Content uploaded by Jerome Koral
Author content
All content in this area was uploaded by Jerome Koral on Oct 11, 2021
Content may be subject to copyright.
Fatigue Measured in Dynamic Versus Isometric Modes
After Trail Running Races of Various Distances
Jerome Koral, Marie Fanget, Laurianne Imbert, Thibault Besson, Djahid Kennouche, Audrey Parent,
Clément Foschia, Jérémy Rossi, and Guillaume Y. Millet
Purpose:Fatigue has previously been investigated in trail running by comparing maximal isometric force before and after the
race. Isometric contractions may not entirely reflect fatigue-induced changes, and therefore dynamic evaluation is warranted.
The aim of the present study was to compare the magnitude of the decrement of maximal isometric force versus maximal
power, force, and velocity after trail running races ranging from 40 to 170 km. Methods:Nineteen trail runners completed
races shorter than 60 km, and 21 runners completed races longer than 100 km. Isometric maximal voluntary contractions
(IMVCs) of knee extensors and plantar flexors and maximal 7-second sprints on a cycle ergometer were performed before and
after the event. Results:Maximal power output (P
max
;−14% [11%], P<.001), theoretical maximum force (F
0
;−11% [14%],
P<.001), and theoretical maximum velocity (−3% [8%], P= .037) decreased significantly after both races. All dynamic
parameters but theoretical maximum velocity decreased more after races longer than 100 km than races shorter than 60 km
(P<.05). Although the changes in IMVCs were significantly correlated (P<.05) with the changes in F
0
and P
max
, reductions
in IMVCs for knee extensors (−29% [16%], P<.001) and plantar flexors (−26% [13%], P<.001) were larger (P<.001) than
the reduction in P
max
and F
0
.Conclusions:After a trail running race, reductions in isometric versus dynamic forces were
correlated, yet they are not interchangeable because the losses in isometric force were 2 to 3 times greater than the reductions
in P
max
and F
0
. This study also shows that the effect of race distance on fatigue measured in isometric mode is true when
measured in dynamic mode.
Keywords:isometric maximal voluntary contraction, dynamic exercise, neuromuscular fatigue assessment, power-force-velocity
profile
Neuromuscular fatigue is usually defined as an exercise-
induced decrease in the maximal isometric force and/or power
output.
1
Yet, physical activities and sport performance such as trail
running not only are based on the capacity to generate isometric
forces but often depend on dynamic parameters such as power
output. Maximal power output reflects the ability of athletes’
neuromuscular system to generate high levels of force and produce
this force at high contraction velocity.
2
This ability of the neuro-
muscular system to produce power can be evaluated and charac-
terized by parabolic power–velocity and linear force–velocity
relationships during a single all-out sprint.
3
Whereas, the power–
force–velocity profile (PFVP) has widely been investigated in
athletes,
4–6
including in recreational marathon runners,
7
to indi-
vidualize training and optimize performance, and it has more rarely
been used in the context of fatigue. Nevertheless, some studies
used PFVP to investigate fatigue due to strength training, that
is, the acute effects of different fatigue protocols
8
or the effects
of 3 interset rest intervals on PRE–POST exercise changes
in PFVP.
9
Since 2003, the town of Chamonix (France) welcomes
thousands of mountain trail runners annually to compete in
different races ranging from 40 to 170 km. Previous studies
showed that neuromuscular fatigue, measured as an exercise-
related decrease in isometric maximal voluntary contraction
(IMVC), is greatly impacted by those races.
10,11
The reasons
why most scientists use the decrement in IMVC as a fatigue
index
12–14
is likely due to the simplicity of its measurement.
Nonetheless, Cheng and Rice
15
showed that the assessment of
IMVC is related to some but not all impairments in neuromuscular
function following dynamic exercise. Various studies,
16,17
includ-
ing some recent investigations from our group,
18,19
also
highlighted that measurements of velocity and power provide
additional information on the etiology of neuromuscular fatigue
induced by dynamic tasks. For example, Krüger et al
19
reported
that IMVC and maximal power output (P
max
) are both indicative
of neuromuscular fatigue, yet depending on the duration and
intensity of exercise, the decrease in IMVC may be greater than
P
max
or vice versa. Thus, they are not interchangeable. Indeed,
these authors
19
suggested that the peripheral changes explaining
the decrease in IMVC are likely attributed to changes within the
muscle, such as a decrease in myoplasmic Ca
2+
concentration
and/or Ca
2+
sensitivity while changes in PFVP would be more
related to metabolic disturbance (ie, increased ADP concentration
and/or decreased ATP concentration) affecting for instance the
rate of cross-bridge dissociation. In addition, Morel et al
20
showed
that low velocity and isometric contractions resulted in a greater
voluntary activation (VA) inhibition (ie, central fatigue) as com-
pared with higher velocity dynamic contractions. This may also
have an effect during testing.
Koral, Imbert, Besson, Kennouche, Rossi, and Millet are with the Laboratoire
Interuniversitaire de Biologie de la Motricité, and Fanget, the Laboratoire SNA-
EPIS, Univ Lyon, UJM-Saint-Etienne, Saint-Etienne, France. Parent is with the
Dept of Biological Sciences, Université du Québec à Montréal (UQÀM), Montreal,
QC, Canada, and CHU Sainte-Justine (CRME) Montréal, QC, Canada. Foschia is
with the Dept of Clinical and Exercise Physiology, Sports Medicine Unit, Faculty of
Medicine, University Hospital of Saint-Etienne, Saint-Etienne, France. Millet is also
with the Inst Universitaire de France, Paris, France. Millet (guillaume.millet@univ-
st-etienne.fr) is corresponding author.
1
International Journal of Sports Physiology and Performance, (Ahead of Print)
https://doi.org/10.1123/ijspp.2020-0940
© 2021 Human Kinetics, Inc. ORIGINAL INVESTIGATION
Brought to you by INSEP | Unauthenticated | Downloaded 10/11/21 09:28 AM UTC
In Krüger et al,
19
the longest fatiguing exercise duration was
90 minutes. Hence, the effect of prolonged (eg, marathon) versus
extreme (ultramarathon) duration tasks on neuromuscular fatigue
measured using both isometric (IMVC) and dynamic (PFVP)
parameters remains unknown. Comparing the effects of prolonged
running exercise of varying distances on IMVC and PFVP could
help to provide more comprehensive insight into the etiology of
neuromuscular fatigue following mountain trail running. Accord-
ingly, the aim of the present study was to compare the magnitude of
the decrement of IMVC versus PFVP parameters (ie, P
max
, theo-
retical maximum force [F
0
] and theoretical maximum velocity
[V
0
]) measured during sprint cycling after trail running races
ranging from 40 to 170 km. More specifically, we wanted to
test 3 hypotheses: (1) based on a previous study,
19
the loss in
P
max
is mostly explained by a decrease in F
0
rather than a decrease
in V
0
, (2) since the distance of the running exercise induces
significant fatigue that negatively affects force
21
thus IMVC and
potentially F
0
/P
max
, the amplitude of the change in PFVP is
dependent on the distance (SHORT vs LONG), (3) a strong
correlation exists between the changes in IMVC and the changes
in F
0
.
Methods
Participants
Out of the 75 runners (49 males and 26 females) who voluntarily
participated in this study, 40 (24 males and 16 females) were able to
complete the testing (Figure 1). They were all registered to one of
the different races of the Ultra-Trail du Mont-Blanc, ranging from
40 km and ± 2300 m of elevation to 170 km and ± 10,000 m
(Table 1). Races were subsequently classified into SHORT (less
than 60 km: Martigny-Combe à Chamonix and Orsières–Champex–
Chamonix) and LONG (more than 100 km: Courmayeur–
Champex–Chamonix, Sur les Traces des Ducs de Savoie, and
Ultra-Trail du Mont-Blanc) distances. Three participants reported
that they ran the entire race with a slower runner (partner or friend)
and did not attempt to complete the race as quickly as possible
during the last 10 km and thus were excluded from the study. All
participants were trained and were not suffering from any chronic
metabolic and muscle diseases. Their characteristics are presented in
Table 1. Written and verbal explanations of the experimental
protocol and associated risks were provided to all participants before
obtaining written informed consent. Ethical approval has been
obtained from the French Ethical Research Committee (CPP Ouest
VI, ethics committee agreement 19.03.14.41740 received on 05/02/
2019) and the study has been registered to ClinicalTrials.gov
(NCT04025138).
Experimental Design
This study was part of a larger study investigating the effect of trail
and ultra-trail racing on various physiological and biomechanical
responses in men and women. Each participant completed one
familiarization and 2 experimental sessions. Participants first vis-
ited our laboratory 5 to 8 weeks before the race to be familiarized
Figure 1 —CONSORT study flow diagram. CONSORT indicates CONsolidated Standards of Reporting Trials; LONG, races longer than 100 km;
POST, after; PRE, before; SHORT, races shorter than 60 km.
2Koral et al
(Ahead of Print)
Brought to you by INSEP | Unauthenticated | Downloaded 10/11/21 09:28 AM UTC
Table 1 Characteristics of the Races and Participants
Race Our classification Distance, km D+, m ITRA category Time (min–max), min Participants (M/F) Age, y Height, m Mass, kg
MCC SHORT 40 2300 S 265–585 5/5 36.3 (9.2) 1.73 (0.08) 65.4 (10.7)
OCC 55 3500 M 378–746 6/3 36.6 (7.5) 1.75 (0.08) 71.1 (15.2)
CCC LONG 101 6100 XL 926–1576 8/3 36.6 (9.2) 1.75 (0.09) 68.9 (9.0)
TDS 145 9100 XXL 1484–2662 1/1 39.5 (6.4) 1.66 (0.15) 56.2 (13.4)
UTMB 170 10,000 XXL 1731–2168 5/3 37.8 (7.0) 1.71 (0.11) 63.2 (10.9)
Abbreviations: CCC, Courmayeur–Champex–Chamonix; F, female; ITRA, International Trail Running Association; LONG, races longer than 100 km; M, male; max, maximum; min, minimum; MCC, Martigny-Combe à
Chamonix; OCC, Orsières–Champex–Chamonix; SHORT, races shorter than 60 km; TDS, Sur les Traces des Ducs de Savoie; UTMB, Ultra-Trail du Mont-Blanc.
(Ahead of Print) 3
Brought to you by INSEP | Unauthenticated | Downloaded 10/11/21 09:28 AM UTC
with the experimental settings. Subsequently, 2 to 3 days before
the race, participants visited the on-site laboratory (National School
of Ski & Mountaineering, Chamonix, France) to perform PRE
tests. Finally, within 60 (15) minutes of finishing the race, they
completed the POST tests and were asked to rate their fatigue and
muscle pain in knee extensor (KE) and plantar flexor (PF) muscles.
Familiarization Session. After a medical examination and a
maximal incremental test on a treadmill, participants performed
2 maximal 7-second duration sprints on cycle ergometer (Monark,
Vansbro, Sweden) separated by a 2-minute rest period. The
resistance was set to 0.5 N·kg
−1
of body mass and 0.7 N·kg
−1
of body mass, for the first and second sprints, respectively.
Afterward, they sat on an isometric knee dynamometer, and
subsequently on a custom-built PF dynamometer, and were famil-
iarized with IMVC measurements of the KE and PF. Participants
were instructed to contract as strongly as possible for ∼4 seconds
during IMVCs.
PRE-Race Tests. Contrary to the familiarization session, and in
order to efficiently record IMVCs, all participants first performed a
standardized warm-up of 10 submaximal isometric contractions,
one near maximal contraction, and subsequently 3 KE IMVC and
3 PF IMVC. All attempts were separated by 30-second rest
(Figure 2). Next, two 7-second sprints with 120-second rest
were performed on the cycle ergometer with the same resistances
as in the familiarization session.
POST-Race Tests. The KE IMVC and PF IMVC of the finishers
were tested on the same ergometers as in PRE. Regarding the
PFVP, the friction loads were reduced to 0.35 and 0.5 N·kg
−1
of
body mass for the first and second sprints, respectively, in order to
take into account the anticipated 30% to 35% reduction in maximal
force production capacity due to fatigue caused by the race.
10,22
PRE and POST resistances were chosen to cover a wide range of
velocities and allow the subjects to reach their P
max
.
23
Data Collection
The PFVP Recordings. All features of the equipment were
described in previous studies.
24,25
The Monark cycle ergometer
was made up of a strain gauge (FN 3030 type; FGP Instrumenta-
tion, Les Clayes-sous-Bois, France) to measure the friction force
and an optical encoder (100 pts/turn, Hengstler type RI 32.0;
Aldingen, Germany) to quantify the flywheel displacement.
Data were sampled at 200 Hz, recorded in LabVIEW software
(NI, Austin, TX) and were filtered with a fourth-order low-pass
Butterworth filter at 12 Hz.
Figure 2 —(A) Experimental protocol, (B) evaluation of KE, and (C) evaluation of PF. Warm-up was not performed during POST sessions. IMVC
indicates isometric maximal voluntary contractions; KE, knee extensors; PF, plantar flexors; POST, after.
4Koral et al
(Ahead of Print)
Brought to you by INSEP | Unauthenticated | Downloaded 10/11/21 09:28 AM UTC
The saddle height was adjusted to allow the participants to
fully extend the leg once seated with the heel on the pedal, and was
kept the same for all sessions. Toe clips were well fastened to avoid
foot displacement from the pedal. Participants started each trial
with right pedal at 45° from the vertical axis and to remain seated at
all time. For each trial, participants were vigorously encouraged to
pedal as fast as possible during the entire sprint.
Isometric Force Recordings. The KE torque was measured on
an isometric knee dynamometer (ARS dynamometry; SP2 Ltd,
LjubIjana, Slovenia) with the hips and the right knee at 90° of
flexion (with 0° referring to the extended neutral position). PF
torque was measured by an instrumented pedal (CS1060 300 Nm;
FGP sensors, Les Clayes-sous-Bois, France). Participants were
seated in a custom-built chair with right hip, knee, and ankle angles
of 90°. The chest was strapped to the chair and heel and forefoot
were securely fixed to the pedal with noncompliant straps to avoid
displacement of the foot during IMVC. All data were recorded and
analyzed using LabChart 8 Software (ADInstruments, Bella Vista,
Australia).
Electrical Nerve Stimulation. Single electrical stimuli were deliv-
ered via constant-current stimulator (DS7A; Digitimer, Welwyn
Garden City, Hertfordshire, United Kingdom) to both the right
femoral (pulse width: 1 ms) and the tibial (pulse width: 0.2 ms)
nerves for KE and PF, respectively. Maximal output voltage was
400 V. Stimulationof the femoral nerve were delivered via a 30-mm
diameter surface cathode manually pressed into the femoral triangle
(Meditrace 100) and a 10 ×5 cm self-adhesive stimulation electrode
(Medicompex SA, Ecublens, Switzerland) located in the gluteal fold
(Figure 2B). Stimulation of the tibial nerve was delivered via a
bipolar bar stimulating electrode with 30-mm anode–cathode spac-
ing (Bipolar Felt Pad Stimulating Electrode part no. E.SB020/4 mm;
Digitimer) placed on the popliteal fossa and parallel to the nerve
(Figure 2C). To determine the optimal intensity of stimulation,
single stimuli were delivered incrementally in relaxed muscle until
the force response plateaued. To ensure supramaximality, a stimulus
intensity of 130% of the intensity to produce the maximal twitch
responses was used. Stimulus intensity was determined at the
beginning of each session.
Data Analysis
Power–Force–Velocity Profile. In order to be consistent with
neuromuscular tests, only the right leg was analyzed. All partici-
pants who did not reach a coefficient of determination R
2
>.85 in
force–velocity relationship were excluded from the study.
The range of velocity data was set between 0% and 98% of
maximal velocity. P
max
was defined using F
0
and V
0
26
:
Pmax =ðF0×V0Þ
4,
where F
0
and V
0
represent the 2 extremes of the FV spectrum and
characterize the dynamic force production capabilities at low and
high velocities, respectively.
27
The slope of FV relationship (S
FV
) was computed as:
SFV =
−F0
V0
:
The power output (P; in Watts) produced at each pedal stroke
during the sprint was computed as presented in Fanget et al
28
by
using the friction force (F
frict
), the inertial force (F
inert
) and the
flywheel linear velocity (V):
P=ðFfrict þFinertÞ×V:
The linear relationship obtained by free deceleration of the flywheel
allowed to define the inertia.
24
PRE–POST variations of all
dynamic parameters, that is, P
max
(ΔP
max
), F
0
(ΔF
0
), V
0
(ΔV
0
)
and S
FV
(ΔS
FV
) were calculated.
Isometric Forces. The maximal torque values were determined as
the highest peak torque recorded from the contractions (out of the 3
IMVCs for PF and KE). PRE–POST variations of KE (ΔKE), PF
(ΔPF), and KE + PF as a surrogate of the lower limb extensors force
production (ΔKE + PF obtained by summing the 2 torques at PRE and
comparing it to the sum of those 2 torques at POST) were calculated.
Electrical Nerve Stimulation. During the last 2 IMVCs for KE
and PF, a superimposed high frequency doublet (Db100) was
delivered on the force plateau. Afterward, the relaxed muscle
was stimulated by resting Db100 and a single twitch (Pt). ΔPt
was used as an index of peripheral fatigue. Percentage of VA was
then assessed with the conventional ratio of the superimposed
Db100 over the size of the control Db100:
VA =1−superimposed Db100
resting Db100 ×100:
Statistical Analysis
All data were reported as the mean (SD). Linear force–velocity
relationships were plotted under least squares regressions. All
statistical tests were performed using Statistica (version 8; StatSoft,
Inc, Tulsa, OK). Data were checked for normality and homogeneity
of variances using Shapiro–Wilk and Levene tests, respectively.
The effect of race distance on change in PFVP and IMVC
parameters was tested using a 2-way repeated-measures analysis
of variance, that is, distance (LONG–SHORT) ×time (PRE–
POST). Main and interaction effects were followed up with the
Newman–Keuls post hoc pairwise comparisons when appropriate.
The relationships between the PFVP parameters, distance, running
time, and IMVC (PF and KE) were investigated by computing the
Pearson correlation coefficients. When the normality and variance
homogeneity assumptions were not satisfied, the nonparametric
tests, Wilcoxon for PRE–POST comparisons, Mann–Whitney for
SHORT–LONG comparisons, and Spearman for correlations (ρ),
were preferred. The PRE–POST differences for SHORT and
LONG distances were compared using a Student test. The thresh-
old to reject the null hypothesis was set at P<.05. Effect size was
calculated using partial eta square (η2
p).
Results
At PRE, P
max
of participants was 669.8 (185.9) W (range = 289–
1112 W), F
0
was 122.6 (28.4) N (range = 70–192 N), and V
0
was
21.7 (2.1) m·s
−1
(range = 16.5–26.3 m·s
−1
). P
max
,F
0
,V
0
, and
S
FV
were not significantly different between SHORT and LONG
(P= .687, P= .981, P= .359, and P= .651, respectively).
At POST, 5 participants from LONG races did not reach a R
2
value higher than .85 (range = .08–.77) and were thus excluded.
Changes in PFVP
There was a significant time effect for all PFVP variables, that is,
P
max
(−14% [11%], P<.001, η2
p=.59), F
0
(−11% [14%], P<.001,
Fatigue in Trail Running: Isometric vs Dynamic Evaluation 5
(Ahead of Print)
Brought to you by INSEP | Unauthenticated | Downloaded 10/11/21 09:28 AM UTC
η2
p=.39), V
0
(−3% [8%], P= .037, η2
p=.12), and S
FV
(−8% [18%],
P= .037, η2
p=.19) decreased significantly from PRE to POST. F
0
,
P
max
, and S
FV
showed a significant distance ×time interaction
(P= .025, η2
p=.13; P= .047, η2
p=.10; P= .032, η2
p=.12, respec-
tively). Post hoc tests showed a significant decrease in P
max
between PRE and POST for both SHORT and LONG (−9%
[9%], P<.001 and −19% [11%], P= .008, respectively). The
decrease in P
max
was significantly greater for LONG than SHORT
(P= .008). F
0
and S
FV
decreased significantly in LONG (−17%
[14%], P<.001 and −15% [19%], P= .004, respectively) but not in
SHORT (−5% [11%], P= .059 and −1% [14%], P= .687, respec-
tively; Figure 3).
The ΔP
max
was correlated to ΔF
0
(ρ= .908, P<.001; Figure 4A)
and ΔS
FV
(ρ= .802, P<.001) but not to ΔV
0
(ρ=−.129, P= .426;
Figure 4B).
Isometric Force Parameters
There was a time effect for both KE (−28% [16%], P<.001,
η2
p=.69) and PF (−24% [16%], P<.001, η2
p=.77) IMVCs but
distance ×time interaction did not reach significance for these
variables (P= .085 and P= .067, respectively). There were both
time effect (−28% [14%], P<.001, η2
p=.77) and distance ×time
interaction (P<.05, η2
p=.11) for KE + PF IMVC. Significant
correlations were found between dynamic properties and isometric
force parameters (Table 2). In particular, ΔKE + PF was correlated
with ΔP
max
(ρ=.612,P<.001; Figure 4C), ΔF
0
(r=.665,P<.001;
Figure 4D)andΔS
FV
(r=.656, P<.001) but not with ΔV
0
(r=−.216; P=.572).
There was a time effect for both VA
KE
(P<.001, η2
p=.43) and
VA
PF
(P<.001, η2
p=.42) but there was no distance ×time inter-
action for these variables. Similarly, both Pt
KE
and Pt
PF
showed a
time effect (P<.001, η2
p=.68 and η2
p=.49, respectively) but the
distance ×time interaction was not reached for this parameter either
(P= .058 for Pt
KE
and P= .821 for Pt
PF
and P= .056 for Pt
KE+PF
).
Discussion
The aims of the present study were to measure the effects of fatigue
induced by prolonged running exercises on dynamic neuromuscu-
lar properties and to investigate whether or not this differs from
“classic”neuromuscular fatigue measures, that is, in isometric
mode. The main findings were that (1) all dynamic parameters
decreased significantly after the race but the changes were more
pronounced for P
max
and F
0
than V
0
; (2) all PFVP parameters but
V
0
decreased more after LONG than SHORT; (3) the changes in
Figure 3 —(A) P
max
, (B) F
0
, (C) V
0
, and (D) S
FV
measured PRE and POST trail running races. Values are presented as mean (SD) and individual data
of each participant. F
0
indicates theoretical maximum force; LONG, races longer than 100 km; P
max
, maximal power; POST, after; PRE, before; S
FV
,
slope of force–velocity relationship; SHORT, races shorter than 60 km; V
0
, theoretical maximum velocity. *Significant difference at P<.01. **Significant
difference at P<.001. $Significant distance ×time interaction.
6Koral et al
(Ahead of Print)
Brought to you by INSEP | Unauthenticated | Downloaded 10/11/21 09:28 AM UTC
Figure 4 —Correlations for short and long distances (A) between ΔP
max
and ΔF
0
, (B) between ΔP
max
and ΔV
0
, (C) between ΔP
max
and ΔKE + PF, and (D) between ΔF
0
and ΔKE + PF. ΔF
0
indicates PRE–POST variations of theoretical maximum force; KE, knee extensors; ΔKE + PF, PRE–POST variations of KE + PF maximal isometric force; LONG, races longer than 100 km; ΔP
max
,
PRE–POST variations of maximal power; PF, plantar flexors; POST, after; PRE, before; SHORT, races shorter than 60 km; ΔV
0
, PRE–POST variations of theoretical maximum velocity.
(Ahead of Print) 7
Brought to you by INSEP | Unauthenticated | Downloaded 10/11/21 09:28 AM UTC
isometric force were significantly correlated to dynamic force and
maximal power, but the reduction in IMVC were approximately
twice as great as the reductions in P
max
and F
0
. Overall, these
findings suggest that the decreases in isometric versus dynamic
forces due to fatigue induced by trail running are related, yet they
are not interchangeable.
Effects of Fatigue Due to Trail Running of Various
Distances on Dynamic Neuromuscular Properties
This study is the first to compare parameters derived from PFVP
following prolonged running exercise of varying durations (518
[140] min vs 1750 [618] min). An original finding of the present
work is that P
max
and F
0
were directly and differently affected by
race distance (Figure 3A and 3B), while V
0
barely declined and
showed no distance ×time interaction. Indeed, losses in P
max
and
F
0
were 2 (for P
max
) to 3 (for F
0
) times as elevated in LONG (−19%
and −17%, respectively) than in SHORT (−9% and −5%, respec-
tively) races (P<.01).
The loss in P
max
was mainly due to a reduction in F
0
rather
than a loss in V
0
as shown by the high (r>.85) coefficient of
correlation between ΔP
max
and ΔF
0
(Figure 4A) and the similar
magnitude of change for P
max
(−14%) and F
0
(−11%). On the
contrary, no correlation was found between ΔV
0
and ΔP
max
(Figure 4B) and V
0
barely, while significantly, declined (−3%).
This finding was in line with the study of Krüger et al
19
that did not
observe any V
0
decrease after a prolonged cycling exercise, that is,
P
max
was also mainly affected by F
0
. Although the present study
does not allow to determine why the change in power output
depends almost exclusively on F
0
, possible explanations can
first be found in central alterations. Indeed, Morel et al
20
reported
that low velocity and isometric contractions (ie, higher force
production) were associated with higher central fatigue (higher
VA inhibition) when high-velocity contractions induced more
peripheral fatigue (larger changes in contractile properties). The
reasons why the change in power output depends more on F
0
could
also be related to what happens in the muscles. The moderate
decrease in V
0
(η2
p=.12) was expected in the present study as
Krüger et al
19
showed that V
0
was altered for brief and intense
rather than prolonged and low-intensity exercise, that is, the lower
the fatiguing exercise intensity, the lower the reduction in V
0
.These
authors suggested that the reduction in V
0
during intense exercise, in
which maximal power output was systematically reached, was due to
an increased concentration in ADP and/or a decreased concentration in
ATP resulting in a decrease rate of cross-bridge formation and a greater
fatigue in fast-twitch motor units. This was unlikely to occur in the
present study, as it is well known that trail running is submaximal and
mainly aerobic. Here, it can be speculated that F
0
(hence P
max
)
reduction was likely due to central alterations and to excitation–
contraction coupling failure (ie, decreased Ca
2+
release by the sarco-
plasmic reticulum) due to glycogen depletion resulting from prolonged
and exhausting exercise.
29
This is in agreement with the decline in both
Pt
KE
and Pt
PF
(Table 3). Repeated eccentric contractions due to
downhill running may have induced muscle damage which could
have also negatively affected the contractility properties.
30
Fatigue in Dynamic Versus Isometric Modes
In the present study, trail running races induced not only a significant
reduction in P
max
(see above), but also in IMVC in both KE and PF
after SHORT and LONG races (Table 3). A decreased IMVC was
expected as our group has consistently reported similar findings over
the past 20 years.
10,11,31–33
However, the present study performed a
novel comparison between isometric and dynamic measurements.
Contrary to Krüger et al,
19
longer and less intense exercise was
not associated with lower P
max
reduction since ΔP
max
was higher
for LONG than SHORT. This suggests that for prolonged exercise,
that is, when energy expenditure comes almost entirely from
oxidative metabolism, longer exercise leads to greater reduction
in P
max
and IMVC, possibly because force reductions measured in
isometric and dynamic modes share common mechanisms, in
particular central fatigue. Indeed, the correlation between ΔF
0
and ΔIMVC was significant in the present study, in contrast to
the results reported by Krüger et al.
19
It is also worth mentioning
Table 2 Correlations Between Dynamic Properties and Isometric Force
Parameters
Parameter ΔP
max
ΔF
0
ΔV
0
ΔS
FV
ΔKE ρ= .518*** ρ= .511*** ρ=−.215 ρ= .513***
ΔPF ρ= .529*** r= .443** ρ=−.071 r= .394*
ΔKE + PF ρ= .612*** r= .665*** ρ=−.216 r= .656***
ΔP
max SHORT
ΔF
0 SHORT
ΔV
0 SHORT
ΔS
FV SHORT
ΔKE
SHORT
r= .599** r= .606** r=−.355 r= .567*
ΔPF
SHORT
r= .293 r= .210 r= .251 r= .133
ΔKE + PF
SHORT
r= .556* r= .529* ρ=−.275 r= .472*
ΔP
max LONG
ΔF
0 LONG
ΔV
0 LONG
ΔS
FV LONG
ΔKE
LONG
ρ= .319 ρ= .479* ρ=−.284 ρ= .509*
ΔPF
LONG
r= .509* r= .537* ρ=−.311 r= .504*
ΔKE + PF
LONG
r= .480** r= .683*** ρ=−.439 r= .714***
Abbreviations: ρ, Spearman correlation coefficients; ΔF
0
, PRE–POST variations of theoretical maximum force; ΔKE, PRE–
POST variations of knee extensors; ΔKE + PF, PRE–POST variations of maximal isometric force KE + PF; LONG, races
longer than 100 km; ΔPF, PRE–POST variations of plantar flexors; ΔP
max
, PRE–POST variations of maximal power;
SHORT, races shorter than 60 km; ΔV
0
, PRE–POST variations of theoretical maximum velocity; ΔS
FV
, PRE–POST
variations of the slope of FV relationship; r, Pearson correlation coefficients.
*Significant differences at P<.05. **Significant differences at P<.01. ***Significant differences at P<.001.
8Koral et al
(Ahead of Print)
Brought to you by INSEP | Unauthenticated | Downloaded 10/11/21 09:28 AM UTC
that the longer the race, the greater the correlation between IMVC
and F
0
(see Table 2).
However, similar to Krüger et al,
19
after 90 minutes of cycling
at moderate intensity, IMVC (over −25%) decreased more than
P
max
(−14%) after trail running races (SHORT and LONG pooled).
The present results strengthen Krüger’s hypothesis concerning
differences between isometric and dynamic modes which suggest
that these indexes of fatigue do not entirely share the same
physiological mechanisms and, as such, are not interchangeable.
As stated above, ΔP
max
and ΔF
0
but not ΔV
0
were affected by
race distance. There was also a distance ×time interaction for KE +
PF IMVC although the 2 to 3 times greater changes in LONG than
in SHORT for ΔP
max
and ΔF
0
were not found for ΔIMVC as it was
only 50% higher for LONG in isometric mode. Also, there were no
significant differences between SHORT and LONG for VA
KE
and
VA
PF
(Table 3) so that the greater P
max
and F
0
reduction in LONG
were caused by central fatigue as anticipated based on previous
research in ultra-trail exercise.
10,11
The distance ×time interaction
did not reached significance for peripheral fatigue either (eg, P=
.056 for Pt
KE+PF
). Interestingly, some correlations with ΔKE versus
ΔPF differed between SHORT and LONG (Table 2). For instance,
ΔKE was significantly correlated with ΔP
max
and ΔF
0
in SHORT
whereas ΔPF was not. On the opposite, ΔPF was significantly
correlated with ΔP
max
and ΔF
0
in LONG; whereas, ΔKE was only
correlated with ΔF
0
. The reasons for these differences are not clear
(eg, there were no effects of using poles or not) and still need to be
investigated.
This study does not come without limitations. First, as the
present data are part of a larger study, there was a delay of 60
(15) minutes between the end of the race and the postrace assess-
ment. Yet the measurements of dynamic and isometric properties
were taken in close temporal proximity (5–10 min); although, the
participants always started with isometric contractions. Second,
other testing methods to characterize the PFVP of our participants
could have been used. Yet, as trail runners are not used to perform
SJ, CMJ, or sprint, and because our main concern was to avoid our
participants from getting injured, we decided to use the cycle
ergometer with the 2-point method recommended by Garcia-
Ramos et al.
34
It allowed us to increase the number of points
for the FV relation as each resistance would extend the total pool of
data. Nevertheless, because of the specificity of our participants, we
had to use medium values (0.5–0.7 N·kg
−1
) which could compro-
mise the precision of the force–velocity profile and are less reliable
than 0.4 to 1.0 N·kg
−1
as described by Garcia-Ramos et al.
34
Third,
the determination of the 2 friction loads at POST was based on the
results of previous studies on IMVC decrement after ultra-trail
running. In the future, loads must be individualized based on
reduction in IMVC or P
max
measured during a first sprint.
23
Practical Applications
When attempting to assess the etiology of fatigue in athletes, the
present paper suggests that it would be a limited approach to only
focus on IMVC, and that integrating PFVP allows a more holistic
insight into neuromuscular function, while also being more represen-
tative of multijoint dynamic performance, such as running. Since both
types of evaluation are fast and easy to perform, we suggest that
scientists and coaches integrate isometric and dynamic measurements,
that is, isometric maximal voluntary contraction and PFVP, to better
quantify/understand acute and chronic fatigue. This is an important
consideration in the context of training and fatigue monitoring.
Conclusions
The present study brings important insight into fatigue etiology in trail
running from short (∼50 km) to longer (up to 170 km) trail running
races. Our findings indicate that all PVFP parameters decreased after
the race, yet the effect of distance differs between the PVFP variables
since the magnitude of losses was greater in LONG than in SHORT
for P
max
and F
0
but not for V
0
.ThelossinP
max
was mostly explained
by a decrease in F
0
. Finally, it is concluded that while P
max
,F
0
,and
IMVC are strongly correlated, they are not interchangeable as the
magnitude of their changes was different, that is, isometric and
dynamic measurements identify distinct fatigue responses.
Acknowledgments
The authors would like to thank the Saint-Etienne University Hospital, the
organizers of the Ultra-Trail du Mont Blanc, the National School of Ski &
Mountaineering for logistical support and all the participants. The authors
also acknowledge the technical help provided by Dr Diana Rimaud and
Prof Léonard Féasson as well as Dr Pierre Samozino for his scientific
contribution and Dr Callum Brownstein for providing suggestions on the
manuscript. This study was funded by an IDEX Lyon fellowship.
References
1. Bigland-Ritchie B, Furbush F, Woods JJ. Fatigue of intermittent
submaximal voluntary contractions: central and peripheral factors.
Table 3 Voluntary and Evoked Isometric Force of KE and PF Before and After the Race
KE PF KE + PF
Parameter Before After Before After Before After
SHORT
MVC, N 251.0 (76.5) 191.3 (67.1) 142.4 (33.9) 108.9 (31.6) 393.4 (106.7) 300.2 (93.4)
Pt, N 62.2 (22.6) 52.0 (23.5) 24.3 (5.9) 20.5 (5.1) 86.5 (25.2) 72.4 (25.3)
VA
PNS
, % 87.9 (9.9) 77.6 (14.4) 93.2 (7.1) 79.9 (15.8) / /
LONG
MVC, N 258.7 (72.3) 172.3 (62.2) 150.4 (32.2) 103.6 (35.3) 410.6 (101.0) 274.5 (89.9)
Pt, N 60.3 (19.4) 44.5 (14.2) 22.2 (4.4) 18.1 (5.9) 81.3 (22.7) 61.7 (18.4)
VA
PNS
, % 85.4 (8.1) 69.9 (18.0) 92.3 (8.7) 82.7 (15.9) / /
Abbreviations: KE, knee extensors; LONG, races longer than 100 km; MVC, maximal voluntary contractions; PF, plantar flexors; PNS, peripheral nerve stimulation; Pt,
peak twitch; SHORT, races shorter than 60 km; VA, voluntary activation.
Fatigue in Trail Running: Isometric vs Dynamic Evaluation 9
(Ahead of Print)
Brought to you by INSEP | Unauthenticated | Downloaded 10/11/21 09:28 AM UTC
J Appl Physiol. 1986;61(2):421–429. PubMed ID: 3745035 doi:10.
1152/jappl.1986.61.2.421
2. Morin JB, Samozino P. Interpreting power-force-velocity profiles for
individualized and specific training. Int J Sports Physiol Perform.
2016;11(2):267–272. PubMed ID: 26694658 doi:10.1123/ijspp.
2015-0638
3. Cross MR, Brughelli M, Samozino P, Morin JB. Methods of power-
force-velocity profiling during sprint running: a narrative review.
Sports Med. 2017;47(7):1255–1269. PubMed ID: 27896682 doi:10.
1007/s40279-016-0653-3
4. Busko K. Power-velocity characteristics and jumping abilities in male
combat athletes. Hum Mov. 2016;17(3):181–184.
5. Nikolaidis PT, Torres-Luque G, Chtourou H, Clemente-Suarez VJ,
Ramirez-Velez R, Heller J. Comparison between jumping vs. cycling
tests of short-term power in elite male handball players: the effect of
age. Mov Sports Sci Sci Mot. 2016;91(1):93–101. doi:10.3917/sm.
091.0093
6. Nikolaidis PT, Papadopoulos VE. Cardiorespiratory power and force-
velocity characteristics in road cycling: the effect of aging and
underlying physiological mechanisms. Med Sport. 2011;15:68–74.
7. Nikolaidis PT, Knechtle B. Force-velocity characteristics and maxi-
mal anaerobic power in male recreational marathon runners. Res
Sports Med. 2019;28(1):99–110. PubMed ID: 31030556 doi:10.
1080/15438627.2019.1608993
8. García-Ramos A, Torrejo´n A, Feriche B, et al. Selective effects of
different fatigue protocols on the function of upper body muscles
assessed through the force-velocity relationship. Eur J Appl Physiol.
2018;118(2):439–447. PubMed ID: 29242994 doi:10.1007/s00421-
017-3786-7
9. González-Hernández JM, Jimenez-Reyes P, Janicijevic D, Tufano
JJ, Marquez G, Garcia-Ramos A. Effect of different interset rest
intervals on mean velocity during the squat and bench press exercises
[published online June 22, 2020]. Sports Biomech. doi:10.1080/
14763141.2020.1766102
10. Temesi J, Rupp T, Martin V, et al. Central fatigue assessed by
transcranial magnetic stimulation in ultratrail running. Med Sci Sports
Exerc. 2014;46(6):1166–1175. PubMed ID: 24195865 doi:10.1249/
MSS.0000000000000207
11. Millet GY, Tomazin K, Verges S, et al. Neuromuscular consequences
of an extreme mountain ultra-marathon. PLoS One. 2011;6(2):
e17059. PubMed ID: 21364944 doi:10.1371/journal.pone.0017059
12. Morris MG, Dawes H, Howells K, Scott OM, Cramp M, Izadi H.
Alterations in peripheral muscle contractile characteristics following high
and low intensity bouts of exercise. Eur J Appl Physiol. 2012; 112(1):
337–343. PubMed ID: 21556817 doi:10.1007/s00421-011-1970-8
13. Temesi J, Arnal PJ, Davranche K, et al. Does central fatigue explain
reduced cycling after complete sleep deprivation? Med Sci Sports
Exerc. 2013;45(12):2243–2253. PubMed ID: 23760468 doi:10.1249/
MSS.0b013e31829ce379
14. Thomas K, Elmeua M, Howatson G, Goodall S. Intensity-dependent
contribution of neuromuscular fatigue after constant-load cycling.
Med Sci Sports Exerc. 2016;48(9):1751–1760. PubMed ID:
27187101 doi:10.1249/MSS.0000000000000950
15. Cheng AJ, Rice CL. Fatigue and recovery of power and isometric
torque following isotonic knee extensions. J Appl Physiol. 2005;
99(4):1446–1452. PubMed ID: 15976360 doi:10.1152/japplphysiol.
00452.2005
16. Babault N, Desbrosses K, Fabre MS, Michaut A, Pousson M.
Neuromuscular fatigue development during maximal concentric
and isometric knee extensions. J Appl Physiol. 2006;100(3):
780–785. PubMed ID: 16282433 doi:10.1152/japplphysiol.
00737.2005
17. Cheng AJ, Rice CL. Isometric torque ad shortening velocity follow-
ing fatigue and recovery of different voluntary tasks in the dorsi-
flexors. Appl Physiol Nutr Metab. 2009;34(5):866–874. PubMed ID:
19935848 doi:10.1139/H09-085
18. Krüger RL, Aboodarda SJ, Samozino P, Rice CL, Millet GY.
Isometric versus dynamic measurements of fatigue: does age matter?
A meta-analysis. Med Sci Sports Exerc. 2018;50(10):2132–2144.
PubMed ID: 29787475 doi:10.1249/MSS.0000000000001666
19. Krüger RL, Aboodarda SJ, Jaimes LM, MacIntosh BR, Samozino P,
Millet GY. Fatigue and recovery measured with dynamic properties
versus isometric force: effects of exercise intensity. J Exp Biol.
2019;222(9):jeb197483. doi:10.1242/jeb.197483
20. Morel B, Clémençon M, Rota S, et al. Contraction velocity influence
the magnitude and etiology of neuromuscular fatigue during repeated
maximal contractions. Scand J Med Sci Sports 2015;25(5):e432–
e441. PubMed ID: 25556533 doi:10.1111/sms.12358
21. Millet GY. Can neuromuscular fatigue explain running strategies
and performance in ultra-marathons?: the flush model. Sports Med.
2011;41(6):489–506. PubMed ID: 21615190 doi:10.2165/11588760-
000000000-00000
22. Temesi J, Arnal PJ, Rupp T, et al. Are females more resistant to extreme
neuromuscular fatigue? Med Sci Sports Exerc. 2015;47(7):1372–1382.
PubMed ID: 25304334 doi:10.1249/MSS.0000000000000540
23. Krüger RL, Peyrard A, di Domenico H, Rupp T, Millet GY, Samo-
zino P. Optimal load for a torque-velocity relationship test during
cycling. Eur J Appl Physiol. 2020;120(11):2455–2466. PubMed ID:
32816143 doi:10.1007/s00421-020-04454-x
24. Arsac LM, Belli A, Lacour JR. Muscle function during brief maximal
exercise: accurate measurements on a friction-loaded cycle ergometer.
Eur J Appl Physiol. 1996;74(1–2):100–106. doi:10.1007/BF00376501
25. Morin JB, Belli A. A simple method for measurement of maximal
downstroke power on friction-loaded cycle ergometer. J Biomech.
2004;37(1):141–145. PubMed ID: 14672578 doi:10.1016/S0021-
9290(03)00253-7
26. Vandewalle H, Peres G, Heller J, Panel J, Monod H. Force-velocity
relationship and maximal power on a cycle ergometer: correlation
with the height of a vertical jump. Eur J Appl Physiol. 1987;56(6):
650–656. doi:10.1007/BF00424805
27. Morin JB, Samozino P. Interpreting power-force-velocity profiles for
individualized and specific training. Int J Sports Physiol Perform.
2016;11(2):267–272. PubMed ID: 26694658 doi:10.1123/ijspp.
2015-0638
28. Fanget M, Rossi J, Samozino P, et al. Dynamic force production
capacities between coronary artery disease patients vs. healthy parti-
cipants on a cycle ergometer. Front Physiol. 2020;10:1639. PubMed
ID: 32038306 doi:10.3389/fphys.2019.01639
29. Ørtenblad N, Nielsen J, Saltin B, Holmberg HC. Role of glycogen
availability in sarcoplasmic reticulum Ca2+ kinetics in human skele-
tal muscle. J Physiol. 2011;589(3):711–725. PubMed ID: 21135051
doi:10.1113/jphysiol.2010.195982
30. Giovanelli N, Biasutti L, Salvadego D, Alemayehu HK, Grassi B,
Lazzer S. Changes in skeletal muscle oxidative capacity after a trail
running race. Int J Sports Physiol Perform. 2020;15(2):278–284.
PubMed ID: 31188647 doi:10.1123/ijspp.2018-0882
31. Millet GY, Lepers R, Maffiuletti NA, Babault N, Martin V, Lattier G.
Alterations of neuromuscular function after an ultramarathon. J Appl
Physiol. 2002;92(2):486–492. PubMed ID: 11796655 doi:10.1152/
japplphysiol.00122.2001
32. Millet GY, Martin V, Lattier G, Ballay Y. Mechanisms contributing
to knee extensor strength loss after prolonged running exercise. J Appl
Physiol. 2003;94(1):193–198. PubMed ID: 12391039 doi:10.1152/
japplphysiol.00600.2002
10 Koral et al
(Ahead of Print)
Brought to you by INSEP | Unauthenticated | Downloaded 10/11/21 09:28 AM UTC
33. Besson T, Rossi J, Marechal M, et al. Neuromuscular fatigue and
recovery after a single vs multi-stage mountain race of the same
distance. Med Sci Sports Exerc. 2020;52(8):1691–1698. PubMed ID:
32079919 doi:10.1249/MSS.0000000000002303
34. García-Ramos A, Torrejo´n A, Morales-Artacho AJ, Pérez-Castilla A,
Jaric S. Optimal resistive forces for maximizing the reliability of leg
muscles’capacities tested on a cycle ergometer. J Appl Biomech.
2018;34(1):47–52. PubMed ID: 28952867 doi:10.1123/jab.2017-0056
Fatigue in Trail Running: Isometric vs Dynamic Evaluation 11
(Ahead of Print)
Brought to you by INSEP | Unauthenticated | Downloaded 10/11/21 09:28 AM UTC