Upload
dangphuc
View
216
Download
0
Embed Size (px)
Citation preview
Title: Energy expenditure derived from micro-technology is not suitable for assessing
internal load in collision-based activities
Submission type: Brief report
Authors: Jamie Highton, Thomas Mullen, Jonathan Norris, Chelsea Oxendale, Craig
Twist*
Affiliation
Department of Sport and Exercise SciencesUniversity of ChesterParkgate RoadChesterCH1 4BJ
*Author for correspondence: Craig Twist ([email protected])
Preferred running head: Metabolic power in contact sports
Abstract word count: 154
Text word count: 1741
Number of figures: 2 Number of tables: 1
1
1
2
3
4
5
6
7
8
9
101112131415
16
17
18
19
20
21
22
23
24
Abstract
This aim of this study was to examine the validity of energy expenditure derived from
micro-technology when measured during a repeated effort rugby protocol. Sixteen
male rugby players completed a repeated effort protocol comprising 3 sets of 6
collisions during which movement activity and energy expenditure (EEGPS) were
measured using micro-technology. In addition, energy expenditure was also estimated
from open circuit spirometry (EEVO2). Whilst related (r = 0.63, 90%CI 0.08-0.89),
there was a systematic underestimation of energy expenditure during the protocol (-
5.94 ± 0.67 kcalmin-1) for EEGPS (7.2 ± 1.0 kcalmin-1) compared to EEVO2 (13.2 ±
2.3 kcalmin-1). High-speed running distance (r = 0.50, 95%CI -0.66-0.84) was
related to EEVO2, while Player Load was not (r = 0.37, 95%CI -0.81-0.68). Whilst
metabolic power might provide a different measure of external load than other
typically used micro-technology metrics (e.g. high-speed running, Player Load), it
underestimates energy expenditure during intermittent team sports that involve
collisions.
Key words: rugby, tackle, training load, GPS, accelerometry
2
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
Introduction
Global positioning systems (GPS) with in-built accelerometers are regularly used in
contact sport training1,2 and competition1,3 to quantify internal and external player
load. A metric recently introduced to these devices is metabolic power (i.e. energy
expenditure), which estimates the metabolic ‘internal’ cost of activities from players’
‘external’ movements.4 Metabolic power values of ~9-11 W·kg-1 have recently been
employed to estimate energy expenditures of ~24-45 kJkg-1 during elite rugby
league5 and Australian football match-play.6 In these studies, metabolic power was
proposed as a more appropriate method than traditional speed-dependant thresholds
on the basis that the latter might underestimate the true metabolic demands associated
with accelerated running. Some authors also suggest this metric might be useful for
informing players’ nutritional programming.1,5 However, the validity of metabolic
power to quantify the internal load of team sport activity has recently been questioned
due to a reported underestimation of energy expenditure owing to an inability to
detect non-ambulatory related activities.7 What remains unknown, however, is the
utility of this GPS-derived metric in collision-based sports, such as rugby. This is
particularly pertinent given that when physical contact is combined with intermittent
running, increases in heart rate, rating of perceived exertion and blood lactate
concentration suggest a greater metabolic response and anaerobic contribution when
compared to intermittent running alone. 8 Accordingly, this study sought to establish
the validity of energy expenditure derived from micro-technology when measured
during a repeated effort rugby protocol.
3
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
Methods
Participants and design
After ethics approval, sixteen university rugby players (age 23.8 ± 4.8 y; stature 1.80
± 0.05 m, body mass 84.5 ± 8.6 kg) volunteered to participate in this study.
Participants completed a single visit comprising a repeated effort protocol during
which movement activity and energy expenditure were measured.
The protocol comprised three sets of six ‘efforts’ with a strictly enforced 60 s standing
recovery between each set. One effort involved an 8 m run at 4 m·s -1 to make a
collision with a standard 30 kg tackle bag (Rhino Products, Leeds, UK). The
participant took the bag to the ground after collision, paused for 2 s, stood to re-
position the bag and then ran backwards at 2.5 m·s-1 to the start position. The protocol
was performed on a 3G artificial grass surface with running speeds controlled by an
audio signal emitted from a CD player. Environmental conditions were 22.8 ± 0.4ºC
and 37.3 ± 1.0% relative humidity.
Direct measures of internal load
Expired air was collected for each work (n = 3) and passive recovery (n = 3) period of
the protocol using a facemask connected to a Douglas bag (Hans Rudolph, UK). A
researcher ran alongside the participant carrying the Douglas bag, ensuring it
remained connected and did not disrupt running or tackling technique. Expired air
was analysed immediately for oxygen and carbon dioxide fractions using a previously
calibrated gas meter (Servomex 5200, Crowborough, UK) and gas volume using a dry
gas meter (Harvard apparatus, Kent, UK). These data were then used to calculate
oxygen uptake (VO2), RER and energy expenditure (EEVO2)10 for each work and rest
4
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
period. These values were then averaged over the whole protocol. The VO2 and
associated energy expenditure whilst standing was measured in all participants (2.9 ±
0.6 kcal) before exercise and subtracted from exercise VO2 to exclude resting
metabolic rate from calculation of internal load. Heart rate was measured using a
monitor (Polar Electro Oy, Kempele, Finland) strapped to the participant’s chest, with
data stored on the GPS unit (see below). Blood was collected from a fingertip
capillary sample 5 min after completing the protocol and analysed for lactate
concentration (Lactate Pro, Arkray, Japan).
GPS and accelerometer measures
A 10 Hz GPS device fitted with a 100 Hz tri-axial accelerometer, gyroscope and
magnetometer (Optimeye S5, Catapult Innovations, Melbourne, Australia) was
securely positioned between the participant’s scapulae using a custom-made vest. The
available satellites and HDOP were 16.3 ± 0.9 (range 15 - 18) and 0.7 ± 0.1 AU
(range 0.5 - 1.2), respectively. Data were later downloaded to a laptop and analysed
(Sprint, Version 5.1, Catapult Sports, VIC, Australia) to calculate distance (m), high-
speed running (>14 kmh-1) distance (m), Player Load (AU) and energy expenditure
(EEGPS; kcalmin-1) for each work and rest period.
Statistical analysis
All data are presented as mean ± SD. Data were checked for normality using the
Shapiro-Wilk statistic (P > 0.05). The difference between energy expenditure
determined via open circuit spirometry and GPS was assessed using a paired-samples
t-test. Agreement between measures of energy expenditure was assessed using 95%
limits of agreement (95%LoA), calculated as the mean difference between measures
5
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
(i.e. systematic bias) ± 1.96 x the SD of the differences between measures (i.e. the
‘random’ error in measurement). Pearson’s correlation coefficients with associated
95% confidence intervals (95%CI) were also performed to explore any relationship
between GPS-derived external load measurements (metabolic power, high-speed
running distance, Player Load) and the measurement of energy expenditure. To assess
the magnitude of the relationship, the following criteria were applied: < 0.1, trivial; >
0.1 – 0.3, small; > 0.3 – 0.5, moderate; > 0.5 – 0.7, large; > 0.7 – 0.9, very large; and
> 0.9 – 1.0, extremely large.9 In all instances the alpha was set to P < 0.05.
Results
Physiological responses to the exercise protocol are reported in Table 1. Mean energy
expenditure over the protocol according to open circuit spirometry and GPS were 13.2
± 2.3 and 7.2 ± 1.0 kcalmin-1, respectively (Figure 1). This amounted to a total
energy expenditure of 82.8 ± 14.1 kcal using open circuit spirometry and 47.4 ± 6.9
kcal using GPS. The 95%LoA indicated a systematic underestimation of energy
expenditure in EEGPS (-5.94 ± 0.67 kcalmin-1), although the measures were
moderately related (r = 0.63, 95%CI 0.08-0.89; Figure 2). A moderate relationship
was also observed between high-speed running distance and EEVO2 (r = 0.50, 95%CI -
0.66-0.84, P < 0.05), but was low between Player Load and EEVO2 (r = 0.37, 95%CI -
0.81-0.68, P > 0.05).
**********Insert figures and table approximately here**********
6
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
Discussion
This study has demonstrated that, despite sharing a moderate relationship, energy
expenditure measured using micro-technology during a repeated effort contact drill in
rugby players is systematically lower than that measured using open-circuit
spirometry. The potential magnitude of underestimation during this specific form of
exercise, according to our 95%LoA, is between 5.27 and 6.61 kcalmin-1.
Extrapolated to, for example, a duration associated with rugby match-play (~40 min),3
then the absolute underestimation of energy expenditure could be 210 to 264 kcal.
The acceptability of this level of agreement is somewhat dependent on the purpose of
measurement. However, to provide some relevant context, the previously estimated
energy expenditure during ~40 min of rugby league match is approximately 908
kcal.11 In this context, future studies should investigate the magnitude of agreement in
energy expenditure using GPS and open-circuit spirometry associated with exercise
more closely simulating the exercise duration of contact team sports.
We reaffirm findings previously reported in team sport athletes1,7 that energy
expenditure is underestimated when calculated using metabolic power from micro-
technology. Our 45% lower energy expenditure using GPS derived metabolic power
is in agreement with Buchheit et al.,7 who employed similar procedures to report a
~51% difference between methods during a soccer-specific circuit. This is noteworthy
because, unlike Buchheit and co-workers,7 our measurements appropriately accounted
for the energetic cost of standing in the direct measure of energy expenditure.12 The
differences between methods in this study are much higher than the 11%
disagreement observed by Walker et al.1 However, rather than the direct measures
using open circuit spirometry as used here, Walker and colleagues1 used
7
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
accelerometer based predictions of energy expenditure that are known to
underestimate true values.13
The underestimation of energy expenditure in the present study might be because the
GPS is unable to detect increased energy transfer during periods of rest associated
with intermittent exercise (i.e. excess post-exercise oxygen consumption; EPOC).7
Indeed, Figure 1 shows that the difference between energy expenditure measures is
almost entirely accounted for by an increased VO2 during the rest periods. However,
the ‘real-time’ agreement between methods was not statistically analysed here,
because of a ‘lag’ in pulmonary VO2 kinetics relative to muscle VO2 (20-35 s).12 As
such, data were analysed over the whole protocol. This also served to incorporate the
inclusion of VO2 measured during recovery, which could in part account for the
replenishment of resources depleted using anaerobic metabolism.13 We acknowledge
that anaerobic metabolism would not be accounted for with measurements of VO2
whilst exercising, but which would presumably be incorporated in the GPS
measurement of metabolic power given it describes the energy required to
resynthesize ATP required for work performed at that time.12 Indeed, the relatively
high blood lactate concentrations observed here (10.5 ± 3.6 mmoll-1) provide an
indirect indication that there was a significant contribution of anaerobic metabolism to
ATP resynthesis during the exercise protocol.
Another potential explanation for the underestimation of energy expenditure by the
GPS is an inability to detect energy expenditure associated with non-locomotor
exertion. Metabolic power is derived from measures of acceleration and deceleration
and their corresponding energy cost predicted from steady state incline running.4 Thus
8
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
the energy cost predicted from acceleration/deceleration associated with contact, or
indeed measurement of energy expenditure with a static exertion, cannot be
determined from GPS. Given that contacts, such as those experienced during certain
team sports, increase the physiological exertion associated with intermittent running,8
it is likely that energy expenditure is increased by this type of activity, yet it is not
incorporated adequately into a calculation of metabolic power. While measures of
instantaneous velocity by the 10 Hz GPS device have acceptable validity and
reliability,15 criterion velocity during accelerations is still underestimated by these
devices.15 That these GPS measures contribute to the calculation of metabolic power
might also explain the observed underestimation.
We observed moderate associations between open-circuit spirometry derived energy
expenditure and metabolic power (r = 0.63) that was similar in magnitude but greater
than high-speed running (r = 0.50). Our data support previous assertions that
metabolic power more adequately reflects the physiological load associated with
intermittent running than traditional speed zones.5 Based on the low, non-significant
relationship (r = 0.37) between Player Load and energy expenditure, we suggest
Player Load does not reflect the metabolic load associated with intermittent exercise
employing collisions. Finally, we conclude that energy expenditure is underestimated
from micro-technology and therefore should be applied with caution when attempting
to quantify training load during intermittent team sports that involve collisions (e.g.
rugby, Australian rules football).
9
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
Acknowledgements: The authors wish to thank Richard Bott and Simon Cushman
for their technical assistance during this study.
References
1. Walker EJ, McAinch AJ, Sweeting A, Aughey RJ. Inertial sensors to estimate the energy expenditure of team-sport athletes. J Sci Med Sports. 2015;19:177-181.
2. Weaving D, Marshall P, Earle K, Nevill A, Abt G. A. combination of internal and external training load measures explains the greatest proportion of variance in certain training modes in professional rugby league. Int J Sports Physiol Perf. 2015;9:905-912.
3. Oxendale C, Twist C, Daniels M, Highton J. The relationship between match-play characteristics of elite rugby league and indirect markers of muscle damage. Int J Sports Physiol Perf. 2015.
4. Osgnach C, Poser S, Bernardini R, Rinaldo R, di Prampero P. Energy cost and metabolic power in elite soccer: A new match analysis approach. Med Sci Sports Exerc. 2010;42:170-178.
5. Kempton T, Sirotic AC, Rampinini E, Coutts AJ. Metabolic power demands of rugby league match-play. Int J Sports Physiol Perf. 2015;10:23-28.
6. Coutts AJ, Kempton T, Sullivan C, Bilsborough J, Cordy J, Rampinini E. Metabolic power and energetic costs of professional Australian football match-play. J Sci Med Sports. 2015;18:219-224.
7. Buchheit M, Manouvrier C, Cassirame J, Morin JB. Monitoring locomotor load in soccer: is metabolic power, powerful? Int J Sports Med. 2015;36:1149-1155.
8. Mullen T, Highton J, Twist C. The physiological and perceptual responses to a forward-specific rugby league simulation protocol with and without physical contact. Int J Sports Physiol Perf. 2015;10:746-753.
9. Hopkins WG. A scale of magnitudes for effect statistics. Retrieved from http://www.sportsci.org/resoruce/stats/index.html
10. Weir JB. New methods for calculating metabolic rate with special reference to protein metabolism. J. Physiol. 1949;109:1-9.
10
218
219
220
221
222
223
224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259
11. Coutts AJ, Reaburn P, Abt G. Heart rate, blood lactate concentration and estimated energy expenditure in a semi-professional rugby league team during a match: a case study. J Sports Sci. 2003;21:97-103.
12. Osgnach C, Paolini E, Roberti V, Vettor M, di Prampero PE. Metabolic power and oxygen consumption in team sports: A brief response to Buchheit et al. Int J Sports Med. 2016;37:77-81.
13. Tabata I, Irisawa K, Kouzaki M, Nishimura K, Ogita F, Miyachi M. Metabolic profile of high intensity intermittent exercises. Med Sci Sports Exerc. 1997;29:390-395.
14. Crouter SE, Churilla JR, Bassett DR. Estimating energy expenditure using accelerometers. Eur J Appl Physiol. 2006;98:601-612.
15. Varley MC, Fairweather IH, Aughey RJ. Validity and reliability of GPS formeasuring instantaneous velocity during acceleration, deceleration, and constant motion. J Sports Sci. 2012;30:121-127.
11
260261262263264265266267268269270271272273274275276277278
279
Table 1. Internal and external load measured during the exercise protocol
Mean ± SDInternal Load
Energy Expenditure (kcal∙min-1) 13.2 ± 2.3Heart rate (%max) 87.4 ± 4.1B[La] (mmol∙l-1) 10.5 ± 3.6
External LoadDistance (m) 391.3 ± 16.8High-speed running (m) 53.9 ± 31.2Metabolic power (kcal∙min-1) 7.2 ± 1.0Metabolic power >20 W∙kg-1 (s) 41 ± 9Player load (AU) 54.4 ± 5.5
12
280
281
282
283
284
Figure legends
Figure 1. Energy expenditure determined using global position system-derived
metabolic power (GPS) and open-circuit spirometry (VO2) during exercise and rest
bouts. White bars denote the metabolic rate associated with standing, whilst the
shaded bars represent the energy expenditure associated with exercise (i.e. delta
energy expenditure). * Significant (P < 0.05) difference between measures.
Figure 2. Relationship between energy expenditure predicted from open circuit
spirometry and GPS
13
285
286
287
288
289
290
291
292
293
294
295
296
297
Figure 1
14
298299
300
301
Figure 2
15
302303
304