Sports Med 2011; 41 (9): 695-708 0112-1642/11/0009-0695/$49.95/0
CURRENT OPINION
ª 2011 Adis Data Information BV. All rights reserved.
A Model to Estimate the Potential for a Physical Activity-Induced Reduction in Healthcare Costs for the Elderly, Based on Pedometer/Accelerometer Data from the Nakanojo Study Yukitoshi Aoyagi1 and Roy J. Shephard2 1 Exercise Sciences Research Group, Tokyo Metropolitan Institute of Gerontology, Itabashi, Tokyo, Japan 2 Faculty of Physical Education and Health, University of Toronto, Toronto, Ontario, Canada
Abstract
The rising healthcare costs associated with an aging population have become an urgent fiscal problem. However, evidence of the efficacy of preventive programmes is limited, since almost all studies have involved only small numbers of highly selected participants. This article examines potential physical activity-induced decreases in healthcare expenses, applying a theoretical model to the Nakanojo Study of habitual physical activity and health in an entire elderly community. The Nakanojo Study has shown substantial associations of health with both step count and the duration of moderate effort (intensity >3 metabolic equivalents [METs]). Participants are classed as ‘dependent’ (n = 800) or ‘independent’ (n = 4400); the latter category is divided arbitrarily into quartiles, based on physical activity patterns (Q1–Q4; n = 1100 for each quartile). The five groups show a graded prevalence of various morbidities, including dependency, depression, osteoporosis, fractures, hypertension, diabetes mellitus, hyperlipidaemia, ischaemic heart diseases, cerebrovascular diseases, cancer and dementia. Consequently, annual healthcare expenditures (based on 2009 published Japanese costs associated with each of these conditions) differ by about yen (f)197 900 ($US1979) per person between dependent individuals and those in group Q1, f20 700 ($US207) between Q1 and Q2, f14 600 ($US146) between Q2 and Q3, and f5300 ($US53) between Q3 and Q4. Accepting a causal relationship between physical activity and health, and assuming that an increase in physical activity induces a benefit that is uniform across conditions and diseases, respective morbidity prevalences and associated healthcare costs seem likely to decrease as physical activity increases. Thus, if the physical activity of only 5% of each group could be increased by a single ranking (pedometer/accelerometer scores of 2000 steps/day and 5–10 min/day at >3 METs and/or an adjusted questionnaire score of 10 MET hours/week), one might predict average savings across this population of about f12 600 ($US126) per person, or 3.7%, of total medical expenses,
Aoyagi & Shephard
696
including f9800 ($US98) of public nursing care insurance costs and an additional f2800 ($US28) of national health insurance expenditures. The impact of various changes in the prevalence of physical activity can be simulated using our model. In principle, savings should increase if more people increase their physical activity, and/or the magnitude of individual increases in physical activity is greater. Nevertheless, our analysis suggests that if even a small fraction of individuals in the three least active groups were to make a single-rank increase in their habitual physical activity as a result of focused health support and the promotion of physical activity, a significant reduction in medical expenses might be anticipated, justifying investment in preventive programmes. We now propose to test the validity of the present simulations on a national basis, obtaining accurate and objective evidence of change in individual physical activity patterns using an advanced design of pedometer/accelerometer.
1. Introduction In Japan, as in other developed countries, the rising healthcare and medical costs associated with an aging population have become an urgent fiscal problem. Preventive medical interventions have sought to address this issue, but scientific evidence of their efficacy remains limited. To date, almost all studies of the cost effectiveness of short-term exercise promotion have involved relatively small numbers of highly selected participants.[1-6] This article offers a theoretical analysis, estimating the potential for a physical activityinduced decrease in healthcare expenses, based on cross-sectional data from the Nakanojo Study, an ongoing investigation that has examined associations between habitual physical activity and health in an entire community of elderly people.[7-27] Details of the Nakanojo Study have been described previously.[9,10] In brief, our potential subjects included all willing community residents ‡65 years of age with the exception of those who were severely demented or bedridden (giving a sample of some 5000 participants). All participants completed a conventional physical activity questionnaire once a year[24] and, in an arbitrarily selected subgroup, about a tenth of the sample, physical activity was assessed continuously, 24 hours per day, for >10 years; this subgroup did not differ from the main sample in terms of either age or sex distribution. Objective measurements were made using a specially adapted uniaxial pedometer/ accelerometer (modified Kenz Lifecorder, Suzuken ª 2011 Adis Data Information BV. All rights reserved.
Co., Ltd, Nagoya, Aichi, Japan). This device compares favourably with other pedometers and accelerometers in terms of reliability and validity,[20] offering consistently accurate estimates of both step count (intramodel reliability 0.998; absolute accuracy – <3%) and the intensity of ambulatory activity under both controlled and freeliving conditions. To date, the primary aim of the Nakanojo Study has been to establish the overall patterns of physical activity most closely associated with good health in the elderly. This article provides a brief summary of the associations that we have observed between habitual physical activity (daily step count and duration of light and moderately vigorous physical activity) and the physical, psychosocial, mental and metabolic components of health in the elderly. Based on this evidence, and accepting also the causal nature of these relationships as established by various experimental studies, a model is proposed to estimate reductions in the costs of healthcare for the community if an arbitrary fraction of the population were to increase their physical activity by the minimum amount associated with significantly better health.
2. Associations between Physical Activity and Health in the Nakanojo Study Physical activities that the seniors of Nakanojo undertake over a typical day (figure 1) can be divided arbitrarily into three categories of Sports Med 2011; 41 (9)
Healthcare Cost Reductions by Physical Activity
697
Activity intensity (METs)
Brisk walking
Housework
Errands
9 High 6 Moderate 3 Low 0 0:00
• Breakfast 2:00
4:00
6:00
8:00
• Lunch 10:00
12:00
• Dinner 14:00
16:00
18:00
20:00
22:00
24:00
Time (h) Getting up
Napping
Going to bed
Fig. 1. A typical example of daily physical activity in the elderly (as seen in 24-hour step-count recordings from the Nakanojo Study) [reproduced from Aoyagi and Shephard,[9] with permission from Adis, a Wolters Kluwer business ª Adis Data Information BV, 1996. All rights reserved]. METs = metabolic equivalents.
intensity: ‘low’ (<3 metabolic equivalents [METs]), ‘moderately vigorous’ (3–6 METs) and ‘high’ (>6 METs). High-intensity activities were pursued for an average of only <1 min/day. Pedometer/ accelerometer records over an entire year were thus summarized as the average step count per day and the daily durations of light (<3 METs) and moderately vigorous (>3 METs) activity. The relationship between step count and the duration of moderately vigorous physical activity included a statistically significant quadratic term,[13] but nevertheless could be approximated by the dotted linear line shown in figure 2. Associations between pedometer/accelerometer data and decreases in various health risks also tended to be nonlinear. However, we were able to clarify ranges of physical activity associated with health advantages of statistical and public health significance by dividing all independent and ostensibly healthy older people from our sample into quartiles (Q1–Q4), based on their year-averaged daily step counts and daily durations of activity >3 METs. Differences of physical activity between immediate quartiles (Q1 vs Q2; Q2 vs Q3; and Q3 vs Q4), and between those members of the community who were dependent and Q1, were counts of 2000 steps/day and periods of 5–10 min/day spent at an intensity >3 METs per category. In order to include in our model conditions known to be affected by ª 2011 Adis Data Information BV. All rights reserved.
physical activity where we do not yet have published pedometer/accelerometer data (coronary artery disease, stroke, cancer and dementia), we have gleaned questionnaire data from both our database and the literature,[28-32] using, as a basis of our grading, increments in reported activity of 10 MET hours/week.[33] Data from our project have indicated associations between many measures of health and both step count and the duration of moderately vigorous physical activity (figure 2). Most authors recognize that, for the elderly, any level of physical activity is better than none, but the relationships found in Nakanojo suggest that statistically and clinically significant health benefits are not observed unless certain minima of habitual physical activity are maintained. Aspects of impaired mental and psychosocial health, such as a depressive mood state[26] and a poor health-related quality of life,[23] are less prevalent in individuals who meet very modest minimum standards of habitual activity (at least 4000–5000 steps/day and/or at least 5–7.5 min/day at an intensity >3 METs), although in these instances, longitudinal research is needed to determine how far physical activity influences mood and how far the converse is the case. Stressful life events, such as a partner’s death (in both men and women) and retirement (predominantly in men), are associated with a Sports Med 2011; 41 (9)
Aoyagi & Shephard
698
low level of habitual physical activity (being seen particularly in men who take little activity >3 METs).[27] In contrast, the activity threshold associated with many aspects of better physical health, such as freedom from arteriosclerosis,[14] osteoporosis,[16] sarcopenia[17] and a low level of physical fitness,[12] amounts to at least 7000–8000 steps/day and/or at least 15–20 min/day at an intensity >3 METs, and the threshold associated with absence of the metabolic syndrome may be even greater.[18] Among our sample of individuals aged 65–74 years, the risks of hypertension and hyperglycaemia were lower only in those who took >10 000 steps/day and/or spent >30 min/day of physical activity >3 METs; in those aged 75–84 years, the corresponding threshold was >8000 steps/day and/or >20 min/day at an intensity >3 METs. Few subjects who met such activity levels showed three or more of the five commonly
accepted metabolic risk factors. On the whole, these various measures of health differed significantly between all quartiles of step count or moderately vigorous physical activity, although this was not always true for Q3 and Q4. 3. Simulation of a Physical ActivityInduced Reduction in Healthcare Costs When calculating a physical activity-related change in healthcare costs, it is important not only to select appropriate candidate diseases, but also to estimate the costs of treating each disease. The development of our model is illustrated in tables I and II. For each of the medical conditions where the Nakanojo Study[16,18,24,26] and others[28-32] have demonstrated significant associations between habitual physical activity and disease prevalence, we applied the average costs to Japanese society
35.0 Male-specific area including more exercise
Metabolic health (age <75 years)
Q4
30.0
25.0 Metabolic health (age ≥75 years)
Q3
20.0
Physical health
15.0
Q2
10.0
Psychosocial health
7.5
Mental health
Q1
Housebound
5.0 Female-specific area including more housework
Year-averaged duration of physical activity >3 METs (min/day)
40.0
2.5
Dependent Bedridden 0
1000
2000
3000
4000 5000 6000 7000 8000 Year-averaged step count (steps/day)
0 9000 10 000 11 000 12 000
Fig. 2. Schematic diagram showing categories of habitual physical activity in elderly Japanese people and the relationships between such activity patterns and health (based on data from the Nakanojo Study) [reproduced from Aoyagi and Shephard,[9] with permission from Adis, a Wolters Kluwer business ª Adis Data Information BV, 1996. All rights reserved]. METs = metabolic equivalents; Q1–Q4 = first through fourth quartiles of physical activity in study participants (n = about 50 for each quartile).
ª 2011 Adis Data Information BV. All rights reserved.
Sports Med 2011; 41 (9)
Condition/disease
Activity
Prevalence
category/groupa
%
Expense, mainly as an outpatientb,c n
¥/disease/year
¥/group/year
¥/person/year baseline
Dependency
Dependent
73
584
1 743 600
1 018 262 400
195 820
Q1
0
0
1 743 600
0
0
-195 820
Q2
0
0
1 743 600
0
0
0
Q3
0
0
1 743 600
0
0
0
Q4
0
0
1 743 600
0
0
0
1 018 262 400
195 820
Total Depression
584
Dependent
6
48
212 280
10 189 440
1 960
Q1
4
44
212 280
9 340 320
1 796
- 163
Q2
2
22
212 280
4 670 160
898
- 898
Q3
1
11
212 280
2 335 080
449
- 449
Q4
0
0
212 280
0
0
- 449
26 535 000
5 103
Total Osteoporosis
125
Dependent
18
144
169 200
24 364 800
4 686
Q1
13
143
169 200
24 195 600
4 653
- 33
Q2
8
88
169 200
14 889 600
2 863
-1 790
Q3
3
33
169 200
5 583 600
1 074
-1 790
Q4
1
11
169 200
1 861 200
358
- 716
70 894 800
13 634
Total Fractures
419
Dependent
15
120
42 750
5 130 000
987
Q1
11
121
42 750
5 172 750
995
8
Q2
7
77
42 750
3 291 750
633
- 362
Q3
3
33
42 750
1 410 750
271
- 362
Q4
1
11
42 750
470 250
90
- 181
15 475 500
2 976
Total
362
Dependent
48
384
179 760
69 027 840
13 275
Q1
36
396
179 760
71 184 960
13 689
415
Q2
24
264
179 760
47 456 640
9 126
- 4 563 Continued next page
699
Sports Med 2011; 41 (9)
Hypertension
difference between immediate groups
Healthcare Cost Reductions by Physical Activity
ª 2011 Adis Data Information BV. All rights reserved.
Table I. Estimates of physical activity-related healthcare costs for a sample of 5200 elderly people (based on data for 2009 from the Nakanojo Study)
Condition/disease
Activity
Prevalence
category/groupa
%
700
ª 2011 Adis Data Information BV. All rights reserved.
Table I. Contd Expense, mainly as an outpatientb,c n
¥/disease/year
¥/group/year
¥/person/year baseline
Q3
12
132
179 760
23 728 320
4 563
- 4 563
Q4
8
88
179 760
15 818 880
3 042
- 1 521
227 216 640
43 696 8 147
Total Diabetes mellitus
1 264
Dependent
16
128
330 960
42 362 880
Q1
12
132
330 960
43 686 720
8 401
255
Q2
8
88
330 960
29 124 480
5 601
- 2 800
Q3
4
44
330 960
14 562 240
2 800
- 2 800
Q4
2
22
330 960
7 281 120
1 400
- 1 400
137 017 440
26 350
Total Hyperlipidaemia
difference between immediate groups
414
Dependent
19
152
163 080
24 788 160
4 767
Q1
14
154
163 080
25 114 320
4 830
63
Q2
9
99
163 080
16 144 920
3 105
- 1 725
Q3
4
44
163 080
7 175 520
1 380
- 1 725
Q4
2
22
163 080
3 587 760
690
- 690
76 810 680
14 771
Total
471
Ischaemic heart
Dependent
12
96
307 080
29 479 680
5 669
disease
Q1
7
77
307 080
23 645 160
4 547
- 1 122
Q2
2
22
307 080
6 755 760
1 299
- 3 248
Q3
1
11
307 080
3 377 880
650
- 650
Q4
1
11
307 080
3 377 880
650
0
66 636 360
12 815
Total
217
Dependent
15
120
228 960
27 475 200
5 284
disease
Q1
9
99
228 960
22 667 040
4 359
- 925
Q2
3
33
228 960
7 555 680
1 453
-2 906
Q3
1
11
228 960
2 518 560
484
-969
Q4
1
11
228 960
2 518 560
484
0
62 735 040
12 064
Total
274
Continued next page
Aoyagi & Shephard
Sports Med 2011; 41 (9)
Cerebrovascular
Condition/disease
Activity
Prevalence
category/groupa
%
Expense, mainly as an outpatientb,c n
¥/disease/year
¥/group/year
¥/person/year baseline
Dementia
Dependent
7
56
266 520
14 925 120
2 870
Q1
4
44
266 520
11 726 880
2 255
- 615
Q2
1
11
266 520
2 931 720
564
- 1 691
Q3
0
0
266 520
0
0
- 564
Q4
0
0
266 520
0
0
0
29 583 720
5 689
Total Cancer
111
Dependent
8
64
162 000
10 368 000
1 994
Q1
6
66
162 000
10 692 000
2 056
62
Q2
4
44
162 000
7 128 000
1 371
- 685
Q3
2
22
162 000
3 564 000
685
- 685
Q4
1
11
162 000
1 782 000
343
- 343
33 534 000
6 449
Total Overall
207
1 276 373 520
245 456
Q1
Dependent
247 425 750
47 582
Q2
139 948 710
26 913
- 20 669
Q3
64 255 950
12 357
- 14 556
36 697 650
7 057
- 5 300
1 764 701 580
339 366
Q4 Grand total a
difference between immediate groups
Healthcare Cost Reductions by Physical Activity
ª 2011 Adis Data Information BV. All rights reserved.
Table I. Contd
-197 875
Dependent = elderly persons who are certified as a recipient of services under the nursing care insurance system (n = 800). Q1–Q4 = first through fourth quartiles of physical activity in independent and ostensibly healthy older people (n = 1100 for each quartile). Older adults are categorized on the bases of step count and the duration of physical activity >3 METs and/or an adjusted physical activity questionnaire score: Dependent = <2000 steps/day and <2.5 min/day and/or <10 MET h/wk; Q1 = 2000–<5000 (mean 4000) steps/day and <7.5 (mean 5) min/day and/or 10–<25 (mean 20) MET h/wk; Q2 = 5000–<7000 (mean 6000) steps/day and 7.5–<15 (mean 10) min/day and/or 25–<35 (mean 30) MET h/wk; Q3 = 7000–<9000 (mean 8000) steps/day and 15–<25 (mean 20) min/day and/or 35–<45 (mean 40) MET h/wk;
b
The period of a treatment for fractures is considered to be 3 months.
c ¥1 = approximately $US0.01. MET(s) = metabolic equivalent(s); ¥ = Japanese yen.
701
Sports Med 2011; 41 (9)
Q4 = ‡9000 (mean 10 000) steps/day and ‡25 (mean 30) min/day and/or ‡45 (mean 50) MET h/wk.
702
ª 2011 Adis Data Information BV. All rights reserved.
Table II. A simulation of physical activity-induced reductions in healthcare costs for a sample of 5200 elderly people (based on data for 2009 from the Nakanojo Study) Condition/disease
Activity
Prevalenceb
category/groupa
%
¥/disease/year
preimprovement Dependency
preimprovement
postimprovement
pre-/postdifference
584
555
1 743 600
1 018 262 400
967 349 280
195 820
186 029
-9 791
0
0
1 743 600
0
0
0
0
0
Q2
0
0
0
1 743 600
0
0
0
0
0
Q3
0
0
0
1 743 600
0
0
0
0
0
Q4
0
0
0
1 743 600
0
0
0
0
0
584
555
1 018 262 400
967 349 280
195 820
186 029
-9 791
Dependent
6
48
47
212 280
10 189 440
10 019 616
1 960
1 927
- 33
Q1
4
44
43
212 280
9 340 320
9 106 812
1 796
1 751
- 45
Q2
2
22
21
212 280
4 670 160
4 553 406
898
876
- 22
Q3
1
11
10
212 280
2 335 080
2 218 326
449
427
- 22
Q4
0
0
0
212 280
0
0
0
0
0
125
122
26 535 000
25 898 160
5 103
4 980
- 122
Dependent
18
144
142
169 200
24 364 800
24 026 400
4 686
4 620
- 65
Q1
13
143
140
169 200
24 195 600
23 730 300
4 653
4 564
- 89
Q2
8
88
85
169 200
14 889 600
14 424 300
2 863
2 774
- 89
Q3
3
33
32
169 200
5 583 600
5 397 480
1 074
1 038
- 36
Q4
1
11
11
169 200
1 861 200
1 861 200
358
358
0
419
410
70 894 800
69 439 680
13 634
13 354
- 280
Dependent
15
120
118
42 750
5 130 000
5 061 600
987
973
- 13
Q1
11
121
119
42 750
5 172 750
5 078 700
995
977
- 18
Q2
7
77
75
42 750
3 291 750
3 197 700
633
615
- 18
Q3
3
33
32
42 750
1 410 750
1 363 725
271
262
-9
Q4
1
11
11
42 750
470 250
470 250
90
90
0
362
355
15 475 500
15 171 975
2 976
2 918
- 58
Total Dependent
48
384
379
179 760
69 027 840
68 164 992
13 275
13 109
- 166
Q1
36
396
389
179 760
71 184 960
69 998 544
13 689
13 461
- 228
Q2
24
264
257
179 760
47 456 640
46 270 224
9 126
8 898
- 228 Continued next page
Aoyagi & Shephard
Sports Med 2011; 41 (9)
Hypertension
¥/person/year postimprovement
0
Total Fractures
¥/group/year preimprovement
73
Dependent
Total Osteoporosis
postimprovement
Q1
Total Depression
Expense, mainly as an outpatientb,c,d
n
Condition/disease
Activity
Prevalenceb
category/groupa
%
¥/disease/year
preimprovement
¥/group/year preimprovement
¥/person/year postimprovement
preimprovement
postimprovement
pre-/postdifference - 76
12
132
130
179 760
23 728 320
23 332 848
4 563
4 487
Q4
8
88
88
179 760
15 818 880
15 818 880
3 042
3 042
0
1 264
1 244
227 216 640
223 585 488
43 696
42 997
- 698
Dependent
16
128
126
330 960
42 362 880
41 833 344
8 147
8 045
- 102
Q1
12
132
130
330 960
43 686 720
42 958 608
8 401
8 261
- 140
Q2
8
88
86
330 960
29 124 480
28 396 368
5 601
5 461
- 140
Q3
4
44
43
330 960
14 562 240
14 198 184
2 800
2 730
- 70
Q4
2
22
22
330 960
7 281 120
7 281 120
1 400
1 400
0
414
407
137 017 440
134 667 624
26 350
25 898
- 452
24 788 160
24 462 000
4 767
4 704
- 63
Total Hyperlipidaemia
postimprovement
Q3
Total Diabetes mellitus
Expense, mainly as an outpatientb,c,d
n
Dependent
19
152
150
Q1
14
154
151
163 080
25 114 320
24 665 850
4 830
4 743
- 86
Q2
9
99
96
163 080
16 144 920
15 696 450
3 105
3 019
- 86
Q3
4
44
43
163 080
7 175 520
6 996 132
1 380
1 345
- 34
Q4
2
22
22
163 080
3 587 760
3 587 760
690
690
0
471
462
76 810 680
75 408 192
14 771
14 502
- 270
Total
163 080
Ischaemic heart
Dependent
12
96
94
307 080
29 479 680
28 865 520
5 669
5 551
- 118
disease
Q1
7
77
74
307 080
23 645 160
22 800 690
4 547
4 385
- 162
Q2
2
22
21
307 080
6 755 760
6 586 866
1 299
1 267
- 32
Q3
1
11
11
307 080
3 377 880
3 377 880
650
650
0
Q4
1
307 080
Total
11
11
217
212
3 377 880
3 377 880
650
650
0
66 636 360
65 008 836
12 815
12 502
- 313
Dependent
15
120
118
228 960
27 475 200
26 925 696
5 284
5 178
- 106
disease
Q1
9
99
96
228 960
22 667 040
21 911 472
4 359
4 214
- 145
Q2
3
33
32
228 960
7 555 680
7 303 824
1 453
1 405
- 48
Q3
1
11
11
228 960
2 518 560
2 518 560
484
484
0
Q4
1
11
11
228 960
2 518 560
2 518 560
484
484
0
274
267
62 735 040
61 178 112
12 064
11 765
- 299
Total
Continued next page
703
Sports Med 2011; 41 (9)
Cerebrovascular
Healthcare Cost Reductions by Physical Activity
ª 2011 Adis Data Information BV. All rights reserved.
Table II. Contd
704
ª 2011 Adis Data Information BV. All rights reserved.
Table II. Contd Condition/disease
Activity
Prevalenceb
category/groupa
%
¥/disease/year
preimprovement Dementia
¥/person/year postimprovement
preimprovement
postimprovement
pre-/postdifference
7
56
55
266 520
14 925 120
14 605 296
2 870
2 809
- 62
4
44
42
266 520
11 726 880
11 287 122
2 255
2 171
- 85
Q2
1
11
10
266 520
2 931 720
2 785 134
564
536
- 28
Q3
0
0
0
266 520
0
0
0
0
0
Q4
0
0
0
266 520
0
0
0
0
0
111
108
29 583 720
28 677 552
5 689
5 515
- 174
Dependent
8
64
63
162 000
10 368 000
10 238 400
1 994
1 969
- 25
Q1
6
66
65
162 000
10 692 000
10 513 800
2 056
2 022
- 34
Q2
4
44
43
162 000
7 128 000
6 949 800
1 371
1 337
- 34
Q3
2
22
21
162 000
3 564 000
3 474 900
685
668
- 17
Q4
1
11
11
162 000
1 782 000
1 782 000
343
343
0
207
203
33 534 000
32 958 900
6 449
6 338
- 111
1 276 373 520
1 221 552 144
245 456
234 914
-10 543
Q1
247 425 750
242 051 898
47 582
46 548
- 1 033
Q2
139 948 710
136 164 072
26 913
26 185
- 728
Q3
64 255 950
62 878 035
12 357
12 092
- 265
Q4
36 697 650
36 697 650
7 057
7 057
0
1 764 701 580
1 699 343 799
339 366
326 797
-12 569
Dependent
Grand total a
¥/group/year preimprovement
Dependent
Total Overall
postimprovement
Q1
Total Cancer
Expense, mainly as an outpatientb,c,d
n
Dependent = elderly persons who are certified as a recipient of services under the nursing care insurance system (n = 800). Q1–Q4 = first through fourth quartiles of physical activity in independent and ostensibly healthy older people (n = 1100 for each quartile). Older adults are categorized on the bases of step count and the duration of physical activity >3 METs and/or an adjusted physical activity questionnaire score: Dependent = <2000 steps/day and <2.5 min/day and/or <10 MET h/wk; Q1 = 2000–<5000 (mean 4000) steps/day and <7.5 (mean 5) min/day and/or 10–<25 (mean 20) MET h/wk; Q2 = 5000–<7000 (mean 6000) steps/day and 7.5–<15 (mean 10) min/day and/or 25–<35 (mean 30) MET h/wk; Q3 = 7000–<9000 (mean 8000) steps/day and 15–<25 (mean 20) min/day and/or 35–<45 (mean 40) MET h/wk; Pre- and post-improvement = respective cases where activity has not and has increased by a single rank (2000 steps/day and 5–10 min/day at >3 METs and/or 10 MET h/wk) in only 5% (n = 40 or 55) of each group (‘dependent’ or ‘independent’, respectively).
c The period of a treatment for fractures is considered to be 3 months. d
¥1 = approximately $US0.01.
MET(s) = metabolic equivalent(s); ¥ = Japanese yen.
Aoyagi & Shephard
Sports Med 2011; 41 (9)
Q4 = ‡9000 (mean 10 000) steps/day and ‡25 (mean 30) min/day and/or ‡45 (mean 50) MET h/wk. b
Healthcare Cost Reductions by Physical Activity
(90% of the actual costs, and reflecting the sums paid directly to service providers by the public nursing care insurance[34] and national health insurance schemes[35]). Conditions included in the analysis comprised dependency, depression, osteoporosis, fractures, hypertension, diabetes mellitus, hyperlipidaemia, ischaemic or coronary heart diseases (angina pectoris and myocardial infarction), stroke or cerebrovascular diseases (cerebral infarction, cerebral haemorrhage and subarachnoid haemorrhage), dementia (vascular dementia and Alzheimer’s disease) and cancer (colon/rectum, lung, breast and endometrial cancers). Tables I and II summarize the costs incurred for elderly Nakanojo residents who required medical assistance or nursing care (designated as ‘dependent’). For both dependent and independent seniors, the two tables include only insurance payments for treatments carried out by a doctor or the outpatient service of a hospital because, to date, our studies have covered mainly free-living seniors; as yet, only a few of the group have been hospitalized or institutionalized for the conditions listed in the two tables. Official records for 2009[34,35] demonstrate that, in Japan, the total insured costs attributable to the 11 conditions that we have listed account for some two-thirds of the benefits paid on behalf of older people. We do not as yet have data relating an individual’s level of physical activity to the remaining causes of ill health and, to this extent, our calculations may underestimate the grand total of savings from an increase of physical activity. However, many of the diseases that we have not as yet considered are less closely related to patterns of habitual physical activity. Participants fell into two main categories (tables I and II): ‘dependent’ (n = 800) and ‘independent’ (n = 4400), the latter being divided arbitrarily into quartiles based on objectively and/or subjectively determined physical activity patterns (Q1–Q4; n = 1100 for each quartile). There were no statistically significant differences in either age (mean 78–80 years) or sex distribution (men 46–51% vs women 49–54%) between the five groups, so that intergroup differences in disease prevalence cannot be attributed to aging. The tables show the number and percentage of individuals in each of ª 2011 Adis Data Information BV. All rights reserved.
705
the five groups receiving services under the nursing care insurance and national health insurance schemes for each condition or disease. The insured medical and healthcare expenses for a given group were calculated by multiplying the average cost of each condition by its prevalence, and data were finally expressed as the average cost for individual members of the group. Table II examines the health effects (and resulting decreases in medical costs) that would arise from a single-rank increase of physical activity (a gain in pedometer/ accelerometer scores of 2000 steps/day and 5–10 min/day at >3 METs and/or an adjusted questionnaire score of 10 MET hours/week) in only 5% of the members of each group. A singlerank increment of activity is chosen as having statistical and clinical health significance, and 5% also reflects the approximate annual rate at which elderly people in Nakanojo have been certified as recipients of services under the nursing care insurance system because of decreased physical activity. The new numbers of seniors per group having each condition or disease after the assumed intervention were calculated as (0.01 · [prevalence rate of the disease concerned in the group concerned] · [95% of persons in the group concerned]) + (0.01 · [prevalence rate of the disease concerned in the next upper group or the group concerned in the case of Q4] · [5% of persons in the group concerned]). Thus, among the 800 individuals who were initially classed as ‘dependent’, the new number of cases of hypertension would be (0.01 · 48 · 760) + (0.01 · 36 · 40) = 379, compared with the original 384 cases. The prevalence of the various conditions shown in tables I and II differed between groups. Assuming a uniform across-group effect for each condition, one might anticipate that the corresponding prevalence rate and thus the estimated medical and healthcare costs would decrease as physical activity increased. Summing across conditions, cost differences (based on 2009 published records[34,35]) would amount to a total of some Japanese yen (f)197 900 ($US1979) per person between dependent and Q1 individuals, f20 700 ($US207) between Q1 and Q2, f14 600 ($US146) between Q2 and Q3, and f5300 ($US53) between Q3 and Q4 (table I). Thus, if there were to be a Sports Med 2011; 41 (9)
Aoyagi & Shephard
706
single-rank increase in physical activity (a gain of 2000 steps/day and 5–10 min/day at >3 METs and/or 10 MET hours/week) in only 5% of each group, the potential savings to the community would average about f12 600 ($US126) per person, 3.7%, of total medical and healthcare expenses, including f9800 ($US98) of public nursing care insurance costs and a further f2800 ($US28) of national health insurance expenditures (table II). Alternatively, an at least equal saving could be achieved if only 5% of seniors in each group were to maintain their habitual physical activity in the face of aging. The estimated savings would in turn justify at least equivalent investments in programmes aimed at maintaining or augmenting physical activity. For example, one might advocate community use of an advanced pedometer/ accelerometer at a cost of f5000–10 000 ($US50–100) per unit; reactivity to the wearing of such a device could induce at least a temporary single-rank increase in physical activity,[36,37] with reductions in medical costs exceeding this expense. The costs and the effectiveness of other interventions needed to induce long-term increases in the physical activity of various percentages of currently sedentary people remain an important issue for further investigation. 4. Limitations of the Analysis There are presently several important limitations in our analysis. We have already indicated that our calculations do not include the cost of inpatient treatments. Although few of our sample were hospitalized, such treatments can be very costly (in Japan, amounting to a third of all health insurance benefits). Moreover, the costs ascribed to a given condition are, necessarily, average costs per individual, although it is probable that much of the total cost will be incurred in the final few months of life.[38] We have also assumed a linear relationship between disease prevalence and physical activity, with the conservative assumption that if a diagnosis is made, the severity of the disease and thus the cost of treatment are equivalent in active and sedentary individuals. However, there is a strong likelihood that the costs of treating many diseases will be greater for ª 2011 Adis Data Information BV. All rights reserved.
someone who is sedentary than for a more active person. Our estimates for the effects of physical activity upon ischaemic heart disease, stroke, cancer and dementia are based on questionnaire data obtained from the Nakanojo database and elsewhere,[28-33] and there remains a need to incorporate into our model reliable objective data for these conditions. There may be some overlap between various categories of expense, thus resulting in an overestimation of total medical expenditures; for instance, some of the costs associated with the treatment of fractures may also have been assigned to osteoporosis. People with some form of illness (and thus more expensive to the health insurance system) could also have become less physically active because of their ailment, exaggerating the apparent influence of physical activity level upon medical costs. In such a situation, an increase in physical activity would not necessarily reduce healthcare costs by the full intercategory difference seen in tables I and II. The associations between physical activity and health that we have introduced into our model are as yet based on cross-sectional studies. For most of the conditions that we have considered, available reports from longitudinal and experimental studies support a causal association, but nevertheless the direction of the association remains to be confirmed for some disorders. Even more importantly, there is no guarantee that an increase of activity quartile would restore an inactive individual to the health status of someone who has maintained a higher level of activity. Economists may also object that an increase in physical activity is likely to cause some prolongation of life. This is a desired outcome. Nevertheless, it could increase both pension and healthcare expenditures (although Fries[38] has argued that a large fraction of an individual’s medical expenses arises from treatments that are given during the final few weeks of life, and such treatments are unlikely to be augmented by an increase of longevity). Finally, even if the community costs of treating a specific condition are decreased by an increase of physical activity, this does not guarantee that overall medical expenditures will fall; those responsible for medical care Sports Med 2011; 41 (9)
Healthcare Cost Reductions by Physical Activity
may simply decide to spend the money thus saved on other health initiatives. This is not necessarily a bad outcome, but it should be recognized when making an overall financial analysis. In transferring the present data from Japan to other developed countries, account will need to be taken not only of nominal exchange rates, but also of substantial differences in the costs incurred by differing systems of healthcare delivery. Nevertheless, the present estimates of savings from an increase of physical activity are of a similar order to those previously calculated for Canadian populations.[29,30] They are also in concordance with the decrease in medical insurance costs as observed in a 1-year quasiexperimental study of a Canadian worksite fitness programme for office workers,[5,6] where about a third of eligible individuals became more active and showed small gains in aerobic fitness. 5. Conclusions and Practical Implications As further information accumulates, the potential savings from a wide range of increases in physical activity can be simulated, using a model of the type shown in table II. The gradients of the physical activity/disease prevalence relationship and the costs of treatment may both differ from one country to another, but analogous models can still be developed. In principle, as a larger proportion of people augment their physical activity, and/or the magnitude of the increase in physical activity becomes larger, greater savings may be predicted. But perhaps the most important inference from table II is that a significant reduction in healthcare costs can still be anticipated if even a small proportion of seniors in each of the least active groups (Q1, Q2 and particularly the ‘dependent’) were to increase their habitual physical activity by a single rank. The resulting savings would in turn justify at least equivalent investments in programmes designed to augment physical activity. Community use of an advanced pedometer/accelerometer, such as the Kenz Lifecorder, allows objective and accurate monitoring of increases in habitual physical activity from such interventions, and the likely savings in medical and healthcare costs can be ª 2011 Adis Data Information BV. All rights reserved.
707
calculated from this information. We thus plan to assess the validity of our proposed model in a national project, looking longitudinally at increases in physical activity patterns and resulting gains of health status. Acknowledgements This article focuses particularly on data from an interdisciplinary study on the habitual physical activity and health of elderly people living in Nakanojo, Gunma, Japan (the Nakanojo Study). This study was supported in part by grants (Grant-in-Aid for Encouragement of Young Scientists: 12770037 and Grant-in-Aid for Scientific Research [C]: 15500503, [C]: 17500493 and [B]: 19300235) from the Japan Society for the Promotion of Science. The authors gratefully acknowledge the expert technical assistance of the research and nursing staff of the Tokyo Metropolitan Institute of Gerontology, The University of Tokyo and the Nakanojo Public Health Center. We would also like to thank the subjects whose participation made the Nakanojo Study possible. No sources of funding were used in preparing this manuscript, and the authors have no conflicts of interest relevant to the content of this manuscript.
References 1. Ackermann RT, Cheadle A, Sandhu N, et al. Community exercise program use and changes in healthcare costs for older adults. Am J Prev Med 2003; 25 (3): 232-7 2. Kemmler W, von Stengel S, Mayer S, et al. Exercise effects on risk factors and health care costs in the elderly: final results of the senior fitness and prevention study (SEFIP) [in German]. Deut Z Sportmed 2010; 61 (11): 264-9 3. Muller-Riemenschneider F, Reinhold T, Willich SN. Costeffectiveness of interventions promoting physical activity. Br J Sports Med 2009; 43 (1): 70-6 4. Nguyen HQ, Ackermann RT, Berke EM, et al. Impact of a managed-Medicare physical activity benefit on health care utilization and costs in older adults with diabetes. Diabetes Care 2007; 30 (1): 43-8 5. Shephard RJ, Corey P, Renzland P, et al. The influence of an employee fitness and lifestyle modification program upon medical care costs. Can J Public Health 1982; 73 (4): 259-63 6. Shephard RJ, Corey P, Renzland P, et al. The impact of changes in fitness and lifestyle upon health care utilization. Can J Public Health 1983; 74 (1): 51-4 7. Aoyagi Y, Shephard RJ. How many days of pedometer monitoring are needed [letter]? Med Sci Sports Exerc 2009; 41 (3): 734 8. Aoyagi Y, Shephard RJ. How should objectively measured physical activity data be used analytically [letter]? Exerc Sport Sci Rev 2009; 37 (2): 109 9. Aoyagi Y, Shephard RJ. Steps per day: the road to senior health? Sports Med 2009; 39 (6): 423-38 10. Aoyagi Y, Shephard RJ. Habitual physical activity and health in the elderly: the Nakanojo Study. Geriatr Gerontol Int 2010; 10 (1 Suppl.): 236S-43S
Sports Med 2011; 41 (9)
708
11. Aoyagi Y, Togo F, Matsuki S, et al. Walking velocity measured over 5 m as a basis of exercise prescription for the elderly: preliminary data from the Nakanojo Study. Eur J Appl Physiol 2004; 93 (1-2): 217-23 12. Aoyagi Y, Park H, Watanabe E, et al. Habitual physical activity and physical fitness in older Japanese adults: the Nakanojo Study. Gerontology 2009; 55 (5): 523-31 13. Aoyagi Y, Park H, Park S, et al. Habitual physical activity and health-related quality of life in older adults: interactions between the amount and intensity of activity (the Nakanojo Study). Qual Life Res 2010; 19 (3): 333-8 14. Aoyagi Y, Park H, Kakiyama T, et al. Yearlong physical activity and regional stiffness of arteries in older adults: the Nakanojo Study. Eur J Appl Physiol 2010; 109 (3): 455-64 15. Aoyagi Y, Park H, Park S, et al. Interactive effects of milk basic protein supplements and habitual physical activity on bone health in older women: a 1-year randomized controlled trial. Int Dairy J 2010; 20 (10): 724-30 16. Park H, Togo F, Watanabe E, et al. Relationship of bone health to yearlong physical activity in older Japanese adults: cross-sectional data from the Nakanojo Study. Osteoporos Int 2007; 18 (3): 285-93 17. Park H, Park S, Shephard RJ, et al. Yearlong physical activity and sarcopenia in older adults: the Nakanojo Study. Eur J Appl Physiol 2010; 109 (5): 953-61 18. Park S, Park H, Togo F, et al. Year-long physical activity and metabolic syndrome in older Japanese adults: crosssectional data from the Nakanojo Study. J Gerontol A Biol Sci Med Sci 2008; 63 (10): 1119-23 19. Shephard RJ, Aoyagi Y. Seasonal variations in physical activity and implications for human health. Eur J Appl Physiol 2009; 107 (3): 251-71 20. Shephard RJ, Aoyagi Y. Objective monitoring of physical activity in older adults: clinical and practical implications. Phys Ther Rev 2010; 15 (3): 170-82 21. Togo F, Watanabe E, Park H, et al. Meteorology and the physical activity of the elderly: the Nakanojo Study. Int J Biometeorol 2005; 50 (2): 83-9 22. Togo F, Watanabe E, Park H, et al. How many days of pedometer use predict the annual activity of the elderly reliably? Med Sci Sports Exerc 2008; 40 (6): 1058-64 23. Yasunaga A, Togo F, Watanabe E, et al. Yearlong physical activity and health-related quality of life in older Japanese adults: the Nakanojo Study. J Aging Phys Act 2006; 14 (3): 288-301 24. Yasunaga A, Park H, Watanabe E, et al. Development and evaluation of the physical activity questionnaire for elderly Japanese: the Nakanojo Study. J Aging Phys Act 2007; 15 (4): 398-411 25. Yasunaga A, Togo F, Watanabe E, et al. Sex, age, season, and habitual physical activity of older Japanese: the Nakanojo Study. J Aging Phys Act 2008; 16 (1): 3-13
ª 2011 Adis Data Information BV. All rights reserved.
Aoyagi & Shephard
26. Yoshiuchi K, Nakahara R, Kumano H, et al. Yearlong physical activity and depressive symptoms in older Japanese adults: cross-sectional data from the Nakanojo Study. Am J Geriatr Psychiatry 2006; 14 (7): 621-4 27. Yoshiuchi K, Inada S, Nakahara R, et al. Stressful life events and habitual physical activity in older adults: 1-year accelerometer data from the Nakanojo Study. Ment Health Phys Act 2010; 3 (1): 23-5 28. Allender S, Foster C, Scarborough P, et al. The burden of physical activity-related ill health in the UK. J Epidemiol Community Health 2007; 61 (4): 344-8 29. Katzmarzyk PT, Gledhill N, Shephard RJ. The economic burden of physical inactivity in Canada. CMAJ 2000; 163 (11): 1435-40 30. Shephard RJ. The cost-effectiveness and cost-benefits of ‘active living’. In: Shephard RJ, Alexander MJL, Cantu RC, et al., editors. Year book of sports medicine, 2002. Philadelphia (PA): CV Mosby, 2002: xvii-xxx 31. US Department of Health and Human Services, Office of Disease Prevention and Health Promotion. Physical activity guidelines for Americans: be active, healthy, and happy. Washington, DC: US Department of Health and Human Services, Office of Disease Prevention and Health Promotion, 2008: ODPHP publication no.: U0036 32. US Department of Health and Human Services, Physical Activity Guidelines Advisory Committee. Physical Activity Guidelines Advisory Committee Report. Washington, DC: US Department of Health and Human Services, Physical Activity Guidelines Advisory Committee, 2008 33. Woolcott JC, Ashe MC, Miller WC, et al., PACC Research Team. Does physical activity reduce seniors’ need for healthcare? A study of 24 281 Canadians. Br J Sports Med 2010; 44 (12): 902-4 34. Town of Nakanojo. Health and welfare plan for senior citizens. 4th report [in Japanese]. Nakanojo: Town of Nakanojo, 2009 35. Gunma Prefecture National Health Insurance Group Federation. Disease classification statistics, based on medical treatments received in May of 2008. 39th issue [in Japanese]. Maebashi: Gunma Prefecture National Health Insurance Group Federation, 2009 36. Clemes SA, Parker RA. Increasing our understanding of reactivity to pedometers in adults. Med Sci Sports Exerc 2009; 41 (3): 674-80 37. Clemes SA, Matchett N, Wane SL. Reactivity: an issue for short-term pedometer studies? Br J Sports Med 2008; 42 (1): 68-70 38. Fries JF. Aging well. Reading (MA): Addison-Wesley, 1980
Correspondence: Dr Yukitoshi Aoyagi, Exercise Sciences Research Group, Tokyo Metropolitan Institute of Gerontology, 35-2 Sakaecho, Itabashi-ku, Tokyo 173-0015, Japan. E-mail:
[email protected]
Sports Med 2011; 41 (9)
Sports Med 2011; 41 (9): 709-719 0112-1642/11/0009-0709/$49.95/0
REVIEW ARTICLE
ª 2011 Adis Data Information BV. All rights reserved.
Skin Conditions in Figure Skaters, Ice-Hockey Players and Speed Skaters Part I – Mechanical Dermatoses Brook E. Tlougan,1 Anthony J. Mancini,2,3,4 Jenny A. Mandell,5 David E. Cohen5 and Miguel R. Sanchez5,6 1 2 3 4 5
Department of Dermatology, Columbia University Medical Center, New York, NY, USA Department of Pediatrics, Feinberg School of Medicine, Northwestern University, Chicago, IL, USA Department of Dermatology, Feinberg School of Medicine, Northwestern University, Chicago, IL, USA Division of Dermatology, Children’s Memorial Hospital, Chicago, IL, USA Ronald O. Perelman Department of Dermatology, School of Medicine, New York University, New York, NY, USA 6 Department of Dermatology, Bellevue Hospital Center, New York, NY, USA
Contents Abstract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Mechanical Dermatoses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Skater’s Nodules. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Pump Bumps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Piezogenic Pedal Papules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Talon Noir . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5 Skate Bite/Lace Bite. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.6 Friction Bullae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.7 Corns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.8 Callosities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.9 Onychocryptosis (Unguis Incarnatus) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.10 Skater’s Toenail/Skater’s Toe. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.11 Lacerations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Abstract
709 710 711 711 712 713 713 714 714 715 715 716 717 717 718
Figure skaters, ice-hockey players and speed skaters experience a range of dermatologic conditions and tissue-related injuries on account of mechanical trauma, infectious pathogens, inflammatory processes and environmental factors related to these competitive pursuits. Sports medicine practitioners, family physicians, dermatologists and coaches should be familiar with these skin conditions to ensure timely and accurate diagnosis and management of affected athletes. This review is Part I of a subsequent companion review and provides a comprehensive review of mechanical dermatoses experienced by ice-skating athletes, including skater’s nodules and its variants, pump bumps, piezogenic pedal papules, talon noir, skate/lace bite, friction bullae,
Tlougan et al.
710
corns and calluses, onychocryptosis, skater’s toe and skate blade-induced lacerations. These injuries result from friction, shear forces, chronic pressure and collisions with surfaces that occur when athletes endure repetitive jump landings, accelerated starts and stops and other manoeuvres during rigorous training and competition. Ill-fitting skates, improper lacing techniques and insufficient lubrication or protective padding of the foot and ankle often contribute to the development of skin conditions that result from these physical and mechanical stresses. As we will explain, simple measures can frequently prevent the development of these conditions. The treatment of skater’s nodules involves reduction in chronic stimulation of the malleoli, and the use of keratolytics and intralesional steroid injections; if malleolar bursitis develops, bursa aspirations may be required. Pump bumps, which result from repetitive friction posteriorly, can be prevented by wearing skates that fit correctly at the heel. Piezogenic pedal papules may be treated conservatively by using heel cups, compressive stockings and by reducing prolonged standing. Talon noir usually resolves without intervention within several weeks. The treatment of skate bite is centred on reducing compression by the skate tongue of the extensor tendons of the anterior ankle, which can be accomplished by use of proper lacing techniques, increasing pliability of the skate tongue and using protective padding, such as Bunga Pads. Anti-inflammatory medications and cold compresses can also help reduce inflammation. Friction bullae are best managed by careful lancing of painful blisters and application of petrolatum or protective dressings to accelerate healing; preventative measures include the use of well fitting skates, proper lacing techniques and moisturewicking socks. Corns and calluses are similarly best prevented by the use of well fitted skates and orthotic devices. Symptomatic, debridement reduces the irritant effect of the thick epidermis, and can be accomplished by soaking the area in warm water followed by paring. Application of creams with high concentrations of urea or salicylic acid can also soften callosities. Cases of onychocryptosis benefit from warm soaks, antibiotic ointments and topical steroids to reduce inflammation, but sometimes chemical or surgical matricectomies are required. Preventative measures of both onychocryptosis and skater’s toe include cutting toenails straight across to allow for a more equal distribution of forces within the toe box. Finally, the prevention and treatment of lacerations, which constitute a potentially fatal type of mechanical injury, require special protective gear and acute surgical intervention with appropriate suturing. The subsequent companion review of skin conditions in ice skaters will discuss infectious, inflammatory and cold-induced dermatoses, with continued emphasis on clinical presentation, diagnosis, treatment and prevention.
1. Introduction Participation in winter sports, particularly figure skating, ice hockey and speed skating, has increased in recent years both at the competitive and recreational level. Cold-related tissue injury, ª 2011 Adis Data Information BV. All rights reserved.
chronic irritation, friction and pressure elicited by equipment or padding, and physiological responses to physical stress can result in the development of skin lesions and disease that dermatologists and sports medicine practitioners should be able to recognize and treat. This review is Part I of Sports Med 2011; 41 (9)
Ice-Skating Dermatoses Part I: Mechanical
a subsequent companion review and discusses the cutaneous manifestations of ice-related sports that result from mechanical injury, highlighting clinical presentation, diagnosis, treatment, and prevention. The most frequently cited area of injury in skaters is the foot and ankle.[1-5] In contrast to footwear used by other ice skaters, figure skate blades have a toe pick that helps to propel the skater upwards during toe jumps, which creates remarkably high pressures on the phalanges of the launching foot. Also, figure skaters are the only athletes to wear high-heeled boots that maintain the ankle in partial plantar flexion, changing the mechanics of the foot. The mechanical stresses on figure skating boots require athletes to replace boots as often as every 6 months.[6] Likewise, these mechanical dynamics between the foot and skate result in overuse injuries, which predominate in singles skaters.[7] In contrast, pairs skaters, ice dancers and synchronized skaters, are more frequently afflicted by acute injuries.[8] To investigate the nature and frequency of dermatoses in ice-skating athletes, we comprehensively reviewed the available literature on cutaneous disease in figure skating, ice hockey and speed skating, from 1979 to 2011, using the PubMed database. We excluded tissue injuries not specifically involving the skin. Search terms were broad and included ‘injuries in figure skating’, ‘ice hockey’, ‘speed skating’ and ‘winter sports’, as well as ‘skin conditions in skating’ and ‘dermatoses in skaters’. Case reports and review articles were included from the dermatology, sports medicine, general medicine and podiatry literature. In Part II of this review, we will focus on the infectious, inflammatory and cold-induced dermatoses of ice-skating athletes with a similar emphasis on diagnosis and treatment. The most common ice-skating-related dermatoses are listed in table I. 2. Mechanical Dermatoses 2.1 Skater’s Nodules
Skater’s nodules, also known as ‘athlete’s nodules’, or ‘double ankle bones’ are cutaneous ª 2011 Adis Data Information BV. All rights reserved.
711
Table I. Common dermatoses seen in figure skaters, ice-hockey players and speed skaters Mechanical dermatoses Skater’s (athlete’s) nodules/pads Malleolar bursitis Pump bumps (Haglund’s deformity) Piezogenic pedal papules Talon noir Skate bite/lace bite Friction bullae Corns (hard/soft) Callosities Onychocryptosis Skater’s toe/toenail Lacerations Infectious dermatoses Tinea pedis Onychomycosis Pitted keratolysis Warts (verrucae vulgaris and plantaris) Folliculitis Inflammatory dermatoses Allergic contact dermatitis Irritant contact dermatitis Palmoplantar eccrine hidradenitis Exercise-induced purpuric eruptions Urticaria Chromhidrosis Cold-induced dermatoses Chilblains (pernio) Cold panniculitis Frostnip/frostbite Raynaud phenomenon Physiological livedo reticularis
masses caused by repetitive friction, pressure and chronic irritation to the skin of the feet and ankles during skating (figure 1).[9] The nodules are frequently seen over the lateral malleoli, the lateral sides of both feet, the skin overlying the Achilles tendon or a combination of these locations. Inflammation of the nodules may cause significant pain that can interfere with skating performance. The surface of the nodules is often hyperkeratotic but smooth. Histopathological examination shows epidermal acanthosis, compact hyperkeratosis and thickening of the dermis with an increase in collagen fibres. The differential diagnosis includes Sports Med 2011; 41 (9)
Tlougan et al.
712
calluses, skater’s pads, hypertrophic scars, dermal neoplasms, granulomas and warts. Treatment of skater’s nodules consists of a reduction in chronic stimulation and the use of keratolytics and intralesional injections of triamcinolone acetonide or occluded ultrapotent topical corticosteroids. No treatment is likely to be effective unless measures are adopted to protect the skin from mechanical forces. Surgical excision is not routinely recommended due to potential complications from residual scarring, possible keloid formation and a high recurrence rate from continued friction after resuming skating.[9] Skater’s pads are callus-like, hyperkeratotic epidermal thickenings that are more superficial than skater’s nodules but also form as a reaction to chronic pressure, friction and irritation (figure 2). Development of pads may be prevented with the use of protective padding or stretching of the skate in the areas of greatest pressure. Once formed, the pads respond to paring of the skin in combination with intralesionally injected or topically applied corticostosteroids and creams, lotions or gels containing urea or other keratolytic agents. Shear forces and contact pressures endured by some competitive skaters may be strong enough to cause malleolar bursitis. Although aseptic trau-
Fig. 2. Skater’s pads are hyperkeratotic epidermal thickenings over the Achilles tendon with accentuated skin lines (as shown in the figure). Skater’s pads can also occur concurrently with ‘pump bumps’.
matic bursitis of the medial malleolus is more common, lateral bursitis may also develop in some skaters. This condition, which presents as an intermittently painful subcutaneous mass over bony prominences, results from damage of soft tissue by exertional pressures from the skate, that eventuates in the formation of an adventitious bursa (malleoli normally do not have bursae) at the site of highest stress. Chronic, repeated stimulation produces symptomatic inflammation that can interfere with the athlete’s training. Usually, only non-surgical interventions, such as stretching of the skate, protective padding, intrabursal corticosteroid injections, bursa aspirations or rest are necessary. Septic bursitis, most often caused by Staphylococcus aureus, is a serious complication that may require surgical debridement and intravenous antibiotics.[3] 2.2 Pump Bumps
Fig. 1. Erythematous indurated nodules on the lateral ankle of this international figure skating competitor, characteristic of Skater’s nodules or double ankle bones. A pump bump can also be appreciated as an erythematous nodule on the back of the heel.
ª 2011 Adis Data Information BV. All rights reserved.
Pump bumps, caused by Haglund’s deformity of the calcaneal tuberosity, present as a bony enlargement located on the back of the heel that occurs when the soft tissue overlying the Achilles tendon becomes inflamed from repetitive friction between the bony prominence and the back of a Sports Med 2011; 41 (9)
Ice-Skating Dermatoses Part I: Mechanical
skate. This can occur if the back of the skate is too rigid or if the boot heel is too wide, which can cause excessive movement of the foot vertically. A painful bursitis can also develop at the site of this chronic irritation. Pump bumps likely have a hereditary component whereby individuals with a high-arched foot, ‘tight’ Achilles tendon or a tendency to walk on the outside of the heel, are more prone to develop the condition, particularly in the context of rigorous ice-skating and/or the use of poorly fitted skates. Clinical presentation is characterized by the presence of a noticeable inflamed bump on the back of one or both heels and tenderness in the area where the Achilles tendon meets the calcaneal tuberosity. If necessary, x-rays can be used to better visualize the structure of the heel bone. Initial treatment is non-surgical and is aimed at reducing bursal inflammation. These include oral anti-inflammatory medications, ice packs and exercises aimed at stretching the heel cord. Skate modifications may also be useful including heel lifts in athletes with high arches, heel pads and custom orthotics. Skaters with this condition should also avoid wearing shoes with a rigid heel back as well as avoid running on hard surfaces and uphill during off-ice training.[1,4,10] 2.3 Piezogenic Pedal Papules
Piezogenic pedal papules are skin-colored herniations of the subcutaneous fat into the dermis formed by mechanical forces generated by high-impact surface collisions during jump landings in figure skating or sustained weight bearing from long-distance running (figure 3).[11] The majority of such papules are asymptomatic, but sometimes pain may be caused by the occlusion of blood vessels or impingement of nerves when the adipose tissue herniates.[11] The papules are typically located on the medial or lateral aspects of both heels, and may not be visible unless the athlete is examined in the standing position. Treatment of both asymptomatic and painful piezogenic pedal papules include surgery, heel cups, compressive stockings, a reduction in prolonged standing and weight bearing or trauma, weight loss and multiple injections of betamethaª 2011 Adis Data Information BV. All rights reserved.
713
*
Fig. 3. Skin-coloured papules (indicated by an *) representing herniations of subcutaneous fat into the dermis on the medial heel accentuated by weight-bearing, classic for piezogenic pedal papules.
sone and bupivacaine, but none of these treatments has proven to be consistently effective.[11,12] 2.4 Talon Noir
Talon noir, also known as black heel or calcaneal petechiae, is a form of traumatic purpura that is common in ice-hockey players due to accelerated stops and starts required by the sport, and in figure skaters who perform jumps (figure 4). These movements produce shear forces that can rupture tiny blood vessels within the dermis, resulting in discrete, haemorrhagical brown or blueblack macules.[13] The cutaneous intracorneal collections of blood commonly appear over the lateral borders of the heel, midfoot, soles and toes. The petechiae may be single or multiple, diffuse or punctiform and can cover larger surface areas.[14] Histological evaluation shows deposition of blood, visible as focal deposits of eosinophilic, amorphous material in the stratum corneum.[14] Some extravasated erythrocytes and scattered hemosiderophages may also be present in the papillary dermis.[14] The diagnosis is confirmed by a ‘scratch test’, which consists of paring the horny layer of the affected area with a scalpel to partially reduce the black purpuric area without bleeding. If superficial, the entire haemorrhage can be scraped away. Dermatoscopy is another useful diagnostic tool that helps to differentiate these lesions from melanoma. The characteristic Sports Med 2011; 41 (9)
Tlougan et al.
714
Fig. 4. Brown-black macule on the distal fifth toe. The diagnosis of talon noir was confirmed by the scratch test, in which the superficial hemorrhage was scraped away using a number 15 scalpel.
dermatoscopic feature is the presence of a redblack homogeneous pattern of pigmentation, often bordered by isolated red-black globules.[14] A parallel ridge pattern occurs in 40% of cases whereas parallel furrow or fibrillar patterns are rare.[15] Any suspicion for melanoma should prompt a biopsy.[13] Talon noir usually resolves without intervention within 2 to 3 weeks.[13,14] Skin lubrication, heel cups, change of footwear, wearing additional pairs of thick socks and temporary cessation of training may be valuable in preventing or expediting resolution of these lesions.[13] 2.5 Skate Bite/Lace Bite
Extensor or anterior tibialis tenosynovitis, also referred to as skate bite or lace bite, is caused by heightened pressure from an inflexible skate tongue on the anterior ankle, which leads to formation of a midline anterior tibialis callus or inflammatory pseudonodules distributed along the extensor hallucis longus tendon that runs from the anterior ankle to the base of the great toe.[1] This condition develops in ice-hockey players and ª 2011 Adis Data Information BV. All rights reserved.
figure skaters with new or very old skates, in which lateral slippage and compression of the skate tongue across the top of the foot and the ankle occur with dorsiflexion.[16] The athlete typically complains of pain and swelling around the dorsal aspects of both feet near the tongue of the skate. The treatment of skate bite is centred on reducing damage by the skate tongue that is compressing the foot. Proper lacing techniques, such that the tongue is maintained in a neutral position, should be discussed, and protective padding, such as Bunga Pads, should be worn outside the ice hockey or skating sock to lessen irritation with motion. Anti-inflammatory medications and ice compresses reduce inflammation. Skaters can also prevent the risk of skate bite by increasing the pliability of the skate tongue prior to skating. This can be accomplished by placing a piece of foam rubber between the top part of the ankle and the skate tongue or by loosening the top portion of the skate. Skates can also be outfitted with specialized expandable or ‘double felt’ tongues that innately provide more padding. Knowledge of the optimized ankle axis position reduces friction during dorsiflexion.[1,17,18] It is valuable to refer the athlete to a skating shop professional who can customize boots, as using makeshift padding can adversely impact the overall fit of the skate. If irritation occurs posteriorly, Achilles tendonitis may also develop but treatment is aimed at relieving pressure on the posterior aspects of the boot. 2.6 Friction Bullae
Friction vesicles and bullae are the most common complaint of marathon runners (figure 5). This type of blister also often afflicts figure skaters, ice-hockey players, speed skaters and other athletes who sustain acute friction on the soles of the feet in an environment of increased temperature, dryness or moisture. Such friction causes horizontal shearing forces that can split the stratum granulosum, leading the separated layers to fill with blood or tissue transudate.[13] Commonly affected sites include the tips of the toes, lateral feet, balls of the feet and posterior heels. Painful blisters can be lanced peripherally, taking care to keep the blister roof intact while draining the Sports Med 2011; 41 (9)
Ice-Skating Dermatoses Part I: Mechanical
Fig. 5. Friction bullae are the most common mechanical dermatosis seen in skaters.
fluid. The application of petrolatum, hydrocolloid dressings or moleskin can accelerate healing. Preventative measures to decrease the risk of developing blisters include use of well fitting skates with adequate space round the toes; on-slip insoles; proper lacing techniques; moisture-wicking socks; pads or padded socks; and application of topical agents, such as petrolatum, drying powders, aluminum chloride and adhesives.[13]
715
Soft corns are characterized by a white macerated appearance, which results from the absorption of moisture due to perspiration, and can be extremely painful (figure 7). They are found between any toes but most commonly between the fourth and fifth toes.[19] Paring with a scalpel reveals a central core in both types of corns, and helps differentiate corns from callosities and plantar verrucae. After paring down a callus, smooth translucent skin remains, in contrast to the base of warts, which show thrombosed capillaries. Enucleation of early corns temporarily reduces pain but continued debridement causes repetitive trauma that can result in subepidermal thickening, bursa formation, sinus tracking over joints and fibrous infiltration around sweat glands and nerve endings, all of which increase the risk of incurring pain. The most important measure to prevent corns is the use of well fitting skates. Orthotic management with silicone orthodigital splints or digital insoles lined with felt may be valuable. Over-the-counter products such as moleskins and toe separators reduce friction but do not eradicate the corn completely. Corticosteroid injections into the corn to atrophy the fibrous tissue provide variable benefits. There is also little evidence to support applications of silver nitrate or 50% pyrogallic acid to dessicate the corn.[20] 2.8 Callosities
Calluses or callosities are broad hyperkeratotic papules and plaques of even thickness, typically
2.7 Corns
A corn is a discrete hyperkeratotic papule with a central conical deposit of keratin that is produced in response to mechanical trauma and stress, typically related to ill-fitting footgear, abnormal repetitive foot manoeuvres and positions and high levels of physical activity. Corns are divided into two types, hard and soft corns. Hard corns, the most common type, usually develop on the dorsolateral aspect of the fifth toe or the dorsum of the interphalangeal joints of the lesser toes, and appear as dry, horny masses of hyperkeratosis with a hard central core (figure 6).[19] ª 2011 Adis Data Information BV. All rights reserved.
Fig. 6. Hyperkeratotic nodule on the dorsolateral aspect of the fifth toe representing a hard corn.
Sports Med 2011; 41 (9)
Tlougan et al.
716
Fig. 7. A soft corn, in its classic location between the fourth and fifth toes, can be extremely painful and results from the absorption of moisture due to interdigital perspiration.
crusting, purulence and protuberant granulation tissue develop at the nail fold and nail plate junction, resulting in severe pain (figure 8).[21] Mild to moderate onychocryptosis (stage 1), characterized by pain and erythema without purulent drainage, may be treated conservatively with soaks of warm, soapy water for 5–10 minutes three times daily, cotton wick elevation of the affected nail corner, topical antibiotic ointment to prevent bacterial infection and mid-tohigh potency steroid cream or ointment to reduce inflammation.[22] Moderate (stage 2) onychocryptosis with infection but no granulation tissue is treated with antibiotics and simple, partial avulsion of the nail plate and removal of any laterally pointing nail spicule with a nail splitter and haemostat.[22] This procedure, however, is effective only in 30% of cases.[22] Application of silver nitrate cautery to the granulation tissue is at best a temporary measure. Persistent stage 2 and severe (stage 3) onychocryptosis may ne-
arising under the metatarsal heads at sites of friction and pressure, often in response to secondary anatomic anomalies. Ice-sport athletes usually choose to maintain calluses given their protective benefits against additional trauma and blistering. Unlike corns, calluses have relatively undefined margins. If symptomatic, debridement reduces the irritant effect of the thick epidermis on sensitive nerve endings, and can be accomplished by soaking the area in warm water followed by paring with a pumice stone, file or scalpel blade. Application of creams with high concentrations of urea or salicylic acid can soften the callus. Like corns, prevention involves wearing appropriate skates and footgear, using padding as needed to reduce friction, and correcting biomechanical foot problems with orthotic devices.[19] 2.9 Onychocryptosis (Unguis Incarnatus)
Ingrown toenails (onychocryptosis) form when the nail groove is punctured by the corresponding nail plate. Invariably, this condition affects the nails of the first toes. Piercing of the skin initially causes erythema and localized tenderness but eventually ª 2011 Adis Data Information BV. All rights reserved.
Fig. 8. Stage 1 onychocryptosis resulting from the nail plate piercing the skin of the lateral nail fold causes localized tenderness and erythema and may be treated conservatively.
Sports Med 2011; 41 (9)
Ice-Skating Dermatoses Part I: Mechanical
cessitate subspecialty evaluation. These patients can require excision/avulsion of the lateral nail plate combined with lateral matrixectomy using phenol, sodium hydroxide, laser or an electrocautery unit.[22] Preventative measures include cutting toenails straight across to allow for a more equal distribution of forces across the toe box. Radiographs may be necessary if subungual exostosis is suspected. 2.10 Skater’s Toenail/Skater’s Toe
Skater’s toenail, also known as skater’s toe, jogger’s toe, jogger’s toenail and tennis toe, is characterized by black discolouration and distal callosity of the toenails (figure 9). The nail on the second toe, which is often the longest toe, is most commonly affected. The abnormal nail changes are caused by repetitive trauma as a result of wearing improperly fitted skates that enable thrusting of the toes into the toe box upon contact of the toe pick with the ice, or during quick stops and restarts. The toenail may become a pincer nail, or may develop onychauxis (thickening of the nail plate), a subungual haematoma or hyperkeratosis.[23] Onycholysis (separation of the nail plate from the underlying nail bed), thickening and secondary fungal infection of the nail may occur with continued injury.[13] Onychomycosis and subungual malignant melanoma are the two
Fig. 9. Skater’s toe, also known as jogger’s toe or tennis toe, is typically located on the second toe. This condition is characterized by discolouration of the toenail and results from repetitive thrusting of the toes into the toebox during manoeuvres utilizing the toe pick.
ª 2011 Adis Data Information BV. All rights reserved.
717
conditions most frequently considered in the differential diagnosis. Melanoma should be strongly suspected in the presence of Hutchinson’s sign, which is defined as pigmentation of the proximal or lateral nail fold usually from extension of longitudinal melanonychia.[13] Cutting toenails straight across and close to the skin is useful in preventing skater’s toe. Other strategies that can be used in the prevention of skater’s toe include tight lacing of the skate to keep the foot from moving forward without restricting circulation, as well as using a long, elevated anterior toe box to allow for dorsiflexion of the toes with minimal forward slippage.[13] 2.11 Lacerations
Figure skaters, ice-hockey players and speed skaters are at risk for sustaining lacerations caused by skate blades on the extremities, head and neck. Well publicized severe lacerations suffered by top-tier athletes in recent years have increased awareness of this potentially catastrophic injury. Recently, a female pair skater sustained a facial laceration during a side-by-side camel spin at a major international competition, a professional ice-hockey player incurred a neck laceration of the carotid artery during a game and an elite speed skater suffered a large thigh laceration during an Olympic trials race. The rigorous and risky athletic proficiency currently required for success at high levels of sport competition disposes skaters to acute injury, which often involves the skin. A study of onsite questionnaire interviews of 236 female and 233 male junior level competitive skaters reported that 25% of female skaters and 27.9% of male skaters had sustained an acute skating injury over the course of their careers.[24] Although the study stressed musculoskeletal injuries, the investigators found that 42.8% of the female skaters and 45.5% of the male skaters reported overuse syndromes, which also escalate the athlete’s risk of skin and soft tissue damage.[24] Lacerations were found to be more common in ice dancing, since the male and female skaters remain in close contact while performing skills that involve rapid changes of direction and hand Sports Med 2011; 41 (9)
Tlougan et al.
718
holds. Pair skaters also experience laceration injuries more frequently than individual skaters because their routines require lifts and throws. Nonetheless, hand lacerations are becoming increasingly common in singles skaters, especially female skaters, since more points may be awarded for difficult variations of camel, upright and layback spins, such as the Bielmann spin. This manoeuvre requires the skater to grab the blade with both hands while the foot is extended over and behind the head. As a result of increasingly intricate and technical routines, the risk of lacerations and other specific injuries has also risen among synchronized skaters.[25] A 2006 study of synchronized skating found that 91 (22%) of 412 injuries in 514 female and 218 male skaters were lacerations. Of these lacerations, 34 occurred on an upper extremity, 33 on a lower extremity, and 23 on the head. Sixty-five (73.9%) lacerations required sutures for closure, while the remainder were superficial enough to be treated with primary first aid.[25] Neck lacerations, though rare, are potentially fatal and may result in injury to nerves, blood vessels and airway components. Neck guards, or ‘neck laceration protectors’, were developed to prevent this injury, and are currently mandated by Hockey Canada and recommended by USA Hockey.[26] A recent survey of more than 26 000 ice-hockey players concluded that neck lacerations are infrequent and often superficial. 485 players (1.8%) reported being cut in the neck by a skate blade while playing ice hockey. Notably, 27% of all skaters who reported a neck laceration were wearing neck protection at the time of injury, indicating that these protective devices do not completely eliminate the risk of neck laceration.[26] Further research is needed to develop appropriate protective measures to avoid this potentially disabling and fatal injury. Short-track speed skaters also have a high prevalence of lacerations as skaters often pass each other at speeds that may result in high-speed collisions.[27] The frequency and severity of injuries is enhanced by the 16–18-inch skate blades worn by skaters, high-fleet speeds (over 50 km/h), and the relatively small ice track.[27] A published survey found that in one season, 64.2% of 95 elite ª 2011 Adis Data Information BV. All rights reserved.
speed skaters had sustained at least one major injury and 12.1% had experienced skin and soft tissue lacerations during competition.[27] In this study, the skaters reported that 25.9% of 108 injuries were lacerations. Lacerations of the skin below the knees were the most frequent injury, constituting 11.1% of all injuries and affecting 12.6% of speed skaters. The arms and hands were the second most common site of laceration, accounting for 6.5% of all reported injuries in these athletes.[27] The rates of these injuries may be lowered by modifying skate boot design and wearing racing suits made of cut-resistant material.[27] Methods for repairing lacerations include conventional skin suturing as well as the use of tissue glues, such as 2-octyl-cyanoacrylate, which may be substituted for sutures in the closure of lacerations and incisions.[28] 3. Conclusion Ice-skating athletes are subject to a range of skin conditions and tissue injuries on account of mechanical forces resulting from environmental conditions and athletic equipment. This set of dermatoses is diverse and includes frictioninduced nodules (skater’s nodules, skater’s pads, malleolar bursitis, and pump bumps), traumatic fat herniations (piezogenic pedal papules), traumatic purpura (talon noir), extensor tendonitis (skate bite), friction bullae, hyperkeratotic papules (corns and calluses), head and neck lacerations, and a range of nail abnormalities (onychocryptosis and skater’s toe). We conclude that awareness and prevention are paramount to managing these conditions, particularly in the highly competitive athlete. We are optimistic about continued research into skate design and protective padding and, each year, new and improved devices are developed to minimize friction-induced cutaneous injuries as well as to protect against and repair acute skin injuries such as lacerations. In the second part of this 2-part series, we will continue our discussion of skin conditions in iceskating athletes, focusing on dermatoses relating to infectious, inflammatory and cold-induced etiologies, with a similar emphasis on clinical presentation, diagnosis, treatment and prevention. Sports Med 2011; 41 (9)
Ice-Skating Dermatoses Part I: Mechanical
Acknowledgements The authors have received no funding and have no conflicts of interest.
References 1. Jaworski CA, Ballantine-Talmadge S. On thin ice: preparing and caring for the ice skater during competition. Curr Sports Med Rep 2008; 7 (3): 133-7 2. Bradley MA. Prevention and treatment of foot and ankle injuries in figure skaters. Curr Sports Med Rep 2006; 5: 258-61 3. Brown TD, Varney TE, Micheli LJ. Malleolar bursitis in figure skaters. Indications for operative and nonoperative treatment. Am J Sports Med 2000; 28 (1): 109-11 4. Lipetz J, Kruse R. Injuries and special concerns of female figure skaters. Clin Sports Med 2000; 19: 369-80 5. Davis MW. Figure skater’s foot. Minn Med 1979; 62: 647-8 6. Porter EB, Young CC, Niedfeldt MW, et al. Sport-specific injuries and medical problems of figure skaters. Wis Med J 2007; 106 (6): 330-4 7. Fortin JD. Competitive figure skating injuries. Pain Phys 2003; 6: 313-8 8. Smith AD, Ludington R. Injuries in elite pair skaters and ice dancers. Am J Sports Med 1989; 17: 482-8 9. Uchiyama M, Tsuboi R, Mitsuhashi Y. Athlete’s nodule. J Dermatol 2009; 36 (11): 608-11 10. Omey ML, Micheli LJ. Foot and ankle problems in the young athlete. Med Sci Sports Exerc 1999; 31 (7): S470-86 11. Redbord KP, Adams BB. Piezogenic pedal papules in a marathon runner. Clin J Sport Med 2006; 16 (1): 81-3 12. Doukas DJ, Holmes J, Leonard JA. A nonsurgical approach to painful piezogenic pedal papules. Cutis 2004; 73 (5): 339-40, 346 13. Mailler-Savage EA, Adams BB. Skin manifestations of running. J Am Acad Dermatol 2006; 55 (2): 290-301 14. Urbina F, Leon L, Sudy E. Black heel, talon noir or calcaneal petechiae? Australas J Dermatol 2008; 49 (3): 148-51 15. Zalaudek I, Argenziano G, Soyer HP, et al. Dermoscopy of subcorneal hematoma. Dermatol Surg 2004; 30 (9): 1229-32
ª 2011 Adis Data Information BV. All rights reserved.
719
16. Janowicz RA. How to evaluate figure skating injuries. Podiatry Today 2006; 19 (4): 64-74 17. Bruening DA, Richards JG. The effects of articulated figure skates on jump landing forces. J Appl Biomech 2006; 22: 285-95 18. Bruening DA, Richards JG. Optimal ankle position for articulated boots. Sports Biomech 2005; 4 (2): 215-25 19. Freeman DB. Corns and calluses resulting from mechanical hyperkeratosis. Am Fam Physician 2002; 65 (11): 2277-80 20. Yates B. Management of the sports patient. In: Turner W, Merriman L, editors. Clinical skills in treating the foot. Philadelphia (PA): Elsevier/Churchill Livingstone, 2005: 393-430 21. Martinez-Nova A, Sanchez-Rodriguez R, Alonso-Pena D. A new onychocryptosis classification and treatment plan. J Am Podiatr Med Assoc 2007; 97 (5): 389-93 22. Heidelbaugh JJ, Lee H. Management of the ingrown toenail. Am Fam Physician 2009; 79 (4): 303-8 23. Mailler EA, Adams BB. The wear and tear of 26.2: dermatological injuries reported on marathon day. Br J Sports Med 2004; 38 (4): 498-501 24. Dubravcic-Simunjak S, Pecina M, Kuipers H, et al. The incidence of injuries in elite junior figure skaters. Am J Sports Med 2003; 31 (4): 511-7 25. Dubravcic-Simunjak S, Kuipers H, Moran J, et al. Injuries in synchronized skating. Int J Sports Med 2006; 27 (6): 493-9 26. Stuart MJ, Link AA, Smith AM, et al. Skate blade neck lacerations: a survey and case follow-up. Clin J Sport Med 2009; 19 (6): 494-7 27. Quinn A, Lun V, McCall J, et al. Injuries in short track speed skating. Am J Sports Med 2003; 31 (4): 507-10 28. Eaglstein WH, Sullivan T. Cyanoacrylates for skin closure. Dermatol Clin 2005; 23 (2): 193-8
Correspondence: Dr Brook E. Tlougan, Columbia University, Department of Dermatology, Herbert Irving Pavilion, 12th Floor, New York, NY 10032, USA. E-mail:
[email protected]
Sports Med 2011; 41 (9)
Sports Med 2011; 41 (9): 721-740 0112-1642/11/0009-0721/$49.95/0
REVIEW ARTICLE
ª 2011 Adis Data Information BV. All rights reserved.
Amputees and Sports A Systematic Review Mihail Bragaru,1 Rienk Dekker,1 Jan H.B. Geertzen1 and Pieter U. Dijkstra1,2 1 From the Department of Rehabilitation Medicine, University Medical Centre Groningen, Groningen, the Netherlands 2 Department of Oral and Maxillofacial Surgery, University Medical Centre Groningen, University of Groningen, Groningen, the Netherlands
Contents Abstract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Review Methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Biomechanical Aspects and Athletic Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Cardiopulmonary Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Psychological Aspects and Quality of Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Sport Participation and Physical Functioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5 Sports Injuries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Limitations of the Current Systematic Review. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Abstract
721 723 724 725 725 725 734 734 735 737 737
Amputation of a limb may have a negative impact on the psychological and physical well-being, mobility and social life of individuals with limb amputations. Participation in sports and/or regular physical activity has a positive effect on the above mentioned areas in able-bodied individuals. Data concerning participation in sports or regular physical activity together with its benefits and risks for individuals with limb amputations are scarce. No systematic review exists that addresses a wide range of outcomes such as biomechanics, cardiopulmonary function, psychology, sport participation and sport injuries. Therefore, the aim of this article is to systematically review the literature about individuals with limb amputations and sport participation. MEDLINE (PubMed), EMBASE, CINAHL and SportDiscus were searched without time or language restrictions using free text words and MeSH terms. The last search date was 31 March 2010. Books, internet sites and references of included papers were checked for papers relevant to the topic under review. Papers were included if the research topic concerned sports and a minimum of ten individuals with limb amputations were part of the study population. Papers were excluded if they included individuals with amputations of body parts other than upper or lower limbs or more distal than the wrist or ankle, or if they consisted of case reports, narrative reviews, books, notes or letters to the editor.
Bragaru et al.
722
Title, abstract and full-text assessments were performed by two independent observers following a list of preset criteria. Of the 3689 papers originally identified, 47 were included in the review. Most of the included studies were older than 10 years and had cross-sectional designs. Study participants were generally younger and often had more traumatic amputations than the general population of individuals with limb amputations. Heterogeneity in population characteristics, intervention types and main outcomes made data pooling impossible. In general, sports were associated with a beneficial effect on the cardiopulmonary system, psychological well-being, social reintegration and physical functioning. Younger individuals with unilateral transtibial amputations achieve better athletic performance and encounter fewer problems when participating in sports compared with older individuals with bilateral transfemoral amputations. Regardless of their amputation level, individuals with limb amputations participate in a wide range of recreational activities. The majority of them were not aware of the sport facilities in their area and were not informed about available recreational activities. Sport prosthetic devices were used mostly by competitive athletes. For football, the injury rate and pattern of the players with an amputation were similar to those of able-bodied players. Individuals with limb amputations appear to benefit both physically and psychologically from participation in sports and/or regular physical activity. Therefore, sports should be included in rehabilitation programmes, and individuals with limb amputations should be encouraged to pursue a physically active life following hospital discharge.
Amputation of a limb may cause permanent disability and decreases mobility temporarily or permanently.[1] Individuals with limb amputations often see themselves as part of a special group that, according to able-bodied people, has special needs and requires additional attention.[2] These perceptions contribute to the relatively high depression and anxiety rates recorded amongst individuals with limb amputations, especially in the first 2 years after amputation.[3-5] Consequently, they will experience difficulties with social participation and in returning to everyday life. Individuals with limb amputations in general are in poor physical condition not only due to the amputation itself but also because of the illness preceding and leading to the amputation. In the US, about 82% of all lower and upper limb amputations are due to vascular conditions, whereas 16% of amputations are due to trauma.[6] The remaining 2% of amputations are necessary due to cancer or inflammation, or represent congenital birth defects. It has been predicted that the numª 2011 Adis Data Information BV. All rights reserved.
ber of individuals with limb amputations will increase as a consequence of the population’s increasing age and increasing incidence of diabetes mellitus and cardiovascular diseases.[7] In general, participation in sports or physical activities is important in maintaining physical fitness.[8,9] Lack of physical exercise is the most important determinant of a deteriorating physical state, often leading to coronary heart disease.[10] Health organizations recommend 30 minutes or more of moderately vigorous daily physical activity.[11] Because of the amputation and the underlying diseases persisting after amputation, individuals with limb amputations tend to be less active than the able-bodied.[12] Participation in sports and an active lifestyle are assumed to be important for individuals with limb amputations as they enhance psychological well-being, selfconfidence and coping behaviour.[13] Publications focusing on sports participation among individuals with limb amputations are generally characterized by a limited number of Sports Med 2011; 41 (9)
Amputees and Sports
participants, anecdotal reports and diverse outcome variables.[14-18] Even though there are a number of reviews[13,19,20] concerning some aspects of the sport participation of individuals with limb amputations, none of them address both upper and lower limb amputations, nor do they offer a full picture of all important variables associated with sport participation, such as participation rate, psychosocial modifications or injury rate. A complete overview may help professionals working in the sector of rehabilitation of individuals with limb amputations to evaluate sports or a physical training programme as part of a treatment programme for their patients, and to better understand the benefits and risks of sports participation for this group. Therefore, the aim of this article was to systematically review the literature on participation in sports and/or physical activity among individuals with upper and/or lower limb amputations and to identify their biomechanical characteristics, cardiopulmonary function, psychological well-being, sport participation, and physical functioning and injury characteristics. 1. Review Methods For this systematic review, sports were defined as ‘‘an activity involving physical exertion, with or without game or competition elements, with a minimal duration of half an hour and where skills and physical endurance are either required or to be improved’’.[21] Physical activity was defined as ‘‘any bodily movement produced by skeletal muscles that require energy expenditure’’.[22] Four databases were searched: PubMed, EMBASE, CINAHL and SportDiscus. The search strategy used consisted of a combination of database-specific MeSH terms, free text, ‘wild cards’ (words truncated by using ‘‘*’’) and Boolean operators (‘‘AND’’, ‘‘OR’’, ‘‘NOT’’). No time or language restrictions were applied and the
723
search was structured into two parts. One part concerned papers that related to amputations, while the second part concerned papers that related to sports and physical activity. The two parts of the search were combined using the Boolean operator ‘‘AND’’. The search details are presented in table S1 online in the Supplemental Digital Content (SDC) 1, http://links.adisonline.com/SMZ/A6. Sports were only searched as MeSH headings and as a general free text word, not by means of specific types of sports, such as running, cycling or basketball. All retrieved papers were combined in a single database and duplicates were removed. The most recent search date was 31 March 2010.1 Papers were selected for this review in three stages after evaluation of the title, abstract and full text. Papers were included if the research topic was sports or physical activity and a minimum of ten individuals with limb amputations were part of the study population. Papers were excluded if they concerned minor amputation (distal to the wrist or ankle), amputation of body parts other than upper and lower limbs (e.g. ear, breast) or endoprostheses. In addition, case reports, narrative reviews, editorials, notes and letters to the editor were excluded. If, after title and abstract assessment, the paper’s inclusion or exclusion remained questionable, the paper was included in the next selection stage. References of papers selected for inclusion in the final assessment stage of the review were checked for relevant citations, which were later retrieved and assessed in the same way. Each assessment was performed by two independent observers. If the observers were not fluent in the language of the paper, a native speaker translated the paper into English with the two observers present. In case of assessor disagreement, a consensus meeting was held. If disagreement persisted, a third observer gave a binding verdict. The full text of a paper was assessed if the paper fulfilled the following inclusion criteria: a minimum of ten (1) individuals
1 Authors’ note: the literature search was updated on 21 February 2011. Following the same inclusion/exclusion criteria as described in section 1, we identified four more studies eligible for inclusion in the current systematic review.[33,50,66,67] These studies were added to table I under the relevant characteristics analyses. The results of these studies did not influence the conclusion of this review and therefore they were not brought up for discussion.
ª 2011 Adis Data Information BV. All rights reserved.
Sports Med 2011; 41 (9)
Bragaru et al.
724
with limb amputations (2) were part of the study population and sport or physical activity was considered (3). Methodological quality was based on the assessment of the following criteria: reporting of inclusion (1) and exclusion (2) criteria; the numbers or percentages of males and females (3); age (4) [as mean and standard deviation or median and inter-quartile range]; cause of amputation (general description of cause [5] and exact number [6] per cause) and level (7); and side (8) of amputation.
they did not fulfil the inclusion criteria, leaving 47 for final inclusion in this systematic review (figure 1). Inter-observer agreement, expressed as Cohen’s Kappa, for the full-text assessment of the 47 included papers was 0.83 (95% CI 0.78, 0.89). The quality of the included papers was moderate, with only four papers fulfilling all eight criteria. The frequency distribution of the methodological quality of all studies is presented in figure S1 online in the SDC 1. In general, there was substantial heterogeneity in interventions, population characteristics and main outcomes between the studies. In order to provide structure to the findings, main outcomes were organized into five categories: Biomechanical aspects and athletic performance: papers in this category had to present data regarding forces or any other biomechanical variables of the subjects or activity. Cardiopulmonary function: papers in this category had to present biometric data recorded during or after physical activity or sports.
2. Results A total of 3689 papers were identified, of which 895 were duplicates. After title and abstract assessment, 85 full-text papers were selected for further assessment. As a result of reference checking, 29 additional potentially relevant papers were identified. In total, 17 papers could not be retrieved for bibliographic reasons or because there was no complete paper available. After fulltext assessment, 50 papers were excluded because 1217 PubMed
90 CINAHL®
1577 EMBASE
805 SportDiscus®
3689 895
Duplicates
2794 2594
Title assessment
200 115
Abstracts assessment
85 Reference checking
29 114 17
Unavailability
97 50
Full-text assessment
47 Fig. 1. Flow chart of the systematic review.
ª 2011 Adis Data Information BV. All rights reserved.
Sports Med 2011; 41 (9)
Amputees and Sports
Psychological aspects and quality of life: papers in this category had to present data regarding psychological aspects and quality of life. Sports participation and physical functioning: papers in this category had to present data about sports participation and modifications in physical functioning following participation in sports or physical activities. Sports injuries: papers in this category had to present data about sports injuries. 2.1 Biomechanical Aspects and Athletic Performance
Ten studies analysed the biomechanical aspects of swimming,[23] running[24-26] and long jump,[27-32] and athletic performance of individuals with upper and lower limb amputations. Video cameras, force plates or Doppler devices were used to measure step length and rate, joint angles, ground reaction force and speed, among other variables. An overview of these papers is presented in table I. Young individuals with unilateral transtibial amputations who were provided with prostheses and were adequately trained were able to run.[24,25] Runners with lower limb amputations demonstrated a difference between the prosthetic and nonprosthetic limbs regarding step length and vertical, mediolateral and horizontal displacement of the centre of mass. The prosthetic and nonprosthetic limbs also differed in these variables from those of able-bodied individuals.[26,32] Long jumpers with transtibial amputations jumped further than those with transfemoral amputations.[27,31] Long jumpers with transtibial amputations who used their prosthetic limb for take-off had a shorter last step and a lower vertical velocity at touchdown than did those jumpers using their sound limb for take-off.[28] Runners with lower limb amputations[25,26] and swimmers with upper limb amputations[23] increase their speed by increasing their pace rather than their step or stroke length. 2.2 Cardiopulmonary Function
Twelve studies analysed cardiopulmonary function in relation to sports or physical activity among ª 2011 Adis Data Information BV. All rights reserved.
725
individuals with limb amputations.[1,17,34-43] Training equipment such as an exercise cycle or rowing ergometer was used. An ECG, spirometer, sphygmomanometer and Doppler device. were used to measure maximal oxygen intake (VO2max), heart rate, blood pressure, anaerobic threshold, and maximum power output. An overview of these papers is presented in table I. The general physical condition of individuals with limb amputations is worse than the reference values for able-bodied people of similar age.[1,34] Nevertheless, individuals with limb amputations have better aerobic and anaerobic power outputs than do individuals with other locomotor disabilities.[39] Participation in sports or physical activity has beneficial influences on the cardiopulmonary system, muscle force and body mass of individuals with limb amputations.[17,35,38,40] The rehabilitation time of individuals with limb amputations was shorter when physical training was part of their rehabilitation programme.[41]
2.3 Psychological Aspects and Quality of Life
Six studies analysed the relationship between sport participation and the psychological aspects and quality of life of individuals with limb amputations.[44-49] Questionnaires or interviews were used to measure motivation to participate in sports, self-esteem and perceived benefits and barriers in physically active individuals with limb amputations. An overview of these papers is presented in table I. Quality of life and self-esteem of individuals with limb amputations who participated in sports and physical activities were higher than those of people with limb amputations who did not participate in these activities.[44,47] Sports and physical activity helped these individuals to increase their number of social contacts and their knowledge about sporting equipment that could facilitate their participation in sports. It also helped them to accept their disability and to improve their motor skills.[48,49] Participation in sports and/or physical activity decreased following the amputation as a direct result of physical constraints and accessibility issues.[49] Sports Med 2011; 41 (9)
726
ª 2011 Adis Data Information BV. All rights reserved.
Table I. Studies analysing characteristics for individuals with limb amputations Study (y)
QS S (n); in/ex
Gender M/W; agea
Amputation characteristics level
cause (G/S) side
Study design Analysis aim
Results
Biomechanical aspects and athletic performance 4
10; N/N
9/1; 39
10 TT
N/N
10 unilateral; LL
CS
Running gait characteristics
60% of young individuals with unilateral TT amputations who used a prosthesis were able to run at speeds ranging from 2.7 to 8.2 m/sec. Speed increase is related to an increase in stride rate Step length of the ProsL is directly related to speed increase
Engsberg et al.[24] (1993)
4
21 (221); N/N
17/4; 11
21 TT
N/N
21 unilateral; LL
CS
Running characteristics of AB and children with TT amputations
Non-ProsL generated greater vertical, anteroposterior and mediolateral forces as compared with ProsL of AB. With increasing speed, the non-ProsL generated greater forward propulsion than the ProsL
Gavron et al.[26] 4 (1995)
12; N/N
12/0; N
12 TT
N/N
12 unilateral; LL
CS
Sprinting characteristics of individuals with TT amputations
ProsL and non-ProsL were asymmetric with respect to stride length and time and vertical displacement of centre of mass. Non-ProsL stride contributed more to horizontal displacement than ProsL stride
Nolan et al.[27] (2000)
4
16; N/N
16/0; N
8 TT/ 8 TF
N/N
16 unilateral; LL
CS
Take-off characteristics during long jumping
Athletes with TT amputations jumped further, had a faster approach speed and a lower centre of mass compared with athletes with TF amputations
Simpson et al.[32] (2001)
4
23; N/N
17/6; (18–36)
20 TT/ 3 TF
N/N
22 unilateral/ 1 bilateral; LL
CS
Locomotor characteristics of long jump
ProsL of individuals with TT amputations had a greater step length compared with their non-ProsL More proximal amputations generated larger interlimb kinematic asymmetry. Individuals with TF amputations increased speed by increasing step length of their non-ProsL. Individuals with TT amputations increased speed by increasing step length of their ProsL
Patritti et al.[31] (2005)
3
34; N/N
11/9; N
20 TT/ 14 TF
N/N
N/LL
CS
Approach velocity of long jumpers
Individuals with TT amputations ran faster and jumped further than individuals with TF amputations
Nolan et al.[28] (2005)
3
14; N/N
14/0; N
7 TT/ 7 TF
N/N
N/LL
CS
Influence of take-off leg during the long jump
Athletes with TT amputations who took off from their ProsL were able to better control their downward velocity at touchdown and had a shorter last stride
Nolan et al.[29] (2006)
4
17; N/N
0/17; N
9 TT/ 8 TF
N/N
17 unilateral/LL
CS
Kinematic characteristics of women with TT and TF amputations during long jump
Approach velocity more strongly influenced the jumped distance of women with TT amputations than that of the women with TF amputations. Women with TF amputations had a lower centre of mass than the ones with TT amputations. Women with TF amputations had greater joint angles (hip, knee and leg) on all jump phases compared to the athletes with TT amputations
Continued next page
Bragaru et al.
Sports Med 2011; 41 (9)
Enoka et al.[25] (1982)
Study (y)
QS S (n); in/ex
Gender M/W; agea
Amputation characteristics level
cause (G/S) side
Study design Analysis aim
Results
Nolan et al.[30] (2007)
4
13; N/N
13/0; N
6 TT/ 7 TF
N/N
13 unilateral/LL
CS
Kinematic and temporal characteristics of long jump
Athletes with TF amputations had slower horizontal velocity and shorter stride length on the second to last and last strides compared with athletes with TT amputations. Athletes with TF amputations lowered their centre of mass in the last stride more than athletes with TT amputations. Athletes with TF amputations had greater joint angles (hip, knee and leg) on all jump phases compared with athletes with TT amputations
Osborough et al.[23] (2009)
4
13; N/N
3/10; [16.9 – 3.1]
13 ED
N/N
13 unilateral/UL
CS
Swimming characteristics of individuals with UL amputations related to their anthropometric characteristics
SF was related to maximum swimming speed (r = 0.72). SF was related to biacromial breadth (r = 0.86), shoulder girth (r = 0.64) and upper-arm length (r = 0.58)
Osborough et al.[33] (2010)
6
13; N/N
3/10; [16.9 – 3.1]
13 ED
Non-PVD/Y 13 unilaterial/UL
CS
Swimming characteristics of individuals with UL amputations related to their inter-arm coordination
Swimming speed (r = 0.59) and SF (r = 0.66) were related to the coordination of the amputated arm
HR and BP decreased in the majority of individuals with limb amputations after 2 wk of indoor training Inspiration frequency and muscle force of hand, shoulder and back increased in the majority of individuals with limb amputations in the second day of skiing
Amputees and Sports
ª 2011 Adis Data Information BV. All rights reserved.
Table I. Contd
Cardiopulmonary function 19; N/N
16/3; N
6 TT/ 11 TF
N/N
17 unilateral/ 2 bilateral; LL
Long
Changes of cardiopulmonary and muscle force characteristics after training
van Alste et al.[34] (1985)
7
39; Y/N
28/11; 67
10 TT/ 13 TF/ 11 KD
PVD/Y
35 unilateral/ 4 bilateral; LL
CS
HR during PA and the Individuals with limb amputations achieved 80% of threshold of prosthetic the predicted HRmax value for AB of same age. ambulation Work capacity of 60 Watts was the threshold of prosthetic ambulation
Pitetti et al.[35] (1987)
6
10; Y/N
N; 39
4 TT/ 3 TF
Non-PVD/Y 8 unilateral/ 2 bilateral; LL
Longitudinal
Changes of cardiopulmonary and work capacity characteristics after aerobic training
HR during rest and exercise decreased. HR during volitional exhaustion increased. . VO2max increased. Wmax increased
Continued next page
727
Sports Med 2011; 41 (9)
Tomaszewska 3 et al.[17] (1965)b
Study (y)
728
ª 2011 Adis Data Information BV. All rights reserved.
Table I. Contd QS S (n); in/ex
Gender M/W; agea
Amputation characteristics level
cause (G/S) side
Results
Alaranta et al.[36] (1988)
5
10; N/N
9/1; 35 (23–57)
3 TT/3 TF, 4 UL
Non-PVD/Y N; 6 LL/4 UL
CS
Suitability of tests evaluating skiers’ physical capacity
Most tests were not suitable for skiers with limb amputations. Recommended tests: rowing ergometer for individuals with TF amputations; ‘walking with sticks’c for individuals with TT amputations and/or individuals with unilateral UL amputations, treadmill running for individuals with bilateral UL amputations
Chin et al.[37] (1997)
5
53; Y/N
40/13; 42
11 TT/ 37 TF/ 5 HD
N/N
CS
Validity of the one-leg cycling ergometer test in determining AT
Kurdibaylo and Bogatykh[38] (1997)
5
78 (90); N/N
78/0; N
34 TT/ 37 TF
Non-PVD/Y 61 unilateral/ 17 bilateral; LL
Longitudinal
Changes of cardiovascular characteristics and body mass after swimming pool exercises
BP decreased as a result of suppressing the influence of the sympathetic nervous system. Body mass decreased
Hutzler et al.[39] 4 (1998)
10 (50); Y/N
10/0; [39d – 9.2]
N
N/N
CS
Difference in power outputs between individuals with limb amputations and individuals with other types of physical disabilities
Individuals with limb amputations had better aerobic and anaerobic power outputs and fatigue indices than individuals with other types of physical disabilities
Chin et al.[40] (2001)
6
24; Y/N
N; [41 – 18.4]
24 TF
Non-PVD/Y 24 unilateral/LL
Longitudinal
Changes in cardiopulmonary characteristics after endurance training based on AT
Chin et al.[1] (2002)
7
31 (49); Y/N
18/13; [26 – 5.7]
10 TT/ 20 TF/ 1 KD
Non-PVD/Y 31 unilateral/LL
Longitudinal
Changes of cardiopulmonary and physical fitness after physical training compared with AB
Kobzev and Khramov[41] (2002)
4
18; N/N
18/0; (19–44)
N
Non-PVD/Y N/LL
CS
Influence of PA level on BP, HR and rehabilitation time
53 unilateral; LL
N/LL
. AT correlated (r = 0.66) with predicted VO2max One-leg cycle ergometer is valid in AT determination
. VO2max increased for the endurance group compared with pre-training and to a control group. AT increased for the endurance group compared with pre-training and to a control group
. VO2max increased to the level of AB. AT increased to the level of AB. Wmax increased to the level of AB
Higher PA level was associated with lower BP and HR Higher PA level was associated with a shorter rehabilitation time
Continued next page
Bragaru et al.
Sports Med 2011; 41 (9)
Study design Analysis aim
Study (y)
QS S (n); in/ex
Gender M/W; agea
Amputation characteristics level
cause (G/S) side
Study design Analysis aim
Results
Huonker et al.[42] (2003)
3
17 (125); Y/N
N; [34 – 11.5]
17 TT
N/N
17 unilateral/LL
CS
Blood flow and diameter of the common femoral artery of the ProsL were smaller compared to non-ProsL and to untrained AB
Chin et al.[43] (2006)
7
49; Y/N
34/15; [67 – 5.6]
43 TF/ 6 HD
19 PVD, 30 non-PVD/Y
49 unilateral/LL
CS
Vascular characteristics of physically active individuals with limb amputations and AB
Amputees and Sports
ª 2011 Adis Data Information BV. All rights reserved.
Table I. Contd
. To identify the Exercise intensity of ‡50% VO2max was the threshold threshold of prosthetic of prosthetic ambulation ambulation
Psychological aspects and quality of life
33 (161); N/N
19/14; N
N
N/N
N
CS
Relationship between self-esteem, locus of controle and PA of individuals with limb amputations
Physically active individuals with limb amputations had higher self-esteem than inactive individuals. Active men with limb amputations had a lower locus of control than inactive men. Active women with limb amputations had a higher locus of control than inactive women
Mastro et al.[45] 3 (1996)
22 (138); Y/N
17/5; 29
N
N/N
N
CS
Athletes’ attitudes towards each other and their ranking of disability preferences
Individuals with limb amputations were ranked highest in the hierarchy of preferences due to the lowest perceived disability
Wetterhahn et al.[46] (2002)
6
56; Y/N
36/20; N
34 TT/ 22 TF
8 PVD, 48 nonPVD/Y
48 unilateral/ 8 bilateral/LL
CS
Relationship between PA and body image
MBSRQ and ABIS scores were significantly higher (p = 0.0001, p = 0.01, respectively) for active individuals with limb amputations compared with an inactive group. A relationship exists between PA level and body image of individuals with limb amputations. No data presented to substantiate the nature of the relationship between PA level and body image. A chronic illness did not influence body image in the active group
Lowther et al.[47] (2002)
2
15; Y/N
15/0; (19–28)
N
N/N
N
CS
Relationship between athletic performance, self-efficacy and psychological skills
High self-efficacy was associated with successful athletic performance. The usage of activationf and relaxation skills was associated with high self-efficacy and successful athletic performance
Continued next page
729
Sports Med 2011; 41 (9)
Valliant et al.[44] 1 (1985)
Study (y)
730
ª 2011 Adis Data Information BV. All rights reserved.
Table I. Contd QS S (n); in/ex
Gender M/W; agea
Amputation characteristics level
cause (G/S) side
Study design Analysis aim
Results
Sporner et al.[48] (2009)
1
57 (132); Y/N
115/17g; [47.4g – 13.4]
N
N/N
N
CS
Relationship between participation in organized sport events and psychosocial characteristics
Participation in organized sporting events increased the knowledge of sporting equipment (92%), mobility skills (84%) and disability acceptance (84%) 98% felt that participation in organized sporting events improved their lives. Increased numbers of friends, interaction with other disabled people and ability to be competitive were seen as benefits of participation in organized sporting events. Participants in organized sporting events had decreased cognitive and physical limitations (CHART) compared with non-participants For 63% of the participants, taking part in organized sport events represents their only sporting activity
Couture et al.[49] (2010)
7
15; Y/N
8/7; [65.1 – 13.9]
11 TT/ 4 TF
PVD/Y
15 unilateral/LL
Longitudinal
Characteristics of leisure activities of individuals with LL amputations
Participation in leisure activities decreased following amputation. Leisure satisfaction of individuals with LL amputations was higher than ILP reference value Individuals with LL amputations encounter more constraints than AB in terms of functional abilities and accessibility when engaging in leisure activities
Tatar[50] (2010)
6
37; Y/N
25/12; N
18 TT/ 19 TF
4 PVD/Y
37 unilateral
CS
Relationship between participation in sports and body image of amputees
Participation in sports increases perceived body image of amputees
Sport participation and physical functioning
6
134; Y/N
103/31; 47
87 TT/ 27 TF
65 PVD, 69 nonPVD/Y
114 unilateral/ 20 bilateral/LL
CS
61% were active in sports Level of sport participation, preferred Fishing and swimming were most frequently performed. 6% used sport prostheses sports and prosthetic use
Kegel et al.[52] (1978)h,i
6
134; Y/N
103/31; 47
81 TT/ 3 KD/ 19 TF/ 5 HD
65 PVD, 69 nonPVD/Y
114 unilateral/ 20 bilateral/LL
CS
Functional capabilities and population characteristics of active individuals with limb amputations
Younger individuals with TT amputations were more active than older individuals with TF amputations
Continued next page
Bragaru et al.
Sports Med 2011; 41 (9)
Kegel et al.[51] (1977)h,i
Study (y)
QS S (n); in/ex
Gender M/W; agea
Amputation characteristics level
cause (G/S) side
Results
Kegel et al.[53] (1980)i
6
100; Y/N
85/15; 45
58 TT/ 25 TF
29 PVD, 71 nonPVD/Y
83 unilateral/ 17 bilateral/LL
CS
Characteristics of sport participation: level, barriers, prosthesis use and complaints
Younger individuals with traumatic limb amputations were more active than older individuals with vascular limb amputations. Barriers to sports participation: pain, embarrassment, lack of special organized programmes for individuals with limb amputations, lack of awareness of the existing sport facilities (93%). Prosthetists did not want to modify the prosthesis according to the suggestions of individuals with limb amputations in 45% of cases
Medhat et al.[54] (1990)
4
131; Y/N
122/9; 58 (24–90)
82 TT/ 61 TF
47 PVD, 84 nonPVD/Y
N/LL
CS
Factors influencing ADL and sport participation
Individuals with TF amputations reported more problems in ADL than did individuals with TT amputations. Sport participation was problematic for individuals with both TT and TF amputations. Least problematic sports: canoeing and swimming
Pohjolainen[55] et al. (1990)
7
175; Y/N
127/48; [62 – 15.8]
93 TT/ 62 TF
142 PVD, 33 nonPVD/Y
155 unilateral/ 20 bilateral/LL
CS
60% used their prosthesis >12 h/day. 15% could walk Characteristics of prosthetic use, walking 2–3 h and 23% could walk more than 1 km. Ischemic pain restricted walking ability and factors influencing walking
Gailey[56] (1992)
5
1214; Y/N
Nj; Nj
Nj
242 PVD, 972 nonPVD/Y
Nj
CS
Nj Characteristics of recreational activities: type, return to, amount and barriers
Burger et al.[57] 7 (1997)
228; Y/N
191/37; [53.3 – 15.4]
114 TT/ 2 KD/ 108 TF/ 4 HD
NonPVD/Y
228 unilateral/LL
CS
Changes in sports participation and preferred recreational activities following amputation
Sport participation decreased. Preferred recreational activities changed towards ‘more energy efficient’ ones
Burger et al.[58] 7 (1997)
223; Y/N
187/36; [54 – 15.4]
115 TT/ 102 TF/ 2 KD/ 4 HD
NonPVD/Y
203 unilateral/ 20 bilateral/LL
CS
Influences of age, amputation level and time since amputation on walking, cycling and independence level
Younger individuals with TT amputations walked longer and were more likely to cycle than older individuals with TF amputations. A shorter time since amputation lead to a higher independence level
Legro et al.[59] (2001)
92; Y/N
79/13; [54.95 – 13.7]
58 TT
N/N
N/LL
CS
Preferred recreational activities and factors influencing the choice of activities
Fishing was the most frequently preferred recreational activity (n = 15). Gender, required energy level, ProsL impact load and age influenced choice of recreational activity
3
Continued next page
731
Sports Med 2011; 41 (9)
Study design Analysis aim
Amputees and Sports
ª 2011 Adis Data Information BV. All rights reserved.
Table I. Contd
Study (y)
732
ª 2011 Adis Data Information BV. All rights reserved.
Table I. Contd Gender M/W; agea
Amputation characteristics level
cause (G/S) side
Study design Analysis aim
Results
Rau et al.[60] (2007)
8
58; Y/Y
58/0; [37 – 10.9]
43 TT/ 15 TF
NonPVD/Y
58 unilateral/LL
RCT
Training programme was effective in increasing Changes of walking walking distance and speed, ProsL maximal load characteristics after and PCI short intensive physiotherapy for individuals with limb amputations compared with controls
Yazicioglu et al.[61] (2007)
8
24; Y/Y
24/0; [28 – 4.6]
24 TT
NonPVD/Y
24 unilateral/LL
CS
Differences in physiological and QoL characteristics of soccer players compared with controls
A difference in physiological characteristics was recorded, favouring players. QoL improved for players. with regards to pain, emotional role and fear of falling
Yari et al.[62] (2008)
8
46; Y/Y
21/25; [55.8 – 12.1]
31 HD/ 15 HP
6 PVD, 40 nonPVD/Y
46 unilateral/LL
CS
Activity level and mobility limitations of individuals with HD and HP amputations
39% participated in sports. Swimming, fitness, sailing and golf were the most practiced sports
Walker et al.[63] 7 (2009)
36 (62); Y/Y
21/15; 32.5k
36 TTk
NonPVD/Y
36 unilateral/LL
CS
Difference in outcome following fibular lengthening or amputation
No difference in sport participation between fibular lengthening patients and patients with amputation. No difference in sports activity between fibular lengthening patients, patients with amputation and control group. Patients with amputations scored significantly better than fibular lengthening patients on the job satisfiers content scale
Karmarkar et al.[64] (2009)
7
42; Y/Y
N; [42.11 – 16]
1 AD/ 20 TT/ 3 KD/ 14 TF/ 4 HD
13 PVD, 27 nonPVD/Y
28 unilateral/ 10 bilaterial/ 8 UL/34 LL
CS
Personal characteristics and functional performance related to mobility device in physical active veterans
Amputation level and degree of difficulty of the intended activity were related to participation in sports Prosthetic users with a more proximal amputation had more problems participating in sports compared with wheelchair users of same amputation level.
Kars et al.[65] (2009)
8
105; Y/Y
71/31; 58.7
1 AD/ 58 TT/ 13 KD/ 27 TF/ 5 HD/ 1 HP
42 PVD, 63 nonPVD/Y
101 unilateral/ 4 bilateral/LL
CS
Participation in sports of individuals with LL amputations
32% participated in sports. Participation in sports before the amputation was related to participation in sports following the amputation. Swimming, fitness and cycling were the most practiced sports. 42% complained about their prosthesis or sport organization and 80% of them found this problem hindering their participation in sports
Continued next page
Bragaru et al.
Sports Med 2011; 41 (9)
QS S (n); in/ex
Study (y)
QS S (n); in/ex
Gender M/W; agea
Amputation characteristics level
cause (G/S) side
Study design Analysis aim
Results Individuals with TT amputations are 40% less active than AB. Rehabilitation physicians significantly overestimate the activity levels of individuals with limb amputations
van den Berg- 8 Emons et al.[66] (2010)
18 (461); Y/Y
17/1; [56 – 13.13]
18 TT
9 PVD, 9 nonPVD/Y
18 unilateral/LL
CS
Activity level of individuals LL amputations
Bekkering et al.[67] (2011)
7
43 (82); Y/N
22/21; [16.1 – 4.4]
4 TT/ 11 KD/ 12 TF/ 16 RP
43 nonPVD/Y
43 unilateral/LL
CS
Difference in physical No difference in physical activity level between young activity between young adults who underwent limb-salvage interventions and the ones who underwent ablative surgery adults undergoing limb-salvage or ablative surgery
3
75; Y/N
72/3; 29 (18–44)
N
17 PVD, 58 nonPVD/N
N; 61 LL/14 UL
CS
Injury characteristics of amputee soccer players
Physical injuries appeared minor compared with emotional benefits induced by sport. 52% of individuals with limb amputations never sustained an injury while playing soccer. The injury pattern of individuals with limb amputations was similar to AB
Melzer et al.[69] 7 (2001)
32 (56); Y/N
32/0; 42.4
N
NonPVD/Y
32 unilateral; LL
CS
Contralateral knee osteoarthritis prevalence in individuals with limb amputations who play or do not play volleyball
Contralateral knee osteoarthritis was significantly more common among individuals with limb amputations compared with controls (p < 0.05). No difference in the prevalence of contralateral knee osteoarthritis was observed between individuals with limb amputations who played volleyball and the ones who did not playl
Bernardi et al.[70] (2003)
1
28 (227); Y/N
N; N
N
N/N
N
CS
Prevalence of SRMP
75% of individuals with limb amputations exhibited SRMP. Individuals with limb amputations are more likely to present SRMP than any other disabled group (adjusted OR = 15.4). Presence of SRMP was associated with a BMI between 24.6 and 30.9 (adjusted OR= 3.4) and more than 7 h/wk training (adjusted OR = 3.8)
Desmond et al.[71] (2008)
6
89; Y/N
62/27; ‡60m
55 TT/ 30 TF/ 4 KD
16 PVD, 73 nonPVD/Y
89 unilateral; LL
CS
Association between pain, prosthesis satisfaction and activity restriction
Pain was associated with prosthesis dissatisfaction Pain presence was not associated with activity restriction
Amputees and Sports
ª 2011 Adis Data Information BV. All rights reserved.
Table I. Contd
Sport injuries Kegel and Malchow[68] (1994)
a
Mean or (range) or [mean – SD].
b
Training programme was not standardized and the measurements were not distributed according to a standard schedule throughout the study duration.
d
Value for the total sample, including individuals with limb amputations.
e
A higher locus of control represents a more externalized person.
Continued next page
733
Sports Med 2011; 41 (9)
c Similar to ‘Nordic walking’.
Bragaru et al.
l
AB = able-bodied; ABIS = Amputee Body-Image Scale; AD = ankle disarticulation; ADL = activities of daily living; AT = anaerobic threshold; BMI = body mass index; BP = blood pressure; CHART = Craig Handicap Assessment Reporting Technique; CS = cross-sectional; ED = elbow disarticulation; G/S = general cause for amputation/specific numbers per cause; HD = hip disarticulation; HP = hemipelvectomy; HR = heart rate; HRmax = maximum HR; ILP = individual leisure profile; in/ex = inclusion/exclusion criteria reported; KD = knee disarticulation; LL = lower limb; M = men; MBSRQ = Multidimensional Body-Self Relations Questionnaire; N = no, data missing or criterion not fulfilled; non-ProsL = non-prosthetic limb; non-PVD = any cause of amputation other than peripheral vascular disease, including trauma, infection or cancer; OR = odds ratio; PA = physical activity; PCI = physiological cost index calculated using the values obtained from the 2 min test as follows: [HRfinish - HRrest]/speed; ProsL = prosthetic limb; PVD = peripheral vascular disease; QoL = quality of life; QS = quality score of the study; RCT = randomised controlled trial; RP = rotationplasty; S (n) = number of subjects with limb amputations (total number of participants); SF = stroke . frequency; SRMP = sport related muscle pain; TF = transfemoral; TT = transtibial; UL = upper limb; VO2max = maximal oxygen uptake volume; W = women; Wmax = maximal power output; Y = yes, data present or criterion fulfilled.
Authors state that the population sample was too small to sustain a ‘statistically reliable conclusion’.
k
m 65.2% were over 60 years of age.
Inconsistent data presented by the author.
Characteristics for the whole study population.
j
Possibly the same study population.
Similar results reported by the three studies for sport participation and use of special sport prostheses. i
Characteristics for the whole population (n = 132). g
h
The ability to increase energy. f
Table I. Contd
734
ª 2011 Adis Data Information BV. All rights reserved.
2.4 Sport Participation and Physical Functioning
Fifteen studies analysed associations between sport participation and/or physical activity and physical functioning of individuals with limb amputations.[51-65] A combination of self-developed and published questionnaires as well as specific tests addressing mobility outcomes were used as measurement tools. The main outcome variables were sport participation rate, the type of preferred physical activity, type and use of prosthesis and modifications of physical functioning following a physical training programme. An overview of these papers is presented in table I. From the included papers, it appears that between 11% and 61% of individuals with lower limb amputations participate in sports and/or physical activities.[51-53,57,62,65] The choice of which sports to take part in was influenced by gender, the specific energy requirement of the sport and the load on the prosthetic limb.[59,64] Fishing, swimming, golfing, walking and cycling were favoured sports. Younger individuals with unilateral transtibial amputations due to nonvascular causes were more active than older individuals with bilateral transfemoral amputations due to vascular causes.[53,57] A short but intensive physical training programme improved the walking distance and speed of individuals with traumatic lower limb amputations.[60] 2.5 Sports Injuries
Four studies analysed the sports injuries suffered by individuals with limb amputations.[68-71] Questionnaires were used to assess the injury rate and injury-related phenomena such as pain or activity restriction. An overview of these papers is presented in table I. The injury pattern and rate among individuals with limb amputations who play football (soccer) appear to be the same as for able-bodied individuals. Sport-related muscle pain occurs more frequently in those with limb amputations than in individuals with other types of locomotor disabilities.[70] The emotional benefits of participating in sports outweighed the possible risk of injury.[68] Sports Med 2011; 41 (9)
Amputees and Sports
The presence of pain did not influence perceived activity restrictions.[71] 3. Discussion The aim of this study was to systematically review the literature on biomechanical characteristics, cardiopulmonary function, psychological well-being, sport participation, and physical functioning and injury characteristics related to sports and/or physical activity among individuals with upper and/or lower limb amputations. Only 47 (1.3%) of 3689 papers initially identified were selected for inclusion in this systematic review. Most of the included studies were older than 10 years, were observational, had cross-sectional designs and used convenience sampling from a single rehabilitation centre. In most studies, the mean age of the study participants was below 65 years, and the study samples consisted of a high percentage of individuals with nonvascular amputations. The general population of individuals with limb amputations has an average age above 65 years, and most of these individuals have vascular amputations.[72] Due to this difference, the results of the current review do not necessarily apply to the general populations of individuals with limb amputations. Age, gender and amputation level were found to influence running and long jumping performance in athletes with limb amputations.[29-31,52] Participation in sports and physical activity positively influences their physical fitness, psychosocial well-being and physical functioning.[35,40,41,44,48,59,60] A more proximal amputation, older age and a vascular cause of amputation may lead to more problems in completing the activities of daily living among individuals with limb amputations.[54,57,64] Various studies have identified different factors influencing participation in sports among individuals with limb amputations without reaching overall agreement on a single one. In clinical practice, the type of sport or physical activity should be chosen according to each patient’s characteristics, needs and physical capabilities. When young individuals with a transtibial amputation are able to run,[25] they can particiª 2011 Adis Data Information BV. All rights reserved.
735
pate in a wide range of sports in which running is a basic component. Athletic performance was determined by the amputation level, with more proximal amputations leading to poorer performance as a result of more pronounced limb asymmetry.[27,31,32] For long jumpers with transtibial amputations, better results were recorded among individuals who used their prosthetic limb for take-off compared with those using their intact limb for take-off.[28] The findings of two studies, one[73] with a small sample size (n = 5) and a literature review,[74] suggest that prosthesis characteristics influence running performance therefore also influence athletic performance. To clarify the influence of prosthesis characteristics on athletic performance, further research is needed in which athletes with limb amputations are repeatedly tested with different types of prostheses. Every athlete with a limb amputation should be assessed individually because each has a unique running style. Individual prosthesis modifications, special components or advice may be required. One study investigated swimming technique among individuals with upper limb amputations.[23] The authors concluded that when swimming at higher speeds (at least 75% of the individual’s maximum swimming speed), stroke frequency was more important than stroke length in gaining and maintaining speed. The similar results found for running[25] may indicate that for increasing speed in running or swimming, athletes with limb amputations rely more on increasing their pace than on the length of their stride or stroke. Because data regarding swimming characteristics are available only from a single study, further research on this topic is needed before drawing conclusions. Cardiopulmonary function of individuals with limb amputations was better when a simple physical exercise programme was included in their rehabilitation programme. The intensity of the programme should be based on each individual’s heart rate during anaerobic threshold and should not exceed 80% of the maximum peak value.[37] Individuals with limb amputations must be subjected to a maximal test to obtain a peak value. This is not always possible because vigorous physical activity may be contraindicated by underlying Sports Med 2011; 41 (9)
736
cardiac problems. Therefore, only individuals with limb amputations who are healthy enough to undergo a peak test should do so. If an individual cannot be subjected to a peak test, clinicians can adjust the value for able-bodied persons of the same age according to the individual’s physical condition. The rehabilitation programmes may vary in duration, intensity, desired results and available rehabilitation time. An ergometer test can be used along with questionnaires (Medical Outcome Study 36-item short-form; SF-36,[75] and Prosthesis Evaluation Questionnaire; PEQ[76]) to assess the ability to walk. Individuals with lower limb amputations who . are able to achieve an exercise intensity of 50% VO2max[43] or 60 Watts can be expected to become successful prosthetic walkers.[34,43] When an individual’s walking prognosis is known, the rehabilitation process can be adapted according to the expected outcome, therefore optimizing the results. The psychological impact of the disability on athletes with limb amputations was found to be smaller as compared with athletes with other disabilities, such as audio-visual impairment or spinal cord injury.[45] This is an interesting finding considering that an amputation is often perceived by the able-bodied as one of the worst physical disabilities.[77] Unfortunately, no similar comparison between different disabilities has been performed in nonsporting or inactive individuals. Therefore, we cannot say if this difference is due to selection bias. Participation in sports and physical activities has a positive influence on self-esteem, perceived body image and locus of control.[44,46,78] In general, the benefits of participation in sports outweigh the inconvenience of the disability. When individuals with limb amputations participate in sports and physical activities, they can set aside the concerns related to their disability. Because the majority of them have an underlying chronic disease, encouraging them to participate in sports may help them to overcome their disability by increasing their selfesteem. By taking part in organized sporting events, they can increase their knowledge of relevant sporting equipment and techniques to improve their performance. In addition, they improve their mobility skills, personal relationships and the accepª 2011 Adis Data Information BV. All rights reserved.
Bragaru et al.
tance of their own disability.[48] When surrounded by other individuals with physical disabilities, persons with limb amputations gain a sense of normality, and they may feel more comfortable with their disability.[79] Participation in sports decreases following amputation.[49,58] In Europe, 11–39% of individuals with limb amputations participate in sports or regular physical activity, while in the US this percentage is 61%.[51-53,57,62,65] This high percentage may be biased by sample characteristics in the US studies, including an average age of 52 years and predominantly traumatic limb amputation in the study samples.[51-53] In general, individuals with limb amputations are older than 65, and more than 80% have a vascular cause for amputation.[72] The difference between European and North American studies may also be related to general differences in sports and physical activity habits between European and North American people.[80] The sports that individuals with limb amputations prefer to take part in are similar regardless of the continent: swimming, cycling, golf, fishing, fitness.[53,59,62,65] Most individuals with limb amputations do not use special sport prostheses because of high costs, lack of knowledge about such prostheses or the feeling that they are unnecessary.[53,59,62,65] A high percentage (42%) of all individuals with limb amputations reported at least one complaint about their prosthesis or about the sport organization in which they participated.[65] Sport participation appears to be hindered to some extent by the unavailability of a suitable prosthesis, poor performance or high cost of the prosthesis, inadequate facilities or insufficient information.[53,65] To increase sport participation, these factors have to be addressed. Individuals with limb amputations could be introduced to sports that do not require prosthesis use, such as wheelchair or sitting sports. Professionals should encourage individuals with limb amputations to participate more in sports or physical activities and advise them in choosing an appropriate sport prosthesis. Several factors were associated with the physical functioning, mobility and activity level following amputation including age,[52,53,57] aetiology,[53] amputation level and previous sport participaSports Med 2011; 41 (9)
Amputees and Sports
tion.[54,59,64,65] However, discrepancies were found concerning the importance of aetiology[55,57,65] and amputation level.[52,53] For example, two studies[55,57] using samples with different proportions of vascular and nonvascular amputations had similar main outcomes. This finding might lead to the conclusion that aetiology has no influence on sport participation and mobility outcomes. This statement contradicts other results on this topic[52,53,81] showing that individuals with nonvascular limb amputations are more active than individuals with vascular amputations. In some studies,[52,53,64] a more proximal amputation was found to lead to a decrease in sport participation. Other studies[62,65] have found similar rates of sport participation regardless of the amputation level. Less discrepancy exists concerning the influence of age on physical functioning, mobility and activity level following amputation.[52,53,57] Rehabilitation practitioners need to consider that a more proximal amputation, older age and the presence of co-morbidity usually lead to a longer and more difficult rehabilitation.[82] Sport-related muscle pain was more frequent amongst individuals with limb amputations than amongst individuals with other physical disabilities.[70] This difference is probably caused by the relatively limited amount of muscular tissue still available, which is subjected to more intense use as compared with individuals with other physical disabilities. Only one study of sport injuries was found that focused completely on individuals with limb amputations.[68] Other papers assessed sport injuries in a mixed group of athletes with different locomotor disabilities.[83-86] Unfortunately, they did not address each disability as a separate category, making it impossible to identify disability-specific injury rates or patterns. Additionally, the sports in which individuals with limb amputations prefer to partake, such as fishing, swimming and golf,[45,53,54] were not investigated concerning injury rates or patterns. 4. Limitations of the Current Systematic Review The literature search used only the generic term ‘sports’, and no separate searches were conducted ª 2011 Adis Data Information BV. All rights reserved.
737
for studies involving individual sports. We assumed that studies relevant to the topic of this review would most likely have the word ‘sport’ or ‘athlete’ somewhere in their content or be registered under the MeSH terms ‘sports’ or ‘physical activity’. During the title assessment phase, papers were excluded if the title had no connection to the topic of the review. It is possible that some papers that did not include the word ‘sport’ or were not included in the ‘sports’ MeSH category may have been incorrectly excluded. Therefore, a reference check of the included papers was performed, resulting in the identification of 29 additional studies. The minimum number of ten participants was arbitrarily chosen to reduce the influence of outliers on outcome and to increase the possibility of generalizing the results. Seventeen papers could not be retrieved due to unavailability or indexing errors. These papers included book chapters, dissertations and oral presentations. If, as in the main sample, only 1% of these missing papers could be included in this review, the effects on the main outcome would be negligible. The findings of this review should be interpreted cautiously because only few studies had a high methodological value. Only one randomized controlled trial was identified. Conducting a randomized controlled trial on individuals with limb amputations may prove difficult because of the limited number of subjects available. Additionally, physical activity tests can only be performed on a healthier subgroup of individuals with limb amputations. Finally, only four studies included individuals with upper limb amputations in their study populations. 5. Conclusion Participating in sports or physical activity is beneficial for individuals with lower limb amputations. The psychosocial benefits for these individuals are at least equal to those experienced by able-bodied persons. Future research should focus on the inclusion of a larger variety of sports and individuals with upper limb amputations in the study population. The influence of prosthetic technical characteristics on athletic performance needs further clarification Sports Med 2011; 41 (9)
Bragaru et al.
738
because only running, long jumping and swimming have been analysed so far. The influence of sports on quality of life needs to be more thoroughly investigated. The determinants of sport participation are controversial. Therefore, more studies investigating these determinants are needed. A physical training programme to improve cardiopulmonary function as part of the rehabilitation of individuals with limb amputations should be developed and tested for its efficacy. Acknowledgements The authors certify that no party having a direct interest in the results of the research supporting this review has or will confer a benefit on their or on any organization with which they are associated. No sources of funding were used to assist in the preparation of this review. The authors have no conflicts of interest that are directly relevant to the content of this review.
References 1. Chin T, Sawamura S, Fujita H, et al. Physical fitness of lower limb amputees. Am J Phys Med Rehabil 2002; 81 (5): 321-5 2. Horgan O, MacLachlan M. Psychosocial adjustment to lower-limb amputation: a review. Disabil Rehabil 2004; 26 (14-15): 837-50 3. Cansever A, Uzun O, Yildiz C, et al. Depression in men with traumatic lower part amputation: a comparison to men with surgical lower part amputation. Mil Med 2003; 168 (2): 106-9 4. Kashani JH, Frank RG, Kashani SR, et al. Depression among amputees. J Clin Psychiatry 1983; 44 (7): 256-8 5. Singh R, Hunter J, Philip A. The rapid resolution of depression and anxiety symptoms after lower limb amputation. Clin Rehabil 2007; 21 (8): 754-9 6. Dillingham TR, Pezzin LE, MacKenzie EJ. Limb amputation and limb deficiency: epidemiology and recent trends in the United States. South Med J 2002; 95 (8): 875-83 7. Ziegler-Graham K, MacKenzie EJ, Ephraim PL, et al. Estimating the prevalence of limb loss in the United States: 2005 to 2050. Arch Phys Med Rehabil 2008; 89 (3): 422-9 8. Bijnen FC, Caspersen CJ, Feskens EJ, et al. Physical activity and 10-year mortality from cardiovascular diseases and all causes: the Zutphen elderly study. Arch Intern Med 1998; 158 (14): 1499-505 9. Shephard RJ. Benefits of sport and physical activity for the disabled: implications for the individual and for society. Scand J Rehabil Med 1991; 23 (2): 51-9 10. Powell KE, Thompson PD, Caspersen CJ, et al. Physical activity and the incidence of coronary heart disease. Annu Rev Public Health 1987; 8: 253-87 11. Pate RR, Pratt M, Blair SN, et al. Physical activity and public health: a recommendation from the centers for disease control and prevention and the American College of Sports Medicine. JAMA 1995; 273 (5): 402-7
ª 2011 Adis Data Information BV. All rights reserved.
12. Modan M, Peles E, Halkin H, et al. Increased cardiovascular disease mortality rates in traumatic lower limb amputees. Am J Cardiol 1998; 82 (10): 1242-7 13. Webster JB, Levy CE, Bryant PR, et al. Sports and recreation for persons with limb deficiency. Arch Phys Med Rehabil 2001; 82 (3 Suppl. 1): 38-44 14. Buckley JG. Sprint kinematics of athletes with lower-limb amputations. Arch Phys Med Rehabil 1999; 80 (5): 501-8 15. Sanderson DJ, Martin PE. Joint kinetics in unilateral belowknee amputee patients during running. Arch Phys Med Rehabil 1996; 77 (12): 1279-85 16. Czerniecki JM, Gitter A. Insights into amputee running: a muscle work analysis. Am J Phys Med Rehabil 1992; 71 (4): 209-18 17. Tomaszewska J, Hildebrandt M. Badania tetna, cisnienia tetniczego krwi, oddychania i sily miesni przeprowadzone podczas kursu narciarskiego u amputowanych na konczynach dolnych. Chir Narzadow Ruchu Ortop Pol 1965; 30 (5): 557-61 18. Laboute E, Lassalle A, Letallec H, et al. Stress fractures in lower limb amputees practicing handi-sport athletics. J Traumatol Sport 2003; 20 (3): 155-61 19. Nolan L. Carbon fibre prostheses and running in amputees: a review. Foot Ankle Surg 2008; 14 (3): 125-9 20. van Velzen JM, van Bennekom CAM, Polomski W, et al. Physical capacity and walking ability after lower limb amputation: a systematic review. Clin Rehabil 2006; 20 (11): 999-1016 21. Kemper HGC, Ooijendijk WTM, Stiggelbout M. Consensus over de Nederlandse norm voor gezond bewegen. Tijdschr Soc Gezondheidsz 2000; 78: 180-3 22. World Health Organization. Physical activity-definition [online]. Available from URL: http://www.who.int/topics/ physical_activity/en/ [Accessed 2008 Dec 4] 23. Osborough CD, Payton CJ, Daly DJ. Relationships between the front crawl stroke parameters of competitive unilateral arm amputee swimmers, with selected anthropometric characteristics. J Appl Biomech 2009; 25 (4): 304-12 24. Engsberg JR, Lee AG, Tedford KG, et al. Normative ground reaction force data for able-bodied and trans-tibial amputee children during running. Prosthet Orthot Int 1993; 17 (2): 83-9 25. Enoka RM, Miller DI, Burgess EM. Below-knee amputee running gait. Am J Phys Med 1982; 61 (2): 66-84 26. Gavron SJ. Biomechanical analysis of elite sprinters with below-knee amputations. BIJAPER 1995; 2 (1): 1-14 27. Nolan L, Lees A. Touch-down and take-off characteristics of the long jump performance of world level above- and belowknee amputee athletes. Ergonomics 2000; 43 (10): 1637-50 28. Nolan L, Lees A. Prosthetic limb versus intact limb take-off in the amputee long jump. In: Proceedings of the XXIII International Symposium on Biomechanics in Sports; 2005 Aug 22-27; Beijing. 2005: 305-8 [online]. Available from URL: http://w4.ub.uni-konstanz.de/cpa/issue/view/ISBS2005 [Accessed 2011 Jul 25] 29. Nolan L, Patritti BL, Simpson KJ. A biomechanical analysis of the long-jump technique of elite female amputee athletes. Med Sci Sports Exerc 2006; 38 (10): 1829-35
Sports Med 2011; 41 (9)
Amputees and Sports
30. Nolan L, Lees A. The influence of lower limb amputation level on the approach in the amputee long jump. J Sports Sci 2007; 25 (4): 393-401 31. Patritti BL, Simpson KJ, Nolan L. Approach velocity profiles of elite male and female lower-limb amputee long jumpers. In: Proceedings of the XXth Congress of the International Society of Biomechanics; 2005 Jul 31-Aug 5; Cleveland (OH). Cleveland (OH): The Cleveland Clinic Foundation 2005: 823 32. Simpson KJ, Williams SL, DelRey P, et al. Locomotor characteristics exhibited during paralympic long jump competitions of classifications ‘‘below-’’ and ‘‘above-knee amputee’’. Int J Appl Sport Sci 2001; 13 (1): 1-17 33. Osborough CD, Payton CJ, Daly DJ. Influence of swimming speed on inter-arm coordination in competitive unilateral arm amputee front crawl swimmers. Hum Mov Sci 2010; 29 (6): 921-31 34. van Alste JA, Cruts HEP, Huisman K, et al. Exercise testing of leg amputees and the result of prosthetic training. Int Rehab Med 1985; 7 (3): 93-8 35. Pitetti KH, Snell PG, Stray-Gundersen J, et al. Aerobic training exercises for individuals who had amputation of the lower limb. J Bone Joint Surg Am 1987; 69 (6): 914-21 36. Alaranta H, Niittymaki S, Karhumaki L, et al. Physische belastungstests bei amputierten skilaufern. Med Sport 1988; 28: 112-6 37. Chin T, Sawamura S, Fujita H, et al. The efficacy of the oneleg cycling test for determining the anaerobic threshold (AT) of lower limb amputees. Prosthet Orthot Int 1997; 21 (2): 141-6 38. Kurdibaylo SF, Bogatykh VG. Swimming as a means of enhancing the adaptive potentials of the disabled after amputation of the lower extremities. Vopr Kurortol Fizioter Lech Fiz Kult 1997; 1: 25-8 [in Russian; online]. Available from URL: http://lib.sportedu.ru/Document. idc?DocID=91670&DocQuerID=5048925&DocTypID= NULL&QF=Simple&Pg=20&Cd=Win&Tr=0&On=0&Doc QuerItmID= [Accessed 2011 Jul 25] 39. Hutzler Y, Ochana S, Bolotin R, et al. Aerobic and anaerobic arm-cranking power outputs of males with lower limb impairments: Relationship with sport participation intensity, age, impairment and functional classification. Spinal Cord 1998; 36 (3): 205-12 40. Chin T, Sawamura S, Fujita H, et al. Effect of endurance training program based on anaerobic threshold (AT) for lower limb amputees. J Rehabil Res Dev 2001; 38 (1): 7-11 41. Kobzev IA, Khramov VV. Some features of reaction of cardiovascular system on physical load in physically handicapped amputee engaged in sports. Sportivnaia Medicina 2002; 7: 13-6 [in Russian; online]. Available from URL: http://lib.sportedu.ru/Document.idc?DocID=91670&Doc QuerID=5048925&DocTypID=NULL&QF=Simple&Pg= 20&Cd=Win&Tr=0&On=0&DocQuerItmID= [Accessed 2011 Jul 25] 42. Huonker M, Schmid A, Schmidt-Trucksass A, et al. Size and blood flow of central and peripheral arteries in highly trained able-bodied and disabled athletes. J Appl Physiol 2003; 95 (2): 685-91 43. Chin T, Sawamura S, Shiba R. Effect of physical fitness on prosthetic ambulation in elderly amputees. Am J Phys Med Rehabil 2006; 85 (12): 992-6
ª 2011 Adis Data Information BV. All rights reserved.
739
44. Valliant PM, Bezzubyk I, Daley L, et al. Psychological impact of sport on disabled athletes. Psychol Rep 1985; 56 (3): 923-9 45. Mastro JV, Burton AW, Rosendahl M. Attitudes of elite athletes with impairments toward one another: a hierarchy of preference. Adapt Phys Act Q 1996; 2 (13): 197-210 46. Wetterhahn KA, Hanson C, Levy CE. Effect of participation in physical activity on body image of amputees. Am J Phys Med Rehabil 2002; 81 (3): 194-201 47. Lowther J, Lane A, Lane H. Self-efficacy and psychological skills during the amputee soccer world cup. Athletic Insight 2002; 4 (2): 23–34 [online]. Available from URL: http:// www.athleticinsight.com/Vol4Iss2/SoccerSelfEfficacy.htm [Accessed 2009 Jul 6] 48. Sporner ML, Fitzgerald SG, Dicianno BE, et al. Psychosocial impact of participation in the national veterans wheelchair games and winter sports clinic. Disabil Rehabil 2009; 31 (5): 410-8 49. Couture M, Caron CD, Desrosiers J. Leisure activities following a lower limb amputation. Disabil Rehabil 2010; 32 (1): 57-64 50. Tatar Y. Body image and its relationship with exercise and sports in turkish lower-limb amputees who use prosthesis. Sci Sports 2010; 25 (6): 312-7 51. Kegel B, Carpenter ML, Burgess EM. A survey of lowerlimb amputees: prostheses, phantom sensations, and psychosocial aspects. Bull Prosthet Res 1977; 10 (27): 43-60 52. Kegel B, Carpenter ML, Burgess EM. Functional capabilities of lower extremity amputees. Arch Phys Med Rehabil 1978; 59 (3): 109-20 53. Kegel B, Webster JC, Burgess EM. Recreational activities of lower extremity amputees: a survey. Arch Phys Med Rehabil 1980; 61: 258-64 54. Medhat A, Huber PM, Medhat MA. Factors that influence the level of activities in persons with lower extremity amputation. Rehabil Nurs 1990; 15 (1): 13-8 55. Pohjolainen T, Alaranta H, Karkkainen M. Prosthetic use and functional and social outcome following major lower limb amputation. Prosthet Orthot Int 1990; 14 (2): 75-9 56. Gailey R. Recreational pursuits for elders with amputation. Top Geriatr Rehabil 1992; 8 (1): 39-58 57. Burger H, Marincek C, Isakov E. Mobility of persons after traumatic lower limb amputation. Disabil Rehabil 1997; 19 (7): 272-7 58. Burger H, Marincek C. The life style of young persons after lower limb amputation caused by injury. Prosthet Orthot Int 1997; 21 (1): 35-9 59. Legro MW, Reiber GE, Czerniecki JM, et al. Recreational activities of lower-limb amputees with prostheses. J Rehabil Res Dev 2001; 38 (3): 319-25 60. Rau B, Bonvin F, de Bie R. Short-term effect of physiotherapy rehabilitation on functional performance of lower limb amputees. Prosthet Orthot Int 2007; 31 (3): 258-70 61. Yazicioglu K, Taskaynatan MA, Guzelkucuk U, et al. Effect of playing football (soccer) on balance, strength, and quality of life in unilateral below-knee amputees. Am J Phys Med Rehabil 2007; 86 (10): 800-5
Sports Med 2011; 41 (9)
740
62. Yari P, Dijkstra PU, Geertzen JHB. Functional outcome of hip disarticulation and hemipelvectomy: a cross-sectional national descriptive study in the Netherlands. Clin Rehabil 2008; 22 (12): 1127-33 63. Walker JL, Knapp D, Minter C, et al. Adult outcomes following amputation or lengthening for fibular deficiency. J Bone Joint Surg Am 2009; 91 (4): 797-804 64. Karmarkar AM, Collins DM, Wichman T, et al. Prosthesis and wheelchair use in veterans with lower-limb amputation. J Rehabil Res Dev 2009; 46 (5): 567-76 65. Kars C, Hofman M, Geertzen JH, et al. Participation in sports by lower limb amputees in the province of Drenthe, the Netherlands. Prosthet Orthot Int 2009; 33 (4): 356-67 66. van den Berg-Emons RJ, Bussmann JB, Stam HJ. Accelerometry-based activity spectrum in persons with chronic physical conditions. Arch Phys Med Rehabil 2010; 91 (12): 1856-61 67. Bekkering WP, Vliet Vlieland TP, Koopman HM, et al. Functional ability and physical activity in children and young adults after limb-salvage or ablative surgery for lower extremity bone tumors. J Surg Oncol 2011; 103 (3): 276-82 68. Kegel B, Malchow D. Incidence of injury in amputees playing soccer. Palaestra 1994; 10 (2): 50-4 69. Melzer I, Yekutiel M, Sukenik S. Comparative study of osteoarthritis of the contralateral knee joint of male amputees who do and do not play volleyball. J Rheumatol 2001; 28 (1): 169-72 70. Bernardi M, Castellano V, Ferrara MS, et al. Muscle pain in athletes with locomotor disability. Med Sci Sports Exerc 2003; 35 (2): 199-206 71. Desmond D, Gallagher P, Henderson-Slater D, et al. Pain and psychosocial adjustment to lower limb amputation amongst prosthesis users. Prosthet Orthot Int 2008; 32 (2): 244-52 72. Rommers GM, Vos LD, Groothoff JW, et al. Epidemiology of lower limb amputees in the north of the Netherlands: aetiology, discharge destination and prosthetic use. Prosthet Orthot Int 1997; 21 (2): 92-9 73. Czerniecki JM, Gitter A, Munro C. Joint moment and muscle power output characteristics of below knee amputees during running: the influence of energy storing prosthetic feet. J Biomech 1991; 24 (1): 63-75
ª 2011 Adis Data Information BV. All rights reserved.
Bragaru et al.
74. Pailler D, Sautreuil P, Piera JB, et al. Evolution in prostheses for sprinters with lower-limb amputation. Ann Readapt Med Phys 2004; 47 (6): 374-81 75. Ware JE, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I: conceptual framework and item selection. Med Care 1992; 30 (6): 473-83 76. Legro MW, Reiber GD, Smith DG, et al. Prosthesis evaluation questionnaire for persons with lower limb amputations: assessing prosthesis-related quality of life. Arch Phys Med Rehabil 1998; 79 (8): 931-8 77. Furst L, Humphrey M. Coping with the loss of a leg. Prosthet Orthot Int 1983; 7 (3): 152-6 78. Durstine JL, Painter P, Franklin BA, et al. Physical activity for the chronically ill and disabled. Sports Med 2000; 30 (3): 207-19 79. Verbrugge LM. The structure of adult friendship choices. Social Forces 1977; 56 (2, Special Issue): 576-97 80. Ford R. European and North American sports differences. Scot J Polit Econ 2000; 47 (4): 431-55 81. Geertzen JHB, Bosmans JC, van der Schans CP, et al. Claimed walking distance of lower limb amputees. Disabil Rehabil 2005; 27 (3): 101-4 82. Pernot HF, de Witte LP, Lindeman E, et al. Daily functioning of the lower extremity amputee: an overview of the literature. Clin Rehabil 1997; 11 (2): 93-106 83. Manonelles Marqueta P, Arguisuelas Martinez M, Santiago Fernandez R, et al. Number of injuries in high athletics competition of Paralympics sportsmen. Arch Med Deporte 2005; 22 (109): 371-9 84. Ferrara MS, Peterson CL. Injuries to athletes with disabilities: identifying injury patterns. Sports Med 2000; 30 (2): 137-43 85. Ferrara MS, Buckley WE, Messner DG, et al. The injury experience and training history of the competitive skier with a disability. Am J Sports Med 1992; 20 (1): 55-60 86. McCormick DP. Injuries in handicapped alpine ski racers. Phys Sportsmed 1985; 13 (12): 93-7
Correspondence: M. Bragaru, MSc, Department of Rehabilitation Medicine, University Medical Centre Groningen, Hanzeplein 1, PO Box 30.001, CB 41, 9700 RB Groningen, the Netherlands. E-mail:
[email protected]
Sports Med 2011; 41 (9)
Sports Med 2011; 41 (9): 741-756 0112-1642/11/0009-0741/$49.95/0
REVIEW ARTICLE
ª 2011 Adis Data Information BV. All rights reserved.
Repeated-Sprint Ability – Part II Recommendations for Training David Bishop,1 Olivier Girard2 and Alberto Mendez-Villanueva3 1 Institute of Sport, Exercise and Active Living (ISEAL), School of Sport and Exercise Science, Victoria University, Melbourne, VIC, Australia 2 ASPETAR – Qatar Orthopaedic and Sports Medicine Hospital, Research and Education Centre, Doha, Qatar 3 Physiology Unit, Sport Science Department, ASPIRE Academy for Sport Excellence, Doha, Qatar
Contents Abstract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Training the Limiting Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Energy Supply. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.1 Phosphocreatine Resynthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.2 Anaerobic Glycolysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.3 Aerobic Metabolism. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 H+ Accumulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Muscle Activation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Specific Training Strategies and Repeated-Sprint Ability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Repeated-Sprint Training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Sprint Training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Small-Sided Games . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4 Resistance Training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Abstract
741 742 743 743 743 744 745 745 747 748 748 750 750 751 752
Short-duration sprints, interspersed with brief recoveries, are common during most team sports. The ability to produce the best possible average sprint performance over a series of sprints (£10 seconds), separated by short (£60 seconds) recovery periods has been termed repeated-sprint ability (RSA). RSA is therefore an important fitness requirement of team-sport athletes, and it is important to better understand training strategies that can improve this fitness component. Surprisingly, however, there has been little research about the best training methods to improve RSA. In the absence of strong scientific evidence, two principal training theories have emerged. One is based on the concept of training specificity and maintains that the best way to train RSA is to perform repeated sprints. The second proposes that training interventions that target the main factors limiting RSA may be a more effective approach. The aim of this review (Part II) is to critically analyse training strategies to improve both RSA and the underlying factors responsible for fatigue during repeated sprints (see Part I of the preceding
Bishop et al.
742
companion article). This review has highlighted that there is not one type of training that can be recommended to best improve RSA and all of the factors believed to be responsible for performance decrements during repeated-sprint tasks. This is not surprising, as RSA is a complex fitness component that depends on both metabolic (e.g. oxidative capacity, phosphocreatine recovery and H+ buffering) and neural factors (e.g. muscle activation and recruitment strategies) among others. While different training strategies can be used in order to improve each of these potential limiting factors, and in turn RSA, two key recommendations emerge from this review; it is important to include (i) some training to improve single-sprint performance (e.g. ‘traditional’ sprint training and strength/power training); and (ii) some high-intensity (80–90% maximal oxygen consumption) interval training to best improve the ability to recover between sprints. Further research is required to establish whether it is best to develop these qualities separately, or whether they can be developed concurrently (without interference effects). While research has identified a correlation between RSA and total sprint distance during soccer, future studies need to address whether training-induced changes in RSA also produce changes in match physical performance.
1. Introduction Short-duration sprints (£10 seconds), interspersed with brief recovery periods, are common during most team sports.[1] The ability to produce the best possible average sprint performance over a series of sprints, separated by short (£60 seconds) recovery periods, is therefore important for all team-sport athletes and has been termed repeated-sprint ability (RSA). While RSA is often equated with a low fatigue index (i.e. the decrease in performance from the first to the last sprint), it is important to note that a good RSA is better described by a high average sprint performance, with or without a low fatigue index (e.g. a marathon runner with a low average sprint performance, but a very low fatigue index, would not be classified as having good repeated-sprint ability) [see also Part I of the preceding companion article[2]]. Mean time recorded during an RSA test predicts the distance of high-intensity running (>19.8 km/h), and the total sprint distance during a professional soccer match.[3] This suggests that improving RSA should result in greater team-sport physical performance, and that it is important to better understand training strategies that can enhance this fitness component. ª 2011 Adis Data Information BV. All rights reserved.
Recently, there has been an increase in scientific research regarding the importance of RSA for team- and racket-sport athletes.[1,3-7] Surprisingly, however, there has been little research about the best training methods to improve this fitness component.[8] In the absence of strong scientific evidence, one concept that has emerged is that the best way to train RSA may be to perform repeated sprints.[9] While such a concept appeals to the concept of training specificity, the scientific evidence in support of this approach is currently lacking. Indeed, many studies have reported significant improvements in RSA with more generic training (e.g. interval training).[9-12] The aim of this review is to critically analyse training strategies to improve both RSA and the underlying factors responsible for fatigue during repeated sprints. In order to obtain the necessary articles for this review, several databases were searched including SportDiscus, PubMed, Web of Science, MEDLINE and Google Scholar. Key search terms used included ‘repeated-sprint ability’, ‘repeatedsprint exercise’, ‘multiple sprints’, ‘team sports’, ‘training’, ‘rugby’, ‘soccer’, ‘football’, ‘basketball’, ‘conditioning’, ‘endurance’ and ‘small-sided games’. Manual searches were also made using the reference lists from recovered articles. Due to Sports Med 2011; 41 (9)
Training to Improve Repeated-Sprint Ability
743
the small number of articles relating to training and RSA, there was no limit to the search period. 2. Training the Limiting Factors During repeated-sprint exercise (RSE), the inability to reproduce performance across sprint repetitions (fatigue) is manifested by a decline in sprint speed (i.e. increased time to cover a fixed distance) or peak/mean power output. Proposed factors responsible for these performance decrements have previously been reviewed[13] (see also Part I of this review[2]) and include limitations to energy supply (e.g. phosphocreatine resynthesis, aerobic and anaerobic glycolysis) and metabolite accumulation (e.g. inorganic phosphate [Pi], H+). Increasing evidence suggests that failure to fully activate the contracting muscle may also limit repeated-sprint performance.[14,15] Training interventions that are able to lessen the influence of these limiting factors should improve RSA. 2.1 Energy Supply 2.1.1 Phosphocreatine Resynthesis
As the brief recovery times between repeated sprints will lead to only a partial restoration of phosphocreatine stores,[16,17] it has been proposed that the ability to resynthesize phosphocreatine may be an important determinant of the ability to
reproduce sprint performance.[17,18] In line with this proposition, strong relationships have been reported between phosphocreatine resynthesis and the recovery of performance during both repeated, 30-second, all-out exercise bouts[17,18] and repeated 6-second sprints (Mendez-Vallanueva A. et al., unpublished data). These findings suggest that the performance of repeated sprints may be improved by training interventions that increase the rate of phosphocreatine resynthesis. The oxidative metabolism pathways are essential for phosphocreatine resynthesis during the recovery from high-intensity exercise.[19] This suggests that individuals with an elevated aerobic fitness (i.e. high maximal oxygen consumption . [VO2max] or lactate threshold) should be able to more rapidly resynthesize phosphocreatine between repeated sprints. Indeed, cross-sectional research[17,20-23] and one training study[24] support the hypothesis that endurance training enhances phosphocreatine resynthesis following low-intensity exercise. Recently, it has also been reported that high-intensity . interval training (6–12 · [2 minutes at ~100% VO2max: 1 minute rest]), can significantly improve phosphocreatine resynthesis during the first 60 seconds following high-intensity exercise (figure 1).[25] In contrast, no changes in the rate of phosphocreatine resynthesis have been reported following interval (8 · [30 seconds at Post-train Pre-train
90
PCr content (mmoL/kg dw)
85 80 75 *†
70 65 †
60 55
†
50
†
45 Rest
Post-ex Post-60 sec Timepoint
Post-180 sec
Fig. 1. Changes in resting and post-exercise phosphocreatine (PCr) content following high-intensity interval training ([25] and Bishop D. et al., unpublished research). dw = dry weight; * indicates significantly different from pre train; - indicates significantly different from rest.
ª 2011 Adis Data Information BV. All rights reserved.
Sports Med 2011; 41 (9)
Bishop et al.
744
. ~130% VO2max: 90 seconds rest]), or intermittentsprint training (15 · [6-second sprint: 1-minute jog recovery]),[12] or training involving repeated, 30-second, all-out efforts (4–7 · [30 seconds ‘allout’: 3–4 minutes rest]).[26] These results can possibly be attributed to the absence of significant changes . in aerobic fitness (as measured by VO2max) with these types of training. Alternatively, these results may be related to the fact that these studies all measured phosphocreatine resynthesis 3-minutes post-exercise, a timepoint when phosphocreatine resynthesis is largely complete and therefore less likely to be influenced by training. Nonetheless, while the optimal training intensity has not yet been established, the limited research to date suggests that improvements in aerobic fitness may be required to improve phosphocreatine resynthesis. As repeated-sprint training has been reported to increase aerobic fitness,[4,27] further research is required to investigate whether this type of training can also increase the fast component (e.g. first 60 seconds) of phosphocreatine resynthesis, and whether such changes are superior to those observed following aerobic training (e.g. interval training[25]). 2.1.2 Anaerobic Glycolysis
The large drop in intramuscular phosphocreatine, along with the concomitant rise in Pi and adenosine monophosphate, stimulates the rapid activation of anaerobic glycolysis at the start of a sprint.[28] As a consequence, anaerobic glycolysis is an important source of adenosine triphosphate (ATP) during a single sprint.[29] During subsequent sprints however, there is a dramatic decrease in the ATP production, via anaerobic glycolysis, during sprint efforts that has been attributed to the acidosis resulting from the maximal anaerobic degradation of glycogen during the early sprints.[18,30] It is therefore unclear whether increasing the maximal anaerobic glycogenolytic and glycolytic rate will lead to improvements in RSA. On one hand, it could be argued that training that increases the ability to supply ATP from anaerobic glycolysis would be detrimental to RSA due to the negative correlation between anaerobic ATP production during the first sprint and sprint decrement during a ª 2011 Adis Data Information BV. All rights reserved.
repeated-sprint test.[29,31] On the other hand, it also needs to be considered that subjects with a greater glycogenolytic rate have also been reported to have a greater initial sprint performance,[29] and that researchers have reported a strong positive correlation between initial sprint performance and both final sprint performance[29] and total sprint performance[32,33] during tests of RSA. Thus, while these findings highlight the difficulties associated with interpreting contrasting effects on the various RSA test measures,[34] they suggest that increasing the anaerobic contribution is likely to improve both initial and mean sprint performance, and thus the ability to perform repeated sprints. It should be noted, however, that some researchers have reported significant increases in glycolytic enzymes following sprint training without a corresponding increase in sprint performance.[35,36] Further research is therefore required to investigate the relationship between improvements in anaerobic ATP production and RSA. As training does not increase the amount of phosphocreatine breakdown during high-intensity exercise,[12,25,37] changes in the ability to produce ATP via anaerobic glycolysis are likely to be well reflected by training-induced changes in indirect measures of anaerobic capacity, such as maximal accumulated oxygen deficit (MAOD). A high rate of anaerobic energy release during exercise has been proposed to be an important stimulus to increase MAOD.[38] This is supported by increases in MAOD in response to high-intensity (20– . 120-second intervals at 100–200% VO2max),[38-40] but . not moderate-intensity (60 minutes at 70% VO2max) endurance training.[40] Furthermore, the greatest changes in MAOD have typically been reported in response to interval training that produces large changes in blood lactate concentration (>10 mmol/L).[38,40] These results are consistent with the observation that traininginduced changes in enzymes important for anaerobic glycolysis (e.g. phosphofructokinase and phosphorylase) are greater following training that involves repeated 30-second bouts than repeated 6-second bouts[41] or continuous training.[42] In the only study to date, 6 weeks of repeated-sprint training did not increase phosphofructokinase Sports Med 2011; 41 (9)
Training to Improve Repeated-Sprint Ability
activity.[43] Greater increases in glycolytic enzymes have typically been reported when high-intensity intervals are separated by long (10–15 minute),[36,44] rather than by short (£4 minute),[45-47] rest periods. This is consistent with the larger increases in peak blood or muscle lactate when 30-second all-out efforts are separated by 10to 15-minute rest periods,[35,48] compared with 3–4-minute rest periods.[37] From this research, it is difficult to determine whether this is an effect of recovery duration per se, or the better maintenance of exercise intensity with longer recoveries. The above research suggests that to increase the anaerobic performance of team-sport athletes one should utilize short (20–30 second), highintensity (all-out) intervals separated by relatively long rest periods (~10 minutes). 2.1.3 Aerobic Metabolism
Several physiological adaptations related to an increased reliance on aerobic metabolism to resynthesize ATP, such as greater mitochondrial respiratory capacity,[49] faster oxygen uptake kinetics,[50,51] an accelerated post-sprint muscle re[52] a higher lactate threshold[53] oxygenation rate, . and a higher VO2max,[51,54-57] have been associated with an enhanced ability to resist fatigue during . repeated sprints. The most studied factor is VO2max that has been reported to be moderately correlated (0.62 < r < 0.68; p < 0.05) with RSA (both mean sprint performance and sprint decrement).[51,54-56] Research . has also shown that subjects with a greater VO2max[58] have a superior ability to resist fatigue during RSE, especially during the latter stages of a repeated-sprint test . when subjects may reach. their VO2max.[59] This suggests that improving VO2max may allow for a greater aerobic contribution to repeated sprints, potentially improving RSA. However, research also indicates .that there is not a linear relationship between VO2max and various repeated-sprint fatigue indices.[32,60] Thus, it may be more important to. develop an ‘optimal’, rather than a to maximal, VO2max. Further research is required . determine what an appropriate level of VO2max is, above which further increases may not be accompanied by comparable improvements in RSA. In addition, the possible links between other aerobic ª 2011 Adis Data Information BV. All rights reserved.
745
fitness indices (e.g. lactate threshold, economy, oxygen kinetics, the velocity associated with . ), which are relatively independent of the VO 2max . VO2max, should be the subject of further research. Many physiologists believe that it is the reduced muscle oxygen levels during. training that provide the stimulus to increase VO2max.[61] As the oxygen level in the muscle decreases with in. creases in exercise intensity up to 100% VO2max, but does not decrease further once the exercise intensity exceeds this point,[62] this suggests that interval training at intensities that approximate . VO . 2max may be most effective for improving VO2max. This is supported by previous studies that have reported greater improvements in . VO2max . after interval training (at approximately the VO2max intensity) when compared with continuous training matched for total work.[61,63-65] It should be noted, however, that most of these studies performed their continuous training at . very low intensities (£56% of the power at VO2max).[61,63,64] When compared with continuous training performed at intensities >60% . of the power at VO2max, interval training has been reported to produce similar improvement in . VO2max.[66-69] These results therefore suggest that if a minimum training intensity is exceeded (>60% . of the power output at VO2max), and total work is equivalent, the choice of either interval or continuous training will result in similar improve. ments in VO2max. However, one advantage of interval training is that it may concurrently develop other factors (e.g. the rate of phosphocreatine resynthesis[25] and muscle buffer capacity[69]). The above research suggests that to increase the aerobic fitness of team-sport athletes, one should utilize . high-intensity interval training (80–90% of VO2max) interspersed with rest periods (e.g. 1 minute) that are shorter than the work periods (e.g. 2 minutes). 2.2 H+ Accumulation
It has been argued that the considerable increases in muscle[58,69,70] and blood[32,71] H+ accumulation observed following sprinting may impair repeated-sprint performance.[72] In support of this, correlations have been observed between Sports Med 2011; 41 (9)
Bishop et al.
746
sprint decrement, and both muscle buffer capacity (Xm) and changes in muscle and blood pH.[32,54,55,58,73] This suggests that RSA may be improved by interventions that can increase the removal of H+ from the muscle.[12,54,73] The removal of intracellular H+ during intense skeletal muscle contractions (such as repeated sprints) occurs via intracellular buffering (bmin vitro) and a number of different membrane transport systems, especially the monocarboxylate transporters (MCTs) [figure 2].[74] The MCTs appear to be the dominant regulator of muscle pH during and after high-intensity exercise.[74] A large increase in muscle H+ and/or lactate during exercise has been proposed to be an important stimulus for adaptations of the muscle pH regulating systems.[75] This is supported by increases in bmin vitro in response to high-intensity interval training (6–10 · [2 minutes at 120–140% of the lactate threshold] : 1 minute rest), but not moderate-intensity, continuous training (~30 minutes at 80–95% lactate threshold).[69] However, greater accumulation of lactate and H+ during training has not always been associated with greater increases in MCTs[12,76] or bmin vitro.[25] Furthermore, research suggests that too large an accumulation of H+ during training (e.g. .interval training performed at intensities >100% VO2max)
may have a detrimental effect on adaptations to the pH regulatory systems within the muscle.[25,77] Thus, while further research is required, it appears that intramuscular accumulation of H+ and/or lactate provides an important stimulus to improve the muscle pH regulating systems; however, maximizing H+ accumulation during training does not maximize these adaptations. It should be noted, however, that most of this research has been conducted on moderately-trained subjects and further research is required to confirm these observations in well trained, teamsport athletes. The above considerations have important implications for the design of training programmes to improve the muscle pH regulating systems and hence, RSA. To increase bmin vitro, it appears important to employ high-intensity interval training . (~80–90% VO2max), interspersed with rest periods that are shorter than the work periods (e.g. 2 minutes of exercise followed by 1 minute of rest), so that the muscle is required to contract while experiencing a .reduced pH.[69,78] Interval training at intensities >VO2max does not appear to provide additional benefits, and has the potential to actually decrease bmin vitro.[25] In addition, the use of rest periods that are longer than the work periods allows greater removal of lactate and H+ prior to Na+
NHE H+
H+
+ Proteins Dipeptides (e.g.carnosine and anserine) Phosphates
HCO3−
MCT La
H2CO3 −
CI NBC CO2 + H2O → H+ + HCO3− Na+
Fig. 2. Muscle (H+) regulation. MCT = monocarboxylate transporters; NBC = sodium-bicarbonate co-transporter; NHE = sodium-hydrogen exchanger.
ª 2011 Adis Data Information BV. All rights reserved.
Sports Med 2011; 41 (9)
Training to Improve Repeated-Sprint Ability
subsequent intervals[79] and typically does not result in a significant increase in bmin vitro.[37,48] While an optimal training volume to improve bmin vitro is yet to be established, it appears that interval training at the above-mentioned intensities, 2–3 times per week, for 3–5 weeks, can result in significant increases in bmin vitro.[69,75,80] In the only study to date, repeated-sprint training (5–8 · [5 · 25–35 m sprints: 21 seconds of rest]) was reported to be less effective than highintensity. interval training (5–8 · [2 minutes at ~100% VO2max: 1–3 minute of rest]) for improving bmin vitro in team-sport athletes, even when matched for total training volume.[27] It is more difficult to recommend the ideal training programme to increase the MCTs as significant increases have been reported following both moderate-[81-83] and high-[76,84] intensity training. However, one factor that these training programmes tend to have in common is that they are associated with only modest posttraining increases in blood lactate concentration (~4–8 mmol/L). When high-intensity training has been employed, the rest periods between highintensity intervals have ranged from 90 to 240 seconds (e.g. a work-to-rest ratio of £1 : 2),[76,85] allowing substantial removal of lactate and H+ prior to subsequent intervals.[79] Thus, in contrast to the high-intensity training required to increase bmin vitro, it appears that both moderate- and high-intensity training can increase the MCTs, but that training should be structured so as to provoke only a modest increase in blood lactate concentration (~4–8 mmol/L). This might explain why, in the only study to date that has recruited well . trained subjects, training at 60–70% VO2max (postexercise blood lactate concentration <1.5 mmol/L) was insufficient to maintain MCT content.[65] Significant changes in MCT content appear more likely when training is performed 2–3 times per week for 6–8 weeks. While no studies to our knowledge have investigated the influence of repeated-sprint training on changes in MCT content, intermittent-sprint training (15 · [6-second sprint: 1-minute jogging recovery]) and. interval training (8 · [30 seconds at 130% VO2max: 90-seconds rest]) have been reported to be equally effective for increasing MCT1 content.[12] Further ª 2011 Adis Data Information BV. All rights reserved.
747
research is therefore required to determine the effects of repeated-sprint training on the muscle lactate transporters. 2.3 Muscle Activation
Sprinting requires considerable levels of neural activation.[86] Among the various potential neurallymediated mechanisms determining RSA (in particular, sprint decrement), the ability to voluntarily fully activate the working musculature and to maintain muscle recruitment and rapid firing over sprint repetitions may critically affect fatigue resistance.[14,15,31,87] This suggests that under conditions of considerable fatigue development (e.g. sprint decrement score and fatigue index >25%) the failure to fully activate the contracting musculature may become an important factor limiting performance during RSE. Other factors, including disruption of optimal temporal sequencing of agonist and antagonist muscle activation (i.e. muscle coordination patterns) and/or motor unit recruitment strategies (e.g. decreased recruitment of fibres with faster conduction velocities), can also potentially limit RSA, as a multitude of different muscles must be activated at the appropriate times and intensities to maximize sprinting efficiency.[88,89] A variety of training methods have been employed to successfully improve the degree of muscle activation (e.g. electromyostimulation, eccentric strength and plyometric training).[90] There is also evidence that such neural adaptations could enhance subsequent athletic performance.[91,92] While such research suggests that training which improves muscle activation has the potential to improve RSA, specific training studies are required before scientifically-based training recommendations can be given. This will not be easy as much of the fatigue experienced during RSE appears to be mediated by metabolic factors (see Part I of the preceding companion review[2]), and such research will need to demonstrate that traininginduced improvements in RSA can be attributed to improvements in actual muscle activation, and not concurrent improvement in metabolic factors. It has also been postulated that the ability for fast torque development depends, among other Sports Med 2011; 41 (9)
Bishop et al.
748
factors, on the specific ability for fast muscle activation at contraction onset (i.e. earlier recruitment of large motor units, increased synchrony, elevated motor unit firing rate[93]). Training-based studies that have reported corresponding increases in the rate of force rise and EMG development[94,95] support this viewpoint. Pending confirmatory research, these adaptations have the potential to improve rapid and forceful field/ on-court movements such as sprinting involving muscle contraction times of less than 250 msec. It is therefore recommended that measurement of rate of force development and concomitant EMG activity (0–200 msec time frame) should be employed in future training studies (during standardized tests on a dynamometer or by exploring the early slope of vertical ground-reaction forcetime curves during running-based RSE) to shed more light on neural adaptations to training targeting an improvement in sprint performance. Although it is tempting to propose that enhancing performance during initial sprint efforts may provide an effective strategy to improve mean sprint performance (e.g. total mechanical work), it also needs to be acknowledged that this is also likely to lead to a higher sprint decrement score.[31,32] Thus, additional training regimens may also need to be implemented to develop those fatigueresistance factors. 3. Specific Training Strategies and Repeated-Sprint Ability 3.1 Repeated-Sprint Training
Anecdotally, repeated-sprint training is a popular training method used by team-sport athletes to improve RSA. However, despite the belief that such specific training will improve RSA more than generic training (e.g. interval training), very few studies have directly compared these two forms of training. We are aware of seven studies that have investigated adaptations to repeatedsprint training (table I). Only five of these studies incorporated a control training group, and only four of these studies recruited team-sport athletes. It is therefore difficult to make solid conclusions about the benefits of repeated-sprint ª 2011 Adis Data Information BV. All rights reserved.
training in comparison to other types of training. Nonetheless, despite the obvious need for further research, some tentative conclusions can be made. . Repeated-sprint training is able to improve VO2max. In the studies performed to date, 5–12 weeks of repeated-sprint training has been reported to result in a 5.0–6.1% increase in . VO2max. Moreover, this increase is similar to that reported in the two studies, which incorporated a control group who performed interval training . (5.2–6.6% increase in VO2max).[4,27] However, as other studies utilizing different types of interval training have reported more than 10% increases . in VO2max,[10,101] further research, comparing repeated-sprint training and these other types of training, is . required to verify the best means to improve VO2max in team-sport athletes. Further research is also required to investigate additional physiological adaptations to repeated-sprint training (e.g. changes in ion regulation, anaerobic capacity, phosphocreatine resynthesis, etc). For example, the limited evidence to date suggests that, compared with repeated- or intermittent-sprint training, interval training produces superior increases in both bmin vitro[27] and Na+/K+ pump isoform content.[12] With respect to RSA, repeated-sprint training has been reported to produce greater improvements in best sprint time[12,27,96] and mean sprint time,[4,12,27,96] compared with interval-based training. In contrast, interval training appears to be superior to repeated-sprint training to decrease (i.e. improve) the sprint decrement score (or the fatigue index).[12,27] However, due to the problems associated with interpreting changes in the sprint decrement score when there are concurrent changes in best sprint time,[102] it is difficult to make universal recommendations. For example, Mohr et al.[12,103] have suggested that the greater improvement in sprint decrement following interval training (termed ‘speed-endurance’ training [SET] by the authors), when compared with intermittent-sprint training (ST) [figure 3], is a sign that interval training is superior for improving RSA. However, this interpretation has been questioned[34] as a closer analysis of their data suggests that the intermittent-sprint-training group had a greater improvement in single-sprint performance Sports Med 2011; 41 (9)
Study (y)
Buchheit et al.[9] (2008)
Subjects
Training programme . VO2max (mL/kg/min)a
9, MA, M, TS
NR
2 · ([5–6 · 30–40 m shuttle sprints: 14–23 sec]: 2 min rest); 2 d/wk, 9 wk . 9–24 · (15–20 sec at 105–115% VO2max: 15–20 sec); 2 d/wk, 9 wk
NR
3–4 · ([4–6 · accelerations/sprints (<5 sec): 30 sec]: 3 min rest); › 2.7 › 0.7 2 d/wk, 4 wk
8, MA, M, TS
Buchheit et al.[96] (2010)
7, MA, M, TS
9, MA, M
. VO2max (%)
best sprint (%)
mean sprint (%) DS (%)
› 0.3 NS › 1.4*
› 1.0 NS › 1.5*
› 19 NS › 44 NS
NR
› 22 › 0.8
› 35 › 39
NR
3–5 · (30 sec all-out shuttle sprints: 4 min rest), 2 d/wk, 4 wk (both groups also performed two other team training sessions)
7, MA, M, TS
Dawson et al.[43] (1998)
Adaptations
type
57.0 – 2.4
4–6 · ([5 · 30 to 80 m sprints: 30–90 sec rest]: 2–4 min rest); 3 d/wk, 6 wk
› 2.4*
› 2.2*
› 16 NS
› 6.1*
NR
› 2.1* › 0.3 NS
NR
› 5.0* › 6.6*
› 4.0* › 0.7 NS
› 4.3* › 2.4*
› 13 NS › 54*
NR
Bravo et al.[4] (2008)
21, MA, M, TS
55.7 – 2.3
3 · ([6 · 40 m sprint: 20 sec rest]: 4 min rest); 2 d/wk, 12 wk
21, MA, M, TS
52.8 – 3.2
4 · (4 min at 95% HRmax: 3 min at 75% HRmax); 2 d/wk, 12 wk (both groups also performed two other team training sessions)
Mohr et al.[12] (2007)
6, MA, M
50.2 – 3.7
15 · (6 sec sprint:1 min jog recovery); 3–5 d/wk, 8 wk
7, MA, M
49.0 – 4.2
8 · (30 sec at 130% max: 90 sec rest); 3–5 d/wk, 8 wk
Schneiker and Bishop[27] (2008)
7, MA, M TS
56.2 – 6.8
7, MA, M, TS
56.6 – 5.3
5–8 · (5 · 25 to 35 m sprints: 21 sec rest); 3 d/wk, 5 wk . 5–8 · (2 min at 110% VO2max: 2 min rest); 3 d/wk, 5 wk
› 1.3* fl 0.5 NS
› 1.6* › 0.6 NS
› 12 NS › 26*
› 5.1* › 5.2*
Serpiello et al.[97] (2011)
10, M, M, F
53.7 – 6.9
3 · ([5 · 4 sec sprint: 16 sec rest]: 4.5 min rest); 3 d/wk, 4 wk (training/tests performed on a non-motorized treadmill)
› 5.5*
› 8.8*
NR
› 2.0
6, MA, M, TS
NR
Control (squad training)
› 0.6 NS fl 0.2 NS
› 1.4 NS › 5.0 NS
› 2 NS › 8 NS
NR
Walklate et al.[98] (2009) Buchheit et al.[99] (2009)
a
NR
Small-sided games (2–4 · 2.5–4 min games) . 12–24 · (15 sec at 105–115% VO2max: 15–20 sec); 2 d/wk, 10 wk
› 3.7* › 3.5*
› 4.6* › 3.4*
› 23 NS › 3 NS
NR
10, MA, M, TS
59.3 – 4.5
Small-sided games 2–6 · (6–13 min games:1–3 min of rest)
9, MA, M, TS
60.2 – 4.6
Generic training (see review for more details)
› 0.6 NS › 1.5 NS
› 0.2 NS fl 0.2 NS
fl 5 NS fl 23 NS
fl 0.7 NS › 2.0 NS
15, MA, M, TS 17, MA, M, TS
Data presented as mean – SD unless NR.
DS = decrement score (or fatigue index); . F = Females; HRmax = maximal heart rate; M = Males; MA = moderate aerobic fitness; max = maximum; NR = not reported; NS = not significant; TS = team-sport athletes; VO2max = maximal oxygen consumption; * indicates significant difference between pre and post (p < 0.05); › indicates improved; fl indicates worsened.
749
Sports Med 2011; 41 (9)
Hill-Haas et al.[100] (2009)
Squad training + 7–15 · (20 sec sprint: 10 sec rest); 2 d/wk, 4 wk
6, MA, M, TS
Training to Improve Repeated-Sprint Ability
ª 2011 Adis Data Information BV. All rights reserved.
Table I. A summary of the characteristics and results of training studies that have investigated changes in repeated-sprint ability following running-based training
Bishop et al.
750
SET ST
b
5.00
5.00
4.90
4.90
4.80
4.80
Times (sec)
Times (sec)
a
4.70 4.60 4.50 4.40
4.70 4.60 4.50 4.40
4.30
4.30 1
2
3
4
5
Sprint number
Mean sprint time
1
2
3
4
Sprint number
5
Mean sprint time
Fig. 3. (a) Pre-training and (b) post-training individual and mean sprint times derived from a repeated-sprint test consisting of five 30 m sprints, separated by 25 sec periods of active recovery during which the subjects jogged back to the starting line. ST performed intermittentsprint training, while SET performed ‘speed-endurance’ training (a type of interval training). [See table II for more details of the training performed]. Post-training there was a significant decrease in initial sprint time for ST only, but a significant decrease in mean sprint time for both groups.[12]
(including the final sprint; 4.5 vs 3.2%) and mean sprint time (4.3% vs 2.4%) [figure 3]. Furthermore, the smaller improvement in the fatigue index by the intermittent-sprint training group is likely to be related to their much-improved first sprint. Thus, it appears that while interval training may be superior at minimizing the decrement during repeated sprints (possibly due to greater physiological adaptations, as outlined in section 2), intermittent- or repeated-sprint training is superior at improving the performance of individual sprints. As a result, the combination of the two (i.e. repeated-sprint training to improve sprint performance plus interval training to improve the recovery between sprints) may be the best strategy to improve RSA. Further research is required to investigate the optimal volume and duration of a repeated-sprint training macro cycle, as anecdotal evidence suggests that too much repeated-sprint training is stressful and may lead to decreases in RSA. 3.2 Sprint Training
Given the improvements in individual sprint times following repeated- and intermittent-sprint training (see table II), a logical question is whether or not similar (or greater) improvements in individual and mean sprint times can be achieved by traditional sprint training (i.e. short sprints interspersed with complete recovery periods[107]). ª 2011 Adis Data Information BV. All rights reserved.
To date, we are unaware of research that has investigated the influence of ‘traditional’ sprint training on RSA. However, it is possible that such training may produce even better improvements in both best sprint time and mean sprint time,[86] and further research is warranted. In support of this, a targeted sprint/agility training protocol (incorporating incomplete rest periods) improved mean sprint time by 2.2% in a group of young soccer players. These changes in mean sprint time were associated with concurrent improvements in single-sprint performance (approximately 2.7% reduction in 10 m sprint time), while no changes in aerobic fitness were observed.[96] Despite the obvious need of more research in this area, these results seem to confirm that in well trained team-sport athletes, maximization of mean repeated-sprint time is linked to improvements in single-sprint performance.[33] 3.3 Small-Sided Games
Recently, there has been an increased emphasis on the use of small-sided games .to improve both team-sport-related fitness (e.g. VO2max, intermittent exercise capacity) and technical skills.[108,109] To date, however, only two studies (table I) have investigated the effects of smallsided-games training on RSA, and both have reported only small, nonsignificant differences in terms of RSE performance enhancement when Sports Med 2011; 41 (9)
Training to Improve Repeated-Sprint Ability
751
compared with generic training.[99,100] For example, when training twice per week for 10 weeks, a similar ~4% improvement in best and mean sprint time has been reported following both smallsided games (2–4 · [2.5–4-minute ‘games’]) and interval training (12–24 · [15 seconds at ~105–115% . VO2max: 15 seconds of rest]).[99] As the small-sided game protocols employed in these studies targeted the development of aerobic fitness, it is likely that the mechanisms responsible for the reported improvements in RSA are also related to improvements in aerobic fitness. In addition, factors other than aerobic fitness, such as neuromuscular factors (e.g. acceleration and turning) that can also be developed with the use of small-sided games, might also explain the observed improvements in RSA.[99,100] However, given the limited research to date, further research is obviously required, especially research comparing the use of smallsided games with other types of training that have previously been demonstrated to improve RSA. Additional research is also required to determine whether small-sided games can be used to improve other factors such as H+ regulation and phosphocreatine resynthesis.
3.4 Resistance Training
While there is good evidence to suggest that resistance training could be beneficial for singlesprint performance,[110-112] the impact of such training on RSA is less clear (table III). To date, three studies have reported that resistance training (2–5 sets of 10–15 maximal repetitions) produces similar increases in mean work during a repeated-sprint test (~12%)[113-115] compared with high-intensity interval training (~13%)[10] or sprint training (~12%).[105] Resistance training also improved both first-sprint performance (8–9%) and the sprint decrement score (~20%).[113,114] The increases in RSA reported in these studies are likely to be accounted for, at least in part, by strength gains. However, factors other than improvements in maximal strength may also be involved as we have observed greater improvements in RSA when sets of resistance training were separated by 20 seconds, compared with 80 seconds of rest, despite half the increase in maximal leg strength (20 vs 46%).[114] This suggests that resistance training that includes a high metabolic load (e.g. blood lactate concentration
Table II. A summary of the characteristics and results of training studies that have investigated changes in cycle repeated-sprint ability (RSA) following different types of training performed on a cycle ergometer Study (y)
Edge et al.[10] (2005)
Subjects
Training programme
type
. VO2max (mL/kg/min)a
10, MA, F
42.4 – 6.3
6–10 · (2 min at 120–140% LT: 1 min rest); 3 d/wk, 5 wk
10, MA, F
41.3 – 7.3
20–30 min at 80–95% LT; 3 d/wk, 5 wk
Bishop and Edge[104] (2005)
11, MA, F
39.0 – 6.4
Glaister et al.[11] (2007)
12, MA, M, TS
46.6 – 4.2
3–12 · (2 min at 130–180% LT: 1 min rest); 3 d/wk + RSA test (5 · 6 sec sprint every 30 sec); 1 d/wk, 8 wk . 20 min at 70% VO2max; 3 d/wk, 6 wk
9, MA, M, TS
52.1 – 3.6
Control (normal recreational activities)
Ortenblad et al.[105] (2000)
9, MA, M
61.3 – 1.7
20 · (10 sec sprint: 50 sec rest); 3 d/wk, 5 wk
6, MA, M
64.0 – 0.5
Control (normal recreational activities)
a
Adaptations sprint 1 (%) [W]
total work (%) [kJ]
DS (%)
. VO2max (%)
› 6.2* › 6.9*
› 13.0*,› 8.5*
› 10 NS fl 16 NS
› 13.2* › 10.4*
› 21.2*
› 28.3*
fl 14 NS
› 14.6*
› 4.0* –
› 9.4* › 1.4 NS
› 46* › 10 NS
› 9.9* –
› 6.6* –
› 12* › 1.0 NS
› 27* –
– –
Data presented as mean – SD.
DS = decrement score (or fatigue index); F = females; LT = lactate threshold (as determined using the. modified Dmax method[106]); M = males; MA = moderate aerobic fitness; NR = not reported; NS = not significant; TS = team-sport athletes; VO2max = maximal oxygen consumption; * indicates significant difference between pre and post (p < 0.05); - indicates significantly greater improvement than the alternate training group; › indicates improved; fl indicates worsened; – indicates no change.
ª 2011 Adis Data Information BV. All rights reserved.
Sports Med 2011; 41 (9)
Bishop et al.
752
Table III. A summary of the characteristics and results of training studies that have investigated changes in cycle repeated-sprint ability (RSA) following different types of resistance training Study (y)
Subjects type
Edge et al.[113] (2006) Hill-Haas et al.[114] (2007)
Robinson et al.[115] (1995)
Training programme
Adaptations
. VO2max (mL/kg/min)a
8, MA, F
42.4 – 9.6
Control
8, MA, F
44.8 – 5.5
6 leg exercises for 2–5 sets · (15–20 RM: 20 sec rest); 3 d/wk, 5 wk
8, MA, F
42.4 – 9.6
6 leg exercises for 2–5 sets · (15–20 RM: 80 sec rest); 3 d/wk, 5 wk
8, MA, F
44.8 – 5.5
6 leg exercises for 2–5 sets · (15–20 RM: 20 sec rest); 3 d/wk, 5 wk
8, MA, M
–
2 leg exercises for 5 sets · (10 RM: 30–180 sec rest); 4 d/wk, 5 wk
. VO2max (%)
sprint 1 (%) [kJ]
total work (%) [kJ]
DS (%)
› 2.7 NS › 8.0 NS
› 3.0 NS › 12.0*,-
› 3 NS – › 22*,–
› 9.3* › 8.4*
› 5.4* › 12.5*,-
› 21* › 23*
– –
› 6.6*
› 8.5*
–
–
Data presented as mean – SD unless no change. . DS = decrement score (or fatigue index); F = females; M = males; MA = moderate aerobic fitness; NS = not significant; VO2max = maximal * oxygen consumption; indicates significant difference between pre and post (p < 0.05); - indicates significantly greater improvement than the alternate training group; › indicates improved; fl indicates worsened; – indicates no change. a
‡10 mmol/L), rather than resistance training which maximizes strength gains (e.g. using 1–4 maximal repetitions), may best improve RSA (possibly via greater improvements in H+ regulation[113]). Further research is required though as the subjects involved in these studies were only moderately trained. Given that success in repeated-sprint activities is also likely to depend on an athlete’s explosive power, further research is also required to investigate the importance of explosive muscle strength training on RSA. 4. Conclusions RSA is an important fitness component of many popular team sports. This review has highlighted that there is not one type of training that can be recommended to best improve RSA and all of the factors believed to be responsible for performance decrements during repeated-sprint tasks. This is not surprising, as RSA is a complex fitness component that depends on both metabolic (e.g. oxidative capacity, phosphocreatine recovery and H+ buffering) and neural factors (e.g. muscle activation and recruitment strategies) among others (figure 4). While different training strategies can be used in order to improve each of these potential limiting factors, and in turn RSA, the concurrent implementation of different forms of ª 2011 Adis Data Information BV. All rights reserved.
training may be the best strategy to improve RSA. However, the currently unknown synergies and interferences resulting from the combination of various training contents[116] on the metabolic, neural and mechanical determinants of RSA make guidelines on how training content should be manipulated and periodized difficult. Nonetheless, two key recommendations can be made based on the existing literature as follows: 1. It is important to include some training to improve single-sprint performance. This should include (i) specific sprint training; (ii) strength/ power . training; and (iii) occasional high-intensity (>VO2max) training (e.g. repeated, 30-second, allout efforts separated by ~10 minutes of recovery) to increase the anaerobic capacity. 2. It is also important to include some interval training to best improve the ability to recover between sprints (if the goal is to improve fatigue . resistance). High-intensity (80–90% VO2max) interval training, interspersed with rest periods (e.g. 1 minute) that are shorter than the work periods (e.g. 2 minutes) is efficient at improving the ability to recover . between sprints by increasing aerobic fitness (VO2max and the lactate threshold), the rate of phosphocreatine resynthesis and bmin vitro. In support of the above recommendation, to date, the greatest improvements in both single and mean sprint performance have been reported Sports Med 2011; 41 (9)
Training to Improve Repeated-Sprint Ability
753
Repeated-sprint ability
Initial sprint performance
Stride length
ATP supply
Strength
Power
Recovery between sprints
Stride frequency
Flexibility
Neural co-ordination
PCr resynthesis
Aerobic fitness
Muscle buffering
Elastic strength
Fig. 4. A summary of factors which should be targeted by training to improve repeated-sprint ability. ATP = adenosine triphosphate; PCr = phosphocreatine.
after training that included both high-intensity interval training and repeated sprints.[104] For most athletes, it is probably impossible to perform all of the above-described training concurrently. It is therefore paramount that a periodized training programme, designed to improve RSA, is structured such that different aspects are emphasized, at different times, in accordance with the competitive demands of each particular sport and the strengths and weaknesses of the individual athlete. As RSA requires a unique blend of power (sprint speed) and endurance (recovery between sprints), it needs to be established whether it is best to develop these qualities separately, or whether they can be developed concurrently (without interference effects). Future studies also need to address whether training-induced changes in RSA actually impact upon field performance. More importantly, as many studies to date have used untrained subjects and/or a cycle ergometer, future research must recruit highly-trained teamsport athletes and be expanded to sport-specific test settings with, in parallel, a high level of standardization and reliability of the measures. Acknowledgements The authors have no conflicts of interest that are directly relevant to the content of this review. No funding was used to assist in the preparation of this review.
ª 2011 Adis Data Information BV. All rights reserved.
References 1. Spencer M, Bishop D, Dawson B, et al. Physiological and metabolic responses of repeated-sprint activities: specific to field-based team sports. Sports Med 2005; 35: 1025-44 2. Girard O, Mendez-Villanueva A, Bishop D. Repeatedsprint ability – part I: factors contributing to fatigue. Sports Med 2011; 41 (8): 673-94 3. Rampinini E, Bishop D, Marcora SM, et al. Validity of simple field tests as indicators of match-related physical performance in top-level professional soccer players. Int J Sports Med 2007; 28: 228-35 4. Ferrari Bravo D, Impellizzeri FM, Rampinini E, et al. Sprint vs. interval training in football. Int J Sports Med 2008; 29: 668-74 5. Spencer M, Bishop D, Lawrence S. Longitudinal assessment of the effects of field-hockey training on repeated sprint ability. J Sci Med Sport 2004; 7: 323-34 6. Perrey S, Racinais S, Saimouaa K, et al. Neural and muscular adjustments following repeated running sprints. Eur J Appl Physiol 2010; 109 (6): 1027-36 7. Impellizzeri FM, Rampinini E, Castagna C, et al. Validity of a repeated-sprint test for football. Int J Sports Med 2008; 29: 899-905 8. Bishop D. Game sense or game nonsense? J Sci Med Sport 2009; 12: 426-7 9. Buchheit M, Millet GP, Parisy A, et al. Supramaximal training and postexercise parasympathetic reactivation in adolescents. Med Sci Sports Exerc 2008; 40: 362-71 10. Edge J, Bishop D, Goodman C. Effects of high- and moderate-intensity training on metabolism and repeated sprints. Med Sci Sports Exerc 2005; 37: 1975-82 11. Glaister M, Stone MH, Stewart AM, et al. The influence of endurance training on multiple sprint cycling performance. J Strength Cond Res 2007; 21: 606-12 12. Mohr M, Krustrup P, Nielsen JJ, et al. Effect of two different intense training regimens on skeletal muscle ion
Sports Med 2011; 41 (9)
Bishop et al.
754
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
transport proteins and fatigue development. Am J Physiol Regul Integr Comp Physiol 2007; 292: R1594-602 Glaister M. Multiple sprint work: physiological responses, mechanisms of fatigue and the influence of aerobic fitness. Sports Med 2005; 35: 757-77 Racinais S, Bishop D, Denis R, et al. Muscle deoxygenation and neural drive to the muscle during repeated sprint cycling. Med Sci Sports Exerc 2007; 39: 268-74 Mendez-Villanueva A, Hamer P, Bishop D. Fatigue responses during repeated sprints matched for initial mechanical output. Med Sci Sports Exerc 2007; 39: 2219-25 Dawson B, Goodman C, Lawrence S, et al. Muscle phosphocreatine repletion following single and repeated short sprint efforts. Scand J Med Sci Sports 1997; 7: 206-13 Bogdanis GC, Nevill ME, Boobis LH, et al. Contribution of phosphocreatine and aerobic metabolism to energy supply during repeated sprint exercise. J Appl Physiol 1996; 80: 876-84 Bogdanis GC, Nevill ME, Boobis LH, et al. Recovery of power output and muscle metabolites following 30 s of maximal sprint cycling in man. J Physiol 1995; 482 (Pt 2): 467-80 Haseler LJ, Hogan MC, Richardson RS. Skeletal muscle phosphocreatine recovery in exercise-trained humans is dependent on O2 availability. J Appl Physiol 1999; 86: 2013-8 Yoshida T, Watari H. 31P-Nuclear magnetic resonance spectroscopy study of the time course of energy metabolism during exercise and recovery. Eur J Appl Physiol 1993; 66: 494-9 McCully KK, Boden BP, Tuchler M, et al. Wrist flexor muscles of elite rowers measured with magnetic resonance spectroscopy. J Appl Physiol 1989; 67 (3): 926-32 McCully KK, Vandenborne K, DeMeirleir K, et al. Muscle metabolism in track athletes, using 31P magnetic resonance spectroscopy. Can J Physiol Pharmacol 1992; 70: 1353-9 Yoshida T, Watari H. Metabolic consequences of repeated exercise in long distance runners. Eur J Appl Physiol 1993; 67: 261-5 McCully KK, Kakihira H, Vandenborne K, et al. Noninvasive measurements of activity-induced changes in muscle metabolism. J Biomech 1991; 21: 153-61 Bishop D, Edge J, Thomas C, et al. Effects of high-intensity training on muscle lactate transporters and postexercise recovery of muscle lactate and hydrogen ions in women. Am J Physiol Regul Integr Comp Physiol 2008; 295: R1991-8 Stathis CG, Febbraio MA, Carey MF, et al. Influence of sprint training on human skeletal muscle purine nucleotide metabolism. J Appl Physiol 1994; 76 (4): 1802-9 Schneiker K, Bishop D. The effects oh high-intensity interval training vs intermittent sprint training on physiological capacities important for team sport performance. In: Burnett A, editor. Science and nutrition in exercise and sport. Melbourne (VIC): Exerc Sport Sci Aust, 2008 Crowther GJ, Carey MF, Kemper WF, et al. Control of glycolysis in contracting muscle. I: turning it on. Am J Physiol 2002; 282: E67-73 Gaitanos GC, Williams C, Boobis LH, et al. Human muscle metabolism during intermittent maximal exercise. J Appl Physiol 1993; 75 (2): 712-9 Sahlin K, Ren JM. Relationship of contraction capacity to metabolic changes during recovery from a fatiguing contraction. J Appl Physiol 1989; 67: 648-54
ª 2011 Adis Data Information BV. All rights reserved.
31. Mendez-Villanueva A, Hamer P, Bishop D. Fatigue in repeated-sprint exercise is related to muscle power factors and reduced neuromuscular activity. Eur J Appl Physiol 2008; 103: 411-9 32. Bishop D, Lawrence S, Spencer M. Predictors of repeatedsprint ability in elite female hockey players. J Sci Med Sport 2003; 6: 199-209 33. Pyne DB, Saunders PU, Montgomery PG, et al. Relationships between repeated sprint testing, speed, and endurance. J Strength Cond Res 2008; 22: 1633-7 34. Bishop D, Schneiker KT. Different interpretation of the effect of two different intense training regimens on repeated sprint ability [letter]. Am J Physiol 2007; 293 (3): R1459 35. Jacobs I, Esbjornsson M, Sylven C, et al. Sprint training effects on muscle myoglobin, enzymes, fibre types, and blood lactate. Med Sci Sports Exerc 1987; 19: 368-74 36. Parra J, Cadefau JA, Rodas G, et al. The distribution of rest periods affects performance and adaptations of energy metabolism induced by high-intensity training in human muscle. Acta Physiol Scand 2000; 169: 157-65 37. Harmer AR, McKenna MJ, Sutton JR, et al. Skeletal muscle metabolic and ionic adaptations during intense exercise following sprint training in humans. J Appl Physiol 2000; 89: 1793-803 38. Medbo JI, Burgers S. Effect of training on the anaerobic capacity. Med Sci Sports Exerc 1990; 22: 501-7 39. Weber CL, Schneider DA. Increases in maximal accumulated oxygen deficit after high-intensity interval training are not gender dependent. J Appl Physiol 2002; 92: 1795-801 40. Tabata I, Nishimura K, Kouzaki M, et al. Effects of moderate-intensity endurance and high-intensity intermittent training on anaerobic capacity and VO2max. Med Sci Sports Exerc 1996; 28: 1327-30 41. Costill DL, Coyle EF, Fink WF, et al. Adaptations in skeletal muscle following strength training. J Appl Physiol 1979; 46: 96-9 42. Phillips SM, Green HJ, Tarnopolsky MA, et al. Progressive effect of endurance training on metabolic adaptations in working skeletal muscle. Am J Physiol 1996; 270 (2 Pt 1): E265-72 43. Dawson B, Fitzsimons M, Green S, et al. Changes in performance, muscle metabolites, enzymes and fibre types after short sprint training. Eur J Appl Physiol 1998; 78: 163-9 44. Rodas G, Ventura JL, Cadefau JA, et al. A short training programme for the rapid improvement of both aerobic and anaerobic metabolism. Eur J Appl Physiol 2000; 82: 480-6 45. Barnett C, Carey M, Proietto J, et al. Muscle metabolism during sprint exercise in man: influence of sprint training. J Sci Med Sport 2004; 7: 314-22 46. Linossier MT, Dormois D, Perier C, et al. Enzyme adaptations of human skeletal muscle during bicycle shortsprint training and detraining. Acta Physiol Scand 1997; 161: 439-45 47. MacDougall JD, Hicks AL, MacDonald JR, et al. Muscle performance and enzymatic adaptations to sprint interval training. J Appl Physiol 1998; 84 (6): 2138-42 48. Nevill ME, Boobis LH, Brooks ST, et al. Effect of training on muscle metabolism during treadmill sprinting. J Appl Physiol 1989; 67: 2376-82
Sports Med 2011; 41 (9)
Training to Improve Repeated-Sprint Ability
49. Thomas C, Sirvent P, Perrey S, et al. Relationships between maximal muscle oxidative capacity and blood lactate removal after supramaximal exercise and fatigue indexes in humans. J Appl Physiol 2004; 97: 2132-8 50. Dupont G, Millet GP, Guinhouya C, et al. Relationship between oxygen uptake kinetics and performance in repeated running sprints. Eur J Appl Physiol 2005; 95: 27-34 51. Rampinini E, Sassi A, Morelli A, et al. Repeated-sprint ability in professional and amateur soccer players. Appl Physiol Nutr Metab 2010; 34: 1048-54 52. Buchheit M, Ufland P. Effect of endurance training on performance and muscle reoxygenation rate during repeatedsprint running. Eur J Appl Physiol 2011; 111 (2): 293-301 53. Fernandes da Silva J, Guglielmo LGA, Bishop D. Relationship between different measures of aerobic fitness and repeated-sprint ability in elite soccer players. J Strength Cond Res 2010; 24: 2115-21 54. Bishop D, Edge J, Goodman C. Muscle buffer capacity and aerobic fitness are associated with repeated-sprint ability in women. Eur J Appl Physiol 2004; 92: 540-7 55. Bishop D, Spencer M. Determinants of repeated-sprint ability in well-trained team-sport athletes and endurancetrained athletes. J Sports Med Phys Fitness 2004; 44: 1-7 56. McMahon S, Wenger HA. The relationship between aerobic fitness and both power output and subsequent recovery during maximal intermittent exercise. J Sci Med Sport 1998; 1 (4): 219-27 57. Tomlin DL, Wenger HA. The relationship between aerobic fitness, power maintenance and oxygen consumption during intense intermittent exercise. J Sci Med Sport 2002; 5 (3): 194-203 58. Bishop D, Edge J. Determinants of repeated-sprint ability in females matched for single-sprint performance. Eur J Appl Physiol 2006; 97: 373-9 59. McGawley K, Bishop D. Anaerobic and aerobic contribution to two, 5 · 6-s repeated-sprint bouts [abstract]. Coach Sport Sci J 2008; 3: 52 60. Hoffman JR. The relationship between aerobic fitness and recovery from high-intensity exercise in infantry soldiers. Mil Med 1997; 162: 484-8 61. Daussin FN, Zoll J, Dufour SP, et al. Effect of interval versus continuous training on cardiorespiratory and mitochondrial functions: relationship to aerobic performance improvements in sedentary subjects. Am J Physiol Regul Integr Comp Physiol 2008; 295: R264-72 62. MacDougall D, Sale D. Continuous vs. interval training: a review for the athlete and the coach. Can J Appl Sport Sci 1981; 6: 93-7 63. Gorostiaga EM, Walter CB, Foster A, et al. Uniqueness of interval and continuous training at the same maintained exercise intensity. Eur J Appl Physiol 1991; 63: 101-7 64. Helgerud J, Hoydal K, Wang E, et al. Aerobic highintensity intervals improve VO2max more than moderate training. Med Sci Sports Exerc 2007; 39: 665-71 65. Eversten F, Medbo JI, Bonen A. Effect of training intensity on muscle lactate transporters and lactate threshold of crosscountry skiers. Acta Physiol Scand 2001; 173: 195-205 66. Cunningham DA, McCrimmon D, Vlach LF. Cardiovascular response to interval and continuous training in women. Eur J Appl Physiol 1979; 41: 187-97
ª 2011 Adis Data Information BV. All rights reserved.
755
67. Eddy DO, Sparks KL, Adelizi DA. The effects of continuous and interval training in women and men. Eur J Appl Physiol 1977; 37: 83-92 68. Poole DC, Gaesser GA. Response of ventilatory and lactate thresholds to continuous and interval training. J Appl Physiol 1985; 58: 1115-21 69. Edge J, Bishop D, Goodman C. The effects of training intensity on muscle buffer capacity in females. Eur J Appl Physiol 2006; 96: 97-105 70. Spencer M, Dawson B, Goodman C, et al. Performance and metabolism in repeated sprint exercise: effect of recovery intensity. Eur J Appl Physiol 2008; 103: 545-52 71. Ratel S, Williams CA, Oliver J, et al. Effects of age and recovery duration on performance during multiple treadmill sprints. Int J Sports Med 2005; 26: 1-8 72. Spriet LL, Lindinger MI, Mckelvie RS, et al. Muscle glycogenolysis and H+ concentration during maximal intermittent cycling. J Appl Physiol 1989; 66 (1): 8-13 73. Bishop D, Edge J, Davis C, et al. Induced metabolic alkalosis affects muscle metabolism and repeated-sprint ability. Med Sci Sports Exerc 2004; 36: 807-13 74. Juel C. Muscle pH regulation: role of training. Acta Physiol Scand 1998; 162: 359-66 75. Weston AR, Myburgh KH, Lindsay FH, et al. Skeletal muscle buffering capacity and endurance performance after high-intensity interval training by well-trained cyclists. Eur J Appl Physiol 1997; 75: 7-13 76. Juel C, Klarskov C, Nielsen JJ, et al. Effect of highintensity intermittent training on lactate and H+ release from human skeletal muscle. Am J Physiol Endocrinol Metab 2004; 286: E245-51 77. Thomas C, Bishop D, Moore-Morris T, et al. Effects of highintensity training on MCT1, MCT4, and NBC expressions in rat skeletal muscles: influence of chronic metabolic alkalosis. Am J Physiol Endocrinol Metab 2007; 293: E916-22 78. Edge J, Bishop D, Goodman C. Effects of chronic NaHCO3 ingestion during interval training on changes to muscle buffer capacity, metabolism, and short-term endurance performance. J Appl Physiol 2006; 101: 918-25 79. Sahlin K, Harris RC, Nylind B, et al. Lactate content and pH in muscle obtained after dynamic exercise. Pflugers Archiv 1976; 367: 143-9 80. Gibala MJ, Little JP, van Essen M, et al. Short-term sprint interval versus traditional endurance training: similar initial adaptations in human skeletal muscle and exercise performance. J Physiol 2006; 575: 901-11 81. Bonen A, McCullagh KJA, Putman CT, et al. Short-term training increases human muscle MCT1 and femoral venous lactate in relation to muscle lactate. Am J Physiol Endocrinol Metab 1998; 274: E102-7 82. Dubouchaud H, Butterfield GE, Wolfel EE, et al. Endurance training, expression, and physiology of LDH, MCT1, and MCT4 in human skeletal muscle. Am J Physiol Endocrinol Metab 2000; 278: E571-9 83. Juel C, Holten MK, Dela F. Effects of strength training on muscle lactate release and MCT1 and MCT4 content in healthy and type 2 diabetic humans. J Physiol 2004; 556 (1): 297-304 84. Burgomaster KA, Cermak NM, Phillips SM, et al. Divergent response of metabolite transport proteins in human skeletal
Sports Med 2011; 41 (9)
Bishop et al.
756
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
muscle after sprint interval training and detraining. Am J Physiol Regul Integr Comp Physiol 2007; 292: R1970-6 Pilegaard H, Domino K, Noland T, et al. Effect of highintensity exercise training on lactate/hydrogen ion transport capacity in human skeletal muscle. Am J Physiol 1999; 276: E255-61 Ross A, Leveritt M, Riek S. Neural influences on sprint running: training adaptations and acute responses. Sports Med 2001; 31: 409-25 Matsuura R, Arimitsu T, Kimura T, et al. Effect of oral administration of sodium bicarbonate on surface EMG activity during repeated cycling sprints. 2007; 101: 409-17 Billaut F, Basset FA, Giacomoni M, et al. Effect of highintensity intermittent cycling sprints on neuromuscular activity. Int J Sports Med 2006; 27: 25-30 Billaut F, Basset FA, Falgairette G. Muscle coordination changes during intermittent cycling sprints. Neurosci Lett 2005; 380: 265-9 Gabriel DA, Kamen G, Frost G. Neural adaptations to resistive exercise: mechanisms and recommendations for training practices. Sports Med 2006; 36: 133-49 Mikkola J, Rusko H, Nummela A, et al. Concurrent endurance and explosive type strength training improves neuromuscular and anaerobic characteristics in young distance runners. Int J Sports Med 2007; 28: 602-11 Murray DP, Brown LE, Zinder SM, et al. Effects of velocity-specific training on rate of velocity development, peak torque, and performance. J Strength Cond Res 2007; 21: 870-4 Van Cutsem M, Duchateau J, Hainaut K. Changes in single motor unit behaviour contribute to the increase in contraction speed after dynamic training in humans. J Physiol 1998; 513 (Pt 1): 295-305 Aagaard P, Simonsen EB, Andersen JL, et al. Increased rate of force development and neural drive of human skeletal muscle following resistance training. J Appl Physiol 2002; 93: 1318-26 Del Balso C, Cafarelli E. Adaptations in the activation of human skeletal muscle induced by short-term isometric resistance training. J Appl Physiol 2007; 103: 402-11 Buchheit M, Mendez-Villaneuva A, Quod M, et al. Improving acceleration and repeated sprint ability in well-trained adolescent handball players: speed vs sprint interval training. Int J Sports Physiol Perform 2010; 5: 152: 64 Serpiello FR, McKenna MJ, Stepto NK, et al. Performance and physiological responses to repeated-sprint exercise: a novel multiple-set approach. Eur J Appl Physiol 2011; 111 (4): 669-78 Walklate BM, O’Brien BJ, Paton CD, et al. Supplementing regular training with short-duration sprint-agility training leads to a substantial increase in repeated sprint-agility performance with national level badminton players. J Strength Cond Res 2009; 23: 1477-81 Buchheit M, Laursen PB, Kuhnle J, et al. Game-based training in young elite handball players. Int J Sports Med 2009; 30: 251-8 Hill-Haas SV, Coutts AJ, Rowsell GJ, et al. Generic versus small-sided game training in soccer. Int J Sports Med 2009; 30: 636-42
ª 2011 Adis Data Information BV. All rights reserved.
101. Helgerud J, Engen LC, Wisloff U, et al. Aerobic endurance training improves soccer performance. Med Sci Sports Exerc 2001; 33 (11): 1925-31 102. Billaut F, Bishop D. Muscle fatigue in males and females during multiple-sprint exercise. Sports Med 2008; 39: 257-78 103. Mohr M, Krustrup P, Nielsen JJ, et al. Reply to Bishop and Schneiker [letter]. Am J Physiol Regul Integr Comp Physiol 2007; 293: R1460 104. Bishop D, Edge J. The effects of a 10-day taper on repeated-sprint performance in females. J Sci Med Sport 2005; 8: 200-9 105. Ortenblad N, Lunde PK, Levin K, et al. Enhanced sarcoplasmic reticulum calcium release following intermittent sprint training. Am J Physiol 2000; 279: R152-60 106. Bishop D, Jenkins DG, Mackinnon LT. The relationship between plasma lactate parameters, Wpeak and 1-h cycling performance in women. Med Sci Sports Exerc 1998; 30 (8): 1270-5 107. Ross A, Leveritt M. Long term metabolic and skeletal muscle adaptations to short-sprint training: implications for sprint training and taper. Sports Med 2001; 31 (15): 1063-82 108. Impellizzeri FM, Marcora SM, Castagna C, et al. Physiological and performance effects of generic versus specific aerobic training in soccer players. Int J Sports Med 2006; 27 (6): 483-92 109. Gabbett TJ. Performance changes following a field conditioning program in junior and senior rugby league players. J Strength Cond Res 2006; 20: 215-21 110. Delecluse C, Van Coppenolle H, Willems E, et al. Influence of high-resistance and high-velocity training on sprint performance. Med Sci Sports Exerc 1995; 27: 1203-9 111. Delecluse C. Influence of strength training on sprint running performance: current findings and implications for training. Sports Med 1997; 24: 147-56 112. Newman MA, Tarpenning KM, Marino FE. Relationships between isokinetic knee strength, single-sprint performance, and repeated-sprint ability in football players. J Strength Cond Res 2004; 18: 867-72 113. Edge J, Hill-Haas S, Goodman C, et al. Effects of resistance training on H+ regulation, buffer capacity, and repeated sprints. Med Sci Sports Exerc 2006; 38: 2004-11 114. Hill-Haas S, Bishop D, Dawson B, et al. Effects of rest interval during high-repetition resistance training on strength, aerobic fitness, and repeated-sprint ability. J Sports Sci 2007; 25 (6): 619-28 115. Robinson JM, Stone MH, Johnson RL, et al. Effects of different weight training exercise/rest intervals on strength, power and high intensity exercise endurance. J Strength Cond Res 1995; 9 (4): 216-21 116. Coffey VG, Jemiolo B, Edge J, et al. Effect of consecutive repeated sprint and resistance exercise bouts on acute adaptive responses in human skeletal muscle. Am J Physiol Regul Integr Comp Physiol 2009; 297: R1441-51
Correspondence: Prof. David Bishop, Institute of Sport, Exercise and Active Living (ISEAL), School of Sport and Exercise Science, Victoria University, PO Box 14428 Melbourne, VIC 8001, Australia. E-mail:
[email protected]
Sports Med 2011; 41 (9)
Sports Med 2011; 41 (9): 757-771 0112-1642/11/0009-0757/$49.95/0
REVIEW ARTICLE
ª 2011 Adis Data Information BV. All rights reserved.
Induction and Decay of Short-Term Heat Acclimation in Moderately and Highly Trained Athletes Andrew T. Garrett,1 Nancy J. Rehrer1 and Mark J. Patterson2 1 School of Physical Education, University of Otago, Dunedin, New Zealand 2 Defence Science Technology Organisation (DSTO), Melbourne, VIC, Australia
Contents Abstract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Literature Search Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Short-Term Heat Acclimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Method of Acclimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Heat-Acclimation and Cardiovascular Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. Effectiveness of Dehydration on Adaptation to the Heat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6. Heat-Shock Proteins and Acclimation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7. Time Course of Acclimation Decay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8. Highly Trained Athletes and Heat Acclimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Abstract
757 758 758 759 761 762 763 764 765 767
A rethinking of current heat-acclimation strategies is required as most research and advice for improving physiological strain in the heat includes maintaining hydration using long-term acclimation protocols (>10 days). Furthermore, these strategies have tended to use untrained and moderately trained participants. Therefore, the aims of this review were to (i) investigate the effectiveness of short-term heat acclimation (STHA) with moderately and highly trained athletes; (ii) determine the importance of fluid regulatory strain, which has a thermally independent role in heat adaptation; (iii) assess the impact of STHA on a marker of thermotolerance (inducible heat-shock protein 70 [HSP70]); and (iv) provide further information on the decay of acclimation to heat. The review suggests that 5-day STHA is effective, and adaptations may be more pronounced after fluid regulatory strain from a dehydration-acclimation regimen. Furthermore, highly trained athletes may have similar physiological gains to those who are less trained using STHA. However, research has tended to focus on untrained or moderately trained participants and more information is required for highly trained populations. HSP70 response is upregulated across STHA. This indicates increased thermotolerance and protective adaptive change that may indicate HSP70 response as a useful marker of heat acclimation. Physiological adaptations after heat acclimation are relatively short term and may vanish only a few days or
Garrett et al.
758
weeks after removal from heat exposure. From a practical perspective 5-day STHA may be the preferred acclimation regimen for moderately and highly trained athletes as it has been shown to be effective, less expensive and less likely to disrupt the tapering for competition in elite performers. Furthermore, updated information on the time course of acclimation decay may allow a reliable estimate of how long individuals can be free from heat exposure before reacclimation is required. This is particularly pertinent in present times as many athletes, civilians and military personnel increasingly have to relocate to different climates of the world, often within a short period of time.
The physiological benefits of heat acclimation are well established, and include a greater cardiovascular stability during exercise in the heat.[1-5] This is indicated by decreased heart rate and increased cardiac output at the same relative intensity-enhancing exercise performance in the heat.[6] The physiological adaptations characteristic of heat acclimation include attenuated body core temperature,[7] a more dilute sweat, earlier onset of sweating and plasma volume expansion.[8] Failure to adequately prepare for hot and humid conditions will result in the earlier onset of fatigue. One of the major factors that must be considered is hydration status.[9] The potential, detrimental effects on performance and possibly on health during prolonged exercise in the heat can be reduced by adequate fluid ingestion.[10-13] Most research on adaptation to heat stress has tended to use long-term acclimation protocols. However, many of the important adaptations to heat stress are cardiovascular, and these occur relatively rapidly.[14] Therefore, using a shortterm heat acclimation (STHA) protocol of 5-days duration may be more appropriate as there is clear evidence that the effects of moderate heat acclimation can be induced by such a regimen.[14-17] The stimulus of heat acclimation is therefore affected by the length of exposure, acclimation protocol and thermal stress. Examples of short(<7 days), medium- (8–14 days) and long-term (>15 days) acclimation regimens are shown in table I. The aims of this review are to (i) investigate the effectiveness of short-term heat acclimation (STHA) with moderately and highly trained athletes; (ii) determine the importance of fluid regª 2011 Adis Data Information BV. All rights reserved.
ulatory strain, which has a thermally independent role in heat adaptation; (iii) assess the impact of STHA on a marker of thermotolerance (inducible heat-shock protein 70 [HSP70]); and (iv) provide further information on the decay of acclimation to heat. 1. Literature Search Methodology No articles were dismissed in this review but, where appropriate, comment has been made on the limitations of the study presented, i.e., limited subject numbers and confounding methodological issues. Historically based journals have been included where appropriate to indicate the limitations in current research; for example, the decay of acclimation. Search periods have been ongoing since 2001 and have been updated to July 2011. Use of the terms ‘short-term heat acclimation’, ‘cardiovascular stress’, ‘dehydration’, ‘moderately/ highly trained’ and ‘HSP70+exercise’ were employed. Search engines included PubMed, Web of Knowledge and directly from the appropriate academic journals: the Journal of Applied Physiology, the Journal of Physiology, the European Journal of Applied Physiology and Medicine and Science in Sports and Exercise. 2. Short-Term Heat Acclimation The adaptive effects of medium- to long-term heat acclimation (>8–12 days) have received much research attention[16,22-26] and it is well established that the physiological responses before and after heat acclimation are dependent on the length of exposure to heat-stress conditions.[31,32] This is an Sports Med 2011; 41 (9)
Short-Term Heat Acclimation
759
Table I. Short-, medium- and long-term heat-acclimation regimens Study
Time (days)
Daily heat-acclimation protocol
Thermal stress
5
90 min; HC (HT)
40C, 60% RH
5
90 min; HC (MT)
40C, 60% RH
STHA (<7days) Garrett et al.[18] [14]
Garrett et al.
4
30–45 min; LIST
30C, 24% RH
4
60 min; step test/HSP70
28C
Patterson et al.[16]
7
90 min; HC
40C, 60% RH
Cotter et al.[15]
6
70 min; HC
39C, 59% RH
5
2 h; HC
40–50C, WBGT
Sunderland et al.
[19]
Kresfelder et al.[20] WBGT
Turk and Worsley
[17]
MTHA (8–14 days) Lorenzo et al.[21]
Weller and Harrison
10
. 90 min; 50% VO2max . 90 min; 50% VO2max
10
70 min; HC/NBC
10
[22]
38C, 30% RH 13C, 30% RH 35C, 18% RH
Regan et al.
10
60 min; HC
39C, 40% RH
Cheung and McLellan[24]
10
40C, 30% RH
Neilsen et al.[25]
8–13
Neilsen et al.[26]
9–12 9
1 h; 3.5 km/h; NBC . 45 min; 45% VO2max . 90 min; 60% VO2max . 50–75% VO2max
10
120 min; 1.34 m/sec
Patterson et al.[16]
22
90 min; HC
40C, 60% RH
Levi et al.[29]
60
Rat model/CE
34C
Levi et al.[30]
30
Rat model/CE
34C
[23]
Houmard et al. Shapiro et al.
[27]
[28]
35C, 87% RH 40C, 10% RH 40C, 27% RH 40C, 30% RH
LTHA (‡15 days)
CE = continuous exposure; HC = hyperthermia controlled (36.5–38.8C); HT = highly trained; HSP70 = heat-shock protein 70; LIST = Loughborough Intermittent Shuttle Test; LTHA = long-term heat acclimation; MT = moderately trained; MTHA = medium-term .heat acclimation; NBC = nuclear, biological and chemical protective clothing; RH = relative humidity; STHA = short-term heat acclimation; VO2max = maximal oxygen consumption; WBGT = Wet Bulb Globe Temperature.
important consideration for using short- or longterm acclimation regimens. For example, increased sweating power is an adaptive response of commonly employed long-term (12–14 days) heatacclimation regimens.[33] Cardiovascular capacity occurs during a shorter acclimation period (<7 days) and this can enable greater heat transfer to and from the skin.[26] Therefore, long-term acclimation protocols may dehydrate people faster and may not be beneficial for all participants. For example, it has been reported that the continued secretion of sweat does not facilitate cooling in workers wearing protective clothing, leading to a more rapid dehydration and work performance decrement.[34] This supports the notion, in certain situations, of using STHA regimens (<7 days) as a model for future research.[14,16,19,20,22] From a practical perspective, completing an STHA ª 2011 Adis Data Information BV. All rights reserved.
regimen for some athletes is advantageous when sustaining quality training and tapering performance several weeks before competition. 3. Method of Acclimation The method of the heat-acclimation regimen is an important consideration and can generally be grouped into the following three types: (i) constant work-rate methods; (ii) self-regulated exercise; and (iii) controlled hyperthermia.[35] The most commonly used method is employing a constant work rate[36] and has been typically used in evaluating performance in the military.[37] However, this procedure should be viewed with caution, as the relative load placed upon participants will be different; hence, variability in the physiological strain and the adaptive response. Self-regulated Sports Med 2011; 41 (9)
Garrett et al.
760
methods allow participants to select their own work rates on the basis of their level of physical conditioning and perceived stress of the acclimation sessions.[38] This method has greater practical application than research focus, as it is difficult to determine that the same workload is placed upon all participants in the study. In contrast, controlled hyperthermia ensures equal thermal strain is placed upon participants, as it involves elevating and maintaining a steady-state body temperature above the sweating threshold using exercise. This is an important consideration when choosing a more complete heat-acclimation method as it has been established that in the absence of an elevated body temperature, exercise by itself will be an inadequate stimulus for adaptation.[39] The controlled hyperthermia method is referred to as the isothermal model of heat acclimation.[40] Table II illustrates the physiological adaptations at rest and end-exercise induced from short- (<7 days), medium- (8–14 days) and long-term (>15 days) acclimation using the controlled hyperthermia technique. This supports the benefits of using this method in heat acclimation. In comparison with the constant and selfregulated methods, the controlled hyperthermia technique may offer a more complete adaptation[35] but it has received limited attention in the litera-
ture. From the information available, body temperature has been elevated and controlled using vapour-barrier suits[41] and it has been employed as a method to acclimate soldiers in the British military.[17,42] Turk and Worsley,[17] with a large sample of participants (n = 51), used the controlled hyperthermia technique for 5 days in a hot environment (36C Wet Bulb Globe Temperature [WBGT]), for 2 hours at a rectal temperature of 38.8C. This resulted in mean 80 – SD5% of participants reaching a satisfactory level of acclimatization that was indicated by greater cardiovascular stability. This procedure was later adapted into a work-rest protocol.[43] More recently, the isothermal strain model has been used to evaluate the importance of skin temperature after medium-term (10-day) heat acclimation.[23] Furthermore, reduced thermal strain and increased work capacity after medium-term (10-day) heat acclimation, using the controlled hyperthermia technique have been reported.[22] Participants (n = 10) maintained an aural temperature between 38C and 38.5C for 70 minutes on each day, by water immersion (40C) and cycle exercise. Patterson[44] reported that STHAs of 7 days (n = 12), in hot and humid conditions (40C, 60% relative humidity [RH]), undertaken using controlled hyperthermia in 90-minute daily bouts,
Table II. Physiological adaptations at rest and end-exercise induced by short- (<7 days), medium- (8–14 days) and long-term (‡15 days) heat acclimation using the controlled hyperthermia techniquea Study
Plasma volume (%)
Rectal temperature (%)
Mean skin temperature
Plasma volume (%)
Mean sweat rate (%)
Work capacity (%)
Garrett et al.[18] b
fl 7.5
fl 0.8
› 4.5
› 1.5
Garrett et al.[14] c
fl 7.0
fl 0.7
› 4.2
› 14.0
Patterson et al.[16]
fl 6.8 R
fl 0.5 R
› 9.8
STHA (<7 days)
Cotter et al.[15]
fl 6.1
fl 0.5 R
Weller and Harrison[22]
fl 2.3
fl 0.5
Patterson et al.[16]
fl 13.5 R
fl 0.8 R
Turk and Worsley[17]
a
2
› 8.3 2
2 2
fl 2.7 2
› 11.6 › 9.8
› 13.3
Unless otherwise shown as R, the variables reviewed are at end exercise.
b
Subjects were highly trained.
c
Subjects were moderately trained.
R = resting; STHA = short-term heat acclimation; › indicates variable increase (%); fl indicates variable decrease (%); 2 indicates variable unchanged (%).
ª 2011 Adis Data Information BV. All rights reserved.
Sports Med 2011; 41 (9)
Short-Term Heat Acclimation
conferred much of the physiological and performance improvement that was evident after 21 days of acclimation. In general, prior to heat-acclimation regimens, participants should refrain from strenuous exercise immediately before and 24 hours prior to exposure, as it has been demonstrated that lower resting-core temperature contributes to reduced physiological strain during acclimation.[45] Consistency in acclimation bouts is also important to control for changes in core temperature and sweating response that may occur if repeated heat exposures were to take place at different times of the day.[46] Specifically, Shido et al.[46] reported that the effect of acclimation on resting rectal temperature was specific to the time of day of acclimation. 4. Heat-Acclimation and Cardiovascular Stability The effects of heat acclimation on gross physiological measures and exercise performance have received much attention,[26,36,44,47,48] but little information exists on the relative importance and mechanisms responsible for the adaptations that occur after repeated exposure to heat-stress conditions. A principal feature of heat acclimation is increased cardiovascular stability,[3,36] indicated especially by lower exercising heart rate,[7] which is often concurrent with hypervolaemia.[6,49] Plasma volume expansion and decreasing exercise heart rate are two rapidly occurring responses to heat acclimation,[16,44] but it is not clear if they are causally related. It was suggested that after artificially induced plasma volume expansion, hypervolaemia may merely be a supportive adaptation to enable lowered heart rate response, but may not improve thermoregulatory function or tolerance in the heat.[50,51] However, it is widely accepted that exercise-induced hypervolaemia, mediated by plasma volume expansion, has the beneficial effect of enhancing cardiovascular and thermoregulatory responses to exercise.[52] Similarly, it is widely assumed that the cardiovascular and thermoregulatory improvement noted early during heat acclimation[26,53,54] is mediated by plasma volume expansion.[2,6,28,49] ª 2011 Adis Data Information BV. All rights reserved.
761
Artificial plasma volume expansion confers some adaptations in temperate[52] and hot conditions[51] but does not necessarily enhance exercising stroke volume[55] or work capacity.[12] Irrespective of whether hypervolaemia does[28] or does not[44] decrease with continued acclimation, there is limited information available on its rate of decay within a few days or weeks of ceasing acclimation. However, the transient nature of plasma volume expansion has been observed to decrease to baseline levels within 2 to 3 days after short-term aerobic training.[56] It has been reported by Convertino et al.[57] that after an 8-day heat-exposure period followed by minimal activity for 7 days, plasma volume expansion had returned back to baseline within 1 week after the stimulus was removed. Hence, after short-term (5 days) heat acclimation, one would expect the plasma volume expansion to return to baseline levels within the first week of decay. Therefore, if hypervolaemia were largely responsible for the increased cardiovascular stability, indicated by decreased exercise heart rate, then this attenuation would be expected to track the induction and decay of hypervolaemia during and after acclimation. If not, it may suggest that more centrally controlled mechanisms have a major role in the improved cardiovascular stability with heat acclimation.[29,30,58-60] An examination of the relationship between the induction and decay of exercise heart rate and hypervolaemia, during and after acclimation requires further investigation in the literature. Moderate exercise in a hot environment reduces stroke volume, and increases heart rate accordingly, to cause little net effect on cardiac output until severe heat stress is imposed.[61] It has been hypothesized that the reduced stroke volume may be due, predominantly, to increased skin blood flow, although the mechanisms mediating this reduction are not fully understood.[62] However, it may be that increased cutaneous capacitance plays a greater role than skin blood flow in reducing blood volume. Hales[63] suggested that a conflict develops for the cardiovascular system between the maintenance of central blood pressure and the thermoregulatory requirement for increased skin blood flow.[63] However, in contrast Sports Med 2011; 41 (9)
762
to Hales’ findings, Nielsen et al.[25] concluded that high core temperature and not circulatory failure was the critical factor for exhaustion during acclimation-exercise bouts in hot and dry conditions.[26] Recent work has provided evidence for a critical core temperature that limited moderate exercise in the heat.[61,64] Central fatigue, or reduced central activation, is proposed as the mechanism underlying the critical core temperature in heat-stress conditions.[65,66] The notion of central fatigue is supported by Horowitz,[59] who suggested that the fundamental adaptive feature of the acclimation mechanisms is increased contractile efficiency in the heart. This is based on investigations of the isolated heart in the rat model.[59] Therefore, current research strongly supports the notion of central, rather than peripheral, mechanisms limiting exercise performance in the heat. The fluid conserving hormones, aldosterone and arginine vasopressin (AVP), play a major role in the sustained control of plasma volume expansion. Aldosterone is produced by the adrenal cortex to facilitate reabsorption of sodium ions (Na+) in the kidney and sweat glands. The release of aldosterone is mediated by plasma potassium (K+) concentration, activity of the renin-angiotensin system and, to a lesser extent, plasma Na+ concentration. AVP is also known as an anti-diuretic hormone and is secreted from the posterior pituitary gland to retain water and thus solute in the kidney. It also maintains blood pressure by causing vasoconstriction. Increased Na+ concentration is considered the major stimulus for AVP secretion during exercise.[67,68] Under heatstress conditions and hypohydration, the hormones aldosterone and AVP respond in a regulatory process to maintain fluid balance. It has been observed that during prolonged exercise in the heat, dehydrated, compared with subjects with progressive hydration, had a marked increase in the response of aldosterone, AVP and plasma cortisol.[69] It was determined that progressive rehydration with water or isotonic solution, prevented the increase in fluid conserving hormones because an absence of osmotic and volaemic stimuli had been created. It has also been established that fluid regulatory adaptations are an important ª 2011 Adis Data Information BV. All rights reserved.
Garrett et al.
factor in heat acclimation,[4,16,70] and it may be that by stressing fluid homeostasis thermal adaptation will be optimized.[40] Therefore, this review puts forward the premise that following a regimen of permissive dehydration during STHA, acclimation by increased fluid-electrolyte retention is facilitated by plasma volume expansion and a cardiovascular response to heat stress. Cardiovascular function during exercise may be mediated by a combination of central and peripheral components. It is generally assumed that improved cardiovascular stability with heat acclimation reflects an increased blood volume,[6,49] possibly by increased ventricular contractility.[29,30,59] However, in the upright position, ~75% of the blood is below the heart at any one time[71] and it may be possible that there is an improved maintenance of central blood volume by way of a more active, improved peripheral vascular function. For example, the calibre of arterioles and pre-capillary sphincters have been postulated as being regulated by local control factors including metabolic, myogenic and endothelial components.[72,73] Using a validated, non-invasive technique (high-frequency ultrasound), the brachial artery can be used for the assessment of peripheral vascular (endothelial) function.[74-76] It has previously been used to determine the effects of exercise training on markers of endothelial function[77] in young men[50,77,78] but there is limited evidence that it has been used to determine the effects of heat exposure on endothelial function in healthy participants. 5. Effectiveness of Dehydration on Adaptation to the Heat The use of permissive dehydration during acclimation contradicts the existing fluid replenishment guidelines for heat acclimation that recommends the maintenance of good hydration status during exposure to heat stress conditions.[7,9,24,79-81] However, the reality for many people undergoing acclimation bouts is that some level of dehydration is normal, if not frequently, inevitable.[82-85] Therefore, the hypothesis that heat acclimation will confer substantial improvements in physiological strain and exercise tolerance Sports Med 2011; 41 (9)
Short-Term Heat Acclimation
for exercise in the heat, and fluid regulatory strain provides a thermally-independent stimulus for such adaptations, requires further attention in the literature. It is well recognized that humans do not voluntarily replace the water that is lost during prolonged exercise in the heat[11,83,86] and, in some individuals with a high sweating rate, total-body water loss may reach 8% of body mass.[87] Furthermore, the influence of severe passive heat stress and hypohydration on exercise has received increasing attention in the literature.[82,88-93] Heatand exercise-induced hypohydration manifests itself as hyperosmotic hypovolaemia[94] that impairs cutaneous blood flow and sweat rate and raises heart rate, core temperature, glycogenolysis, perceived exertion and permeability of tight membranes.[8,95-100] The consequence is impaired exercise tolerance and performance in temperate condition, which has previously been debated.[101] Dehydration during prolonged exercise in the heat increases the response of the fluid- and stressregulatory hormones aldosterone, AVP and cortisol[102] as well as thirst,[103,104] such that hydration is regulated at a lower level until after exercise and/or heat stress.[69,105] The osmolality and volume effects of hypohydration incur fluid-regulatory responses that could partially mediate the hypervolaemia and improved fluid-regulatory efficiency that is observed with training- and heat-acclimation adaptations, and which helps attenuate cardiovascular strain in exercise.[106] Therefore, we believe that in contrast to current recommendations of avoiding dehydration during exercise and heat acclimation, permissive dehydration independently facilitates acclimation by increased fluid-electrolyte retention, plasma volume expansion and cardiovascular response to heat stress. 6. Heat-Shock Proteins and Acclimation Long-term systemic protective adaptation occurs after repeated exposure to heat stress and is referred to as acclimation using artificial conditions (e.g. heat chamber) and acclimatization in a natural environmental climate.[33] This allows an organism to perform increased work in the heat because of improvements in heat dissipation and ª 2011 Adis Data Information BV. All rights reserved.
763
lower core temperature, and is one of two main thermo-protective pathways to combat the effects of heat stress. The other major pathway is the heatshock protein response (HSPR), that is mediated primarily by heat-shock proteins (HSPs).[59] HSP expression is associated with resistance to stress and may be central to the understanding at a cellular level of thermotolerance after acclimatization.[107] The most inducible HSP is the 70 kDa family (HSP70).[108] Proteins in the HSP70 group share common protein sequences but are synthesized by different stimuli. For example, the highly inducible HSP72 kDa protein responds to multiple stressors including increased temperature[109] exercise[110] and reduced glucose availability,[111] whereas the HSP73 kDa protein is constitutively produced.[112] Therefore, HSP70 is a stress protein family whose inducible form is HSP72[111] and has a major role in preventing thermal injury by enhancing the signalling pathways of the cytoprotective mechanisms[59,113,114] and the immune system.[115] HSP accumulation was originally considered a cellular marker of stress that occurred in humans after exercise in the heat,[108,116] but more recently it has been demonstrated that HSPR has a major role in protecting cells from other forms of stress in preparing them to survive environmental challenges.[112] HSPs respond in cells exposed to hyperthermia, hypoxia, starvation and oxygen stress.[117] Furthermore, mechanisms for HSP70 response include hyperthermia[59,107] and exercise.[108,116,118] However, hyperthermia occurs during exercise but exercise-induced HSP70 can be independent of changes in body temperature.[119] Skidmore et al.[119] investigated the role of internal temperature on HSP70 induction in the skeletal muscle of rats during exercise and heat stress. They observed that the increased HSP70 levels were independent of core temperature and suggested that factors other than heat stress may contribute to HSP70 expression during exercise. For example, exercise generates many metabolic changes independent of hyperthermia that have been demonstrated to induce HSP70, such as glycogen depletion.[120] Furthermore, Kee-Bum et al.[112] demonstrated that metabolic stress may be partly responsible for the increased exercise-associated Sports Med 2011; 41 (9)
Garrett et al.
764
skeletal and cardiac tissue HSP70 levels in rats exercising in hot temperatures. They reported similar reductions in body mass during the course of the experimental period, due to fluid loss by increased evaporative heat loss. This indicates that the HSP70 expression in skeletal and cardiac tissue may not be entirely dependent on rectal temperature, and hypohydration may be partly responsible. Kresfelder et al.[20] investigated the possibility of using serum HSP70 as a marker of acclimation in humans (table I). They reported that physiological parameters denoting acclimation were correlated to levels of serum HSP70, thus providing a useful marker of heat acclimation. However, more information needs to be made available in the literature on whether HSPR will be impacted by the relatively brief submaximal stress nature of heat acclimation. 7. Time Course of Acclimation Decay Research on heat acclimation has focused almost exclusively on its induction, such that there is less information available on the time course for the decay of adaptation to the heat, often with conflicting results between researchers.[7,14,28,48,70,121-128] Of the limited information available, the characteristic adaptations to the heat have been shown to return to normal values 3 weeks following cessation of acclimation exposure.[7,121-123] Therefore, physiological adaptations after heat acclimation are relatively short term and may vanish only a few days or weeks after removal from heat exposure.[7,35] It is well established that the characteristic adaptations to the heat such as decreased heart rate, attenuated core temperature, a more dilute sweat and the earlier onset of sweating, have been shown to return to normal values after 3-weeks post-heat exposure.[7,121-123] It has been suggested that the rate of heat-acclimation decay is such that for every 2 days spent without working in the heat, 1 day of acclimatization is lost.[129] Furthermore, in a more recent review on heat adaptation,[35] it was recommended that one additional heat exposure be used for each 5 days away from significant exposure. However, it has been reported[128] that 16 male participants, were acclimated to dry ª 2011 Adis Data Information BV. All rights reserved.
heat (mean – SD 46.1 – 0.1C; 17.9 – 0.1% relative humidity), using a long-term acclimation protocol of 10 consecutive days. Using the hyperthermia controlled technique (38.5C rectal temperature) during acclimation, the participants were divided into two groups and re-exposed to work in the heat at 12 (n = 8) and 26 days (n = 8). They reported that heat acclimation was reattained after 2 and 4 days, respectively. It was concluded that after acclimation, the time spent in cooler climates maybe up to 1 month before extensive reacclimation is required.[128] It is widely acknowledged that the first heat adaptations to decay are the first to occur, which are from cardiovascular origin.[14,48,122,124,127] For example, Saat et al.[124] reported a greater percentage loss of acclimation for heart rate than rectal temperature on the fourth day after ceasing acclimation. Similarly, in the research groups of Pandolf et al.[48] and Williams et al.[122] it was reported that the percentage loss of acclimation is greater for heart rate and mean sweat rate than rectal temperature during the decay of acclimation from 6 to 21 days,[48,122] indicating that the change in rectal temperature may not be dependent on heart rate and mean sweat rate per se. It further suggests that physiological changes that take longer to develop (sudomotor habituation and improved sweating efficiency[35]) will have a slower rate of decay. Therefore, due to incomplete knowledge in the literature, the decay of shortterm acclimation requires further investigation, as it may add to our understanding of how the attenuation of cardiovascular strain associated with heat acclimation is mediated. Of the limited research available on the decay of acclimation, there are many confounding methodological issues such as the heat exposure type, training status, number and duration of acclimations. For example, in a review by Pandolf,[127] it stated that the retention of acclimation remained longer after exposure for dry versus humid heat and was associated with high aerobic fitness. The interval between heat-stress tests is particularly important because, if it is too brief, it may actually constitute adaptation stimuli.[130] It is suggested that the interval between heatstress tests is an experimental design problem for Sports Med 2011; 41 (9)
Short-Term Heat Acclimation
the research groups of Saat et al.[124] and Pandolf et al.,[48] who had performed heat-stress tests during the decay of acclimation at 4 and 3 days, respectively. The work of Barnett and Maughan[130] reported that in order to ensure there is no heatacclimation effect, repeated exercise exposures should be performed at 1-week intervals. Therefore, a lack of consistency in experimental standardization has resulted in many issues with the decay of acclimation still awaiting further definitive research.[14,35,127] 8. Highly Trained Athletes and Heat Acclimation It has been established that the efficacy of a heat acclimation may be dependent on the fitness status of the individual.[24] In trained athletes it has been reported that the higher the background adaptation, the lower the adaptation response.[40] The effectiveness of STHA protocols has tended to use untrained[17,22] or moderately trained[14,23,131,132] participants (table I). Therefore, the physiological and performance adaptations associated with heat acclimation for highly trained athletes require further investigation. Endurance-trained athletes behave physiologically as if they were already heat acclimatized.[35] Interestingly, a recent study by Lorenzo et al.,[21] supported an earlier study by Schoon et al.[133] that demonstrated heat acclimation improves endurance performance in hot and temperate-cool conditions. Heat-induced hypervolaemia was achieved in these studies, which was closely related to the performance improvement. This supports the notion of using heat acclimation as a training method to enhance physical conditioning programmes; however, this requires further investigation. In summary, highly trained athletes may have less adaptive potential in comparison with untrained or moderately trained participants. Often it is highly trained athletes that are in need of heat acclimation and they have to rely on vast literature from lesser trained people who have more room to improve. However, of the limited research available on highly trained athletes,[18] it has been demonstrated that they may have similar gains to those seen in the lesser trained using STHA (table II). ª 2011 Adis Data Information BV. All rights reserved.
765
Therefore, due to insufficient and valid knowledge, an important aim of this review was to examine the extent to which highly trained athletes adapt to STHA. STHA may be the preferred regimen for highly trained athletes, as it potentially provides less disruption of quality training near to competition and is less expensive than long-term protocols. This is based on two principles of adaptation that (i) controlled hyperthermia maintains strain across adaptation;[35] and (ii) previous findings from work with moderately trained athletes,[14] tentatively support the use of a permissive dehydration regimen in the adaptation. However, the effectiveness of STHA, using the controlled hyperthermia technique, has tended to use untrained[17,22] or moderately trained[14,23,131,132] participants. Pandolf et al.[48] reported that well trained participants with a high level of aerobic fitness (>65 mL/kg/min), adapt more readily to heat exposure and could be acclimated after 4 days, achieving greater cardiac stability and lower body temperature during a subsequent heat stress.[48] In contrast, slower adaptation is seen in individuals with low basal fitness.[134] However, individuals with lowto moderate-aerobic conditioning (<55 mL/kg/min) may have greater potential benefit from heat acclimation than individuals with a high level of aerobic fitness. It has been demonstrated that individuals with a low as opposed to a high max. imal oxygen consumption (VO2max) experienced larger decreases in heart rate and rectal temperature after acclimation.[135,136] . Improvements in aerobic fitness (VO2max) and the physiological adaptation that occur, such as increased heat loss capacity and decreased rectal temperature, have been associated with increased tolerance to exercise in the heat.[37] For example, it was reported that endurance-trained males who performed regular exercise under sunny desert conditions were better adapted and acclimatized than their lesser trained counterparts. This was indicated by greater cardiovascular stability against a fixed work load.[137] Caderette et al.[135] demonstrated that during heat stress, decreased cardiovascular and thermoregulatory strain occurred with increasing fitness before acclimation. Sports Med 2011; 41 (9)
766
Armstrong et al.[47] investigated well trained distance runners (mean – SD best marathon time: 2:38:00 – 0.05; n = 4) during spring and summer training (30.3C) in the northeastern US. They observed equivalent heat tolerance both before and after summer training indicated by no significant difference in heart rate, rectal temperature, sweat Na+ and K+ concentration, plasma Na+ and K+, or change in plasma volume during exercise. They concluded that highly trained endurance athletes do not need special thermal preparation to facilitate heat tolerance during seasonal climatic changes.[47] However, the small number of subjects (n = 4) indicates this work should be viewed with caution. The work of Cheung et al.[24] used untrained (<50 mL/kg/min; n = 8) and trained (>55 mL/kg/min; n = 7) participants who were stress tested before and after heat acclimation while they were euhydrated and hypohydrated (~2.5% body mass) by exercise and fluid restriction on the day preceding the trials. This protocol involved 1 hour of treadmill exercise at 40C, 30% RH, for 2 weeks of daily heat acclimation wearing nuclear, biological and chemical protective clothing. Heat acclimation increased sweat rate and decreased rectal temperature and mean skin temperature in trained participants but had no effect on exercise tolerance time. Untrained participants increased sweat rate but did not alter heart rate and rectal temperature or exercise tolerance time. For trained and untrained participants, regardless of the acclimation state, hypohydration increased heart rate and rectal temperature with a decreased exercise tolerance time. The rate of rise in mean skin temperature was less, whereas the change in rectal temperature and exercise-tolerance time were greater in trained than in untrained participants. Therefore, it was concluded that long-term aerobic fitness resulted in a significant improvement in exercise heat tolerance, regardless of hydration or acclimation status.[24] STHA (<7 days) is an ideal protocol for exercise in the heat as many of the important adaptations to heat stress are cardiovascular, and these occur relatively rapidly. However, increased sweating power is an adaptive response of commonly employed long-term (12–14 day) heat-acclimation ª 2011 Adis Data Information BV. All rights reserved.
Garrett et al.
regimens.[33] Examples of short- (<7 days), medium(8–14 days) and long-term (>15 days) acclimation regimens are shown earlier in section 1 (table II). From a practical perspective, completing a STHA regimen for highly trained athletes is easier when sustaining quality training and tapering performance in the weeks before competition. Financial considerations may also dictate a preference for STHA in comparison with the potential extra expense of long-term regimens. Furthermore, the controlled hyperthermia technique may offer a more complete adaptation[35] and is referred to as the isothermal strain model.[40] Of the limited information available, STHA using the controlled hyperthermia technique, has been employed as a method to acclimate untrained soldiers in the British military.[17,42] Turk and Worsley[17] exposed a large group (n = 51) to a hot environment for 5 days (36C WBGT), for 2 hours per day at rectal temperature of 38.8C, resulting in mean 80 – SD5% of participants reaching a satisfactory level of acclimatization that was indicated by greater cardiovascular stability. Furthermore, reduced thermal strain and increased work capacity after mediumterm (10-day) heat acclimation using the isothermal strain model with untrained participants have been reported.[22] Participants (n = 10) maintained an aural temperature between 38 and 38.5C for 70 minutes on each day, by water immersion (40C) and cycle exercise. Using a cycle ergometer, they reported an end-exercise decrease in heart rate (2.3%), rectal temperature (0.5%) and increased work capacity (11.6%). Patterson et al.[16] used STHA (7 days), with moderately trained participants (n = 12), in hot and humid conditions (40C, 60% RH), undertaken using controlled hyperthermia in 90-minute daily bouts. They reported a decrease in resting heart rate (6.8%), rectal temperature (0.5%), plasma volume expansion (9.8%) and increased work capacity (11.6%).[16] In summary, the effectiveness of STHA using the controlled hyperthermia technique has been demonstrated with the untrained[17,22] and moderately trained[14,23,131,132] participants. Therefore, despite highly trained participants having a lesser adaptive potential, it may be expected that it would be as effective with this population group as well.[18] However, due to the lack of information currently available the Sports Med 2011; 41 (9)
Short-Term Heat Acclimation
effectiveness of STHA (5 days), using highly trained participants requires further research. 9. Conclusion In summary, we support the premises that STHA and the strain experienced by the cardiovascular system, may have a more important role in the stimulus of the adaptive responses to exercise in the heat than is generally accepted. Furthermore, fluid regulatory strain may have a thermally independent role in heat adaptation but more work is required in this area. Thermotolerance (HSP70) maybe increased by heat acclimation in a moderately to highly trained population. HSPR is upregulated across heat acclimation indicating increased thermotolerance and a protective adaptive change. Therefore, HSP70 response may be a useful marker of heat acclimation. Of the limited research available, highly trained athletes may have similar physiological gains to the lesser trained using STHA. However, due to the lack of available knowledge, the effectiveness of STHA for highly trained athletes requires further investigation as most previous work has used untrained or moderately trained populations. This review supports the notion that because research on heat acclimation has primarily focused on its induction, the retention (or decay) requires further attention. This may provide additional information on the adaptive response to exercise in the heat. Acknowledgements This review was supported by grants from the Australian Defence Science Technology Organisation, Melbourne, VIC, Australia, and the School of Physical Education, the University of Otago, Dunedin, New Zealand. The authors thank James Cotter for his contribution to the concepts discussed in this review. The authors have no conflicts of interest that are directly relevant to the content of this review.
References 1. Sawka MN, Young AJ, Caderette BS, et al. Influence of heat stress and acclimation on maximal aerobic power. Eur J Appl Physiol 1985; 53: 294-8 2. Senay LC, Mitchell D, Wyndam CH. Acclimatization in a hot, humid environment: body fluid adjustments. J Appl Physiol 1976; 40: 786-96
ª 2011 Adis Data Information BV. All rights reserved.
767
3. Strydom NB, Williams CG. Effect of physical conditioning on the state of heat acclimation on Banto labourers. J Appl Physiol 1969; 27: 262-5 4. Wyndham CH, Benade AJA, Williams CG, et al. Changes in central circulation and body fluid spaces during acclimatization to the heat. J Appl Physiol 1968; 25: 586-93 5. Armstrong LE, Dziados JE. Effects of heat exposure on the exercising adult. In: Bernhardt DB, editor. Sports physical therapy. New York: Churchill Livingstone; 1986 6. Harrison MH. Effects of thermal stress and exercise on blood volume in humans. Physiol Rev 1985; 65: 149-209 7. Armstrong LE, Maresh CM. The induction and decay of heat acclimatization in trained athletes. Sports Med 1991; 12 (5): 302-12 8. Mack GW, Nadel ER. Body fluid balance during heat stress in humans. In: Fregly MJ, Blatteis CM, editors. Handbook of physiology. Bethesda (MD): Am Physiol Soc, 1996; 187-4 9. Casa DJ, Armstrong LE, Hillman SK, et al. National Athletic Trainers’ Association: position statement. Fluid replacement for athletes. J Athl Train 2000; 35: 121-224 10. Coyle EF, Montain SJ. Benefits of fluid replacement [abstract]. Med Sci Sports Exerc 1992; 24: S324 11. Armstrong LE, Maresh CM, Gabaree CV, et al. Thermal and circulatory responses during exercise: effects of hypohydration, dehydration, and water intake. J Appl Physiol 1997; 86 (6): 2028-35 12. Sawka MN. Hypohydration and exercise: effects of heat acclimatization, gender and environment. J Appl Physiol 1983; 55: 1147-53 13. Shirreffs SM. Markers of hydration status. J Sports Med Phys Fitness 2000; 40: 80-4 14. Garrett AT, Goossens NG, Rehrer NJ, et al. Induction and decay of short-term heat acclimation. Eur J Appl Physiol 2009; 107 (6): 659-71 15. Cotter JD, Patterson MJ, Taylor NAS. Sweat distribution before and after repeated heat exposure. Eur J Appl Physiol 1997; 76: 181-6 16. Patterson MJ, Stocks JM, Taylor NAS. Sustained and generalized extracellular fluid expansion following heat acclimation. J Physiol 2004; 559: 327-34 17. Turk J, Worsley DE. A technique for the rapid acclimatization to heat for the army: Farnborough: Army Personnel Research Establishment, Ministry of Defence; 1974. Report no.: 12/74 18. Garrett AT, Creasy R, Rehrer NJ, et al. Effectiveness of short-term heat acclimation for highly trained athletes. Eur J Appl Physiol. In press 19. Sunderland C, Morris JG, Nevill ME. A heat acclimation protocol for team sports. Br J Sports Med 2008 May 1; 42 (5): 327-33 20. Kresfelder TL, Claassen N, Cronje´ MJ. Hsp70 Induction and hsp70 Gene polymorphisms as indicators of acclimatization under hyperthermic conditions. J Therm Biol 2006; 31 (5): 406-15 21. Lorenzo S, Halliwill JR, Sawka MN, et al. Heat acclimation improves exercise performance. J Appl Physiol 2010; 109: 1140-7 22. Weller AS, Harrison MH. Influence of heat acclimation on physiological strain during exercise-heat stress in men
Sports Med 2011; 41 (9)
Garrett et al.
768
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35. 36.
37.
38.
39.
wearing clothing of limited water vapour permeability [abstract]. J Physiol 2001; 531: 51P Regan JM, Macfarlane DJ, Taylor NAS. An evaluation of the role of skin temperature during heat adaptation. Acta Physiol Scand 1996; 158: 365-75 Cheung SS, McLellan TM. Heat acclimation, aerobic fitness, and hydration effects on tolerance during uncompensable heat stress. J Appl Physiol 1998; 84 (5): 1731-9 Nielsen B, Strange S, Christensen NJ, et al. Acute and adaptive responses in humans to exercise in a warm, humid environment. Pfluger’s Archiv 1997; 434: 49-56 Nielsen B, Hales JRS, Strange S, et al. Human circulatory and thermoregulatory adaptations with heat acclimation and exercise in a hot, dry environment. J Physiol 1993; 460: 467-85 Houmard JA, Costill DL, Davis JA, et al. The influence of exercise intensity on heat acclimation in trained subjects. Med Sci Sports Exerc 1990; 22 (1): 615-20 Shapiro Y, Hubbard RW, Kimbrough CM, et al. Physiological and hematologic responses to summer and winter dry-heat acclimation. J Appl Physiol 1981; 50: 792-8 Levi E, Vivi A, Hasis Y, et al. Chronic heat improves mechanical and metabolic response of trained rat heart on ischemia and reperfusion. Am J Physiol 1997: H2085-94 Levi E, Vivi A, Hasis Y, et al. Heat acclimation improves cardiac mechanics and metabolic performance during ischemia and reperfusion. J Appl Physiol 1993; 75 (2): 833-9 Fox RH, Goldsmith R, Kidd DJ, et al. Blood flow and other thermoregulatory changes with acclimatization to heat. J Physiol 1963; 166: 548-62 Fox RH, Goldsmith R, Hampton IFG, et al. The nature of the increase in sweating capacity produced by heat acclimatization. J Physiol 1964; 171: 368-76 Sawka MN, Wenger CB, Pandolf KB. Thermoregulatory responses to acute exercise-heat stress and heat acclimation. In: Fregly MJ, Blatteis CM, editors. Handbook of physiology: New York (NY): Am Physiol Soc, Oxford University Press; 1996: 157-85 McLellan TM, Jacobs I, Bain JB. Influence of temperature and metabolic rate on work performance with Canadian Forces NBC clothing. Aviat Space Environ Med 1993; 64: 587-94 Taylor NAS. Principles and practices of heat adaptation. J Human-Environ Sys 2000; 4 (1): 11-22 Greenleaf JE, Greenleaf CJ. Human acclimation and acclimatization to heat: compendium of research. Moffett Field (CA): Biotechnology Division, National Aeronautics and Space Administration, Ames Research Center, 1970 Armstrong LE, Pandolf KB. Physical training, cardiorespiratory physical fitness and exercise-heat tolerance. In: Pandolf KB, Sawka MN, Gonzalez RR, editors. Human performance physiology and environmental medicine at terrestrial extremes. Indianapolis (IN): Benchmark Press, Inc. 1988: 199-266 Armstrong LE, Hubbard RW, DeLuca JP, et al. Self-paced heat acclimation procedures. Natick (MA): U.S. Army Research Institute of Environmental Medicine, 1986 Hessemer V, Zeh A, Bruck A. Effects of passive heat adaptation and moderate sweatless conditioning on re-
ª 2011 Adis Data Information BV. All rights reserved.
40.
41. 42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52. 53.
54.
55.
56.
57.
58.
sponses to cold and heat. Eur J Appl Physiol 1986; 55: 281-9 Taylor NAS, Cotter JD. Heat adaptation: guidelines for the optimization of human performance. Int SportMed J 2006; 7 (2): 1-37 Fox RH, Goldsmith R, Kidd DJ, et al. Acclimatization of the sweating mechanism in man. J Physiol 1961; 157: 56-7 Turk J. A reliable test of heat acclimatization for the army. Farnborough: Army Personnel Research Establishment, Ministry of Defence, 1974 Havenith G, van Middendorp H. Determination of individual state of acclimatization: IZF Report. Soesterberg: TNO Institute for Perception, 1986 Patterson MJ. The effect of heat acclimation on body-fluid regulation [PhD]. Wollongong (NSW): University of Wollongong, 1999 Kampmann B, Brode P, Schutte M, et al. Lowering of resting core temperature during acclimation is influenced by exercise stimulus. Eur J Appl Physiol 2008; 104: 321-7 Shido O, Sugimoto N, Tanabe M, et al. Core temperature and sweating onset in humans acclimated to heat at a fixed daily time. Am J Physiol 1999; R1095-R101 Armstrong LE, Hubbard RW, DeLuca JP, et al. Heat acclimation during summer running in the northeastern United States. Med Sci Sports Exerc 1987; 19 (2): 131-6 Pandolf KB, Burse RL, Goldman RF. Role of physical fitness in heat acclimatization, decay and reintroduction. Ergonomics 1977; 20: 399-408 Sawka MN, Convertino VA, Eichner RE, et al. Blood volume: importance and adaptations to exercise training, environmental stresses, and trauma/sickness. Med Sci Sports Exerc 2000; 32 (2): 332-48 O’Sullivan SE. The effects of exercise training on markers of endothelial function in young healthy men. Int J Sport Med 2003; 24: 404-9 Sawka MN, Hubbard RW, Francesconi RP, et al. Effects of acute plasma volume expansion on altering exerciseheat performance. Eur J Appl Physiol 1983; 51: 303-12 Fellman N. Hormonal and plasma volume alterations following endurance exercise. Sports Med 1992; 13 (1): 37-49 Armstrong LE. Keeping your cool in Barcelona: the effect of heat, humidity and dehydration on athletic performance, strength and endurance. Colorado Springs (CO): U.S. Olympic Committee, 1992 Hargreaves M, Febbraio M. Limits to exercise performance in the heat. Int J Sports Med 1998; 19 (Suppl. 2): S115-6 Montain SJ, Coyle EF. Fluid ingestion during exercise increases skin blood flow independent of increases in blood volume. J Appl Physiol 1992; 73 (3): 903-10 Lamb DR, Murray R, editors. Perspectives in exercise science and sports medicine: prolonged exercise. Indianapolis (IN): Benchmark Press Inc., 1988 Convertino VA, Greenleaf JE, Bernauer EM. Role of exercise and thermal factors in the mechanism of hypervolemia. J Appl Physiol 1980; 48 (4): 657-64 Horowitz M, Meiri U. Central and peripheral contributions for the control of heart rate during heat acclimation. Pflugers’s Archiv 1993; 422: 386-92
Sports Med 2011; 41 (9)
Short-Term Heat Acclimation
59. Horowitz M. From molecular and cellular to integrative heat defense during exposure to chronic heat. Comp Biochem Physiol 2002; part A (131): 475-83 60. Mirit E, Gross C, Hasin Y, et al. Changes in cardiac mechanics with heat acclimation: adrenergic signaling and SR-Ca regulatory proteins. Am J Physiol Regul Integr Comp Physiol 2000; 279: R77-85 61. Gonzalez-Alonso J, Teller C, Andersen SL, et al. Influence of body temperature on the development of fatigue during prolonged exercise in the heat. J Appl Physiol 1999b; 86 (3): 1032-9 62. Gonzalez-Alonso J, Mora-Rodriguez R, Coyle EF. Supine exercise restores arterial blood pressure and skin blood flow despite dehydration and hyperthermia. Am J Physiol 1999; 277 (2 Pt 2): H576-83 63. Hales J. Hyperthermia and heat illness. Thermoregulation 1997; 813: 534-44 64. Walters TJ, Ryan KL, Tate LM, et al. Exercise in the heat is limited by a critical internal temperature. J Appl Physiol 2000; 89 (2): 799-806 65. Nybo L, Nielsen B. Hyperthermia and central fatigue during prolonged exercise in humans. J Appl Physiol 2001; 91: 1055-60 66. Morrison S, Sleivert GG, Cheung SS. Passive hyperthermia reduces voluntary activation and isometric force production. Eur J Appl Physiol 2004; 91 (5-6): 729-36 67. Convertino VA, Keil LC, Bernauer EM, et al. Plasma volume, osmolality, vasopressin, and renin activity during graded exercise in man. J Appl Physiol: Respirat Environ Exercise Physiol 1981; 50 (1): 123-8 68. Hew-Butler T. Arginine vasopressin, fluid balance and exercise: is exercise-associated hyponatraemia a disorder of arginine vasopressin secretion? Sports Med 2010; 40 (6): 459-79 69. Brandenberger G, Candas V, Follenius M, et al. Vascular fluid shifts and endocrine responses to exercise in the heat. Eur J Appl Physiol 1986; 55: 123-9 70. Bass DE, Kleeman CR, Quinn M, et al. Mechanisms to acclimatization to heat in man. Medicine 1955; 34: 323-80 71. Rowell LB. Cardiovascular adjustments to thermal stress. In: Shepard JP, Abboud FM, editors. Handbook of physiology: the cardiovascular system. Bethesda (MD): Am Physiol Soc, 1983: 967-1023 72. Laughlin MH. Cardiovascular responses to exercise. Advan Physiol Edu 1999; 277 (6): S244-59 73. Laughlin MH, Korthuis RJ, Duncker DJ, et al. Control of bloodflow to cardiac and skeletal muscle during exercise. In: Fregly MJ, Blatteis CM, editors. Handbook of physiology. Bethesda (MD): Am Physiol Soc; 1996: 705-69 74. Sorensen KE, Celermajer D, Spiegelhalter DJ. Noninvasive measurement of human endothelium dependent arterial responses: accuracy and reproducibility. Br Heart J 1995; 74: 247-53 75. Celermajer DS, Sorensen KE, Gooch VM, et al. Noninvasive detection of endothelial function in children and adults at risk of atherosclerosis. Lancet 1992; 340: 1111-5 76. Corretti MC, Anderson TJ, Benjamin EJ, et al. Guidelines for the ultrasound assessment of endothelial-dependent flow-mediated vasodilation of the brachial artery. J Am Coll Cardiol 2002; 39: 257-65
ª 2011 Adis Data Information BV. All rights reserved.
769
77. Clarkson P, Montgomery HE, Mullan MJ, et al. Exercise training enhances endothelial function in young men. J Am Coll Cardiol 1999; 33 (5): 1379-85 78. Rabobowchuk M, McGowan CL, de Groot PC, et al. Endothelial function of young healthy males following whole body resistance training. J Appl Physiol 2005; 98: 2185-90 79. Wendt D, Van Loon LJ, Lichtenbelt WD. Thermoregulation during exercise in the heat: strategies for maintaining health and performance. Sports Med 2007; 37 (8): 669-82 80. Convertino VA, Armstrong LE, Coyle EF, et al. ACSM: position stand on exercise and fluid replacement. Med Sci Sports Exerc 1996; 28: i-vii 81. Kamijo Y, Nose H. Heat illness during working and preventive considerations from body fluid homeostasis. Ind Health 2006; 44 (3): 345-58 82. Noakes TD. Drinking guidelines for exercise: what is the evidence that athletes should either drink ‘‘as much as tolerable’’ or ‘‘to replace all the weight lost during exercise’’ or ‘‘ad libitum’’. J Sports Sci 2007; 25 (7): 781-96 83. Greenleaf JE. Problem: thirst, drinking behaviour, and involuntary dehydration. Med Sci Sports Exerc 1992; 24: 645-6 84. Noakes TD, Adams BA, Myburgh KH, et al. The danger of inadequate water intake during prolonged exercise. Eur J Appl Physiol 1988; 57: 210-9 85. Noakes TD. Exercise in the heat: old ideas, new dogmas. Int SportMed J 2006; 7 (1): 58-74 86. Szlyk PC, Sils I, Francesconi RN, et al. Effects on water temperature and flavouring on voluntary dehydration in man. Physiol Behav 1989; 45: 639-47 87. Armstrong LE, Hubbard RW, Jones BH, et al. Preparing Alberto Salazar for the heat of the Olympic Marathon. Phys Sports Med 1986; 14 (3): 73-81 88. Edwards AM, Noakes TD. Dehydration: cause of fatigue or sign of pacing in elite soccer? Sports Med 2009; 39 (1): 1-13 89. Fan J-L, Cotter JD, Lucas RAI, et al. Human cardiorespiratory and cerebrovascular function during severe passive hyperthermia: effects of mild hypohydration. J Appl Physiol 2008 August 1; 105 (2): 433-45 90. Judelson DA, Maresh CM, Yamamoto LM, et al. Effect of hydration state on resistance exercise-induced endocrine markers of anabolism, catabolism, and metabolism. J Appl Physiol 2008 Sep 1; 105 (3): 816-24 91. Osterberg KL, Pallardy SE, Johnson RJ, et al. Carbohydrate exerts a mild influence on fluid retention following exercise-induced dehydration. J Appl Physiol 2010 Feb 1; 108 (2): 245-50 92. Godek SF, Bartolozzi AR, Peduzzi C, et al. Fluid consumption and sweating in National Football League and collegiate football players with different access to fluids during practice. J Athl Train 2010; 45 (2): 128-35 93. Edwards AM, Mann ME, Marfell-Jones MJ, et al. Influence of moderate dehydration on soccer performance: physiological responses to 45 min of outdoor match-play and the immediate subsequent performance of sportspecific and mental concentration tests. Br J Sports Med 2007; 41: 385-91 94. Nadel ER, Fortney SM, Wenger CB. Effect of hydration state on circulatory and thermal regulations. J Appl Physiol 1980; 49: 715-21
Sports Med 2011; 41 (9)
770
95. Gonzalez-Alonso J, Calbert JAL, Nielsen B. Muscle blood flow is reduced with dehydration during prolonged exercise in humans. J Physiol 1998; 513.3: 895-905 96. Gonzalez-Alonso J, Calbert JAL, Nielsen B. Metabolic and thermodynamic responses to dehydration-induced reductions in muscle blood flow in exercising humans. J Physiol 1999; 520.2: 577-89 97. Gonzalez-Alonso J, Calbert JAL. Reductions in systematic and skeletal muscle blood flow and oxygen delivery limit maximal aerobic capacity in humans. Circ 2003; 107: 824-30 98. Watson G, Judelson DA, Armstrong LE, et al. Influence of diuretic-induced dehydration on competitive sprint and power performance. Med Sci Sports Exerc 2005; 37 (7): 1168-74 99. Maughan RJ. Impact of mild dehydration on wellness and on exercise performance. Eur J Clin Nutr 2003; 57 (Suppl. 2): S19-23 100. Blatteis CM. Thermoregulation in complex situations: combined heat exposure, infectious fever and water deprivation. Int J Biometeorol 2000; 44: 31-43 101. Sawka MN, Noakes TD. Does dehydration impair exercise performance? Med Sci Sports Exerc 2007; 39 (8): 1209-17 102. Kenefick RW, Maresh CM, Armstrong LE, et al. Rehydration with fluid of varying tonicities: effects on fluid regulatory hormones and exercise performance in the heat. J Appl Physiol 2007 May 1; 102 (5): 1899-905 103. Sawka MN, Francesconi RP, Drolet L, et al. Thirst and fluid intake following graded hypohydration levels in humans. Physiol Behav 1987; 40: 229-36 104. Maresh CM, Gabaree-Boulant CL, Armstrong LE, et al. Effect of hydration status on thirst, drinking, and related hormonal responses during low-intensity exercise in the heat. J Appl Physiol 2004 July 1; 97 (1): 39-44 105. Brandenberger G, Candas V, Follenius M, et al. The influence of initial state of hydration on endocrine responses to exercise in the heat. Eur J Appl Physiol 1989; 58: 674-9 106. Hopper MK, Coggan AR, Coyle EF. Exercise stroke volume in relation to plasma-volume expansion. J Appl Physiol 1988; 64 (1): 404-8 107. Feder ME, Hofmann GE. Heat shock proteins, molecular chaperones, and the stress response: evolutionary and ecological physiology. Ann Rev Physiol 1999; 276: R550-R8 108. Moseley PL. Heat shock proteins and heat adaptation of the whole organism. J Appl Physiol 1997; 83 (5): 1413-7 109. Mizzen LA, Welch WJ. Characterization of the thermotolerant cell: effects on protein synthesis activity and the regulation of heat-shock protein 70 expression. J Cell Biol 1988; 106: 1105-16 110. Walsh RC, Koukoulas I, Garnham A, et al. Exercise increases serum HSP72 in humans. Cell Stress Chaperones 2001; 6 (4): 386-93 111. Febbraio M, Mesa JL, Chung J, et al. Glucose ingestion attenuates the exercise-induced increase in circulating heat shock protein 72 and heat shock protein 60 in humans. Cell Stress Chaperones 2004; 9 (4): 390-6 112. Kee-Bum K, Mun-Hee K, Dong-Jun L. The effect of exercise in cool, control and hot environments on cardioprotective HSP70 induction. J Physiol Anthropol Appl Human Sci 2004; 6: 225-30
ª 2011 Adis Data Information BV. All rights reserved.
Garrett et al.
113. Kregel KC. Heat shock proteins: modifying factors in physiological stress responses and acquired thermotolerance. J Appl Physiol 2002; 92: 2177-86 114. Gabai VL, Sherman MY. Invited review. Interplay between molecular chaperones and signaling pathways in survival of heat shock. J Appl Physiol 2002; 92: 1743-8 115. Schneider EM, Niess AM, Lorenz I, et al. Inducible HSP70 expression analysis after heat and physical exercise: transcriptional, protein expression, and subcellular localization. Ann NY Acad Sci 2002; 973: 8-12 116. Ryan AJ, Gisolfi CV, Mosely PL. Synthesis of 70-K stress protein by human leukocytes: effect of exercise in the heat. J Appl Physiol 1991; 70: 466-71 117. Gething MJ, Sambrook J. Protein folding in the cell. Nature 1992; 355: 33-45 118. Liu Y, Sabine M, Optiz-Gress C, et al. Human skeletal muscle HSP70 response to training in highly trained rowers. J Appl Physiol 1999; 86 (1): 101-4 119. Skidmore R, Gutierrez JA, Guerriero JR V, et al. HSP70 induction during exercise and heat stress in rats: role of internal temperature. Am J Physiol 1995: R92-R7 120. Febbraio M, Steensberg A, Walsh RC, et al. Reduced glycogen availability is associated with an elevation in HSP72 in contracting human skeletal muscle. J Physiol 2002; 538 (3): 911-7 121. Adams JM, Fox RH, Grimby G, et al. Acclimatization to heat and its rate of decay in man. J Physiol 1960; 152: 26P-7P 122. Williams CG, Wyndham CH, Morrison JF. Rate of loss of acclimation in summer and winter. J Appl Physiol 1967; 22: 21-6 123. Wyndham CH, Jacobs GE. Loss of acclimatization after six days of work in cool conditions on the surface of a mine. J Appl Physiol 1957; 11: 197-9 124. Saat M, Sirisinghe RG, Singh R, et al. Decay of heat acclimation during exercise in cold and exposure to cold environment. Eur J Appl Physiol 2005; 95: 313-20 125. Henschel A, Taylor HL, Keys A. The persistence of heat acclimatization in man. Am J Physiol 1943; 140: 321-5 126. Lind AR, Bass DE. Optimal exposure time for development of acclimatization to the heat. Fed Proc 1963; 22: 704-8 127. Pandolf KB. Time course of heat acclimation and its decay. Int J Sport Med 1998; 19: S157-60 128. Weller A, Linnane D, Jonkman A, et al. Quantification of the decay and re-induction of heat acclimation in dry-heat following 12 and 26 days without exposure to heat stress. Eur J Appl Physiol 2007; 102 (1): 57-66 129. Givoni B, Goldman RF. Predicting rectal temperature response to work, environment, and clothing. J Appl Physiol 1972; 32 (6): 812-22 130. Barnett A, Maughan R. Response of unacclimatized males to repeated bouts of exercise in the heat. Br J Sports Med 1993; 27 (1): 39-44 131. Creasy R. Post-exercise sauna bathing does not improve 2000-metre rowing performance [BPhEd]. Dunedin: University of Otago, 2002 132. Taylor NAS, Patterson MJ, Regan JM. Heat acclimation procedures: preparation for humid heat exposure. Wollongong (NSW): Applied Physiology Research Laboratory, University of Wollongong, 1995
Sports Med 2011; 41 (9)
Short-Term Heat Acclimation
133. Scoon GSM, Hopkins WG, Mayhew S, et al. Effect of postexercise sauna bathing on the endurance performance of competitive male runners. J Sci Med Sport 2007; 10 (4): 259-62 134. Kok R. Heat tolerance of Bantu undergoing acclimatization [abstract]. South Afr Med J 1973; 47: 960 135. Cadarette BS, Sawka MN, Toner MM, et al. Aerobic fitness and the hypohydration responses to exercise-heat stress. Aviat Space Environ Med 1984; 55: 507-12 136. Shvartz E, Shapiro Y, Magazanik A, et al. Heat acclimation, physical fitness, and responses to exercise in
ª 2011 Adis Data Information BV. All rights reserved.
771
temperate and hot environments. J Appl Physiol 1977; 43: 678-83 137. Wells CL, Constable SH, Haan AL. Training and acclimatization: effects on responses to exercise in a desert environment. Aviat Space Environ Med 1980; 51: 105-12
Correspondence: Dr Andrew Garrett, School of Physical Education, University of Otago, 56 Union St West, Dunedin, New Zealand. E-mail:
[email protected]
Sports Med 2011; 41 (9)
Sports Med 2011; 41 (9): 773-792 0112-1642/11/0009-0773/$49.95/0
RESEARCH REVIEW ARTICLE
ª 2011 Adis Data Information BV. All rights reserved.
Effects of Acute Carbohydrate Supplementation on Endurance Performance A Meta-Analysis Tom J. Vandenbogaerde and Will G. Hopkins Sport Performance Research Institute of NZ, AUT University, Auckland, New Zealand
Abstract
Research on the performance effects of acute carbohydrate supplementation is comprehensive. Here we present the first meta-analytic review of this research. Methods: Eighty-eight randomized crossover studies in which carbohydrate supplements were consumed with or without protein before and/or during exercise provided 155 estimates for performance effects in time-to-exhaustion tests or in time trials with or without a preload. For the mixed-model metaanalysis, all effects were converted into percentage changes in mean power in a non-preloaded time trial and weighted using percentage standard errors derived from exact p-values (in a minority of studies) or from estimated errors of measurement (in all other studies). Publication bias was assessed with a plot of t-values for the random-effect solutions versus standard errors. Probabilistic inferences were derived with reference to thresholds for small, moderate and large effects on performance of 0.5, 1.5 and 2.7%. Results: Publication bias was reduced by excluding studies with a standard error >1.25%. In the remaining 73 studies and 122 estimates, the metaanalysed performance effects of carbohydrate supplements ranged from clear large improvements of ~6% to clear moderate impairments of ~2%. The best supplement inferred from the analysis consisted of a ~3–10% carbohydrateplus-protein drink providing ~0.7 g/kg/h glucose polymers, ~0.2 g/kg/h fructose and ~0.2 g/kg/h protein. Substantial increases in the benefit of a supplement were probably small with an additional 9-hour fast and with the inclusion of ~0.2 g/kg/h of protein, probably small to moderate with ingesting the first bolus not at the start of exercise but 1–4 hours before exercise, and possibly small with increasing the frequency of ingestion by three boluses per hour. Substantial reductions in the benefit of a supplement were possibly moderate with a supplement providing >0.25 g/kg/h fructose, and possibly small with an increase in ambient temperature of 10C. The effect in subjects with maximal oxygen consumption higher by 10 mL/kg/min was probably trivial, and the effects of exercise duration were dependent on the concentration of
Vandenbogaerde & Hopkins
774
carbohydrate plus protein in the supplement. The effect of including salt was unexpectedly trivial, and the effect of gender was unclear. Conclusions: Carbohydrate supplements with an appropriate composition and administration regimen can have large benefits on endurance performance. More research and better reporting are required to investigate the moderating effects of gender and salt.
The majority of publications on the performance effects of acute ingestion of carbohydrate and carbohydrate-protein supplements show an ergogenic effect (for reviews see Bosch[1] or Jeukendrup[2]). The underlying mechanism is believed to be a fatigue delay that arises from the maintenance of high rates of carbohydrate oxidation necessary to sustain exercise intensity. In view of the long history of this research, the lack of a published meta-analysis is surprising. Metaanalysts may have been overawed by the diversity in performance tests and the poor reporting of inferential statistics. We have devised ways for converting performance outcomes in time-toexhaustion tests and time trials to a common metric and for dealing with the lack of inferential information. We have derived meta-analysed effects of acute carbohydrate supplementation on performance with moderating effects for differences in subject characteristics (gender and level of athletes), supplement characteristics (type, timing and amount of carbohydrate ingestion, and the inclusion of protein and/or electrolytes), exercise protocols (any preload, and the type and duration of exercise), ambient temperature and fasting time. 1. Methods 1.1 Study Selection
We used Google Scholar to search for crossover investigations of acute effects of carbohydrate and/or carbohydrate-protein supplements on performance published between 1979 and 2009. Reference lists in reviews and research articles were also examined. Table I shows the study-estimate characteristics for studies that were included in the analysis. We considered all studies in which ª 2011 Adis Data Information BV. All rights reserved.
carbohydrate supplements were consumed with or without protein on the day of a physical performance test, including consumption before and/ or during the test. Studies were excluded for the following reasons: published only as conference abstracts; substantial rest intervals in the preload (e.g. rest : work>1 : 4); >5 minutes rest between preload and performance test; performance tests with other than continuous exercise; an inappropriate control (e.g. no fluid consumed in the control); unrealistically high error and performance effect (probably arising from use of a poor ergometer); investigations of supplements with carbohydrates other than glucose, sucrose, fructose or glucose polymers (including maltodextrins); investigations of supplements containing any substances other than carbohydrate, protein and electrolytes, although we included supplementation of high-carbohydrate foods with known approximate content of carbohydrates and protein; a substantial preload preceding a time-to-exhaustion test; glycogen-depleting protocols other than an overnight fast before the start of exercise; and a study of children. A list of these references is available on request. 1.2 Data Extraction 1.2.1 Performance Measures
To perform a meta-analysis, the magnitudes of effects from all relevant studies need to be expressed in a common metric. The most appropriate metric for athletic performance is power output in a time trial, because the effect can then be applied directly to competitive performance.[91] We therefore converted effects on performance in timeto-exhaustion tests and preloaded time trials to effects on power output in non-preloaded time trials. Sports Med 2011; 41 (9)
Foster et al.[3] (1979)
8 M, 8 F
57
Glucose
8 M, 8 F
57
Milkg
25 8.3
Exercise duration (min)d
Power effect (%)e
Power SE (%)f
TTE
100%
6
0.4
0.7
TTE
100%
6
0.4
0.7
8 M, 8 F
57
Glucose
TTE
80%
53
-2.3
0.7
8 M, 8 F
57
Milkg
8.3
TTE
80%
53
-0.1
0.7
Ivy et al.[4] (1979)
7 M, 2 F
60
Polycose
7.5
TT
120 min
120
2.0
1.1
Felig et al.[5] (1982)
10 M
49
Glucose
5.0
TTE
60–65%
164
0.7
0.9
9M
49
Glucose
10
TTE
60–65%
148
1.3
1.0
9 M, 1 F
59
Glucose polymers
11
TTE
74%
134
1.9
0.6 0.8
Coyle et al.[6] (1983) Bjorkman[7] (1984)
25
8M
56
Glucose
7.0
TTE
68%
116
2.7
8M
56
Fructose
7.0
TTE
68%
116
-0.3
1.1
Decombaz et al.[8] (1985)
10 M + F
63
Fructose
TT
15 min
29
-0.1
1.0
Coyle et al.[9] (1986)
7M
70
Glucose polymers
14
TTE
71%
181
3.9
1.0
Gleeson et al.[10] (1986)
6M
47
Glucose
18
TTE
73%
96
1.6
0.4
Flynn et al.[11] (1987)
8M
64
CHO mixh
TT
120 min
120
-1.0
1.8
8M
64
CHO mixh
10
TT
120 min
120
-4.1
1.8
8M
64
CHO mixh
10
TT
120 min
120
0.4
1.8
6M
61
Glucose
21.4
TTE
75%
93
0.0
1.2
6M
61
Fructose
21.4
TTE
75%
93
-0.3
1.2
13 M
45
Glucose polymers
5.0
TT
480 pedal revs
51
-0.2
0.9
13 M
45
CHO mixh
6.0
TT
480 pedal revs
65
-0.9
0.9
13 M
45
CHO mixh
7.0
TT
480 pedal revs
65
1.4
0.9
10 M
60
CHO mixh
11.3
TT
15 min
45
3.9
1.0
Hargreaves et al.[12] (1987) Murray et al.[13] (1987)
Neufer et al.[14] (1987)
20
5.0
10 M
60
Sucrose
11.3
TT
15 min
45
4.1
1.0
Sasaki et al.[15] (1987)
5M
58
Sucrose
18
TTE
80%
40
5.8
1.3
Davis et al.[16] (1988)
7M
63
Glucose
6.0
TT
270 pedal revs
27
-0.2
1.2
7M
63
Glucose
6.0
TT
270 pedal revs
159
4.1
2.1
19 M
64
CHO mixh
6.0
TT
~120 min
129
0.5
0.8
19 M
64
Glucose
2.5
TT
~120 min
129
0.5
0.8
8M
60
CHO mixh
5.0
TT
12 min
68
3.4
1.2
8M
60
CHO mixh
6.0
TT
12 min
68
2.9
1.2
8M
60
CHO mixh
7.5
TT
12 min
68
5.0
1.2
6M
65
Exceed
TTE
70%
169
2.7
0.7
Davis et al.[17] (1988)
Coggan and Coyle[19] (1989)
11
Continued next page
775
Sports Med 2011; 41 (9)
Mitchell et al.[18] (1988)
Carbohydrate and Performance
ª 2011 Adis Data Information BV. All rights reserved.
Table I. Study-estimate characteristics, shown in chronological order by year and alphabetical order within years . Study (y) Sample Supplement CHO + TT or Exercise protocolc VO2maxa size protein (%)b TTE
Study (y) Maughan et al.[20] (1989)
Mitchell et al.[21] (1989)
Murray et al.[22] (1989)
Murray et al.[23] (1989)
Sample size
. VO2maxa
Supplement
CHO + protein (%)b
TT or TTE
Exercise protocolc
776
ª 2011 Adis Data Information BV. All rights reserved.
Table I. Contd Exercise duration (min)d
Power effect (%)e
Power SE (%)f
6M
53
Dioralyte
4.0
TTE
70%
76
3.0
1.2
6M
53
CHO mixh
35.5
TTE
70%
76
0.7
1.2 1.2
6M
53
Glucose syrup
35.5
TTE
70%
76
0.6
6M
53
Fructose syrup
35.5
TTE
70%
76
-2.4
1.2
10 M
63
CHO mixh
6.1
TT
15 min
94
2.1
1.2
10 M
63
CHO mixh
12.2
TT
15 min
94
4.7
1.2
10 M
63
CHO mixh
18.2
TT
15 min
94
2.8
1.2
10 M
63
CHO mixh
12.2
TT
15 min
94
3.6
1.2
7 M, 5 F
48
Sucose
6.0
TT
1200 pedal revs
73
1.2
0.5
7 M, 5 F
48
Sucrose
8.0
TT
1200 pedal revs
73
0.6
0.5
7 M, 5 F
48
Sucrose
TT
1200 pedal revs
73
0.1
0.5
9 M, 3 F
53
Fructose
6.0
TT
600 pedal revs
183
-1.3
0.6
9 M, 3 F
53
Sucrose
6.0
TT
600 pedal revs
183
0.1
0.6
9 M, 3 F
53
Fructose
6.0
TT
600 pedal revs
183
-1.5
0.6
10
Sherman et al.[24] (1989)
8 M, 2 F
59
CHO mixh
TT
~45 min
103
5.4
1.0
Powers et al.[25] (1990)
9 M+F
63
Exceed
7.0
TTE
85%
36
1.0
1.0
9 M+F
63
Exceed
7.0
TTE
85%
40
-0.2
1.0
Williams et al.[26] (1990)
9M
59
CHO mixh
5.0
TT
30 km
129
3.6
1.6
9M
59
CHO mixh
5.0
TT
30 km
129
2.7
1.6
Murray et al.[27] (1991)
8 M, 2 F
51
Glucose
6.0
TT
4.8 km
95
7.3
2.2
8 M, 2 F
51
Glucose
12
TT
4.8 km
95
5.2
2.2
8 M, 2 F
51
Glucose
18
TT
4.8 km
95
7.8
2.2
9M
58
Glucose
20
TT
~45 min
127
2.7
1.1
9M
58
CHO mixh
39.9
TT
~45 min
127
2.6
1.1
Sherman et al.[28] (1991) Thomas et al.[29] (1991)
7M
55
Glucose
17.5
TTE
65–70%
99
1.5
1.1
7M
55
Lentilsg
25
TTE
65–70%
99
3.0
1.1
7M
55
Potatog
18.5
TTE
65–70%
99
-0.3
1.1
8 M, 1 F
63
Exceed
5.1
TTE
70%
201
2.1
0.9
Sports Med 2011; 41 (9)
8 M, 1 F
63
Exceed
2.6
TTE
70%
201
3.7
0.9
8 M, 1 F
63
CHO mixh
7.1
TTE
70%
201
4.8
0.9
Millard-Stafford et al.[31] (1992)
8M
65
CHO mixh
7.0
TT
5 km
184
2.1
0.8
Wilber and Moffatt[32] (1992)
10 M
60
Exceed
7.0
TTE
80%
79
2.7
0.9
Zachwieja et al.[33] (1992)
8M
65
CHO mixh
9.5
TT
15 min
90
3.3
1.5
8M
65
CHO mixh
9.5
TT
15 min
90
2.8
1.5
Continued next page
Vandenbogaerde & Hopkins
Wright et al.[30] (1991)
20
. VO2maxa
Supplement
10 M
60
CHO mixh
10 M
60
CHO mixh
10 M 7M
60 50
CHO mixh Glucose
8.3 10
Study (y)
Sample size
Cole et al.[34] (1993)
Nishibata et al.[35] (1993)
TT or TTE
Exercise protocolc
Exercise duration (min)d
6.0
TT
15 min
111
0.7
1.0
8.3
TT
15 min
111
0.6
1.0
TT TTE
15 min 73%
111 98
0.6 -0.8
1.0 1.1
CHO + protein (%)b
Power effect (%)e
Power SE (%)f
Tsintzas et al.[36] (1993)
4 M, 3 F
64
CHO mixh
5.0
TT
30 km
131
2.3
0.6
Widrick et al.[37] (1993)
8M
58
CHO mixh
9.1
TT
70 km
119
1.8
1.2
Bacharach et al.[38] (1994)
12 M
68
CHO mixh
6.4
TT
500 pedal revs
63
2.2
0.7
12 M
68
CHO mixh
10
TT
500 pedal revs
63
3.0
1.0
Chryssanthopoulos et al.[39] (1994)
5 M, 4 F
63
Glucose
25
TTE
70%
121
1.5
0.7
Anantaraman et al.[40] (1995)
3 M, 2 F
57
Glucose polymers
10
TT
60 min
60
10.5
2.8
3 M, 2 F
57
Glucose polymers
10
TT
60 min
60
7.0
2.8
Ball et al.[41] (1995)
8M
62
Glucose polymers
7.0
TT
30 sec
49
5.6
1.6
Below et al.[42] (1995)
8M
63
CHO mixh
5.9
TT
~10 min
87
0.8
1.2 1.2
8M
63
Maltodextrin
39.5
TT
~10 min
88
1.0
El-Sayed et al.[43] (1995)
9M
61
Glucose
7.5
TT
10 min
38
4.5
1.1
Kang et al.[44] (1995)
7M
62
Gatorade
6.0
TTE
71%
154
2.8
1.1
Tsintzas et al.[45] (1995)
7M
54
CHO mixh
5.5
TT
42.2 km
194
2.1
0.8
7M
54
CHO mixh
6.9
TT
42.2 km
194
0.8
0.8
9M
63
Glucose
5.0
TT
100 km
160
-0.2
2.2
9M
63
Glucose + protein
5.5
TT
100 km
160
2.2
2.2
Madsen et al.[46] (1996) Maughan et al.[47] (1996) McConell et al.[48] (1996) Tsintzas et al.[49] (1996)
12 M
59
Glucose
3.6
TTE
70%
101
2.6
0.9
12 M
59
Glucose
1.6
TTE
70%
101
1.3
0.9
8M
69
CHO mixh
7.0
TT
15 min
92
4.3
1.2
8M
69
Glucose polymers
7.0
TT
15 min
92
1.8
1.2
10 M
57
CHO mixh
3.5
TTE
70%
110
2.0
0.8
10 M
57
Lucozade
4.5
TTE
70%
110
1.6
0.8
Tsintzas et al.[50] (1996)
8M
57
CHO mixh
5.5
TTE
70%
104
3.9
1.1
El-Sayed et al.[51] (1997)
8M
67
Glucose
8.0
TT
60 min
60
3.0
1.2
Jeukendrup et al.[52] (1997)
17 M, 2 F
74
Isostar
7.6
TT
1039 kJ
60
2.2
0.6
12 M
66
Gatorade
6.0
TT
1.6 km
57
1.6
0.7
12 M
66
Powerade
8.0
TT
1.6 km
57
1.9
0.7
Jeukendrup et al.[54] (1998)
7M
74
CHO mixh
10.5
TT
271 kJ
33
0.7
1.5
Kovacs et al.[55] (1998)
15 M
60
CHO mixh
6.8
TT
1083 kJ
62
1.0
0.8
Continued next page
777
Sports Med 2011; 41 (9)
Millard-Stafford et al.[53] (1997)
Carbohydrate and Performance
ª 2011 Adis Data Information BV. All rights reserved.
Table I. Contd
Study (y)
Sample size
. VO2maxa
Supplement
CHO + protein (%)b
TT or TTE
Exercise protocolc
778
ª 2011 Adis Data Information BV. All rights reserved.
Table I. Contd Exercise duration (min)d
Power effect (%)e
Power SE (%)f
Palmer et al.[56] (1998)
11 M, 3 F
67
Energade
TT
20 km
28
0.1
0.9
Sugiura and Kobayashi[57] (1998)
8M
56
Glucose polymers
20
TT
40 sec
105
4.9
0.9
20
1.0
6.8
8M
56
Fructose + protein
TT
40 sec
105
3.6
Jarvis et al.[58] (1999)
0 M, 10 F
61
Exceed
7.0
TT
30 sec
72
1.0
1.0
McConell et al.[59] (1999)
8M
67
CHO mixh
8.0
TTE
69%
152
4.2
1.1
Angus et al.[60] (2000)
8M
65
Gatorade
6.0
TT
35 kJ/kg
178
7.2
2.1
Febbraio et al.[61] (2000)
7M
63
Lucozade
4.5
TT
7 kJ/kg
148
9.9
3.6
7M
63
Lucozade
4.5
TT
7 kJ/kg
148
4.0
3.6
7M
63
Lucozade
9.0
TT
7 kJ/kg
148
11.9
3.6
McConell et al.[62] (2000)
13 M
66
Glucose
5.5
TTE
83%
70
-0.2
0.8
Mitchell et al.[63] (2000)
10 M
55
CHO mixh
8.0
TT
10 km
42
0.5
1.0
10 M
55
CHO mixh
8.0
TT
10 km
42
0.2
1.0
10 M
55
Glucose
6.0
TT
10 km
42
0.5
1.0
10 M
55
Banana
6.0
TT
10 km
42
0.8
1.0
10 M
55
CHO mixh
6.0
TT
10 km
42
0.3
1.0 1.0
Bishop et al.[64] (2001)
9M
53
Glucose
5.0
TTE
75%
77
3.2
Chinevere et al.[65] (2002)
9M
62
Polydextrose
7.3
TT
449 kJ
92
9.1
1.1
Chryssanthopoulos et al.[66] (2002)
10 M
59
Lucozade
6.9
TTE
70%
112
1.8
0.9
Carter et al.[67] (2003) Jentjens et al.[68] (2003)
7M
60
Glucose polymers
6.4
TTE
60%
123
4.0
1.4
8M
60
Glucose polymers
6.4
TTE
60%
51
3.3
1.3
64
Glucose
TT
691 kJ
46
-1.1
1.1
64
Glucose
15
TT
691 kJ
46
-0.8
1.1
8M
64
Glucose
40
TT
691 kJ
46
0.0
1.2
Desbrow et al.[69] (2004)
9M
65
Gatorade
6.0
TT
1053 kJ
63
0.0
1.1
Nikolopoulos et al.[70] (2004)
8M
66
Lucozade
6.4
TTE
85%
51
1.4
1.1
Saunders et al.[71] (2004)
15 M
53
Acceleradeg
9.1
TTE
75%
82
3.3
0.8
[72]
10 M
66
Powergel
17.1
TT
21 km
74
0.3
0.4
90 M, 8 F
60
Sucrose
6.9
TT
18 km
78
-0.4
0.3
Burke et al.
(2005)
Van Nieuwenhoven et al.[73] (2005) Gusbakti[74] (2006) Sports Med 2011; 41 (9)
Romano-Ely et al.[75] (2006) Van Essen and Gibala
[76]
(2006)
10 M
45
CHO mixh
10 M
45
CHO mixh
14 M
60
CHO mixh
10 M
63
Sucrose
10 M
63
CHO + proteini
5.0
6.0
TTE
64%
66
2.9
0.6
TTE
64%
66
5.8
1.8
9.3
TTE
70%
96
0.4
1.3
6.0
TT
80 km
141
9.5
2.0
8.0
TT
80 km
141
8.2
1.6
12
Continued next page
Vandenbogaerde & Hopkins
9M 9M
Study (y)
Sample size
. VO2maxa
Supplement
CHO + protein (%)b
TT or TTE
Exercise protocolc
Exercise duration (min)d
Power effect (%)e
Power SE (%)f
Saunders et al.[77] (2007)
8 M, 5 F
63
CHO + proteini
TTE
75%
103
1.5
Abbiss et al.[78] (2008)
10 M
62
Sucrose gel
25
TT
16.1 km
42
5.6
2.5
10 M
62
Sucrose gel
25
TT
16.1 km
93
3.3
1.0
Campbell et al.[79] (2008)
Currell and Jeukendrup[80] (2008)
9.1
0.7
8 M, 8 F
62
Gatorade
5.6
TT
10 km
108
1.3
0.6
8 M, 8 F
62
CHO gel
5.6
TT
10 km
108
1.6
0.6
8 M, 8 F
62
Sport beans
5.6
TT
10 km
108
1.7
0.6
8M
65
Glucose
14.4
TT
983 kJ
82
7.8
1.1
8M
65
CHO mixh
14.4
TT
983 kJ
82
14.9
1.6
Hulston and Jeukendrup[81] (2008)
10 M
66
Glucose
6.4
TT
847 kJ
77
2.6
0.9
Jeukendrup et al.[82] (2008)
12 M
66
CHO mixh
5.9
TT
16 km
25
-0.5
0.6
Osterberg et al.[83] (2008)
13 M
56
CHO mixh
6.0
TT
514 kJ
102
2.5
1.1
13 M
56
CHO + proteini
9.1
TT
514 kJ
102
1.4
1.1
10 M
62
Sucrose gel
9.4
TT
16.1 km
41
0.1
0.9
10 M
62
Sucrose gel
5.8
TT
16.1 km
88
3.1
1.3
Peake et al.[84] (2008)
11 M
53
CHO mixh
7.8
TTE
75%
107
1.1
0.6
11 M
53
CHO mixh
9.7
TTE
75%
107
1.4
0.6
11 M
53
CHO + proteini
9.7
TTE
75%
107
1.9
0.6
Breen et al.[86] (2009)
12 M
63
CHO + proteini
6.8
TT
880 kJ
87
0.1
0.5
Hulston and Jeukendrup[87] (2009)
10 M
62
CHO mixh
6.0
TT
847 kJ
107
5.7
1.5
10 M
62
CHO mixh
6.0
TT
847 kJ
107
5.5
1.4
9M
56
Maltodextrin
6.0
TTE
66%
93
-0.3
1.0
7M
60
Maltodextrin
7.5
TT
90 min
90
4.6
0.9
CHO + proteini
7.8
TT
60 km
135
0.9
0.9
Valentine et al.[85] (2008)
Lacerda et al.
[88]
(2009)
Robson-Ansley et al.[89] (2009)
13 M 61 Saunders et al.[90] (2009) . a Data for VO2max (mL/kg/min) are adjusted to 100% M.
Carbohydrate and Performance
ª 2011 Adis Data Information BV. All rights reserved.
Table I. Contd
Total concentration of CHO + protein; percentage unit is grams per 100 mL of total fluid consumed. . c Data shown for TT are measures of work performed, distance travelled or test duration; data for TTE are intensity expressed as percentage of VO2max. b
Exercise duration = test duration + (fractional utilization) (preload duration), where fractional utilization is the endurance capacity used up by preload (see Methods).
e
Effect of the supplement versus the reference drink on mean power in a non-preloaded time trial.
f
Standard error of the effect (for calculation see Methods).
g
Food or drink containing protein.
h
Combination of CHO.
i
CHO-protein supplement.
CHO = carbohydrate; F = female; . M = male; revs = revolutions; SE = standard error of the effect (for calculation see Methods); TT = time trial presented as test duration or workload; TTE = time-to-exhaustion test; VO2max = maximal oxygen consumption; ~ indicates approximately.
779
Sports Med 2011; 41 (9)
d
Vandenbogaerde & Hopkins
780
Effects on time to exhaustion can be converted into effects in a time trial when the relationship between power output and duration is known.[92] The exercise in all studies that qualified for inclusion with time-to-exhaustion protocols was performed at or below maximal oxygen consumption . (VO2max). An appropriate relationship between power and duration for such exercise is that of Leger and Mercier (equation 1):[93]
P ¼ a b lnðTÞ
(Eq: 1Þ
where P is the power expressed as a percentage of . VO2max, T is the duration, and a and b are constants for a given individual and mode of exercise. We assumed that this relationship could be applied with mean values of a and b for the subjects in a given study, which we derived by solving for a and b using values of power and duration for two intensities of exercise. The first intensity was given by the study itself: the time to exhaustion in the . at the given percentage of . control condition VO2max. We used VO2max as the second intensity, and we estimated the time to exhaustion .at this intensity by regressing mean values of VO2max against log of mean time to exhaustion from the studies reviewed by Billat and Koralzstein.[94] These studies provided values differing in gender (females, males) and exercise mode (cycling, running), but there appeared to be little effect of gender or mode on the regression. We concluded that there was no need to adjust for gender . and mode in predicting time to exhaustion at VO2max. The values of a and b for the given study were then used to convert the times to exhaustion in control and treatment conditions into the ratio of power outputs in equivalent time trials using the relationship (equation 2):
Pt =Pc ¼ ða b lnðTc Þ=ða b lnðTt ÞÞ
(Eq: 2Þ
where c and t are control and treatment conditions. The percentage effect of the supplement on timetrial power was then calculated as 100 (Pt/Pc – 1). We performed a sensitivity analysis to investigate the effect of error in the prediction of time to ex. haustion at VO2max equal to twice the standard error of the estimate. We observed ~5% change in the effect on time-trial power (e.g. 1.9% became 2.0%), so we are confident that this approach to
ª 2011 Adis Data Information BV. All rights reserved.
converting effects on times to exhaustion into effects on time-trial times is trustworthy. Effects on performance time in time trials were first converted to effects on mean power output by using the power-speed relationship (equation 3):
P ¼ k Sx ¼ a ðD=TÞx
(Eq: 3Þ
where P is power, S is speed, D is distance, T is performance time, and k and x are constants.[91] Thus Pt/Pc = (Dt/Dc)x or (Tc/Tt)x [performance in time trials that was measured as power output or work done did not require this conversion]. The constant x is 1.0 for running, but x varies between cycle ergometers, depending on the way they simulate distance travelled. For the Monark, x = 1.0.[90] For the Velodyne, x = 2.8; for the Velotron, x = 2.0 and 2.5 for time trials that include and do not include climbs (Paton CD and Hopkins WG, unpublished observations). A value of x was not available for the Politecnica ergometer so we used the power-speed relationship P = 9.65 S – 86.7.[91] The approach for then converting the performance effect on power in a preloaded time trial into an effect in a non-preloaded time trial was novel and based on the assumption that the factor increase in error of measurement produced by a preload would apply also to the effects on performance. The conversion involved the following steps: estimate the typical (standard) error of measurement in all time trials where an exact p-value or confidence limits were provided; estimate the fraction of endurance capacity utilized in the preload; derive the relationship between error and fractional utilization; use the relationship to predict the factor reduction in the error for a study without a preload (fractional utilization = 0); and finally, for all time trials, use this factor to convert the effect with the preload into an effect without a preload. The typical error of measurement was calculated as SEM O(sample size)/O2, where SEM was the standard error of the mean effect estimated from the exact p-value and/or confidence limits via the t-statistic and its degrees of freedom (sample size – 1). The fractional utilization was calculated as (duration of preload)/(time to exhaustion at the preload intensity); time to exhaustion at the .preload intensity was assumed to be dependent on VO2max and was calculated with the Leger equation from estimates of a and b, which we
Sports Med 2011; 41 (9)
Carbohydrate and Performance
derived .by regressing values of a and b against values of VO2max from the time-to-exhaustion studies (see previous paragraph). For the relationship between error and fractional utilization, we started with fractional utilization as a linear predictor of log(error), but we found substantial improvement in the prediction by including log(duration) interacted with fractional utilization as an extra predictor. A meta-analysis also requires calculation of a standard error for each study estimate. For studies that provided an exact p-value or confidence limits, we calculated the standard error for the effect on mean power in the equivalent non-preloaded time trial via the t-statistic and its degrees of freedom. Each of these studies also provided an estimate of the typical error for mean power in a time trial via the relationship between typical error and standard error (see previous paragraph). The typical error from the time-toexhaustion studies was then averaged (via variances), allocated to time-to-exhaustion studies that did not report an exact p-value or confidence limits, and the standard error was calculated via the relationship between typical and standard error. For time-trial studies that did not report an exact p-value or confidence limits, the typical error predicted for zero fractional utilization in the relationship between error and fractional utilization (see previous paragraph) was used to calculate the standard error for the estimate of mean power in non-preloaded time trials. 1.2.2 Study Characteristics
Table II shows the mean study and study. estimate characteristics. VO2max was included as a predictor in the meta-analytic model and was therefore adjusted to the value expected for 100% males in studies of cyclists via a preliminary multiple linear regression in which sex and mode of exercise. (running or cycling) were linear predictors of VO2max (without interaction). In this preliminary analysis, and in the main meta-analysis, sex was coded as a variable with values equal to the fraction of males in the sample. All food and fluid consumed after the last meal before exercise was combined into an average supplement composition, and this supplement was assumed to be distributed equally across the boluses. The metaª 2011 Adis Data Information BV. All rights reserved.
781
analysis included the timing of intake of the first supplement bolus before the start of exercise, coded as a quadratic to determine the optimal timing. We included the frequency of supplement ingestion as the number of boluses divided by the time from intake of the first bolus to the end of exercise, on the assumption that equal boluses were ingested at equal intervals. Exercise duration for times to exhaustion was included as the log transformation of the test duration of the placebo treatment; for time trials, exercise duration was adjusted to include any preload by adding (fractional utilization) (preload duration) to the test duration before log transformation. Blinding of the treatment was coded as a variable with values 0 or 1. Table III shows the supplement ingestion regimens for the study estimates. The measures that were included in the analysis as predictors were: the total percentage concentration of carbohydrate and protein (grams of carbohydrate plus protein per 100 mL of total fluid ingested); the rates of ingestion of each of glucose, sucrose, fructose, and glucose polymers (grams per kilogram of body mass per hour, where hour referred to the time measured from the first ingestion of carbohydrate to the end of exercise); and the inclusion of salt (NaCl), coded as 0 or 1. We were able to obtain some data missing from the manuscripts by emailing authors. Where we could not retrieve data, we. derived values as follows. Missing . data for VO2max were assigned the mean VO2max from other studies with similar subjects. Studies without a stated ambient temperature were assigned 21.3C, which was the average for all performance tests that appeared to be performed at normal room temperatures. Missing data for supplement characteristics were either retrieved from the manufacturer’s website or were assigned the average of all supplements with similar carbohydrate concentration. Where data on salt inclusion were missing, we assumed that commercially available supplements contained salt, and that supplements that were made in a laboratory did not. A spreadsheet containing individual values of the characteristics summarized in tables II and III can be obtained from the authors.
Sports Med 2011; 41 (9)
Vandenbogaerde & Hopkins
782
1.2.3 Publication Bias and Outliers
Published effects are on average larger than true values, owing to the misuse of statistical significance as a criterion for publication.[95] To reduce the effects of publication bias, we plotted the
standard error of each study estimate versus the t-value for the solution for the between-study random effect (figure 1).[96] A line was drawn at a value for the standard error that divided the scatter into a symmetrical plot on the left and an
Table II. Study and study-estimate characteristics. Data are either counts, mean – SD or proportions (%) Studies with SE <1.25
All studies TT
TTE
TT
TTE
Study characteristics No. of studies
56
32
44
29
Estimates per study
1.8 – 0.9
1.6 – 0.9
1.7 – 1.0
1.6 – 0.9
Sample size
11.7 – 12.1
9.2 – 2.7
12.6 – 13.5
9.3 – 2.6
Males (%)
93
95
92
94
Mean body mass (kg) . Mean VO2max (mL/kg/min)a
72.1 – 4.1
70.6 – 4.6
71.9 – 4.1
70.8 – 4.1
61.6 – 5.3
58.3 – 6.0
61.4 – 5.3
58.2 – 6.3
Cycling (%)b
84
81
82
83
Fasting period (h)c
7.4 – 4.3
9.7 – 3.9
7.6 – 4.2
9.8 – 3.9
Ambient temperature (C)
22.5 – 3.9
21.8 – 3.1
23.1 – 4.1
21.3 – 2.1
Intake-time first bolus (h)d
0.5 – 0.7
0.5 – 0.6
0.5 – 0.8
0.5 – 0.6
No. of boluses
5.5 – 3.0
5.7 – 3.4
5.0 – 3.0
5.7 – 3.5
Boluses per he
3.0 – 1.4
3.0 – 1.7
2.8 – 1.5
2.9 – 1.7
Preload intensity (%)f
68 – 7
–
68 – 7
–
Preload duration (min)f
89 – 30
–
84 – 30
–
Fractional utilizationg
0.73 – 0.42
–
0.73 – 0.45
–
Test duration (min)h
53 – 50
103 – 41
47 – 44
106 – 41
Exercise duration (min)i
92 – 41
103 – 41
85 – 41
106 – 41
Study-estimate characteristics No. of study estimates
103
52
75
47
Blinded (%)j
90
65
91
62
Change in power (%)k
2.7 – 3.1
1.7 – 1.8
1.7 – 2.0
1.5 – 1.6
SEl
1.2 – 0.7
1.0 – 0.3
0.9 – 0.3
0.9 – 0.2
Typical errorl
2.7 – 1.1
2.0 – 0.5
2.1 – 0.4
1.9 – 0.3
a
Adjusted to 100% males.
b
Percentage of studies that used cycling exercise.
c
Time between last feeding defined by the researchers as a meal and start of exercise.
d
Measured to the beginning of the adjusted period of exercise.
e
Number of boluses/actual time from first bolus to end of exercise.
f
For 34 and 26 preloaded TTs in all studies and studies with SE <1.25%, respectively.
g
Endurance capacity used up by preload (see Methods).
h
For the placebo treatment.
i
Test duration + (fractional utilization) (preload duration).
j
Percentage of study estimates from placebo-controlled (at least single-blind) designs.
k
After conversion into non-preloaded TT (see Methods).
l
Adjusted for fractional utilization = 0 (see Methods).
. SE = standard error of the effect (for calculation see Methods); TT = time trial; TTE = time-to-exhaustion test; VO2max = maximal oxygen consumption.
ª 2011 Adis Data Information BV. All rights reserved.
Sports Med 2011; 41 (9)
Carbohydrate and Performance
783
Table III. Supplement ingestion regimens for the study estimates. Data are mean – SD or proportions (%) Studies with SE <1.25
All studies TT (n = 103)
TTE (n = 52)
TT (n = 75)
CHO + protein (%)a
9.6 – 6.9
11.6 – 8.7
10.0 – 7.6
TTE (n = 47) 11.7 – 9.0
Volume (L)
1.3 – 0.7
1.5 – 1.5
1.1 – 0.6
1.5 – 1.6
Glucose (g/kg/h)
0.4 – 0.5
0.4 – 0.7
0.3 – 0.5
0.5 – 0.7
Sucrose (g/kg/h)
0.3 – 0.5
0.3 – 0.6
0.2 – 0.4
0.2 – 0.5
Fructose (g/kg/h)
0.2 – 0.3
0.3 – 1.0
0.2 – 0.3
0.3 – 1.0
Glucose polymers (g/kg/h)
0.4 – 0.8
0.5 – 0.9
0.4 – 0.8
0.5 – 0.9
Protein (g/kg/h)
0.01 – 0.07
0.04 – 0.10
0.01 – 0.05
0.03 – 0.10
Protein included (%)
5.8
13
5.3
15
Salted (%)b
59
56
55
51
a
Percentage unit is total grams per 100 mL of total fluid consumed.
b
Percentage of study estimates with salt in the supplement.
CHO + protein = total concentration of carbohydrate plus protein; SE = standard error of the effect (for calculation see Methods); TT = time trial; TTE = time-to-exhaustion test.
1.3 Meta-Analytic Model
The meta-analyses were performed with the mixed linear modelling procedure (Proc Mixed) in the Statistical Analysis System (Version 9.2, SAS Institute, Cary, NC, USA). Percentage effects on mean power output were converted to factors (= 1+effect/100), log transformed for the analysis, then back transformed to percentages. The fixed effects in the meta-analytic model consisted of binary and continuous predictors representing study characteristics. The binary predictors were mode of exercise (1 = cycling, 0 = running), type of performance test (1 = time trial, 0 = time to exhaustion), blinding (1 = blind), salt in drink (1 = included), and variables for each level (low, moderate and high) of each carbohydrate, protein, and concentration of carbohydrate plus protein, as defined in table IV. The continuous predictors were: ambient temperature; adjusted . VO2max; time of ingestion of first bolus before exercise (as a quadratic); fasting period before exercise; fraction of males in the sample; number ª 2011 Adis Data Information BV. All rights reserved.
of boluses per hour; the amount of glucose, sucrose, fructose and glucose polymers ingested; and the total concentration of carbohydrate plus protein. Values of the predictor variables characterizing the composition of the carbohydrate supplement were the values in the supplement treatment minus the values in the control or reference treatment. In a further analysis, we included a predictor to account for the possibility of a synergistic effect of co-ingesting two or more carbohydrates. The predictor for synergism was coded 0 for ingestion of one carbohydrate source, and 1 for
4
2 t-Value
asymmetrical plot on the right. Asymmetrical scatter is very likely the result of a publication trend towards positive effects, so the meta-analysis was also performed only for those study estimates falling on the left of the line. T-values of the study estimates were also used to assess the presence of outliers.[96]
0 −2 −4 0
1
2
3
4
Standard error (%) Fig. 1. Scatter plot to investigate outlier studies and publication bias. The t-value of the solution of the between-study random effect is plotted against the standard error for the study estimate. Dashed line at a standard error of ~1.25% divides the plot into a region with symmetric scatter (in the vertical direction) to the left and a region to the right where a dearth of t-values within the dashed curve is apparent.
Sports Med 2011; 41 (9)
Vandenbogaerde & Hopkins
784
ingestion of two or more carbohydrate sources. The random effects in the model specified withinstudy standard deviation and between-study standard deviation; negative estimates for variances were allowed, and standard deviation derived from negative variances were expressed as negative standard deviation. At the suggestion of a reviewer, we used the above mixed model to analyse the studies in two groups: those where ingestion began earlier than 15 minutes before exercise and those where ingestion began at 15 minutes or later. To allow for greater non-linearity in the effect of time of commencement of ingestion, we also modelled its effect as a cubic with the full data set. Uncertainty in the meta-analysed estimates is reported as 90% confidence limits. We made probabilistic magnitude-based inferences about the true values of outcomes, as described elsewhere.[96] In brief, an outcome was deemed unclear if its confidence interval overlapped thresholds for smallest worthwhile positive and negative effects; equivalently, effects were unclear if chances of the true value being substantially positive and negative were both >5%. The magnitude of a clear effect was reported as the magnitude of its observed value, sometimes with an assertion about the probability that the true value was substantial. The thresholds for small, moderate, large and very large effects on performance were assumed to be 0.3, 0.9, 1.6 and 2.5 of the race-to-race within-athlete variability in competitive performance of elite athletes.[96] Although the meta-analysed studies used mainly sub-elite athletes or active nonathletes, thresholds for elite athletes were used here, as in all previous studies using this approach to inferences, under the assumption that the effects will apply to elites and that medal winning by elites is more important than that by non-elites. The variabilities in mean power output for endurance running and cycling in races are ~0.8%[97] and ~3.5%,[98] respectively, giving smallest effects of 0.25% and ~1.0%. We chose 0.5% as a value to apply to high-intensity endurance sports generally; thresholds for moderate, large and very large effects were therefore 1.5%, 2.7% and 4.2%. The magnitudes of the effects of continuous predictors . (VO2max, ambient temperature, exercise duraª 2011 Adis Data Information BV. All rights reserved.
Table IV. Components of the meta-analytic model expressed as additive percentage effects (–90% confidence limits) on mean power in a non-preloaded time trial. Effects shown are for a difference between levels (e.g. female – male) or for an increase of ~2 SD (e.g. . VO2max per 10 mL/kg/min) All studies
Studies with SE <1.25%
0.8, –1.9
0.6, –1.8
low
1.1, –1.9
1.1, –1.7
moderate
1.2, –1.8
1.1, –1.6
high
0.5, –2.0
0.3, –1.7
Supplement composition Intercepta CHO + proteinb
Glucosec low moderate high
-0.1, –0.7
-0.2, –0.6
0.4, –1.0
-0.6, –0.9
-0.9, –1.3
-0.3, –1.2
Sucrosec -0.2, –0.6
-0.2, –0.5
moderate
0.3, –0.9
-0.7, –0.8
high
1.4, –1.7
-0.2, –1.5
low
Fructosec 0.4, –0.6
0.3, –0.6
moderate
-1.2; –1.3
-2.2, –1.1
high
-1.7, –2.3
-2.2, –2.1
low
Glucose polymersc -0.3, –0.7
-0.2, –0.7
moderate
0.4, –1.0
0.0, –1.0
high
1.5, –1.4
0.6, –1.4
low
0.1, –1.1
0.2, –1.0
moderate
1.9, –1.4
1.1, –1.2
high
1.0, –3.9
0.6, –3.3
low
Proteind
Study characteristics Female – male . VO2maxe (per 10 mL/kg/min)
0.3, –2.3
-0.4, –2.0
-0.4, –0.7
-0.1, –0.4
Ambient temp. (per 10 C)
-0.5, –0.9
-0.5, –0.8
0.0, –1.0
-0.2, –0.9
-1.0, –0.8
-0.5, –0.8
low CHO + proteinc
1.1, –1.7
1.2, –1.7
moderate CHO + protein
1.2, –0.7
0.2, –0.7
-0.4, –0.8
-0.5, –0.7
Fasting (per 9 h)
0.5, –0.9
1.1, –0.8
Blind – not blind
0.2, –1.0
0.1, –1.0
Inclusion of salt
-0.4, –0.6
-0.3, –0.6
0.9, –0.9
0.7, –0.8
Running – cycling TTE – TT Exercise durationf (per 3-fold increase)
high CHO + protein
Boluses (per 3 per h)
Continued next page
Sports Med 2011; 41 (9)
Carbohydrate and Performance
785
Table IV. Contd All studies
Studies with SE <1.25%
Time of first bolus ingestiong linear (per h) quadratic (per h2) maximum effect (at ~4 h)
1.4, –0.9
1.1, –0.8
-0.3, –0.2
-0.1, –0.2
1.4, –2.3
2.4, –2.1
1.7, –0.4
1.4, –0.3
Random Variation Between-study SD
0.4, –0.6 -0.4, –0.5 . Intercept for a male subject with a VO2max of 60 mL/kg/min performing a 100 min cycling time trial at 22C after a 4 h fast, who knowingly ingests one litre per 70 kg per h of a supplement without salt as three boluses per h, with the first bolus ingestion at the start of exercise.
Within-study SD a
b
Thresholds (%): 0
c
Thresholds (g/kg/h): 0
d
Thresholds (g/kg/h): 0
e
Adjusted to 100% males.
f
Test duration + (fractional utilization) (preload duration).
g
Before the start of exercise.
CHO + protein = total concentration of carbohydrate plus protein; SE = standard error . of the effect (for calculation see Methods); temp. = temperature; VO2max = maximal oxygen consumption; ~ indicates approximately.
tion, fasting period, rate and timing of bolus ingestion) were assessed for an increase of ~2 between-study standard deviations of the predictor (or of the log of the predictor, for those that were log-transformed for the analysis).[96] By analogy with this approach, we assessed the magnitude of the between-study random effect for twice its standard deviation. 2. Results Figure 1 shows the plot that was used to assess the presence of outliers and publication bias. The plot showed an asymmetrical scatter for studies with an error more than ~1.25%. The metaanalysis was then repeated for studies with a standard error <1.25% to reduce publication bias. None of the t-values were considered sufficiently large to warrant exclusion of the study estimate. Table IV shows the components of the metaanalytic model expressed as additive percentage effects (–90% confidence limits) on mean power in a non-preloaded time trial. Effects for suppleª 2011 Adis Data Information BV. All rights reserved.
ment composition were generally reduced by up to ~1.5% after adjusting for publication bias. Combining these effects into a meta-analysed performance effect reduced the effect by up to 2% (e.g. from 5% to 3%). Here we describe the effects only for studies with a standard error <1.25%. The effects shown for the supplement composition in table IV have to be added to an intercept for a given condition, which can be illustrated with the following example. . The intercept is 0.6% for a male subject with a VO2max of 60 mL/kg/min performing a 100-minute cycling time trial with no preload at 22C after a 4-hour fast who knowingly ingests one litre per 70 kg per hour of a supplement without salt in six boluses per hour, with the first bolus ingestion at the start of exercise. For a supplement providing 1.1 g/kg/h glucose polymers and 0.1 g/kg/h protein in a drink with an 8% concentration of carbohydrate plus protein, the meta-analysed effect on performance is 5.3% (90% confidence limits, –2.6%). The best effect for this condition was a performance increase of 6.5% (–2.5%) for a supplement providing ~0.7 g/kg/h glucose polymers, ~0.2 g/kg/h fructose and ~0.2 g/kg/h protein. Carbohydrate supplementation regimens may also impair performance. The largest impairment (-2.1%, –2.3%) was found for a >15% fructose supplement providing >0.5 g/kg/h ingested as one bolus per hour with the first bolus ingestion at the start of exercise. Reductions in the performance effect of carbohydrate supplementation were possibly moderate with a supplement providing >0.25 g/kg/h fructose. Reductions were possibly small with an increase in ambient temperature of 10C and in time-to-exhaustion tests compared with time trials. A 3-fold increase in exercise duration also reduced the performance benefit when supplements with high carbohydrate-plus-protein concentration were ingested; with low concentrations the performance benefit was moderately increased for longer exercise durations, but this effect was unclear. Increases in the effectiveness of carbohydrate ingestion on performance were likely to be small with an additional 9-hour fast and with ingesting moderate rates of protein. Increases were possibly small with increasing the frequency of supplement ingestion by three boluses per hour Sports Med 2011; 41 (9)
Vandenbogaerde & Hopkins
786
and likely small-moderate with ingesting the first bolus not at the start of exercise but 1–4 hours before exercise. Effects were likely trivial for including. salt in the supplement and for subjects with a VO2max higher by 10 mL/kg/min; effects of gender, exercise mode, exercise duration when supplements with low or moderate carbohydrateplus-protein concentrations were ingested and blinding of treatment were unclear. The magnitude of the between-study random variation (interpreted as two between-study standard deviations) was large and positive. The magnitude of the within-study random variation was small and negative, although its confidence interval allowed for trivial positive variation. The estimate for synergism arising from coingestion of two or more carbohydrates had a negative trivial value; the effect was unclear, but at most small (-0.3%, –1.0%). The contributions of the components of the meta-analytic model in studies where ingestion began earlier than 15 minutes before exercise were very similar to those in the analysis of the full data set. A substantial difference was a greater contribution of consumption of multiple boluses (per 3 per hour: 1.9%, –1.2%). The only other major difference was a moderate but unclear benefit of high sucrose ingestion (2.8%, –4.5%). Contributions of components in studies where ingestion began £15 minutes before the start of exercise were generally less clear, but the combinations investigated for the full data set produced similar effects on performance and a similar effect of time of commencement of ingestion (over the range 15 minutes before exercise to 30 minutes after). In the analysis where we modelled the effect of time of ingestion of first bolus as a cubic, the coefficient of the cubic was practically zero, the maximum still occurred at 4 hours before exercise and there was almost exactly the same gradual decrease between 4 hours before exercise and 1 hour post-exercise. 3. Discussion This study is the first meta-analytic review of the effects of acute carbohydrate supplementation on performance. To our knowledge, this metaª 2011 Adis Data Information BV. All rights reserved.
analysis is also the first in our discipline to account for publication bias. This bias was substantial (up to 2% increase in the performance effect of supplements), which underscores the value of metaanalysis in providing more realistic performance effects of treatments than would be obtained in narrative reviews or from a researcher’s impression of the effect of supplements. After adjusting for publication bias, the meta-analysed performance effects of carbohydrate supplements ranged from clear improvements of ~6% to clear impairments of ~2%. The best supplement derived from the analysis provided ~0.7 g/kg/h glucose polymers, ~0.2 g/kg/h fructose and ~0.2 g/kg/h protein. A possible explanation for this combination of carbohydrates is that the increase in carbohydrate oxidation rate with ingestion of several varieties of carbohydrate helps the athlete to sustain exercise intensity (for review see Jeukendrup[2]). Our analysis did not provide evidence for an additional synergistic effect of ingesting several carbohydrate sources (other than a simple additive effect), although we cannot exclude a small positive or negative synergistic effect. The meta-analysis also shows that the best single source of carbohydrate consumed at a high rate is glucose polymers. This superiority is apparently not due to the lower osmolarity of glucose polymers, because osmolarity has little effect on gastric emptying and carbohydrate oxidation.[99] One possible explanation is that it may cause less gastrointestinal distress compared with the other carbohydrates.[100] Inclusion of supplement osmolarity and gastrointestinal distress in the metaanalytic model might have helped resolve this issue, but unfortunately we were unable to code osmolarity and gastrointestinal distress for enough studies from the limited information provided by authors and manufacturers. Recent studies, including some included in this meta-analysis, have provided evidence for higher rates of oxidation of carbohydrate and enhanced performance with ingestion of multiple forms of carbohydrate, mediated via multiple carbohydrate transporters in the intestinal epithelium.[2] Inspection of the effects of the different kinds of carbohydrate in table IV for the studies with standard error <1.25% shows that most are negative, Sports Med 2011; 41 (9)
Carbohydrate and Performance
the only exceptions being high rates of glucose polymers and low rates of fructose (hence our recommendation above). Thus, the meta-analysis provides only limited evidence for the benefit of ingestion of multiple forms of carbohydrate, namely fructose plus glucose polymers. Although the confidence limits allow for the possibility of benefits from including glucose and sucrose, it is possible that the higher carbohydrate oxidation rates with multiple forms of carbohydrate are not accompanied proportionally by enhanced performance. Our meta-analysis also suggests that more than low rates of consumption of fructose should be avoided. Indeed, the largest performance impairment derived from the analysis would occur with ingestion of a single moderate bolus of fructose at the start of exercise. Possible explanations for this harmful effect are gastrointestinal distress[23] and the conversion of fructose to glucose, which may be too slow to maintain high carbohydrate oxidation rates in the later stages of exercise.[101] However, several authors have reported reduced gastrointestinal distress and enhanced performance when higher doses of fructose are ingested with other carbohydrates.[102,103] This research was not included in the meta-analysis because of the complexity of the performance tests. Whether or not protein should be included in a carbohydrate supplement is a topical issue. The effects of protein in this meta-analysis are effects to be added to those of carbohydrate, which was always ingested with any protein in the analysed studies. Our meta-analysis suggests that the inclusion of moderate amounts of protein in a carbohydrate supplement substantially increases performance. From the perspective of energy intake, moderate rates of protein ingestion are similar to low rates of carbohydrate ingestion. Effects for these corresponding levels of protein and carbohydrate in table IV suggest that protein is potentially more effective than any of the carbohydrates and may therefore mediate its effect on performance via more than simply provision of energy. One possibility is a placebo effect with protein, because it is difficult to blind subjects to the addition of protein in a carbohydrate beverage. The overall placebo effect in the meta-analysis was trivial, but there is substantial uncertainty in ª 2011 Adis Data Information BV. All rights reserved.
787
the estimate and, in any case, subjects might get an extra placebo effect in protein-supplement studies from second guessing that they have received a treatment that is potentially beneficial. Another possible explanation is a delay in central fatigue: researchers have suggested that the increase in amino-acid availability with protein ingestion may result in a decrease in serotonin levels and thereby delay central fatigue.[104] A possible explanation for the reduction in the performance benefit with ingesting high concentrations of carbohydrate and protein is slower gastric emptying,[105] which would limit the rate of carbohydrate absorption. High concentrations of carbohydrate and protein could also compromise fluid balance, either because the high concentration represents inadequate fluid intake or because the high osmolarity could draw fluid from the circulation into the gut.[106] The trivial effect of salt inclusion in the supplement was unexpected: researchers have demonstrated that salt increases the net intestinal absorption rate of carbohydrate and that it promotes retention of ingested fluids,[107] which should have a beneficial effect on performance. Our variable representing salt was crude because of poor reporting in many of the studies. We expect that inclusion of salt in a supplement is beneficial provided it is not present at high concentrations. Various study characteristics clearly moderated the performance effect of carbohydrate supplements. A possible mechanism for the reduction in the performance effect of carbohydrate with an increase in ambient temperature is the increase in core temperature that has been demonstrated during high-intensity exercise with carbohydrate ingestion compared with placebo:[108] this increase in core temperature may cause an earlier onset of fatigue in a warmer environment; another possible mechanism is a decrease in the contribution of exogenous carbohydrate to substrate utilization at higher environmental temperatures.[109] Both of these mechanisms could arise from changes in blood flow to gut, skin and muscle. The effect of the timing of first bolus ingestion appeared to be modelled adequately with a quadratic: the cubic produced little additional Sports Med 2011; 41 (9)
788
difference in the effects of time of ingestion before versus during exercise, even though there could be substantial physiological differences between ingesting carbohydrates either before or during exercise. If more studies were available where ingestion began immediately before or during exercise, meta-analysis might reveal different effects for some carbohydrates and/or protein compared with their effects when ingestion begins earlier. However, the smaller performance gains from such ingestion regimens would make further investigation of this issue academic. The benefits with increasing the frequency of bolus ingestion and with ingesting the first bolus of the supplement earlier than at the start of exercise may have several explanations: a reduction in gastrointestinal distress; a change in glucose metabolism mediated by changes in the insulin response;[110] or possibly even effects of ongoing stimulation of carbohydrate receptors in the mouth.[111] A longer fast increased the performance effect of carbohydrate ingestion, probably because of a reduction in hepatic glycogen stores. This finding is in agreement with the ~1% enhancement arising from an additional 9-hour fast on performance in a 21 km running time trial in Wong et al.,[112] a study we could not include in the analysis because of its design. Although the effect had considerable uncertainty, a 3-fold increase in exercise duration increased the performance benefit of a supplement when drinks of low carbohydrate-plus-protein concentration were ingested. Gastrointestinal distress and/or a delay in gastric emptying may explain why the opposite (and clear) effect was seen for supplements with high concentrations. Low concentration drinks may provide sufficient energy to maintain high carbohydrate oxidation rates and, the longer the test, the more performance may depend on this provision of energy. There may be some subtle confounding effects of exercise duration in timeto-exhaustion tests arising from substantial performance enhancements and impairments; the meta-analytic model did not account for any differences in energy consumed with the substantial changes in exercise duration that can occur with this protocol. Moderating effects of gender and blinding were unclear, which indicates that more ª 2011 Adis Data Information BV. All rights reserved.
Vandenbogaerde & Hopkins
research is necessary to investigate these predictors. Researchers should be cautious to extrapolate results of this meta-analysis to high-intensity shortduration tests, as there was only one test with an exercise duration (adjusted to zero preload) <25 minutes (table I). Carbohydrate supplementation in short-duration exercise is unlikely to benefit performance via increased carbohydrate oxidation, but other mechanisms may be involved, including activation of brain regions via carbohydrate receptors in the mouth.[111] Similarly the longest duration of exercise (including preloads) in the meta-analysis was 3.3 hours (table I), so extrapolation to ultra-endurance exercise is also inappropriate, although clearly there will be beneficial effects for such exercise from carbohydrate supplementation. The meta-analysis did not include studies with substantial rest intervals either because of difficulties with calculation of a fractional utilization or because of interactions with recovery from exercise. All performance tests included in the analysis consisted of continuous cycling or running. The small difference in the performance effect of supplements in time trials versus time-toexhaustion tests is probably an artefact of the ways we have converted time-to-exhaustion tests and preloaded time trials into non-preloaded time trials. In view of the complexity of these conversions, the small magnitude of this difference is reassuring. Insights into the practical application of the findings of the meta-analysis can also be gleaned from a consideration of the between- and withinstudy standard deviations (see end of table IV). These standard deviations represent unexplained variation in the mean effect of the protocol from study to study and within a study; as such, their magnitude is the typical deviation from the metaanalysed mean effects that a researcher or practitioner can expect to experience in his or her setting. The unexplained variation arises in several ways: poor measuring or reporting of covariates; unknown covariates modulating the effect on performance; and a meta-analytic model that does not fully capture the complexity of the underlying reality, including non-linear effects and interactions. The large between-study difference Sports Med 2011; 41 (9)
Carbohydrate and Performance
means that researchers and practitioners might not see beneficial effects of some carbohydrate protocols in their setting. The magnitude of the within-study random variation was negative, which implies variation between estimates within studies being smaller than expected from the standard errors of the estimates; this outcome is probably a consequence of our estimates of measurement error being too large. It follows that our estimates of uncertainty (confidence intervals) are in general conservative, and that our outcomes are trustworthy, at least as far as random sampling variation is concerned. It is also important to understand that some individual athletes may obtain no benefit or even impairment in performance from carbohydrate ingestion, even with those protocols that are clearly beneficial. Meta-analysis cannot address the question of individual responses to treatments until researchers provide complete inferential information about experimental and control groups in the form of confidence limits, exact p-values or, best of all, standard deviations of change scores. In crossover studies, researchers would need to include an extra trial in one or more of the treatment conditions to permit estimation of individual responses.[113,114] We have simplified the various protocols in the studies to address the obvious questions that an athlete would ask about energy supplements for an endurance competition: what should I take, when do I start and how often should I take it? This meta-analysis provides a new best practice that researchers can use to further improve supplementation protocols. 4. Conclusion Carbohydrate supplementation can have large performance benefits in endurance exercise. A good supplementation regimen is to ingest carbohydrate before and during exercise in many boluses with the first bolus up to 4 hours before the start of exercise. Supplements or foods containing high concentrations of carbohydrate or more than small amounts of fructose should be avoided. A carbohydrate ingestion regimen providing ~0.7 g/kg/h glucose polymers, ~0.2 g/kg/h ª 2011 Adis Data Information BV. All rights reserved.
789
fructose and ~0.2 g/kg/h protein may give the largest ergogenic effect. This meta-analysis did not account for individual responses: individual athletes may obtain no benefit or even impairment in performance with carbohydrate ingestion. In future, research should focus on females, short-duration exercise, ultra-endurance exercise, better reporting of inferential statistics and, in crossover studies, inclusion of additional trials. Acknowledgements No funding additional to government occupational salaries was provided for the preparation of this review. The authors have no conflicts of interest. No other people made substantial contributions worthy of acknowledgement.
References 1. Bosch AN. Carbohydrate ingestion during exercise: an aid to endurance performance [review]. Int Sportmed J 2007; 8: 24-30 2. Jeukendrup AE. Carbohydrate feeding during exercise. Eur J Sport Sci 2008; 8: 77-86 3. Foster C, Costill DL, Fink WJ. Effects of preexercise feedings on endurance performance. Med Sci Sports 1979; 11: 1-5 4. Ivy JL, Costill DL, Fink WJ, et al. Influence of caffeine and carbohydrate feedings on endurance performance. Med Sci Sports 1979; 11: 6-11 5. Felig P, Cherif A, Minagawa A, et al. Hypoglycemia during prolonged exercise in normal men. N Engl J Med 1982; 306: 895-900 6. Coyle EF, Hagberg JM, Hurley BF, et al. Carbohydrate feeding during prolonged strenuous exercise can delay fatigue. J Appl Physiol 1983; 55: 230-5 7. Bjorkman O. Influence of glucose and fructose ingestion on the capacity for long-term exercise in well-trained men. Clin Physiol 1984; 4: 483-94 8. Decombaz J, Sartori D, Arnaud MJ, et al. Oxidation and metabolic effects of fructose or glucose ingested before exercise. Int J Sports Med 1985; 6: 282-6 9. Coyle EF, Coggan AR, Hemmert MK, et al. Muscle glycogen utilization during prolonged strenuous exercise when fed carbohydrate. J Appl Physiol 1986; 61: 165-72 10. Gleeson M, Maughan RJ, Greenhaff PL. Comparison of the effects of pre-exercise feeding of glucose, glycerol and placebo on endurance and fuel homeostasis in man. Eur J Appl Physiol 1986; 55: 645-53 11. Flynn MG, Costill DL, Hawley JA, et al. Influence of selected carbohydrate drinks on cycling performance and glycogen use. Med Sci Sports Exerc 1987; 19: 37-40 12. Hargreaves M, Costill DL, Fink WJ, et al. Effect of preexercise carbohydrate feedings on endurance cycling performance. Med Sci Sports Exerc 1987; 19: 33-6 13. Murray R, Eddy DE, Murray TW, et al. The effect of fluid and carbohydrate feedings during intermittent cycling exercise. Med Sci Sports Exerc 1987; 19: 597-604
Sports Med 2011; 41 (9)
790
14. Neufer PD, Costill DL, Flynn MG, et al. Improvements in exercise performance: effects of carbohydrate feedings and diet. J Appl Physiol 1987; 62: 983-8 15. Sasaki H, Maeda J, Usui S, et al. Effect of sucrose and caffeine ingestion on performance of prolonged strenuous running. Int J Sports Med 1987; 8: 261-5 16. Davis JM, Burgess WA, Slentz CA, et al. Effects of ingesting 6% and 12% glucose/electrolyte beverages during prolonged intermittent cycling in the heat. Eur J Appl Physiol 1988; 57: 563-9 17. Davis JM, Lamb DR, Pate RR, et al. Carbohydrateelectrolyte drinks: effects on endurance cycling in the heat. Am J Clin Nutr 1988; 48: 1023-30 18. Mitchell JB, Costill DL, Houmard JA, et al. Effects of carbohydrate ingestion on gastric emptying and exercise performance. Med Sci Sports Exerc 1988; 20: 110-5 19. Coggan AR, Coyle EF. Metabolism and performance following carbohydrate ingestion late in exercise. Med Sci Sports Exerc 1989; 21: 59-65 20. Maughan RJ, Fenn CE, Leiper JB. Effects of fluid, electrolyte and substrate ingestion on endurance capacity. Eur J Appl Physiol 1989; 58: 481-6 21. Mitchell JB, Costill DL, Houmard JA, et al. Influence of carbohydrate dosage on exercise performance and glycogen metabolism. J Appl Physiol 1989; 67: 1843-9 22. Murray R, Seifert JG, Eddy DE, et al. Carbohydrate feeding and exercise: effect of beverage carbohydrate content. Eur J Appl Physiol 1989; 59: 152-8 23. Murray R, Paul GL, Seifert JG, et al. The effects of glucose, fructose, and sucrose ingestion during exercise. Med Sci Sports Exerc 1989; 21: 275-82 24. Sherman WM, Brodowicz G, Wright DA, et al. Effects of 4 h preexercise carbohydrate feedings on cycling performance. Med Sci Sports Exerc 1989; 21: 598-604 25. Powers SK, Lawler J, Dodd S, et al. Fluid replacement drinks during high intensity exercise effects on minimizing exercise-induced disturbances in homeostasis. Eur J Appl Physiol 1990; 60: 54-60 26. Williams C, Nute MG, Broadbank L, et al. Influence of fluid intake on endurance running performance. Eur J Appl Physiol 1990; 60: 112-9 27. Murray R, Paul GL, Seifert JG, et al. Responses to varying rates of carbohydrate ingestion during exercise. Med Sci Sports Exerc 1991; 23: 713-8 28. Sherman WM, Peden MC, Wright DA. Carbohydrate feedings 1 h before exercise improves cycling performance. Am J Clin Nutr 1991; 54: 866-70 29. Thomas DE, Brotherhood JR, Brand JC. Carbohydrate feeding before exercise: effect of glycemic index. Int J Sports Med 1991; 12: 180-6 30. Wright DA, Sherman WM, Dernbach AR. Carbohydrate feedings before, during, or in combination improve cycling endurance performance. J Appl Physiol 1991; 71: 1082-8 31. Millard-Stafford ML, Sparling PB, Rosskopf LB, et al. Carbohydrate-electrolyte replacement improves distance running performance in the heat. Med Sci Sports Exerc 1992; 24: 934-40
ª 2011 Adis Data Information BV. All rights reserved.
Vandenbogaerde & Hopkins
32. Wilber RL, Moffatt RJ. Influence of carbohydrate ingestion on blood glucose and performance in runners. Int J Sport Nutr 1992; 2: 317-27 33. Zachwieja JJ, Costill DL, Beard GC, et al. The effects of a carbonated carbohydrate drink on gastric emptying, gastrointestinal distress, and exercise performance. Int J Sport Nutr 1992; 2: 239-50 34. Cole KJ, Grandjean PW, Sobszak RJ, et al. Effect of carbohydrate composition on fluid balance, gastric emptying, and exercise performance. Int J Sport Nutr 1993; 3: 408-17 35. Nishibata I, Sadamoto T, Mutoh Y, et al. Glucose ingestion before and during exercise does not enhance performance of daily repeated endurance exercise. Eur J Appl Physiol 1993; 66: 65-9 36. Tsintzas K, Liu R, Williams C, et al. The effect of carbohydrate ingestion on performance during a 30-km race. Int J Sport Nutr 1993; 3: 127-39 37. Widrick JJ, Costill DL, Fink WJ, et al. Carbohydrate feedings and exercise performance: effect of initial muscle glycogen concentration. J Appl Physiol 1993; 74: 2998-3005 38. Bacharach DW, Von Duvillard SP, Rundell KW, et al. Carbohydrate drinks and cycling performance. J Sports Med Phys Fit 1994; 34: 161-8 39. Chryssanthopoulos C, Hennessy LC, Williams C. The influence of pre-exercise glucose ingestion on endurance running capacity. Br J Sports Med 1994; 28: 105-9 40. Anantaraman R, Carmines AA, Gaesser GA, et al. Effects of carbohydrate supplementation on performance during 1 hour of high-intensity exercise. Int J Sports Med 1995; 16: 461-5 41. Ball TC, Headley SA, Vanderburgh PM, et al. Periodic carbohydrate replacement during 50 min of high-intensity cycling improves subsequent sprint performance. Int J Sport Nutr 1995; 5: 151-8 42. Below PR, Mora-Rodriguez R, Gonzalez-Alonso J, et al. Fluid and carbohydrate ingestion independently improve performance during 1 h of intense exercise. Med Sci Sports Exerc 1995; 27: 200-10 43. El-Sayed MS, Rattu AJM, Roberts I. Effects of carbohydrate feeding before and during prolonged exercise on subsequent maximal exercise performance capacity. Int J Sport Nutr 1995; 5: 215-24 44. Kang J, Robertson RJ, Denys BG, et al. Effect of carbohydrate ingestion subsequent to carbohydrate supercompensation on endurance performance. Int J Sport Nutr 1995; 5: 329-43 45. Tsintzas OK, Williams C, Singh R, et al. Influence of carbohydrate-electrolyte drinks on marathon running performance. Eur J Appl Physiol 1995; 70: 154-60 46. Madsen K, Maclean DA, Kiens B, et al. Effects of glucose, glucose plus branched-chain amino acids, or placebo on bike performance over 100 km. J Appl Physiol 1996; 81: 2644-50 47. Maughan RJ, Bethell LR, Leiper JB. Effects of ingested fluids on exercise capacity and on cardiovascular and metabolic responses to prolonged exercise in man. Exp Physiol 1996; 81: 847-59
Sports Med 2011; 41 (9)
Carbohydrate and Performance
48. McConell G, Kloot K, Hargreaves M. Effect of timing of carbohydrate ingestion on endurance exercise performance. Med Sci Sports Exerc 1996; 28: 1300-4 49. Tsintzas OK, Williams C, Wilson W, et al. Influence of carbohydrate supplementation early in exercise on endurance running capacity. Med Sci Sports Exerc 1996; 28: 1373-9 50. Tsintzas OK, Williams C, Boobis L, et al. Carbohydrate ingestion and single muscle fiber glycogen metabolism during prolonged running in men. J Appl Physiol 1996; 81: 801-9 51. El-Sayed MS, Balmer J, Rattu AJM. Carbohydrate ingestion improves endurance performance during a 1 h simulated cycling time trial. J Sports Sci 1997; 15: 223-30 52. Jeukendrup A, Brouns F, Wagenmakers AJM, et al. Carbohydrate-electrolyte feedings improve 1 h time trial cycling performance. Int J Sports Med 1997; 18: 125-9 53. Millard-Stafford M, Rosskopf LB, Snow TK, et al. Water versus carbohydrate-electrolyte ingestion before and during a 15-km run in the heat. Int J Sport Nutr 1997; 7: 26-38 54. Jeukendrup AE, Thielen JJ, Wagenmakers AJ, et al. Effect of medium-chain triacylglycerol and carbohydrate ingestion during exercise on substrate utilization and subsequent cycling performance. Am J Clin Nutr 1998; 67: 397-404 55. Kovacs EMR, Stegen J, Brouns F. Effect of caffeinated drinks on substrate metabolism, caffeine excretion, and performance. J Appl Physiol 1998; 85: 709-15 56. Palmer GS, Clancy MC, Hawley JA, et al. Carbohydrate ingestion immediately before exercise does not improve 20 km time trial performance in well trained cyclists. Int J Sports Med 1998; 19: 415-8 57. Sugiura K, Kobayashi K. Effect of carbohydrate ingestion on sprint performance following continuous and intermittent exercise. Med Sci Sports Exerc 1998; 30: 1624-30 58. Jarvis AT, Felix SD, Sims S, et al. Carbohydrate supplementation fails to improve the sprint performance of female cyclists. J Exerc Physiol 1999; 1: 1424-40 59. McConell G, Snow RJ, Proietto J, et al. Muscle metabolism during prolonged exercise in humans: influence of carbohydrate availability. J Appl Physiol 1999; 87: 1083-6 60. Angus DJ, Hargreaves M, Dancey J, et al. Effect of carbohydrate or carbohydrate plus medium-chain triglyceride ingestion on cycling time trial performance. J Appl Physiol 2000; 88: 113-9 61. Febbraio MA, Chiu A, Angus DJ, et al. Effects of carbohydrate ingestion before and during exercise on glucose kinetics and performance. J Appl Physiol 2000; 89: 2220-6 62. McConell GK, Canny BJ, Daddo MC, et al. Effect of carbohydrate ingestion on glucose kinetics and muscle metabolism during intense endurance exercise. J Appl Physiol 2000; 89: 1690-8 63. Mitchell JB, Braun WA, Pizza FX, et al. Pre-exercise carbohydrate and fluid ingestion: influence of glycemic response on 10-km treadmill running performance in the heat. J Sports Med Phys Fit 2000; 40: 41-50 64. Bishop NC, Blannin AK, Walsh NP, et al. Carbohydrate beverage ingestion and neutrophil degranulation re-
ª 2011 Adis Data Information BV. All rights reserved.
791
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
sponses following cycling to fatigue at 75% VO2max. Int J Sports Med 2001; 22: 226-31 Chinevere TD, Sawyer RD, Creer AR, et al. Effects of L-tyrosine and carbohydrate ingestion on endurance exercise performance. J Appl Physiol 2002; 93: 1590-7 Chryssanthopoulos C, Williams C, Nowitz A, et al. The effect of a high carbohydrate meal on endurance running capacity. Int J Sport Nutr Exerc Metab 2002; 12: 157-71 Carter J, Jeukendrup AE, Mundel T, et al. Carbohydrate supplementation improves moderate and high-intensity exercise in the heat. Pflugers Arch 2003; 446: 211-9 Jentjens R, Cale C, Gutch C, et al. Effects of pre-exercise ingestion of differing amounts of carbohydrate on subsequent metabolism and cycling performance. Eur J Appl Physiol 2003; 88: 444-52 Desbrow B, Anderson S, Barrett J, et al. Carbohydrateelectrolyte feedings and 1 h time trial cycling performance. Int J Sport Nutr Exerc Metab 2004; 14: 541-9 Nikolopoulos V, Arkinstall MJ, Hawley JA. Reduced neuromuscular activity with carbohydrate ingestion during constant load cycling. Int J Sport Nutr Exerc Metab 2004; 14: 161-70 Saunders MJ, Kane MD, Todd M. Effects of a carbohydrateprotein beverage on cycling endurance and muscle damage. Med Sci Sports Exerc 2004; 36: 1233-8 Burke LM, Wood C, Pyne DB, et al. Effect of carbohydrate intake on half-marathon performance of well-trained runners. Int J Sport Nutr Exerc Metab 2005; 15: 573-89 Van Nieuwenhoven MA, Brouns F, Kovacs EMR. The effect of two sports drinks and water on GI complaints and performance during an 18-km run. Int J Sports Med 2005; 26: 281-5 Gusbakti R. Ingestion of carbohydrate-electrolyte beverage improves exercise performance. Biomed Res 2006; 17: 183-7 Romano-Ely BC, Todd M, Saunders MJ, et al. Effect of an isocaloric carbohydrate-protein-antioxidant drink on cycling performance. Med Sci Sports Exerc 2006; 38: 1608-16 Van Essen M, Gibala MJ. Failure of protein to improve time trial performance when added to a sports drink. Med Sci Sports Exerc 2006; 38: 1476-83 Saunders MJ, Luden ND, Herrick JE. Consumption of an oral carbohydrate-protein gel improves cycling endurance and prevents postexercise muscle damage. J Strength Cond Res 2007; 21: 678-84 Abbiss CR, Peiffer JJ, Peake JM, et al. Effect of carbohydrate ingestion and ambient temperature on muscle fatigue development in endurance-trained male cyclists. J Appl Physiol 2008; 104: 1021-8 Campbell C, Prince D, Braun M, et al. Carbohydratesupplement form and exercise performance. Int J Sport Nutr Exerc Metab 2008; 18: 179-90 Currell K, Jeukendrup AE. Superior endurance performance with ingestion of multiple transportable carbohydrates. Med Sci Sports Exerc 2008; 40: 275-81 Hulston CJ, Jeukendrup AE. Substrate metabolism and exercise performance with caffeine and carbohydrate intake. Med Sci Sports Exerc 2008; 40: 2096-104
Sports Med 2011; 41 (9)
792
82. Jeukendrup AE, Hopkins S, Arago´n-Vargas LF, et al. No effect of carbohydrate feeding on 16 km cycling time trial performance. Eur J Appl Physiol 2008; 104: 831-7 83. Osterberg KL, Zachwieja JJ, Smith JW. Carbohydrate and carbohydrate+protein for cycling time-trial performance. J Sports Sci 2008; 26: 227-33 84. Peake J, Peiffer JJ, Abbiss CR, et al. Carbohydrate gel ingestion and immunoendocrine responses to cycling in temperate and hot conditions. Int J Sport Nutr Exerc Metab 2008; 18: 229-46 85. Valentine RJ, Saunders MJ, Todd MK, et al. Influence of carbohydrate-protein beverage on cycling endurance and indices of muscle disruption. Int J Sport Nutr Exerc Metab 2008; 18: 363-78 86. Breen L, Tipton KD, Jeukendrup AE. No effect of carbohydrate-protein on cycling performance and indices of recovery. Med Sci Sports Exerc 2010; 42: 1140-8 87. Hulston CJ, Jeukendrup AE. No placebo effect from carbohydrate intake during prolonged exercise. Int J Sport Nutr Exerc Metab 2009; 19: 275-84 88. Lacerda ACR, Alecrim P, Damasceno WC, et al. Carbohydrate ingestion during exercise does not delay the onset of fatigue during submaximal cycle exercise. J Strength Cond Res 2009; 23: 1276-81 89. Robson-Ansley P, Barwood M, Eglin C, et al. The effect of carbohydrate ingestion on the interleukin-6 response to a 90-minute run time trial. Int J Sports Physiol Perform 2009; 4: 186-94 90. Saunders MJ, Moore RW, Kies AK, et al. Carbohydrate and protein hydrolysate coingestions improvement of late-exercise time-trial performance. Int J Sport Nutr Exerc Metab 2009; 19: 136-49 91. Hopkins WG, Schabort EJ, Hawley JA. Reliability of power in physical performance tests. Sports Med 2001; 31: 211-34 92. Hinckson EA, Hopkins WG. Reliability of time to exhaustion analyzed with critical-power and log-log modeling. Med Sci Sports Exerc 2005; 37: 696-701 93. Leger L, Mercier D. Gross energy cost of horizontal treadmill and track running. Sports Med 1984; 1: 270-7 94. Billat LV, Koralsztein JP. Significance of the velocity at VO2max and time to exhaustion at this velocity. Sports Med 1996; 22: 90-108 95. Easterbrook PJ, Gopalan R, Berlin JA, et al. Publication bias in clinical research. Lancet 1991; 337: 867-72 96. Hopkins WG, Marshall SW, Batterham AM, et al. Progressive statistics for studies in sports medicine and exercise science. Med Sci Sports Exerc 2009; 41: 3-12 97. Hopkins WG. Competitive performance of elite track-andfield athletes: variability and smallest worthwhile enhancements. Sportscience 2005; 9: 17-20 98. Paton CD, Hopkins WG. Variation in performance of elite cyclists from race to race. Eur J Sport Sci 2006; 6: 25-31 99. Jeukendrup AE, Jentjens R. Oxidation of carbohydrate feedings during prolonged exercise: current thoughts, guidelines and directions for future research. Sports Med 2000; 29: 407-24
ª 2011 Adis Data Information BV. All rights reserved.
Vandenbogaerde & Hopkins
100. Shi X, Horn MK, Osterberg KL, et al. Gastrointestinal discomfort during intermittent high-intensity exercise: effect of carbohydrate-electrolyte beverage. Int J Sport Nutr Exerc Metab 2004; 14: 673-83 101. Massicotte D, Peronnet F, Brisson G, et al. Oxidation of a glucose polymer during exercise: comparison with glucose and fructose. J Appl Physiol 1989; 66: 179-83 102. Rowlands DS, Thorburn MS, Thorp RM, et al. Effect of graded fructose coingestion with maltodextrin on exogenous 14C-fructose and 13C-glucose oxidation efficiency and high-intensity cycling performance. J Appl Physiol 2008; 104: 1709-19 103. O’Brien W, Rowlands D. Fructose-maltodextrin ratio in a carbohydrate-electrolyte solution differentially affects exogenous carbohydrate oxidation rate, gut comfort and performance. Am J Physiol Gastrointest Liver Physiol 2011; 300: G181-9 104. Blomstrand E. A role for branched-chain amino acids in reducing central fatigue. J Nutr 2006; 136: 544S-7S 105. Hunt JN, Stubbs DF. The volume and energy content of meals as determinants of gastric emptying. J Physiol 1975; 245: 209-25 106. Rehrer NJ. The maintenance of fluid balance during exercise. Int J Sports Med 1994; 15: 122-5 107. Rehrer NJ, Beckers EJ, Brouns F, et al. Effects of electrolytes in carbohydrate beverages on gastric emptying and secretion. Med Sci Sports Exerc 1993; 25: 42-51 108. Morris JG, Nevill ME, Thompson D, et al. The influence of a 6.5% carbohydrate-electrolyte solution on performance of prolonged intermittent high-intensity running at 30 degrees C. J Sports Sci 2003; 21: 371-81 109. Jentjens R, Wagenmakers AJM, Jeukendrup AE. Heat stress increases muscle glycogen use but reduces the oxidation of ingested carbohydrates during exercise. J Appl Physiol 2002; 92: 1562-72 110. Burke LM, Collier GR, Davis PG, et al. Muscle glycogen storage after prolonged exercise: effect of the frequency of carbohydrate feedings. Am J Clin Nutr 1996; 64: 115-9 111. Chambers ES, Bridge MW, Jones DA. Carbohydrate sensing in the human mouth: effects on exercise performance and brain activity. J Physiol 2009; 587: 1779-94 112. Wong SH, Chan OW, Chen YJ, et al. Effect of preexercise glycemic-index meal on running when CHO-electrolyte solution is consumed during exercise. Int J Sport Nutr Exerc Metab 2009; 19: 222-42 113. Hopkins WG. Research designs: choosing and fine-tuning a design for your study. Sportscience 2008; 12: 12-21 114. Rowlands DS, Thorp RM, Rossler K, et al. Effect of proteinrich feeding on recovery following intense exercise. Int J Sport Nutr Exerc Metab 2007; 17: 521-43
Correspondence: Professor Will G. Hopkins, Sport Performance Research Institute of NZ, AUT University, Private Bag 92006, Victoria Street West, Auckland 1142, New Zealand. E-mail:
[email protected]
Sports Med 2011; 41 (9)
