baseline pain-free walking distance.
Figure S7 Effect of intensive walking exercise on the
ankle–brachial index.
Figure S8 Effect of intensive walking exercise on the
post-exercise ankle–brachial index.
Figure S9 Effect of intensive walking exercise on the
Walking Impairment Questionnaire distance score.
X. LYU et al.
Figure S10 Effect of intensive walking exercise on the
Walking Impairment Questionnaire speed score.
Figure S11 Effect of intensive walking exercise on the
Walking Impairment Questionnaire stair-climbing score.
Figure S12 Subgroup analyses of the effect of intensive
walking exercise on Walking Impairment Questionnaire
distance score by different peripheral arterial disease
symptoms.
Figure S13 Subgroup analyses of the effect of intensive
walking exercise on Walking Impairment Questionnaire
© 2015 Ruijin Hospital, Shanghai Jiaotong University School of Medicine and John Wiley Sons & Australia, Ltd
Walking for peripheral arterial disease
speed score by different peripheral arterial disease
symptoms.
Figure S14 Subgroup analyses of the effect of intensive
walking exercise on Walking Impairment Questionnaire
stair-climbing score by different peripheral arterial
disease symptoms.
Table S1 Subgroup analyses of maximal walking
distance.
Table S2 Subgroup analyses of pain-free walking
distance.
377