BMJ Open · For peer review only 2 Abstract Incremental shuttle-walking and treadmill-walking tests...
Transcript of BMJ Open · For peer review only 2 Abstract Incremental shuttle-walking and treadmill-walking tests...
For peer review only
Pilot investigation of the oxygen demands and metabolic cost of incremental shuttle-walking and treadmill-walking
in patients with cardiovascular disease.
Journal: BMJ Open
Manuscript ID: bmjopen-2014-005216
Article Type: Research
Date Submitted by the Author: 10-Mar-2014
Complete List of Authors: Almodhy, Meshal; Universiity of Essex, Biological Sciences Beneke, Ralph; University of Marburg, Institute of Sports Medicine Cardosos, Fernando; Universiity of Essex, Biological Sciences
Sandercock, Gavin; University of Essex, Biological Sciences Taylor, Matthew; Universiity of Essex, Biological Sciences
<b>Primary Subject Heading</b>:
Cardiovascular medicine
Secondary Subject Heading: Sports and exercise medicine
Keywords: REHABILITATION MEDICINE, Myocardial infarction < CARDIOLOGY, Ischaemic heart disease < CARDIOLOGY
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open on D
ecember 7, 2020 by guest. P
rotected by copyright.http://bm
jopen.bmj.com
/B
MJ O
pen: first published as 10.1136/bmjopen-2014-005216 on 16 S
eptember 2014. D
ownloaded from
For peer review only
1
Pilot investigation of the oxygen demands and metabolic cost of incremental shuttle-
walking and treadmill-walking in patients with cardiovascular disease.
Sandercock, GRH1
Almodhy, M1., Beneke, R
2., Cardoso, F
1., Taylor, MJ.
1
1Centre for Sports & Exercise Sciences, School of Biological Sciences, University of Essex
Colchester, UK CO4 3SQ
2Institut für Sportwissenschaft und Motologie Medezin, Philipps-Universität Marburg,
Germany.
Oxygen demands of shuttle-walking and treadmill walking
Key words: Cardiovascular disease; metabolism; cardiorespiratory fitness; pilot study
Word Count: 2288
Page 1 of 17
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
2
Abstract
Incremental shuttle-walking and treadmill-walking tests are both used to assess patients with
cardiovascular disease, but it remains unknown whether their energetic costs are comparable..
Patients completed the incremental shuttle walk test (on the ground) and the treadmill-shuttle test
in a randomised order with one week between trials. The 12 stage protocol starts at a walking
speed of 0.5 m.s-1 and increases by 0.17 m
.s-1 each minute. An identical incremental protocol was
programmed into the treadmill. During both tests a portable gas analyser was used to record
expired gas allowing for calculation of metabolic power and energy cost of walking
Average overall energy cost per meter was higher during treadmill-walking (3.22 ± 0.55 J.kg
.m
-1)
than during shuttle-walking (3.00 ± 0.41 J.kg
.m
-1). There were significant post hoc effects at 0.67
m.s-1 and 0.84 m
.s-1, where the energy cost of treadmill-walking was significantly higher than that
of shuttle-walking. This pattern was reversed at higher walking speeds of 1.52 m.s-1 and 1.69 m
.s-
1 where shuttle-walking had a greater energy cost per meter than treadmill-walking
These data demonstrate that the energetic demand of shuttle-walking is fundamentally different
from that of treadmill walking. We suggest that direct comparisons of results from these two
exercise modalities should be made with great caution
Objective
Incremental shuttle-walking and treadmill-walking tests are both used to assess patients with
cardiovascular disease, but it remains unknown whether their energetic costs are comparable..
Design
Patients completed the incremental shuttle walk test (on the ground) and the treadmill-shuttle test
in a randomised order with one week between trials. The 12 stage protocol starts at a walking
Page 2 of 17
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
3
speed of 0.5 m.s-1 and increases by 0.17 m
.s-1 each minute. An identical incremental protocol was
programmed into the treadmill. During both tests a portable gas analyser was used to record
expired gas allowing for calculation of metabolic power and energy cost of walking
Results
Average overall energy cost per meter was higher during treadmill-walking (3.22 ± 0.55 J.kg
.m
-1)
than during shuttle-walking (3.00 ± 0.41 J.kg
.m
-1). There were significant post hoc effects at 0.67
m.s-1 and 0.84 m
.s-1, where the energy cost of treadmill-walking was significantly higher than that
of shuttle-walking. This pattern was reversed at higher walking speeds of 1.52 m.s-1 and 1.69 m
.s-
1 where shuttle-walking had a greater energy cost per meter than treadmill-walking
Conclusion
These data demonstrate that the energetic demand of shuttle-walking is fundamentally different
from that of treadmill walking. We suggest that direct comparisons of results from these two
exercise modalities should be made with great caution.
Page 3 of 17
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
4
INTRODUCTION
Since its conception as an alternative to incremental treadmill testing of chronic obstructive
pulmonary disease (COPD) patients, the incremental shuttle-walking test (ISWT) has gained
popularity as an estimate of functional capacity in numerous clinical populations. The ISWT
appears adequately reliable[1], and is sensitive to changes in functional capacity[2 3]. However,
the ISWT’s validity as an estimate of cardiovascular fitness is only moderate[3] and the use of
the test to estimate oxygen consumption exercise capacity in Metabolic Equivalents (METs) is
questionable[4].
Woolf-May et al.2 found acceptable agreement between the energy cost of treadmill walking and
the ISWT in healthy volunteers using linear regression analyses, but the relationship was not
assessed in cardiac patients. The authors also reported that at walking speeds >1m·s-1, the ISWT
significantly underestimated the actual metabolic demands of shuttle-walking compared with
reference values calculated during treadmill walking in healthy adults. [5] The authors suggested
that cardiac patients may have higher energy demands than predicted, due to poorer walking
economy but made no such comparison of walking economy in cardiac patients between ISWT
and treadmill walking. They also made no calculations of economy during ISWT.
Treadmill and shuttle-walking tests are routinely used to assess cardiovascular disease patients
and we have previously reported discrete values for change in fitness measured using these tests.
[27] Prior to undertaking a proposed multicentre study to identify predictors of change in
cardiorespiratory fitness due to cardiac rehabilitation we performed the present pilot study. Our
aim was to determine whether the metabolic demands and the energy cost of the tests were
comparable in this patient group to the extent that we could combine data from these tests.
Page 4 of 17
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
5
METHODS
Participants (n=8; 7 males; 67±5.2 years: 86.6 ±10.1 kg) were stable cardiac patients attending
community-based rehabilitation. The study was approved by the Faculty ethics board and all
patients gave written, informed consent.
Equipment
The ISWT was performed on a non-slip floor using two cones placed 9 m apart and a portable
CD player. The treadmill test was performed on a motorised treadmill (Quaser, HP Cosmos,
Nussdorf,Germany). During both tests a portable gas analyser (K4b2 Mobile Breath by Breath
Metabolic System, COSMED Pulmonary Function Equipment, Rome, Italy) was used to record
expired gas. This was calibrated using gases of a known concentration and a syringe before each
test.
Protocol
Patients completed the ISWT and the treadmill test in a randomised order with one week
between trials. The ISWT was performed in accordance with national recommendations for
cardiac patients[6]. Briefly, the 12 stage protocol starts at a walking speed of 0.5 m·s-1 (1.12
mph) and increases by 0.17 m·s-1 (0.48 mph) each minute. An identical incremental protocol was
programmed into the treadmill. Patients were accustomed to treadmill walking and following a
brief period of familiarisation this test was also performed by each patient.
Calculation of metabolic power and energy cost of walking
Metabolic power was calculated via indirect calorimetry from and above rest,
caloric equivalent, and body mass by: metabolic power [W.kg
-1] = [ml
.kg
-1.s-1] . respiratory
&VO2 &VCO2
&VO2
Page 5 of 17
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
6
exchange ratio adjusted caloric equivalent [J.ml
-1]. [7 8] To analyse the relationship between
speed and metabolic power of walking the metabolic power was predicted as a quadratic function
of speed: metabolic power = a + b v². [7 9],[10],[11],[12-14]. The energy cost of walking per
metre distance was calculated by: energy cost [J.kg
-1.m
-1] = metabolic power [W
.kg
-1] . speed
-1
[m.s-1]. [7 8 15]
Statistical Analyses
Descriptive results are presented as mean ± SD. A test modality-by-walking speed analysis of
variance (ANOVA) with shuttle vs. treadmill walking as within-subjects factor and walking
speed as the between-subjects factor was performed. Significant interactions and main effects
were further analysed using one-way ANOVA and paired samples t-tests as appropriate. Non-
linear regression models were used to identify significant interrelationships between metabolic
power, energy cost per meter and walking speed, respectively. All analyses were completed
using SPSS v19.0 (SPSS Inc. and IBM Company. Chicago, IL) and statistical significance was
defined as p< 0.05.
RESULTS
Figure 1 shows the metabolic cost at each of seven stages completed by at least 7 patients. There
was a significant main effect for walking speed on oxygen uptake and a significant interaction
between treadmill-walking and shuttle-walking on the ground. Oxygen uptake was higher in
treadmill-walking than shuttle-walking at 0.67 and 0.84 m·s-1 (p < 0.05; n = 8) but the
significantly steeper increases in oxygen demand during shuttle-walking meant the opposite was
true at 1.69 m·s-1 (p < 0.05; n = 7).
Page 6 of 17
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
7
**Insert Fig 1 here**
Figure 2 shows the metabolic power of treadmill-walking and shuttle-walking. There was a main
effect for walking velocity on metabolic power during both treadmill and shuttle walking (p <
0.05). The different effects of walking modality on metabolic power were more pronounced if
power was predicted as a function of walking speed with power-treadmill walking = 2.028 +
1.115 v² and power-shuttle walking on the ground = 1.126 + 1.665 v² where 99 % of the variance
of power were explained by the quadratic curve fits in both modalities (both p < 0.001). The
difference in response to each modality was indicated by a significant interaction between
modality and speed. There were significantly higher metabolic power requirements for treadmill
walking at 0.67 m·s-1 and 0.84 m·s
-1 (p < 0.05; n = 8). Due to the steeper increase observed in
shuttle-walking the metabolic power was significantly higher at 1.52 and 1.69 m⋅s-1 (p < 0.05;
n=7) compared with treadmill-walking.
**Insert Fig 2 here**
There were significant main effects for both modality and velocity in relative energy cost (per
metre) of walking again well described as a function of speed by the above approximated
parameters for both walking modalities (energy cost-treadmill walking = 2.028 / v + 1.115 v and
energy cost shuttle-walking on the ground = 1.126 / v + 1.665 v; both p < 0.001). Average
overall energy cost per meter (kg·m-1) was higher during treadmill-walking (3.22 ± 0.55 J·kg·m
-1)
than during shuttle-walking (3.00 ± 0.41 J·kg·m-1). There were significant post hoc effects at 0.67
m·s-1 and 0.84 m·s
-1 (n=8), where the energy cost of treadmill-walking was higher than that of
Page 7 of 17
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
8
shuttle-walking. Again, this pattern was reversed at higher walking speeds of 1.52 m·s-1 and 1.69
m·s-1 (n=7) where shuttle-walking had a greater energy cost per meter than treadmill-walking.
**Insert Fig 3 here**
DISCUSSION
This is the first comparative investigation of the metabolic demands and energy cost per meter
walking of incremental treadmill-walking and shuttle-walking in cardiac rehabilitation patients.
The primary finding is that identical walking protocols performed on a treadmill or by shuttle-
walking are not comparable in terms of their oxygen requirements or energy cost. These findings
suggest that cardiorespiratory fitness measured using these different exercise testing modalities
should be compared with caution. Additionally, evaluating exercise capacity of patients with
cardiovascular disease using shuttle-walking by using MET values derived from treadmill-
walking may lead to potentially misleading results and poor clinical decision making.
The ISWT is the most used field test of exercise capacity in UK cardiac rehabilitation
centres[16] and is recommended nationally for this purpose[6]. Test results are used in patient
risk stratification[6], in exercise prescription and to assess the efficacy of exercise training during
outpatient rehabilitation. Current recommendations suggest patients be classed as high risk if
their exercise capacity is <5 METs. Failure to reach this criterion standard may lead to prevent
patients from entering community-based exercise training[17] or being readmitted to hospital-
based programmes.
The present data suggest that the ISWT stage deemed to be equivalent to 5 METs (level 7,
walking speed 1.52 m·s-1), actually has a significantly higher oxygen requirement, electing
greater metabolic power and energy cost per metre than walking at the equivalent speed on a
treadmill. These findings agree with previous data in cardiac patients[4] showing that oxygen
Page 8 of 17
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
9
uptake during shuttle-walking was higher than the value estimated using the American College
of Sports Medicine walking equations. We concur with these data suggesting that oxygen
requirements are higher in shuttle-walking than those estimated from treadmill walking from
level 7 (1.52 m·s-1) onwards.
This level and walking speed is of particular importance as it is the value deemed to indicate a
functional capacity of 5 METs a value used in risk stratification and triaging of patients to
hospital based rehabilitation programmes designed for ‘high risk’ patients or to low risk,
community based exercise. The higher potential ‘real’ energy cost may lead to false-positive
assessment of high risk leading to costly treatment option such as re-enrolment in outpatient
rehabilitation whereas there is good evidence that fitter patients can be successfully ‘fast tracked’
to community rehabilitation saving capacity and money to the health providers [18]. The ISWT
is also used to provide feedback to patients on gains in functional capacity by its application pre-
and post-rehabilitation. The steeper increases in energy requirements beyond 5 METs may make
it difficult to show small improvements in functional capacity reported as estimated MET.
In cardiac patients measured before outpatient rehabilitation, exercise capacity tends to be lower
when estimated from ISWT[19] than when patients are assessed using standard treadmill
protocols [20 21]. The patterns of change in walking energy cost per metre on the treadmill
show the expected pattern. Slow speeds are associated with higher cost per metre which
decreases as optimal (comfortable) walking speed approaches. Continuing to increase walking
speed above this pace requires a greater cost per metre. In contrast to this, the energy cost per
metre in shuttle-walking decreases only very little and only following the first (very slow)
walking pace in the initial stage. The energy cost then increases stage-by-stage throughout the
protocol. The cost is only consistent between treadmill and shuttle-walking between 1.2 - 1.4
Page 9 of 17
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
10
m·s-1 (close to comfortable walking speed) and the increase in energy requirements is much
greater in shuttle walking. When using the ISWT, patients’ cardiorespiratory fitness may need to
be estimated from maximal walking speed below (Stage 2, 0.84 m·s-1) or above (Stage 6, 1.51
m·s-1) the speeds at which treadmill and shuttle-walking agree. The differences in oxygen
consumption at these speeds are simply not comparable, and were large enough to produce
statistically significant differences in the very small numbers (n=8, n=9) used in this pilot
investigation. Given that our multicentre study involves comparisons of hundreds of patients we
plan to report cardiorespiratory fitness values separately according to test modality. The classical
description of the energy cost during locomotion is of a U-shaped relationship[22] – as speed
increases or decreases from the optimal (1.11-1.3 m.s-1 [22-24]) the energy cost of locomotion
increases. For the treadmill protocol our data support this relationship. At slow speeds (0.6-0.8
m.s-1) energy cost was greater than at optimal speeds (1.2-1.4 m
.s-1). As walking speed increased
(1.6-1.8 m.s-1) the energy cost again began to increase. This is comparable to Berryman et al[25]
who reported a similar energy cost pattern for their subjects (healthy elderly aged 68.9 ± 4.6 yrs)
when walking on a treadmill at speeds ranging from 0.67 – 1.56 m.s-1 and the optimal walking
speed was 1.33 m.s-1. A possible hypothesis for this increased energy cost at slower walking
speeds is the increased displacement of the Centre of Mass (CoM) in the mediolateral direction
which progressively reduces as walking speed increases[26]. Conversely the vertical
displacement of the CoM increases with speed, however reducing the vertical CoM displacement
is not a criterion for choosing a comfortable walking speed[26] but it appears that minimising the
mediolateral displacement is more important.
Furthermore our results also suggest that at lower speeds (0.5 – 0.84 m.s-1), the energy cost of
walking on a treadmill is greater than on the ground. This partly agrees with Berryman et al[25]
Page 10 of 17
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
11
who showed that there was greater energy cost of treadmill-walking compared to ground walking
at all the speeds they tested. The explanation for the increased energy cost for treadmill-walking
may be due to a greater need for stabilisation, via muscular contraction, than when walking on
the ground[25]. In addition to advising caution in comparison of test results, we also propose,
that walking tests and clinical cut-offs should be developed on the same testing modality (i.e.
treadmill or ground) as it is proposed to be used in clinic.
The ISWT was originally designed for patients with COPD who typically have much lower
functional capacity than patients referred for cardiac rehabilitation. The relatively higher energy
cost of the (very slow) initial walking speeds may, consequently, lead to undue fatigue. A
modified version of the test starting closer to ‘normal’ walking pace may elicit higher
performance in cardiac rehabilitation patients.
The overall increases in cardiorespiratory fitness in UK patients are low[19 27] compared with
international data[21]. It is of note that this estimate of UK patient gains was based largely on
data from the ISWT and that slightly larger gains reported by the one centre which used treadmill
testing to assess fitness. There appears to be smaller change in exercise capacity reported when
the ISWT is used[2 3 19 28-30] compared with treadmill testing[20]; although it should be noted
that there is a distinct paucity of UK studies using the latter protocol.
Study limitations and conclusions
While we highlight a potential weakness in the ISWT, it is still of clinical utility in measuring
functional capacity and in exercise prescription. Feedback to patients regarding changes in
distance walked during cardiac rehabilitation programs are a valuable motivational tool which is
Page 11 of 17
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
12
easily understood. In conclusion, we do not recommend the direction comparison of estimates of
cardiorespiratory fitness made using shuttle-walking and treadmill walking.
References
1. Pepera G, McAllister J, Sandercock G. Long-term reliability of the incremental shuttle walking test in
clinically stable cardiovascular disease patients. Physiotherapy 2010;96(3):222-7 doi:
10.1016/j.physio.2009.11.010[published Online First: Epub Date]|.
2. Woolf-May K, Bird S. Physical activity levels during phase IV cardiac rehabilitation in a group of male
myocardial infarction patients. Br J Sports Med 2005;39(3):e12; discussion e12
3. Fowler SJ, Singh S. Reproducibility and validity of the incremental shuttle walking test in patients
following coronary artery bypass surgery. Physiotherapy 2005;91:22-27
4. Woolf-May K, Ferrett D. Metabolic equivalents during the 10-m shuttle walking test for post-
myocardial infarction patients. Br J Sports Med 2008;42(1):36-41; discussion 41
5. ACSM. ACSM's guidelines for exercise testing and prescription. 8th ed. Philadelphia: Lippincott
Williams & Wilkins, 2010.
6. SIGN. SIGN 57 Cardiac rehabilitation. A National clinical guideline. Edinburgh: Royal College of
Physicians, 2002.
7. Beneke R, Meyer K. Walking performance and economy in chronic heart failure patients pre and post
exercise training. European journal of applied physiology and occupational physiology
1997;75(3):246-51
8. di Prampero PE. The energy cost of human locomotion on land and in water. International journal of
sports medicine 1986;7(2):55-72 doi: 10.1055/s-2008-1025736[published Online First: Epub
Date]|.
9. Bobbert AC. Energy expenditure in level and grade walking. Journal of Applied Physiology
1960;15:1015-21
10. Cotes JE, Meade F. The energy expenditure and mechanical energy demand in walking. Ergonomics
1960;3:97-119
11. Ralston HJ. Energy-speed relation and optimal speed during level walking. Internationale Zeitschrift
fur angewandte Physiologie, einschliesslich Arbeitsphysiologie 1958;17(4):277-83
12. van der Walt WH, Wyndham CH. An equation for prediction of energy expenditure of walking and
running. J Appl Physiol 1973;34(5):559-63
13. Zaciorskij VM. Biomechanische Grundlagen der Ausdauer. Berlin: Sportverlag Berlin, 1987.
14. Zarrugh MY, Todd FN, Ralston HJ. Optimization of energy expenditure during level walking. European
journal of applied physiology and occupational physiology 1974;33(4):293-306
15. Brueckner JC, Atchou G, Capelli C, et al. The energy cost of running increases with the distance
covered. European journal of applied physiology and occupational physiology 1991;62(6):385-9
16. Brodie D, Bethell H, Breen S. Cardiac rehabilitation in England: a detailed national survey. Eur J
Cardiovasc Prev Rehabil 2006;13(1):122-8
17. BACR. Standards and Core Components for Cardiac Rehabilitation (2007). Secondary Standards and
Core Components for Cardiac Rehabilitation (2007) 2007.
http://www.bcs.com/documents/affiliates/bacr/BACR%20Standards%202007.pdf.
Page 12 of 17
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
13
18. Robinson HJ, Samani NJ, Singh SJ. Can low risk cardiac patients be 'fast tracked' to Phase IV
community exercise schemes for cardiac rehabilitation? A randomised controlled trial. Int J
Cardiol 2011;146(2):159-63 doi: DOI 10.1016/j.ijcard.2009.06.027[published Online First: Epub
Date]|.
19. Almodhy MY, Sandercock GR, Richards L. Changes in cardiorespiratory fitness in patients receiving
supervised outpatient cardiac rehabilitation either once or twice a week. International journal of
cardiology 2012 doi: 10.1016/j.ijcard.2012.06.071[published Online First: Epub Date]|.
20. Sharma A, McLeod AA. Cardiac rehabilitation after coronary artery bypass graft surgery: its effect on
ishcaemia, functional capacity and a mulitvariate index of prognosis. Coronary Health Care 2001
5:189-93
21. Sandercock G, Hurtado V, Cardoso F. Changes in cardiorespiratory fitness in cardiac rehabilitation
patients: A meta-analysis. International journal of cardiology 2011 doi:
10.1016/j.ijcard.2011.11.068[published Online First: Epub Date]|.
22. Workman JM, Armstrong BW. Metabolic cost of walking: equation and model. J Appl Physiol
1986;61(4):1369-74
23. Bunc V, Dlouha R. Energy cost of treadmill walking. The Journal of sports medicine and physical
fitness 1997;37(2):103-9
24. Mian OS, Thom JM, Ardigo LP, Narici MV, Minetti AE. Metabolic cost, mechanical work, and
efficiency during walking in young and older men. Acta Physiol (Oxf) 2006;186(2):127-39 doi:
10.1111/j.1748-1716.2006.01522.x[published Online First: Epub Date]|.
25. Berryman N, Gayda M, Nigam A, Juneau M, Bherer L, Bosquet L. Comparison of the metabolic energy
cost of overground and treadmill walking in older adults. European journal of applied physiology
2012;112(5):1613-20 doi: DOI 10.1007/s00421-011-2102-1[published Online First: Epub Date]|.
26. Orendurff MS, Segal AD, Klute GK, Berge JS, Rohr ES, Kadel NJ. The effect of walking speed on center
of mass displacement. J Rehabil Res Dev 2004;41(6A):829-34 doi: Doi
10.1682/Jrrd.2003.10.0150[published Online First: Epub Date]|.
27. Sandercock GR, Cardoso F, Almodhy M, Pepera G. Cardiorespiratory fitness changes in patients
receiving comprehensive outpatient cardiac rehabilitation in the UK: a multicentre study. Heart
2012 doi: 10.1136/heartjnl-2012-303055[published Online First: Epub Date]|.
28. Sandercock GR, Grocott-Mason R, Brodie DA. Changes in short-term measures of heart rate
variability after eight weeks of cardiac rehabilitation. Clin Auton Res 2007;17(1):39-45
29. Arnold H, Sewel L, Singh S. A comparison of once versus twice per week cardiac rehabilitation. Br J
Cardiol 2007;14:45-48
30. Arnott AS. Assessment of functional capacity in cardiac rehabilitation. Coronary Health Care
1997;1:30-36
Page 13 of 17
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
14
Figure legends
Figure 1 The oxygen uptake of treadmill-walking (black line) and shuttle-walking (grey line) at
each of the seven stages; * = treadmill-walking different from shuttle walking, p < 0.05.
Figure 2 Metabolic power above rest (PN) of treadmill-walking (black line) and shuttle-walking
(grey line) at each of the seven stages, * = treadmill-walking different from shuttle walking, p <
0.05.
Figure 3 Energy cost above rest (CN) per meter distance of treadmill-walking (black line) and
shuttle-walking (grey line) at each of the seven stages, * = treadmill-walking different from
shuttle-walking, p < 0.05.
Page 14 of 17
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
Velocity (m s-1)
0.6 0.8 1.0 1.2 1.4 1.6 1.8
Oxygen Uptake (ml kg-1 min-1)
0
8
12
16
20
24
**
*
Figure 1
Page 15 of 17
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
Velocity (m s-1)
0.6 0.8 1.0 1.2 1.4 1.6 1.8
PN (W kg-1)
0
2
3
4
5
6
7
**
*
*
Figure 2
Page 16 of 17
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
Velocity (m s-1)
0.6 0.8 1.0 1.2 1.4 1.6 1.8
CN (J kg-1 m
-1)
0.0
2.4
2.7
3.0
3.3
3.6
3.9
4.2
4.5
*
*
*
*
Figure 3
Page 17 of 17
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
Pilot investigation of the oxygen demands and metabolic cost of incremental shuttle-walking and treadmill-walking
in patients with cardiovascular disease.
Journal: BMJ Open
Manuscript ID: bmjopen-2014-005216.R1
Article Type: Research
Date Submitted by the Author: 20-Jun-2014
Complete List of Authors: Almodhy, Meshal; Universiity of Essex, Biological Sciences Beneke, Ralph; University of Marburg, Institute of Sports Medicine Cardosos, Fernando; Universiity of Essex, Biological Sciences
Taylor, Matthew; Universiity of Essex, Biological Sciences Sandercock, Gavin; University of Essex, Biological Sciences
<b>Primary Subject Heading</b>:
Cardiovascular medicine
Secondary Subject Heading: Sports and exercise medicine
Keywords: REHABILITATION MEDICINE, Myocardial infarction < CARDIOLOGY, Ischaemic heart disease < CARDIOLOGY
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open on D
ecember 7, 2020 by guest. P
rotected by copyright.http://bm
jopen.bmj.com
/B
MJ O
pen: first published as 10.1136/bmjopen-2014-005216 on 16 S
eptember 2014. D
ownloaded from
For peer review only
1
Pilot investigation of the oxygen demands and metabolic cost of incremental shuttle-1
walking and treadmill-walking in patients with cardiovascular disease. 2
3
Almodhy, M1., Beneke, R
2., Cardoso, F
1., Taylor, MJD.
1 Sandercock, GRH
1 * 4
5
*Corresponding Author 6
1Centre for Sports & Exercise Sciences, School of Biological Sciences, University of Essex 7
Colchester, UK CO4 3SQ 8
2Institut für Sportwissenschaft und Motologie Medezin, Philipps-Universität Marburg, 9
Germany. 10
11
12
13
Oxygen demands of shuttle-walking and treadmill walking 14
Key words: Cardiovascular disease; metabolism; cardiorespiratory fitness; pilot study 15
Word Count: 2498 16
17
Page 1 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
2
Abstract 1
Objective: To determine if the metabolic cost of the incremental shuttle-walking test protocol is 2
the same as treadmill-walking. 3
Setting: Primary care (Community-based cardiac rehabilitation) 4
Participants: Eight caucasian cardiac rehabilitation patients (7 males) with a mean age 67±5.2 5
years. 6
Primary and secondary outcome measures: Oxygen consumption, metabolic power and 7
energy cost of walking during treadmill- and shuttle-walking performed in a balanced order with 8
one week between trials. 9
Results: Average overall energy cost per meter was higher during treadmill-walking (3.22 ± 10
0.55 J.kg
.m
-1) than during shuttle-walking (3.00 ± 0.41 J
.kg
.m
-1). There were significant post hoc 11
effects at 0.67 m.s
-1 and 0.84 m
.s
-1, where the energy cost of treadmill-walking was significantly 12
higher than that of shuttle-walking. This pattern was reversed at walking speeds of 1.52 m.s
-1 and 13
1.69 m.s
-1 where shuttle-walking had a greater energy cost per meter than treadmill-walking. At 14
all walking speeds, the energy cost of shuttle-walking was higher than predicted. 15
Conclusion: The energetic demands of shuttle-walking were fundamentally different from that 16
of treadmill walking and should not be directly compared. We also warn against estimating 17
cardiorespiratory fitness using predictions of energy cost during shuttle-walking via walking-18
speed equations. 19
20
21
22
23
Page 2 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
3
ARTICLE SUMMARY – Strengths and limitations of this study. 1
We provide the first direct comparison of the metabolic cost of shuttle-walking and treadmill-2
walking in cardiac patients. 3
Our data suggest metabolic demands of these exercise modalities appear fundamentally different 4
We suggest suggesting current methods to estimate metabolic cost of shuttle-walking are flawed 5
and warn against risk stratification of cardiac patients based on estimates of cardiorespiratory 6
fitness from the incremental shuttle-walking test. 7
The sample size limits generalisability particularly in female patients who are not represented at 8
higher walking speeds – a larger study of metabolic demands of the incremental shuttle-walking 9
test is warranted. 10
11
Page 3 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
4
INTRODUCTION 1
Since its conception as an alternative to incremental treadmill testing of chronic obstructive 2
pulmonary disease (COPD) patients, the incremental shuttle-walking test (ISWT) has gained 3
popularity as an estimate of functional capacity in numerous clinical populations. The ISWT 4
appears adequately reliable,[1] and is sensitive to changes in functional capacity.[2 3] However, 5
the ISWT’s validity as an estimate of cardiovascular fitness is only moderate[3] and the use of 6
the test to estimate oxygen consumption exercise capacity in Metabolic Equivalents (METs) is 7
questionable.[4] 8
Woolf-May et al. [4] reported acceptable agreement between the energy cost of treadmill 9
walking and the ISWT in healthy volunteers using linear regression analyses, but did not assess 10
this relationship in cardiac patients. The authors reported higher energy demands of shuttle-11
walking in cardiac patients compared with healthy controls. They suggested this may be due to 12
poorer walking economy in the former, they did not report walking economy during ISWT or 13
make comparisons between shuttle- and treadmill-walking economy. 14
Treadmill and shuttle-walking tests are routinely used to assess cardiovascular disease patients 15
and we have previously reported discrete values for change in fitness measured using these 16
tests.[5] Prior to undertaking a proposed multicentre study to identify predictors of change in 17
cardiorespiratory fitness due to cardiac rehabilitation we performed the present pilot study. We 18
examined whether there were differences in the metabolic demands and energy cost of treadmill 19
and shuttle walking in cardiac rehabilitation patients in order to determine whether we could 20
combine data from these tests in our multicentre study. We also compared metabolic demands of 21
the ISWT with predicted values[6] and published estimates [4 7]. 22
23
Page 4 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
5
METHODS 1
Participants (n=8; 7 males; 67±5.2 years: 86.6 ±10.1 kg) were stable cardiac patients attending 2
community-based rehabilitation. The study was approved by the Faculty ethics board at The 3
University of Essex. All patients gave written, informed consent. 4
Equipment 5
The ISWT was performed on a non-slip floor using two cones placed 9 m apart and a portable 6
CD player. The treadmill test was performed on a motorised treadmill (Quaser, HP Cosmos, 7
Nussdorf, Germany). During both tests a portable gas analyser (K4b2 Mobile Breath by Breath 8
Metabolic System, COSMED Pulmonary Function Equipment, Rome, Italy) was used to record 9
expired gas collected via a face and nose mask (Hans Rudolph, Shawnee, Kansas. US). This was 10
calibrated using gases of a known concentration and a syringe before each test. 11
Protocol 12
Patients completed the ISWT and the treadmill test in a balanced order with one week between 13
trials. The ISWT was performed in accordance with national recommendations for cardiac 14
patients[8]. Briefly, the 12 stage protocol starts at a walking speed of 0.5 m·s-1
(1.12 mph) and 15
increases by 0.17 m·s-1
(0.38 mph) each minute. An identical incremental protocol was 16
programmed into the treadmill. Patients were accustomed to treadmill walking but received a 17
brief period of familiarization in which they were required to walk without holding the treadmill 18
handles before the ISWT protocol was also performed. 19
Calculation of metabolic power and energy cost of walking 20
Metabolic power was calculated via indirect calorimetry from and above rest caloric 21
equivalent, and body mass: metabolic power [W.kg
-1] = [ml
.kg
-1.s
-1]
. respiratory exchange 22
ratio adjusted caloric equivalent [J.ml
-1]. [9 10] To analyse the relationship between speed and 23
&VO2 &VCO2
&VO2
Page 5 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
6
metabolic power of walking the metabolic power was predicted as a quadratic function of speed: 1
metabolic power = a + b v². [9 11-15]. The energy cost of walking per metre distance was 2
calculated by: energy cost [J.kg
-1.m
-1] = metabolic power [W
.kg
-1]
. speed
-1 [m
.s
-1]. [9 10 16] 3
4
Statistical Analyses 5
Descriptive results are presented as mean ± SD. A test modality-by-walking speed analysis of 6
variance (ANOVA) with shuttle vs. treadmill walking as within-subjects factor and walking 7
speed as the between-subjects factor was performed. Significant interactions and main effects 8
were further analysed using one-way ANOVA and paired samples t-tests as appropriate. Based 9
on the classical descriptions of walking energy cost [17-19], non-linear regression models were 10
chosen to identify significant interrelationships between metabolic power, energy cost per meter 11
and walking speed, respectively. All analyses were completed using SPSS v19.0 (SPSS Inc. and 12
IBM Company. Chicago, IL) and statistical significance was defined as p < 0.05. 13
14
RESULTS 15
Figure 1 shows the metabolic cost at each of seven stages completed by at least 7 patients. There 16
was a significant main effect for walking speed on oxygen uptake and a significant interaction 17
between treadmill-walking and shuttle-walking on the ground. Oxygen uptake was higher in 18
treadmill-walking than shuttle-walking at 0.67 m·s-1
and 0.84 m·s-1
(p < 0.05; n = 8) but the 19
significantly steeper increases in oxygen demand during shuttle-walking meant the opposite was 20
true at 1.69 m·s-1
(p < 0.05; n = 7). 21
**Insert Fig 1 here** 22
23
Page 6 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
7
Figure 2 shows the metabolic power of treadmill-walking and shuttle-walking. There was a main 1
effect for walking velocity on metabolic power during both treadmill and shuttle walking (p < 2
0.05). The different effects of walking modality on metabolic power were more pronounced if 3
power was predicted as a function of walking speed with power-treadmill walking = 2.028 + 4
1.115 v² and power-shuttle walking on the ground = 1.126 + 1.665 v² where 99 % of the variance 5
of power were explained by the quadratic curve fits in both modalities (both p < 0.001). The 6
difference in response to each modality was indicated by a significant interaction between 7
modality and speed. There were significantly higher metabolic power requirements for treadmill 8
walking at 0.67 m·s-1
and 0.84 m·s-1
(p < 0.05; n = 8). Due to the steeper increase observed in 9
shuttle-walking the metabolic power was significantly higher at 1.52 and 1.69 m⋅s-1
(p < 0.05; 10
n=7) compared with treadmill-walking. 11
12
**Insert Fig 2 here** 13
There were significant main effects for both modality and velocity in relative energy cost (per 14
metre) of walking again well described as a function of speed by the above approximated 15
parameters for both walking modalities (energy cost-treadmill walking = 2.028 / v + 1.115 v and 16
energy cost shuttle-walking on the ground = 1.126 / v + 1.665 v; both p < 0.001). Average 17
overall energy cost per meter (kg·m-1
) was higher during treadmill-walking (3.22 ± 0.55 J·kg·m-
18
1) than during shuttle-walking (3.00 ± 0.41 J·kg·m
-1). There were significant post hoc effects at 19
0.67 m·s-1
and 0.84 m·s-1
(n=8), where the energy cost of treadmill-walking was higher than that 20
of shuttle-walking. Again, this pattern was reversed at higher walking speeds of 1.52 m·s-1
and 21
1.69 m·s-1
where shuttle-walking had a greater energy cost per meter (for the n=7 patients 22
achieving this level) than treadmill-walking. 23
Page 7 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
8
**Insert 1
2
Figure legends. 3
Figure legends 4
Figure 1 The oxygen uptake of treadmill-walking (black line) and shuttle-walking (grey line) at 5
each of the seven stages; * = treadmill-walking different from shuttle walking, p < 0.05. 6
7
Figure 2 Metabolic power above rest (PN) of treadmill-walking (black line) and shuttle-walking 8
(grey line) at each of the seven stages, * = treadmill-walking different from shuttle walking, p < 9
0.05. 10
11
Figure 3 Energy cost above rest (CN) per meter distance of treadmill-walking (black line) and 12
shuttle-walking (grey line) at each of the seven stages, * = treadmill-walking different from 13
shuttle-walking, p < 0.05. 14
15
Table 1. Predicted values, published data and measures of energy expenditure (METs) during 16
the incremental shuttle-walking test. 17
18
Legend: ISWT – Incremental Shuttle Walking Test, ACSM – American College of Sports 19
Medicine [6]. Published ISWT METs in cardiac patients from Woolf-May & Ferrett [4]. 20
*n=7 subjects only. Predicted METs calculated using formula for walking or jogging** from 21
ACSM [6] 22
3 here** 23
Page 8 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
9
DISCUSSION 1
This is the first comparative investigation of the metabolic demands and energy cost per meter 2
walking of incremental treadmill-walking and shuttle-walking in cardiac rehabilitation patients. 3
We found differences in the oxygen requirements and energy cost of shuttle and treadmill 4
walking large enough to suggest results from these exercise modalities should not be pooled in 5
any future analyses. 6
Economy and energy-requirements recorded during level 1 are difficult to interpret 7
as they are most affected by oxygen kinetics and because of long standing phases due to slow 8
walking speed and were excluded from our figures. The patterns of change in walking energy 9
cost per metre on the treadmill show the expected pattern. Slow speeds are associated with 10
higher cost per metre which decreases as optimal (comfortable) walking speed approaches. 11
Continuing to increase walking speed above this pace requires a greater cost per metre. In 12
contrast to this, the energy cost per metre in shuttle-walking decreases only very little and only 13
following the first (very slow) walking pace in the initial stage. The energy cost then increases 14
stage-by-stage throughout the protocol. The cost is only consistent between treadmill and shuttle-15
walking between 1.2 - 1.4 m·s-1
(close to comfortable walking speed) and the increase in energy 16
requirements is much greater in shuttle walking. Based on these pilot data, we intend to report 17
cardiorespiratory fitness values separately according to test modality and recommend this 18
practice to others. 19
20
The classical description of the energy cost during locomotion is of a U-shaped relationship[17] 21
– as speed increases or decreases from the optimal (1.11-1.3 m.s
-1 [17-19]) the energy cost of 22
locomotion increases. For the treadmill protocol our data support this relationship. At slow 23
Page 9 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
10
speeds (0.6-0.8 m.s
-1) energy cost was greater than at optimal speeds (1.2-1.4 m
.s
-1). As walking 1
speed increased (1.6-1.8 m.s
-1) the energy cost again began to increase. This is comparable to 2
Berryman et al[20] who reported a similar energy cost pattern for their subjects (healthy elderly 3
aged 68.9 ± 4.6 yrs) when walking on a treadmill at speeds ranging from 0.67 – 1.56 m.s
-1 and 4
the optimal walking speed was 1.33 m.s
-1. Furthermore our results also suggest that at lower 5
speeds (0.50-0.84 m.s
-1), the energy cost of walking on a treadmill is greater than on the ground. 6
Berryman et al[20] also showed that there was greater energy cost of treadmill-walking 7
compared to ground walking at all the speeds they tested. The explanation for the increased 8
energy cost for treadmill-walking may be due to a greater need for stabilisation, via muscular 9
contraction, than when walking on the ground[20]. 10
Conversely, the oxygen requirement of shuttle-walking are comparatively higher 11
from level 7 (1.52 m·s-1
) onwards than for treadmill walking at the same speed. The 12
requirements are also much higher (18 ml·kg-1
·min-1
) than the value predicted by the ACSM 13
walking speed equations [6] (12.6 ml·kg-1
·min-1
) which are used to estimate cardiorespiratory 14
fitness from ISWT performance[4]. In addition to advising caution in comparison of test results, 15
we also propose, that walking tests and clinical cut-offs should be developed on the same testing 16
modality (i.e. treadmill or ground) as it is proposed to be used in clinic. 17
Cardiac patients’ exercise capacity is commonly expressed as metabolic equivalents 18
(METs) and a comparison of MET values at all ISWT stages, those reported previously[4] and 19
the ACSM-predicted values are shown in table 1. It should be noted that the values predicted 20
using the ACSM walking equations by Woolf-May and Ferret[4] are incorrect. The MET values 21
they reported in cardiac patients are almost double the predicted values using the ACSM 22
equations and much higher than those reported presently. Woolf-May and Ferrett’s [4] MET 23
Page 10 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
11
values further appear anomalous as they are more than double that age-matched controls and 1
significantly higher than recently-reported values in cardiac patients during the ISWT [7]. These 2
latter values [7] do, however, broadly agree with those reported presently. 3
**Insert Table 1 here** 4
Current recommendations suggest patients be classed as high risk if their exercise capacity is <5 5
METs. Failure to reach this criterion standard may lead to prevent patients from entering 6
community-based rehabilitation [21]. Woolf-May and Ferret’s [4] suggestion that ISWT level 4 7
elicits a 5 MET energy cost in cardiac patients is inconsistent with more-recent data from 8
Meadows et al. [7] and those of the present study; both of which suggest the 5 MET threshold is 9
nearer Level 7 or 8. 10
11
Fitter patients can be successfully ‘fast tracked’ to community rehabilitation saving capacity and 12
money to the health providers [22]. However, where exactly in the ISWT protocol this threshold 13
occurs should be determined in a larger, more representative cohort of cardiac patients. 14
Beyond level 7 (1.52 m·s-1
, 3.8 mph) shuttle-walking incurred an additional extra energy cost 15
compared with treadmill walking which may make it difficult to show small improvements in 16
functional capacity if reported as estimated MET values. The exercise capacity of cardiac 17
patients measured before outpatient rehabilitation tends to be lower when estimated from ISWT 18
[23] than when standard treadmill protocols are used. [24 25] 19
20
Study limitations and conclusions 21
Along with sample size, this study is also limited due to including predominantly male patients 22
and indeed only male’s data at the highest walking speeds. The comparison of treadmill and 23
Page 11 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
12
shuttle-walking may have been improved by increasing treadmill gradient as is common-1
practice. We omitted to do this for comparability with previous work.[4 6] The accuracy of 2
energy costs calculations would also be improved by including a resting metabolic measure pre-3
exercise instead of an assumed value (4 ml.kg
-1.m
-1). 4
5
In conclusion, the ISWT may have clinical utility as measure of functional capacity to use in 6
exercise prescription and patient monitoring, but we question its use as an estimate of 7
cardiorespiratory fitness in cardiac patients. Importantly, previous estimates of the ISWT’s 8
energy cost appear erroneous and we warn against any clinical decision-making or risk 9
stratification based on the 5 MET threshold estimated from the ISWT. We recommend a more 10
accurate assessment of the ISWT’s metabolic requirements be performed in a larger, more 11
generalisable sample of cardiac patients. 12
13
14
15
16
17
18
19
20
21
Contributorship statement: Sandercock, Beneke and Taylor devised the experimental design. 22
Almodhy & Cardoso collected and analysed the data. Beneke and Taylor performed the 23
Page 12 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
13
metabolic modeling and advanced statistical analysis. Sandercock and Taylor drafted the 1
manuscript. Beneke, Almodhy and Cardoso revised the manuscript. All authors contributed to 2
the final preparation and drafting of the manuscript. 3
Competing interests None. 4
Funding. None. 5
Data sharing. No additional data available 6
7
8
9
10
11
12
13
14
15
16
17
18
References 19
20 21
Page 13 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
14
1. Pepera G, McAllister J, Sandercock G. Long-term reliability of the incremental shuttle 1
walking test in clinically stable cardiovascular disease patients. Physiotherapy 2
2010;96(3):222-7 3
2. Woolf-May K, Bird S. Physical activity levels during phase IV cardiac rehabilitation in a 4
group of male myocardial infarction patients. Br J Sports Med 2005;39(3):e12; discussion 5
e12 6
3. Fowler SJ, Singh S. Reproducibility and validity of the incremental shuttle walking test in 7
patients following coronary artery bypass surgery. Physiotherapy 2005;91:22-27 8
4. Woolf-May K, Ferrett D. Metabolic equivalents during the 10-m shuttle walking test for post-9
myocardial infarction patients. Br J Sports Med 2008;42(1):36-41; discussion 41 10
5. Sandercock GR, Cardoso F, Almodhy M, et al. Cardiorespiratory fitness changes in patients 11
receiving comprehensive outpatient cardiac rehabilitation in the UK: a multicentre study. 12
Heart 2012 doi: 10.1136/heartjnl-2012-303055[published Online First: Epub Date]|. 13
6. ACSM. ACSM's guidelines for exercise testing and prescription. 8th ed. Philadelphia: 14
Lippincott Williams & Wilkins, 2010. 15
7. Woolf-May K, Meadows S. Exploring adaptations to the modified shuttle walking test. BMJ 16
Open 2013;3(5) doi: 10.1136/bmjopen-2013-002821[published Online First: Epub Date]|. 17
8. SIGN. SIGN 57 Cardiac Rehabilitation. A National Clinical Guideline. Edinburgh: Royal 18
College of Physicians, 2002. 19
9. Beneke R, Meyer K. Walking performance and economy in chronic heart failure patients pre 20
and post exercise training. Eur J Appl Physiol Occup Physiol 1997;75(3):246-51 21
Page 14 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
15
10. di Prampero PE. The energy cost of human locomotion on land and in water. International 1
journal of sports medicine 1986;7(2):55-72 doi: 10.1055/s-2008-1025736[published Online 2
First: Epub Date]|. 3
11. Bobbert AC. Energy expenditure in level and grade walking. Journal of Applied Physiology 4
1960;15:1015-21 5
12. Ralston HJ. Energy-speed relation and optimal speed during level walking. Internationale 6
Zeitschrift fur angewandte Physiologie, einschliesslich Arbeitsphysiologie 1958;17(4):277-7
83 8
13. van der Walt WH, Wyndham CH. An equation for prediction of energy expenditure of 9
walking and running. J Appl Physiol 1973;34(5):559-63 10
14. Zaciorskij VM. Biomechanische Grundlagen der Ausdauer. Berlin: Sportverlag Berlin, 1987. 11
15. Zarrugh MY, Todd FN, Ralston HJ. Optimization of energy expenditure during level 12
walking. European journal of applied physiology and occupational physiology 13
1974;33(4):293-306 14
16. Brueckner JC, Atchou G, Capelli C, et al. The energy cost of running increases with the 15
distance covered. European journal of applied physiology and occupational physiology 16
1991;62(6):385-9 17
17. Workman JM, Armstrong BW. Metabolic cost of walking: equation and model. J Appl 18
Physiol 1986;61(4):1369-74 19
18. Bunc V, Dlouha R. Energy cost of treadmill walking. The Journal of sports medicine and 20
physical fitness 1997;37(2):103-9 21
Page 15 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
16
19. Mian OS, Thom JM, Ardigo LP, et al. Metabolic cost, mechanical work, and efficiency 1
during walking in young and older men. Acta Physiol (Oxf) 2006;186(2):127-39 doi: 2
10.1111/j.1748-1716.2006.01522.x[published Online First: Epub Date]|. 3
20. Berryman N, Gayda M, Nigam A, et al. Comparison of the metabolic energy cost of 4
overground and treadmill walking in older adults. European journal of applied physiology 5
2012;112(5):1613-20 doi: DOI 10.1007/s00421-011-2102-1[published Online First: Epub 6
Date]|. 7
21. BACR. Standards and Core Components for Cardiac Rehabilitation (2007). Secondary 8
Standards and Core Components for Cardiac Rehabilitation (2007) 2007. 9
http://www.bcs.com/documents/affiliates/bacr/BACR%20Standards%202007.pdf. 10
22. Robinson HJ, Samani NJ, Singh SJ. Can low risk cardiac patients be 'fast tracked' to Phase 11
IV community exercise schemes for cardiac rehabilitation? A randomised controlled trial. 12
Int J Cardiol 2011;146(2):159-63 doi: DOI 10.1016/j.ijcard.2009.06.027[published Online 13
First: Epub Date]|. 14
23. Almodhy MY, Sandercock GR, Richards L. Changes in cardiorespiratory fitness in patients 15
receiving supervised outpatient cardiac rehabilitation either once or twice a week. Int J 16
Cardiol 2012 doi: 10.1016/j.ijcard.2012.06.071[published Online First: Epub Date]|. 17
24. Sharma A, McLeod AA. Cardiac rehabilitation after coronary artery bypass graft surgery: its 18
effect on ishcaemia, functional capacity and a mulitvariate index of prognosis. Coronary 19
Health Care 2001 5:189-93 20
25. Sandercock G, Hurtado V, Cardoso F. Changes in cardiorespiratory fitness in cardiac 21
rehabilitation patients: A meta-analysis. Int J Cardiol 2011 doi: 22
10.1016/j.ijcard.2011.11.068[published Online First: Epub Date]|. 23
Page 16 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
17
1
2
Table 1. 3 4 5 ISWT
Protocol
Level
Walking
Speed
(m·s-1
)
ACSM
Predicted
METs
Published
ISWT
METs
Recorded METs:
Treadmill Walking.
Mean (Range)
Recorded METs:
Shuttle Walking.
Mean (Range)
1 0.50 1.9 3.0 2.3 (1.6-2.6) 2.0 (1.6-2.2)
2 0.67 2.1 3.7 3.3 (2.8-4.0) 2.7 (2.5-3.1)
3 0.84 2.4 4.4 3.6 (3.1-4.3) 3.1 (2.8-3.3)
4 1.01 2.7 5.1 3.8 (3.2-4.6) 3.6 (3.2-3.8)
5 1.18 3.0 5.9 4.0 (3.6-4.7) 4.0 (3.6-4.6)
6 1.35 3.3 6.6 4.4 (4.3-5.9) 4.4 (4.0-4.9)
7 1.52 3.6 7.3 5.0 (4.6-6.2) 5.3 (4.8-5.6)
8 1.69 3.9 8.0 5.5 (5.0-6.7) 6.1 (5.7-6.6)
9 1.86 4.2 8.7 -- --
10 2.03 4.5/7.9** 9.4 -- --
11 2.20 4.8/8.5** 10.2 -- --
12 2.37 5.1/9.1** 10.9 -- --
6 7
8
9
10
11
12
13
14
Page 17 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
18
1
2
Figure legends. 3
Figure legends 4
Figure 1 The oxygen uptake of treadmill-walking (black line) and shuttle-walking (grey line) at 5
each of the seven stages; * = treadmill-walking different from shuttle walking, p < 0.05. 6
7
Figure 2 Metabolic power above rest (PN) of treadmill-walking (black line) and shuttle-walking 8
(grey line) at each of the seven stages, * = treadmill-walking different from shuttle walking, p < 9
0.05. 10
11
Figure 3 Energy cost above rest (CN) per meter distance of treadmill-walking (black line) and 12
shuttle-walking (grey line) at each of the seven stages, * = treadmill-walking different from 13
shuttle-walking, p < 0.05. 14
15
Table 1. Predicted values, published data and measures of energy expenditure (METs) during 16
the incremental shuttle-walking test. 17
18
Legend: ISWT – Incremental Shuttle Walking Test, ACSM – American College of Sports 19
Medicine [6]. Published ISWT METs in cardiac patients from Woolf-May & Ferrett [4]. 20
*n=7 subjects only. Predicted METs calculated using formula for walking or jogging** from 21
ACSM [6] 22
23
Page 18 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
19
1
2
Page 19 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
1
Pilot investigation of the oxygen demands and metabolic cost of incremental shuttle-1
walking and treadmill-walking in patients with cardiovascular disease. 2
3
Almodhy, M1., Beneke, R
2., Cardoso, F
1., Taylor, MJD.
1 Sandercock, GRH
1 * 4
5
*Corresponding Author 6
1Centre for Sports & Exercise Sciences, School of Biological Sciences, University of Essex 7
Colchester, UK CO4 3SQ 8
2Institut für Sportwissenschaft und Motologie Medezin, Philipps-Universität Marburg, 9
Germany. 10
11
12
13
Oxygen demands of shuttle-walking and treadmill walking 14
Key words: Cardiovascular disease; metabolism; cardiorespiratory fitness; pilot study 15
Word Count: 2498 16
17
Page 20 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
2
Abstract 1
Objective: To determine if the metabolic cost of the incremental shuttle-walking test protocol is 2
the same as treadmill-walking. 3
Setting: Primary care (Community-based cardiac rehabilitation) 4
Participants: Eight caucasian cardiac rehabilitation patients (7 males) with a mean age 67±5.2 5
years. 6
Primary and secondary outcome measures: Oxygen consumption, metabolic power and 7
energy cost of walking during treadmill- and shuttle-walking performed in a balanced order with 8
one week between trials. 9
Results: Average overall energy cost per meter was higher during treadmill-walking (3.22 ± 10
0.55 J.kg
.m
-1) than during shuttle-walking (3.00 ± 0.41 J
.kg
.m
-1). There were significant post hoc 11
effects at 0.67 m.s
-1 and 0.84 m
.s
-1, where the energy cost of treadmill-walking was significantly 12
higher than that of shuttle-walking. This pattern was reversed at walking speeds of 1.52 m.s
-1 and 13
1.69 m.s
-1 where shuttle-walking had a greater energy cost per meter than treadmill-walking. At 14
all walking speeds, the energy cost of shuttle-walking was higher than predicted. 15
Conclusion: The energetic demands of shuttle-walking were fundamentally different from that 16
of treadmill walking and should not be directly compared. We also warn against estimating 17
cardiorespiratory fitness using predictions of energy cost during shuttle-walking via walking-18
speed equations. 19
20
21
22
23
Page 21 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
3
ARTICLE SUMMARY – Strengths and limitations of this study. 1
We provide the first direct comparison of the metabolic cost of shuttle-walking and treadmill-2
walking in cardiac patients. 3
Our data suggest metabolic demands of these exercise modalities appear fundamentally different 4
We suggest suggesting current methods to estimate metabolic cost of shuttle-walking are flawed 5
and warn against risk stratification of cardiac patients based on estimates of cardiorespiratory 6
fitness from the incremental shuttle-walking test. 7
The sample size limits generalisability particularly in female patients who are not represented at 8
higher walking speeds – a larger study of metabolic demands of the incremental shuttle-walking 9
test is warranted. 10
11
Page 22 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
4
INTRODUCTION 1
Since its conception as an alternative to incremental treadmill testing of chronic obstructive 2
pulmonary disease (COPD) patients, the incremental shuttle-walking test (ISWT) has gained 3
popularity as an estimate of functional capacity in numerous clinical populations. The ISWT 4
appears adequately reliable,[1] and is sensitive to changes in functional capacity.[2 3] However, 5
the ISWT’s validity as an estimate of cardiovascular fitness is only moderate[3] and the use of 6
the test to estimate oxygen consumption exercise capacity in Metabolic Equivalents (METs) is 7
questionable.[4] 8
Woolf-May et al. [4] reported acceptable agreement between the energy cost of treadmill 9
walking and the ISWT in healthy volunteers using linear regression analyses, but did not assess 10
this relationship in cardiac patients. The authors reported higher energy demands of shuttle-11
walking in cardiac patients compared with healthy controls. They suggested this may be due to 12
poorer walking economy in the former, they did not report walking economy during ISWT or 13
make comparisons between shuttle- and treadmill-walking economy. 14
Treadmill and shuttle-walking tests are routinely used to assess cardiovascular disease patients 15
and we have previously reported discrete values for change in fitness measured using these 16
tests.[5] Prior to undertaking a proposed multicentre study to identify predictors of change in 17
cardiorespiratory fitness due to cardiac rehabilitation we performed the present pilot study. We 18
examined whether there were differences in the metabolic demands and energy cost of treadmill 19
and shuttle walking in cardiac rehabilitation patients in order to determine whether we could 20
combine data from these tests in our multicentre study. We also compared metabolic demands of 21
the ISWT with predicted values[6] and published estimates [4 7]. 22
23
Page 23 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
5
METHODS 1
Participants (n=8; 7 males; 67±5.2 years: 86.6 ±10.1 kg) were stable cardiac patients attending 2
community-based rehabilitation. The study was approved by the Faculty ethics board at The 3
University of Essex. All patients gave written, informed consent. 4
Equipment 5
The ISWT was performed on a non-slip floor using two cones placed 9 m apart and a portable 6
CD player. The treadmill test was performed on a motorised treadmill (Quaser, HP Cosmos, 7
Nussdorf, Germany). During both tests a portable gas analyser (K4b2 Mobile Breath by Breath 8
Metabolic System, COSMED Pulmonary Function Equipment, Rome, Italy) was used to record 9
expired gas collected via a face and nose mask (Hans Rudolph, Shawnee, Kansas. US). This was 10
calibrated using gases of a known concentration and a syringe before each test. 11
Protocol 12
Patients completed the ISWT and the treadmill test in a balanced order with one week between 13
trials. The ISWT was performed in accordance with national recommendations for cardiac 14
patients[8]. Briefly, the 12 stage protocol starts at a walking speed of 0.5 m·s-1
(1.12 mph) and 15
increases by 0.17 m·s-1
(0.38 mph) each minute. An identical incremental protocol was 16
programmed into the treadmill. Patients were accustomed to treadmill walking but received a 17
brief period of familiarization in which they were required to walk without holding the treadmill 18
handles before the ISWT protocol was also performed. 19
Calculation of metabolic power and energy cost of walking 20
Metabolic power was calculated via indirect calorimetry from and above rest caloric 21
equivalent, and body mass: metabolic power [W.kg
-1] = [ml
.kg
-1.s
-1]
. respiratory exchange 22
ratio adjusted caloric equivalent [J.ml
-1]. [9 10] To analyse the relationship between speed and 23
&VO2 &VCO2
&VO2
Page 24 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
6
metabolic power of walking the metabolic power was predicted as a quadratic function of speed: 1
metabolic power = a + b v². [9 11-15]. The energy cost of walking per metre distance was 2
calculated by: energy cost [J.kg
-1.m
-1] = metabolic power [W
.kg
-1]
. speed
-1 [m
.s
-1]. [9 10 16] 3
4
Statistical Analyses 5
Descriptive results are presented as mean ± SD. A test modality-by-walking speed analysis of 6
variance (ANOVA) with shuttle vs. treadmill walking as within-subjects factor and walking 7
speed as the between-subjects factor was performed. Significant interactions and main effects 8
were further analysed using one-way ANOVA and paired samples t-tests as appropriate. Based 9
on the classical descriptions of walking energy cost [17-19], non-linear regression models were 10
chosen to identify significant interrelationships between metabolic power, energy cost per meter 11
and walking speed, respectively. All analyses were completed using SPSS v19.0 (SPSS Inc. and 12
IBM Company. Chicago, IL) and statistical significance was defined as p < 0.05. 13
14
RESULTS 15
Figure 1 shows the metabolic cost at each of seven stages completed by at least 7 patients. There 16
was a significant main effect for walking speed on oxygen uptake and a significant interaction 17
between treadmill-walking and shuttle-walking on the ground. Oxygen uptake was higher in 18
treadmill-walking than shuttle-walking at 0.67 m·s-1
and 0.84 m·s-1
(p < 0.05; n = 8) but the 19
significantly steeper increases in oxygen demand during shuttle-walking meant the opposite was 20
true at 1.69 m·s-1
(p < 0.05; n = 7). 21
**Insert Fig 1 here** 22
23
Page 25 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
7
Figure 2 shows the metabolic power of treadmill-walking and shuttle-walking. There was a main 1
effect for walking velocity on metabolic power during both treadmill and shuttle walking (p < 2
0.05). The different effects of walking modality on metabolic power were more pronounced if 3
power was predicted as a function of walking speed with power-treadmill walking = 2.028 + 4
1.115 v² and power-shuttle walking on the ground = 1.126 + 1.665 v² where 99 % of the variance 5
of power were explained by the quadratic curve fits in both modalities (both p < 0.001). The 6
difference in response to each modality was indicated by a significant interaction between 7
modality and speed. There were significantly higher metabolic power requirements for treadmill 8
walking at 0.67 m·s-1
and 0.84 m·s-1
(p < 0.05; n = 8). Due to the steeper increase observed in 9
shuttle-walking the metabolic power was significantly higher at 1.52 and 1.69 m⋅s-1
(p < 0.05; 10
n=7) compared with treadmill-walking. 11
12
**Insert Fig 2 here** 13
There were significant main effects for both modality and velocity in relative energy cost (per 14
metre) of walking again well described as a function of speed by the above approximated 15
parameters for both walking modalities (energy cost-treadmill walking = 2.028 / v + 1.115 v and 16
energy cost shuttle-walking on the ground = 1.126 / v + 1.665 v; both p < 0.001). Average 17
overall energy cost per meter (kg·m-1
) was higher during treadmill-walking (3.22 ± 0.55 J·kg·m-
18
1) than during shuttle-walking (3.00 ± 0.41 J·kg·m
-1). There were significant post hoc effects at 19
0.67 m·s-1
and 0.84 m·s-1
(n=8), where the energy cost of treadmill-walking was higher than that 20
of shuttle-walking. Again, this pattern was reversed at higher walking speeds of 1.52 m·s-1
and 21
1.69 m·s-1
where shuttle-walking had a greater energy cost per meter (for the n=7 patients 22
achieving this level) than treadmill-walking. 23
Page 26 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
8
**Insert Fig 3 here** 1
DISCUSSION 2
This is the first comparative investigation of the metabolic demands and energy cost per meter 3
walking of incremental treadmill-walking and shuttle-walking in cardiac rehabilitation patients. 4
We found differences in the oxygen requirements and energy cost of shuttle and treadmill 5
walking large enough to suggest results from these exercise modalities should not be pooled in 6
any future analyses. 7
Economy and energy-requirements recorded during level 1 are difficult to interpret 8
as they are most affected by oxygen kinetics and because of long standing phases due to slow 9
walking speed and were excluded from our figures. The patterns of change in walking energy 10
cost per metre on the treadmill show the expected pattern. Slow speeds are associated with 11
higher cost per metre which decreases as optimal (comfortable) walking speed approaches. 12
Continuing to increase walking speed above this pace requires a greater cost per metre. In 13
contrast to this, the energy cost per metre in shuttle-walking decreases only very little and only 14
following the first (very slow) walking pace in the initial stage. The energy cost then increases 15
stage-by-stage throughout the protocol. The cost is only consistent between treadmill and shuttle-16
walking between 1.2 - 1.4 m·s-1
(close to comfortable walking speed) and the increase in energy 17
requirements is much greater in shuttle walking. Based on these pilot data, we intend to report 18
cardiorespiratory fitness values separately according to test modality and recommend this 19
practice to others. 20
21
The classical description of the energy cost during locomotion is of a U-shaped relationship[17] 22
– as speed increases or decreases from the optimal (1.11-1.3 m.s
-1 [17-19]) the energy cost of 23
Page 27 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
9
locomotion increases. For the treadmill protocol our data support this relationship. At slow 1
speeds (0.6-0.8 m.s
-1) energy cost was greater than at optimal speeds (1.2-1.4 m
.s
-1). As walking 2
speed increased (1.6-1.8 m.s
-1) the energy cost again began to increase. This is comparable to 3
Berryman et al[20] who reported a similar energy cost pattern for their subjects (healthy elderly 4
aged 68.9 ± 4.6 yrs) when walking on a treadmill at speeds ranging from 0.67 – 1.56 m.s
-1 and 5
the optimal walking speed was 1.33 m.s
-1. Furthermore our results also suggest that at lower 6
speeds (0.50-0.84 m.s
-1), the energy cost of walking on a treadmill is greater than on the ground. 7
Berryman et al[20] also showed that there was greater energy cost of treadmill-walking 8
compared to ground walking at all the speeds they tested. The explanation for the increased 9
energy cost for treadmill-walking may be due to a greater need for stabilisation, via muscular 10
contraction, than when walking on the ground[20]. 11
Conversely, the oxygen requirement of shuttle-walking are comparatively higher 12
from level 7 (1.52 m·s-1
) onwards than for treadmill walking at the same speed. The 13
requirements are also much higher (18 ml·kg-1
·min-1
) than the value predicted by the ACSM 14
walking speed equations [6] (12.6 ml·kg-1
·min-1
) which are used to estimate cardiorespiratory 15
fitness from ISWT performance[4]. In addition to advising caution in comparison of test results, 16
we also propose, that walking tests and clinical cut-offs should be developed on the same testing 17
modality (i.e. treadmill or ground) as it is proposed to be used in clinic. 18
Cardiac patients’ exercise capacity is commonly expressed as metabolic equivalents 19
(METs) and a comparison of MET values at all ISWT stages, those reported previously[4] and 20
the ACSM-predicted values are shown in table 1. It should be noted that the values predicted 21
using the ACSM walking equations by Woolf-May and Ferret[4] are incorrect. The MET values 22
they reported in cardiac patients are almost double the predicted values using the ACSM 23
Page 28 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
10
equations and much higher than those reported presently. Woolf-May and Ferrett’s [4] MET 1
values further appear anomalous as they are more than double that age-matched controls and 2
significantly higher than recently-reported values in cardiac patients during the ISWT [7]. These 3
latter values [7] do, however, broadly agree with those reported presently. 4
**Insert Table 1 here** 5
Current recommendations suggest patients be classed as high risk if their exercise capacity is <5 6
METs. Failure to reach this criterion standard may lead to prevent patients from entering 7
community-based rehabilitation [21]. Woolf-May and Ferret’s [4] suggestion that ISWT level 4 8
elicits a 5 MET energy cost in cardiac patients is inconsistent with more-recent data from 9
Meadows et al. [7] and those of the present study; both of which suggest the 5 MET threshold is 10
nearer Level 7 or 8. 11
12
Fitter patients can be successfully ‘fast tracked’ to community rehabilitation saving capacity and 13
money to the health providers [22]. However, where exactly in the ISWT protocol this threshold 14
occurs should be determined in a larger, more representative cohort of cardiac patients. 15
Beyond level 7 (1.52 m·s-1
, 3.8 mph) shuttle-walking incurred an additional extra energy cost 16
compared with treadmill walking which may make it difficult to show small improvements in 17
functional capacity if reported as estimated MET values. The exercise capacity of cardiac 18
patients measured before outpatient rehabilitation tends to be lower when estimated from ISWT 19
[23] than when standard treadmill protocols are used. [24 25] 20
21
Study limitations and conclusions 22
Page 29 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
11
Along with sample size, this study is also limited due to including predominantly male patients 1
and indeed only male’s data at the highest walking speeds. The comparison of treadmill and 2
shuttle-walking may have been improved by increasing treadmill gradient as is common-3
practice. We omitted to do this for comparability with previous work.[4 6] The accuracy of 4
energy costs calculations would also be improved by including a resting metabolic measure pre-5
exercise instead of an assumed value (4 ml.kg
-1.m
-1). 6
7
In conclusion, the ISWT may have clinical utility as measure of functional capacity to use in 8
exercise prescription and patient monitoring, but we question its use as an estimate of 9
cardiorespiratory fitness in cardiac patients. Importantly, previous estimates of the ISWT’s 10
energy cost appear erroneous and we warn against any clinical decision-making or risk 11
stratification based on the 5 MET threshold estimated from the ISWT. We recommend a more 12
accurate assessment of the ISWT’s metabolic requirements be performed in a larger, more 13
generalisable sample of cardiac patients. 14
15
Contributorship statement: Sandercock, Beneke and Taylor devised the experimental design. 16
Almodhy & Cardoso collected and analysed the data. Beneke and Taylor performed the 17
metabolic modeling and advanced statistical analysis. Sandercock and Taylor drafted the 18
manuscript. Beneke, Almodhy and Cardoso revised the manuscript. All authors contributed to 19
the final preparation and drafting of the manuscript. 20
21
Page 30 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
12
Competing interests None. 1
Funding. None. 2
Data sharing. Not applicable – pilot study data. 3
4
References 5
6 7
1. Pepera G, McAllister J, Sandercock G. Long-term reliability of the incremental shuttle 8
walking test in clinically stable cardiovascular disease patients. Physiotherapy 9
2010;96(3):222-7 10
2. Woolf-May K, Bird S. Physical activity levels during phase IV cardiac rehabilitation in a 11
group of male myocardial infarction patients. Br J Sports Med 2005;39(3):e12; discussion 12
e12 13
3. Fowler SJ, Singh S. Reproducibility and validity of the incremental shuttle walking test in 14
patients following coronary artery bypass surgery. Physiotherapy 2005;91:22-27 15
4. Woolf-May K, Ferrett D. Metabolic equivalents during the 10-m shuttle walking test for post-16
myocardial infarction patients. Br J Sports Med 2008;42(1):36-41; discussion 41 17
5. Sandercock GR, Cardoso F, Almodhy M, Pepera G. Cardiorespiratory fitness changes in 18
patients receiving comprehensive outpatient cardiac rehabilitation in the UK: a multicentre 19
study. Heart 2012 doi: 10.1136/heartjnl-2012-303055[published Online First: Epub Date]|. 20
6. ACSM. ACSM's guidelines for exercise testing and prescription. 8th ed. Philadelphia: 21
Lippincott Williams & Wilkins, 2010. 22
Page 31 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
13
7. Woolf-May K, Meadows S. Exploring adaptations to the modified shuttle walking test. BMJ 1
Open 2013;3(5) doi: 10.1136/bmjopen-2013-002821[published Online First: Epub Date]|. 2
8. SIGN. SIGN 57 Cardiac Rehabilitation. A National Clinical Guideline. Edinburgh: Royal 3
College of Physicians, 2002. 4
9. Beneke R, Meyer K. Walking performance and economy in chronic heart failure patients pre 5
and post exercise training. Eur J Appl Physiol Occup Physiol 1997;75(3):246-51 6
10. di Prampero PE. The energy cost of human locomotion on land and in water. International 7
journal of sports medicine 1986;7(2):55-72 doi: 10.1055/s-2008-1025736[published Online 8
First: Epub Date]|. 9
11. Bobbert AC. Energy expenditure in level and grade walking. Journal of Applied Physiology 10
1960;15:1015-21 11
12. Ralston HJ. Energy-speed relation and optimal speed during level walking. Internationale 12
Zeitschrift fur angewandte Physiologie, einschliesslich Arbeitsphysiologie 1958;17(4):277-13
83 14
13. van der Walt WH, Wyndham CH. An equation for prediction of energy expenditure of 15
walking and running. J Appl Physiol 1973;34(5):559-63 16
14. Zaciorskij VM. Biomechanische Grundlagen der Ausdauer. Berlin: Sportverlag Berlin, 1987. 17
15. Zarrugh MY, Todd FN, Ralston HJ. Optimization of energy expenditure during level 18
walking. European journal of applied physiology and occupational physiology 19
1974;33(4):293-306 20
16. Brueckner JC, Atchou G, Capelli C, et al. The energy cost of running increases with the 21
distance covered. European journal of applied physiology and occupational physiology 22
1991;62(6):385-9 23
Page 32 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
14
17. Workman JM, Armstrong BW. Metabolic cost of walking: equation and model. J Appl 1
Physiol 1986;61(4):1369-74 2
18. Bunc V, Dlouha R. Energy cost of treadmill walking. The Journal of sports medicine and 3
physical fitness 1997;37(2):103-9 4
19. Mian OS, Thom JM, Ardigo LP, Narici MV, Minetti AE. Metabolic cost, mechanical work, 5
and efficiency during walking in young and older men. Acta Physiol (Oxf) 6
2006;186(2):127-39 doi: 10.1111/j.1748-1716.2006.01522.x[published Online First: Epub 7
Date]|. 8
20. Berryman N, Gayda M, Nigam A, Juneau M, Bherer L, Bosquet L. Comparison of the 9
metabolic energy cost of overground and treadmill walking in older adults. European journal 10
of applied physiology 2012;112(5):1613-20 doi: DOI 10.1007/s00421-011-2102-11
1[published Online First: Epub Date]|. 12
21. BACR. Standards and Core Components for Cardiac Rehabilitation (2007). Secondary 13
Standards and Core Components for Cardiac Rehabilitation (2007) 2007. 14
http://www.bcs.com/documents/affiliates/bacr/BACR%20Standards%202007.pdf. 15
22. Robinson HJ, Samani NJ, Singh SJ. Can low risk cardiac patients be 'fast tracked' to Phase 16
IV community exercise schemes for cardiac rehabilitation? A randomised controlled trial. 17
Int J Cardiol 2011;146(2):159-63 doi: DOI 10.1016/j.ijcard.2009.06.027[published Online 18
First: Epub Date]|. 19
23. Almodhy MY, Sandercock GR, Richards L. Changes in cardiorespiratory fitness in patients 20
receiving supervised outpatient cardiac rehabilitation either once or twice a week. Int J 21
Cardiol 2012 doi: 10.1016/j.ijcard.2012.06.071[published Online First: Epub Date]|. 22
Page 33 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
15
24. Sharma A, McLeod AA. Cardiac rehabilitation after coronary artery bypass graft surgery: its 1
effect on ishcaemia, functional capacity and a mulitvariate index of prognosis. Coronary 2
Health Care 2001 5:189-93 3
25. Sandercock G, Hurtado V, Cardoso F. Changes in cardiorespiratory fitness in cardiac 4
rehabilitation patients: A meta-analysis. Int J Cardiol 2011 doi: 5
10.1016/j.ijcard.2011.11.068[published Online First: Epub Date]|. 6
7
8
9
10
11
Figure legends. 12
Figure legends 13
Figure 1 The oxygen uptake of treadmill-walking (black line) and shuttle-walking (grey line) at 14
each of the seven stages; * = treadmill-walking different from shuttle walking, p < 0.05. 15
16
Figure 2 Metabolic power above rest (PN) of treadmill-walking (black line) and shuttle-walking 17
(grey line) at each of the seven stages, * = treadmill-walking different from shuttle walking, p < 18
0.05. 19
20
Figure 3 Energy cost above rest (CN) per meter distance of treadmill-walking (black line) and 21
shuttle-walking (grey line) at each of the seven stages, * = treadmill-walking different from 22
shuttle-walking, p < 0.05. 23
Page 34 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
16
1
Table 1. Predicted values, published data and measures of energy expenditure (METs) during 2
the incremental shuttle-walking test. 3
4
Legend: ISWT – Incremental Shuttle Walking Test, ACSM – American College of Sports 5
Medicine [6]. Published ISWT METs in cardiac patients from Woolf-May & Ferrett [4]. 6
*n=7 subjects only. Predicted METs calculated using formula for walking or jogging** from 7
ACSM [6] 8
9
10
Page 35 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
90x70mm (300 x 300 DPI)
Page 36 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
Figure 2 Metabolic power above rest (PN) of treadmill-walking (black line) and shuttle-walking (grey line) at each of the seven stages, * = treadmill-walking different from shuttle walking, p < 0.05.
90x71mm (300 x 300 DPI)
Page 37 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
Figure 2 Metabolic power above rest (PN) of treadmill-walking (black line) and shuttle-walking (grey line) at each of the seven stages, * = treadmill-walking different from shuttle walking, p < 0.05.
90x71mm (300 x 300 DPI)
Page 38 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
STARD checklist for reporting of studies of diagnostic accuracy
(version January 2003)
Section and Topic Item
#
On page #
TITLE/ABSTRACT/
KEYWORDS
1 Identify the article as a study of diagnostic accuracy (recommend MeSH
heading 'sensitivity and specificity').
2
INTRODUCTION 2 State the research questions or study aims, such as estimating diagnostic
accuracy or comparing accuracy between tests or across participant
groups.
3
METHODS
Participants 3 The study population: The inclusion and exclusion criteria, setting and
locations where data were collected.
5
4 Participant recruitment: Was recruitment based on presenting symptoms,
results from previous tests, or the fact that the participants had received
the index tests or the reference standard?
5
5 Participant sampling: Was the study population a consecutive series of
participants defined by the selection criteria in item 3 and 4? If not,
specify how participants were further selected.
5
6 Data collection: Was data collection planned before the index test and
reference standard were performed (prospective study) or after
(retrospective study)?
5
Test methods 7 The reference standard and its rationale. 5
8 Technical specifications of material and methods involved including how
and when measurements were taken, and/or cite references for index
tests and reference standard.
5
9 Definition of and rationale for the units, cut-offs and/or categories of the
results of the index tests and the reference standard.
5, 9, 10
10 The number, training and expertise of the persons executing and reading
the index tests and the reference standard.
5
11 Whether or not the readers of the index tests and reference standard
were blind (masked) to the results of the other test and describe any
other clinical information available to the readers.
5
Statistical methods 12 Methods for calculating or comparing measures of diagnostic accuracy,
and the statistical methods used to quantify uncertainty (e.g. 95%
confidence intervals).
6,7,10
13 Methods for calculating test reproducibility, if done. -
RESULTS
Participants 14 When study was performed, including beginning and end dates of
recruitment.
6
15 Clinical and demographic characteristics of the study population (at least
information on age, gender, spectrum of presenting symptoms).
6
16 The number of participants satisfying the criteria for inclusion who did or
did not undergo the index tests and/or the reference standard; describe
why participants failed to undergo either test (a flow diagram is strongly
recommended).
6
Test results 17 Time-interval between the index tests and the reference standard, and
any treatment administered in between.
5
18 Distribution of severity of disease (define criteria) in those with the target
condition; other diagnoses in participants without the target condition.
5,10
19 A cross tabulation of the results of the index tests (including
indeterminate and missing results) by the results of the reference
standard; for continuous results, the distribution of the test results by the
results of the reference standard.
Table 1,
P10
20 Any adverse events from performing the index tests or the reference
standard.
6
Estimates 21 Estimates of diagnostic accuracy and measures of statistical uncertainty
(e.g. 95% confidence intervals).
9,10,11
22 How indeterminate results, missing data and outliers of the index tests
were handled.
10
23 Estimates of variability of diagnostic accuracy between subgroups of
participants, readers or centers, if done.
-
24 Estimates of test reproducibility, if done. -
DISCUSSION 25 Discuss the clinical applicability of the study findings. 10,11,12
Page 39 of 39
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
Pilot investigation of the oxygen demands and metabolic cost of incremental shuttle-walking and treadmill-walking
in patients with cardiovascular disease.
Journal: BMJ Open
Manuscript ID: bmjopen-2014-005216.R2
Article Type: Research
Date Submitted by the Author: 09-Aug-2014
Complete List of Authors: Almodhy, Meshal; Universiity of Essex, Biological Sciences Beneke, Ralph; University of Marburg, Institute of Sports Medicine Cardosos, Fernando; Universiity of Essex, Biological Sciences
Taylor, Matthew; Universiity of Essex, Biological Sciences Sandercock, Gavin; University of Essex, Biological Sciences
<b>Primary Subject Heading</b>:
Cardiovascular medicine
Secondary Subject Heading: Sports and exercise medicine
Keywords: REHABILITATION MEDICINE, Myocardial infarction < CARDIOLOGY, Ischaemic heart disease < CARDIOLOGY
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open on D
ecember 7, 2020 by guest. P
rotected by copyright.http://bm
jopen.bmj.com
/B
MJ O
pen: first published as 10.1136/bmjopen-2014-005216 on 16 S
eptember 2014. D
ownloaded from
For peer review only
1
Pilot investigation of the oxygen demands and metabolic cost of incremental shuttle-1
walking and treadmill-walking in patients with cardiovascular disease. 2
3
Almodhy, M1., Beneke, R
2., Cardoso, F
1., Taylor, MJD.
1 Sandercock, GRH
1 * 4
5
*Corresponding Author 6
1Centre for Sports & Exercise Sciences, School of Biological Sciences, University of Essex 7
Colchester, UK CO4 3SQ 8
2Institut für Sportwissenschaft und Motologie Medezin, Philipps-Universität Marburg, 9
Germany. 10
11
12
13
Oxygen demands of shuttle-walking and treadmill walking 14
Key words: Cardiovascular disease; metabolism; cardiorespiratory fitness; pilot study 15
Word Count: 2498 16
17
Page 1 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
2
Abstract 1
Objective: To determine if the metabolic cost of the incremental shuttle-walking test protocol is 2
the same as treadmill-walking or predicted values walking speed equations. 3
Setting: Primary care (Community-based cardiac rehabilitation) 4
Participants: Eight caucasian cardiac rehabilitation patients (7 males) with a mean age 67±5.2 5
years. 6
Primary and secondary outcome measures: Oxygen consumption, metabolic power and 7
energy cost of walking during treadmill- and shuttle-walking performed in a balanced order with 8
one week between trials. 9
Results: Average overall energy cost per meter was higher during treadmill-walking (3.22 ± 10
0.55 J.kg
.m
-1) than during shuttle-walking (3.00 ± 0.41 J
.kg
.m
-1). There were significant post hoc 11
effects at 0.67 m.s
-1 (p < 0.004) and 0.84 m
.s
-1 (p < 0.001), where the energy cost of treadmill-12
walking was significantly higher than that of shuttle-walking. This pattern was reversed at 13
walking speeds 1.52 m·s-1
(p < 0.042) and 1.69 m·s-1
(p < 0.007) where shuttle-walking had a 14
greater energy cost per meter than treadmill-walking. At all walking speeds, the energy cost of 15
shuttle-walking was higher than predicted using the American College of Sports Medicine 16
walking equations. 17
Conclusion: The energetic demands of shuttle-walking were fundamentally different from that 18
of treadmill walking and should not be directly compared. We warn against estimating the 19
metabolic cost of the incremental shuttle-walking test using the current walking-speed equations. 20
21
22
23
Page 2 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
3
1
ARTICLE SUMMARY – Strengths and limitations of this study. 2
We provide the first direct comparison of the metabolic cost of shuttle-walking and treadmill-3
walking in cardiac patients. 4
Our data suggest metabolic demands of these exercise modalities appear fundamentally different 5
We suggest suggesting current methods to estimate metabolic cost of shuttle-walking are flawed 6
We warn against risk stratification of cardiac patients based on estimated oxygen costs using the 7
American College of Sports Medicine walking equations during the incremental shuttle-walking 8
test. 9
The sample size limits generalisability particularly in female patients who are not represented at 10
higher walking speeds – a larger study of metabolic cost of the incremental shuttle-walking test 11
is warranted. 12
13
Page 3 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
4
INTRODUCTION 1
Since its conception as an alternative to incremental treadmill testing of chronic obstructive 2
pulmonary disease (COPD) patients, the incremental shuttle-walking test (ISWT) has gained 3
popularity as an estimate of functional capacity in numerous clinical populations. The ISWT 4
appears adequately reliable,[1] and is sensitive to changes in functional capacity.[2 3] However, 5
the ISWT’s validity as an estimate of cardiovascular fitness is only moderate[3] and the use of 6
the test to estimate oxygen consumption exercise capacity in Metabolic Equivalents (METs) is 7
questionable.[4] 8
Woolf-May et al. [4] reported acceptable agreement between the energy cost of treadmill 9
walking and the ISWT in healthy volunteers using linear regression analyses, but did not assess 10
this relationship in cardiac patients. The authors reported higher energy demands of shuttle-11
walking in cardiac patients compared with healthy controls. They suggested this may be due to 12
poorer walking economy in the former, they did not report walking economy during ISWT or 13
make comparisons between shuttle- and treadmill-walking economy. 14
Treadmill and shuttle-walking tests are routinely used to assess cardiovascular disease patients 15
and we have previously reported discrete values for change in fitness measured using these 16
tests.[5] Prior to undertaking a proposed multicentre study to identify predictors of change in 17
cardiorespiratory fitness due to cardiac rehabilitation we performed the present pilot study. We 18
examined whether there were differences in the metabolic demands and energy cost of treadmill 19
and shuttle walking in cardiac rehabilitation patients in order to determine whether we could 20
combine data from these tests in our multicentre study. We also compared metabolic cost of the 21
ISWT with values predicted from treadmill-walking equations[6] and published estimates [4 7]. 22
23
Page 4 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
5
METHODS 1
Participants (n=8; 7 males; 67±5.2 years: 86.6 ±10.1 kg) were stable cardiac patients attending 2
community-based rehabilitation following elective cardiac revascularization. The study was 3
approved by the Faculty ethics board at The University of Essex. All patients gave written, 4
informed consent. 5
Equipment 6
The ISWT was performed on a non-slip floor using two cones placed 9 m apart and a portable 7
CD player. The treadmill test was performed on a motorised treadmill (Quaser, HP Cosmos, 8
Nussdorf, Germany). During both tests a portable gas analyser (K4b2 Mobile Breath by Breath 9
Metabolic System, COSMED Pulmonary Function Equipment, Rome, Italy) was used to record 10
expired gas collected via a face and nose mask (Hans Rudolph, Shawnee, Kansas. US). This was 11
calibrated using gases of a known concentration and a syringe before each test. 12
Protocol 13
Patients completed the ISWT and the treadmill test in a balanced order with one week between 14
trials. The ISWT was performed in accordance with national recommendations for cardiac 15
patients[8]. Briefly, the 12 stage protocol starts at a walking speed of 0.5 m·s-1
(1.12 mph) and 16
increases by 0.17 m·s-1
(0.38 mph) each minute. An identical incremental protocol was 17
programmed into the treadmill. Patients were accustomed to treadmill walking but received a 18
brief period of familiarization in which they were required to walk without holding the treadmill 19
handles before the ISWT protocol was also performed. 20
Calculation of metabolic power and energy cost of walking 21
We assumed a standard resting metabolic rate of 4 ml.kg
-1.min
-1 based on reference standards[9]. 22
Metabolic power was then calculated via indirect calorimetry from and above rest 23 &VO2 &VCO2
Page 5 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
6
and from body mass: metabolic power [W.kg
-1] = ( - rest) [ml
.kg
-1.s
-1]
. respiratory 1
exchange ratio adjusted caloric equivalent [J.ml
-1]. [10 11] To analyse the relationship between 2
speed and metabolic power of walking the metabolic power was predicted as a quadratic function 3
of speed: metabolic power = a + b v² [10 12-16]. The energy cost of walking per metre distance 4
was calculated by: energy cost [J.kg
-1.m
-1] = metabolic power [W
.kg
-1]
. speed
-1 [m
.s
-1]. [10 11 17] 5
6
Statistical Analyses 7
Descriptive results are presented as mean ± SD. A test modality-by-walking speed analysis of 8
variance (ANOVA) with shuttle vs. treadmill walking as within-subjects factor and walking 9
speed as the between-subjects factor was performed. Significant interactions and main effects 10
were further analysed using one-way ANOVA and paired samples t-tests as appropriate. Based 11
on the classical descriptions of walking energy cost [18-20], non-linear regression models were 12
chosen to identify significant interrelationships between metabolic power, energy cost per meter 13
and walking speed, respectively. All analyses were completed using SPSS v19.0 (SPSS Inc. and 14
IBM Company. Chicago, IL) and statistical significance was defined as p < 0.05. 15
16
RESULTS 17
Figure 1 shows the oxygen uptake at each of seven stages completed by at least 7 patients. There 18
was a significant main effect for walking speed on oxygen uptake and a significant interaction 19
between treadmill-walking and shuttle-walking on the ground. Oxygen uptake was higher in 20
treadmill-walking than shuttle-walking at 0.67 m·s-1
(p=0.006; n=8) and 0.84 m·s-1
(p=0.003; n 21
= 8) but the significantly steeper increases in oxygen demand during shuttle-walking meant the 22
opposite was true at 1.69 m·s-1
(p < 0.006; n = 7). 23
&VO2 &VO2
Page 6 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
7
**Insert Fig 1 here** 1
2
Figure 2 shows the metabolic power of treadmill-walking and shuttle-walking. There was a main 3
effect for walking velocity on metabolic power during both treadmill and shuttle walking (p < 4
0.05). The different effects of walking modality on metabolic power were more pronounced if 5
power was predicted as a function of walking speed with power-treadmill walking = 2.028 + 6
1.115 v² and power-shuttle walking on the ground = 1.126 + 1.665 v² where 99% of the variance 7
of power were explained by the quadratic curve fits in both modalities (both p < 0.001). The 8
difference in response to each modality was indicated by a significant interaction between 9
modality and speed. There were significantly higher metabolic power requirements for treadmill 10
walking at 0.67 m·s-1
(p < 0.004; n = 8) and 0.84 m·s-1
(p < 0.001; n = 8). Due to the steeper 11
increase observed in shuttle-walking the metabolic power was significantly higher at 1.52 m⋅s-1
12
(p=0.042; n=7) and 1.69 m⋅s-1
(p=0.007; n=7) compared with treadmill-walking. 13
14
**Insert Figure 2 here** 15
There were significant main effects for both modality and velocity in relative energy cost (per 16
metre) of walking again well described as a function of speed by the above approximated 17
parameters for both walking modalities (energy cost-treadmill walking = 2.028 / v + 1.115 v and 18
energy cost shuttle-walking on the ground = 1.126 / v + 1.665 v; both p < 0.001). Average 19
overall energy cost per meter (kg·m-1
) was higher during treadmill-walking (3.22 ± 0.55 J·kg·m-
20
1) than during shuttle-walking (3.00 ± 0.41 J·kg·m
-1). There were significant post hoc effects at 21
0.67 m·s-1
(p < 0.004; n=8) and 0.84 m·s-1
(p < 0.001; n=8), where the energy cost of treadmill-22
walking was higher than that of shuttle-walking. Again, this pattern was reversed at higher 23
Page 7 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
8
walking speeds of 1.52 m·s-1
(p < 0.042) and 1.69 m·s-1
(p < 0.007) where shuttle-walking had a 1
greater energy cost per meter (for the n=7 patients achieving this level) than treadmill-walking. 2
3
**Insert Figure 3 here** 4
DISCUSSION 5
This is the first comparative investigation of the metabolic demands and energy cost per meter 6
walking of incremental treadmill-walking and shuttle-walking in cardiac rehabilitation patients. 7
We found differences in the oxygen requirements and energy cost of shuttle and treadmill 8
walking large enough to suggest results from these exercise modalities should not be pooled in 9
any future analyses. 10
Economy and energy-requirements recorded during level 1 are difficult to interpret 11
as they are most affected by oxygen kinetics and patients’ unusually long stance phase during 12
their gait cycle at this very slow walking speed and were excluded from our figures. The change 13
in walking energy cost per metre on the treadmill show the expected pattern. Slow speeds are 14
associated with higher cost per metre which decreases as optimal (comfortable) walking speed 15
approaches. Continuing to increase walking speed above this pace requires a greater cost per 16
metre. In contrast to this, the energy cost per metre in shuttle-walking decreases only very little 17
and only following the first (very slow) walking pace in the initial stage. The energy cost then 18
increases stage-by-stage throughout the protocol. The cost is only consistent between treadmill 19
and shuttle-walking between 1.2 - 1.4 m·s-1
(close to comfortable walking speed) and the 20
increase in energy requirements is much greater in shuttle walking. Based on these pilot data, we 21
intend to report cardiorespiratory fitness values separately according to test modality and 22
recommend this practice to others. 23
Page 8 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
9
The classical description of the energy cost during locomotion is of a U-shaped relationship[18] 1
– as speed increases or decreases from the optimal (1.11-1.3 m.s
-1 [18-20]) the energy cost of 2
locomotion increases. For the treadmill protocol our data support this relationship. At slow 3
speeds (0.6-0.8 m.s
-1) energy cost was greater than at optimal speeds (1.2-1.4 m
.s
-1). As walking 4
speed increased (1.6-1.8 m.s
-1) the energy cost again began to increase. This is comparable to 5
Berryman et al[21] who reported a similar energy cost pattern for their subjects (healthy elderly 6
aged 68.9 ± 4.6 yrs) when walking on a treadmill at speeds ranging from 0.67 – 1.56 m.s
-1 and 7
the optimal walking speed was 1.33 m.s
-1. Furthermore our results also suggest that at lower 8
speeds (0.50-0.84 m.s
-1), the energy cost of walking on a treadmill is greater than on the ground. 9
Berryman et al[21] also showed that there was greater energy cost of treadmill-walking 10
compared to ground walking at all the speeds they tested. The explanation for the increased 11
energy cost for treadmill-walking may be due to a greater need for stabilisation, via muscular 12
contraction, than when walking on the ground[21]. 13
Conversely, the oxygen requirement of shuttle-walking are comparatively higher 14
from level 7 (1.52 m·s-1
) onwards than for treadmill walking at the same speed. The 15
requirements are also much higher (18 ml·kg-1
·min-1
) than the value predicted by the ACSM 16
walking speed equations [6] (12.6 ml·kg-1
·min-1
) which are used to estimate cardiorespiratory 17
fitness from ISWT performance[4]. In addition to differences in oxygen requirements of ground 18
and treadmill walking, shuttle-walking may have a higher cost due to repeated acceleration, 19
deceleration phases or the negotiation of turns[7]. We propose, therefore that any clinical cut-20
offs for walking tests should be developed using the same testing modality as that for which they 21
are proposed for use in (i.e. treadmill or ground). 22
Page 9 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
10
Cardiac patients’ exercise capacity is commonly expressed as metabolic equivalents 1
(METs). We calculated metabolic cost in METs (Gross [ml.kg
-1.min
-1] /3.5) and compared 2
MET values at all ISWT stages with those reported previously[4] and the ACSM-predicted 3
values (Table 1). It should be noted that the values predicted using the ACSM walking equations 4
by Woolf-May and Ferret[4] are incorrect. The MET values they reported in cardiac patients are 5
almost double the predicted values using the ACSM equations and much higher than those 6
reported presently. Woolf-May and Ferrett’s [4] MET values further appear anomalous as they 7
are more than double that age-matched controls and significantly higher than recently-reported 8
values in cardiac patients during the ISWT [7]. These latter values [7] do, however, broadly 9
agree with those reported presently. 10
**Insert Table 1 here** 11
Current recommendations suggest patients be classed as high risk if their exercise capacity is <5 12
METs. Failure to reach this criterion standard may lead to prevent patients from entering 13
community-based rehabilitation [22]. Woolf-May and Ferret’s [4] suggestion that ISWT level 4 14
elicits a 5 MET energy cost in cardiac patients is inconsistent with more-recent data from Woolf-15
May & Meadows [7] and those of the present study; both of which suggest the 5 MET threshold 16
is nearer Level 7 or 8. 17
18
Fitter patients can be successfully ‘fast tracked’ to community rehabilitation saving capacity and 19
money to the health providers [23]. However, where exactly in the ISWT protocol this threshold 20
occurs should be determined in a larger, more representative cohort of cardiac patients. 21
Beyond level 7 (1.52 m·s-1
, 3.8 mph) shuttle-walking incurred an additional extra energy cost 22
compared with treadmill walking which may make it difficult to show small improvements in 23
&VO2
Page 10 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
11
functional capacity if reported as estimated MET values. The exercise capacity of cardiac 1
patients measured before outpatient rehabilitation tends to be lower when estimated from ISWT 2
[24] than when standard treadmill protocols are used. [25 26] 3
4
Study limitations and conclusions 5
Along with sample size, this study is also limited due to including predominantly male patients 6
and indeed only male’s data at the highest walking speeds. The comparison of treadmill and 7
shuttle-walking may have been improved by increasing treadmill gradient as is common-8
practice. We omitted to do this for comparability with previous work. [4 6] The accuracy of 9
energy costs calculations would also be improved by including a resting metabolic measure pre-10
exercise instead of an assumed value of 4 ml.kg
-1.m
-1. [9] 11
12
In conclusion, the ISWT may have clinical utility as measure of functional capacity to use in 13
exercise prescription and patient monitoring, but we question its use as an estimate of 14
cardiorespiratory fitness in cardiac patients. Importantly, the ACSM walking equations grossly 15
underestimate the actual energy cost of shuttle-walking and should not be used in research or 16
clinical practice. Our comparison using metabolic equivalents (METs) also reveals that some 17
published [4] estimates of the ISWT’s energy cost in cardiac patients appear erroneously high. 18
Given these two shortcomings, we strongly warn against clinical decision-making or patient risk 19
stratification based on achieving the 5 MET threshold estimated using the ISWT. We 20
recommend a more accurate assessment of the ISWT’s energy cost be performed in a larger, 21
more generalisable sample of cardiac patients. 22
23
Page 11 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
12
Contributorship statement: Sandercock, Beneke and Taylor devised the experimental design. 1
Almodhy & Cardoso collected and analysed the data. Beneke performed the metabolic modeling 2
and advanced statistical analysis. Sandercock and Taylor drafted the manuscript. Beneke, 3
Almodhy and Cardoso revised the manuscript. All authors contributed to the final preparation 4
and drafting of the manuscript. 5
Competing interests None. 6
Funding. None. 7
Data sharing. No additional data available 8
9
10
11
12
13
14
15
16
17
18
19
20
21
Page 12 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
13
References 1
2
3
1. Pepera G, McAllister J, Sandercock G. Long-term reliability of the incremental shuttle 4
walking test in clinically stable cardiovascular disease patients. Physiotherapy 5
2010;96(3):222-7 6
2. Woolf-May K, Bird S. Physical activity levels during phase IV cardiac rehabilitation in a 7
group of male myocardial infarction patients. Br J Sports Med 2005;39(3):e12; discussion 8
e12 9
3. Fowler SJ, Singh S. Reproducibility and validity of the incremental shuttle walking test in 10
patients following coronary artery bypass surgery. Physiotherapy 2005;91:22-27 11
4. Woolf-May K, Ferrett D. Metabolic equivalents during the 10-m shuttle walking test for post-12
myocardial infarction patients. Br J Sports Med 2008;42(1):36-41; discussion 41 13
5. Sandercock GR, Cardoso F, Almodhy M, et al. Cardiorespiratory fitness changes in patients 14
receiving comprehensive outpatient cardiac rehabilitation in the UK: a multicentre study. 15
Heart 2012 doi: 10.1136/heartjnl-2012-303055[published Online First: Epub Date]|. 16
6. ACSM. ACSM's guidelines for exercise testing and prescription. 8th ed. Philadelphia: 17
Lippincott Williams & Wilkins, 2010. 18
7. Woolf-May K, Meadows S. Exploring adaptations to the modified shuttle walking test. BMJ 19
Open 2013;3(5) doi: 10.1136/bmjopen-2013-002821[published Online First: Epub Date]|. 20
8. SIGN. SIGN 57 Cardiac Rehabilitation. A National Clinical Guideline. Edinburgh: Royal 21
College of Physicians, 2002. 22
Page 13 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
14
9. Wissenschaftliche Tabellen Geigy Teilband Körperflüssigkeiten. (Scientific Tables Geigy, 1
Volume Body Fluids.) In: Lentner C, ed. Giegy Scientific Tables. 7th ed. Basel: Ciba-Geigy, 2
1985:225-28. 3
10. Beneke R, Meyer K. Walking performance and economy in chronic heart failure patients pre 4
and post exercise training. Eur J Appl Physiol Occup Physiol 1997;75(3):246-51 5
11. di Prampero PE. The energy cost of human locomotion on land and in water. International 6
journal of sports medicine 1986;7(2):55-72 doi: 10.1055/s-2008-1025736[published Online 7
First: Epub Date]|. 8
12. Bobbert AC. Energy expenditure in level and grade walking. Journal of Applied Physiology 9
1960;15:1015-21 10
13. Ralston HJ. Energy-speed relation and optimal speed during level walking. Internationale 11
Zeitschrift fur angewandte Physiologie, einschliesslich Arbeitsphysiologie 1958;17(4):277-12
83 13
14. van der Walt WH, Wyndham CH. An equation for prediction of energy expenditure of 14
walking and running. J Appl Physiol 1973;34(5):559-63 15
15. Zaciorskij VM. Biomechanische Grundlagen der Ausdauer. Berlin: Sportverlag Berlin, 1987. 16
16. Zarrugh MY, Todd FN, Ralston HJ. Optimization of energy expenditure during level 17
walking. European journal of applied physiology and occupational physiology 18
1974;33(4):293-306 19
17. Brueckner JC, Atchou G, Capelli C, et al. The energy cost of running increases with the 20
distance covered. European journal of applied physiology and occupational physiology 21
1991;62(6):385-9 22
Page 14 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
15
18. Workman JM, Armstrong BW. Metabolic cost of walking: equation and model. J Appl 1
Physiol 1986;61(4):1369-74 2
19. Bunc V, Dlouha R. Energy cost of treadmill walking. The Journal of sports medicine and 3
physical fitness 1997;37(2):103-9 4
20. Mian OS, Thom JM, Ardigo LP, et al. Metabolic cost, mechanical work, and efficiency 5
during walking in young and older men. Acta Physiol (Oxf) 2006;186(2):127-39 doi: 6
10.1111/j.1748-1716.2006.01522.x[published Online First: Epub Date]|. 7
21. Berryman N, Gayda M, Nigam A, et al. Comparison of the metabolic energy cost of 8
overground and treadmill walking in older adults. European journal of applied physiology 9
2012;112(5):1613-20 doi: DOI 10.1007/s00421-011-2102-1[published Online First: Epub 10
Date]|. 11
22. BACR. Standards and Core Components for Cardiac Rehabilitation (2007). Secondary 12
Standards and Core Components for Cardiac Rehabilitation (2007) 2007. 13
http://www.bcs.com/documents/affiliates/bacr/BACR%20Standards%202007.pdf. 14
23. Robinson HJ, Samani NJ, Singh SJ. Can low risk cardiac patients be 'fast tracked' to Phase 15
IV community exercise schemes for cardiac rehabilitation? A randomised controlled trial. 16
Int J Cardiol 2011;146(2):159-63 doi: DOI 10.1016/j.ijcard.2009.06.027[published Online 17
First: Epub Date]|. 18
24. Almodhy MY, Sandercock GR, Richards L. Changes in cardiorespiratory fitness in patients 19
receiving supervised outpatient cardiac rehabilitation either once or twice a week. Int J 20
Cardiol 2012 doi: 10.1016/j.ijcard.2012.06.071[published Online First: Epub Date]|. 21
Page 15 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
16
25. Sharma A, McLeod AA. Cardiac rehabilitation after coronary artery bypass graft surgery: its 1
effect on ishcaemia, functional capacity and a mulitvariate index of prognosis. Coronary 2
Health Care 2001 5:189-93 3
26. Sandercock G, Hurtado V, Cardoso F. Changes in cardiorespiratory fitness in cardiac 4
rehabilitation patients: A meta-analysis. Int J Cardiol 2011 doi: 5
10.1016/j.ijcard.2011.11.068[published Online First: Epub Date]|. 6
7
8
9
Page 16 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
17
Table 1. 1 2 3 ISWT
Protocol
Level
Walking
Speed
(m·s-1
)
ACSM
Predicted
METs
Published
ISWT
METs
Recorded METs:
Treadmill Walking.
Mean (Range)
Recorded METs:
Shuttle Walking.
Mean (Range)
1 0.50 1.9 3.0 2.3 (1.6-2.6) 2.0 (1.6-2.2)
2 0.67 2.1 3.7 3.3 (2.8-4.0) 2.7 (2.5-3.1)
3 0.84 2.4 4.4 3.6 (3.1-4.3) 3.1 (2.8-3.3)
4 1.01 2.7 5.1 3.8 (3.2-4.6) 3.6 (3.2-3.8)
5 1.18 3.0 5.9 4.0 (3.6-4.7) 4.0 (3.6-4.6)
6 1.35 3.3 6.6 4.4 (4.3-5.9) 4.4 (4.0-4.9)
7 1.52 3.6 7.3 5.0 (4.6-6.2) 5.3 (4.8-5.6)
8 1.69 3.9 8.0 5.5 (5.0-6.7) 6.1 (5.7-6.6)
9 1.86 4.2 8.7 -- --
10 2.03 4.5/7.9** 9.4 -- --
11 2.20 4.8/8.5** 10.2 -- --
12 2.37 5.1/9.1** 10.9 -- --
4 5
6
7
8
9
10
11
12
13
14
Page 17 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
18
Figure legends 1
Figure 1 The oxygen uptake of treadmill-walking (black line) and shuttle-walking (grey line) at 2
each of the seven stages; * = treadmill-walking different from shuttle walking, p < 0.05. 3
4
Figure 2 Metabolic power above rest (PN) of treadmill-walking (black line) and shuttle-walking 5
(grey line) at each of the seven stages, * = treadmill-walking different from shuttle walking, p < 6
0.05. 7
8
Figure 3 Energy cost above rest (CN) per meter distance of treadmill-walking (black line) and 9
shuttle-walking (grey line) at each of the seven stages, * = treadmill-walking different from 10
shuttle-walking, p < 0.05. 11
12
Table 1. Comparison of predicted values, published values and measured metabolic cost (METs) 13
of the incremental shuttle-walking test. 14
15
Legend: MET – Metabolic Equivalent (calculated as: gross [ml.kg
-1.s
-1]/3.5) ISWT – 16
Incremental Shuttle Walking Test, ACSM – American College of Sports Medicine [6]. Published 17
ISWT METs in cardiac patients from Woolf-May & Ferrett [4]. 18
*n=7 subjects only. Predicted METs calculated using formula for walking or jogging** from 19
ACSM [6] 20
21
&VO2
Page 18 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
1
Pilot investigation of the oxygen demands and metabolic cost of incremental shuttle-1
walking and treadmill-walking in patients with cardiovascular disease. 2
3
Almodhy, M1., Beneke, R
2., Cardoso, F
1., Taylor, MJD.
1 Sandercock, GRH
1 * 4
5
*Corresponding Author 6
1Centre for Sports & Exercise Sciences, School of Biological Sciences, University of Essex 7
Colchester, UK CO4 3SQ 8
2Institut für Sportwissenschaft und Motologie Medezin, Philipps-Universität Marburg, 9
Germany. 10
11
12
13
Oxygen demands of shuttle-walking and treadmill walking 14
Key words: Cardiovascular disease; metabolism; cardiorespiratory fitness; pilot study 15
Word Count: 2498 16
17
Page 19 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
2
Abstract 1
Objective: To determine if the metabolic cost of the incremental shuttle-walking test protocol is 2
the same as treadmill-walking or predicted values walking speed equations. 3
Setting: Primary care (Community-based cardiac rehabilitation) 4
Participants: Eight caucasian cardiac rehabilitation patients (7 males) with a mean age 67±5.2 5
years. 6
Primary and secondary outcome measures: Oxygen consumption, metabolic power and 7
energy cost of walking during treadmill- and shuttle-walking performed in a balanced order with 8
one week between trials. 9
Results: Average overall energy cost per meter was higher during treadmill-walking (3.22 ± 10
0.55 J.kg
.m
-1) than during shuttle-walking (3.00 ± 0.41 J
.kg
.m
-1). There were significant post hoc 11
effects at 0.67 m.s
-1 (p < 0.004) and 0.84 m
.s
-1 (p < 0.001), where the energy cost of treadmill-12
walking was significantly higher than that of shuttle-walking. This pattern was reversed at 13
walking speeds 1.52 m·s-1
(p < 0.042) and 1.69 m·s-1
(p < 0.007) where shuttle-walking had a 14
greater energy cost per meter than treadmill-walking. At all walking speeds, the energy cost of 15
shuttle-walking was higher than predicted using the American College of Sports Medicine 16
walking equations. 17
Conclusion: The energetic demands of shuttle-walking were fundamentally different from that 18
of treadmill walking and should not be directly compared. We warn against estimating the 19
metabolic cost of the incremental shuttle-walking test using the current walking-speed equations. 20
21
22
23
Page 20 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
3
1
ARTICLE SUMMARY – Strengths and limitations of this study. 2
We provide the first direct comparison of the metabolic cost of shuttle-walking and treadmill-3
walking in cardiac patients. 4
Our data suggest metabolic demands of these exercise modalities appear fundamentally different 5
We suggest suggesting current methods to estimate metabolic cost of shuttle-walking are flawed 6
We warn against risk stratification of cardiac patients based on estimated oxygen costs using the 7
American College of Sports Medicine walking equations during the incremental shuttle-walking 8
test. 9
The sample size limits generalisability particularly in female patients who are not represented at 10
higher walking speeds – a larger study of metabolic cost of the incremental shuttle-walking test 11
is warranted. 12
13
Page 21 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
4
INTRODUCTION 1
Since its conception as an alternative to incremental treadmill testing of chronic obstructive 2
pulmonary disease (COPD) patients, the incremental shuttle-walking test (ISWT) has gained 3
popularity as an estimate of functional capacity in numerous clinical populations. The ISWT 4
appears adequately reliable,[1] and is sensitive to changes in functional capacity.[2 3] However, 5
the ISWT’s validity as an estimate of cardiovascular fitness is only moderate[3] and the use of 6
the test to estimate oxygen consumption exercise capacity in Metabolic Equivalents (METs) is 7
questionable.[4] 8
Woolf-May et al. [4] reported acceptable agreement between the energy cost of treadmill 9
walking and the ISWT in healthy volunteers using linear regression analyses, but did not assess 10
this relationship in cardiac patients. The authors reported higher energy demands of shuttle-11
walking in cardiac patients compared with healthy controls. They suggested this may be due to 12
poorer walking economy in the former, they did not report walking economy during ISWT or 13
make comparisons between shuttle- and treadmill-walking economy. 14
Treadmill and shuttle-walking tests are routinely used to assess cardiovascular disease patients 15
and we have previously reported discrete values for change in fitness measured using these 16
tests.[5] Prior to undertaking a proposed multicentre study to identify predictors of change in 17
cardiorespiratory fitness due to cardiac rehabilitation we performed the present pilot study. We 18
examined whether there were differences in the metabolic demands and energy cost of treadmill 19
and shuttle walking in cardiac rehabilitation patients in order to determine whether we could 20
combine data from these tests in our multicentre study. We also compared metabolic cost of the 21
ISWT with values predicted from treadmill-walking equations[6] and published estimates [4 7]. 22
23
Page 22 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
5
METHODS 1
Participants (n=8; 7 males; 67±5.2 years: 86.6 ±10.1 kg) were stable cardiac patients attending 2
community-based rehabilitation following elective cardiac revascularization. The study was 3
approved by the Faculty ethics board at The University of Essex. All patients gave written, 4
informed consent. 5
Equipment 6
The ISWT was performed on a non-slip floor using two cones placed 9 m apart and a portable 7
CD player. The treadmill test was performed on a motorised treadmill (Quaser, HP Cosmos, 8
Nussdorf, Germany). During both tests a portable gas analyser (K4b2 Mobile Breath by Breath 9
Metabolic System, COSMED Pulmonary Function Equipment, Rome, Italy) was used to record 10
expired gas collected via a face and nose mask (Hans Rudolph, Shawnee, Kansas. US). This was 11
calibrated using gases of a known concentration and a syringe before each test. 12
Protocol 13
Patients completed the ISWT and the treadmill test in a balanced order with one week between 14
trials. The ISWT was performed in accordance with national recommendations for cardiac 15
patients[8]. Briefly, the 12 stage protocol starts at a walking speed of 0.5 m·s-1
(1.12 mph) and 16
increases by 0.17 m·s-1
(0.38 mph) each minute. An identical incremental protocol was 17
programmed into the treadmill. Patients were accustomed to treadmill walking but received a 18
brief period of familiarization in which they were required to walk without holding the treadmill 19
handles before the ISWT protocol was also performed. 20
Calculation of metabolic power and energy cost of walking 21
We assumed a standard resting metabolic rate of 4 ml.kg
-1.min
-1 based on reference standards[9]. 22
Metabolic power was then calculated via indirect calorimetry from and above rest 23 &VO2 &VCO2
Page 23 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
6
and from body mass: metabolic power [W.kg
-1] = ( - rest) [ml
.kg
-1.s
-1]
. respiratory 1
exchange ratio adjusted caloric equivalent [J.ml
-1]. [10 11] To analyse the relationship between 2
speed and metabolic power of walking the metabolic power was predicted as a quadratic function 3
of speed: metabolic power = a + b v² [10 12-16]. The energy cost of walking per metre distance 4
was calculated by: energy cost [J.kg
-1.m
-1] = metabolic power [W
.kg
-1]
. speed
-1 [m
.s
-1]. [10 11 17] 5
6
Statistical Analyses 7
Descriptive results are presented as mean ± SD. A test modality-by-walking speed analysis of 8
variance (ANOVA) with shuttle vs. treadmill walking as within-subjects factor and walking 9
speed as the between-subjects factor was performed. Significant interactions and main effects 10
were further analysed using one-way ANOVA and paired samples t-tests as appropriate. Based 11
on the classical descriptions of walking energy cost [18-20], non-linear regression models were 12
chosen to identify significant interrelationships between metabolic power, energy cost per meter 13
and walking speed, respectively. All analyses were completed using SPSS v19.0 (SPSS Inc. and 14
IBM Company. Chicago, IL) and statistical significance was defined as p < 0.05. 15
16
RESULTS 17
Figure 1 shows the oxygen uptake at each of seven stages completed by at least 7 patients. There 18
was a significant main effect for walking speed on oxygen uptake and a significant interaction 19
between treadmill-walking and shuttle-walking on the ground. Oxygen uptake was higher in 20
treadmill-walking than shuttle-walking at 0.67 m·s-1
(p=0.006; n=8) and 0.84 m·s-1
(p=0.003; n 21
= 8) but the significantly steeper increases in oxygen demand during shuttle-walking meant the 22
opposite was true at 1.69 m·s-1
(p < 0.006; n = 7). 23
&VO2 &VO2
Page 24 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
7
**Insert Fig 1 here** 1
2
Figure 2 shows the metabolic power of treadmill-walking and shuttle-walking. There was a main 3
effect for walking velocity on metabolic power during both treadmill and shuttle walking (p < 4
0.05). The different effects of walking modality on metabolic power were more pronounced if 5
power was predicted as a function of walking speed with power-treadmill walking = 2.028 + 6
1.115 v² and power-shuttle walking on the ground = 1.126 + 1.665 v² where 99% of the variance 7
of power were explained by the quadratic curve fits in both modalities (both p < 0.001). The 8
difference in response to each modality was indicated by a significant interaction between 9
modality and speed. There were significantly higher metabolic power requirements for treadmill 10
walking at 0.67 m·s-1
(p < 0.004; n = 8) and 0.84 m·s-1
(p < 0.001; n = 8). Due to the steeper 11
increase observed in shuttle-walking the metabolic power was significantly higher at 1.52 m⋅s-1
12
(p=0.042; n=7) and 1.69 m⋅s-1
(p=0.007; n=7) compared with treadmill-walking. 13
14
**Insert Figure 2 here** 15
There were significant main effects for both modality and velocity in relative energy cost (per 16
metre) of walking again well described as a function of speed by the above approximated 17
parameters for both walking modalities (energy cost-treadmill walking = 2.028 / v + 1.115 v and 18
energy cost shuttle-walking on the ground = 1.126 / v + 1.665 v; both p < 0.001). Average 19
overall energy cost per meter (kg·m-1
) was higher during treadmill-walking (3.22 ± 0.55 J·kg·m-
20
1) than during shuttle-walking (3.00 ± 0.41 J·kg·m
-1). There were significant post hoc effects at 21
0.67 m·s-1
(p < 0.004; n=8) and 0.84 m·s-1
(p < 0.001; n=8), where the energy cost of treadmill-22
walking was higher than that of shuttle-walking. Again, this pattern was reversed at higher 23
Page 25 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
8
walking speeds of 1.52 m·s-1
(p < 0.042) and 1.69 m·s-1
(p < 0.007) where shuttle-walking had a 1
greater energy cost per meter (for the n=7 patients achieving this level) than treadmill-walking. 2
3
**Insert Figure 3 here** 4
DISCUSSION 5
This is the first comparative investigation of the metabolic demands and energy cost per meter 6
walking of incremental treadmill-walking and shuttle-walking in cardiac rehabilitation patients. 7
We found differences in the oxygen requirements and energy cost of shuttle and treadmill 8
walking large enough to suggest results from these exercise modalities should not be pooled in 9
any future analyses. 10
Economy and energy-requirements recorded during level 1 are difficult to interpret 11
as they are most affected by oxygen kinetics and patients’ unusually long stance phase during 12
their gait cycle at this very slow walking speed and were excluded from our figures. The change 13
in walking energy cost per metre on the treadmill show the expected pattern. Slow speeds are 14
associated with higher cost per metre which decreases as optimal (comfortable) walking speed 15
approaches. Continuing to increase walking speed above this pace requires a greater cost per 16
metre. In contrast to this, the energy cost per metre in shuttle-walking decreases only very little 17
and only following the first (very slow) walking pace in the initial stage. The energy cost then 18
increases stage-by-stage throughout the protocol. The cost is only consistent between treadmill 19
and shuttle-walking between 1.2 - 1.4 m·s-1
(close to comfortable walking speed) and the 20
increase in energy requirements is much greater in shuttle walking. Based on these pilot data, we 21
intend to report cardiorespiratory fitness values separately according to test modality and 22
recommend this practice to others. 23
Page 26 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
9
The classical description of the energy cost during locomotion is of a U-shaped relationship[18] 1
– as speed increases or decreases from the optimal (1.11-1.3 m.s
-1 [18-20]) the energy cost of 2
locomotion increases. For the treadmill protocol our data support this relationship. At slow 3
speeds (0.6-0.8 m.s
-1) energy cost was greater than at optimal speeds (1.2-1.4 m
.s
-1). As walking 4
speed increased (1.6-1.8 m.s
-1) the energy cost again began to increase. This is comparable to 5
Berryman et al[21] who reported a similar energy cost pattern for their subjects (healthy elderly 6
aged 68.9 ± 4.6 yrs) when walking on a treadmill at speeds ranging from 0.67 – 1.56 m.s
-1 and 7
the optimal walking speed was 1.33 m.s
-1. Furthermore our results also suggest that at lower 8
speeds (0.50-0.84 m.s
-1), the energy cost of walking on a treadmill is greater than on the ground. 9
Berryman et al[21] also showed that there was greater energy cost of treadmill-walking 10
compared to ground walking at all the speeds they tested. The explanation for the increased 11
energy cost for treadmill-walking may be due to a greater need for stabilisation, via muscular 12
contraction, than when walking on the ground[21]. 13
Conversely, the oxygen requirement of shuttle-walking are comparatively higher 14
from level 7 (1.52 m·s-1
) onwards than for treadmill walking at the same speed. The 15
requirements are also much higher (18 ml·kg-1
·min-1
) than the value predicted by the ACSM 16
walking speed equations [6] (12.6 ml·kg-1
·min-1
) which are used to estimate cardiorespiratory 17
fitness from ISWT performance[4]. In addition to differences in oxygen requirements of ground 18
and treadmill walking, shuttle-walking may have a higher cost due to repeated acceleration, 19
deceleration phases or the negotiation of turns[7]. We propose, therefore that any clinical cut-20
offs for walking tests should be developed using the same testing modality as that for which they 21
are proposed for use in (i.e. treadmill or ground). 22
Page 27 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
10
Cardiac patients’ exercise capacity is commonly expressed as metabolic equivalents 1
(METs). We calculated metabolic cost in METs (Gross [ml.kg
-1.min
-1] /3.5) and compared 2
MET values at all ISWT stages with those reported previously[4] and the ACSM-predicted 3
values (Table 1). It should be noted that the values predicted using the ACSM walking equations 4
by Woolf-May and Ferret[4] are incorrect. The MET values they reported in cardiac patients are 5
almost double the predicted values using the ACSM equations and much higher than those 6
reported presently. Woolf-May and Ferrett’s [4] MET values further appear anomalous as they 7
are more than double that age-matched controls and significantly higher than recently-reported 8
values in cardiac patients during the ISWT [7]. These latter values [7] do, however, broadly 9
agree with those reported presently. 10
**Insert Table 1 here** 11
Current recommendations suggest patients be classed as high risk if their exercise capacity is <5 12
METs. Failure to reach this criterion standard may lead to prevent patients from entering 13
community-based rehabilitation [22]. Woolf-May and Ferret’s [4] suggestion that ISWT level 4 14
elicits a 5 MET energy cost in cardiac patients is inconsistent with more-recent data from Woolf-15
May & Meadows [7] and those of the present study; both of which suggest the 5 MET threshold 16
is nearer Level 7 or 8. 17
18
Fitter patients can be successfully ‘fast tracked’ to community rehabilitation saving capacity and 19
money to the health providers [23]. However, where exactly in the ISWT protocol this threshold 20
occurs should be determined in a larger, more representative cohort of cardiac patients. 21
Beyond level 7 (1.52 m·s-1
, 3.8 mph) shuttle-walking incurred an additional extra energy cost 22
compared with treadmill walking which may make it difficult to show small improvements in 23
&VO2
Page 28 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
11
functional capacity if reported as estimated MET values. The exercise capacity of cardiac 1
patients measured before outpatient rehabilitation tends to be lower when estimated from ISWT 2
[24] than when standard treadmill protocols are used. [25 26] 3
4
Study limitations and conclusions 5
Along with sample size, this study is also limited due to including predominantly male patients 6
and indeed only male’s data at the highest walking speeds. The comparison of treadmill and 7
shuttle-walking may have been improved by increasing treadmill gradient as is common-8
practice. We omitted to do this for comparability with previous work. [4 6] The accuracy of 9
energy costs calculations would also be improved by including a resting metabolic measure pre-10
exercise instead of an assumed value of 4 ml.kg
-1.m
-1. [9] 11
12
In conclusion, the ISWT may have clinical utility as measure of functional capacity to use in 13
exercise prescription and patient monitoring, but we question its use as an estimate of 14
cardiorespiratory fitness in cardiac patients. Importantly, the ACSM walking equations grossly 15
underestimate the actual energy cost of shuttle-walking and should not be used in research or 16
clinical practice. Our comparison using metabolic equivalents (METs) also reveals that some 17
published [4] estimates of the ISWT’s energy cost in cardiac patients appear erroneously high. 18
Given these two shortcomings, we strongly warn against clinical decision-making or patient risk 19
stratification based on achieving the 5 MET threshold estimated using the ISWT. We 20
recommend a more accurate assessment of the ISWT’s energy cost be performed in a larger, 21
more generalisable sample of cardiac patients. 22
23
Page 29 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
12
Contributorship statement: Sandercock, Beneke and Taylor devised the experimental design. 1
Almodhy & Cardoso collected and analysed the data. Beneke performed the metabolic modeling 2
and advanced statistical analysis. Sandercock and Taylor drafted the manuscript. Beneke, 3
Almodhy and Cardoso revised the manuscript. All authors contributed to the final preparation 4
and drafting of the manuscript. 5
Competing interests None. 6
Funding. None. 7
Data sharing. No additional data available 8
9
10
11
12
13
14
15
16
17
18
19
20
21
Page 30 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
13
References 1
2
3
1. Pepera G, McAllister J, Sandercock G. Long-term reliability of the incremental shuttle 4
walking test in clinically stable cardiovascular disease patients. Physiotherapy 5
2010;96(3):222-7 6
2. Woolf-May K, Bird S. Physical activity levels during phase IV cardiac rehabilitation in a 7
group of male myocardial infarction patients. Br J Sports Med 2005;39(3):e12; discussion 8
e12 9
3. Fowler SJ, Singh S. Reproducibility and validity of the incremental shuttle walking test in 10
patients following coronary artery bypass surgery. Physiotherapy 2005;91:22-27 11
4. Woolf-May K, Ferrett D. Metabolic equivalents during the 10-m shuttle walking test for post-12
myocardial infarction patients. Br J Sports Med 2008;42(1):36-41; discussion 41 13
5. Sandercock GR, Cardoso F, Almodhy M, Pepera G. Cardiorespiratory fitness changes in 14
patients receiving comprehensive outpatient cardiac rehabilitation in the UK: a multicentre 15
study. Heart 2012 doi: 10.1136/heartjnl-2012-303055[published Online First: Epub Date]|. 16
6. ACSM. ACSM's guidelines for exercise testing and prescription. 8th ed. Philadelphia: 17
Lippincott Williams & Wilkins, 2010. 18
7. Woolf-May K, Meadows S. Exploring adaptations to the modified shuttle walking test. BMJ 19
Open 2013;3(5) doi: 10.1136/bmjopen-2013-002821[published Online First: Epub Date]|. 20
8. SIGN. SIGN 57 Cardiac Rehabilitation. A National Clinical Guideline. Edinburgh: Royal 21
College of Physicians, 2002. 22
Page 31 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
14
9. Wissenschaftliche Tabellen Geigy Teilband Körperflüssigkeiten. (Scientific Tables Geigy, 1
Volume Body Fluids.) In: Lentner C, ed. Giegy Scientific Tables. 7th ed. Basel: Ciba-Geigy, 2
1985:225-28. 3
10. Beneke R, Meyer K. Walking performance and economy in chronic heart failure patients pre 4
and post exercise training. Eur J Appl Physiol Occup Physiol 1997;75(3):246-51 5
11. di Prampero PE. The energy cost of human locomotion on land and in water. International 6
journal of sports medicine 1986;7(2):55-72 doi: 10.1055/s-2008-1025736[published Online 7
First: Epub Date]|. 8
12. Bobbert AC. Energy expenditure in level and grade walking. Journal of Applied Physiology 9
1960;15:1015-21 10
13. Ralston HJ. Energy-speed relation and optimal speed during level walking. Internationale 11
Zeitschrift fur angewandte Physiologie, einschliesslich Arbeitsphysiologie 1958;17(4):277-12
83 13
14. van der Walt WH, Wyndham CH. An equation for prediction of energy expenditure of 14
walking and running. J Appl Physiol 1973;34(5):559-63 15
15. Zaciorskij VM. Biomechanische Grundlagen der Ausdauer. Berlin: Sportverlag Berlin, 1987. 16
16. Zarrugh MY, Todd FN, Ralston HJ. Optimization of energy expenditure during level 17
walking. European journal of applied physiology and occupational physiology 18
1974;33(4):293-306 19
17. Brueckner JC, Atchou G, Capelli C, et al. The energy cost of running increases with the 20
distance covered. European journal of applied physiology and occupational physiology 21
1991;62(6):385-9 22
Page 32 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
15
18. Workman JM, Armstrong BW. Metabolic cost of walking: equation and model. J Appl 1
Physiol 1986;61(4):1369-74 2
19. Bunc V, Dlouha R. Energy cost of treadmill walking. The Journal of sports medicine and 3
physical fitness 1997;37(2):103-9 4
20. Mian OS, Thom JM, Ardigo LP, Narici MV, Minetti AE. Metabolic cost, mechanical work, 5
and efficiency during walking in young and older men. Acta Physiol (Oxf) 6
2006;186(2):127-39 doi: 10.1111/j.1748-1716.2006.01522.x[published Online First: Epub 7
Date]|. 8
21. Berryman N, Gayda M, Nigam A, Juneau M, Bherer L, Bosquet L. Comparison of the 9
metabolic energy cost of overground and treadmill walking in older adults. European journal 10
of applied physiology 2012;112(5):1613-20 doi: DOI 10.1007/s00421-011-2102-11
1[published Online First: Epub Date]|. 12
22. BACR. Standards and Core Components for Cardiac Rehabilitation (2007). Secondary 13
Standards and Core Components for Cardiac Rehabilitation (2007) 2007. 14
http://www.bcs.com/documents/affiliates/bacr/BACR%20Standards%202007.pdf. 15
23. Robinson HJ, Samani NJ, Singh SJ. Can low risk cardiac patients be 'fast tracked' to Phase 16
IV community exercise schemes for cardiac rehabilitation? A randomised controlled trial. 17
Int J Cardiol 2011;146(2):159-63 doi: DOI 10.1016/j.ijcard.2009.06.027[published Online 18
First: Epub Date]|. 19
24. Almodhy MY, Sandercock GR, Richards L. Changes in cardiorespiratory fitness in patients 20
receiving supervised outpatient cardiac rehabilitation either once or twice a week. Int J 21
Cardiol 2012 doi: 10.1016/j.ijcard.2012.06.071[published Online First: Epub Date]|. 22
Page 33 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
16
25. Sharma A, McLeod AA. Cardiac rehabilitation after coronary artery bypass graft surgery: its 1
effect on ishcaemia, functional capacity and a mulitvariate index of prognosis. Coronary 2
Health Care 2001 5:189-93 3
26. Sandercock G, Hurtado V, Cardoso F. Changes in cardiorespiratory fitness in cardiac 4
rehabilitation patients: A meta-analysis. Int J Cardiol 2011 doi: 5
10.1016/j.ijcard.2011.11.068[published Online First: Epub Date]|. 6
7
8
9
Page 34 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
17
Table 1. 1 2 3 ISWT
Protocol
Level
Walking
Speed
(m·s-1
)
ACSM
Predicted
METs
Published
ISWT
METs
Recorded METs:
Treadmill Walking.
Mean (Range)
Recorded METs:
Shuttle Walking.
Mean (Range)
1 0.50 1.9 3.0 2.3 (1.6-2.6) 2.0 (1.6-2.2)
2 0.67 2.1 3.7 3.3 (2.8-4.0) 2.7 (2.5-3.1)
3 0.84 2.4 4.4 3.6 (3.1-4.3) 3.1 (2.8-3.3)
4 1.01 2.7 5.1 3.8 (3.2-4.6) 3.6 (3.2-3.8)
5 1.18 3.0 5.9 4.0 (3.6-4.7) 4.0 (3.6-4.6)
6 1.35 3.3 6.6 4.4 (4.3-5.9) 4.4 (4.0-4.9)
7 1.52 3.6 7.3 5.0 (4.6-6.2) 5.3 (4.8-5.6)
8 1.69 3.9 8.0 5.5 (5.0-6.7) 6.1 (5.7-6.6)
9 1.86 4.2 8.7 -- --
10 2.03 4.5/7.9** 9.4 -- --
11 2.20 4.8/8.5** 10.2 -- --
12 2.37 5.1/9.1** 10.9 -- --
4 5
6
7
8
9
10
11
12
13
14
Page 35 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
18
Figure legends 1
Figure 1 The oxygen uptake of treadmill-walking (black line) and shuttle-walking (grey line) at 2
each of the seven stages; * = treadmill-walking different from shuttle walking, p < 0.05. 3
4
Figure 2 Metabolic power above rest (PN) of treadmill-walking (black line) and shuttle-walking 5
(grey line) at each of the seven stages, * = treadmill-walking different from shuttle walking, p < 6
0.05. 7
8
Figure 3 Energy cost above rest (CN) per meter distance of treadmill-walking (black line) and 9
shuttle-walking (grey line) at each of the seven stages, * = treadmill-walking different from 10
shuttle-walking, p < 0.05. 11
12
Table 1. Comparison of predicted values, published values and measured metabolic cost (METs) 13
of the incremental shuttle-walking test. 14
15
Legend: MET – Metabolic Equivalent (calculated as: gross [ml.kg
-1.s
-1]/3.5) ISWT – 16
Incremental Shuttle Walking Test, ACSM – American College of Sports Medicine [6]. Published 17
ISWT METs in cardiac patients from Woolf-May & Ferrett [4]. 18
*n=7 subjects only. Predicted METs calculated using formula for walking or jogging** from 19
ACSM [6] 20
21
&VO2
Page 36 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
90x70mm (300 x 300 DPI)
Page 37 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
Figure 2 Metabolic power above rest (PN) of treadmill-walking (black line) and shuttle-walking (grey line) at each of the seven stages, * = treadmill-walking different from shuttle walking, p < 0.05.
90x71mm (300 x 300 DPI)
Page 38 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
Figure 2 Metabolic power above rest (PN) of treadmill-walking (black line) and shuttle-walking (grey line) at each of the seven stages, * = treadmill-walking different from shuttle walking, p < 0.05.
90x71mm (300 x 300 DPI)
Page 39 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from
For peer review only
STARD checklist for reporting of studies of diagnostic accuracy
(version January 2003)
Section and Topic Item
#
On page #
TITLE/ABSTRACT/
KEYWORDS
1 Identify the article as a study of diagnostic accuracy (recommend MeSH
heading 'sensitivity and specificity').
2
INTRODUCTION 2 State the research questions or study aims, such as estimating diagnostic
accuracy or comparing accuracy between tests or across participant
groups.
3
METHODS
Participants 3 The study population: The inclusion and exclusion criteria, setting and
locations where data were collected.
5
4 Participant recruitment: Was recruitment based on presenting symptoms,
results from previous tests, or the fact that the participants had received
the index tests or the reference standard?
5
5 Participant sampling: Was the study population a consecutive series of
participants defined by the selection criteria in item 3 and 4? If not,
specify how participants were further selected.
5
6 Data collection: Was data collection planned before the index test and
reference standard were performed (prospective study) or after
(retrospective study)?
5
Test methods 7 The reference standard and its rationale. 5
8 Technical specifications of material and methods involved including how
and when measurements were taken, and/or cite references for index
tests and reference standard.
5
9 Definition of and rationale for the units, cut-offs and/or categories of the
results of the index tests and the reference standard.
5, 9, 10
10 The number, training and expertise of the persons executing and reading
the index tests and the reference standard.
5
11 Whether or not the readers of the index tests and reference standard
were blind (masked) to the results of the other test and describe any
other clinical information available to the readers.
5
Statistical methods 12 Methods for calculating or comparing measures of diagnostic accuracy,
and the statistical methods used to quantify uncertainty (e.g. 95%
confidence intervals).
6,7,10
13 Methods for calculating test reproducibility, if done. -
RESULTS
Participants 14 When study was performed, including beginning and end dates of
recruitment.
6
15 Clinical and demographic characteristics of the study population (at least
information on age, gender, spectrum of presenting symptoms).
6
16 The number of participants satisfying the criteria for inclusion who did or
did not undergo the index tests and/or the reference standard; describe
why participants failed to undergo either test (a flow diagram is strongly
recommended).
6
Test results 17 Time-interval between the index tests and the reference standard, and
any treatment administered in between.
5
18 Distribution of severity of disease (define criteria) in those with the target
condition; other diagnoses in participants without the target condition.
5,10
19 A cross tabulation of the results of the index tests (including
indeterminate and missing results) by the results of the reference
standard; for continuous results, the distribution of the test results by the
results of the reference standard.
Table 1,
P10
20 Any adverse events from performing the index tests or the reference
standard.
6
Estimates 21 Estimates of diagnostic accuracy and measures of statistical uncertainty
(e.g. 95% confidence intervals).
9,10,11
22 How indeterminate results, missing data and outliers of the index tests
were handled.
10
23 Estimates of variability of diagnostic accuracy between subgroups of
participants, readers or centers, if done.
-
24 Estimates of test reproducibility, if done. -
DISCUSSION 25 Discuss the clinical applicability of the study findings. 10,11,12
Page 40 of 40
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 7, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2014-005216 on 16 Septem
ber 2014. Dow
nloaded from