Content uploaded by Bent R Rønnestad
Author content
All content in this area was uploaded by Bent R Rønnestad on Oct 24, 2019
Content may be subject to copyright.
1
As accepted for publication in Int J Sports Physiol Perform. 2019 Oct 15:1-7. 1
doi: 10.1123/ijspp.2018-0360. 2
3
Article type: Original Investigation 4
5
6
Increasing Oxygen Uptake in Well-Trained Cross-7
Country Skiers During Work Intervals With a Fast 8
Start 9
10
11
Running head: Optimizing high-intensity intervals 12
13
14
Authors: Bent R. Rønnestad1, Tue Rømer1, and Joar Hansen1 15
1Inland Norway University of Applied Sciences, Lillehammer, Norway 16
17
Abstract Word Count: 250 words 18
Text-Only Word Count: 4593 words 19
Number of figures: 2 20
Number of tables: 1 21
22
Corresponding author: 23
Bent R. Rønnestad 24
Inland Norway University of Applied Sciences, Lillehammer, Norway 25
E-mail: bent.ronnestad@inn.no 26
Phone: +47 61288193 27
Fax: +47 61288200 28
29
30
2
Abstract 31
Purpose: Accumulated time at a high percentage of peak oxygen consumption (VO2peak) is 32
important for improving performance in endurance athletes. The present study compared the 33
acute effect of a roller-ski skating session containing work intervals with a fast start followed 34
by decreasing speed (DEC) with a traditional session where the work intervals had a constant 35
speed (similar to the mean speed of DEC; TRAD) on physiological responses, rate of 36
perceived exertion (RPE) and leg press peak power (LPPP). Methods: Eleven well-trained 37
cross-country skiers performed DEC and TRAD in a randomized order (5x5-min work 38
intervals, 3 min relief). Each 5-min work interval in the DEC protocol started with 1.5 min at 39
100% of maximal aerobic speed (MAS) followed by 3.5 min at 85% of MAS, while the 40
TRAD protocol had a constant speed at 90% of MAS. Results: DEC induced a higher VO2 41
than TRAD, measured as both peak and average of all work intervals during the session (98.2 42
± 2.1 vs. 95.4 ± 3.1% VO2peak, respectively and 87.6 ± 1.9 vs. 86.1 ± 3.2% VO2peak, 43
respectively) with a lower mean RPE after DEC compared to TRAD (16.1 ± 1.0 vs. 16.5 ± 44
0.7, respectively) (all p<0.05). There were no differences between sessions for mean heart 45
rate, blood lactate concentration or LPPP. Conclusion: DEC induced a higher mean VO2 and 46
a lower RPE compared with TRAD, despite similar mean speed, indicating that DEC can be a 47
good strategy for interval sessions aiming to accumulate more time at a high percentage of 48
VO2peak. 49
50
KEY WORDS 51
Endurance training, High-intensity aerobic training, Intense exercise, Roller-skiing 52
53
54
55
3
INTRODUCTION 56
Performance in cross-country (XC) skiing is highly related to maximal oxygen consumption 57
(VO2max).1,2 The high VO2max values in XC skiers could be related to numerous factors like 58
genetics, training volume, training periodization and amount of high-intensity aerobic interval 59
training (HIT). To the best of our knowledge, there is only one previous study focusing on 60
optimizing a single HIT session for well-trained XC skiers. This study observed that longer 61
intervals (5-10 min) were more effective than shorter intervals (2-4 min) in improving 62
endurance performance after 8 weeks.3 The importance of optimizing the HIT session for XC 63
skiers is emphasized in a recent review paper where Sandbakk and Holmberg4 suggested that 64
well-trained XC skiers could benefit from increasing the quality of HIT sessions. In this 65
context, the quality of a HIT session can be defined by mean VO2 or accumulated training 66
time ≥ 90% VO2max5-7 possibly due to the large stimulus for myocardial morphological 67
adaptations that increases maximal cardiac stroke volume and also more peripheral skeletal 68
muscle adaptations.6 69
70
In order to optimize exercise time ≥ 90% VO2max, a speed between 90 to 100% of maximal 71
aerobic speed (MAS) is recommended.8 However, continuous work at MAS can only be 72
sustained for ~6 minutes in well-trained runners and cyclists.9,10 Therefore, there is a quest for 73
developing HIT sessions that optimize time ≥ 90% VO2max by balancing work and recovery 74
durations and intensities.11 It has been recognized that a fast-start strategy might speed the 75
VO2 kinetics during both running and kayaking.12,13 Furthermore, by using cycling in 76
untrained to moderately-trained participants, it has been observed that fast start intervals 77
acutely increase VO2.14-16 To the best of our knowledge, the VO2 in roller-ski skating 78
intervals, involving four active limbs with a large muscle mass, have not been investigated. 79
Given the differences in VO2 kinetics during exercise with a large muscle mass (i.e., running) 80
compared with a smaller muscle mass (i.e., cycling),17,18 it would be of interest to investigate 81
the effects of a fast start strategy while roller-ski skating on time ≥ 90% VO2max. Furthermore, 82
acute physiological effects of a fast start with a subsequent speed reduction during work 83
intervals in a HIT session has not yet been investigated in well-trained endurance athletes. 84
85
Therefore, the primary purpose of the present study was to compare the acute effects of a HIT 86
session containing work intervals with a fast start and subsequent reduction in speed with a 87
4
traditional HIT session with similar mean speed during the work intervals but performed at a 88
constant speed on physiological responses during roller-ski skating in well-trained XC skiers. 89
90
METHODS 91
Subjects 92
Eleven well-trained19 male skiers (age 23.5±3.5 years, height 184±6 cm, body mass 77.9±7.2 93
kg, peak oxygen consumption (VO2peak) 70.6± 5.7 mLmin-1kg-1) competing in XC skiing or 94
biathlon volunteered for the study, which was performed according to the ethical standards 95
established by the Helsinki Declaration of 1975 and approved by the local ethical committee 96
of the Department of Sports Science, Lillehammer University College. All participants 97
provided informed consent. The self-reported amount of endurance training hours during the 98
year preceding the experiment was 579±85 h. These hours were categorized into a three-zone 99
model,20 of which 89±3%, 6±2% and 5±1% was performed in heart rate (HR) zone 1 (60%-100
82% of HRmax), zone 2 (83%-87% of HRmax) and zone 3 (88%-100% of HRmax). The mean 101
weekly endurance training time during the three months preceding the start of the experiment 102
was 14.0 ± 1.8 h with an intensity distribution similar to the mean values of the entire year. 103
The experiment was performed in the last half of the skiers’ preparatory period (September). 104
105
Experimental design 106
The participants visited the test laboratory on five separate occasions and roller-ski skating 107
was the exercise mode performed throughout. The first test session was a preliminary test to 108
determine MAS, VO2peak and HRpeak.On the subsequent four visits, two different 5x5-min HIT 109
sessions were performed twice. One HIT session consisted of work intervals with a fast start 110
followed by decreasing speed (DEC) and the other HIT session employed work-matched (± 111
0.1 km·h-1) intervals with a constant speed (TRAD). After performing DEC and TRAD once 112
each in a randomized order, the two sessions were both repeated once more, again in a 113
randomized order. The HIT session with the highest mean VO2 from the two repeated days of 114
testing (within condition) was used in statistical analyses. The recovery demand of 115
neuromuscular function was assessed by measuring peak power output during a seated leg 116
press before and after the HIT sessions. 117
118
Preliminary testing 119
5
Preliminary roller-ski testing and all HIT sessions (including the warm-up) were performed 120
while skating on a treadmill (Lode Valiant Special, Lode B.V. Groningen, the Netherlands) 121
using the same roller-skis (IDT Skate Elite RM2, IDT Solutions AS, Lena, Norway), the same 122
poles (SWIX Triac 2.0, Swix Sport AS, Lillehammer, Norway, length between 160 and 177.5 123
cm), and under similar environmental conditions (17–20 °C) with a fan ensuring circulating 124
air. The athletes used their own ski boots and could freely choose between the two 125
predominant uphill skating techniques (V1, where the skiers use their poles on every second 126
leg push-off and V2, where the poles are used on every leg push-off). 127
128
After a 10-min warm-up on the treadmill at an inclination of 3% and a velocity of 12 km·h-1 129
(68 ± 5% of HRpeak) the test started, using a constant inclination of 9%. The first 5-min bout 130
started at a velocity of 7 km·h-1 and increased by 1 km·h-1 for each 5-min bout until a blood 131
lactate concentration ([La−]) above 4.0 mmol·L-1 was measured (4.5 ± 1 bouts). Capillary 132
blood samples were taken from a fingertip during a 1-min break in between each 5-min bout, 133
and analyzed for whole blood [La–] (Biosen C-line, EKF Diagnostics, Barleben, Germany). 134
The average VO2 from the two last minutes of each 5-min bout was used for a subsequent 135
calculation of MAS based on the relationship between VO2 and workload. VO2 was measured 136
with a sampling time of 30 s, using a computerized metabolic system with mixing chamber 137
(Oxycon Pro, Erich Jaeger, Hoechberg, Germany). The flow turbine (Triple V, Erich Jeger) 138
was calibrated with a 3L, 5530 series, calibration syringe (Hans Rudolph, Kansas City, 139
Missouri, USA). The same metabolic system with identical calibration routines was used 140
during the subsequent HIT sessions. The speed at 4.0 mmol·L-1 [La–] was calculated from the 141
relationship between [La–] and speed using linear regression between the closest workload 142
below and above 4.0 mmol·L-1 from the preliminary testing. 143
144
Ten minutes after the submaximal test an incremental test was performed, starting at a 9% 145
inclination and a speed of 7 km·h-1 which increased by 1 km·h-1 every minute until exhaustion 146
(average duration of 9.1 ± 0.8 minutes). VO2peak was calculated as the average of the two 147
highest 30-s measurements and HRpeak was the highest 1-s value recorded (Polar RCX5, 148
Polar, Kempele, Finland). MAS was defined as the speed where the horizontal line 149
representing VO2peak met the extrapolated linear regression representing the submaximal 150
VO2/speed relationship. Work rates at 4.0 mmol·L-1 [La−] and MAS were calculated as the 151
sum of the power against gravity and the power against rolling friction (friction coefficient = 152
0.0237) at the respective velocities as described previously.21 153
6
154
HIT sessions 155
The HIT sessions were performed at a fixed time of day (± 1 h) interspersed with 3-5 days 156
containing only easy training. The participants were instructed to consume the same meal 157
before each visit to the lab and were not allowed to eat during the hour preceding a session or 158
to consume coffee or other products containing caffeine during the three hours preceding the 159
tests. The roller-ski warm-up and the subsequent HIT sessions were performed on a constant 160
treadmill inclination of 9%. The first 5 min of the warm-up was performed at a speed of 7 161
km·h-1, followed by a 5-min gradual increase in speed, based on individual preferences, 162
towards the speed at 4.0 mmol·L-1 [La−]. The speed at 4.0 mmol·L-1 [La−] was maintained for 163
4 min, before a new increase in speed towards the starting velocity of the first work interval 164
was reached (after ~3 s) and maintained for 1.5 min. A 5-min active recovery period (3% 165
inclination and 10 km·h-1) concluded the warm-up. This means that there was a difference in 166
the warm-up protocol, where DEC involved exercise at 12.6±0.9 km·h-1 and TRAD involved 167
exercise at 11.3±0.8 km·h-1 for the final 1.5 min. The rationale for this difference is that 168
anecdotally there is a common practice amongst XC-skiers to include efforts of the start pace 169
of their work intervals during the warm-up, and that it has been recommended to perform at 170
least one race-pace effort during the warm-up for middle-distance runners22 (which is 171
comparable with the present exercise setting). Furthermore, the 5-min active recovery period 172
before starting the first work interval is likely to minimize any potential physiological 173
difference induced by the intensity difference in the warm-up. However, to verify that the 174
different warm-up procedures did not affect data collection in the first 5-min work interval, 175
seven XC skiers performed the present warm-up protocol in two different sessions within one 176
week. The only difference was 1.5 min at either 12.6 km·h-1 or 11.3 km·h-1 followed by the 5-177
min active recovery before they performed the first 5-min TRAD work interval at a velocity 178
of 11.3 km·h-1. The order was randomized and there were no differences between using 12.6 179
and 11.3 km·h-1 during the warm-up protocol on mean VO2 during the first 1.5 min 180
(3414±120 vs. 3417±237 mL·min-1, respectively, p=0.98) or mean VO2 during the entire 5-181
min work interval (4129±212 vs. 4127±197 mL·min-1, respectively, p=0.96). 182
183
The athletes could freely choose between V1 and V2 skating techniques during all sessions, as 184
there appears to be no difference in work economy or performance at these speeds and 185
inclines.23 Anecdotally, the individual skier was quite consistent in the choice of skating 186
technique from session to session, but there were individual differences between the skiers. 187
7
Between the 5-min work intervals there was a 3-min relief period where the two first minutes 188
were passive and the last minute was performed at 7 km·h-1 and an incline of 9%. The speed 189
was increased to the starting velocity over the final 5 s of each rest period. Each 5-min work 190
interval in the DEC protocol started with 1.5 min at 100% of MAS followed by 3.5 min at 191
85% of MAS. This protocol was based on a previous study indicating that exercise for 1-1.5 192
min at MAS is needed before cyclists reach 90-95% of VO2max.14 It was therefore anticipated 193
that this exercise intensity and duration would speed the VO2 kinetics during the DEC session 194
without resulting in too much fatigue. Furthermore, it has also been indicated that this high 195
VO2 can be maintained for ~16 min even after reducing exercise intensity to ~80% of MAS.14 196
Each work interval in the TRAD protocol was performed with a constant speed at 90% of 197
MAS, resulting in a similar mean speed as in DEC (11.34 and 11.28 km·h-1). Continuous 198
running to exhaustion and intermittent running, both at an exercise intensity of 92% of MAS, 199
have previously elicited a substantial time spent at 90% of VO2max24, which was the rationale 200
for selecting a mean intensity of 90% of MAS for DEC and TRAD in the current study. Other 201
reasons for this choice were that, anecdotally, XC-skiers do not usually perform HIT sessions 202
to exhaustion and it was important that all participants were able to complete the HIT 203
sessions. 204
205
VO2 and HR during the work intervals were recorded at 15-s intervals. The highest 15-s 206
measurement across all intervals was used as maximum VO2 and HR during each HIT 207
session. Time ≥ 90% VO2peak was calculated as the sum of VO2 values (averaged over 15 s) 208
that were superior or equal to 90% of the reference value for VO2peak obtained from the 209
incremental exercise test to exhaustion. The same procedure was used for determination of 210
time ≥ 90% HRpeak. Immediately after each work interval a capillary blood sample was 211
collected from the fingertip and analyzed for [La-] (Biosen C-line, EKF Diagnostics, 212
Barleben, Germany) and rate of perceived exertion (RPE) was recorded using Borg’s 6-20 213
scale.25 Mean VO2 during each HIT session was calculated as the mean value across all work 214
intervals. To evaluate the development of VO2 in the work intervals, the mean values of all 215
five intervals were used. This was calculated as the mean percentage of VO2peak during the 216
first two minutes, the third minute and the last two minutes. 217
218
Leg press peak power 219
Lower-body peak power was measured before and after the HIT sessions using a pneumatic 220
bilateral seated leg press machine (Keiser A420, Keiser Sports Health Equipment Inc, Fresno, 221
8
California, USA). This equipment has been described previously and has been found to have a 222
high level of reproducibility.26 Testing before the HIT session was performed after a 10-min 223
cycling warm-up at a RPE of 11-12 on the 6-20 Borg scale and was repeated 5 min after the 224
last work interval (without any physical activity during this 5-min period). The participants 225
sat with knees flexed at 90° to 96° and the hips flexed at approximately 45° with the 226
individual seating position being identical for all tests. Following the 10-min cycling warm-227
up, two warm-up repetitions were performed at 44 kg. The power testing consisted of a single 228
trial of ten increasingly heavy lifts with standardized recovery periods (gradually increasing 229
from 5 to 45 s) and loads starting at 44 kg and ending at 279 kg. During all lifts the 230
participants were instructed to exert force “as fast as possible”. Average concentric 231
mechanical power of each lift was calculated in the manufacturer’s software and based on 232
these calculations peak power output was calculated and used for statistical analysis. Two 233
warm-up repetitions at 44 kg were performed for the testing after the HIT sessions. 234
Familiarization to this test procedure was provided both before and after the preliminary 235
testing on test day 1. 236
237
Statistics 238
All values presented in the text and tables are mean±SD. Two-way repeated measures 239
ANOVA with Bonferroni post hoc tests were performed to evaluate differences between DEC 240
and TRAD in the mean percentage of VO2peak during the first two minutes, the third minute 241
and during the two last minutes of all work intervals (time vs. condition). Student`s two-tailed 242
paired t-tests were used to evaluate potential differences between DEC and TRAD in RPE and 243
physiological responses. The initial statistical significance level was set to p ≤ 0.05, however, 244
Holm-Bonferroni sequential adjustments were applied when evaluating physiological 245
responses due to multiple comparisons.27 ANOVA analyses were performed in GraphPad 246
Prism 7 (GraphPad Software Inc., CA, USA) and Student’s t-tests were performed in Excel 247
2010 (Microsoft Corporation, Redmond, WA, USA). Effect size (ES) of DEC was calculated 248
when there were significant differences between DEC and TRAD with the following formula: 249
([DEC mean – TRAD mean]/ SD). The scale proposed by Rhea28 for well-trained subjects 250
was used to interpret the magnitude of the treatment effect; 0.0-0.24 trivial, 0.25-0.49 small, 251
0.5-1.0 moderate, >1.0 large. 252
253
254
9
Results 255
The submaximal test on the first test day showed that the mean velocity at 4 mmol·L-1 [La−] 256
was 10.1±0.8 km·h-1, equal to an external workload of 250±32 W, while the calculated MAS 257
was 12.6±0.9 km·h-1, equal to an external workload of 311±30 W. Responses during DEC and 258
TRAD are shown in Table 1 and individual data for maximum and mean VO2 across all work 259
intervals are presented in Figure 1. Maximum VO2 was higher during DEC compared with 260
TRAD (p=0.0003, Table 1), with a moderate ES. DEC also induced a higher mean VO2 than 261
TRAD (p=0.0293) and again ES was moderate. As shown in Figure 2, mean VO2, expressed 262
as percentage of VO2peak, across all work intervals was higher during the two first minutes of 263
DEC compared to TRAD (83.6±4.4% vs. 79.6±5.1%, respectively, p<0.0001; ES=0.78). 264
There was no difference between DEC and TRAD during the third minute (87.7±4.7% vs. 265
87.6±5.5%, respectively), while TRAD was higher than DEC during the last two minutes 266
(88.8±5.3% vs. 86.6±4.6%, respectively, p=0.0015; ES=0.44). Both the maximum and mean 267
RPE for all work intervals were lower in DEC than TRAD (p=0.036 and p=0.045, 268
respectively; Table 1) and ES values were small ES. There was no statistical difference 269
between DEC and TRAD in time ≥ 90% of VO2peak, time ≥ 90% of HRpeak, maximum HR 270
measurement, maximum or mean La- (Table 1). Furthermore, there were no differences 271
between DEC and TRAD in changes in leg press peak power from before (773 ± 104 and 776 272
± 112 W, respectively) to after (788 ± 107 and 801 ± 113 W, respectively) the HIT sessions. 273
274
(Insert Table 1 approximately here) 275
(Insert Figure 1 approximately here) 276
(Insert Figure 2 approximately here) 277
278
Discussion 279
The main finding of the present study was that DEC induced higher maximum and average 280
VO values, and a lower RPE, than TRAD. In addition, the ES analyses showed a moderate 281
practical effect of DEC on VO2 and a small practical effect on RPE, despite having similar 282
mean speed during all work intervals as TRAD. 283
284
In DEC the work intervals commenced with 1.5 min at a higher speed than TRAD and this 285
resulted in a larger mean VO2 during the two first minutes (measured as mean values for all 286
10
work intervals). This is in agreement with the observation that starting a 2-5-min exercise 287
with a higher power output increases the overall VO2 compared with a more even intensity 288
distribution.13,29 During the third minute of all work intervals there was no difference between 289
DEC and TRAD in mean VO2, despite a higher speed during TRAD. This higher speed during 290
TRAD eventually led to a higher mean VO2 during the last two minutes of all work intervals. 291
Interestingly, the ES showed that DEC had a moderate practical effect on VO2 during the first 292
two minutes of the work intervals, while TRAD only had a small practical effect during the 293
two last minutes. To the best of our knowledge, this is the first study to demonstrate this 294
potential advantage of DEC vs. TRAD by using similar mean speed and exercise duration in 295
roller-ski skating intervals. In agreement with the present findings, previous studies utilizing 296
only lower-body exercise through cycling have observed longer times spent at a VO2 >85% of 297
VO2max when starting the exercise with a high power followed by reduced power in untrained 298
to moderately-trained participants.14-16 However, in two of these studies the sessions differed 299
in duration and mean exercise intensity, making the findings somewhat challenging to 300
interpret.14,15 Since the present study utilized a similar mean speed for both DEC and TRAD, 301
it can be suggested that starting the work interval with a higher speed, followed by a 302
subsequent reduction in speed, may lead to a higher mean VO2 and thus a better stimulus on 303
central and peripheral factors involved in VO2 than using similar mean intensity in a flat 304
distribution. In agreement with our findings, Zadow et al.16 used trained cyclists and observed 305
longer time above 85% of VO2max when using a 15-s all-out strategy in the beginning of 3-306
min work intervals compared to a more even distribution of the power output. 307
308
The larger VO2 response in DEC might be related to the quicker rise in VO2 with a fast start 309
compared to a slower start.30 The absolute rate at which VO2 rises at the onset of exercise is a 310
positive function of the difference between the current and the required steady-state VO2 in 311
the working muscle.31 Therefore, the larger difference between current and required VO2 in 312
the working muscles during the beginning of the work intervals in DEC is likely to speed the 313
VO2 response. Additionally, in later intervals slow VO2 kinetics of the initially recruited type 314
II fibers, reduced contractile efficiency, and a gradual increase in O2 demand from the 315
recovery processes in fatigued fibers32 might contribute to explain the higher mean VO2 in 316
DEC than TRAD. The present study may be limited by the choice of increasing the ecological 317
validity by finalizing the warm-up with 1.5-min at the starting velocity of the work interval, 318
which induced a difference in the lead-in to the first work interval. This higher exercise 319
11
intensity in DEC may have sped up the VO2 kinetic response compared to TRAD. However, 320
the 5-min active recovery period before starting the first work interval can be suggested to 321
minimize any potential physiological difference induced by the intensity difference in the 322
warm-up. This is supported by the subsequent preliminary study showing no effect of the two 323
warm-up procedures on mean VO2 during the first 1.5-min or mean VO2 during the 5-min 324
work interval. In addition, any potential difference in physiological response due to the 325
difference in warm-up should mainly be localized to the beginning of the first interval, while 326
the present statistics are based on mean values from all 5 work intervals. 327
328
The time ≥ 90% VO2peak in DEC and TRAD was 12.0 min and 10.8 min, respectively. To the 329
best of our knowledge no previous study has used a similar protocol focusing on time ≥ 90% 330
VO2peak. The study with the closest approach was performed on sub-elite runners who 331
performed ~5 intervals with a duration of ~5 min with ~3 min of recovery in between.24 332
Interestingly, their values were in the same range as our TRAD group in time ≥ 90% VO2peak 333
(10.5 vs. 10.8 min), indicating similar responses in the cardiovascular system in the exercise 334
mode of running and skating on roller-skis.In the present study, there was no statistical 335
difference between DEC and TRAD on time ≥ 90% VO2peak. However, the mean value 336
showed an advantage of ~70 s more time ≥ 90% VO2peak for DEC compared to TRAD. The 337
smallest worthwhile enhancement in performance time for elite skiers is 0.4%33, and therefore 338
it can be hypothesized that in elite sport, this could be a relevant advantage in training 339
stimulus. Indeed, recreationally-trained cyclists that spent ~100 s more time above ~90% 340
VO2max per training session achieved the largest improvement in VO2max and power output at 341
the lactate threshold.7 The explanation for a higher mean VO2 in DEC vs. TRAD with no 342
significant difference between sessions in time ≥ 90% VO2peak remains speculative. It could be 343
that the reduction in speed to 85% of MAS during the last 3.5 min of each DEC work interval 344
was too low to support a substantial time ≥ 90% VO2peak, but sufficient to give a higher mean 345
VO2 than the TRAD session. Individual differences in fractional utilization of VO2peak at 346
lactate threshold could in theory contribute to the large variation in time ≥ 90% VO2peak 347
despite all participants exercising at the same percentage of MAS.34,35 Indeed, a previous 348
study has also observed large individual variations in time ≥ 90% VO2peak even though a 349
similar percentage of MAS was used.10 In order to optimize the DEC session to induce the 350
longest possible time ≥ 90% VO2peak, the individual athlete should probably test different 351
speeds in the different phases of the work interval. Furthermore, maybe future studies should 352
12
consider taking into account individual differences in fractional utilization of VO2peak at 353
lactate threshold. 354
355
Both the peak and mean RPE was lower in DEC than TRAD, indicating that the athletes 356
perceived the DEC protocol to be less demanding than the TRAD protocol, despite similar 357
mean speed and a higher VO2 during the work intervals. This was somewhat unexpected, 358
since RPE during interval exercise is typically associated with HR and [La-]36, and there were 359
no differences between DEC and TRAD in these variables. A reduction in RPE by using a fast 360
start strategy has been observed previously12, but contrasting findings exists.16 Billat et al.14 361
suggested that RPE is directly related to the power output at any instant of the exercise. 362
Therefore, it can be suggested that the assumed higher perceived effort induced by the higher 363
speed during the first 1.5 min in DEC is counteracted by the lower speed during the last 3.5 364
min, leading to a lower RPE overall. The reduced RPE can be linked to previous observations 365
of improved exercise tolerance30 and performance13,29,37 with a fast start strategy compared to 366
a more even distribution strategy. Consequently, we can hypothesize that the higher VO2 in 367
the first 2 min saves the limited anaerobic capacity and therefore the athletes are further away 368
from exhaustion when finishing the work intervals and thus experience a lower exertion. 369
Furthermore, changing the exercise intensity and breaking up a monotonous exercise can 370
increase the rate of perceived enjoyment38 and may thus also contribute to explain the lower 371
RPE in DEC vs. TRAD. The lower RPE after DEC is likely to be an important practical 372
finding since we know that well-trained trained XC skiers perform a high training volume 373
(usually from 12 to 25 weekly training hours) including ~ two weekly HIT sessions.2,39 When 374
performing such a high training volume year after year with regular HIT sessions, it could be 375
hypothesized that a lower RPE of the DEC protocol may be beneficial in terms of mental 376
wellbeing, and avoiding overtraining.40 Importantly, the latter has yet to be investigated in 377
longitudinal studies. 378
379
It has been shown that a HIT session can acutely impair neuromuscular function.41 Power 380
output in the lower-body has been found to be a valid tool to measure both fatigue and 381
recovery of neuromuscular function following HIT sessions.42 A decline in leg press peak 382
power was expected from before to after the HIT session, but this was not observed in any of 383
the sessions. It could be speculated that the warm-up to the leg press before the HIT session 384
13
was insufficient and thus limited the leg press performance. However, the similar warm-up in 385
DEC and TRAD and no differences between them in changes in leg press peak power from 386
before to after the HIT session indicates no differences between sessions in recovery demand 387
of the contractile function in the lower-body muscles. It also indicates that if earlier 388
recruitment and fatigue of type II fibers contributes to the increased mean VO2 in DEC, then 389
the type II fibers involved in leg press peak power had recovered during the 5-min break 390
between the HIT session and the peak power test. The present study did not perform any 391
measurement of upper-body recovery and did not investigate other important factors that need 392
recovery, like the glycogen stores. 393
394
Practical application 395
The DEC strategy may be a good alternative to TRAD if the main aim of the HIT session is to 396
accumulate more time at a high percentage of VO2peak. However, whether this culminates into 397
superior long-term adaptations needs to be investigated in longitudinal studies. Importantly, a 398
lower RPE was noted after DEC compared to TRAD, which can be important to consider in 399
terms of mental wellbeing. In terms of recovery of the contractile function in the lower body, 400
there seems to be no difference between DEC and TRAD. However, the long-term demands 401
of recovery are uncertain, so it is important to carefully monitor the training response and 402
recovery needs. The present DEC protocol utilized a similar approach for all athletes where 403
all work intervals started with 1.5 min at 100% of MAS followed by 3.5 min at 85% of MAS. 404
This approach did not take into account differences in fractional utilization of VO2peak at 405
lactate threshold and was likely not optimal for all athletes. Therefore, if a coach or athlete 406
plans to test out this principle of organizing the HIT session, the exact exercise intensities and 407
durations during the DEC work interval should be individualized in order to achieve the 408
desired stimulus. 409
410
Conclusion 411
DEC induced higher maximum and average VO2 values during the session, as well as lower 412
RPE, than TRAD. This was supported by the ES showing a moderate practical effect of DEC 413
on both maximum and average VO2 and a small practical effect on the RPE. Finally, there 414
14
was no difference between DEC and TRAD in leg press peak power output after the HIT 415
sessions. 416
417
ACKNOWLEDGMENTS 418
We thank Martin Winger, Matias Gjestvang Brandt and Krister Flobergseter for great 419
technical help during data sampling. No funding was obtained for this study. The authors have 420
no professional relationships with companies or manufacturers who will benefit from the 421
results of this study. 422
423
424
425
REFERENCES 426
1. Holmberg H-C. The elite cross-country skier provides unique insights into human 427
exercise physiology. Scand J Med Sci Sport. 2015;25:100–109. 428
429
2. Sandbakk Ø, Hegge AM, Losnegard T, Skattebo Ø, Tønnessen E, Holmberg H-C. The 430
physiological capacity of the world’s highest ranked female cross-country skiers. 431
Med Sci Sports Exerc. 2016;48:1091–1100. 432
433
3. Sandbakk Ø, Sandbakk SB, Ettema G, Welde B. Effects of intensity and duration in 434
aerobic high-intensity interval training in highly trained junior cross-country skiers. J 435
Strength Cond Res. 2013;27:1974-1980. 436
437
4. Sandbakk Ø and Holmberg H-C. Physiological capacity and training routines of elite 438
cross-country skiers: approaching the upper limits of human endurance. Int J Sports 439
Physiol Perform. 2017;12:1003-1011. 440
441
5. Thevenet D, Tardieu-Berger M, Berthoin S, and Prioux J. Influence of recovery mode 442
(passive vs. active) on time spent at maximal oxygen uptake during an intermittent 443
session in young and endurance-trained athletes. Eur J Appl Physiol. 2007;99:133-444
142. 445
446
6. Midgley AW, McNaughton LR, Wilkinson M. Is there an optimal training intensity 447
for enhancing the maximal oxygen uptake of distance runners?: empirical research 448
findings, current opinions, physiological rationale and practical recommendations. 449
Sports Med. 2006;36:117–132. 450
451
7. Turnes T, de Aguiar RA, Cruz RS, Caputo F. Interval training in the boundaries of 452
severe domain: effects on aerobic parameters. Eur J Appl Physiol. 2016;116:161-169. 453
454
8. Hill DW, Williams CS, Burt SE. Responses to exercise at 92% and 100% of the 455
velocity associated with VO2max. J Sport Sci Med. 1997;18:325–329. 456
457
15
9. Billat V, Bernard O, Pinoteau J, Petit B, Koralsztein JP Time to exhaustion at VO2max 458
and lactate steady state velocity in sub elite long-distance runners. Arch Int Physiol 459
Biochim Biophys. 1994;102(3):215-219. 460
461
10. Rønnestad BR, Hansen J. Optimizing interval training at power output associated with 462
peak oxygen uptake in well-trained cyclists. J Strength Cond Res. 2016;30(4):999-463
1006. 464
465
11. Buchheit M, Laursen PB. High-intensity interval training, solutions to the 466
programming puzzle: Part I: cardiopulmonary emphasis. Sports Med. 2013;43:313-467
338. 468
469
12. Ariyoshi M, Tanaka H, Kanamori K, Obara S, Yoshitake H, Yamaji K, Shephard RJ. 470
Influence of running pace upon performance: effects upon oxygen intake, blood 471
lactate, and rating of perceived exertion. Can J Appl Sport Sci. 1979;4:210–213. 472
473
13. Bishop D, Bonetti D, Dawson B. The influence of pacing strategy on VO2 and 474
supramaximal kayak performance. Med Sci Sports Exerc. 2002;34:1041–1047. 475
476
14. Billat V, Petot H, Karp JR, Sarre G, Morton RH, Mille-Hamard L. The sustainability 477
of VO2max: effect of decreasing the workload. Eur J Appl Physiol. 2013;113:385-394. 478
479
15. Lisboa FD, Salvador AF, Raimundo JA, Pereira KL, de Aguiar RA, Caputo F. 480
Decreasing power output increases aerobic contribution during low-volume severe-481
intensity intermittent exercise. J Strength Cond Res. 2015;29:2434-2440. 482
483
16. Zadow EK, Gordon N, Abbiss CR, Peiffer JJ. Pacing, the missing piece of the puzzle 484
to high-intensity interval training. Int J Sports Med. 2015;36:215-219. 485
486
17. Caputo F and Denadai BS. Exercise mode affects the time to achieve VO2max without 487
influencing maximal exercise time at the intensity associated with VO2max in 488
triathletes. Int J Sports Med. 2006; 27:798-803. 489
490
18. Carter H, Jones AM, Barstow TJ, Burnley M, Williams CA, and Doust JH. Oxygen 491
uptake kinetics in treadmill running and cycle ergometry: a comparison. J Appl 492
Physiol. 2000; 89:899-907. 493
494
19. De Pauw K, Roelands B, Cheung SS, de Geus B, Rietjens G, Meeusen R. Guidelines 495
to classify subject groups in sport-science research. Int J Sports Physiol Perform. 496
2013;8(2):111-122. 497
498
20. Sylta Ø, Tønnessen E, Seiler S. From heart-rate data to training quantification: a 499
comparison of 3 methods of training-intensity analysis. Int J Sports Physiol Perform. 500
2014;9(1):100-107. 501
502
21. Sandbakk Ø, Holmberg H-C, Leirdahl S, Ettema, G. Metabolic rate and gross 503
efficiency at high work rates in world class and national level sprint skiers. Eur J Appl 504
Physiol. 2010;109:473-481. 505
506
16
22. McGowan CJ, Pyne DB, Thompson KG, Rattray B. Warm-up strategies for sport 507
and exercise: mechanisms and applications. Sports Med. 2015;45(11):1523-1546. 508
509
23. Losnegard T, Myklebust H, Hallén J. Anaerobic capacity as a determinant of 510
performance in sprint skiing. Med Sci Sports Exerc. 2012;44:673-681. 511
512
24. Demarie S, Koralsztein JP, Billat V. Time limit and time at VO2max during a 513
continuous and an intermittent run. J Sports Med Phys Fitness. 2000;40(2):96-102. 514
515
25. Borg GA. Psychophysical bases of perceived exertion. Med Sci Sports Exerc. 516
2008;14:377-381. 517
518
26. Earles DR, Judge JO, Gunnarsson OT. Velocity training induces power-specific 519
adaptations in highly functioning older adults. Arch Phys Med Rehabil. 2000;82:872-520
878. 521
522
27. Holm S. A simple sequentially rejective multiple test procedure. Scand J Stat. 523
1979;6:65-70. 524
525
28. Rhea MR. Determining the magnitude of treatment effects in strength training 526
research through the use of the effect size. J Strength Cond Res. 2004;18:918–920. 527
528
29. Aisbett B, Lerossignol P, McConell GK, Abbiss CR, Snow R. Influence of all-out and 529
fast start on 5-min cycling time trial performance. Med Sci Sports Exerc. 530
2009;41:1965-1971. 531
532
30. Jones AM, Wilkerson DP, Vanhatalo A, Burnley M. Influence of pacing strategy on 533
O2 uptake and exercise tolerance. Scand J Med Sci Sports. 2008;18:615-626. 534
535
31. Margaria R, Mangili F, Cuttica F, Cerretelli P. The kinetics of the oxygen 536
consumption at the onset of muscular exercise in man. Ergonomics. 1965;8:49–54. 537
538
32. Vanhatalo A, Poole DC, DiMenna FJ, Bailey SJ, Jones AM. Muscle fiber recruitment 539
and the slow component of O2 uptake: constant work rate vs. all-out sprint exercise. 540
Am J Physiol Regul Integr Comp Physiol. 2011;300:R700-R707. 541
542
33. Spencer M, Losnegard T, Hallén J, Hopkins WG. Variability and predictability of 543
performance times of elite cross-country skiers. Int J Sports Physiol Perform. 544
2014;9(1):5-11. 545
546
34. Coyle EF, Coggan AR, Hopper MK, Walters TJ. Determinants of endurance in well-547
trained cyclists. J Appl Physiol (1985). 1988;64(6):2622-2630. 548
549
35. Coyle EF. Integration of the physiological factors determining endurance performance 550
ability. Exerc Sport Sci Rev. 1995;23:25-63. 551
552
36. Oliveira BRR, Slama FA, Deslandes AC, Furtado ES, Santos TM. Continuous and 553
high-intensity interval training: which promotes higher pleasure? PLoS One. 2013; 554
8:e79965. 555
556
17
37. Bailey SJ, Vanhatalo A, DiMenna FJ, Wilkerson DP, Jones AM. Fast-start strategy 557
improves VO2 kinetics and high-intensity exercise performance. Med Sci Sports 558
Exerc. 2011;43:457–467. 559
560
38. Thum JS, Parsons G, Whittle T, Astorino TA. High-intensity interval training elicits 561
higher enjoyment than moderate intensity continuous exercise. PLoS One. 562
2017;12(1):e0166299. 563
564
39. Losnegard T, Myklebust H, Spencer M, Hallén J. Seasonal variations in VO2max, O2-565
cost, O2-deficit, and performance in elite cross-country skiers. J Strength Cond Res. 566
2013;27(7):1780-90. 567
568
40. Fessi MS, Nouira S, Dellal A, Owen A, Elloumi M, Moalla W. Changes of the 569
psychophysical state and feeling of wellness of professional soccer players during pre-570
season and in-season periods. Res Sports Med. 2016;24(4):375-386. 571
572
41. Leveritt M, Abernethy PJ. Acute effects of high-intensity endurance exercise on 573
subsequent resistance activity. J Strength Cond Res. 1999;13:47–51. 574
575
42. Wiewelhove T, Raeder C, Meyer T, Kellmann M, Pfeiffer M, Ferrauti A. Markers for 576
routine assessment of fatigue and recovery in male and female team sport athletes 577
during high-intensity interval training. PLoS One. 2015;10(10):e0139801. 578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
18
FIGURES 595
596
Figure 1 Individual data points (dotted lines) and mean value (solid line) for maximum 597
oxygen consumption (percent of peak oxygen consumption (VO2peak); panel a) and mean 598
oxygen consumption (percent of VO2peak; panel b) during a 5x5-min HIT session with a fast 599
start and declining speed (DEC) or a more traditional evenly-distributed speed with the same 600
mean speed as DEC (TRAD). # Significant difference between sessions (p ˂ 0.05). 601
602
603
604
605
Figure 2 Average VO2 for all 5 work intervals during the HIT session with a fast start and 606
declining speed (DEC) or a more traditional evenly-distributed speed with the same mean 607
speed as DEC (TRAD; panel a). The statistical analyses were performed by comparing mean 608
values during the first two minutes, the third minute and during the two last minutes of all 609
work intervals (panel b). 610
# Significant difference between sessions for mean values during the first and last 2-min 611
periods (p ˂ 0.05). 612
19
Table 1 Data from the 5x5-min HIT sessions with a fast start and declining speed (DEC) or a 613
more traditional evenly-distributed speed with the same mean speed as DEC (TRAD). 614
VO2peak: peak oxygen consumption during the incremental VO2peak test; HRpeak: peak heart 615
rate during the incremental VO2peak test; [La-]: blood lactate concentration after the work 616
intervals; RPE: rate of perceived exertion after the work intervals. Values are mean±SD. 617
#Different from TRAD (p<0.05). 618
619
DEC
TRAD
Effect size
Mean velocity (km
h-1) 11.3 ± 0.8
11.3 ± 0.8
External workload (W) 278 ± 31 280 ± 27
Maximum VO2 measure (%
VO2peak)
98.2 ± 2.1
#
95.4 ± 3.1
0.95
Mean VO2 (% VO2peak) 87.6 ± 1.9
#
86.1 ± 3.2
0.56
Time ≥90% of VO2peak (min) 11.95 ± 4.08 10.84 ± 5.72
Maximum HR measure (%
HRpeak)
97.7 ± 2.2
97.2 ± 1.8
Mean HR (% HRpeak) 92.9 ± 2.2
92.5 ± 1.5
Time ≥90% of HRpeak (min) 20.32 ± 2.52 19.75 ± 1.84
Maximum [La
-
] measure
(mmolL-1)
9.25 ± 3.76
8.87 ± 2.86
Mean [La-] (mmol
L-1) 7.84 ± 2.78
7.62 ± 2.16
Maximum RPE measure 17.5 ± 1.4
#
18.1 ± 0.8
0.48
Mean RPE 16.1 ± 1.0
#
16.5 ± 0.7
0.41
