Controversies in

Cardiovascular Medicine
In Clinical Trials, Is the 6-Minute
Walk Test a Better Functional Test of
Interventions for Peripheral Artery
Disease Than Treadmill Walking Tests?
The Treadmill Is a Better Functional Test Than the 6-Minute
Walk Test in Therapeutic Trials of Patients With Peripheral
Artery Disease
William R. Hiatt, MD; R. Kevin Rogers, MD; Eric P. Brass, MD, PhD

atients with peripheral artery disease (PAD) have a severe

functional limitation in community walking ability, peak
exercise performance, and activities of daily living.13 The
limb manifestations of PAD that are associated with reduced
exercise performance not only are classic intermittent claudication but also include atypical symptoms ranging from leg
discomfort during exercise to exercise limitations without any
clearly referable symptoms in the lower extremity.4 Despite
the diverse nature of limb symptoms during exercise, patients
with PAD uniformly have a severe reduction in peak and
endurance exercise performance.5,6 Thus, in this article, the
term PAD-limited is used to define the exercise impairment
that is well described in PAD, replacing the term claudicationlimited to better characterize the diverse nature of exerciselimiting symptoms described in the disease.

Response by McDermott et al p 78
The quantification of the exercise limitation in PAD and
the implications for daily walking ability and community
activities can be conceptualized by understanding the metabolic work required to perform a range of activities. The
metabolic equivalent of task (MET) is based on a multiple

of the resting metabolic rate (3.5 mL oxygenmin1kg1)

and can be used to characterize a given activity. The MET
can be used to rank physical activities from very mild
(walking at a slow pace is 2 METs) to intense (running at
6 mph is 10 METs).7 Common activities such as walking
on level ground at a speed >3 mph, vacuuming, shopping
for groceries, and transporting packages all require 4 to 5
METs.7 Measurements of oxygen consumption can also be
made to characterize both submaximal and maximal workloads and related to METs. Typically, patients with PAD
have a PAD-limited peak exercise performance of 15 mL
oxygenmin1kg1 or the equivalent of 4 to 5 METs, which
is half that expected in an age-matched healthy control
subject.6 Thus, limitations in peak exercise performance
can translate directly to an inability to complete common
activities in the PAD population.8 Furthermore, the higher
the percentage of maximal capacity required to perform a
daily task is, the harder the task will seem to the patient and
the shorter the patient will be able to engage in the activity.
Practically, this is equivalent to patient-perceived endurance
while performing work and is dependent on the intensity
of the specific activity. As a result, characterization of peak

The opinions expressed in this article are not necessarily those of the editors or of the American Heart Association.
From the University of Colorado School of Medicine, Department of Medicine, Division of Cardiology and CPC Clinical Research, Aurora (W.R.H.,
R.K.R.); and Harbor-UCLA Center for Clinical Pharmacology, Torrance, CA (E.P.B.).
This article is Part II of a 2-part article. Part I appears on p 61.
Correspondence to William R. Hiatt, MD, Professor of Medicine/Cardiology, University of Colorado School of Medicine, C/O CPC Clinical Research,
13199 E Montview Blvd, Ste 200, Aurora, CO 80045. E-mail Will.Hiatt@UCDenver.edu
(Circulation. 2014;130:69-78.)
2014 American Heart Association, Inc.
Circulation is available at http://circ.ahajournals.org

DOI: 10.1161/CIRCULATIONAHA.113.007003

Downloaded from http://circ.ahajournals.org/

by guest on July 30, 2015
69

70CirculationJuly 1, 2014

exercise performance can be used to assess a patients ability

to conduct a variety of normal ambulatory activities, both in
an absolute sense (can or cannot do) and as related to endurance or perceived exertion for a given task.
The primary treatment goal in ameliorating symptoms of
PAD, independently of the type of intervention (eg, revascularization, exercise training or pharmacological therapy), is
to improve the patients ambulatory function and quality of
life. The effectiveness of an intervention can be understood
on the basis of improvements in peak and submaximal exercise capacity because these in turn are determinants of ambulatory function. This improvement can best be assessed by
direct, formal exercise testing. Surrogates of clinical benefit
used to assess the results of a revascularization, including
changes in hemodynamics such as the ankle-brachial index
or changes in vascular anatomy as assessed by vascular
imaging such as duplex ultrasound, magnetic resonance,
or computed tomography imaging, do not directly provide
insight into the resultant functional improvements.6,9 In contrast, an exercise test is a direct surrogate for what matters
to the patient: improved physical functioning and ability
to perform and sustain activities important to that patients
quality of life.

Pathophysiology of the Exercise

Limitation in PAD
In healthy subjects, maximal exercise capacity reflects the
integrated physiological limitation of central hemodynamics (ability of the lungs to allow the transfer of oxygen to
the blood and the ability of the heart to deliver blood to the
lower extremities), the ability of the muscle to extract oxygen and use it for energy production, and the ability of the
muscle to translate chemical energy into mechanical work.10
Each component of this system is essential, and changes in
any one can result in a change in peak exercise performance.
For example, exercise training in healthy subjects induces an
array of adaptations, including an increase in muscle mitochondrial content and thus an improvement in the ability to
generate energy with the delivered oxygen.11,12 In disease
states, impairment in any component will result in a physiological limitation reflected in reduced exercise capacity
such as in heart failure or lung disease that affect central
hemodynamics.13
Similarly, in PAD, the complex insult of chronic ischemia
results in changes in both blood delivery and efficient use of
the oxygen by skeletal muscle, leading to a unique pathophysiological limitation.14,15 Patients with PAD have a limitation in
blood flow to the skeletal muscle, resulting in a mismatch of
oxygen supply to meet metabolic demand.16 In the absence of
critical limb ischemia, symptoms occur only during exercise
because blood delivery is adequate to meet resting bioenergetic requirements. However, inadequate blood flow and oxygen delivery during exercise do not fully define the functional
deficit in PAD. There are secondary sequelae to the perfusion defect that lead to diverse changes to the distal perfusion

bed, including to the microvasculature and muscle function.17

Chronic oxidant stress may be a common injury mechanism in PAD. These secondary injury phenomena include
mitochondrial dysfunction such that oxygen consumption is
impaired even when oxygen delivery is adequate at the onset
of exercise.18,19
Further evidence of the complex pathophysiology of claudication is the observation that exercise training results in a
substantial increase in peak walking time (PWT), which is
often greater than the increase in peak oxygen consumption, despite correlation in the change between the 2 parameters.20,21 The improvement in peak oxygen consumption is
consistent with a classic training response, but other factors
such as improved bioenergetics (increase in ATP generated
per oxygen consumed) may also contribute to the improved
performance in patients with PAD after training. These exercise improvements are not correlated with changes in limb
hemodynamics but may be associated with changes in skeletal muscle metabolism and capillary density.22,23 In contrast,
revascularization results in substantial improvements in limb
blood flow and oxygen delivery, but these improvements in
limb flow are not well correlated with the change in exercise
performance.24
The recognition that a limitation in maximal exercise performance is a direct result of a complex pathophysiology and
that both endurance and submaximal exercise performance
are inherently impaired when maximal performance is limited
has important implications for considering how to assess function and response to treatment in these patients.

Optimal Functional Assessment

of Interventions for PAD
Several characteristics of an optimal functional test can be
postulated to assess the results of treatment of or an intervention in PAD. This is particularly important when the test
serves as a primary or secondary end point in a clinical trial
of a new treatment used for regulatory approval. Table1 summarizes several key concepts in functional testing. As noted,
patients with PAD have a severe limitation in functional
capacity related to a complex pathophysiology. The optimal
test should directly reflect the patients limitation and the
underlying disease characteristics. Ideally, the functional test
should be able to assess the impact of an intervention even
when the mechanism of benefit is not fully understood (eg,
cilostazol).
Patients with PAD have a wide range of functional limitations, from severe challenges in ambulation (unable to walk
50 m) to minimal physical restrictions (able to walk 1 km but
limited when jogging). Therefore an optimal test should have
acceptable performance characteristics across a wide dynamic
range of exercise limitations typical of the disease population.
In the context of a clinical trial, the baseline performance should
be reliably established to assess changes in performance with an
intervention. Furthermore, when patients undergo repeat testing
over a period of time in the absence of an intervention, the test

Downloaded from http://circ.ahajournals.org/ by guest on July 30, 2015

Hiatt et al 6-Minute Walk for PAD Outcomes 71

Table 1. Characteristics of an Optimal Functional Test in Clinical Trials for PAD

Characteristic

Optimal Test

Characterizes functional limitation associated with disease state

Test reflects underlying disease pathophysiology

Large dynamic range

Reliably assesses broad range of functional limitations in the disease population

Baseline

Baseline value clearly established

Reproducible

Provides stable results over time in the absence of an intervention or disease progression

Correlates with patient-reported outcomes

Directly correlates with patient-reported functional status and physical limitations

Sensitive to detecting treatment effect

Detects a change in function with an active treatment with acceptable effect size
Pooled standard deviation of the test is small relative to the treatment effect
Has implications for sample size calculations

Established minimum clinically important difference

Defines the amount of change below which the change is not clinically relevant
Used to differentiate a statistical difference from a clinically meaningful difference.

Safe and well tolerated

Has low risk of cardiovascular or orthopedic events

Is acceptable and not burdensome to patients

Used in multicenter trials

Extensive experience in performing the test across diverse countries and study sites
Site training produces consistent results across centers

Cost

Acceptable costs of equipment and personnel to conduct the test

Equipment available at a wide range of sites

PAD indicates peripheral artery disease.

should provide reproducible results. As a result, in the context

of a clinical trial, patients randomized to a placebo group should
optimally have stable responses over time.
Critically, any change in test performance should be directly
correlated with a patients perception of his or her limitation
and change in ambulatory functional status. Data supporting
this relationship between laboratory test performance and
patients outpatient, real-world function are critical to evaluating the utility of the laboratory measurement. In this context,
the exercise test is really a surrogate for the primary outcome of
interest, that is, improvement in the patients functional status
associated with a wider range of physical activities. In addition,
a quantified unit of change on the test should be interpretable
in the context of a minimum detectable difference, which is
related to the test variance, and minimum clinically important
difference, which is the smallest change in the exercise test that
the patient would report as clinically important.6,25,26
When used to assess the results of an intervention, all functional tests have inherent variability. Typically, there are 2 key
factors: the actual patient variability in the response to a treatment and the assay reproducibility of the test. The reproducibility of the test can be described by the repeat test coefficient
of variance in which the same subject undergoes repeat testing
under stable clinical conditions over a period of days or weeks.
However, in the context of an intervention, the standardized
effect size is a useful way to relate the ability to detect change
on the basis of the heterogeneity of the studied population. The
standardized effect size is the difference in treatment responses
between the active and control arms of a trial normalized to the
pooled standard deviation of the test (variance of the control
and experimental groups combined).27 The population variability (pooled standard deviation) reflects both the test variability
and the response variability. Thus, if a treatment has a very large
effect (eg, exercise training), the sample size to demonstrate a

statistically significant difference between groups can be relatively small if the variance in the response across the population is small. A larger variance will result in a larger sample
size to achieve statistical significance. In contrast, if the treatment effect is relatively small (eg, drug therapy), then controlling the variability and pooled standard deviation of the test is
critically important to identify a statistically significant result
with a limited sample size. The standardized effect size can be
divided into treatments with low, medium, and large effects,
but this approach to grading the magnitude of response on a
treadmill (or other functional test) may not relate the degree
of improvement in the test to what the patient would regard as
important in his or her daily activity.
Any functional test performed in PAD should be safe and
well tolerated. The test should have minimal risk in this
patient population with underlying cardiovascular disease
and increased risk of cardiovascular events.28 The burden of
the test should be acceptable for patients because in a clinical
trial the test will be repeated over time. Equipment for the test
should be readily available and should be standardized across
study sites when used in the context of a multicenter clinical
trial. The cost of equipment and performance of an individual
test should be reasonable. Finally, personnel at the study sites
will typically need to be trained in conducting a standardized and reproducible test. The testing procedures should be
acceptable to study sites in a clinical trial.

Treadmill Exercise Testing

Methodologies in PAD
PAD Treadmill Protocols
Standardized methods for treadmill exercise testing have
been developed for a variety of populations.29 In PAD, there

Downloaded from http://circ.ahajournals.org/ by guest on July 30, 2015

72CirculationJuly 1, 2014

are 2 basic treadmill exercise protocols: the constant-load

test and the graded test. The constant-load test is performed
at a single work rate, typically 2 mph (3.2 km/h) at a grade
of 10% to 12%.30,31 The constant-load protocol evolved out
of the vascular laboratory many years ago where an exercise
test was used to provoke a drop in ankle pressure to facilitate the diagnosis in patients with mild PAD. The results
of a constant-load test are often expressed in units of distance walked, but this approach does not permit an accurate assessment across the range of functional impairment
in the broad population of PAD subjects. For example, the
often used workload of 10% to 12% grade is quite extreme
and often exceeds a patients individual functional limitation, resulting in an inability to conduct the test. In contrast,
some patients with mild PAD will be able to walk for an
extended period on the constant-load protocol without being
limited by claudication and thus precluding assessment of
their maximal limitation. Thus, only a limited spectrum of
the PAD population can be assessed with the constant-workload treadmill test.
In contrast, the most common graded test used in PAD
begins the workload at 2 mph, 0% grade.6,32 This work rate
is typical of low-intensity (2 METs) activities of daily living and is tolerated by the vast majority of patients with
symptomatic PAD. Every 2 minutes, the grade is increased
2% until the patient is limited by maximally tolerated claudication pain and can walk no further. If a patient with
PAD does not have classic claudication, the same test can
be done. In this situation, the patient is limited by his or
her atypical or nonspecific symptoms, but nonetheless premature, muscle fatigue. With a progressive increase in the
workload, each patient is taken to an individually defined,
PAD-limited peak exercise performance so that even
patients with mild PAD will reach a PAD-limited workload.6,32 Thus, the graded treadmill test has a large dynamic
range that can reproducibly define an individual patients
peak PAD-limited PWT and claudication onset time that
occurs before peak workloads.33 In addition, measurements
of oxygen consumption have been used to validate the PWT
and to reflect the underlying pathophysiology of altered
oxygen delivery and consumption. In research programs,
measurement of ventilatory gas exchange can be included
in the test,34 but this requires more specialized equipment,
is more demanding of subjects, and thus is not practical in
most clinical trials.
The use of graded exercise testing is well established and
accepted in clinical medicine. Graded testing has been used
in clinical cardiology for decades to help evaluate cardiac
ischemia and functional limitations in patients with coronary
artery disease.35 The philosophy of increasing workload to
create a metabolic imbalance in heart muscle is at the core of
the Bruce protocol and other graded tests in cardiology and is
directly analogous to the situation in the lower limb of patients
with PAD.36 This history of extensive previous use also means
that high-quality graded treadmills are widely available. Thus,

PWT on a graded treadmill test represents a true physiological

maximal performance based on the individual patients limb
pathophysiology that can easily be performed in diverse clinical settings.

Conduct of the Exercise Test

The utility of any exercise test is dependent on proper
implementation. Thus, it is important that the graded
treadmill exercise test be performed under standardized
and reproducible conditions. Typical instructions are to
have patients tested in the morning after an overnight fast
so that food effects can be minimized. The patient should
not walk long distances to the testing laboratory (to avoid
claudication), should be rested, and should avoid smoking
cigarettes and drinking alcohol before the test. Comfortable
clothes should be worn, and the timing device used to
assess test duration should be hidden from the patient so as
not to bias their performance. The patients should initiate
the test by straddling the moving belt at 2 mph and start
walking at that speed at time 0. This will require familiarization of all patients before the conduct of a formal test
for analysis. During the conduct of the test, patients should
not hold onto the treadmill bars for support or off-loading
because this will alter the subjects performance and introduce variability in the test. Testing laboratories should consistently instruct patients to walk to maximally tolerated
claudication or other limb pain/limitation. For example,
some investigators instruct their patients to stop walking
on the treadmill when they experience a usual intensity of
claudication or other limb pain. However, these instructions will, by definition, artificially limit that patients peak
performance below their true pathophysiological maximal
limitation. This approach will result in data that are less
reproducible and less sensitive to the intervention with less
predictive value with respect to ambulatory performance.
Thus, all patients should be pushed to maximally tolerated pain that results in an inability to walk any further.
Although the safety of treadmill testing has not been evaluated extensively in the PAD population, 1 study included
comprehensive hemodynamic and ECG monitoring during
exercise in a large cohort and identified no safety concerns,
including no changes indicative of cardiac ischemia during testing.37 This finding is consistent with the reported
adverse event experience from clinical trials using peak
treadmill testing involving thousands of patients with PAD.
However, most patients with PAD have underlying coronary artery disease; therefore, ECG monitoring during all
peak treadmill exercise tests in this population is recommended to enhance patient safety.

Challenges in Performing a Graded Treadmill Test

in a Clinical Trial
There are several challenges in performing a graded treadmill test across multiple sites in a clinical trial. Not all sites
maintain appropriate equipment; some treadmills are simply

Downloaded from http://circ.ahajournals.org/ by guest on July 30, 2015

Hiatt et al 6-Minute Walk for PAD Outcomes 73

unable to perform the test according to the study protocol.38

For example, treadmills may not be programmed correctly,
may not increment the grade correctly, or may rely on manual changes in grade that are often subject to error. Treadmill
speed must be calibrated and stable over the entire testing
range. These issues need to be identified and corrected before
a site enrolls patients in a clinical trial. In addition, site personnel and study subjects may not be fully trained on the proper
conduct of the test. All subjects, particularly patients with a
chronic disease, must be familiarized with the testing procedures before performing an exercise test for data analysis. In
addition, site staff should use standardized methods for initiating the test (proper transition from rest to 2 mph, 0% grade)
and completing the test by pushing the patient to a maximal,
PAD-limited performance.
Thus, assessing the treadmill testing equipment and staff
competency at a site is critically important to achieve uniform treadmill end points of claudication onset time and PWT
across a multicenter trial. Routine clinical trial site regulatory monitoring usually does not involve an evaluation of the
actual conduct of the test. Several methods have been proposed to ensure that study sites are properly trained and comply with standardized testing procedures. One approach to this
standardization is the site end-point evaluation visit, which is
an extension of the usual monitoring visit. The site end-point
evaluation visit evaluates the capability and equipment of the
site to perform a test and observes the performance of the test
at the site using mock testing methodologies. This allows further training of the site staff to provide a uniform approach
across sites to reproducibly perform the exercise test. Using
these methodologies and defining other aspects of test performance have provided much greater control of the variance of
the exercise test between and within individuals and sites in
the conduct of a clinical trial. Of note, this approach to control
of testing conditions is likely applicable to all exercise testing
modalities, including the 6-minute walk test.

Treadmill Reliability
The test characteristics of treadmill testing have been studied extensively in PAD. For example, numerous studies have
evaluated the reliability of the constant-load compared with
the graded treadmill test. A meta-regression analysis of 8
studies compared the 2 tests using the intraclass correlation
coefficient to determine the consistency between measurements over time.39 It should also be noted that these 2 exercise
tests have different relationships between workload and PWT.
The constant-load test has a linear relationship, whereas the
graded test has a curvilinear relationship (higher work rates
are associated with progressively shorter incremental walking times).39 The reliability of PWT was significantly better
with the graded test (intraclass correlation coefficient, 0.95)
compared with the constant-load test (intraclass correlation
coefficient, 0.90). Consistent with the limitations enumerated
above, the reliability of the constant-load test was also dependent on the fixed grade of the treadmill (poor at 0% grade and

better at 12% grade), whereas the graded test performed well

across the full dynamic range of patients studied. Other studies have also confirmed better reproducibility with the graded
compared with the constant-load test in PAD.40 Importantly, in
evaluations of patients with PAD, the high baseline reproducibility of the graded treadmill has been confirmed with the use
of data from a multicenter trial.41 The intrasubject coefficient
of variation averaged 15% and was independent of the baseline exercise performance (the test was equally reproducible
across the full range of baseline PWT values). Thus, although
both treadmill protocols are acceptable, the graded test better
reflects the mechanism of the performance limitations in PAD,
has the greatest reliability, and allows testing a wide range of
functional disease severity in PAD.
A major challenge of any treadmill test in PAD is an apparent
improvement in peak or submaximal performance measures
over time in the absence of an intervention. This phenomenon has also been called the placebo response because of the
increase in performance observed in patients randomized to a
placebo group over 6 months of observation.31 In prior studies
using a constant-load exercise test, the placebo response has
varied from 25% to 100%.42,43 However, contemporary multicenter trials using the graded treadmill test and site training
(site end-point evaluation visit) have controlled this placebo
response to 9% to 12% at 3 months and 13% to 23% at 6
months.44 The cause of this placebo response is likely multifactorial. The subjects comfort and adjustment to walking on
the treadmill, including optimization of gait, likely contribute.
In principle, if the maximum performance is properly established, variability over time should be very small because of
the fixed physiological limitation. The observed changes are
also consistent with a change in walking efficiency as a contributing factor.
An additional factor affecting the variance of any exercise
test in a clinical trial is the definition of the baseline performance before randomization to an intervention. A recent study
carefully evaluated the definition of baseline exercise performance for the graded treadmill test. In the context of a multicenter clinical trial, patients performed 3 consecutive treadmill
tests 3 to 10 days apart before randomization to treatment with
an investigational drug, an active comparator drug, or placebo.41
The data were evaluated by assessing various combinations of
baseline definitions. The methods that resulted in the lowest
placebo response and the largest standardized effect size were
defining baseline either by the first of the 3 tests or by the highest of the 3 tests, whereas the characteristics of averaging the
3 tests or any set of 2 tests were inferior.41 Using the highest
baseline value (which we recommend) supports the concepts
that the patient cannot exceed his or her true pathophysiologic
limitation on a graded treadmill test and that this limit will be
achieved with proper test execution.

Treadmill Sensitivity to Treatment Effect

When used as a primary end point, PWT assessed with a
graded treadmill test is sensitive to change with a variety

Downloaded from http://circ.ahajournals.org/ by guest on July 30, 2015

74CirculationJuly 1, 2014

of interventions. Statistically significant responses relative

to a control group have been documented for exercise rehabilitation,20,21 revascularization,45,46 and a variety of medications.26,37,44,47 The treadmill has been used in several drug trials
of agents, including those with small effect sizes. For example, a meta-analysis of propionyl-L-carnitine demonstrated an
effect size in a range of 0.11 to 0.22, whereas phosphodiesterase-3 inhibitors have effect size estimates in the range of
0.28 to 0.32.26,37,47 Indeed, no intervention demonstrated to be
effective has failed to demonstrate improvement in 1 or more
clinical trials using graded treadmill testing.

The Treadmill as a Surrogate for Patient

Ambulatory Function and Quality of Life
Treadmill performance has been related to several measures
of ambulatory activity. Performance on the graded treadmill is
highly correlated with the Walking Impairment Questionnaire,
which is a patient-based assessment of challenges of walking defined distances, walking speeds, and severity of calf
claudication pain during community walking activities.3 The
Walking Impairment Questionnaire score can range from 0.00
to 1.00 (or 0% to 100%), with higher scores indicating better
walking ability. There are also subscales for walking distance,
speed, stairs, and claudication pain severity. The relationship
between improvements in treadmill performance and change in
Walking Impairment Questionnaire scores was further evaluated in a study of exercise training.48 This study demonstrated
that a change of 0.1 in Walking Impairment Questionnaire
score (mean of the subscales) corresponded to a 345-m change
in peak treadmill walking distance with exercise training.
Although the minimum clinically important difference was not
formally assessed, there was a suggestion that different quartiles in treadmill responses could identify potential thresholds
for patient-assessed improvements in walking impairment and
functional status. In addition, patient-reported outcomes have
been assessed in trials of exercise training, drugs, and revascularization. In all these studies of a variety of interventions,
changes in PWT on a graded treadmill test were well correlated
with improvements in patient-assessed outcomes and healthrelated quality of life.20,21,26,49 Maximal walking distance on a
progressive treadmill protocol was also correlated with outdoor
walking distance measured with a global positioning system.50
These relationships between the graded treadmill and patientreported outcomes provide evidence that understanding peak
PAD-limited exercise capacity is a relevant parameter in also
understanding a patients daily activities that are performed at
work intensities well below the patients peak capability. Thus,
treadmill testing provides a facile laboratory measure reflecting
patients ambulatory status and function. The wide acceptance
of treadmill testing generally, and application to PAD in particular, led the European Agency for the Evaluation of Medicinal
Products to recommend treadmill testing in its guidance on
clinical investigation of medical products for PAD.51,52

Assessment of Performance and Response

to Treatment With the Treadmill
Versus the 6-Minute Walk Test
The 6-minute walk test is an attractive assessment modality on many levels. It is simple to perform, requires minimal
equipment, and has been used in a variety of disease states,
including in patients with pulmonary disease and heart failure,
but less so as an end point in PAD.5356 In PAD, although the
graded treadmill test and 6-minute walk test are correlated,
the relationship is relatively weak (r2=0.28), indicating that
peak performance on a graded treadmill may reflect different
pathophysiological limitations than does a submaximal effort
with the 6-minute walk test.57
In contrast to the extensive experience with the treadmill,
equivalent data for performance of the 6-minute walk test in
patients with PAD are not available. However, data from other
disease populations suggest that there are several factors in the
conduct of the 6-minute walk test that will affect outcome. For
example, in patients with chronic lung disease, course length
can influence the measurement.58 As with a treadmill test, precision of the 6-minute walk test is improved with the use of the
higher of 2 baseline measurements59 and with test familiarization.60 In chronic lung disease, a minimal clinically important
difference has been defined for the 6-minute walk test predicting mortality (a loss of 30 m).61 In contrast, in patients with
heart disease, the 6-minute walk test demonstrated poor testretest reliability, as well as a placebo response and time-ofday variability.62,63 A minimum clinically important difference
has been suggested in chronic lung disease and in heart failure
that defined thresholds for change in a patient-reported outcome scale, but in heart failure, this difference was the same
as the variance of the test and hence the minimum detectable
difference.55,64 Thus, the test would be of marginal utility in
assessing an intervention with an effect size that approximated
the minimum clinically important difference.
When used in PAD, the 6-minute walk test has demonstrated a loss of patient performance over several years of
follow-up that is related to a variety of clinical and systemic
factors such as level of chronic inflammation.65,66 However,
any relationship between this loss of walking performance
over time and patient assessment of changes in quality of life
was not assessed. Similar data on functional decline have not
been obtained with the treadmill test, but this particular outcome (preventing clinical disease progression) has not been
the target of interventions tested in PAD randomized, clinical
trials. In addition, when evaluated in the context of an intervention, an endurance test such as the 6-minute walk has no
fixed physiological correlate to support a change in function
because the test is performed under submaximal exercise conditions. Thus, performance (6-minute distance walked) may
be susceptible to environmental and intervention-independent
factors. Additionally, patient and investigator expectations of
benefit may affect the submaximal test stability over time.
This is particularly important in the context of a randomized
trial in which blinding to the intervention may be impossible

Downloaded from http://circ.ahajournals.org/ by guest on July 30, 2015

Hiatt et al 6-Minute Walk for PAD Outcomes 75

(eg, exercise training) or compromised by factors such as drug

adverse events. In contrast to the abundant data validating
treadmill performance as a surrogate of ambulatory patient
function, there are fewer data exploring this relationship for
the 6-mnute walk test in PAD subjects.1
In PAD, very few studies have offered a head-to-head comparison of the 6-minute walk with graded treadmill testing to
evaluate an intervention, but 2 studies have been published
by McDermott et al.67,68 In 1 single-site study, the response
to supervised exercise training over 6 months was evaluated
with both tests.67 In the control group, both measures were
relatively stable over 6 months, with the 6-minute walk test
demonstrating a loss of walking distance of 5% the graded
treadmill demonstrating a 7.6% increase. With the exercise
intervention, the 6-minute walking distance improved 6.4%
(P<0.001 versus baseline), whereas the graded treadmill test
improved 52% (P<0.001 versus baseline). Variance estimates
of the within-group changes in the exercise group for both
tests were determined from data provided in the article, leading to a calculation of a standardized effect size (difference
in treatment response between groups divided by the withinsubject standard deviation). The between-group comparisons
of the exercise and control group changes were statistically
significant for both the 6-minute walk test and the treadmill

test. On the basis of those changes and data from the publication, the 6-minute walk test had an estimated effect size
of 0.70 compared with an estimate of 1.01 for the graded
treadmill. These differences in effect size evaluating the same
intervention have implications for future trials in which the
graded treadmill test (greater observed effect size) would
require fewer patients to demonstrate statistical significance
than the 6-minute walk test. The other single-site study evaluated home-based exercise and suggested that the estimate
of the treatment effect (effect size) was 2.4 times greater
with the graded treadmill test compared with the 6-minute
walk test.68 Taking these results taken together and combining
them with the characterization of each test in PAD and other
conditions indicate that the use of the 6-minute walk test as
a primary end point in assessment of an intervention would
require a larger sample size than if the graded treadmill were
used. In contrast to the substantial multicenter experience
with treadmill testing in a real-world setting, the above data
are from a single-site study. Experience suggests that singlesite test performance should be extrapolated with caution to
multicenter trials. These issues, including experience with
the treadmill and 6-minute walk test in multicenter trials, are
summarized in Table2.69

Table 2. Characterization of the Exercise Treadmill Test Compared With the 6-Minute Walk Test in Clinical Trials in PAD
Characteristic

Graded Treadmill

6-Minute Walk

Characterizes disease functional limitation

Results informative for both submaximal and maximal

performance

Characterizes submaximal walking endurance

Dynamic range

Accurate assessment over a wide range of exercise

limitations

Less well studied in PAD

Baseline

Highest value of a series of tests at baseline facilitates

best estimate of treatment effect41

Not established in PAD

In pulmonary disease, higher of 2 tests recommended59

Experience

Assessed in multicenter trials

Assessed in single-site studies

Reproducible
Placebo response

Controlled with proper conduct of the test67

Controlled with proper conduct of the test67

Variability

Coefficient of variation 12.6%

Coefficient of variation 10.4%57

Correlation of performance parameter with

patient-reported outcomes after treatment

Correlations observed in studies of drugs, exercise, and

revascularization20,21,26,49

Correlation not evaluated but improvement seen in

exercise studies67,68

57

Sensitive to treatment effect

Effect size, medication

0.110.3226,47

Not well studied

Effect size, exercise training

0.871.0146,67

0.7067

Effect size, revascularization

0.47

Minimum clinically important difference

Not established in PAD

Not established in PAD; thresholds suggested in

chronic lung disease and heart failure54,55,64

Safe and well tolerated

Extensive experience without unacceptable dropouts

resulting from adverse events
Cardiac safety well established in PAD population37

Not well studied but test is less burdensome than a

treadmill

Multicenter clinical trials

Broad experience in diverse sites across the world69

Study sites require training

Most experience limited to single-site studies

Likely more acceptable to diverse sites than treadmill
testing

Costs

Requires a programmable treadmill and ECG

monitoring

Requires no specialized equipment

46

Not well studied

PAD indicates peripheral artery disease.

Downloaded from http://circ.ahajournals.org/ by guest on July 30, 2015

76CirculationJuly 1, 2014

Conclusions
The primary goal of any intervention for symptomatic PAD
is to improve the patients ambulatory performance of daily
activities and quality of life. The 6-minute walk test may
have minor advantages over treadmill testing with respect
to study site and patient preferences, considerations that
deserve further evaluation. Its use for other indications also
suggests that it is a useful measure of physical performance.
Nonetheless, the theoretical advantages of the graded treadmill test have been widely confirmed in >30 years of clinical
research. It has a sound physiological basis, has been widely
used in multicenter trials, and has broad acceptance in clinical practice for other conditions. In PAD, the graded treadmill intrasubject and intersubject test characteristics are well
established and support the utility of the test while facilitating
sound power estimates for clinical trials. In contrast to the
6-minute walk test, the validity of treadmill testing in PAD
as a surrogate for patient ambulatory function and quality
of life is well established. Intervention trials in PAD have
repeatedly demonstrated the ability of the graded treadmill
to safely and robustly quantify changes associated with efficacious interventions. Consistent with theoretical considerations and work in other populations, limited data in patients
with PAD support a greater sensitivity to detect change with
an intervention (based on observed effect sizes) with the use
of the graded treadmill test versus the 6-minute walk test.
Future work should further evaluate the 6-minute walk test in
assessing other interventions such as drugs, biological agents,
and revascularization, particularly in terms of the sensitivity
of the instrument to clinically meaningful change in performance. In addition, future studies should better define the
relationship between treadmill performance and optimized
instruments for assessing ambulatory function in patients
with PAD with the goal of developing a value for minimum
clinically important difference.

Disclosures
Dr Hiatt reports grant awards to CPC Clinical Research (a nonprofit academic research organization and affiliate of the University of Colorado)
from the following sponsors: Aastrom, AstraZeneca, National Institutes
of Health, CSI, Cytokinetics, DNAVEC, Kowa, Kyushu University,
Pluristem, Regeneron, Rigel, Takeda. Dr Rogers reports no conflicts. Dr
Brass reports being a consultant to GlaxoSmithKline, Novartis, McNeil
Consumer Pharmaceuticals, Novo Nordisk, 3D Communications,
Catabasis Pharmaceuticals, Allergan, NovaDigm Therapeutics,
NanoBio, Bayer, Endo Pharmaceuticals, Amgen, Boston Scientific,
World Self-Medication Institute, Merck, NPS Pharmaceuticals,
HeartWare International, Takeda Pharmaceuticals, Genzyme, Consumer
Healthcare Products Association, DepoMed, Cangene, Aveo Oncology,
BioMarin, Galderma, QRx Pharma, University of Washington, Amarin,
EnteroMedics, and Trius Therapeutics. Dr Brass has equity in Calistoga
Pharmaceuticals and Catabasis Pharmaceuticals.

References
1. McDermott MM, Liu K, Guralnik JM, Martin GJ, Criqui MH, Greenland
P. Measurement of walking endurance and walking velocity with questionnaire: validation of the walking impairment questionnaire in men and
women with peripheral arterial disease. J Vasc Surg. 1998;28:10721081.

2. Sieminski DJ, Gardner AW. The relationship between free-living daily

physical activity and the severity of peripheral arterial occlusive disease.
Vasc Med. 1997;2:286291.
3. Regensteiner JG, Steiner JF, Panzer RJ, Hiatt WR. Evaluation of walking
impairment by questionnaire in patients with peripheral arterial disease.
J Vasc Med Biol. 1990;2:142152.
4. McDermott MM, Mehta S, Greenland P. Exertional leg symptoms other
than intermittent claudication are common in peripheral arterial disease.
Arch Intern Med. 1999;159:387392.
5. McDermott MM, Greenland P, Liu K, Guralnik JM, Criqui MH, Dolan NC,
Chan C, Celic L, Pearce WH, Schneider JR, Sharma L, Clark E, Gibson D,
Martin GJ. Leg symptoms in peripheral arterial disease: associated clinical
characteristics and functional impairment. JAMA. 2001;286:15991606.
6. Hiatt WR, Nawaz D, Regensteiner JG, Hossack KF. The evaluation
of exercise performance in patients with peripheral vascular disease.
J Cardiopulm Rehabil. 1988;12:525532.
7. Ainsworth BE, Haskell WL, Whitt MC, Irwin ML, Swartz AM, Strath SJ,
OBrien WL, Bassett DR Jr, Schmitz KH, Emplaincourt PO, Jacobs DR Jr,
Leon AS. Compendium of physical activities: an update of activity codes
and MET intensities. Med Sci Sports Exerc. 2000;32(suppl):S498S504.
8. Leon AS, Jacobs DR Jr, DeBacker G, Taylor HL. Relationship of physical characteristics and life habits to treadmill exercise capacity. Am J
Epidemiol. 1981;113:653660.
9. Brass EP, Hiatt WR, Green S. Skeletal muscle metabolic changes in
peripheral arterial disease contribute to exercise intolerance: a point-counterpoint discussion. Vasc Med. 2004;9:293301.
10. Wasserman K. Coupling of external to cellular respiration during exercise:
the wisdom of the body revisited. Am J Physiol. 1994;266(pt 1):E519E539.
11. Turner DL, Hoppeler H, Claassen H, Vock P, Kayser B, Schena F, Ferretti G.
Effects of endurance training on oxidative capacity and structural composition of human arm and leg muscles. Acta Physiol Scand. 1997;161:459464.
12. Tonkonogi M, Sahlin K. Physical exercise and mitochondrial function in
human skeletal muscle. Exerc Sport Sci Rev. 2002;30:129137.
13. Wagner PD. Determinants of maximal oxygen transport and utilization.
Annu Rev Physiol. 1996;58:2150.
14. Brass EP. Intermittent claudication: new targets for drug development.
Drugs. 2013;73:9991014.
15. Muller MD, Reed AB, Leuenberger UA, Sinoway LI. Physiology in medicine: peripheral arterial disease. J Appl Physiol (1985). 2013;115:12191226.
16. Lassen NA, Kampp M. Calf muscle blood flow during walking studied by
the xe-133 method in normals and in patients with intermittent claudication. Scand J Clin Lab Invest. 1965;17:447453.
17. Brevetti G, Giugliano G, Brevetti L, Hiatt WR. Inflammation in peripheral
artery disease. Circulation. 2010;122:18621875.
18. Brass EP, Hiatt WR, Gardner AW, Hoppel CL. Decreased NADH dehydrogenase and ubiquinol-cytochrome c oxidoreductase in peripheral arterial
disease. Am J Physiol. 2001;280:H603H609.
19. Hiatt WR, Wolfel EE, Regensteiner JG, Brass EP. Skeletal muscle carnitine metabolism in patients with unilateral peripheral arterial disease.
J Appl Physiol (1985). 1992;73:346353.
20. Hiatt WR, Wolfel EE, Meier RH, Regensteiner JG. Superiority of treadmill walking exercise versus strength training for patients with peripheral
arterial disease: implications for the mechanism of the training response.
Circulation. 1994;90:18661874.
21. Hiatt WR, Regensteiner JG, Hargarten ME, Wolfel EE, Brass EP. Benefit
of exercise conditioning for patients with peripheral arterial disease.
Circulation. 1990;81:602609.
22. 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 (1985). 1996;81:780788.
23. Robbins JL, Jones WS, Duscha BD, Allen JD, Kraus WE, Regensteiner
JG, Hiatt WR, Annex BH. Relationship between leg muscle capillary density and peak hyperemic blood flow with endurance capacity in peripheral
artery disease. J Appl Physiol (1985). 2011;111:8186.
24. Pernow B, Saltin B, Wahren J, Cronestrand R, Ekestrom S. Leg blood
flow and muscle metabolism in occlusive arterial disease of the leg before
and after reconstructive surgery. Clin Sci Mol Med. 1975;49:265275.
25. Regensteiner JG, Steiner JF, Hiatt WR. Exercise training improves functional status in patients with peripheral arterial disease. J Vasc Surg.
1996;23:104115.

Downloaded from http://circ.ahajournals.org/ by guest on July 30, 2015

Hiatt et al 6-Minute Walk for PAD Outcomes 77

26. Regensteiner JG, Ware JE Jr, McCarthy WJ, Zhang P, Forbes WP, Heckman
J, Hiatt WR. Effect of cilostazol on treadmill walking, community-based
walking ability, and health-related quality of life in patients with intermittent claudication due to peripheral arterial disease: meta-analysis of six
randomized controlled trials. J Am Geriatr Soc. 2002;50:19391946.
27. Cohen J. Statistical Power Analyses for the Behavoioral Sciences.
Hillsdale, NJ: Lawrence Erlbaum Associates; 1988.
28. Brass EP, Hiatt WR. Review of mortality and cardiovascular event rates
in patients enrolled in clinical trials for claudication therapies. Vasc Med.
2006;11:141145.
29. Casaburi R. A primer in cardiopulmonary exercise testing. Monaldi Arch
Chest.Dis. 1993;48:266271.
30. Degischer S, Labs KH, Aschwanden M, Tschoepl M, Jaeger KA.

Reproducibility of constant-load treadmill testing with various treadmill
protocols and predictability of treadmill test results in patients with intermittent claudication. J Vasc Surg. 2002;36:8388.
31. Hiatt WR, Hirsch AT, Regensteiner JG, Brass EP. Clinical trials for claudication: assessment of exercise performance, functional status, and clinical
end points: Vascular Clinical Trialists. Circulation. 1995;92:614621.
32. Gardner AW, Skinner JS, Cantwell BW, Smith LK. Progressive vs singlestage treadmill tests for evaluation of claudication. Med Sci Sports Exerc.
1991;23:402408.
33. Hiatt WR, Goldstone J, Smith SC, Jr., McDermott M, Moneta G, Oka
R, Newman AB, Pearce WH. Atherosclerotic Peripheral Vascular
Disease Symposium II: nomenclature for vascular diseases. Circulation.
2008;118:28262829.
34. Hansen JE, Sue DY, Oren A, Wasserman K. Relation of oxygen uptake
to work rate in normal men and men with circulatory disorders. Am J
Cardiol. 1987;59:669674.
35. Bruce RA, Kusumi F, Hosmer D. Maximal oxygen intake and nomographic assessment of functional aerobic impairment in cardiovascular
disease. Am Heart J. 1973;85:546562.
36. Bruce RA. Exercise testing of patients with coronary heart disease: principles and normal standards for evaluation. Ann Clin Res. 1971;3:323332.
37. Brass EP, Cooper LT, Morgan RE, Hiatt WR. A phase II dose-ranging
study of the phosphodiesterase inhibitor K-134 in patients with peripheral
artery disease and claudication. J Vasc Surg. 2012;55:381389.e1.
38. Hiatt WR, Cox L, Greenwalt M, Griffin A, Schechter C. Quality of the
assessment of primary and secondary endpoints in claudication and critical leg ischemia trials. Vasc Med. 2005;10:207213.
39. Nicola SP, Viechtbauer W, Kruidenier LM, Candel MJ, Prins MH, Teijink
JA. Reliability of treadmill testing in peripheral arterial disease: a metaregression analysis. J Vasc Surg. 2009;50:322329.
40. Labs KH, Nehler MR, Roessner M, Jaeger KA, Hiatt WR. Reliability of
treadmill testing in peripheral arterial disease: a comparison of a constant
load with a graded load treadmill protocol. Vasc Med. 1999;4:239246.
41. Brass EP, Jiao J, Hiatt W. Optimal assessment of baseline treadmill walking performance in claudication clinical trials. Vasc Med. 2007;12:97103.
42. Porter JM, Cutler BS, Lee BY, Reich T, Reichle FA, Scogin JT, Strandness
DE. Pentoxifylline efficacy in the treatment of intermittent claudication:
multicenter controlled double-blind trial with objective assessment of
chronic occlusive arterial disease patients. Am Heart J. 1982;104:6672.
43. Skinner JS, Strandness DE Jr. Exercise and intermittent claudication, I:
effect of repetition and intensity of exercise. Circulation. 1967;36:1522.
44. Brass EP, Anthony R, Cobb FR, Koda I, Jiao J, Hiatt WR. The novel
phosphodiesterase inhibitor NM-702 improves claudication-limited exercise performance in patients with peripheral arterial disease. J Am Coll
Cardiol. 2006;48:25392545.
45. Regensteiner JG, Hargarten ME, Rutherford RB, Hiatt WR. Functional
benefits of peripheral vascular bypass surgery for patients with intermittent claudication. Angiology. 1993;44:110.
46. Murphy TP, Cutlip DE, Regensteiner JG, Mohler ER, Cohen DJ,

Reynolds MR, Massaro JM, Lewis BA, Cerezo J, Oldenburg NC, Thum
CC, Goldberg S, Jaff MR, Steffes MW, Comerota AJ, Ehrman J, TreatJacobson D, Walsh ME, Collins T, Badenhop DT, Bronas U, Hirsch AT.
Supervised exercise versus primary stenting for claudication resulting
from aortoiliac peripheral artery disease: six-month outcomes from the
Claudication: Exercise Versus Endoluminal Revascularization (CLEVER)
Study. Circulation. 2012;125:130139.

47. Brass EP, Koster D, Hiatt WR, Amato A. A systematic review and metaanalysis of propionyl-L-carnitine effects on exercise performance in
patients with claudication. Vasc Med. 2013;18:312.
48. Nicola SP, Kruidenier LM, Rouwet EV, Graffius K, Prins MH, Teijink
JA. The walking impairment questionnaire: an effective tool to assess the
effect of treatment in patients with intermittent claudication. J Vasc Surg.
2009;50:8994.
49. Murphy TP, Reynolds MR, Cohen DJ, Regensteiner JG, Massaro JM,
Cutlip DE, Mohler ER, Cerezo J, Oldenburg NC, Thum CC, Goldberg S,
Hirsch AT. Correlation of patient-reported symptom outcomes and treadmill test outcomes after treatment for aortoiliac claudication. J Vasc Interv
Radiol. 2013;24:14271435.
50. Le Faucheur A, Abraham P, Jaquinandi V, Bouy P, Saumet JL, NouryDesvaux B. Measurement of walking distance and speed in patients with
peripheral arterial disease: a novel method using a global positioning system. Circulation. 2008;117:897904.
51. Committee for Proprietary Medical Products (CPMP). Note for guidance
on clinical investigation of medicinal products for the treatment of peripheral arterial occlusive disease. http://www.ema.europa.eu/docs/en_GB/
document_library/Scientific_guideline/2009/09/WC500003341.pdf.
Accessed April 1, 2014.
52. Labs KH, Dormandy JA, Jaeger KA, Stuerzebecher C, Hiatt WR. TransAtlantic Conference on Clinical Trial Guidelines in PAOD (peripheral
arterial occlusive disease) clinical trial methodology. Eur J Vasc Endovasc
Surg. 1999;18:253265.
53. ATS statement: guidelines for the six-minute walk test. Am J Respir Crit
Care Med. 2002;166:111117.
54. Solway S, Brooks D, Lacasse Y, Thomas S. A qualitative systematic overview of the measurement properties of functional walk tests used in the
cardiorespiratory domain. Chest. 2001;119:256270.
55. Redelmeier DA, Bayoumi AM, Goldstein RS, Guyatt GH. Interpreting
small differences in functional status: the six minute walk test in chronic
lung disease patients. Am J Respir Crit Care Med. 1997;155:12781282.
56. Bittner V, Weiner DH, Yusuf S, Rogers WJ, McIntyre KM, Bangdiwala SI,
Kronenberg MW, Kostis JB, Kohn RM, Guillotte M. Prediction of mortality and morbidity with a 6-minute walk test in patients with left ventricular
dysfunction: SOLVD Investigators. JAMA. 1993;270:17021707.
57. Montgomery PS, Gardner AW. The clinical utility of a six-minute walk
test in peripheral arterial occlusive disease patients. J Am Geriatr Soc.
1998;46:706711.
58. Beekman E, Mesters I, Hendriks EJ, Klaassen MP, Gosselink R, van
Schayck OC, de Bie RA. Course length of 30 metres versus 10 metres has
a significant influence on six-minute walk distance in patients with COPD:
an experimental crossover study. J Physiother. 2013;59:169176.
59. Chandra D, Wise RA, Kulkarni HS, Benzo RP, Criner G, Make B, Slivka
WA, Ries AL, Reilly JJ, Martinez FJ, Sciurba FC. Optimizing the six
minute walk test as a measure of exercise capacity in COPD. Chest.
2012;142:15451552.
60. Jenkins S, Cecins NM. Six-minute walk test in pulmonary rehabilitation:
do all patients need a practice test? Respirology. 2010;15:11921196.
61. Polkey MI, Spruit MA, Edwards LD, Watkins ML, Pinto-Plata V, Vestbo
J, Calverley PM, Tal-Singer R, Agusti A, Bakke PS, Coxson HO, Lomas
DA, MacNee W, Rennard S, Silverman EK, Miller BE, Crim C, Yates J,
Wouters EF, Celli B. Six-minute-walk test in chronic obstructive pulmonary disease: minimal clinically important difference for death or hospitalization. Am J Respir Crit Care Med. 2013;187:382386.
62. Hanson LC, McBurney H, Taylor NF. The retest reliability of the six-minute walk test in patients referred to a cardiac rehabilitation programme.
Physiother Res Int. 2012;17:5561.
63. Kervio G, Ville NS, Leclercq C, Daubert JC, Carr F. Intensity and daily
reliability of the six-minute walk test in moderate chronic heart failure
patients. Arch Phys Med Rehabil. 2004;85:15131518.
64. Shoemaker MJ, Curtis AB, Vangsnes E, Dickinson MG. Clinically meaningful change estimates for the six-minute walk test and daily activity in individuals with chronic heart failure. Cardiopulm Phys Ther J.
2013;24:2129.
65. McDermott MM, Guralnik JM, Ferrucci L, Criqui MH, Greenland P, Tian
L, Liu K, Tan J. Functional decline in lower-extremity peripheral arterial disease: associations with comorbidity, gender, and race. J Vasc Surg.
2005;42:11311137.

Downloaded from http://circ.ahajournals.org/ by guest on July 30, 2015

78CirculationJuly 1, 2014

66. McDermott MM, Liu K, Guralnik JM, Ferrucci L, Green D, Greenland

P, Tian L, Criqui MH, Lo C, Rifai N, Ridker PM, Zheng J, Pearce W.
Functional decline in patients with and without peripheral arterial disease:
predictive value of annual changes in levels of C-reactive protein and
D-dimer. J Gerontol A Biol Sci Med Sci. 2006;61:374379.
67. McDermott MM, Ades P, Guralnik JM, Dyer A, Ferrucci L, Liu K, Nelson
M, Lloyd-Jones D, Van Horn L, Garside D, Kibbe M, Domanchuk K, Stein
JH, Liao Y, Tao H, Green D, Pearce WH, Schneider JR, McPherson D, Laing
ST, McCarthy WJ, Shroff A, Criqui MH. Treadmill exercise and resistance
training in patients with peripheral arterial disease with and without intermittent claudication: a randomized controlled trial. JAMA. 2009;301:165174.

68. McDermott MM, Liu K, Guralnik JM, Criqui MH, Spring B, Tian
L, Domanchuk K, Ferrucci L, Lloyd-Jones D, Kibbe M, Tao H,
Zhao L, Liao Y, Rejeski WJ. Home-based walking exercise intervention in peripheral artery disease: a randomized clinical trial. JAMA.
2013;310:5765.
69. Jaff MR, Dale RA, Creager MA, Lipicky RJ, Constant J, Campbell LA,
Hiatt WR. Anti-chlamydial antibiotic therapy for symptom improvement
in peripheral artery disease: prospective evaluation of rifalazil effect
on vascular symptoms of intermittent claudication and other endpoints
in Chlamydia pneumoniae seropositive patients (PROVIDENCE-1).
Circulation. 2009;119:452458.

Response to Hiatt et al
Mary M. McDermott, MD; Jack M. Guralnik, MD, PhD; Michael H. Criqui, MD, MPH;
Kiang Lui, PhD; Melina R. Kibbe, MD; Luigi Ferrucci, MD, PhD
We agree with Hiatt et al that the primary goal of therapeutic interventions for patients with peripheral artery disease (PAD)
is to improve physical functioning and to increase engagement in activities important to the PAD patients quality of life.
Among PAD patients, growing evidence demonstrates that the 6-minute walk more directly measures outcomes that are
important and relevant to daily functioning and quality of life than treadmill testing. Treadmill walking is an artificial form
of walking, requiring maintenance of a constant rhythmic gait to keep up with the steady pace of the treadmill. This artificial walking does not necessarily translate into easier walking in hall corridors or improved ability to navigate curbs and
uneven sidewalks in daily life. Meaningful clinically important differences have been defined for the 6-minute walk and
are anchored in patient-reported outcomes that include mobility and quality of life. An excellent coefficient of variation
for repeated 6-minute walk testing has been achieved in a PAD population and is better than that reported by Hiatt et al for
treadmill testing. Thirty years of experience measuring peak treadmill walking time does not alone justify its continued use
as a primary outcome. The evidence suggests that peak treadmill walking time does not optimally measure outcomes that
matter most to PAD patients.

Downloaded from http://circ.ahajournals.org/ by guest on July 30, 2015

The Treadmill Is a Better Functional Test Than the 6-Minute Walk Test in Therapeutic
Trials of Patients With Peripheral Artery Disease
William R. Hiatt, R. Kevin Rogers and Eric P. Brass
Circulation. 2014;130:69-78
doi: 10.1161/CIRCULATIONAHA.113.007003
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright 2014 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539

The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circ.ahajournals.org/content/130/1/69

Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial
Office. Once the online version of the published article for which permission is being requested is located,
click Request Permissions in the middle column of the Web page under Services. Further information about
this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation is online at:
http://circ.ahajournals.org//subscriptions/

Downloaded from http://circ.ahajournals.org/ by guest on July 30, 2015