Acute effects of repeated cycling sprints in hypoxia induced by voluntary hypoventilation

Abstract

Purpose:
This study aimed to investigate the acute responses to repeated-sprint exercise (RSE) in hypoxia induced by voluntary hypoventilation at low lung volume (VHL).

Methods:
Nine well-trained subjects performed two sets of eight 6-s sprints on a cycle ergometer followed by 24 s of inactive recovery. RSE was randomly carried out either with normal breathing (RSN) or with VHL (RSH-VHL). Peak (PPO) and mean power output (MPO) of each sprint were measured. Arterial oxygen saturation, heart rate (HR), gas exchange and muscle concentrations of oxy-([O2Hb]) and deoxyhaemoglobin/myoglobin ([HHb]) were continuously recorded throughout exercise. Blood lactate concentration ([La]) was measured at the end of the first (S1) and second set (S2).

Results:
There was no difference in PPO and MPO between conditions in all sprints. Arterial oxygen saturation (87.7 ± 3.6 vs 96.9 ± 1.8% at the last sprint) and HR were lower in RSH-VHL than in RSN during most part of exercise. The changes in [O2Hb] and [HHb] were greater in RSH-VHL at S2. Oxygen uptake was significantly higher in RSH-VHL than in RSN during the recovery periods following sprints at S2 (3.02 ± 0.4 vs 2.67 ± 0.5 L min(-1) on average) whereas [La] was lower in RSH-VHL at the end of exercise (10.3 ± 2.9 vs 13.8 ± 3.5 mmol.L(-1); p < 0.01).

Conclusions:
This study shows that performing RSE with VHL led to larger arterial and muscle deoxygenation than with normal breathing while maintaining similar power output. This kind of exercise may be worth using for performing repeated sprint training in hypoxia.

Full text

Vol.:(0123456789)
1 3
Eur J Appl Physiol
DOI 10.1007/s00421-017-3729-3
ORIGINAL ARTICLE
Acute effects ofrepeated cycling sprints inhypoxia induced
byvoluntary hypoventilation
XavierWoorons1,2 · PatrickMucci1· JulienAucouturier1· AgatheAnthierens1·
GrégoireP.Millet3
Received: 12 June 2017 / Accepted: 21 September 2017
© Springer-Verlag GmbH Germany 2017
exercise. The changes in [O2Hb] and [HHb] were greater in
RSH-VHL at S2. Oxygen uptake was significantly higher
in RSH-VHL than in RSN during the recovery periods fol-
lowing sprints at S2 (3.02 ± 0.4 vs 2.67 ± 0.5L min−1 on
average) whereas [La] was lower in RSH-VHL at the end
of exercise (10.3 ± 2.9 vs 13.8 ± 3.5mmol.L−1; p < 0.01).
Conclusions This study shows that performing RSE with
VHL led to larger arterial and muscle deoxygenation than
with normal breathing while maintaining similar power out-
put. This kind of exercise may be worth using for performing
repeated sprint training in hypoxia.
Keywords Hypoventilation· Hypoxia· Hypoxemia·
Repeated sprints· Exercise
Abbreviations
[HHb] Muscle concentrations of
deoxyhaemoglobin/myoglobin
[La] Blood lactate concentration
[O2Hb] Muscle concentrations of oxyhaemoglobin/
myoglobin
[tHb] Total haemoglobin/myoglobin
ANOVA Analysis of variance
FRC Functional residual capacity
HR Heart rate
MPO Mean power output
NB Normal breathing
NIRS Near-infrared spectroscopy
PPO Peak power output
RPE Rating of perceived exertion
RSE Repeated-sprint exercise
RSH Repeated sprints in hypoxia
RSH-VHL Repeated sprints in hypoxia induced by vol-
untary hypoventilation at low lung volume
RSN Repeated sprints in normoxia
Abstract
Purpose This study aimed to investigate the acute
responses to repeated-sprint exercise (RSE) in hypoxia
induced by voluntary hypoventilation at low lung volume
(VHL).
Methods Nine well-trained subjects performed two sets of
eight 6-s sprints on a cycle ergometer followed by 24s of
inactive recovery. RSE was randomly carried out either with
normal breathing (RSN) or with VHL (RSH-VHL). Peak
(PPO) and mean power output (MPO) of each sprint were
measured. Arterial oxygen saturation, heart rate (HR), gas
exchange and muscle concentrations of oxy-([O2Hb]) and
deoxyhaemoglobin/myoglobin ([HHb]) were continuously
recorded throughout exercise. Blood lactate concentration
([La]) was measured at the end of the first (S1) and second
set (S2).
Results There was no difference in PPO and MPO
between conditions in all sprints. Arterial oxygen satura-
tion (87.7 ± 3.6 vs 96.9 ± 1.8% at the last sprint) and HR
were lower in RSH-VHL than in RSN during most part of
Communicated by Susan Hopkins.
* Xavier Woorons
xavier.woorons@gmail.com
1 URePSSS, Unité de Recherche Pluridisciplinaire Sport,
Santé, Société, Faculté des Sciences du Sport et de l’EP,
University ofLille, 9 rue de l’Université, EA-7369,
59790Ronchin, France
2 ARPEH, Association pour la Recherche et la
Promotion de l’Entraînement en Hypoventilation,
18RueSaintGabriel59800Lille, France
3 ISSUL, Institute ofSports Sciences, University ofLausanne,
BuildingGeopolis,CampusDorigny1015Lausanne,
Switzerland

Eur J Appl Physiol
1 3
SpO2 Arterial oxygen saturation
.
VE
Expired ventilation
.
V
E∕
.
VCO
2
Ventilatory equivalent for carbon dioxide
VHL Voluntary hypoventilation at low lung
volume
.
VO2
Oxygen uptake
.
V
O
2max
Maximal oxygen uptake
Introduction
Repeated-sprint exercise (RSE), i.e., short-duration sprints
(< 10s) interspersed with brief recovery periods (< 60s),
is a common practice in most team and racket sports. This
kind of exercise is effective to improve the ability to recover
and to reproduce performance in subsequent sprints, namely
repeated-sprint ability (Bishop etal. 2011). Recently, several
studies demonstrated that training by performing repeated
sprints in hypoxia (RSH) could be even more effective than
the same type of training in normoxia (RSN) for improv-
ing peak and/or mean power output during repeated-sprint
tests (Brocherie etal. 2015; Faiss etal. 2013a, 2015; Gat-
terer etal. 2014; Kasai etal. 2015; Millet etal. 2013). A
recent meta-analysis confirmed that RSH is more efficient
than RSN to improve mean repeated-sprint performance,
while additional positive (but non-significant) effects on
best repeated-sprint and maximal oxygen uptake (
.
V
O
2max)
were also reported (Brocherie etal. 2017a). The postulated
underlying mechanisms (i.e., compensatory vasodilatation,
microvascular oxygen delivery [fast-twitch fibres]) and spe-
cific skeletal muscle molecular adaptations mediated by the
oxygen sensing pathway (Brocherie etal. 2017b) are likely
to be fibre-type and intensity dependant (Faiss etal. 2013b).
So far, the vast majority of the studies that investigated
the effects of RSH have used normobaric hypoxia by reduc-
ing the ambient fraction of oxygen (Brocherie etal. 2017a).
However, it is remarkable that in one study, hypoxia was
induced by voluntary hypoventilation at low lung volume
(VHL) that is by performing breath holdings near functional
residual capacity (Trincat etal. 2017). It is also noticea-
ble that in this study, like in almost all RSH studies, the
repeated-sprint performance enhancement was larger than
in RSN. Over the last 10years, it has repeatedly and clearly
been shown that exercising with VHL could lead to severe
hypoxemia (Woorons etal. 2007, 2008, 2011, 2014) and
consequently to muscle tissue deoxygenation (Woorons etal.
2010) and could, therefore, represent a valuable and practi-
cal way to train under simulated hypoxic conditions.
Until recently, VHL had always been performed at sub-
maximal exercise intensities (65–80%
.
VO2max
). In addi-
tion to the hypoxemic effect, higher carbon dioxide con-
centrations (i.e., hypercapnic effect), lower level of pH
and increased lactate concentration have consistently been
reported when exercising with VHL at moderate intensity,
compared with the same exercise with normal breathing
(NB) (Woorons etal. 2007, 2010, 2011, 2014). Thus, the
main feature of this type of exercise is to provoke a com-
bined lactic and respiratory acidosis as a result of both the
hypoxic and hypercapnic effects. The latest studies that
investigated the effects of VHL training at maximal velocity
(Trincat etal. 2017) or at supramaximal intensity (Woorons
etal. 2016) showed that it is possible to have a different
approach to this kind of training. Nevertheless, little is
known on the acute physiological effects of an approach to
VHL training involving high exercise intensities. In particu-
lar, it is questionable whether, in these conditions, metabolic
acidosis is greater than during the same exercise with NB.
To our knowledge, the first and only training study on the
effects of RSH induced by VHL (RSH-VHL) was conducted
in swimming (Trincat etal. 2017). Yet it could be interesting
to define the acute physiological consequences of RSH-VHL
in land-based activities (e.g., cycling or running) which are
the core component of conditioning in team-sport players.
Furthermore, it was recently shown that single-sprint per-
formance is generally not affected by hypoxia whereas larger
alterations in repeated-sprint ability (i.e., with earlier and
larger performance decrements) occur at altitudes above
3000m (or inspired fraction of oxygen < 14.5%), when
compared to normoxic conditions (Girard etal. 2017). This
is unknown for RSH-VHL and is of practical importance.
Thus, the main objective of the present study was to ana-
lyse the acute effects of RSH-VHL on performance, muscle
and blood oxygenation as well as the level of blood lactate
concentration. In these conditions, we hypothesized that
the decrement in power output could be greater than during
RSN. We also assumed that RSH-VHL could lead to severe
blood and muscle deoxygenation during most of the exer-
cise. Finally, we postulated that RSH-VHL could provoke an
increased blood lactate concentration, compared with RSN.
Methods
Subjects
Nine well-trained subjects, eight men and one woman, were
recruited for this study. They were all non-smokers and sea-
level residents. None of them reported to have recently been
exposed to altitude (i.e., > 500m above sea level). Four of
the subjects (including the woman) were cyclists who trained
4–5 times per week and regularly participated in competition
at a regional level. The five other subjects were involved in
team sports (four basketball players and one rugby player)
with 3–4 training sessions/week. Their physical charac-
teristics (mean ± SD) were age 27.2 ± 9.3years, height
179.0 ± 8.4cm, and body mass 74.7 ± 9.6kg. Subjects were

Eur J Appl Physiol
1 3
instructed not to participate in any strenuous exercise other
than prescribed by the study protocol from the 48h before
the start of the experiment until the end of the study. They
were also instructed to be adequately hydrated and to not
have eaten for 3h prior to each test. The participants gave
their written informed consent to participate in this study,
which was approved by the local Research Ethics Commit-
tee and complied with the Declaration of Helsinki (2008).
Protocol
Before the start of the experiment, the subjects participated
in a first session to familiarize with the testing procedures
as well as the VHL technique. This breathing technique has
already been used in several studies (Trincat etal. 2017;
Woorons etal. 2008, 2010, 2014, 2016) and is recognized
as one of the various hypoxic training methods (Girard
etal. 2017). It consists of repeating short bouts of breath
holding while exercising. Unlike a simple apnoea, in which
breath is generally held at high lung volume (i.e., total
lung capacity) and for as long as possible, VHL requires
a rigorous control of breathing to obtain a hypoxic effect
while avoiding asphyxia. It stands on four phases: inhala-
tion, exhalation, breath holding and a second exhalation.
After inhalation, the first exhalation is carried out naturally,
without forcing in order to reach the functional residual
capacity (FRC). While FRC represents about one-half of
the total lung capacity at rest (Quanjer etal. 1993), it is
only about one-third of this lung volume during exercise,
in particular when intensities are high. Indeed, FRC has
been shown to be reduced with the increase in exercise
intensity (Sharratt etal. 1987). In the present study, one
can therefore consider that hypoventilation was performed
at (relatively) low lung volume. After the first exhalation, a
breath holding of a few seconds must be performed and is
followed by a second exhalation down to residual volume
which aims at the evacuation of the carbon dioxide accu-
mulated within the lungs.
After the familiarization phase, the subjects participated
in two testing sessions which consisted of performing two
sets of eight “all-out” 6-s sprints on a cycle ergometer
(894E, Monark Exercise AB, Vansbro, Sweden). The 6-s
sprints were separated by 24s of inactive recovery (depar-
ture every 30s) and 3min of passive rest were observed
between the two sets. Before starting the first set, subjects
performed a 5-min warm-up at low intensity followed by a
single 6-s sprint. Exercise began after a 5-min period of rest.
At the first sprint of set 1, subjects were required to achieve
at least 95% of the mean power output (MPO) reached in
the single sprint as recommended (Girard etal. 2011). If
not, they had to restart the set again after a 5-min period of
rest. Sprints were carried out with NB in one of the two test-
ing sessions (RSN) and with VHL in the other (RSH-VHL).
During RSH-VHL, the subjects were required to do a nor-
mal exhalation at the start of each sprint, then to hold their
breath until the end of the 6-s exertion and finally to perform
a second exhalation to empty the remaining air from the
lungs. In the pre-experimental phase, we tested whether it
could be possible to hold breath at the residual volume to
obtain a greater stimulus. It appeared that the manoeuvre
could not be conducted that way. In this condition, subjects
were not able to hold their breath for more than 4–5s and
made micro-inhalations which prevented SpO2 to fall down
below 90%.
The order of exercise conditions was randomized, but
counterbalanced so that five subjects carried out the first
testing session in RSH-VHL and the four others in RSN.
The two sessions took place at the same time of the day and
were separated by 48h. The mechanical resistance was set
at 0.075kg−1 body mass (measured at the first session) and
subjects were instructed to stay seated. Computer-generated
audio signals provided a 5-s countdown to the start of each
sprint. All participants were strongly encouraged during the
test to maintain their maximum pedal rate.
Measurements
Performance
For each 6-s sprint, peak power output (PPO) and MPO
were measured with the Monark Anaerobic Test software
(Monark, Varberg, Sweden). For both sets, we also assessed
fatigue by calculating (in absolute value) the percentage dec-
rement score as follows:
where total sprint MPO is the sum of sprint MPO from all
sprints of the set and ideal sprint MPO = number of sprints
(i.e., 8) × highest sprint MPO of the set.
This formula has been found to be the most valid and
reliable method for quantifying fatigue in tests of multiple-
sprint performance (Glaister etal. 2008).
Arterial oxygen saturation andheart rate
Arterial oxygen saturation (SpO2) and heart rate (HR) were
continuously measured during set 1 and set 2 with the pulse
oximeter Nellcor N-595 (Pleasanton, CA, USA) and with
the adhesive forehead sensor Max-Fast (Nellcor, Pleasanton)
which was applied above the right orbital area. This sen-
sor has already been used during intense cycling exercise
in hypoxia (Amann etal. 2007; Romer etal. 2007) and has
|(100 ×(total sprint MPO of the set∕
ideal sprint MPO of the set ideal sprint MPO of the set
))
−100|,

Eur J Appl Physiol
1 3
been shown to provide accurate and reliable estimations of
both arterial oxygen saturation and HR (Fernandez etal.
2007; Schallom etal. 2007). An adjustable headband was
placed over the forehead sensor to ensure gentle, consistent
pressure on the sensor device. Arterial oxygen saturation
and HR were recorded in real time every 2s by the oximeter
and collected using a data acquisition system (Score Analy-
sis Software, Nellcor, Pleasanton). Data were then averaged
and analyzed over 6s, which corresponded to the duration
of each sprint. It is important to note that in this protocol,
SpO2 began to drop at the end or just after the 6-s sprints.
Moreover, the highest values of HR were recorded after each
sprint. For these reasons, we chose to present the results for
the 6-s periods in which the minimum values of SpO2 and
the maximum values of HR were reached.
Gas exchange
Gas exchange was continuously recorded during the whole
exercise through a breath-by-breath portable system (K4b2,
Cosmed, Rome, Italy). Before the beginning of exercise, we
performed the standardized calibration procedures as rec-
ommended by the manufacturer. These included air calibra-
tion, turbine calibration with a standard 3000-mL syringe,
gas calibration with a certified commercial gas preparation
(oxygen: 16%, carbon dioxide: 5%) and delay calibration
to ensure accurate readings during the testing and to check
the alignment between the gas flow and gas concentrations.
The breath-by-breath measurements were performed for
tidal volume, breathing frequency, expired ventilation (
.
VE),
oxygen uptake (
.
V
O
2
)
,
carbon dioxide production, end-tidal
oxygen and carbon dioxide end-tidal pressures. The ventila-
tory equivalent for oxygen and carbon dioxide were calcu-
lated. Since gas exchange could obviously not be measured
during sprints with breath holdings, data were analysed in
the last 20s of the recovery periods.
Near‑infrared spectroscopy
Muscle oxygenation was assessed using a near-infrared
spectroscopy (NIRS) technique which was well described
elsewhere (Boushel and Piantadosi 2000). The NIRS device
(Portamon Artinis, Zetten, The Netherlands) was used to
measure changes in muscle oxygenation by placing a triple
optode sensor on the left-leg vastus lateralis muscle (at mid-
thigh) with an interoptode spacing of 40mm. The probe
was attached to the skin with double-sided tape and firmly
fastened with an opaque cotton elastic band wrapped around
subjects’ thigh. Position of the probe was marked at the first
testing session with a permanent pen for accurate reposi-
tioning at the second testing session. A standard differential
pathlength factor (DPF) of 4.0 was used in lack of any clear
standard value for human quadriceps muscle during cycling
sprints (Racinais etal. 2007). All signals were recorded with
a sampling frequency of 20Hz. They were then down sam-
pled at 1Hz to remove possible artefacts and smooth the
pedalling-induced perturbations. Muscle concentrations of
deoxyhaemoglobin/myoglobin ([HHb]) and total haemo-
globin/myoglobin ([tHb]) were recorded. We also chose to
analyse the muscle concentrations of oxyhaemoglobin/myo-
globin ([O2Hb]) even though it has been reported that this
factor might be confounded by rapid volume changes during
sprints (Grassi etal. 2003). For the two sets, the change (Δ)
in [HHb], [tHb] and [O2Hb] were calculated from the resting
values recorded over the 2min following the warm-up. Mus-
cle measurements were, therefore, normalized from these
recordings (arbitrarily defined as 0µm). For all variables,
data were averaged over 6s and analysed for the greater
changes obtained over a 6-s period during or just after each
sprint.
Rating ofperceived exertion (RPE) andblood lactate
concentration [La]
In both exercise conditions, just after the end of each set,
RPE was obtained using the Borg scale (range 0–10). In
addition, a blood sample was taken from the earlobe of the
subjects 90s after the end of the first (S1) and the second
set (S2) as well as 180s after the end of S2 to obtain [La]
(Lactate Pro, Arkray, Japan). The highest value of the two
samples collected at the end of S2 was kept for the analyses.
Statistics
All the results are expressed as mean ± SD. To determine
whether there was a difference in PPO, MPO, SpO2, HR,
.
VO2
and NIRS data between exercise conditions for each
sprint and within each set, we performed two-way analyses
of variance (ANOVA) for repeated measures. We performed
one-way ANOVA for repeated measures to determine
whether there were differences in the same data between
sprints of S1 and S2 within each condition. To compare the
percentage decrement score, RPE, [La] and gas exchange
data (except
.
V
O
2
) between and within conditions at the end
of each set (i.e., over the last 20s of the recovery period of
the 8th repetition for gas exchange), we also used two-way
ANOVA for repeated measures. When a significant effect
was found, the Bonferroni (two-way ANOVA) or the Tukey
post hoc test (one-way ANOVA) was carried out to localize
the differences. We used the Student t test to compare the
single sprint MPO of the first sprint between the two exer-
cise conditions. The relationships between the differences
(Δ) in
.
VO2
, HR, SpO2, [La], Δ[HHb] and Δ[O2Hb] between
RSH-VHL and RSN at the end of S2 were analysed by per-
forming Person linear regressions. In particular we tested the
following relationships: Δ
.
V
O
2
/Δ[La], ΔSpO2/Δ(Δ[HHb]),

Eur J Appl Physiol
1 3
ΔSpO2/Δ(Δ[O2Hb]), ΔSpO2/ΔHR, ΔSpO2/Δ
.
V
O
2
. We also
analysed the relationships between [La] and
.
VE∕
.
V
CO2
within each exercise condition. All analyses were made
using Sigmastat 3.5 software (Systat Software, CA, USA).
Null hypothesis was rejected at p < 0.05.
Results
Performance
Single-sprint MPO was not different between RSH-VHL
(858 ± 138W) and RSN (879 ± 154 W) (p = 0.20). Mean
power output of the first sprint of S1 relative to single-sprint
MPO was 96.5 ± 1.1% in RSH-VHL and 98.3 ± 2.6% in RSN.
There was no difference between conditions in both PPO and
MPO either at S1 or S2 (Fig.1a, b). There was also no differ-
ence between sprints of same number at S1 and S2 in the two
exercise conditions. The percentage decrement score was
not different between RSH-VHL and RSN at S1 (6.2 ± 2.2%
vs 7.7 ± 4.1%; p = 0.26) and was lower in RSH-VHL than in
RSN at S2 (9.5 ± 3.8% vs 12.1 ± 4.8%; p < 0.01). Finally, the
percentage decrement score was higher at S2 than at S1 in
both conditions (RSN: p < 0.01; RSH-VHL: p = 0.03).
Arterial oxygen saturation andHR
Arterial oxygen saturation was significantly lower in RSH-
VHL than in RSN from the 2nd sprint of both S1 and S2
(Fig.2a). Heart rate was lower in RSH-VHL than in RSN
from the 4th sprint of S1 and from the 2nd sprint of S2
(Fig.2b). The one-way ANOVA for repeated measures
showed a significant effect for SpO2 and HR (p < 0.01), but
there was no difference in SpO2 between sprints of same
number at S1 and S2 in both conditions. On the other hand,
HR was higher at all sprints of S2 compared with sprints of
same number at S1 during RSN and higher from the 1st to
the 6th sprint of S2 during RSH-VHL. There was a signifi-
cant correlation between ΔSpO2 and ΔHR at the end of S2
(r = 0.84, p < 0.01). The absolute and relative times spent at
different levels of SpO2 are presented in Table1.
Gas exchange
Oxygen uptake was not different between conditions at S1
and higher in RSH-VHL than in RSN from the 2nd sprint to
the 8th sprint of S2 (Fig.2c). Furthermore, there was no dif-
ference in
.
V
O
2
between sprints of same number at S1 and S2
in both conditions except for the 1st sprint of RSH-VHL in
which
.
V
O
2
was higher at S2. Breathing frequency, the ven-
tilatory equivalent for oxygen and end-tidal oxygen pressure
were lower whereas tidal volume and end-tidal carbon diox-
ide pressure were higher in RSH-VHL than in RSN at the
8th repetition of both S1 and S2 (Table2). Carbon dioxide
production was higher and
.
VE∕
.
V
CO2
lower in RSH-VHL
at the end of S2 only. Minute ventilation was not different
between conditions in both sets. End-tidal carbon dioxide
pressure was lower and
.
V
E∕
.
VCO
2
higher at S2 than at S1
in both conditions whereas carbon dioxide production was
lower at S2 than at S1 in RSN only. There was no difference
between S1 and S2 for all the other ventilatory variables in
the two exercise conditions.
Muscle oxygenation
There was no difference in Δ[O2Hb] and Δ[HHb] between
exercise conditions at S1 (Fig.3a, b) and no difference in
Δ[tHb] both at S1 and S2 (Fig.3c). On the other hand,
compared to normalized resting values, the decrease in
[O2Hb] was greater in RSH-VHL than in RSN at sprints
1, 5, 6, 7 and 8 of S2 whereas [HHb] was higher in RSH-
VHL than in RSN from the 5th to the 8th sprints of S2.
400
500
600
700
800
900
1000
1100
1200
12345678
12345678
600
700
800
900
1000
1100
1200
1300
1234567812345678
Sprint number
Set 1Set 2
MPO (W)PPO (W)
C: 0.26, I: 0.12 C: 0.7, I: 0.37
C: 0.17, I: 0.53 C: 0.12, I: 0.07
(a)
(b)
Fig. 1 Peak power output (PPO) (a) and mean power output (MPO)
(b) achieved at each sprint of the first and the second set of the
repeated-sprint exercise performed with voluntary hypoventilation
at low lung volume (RSH-VHL) and with normal breathing (RSN).
RSH-VHL: white bars; RSN: black bars

Eur J Appl Physiol
1 3
The one-way ANOVA for repeated measures showed a
significant effect for Δ[O2Hb], Δ[HHb] and Δ[tHb] in the
two exercise conditions (p < 0.01). However, there were
no significant differences for these variables between
sprints of same number at S1 and S2 both in RSH-VHL
and in RSN. Figure3d–f show the typical NIRS signals of
[O2Hb], [HHb] and [tHb] recorded during one set of RSH-
VHL and RSN. Finally, we found a positive correlation
between Δ(Δ[O2Hb]) and ΔSpO2 (r = 0.89, p < 0.01) and
an inverse relationship between Δ(Δ[HHb]) and ΔSpO2
(r = − 0.71, p = 0.03) at the end of S2.
Rating ofperceived exertion and[La]
Blood lactate concentration was not different between exer-
cise conditions after S1 (p = 0.45) and was lower in RSH-
VHL than in RSN after S2 (p < 0.01) (Fig.4). There was no
difference in RPE between RSH-VHL and RSN both at S1
(8.9 ± 0.3 vs 8.3 ± 0.7) and S2 (9.4 ± 0.6 vs 9.3 ± 0.6). Blood
lactate concentration and RPE were both significantly higher
at S2 than at S1 in the two exercise conditions (p < 0.05).
There was no relationship between Δ[La] and Δ
.
VO2
at the
end of S2 (r = − 0.12, p = 0.76). On the other hand, [La]
was positively correlated with
.
VE∕
.
V
CO2
at the end of S2
in RSN (r = 0.83, p < 0.01) but not in RSH-VHL (r = 0.41,
p = 0.27).
Discussion
This study was the first to investigate the acute effects of
repeated sprints in hypoxia induced by VHL. It brought
out several major findings. First, performing RSH-VHL
in cycling did not provoke a greater fatigue (i.e., decrease
in power output) than the same exercise performed under
normal breathing conditions. Second, RSH-VHL induced
a greater muscle deoxygenation than RSN in the second
part of exercise. Third, lactate concentration was lower in
RSH-VHL than in RSN after the end of exercise. Finally,
oxygen uptake was significantly higher in RSH-VHL during
the recovery periods following sprints in the second half of
exercise.
The first and unexpected result of the present study was
that PPO and MPO were not different between the two exer-
cise conditions throughout the whole RSE. Yet we postu-
lated a greater decrement in performance in RSH-VHL due
to the combined respiratory and metabolic acidosis that was
consistently reported during exercise with VHL (Woorons
etal. 2007, 2010, 2014). Unlike our hypothesis, the per-
centage decrement score was similar in both exercise condi-
tions at S1 and was surprisingly lower in RSH-VHL at S2.
Noticeably, during RSE performed in moderate normobaric
hypoxia (i.e., below 3000m or inspired fraction of oxygen
> 14.5%), fatigue development was not greater than in nor-
moxic conditions (Girard etal. 2017). On the other hand, in
severe hypoxia (i.e., altitude > 3000m; inspired fraction of
oxygen < 14.5%), the decrease in performance has repeat-
edly been found larger than in normoxia (Billaut etal. 2013;
Brocherie etal. 2016; Morrison etal. 2015). Our results
80
84
88
92
96
100
*****
**
*******
SpO2 (%)
12345678 12345678
120
140
160
180
200
1234567812345678
HR (bpm)
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Sprint number
Set
1S
et 2
VO2(L.min-1)
*****
**
*******
12345678
12345678
†
†
††† †
††
†
†
*
†
*
†
*
†*
†*
†
C+I: < 0.01 C+I: < 0.01
C: 0.04; I: 0.03 C: 0.01; I: 0.24
(a)
(b)
(c)
C: 0.06; I: 0.91 C: < 0.01; I: 0.97
.
Fig. 2 Minimum arterial oxygen saturation (SpO2) (a) and maxi-
mum heart rate (HR) (b) measured at each sprint of set 1 (S1) and
set 2 (S2) of the repeated-sprint exercise performed with voluntary
hypoventilation at low lung volume (RSH-VHL) and with normal
breathing (RSN). c Oxygen uptake (
.
V
O
2
) measured over the last 20s
of the recovery periods of each sprint during RSH-VHL and RSN.
Open circles: RSH-VHL; filled squares: RSN. *Significant difference
with RSN; †significant difference with sprint of same number at S1
(p < 0.05)

Eur J Appl Physiol
1 3
may, however, be taken with care. Although MPO of the
first sprint of S1 was not different between conditions and
reached at least 95% of MPO of the single sprint, a slight
pacing may have occurred in some subjects at the beginning
of RSH-VHL. The fact that during RSH-VHL the subjects
could maintain a power output as high as in RSN is a key
result since the efficiency of RSH relies on the repetition of
maximal intensity efforts in order to recruit and induce phys-
iological adaptation in fast-twitch fibres (Faiss etal. 2013b).
Our second hypothesis that performing a cycling RSE
with VHL could provoke a larger blood and muscle deoxy-
genation than the same exercise with normal breathing is
partly validated. Arterial oxygen saturation decreased to
87% at the end of exercise which represents a severe hypox-
emia (Dempsey and Wagner 1999). This level of blood oxy-
genation is very similar to what has been reported in VHL
studies in different exercise modalities (Woorons etal. 2008,
2011, 2014). Noteworthy, SpO2 values of about 90% have
been reported during RSE performed at simulated altitude
of 1800–2000m (Brocherie etal. 2016; Goods etal. 2014)
whereas in severe hypoxia (i.e., 3000m, inspired fraction of
oxygen = 14%), SpO2 values of 82–85% have been recorded
(Billaut etal. 2013; Goods etal. 2014). Arterial oxygen satu-
ration values as low as 75–78% have even been reported
during RSH at 3000m (Bowtell etal. 2014; Morrison etal.
2015). The greater increase in [HHb] and decrease in [O2Hb]
recorded in the second part of exercise also support the fact
that VHL has an impact on muscle oxygenation and leads
to tissue hypoxia (Woorons etal. 2010). This is of particular
importance since muscle deoxygenation has been shown to
be a paramount physiological response for inducing specific
skeletal muscle adaptations leading to greater resistance to
fatigue and increased performance during repeated sprint
tests after RSH training (Faiss etal. 2013b). The present
results must, however, be pondered. First, the minimum
SpO2 reached at each sprint remained in the range of 88–90%
during most part of RSH-VHL. Furthermore, regarding the
total time spent at low levels of SpO2 (Table1), it is clear
Table 1 Time spent at different
levels of SpO2 during repeated
sprint exercise
SpO2 arterial oxygen saturation, RSH‑VHL repeated-sprint exercise with voluntary hypoventilation at low
lung volume, RSN repeated-sprint exercise with normal breathing
*Significantly different from RSN, values are mean ± SD, p < 0.05
Range of SpO2 (%) Time (s) Time (%)
RSH-VHL RSN RSH-VHL RSN
[98–100] 167.5 ± 99.4* 373.5 ± 139.6 33.9 ± 20.2* 73.7 ± 27.6
[95–97] 175.0 ± 61.4 119.0 ± 121.4 35.3 ± 12.2 23.4 ± 23.9
[92–94] 88.5 ± 40.0* 14.5 ± 22.2 17.9 ± 8.1* 2.9 ± 4.4
[89–91] 39.0 ± 19.8* 0.0 ± 0.0 7.9 ± 4.0* 0.0 ± 0.0
≤ 88 25.5 ± 19.9* 0.0 ± 0.0 5.1 ± 3.9* 0.0 ± 0.0
Table 2 Gas exchange at the
end of S1 and S2
Values are mean ± SD
RSH‑VHL repeated sprints in hypoxia induced by voluntary hypoventilation at low lung volume, RSN
repeated sprints with normal breathing, S1 first set, S2 second set, T time effect, C condition effect, T × C
interaction effect (time × condition), Vt tidal volume, Bf breathing frequency,
.
VE
expired ventilation,
.
V
CO
2
carbon dioxide production,
.
V
E
∕
.
VO
2
ventilatory equivalent for oxygen,
.
V
E
∕
.
VCO
2
ventilatory equivalent
for carbon dioxide, PETCO2 end-tidal carbon dioxide pressure, PETO2 end-tidal oxygen pressure
*Significantly different from the set in RSN, †significantly different from S1 within condition, p < 0.05
% RSH-VHL RSN ANOVA P value
S1 S2 S1 S2 T C T × C
Vt (L) 2.50 ± 0.5* 2.52 ± 0.5* 2.34 ± 0.7 2.15 ± 0.6 0.11 < 0.01 0.02
Bf (breaths.min−1)44.5 ± 5.8* 47.1 ± 9.7* 58.4 ± 17.0 62.9 ± 13.5 0.02 < 0.01 0.68
.
VE
(L.min−1)108.5 ± 16.8 115.8 ± 19.8 120.0 ± 20.5 123.1 ± 21.6 0.09 0.13 0.46
.
V
CO
2
(L.min−1)3.15 ± 0.4 3.01 ± 0.3* 3.06 ± 0.4 2.56 ± 0.4†0.01 0.01 < 0.01
.
V
E
∕
.
VO
2
(L.min−1)35.6 ± 5.1* 37.1 ± 5.1* 43.9 ± 7.6 46.0 ± 8.9 0.14 0.01 0.88
.
V
E
∕
.
VCO
2
(L.min−1)33.6 ± 3.6 37.7 ± 5.8*,† 38.7 ± 4.5 47.2 ± 7.9†< 0.01 < 0.01 0.15
PETO2 (mmHg) 111.5 ± 4.8* 112.3 ± 4.6* 116.9 ± 3.4 118.9 ± 3.2 0.12 < 0.01 0.54
PETCO2 (mmHg) 37.1 ± 3.7* 33.8 ± 4.2*,† 34.1 ± 2.9 28.9 ± 4.5†< 0.01 < 0.01 0.18

Eur J Appl Physiol
1 3
that the hypoxic dose was low when performing RSH-VHL,
particularly if we use the new metrics based on the time
spent at low saturation level (Millet etal. 2016). It is also
important to note that there was a large interindividual vari-
ability for arterial desaturation. While SpO2 dropped down
below 85% in some subjects (lowest SpO2 recorded = 80%),
it remained above 90% in others. Finally, it is noticeable that
muscle tissue deoxygenation did not occur during the entire
exercise. Altogether, it is therefore questionable whether or
not RSH-VHL could lead to the same physiological adapta-
tions as RSH performed in ambient hypoxia.
The lower [La] in RSH-VHL at the end of exercise rep-
resents another surprising result. So far, the majority of the
studies on VHL exercise have reported an increased glyco-
lytic activity and higher [La] compared with exercise with
NB, mainly because of the hypoxemic state (Woorons etal.
2010, 2014). However, in these studies, exercise was carried
out at submaximal intensity. One may argue that the room
for increased anaerobic glycolysis is greater when exercise is
-10
-5
0
5
10
15
20
25
12345678 12345678-15
-10
-5
0
5
10
15
20
-5
0
5
15
10
20
25
30
-30
-25
-20
-15
-10
-5
0
5
[tHb] (µm)
Sprint number
Set 1Set 2
30 60 90 120 150 180210 240 270
Time (s)
10
15
20
25
30
35
40
1234567812345678
[HHb] (µm)
*
*
*
*
5
45
-40
-35
-30
-25
-20
-15
-10
-5
1234567812345678
[O2Hb] (µm)
*** **
-45
C: 0.66, I: 0.94 C: 0.67, I: 0.08
C: 0.10, I: 0.24 C: 0.03, I: 0.39
C: 0.12, I: 0.70 C: 0.04, I: 0.35
(a)
(b)
(c)
(d)
(e)
(f)
30 60 90 120 150 180210 240 270
30 60 90 120 150 180210 240 270
[O2Hb] (µm)
[HHb] (µm)
[tHb] (µm)
Fig. 3 Changes in concentrations of oxyhaemoglobin/myoglobin
([O2Hb]) (a), deoxyhaemoglobin/myoglobin ([HHb]) (b) and total
haemoglobin/myoglobin ([tHb]) (c) measured by near-infrared spec-
troscopy (NIRS) at each sprint of set 1 and set 2 of the repeated-
sprint exercise performed with voluntary hypoventilation at low lung
volume (RSH-VHL, open circles) and with normal breathing (RSN,
filled squares). *Significant difference with RSN (p < 0.05). Typical
NIRS signals (after treatment) of [O2Hb] (d), [HHb] (e) and [tHb]
(f) during one set in RSH-VHL (dashed line) and RSN (solid line)
expressed as changes in µM from resting baseline set to 0µM

Eur J Appl Physiol
1 3
carried out at submaximal rather than at maximal intensity.
Still, it has been found that RSE performed at different levels
of normobaric hypoxia could lead to higher [La] than the
same exercise in normoxia (Bowtell etal. 2014). Thus, in the
present study, we could expect at least a maintenance of [La]
in RSH-VHL compared with RSN. Considering that power
output was not different between exercise conditions and that
[La], and maybe the contribution of anaerobic glycolysis was
lower in RSH-VHL, our findings raise the question of how
energy was supplied during this specific exercise.
During RSE, energy supply is mainly ensured by phos-
phocreatine (Bogdanis etal. 1996). However, both anaero-
bic and aerobic glycolysis have been reported to play a role,
in particular in the first part of RSE for the former and in the
second part for the latter (Girard etal. 2011). The reduced
[La] after RSH-VHL may not reflect a lower contribution
from anaerobic glycolysis. In this regard, the measurements
of gas exchange were of a great interest since they revealed
a markedly greater
.
VO2
in RSH-VHL during the recovery
periods following sprints at S2. This is an important finding
because an enhanced oxidative pathway is essential for the
maintenance of high exercise intensities and for reducing
fatigue during RSE (Bishop etal. 2011). Thus, the lower
[La] in RSH-VHL could be explained either by an enhanced
lactate oxidation or a greater phosphocreatine resynthesis
during the resting periods between sprints.
Even though the higher
.
VO2
during the recovery periods
of RSH-VHL is an interesting finding, it was not so sur-
prising since this phenomenon had already been reported in
a previous VHL study (Woorons etal. 2011). The authors
hypothesized that it could have been caused by a greater
work and consequently a larger oxygen cost of respiratory
muscles during the hyperventilation phase following the
periods with VHL. They also postulated that the elevated
.
VO2
could be the consequence of the oxygen debt contracted
during the breath holding periods. Since
.
VE
was not dif-
ferent between exercise conditions in the present study, the
first assumption can be rejected whereas the “oxygen debt”
hypothesis represents a likely explanation for the excess
oxygen consumption recorded after the 6-s sprints at S2
(Gaesser and Brooks 1984).
The mechanism by which
.
VO2
was increased during RSH-
VHL cannot be clearly established with the present data. An
improved blood perfusion is unlikely since Δ[tHb] was not
different between exercise conditions. On the other hand, the
greater Δ[HHb] in the second part of S2 suggests that a better
oxygen extraction may have occurred in RSH-VHL. [HHb] is
closely associated with changes in venous oxygen content and
has been shown to be a reliable estimator of changes in fractional
oxygen extraction (Grassi etal. 2003). However, one could argue
that the greater Δ[HHb] in RSH-VHL was a compensation for
the lower SpO2, as suggested by the significant inverse correla-
tion between Δ[HHb] and ΔSpO2. Therefore, we may assume
that the increase in
.
VO2
was partly the consequence of a greater
cardiac output as already reported (Woorons etal. 2011). Con-
sidering that HR was lower in RSH-VHL than in RSN during
most part of exercise, a marked increase in stroke volume may
have occurred in the former condition like what happens during
moderate exercise with VHL (Woorons etal. 2011).
The lower HR in RSH-VHL was not really expected
either. So far, HR has been found unchanged or even
increased either during or after moderate VHL exercise not
exceeding 5min (Woorons etal. 2007, 2011). The author
attributed this augmentation to the hyperventilation phase
that follows the period with VHL. The same phenomenon
did not occur in this study since
.
VE
reached high levels in
both exercise conditions and was not different between RSH-
VHL and RSN. Noteworthy a decrease in HR has already
been found during a short-lasting supramaximal cycling
exercise performed with apnea (Ahn etal. 1989). These
authors postulated that this decrease could be due to an arte-
rial chemoreceptor mechanism under the effect of the fall in
arterial oxygen saturation and consequently hypoxia. Based
on the significant positive correlation between ΔSpO2 and
ΔHR, it is possible that the same physiological mechanism
occurred in the present study.
While
.
VE
was not different between exercise conditions
after the last sprint of each set, two observations are worth
mentioning. First the ventilatory pattern was not the same in
both conditions. RSN was characterized by a high breathing
frequency whereas tidal volume was greater during RSH-
VHL. This finding may not be trivial since one could specu-
late that large inspirations might lead to an increased stroke
volume through a “pump effect” and might therefore partly
explain the higher
.
VO2
in RSH-VHL (Woorons etal. 2010).
Second, when pertained to the metabolic oxygen demand,
.
VE
was much higher in RSN than in RSH-VHL as indicated
0
2
4
6
8
10
12
14
16
18
[La] (mmol.L-1)
Set 1Set 2
*†
†
C: 0.01
I: 0.03
T: < 0.01
Fig. 4 Blood lactate concentrations at the end of the first and the
second set of the repeated-sprint exercise performed with voluntary
hypoventilation at low lung volume (white bars) and with normal
breathing (black bars). *Significant difference with RSN; †significant
difference with S1 (p < 0.05)

Eur J Appl Physiol
1 3
by the greater ventilatory equivalents for oxygen and car-
bon dioxide in the former condition. This phenomenon is in
accordance with the fact that RSN generated a greater lactic
acidosis than RSH-VHL at the end of exercise. Hyperventi-
lation is a well-known physiological response to eliminate
the excess H+. The metabolic acidosis-induced hyperventila-
tion in RSN is reinforced by the strong correlation between
[La] and
.
VE∕
.
V
CO2
for this exercise condition and that was
absent in RSH-VHL.
The fact that the physiological parameters were not meas-
ured at the same time during exercise constitutes the main
limitation of the present study. Consequently, some of the
results may appear difficult to reconcile and in contradiction
with what has been reported so far. For instance, RSH-VHL
elicited hypoxaemia and tissue deoxygenation whereas the
results show normocapnic conditions and lower [La]. How-
ever, unlike blood and muscle oxygenation, gas exchange
could not be measured during sprints because of the breath
holding. Data, therefore, show what happened during the
recovery periods. Yet it is remarkable that the partial pres-
sure of carbon dioxide rapidly returns to normal conditions
after VHL exercise (Woorons etal. 2011). It is likely that the
6-s sprints with VHL elicited hypercapnia since hypoventila-
tion systematically leads to elevated carbon dioxide partial
pressure (West 1997). The lower [La] recorded after RSH-
VHL probably also reflects what occurred during the recov-
ery periods with duration four times longer than the sprint-
ing bouts (ratio 1:4) and with
.
VO2
that was much higher
than in RSN. To overcome these limitations, it would be
interesting to take a blood sample just after the end of the
sprints through catheterization, to rapidly obtain blood gases
and lactate concentration. Another limitation of this protocol
was that RPE was indistinctly measured. The assessment of
both perceived dyspnoea and pain in the legs would have
been useful, in particular because subjects especially com-
plained from the former after RSH-VHL and from the latter
after RSN, probably due to the greater lactic acidosis.
This study provides interesting information for the
implementation of RSH-VHL. First it shows that this kind
of training could be managed like a “classical” RSH con-
sidering that RSH-VHL did not induce greater fatigue nor
was perceived harder than RSN. Second, it is recommended
to perform at least two sets including six sprints or more
to expect a benefit from this method since the main physi-
ological effects occurred in the second part of RSH-VHL.
Third it is not advised to exceed a 6-s time for the sprints
with VHL. Some of the subjects were not always capable of
holding their breath for the whole duration and had to take
an inhalation between the 5th and 6ths. One could however
consider using sprint durations beyond 6s since significant
performance gains have been reported after RSH with 10-s
sprints followed by 20s of recovery (Faiss etal. 2013a, b,
2015; Gatterer etal. 2014). The feasibility of using such
ratio with RSH-VHL remains, however, to be tested.
In conclusion, this study demonstrated that VHL repre-
sents an interesting way to induce significant arterial and
muscle deoxygenation during RSE whilst maintaining high
intensity level. Oxygen uptake was surprisingly higher in
RSH-VHL than in RSN during the recovery periods follow-
ing sprints at S2. This increase was accompanied by an even
more unexpected reduction in [La] at the end of exercise. On
the basis of the present findings, further studies are required
to investigate the use of VHL as an effective training strategy
to improve repeated sprint ability.
Acknowledgements We would like to sincerely thank all the subjects
who participated in this study for their hard efforts and dedicated time.
References
Ahn B, Nishibayashi Y, Okita S, Masuda A, Takaishi S, Paulev PE,
Honda Y (1989) Heart rate response to breath-holding dur-
ing supramaximal exercise. Eur J Appl Physiol Occup Physiol
59:146–151
Amann M, Romer LM, Subudhi AW, Pegelow DF, Dempsey JA (2007)
Severity of arterial hypoxaemia affects the relative contributions
of peripheral muscle fatigue to exercise performance in healthy
humans. J Physiol 581:389–403
Billaut F, Kerris JP, Rodriguez RF, Martin DT, Gore CJ, Bishop
DJ (2013) Interaction of central and peripheral factors during
repeated sprints at different levels of arterial O2 saturation. PLoS
One 8:e77297. doi:10.1371/journal.pone.0077297
Bishop D, Girard O, Mendez-Villanueva A (2011) Repeated-sprint
ability—part II: recommendations for training. Sport Med
41:741–756 (Review)
Bogdanis GC, Nevill ME, Boobis LH etal (1996) Contribution of
phosphocreatine and aerobic metabolism to energy supply during
repeated sprint exercise. J Appl Physiol 80:876–884
Boushel R, Piantadosi CA (2000) Near-infrared spectroscopy for moni-
toring muscle oxygenation. Acta Physiol Scand 168:615–622
Bowtell JL, Cooke K, Turner R, Mileva KN, Sumners DP (2014) Acute
physiological and performance responses to repeated sprints in
varying degrees of hypoxia. J Sci Med Sport 17:399–403
Brocherie F, Girard O, Faiss R, Millet GP (2015) High-intensity inter-
mittent training in hypoxia: a double-blinded, placebo-controlled
field study in youth football players. J Strength Cond Res 29:226–
237. doi:10.1519/JSC.0000000000000590
Brocherie F, Millet GP, Morin JB, Girard O (2016) Mechanical altera-
tions to repeated treadmill sprints in normobaric hypoxia. Med Sci
Sport Exerc 48:1570–1579. doi:10.1249/MSS.000000000000093
Brocherie F, Girard O, Faiss R, Millet GP (2017a) Effects of repeated-
sprint training in hypoxia on sea-level performance: a meta-anal-
ysis. Sport Med doi:10.1007/s40279-017-0685-3 (Review)
Brocherie F, Millet GP, D’Hulst G, Van Thienen R, Deldicque L,
Girard O (2017b) Repeated maximal-intensity hypoxic exercise
superimposed to hypoxic residence boosts skeletal muscle tran-
scriptional responses in elite team-sport athletes. Acta Physiol.
doi:10.1111/apha.12851
Dempsey JA, Wagner PD (1999) Exercise-induced arterial hypoxemia.
J Appl Physiol 87:1997–2006
Faiss R, Léger B, Vesin JM, Fournier PE, Eggel Y, Dériaz O, Millet
GP (2013a) Significant molecular and systemic adaptations after
repeated sprint training in hypoxia. PLoS One 8:e56522

Eur J Appl Physiol
1 3
Faiss R, Girard O, Millet GP (2013b) Advancing hypoxic training in
team sports: from intermittent hypoxic training to repeated sprint
training in hypoxia. Br J Sport Med 47(Suppl 1):i45–i50
Faiss R, Willis S, Born DP, Sperlich B, Vesin JM, Holmberg HC, Mil-
let GP (2015) Repeated double-poling sprint training in hypoxia
by competitive cross-country skiers. Med Sci Sport Exerc
47:809–817
Fernandez M, Burns K, Calhoun B, George S, Martin B, Weaver C
(2007) Evaluation of a new pulse oximeter sensor. Am J Crit Care
16:146–152
Gaesser GA, Brooks GA (1984) Metabolic bases of excess post-
exercise oxygen consumption: a review. Med Sci Sport Exerc
16:29–43
Gatterer H, Philippe M, Menz V, Mosbach F, Faulhaber M, Burtscher
M (2014) Shuttle-run sprint training in hypoxia for youth elite
soccer players: a pilot study. J Sport Sci Med 13:731–735
Girard O, Mendez-Villanueva A, Bishop D (2011) Repeated-sprint
ability—part I: factors contributing to fatigue. Sport Med 41:673–
694 (Review)
Girard O, Brocherie F, Millet GP (2017) Effects of altitude/hypoxia on
single- and multiple-sprint performance: a comprehensive review.
Sport Med. doi:10.1007/s40279-017-0733-z
Glaister M, Howatson G, Pattison JR, McInnes G (2008) The reliability
and validity of fatigue measures during multiple-sprint work: an
issue revisited. J Strength Cond Res 22:1597–1601
Goods P SR, Dawson BT, Landers GJ, Gore CJ, Peeling P (2014) Effect
of different simulated altitudes on repeat-sprint performance in
team-sport athletes. Int J Sport Physiol Perform 9:857–862.
doi:10.1123/ijspp.2013-0423
Grassi B, Pogliaghi S, Rampichini S, Quaresima V, Ferrari M, Mar-
coni C, Cerretelli P (2003) Muscle oxygenation and pulmonary
gas exchange kinetics during cycling exercise on transitions in
humans. J Appl Physiol 95:149–158
Kasai N, Mizuno S, Ishimoto S, Sakamoto E, Maruta M, Goto K
(2015) Effect of training in hypoxia on repeated sprint perfor-
mance in female athletes. Springerplus 4:310. doi:10.1186/
s40064-015-1041-4
Millet GP, Faiss R, Brocherie F, Girard O (2013) Hypoxic training and
team sports: a challenge to traditional methods? Br J Sport Med
47(Suppl 1):i6–i7. doi:10.1136/bjsports-2013-092793
Millet GP, Brocherie F, Girard O, Wehrlin JP etal (2016) Commentar-
ies on viewpoint: time for a new metric for hypoxic dose? J Appl
Physiol 121:356–358. doi:10.1152/japplphysiol.00460.2016
Morrison J, McLellan C, Minahan C (2015) A Clustered Repeated-
sprint running protocol for team-sport athletes performed in nor-
mobaric hypoxia. J Sport Sci Med 14:857–863
Quanjer PH, Tammeling GJ, Cotes JE, Pedersen OF, Peslin R, Yernault
JC (1993) Lung volumes and forced ventilatory flows. Eur Respir
J 16:5–40. doi:10.1183/09041950.005s1693
Racinais S, Bishop D, Denis R, Lattier G, Mendez-Villaneuva A etal
(2007) Muscle deoxygenation and neural drive to the muscle dur-
ing repeated sprint cycling. Med Sci Sport Exerc 39:268–274
Romer LM, Haverkamp HC, Amann M, Lovering AT, Pegelow DF,
Dempsey JA (2007) Effect of acute severe hypoxia on peripheral
fatigue and endurance capacity in healthy humans. Am J Physiol
Regul Integr Comp Physiol 292:R598–R606
Schallom L, Sona C, McSweeney M, Mazuski J (2007) Comparison of
forehead and digit oximetry in surgical/trauma patients at risk for
decreased peripheral perfusion. Heart Lung 36:188–194
Sharratt MT, Henke KG, Aaron EA, Pegelow DF, Dempsey JA (1987)
Exercise-induced changes in functional residual capacity. Respir
Physiol 70:313–326
Trincat L, Woorons X, Millet GP (2017) Repeated sprint training in
hypoxia induced by voluntary hypoventilation in swimming. Int
J Sports Physiol Perform 12:329–335
West JB (1997) Respiratory physiology. The essentials, 5thedn. Wil-
liams & Wilkins, Baltimore
Woorons X, Mollard P, Pichon A, Duvallet A, Richalet JP, Lamberto
C (2007) Prolonged expiration down to residual volume leads to
severe arterial hypoxemia in athletes during submaximal exercise.
Respir Physiol Neurobiol 15:75–82
Woorons X, Mollard P, Pichon A, Duvallet A, Richalet J-P, Lamberto C
(2008) Effects of a 4-week training with voluntary hypoventilation
carried out at low pulmonary volumes. Respir Physiol Neurobiol
160:123–130
Woorons X, Bourdillon N, Vandewalle H, Lamberto C, Mollard P,
Richalet JP, Pichon A (2010) Exercise with hypoventilation
induces lower muscle oxygenation and higher blood lactate con-
centration: role of hypoxia and hypercapnia. Eur J Appl Physiol
110:367–377
Woorons X, Bourdillon N, Lamberto C, Vandewalle H, Richalet JP,
Mollard P, Pichon A (2011) Cardiovascular responses during
hypoventilation at exercise. Int J Sport Med 32:438–445
Woorons X, Gamelin FX, Lamberto C, Pichon A, Richalet JP (2014)
Swimmers can train in hypoxia at sea level through voluntary
hypoventilation. Respir Physiol Neurobiol 190:33–39
Woorons X, Mucci P, Richalet JP, Pichon A (2016) Hypoventilation
training at supramaximal intensity improves swimming perfor-
mance. Med Sci Sport Exerc 48:119–128