ORIGINAL RESEARCH

published: 12 April 2019

doi: 10.3389/fphys.2019.00409

Relationship of Pre-season Training

Load With In-Season Biochemical
Markers, Injuries and Performance in
Professional Soccer Players
Sullivan Coppalle 1 , Guillaume Rave 2,3 , Abderraouf Ben Abderrahman 4 , Ajmol Ali 5 ,
Iyed Salhi 4 , Sghaier Zouita 4 , Amira Zouita 4 , Matt Brughelli 6 , Urs Granacher 7* and
Hassane Zouhal 1*
1
M2S (Movement, Sport and Health Sciences Laboratory), University of Rennes 2, Rennes, France, 2 Stade Lavallois
Mayenne Football Club, Laval, France, 3 Department of Performance, Université d’Angers, Angers, France, 4 ISSEP Ksar
Said, University of Manouba, Manouba, Tunisia, 5 School of Sport, Exercise and Nutrition, Massey University, Auckland,
New Zealand, 6 Sports Performance Research Institute New Zealand, AUT Millennium, Auckland University of Technology,
Auckland, New Zealand, 7 Division of Training and Movement Sciences, University of Potsdam, Potsdam, Germany

Edited by:
Martin Burtscher,
Introduction: There is controversy in the literature in regards of the link between
University of Innsbruck, Austria training load and injury rate. Thus, the aims of this non-interventional study were to
Reviewed by: evaluate relationships between pre-season training load with biochemical markers,
Luis Manuel Rama, injury incidence and performance during the first month of the competitive period in
University of Coimbra, Portugal
Pantelis Theodoros Nikolaidis, professional soccer players.
Hellenic Military Academy, Greece
Materials and Methods: Healthy professional soccer players were enrolled in this
*Correspondence:
Urs Granacher
study over two pre-season periods. Data sets were available from 26 players during the
[email protected] first season (2014–2015) and 24 players during the second season (2015–2016) who
Hassane Zouhal
completed two pre-season periods (6 weeks each). External training load was assessed
[email protected]
from all athletes during training using Global Positioning System (GPS). Internal training
Specialty section: load was monitored after each training session using rate of perceived exertion (RPE).
This article was submitted to
Before and after each pre-season, blood samples were taken to determine plasma
Exercise Physiology,
a section of the journal lactate dehydrogenase (LDH), creatine kinase (CK) and C-reactive protein (CRP). Injury
Frontiers in Physiology incidence and overall performance (ranking of the team after the first five official games
Received: 22 September 2018 of the championship) were recorded for both seasons separately.
Accepted: 26 March 2019
Published: 12 April 2019 Results: There was no statistically significant difference in mean RPE values of the
Citation: two-preparation periods (2737 ± 452 and 2629 ± 786 AU, p = 0.492). The correlational
Coppalle S, Rave G,
Ben Abderrahman A, Ali A, Salhi I,
analysis did not reveal significant associations between internal and external training load
Zouita S, Zouita A, Brughelli M, (RPE and GPS data) and biological markers. There was a significant positive correlation
Granacher U and Zouhal H (2019)
between RPE and LDH during the 2015/2016 season (r = 0.974, p = 0.001). In addition,
Relationship of Pre-season Training
Load With In-Season Biochemical a significant negative correlation was found between total distance >20 km/h and CRP
Markers, Injuries and Performance during the 2015–2016 season (r = −0.863, p = 0.027). The injury rates for the two
in Professional Soccer Players.
Front. Physiol. 10:409.
seasons were 1.76 and 1.06 per 1000 h exposure for the 2014–2015 and 2015–2016
doi: 10.3389/fphys.2019.00409 seasons, respectively (p = 0.127).

Frontiers in Physiology | www.frontiersin.org 1 April 2019 | Volume 10 | Article 409

Coppalle et al. Training Load in Elite Soccer Players

Conclusion: Our study showed that pre-season training load is not associated with
overall team performance. This association is most likely multifactorial and other factors
(e.g., technical and tactical level of the team, opponents, environment) may play an
important role for the collective team performance. Our findings may help coaches to
better prepare their athletes during pre-season.
Keywords: football, blood sample, monitoring, global positioning system, elite athletes

INTRODUCTION predictor for injuries compared with absolute training load

(Gabbett and Whiteley, 2017; Jones et al., 2017). It has been
Physiological demands in soccer have changed over the past proposed that the pre-competitive period is the period during
years. In fact, the intensity level of soccer matches has increased a season with the highest training load. This was substantiated
tremendously (Carling et al., 2010; Clemente et al., 2013). by Jeong et al. (2011), who showed that average physiological
Today, the average percentage of the maximum heart rate in demands during the pre-season (mean heart rate 124 ± 7
football training varies between 63.5% (Clemente and Nikolaidis, beats/min; training load 4,343 ± 329 AU) were higher compared
2016) and 87.1% (Suarez Arrones et al., 2014) depending on with in-season values (heart rate 112 ± 7 beats/min; training
the expertise level. To withstand the demands of training load 1,703 ± 173 AU). From this it follows that training load is
and competition, high physical fitness levels are needed to higher and/or fitness level lower during the pre-season (Hawkins
cope with the increasing number of matches during a season et al., 2001; Jeong et al., 2011). Several authors highlighted the
and to prevent injuries. Nédélec et al. (2012) postulated impact of the pre-season training load on the incidence of soccer
that players from the Spanish national squad played on injuries. Hawkins et al. (2001), reported a greater proportion of
average 70 games during the 2009–2010 in preparation of overuse injuries, including tendinitis and paratendinitis, during
the 2010 World Cup. Recently, studies used new technologies the pre-season period compared with the in season (10.2 vs.
[e.g., Global Positioning System (GPS)] to analyze activity of 5.8%, p < 0.01) highlighting the need to further address this
soccer players during training and competition and to deduce matter. Woods et al. (2002) were able to show that 17% of
information on underlying physical demands using these data the overall number of injuries over the course of two seasons
sets (Carling et al., 2005; Carling, 2010; Nikolaidis et al., 2018). occurred during the pre-season. In addition, Noya Salces et al.
For instance, Bradley et al. (2013) reported that players in (2014), observed that injury incidence during training was
the English Premier League covered on average 681 m at higher in pre-season and tended to decrease throughout the
running speeds ranging from 19.8 to 25.1 km/h. In addition, season (p < 0.05) in Spanish professional soccer players. More
sprint distances of 248 m at velocities >25.1 km/h were recently, Eliakim et al. (2018) in prospective study conducted
registered during matches. within Israeli professional soccer team, demonstrated that an
The number of the matches played, the intensity and the inappropriate pre-season training period as indicated by lower
physical demands of these matches and related soccer training, improvement in aerobic fitness was associated with a higher
may have an impact on fatigue (Clemente et al., 2017a,b) and incidence of players’ injuries throughout the competitive soccer
soccer-related injuries. In this regard, Wong and Hong (2005), season. This suggests a possible role for high quality pre-
defined injuries in soccer players as “any condition that causes season soccer training not only for shaping physical fitness, but
a player to be removed from a game, miss a game, or to be also (among other factors) for injury prevention. However, the
disabled enough to come to the medical tent.” The injury rates pre-season is considered to be the period with high training
during competition ranged from 8.7 to 65.9 injuries per 1,000 h load and concomitant increased risk of sustaining injuries
of exposure (Pfirrmann et al., 2016). In addition, Ekstrand (2016) (Jones et al., 2017).
analyzed the injury incidence rates among teams that played the Besides the impact of training load on injuries, there is
Champions League during the 2015–2016 season and reported evidence that training load is also associated with physical
2.3 injuries per 1,000 h of training and 20 injuries per 1,000 h stress during the preparation period (Clemente et al., 2017a).
of match exposure. Previously, several studies tried to examine This has been established in soccer players through the
the physiological causes of injuries in soccer. From an analysis assessment of blood markers as indicators for inflammation
of 6,000 injuries within 91 professional English football clubs and muscle damage (Malone et al., 2018; Pascoal et al.,
(Hawkins et al., 2001), the majority of injuries were classified as 2018). Lactate dehydrogenase (LDH), creatine kinase (CK)
strains (37%), as sprains (19%) and the lower extremity being and C-reactive protein (CRP) are biochemical markers often
the site of 87% of the injuries reported. Junge and Dvorak used to quantify muscle damage and inflammation (Djaoui
(2004), consider that 20–25% of all non-contact injuries are re- et al., 2017). Anąelković et al. (2015) monitored these three
injuries of the same type and location. Negative adaptations to blood parameters (CK, CRP, and LDH) during a competitive
exercise training are dose related, with the highest incidence of half season in elite soccer players. For instance, Pascoal
illness and injuries occurring when training loads reach high et al. (2018) showed a 64% increase in CK over an 11-
levels (Foster, 1998; Owen et al., 2015). More recently, there is week preparation period in soccer players. However, to the
evidence that the ratio between acute and chronic loads is a better best of the authors’ knowledge, no study compared two

Frontiers in Physiology | www.frontiersin.org 2 April 2019 | Volume 10 | Article 409

Coppalle et al. Training Load in Elite Soccer Players

preparation periods monitored in the same soccer club and in-season, 5–6 training sessions and one match per week were
examined the influence of training load on injuries and scheduled which amounted to an overall exposure of 8 h per
overall performance during the first month of the first week. In-season training sessions consisted of 30 min lower
competitive period. limb eccentric strength training and balance training followed by
Therefore, the aims of this non-interventional study were to 60 min of soccer-specific technical and tactical drills including
evaluate the effects of pre-season training load on biochemical small-sided games, high intensity running and tactical exercises.
markers, injuries and performance during the first month of Thus, training intensity and volume were higher during pre-
the competitive period in professional soccer players. Hence, the compared with in-season. Training volume and intensity as well
possible link between the training load during pre-season period as recovery periods were individualized and could fluctuate from
and collective performance during this period will be explored. one session to the other. However, total training time was the
same across the two seasons.

MATERIALS AND METHODS Quantification of Training Load

Internal Training Load [Rating of Perceived Exertion
Participants (RPE)]
Initially, 35 professional soccer players were enrolled in The training load was quantified on a daily basis by means of the
this study. Full data sets were obtained from 26 players session rating of perceived exertion (session-RPE) using Borg’s
(age: 26.2 ± 5.1 years; height: 179.7 ± 5.1 cm; mass: 6–20 scale (Foster, 1998; Foster et al., 2001). To ensure that
76.2 ± 5.7 kg) during the first season (2014–2015) and 24 the perceived exertion rating was reflective of the entire session
players (age: 25.9 ± 5.2 years; height: 179.0 ± 5.6 cm; mass: rather than the last effort, data was collected 15–20 min following
76.4 ± 5.6 kg) during the second season (2015–2016). A total each training session. Prior to the start of the study, all players
of 14 players participated in both seasons. Coaches, strength and were familiarized with the Borg scale. Moreover, the included
conditioning professionals, and the medical staff were the same fitness coach and exercise scientist verified each player’s answers.
during both seasons. The perception of effort was calculated according to Foster et al.
All participants were players from a second division French (2001) in arbitrary units (AU). Rating of perceived exertion on
soccer club. Over the course of the study, players received the Borg scale was multiplied with effective duration (min) of a
a balanced training program with endurance, speed, agility, single training session.
strength, technical, and tactical aspects that was delivered by
a professional coach. All players were notified of the research External Training Load (Training and Match Exposure)
protocol, benefits and risks before providing written informed [Global Positioning System (GPS)]
consent. The protocol was fully approved by the Medical Center Data of training and match exposure were collected on a weekly
of Stade Lavallois Mayenne Football Club and ethics committee basis. For team match exposure, the total match exposure time in
of the University of Rennes 2. hours for a team was calculated using the following formula:
Procedures (NM∗ PM∗ DM)/60.
To study the training load during soccer, a non-interventional
study was designed (i.e., no intervention during training) and Where NM is the number of team matches played per
data was collected during the pre-season of two competitive week, PM is the number of players on the team (normally 11)
seasons (2014–2015 and 2015–2016). During the 5-week off- and DM represents match duration in minutes (normally 90)
season, all players were asked to perform three sessions per (Fuller et al., 2006).
week of unsupervised low volume and low intensity aerobic For team training exposure, the total training exposure time in
training. This is the traditional duration (4–6 weeks) of an hours was computed using the sum of the values for (PT∗ DT)/60
off-season period in professional soccer (Silva et al., 2016). for every training session throughout the study. PT is the number
Two time points T1 (representing baseline conditions in June) of players attending a training session and DT is the duration of
and T2 (representing the end of the pre-training period in a single training session in minutes (Fuller et al., 2006).
September) were considered as temporal benchmarks relevant Total distance covered and distances covered at different
to the soccer season. Blood samples were collected at each intensities were collected on a daily basis using GPS technology
time point to monitor muscle damage (LDH and CK) and (GPSPORT, 15 Hz). The selected running speed intensities
inflammation (CRP). Internal [ratings of perceived exertion were <12 km/h; 12–16 km/h; 16–20 km/h; 20–25 km/h
(RPE)] and external training loads (GPS) were recorded and >25 km (Bradley et al., 2013). The transmitters were
after each training session and match during the two pre- installed on the players just before each session and removed
season periods. immediately afterward. GPS data were analyzed immediately
after each session.
Training Program
During pre-season, players performed on average between 6 and Biological Analyses
8 training sessions and one game per week. This resulted in Two blood samples were collected during each pre-season. The
an overall weekly training and match exposure of 11 h. During first one was considered as baseline. The second one was taken at

Frontiers in Physiology | www.frontiersin.org 3 April 2019 | Volume 10 | Article 409

Coppalle et al. Training Load in Elite Soccer Players

the end of the pre-season period. Plasma CK, CRP, and LDH were into four categories that have been used in previous studies
measured. The test conditions were standardized. Training loads (Fuller et al., 2006; Clarsen et al., 2012): minimal (≤3 days).
before the test days were kept low and the same procedures were mild (4–7 days), moderate (8–28 days) and severe (>28 days).
applied before all test days. The blood samples were collected (in In addition, recurrent injuries defined as injuries of the same
tubes containing EDTA) in the morning after an overnight fast type and location that occurred after the player recovered and
and on the same day of the week (Wednesday at 8.00 am). returned to full participation were recorded. Recurrent injuries
The blood samples were collected in tubes containing EDTA were classified as less severe equally severe or more severe in
and centrifuged for 10 min at 4◦ and 3000 rpm and the plasma comparison to the original injury (Clarsen et al., 2012).
was stored frozen at −80◦ C until the final analysis. The CRP, Injury rate was calculated as the number of injuries per 1000 h
CK, and LDH activities were determined using a multiparametric of exposure (Fuller et al., 2006).
analyzer (Konelab 30TM , Thermo Electron Corporation). CRP
activity was determined using the immunoturbidimetry method. Overall Performance
The intra-assay coefficient of variation for the CRP kit was 1.7%. To quantify overall performance, results of the team during
CK activity was determined by the UV method (IFCC) using the first five games of the competitive period were considered
the N-acetyl-cysteine. The intra-assay coefficient of variation for (ranking. point’s won. goals scored. goals conceded). The first five
the CK kit was 1.8%. LDH activity was determined by applying games of the season are indicative of the overall performance of
the enzymatic rate method (IFCC). The intra-assay coefficient of the competitive start of the season and it may therefore reflect the
variation for the LDH kit was 1.1%. training effect of the pre-season period. Kelly and Coutts (2007)
showed that three factors primarily determine intensity level of a
Assessment of Injury Rates Over the Two Pre-season match. These are (a) performance level of the opponent, (b) the
Periods number of training days during the week, and (c) match location.
The medical staff of the soccer team reported and validated With regards to the level of the opponent, we cannot ascertain
each injury in accordance with the Fédération Internationale that the performance level was similar between seasons. However,
de Football Association (FIFA) Consensus Statement (FIFA all teams played in the same professional league. The number of
Consensus, 2006). The protocol was used to record the type, days between games was the same for all teams at the beginning
location, and severity of each injury. Responsible researchers and of the two seasons. During the first season, the team played three
medical staff checked the database on a weekly basis. times at home and twice away. During the second season, the
In addition, exposition time was individually registered as team played twice at home and three times away.
each player’s participation during training and matches. The FIFA
standard injury form used to record players injuries was received Statistical Analyses
on a weekly basis from the medical staff team. Recorded injuries Results are expressed as means and standard deviations (SD). The
included any event resulting in the player being unable to train statistical analyses were performed using Statistical Package for
fully or to play matches (time-loss injuries) and the player was Social Sciences for Windows version 16.0 (SPSS Inc. Chicago).
considered injured until the team’s medical staff allowed return The power analysis (point biserial model) was computed with
to training and competition. an assumed Type I error of 0.05, a Type II error rate of
0.20 (80% statistical power), and a large effect size based
Location of Injuries on a previous study with similar study design from Thorpe
In this study, we used the following 12 categories to document and Sunderland (2012) who observed a significant and large
location of injuries. Foot, ankle, lower leg, knee, thigh, hip/groin, sized correlation between the number of sprints performed
upper extremities shoulder/clavicle, lumbar/sacrum/pelvis, during a match and CK values (r = 0.88, p = 0.019). The
head/face/neck/cervical, abdomen and sternum/rib/dorsal. This analysis revealed that 26 participants would be sufficient to
procedure has been applied in previous studies (Hawkins et al., conduct our study. All included variables were tested for
2001; Woods et al., 2002; Fuller et al., 2006). normality of distribution before analysis using the Kolmogorov-
Smirnov test. Student’s t-test for independent samples was
Types of Injury applied to contrast all variables between the two pre-season
Injuries were classified into seven categories in accordance periods. All participants were included in a two-way analysis
with the Consensus Statement for soccer (Fuller et al., 2006). of covariance (ANCOVA) to estimate training load effects on
These included fractures and bone stress joints (non-bone) and the respective outcome variables (LDH, CK, and CRP). Baseline
ligaments, muscles and tendons, contusions, lacerations and skin values were used as covariates. Injury rates were calculated as the
lesions, central/peripheral nervous system and other injuries. number of injuries per 1000 h of training and match exposure
In addition, injuries were also classified as traumatic (those (Fuller et al., 2006). Significant differences were assumed when
with an acute onset) or overuse injuries (those without p < 0.05. Effect sizes were calculated by converting partial eta-
any known trauma). squared to Cohen’s d to document the size of the statistical
The severity of each injury was defined according to the effects observed and defined as small (0.00 ≤ d ≤ 0.49),
number of days elapsed from the date of injury to the date medium (0.50 ≤ d ≤ 0.79), and large (d ≥ 0.80). Correlations
of the player’s return to full participation in team training or between the independent variables LDH, CK, and CRP and the
availability for competition. The injury severity was classified dependent variables external training load (i.e., GPS data) and

Frontiers in Physiology | www.frontiersin.org 4 April 2019 | Volume 10 | Article 409

Coppalle et al. Training Load in Elite Soccer Players

internal training load (i.e., RPE) were determined using simple RPE and GPS Data
regression. The magnitude of the effect for the correlations was There were no statistically significant differences between the two
determined using the modified scale as proposed by Hopkins: seasons in regards of external training load (GPS data, Table 2)
r < 0.1, trivial; 0.1–0.3, small; >0.3–0.5, moderate; >0.5–0.7, and internal training load (2737 ± 452 and 2629 ± 786 AU;
large; >0.7–0.9, very large; >0.9, nearly perfect; and 1 perfect p = 0.492, d = 0.109, small). The daily mean training internal
(Hopkins, 2002). load (RPE) of the team was 456 for the 2014–2015 and 438 AU
for the 2015–2016 preseason (p = 0.465, d = 0.235, small). The
highest internal training loads were achieved during the second
RESULTS week of both pre-seasons (4060 AU for the first season (p = 0.006,
d = 0.577, medium) and 4789 AU for the second season (p = 0.005,
d = 0.601, medium). During the 2014–2015 season, total distance
Overall Performance
covered at running velocities >12 km/h was highest in week 3
In terms of overall performance, statistically significant
(p = 0.032, d = 0.453, small). During the 2015–2016 season, the
differences were found between the two seasons (Table 1). After
highest value was reached in week 4 (p = 0.039, d = 0.399, small)
the first five soccer matches of the season, there was a difference
(2527 m and 2432 m).
of 6 points and 12 places in the table in favor of the second season
(p = 0.022, d = 0.332, small). More, while no win was recorded
during the first five games of the first season, three of the first five Biological Data
games were won in the second season. Finally, the average goal There were no significant differences in CK and CRP
was positive during the 2015–2016 seasons which was not the development from the beginning to the end of the each
case during the 2014–2015 season. pre-season (p > 0.05) (Table 3). However. LDH significantly

TABLE 1 | Statistics of the soccer team after the first 5 matches during the two seasons 2014–2015 (n = 26 players) and 2015–2016 (n = 24 players).

Season Rank /20 teams Pts Matches Win Draw Lost Goal + Goal − Difference

2014–2015 16 4 5 0 4 1 2 3 −1
2015–2016 4 10 5 3 1 1 7 5 2

TABLE 2 | Indicators of external (GPS data) and internal load (RPE) of players determined through the 6 weeks of the pre-season periods in 2014–2015 (n = 26 players)
and 2015–2016 (n = 24 players).

Indicators of external load (GPS data)

Indicators of internal
12–16 km/h 16–20 km/h >20 km/h TD > 12 km/h (m) load (Session RPE)

Week 1
2014–2015 965.2 ± 189.2 162.8 ± 44.8 46.7 ± 18.4 1174 ± 252.6 2315 ± 424.4
2015–2016 581.7 ± 115.3 132.7 ± 23.1 9.3 ± 4.3 723.6 ± 142.7 1657 ± 474.5
Week 2
2014–2015 1812.0 ± 727.7 483.0 ± 240.7 207.5 ± 90.3 2503 ± 1058.8 4060 ± 688.4
2015–2016 1325 ± 309 455.3 ± 66.5 214.1 ± 61.1 1994.3 ± 436.2 4789 ± 1263
Week 3
2014–2015 1528 ± 218.4 682.4 ± 126.3 316.2 ± 37.2 2527 ± 382 2772 ± 405.2
2015–2016 1180.6 ± 394 435.7 ± 145.2 192.2 ± 64.1 1808.5 ± 602.8 2828 ± 546.7
Week 4
2014–2015 851.0 ± 279 410.5 ± 248.4 299.3 ± 230.6 1561 ± 759 2462 ± 394.7
2015–2016 1131.6 ± 377 477 ± 159 239 ± 79.6 1847.5 ± 615.8 2003 ± 443.4
Week 5
2014–2015 1069.9 ± 142.9 340.8 ± 54.3 208.8 ± 75.2 1619 ± 272.4 2579 ± 236.1
2015–2016 934 ± 311.3 343.8 ± 114.6 234.5 ± 78.2 1512.2 ± 504.1 2583 ± 645.8
Week 6
2014–2015 676.1 ± 237.4 250.9 ± 58.7 155.6 ± 50.84 1082.6 ± 346.9 2233 ± 363
2015–2016 935.7 ± 311.9 319.3 ± 106.4 159.9 ± 53.3 1415 ± 471.7 1914 ± 266.4
Means
2014–2015 1150.4 ± 432.6 388.4 ± 183.3 205 ± 98.8 1744.4 ± 632.6 2737 ± 452
2015–2016 1014 ± 259.8 360.6 ± 128.1 174.9 ± 86.2 1550.2 ± 459.4 2629 ± 786
P-value 0.345 0.788 0.698 0.954 0.492

TD, total distance; RPE, ratings of perceived exertion; AU, arbitrary unit.

Frontiers in Physiology | www.frontiersin.org 5 April 2019 | Volume 10 | Article 409

Coppalle et al. Training Load in Elite Soccer Players

TABLE 3 | Biological concentrations (LDH, CK, and CRP) of soccer players determined before and after the pre-season periods 2014–2015 and 2015–2016.

2014/2015 P-value Effect size 2015/2016 P-value Effect size

(n = 26 players) (n = 24 players)

Before After Before After

LDH (UI.L−1 ) 188.9 ± 30.0 190.1 ± 32.4 0.791 0.874 174.4 ± 25.0 212.6 ± 34.8 0.007∗∗ 0.904
CK (UI.L−1 ) 339.9 ± 178.0 413.9 ± 335.6 0.290 0.393 256.1 ± 170.7 380.2 ± 176.1 0.079 0.433
CRP (mg.L−1 ) 0.4 ± 0.4 0.79 ± 1.08 0.130 0.207 1.2 ± 1.9 0.9 ± 0.5 0.569 0.197

Data are presented as mean ± SD. ∗∗ Significant difference between before and after season period with p < 0.01.

TABLE 4 | Exposure time during training and matches and injury rates during the TABLE 5 | Type and duration of injuries recorded during the two pre-season
two pre-season periods 2014–2015 (n = 26 players) and 2015–2016 (n = 24 periods and the five first matches of the season.
players) and during the first five matches of the season.
2014–2015 2015–2016
Pre-season Number of Exposure time Exposure time Injury
periods injuries training(a) match(b) rate(c) Traumatic injuries 3 0
Non-traumatic 2 3
2014–2015 5 2664 165 165 1.76 injuries
2015–2016 3 2664 1.06 Total 5 3
(a) Exposure training time calculated using number of players per session (around 24
Traumatic Non- Traumatic Non-
players). Number of training session (around 74 for the studied period. 12 weeks) traumatic traumatic
duration of session per min (90 min). (b) Exposure match time was calculated using
of number of played matches (10 matches for the examined period). Number of Average duration 23.0 ± 31.2 27.5 ± 14.9 0 52.3 ± 54.0
players during the game (11 players) and duration of match per min (90 min). (days)
(c) Injury rate was calculated as the number of injuries divided by hours of exposure All injuries All injuries
and multiplied with 1000. 24.8 ± 23.4 31.9 ± 36.4

increased from 174.37 ± 25.04 to 212.55 ± 34.81 UI.L−1 Traumatic Non- Traumatic Non-
traumatic traumatic
(p = 0.007, d = 0.904, large) during the 2015/2016-
preparation period. Total cumulative 69 55 0 157
duration (days) All injuries All injuries
Correlations Between the Parameters 124 157

Measured
Eight injuries were recorded during both periods (5 in 2014–
performance during the first month of the competitive period
2015 and 3 in 2015/2016). The injury rates for the two seasons
in professional soccer players. To our knowledge, this is the
amounted to 1.76 and 1.06 for 2014/2015 and 2015/2016,
first attempt to investigate this issue in a professional soccer
respectively (Table 4). The difference between the two seasons
team. Our findings imply that training load during the pre-
was not statistically significant (p = 0.127, d = 0.339, small).
season does not have an effect on overall team performance
Table 5 contains type and duration of injuries recorded during
during the first five official matches of the season. Moreover,
the two seasons.
no significant relationship was found between training load and
The relationship between parameters of external and internal
injury incidence, inflammation and muscle damage markers.
training load and biological markers are presented in Table 6.
Thus, skill-related (technical) and tactical factors may play a
Only two statistically significant correlations were found. In fact,
role in the collective performance during the first month of the
there was a near perfect correlation between RPE and LDH PRE
competitive season.
during the 2015/2016 season (n = 17; r = 0.974, p = 0.001,
nearly perfect) (Figure 1). There was a significant negative
correlation between total distance >20 km/h and CRP POST Training Load
during the 2015/2016 season (n = 19; r = −0.863, p = 0.027, very Internal training load was similar during the two pre-seasons
large) (Figure 2). (456 AU in 2014–2015 vs. 438 AU in 2015–2016). Other
No significant correlation was observed between the training researchers found similar values in professional soccer players.
load and the overall performance of the team and the injury rates For instance, Malone et al. (2015), analyzed internal training
recorded during the two pre-seasons. load (i.e., RPE) among professional English soccer players and
observed that during the pre-season, the training load amounted
to 447 ± 209 AU. There is equivocal data in the literature with
DISCUSSION some reporting similar results (Redkva et al., 2017), and others
reporting lower values (321 AU for Jeong et al., 2011 and for
The aim of this study was to examine the effects of pre-season Clemente et al., 2017a, 308 AU (1 match per week) and 245 AU
training load on biochemical markers, injury incidence and (2 matches per week). However, these discrepancies in findings

Frontiers in Physiology | www.frontiersin.org 6 April 2019 | Volume 10 | Article 409

Coppalle et al. Training Load in Elite Soccer Players

TABLE 6 | Relationship between external (GPS data), internal indicators (RPE) and biological parameters (LDH, CPK, and CRP).

Indicators LDH LDH CK CK CRP CRP

PRE POST PRE POST PRE POST

Session RPE 2014/2015 r 0.736 0.757 0.041 −0.129 0.008 0.733

P 0.095 0.082 0.939 0.807 0.989 0.097
2015/2016 r 0.974∗∗ 0.510 0.179 0.353 0.394 −2.211
P 0.001 0.734 0.734 0.492 0.440 0.688
Total distance >12 km/h 2014/2015 r −0.205 0.043 −0.739 −0.680 −0.510 0.266
P 0.696 0.935 0.093 0.137 0.310 0.611
2015–2016 r 0.004 −0.468 −0.270 −0.359 −0.688 −0.658
P 0.994 0.350 0.605 0.484 0.131 0.155
Total distance 12–16 km/h 2014/2015 r 0.056 0.425 −0.570 −0.583 −0.443 0.656
P 0.916 0.400 0.238 0.224 0.379 0.157
2015/2016 r 0.103 −0.302 −0.151 −0.299 −0.599 −0.559
P 0.846 0.561 0.776 0.565 0.209 0.249
Total distance 16–20 km/h 2014/2015 r −0.421 −0.280 −0.714 −0.596 −0.407 −0.147
P 0.406 0.591 0.111 0.212 0.423 0.781
2015/2016 r −0.039 −0.511 −0.309 −0.381 −0.701 −0.633
P 0.942 0.300 0.551 0.457 0.120 0.177
Total distance >20 km/h 2014/2015 r −0.419 −0.565 −0.572 −0.452 −0.381 −0.434
P 0.408 0.243 0.235 0.368 0.456 0.390
2015/2016 r −0.229 −0.808 −0.517 0.438 −0.803 −0.863∗
P 0.662 0.052 0.293 0.385 0.055 0.027

r: correlations of Pearson; p: bilateral difference; ∗ Significant correlations. ∗∗ p < 0.01 and ∗p < 0.05.

can most likely be explained by the fact that Jeong et al. (2011) the performance level of the opponent may also have an
used data from only 1 week during the pre-season. Clemente impact on match performance even though the teams were
et al. (2017a) monitored the internal training load in professional from the same league. Fatigue-related declines in physical
soccer players from the Portuguese premier league during one performance have been reported during a soccer match. In fact,
season including the pre-season period. Carling et al. (2010) showed that the covered sprint distances
Global Positioning System data from this study were similar
to those reported by other authors. Scott et al. (2013) monitored
external training load in professional soccer players and observed
distances of 544 m per session at a high running velocity
(>14 km/h) and 132 m per session at a high running velocity
of >19.8 km/h. More recently, Clemente et al. (2018) monitored
one typical training week during the in-season period in
two professional teams from Portugal and the Netherlands.
These authors measured average distances of 585 and 213 m,
respectively for velocities of 14–20 km/h and >20 km/h. In our
study, distances at these intensities were even higher compared
with Scott et al. (2013) and lower compared with Clemente et al.
(2018). Data amounted to 388 and 205 m for the 2014–2015 and
360 and 175 m for the 2015–2016 seasons, respectively. These
differences can mainly be explained by several experimental
factors such as the period of the monitored season and the applied
test device (e.g., GPS).

Overall Performance
In the current study, the first five games of the season were
chosen to represent the overall performance. Team performance
was largely different between the two seasons while training
load was not (Table 1). Soccer match-performance seems to
FIGURE 1 | Relationship between RPE and LDH determined before the
depend on the successful interaction of physical, tactical and pre-season period 2015/2016 (N = 17; r = 0.974. p = 0.001, nearly perfect).
technical characteristics of the game (Carling, 2010). In addition,

Frontiers in Physiology | www.frontiersin.org 7 April 2019 | Volume 10 | Article 409

Coppalle et al. Training Load in Elite Soccer Players

professional soccer players. In fact, Hawkins et al. (2001)

showed a greater proportion of overuse injuries (e.g., tendonitis
and paratendonitis) during the pre- compared with the in-
season (10.2% vs. 5.8%, p < 0.01). Moreover, Woods et al.
(2002) emphasized that over the course of two soccer seasons,
17% of the overall number of injuries occurred during the
pre-season. In addition, Noya Salces et al. (2014) reported
higher injury incidence rates during training in the pre-
season compared with the rest of the season in Spanish
professional soccer players. Ekstrand (2017), showed that during
the 2016–2017 season, injury rates in professional European
soccer leagues averaged 2.3 injuries per 1,000 h of training.
However, in professional soccer players from Israel, Eliakim
et al. (2018) reported higher match injury rates (9.4 per
1,000 match hours) compared with injuries sustained during
training (4.7 per 1,000 training hours) (Eliakim et al., 2018).
These authors further postulated that most of the recorded
injuries were overuse injuries of the lower limbs (71%)
(Eliakim et al., 2018).
The number of injuries recorded in the current study
was relatively low and did not allow us to detect the role
that players’ absence could have on collective performance.
FIGURE 2 | Relationship between total distance >20 km/h and CRP
Training load on both pre-season periods seemed to have a
determined after the pre-season period 2015/2016 (N = 19; r = –0.863, little impact on injuries, since no significant relationship was
p = 0.027, very large). observed. However, recently, Eliakim et al. (2018) examined
the effect of pre-season fitness on injury rate during two
consecutive competitive seasons among Israeli professional
at high intensity decreased from the first 15 min to the last soccer players. These authors observed that although there were
15 min of a match (total high intensity sprinting: 468 m no differences in initial fitness characteristics (in the beginning
vs. 411 m). Of note, technical performance during the match of pre-season training) between injured and non-injured players,
was not affected even if three matches were played within improvements in VO2 max during the pre-season training period
7 days (Carling et al., 2010). In addition, Janković et al. were significantly lower among injured players compared to
(2011) showed that the covered distances during a match are non-injured players.
not associated with the final results of the match. Technical In the current study, the injury rate over the two seasons
parameters of the players and the tactical efficiency appear was lower than two injuries per 1,000 h of exposure (Table 5).
to be of larger importance for the outcome of a soccer Further, we were not able to detect significant correlations
match. Top teams score on average more goals relative to between injury rates overall team performance. Thus, the
the number of shots on the goal (Lago, 2009). According to rather low injury rate during the season 2015–2016 does not
the same authors, better teams also have higher rate of ball appear to play a major role in the collective performance
possession. They showed that ball possession was an important during competition. Dauty and Collon (2011), also showed that
parameter for the match result. Saito et al. (2013) showed there was no relationship between the number of injuries and
that the number of passes and especially, the number of the final ranking in a professional soccer French team lower
pass correctly carried out had an importance on the result injury incidence was strongly correlated with team ranking
of the match. Therefore, more than physical fitness, technical, position (Eirale et al., 2013). Furthermore. Hägglund et al.
tactical and efficacy parameters together with the performance (2013), reported a strong correlation between number of injuries
level of the opposing team appear to have a major impact and final ranking after having followed 24 professional soccer
on the collective team performance that is characterized by teams over 11 years.
the match results.
Biological Follow-Up
The results of inflammation and muscle damage markers
Relationship Between Training Load and measured in this study were similar to those of the data
Injury Incidence reported in the literature. These experiments were conducted
In this study, the average of reported training injuries amounted during the pre-season (Djaoui et al., 2017). In our study, the
to 1.76 per 1,000 h exposure in the seasons 2014–2015 increase of the different parameters was not significant unlike
and 1.06 per 1,000 h of exposure in 2015–2016. These other studies. This divergence could mainly be explained by the
injury rates were recorded during July and August of each fact that in the current study for technical reasons, the blood
season and were below those that were recently reported for samplings were done 1 week after the end of the pre-season

Frontiers in Physiology | www.frontiersin.org 8 April 2019 | Volume 10 | Article 409

Coppalle et al. Training Load in Elite Soccer Players

period, which may be sufficient to allow the various markers RPE, remains the best way to track internal and external training
to recover their initial values. It is also possible that the little load. Training load during the pre-season appears to influence
variability between the different training weeks does not allow for team results to a lesser extend than the technical and tactical
a large variation in blood markers. Hence, another reason may level of the team. This study may suggest that the technical level
be the great heterogeneity of the team. However, in the current and the quality of the players in a team are essential factors with
study, a significant and nearly perfect correlation was found regards to the collective performance.
between RPE and LDH before training started in the 2015/2016
season. In addition, a significant and very large association was
observed between total distance covered (>20 km/h) and CRP CONCLUSION
after the pre-season period 2015/2016. Several studies showed
that intensified training or match exposure could influence the The results of this study showed that training load during the
increase of inflammation and muscle damage markers (Nédélec pre-season period was not related to overall performance and
et al., 2012; Malone et al., 2018; Souglis et al., 2018). In this injury rates of professional soccer players during the first months
regard, the first training session after the off-season may induce of the competitive season. With reference to our findings, it
microtrauma (muscle damage) to structural and contractile can be hypothesized that technical and tactical factors together
components within the muscle fiber. This again results in with performance level of the opponent could have an impact
increased LDH and may affect players’ perceived exertion, even on success in competition. More research with larger cohorts is
if training intensity was kept rather low during this early stage needed to verify these findings.
of the season (Meister et al., 2014). It is well known that
running speeds, accelerations and decelerations over a certain
magnitude (moderate to high) and over a certain period may ETHICS STATEMENT
induce muscle inflammation as indicated by the CRP values
(Young et al., 2012). The protocol was fully approved by the Medical Center of Stade
Lavallois Mayenne Football Club and Ethics Committee of the
Limitations, Strengths and Practical University of Rennes 2.
Applications
Several limitations should be acknowledged in the current study:
(i) blood samples were taken 1 week after the end of the training AUTHOR CONTRIBUTIONS
period and only three blood markers were measured. However,
more markers should be monitored over the entire duration of SC, GR, and HZ conceived and designed the research. SC,
the season (Djaoui et al., 2017). (ii) During the two examined GR, and HZ conducted the experiments. IS, SZ, AZ, and ABA
seasons, only a single competitive soccer team was monitored. In analyzed the data. SC, AA, MB, UG, ABA, SZ, AZ, and HZ wrote
addition, there was fluctuation in the number of the players over the manuscript. All authors read and approved the manuscript.
the two seasons due to traded players. Moreover, only the first
five games of the season were computed to analyze overall team
performance. It will be interesting to monitor several teams from FUNDING
different countries and championships and to extend the analysis
to one or more entire seasons; and (iii) due to a relatively small The authors acknowledge the support of the Deutsche
cohort, the rather low overall number of injuries may have failed Forschungsgemeinschaft (DFG) and Open Access Publishing
to highlight a relationship between training load and injuries. Fund of the University of Potsdam, Germany.
Monitoring injuries and training load during the entire season
from more teams may provide useful results for strength and
conditioning coaches in soccer and medical staff (Soligard et al., ACKNOWLEDGMENTS
2016; Foster et al., 2017).
The monitoring of the training load in professional soccer The authors would like to thank the players and both physical and
players using, for example, simultaneous measures of GPS and medical staff for their participation in this study.

REFERENCES standards of english professional soccer. Hum. Mov. Sci. 32, 808–821. doi:
10.1016/j.humov.2013.06.002
Anąelković, M., Baralić, I., Ðorąević, B., Stevuljević, J. K., Radivojević, Carling, C. (2010). Analysis of physical activity profiles when running with the
N., Dikić, N., et al. (2015). Hematological and biochemical ball in a professional soccer team. J. Sports Sci. 28, 319–326. doi: 10.1080/
parameters in elite soccer players during a competitive half 02640410903473851
season. J. Med. Biochem. 34, 460–466. doi: 10.2478/jomb-2014- Carling, C., Orhant, E., and Le Gall, F. (2010). Match injuries in professional soccer:
0057 inter-seasonal variation and effects of competition type, match congestion
Bradley, P. S., Carling, C., Diaz, A. G., Hood, P., Barnes, C., Ade, J., et al. (2013). and positional role. Int. J. Sports Med. 31, 271–276. doi: 10.1055/s-0029-124
Match performance and physical capacity of players in the top three competitive 3646

Frontiers in Physiology | www.frontiersin.org 9 April 2019 | Volume 10 | Article 409

Coppalle et al. Training Load in Elite Soccer Players

Carling, C., Williams, A. M., and Reilly, T. P. (2005). The Handbook of Soccer Match Janković, A., Leontijević, B., Jelušić, V., Pašić, M., and Mićović, B. (2011).
Analysis: A Systematic Approach to Performance Enhancement. Abingdon: Influence of tactics efficiency on results in serbian soccer super league in season
Routledge. 2009/2010. J. Phys. Educ. Sport Citius Altius Fortius 11, 32–41.
Clarsen, B., Myklebust, G., and Bahr, R. (2012). Development and validation Jeong, T. S., Reilly, T., Morton, J., Bae, S. W., and Drust, B. (2011). Quantification
of a new method for the registration of overuse injuries in sports injury of the physiological loading of one week of “pre-season” and one week of “in-
epidemiology: the oslo sports trauma research centre (OSTRC) overuse injury season” training in professional soccer players. J. Sports Sci. 29, 1161–1166.
questionnaire. Br. J. Sports Med. 47, 495–502. doi: 10.1136/bjsports-2012- doi: 10.1080/02640414.2011.583671
091524 Jones, C. M., Griffiths, P. C., and Mellalieu, S. D. (2017). Training load and fatigue
Clemente, F. M., Couceiro, M. S., Martins, F. M. L., Ivanova, M. O., and Mendes, R. marker associations with injury and illness: a systematic review of longitudinal
(2013). Activity profiles of soccer players during the 2010 world cup. J. Hum. studies. Sports Med. 47, 943–974. doi: 10.1007/s40279-016-0619-5
Kinet. 38, 201–211. doi: 10.2478/hukin-2013-0060 Junge, A., and Dvorak, J. (2004). Soccer injuries. Sports Med. 34, 929–938. doi:
Clemente, F. M., and Nikolaidis, P. T. (2016). Profile of 1-month training load in 10.2165/00007256-200434130-00004
male and female football and futsal players. SpringerPlus 5:694. doi: 10.1186/ Kelly, V. G., and Coutts, A. J. (2007). Planning and monitoring training loads
s40064-016-2327-x during the competition phase in team sports. Natl. Strength Condit. Assoc. 29,
Clemente, F. M., Mendes, M., Nikolaidis, P. T., Calvete, F., Carriço, S., and 32–37. doi: 10.1080/02640414.2014.925572
Owen, A. L. (2017a). Internal training load and its longitudinal relationship Lago, C. (2009). The influence of match location, quality of opposition, and match
with seasonal player wellness in elite professional soccer. Physiol. Behav. 179, status on possession strategies in professional association football. J. Sports Sci.
262–267. doi: 10.1016/j.physbeh.2017.06.021 27, 1463–1469. doi: 10.1080/02640410903131681
Clemente, F. M., and Nikolaidis, P. T., van der Linden, C. M. I., Silva, B. (2017b). Malone, J. J., Di Michele, R., Morgans, R., Burgess, D., Morton, J. P., and Drust, B.
Effects of small-sided soccer games on internal and external load and lower (2015). Seasonal training-load quantification in elite english premier league
limb power: a pilot study in collegiate players. Hum. Mov. 18, 50–57. doi: soccer players. Int. J. Sports Physiol. Perform. 10, 489–497. doi: 10.1123/ijspp.
10.1515/humo-2017-0007 2014-0352
Clemente, F. M., Owen, A. L., Serra-Olivares, J., Nikolaidis, P. T., van der Linden, Malone, S., Mendes, B., Hugues, B., Roe, M., Devenney, S., Collins, K., et al.
C. M. I., and Mendes, B. (2018). Characterization of the weekly external load (2018). Decrements in neuromuscular performance and increases in creatine
profile of professional soccer teams from portugal and the Netherlands. J. Hum. kinase impact training outputs in elite soccer players. J. Strength Cond. Res. 32,
Kinet. 66, 155–201. doi: 10.2478/hukin-2018-2054. 1342–1351. doi: 10.1519/JSC.0000000000001997
Dauty, M., and Collon, S. (2011). Incidence of injuries in French professional Meister, S., Der Funten, K., and Meyer, T. (2014). Repeated monitoring of
soccer players. Int. J. Sports Med. 32, 965–969. doi: 10.1055/s-0031-1283188 blood parameters for evaluating strain and over- load in elite professional
Djaoui, L., Haddad, M., Chamari, K., and Dellal, A. (2017). Monitoring training players: is it justified? J. Sports Sci. 32, 1328–1331. doi: 10.1080/02640414.2014.
load and fatigue in soccer players with physiological markers. Physiol. Behav. 927070
181, 86–94. doi: 10.1016/j.physbeh.2017.09.004 Nédélec, M., McCall, A., Carling, C., Legall, F., Berthoin, S., and Dupont, G. (2012).
Eirale, C., Tol, J. L., Farooq, A., Smiley, F., and Chalabi, H. (2013). Low injury rate Recovery in soccer. Sports Med. 42, 997–1015.
strongly correlates with team success in Qatari professional football. Br. J. Sports Nikolaidis, P. T., Clemente, F. M., Van der Linden, C. I., Rosemann, T., and
Med. Bjsports 47, 807–808. doi: 10.1136/bjsports-2012-091040 Knechtle, B. (2018). Validity and reliability of 10-Hz global positioning system
Ekstrand, J. (2016). UEFA Elite Club Injury Study Report 2015/16. UEFA: Nyon. to assess in-line movement and change of direction. Front. Physiol. 9:228. doi:
Ekstrand, J. (2017). “Overview of football injuries,” in Encyclopidia of Football 10.3389/fphys.2018.00228
Medicine, Vol. 2 (New York, NY: Thieme Medical Publishers), 1–13. Noya Salces, J., Gómez-Carmona, P. M., Moliner-Urdiales, D., Gracia-Marco, L.,
Eliakim, E. Doron, O., Meckel, Y., Nemet, D., and Eliakim A. (2018). Pre-season and Sillero-Quintana, M. (2014). An examination of injuries in Spanish
fitness level and injury rate in professional soccer – a prospective study. Sports Professional Soccer League. J. Sports Med. Phys. Fit. 54, 765–771.
Med. Int. Open 2, E84–E90. doi: 10.1055/a-0631-9346 Owen, A. L., Forsyth, J. J., Wong, D. P., Dellal, A., Connelly, S. P., and Chamari, K.
FIFA Consensus (2006). Nutrition for football: the FIFA/F-MARC consensus (2015). Heart rate–based training intensity and its impact on injury incidence
conference. J. Sports Sci. 24, 663–664. doi: 10.1080/02640410500482461 among elite-level professional soccer players. J. Strength Cond. Res. 29, 1705–
Foster, C. (1998). Monitoring training in athletes with reference to overtraining 1712. doi: 10.1519/JSC.0000000000000810
syndrome. Med. Sci. Sport Exerc. 30, 1164–1168. doi: 10.1097/00005768- Pascoal, E. H. F., Borges, J. H., Franciscon, C. A., Spigolon, L. M. P., and Borin, J. P.
199807000-00023 (2018). Pre-Season training affects negatively the immunological parameters
Foster, C., Florhaug, J. A., Franklin, J., Gottschall, L., Hrovatin, L. A., Parker, S., et al. and creatine kinase but not power performance in young soccer players. Arch.
(2001). A new approach to monitoring exercise training. J. Strength Condit. Res. Sports Med. 2, 94–102.
15, 109–115. Pfirrmann, D., Herbst, M., Ingelfinger, P., Simon, P., and Tug, S. (2016). Analysis
Foster, C., Rodriguez-Marroyo, J. A., and de Koning Monitoring, J. J. (2017). of injury incidences in male professional adult and elite youth soccer players:
Training loads: the past, the present, and the future. Int. J. Sports Physiol. a systematic review. J. Athl. Train. 51, 410–424. doi: 10.4085/1062-6050-
Perform. 12, S22–S28. doi: 10.1123/IJSPP.2016-0388. 51.6.03
Fuller, C. W., Ekstrand, J., Junge, A., Andersen, T. E., Bahr, R., Dvorak, J., Redkva, P. E., Gregorio da Silva, S., Paes, M. R., and Dos-Santos, J. W. (2017). The
et al. (2006). Consensus statement on injury definitions and data collection relationship between coach and player training load perceptions in professional
procedures in studies of football (soccer) injuries. Scand. J. Med. Sci. Sports 16, soccer. Percept. Motor Skills 124, 264–276. doi: 10.1177/0031512516678727
83–92. doi: 10.1111/j.1600-0838.2006.00528.x Saito, K., Yoshimura, M., and Ogiwara, T. (2013). Pass appearance time and pass
Gabbett, T. J., and Whiteley, R. (2017). Two training-load paradoxes: can we work attempts by teams qualifying for the second stage of FIFA World Cup 2010 in
harder and smarter, can physical preparation and medical be teammates? Int. J. South Africa. Football Sci. 10, 65–69.
Sports Physiol. Perform. 12(Suppl 2), S2–S50. doi: 10.1123/ijspp.2016-0321 Scott, B. R., Lockie, R. G., Knight, T. J., Clark, A. C., and Janse de Jonge, X. A.
Hägglund, M., Waldén, M., Magnusson, H., Kristenson, K., Bengtsson, H., and (2013). A comparison of methods to quantify the in-season training load of
Ekstrand, J. (2013). Injuries affect team performance negatively in professional professional soccer players. Int. J. Sports Physiol. Perform. 8, 195–202. doi:
football: an 11-year follow-up of the UEFA champions league injury study. Br. 10.1123/ijspp.8.2.195
J. Sports Med. Bjsports 47, 738–742. doi: 10.1136/bjsports-2013-092215 Silva, J. R., Brito, J., Akenhead, R., and Nassis, G. P. (2016). The transition period in
Hawkins, R. D., Hulse, M. A., Wilkinson, C., Hodson, A., and Gibson, M. soccer: a window of opportunity. Sports Med. 46, 305–313. doi: 10.1007/s40279-
(2001). The association football medical research programme: an audit of 015-0419-3
injuries in professional football. Br. J. Sports Med. 35, 43–47. doi: 10.1136/bjsm. Soligard, T., Schwellnus, M., Alonso, J. M., Bahr, R., Clarsen, B., Dijkstra, H. P.,
35.1.43 et al. (2016). How much is too much? (part 1) international olympic committee
Hopkins, W. G. (2002). A scale of magnitudes for effect statistics. New View Stat. consensus statement on load in sport and risk of injury. Br. J. Sports Med. 50,
502:411. 1030–1041. doi: 10.1136/bjsports-2016-096581

Frontiers in Physiology | www.frontiersin.org 10 April 2019 | Volume 10 | Article 409

Coppalle et al. Training Load in Elite Soccer Players

Souglis, A., Bogdanis, G. C., Chryssanthopoulos, C., Apostolidis, N., and Geladas, Young, W. B., Hepner, J., and Robbins, D. W. (2012). Movement demands in
N. D. (2018). Time course of oxidative stress, inflammation, and muscle Australian rules football as indicators of muscle damage. J. Strength Cond. Res.
damage markers for 5 days after a soccer match: effects of sex and playing 26, 492–496. doi: 10.1519/JSC.0b013e318225a1c4
position. J. Strength Cond. Res. 32, 2045–2054. doi: 10.1519/JSC.000000000000
2436 Conflict of Interest Statement: GR was employed by company Stade Lavallois
Suarez Arrones, L., Torreno, N., Requena, B., Saez de Villareal, E., Casamichana, D., Mayenne Football Club.
Barbero-Alvarez, E., et al. (2014). Match-play activity profile in professional
soccer players during official games and the relationship between external and The remaining authors declare that the research was conducted in the absence of
internal load. J. Sports Med. Phys. Fit. 55, 1417–1422. any commercial or financial relationships that could be construed as a potential
Thorpe, R., and Sunderland, C. (2012). Muscle damage, endocrine, and immune conflict of interest.
marker response to a soccer match. J. Strength Cond. Res. 26, 2783–2790. doi:
10.1519/JSC.0b013e318241e174 Copyright © 2019 Coppalle, Rave, Ben Abderrahman, Ali, Salhi, Zouita, Zouita,
Wong, P., and Hong, Y. (2005). Soccer injury in the lower extremities. Br. J. Sports Brughelli, Granacher and Zouhal. This is an open-access article distributed under the
Med. 39, 473–482. doi: 10.1136/bjsm.2004.015511 terms of the Creative Commons Attribution License (CC BY). The use, distribution
Woods, C., Hawkins, R., Hulse, M., and Hodson, A. (2002). The football association or reproduction in other forums is permitted, provided the original author(s) and
medical research programme: an audit of injuries in professional football– the copyright owner(s) are credited and that the original publication in this journal
analysis of preseason injuries. Br. J. Sports Med. 36, 436–441. doi: 10.1136/bjsm. is cited, in accordance with accepted academic practice. No use, distribution or
36.6.436 reproduction is permitted which does not comply with these terms.

Frontiers in Physiology | www.frontiersin.org 11 April 2019 | Volume 10 | Article 409