Original Article

Physiological responses during

downhill walking: A new exercise
modality for subjects with chronic
obstructive pulmonary disease?

Chronic Respiratory Disease

2015, Vol. 12(2) 155164
The Author(s) 2015
Reprints and permission:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/1479972315575717
crd.sagepub.com

Carlos Augusto Camillo1, Chris Burtin1,2, Miek Hornikx1,

Heleen Demeyer1, Kristien De Bent3,
Hans van Remoortel1, Christian R Osadnik1,4,5,
Wim Janssens3 and Thierry Troosters1,3

Abstract
Skeletal muscle quadriceps low-frequency fatigue (LFF) during exercise promotes improvements in
exercise capacity with exercise training. In healthy subjects, eccentric muscle work induced by
downhill walking (DW) generates higher muscular stress, whilst metabolic cost is lower compared to
level walking (LW). We investigated quadriceps LFF and metabolic cost of DW in patients with chronic
obstructive pulmonary disease. Ten participants (67 + 7 years, FEV1 51 + 15% predicted) performed
DW, DW carrying a load (DWL) of 10% body weight via vest and LW, in random order. Quadriceps
potentiated twitch force (TWqpot) was assessed before and after each walk, and muscle damage was
assessed before and 24 hours after each walk via serum creatine kinase (CK) levels. Ventilation (VE) and
oxygen consumption (VO2) were measured via breath-by-breath analysis during each walk. DW and DWL
resulted in a greater decrease in TWqpot (30 + 14 N in DW, p < 0.05; and 22 + 16 N in DWL,
p < 0.05) compared to LW (3 + 21 N, p > 0.05). CK levels only increased 24 hours following DW
and DWL (p < 0.05). DW and DWL showed lower VE and VO2 than LW (p < 0.05). DW is associated
with enhanced quadriceps LFF and lower cardiorespiratory costs than LW. The addition of a chest load to
DW does not seem to enhance these effects.
Keywords
Pulmonary disease, chronic obstructive, downhill walking, exercise, fatigue, cardiorespiratory costs

Introduction
Skeletal muscle weakness and exercise capacity impairment are common and important features of chronic
obstructive pulmonary disease (COPD).1 Skeletal muscle dysfunction is associated with poor exercise
capacity and is a predictor of mortality. High-intensity
exercise training is recommended to improve cardiovascular and skeletal muscle function,1,2 however,
treatment responses are variable.3
The development of low-frequency fatigue (LFF)
may be an important determinant of optimal response
to exercise. LFF occurs when the muscle force response
to low-frequency stimulation decreases in association
with a slow (hours or days) recovery.4 It is characterized

KU Leuven, Department of Rehabilitation Sciences, Leuven,

Belgium
2
Hasselt University, Rehabilitation Research Centre, Biomedical
Research Institute, Faculty of Medicine and Life Sciences,
Diepenbeek, Belgium
3
University Hospital Leuven, Respiratory Division and
Rehabilitation, Leuven, Belgium
4
Monash University, Department of Physiotherapy, Victoria,
Australia
5
Institute for Breathing and Sleep, Victoria, Australia
Corresponding author:
Thierry Troosters, Herestraat 49 bus 706, Onderwijs &
Navorsing I, Labo Pneumologie, B-3000 Leuven, Belgium.
Email: thierry.troosters@med.kuleuven.be

156

by decreased intracellular calcium ion concentration5,6

and muscle damage.7 LFF can be effectively stimulated
via eccentric muscle training due to its capacity to
induce skeletal muscle damage.8,9 The benefits of inducing LFF in people with COPD have been recently
demonstrated in a study of exercise training by Burtin
and colleagues.10 In this study, the development of
quadriceps LFF following a single exercise session (in
approximately 30% of participants) was associated with
superior improvements in functional exercise capacity
and symptoms related to quality of life at the end of the
12-week high-intensity rehabilitation programme.10,11
This highlights the potential importance of identifying
treatment modalities capable of eliciting LFF.
Downhill walking (DW) is an exercise modality
with a relatively high eccentric component embedded
in a usual activity (walking).12 Eccentric muscle work
occurs due to the braking action of quadriceps
femoris required to slow the individual down during
movement. Work may be further increased via the
addition of a load to the chest (i.e. DW carrying load
(DWL)).12 DW is of particular interest for individuals
with COPD as it offers a unique opportunity to efficiently induce skeletal muscle stress whilst simultaneously minimizing ventilatory demand during
exercise.13,14
The primary aim of this study was therefore to
determine the extent to which DW and DWL induce
quadriceps LFF compared to level walking (LW) in
individuals with stable COPD. The secondary aims
were to investigate the cardiorespiratory costs, dyspnoea and fatigue response and development of
delayed onset muscle soreness following DW, DWL
and LW.

Methods
Study population
Twelve individuals with COPD15 were recruited from
the outpatient clinic of University Hospital Gasthuisberg, Leuven, Belgium. All patients were familiar
with treadmill walking and were free of exacerbations
in the previous month before study commencement.
Individuals were excluded if they had underlying disease or musculoskeletal limitations that would impair
test procedures, if they required supplemental oxygen
during exercise or could not walk at least 15 minutes
continuously. Those with a body mass index (BMI)
greater than 30 kg/m2 were also excluded due to the
potential for musculoskeletal injury relating to the
extra load requirements of the DWL intervention.

Chronic Respiratory Disease 12(2)

Figure 1. Equipment set-up for DW. Above: Modified

treadmill set-up for DW (portable metabolic monitor in situ);
below: custom-made bracket to achieve 10% negative incline
(approximate cost 200). DW: downhill walking.

Procedure
This study was approved by the ethics committee of
UZ Gasthuisberg, Leuven, Belgium. Written informed
consent was obtained from all patients. Participants
attended clinic on four visits, each separated by 1 week.
The first session comprised familiarization with DW
and baseline measurement of spirometry, functional
exercise tolerance (6-minute walking test (6MWT)),16
cycle maximal exercise capacity,17 handgrip force,18
isometric quadriceps femoris strength,19 usual dyspnoea (modified Medical Research Council dyspnoea
scale)20 and health status (COPD assessment test).21
Thereafter, patients performed three walking protocols
on a treadmill (DW, LW and DWL) on the same day
and time of each week in random sequence.

Interventions
DW was performed using a modified treadmill set-up
(Figure 1) at a negative inclination of 10%. This was
based on previous research in healthy subjects demonstrating a nadir cardiopulmonary demand and heart rate
response at 10% compared to other negative inclinations (range 0 to 18%) at the slower of the two

156

by decreased intracellular calcium ion concentration5,6

and muscle damage.7 LFF can be effectively stimulated
via eccentric muscle training due to its capacity to
induce skeletal muscle damage.8,9 The benefits of inducing LFF in people with COPD have been recently
demonstrated in a study of exercise training by Burtin
and colleagues.10 In this study, the development of
quadriceps LFF following a single exercise session (in
approximately 30% of participants) was associated with
superior improvements in functional exercise capacity
and symptoms related to quality of life at the end of the
12-week high-intensity rehabilitation programme.10,11
This highlights the potential importance of identifying
treatment modalities capable of eliciting LFF.
Downhill walking (DW) is an exercise modality
with a relatively high eccentric component embedded
in a usual activity (walking).12 Eccentric muscle work
occurs due to the braking action of quadriceps
femoris required to slow the individual down during
movement. Work may be further increased via the
addition of a load to the chest (i.e. DW carrying load
(DWL)).12 DW is of particular interest for individuals
with COPD as it offers a unique opportunity to efficiently induce skeletal muscle stress whilst simultaneously minimizing ventilatory demand during
exercise.13,14
The primary aim of this study was therefore to
determine the extent to which DW and DWL induce
quadriceps LFF compared to level walking (LW) in
individuals with stable COPD. The secondary aims
were to investigate the cardiorespiratory costs, dyspnoea and fatigue response and development of
delayed onset muscle soreness following DW, DWL
and LW.

Methods
Study population
Twelve individuals with COPD15 were recruited from
the outpatient clinic of University Hospital Gasthuisberg, Leuven, Belgium. All patients were familiar
with treadmill walking and were free of exacerbations
in the previous month before study commencement.
Individuals were excluded if they had underlying disease or musculoskeletal limitations that would impair
test procedures, if they required supplemental oxygen
during exercise or could not walk at least 15 minutes
continuously. Those with a body mass index (BMI)
greater than 30 kg/m2 were also excluded due to the
potential for musculoskeletal injury relating to the
extra load requirements of the DWL intervention.

Chronic Respiratory Disease 12(2)

Figure 1. Equipment set-up for DW. Above: Modified

treadmill set-up for DW (portable metabolic monitor in situ);
below: custom-made bracket to achieve 10% negative incline
(approximate cost 200). DW: downhill walking.

Procedure
This study was approved by the ethics committee of
UZ Gasthuisberg, Leuven, Belgium. Written informed
consent was obtained from all patients. Participants
attended clinic on four visits, each separated by 1 week.
The first session comprised familiarization with DW
and baseline measurement of spirometry, functional
exercise tolerance (6-minute walking test (6MWT)),16
cycle maximal exercise capacity,17 handgrip force,18
isometric quadriceps femoris strength,19 usual dyspnoea (modified Medical Research Council dyspnoea
scale)20 and health status (COPD assessment test).21
Thereafter, patients performed three walking protocols
on a treadmill (DW, LW and DWL) on the same day
and time of each week in random sequence.

Interventions
DW was performed using a modified treadmill set-up
(Figure 1) at a negative inclination of 10%. This was
based on previous research in healthy subjects demonstrating a nadir cardiopulmonary demand and heart rate
response at 10% compared to other negative inclinations (range 0 to 18%) at the slower of the two

Camillo et al.

walking speeds investigated (5.4 km/hour).14 DWL

was performed the same as DW with the addition
of a vest, loaded with equal weight distribution around
the chest equivalent to 10% of body weight.22 LW
consisted of conventional walking on a treadmill
without any inclination. Participants were encouraged to walk as long as possible up to 20 minutes
(minimum 15 minutes) for each modality. Velocity
was kept constant for the three modalities (75% of
the average walking velocity achieved during the initial 6MWT).23 Participants were allowed to hold
onto side hand bars during walking, if necessary.

Data collection
Markers of fatigue. The protocol used to verify the
presence or absence of quadriceps femoris fatigue has
been previously published by our research group.11
Briefly, subjects sat in a recumbent chair with hips
extended at 120 , knees flexed at 90 and arms
crossed in front of the chest. Unpotentiated quadriceps twitch contraction (TWqunpot), maximal voluntary contraction (MVC) and potentiated quadriceps
twitch contraction (TWqpot) were measured in the
order described before, 15 and 40 minutes after each
walk, and contractile force (in newtons) was recorded.
At least five repetitions were made, and the average of
the two best attempts (within 5% difference) was used
for analysis. The femoral nerve was stimulated through
a 45 mm figure-of-eight coil powered by a double
Magstim stimulator (Magstim Co Ltd, Whitland, Dyed,
Wales, UK). A regularly calibrated strain gauge force
transducer (DS Europe, model 546QD, Milan, Italy)
was used to register force. Signals were amplified
(model 811A amplifiers; Hewlett-Packard, Palo Alto,
California, USA) and stored on a computer. Consistent
with previous research, a TWqpot decrease greater than
15% at 150 was defined a priori as significant LFF. This
threshold is equivalent to the upper limit of variability of
two consecutive measurements of the TWqpot.24
Markers of cardiorespiratory cost. Breath-by-breath
measurement of gas exchange, oxyhaemoglobin
saturation, ventilation (VE) and heart rate was performed during each walk via a portable device (Oxycon mobile, Carefusion, San Diego, California, USA)
and averaged over 1 minute intervals. Blood levels of
lactate were measured before and 2 minutes after each
walk (colorimetric/enzymatic method, lactate oxidase/
perioxidase, Hitachi (Japan)/Roche (Switzerland)
COBAS c702).

157

Other outcomes. Serum levels of creatine kinase (CK)

were collected before and 24 hours after the walks as
markers of muscle damage25,26 (colorimetric/International Federation of Clinical Chemistry method,
COBAS c702). Dyspnoea and leg discomfort were
assessed with a modified Borg dyspnoea scale27 every
2 minutes. Over the 7 days after test completion (at
the same hour as per testing), participants documented
perceived muscle soreness via a 15 cm visual analogue scale (VAS; higher scores worse).28

Analysis
Statistical analyses were performed with Statistical
Analysis Software (SAS 9.3, SAS Institute Inc, Cary,
North Carolina, USA). Data are described as mean
(SD) or median (interquartile range) according to data
distribution. Data normality was assessed by the ShapiroWilk test. One-way repeated measures analysis
of variance was used to compare change over time
within walks for TWqpot, TWqunpot and MVC. The
magnitude of change in TWqpot force (N) from baseline to 15 minutes was compared between interventions via separate paired t-tests and reported as
measures of effect size (Cohens d). Effect size
classes were defined according to calculated d values
(small (d  0.2), moderate (d 0.5) or large (d 
0.8)).29 The incidence of quadriceps LFF was compared across the three interventions via w2 test and
between individual interventions via Fishers exact
tests. Lactate and CK level changes were analysed via
paired t-tests. Statistical significance was defined as
p < 0.05 for all analyses.

Results
Two participants were excluded from the study after
recruitment. One withdrew consent, whilst the other
was unable to walk the minimum required distance
(15 minutes) during LW. Ten participants completed
the study protocol and were included in the analyses.
Demographic characteristics are detailed in Table 1.
Participants were mostly males (70%), with normal
BMI (23 + 4kg/m2) and decreased exercise capacity
(88 + 10% predicted 6MWT; 64 + 20% predicted
maximal workload during cardiopulmonary exercise
capacity). Mean walking speed across all interventions was 4.2 + 0.7 km/hour. Eight participants
achieved the maximum target of 20 minutes treadmill
walking (one walked 160 1000 in DWL whilst the other
160 in LW).

158

Chronic Respiratory Disease 12(2)

Table 1. Participant characteristics (n 10).a

Age (years)
Gender (male/female)
BMI (kg/m2)
CAT questionnaire
mMRC scale
CK (U/L)
Pulmonary function
FEV1 (L)
FEV1 (% of predicted)
FVC (L)
FEV1/FVC
GOLD criteria (1/2/3/4)
Muscle function
Quadriceps force (Nm)
Handgrip Force (N)
Exercise capacity
6 MWT (metre)
Wmax (Watts)
VO2max (L/min)
HRmax (beats/min)
VEmax (L/min)

67 + 7
7/3
23 + 4
14[11  21]
2[1  2]
109 + 55
1.4 + 0.4
51 + 15
3.4 + 0.6
0.50 + 0.16
1/3/5/1
134 + 39
329 + 103
558 + 93
86 + 31
1.35 + 0.40
127 + 15
49 + 12

BMI: body mass index; CAT: chronic obstructive pulmonary disease assessment test; mMRC: modified Medical Research Council
dyspnoea scale; FEV1: forced expiratory volume in 1 second; FVC:
forced vital capacity; 6MWT: 6-minute walk test; Wmax: maximal
workload; VO2max: maximal oxygen uptake; HRmax: maximal heart
rate; VEmax: maximal ventilation; GOLD: Global Initiative for
Chronic Obstructive Lung Disease.
a
Data are mean + standard deviation or median (interquartile
range).

Contractile muscle fatigue

Both the DW and DWL interventions were associated
with statistically significant decreases in TWqpot,
TWqunpot and MVC over time. These changes were
not observed following LW (Table 2). A decrease in
TWqpot greater than 15% (a priori definition of LFF)
was observed at 15 and 40 minutes in 9/10 participants
following DW, 6/10 participants following DWL and
only 3/10 participants following LW (w2 7.5, p
0.02). This incidence of LFF was significantly greater
for both DW and DWL compared to LW but did not
differ between DW and DWL. Although the cut point
of 15% is somewhat arbitrarily chosen and could
be criticized, the present findings do not depend on
this threshold. When approached as a continuous variable, contractile muscle fatigue was observed more
evidently in DW. The mean decrease in TWqpot was
22 + 6% at 15 min and 24 + 7% at 40 min for
DW (p < 0.0001), 16 + 12% at 15 min and
16 + 10% at 40 min for DWL (p < 0.0001) and

5 + 9% at 15 min and 7 + 10% at 40 min for

LW (p 0.62) (Figure 2). In comparison with LW,
these reductions from baseline to 15 minutes in
TWqpot force were significantly greater for both
DW (p 0.0008; Cohens d 1.9) and DWL
(p 0.008; Cohens d 1.01) but not significantly
different between DW and DWL (p 0.18; Cohens
d 0.5). Similar patterns of change across the three
interventions were observed for TWqunpot and MVC
(Table 2). Serum CK levels were significantly greater
following DW (49 + 53%) and DWL (50 +
60%) but not LW (14 + 24%) after 24 hours.
The inducement of LFF did not appear related to
the onset of muscle soreness in the week following the
test protocols. VAS scores were generally low (median 1/15, range 06 for all interventions) without considerable variability within or between individuals.
No clinically or statistically significant differences
in VAS scores were observed between the three modalities on any day after the interventions (p > 0.05).
There was a small, statistically significant decrease
(improvement) in VAS scores from day 1 to day 7
(only) following DWL.

Cardiorespiratory responses during the walking

modalities
A detailed overview of the cardiorespiratory
responses to the three walking modalities is provided
in Table 3. No differences in resting oxygen consumption (VO2), VE and heart rate were observed
between DW, DWL and LW. Steady-state conditions,
defined as a lack of variation greater than 10% during
the final 5 minutes of measurement, were reached in
all patients in all modalities. DW and DWL were
associated with significantly lower VO2 and VE during steady state than LW. During LW, participants
reached 74 + 14% of their individual VO2 peak and
74 + 22% of VE peak. These values were significantly lower during DW (64 + 16% of VO2 peak;
65 + 20% of VE peak) and DWL (68 + 15% of
VO2 peak; 68 + 16% of VE peak; Figure 3). No significant differences in heart rate were observed during
steady state between the three modalities. Lactate
levels were generally low and did not change from
resting values for any modality. There were no significant differences in perceived dyspnoea and fatigue
between the three modalities in the final minute of the
walks (p > 0.05); however, small, statistically significant increases were observed from baseline to test
completion for all modalities (Table 3).

Camillo et al.

159

Table 2. Change in quadriceps muscle force and CK.a

Parameter

Time

TWqpot (N)

Pre
15 minutes
40 minutes
Pre
15 minutes
40 minutes
Pre
15 minutes
40 minutes
Pre
24 hours

TWqunpot (N)

MVC (N)

CK (U/L)

LW

DW

DWL

132.8 + 32.7
129.4 + 27.4
127.0 + 27.9
83.1 + 13.3
78.7 + 18.4
79.2 + 18.3
375.6 + 107.6
363.9 + 107.0
366.5 + 99.5
119 + 65
130 + 54

145.2 + 33.6
115.1 + 27.1b
111.4 + 25.5c
85.7 + 20.2
75.4 + 21.8
73.6 + 17.8c
375.1 + 116.3
339.9 + 105.2
327.6 + 99.5c
118 + 65
172 + 96d

146.5 + 36.1
124.4 + 36.5b
121.7 + 31.2c
93.4 + 24.7
73.1 + 16.4
74.9 + 16.7c
371.8 + 115.5
357.8 + 124.2
340.4 + 105.9c
135 + 112
176 + 107d

TWqpot: potentiated quadriceps twitch contraction; TWqunpot: unpotentiated quadriceps twitch contraction; CK: creatine kinase;
Pre: values collected prior the walks; 15 min: values collected 15 minutes after the end of the walks; 40 min: values collected 40 minutes
after the end of the walks; 24 h: values collected 24 hours after the end of the walks; DW: downhill walking; DWL: downhill walking
carrying load; LW: level walking.
a
Data are mean + standard deviation.
b
Significant between-group difference in magnitude of change from baseline compared to LW (p < 0.05).
c
Significant within-group effect over time (analysis of variance) for corresponding intervention (p < 0.05).
d
Significant within-group change from baseline (p < 0.05).

Figure 2. Change in potentiated quadriceps twitch muscle force. (a) level walking; (b) downhill walking; (c) downhill
walking with load. Data are represented as percentage of initial values. Bars represent mean change. Lines show
individual responses. Dotted line represents the threshold of fatigue (15% of decrease in muscle force from the initial
values). PRE: values before the walk; 15 min: potentiated twitch force 15 minutes after walking; 40 min: potentiated
twitch force 40 minutes after walking. *p < 0.05: versus PRE.

160

Chronic Respiratory Disease 12(2)

Table 3. Responses in physiological measurements during the walks.a

Parameter

Time

LW

DW

VO2 (L/min)

Rest
Steady
Rest
Steady
Rest
Steady
Rest
Steady
Rest
Steady
Rest
Steady
Pre
Post
Pre
Post
Pre
Post

0.35 + 0.09
1.02 + 0.22b
0.28 + 0.09
0.89 + 0.27b
14.0 + 3.8
35.2 + 8.12b
26 + 7
67 + 23b
86 + 15
107 + 16b
98 + 1
94 + 4b
0 [0  1.5]
2 [0  3.5]b
0 [0  1.5]
1 [0  3.5]b
1.7 + 1
1.6 + 1

0.3 + 0.05
0.85 + 0.22b,c
0.27 + 0.06
0.70 + 0.21b,c
13.3 + 2.4
29.6 + 6.5b,c
26 + 8
57 + 19b,c
84 + 10
105 + 11b
99 + 1
95 + 4b
0 [0  0.5]
1 [0.4  4]b
0 [0  0.9]
2 [0.4  3]b
1.6 + 0.7
1.4 + 0.6

VCO2 (L/min)
VE (L/min)
VE/MVV (%)
HR (beats/min)
SpO2 (%)
Borg dyspnoea
Borg fatigue
Lactate (mmol/L)

DWL
0.3 + 0.1
0,9 + 0.2b,c
0.28 + 0.07
0.75 + 0.17b,c
13.7 + 2.8
31.5 + 4.7b,c
26 + 8
60 + 17b, c
85 + 9
107 + 14b
98 + 1
94 + 5b
0.2 [0  2]
3 [0.5  6]b
0.2 [0  2]
2.5 [0.9  4]b
1.7 + 0.7
1.3 + 0.8

VO2: oxygen uptake; VE: ventilation; MVV: maximal voluntary ventilation; Borg: perceived effort; SpO2: oxyhaemoglobin saturation;
Rest: immediately before the walks; Steady: values achieved during steady state walking; Pre: values collected prior the walks; Post:
values collected immediately after the end of the walks; LW: level walking; DW: downhill walking; DWL: downhill walking carrying load;
VCO2: carbon dioxide elimination; HR: heart rate.
a
Data are mean + standard deviation or median [interquartile range].
b
Significant within-group difference, steady state versus rest (p < 0.05).
c
Significant between-group difference at steady state compared to LW (p < 0.05).

Discussion
Contractile muscle fatigue
This study demonstrates that DW induces greater quadriceps LFF than LW in individuals with COPD, as measured by both twitch contraction and a serum biomarker.
The addition of an extra load around the chest does not
appear to affect contractile muscle fatigue or markers of
muscle damage. Conventional (flat) walking was associated with virtually no LFF. This is consistent with previous findings in this patient group.30,31 Importantly,
walking downhill proved to be both a reliable and an
efficient means of inducing quadriceps fatigue, with 9
of 10 participants having detectable quadriceps LFF
following a single 20-minute exercise session.
The observed difference in LFF between DW and
LW may have been affected by gait pattern alterations
in response to the negative incline. This study did not,
however, seek to investigate biomechanical adaptations
in people with COPD. DW is known to elicit greater
eccentric quadriceps muscle contraction than walking
on a flat surface in healthy individuals.12,22,32 The repetitive eccentric contractions inherent in DW therefore
predisposes to more muscle damage than concentric

contractions.8,9 This was indirectly confirmed in our

study via elevated serum CK levels.25,33 Our finding
that LFF and elevated CK were observed during both
DW conditions but not LW supports the hypothesis
that a higher eccentric: concentric muscle contraction
ratio occurs during DW in patients with COPD.
DWL had less of an effect on LFF than DW
observed by a lack of statistical difference and moderate effect size when comparing the decrease in 150
Twqpot between the two modalities. This result was
not expected. Running on a decline carrying a load has
been demonstrated to increase the eccentric: concentric
muscle contraction ratio in young healthy subjects;12
however, one cannot assume this applies equally to
an elderly COPD population. For example, the selfselected walking speed, cardiopulmonary responses
and muscle composition of young healthy subjects
would differ considerably to older adults living with
chronic health conditions. It is plausible that the effect
observed following DW could be a potential ceiling
upon which further improvement may not be likely
to occur. It is also possible that the addition of a chest
load resulted in alterations to gait pattern (e.g. cadence

Camillo et al.

161

Figure 3. Oxygen uptake and V0 E in the three modalities. (a) Oxygen consumption during walking; (b) Ventilation during
walking. Data are mean + standard deviation. Dotted line represents the maximal value achieved during maximal cycling
exercise test. VO2 oxygen uptake; VE minute ventilation; LW level walking; DW downhill walking; DWL
downhill walking carrying load. # p < 0.05 DW versus LW at iso-time; y p < 0.05 DWL versus LW at iso-time.

and stride length) and/or muscle activation to reduce

the quadriceps braking demand. Neither of these factors were monitored in this study. It is, however, useful
for clinicians to know that LFF can be elicited during
walking without the necessity of adding extra weight
during a walking activity in individuals with COPD.
In order to verify the clinical feasibility of using
DW as an exercise training modality, we investigated
skeletal muscle soreness in the week after the session.
Despite knowledge that high-intensity eccentric exercise may enhance delayed onset of muscle soreness,25,26 we found no significant differences in the
perception of muscle soreness and very low ratings
of discomfort overall.

Cardiorespiratory cost of DW
Both DW and DWL elicited lower cardiorespiratory
demands (VO2, VE and VE/maximum voluntary

ventilation (MVV) than LW in this patient group.

This finding is consistent with previous data in young
and old healthy adults.13,34,35 Our data confirm, in a
small sample, that this phenomenon applies equally
to individuals who are chronically ill and ventilatory
limited. The mild reduction in cardiorespiratory cost
of activity associated with DW and DWL is appealing
for individuals with COPD as many are symptom (i.e.
breathlessness) limited on exertion.2,31 The clinical
significance of the difference in absolute reduction
of VE between the DW and LW interventions
(approximately 5 L/min) is difficult to ascertain. The
average difference of VE/MVV between the DW and
LW interventions was, however, 10 + 9%. One could
speculate, for example, that individuals with more
severe airways disease might derive greater benefit
from DW than others.
Our data also compare favourably with previous
research of the metabolic cost of exercise training

162

modalities in COPD by Probst and colleagues.23 In this

study, the mean VO2 levels achieved during treadmill
walking (set at 75% of mean 6MWT speed) were
0.98 + 0.32 L/min. Interestingly, mean VO2 data from
the DW and DWL interventions in our study compare
similarly to the lower energy demands observed during
cycling (0.92 + 0.25 L/min) by Probst and colleagues.
It is important to note that exercise intensities above
40% of VO2 reserve are recommended to develop or
maintain physical fitness.36 Our data showed that
DW and DWL achieved 50% and 58% of VO2 reserve
(60 and 64% of VO2 peak), respectively. Thus, despite
eliciting a lower VO2 than LW, it would be appropriate
to consider DW a suitable modality to achieve exercise
of a moderate intensity.
One could speculate the metabolic work could have
been larger in the downhill modalities as they were
associated with higher CK and greater contractile muscle fatigue. This, however, would be difficult to couple
with the lower cardiopulmonary stress and the fact that
the mechanical work of DW compared to LW is clearly
less. It is possible that higher CK levels and incidence
of contractile muscle fatigue were due to the eccentric
nature of muscle contraction. Finally, the lack of differences observed in the heart rate between the DW and
LW corroborates with previous investigation.37 Some
mechanisms such as the decreased mechanical work
required during eccentric contractions for a given oxygen uptake38 and/or increased sympathetic modulation
and fibre activation during eccentric contraction39 may
explain this uncoupling.

Safety considerations
To the best of the authors knowledge, this is the first
study to investigate DW in individuals with COPD.
Little is known about the potential side effects of DW
apart from investigations of its effect on muscle
damage and delayed onset of muscle soreness.25,26,35,40
We had no reports of serious adverse events but tolerable discomfort was reported from participants in the
hip (n 1) and knee (n 2) during DWL. One participant was excluded due to dyspnoea and oxygen desaturation after 5 minutes of the LW intervention. Another
two participants required rests due to dyspnoea (one
subject walked 160 1000 of DWL; the other 160 of LW),
but both were able to complete 20 minutes of DW.

Limitations
Some methodological limitations may affect the
implications arising from this study. The modest

Chronic Respiratory Disease 12(2)

sample size should be considered when applying the

findings to the broader COPD population, whilst the
lack of a healthy, age-matched control group mean
results cannot be extrapolated beyond a COPD context.
However, as a large premise for investigating DW in
COPD is the preservation of ventilatory function on
exertion, this may be less important for a normal population. Interestingly, our observed changes in metabolic data (e.g. VO2 and VE) were fundamentally
similar to those of Gault and colleagues who investigated healthy subjects of similar age to our cohort.35
We considered the increase in serum CK levels
24 hours after DW and DWL reflective of induced
muscle damage (and LFF). This response following
eccentric muscle contractions is well supported in the
medical literature.8,9,25,33 In the absence of a truly
confirmatory muscle biopsy analysis, however, this
biomarker may be more accurately considered a surrogate biomarker of LFF.
The walking protocols were implemented at one
relative set speed, despite the potential for variability
in cardiorespiratory and fatigue responses according
to treadmill inclination, speed and intensity.13,37 The
responses to the protocols at different speeds would
certainly be of some clinical value but was beyond the
scope of this study. Importantly, our study design was
able to verify LFF and cardiorespiratory demands in
all three modalities. It is also possible that, despite a
fixed speed across walking conditions, training load
may have differed between test conditions. The
recruitment and activation of different muscle groups
may therefore have contributed to changes observed
in the downhill conditions compared to LW. Whilst
we did not measure muscle activity during the protocol, the combined increase in serum CK levels and
reductions in quadriceps force (Twqpot) and cardiopulmonary responses suggest a true quadriceps fatiguing effect of DW.
Finally, it is possible that recurrent eccentric braking contractions, inherent in DW, may become a fatiguing stimulus that negatively impacts on skeletal
muscle in COPD. From a LFF perspective, recovery
can take hours to days,4 meaning that inadequate
recovery may be possible when performing repeated
exercise training sessions over short time periods.
This is not, however, supported by recent data from
Burtin and colleagues10 who demonstrated that individuals with COPD who developed fatigue during
exercise training and completed three sessions per
week improved more upon completion of the training
programme than those who did not fatigue.

Camillo et al.

Future perspectives and clinical application

In individuals with COPD, the development of LFF
after a single exercise session relates to significant
improvements in exercise capacity and health-related
quality of life following a physical exercise training
programme.10 This study demonstrates that DW elicits LFF more than conventional LW. These results
therefore provide a clear rationale for further exploration into the potential effectiveness of rehabilitation incorporating a DW exercise modality. The
intervention is simple to adapt into existing rehabilitation models and may prove important for individuals who do not usually fatigue during training. The
effect of inducing skeletal muscle fatigue on exercise compliance must also be considered.

Conclusion
Individuals with COPD develop more quadriceps LFF
during DW compared to LW and have lower cardiorespiratory costs. It is important for future investigations to evaluate the effectiveness of DW as part of
a comprehensive rehabilitation programme on broader,
important health outcomes.
Acknowledgements
The authors would like to thank physiotherapists Ilse
Muylaert, Iris Coosemans, Veronica Barbier and the whole
staff of the Respiratory Rehabilitation Department and the
Pulmonary Function Department at the University Hospitals Gasthuisberg, Leuven, Belgium, for helping with the
logistics and execution of the procedures in this study.
We also would like to thank Patricia Besem, Rob Verlinden
and Carthy L Aguillon for helping with the data collection.
Finally, we thank Maarten Spruyt and Willem Dewit for the
help with blood samples.

Funding
This work was supported by the Flemish Research Foundation (G.0871.13) and PROactive IMI-JU (115011). CAC is
funded by The National Council for Scientific and Technological Development (CNPq), Brazil (202425/2011-8). CO
is the recipient of a European Respiratory Society Fellowship (LTRF 2014 3132).

References
1. Spruit MA, Singh SJ, Garvey C, et al. An official
American Thoracic Society/European Respiratory
Society statement: key concepts and advances in pulmonary rehabilitation. Am J Respir Crit Care Med
2013; 188: e13e64.

163
2. Troosters T, Casaburi R, Gosselink R, et al. Pulmonary
rehabilitation in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2005; 172: 1938.
3. Troosters T, Gosselink R and Decramer M. Exercise
training in COPD: how to distinguish responders from
nonresponders. J Cardiopulm Rehabil 2001; 21:
1017.
4. Jones DA. High-and low-frequency fatigue revisited.
Acta Physiol Scand 1996; 156: 265270.
5. Allen DG, Lee JA and Westerblad H. Intracellular calcium and tension during fatigue in isolated single muscle fibres from Xenopus laevis. J Physiol 1989; 415:
433458.
6. Westerblad H, Lee JA, Lannergren J, et al. Cellular
mechanisms of fatigue in skeletal muscle. Am J Physiol 1991; 261: C195C209.
7. Jones DA. Muscle fatigue due to changes beyond the
neuromuscular junction. Ciba Found Symp 1981; 82:
178196.
8. Smith IC and Newham DJ. Fatigue and functional performance of human biceps muscle following concentric or eccentric contractions. J Appl Physiol
2007; 102: 207213.
9. Newham DJ, Mills KR, Quigley BM, et al. Pain and
fatigue after concentric and eccentric muscle contractions. Clin Sci (Lond) 1983; 64: 5562.
10. Burtin C, Saey D, Saglam M, et al. Effectiveness of
exercise training in patients with COPD: the role of
muscle fatigue. Eur Respir J 2012; 40: 338344.
11. Mador MJ, Mogri M and Patel A. Contractile fatigue
of the quadriceps muscle predicts improvement in
exercise performance after pulmonary rehabilitation.
J Cardiopulm Rehabil Prev 2014; 34: 5461.
12. Abe D, Fukuoka Y, Muraki S, et al. Effects of load and
gradient on energy cost of running. J Physiol Anthropol 2011; 30: 153160.
13. Wanta DM, Nagle FJ and Webb P. Metabolic response
to graded downhill walking. Med Sci Sports Exerc
1993; 25: 159162.
14. Liefeldt G, Noakes TD and Dennis SC. Oxygen delivery does not limit peak running speed during incremental downhill running to exhaustion. Eur J Appl Physiol
Occup Physiol 1992; 64: 493496.
15. Vestbo J, Hurd SS, Agusti AG, et al. Global strategy
for the diagnosis, management, and prevention of
chronic obstructive pulmonary disease: GOLD executive summary. Am J Respir Crit Care Med 2013; 187:
347365.
16. Brooks D, Solway S and Gibbons WJ. ATS statement
on six-minute walk test. Am J Respir Crit Care Med
2003; 167: 1287.

164
17. ATS/ACCP Statement on cardiopulmonary exercise
testing. Am J Respir Crit Care Med 2003; 167:
211277.
18. Mathiowetz V, Kashman N, Volland G, et al. Grip and
pinch strength: normative data for adults. Arch Phys
Med Rehabil 1985; 66: 6974.
19. Decramer M, Lacquet LM, Fagard R, et al. Corticosteroids contribute to muscle weakness in chronic airflow
obstruction. Am J Respir Crit Care Med 1994; 150:
1116.
20. Bestall JC, Paul EA, Garrod R, et al. Usefulness of the
medical research council (MRC) dyspnoea scale as a
measure of disability in patients with chronic obstructive pulmonary disease. Thorax 1999; 54: 581586.
21. Jones PW, Harding G, Berry P, et al. Development and
first validation of the COPD assessment test. Eur
Respir J 2009; 34: 648654.
22. Bourdin M, Belli A, Arsac LM, et al. Effect of vertical loading on energy cost and kinematics of running in trained male subjects. J Appl Physiol 1995;
79: 20782085.
23. Probst VS, Troosters T, Pitta F, et al. Cardiopulmonary
stress during exercise training in patients with COPD.
Eur Respir J 2006; 27: 11101118.
24. Saey D, Debigare R, Leblanc P, et al. Contractile leg
fatigue after cycle exercise: a factor limiting exercise
in patients with chronic obstructive pulmonary disease.
Am J Respir Crit Care Med 2003; 168: 425430.
25. Clarkson PM and Hubal MJ. Exercise-induced muscle
damage in humans. Am J Phys Med Rehabil 2002; 81:
S52S69.
26. Byrnes WC, Clarkson PM, White JS, et al. Delayed
onset muscle soreness following repeated bouts of
downhill running. J Appl Physiol 1985; 59: 710715.
27. Borg GA. Psychophysical bases of perceived exertion.
Med Sci Sports Exerc 1982; 14: 377381.
28. Price DD, McGrath PA, Rafii A, et al. The validation
of visual analogue scales as ratio scale measures for
chronic and experimental pain. Pain 1983; 17: 4556.
29. Cohen J. Statistical Power Analysis for the Behavioral
Sciences. Hilsdale: Lauwrence Erlbaum Associates
Publisher, 1988.

Chronic Respiratory Disease 12(2)

30. Man WD, Soliman MG, Gearing J, et al. Symptoms
and quadriceps fatigability after walking and cycling
in chronic obstructive pulmonary disease. Am J Respir
Crit Care Med 2003; 168: 562567.
31. Pepin V, Saey D, Laviolette L, et al. Exercise capacity
in chronic obstructive pulmonary disease: mechanisms
of limitation. COPD 2007; 4: 195204.
32. Abe D, Yanagawa K and Niihata S. Effects of load carriage, load position, and walking speed on energy cost
of walking. Appl Ergon 2004; 35: 329335.
33. Brancaccio P, Maffulli N and Limongelli FM. Creatine
kinase monitoring in sport medicine. Br Med Bull
2007; 8182: 209230.
34. Pivarnik JM and Sherman NW. Responses of aerobically fit men and women to uphill/downhill walking
and slow jogging. Med Sci Sports Exerc 1990; 22:
127130.
35. Gault ML, Clements RE and Willems ME. Cardiovascular responses during downhill treadmill walking at
self-selected intensity in older adults. J Aging Phys Act
2013; 21: 335347.
36. Garber CE, Blissmer B, Deschenes MR, et al. American College of Sports Medicine position stand. Quantity and quality of exercise for developing and
maintaining cardiorespiratory, musculoskeletal, and
neuromotor fitness in apparently healthy adults: guidance for prescribing exercise. Med Sci Sports Exerc
2011; 43: 13341359.
37. Navalta JW, Sedlock DA and Park K-S. Physiological responses to downhill walking in older and
younger individuals. J Appl Physiol (Online) 2004;
7: 4551.
38. Isner-Horobeti ME, Dufour SP, Vautravers P, et al.
Eccentric exercise training: modalities, applications
and perspectives. Sports Med 2013; 43: 483512.
39. Bigland-Ritchie B and Woods JJ. Integrated electromyogram and oxygen uptake during positive and negative work. J Physiol 1976; 260: 267277.
40. Schwane JA, Williams JS and Sloan JH. Effects of
training on delayed muscle soreness and serum creatine kinase activity after running. Med Sci Sports Exerc
1987; 19: 584590.