Content uploaded by Øystein Sylta
Author content
All content in this area was uploaded by Øystein Sylta on Feb 28, 2024
Content may be subject to copyright.
. . . Published ahead of Print
Medicine & Science in Sports & Exercise® Published ahead of Print contains articles in unedited
manuscript form that have been peer reviewed and accepted for publication. This manuscript will undergo
copyediting, page composition, and review of the resulting proof before it is published in its final form.
Please note that during the production process errors may be discovered that could affect the content.
Copyright © 2017 American College of Sports Medicine
Effects of HIT on Physiological and Hormonal Adaptions
in Well-Trained Cyclists
Øystein Sylta1, Espen Tønnessen2, Øyvind Sandbakk3, Daniel Hammarström4,
Jørgen Danielsen3, Knut Skovereng3, Bent R. Rønnestad4, and Stephen Seiler1
1Faculty of Health and Sport Sciences, University of Agder, Kristiansand, Norway; 2The
Norwegian Olympic Federation, Oslo, Norway; 3Centre for Elite Sports Research, Department of
Neuroscience, Norwegian University of Science and Technology, Trondheim, Norway; 4Section
for Sport Science, Lillehammer University College, Lillehammer, Norway
Accepted for Publication: 16 January 2017
PAP coversheet
ABSTRACT
PURPOSE: Investigate development of specific performance adaptions and hormonal responses
every 4th week during a 12-week HIT period in groups with different interval-training
prescriptions. METHODS: Sixty-three well-trained cyclists performing a 12-week intervention
consisting of 2-3 HIT sessions.week-1 in addition to ad libitum low intensity training. Groups
were matched for total training load, but increasing HIT (INC) group (n=23) performed interval-
sessions as 4x16 min in week 1-4, 4x8 min in week 5-8 and 4x4 min in week 9-12. Decreasing
HIT (DEC) group (n=20) performed interval-sessions in the opposite order as INC, and mixed
HIT (MIX) group (n=20) performed all interval-sessions in a mixed distribution during 12
weeks. Cycling-tests and measures of resting blood-hormones were conducted pre, week 4, 8 and
12. RESULTS: INC and MIX achieved >70% of total change in workload eliciting 4 mMol.L-1
[la-] (Power4mM) and 𝑉
O2peak during week 1-4, versus only 34-38% in DEC. INC induced larger
improvement vs. DEC during week 1-4 in Power4mM (ES: 0.7) and 𝑉
O2peak (ES: 0.8). All groups
increased similarly in peak power output (PPO) during week 1-4 (64-89% of total change). All
groups’ pooled, total- and free-testosterone and free-testosterone/cortisol-ratio decreased by
22±15%, 13±23% and 14±31% (all P<0.05), and insulin-like growth factor-1 increased by
10±14% (P<0.05) during week 1-4. CONCLUSIONS: Most of progression in Power4mM,
𝑉
O2peak and PPO was achieved during weeks 1-4 in INC and MIX, and accompanied by changes
in resting blood-hormones consistent with increased but compensable stress load. In these well-
trained subjects, accumulating 2-3 h.week-1 performing 4x16 min work bouts at best effort
induces greater adaptions in Power4mM and 𝑉
O2peak than accumulating ~1 h.week-1 performing
best effort intervals as 4x4 min.
Abstract
Effects of HIT on Physiological and Hormonal Adaptions in Well-Trained
1
Cyclists
2
3
Øystein Sylta1, Espen Tønnessen2, Øyvind Sandbakk3, Daniel Hammarström4, Jørgen
4
Danielsen3, Knut Skovereng3, Bent R. Rønnestad4, and Stephen Seiler1
5
6
1Faculty of Health and Sport Sciences, University of Agder, Kristiansand, Norway; 2The
7
Norwegian Olympic Federation, Oslo, Norway; 3Centre for Elite Sports Research, Department of
8
Neuroscience, Norwegian University of Science and Technology, Trondheim, Norway; 4Section
9
for Sport Science, Lillehammer University College, Lillehammer, Norway
10
11
12
Running title: Interval training organization
13
14
Corresponding author:
15
Øystein Sylta
16
University of Agder, Faculty of Health and Sport Science,
17
Postboks 442, 4604 Kristiansand
18
Norway
19
E-mail: oystein.sylta@uia.no
20
Telephone 0047 92252792
21
Fax number 0047 38141001
22
23
24
This study was funded in part by the Norwegian Olympic Committee, Oslo, Norway. We declare that the
25
results of the study are presented clearly, honestly, and without fabrication, falsification, or inappropriate
26
data manipulation. This study was funded by the affiliated Universities and The Norwegian Olympic
27
Federation. None of the authors has any relevant conflicts of interest. All were involved in designing the
28
study and writing the manuscript and/or acquisition and interpretation of data. The results of the present
29
study do not constitute endorsement by the American College of Sports Medicine.
30
Manuscript FINAL
ABSTRACT
31
PURPOSE: Investigate development of specific performance adaptions and hormonal responses
32
every 4th week during a 12-week HIT period in groups with different interval-training
33
prescriptions. METHODS: Sixty-three well-trained cyclists performing a 12-week intervention
34
consisting of 2-3 HIT sessions.week-1 in addition to ad libitum low intensity training. Groups
35
were matched for total training load, but increasing HIT (INC) group (n=23) performed interval-
36
sessions as 4x16 min in week 1-4, 4x8 min in week 5-8 and 4x4 min in week 9-12. Decreasing
37
HIT (DEC) group (n=20) performed interval-sessions in the opposite order as INC, and mixed
38
HIT (MIX) group (n=20) performed all interval-sessions in a mixed distribution during 12
39
weeks. Cycling-tests and measures of resting blood-hormones were conducted pre, week 4, 8 and
40
12. RESULTS: INC and MIX achieved >70% of total change in workload eliciting 4 mMol.L-1
41
[la-] (Power4mM) and 𝑉
O2peak during week 1-4, versus only 34-38% in DEC. INC induced larger
42
improvement vs. DEC during week 1-4 in Power4mM (ES: 0.7) and 𝑉
O2peak (ES: 0.8). All groups
43
increased similarly in peak power output (PPO) during week 1-4 (64-89% of total change). All
44
groups’ pooled, total- and free-testosterone and free-testosterone/cortisol-ratio decreased by
45
22±15%, 13±23% and 14±31% (all P<0.05), and insulin-like growth factor-1 increased by
46
10±14% (P<0.05) during week 1-4. CONCLUSIONS: Most of progression in Power4mM,
47
𝑉
O2peak and PPO was achieved during weeks 1-4 in INC and MIX, and accompanied by changes
48
in resting blood-hormones consistent with increased but compensable stress load. In these well-
49
trained subjects, accumulating 2-3 h.week-1 performing 4x16 min work bouts at best effort
50
induces greater adaptions in Power4mM and 𝑉
O2peak than accumulating ~1 h.week-1 performing
51
best effort intervals as 4x4 min. KEY WORDS: blood hormones, cycling, endurance
52
performance, lactate threshold, maximal oxygen consumption, training intensity
53
54
INTRODUCTION
55
A famous Norwegian coach of World Champions from 4 different endurance sports said “elite
56
endurance athletes must train a lot and they must train smart”. This advice is simple, but research
57
over several decades suggests that translating it into best practice is quite complex. Elite
58
endurance athletes organize their training around a high volume of low-intensity training (LIT,
59
defined as a workload eliciting a stable blood lactate concentration ([la-]) of less than 2 mMol.L-
60
1). This high volume of LIT is infused with smaller proportions of both moderate- (MIT, 2-4
61
mMol.L-1 [la-]) and high-intensity training (HIT, >4 mMol.L-1 [la-]). Training within these three
62
intensity categories, LIT, MIT, and HIT, is usually distributed either in a pyramidal or polarized
63
model (27, 32). Most retrospective studies on elite endurance athletes report a pyramidal training
64
distribution with approximately ≥80% LIT, 5-15% MIT and ≤10% HIT throughout the
65
preparation phase, e.g. (2, 25, 34). However, short term experimental studies demonstrate
66
superior responses to a polarized, compared to a pyramidal model (20, 31). This finding aligns
67
with the more polarized pattern observed among international medal winning athletes in the pre-
68
competition and competition period (3, 34). Adding or manipulating HIT, in combination with a
69
high volume of LIT, has been found to induce 2-12% average performance improvements in
70
groups of well-trained cyclists of varying performance levels over timeframes from a few weeks
71
to three months (18, 23, 29, 33). The primary physiological adaptations reported during these
72
relatively short intervention periods are increases in power output at lactate threshold (LT) and
73
maximal oxygen uptake (𝑉
O2max). Importantly, these effects are often only reported as net
74
changes from pre- to post- intervention period. There is still limited evidence available
75
concerning the time-course of adaptive development during a longer training cycle, and how this
76
development trajectory might be influenced by the organization and execution of the HIT
77
component during the training cycle.
78
79
During standardized HIT sessions, we have previously observed that relatively small changes in
80
exercise intensity are associated with large changes in tolerable accumulated exercise duration
81
(29, 33). Data from these studies and others raise important questions about how work intensity
82
and accumulated duration of HIT interact to signal physiological adaptation. For example,
83
Helgerud et al, 2007 (13) found that a total accumulated HIT duration of ~10-15 min at ~90-95%
84
of maximal heart rate (HRmax) had a greater impact on endurance performance than accumulating
85
~25 min at ~85% HRmax during a 3 session.week-1 interval training program lasting 8 weeks.
86
However, other studies conclude that accumulating ~30-45 min at ~90% HRmax twice per week
87
is a more effective HIT prescription than accumulating 15-20 min at ~95% HRmax (26, 29).
88
Discrepancies in reported results might be explained by the characteristics of the added HIT
89
stimuli, baseline performance level, age and small sample sizes.
90
91
Conceptually, optimization of endurance training can be seen as an attempt to maximize positive
92
adaptive signaling effects of training frequency, volume, and intensity adjustments while
93
managing accompanying psychological and physiological stress loads at tolerable levels.
94
Testosterone (T) and cortisol (C) have been suggested to be important mediators of the adaptive
95
response to endurance training, and considered as useful biomarkers of anabolic and catabolic
96
hormonal control, respectively (4, 5, 11, 14, 39). However, the relationship between the time-
97
course of training adaptations during a training cycle and the parallel time-course of potential
98
changes in resting T and C is not well established. Pre to post intervention comparisons do not
99
paint a consistent picture. For example, a 14-day mesocycle with frequent HIT sessions induced
100
both endurance adaptions and increases in serum T concentration in male junior triathletes (39).
101
In contrast, others have reported significant adaptive responses to a training program that also
102
induced declining T and increasing C concentrations indicative of an increased catabolic state
103
(14). Discrepancies among studies may be due to differences in the baseline training status of
104
participants, or the training dose administered. Further, a decrease in the ratio between free
105
testosterone and cortisol (FTCR) has been proposed as a marker of the overtraining syndrome (1,
106
10), although doubt has been cast as to whether FTCR is able to differentiate between functional
107
overreaching and overtraining (36, 37). In addition, increased human growth hormone (HGH)
108
has been reported in endurance trained subjects, and elevated 24 h HGH secretion rates
109
combined with increased plasma levels of insulin-like growth factor-1 (IGF-I) have also been
110
found to correlate positively with 𝑉
O2max (8, 21). This finding is consistent with the observation
111
that a one-year exercise training program approximately doubled resting HGH concentration in
112
untrained women (38). However, the effect of multiple training-cycles with different intensities
113
and accumulated HIT duration on hormonal responses in well-trained endurance athletes remains
114
to be thoroughly investigated.
115
116
The aims of the present study were therefore to compare the influence of three different 12-week
117
training programs differing in HIT load intensification structure on: 1 - the development of
118
specific endurance adaptations, 2 - the potential interactions among the different HIT
119
prescriptions, and 3 - the time-course of changes in resting anabolic and catabolic hormones over
120
12 weeks divided in three mesocycles.
121
122
123
METHODS
124
This study was conducted as a multicenter trial, with all participants completing a 12-week
125
training period, divided in three 4-week cycles. These data were collected in parallel with data
126
from a newly published study where the main purpose was to compare the effects of different
127
periodized HIT models in well-trained endurance athletes (33).
128
129
Subjects
130
Sixty-nine experienced male competitive cyclists (age: 38±8 years, 𝑉
O2peak: 62±6 mL.kg-1.min-1,
131
training experience: 6±4 years) completed the intervention period, with 63 included in the final
132
analyses. Six subjects were excluded due to absence from post-testing, and/or <70% compliance
133
with prescribed interval sessions. Based on peak power output (PPO), training volume and
134
cycling experience, subjects were categorized as well-trained (15). The study was approved by
135
the ethics committee of the Faculty for Health and Sport Science, University of Agder, and
136
registered with the Norwegian Social Science Data Services (NSD). All athletes provided their
137
written informed consent to participate in the study.
138
139
Pre-intervention period
140
A 6-week pre-intervention period was conducted in order to ensure an approximately equal
141
training status, and familiarize subjects with testing protocols and interval sessions included in
142
the intervention period (Figure 1). Subjects were instructed to perform only one interval session
143
each week, combined with ad libitum LIT volume. Pre-testing was performed at the end of the
144
pre-intervention period (mid-December), and subjects were thereafter randomized in a stratified
145
manner based on age, cycling experience and peak oxygen uptake (𝑉
O2peak) into one of three
146
different training groups; increasing HIT (INC) (n=23), decreasing HIT (DEC) (n=20) or mixed
147
HIT (MIX) (n=20) group.
148
149
Intervention period
150
The training intervention was performed from early January to the end of March, and consisted
151
of 12 weeks, divided in three 4-week cycles. Subjects were instructed to follow a training load
152
structure within each cycle as follows; week 1; medium LIT volume and two supervised interval
153
sessions, week 2 and 3; high LIT volume and three supervised interval sessions, week 4; reduced
154
LIT volume by 50% compared to the previous two weeks and 1-2 laboratory testing sessions. All
155
interval sessions was performed indoors as supervised group training, and included a 20-30 min
156
low-intensity (55-70% HRmax) warm-up, followed by four interval bouts of either 4, 8 or 16 min
157
separated by 2 min rest, and concluded with 10-30 min low-intensity (55-70% HRmax) cool-
158
down. During interval sessions, subjects were instructed to cycle at their maximal sustainable
159
intensity during all four interval bouts (isoeffort) (28, 29) such that they completed the described
160
session structure (all four interval bouts completed with only 2 min rest), and with consistent or
161
slightly progressive power output from the 1st to the 4th interval bout. In total, each participant
162
was prescribed 24 supervised interval sessions during the 12-week intervention period, in
163
addition to testing and self-organized ad libitum LIT. Figure 1 shows the study design and
164
interval session prescriptions in each group during the intervention. INC group performed eight
165
interval sessions as 4x16 min in cycle 1, eight interval sessions as 4x8 min in cycle 2 and eight
166
interval sessions as 4x4 min in cycle 3. DEC group performed interval sessions in the opposite
167
cycle order as INC, and MIX group organized all 24 interval sessions (eight in each cycle) in a
168
mixed distribution; session 1 as 4x16 min, session 2 as 4x8 min, session 3 as 4x4 min, session 4
169
as 4x16 min and so on.
170
171
- - Figure 1 - -
172
173
Although all sessions were performed with isoeffort instructions, the different interval session
174
prescriptions differing in interval bout duration and total accumulated HIT duration, induced
175
significantly different power output, [la-], heart rate (HR) and rating of perceived exertion (RPE)
176
responses (Table 1). During each interval session, independent of prescription, there were
177
significant increased HR and RPE responses from interval bout 1 to 4 (data not presented). The
178
evolution of power output was, in keeping with the instructions given to subjects, maintained
179
relatively constant over the 4 interval bouts. However, sub-analyses revealed that relatively few
180
subjects (n=6) typically showed a decreasing power development over 4x16 min, whilst in
181
contrast, 23 of 63 subjects typically reduced their power output by the end of 4x4 min sessions.
182
Data in Table 1 are presented as average values during all four interval bouts for all three groups
183
pooled. There were no differences across groups, although different interval prescriptions (4x16
184
and 4x4 min) were performed in opposite sequence (cycle 1 and 3) for INC and DEC,
185
respectively.
186
187
- - Table 1 - -
188
189
190
Testing procedures
191
Cycling test
192
Testing weeks included a laboratory-based cycling-test, which were conducted pre-intervention,
193
and at week 4, 8 and 12 during the intervention period (see Figure 1). Subjects were instructed to
194
perform only LIT during the 48 h preceding each test and to consume the same type of pre-test
195
meal. Subjects were not permitted to eat during the last hour, or consume caffeine during the last
196
3 h preceding each test.
197
198
Briefly, four to six steady state submaximal 5-min steps were performed on a bicycle ergometer
199
to identify the workload eliciting 4 mMol.L-1 [la-] (Power4mM) and gross efficiency (GE). The test
200
started at 125 W, increased 50 W (25 W if [la-] was ≥3 mMol.L-1) every fifth minute, and
201
terminated when [la-] reached ≥4 mMol.L-1. Our purpose of reporting Power4mM was not to
202
determine LT or maximal lactate steady state (MLSS), but having a fixed value for
203
measurements of changes during different test periods. After 10 min recovery, an incremental
204
test, starting at 3 W.kg-1 and subsequently increased by 25 W every minute until voluntary
205
exhaustion, was performed to determine 𝑉
O2peak and PPO (calculated as mean power output of
206
the last completed minute). Finally, after 15 min recovery, a 30 s all-out Wingate test was
207
performed to identify mean power during 30 s (Power30s). A detailed description of all testing
208
protocols, instruments and materials has recently been described elsewhere (33).
209
210
Serum hormone concentrations
211
Venous blood samples were collected from a sub-group of twenty-nine subjects to assess
212
hormonal responses (INC; n=9, DEC; n=10, MIX; n=10). For each testing session (pre, week 4,
213
8 and 12) all subjects reported to the laboratory between 07.00 and 09.00 AM in a rested, fasted
214
state, and were only allowed to perform LIT 48 h preceding blood-tests. Approximately 10 mL
215
venous blood was collected from an antecubital vein using Vacutainer tubes (Becton Dickinson,
216
Franklin Lanes, USA). Samples were stored at room temperature (20-22°C) for 30-60 min before
217
being centrifuged for 10 min at 3000 rpm (StatSpin Express 4, Beckman Coulter, USA). The
218
supernatant serum was pipetted into 1 mL aliquots and immediately frozen at –20°C until
219
analyses. Serum was analyzed for total testosterone (TT), free testosterone (FT), C, IGF-1, IGF-
220
BP3, HGH, sexual hormone binding globulin (SHBG) and prolactin (PRL). The FTCR was
221
calculated using the method of Banfi & Dolci (2006) (1). Given the sensitivity of resting HGH to
222
natural variations or dietary status (although subjects were in a fasted state), subjects with
223
extreme outlier values (identified through boxplot analyses in SPSS) were excluded from HGH
224
analyses. Four, 1 and 3 subjects were excluded from the INC, DEC and MIX group, respectively.
225
Sub-analyses were executed to ensure that this sub-group of 29 subjects (both pooled and divided
226
in intervention groups) was representative to the main findings of specific performance responses
227
in the present study (not presented).
228
229
Statistical analyses
230
Data were analyzed using SPSS 22.0 (SPSS Inc, Chicago, IL, USA) and are presented as mean ±
231
standard deviation (SD) or 95% confidence intervals (95% CI). Training characteristics and
232
differences in blood hormone responses between groups were compared using a one-way
233
between-groups analysis of variance (ANOVA). A GLM repeated measures model (ANOVA)
234
was used to assess statistical differences in physiological test variables and blood hormones from
235
pre to week 4, 8 and 12 within each group. Statistical comparisons were followed by Bonferroni
236
post hoc corrections if there was a significant within-group difference. A univariate General
237
Linear Model (GLM) (analysis of covariance (ANCOVA)) was used to assess differences in
238
physiological baseline characteristics and delta changes (pre – week 4, week 4 – 8 and week 8 –
239
12) in physiological test variables between each training group. For physiological test-variables,
240
GLM analyses were adjusted for the influence of different covariates (test-location and pre
241
Power4mM (w.kg-1)), and presented as adjusted values. Effect size (ES) was calculated according
242
to Cohen’s d (0.2=small, 0.5=medium, 0.8=large) (7). Medium or large ES (≥0.5) are discussed
243
as tendencies if comparisons are non-significant. A total of <2% of all data variables were
244
missing, and treated as “last observation carried forward”. For all comparisons, statistical
245
significance was accepted as α ≤0.05.
246
247
RESULTS
248
Body mass
249
There were no significant differences in body mass among groups at pre. After 12 weeks, there
250
was a significant body mass reduction in INC (80.3±7.4 vs. 79.0±7.6 kg), DEC (79.7±7.8 vs.
251
78.5±7.5 kg) and MIX (79.7±8.9 vs. 78.2±8.8 kg) (all P<0.05). All physiological and
252
performance adaptions are further presented as absolute values, hence relative values with
253
respect to body mass are therefore slightly different.
254
255
Training characteristics
256
There were no differences among groups in any training variables at pre. Weekly training
257
volume did not change in the three cycles and was 9.8±3.2, 10.0±3.2, and 10.7±3.1 h.week-1 in
258
cycles 1-3, respectively. For detailed training characteristics see Sylta et al (2016) (33). The only
259
difference among groups was the intensity x accumulated duration of HIT within cycle 1-3
260
(Table 1, Figure 1 and Figure 2, A-C). INC, DEC and MIX completed on average 95±5%,
261
94±8% and 93±9% of their 24 prescribed interval sessions, respectively. Overall, the 3 HIT
262
prescriptions were executed with even pacing, as prescribed. Mean power output was within +/-
263
3 W from work bout 1 to 4 within each prescription. However, at the individual level, execution
264
of the 4x4 min prescription was more often associated with a negative pacing pattern (observed
265
in ~1/3 of subjects) where power output declined >2% from the first to last work bout.
266
267
Adaptation time-course
268
Of the total change in Power4mM and 𝑉
O2peak during 12 weeks, INC achieved 98±80% and
269
70±80%, and MIX 147±74% and 92±74%, respectively, while DEC achieved only 34±83% and
270
38±91%, during the first 4 weeks of intensified training (Figure 2). However, changes in PPO
271
during cycle 1 were similar, 77±52%, 64±86% and 89±88% of total change in INC, MIX and
272
DEC groups, respectively. There was a significant change in Power30s in DEC during cycle 1.
273
Only small changes occurred during 12 weeks in all groups with respect to GE, and will not be
274
any further discussed. See Table 2 for more details.
275
276
Individual adaption variation was very large in all test variables in this cohort. For example,
277
overall mean improvement in PPO from pre to week 12 was 6±7% (P<0.05). However, the
278
individual range was from -9 to 36%, a range which is representative for all test variables
279
presented.
280
281
- - Figure 2 - -
282
283
- - Table 2 - -
284
285
Group comparisons
286
During cycle 1, INC and MIX significantly increased PPO, Power4mM and 𝑉
O2peak (all P<0.05),
287
while DEC significantly increased PPO and Power30s (all P<0.05). There were no significant
288
differences in delta changes in any test variables across INC (4x16 min), DEC (4x4 min), or
289
MIX (Table 2 and Figure 3). However, INC (4x16 min) revealed a moderate ES compared to
290
DEC (4x4 min) when comparing delta changes in Power4mM (ES: 0.7) and 𝑉
O2peak (ES: 0.7). A
291
similar analysis of PPO and Power30s revealed no differences between INC and DEC.
292
293
During cycle 2, DEC increased significantly in Power4mM (P<0.05). No further significant
294
changes were observed in any test variables in INC (4x8 min), DEC (4x8 min), or MIX (all
295
P>0.05), and there were no significant differences in delta changes between groups (Table 2,
296
Figure 3).
297
298
During cycle 3, DEC significantly increased 𝑉
O2peak (P<0.05). No further significant changes
299
were observed for any test variables in INC (4x4 min), DEC (4x16 min), or MIX (all P>0.05),
300
and there were no significant differences in delta changes between groups (Table 2, Figure 3).
301
However, in this final 4-week cycle, DEC (4x16 min) revealed a moderate ES compared to INC
302
(4x4 min) when comparing delta changes in both Power4mM (ES: 0.5) and 𝑉
O2peak (ES: 0.5).
303
304
- - Figure 3 - -
305
306
Blood hormones
307
The sub-sample of 29 subjects assessed for anabolic and catabolic hormonal responses in
308
addition to physiological tests were representative of the total sample in terms of both adaptive
309
time-course and group comparisons. There were no significant differences among INC, DEC and
310
MIX at pre for any blood hormone measured.
311
312
Pooling the three training groups, TT, FT and FTCR decreased by 22±15%, 13±23% and
313
14±31%, respectively by the end of the first 4-week training cycle (all P<0.05). IGF-1 increased
314
10±14% (P<0.05). In contrast, comparing pre to week 12, TT, IGF-1 and IGF-BP3 increased
315
24±31%, 11±18% and 8±13%, respectively (all P<0.05, Figure 4).
316
317
Hormonal changes are presented in Figure 4 as delta changes in each group across 12 weeks.
318
Most important findings are:
319
TT decreased 27±15%, 25±14% and 16±15% during cycle 1 in INC, DEC and MIX
320
groups, respectively (all P<0.05), and returned to pre-intervention levels by the end of
321
cycle 2 (P>0.05 vs. pre). MIX group had 42±24% elevated TT at the end of cycle 3
322
compared to pre (P<0.05).
323
FT decreased 24±15% in INC during cycle 1 (P<0.05) and returned to pre-intervention
324
level by cycle 3. The decline in FT was significantly higher in INC compared to DEC
325
(24±15% vs. 1±29%) during cycle 1 (P<0.05, ES: 1.0).
326
FTCR decreased 22±27%, 12±25% and 8±41% during cycle 1 in INC, DEC and MIX
327
groups, respectively (all P>0.05). A comparison of INC (4x16 min) (22±27%) vs. DEC
328
(4x4 min) (12±25%) during cycle 1, revealed an effect size of 0.4 (P>0.05). A
329
comparison of DEC (4x16 min) (decreased 4±20%) vs. INC (4x4 min) (increased
330
18±34%) in the final cycle revealed a significant difference (P<0.05, ES: 0.9).
331
HGH increased 38±80% in INC (4x16 min) compared to 19±45% in DEC (4x4 min)
332
during cycle 1 (P>0.05, ES: 0.5).
333
334
- - Figure 4 - -
335
336
Discussion
337
This study can be summarized with three key findings:
338
1. HIT performed during the initial 4-weeks of training appears to have larger impact on
339
specific performance outcomes than what occurs later in the periodized mesocycles. Both
340
INC (4x16 min) and MIX reached ≥70% of total development in Power4mM and 𝑉
O2peak,
341
while DEC (4x4 min) reached ≥89% of total development in PPO and Power30s already
342
during cycle 1.
343
2. Performing 2-3 weekly HIT sessions with an interval prescription of 4x16 min, induced
344
greater adaptions in Power4mM and 𝑉
O2peak compared to the same frequency of a 4x4 min
345
prescription whether prescribed early or late during a 12-week periodization plan.
346
3. The first 4 training weeks, which were associated with the largest progression in
347
Power4mM, 𝑉
O2peak, PPO and Power30s in specific groups, were also characterized by
348
decreases in anabolic hormones in all groups. In training cycles 2 and 3, resting hormone
349
values rebounded to baseline levels or even increased, but this rebound was accompanied
350
by smaller adaption magnitude.
351
352
Our first key finding is that ≥70% of the progression in Power4mM and 𝑉
O2peak was achieved
353
already during the initial 4 weeks of training for both INC (4x16 min) and MIX group, while
354
DEC (4x4 min) reached ≥89% of total development in PPO and Power30s in cycle 1. During this
355
period, all groups increased ~2-6% in Power4mM, 𝑉
O2peak and PPO, a magnitude comparable to
356
previous studies of similar length (18, 24).
357
358
To stimulate improvements in endurance capacity in already well-trained athletes it appears
359
necessary to increase the total training volume (3, 9, 25), increase intensity of the aerobic
360
endurance training (17, 19) or reorganize HIT training in, for example, block periods to provide
361
an adequate stimuli (23, 24). In the present study, subjects increased the HIT frequency from 1
362
weekly session during the pre-intervention period, to 2-3 weekly sessions during the intervention
363
period. On average, this intensification provided a sufficient stimulus to elicit physiological
364
improvements in Power4mM, 𝑉
O2peak and PPO (Figure 2/Table 2). This finding alone is not
365
surprising. However, most previous training intervention studies present only pre to post results
366
during similar timeframes (13, 23, 26, 29, 31). We therefore argue that by providing a time-
367
course with more frequent testing (e.g. every 4th week) more accurate prediction of training
368
effects over more extended timeframes can be achieved. Bearing in mind that most of the
369
positive effect in specific variables was achieved already during the initial 4 weeks of training
370
intensification, our results highlight that extrapolating short term adaptation rates from a training
371
intervention involving HIT to even modestly longer time frames is ill-advised. In this context, it
372
is interesting that 4-week cycles are quite commonly used in elite endurance sport, often
373
characterized by 3-week training load builds and 1-week load reductions. Our findings are also
374
consistent with training descriptions of elite endurance athletes, who use HIT consistently but
375
relatively sparingly when examined over an entire training year (3, 25, 34).
376
377
Group comparisons
378
Our second key finding is that accumulating 2-3 h.week-1 at the “lower” end of the HIT range
379
performing intervals as 4x16 min, tended to elicit superior adaptions in Power4mM and 𝑉
O2peak
380
compared to accumulating ~1 h.week-1 at the “higher” end of the HIT range performing a 4x4
381
min interval prescription.
382
383
During the first training cycle, a 4x16 min “isoeffort” interval training prescription (INC group)
384
tended to induce greater adaptations in Power4mM and 𝑉
O2peak compared to a 4x4 min interval
385
prescription (DEC group). The ES of the relative improvement in Power4mM and 𝑉
O2peak revealed
386
a moderate effect of 4x16 min vs. 4x4 min prescription. Even in the final cycle when, in theory,
387
much of the short-term adaptation potential had been realized, we found a similar tendency.
388
These results are in line with previous findings from our research group. Both Seiler et al (2013)
389
(29) and Sandbakk et al (2013) (26) found that a HIT prescription accumulating more minutes at
390
a slightly lower intensity level compared to a 4x4 min prescription, induced greater overall
391
adaptive response, inclusive 𝑉
O2max, in recreational to well-trained athletes. The present study
392
was however performed on a much larger group of well-trained subjects (n=69) during a longer
393
time-frame. Furthermore, a case study of a professional cyclist suggests that increasing HIT time
394
by slightly decreasing intensity during 2-3 weekly interval sessions, in combination with an
395
increase in total training volume, increased 𝑉
O2max from 82 to 90 ml.kg-1.min-1 during a 3-month
396
period (30). However, in contrast to our results, Helgerud et al (2007) (13) observed that 4x4 min
397
intervals at 90-95% HRmax lead to larger improvements in endurance capacity compared to LT
398
training at ~85% HRmax. The training groups in the study by Helgerud and colleagues were
399
however matched for total work (isoenergetic) in contrast to our “maximal overall effort”
400
(isoeffort) model. Consequently, the LT training sessions were only modestly longer in
401
accumulated duration than the 4x4 min sessions. This form of matching is not consistent with
402
how athletes manage intensity and accumulated duration in their daily training. We argue that
403
matching training for overall effort is more representative of this process in well-trained athletes.
404
405
During cycle 1, DEC was the only group which significantly improved in both PPO and
406
Power30s. This may be because those variables are more specific to a 4x4 min interval
407
prescription due to higher power output (Table 1). PPO performed as an incremental test is a
408
function of both aerobic and anaerobic energy supply. Therefore, an individual can increase in
409
PPO without any change in aerobic energy supply. Due to no or only small aerobic adaptions in
410
DEC during cycle 1, we speculate that the observed increase in PPO was a result of anaerobic
411
energy supply adaptions or other adaptions related to postponing accumulation of fatigue
412
metabolites.
413
414
Blood hormones
415
The third key finding is that large progression in Power4mM, 𝑉
O2peak, PPO and Power30s in
416
specific groups during the first 4 weeks was accompanied with a decrease in anabolic hormones
417
in all groups, which thereafter rebounded to baseline levels in cycles 2 and 3, when adaption
418
magnitude was reduced.
419
420
During the first 4-week cycle, both TT, FT and FTCR decreased significantly. Although an
421
anabolic response (increased T/decreased C) is most likely expected together with physiological
422
adaptions, reduced serum concentrations of T (measured in a fasted rested state) after a
423
successful period of intensive training have also been observed elsewhere (5, 12). However, an
424
acute increase in the circulating concentration of T is also a normal observation directly after
425
high intensity endurance exercise (35). Up-regulation of T has been suggested to be associated
426
with increased androgen receptor expression (AR) (22). Therefore, we speculate whether
427
increased expression of AR can partially explain the present temporary reduction (measured after
428
4 weeks) in serum T, due to increased binding of T to AR and therefore increased uptake of T in
429
muscle cells (16). Speculating further, this increased T uptake could, in turn, amplify the
430
intracellular signal for endurance adaptation. The present results suggest that in well-trained
431
cyclists, a modest reduction in T levels during intensified training do not need to predict
432
decreased performance.
433
434
For all groups pooled together during the entire 12-week training period, we observed a
435
significant increase in both TT and IGF-1/BP3. The observed anabolic response was
436
accompanied by improvements in key components of performance, such as PPO, Power4mM and
437
𝑉
O2peak. This is in agreement with previous findings that have demonstrated that training periods
438
with frequent HIT sessions increase T levels (39), and that increased IGF-1 correlates positively
439
with improvements in 𝑉
O2max (8, 21).
440
441
When comparing between groups, superior adaptations in Power4mM and 𝑉
O2peak were observed
442
in INC (4x16 min) compared to DEC (4x4 min) during the first training cycle. Simultaneously,
443
we also observed a large ES and a significant difference when comparing the decrease in FT in
444
INC and DEC group, which may indicate a functional, controlled overreaching in INC group,
445
and may explain absence of physiological adaption in DEC. On the other hand, decreased T in
446
combination with increased C has been proposed as an early marker of the overtraining
447
syndrome, and a change in FTCR of >30% as a boundary to diagnose overtraining (36, 37). In
448
the present study, FTCR decreased by 22% after performing cycle 1 with 4x16 min interval
449
prescription (INC), compared to 12% after a 4x4 min interval prescription (DEC) (ES: 0.4). This
450
pattern was confirmed during the final cycle, where a 4x16 min interval prescription (DEC) was
451
followed by a 4% decreased in FTCR, compared to an 18% increase after a 4x4 min interval
452
prescription (INC) (ES: 1.0). This suggests that the 2-3 weekly sessions of 4x16 min were very
453
demanding, but may be necessary to stimulate large aerobic enhancements in already well-
454
trained cyclists. The latter is supported by the fact that superior endurance adaptations have been
455
observed after implementing periods with very demanding HIT blocks, compared to a more even
456
distribution of the same training volume and exercise intensity distribution (23). Although FTCR
457
decreased, we found a 38% increase in the anabolic hormone HGH in INC (4x16 min) vs. 19%
458
in DEC (4x4 min) group (moderate ES) during cycle 1. It has been suggested that circulating
459
HGH may act as a positive stimulus for expansion of plasma volume and erythropoiesis (6).
460
Altogether, the hormonal data from the first training cycle indicate that differences in hormonal
461
changes induced by the different HIT training cycles may contribute to the observed differences
462
in adaptations between the training groups.
463
464
465
Methodological considerations
466
This present intervention period aimed to simulate a preparation period leading up to the
467
competition period, and not peak performance. We assume that athletes switch their training-
468
focus after a similar period, for example by competing regularly. The intention with interval
469
sessions was therefore mainly to build general aerobic performance capacity. Performed
470
intensities differed in all interval prescriptions (Table 1). The 4x16 min was executed at an
471
average power output just below Power4mM and almost all subjects managed to achieve a
472
constant or slightly increasing power output evolution from first to fourth interval bout. We
473
suggest that the 4x16 min intensity is near power output at LT or MLSS, but still in the lower
474
range of the HIT zone, and therefore almost exclusively sustained through aerobic metabolism.
475
However, the 4x4 min prescription was executed 15-20% above Power4mM and therefore in the
476
upper range of the HIT zone or near maximal aerobic intensities. In addition, subjects more often
477
failed our “steady or increasing” prescription during 4x4 min intervals, indicative of more
478
“anaerobic” intracellular metabolic conditions that may not be conducive to optimal adaptive
479
signaling of aerobic metabolic adaptations. These differences may explain why we observed
480
different specific performance adaptions comparing a 4x16 min vs. 4x4 min interval prescription,
481
especially during cycle 1. Our result suggesting that it is advantageous for well-trained
482
endurance athletes to accumulate a large training volume at or near MLSS intensity, is contrary
483
to intervention studies (31) or retrospective descriptions of elite athletes during a competition
484
period (34) emphasizing the advantages of a polarized training model. Hence, more research
485
evaluating the effects of large volume of training near MLSS intensities in elite athletes is
486
needed.
487
488
We acknowledge the lack of an all-out time-trial performance test every 4th week during the
489
present intervention study. However, we argue PPO in addition to physiological variables, to be
490
strong predictors of cycling performance.
491
492
CONCLUSION
493
The results of the current study suggest that most of the progression in Power4mM, 𝑉
O2peak, PPO
494
and Power30s during a 12 week HIT intervention were achieved already during the initial 4 weeks
495
of training. However, the magnitude of adaption was dependent on the specific interval training
496
prescription, independent of timing of prescription. Accumulating 2-3 h per week performing
497
intervals as 4x16 min appears to induce greater adaptions in Power4mM and 𝑉
O2peak compared to
498
accumulating ~1 h per week performing intervals as 4x4 min. Resting levels of anabolic
499
hormones were found to first decline and then rebound over 12 weeks, with the period of decline
500
associated with greater adaption.
501
502
ACKNOWLEDGEMENTS
503
We would like to thank Dr. Michael Vogt and Professor Primus Mullis for their positive spirit
504
and important contributions to the planning of the study and analyses of blood hormones. Sadly,
505
they both passed away during the process of this study. We would also like to thank and
506
acknowledge the enthusiastic group of test cyclists who made this study possible. This study
507
was funded in part by the Norwegian Olympic Committee, Oslo, Norway. We declare that the
508
results of the study are presented clearly, honestly, and without fabrication, falsification, or
509
inappropriate data manipulation.
510
511
CONFLICT OF INTEREST
512
This study was funded by the affiliated Universities and The Norwegian Olympic Federation.
513
None of the authors has any relevant conflicts of interest. All were involved in designing the
514
study and writing the manuscript and/or acquisition and interpretation of data. The results of the
515
present study do not constitute endorsement by the American College of Sports Medicine.
516
517
REFERENCES
518
1. Banfi G, Dolci A. Free testosterone/cortisol ratio in soccer: usefulness of a categorization
519
of values. The Journal of sports medicine and physical fitness. 2006;46(4):611-6.
520
2. Billat V, Lepretre PM, Heugas AM, Laurence MH, Salim D, Koralsztein JP. Training and
521
bioenergetic characteristics in elite male and female Kenyan runners. Medicine and
522
science in sports and exercise. 2003;35(2):297-304; discussion 5-6.
523
3. Billat VL, Demarle A, Slawinski J, Paiva M, Koralsztein JP. Physical and training
524
characteristics of top-class marathon runners. Medicine and science in sports and
525
exercise. 2001;33(12):2089-97.
526
4. Bonifazi M, Bela E, Carli G et al. Influence of training on the response of androgen
527
plasma concentrations to exercise in swimmers. European journal of applied physiology
528
and occupational physiology. 1995;70(2):109-14.
529
5. Chennaoui M, Gomez-Marino D, Drogou C, Bourrilhon C, Sautivet S, Guezennec CY.
530
Hormonal and metabolic adaptation in professional cyclists during training. Canadian
531
journal of applied physiology = Revue canadienne de physiologie appliquee.
532
2004;29(6):714-30.
533
6. Christ ER, Cummings MH, Westwood NB et al. The importance of growth hormone in
534
the regulation of erythropoiesis, red cell mass, and plasma volume in adults with growth
535
hormone deficiency. The Journal of clinical endocrinology and metabolism.
536
1997;82(9):2985-90.
537
7. Cohen JW. Statistical power analysis for the behavioral sciences. 2en ed. Hillsdale (NJ):
538
Lawrence Erlbaum Associates; 1988. p. 567.
539
8. Eliakim A, Brasel JA, Mohan S, Barstow TJ, Berman N, Cooper DM. Physical fitness,
540
endurance training, and the growth hormone-insulin-like growth factor I system in
541
adolescent females. The Journal of clinical endocrinology and metabolism.
542
1996;81(11):3986-92.
543
9. Fiskerstrand Â, Seiler KS. Training and performance characteristics among Norwegian
544
International Rowers 1970–2001. Scandinavian Journal of Medicine and Science in
545
Sports. 2004;14(5):303-10.
546
10. Fry RW, Morton AR, Keast D. Overtraining in athletes. An update. Sports Med.
547
1991;12(1):32-65.
548
11. Grandys M, Majerczak J, Duda K, Zapart-Bukowska J, Kulpa J, Zoladz JA. Endurance
549
training of moderate intensity increases testosterone concentration in young, healthy men.
550
International journal of sports medicine. 2009;30(7):489-95.
551
12. Hackney AC, Sharp RL, Runyan WS, Ness RJ. Relationship of resting prolactin and
552
testosterone in males during intensive training. British journal of sports medicine.
553
1989;23(3):194.
554
13. Helgerud J, Hoydal K, Wang E et al. Aerobic high-intensity intervals improve VO2max
555
more than moderate training. Medicine and science in sports and exercise.
556
2007;39(4):665-71.
557
14. Hoogeveen AR, Zonderland ML. Relationships between testosterone, cortisol and
558
performance in professional cyclists. International journal of sports medicine.
559
1996;17(6):423-8.
560
15. Jeukendrup AE, Craig NP, Hawley JA. The bioenergetics of World Class Cycling.
561
Journal of science and medicine in sport / Sports Medicine Australia. 2000;3(4):414-33.
562
16. Kraemer WJ, Spiering BA, Volek JS et al. Androgenic responses to resistance exercise:
563
effects of feeding and L-carnitine. Medicine and science in sports and exercise.
564
2006;38(7):1288-96.
565
17. Laursen PB, Jenkins DG. The Scientific Basis for High-Intensity Interval Training.
566
Sports Med. 2002;32(1):53-73.
567
18. Laursen PB, Shing CM, Peake JM, Coombes JS, Jenkins DG. Interval training program
568
optimization in highly trained endurance cyclists. Medicine and science in sports and
569
exercise. 2002;34(11):1801-7.
570
19. Midgley AW, McNaughton LR, Wilkinson M. Is there an optimal training intensity for
571
enhancing the maximal oxygen uptake of distance runners?: empirical research findings,
572
current opinions, physiological rationale and practical recommendations. Sports Med.
573
2006;36(2):117-32.
574
20. Neal CM, Hunter AM, Brennan L et al. Six weeks of a polarized training-intensity
575
distribution leads to greater physiological and performance adaptations than a threshold
576
model in trained cyclists. Journal of applied physiology. 2013;114(4):461-71.
577
21. Poehlman ET, Copeland KC. Influence of physical activity on insulin-like growth factor-
578
I in healthy younger and older men. The Journal of clinical endocrinology and
579
metabolism. 1990;71(6):1468-73.
580
22. Ratamess NA, Kraemer WJ, Volek JS et al. Androgen receptor content following heavy
581
resistance exercise in men. The Journal of steroid biochemistry and molecular biology.
582
2005;93(1):35-42.
583
23. Rønnestad BR, Ellefsen S, Nygaard H et al. Effects of 12 weeks of block periodization on
584
performance and performance indices in well-trained cyclists. Scandinavian journal of
585
medicine & science in sports. 2014;24(2):327-35.
586
24. Rønnestad BR, Hansen J, Ellefsen S. Block periodization of high-intensity aerobic
587
intervals provides superior training effects in trained cyclists. Scandinavian journal of
588
medicine & science in sports. 2014;24(1):34-42.
589
25. Sandbakk O, Holmberg HC, Leirdal S, Ettema G. The physiology of world-class sprint
590
skiers. Scandinavian journal of medicine & science in sports. 2011;21(6):e9-16.
591
26. Sandbakk O, Sandbakk SB, Ettema G, Welde B. Effects of intensity and duration in
592
aerobic high-intensity interval training in highly trained junior cross-country skiers.
593
Journal of strength and conditioning research / National Strength & Conditioning
594
Association. 2013;27(7):1974-80.
595
27. Seiler S. What is best practice for training intensity and duration distribution in endurance
596
athletes? International journal of sports physiology and performance. 2010;5(3):276-91.
597
28. Seiler S, Hetlelid KJ. The impact of rest duration on work intensity and RPE during
598
interval training. Medicine and science in sports and exercise. 2005;37(9):1601-7.
599
29. Seiler S, Joranson K, Olesen BV, Hetlelid KJ. Adaptations to aerobic interval training:
600
interactive effects of exercise intensity and total work duration. Scandinavian journal of
601
medicine & science in sports. 2013;23(1):74-83.
602
30. Seiler S, Tønnessen E. Intervals, thresholds, and long slow distance: the role of intensity
603
and duration in endurance training. Sportscience. 2009;13:32-53.
604
31. Stoggl T, Sperlich B. Polarized training has greater impact on key endurance variables
605
than threshold, high intensity, or high volume training. Frontiers in physiology.
606
2014;5:33.
607
32. Stoggl TL, Sperlich B. The training intensity distribution among well-trained and elite
608
endurance athletes. Frontiers in physiology. 2015;6:295.
609
33. Sylta O, Tonnessen E, Hammarstrom D et al. The Effect of Different High-Intensity
610
Periodization Models on Endurance Adaptations. Medicine and science in sports and
611
exercise. 2016.
612
34. Tønnessen E, Sylta O, Haugen TA, Hem E, Svendsen IS, Seiler S. The road to gold:
613
training and peaking characteristics in the year prior to a gold medal endurance
614
performance. PloS one. 2014;9(7):e101796.
615
35. Tremblay MS, Copeland JL, Van Helder W. Influence of exercise duration on post-
616
exercise steroid hormone responses in trained males. European journal of applied
617
physiology. 2005;94(5-6):505-13.
618
36. Urhausen A, Gabriel H, Kindermann W. Blood hormones as markers of training stress
619
and overtraining. Sports Med. 1995;20(4):251-76.
620
37. Urhausen A, Kindermann W. Diagnosis of Overtraining. Sports Med. 2002;32(2):95-102.
621
38. Weltman A, Weltman JY, Schurrer R, Evans WS, Veldhuis JD, Rogol AD. Endurance
622
training amplifies the pulsatile release of growth hormone: effects of training intensity.
623
Journal of applied physiology. 1992;72(6):2188-96.
624
39. Zinner C, Wahl P, Achtzehn S, Reed JL, Mester J. Acute hormonal responses before and
625
after 2 weeks of HIT in well-trained junior triathletes. International journal of sports
626
medicine. 2014;35(4):316-22.
627
628
629
FIGURE LEGENDS
630
631
FIGURE 1: Study protocol. A 6-week pre-intervention period, consisting of ad libitum LIT and
632
one prescribed interval session each week, in addition to pre-test and randomization (R), was
633
followed by a 12-week intervention period divided in three 4-week cycles with different interval
634
session prescriptions for the increasing HIT (INC) (n=23) decreasing HIT (DEC) (n=20) and
635
mixed HIT (MIX) (n=20) groups. Testing was performed pre-intervention, during week 4, 8 and
636
12. Figure redrawn from Sylta et al (2016) (33).
637
638
FIGURE 2, A-C: Mean (SD) high-intensity training (HIT) duration each week during a 12-week
639
training period in (A) increasing HIT (INC) (N=23), (B) decreasing HIT (DEC) (N=20) and (C)
640
mixed HIT (MIX) (N=20) training group. T=test. See Figure 1 for detailed interval training
641
prescriptions during each cycle. D-L: Mean and 95% CI for delta changes in peak power output
642
(PPO), peak oxygen uptake (V
O2peak) (ml.min-1) and power at 4 mMol.L-1 lactate (Power4mM ) at
643
pre, after 4, 8 and 12 weeks of training in INC (D/G/J), DEC (E/H/K) and MIX (F/I/L) training
644
group, respectively. * P<0.05 for changes from pre.
645
646
FIGURE 3: Mean and 95% CI for delta changes (%) in (A) power at 4 mMol.L-1 lactate
647
(Power4mM) and (B) peak oxygen uptake (V
O2peak) (ml.min-1) in increasing HIT (INC), decreasing
648
HIT (DEC) and mixed HIT (MIX) training group, during cycle 1 (week 1-4), cycle 2 (week 5-8)
649
and cycle 3 (week 9-12), respectively. Values inside boxes represent interval training
650
prescriptions during each cycle. * P<0.05 for changes within cycle.
651
652
FIGURE 4: Mean change in blood hormones at pre, after 4, 8 and 12 weeks of training in
653
increasing HIT (INC) (N=9), decreasing HIT (DEC) (N=10) and mixed HIT (MIX) (N=10)
654
training group, respectively. * P<0.05 for changes from last observation, # P<0.05 for changes
655
from pre.
656
657
TABLE 1. Physiological and perceptual responses during interval sessions executed as 4x16,
1
4x8 and 4x4 min during a 12 week intervention period.
2
4x16 min
4x8 min
4x4 min
P-value*
Power (W)§
276 (25)
308 (29)
342 (33)
<0.001
Power (W.kg-1)§
3.5 (0.4)
3.9 (0.4)
4.3 (0.4)
<0.001
Percent of PPO (%)§
65 (4)
71 (4)
80 (4)
<0.001
Percent of Power4mM (%)§
97 (8)
106 (8)
118 (9)
<0.001
Percent of Power40min (%)§
95 (5)
106 (5)
117 (6)
<0.001
Blood lactate (mMol.L-1)#
4.7 (1.6)
9.2 (2.4)
12.7 (2.7)
<0.001
Interval bout HRmean (% HRpeak)§
86 (3)
88 (2)
89 (2)
<0.001
Interval bout HRpeak (% HRpeak)§
89 (2)
91 (2)
94 (2)
<0.001
RPE (6-20)§
15.0 (1.1)
16.2 (0.8)
17.1 (0.9)
<0.001
sRPE 30min post session (1-10)β
6.3 (1.0)
6.9 (1.0)
7.7 (1.2)
<0.001
All values are calculated as the mean of means (SD) of up to 24 training sessions in 63
3
subjects. § All values of power, heart rate (HR) and rate of perceived exertion (RPE) represent
4
a mean of all 4 interval laps. Reference values for Power at 4 mMol.L-1 blood lactate
5
(Power4mM) are mean of 4 tests performed at pre, week 4, 8, and 12. Reference value for 40
6
min time-trial power (Power40min) is mean of pre and post test results. # Blood lactate was
7
measured randomly among a subset of 56 subjects after interval lap 3 and 4, and a total of 531
8
samples (~10 per participant) were collected. β Session RPE (sRPE) was quantified 30 min
9
post exercise. * One way repeated measure ANOVA comparing responses to HIT
10
prescriptions. There were no significant differences in responses across intervention groups,
11
although different interval prescriptions (4x16 and 4x4 min) were performed in opposite
12
sequence (cycle 1 and 3) for INC and DEC, respectively.
13
14
Table 1
TABLE 2: Pre-intervention values and absolute mean changes from last cycle in performance
1
and physiological variables in the Increasing HIT (INC) (N=23), Decreasing HIT (DEC)
2
(N=20) and Mixed HIT (MIX) (N=20) groups during the 12 week intervention period. All
3
values are mean (95% CI).
4
Mean
Pre
Mean change,
Pre-Cycle 1
Mean change,
Cycle 1-2
Mean change,
Cycle 2-3
Power4mM (W)
INC
DEC
MIX
277 (266, 287)
283 (274, 293)
287 (273, 302)
16 (6, 25)*
5 (-5, 15)
8 (0, 17)
2 (-4, 9)
5 (-3, 14)
-2 (-8, 4)
-2 (-10, 6)
4 (-5, 13)
-1 (-11, 8)
𝑉
O2peak (ml.min-1)
INC
DEC
MIX
4947 (4749, 5146)
4794 (4594, 4994)
4858 (4609, 5108)
196 (77, 316)*
83 (-51, 217)
137 (9, 266)*
97 (-18, 211)
48 (-124, 220)
-7 (-148, 134)
-10 (-142, 121)
71 (-118, 260)
10 (-183, 202)
Gross eff. (%)
INC
DEC
MIX
18.8 (18.4, 19.3)
19.3 (18.9, 19.7)
19.1 (18.7, 19.5)
-0.3 (-0.7, 0.2)
-0.2 (-0.7, 0.3)
0.1 (-0.4, 0.5)
-0.3 (-0.7, 0.2)
-0.1 (-0.6, 0.3)
-0.4 (-0.9, 0.2)
0.0 (-0.4, 0.4)
0.0 (-0.4, 0.4)
0.0 (-0.5, 0.6)
PPO (W)
INC
DEC
MIX
418 (403, 433)
414 (401, 427)
417 (402, 433)
22 (14, 30)*
21 (8, 34)*
14 (1, 27)*
3 (-6, 11)
3 (-7, 12)
1 (-10, 12)
4 (-4, 12)
-1 (-7, 6)
6 (-10, 21)
Power30s (W)
INC
DEC
MIX
852 (827, 878)
824 (787, 862)
820 (773, 867)
10 (-5, 24)
21 (1, 42)*
19 (-8, 45)
0 (-13, 12)
0 (-11, 11)
0 (-17, 16)
1 (-15, 17)
0 (-15, 15)
-4 (-22, 14)
Power4mM = Power corresponding to 4mMol.L-1 lactate, 𝑉
O2peak = Peak oxygen uptake, PPO =
5
Peak Power Output, Power30s = Mean power during 30 s all out test. * = P<0.05 vs. last cycle.
6
There were no sig. between-group differences in relation to pre values or mean changes.
7
Table 2
